igneum-network/igneum
1
0
Fork 0
Code Issues Pull requests Releases Activity

Merge master 63e11d345 into ci-int-suite under the master-landing lock

Browse source
This commit is contained in:
igneum-labs 2026-10-08 20:10:21 +00:00
commit 55d0830f73
191 changed files with 27027 additions and 882 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
docs/analysis/class-v6/census-packs.md
View file

@ -86,4 +86,15 @@ Per site: 1.000 for every site under 1 GiB; under 256 MiB 1.000 on a quarter sit
**Against the one acceptance rule** (a hash change passes only when the best adversarial implementation becomes worse relative to the best practical GPU implementation, within pre-agreed cost and verification limits): on the address axis the v6 candidate is the class v5 object (both uniform inside their windows to 2.7e-7 at the worst address; both with the era windows' cache shares, which a GPU's L2 sees as well as a chip's SRAM), and the v6 candidate is better on the one address bias the record carried (the stride-bit class, closed by the fold: section 0). A03's class is closed on both by (c') and (a'); nothing in it names a generator cause on v5 or v6 because the concentration is absent at 1e-6 resolution. The window draw's cache shares are layer 8 of the era layout by design (the research lane's reconcile: 0.72 is the Devnet 3 program at about p98 of the draw, mean 0.58), and the "every site on the whole dataset" variant that flattens them (win = 0 everywhere) is the D2 experiment already named, at zero GPU cost.
Owed: the x16 rows for the four chain seeds (the 2^18 rows stand; 4 minutes each on 32 cores when the pool has them); a sample-free cache figure (an LRU trace rather than a top-K ranking) if anyone wants the measured column to equal the model to the third digit.
The x16 rows for the four chain seeds (19:20 UK): worst site 8 to 9 of 33,554,432 reads on every one (2.7e-7), worst cell 5 to 6 of 4,194,304; the pooled 256 MiB figure 0.345 and 0.369 on class v5's p18 and p19 (the window model 0.328 and 0.328), 0.335 on the v6 class's p18 (0.328), with the v6 p19 row in the TSV. Owed: a sample-free cache figure (an LRU trace rather than a top-K ranking) if anyone wants the measured column to equal the model to the third digit.
## 7. The post-review object: the freeze's (c''') read (8 October 2026, 19:5x to 20:34 UK)
The external review's fixes moved the v6 draw (every v6-flagged class reserves five non-load slots as shuffles over the five lane dimensions in a drawn order; a `mad` never names its destination as its second source; the id recipe carries `v6draw/`, then generator 6 and `ProgramClass::V6`), so the packs were re-exported. The harness was rebuilt on each tip (class-v6-census-all2 at c566ae2b1 on 04442d9ca; class-v6-census-all3 at 6debbac1a on b24dfc162, whose draw and acceptance the tips 9415e9c89 and ac86d791 keep), the family predicate setting the window, layer 8 off and the width aside as the crate's own predicate now does, and every id reproduced from the pack's seeds before its row. The read-back from build-1 at 20:24 UK (the founder's rule: a pack is named only when read back): hl-v6-all.tgz sha256 6df3d4307202fe68b9e846b38c8c7758effd65daca47c65103401f9092a6ddb5, program 0x4de7b836cc40a4ea, generator 6, class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw, attempt 0.
| Object | Harness | (A) the rule with (c''') on the pack's program | (B) attempts census, 256 chain-shaped seeds, the pack's era and the node1 state | (C) the window trace, 16 seeds at 2^22 | Verdict |
|---|---|---|---|---|---|
| class-v6 04442d9ca's all class (program 46a8008fac7787a5, generator 5; the first re-export's draw) | all2-c566ae2b, build-6 | accepted; min site ratio 0.99994; bucket +5.00 sigma; worst free bit 3.40; no site over 6 sigma | 256 of 256, 0 exhausted, r 0.129, mean attempt 0.15, max 2, (c''') 0; parts (a') 23, (b) 7, (a) 4, reg64 liveness 3, (c) 1 | per-site distinct ratio 1.0026 to 1.0027 (every site); the item share 1.11 to 1.41x flat against the foldrw control's 1.10 to 1.37x on the same seeds, paired 1.015 to 1.032x | PASS |
| **hl-v6-all, the freeze's object (4de7b836cc40a4ea, generator 6; the same instructions under the generator-6 id)** | all3-6debbac1, build-4 (build-6's pool held by lane D at release class) | accepted; min site ratio 0.99994; bucket +5.00; worst free bit 3.40; no site over 6 sigma | 256 of 256, 0 exhausted, r 0.129, mean 0.15, max 2, (c''') 0 of 294 | the 04442d9ca trace stands (the same draw) | **PASS** |
The review's draw changes read in the parts: the reg64 liveness rule now fires on 3 of 294 candidates (the probe in the crate), and the rejection rate on the all class falls from 0.117 (the rw1 table alone) to 0.129 with the five reserved shuffles, every seed still accepted inside two attempts.

377
docs/analysis/class-v6/coexistence-model.md
View file

@ -21,8 +21,23 @@ manufacturing, deployment and operation only); the paid-development cases sit be
## 0. The result in one page
**Amended 17:4x UK on the coordinator's pin (17:38): the model's die ticket is the adversary lane's complete-machine
figure, USD 2.94 per MH/s (power, cooling, board, the node it must run, yield and packaging), under the plan's
evidence standard (complete-board overhead, same-node and advanced-node results kept separate); lane B's USD 0.6
silicon-plus-board row stays as the silicon floor, a lower bound and not a machine. **Amended again 17:2x UK on the adversary lane's reconciliation (its section 15 at d3ad1549): the USD 2.94 figure
was a 300 W machine (the die throttled to 383 MH/s by its shadow core); the maker raises the power budget until the
power train and the core's silicon bind, and the reconciled unit ticket is USD 1.0 per MH/s (0.5 to 1.6) at 1 to 1.5
kW, the joules unchanged. At the reconciled ticket the die fails (a), (b), (c) and (f) and passes only (d), (e) and
(g) (section 12c, the headline); the 12b pass of (a) and (b) was the 300 W artefact. The served form: "a sunk USD
10 M die fleet holds 42 percent of a growing chain and 62 percent of a flat one on landing (27 to 93 across the
reconciled bounds) and takes every flat or shrinking chain by year 3 to 5; its success statement does not hold at
any reconciled ticket, and what holds the die is the project economics (NRE of USD 50 to 250 per MH/s against USD 1
of unit cost)". Items 1 and 6 below are the first cut's reading at the USD 0.6 silicon floor, which the reconciled
ticket lands near; section 12c is the headline and 12b the superseded 300 W row.**
1. **The DRAM-board chip (a GPU's memory system with a programmable core) passes the success statement on today's
rows; the N2 SRAM die fails it once it exists with its development sunk.** With development sunk, a 3-year life
rows; the N2 SRAM die fails it once it exists with its development sunk AT THE SILICON FLOOR (USD 0.6 per MH/s;
at the pinned machine ticket of USD 2.94 it takes the hybrid's verdict, section 12b).** With development sunk, a 3-year life
and power at USD 0.06 per kWh, the board's cost per accepted MH/s-hour is 307 micro-USD against the best GPU
owner's 156 to 204 at the same electricity and 415 to 588 for a new entrant (0.5x to 0.7x the owner, 1.4x to 1.9x
the entrant); the die's is 72 (2.2x to 2.8x the owner, 5.8x to 8.2x the entrant; at the GPU's 0.12, 3.6x to 4.6x
@ -56,9 +71,12 @@ manufacturing, deployment and operation only); the paid-development cases sit be
not under about a quarter of the GPU entrant's; (c) a fleet above a third of the chain costing more than a year's
miner revenue; (d) GPUs keeping a resale market and a use outside mining; (e) the per-joule gap at the honest knee
under about 3x; (f) the supplier's margin a normal return that does not rise with the halvings; (g) a second
income (proving) for the commodity side that the specialised supplier cannot enter. The board passes all seven at
a 1 to 3 year life; the die fails (a), (b), (c), (e) and (f) at every life, price path and electricity price once
it exists. **Against the pass line**: the board's pass does not need a small network (it holds at 42 TH/s and IGN
income (proving) for the commodity side that the specialised supplier's hash engine cannot earn (amended 19:5x UK
on review B's F09(g): its owner can buy into it with companion GPUs at the GPU entrant's cost, so what remains to
the commodity side is the sunk card's hardware term on the proving income, section 5a). The board passes six of
seven at a 1 to 3 year life (it fails (c), as every chip does: corrected 19:5x UK on F09(c), the first cut's cell
contradicted its own inequality); the die fails (a), (b), (c), (e) and (f) at every life, price path and
electricity price once it exists at the silicon floor, and (a), (b), (c), (f) at the reconciled ticket (12c). **Against the pass line**: the board's pass does not need a small network (it holds at 42 TH/s and IGN
1.00), token appreciation (it holds on the flat and shrinking paths) or scheduled ASIC death (its life axis is 1 to
5 years and the 180-day rotation is not what holds it); the die's failure is not cured by any of the three either,
and the only condition that holds the die is that nobody pays to build it (the surface: IGN 0.73 for a USD 150 M
@ -199,6 +217,40 @@ day with a fault) the internal rows are 0.4 to 0.9 USD a day, still above mining
classes. The condition this adds to section 12 is (g): a second income for the commodity side that the specialised
supplier cannot enter; it holds while proving demand exists and the 16 GB and larger tiers can prove.
### 5a. Amendment, 19:5x UK (review B's F09(g)): the hybrid operator who owns companion GPUs or buys proofs
The first cut treated proving income as closed to the specialised supplier because its hash engine cannot prove. A
company can own companion GPUs, or buy proofs; the single-device limitation does not exclude the operator. The test:
for a hybrid operator (the chip for the hash, companion GPUs bought at MSRP for the proving), the proving surplus per
card-day is the proving income less the companion card's entrant cost per day (hardware over two years with resale,
plus power at 0.12 at the proving watts); for the sunk GPU owner it is the income less power alone. At IGN 0.10, the
pool shared by a 5,000-card fleet, external demand USD 2,000 a day at launch and 20,000 at a spike (build-4, 19:54 UK;
all modelled; the 12 GB tier's internal income the measured zero):
| Class | Proving income per card-day, launch / spike | Companion card's hardware per day at MSRP | Power per day at 0.12 | The hybrid operator's surplus, launch / spike | The sunk owner's surplus, launch / spike |
|---|---|---|---|---|---|
| RTX 5090 | 10.49 / 13.61 | 1.23 | 0.86 | 8.40 / 11.52 | 9.63 / 12.75 |
| RTX 5080 | 8.26 / 10.72 | 0.68 | 0.63 | 6.95 / 9.40 | 7.63 / 10.09 |
| RTX 5070 Ti | 9.44 / 12.25 | 0.51 | 0.58 | 8.36 / 11.16 | 8.87 / 11.68 |
| RTX 5070 (12 GB) | 0.45 / 4.55 | 0.38 | 0.40 | -0.32 / 3.77 | 0.05 / 4.15 |
| RTX 4090 | 10.49 / 13.61 | 1.20 | 0.81 | 8.48 / 11.60 | 9.69 / 12.81 |
| RTX 4080 | 8.81 / 11.43 | 0.82 | 0.63 | 7.36 / 9.98 | 8.18 / 10.80 |
| RTX 4070 (12 GB) | 0.18 / 1.80 | 0.45 | 0.32 | -0.59 / 1.04 | -0.14 / 1.49 |
| RTX 3090 (used) | 4.44 / 5.76 | 1.44 | 0.66 | 2.34 / 3.66 | 3.77 / 5.09 |
| H100 (hosted) | 22.04 / 28.59 | 17.11 | 1.01 | 3.92 / 10.47 | 21.03 / 27.58 |
| A100 (used) | 13.22 / 17.15 | 8.90 | 0.72 | 3.60 / 7.53 | 12.50 / 16.43 |
Reading: at launch-shape demand the proving income on a 16 GB or larger consumer card is 4x to 20x the companion
card's own daily cost, so a hybrid operator buys companion GPUs and captures the proving income with a surplus within
10 to 15 percent of the sunk owner's (the hardware term, USD 0.5 to 1.4 a day on a consumer card); the datacentre
parts are the exception (the hybrid keeps a fifth of the owner's surplus). Buying proofs instead of making them
yields no income and is not a route to it. **Condition (g) is therefore not an exclusive advantage of the commodity
side**: proving is a second market open to anyone who buys GPUs, which the chip's owner can and would; what remains to
the commodity side is the sunk card's hardware term on that income (10 to 15 percent of the surplus on consumer
cards), and the structural point that the proving fleet is GPUs whoever owns them. Condition (g) in section 12 is
re-read accordingly: it holds for the GPU as a device (the proving fleet is GPUs) and not for the GPU owner as a
business against a hybrid operator; it does not move any chip's verdict, since no chip passed (c) or (f) on it.
## 6. Private mining and hardware sales; cheaper derivative chips (D4)
From the surface (the floor file 4.4, the mission lane's shape), `p*` is the break-even price in USD per IGN:
@ -255,6 +307,77 @@ flat or shrinking path by year 3 to 5. The pass line's "small network": the boar
1.00 (the growing path's year 4), so it does not need one; the die's failure is worst in a small network, and a large
one only delays it.
## 8a. Amendment, 17:1x UK: the market-structure axis, at the adversary lane's placed energies
The complete 2.0 plan (docs/plans/igneum-2.0.md at b943c047, the addendum) adds MARKET STRUCTURE as a scenario axis:
private supply (the operator self-mining a fixed fleet), public hardware sales (one manufacturer selling units at
twice their hardware term, buyers entering at cost up to a year's production), multiple suppliers (three derivative
designs at a 1.2x markup and three times the production), and concentration (the largest single supplier's share of
the chain's hash). The three chips are scored at the adversary lane's PLACED whole-machine energies (the design's
10.0r: the board 1.6x node-for-node and 1.9x a node ahead, the stored-half hybrid 2.4x and 2.9x, the die 2.4x and
3.3x, per joule against the 5090 at its lock; per-dollar unchanged), which replace the chip-model energies of
sections 2 and 12 for this axis. Budget USD 10 M (a fleet, or a year's production); entry at year 2; the bought
fleet is sunk and stays while revenue covers its power; the per-class GPU supply curve of section 9. Script
`scratchpad/market.py` on build-4, 17:07 UK; all modelled.
First, the chips re-scored at the placed energies (sunk, 3 years, at 0.06):
| Chip, placed energy | Cost per accepted MH/s-hour, node-for-node / a node ahead | (a) against the best Blackwell owner at 0.12 (line 1.5x) | (b) hardware over the 5080 entrant's (line 0.25) | (e) joules at the knee (line 3x): vs the 5090 lock / the 5070 Ti knee | Label |
|---|---|---|---|---|---|
| GDDR7 board, 1.6x / 1.9x | 363 / 343 micro-USD | 0.72x / 0.76x: PASSES | 0.52: PASSES | 1.6x / 1.2x; 1.9x / 1.4x: PASSES | modelled on the adversary lane's rows |
| Stored-half hybrid, 2.4x / 2.9x | 186 / 172 | 1.41x / 1.52x: ON THE LINE | 0.23: FAILS by a hair | 2.4x / 1.75x; 2.9x / 2.1x: PASSES | the same |
| N2 SRAM die, 2.4x / 3.3x | 115 / 92 | 2.29x / 2.84x: FAILS | 0.07: FAILS | 2.4x / 1.75x; 3.3x / 2.4x: PASSES today, on the line a node ahead | the same |
At the placed energies the die passes (e) where the chip-model energy failed it, and still fails (a), (b), (c) and
(f): its ticket, not its joules, is what fails it, which is the floor file's reading from the other side.
The five-year run with the axis (chip share of the hash / the supplier's margin / the largest single supplier's
share / GPU classes above water, at years 2, 3 and 5):
| Chip | Structure | Growing x2 from 0.10 | Flat 0.10 | Shrinking x0.5 from 0.30 |
|---|---|---|---|---|
| Board | private supply | 9 / 59 / 9 / 14; 7 / 52 / 7 / 15; 5 / 63 / 5 / 16 | 14 / 49 / 14 / 12; 16 / 23 / 16 / 11; 24 / 12 / 24 / 6 | 10 / 55 / 10 / 13; 22 / 15 / 22 / 8; 91 / -57 / 91 / 2 |
| Board | hardware sales | 4 / 60 / 4 / 14; 5 / 52 / 5 / 15; 8 / 63 / 8 / 16 | 3 / 49 / 3 / 12; 3 / 23 / 3 / 11; 6 / 12 / 6 / 6 | 3 / 55 / 3 / 13; 8 / 20 / 8 / 11; 39 / -23 / 39 / 4 |
| Board | multiple suppliers | 24 / 56 / 8 / 13; 31 / 45 / 10 / 14; 42 / 57 / 14 / 15 | 25 / 46 / 8 / 12; 37 / 18 / 12 / 11; 66 / 2 / 22 / 6 | 25 / 52 / 8 / 12; 45 / 7 / 15 / 6; 100 / -236 / 33 / 0 |
| Hybrid | private supply | 19 / 78 / 19 / 13; 16 / 75 / 16 / 15; 10 / 81 / 10 / 16 | 29 / 72 / 29 / 11; 39 / 57 / 39 / 11; 64 / 51 / 64 / 6 | 23 / 76 / 23 / 13; 42 / 52 / 42 / 6; 100 / -65 / 100 / 0 |
| Hybrid | hardware sales | 19 / 78 / 19 / 13; 30 / 72 / 30 / 14; 35 / 79 / 35 / 15 | 29 / 72 / 29 / 11; 49 / 54 / 49 / 8; 89 / 38 / 89 / 4 | 23 / 76 / 23 / 13; 50 / 52 / 50 / 6; 100 / -121 / 100 / 0 |
| Hybrid | multiple suppliers | 49 / 74 / 16 / 12; 70 / 64 / 23 / 12; 78 / 70 / 26 / 13 | 69 / 64 / 23 / 6; 95 / 31 / 32 / 3; 100 / -42 / 33 / 0 | 57 / 70 / 19 / 11; 97 / 22 / 32 / 2; 100 / -158 / 33 / 0 |
| Die | private supply | 50 / 84 / 50 / 12; 43 / 81 / 43 / 13; 30 / 89 / 30 / 15 | 70 / 78 / 70 / 6; 89 / 64 / 89 / 6; 100 / 48 / 100 / 0 | 58 / 82 / 58 / 11; 96 / 55 / 96 / 3; 100 / -177 / 100 / 0 |
| Die | hardware sales | 50 / 84 / 50 / 12; 71 / 77 / 71 / 12; 79 / 84 / 79 / 12 | 70 / 78 / 70 / 6; 99 / 50 / 99 / 2; 100 / -2 / 100 / 0 | 58 / 82 / 58 / 11; 100 / 42 / 100 / 1; 100 / -57 / 100 / 0 |
| Die | multiple suppliers | 95 / 75 / 32 / 6; 99 / 52 / 33 / 2; 100 / 61 / 33 / 2 | 100 / 52 / 33 / 0; 100 / 5 / 33 / 0; 100 / -57 / 33 / 0 | 98 / 67 / 33 / 3; 100 / -29 / 33 / 0; 100 / -57 / 33 / 0 |
Reading, per structure:
1. **Private supply** is the first cut's row: the board coexists (5 to 24 percent of the chain at a 12 to 63 percent
margin, 6 to 16 classes above water) except in a shrinking chain's last year; the hybrid coexists in a growing
chain (10 to 19 percent) and takes a flat or shrinking one (64 to 100 percent by year 5); the die takes the chain
on every flat or shrinking path by year 3 to 5 and holds 30 to 50 percent of a growing one.
2. **Public hardware sales** LOWER the board's and raise the hybrid's and the die's share: a manufacturer's markup
doubles the buyer's hardware term, which is most of the board's cost (its share falls to 3 to 8 percent, with 4 to
16 classes above water) and a small part of the die's (its cumulative production takes 79 to 100 percent of the
chain by year 3 to 5 on every path). Selling the die is worse for coexistence than mining it privately, because a
sold fleet keeps growing while the price covers its power and a private one is fixed.
3. **Multiple suppliers** raise every chip's share (three times the production at a lower markup) and lower the
largest supplier's share to a third of it: the board's three suppliers hold 24 to 66 percent of the chain between
them at 8 to 22 percent each with 6 to 15 classes above water in the growing and flat paths (coexistence with a
less concentrated supply), and take a shrinking chain outright; the hybrid's and the die's three suppliers take the
chain on every path but the growing one (49 to 100 percent), with the largest at a third.
4. **Concentration** is the fourth column: a private fleet is one operator and one supplier at its whole share; a
sold fleet is one supplier at its whole share and many operators; three suppliers cap any one at a third. The
dependence table of section 11 reads the same fact in wafers: the chain is 17 to 128 N2 wafers from one supplier
at any price in the window.
The condition this adds to section 12: (h) the supply structure. The success statement holds for the board under
every structure in the growing and flat paths (its largest supplier never above 24 percent of the chain, its margin
a normal return, 6 to 16 classes above water) and fails for it only in a shrinking chain's last year; it holds for the
hybrid only under private supply in a growing chain; it fails for the die under every structure on every path except
private supply in a growing chain, and a sold or multi-sourced die is worse than a privately mined one. Hardware
sales and multiple suppliers, which the GPU side has by nature, are what make a chip's supply grow with the price and
are therefore not a mitigation for the die; they are one for the board, whose cost is the hardware they mark up.
The figures above and every threshold in this document live in `docs/analysis/class-v6/coexistence-workbook.md` with
their assumptions in columns; none is served as a boundary.
## 9. Miners react with no fixed shares (D4: new)
The reaction rule replacing the first cut's single rule: each class participates with a fraction of its base, the
@ -298,24 +421,49 @@ operator simulation's D1), a GPU fleet of the same hash is tens of thousands of
## 12. The conditions under which the success statement holds (the result)
| Condition | The number on today's rows | DRAM board | SRAM die |
|---|---|---|---|
| (a) the chip's all-in cost per accepted unit at its own electricity within about 1.5x of the best GPU owner's at the GPU's electricity | the owner at 0.12: 263 to 332 micro-USD (5070 Ti, 5080) | 307 at 3 y, 750 at 1 y: PASSES | 72 to 134: FAILS at every life |
| (b) the chip's annualised hardware per MH/s not below about a quarter of the GPU entrant's | the 5080 entrant's hardware 460 micro-USD | 239 to 677: PASSES | 33 to 95: FAILS |
| (c) a fleet above a third of the chain's hash costs more than a year's miner revenue | at IGN 0.10 the chain is 9.6 TH/s, a third 3.2 | USD 18 M of boards against USD 40 to 80 M: PASSES above IGN 0.05 | USD 2.5 M of dies: FAILS at every price under about 3 |
| (d) GPUs keep a resale market and a use outside mining | resale 25 to 55 percent after two years; rental 3x to 5x the mining cost | PASSES (the GPU side's property) | PASSES (the same) |
| (e) the per-joule gap at the honest knee stays under about 3x | Blackwell at the knee 1.70 to 2.06 microjoules | 2.2x to 2.6x: PASSES | 3.7x to 4.5x: FAILS (the bare-lane floor 10x to 12x) |
| (f) the supplier's gross margin is a normal return (under about 70 percent) and does not rise with the halvings | section 8 | 27 to 69 percent, falling with the halvings: PASSES | 85 to 93 percent, rising, or a loss once it is the chain: FAILS |
| (g) a second income (proving) for the commodity side that the specialised supplier cannot enter | section 5: USD 3 to 7 a card-day at launch demand on the 16 GB and larger tiers; 0 for any hash engine | PASSES (the board cannot prove either, which is the GPU's advantage over it) | PASSES (the same) |
| Condition | The number on today's rows | DRAM board (chip-model ticket USD 5.6; at the final USD 4.84 see 12b) | SRAM die at the 300 W machine ticket, USD 2.94 (12b, SUPERSEDED by the reconciled USD 1.0 of 12c, at which (a) 2.06x and (b) 0.09 FAIL and (c) is 12.6x short) | SRAM die at the silicon floor, USD 0.6 (the lower bound, not a machine; the reconciled ticket lands near it) |
|---|---|---|---|---|
| (a) the chip's all-in cost per accepted unit at its own electricity within about 1.5x of the best GPU owner's at the GPU's electricity | the owner at 0.12: 263 to 332 micro-USD (5070 Ti, 5080) | 307 at 3 y, 750 at 1 y: PASSES | 211 / 190 (node-for-node / a node ahead): 1.24x / 1.39x, PASSES | 72 to 134: FAILS at every life |
| (b) the chip's annualised hardware per MH/s not below about a quarter of the GPU entrant's | the 5080 entrant's hardware 460 micro-USD | 239 to 677: PASSES | 0.27 of it: PASSES by a hair | 33 to 95 (0.07): FAILS |
| (c) a fleet above a third of the chain's hash costs more than a year's miner revenue (the inequality: a third of the equilibrium hash x the ticket in USD per MH/s > the year's miner revenue) | at IGN 0.03 / 0.10 / 0.30 / 1.00 the chain is 5.6 / 9.6 / 19.7 / 42.3 TH/s, a third 1.9 / 3.2 / 6.6 / 14.1; the year's revenue USD 12 / 40 / 120 / 400 M | USD 10 / 18 / 37 / 79 M of boards at USD 5.6 per MH/s: **FAILS at every price by 1.2x to 5.1x** (the first cut's "PASSES above IGN 0.05" was a hand-written cell contradicting the inequality, corrected 19:5x UK on review B's F09(c); at the final USD 4.84, 15 M against 40 M, 2.6x short) | USD 9.4 M against 40 M: FAILS by 4.3x (2.2x to 9.7x across IGN 0.03 to 1.00) | USD 1.5 / 2.5 / 5.3 / 11.3 M of dies at USD 0.8: FAILS by 8x to 35x |
| (d) GPUs keep a resale market and a use outside mining | resale 25 to 55 percent after two years; rental 3x to 5x the mining cost | PASSES (the GPU side's property) | PASSES (the same) | PASSES (the same) |
| (e) the per-joule gap at the honest knee stays under about 3x | Blackwell at the knee 1.70 to 2.06 microjoules | 2.2x to 2.6x: PASSES | 2.3x node-for-node: PASSES; 3.1x a node ahead: on the line | 3.7x to 4.5x at the chip-model energy: FAILS (the bare-lane floor 10x to 12x) |
| (f) the supplier's gross margin is a normal return (under about 70 percent) and does not rise with the halvings | section 8 and 12b | 27 to 69 percent, falling with the halvings: PASSES | 72 to 79 percent on the growing path, 49 to 69 flat, a loss once it is the chain: FAILS | 85 to 93 percent, rising, or a loss once it is the chain: FAILS |
| (g) a second income (proving) that the specialised hardware cannot earn and its owner can buy into only at the GPU entrant's cost (amended 19:5x UK, F09(g), section 5a) | section 5: USD 3 to 7 a card-day at launch demand on the 16 GB and larger tiers; 0 for any hash engine; a hybrid operator with companion GPUs keeps 85 to 90 percent of the sunk owner's proving surplus on consumer cards | holds for the GPU as a device (the proving fleet is GPUs whoever owns them); NOT an exclusive advantage of the GPU owner as a business: the hybrid operator captures it | the same | the same |
**The result.** The success statement holds for the stored-dataset DRAM-board chip at a 1 to 3 year life on today's
rows, in every price path, at every electricity price on the axis, with the GPU side's own generation curve narrowing
the gap further and proving as a second income the chip cannot enter; its pass needs no small network (it holds at 42
TH/s and IGN 1.00), no token appreciation (it holds on the flat and shrinking paths) and no scheduled ASIC death (the
life axis, not the rotation, is what the board lives on). The statement does not hold for the N2 SRAM die once that
die exists with its development sunk, at any life, price path or electricity price; a small network makes it worse,
appreciation only delays it, and the rotation does not touch a programmable die. **The model names where it fails: a
sunk SRAM die fleet of USD 10 M or more at any price in the window, and of USD 1 M in a shrinking chain.** The only
**Every (c) cell, generated from the inequality and its inputs (review B's F09(c), 19:5x UK; `sim/economy/coexist/market.py`
M3 and the equilibrium hashes of section 11):** a third of the chain's hash at the per-class equilibrium, times the
chip's ticket, against the year's miner revenue.
| Ticket (USD per MH/s) | IGN 0.03 (USD 12 M a year; a third of the chain 1.9 TH/s) | 0.10 (40 M; 3.2 TH/s) | 0.30 (120 M; 6.6 TH/s) | 1.00 (400 M; 14.1 TH/s) | Verdict |
|---|---|---|---|---|---|
| GDDR7 board, chip model 5.6 | 10.4 M, 1.2x short | 17.8 M, 2.2x short | 36.8 M, 3.3x short | 78.9 M, 5.1x short | FAILS at every price |
| GDDR7 machine, final 4.84 | 9.0 M, 1.3x | 15.4 M, 2.6x | 31.8 M, 3.8x | 68.2 M, 5.9x | FAILS |
| Hybrid, chip model 2.44 | 4.5 M, 2.6x | 7.8 M, 5.1x | 16.0 M, 7.5x | 34.4 M, 11.6x | FAILS |
| Hybrid, final 2.80 | 5.2 M, 2.3x | 8.9 M, 4.5x | 18.4 M, 6.5x | 39.5 M, 10.1x | FAILS |
| SRAM die, reconciled 1.0 | 1.9 M, 6.5x | 3.2 M, 12.6x | 6.6 M, 18x | 14.1 M, 28x | FAILS |
| SRAM die, silicon floor 0.8 with the board (0.6 bare) | 1.5 M, 8x | 2.5 M, 16x | 5.3 M, 23x | 11.3 M, 35x | FAILS |
Condition (c) as written is met by no chip at any price in the window: a chip fleet holding a third of the chain always
costs less than a year of the chain's miner revenue, because every chip in the model is 3x to 15x cheaper per MH/s than
the GPUs that set the equilibrium. The condition therefore does not discriminate between the chips; what does is the
ratio by which each fails (1.2x to 5.9x for the DRAM board against 8x to 35x for the die), which is the dependence
on a single supplier of section 11 read as money. The board's verdict is six of seven at both tickets (it fails (c)
only), not seven; the first cut's seven was the hand-written cell.
**The result (as amended on the pinned ticket, 17:4x UK, and on F09(c) at 19:5x).** The success statement holds for the stored-dataset
DRAM-board chip on six of the seven conditions at a 1 to 3 year life on today's rows, in every price path, at every electricity price on the axis,
with the GPU side's own generation curve narrowing the gap further and proving as a second income the chip cannot
enter; its pass needs no small network (it holds at 42 TH/s and IGN 1.00), no token appreciation (it holds on the flat
and shrinking paths) and no scheduled ASIC death (the life axis, not the rotation, is what the board lives on). For
the N2 SRAM die at the pinned complete-machine ticket (USD 2.94 per MH/s) the statement holds under private supply in
a growing chain only (16 / 14 / 8 percent of the chain, 13 to 16 GPU classes above water) and fails on flat and
shrinking paths by year 5 and under hardware sales or multiple suppliers on those paths, on conditions (c) and (f);
at the silicon floor (USD 0.6, the lower bound) it fails at any life, price path or electricity price. A small network
makes the die worse at either ticket, appreciation only delays it, and the rotation does not touch a programmable
die. **The model names where it fails: a sunk SRAM die fleet of USD 10 M in a flat or shrinking chain at the pinned
ticket (16 to 25 percent of the chain on landing, 56 to 100 percent by year 5), and at any price in the window at the
silicon floor (50 to 70 percent on landing).** The only
condition that holds the die is that nobody pays to build it (section 6: IGN 0.73 for a USD 150 M project at a third
of the chain over three years, 0.22 taking the chain, 0.12 to 0.16 for a revision), which is a statement about an
investor's decision and is carried as such, not as a level the chain stays below.
@ -327,6 +475,190 @@ finding), the share detector that makes (c) visible the week it fails, and the d
to 3,000 per die at the schedule, which moves (c) by 2x to 4x and nothing else). Nothing in the hash moves (a), (b) or
(e) for the die by the factor they need.
## 12a. Amendment, 17:0x UK: the third chip, the adversary lane's hybrid board, scored on the conditions and the surface
The adversary lane's D2(b) first row (17:3x UK, harness `tools/chip-model/mf/flow/d2b.py` on class-v6-adversary): the
hottest half of the items in 1 GiB of N2 SRAM beside the 16-device DRAM board serves 72 percent of the reads; the
activate-bound board runs 3.6x the hashes on the same devices (about 600 MH/s a board); 2.69x per joule against the
5090 at its lock, node for node (2.20 to 3.03); USD 1.88 per MH/s of silicon and memory (USD 2.44 with the system);
USD 250 of SRAM a board (claimed); the uniform-store bound 2.30x and USD 3.34; the hottest three quarters 3.10x and
USD 0.83; a node ahead 3.33x; the 5.5 GiB floor raises the SRAM ticket to USD 690 (USD 2.9 per MH/s) and not the
ratio. Its joules per hash here: 2.33 microjoules (lane 4's 5090 class v5 lock floor) over 2.69, 0.87. Scored on this
file's rows (`scratchpad/hybrid.py`, build-4, 17:03 UK; all modelled):
| Hybrid row (sunk) | 1 y / 3 y / 5 y at 0.06, micro-USD per accepted MH/s-hour | 3 y at 0.12 / 0.25 | Hardware / power at 0.06, 3 y |
|---|---|---|---|
| Hottest half (the first row) | 371 / **178** / 139 | 232 / 349 | 104 / 71 |
| Uniform-store bound | 616 / 272 / 203 | 335 / 472 | 186 / 84 |
| Hottest three quarters | 195 / 109 / 92 | 156 / 258 | 46 / 62 |
| A node ahead | 357 / 164 / 125 | 207 / 302 | 104 / 58 |
| At the 5.5 GiB floor (SRAM USD 690) | 533 / 235 / 175 | 289 / 406 | 161 / 71 |
| The pure board (reference) | 750 / 307 / 219 | 356 / 463 | 239 / 65 |
| The SRAM die (reference) | 134 / 72 / 60 | 101 / 163 | 33 / 38 |
| Condition | The hybrid at 3 years, sunk | Verdict |
|---|---|---|
| (a) all-in within 1.5x of the best GPU owner at the GPU's electricity | 178 against the 5070 Ti owner's 263 at 0.12: 1.48x; against the 5080's 332: 1.87x; at 0.06, 0.9x and 1.1x | ON THE LINE: passes against the 5070 Ti at 0.12 and at any card at 0.06, fails against the 5080 and above at 0.12 and against every card at 0.25 (1.5x to 3.4x) |
| (b) hardware not under a quarter of the GPU entrant's | 104 against the 5080 entrant's 459: 0.23 | FAILS by a hair (0.25 is the line); the uniform-store and floor rows pass (0.41, 0.35), the hottest-three-quarters row fails (0.10) |
| (c) a third of the chain costs more than a year's miner revenue | USD 4.5 / 7.8 / 16 / 34 M for a third at IGN 0.03 / 0.10 / 0.30 / 1.00 against USD 12 / 40 / 120 / 400 M | FAILS at every price (a third of the chain costs a tenth to a third of a year's revenue); the 5.5 GiB floor raises it 1.2x |
| (d) GPUs keep resale and a use outside mining | the GPU side's property | PASSES |
| (e) the per-joule gap at the honest knee under 3x | 2.69x against the 5090 at the lock, 2.38x against the 5080's knee, 1.96x against the 5070 Ti's; 3.33x a node ahead | PASSES today, on the line a node ahead |
| (f) a normal margin that does not rise with the halvings | the five-year runs below: 57 to 83 percent, rising on the flat path | FAILS on the flat and shrinking paths, passes (falling, 75 to 83) on the growing one |
| (g) a second income the supplier cannot enter | a hash engine: 0 | PASSES (the GPU's advantage over it) |
Five years with a sunk USD 10 M hybrid fleet (4.1 TH/s at USD 2.44 per MH/s), the per-class supply curve:
| Path | Year 2 | Year 3 | Year 5 | GPU classes above water |
|---|---|---|---|---|
| Growing x2 from 0.10 | 19 percent; margin 79 | 17 percent; 76 | 10 percent; 83 | 13, 15, 16 |
| Flat 0.10 | 29 percent; 73 | 39 percent; 59 | 64 percent; 57 | 11, 11, 6 |
| Shrinking x0.5 from 0.30 | 23 percent; 77 | 42 percent; 55 | 100 percent; -47 | 13, 6, 0 |
On the surface (operator self-mining, `T0` 2 years, edge 2.7x, hardware USD 2.44 per MH/s; the development cost the
board's plus a share of a 1 GiB SRAM die's, three rows): `p*` in USD per IGN at `L` 1 / 3 years and `q` 0.3 / 1.0:
`C_dev` 50 M: 0.80 / 0.26 at a third, 0.24 / 0.08 taking the chain; 125 M: 1.99 / 0.66 and 0.60 / 0.20; 225 M: 3.59 /
1.18 and 1.08 / 0.36. The margin is 0.87 at 3 years and 0.66 at 1.
Reading: the hybrid sits between the two references on every row, nearer the die on the ticket and nearer the board
on the joules. It passes (d), (e) and (g), sits on the line of (a) (it passes against the cheapest Blackwell card at
the GPU's 0.12 and against every card at 0.06, and fails at 0.25), fails (b) by a hair on its first row and (c) at
every price, and fails (f) on the flat and shrinking paths. **The success statement therefore holds for the pure
board only, as the research lane's first reading said; the hybrid is a coexistence case in a growing chain at cheap
GPU electricity (10 to 19 percent of the chain at 76 to 83 percent margins, 13 to 16 classes above water) and a
failure case in a flat or shrinking one (64 to 100 percent by year 5).** What distinguishes it from the die is that
its ticket is a board's (USD 2.44 per MH/s, 8 to 34 M for a third of the chain) rather than a wafer's, so condition
(c) fails by 3x to 10x rather than 20x to 100x, and the dataset floor moves it (1.2x at 5.5 GiB, more at 8.5 and
11.5 as the SRAM half grows) where it did not move the die at all; the floor is a real lever on this chip and the
schedule's ticket argument (the floor file 3.2) applies to it at the SRAM's USD 250 per GiB.
## 12b. Amendment, 17:1x UK: the re-run at the adversary lane's FINAL placed energies and corrected tickets, and one verdict that moves
The adversary lane's final rows (the design's 10.0t): the complete GDDR7 machine 1.5x per joule node-for-node (1.3x to
1.7x), 1.8x a node ahead, USD 4.84 per MH/s; the stored-half hybrid at the reconciled mean hit rate 0.581, 1.93x
(1.65x to 2.12x), 2.40x a node ahead, USD 2.80 per MH/s (the p98 program 2.09x and USD 1.88); the N2 die 2.3x / 3.1x
at **USD 2.94 per MH/s as a complete machine** (2.14 at the smaller ticket). The tickets are taken as given (complete
machines, no system markup added). The die's ticket is the material change: the chip model's USD 0.6 per MH/s
(lane B's USD 0.25 to 0.4 of silicon plus the board, sections 2 and 12) against the adversary lane's 2.94 for the
whole machine, a 5x difference that this file carries as two rows with their sources rather than resolving.
Re-scored (`scratchpad/market.py`, build-4, 17:14 UK; all modelled; sunk, 3 years, at 0.06):
| Chip at the final rows | Cost per accepted MH/s-hour, node-for-node / a node ahead | (a) vs the best Blackwell owner at 0.12 (line 1.5x) | (b) hardware over the 5080 entrant's (line 0.25) | (c) a third of the chain at IGN 0.03 / 0.10 / 0.30 / 1.00 against a year's revenue | (e) joules vs the 5090 lock / the 5070 Ti knee (line 3x) | Verdict |
|---|---|---|---|---|---|---|
| GDDR7 machine, 1.5x / 1.8x, USD 4.84 | 338 / 317 | 0.78x / 0.83x: PASSES | 0.45: PASSES | USD 9 / 15 / 32 / 68 M against 12 / 40 / 120 / 400: short 1.3x to 5.9x, FAILS | 1.5x / 1.1x; 1.8x / 1.3x: PASSES | passes (a), (b), (d), (e), (g); (c) fails by 1.3x to 5.9x (it failed by less at the chip-model ticket, 5.6); (f) 19 to 66 percent, passes: **the board's verdict holds** |
| Stored-half hybrid, 1.93x / 2.40x, USD 2.80 | 221 / 202 | 1.19x / 1.30x: PASSES (was on the line) | 0.26: PASSES by a hair (was 0.23, a fail by a hair) | 5 / 9 / 18 / 40 M: short 2.3x to 10x, FAILS | 1.9x / 1.4x; 2.4x / 1.75x: PASSES | (c) and (f) fail: **its verdict holds** (coexistence under private supply in a growing chain, failure on flat and shrinking paths) |
| The hybrid at the p98 program, 2.09x, USD 1.88 | 174 / 156 | 1.51x / 1.69x: on the line / fails | 0.18: FAILS | 3.5 / 6 / 12 / 27 M: short 3.4x to 15x | 2.1x / 1.5x: passes | the first reading's row (12a) |
| N2 SRAM die, 2.3x / 3.1x, USD 2.94 (the adversary lane's complete machine) | 211 / 190 | 1.24x / 1.39x: **PASSES** (was 2.3x to 2.8x at USD 0.6) | 0.27: **PASSES by a hair** (was 0.07) | 5.5 / 9.4 / 19 / 41 M: short 2.2x to 9.7x, FAILS (was 20x to 100x) | 2.3x / 1.7x: passes; 3.1x / 2.3x: on the line a node ahead | **the die's verdict MOVES at this ticket**: from "fails (a), (b), (c), (e), (f) at every life, path and tariff" to the hybrid's verdict, "passes (a), (b), (d), (e), (g); fails (c) and (f); coexists under private supply in a growing chain and takes a flat or shrinking chain by year 5" |
| N2 SRAM die at the smaller ticket, USD 2.14 | 177 / 155 | 1.49x / 1.69x: on the line / fails | 0.20: FAILS | 4 / 7 / 14 / 30 M: short 3x to 13x | the same | between the two rows above |
| N2 SRAM die at the chip model's ticket, USD 0.6 (sections 2 and 12) | 72 | 2.2x to 4.6x: FAILS | 0.07: FAILS | 2.3 M at IGN 0.10: short 16x | 3.7x to 4.5x at the chip-model energy: FAILS | the first cut's verdict |
The five-year run at the final rows (a sunk USD 10 M fleet; chip share / margin / largest supplier / GPU classes at
years 2, 3, 5):
| Chip | Structure | Growing x2 from 0.10 | Flat 0.10 | Shrinking x0.5 from 0.30 |
|---|---|---|---|---|
| Board (USD 4.84) | private | 10 / 61 / 10 / 13; 8 / 56 / 8 / 15; 5 / 66 / 5 / 16 | 16 / 52 / 16 / 12; 18 / 28 / 18 / 11; 27 / 19 / 27 / 6 | 11 / 58 / 11 / 13; 25 / 19 / 25 / 8; 98 / -56 / 98 / 1 |
| Board | hardware sales | 6 / 62 / 6 / 14; 8 / 56 / 8 / 15; 13 / 65 / 13 / 16 | 5 / 53 / 5 / 12; 6 / 28 / 6 / 11; 10 / 19 / 10 / 6 | 6 / 58 / 6 / 13; 13 / 24 / 13 / 11; 60 / -21 / 60 / 3 |
| Board | multiple suppliers | 27 / 58 / 9 / 13; 37 / 48 / 12 / 13; 47 / 58 / 16 / 15 | 31 / 48 / 10 / 11; 45 / 19 / 15 / 8; 78 / 0 / 26 / 6 | 30 / 54 / 10 / 12; 66 / 13 / 22 / 6; 100 / -194 / 33 / 0 |
| Hybrid (USD 2.80) | private | 17 / 74 / 17 / 13; 15 / 70 / 15 / 15; 9 / 78 / 9 / 16 | 26 / 67 / 26 / 12; 35 / 50 / 35 / 11; 58 / 45 / 58 / 6 | 20 / 72 / 20 / 13; 39 / 43 / 39 / 6; 100 / -68 / 100 / 0 |
| Hybrid | hardware sales | 17 / 74 / 17 / 13; 27 / 68 / 27 / 14; 32 / 75 / 32 / 15 | 25 / 67 / 25 / 12; 41 / 49 / 41 / 9; 76 / 38 / 76 / 6 | 20 / 72 / 20 / 13; 42 / 43 / 42 / 6; 100 / -92 / 100 / 0 |
| Hybrid | multiple suppliers | 44 / 71 / 15 / 12; 62 / 59 / 21 / 12; 72 / 67 / 24 / 13 | 64 / 60 / 21 / 8; 91 / 28 / 30 / 6; 100 / -39 / 33 / 0 | 52 / 67 / 17 / 11; 94 / 18 / 31 / 3; 100 / -149 / 33 / 0 |
| Die (USD 2.94) | private | 16 / 75 / 16 / 13; 14 / 72 / 14 / 15; 8 / 79 / 8 / 16 | 25 / 69 / 25 / 12; 33 / 53 / 33 / 11; 56 / 49 / 56 / 6 | 20 / 73 / 20 / 13; 38 / 46 / 38 / 6; 100 / -51 / 100 / 0 |
| Die | hardware sales | 16 / 75 / 16 / 13; 26 / 69 / 26 / 14; 31 / 77 / 31 / 15 | 25 / 69 / 25 / 12; 41 / 51 / 41 / 10; 76 / 41 / 76 / 6 | 20 / 73 / 20 / 13; 41 / 46 / 41 / 6; 100 / -74 / 100 / 0 |
| Die | multiple suppliers | 43 / 72 / 14 / 12; 61 / 61 / 20 / 12; 70 / 70 / 23 / 13 | 63 / 62 / 21 / 8; 90 / 32 / 30 / 6; 100 / -33 / 33 / 0 | 50 / 69 / 17 / 12; 94 / 22 / 31 / 3; 100 / -203 / 33 / 0 |
Reading: at the adversary lane's ticket the die's five-year rows are the hybrid's within a few points (a USD 10 M
fleet is 3.4 TH/s at USD 2.94 against 16 TH/s at 0.6), so the die and the hybrid share one verdict and the board
keeps its own. **What moved and why**: the die's whole failure in sections 2, 8 and 12 was its ticket (USD 0.6 per
MH/s of silicon and board, lane B's die at USD 500 for 5.5 GH/s), which made a third of the chain cost a tenth to a
thirtieth of a year's revenue and put a USD 10 M fleet at 70 to 100 percent of the chain; at the adversary lane's
USD 2.94 for the complete machine the same fleet is 16 to 25 percent and conditions (a) and (b) pass. The joules
never decided it (both lanes read the die at 2.3x to 3.3x per joule at the knee, inside condition (e)). The two
tickets: lane B priced the die's silicon, the adversary lane priced a machine (power, cooling, board, the node it
must run, the yield and the packaging). **The coordinator's pin (17:38 UK): the model's die ticket is the adversary
lane's complete-machine figure, USD 2.94, under the plan's evidence standard (complete-board overhead, same-node and
advanced-node results kept separate); the USD 0.6 row stays as the silicon floor, a lower bound and not a machine.**
The served reading until the adversary lane's one-line reconciliation of what its ticket includes (asked by 18:30 UK)
is the range: "a sunk SRAM die fleet of USD 10 M holds 16 to 70 percent of the chain on landing depending on whether
its machine costs USD 3 or USD 0.6 per MH/s, and takes a flat or shrinking chain by year 5 at either".
The success statement at the final rows: holds for the board under every structure in growing and flat paths; holds
for the hybrid and the die (at the adversary lane's ticket) under private supply in a growing chain only, and fails
for both on flat and shrinking paths and under hardware sales or multiple suppliers on those paths; fails for the die
at the chip model's ticket everywhere. The conditions that fail the die at either ticket are (c) and (f), the ticket
and the margin, which the dataset floor moves 1.2x to 2x and the share detector makes visible; nothing in the hash
moves them further.
## 12c. Amendment, 17:2x UK: the die at the adversary lane's RECONCILED ticket, and the range narrowed
The adversary lane's reconciliation (multi-family-adversary.md section 15 at d3ad1549): the two figures priced
different machines. Lane B's USD 0.25 to 0.4 of silicon (0.6 with a board) is the 2 GiB die at zero shadow, 2,100
MH/s per reticle at 300 W with no shadow core; the USD 2.94 is the same die carrying the class v4 shadow on the
placed core (6.55 pJ per lane-op at N5, 13x the die's own energy per hash) with the machine held at 300 W, so it makes
383 MH/s and the fixed tickets divide by 383. Neither is the maker's choice: the power budget is raised until the
power train and the core's silicon bind (the die's own read ceiling, 8.3 GH/s, is never reached). Components (the
die USD 500, the shadow core's silicon USD 0.55 per MH/s on the placed slot or 0.11 to 0.2 pipelined, the package
USD 60 to 200, the board USD 150 plus PSU and cooling at USD 0.15 per W, the host USD 15; claimed and approximate as
the adversary lane labels them): 852 MH/s at 600 W, 1,420 at 1 kW, 2,130 at 1.5 kW, 2,840 at 2 kW; USD 1.63 / 1.20 /
0.98 / 0.88 per MH/s on the placed core, 1.07 / 0.73 / 0.56 / 0.47 pipelined with the organic package. **The
reconciled unit ticket: about USD 1.0 per MH/s (0.5 to 1.6) at 1 to 1.5 kW per machine**; per joule unchanged (2.3x
node-for-node, 3.1x a node ahead). The adversary lane adds that a maker's first batch of 1,000 dies (1 to 2 TH/s,
several times the chain's hash at the rental equilibrium) pays the project (USD 100 M to 500 M) as USD 50 to 250 per
MH/s of NRE against USD 1 of unit cost, plus first-yield losses of USD 100 to 400 per die: the die's economics are
project economics (the surface, section 6), and the unit ticket of a fleet that exists is USD 1.0.
Re-scored at the reconciled ticket and its bounds (`scratchpad/market.py`, build-4, 17:23 UK; sunk, 3 years, at
0.06; all modelled):
| Die ticket | Cost per accepted MH/s-hour, node-for-node / a node ahead | (a) vs the best Blackwell owner at 0.12 (line 1.5x) | (b) hardware over the 5080 entrant's (line 0.25) | (c) a third of the chain at IGN 0.03 / 0.10 / 0.30 / 1.00, short of a year's revenue by | (e) | (f) margin, growing path |
|---|---|---|---|---|---|---|
| **USD 1.0, reconciled** | 128 / 106 | 2.06x / 2.48x: FAILS | 0.09: FAILS | 6.5x / 12.6x / 18x / 28x: FAILS | 2.3x passes; 3.1x on the line | 80 to 88 percent: FAILS |
| USD 0.5, the low bound | 106 / 84 | 2.48x / 3.12x: FAILS | 0.05: FAILS | 13x to 57x: FAILS | the same | 81 to 89: FAILS |
| USD 1.6, the high bound | 153 / 132 | 1.71x / 1.99x: FAILS | 0.15: FAILS | 4.0x to 18x: FAILS | the same | 78 to 85: FAILS |
| USD 2.94 (the 300 W machine, 12b) | 211 / 190 | 1.24x / 1.39x: passed | 0.27: passed by a hair | 2.2x to 9.7x: fails | the same | 72 to 79: fails |
| USD 0.6 (the silicon floor, zero shadow) | 72 | 2.2x to 4.6x: fails | 0.07: fails | 16x: fails | 3.7x to 4.5x at the chip-model energy: fails | 85 to 93: fails |
The five-year run at the reconciled ticket (a sunk USD 10 M fleet, 10 TH/s; share / margin / largest supplier / GPU
classes above water at years 2, 3, 5; the bounds in brackets as the year-2 share):
| Structure | Growing x2 from 0.10 | Flat 0.10 | Shrinking x0.5 from 0.30 |
|---|---|---|---|
| Private supply | 42 / 83 / 42 / 13; 36 / 80 / 36 / 13; 24 / 88 / 24 / 16 (27 to 68 at year 2) | 62 / 77 / 62 / 8; 80 / 65 / 80 / 6; 98 / 53 / 98 / 2 (40 to 93) | 49 / 81 / 49 / 12; 88 / 58 / 88 / 5; 100 / -147 / 100 / 0 (33 to 78) |
| Hardware sales | 42 / 83 / 42 / 13; 60 / 77 / 60 / 12; 70 / 83 / 70 / 13 | 62 / 77 / 62 / 8; 95 / 53 / 95 / 3; 100 / 4 / 100 / 0 | 49 / 81 / 49 / 12; 97 / 47 / 97 / 2; 100 / -69 / 100 / 0 |
| Multiple suppliers | 84 / 76 / 28 / 6; 98 / 57 / 33 / 5; 99 / 65 / 33 / 5 | 99 / 58 / 33 / 2; 100 / 11 / 33 / 0; 100 / -64 / 33 / 0 | 95 / 70 / 32 / 6; 100 / -12 / 33 / 0; 100 / -69 / 33 / 0 |
**The verdict at the reconciled ticket, which replaces 12b's as the die's headline**: the die fails (a), (b), (c)
and (f) at USD 1.0 and at both bounds, and passes only (d), (e) and (g); the 12b pass of (a) and (b) was an artefact
of the 300 W machine (a die throttled to 383 MH/s), and the pinned complete-machine ticket, once reconciled, lands
nearer lane B's floor than the 300 W figure. A sunk USD 10 M fleet holds 42 percent of a growing chain on landing (27
to 68 at the bounds) and 62 percent of a flat one (40 to 93), and takes every flat or shrinking chain by year 3 to 5
under every structure; in a growing chain under private supply it dilutes to 24 percent by year 5 with 16 classes
above water, which is the one cell where the share is a coexistence share while the conditions still fail. **The range
narrows from "16 to 70 percent on landing" to "42 percent of a growing chain and 62 percent of a flat one on landing
(27 to 93 across the reconciled bounds), and 100 percent of a flat or shrinking chain by year 3 to 5"; the die's
success statement does not hold at any reconciled ticket, and the only condition that holds the die stays the
project economics of section 6** (the adversary lane's NRE of USD 50 to 250 per MH/s against USD 1 of unit cost says
the same thing from the maker's side). The hybrid's verdict (12a, at its own reconciled USD 2.80) and the board's
(six of seven at USD 4.84) are unchanged.
**The surface row beside it (the die's economics are project economics).** The operator self-mining, `T0` 2 years,
edge 2.3x, the unit ticket USD 1.0 per MH/s (the 0.5 and 1.6 ends in brackets), the first batch's yield loss (1,000
dies at USD 100 to 400 each, USD 0.6 to 0.9 M) added to `C_dev` and moving `p*` under 1 percent; `p*` in USD per IGN
at `L` 1 / 3 years and `q` 0.3 / 1.0 (build-4, 17:25 UK; modelled):
| `C_dev` | `L` 1 y, `q` 0.3 | `L` 1 y, `q` 1.0 | `L` 3 y, `q` 0.3 | `L` 3 y, `q` 1.0 | NRE per MH/s on a 1,000-die first batch (1.42 to 2.13 TH/s at 1 to 1.5 kW) |
|---|---|---|---|---|---|
| USD 100 M | 1.28 (1.19 to 1.40) | 0.38 (0.36 to 0.42) | 0.50 (0.49 to 0.51) | 0.15 (0.15 to 0.15) | USD 47 to 70 |
| USD 250 M | 3.17 (2.95 to 3.48) | 0.95 (0.89 to 1.05) | 1.24 (1.22 to 1.28) | 0.37 (0.37 to 0.38) | USD 117 to 176 |
| USD 500 M | 6.33 (5.90 to 6.95) | 1.90 (1.77 to 2.09) | 2.48 (2.43 to 2.55) | 0.75 (0.73 to 0.77) | USD 235 to 352 |
Reading: the unit ticket's whole band (0.5 to 1.6) moves `p*` by under 10 percent at a 1-year life and under 3
percent at 3 years; the project cost moves it 5x and the share 3x. The first batch's NRE of USD 47 to 352 per MH/s
against a unit cost of USD 1 is the surface's statement from the maker's side: a die project is attractive only where
the price its maker expects over the life exceeds IGN 0.15 to 0.75 taking the whole chain over three years, or 0.50
to 2.48 at a third, and it is never the unit ticket that decides. The coexistence verdict (the die fails once built)
and the surface verdict (it is built only above those prices) are the two halves of the same row; the chain's
instruments on the first are the honest side's efficiency and the share detector, on the second nothing in the hash.
## 13. The D4 checklist, line by line
| Item | Where | Status |
@ -343,6 +675,9 @@ to 3,000 per die at the schedule, which moves (c) by 2x to 4x and nothing else).
| The outputs: cost advantage, replacement economics, accessible supply, break-even electricity per class (17), supplier and operator dependence as its own table | sections 1, 4, 11, 3, 11 | in |
| The tariff advantage shown separately from the hardware advantage, the 6.25x illustration first | section 1 | in |
| The operator simulation | section 10 and its own file | in (first run) |
| A third chip, the adversary lane's hybrid board, on the conditions and the surface | section 12a (amendment, 17:0x UK) | in |
| Market structure as a scenario axis (private supply, hardware sales, multiple suppliers, concentration), a row per structure for each of the three chips, at the adversary lane's placed energies | section 8a (amendment, 17:1x UK) | in |
| The sensitivity workbook: every threshold with its assumptions in columns, never served as a boundary | `docs/analysis/class-v6/coexistence-workbook.md` (171 rows, 17:1x UK) | in |
| The k lane's placed rows and the adversary lane's whole-machine rows | the chip rows stand until they land | owed by others |
| Per-card agents, the hybrid mode, the measured stage times in the simulation | the simulation's section 4 | second cut |

201
docs/analysis/class-v6/coexistence-workbook.md Normal file
View file

@ -0,0 +1,201 @@
# The coexistence sensitivity workbook: every threshold figure with the assumptions that produce it
8 October 2026, 17:1x UK, branch `class-v6-floor-sram`, floor lane 3, Igneum 2.0 D4 (the research lane's pin of 17:xx UK). Every row here is a figure produced today by the floor file (`docs/analysis/class-v6/floor/sram-and-floor.md`), the coexistence model (`docs/analysis/class-v6/coexistence-model.md`) or this lane's scripts, with the assumption that produces it in its own column, so that a reader can change one column and see the figure move. **Nothing in this table is a boundary the chain stays below; every figure is modelled unless its label says otherwise, and the model document cites this workbook instead of stating a threshold in prose.** Generated by `scratchpad/market.py` on build-4 (the rows, 17:07 UK) from the same code that produced the documents' tables; the condition lines are judgement lines, labelled as such. The founder is not named.
## How to read a row
`figure` is the quantity; `value` and `unit` its number; the next columns are the assumptions that produce it (`n/a` where the figure does not depend on that assumption); `GPU_cost_basis` says whether the GPU side is priced at MSRP or street and which card and operating point; `emission_years` says which years of the spec's constant the figure integrates; `price_assumption` carries an IGN price where one is assumed and `solved` where the price is the output; `source_section` is where the figure is used; `label` is measured, modelled, approximate or a judgement line.
The standing assumptions behind every surface row unless the row says otherwise: the spec's emission (0.77 / 0.80 / 0.40 / 0.40 / 0.20 / 0.20 / 0.10 / 0.10 B IGN to miners in years 1 to 8), the project starting at launch and mining for `life_y` years after `pre_production_y`, mid-year discounting, the chip's margin `1 - (0.000084 / edge + capex / (life x 8,766)) / 0.00092` with the 5090 at the lock over two years at MSRP as the GPU all-in cost (USD 0.00092 per MH/s-hour, 0.000084 of it electricity), the fleet capital and residual dropped (under 2 percent of revenue). The break-even electricity rows assume 97 percent accepted work, a 1 percent pool fee, a 5 percent wear allowance on the used price for the owner, a two-year life with the table's resale for the entrant, and the chip at USD 0.06 per kWh plus USD 0.02 of hosting with development sunk.
## The rows
| figure | value | unit | discount | life_y | share_q | project_cost_USD_M | pre_production_y | edge_per_joule | chip_capex_USD_per_MHs | GPU_cost_basis | electricity_USD_kWh | emission_years | price_assumption | source_section | label |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| lower threshold: no chip project below | 23 | USD M a year of miner revenue at launch | 0.10 | 4 (years 3 to 6) | 1.0 | 20 | 2 | 5x | n/a (margin 1 - 1/e) | n/a | n/a | y3 to y6 at the constant | p* 0.029 USD per IGN | floor 4.2 / 4.3 | modelled |
| upper threshold: every DRAM-board project above | 340 | USD M a year of miner revenue at launch | 0.10 | 4 (years 3 to 6) | 0.3 | 75 | 2 | 3x | n/a (margin 1 - 1/e) | n/a | n/a | y3 to y6 at the constant | p* 0.441 USD per IGN | floor 4.2 / 4.3 | modelled |
| SRAM project taking the chain | 98 | USD M a year of miner revenue at launch | 0.10 | 4 (years 3 to 6) | 1.0 | 100 | 2 | 13x | n/a (margin 1 - 1/e) | n/a | n/a | y3 to y6 at the constant | p* 0.127 USD per IGN | floor 4.2 / 4.3 | modelled |
| SRAM project at a third | 566 | USD M a year of miner revenue at launch | 0.10 | 4 (years 3 to 6) | 0.3 | 150 | 2 | 5x | n/a (margin 1 - 1/e) | n/a | n/a | y3 to y6 at the constant | p* 0.735 USD per IGN | floor 4.2 / 4.3 | modelled |
| development cost range, cheapest DRAM-board chip | 20 to 75 | USD M | n/a | n/a | n/a | 20 to 75 | n/a | n/a | n/a | n/a | n/a | n/a | n/a | floor 4.3 (floor lane 5: no 28 nm GDDR7 PHY) | approximate |
| surface p*, operator self-mining | 0.224 | USD per IGN | 0.10 | 0.5 | 0.3 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.067 | USD per IGN | 0.10 | 0.5 | 1.0 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.105 | USD per IGN | 0.10 | 1 | 0.3 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.031 | USD per IGN | 0.10 | 1 | 1.0 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.070 | USD per IGN | 0.10 | 2 | 0.3 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.021 | USD per IGN | 0.10 | 2 | 1.0 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.054 | USD per IGN | 0.10 | 3 | 0.3 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.016 | USD per IGN | 0.10 | 3 | 1.0 | 20 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.493 | USD per IGN | 0.10 | 0.5 | 0.3 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.148 | USD per IGN | 0.10 | 0.5 | 1.0 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.230 | USD per IGN | 0.10 | 1 | 0.3 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.069 | USD per IGN | 0.10 | 1 | 1.0 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.117 | USD per IGN | 0.10 | 2 | 0.3 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.035 | USD per IGN | 0.10 | 2 | 1.0 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.095 | USD per IGN | 0.10 | 3 | 0.3 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.028 | USD per IGN | 0.10 | 3 | 1.0 | 20 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.841 | USD per IGN | 0.10 | 0.5 | 0.3 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.252 | USD per IGN | 0.10 | 0.5 | 1.0 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.392 | USD per IGN | 0.10 | 1 | 0.3 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.118 | USD per IGN | 0.10 | 1 | 1.0 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.261 | USD per IGN | 0.10 | 2 | 0.3 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.078 | USD per IGN | 0.10 | 2 | 1.0 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.201 | USD per IGN | 0.10 | 3 | 0.3 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.060 | USD per IGN | 0.10 | 3 | 1.0 | 75 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.849 | USD per IGN | 0.10 | 0.5 | 0.3 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.555 | USD per IGN | 0.10 | 0.5 | 1.0 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.862 | USD per IGN | 0.10 | 1 | 0.3 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.259 | USD per IGN | 0.10 | 1 | 1.0 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.437 | USD per IGN | 0.10 | 2 | 0.3 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.131 | USD per IGN | 0.10 | 2 | 1.0 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.355 | USD per IGN | 0.10 | 3 | 0.3 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.107 | USD per IGN | 0.10 | 3 | 1.0 | 75 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.681 | USD per IGN | 0.10 | 0.5 | 0.3 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.504 | USD per IGN | 0.10 | 0.5 | 1.0 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.784 | USD per IGN | 0.10 | 1 | 0.3 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.235 | USD per IGN | 0.10 | 1 | 1.0 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.521 | USD per IGN | 0.10 | 2 | 0.3 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.156 | USD per IGN | 0.10 | 2 | 1.0 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.402 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.121 | USD per IGN | 0.10 | 3 | 1.0 | 150 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 3.699 | USD per IGN | 0.10 | 0.5 | 0.3 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.110 | USD per IGN | 0.10 | 0.5 | 1.0 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.725 | USD per IGN | 0.10 | 1 | 0.3 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.517 | USD per IGN | 0.10 | 1 | 1.0 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.874 | USD per IGN | 0.10 | 2 | 0.3 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.262 | USD per IGN | 0.10 | 2 | 1.0 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.711 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.213 | USD per IGN | 0.10 | 3 | 1.0 | 150 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 5.604 | USD per IGN | 0.10 | 0.5 | 0.3 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.681 | USD per IGN | 0.10 | 0.5 | 1.0 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 2.613 | USD per IGN | 0.10 | 1 | 0.3 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.784 | USD per IGN | 0.10 | 1 | 1.0 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.738 | USD per IGN | 0.10 | 2 | 0.3 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.521 | USD per IGN | 0.10 | 2 | 1.0 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.339 | USD per IGN | 0.10 | 3 | 0.3 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.402 | USD per IGN | 0.10 | 3 | 1.0 | 500 | 1 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 12.329 | USD per IGN | 0.10 | 0.5 | 0.3 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 3.699 | USD per IGN | 0.10 | 0.5 | 1.0 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 5.749 | USD per IGN | 0.10 | 1 | 0.3 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 1.725 | USD per IGN | 0.10 | 1 | 1.0 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 2.913 | USD per IGN | 0.10 | 2 | 0.3 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.874 | USD per IGN | 0.10 | 2 | 1.0 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 2.369 | USD per IGN | 0.10 | 3 | 0.3 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, operator self-mining | 0.711 | USD per IGN | 0.10 | 3 | 1.0 | 500 | 2 | 5x | 0.5 | 5090 at the lock over 2 y, MSRP (0.00092 per MH/s-h) | 0.06 chip | T0+1 onward at the constant | solved | floor 4.4 A | modelled |
| surface p*, edge axis | 0.732 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 2x | 0.5 | MSRP basis | 0.06 | constant | solved | floor 4.4 A2 | modelled |
| surface p*, edge axis | 0.720 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 3x | 0.5 | MSRP basis | 0.06 | constant | solved | floor 4.4 A2 | modelled |
| surface p*, edge axis | 0.703 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 13x | 0.5 | MSRP basis | 0.06 | constant | solved | floor 4.4 A2 | modelled |
| surface p*, chip capex axis | 0.789 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 5x | 2.8 | MSRP basis | 0.06 | constant | solved | floor 4.4 A2 | modelled |
| surface p*, chip capex axis | 0.881 | USD per IGN | 0.10 | 3 | 0.3 | 150 | 2 | 5x | 5.0 | MSRP basis | 0.06 | constant | solved | floor 4.4 A2 | modelled |
| surface p*, discount axis | 0.611 | USD per IGN | 0.05 | 3 | 0.3 | 150 | 2 | 5x | 0.5 | MSRP basis | 0.06 | constant | solved | this workbook | modelled |
| surface p*, discount axis | 0.821 | USD per IGN | 0.15 | 3 | 0.3 | 150 | 2 | 5x | 0.5 | MSRP basis | 0.06 | constant | solved | this workbook | modelled |
| surface p*, street GPU prices (capex term x2: GPU all-in 0.00176) | 0.372 | USD per IGN (approximate: p* scales with the GPU all-in cost) | 0.10 | 3 | 0.3 | 150 | 2 | 5x | 0.5 | street (2x MSRP) | 0.06 | constant | solved | this workbook | approximate |
| surface p*, the hybrid board | 0.797 | USD per IGN | 0.10 | 1 | 0.3 | 50 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.239 | USD per IGN | 0.10 | 1 | 1.0 | 50 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.263 | USD per IGN | 0.10 | 3 | 0.3 | 50 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.079 | USD per IGN | 0.10 | 3 | 1.0 | 50 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 1.992 | USD per IGN | 0.10 | 1 | 0.3 | 125 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.598 | USD per IGN | 0.10 | 1 | 1.0 | 125 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.658 | USD per IGN | 0.10 | 3 | 0.3 | 125 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.197 | USD per IGN | 0.10 | 3 | 1.0 | 125 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 3.586 | USD per IGN | 0.10 | 1 | 0.3 | 225 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 1.076 | USD per IGN | 0.10 | 1 | 1.0 | 225 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 1.184 | USD per IGN | 0.10 | 3 | 0.3 | 225 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| surface p*, the hybrid board | 0.355 | USD per IGN | 0.10 | 3 | 1.0 | 225 | 2 | 2.7x | 2.44 | MSRP basis | 0.06 | constant | solved | model 12a | modelled |
| break-even electricity, RTX 5090 vs GDDR7 board 1 y, owner / entrant | 26.9 / 9.7 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.9x joules | 5.59 | measured; used price 2200 USD, new 2600, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5090 vs GDDR7 board 3 y, owner / entrant | 8.7 / -8.6 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.9x joules | 5.59 | measured; used price 2200 USD, new 2600, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5090 vs SRAM die 1 y, owner / entrant | 1.5 / -15.7 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 5.1x joules | 0.78 | measured; used price 2200 USD, new 2600, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5090 vs SRAM die 3 y, owner / entrant | -1.0 / -18.3 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 5.1x joules | 0.78 | measured; used price 2200 USD, new 2600, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5080 vs GDDR7 board 1 y, owner / entrant | 31.5 / 13.6 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.6x joules | 5.59 | measured; used price 900 USD, new 1100, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5080 vs GDDR7 board 3 y, owner / entrant | 10.8 / -7.1 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.6x joules | 5.59 | measured; used price 900 USD, new 1100, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5080 vs SRAM die 1 y, owner / entrant | 2.7 / -15.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.5x joules | 0.78 | measured; used price 900 USD, new 1100, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5080 vs SRAM die 3 y, owner / entrant | -0.1 / -18.0 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.5x joules | 0.78 | measured; used price 900 USD, new 1100, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 Ti vs GDDR7 board 1 y, owner / entrant | 39.5 / 24.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.2x joules | 5.59 | modelled knee; used price 650 USD, new 800, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 Ti vs GDDR7 board 3 y, owner / entrant | 14.5 / -0.1 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.2x joules | 5.59 | modelled knee; used price 650 USD, new 800, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 Ti vs SRAM die 1 y, owner / entrant | 4.7 / -9.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 3.7x joules | 0.78 | modelled knee; used price 650 USD, new 800, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 Ti vs SRAM die 3 y, owner / entrant | 1.2 / -13.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 3.7x joules | 0.78 | modelled knee; used price 650 USD, new 800, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 vs GDDR7 board 1 y, owner / entrant | 37.6 / 18.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.2x joules | 5.59 | modelled knee; used price 450 USD, new 560, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 vs GDDR7 board 3 y, owner / entrant | 13.3 / -5.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.2x joules | 5.59 | modelled knee; used price 450 USD, new 560, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 vs SRAM die 1 y, owner / entrant | 3.8 / -14.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 3.8x joules | 0.78 | modelled knee; used price 450 USD, new 560, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5070 vs SRAM die 3 y, owner / entrant | 0.4 / -18.3 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 3.8x joules | 0.78 | modelled knee; used price 450 USD, new 560, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 Ti 16G vs GDDR7 board 1 y, owner / entrant | 25.9 / -1.0 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 3.0x joules | 5.59 | modelled knee; used price 360 USD, new 450, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 Ti 16G vs GDDR7 board 3 y, owner / entrant | 7.9 / -19.0 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 3.0x joules | 5.59 | modelled knee; used price 360 USD, new 450, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 Ti 16G vs SRAM die 1 y, owner / entrant | 0.9 / -26.0 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 5.1x joules | 0.78 | modelled knee; used price 360 USD, new 450, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 Ti 16G vs SRAM die 3 y, owner / entrant | -1.7 / -28.6 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 5.1x joules | 0.78 | modelled knee; used price 360 USD, new 450, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 vs GDDR7 board 1 y, owner / entrant | 28.8 / 6.7 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.8x joules | 5.59 | modelled knee; used price 250 USD, new 310, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 vs GDDR7 board 3 y, owner / entrant | 9.5 / -12.5 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.8x joules | 5.59 | modelled knee; used price 250 USD, new 310, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 vs SRAM die 1 y, owner / entrant | 2.0 / -20.1 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.8x joules | 0.78 | modelled knee; used price 250 USD, new 310, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 5060 vs SRAM die 3 y, owner / entrant | -0.7 / -22.8 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.8x joules | 0.78 | modelled knee; used price 250 USD, new 310, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4090 vs GDDR7 board 1 y, owner / entrant | 16.5 / -5.6 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.5x joules | 5.59 | modelled knee; used price 1300 USD, new 1700, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4090 vs GDDR7 board 3 y, owner / entrant | 4.7 / -17.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.5x joules | 5.59 | modelled knee; used price 1300 USD, new 1700, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4090 vs SRAM die 1 y, owner / entrant | 0.0 / -22.1 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 7.8x joules | 0.78 | modelled knee; used price 1300 USD, new 1700, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4090 vs SRAM die 3 y, owner / entrant | -1.6 / -23.8 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 7.8x joules | 0.78 | modelled knee; used price 1300 USD, new 1700, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4080 vs GDDR7 board 1 y, owner / entrant | 18.0 / -1.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.4x joules | 5.59 | modelled knee; used price 700 USD, new 1000, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4080 vs GDDR7 board 3 y, owner / entrant | 5.9 / -13.3 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.4x joules | 5.59 | modelled knee; used price 700 USD, new 1000, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4080 vs SRAM die 1 y, owner / entrant | 1.1 / -18.0 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 7.6x joules | 0.78 | modelled knee; used price 700 USD, new 1000, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4080 vs SRAM die 3 y, owner / entrant | -0.6 / -19.7 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 7.6x joules | 0.78 | modelled knee; used price 700 USD, new 1000, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4070 vs GDDR7 board 1 y, owner / entrant | 18.1 / 3.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.5x joules | 5.59 | measured; used price 400 USD, new 550, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4070 vs GDDR7 board 3 y, owner / entrant | 6.2 / -8.7 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.5x joules | 5.59 | measured; used price 400 USD, new 550, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4070 vs SRAM die 1 y, owner / entrant | 1.5 / -13.3 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 7.8x joules | 0.78 | measured; used price 400 USD, new 550, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4070 vs SRAM die 3 y, owner / entrant | -0.1 / -15.0 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 7.8x joules | 0.78 | measured; used price 400 USD, new 550, resale 0.4 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4060 Ti 16G vs GDDR7 board 1 y, owner / entrant | 16.4 / -4.3 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.8x joules | 5.59 | modelled knee; used price 290 USD, new 420, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4060 Ti 16G vs GDDR7 board 3 y, owner / entrant | 5.3 / -15.5 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.8x joules | 5.59 | modelled knee; used price 290 USD, new 420, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4060 Ti 16G vs SRAM die 1 y, owner / entrant | 0.9 / -19.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 8.3x joules | 0.78 | modelled knee; used price 290 USD, new 420, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 4060 Ti 16G vs SRAM die 3 y, owner / entrant | -0.7 / -21.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 8.3x joules | 0.78 | modelled knee; used price 290 USD, new 420, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3090 (used) vs GDDR7 board 1 y, owner / entrant | 13.8 / -5.1 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 5.7x joules | 5.59 | modelled cap; used price 650 USD, new 900, resale 0.3 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3090 (used) vs GDDR7 board 3 y, owner / entrant | 4.4 / -14.5 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 5.7x joules | 5.59 | modelled cap; used price 650 USD, new 900, resale 0.3 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3090 (used) vs SRAM die 1 y, owner / entrant | 0.7 / -18.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 9.8x joules | 0.78 | modelled cap; used price 650 USD, new 900, resale 0.3 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3090 (used) vs SRAM die 3 y, owner / entrant | -0.6 / -19.5 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 9.8x joules | 0.78 | modelled cap; used price 650 USD, new 900, resale 0.3 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3080 (used) vs GDDR7 board 1 y, owner / entrant | 16.0 / 5.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 5.3x joules | 5.59 | modelled cap; used price 330 USD, new 450, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3080 (used) vs GDDR7 board 3 y, owner / entrant | 5.9 / -4.2 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 5.3x joules | 5.59 | modelled cap; used price 330 USD, new 450, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3080 (used) vs SRAM die 1 y, owner / entrant | 2.0 / -8.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 9.1x joules | 0.78 | modelled cap; used price 330 USD, new 450, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3080 (used) vs SRAM die 3 y, owner / entrant | 0.5 / -9.6 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 9.1x joules | 0.78 | modelled cap; used price 330 USD, new 450, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3060 (used) vs GDDR7 board 1 y, owner / entrant | 11.7 / 4.4 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 7.3x joules | 5.59 | modelled cap; used price 190 USD, new 260, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3060 (used) vs GDDR7 board 3 y, owner / entrant | 4.3 / -3.0 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 7.3x joules | 5.59 | modelled cap; used price 190 USD, new 260, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3060 (used) vs SRAM die 1 y, owner / entrant | 1.4 / -5.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 12.5x joules | 0.78 | modelled cap; used price 190 USD, new 260, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RTX 3060 (used) vs SRAM die 3 y, owner / entrant | 0.4 / -6.9 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 12.5x joules | 0.78 | modelled cap; used price 190 USD, new 260, resale 0.25 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RX 9070 XT vs GDDR7 board 1 y, owner / entrant | 7.1 / -4.5 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 10.0x joules | 5.59 | measured; used price 520 USD, new 650, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RX 9070 XT vs GDDR7 board 3 y, owner / entrant | 1.7 / -9.9 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 10.0x joules | 5.59 | measured; used price 520 USD, new 650, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RX 9070 XT vs SRAM die 1 y, owner / entrant | -0.4 / -12.0 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 17.2x joules | 0.78 | measured; used price 520 USD, new 650, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, RX 9070 XT vs SRAM die 3 y, owner / entrant | -1.1 / -12.8 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 17.2x joules | 0.78 | measured; used price 520 USD, new 650, resale 0.45 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, H100 (hosted) vs GDDR7 board 1 y, owner / entrant | -21.0 / -360.1 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 2.5x joules | 5.59 | modelled lock; used price 18000 USD, new 25000, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, H100 (hosted) vs GDDR7 board 3 y, owner / entrant | -42.3 / -381.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 2.5x joules | 5.59 | modelled lock; used price 18000 USD, new 25000, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, H100 (hosted) vs SRAM die 1 y, owner / entrant | -50.6 / -389.7 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 4.3x joules | 0.78 | modelled lock; used price 18000 USD, new 25000, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, H100 (hosted) vs SRAM die 3 y, owner / entrant | -53.6 / -392.6 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 4.3x joules | 0.78 | modelled lock; used price 18000 USD, new 25000, resale 0.5 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, A100 (used) vs GDDR7 board 1 y, owner / entrant | 5.2 / -188.9 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 3.7x joules | 5.59 | modelled; used price 6000 USD, new 10000, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, A100 (used) vs GDDR7 board 3 y, owner / entrant | -9.5 / -203.6 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 3.7x joules | 5.59 | modelled; used price 6000 USD, new 10000, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, A100 (used) vs SRAM die 1 y, owner / entrant | -15.3 / -209.4 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 6.3x joules | 0.78 | modelled; used price 6000 USD, new 10000, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, A100 (used) vs SRAM die 3 y, owner / entrant | -17.3 / -211.4 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 6.3x joules | 0.78 | modelled; used price 6000 USD, new 10000, resale 0.35 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, Apple M5 Max vs GDDR7 board 1 y, owner / entrant | 3.3 / -219.2 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 1.8x joules | 5.59 | measured (reported); used price 3200 USD, new 4000, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, Apple M5 Max vs GDDR7 board 3 y, owner / entrant | -27.0 / -249.5 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 1.8x joules | 5.59 | measured (reported); used price 3200 USD, new 4000, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, Apple M5 Max vs SRAM die 1 y, owner / entrant | -38.9 / -261.4 | cents per kWh | n/a | 1 | n/a | sunk | n/a | 3.0x joules | 0.78 | measured (reported); used price 3200 USD, new 4000, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| break-even electricity, Apple M5 Max vs SRAM die 3 y, owner / entrant | -43.2 / -265.7 | cents per kWh | n/a | 3 | n/a | sunk | n/a | 3.0x joules | 0.78 | measured (reported); used price 3200 USD, new 4000, resale 0.55 | chip at 0.06 + 0.02 hosting | n/a | n/a | model 3 | modelled |
| condition (a) line | 1.5 | x the best GPU owner's cost at the GPU's electricity | n/a | n/a | n/a | n/a | n/a | n/a | n/a | owner at 0.12: 263 to 332 uUSD | 0.12 GPU / 0.06 chip | n/a | n/a | model 12 | a judgement line, not a measurement |
| condition (b) line | 0.25 | of the GPU entrant's annualised hardware | n/a | 2 (GPU) | n/a | n/a | n/a | n/a | n/a | 5080 entrant hardware 459 uUSD | n/a | n/a | n/a | model 12 | a judgement line |
| condition (c) line | 1.0 | years of miner revenue to buy a third of the chain's hash | n/a | n/a | n/a | n/a | n/a | n/a | per chip | GPU equilibrium hash at the per-class curve | 0.12 | year 3 | 0.03 to 1.00 | model 12 | a judgement line |
| condition (e) line | 3.0 | x per joule at the honest knee | n/a | n/a | n/a | n/a | n/a | n/a | n/a | Blackwell knee 1.70 to 2.06 uJ (5070 Ti modelled, 5080 measured) | n/a | n/a | n/a | model 12 | a judgement line |
| condition (f) line | 0.70 | gross margin, falling with the halvings | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | model 12 | a judgement line |
## Rows carried from the first-cut documents that the script does not regenerate
| figure | value | unit | assumptions | source_section | label |
|---|---|---|---|---|---|
| The record's SRAM read energy | 1.0 (0.5 to 2.0) | nJ per 64-byte read | the wire at 1.3 pJ per bit across a 24 mm die, a 64-byte atom | floor 2.1 | approximate (lane B) |
| The die's read energy at the hash's width | 0.25 / 0.38 / 0.55 / 0.88 | nJ at W = 1 / 4 / 8 / 16 | the same wire, 0.1 nJ macro, 0.05 controller, 48 + 32 W bits per read | floor 2.2 | modelled |
| The SRAM cost curve | about 250 | USD per GiB of store, flat to 2031 | N2 38 Mb/mm^2 (claimed), USD 500 per reticle die, +8 percent density and +10 percent wafer price per node | floor 3.1 | approximate |
| USD 5,000 of silicon per store | 20 GiB (10 reticles) at N2, about 21 GiB in 2031 | the floor size that costs it | the cost curve above | floor 3.2 | modelled |
| The dataset floor's honest-side cost | 4 / 8 / 10 percent per hash at 2 / 4 / 8 GiB at the 5090's knee; 3.5 percent of rate at 5.5 GiB at stock | the 5090 at the 1,300 MHz lock; a rented 5090 at stock (host capped at 328 W) | the hash lane's and the fleet lane's rows | floor 3.2 | measured |
| The W = 8 energy cost on the card | +4.8 percent per hash at stock | the 5090 under the class v5 state term, PC 1 | one sector per load, the second sector 1.2 nJ | floor 2.2 | measured |
| The W = 16 energy cost on the card | +25 percent (5090), +34 percent (4090) per hash at stock | the hinted one-request form, rate held | floor lane 5's rented rows | floor 2.2c | measured |
| The k lane's programmable core | 3.5 pJ per lane-op at N3, 2.5 at N2 | synthesis-only, pre-place; x0.5 ASAP7 to N3 | 0.36 microjoules of class v4 shadow at N3; absolute k 0.31 stock, 0.56 lock | floor 2.2b | synthesised (the scaling approximate) |
| The adversary lane's placed whole-machine energies (first) | board 1.6x / 1.9x, hybrid 2.4x / 2.9x, die 2.4x / 3.3x | per joule vs the 5090 at the lock, node-for-node / a node ahead | the design's 10.0r | model 8a | the adversary lane's rows |
| The adversary lane's FINAL placed energies and tickets | board 1.5x / 1.8x at USD 4.84 per MH/s; hybrid 1.93x / 2.40x at 2.80 (p98 2.09x at 1.88); die 2.3x / 3.1x at 2.94 (2.14 at the smaller ticket) | complete machines | the design's 10.0t | model 12b | the adversary lane's rows |
| The die's ticket, reconciled (the adversary lane's section 15, 17:2x UK) | **1.0 (0.5 to 1.6)** per MH/s at 1 to 1.5 kW per machine: the die USD 500, the shadow core's silicon 0.55 per MH/s placed or 0.11 to 0.2 pipelined, the package 60 to 200, the board 150 plus PSU and cooling at 0.15 per W, the host 15; 852 / 1,420 / 2,130 / 2,840 MH/s at 600 W / 1 / 1.5 / 2 kW. The earlier 2.94 was a 300 W machine (the die throttled to 383 MH/s); the 0.6 is the zero-shadow silicon floor | USD per MH/s | the power budget (the maker raises it until the power train and the core's silicon bind); the core's pipelining; the package | model 12c | reconciled; the components claimed and approximate as the adversary lane labels them |
| The die's cost per accepted MH/s-hour at the reconciled ticket | 128 / 106 (node-for-node / a node ahead); 106 / 84 at USD 0.5; 153 / 132 at 1.6 | micro-USD, sunk, 3 y, at 0.06 | the ticket | model 12c | modelled |
| Conditions (a), (b), (c) on the die at the reconciled ticket | (a) 2.06x / 2.48x (fails); (b) 0.09 (fails); (c) 12.6x short at IGN 0.10 (fails) | x the best owner; of the 5080 entrant's hardware; a third of the chain against a year's revenue | the ticket | model 12c | modelled |
| A sunk USD 10 M die fleet's share on landing at the reconciled ticket | 42 percent of a growing chain (27 to 68 at the bounds), 62 of a flat one (40 to 93), 49 of a shrinking one (33 to 78); 100 percent of flat and shrinking chains by year 3 to 5 | percent of the hash | the ticket; private supply | model 12c | modelled |
| The die's project economics (the adversary lane) | NRE USD 50 to 250 per MH/s for a first batch of 1,000 dies (1 to 2 TH/s) against USD 1 of unit cost; first-yield losses USD 100 to 400 per die | USD per MH/s; USD per die | the project cost 100 to 500 M; the batch size | model 12c and the surface (floor 4.4) | approximate |
| The die's cost per accepted MH/s-hour at each ticket | 72 at USD 0.6; 211 / 190 at 2.94 (node-for-node / a node ahead); 177 / 155 at 2.14 | micro-USD, sunk, 3 y, at 0.06 | the ticket | model 12b | modelled |
| Condition (a) on the die at each ticket | 2.2x to 4.6x at USD 0.6 (fails); 1.24x / 1.39x at 2.94 (passes); 1.49x / 1.69x at 2.14 (on the line) | x the best Blackwell owner at 0.12 | the ticket | model 12b | modelled |
| Condition (b) on the die at each ticket | 0.07 at USD 0.6 (fails); 0.27 at 2.94 (passes by a hair); 0.20 at 2.14 (fails) | of the 5080 entrant's hardware | the ticket | model 12b | modelled |
| Condition (c) on the die at each ticket, IGN 0.10 | USD 2.3 M for a third of the chain at 0.6 (16x short of a year's revenue); 9.4 M at 2.94 (4.3x); 6.8 M at 2.14 (5.9x) | USD M against USD 40 M | the ticket; the per-class equilibrium hash | model 12b | modelled |
| A sunk USD 10 M die fleet's share of the chain on landing, year 2 | 70 percent at USD 0.6 (flat path), 50 (growing); 25 and 16 at 2.94 | percent of the hash | the ticket; private supply | model 8 and 12b | modelled |

211
docs/analysis/class-v6/eco-05-results.md Normal file
View file

@ -0,0 +1,211 @@
# ECO-05 results: the full cube on the frozen register, the boundaries, and the failed worlds explained
Case ECO-05 "Stress success, contraction and cheap electricity", registry row `docs/plans/igneum-2.0-test-registry.json`
suite ECO, gate G4, profile P12, GATE; this run moves its status from NOT RUN to RUN, VERDICT FAIL (the envelope is
not met in the sustainable worlds as the register froze them; the failure regions are explicit; nothing was weakened
to pass). 8 October 2026, 18:3x to 18:5x UK, branch `class-v6-floor-sram`, floor lane 3. The register
(`eco-05-scenarios.md`, frozen at ece460a5, 18:37 UK) is cited and unchanged. **Every row is modelled** on the labelled
inputs of `sim/economy/coexist/inputs/`; the driver is `sim/economy/coexist/eco05.py`, run on build-4 at 18:39 UK at
the assumed proving efficiency (0.5) and at the measured day's (0.055); the full cubes (2,592 cells each) are
`sim/economy/coexist/out/eco05-cube-eff050.tsv` and `eco05-cube-eff0055.tsv`, every cell published. The founder is
not named.
## 0. The verdict in one page
1. **Sustainability, as found (the register's section 5 expectation contradicted and reported, not revised).** Under
the register's rule (three purchasable configurations across two vendors with positive new-entry economics at MSRP,
and 75 percent of the entry cohort with positive marginal operating economics, at the GPU side's own equilibrium)
**only one of the sixteen revenue-tariff worlds is sustainable: 10R at USD 0.03 per kWh** (17 classes with positive
entry across three vendors, 100 percent operating). Every other world fails the two-vendor test: at the equilibrium
set by NVIDIA's cards (revenue per MH/s-hour 321 to 2,301 micro-USD), no AMD or Intel card has positive new-entry
economics at MSRP, because on this hash the measured RX 9070 XT (7.9 microjoules at its knee), the RX 7600 (6.2
estimated, 8.1 measured at stock) and the Arc B580 (10.4, watts estimated) cost 4x to 6x a Blackwell card per joule
and their entrants 1,400 to 2,100 micro-USD against an equilibrium under 1,100 at every world below 10R. The
operating test passes in 30 of 48 world-demand cells (68 to 100 percent of the cohort above water); the 0.25R worlds
at 0.25 and 0.40 and the R and 4R worlds at 0.40 without proving demand fall to 39 to 46 percent (collapse worlds,
retained as such). **This is a failure region of the chain's own commodity participation, independent of any
specialist: across vendors, sustained new entry on this hash exists only at the cheapest tariff and the largest
revenue.** The expected declaration in the register's section 5 (sustainable at R and above for 0.03 to 0.25) was
wrong on the two-vendor count and is reported so.
2. **The envelope in the sustainable worlds: no specialist passes.** In the 54 cells of the one sustainable world
(10R, 0.03; three demand states x three lives x three development cases x two business models), per specialist:
| Specialist | Cells passing the median (1.5) | The p90 (1.75) | The core (2.0) | All three | Median range | The worst core cell |
|---|---|---|---|---|---|---|
| N2 SRAM die (USD 1.0, 2.3x) | 6 of 54 (development USD 75 M charged, life 1 only) | 0 | 0 | **0** | 1.03 to 11.7 | 23.6 (the used RTX 3090, life 5, zero development) |
| Stored-half hybrid (2.80, 1.93x) | 12 | 0 | 6 | **0** | 0.69 to 6.8 | 13.6 |
| GDDR7 machine (4.84, 1.5x) | 21 | 0 | 12 | **0** | 0.50 to 4.4 | 8.9 |
The p90 fails everywhere because the cohort's dearest entrants (the hosted H100 at 8,500 micro-USD, the used A100,
the Mac, the used 3090) sit 9x to 50x above any specialist; the consumer-16 cut (the register's cohort less the
three that never enter) lowers the medians by 5 to 10 percent and changes no verdict.
3. **The boundaries (life 3, development zero, private mining, normal demand; the ratio does not move with revenue
at zero development because neither side's cost does):**
| Specialist | Median at 0.03 / 0.10 / 0.25 / 0.40 | Consumer-16 median | The best competitive-core cell (the 5090 at 0.03, the 5080 above) | The worst core cell |
|---|---|---|---|---|
| Die | 9.80 / 7.38 / 5.80 / 5.26 | 9.28 / 7.22 / 5.55 / 4.88 | 5.14 / 3.87 / 2.98 / 2.67 | 19.7 (the used 3090 at 0.03) to 10.7 (the B580 at 0.40) |
| Hybrid | 5.12 / 4.61 / 4.13 / 3.94 | 4.84 / 4.51 / 3.95 / 3.66 | 2.68 / 2.42 / 2.12 / 2.00 | 10.3 to 7.8 |
| GDDR7 machine | 3.24 / 3.11 / 2.95 / 2.88 | 3.07 / 3.04 / 2.82 / 2.67 | **1.70 / 1.63 / 1.51 / 1.46** | 6.5 to 5.6 |
Read: the one cell class inside the envelope anywhere is the best Blackwell card against the GDDR7 machine (1.46
to 1.70, under 1.5 from 0.25 per kWh up); every other GPU class as a NEW ENTRANT AT MSRP costs 2x to 20x any
specialist. The dearer the tariff the smaller the ratio, because power is a larger share of the GPU entrant's
cost than of the specialist's and the specialist's hardware term does not shrink.
4. **The heterogeneous tariff (the specialist at 0.03 against each GPU tariff, published beside, not the envelope's
test):** at R the die's median reads 9.8 / 13.1 / 19.8 / 26.7 against GPUs at 0.03 / 0.10 / 0.25 / 0.40, the
hybrid 5.1 to 13.9, the GDDR7 machine 3.2 to 8.8; the review's electricity axis multiplies the gap as it said.
5. **The adversarial combinations (at R, life 3; the same at 4R):** the die at USD 0.5 and a node ahead against GPUs at
0.40 with itself at 0.03, median 42 and p90 95 (the worst case in the cube); the die taking the whole chain with USD
75 M of development charged, median 2.7 (the development term is what brings it nearest the envelope, and the
standard forbids relying on research-cost recovery); hardware sales at zero development 5.9; the GPU entrant at
street 7.1; a 1.5x GPU generation against the die a node ahead 8.0; the die at USD 1.6 6.4; the GDDR7 machine
against street GPU prices at 0.25, 2.9. At the measured 5.5 percent proving efficiency the sustainability counts
and every ratio are unchanged (the ratio is a cost ratio; the proving income enters only the operating test, where
the measured day lowers no world below the line that was not already below it).
6. **What the failure means, and what it is not.** The envelope compares a GPU bought new at list price and run for
two years against a specialist that exists with its development sunk; on today's rows that comparison fails for
every specialist including the one the coexistence model passes on its seven conditions (the DRAM-board chip), and
it fails first on the GPU side's own spread: the cohort's median new entrant (a used 3080 or 3060, 940 to 1,254
micro-USD at 0.03 to 0.10) is 2x to 3x the best Blackwell entrant (493 to 656), so no specialist that beats the
best card by 1.5x can be within 1.5x of the cohort's median. **The result does not rest on a small network, token
appreciation or chip death** (the ratios are revenue-independent at zero development, the one sustainable world is
the largest, and the 180-day rotation enters nowhere); it rests on the measured energy of the AMD, Intel and older
NVIDIA classes on this hash and on the standard's choice of the cohort-wide median at MSRP as the test. The
coexistence model's seven-condition verdict (the board coexists, the die does not once built) is the operating
reading; ECO-05's envelope is the stricter new-entry reading, and the chain fails it as frozen.
## 0a. Amendment, 18:4x UK: the cube re-run at the corrected R (USD 31.65 M a year, the register's section 7a)
The second explorer read settled the reward at 1.6384 ETC (era 6 of ECIP-1017, confirmed from block 25,497,000's own
record) and R at USD 31.65 M a year, 1.54x below the first page's figure. The driver re-ran on build-4 at 18:48 UK at
both proving efficiencies (`sim/economy/coexist/out/eco05-cube-eff050-R31.65M.tsv` and `-eff0055-R31.65M.tsv`; the
first run's cubes stay beside them, labelled by their R). **The verdict is unchanged: RUN, FAIL.** What moved:
| Item | At R = USD 48.7 M (the first run) | At R = USD 31.65 M (corrected) |
|---|---|---|
| Sustainable worlds | one of sixteen: 10R at 0.03 (17 classes, 3 vendors, 100 percent) | one of sixteen: 10R at 0.03 (16 classes, 2 vendors, 100 percent; the equilibrium 948 micro-USD per MH/s-hour, 38 TH/s) |
| 4R at 0.03 | 14 to 15 classes, 1 vendor | 12 to 15 classes, 1 vendor |
| The envelope in the sustainable world (cells passing all three of 54) | die 0, hybrid 0, machine 0 | die 0, hybrid 0, machine 0 (the die's core passes in 3 cells, the USD 75 M development rows at life 1, which the standard forbids as a prop) |
| The boundary medians and best-core cells | as section 0.3 | identical (revenue-independent at zero development) |
| Adversarial row 2, the die taking the whole chain with USD 75 M charged, at R and 0.10 | median 2.7 | median 2.1 (the development term is larger against a smaller chain; still above 1.5 and the p90 8.0) |
| Every other adversarial row | | unchanged |
The one world that is sustainable at the corrected R is USD 316 M a year of miner revenue at 3 cents per kWh, which
is 10x a public GPU-mined chain's revenue at the cheapest tariff on the axis; the register's own reading of that world
(section 5, "sustainable expected" from R upward at 0.03 to 0.25) is contradicted by the driver at both values of R
and is reported as such.
## 0b. Batch 02, 21:0x UK: the re-run on the RX 7600's class v6 cell and the Arc B580's class v6 cell (the 7600's knee MODELLED; the metered knee is batch 03 in the morning)
The AMD-and-Intel lane's correction (20:2x UK): the reference population's RX 7600 cell was the class v4 sub-version
3 row (13.9 MH/s at 113 W); on class v6 the card reads 15.79 MH/s measured (PC 1), so the stock cell is 7.2
microjoules, the 9070 XT's 24 percent knee on it 5.4 (modelled) and the -30 percent power floor about 5.0; the Arc
B580 reads 11.0 MH/s measured on class v6 (PC 2, 19:50 UK) with its watts unread (about 110 W estimated, 10.0
microjoules estimated). The metered 7600 knee sits behind an overnight grid job on PC 1's runner and comes at about
05:30 UK as batch 03. Batch 02 therefore runs on the modelled 7600 knee (5.4) and the B580's class v6 cell, labelled
modelled, as the default; the frozen grid, R and envelope are unchanged and batch 01's FAIL stands as recorded.
Run on build-4 at 20:59 UK with the driver's per-class override (its first output line reads the override back:
`# overrides: {"RX 7600 8 GB": {"uJ": 5.4, "mhs": 15.79}, "Intel Arc B580": {"uJ": 10.0, "mhs": 11.0}}`); the cubes
are `sim/economy/coexist/out/eco05-cube-eff050-R31.65M-b02-modelled-7600.tsv` and `-eff0055-`, 2,592 cells each.
A first attempt at 20:57 UK ran master's copy of the driver, which predates the override hook, and reproduced batch
01's counts; it is discarded and the hook travels with this batch.
| Item | Batch 01 (corrected R; the 7600 at class v4's 6.2 / 13.9, the B580 at 10.4 / 10.7) | Batch 02 (the 7600 at 5.4 / 15.79 modelled, the B580 at 10.0 / 11.0) |
|---|---|---|
| Sustainable world-demand cells (of 48) | 3: 10R at 0.03 (16 classes, 2 vendors) | **6: 10R at 0.03 and 10R at 0.10** (0.10: 14 to 16 classes, 2 vendors, 100 percent operating; the equilibrium 1,225 micro-USD per MH/s-hour, 29.5 TH/s), at both proving efficiencies |
| The world that changed, and why | | the RX 7600 at the modelled knee has a new-entry cost of 1,206 micro-USD per MH/s-hour at 0.10 against the equilibrium's 1,225, so AMD becomes the second vendor with positive entry by 1.6 percent; at the class v4 cell (1,393) it did not; the B580 (2,105) does not at any world below 10R at 0.03 |
| The envelope in the sustainable worlds | 0 of 54 cells per specialist pass all three lines | **0 of 108**: die 12 / 0 / 9 of 108 on median / p90 / core, hybrid 24 / 0 / 12, board 39 / 0 / 24 |
| The new world's medians (life 3, zero development, private, normal) | | die 7.34, hybrid 4.59, board 3.09; best core cells 3.87 / 2.42 / 1.63 (the 5080); worst 13.1 / 8.2 / 5.5 (the used 3090) |
| Verdict | RUN, FAIL | **RUN, FAIL, unchanged**; one more sustainable world, no cell passing the envelope |
The sensitivity of the added world: it rests on a 1.6 percent margin between the 7600's modelled entry cost and the
equilibrium, so the metered knee (batch 03) can take it away (a knee above about 5.6 microjoules does) or keep it (at
or below 5.4); the envelope verdict does not depend on it either way. The two-vendor failure region of section 0.1
stands at every world below 10R.
## 1. The sustainability declaration as found (no specialist present; the first run at R = 48.7 M, the corrected run's rows in the TSV)
| Revenue | Tariff | Demand | Sustainable | r (micro-USD per MH/s-h) | GPU hash (TH/s) | Classes with positive entry | Vendors | Cohort above water |
|---|---|---|---|---|---|---|---|---|
| 0.25R | 0.03 | none / normal / spiking | no / no / no | 321 | 4.3 | 7 / 9 / 13 | 1 | 95 percent |
| 0.25R | 0.10 | the same | no | 467 | 3.0 | 7 / 9 / 13 | 1 | 57 / 91 / 91 |
| 0.25R | 0.25 | | no | 710 | 2.0 | 7 / 9 / 13 | 1 | 45 / 68 / 91 |
| 0.25R | 0.40 | | no | 951 | 1.5 | 7 / 9 / 13 | 1 | 39 / 45 / 91 |
| R | 0.03 | | no | 476 | 12.8 | 11 / 14 / 15 | 1 | 99 |
| R | 0.10 | | no | 604 | 10.1 | 11 / 12 / 15 | 1 | 68 / 91 / 91 |
| R | 0.25 | | no | 873 | 6.4 | 10 / 12 / 15 | 1 | 45 / 68 / 91 |
| R | 0.40 | | no | 1,079 | 5.2 | 10 / 12 / 15 | 1 | 45 / 45 / 91 |
| 4R | 0.03 | | no | 833 | 30.9 | 14 / 15 / 15 | 1 | 100 |
| 4R | 0.10 | | no | 1,028 | 21.9 | 12 / 15 / 15 | 1 | 98 |
| 4R | 0.25 | | no | 1,316 | 16.9 | 12 / 13 / 15 | 1 | 69 / 92 / 92 |
| 4R | 0.40 | | no | 1,530 | 14.6 | 12 / 12 / 15 | 1 | 46 / 69 / 92 |
| **10R** | **0.03** | | **yes / yes / yes** | 1,284 | 43.3 | 17 | 3 | 100 |
| 10R | 0.10 | | no | 1,354 | 41.0 | 15 | 1 | 100 |
| 10R | 0.25 | | no | 1,887 | 30.8 | 13 / 15 / 15 | 1 | 97 |
| 10R | 0.40 | | no | 2,301 | 25.3 | 13 / 14 / 15 | 1 | 69 / 92 / 92 |
Where the two-vendor test fails, the one vendor with positive entry is NVIDIA; the count of NVIDIA classes with
positive entry is 7 to 15 of 16. The proving demand (normal, spiking) adds 2 to 6 classes to the entry count and up to
46 points to the operating share, which is the second income doing what the coexistence model said it would; it never
brings an AMD or Intel card over the entry line below 10R.
## 2. The cube
Every cell is in `sim/economy/coexist/out/eco05-cube-eff050.tsv` (columns: specialist, revenue, tariff, life,
development, business model, demand, sustainable, r, GPU hash, the two sustainability counts, the specialist's cost
and its hardware, power and development terms, the median and p90 over the 19 classes, the consumer-16 median and p90,
the best and worst competitive-core cells with their class, and the three envelope verdicts). The cells of the one
sustainable world for the die (the first 54 rows of its block) read, as medians: life 1 at zero development 5.4
(private) and 3.2 (sales), at USD 20 M 2.8 and 2.0, at USD 75 M 1.2 and 1.0; life 3 at zero 9.8 and 6.8, at 20 M 6.2
and 4.9, at 75 M 3.1 and 2.7; life 5 at zero 11.7 and 8.8, at 75 M 4.6 and 4.1. The only die cells under the median
line are those where USD 75 M of development is charged against a 1-year fleet, and they fail the p90 and the core;
the standard says that recovery may not prop up a result, and it does not here.
## 3. The failed worlds, explained (life 3, development zero, private mining, normal demand)
| Specialist | Tariff | The worst core cell and its ratio | That GPU entrant's cost (micro-USD) | The specialist's cost: total (hardware / power) | The cohort's median class and cost | The best core cell | Which term fails |
|---|---|---|---|---|---|---|---|
| Die | 0.03 | the used RTX 3090, 19.7 | 1,889 | 96 (43 / 52) | the used RTX 3080, 940 | the 5090, 5.14 | the GPU's hardware: a used 3090 at USD 900 for 37.8 MH/s is 1,700 micro-USD of annualised hardware against the die's 43 |
| Die | 0.10 | the used RTX 3090, 13.1 | 2,217 | 170 (43 / 125) | the used 3060, 1,254 | the 5080, 3.87 | hardware on the GPU side; the die's power term is 125 at a 2.3x edge |
| Die | 0.25 | the Arc B580, 10.9 | 3,587 | 328 (43 / 282) | the used 3080, 1,902 | the 5080, 2.98 | the B580's 10.4 microjoules (watts estimated) at 0.25 is 2,700 of power alone |
| Die | 0.40 | the Arc B580, 10.7 | 5,211 | 486 (43 / 439) | the used 3080, 2,558 | the 5080, 2.67 | power on both sides; the ratio falls toward the joule ratio (4.5x) as the tariff rises |
| Hybrid | 0.03 to 0.40 | the used 3090 10.3 to the B580 8.0 | the same | 184 to 649 (120 / 62 to 523) | the same | the 5090 2.68 to the 5080 2.00 | hardware at the cheap tariffs, power at the dear |
| GDDR7 machine | 0.03 to 0.40 | the used 3090 6.5 to the B580 5.9 | the same | 290 to 888 (207 / 80 to 673) | the same | **the 5090 1.70 to the 5080 1.46** | the machine's hardware (207) is a third of the best Blackwell entrant's and its power 1.5x less; the median class is a used Ampere or Ada card at 3x the machine |
The pattern: the envelope's median is set by the cohort's middle (the used 3080, the used 3060, the 4070: 940 to 2,558
micro-USD), which is 2x to 3x the best Blackwell card as a new entrant because of their joules (4.2 to 5.8 against
1.7 to 2.1) and, at the cheap tariffs, their hardware per MH/s. No specialist within the envelope against the best
card can be within it against the middle. The core's worst cell is the used 3090 at the cheap tariffs (hardware) and
the Arc B580 at the dear ones (power), the two classes whose rows carry the register's BLOCKED labels (the 3090's
cap modelled, the B580's watts estimated); replacing them with measured rows moves the worst cell but not the verdict,
since the next-worst core cells (the RX 7600 at 10.2, the 9070 XT at 8.6 against the die) are also above 2.0.
## 4. The boundary of the envelope, as a question
What would pass: for the GDDR7 machine the best-core cell passes 1.5 from 0.25 per kWh and 2.0 everywhere, and its
consumer-16 median would pass 1.5 if the cohort were the Blackwell cards alone (1.46 to 1.9); for the hybrid and the
die no cohort of today's cards passes at any tariff with development sunk. So the envelope as frozen is passable by the
DRAM-board specialist only against a Blackwell-only cohort, which the standard does not allow to be chosen after the
fact, and is not passable by the SRAM die or the hybrid against any cohort of today's cards. That boundary is the
honest statement of the chain's position against a funded programmable competitor: the stored-dataset board coexists
with the best honest cards and no other, and the SRAM die with none.
## 5. Uncertainty and the model's range
- The die's ticket band (0.5 to 1.6) moves its medians from 7.2 (at 1.6) to 10.7 (at 0.5) at 0.10, the hybrid's and
the machine's rows have no band; the direction of every verdict is unchanged across the band.
- The modelled knees (Ada, Ampere) and the two estimated core rows move the worst core cell by up to 2x and the
cohort median by under 10 percent; no verdict moves.
- Street against MSRP for the GPU entrant raises every ratio 10 to 20 percent (adversarial row 4).
- The proving efficiency (0.5 assumed, 0.055 measured) changes no ratio and no sustainability verdict.
- The one sustainable world's existence rests on the register's two-vendor rule and the measured AMD and Intel rows;
a measured RX 7600 knee under 5 microjoules or a B580 under 8 would add worlds at 0.03 and 0.10, not above.
## 6. Status for the registry
ECO-05: RUN, FAIL against the frozen envelope in the sustainable worlds (0 of 54 cells per specialist pass all three
lines); the sustainability declaration as found differs from the register's expectation and is reported; the failure
regions are explicit (section 1 for commodity participation across vendors; sections 0.3 and 3 for the envelope);
collapse worlds retained as safety and exit tests; nothing imposed. Owed before G4: the second explorer read of R; the
RX 7600's own knee and the B580's metered watts; the independent reviewer's adversarial combinations (ECO-08).

134
docs/analysis/class-v6/eco-05-scenarios.md Normal file
View file

@ -0,0 +1,134 @@
# ECO-05 scenario register: stress success, contraction and cheap electricity (frozen before any result)
Case ECO-05, registry row `docs/plans/igneum-2.0-test-registry.json` suite ECO, gate G4, profile P12 ("Five-year
coexistence envelope"), priority GATE, owner the economics lead and an independent reviewer, status NOT RUN at the
time of this register. Written 8 October 2026, 18:4x UK, branch `class-v6-floor-sram`, floor lane 3, on the research
lane's word of 18:3x UK under the approved Test and Acceptance Standard. **This file is frozen before any ECO-05
result is computed; the result cube (`eco-05-results.md`) cites it and changes nothing in it. Unknown inputs are
BLOCKED, never guessed; nothing is weakened to pass.** The model is the coexistence model
(`docs/analysis/class-v6/coexistence-model.md`), its reference population (`reference-population.md`) and the
profitability surface (`floor/sram-and-floor.md` section 4.4); every row of those is modelled unless labelled measured.
The founder is not named.
## 1. The sourced revenue reference R
| Item | Value | Source and date | Label |
|---|---|---|---|
| The reference chain | Ethereum Classic (ETC): a public GPU-mined chain (Etchash, GPU and ASIC mined, the GPU share not published; the standard's default when no better source exists) | | the standard's default |
| ETC block reward | 2.5600 ETC per block | miningboard.com, https://miningboard.com/mining/ethereum-classic/halving, read 8 October 2026 (the page's data "15 minutes ago" at 18:3x UK); the page states a constant reward with no halving schedule, which conflicts with ETC's published 5M20 reduction schedule (a 20 percent cut every 5,000,000 blocks); the page's figure is carried as read and the conflict is a limitation | claimed (the page) |
| Block time | 13.0 s | the same page | claimed |
| ETC price | USD 7.83 (down 6.45 percent in 24 hours) | the same page, the same reading | claimed, point in time |
| Network hashrate | 143.42 TH/s | the same page | claimed |
| Blocks a year | 31,557,600 / 13.0 = 2,427,508 | arithmetic | modelled |
| ETC minted a year | 2,427,508 x 2.56 = 6,214,420 ETC | arithmetic | modelled |
| **R, the trailing-year miner revenue** | **USD 48.7 M a year** (6,214,420 x 7.83); USD 133,000 a day | arithmetic on the page's figures | modelled |
| R's range | USD 40 M to 55 M a year on the year's price range (USD 6.51 in August 2026 to 8.84 in May 2026, the same site's monthly pages via the search summary of 8 October 2026) | the search summary; a point-in-time price times a year's emission is not a trailing sum, which is a limitation | approximate |
| The IGN price that the spec's year-1 emission implies for R | 0.77 B IGN to miners in year 1 (3,168,808,781 base units per DAA second, 10^8 per IGN, 80 percent to miners, the 30-day ramp): **USD 0.063 per IGN** at R; 0.016 at 0.25R, 0.25 at 4R, 0.63 at 10R | the spec's constant; arithmetic | modelled; never an assumed appreciation |
| The rejected alternative | Kaspa: ASIC-mined since 2023 (not a GPU reference); its public page read the same way (https://miningboard.com/mining/kaspa/halving, 8 October 2026: 93 KAS per 1.0 s block, USD 0.04, 340 PH/s) gives USD 117 M a year, carried here only as a cross-check of R's order of magnitude | claimed; rejected as a GPU reference |
## 2. The mandatory worlds (the standard's P12 list, verbatim as axes)
| Axis | Values | Note |
|---|---|---|
| Revenue | 0.25R, R, 4R, 10R (USD 12.2 M, 48.7 M, 195 M, 487 M a year of miner revenue; IGN 0.016, 0.063, 0.25, 0.63) | the year-1 emission; the halvings in later years are the five-year run's, not this grid's |
| Electricity | USD 0.03, 0.10, 0.25, 0.40 per kWh | the GPU side's tariff; the matched-tariff test puts the specialist at the same tariff; the heterogeneous test (section 4) puts the specialist at 0.03 against each GPU tariff |
| Productive life | 1, 3, 5 years | the specialist's; the GPU entrant's horizon is 2 years with resale as the reference population's method |
| Development cost | zero (sunk, the mandatory case), USD 20 M, USD 75 M | amortised over the specialist's fleet and life; the fleet is a third of the world's GPU-equilibrium hash (declared here; the whole-chain case is reported beside it as an adversarial combination) |
| Business model | private mining (the operator self-mining at the unit ticket), hardware sales (the buyer pays twice the hardware term; the manufacturer's half) | |
| External proving demand | none, normal (USD 2,000 a day at the base fee, spec 05's launch grid, 90 percent to provers), spiking (x10) | a second income for the GPU cohort's proving-capable classes (16 GB and up; the 12 GB tier a measured zero), none for any hash engine |
| Grid size | 4 x 4 x 3 x 3 x 2 x 3 = 864 worlds per specialist | |
## 3. The specialists and the honest cohort
| Side | Rows | Energy and ticket | Label |
|---|---|---|---|
| Specialist 1 | the N2 SRAM die with the shadow core | 2.3x per joule vs the 5090 at its lock node-for-node (3.1x a node ahead), the reconciled unit ticket USD 1.0 per MH/s with the band 0.5 to 1.6 as adversarial ends | modelled (the adversary lane's reconciled rows; the chip model's energy) |
| Specialist 2 | the stored-half hybrid board | 1.93x, USD 2.80 per MH/s | modelled |
| Specialist 3 | the complete GDDR7 machine | 1.5x, USD 4.84 per MH/s | modelled |
| The honest cohort | the reference population's 19 classes (block 1 the plan's eight, block 2 the eleven others), each as the existing owner and the new entrant at MSRP (the matched-tariff new-entry test) and at street | measured where the population says so, modelled knees elsewhere | per cell |
| The competitive core | RTX 5090, RTX 5080, RTX 4090, RTX 3090, RX 9070 XT, RX 7600 8 GB, Intel Arc B580: seven configurations across three advertised vendors and five discrete generations (Blackwell, Ada, Ampere, RDNA 4, RDNA 3, Xe2) | declared before any result | the standard's P02 asks at least six across every advertised vendor and two generations |
| The entry cohort | the 19 classes weighted by the reference population's installed-base counts (approximate) | | |
## 4. The measure and the pass envelope (verbatim)
- Cost per accepted MH/s-hour as the reference population defines it (annualised hardware plus power plus hosting,
failures and fees, over 97 percent accepted work; the specialist with farm hosting USD 0.02 per kWh on top and its
development cost amortised as section 2 says).
- The cell ratio: the GPU new entrant's total cost per accepted work at MSRP over the specialist's total cost per
accepted work, at the SAME tariff (matched tariff). The heterogeneous-tariff cells (the specialist at 0.03 against
each GPU tariff) are published beside them and are not the envelope's test.
- **The pass envelope, in every mandatory sustainable world: the matched-tariff median over the cohort at most 1.5;
the 90th percentile at most 1.75; no competitive-core cell above 2.0.** Published for every cell.
- Sustainable worlds (declared before running, section 5): a world has enough funded demand for rational ongoing
service when, at the GPU side's own equilibrium (revenue per MH/s-hour set by the per-class supply curve with the
installed base as the cap), at least three purchasable GPU configurations across at least two advertised vendors
have positive modelled new-entry economics at MSRP, and at least 75 percent of the entry cohort by installed count
has positive marginal operating economics (owner cost below revenue per accepted unit). Worlds that fail that test
are collapse worlds and are retained as explicit safety and exit tests, never as profitable successes.
- Not imposed anywhere: GPU market share, specialist production limits, token appreciation, full research-cost
recovery, automatic chip death. The 180-day rotation enters only as the 1-year life's reason, not as a death.
## 5. The declaration of sustainable worlds, before running
By the rule in section 4, read off the reference population's owner and entrant rows at each tariff and the
per-class equilibrium (the coexistence model's section 9 rule), without any specialist present:
| Revenue | 0.03 | 0.10 | 0.25 | 0.40 |
|---|---|---|---|---|
| 0.25R (USD 12.2 M a year) | to be read by the driver and published; expected collapse (the equilibrium hash under 2 TH/s, fewer than three classes with positive entry) | the same | collapse expected | collapse expected |
| R (48.7 M) | sustainable expected | sustainable expected | to be read | collapse expected |
| 4R (195 M) | sustainable expected | sustainable expected | sustainable expected | to be read |
| 10R (487 M) | sustainable expected | sustainable expected | sustainable expected | sustainable expected |
The driver reads every cell and publishes the declaration as found; an expectation above that the driver contradicts
is reported as such, not revised. The sustainability test is applied with no specialist present (the GPU side's own
economics), so that a world is not declared unsustainable because a specialist exists.
## 6. The adversarial combinations (the independent reviewer may add; none removed)
1. The die at the band's cheap end (USD 0.5 per MH/s) and a node ahead (3.1x), with development zero and private
mining, at the GPU's dearest tariff (0.40) and the specialist's cheapest (0.03): the heterogeneous worst case.
2. The die taking the whole chain (development amortised over 100 percent of the equilibrium hash) rather than a third.
3. Hardware sales at zero development: the manufacturer's half with no project to recover.
4. The GPU entrant at street rather than MSRP (the market's price, not the list).
5. A GPU generation at 1.5x per joule at the same price (the coexistence model's year-3 step) against the specialist a
node ahead.
6. The spiking proving demand credited to the GPU side at the measured 5.5 percent proving efficiency rather than the
assumed 0.5.
## 7. Outputs the results file must carry
The full cube (every cell: world, specialist, median, p90, max core cell, the three envelope verdicts, the
sustainability verdict and its two counts), boundary tables (the tariff and revenue at which each specialist crosses
1.5, 1.75 and 2.0), a failed-world explanation per failing cell (which cohort classes carry the median, which core
cell breaks 2.0 and by what term, hardware or power), the adversarial combinations beside the grid, and the model's
uncertainty (the die's ticket band, the modelled knees, the approximate prices) stated as a range on the median.
## 7a. Amendment, 18:4x UK: the second explorer read of R (the register's section 1 above is kept as frozen; this section records the correction and its minute)
| Item | Value | Source and date | Label |
|---|---|---|---|
| Block height | 25,497,034 | the Ethereum Classic Blockscout explorer's own stats endpoint, https://etc.blockscout.com/api/v2/stats, last updated 2026-10-08T17:47:09Z, read 18:4x UK | measured by the explorer |
| Average block time | 12.758 s | the same | measured by the explorer |
| ETC price | USD 7.81 | the same | point in time |
| **The block reward, from a block's own record** | **1.6384 ETC** (1,638,421,000,000,000,000 wei, "Miner Reward", block 25,497,000, timestamp 2026-10-08T17:17:59Z) | https://etc.blockscout.com/api/v2/blocks/25497000, read 18:4x UK | measured by the explorer |
| The 5M20 question, settled | ETC's ECIP-1017 reduction (20 percent every 5,000,000 blocks) is in force: era 6 began at block 25,000,000 (about June 2026 at 12.8 s blocks) with 5 x 0.8^5 = 1.6384 ETC; the first page's 2.56 was era 4's figure (blocks 15 M to 20 M) and its "no halving schedule" line was wrong | arithmetic on ECIP-1017 against the explorer's block | settled |
| Blocks a year | 31,557,600 / 12.758 = 2,473,554 | arithmetic | modelled |
| ETC minted a year | 2,473,554 x 1.6384 = 4,052,672 ETC | arithmetic | modelled |
| **R, corrected** | **USD 31.65 M a year** (4,052,672 x 7.81); USD 86,700 a day; the first page's USD 48.7 M was 1.54x too high | arithmetic on the explorer's figures | modelled |
| The implied IGN price at R | USD 0.041 per IGN (0.77 B to miners); 0.010 / 0.164 / 0.411 at 0.25R / 4R / 10R | the spec's constant | modelled |
| The remaining limitation on R | a point-in-time price times a year's emission at today's reward, not a trailing sum over a year in which the reward fell from 2.048 to 1.6384 at block 25 M and the price ranged USD 6.5 to 8.8; a trailing sum would read about USD 35 M to 45 M | approximate | stated, not BLOCKED |
The grid, the envelope, the rule and the declaration are unchanged; the cube is re-run at the corrected R (section
0a of `eco-05-results.md`) and both runs are published.
## 8. What is BLOCKED
- The ETC page's "no halving schedule" against ETC's published 5M20 schedule: R is carried at the page's reward and
flagged; a second explorer read would settle it and is owed before G4.
- The GPU share of ETC's hashrate (Etchash is ASIC-mined in part): unknown; R is the chain's whole miner revenue.
- The Arc B580's watts (estimated, about 110 W) and the RX 7600's knee (estimated by the 9070 XT's grid): those two
core cells carry an estimated energy and are published with that label; the Intel lane's meter and the 7600's own
grid replace them.
- No chip has been measured; every specialist row is modelled on the chip model's method and the adversary lane's
placed rows.

319
docs/analysis/class-v6/floor/sram-and-floor.md
View file

@ -22,20 +22,23 @@ schedule and which served claim.
a wordline sensed) does not. At today's read width (`W = 1`, 4 bytes) the die moves about 80 bits per dependent read
(address, control, data), not 560: **0.25 nJ per read, 8.3 GH/s per die, 66x the 5090 at zero shadow** (modelled on
the record's own wire figure). The record's 17x is the `W = 16` row. With the class v4 shadow at `k = 0.5` the same
chip reads 5.7x (the record's 4.8x), at `k = 1` 3.0x. **On the k lane's synthesised core (13:1x UK, section 2.2b:
about 0.11 microjoules of shadow per hash at N3, absolute `k` about 0.1, the shuffle row open) the same die reads
21x at W = 4 and 18x at W = 8 under class v4, 10x and 9.7x at the honest cards' whole latency shadow, 6x to 7x at
the 5090's knee under the full shadow.** On the record's claimed band the shadow is the whole hold and the width
moves the shadowed number 0.2x to 0.7x; on the synthesised core the memory is a third to a half of the die's energy
and the width is worth 1.3x. The public 2x is not reachable against this die either way.
chip reads 5.7x (the record's 4.8x), at `k = 1` 3.0x. **On the k lane's synthesised programmable core (13:5x UK,
section 2.2b: 3.5 pJ per lane-op at N3, 2.5 at N2, pre-place; 0.36 microjoules of class v4 shadow per hash at N3,
absolute `k` 0.31 at stock and 0.56 at the lock) the die reads 7.7x to 8.1x under class v4 at stock and 5.3x to
5.6x at the knee, and 3.4x to 3.5x at the honest cards' whole latency shadow at stock, 2.2x at the 5090's knee (an
N2 core: 10x to 11x, 6.9x to 7.4x, 4.7x to 4.8x, 2.9x to 3.0x).** The bare-lane floor (0.11 microjoules, `k` 0.1)
read 18x to 21x and 10x; the record's claimed band and that floor bracket the core, which sits inside the record's
band at its low end. The shadow is most of the hold; the memory is a tenth to a fifth of the die's energy per hash
and the read width is worth 1.05x on the full-shadow number. The public 2x is reached only at the knee under the
full shadow on an N3 core (2.2x), not at stock and not at N2.
2. **The one lever on the wire is the read width, and it is free on the honest cards up to 16 bytes.** `W = 4` words
(16 bytes) is measured within 2.7 percent on the 5090 and the 9070 XT and within 1 percent on the M5 Max (the
record, 5 October); it moves the die to 176 bits per read, 0.38 nJ, **44x at zero shadow, 5.6x at k = 0.5**. `W = 8`
(32 bytes, exactly the GDDR7 sector the 5090 already fetches) has no row of its own but is free by the measured
rows: the 5090's ceiling is about 18 G sectors a second (w16 moved 573 GB/s of sectors at 139.8 MH/s, w64 589 GB/s
at 71.9, read-width.md), an aligned 32-byte read is one sector per load exactly as w16 is, AMD moves its 64-byte line
either way, and the M5 Max read 1.03 of its v2 rate at both w16 and w64; it reads 0.55 nJ, **31x and 5.4x**, and one
PC 1 row confirms it. `W = 16` costs the 5090 47 percent of its rate (two sectors per load, measured) and is dead. The fold is keyed by the destination
(32 bytes, the GDDR7 sector) was measured on the 5090 at stock at 15:03 UK (the hash lane, PC 1): the rate held to
0.01 MH/s but the energy rose 4.8 percent (the second sector of the pair costs the card about 1.2 nJ a read); it
reads 0.55 nJ on the die, **31x and 5.4x**, and with the shadow on lands within 0.2x of W = 4 while costing the card
5 percent, so **W = 4 pins** (the lock row owed). `W = 16` costs the 5090 47 percent of its rate in the exported
form and 25 percent of its energy in the hinted form (measured, section 2.2c) and is dead. The fold is keyed by the destination
register through its first word (`verify::fold_words`), so a chip cannot pre-fold a wide read into a narrower table:
the raw words must cross the die. A chain of 2, 4 or 8 dependent atoms per step moves nothing: the ratio is per
read on both sides. Banking plus a sequencer cannot localise: the chain alternates the 16 load sites and their
@ -51,8 +54,10 @@ schedule and which served claim.
powered, the store's USD per MH/s does not rise with the dataset (USD 0.25 to 0.5 at every size). The floor
raises the minimum unit from USD 500 to USD 5,000, which is 2.5 times a 5090's street price. And the floor has a
cost on the honest side that the record read as zero: the 5090 at its knee pays 4, 8 and 10 percent more energy
per hash at 2, 4 and 8 GiB than at 1 GiB (measured, PC 1, 12:58 UK; 1 to 4 percent unlocked), so each step moves
the honest denominator toward the chip by the same share. **The schedule that
per hash at 2, 4 and 8 GiB than at 1 GiB (measured, PC 1, 12:58 UK; 1 to 4 percent unlocked), and **14.3 percent
at the schedule's own 5.5 GiB step on the multiply-shift mapping (measured, PC 1, 17:4x UK; 4.8 at stock), over a
10 percent budget at the lock**, so each step moves the honest denominator toward the chip by that share and the
schedule's first step already costs a tuned 5090 more than the chip's joules move at any step. **The schedule that
keeps the most honest cards is lane A's: 5.5 GiB at the v6 epoch (3 reticles, USD 1,500), 8.5 GiB two years on (5
reticles, USD 2,500, the last size inside the 12 GB tier's room), 11.5 GiB at four years (6 reticles, USD 3,000,
the last inside the 16 GB tier); each step sits just under a card tier's room and just over a reticle multiple.**
@ -74,10 +79,11 @@ schedule and which served claim.
stays pinned and the data must travel), and the note that 3D-stacked SRAM (SoIC-class, a vertical hop at about 0.1
pJ per bit, approximate) would make a multi-die store read as one die, which weakens the floor on joules further.
The honest line: against this die the memory lever is worth a factor of two at zero shadow and 1.3x with the shadow
on at a synthesised core (a tenth on the record's claimed band); the floor is a ticket; the project cost against the
chain's revenue is the wall; and the number the public text can carry is a range, 6x to 10x at the honest cards'
whole latency shadow on the synthesised core (2x to 4x on the record's claimed band, marked), never "under 2x".
The honest line: against this die the memory lever is worth a factor of two at zero shadow and 1.05x with the shadow
on at the synthesised programmable core; the floor is a ticket; the project cost against the chain's revenue is the
wall; and the number the public text can carry is 3.4x to 3.5x at the honest cards' whole latency shadow at stock
and 2.2x at the 5090's knee on an N3 core (4.7x and 2.9x on an N2 core; 6x to 10x on the bare-lane floor, marked as
the worst case), with "under 2x" reached only at the knee on the N3 core.
## 1. Inputs
@ -98,6 +104,7 @@ whole latency shadow on the synthesised core (2x to 4x on the record's claimed b
| The card population | by count of the bench table's 32 measured consumer cards: 6 GB 3 percent, 8 GB 22, 10 and 11 GB about 9, 12 GB 22, 16 GB 28, 24 GB and up 16; the fleet's hashrate-weighted census unread | approximate | lane A 4.2a via the design document 3.4 |
| The Apple rate cost of the working set | -12 percent at 2 GiB, -20 at 4, -22 at 8 against 1 GiB on the M5 Max | measured, 10:40 UTC | the design document 3.3 |
| The 5090's cost of the working set | against the 1 GiB control (137.65 MH/s at 312.2 W unlocked, 127.39 at 212.6 W at the 1,300 MHz lock): 2 GiB -2.8 percent unlocked and -4.9 at the lock, 4 GiB -3.8 and -11.2, 8 GiB -4.3 and -14.1; MH/W at the lock 0.599, 0.576, 0.553, 0.543 (4, 8 and 10 percent per hash in energy at the knee) | measured, 12:49 to 12:58 UK, PC 1, label ca4-v6 | the hash lane's message of 13:0x UK |
| The 5.5 GiB step on the 5090 at stock | a rented 5090 (RunPod, driver 570.195.03), the kit's own CUDA worker, 250 and 500 batches of 2^24: the 1 GiB control 141.48 / 141.38 MH/s at about 328 W steady (2.30 to 2.32 microjoules), the 5.5 GiB pack 136.56 / 136.54 MH/s at the same watts (2.40 on the steady watts): -3.5 percent of rate, about +4 percent of energy per hash; the non-power-of-two mapping (`(src * words) >> 32`, 92,274,688 items) is not a cliff on sm_120; fingerprints stable across both passes (23ced07a4d28b465 becomes the pin); memory.used peak 6,522 MiB. Caveat: the host capped the card at about 328 W on both packs, so the microjoules are capped-card numbers, the rate and fingerprints stand | measured, 15:19 to 15:23 UK (the fleet lane, ca4-v6-size) | the fleet lane's message of 15:2x UK; `~/igneum-fleet/ca4-v6-size.log` |
| The emission | 3,168,808,781 base units per DAA second, 10^8 base units per IGN (1.0 B IGN a year in years 1 and 2), a 30-day ramp (about 37 M IGN never minted), halving every 63,115,200 DAA s, 80 percent to miners | spec | spec 05 (the economics table, section 2.5) |
| The mission lane's break-even model | revenue `s x E2 x p`; cap = `C_proj / s` in years 1 to 2 at `s` = 0.30; its E2 of 4.18 B IGN is not this emission constant's figure (1.57 B to miners), so this file recomputes from the constant and carries both | model | mission/future.md 2.2 |
| The re-fill | a 1 GiB rebuild 157 G ops, 13.4 ms on a 5090, 0.16 to 0.5 J on a chip core; class v5 refreshes per epoch (3,600 s) | measured / modelled | the invention lane 2.11, 2.15; the research file row 7 |
@ -118,7 +125,7 @@ rate. With the shadow: `edge = (2.26 + F) / (E_hash0 + k F)`. All modelled; the
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 (today, 4 B) | 80 | 0.25 | 8.3 | 0.036 | **66x** | 22x | 5.7x / 3.0x | 4.1x / 2.1x | 4.0x | 0 | modelled |
| 4 (16 B) | 176 | 0.38 | 5.6 | 0.054 | **44x** | 14x | 5.6x / 2.9x | 4.0x / 2.1x | 3.9x | 5090 +2.7 percent rate, 9070 XT -1.4, M5 Max within 1 (measured) | modelled chip, measured cards |
| 8 (32 B, the GDDR7 sector) | 304 | 0.55 | 3.9 | 0.078 | **31x** | 10x | 5.4x / 2.9x | 4.0x / 2.0x | 3.6x | free by the measured rows: one sector per load on NVIDIA as at w16 (the ceiling about 18 G sectors a second: 573 GB/s at w16, 589 at w64), one 64 B line on AMD, and the M5 Max at 1.03 of v2 at both w16 and w64; a PC 1 row confirms (the crate's width set is {1, 4, 16} words, so a pack at 8 needs a generator and emitter line first, the hash lane's caveat) | modelled chip; the card by interpolation of measured rows |
| 8 (32 B, the GDDR7 sector) | 304 | 0.55 | 3.9 | 0.078 | **31x** (32.5x net) | 10x | 5.4x / 2.9x | 4.0x / 2.0x | 3.6x | **MEASURED 15:03 to 15:06 UK (the hash lane, PC 1, the 5090 at stock under the class v5 state term, 250 batches of 2^24, every row PASS): rate held to 0.01 MH/s (117.54 against 117.54 at W = 4) at 451.5 W against 430.9, 3.84 against 3.67 microjoules, +4.8 percent of energy per hash; the second sector of the pair costs the card about 1.2 nJ a read, not 0.** This file's interpolation ("free") was wrong on the energy. **The lock row landed (run b, 16:26 to 16:46 UTC, PC 1, every fingerprint equal to the morning's; PCIe gen 4 since the eGPU swap, rows compare within the job): at 1,300 MHz W = 4 reads 140.7 MH/s at 322.8 W, 2.29 microjoules; W = 8 119.1 at 303.7 W, 2.55, +11.2 percent per hash (the lock widens the loss from +5.6 at stock); W = 16 67.7 at 277.0 W, 4.09, +78 percent (+70 at stock). Both stay out, as regression controls.** With the card at 1.048x and the programmable core: 7.9x / 5.5x under class v4, 3.5x / 2.2x at the full shadow, the same as W = 4 within 0.2x; the DRAM chips' second sector costs them the DRAM's 1.15 nJ, so the GDDR7 chip's zero-shadow edge falls 5.1x to 4.7x. **W = 8 buys the chain nothing with the shadow on and costs the card 5 percent: W = 4 pins** | modelled chip; measured card |
| 16 (64 B, the record's atom) | 560 | 0.88 | 2.4 | 0.125 | 19x (the record's 17x at 1.0 nJ) | 6.2x | 5.0x / 2.7x | 3.8x / 2.0x | 3.2x | the 5090 -47 percent of rate (measured): dead | modelled |
Reading: the record's 1.0 nJ and 17x are the 64-byte row; at the hash's own width the die is 3.5x stronger at zero
@ -161,13 +168,34 @@ pays 0.36 (N3) or 0.25 (N2). The re-fold, `E_chip = E_hash0 + shadow`, against t
| 8 | **18x / 12x** | 21x / 15x | **9.7x / 6.1x** | 13x / 8.2x | 7.5x / 8.9x | the same |
| 16 | 14x / 9.9x | 16x / 11x | 8.8x / 5.5x | 11x / 7.1x | 6.0x / 6.8x | the same |
Reading, which replaces "the shadow is the whole hold" of section 0 where the synthesised figure stands: at a
measured-class `k` of 0.1 the shadow holds the die to 14x to 23x under class v4 and 6x to 11x at the honest cards'
whole latency shadow, not 2x to 6x; the die's memory energy is then a third to a half of its energy per hash, and the
read width is worth 1.3x (W = 1 to W = 8) on the shadowed number, not 0.2x. The public 2x is not reachable against
this die by any shadow at a synthesised core; what holds it is the project cost against the chain's revenue (section
4), which the shadowed edge moves by 1.4x across the whole band. The rows carry the k lane's labels (ASAP7 measured by
synthesis, the node scaling approximate, the shuffle open); its own re-fold replaces this table when it lands.
That table is the bare-lane FLOOR (no fetch, decode or register-file clocking). **The programmable-core row (floor
lane 2, 13:5x UK, synthesis-only, pre-place: imem, fetch, decode, the era registers, a 32 x 32-bit register file per
lane, every class unit including the cross-lane shuffle butterfly and the load fold, 8 lanes, a 256-instruction
program at the class v4 weights, gate-level random-input VCD) reads 6.9 pJ per lane-op at ASAP7, 4.8 at N5, 3.5 at N3,
2.5 at N2 (absolute; 2.4 of the 6.9 is the register file and imem clocking without gating), 3.1x the bare lane.** The
chip's class v4 shadow is then 0.36 microjoules per hash at N3 (0.26 at N2), the full 330,000-op shadow 1.16 (0.83);
absolute `k` at N3 0.31 against the 5090 at stock, 0.56 at the lock, 0.50 against the M5 Max. Placement (+20 to +40
percent, approximate) and register-file clock gating (most of the 2.4 pJ) pull opposite ways; the placed 8-lane and
32-lane rows follow by 15:20 UK. The re-fold on the programmable core, which is this file's headline for the close:
| `W` | Class v4 shadow, stock / lock, N3 core | The same, N2 core | Full shadow, stock / lock, N3 | The same, N2 | M5 Max class v4, N3 / N2 | W = 16 net (card 1.25x), v4 / full, N3 | Label |
|---|---|---|---|---|---|---|---|
| 1 | 8.5x / 5.9x | 11x / 7.8x | 3.6x / 2.2x | 4.9x / 3.1x | 3.5x / 4.7x | 9.9x / 4.0x | modelled on synthesised pJ, pre-place |
| 4 | **8.1x / 5.6x** | 11x / 7.4x | **3.5x / 2.2x** | 4.8x / 3.0x | 3.4x / 4.5x | 9.5x / 4.0x | the same |
| 8 | **7.7x / 5.3x** | 10x / 6.9x | **3.4x / 2.2x** | 4.7x / 2.9x | 3.2x / 4.1x | 9.0x / 3.9x | the same |
| 16 | 6.9x / 4.8x | 8.7x / 6.0x | 3.3x / 2.1x | 4.5x / 2.8x | 2.9x / 3.6x | 8.1x / 3.8x | the same |
| The DRAM chips at the same core (N3): GDDR7 | 4.1x | | 2.6x | | | | modelled |
| HBM3, one stack | 4.9x | | 2.9x | | | | modelled |
Reading: on the programmable core the shadow is most of the hold again. The SRAM die reads 7.7x to 8.1x under class
v4 at stock and 5.3x to 5.6x at the knee (N3; 10x to 11x and 6.9x to 7.4x at N2), and **3.4x to 3.5x at the honest
cards' whole latency shadow at stock, 2.2x at the 5090's knee** (N2: 4.7x to 4.8x and 2.9x to 3.0x); the memory is a
tenth to a fifth of the die's energy per hash, and the read width is worth 1.05x on the full-shadow number (W = 4 to
W = 8). The record's claimed band (`k` 0.3 to 0.8) and the bare-lane floor (`k` 0.1) bracket it; the synthesised
core sits at `k` 0.31 at stock and 0.56 at the lock, inside the record's band at its low end. The public 2x is
reached only at the knee under the full shadow on an N3 core (2.2x) and not at N2 (2.9x). What holds the die is still
the project cost against the chain's revenue (section 4), which the shadowed edge moves 1.4x across the band. The
placed rows replace this table when they land.
Sensitivity: the wire at 0.65 pJ per bit halves every wire term (`W = 1` 0.20 nJ, 83x; `W = 8` 0.35 nJ, 49x); at 2.6
pJ per bit doubles it (`W = 1` 0.35 nJ, 47x; `W = 8` 0.95 nJ, 18x). The macro at 0.2 nJ adds 0.1 nJ to every row. The
@ -192,10 +220,29 @@ the measured w64 row stands (71.9 MH/s, 47 percent of rate, 1.89x the energy per
| HBM3, one stack | 0.321 (7.5x) | 0.358 | 6.7x | 12.7x | 7.2x | 5.9x | modelled |
| SRAM die, one N2 reticle | 0.036 (66x) | 0.125 | **19x** | 36x | 14x | 8.8x | modelled |
Reading: W = 16 is the one width that moves the die by more than 2x at zero shadow (66x to 19x) and it costs the DRAM
chips 11 to 15 percent; it is worth that only if the 5090 passes, since at the measured w64 row every chip's edge
nearly doubles instead. The decision is the PC 1 measurement, not this table; if it passes, W = 16 replaces W = 8 as
this file's width recommendation and the die's shadowed rows read 14x (class v4) and 8.8x (full).
**Measured, 13:47 UK (floor lane 5, rented cards at stock, the hinted one-request form `ld.global.L2::64B` on the
item's first load, bit-exact):** the RTX 4090 holds its 4-byte rate at W = 16 (63.65 against 64.73 MH/s, 98 percent;
the exported form loses 48) but draws 296 W against 225 at the same rate, **+34 percent of energy per hash** (the
second sector costs the card about 8.6 nJ through L2 and the crossbar, not the DRAM's 1.15); the H100's best form is
-21 percent of rate and +33 percent of energy; the 3090's best -15 percent of rate. **The 5090 stock row (13:40 UK,
PC 1, the hinted form at the sparse shape sp170-w4): the 4-byte rate held (145.7 against 142.6 MH/s, +2 percent) at 401
W against 313 to 319, energy per hash 2.21 to 2.75 microjoules, +25 percent; the second sector costs the 5090 about
4.7 nJ through its fabric against the DRAM's 1.15.** The knee row is unread (a rented pod refuses the lock) and is the
one row that could move the verdict (it would need the second sector under about 2 nJ). The net, with the 5090 at
1.25x its energy and its rate held (the 4090's 1.34x in brackets):
| Chip | W = 16, zero shadow, net | Class v4 shadow (synthesised core), net | Full shadow, net | The same chip at W = 8 (card free) | Label |
|---|---|---|---|---|---|
| GDDR7 board | 5.5x (5.8x), up from 5.1x | 5.9x (6.3x) | 5.3x (5.5x) | 5.1x / 5.1x / 4.7x | modelled chip, measured card |
| HBM3, one stack | 8.4x (9.0x), up from 7.5x | 8.4x (8.8x) | 6.7x (7.0x) | 7.5x / 7.2x / 5.9x | the same |
| SRAM die | 24x (26x), from 66x | 16.7x (17.6x) | 9.9x (10.4x) | 31x / 17.9x / 9.7x | the same |
Reading: W = 16 is dead on the measured energy, not on the rate. It costs the honest card 25 to 34 percent of its
energy per hash for the DRAM chips' edges to RISE (the chip pays the second sector at the DRAM's 1.15 nJ, the card at
4.7 to 8.6) and for the SRAM die's shadowed rows to land where W = 8 already puts them for free (16.7x against 17.9x
under class v4, 9.9x against 9.7x at the full shadow). The one thing it buys is the die's zero-shadow number (24x
against 31x), which no served line carries. **W = 8 is this file's width; "never 16" stands on measured rows at stock
on the 5090, the 4090, the H100 and the 3090.** The knee row goes in as an amendment.
### 2.3 What cannot localise, checked
@ -274,8 +321,35 @@ moves the minimum ticket.
The honest side pays the floor in joules, which the record read as zero for the NVIDIA tiers: the 5090 at the 1,300
MHz lock loses 4.9, 11.2 and 14.1 percent of rate at 2, 4 and 8 GiB against 1 GiB (unlocked 2.8, 3.8 and 4.3), which
is 4, 8 and 10 percent more energy per hash at the knee (measured, PC 1, 12:49 to 12:58 UK, the hash lane's ca4-v6
rows: the page-walk cost the unlocked card hides). So the 5.5 GiB step costs a tuned 5090 about 9 percent per hash and
the 8.5 and 11.5 steps 10 to 11 (the curve past 8 GiB unmeasured, read as flattening), against the chip's ticket rising
rows: the page-walk cost the unlocked card hides). **The 5.5 GiB step itself is measured at stock (the fleet lane,
a rented 5090, 15:19 to 15:23 UK): -3.5 percent of rate at the same watts against the 1 GiB control (136.56 against
141.48 MH/s, both passes), about +4 percent of energy per hash, on the multiply-shift mapping of spec 1.13.3 option (a)
with no cliff on sm_120 (the worker's check line: "not a power of two: loads are (src * words) >> 32", self-test PASS
96 of 96); it sits on the 2 and 4 GiB stock rows (2.8 and 3.8 percent) where the interpolation put it, so the schedule's
non-power-of-two floors cost nothing beyond the size.** The 5.5 GiB knee row and the RX 7600's 8 GB reading (its 1.8
GB of headroom is the question for the 8 GB tier at the v6 epoch) come from PC 1 after its queue, about 16:00 to 16:30
UK, as amendments. **The 5.5 GiB knee row landed (PC 1, 16:46 to 16:50 UTC, the pinned class v3 program at both
sizes, fingerprints equal to the rented card's; PCIe gen 4 since the eGPU swap, rows compare within the job): at the
1,300 MHz lock 1 GiB 137.0 MH/s at 218.2 W, 1.59 microjoules; 5.5 GiB 116.5 at 212.1 W, 1.82: +14.3 percent of energy
per hash and 15 percent less rate; unlocked 2.25 against 2.36, +4.8 percent (the rented 5090 read +4).** So the 5.5
GiB step costs a tuned 5090 14 percent per hash at its knee (measured; the interpolation of 9 was low) and 5 at
stock, which is over a 10 percent budget at the lock: the lock takes the card into the DRAM-latency regime where the
larger footprint's TLB and page misses show. The 8.5 and 11.5 steps read 15 to 18 at the knee by the 8 GiB row's
shape (unmeasured past 8 GiB). **The 8 GB tier's own reading landed (the founder's RX 7600, PC 1, OpenCL on gfx1102,
30 dispatches of 2^24 per size, rate only, the AMD script carrying no wattmeter; the hash lane, 18:2x UK): the 5.5
GiB dataset FITS on the card's 8,176 MiB (6,732 MiB max alloc, the ds55 pack built and ran) and the rate is 15.75 /
15.46 / 15.34 / 15.35 MH/s at 1 / 2 / 4 / 5.5 GiB (-1.9 / -2.6 / -2.5 percent), fingerprints equal to the 5090's at
every size. The 8 GB tier pays under 3 percent of rate for the 5.5 GiB step where the locked 5090 pays 15: a 16-CU
card is latency-bound by its own lane count at every size and the footprint never reaches its limit; with the rate
flat its energy per hash moves with its watts alone (113 W at stock, today's card-in reading, approximate).** **The mapping control landed (PC 1, the same program byte for byte, 18:3x UK): at the 1,300 MHz lock the 4 GiB
mask path reads 1.780 microjoules and the 4 GiB multiply-shift path (2^30 minus 2^16 words) 1.782, +0.1 percent, so
the mapping term is zero at the knee and the dataset size is the whole cost; against this card's 1 GiB rows (lock
1.593, stock 2.248) the 4 GiB step is +11.7 percent at the knee and +0.7 at stock, the 5.5 GiB step +13.4 and
+4.3. Under the standard's P03 (no more than 5 percent of joules per accepted work against the paired tuned
baseline) no step past 1 GiB passes at the knee on a 5090; the decision is the research lane's.** So the
v6 epoch's 5.5 GiB floor holds the 8 GB tier in fact (section 3.2's worst-case "OUT" rested on the 75 percent rule
with the cache resident; the measured allocation is 82 percent of the card with the cache freed, and the card mines
at 97 percent of its 1 GiB rate), against the chip's ticket rising
from USD 500 to 1,500, 2,500 and 3,000 and its joules not at all; every chip edge in section 2 rises by the same 4 to 11
percent against a card at its knee. The 9070 XT and the smaller NVIDIA tiers have no size rows (approximate: the
5090's curve). Per step the floor buys the chain USD 1,000 of chip ticket for about 1 percent of the tuned 5090's
@ -289,7 +363,7 @@ a step forward. The proposed floor schedule, by height like a class epoch and a
| Step | Floor | Height (DAA seconds from the v6 epoch) | The state size that brings it forward instead | Reticles (N2 / 2031) | Silicon per store | Tiers that leave | Apple rate cost on top |
|---|---|---|---|---|---|---|---|
| the v6 epoch | 5.5 GiB | 0 | 92 M records | 3 / 3 | USD 1,500 | the 6 GB tier; the 8 GB tier only under the worst-case working set | about -21 percent against 1 GiB (between the measured 4 and 8 GiB rows) |
| the v6 epoch | 5.5 GiB | 0 | 92 M records | 3 / 3 | USD 1,500 | the 6 GB tier; the 8 GB tier HOLDS (measured on the RX 7600: the dataset fits at 6,732 MiB of 8,176, the rate -2.5 percent) | about -21 percent against 1 GiB on the Mac (between the measured 4 and 8 GiB rows); the 5090 -15 percent of rate at its knee (+14.3 percent per hash), -3.5 at stock; the RX 7600 -2.5 (all measured) |
| two years on | 8.5 GiB | 63,115,200 | 143 M records | 5 / 4 | USD 2,500 | the 8 GB tier (22 percent), the 10 GB 3080, the Apple 16 GB laptop; the 12 GB tier holds with the cache freed | about -22 percent (the measured 8 GiB row) |
| four years on | 11.5 GiB | 126,230,400 | 193 M records | 6 / 5 | USD 3,000 | the 12 GB tier (22 percent) and the 11 GB 1080 Ti and 2080 Ti; the 16 GB tier holds with the cache freed | -22 to -25 percent (unmeasured past 8 GiB) |
| the 16 GiB step | 16 GiB | whenever the state passes 268 M records, never by the calendar inside the window | 268 M | 8 / 7 | USD 4,000 | the 16 GB tier (28 percent), the 24 GB tier with the cache resident, Apple 32 GB | the same |
@ -370,7 +444,154 @@ a day, a cap of about USD 0.9 B) every DRAM-board project starts at a third of t
starts at about the same level (USD 330 M a year, 4.2) at 13x or a whole-chain maker.** The band between is where the
detector and the issuance trigger live (the history's USD 20 K to 50 K a day for compute-bound hashes sits at this
file's lower threshold). The corrected floor moves the lower number 4x (from USD 100 M a year on the SRAM project to
23 M) and the upper number not at all.
23 M) and the upper number not at all. **Superseded in form at 15:4x UK (the founder's accepted external review):
the two numbers are read as points on the surface of 4.4, which states the conditions under which development is
attractive rather than a level the chain stays below; the points themselves stand (a USD 20 M project taking the
chain above about USD 13 to 22 M a year over a 3-year life, at a third above 43 to 75 M; the SRAM project above
168 to 561 M).**
### 4.4 The profitability surface, first cut (16:0x UK; the founder's accepted external review of 15:4x: the single threshold comes out and a surface goes in, stated as the conditions under which development is attractive, never as "the chain stays below X")
The model (`scratchpad/surface.py`, run on build-3; every row modelled): an entrant pays its development cost `C_dev`
at time zero, mines nothing for a pre-production period `T0`, then holds a captured share `q` of the hash for a
productive life `L` and earns `q x m x R(y)` in each year, where `R(y)` is the year's miner emission at the price `p`
(the spec's constant: 0.77 / 0.80 / 0.40 / 0.40 / 0.20 / 0.20 B IGN in years 1 to 6) and `m` is the operating margin.
GPU miners enter and exit at their all-in cost (the 5090 at the lock over two years: USD 0.00092 per MH/s-hour,
0.000084 of it electricity, research 16.1), so while any GPU mines the revenue per MH/s-hour is that cost, the
network's hash is `R / 0.00092` and the chip's margin is `1 - (0.000084 / e + capex / (L x 8,766)) / 0.00092`: 0.96
at a 5x edge, USD 0.5 per MH/s of silicon and a 3-year life, 0.86 at a 0.5-year life, 0.29 for the GDDR7 board at
USD 2.8 per MH/s and 0.5 years. The fleet capital is `q x hash x capex` paid at `T0` (under USD 10 M at every price
in the tables), with 20 percent residual at the end of life. Discount 10 percent a year, mid-year. The surface is
read as the break-even price `p*` (USD per IGN) with the launch-year miner revenue at that price in brackets (USD M a
year, 0.77 B IGN x `p*`); development is attractive where the price the entrant expects over the life exceeds `p*`.
**Table A. The operator, self-mining, a first design (bears the whole development cost).** Edge 5x, silicon USD
0.5 per MH/s. The 0.5-year life is the fixed-lane chip under the 180-day rotation; 3 years is the programmable chip's
default.
| `C_dev` | `T0` | `L` 0.5 y: `q` 0.1 / 0.3 / 1.0 | `L` 1 y | `L` 2 y | `L` 3 y |
|---|---|---|---|---|---|
| USD 20 M (a DRAM board) | 1 y | 0.77 / 0.26 / 0.08 (590 / 197 / 59) | 0.33 / 0.11 / 0.03 (256 / 85 / 26) | 0.22 / 0.07 / 0.02 (167 / 56 / 17) | 0.17 / 0.055 / 0.017 (128 / 43 / 13) |
| USD 20 M | 2 y | 1.69 / 0.56 / 0.17 | 0.73 / 0.24 / 0.07 | 0.36 / 0.12 / 0.04 | 0.29 / 0.10 / 0.03 (224 / 75 / 22) |
| USD 75 M | 1 y | 2.9 / 0.96 / 0.29 | 1.25 / 0.42 / 0.13 | 0.81 / 0.27 / 0.08 | 0.62 / 0.21 / 0.06 (479 / 160 / 48) |
| USD 75 M | 2 y | 6.3 / 2.1 / 0.63 | 2.7 / 0.92 / 0.28 | 1.35 / 0.45 / 0.14 | 1.09 / 0.36 / 0.11 (842 / 281 / 84) |
| USD 150 M (the SRAM die at N2) | 1 y | 5.7 / 1.9 / 0.58 | 2.5 / 0.83 / 0.25 | 1.63 / 0.54 / 0.16 | 1.24 / 0.41 / 0.12 (957 / 319 / 96) |
| USD 150 M | 2 y | 12.6 / 4.2 / 1.27 | 5.5 / 1.83 / 0.55 | 2.7 / 0.90 / 0.27 | 2.19 / 0.73 / 0.22 (1,684 / 561 / 168) |
| USD 500 M | 2 y | 42 / 14 / 4.2 | 18 / 6.1 / 1.83 | 9.0 / 3.0 / 0.90 | 7.3 / 2.4 / 0.73 (5,612 / 1,871 / 561) |
The axes the table does not show: the edge moves `p*` under 5 percent across 2x to 13x (the margin is 0.93 to 0.97
at every edge once the chip's power is a tenth of a GPU's, so the per-joule number is nearly irrelevant to the
decision); the silicon per MH/s matters only for short lives (at USD 2.8 per MH/s the 0.5-year chip never pays, the
1-year chip needs 4.7; at USD 5 neither the 0.5 nor the 1-year chip pays).
**Table B. The manufacturer selling hardware** (bears `C_dev`, takes half the operators' mining profit through the
unit price; operators bear the fleet and the power). `p*` in USD per IGN, edge 5x.
| `C_dev` | `T0` | `L` 1 y: `q` 0.3 / 1.0 | `L` 2 y | `L` 3 y |
|---|---|---|---|---|
| USD 20 M | 1 y | 0.24 / 0.07 | 0.15 / 0.045 | 0.11 / 0.034 |
| USD 75 M | 2 y | 1.95 / 0.59 | 0.93 / 0.28 | 0.75 / 0.22 |
| USD 150 M | 2 y | 3.9 / 1.17 | 1.86 / 0.56 | 1.50 / 0.45 |
| USD 500 M | 2 y | 13.0 / 3.9 | 6.2 / 1.86 | 5.0 / 1.50 |
**Table C. The revision entrant and the shared-cost entrant** (`q` 0.3, edge 5x): a second design at 0.3 x `C_dev`
and a 1-year `T0`; a design whose cost three entrants share, `T0` 2 years.
| First design `C_dev` | Revision, `L` 1 / 2 / 3 y | Shared by three, `L` 1 / 2 / 3 y |
|---|---|---|
| USD 75 M | 0.125 / 0.08 / 0.06 | 0.31 / 0.15 / 0.12 |
| USD 150 M | 0.25 / 0.16 / 0.12 | 0.61 / 0.30 / 0.24 |
| USD 500 M | 0.83 / 0.54 / 0.41 | 2.0 / 1.0 / 0.81 |
**Table D. The hybrid** (self-mines the first productive year at `q` 0.3, then sells at half the profit): `C_dev` 75 M,
`L` 2 / 3 years: `p*` 0.64 / 0.55 (USD 494 / 422 M a year); 150 M: 1.28 / 1.10 (988 / 845 M a year). It sits
between the operator and the manufacturer rows, as it should.
**Table E. NPV at the price axis** (USD M; `C_dev` 150 M, `T0` 2 y, `q` 0.3, edge 5x):
| `L` | IGN 0.03 | 0.10 | 0.30 | 1.00 | 3.00 |
|---|---|---|---|---|---|
| 0.5 y | -149 | -146 | -139 | -114 | -43 |
| 1 y | -148 | -142 | -125 | -68 | +96 |
| 2 y | -145 | -133 | -100 | +16 | +349 |
| 3 y | -144 | -129 | -88 | +56 | +467 |
**The conditions under which development is attractive, read off the surface:**
1. A chip project is attractive when the price its maker expects to hold over the productive life exceeds `p*`, and
`p*` scales almost exactly with `C_dev / (q x L_eff)` where `L_eff` is the discounted life after `T0`: 5x on the
project cost, 3x on the share from a third to the whole chain, 2x to 4x on the life from 1 to 3 years, 2x on the
pre-production period from 1 to 2 years, and under 5 percent on the per-joule edge from 2x to 13x.
2. The fixed-lane chip (`L` 0.5 years under the 180-day rotation) needs 4x the price of the programmable chip at
`L` 3 years for the same `C_dev`, and never pays at the DRAM board's USD 2.8 per MH/s; the rotation's value is
that factor, not a wall.
3. The cheapest attractive project at the launch-shape emission is a USD 20 M DRAM-board design that takes the whole
chain for three years: `p*` about USD 0.02 to 0.03 per IGN (USD 13 to 22 M a year of miner revenue). At a third
of the chain it needs 0.055 to 0.10 (43 to 75 M a year). The SRAM die at N2 (USD 150 M, `T0` 2 years) needs 0.22
taking the chain or 0.73 at a third over three years (168 to 561 M a year); a revision of it 0.12 to 0.16; one
shared by three entrants 0.24 to 0.30.
4. Selling hardware raises `p*` about 2x over self-mining at the same `C_dev` (the manufacturer keeps half the
profit), so the first entrant self-mines and sells only once the design is paid; the history's pattern.
5. What the chain controls in this surface is `L` (the rotation's 180 days against a fixed lane; nothing against a
programmable one), the honest cost per MH/s-hour that sets the revenue per hash (the operating point and the
dataset floor), and the share detector that makes `q` visible; it does not control `C_dev`, `T0` or `p`.
Unverified: the GPU all-in cost (the 5090 at MSRP; street prices double the capex term and halve every `p*`), the
half-profit split for hardware sales, the 20 percent residual, the 10 percent discount; the second cut adds GPU
re-entry at a higher price (which raises the hash and lowers the chip's revenue per MH/s only if GPU costs fall),
the per-year `q` path instead of a constant, and the emission beyond year 8.
## 4a. Amendment, 20:0x UK (review B's F05): the address coupling's cost on the chip, the literal 63-fold against the cached prefix tree
Review B's F05 (findings.json on review-b-landing): the connected-state class's full-chain window computes a
rotate-XOR fold over the 63 registers other than the source and XORs the source; it connects every register
syntactically, but a physical implementation need not reread and fold all 63 on each load. With `a[k] = ROL32(r[k],
63 - k)`, `S` the XOR of all `a[k]` and `P_s` the XOR of `a[k]` for `k < s`, the source expression is exactly `r[s] XOR
ROR32(P_s, 1) XOR S XOR P_s XOR a[s]` (the review checked 33,024 comparisons including 4,096 state updates with no
mismatch), and a prefix-XOR tree gives point updates and prefix queries in logarithmic time. The review's own
caveat: cached prefix state, port bandwidth and update costs must all be priced; this is not evidence the hash is
cheap. The chip-side cost of both designs at the adversary lane's PLACED per-lane-op figures (its mf core: 9.36 pJ per
lane-op at ASAP7 placed and routed, 6.55 at N5, 4.72 at N3; the window's gated file charging per write, the k lane's
+1.2 pJ per lane-op at N5 for the 64-register window; all modelled, no chip measured):
| Design | State per lane | Ports and traffic | Ops per hash for the fold (448-instruction text, 128 loads) | Energy per hash for the fold at N5 (6.55 pJ per lane-op) / N3 (4.72) | Against the class's shadow (55,296 lane-ops: 362 nJ at N5, 261 at N3) | Label |
|---|---|---|---|---|---|---|
| The literal 63-fold on every load | the 64 x 32-bit window only | 63 register reads per load (2,016 bits over the file's read ports, or a 63-cycle sequential read hidden by lane count) | 63 rotate-XORs x 128 loads = 8,064 | 52.8 nJ / 38.1 nJ | +14.6 percent of the chip's shadow energy; `E_chip` 0.466 + 0.362 + 0.053 = 0.881 microjoules (N5) | modelled |
| The cached prefix tree (a Fenwick tree over the 64 `a[k]`, plus the running `S`) | +64 x 32 bits of prefix state and one 32-bit `S`: 260 bytes per lane, the window doubled (the k lane's +1.2 pJ per lane-op window term scales with it, +1.2 at N5, approximate) | per instruction: log2(64) = 6 read-modify-writes of prefix nodes plus the `S` update (7 ops, 7 x 32 bits each way); per load: 6 node reads for `P_s`, the rotation, 4 XORs (11 ops) | 7 x 448 updates + 11 x 128 queries = 4,544 | 29.8 nJ / 21.4 nJ, plus the doubled window's +1.2 pJ on 55,296 ops: +66 nJ at N5 if the whole file widens, +0 if the prefix state sits in a separate gated macro written 7 times per instruction (the designer's choice; the latter is the cheaper and is carried) | +8.2 percent of the shadow; `E_chip` 0.858 microjoules (N5) | modelled |
| The difference | +260 B per lane | 7 ports' traffic per instruction against 63 reads per load | 1.8x fewer ops for the fold | 23 nJ per hash saved at N5 | **the chip's energy per hash falls 2.6 percent (0.881 to 0.858); the connected-state ratio of that lane's section 5 moves from 2.69x to 2.76x at the lock** (E_GPU 2.37 microjoules with the window at the lock, measured +1.6 percent) | modelled |
**Corrected 20:0x UK on the adversary lane's check against its placed core (its section 17 at 319e79c6):** the
Fenwick row above holds only if the prefix state is updated by the base program's writes alone. On the placed core
every register write costs the tree seven read-modify-writes, fourteen macro accesses per eight lanes at 3.5 pJ each
(2.0 to 7.0) plus seven xor-rotates, and the shadow block writes the same window about 800 times per load; updated on
every instruction the tree costs about 5 microjoules per hash, seven times the shadow, so the Fenwick tree never wins
on a macro-file core. The adversary lane's two cheaper forms: the literal 63-register fold at each load, 8,064
lane-ops and 63 SIMD macro reads per load shared by eight lanes (3.5 nJ per hash of accesses), 56 nJ per hash (48 to
72), +8.4 percent of the shadow, within the band of this file's 53; and the running total (the fold is linear over
GF(2): a write to register `j` changes it by a fixed rotation of old XOR new, and the old value is already read in the
core's phase 0), one 32-bit register per lane and an xor-rotate pair per instruction at about 1.0 pJ, 0.10 microjoules
per hash if applied on every instruction and 5 nJ if the shadow block's registers are exempt from the fold. The
corrected table:
| Form | Ops and state | Energy per hash for the fold (N5, placed) | Against the shadow | When it is the cheapest |
|---|---|---|---|---|
| The literal 63-fold | 8,064 lane-ops, no extra state | 56 nJ (48 to 72); this file's 53 | +8.4 percent | **at the shipped shape (the shadow writes the window)** |
| The Fenwick prefix tree | 4,544 fold ops, 260 B per lane, seven RMWs per write | 30 nJ on the base program's 512 writes; about 5 microjoules on every instruction's write | +8 percent, or 7x the shadow | never on a macro-file core |
| The running total | one 32-bit register per lane, an xor-rotate pair per write | 0.10 microjoules on every instruction; 5 nJ with the shadow exempt | +28 percent, or +1.4 percent | only if the shadow's registers are exempt from the fold, which the shipped class does not grant |
Reading, corrected: the adversary's cheapest form at the shipped shape is the literal fold, so this file's row stands at
+8.4 percent of the shadow and 2.6 percent of the GDDR7 board's energy per hash; the complete machine moves at most
from 1.5x to 1.42x to 1.49x node-for-node (1.7x to 1.8x a node ahead), inside the SRAM term's band, and the
connected-state verdict (KILL as a class, that lane's section 5: 1.08x node-for-node against a 1.25x gate) is
unchanged either way: the fold was never the chip's cost, the shadow's op count was. The GPU side: the measured window cost on the
card (+1.6 percent of energy per hash at the 5090's lock, +5.9 unlocked, that lane's section 4) was taken with the
compiler's own code for the fold, which may already hoist part of it; a byte-preserving incremental form on the GPU
would be the matching experiment and is the review's "cheapest alternative first". The review's required regressions
(native reference against incremental equivalence across every source register and real instruction updates; full
kernel equivalence, registers, spills, wall power and accepted throughput on the target cards; the adversary's design
carrying the prefix cost and never an assumed 63-read cost) are the connected-state lane's and the adversary lane's
to run; this file carries the price only, and the price says the class's verdict is unchanged by F05. Nothing
unmeasured and nonlinear is added in response, as the review asks.
## 5. Lever 4: everything else that reaches the die, checked
@ -393,25 +614,27 @@ file's lower threshold). The corrected floor moves the lower number 4x (from USD
| Home miner, one 8 GB card | Lever 1 costs it nothing at `W = 4` (measured on larger cards; its own row unmeasured) and under 5 percent at `W = 8` by the sector argument (unmeasured). Lever 2 retires it at the year-2 step (8.5 GiB) on this file's schedule, as on lane A's; a 6 GiB v6 epoch retires it two years earlier. Lever 3 is not its concern: no chip exists, and the first one needs IGN at USD 0.13 to 2.9 | the schedule is the founder's number; the `W = 8` measurement is a PC 1 job |
| One 12 GB card | holds to the year-4 step (11.5 GiB) with the cache freed after the build; out at 11.0 to 11.5 with it resident | the same |
| One 16 GB card (the 9070 XT, the 5070 Ti) | holds every step to 11.5 GiB; out at the 16 GiB step | the same |
| One 24 or 32 GB card, the 5090 at its knee | holds every step of the schedule; the 24 GB tier leaves at 16 GiB with the cache resident; the 32 GB card at 22 GiB. It pays the floor in joules at the knee: 4, 8 and 10 percent per hash at 2, 4 and 8 GiB (measured), about 9 to 11 percent across the schedule; unlocked 3 to 4 percent. Against the die with the shadow on at the lock (the card 1.67 microjoules plus F 0.65, measured; the die at `W = 4`): under the record's convention (`k` against the GPU's op cost at the same operating point, 6.2 pJ at the lock) 6.1x at `k = 0.5` and 3.3x at `k = 1`; under the absolute convention (the die's core costs what it costs, `k` fixed against the stock 11.3 pJ) 3.8x and 2.0x. The record's convention flatters the die at the lock by 1.6x; the absolute one is the physics and is what this file carries for the served line | the Ember knob carries the lock; the research lane picks the convention for its tables |
| One 24 or 32 GB card, the 5090 at its knee | holds every step of the schedule; the 24 GB tier leaves at 16 GiB with the cache resident; the 32 GB card at 22 GiB. It pays the floor in joules at the knee: 4, 8 and 10 percent per hash at 2, 4 and 8 GiB and 14.3 percent at the 5.5 GiB step itself (all measured), 15 to 18 percent read for the 8.5 and 11.5 steps (unmeasured past 8 GiB); unlocked 3 to 5 percent. A tuned 5090 pays more for the floor than for any width or shadow draw in the band. Against the die with the shadow on at the lock (the card 1.67 microjoules plus F 0.65, measured; the die at `W = 4`): under the record's convention (`k` against the GPU's op cost at the same operating point, 6.2 pJ at the lock) 6.1x at `k = 0.5` and 3.3x at `k = 1`; under the absolute convention (the die's core costs what it costs, `k` fixed against the stock 11.3 pJ) 3.8x and 2.0x. The record's convention flatters the die at the lock by 1.6x; the absolute one is the physics and is what this file carries for the served line | the Ember knob carries the lock; the research lane picks the convention for its tables |
| The unified-memory SoC | a 16 GB Mac leaves at the year-2 step; 32 GB at the 16 GiB step; every Mac pays -21 to -25 percent of rate from the v6 epoch on (measured to 8 GiB). Against the die at class v4 and `k = 0.5` the M5 Max reads 3.6x to 4.0x | finding 3 of lane B: the reference joule |
| A rig | nothing changes until a project pays (section 4); the 180-day rotation and the floor do not move the chip's per-MH/s cost | the issuance trigger and the detector, as the record |
| A pool user | the chip fleet that matters is one package; the share-pattern detector is the warning | the detector before the public testnet |
| The public claim | "under 2x" is not reachable against the die by any shadow: on the k lane's synthesised core 10x at the honest cards' whole latency shadow at stock, 6x to 7x at the 5090's knee, 18x to 21x under class v4; on the record's claimed band (marked) 2x at `k = 1` and 4x at `k = 0.5`; 20x to 60x without the shadow. The number a miner can act on is the project's price threshold: USD 0.13 to 2.9 per IGN | section 7's served sentence |
| The public claim | On the k lane's synthesised programmable core (pre-place) the die reads 3.4x to 3.5x at the honest cards' whole latency shadow at stock and 2.2x at the 5090's knee (N3; an N2 core 4.7x and 2.9x), 7.7x to 8.1x under class v4 at stock; the bare-lane floor (10x at the full shadow, 18x to 21x under class v4) is the marked worst case; 20x to 60x without the shadow. "Under 2x" is reached only at the knee under the full shadow on an N3 core. The number a miner can act on is the project's price threshold: USD 0.13 to 2.9 per IGN | section 7's served sentence |
## 7. Recommendation, for the 15:45 close
The lever is the read width: pin `W = 4` words at genesis on the measured rows (the 5090 +2.7 percent, the 9070 XT
-1.4, the M5 Max within 1) and move to `W = 8` once one PC 1 row confirms what the measured rows already say (one
sector per load on NVIDIA, one line on AMD, Apple flat to w64), which moves the die from 66x to 31x to 44x at zero
shadow and by 0.2x to 0.4x with the shadow on, and keep the per-lane scratch resident so the lane stays pinned. The schedule is 5.5 GiB at the v6 epoch, 8.5 two years on and
The lever is the read width: pin `W = 4` words at genesis on the measured rows (the 5090 +2.7 percent of rate, the
9070 XT -1.4, the M5 Max within 1), which moves the die from 66x to 44x at zero shadow and nothing with the shadow
on; `W = 8` is measured rate-free but 4.8 percent dearer in energy on the 5090 at stock and buys nothing with the
shadow on, so it does not pin unless its lock row reverses that; `W = 16` never. Keep the per-lane scratch resident
so the lane stays pinned. The schedule is 5.5 GiB at the v6 epoch, 8.5 two years on and
11.5 at four, each just under a card tier's room and just over a reticle multiple (3, 5 and 6 dies; USD 1,500, 2,500
and 3,000 of silicon), and USD 5,000 is not recommended because it is 20 GiB and retires every card under 32 GB and
every Mac under 64 GB whenever it lands. The served claim should say that the strongest chip we can model for 2027
to 2028, an SRAM store on 2 nm, would reach about 6x to 10x per joule against an RTX 5090 paying its whole latency
shadow with a core at the synthesised cost of a 3 nm lane (2x to 4x on the claimed band the record carried, marked),
20x to 60x without that shadow, and that such a project pays only above about USD 0.4 per IGN at a third of the
network or USD 0.13 for a maker who takes the whole chain, which is the pattern the share detector watches for. The floor is a ticket (USD 500 to 3,000 of silicon per
to 2028, an SRAM store on 2 nm, would reach about 3.5x per joule against an RTX 5090 at stock paying its whole latency
shadow and about 2.2x against one locked at its knee, with a programmable core at the synthesised cost of a 3 nm
lane (about 4.7x and 2.9x with a 2 nm core; up to 10x on the bare-lane floor, marked as the worst case), 20x to 60x
without that shadow, and that such a project pays only above about USD 0.4 per IGN at a third of the network or USD
0.13 for a maker who takes the whole chain, which is the pattern the share detector watches for. The floor is a ticket (USD 500 to 3,000 of silicon per
store across the schedule) and not a per-joule or per-MH/s lever, and the text should not claim otherwise. Nothing in
this file moves a served number; the `W = 8` row and the Apple rate curve past 8 GiB are the two measurements owed.
@ -420,10 +643,12 @@ this file moves a served number; the `W = 8` row and the Apple rate curve past 8
- Every chip-side figure is modelled; the macro access (0.1 nJ, width-independent) and the wire (1.3 pJ per bit at 24
mm) are the record's approximations and move every zero-shadow row by 2x either way; the shadowed rows move under
0.3x.
- `W = 8` has no row of its own: its card cost is read by interpolation of the measured w16 and w64 rows (one sector
per load on NVIDIA, one line on AMD, the M5 Max flat at 1.03 through w64). One PC 1 row at stock and at the lock
confirms it; the crate's width set is {1, 4, 16} words, so the pack needs a generator and emitter line first (the
hash lane's caveat, 13:0x UK). Not run by this lane (no builds on the Mac; no GPU on build-3 or build-4).
- `W = 8` is measured at stock (15:03 UK: rate held, +4.8 percent energy; run b +5.6) and at the 1,300 MHz lock
(run b, 17:4x UK: +11.2 percent per hash); `W = 16` +70 percent at stock and +78 at the lock. This file's earlier
interpolation ("free") was wrong on the energy and is withdrawn; the width pin is `W = 4` on measured rows at both
operating points, and nothing is owed on it. Every PC 1 row since the 14:2x eGPU swap reads about 13 percent under the morning's rate on the same
packs (a host-side change, the PCIe link to be read), so rows compare within a job only. The 9070 XT and the M5 Max
have no W = 8 row.
- The Apple rate curve past 8 GiB (the 11.5 and 16 GiB steps): unmeasured; read as -22 to -25 percent.
- The 5090's size rows at 2, 4 and 8 GiB landed at 12:58 UK (the hash lane, PC 1, label ca4-v6) and are in section 3.2;
the 9070 XT and the smaller NVIDIA tiers have no size rows (their page reach is read as the 5090's, approximate).

25
docs/analysis/class-v6/logs/census-packs/addrsite/v5-seed18-x16.tsv Normal file
View file

@ -0,0 +1,25 @@
# addrsite: program f1ca3ca995c57b7a class mx8-erad810f22d+sh256x27+state attempt 0 generator v5 day 2026-10-03 dataset 2^28 words (memory-hard) headers 4 warps 32768 nonces/header 1048576 threads 32
# per (iteration, site) cell: the most read address and its share (A03's form); cells over 0.1 percent listed
it site instr reads distinct top_addr top_count top_share
0 0 3 4194304 4129509 0x0e6884ac 4 0.000001
0 1 6 4194304 4066280 0x058b9a8a 4 0.000001
# worst cell: iteration 0 site 10 address 0x0adbc6e5 in 5 reads, share 0.000001
# per site, pooled over the 8 iterations (hit rates: an ideal cache of the site's top K items, K = bytes / 64; exp = min(1, K / window items))
site instr win window_items reads distinct top_addr top_count top_share hit_1GiB hit_256MiB hit_64MiB exp_1GiB exp_256MiB exp_64MiB
0 3 1 8388608 33554432 29690148 0x0ccee0cd 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
1 6 2 4194304 33554432 26405457 0x056b9fd1 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
2 11 0 16777216 33554432 31543021 0x0b25ed16 6 0.000000 1.0000 0.4839 0.1625 1.0000 0.2500 0.0625
3 18 0 16777216 33554432 31544239 0x0fc85a59 5 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
4 27 0 16777216 33554432 31541959 0x06a04d7c 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
5 34 2 4194304 33554432 26406629 0x0f67e2d5 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
6 37 0 16777216 33554432 31543195 0x032edc74 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
7 40 1 8388608 33554432 29688023 0x08eacd44 7 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
8 41 0 16777216 33554432 31544547 0x0614c7e4 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
9 42 0 16777216 33554432 31539943 0x09d0f913 6 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
10 48 2 4194304 33554432 26408669 0x0a2b7cc3 9 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
11 50 2 4194304 33554432 26406921 0x078c4109 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
12 55 2 4194304 33554432 26407721 0x08e928d8 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
13 57 1 8388608 33554432 29689588 0x0e83878c 6 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
14 60 0 16777216 33554432 31543105 0x046b44e4 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
15 62 2 4194304 33554432 26405478 0x0042b970 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
# program: reads 536870912, distinct items 16777216, worst site 10 address 0x0a2b7cc3 in 9 reads (share 0.000000); one cache over the program's reads: 1 GiB 1.0000, 256 MiB 0.3447, 64 MiB 0.0988; the per-site ideal caches summed: 1.0000, 0.7171, 0.2526; 253.6 s
Can't render this file because it has a wrong number of fields in line 3.

25
docs/analysis/class-v6/logs/census-packs/addrsite/v5-seed19-x16.tsv Normal file
View file

@ -0,0 +1,25 @@
# addrsite: program 116d921a2211b1e2 class mx8-erad810f22d+sh256x27+state attempt 5 generator v5 day 2026-10-03 dataset 2^28 words (memory-hard) headers 4 warps 32768 nonces/header 1048576 threads 32
# per (iteration, site) cell: the most read address and its share (A03's form); cells over 0.1 percent listed
it site instr reads distinct top_addr top_count top_share
0 0 5 4194304 4065678 0x038f5d3d 5 0.000001
0 1 7 4194304 4129564 0x0b33f0b8 4 0.000001
# worst cell: iteration 2 site 5 address 0x00c909c0 in 6 reads, share 0.000001
# per site, pooled over the 8 iterations (hit rates: an ideal cache of the site's top K items, K = bytes / 64; exp = min(1, K / window items))
site instr win window_items reads distinct top_addr top_count top_share hit_1GiB hit_256MiB hit_64MiB exp_1GiB exp_256MiB exp_64MiB
0 5 2 4194304 33554432 26387138 0x02a31bfe 9 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
1 7 1 8388608 33554432 29690800 0x0d592300 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
2 8 1 8388608 33554432 29690555 0x0c5c8fd2 7 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
3 20 2 4194304 33554432 26404161 0x00e867f2 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
4 23 1 8388608 33554432 29686249 0x03e57fd7 7 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
5 25 2 4194304 33554432 26406824 0x0069d880 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
6 34 1 8388608 33554432 29689811 0x0bf7316e 6 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
7 36 2 4194304 33554432 26405255 0x021ba3b2 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
8 37 2 4194304 33554432 26403259 0x073f8e5d 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
9 40 1 8388608 33554432 29689189 0x0e7354d2 8 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
10 43 2 4194304 33554432 26406153 0x0f506142 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
11 45 0 16777216 33554432 31542555 0x0c5ffb53 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
12 46 2 4194304 33554432 26402430 0x0a4d29f3 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
13 48 1 8388608 33554432 29684129 0x07b2bb0a 6 0.000000 1.0000 0.6954 0.2365 1.0000 0.5000 0.1250
14 51 2 4194304 33554432 26406067 0x0a7f2347 9 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
15 59 2 4194304 33554432 26404385 0x0be2218d 9 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
# program: reads 536870912, distinct items 16777216, worst site 0 address 0x02a31bfe in 9 reads (share 0.000000); one cache over the program's reads: 1 GiB 1.0000, 256 MiB 0.3685, 64 MiB 0.1035; the per-site ideal caches summed: 1.0000, 0.8535, 0.3045; 245.3 s
Can't render this file because it has a wrong number of fields in line 3.

41
docs/analysis/class-v6/logs/census-packs/addrsite/v6-seed18-x16.tsv Normal file
View file

@ -0,0 +1,41 @@
# addrsite: program 7202548bfc7f98e9 class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 0 generator v5 day 2026-10-03 dataset 2^28 words (memory-hard) headers 4 warps 32768 nonces/header 1048576 threads 32
# per (iteration, site) cell: the most read address and its share (A03's form); cells over 0.1 percent listed
it site instr reads distinct top_addr top_count top_share
0 0 3 4194304 4129479 0x08e347f5 4 0.000001
0 1 3 4194304 4129572 0x0c06f9a4 4 0.000001
# worst cell: iteration 5 site 24 address 0x0aac567d in 6 reads, share 0.000001
# per site, pooled over the 8 iterations (hit rates: an ideal cache of the site's top K items, K = bytes / 64; exp = min(1, K / window items))
site instr win window_items reads distinct top_addr top_count top_share hit_1GiB hit_256MiB hit_64MiB exp_1GiB exp_256MiB exp_64MiB
0 3 1 8388608 33554432 29689019 0x0deb06b7 7 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
1 3 1 8388608 33554432 29689729 0x0ada2fc4 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
2 6 2 4194304 33554432 26405765 0x0760aefa 7 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
3 6 2 4194304 33554432 26406636 0x07c4d05a 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
4 11 0 16777216 33554432 31541448 0x05c28c64 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
5 11 0 16777216 33554432 31542278 0x04f48b4f 6 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
6 18 0 16777216 33554432 31541498 0x000bc4bc 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
7 18 0 16777216 33554432 31542500 0x00b738d2 5 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
8 27 0 16777216 33554432 31540858 0x0fa18223 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
9 27 0 16777216 33554432 31539145 0x037f6563 5 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
10 34 2 4194304 33554432 26402753 0x0eecd61a 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
11 34 2 4194304 33554432 26405703 0x0eeec3d9 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
12 37 0 16777216 33554432 31541771 0x094de445 5 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
13 37 0 16777216 33554432 31542978 0x002bd9f0 5 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
14 40 1 8388608 33554432 29689644 0x0df000e0 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
15 40 1 8388608 33554432 29691134 0x0be554c4 6 0.000000 1.0000 0.6953 0.2364 1.0000 0.5000 0.1250
16 41 0 16777216 33554432 31543936 0x0f528b80 6 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
17 41 0 16777216 33554432 31539898 0x00ee77fb 6 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
18 42 0 16777216 33554432 31540916 0x0056a1bd 5 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
19 42 0 16777216 33554432 31541048 0x0100a7e9 5 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
20 48 2 4194304 33554432 26403608 0x0b8db743 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
21 48 2 4194304 33554432 26404624 0x094a23d8 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
22 50 2 4194304 33554432 26406113 0x0558a956 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
23 50 2 4194304 33554432 26407223 0x053efa2e 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
24 55 2 4194304 33554432 26404148 0x098834b1 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
25 55 2 4194304 33554432 26399829 0x09bd70df 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
26 57 1 8388608 33554432 29687219 0x08c6e7ef 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
27 57 1 8388608 33554432 29689432 0x0c744ea8 6 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
28 60 0 16777216 33554432 31543251 0x070a54cd 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
29 60 0 16777216 33554432 31544208 0x0d02d4ec 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
30 62 2 4194304 33554432 26405366 0x0247a73f 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
31 62 2 4194304 33554432 26405501 0x03316e03 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
# program: reads 1073741824, distinct items 16777216, worst site 3 address 0x07c4d05a in 8 reads (share 0.000000); one cache over the program's reads: 1 GiB 1.0000, 256 MiB 0.3350, 64 MiB 0.0936; the per-site ideal caches summed: 1.0000, 0.7171, 0.2526; 476.8 s
Can't render this file because it has a wrong number of fields in line 3.

41
docs/analysis/class-v6/logs/census-packs/addrsite/v6-seed19-x16.tsv Normal file
View file

@ -0,0 +1,41 @@
# addrsite: program dd6b285ae45e7b9b class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 1 generator v5 day 2026-10-03 dataset 2^28 words (memory-hard) headers 4 warps 32768 nonces/header 1048576 threads 32
# per (iteration, site) cell: the most read address and its share (A03's form); cells over 0.1 percent listed
it site instr reads distinct top_addr top_count top_share
0 0 5 4194304 4162021 0x04a8794d 3 0.000001
0 1 5 4194304 4161582 0x005ae5cc 3 0.000001
# worst cell: iteration 4 site 29 address 0x063d7266 in 6 reads, share 0.000001
# per site, pooled over the 8 iterations (hit rates: an ideal cache of the site's top K items, K = bytes / 64; exp = min(1, K / window items))
site instr win window_items reads distinct top_addr top_count top_share hit_1GiB hit_256MiB hit_64MiB exp_1GiB exp_256MiB exp_64MiB
0 5 0 16777216 33554432 31542070 0x0cb5e8e9 5 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
1 5 0 16777216 33554432 31542481 0x0d7d8f79 5 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
2 7 0 16777216 33554432 31540989 0x0ab53bfd 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
3 7 0 16777216 33554432 31541158 0x082ff901 5 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
4 9 0 16777216 33554432 31542223 0x04527b7c 6 0.000000 1.0000 0.4841 0.1626 1.0000 0.2500 0.0625
5 9 0 16777216 33554432 31540778 0x0756b940 5 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
6 19 1 8388608 33554432 29689935 0x020c7d00 7 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
7 19 1 8388608 33554432 29689915 0x04817391 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
8 20 2 4194304 33554432 26408371 0x01110531 9 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
9 20 2 4194304 33554432 26407942 0x023bd59f 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
10 25 1 8388608 33554432 29692542 0x076f96d9 7 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
11 25 1 8388608 33554432 29688307 0x0310068d 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
12 26 0 16777216 33554432 31540886 0x0c1976e2 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
13 26 0 16777216 33554432 31544918 0x04747772 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
14 30 1 8388608 33554432 29689661 0x0ff00d44 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
15 30 1 8388608 33554432 29686510 0x09b5ebd7 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
16 33 0 16777216 33554432 31541134 0x0bc0faa1 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
17 33 0 16777216 33554432 31541146 0x01b87eb8 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
18 42 2 4194304 33554432 26406681 0x026c2dc5 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
19 42 2 4194304 33554432 26402342 0x00c31321 9 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
20 44 1 8388608 33554432 29691469 0x03be570e 7 0.000000 1.0000 0.6954 0.2364 1.0000 0.5000 0.1250
21 44 1 8388608 33554432 29689278 0x062f3a11 7 0.000000 1.0000 0.6953 0.2364 1.0000 0.5000 0.1250
22 47 2 4194304 33554432 26404322 0x057c28b5 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
23 47 2 4194304 33554432 26406025 0x051b36f8 7 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
24 52 1 8388608 33554432 29693023 0x0c6b67d3 7 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
25 52 1 8388608 33554432 29687741 0x0830221a 6 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
26 53 1 8388608 33554432 29689829 0x03a1abed 7 0.000000 1.0000 0.6954 0.2363 1.0000 0.5000 0.1250
27 53 1 8388608 33554432 29690972 0x0234bd46 7 0.000000 1.0000 0.6953 0.2363 1.0000 0.5000 0.1250
28 55 2 4194304 33554432 26404287 0x04753640 8 0.000000 1.0000 1.0000 0.3657 1.0000 1.0000 0.2500
29 55 2 4194304 33554432 26405954 0x07530caa 8 0.000000 1.0000 1.0000 0.3658 1.0000 1.0000 0.2500
30 62 0 16777216 33554432 31540461 0x027cd506 6 0.000000 1.0000 0.4840 0.1626 1.0000 0.2500 0.0625
31 62 0 16777216 33554432 31543136 0x0578c56a 6 0.000000 1.0000 0.4840 0.1625 1.0000 0.2500 0.0625
# program: reads 1073741824, distinct items 16777216, worst site 8 address 0x01110531 in 9 reads (share 0.000000); one cache over the program's reads: 1 GiB 1.0000, 256 MiB 0.3730, 64 MiB 0.1013; the per-site ideal caches summed: 1.0000, 0.6923, 0.2410; 510.2 s
Can't render this file because it has a wrong number of fields in line 3.

259
docs/analysis/class-v6/logs/census-packs/hl-v6-all-g6/census.tsv Normal file
View file

@ -0,0 +1,259 @@
# census class mx8+sh256x27+state+reg64c+fold+rw eras [0] seeds 256 era-widths 4 state /srv/builds/v6-census/packs/node1-state.igsd1 erahex edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 weights base bittest off bin be3e32dcaa3d958f start 2026-10-08T19:30:57Z host igneum-build-4 threads 16
cand era seed candidates accepted attempt bit_z bit_site bit_bit reasons
census 0 9 1 1 0 2.8 6 25
census 0 3 1 1 0 2.8 14 3
census 0 2 1 1 0 3.5 3 15
census 0 1 1 1 0 nan - -
census 0 0 1 1 0 nan - -
census 0 7 1 1 0 3.4 1 23
census 0 4 1 1 0 2.9 15 4
census 0 12 1 1 0 3.2 9 13
census 0 20 1 1 0 3.2 10 7
census 0 8 1 1 0 2.8 9 12
census 0 6 1 1 0 nan - -
census 0 11 1 1 0 nan - -
census 0 16 1 1 0 nan - -
census 0 17 1 1 0 nan - -
census 0 22 1 1 0 3.1 6 11
census 0 21 2 1 1 3.5 0 18 reg64=1;
census 0 14 1 1 0 3.5 12 21
census 0 25 2 1 1 nan - - a=1;
census 0 30 1 1 0 nan - -
census 0 29 1 1 0 nan - -
census 0 32 1 1 0 4.4 10 21
census 0 10 1 1 0 3.0 0 2
census 0 5 1 1 0 nan - -
census 0 18 1 1 0 nan - -
census 0 15 1 1 0 nan - -
census 0 13 1 1 0 2.8 12 13
census 0 28 1 1 0 nan - -
census 0 23 1 1 0 nan - -
census 0 26 1 1 0 3.2 7 9
census 0 24 1 1 0 3.7 13 12
census 0 36 2 1 1 3.2 14 3 b=1;
census 0 33 1 1 0 nan - -
census 0 34 1 1 0 3.0 12 1
census 0 43 1 1 0 nan - -
census 0 42 2 1 1 nan - - a'=1;
census 0 39 1 1 0 nan - -
census 0 44 2 1 1 nan - - a'=1;
census 0 45 1 1 0 nan - -
census 0 19 1 1 0 nan - -
census 0 27 1 1 0 3.0 9 8
census 0 41 1 1 0 nan - -
census 0 37 1 1 0 3.1 13 8
census 0 52 1 1 0 nan - -
census 0 49 1 1 0 nan - -
census 0 46 1 1 0 4.4 2 3
census 0 53 2 1 1 nan - - b=1;
census 0 31 2 1 1 3.2 1 5 b=1;
census 0 51 2 1 1 nan - - reg64=1;
census 0 38 1 1 0 nan - -
census 0 35 1 1 0 4.2 12 12
census 0 47 1 1 0 3.1 4 16
census 0 48 1 1 0 nan - -
census 0 57 1 1 0 3.5 6 24
census 0 55 1 1 0 nan - -
census 0 56 2 1 1 2.8 2 9 a'=1;
census 0 62 1 1 0 3.2 9 19
census 0 60 1 1 0 nan - -
census 0 50 1 1 0 3.3 9 9
census 0 59 1 1 0 nan - -
census 0 67 2 1 1 nan - - a=1;
census 0 69 1 1 0 3.2 14 25
census 0 66 1 1 0 3.4 6 11
census 0 72 1 1 0 nan - -
census 0 71 2 1 1 2.8 7 17 a=1;
census 0 74 1 1 0 3.0 4 6
census 0 73 1 1 0 nan - -
census 0 63 1 1 0 nan - -
census 0 78 1 1 0 nan - -
census 0 54 1 1 0 3.5 3 13
census 0 40 1 1 0 2.8 9 25
census 0 58 1 1 0 3.0 8 17
census 0 61 1 1 0 3.3 3 11
census 0 80 2 1 1 nan - - a'=1;
census 0 76 1 1 0 3.2 2 23
census 0 81 1 1 0 nan - -
census 0 64 1 1 0 3.6 9 8
census 0 82 1 1 0 nan - -
census 0 70 1 1 0 nan - -
census 0 77 1 1 0 nan - -
census 0 79 1 1 0 nan - -
census 0 75 1 1 0 2.9 4 16
census 0 83 1 1 0 nan - -
census 0 84 1 1 0 nan - -
census 0 65 1 1 0 nan - -
census 0 68 1 1 0 nan - -
census 0 91 1 1 0 3.2 13 8
census 0 94 1 1 0 nan - -
census 0 95 2 1 1 3.0 13 12 a'=1;
census 0 89 1 1 0 nan - -
census 0 87 2 1 1 nan - - a'=1;
census 0 88 1 1 0 nan - -
census 0 90 2 1 1 nan - - b=1;
census 0 92 1 1 0 nan - -
census 0 96 1 1 0 3.1 8 7
census 0 93 1 1 0 3.0 7 20
census 0 86 1 1 0 3.0 7 19
census 0 101 1 1 0 nan - -
census 0 98 1 1 0 nan - -
census 0 100 1 1 0 nan - -
census 0 99 1 1 0 3.1 2 23
census 0 103 1 1 0 nan - -
census 0 104 1 1 0 nan - -
census 0 106 1 1 0 nan - -
census 0 108 1 1 0 nan - -
census 0 114 1 1 0 2.9 7 8
census 0 115 2 1 1 nan - - b=1;
census 0 97 1 1 0 nan - -
census 0 105 1 1 0 nan - -
census 0 109 1 1 0 nan - -
census 0 110 1 1 0 3.4 7 15
census 0 117 1 1 0 nan - -
census 0 116 1 1 0 3.1 1 5
census 0 121 1 1 0 nan - -
census 0 122 1 1 0 nan - -
census 0 123 1 1 0 nan - -
census 0 124 1 1 0 3.6 0 15
census 0 128 1 1 0 nan - -
census 0 112 1 1 0 3.3 5 4
census 0 130 1 1 0 3.2 11 8
census 0 126 1 1 0 4.3 8 3
census 0 131 1 1 0 3.3 5 21
census 0 107 1 1 0 nan - -
census 0 125 2 1 1 3.3 1 6 b=1;
census 0 85 1 1 0 3.1 14 24
census 0 120 1 1 0 nan - -
census 0 127 2 1 1 nan - - a=1;
census 0 129 2 1 1 3.2 6 6 a'=1;
census 0 102 1 1 0 3.5 0 13
census 0 118 1 1 0 nan - -
census 0 119 1 1 0 3.1 3 18
census 0 133 1 1 0 2.8 10 0
census 0 135 1 1 0 3.0 5 15
census 0 134 1 1 0 3.7 11 4
census 0 137 1 1 0 nan - -
census 0 138 1 1 0 2.7 12 19
census 0 146 1 1 0 3.4 15 22
census 0 147 1 1 0 nan - -
census 0 113 1 1 0 nan - -
census 0 132 1 1 0 3.4 15 1
census 0 136 1 1 0 nan - -
census 0 111 1 1 0 3.3 3 27
census 0 139 1 1 0 3.4 11 3
census 0 140 2 1 1 nan - - a'=1;
census 0 141 1 1 0 nan - -
census 0 151 1 1 0 nan - -
census 0 150 1 1 0 nan - -
census 0 142 2 1 1 4.0 6 0 a'=1;
census 0 143 1 1 0 nan - -
census 0 149 1 1 0 nan - -
census 0 144 1 1 0 3.7 3 22
census 0 157 1 1 0 3.5 1 16
census 0 159 1 1 0 nan - -
census 0 162 1 1 0 3.1 2 16
census 0 156 1 1 0 nan - -
census 0 160 1 1 0 2.5 1 17
census 0 164 1 1 0 3.8 7 25
census 0 145 1 1 0 2.7 14 24
census 0 161 1 1 0 nan - -
census 0 165 1 1 0 3.0 11 5
census 0 153 1 1 0 nan - -
census 0 158 1 1 0 3.1 13 1
census 0 148 1 1 0 3.7 10 0
census 0 154 2 1 1 3.3 12 17 a'=1;
census 0 163 2 1 1 3.7 0 19 a'=1;
census 0 155 1 1 0 nan - -
census 0 167 3 1 2 3.4 2 14 a'=2;
census 0 173 3 1 2 nan - - a'=2;
census 0 171 1 1 0 nan - -
census 0 166 1 1 0 nan - -
census 0 175 1 1 0 2.9 1 24
census 0 170 1 1 0 nan - -
census 0 169 2 1 1 nan - - a'=1;
census 0 172 2 1 1 3.0 15 16 a'=1;
census 0 177 1 1 0 3.1 14 0
census 0 176 1 1 0 nan - -
census 0 168 1 1 0 3.6 6 25
census 0 183 1 1 0 nan - -
census 0 188 1 1 0 3.2 0 12
census 0 174 1 1 0 nan - -
census 0 180 1 1 0 3.5 3 19
census 0 181 1 1 0 3.1 1 14
census 0 186 2 1 1 nan - - a'=1;
census 0 191 1 1 0 nan - -
census 0 185 1 1 0 nan - -
census 0 184 2 1 1 3.0 2 25 a'=1;
census 0 189 1 1 0 nan - -
census 0 192 1 1 0 4.1 3 23
census 0 152 1 1 0 3.1 11 25
census 0 182 1 1 0 nan - -
census 0 187 1 1 0 3.6 4 14
census 0 179 1 1 0 3.1 10 14
census 0 190 2 1 1 3.3 14 13 a'=1;
census 0 196 1 1 0 3.0 3 21
census 0 195 1 1 0 2.9 2 10
census 0 194 1 1 0 3.7 2 6
census 0 200 1 1 0 3.1 5 21
census 0 198 1 1 0 nan - -
census 0 201 1 1 0 nan - -
census 0 197 1 1 0 nan - -
census 0 193 1 1 0 3.5 4 23
census 0 199 1 1 0 2.9 14 0
census 0 178 1 1 0 4.0 14 14
census 0 213 1 1 0 2.8 13 3
census 0 203 1 1 0 nan - -
census 0 206 2 1 1 nan - - a'=1;
census 0 205 1 1 0 3.1 7 7
census 0 214 1 1 0 3.7 7 24
census 0 215 1 1 0 3.0 12 7
census 0 202 1 1 0 nan - -
census 0 204 1 1 0 3.8 8 24
census 0 208 1 1 0 nan - -
census 0 209 1 1 0 nan - -
census 0 211 1 1 0 nan - -
census 0 207 1 1 0 nan - -
census 0 210 1 1 0 3.3 1 20
census 0 216 1 1 0 3.5 7 4
census 0 217 1 1 0 nan - -
census 0 223 1 1 0 nan - -
census 0 225 1 1 0 3.2 14 23
census 0 229 1 1 0 3.0 5 17
census 0 226 1 1 0 3.6 12 11
census 0 221 1 1 0 nan - -
census 0 227 1 1 0 3.4 14 24
census 0 224 1 1 0 3.0 12 11
census 0 219 1 1 0 3.0 9 22
census 0 222 1 1 0 3.0 14 10
census 0 220 1 1 0 nan - -
census 0 218 2 1 1 2.9 0 7 a'=1;
census 0 228 1 1 0 3.0 8 9
census 0 232 1 1 0 3.4 4 15
census 0 237 1 1 0 3.1 7 12
census 0 212 1 1 0 nan - -
census 0 230 1 1 0 nan - -
census 0 241 1 1 0 nan - -
census 0 231 1 1 0 2.9 12 13
census 0 238 1 1 0 nan - -
census 0 240 2 1 1 3.4 11 17 b=1;
census 0 243 1 1 0 3.4 5 5
census 0 242 1 1 0 nan - -
census 0 236 1 1 0 nan - -
census 0 234 1 1 0 nan - -
census 0 245 1 1 0 3.0 13 2
census 0 244 2 1 1 3.0 3 8 reg64=1;
census 0 246 2 1 1 nan - - a'=1;
census 0 248 1 1 0 3.8 0 16
census 0 235 1 1 0 3.3 8 16
census 0 239 1 1 0 3.4 6 18
census 0 247 1 1 0 nan - -
census 0 254 1 1 0 nan - -
census 0 251 1 1 0 3.0 11 0
census 0 253 1 1 0 nan - -
census 0 233 2 1 1 3.2 7 11 c=1;
census 0 255 1 1 0 2.6 6 14
census 0 250 1 1 0 3.9 14 21
census 0 249 1 1 0 nan - -
census 0 252 1 1 0 3.0 0 24
# end 2026-10-08T19:33:50Z rows=257
Can't render this file because it has a wrong number of fields in line 2.

1
docs/analysis/class-v6/logs/census-packs/hl-v6-all-g6/run.txt Normal file
View file

@ -0,0 +1 @@
# pack hl-v6-all-g6 class mx8+sh256x27+state+reg64c+fold+rw epoch edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 day 69676e65756d2d6461792ffa50000000000000 era edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 state /srv/builds/v6-census/packs/node1-state.igsd1 day-index 20730 era-widths 4 bin be3e32dcaa3d958f f8 host igneum-build-4 start 2026-10-08T19:30:54Z

67
docs/analysis/class-v6/logs/census-packs/hl-v6-all-g6/show.txt Normal file
View file

@ -0,0 +1,67 @@
seed "igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07" generator v6 class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 0 program id 4de7b836cc40a4ea seed words 667d0fbd 7b8e5963 31c67e5e 4529ddc6 ef19d6d8 accf6211 da0aed32 abc6df31
op mix load=16 add=10 xor=10 mad=7 shfl=7 rotl=6 rotr=3 sub=3 mul=2 loads/hash 128 bytes/hash 512
era d810f22d (igneum-era/0/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07): width 4 B, stride mul 0x9ad30d99 rot 29, interleave [0, 2, 12, 13], windows (site:shrink:offset) 7:1:1 8:1:1 9:0:0 10:1:1 11:2:1 13:1:0 29:0:0 30:1:0 31:1:0 44:1:0 46:1:0 47:2:0 52:0:0 56:0:0 58:1:1 63:0:0
0: rotl dst=5 src=4 src2=3 imm=0xde0bea4e imm2=0xbaab6229 rot=2 bit=25 mask=2
1: shfl dst=2 src=6 src2=5 imm=0xca149f6a imm2=0x5b623116 rot=4 bit=16 mask=8
2: sub dst=1 src=6 src2=5 imm=0x512ddc79 imm2=0xe99741c7 rot=13 bit=21 mask=2
3: rotr dst=1 src=5 src2=7 imm=0x7819f5ec imm2=0x673c2157 rot=17 bit=21 mask=2
4: mad dst=1 src=7 src2=5 imm=0x7826a6a7 imm2=0x946f7818 rot=18 bit=21 mask=8
5: xor dst=0 src=2 src2=0 imm=0xc38c81bc imm2=0x5f6a0ed2 rot=5 bit=19 mask=16
6: shfl dst=7 src=0 src2=4 imm=0x80c84659 imm2=0x5c61dcf7 rot=3 bit=20 mask=4
7: load dst=4 src=2 src2=4 imm=0x21579975 imm2=0x85668475 rot=6 bit=3 mask=8
8: load dst=3 src=1 src2=1 imm=0x9f1c1f9c imm2=0x4454980f rot=22 bit=11 mask=16
9: load dst=0 src=7 src2=4 imm=0xd8224ae2 imm2=0xe075297c rot=17 bit=11 mask=1
10: load dst=5 src=4 src2=0 imm=0xdf580906 imm2=0x65aa4311 rot=19 bit=0 mask=4
11: load dst=6 src=5 src2=2 imm=0xe6fbc6d2 imm2=0x1383d3ad rot=18 bit=27 mask=16
12: shfl dst=4 src=3 src2=6 imm=0xd8fde872 imm2=0xed8a496f rot=13 bit=24 mask=16
13: load dst=5 src=4 src2=2 imm=0x697b3d00 imm2=0x8b965b57 rot=9 bit=28 mask=1
14: xor dst=3 src=6 src2=1 imm=0x7af26c63 imm2=0x877c7586 rot=27 bit=26 mask=16
15: rotr dst=0 src=2 src2=5 imm=0xd34b431b imm2=0xb740221a rot=24 bit=0 mask=8
16: add dst=0 src=2 src2=6 imm=0xe364a1b9 imm2=0x26f3ad8a rot=15 bit=26 mask=2
17: rotl dst=5 src=7 src2=3 imm=0xea2f4247 imm2=0xf66e7017 rot=3 bit=7 mask=1
18: shfl dst=0 src=4 src2=5 imm=0x735eeb8d imm2=0x849ae6ee rot=28 bit=5 mask=1
19: add dst=0 src=7 src2=6 imm=0x6142ad66 imm2=0x52334d12 rot=24 bit=23 mask=1
20: xor dst=2 src=6 src2=7 imm=0xd7d553a5 imm2=0xb1871e63 rot=20 bit=5 mask=1
21: mad dst=3 src=1 src2=7 imm=0x2325a07c imm2=0x97df29e4 rot=31 bit=27 mask=1
22: rotl dst=3 src=1 src2=0 imm=0x65c76dab imm2=0x8f30d21d rot=24 bit=6 mask=1
23: add dst=2 src=1 src2=3 imm=0x41c9a9f2 imm2=0xd2c4025f rot=9 bit=18 mask=16
24: mad dst=2 src=0 src2=6 imm=0xf4186cbd imm2=0x659fc3d3 rot=28 bit=26 mask=8
25: mul dst=1 src=4 src2=7 imm=0x83a58861 imm2=0xd89ef484 rot=15 bit=0 mask=8
26: rotl dst=0 src=1 src2=2 imm=0x41f831bf imm2=0xb48420ae rot=3 bit=30 mask=16
27: rotr dst=3 src=5 src2=6 imm=0xc7ce690c imm2=0xc5c46d76 rot=26 bit=30 mask=8
28: mul dst=6 src=5 src2=2 imm=0x33dc553b imm2=0x0fbaf177 rot=3 bit=31 mask=8
29: load dst=6 src=3 src2=2 imm=0x945db3b8 imm2=0xe90eb2e5 rot=27 bit=28 mask=16
30: load dst=6 src=5 src2=2 imm=0x50d18cd2 imm2=0x97ba3fc3 rot=22 bit=18 mask=2
31: load dst=1 src=0 src2=4 imm=0xbee8421f imm2=0x070888a8 rot=20 bit=28 mask=4
32: xor dst=4 src=3 src2=6 imm=0xb9433095 imm2=0xf2e46d55 rot=21 bit=15 mask=2
33: xor dst=2 src=3 src2=3 imm=0x7bc88198 imm2=0x58f75b87 rot=18 bit=24 mask=2
34: xor dst=3 src=5 src2=5 imm=0x1d4f15de imm2=0x5892a9e6 rot=7 bit=28 mask=2
35: add dst=7 src=4 src2=5 imm=0x7d47f873 imm2=0x7b5b5474 rot=17 bit=1 mask=16
36: sub dst=1 src=0 src2=0 imm=0xfc2acc3f imm2=0xccf564a5 rot=6 bit=6 mask=1
37: add dst=0 src=3 src2=7 imm=0xc021c42e imm2=0xcac8f06d rot=11 bit=27 mask=1
38: shfl dst=5 src=3 src2=3 imm=0x780e96fc imm2=0xb8180e9d rot=26 bit=31 mask=1
39: mad dst=2 src=7 src2=1 imm=0x8eb8b2c8 imm2=0x2d6070bc rot=6 bit=10 mask=16
40: rotl dst=2 src=6 src2=4 imm=0x92d9a412 imm2=0x0daf96ea rot=8 bit=4 mask=4
41: xor dst=0 src=7 src2=3 imm=0x221869be imm2=0x6ced15b7 rot=8 bit=10 mask=4
42: sub dst=6 src=0 src2=1 imm=0xc6da7465 imm2=0x8ace05f3 rot=20 bit=3 mask=2
43: mad dst=6 src=3 src2=5 imm=0x47d59dfc imm2=0xb0607786 rot=1 bit=21 mask=4
44: load dst=6 src=3 src2=5 imm=0x756c716d imm2=0xe09f54e9 rot=19 bit=0 mask=1
45: xor dst=4 src=3 src2=4 imm=0xf80bae18 imm2=0x63cc1e4e rot=22 bit=29 mask=8
46: load dst=6 src=2 src2=6 imm=0x05783a87 imm2=0x353f1d79 rot=9 bit=19 mask=8
47: load dst=6 src=1 src2=2 imm=0x8818cd6f imm2=0x00d8a3cd rot=18 bit=27 mask=8
48: shfl dst=2 src=5 src2=6 imm=0xeed1873a imm2=0xa8bae6df rot=25 bit=17 mask=1
49: mad dst=4 src=2 src2=7 imm=0x84f12ddf imm2=0x81ef22e1 rot=28 bit=30 mask=1
50: xor dst=6 src=2 src2=5 imm=0x3bba8f2e imm2=0x383b9260 rot=25 bit=20 mask=2
51: add dst=6 src=1 src2=1 imm=0x3bfc68c2 imm2=0xfb84f451 rot=21 bit=21 mask=16
52: load dst=3 src=0 src2=2 imm=0xbefc9e84 imm2=0xb1a7db6b rot=11 bit=30 mask=8
53: rotl dst=6 src=0 src2=4 imm=0x0ac02513 imm2=0x6f981f52 rot=21 bit=12 mask=1
54: add dst=4 src=1 src2=6 imm=0xda6a6f1a imm2=0xac6be8e3 rot=19 bit=23 mask=8
55: add dst=7 src=4 src2=7 imm=0xce79bad0 imm2=0x7e7f6a00 rot=13 bit=29 mask=2
56: load dst=6 src=2 src2=1 imm=0x0757ab14 imm2=0xf03673fe rot=28 bit=16 mask=4
57: mad dst=5 src=7 src2=7 imm=0xa0e30552 imm2=0x227f94a6 rot=8 bit=8 mask=4
58: load dst=7 src=4 src2=1 imm=0xffd6dd93 imm2=0xdecd4794 rot=27 bit=6 mask=4
59: shfl dst=5 src=3 src2=7 imm=0x028a6dd2 imm2=0x0dd268e0 rot=21 bit=11 mask=2
60: xor dst=0 src=4 src2=3 imm=0x96d9f2af imm2=0x3a45a321 rot=9 bit=10 mask=2
61: add dst=0 src=3 src2=4 imm=0x715ae692 imm2=0x8f229cc1 rot=23 bit=26 mask=16
62: add dst=5 src=2 src2=3 imm=0x2e6ee746 imm2=0xb2574178 rot=22 bit=17 mask=2
63: load dst=2 src=3 src2=4 imm=0xeb76079d imm2=0xfe5e2b6e rot=27 bit=9 mask=8

20
docs/analysis/class-v6/logs/census-packs/hl-v6-all-g6/sitestats.tsv Normal file
View file

@ -0,0 +1,20 @@
program id 4de7b836cc40a4ea class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 0 generator v6
rule: ACCEPTED, distinct 128.000 per hash, saturated 0, bias max 65
site instr win off width distinct ratio under_c2 under_c3 bucket_ratio bucket_z bucket_id bit_z bit top_count
0 7 1 1 1 1044547 1.00006 0 0 1.719 4.07 35439 1.80 9 3
1 8 1 1 1 1044420 0.99994 0 0 1.812 4.60 42157 -1.89 24 3
2 9 0 0 1 1046542 1.00001 0 0 2.250 5.00 3711 -2.12 18 3
3 10 1 1 1 1044485 1.00000 0 0 1.844 4.77 43630 -2.79 22 3
4 11 2 1 1 1040456 1.00007 0 0 1.609 4.88 23903 -2.46 12 3
5 13 1 0 1 1044634 1.00015 0 0 1.750 4.24 18432 1.88 20 3
6 29 0 0 1 1046553 1.00002 0 0 2.188 4.75 23731 -2.15 23 3
7 30 1 0 1 1044584 1.00010 0 0 1.719 4.07 8130 -2.33 8 3
8 31 1 0 1 1044545 1.00006 0 0 1.812 4.60 9229 -2.38 21 3
9 44 1 0 1 1044567 1.00008 0 0 1.875 4.95 23182 -1.85 15 3
10 46 1 0 1 1044495 1.00001 0 0 1.781 4.42 13046 2.17 18 3
11 47 2 0 1 1040577 1.00019 0 0 1.484 3.88 14746 -2.32 4 3
12 52 0 0 1 1046492 0.99997 0 0 2.188 4.75 43209 3.40 5 3
13 56 0 0 1 1046475 0.99995 0 0 2.188 4.75 11372 1.89 25 3
14 58 1 1 1 1044526 1.00004 0 0 1.812 4.60 42089 2.88 18 3
15 63 0 0 1 1046517 0.99999 0 0 2.250 5.00 34545 -2.70 19 3
summary: sites 16, min ratio 0.99994, under 0.98: 0, under 0.995: 0, bucket z max 5.00, bit z max 3.40, sites over 6 sigma 0, 2.1 s
Can't render this file because it has a wrong number of fields in line 3.

259
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/census.tsv Normal file
View file

@ -0,0 +1,259 @@
# census class mx8+sh256x27+state+reg64c+fold+rw eras [0] seeds 256 era-widths 4 state /srv/builds/v6-census/packs/node1-state.igsd1 erahex edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 weights base bittest off bin 6bb243c738a4acf6 start 2026-10-08T18:54:41Z host igneum-build-6 threads 16
cand era seed candidates accepted attempt bit_z bit_site bit_bit reasons
census 0 8 1 1 0 2.8 9 12
census 0 4 1 1 0 2.9 15 4
census 0 14 1 1 0 3.5 12 21
census 0 9 1 1 0 2.8 6 25
census 0 5 1 1 0 nan - -
census 0 0 1 1 0 nan - -
census 0 1 1 1 0 nan - -
census 0 3 1 1 0 2.8 14 3
census 0 12 1 1 0 3.2 9 13
census 0 10 1 1 0 3.0 0 2
census 0 13 1 1 0 2.8 12 13
census 0 11 1 1 0 nan - -
census 0 7 1 1 0 3.4 1 23
census 0 6 1 1 0 nan - -
census 0 2 1 1 0 3.5 3 15
census 0 15 1 1 0 nan - -
census 0 16 1 1 0 nan - -
census 0 20 1 1 0 3.2 10 7
census 0 25 2 1 1 nan - - a=1;
census 0 17 1 1 0 nan - -
census 0 22 1 1 0 3.1 6 11
census 0 24 1 1 0 3.7 13 12
census 0 18 1 1 0 nan - -
census 0 23 1 1 0 nan - -
census 0 26 1 1 0 3.2 7 9
census 0 28 1 1 0 nan - -
census 0 29 1 1 0 nan - -
census 0 30 1 1 0 nan - -
census 0 19 1 1 0 nan - -
census 0 27 1 1 0 3.0 9 8
census 0 31 2 1 1 3.2 1 5 b=1;
census 0 21 2 1 1 3.5 0 18 reg64=1;
census 0 35 1 1 0 4.2 12 12
census 0 32 1 1 0 4.4 10 21
census 0 34 1 1 0 3.0 12 1
census 0 33 1 1 0 nan - -
census 0 37 1 1 0 3.1 13 8
census 0 36 2 1 1 3.2 14 3 b=1;
census 0 41 1 1 0 nan - -
census 0 38 1 1 0 nan - -
census 0 39 1 1 0 nan - -
census 0 42 2 1 1 nan - - a'=1;
census 0 40 1 1 0 2.8 9 25
census 0 44 2 1 1 nan - - a'=1;
census 0 46 1 1 0 4.4 2 3
census 0 43 1 1 0 nan - -
census 0 45 1 1 0 nan - -
census 0 47 1 1 0 3.1 4 16
census 0 50 1 1 0 3.3 9 9
census 0 53 2 1 1 nan - - b=1;
census 0 48 1 1 0 nan - -
census 0 49 1 1 0 nan - -
census 0 52 1 1 0 nan - -
census 0 54 1 1 0 3.5 3 13
census 0 57 1 1 0 3.5 6 24
census 0 58 1 1 0 3.0 8 17
census 0 59 1 1 0 nan - -
census 0 56 2 1 1 2.8 2 9 a'=1;
census 0 60 1 1 0 nan - -
census 0 55 1 1 0 nan - -
census 0 61 1 1 0 3.3 3 11
census 0 62 1 1 0 3.2 9 19
census 0 63 1 1 0 nan - -
census 0 51 2 1 1 nan - - reg64=1;
census 0 64 1 1 0 3.6 9 8
census 0 65 1 1 0 nan - -
census 0 67 2 1 1 nan - - a=1;
census 0 66 1 1 0 3.4 6 11
census 0 69 1 1 0 3.2 14 25
census 0 70 1 1 0 nan - -
census 0 68 1 1 0 nan - -
census 0 71 2 1 1 2.8 7 17 a=1;
census 0 74 1 1 0 3.0 4 6
census 0 72 1 1 0 nan - -
census 0 73 1 1 0 nan - -
census 0 76 1 1 0 3.2 2 23
census 0 75 1 1 0 2.9 4 16
census 0 77 1 1 0 nan - -
census 0 78 1 1 0 nan - -
census 0 79 1 1 0 nan - -
census 0 80 2 1 1 nan - - a'=1;
census 0 81 1 1 0 nan - -
census 0 85 1 1 0 3.1 14 24
census 0 83 1 1 0 nan - -
census 0 84 1 1 0 nan - -
census 0 86 1 1 0 3.0 7 19
census 0 82 1 1 0 nan - -
census 0 87 2 1 1 nan - - a'=1;
census 0 88 1 1 0 nan - -
census 0 89 1 1 0 nan - -
census 0 90 2 1 1 nan - - b=1;
census 0 91 1 1 0 3.2 13 8
census 0 92 1 1 0 nan - -
census 0 93 1 1 0 3.0 7 20
census 0 94 1 1 0 nan - -
census 0 95 2 1 1 3.0 13 12 a'=1;
census 0 96 1 1 0 3.1 8 7
census 0 97 1 1 0 nan - -
census 0 98 1 1 0 nan - -
census 0 100 1 1 0 nan - -
census 0 101 1 1 0 nan - -
census 0 102 1 1 0 3.5 0 13
census 0 99 1 1 0 3.1 2 23
census 0 103 1 1 0 nan - -
census 0 104 1 1 0 nan - -
census 0 105 1 1 0 nan - -
census 0 106 1 1 0 nan - -
census 0 109 1 1 0 nan - -
census 0 107 1 1 0 nan - -
census 0 108 1 1 0 nan - -
census 0 111 1 1 0 3.3 3 27
census 0 110 1 1 0 3.4 7 15
census 0 112 1 1 0 3.3 5 4
census 0 113 1 1 0 nan - -
census 0 114 1 1 0 2.9 7 8
census 0 117 1 1 0 nan - -
census 0 116 1 1 0 3.1 1 5
census 0 115 2 1 1 nan - - b=1;
census 0 118 1 1 0 nan - -
census 0 119 1 1 0 3.1 3 18
census 0 121 1 1 0 nan - -
census 0 120 1 1 0 nan - -
census 0 122 1 1 0 nan - -
census 0 123 1 1 0 nan - -
census 0 124 1 1 0 3.6 0 15
census 0 125 2 1 1 3.3 1 6 b=1;
census 0 126 1 1 0 4.3 8 3
census 0 127 2 1 1 nan - - a=1;
census 0 129 2 1 1 3.2 6 6 a'=1;
census 0 128 1 1 0 nan - -
census 0 134 1 1 0 3.7 11 4
census 0 132 1 1 0 3.4 15 1
census 0 131 1 1 0 3.3 5 21
census 0 130 1 1 0 3.2 11 8
census 0 133 1 1 0 2.8 10 0
census 0 137 1 1 0 nan - -
census 0 135 1 1 0 3.0 5 15
census 0 136 1 1 0 nan - -
census 0 138 1 1 0 2.7 12 19
census 0 139 1 1 0 3.4 11 3
census 0 140 2 1 1 nan - - a'=1;
census 0 141 1 1 0 nan - -
census 0 143 1 1 0 nan - -
census 0 142 2 1 1 4.0 6 0 a'=1;
census 0 145 1 1 0 2.7 14 24
census 0 144 1 1 0 3.7 3 22
census 0 147 1 1 0 nan - -
census 0 148 1 1 0 3.7 10 0
census 0 149 1 1 0 nan - -
census 0 146 1 1 0 3.4 15 22
census 0 150 1 1 0 nan - -
census 0 151 1 1 0 nan - -
census 0 152 1 1 0 3.1 11 25
census 0 153 1 1 0 nan - -
census 0 154 2 1 1 3.3 12 17 a'=1;
census 0 155 1 1 0 nan - -
census 0 156 1 1 0 nan - -
census 0 157 1 1 0 3.5 1 16
census 0 158 1 1 0 3.1 13 1
census 0 159 1 1 0 nan - -
census 0 160 1 1 0 2.5 1 17
census 0 161 1 1 0 nan - -
census 0 162 1 1 0 3.1 2 16
census 0 166 1 1 0 nan - -
census 0 163 2 1 1 3.7 0 19 a'=1;
census 0 165 1 1 0 3.0 11 5
census 0 167 3 1 2 3.4 2 14 a'=2;
census 0 164 1 1 0 3.8 7 25
census 0 168 1 1 0 3.6 6 25
census 0 169 2 1 1 nan - - a'=1;
census 0 170 1 1 0 nan - -
census 0 171 1 1 0 nan - -
census 0 173 3 1 2 nan - - a'=2;
census 0 172 2 1 1 3.0 15 16 a'=1;
census 0 174 1 1 0 nan - -
census 0 175 1 1 0 2.9 1 24
census 0 176 1 1 0 nan - -
census 0 177 1 1 0 3.1 14 0
census 0 178 1 1 0 4.0 14 14
census 0 180 1 1 0 3.5 3 19
census 0 182 1 1 0 nan - -
census 0 179 1 1 0 3.1 10 14
census 0 183 1 1 0 nan - -
census 0 181 1 1 0 3.1 1 14
census 0 184 2 1 1 3.0 2 25 a'=1;
census 0 187 1 1 0 3.6 4 14
census 0 186 2 1 1 nan - - a'=1;
census 0 185 1 1 0 nan - -
census 0 188 1 1 0 3.2 0 12
census 0 189 1 1 0 nan - -
census 0 191 1 1 0 nan - -
census 0 190 2 1 1 3.3 14 13 a'=1;
census 0 192 1 1 0 4.1 3 23
census 0 193 1 1 0 3.5 4 23
census 0 195 1 1 0 2.9 2 10
census 0 196 1 1 0 3.0 3 21
census 0 197 1 1 0 nan - -
census 0 194 1 1 0 3.7 2 6
census 0 199 1 1 0 2.9 14 0
census 0 198 1 1 0 nan - -
census 0 200 1 1 0 3.1 5 21
census 0 203 1 1 0 nan - -
census 0 202 1 1 0 nan - -
census 0 201 1 1 0 nan - -
census 0 204 1 1 0 3.8 8 24
census 0 205 1 1 0 3.1 7 7
census 0 206 2 1 1 nan - - a'=1;
census 0 207 1 1 0 nan - -
census 0 209 1 1 0 nan - -
census 0 208 1 1 0 nan - -
census 0 211 1 1 0 nan - -
census 0 214 1 1 0 3.7 7 24
census 0 210 1 1 0 3.3 1 20
census 0 213 1 1 0 2.8 13 3
census 0 212 1 1 0 nan - -
census 0 216 1 1 0 3.5 7 4
census 0 215 1 1 0 3.0 12 7
census 0 217 1 1 0 nan - -
census 0 218 2 1 1 2.9 0 7 a'=1;
census 0 219 1 1 0 3.0 9 22
census 0 221 1 1 0 nan - -
census 0 220 1 1 0 nan - -
census 0 222 1 1 0 3.0 14 10
census 0 223 1 1 0 nan - -
census 0 225 1 1 0 3.2 14 23
census 0 224 1 1 0 3.0 12 11
census 0 227 1 1 0 3.4 14 24
census 0 229 1 1 0 3.0 5 17
census 0 226 1 1 0 3.6 12 11
census 0 228 1 1 0 3.0 8 9
census 0 230 1 1 0 nan - -
census 0 231 1 1 0 2.9 12 13
census 0 232 1 1 0 3.4 4 15
census 0 235 1 1 0 3.3 8 16
census 0 234 1 1 0 nan - -
census 0 236 1 1 0 nan - -
census 0 237 1 1 0 3.1 7 12
census 0 233 2 1 1 3.2 7 11 c=1;
census 0 238 1 1 0 nan - -
census 0 239 1 1 0 3.4 6 18
census 0 240 2 1 1 3.4 11 17 b=1;
census 0 241 1 1 0 nan - -
census 0 243 1 1 0 3.4 5 5
census 0 242 1 1 0 nan - -
census 0 246 2 1 1 nan - - a'=1;
census 0 247 1 1 0 nan - -
census 0 245 1 1 0 3.0 13 2
census 0 250 1 1 0 3.9 14 21
census 0 248 1 1 0 3.8 0 16
census 0 251 1 1 0 3.0 11 0
census 0 249 1 1 0 nan - -
census 0 252 1 1 0 3.0 0 24
census 0 253 1 1 0 nan - -
census 0 244 2 1 1 3.0 3 8 reg64=1;
census 0 255 1 1 0 2.6 6 14
census 0 254 1 1 0 nan - -
# end 2026-10-08T18:56:51Z rows=257
Can't render this file because it has a wrong number of fields in line 2.

1
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/run.txt Normal file
View file

@ -0,0 +1 @@
# pack hl-v6-all-pr class mx8+sh256x27+state+reg64c+fold+rw epoch edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 day 69676e65756d2d6461792ffa50000000000000 era edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 state /srv/builds/v6-census/packs/node1-state.igsd1 day-index 20730 era-widths 4 bin 6bb243c738a4acf6 f8 c8277b9779b83c01 host igneum-build-6 start 2026-10-08T18:54:39Z

67
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/show.txt Normal file
View file

@ -0,0 +1,67 @@
seed "igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07" generator v5 class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 0 program id 46a8008fac7787a5 seed words 667d0fbd 7b8e5963 31c67e5e 4529ddc6 ef19d6d8 accf6211 da0aed32 abc6df31
op mix load=16 add=10 xor=10 mad=7 shfl=7 rotl=6 rotr=3 sub=3 mul=2 loads/hash 128 bytes/hash 512
era d810f22d (igneum-era/0/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07): width 4 B, stride mul 0x9ad30d99 rot 29, interleave [0, 2, 12, 13], windows (site:shrink:offset) 7:1:1 8:1:1 9:0:0 10:1:1 11:2:1 13:1:0 29:0:0 30:1:0 31:1:0 44:1:0 46:1:0 47:2:0 52:0:0 56:0:0 58:1:1 63:0:0
0: rotl dst=5 src=4 src2=3 imm=0xde0bea4e imm2=0xbaab6229 rot=2 bit=25 mask=2
1: shfl dst=2 src=6 src2=5 imm=0xca149f6a imm2=0x5b623116 rot=4 bit=16 mask=8
2: sub dst=1 src=6 src2=5 imm=0x512ddc79 imm2=0xe99741c7 rot=13 bit=21 mask=2
3: rotr dst=1 src=5 src2=7 imm=0x7819f5ec imm2=0x673c2157 rot=17 bit=21 mask=2
4: mad dst=1 src=7 src2=5 imm=0x7826a6a7 imm2=0x946f7818 rot=18 bit=21 mask=8
5: xor dst=0 src=2 src2=0 imm=0xc38c81bc imm2=0x5f6a0ed2 rot=5 bit=19 mask=16
6: shfl dst=7 src=0 src2=4 imm=0x80c84659 imm2=0x5c61dcf7 rot=3 bit=20 mask=4
7: load dst=4 src=2 src2=4 imm=0x21579975 imm2=0x85668475 rot=6 bit=3 mask=8
8: load dst=3 src=1 src2=1 imm=0x9f1c1f9c imm2=0x4454980f rot=22 bit=11 mask=16
9: load dst=0 src=7 src2=4 imm=0xd8224ae2 imm2=0xe075297c rot=17 bit=11 mask=1
10: load dst=5 src=4 src2=0 imm=0xdf580906 imm2=0x65aa4311 rot=19 bit=0 mask=4
11: load dst=6 src=5 src2=2 imm=0xe6fbc6d2 imm2=0x1383d3ad rot=18 bit=27 mask=16
12: shfl dst=4 src=3 src2=6 imm=0xd8fde872 imm2=0xed8a496f rot=13 bit=24 mask=16
13: load dst=5 src=4 src2=2 imm=0x697b3d00 imm2=0x8b965b57 rot=9 bit=28 mask=1
14: xor dst=3 src=6 src2=1 imm=0x7af26c63 imm2=0x877c7586 rot=27 bit=26 mask=16
15: rotr dst=0 src=2 src2=5 imm=0xd34b431b imm2=0xb740221a rot=24 bit=0 mask=8
16: add dst=0 src=2 src2=6 imm=0xe364a1b9 imm2=0x26f3ad8a rot=15 bit=26 mask=2
17: rotl dst=5 src=7 src2=3 imm=0xea2f4247 imm2=0xf66e7017 rot=3 bit=7 mask=1
18: shfl dst=0 src=4 src2=5 imm=0x735eeb8d imm2=0x849ae6ee rot=28 bit=5 mask=1
19: add dst=0 src=7 src2=6 imm=0x6142ad66 imm2=0x52334d12 rot=24 bit=23 mask=1
20: xor dst=2 src=6 src2=7 imm=0xd7d553a5 imm2=0xb1871e63 rot=20 bit=5 mask=1
21: mad dst=3 src=1 src2=7 imm=0x2325a07c imm2=0x97df29e4 rot=31 bit=27 mask=1
22: rotl dst=3 src=1 src2=0 imm=0x65c76dab imm2=0x8f30d21d rot=24 bit=6 mask=1
23: add dst=2 src=1 src2=3 imm=0x41c9a9f2 imm2=0xd2c4025f rot=9 bit=18 mask=16
24: mad dst=2 src=0 src2=6 imm=0xf4186cbd imm2=0x659fc3d3 rot=28 bit=26 mask=8
25: mul dst=1 src=4 src2=7 imm=0x83a58861 imm2=0xd89ef484 rot=15 bit=0 mask=8
26: rotl dst=0 src=1 src2=2 imm=0x41f831bf imm2=0xb48420ae rot=3 bit=30 mask=16
27: rotr dst=3 src=5 src2=6 imm=0xc7ce690c imm2=0xc5c46d76 rot=26 bit=30 mask=8
28: mul dst=6 src=5 src2=2 imm=0x33dc553b imm2=0x0fbaf177 rot=3 bit=31 mask=8
29: load dst=6 src=3 src2=2 imm=0x945db3b8 imm2=0xe90eb2e5 rot=27 bit=28 mask=16
30: load dst=6 src=5 src2=2 imm=0x50d18cd2 imm2=0x97ba3fc3 rot=22 bit=18 mask=2
31: load dst=1 src=0 src2=4 imm=0xbee8421f imm2=0x070888a8 rot=20 bit=28 mask=4
32: xor dst=4 src=3 src2=6 imm=0xb9433095 imm2=0xf2e46d55 rot=21 bit=15 mask=2
33: xor dst=2 src=3 src2=3 imm=0x7bc88198 imm2=0x58f75b87 rot=18 bit=24 mask=2
34: xor dst=3 src=5 src2=5 imm=0x1d4f15de imm2=0x5892a9e6 rot=7 bit=28 mask=2
35: add dst=7 src=4 src2=5 imm=0x7d47f873 imm2=0x7b5b5474 rot=17 bit=1 mask=16
36: sub dst=1 src=0 src2=0 imm=0xfc2acc3f imm2=0xccf564a5 rot=6 bit=6 mask=1
37: add dst=0 src=3 src2=7 imm=0xc021c42e imm2=0xcac8f06d rot=11 bit=27 mask=1
38: shfl dst=5 src=3 src2=3 imm=0x780e96fc imm2=0xb8180e9d rot=26 bit=31 mask=1
39: mad dst=2 src=7 src2=1 imm=0x8eb8b2c8 imm2=0x2d6070bc rot=6 bit=10 mask=16
40: rotl dst=2 src=6 src2=4 imm=0x92d9a412 imm2=0x0daf96ea rot=8 bit=4 mask=4
41: xor dst=0 src=7 src2=3 imm=0x221869be imm2=0x6ced15b7 rot=8 bit=10 mask=4
42: sub dst=6 src=0 src2=1 imm=0xc6da7465 imm2=0x8ace05f3 rot=20 bit=3 mask=2
43: mad dst=6 src=3 src2=5 imm=0x47d59dfc imm2=0xb0607786 rot=1 bit=21 mask=4
44: load dst=6 src=3 src2=5 imm=0x756c716d imm2=0xe09f54e9 rot=19 bit=0 mask=1
45: xor dst=4 src=3 src2=4 imm=0xf80bae18 imm2=0x63cc1e4e rot=22 bit=29 mask=8
46: load dst=6 src=2 src2=6 imm=0x05783a87 imm2=0x353f1d79 rot=9 bit=19 mask=8
47: load dst=6 src=1 src2=2 imm=0x8818cd6f imm2=0x00d8a3cd rot=18 bit=27 mask=8
48: shfl dst=2 src=5 src2=6 imm=0xeed1873a imm2=0xa8bae6df rot=25 bit=17 mask=1
49: mad dst=4 src=2 src2=7 imm=0x84f12ddf imm2=0x81ef22e1 rot=28 bit=30 mask=1
50: xor dst=6 src=2 src2=5 imm=0x3bba8f2e imm2=0x383b9260 rot=25 bit=20 mask=2
51: add dst=6 src=1 src2=1 imm=0x3bfc68c2 imm2=0xfb84f451 rot=21 bit=21 mask=16
52: load dst=3 src=0 src2=2 imm=0xbefc9e84 imm2=0xb1a7db6b rot=11 bit=30 mask=8
53: rotl dst=6 src=0 src2=4 imm=0x0ac02513 imm2=0x6f981f52 rot=21 bit=12 mask=1
54: add dst=4 src=1 src2=6 imm=0xda6a6f1a imm2=0xac6be8e3 rot=19 bit=23 mask=8
55: add dst=7 src=4 src2=7 imm=0xce79bad0 imm2=0x7e7f6a00 rot=13 bit=29 mask=2
56: load dst=6 src=2 src2=1 imm=0x0757ab14 imm2=0xf03673fe rot=28 bit=16 mask=4
57: mad dst=5 src=7 src2=7 imm=0xa0e30552 imm2=0x227f94a6 rot=8 bit=8 mask=4
58: load dst=7 src=4 src2=1 imm=0xffd6dd93 imm2=0xdecd4794 rot=27 bit=6 mask=4
59: shfl dst=5 src=3 src2=7 imm=0x028a6dd2 imm2=0x0dd268e0 rot=21 bit=11 mask=2
60: xor dst=0 src=4 src2=3 imm=0x96d9f2af imm2=0x3a45a321 rot=9 bit=10 mask=2
61: add dst=0 src=3 src2=4 imm=0x715ae692 imm2=0x8f229cc1 rot=23 bit=26 mask=16
62: add dst=5 src=2 src2=3 imm=0x2e6ee746 imm2=0xb2574178 rot=22 bit=17 mask=2
63: load dst=2 src=3 src2=4 imm=0xeb76079d imm2=0xfe5e2b6e rot=27 bit=9 mask=8

20
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/sitestats.tsv Normal file
View file

@ -0,0 +1,20 @@
program id 46a8008fac7787a5 class mx8-erad810f22d+sh256x27+state+reg64c+fold+rw attempt 0 generator v5
rule: ACCEPTED, distinct 128.000 per hash, saturated 0, bias max 65
site instr win off width distinct ratio under_c2 under_c3 bucket_ratio bucket_z bucket_id bit_z bit top_count
0 7 1 1 1 1044547 1.00006 0 0 1.719 4.07 35439 1.80 9 3
1 8 1 1 1 1044420 0.99994 0 0 1.812 4.60 42157 -1.89 24 3
2 9 0 0 1 1046542 1.00001 0 0 2.250 5.00 3711 -2.12 18 3
3 10 1 1 1 1044485 1.00000 0 0 1.844 4.77 43630 -2.79 22 3
4 11 2 1 1 1040456 1.00007 0 0 1.609 4.88 23903 -2.46 12 3
5 13 1 0 1 1044634 1.00015 0 0 1.750 4.24 18432 1.88 20 3
6 29 0 0 1 1046553 1.00002 0 0 2.188 4.75 23731 -2.15 23 3
7 30 1 0 1 1044584 1.00010 0 0 1.719 4.07 8130 -2.33 8 3
8 31 1 0 1 1044545 1.00006 0 0 1.812 4.60 9229 -2.38 21 3
9 44 1 0 1 1044567 1.00008 0 0 1.875 4.95 23182 -1.85 15 3
10 46 1 0 1 1044495 1.00001 0 0 1.781 4.42 13046 2.17 18 3
11 47 2 0 1 1040577 1.00019 0 0 1.484 3.88 14746 -2.32 4 3
12 52 0 0 1 1046492 0.99997 0 0 2.188 4.75 43209 3.40 5 3
13 56 0 0 1 1046475 0.99995 0 0 2.188 4.75 11372 1.89 25 3
14 58 1 1 1 1044526 1.00004 0 0 1.812 4.60 42089 2.88 18 3
15 63 0 0 1 1046517 0.99999 0 0 2.250 5.00 34545 -2.70 19 3
summary: sites 16, min ratio 0.99994, under 0.98: 0, under 0.995: 0, bucket z max 5.00, bit z max 3.40, sites over 6 sigma 0, 3.8 s
Can't render this file because it has a wrong number of fields in line 3.

17
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/uniform-all.tsv Normal file
View file

@ -0,0 +1,17 @@
class era seed attempt program_id min_site_ratio min_site top0.1_share control_share ratio_to_control max_item_reads control_max reads
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 6 1 f20d054abb283da5 1.00262 24 0.001685 0.001447 1.1647 128 115 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 0 0 00fda3d2b8612168 1.00263 15 0.001828 0.001447 1.2634 141 111 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 5 0 12879a9b6ced301f 1.00260 28 0.001829 0.001446 1.2651 144 108 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 2 0 3902e732d1eb8fe5 1.00263 30 0.001758 0.001446 1.2159 144 113 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 1 0 fb48292005a60da8 1.00267 0 0.001974 0.001447 1.3642 156 112 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 3 0 16990a318d0df39f 1.00265 28 0.001862 0.001447 1.2873 141 113 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 4 0 61b13884809d7107 1.00267 17 0.001973 0.001446 1.3643 150 111 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 7 0 3e32263b35b3b80c 1.00258 5 0.002045 0.001446 1.4142 155 109 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 14 0 4756d11f3858fee8 1.00264 16 0.001828 0.001447 1.2635 136 110 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 9 0 cbbf99686239994e 1.00255 2 0.001756 0.001447 1.2134 131 113 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 11 0 4fc1dd26390dbfb9 1.00266 17 0.001829 0.001446 1.2649 143 109 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 8 0 b71612f2b27a6a95 1.00265 30 0.001612 0.001446 1.1148 122 108 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 10 0 c21fa18b38d5a7ae 1.00260 7 0.001642 0.001446 1.1359 127 112 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 12 0 0d47c44b60dfbe2a 1.00263 5 0.001685 0.001446 1.1650 129 113 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 13 0 46581bd04324f6c4 1.00265 24 0.001683 0.001446 1.1638 129 112 1073741824
mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 15 1 2985831a995342a5 1.00264 7 0.001831 0.001446 1.2659 136 115 1073741824
1 class era seed attempt program_id min_site_ratio min_site top0.1_share control_share ratio_to_control max_item_reads control_max reads
2 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 6 1 f20d054abb283da5 1.00262 24 0.001685 0.001447 1.1647 128 115 1073741824
3 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 0 0 00fda3d2b8612168 1.00263 15 0.001828 0.001447 1.2634 141 111 1073741824
4 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 5 0 12879a9b6ced301f 1.00260 28 0.001829 0.001446 1.2651 144 108 1073741824
5 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 2 0 3902e732d1eb8fe5 1.00263 30 0.001758 0.001446 1.2159 144 113 1073741824
6 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 1 0 fb48292005a60da8 1.00267 0 0.001974 0.001447 1.3642 156 112 1073741824
7 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 3 0 16990a318d0df39f 1.00265 28 0.001862 0.001447 1.2873 141 113 1073741824
8 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 4 0 61b13884809d7107 1.00267 17 0.001973 0.001446 1.3643 150 111 1073741824
9 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 7 0 3e32263b35b3b80c 1.00258 5 0.002045 0.001446 1.4142 155 109 1073741824
10 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 14 0 4756d11f3858fee8 1.00264 16 0.001828 0.001447 1.2635 136 110 1073741824
11 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 9 0 cbbf99686239994e 1.00255 2 0.001756 0.001447 1.2134 131 113 1073741824
12 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 11 0 4fc1dd26390dbfb9 1.00266 17 0.001829 0.001446 1.2649 143 109 1073741824
13 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 8 0 b71612f2b27a6a95 1.00265 30 0.001612 0.001446 1.1148 122 108 1073741824
14 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 10 0 c21fa18b38d5a7ae 1.00260 7 0.001642 0.001446 1.1359 127 112 1073741824
15 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 12 0 0d47c44b60dfbe2a 1.00263 5 0.001685 0.001446 1.1650 129 113 1073741824
16 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 13 0 46581bd04324f6c4 1.00265 24 0.001683 0.001446 1.1638 129 112 1073741824
17 mx8+sh256x27+state+reg64c+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 15 1 2985831a995342a5 1.00264 7 0.001831 0.001446 1.2659 136 115 1073741824

17
docs/analysis/class-v6/logs/census-packs/hl-v6-all-pr/uniform-foldrw.tsv Normal file
View file

@ -0,0 +1,17 @@
class era seed attempt program_id min_site_ratio min_site top0.1_share control_share ratio_to_control max_item_reads control_max reads
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 12 0 b442199389375494 1.00274 11 0.001874 0.001647 1.1378 74 66 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 9 0 c65de5f57750b3d8 1.00265 3 0.001950 0.001647 1.1837 81 67 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 5 0 c75a8dcb0e3abba3 1.00265 4 0.002030 0.001646 1.2335 81 67 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 3 0 e2f4a6bc767cb223 1.00264 14 0.002077 0.001645 1.2621 82 68 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 0 0 8761a2b8abcaf2fa 1.00270 6 0.002029 0.001646 1.2326 83 72 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 1 0 66de8ae8ad3112ba 1.00241 0 0.002180 0.001646 1.3243 85 66 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 14 0 955fbdfb42ffc77a 1.00262 7 0.002029 0.001644 1.2339 82 69 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 2 0 8eb519706ce24ac9 1.00266 4 0.001959 0.001647 1.1899 80 68 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 13 0 87309135d1064a16 1.00263 12 0.001874 0.001646 1.1388 78 64 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 8 0 d3170c83bca3e2f9 1.00264 11 0.001802 0.001645 1.0956 71 66 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 15 1 5a6f1d91e4491c09 1.00268 7 0.002034 0.001645 1.2365 80 70 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 10 0 c29a5131a2facdf8 1.00262 0 0.001843 0.001646 1.1192 72 66 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 7 0 45fe3f5104357d5e 1.00260 3 0.002256 0.001647 1.3700 92 68 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 6 1 70b85efeb5e48509 1.00265 3 0.001879 0.001645 1.1422 78 65 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 4 0 9f2e68c20d6f4f0b 1.00260 6 0.002181 0.001645 1.3260 87 68 536870912
mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 11 0 a43a073e5a7b7ffd 1.00265 8 0.002030 0.001646 1.2332 84 66 536870912
1 class era seed attempt program_id min_site_ratio min_site top0.1_share control_share ratio_to_control max_item_reads control_max reads
2 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 12 0 b442199389375494 1.00274 11 0.001874 0.001647 1.1378 74 66 536870912
3 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 9 0 c65de5f57750b3d8 1.00265 3 0.001950 0.001647 1.1837 81 67 536870912
4 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 5 0 c75a8dcb0e3abba3 1.00265 4 0.002030 0.001646 1.2335 81 67 536870912
5 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 3 0 e2f4a6bc767cb223 1.00264 14 0.002077 0.001645 1.2621 82 68 536870912
6 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 0 0 8761a2b8abcaf2fa 1.00270 6 0.002029 0.001646 1.2326 83 72 536870912
7 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 1 0 66de8ae8ad3112ba 1.00241 0 0.002180 0.001646 1.3243 85 66 536870912
8 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 14 0 955fbdfb42ffc77a 1.00262 7 0.002029 0.001644 1.2339 82 69 536870912
9 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 2 0 8eb519706ce24ac9 1.00266 4 0.001959 0.001647 1.1899 80 68 536870912
10 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 13 0 87309135d1064a16 1.00263 12 0.001874 0.001646 1.1388 78 64 536870912
11 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 8 0 d3170c83bca3e2f9 1.00264 11 0.001802 0.001645 1.0956 71 66 536870912
12 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 15 1 5a6f1d91e4491c09 1.00268 7 0.002034 0.001645 1.2365 80 70 536870912
13 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 10 0 c29a5131a2facdf8 1.00262 0 0.001843 0.001646 1.1192 72 66 536870912
14 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 7 0 45fe3f5104357d5e 1.00260 3 0.002256 0.001647 1.3700 92 68 536870912
15 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 6 1 70b85efeb5e48509 1.00265 3 0.001879 0.001645 1.1422 78 65 536870912
16 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 4 0 9f2e68c20d6f4f0b 1.00260 6 0.002181 0.001645 1.3260 87 68 536870912
17 mx8+sh256x27+state+fold+rw 0:edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 11 0 a43a073e5a7b7ffd 1.00265 8 0.002030 0.001646 1.2332 84 66 536870912

259
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/census.tsv Normal file
View file

@ -0,0 +1,259 @@
# census class mx8+sh256x27+state+rw eras [0] seeds 256 era-widths 4 state /srv/builds/v6-census/packs/node1-state.igsd1 erahex edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 weights base bittest off bin 50903d300aa46f20 start 2026-10-08T15:22:12Z host igneum-build-3 threads 8
cand era seed candidates accepted attempt bit_z bit_site bit_bit reasons
census 0 4 1 1 0 nan - -
census 0 0 1 1 0 3.6 6 23
census 0 6 3 1 2 nan - - b=1;a'=1;
census 0 1 1 1 0 nan - -
census 0 7 1 1 0 3.3 12 9
census 0 3 1 1 0 nan - -
census 0 2 1 1 0 nan - -
census 0 5 1 1 0 nan - -
census 0 9 1 1 0 nan - -
census 0 8 1 1 0 nan - -
census 0 10 1 1 0 nan - -
census 0 12 2 1 1 nan - - a'=1;
census 0 11 1 1 0 nan - -
census 0 13 2 1 1 3.2 5 20 a'=1;
census 0 15 1 1 0 3.4 14 11
census 0 14 1 1 0 2.8 10 14
census 0 16 1 1 0 3.4 2 21
census 0 19 1 1 0 nan - -
census 0 23 1 1 0 nan - -
census 0 18 1 1 0 nan - -
census 0 17 2 1 1 3.4 15 0 a'=1;
census 0 20 1 1 0 2.8 9 14
census 0 22 1 1 0 3.2 8 13
census 0 21 1 1 0 nan - -
census 0 24 1 1 0 3.2 11 7
census 0 28 1 1 0 nan - -
census 0 27 1 1 0 nan - -
census 0 26 2 1 1 3.4 15 15 a'=1;
census 0 29 1 1 0 nan - -
census 0 30 1 1 0 nan - -
census 0 25 1 1 0 2.8 3 21
census 0 31 1 1 0 nan - -
census 0 32 2 1 1 3.2 14 26 a'=1;
census 0 33 1 1 0 2.5 1 15
census 0 35 1 1 0 3.0 6 25
census 0 37 1 1 0 2.8 13 2
census 0 34 1 1 0 nan - -
census 0 36 1 1 0 3.6 5 13
census 0 39 1 1 0 nan - -
census 0 38 1 1 0 nan - -
census 0 42 1 1 0 3.3 3 12
census 0 40 1 1 0 nan - -
census 0 44 1 1 0 nan - -
census 0 43 1 1 0 nan - -
census 0 41 1 1 0 nan - -
census 0 45 1 1 0 3.3 3 17
census 0 47 1 1 0 nan - -
census 0 46 2 1 1 3.4 2 10 a=1;
census 0 48 1 1 0 2.5 5 3
census 0 50 1 1 0 nan - -
census 0 49 2 1 1 nan - - a'=1;
census 0 51 1 1 0 2.7 11 12
census 0 52 1 1 0 nan - -
census 0 53 1 1 0 3.5 14 25
census 0 54 1 1 0 nan - -
census 0 55 1 1 0 nan - -
census 0 57 1 1 0 2.7 10 8
census 0 56 1 1 0 nan - -
census 0 58 1 1 0 3.2 10 11
census 0 59 1 1 0 nan - -
census 0 62 1 1 0 3.0 6 25
census 0 60 1 1 0 nan - -
census 0 61 1 1 0 nan - -
census 0 63 1 1 0 nan - -
census 0 64 1 1 0 nan - -
census 0 66 1 1 0 nan - -
census 0 65 1 1 0 3.0 4 15
census 0 67 1 1 0 nan - -
census 0 68 1 1 0 nan - -
census 0 70 1 1 0 3.0 4 3
census 0 69 1 1 0 nan - -
census 0 71 1 1 0 3.5 14 2
census 0 72 1 1 0 3.0 0 16
census 0 73 1 1 0 3.0 5 5
census 0 76 1 1 0 nan - -
census 0 75 2 1 1 nan - - a'=1;
census 0 74 1 1 0 nan - -
census 0 78 1 1 0 3.3 10 23
census 0 77 1 1 0 nan - -
census 0 79 1 1 0 nan - -
census 0 80 1 1 0 nan - -
census 0 81 1 1 0 nan - -
census 0 84 1 1 0 3.0 10 20
census 0 82 2 1 1 2.9 6 16 a=1;
census 0 83 1 1 0 3.0 3 6
census 0 85 1 1 0 nan - -
census 0 86 1 1 0 nan - -
census 0 87 1 1 0 nan - -
census 0 88 2 1 1 nan - - a'=1;
census 0 89 1 1 0 nan - -
census 0 92 1 1 0 3.1 13 22
census 0 90 1 1 0 3.0 6 16
census 0 91 1 1 0 3.2 0 19
census 0 93 1 1 0 3.2 13 3
census 0 94 1 1 0 nan - -
census 0 95 1 1 0 nan - -
census 0 96 1 1 0 3.1 15 27
census 0 98 1 1 0 nan - -
census 0 99 1 1 0 nan - -
census 0 97 1 1 0 3.0 8 5
census 0 100 1 1 0 nan - -
census 0 101 1 1 0 nan - -
census 0 102 1 1 0 nan - -
census 0 103 1 1 0 nan - -
census 0 104 1 1 0 7.0 13 1
census 0 105 1 1 0 nan - -
census 0 107 1 1 0 nan - -
census 0 108 1 1 0 nan - -
census 0 106 1 1 0 3.2 2 7
census 0 109 1 1 0 nan - -
census 0 110 1 1 0 nan - -
census 0 111 1 1 0 nan - -
census 0 112 1 1 0 nan - -
census 0 113 1 1 0 nan - -
census 0 116 1 1 0 nan - -
census 0 115 1 1 0 nan - -
census 0 114 1 1 0 nan - -
census 0 117 2 1 1 nan - - a'=1;
census 0 118 1 1 0 nan - -
census 0 119 1 1 0 3.2 4 0
census 0 120 1 1 0 nan - -
census 0 121 1 1 0 3.7 8 19
census 0 122 1 1 0 3.1 3 25
census 0 124 1 1 0 nan - -
census 0 125 1 1 0 2.9 0 5
census 0 123 1 1 0 3.8 15 9
census 0 126 1 1 0 3.3 8 9
census 0 127 1 1 0 nan - -
census 0 128 1 1 0 nan - -
census 0 129 1 1 0 2.9 10 24
census 0 130 1 1 0 3.9 13 15
census 0 131 1 1 0 3.1 3 7
census 0 132 1 1 0 nan - -
census 0 134 1 1 0 nan - -
census 0 133 1 1 0 nan - -
census 0 135 1 1 0 3.7 11 6
census 0 136 2 1 1 3.6 14 6 a'=1;
census 0 137 1 1 0 2.7 3 5
census 0 138 1 1 0 3.3 1 5
census 0 142 1 1 0 4.3 3 20
census 0 139 1 1 0 3.3 1 5
census 0 140 2 1 1 3.8 13 10 a'=1;
census 0 141 2 1 1 2.7 6 3 a'=1;
census 0 143 1 1 0 nan - -
census 0 144 1 1 0 nan - -
census 0 145 1 1 0 nan - -
census 0 151 1 1 0 3.3 14 7
census 0 147 1 1 0 nan - -
census 0 146 1 1 0 3.2 14 22
census 0 149 1 1 0 nan - -
census 0 148 2 1 1 nan - - a=1;
census 0 150 1 1 0 nan - -
census 0 152 1 1 0 nan - -
census 0 153 1 1 0 2.9 4 26
census 0 154 1 1 0 nan - -
census 0 156 1 1 0 nan - -
census 0 155 1 1 0 nan - -
census 0 157 1 1 0 4.2 8 21
census 0 158 1 1 0 nan - -
census 0 159 1 1 0 3.0 6 22
census 0 160 1 1 0 3.6 6 4
census 0 161 1 1 0 2.9 1 13
census 0 162 1 1 0 3.2 3 19
census 0 163 1 1 0 3.5 15 5
census 0 165 1 1 0 3.7 6 17
census 0 164 1 1 0 3.0 9 26
census 0 166 1 1 0 2.9 7 0
census 0 168 1 1 0 nan - -
census 0 167 1 1 0 2.6 8 16
census 0 169 2 1 1 2.7 10 17 a'=1;
census 0 172 1 1 0 nan - -
census 0 171 2 1 1 3.0 13 18 b=1;
census 0 170 3 1 2 nan - - a'=2;
census 0 173 1 1 0 nan - -
census 0 175 1 1 0 nan - -
census 0 174 1 1 0 nan - -
census 0 176 1 1 0 nan - -
census 0 177 1 1 0 2.9 11 1
census 0 178 1 1 0 3.2 6 3
census 0 180 1 1 0 2.7 10 18
census 0 179 1 1 0 nan - -
census 0 181 1 1 0 nan - -
census 0 183 2 1 1 nan - - a'=1;
census 0 184 1 1 0 nan - -
census 0 182 1 1 0 nan - -
census 0 187 1 1 0 nan - -
census 0 186 1 1 0 nan - -
census 0 188 1 1 0 3.0 1 22
census 0 185 1 1 0 nan - -
census 0 189 1 1 0 nan - -
census 0 190 1 1 0 nan - -
census 0 191 1 1 0 nan - -
census 0 192 1 1 0 nan - -
census 0 194 1 1 0 nan - -
census 0 195 2 1 1 nan - - a'=1;
census 0 193 1 1 0 3.3 12 13
census 0 196 1 1 0 3.9 2 18
census 0 198 1 1 0 3.0 12 10
census 0 197 1 1 0 nan - -
census 0 199 1 1 0 nan - -
census 0 200 3 1 2 nan - - a=2;
census 0 201 1 1 0 3.0 7 11
census 0 204 1 1 0 3.2 3 24
census 0 202 1 1 0 3.3 5 2
census 0 203 1 1 0 nan - -
census 0 205 1 1 0 nan - -
census 0 206 1 1 0 nan - -
census 0 207 1 1 0 nan - -
census 0 208 1 1 0 3.2 6 12
census 0 209 1 1 0 2.9 11 3
census 0 210 2 1 1 3.1 14 10 a'=1;
census 0 211 1 1 0 nan - -
census 0 212 1 1 0 3.7 10 20
census 0 213 1 1 0 3.1 0 9
census 0 216 2 1 1 3.1 2 1 a'=1;
census 0 214 1 1 0 nan - -
census 0 215 1 1 0 nan - -
census 0 218 1 1 0 2.9 3 3
census 0 217 1 1 0 3.2 2 27
census 0 219 1 1 0 3.3 2 22
census 0 220 2 1 1 nan - - a=1;
census 0 221 1 1 0 nan - -
census 0 222 1 1 0 nan - -
census 0 223 1 1 0 3.1 8 4
census 0 224 1 1 0 nan - -
census 0 226 1 1 0 nan - -
census 0 225 2 1 1 nan - - b=1;
census 0 228 1 1 0 nan - -
census 0 227 1 1 0 3.6 15 18
census 0 229 1 1 0 nan - -
census 0 230 1 1 0 nan - -
census 0 232 1 1 0 nan - -
census 0 233 2 1 1 2.9 7 17 b=1;
census 0 234 1 1 0 nan - -
census 0 235 1 1 0 nan - -
census 0 236 1 1 0 nan - -
census 0 231 2 1 1 nan - - c'''=1;
census 0 237 1 1 0 nan - -
census 0 238 1 1 0 nan - -
census 0 239 1 1 0 3.2 1 22
census 0 240 1 1 0 nan - -
census 0 241 1 1 0 nan - -
census 0 242 1 1 0 nan - -
census 0 243 1 1 0 nan - -
census 0 245 3 1 2 nan - - b=1;a=1;
census 0 244 1 1 0 3.9 13 16
census 0 246 2 1 1 2.5 5 9 a'=1;
census 0 247 1 1 0 nan - -
census 0 249 1 1 0 3.6 6 21
census 0 250 1 1 0 nan - -
census 0 248 1 1 0 nan - -
census 0 251 1 1 0 nan - -
census 0 255 1 1 0 nan - -
census 0 252 1 1 0 nan - -
census 0 253 1 1 0 nan - -
census 0 254 1 1 0 nan - -
# end 2026-10-08T15:25:06Z rows=257
Can't render this file because it has a wrong number of fields in line 2.

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p10.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (9,10,11,16,18,19,20,31, waited 1018 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p10 at 2^24 (window control, by site)
[2026-10-08T15:25:15Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "10..10", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p10"]
[2026-10-08T15:25:15Z] census warps: programs [10] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:25:15Z] day 20730: cache filled in 0.53 s, fnv 448274a57f508cbc
[2026-10-08T15:25:25Z] table: 16777216 items derived with their 8 lines in 9.6 s; library comparison all (0 mismatches)
[2026-10-08T15:25:28Z] program p10-attack-f8: epoch seed 09368273d8db2033d44dd7ded3ddc0b77f0534c0eda2df0738322f27703dbf14 era seed 765f89f9870a30625296d2e6a5c5c5e669f5c92126edf631cc3dc00cffdfadf4 id 44d41a7fb8b270c2 attempt 0 class mx8-era8e8bdbb5+sh256x27+state+rw op mix load=16 rotl=11 add=8 xor=7 mad=6 sub=6 shfl=4 mul=3 rotr=3; era stride mul 0x5732bb5d rot 17 interleave [3, 5, 8, 13] windows 1:1:1 2:1:1 4:0:0 5:2:2 8:0:0 18:2:3 24:0:0 25:2:1 28:2:1 37:2:0 39:2:3 41:1:0 45:1:0 52:0:0 53:1:0 57:1:1
[2026-10-08T15:25:28Z] load site instr 1: src r0 win 1 off 1; last base writer add at 0; writers back to the last injecting one: add@0; shadow writes of r0 per rep: add=9 mad=6 mul=3 rotl=6 rotr=2 shfl=2 sub=2 xor=7
[2026-10-08T15:25:28Z] load site instr 2: src r1 win 1 off 1; last base writer load at 1; writers back to the last injecting one: load@1; shadow writes of r1 per rep: add=5 mad=4 mul=1 mulhi=1 rotl=3 rotr=8 shfl=2 sub=4 xor=6
[2026-10-08T15:25:28Z] load site instr 4: src r3 win 0 off 0; last base writer load at 2; writers back to the last injecting one: load@2; shadow writes of r3 per rep: add=5 mad=2 mul=1 rotl=6 rotr=6 shfl=1 sub=8 xor=3
[2026-10-08T15:25:28Z] load site instr 5: src r1 win 2 off 2; last base writer mad at 3; writers back to the last injecting one: mad@3; shadow writes of r1 per rep: add=5 mad=4 mul=1 mulhi=1 rotl=3 rotr=8 shfl=2 sub=4 xor=6
[2026-10-08T15:25:28Z] load site instr 8: src r6 win 0 off 0; last base writer sub at 7; writers back to the last injecting one: sub@7; shadow writes of r6 per rep: add=3 mad=5 mul=2 mulhi=1 rotl=6 rotr=3 shfl=1 sub=2 xor=3
[2026-10-08T15:25:28Z] load site instr 18: src r5 win 2 off 3; last base writer add at 9; writers back to the last injecting one: add@9; shadow writes of r5 per rep: add=1 mad=5 mul=2 rotl=7 rotr=3 shfl=1 xor=4
[2026-10-08T15:25:28Z] load site instr 24: src r4 win 0 off 0; last base writer mad at 14; writers back to the last injecting one: mad@14; shadow writes of r4 per rep: add=7 mad=8 mul=1 rotl=3 rotr=1 shfl=1 sub=4 xor=5
[2026-10-08T15:25:28Z] load site instr 25: src r6 win 2 off 1; last base writer shfl at 15; writers back to the last injecting one: shfl@15; shadow writes of r6 per rep: add=3 mad=5 mul=2 mulhi=1 rotl=6 rotr=3 shfl=1 sub=2 xor=3
[2026-10-08T15:25:28Z] load site instr 28: src r1 win 2 off 1; last base writer sub at 26; writers back to the last injecting one: sub@26; shadow writes of r1 per rep: add=5 mad=4 mul=1 mulhi=1 rotl=3 rotr=8 shfl=2 sub=4 xor=6
[2026-10-08T15:25:28Z] load site instr 37: src r1 win 2 off 0; last base writer mad at 33; writers back to the last injecting one: mad@33; shadow writes of r1 per rep: add=5 mad=4 mul=1 mulhi=1 rotl=3 rotr=8 shfl=2 sub=4 xor=6
[2026-10-08T15:25:28Z] load site instr 39: src r5 win 2 off 3; last base writer mad at 38; writers back to the last injecting one: mad@38; shadow writes of r5 per rep: add=1 mad=5 mul=2 rotl=7 rotr=3 shfl=1 xor=4
[2026-10-08T15:25:28Z] load site instr 41: src r4 win 1 off 0; last base writer xor at 30; writers back to the last injecting one: xor@30; shadow writes of r4 per rep: add=7 mad=8 mul=1 rotl=3 rotr=1 shfl=1 sub=4 xor=5
[2026-10-08T15:25:28Z] load site instr 45: src r2 win 1 off 0; last base writer rotl at 36; writers back to the last injecting one: rotl@36 rotr@32 load@28; shadow writes of r2 per rep: add=6 mad=6 mul=2 rotl=5 rotr=5 sub=6 xor=6
[2026-10-08T15:25:28Z] load site instr 52: src r3 win 0 off 0; last base writer load at 41; writers back to the last injecting one: load@41; shadow writes of r3 per rep: add=5 mad=2 mul=1 rotl=6 rotr=6 shfl=1 sub=8 xor=3
[2026-10-08T15:25:28Z] load site instr 53: src r6 win 1 off 0; last base writer xor at 49; writers back to the last injecting one: xor@49; shadow writes of r6 per rep: add=3 mad=5 mul=2 mulhi=1 rotl=6 rotr=3 shfl=1 sub=2 xor=3
[2026-10-08T15:25:28Z] load site instr 57: src r4 win 1 off 1; last base writer load at 52; writers back to the last injecting one: load@52; shadow writes of r4 per rep: add=7 mad=8 mul=1 rotl=3 rotr=1 shfl=1 sub=4 xor=5
[2026-10-08T15:25:28Z] window layer: site item windows (first, items) (0x800000,2^23) (0x800000,2^23) (0x0,2^24) (0x800000,2^22) (0x0,2^24) (0xc00000,2^22) (0x0,2^24) (0x400000,2^22) (0x400000,2^22) (0x0,2^22) (0xc00000,2^22) (0x0,2^23) (0x0,2^23) (0x0,2^24) (0x0,2^23) (0x800000,2^23); expected reads per item by quarter 112.000 144.000 112.000 144.000 (flat uniform 128.000)
[2026-10-08T15:28:13Z] 524288 warps (16777216 nonces) interpreted in 165.5 s (2.526 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:28:13Z] p10-attack-f8 distinct lines per hash: n 16777216 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8711
[2026-10-08T15:28:13Z] p10-attack-f8 distinct items per hash: n 16777216 min 126 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:28:13Z] p10-attack-f8 distinct lines per warp: n 524288 min 32566 p1 32605 median 32637 p99 32664 max 32688 mean 32636.3316
[2026-10-08T15:28:13Z] p10-attack-f8 distinct items per warp: n 524288 min 4089 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4936
[2026-10-08T15:28:13Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:28:13Z] positions: largest hi16-bucket (256 items) share 0.0070% at p23 (iteration 1, site 7, instr 25); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p5 (site 5, instr 18)
[2026-10-08T15:28:13Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p15 (site 15, instr 57); saturated sources at one position 0 of 2048 at p0 (site 0, instr 1); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:28:14Z] p10-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 214 (bin 7211211) z_max +7.60 min 61 z_min -5.92 chi2/dof 3.09410 chi2_z +6065.15 top0.1% 0.14517% top0.5% 0.69835% top1% 1.37056%
[2026-10-08T15:28:14Z] p10-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.09456 largest 179 (item 0x207d2) at +6.33 sigma smallest at -5.00; buckets64 chi2/dof 7.08579 largest 8352 (bucket 2138) at +13.98 sigma smallest at -7.05
[2026-10-08T15:28:24Z] p10-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 214 (bin 16348164) z_max +7.60 min 61 z_min -5.92 chi2/dof 3.00028 chi2_z +5793.42 top0.1% 0.14333% top0.5% 0.69129% top1% 1.35803%
[2026-10-08T15:28:24Z] p10-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99993 largest +5.83 sigma; buckets64 chi2/dof 0.99937 largest +4.79 smallest -4.56
[2026-10-08T15:28:24Z] 6-sigma p10-attack-f8 items buckets64 against the window density: largest bucket +13.98 sigma, smallest -7.05 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:28:24Z] hot-set p10-attack-f8 items (windowed control): f 0.1% S_f 0.14517% E_f(control) 0.14333% X_f +0.00184% X_f/f +0.0184 -> no hot set
[2026-10-08T15:28:24Z] hot-set p10-attack-f8 items (windowed control): f 0.5% S_f 0.69835% E_f(control) 0.69129% X_f +0.00707% X_f/f +0.0141 -> no hot set
[2026-10-08T15:28:24Z] hot-set p10-attack-f8 items (windowed control): f 1.0% S_f 1.37056% E_f(control) 1.35803% X_f +0.01253% X_f/f +0.0125 -> no hot set
[2026-10-08T15:28:24Z] hot-set p10-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:28:33Z] p10-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 193 (bin 5932401) z_max +5.75 min 76 z_min -4.60 chi2/dof 1.00015 chi2_z +0.44 top0.1% 0.13106% top0.5% 0.63250% top1% 1.24339%
[2026-10-08T15:28:33Z] hot-set p10-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.14517% E_f(control) 0.13106% X_f +0.01412% X_f/f +0.1412 -> no hot set
[2026-10-08T15:28:33Z] hot-set p10-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.69835% E_f(control) 0.63250% X_f +0.06586% X_f/f +0.1317 -> no hot set
[2026-10-08T15:28:33Z] hot-set p10-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.37056% E_f(control) 1.24339% X_f +0.12717% X_f/f +0.1272 -> no hot set
[2026-10-08T15:28:33Z] hot-set p10-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:28:33Z] ratio p10-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0128x / 1.0102x / 1.0092x (gate 1.2x at 0.1%: within); over the FLAT control 1.1077x / 1.1041x / 1.1023x
[2026-10-08T15:28:34Z] attribution: hot threshold count >= 182 marks 17608 items (0.1050% of items) holding 3268839 reads (0.1522% of reads); top 8 items t=0x6e08cb:214 t=0x7e1623:212 t=0x7d024e:211 t=0x510aba:211 t=0xdd4a23:210 t=0x7f0244:210 t=0x68d58f:210 t=0x760a61:209
[2026-10-08T15:30:29Z] attribution: reads on hot items per position (flat expectation 0.1050% of each position's 16777216 reads); the 24 largest: p23(it1 s7)=0.345% p55(it3 s7)=0.343% p104(it6 s8)=0.343% p56(it3 s8)=0.342% p87(it5 s7)=0.342% p24(it1 s8)=0.342% p40(it2 s8)=0.342% p120(it7 s8)=0.342% p8(it0 s8)=0.341% p7(it0 s7)=0.341% p103(it6 s7)=0.341% p72(it4 s8)=0.341% p88(it5 s8)=0.341% p71(it4 s7)=0.341% p119(it7 s7)=0.341% p39(it2 s7)=0.340% p75(it4 s11)=0.200% p91(it5 s11)=0.199% p43(it2 s11)=0.199% p27(it1 s11)=0.199% p123(it7 s11)=0.198% p107(it6 s11)=0.198% p59(it3 s11)=0.197% p14(it0 s14)=0.194%
[2026-10-08T15:30:29Z] attribution: hot reads by iteration it0=0.149% it1=0.153% it2=0.153% it3=0.153% it4=0.153% it5=0.153% it6=0.152% it7=0.152% ; by site s0=0.091% s1=0.092% s2=0.131% s3=0.000% s4=0.131% s5=0.183% s6=0.160% s7=0.342% s8=0.342% s9=0.000% s10=0.183% s11=0.195% s12=0.171% s13=0.131% s14=0.193% s15=0.092%
[2026-10-08T15:30:29Z] site 0 (instr 1, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0914% (flat expectation 0.1050%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] site 1 (instr 2, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0918% (flat expectation 0.1050%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] site 2 (instr 4, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1312% (flat expectation 0.1050%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:30:29Z] site 3 (instr 5, src r1, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 4 (instr 8, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1308% (flat expectation 0.1050%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:30:29Z] site 5 (instr 18, src r5, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.1826% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 6 (instr 24, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1602% (flat expectation 0.1050%); index entropy 15.962 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0023% (window expectation 0.0015%)
[2026-10-08T15:30:29Z] site 7 (instr 25, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.3417% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 8 (instr 28, src r1, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.3418% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 9 (instr 37, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 10 (instr 39, src r5, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.1828% (flat expectation 0.1050%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:30:29Z] site 11 (instr 41, src r4, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.1949% (flat expectation 0.1050%); index entropy 14.986 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0038% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] site 12 (instr 45, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.1713% (flat expectation 0.1050%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] site 13 (instr 52, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1307% (flat expectation 0.1050%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:30:29Z] site 14 (instr 53, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.1927% (flat expectation 0.1050%); index entropy 14.994 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0048% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] site 15 (instr 57, src r4, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0916% (flat expectation 0.1050%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:30:29Z] attribution: item 0x6e08cb (214 reads) read from positions p23(it1 s7)x10 p39(it2 s7)x9 p30(it1 s14)x8 p119(it7 s7)x8 p11(it0 s11)x7 p120(it7 s8)x7 p24(it1 s8)x6 p40(it2 s8)x6 p43(it2 s11)x6 p70(it4 s6)x6 p87(it5 s7)x6 p88(it5 s8)x6
[2026-10-08T15:30:29Z] attribution: item 0x7e1623 (212 reads) read from positions p103(it6 s7)x10 p71(it4 s7)x9 p72(it4 s8)x9 p87(it5 s7)x9 p23(it1 s7)x8 p56(it3 s8)x8 p62(it3 s14)x8 p8(it0 s8)x7 p75(it4 s11)x7 p120(it7 s8)x7 p123(it7 s11)x5 p28(it1 s12)x5
[2026-10-08T15:30:29Z] attribution: item 0x7d024e (211 reads) read from positions p8(it0 s8)x8 p40(it2 s8)x8 p55(it3 s7)x8 p56(it3 s8)x7 p14(it0 s14)x6 p23(it1 s7)x6 p87(it5 s7)x6 p119(it7 s7)x6 p120(it7 s8)x6 p124(it7 s12)x6 p7(it0 s7)x5 p28(it1 s12)x5
[2026-10-08T15:30:29Z] attribution: item 0x510aba (211 reads) read from positions p55(it3 s7)x9 p87(it5 s7)x9 p7(it0 s7)x8 p107(it6 s11)x8 p14(it0 s14)x6 p56(it3 s8)x6 p120(it7 s8)x6 p8(it0 s8)x5 p23(it1 s7)x5 p28(it1 s12)x5 p59(it3 s11)x5 p60(it3 s12)x5
[2026-10-08T15:30:29Z] attribution: item 0xdd4a23 (210 reads) read from positions p53(it3 s5)x10 p74(it4 s10)x9 p117(it7 s5)x8 p5(it0 s5)x7 p42(it2 s10)x7 p106(it6 s10)x7 p122(it7 s10)x6 p69(it4 s5)x6 p90(it5 s10)x6 p101(it6 s5)x6 p26(it1 s10)x6 p37(it2 s5)x5
[2026-10-08T15:30:29Z] attribution: item 0x7f0244 (210 reads) read from positions p119(it7 s7)x10 p23(it1 s7)x9 p71(it4 s7)x8 p104(it6 s8)x8 p7(it0 s7)x7 p8(it0 s8)x7 p24(it1 s8)x7 p39(it2 s7)x6 p72(it4 s8)x6 p30(it1 s14)x5 p44(it2 s12)x5 p56(it3 s8)x5
[2026-10-08T15:30:29Z] attribution: item 0x68d58f (210 reads) read from positions p104(it6 s8)x9 p119(it7 s7)x9 p56(it3 s8)x8 p55(it3 s7)x7 p71(it4 s7)x7 p87(it5 s7)x7 p108(it6 s12)x7 p126(it7 s14)x6 p88(it5 s8)x6 p120(it7 s8)x6 p7(it0 s7)x5 p8(it0 s8)x5
[2026-10-08T15:30:29Z] attribution: item 0x760a61 (209 reads) read from positions p7(it0 s7)x10 p88(it5 s8)x10 p126(it7 s14)x9 p23(it1 s7)x9 p39(it2 s7)x9 p87(it5 s7)x8 p8(it0 s8)x7 p24(it1 s8)x6 p119(it7 s7)x6 p120(it7 s8)x6 p28(it1 s12)x5 p43(it2 s11)x5
[2026-10-08T15:30:29Z] hottest item p10-attack-f8 (0x6e08cb): 214 reads (0.0000% of all); predicted source none
[2026-10-08T15:30:29Z] program p10-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
Terminated
lease: released 8 pool cores after 1333 s, exit 0

142
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p10/warps-p10-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,142 @@
# distinct lines_hash: value count
lines_hash 1008 1
lines_hash 1013 3
lines_hash 1014 56
lines_hash 1015 904
lines_hash 1016 7149
lines_hash 1020 136
lines_hash 1021 4786
lines_hash 1022 115177
lines_hash 1023 1850677
lines_hash 1024 14798327
# distinct items_hash: value count
items_hash 126 1
items_hash 127 8112
items_hash 128 16769103
# distinct lines_warp: value count
lines_warp 32566 1
lines_warp 32568 1
lines_warp 32571 1
lines_warp 32572 1
lines_warp 32573 1
lines_warp 32576 3
lines_warp 32577 2
lines_warp 32578 2
lines_warp 32579 3
lines_warp 32580 7
lines_warp 32581 5
lines_warp 32582 9
lines_warp 32583 11
lines_warp 32584 13
lines_warp 32585 18
lines_warp 32586 25
lines_warp 32587 33
lines_warp 32588 31
lines_warp 32589 48
lines_warp 32590 53
lines_warp 32591 65
lines_warp 32592 70
lines_warp 32593 102
lines_warp 32594 135
lines_warp 32595 150
lines_warp 32596 160
lines_warp 32597 231
lines_warp 32598 264
lines_warp 32599 325
lines_warp 32600 383
lines_warp 32601 464
lines_warp 32602 561
lines_warp 32603 698
lines_warp 32604 768
lines_warp 32605 932
lines_warp 32606 1024
lines_warp 32607 1309
lines_warp 32608 1464
lines_warp 32609 1737
lines_warp 32610 1964
lines_warp 32611 2303
lines_warp 32612 2743
lines_warp 32613 2974
lines_warp 32614 3466
lines_warp 32615 3852
lines_warp 32616 4329
lines_warp 32617 4913
lines_warp 32618 5432
lines_warp 32619 6124
lines_warp 32620 6713
lines_warp 32621 7369
lines_warp 32622 8119
lines_warp 32623 8821
lines_warp 32624 9607
lines_warp 32625 10547
lines_warp 32626 11011
lines_warp 32627 11863
lines_warp 32628 12499
lines_warp 32629 13466
lines_warp 32630 14209
lines_warp 32631 14663
lines_warp 32632 15361
lines_warp 32633 15921
lines_warp 32634 16283
lines_warp 32635 16345
lines_warp 32636 16567
lines_warp 32637 16768
lines_warp 32638 16779
lines_warp 32639 16534
lines_warp 32640 16189
lines_warp 32641 15912
lines_warp 32642 15613
lines_warp 32643 15093
lines_warp 32644 14562
lines_warp 32645 13896
lines_warp 32646 13299
lines_warp 32647 12375
lines_warp 32648 11433
lines_warp 32649 10630
lines_warp 32650 9710
lines_warp 32651 8997
lines_warp 32652 8250
lines_warp 32653 7209
lines_warp 32654 6464
lines_warp 32655 5662
lines_warp 32656 5024
lines_warp 32657 4304
lines_warp 32658 3645
lines_warp 32659 3147
lines_warp 32660 2771
lines_warp 32661 2244
lines_warp 32662 1905
lines_warp 32663 1517
lines_warp 32664 1263
lines_warp 32665 1033
lines_warp 32666 801
lines_warp 32667 581
lines_warp 32668 496
lines_warp 32669 405
lines_warp 32670 324
lines_warp 32671 229
lines_warp 32672 163
lines_warp 32673 135
lines_warp 32674 114
lines_warp 32675 70
lines_warp 32676 48
lines_warp 32677 31
lines_warp 32678 23
lines_warp 32679 20
lines_warp 32680 18
lines_warp 32681 12
lines_warp 32682 8
lines_warp 32683 2
lines_warp 32684 1
lines_warp 32685 1
lines_warp 32686 3
lines_warp 32688 1
# distinct items_warp: value count
items_warp 4089 1
items_warp 4090 14
items_warp 4091 96
items_warp 4092 862
items_warp 4093 6889
items_warp 4094 40446
items_warp 4095 159941
items_warp 4096 316039

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p10/warps-p10-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 1 0.000036 0.000000
1 0 1 2 0.000037 0.000000
2 0 2 4 0.000019 0.000000
3 0 3 5 0.000069 0.000000
4 0 4 8 0.000020 0.000000
5 0 5 18 0.000068 0.000000
6 0 6 24 0.000019 0.000000
7 0 7 25 0.000069 0.000000
8 0 8 28 0.000068 0.000000
9 0 9 37 0.000068 0.000000
10 0 10 39 0.000069 0.000000
11 0 11 41 0.000036 0.000000
12 0 12 45 0.000036 0.000000
13 0 13 52 0.000020 0.000000
14 0 14 53 0.000050 0.000000
15 0 15 57 0.000036 0.000000
16 1 0 1 0.000036 0.000000
17 1 1 2 0.000036 0.000000
18 1 2 4 0.000020 0.000000
19 1 3 5 0.000068 0.000000
20 1 4 8 0.000020 0.000000
21 1 5 18 0.000069 0.000000
22 1 6 24 0.000027 0.000000
23 1 7 25 0.000070 0.000000
24 1 8 28 0.000068 0.000000
25 1 9 37 0.000070 0.000000
26 1 10 39 0.000069 0.000000
27 1 11 41 0.000043 0.000000
28 1 12 45 0.000037 0.000000
29 1 13 52 0.000020 0.000000
30 1 14 53 0.000050 0.000000
31 1 15 57 0.000036 0.000000
32 2 0 1 0.000037 0.000000
33 2 1 2 0.000036 0.000000
34 2 2 4 0.000020 0.000000
35 2 3 5 0.000069 0.000000
36 2 4 8 0.000019 0.000000
37 2 5 18 0.000069 0.000000
38 2 6 24 0.000026 0.000000
39 2 7 25 0.000068 0.000000
40 2 8 28 0.000069 0.000000
41 2 9 37 0.000068 0.000000
42 2 10 39 0.000069 0.000000
43 2 11 41 0.000042 0.000000
44 2 12 45 0.000037 0.000000
45 2 13 52 0.000020 0.000000
46 2 14 53 0.000051 0.000000
47 2 15 57 0.000036 0.000000
48 3 0 1 0.000036 0.000000
49 3 1 2 0.000037 0.000000
50 3 2 4 0.000020 0.000000
51 3 3 5 0.000068 0.000000
52 3 4 8 0.000020 0.000000
53 3 5 18 0.000069 0.000000
54 3 6 24 0.000026 0.000000
55 3 7 25 0.000069 0.000000
56 3 8 28 0.000069 0.000000
57 3 9 37 0.000068 0.000000
58 3 10 39 0.000068 0.000000
59 3 11 41 0.000043 0.000000
60 3 12 45 0.000037 0.000000
61 3 13 52 0.000019 0.000000
62 3 14 53 0.000051 0.000000
63 3 15 57 0.000036 0.000000
64 4 0 1 0.000036 0.000000
65 4 1 2 0.000037 0.000000
66 4 2 4 0.000019 0.000000
67 4 3 5 0.000069 0.000000
68 4 4 8 0.000020 0.000000
69 4 5 18 0.000068 0.000000
70 4 6 24 0.000026 0.000000
71 4 7 25 0.000069 0.000000
72 4 8 28 0.000068 0.000000
73 4 9 37 0.000070 0.000000
74 4 10 39 0.000068 0.000000
75 4 11 41 0.000043 0.000000
76 4 12 45 0.000037 0.000000
77 4 13 52 0.000019 0.000000
78 4 14 53 0.000051 0.000000
79 4 15 57 0.000036 0.000000
80 5 0 1 0.000036 0.000000
81 5 1 2 0.000036 0.000000
82 5 2 4 0.000019 0.000000
83 5 3 5 0.000068 0.000000
84 5 4 8 0.000019 0.000000
85 5 5 18 0.000069 0.000000
86 5 6 24 0.000026 0.000000
87 5 7 25 0.000069 0.000000
88 5 8 28 0.000069 0.000000
89 5 9 37 0.000068 0.000000
90 5 10 39 0.000068 0.000000
91 5 11 41 0.000043 0.000000
92 5 12 45 0.000036 0.000000
93 5 13 52 0.000020 0.000000
94 5 14 53 0.000051 0.000000
95 5 15 57 0.000036 0.000000
96 6 0 1 0.000036 0.000000
97 6 1 2 0.000036 0.000000
98 6 2 4 0.000020 0.000000
99 6 3 5 0.000068 0.000000
100 6 4 8 0.000019 0.000000
101 6 5 18 0.000068 0.000000
102 6 6 24 0.000027 0.000000
103 6 7 25 0.000069 0.000000
104 6 8 28 0.000070 0.000000
105 6 9 37 0.000069 0.000000
106 6 10 39 0.000068 0.000000
107 6 11 41 0.000042 0.000000
108 6 12 45 0.000036 0.000000
109 6 13 52 0.000019 0.000000
110 6 14 53 0.000049 0.000000
111 6 15 57 0.000036 0.000000
112 7 0 1 0.000036 0.000000
113 7 1 2 0.000036 0.000000
114 7 2 4 0.000019 0.000000
115 7 3 5 0.000068 0.000000
116 7 4 8 0.000020 0.000000
117 7 5 18 0.000069 0.000000
118 7 6 24 0.000027 0.000000
119 7 7 25 0.000070 0.000000
120 7 8 28 0.000068 0.000000
121 7 9 37 0.000069 0.000000
122 7 10 39 0.000068 0.000000
123 7 11 41 0.000043 0.000000
124 7 12 45 0.000036 0.000000
125 7 13 52 0.000020 0.000000
126 7 14 53 0.000050 0.000000
127 7 15 57 0.000036 0.000000

77
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p15.log Normal file
View file

@ -0,0 +1,77 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (8,9,10,11,16,18,19,20, waited 70 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p15 at 2^24 (window control, by site)
[2026-10-08T15:09:28Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "15..15", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p15"]
[2026-10-08T15:09:28Z] census warps: programs [15] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:09:28Z] day 20730: cache filled in 0.35 s, fnv 448274a57f508cbc
[2026-10-08T15:09:36Z] table: 16777216 items derived with their 8 lines in 7.5 s; library comparison all (0 mismatches)
[2026-10-08T15:09:38Z] program p15-attack-f8: epoch seed 65d1bc6af696d940f57d7d04469e29dd7c1eb69055619045911121b1dcec8b69 era seed e6b5324458cfd3c3326baed2aabd49cc361a17bc5efdde9e291daddad21ec5c7 id 69baaa1fb258acb9 attempt 1 class mx8-era676c3949+sh256x27+state+rw op mix load=16 add=15 sub=8 rotr=7 xor=6 mul=4 rotl=4 mad=3 mulhi=1; era stride mul 0x26f35c81 rot 26 interleave [0, 2, 14, 15] windows 2:2:2 5:2:2 6:0:0 7:1:1 8:0:0 12:0:0 15:2:0 24:2:2 29:0:0 30:1:0 36:2:3 40:2:3 44:0:0 48:0:0 56:0:0 63:1:1
[2026-10-08T15:09:38Z] load site instr 2: src r7 win 2 off 2; last base writer sub at 1; writers back to the last injecting one: sub@1; shadow writes of r7 per rep: add=4 mad=8 mul=1 rotl=6 rotr=1 shfl=1 sub=3 xor=6
[2026-10-08T15:09:38Z] load site instr 5: src r6 win 2 off 2; last base writer xor at 4; writers back to the last injecting one: xor@4; shadow writes of r6 per rep: add=2 mad=6 mul=1 rotl=3 rotr=2 shfl=1 sub=3 xor=9
[2026-10-08T15:09:38Z] load site instr 6: src r7 win 0 off 0; last base writer load at 5; writers back to the last injecting one: load@5; shadow writes of r7 per rep: add=4 mad=8 mul=1 rotl=6 rotr=1 shfl=1 sub=3 xor=6
[2026-10-08T15:09:38Z] load site instr 7: src r4 win 1 off 1; last base writer load at 6; writers back to the last injecting one: load@6; shadow writes of r4 per rep: add=7 mad=5 mul=4 mulhi=1 rotl=3 rotr=4 sub=3 xor=7
[2026-10-08T15:09:38Z] load site instr 8: src r7 win 0 off 0; last base writer load at 7; writers back to the last injecting one: load@7; shadow writes of r7 per rep: add=4 mad=8 mul=1 rotl=6 rotr=1 shfl=1 sub=3 xor=6
[2026-10-08T15:09:38Z] load site instr 12: src r6 win 0 off 0; last base writer rotr at 10; writers back to the last injecting one: rotr@10 xor@4; shadow writes of r6 per rep: add=2 mad=6 mul=1 rotl=3 rotr=2 shfl=1 sub=3 xor=9
[2026-10-08T15:09:38Z] load site instr 15: src r7 win 2 off 0; last base writer add at 13; writers back to the last injecting one: add@13; shadow writes of r7 per rep: add=4 mad=8 mul=1 rotl=6 rotr=1 shfl=1 sub=3 xor=6
[2026-10-08T15:09:38Z] load site instr 24: src r1 win 2 off 2; last base writer add at 23; writers back to the last injecting one: add@23; shadow writes of r1 per rep: add=4 mad=7 mul=4 rotl=5 rotr=3 shfl=1 sub=8 xor=1
[2026-10-08T15:09:38Z] load site instr 29: src r0 win 0 off 0; last base writer xor at 14; writers back to the last injecting one: xor@14; shadow writes of r0 per rep: add=4 mad=1 mulhi=1 rotl=7 rotr=5 shfl=2 sub=2 xor=5
[2026-10-08T15:09:38Z] load site instr 30: src r5 win 1 off 0; last base writer load at 8; writers back to the last injecting one: load@8; shadow writes of r5 per rep: add=9 mad=6 mul=1 mulhi=4 rotl=5 rotr=5 shfl=4 sub=6 xor=5
[2026-10-08T15:09:38Z] load site instr 36: src r1 win 2 off 3; last base writer sub at 35; writers back to the last injecting one: sub@35; shadow writes of r1 per rep: add=4 mad=7 mul=4 rotl=5 rotr=3 shfl=1 sub=8 xor=1
[2026-10-08T15:09:38Z] load site instr 40: src r2 win 2 off 3; last base writer xor at 37; writers back to the last injecting one: xor@37; shadow writes of r2 per rep: add=5 mad=7 mul=1 mulhi=2 rotl=5 rotr=2 shfl=2 sub=5 xor=7
[2026-10-08T15:09:38Z] load site instr 44: src r1 win 0 off 0; last base writer mad at 41; writers back to the last injecting one: mad@41; shadow writes of r1 per rep: add=4 mad=7 mul=4 rotl=5 rotr=3 shfl=1 sub=8 xor=1
[2026-10-08T15:09:38Z] load site instr 48: src r4 win 0 off 0; last base writer rotr at 46; writers back to the last injecting one: rotr@46 load@44; shadow writes of r4 per rep: add=7 mad=5 mul=4 mulhi=1 rotl=3 rotr=4 sub=3 xor=7
[2026-10-08T15:09:38Z] load site instr 56: src r7 win 0 off 0; last base writer rotr at 43; writers back to the last injecting one: rotr@43 xor@38; shadow writes of r7 per rep: add=4 mad=8 mul=1 rotl=6 rotr=1 shfl=1 sub=3 xor=6
[2026-10-08T15:09:38Z] load site instr 63: src r0 win 1 off 1; last base writer rotr at 51; writers back to the last injecting one: rotr@51 xor@14; shadow writes of r0 per rep: add=4 mad=1 mulhi=1 rotl=7 rotr=5 shfl=2 sub=2 xor=5
[2026-10-08T15:09:38Z] window layer: site item windows (first, items) (0x800000,2^22) (0x800000,2^22) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x800000,2^22) (0x0,2^24) (0x0,2^23) (0xc00000,2^22) (0xc00000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x800000,2^23); expected reads per item by quarter 104.000 72.000 184.000 152.000 (flat uniform 128.000)
[2026-10-08T15:12:20Z] 524288 warps (16777216 nonces) interpreted in 162.2 s (2.475 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:12:20Z] p15-attack-f8 distinct lines per hash: n 16777216 min 976 p1 1008 median 1024 p99 1024 max 1024 mean 1021.8695
[2026-10-08T15:12:20Z] p15-attack-f8 distinct items per hash: n 16777216 min 122 p1 126 median 128 p99 128 max 128 mean 127.7492
[2026-10-08T15:12:20Z] p15-attack-f8 distinct lines per warp: n 524288 min 32433 p1 32508 median 32573 p99 32627 max 32676 mean 32572.3743
[2026-10-08T15:12:20Z] p15-attack-f8 distinct items per warp: n 524288 min 4070 p1 4080 median 4088 p99 4093 max 4096 mean 4087.4405
[2026-10-08T15:12:20Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:12:20Z] positions: largest hi16-bucket (256 items) share 0.0070% at p1 (iteration 0, site 1, instr 5); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p46 (site 14, instr 56)
[2026-10-08T15:12:20Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p64 (site 0, instr 2); saturated sources at one position 0 of 2048 at p0 (site 0, instr 2); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:12:20Z] p15-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 261 (bin 9132859) z_max +11.76 min 31 z_min -8.57 chi2/dof 15.38407 chi2_z +41660.73 top0.1% 0.17606% top0.5% 0.84963% top1% 1.66929%
[2026-10-08T15:12:21Z] p15-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.11137 largest 166 (item 0xce266) at +6.08 sigma smallest at -4.95; buckets64 chi2/dof 8.00498 largest 7376 (bucket 13897) at +8.83 sigma smallest at -6.69
[2026-10-08T15:12:29Z] p15-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 260 (bin 9655594) z_max +11.67 min 30 z_min -8.66 chi2/dof 15.49839 chi2_z +41991.82 top0.1% 0.17612% top0.5% 0.85017% top1% 1.67049%
[2026-10-08T15:12:29Z] p15-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99951 largest +5.98 sigma; buckets64 chi2/dof 1.00160 largest +4.57 smallest -4.34
[2026-10-08T15:12:29Z] 6-sigma p15-attack-f8 items buckets64 against the window density: largest bucket +8.83 sigma, smallest -6.69 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:12:29Z] hot-set p15-attack-f8 items (windowed control): f 0.1% S_f 0.17606% E_f(control) 0.17612% X_f -0.00006% X_f/f -0.0006 -> no hot set
[2026-10-08T15:12:29Z] hot-set p15-attack-f8 items (windowed control): f 0.5% S_f 0.84963% E_f(control) 0.85017% X_f -0.00054% X_f/f -0.0011 -> no hot set
[2026-10-08T15:12:29Z] hot-set p15-attack-f8 items (windowed control): f 1.0% S_f 1.66929% E_f(control) 1.67049% X_f -0.00120% X_f/f -0.0012 -> no hot set
[2026-10-08T15:12:29Z] hot-set p15-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:12:37Z] p15-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 194 (bin 10872769) z_max +5.83 min 71 z_min -5.04 chi2/dof 0.99996 chi2_z -0.11 top0.1% 0.13110% top0.5% 0.63250% top1% 1.24332%
[2026-10-08T15:12:37Z] hot-set p15-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17606% E_f(control) 0.13110% X_f +0.04496% X_f/f +0.4496 -> no hot set
[2026-10-08T15:12:37Z] hot-set p15-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.84963% E_f(control) 0.63250% X_f +0.21714% X_f/f +0.4343 -> no hot set
[2026-10-08T15:12:37Z] hot-set p15-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.66929% E_f(control) 1.24332% X_f +0.42597% X_f/f +0.4260 -> no hot set
[2026-10-08T15:12:37Z] hot-set p15-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:12:37Z] ratio p15-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9997x / 0.9994x / 0.9993x (gate 1.2x at 0.1%: within); over the FLAT control 1.3430x / 1.3433x / 1.3426x
[2026-10-08T15:12:38Z] attribution: hot threshold count >= 221 marks 18076 items (0.1077% of items) holding 4067931 reads (0.1894% of reads); top 8 items t=0x8b5b3b:261 t=0xbfed2d:254 t=0xba3bd7:254 t=0xa44071:254 t=0x866d81:252 t=0xb9c061:251 t=0x91de3e:250 t=0x8ca132:250
[2026-10-08T15:14:00Z] attribution: reads on hot items per position (flat expectation 0.1077% of each position's 16777216 reads); the 24 largest: p81(it5 s1)=0.531% p113(it7 s1)=0.530% p33(it2 s1)=0.529% p97(it6 s1)=0.529% p23(it1 s7)=0.529% p17(it1 s1)=0.529% p65(it4 s1)=0.528% p119(it7 s7)=0.528% p87(it5 s7)=0.528% p16(it1 s0)=0.528% p1(it0 s1)=0.528% p96(it6 s0)=0.527% p55(it3 s7)=0.527% p80(it5 s0)=0.527% p48(it3 s0)=0.527% p49(it3 s1)=0.527% p32(it2 s0)=0.527% p112(it7 s0)=0.527% p39(it2 s7)=0.526% p7(it0 s7)=0.526% p64(it4 s0)=0.526% p71(it4 s7)=0.525% p0(it0 s0)=0.525% p103(it6 s7)=0.525%
[2026-10-08T15:14:00Z] attribution: hot reads by iteration it0=0.189% it1=0.190% it2=0.189% it3=0.189% it4=0.189% it5=0.190% it6=0.189% it7=0.190% ; by site s0=0.527% s1=0.529% s2=0.131% s3=0.263% s4=0.131% s5=0.130% s6=0.000% s7=0.527% s8=0.132% s9=0.000% s10=0.000% s11=0.000% s12=0.132% s13=0.132% s14=0.132% s15=0.264%
[2026-10-08T15:14:00Z] site 0 (instr 2, src r7, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.5268% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 1 (instr 5, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.5287% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 2 (instr 6, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1315% (flat expectation 0.1077%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 3 (instr 7, src r4, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2629% (flat expectation 0.1077%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:00Z] site 4 (instr 8, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1314% (flat expectation 0.1077%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 5 (instr 12, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1298% (flat expectation 0.1077%); index entropy 15.991 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0020% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 6 (instr 15, src r7, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 7 (instr 24, src r1, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.5269% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 8 (instr 29, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1325% (flat expectation 0.1077%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 9 (instr 30, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1077%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:14:00Z] site 10 (instr 36, src r1, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 11 (instr 40, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1077%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:00Z] site 12 (instr 44, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1325% (flat expectation 0.1077%); index entropy 15.936 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0025% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 13 (instr 48, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1321% (flat expectation 0.1077%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 14 (instr 56, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1315% (flat expectation 0.1077%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:00Z] site 15 (instr 63, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2644% (flat expectation 0.1077%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:00Z] attribution: item 0x8b5b3b (261 reads) read from positions p17(it1 s1)x9 p80(it5 s0)x9 p48(it3 s0)x8 p87(it5 s7)x8 p15(it0 s15)x7 p33(it2 s1)x7 p39(it2 s7)x7 p115(it7 s3)x6 p50(it3 s2)x6 p55(it3 s7)x6 p63(it3 s15)x6 p71(it4 s7)x6
[2026-10-08T15:14:00Z] attribution: item 0xbfed2d (254 reads) read from positions p39(it2 s7)x13 p96(it6 s0)x9 p65(it4 s1)x8 p23(it1 s7)x7 p33(it2 s1)x7 p51(it3 s3)x7 p87(it5 s7)x7 p115(it7 s3)x6 p47(it2 s15)x6 p49(it3 s1)x6 p55(it3 s7)x6 p64(it4 s0)x6
[2026-10-08T15:14:00Z] attribution: item 0xba3bd7 (254 reads) read from positions p96(it6 s0)x12 p64(it4 s0)x10 p127(it7 s15)x7 p0(it0 s0)x7 p33(it2 s1)x7 p87(it5 s7)x7 p97(it6 s1)x7 p16(it1 s0)x6 p39(it2 s7)x6 p55(it3 s7)x6 p65(it4 s1)x6 p23(it1 s7)x5
[2026-10-08T15:14:00Z] attribution: item 0xa44071 (254 reads) read from positions p81(it5 s1)x10 p16(it1 s0)x8 p97(it6 s1)x8 p64(it4 s0)x7 p87(it5 s7)x7 p119(it7 s7)x6 p33(it2 s1)x6 p0(it0 s0)x6 p103(it6 s7)x6 p19(it1 s3)x5 p23(it1 s7)x5 p32(it2 s0)x5
[2026-10-08T15:14:00Z] attribution: item 0x866d81 (252 reads) read from positions p33(it2 s1)x10 p97(it6 s1)x10 p39(it2 s7)x9 p55(it3 s7)x9 p48(it3 s0)x8 p112(it7 s0)x8 p96(it6 s0)x7 p17(it1 s1)x6 p49(it3 s1)x6 p64(it4 s0)x6 p80(it5 s0)x6 p81(it5 s1)x6
[2026-10-08T15:14:00Z] attribution: item 0xb9c061 (251 reads) read from positions p119(it7 s7)x11 p16(it1 s0)x10 p87(it5 s7)x10 p65(it4 s1)x9 p80(it5 s0)x9 p49(it3 s1)x8 p97(it6 s1)x8 p64(it4 s0)x7 p113(it7 s1)x7 p7(it0 s7)x6 p48(it3 s0)x6 p71(it4 s7)x6
[2026-10-08T15:14:00Z] attribution: item 0x91de3e (250 reads) read from positions p17(it1 s1)x9 p49(it3 s1)x9 p7(it0 s7)x8 p97(it6 s1)x8 p31(it1 s15)x7 p33(it2 s1)x7 p48(it3 s0)x7 p64(it4 s0)x7 p119(it7 s7)x6 p23(it1 s7)x6 p71(it4 s7)x6 p79(it4 s15)x5
[2026-10-08T15:14:00Z] attribution: item 0x8ca132 (250 reads) read from positions p49(it3 s1)x16 p16(it1 s0)x10 p81(it5 s1)x9 p39(it2 s7)x8 p113(it7 s1)x8 p33(it2 s1)x7 p64(it4 s0)x7 p80(it5 s0)x7 p55(it3 s7)x6 p0(it0 s0)x6 p82(it5 s2)x6 p97(it6 s1)x6
[2026-10-08T15:14:00Z] hottest item p15-attack-f8 (0x8b5b3b): 261 reads (0.0000% of all); predicted source none
[2026-10-08T15:14:00Z] program p15-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
lease: released 8 pool cores after 342 s, exit 0

287
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p15/warps-p15-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,287 @@
# distinct lines_hash: value count
lines_hash 976 3
lines_hash 982 1
lines_hash 983 4
lines_hash 984 40
lines_hash 990 8
lines_hash 991 144
lines_hash 992 1374
lines_hash 997 11
lines_hash 998 175
lines_hash 999 3216
lines_hash 1000 26008
lines_hash 1004 3
lines_hash 1005 113
lines_hash 1006 2566
lines_hash 1007 42449
lines_hash 1008 351114
lines_hash 1011 1
lines_hash 1012 29
lines_hash 1013 943
lines_hash 1014 22346
lines_hash 1015 361739
lines_hash 1016 2934906
lines_hash 1019 1
lines_hash 1020 106
lines_hash 1021 3786
lines_hash 1022 89406
lines_hash 1023 1436441
lines_hash 1024 11500283
# distinct items_hash: value count
items_hash 122 3
items_hash 123 45
items_hash 124 1526
items_hash 125 29410
items_hash 126 396245
items_hash 127 3319964
items_hash 128 13030023
# distinct lines_warp: value count
lines_warp 32433 1
lines_warp 32441 1
lines_warp 32444 2
lines_warp 32446 1
lines_warp 32449 2
lines_warp 32450 1
lines_warp 32451 3
lines_warp 32452 3
lines_warp 32453 2
lines_warp 32454 5
lines_warp 32456 2
lines_warp 32457 5
lines_warp 32458 3
lines_warp 32459 2
lines_warp 32460 6
lines_warp 32461 7
lines_warp 32462 5
lines_warp 32463 4
lines_warp 32464 3
lines_warp 32465 3
lines_warp 32466 9
lines_warp 32467 9
lines_warp 32468 13
lines_warp 32469 16
lines_warp 32470 10
lines_warp 32471 19
lines_warp 32472 14
lines_warp 32473 19
lines_warp 32474 26
lines_warp 32475 20
lines_warp 32476 34
lines_warp 32477 27
lines_warp 32478 21
lines_warp 32479 29
lines_warp 32480 37
lines_warp 32481 35
lines_warp 32482 45
lines_warp 32483 61
lines_warp 32484 57
lines_warp 32485 59
lines_warp 32486 70
lines_warp 32487 74
lines_warp 32488 70
lines_warp 32489 90
lines_warp 32490 77
lines_warp 32491 101
lines_warp 32492 114
lines_warp 32493 130
lines_warp 32494 132
lines_warp 32495 147
lines_warp 32496 171
lines_warp 32497 157
lines_warp 32498 167
lines_warp 32499 212
lines_warp 32500 227
lines_warp 32501 271
lines_warp 32502 293
lines_warp 32503 297
lines_warp 32504 333
lines_warp 32505 348
lines_warp 32506 381
lines_warp 32507 397
lines_warp 32508 457
lines_warp 32509 512
lines_warp 32510 553
lines_warp 32511 575
lines_warp 32512 601
lines_warp 32513 703
lines_warp 32514 678
lines_warp 32515 772
lines_warp 32516 789
lines_warp 32517 879
lines_warp 32518 975
lines_warp 32519 997
lines_warp 32520 1109
lines_warp 32521 1102
lines_warp 32522 1295
lines_warp 32523 1295
lines_warp 32524 1455
lines_warp 32525 1540
lines_warp 32526 1602
lines_warp 32527 1764
lines_warp 32528 1759
lines_warp 32529 1867
lines_warp 32530 2090
lines_warp 32531 2158
lines_warp 32532 2216
lines_warp 32533 2385
lines_warp 32534 2457
lines_warp 32535 2687
lines_warp 32536 2830
lines_warp 32537 2898
lines_warp 32538 3135
lines_warp 32539 3316
lines_warp 32540 3427
lines_warp 32541 3638
lines_warp 32542 3639
lines_warp 32543 3899
lines_warp 32544 4111
lines_warp 32545 4320
lines_warp 32546 4525
lines_warp 32547 4618
lines_warp 32548 4814
lines_warp 32549 4998
lines_warp 32550 5193
lines_warp 32551 5388
lines_warp 32552 5494
lines_warp 32553 5695
lines_warp 32554 5949
lines_warp 32555 6158
lines_warp 32556 6197
lines_warp 32557 6300
lines_warp 32558 6469
lines_warp 32559 6844
lines_warp 32560 6931
lines_warp 32561 7010
lines_warp 32562 7125
lines_warp 32563 7259
lines_warp 32564 7409
lines_warp 32565 7484
lines_warp 32566 7621
lines_warp 32567 7803
lines_warp 32568 7983
lines_warp 32569 7870
lines_warp 32570 7928
lines_warp 32571 8064
lines_warp 32572 8021
lines_warp 32573 8005
lines_warp 32574 8264
lines_warp 32575 8285
lines_warp 32576 8304
lines_warp 32577 8007
lines_warp 32578 8239
lines_warp 32579 8214
lines_warp 32580 8084
lines_warp 32581 8017
lines_warp 32582 7925
lines_warp 32583 7868
lines_warp 32584 7689
lines_warp 32585 7658
lines_warp 32586 7521
lines_warp 32587 7386
lines_warp 32588 7177
lines_warp 32589 7187
lines_warp 32590 6828
lines_warp 32591 6792
lines_warp 32592 6417
lines_warp 32593 6480
lines_warp 32594 6079
lines_warp 32595 5978
lines_warp 32596 5899
lines_warp 32597 5717
lines_warp 32598 5249
lines_warp 32599 5226
lines_warp 32600 4993
lines_warp 32601 4704
lines_warp 32602 4525
lines_warp 32603 4423
lines_warp 32604 4119
lines_warp 32605 3854
lines_warp 32606 3703
lines_warp 32607 3491
lines_warp 32608 3267
lines_warp 32609 3054
lines_warp 32610 2899
lines_warp 32611 2767
lines_warp 32612 2675
lines_warp 32613 2381
lines_warp 32614 2223
lines_warp 32615 2102
lines_warp 32616 2015
lines_warp 32617 1819
lines_warp 32618 1658
lines_warp 32619 1557
lines_warp 32620 1383
lines_warp 32621 1284
lines_warp 32622 1107
lines_warp 32623 1019
lines_warp 32624 969
lines_warp 32625 869
lines_warp 32626 778
lines_warp 32627 673
lines_warp 32628 605
lines_warp 32629 542
lines_warp 32630 508
lines_warp 32631 426
lines_warp 32632 381
lines_warp 32633 356
lines_warp 32634 290
lines_warp 32635 284
lines_warp 32636 242
lines_warp 32637 190
lines_warp 32638 189
lines_warp 32639 144
lines_warp 32640 125
lines_warp 32641 96
lines_warp 32642 95
lines_warp 32643 75
lines_warp 32644 85
lines_warp 32645 57
lines_warp 32646 54
lines_warp 32647 36
lines_warp 32648 36
lines_warp 32649 34
lines_warp 32650 23
lines_warp 32651 17
lines_warp 32652 16
lines_warp 32653 21
lines_warp 32654 5
lines_warp 32655 9
lines_warp 32656 5
lines_warp 32657 5
lines_warp 32658 1
lines_warp 32659 5
lines_warp 32660 2
lines_warp 32661 2
lines_warp 32663 2
lines_warp 32664 1
lines_warp 32666 1
lines_warp 32671 1
lines_warp 32676 1
# distinct items_warp: value count
items_warp 4070 1
items_warp 4071 1
items_warp 4072 1
items_warp 4073 11
items_warp 4074 34
items_warp 4075 68
items_warp 4076 168
items_warp 4077 420
items_warp 4078 852
items_warp 4079 1940
items_warp 4080 3740
items_warp 4081 7250
items_warp 4082 12908
items_warp 4083 20983
items_warp 4084 32635
items_warp 4085 46124
items_warp 4086 58972
items_warp 4087 68730
items_warp 4088 72493
items_warp 4089 67136
items_warp 4090 54603
items_warp 4091 38098
items_warp 4092 22280
items_warp 4093 10340
items_warp 4094 3583
items_warp 4095 826
items_warp 4096 91

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p15/warps-p15-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 2 0.000069 0.000000
1 0 1 5 0.000070 0.000000
2 0 2 6 0.000020 0.000000
3 0 3 7 0.000036 0.000000
4 0 4 8 0.000020 0.000000
5 0 5 12 0.000022 0.000000
6 0 6 15 0.000070 0.000000
7 0 7 24 0.000070 0.000000
8 0 8 29 0.000019 0.000000
9 0 9 30 0.000036 0.000000
10 0 10 36 0.000068 0.000000
11 0 11 40 0.000069 0.000000
12 0 12 44 0.000028 0.000000
13 0 13 48 0.000020 0.000000
14 0 14 56 0.000020 0.000000
15 0 15 63 0.000036 0.000000
16 1 0 2 0.000069 0.000000
17 1 1 5 0.000068 0.000000
18 1 2 6 0.000019 0.000000
19 1 3 7 0.000036 0.000000
20 1 4 8 0.000020 0.000000
21 1 5 12 0.000022 0.000000
22 1 6 15 0.000068 0.000000
23 1 7 24 0.000069 0.000000
24 1 8 29 0.000020 0.000000
25 1 9 30 0.000036 0.000000
26 1 10 36 0.000068 0.000000
27 1 11 40 0.000069 0.000000
28 1 12 44 0.000027 0.000000
29 1 13 48 0.000020 0.000000
30 1 14 56 0.000019 0.000000
31 1 15 63 0.000037 0.000000
32 2 0 2 0.000068 0.000000
33 2 1 5 0.000069 0.000000
34 2 2 6 0.000020 0.000000
35 2 3 7 0.000037 0.000000
36 2 4 8 0.000020 0.000000
37 2 5 12 0.000023 0.000000
38 2 6 15 0.000070 0.000000
39 2 7 24 0.000069 0.000000
40 2 8 29 0.000020 0.000000
41 2 9 30 0.000036 0.000000
42 2 10 36 0.000068 0.000000
43 2 11 40 0.000068 0.000000
44 2 12 44 0.000028 0.000000
45 2 13 48 0.000020 0.000000
46 2 14 56 0.000020 0.000000
47 2 15 63 0.000037 0.000000
48 3 0 2 0.000069 0.000000
49 3 1 5 0.000069 0.000000
50 3 2 6 0.000019 0.000000
51 3 3 7 0.000036 0.000000
52 3 4 8 0.000020 0.000000
53 3 5 12 0.000022 0.000000
54 3 6 15 0.000068 0.000000
55 3 7 24 0.000068 0.000000
56 3 8 29 0.000020 0.000000
57 3 9 30 0.000036 0.000000
58 3 10 36 0.000069 0.000000
59 3 11 40 0.000068 0.000000
60 3 12 44 0.000027 0.000000
61 3 13 48 0.000019 0.000000
62 3 14 56 0.000019 0.000000
63 3 15 63 0.000037 0.000000
64 4 0 2 0.000069 0.000000
65 4 1 5 0.000069 0.000000
66 4 2 6 0.000020 0.000000
67 4 3 7 0.000037 0.000000
68 4 4 8 0.000020 0.000000
69 4 5 12 0.000022 0.000000
70 4 6 15 0.000070 0.000000
71 4 7 24 0.000068 0.000000
72 4 8 29 0.000019 0.000000
73 4 9 30 0.000036 0.000000
74 4 10 36 0.000069 0.000000
75 4 11 40 0.000068 0.000000
76 4 12 44 0.000029 0.000000
77 4 13 48 0.000020 0.000000
78 4 14 56 0.000020 0.000000
79 4 15 63 0.000037 0.000000
80 5 0 2 0.000068 0.000000
81 5 1 5 0.000068 0.000000
82 5 2 6 0.000019 0.000000
83 5 3 7 0.000036 0.000000
84 5 4 8 0.000020 0.000000
85 5 5 12 0.000022 0.000000
86 5 6 15 0.000068 0.000000
87 5 7 24 0.000069 0.000000
88 5 8 29 0.000020 0.000000
89 5 9 30 0.000036 0.000000
90 5 10 36 0.000068 0.000000
91 5 11 40 0.000068 0.000000
92 5 12 44 0.000028 0.000000
93 5 13 48 0.000020 0.000000
94 5 14 56 0.000019 0.000000
95 5 15 63 0.000037 0.000000
96 6 0 2 0.000068 0.000000
97 6 1 5 0.000068 0.000000
98 6 2 6 0.000019 0.000000
99 6 3 7 0.000036 0.000000
100 6 4 8 0.000019 0.000000
101 6 5 12 0.000023 0.000000
102 6 6 15 0.000069 0.000000
103 6 7 24 0.000069 0.000000
104 6 8 29 0.000020 0.000000
105 6 9 30 0.000036 0.000000
106 6 10 36 0.000068 0.000000
107 6 11 40 0.000070 0.000000
108 6 12 44 0.000027 0.000000
109 6 13 48 0.000019 0.000000
110 6 14 56 0.000019 0.000000
111 6 15 63 0.000037 0.000000
112 7 0 2 0.000068 0.000000
113 7 1 5 0.000069 0.000000
114 7 2 6 0.000019 0.000000
115 7 3 7 0.000036 0.000000
116 7 4 8 0.000019 0.000000
117 7 5 12 0.000023 0.000000
118 7 6 15 0.000069 0.000000
119 7 7 24 0.000070 0.000000
120 7 8 29 0.000020 0.000000
121 7 9 30 0.000036 0.000000
122 7 10 36 0.000069 0.000000
123 7 11 40 0.000068 0.000000
124 7 12 44 0.000028 0.000000
125 7 13 48 0.000020 0.000000
126 7 14 56 0.000020 0.000000
127 7 15 63 0.000036 0.000000

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p212.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (8,24,25,26,27,28,29,30, waited 655 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p212 at 2^24 (window control, by site)
[2026-10-08T15:19:15Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "212..212", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p212"]
[2026-10-08T15:19:15Z] census warps: programs [212] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:19:16Z] day 20730: cache filled in 1.00 s, fnv 448274a57f508cbc
[2026-10-08T15:19:26Z] table: 16777216 items derived with their 8 lines in 10.2 s; library comparison all (0 mismatches)
[2026-10-08T15:19:29Z] program p212-attack-f8: epoch seed 3ff9365abc8f8f22cf607c20f8f875bd0121f0c8f0409064ad5a81ea19c737ef era seed 5bbce245f76d64247ec5b4238b5e68fa00f064982902e13b15d77abab285e8d4 id 45c96a4e0afede43 attempt 0 class mx8-era061e9baf+sh256x27+state+rw op mix load=16 add=10 mad=9 rotl=8 xor=6 mul=5 rotr=3 sub=3 mulhi=2 shfl=2; era stride mul 0xd56f9b19 rot 8 interleave [5, 7, 13, 15] windows 3:1:1 5:2:0 8:0:0 14:2:2 18:0:0 20:0:0 24:1:1 29:0:0 30:2:0 34:1:1 38:0:0 47:0:0 49:2:0 50:0:0 56:0:0 59:2:0
[2026-10-08T15:19:29Z] load site instr 3: src r5 win 1 off 1; last base writer mad at 2; writers back to the last injecting one: mad@2; shadow writes of r5 per rep: add=1 mad=4 mul=3 rotl=6 rotr=4 shfl=1 sub=4 xor=5
[2026-10-08T15:19:29Z] load site instr 5: src r2 win 2 off 0; last base writer rotl at 1; writers back to the last injecting one: rotl@1 xor@53; shadow writes of r2 per rep: add=5 mad=3 mul=1 mulhi=2 rotl=2 rotr=6 sub=6 xor=5
[2026-10-08T15:19:29Z] load site instr 8: src r5 win 0 off 0; last base writer rotr at 7; writers back to the last injecting one: rotr@7 mad@2; shadow writes of r5 per rep: add=1 mad=4 mul=3 rotl=6 rotr=4 shfl=1 sub=4 xor=5
[2026-10-08T15:19:29Z] load site instr 14: src r6 win 2 off 2; last base writer rotl at 9; writers back to the last injecting one: rotl@9 add@58; shadow writes of r6 per rep: add=7 mad=3 rotl=6 rotr=9 shfl=5 sub=5 xor=5
[2026-10-08T15:19:29Z] load site instr 18: src r0 win 0 off 0; last base writer sub at 17; writers back to the last injecting one: sub@17; shadow writes of r0 per rep: add=5 mad=5 mul=1 mulhi=2 rotl=5 rotr=5 sub=4 xor=11
[2026-10-08T15:19:29Z] load site instr 20: src r4 win 0 off 0; last base writer load at 5; writers back to the last injecting one: load@5; shadow writes of r4 per rep: add=6 mad=5 rotl=6 rotr=4 sub=6 xor=4
[2026-10-08T15:19:29Z] load site instr 24: src r2 win 1 off 1; last base writer mad at 21; writers back to the last injecting one: mad@21; shadow writes of r2 per rep: add=5 mad=3 mul=1 mulhi=2 rotl=2 rotr=6 sub=6 xor=5
[2026-10-08T15:19:29Z] load site instr 29: src r2 win 0 off 0; last base writer xor at 27; writers back to the last injecting one: xor@27; shadow writes of r2 per rep: add=5 mad=3 mul=1 mulhi=2 rotl=2 rotr=6 sub=6 xor=5
[2026-10-08T15:19:29Z] load site instr 30: src r5 win 2 off 0; last base writer add at 28; writers back to the last injecting one: add@28; shadow writes of r5 per rep: add=1 mad=4 mul=3 rotl=6 rotr=4 shfl=1 sub=4 xor=5
[2026-10-08T15:19:29Z] load site instr 34: src r3 win 1 off 1; last base writer load at 24; writers back to the last injecting one: load@24; shadow writes of r3 per rep: add=6 mad=3 rotl=2 rotr=2 shfl=2 sub=8 xor=3
[2026-10-08T15:19:29Z] load site instr 38: src r3 win 0 off 0; last base writer rotl at 35; writers back to the last injecting one: rotl@35 load@24; shadow writes of r3 per rep: add=6 mad=3 rotl=2 rotr=2 shfl=2 sub=8 xor=3
[2026-10-08T15:19:29Z] load site instr 47: src r1 win 0 off 0; last base writer mad at 42; writers back to the last injecting one: mad@42; shadow writes of r1 per rep: add=1 mad=5 mul=2 mulhi=1 rotl=3 rotr=7 shfl=1 sub=2 xor=4
[2026-10-08T15:19:29Z] load site instr 49: src r0 win 2 off 0; last base writer mad at 46; writers back to the last injecting one: mad@46; shadow writes of r0 per rep: add=5 mad=5 mul=1 mulhi=2 rotl=5 rotr=5 sub=4 xor=11
[2026-10-08T15:19:29Z] load site instr 50: src r4 win 0 off 0; last base writer mad at 45; writers back to the last injecting one: mad@45; shadow writes of r4 per rep: add=6 mad=5 rotl=6 rotr=4 sub=6 xor=4
[2026-10-08T15:19:29Z] load site instr 56: src r5 win 0 off 0; last base writer load at 47; writers back to the last injecting one: load@47; shadow writes of r5 per rep: add=1 mad=4 mul=3 rotl=6 rotr=4 shfl=1 sub=4 xor=5
[2026-10-08T15:19:29Z] load site instr 59: src r3 win 2 off 0; last base writer add at 40; writers back to the last injecting one: add@40; shadow writes of r3 per rep: add=6 mad=3 rotl=2 rotr=2 shfl=2 sub=8 xor=3
[2026-10-08T15:19:29Z] window layer: site item windows (first, items) (0x800000,2^23) (0x0,2^22) (0x0,2^24) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^22) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^22); expected reads per item by quarter 192.000 64.000 144.000 112.000 (flat uniform 128.000)
[2026-10-08T15:21:51Z] 524288 warps (16777216 nonces) interpreted in 142.3 s (2.172 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:21:51Z] p212-attack-f8 distinct lines per hash: n 16777216 min 984 p1 1015 median 1024 p99 1024 max 1024 mean 1022.8721
[2026-10-08T15:21:51Z] p212-attack-f8 distinct items per hash: n 16777216 min 123 p1 127 median 128 p99 128 max 128 mean 127.8746
[2026-10-08T15:21:51Z] p212-attack-f8 distinct lines per warp: n 524288 min 32502 p1 32553 median 32605 p99 32647 max 32681 mean 32604.0968
[2026-10-08T15:21:51Z] p212-attack-f8 distinct items per warp: n 524288 min 4080 p1 4086 median 4092 p99 4095 max 4096 mean 4091.4352
[2026-10-08T15:21:51Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:21:51Z] positions: largest hi16-bucket (256 items) share 0.0073% at p76 (iteration 4, site 12, instr 49); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 3)
[2026-10-08T15:21:51Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p30 (site 14, instr 56); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:21:51Z] p212-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 316 (bin 1167415) z_max +16.62 min 25 z_min -9.10 chi2/dof 18.88257 chi2_z +51793.46 top0.1% 0.21052% top0.5% 1.00105% top1% 1.94764%
[2026-10-08T15:21:51Z] p212-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.71190 largest 316 (item 0x11d037) at +8.95 sigma smallest at -5.56; buckets64 chi2/dof 46.44376 largest 16383 (bucket 32768) at +36.94 sigma smallest at -12.98
[2026-10-08T15:22:01Z] p212-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 265 (bin 1231845) z_max +12.11 min 27 z_min -8.93 chi2/dof 18.00153 chi2_z +49241.69 top0.1% 0.18302% top0.5% 0.88403% top1% 1.73787%
[2026-10-08T15:22:01Z] p212-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99971 largest +5.50 sigma; buckets64 chi2/dof 0.99965 largest +4.64 smallest -4.17
[2026-10-08T15:22:01Z] 6-sigma p212-attack-f8 items buckets64 against the window density: largest bucket +36.94 sigma, smallest -12.98 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:22:01Z] hot-set p212-attack-f8 items (windowed control): f 0.1% S_f 0.21052% E_f(control) 0.18302% X_f +0.02750% X_f/f +0.2750 -> no hot set
[2026-10-08T15:22:01Z] hot-set p212-attack-f8 items (windowed control): f 0.5% S_f 1.00105% E_f(control) 0.88403% X_f +0.11702% X_f/f +0.2340 -> no hot set
[2026-10-08T15:22:01Z] hot-set p212-attack-f8 items (windowed control): f 1.0% S_f 1.94764% E_f(control) 1.73787% X_f +0.20977% X_f/f +0.2098 -> no hot set
[2026-10-08T15:22:01Z] hot-set p212-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:22:09Z] p212-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 194 (bin 15903188) z_max +5.83 min 73 z_min -4.86 chi2/dof 0.99977 chi2_z -0.68 top0.1% 0.13105% top0.5% 0.63246% top1% 1.24332%
[2026-10-08T15:22:09Z] hot-set p212-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.21052% E_f(control) 0.13105% X_f +0.07946% X_f/f +0.7946 -> no hot set
[2026-10-08T15:22:09Z] hot-set p212-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 1.00105% E_f(control) 0.63246% X_f +0.36860% X_f/f +0.7372 -> no hot set
[2026-10-08T15:22:09Z] hot-set p212-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.94764% E_f(control) 1.24332% X_f +0.70432% X_f/f +0.7043 -> no hot set
[2026-10-08T15:22:09Z] hot-set p212-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:22:09Z] ratio p212-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.1502x / 1.1324x / 1.1207x (gate 1.2x at 0.1%: within); over the FLAT control 1.6063x / 1.5828x / 1.5665x
[2026-10-08T15:22:10Z] attribution: hot threshold count >= 262 marks 18212 items (0.1086% of items) holding 4896809 reads (0.2280% of reads); top 8 items t=0x11d037:316 t=0x342803:313 t=0x019003:313 t=0x290014:310 t=0x05800a:310 t=0x256001:309 t=0x04400e:309 t=0x3f2002:306
[2026-10-08T15:23:15Z] attribution: reads on hot items per position (flat expectation 0.1086% of each position's 16777216 reads); the 24 largest: p108(it6 s12)=0.927% p124(it7 s12)=0.927% p12(it0 s12)=0.926% p28(it1 s12)=0.925% p92(it5 s12)=0.925% p44(it2 s12)=0.925% p76(it4 s12)=0.924% p60(it3 s12)=0.922% p56(it3 s8)=0.492% p120(it7 s8)=0.491% p17(it1 s1)=0.491% p65(it4 s1)=0.491% p81(it5 s1)=0.490% p72(it4 s8)=0.490% p95(it5 s15)=0.490% p49(it3 s1)=0.490% p31(it1 s15)=0.489% p1(it0 s1)=0.489% p104(it6 s8)=0.488% p127(it7 s15)=0.488% p40(it2 s8)=0.488% p24(it1 s8)=0.488% p79(it4 s15)=0.488% p97(it6 s1)=0.488%
[2026-10-08T15:23:15Z] attribution: hot reads by iteration it0=0.228% it1=0.228% it2=0.227% it3=0.228% it4=0.228% it5=0.228% it6=0.228% it7=0.228% ; by site s0=0.000% s1=0.489% s2=0.123% s3=0.000% s4=0.122% s5=0.123% s6=0.000% s7=0.122% s8=0.489% s9=0.000% s10=0.122% s11=0.123% s12=0.925% s13=0.400% s14=0.123% s15=0.488%
[2026-10-08T15:23:15Z] site 0 (instr 3, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1086%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:23:15Z] site 1 (instr 5, src r2, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4892% (flat expectation 0.1086%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0063% (window expectation 0.0061%)
[2026-10-08T15:23:15Z] site 2 (instr 8, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1226% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 3 (instr 14, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1086%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:23:15Z] site 4 (instr 18, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1224% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 5 (instr 20, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1228% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 6 (instr 24, src r2, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1086%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:23:15Z] site 7 (instr 29, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1219% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 8 (instr 30, src r5, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4892% (flat expectation 0.1086%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:23:15Z] site 9 (instr 34, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1086%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:23:15Z] site 10 (instr 38, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1220% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 11 (instr 47, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1228% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 12 (instr 49, src r0, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.9248% (flat expectation 0.1086%); index entropy 13.999 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0068% (window expectation 0.0061%)
[2026-10-08T15:23:15Z] site 13 (instr 50, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.4001% (flat expectation 0.1086%); index entropy 15.981 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0028% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 14 (instr 56, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1227% (flat expectation 0.1086%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:23:15Z] site 15 (instr 59, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4881% (flat expectation 0.1086%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:23:15Z] attribution: item 0x11d037 (316 reads) read from positions p92(it5 s12)x13 p76(it4 s12)x12 p108(it6 s12)x12 p12(it0 s12)x11 p60(it3 s12)x11 p28(it1 s12)x9 p44(it2 s12)x9 p88(it5 s8)x8 p120(it7 s8)x8 p127(it7 s15)x7 p8(it0 s8)x7 p13(it0 s13)x7
[2026-10-08T15:23:15Z] attribution: item 0x342803 (313 reads) read from positions p92(it5 s12)x17 p108(it6 s12)x14 p60(it3 s12)x13 p76(it4 s12)x11 p49(it3 s1)x9 p31(it1 s15)x8 p44(it2 s12)x8 p56(it3 s8)x8 p12(it0 s12)x7 p28(it1 s12)x7 p65(it4 s1)x7 p81(it5 s1)x7
[2026-10-08T15:23:15Z] attribution: item 0x019003 (313 reads) read from positions p108(it6 s12)x16 p28(it1 s12)x11 p44(it2 s12)x11 p76(it4 s12)x11 p88(it5 s8)x11 p92(it5 s12)x11 p29(it1 s13)x9 p1(it0 s1)x9 p8(it0 s8)x8 p33(it2 s1)x8 p60(it3 s12)x8 p56(it3 s8)x7
[2026-10-08T15:23:15Z] attribution: item 0x290014 (310 reads) read from positions p28(it1 s12)x13 p60(it3 s12)x12 p124(it7 s12)x11 p108(it6 s12)x11 p33(it2 s1)x10 p92(it5 s12)x10 p44(it2 s12)x9 p97(it6 s1)x9 p12(it0 s12)x8 p24(it1 s8)x8 p13(it0 s13)x7 p15(it0 s15)x7
[2026-10-08T15:23:15Z] attribution: item 0x05800a (310 reads) read from positions p12(it0 s12)x12 p44(it2 s12)x11 p28(it1 s12)x10 p31(it1 s15)x10 p108(it6 s12)x10 p13(it0 s13)x9 p76(it4 s12)x9 p88(it5 s8)x9 p127(it7 s15)x9 p124(it7 s12)x9 p92(it5 s12)x8 p17(it1 s1)x7
[2026-10-08T15:23:15Z] attribution: item 0x256001 (309 reads) read from positions p92(it5 s12)x14 p28(it1 s12)x12 p124(it7 s12)x11 p12(it0 s12)x10 p60(it3 s12)x10 p44(it2 s12)x9 p97(it6 s1)x9 p120(it7 s8)x9 p17(it1 s1)x8 p29(it1 s13)x8 p127(it7 s15)x7 p31(it1 s15)x7
[2026-10-08T15:23:15Z] attribution: item 0x04400e (309 reads) read from positions p28(it1 s12)x12 p24(it1 s8)x10 p127(it7 s15)x9 p12(it0 s12)x9 p13(it0 s13)x9 p60(it3 s12)x9 p1(it0 s1)x9 p92(it5 s12)x9 p109(it6 s13)x9 p72(it4 s8)x8 p76(it4 s12)x8 p40(it2 s8)x7
[2026-10-08T15:23:15Z] attribution: item 0x3f2002 (306 reads) read from positions p92(it5 s12)x15 p108(it6 s12)x12 p124(it7 s12)x11 p76(it4 s12)x10 p60(it3 s12)x9 p24(it1 s8)x8 p49(it3 s1)x8 p1(it0 s1)x8 p12(it0 s12)x7 p47(it2 s15)x7 p109(it6 s13)x7 p17(it1 s1)x6
[2026-10-08T15:23:15Z] hottest item p212-attack-f8 (0x11d037): 316 reads (0.0000% of all); predicted source none
[2026-10-08T15:23:15Z] program p212-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
Terminated
lease: released 8 pool cores after 895 s, exit 0

225
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p212/warps-p212-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,225 @@
# distinct lines_hash: value count
lines_hash 984 1
lines_hash 991 6
lines_hash 992 54
lines_hash 998 18
lines_hash 999 340
lines_hash 1000 2918
lines_hash 1004 2
lines_hash 1005 33
lines_hash 1006 632
lines_hash 1007 11290
lines_hash 1008 93527
lines_hash 1011 1
lines_hash 1012 18
lines_hash 1013 497
lines_hash 1014 12629
lines_hash 1015 204879
lines_hash 1016 1664771
lines_hash 1019 7
lines_hash 1020 132
lines_hash 1021 4203
lines_hash 1022 101653
lines_hash 1023 1630550
lines_hash 1024 13049055
# distinct items_hash: value count
items_hash 123 1
items_hash 124 60
items_hash 125 3276
items_hash 126 105484
items_hash 127 1882795
items_hash 128 14785600
# distinct lines_warp: value count
lines_warp 32502 2
lines_warp 32504 1
lines_warp 32506 1
lines_warp 32507 2
lines_warp 32508 1
lines_warp 32509 5
lines_warp 32510 2
lines_warp 32511 3
lines_warp 32512 3
lines_warp 32513 4
lines_warp 32514 7
lines_warp 32515 5
lines_warp 32516 10
lines_warp 32517 11
lines_warp 32518 8
lines_warp 32519 12
lines_warp 32520 10
lines_warp 32521 13
lines_warp 32522 13
lines_warp 32523 21
lines_warp 32524 13
lines_warp 32525 29
lines_warp 32526 21
lines_warp 32527 32
lines_warp 32528 30
lines_warp 32529 36
lines_warp 32530 47
lines_warp 32531 45
lines_warp 32532 56
lines_warp 32533 68
lines_warp 32534 66
lines_warp 32535 90
lines_warp 32536 88
lines_warp 32537 75
lines_warp 32538 100
lines_warp 32539 141
lines_warp 32540 170
lines_warp 32541 147
lines_warp 32542 168
lines_warp 32543 213
lines_warp 32544 211
lines_warp 32545 212
lines_warp 32546 248
lines_warp 32547 287
lines_warp 32548 346
lines_warp 32549 364
lines_warp 32550 458
lines_warp 32551 493
lines_warp 32552 527
lines_warp 32553 563
lines_warp 32554 613
lines_warp 32555 706
lines_warp 32556 703
lines_warp 32557 847
lines_warp 32558 880
lines_warp 32559 980
lines_warp 32560 1092
lines_warp 32561 1156
lines_warp 32562 1232
lines_warp 32563 1484
lines_warp 32564 1571
lines_warp 32565 1633
lines_warp 32566 1833
lines_warp 32567 1941
lines_warp 32568 2168
lines_warp 32569 2209
lines_warp 32570 2482
lines_warp 32571 2728
lines_warp 32572 2773
lines_warp 32573 3109
lines_warp 32574 3124
lines_warp 32575 3446
lines_warp 32576 3720
lines_warp 32577 4064
lines_warp 32578 4233
lines_warp 32579 4451
lines_warp 32580 4615
lines_warp 32581 5025
lines_warp 32582 5346
lines_warp 32583 5439
lines_warp 32584 5758
lines_warp 32585 6124
lines_warp 32586 6199
lines_warp 32587 6662
lines_warp 32588 6893
lines_warp 32589 7297
lines_warp 32590 7295
lines_warp 32591 7677
lines_warp 32592 8228
lines_warp 32593 8267
lines_warp 32594 8650
lines_warp 32595 8889
lines_warp 32596 8993
lines_warp 32597 9228
lines_warp 32598 9580
lines_warp 32599 9478
lines_warp 32600 9731
lines_warp 32601 10086
lines_warp 32602 10104
lines_warp 32603 10209
lines_warp 32604 10229
lines_warp 32605 10226
lines_warp 32606 10470
lines_warp 32607 10324
lines_warp 32608 10254
lines_warp 32609 10404
lines_warp 32610 10097
lines_warp 32611 10131
lines_warp 32612 10164
lines_warp 32613 9870
lines_warp 32614 9903
lines_warp 32615 9543
lines_warp 32616 9471
lines_warp 32617 9293
lines_warp 32618 8928
lines_warp 32619 8676
lines_warp 32620 8271
lines_warp 32621 7976
lines_warp 32622 7781
lines_warp 32623 7393
lines_warp 32624 6985
lines_warp 32625 6728
lines_warp 32626 6294
lines_warp 32627 5894
lines_warp 32628 5621
lines_warp 32629 5276
lines_warp 32630 5083
lines_warp 32631 4572
lines_warp 32632 4308
lines_warp 32633 3942
lines_warp 32634 3646
lines_warp 32635 3246
lines_warp 32636 3123
lines_warp 32637 2742
lines_warp 32638 2525
lines_warp 32639 2295
lines_warp 32640 2129
lines_warp 32641 1865
lines_warp 32642 1634
lines_warp 32643 1437
lines_warp 32644 1325
lines_warp 32645 1147
lines_warp 32646 1049
lines_warp 32647 850
lines_warp 32648 757
lines_warp 32649 638
lines_warp 32650 555
lines_warp 32651 490
lines_warp 32652 399
lines_warp 32653 368
lines_warp 32654 292
lines_warp 32655 240
lines_warp 32656 186
lines_warp 32657 152
lines_warp 32658 142
lines_warp 32659 107
lines_warp 32660 78
lines_warp 32661 66
lines_warp 32662 66
lines_warp 32663 44
lines_warp 32664 47
lines_warp 32665 24
lines_warp 32666 21
lines_warp 32667 19
lines_warp 32668 13
lines_warp 32669 7
lines_warp 32670 9
lines_warp 32671 3
lines_warp 32672 4
lines_warp 32673 6
lines_warp 32674 1
lines_warp 32676 2
lines_warp 32677 1
lines_warp 32678 1
lines_warp 32681 1
# distinct items_warp: value count
items_warp 4080 4
items_warp 4081 32
items_warp 4082 91
items_warp 4083 297
items_warp 4084 903
items_warp 4085 2361
items_warp 4086 5729
items_warp 4087 12601
items_warp 4088 25498
items_warp 4089 44388
items_warp 4090 69165
items_warp 4091 91146
items_warp 4092 99466
items_warp 4093 86491
items_warp 4094 56455
items_warp 4095 24419
items_warp 4096 5242

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p212/warps-p212-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 3 0.000036 0.000000
1 0 1 5 0.000069 0.000000
2 0 2 8 0.000020 0.000000
3 0 3 14 0.000070 0.000000
4 0 4 18 0.000019 0.000000
5 0 5 20 0.000020 0.000000
6 0 6 24 0.000036 0.000000
7 0 7 29 0.000019 0.000000
8 0 8 30 0.000069 0.000000
9 0 9 34 0.000036 0.000000
10 0 10 38 0.000019 0.000000
11 0 11 47 0.000019 0.000000
12 0 12 49 0.000072 0.000000
13 0 13 50 0.000027 0.000000
14 0 14 56 0.000020 0.000000
15 0 15 59 0.000069 0.000000
16 1 0 3 0.000037 0.000000
17 1 1 5 0.000071 0.000000
18 1 2 8 0.000019 0.000000
19 1 3 14 0.000068 0.000000
20 1 4 18 0.000019 0.000000
21 1 5 20 0.000020 0.000000
22 1 6 24 0.000036 0.000000
23 1 7 29 0.000020 0.000000
24 1 8 30 0.000068 0.000000
25 1 9 34 0.000036 0.000000
26 1 10 38 0.000019 0.000000
27 1 11 47 0.000020 0.000000
28 1 12 49 0.000073 0.000000
29 1 13 50 0.000029 0.000000
30 1 14 56 0.000019 0.000000
31 1 15 59 0.000068 0.000000
32 2 0 3 0.000036 0.000000
33 2 1 5 0.000068 0.000000
34 2 2 8 0.000019 0.000000
35 2 3 14 0.000068 0.000000
36 2 4 18 0.000019 0.000000
37 2 5 20 0.000019 0.000000
38 2 6 24 0.000036 0.000000
39 2 7 29 0.000019 0.000000
40 2 8 30 0.000068 0.000000
41 2 9 34 0.000036 0.000000
42 2 10 38 0.000020 0.000000
43 2 11 47 0.000019 0.000000
44 2 12 49 0.000072 0.000000
45 2 13 50 0.000028 0.000000
46 2 14 56 0.000019 0.000000
47 2 15 59 0.000068 0.000000
48 3 0 3 0.000036 0.000000
49 3 1 5 0.000068 0.000000
50 3 2 8 0.000019 0.000000
51 3 3 14 0.000068 0.000000
52 3 4 18 0.000019 0.000000
53 3 5 20 0.000020 0.000000
54 3 6 24 0.000036 0.000000
55 3 7 29 0.000019 0.000000
56 3 8 30 0.000069 0.000000
57 3 9 34 0.000036 0.000000
58 3 10 38 0.000020 0.000000
59 3 11 47 0.000019 0.000000
60 3 12 49 0.000072 0.000000
61 3 13 50 0.000029 0.000000
62 3 14 56 0.000019 0.000000
63 3 15 59 0.000069 0.000000
64 4 0 3 0.000036 0.000000
65 4 1 5 0.000069 0.000000
66 4 2 8 0.000019 0.000000
67 4 3 14 0.000070 0.000000
68 4 4 18 0.000020 0.000000
69 4 5 20 0.000019 0.000000
70 4 6 24 0.000036 0.000000
71 4 7 29 0.000019 0.000000
72 4 8 30 0.000068 0.000000
73 4 9 34 0.000037 0.000000
74 4 10 38 0.000019 0.000000
75 4 11 47 0.000021 0.000000
76 4 12 49 0.000073 0.000000
77 4 13 50 0.000029 0.000000
78 4 14 56 0.000019 0.000000
79 4 15 59 0.000069 0.000000
80 5 0 3 0.000036 0.000000
81 5 1 5 0.000069 0.000000
82 5 2 8 0.000019 0.000000
83 5 3 14 0.000069 0.000000
84 5 4 18 0.000019 0.000000
85 5 5 20 0.000019 0.000000
86 5 6 24 0.000036 0.000000
87 5 7 29 0.000019 0.000000
88 5 8 30 0.000068 0.000000
89 5 9 34 0.000037 0.000000
90 5 10 38 0.000020 0.000000
91 5 11 47 0.000019 0.000000
92 5 12 49 0.000073 0.000000
93 5 13 50 0.000027 0.000000
94 5 14 56 0.000020 0.000000
95 5 15 59 0.000069 0.000000
96 6 0 3 0.000036 0.000000
97 6 1 5 0.000068 0.000000
98 6 2 8 0.000019 0.000000
99 6 3 14 0.000068 0.000000
100 6 4 18 0.000019 0.000000
101 6 5 20 0.000019 0.000000
102 6 6 24 0.000036 0.000000
103 6 7 29 0.000020 0.000000
104 6 8 30 0.000068 0.000000
105 6 9 34 0.000036 0.000000
106 6 10 38 0.000020 0.000000
107 6 11 47 0.000020 0.000000
108 6 12 49 0.000072 0.000000
109 6 13 50 0.000028 0.000000
110 6 14 56 0.000019 0.000000
111 6 15 59 0.000069 0.000000
112 7 0 3 0.000036 0.000000
113 7 1 5 0.000069 0.000000
114 7 2 8 0.000020 0.000000
115 7 3 14 0.000069 0.000000
116 7 4 18 0.000020 0.000000
117 7 5 20 0.000019 0.000000
118 7 6 24 0.000037 0.000000
119 7 7 29 0.000020 0.000000
120 7 8 30 0.000068 0.000000
121 7 9 34 0.000036 0.000000
122 7 10 38 0.000019 0.000000
123 7 11 47 0.000020 0.000000
124 7 12 49 0.000073 0.000000
125 7 13 50 0.000029 0.000000
126 7 14 56 0.000019 0.000000
127 7 15 59 0.000069 0.000000

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p225.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (24,25,26,27,28,29,30,31, waited 100 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p225 at 2^24 (window control, by site)
[2026-10-08T15:10:01Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "225..225", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p225"]
[2026-10-08T15:10:01Z] census warps: programs [225] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:10:02Z] day 20730: cache filled in 0.45 s, fnv 448274a57f508cbc
[2026-10-08T15:10:11Z] table: 16777216 items derived with their 8 lines in 9.2 s; library comparison all (0 mismatches)
[2026-10-08T15:10:14Z] program p225-attack-f8: epoch seed fb1148e75003cb29c313b7e815e1786e78334aa7bfccd7bb47191af0f1c6f97e era seed f3a9e409ea5db0210c63a2f412442f560b7619672a5bb0941345cba4fa8d2c67 id 0cc72c2c3b0a516e attempt 0 class mx8-era3f991ed2+sh256x27+state+rw op mix load=16 add=15 rotr=9 mad=8 shfl=5 xor=5 rotl=3 sub=3; era stride mul 0x2ef225c7 rot 2 interleave [0, 12, 14, 15] windows 5:2:1 9:0:0 13:1:1 20:1:1 23:1:1 25:0:0 30:1:1 31:0:0 36:0:0 37:2:2 40:0:0 47:0:0 49:0:0 53:2:0 54:0:0 56:0:0
[2026-10-08T15:10:14Z] load site instr 5: src r6 win 2 off 1; last base writer shfl at 4; writers back to the last injecting one: shfl@4; shadow writes of r6 per rep: add=9 mad=5 mul=3 rotl=5 rotr=5 shfl=3 sub=4 xor=4
[2026-10-08T15:10:14Z] load site instr 9: src r7 win 0 off 0; last base writer xor at 1; writers back to the last injecting one: xor@1; shadow writes of r7 per rep: add=6 mad=7 mul=1 mulhi=1 rotl=5 rotr=3 shfl=1 sub=2 xor=8
[2026-10-08T15:10:14Z] load site instr 13: src r0 win 1 off 1; last base writer rotl at 2; writers back to the last injecting one: rotl@2 rotl@0 rotr@62 load@56; shadow writes of r0 per rep: add=9 mad=4 mulhi=3 rotr=3 sub=5 xor=6
[2026-10-08T15:10:14Z] load site instr 20: src r0 win 1 off 1; last base writer shfl at 19; writers back to the last injecting one: shfl@19; shadow writes of r0 per rep: add=9 mad=4 mulhi=3 rotr=3 sub=5 xor=6
[2026-10-08T15:10:14Z] load site instr 23: src r7 win 1 off 1; last base writer add at 21; writers back to the last injecting one: add@21; shadow writes of r7 per rep: add=6 mad=7 mul=1 mulhi=1 rotl=5 rotr=3 shfl=1 sub=2 xor=8
[2026-10-08T15:10:14Z] load site instr 25: src r3 win 0 off 0; last base writer add at 14; writers back to the last injecting one: add@14; shadow writes of r3 per rep: add=6 mad=7 mul=3 rotl=4 rotr=3 shfl=2 sub=3 xor=11
[2026-10-08T15:10:14Z] load site instr 30: src r1 win 1 off 1; last base writer load at 13; writers back to the last injecting one: load@13; shadow writes of r1 per rep: add=3 mad=7 mul=1 rotl=4 rotr=8 shfl=1 sub=1 xor=5
[2026-10-08T15:10:14Z] load site instr 31: src r2 win 0 off 0; last base writer load at 23; writers back to the last injecting one: load@23; shadow writes of r2 per rep: add=3 mad=2 mul=2 rotl=4 rotr=2 sub=2 xor=5
[2026-10-08T15:10:14Z] load site instr 36: src r4 win 0 off 0; last base writer rotr at 22; writers back to the last injecting one: rotr@22 load@20; shadow writes of r4 per rep: add=9 mad=6 mul=4 rotl=2 rotr=7 sub=2 xor=4
[2026-10-08T15:10:14Z] load site instr 37: src r6 win 2 off 2; last base writer load at 25; writers back to the last injecting one: load@25; shadow writes of r6 per rep: add=9 mad=5 mul=3 rotl=5 rotr=5 shfl=3 sub=4 xor=4
[2026-10-08T15:10:14Z] load site instr 40: src r0 win 0 off 0; last base writer shfl at 27; writers back to the last injecting one: shfl@27; shadow writes of r0 per rep: add=9 mad=4 mulhi=3 rotr=3 sub=5 xor=6
[2026-10-08T15:10:14Z] load site instr 47: src r2 win 0 off 0; last base writer load at 37; writers back to the last injecting one: load@37; shadow writes of r2 per rep: add=3 mad=2 mul=2 rotl=4 rotr=2 sub=2 xor=5
[2026-10-08T15:10:14Z] load site instr 49: src r3 win 0 off 0; last base writer add at 35; writers back to the last injecting one: add@35; shadow writes of r3 per rep: add=6 mad=7 mul=3 rotl=4 rotr=3 shfl=2 sub=3 xor=11
[2026-10-08T15:10:14Z] load site instr 53: src r0 win 2 off 0; last base writer rotr at 48; writers back to the last injecting one: rotr@48 shfl@27; shadow writes of r0 per rep: add=9 mad=4 mulhi=3 rotr=3 sub=5 xor=6
[2026-10-08T15:10:14Z] load site instr 54: src r5 win 0 off 0; last base writer add at 39; writers back to the last injecting one: add@39; shadow writes of r5 per rep: add=3 mad=3 mulhi=1 rotr=5 shfl=5 sub=7 xor=7
[2026-10-08T15:10:14Z] load site instr 56: src r2 win 0 off 0; last base writer load at 53; writers back to the last injecting one: load@53; shadow writes of r2 per rep: add=3 mad=2 mul=2 rotl=4 rotr=2 sub=2 xor=5
[2026-10-08T15:10:14Z] window layer: site item windows (first, items) (0x400000,2^22) (0x0,2^24) (0x800000,2^23) (0x800000,2^23) (0x800000,2^23) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x0,2^24); expected reads per item by quarter 104.000 104.000 168.000 136.000 (flat uniform 128.000)
[2026-10-08T15:13:09Z] 524288 warps (16777216 nonces) interpreted in 174.9 s (2.669 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:13:09Z] p225-attack-f8 distinct lines per hash: n 16777216 min 992 p1 1016 median 1024 p99 1024 max 1024 mean 1023.3705
[2026-10-08T15:13:09Z] p225-attack-f8 distinct items per hash: n 16777216 min 124 p1 127 median 128 p99 128 max 128 mean 127.9369
[2026-10-08T15:13:09Z] p225-attack-f8 distinct lines per warp: n 524288 min 32525 p1 32578 median 32621 p99 32656 max 32685 mean 32620.2786
[2026-10-08T15:13:09Z] p225-attack-f8 distinct items per warp: n 524288 min 4083 p1 4089 median 4094 p99 4096 max 4096 mean 4093.4769
[2026-10-08T15:13:09Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:13:09Z] positions: largest hi16-bucket (256 items) share 0.0070% at p0 (iteration 0, site 0, instr 5); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p18 (site 2, instr 13)
[2026-10-08T15:13:09Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p81 (site 1, instr 9); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:13:09Z] p225-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 243 (bin 9328126) z_max +10.16 min 53 z_min -6.63 chi2/dof 6.50035 chi2_z +15930.71 top0.1% 0.16220% top0.5% 0.78187% top1% 1.53524%
[2026-10-08T15:13:09Z] p225-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00059 largest 243 (item 0x8e55fe) at +5.79 sigma smallest at -5.00; buckets64 chi2/dof 0.99854 largest 11205 (bucket 137946) at +4.37 sigma smallest at -4.19
[2026-10-08T15:13:20Z] p225-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 243 (bin 8547428) z_max +10.16 min 56 z_min -6.36 chi2/dof 6.50051 chi2_z +15931.17 top0.1% 0.16224% top0.5% 0.78191% top1% 1.53535%
[2026-10-08T15:13:20Z] p225-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99989 largest +6.43 sigma; buckets64 chi2/dof 0.99991 largest +4.49 smallest -4.67
[2026-10-08T15:13:20Z] 6-sigma p225-attack-f8 items buckets64 against the window density: largest bucket +4.37 sigma, smallest -4.19 -> within 6 sigma
[2026-10-08T15:13:20Z] hot-set p225-attack-f8 items (windowed control): f 0.1% S_f 0.16220% E_f(control) 0.16224% X_f -0.00003% X_f/f -0.0003 -> no hot set
[2026-10-08T15:13:20Z] hot-set p225-attack-f8 items (windowed control): f 0.5% S_f 0.78187% E_f(control) 0.78191% X_f -0.00004% X_f/f -0.0001 -> no hot set
[2026-10-08T15:13:20Z] hot-set p225-attack-f8 items (windowed control): f 1.0% S_f 1.53524% E_f(control) 1.53535% X_f -0.00011% X_f/f -0.0001 -> no hot set
[2026-10-08T15:13:20Z] hot-set p225-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:13:30Z] p225-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 194 (bin 3106487) z_max +5.83 min 73 z_min -4.86 chi2/dof 1.00005 chi2_z +0.15 top0.1% 0.13112% top0.5% 0.63249% top1% 1.24335%
[2026-10-08T15:13:30Z] hot-set p225-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.16220% E_f(control) 0.13112% X_f +0.03108% X_f/f +0.3108 -> no hot set
[2026-10-08T15:13:30Z] hot-set p225-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.78187% E_f(control) 0.63249% X_f +0.14938% X_f/f +0.2988 -> no hot set
[2026-10-08T15:13:30Z] hot-set p225-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.53524% E_f(control) 1.24335% X_f +0.29189% X_f/f +0.2919 -> no hot set
[2026-10-08T15:13:30Z] hot-set p225-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:13:30Z] ratio p225-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9998x / 1.0000x / 0.9999x (gate 1.2x at 0.1%: within); over the FLAT control 1.2370x / 1.2362x / 1.2348x
[2026-10-08T15:13:31Z] attribution: hot threshold count >= 203 marks 19961 items (0.1190% of items) holding 4129656 reads (0.1923% of reads); top 8 items t=0x8e55fe:243 t=0xbc3cfe:236 t=0xb079ef:236 t=0x830b64:235 t=0xabf74c:234 t=0xa0eb4a:234 t=0xa849ca:233 t=0xa8e0ac:232
[2026-10-08T15:14:45Z] attribution: reads on hot items per position (flat expectation 0.1190% of each position's 16777216 reads); the 24 largest: p121(it7 s9)=0.588% p25(it1 s9)=0.588% p9(it0 s9)=0.588% p89(it5 s9)=0.586% p105(it6 s9)=0.585% p57(it3 s9)=0.585% p73(it4 s9)=0.585% p41(it2 s9)=0.585% p20(it1 s4)=0.296% p86(it5 s6)=0.296% p83(it5 s3)=0.295% p2(it0 s2)=0.295% p102(it6 s6)=0.294% p67(it4 s3)=0.294% p66(it4 s2)=0.294% p6(it0 s6)=0.294% p70(it4 s6)=0.294% p116(it7 s4)=0.293% p19(it1 s3)=0.293% p100(it6 s4)=0.293% p118(it7 s6)=0.293% p52(it3 s4)=0.293% p54(it3 s6)=0.293% p51(it3 s3)=0.293%
[2026-10-08T15:14:45Z] attribution: hot reads by iteration it0=0.192% it1=0.193% it2=0.192% it3=0.192% it4=0.192% it5=0.193% it6=0.192% it7=0.192% ; by site s0=0.000% s1=0.147% s2=0.293% s3=0.293% s4=0.293% s5=0.146% s6=0.293% s7=0.146% s8=0.146% s9=0.586% s10=0.147% s11=0.146% s12=0.147% s13=0.000% s14=0.147% s15=0.147%
[2026-10-08T15:14:45Z] site 0 (instr 5, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1190%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:45Z] site 1 (instr 9, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1465% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 2 (instr 13, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2931% (flat expectation 0.1190%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:45Z] site 3 (instr 20, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2928% (flat expectation 0.1190%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:45Z] site 4 (instr 23, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2927% (flat expectation 0.1190%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:45Z] site 5 (instr 25, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1464% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 6 (instr 30, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2935% (flat expectation 0.1190%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:14:45Z] site 7 (instr 31, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1462% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 8 (instr 36, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1465% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 9 (instr 37, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.5861% (flat expectation 0.1190%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0063% (window expectation 0.0061%)
[2026-10-08T15:14:45Z] site 10 (instr 40, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1468% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 11 (instr 47, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1459% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 12 (instr 49, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1468% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 13 (instr 53, src r0, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1190%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:14:45Z] site 14 (instr 54, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1466% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] site 15 (instr 56, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1469% (flat expectation 0.1190%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:14:45Z] attribution: item 0x8e55fe (243 reads) read from positions p121(it7 s9)x7 p25(it1 s9)x7 p105(it6 s9)x7 p38(it2 s6)x5 p57(it3 s9)x5 p67(it4 s3)x5 p100(it6 s4)x5 p115(it7 s3)x5 p118(it7 s6)x5 p6(it0 s6)x4 p17(it1 s1)x4 p22(it1 s6)x4
[2026-10-08T15:14:45Z] attribution: item 0xbc3cfe (236 reads) read from positions p25(it1 s9)x11 p89(it5 s9)x8 p73(it4 s9)x7 p115(it7 s3)x6 p52(it3 s4)x6 p9(it0 s9)x5 p35(it2 s3)x5 p3(it0 s3)x4 p33(it2 s1)x4 p41(it2 s9)x4 p43(it2 s11)x4 p18(it1 s2)x4
[2026-10-08T15:14:45Z] attribution: item 0xb079ef (236 reads) read from positions p25(it1 s9)x10 p9(it0 s9)x8 p4(it0 s4)x6 p57(it3 s9)x6 p105(it6 s9)x6 p121(it7 s9)x6 p34(it2 s2)x5 p41(it2 s9)x5 p51(it3 s3)x5 p68(it4 s4)x5 p83(it5 s3)x5 p126(it7 s14)x4
[2026-10-08T15:14:45Z] attribution: item 0x830b64 (235 reads) read from positions p18(it1 s2)x10 p73(it4 s9)x8 p121(it7 s9)x7 p105(it6 s9)x7 p25(it1 s9)x6 p41(it2 s9)x6 p57(it3 s9)x6 p3(it0 s3)x5 p4(it0 s4)x5 p6(it0 s6)x5 p9(it0 s9)x5 p31(it1 s15)x5
[2026-10-08T15:14:45Z] attribution: item 0xabf74c (234 reads) read from positions p73(it4 s9)x9 p9(it0 s9)x8 p105(it6 s9)x8 p20(it1 s4)x6 p41(it2 s9)x6 p54(it3 s6)x6 p3(it0 s3)x6 p99(it6 s3)x6 p126(it7 s14)x5 p25(it1 s9)x5 p26(it1 s10)x5 p70(it4 s6)x5
[2026-10-08T15:14:45Z] attribution: item 0xa0eb4a (234 reads) read from positions p73(it4 s9)x10 p57(it3 s9)x9 p121(it7 s9)x8 p89(it5 s9)x7 p4(it0 s4)x6 p66(it4 s2)x6 p103(it6 s7)x6 p9(it0 s9)x5 p28(it1 s12)x5 p41(it2 s9)x5 p59(it3 s11)x5 p106(it6 s10)x4
[2026-10-08T15:14:45Z] attribution: item 0xa849ca (233 reads) read from positions p73(it4 s9)x10 p121(it7 s9)x9 p41(it2 s9)x8 p54(it3 s6)x8 p9(it0 s9)x6 p25(it1 s9)x6 p57(it3 s9)x6 p87(it5 s7)x5 p89(it5 s9)x5 p91(it5 s11)x5 p114(it7 s2)x5 p123(it7 s11)x4
[2026-10-08T15:14:45Z] attribution: item 0xa8e0ac (232 reads) read from positions p102(it6 s6)x9 p89(it5 s9)x8 p22(it1 s6)x7 p105(it6 s9)x7 p121(it7 s9)x6 p35(it2 s3)x6 p41(it2 s9)x5 p57(it3 s9)x5 p84(it5 s4)x5 p123(it7 s11)x5 p103(it6 s7)x4 p6(it0 s6)x4
[2026-10-08T15:14:45Z] hottest item p225-attack-f8 (0x8e55fe): 243 reads (0.0000% of all); predicted source none
[2026-10-08T15:14:45Z] program p225-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
Terminated
lease: released 8 pool cores after 384 s, exit 0

195
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p225/warps-p225-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,195 @@
# distinct lines_hash: value count
lines_hash 992 9
lines_hash 998 3
lines_hash 999 46
lines_hash 1000 399
lines_hash 1005 4
lines_hash 1006 175
lines_hash 1007 2897
lines_hash 1008 24867
lines_hash 1012 12
lines_hash 1013 255
lines_hash 1014 6616
lines_hash 1015 108441
lines_hash 1016 885785
lines_hash 1019 3
lines_hash 1020 137
lines_hash 1021 4557
lines_hash 1022 108834
lines_hash 1023 1736512
lines_hash 1024 13897664
# distinct items_hash: value count
items_hash 124 9
items_hash 125 448
items_hash 126 27943
items_hash 127 1001109
items_hash 128 15747707
# distinct lines_warp: value count
lines_warp 32525 1
lines_warp 32528 1
lines_warp 32532 1
lines_warp 32534 1
lines_warp 32535 2
lines_warp 32536 2
lines_warp 32537 2
lines_warp 32538 3
lines_warp 32539 3
lines_warp 32540 2
lines_warp 32541 6
lines_warp 32542 2
lines_warp 32543 3
lines_warp 32544 8
lines_warp 32545 4
lines_warp 32546 5
lines_warp 32547 10
lines_warp 32548 8
lines_warp 32549 16
lines_warp 32550 12
lines_warp 32551 22
lines_warp 32552 12
lines_warp 32553 13
lines_warp 32554 26
lines_warp 32555 38
lines_warp 32556 45
lines_warp 32557 32
lines_warp 32558 43
lines_warp 32559 56
lines_warp 32560 69
lines_warp 32561 78
lines_warp 32562 80
lines_warp 32563 110
lines_warp 32564 83
lines_warp 32565 144
lines_warp 32566 142
lines_warp 32567 167
lines_warp 32568 194
lines_warp 32569 227
lines_warp 32570 290
lines_warp 32571 279
lines_warp 32572 347
lines_warp 32573 417
lines_warp 32574 451
lines_warp 32575 510
lines_warp 32576 554
lines_warp 32577 605
lines_warp 32578 694
lines_warp 32579 782
lines_warp 32580 891
lines_warp 32581 1022
lines_warp 32582 1075
lines_warp 32583 1196
lines_warp 32584 1409
lines_warp 32585 1575
lines_warp 32586 1804
lines_warp 32587 1887
lines_warp 32588 2041
lines_warp 32589 2266
lines_warp 32590 2514
lines_warp 32591 2728
lines_warp 32592 2812
lines_warp 32593 3306
lines_warp 32594 3487
lines_warp 32595 3823
lines_warp 32596 4060
lines_warp 32597 4363
lines_warp 32598 4904
lines_warp 32599 5335
lines_warp 32600 5458
lines_warp 32601 5886
lines_warp 32602 6357
lines_warp 32603 6627
lines_warp 32604 7188
lines_warp 32605 7596
lines_warp 32606 8112
lines_warp 32607 8259
lines_warp 32608 8748
lines_warp 32609 9192
lines_warp 32610 9476
lines_warp 32611 9973
lines_warp 32612 10361
lines_warp 32613 10467
lines_warp 32614 10866
lines_warp 32615 11241
lines_warp 32616 11711
lines_warp 32617 11901
lines_warp 32618 11960
lines_warp 32619 12066
lines_warp 32620 12176
lines_warp 32621 12407
lines_warp 32622 12324
lines_warp 32623 12613
lines_warp 32624 12560
lines_warp 32625 12527
lines_warp 32626 12268
lines_warp 32627 12127
lines_warp 32628 11876
lines_warp 32629 11545
lines_warp 32630 11369
lines_warp 32631 10959
lines_warp 32632 10500
lines_warp 32633 10289
lines_warp 32634 9813
lines_warp 32635 9206
lines_warp 32636 8974
lines_warp 32637 8483
lines_warp 32638 8074
lines_warp 32639 7260
lines_warp 32640 6830
lines_warp 32641 6394
lines_warp 32642 5964
lines_warp 32643 5359
lines_warp 32644 5016
lines_warp 32645 4478
lines_warp 32646 3994
lines_warp 32647 3637
lines_warp 32648 3195
lines_warp 32649 2814
lines_warp 32650 2531
lines_warp 32651 2198
lines_warp 32652 1921
lines_warp 32653 1689
lines_warp 32654 1398
lines_warp 32655 1238
lines_warp 32656 997
lines_warp 32657 844
lines_warp 32658 750
lines_warp 32659 649
lines_warp 32660 497
lines_warp 32661 441
lines_warp 32662 334
lines_warp 32663 277
lines_warp 32664 192
lines_warp 32665 185
lines_warp 32666 137
lines_warp 32667 120
lines_warp 32668 60
lines_warp 32669 61
lines_warp 32670 40
lines_warp 32671 38
lines_warp 32672 33
lines_warp 32673 22
lines_warp 32674 15
lines_warp 32675 13
lines_warp 32676 15
lines_warp 32677 5
lines_warp 32678 2
lines_warp 32679 4
lines_warp 32680 1
lines_warp 32681 3
lines_warp 32683 1
lines_warp 32685 1
# distinct items_warp: value count
items_warp 4083 1
items_warp 4084 2
items_warp 4085 29
items_warp 4086 101
items_warp 4087 471
items_warp 4088 1711
items_warp 4089 5341
items_warp 4090 14688
items_warp 4091 36018
items_warp 4092 71262
items_warp 4093 112856
items_warp 4094 134368
items_warp 4095 105602
items_warp 4096 41838

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p225/warps-p225-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 5 0.000070 0.000000
1 0 1 9 0.000020 0.000000
2 0 2 13 0.000037 0.000000
3 0 3 20 0.000037 0.000000
4 0 4 23 0.000036 0.000000
5 0 5 25 0.000019 0.000000
6 0 6 30 0.000036 0.000000
7 0 7 31 0.000020 0.000000
8 0 8 36 0.000019 0.000000
9 0 9 37 0.000069 0.000000
10 0 10 40 0.000020 0.000000
11 0 11 47 0.000019 0.000000
12 0 12 49 0.000020 0.000000
13 0 13 53 0.000069 0.000000
14 0 14 54 0.000019 0.000000
15 0 15 56 0.000020 0.000000
16 1 0 5 0.000069 0.000000
17 1 1 9 0.000020 0.000000
18 1 2 13 0.000036 0.000000
19 1 3 20 0.000036 0.000000
20 1 4 23 0.000036 0.000000
21 1 5 25 0.000020 0.000000
22 1 6 30 0.000036 0.000000
23 1 7 31 0.000020 0.000000
24 1 8 36 0.000019 0.000000
25 1 9 37 0.000068 0.000000
26 1 10 40 0.000020 0.000000
27 1 11 47 0.000020 0.000000
28 1 12 49 0.000019 0.000000
29 1 13 53 0.000069 0.000000
30 1 14 54 0.000019 0.000000
31 1 15 56 0.000020 0.000000
32 2 0 5 0.000069 0.000000
33 2 1 9 0.000019 0.000000
34 2 2 13 0.000036 0.000000
35 2 3 20 0.000036 0.000000
36 2 4 23 0.000036 0.000000
37 2 5 25 0.000019 0.000000
38 2 6 30 0.000036 0.000000
39 2 7 31 0.000019 0.000000
40 2 8 36 0.000020 0.000000
41 2 9 37 0.000069 0.000000
42 2 10 40 0.000019 0.000000
43 2 11 47 0.000019 0.000000
44 2 12 49 0.000019 0.000000
45 2 13 53 0.000069 0.000000
46 2 14 54 0.000020 0.000000
47 2 15 56 0.000019 0.000000
48 3 0 5 0.000069 0.000000
49 3 1 9 0.000019 0.000000
50 3 2 13 0.000036 0.000000
51 3 3 20 0.000036 0.000000
52 3 4 23 0.000036 0.000000
53 3 5 25 0.000020 0.000000
54 3 6 30 0.000036 0.000000
55 3 7 31 0.000019 0.000000
56 3 8 36 0.000019 0.000000
57 3 9 37 0.000068 0.000000
58 3 10 40 0.000020 0.000000
59 3 11 47 0.000019 0.000000
60 3 12 49 0.000019 0.000000
61 3 13 53 0.000069 0.000000
62 3 14 54 0.000020 0.000000
63 3 15 56 0.000019 0.000000
64 4 0 5 0.000068 0.000000
65 4 1 9 0.000020 0.000000
66 4 2 13 0.000036 0.000000
67 4 3 20 0.000036 0.000000
68 4 4 23 0.000036 0.000000
69 4 5 25 0.000020 0.000000
70 4 6 30 0.000036 0.000000
71 4 7 31 0.000020 0.000000
72 4 8 36 0.000020 0.000000
73 4 9 37 0.000068 0.000000
74 4 10 40 0.000019 0.000000
75 4 11 47 0.000019 0.000000
76 4 12 49 0.000019 0.000000
77 4 13 53 0.000069 0.000000
78 4 14 54 0.000020 0.000000
79 4 15 56 0.000019 0.000000
80 5 0 5 0.000069 0.000000
81 5 1 9 0.000020 0.000000
82 5 2 13 0.000037 0.000000
83 5 3 20 0.000036 0.000000
84 5 4 23 0.000036 0.000000
85 5 5 25 0.000020 0.000000
86 5 6 30 0.000036 0.000000
87 5 7 31 0.000019 0.000000
88 5 8 36 0.000019 0.000000
89 5 9 37 0.000069 0.000000
90 5 10 40 0.000019 0.000000
91 5 11 47 0.000019 0.000000
92 5 12 49 0.000020 0.000000
93 5 13 53 0.000068 0.000000
94 5 14 54 0.000019 0.000000
95 5 15 56 0.000019 0.000000
96 6 0 5 0.000068 0.000000
97 6 1 9 0.000019 0.000000
98 6 2 13 0.000036 0.000000
99 6 3 20 0.000036 0.000000
100 6 4 23 0.000036 0.000000
101 6 5 25 0.000020 0.000000
102 6 6 30 0.000037 0.000000
103 6 7 31 0.000020 0.000000
104 6 8 36 0.000020 0.000000
105 6 9 37 0.000069 0.000000
106 6 10 40 0.000019 0.000000
107 6 11 47 0.000019 0.000000
108 6 12 49 0.000019 0.000000
109 6 13 53 0.000069 0.000000
110 6 14 54 0.000020 0.000000
111 6 15 56 0.000019 0.000000
112 7 0 5 0.000069 0.000000
113 7 1 9 0.000019 0.000000
114 7 2 13 0.000036 0.000000
115 7 3 20 0.000037 0.000000
116 7 4 23 0.000036 0.000000
117 7 5 25 0.000019 0.000000
118 7 6 30 0.000037 0.000000
119 7 7 31 0.000019 0.000000
120 7 8 36 0.000019 0.000000
121 7 9 37 0.000068 0.000000
122 7 10 40 0.000019 0.000000
123 7 11 47 0.000019 0.000000
124 7 12 49 0.000020 0.000000
125 7 13 53 0.000069 0.000000
126 7 14 54 0.000019 0.000000
127 7 15 56 0.000019 0.000000

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p34.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (8,9,10,11,16,18,19,20, waited 342 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p34 at 2^24 (window control, by site)
[2026-10-08T15:14:01Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "34..34", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p34"]
[2026-10-08T15:14:01Z] census warps: programs [34] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:14:02Z] day 20730: cache filled in 0.50 s, fnv 448274a57f508cbc
[2026-10-08T15:14:12Z] table: 16777216 items derived with their 8 lines in 10.4 s; library comparison all (0 mismatches)
[2026-10-08T15:14:15Z] program p34-attack-f8: epoch seed 5ec98735ab1ac5c148faec5c06057e78ef0ecf0c71385f07e861d37f3d8c4ff1 era seed e0aa83ed164a680fc6d7ceeb411d23e845a004df139688888299bcc38ea7df43 id e8ffbc5c34dbc5f2 attempt 0 class mx8-erace76cc36+sh256x27+state+rw op mix load=16 rotl=10 add=7 rotr=7 sub=7 mad=6 xor=6 mul=2 shfl=2 mulhi=1; era stride mul 0x2e362f07 rot 7 interleave [3, 6, 11, 14] windows 4:0:0 13:1:1 16:2:3 17:0:0 23:2:1 25:0:0 31:0:0 33:2:3 36:1:1 45:1:0 49:0:0 52:2:0 53:0:0 54:0:0 57:0:0 60:0:0
[2026-10-08T15:14:15Z] load site instr 4: src r0 win 0 off 0; last base writer add at 1; writers back to the last injecting one: add@1; shadow writes of r0 per rep: add=5 mad=2 mul=1 mulhi=2 rotl=3 rotr=4 shfl=1 sub=4 xor=6
[2026-10-08T15:14:15Z] load site instr 13: src r7 win 1 off 1; last base writer mad at 10; writers back to the last injecting one: mad@10; shadow writes of r7 per rep: add=5 mad=6 mul=2 rotl=3 rotr=3 shfl=3 sub=2 xor=2
[2026-10-08T15:14:15Z] load site instr 16: src r6 win 2 off 3; last base writer rotl at 12; writers back to the last injecting one: rotl@12 rotr@8 load@4; shadow writes of r6 per rep: add=1 mad=2 mul=2 mulhi=1 rotl=4 rotr=4 shfl=2 sub=5 xor=3
[2026-10-08T15:14:15Z] load site instr 17: src r1 win 0 off 0; last base writer add at 2; writers back to the last injecting one: add@2; shadow writes of r1 per rep: add=13 mad=5 mul=2 rotl=4 rotr=6 sub=3 xor=9
[2026-10-08T15:14:15Z] load site instr 23: src r1 win 2 off 1; last base writer xor at 19; writers back to the last injecting one: xor@19; shadow writes of r1 per rep: add=13 mad=5 mul=2 rotl=4 rotr=6 sub=3 xor=9
[2026-10-08T15:14:15Z] load site instr 25: src r0 win 0 off 0; last base writer load at 23; writers back to the last injecting one: load@23; shadow writes of r0 per rep: add=5 mad=2 mul=1 mulhi=2 rotl=3 rotr=4 shfl=1 sub=4 xor=6
[2026-10-08T15:14:15Z] load site instr 31: src r6 win 0 off 0; last base writer mad at 29; writers back to the last injecting one: mad@29; shadow writes of r6 per rep: add=1 mad=2 mul=2 mulhi=1 rotl=4 rotr=4 shfl=2 sub=5 xor=3
[2026-10-08T15:14:15Z] load site instr 33: src r3 win 2 off 3; last base writer add at 27; writers back to the last injecting one: add@27; shadow writes of r3 per rep: add=11 mad=6 mul=1 rotl=2 rotr=1 shfl=2 sub=7 xor=3
[2026-10-08T15:14:15Z] load site instr 36: src r7 win 1 off 1; last base writer load at 33; writers back to the last injecting one: load@33; shadow writes of r7 per rep: add=5 mad=6 mul=2 rotl=3 rotr=3 shfl=3 sub=2 xor=2
[2026-10-08T15:14:15Z] load site instr 45: src r4 win 1 off 0; last base writer mad at 43; writers back to the last injecting one: mad@43; shadow writes of r4 per rep: add=5 mad=8 mul=2 mulhi=1 rotl=7 rotr=4 shfl=3 sub=3 xor=3
[2026-10-08T15:14:15Z] load site instr 49: src r6 win 0 off 0; last base writer rotl at 42; writers back to the last injecting one: rotl@42 mad@29; shadow writes of r6 per rep: add=1 mad=2 mul=2 mulhi=1 rotl=4 rotr=4 shfl=2 sub=5 xor=3
[2026-10-08T15:14:15Z] load site instr 52: src r3 win 2 off 0; last base writer mad at 48; writers back to the last injecting one: mad@48; shadow writes of r3 per rep: add=11 mad=6 mul=1 rotl=2 rotr=1 shfl=2 sub=7 xor=3
[2026-10-08T15:14:15Z] load site instr 53: src r0 win 0 off 0; last base writer add at 51; writers back to the last injecting one: add@51; shadow writes of r0 per rep: add=5 mad=2 mul=1 mulhi=2 rotl=3 rotr=4 shfl=1 sub=4 xor=6
[2026-10-08T15:14:15Z] load site instr 54: src r1 win 0 off 0; last base writer xor at 39; writers back to the last injecting one: xor@39; shadow writes of r1 per rep: add=13 mad=5 mul=2 rotl=4 rotr=6 sub=3 xor=9
[2026-10-08T15:14:15Z] load site instr 57: src r5 win 0 off 0; last base writer load at 52; writers back to the last injecting one: load@52; shadow writes of r5 per rep: add=3 mad=3 mul=2 rotl=1 rotr=9 shfl=2 sub=7 xor=9
[2026-10-08T15:14:15Z] load site instr 60: src r0 win 0 off 0; last base writer rotl at 58; writers back to the last injecting one: rotl@58 add@51; shadow writes of r0 per rep: add=5 mad=2 mul=1 mulhi=2 rotl=3 rotr=4 shfl=1 sub=4 xor=6
[2026-10-08T15:14:15Z] window layer: site item windows (first, items) (0x0,2^24) (0x800000,2^23) (0xc00000,2^22) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x800000,2^23) (0x0,2^23) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^24); expected reads per item by quarter 120.000 120.000 104.000 168.000 (flat uniform 128.000)
[2026-10-08T15:16:57Z] 524288 warps (16777216 nonces) interpreted in 161.7 s (2.468 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:16:57Z] p34-attack-f8 distinct lines per hash: n 16777216 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8711
[2026-10-08T15:16:57Z] p34-attack-f8 distinct items per hash: n 16777216 min 126 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:16:57Z] p34-attack-f8 distinct lines per warp: n 524288 min 32567 p1 32605 median 32637 p99 32664 max 32688 mean 32636.2143
[2026-10-08T15:16:57Z] p34-attack-f8 distinct items per warp: n 524288 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4811
[2026-10-08T15:16:57Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:16:57Z] positions: largest hi16-bucket (256 items) share 0.0070% at p66 (iteration 4, site 2, instr 16); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 4)
[2026-10-08T15:16:57Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p30 (site 14, instr 57); saturated sources at one position 0 of 2048 at p0 (site 0, instr 4); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:16:57Z] p34-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 256 (bin 14646730) z_max +11.31 min 51 z_min -6.81 chi2/dof 5.64798 chi2_z +13461.98 top0.1% 0.16418% top0.5% 0.78976% top1% 1.54922%
[2026-10-08T15:16:57Z] p34-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.15168 largest 256 (item 0xdf7dca) at +6.79 sigma smallest at -5.20; buckets64 chi2/dof 1.62408 largest 8123 (bucket 14888) at +5.06 sigma smallest at -5.18
[2026-10-08T15:17:06Z] p34-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 234 (bin 14399781) z_max +9.37 min 55 z_min -6.45 chi2/dof 5.49923 chi2_z +13031.17 top0.1% 0.16230% top0.5% 0.78206% top1% 1.53546%
[2026-10-08T15:17:06Z] p34-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00023 largest +5.98 sigma; buckets64 chi2/dof 0.99853 largest +4.82 smallest -4.70
[2026-10-08T15:17:06Z] 6-sigma p34-attack-f8 items buckets64 against the window density: largest bucket +5.06 sigma, smallest -5.18 -> within 6 sigma
[2026-10-08T15:17:06Z] hot-set p34-attack-f8 items (windowed control): f 0.1% S_f 0.16418% E_f(control) 0.16230% X_f +0.00188% X_f/f +0.0188 -> no hot set
[2026-10-08T15:17:06Z] hot-set p34-attack-f8 items (windowed control): f 0.5% S_f 0.78976% E_f(control) 0.78206% X_f +0.00770% X_f/f +0.0154 -> no hot set
[2026-10-08T15:17:06Z] hot-set p34-attack-f8 items (windowed control): f 1.0% S_f 1.54922% E_f(control) 1.53546% X_f +0.01376% X_f/f +0.0138 -> no hot set
[2026-10-08T15:17:06Z] hot-set p34-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:17:13Z] p34-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 191 (bin 6265201) z_max +5.57 min 71 z_min -5.04 chi2/dof 0.99956 chi2_z -1.28 top0.1% 0.13106% top0.5% 0.63237% top1% 1.24316%
[2026-10-08T15:17:13Z] hot-set p34-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.16418% E_f(control) 0.13106% X_f +0.03312% X_f/f +0.3312 -> no hot set
[2026-10-08T15:17:13Z] hot-set p34-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.78976% E_f(control) 0.63237% X_f +0.15739% X_f/f +0.3148 -> no hot set
[2026-10-08T15:17:13Z] hot-set p34-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.54922% E_f(control) 1.24316% X_f +0.30606% X_f/f +0.3061 -> no hot set
[2026-10-08T15:17:13Z] hot-set p34-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:17:13Z] ratio p34-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0116x / 1.0098x / 1.0090x (gate 1.2x at 0.1%: within); over the FLAT control 1.2527x / 1.2489x / 1.2462x
[2026-10-08T15:17:14Z] attribution: hot threshold count >= 206 marks 16938 items (0.1010% of items) holding 3558923 reads (0.1657% of reads); top 8 items t=0xdf7dca:256 t=0xe398c8:242 t=0xc25788:239 t=0xd4c29d:237 t=0xe81613:236 t=0xd10b9c:236 t=0xe9ac00:235 t=0xcbdb03:235
[2026-10-08T15:18:22Z] attribution: reads on hot items per position (flat expectation 0.1010% of each position's 16777216 reads); the 24 largest: p98(it6 s2)=0.495% p82(it5 s2)=0.495% p87(it5 s7)=0.494% p103(it6 s7)=0.494% p71(it4 s7)=0.494% p114(it7 s2)=0.493% p34(it2 s2)=0.493% p55(it3 s7)=0.493% p66(it4 s2)=0.493% p50(it3 s2)=0.493% p23(it1 s7)=0.493% p119(it7 s7)=0.492% p7(it0 s7)=0.492% p2(it0 s2)=0.491% p39(it2 s7)=0.491% p18(it1 s2)=0.489% p97(it6 s1)=0.313% p17(it1 s1)=0.311% p113(it7 s1)=0.311% p81(it5 s1)=0.310% p65(it4 s1)=0.310% p49(it3 s1)=0.310% p33(it2 s1)=0.310% p1(it0 s1)=0.308%
[2026-10-08T15:18:22Z] attribution: hot reads by iteration it0=0.165% it1=0.165% it2=0.165% it3=0.166% it4=0.166% it5=0.166% it6=0.166% it7=0.166% ; by site s0=0.124% s1=0.310% s2=0.493% s3=0.123% s4=0.000% s5=0.123% s6=0.123% s7=0.493% s8=0.246% s9=0.000% s10=0.123% s11=0.000% s12=0.123% s13=0.123% s14=0.123% s15=0.123%
[2026-10-08T15:18:22Z] site 0 (instr 4, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1242% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 1 (instr 13, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3105% (flat expectation 0.1010%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:18:22Z] site 2 (instr 16, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.4928% (flat expectation 0.1010%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:18:22Z] site 3 (instr 17, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1234% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 4 (instr 23, src r1, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1010%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:18:22Z] site 5 (instr 25, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1227% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 6 (instr 31, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1231% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 7 (instr 33, src r3, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.4928% (flat expectation 0.1010%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:18:22Z] site 8 (instr 36, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2463% (flat expectation 0.1010%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:18:22Z] site 9 (instr 45, src r4, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1010%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:18:22Z] site 10 (instr 49, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1230% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 11 (instr 52, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1010%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:18:22Z] site 12 (instr 53, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1232% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 13 (instr 54, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1231% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 14 (instr 57, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1235% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] site 15 (instr 60, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1230% (flat expectation 0.1010%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:18:22Z] attribution: item 0xdf7dca (256 reads) read from positions p66(it4 s2)x13 p50(it3 s2)x11 p55(it3 s7)x9 p82(it5 s2)x8 p98(it6 s2)x8 p34(it2 s2)x7 p71(it4 s7)x7 p7(it0 s7)x6 p18(it1 s2)x6 p53(it3 s5)x6 p23(it1 s7)x5 p39(it2 s7)x5
[2026-10-08T15:18:22Z] attribution: item 0xe398c8 (242 reads) read from positions p7(it0 s7)x11 p55(it3 s7)x10 p2(it0 s2)x9 p71(it4 s7)x9 p114(it7 s2)x8 p119(it7 s7)x8 p39(it2 s7)x7 p18(it1 s2)x6 p50(it3 s2)x6 p66(it4 s2)x6 p72(it4 s8)x6 p103(it6 s7)x6
[2026-10-08T15:18:22Z] attribution: item 0xc25788 (239 reads) read from positions p34(it2 s2)x8 p55(it3 s7)x7 p66(it4 s2)x7 p119(it7 s7)x7 p7(it0 s7)x6 p23(it1 s7)x6 p24(it1 s8)x6 p72(it4 s8)x6 p87(it5 s7)x6 p98(it6 s2)x6 p49(it3 s1)x5 p50(it3 s2)x5
[2026-10-08T15:18:22Z] attribution: item 0xd4c29d (237 reads) read from positions p82(it5 s2)x12 p71(it4 s7)x10 p50(it3 s2)x9 p2(it0 s2)x7 p23(it1 s7)x7 p34(it2 s2)x7 p65(it4 s1)x7 p119(it7 s7)x6 p66(it4 s2)x6 p87(it5 s7)x6 p88(it5 s8)x6 p114(it7 s2)x6
[2026-10-08T15:18:22Z] attribution: item 0xe81613 (236 reads) read from positions p1(it0 s1)x9 p71(it4 s7)x9 p39(it2 s7)x7 p50(it3 s2)x7 p88(it5 s8)x7 p98(it6 s2)x7 p114(it7 s2)x7 p119(it7 s7)x6 p34(it2 s2)x6 p49(it3 s1)x6 p66(it4 s2)x6 p113(it7 s1)x6
[2026-10-08T15:18:22Z] attribution: item 0xd10b9c (236 reads) read from positions p18(it1 s2)x11 p2(it0 s2)x9 p17(it1 s1)x8 p119(it7 s7)x8 p49(it3 s1)x7 p7(it0 s7)x6 p40(it2 s8)x6 p55(it3 s7)x5 p66(it4 s2)x5 p71(it4 s7)x5 p82(it5 s2)x5 p114(it7 s2)x5
[2026-10-08T15:18:22Z] attribution: item 0xe9ac00 (235 reads) read from positions p2(it0 s2)x8 p34(it2 s2)x8 p50(it3 s2)x8 p98(it6 s2)x8 p39(it2 s7)x7 p82(it5 s2)x7 p1(it0 s1)x6 p24(it1 s8)x6 p55(it3 s7)x6 p66(it4 s2)x6 p87(it5 s7)x6 p103(it6 s7)x6
[2026-10-08T15:18:22Z] attribution: item 0xcbdb03 (235 reads) read from positions p39(it2 s7)x9 p2(it0 s2)x8 p7(it0 s7)x8 p18(it1 s2)x8 p66(it4 s2)x8 p87(it5 s7)x7 p82(it5 s2)x6 p103(it6 s7)x6 p113(it7 s1)x6 p1(it0 s1)x5 p38(it2 s6)x5 p114(it7 s2)x5
[2026-10-08T15:18:22Z] hottest item p34-attack-f8 (0xdf7dca): 256 reads (0.0000% of all); predicted source none
[2026-10-08T15:18:22Z] program p34-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
Terminated
lease: released 8 pool cores after 603 s, exit 0

139
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p34/warps-p34-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,139 @@
# distinct lines_hash: value count
lines_hash 1008 2
lines_hash 1013 4
lines_hash 1014 66
lines_hash 1015 905
lines_hash 1016 7421
lines_hash 1019 3
lines_hash 1020 142
lines_hash 1021 4647
lines_hash 1022 115079
lines_hash 1023 1850370
lines_hash 1024 14798577
# distinct items_hash: value count
items_hash 126 2
items_hash 127 8396
items_hash 128 16768818
# distinct lines_warp: value count
lines_warp 32567 1
lines_warp 32575 1
lines_warp 32576 2
lines_warp 32577 4
lines_warp 32578 5
lines_warp 32579 2
lines_warp 32580 4
lines_warp 32581 7
lines_warp 32582 10
lines_warp 32583 17
lines_warp 32584 16
lines_warp 32585 12
lines_warp 32586 22
lines_warp 32587 33
lines_warp 32588 36
lines_warp 32589 44
lines_warp 32590 64
lines_warp 32591 49
lines_warp 32592 95
lines_warp 32593 109
lines_warp 32594 132
lines_warp 32595 158
lines_warp 32596 184
lines_warp 32597 251
lines_warp 32598 297
lines_warp 32599 318
lines_warp 32600 400
lines_warp 32601 479
lines_warp 32602 546
lines_warp 32603 654
lines_warp 32604 804
lines_warp 32605 919
lines_warp 32606 1122
lines_warp 32607 1302
lines_warp 32608 1496
lines_warp 32609 1773
lines_warp 32610 2097
lines_warp 32611 2290
lines_warp 32612 2751
lines_warp 32613 3102
lines_warp 32614 3492
lines_warp 32615 3883
lines_warp 32616 4442
lines_warp 32617 4943
lines_warp 32618 5594
lines_warp 32619 6188
lines_warp 32620 6817
lines_warp 32621 7565
lines_warp 32622 8188
lines_warp 32623 9074
lines_warp 32624 9622
lines_warp 32625 10407
lines_warp 32626 11148
lines_warp 32627 12104
lines_warp 32628 12751
lines_warp 32629 13145
lines_warp 32630 14000
lines_warp 32631 14790
lines_warp 32632 15206
lines_warp 32633 16002
lines_warp 32634 16187
lines_warp 32635 16515
lines_warp 32636 16566
lines_warp 32637 16548
lines_warp 32638 16880
lines_warp 32639 16578
lines_warp 32640 16604
lines_warp 32641 16193
lines_warp 32642 15448
lines_warp 32643 14977
lines_warp 32644 14407
lines_warp 32645 13669
lines_warp 32646 13114
lines_warp 32647 12056
lines_warp 32648 11437
lines_warp 32649 10555
lines_warp 32650 9663
lines_warp 32651 8866
lines_warp 32652 7896
lines_warp 32653 7214
lines_warp 32654 6455
lines_warp 32655 5618
lines_warp 32656 4884
lines_warp 32657 4200
lines_warp 32658 3622
lines_warp 32659 3189
lines_warp 32660 2678
lines_warp 32661 2217
lines_warp 32662 1844
lines_warp 32663 1571
lines_warp 32664 1258
lines_warp 32665 975
lines_warp 32666 797
lines_warp 32667 634
lines_warp 32668 488
lines_warp 32669 348
lines_warp 32670 312
lines_warp 32671 224
lines_warp 32672 174
lines_warp 32673 117
lines_warp 32674 92
lines_warp 32675 74
lines_warp 32676 54
lines_warp 32677 33
lines_warp 32678 32
lines_warp 32679 16
lines_warp 32680 15
lines_warp 32681 8
lines_warp 32682 6
lines_warp 32684 4
lines_warp 32685 4
lines_warp 32686 1
lines_warp 32687 1
lines_warp 32688 1
# distinct items_warp: value count
items_warp 4090 11
items_warp 4091 90
items_warp 4092 942
items_warp 4093 7316
items_warp 4094 41790
items_warp 4095 162230
items_warp 4096 311909

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p34/warps-p34-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 4 0.000019 0.000000
1 0 1 13 0.000037 0.000000
2 0 2 16 0.000068 0.000000
3 0 3 17 0.000019 0.000000
4 0 4 23 0.000068 0.000000
5 0 5 25 0.000020 0.000000
6 0 6 31 0.000019 0.000000
7 0 7 33 0.000068 0.000000
8 0 8 36 0.000036 0.000000
9 0 9 45 0.000037 0.000000
10 0 10 49 0.000019 0.000000
11 0 11 52 0.000069 0.000000
12 0 12 53 0.000019 0.000000
13 0 13 54 0.000019 0.000000
14 0 14 57 0.000019 0.000000
15 0 15 60 0.000020 0.000000
16 1 0 4 0.000019 0.000000
17 1 1 13 0.000036 0.000000
18 1 2 16 0.000070 0.000000
19 1 3 17 0.000019 0.000000
20 1 4 23 0.000068 0.000000
21 1 5 25 0.000019 0.000000
22 1 6 31 0.000020 0.000000
23 1 7 33 0.000069 0.000000
24 1 8 36 0.000036 0.000000
25 1 9 45 0.000037 0.000000
26 1 10 49 0.000020 0.000000
27 1 11 52 0.000068 0.000000
28 1 12 53 0.000019 0.000000
29 1 13 54 0.000019 0.000000
30 1 14 57 0.000019 0.000000
31 1 15 60 0.000020 0.000000
32 2 0 4 0.000020 0.000000
33 2 1 13 0.000037 0.000000
34 2 2 16 0.000068 0.000000
35 2 3 17 0.000019 0.000000
36 2 4 23 0.000068 0.000000
37 2 5 25 0.000020 0.000000
38 2 6 31 0.000019 0.000000
39 2 7 33 0.000069 0.000000
40 2 8 36 0.000036 0.000000
41 2 9 45 0.000036 0.000000
42 2 10 49 0.000019 0.000000
43 2 11 52 0.000069 0.000000
44 2 12 53 0.000020 0.000000
45 2 13 54 0.000020 0.000000
46 2 14 57 0.000019 0.000000
47 2 15 60 0.000019 0.000000
48 3 0 4 0.000019 0.000000
49 3 1 13 0.000036 0.000000
50 3 2 16 0.000069 0.000000
51 3 3 17 0.000019 0.000000
52 3 4 23 0.000068 0.000000
53 3 5 25 0.000019 0.000000
54 3 6 31 0.000019 0.000000
55 3 7 33 0.000069 0.000000
56 3 8 36 0.000037 0.000000
57 3 9 45 0.000036 0.000000
58 3 10 49 0.000020 0.000000
59 3 11 52 0.000069 0.000000
60 3 12 53 0.000020 0.000000
61 3 13 54 0.000019 0.000000
62 3 14 57 0.000019 0.000000
63 3 15 60 0.000019 0.000000
64 4 0 4 0.000020 0.000000
65 4 1 13 0.000036 0.000000
66 4 2 16 0.000070 0.000000
67 4 3 17 0.000019 0.000000
68 4 4 23 0.000069 0.000000
69 4 5 25 0.000019 0.000000
70 4 6 31 0.000019 0.000000
71 4 7 33 0.000068 0.000000
72 4 8 36 0.000037 0.000000
73 4 9 45 0.000036 0.000000
74 4 10 49 0.000020 0.000000
75 4 11 52 0.000068 0.000000
76 4 12 53 0.000019 0.000000
77 4 13 54 0.000019 0.000000
78 4 14 57 0.000020 0.000000
79 4 15 60 0.000020 0.000000
80 5 0 4 0.000020 0.000000
81 5 1 13 0.000036 0.000000
82 5 2 16 0.000069 0.000000
83 5 3 17 0.000020 0.000000
84 5 4 23 0.000069 0.000000
85 5 5 25 0.000020 0.000000
86 5 6 31 0.000019 0.000000
87 5 7 33 0.000069 0.000000
88 5 8 36 0.000036 0.000000
89 5 9 45 0.000036 0.000000
90 5 10 49 0.000019 0.000000
91 5 11 52 0.000069 0.000000
92 5 12 53 0.000020 0.000000
93 5 13 54 0.000019 0.000000
94 5 14 57 0.000019 0.000000
95 5 15 60 0.000019 0.000000
96 6 0 4 0.000020 0.000000
97 6 1 13 0.000036 0.000000
98 6 2 16 0.000068 0.000000
99 6 3 17 0.000019 0.000000
100 6 4 23 0.000069 0.000000
101 6 5 25 0.000019 0.000000
102 6 6 31 0.000020 0.000000
103 6 7 33 0.000068 0.000000
104 6 8 36 0.000036 0.000000
105 6 9 45 0.000037 0.000000
106 6 10 49 0.000020 0.000000
107 6 11 52 0.000069 0.000000
108 6 12 53 0.000020 0.000000
109 6 13 54 0.000019 0.000000
110 6 14 57 0.000020 0.000000
111 6 15 60 0.000019 0.000000
112 7 0 4 0.000020 0.000000
113 7 1 13 0.000036 0.000000
114 7 2 16 0.000068 0.000000
115 7 3 17 0.000019 0.000000
116 7 4 23 0.000069 0.000000
117 7 5 25 0.000019 0.000000
118 7 6 31 0.000019 0.000000
119 7 7 33 0.000069 0.000000
120 7 8 36 0.000037 0.000000
121 7 9 45 0.000036 0.000000
122 7 10 49 0.000020 0.000000
123 7 11 52 0.000068 0.000000
124 7 12 53 0.000020 0.000000
125 7 13 54 0.000019 0.000000
126 7 14 57 0.000020 0.000000
127 7 15 60 0.000020 0.000000

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p4.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (24,25,26,27,28,29,30,31, waited 393 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p4 at 2^24 (window control, by site)
[2026-10-08T15:14:48Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "4..4", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p4"]
[2026-10-08T15:14:48Z] census warps: programs [4] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:14:48Z] day 20730: cache filled in 0.45 s, fnv 448274a57f508cbc
[2026-10-08T15:14:59Z] table: 16777216 items derived with their 8 lines in 10.4 s; library comparison all (0 mismatches)
[2026-10-08T15:15:02Z] program p4-attack-f8: epoch seed 1afaec86f76841dbd11b07f8e797e76ea7741a919b552517be524dff6279e685 era seed 855a2e525d282e8eed84759f4963aed36eac73bfe1b5f4f3fac853568f03260f id cbea50e8712cf888 attempt 0 class mx8-erad8bbc64a+sh256x27+state+rw op mix load=16 rotl=9 rotr=9 add=7 sub=7 shfl=5 xor=5 mad=4 mul=2; era stride mul 0x2477779f rot 13 interleave [5, 10, 11, 15] windows 3:0:0 8:2:1 11:0:0 14:1:0 19:2:1 20:0:0 26:1:0 34:0:0 35:1:1 36:2:2 40:2:3 43:2:2 44:0:0 55:2:1 57:1:0 61:2:3
[2026-10-08T15:15:02Z] load site instr 3: src r0 win 0 off 0; last base writer sub at 2; writers back to the last injecting one: sub@2; shadow writes of r0 per rep: add=5 mad=2 mul=1 rotl=2 rotr=4 sub=5 xor=4
[2026-10-08T15:15:02Z] load site instr 8: src r5 win 2 off 1; last base writer load at 3; writers back to the last injecting one: load@3; shadow writes of r5 per rep: add=7 mad=5 mulhi=1 rotl=8 rotr=6 shfl=3 sub=7 xor=5
[2026-10-08T15:15:02Z] load site instr 11: src r0 win 0 off 0; last base writer shfl at 9; writers back to the last injecting one: shfl@9; shadow writes of r0 per rep: add=5 mad=2 mul=1 rotl=2 rotr=4 sub=5 xor=4
[2026-10-08T15:15:02Z] load site instr 14: src r7 win 1 off 0; last base writer load at 11; writers back to the last injecting one: load@11; shadow writes of r7 per rep: add=4 mad=1 mul=2 mulhi=1 rotl=6 rotr=4 shfl=1 sub=6 xor=5
[2026-10-08T15:15:02Z] load site instr 19: src r3 win 2 off 1; last base writer sub at 1; writers back to the last injecting one: sub@1; shadow writes of r3 per rep: add=9 mad=6 mul=2 rotl=2 rotr=4 sub=8 xor=1
[2026-10-08T15:15:02Z] load site instr 20: src r1 win 0 off 0; last base writer xor at 18; writers back to the last injecting one: xor@18; shadow writes of r1 per rep: add=3 mad=7 mul=2 mulhi=1 rotl=6 sub=2 xor=7
[2026-10-08T15:15:02Z] load site instr 26: src r1 win 1 off 0; last base writer xor at 24; writers back to the last injecting one: xor@24; shadow writes of r1 per rep: add=3 mad=7 mul=2 mulhi=1 rotl=6 sub=2 xor=7
[2026-10-08T15:15:02Z] load site instr 34: src r1 win 0 off 0; last base writer shfl at 32; writers back to the last injecting one: shfl@32; shadow writes of r1 per rep: add=3 mad=7 mul=2 mulhi=1 rotl=6 sub=2 xor=7
[2026-10-08T15:15:02Z] load site instr 35: src r4 win 1 off 1; last base writer rotl at 33; writers back to the last injecting one: rotl@33 load@19; shadow writes of r4 per rep: add=8 mad=6 mulhi=1 rotl=8 rotr=3 shfl=2 sub=2 xor=4
[2026-10-08T15:15:02Z] load site instr 36: src r7 win 2 off 2; last base writer sub at 30; writers back to the last injecting one: sub@30; shadow writes of r7 per rep: add=4 mad=1 mul=2 mulhi=1 rotl=6 rotr=4 shfl=1 sub=6 xor=5
[2026-10-08T15:15:02Z] load site instr 40: src r7 win 2 off 3; last base writer rotr at 37; writers back to the last injecting one: rotr@37 sub@30; shadow writes of r7 per rep: add=4 mad=1 mul=2 mulhi=1 rotl=6 rotr=4 shfl=1 sub=6 xor=5
[2026-10-08T15:15:02Z] load site instr 43: src r2 win 2 off 2; last base writer add at 31; writers back to the last injecting one: add@31; shadow writes of r2 per rep: add=10 mad=3 rotl=1 rotr=5 shfl=1 sub=3 xor=10
[2026-10-08T15:15:02Z] load site instr 44: src r3 win 0 off 0; last base writer mad at 28; writers back to the last injecting one: mad@28; shadow writes of r3 per rep: add=9 mad=6 mul=2 rotl=2 rotr=4 sub=8 xor=1
[2026-10-08T15:15:02Z] load site instr 55: src r1 win 2 off 1; last base writer rotl at 49; writers back to the last injecting one: rotl@49 shfl@32; shadow writes of r1 per rep: add=3 mad=7 mul=2 mulhi=1 rotl=6 sub=2 xor=7
[2026-10-08T15:15:02Z] load site instr 57: src r5 win 1 off 0; last base writer add at 52; writers back to the last injecting one: add@52; shadow writes of r5 per rep: add=7 mad=5 mulhi=1 rotl=8 rotr=6 shfl=3 sub=7 xor=5
[2026-10-08T15:15:02Z] load site instr 61: src r4 win 2 off 3; last base writer rotr at 60; writers back to the last injecting one: rotr@60 sub@56; shadow writes of r4 per rep: add=8 mad=6 mulhi=1 rotl=8 rotr=3 shfl=2 sub=2 xor=4
[2026-10-08T15:15:02Z] window layer: site item windows (first, items) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x0,2^23) (0x400000,2^22) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x800000,2^23) (0x800000,2^22) (0xc00000,2^22) (0x800000,2^22) (0x0,2^24) (0x400000,2^22) (0x0,2^23) (0xc00000,2^22); expected reads per item by quarter 88.000 184.000 120.000 120.000 (flat uniform 128.000)
[2026-10-08T15:17:41Z] 524288 warps (16777216 nonces) interpreted in 159.7 s (2.437 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:17:41Z] p4-attack-f8 distinct lines per hash: n 16777216 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8710
[2026-10-08T15:17:41Z] p4-attack-f8 distinct items per hash: n 16777216 min 126 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:17:41Z] p4-attack-f8 distinct lines per warp: n 524288 min 32569 p1 32605 median 32636 p99 32664 max 32687 mean 32636.0724
[2026-10-08T15:17:41Z] p4-attack-f8 distinct items per warp: n 524288 min 4089 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4626
[2026-10-08T15:17:41Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:17:41Z] positions: largest hi16-bucket (256 items) share 0.0070% at p111 (iteration 6, site 15, instr 61); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p70 (site 6, instr 26)
[2026-10-08T15:17:41Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p26 (site 10, instr 40); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:17:42Z] p4-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 254 (bin 4344343) z_max +11.14 min 45 z_min -7.34 chi2/dof 10.50157 chi2_z +27519.49 top0.1% 0.17610% top0.5% 0.85014% top1% 1.67052%
[2026-10-08T15:17:42Z] p4-attack-f8 item histogram against the window density: full 2^24 chi2/dof 0.99972 largest 183 (item 0xd2145b) at +5.75 sigma smallest at -4.84; buckets64 chi2/dof 1.00058 largest 8099 (bucket 226969) at +4.78 sigma smallest at -4.62
[2026-10-08T15:17:52Z] p4-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 255 (bin 4207222) z_max +11.23 min 44 z_min -7.42 chi2/dof 10.50127 chi2_z +27518.60 top0.1% 0.17613% top0.5% 0.85025% top1% 1.67059%
[2026-10-08T15:17:52Z] p4-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00040 largest +5.76 sigma; buckets64 chi2/dof 1.00665 largest +4.85 smallest -4.75
[2026-10-08T15:17:52Z] 6-sigma p4-attack-f8 items buckets64 against the window density: largest bucket +4.78 sigma, smallest -4.62 -> within 6 sigma
[2026-10-08T15:17:52Z] hot-set p4-attack-f8 items (windowed control): f 0.1% S_f 0.17610% E_f(control) 0.17613% X_f -0.00003% X_f/f -0.0003 -> no hot set
[2026-10-08T15:17:52Z] hot-set p4-attack-f8 items (windowed control): f 0.5% S_f 0.85014% E_f(control) 0.85025% X_f -0.00011% X_f/f -0.0002 -> no hot set
[2026-10-08T15:17:52Z] hot-set p4-attack-f8 items (windowed control): f 1.0% S_f 1.67052% E_f(control) 1.67059% X_f -0.00007% X_f/f -0.0001 -> no hot set
[2026-10-08T15:17:52Z] hot-set p4-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:18:02Z] p4-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 195 (bin 12547865) z_max +5.92 min 75 z_min -4.68 chi2/dof 1.00066 chi2_z +1.92 top0.1% 0.13106% top0.5% 0.63244% top1% 1.24332%
[2026-10-08T15:18:02Z] hot-set p4-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17610% E_f(control) 0.13106% X_f +0.04504% X_f/f +0.4504 -> no hot set
[2026-10-08T15:18:02Z] hot-set p4-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.85014% E_f(control) 0.63244% X_f +0.21770% X_f/f +0.4354 -> no hot set
[2026-10-08T15:18:02Z] hot-set p4-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.67052% E_f(control) 1.24332% X_f +0.42720% X_f/f +0.4272 -> no hot set
[2026-10-08T15:18:02Z] hot-set p4-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:18:02Z] ratio p4-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9998x / 0.9999x / 1.0000x (gate 1.2x at 0.1%: within); over the FLAT control 1.3436x / 1.3442x / 1.3436x
[2026-10-08T15:18:03Z] attribution: hot threshold count >= 221 marks 18256 items (0.1088% of items) holding 4108579 reads (0.1913% of reads); top 8 items t=0x7a64ac:254 t=0x424a17:254 t=0x6bdeed:252 t=0x689403:252 t=0x7fe58f:251 t=0x75a1f2:251 t=0x4d6140:251 t=0x7d250a:250
[2026-10-08T15:19:13Z] attribution: reads on hot items per position (flat expectation 0.1088% of each position's 16777216 reads); the 24 largest: p13(it0 s13)=0.535% p45(it2 s13)=0.535% p52(it3 s4)=0.534% p33(it2 s1)=0.534% p4(it0 s4)=0.534% p100(it6 s4)=0.534% p29(it1 s13)=0.533% p65(it4 s1)=0.533% p77(it4 s13)=0.532% p109(it6 s13)=0.532% p68(it4 s4)=0.532% p17(it1 s1)=0.532% p125(it7 s13)=0.532% p81(it5 s1)=0.532% p84(it5 s4)=0.532% p20(it1 s4)=0.531% p116(it7 s4)=0.531% p49(it3 s1)=0.531% p93(it5 s13)=0.531% p1(it0 s1)=0.531% p113(it7 s1)=0.531% p97(it6 s1)=0.531% p61(it3 s13)=0.530% p36(it2 s4)=0.530%
[2026-10-08T15:19:13Z] attribution: hot reads by iteration it0=0.192% it1=0.191% it2=0.191% it3=0.191% it4=0.191% it5=0.191% it6=0.191% it7=0.191% ; by site s0=0.133% s1=0.532% s2=0.133% s3=0.266% s4=0.532% s5=0.133% s6=0.266% s7=0.133% s8=0.000% s9=0.000% s10=0.000% s11=0.000% s12=0.134% s13=0.533% s14=0.267% s15=0.000%
[2026-10-08T15:19:13Z] site 0 (instr 3, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1332% (flat expectation 0.1088%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:19:13Z] site 1 (instr 8, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5316% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 2 (instr 11, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1334% (flat expectation 0.1088%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:19:13Z] site 3 (instr 14, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2659% (flat expectation 0.1088%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:19:13Z] site 4 (instr 19, src r3, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5321% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 5 (instr 20, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1331% (flat expectation 0.1088%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:19:13Z] site 6 (instr 26, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2659% (flat expectation 0.1088%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:19:13Z] site 7 (instr 34, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1331% (flat expectation 0.1088%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:19:13Z] site 8 (instr 35, src r4, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1088%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:19:13Z] site 9 (instr 36, src r7, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 10 (instr 40, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 11 (instr 43, src r2, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0063% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 12 (instr 44, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1335% (flat expectation 0.1088%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:19:13Z] site 13 (instr 55, src r1, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5325% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] site 14 (instr 57, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2669% (flat expectation 0.1088%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0032% (window expectation 0.0031%)
[2026-10-08T15:19:13Z] site 15 (instr 61, src r4, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1088%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:19:13Z] attribution: item 0x7a64ac (254 reads) read from positions p45(it2 s13)x15 p68(it4 s4)x8 p14(it0 s14)x7 p33(it2 s1)x7 p65(it4 s1)x7 p97(it6 s1)x7 p49(it3 s1)x6 p52(it3 s4)x6 p62(it3 s14)x6 p100(it6 s4)x6 p109(it6 s13)x6 p30(it1 s14)x5
[2026-10-08T15:19:13Z] attribution: item 0x424a17 (254 reads) read from positions p17(it1 s1)x9 p81(it5 s1)x9 p97(it6 s1)x9 p109(it6 s13)x9 p29(it1 s13)x8 p125(it7 s13)x7 p4(it0 s4)x7 p13(it0 s13)x7 p77(it4 s13)x7 p84(it5 s4)x7 p93(it5 s13)x7 p116(it7 s4)x7
[2026-10-08T15:19:13Z] attribution: item 0x6bdeed (252 reads) read from positions p68(it4 s4)x8 p116(it7 s4)x7 p61(it3 s13)x7 p93(it5 s13)x7 p4(it0 s4)x6 p35(it2 s3)x6 p65(it4 s1)x6 p86(it5 s6)x6 p1(it0 s1)x6 p109(it6 s13)x5 p125(it7 s13)x5 p13(it0 s13)x5
[2026-10-08T15:19:13Z] attribution: item 0x689403 (252 reads) read from positions p84(it5 s4)x11 p100(it6 s4)x10 p49(it3 s1)x10 p1(it0 s1)x9 p61(it3 s13)x8 p33(it2 s1)x8 p52(it3 s4)x8 p29(it1 s13)x7 p67(it4 s3)x7 p68(it4 s4)x7 p81(it5 s1)x7 p13(it0 s13)x6
[2026-10-08T15:19:13Z] attribution: item 0x7fe58f (251 reads) read from positions p36(it2 s4)x11 p125(it7 s13)x9 p49(it3 s1)x9 p29(it1 s13)x8 p45(it2 s13)x8 p61(it3 s13)x8 p84(it5 s4)x8 p4(it0 s4)x7 p35(it2 s3)x7 p109(it6 s13)x7 p13(it0 s13)x6 p70(it4 s6)x6
[2026-10-08T15:19:13Z] attribution: item 0x75a1f2 (251 reads) read from positions p68(it4 s4)x10 p93(it5 s13)x10 p65(it4 s1)x9 p17(it1 s1)x7 p109(it6 s13)x7 p62(it3 s14)x6 p84(it5 s4)x6 p94(it5 s14)x6 p100(it6 s4)x6 p6(it0 s6)x5 p13(it0 s13)x5 p30(it1 s14)x5
[2026-10-08T15:19:13Z] attribution: item 0x4d6140 (251 reads) read from positions p4(it0 s4)x9 p65(it4 s1)x9 p81(it5 s1)x9 p100(it6 s4)x8 p13(it0 s13)x7 p33(it2 s1)x7 p52(it3 s4)x7 p61(it3 s13)x7 p125(it7 s13)x6 p17(it1 s1)x6 p20(it1 s4)x6 p30(it1 s14)x6
[2026-10-08T15:19:13Z] attribution: item 0x7d250a (250 reads) read from positions p13(it0 s13)x13 p17(it1 s1)x10 p97(it6 s1)x10 p45(it2 s13)x9 p33(it2 s1)x8 p81(it5 s1)x8 p109(it6 s13)x8 p116(it7 s4)x8 p77(it4 s13)x7 p125(it7 s13)x6 p4(it0 s4)x6 p36(it2 s4)x6
[2026-10-08T15:19:13Z] hottest item p4-attack-f8 (0x7a64ac): 254 reads (0.0000% of all); predicted source none
[2026-10-08T15:19:13Z] program p4-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
Terminated
lease: released 8 pool cores after 658 s, exit 0

143
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p4/warps-p4-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,143 @@
# distinct lines_hash: value count
lines_hash 1008 4
lines_hash 1013 5
lines_hash 1014 56
lines_hash 1015 951
lines_hash 1016 7838
lines_hash 1019 2
lines_hash 1020 124
lines_hash 1021 4724
lines_hash 1022 114762
lines_hash 1023 1847930
lines_hash 1024 14800820
# distinct items_hash: value count
items_hash 126 4
items_hash 127 8850
items_hash 128 16768362
# distinct lines_warp: value count
lines_warp 32569 1
lines_warp 32570 1
lines_warp 32572 1
lines_warp 32573 2
lines_warp 32575 2
lines_warp 32576 2
lines_warp 32577 8
lines_warp 32578 3
lines_warp 32579 6
lines_warp 32580 4
lines_warp 32581 4
lines_warp 32582 9
lines_warp 32583 12
lines_warp 32584 18
lines_warp 32585 14
lines_warp 32586 15
lines_warp 32587 28
lines_warp 32588 41
lines_warp 32589 44
lines_warp 32590 48
lines_warp 32591 79
lines_warp 32592 93
lines_warp 32593 104
lines_warp 32594 124
lines_warp 32595 188
lines_warp 32596 200
lines_warp 32597 273
lines_warp 32598 284
lines_warp 32599 329
lines_warp 32600 388
lines_warp 32601 476
lines_warp 32602 614
lines_warp 32603 732
lines_warp 32604 847
lines_warp 32605 984
lines_warp 32606 1164
lines_warp 32607 1311
lines_warp 32608 1581
lines_warp 32609 1863
lines_warp 32610 2044
lines_warp 32611 2373
lines_warp 32612 2772
lines_warp 32613 3072
lines_warp 32614 3561
lines_warp 32615 4126
lines_warp 32616 4429
lines_warp 32617 5050
lines_warp 32618 5645
lines_warp 32619 6275
lines_warp 32620 6947
lines_warp 32621 7719
lines_warp 32622 8352
lines_warp 32623 9060
lines_warp 32624 9705
lines_warp 32625 10622
lines_warp 32626 11319
lines_warp 32627 12014
lines_warp 32628 12882
lines_warp 32629 13344
lines_warp 32630 14300
lines_warp 32631 14895
lines_warp 32632 15415
lines_warp 32633 15761
lines_warp 32634 16172
lines_warp 32635 16461
lines_warp 32636 16617
lines_warp 32637 16602
lines_warp 32638 16611
lines_warp 32639 16645
lines_warp 32640 16157
lines_warp 32641 15767
lines_warp 32642 15472
lines_warp 32643 15057
lines_warp 32644 14217
lines_warp 32645 13687
lines_warp 32646 12858
lines_warp 32647 12201
lines_warp 32648 11216
lines_warp 32649 10437
lines_warp 32650 9654
lines_warp 32651 8802
lines_warp 32652 7896
lines_warp 32653 7152
lines_warp 32654 6174
lines_warp 32655 5470
lines_warp 32656 4886
lines_warp 32657 4195
lines_warp 32658 3510
lines_warp 32659 2999
lines_warp 32660 2596
lines_warp 32661 2193
lines_warp 32662 1772
lines_warp 32663 1417
lines_warp 32664 1259
lines_warp 32665 1059
lines_warp 32666 805
lines_warp 32667 654
lines_warp 32668 479
lines_warp 32669 377
lines_warp 32670 301
lines_warp 32671 220
lines_warp 32672 181
lines_warp 32673 130
lines_warp 32674 100
lines_warp 32675 62
lines_warp 32676 48
lines_warp 32677 39
lines_warp 32678 33
lines_warp 32679 25
lines_warp 32680 13
lines_warp 32681 12
lines_warp 32682 6
lines_warp 32683 8
lines_warp 32684 1
lines_warp 32685 1
lines_warp 32686 1
lines_warp 32687 2
# distinct items_warp: value count
items_warp 4089 1
items_warp 4090 15
items_warp 4091 120
items_warp 4092 1026
items_warp 4093 7986
items_warp 4094 44278
items_warp 4095 164416
items_warp 4096 306446

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p4/warps-p4-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 3 0.000020 0.000000
1 0 1 8 0.000069 0.000000
2 0 2 11 0.000020 0.000000
3 0 3 14 0.000037 0.000000
4 0 4 19 0.000068 0.000000
5 0 5 20 0.000020 0.000000
6 0 6 26 0.000037 0.000000
7 0 7 34 0.000019 0.000000
8 0 8 35 0.000036 0.000000
9 0 9 36 0.000069 0.000000
10 0 10 40 0.000070 0.000000
11 0 11 43 0.000069 0.000000
12 0 12 44 0.000020 0.000000
13 0 13 55 0.000068 0.000000
14 0 14 57 0.000036 0.000000
15 0 15 61 0.000068 0.000000
16 1 0 3 0.000020 0.000000
17 1 1 8 0.000068 0.000000
18 1 2 11 0.000019 0.000000
19 1 3 14 0.000036 0.000000
20 1 4 19 0.000069 0.000000
21 1 5 20 0.000020 0.000000
22 1 6 26 0.000037 0.000000
23 1 7 34 0.000020 0.000000
24 1 8 35 0.000037 0.000000
25 1 9 36 0.000070 0.000000
26 1 10 40 0.000070 0.000000
27 1 11 43 0.000069 0.000000
28 1 12 44 0.000019 0.000000
29 1 13 55 0.000069 0.000000
30 1 14 57 0.000036 0.000000
31 1 15 61 0.000069 0.000000
32 2 0 3 0.000020 0.000000
33 2 1 8 0.000068 0.000000
34 2 2 11 0.000019 0.000000
35 2 3 14 0.000036 0.000000
36 2 4 19 0.000069 0.000000
37 2 5 20 0.000020 0.000000
38 2 6 26 0.000036 0.000000
39 2 7 34 0.000020 0.000000
40 2 8 35 0.000036 0.000000
41 2 9 36 0.000069 0.000000
42 2 10 40 0.000069 0.000000
43 2 11 43 0.000068 0.000000
44 2 12 44 0.000020 0.000000
45 2 13 55 0.000069 0.000000
46 2 14 57 0.000037 0.000000
47 2 15 61 0.000069 0.000000
48 3 0 3 0.000020 0.000000
49 3 1 8 0.000068 0.000000
50 3 2 11 0.000020 0.000000
51 3 3 14 0.000037 0.000000
52 3 4 19 0.000068 0.000000
53 3 5 20 0.000020 0.000000
54 3 6 26 0.000036 0.000000
55 3 7 34 0.000020 0.000000
56 3 8 35 0.000036 0.000000
57 3 9 36 0.000069 0.000000
58 3 10 40 0.000068 0.000000
59 3 11 43 0.000069 0.000000
60 3 12 44 0.000019 0.000000
61 3 13 55 0.000068 0.000000
62 3 14 57 0.000036 0.000000
63 3 15 61 0.000069 0.000000
64 4 0 3 0.000020 0.000000
65 4 1 8 0.000068 0.000000
66 4 2 11 0.000019 0.000000
67 4 3 14 0.000036 0.000000
68 4 4 19 0.000070 0.000000
69 4 5 20 0.000019 0.000000
70 4 6 26 0.000036 0.000000
71 4 7 34 0.000019 0.000000
72 4 8 35 0.000037 0.000000
73 4 9 36 0.000070 0.000000
74 4 10 40 0.000069 0.000000
75 4 11 43 0.000069 0.000000
76 4 12 44 0.000019 0.000000
77 4 13 55 0.000069 0.000000
78 4 14 57 0.000036 0.000000
79 4 15 61 0.000068 0.000000
80 5 0 3 0.000019 0.000000
81 5 1 8 0.000070 0.000000
82 5 2 11 0.000019 0.000000
83 5 3 14 0.000036 0.000000
84 5 4 19 0.000068 0.000000
85 5 5 20 0.000020 0.000000
86 5 6 26 0.000036 0.000000
87 5 7 34 0.000019 0.000000
88 5 8 35 0.000036 0.000000
89 5 9 36 0.000069 0.000000
90 5 10 40 0.000069 0.000000
91 5 11 43 0.000069 0.000000
92 5 12 44 0.000019 0.000000
93 5 13 55 0.000068 0.000000
94 5 14 57 0.000037 0.000000
95 5 15 61 0.000069 0.000000
96 6 0 3 0.000019 0.000000
97 6 1 8 0.000068 0.000000
98 6 2 11 0.000019 0.000000
99 6 3 14 0.000036 0.000000
100 6 4 19 0.000068 0.000000
101 6 5 20 0.000020 0.000000
102 6 6 26 0.000037 0.000000
103 6 7 34 0.000019 0.000000
104 6 8 35 0.000036 0.000000
105 6 9 36 0.000068 0.000000
106 6 10 40 0.000068 0.000000
107 6 11 43 0.000068 0.000000
108 6 12 44 0.000019 0.000000
109 6 13 55 0.000069 0.000000
110 6 14 57 0.000036 0.000000
111 6 15 61 0.000070 0.000000
112 7 0 3 0.000019 0.000000
113 7 1 8 0.000068 0.000000
114 7 2 11 0.000019 0.000000
115 7 3 14 0.000036 0.000000
116 7 4 19 0.000069 0.000000
117 7 5 20 0.000020 0.000000
118 7 6 26 0.000036 0.000000
119 7 7 34 0.000020 0.000000
120 7 8 35 0.000036 0.000000
121 7 9 36 0.000069 0.000000
122 7 10 40 0.000068 0.000000
123 7 11 43 0.000070 0.000000
124 7 12 44 0.000019 0.000000
125 7 13 55 0.000068 0.000000
126 7 14 57 0.000037 0.000000
127 7 15 61 0.000069 0.000000

78
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p8.log Normal file
View file

@ -0,0 +1,78 @@
lease: 0 of 2 pool cores free (24 leased); waiting
lease: holding 8 pool cores (8,24,25,26,27,28,29,30, waited 906 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 known-failed p8 at 2^24 (window control, by site)
[2026-10-08T15:23:22Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "8..8", "--nonces", "16777216", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-p8"]
[2026-10-08T15:23:22Z] census warps: programs [8] day 20730 nonces 16777216 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:23:23Z] day 20730: cache filled in 0.36 s, fnv 448274a57f508cbc
[2026-10-08T15:23:32Z] table: 16777216 items derived with their 8 lines in 8.7 s; library comparison all (0 mismatches)
[2026-10-08T15:23:34Z] program p8-attack-f8: epoch seed 3d53a453d039f63e75b785ad66bb213ad58b2791d73e617326017d7c0a113cc7 era seed 95a01b3804dcabcf9733d5486463bb492d592dd33f33d93a25a161fc88119f54 id 3ed6782cb437996c attempt 0 class mx8-erae58ec3a4+sh256x27+state+rw op mix load=16 rotl=10 add=9 sub=8 mad=7 rotr=5 xor=3 mul=2 mulhi=2 shfl=2; era stride mul 0x573e71b3 rot 22 interleave [1, 4, 7, 12] windows 5:2:2 6:0:0 11:2:2 14:2:2 15:0:0 17:0:0 20:0:0 26:2:0 30:1:0 34:0:0 38:2:1 42:2:1 47:2:1 51:0:0 55:0:0 56:0:0
[2026-10-08T15:23:34Z] load site instr 5: src r5 win 2 off 2; last base writer add at 3; writers back to the last injecting one: add@3; shadow writes of r5 per rep: add=8 mad=8 mul=1 mulhi=1 rotl=4 rotr=7 shfl=1 sub=2 xor=5
[2026-10-08T15:23:34Z] load site instr 6: src r3 win 0 off 0; last base writer load at 5; writers back to the last injecting one: load@5; shadow writes of r3 per rep: add=10 mad=3 mul=1 mulhi=1 rotl=3 rotr=6 sub=4 xor=5
[2026-10-08T15:23:34Z] load site instr 11: src r6 win 2 off 2; last base writer load at 6; writers back to the last injecting one: load@6; shadow writes of r6 per rep: add=4 mad=5 mul=3 mulhi=1 rotl=2 rotr=6 shfl=2 sub=1 xor=6
[2026-10-08T15:23:34Z] load site instr 14: src r1 win 2 off 2; last base writer add at 13; writers back to the last injecting one: add@13; shadow writes of r1 per rep: add=6 mad=7 mul=2 mulhi=1 rotl=5 rotr=5 shfl=1 sub=5 xor=6
[2026-10-08T15:23:34Z] load site instr 15: src r7 win 0 off 0; last base writer load at 14; writers back to the last injecting one: load@14; shadow writes of r7 per rep: add=7 mad=6 rotl=2 rotr=5 shfl=1 sub=7 xor=4
[2026-10-08T15:23:34Z] load site instr 17: src r5 win 0 off 0; last base writer load at 15; writers back to the last injecting one: load@15; shadow writes of r5 per rep: add=8 mad=8 mul=1 mulhi=1 rotl=4 rotr=7 shfl=1 sub=2 xor=5
[2026-10-08T15:23:34Z] load site instr 20: src r0 win 0 off 0; last base writer rotl at 18; writers back to the last injecting one: rotl@18 load@17; shadow writes of r0 per rep: add=2 mad=4 mul=3 mulhi=1 rotl=6 rotr=3 shfl=1 sub=2 xor=5
[2026-10-08T15:23:34Z] load site instr 26: src r2 win 2 off 0; last base writer sub at 16; writers back to the last injecting one: sub@16; shadow writes of r2 per rep: add=6 mad=6 mul=1 mulhi=1 rotl=1 rotr=4 shfl=3 sub=1 xor=7
[2026-10-08T15:23:34Z] load site instr 30: src r1 win 1 off 0; last base writer sub at 28; writers back to the last injecting one: sub@28; shadow writes of r1 per rep: add=6 mad=7 mul=2 mulhi=1 rotl=5 rotr=5 shfl=1 sub=5 xor=6
[2026-10-08T15:23:34Z] load site instr 34: src r6 win 0 off 0; last base writer mad at 33; writers back to the last injecting one: mad@33; shadow writes of r6 per rep: add=4 mad=5 mul=3 mulhi=1 rotl=2 rotr=6 shfl=2 sub=1 xor=6
[2026-10-08T15:23:34Z] load site instr 38: src r4 win 2 off 1; last base writer load at 20; writers back to the last injecting one: load@20; shadow writes of r4 per rep: add=6 mad=3 mul=1 mulhi=1 rotl=2 rotr=7 shfl=1 sub=3 xor=5
[2026-10-08T15:23:34Z] load site instr 42: src r5 win 2 off 1; last base writer rotr at 39; writers back to the last injecting one: rotr@39 rotr@37 add@35; shadow writes of r5 per rep: add=8 mad=8 mul=1 mulhi=1 rotl=4 rotr=7 shfl=1 sub=2 xor=5
[2026-10-08T15:23:34Z] load site instr 47: src r4 win 2 off 1; last base writer rotr at 40; writers back to the last injecting one: rotr@40 load@20; shadow writes of r4 per rep: add=6 mad=3 mul=1 mulhi=1 rotl=2 rotr=7 shfl=1 sub=3 xor=5
[2026-10-08T15:23:34Z] load site instr 51: src r7 win 0 off 0; last base writer add at 36; writers back to the last injecting one: add@36; shadow writes of r7 per rep: add=7 mad=6 rotl=2 rotr=5 shfl=1 sub=7 xor=4
[2026-10-08T15:23:34Z] load site instr 55: src r0 win 0 off 0; last base writer add at 53; writers back to the last injecting one: add@53; shadow writes of r0 per rep: add=2 mad=4 mul=3 mulhi=1 rotl=6 rotr=3 shfl=1 sub=2 xor=5
[2026-10-08T15:23:34Z] load site instr 56: src r3 win 0 off 0; last base writer xor at 54; writers back to the last injecting one: xor@54; shadow writes of r3 per rep: add=10 mad=3 mul=1 mulhi=1 rotl=3 rotr=6 sub=4 xor=5
[2026-10-08T15:23:34Z] window layer: site item windows (first, items) (0x800000,2^22) (0x0,2^24) (0x800000,2^22) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x0,2^23) (0x0,2^24) (0x400000,2^22) (0x400000,2^22) (0x400000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24); expected reads per item by quarter 112.000 176.000 160.000 64.000 (flat uniform 128.000)
[2026-10-08T15:26:17Z] 524288 warps (16777216 nonces) interpreted in 163.2 s (2.490 ms per warp per thread); Epoch::hash_warp agreement on 589 warps: 0 mismatches
[2026-10-08T15:26:17Z] p8-attack-f8 distinct lines per hash: n 16777216 min 983 p1 1008 median 1024 p99 1024 max 1024 mean 1021.8727
[2026-10-08T15:26:17Z] p8-attack-f8 distinct items per hash: n 16777216 min 123 p1 126 median 128 p99 128 max 128 mean 127.7496
[2026-10-08T15:26:17Z] p8-attack-f8 distinct lines per warp: n 524288 min 32438 p1 32509 median 32573 p99 32627 max 32671 mean 32572.4507
[2026-10-08T15:26:17Z] p8-attack-f8 distinct items per warp: n 524288 min 4073 p1 4080 median 4088 p99 4093 max 4096 mean 4087.4499
[2026-10-08T15:26:17Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:26:17Z] positions: largest hi16-bucket (256 items) share 0.0070% at p114 (iteration 7, site 2, instr 11); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p88 (site 8, instr 30)
[2026-10-08T15:26:17Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p6 (site 6, instr 20); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:26:17Z] p8-attack-f8 item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 258 (bin 4211841) z_max +11.49 min 27 z_min -8.93 chi2/dof 16.23608 chi2_z +44128.41 top0.1% 0.17211% top0.5% 0.82770% top1% 1.62358%
[2026-10-08T15:26:17Z] p8-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.28391 largest 195 (item 0xd097) at +7.84 sigma smallest at -5.22; buckets64 chi2/dof 18.97856 largest 9058 (bucket 1654) at +22.32 sigma smallest at -7.81
[2026-10-08T15:26:30Z] p8-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 246 (bin 4332293) z_max +10.43 min 25 z_min -9.10 chi2/dof 16.00091 chi2_z +43447.28 top0.1% 0.16920% top0.5% 0.81646% top1% 1.60412%
[2026-10-08T15:26:30Z] p8-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99990 largest +5.88 sigma; buckets64 chi2/dof 0.99913 largest +4.74 smallest -4.38
[2026-10-08T15:26:30Z] 6-sigma p8-attack-f8 items buckets64 against the window density: largest bucket +22.32 sigma, smallest -7.81 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:26:30Z] hot-set p8-attack-f8 items (windowed control): f 0.1% S_f 0.17211% E_f(control) 0.16920% X_f +0.00291% X_f/f +0.0291 -> no hot set
[2026-10-08T15:26:30Z] hot-set p8-attack-f8 items (windowed control): f 0.5% S_f 0.82770% E_f(control) 0.81646% X_f +0.01124% X_f/f +0.0225 -> no hot set
[2026-10-08T15:26:30Z] hot-set p8-attack-f8 items (windowed control): f 1.0% S_f 1.62358% E_f(control) 1.60412% X_f +0.01946% X_f/f +0.0195 -> no hot set
[2026-10-08T15:26:30Z] hot-set p8-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:26:42Z] p8-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 2147483648 mean 128.000 sigma 11.314 max 197 (bin 2508372) z_max +6.10 min 72 z_min -4.95 chi2/dof 1.00022 chi2_z +0.62 top0.1% 0.13107% top0.5% 0.63247% top1% 1.24336%
[2026-10-08T15:26:42Z] hot-set p8-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17211% E_f(control) 0.13107% X_f +0.04104% X_f/f +0.4104 -> no hot set
[2026-10-08T15:26:42Z] hot-set p8-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.82770% E_f(control) 0.63247% X_f +0.19523% X_f/f +0.3905 -> no hot set
[2026-10-08T15:26:42Z] hot-set p8-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.62358% E_f(control) 1.24336% X_f +0.38022% X_f/f +0.3802 -> no hot set
[2026-10-08T15:26:42Z] hot-set p8-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:26:42Z] ratio p8-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0172x / 1.0138x / 1.0121x (gate 1.2x at 0.1%: within); over the FLAT control 1.3131x / 1.3087x / 1.3058x
[2026-10-08T15:26:43Z] attribution: hot threshold count >= 215 marks 20168 items (0.1202% of items) holding 4425205 reads (0.2061% of reads); top 8 items t=0x404481:258 t=0x5a91b3:252 t=0x419a7c:252 t=0x410d02:251 t=0x422bbd:250 t=0x429429:247 t=0x400bf4:247 t=0x4335b0:246
[2026-10-08T15:28:04Z] attribution: reads on hot items per position (flat expectation 0.1202% of each position's 16777216 reads); the 24 largest: p90(it5 s10)=0.559% p108(it6 s12)=0.559% p106(it6 s10)=0.559% p74(it4 s10)=0.558% p44(it2 s12)=0.558% p92(it5 s12)=0.558% p28(it1 s12)=0.558% p124(it7 s12)=0.557% p10(it0 s10)=0.557% p12(it0 s12)=0.557% p76(it4 s12)=0.557% p122(it7 s10)=0.557% p26(it1 s10)=0.556% p60(it3 s12)=0.556% p58(it3 s10)=0.556% p75(it4 s11)=0.556% p91(it5 s11)=0.555% p42(it2 s10)=0.555% p59(it3 s11)=0.555% p107(it6 s11)=0.554% p11(it0 s11)=0.554% p43(it2 s11)=0.553% p27(it1 s11)=0.553% p123(it7 s11)=0.553%
[2026-10-08T15:28:04Z] attribution: hot reads by iteration it0=0.206% it1=0.206% it2=0.206% it3=0.206% it4=0.206% it5=0.206% it6=0.206% it7=0.206% ; by site s0=0.025% s1=0.144% s2=0.025% s3=0.024% s4=0.145% s5=0.144% s6=0.145% s7=0.000% s8=0.276% s9=0.267% s10=0.557% s11=0.554% s12=0.557% s13=0.145% s14=0.144% s15=0.144%
[2026-10-08T15:28:04Z] site 0 (instr 5, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0245% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 1 (instr 6, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1443% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 2 (instr 11, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0247% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 3 (instr 14, src r1, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0245% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 4 (instr 15, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1446% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 5 (instr 17, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1442% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 6 (instr 20, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1451% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 7 (instr 26, src r2, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 8 (instr 30, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2764% (flat expectation 0.1202%); index entropy 15.000 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0033% (window expectation 0.0031%)
[2026-10-08T15:28:04Z] site 9 (instr 34, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2667% (flat expectation 0.1202%); index entropy 15.713 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0063% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 10 (instr 38, src r4, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5570% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 11 (instr 42, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5542% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 12 (instr 47, src r4, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5575% (flat expectation 0.1202%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0064% (window expectation 0.0061%)
[2026-10-08T15:28:04Z] site 13 (instr 51, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1446% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 14 (instr 55, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1444% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] site 15 (instr 56, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1443% (flat expectation 0.1202%); index entropy 16.000 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0017% (window expectation 0.0015%)
[2026-10-08T15:28:04Z] attribution: item 0x404481 (258 reads) read from positions p26(it1 s10)x13 p12(it0 s12)x10 p90(it5 s10)x10 p124(it7 s12)x8 p91(it5 s11)x8 p11(it0 s11)x7 p28(it1 s12)x7 p43(it2 s11)x7 p74(it4 s10)x7 p75(it4 s11)x7 p105(it6 s9)x7 p10(it0 s10)x6
[2026-10-08T15:28:04Z] attribution: item 0x5a91b3 (252 reads) read from positions p58(it3 s10)x10 p60(it3 s12)x10 p90(it5 s10)x9 p75(it4 s11)x8 p122(it7 s10)x8 p91(it5 s11)x7 p92(it5 s12)x7 p123(it7 s11)x6 p11(it0 s11)x6 p24(it1 s8)x6 p26(it1 s10)x6 p27(it1 s11)x6
[2026-10-08T15:28:04Z] attribution: item 0x419a7c (252 reads) read from positions p75(it4 s11)x11 p27(it1 s11)x10 p26(it1 s10)x7 p88(it5 s8)x7 p90(it5 s10)x7 p108(it6 s12)x7 p122(it7 s10)x7 p124(it7 s12)x6 p12(it0 s12)x6 p74(it4 s10)x6 p91(it5 s11)x6 p121(it7 s9)x6
[2026-10-08T15:28:04Z] attribution: item 0x410d02 (251 reads) read from positions p124(it7 s12)x10 p122(it7 s10)x8 p123(it7 s11)x8 p36(it2 s4)x7 p90(it5 s10)x7 p92(it5 s12)x7 p10(it0 s10)x6 p28(it1 s12)x6 p44(it2 s12)x6 p57(it3 s9)x6 p74(it4 s10)x6 p75(it4 s11)x6
[2026-10-08T15:28:04Z] attribution: item 0x422bbd (250 reads) read from positions p107(it6 s11)x12 p60(it3 s12)x9 p108(it6 s12)x9 p10(it0 s10)x8 p27(it1 s11)x8 p90(it5 s10)x8 p92(it5 s12)x8 p76(it4 s12)x7 p124(it7 s12)x7 p59(it3 s11)x6 p106(it6 s10)x6 p58(it3 s10)x5
[2026-10-08T15:28:04Z] attribution: item 0x429429 (247 reads) read from positions p42(it2 s10)x11 p44(it2 s12)x10 p76(it4 s12)x9 p108(it6 s12)x7 p11(it0 s11)x6 p74(it4 s10)x6 p75(it4 s11)x6 p91(it5 s11)x6 p105(it6 s9)x6 p107(it6 s11)x6 p123(it7 s11)x5 p9(it0 s9)x5
[2026-10-08T15:28:04Z] attribution: item 0x400bf4 (247 reads) read from positions p11(it0 s11)x10 p43(it2 s11)x9 p58(it3 s10)x9 p123(it7 s11)x8 p124(it7 s12)x8 p60(it3 s12)x7 p73(it4 s9)x7 p9(it0 s9)x6 p76(it4 s12)x6 p108(it6 s12)x6 p121(it7 s9)x6 p122(it7 s10)x6
[2026-10-08T15:28:04Z] attribution: item 0x4335b0 (246 reads) read from positions p43(it2 s11)x10 p60(it3 s12)x9 p28(it1 s12)x8 p59(it3 s11)x8 p124(it7 s12)x7 p10(it0 s10)x7 p42(it2 s10)x7 p108(it6 s12)x6 p26(it1 s10)x6 p11(it0 s11)x6 p58(it3 s10)x5 p73(it4 s9)x5
[2026-10-08T15:28:04Z] hottest item p8-attack-f8 (0x404481): 258 reads (0.0000% of all); predicted source none
[2026-10-08T15:28:04Z] program p8-attack-f8 DONE: 16777216 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
Terminated
lease: released 8 pool cores after 1188 s, exit 0

280
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p8/warps-p8-attack-f8-d20730-n16777216-none-distinct.txt Normal file
View file

@ -0,0 +1,280 @@
# distinct lines_hash: value count
lines_hash 983 2
lines_hash 984 24
lines_hash 990 9
lines_hash 991 112
lines_hash 992 925
lines_hash 997 6
lines_hash 998 154
lines_hash 999 2557
lines_hash 1000 21635
lines_hash 1004 3
lines_hash 1005 90
lines_hash 1006 2404
lines_hash 1007 40704
lines_hash 1008 337307
lines_hash 1011 1
lines_hash 1012 37
lines_hash 1013 956
lines_hash 1014 22239
lines_hash 1015 366931
lines_hash 1016 2971882
lines_hash 1019 4
lines_hash 1020 100
lines_hash 1021 3710
lines_hash 1022 89392
lines_hash 1023 1433279
lines_hash 1024 11482753
# distinct items_hash: value count
items_hash 123 26
items_hash 124 1046
items_hash 125 24352
items_hash 126 380508
items_hash 127 3362046
items_hash 128 13009238
# distinct lines_warp: value count
lines_warp 32438 1
lines_warp 32441 1
lines_warp 32442 1
lines_warp 32443 1
lines_warp 32445 1
lines_warp 32446 1
lines_warp 32449 2
lines_warp 32450 1
lines_warp 32453 1
lines_warp 32454 2
lines_warp 32456 2
lines_warp 32457 2
lines_warp 32458 1
lines_warp 32459 3
lines_warp 32460 5
lines_warp 32461 5
lines_warp 32462 5
lines_warp 32463 3
lines_warp 32464 12
lines_warp 32465 8
lines_warp 32466 6
lines_warp 32467 14
lines_warp 32468 13
lines_warp 32469 12
lines_warp 32470 12
lines_warp 32471 15
lines_warp 32472 11
lines_warp 32473 8
lines_warp 32474 17
lines_warp 32475 17
lines_warp 32476 16
lines_warp 32477 21
lines_warp 32478 22
lines_warp 32479 30
lines_warp 32480 32
lines_warp 32481 34
lines_warp 32482 33
lines_warp 32483 38
lines_warp 32484 50
lines_warp 32485 67
lines_warp 32486 57
lines_warp 32487 69
lines_warp 32488 76
lines_warp 32489 90
lines_warp 32490 75
lines_warp 32491 84
lines_warp 32492 111
lines_warp 32493 126
lines_warp 32494 123
lines_warp 32495 140
lines_warp 32496 147
lines_warp 32497 154
lines_warp 32498 173
lines_warp 32499 202
lines_warp 32500 227
lines_warp 32501 234
lines_warp 32502 270
lines_warp 32503 258
lines_warp 32504 319
lines_warp 32505 307
lines_warp 32506 373
lines_warp 32507 376
lines_warp 32508 428
lines_warp 32509 457
lines_warp 32510 510
lines_warp 32511 575
lines_warp 32512 617
lines_warp 32513 656
lines_warp 32514 698
lines_warp 32515 738
lines_warp 32516 806
lines_warp 32517 884
lines_warp 32518 907
lines_warp 32519 995
lines_warp 32520 1078
lines_warp 32521 1160
lines_warp 32522 1268
lines_warp 32523 1320
lines_warp 32524 1402
lines_warp 32525 1534
lines_warp 32526 1608
lines_warp 32527 1779
lines_warp 32528 1806
lines_warp 32529 1913
lines_warp 32530 2020
lines_warp 32531 2168
lines_warp 32532 2274
lines_warp 32533 2443
lines_warp 32534 2529
lines_warp 32535 2678
lines_warp 32536 2768
lines_warp 32537 2873
lines_warp 32538 2992
lines_warp 32539 3260
lines_warp 32540 3390
lines_warp 32541 3458
lines_warp 32542 3758
lines_warp 32543 3849
lines_warp 32544 4079
lines_warp 32545 4308
lines_warp 32546 4499
lines_warp 32547 4626
lines_warp 32548 4740
lines_warp 32549 4987
lines_warp 32550 5132
lines_warp 32551 5343
lines_warp 32552 5515
lines_warp 32553 5655
lines_warp 32554 6027
lines_warp 32555 6078
lines_warp 32556 6087
lines_warp 32557 6453
lines_warp 32558 6604
lines_warp 32559 6639
lines_warp 32560 7036
lines_warp 32561 7048
lines_warp 32562 7024
lines_warp 32563 7504
lines_warp 32564 7382
lines_warp 32565 7586
lines_warp 32566 7610
lines_warp 32567 7800
lines_warp 32568 7889
lines_warp 32569 8138
lines_warp 32570 7907
lines_warp 32571 8048
lines_warp 32572 8287
lines_warp 32573 8263
lines_warp 32574 8446
lines_warp 32575 8070
lines_warp 32576 7993
lines_warp 32577 8314
lines_warp 32578 8173
lines_warp 32579 8005
lines_warp 32580 8057
lines_warp 32581 8023
lines_warp 32582 7881
lines_warp 32583 7937
lines_warp 32584 7986
lines_warp 32585 7567
lines_warp 32586 7513
lines_warp 32587 7464
lines_warp 32588 7168
lines_warp 32589 7210
lines_warp 32590 7100
lines_warp 32591 6723
lines_warp 32592 6632
lines_warp 32593 6466
lines_warp 32594 6219
lines_warp 32595 6007
lines_warp 32596 5889
lines_warp 32597 5588
lines_warp 32598 5422
lines_warp 32599 5217
lines_warp 32600 5002
lines_warp 32601 4679
lines_warp 32602 4569
lines_warp 32603 4437
lines_warp 32604 4222
lines_warp 32605 3915
lines_warp 32606 3646
lines_warp 32607 3517
lines_warp 32608 3342
lines_warp 32609 3146
lines_warp 32610 2975
lines_warp 32611 2761
lines_warp 32612 2560
lines_warp 32613 2292
lines_warp 32614 2207
lines_warp 32615 1987
lines_warp 32616 1933
lines_warp 32617 1725
lines_warp 32618 1594
lines_warp 32619 1456
lines_warp 32620 1346
lines_warp 32621 1238
lines_warp 32622 1110
lines_warp 32623 980
lines_warp 32624 894
lines_warp 32625 856
lines_warp 32626 741
lines_warp 32627 721
lines_warp 32628 604
lines_warp 32629 538
lines_warp 32630 514
lines_warp 32631 422
lines_warp 32632 398
lines_warp 32633 327
lines_warp 32634 289
lines_warp 32635 269
lines_warp 32636 256
lines_warp 32637 195
lines_warp 32638 165
lines_warp 32639 149
lines_warp 32640 131
lines_warp 32641 105
lines_warp 32642 94
lines_warp 32643 90
lines_warp 32644 72
lines_warp 32645 57
lines_warp 32646 49
lines_warp 32647 42
lines_warp 32648 35
lines_warp 32649 28
lines_warp 32650 24
lines_warp 32651 23
lines_warp 32652 13
lines_warp 32653 17
lines_warp 32654 16
lines_warp 32655 9
lines_warp 32656 4
lines_warp 32657 5
lines_warp 32658 2
lines_warp 32659 2
lines_warp 32660 1
lines_warp 32661 6
lines_warp 32662 1
lines_warp 32663 3
lines_warp 32664 1
lines_warp 32671 1
# distinct items_warp: value count
items_warp 4073 12
items_warp 4074 26
items_warp 4075 65
items_warp 4076 146
items_warp 4077 345
items_warp 4078 793
items_warp 4079 1736
items_warp 4080 3691
items_warp 4081 7120
items_warp 4082 12800
items_warp 4083 21036
items_warp 4084 32332
items_warp 4085 45635
items_warp 4086 59199
items_warp 4087 69463
items_warp 4088 72348
items_warp 4089 68100
items_warp 4090 55139
items_warp 4091 37937
items_warp 4092 21924
items_warp 4093 10052
items_warp 4094 3474
items_warp 4095 807
items_warp 4096 108

129
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-p8/warps-p8-attack-f8-d20730-n16777216-none-positions.txt Normal file
View file

@ -0,0 +1,129 @@
# position iteration site instr max_hi16_bucket_share saturated_share
0 0 0 5 0.000068 0.000000
1 0 1 6 0.000020 0.000000
2 0 2 11 0.000069 0.000000
3 0 3 14 0.000069 0.000000
4 0 4 15 0.000019 0.000000
5 0 5 17 0.000020 0.000000
6 0 6 20 0.000020 0.000000
7 0 7 26 0.000068 0.000000
8 0 8 30 0.000036 0.000000
9 0 9 34 0.000066 0.000000
10 0 10 38 0.000069 0.000000
11 0 11 42 0.000068 0.000000
12 0 12 47 0.000069 0.000000
13 0 13 51 0.000019 0.000000
14 0 14 55 0.000020 0.000000
15 0 15 56 0.000019 0.000000
16 1 0 5 0.000068 0.000000
17 1 1 6 0.000020 0.000000
18 1 2 11 0.000068 0.000000
19 1 3 14 0.000069 0.000000
20 1 4 15 0.000020 0.000000
21 1 5 17 0.000019 0.000000
22 1 6 20 0.000020 0.000000
23 1 7 26 0.000069 0.000000
24 1 8 30 0.000037 0.000000
25 1 9 34 0.000067 0.000000
26 1 10 38 0.000069 0.000000
27 1 11 42 0.000068 0.000000
28 1 12 47 0.000069 0.000000
29 1 13 51 0.000020 0.000000
30 1 14 55 0.000020 0.000000
31 1 15 56 0.000020 0.000000
32 2 0 5 0.000068 0.000000
33 2 1 6 0.000020 0.000000
34 2 2 11 0.000068 0.000000
35 2 3 14 0.000069 0.000000
36 2 4 15 0.000019 0.000000
37 2 5 17 0.000019 0.000000
38 2 6 20 0.000020 0.000000
39 2 7 26 0.000069 0.000000
40 2 8 30 0.000036 0.000000
41 2 9 34 0.000068 0.000000
42 2 10 38 0.000069 0.000000
43 2 11 42 0.000068 0.000000
44 2 12 47 0.000068 0.000000
45 2 13 51 0.000020 0.000000
46 2 14 55 0.000020 0.000000
47 2 15 56 0.000019 0.000000
48 3 0 5 0.000068 0.000000
49 3 1 6 0.000019 0.000000
50 3 2 11 0.000069 0.000000
51 3 3 14 0.000068 0.000000
52 3 4 15 0.000020 0.000000
53 3 5 17 0.000020 0.000000
54 3 6 20 0.000020 0.000000
55 3 7 26 0.000070 0.000000
56 3 8 30 0.000036 0.000000
57 3 9 34 0.000068 0.000000
58 3 10 38 0.000069 0.000000
59 3 11 42 0.000069 0.000000
60 3 12 47 0.000068 0.000000
61 3 13 51 0.000019 0.000000
62 3 14 55 0.000019 0.000000
63 3 15 56 0.000020 0.000000
64 4 0 5 0.000069 0.000000
65 4 1 6 0.000019 0.000000
66 4 2 11 0.000069 0.000000
67 4 3 14 0.000069 0.000000
68 4 4 15 0.000019 0.000000
69 4 5 17 0.000020 0.000000
70 4 6 20 0.000020 0.000000
71 4 7 26 0.000069 0.000000
72 4 8 30 0.000036 0.000000
73 4 9 34 0.000067 0.000000
74 4 10 38 0.000068 0.000000
75 4 11 42 0.000069 0.000000
76 4 12 47 0.000069 0.000000
77 4 13 51 0.000019 0.000000
78 4 14 55 0.000020 0.000000
79 4 15 56 0.000020 0.000000
80 5 0 5 0.000070 0.000000
81 5 1 6 0.000019 0.000000
82 5 2 11 0.000068 0.000000
83 5 3 14 0.000069 0.000000
84 5 4 15 0.000019 0.000000
85 5 5 17 0.000019 0.000000
86 5 6 20 0.000019 0.000000
87 5 7 26 0.000069 0.000000
88 5 8 30 0.000036 0.000000
89 5 9 34 0.000067 0.000000
90 5 10 38 0.000069 0.000000
91 5 11 42 0.000068 0.000000
92 5 12 47 0.000069 0.000000
93 5 13 51 0.000019 0.000000
94 5 14 55 0.000019 0.000000
95 5 15 56 0.000020 0.000000
96 6 0 5 0.000069 0.000000
97 6 1 6 0.000020 0.000000
98 6 2 11 0.000068 0.000000
99 6 3 14 0.000069 0.000000
100 6 4 15 0.000019 0.000000
101 6 5 17 0.000019 0.000000
102 6 6 20 0.000020 0.000000
103 6 7 26 0.000069 0.000000
104 6 8 30 0.000036 0.000000
105 6 9 34 0.000067 0.000000
106 6 10 38 0.000069 0.000000
107 6 11 42 0.000069 0.000000
108 6 12 47 0.000069 0.000000
109 6 13 51 0.000019 0.000000
110 6 14 55 0.000019 0.000000
111 6 15 56 0.000020 0.000000
112 7 0 5 0.000068 0.000000
113 7 1 6 0.000020 0.000000
114 7 2 11 0.000070 0.000000
115 7 3 14 0.000069 0.000000
116 7 4 15 0.000019 0.000000
117 7 5 17 0.000020 0.000000
118 7 6 20 0.000020 0.000000
119 7 7 26 0.000069 0.000000
120 7 8 30 0.000036 0.000000
121 7 9 34 0.000068 0.000000
122 7 10 38 0.000068 0.000000
123 7 11 42 0.000069 0.000000
124 7 12 47 0.000068 0.000000
125 7 13 51 0.000019 0.000000
126 7 14 55 0.000020 0.000000
127 7 15 56 0.000019 0.000000

900
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/f8-seeds.log Normal file
View file

@ -0,0 +1,900 @@
lease: 0 of 4 pool cores free (24 leased); waiting
lease: holding 8 pool cores (12,13,14,15,17,21,22,23, waited 544 s, class v5, mem 8 GB): class v6 census: pack hl-v6-rw F8 census 16 seeds p18 to p33 at 2^22 (window control, by site)
[2026-10-08T15:17:26Z] attack-f8 0.1.0 (igneum-pow 4); args ["census", "--programs", "18..33", "--nonces", "4194304", "--control", "window", "--by-site", "--class", "v5", "--state", "/srv/builds/v6-census/packs/node1-state.igsd1", "--load-class", "mx8+sh256x27+state+rw", "--era-widths", "4", "--day", "20730", "--threads", "8", "--out", "/srv/builds/v6-census/packs-out/hl-v6-rw/f8-seeds"]
[2026-10-08T15:17:26Z] census warps: programs [18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33] day 20730 nonces 4194304 threads 8 plant none validate all check_every 997 diag true
[2026-10-08T15:17:26Z] day 20730: cache filled in 0.44 s, fnv 448274a57f508cbc
[2026-10-08T15:17:37Z] table: 16777216 items derived with their 8 lines in 10.6 s; library comparison all (0 mismatches)
[2026-10-08T15:17:40Z] program p18-attack-f8: epoch seed aa4f71160ecbb87bde40eb5e5731ae84346adcd48b533b26f35c0925d7f40784 era seed 82bb5b72dfbeb137505c844e103906c8b93aacc7f61d3fce1b24d8c0e149c8e2 id b9ea7b20b6c5bef4 attempt 0 class mx8-era74045749+sh256x27+state+rw op mix load=16 add=10 mad=10 sub=6 rotl=5 shfl=4 xor=4 mul=3 mulhi=3 rotr=3; era stride mul 0x83e41bc3 rot 22 interleave [1, 8, 10, 13] windows 2:1:1 10:0:0 15:1:1 18:0:0 22:0:0 24:2:0 27:2:0 30:0:0 31:2:2 39:1:1 43:2:0 46:2:3 47:2:2 49:1:1 56:2:0 63:2:0
[2026-10-08T15:17:40Z] window layer: site item windows (first, items) (0x800000,2^23) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x0,2^22) (0x0,2^24) (0x800000,2^22) (0x800000,2^23) (0x0,2^22) (0xc00000,2^22) (0x800000,2^22) (0x800000,2^23) (0x0,2^22) (0x0,2^22); expected reads per item by quarter 48.000 8.000 40.000 32.000 (flat uniform 32.000)
[2026-10-08T15:18:19Z] 131072 warps (4194304 nonces) interpreted in 38.8 s (2.371 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:18:19Z] p18-attack-f8 distinct lines per hash: n 4194304 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8706
[2026-10-08T15:18:19Z] p18-attack-f8 distinct items per hash: n 4194304 min 126 p1 128 median 128 p99 128 max 128 mean 127.9994
[2026-10-08T15:18:19Z] p18-attack-f8 distinct lines per warp: n 131072 min 32567 p1 32604 median 32636 p99 32663 max 32686 mean 32635.4753
[2026-10-08T15:18:19Z] p18-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.3887
[2026-10-08T15:18:19Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:18:19Z] positions: largest hi16-bucket (256 items) share 0.0206% at p26 (iteration 1, site 10, instr 43); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 2)
[2026-10-08T15:18:19Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p10 (site 10, instr 43); saturated sources at one position 0 of 2048 at p0 (site 0, instr 2); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:18:19Z] p18-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 105 (bin 107390) z_max +12.90 min 0 z_min -5.66 chi2/dof 8.23472 chi2_z +20953.99 top0.1% 0.25140% top0.5% 1.15492% top1% 2.20526%
[2026-10-08T15:18:19Z] p18-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.15684 largest 105 (item 0x1a37e) at +8.23 sigma smallest at -4.76; buckets64 chi2/dof 11.01245 largest 4339 (bucket 4031) at +22.86 sigma smallest at -8.53
[2026-10-08T15:18:22Z] p18-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 88 (bin 826391) z_max +9.90 min 0 z_min -5.66 chi2/dof 7.99973 chi2_z +20273.38 top0.1% 0.21812% top0.5% 1.02501% top1% 1.98765%
[2026-10-08T15:18:22Z] p18-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00040 largest +6.36 sigma; buckets64 chi2/dof 0.99818 largest +4.38 smallest -4.35
[2026-10-08T15:18:22Z] 6-sigma p18-attack-f8 items buckets64 against the window density: largest bucket +22.86 sigma, smallest -8.53 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:18:22Z] hot-set p18-attack-f8 items (windowed control): f 0.1% S_f 0.25140% E_f(control) 0.21812% X_f +0.03328% X_f/f +0.3328 -> no hot set
[2026-10-08T15:18:22Z] hot-set p18-attack-f8 items (windowed control): f 0.5% S_f 1.15492% E_f(control) 1.02501% X_f +0.12991% X_f/f +0.2598 -> no hot set
[2026-10-08T15:18:22Z] hot-set p18-attack-f8 items (windowed control): f 1.0% S_f 2.20526% E_f(control) 1.98765% X_f +0.21761% X_f/f +0.2176 -> no hot set
[2026-10-08T15:18:22Z] hot-set p18-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:18:24Z] p18-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 67 (bin 12430837) z_max +6.19 min 6 z_min -4.60 chi2/dof 1.00007 chi2_z +0.20 top0.1% 0.16452% top0.5% 0.77438% top1% 1.50135%
[2026-10-08T15:18:24Z] hot-set p18-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.25140% E_f(control) 0.16452% X_f +0.08689% X_f/f +0.8689 -> no hot set
[2026-10-08T15:18:24Z] hot-set p18-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 1.15492% E_f(control) 0.77438% X_f +0.38054% X_f/f +0.7611 -> no hot set
[2026-10-08T15:18:24Z] hot-set p18-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 2.20526% E_f(control) 1.50135% X_f +0.70391% X_f/f +0.7039 -> no hot set
[2026-10-08T15:18:24Z] hot-set p18-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:18:24Z] ratio p18-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.1526x / 1.1267x / 1.1095x (gate 1.2x at 0.1%: within); over the FLAT control 1.5281x / 1.4914x / 1.4689x
[2026-10-08T15:18:25Z] attribution: hot threshold count >= 77 marks 18143 items (0.1081% of items) holding 1454885 reads (0.2710% of reads); top 8 items t=0x01a37e:105 t=0x031d3f:102 t=0x01f024:102 t=0x00bd55:101 t=0x028221:100 t=0x026ff2:100 t=0x005330:100 t=0x03072a:99
[2026-10-08T15:18:41Z] attribution: reads on hot items per position (flat expectation 0.1081% of each position's 4194304 reads); the 24 largest: p42(it2 s10)=1.592% p90(it5 s10)=1.589% p106(it6 s10)=1.588% p26(it1 s10)=1.588% p74(it4 s10)=1.586% p58(it3 s10)=1.585% p122(it7 s10)=1.583% p10(it0 s10)=1.579% p37(it2 s5)=0.556% p22(it1 s6)=0.555% p95(it5 s15)=0.554% p118(it7 s6)=0.553% p85(it5 s5)=0.553% p14(it0 s14)=0.552% p117(it7 s5)=0.552% p21(it1 s5)=0.552% p101(it6 s5)=0.551% p78(it4 s14)=0.551% p15(it0 s15)=0.550% p5(it0 s5)=0.550% p94(it5 s14)=0.550% p63(it3 s15)=0.550% p62(it3 s14)=0.550% p31(it1 s15)=0.549%
[2026-10-08T15:18:41Z] attribution: hot reads by iteration it0=0.271% it1=0.271% it2=0.271% it3=0.271% it4=0.270% it5=0.271% it6=0.271% it7=0.271% ; by site s0=0.000% s1=0.137% s2=0.000% s3=0.138% s4=0.138% s5=0.551% s6=0.549% s7=0.139% s8=0.000% s9=0.000% s10=1.586% s11=0.000% s12=0.000% s13=0.000% s14=0.548% s15=0.549%
[2026-10-08T15:18:41Z] site 0 (instr 2, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:18:41Z] site 1 (instr 10, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1373% (flat expectation 0.1081%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:18:41Z] site 2 (instr 15, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:18:41Z] site 3 (instr 18, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1378% (flat expectation 0.1081%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:18:41Z] site 4 (instr 22, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1376% (flat expectation 0.1081%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:18:41Z] site 5 (instr 24, src r2, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.5510% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 6 (instr 27, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.5493% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 7 (instr 30, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1387% (flat expectation 0.1081%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:18:41Z] site 8 (instr 31, src r1, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 9 (instr 39, src r6, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:18:41Z] site 10 (instr 43, src r7, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 1.5862% (flat expectation 0.1081%); index entropy 13.718 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0191% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 11 (instr 46, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 12 (instr 47, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0001% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 13 (instr 49, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1081%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:18:41Z] site 14 (instr 56, src r2, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.5483% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] site 15 (instr 63, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.5493% (flat expectation 0.1081%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:18:41Z] attribution: item 0x01a37e (105 reads) read from positions p58(it3 s10)x9 p26(it1 s10)x6 p90(it5 s10)x6 p122(it7 s10)x4 p74(it4 s10)x4 p21(it1 s5)x3 p42(it2 s10)x3 p49(it3 s1)x3 p63(it3 s15)x3 p86(it5 s6)x3 p1(it0 s1)x3 p95(it5 s15)x3
[2026-10-08T15:18:41Z] attribution: item 0x031d3f (102 reads) read from positions p10(it0 s10)x6 p106(it6 s10)x6 p90(it5 s10)x5 p122(it7 s10)x4 p86(it5 s6)x4 p30(it1 s14)x3 p54(it3 s6)x3 p58(it3 s10)x3 p74(it4 s10)x3 p78(it4 s14)x3 p85(it5 s5)x3 p101(it6 s5)x3
[2026-10-08T15:18:41Z] attribution: item 0x01f024 (102 reads) read from positions p90(it5 s10)x9 p26(it1 s10)x6 p42(it2 s10)x5 p10(it0 s10)x4 p54(it3 s6)x4 p58(it3 s10)x4 p94(it5 s14)x4 p106(it6 s10)x4 p127(it7 s15)x3 p30(it1 s14)x3 p53(it3 s5)x3 p118(it7 s6)x3
[2026-10-08T15:18:41Z] attribution: item 0x00bd55 (101 reads) read from positions p42(it2 s10)x6 p74(it4 s10)x6 p10(it0 s10)x5 p58(it3 s10)x5 p15(it0 s15)x4 p38(it2 s6)x4 p54(it3 s6)x4 p62(it3 s14)x3 p90(it5 s10)x3 p95(it5 s15)x3 p106(it6 s10)x3 p111(it6 s15)x3
[2026-10-08T15:18:41Z] attribution: item 0x028221 (100 reads) read from positions p106(it6 s10)x8 p10(it0 s10)x5 p127(it7 s15)x4 p26(it1 s10)x4 p42(it2 s10)x4 p58(it3 s10)x4 p90(it5 s10)x4 p110(it6 s14)x4 p122(it7 s10)x4 p101(it6 s5)x3 p126(it7 s14)x3 p35(it2 s3)x3
[2026-10-08T15:18:41Z] attribution: item 0x026ff2 (100 reads) read from positions p74(it4 s10)x10 p106(it6 s10)x7 p101(it6 s5)x5 p79(it4 s15)x4 p90(it5 s10)x4 p118(it7 s6)x3 p26(it1 s10)x3 p30(it1 s14)x3 p37(it2 s5)x3 p46(it2 s14)x3 p78(it4 s14)x3 p86(it5 s6)x3
[2026-10-08T15:18:41Z] attribution: item 0x005330 (100 reads) read from positions p106(it6 s10)x8 p58(it3 s10)x7 p10(it0 s10)x5 p26(it1 s10)x5 p53(it3 s5)x5 p42(it2 s10)x4 p126(it7 s14)x3 p21(it1 s5)x3 p22(it1 s6)x3 p46(it2 s14)x3 p67(it4 s3)x3 p90(it5 s10)x3
[2026-10-08T15:18:41Z] attribution: item 0x03072a (99 reads) read from positions p122(it7 s10)x8 p42(it2 s10)x5 p46(it2 s14)x5 p74(it4 s10)x5 p6(it0 s6)x4 p26(it1 s10)x4 p53(it3 s5)x4 p106(it6 s10)x4 p10(it0 s10)x3 p30(it1 s14)x3 p31(it1 s15)x3 p79(it4 s15)x3
[2026-10-08T15:18:41Z] hottest item p18-attack-f8 (0x01a37e): 105 reads (0.0000% of all); predicted source none
[2026-10-08T15:18:41Z] program p18-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
[2026-10-08T15:18:44Z] program p19-attack-f8: epoch seed c6dce23f9819f5f9345525f70dc31b7f28b6e4d3a5bfaa0eb8b0c0073658bf2b era seed 37cc6fa9e2e7baa2c206142573c2799bf06f19c1ebfab3990d9b4003b3b4dad2 id fa128bbce81d993d attempt 1 class mx8-era09f19b22+sh256x27+state+rw op mix load=16 rotl=12 xor=11 add=7 mad=7 rotr=5 sub=3 mul=1 mulhi=1 shfl=1; era stride mul 0x69944467 rot 9 interleave [6, 9, 11, 12] windows 1:1:0 4:2:3 8:1:1 13:0:0 18:1:1 20:0:0 28:0:0 29:2:3 40:0:0 45:2:0 46:1:0 56:1:0 57:2:0 58:1:0 60:1:0 62:1:1
[2026-10-08T15:18:44Z] window layer: site item windows (first, items) (0x0,2^23) (0xc00000,2^22) (0x800000,2^23) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x0,2^24) (0x0,2^22) (0x0,2^23) (0x0,2^23) (0x0,2^22) (0x0,2^23) (0x0,2^23) (0x800000,2^23); expected reads per item by quarter 44.000 28.000 20.000 36.000 (flat uniform 32.000)
[2026-10-08T15:19:21Z] 131072 warps (4194304 nonces) interpreted in 37.5 s (2.286 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:19:21Z] p19-attack-f8 distinct lines per hash: n 4194304 min 984 p1 1008 median 1024 p99 1024 max 1024 mean 1021.8707
[2026-10-08T15:19:21Z] p19-attack-f8 distinct items per hash: n 4194304 min 123 p1 126 median 128 p99 128 max 128 mean 127.7493
[2026-10-08T15:19:21Z] p19-attack-f8 distinct lines per warp: n 131072 min 32443 p1 32509 median 32574 p99 32627 max 32665 mean 32572.5577
[2026-10-08T15:19:21Z] p19-attack-f8 distinct items per warp: n 131072 min 4073 p1 4080 median 4088 p99 4093 max 4096 mean 4087.4626
[2026-10-08T15:19:21Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:19:21Z] positions: largest hi16-bucket (256 items) share 0.0081% at p119 (iteration 7, site 7, instr 29); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 1)
[2026-10-08T15:19:21Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p90 (site 10, instr 46); saturated sources at one position 0 of 2048 at p0 (site 0, instr 1); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:19:21Z] p19-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 81 (bin 3168960) z_max +8.66 min 1 z_min -5.48 chi2/dof 3.50289 chi2_z +7249.14 top0.1% 0.20310% top0.5% 0.95167% top1% 1.84228%
[2026-10-08T15:19:21Z] p19-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00389 largest 61 (item 0x77e0b5) at +6.24 sigma smallest at -4.52; buckets64 chi2/dof 1.00389 largest 2000 (bucket 118629) at +4.91 sigma smallest at -4.52
[2026-10-08T15:19:24Z] p19-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 85 (bin 1425407) z_max +9.37 min 2 z_min -5.30 chi2/dof 3.49995 chi2_z +7240.64 top0.1% 0.20291% top0.5% 0.95108% top1% 1.84117%
[2026-10-08T15:19:24Z] p19-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00011 largest +6.24 sigma; buckets64 chi2/dof 1.00021 largest +5.09 smallest -4.61
[2026-10-08T15:19:24Z] 6-sigma p19-attack-f8 items buckets64 against the window density: largest bucket +4.91 sigma, smallest -4.52 -> within 6 sigma
[2026-10-08T15:19:24Z] hot-set p19-attack-f8 items (windowed control): f 0.1% S_f 0.20310% E_f(control) 0.20291% X_f +0.00019% X_f/f +0.0019 -> no hot set
[2026-10-08T15:19:24Z] hot-set p19-attack-f8 items (windowed control): f 0.5% S_f 0.95167% E_f(control) 0.95108% X_f +0.00059% X_f/f +0.0012 -> no hot set
[2026-10-08T15:19:24Z] hot-set p19-attack-f8 items (windowed control): f 1.0% S_f 1.84228% E_f(control) 1.84117% X_f +0.00111% X_f/f +0.0011 -> no hot set
[2026-10-08T15:19:24Z] hot-set p19-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:19:26Z] p19-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 67 (bin 6805290) z_max +6.19 min 8 z_min -4.24 chi2/dof 1.00034 chi2_z +0.97 top0.1% 0.16461% top0.5% 0.77434% top1% 1.50119%
[2026-10-08T15:19:26Z] hot-set p19-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.20310% E_f(control) 0.16461% X_f +0.03849% X_f/f +0.3849 -> no hot set
[2026-10-08T15:19:26Z] hot-set p19-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.95167% E_f(control) 0.77434% X_f +0.17734% X_f/f +0.3547 -> no hot set
[2026-10-08T15:19:26Z] hot-set p19-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.84228% E_f(control) 1.50119% X_f +0.34108% X_f/f +0.3411 -> no hot set
[2026-10-08T15:19:26Z] hot-set p19-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:19:26Z] ratio p19-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0010x / 1.0006x / 1.0006x (gate 1.2x at 0.1%: within); over the FLAT control 1.2338x / 1.2290x / 1.2272x
[2026-10-08T15:19:27Z] attribution: hot threshold count >= 63 marks 17554 items (0.1046% of items) holding 1139339 reads (0.2122% of reads); top 8 items t=0x305ac0:81 t=0x288a00:80 t=0x239635:80 t=0x054873:80 t=0x3dcc9f:79 t=0x12db77:79 t=0x03a55d:79 t=0x039629:79
[2026-10-08T15:19:47Z] attribution: reads on hot items per position (flat expectation 0.1046% of each position's 4194304 reads); the 24 largest: p12(it0 s12)=0.620% p76(it4 s12)=0.620% p28(it1 s12)=0.618% p121(it7 s9)=0.616% p124(it7 s12)=0.614% p73(it4 s9)=0.613% p44(it2 s12)=0.612% p92(it5 s12)=0.611% p57(it3 s9)=0.611% p89(it5 s9)=0.610% p108(it6 s12)=0.610% p25(it1 s9)=0.609% p41(it2 s9)=0.608% p105(it6 s9)=0.608% p60(it3 s12)=0.606% p9(it0 s9)=0.606% p106(it6 s10)=0.319% p26(it1 s10)=0.316% p122(it7 s10)=0.316% p11(it0 s11)=0.315% p74(it4 s10)=0.313% p123(it7 s11)=0.311% p42(it2 s10)=0.311% p64(it4 s0)=0.310%
[2026-10-08T15:19:47Z] attribution: hot reads by iteration it0=0.213% it1=0.212% it2=0.212% it3=0.211% it4=0.213% it5=0.212% it6=0.212% it7=0.213% ; by site s0=0.307% s1=0.005% s2=0.002% s3=0.153% s4=0.002% s5=0.154% s6=0.154% s7=0.005% s8=0.154% s9=0.610% s10=0.312% s11=0.310% s12=0.614% s13=0.306% s14=0.306% s15=0.002%
[2026-10-08T15:19:47Z] site 0 (instr 1, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3067% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 1 (instr 4, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0047% (flat expectation 0.1046%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:19:47Z] site 2 (instr 8, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0023% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 3 (instr 13, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1531% (flat expectation 0.1046%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:19:47Z] site 4 (instr 18, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0024% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 5 (instr 20, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1536% (flat expectation 0.1046%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:19:47Z] site 6 (instr 28, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1542% (flat expectation 0.1046%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:19:47Z] site 7 (instr 29, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0045% (flat expectation 0.1046%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:19:47Z] site 8 (instr 40, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1536% (flat expectation 0.1046%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:19:47Z] site 9 (instr 45, src r0, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6100% (flat expectation 0.1046%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:19:47Z] site 10 (instr 46, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3121% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 11 (instr 56, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3098% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 12 (instr 57, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6138% (flat expectation 0.1046%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:19:47Z] site 13 (instr 58, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3064% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 14 (instr 60, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3058% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] site 15 (instr 62, src r6, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0025% (flat expectation 0.1046%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:19:47Z] attribution: item 0x305ac0 (81 reads) read from positions p89(it5 s9)x5 p14(it0 s14)x4 p92(it5 s12)x4 p105(it6 s9)x4 p108(it6 s12)x3 p57(it3 s9)x3 p74(it4 s10)x3 p12(it0 s12)x2 p40(it2 s8)x2 p43(it2 s11)x2 p45(it2 s13)x2 p58(it3 s10)x2
[2026-10-08T15:19:47Z] attribution: item 0x288a00 (80 reads) read from positions p41(it2 s9)x5 p25(it1 s9)x4 p76(it4 s12)x4 p121(it7 s9)x3 p12(it0 s12)x3 p60(it3 s12)x3 p85(it5 s5)x3 p86(it5 s6)x3 p92(it5 s12)x3 p108(it6 s12)x3 p112(it7 s0)x3 p57(it3 s9)x2
[2026-10-08T15:19:47Z] attribution: item 0x239635 (80 reads) read from positions p28(it1 s12)x4 p109(it6 s13)x3 p11(it0 s11)x3 p64(it4 s0)x3 p108(it6 s12)x3 p0(it0 s0)x2 p124(it7 s12)x2 p9(it0 s9)x2 p29(it1 s13)x2 p32(it2 s0)x2 p45(it2 s13)x2 p46(it2 s14)x2
[2026-10-08T15:19:47Z] attribution: item 0x054873 (80 reads) read from positions p27(it1 s11)x4 p28(it1 s12)x4 p44(it2 s12)x4 p9(it0 s9)x3 p12(it0 s12)x3 p29(it1 s13)x3 p107(it6 s11)x3 p118(it7 s6)x3 p121(it7 s9)x2 p11(it0 s11)x2 p14(it0 s14)x2 p24(it1 s8)x2
[2026-10-08T15:19:47Z] attribution: item 0x3dcc9f (79 reads) read from positions p60(it3 s12)x6 p28(it1 s12)x4 p57(it3 s9)x4 p121(it7 s9)x4 p37(it2 s5)x3 p42(it2 s10)x3 p46(it2 s14)x3 p61(it3 s13)x3 p73(it4 s9)x3 p92(it5 s12)x3 p124(it7 s12)x2 p14(it0 s14)x2
[2026-10-08T15:19:47Z] attribution: item 0x12db77 (79 reads) read from positions p41(it2 s9)x4 p105(it6 s9)x4 p124(it7 s12)x3 p12(it0 s12)x3 p27(it1 s11)x3 p32(it2 s0)x3 p46(it2 s14)x3 p64(it4 s0)x3 p73(it4 s9)x3 p112(it7 s0)x3 p9(it0 s9)x2 p10(it0 s10)x2
[2026-10-08T15:19:47Z] attribution: item 0x03a55d (79 reads) read from positions p89(it5 s9)x5 p42(it2 s10)x4 p73(it4 s9)x4 p105(it6 s9)x3 p12(it0 s12)x3 p43(it2 s11)x3 p57(it3 s9)x3 p58(it3 s10)x3 p8(it0 s8)x2 p27(it1 s11)x2 p28(it1 s12)x2 p32(it2 s0)x2
[2026-10-08T15:19:47Z] attribution: item 0x039629 (79 reads) read from positions p57(it3 s9)x5 p0(it0 s0)x3 p25(it1 s9)x3 p32(it2 s0)x3 p35(it2 s3)x3 p45(it2 s13)x3 p89(it5 s9)x3 p105(it6 s9)x3 p108(it6 s12)x3 p124(it7 s12)x2 p29(it1 s13)x2 p41(it2 s9)x2
[2026-10-08T15:19:47Z] hottest item p19-attack-f8 (0x305ac0): 81 reads (0.0000% of all); predicted source none
[2026-10-08T15:19:47Z] program p19-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:19:50Z] program p20-attack-f8: epoch seed dcb66b784dffed6161fcfeb14601d08ad17195f25e240d7dd799524e3067550e era seed 686b6162e4c68cc4d1d74c883cbca002cfb50b78d47c5265ecc0d890d0a75f5d id 0aa00969fcd4c14c attempt 0 class mx8-erade9a6760+sh256x27+state+rw op mix load=16 rotr=8 sub=8 xor=8 add=7 rotl=6 mad=5 shfl=4 mul=2; era stride mul 0xaa378839 rot 28 interleave [0, 3, 7, 10] windows 5:1:0 6:2:1 8:0:0 15:0:0 18:0:0 20:0:0 21:1:0 22:2:2 27:2:0 28:1:0 36:2:0 45:0:0 49:1:1 50:2:2 55:2:1 60:1:1
[2026-10-08T15:19:50Z] window layer: site item windows (first, items) (0x0,2^23) (0x400000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^23) (0x800000,2^22) (0x0,2^22) (0x0,2^23) (0x0,2^22) (0x0,2^24) (0x800000,2^23) (0x800000,2^22) (0x400000,2^22) (0x800000,2^23); expected reads per item by quarter 38.000 38.000 34.000 18.000 (flat uniform 32.000)
[2026-10-08T15:20:28Z] 131072 warps (4194304 nonces) interpreted in 38.6 s (2.356 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:20:28Z] p20-attack-f8 distinct lines per hash: n 4194304 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8712
[2026-10-08T15:20:28Z] p20-attack-f8 distinct items per hash: n 4194304 min 126 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:20:28Z] p20-attack-f8 distinct lines per warp: n 131072 min 32571 p1 32605 median 32637 p99 32664 max 32686 mean 32636.1400
[2026-10-08T15:20:28Z] p20-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4672
[2026-10-08T15:20:28Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:20:28Z] positions: largest hi16-bucket (256 items) share 0.0082% at p56 (iteration 3, site 8, instr 27); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 5)
[2026-10-08T15:20:28Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p15 (site 15, instr 60); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:20:28Z] p20-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 76 (bin 3610005) z_max +7.78 min 1 z_min -5.48 chi2/dof 3.12464 chi2_z +6153.60 top0.1% 0.18453% top0.5% 0.86740% top1% 1.68314%
[2026-10-08T15:20:29Z] p20-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00035 largest 76 (item 0x371595) at +6.16 sigma smallest at -4.54; buckets64 chi2/dof 1.00269 largest 1304 (bucket 201846) at +4.48 sigma smallest at -4.56
[2026-10-08T15:20:31Z] p20-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 76 (bin 4338426) z_max +7.78 min 1 z_min -5.48 chi2/dof 3.12509 chi2_z +6154.91 top0.1% 0.18447% top0.5% 0.86732% top1% 1.68301%
[2026-10-08T15:20:31Z] p20-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00042 largest +6.16 sigma; buckets64 chi2/dof 1.00149 largest +4.91 smallest -4.77
[2026-10-08T15:20:31Z] 6-sigma p20-attack-f8 items buckets64 against the window density: largest bucket +4.48 sigma, smallest -4.56 -> within 6 sigma
[2026-10-08T15:20:31Z] hot-set p20-attack-f8 items (windowed control): f 0.1% S_f 0.18453% E_f(control) 0.18447% X_f +0.00006% X_f/f +0.0006 -> no hot set
[2026-10-08T15:20:31Z] hot-set p20-attack-f8 items (windowed control): f 0.5% S_f 0.86740% E_f(control) 0.86732% X_f +0.00008% X_f/f +0.0002 -> no hot set
[2026-10-08T15:20:31Z] hot-set p20-attack-f8 items (windowed control): f 1.0% S_f 1.68314% E_f(control) 1.68301% X_f +0.00013% X_f/f +0.0001 -> no hot set
[2026-10-08T15:20:31Z] hot-set p20-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:20:33Z] p20-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 68 (bin 9738263) z_max +6.36 min 7 z_min -4.42 chi2/dof 1.00036 chi2_z +1.03 top0.1% 0.16465% top0.5% 0.77453% top1% 1.50148%
[2026-10-08T15:20:33Z] hot-set p20-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.18453% E_f(control) 0.16465% X_f +0.01988% X_f/f +0.1988 -> no hot set
[2026-10-08T15:20:33Z] hot-set p20-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.86740% E_f(control) 0.77453% X_f +0.09287% X_f/f +0.1857 -> no hot set
[2026-10-08T15:20:33Z] hot-set p20-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.68314% E_f(control) 1.50148% X_f +0.18166% X_f/f +0.1817 -> no hot set
[2026-10-08T15:20:33Z] hot-set p20-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:20:33Z] ratio p20-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0003x / 1.0001x / 1.0001x (gate 1.2x at 0.1%: within); over the FLAT control 1.1207x / 1.1199x / 1.1210x
[2026-10-08T15:20:34Z] attribution: hot threshold count >= 57 marks 20870 items (0.1244% of items) holding 1223994 reads (0.2280% of reads); top 8 items t=0x371595:76 t=0x270f1a:73 t=0x6cb549:71 t=0x670bd2:71 t=0x3df9f5:71 t=0x15645e:71 t=0x79421d:70 t=0x785ee5:70
[2026-10-08T15:20:52Z] attribution: reads on hot items per position (flat expectation 0.1244% of each position's 4194304 reads); the 24 largest: p9(it0 s9)=0.372% p86(it5 s6)=0.372% p10(it0 s10)=0.371% p72(it4 s8)=0.371% p113(it7 s1)=0.371% p80(it5 s0)=0.371% p16(it1 s0)=0.371% p8(it0 s8)=0.370% p110(it6 s14)=0.370% p88(it5 s8)=0.370% p32(it2 s0)=0.370% p40(it2 s8)=0.370% p120(it7 s8)=0.370% p118(it7 s6)=0.370% p54(it3 s6)=0.370% p122(it7 s10)=0.370% p17(it1 s1)=0.370% p6(it0 s6)=0.369% p46(it2 s14)=0.369% p96(it6 s0)=0.369% p48(it3 s0)=0.369% p49(it3 s1)=0.368% p65(it4 s1)=0.368% p104(it6 s8)=0.368%
[2026-10-08T15:20:52Z] attribution: hot reads by iteration it0=0.228% it1=0.228% it2=0.228% it3=0.228% it4=0.228% it5=0.228% it6=0.228% it7=0.229% ; by site s0=0.369% s1=0.368% s2=0.193% s3=0.193% s4=0.194% s5=0.195% s6=0.368% s7=0.036% s8=0.369% s9=0.366% s10=0.367% s11=0.193% s12=0.018% s13=0.035% s14=0.367% s15=0.017%
[2026-10-08T15:20:52Z] site 0 (instr 5, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3687% (flat expectation 0.1244%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:20:52Z] site 1 (instr 6, src r3, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.3675% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 2 (instr 8, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1927% (flat expectation 0.1244%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:20:52Z] site 3 (instr 15, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1934% (flat expectation 0.1244%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:20:52Z] site 4 (instr 18, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1939% (flat expectation 0.1244%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:20:52Z] site 5 (instr 20, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1951% (flat expectation 0.1244%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:20:52Z] site 6 (instr 21, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3679% (flat expectation 0.1244%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:20:52Z] site 7 (instr 22, src r4, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0356% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 8 (instr 27, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.3688% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 9 (instr 28, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3661% (flat expectation 0.1244%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:20:52Z] site 10 (instr 36, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.3675% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 11 (instr 45, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1932% (flat expectation 0.1244%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:20:52Z] site 12 (instr 49, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0179% (flat expectation 0.1244%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:20:52Z] site 13 (instr 50, src r7, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0351% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 14 (instr 55, src r2, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.3671% (flat expectation 0.1244%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:20:52Z] site 15 (instr 60, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0174% (flat expectation 0.1244%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:20:52Z] attribution: item 0x371595 (76 reads) read from positions p40(it2 s8)x7 p120(it7 s8)x3 p56(it3 s8)x3 p64(it4 s0)x3 p80(it5 s0)x3 p88(it5 s8)x3 p96(it6 s0)x3 p6(it0 s6)x2 p8(it0 s8)x2 p10(it0 s10)x2 p16(it1 s0)x2 p18(it1 s2)x2
[2026-10-08T15:20:52Z] attribution: item 0x270f1a (73 reads) read from positions p26(it1 s10)x6 p56(it3 s8)x4 p120(it7 s8)x3 p10(it0 s10)x3 p43(it2 s11)x3 p106(it6 s10)x3 p20(it1 s4)x2 p22(it1 s6)x2 p36(it2 s4)x2 p38(it2 s6)x2 p41(it2 s9)x2 p54(it3 s6)x2
[2026-10-08T15:20:52Z] attribution: item 0x6cb549 (71 reads) read from positions p49(it3 s1)x6 p30(it1 s14)x4 p102(it6 s6)x4 p14(it0 s14)x3 p41(it2 s9)x3 p81(it5 s1)x3 p97(it6 s1)x3 p126(it7 s14)x2 p6(it0 s6)x2 p17(it1 s1)x2 p46(it2 s14)x2 p54(it3 s6)x2
[2026-10-08T15:20:52Z] attribution: item 0x670bd2 (71 reads) read from positions p46(it2 s14)x4 p4(it0 s4)x3 p14(it0 s14)x3 p16(it1 s0)x3 p30(it1 s14)x3 p49(it3 s1)x3 p64(it4 s0)x3 p89(it5 s9)x3 p97(it6 s1)x3 p126(it7 s14)x2 p53(it3 s5)x2 p65(it4 s1)x2
[2026-10-08T15:20:52Z] attribution: item 0x3df9f5 (71 reads) read from positions p8(it0 s8)x6 p40(it2 s8)x6 p106(it6 s10)x4 p26(it1 s10)x4 p56(it3 s8)x4 p89(it5 s9)x4 p90(it5 s10)x3 p0(it0 s0)x2 p6(it0 s6)x2 p3(it0 s3)x2 p41(it2 s9)x2 p86(it5 s6)x2
[2026-10-08T15:20:52Z] attribution: item 0x15645e (71 reads) read from positions p26(it1 s10)x4 p116(it7 s4)x4 p72(it4 s8)x3 p123(it7 s11)x2 p8(it0 s8)x2 p20(it1 s4)x2 p24(it1 s8)x2 p25(it1 s9)x2 p42(it2 s10)x2 p52(it3 s4)x2 p53(it3 s5)x2 p64(it4 s0)x2
[2026-10-08T15:20:52Z] attribution: item 0x79421d (70 reads) read from positions p17(it1 s1)x5 p94(it5 s14)x5 p80(it5 s0)x4 p118(it7 s6)x4 p30(it1 s14)x3 p41(it2 s9)x3 p49(it3 s1)x3 p78(it4 s14)x3 p1(it0 s1)x2 p81(it5 s1)x2 p84(it5 s4)x2 p110(it6 s14)x2
[2026-10-08T15:20:52Z] attribution: item 0x785ee5 (70 reads) read from positions p97(it6 s1)x4 p54(it3 s6)x4 p14(it0 s14)x3 p49(it3 s1)x3 p62(it3 s14)x3 p65(it4 s1)x3 p86(it5 s6)x3 p59(it3 s11)x2 p118(it7 s6)x2 p6(it0 s6)x2 p4(it0 s4)x2 p57(it3 s9)x2
[2026-10-08T15:20:52Z] hottest item p20-attack-f8 (0x371595): 76 reads (0.0000% of all); predicted source none
[2026-10-08T15:20:52Z] program p20-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:20:55Z] program p21-attack-f8: epoch seed fcfbea40c3b0b1e3301df7ba6b90a9c36248b2092ffb23d1e0f597672885808c era seed 4a36255eb732cd2db4f2c34b31ca5b0e405cd2f8442efa9edabcd2fc0347e198 id a730af697217c12b attempt 0 class mx8-era22a7f5e9+sh256x27+state+rw op mix load=16 mad=13 xor=10 rotl=8 rotr=6 sub=6 add=3 shfl=2; era stride mul 0x48825377 rot 22 interleave [3, 5, 6, 8] windows 3:2:2 4:0:0 6:1:0 16:2:3 17:0:0 29:2:0 31:2:2 33:1:0 34:1:0 35:2:0 36:2:0 38:2:2 47:0:0 52:0:0 60:2:3 61:0:0
[2026-10-08T15:20:55Z] window layer: site item windows (first, items) (0x800000,2^22) (0x0,2^24) (0x0,2^23) (0xc00000,2^22) (0x0,2^24) (0x0,2^22) (0x800000,2^22) (0x0,2^23) (0x0,2^23) (0x0,2^22) (0x0,2^22) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x0,2^24); expected reads per item by quarter 46.000 22.000 34.000 26.000 (flat uniform 32.000)
[2026-10-08T15:21:29Z] 131072 warps (4194304 nonces) interpreted in 34.2 s (2.087 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:21:29Z] p21-attack-f8 distinct lines per hash: n 4194304 min 1013 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8713
[2026-10-08T15:21:29Z] p21-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:21:29Z] p21-attack-f8 distinct lines per warp: n 131072 min 32573 p1 32605 median 32636 p99 32664 max 32688 mean 32636.0746
[2026-10-08T15:21:29Z] p21-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4629
[2026-10-08T15:21:29Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:21:29Z] positions: largest hi16-bucket (256 items) share 0.0110% at p90 (iteration 5, site 10, instr 36); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 3)
[2026-10-08T15:21:29Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p24 (site 8, instr 34); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:21:29Z] p21-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 87 (bin 32206) z_max +9.72 min 4 z_min -4.95 chi2/dof 3.66172 chi2_z +7709.17 top0.1% 0.21425% top0.5% 1.00280% top1% 1.93913%
[2026-10-08T15:21:29Z] p21-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.02490 largest 58 (item 0xe2b8ef) at +6.28 sigma smallest at -4.46; buckets64 chi2/dof 2.59097 largest 3380 (bucket 2027) at +8.04 sigma smallest at -6.62
[2026-10-08T15:21:31Z] p21-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 84 (bin 1985802) z_max +9.19 min 3 z_min -5.13 chi2/dof 3.62434 chi2_z +7600.91 top0.1% 0.21039% top0.5% 0.98792% top1% 1.91270%
[2026-10-08T15:21:31Z] p21-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99920 largest +6.17 sigma; buckets64 chi2/dof 1.00324 largest +4.97 smallest -4.93
[2026-10-08T15:21:31Z] 6-sigma p21-attack-f8 items buckets64 against the window density: largest bucket +8.04 sigma, smallest -6.62 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:21:31Z] hot-set p21-attack-f8 items (windowed control): f 0.1% S_f 0.21425% E_f(control) 0.21039% X_f +0.00386% X_f/f +0.0386 -> no hot set
[2026-10-08T15:21:31Z] hot-set p21-attack-f8 items (windowed control): f 0.5% S_f 1.00280% E_f(control) 0.98792% X_f +0.01487% X_f/f +0.0297 -> no hot set
[2026-10-08T15:21:31Z] hot-set p21-attack-f8 items (windowed control): f 1.0% S_f 1.93913% E_f(control) 1.91270% X_f +0.02643% X_f/f +0.0264 -> no hot set
[2026-10-08T15:21:31Z] hot-set p21-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:21:33Z] p21-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 1161920) z_max +5.83 min 7 z_min -4.42 chi2/dof 1.00022 chi2_z +0.64 top0.1% 0.16457% top0.5% 0.77429% top1% 1.50119%
[2026-10-08T15:21:33Z] hot-set p21-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.21425% E_f(control) 0.16457% X_f +0.04968% X_f/f +0.4968 -> no hot set
[2026-10-08T15:21:33Z] hot-set p21-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 1.00280% E_f(control) 0.77429% X_f +0.22850% X_f/f +0.4570 -> no hot set
[2026-10-08T15:21:33Z] hot-set p21-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.93913% E_f(control) 1.50119% X_f +0.43794% X_f/f +0.4379 -> no hot set
[2026-10-08T15:21:33Z] hot-set p21-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:21:33Z] ratio p21-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0183x / 1.0151x / 1.0138x (gate 1.2x at 0.1%: within); over the FLAT control 1.3019x / 1.2951x / 1.2917x
[2026-10-08T15:21:34Z] attribution: hot threshold count >= 66 marks 20414 items (0.1217% of items) holding 1390285 reads (0.2590% of reads); top 8 items t=0x007dce:87 t=0x217f19:85 t=0x02c3ff:84 t=0x331018:83 t=0x22b53a:83 t=0x21137d:83 t=0x1038a3:83 t=0x0153cf:83
[2026-10-08T15:21:49Z] attribution: reads on hot items per position (flat expectation 0.1217% of each position's 4194304 reads); the 24 largest: p10(it0 s10)=0.860% p58(it3 s10)=0.859% p122(it7 s10)=0.858% p26(it1 s10)=0.858% p106(it6 s10)=0.858% p74(it4 s10)=0.854% p42(it2 s10)=0.852% p90(it5 s10)=0.851% p69(it4 s5)=0.704% p37(it2 s5)=0.699% p101(it6 s5)=0.699% p57(it3 s9)=0.697% p41(it2 s9)=0.697% p73(it4 s9)=0.696% p5(it0 s5)=0.695% p121(it7 s9)=0.694% p117(it7 s5)=0.692% p25(it1 s9)=0.690% p105(it6 s9)=0.690% p53(it3 s5)=0.690% p9(it0 s9)=0.688% p85(it5 s5)=0.686% p89(it5 s9)=0.685% p21(it1 s5)=0.680%
[2026-10-08T15:21:49Z] attribution: hot reads by iteration it0=0.259% it1=0.258% it2=0.259% it3=0.259% it4=0.261% it5=0.257% it6=0.259% it7=0.259% ; by site s0=0.000% s1=0.175% s2=0.345% s3=0.000% s4=0.173% s5=0.693% s6=0.000% s7=0.345% s8=0.347% s9=0.692% s10=0.856% s11=0.000% s12=0.171% s13=0.172% s14=0.000% s15=0.172%
[2026-10-08T15:21:49Z] site 0 (instr 3, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0002% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 1 (instr 4, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1746% (flat expectation 0.1217%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:21:49Z] site 2 (instr 6, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3451% (flat expectation 0.1217%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:21:49Z] site 3 (instr 16, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 4 (instr 17, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1730% (flat expectation 0.1217%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:21:49Z] site 5 (instr 29, src r0, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6932% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 6 (instr 31, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0002% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 7 (instr 33, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3450% (flat expectation 0.1217%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:21:49Z] site 8 (instr 34, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3471% (flat expectation 0.1217%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:21:49Z] site 9 (instr 35, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6923% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 10 (instr 36, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.8562% (flat expectation 0.1217%); index entropy 13.949 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0095% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 11 (instr 38, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0002% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 12 (instr 47, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1715% (flat expectation 0.1217%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:21:49Z] site 13 (instr 52, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1723% (flat expectation 0.1217%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:21:49Z] site 14 (instr 60, src r4, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1217%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:21:49Z] site 15 (instr 61, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1723% (flat expectation 0.1217%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:21:49Z] attribution: item 0x007dce (87 reads) read from positions p42(it2 s10)x5 p90(it5 s10)x4 p101(it6 s5)x4 p117(it7 s5)x4 p41(it2 s9)x3 p69(it4 s5)x3 p87(it5 s7)x3 p5(it0 s5)x2 p10(it0 s10)x2 p13(it0 s13)x2 p18(it1 s2)x2 p26(it1 s10)x2
[2026-10-08T15:21:49Z] attribution: item 0x217f19 (85 reads) read from positions p122(it7 s10)x5 p26(it1 s10)x5 p105(it6 s9)x4 p10(it0 s10)x3 p23(it1 s7)x3 p88(it5 s8)x3 p90(it5 s10)x3 p106(it6 s10)x3 p116(it7 s4)x3 p73(it4 s9)x2 p121(it7 s9)x2 p58(it3 s10)x2
[2026-10-08T15:21:49Z] attribution: item 0x02c3ff (84 reads) read from positions p122(it7 s10)x4 p117(it7 s5)x4 p26(it1 s10)x3 p42(it2 s10)x3 p57(it3 s9)x3 p58(it3 s10)x3 p106(it6 s10)x3 p2(it0 s2)x3 p114(it7 s2)x3 p121(it7 s9)x3 p18(it1 s2)x2 p21(it1 s5)x2
[2026-10-08T15:21:49Z] attribution: item 0x331018 (83 reads) read from positions p5(it0 s5)x6 p26(it1 s10)x4 p114(it7 s2)x4 p58(it3 s10)x3 p88(it5 s8)x3 p90(it5 s10)x3 p119(it7 s7)x3 p1(it0 s1)x2 p121(it7 s9)x2 p37(it2 s5)x2 p39(it2 s7)x2 p50(it3 s2)x2
[2026-10-08T15:21:49Z] attribution: item 0x22b53a (83 reads) read from positions p122(it7 s10)x5 p42(it2 s10)x4 p89(it5 s9)x4 p5(it0 s5)x3 p25(it1 s9)x3 p26(it1 s10)x3 p53(it3 s5)x3 p74(it4 s10)x3 p90(it5 s10)x3 p101(it6 s5)x3 p117(it7 s5)x3 p15(it0 s15)x2
[2026-10-08T15:21:49Z] attribution: item 0x21137d (83 reads) read from positions p121(it7 s9)x5 p25(it1 s9)x4 p106(it6 s10)x4 p10(it0 s10)x3 p18(it1 s2)x3 p26(it1 s10)x3 p85(it5 s5)x3 p90(it5 s10)x3 p101(it6 s5)x3 p13(it0 s13)x2 p37(it2 s5)x2 p42(it2 s10)x2
[2026-10-08T15:21:49Z] attribution: item 0x1038a3 (83 reads) read from positions p122(it7 s10)x5 p90(it5 s10)x5 p9(it0 s9)x4 p42(it2 s10)x4 p57(it3 s9)x4 p74(it4 s10)x4 p10(it0 s10)x3 p18(it1 s2)x3 p82(it5 s2)x3 p23(it1 s7)x2 p25(it1 s9)x2 p39(it2 s7)x2
[2026-10-08T15:21:49Z] attribution: item 0x0153cf (83 reads) read from positions p117(it7 s5)x5 p21(it1 s5)x5 p72(it4 s8)x4 p90(it5 s10)x4 p13(it0 s13)x3 p36(it2 s4)x3 p58(it3 s10)x3 p73(it4 s9)x3 p9(it0 s9)x2 p10(it0 s10)x2 p25(it1 s9)x2 p26(it1 s10)x2
[2026-10-08T15:21:49Z] hottest item p21-attack-f8 (0x007dce): 87 reads (0.0000% of all); predicted source none
[2026-10-08T15:21:49Z] program p21-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
[2026-10-08T15:21:51Z] program p22-attack-f8: epoch seed 382043fe9821bc62ecc3e2ee38a05f73017845fb220c0b3895b3f77a11885dab era seed ef38feca26aca5746137933bf88c08439fbfb5b2686f79e62773769183b18c79 id 1faa9e3199040c5b attempt 0 class mx8-era534b22c8+sh256x27+state+rw op mix load=16 xor=8 add=7 mad=7 rotr=7 sub=6 mul=5 rotl=5 mulhi=2 shfl=1; era stride mul 0x1d32e3af rot 25 interleave [7, 9, 11, 15] windows 3:0:0 5:1:0 7:1:1 16:0:0 17:2:1 18:0:0 20:2:3 22:2:3 33:2:2 34:0:0 35:0:0 37:1:0 42:2:1 51:0:0 59:1:1 60:1:0
[2026-10-08T15:21:51Z] window layer: site item windows (first, items) (0x0,2^24) (0x0,2^23) (0x800000,2^23) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0xc00000,2^22) (0xc00000,2^22) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^23) (0x400000,2^22) (0x0,2^24) (0x800000,2^23) (0x0,2^23); expected reads per item by quarter 24.000 40.000 28.000 36.000 (flat uniform 32.000)
[2026-10-08T15:22:21Z] 131072 warps (4194304 nonces) interpreted in 29.9 s (1.825 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:22:21Z] p22-attack-f8 distinct lines per hash: n 4194304 min 992 p1 1016 median 1024 p99 1024 max 1024 mean 1023.3726
[2026-10-08T15:22:21Z] p22-attack-f8 distinct items per hash: n 4194304 min 124 p1 127 median 128 p99 128 max 128 mean 127.9371
[2026-10-08T15:22:21Z] p22-attack-f8 distinct lines per warp: n 131072 min 32525 p1 32578 median 32621 p99 32656 max 32678 mean 32620.4310
[2026-10-08T15:22:21Z] p22-attack-f8 distinct items per warp: n 131072 min 4083 p1 4089 median 4094 p99 4096 max 4096 mean 4093.4894
[2026-10-08T15:22:21Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:22:21Z] positions: largest hi16-bucket (256 items) share 0.0079% at p108 (iteration 6, site 12, instr 42); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 3)
[2026-10-08T15:22:21Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p20 (site 4, instr 17); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:22:21Z] p22-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 76 (bin 6516684) z_max +7.78 min 4 z_min -4.95 chi2/dof 2.16880 chi2_z +3385.21 top0.1% 0.18693% top0.5% 0.87606% top1% 1.69611%
[2026-10-08T15:22:22Z] p22-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.04607 largest 55 (item 0xc7d0a) at +6.33 sigma smallest at -4.35; buckets64 chi2/dof 3.76366 largest 1850 (bucket 22288) at +8.01 sigma smallest at -6.36
[2026-10-08T15:22:24Z] p22-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 78 (bin 7468672) z_max +8.13 min 4 z_min -4.95 chi2/dof 2.24999 chi2_z +3620.34 top0.1% 0.18806% top0.5% 0.88125% top1% 1.70626%
[2026-10-08T15:22:24Z] p22-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99990 largest +6.43 sigma; buckets64 chi2/dof 1.00168 largest +4.46 smallest -4.65
[2026-10-08T15:22:24Z] 6-sigma p22-attack-f8 items buckets64 against the window density: largest bucket +8.01 sigma, smallest -6.36 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:22:24Z] hot-set p22-attack-f8 items (windowed control): f 0.1% S_f 0.18693% E_f(control) 0.18806% X_f -0.00112% X_f/f -0.0112 -> no hot set
[2026-10-08T15:22:24Z] hot-set p22-attack-f8 items (windowed control): f 0.5% S_f 0.87606% E_f(control) 0.88125% X_f -0.00518% X_f/f -0.0104 -> no hot set
[2026-10-08T15:22:24Z] hot-set p22-attack-f8 items (windowed control): f 1.0% S_f 1.69611% E_f(control) 1.70626% X_f -0.01015% X_f/f -0.0102 -> no hot set
[2026-10-08T15:22:24Z] hot-set p22-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:22:26Z] p22-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 11627338) z_max +5.83 min 7 z_min -4.42 chi2/dof 0.99974 chi2_z -0.75 top0.1% 0.16469% top0.5% 0.77435% top1% 1.50119%
[2026-10-08T15:22:26Z] hot-set p22-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.18693% E_f(control) 0.16469% X_f +0.02224% X_f/f +0.2224 -> no hot set
[2026-10-08T15:22:26Z] hot-set p22-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.87606% E_f(control) 0.77435% X_f +0.10172% X_f/f +0.2034 -> no hot set
[2026-10-08T15:22:26Z] hot-set p22-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.69611% E_f(control) 1.50119% X_f +0.19493% X_f/f +0.1949 -> no hot set
[2026-10-08T15:22:26Z] hot-set p22-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:22:26Z] ratio p22-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9940x / 0.9941x / 0.9941x (gate 1.2x at 0.1%: within); over the FLAT control 1.1351x / 1.1314x / 1.1298x
[2026-10-08T15:22:27Z] attribution: hot threshold count >= 58 marks 17591 items (0.1049% of items) holding 1050789 reads (0.1957% of reads); top 8 items t=0x636fcc:76 t=0x4fe751:74 t=0x4b11b7:74 t=0x48dfbc:74 t=0x532bb3:73 t=0xe41c8a:72 t=0x5765f4:72 t=0x511550:72
[2026-10-08T15:22:41Z] attribution: reads on hot items per position (flat expectation 0.1049% of each position's 4194304 reads); the 24 largest: p76(it4 s12)=0.568% p20(it1 s4)=0.566% p44(it2 s12)=0.564% p100(it6 s4)=0.563% p28(it1 s12)=0.562% p60(it3 s12)=0.561% p52(it3 s4)=0.560% p84(it5 s4)=0.560% p92(it5 s12)=0.559% p68(it4 s4)=0.559% p116(it7 s4)=0.555% p12(it0 s12)=0.555% p36(it2 s4)=0.555% p4(it0 s4)=0.553% p108(it6 s12)=0.550% p124(it7 s12)=0.550% p27(it1 s11)=0.284% p43(it2 s11)=0.283% p75(it4 s11)=0.282% p127(it7 s15)=0.282% p123(it7 s11)=0.281% p111(it6 s15)=0.281% p91(it5 s11)=0.280% p11(it0 s11)=0.280%
[2026-10-08T15:22:41Z] attribution: hot reads by iteration it0=0.195% it1=0.196% it2=0.196% it3=0.195% it4=0.197% it5=0.196% it6=0.195% it7=0.195% ; by site s0=0.159% s1=0.272% s2=0.038% s3=0.160% s4=0.559% s5=0.159% s6=0.076% s7=0.076% s8=0.000% s9=0.159% s10=0.159% s11=0.280% s12=0.559% s13=0.159% s14=0.038% s15=0.278%
[2026-10-08T15:22:41Z] site 0 (instr 3, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1590% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 1 (instr 5, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2719% (flat expectation 0.1049%); index entropy 14.900 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0052% (window expectation 0.0031%)
[2026-10-08T15:22:41Z] site 2 (instr 7, src r6, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0383% (flat expectation 0.1049%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:22:41Z] site 3 (instr 16, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1599% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 4 (instr 17, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5589% (flat expectation 0.1049%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:22:41Z] site 5 (instr 18, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1586% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 6 (instr 20, src r1, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0759% (flat expectation 0.1049%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:22:41Z] site 7 (instr 22, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0758% (flat expectation 0.1049%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:22:41Z] site 8 (instr 33, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1049%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:22:41Z] site 9 (instr 34, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1592% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 10 (instr 35, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1594% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 11 (instr 37, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2803% (flat expectation 0.1049%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:22:41Z] site 12 (instr 42, src r7, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5586% (flat expectation 0.1049%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:22:41Z] site 13 (instr 51, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1592% (flat expectation 0.1049%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:22:41Z] site 14 (instr 59, src r4, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0382% (flat expectation 0.1049%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:22:41Z] site 15 (instr 60, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2784% (flat expectation 0.1049%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:22:41Z] attribution: item 0x636fcc (76 reads) read from positions p36(it2 s4)x4 p4(it0 s4)x4 p108(it6 s12)x4 p111(it6 s15)x3 p11(it0 s11)x3 p28(it1 s12)x3 p44(it2 s12)x3 p92(it5 s12)x3 p95(it5 s15)x2 p12(it0 s12)x2 p52(it3 s4)x2 p60(it3 s12)x2
[2026-10-08T15:22:41Z] attribution: item 0x4fe751 (74 reads) read from positions p68(it4 s4)x4 p11(it0 s11)x3 p12(it0 s12)x3 p92(it5 s12)x3 p100(it6 s4)x3 p124(it7 s12)x2 p13(it0 s13)x2 p17(it1 s1)x2 p20(it1 s4)x2 p36(it2 s4)x2 p44(it2 s12)x2 p60(it3 s12)x2
[2026-10-08T15:22:41Z] attribution: item 0x4b11b7 (74 reads) read from positions p28(it1 s12)x5 p84(it5 s4)x4 p36(it2 s4)x4 p76(it4 s12)x4 p68(it4 s4)x3 p1(it0 s1)x3 p73(it4 s9)x2 p29(it1 s13)x2 p15(it0 s15)x2 p59(it3 s11)x2 p60(it3 s12)x2 p65(it4 s1)x2
[2026-10-08T15:22:41Z] attribution: item 0x48dfbc (74 reads) read from positions p83(it5 s3)x5 p68(it4 s4)x4 p76(it4 s12)x4 p4(it0 s4)x3 p60(it3 s12)x3 p28(it1 s12)x2 p65(it4 s1)x2 p75(it4 s11)x2 p77(it4 s13)x2 p81(it5 s1)x2 p84(it5 s4)x2 p92(it5 s12)x2
[2026-10-08T15:22:41Z] attribution: item 0x532bb3 (73 reads) read from positions p68(it4 s4)x4 p124(it7 s12)x4 p12(it0 s12)x3 p20(it1 s4)x3 p28(it1 s12)x3 p36(it2 s4)x3 p75(it4 s11)x3 p100(it6 s4)x3 p116(it7 s4)x3 p4(it0 s4)x2 p27(it1 s11)x2 p31(it1 s15)x2
[2026-10-08T15:22:41Z] attribution: item 0xe41c8a (72 reads) read from positions p6(it0 s6)x6 p118(it7 s6)x4 p87(it5 s7)x4 p7(it0 s7)x3 p50(it3 s2)x3 p55(it3 s7)x3 p103(it6 s7)x2 p23(it1 s7)x2 p30(it1 s14)x2 p34(it2 s2)x2 p35(it2 s3)x2 p45(it2 s13)x2
[2026-10-08T15:22:41Z] attribution: item 0x5765f4 (72 reads) read from positions p28(it1 s12)x5 p31(it1 s15)x3 p49(it3 s1)x3 p52(it3 s4)x3 p68(it4 s4)x3 p97(it6 s1)x3 p127(it7 s15)x3 p4(it0 s4)x2 p10(it0 s10)x2 p17(it1 s1)x2 p21(it1 s5)x2 p36(it2 s4)x2
[2026-10-08T15:22:41Z] attribution: item 0x511550 (72 reads) read from positions p60(it3 s12)x6 p124(it7 s12)x3 p16(it1 s0)x3 p79(it4 s15)x3 p10(it0 s10)x2 p26(it1 s10)x2 p28(it1 s12)x2 p35(it2 s3)x2 p36(it2 s4)x2 p52(it3 s4)x2 p65(it4 s1)x2 p80(it5 s0)x2
[2026-10-08T15:22:41Z] hottest item p22-attack-f8 (0x636fcc): 76 reads (0.0000% of all); predicted source none
[2026-10-08T15:22:41Z] program p22-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
[2026-10-08T15:22:44Z] program p23-attack-f8: epoch seed 01aa1485fcb5d59223ca40602086e618277decf440188c5a55898f635f7be34f era seed 00951c99e7ef952fd52611b8de7c07cc705cdec9e129a31a19d696c99909f052 id cb2341129841e67d attempt 0 class mx8-era54807265+sh256x27+state+rw op mix load=16 xor=13 mad=9 rotl=8 add=7 rotr=4 sub=3 mul=2 shfl=2; era stride mul 0xd88595c9 rot 26 interleave [0, 1, 4, 14] windows 9:2:1 14:2:2 16:2:0 18:0:0 19:1:0 21:2:0 22:1:1 24:1:1 34:2:1 40:1:1 43:1:1 44:0:0 49:1:0 55:2:1 62:1:0 63:2:1
[2026-10-08T15:22:44Z] window layer: site item windows (first, items) (0x400000,2^22) (0x800000,2^22) (0x0,2^22) (0x0,2^24) (0x0,2^23) (0x0,2^22) (0x800000,2^23) (0x800000,2^23) (0x400000,2^22) (0x800000,2^23) (0x800000,2^23) (0x0,2^24) (0x0,2^23) (0x400000,2^22) (0x0,2^23) (0x400000,2^22); expected reads per item by quarter 32.000 48.000 28.000 20.000 (flat uniform 32.000)
[2026-10-08T15:23:14Z] 131072 warps (4194304 nonces) interpreted in 30.3 s (1.847 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:23:14Z] p23-attack-f8 distinct lines per hash: n 4194304 min 1014 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8709
[2026-10-08T15:23:14Z] p23-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:23:14Z] p23-attack-f8 distinct lines per warp: n 131072 min 32564 p1 32604 median 32636 p99 32664 max 32689 mean 32635.9784
[2026-10-08T15:23:14Z] p23-attack-f8 distinct items per warp: n 131072 min 4089 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4488
[2026-10-08T15:23:14Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:23:14Z] positions: largest hi16-bucket (256 items) share 0.0081% at p49 (iteration 3, site 1, instr 14); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 9)
[2026-10-08T15:23:14Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p93 (site 13, instr 55); saturated sources at one position 0 of 2048 at p0 (site 0, instr 9); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:23:14Z] p23-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 91 (bin 4653593) z_max +10.43 min 2 z_min -5.30 chi2/dof 4.24900 chi2_z +9410.10 top0.1% 0.21801% top0.5% 1.02425% top1% 1.98551%
[2026-10-08T15:23:15Z] p23-attack-f8 item histogram against the window density: full 2^24 chi2/dof 0.99971 largest 61 (item 0x9d9237) at +6.24 sigma smallest at -4.62; buckets64 chi2/dof 0.99341 largest 1449 (bucket 223780) at +4.72 sigma smallest at -4.73
[2026-10-08T15:23:16Z] p23-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 91 (bin 8344575) z_max +10.43 min 2 z_min -5.30 chi2/dof 4.25035 chi2_z +9414.03 top0.1% 0.21799% top0.5% 1.02418% top1% 1.98537%
[2026-10-08T15:23:16Z] p23-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99985 largest +6.26 sigma; buckets64 chi2/dof 1.00123 largest +4.40 smallest -4.35
[2026-10-08T15:23:16Z] 6-sigma p23-attack-f8 items buckets64 against the window density: largest bucket +4.72 sigma, smallest -4.73 -> within 6 sigma
[2026-10-08T15:23:16Z] hot-set p23-attack-f8 items (windowed control): f 0.1% S_f 0.21801% E_f(control) 0.21799% X_f +0.00002% X_f/f +0.0002 -> no hot set
[2026-10-08T15:23:16Z] hot-set p23-attack-f8 items (windowed control): f 0.5% S_f 1.02425% E_f(control) 1.02418% X_f +0.00007% X_f/f +0.0001 -> no hot set
[2026-10-08T15:23:16Z] hot-set p23-attack-f8 items (windowed control): f 1.0% S_f 1.98551% E_f(control) 1.98537% X_f +0.00014% X_f/f +0.0001 -> no hot set
[2026-10-08T15:23:16Z] hot-set p23-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:23:18Z] p23-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 66 (bin 6982480) z_max +6.01 min 7 z_min -4.42 chi2/dof 0.99986 chi2_z -0.40 top0.1% 0.16456% top0.5% 0.77432% top1% 1.50116%
[2026-10-08T15:23:18Z] hot-set p23-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.21801% E_f(control) 0.16456% X_f +0.05345% X_f/f +0.5345 -> no hot set
[2026-10-08T15:23:18Z] hot-set p23-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 1.02425% E_f(control) 0.77432% X_f +0.24993% X_f/f +0.4999 -> no hot set
[2026-10-08T15:23:18Z] hot-set p23-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.98551% E_f(control) 1.50116% X_f +0.48434% X_f/f +0.4843 -> no hot set
[2026-10-08T15:23:18Z] hot-set p23-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:23:18Z] ratio p23-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0001x / 1.0001x / 1.0001x (gate 1.2x at 0.1%: within); over the FLAT control 1.3248x / 1.3228x / 1.3226x
[2026-10-08T15:23:19Z] attribution: hot threshold count >= 67 marks 22672 items (0.1351% of items) holding 1565390 reads (0.2916% of reads); top 8 items t=0x6149d0:91 t=0x470219:91 t=0x64e303:87 t=0x7e53b8:85 t=0x469319:85 t=0x7fe875:84 t=0x736597:84 t=0x5b50fa:84
[2026-10-08T15:23:32Z] attribution: reads on hot items per position (flat expectation 0.1351% of each position's 4194304 reads); the 24 largest: p109(it6 s13)=0.784% p31(it1 s15)=0.784% p88(it5 s8)=0.783% p127(it7 s15)=0.781% p111(it6 s15)=0.781% p125(it7 s13)=0.780% p24(it1 s8)=0.780% p29(it1 s13)=0.780% p40(it2 s8)=0.780% p48(it3 s0)=0.780% p13(it0 s13)=0.780% p120(it7 s8)=0.779% p63(it3 s15)=0.779% p0(it0 s0)=0.779% p96(it6 s0)=0.778% p93(it5 s13)=0.777% p64(it4 s0)=0.777% p112(it7 s0)=0.776% p16(it1 s0)=0.776% p79(it4 s15)=0.776% p32(it2 s0)=0.776% p45(it2 s13)=0.776% p61(it3 s13)=0.775% p95(it5 s15)=0.775%
[2026-10-08T15:23:32Z] attribution: hot reads by iteration it0=0.291% it1=0.293% it2=0.291% it3=0.291% it4=0.291% it5=0.292% it6=0.292% it7=0.291% ; by site s0=0.777% s1=0.000% s2=0.000% s3=0.195% s4=0.389% s5=0.000% s6=0.000% s7=0.000% s8=0.777% s9=0.000% s10=0.000% s11=0.194% s12=0.391% s13=0.778% s14=0.387% s15=0.777%
[2026-10-08T15:23:32Z] site 0 (instr 9, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.7768% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 1 (instr 14, src r6, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 2 (instr 16, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 3 (instr 18, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1953% (flat expectation 0.1351%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:23:32Z] site 4 (instr 19, src r0, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3886% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 5 (instr 21, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 6 (instr 22, src r2, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 7 (instr 24, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 8 (instr 34, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.7769% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 9 (instr 40, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 10 (instr 43, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 11 (instr 44, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1944% (flat expectation 0.1351%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:23:32Z] site 12 (instr 49, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3909% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 13 (instr 55, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.7783% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] site 14 (instr 62, src r4, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3871% (flat expectation 0.1351%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:23:32Z] site 15 (instr 63, src r1, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.7769% (flat expectation 0.1351%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:23:32Z] attribution: item 0x6149d0 (91 reads) read from positions p15(it0 s15)x6 p44(it2 s12)x5 p45(it2 s13)x4 p95(it5 s15)x4 p0(it0 s0)x3 p16(it1 s0)x3 p24(it1 s8)x3 p43(it2 s11)x3 p96(it6 s0)x3 p104(it6 s8)x3 p109(it6 s13)x3 p111(it6 s15)x2
[2026-10-08T15:23:32Z] attribution: item 0x470219 (91 reads) read from positions p125(it7 s13)x6 p31(it1 s15)x4 p95(it5 s15)x4 p96(it6 s0)x4 p120(it7 s8)x4 p13(it0 s13)x3 p27(it1 s11)x3 p45(it2 s13)x3 p88(it5 s8)x3 p104(it6 s8)x3 p109(it6 s13)x3 p111(it6 s15)x3
[2026-10-08T15:23:32Z] attribution: item 0x64e303 (87 reads) read from positions p45(it2 s13)x4 p52(it3 s4)x4 p110(it6 s14)x3 p64(it4 s0)x3 p93(it5 s13)x3 p96(it6 s0)x3 p108(it6 s12)x2 p127(it7 s15)x2 p31(it1 s15)x2 p36(it2 s4)x2 p40(it2 s8)x2 p11(it0 s11)x2
[2026-10-08T15:23:32Z] attribution: item 0x7e53b8 (85 reads) read from positions p45(it2 s13)x7 p111(it6 s15)x6 p32(it2 s0)x5 p100(it6 s4)x5 p64(it4 s0)x4 p15(it0 s15)x3 p63(it3 s15)x3 p72(it4 s8)x3 p109(it6 s13)x3 p112(it7 s0)x3 p4(it0 s4)x2 p35(it2 s3)x2
[2026-10-08T15:23:32Z] attribution: item 0x469319 (85 reads) read from positions p32(it2 s0)x4 p47(it2 s15)x4 p127(it7 s15)x3 p8(it0 s8)x3 p15(it0 s15)x3 p29(it1 s13)x3 p46(it2 s14)x3 p0(it0 s0)x3 p95(it5 s15)x3 p96(it6 s0)x3 p111(it6 s15)x3 p93(it5 s13)x2
[2026-10-08T15:23:32Z] attribution: item 0x7fe875 (84 reads) read from positions p93(it5 s13)x5 p127(it7 s15)x4 p45(it2 s13)x4 p63(it3 s15)x4 p12(it0 s12)x3 p77(it4 s13)x3 p79(it4 s15)x3 p84(it5 s4)x3 p95(it5 s15)x3 p104(it6 s8)x3 p0(it0 s0)x2 p13(it0 s13)x2
[2026-10-08T15:23:32Z] attribution: item 0x736597 (84 reads) read from positions p125(it7 s13)x5 p111(it6 s15)x4 p40(it2 s8)x3 p48(it3 s0)x3 p63(it3 s15)x3 p72(it4 s8)x3 p80(it5 s0)x3 p120(it7 s8)x3 p78(it4 s14)x2 p35(it2 s3)x2 p43(it2 s11)x2 p45(it2 s13)x2
[2026-10-08T15:23:32Z] attribution: item 0x5b50fa (84 reads) read from positions p13(it0 s13)x4 p31(it1 s15)x4 p20(it1 s4)x3 p46(it2 s14)x3 p47(it2 s15)x3 p77(it4 s13)x3 p78(it4 s14)x3 p127(it7 s15)x3 p4(it0 s4)x2 p8(it0 s8)x2 p12(it0 s12)x2 p15(it0 s15)x2
[2026-10-08T15:23:32Z] hottest item p23-attack-f8 (0x6149d0): 91 reads (0.0000% of all); predicted source none
[2026-10-08T15:23:32Z] program p23-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:23:34Z] program p24-attack-f8: epoch seed 2496c474e54e30c09afd030f09d214cecb776e99e3bb9f2d3131d477d5e5d592 era seed f2faf296591ad403fe136f396bc2f9df54c95a466a86325e46e42dbb9c083e4b id fd5d5b3afe0515ee attempt 1 class mx8-era96203cac+sh256x27+state+rw op mix load=16 add=12 rotl=9 xor=9 mad=8 rotr=3 sub=3 mul=2 mulhi=1 shfl=1; era stride mul 0x44fd49d7 rot 3 interleave [0, 4, 7, 9] windows 5:0:0 8:1:0 11:2:3 13:1:0 16:1:0 20:2:2 23:1:0 24:0:0 36:0:0 38:0:0 39:2:1 42:0:0 43:1:1 46:2:2 49:2:3 62:0:0
[2026-10-08T15:23:34Z] window layer: site item windows (first, items) (0x0,2^24) (0x0,2^23) (0xc00000,2^22) (0x0,2^23) (0x0,2^23) (0x800000,2^22) (0x0,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x800000,2^23) (0x800000,2^22) (0xc00000,2^22) (0x0,2^24); expected reads per item by quarter 28.000 36.000 32.000 32.000 (flat uniform 32.000)
[2026-10-08T15:24:07Z] 131072 warps (4194304 nonces) interpreted in 32.6 s (1.991 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:24:07Z] p24-attack-f8 distinct lines per hash: n 4194304 min 992 p1 1016 median 1024 p99 1024 max 1024 mean 1023.3476
[2026-10-08T15:24:07Z] p24-attack-f8 distinct items per hash: n 4194304 min 124 p1 127 median 128 p99 128 max 128 mean 127.9340
[2026-10-08T15:24:07Z] p24-attack-f8 distinct lines per warp: n 131072 min 32533 p1 32577 median 32620 p99 32656 max 32682 mean 32619.6995
[2026-10-08T15:24:07Z] p24-attack-f8 distinct items per warp: n 131072 min 4084 p1 4089 median 4094 p99 4096 max 4096 mean 4093.4009
[2026-10-08T15:24:07Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:24:07Z] positions: largest hi16-bucket (256 items) share 0.0078% at p125 (iteration 7, site 13, instr 46); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 5)
[2026-10-08T15:24:07Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p2 (site 2, instr 11); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:24:07Z] p24-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 72 (bin 8234647) z_max +7.07 min 5 z_min -4.77 chi2/dof 1.25104 chi2_z +727.08 top0.1% 0.17277% top0.5% 0.80940% top1% 1.56643%
[2026-10-08T15:24:07Z] p24-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00119 largest 61 (item 0x3a13a2) at +6.24 sigma smallest at -4.60; buckets64 chi2/dof 1.00319 largest 2256 (bucket 144572) at +4.60 sigma smallest at -4.02
[2026-10-08T15:24:11Z] p24-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 70 (bin 5361141) z_max +6.72 min 6 z_min -4.60 chi2/dof 1.25001 chi2_z +724.09 top0.1% 0.17269% top0.5% 0.80911% top1% 1.56611%
[2026-10-08T15:24:11Z] p24-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00014 largest +5.83 sigma; buckets64 chi2/dof 0.99692 largest +4.70 smallest -4.56
[2026-10-08T15:24:11Z] 6-sigma p24-attack-f8 items buckets64 against the window density: largest bucket +4.60 sigma, smallest -4.02 -> within 6 sigma
[2026-10-08T15:24:11Z] hot-set p24-attack-f8 items (windowed control): f 0.1% S_f 0.17277% E_f(control) 0.17269% X_f +0.00009% X_f/f +0.0009 -> no hot set
[2026-10-08T15:24:11Z] hot-set p24-attack-f8 items (windowed control): f 0.5% S_f 0.80940% E_f(control) 0.80911% X_f +0.00029% X_f/f +0.0006 -> no hot set
[2026-10-08T15:24:11Z] hot-set p24-attack-f8 items (windowed control): f 1.0% S_f 1.56643% E_f(control) 1.56611% X_f +0.00032% X_f/f +0.0003 -> no hot set
[2026-10-08T15:24:11Z] hot-set p24-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:24:14Z] p24-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 8844360) z_max +5.83 min 7 z_min -4.42 chi2/dof 1.00036 chi2_z +1.03 top0.1% 0.16462% top0.5% 0.77451% top1% 1.50146%
[2026-10-08T15:24:14Z] hot-set p24-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17277% E_f(control) 0.16462% X_f +0.00816% X_f/f +0.0816 -> no hot set
[2026-10-08T15:24:14Z] hot-set p24-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.80940% E_f(control) 0.77451% X_f +0.03489% X_f/f +0.0698 -> no hot set
[2026-10-08T15:24:14Z] hot-set p24-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.56643% E_f(control) 1.50146% X_f +0.06497% X_f/f +0.0650 -> no hot set
[2026-10-08T15:24:14Z] hot-set p24-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:24:14Z] ratio p24-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0005x / 1.0004x / 1.0002x (gate 1.2x at 0.1%: within); over the FLAT control 1.0495x / 1.0451x / 1.0433x
[2026-10-08T15:24:15Z] attribution: hot threshold count >= 53 marks 23287 items (0.1388% of items) holding 1272595 reads (0.2370% of reads); top 8 items t=0x7da697:72 t=0x4ec65f:70 t=0x4744df:69 t=0x7d23e2:68 t=0x7b683d:67 t=0x6a0007:67 t=0x69b4e6:67 t=0x66a2d7:67
[2026-10-08T15:24:32Z] attribution: reads on hot items per position (flat expectation 0.1388% of each position's 4194304 reads); the 24 largest: p10(it0 s10)=0.723% p90(it5 s10)=0.722% p42(it2 s10)=0.717% p106(it6 s10)=0.715% p58(it3 s10)=0.715% p26(it1 s10)=0.713% p122(it7 s10)=0.709% p74(it4 s10)=0.709% p35(it2 s3)=0.366% p86(it5 s6)=0.364% p36(it2 s4)=0.363% p22(it1 s6)=0.363% p54(it3 s6)=0.362% p4(it0 s4)=0.362% p102(it6 s6)=0.362% p38(it2 s6)=0.362% p6(it0 s6)=0.361% p83(it5 s3)=0.361% p99(it6 s3)=0.361% p97(it6 s1)=0.361% p17(it1 s1)=0.360% p118(it7 s6)=0.360% p115(it7 s3)=0.359% p1(it0 s1)=0.359%
[2026-10-08T15:24:32Z] attribution: hot reads by iteration it0=0.238% it1=0.237% it2=0.238% it3=0.237% it4=0.236% it5=0.237% it6=0.237% it7=0.236% ; by site s0=0.214% s1=0.358% s2=0.073% s3=0.359% s4=0.359% s5=0.069% s6=0.362% s7=0.214% s8=0.216% s9=0.216% s10=0.715% s11=0.214% s12=0.070% s13=0.069% s14=0.072% s15=0.215%
[2026-10-08T15:24:32Z] site 0 (instr 5, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2143% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] site 1 (instr 8, src r0, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3584% (flat expectation 0.1388%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:24:32Z] site 2 (instr 11, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0726% (flat expectation 0.1388%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:24:32Z] site 3 (instr 13, src r3, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3585% (flat expectation 0.1388%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:24:32Z] site 4 (instr 16, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3588% (flat expectation 0.1388%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:24:32Z] site 5 (instr 20, src r3, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0687% (flat expectation 0.1388%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:24:32Z] site 6 (instr 23, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3615% (flat expectation 0.1388%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:24:32Z] site 7 (instr 24, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2140% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] site 8 (instr 36, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2159% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] site 9 (instr 38, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2155% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] site 10 (instr 39, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.7153% (flat expectation 0.1388%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:24:32Z] site 11 (instr 42, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2135% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] site 12 (instr 43, src r2, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0704% (flat expectation 0.1388%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:24:32Z] site 13 (instr 46, src r4, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0686% (flat expectation 0.1388%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:24:32Z] site 14 (instr 49, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0720% (flat expectation 0.1388%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:24:32Z] site 15 (instr 62, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2146% (flat expectation 0.1388%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:24:32Z] attribution: item 0x7da697 (72 reads) read from positions p122(it7 s10)x4 p51(it3 s3)x3 p54(it3 s6)x3 p81(it5 s1)x3 p84(it5 s4)x3 p4(it0 s4)x2 p8(it0 s8)x2 p22(it1 s6)x2 p42(it2 s10)x2 p52(it3 s4)x2 p58(it3 s10)x2 p70(it4 s6)x2
[2026-10-08T15:24:32Z] attribution: item 0x4ec65f (70 reads) read from positions p58(it3 s10)x6 p42(it2 s10)x3 p51(it3 s3)x3 p52(it3 s4)x3 p115(it7 s3)x3 p6(it0 s6)x2 p25(it1 s9)x2 p33(it2 s1)x2 p41(it2 s9)x2 p47(it2 s15)x2 p54(it3 s6)x2 p71(it4 s7)x2
[2026-10-08T15:24:32Z] attribution: item 0x4744df (69 reads) read from positions p10(it0 s10)x3 p22(it1 s6)x3 p42(it2 s10)x3 p54(it3 s6)x3 p68(it4 s4)x3 p113(it7 s1)x3 p3(it0 s3)x2 p20(it1 s4)x2 p26(it1 s10)x2 p38(it2 s6)x2 p57(it3 s9)x2 p65(it4 s1)x2
[2026-10-08T15:24:32Z] attribution: item 0x7d23e2 (68 reads) read from positions p42(it2 s10)x4 p118(it7 s6)x3 p10(it0 s10)x3 p54(it3 s6)x3 p4(it0 s4)x2 p8(it0 s8)x2 p16(it1 s0)x2 p17(it1 s1)x2 p47(it2 s15)x2 p49(it3 s1)x2 p58(it3 s10)x2 p70(it4 s6)x2
[2026-10-08T15:24:32Z] attribution: item 0x7b683d (67 reads) read from positions p90(it5 s10)x7 p38(it2 s6)x6 p10(it0 s10)x4 p26(it1 s10)x4 p80(it5 s0)x3 p84(it5 s4)x3 p116(it7 s4)x2 p39(it2 s7)x2 p56(it3 s8)x2 p58(it3 s10)x2 p99(it6 s3)x2 p70(it4 s6)x1
[2026-10-08T15:24:32Z] attribution: item 0x6a0007 (67 reads) read from positions p122(it7 s10)x3 p25(it1 s9)x3 p49(it3 s1)x3 p67(it4 s3)x3 p102(it6 s6)x3 p10(it0 s10)x2 p16(it1 s0)x2 p26(it1 s10)x2 p35(it2 s3)x2 p42(it2 s10)x2 p68(it4 s4)x2 p70(it4 s6)x2
[2026-10-08T15:24:32Z] attribution: item 0x69b4e6 (67 reads) read from positions p42(it2 s10)x4 p33(it2 s1)x3 p67(it4 s3)x3 p86(it5 s6)x3 p102(it6 s6)x3 p122(it7 s10)x2 p15(it0 s15)x2 p25(it1 s9)x2 p26(it1 s10)x2 p36(it2 s4)x2 p65(it4 s1)x2 p73(it4 s9)x2
[2026-10-08T15:24:32Z] attribution: item 0x66a2d7 (67 reads) read from positions p6(it0 s6)x4 p116(it7 s4)x4 p10(it0 s10)x3 p36(it2 s4)x3 p74(it4 s10)x3 p97(it6 s1)x3 p113(it7 s1)x3 p122(it7 s10)x3 p1(it0 s1)x2 p35(it2 s3)x2 p42(it2 s10)x2 p52(it3 s4)x2
[2026-10-08T15:24:32Z] hottest item p24-attack-f8 (0x7da697): 72 reads (0.0000% of all); predicted source none
[2026-10-08T15:24:32Z] program p24-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:24:35Z] program p25-attack-f8: epoch seed 63fd22da35b053e5be826537be700df1e3e18c59a4ff8fe804786176a9c9ac6d era seed 8a38fd4f1e79da8d3fa48b06af7a89d0da76aaaf0875d4ee294acceb6dd59af3 id 8b83f49ba0b49e23 attempt 0 class mx8-era42c17321+sh256x27+state+rw op mix load=16 add=9 xor=9 mad=7 rotl=7 rotr=6 shfl=4 sub=3 mulhi=2 mul=1; era stride mul 0xc619dd6b rot 14 interleave [2, 6, 8, 15] windows 5:0:0 11:1:0 14:0:0 17:0:0 28:2:3 29:0:0 34:0:0 35:2:1 37:0:0 40:1:0 42:0:0 43:2:0 48:2:1 51:2:3 55:0:0 62:1:0
[2026-10-08T15:24:35Z] window layer: site item windows (first, items) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x0,2^24) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x0,2^22) (0x400000,2^22) (0xc00000,2^22) (0x0,2^24) (0x0,2^23); expected reads per item by quarter 36.000 44.000 16.000 32.000 (flat uniform 32.000)
[2026-10-08T15:25:14Z] 131072 warps (4194304 nonces) interpreted in 39.2 s (2.395 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:25:14Z] p25-attack-f8 distinct lines per hash: n 4194304 min 1014 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8705
[2026-10-08T15:25:14Z] p25-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:25:14Z] p25-attack-f8 distinct lines per warp: n 131072 min 32573 p1 32605 median 32636 p99 32664 max 32684 mean 32635.9177
[2026-10-08T15:25:14Z] p25-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4479
[2026-10-08T15:25:14Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:25:14Z] positions: largest hi16-bucket (256 items) share 0.0082% at p125 (iteration 7, site 13, instr 51); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p111 (site 15, instr 62)
[2026-10-08T15:25:14Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 1 of 2048 at p0 (site 0, instr 5); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:25:14Z] p25-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 86 (bin 4378477) z_max +9.55 min 1 z_min -5.48 chi2/dof 4.25071 chi2_z +9415.06 top0.1% 0.20288% top0.5% 0.95099% top1% 1.84118%
[2026-10-08T15:25:14Z] p25-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00034 largest 86 (item 0x42cf6d) at +6.33 sigma smallest at -4.37; buckets64 chi2/dof 1.00059 largest 1169 (bucket 137660) at +4.53 sigma smallest at -4.55
[2026-10-08T15:25:17Z] p25-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 84 (bin 6911207) z_max +9.19 min 1 z_min -5.48 chi2/dof 4.25074 chi2_z +9415.15 top0.1% 0.20287% top0.5% 0.95099% top1% 1.84119%
[2026-10-08T15:25:17Z] p25-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99945 largest +6.03 sigma; buckets64 chi2/dof 1.00250 largest +4.84 smallest -4.27
[2026-10-08T15:25:17Z] 6-sigma p25-attack-f8 items buckets64 against the window density: largest bucket +4.53 sigma, smallest -4.55 -> within 6 sigma
[2026-10-08T15:25:17Z] hot-set p25-attack-f8 items (windowed control): f 0.1% S_f 0.20288% E_f(control) 0.20287% X_f +0.00001% X_f/f +0.0001 -> no hot set
[2026-10-08T15:25:17Z] hot-set p25-attack-f8 items (windowed control): f 0.5% S_f 0.95099% E_f(control) 0.95099% X_f -0.00001% X_f/f -0.0000 -> no hot set
[2026-10-08T15:25:17Z] hot-set p25-attack-f8 items (windowed control): f 1.0% S_f 1.84118% E_f(control) 1.84119% X_f -0.00001% X_f/f -0.0000 -> no hot set
[2026-10-08T15:25:17Z] hot-set p25-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:25:20Z] p25-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 69 (bin 1831153) z_max +6.54 min 6 z_min -4.60 chi2/dof 0.99956 chi2_z -1.29 top0.1% 0.16455% top0.5% 0.77447% top1% 1.50142%
[2026-10-08T15:25:20Z] hot-set p25-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.20288% E_f(control) 0.16455% X_f +0.03833% X_f/f +0.3833 -> no hot set
[2026-10-08T15:25:20Z] hot-set p25-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.95099% E_f(control) 0.77447% X_f +0.17652% X_f/f +0.3530 -> no hot set
[2026-10-08T15:25:20Z] hot-set p25-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.84118% E_f(control) 1.50142% X_f +0.33976% X_f/f +0.3398 -> no hot set
[2026-10-08T15:25:20Z] hot-set p25-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:25:20Z] ratio p25-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0000x / 1.0000x / 1.0000x (gate 1.2x at 0.1%: within); over the FLAT control 1.2329x / 1.2279x / 1.2263x
[2026-10-08T15:25:21Z] attribution: hot threshold count >= 63 marks 17204 items (0.1025% of items) holding 1116078 reads (0.2079% of reads); top 8 items t=0x42cf6d:86 t=0x6eded0:85 t=0x7dd4a0:82 t=0x77ab1b:81 t=0x5e50a1:81 t=0x6e1c13:80 t=0x78de73:79 t=0x6cb009:79
[2026-10-08T15:25:47Z] attribution: reads on hot items per position (flat expectation 0.1025% of each position's 4194304 reads); the 24 largest: p92(it5 s12)=0.607% p124(it7 s12)=0.606% p103(it6 s7)=0.606% p87(it5 s7)=0.604% p39(it2 s7)=0.603% p44(it2 s12)=0.603% p55(it3 s7)=0.600% p23(it1 s7)=0.600% p60(it3 s12)=0.600% p28(it1 s12)=0.599% p7(it0 s7)=0.599% p119(it7 s7)=0.598% p76(it4 s12)=0.598% p108(it6 s12)=0.597% p71(it4 s7)=0.597% p12(it0 s12)=0.588% p97(it6 s1)=0.307% p33(it2 s1)=0.307% p113(it7 s1)=0.306% p111(it6 s15)=0.305% p63(it3 s15)=0.305% p127(it7 s15)=0.304% p9(it0 s9)=0.304% p57(it3 s9)=0.304%
[2026-10-08T15:25:47Z] attribution: hot reads by iteration it0=0.207% it1=0.208% it2=0.208% it3=0.209% it4=0.208% it5=0.208% it6=0.208% it7=0.208% ; by site s0=0.152% s1=0.304% s2=0.152% s3=0.151% s4=0.000% s5=0.151% s6=0.151% s7=0.601% s8=0.152% s9=0.302% s10=0.152% s11=0.006% s12=0.600% s13=0.000% s14=0.151% s15=0.302%
[2026-10-08T15:25:47Z] site 0 (instr 5, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1517% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 1 (instr 11, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3043% (flat expectation 0.1025%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:25:47Z] site 2 (instr 14, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1519% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 3 (instr 17, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1510% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 4 (instr 28, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0003% (flat expectation 0.1025%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:25:47Z] site 5 (instr 29, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1506% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 6 (instr 34, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1512% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 7 (instr 35, src r4, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.6008% (flat expectation 0.1025%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:25:47Z] site 8 (instr 37, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1515% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 9 (instr 40, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3018% (flat expectation 0.1025%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:25:47Z] site 10 (instr 42, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1523% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 11 (instr 43, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0060% (flat expectation 0.1025%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:25:47Z] site 12 (instr 48, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.5997% (flat expectation 0.1025%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0068% (window expectation 0.0061%)
[2026-10-08T15:25:47Z] site 13 (instr 51, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0002% (flat expectation 0.1025%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:25:47Z] site 14 (instr 55, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1507% (flat expectation 0.1025%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:25:47Z] site 15 (instr 62, src r0, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3022% (flat expectation 0.1025%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:25:47Z] attribution: item 0x42cf6d (86 reads) read from positions p12(it0 s12)x4 p39(it2 s7)x4 p92(it5 s12)x4 p108(it6 s12)x4 p60(it3 s12)x3 p111(it6 s15)x3 p119(it7 s7)x3 p121(it7 s9)x3 p124(it7 s12)x2 p17(it1 s1)x2 p33(it2 s1)x2 p49(it3 s1)x2
[2026-10-08T15:25:47Z] attribution: item 0x6eded0 (85 reads) read from positions p23(it1 s7)x6 p31(it1 s15)x4 p127(it7 s15)x3 p7(it0 s7)x3 p80(it5 s0)x3 p103(it6 s7)x3 p111(it6 s15)x3 p12(it0 s12)x2 p28(it1 s12)x2 p33(it2 s1)x2 p37(it2 s5)x2 p49(it3 s1)x2
[2026-10-08T15:25:47Z] attribution: item 0x7dd4a0 (82 reads) read from positions p60(it3 s12)x5 p35(it2 s3)x4 p87(it5 s7)x4 p39(it2 s7)x3 p44(it2 s12)x3 p47(it2 s15)x3 p49(it3 s1)x3 p55(it3 s7)x3 p90(it5 s10)x3 p122(it7 s10)x3 p7(it0 s7)x2 p8(it0 s8)x2
[2026-10-08T15:25:47Z] attribution: item 0x77ab1b (81 reads) read from positions p87(it5 s7)x5 p41(it2 s9)x4 p1(it0 s1)x3 p17(it1 s1)x3 p28(it1 s12)x3 p44(it2 s12)x3 p108(it6 s12)x3 p7(it0 s7)x2 p8(it0 s8)x2 p12(it0 s12)x2 p21(it1 s5)x2 p47(it2 s15)x2
[2026-10-08T15:25:47Z] attribution: item 0x5e50a1 (81 reads) read from positions p57(it3 s9)x5 p44(it2 s12)x4 p71(it4 s7)x3 p92(it5 s12)x3 p103(it6 s7)x3 p121(it7 s9)x3 p126(it7 s14)x2 p7(it0 s7)x2 p23(it1 s7)x2 p28(it1 s12)x2 p33(it2 s1)x2 p41(it2 s9)x2
[2026-10-08T15:25:47Z] attribution: item 0x6e1c13 (80 reads) read from positions p103(it6 s7)x6 p31(it1 s15)x4 p39(it2 s7)x4 p44(it2 s12)x4 p42(it2 s10)x3 p81(it5 s1)x3 p115(it7 s3)x2 p5(it0 s5)x2 p7(it0 s7)x2 p18(it1 s2)x2 p33(it2 s1)x2 p56(it3 s8)x2
[2026-10-08T15:25:47Z] attribution: item 0x78de73 (79 reads) read from positions p12(it0 s12)x4 p76(it4 s12)x4 p119(it7 s7)x3 p7(it0 s7)x3 p23(it1 s7)x3 p28(it1 s12)x3 p71(it4 s7)x3 p92(it5 s12)x3 p108(it6 s12)x3 p26(it1 s10)x2 p32(it2 s0)x2 p33(it2 s1)x2
[2026-10-08T15:25:47Z] attribution: item 0x6cb009 (79 reads) read from positions p103(it6 s7)x5 p121(it7 s9)x4 p23(it1 s7)x3 p60(it3 s12)x3 p71(it4 s7)x3 p76(it4 s12)x3 p108(it6 s12)x3 p124(it7 s12)x2 p25(it1 s9)x2 p38(it2 s6)x2 p39(it2 s7)x2 p44(it2 s12)x2
[2026-10-08T15:25:47Z] hottest item p25-attack-f8 (0x42cf6d): 86 reads (0.0000% of all); predicted source none
[2026-10-08T15:25:47Z] program p25-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:25:50Z] program p26-attack-f8: epoch seed 6dec36787b113f7afe8b37d56530234466c86920158fb4e8b594a803f970f24a era seed b63aa0da8df2b2d408aafbe7e3284d0597483f81e1ee4420e46e0a24dd1cc03b id f016098c74eae0f7 attempt 0 class mx8-era5a786fb2+sh256x27+state+rw op mix load=16 add=11 sub=9 rotl=8 rotr=8 xor=6 mad=4 shfl=2; era stride mul 0x9f5f0bdb rot 19 interleave [2, 7, 11, 12] windows 3:1:1 10:0:0 11:0:0 18:1:1 21:1:0 27:0:0 31:0:0 40:1:1 45:0:0 47:0:0 52:0:0 56:2:1 57:0:0 58:0:0 62:2:3 63:0:0
[2026-10-08T15:25:50Z] window layer: site item windows (first, items) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x800000,2^23) (0x0,2^23) (0x0,2^24) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x0,2^24); expected reads per item by quarter 24.000 32.000 32.000 40.000 (flat uniform 32.000)
[2026-10-08T15:26:30Z] 131072 warps (4194304 nonces) interpreted in 39.9 s (2.432 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:26:30Z] p26-attack-f8 distinct lines per hash: n 4194304 min 1013 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8712
[2026-10-08T15:26:30Z] p26-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:26:30Z] p26-attack-f8 distinct lines per warp: n 131072 min 32575 p1 32605 median 32637 p99 32664 max 32689 mean 32636.2953
[2026-10-08T15:26:30Z] p26-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4885
[2026-10-08T15:26:30Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:26:30Z] positions: largest hi16-bucket (256 items) share 0.0078% at p43 (iteration 2, site 11, instr 56); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p6 (site 6, instr 31)
[2026-10-08T15:26:30Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p11 (site 11, instr 56); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:26:30Z] p26-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 74 (bin 14374138) z_max +7.42 min 2 z_min -5.30 chi2/dof 2.00004 chi2_z +2896.43 top0.1% 0.18752% top0.5% 0.87753% top1% 1.69734%
[2026-10-08T15:26:30Z] p26-attack-f8 item histogram against the window density: full 2^24 chi2/dof 0.99987 largest 56 (item 0x2719f9) at +6.53 sigma smallest at -4.77; buckets64 chi2/dof 0.99978 largest 1708 (bucket 10382) at +4.39 sigma smallest at -4.84
[2026-10-08T15:26:33Z] p26-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 76 (bin 14104135) z_max +7.78 min 4 z_min -4.95 chi2/dof 2.00058 chi2_z +2897.98 top0.1% 0.18749% top0.5% 0.87758% top1% 1.69739%
[2026-10-08T15:26:33Z] p26-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00080 largest +6.33 sigma; buckets64 chi2/dof 0.99670 largest +4.74 smallest -4.40
[2026-10-08T15:26:33Z] 6-sigma p26-attack-f8 items buckets64 against the window density: largest bucket +4.39 sigma, smallest -4.84 -> within 6 sigma
[2026-10-08T15:26:33Z] hot-set p26-attack-f8 items (windowed control): f 0.1% S_f 0.18752% E_f(control) 0.18749% X_f +0.00003% X_f/f +0.0003 -> no hot set
[2026-10-08T15:26:33Z] hot-set p26-attack-f8 items (windowed control): f 0.5% S_f 0.87753% E_f(control) 0.87758% X_f -0.00005% X_f/f -0.0001 -> no hot set
[2026-10-08T15:26:33Z] hot-set p26-attack-f8 items (windowed control): f 1.0% S_f 1.69734% E_f(control) 1.69739% X_f -0.00005% X_f/f -0.0001 -> no hot set
[2026-10-08T15:26:33Z] hot-set p26-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:26:37Z] p26-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 69 (bin 7725511) z_max +6.54 min 7 z_min -4.42 chi2/dof 1.00040 chi2_z +1.15 top0.1% 0.16458% top0.5% 0.77439% top1% 1.50132%
[2026-10-08T15:26:37Z] hot-set p26-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.18752% E_f(control) 0.16458% X_f +0.02294% X_f/f +0.2294 -> no hot set
[2026-10-08T15:26:37Z] hot-set p26-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.87753% E_f(control) 0.77439% X_f +0.10314% X_f/f +0.2063 -> no hot set
[2026-10-08T15:26:37Z] hot-set p26-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.69734% E_f(control) 1.50132% X_f +0.19602% X_f/f +0.1960 -> no hot set
[2026-10-08T15:26:37Z] hot-set p26-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:26:37Z] ratio p26-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0002x / 0.9999x / 1.0000x (gate 1.2x at 0.1%: within); over the FLAT control 1.1394x / 1.1332x / 1.1306x
[2026-10-08T15:26:38Z] attribution: hot threshold count >= 58 marks 18657 items (0.1112% of items) holding 1115799 reads (0.2078% of reads); top 8 items t=0xdb54fa:74 t=0xfecebb:73 t=0xfaf12c:73 t=0xe0fa47:73 t=0xdd2935:73 t=0xda050c:73 t=0xd9e89e:73 t=0xcb418b:73
[2026-10-08T15:26:56Z] attribution: reads on hot items per position (flat expectation 0.1112% of each position's 4194304 reads); the 24 largest: p46(it2 s14)=0.665% p126(it7 s14)=0.662% p62(it3 s14)=0.660% p14(it0 s14)=0.658% p30(it1 s14)=0.655% p110(it6 s14)=0.654% p78(it4 s14)=0.652% p94(it5 s14)=0.652% p67(it4 s3)=0.338% p51(it3 s3)=0.338% p80(it5 s0)=0.335% p64(it4 s0)=0.335% p115(it7 s3)=0.334% p119(it7 s7)=0.334% p16(it1 s0)=0.334% p71(it4 s7)=0.333% p3(it0 s3)=0.332% p35(it2 s3)=0.332% p112(it7 s0)=0.332% p99(it6 s3)=0.332% p83(it5 s3)=0.331% p19(it1 s3)=0.331% p23(it1 s7)=0.331% p39(it2 s7)=0.331%
[2026-10-08T15:26:56Z] attribution: hot reads by iteration it0=0.208% it1=0.207% it2=0.208% it3=0.208% it4=0.208% it5=0.208% it6=0.207% it7=0.209% ; by site s0=0.331% s1=0.166% s2=0.167% s3=0.334% s4=0.002% s5=0.167% s6=0.167% s7=0.331% s8=0.168% s9=0.167% s10=0.165% s11=0.004% s12=0.167% s13=0.167% s14=0.657% s15=0.166%
[2026-10-08T15:26:56Z] site 0 (instr 3, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3309% (flat expectation 0.1112%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:26:56Z] site 1 (instr 10, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1658% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 2 (instr 11, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1674% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 3 (instr 18, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3336% (flat expectation 0.1112%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:26:56Z] site 4 (instr 21, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0019% (flat expectation 0.1112%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:26:56Z] site 5 (instr 27, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1673% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 6 (instr 31, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1668% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 7 (instr 40, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3305% (flat expectation 0.1112%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:26:56Z] site 8 (instr 45, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1680% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 9 (instr 47, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1672% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 10 (instr 52, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1654% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 11 (instr 56, src r4, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0038% (flat expectation 0.1112%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:26:56Z] site 12 (instr 57, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1667% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 13 (instr 58, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1666% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] site 14 (instr 62, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.6572% (flat expectation 0.1112%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:26:56Z] site 15 (instr 63, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1660% (flat expectation 0.1112%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:26:56Z] attribution: item 0xdb54fa (74 reads) read from positions p64(it4 s0)x5 p14(it0 s14)x4 p87(it5 s7)x3 p80(it5 s0)x3 p99(it6 s3)x2 p16(it1 s0)x2 p23(it1 s7)x2 p42(it2 s10)x2 p44(it2 s12)x2 p51(it3 s3)x2 p3(it0 s3)x2 p79(it4 s15)x2
[2026-10-08T15:26:56Z] attribution: item 0xfecebb (73 reads) read from positions p80(it5 s0)x4 p14(it0 s14)x3 p30(it1 s14)x3 p126(it7 s14)x3 p1(it0 s1)x2 p13(it0 s13)x2 p19(it1 s3)x2 p39(it2 s7)x2 p42(it2 s10)x2 p46(it2 s14)x2 p51(it3 s3)x2 p60(it3 s12)x2
[2026-10-08T15:26:56Z] attribution: item 0xfaf12c (73 reads) read from positions p26(it1 s10)x3 p46(it2 s14)x3 p78(it4 s14)x3 p80(it5 s0)x3 p126(it7 s14)x2 p10(it0 s10)x2 p14(it0 s14)x2 p17(it1 s1)x2 p30(it1 s14)x2 p32(it2 s0)x2 p62(it3 s14)x2 p90(it5 s10)x2
[2026-10-08T15:26:56Z] attribution: item 0xe0fa47 (73 reads) read from positions p14(it0 s14)x4 p78(it4 s14)x3 p3(it0 s3)x3 p19(it1 s3)x3 p38(it2 s6)x3 p40(it2 s8)x3 p64(it4 s0)x3 p87(it5 s7)x2 p117(it7 s5)x2 p30(it1 s14)x2 p7(it0 s7)x2 p47(it2 s15)x2
[2026-10-08T15:26:56Z] attribution: item 0xdd2935 (73 reads) read from positions p14(it0 s14)x4 p94(it5 s14)x4 p110(it6 s14)x4 p79(it4 s15)x3 p124(it7 s12)x2 p7(it0 s7)x2 p16(it1 s0)x2 p22(it1 s6)x2 p37(it2 s5)x2 p55(it3 s7)x2 p62(it3 s14)x2 p78(it4 s14)x2
[2026-10-08T15:26:56Z] attribution: item 0xda050c (73 reads) read from positions p62(it3 s14)x6 p127(it7 s15)x3 p30(it1 s14)x3 p13(it0 s13)x2 p14(it0 s14)x2 p42(it2 s10)x2 p45(it2 s13)x2 p51(it3 s3)x2 p63(it3 s15)x2 p67(it4 s3)x2 p78(it4 s14)x2 p94(it5 s14)x2
[2026-10-08T15:26:56Z] attribution: item 0xd9e89e (73 reads) read from positions p46(it2 s14)x5 p78(it4 s14)x4 p0(it0 s0)x3 p32(it2 s0)x3 p81(it5 s1)x3 p87(it5 s7)x2 p126(it7 s14)x2 p21(it1 s5)x2 p24(it1 s8)x2 p29(it1 s13)x2 p7(it0 s7)x2 p33(it2 s1)x2
[2026-10-08T15:26:56Z] attribution: item 0xcb418b (73 reads) read from positions p78(it4 s14)x6 p62(it3 s14)x2 p126(it7 s14)x2 p10(it0 s10)x2 p12(it0 s12)x2 p16(it1 s0)x2 p19(it1 s3)x2 p30(it1 s14)x2 p32(it2 s0)x2 p37(it2 s5)x2 p51(it3 s3)x2 p56(it3 s8)x2
[2026-10-08T15:26:56Z] hottest item p26-attack-f8 (0xdb54fa): 74 reads (0.0000% of all); predicted source none
[2026-10-08T15:26:56Z] program p26-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:26:59Z] program p27-attack-f8: epoch seed f21a55035379ff8e1483e6e8360b946c6539fca60d4b95ae9a6c00d079ab0672 era seed b8fa171fbe2684a597b72fe402913c998dc75093a244e5f7fcb4a893cb1fb9a2 id ce3854c18fb6f409 attempt 2 class mx8-era0f2c01fb+sh256x27+state+rw op mix load=16 xor=11 rotl=8 sub=7 mad=6 add=5 rotr=5 mul=3 shfl=2 mulhi=1; era stride mul 0xab4deb19 rot 15 interleave [4, 5, 11, 13] windows 11:1:0 12:1:1 18:2:3 20:1:0 21:2:3 24:0:0 28:2:3 31:0:0 42:0:0 45:1:0 48:0:0 49:1:0 53:0:0 54:1:1 58:2:1 59:1:1
[2026-10-08T15:26:59Z] window layer: site item windows (first, items) (0x0,2^23) (0x800000,2^23) (0xc00000,2^22) (0x0,2^23) (0xc00000,2^22) (0x0,2^24) (0xc00000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x800000,2^23) (0x400000,2^22) (0x800000,2^23); expected reads per item by quarter 26.000 34.000 22.000 46.000 (flat uniform 32.000)
[2026-10-08T15:27:47Z] 131072 warps (4194304 nonces) interpreted in 47.9 s (2.926 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:27:47Z] p27-attack-f8 distinct lines per hash: n 4194304 min 1013 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8712
[2026-10-08T15:27:47Z] p27-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:27:47Z] p27-attack-f8 distinct lines per warp: n 131072 min 32573 p1 32605 median 32636 p99 32664 max 32684 mean 32636.0618
[2026-10-08T15:27:47Z] p27-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4581
[2026-10-08T15:27:47Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:27:47Z] positions: largest hi16-bucket (256 items) share 0.0081% at p34 (iteration 2, site 2, instr 18); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 11)
[2026-10-08T15:27:47Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 1 of 2048 at p0 (site 0, instr 11); saturated sources at one position 0 of 2048 at p0 (site 0, instr 11); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:27:47Z] p27-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 85 (bin 14726306) z_max +9.37 min 2 z_min -5.30 chi2/dof 3.62522 chi2_z +7603.44 top0.1% 0.21036% top0.5% 0.98815% top1% 1.91300%
[2026-10-08T15:27:47Z] p27-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00039 largest 57 (item 0x11eca4) at +6.08 sigma smallest at -4.29; buckets64 chi2/dof 0.99906 largest 1573 (bucket 156521) at +4.40 sigma smallest at -4.40
[2026-10-08T15:27:49Z] p27-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 87 (bin 16553139) z_max +9.72 min 2 z_min -5.30 chi2/dof 3.62584 chi2_z +7605.24 top0.1% 0.21034% top0.5% 0.98780% top1% 1.91249%
[2026-10-08T15:27:49Z] p27-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99992 largest +6.05 sigma; buckets64 chi2/dof 0.99652 largest +4.39 smallest -4.34
[2026-10-08T15:27:49Z] 6-sigma p27-attack-f8 items buckets64 against the window density: largest bucket +4.40 sigma, smallest -4.40 -> within 6 sigma
[2026-10-08T15:27:49Z] hot-set p27-attack-f8 items (windowed control): f 0.1% S_f 0.21036% E_f(control) 0.21034% X_f +0.00002% X_f/f +0.0002 -> no hot set
[2026-10-08T15:27:49Z] hot-set p27-attack-f8 items (windowed control): f 0.5% S_f 0.98815% E_f(control) 0.98780% X_f +0.00035% X_f/f +0.0007 -> no hot set
[2026-10-08T15:27:49Z] hot-set p27-attack-f8 items (windowed control): f 1.0% S_f 1.91300% E_f(control) 1.91249% X_f +0.00051% X_f/f +0.0005 -> no hot set
[2026-10-08T15:27:49Z] hot-set p27-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:27:51Z] p27-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 10197036) z_max +5.83 min 6 z_min -4.60 chi2/dof 0.99912 chi2_z -2.56 top0.1% 0.16449% top0.5% 0.77436% top1% 1.50119%
[2026-10-08T15:27:51Z] hot-set p27-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.21036% E_f(control) 0.16449% X_f +0.04587% X_f/f +0.4587 -> no hot set
[2026-10-08T15:27:51Z] hot-set p27-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.98815% E_f(control) 0.77436% X_f +0.21378% X_f/f +0.4276 -> no hot set
[2026-10-08T15:27:51Z] hot-set p27-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.91300% E_f(control) 1.50119% X_f +0.41181% X_f/f +0.4118 -> no hot set
[2026-10-08T15:27:51Z] hot-set p27-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:27:51Z] ratio p27-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0001x / 1.0004x / 1.0003x (gate 1.2x at 0.1%: within); over the FLAT control 1.2789x / 1.2761x / 1.2743x
[2026-10-08T15:27:53Z] attribution: hot threshold count >= 65 marks 20099 items (0.1198% of items) holding 1345303 reads (0.2506% of reads); top 8 items t=0xeea162:85 t=0xe0b4a2:85 t=0xf52110:83 t=0xed8860:82 t=0xdc9ba5:82 t=0xf7dd2e:81 t=0xecd2a6:81 t=0xe9fa6c:81
[2026-10-08T15:28:11Z] attribution: reads on hot items per position (flat expectation 0.1198% of each position's 4194304 reads); the 24 largest: p100(it6 s4)=0.710% p86(it5 s6)=0.703% p84(it5 s4)=0.703% p6(it0 s6)=0.702% p118(it7 s6)=0.702% p102(it6 s6)=0.700% p114(it7 s2)=0.700% p68(it4 s4)=0.699% p38(it2 s6)=0.699% p22(it1 s6)=0.699% p34(it2 s2)=0.699% p66(it4 s2)=0.699% p18(it1 s2)=0.699% p82(it5 s2)=0.699% p2(it0 s2)=0.698% p36(it2 s4)=0.698% p52(it3 s4)=0.698% p4(it0 s4)=0.697% p50(it3 s2)=0.693% p116(it7 s4)=0.693% p70(it4 s6)=0.693% p54(it3 s6)=0.692% p98(it6 s2)=0.690% p20(it1 s4)=0.690%
[2026-10-08T15:28:11Z] attribution: hot reads by iteration it0=0.250% it1=0.250% it2=0.251% it3=0.250% it4=0.251% it5=0.252% it6=0.251% it7=0.251% ; by site s0=0.000% s1=0.349% s2=0.697% s3=0.000% s4=0.698% s5=0.174% s6=0.699% s7=0.174% s8=0.175% s9=0.000% s10=0.173% s11=0.000% s12=0.173% s13=0.349% s14=0.000% s15=0.348%
[2026-10-08T15:28:11Z] site 0 (instr 11, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0001% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 1 (instr 12, src r6, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3487% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 2 (instr 18, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.6970% (flat expectation 0.1198%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:28:11Z] site 3 (instr 20, src r0, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0001% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 4 (instr 21, src r1, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.6983% (flat expectation 0.1198%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:28:11Z] site 5 (instr 24, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1737% (flat expectation 0.1198%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:28:11Z] site 6 (instr 28, src r7, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.6987% (flat expectation 0.1198%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:28:11Z] site 7 (instr 31, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1741% (flat expectation 0.1198%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:28:11Z] site 8 (instr 42, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1749% (flat expectation 0.1198%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:28:11Z] site 9 (instr 45, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0001% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 10 (instr 48, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1729% (flat expectation 0.1198%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:28:11Z] site 11 (instr 49, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0001% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 12 (instr 53, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1734% (flat expectation 0.1198%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:28:11Z] site 13 (instr 54, src r4, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3488% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] site 14 (instr 58, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0002% (flat expectation 0.1198%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:28:11Z] site 15 (instr 59, src r2, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3481% (flat expectation 0.1198%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:28:11Z] attribution: item 0xeea162 (85 reads) read from positions p98(it6 s2)x5 p82(it5 s2)x4 p4(it0 s4)x4 p114(it7 s2)x4 p22(it1 s6)x3 p52(it3 s4)x3 p66(it4 s2)x3 p84(it5 s4)x3 p6(it0 s6)x2 p12(it0 s12)x2 p20(it1 s4)x2 p31(it1 s15)x2
[2026-10-08T15:28:11Z] attribution: item 0xe0b4a2 (85 reads) read from positions p36(it2 s4)x4 p2(it0 s2)x3 p4(it0 s4)x3 p22(it1 s6)x3 p34(it2 s2)x3 p54(it3 s6)x3 p102(it6 s6)x3 p1(it0 s1)x3 p109(it6 s13)x3 p118(it7 s6)x3 p103(it6 s7)x2 p122(it7 s10)x2
[2026-10-08T15:28:11Z] attribution: item 0xf52110 (83 reads) read from positions p6(it0 s6)x5 p18(it1 s2)x5 p66(it4 s2)x4 p84(it5 s4)x4 p4(it0 s4)x3 p22(it1 s6)x3 p54(it3 s6)x3 p63(it3 s15)x3 p114(it7 s2)x3 p12(it0 s12)x2 p21(it1 s5)x2 p31(it1 s15)x2
[2026-10-08T15:28:11Z] attribution: item 0xed8860 (82 reads) read from positions p20(it1 s4)x4 p36(it2 s4)x4 p49(it3 s1)x4 p114(it7 s2)x4 p118(it7 s6)x3 p38(it2 s6)x3 p54(it3 s6)x3 p86(it5 s6)x3 p93(it5 s13)x3 p102(it6 s6)x3 p1(it0 s1)x2 p22(it1 s6)x2
[2026-10-08T15:28:11Z] attribution: item 0xdc9ba5 (82 reads) read from positions p86(it5 s6)x5 p4(it0 s4)x4 p70(it4 s6)x4 p2(it0 s2)x3 p38(it2 s6)x3 p114(it7 s2)x3 p118(it7 s6)x3 p63(it3 s15)x2 p117(it7 s5)x2 p36(it2 s4)x2 p47(it2 s15)x2 p52(it3 s4)x2
[2026-10-08T15:28:11Z] attribution: item 0xf7dd2e (81 reads) read from positions p2(it0 s2)x5 p22(it1 s6)x4 p52(it3 s4)x4 p4(it0 s4)x3 p36(it2 s4)x3 p82(it5 s2)x3 p127(it7 s15)x3 p6(it0 s6)x2 p20(it1 s4)x2 p29(it1 s13)x2 p38(it2 s6)x2 p65(it4 s1)x2
[2026-10-08T15:28:11Z] attribution: item 0xecd2a6 (81 reads) read from positions p34(it2 s2)x4 p2(it0 s2)x3 p22(it1 s6)x3 p31(it1 s15)x3 p54(it3 s6)x3 p98(it6 s2)x3 p102(it6 s6)x3 p111(it6 s15)x3 p116(it7 s4)x3 p6(it0 s6)x2 p26(it1 s10)x2 p47(it2 s15)x2
[2026-10-08T15:28:11Z] attribution: item 0xe9fa6c (81 reads) read from positions p54(it3 s6)x4 p13(it0 s13)x3 p18(it1 s2)x3 p29(it1 s13)x3 p50(it3 s2)x3 p65(it4 s1)x3 p77(it4 s13)x3 p84(it5 s4)x3 p106(it6 s10)x3 p109(it6 s13)x3 p116(it7 s4)x3 p2(it0 s2)x2
[2026-10-08T15:28:11Z] hottest item p27-attack-f8 (0xeea162): 85 reads (0.0000% of all); predicted source none
[2026-10-08T15:28:11Z] program p27-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:28:13Z] program p28-attack-f8: epoch seed 0c5625ee7c9e721e1624038038cf255cab17101e7b0be28bf08b1f666fa265ae era seed 4fd9c0443b974a122c7aef9137055ae16f4c52656486796f0d4308a6837ec6e1 id 3734c9f7316ac786 attempt 0 class mx8-eraf18f3410+sh256x27+state+rw op mix load=16 add=14 xor=8 mad=7 rotr=6 rotl=5 sub=4 mul=3 shfl=1; era stride mul 0x347b218b rot 26 interleave [0, 6, 12, 13] windows 1:1:1 7:0:0 9:2:1 11:0:0 14:0:0 24:0:0 25:2:2 26:1:0 34:2:3 37:0:0 40:2:0 44:2:1 50:1:1 51:0:0 53:2:1 58:1:1
[2026-10-08T15:28:13Z] window layer: site item windows (first, items) (0x800000,2^23) (0x0,2^24) (0x400000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x800000,2^22) (0x0,2^23) (0xc00000,2^22) (0x0,2^24) (0x0,2^22) (0x400000,2^22) (0x800000,2^23) (0x0,2^24) (0x400000,2^22) (0x800000,2^23); expected reads per item by quarter 24.000 40.000 32.000 32.000 (flat uniform 32.000)
[2026-10-08T15:28:50Z] 131072 warps (4194304 nonces) interpreted in 37.2 s (2.272 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:28:50Z] p28-attack-f8 distinct lines per hash: n 4194304 min 1007 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8710
[2026-10-08T15:28:50Z] p28-attack-f8 distinct items per hash: n 4194304 min 126 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:28:50Z] p28-attack-f8 distinct lines per warp: n 131072 min 32563 p1 32605 median 32637 p99 32664 max 32692 mean 32636.2251
[2026-10-08T15:28:50Z] p28-attack-f8 distinct items per warp: n 131072 min 4091 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4857
[2026-10-08T15:28:50Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:28:50Z] positions: largest hi16-bucket (256 items) share 0.0080% at p94 (iteration 5, site 14, instr 53); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 1)
[2026-10-08T15:28:50Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 1 of 2048 at p0 (site 0, instr 1); saturated sources at one position 0 of 2048 at p0 (site 0, instr 1); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:28:51Z] p28-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 78 (bin 5658797) z_max +8.13 min 4 z_min -4.95 chi2/dof 2.00001 chi2_z +2896.34 top0.1% 0.18751% top0.5% 0.87758% top1% 1.69725%
[2026-10-08T15:28:51Z] p28-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00014 largest 54 (item 0x30a0da) at +6.12 sigma smallest at -4.43; buckets64 chi2/dof 1.00036 largest 2250 (bucket 181019) at +4.46 sigma smallest at -4.57
[2026-10-08T15:28:54Z] p28-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 77 (bin 6910839) z_max +7.95 min 3 z_min -5.13 chi2/dof 1.99988 chi2_z +2895.95 top0.1% 0.18740% top0.5% 0.87719% top1% 1.69708%
[2026-10-08T15:28:54Z] p28-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99990 largest +6.12 sigma; buckets64 chi2/dof 0.99627 largest +4.86 smallest -4.64
[2026-10-08T15:28:54Z] 6-sigma p28-attack-f8 items buckets64 against the window density: largest bucket +4.46 sigma, smallest -4.57 -> within 6 sigma
[2026-10-08T15:28:54Z] hot-set p28-attack-f8 items (windowed control): f 0.1% S_f 0.18751% E_f(control) 0.18740% X_f +0.00011% X_f/f +0.0011 -> no hot set
[2026-10-08T15:28:54Z] hot-set p28-attack-f8 items (windowed control): f 0.5% S_f 0.87758% E_f(control) 0.87719% X_f +0.00038% X_f/f +0.0008 -> no hot set
[2026-10-08T15:28:54Z] hot-set p28-attack-f8 items (windowed control): f 1.0% S_f 1.69725% E_f(control) 1.69708% X_f +0.00017% X_f/f +0.0002 -> no hot set
[2026-10-08T15:28:54Z] hot-set p28-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:28:57Z] p28-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 66 (bin 10920895) z_max +6.01 min 6 z_min -4.60 chi2/dof 1.00083 chi2_z +2.41 top0.1% 0.16456% top0.5% 0.77447% top1% 1.50154%
[2026-10-08T15:28:57Z] hot-set p28-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.18751% E_f(control) 0.16456% X_f +0.02294% X_f/f +0.2294 -> no hot set
[2026-10-08T15:28:57Z] hot-set p28-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.87758% E_f(control) 0.77447% X_f +0.10310% X_f/f +0.2062 -> no hot set
[2026-10-08T15:28:57Z] hot-set p28-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.69725% E_f(control) 1.50154% X_f +0.19570% X_f/f +0.1957 -> no hot set
[2026-10-08T15:28:57Z] hot-set p28-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:28:57Z] ratio p28-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0006x / 1.0004x / 1.0001x (gate 1.2x at 0.1%: within); over the FLAT control 1.1394x / 1.1331x / 1.1303x
[2026-10-08T15:28:58Z] attribution: hot threshold count >= 58 marks 18678 items (0.1113% of items) holding 1116941 reads (0.2080% of reads); top 8 items t=0x5658ad:78 t=0x7c83ef:74 t=0x423af3:74 t=0x529a4f:73 t=0x774d28:72 t=0x772ad4:72 t=0x706c5a:72 t=0x6f3ac4:72
[2026-10-08T15:29:40Z] attribution: reads on hot items per position (flat expectation 0.1113% of each position's 4194304 reads); the 24 largest: p78(it4 s14)=0.667% p123(it7 s11)=0.665% p11(it0 s11)=0.664% p14(it0 s14)=0.664% p30(it1 s14)=0.664% p82(it5 s2)=0.662% p110(it6 s14)=0.662% p94(it5 s14)=0.662% p62(it3 s14)=0.660% p107(it6 s11)=0.659% p2(it0 s2)=0.659% p46(it2 s14)=0.659% p43(it2 s11)=0.658% p91(it5 s11)=0.658% p98(it6 s2)=0.658% p18(it1 s2)=0.658% p126(it7 s14)=0.657% p27(it1 s11)=0.656% p66(it4 s2)=0.656% p34(it2 s2)=0.655% p50(it3 s2)=0.655% p59(it3 s11)=0.653% p114(it7 s2)=0.653% p75(it4 s11)=0.652%
[2026-10-08T15:29:40Z] attribution: hot reads by iteration it0=0.209% it1=0.208% it2=0.208% it3=0.207% it4=0.208% it5=0.208% it6=0.208% it7=0.208% ; by site s0=0.005% s1=0.167% s2=0.657% s3=0.167% s4=0.167% s5=0.167% s6=0.005% s7=0.326% s8=0.005% s9=0.168% s10=0.000% s11=0.658% s12=0.005% s13=0.167% s14=0.662% s15=0.005%
[2026-10-08T15:29:40Z] site 0 (instr 1, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0048% (flat expectation 0.1113%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:29:40Z] site 1 (instr 7, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1668% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 2 (instr 9, src r1, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.6568% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 3 (instr 11, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1669% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 4 (instr 14, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1668% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 5 (instr 24, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1670% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 6 (instr 25, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0047% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 7 (instr 26, src r3, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3262% (flat expectation 0.1113%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:29:40Z] site 8 (instr 34, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0048% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 9 (instr 37, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1679% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 10 (instr 40, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 11 (instr 44, src r2, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.6581% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 12 (instr 50, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0047% (flat expectation 0.1113%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:29:40Z] site 13 (instr 51, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1666% (flat expectation 0.1113%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:29:40Z] site 14 (instr 53, src r6, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.6618% (flat expectation 0.1113%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:29:40Z] site 15 (instr 58, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0047% (flat expectation 0.1113%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:29:40Z] attribution: item 0x5658ad (78 reads) read from positions p30(it1 s14)x6 p78(it4 s14)x4 p98(it6 s2)x4 p110(it6 s14)x3 p43(it2 s11)x3 p2(it0 s2)x3 p66(it4 s2)x3 p62(it3 s14)x2 p123(it7 s11)x2 p39(it2 s7)x2 p41(it2 s9)x2 p11(it0 s11)x2
[2026-10-08T15:29:40Z] attribution: item 0x7c83ef (74 reads) read from positions p30(it1 s14)x5 p91(it5 s11)x5 p46(it2 s14)x4 p110(it6 s14)x4 p78(it4 s14)x3 p113(it7 s1)x3 p123(it7 s11)x3 p126(it7 s14)x2 p18(it1 s2)x2 p27(it1 s11)x2 p34(it2 s2)x2 p66(it4 s2)x2
[2026-10-08T15:29:40Z] attribution: item 0x423af3 (74 reads) read from positions p11(it0 s11)x4 p50(it3 s2)x4 p2(it0 s2)x4 p110(it6 s14)x3 p27(it1 s11)x3 p75(it4 s11)x3 p91(it5 s11)x3 p57(it3 s9)x2 p126(it7 s14)x2 p34(it2 s2)x2 p43(it2 s11)x2 p45(it2 s13)x2
[2026-10-08T15:29:40Z] attribution: item 0x529a4f (73 reads) read from positions p14(it0 s14)x4 p98(it6 s2)x4 p18(it1 s2)x3 p65(it4 s1)x3 p110(it6 s14)x3 p2(it0 s2)x2 p11(it0 s11)x2 p20(it1 s4)x2 p39(it2 s7)x2 p43(it2 s11)x2 p46(it2 s14)x2 p51(it3 s3)x2
[2026-10-08T15:29:40Z] attribution: item 0x774d28 (72 reads) read from positions p18(it1 s2)x5 p50(it3 s2)x4 p78(it4 s14)x4 p2(it0 s2)x3 p14(it0 s14)x3 p43(it2 s11)x3 p59(it3 s11)x3 p91(it5 s11)x3 p94(it5 s14)x3 p9(it0 s9)x2 p11(it0 s11)x2 p17(it1 s1)x2
[2026-10-08T15:29:40Z] attribution: item 0x772ad4 (72 reads) read from positions p110(it6 s14)x5 p30(it1 s14)x4 p11(it0 s11)x3 p66(it4 s2)x3 p69(it4 s5)x3 p82(it5 s2)x3 p98(it6 s2)x3 p123(it7 s11)x2 p9(it0 s9)x2 p18(it1 s2)x2 p27(it1 s11)x2 p33(it2 s1)x2
[2026-10-08T15:29:40Z] attribution: item 0x706c5a (72 reads) read from positions p11(it0 s11)x4 p78(it4 s14)x4 p14(it0 s14)x3 p82(it5 s2)x3 p91(it5 s11)x3 p107(it6 s11)x3 p114(it7 s2)x3 p123(it7 s11)x3 p126(it7 s14)x2 p18(it1 s2)x2 p34(it2 s2)x2 p37(it2 s5)x2
[2026-10-08T15:29:40Z] attribution: item 0x6f3ac4 (72 reads) read from positions p11(it0 s11)x4 p123(it7 s11)x3 p18(it1 s2)x3 p59(it3 s11)x3 p78(it4 s14)x3 p82(it5 s2)x3 p91(it5 s11)x3 p94(it5 s14)x3 p103(it6 s7)x3 p114(it7 s2)x3 p65(it4 s1)x2 p43(it2 s11)x2
[2026-10-08T15:29:40Z] hottest item p28-attack-f8 (0x5658ad): 78 reads (0.0000% of all); predicted source none
[2026-10-08T15:29:40Z] program p28-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:29:42Z] program p29-attack-f8: epoch seed 7d7e131e5018c707926e608ebe0956cb78afc3f32b2c4cae3edc5dbf64bbc616 era seed da985cca265c3db28d22ba96b335dadc64e9036cfc46454f47994516c94598cb id 2e84da6ea8f3a4d3 attempt 0 class mx8-era4bbbf36c+sh256x27+state+rw op mix load=16 add=9 mad=9 xor=8 rotl=7 rotr=4 sub=4 mul=3 shfl=3 mulhi=1; era stride mul 0xe597ba17 rot 25 interleave [4, 5, 9, 14] windows 5:2:3 6:1:0 7:1:0 11:0:0 16:1:1 21:1:1 34:0:0 38:0:0 42:0:0 43:1:1 47:0:0 50:1:0 53:1:1 58:0:0 59:2:0 60:1:0
[2026-10-08T15:29:42Z] window layer: site item windows (first, items) (0xc00000,2^22) (0x0,2^23) (0x0,2^23) (0x0,2^24) (0x800000,2^23) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^23) (0x800000,2^23) (0x0,2^24) (0x0,2^22) (0x0,2^23); expected reads per item by quarter 36.000 28.000 28.000 36.000 (flat uniform 32.000)
[2026-10-08T15:30:50Z] 131072 warps (4194304 nonces) interpreted in 67.9 s (4.144 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:30:50Z] p29-attack-f8 distinct lines per hash: n 4194304 min 983 p1 1008 median 1024 p99 1024 max 1024 mean 1021.8685
[2026-10-08T15:30:50Z] p29-attack-f8 distinct items per hash: n 4194304 min 123 p1 126 median 128 p99 128 max 128 mean 127.7491
[2026-10-08T15:30:50Z] p29-attack-f8 distinct lines per warp: n 131072 min 32451 p1 32509 median 32574 p99 32627 max 32664 mean 32572.7225
[2026-10-08T15:30:50Z] p29-attack-f8 distinct items per warp: n 131072 min 4072 p1 4080 median 4088 p99 4093 max 4096 mean 4087.4815
[2026-10-08T15:30:50Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:30:50Z] positions: largest hi16-bucket (256 items) share 0.0078% at p0 (iteration 0, site 0, instr 5); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 5)
[2026-10-08T15:30:50Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p16 (site 0, instr 5); saturated sources at one position 0 of 2048 at p0 (site 0, instr 5); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:30:50Z] p29-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 72 (bin 4130270) z_max +7.07 min 5 z_min -4.77 chi2/dof 1.50795 chi2_z +1471.18 top0.1% 0.17673% top0.5% 0.82859% top1% 1.60495%
[2026-10-08T15:30:50Z] p29-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00438 largest 60 (item 0x8093ce) at +6.05 sigma smallest at -4.50; buckets64 chi2/dof 1.00890 largest 2511 (bucket 254637) at +4.31 sigma smallest at -4.38
[2026-10-08T15:30:52Z] p29-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 70 (bin 14517762) z_max +6.72 min 5 z_min -4.77 chi2/dof 1.50016 chi2_z +1448.62 top0.1% 0.17642% top0.5% 0.82777% top1% 1.60347%
[2026-10-08T15:30:52Z] p29-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00067 largest +6.61 sigma; buckets64 chi2/dof 0.99837 largest +4.83 smallest -4.23
[2026-10-08T15:30:52Z] 6-sigma p29-attack-f8 items buckets64 against the window density: largest bucket +4.31 sigma, smallest -4.38 -> within 6 sigma
[2026-10-08T15:30:52Z] hot-set p29-attack-f8 items (windowed control): f 0.1% S_f 0.17673% E_f(control) 0.17642% X_f +0.00030% X_f/f +0.0030 -> no hot set
[2026-10-08T15:30:52Z] hot-set p29-attack-f8 items (windowed control): f 0.5% S_f 0.82859% E_f(control) 0.82777% X_f +0.00082% X_f/f +0.0016 -> no hot set
[2026-10-08T15:30:52Z] hot-set p29-attack-f8 items (windowed control): f 1.0% S_f 1.60495% E_f(control) 1.60347% X_f +0.00148% X_f/f +0.0015 -> no hot set
[2026-10-08T15:30:52Z] hot-set p29-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:30:54Z] p29-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 67 (bin 212093) z_max +6.19 min 6 z_min -4.60 chi2/dof 0.99974 chi2_z -0.74 top0.1% 0.16453% top0.5% 0.77416% top1% 1.50104%
[2026-10-08T15:30:54Z] hot-set p29-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17673% E_f(control) 0.16453% X_f +0.01220% X_f/f +0.1220 -> no hot set
[2026-10-08T15:30:54Z] hot-set p29-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.82859% E_f(control) 0.77416% X_f +0.05443% X_f/f +0.1089 -> no hot set
[2026-10-08T15:30:54Z] hot-set p29-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.60495% E_f(control) 1.50104% X_f +0.10391% X_f/f +0.1039 -> no hot set
[2026-10-08T15:30:54Z] hot-set p29-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:30:54Z] ratio p29-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0017x / 1.0010x / 1.0009x (gate 1.2x at 0.1%: within); over the FLAT control 1.0741x / 1.0703x / 1.0692x
[2026-10-08T15:30:55Z] attribution: hot threshold count >= 54 marks 26069 items (0.1554% of items) holding 1450567 reads (0.2702% of reads); top 8 items t=0x3f05de:72 t=0xf4133c:70 t=0x3ce0da:69 t=0xde14c5:68 t=0xd438dd:68 t=0x392b03:68 t=0x1b4610:68 t=0x1868d6:68
[2026-10-08T15:31:12Z] attribution: reads on hot items per position (flat expectation 0.1554% of each position's 4194304 reads); the 24 largest: p110(it6 s14)=0.486% p94(it5 s14)=0.486% p78(it4 s14)=0.483% p46(it2 s14)=0.483% p14(it0 s14)=0.483% p112(it7 s0)=0.482% p62(it3 s14)=0.481% p126(it7 s14)=0.480% p30(it1 s14)=0.479% p80(it5 s0)=0.476% p32(it2 s0)=0.474% p48(it3 s0)=0.471% p96(it6 s0)=0.470% p0(it0 s0)=0.470% p16(it1 s0)=0.470% p64(it4 s0)=0.465% p82(it5 s2)=0.247% p40(it2 s8)=0.247% p26(it1 s10)=0.247% p97(it6 s1)=0.247% p81(it5 s1)=0.247% p49(it3 s1)=0.247% p8(it0 s8)=0.246% p90(it5 s10)=0.246%
[2026-10-08T15:31:12Z] attribution: hot reads by iteration it0=0.270% it1=0.270% it2=0.270% it3=0.270% it4=0.269% it5=0.271% it6=0.270% it7=0.271% ; by site s0=0.472% s1=0.244% s2=0.242% s3=0.238% s4=0.239% s5=0.238% s6=0.239% s7=0.240% s8=0.244% s9=0.238% s10=0.244% s11=0.241% s12=0.237% s13=0.240% s14=0.483% s15=0.243%
[2026-10-08T15:31:12Z] site 0 (instr 5, src r0, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.4722% (flat expectation 0.1554%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:31:12Z] site 1 (instr 6, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2440% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 2 (instr 7, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2424% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 3 (instr 11, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2383% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 4 (instr 16, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2393% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 5 (instr 21, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2380% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 6 (instr 34, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2388% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 7 (instr 38, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2397% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 8 (instr 42, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2439% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 9 (instr 43, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2385% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 10 (instr 47, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2439% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 11 (instr 50, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2412% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 12 (instr 53, src r2, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2369% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] site 13 (instr 58, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2403% (flat expectation 0.1554%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:31:12Z] site 14 (instr 59, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4827% (flat expectation 0.1554%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:31:12Z] site 15 (instr 60, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2428% (flat expectation 0.1554%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:31:12Z] attribution: item 0x3f05de (72 reads) read from positions p62(it3 s14)x7 p126(it7 s14)x3 p14(it0 s14)x3 p26(it1 s10)x3 p49(it3 s1)x3 p78(it4 s14)x3 p79(it4 s15)x3 p11(it0 s11)x2 p30(it1 s14)x2 p35(it2 s3)x2 p46(it2 s14)x2 p61(it3 s13)x2
[2026-10-08T15:31:12Z] attribution: item 0xf4133c (70 reads) read from positions p32(it2 s0)x4 p16(it1 s0)x3 p36(it2 s4)x3 p92(it5 s12)x3 p4(it0 s4)x2 p5(it0 s5)x2 p13(it0 s13)x2 p20(it1 s4)x2 p26(it1 s10)x2 p35(it2 s3)x2 p44(it2 s12)x2 p55(it3 s7)x2
[2026-10-08T15:31:12Z] attribution: item 0x3ce0da (69 reads) read from positions p78(it4 s14)x6 p62(it3 s14)x4 p30(it1 s14)x3 p46(it2 s14)x3 p63(it3 s15)x3 p65(it4 s1)x3 p88(it5 s8)x3 p94(it5 s14)x3 p119(it7 s7)x2 p14(it0 s14)x2 p15(it0 s15)x2 p38(it2 s6)x2
[2026-10-08T15:31:12Z] attribution: item 0xde14c5 (68 reads) read from positions p0(it0 s0)x4 p20(it1 s4)x3 p48(it3 s0)x3 p119(it7 s7)x3 p125(it7 s13)x2 p8(it0 s8)x2 p10(it0 s10)x2 p12(it0 s12)x2 p16(it1 s0)x2 p32(it2 s0)x2 p41(it2 s9)x2 p42(it2 s10)x2
[2026-10-08T15:31:12Z] attribution: item 0xd438dd (68 reads) read from positions p41(it2 s9)x5 p100(it6 s4)x3 p32(it2 s0)x3 p44(it2 s12)x3 p85(it5 s5)x3 p104(it6 s8)x2 p117(it7 s5)x2 p16(it1 s0)x2 p28(it1 s12)x2 p5(it0 s5)x2 p8(it0 s8)x2 p9(it0 s9)x2
[2026-10-08T15:31:12Z] attribution: item 0x392b03 (68 reads) read from positions p14(it0 s14)x4 p78(it4 s14)x4 p114(it7 s2)x4 p62(it3 s14)x3 p65(it4 s1)x3 p126(it7 s14)x2 p29(it1 s13)x2 p42(it2 s10)x2 p46(it2 s14)x2 p79(it4 s15)x2 p81(it5 s1)x2 p82(it5 s2)x2
[2026-10-08T15:31:12Z] attribution: item 0x1b4610 (68 reads) read from positions p110(it6 s14)x4 p115(it7 s3)x3 p14(it0 s14)x3 p31(it1 s15)x2 p33(it2 s1)x2 p34(it2 s2)x2 p63(it3 s15)x2 p65(it4 s1)x2 p75(it4 s11)x2 p78(it4 s14)x2 p81(it5 s1)x2 p94(it5 s14)x2
[2026-10-08T15:31:12Z] attribution: item 0x1868d6 (68 reads) read from positions p49(it3 s1)x4 p30(it1 s14)x3 p62(it3 s14)x3 p2(it0 s2)x2 p6(it0 s6)x2 p17(it1 s1)x2 p19(it1 s3)x2 p24(it1 s8)x2 p31(it1 s15)x2 p47(it2 s15)x2 p58(it3 s10)x2 p67(it4 s3)x2
[2026-10-08T15:31:12Z] hottest item p29-attack-f8 (0x3f05de): 72 reads (0.0000% of all); predicted source none
[2026-10-08T15:31:12Z] program p29-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:31:15Z] program p30-attack-f8: epoch seed ab7cd3cd0869be5274ceb894a768e82d2f4d97d33833f7746864138744b7c559 era seed e1de2b867d7439f7e80693cec571bf3b22c8dd27a5e1120209515a5080585e8e id 40906acb1cdfc01c attempt 0 class mx8-era8907e324+sh256x27+state+rw op mix load=16 rotr=9 sub=9 xor=9 add=8 mad=5 mul=3 rotl=3 mulhi=1 shfl=1; era stride mul 0xac8f1c1b rot 18 interleave [3, 4, 11, 12] windows 3:2:0 5:0:0 13:2:0 15:2:1 16:0:0 17:2:0 18:0:0 22:0:0 25:1:0 26:0:0 27:0:0 33:0:0 36:2:3 43:0:0 48:0:0 50:1:0
[2026-10-08T15:31:15Z] window layer: site item windows (first, items) (0x0,2^22) (0x0,2^24) (0x0,2^22) (0x400000,2^22) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0xc00000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^23); expected reads per item by quarter 50.000 34.000 18.000 26.000 (flat uniform 32.000)
[2026-10-08T15:31:48Z] 131072 warps (4194304 nonces) interpreted in 33.5 s (2.046 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:31:48Z] p30-attack-f8 distinct lines per hash: n 4194304 min 991 p1 1015 median 1024 p99 1024 max 1024 mean 1022.8682
[2026-10-08T15:31:48Z] p30-attack-f8 distinct items per hash: n 4194304 min 124 p1 127 median 128 p99 128 max 128 mean 127.8741
[2026-10-08T15:31:48Z] p30-attack-f8 distinct lines per warp: n 131072 min 32505 p1 32553 median 32605 p99 32647 max 32678 mean 32603.9759
[2026-10-08T15:31:48Z] p30-attack-f8 distinct items per warp: n 131072 min 4080 p1 4086 median 4092 p99 4095 max 4096 mean 4091.4219
[2026-10-08T15:31:48Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:31:48Z] positions: largest hi16-bucket (256 items) share 0.0078% at p114 (iteration 7, site 2, instr 13); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p106 (site 10, instr 27)
[2026-10-08T15:31:48Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p37 (site 5, instr 17); saturated sources at one position 0 of 2048 at p0 (site 0, instr 3); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:31:48Z] p30-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 91 (bin 1459257) z_max +10.43 min 1 z_min -5.48 chi2/dof 5.37801 chi2_z +12680.06 top0.1% 0.22583% top0.5% 1.06168% top1% 2.05856%
[2026-10-08T15:31:48Z] p30-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00153 largest 43 (item 0xa33651) at +5.89 sigma smallest at -4.81; buckets64 chi2/dof 1.00426 largest 1305 (bucket 191849) at +4.51 sigma smallest at -4.27
[2026-10-08T15:31:50Z] p30-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 88 (bin 2489228) z_max +9.90 min 2 z_min -5.30 chi2/dof 5.37369 chi2_z +12667.57 top0.1% 0.22558% top0.5% 1.06084% top1% 2.05710%
[2026-10-08T15:31:50Z] p30-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 0.99919 largest +6.08 sigma; buckets64 chi2/dof 0.99635 largest +4.71 smallest -4.17
[2026-10-08T15:31:50Z] 6-sigma p30-attack-f8 items buckets64 against the window density: largest bucket +4.51 sigma, smallest -4.27 -> within 6 sigma
[2026-10-08T15:31:50Z] hot-set p30-attack-f8 items (windowed control): f 0.1% S_f 0.22583% E_f(control) 0.22558% X_f +0.00026% X_f/f +0.0026 -> no hot set
[2026-10-08T15:31:50Z] hot-set p30-attack-f8 items (windowed control): f 0.5% S_f 1.06168% E_f(control) 1.06084% X_f +0.00085% X_f/f +0.0017 -> no hot set
[2026-10-08T15:31:50Z] hot-set p30-attack-f8 items (windowed control): f 1.0% S_f 2.05856% E_f(control) 2.05710% X_f +0.00146% X_f/f +0.0015 -> no hot set
[2026-10-08T15:31:50Z] hot-set p30-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:31:52Z] p30-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 69 (bin 8710271) z_max +6.54 min 6 z_min -4.60 chi2/dof 0.99959 chi2_z -1.19 top0.1% 0.16457% top0.5% 0.77431% top1% 1.50118%
[2026-10-08T15:31:52Z] hot-set p30-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.22583% E_f(control) 0.16457% X_f +0.06127% X_f/f +0.6127 -> no hot set
[2026-10-08T15:31:52Z] hot-set p30-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 1.06168% E_f(control) 0.77431% X_f +0.28738% X_f/f +0.5748 -> no hot set
[2026-10-08T15:31:52Z] hot-set p30-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 2.05856% E_f(control) 1.50118% X_f +0.55738% X_f/f +0.5574 -> no hot set
[2026-10-08T15:31:52Z] hot-set p30-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:31:52Z] ratio p30-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0011x / 1.0008x / 1.0007x (gate 1.2x at 0.1%: within); over the FLAT control 1.3723x / 1.3711x / 1.3713x
[2026-10-08T15:31:53Z] attribution: hot threshold count >= 70 marks 18490 items (0.1102% of items) holding 1332351 reads (0.2482% of reads); top 8 items t=0x164439:91 t=0x17c18e:89 t=0x35454a:88 t=0x184dc7:88 t=0x13f3f4:88 t=0x02cbe5:86 t=0x3bcf78:85 t=0x3aa44f:85
[2026-10-08T15:32:10Z] attribution: reads on hot items per position (flat expectation 0.1102% of each position's 4194304 reads); the 24 largest: p117(it7 s5)=0.647% p21(it1 s5)=0.643% p37(it2 s5)=0.642% p85(it5 s5)=0.641% p69(it4 s5)=0.641% p2(it0 s2)=0.638% p0(it0 s0)=0.637% p32(it2 s0)=0.637% p16(it1 s0)=0.636% p114(it7 s2)=0.636% p48(it3 s0)=0.636% p80(it5 s0)=0.635% p5(it0 s5)=0.635% p34(it2 s2)=0.635% p64(it4 s0)=0.635% p82(it5 s2)=0.634% p53(it3 s5)=0.634% p96(it6 s0)=0.633% p101(it6 s5)=0.633% p112(it7 s0)=0.632% p18(it1 s2)=0.631% p98(it6 s2)=0.629% p66(it4 s2)=0.629% p50(it3 s2)=0.627%
[2026-10-08T15:32:10Z] attribution: hot reads by iteration it0=0.249% it1=0.248% it2=0.249% it3=0.248% it4=0.249% it5=0.248% it6=0.247% it7=0.248% ; by site s0=0.635% s1=0.159% s2=0.632% s3=0.000% s4=0.158% s5=0.639% s6=0.159% s7=0.157% s8=0.321% s9=0.158% s10=0.159% s11=0.159% s12=0.000% s13=0.158% s14=0.159% s15=0.317%
[2026-10-08T15:32:10Z] site 0 (instr 3, src r2, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6353% (flat expectation 0.1102%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:32:10Z] site 1 (instr 5, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1586% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 2 (instr 13, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6324% (flat expectation 0.1102%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:32:10Z] site 3 (instr 15, src r7, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1102%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:32:10Z] site 4 (instr 16, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1584% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 5 (instr 17, src r1, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.6393% (flat expectation 0.1102%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:32:10Z] site 6 (instr 18, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1586% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 7 (instr 22, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1574% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 8 (instr 25, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3212% (flat expectation 0.1102%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:32:10Z] site 9 (instr 26, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1576% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 10 (instr 27, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1593% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 11 (instr 33, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1592% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 12 (instr 36, src r6, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1102%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:32:10Z] site 13 (instr 43, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1578% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 14 (instr 48, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1590% (flat expectation 0.1102%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:32:10Z] site 15 (instr 50, src r1, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3166% (flat expectation 0.1102%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:32:10Z] attribution: item 0x164439 (91 reads) read from positions p80(it5 s0)x4 p96(it6 s0)x4 p16(it1 s0)x3 p34(it2 s2)x3 p48(it3 s0)x3 p69(it4 s5)x3 p95(it5 s15)x3 p104(it6 s8)x3 p112(it7 s0)x2 p4(it0 s4)x2 p14(it0 s14)x2 p21(it1 s5)x2
[2026-10-08T15:32:10Z] attribution: item 0x17c18e (89 reads) read from positions p114(it7 s2)x4 p21(it1 s5)x3 p34(it2 s2)x3 p69(it4 s5)x3 p80(it5 s0)x3 p85(it5 s5)x3 p117(it7 s5)x3 p120(it7 s8)x2 p5(it0 s5)x2 p8(it0 s8)x2 p16(it1 s0)x2 p24(it1 s8)x2
[2026-10-08T15:32:10Z] attribution: item 0x35454a (88 reads) read from positions p64(it4 s0)x5 p127(it7 s15)x3 p37(it2 s5)x3 p80(it5 s0)x3 p85(it5 s5)x3 p98(it6 s2)x3 p117(it7 s5)x3 p0(it0 s0)x2 p21(it1 s5)x2 p24(it1 s8)x2 p31(it1 s15)x2 p34(it2 s2)x2
[2026-10-08T15:32:10Z] attribution: item 0x184dc7 (88 reads) read from positions p53(it3 s5)x4 p117(it7 s5)x3 p90(it5 s10)x3 p96(it6 s0)x3 p112(it7 s0)x3 p2(it0 s2)x2 p6(it0 s6)x2 p14(it0 s14)x2 p18(it1 s2)x2 p21(it1 s5)x2 p32(it2 s0)x2 p34(it2 s2)x2
[2026-10-08T15:32:10Z] attribution: item 0x13f3f4 (88 reads) read from positions p5(it0 s5)x5 p117(it7 s5)x3 p31(it1 s15)x3 p53(it3 s5)x3 p71(it4 s7)x3 p96(it6 s0)x3 p18(it1 s2)x2 p20(it1 s4)x2 p24(it1 s8)x2 p32(it2 s0)x2 p37(it2 s5)x2 p40(it2 s8)x2
[2026-10-08T15:32:10Z] attribution: item 0x02cbe5 (86 reads) read from positions p91(it5 s11)x4 p114(it7 s2)x4 p32(it2 s0)x3 p82(it5 s2)x3 p85(it5 s5)x3 p88(it5 s8)x3 p112(it7 s0)x3 p127(it7 s15)x3 p2(it0 s2)x2 p7(it0 s7)x2 p15(it0 s15)x2 p21(it1 s5)x2
[2026-10-08T15:32:10Z] attribution: item 0x3bcf78 (85 reads) read from positions p66(it4 s2)x5 p34(it2 s2)x3 p69(it4 s5)x3 p96(it6 s0)x3 p98(it6 s2)x3 p113(it7 s1)x3 p4(it0 s4)x2 p5(it0 s5)x2 p8(it0 s8)x2 p14(it0 s14)x2 p15(it0 s15)x2 p17(it1 s1)x2
[2026-10-08T15:32:10Z] attribution: item 0x3aa44f (85 reads) read from positions p32(it2 s0)x6 p18(it1 s2)x5 p64(it4 s0)x5 p8(it0 s8)x4 p21(it1 s5)x4 p2(it0 s2)x3 p37(it2 s5)x3 p53(it3 s5)x3 p114(it7 s2)x3 p5(it0 s5)x2 p16(it1 s0)x2 p23(it1 s7)x2
[2026-10-08T15:32:10Z] hottest item p30-attack-f8 (0x164439): 91 reads (0.0000% of all); predicted source none
[2026-10-08T15:32:10Z] program p30-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:32:13Z] program p31-attack-f8: epoch seed e077e8e988982c1de9776ce91bfd55367259faab9d3fa716c195c06f59bd32a5 era seed ce9eee734d40682694fc02df739b73c8de2fea370be1dc9c1667288159dd741c id 31cac1b1f738a5fe attempt 0 class mx8-era8ea479d1+sh256x27+state+rw op mix load=16 add=10 xor=9 rotr=7 mad=5 rotl=5 shfl=5 sub=5 mulhi=2; era stride mul 0xd4aea2c7 rot 14 interleave [2, 5, 12, 14] windows 4:0:0 6:2:0 12:0:0 15:1:1 17:1:1 19:0:0 20:1:1 22:0:0 30:2:0 34:1:0 42:0:0 46:1:1 53:1:0 54:0:0 59:2:3 60:2:1
[2026-10-08T15:32:13Z] window layer: site item windows (first, items) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x800000,2^23) (0x800000,2^23) (0x0,2^24) (0x800000,2^23) (0x0,2^24) (0x0,2^22) (0x0,2^23) (0x0,2^24) (0x800000,2^23) (0x0,2^23) (0x0,2^24) (0xc00000,2^22) (0x400000,2^22); expected reads per item by quarter 36.000 28.000 28.000 36.000 (flat uniform 32.000)
[2026-10-08T15:32:52Z] 131072 warps (4194304 nonces) interpreted in 39.4 s (2.403 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:32:52Z] p31-attack-f8 distinct lines per hash: n 4194304 min 1014 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8715
[2026-10-08T15:32:52Z] p31-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:32:52Z] p31-attack-f8 distinct lines per warp: n 131072 min 32577 p1 32605 median 32637 p99 32664 max 32683 mean 32636.2858
[2026-10-08T15:32:52Z] p31-attack-f8 distinct items per warp: n 131072 min 4090 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4930
[2026-10-08T15:32:52Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:32:52Z] positions: largest hi16-bucket (256 items) share 0.0082% at p49 (iteration 3, site 1, instr 6); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 4)
[2026-10-08T15:32:52Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 1 of 2048 at p0 (site 0, instr 4); saturated sources at one position 0 of 2048 at p0 (site 0, instr 4); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:32:52Z] p31-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 72 (bin 12631082) z_max +7.07 min 5 z_min -4.77 chi2/dof 1.50067 chi2_z +1450.08 top0.1% 0.17650% top0.5% 0.82781% top1% 1.60345%
[2026-10-08T15:32:52Z] p31-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.00051 largest 72 (item 0xc0bc2a) at +6.00 sigma smallest at -4.35; buckets64 chi2/dof 0.99725 largest 1976 (bucket 91211) at +4.35 sigma smallest at -4.35
[2026-10-08T15:32:54Z] p31-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 71 (bin 1936166) z_max +6.89 min 6 z_min -4.60 chi2/dof 1.50049 chi2_z +1449.59 top0.1% 0.17656% top0.5% 0.82800% top1% 1.60362%
[2026-10-08T15:32:54Z] p31-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00053 largest +5.86 sigma; buckets64 chi2/dof 0.99928 largest +4.51 smallest -4.06
[2026-10-08T15:32:54Z] 6-sigma p31-attack-f8 items buckets64 against the window density: largest bucket +4.35 sigma, smallest -4.35 -> within 6 sigma
[2026-10-08T15:32:54Z] hot-set p31-attack-f8 items (windowed control): f 0.1% S_f 0.17650% E_f(control) 0.17656% X_f -0.00006% X_f/f -0.0006 -> no hot set
[2026-10-08T15:32:54Z] hot-set p31-attack-f8 items (windowed control): f 0.5% S_f 0.82781% E_f(control) 0.82800% X_f -0.00018% X_f/f -0.0004 -> no hot set
[2026-10-08T15:32:54Z] hot-set p31-attack-f8 items (windowed control): f 1.0% S_f 1.60345% E_f(control) 1.60362% X_f -0.00017% X_f/f -0.0002 -> no hot set
[2026-10-08T15:32:54Z] hot-set p31-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:32:56Z] p31-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 2122700) z_max +5.83 min 7 z_min -4.42 chi2/dof 1.00053 chi2_z +1.53 top0.1% 0.16464% top0.5% 0.77456% top1% 1.50149%
[2026-10-08T15:32:56Z] hot-set p31-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17650% E_f(control) 0.16464% X_f +0.01186% X_f/f +0.1186 -> no hot set
[2026-10-08T15:32:56Z] hot-set p31-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.82781% E_f(control) 0.77456% X_f +0.05325% X_f/f +0.1065 -> no hot set
[2026-10-08T15:32:56Z] hot-set p31-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.60345% E_f(control) 1.50149% X_f +0.10197% X_f/f +0.1020 -> no hot set
[2026-10-08T15:32:56Z] hot-set p31-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:32:56Z] ratio p31-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9997x / 0.9998x / 0.9999x (gate 1.2x at 0.1%: within); over the FLAT control 1.0720x / 1.0687x / 1.0679x
[2026-10-08T15:32:57Z] attribution: hot threshold count >= 54 marks 25522 items (0.1521% of items) holding 1419831 reads (0.2645% of reads); top 8 items t=0xc0bc2a:72 t=0x3fd9ec:71 t=0x358e7a:71 t=0x30b1d3:69 t=0x1e8074:69 t=0xd8f5c9:68 t=0x3c2ebc:68 t=0x1a81fb:68
[2026-10-08T15:33:13Z] attribution: reads on hot items per position (flat expectation 0.1521% of each position's 4194304 reads); the 24 largest: p46(it2 s14)=0.480% p62(it3 s14)=0.477% p14(it0 s14)=0.475% p126(it7 s14)=0.475% p78(it4 s14)=0.473% p97(it6 s1)=0.472% p94(it5 s14)=0.471% p30(it1 s14)=0.470% p110(it6 s14)=0.469% p40(it2 s8)=0.469% p81(it5 s1)=0.468% p113(it7 s1)=0.467% p49(it3 s1)=0.466% p1(it0 s1)=0.466% p104(it6 s8)=0.466% p56(it3 s8)=0.466% p120(it7 s8)=0.466% p72(it4 s8)=0.465% p17(it1 s1)=0.465% p24(it1 s8)=0.463% p65(it4 s1)=0.463% p8(it0 s8)=0.462% p33(it2 s1)=0.462% p88(it5 s8)=0.461%
[2026-10-08T15:33:13Z] attribution: hot reads by iteration it0=0.264% it1=0.265% it2=0.265% it3=0.264% it4=0.265% it5=0.264% it6=0.265% it7=0.264% ; by site s0=0.236% s1=0.466% s2=0.235% s3=0.239% s4=0.237% s5=0.235% s6=0.236% s7=0.235% s8=0.465% s9=0.234% s10=0.235% s11=0.236% s12=0.234% s13=0.234% s14=0.474% s15=0.001%
[2026-10-08T15:33:13Z] site 0 (instr 4, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2363% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 1 (instr 6, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4660% (flat expectation 0.1521%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:33:13Z] site 2 (instr 12, src r2, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2351% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 3 (instr 15, src r5, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2392% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 4 (instr 17, src r0, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2368% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 5 (instr 19, src r4, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2348% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 6 (instr 20, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2359% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 7 (instr 22, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2345% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 8 (instr 30, src r3, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.4647% (flat expectation 0.1521%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:33:13Z] site 9 (instr 34, src r5, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2337% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 10 (instr 42, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2352% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 11 (instr 46, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.2363% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 12 (instr 53, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.2337% (flat expectation 0.1521%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:33:13Z] site 13 (instr 54, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2342% (flat expectation 0.1521%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:33:13Z] site 14 (instr 59, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.4737% (flat expectation 0.1521%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:33:13Z] site 15 (instr 60, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0013% (flat expectation 0.1521%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:33:13Z] attribution: item 0xc0bc2a (72 reads) read from positions p62(it3 s14)x4 p110(it6 s14)x3 p23(it1 s7)x3 p46(it2 s14)x3 p86(it5 s6)x3 p94(it5 s14)x3 p0(it0 s0)x2 p118(it7 s6)x2 p27(it1 s11)x2 p35(it2 s3)x2 p36(it2 s4)x2 p38(it2 s6)x2
[2026-10-08T15:33:13Z] attribution: item 0x3fd9ec (71 reads) read from positions p40(it2 s8)x4 p49(it3 s1)x4 p56(it3 s8)x4 p105(it6 s9)x3 p17(it1 s1)x3 p28(it1 s12)x3 p1(it0 s1)x3 p88(it5 s8)x3 p104(it6 s8)x3 p71(it4 s7)x2 p8(it0 s8)x2 p55(it3 s7)x2
[2026-10-08T15:33:13Z] attribution: item 0x358e7a (71 reads) read from positions p1(it0 s1)x4 p8(it0 s8)x4 p24(it1 s8)x4 p65(it4 s1)x3 p82(it5 s2)x3 p108(it6 s12)x3 p12(it0 s12)x2 p17(it1 s1)x2 p56(it3 s8)x2 p66(it4 s2)x2 p72(it4 s8)x2 p76(it4 s12)x2
[2026-10-08T15:33:13Z] attribution: item 0x30b1d3 (69 reads) read from positions p33(it2 s1)x5 p81(it5 s1)x5 p17(it1 s1)x4 p88(it5 s8)x4 p1(it0 s1)x3 p24(it1 s8)x3 p44(it2 s12)x3 p92(it5 s12)x3 p8(it0 s8)x2 p40(it2 s8)x2 p42(it2 s10)x2 p65(it4 s1)x2
[2026-10-08T15:33:13Z] attribution: item 0x1e8074 (69 reads) read from positions p113(it7 s1)x4 p17(it1 s1)x3 p33(it2 s1)x3 p40(it2 s8)x3 p57(it3 s9)x3 p65(it4 s1)x3 p88(it5 s8)x3 p101(it6 s5)x2 p120(it7 s8)x2 p24(it1 s8)x2 p8(it0 s8)x2 p48(it3 s0)x2
[2026-10-08T15:33:13Z] attribution: item 0xd8f5c9 (68 reads) read from positions p118(it7 s6)x4 p78(it4 s14)x3 p110(it6 s14)x3 p11(it0 s11)x2 p23(it1 s7)x2 p32(it2 s0)x2 p38(it2 s6)x2 p46(it2 s14)x2 p62(it3 s14)x2 p67(it4 s3)x2 p68(it4 s4)x2 p83(it5 s3)x2
[2026-10-08T15:33:13Z] attribution: item 0x3c2ebc (68 reads) read from positions p65(it4 s1)x4 p81(it5 s1)x3 p49(it3 s1)x3 p103(it6 s7)x2 p124(it7 s12)x2 p25(it1 s9)x2 p33(it2 s1)x2 p37(it2 s5)x2 p40(it2 s8)x2 p41(it2 s9)x2 p8(it0 s8)x2 p56(it3 s8)x2
[2026-10-08T15:33:13Z] attribution: item 0x1a81fb (68 reads) read from positions p113(it7 s1)x5 p56(it3 s8)x4 p104(it6 s8)x4 p17(it1 s1)x3 p28(it1 s12)x3 p2(it0 s2)x2 p5(it0 s5)x2 p25(it1 s9)x2 p26(it1 s10)x2 p39(it2 s7)x2 p57(it3 s9)x2 p81(it5 s1)x2
[2026-10-08T15:33:13Z] hottest item p31-attack-f8 (0xc0bc2a): 72 reads (0.0000% of all); predicted source none
[2026-10-08T15:33:13Z] program p31-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:33:15Z] program p32-attack-f8: epoch seed 13da69e4bc3b4af5591aaead41f6b16002dfab790760e2b97cd6f26c70ba5586 era seed d84a34744b2247e02e91a9222573f7ce70d94467458fbc1941765bbf1cdddd56 id e0ef2a8d99bbe50a attempt 0 class mx8-era0fe73018+sh256x27+state+rw op mix load=16 rotl=10 xor=9 add=7 mad=7 rotr=5 sub=4 shfl=3 mul=2 mulhi=1; era stride mul 0xdc6791fd rot 17 interleave [6, 11, 14, 15] windows 2:1:0 5:1:1 10:0:0 19:1:0 29:2:3 34:2:0 35:1:1 42:2:0 45:1:1 48:1:0 49:0:0 51:1:1 55:1:1 58:1:0 60:2:2 61:2:1
[2026-10-08T15:33:15Z] window layer: site item windows (first, items) (0x0,2^23) (0x800000,2^23) (0x0,2^24) (0x0,2^23) (0xc00000,2^22) (0x0,2^22) (0x800000,2^23) (0x0,2^22) (0x800000,2^23) (0x0,2^23) (0x0,2^24) (0x800000,2^23) (0x800000,2^23) (0x0,2^23) (0x800000,2^22) (0x400000,2^22); expected reads per item by quarter 36.000 28.000 32.000 32.000 (flat uniform 32.000)
[2026-10-08T15:33:47Z] 131072 warps (4194304 nonces) interpreted in 31.7 s (1.935 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:33:47Z] p32-attack-f8 distinct lines per hash: n 4194304 min 1008 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8702
[2026-10-08T15:33:47Z] p32-attack-f8 distinct items per hash: n 4194304 min 126 p1 128 median 128 p99 128 max 128 mean 127.9994
[2026-10-08T15:33:47Z] p32-attack-f8 distinct lines per warp: n 131072 min 32567 p1 32605 median 32637 p99 32664 max 32687 mean 32636.2704
[2026-10-08T15:33:47Z] p32-attack-f8 distinct items per warp: n 131072 min 4091 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4938
[2026-10-08T15:33:47Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:33:47Z] positions: largest hi16-bucket (256 items) share 0.0080% at p110 (iteration 6, site 14, instr 60); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p77 (site 13, instr 58)
[2026-10-08T15:33:47Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p109 (site 13, instr 58); saturated sources at one position 0 of 2048 at p0 (site 0, instr 2); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:33:47Z] p32-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 67 (bin 7535) z_max +6.19 min 6 z_min -4.60 chi2/dof 1.24990 chi2_z +723.79 top0.1% 0.17264% top0.5% 0.80903% top1% 1.56597%
[2026-10-08T15:33:47Z] p32-attack-f8 item histogram against the window density: full 2^24 chi2/dof 0.99967 largest 64 (item 0xdd0214) at +5.66 sigma smallest at -4.42; buckets64 chi2/dof 0.99997 largest 2012 (bucket 129128) at +5.20 sigma smallest at -4.46
[2026-10-08T15:33:49Z] p32-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 77 (bin 434422) z_max +7.95 min 5 z_min -4.77 chi2/dof 1.25040 chi2_z +725.23 top0.1% 0.17270% top0.5% 0.80938% top1% 1.56648%
[2026-10-08T15:33:49Z] p32-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00037 largest +6.83 sigma; buckets64 chi2/dof 1.00116 largest +5.25 smallest -4.24
[2026-10-08T15:33:49Z] 6-sigma p32-attack-f8 items buckets64 against the window density: largest bucket +5.20 sigma, smallest -4.46 -> within 6 sigma
[2026-10-08T15:33:49Z] hot-set p32-attack-f8 items (windowed control): f 0.1% S_f 0.17264% E_f(control) 0.17270% X_f -0.00006% X_f/f -0.0006 -> no hot set
[2026-10-08T15:33:49Z] hot-set p32-attack-f8 items (windowed control): f 0.5% S_f 0.80903% E_f(control) 0.80938% X_f -0.00035% X_f/f -0.0007 -> no hot set
[2026-10-08T15:33:49Z] hot-set p32-attack-f8 items (windowed control): f 1.0% S_f 1.56597% E_f(control) 1.56648% X_f -0.00051% X_f/f -0.0005 -> no hot set
[2026-10-08T15:33:49Z] hot-set p32-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:33:51Z] p32-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 65 (bin 2983295) z_max +5.83 min 7 z_min -4.42 chi2/dof 0.99973 chi2_z -0.79 top0.1% 0.16465% top0.5% 0.77438% top1% 1.50129%
[2026-10-08T15:33:51Z] hot-set p32-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.17264% E_f(control) 0.16465% X_f +0.00800% X_f/f +0.0800 -> no hot set
[2026-10-08T15:33:51Z] hot-set p32-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.80903% E_f(control) 0.77438% X_f +0.03465% X_f/f +0.0693 -> no hot set
[2026-10-08T15:33:51Z] hot-set p32-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.56597% E_f(control) 1.50129% X_f +0.06468% X_f/f +0.0647 -> no hot set
[2026-10-08T15:33:51Z] hot-set p32-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:33:51Z] ratio p32-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 0.9997x / 0.9996x / 0.9997x (gate 1.2x at 0.1%: within); over the FLAT control 1.0486x / 1.0447x / 1.0431x
[2026-10-08T15:33:52Z] attribution: hot threshold count >= 53 marks 23277 items (0.1387% of items) holding 1271372 reads (0.2368% of reads); top 8 items t=0x3eab39:67 t=0x2d0a69:67 t=0x17853d:67 t=0x16dccb:67 t=0x16774e:67 t=0x001d6f:67 t=0x3839e6:66 t=0x33420b:66
[2026-10-08T15:34:05Z] attribution: reads on hot items per position (flat expectation 0.1387% of each position's 4194304 reads); the 24 largest: p23(it1 s7)=0.722% p69(it4 s5)=0.721% p85(it5 s5)=0.720% p21(it1 s5)=0.719% p103(it6 s7)=0.719% p39(it2 s7)=0.719% p37(it2 s5)=0.719% p7(it0 s7)=0.718% p53(it3 s5)=0.716% p101(it6 s5)=0.716% p119(it7 s7)=0.716% p71(it4 s7)=0.715% p87(it5 s7)=0.715% p55(it3 s7)=0.711% p5(it0 s5)=0.710% p117(it7 s5)=0.710% p77(it4 s13)=0.367% p35(it2 s3)=0.366% p80(it5 s0)=0.366% p67(it4 s3)=0.366% p121(it7 s9)=0.364% p89(it5 s9)=0.364% p64(it4 s0)=0.364% p109(it6 s13)=0.364%
[2026-10-08T15:34:05Z] attribution: hot reads by iteration it0=0.236% it1=0.237% it2=0.237% it3=0.236% it4=0.238% it5=0.237% it6=0.237% it7=0.236% ; by site s0=0.360% s1=0.069% s2=0.215% s3=0.362% s4=0.068% s5=0.716% s6=0.068% s7=0.717% s8=0.068% s9=0.361% s10=0.214% s11=0.068% s12=0.069% s13=0.362% s14=0.068% s15=0.003%
[2026-10-08T15:34:05Z] site 0 (instr 2, src r6, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3599% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 1 (instr 5, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0691% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 2 (instr 10, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2150% (flat expectation 0.1387%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:34:05Z] site 3 (instr 19, src r7, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3617% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 4 (instr 29, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.0683% (flat expectation 0.1387%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:34:05Z] site 5 (instr 34, src r4, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.7164% (flat expectation 0.1387%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:34:05Z] site 6 (instr 35, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0679% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 7 (instr 42, src r0, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.7168% (flat expectation 0.1387%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:34:05Z] site 8 (instr 45, src r3, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0683% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 9 (instr 48, src r2, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3607% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 10 (instr 49, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.2141% (flat expectation 0.1387%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:34:05Z] site 11 (instr 51, src r7, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0684% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 12 (instr 55, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.0687% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0034% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 13 (instr 58, src r3, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.3622% (flat expectation 0.1387%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:34:05Z] site 14 (instr 60, src r0, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.0684% (flat expectation 0.1387%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0068% (window expectation 0.0061%)
[2026-10-08T15:34:05Z] site 15 (instr 61, src r5, k_off 2 offset 1, window 2^22 items): share of its reads into the top 0.1% 0.0031% (flat expectation 0.1387%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:34:05Z] attribution: item 0x3eab39 (67 reads) read from positions p5(it0 s5)x4 p64(it4 s0)x4 p16(it1 s0)x3 p37(it2 s5)x3 p90(it5 s10)x3 p3(it0 s3)x2 p48(it3 s0)x2 p51(it3 s3)x2 p80(it5 s0)x2 p85(it5 s5)x2 p87(it5 s7)x2 p0(it0 s0)x2
[2026-10-08T15:34:05Z] attribution: item 0x2d0a69 (67 reads) read from positions p23(it1 s7)x4 p67(it4 s3)x4 p87(it5 s7)x4 p89(it5 s9)x4 p32(it2 s0)x3 p39(it2 s7)x3 p85(it5 s5)x3 p112(it7 s0)x3 p7(it0 s7)x2 p29(it1 s13)x2 p37(it2 s5)x2 p53(it3 s5)x2
[2026-10-08T15:34:05Z] attribution: item 0x17853d (67 reads) read from positions p61(it3 s13)x5 p103(it6 s7)x5 p7(it0 s7)x3 p37(it2 s5)x3 p39(it2 s7)x3 p2(it0 s2)x2 p5(it0 s5)x2 p16(it1 s0)x2 p48(it3 s0)x2 p51(it3 s3)x2 p57(it3 s9)x2 p73(it4 s9)x2
[2026-10-08T15:34:05Z] attribution: item 0x16dccb (67 reads) read from positions p119(it7 s7)x5 p19(it1 s3)x4 p53(it3 s5)x4 p103(it6 s7)x4 p13(it0 s13)x3 p55(it3 s7)x3 p71(it4 s7)x3 p105(it6 s9)x3 p0(it0 s0)x2 p5(it0 s5)x2 p35(it2 s3)x2 p42(it2 s10)x2
[2026-10-08T15:34:05Z] attribution: item 0x16774e (67 reads) read from positions p23(it1 s7)x6 p87(it5 s7)x5 p21(it1 s5)x4 p103(it6 s7)x4 p9(it0 s9)x3 p37(it2 s5)x3 p55(it3 s7)x3 p69(it4 s5)x3 p96(it6 s0)x3 p125(it7 s13)x3 p51(it3 s3)x2 p73(it4 s9)x2
[2026-10-08T15:34:05Z] attribution: item 0x001d6f (67 reads) read from positions p53(it3 s5)x5 p5(it0 s5)x3 p23(it1 s7)x3 p25(it1 s9)x3 p87(it5 s7)x3 p101(it6 s5)x3 p115(it7 s3)x3 p119(it7 s7)x3 p7(it0 s7)x2 p9(it0 s9)x2 p32(it2 s0)x2 p37(it2 s5)x2
[2026-10-08T15:34:05Z] attribution: item 0x3839e6 (66 reads) read from positions p101(it6 s5)x6 p119(it7 s7)x4 p37(it2 s5)x4 p69(it4 s5)x3 p71(it4 s7)x3 p7(it0 s7)x2 p13(it0 s13)x2 p16(it1 s0)x2 p21(it1 s5)x2 p25(it1 s9)x2 p32(it2 s0)x2 p39(it2 s7)x2
[2026-10-08T15:34:05Z] attribution: item 0x33420b (66 reads) read from positions p103(it6 s7)x4 p115(it7 s3)x3 p23(it1 s7)x3 p37(it2 s5)x3 p71(it4 s7)x3 p0(it0 s0)x3 p90(it5 s10)x2 p29(it1 s13)x2 p39(it2 s7)x2 p48(it3 s0)x2 p57(it3 s9)x2 p69(it4 s5)x2
[2026-10-08T15:34:05Z] hottest item p32-attack-f8 (0x3eab39): 67 reads (0.0000% of all); predicted source none
[2026-10-08T15:34:05Z] program p32-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) clear; hot-set clear; verdict PASS (no test fired)
[2026-10-08T15:34:08Z] program p33-attack-f8: epoch seed 89a8f48f468a8673ddfc9624d7f7f9f6355e7a827e94e2093ff7832640243afd era seed 2dd8c4a6f6b3216c4bc88647a94277d32a8a35cafe374968935311c1d6150c6d id 8ad82116c871037f attempt 0 class mx8-erae8a44951+sh256x27+state+rw op mix load=16 add=11 rotl=9 mad=7 rotr=6 shfl=6 xor=5 sub=2 mul=1 mulhi=1; era stride mul 0x3e376d09 rot 6 interleave [0, 3, 4, 9] windows 6:2:2 15:0:0 16:2:3 19:2:2 22:0:0 24:0:0 25:2:0 28:0:0 30:1:0 32:1:1 36:0:0 37:0:0 48:0:0 54:0:0 58:2:2 60:0:0
[2026-10-08T15:34:08Z] window layer: site item windows (first, items) (0x800000,2^22) (0x0,2^24) (0xc00000,2^22) (0x800000,2^22) (0x0,2^24) (0x0,2^24) (0x0,2^22) (0x0,2^24) (0x0,2^23) (0x800000,2^23) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x0,2^24) (0x800000,2^22) (0x0,2^24); expected reads per item by quarter 30.000 22.000 46.000 30.000 (flat uniform 32.000)
[2026-10-08T15:34:55Z] 131072 warps (4194304 nonces) interpreted in 46.9 s (2.865 ms per warp per thread); Epoch::hash_warp agreement on 195 warps: 0 mismatches
[2026-10-08T15:34:55Z] p33-attack-f8 distinct lines per hash: n 4194304 min 1013 p1 1023 median 1024 p99 1024 max 1024 mean 1023.8710
[2026-10-08T15:34:55Z] p33-attack-f8 distinct items per hash: n 4194304 min 127 p1 128 median 128 p99 128 max 128 mean 127.9995
[2026-10-08T15:34:55Z] p33-attack-f8 distinct lines per warp: n 131072 min 32576 p1 32605 median 32636 p99 32664 max 32682 mean 32636.0161
[2026-10-08T15:34:55Z] p33-attack-f8 distinct items per warp: n 131072 min 4089 p1 4093 median 4096 p99 4096 max 4096 mean 4095.4575
[2026-10-08T15:34:55Z] uniform expectation: distinct lines per hash 1023.875 of 1024 reads, per warp 32640.3 of 32768 reads (2^22 lines); distinct items per hash 127.9995 of 128, per warp 4095.50 of 4096 (2^24 items)
[2026-10-08T15:34:55Z] positions: largest hi16-bucket (256 items) share 0.0080% at p48 (iteration 3, site 0, instr 6); windowed expectation 0.0061%; largest saturated-source share 0.0000% at p0 (site 0, instr 6)
[2026-10-08T15:34:55Z] acceptance-style (2,048 evaluations, the rule's sample size): the most repeated address at one position 2 of 2048 at p13 (site 13, instr 54); saturated sources at one position 0 of 2048 at p0 (site 0, instr 6); uniform expectation: repeats 1 to 2, saturated 0
[2026-10-08T15:34:55Z] p33-attack-f8 item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 95 (bin 16252935) z_max +11.14 min 2 z_min -5.30 chi2/dof 3.60197 chi2_z +7536.10 top0.1% 0.21125% top0.5% 0.99112% top1% 1.91961%
[2026-10-08T15:34:55Z] p33-attack-f8 item histogram against the window density: full 2^24 chi2/dof 1.24198 largest 95 (item 0xf80007) at +11.87 sigma smallest at -4.56; buckets64 chi2/dof 1.07223 largest 2227 (bucket 253952) at +7.01 sigma smallest at -4.45
[2026-10-08T15:34:57Z] p33-attack-f8 WINDOWED CONTROL item histogram (flat): bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 83 (bin 8804640) z_max +9.02 min 3 z_min -5.13 chi2/dof 3.37499 chi2_z +6878.70 top0.1% 0.21033% top0.5% 0.98808% top1% 1.91286%
[2026-10-08T15:34:57Z] p33-attack-f8 WINDOWED CONTROL against the window density: full chi2/dof 1.00007 largest +6.02 sigma; buckets64 chi2/dof 0.99960 largest +4.61 smallest -4.34
[2026-10-08T15:34:57Z] 6-sigma p33-attack-f8 items buckets64 against the window density: largest bucket +7.01 sigma, smallest -4.45 -> FLAGGED (beyond 6 sigma)
[2026-10-08T15:34:57Z] hot-set p33-attack-f8 items (windowed control): f 0.1% S_f 0.21125% E_f(control) 0.21033% X_f +0.00092% X_f/f +0.0092 -> no hot set
[2026-10-08T15:34:57Z] hot-set p33-attack-f8 items (windowed control): f 0.5% S_f 0.99112% E_f(control) 0.98808% X_f +0.00304% X_f/f +0.0061 -> no hot set
[2026-10-08T15:34:57Z] hot-set p33-attack-f8 items (windowed control): f 1.0% S_f 1.91961% E_f(control) 1.91286% X_f +0.00675% X_f/f +0.0067 -> no hot set
[2026-10-08T15:34:57Z] hot-set p33-attack-f8 items (windowed control): verdict clear
[2026-10-08T15:34:59Z] p33-attack-f8 FLAT CONTROL item histogram: bins 16777216 reads 536870912 mean 32.000 sigma 5.657 max 67 (bin 10780823) z_max +6.19 min 7 z_min -4.42 chi2/dof 1.00074 chi2_z +2.13 top0.1% 0.16462% top0.5% 0.77454% top1% 1.50158%
[2026-10-08T15:34:59Z] hot-set p33-attack-f8 items (flat control, the auditor's first view): f 0.1% S_f 0.21125% E_f(control) 0.16462% X_f +0.04664% X_f/f +0.4664 -> no hot set
[2026-10-08T15:34:59Z] hot-set p33-attack-f8 items (flat control, the auditor's first view): f 0.5% S_f 0.99112% E_f(control) 0.77454% X_f +0.21658% X_f/f +0.4332 -> no hot set
[2026-10-08T15:34:59Z] hot-set p33-attack-f8 items (flat control, the auditor's first view): f 1.0% S_f 1.91961% E_f(control) 1.50158% X_f +0.41803% X_f/f +0.4180 -> no hot set
[2026-10-08T15:34:59Z] hot-set p33-attack-f8 items (flat control, the auditor's first view): verdict clear
[2026-10-08T15:34:59Z] ratio p33-attack-f8: top 0.1% / 0.5% / 1% share over the WINDOW-MODEL control 1.0044x / 1.0031x / 1.0035x (gate 1.2x at 0.1%: within); over the FLAT control 1.2833x / 1.2796x / 1.2784x
[2026-10-08T15:35:00Z] attribution: hot threshold count >= 65 marks 21599 items (0.1287% of items) holding 1447596 reads (0.2696% of reads); top 8 items t=0xf80007:95 t=0xb0c016:86 t=0xce0001:85 t=0xc00003:85 t=0xf80003:84 t=0xf40002:84 t=0xf20001:84 t=0xef0002:84
[2026-10-08T15:35:21Z] attribution: reads on hot items per position (flat expectation 0.1287% of each position's 4194304 reads); the 24 largest: p46(it2 s14)=0.703% p51(it3 s3)=0.700% p14(it0 s14)=0.700% p126(it7 s14)=0.698% p99(it6 s3)=0.698% p35(it2 s3)=0.697% p83(it5 s3)=0.697% p3(it0 s3)=0.696% p110(it6 s14)=0.696% p115(it7 s3)=0.696% p96(it6 s0)=0.696% p112(it7 s0)=0.695% p32(it2 s0)=0.694% p16(it1 s0)=0.694% p78(it4 s14)=0.694% p62(it3 s14)=0.694% p30(it1 s14)=0.693% p48(it3 s0)=0.693% p19(it1 s3)=0.691% p80(it5 s0)=0.691% p94(it5 s14)=0.691% p64(it4 s0)=0.691% p67(it4 s3)=0.690% p0(it0 s0)=0.689%
[2026-10-08T15:35:21Z] attribution: hot reads by iteration it0=0.270% it1=0.269% it2=0.270% it3=0.269% it4=0.268% it5=0.270% it6=0.270% it7=0.270% ; by site s0=0.693% s1=0.186% s2=0.181% s3=0.696% s4=0.186% s5=0.186% s6=0.000% s7=0.187% s8=0.000% s9=0.372% s10=0.186% s11=0.186% s12=0.186% s13=0.187% s14=0.696% s15=0.187%
[2026-10-08T15:35:21Z] site 0 (instr 6, src r3, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.6929% (flat expectation 0.1287%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:35:21Z] site 1 (instr 15, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1859% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 2 (instr 16, src r2, k_off 2 offset 3, window 2^22 items): share of its reads into the top 0.1% 0.1808% (flat expectation 0.1287%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0068% (window expectation 0.0061%)
[2026-10-08T15:35:21Z] site 3 (instr 19, src r0, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.6957% (flat expectation 0.1287%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:35:21Z] site 4 (instr 22, src r3, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1864% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 5 (instr 24, src r1, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1858% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 6 (instr 25, src r7, k_off 2 offset 0, window 2^22 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1287%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0066% (window expectation 0.0061%)
[2026-10-08T15:35:21Z] site 7 (instr 28, src r5, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1865% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 8 (instr 30, src r3, k_off 1 offset 0, window 2^23 items): share of its reads into the top 0.1% 0.0000% (flat expectation 0.1287%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:35:21Z] site 9 (instr 32, src r1, k_off 1 offset 1, window 2^23 items): share of its reads into the top 0.1% 0.3723% (flat expectation 0.1287%); index entropy 14.999 bits of 15 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0035% (window expectation 0.0031%)
[2026-10-08T15:35:21Z] site 10 (instr 36, src r0, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1857% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 11 (instr 37, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1862% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 12 (instr 48, src r6, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1865% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 13 (instr 54, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1865% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0019% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] site 14 (instr 58, src r5, k_off 2 offset 2, window 2^22 items): share of its reads into the top 0.1% 0.6961% (flat expectation 0.1287%); index entropy 14.000 bits of 14 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0067% (window expectation 0.0061%)
[2026-10-08T15:35:21Z] site 15 (instr 60, src r7, k_off 0 offset 0, window 2^24 items): share of its reads into the top 0.1% 0.1868% (flat expectation 0.1287%); index entropy 15.999 bits of 16 uniform-on-window; saturated source 0.0000%; largest 256-item bucket 0.0018% (window expectation 0.0015%)
[2026-10-08T15:35:21Z] attribution: item 0xf80007 (95 reads) read from positions p82(it5 s2)x12 p34(it2 s2)x10 p50(it3 s2)x8 p2(it0 s2)x7 p18(it1 s2)x7 p66(it4 s2)x7 p98(it6 s2)x7 p114(it7 s2)x6 p1(it0 s1)x2 p29(it1 s13)x2 p73(it4 s9)x2 p23(it1 s7)x2
[2026-10-08T15:35:21Z] attribution: item 0xb0c016 (86 reads) read from positions p19(it1 s3)x6 p48(it3 s0)x5 p78(it4 s14)x5 p126(it7 s14)x4 p80(it5 s0)x4 p0(it0 s0)x3 p124(it7 s12)x3 p30(it1 s14)x3 p46(it2 s14)x3 p64(it4 s0)x3 p16(it1 s0)x3 p93(it5 s13)x3
[2026-10-08T15:35:21Z] attribution: item 0xce0001 (85 reads) read from positions p18(it1 s2)x11 p114(it7 s2)x8 p66(it4 s2)x8 p82(it5 s2)x8 p2(it0 s2)x6 p34(it2 s2)x6 p98(it6 s2)x3 p5(it0 s5)x2 p42(it2 s10)x2 p50(it3 s2)x2 p60(it3 s12)x2 p108(it6 s12)x2
[2026-10-08T15:35:21Z] attribution: item 0xc00003 (85 reads) read from positions p2(it0 s2)x10 p114(it7 s2)x9 p18(it1 s2)x8 p34(it2 s2)x8 p50(it3 s2)x7 p66(it4 s2)x6 p82(it5 s2)x6 p98(it6 s2)x4 p92(it5 s12)x3 p44(it2 s12)x2 p119(it7 s7)x2 p11(it0 s11)x1
[2026-10-08T15:35:21Z] attribution: item 0xf80003 (84 reads) read from positions p98(it6 s2)x11 p2(it0 s2)x10 p114(it7 s2)x10 p50(it3 s2)x7 p18(it1 s2)x6 p66(it4 s2)x6 p82(it5 s2)x5 p34(it2 s2)x3 p41(it2 s9)x2 p95(it5 s15)x2 p127(it7 s15)x2 p11(it0 s11)x1
[2026-10-08T15:35:21Z] attribution: item 0xf40002 (84 reads) read from positions p34(it2 s2)x8 p82(it5 s2)x8 p98(it6 s2)x8 p2(it0 s2)x7 p18(it1 s2)x7 p50(it3 s2)x7 p114(it7 s2)x7 p66(it4 s2)x5 p119(it7 s7)x3 p89(it5 s9)x2 p111(it6 s15)x2 p7(it0 s7)x1
[2026-10-08T15:35:21Z] attribution: item 0xf20001 (84 reads) read from positions p82(it5 s2)x9 p18(it1 s2)x8 p114(it7 s2)x7 p34(it2 s2)x7 p98(it6 s2)x6 p66(it4 s2)x4 p2(it0 s2)x3 p50(it3 s2)x3 p57(it3 s9)x3 p85(it5 s5)x3 p9(it0 s9)x2 p79(it4 s15)x2
[2026-10-08T15:35:21Z] attribution: item 0xef0002 (84 reads) read from positions p34(it2 s2)x11 p18(it1 s2)x10 p98(it6 s2)x10 p50(it3 s2)x9 p2(it0 s2)x7 p82(it5 s2)x7 p66(it4 s2)x5 p17(it1 s1)x2 p31(it1 s15)x2 p52(it3 s4)x2 p114(it7 s2)x2 p121(it7 s9)x2
[2026-10-08T15:35:21Z] hottest item p33-attack-f8 (0xf80007): 95 reads (0.0000% of all); predicted source none
[2026-10-08T15:35:21Z] program p33-attack-f8 DONE: 4194304 nonces; 6-sigma (windowed) FLAGGED; hot-set clear; verdict FLAGGED
[2026-10-08T15:35:21Z] SEED p18-attack-f8 id b9ea7b20b6c5bef4 attempt 0 S0.1 0.25140% null0.1 0.21812% ratio 1.1526x S1.0 2.20526% null1.0 1.98765% hottest 0x01a37e:105 source none by-site s0=0.000% s1=0.137% s2=0.000% s3=0.138% s4=0.138% s5=0.551% s6=0.549% s7=0.139% s8=0.000% s9=0.000% s10=1.586% s11=0.000% s12=0.000% s13=0.000% s14=0.548% s15=0.549%
[2026-10-08T15:35:21Z] SEED p19-attack-f8 id fa128bbce81d993d attempt 1 S0.1 0.20310% null0.1 0.20291% ratio 1.0010x S1.0 1.84228% null1.0 1.84117% hottest 0x305ac0:81 source none by-site s0=0.307% s1=0.005% s2=0.002% s3=0.153% s4=0.002% s5=0.154% s6=0.154% s7=0.005% s8=0.154% s9=0.610% s10=0.312% s11=0.310% s12=0.614% s13=0.306% s14=0.306% s15=0.002%
[2026-10-08T15:35:21Z] SEED p20-attack-f8 id 0aa00969fcd4c14c attempt 0 S0.1 0.18453% null0.1 0.18447% ratio 1.0003x S1.0 1.68314% null1.0 1.68301% hottest 0x371595:76 source none by-site s0=0.369% s1=0.368% s2=0.193% s3=0.193% s4=0.194% s5=0.195% s6=0.368% s7=0.036% s8=0.369% s9=0.366% s10=0.367% s11=0.193% s12=0.018% s13=0.035% s14=0.367% s15=0.017%
[2026-10-08T15:35:21Z] SEED p21-attack-f8 id a730af697217c12b attempt 0 S0.1 0.21425% null0.1 0.21039% ratio 1.0183x S1.0 1.93913% null1.0 1.91270% hottest 0x007dce:87 source none by-site s0=0.000% s1=0.175% s2=0.345% s3=0.000% s4=0.173% s5=0.693% s6=0.000% s7=0.345% s8=0.347% s9=0.692% s10=0.856% s11=0.000% s12=0.171% s13=0.172% s14=0.000% s15=0.172%
[2026-10-08T15:35:21Z] SEED p22-attack-f8 id 1faa9e3199040c5b attempt 0 S0.1 0.18693% null0.1 0.18806% ratio 0.9940x S1.0 1.69611% null1.0 1.70626% hottest 0x636fcc:76 source none by-site s0=0.159% s1=0.272% s2=0.038% s3=0.160% s4=0.559% s5=0.159% s6=0.076% s7=0.076% s8=0.000% s9=0.159% s10=0.159% s11=0.280% s12=0.559% s13=0.159% s14=0.038% s15=0.278%
[2026-10-08T15:35:21Z] SEED p23-attack-f8 id cb2341129841e67d attempt 0 S0.1 0.21801% null0.1 0.21799% ratio 1.0001x S1.0 1.98551% null1.0 1.98537% hottest 0x6149d0:91 source none by-site s0=0.777% s1=0.000% s2=0.000% s3=0.195% s4=0.389% s5=0.000% s6=0.000% s7=0.000% s8=0.777% s9=0.000% s10=0.000% s11=0.194% s12=0.391% s13=0.778% s14=0.387% s15=0.777%
[2026-10-08T15:35:21Z] SEED p24-attack-f8 id fd5d5b3afe0515ee attempt 1 S0.1 0.17277% null0.1 0.17269% ratio 1.0005x S1.0 1.56643% null1.0 1.56611% hottest 0x7da697:72 source none by-site s0=0.214% s1=0.358% s2=0.073% s3=0.359% s4=0.359% s5=0.069% s6=0.362% s7=0.214% s8=0.216% s9=0.216% s10=0.715% s11=0.214% s12=0.070% s13=0.069% s14=0.072% s15=0.215%
[2026-10-08T15:35:21Z] SEED p25-attack-f8 id 8b83f49ba0b49e23 attempt 0 S0.1 0.20288% null0.1 0.20287% ratio 1.0000x S1.0 1.84118% null1.0 1.84119% hottest 0x42cf6d:86 source none by-site s0=0.152% s1=0.304% s2=0.152% s3=0.151% s4=0.000% s5=0.151% s6=0.151% s7=0.601% s8=0.152% s9=0.302% s10=0.152% s11=0.006% s12=0.600% s13=0.000% s14=0.151% s15=0.302%
[2026-10-08T15:35:21Z] SEED p26-attack-f8 id f016098c74eae0f7 attempt 0 S0.1 0.18752% null0.1 0.18749% ratio 1.0002x S1.0 1.69734% null1.0 1.69739% hottest 0xdb54fa:74 source none by-site s0=0.331% s1=0.166% s2=0.167% s3=0.334% s4=0.002% s5=0.167% s6=0.167% s7=0.331% s8=0.168% s9=0.167% s10=0.165% s11=0.004% s12=0.167% s13=0.167% s14=0.657% s15=0.166%
[2026-10-08T15:35:21Z] SEED p27-attack-f8 id ce3854c18fb6f409 attempt 2 S0.1 0.21036% null0.1 0.21034% ratio 1.0001x S1.0 1.91300% null1.0 1.91249% hottest 0xeea162:85 source none by-site s0=0.000% s1=0.349% s2=0.697% s3=0.000% s4=0.698% s5=0.174% s6=0.699% s7=0.174% s8=0.175% s9=0.000% s10=0.173% s11=0.000% s12=0.173% s13=0.349% s14=0.000% s15=0.348%
[2026-10-08T15:35:21Z] SEED p28-attack-f8 id 3734c9f7316ac786 attempt 0 S0.1 0.18751% null0.1 0.18740% ratio 1.0006x S1.0 1.69725% null1.0 1.69708% hottest 0x5658ad:78 source none by-site s0=0.005% s1=0.167% s2=0.657% s3=0.167% s4=0.167% s5=0.167% s6=0.005% s7=0.326% s8=0.005% s9=0.168% s10=0.000% s11=0.658% s12=0.005% s13=0.167% s14=0.662% s15=0.005%
[2026-10-08T15:35:21Z] SEED p29-attack-f8 id 2e84da6ea8f3a4d3 attempt 0 S0.1 0.17673% null0.1 0.17642% ratio 1.0017x S1.0 1.60495% null1.0 1.60347% hottest 0x3f05de:72 source none by-site s0=0.472% s1=0.244% s2=0.242% s3=0.238% s4=0.239% s5=0.238% s6=0.239% s7=0.240% s8=0.244% s9=0.238% s10=0.244% s11=0.241% s12=0.237% s13=0.240% s14=0.483% s15=0.243%
[2026-10-08T15:35:21Z] SEED p30-attack-f8 id 40906acb1cdfc01c attempt 0 S0.1 0.22583% null0.1 0.22558% ratio 1.0011x S1.0 2.05856% null1.0 2.05710% hottest 0x164439:91 source none by-site s0=0.635% s1=0.159% s2=0.632% s3=0.000% s4=0.158% s5=0.639% s6=0.159% s7=0.157% s8=0.321% s9=0.158% s10=0.159% s11=0.159% s12=0.000% s13=0.158% s14=0.159% s15=0.317%
[2026-10-08T15:35:21Z] SEED p31-attack-f8 id 31cac1b1f738a5fe attempt 0 S0.1 0.17650% null0.1 0.17656% ratio 0.9997x S1.0 1.60345% null1.0 1.60362% hottest 0xc0bc2a:72 source none by-site s0=0.236% s1=0.466% s2=0.235% s3=0.239% s4=0.237% s5=0.235% s6=0.236% s7=0.235% s8=0.465% s9=0.234% s10=0.235% s11=0.236% s12=0.234% s13=0.234% s14=0.474% s15=0.001%
[2026-10-08T15:35:21Z] SEED p32-attack-f8 id e0ef2a8d99bbe50a attempt 0 S0.1 0.17264% null0.1 0.17270% ratio 0.9997x S1.0 1.56597% null1.0 1.56648% hottest 0x3eab39:67 source none by-site s0=0.360% s1=0.069% s2=0.215% s3=0.362% s4=0.068% s5=0.716% s6=0.068% s7=0.717% s8=0.068% s9=0.361% s10=0.214% s11=0.068% s12=0.069% s13=0.362% s14=0.068% s15=0.003%
[2026-10-08T15:35:21Z] SEED p33-attack-f8 id 8ad82116c871037f attempt 0 S0.1 0.21125% null0.1 0.21033% ratio 1.0044x S1.0 1.91961% null1.0 1.91286% hottest 0xf80007:95 source none by-site s0=0.693% s1=0.186% s2=0.181% s3=0.696% s4=0.186% s5=0.186% s6=0.000% s7=0.187% s8=0.000% s9=0.372% s10=0.186% s11=0.186% s12=0.186% s13=0.187% s14=0.696% s15=0.187%
[2026-10-08T15:35:21Z] CENSUS PASS: 16 seeds at 4194304 nonces, 0 over 1.2x of the window model at the top 0.1%
[2026-10-08T15:35:21Z] SEED CENSUS: 16 programs at 4194304 nonces each: hot set (X_f >= f, windowed control) in 0; top 0.1% beyond 1.2x of the window-model control in 0; windowed buckets64 beyond 6 sigma in 4; acceptance-style most-repeated address at one position >= 3 / 11 / 21 of 2048 in 0 / 0 / 0; saturated sources at one position >= 3 / 11 / 21 of 2048 in 0 / 0 / 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p18-attack-f8 id b9ea7b20b6c5bef4 attempt 0 ratio_0.1% window 1.153x flat 1.528x X_0.1% +0.0333% X_1% +0.2176% hot 0 z_w +22.9 max bucket share 0.0206% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p21-attack-f8 id a730af697217c12b attempt 0 ratio_0.1% window 1.018x flat 1.302x X_0.1% +0.0039% X_1% +0.0264% hot 0 z_w +8.0 max bucket share 0.0110% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p33-attack-f8 id 8ad82116c871037f attempt 0 ratio_0.1% window 1.004x flat 1.283x X_0.1% +0.0009% X_1% +0.0067% hot 0 z_w +7.0 max bucket share 0.0080% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p29-attack-f8 id 2e84da6ea8f3a4d3 attempt 0 ratio_0.1% window 1.002x flat 1.074x X_0.1% +0.0003% X_1% +0.0015% hot 0 z_w +4.3 max bucket share 0.0078% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p30-attack-f8 id 40906acb1cdfc01c attempt 0 ratio_0.1% window 1.001x flat 1.372x X_0.1% +0.0003% X_1% +0.0015% hot 0 z_w +4.5 max bucket share 0.0078% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p19-attack-f8 id fa128bbce81d993d attempt 1 ratio_0.1% window 1.001x flat 1.234x X_0.1% +0.0002% X_1% +0.0011% hot 0 z_w +4.9 max bucket share 0.0081% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p28-attack-f8 id 3734c9f7316ac786 attempt 0 ratio_0.1% window 1.001x flat 1.139x X_0.1% +0.0001% X_1% +0.0002% hot 0 z_w +4.5 max bucket share 0.0080% max sat share 0.0000% acc addr 1 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p24-attack-f8 id fd5d5b3afe0515ee attempt 1 ratio_0.1% window 1.001x flat 1.050x X_0.1% +0.0001% X_1% +0.0003% hot 0 z_w +4.6 max bucket share 0.0078% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p20-attack-f8 id 0aa00969fcd4c14c attempt 0 ratio_0.1% window 1.000x flat 1.121x X_0.1% +0.0001% X_1% +0.0001% hot 0 z_w +4.5 max bucket share 0.0082% max sat share 0.0000% acc addr 2 acc sat 0
[2026-10-08T15:35:21Z] SEED CENSUS top: p26-attack-f8 id f016098c74eae0f7 attempt 0 ratio_0.1% window 1.000x flat 1.139x X_0.1% +0.0000% X_1% -0.0001% hot 0 z_w +4.4 max bucket share 0.0078% max sat share 0.0000% acc addr 2 acc sat 0
Terminated
lease: released 8 pool cores after 1619 s, exit 0

1
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/run.txt Normal file
View file

@ -0,0 +1 @@
# pack hl-v6-rw class mx8+sh256x27+state+rw epoch edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 day 69676e65756d2d6461792ffa50000000000000 era edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 state /srv/builds/v6-census/packs/node1-state.igsd1 day-index 20730 era-widths 4 bin 50903d300aa46f20 f8 9ba2210b5d14376b host igneum-build-3 start 2026-10-08T15:08:15Z

67
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/show.txt Normal file
View file

@ -0,0 +1,67 @@
seed "igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07" generator v5 class mx8-erad810f22d+sh256x27+state+rw attempt 0 program id 30628f8adcf6035e seed words 667d0fbd 7b8e5963 31c67e5e 4529ddc6 ef19d6d8 accf6211 da0aed32 abc6df31
op mix load=16 xor=11 rotl=8 add=6 rotr=6 mul=5 mad=4 shfl=4 sub=4 loads/hash 128 bytes/hash 512
era d810f22d (igneum-era/0/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07): width 4 B, stride mul 0x9ad30d99 rot 29, interleave [0, 2, 12, 13], windows (site:shrink:offset) 7:2:1 8:1:1 9:1:1 10:1:1 11:0:0 13:1:1 29:0:0 30:2:2 31:1:1 44:1:1 46:2:0 47:0:0 52:0:0 56:0:0 58:2:0 63:1:1
0: shfl dst=4 src=5 src2=7 imm=0xea86e152 imm2=0x5810667a rot=27 bit=13 mask=2
1: add dst=7 src=0 src2=6 imm=0xbaab6229 imm2=0xed861989 rot=26 bit=22 mask=4
2: sub dst=3 src=6 src2=2 imm=0x5b623116 imm2=0xff12e5b2 rot=12 bit=24 mask=1
3: rotr dst=4 src=1 src2=1 imm=0xe99741c7 imm2=0xf5fa5009 rot=1 bit=21 mask=1
4: rotl dst=2 src=0 src2=4 imm=0x673c2157 imm2=0xee02465f rot=20 bit=22 mask=2
5: xor dst=4 src=7 src2=7 imm=0x946f7818 imm2=0x45d3399e rot=9 bit=2 mask=16
6: shfl dst=0 src=2 src2=4 imm=0x5f6a0ed2 imm2=0x7043a636 rot=19 bit=31 mask=8
7: load dst=6 src=7 src2=1 imm=0x5c61dcf7 imm2=0x7466aa40 rot=19 bit=9 mask=2
8: load dst=1 src=4 src2=5 imm=0x85668475 imm2=0xdb8cc483 rot=29 bit=7 mask=4
9: load dst=1 src=2 src2=4 imm=0x4454980f imm2=0xebd31581 rot=10 bit=28 mask=16
10: load dst=7 src=0 src2=2 imm=0xe075297c imm2=0x5779c44c rot=10 bit=22 mask=2
11: load dst=7 src=1 src2=6 imm=0x65aa4311 imm2=0x4fe48ea9 rot=15 bit=9 mask=1
12: rotr dst=5 src=4 src2=2 imm=0x1383d3ad imm2=0xf3094b29 rot=8 bit=9 mask=2
13: load dst=7 src=6 src2=2 imm=0xed8a496f imm2=0x3072c3c6 rot=28 bit=19 mask=8
14: mad dst=6 src=5 src2=0 imm=0x8b965b57 imm2=0xcfeca6c1 rot=12 bit=27 mask=16
15: sub dst=3 src=5 src2=3 imm=0x877c7586 imm2=0xa9cb2a03 rot=2 bit=29 mask=1
16: xor dst=2 src=7 src2=3 imm=0xb740221a imm2=0x89d38d6d rot=6 bit=31 mask=8
17: mad dst=0 src=6 src2=1 imm=0x26f3ad8a imm2=0x27256f15 rot=18 bit=5 mask=2
18: xor dst=7 src=5 src2=7 imm=0xf66e7017 imm2=0xb9e3577e rot=9 bit=12 mask=16
19: mad dst=7 src=0 src2=5 imm=0x849ae6ee imm2=0x02b358f9 rot=24 bit=18 mask=8
20: rotr dst=6 src=7 src2=6 imm=0x52334d12 imm2=0x8c9f0ef8 rot=11 bit=23 mask=4
21: xor dst=2 src=6 src2=5 imm=0xb1871e63 imm2=0xb2e40191 rot=5 bit=13 mask=4
22: add dst=6 src=5 src2=4 imm=0x97df29e4 imm2=0xe60fea84 rot=11 bit=16 mask=2
23: rotl dst=1 src=0 src2=3 imm=0x8f30d21d imm2=0x2df685a0 rot=31 bit=0 mask=16
24: mul dst=6 src=1 src2=2 imm=0xd2c4025f imm2=0x5269eb4d rot=31 bit=21 mask=16
25: mul dst=2 src=6 src2=5 imm=0x659fc3d3 imm2=0x9cec0e12 rot=6 bit=17 mask=4
26: add dst=7 src=0 src2=1 imm=0xd89ef484 imm2=0x20be3846 rot=12 bit=9 mask=16
27: xor dst=4 src=5 src2=7 imm=0xb48420ae imm2=0x3d1f2485 rot=14 bit=28 mask=1
28: xor dst=3 src=2 src2=4 imm=0xc5c46d76 imm2=0x700044b5 rot=22 bit=10 mask=2
29: load dst=1 src=0 src2=3 imm=0x0fbaf177 imm2=0xfff4f2ed rot=20 bit=31 mask=2
30: load dst=2 src=4 src2=0 imm=0xe90eb2e5 imm2=0xb0f9eb79 rot=27 bit=14 mask=4
31: load dst=1 src=5 src2=2 imm=0x97ba3fc3 imm2=0x7894e657 rot=3 bit=30 mask=16
32: mad dst=7 src=2 src2=7 imm=0x070888a8 imm2=0xe403240e rot=2 bit=16 mask=2
33: xor dst=2 src=0 src2=5 imm=0xf2e46d55 imm2=0x29701828 rot=31 bit=28 mask=8
34: rotl dst=2 src=3 src2=0 imm=0x58f75b87 imm2=0x343b7aee rot=12 bit=14 mask=4
35: sub dst=2 src=5 src2=6 imm=0x5892a9e6 imm2=0xc9824c94 rot=19 bit=26 mask=4
36: add dst=4 src=2 src2=3 imm=0x7b5b5474 imm2=0x45cfc5dd rot=17 bit=18 mask=8
37: mul dst=6 src=5 src2=7 imm=0xccf564a5 imm2=0x873ad101 rot=7 bit=11 mask=4
38: xor dst=7 src=0 src2=6 imm=0xcac8f06d imm2=0x6b97c683 rot=18 bit=28 mask=2
39: rotr dst=7 src=2 src2=4 imm=0xb8180e9d imm2=0x32bbd117 rot=23 bit=19 mask=4
40: rotl dst=2 src=3 src2=0 imm=0x2d6070bc imm2=0x68ff101e rot=13 bit=18 mask=1
41: rotl dst=7 src=0 src2=2 imm=0x0daf96ea imm2=0x36f37be1 rot=5 bit=0 mask=8
42: rotr dst=4 src=3 src2=6 imm=0x6ced15b7 imm2=0x6df7aed4 rot=19 bit=4 mask=1
43: xor dst=7 src=3 src2=5 imm=0x8ace05f3 imm2=0xd378ec12 rot=23 bit=10 mask=4
44: load dst=0 src=7 src2=4 imm=0xb0607786 imm2=0xc4acabbc rot=13 bit=7 mask=16
45: xor dst=1 src=6 src2=5 imm=0xe09f54e9 imm2=0x83e825bf rot=23 bit=14 mask=8
46: load dst=3 src=0 src2=0 imm=0x63cc1e4e imm2=0xa1be8118 rot=12 bit=6 mask=2
47: load dst=6 src=3 src2=7 imm=0x353f1d79 imm2=0x3b2e7456 rot=18 bit=18 mask=1
48: rotl dst=4 src=2 src2=7 imm=0x00d8a3cd imm2=0x231866d2 rot=21 bit=20 mask=1
49: shfl dst=5 src=0 src2=2 imm=0xa8bae6df imm2=0xf572bdb9 rot=14 bit=7 mask=1
50: rotl dst=0 src=7 src2=7 imm=0x81ef22e1 imm2=0x74438fc5 rot=28 bit=18 mask=4
51: sub dst=6 src=0 src2=6 imm=0x383b9260 imm2=0x11e17c61 rot=12 bit=19 mask=16
52: load dst=5 src=0 src2=2 imm=0xfb84f451 imm2=0x11cd863e rot=21 bit=20 mask=8
53: shfl dst=6 src=5 src2=4 imm=0xb1a7db6b imm2=0x76686b9b rot=12 bit=4 mask=16
54: xor dst=3 src=6 src2=3 imm=0x6f981f52 imm2=0xd99aeba2 rot=12 bit=27 mask=1
55: rotr dst=2 src=1 src2=2 imm=0xac6be8e3 imm2=0x18d67dbb rot=26 bit=10 mask=4
56: load dst=1 src=5 src2=0 imm=0x7e7f6a00 imm2=0x6f0747da rot=25 bit=28 mask=1
57: add dst=5 src=0 src2=4 imm=0xf03673fe imm2=0xa75cd60d rot=16 bit=12 mask=8
58: load dst=5 src=1 src2=2 imm=0x227f94a6 imm2=0x0e8344f9 rot=20 bit=10 mask=2
59: rotl dst=1 src=4 src2=3 imm=0xdecd4794 imm2=0x8dfb96bb rot=21 bit=7 mask=4
60: mul dst=3 src=2 src2=2 imm=0x0dd268e0 imm2=0x53034ca9 rot=1 bit=8 mask=8
61: mul dst=6 src=4 src2=7 imm=0x3a45a321 imm2=0x9bc59a5f rot=25 bit=11 mask=1
62: add dst=5 src=4 src2=2 imm=0x8f229cc1 imm2=0xcaac64a2 rot=17 bit=13 mask=8
63: load dst=3 src=5 src2=6 imm=0xb2574178 imm2=0xcbcc798d rot=28 bit=0 mask=16

20
docs/analysis/class-v6/logs/census-packs/hl-v6-rw/sitestats.tsv Normal file
View file

@ -0,0 +1,20 @@
program id 30628f8adcf6035e class mx8-erad810f22d+sh256x27+state+rw attempt 0 generator v5
rule: ACCEPTED, distinct 128.000 per hash, saturated 0, bias max 51
site instr win off width distinct ratio under_c2 under_c3 bucket_ratio bucket_z bucket_id bit_z bit top_count
0 7 2 1 1 1040390 1.00001 0 0 1.578 4.62 21481 -2.71 24 4
1 8 1 1 1 1044551 1.00007 0 0 1.812 4.60 42831 2.89 15 3
2 9 1 1 1 1044565 1.00008 0 0 1.812 4.60 44242 2.15 20 3
3 10 1 1 1 1044596 1.00011 0 0 1.750 4.24 45914 -1.31 10 3
4 11 0 0 1 1046426 0.99990 0 0 2.188 4.75 43244 2.52 6 3
5 13 1 1 1 1044427 0.99995 0 0 1.875 4.95 39979 -1.81 12 3
6 29 0 0 1 1046580 1.00005 0 0 2.188 4.75 58815 -2.54 5 3
7 30 2 2 1 1040423 1.00004 0 0 1.547 4.38 47083 2.51 15 3
8 31 1 1 1 1044400 0.99992 0 0 1.875 4.95 42004 2.17 8 3
9 44 1 1 1 1044545 1.00006 0 0 1.781 4.42 55440 2.87 7 3
10 46 2 0 1 1040493 1.00010 0 0 1.547 4.38 15496 -2.38 9 3
11 47 0 0 1 1046444 0.99992 0 0 2.188 4.75 10355 -2.19 17 3
12 52 0 0 1 1046513 0.99999 0 0 2.375 5.50 35652 -1.70 23 3
13 56 0 0 1 1046445 0.99992 0 0 2.188 4.75 17864 -2.41 11 3
14 58 2 0 1 1040455 1.00007 0 0 1.516 4.12 14233 2.30 13 3
15 63 1 1 1 1044479 1.00000 0 0 1.781 4.42 33129 -2.71 1 3
summary: sites 16, min ratio 0.99990, under 0.98: 0, under 0.995: 0, bucket z max 5.50, bit z max 2.89, sites over 6 sigma 0, 3.3 s
Can't render this file because it has a wrong number of fields in line 3.

119
docs/analysis/class-v6/reference-population.md Normal file
View file

@ -0,0 +1,119 @@
# The reference GPU population and the cost per accepted unit of work
8 October 2026, 18:0x UK, branch `class-v6-floor-sram`, floor lane 3 for the research lane (the pin appended to
docs/plans/igneum-2.0.md by the coordinator at 18:16 UK: the served reference population, one row per class, two tests
per class, every cell labelled). Beside `docs/analysis/class-v6/coexistence-model.md` (its section 2 is most of this)
and the workbook. **Every energy cell carries its label (measured or modelled or estimated, with where); every cost cell
is arithmetic on labelled inputs and is modelled; prices are approximate street and list figures of October 2026.**
The script is `scratchpad/refpop.py`, run on build-4 at 18:02 UK. Apple is reported and not headlined. Nothing here is
a boundary; the founder is not named.
## 1. The measure
Cost per accepted MH/s-hour, in micro-USD (USD x 10^-6): annualised hardware plus power plus hosting, failures and fees,
divided by the year's accepted work. Accepted work is 97 percent of raw (rejects 0.5 percent, downtime 2.0, epoch
preparation and propagation 0.5; approximate from the devnet's share rates); the pool fee 1 percent. Two tests per
class:
- **The existing owner**: hardware sunk; pays power at the card's floor joules (the knee with the knobs, lane 4's
class v5 table and the tiers file `app/igneum-app/tiers/class-v5-tiers.json`), a wear allowance of 5 percent of the
used price a year, and fees; the alternative use (the card's rental yield on a public market, approximate) is
reported in its own column and not deducted, so the owner's number is the floor of what the card costs to run.
- **The new entrant**: buys at MSRP (the list price) or at street (October 2026, approximate), operates two years,
resells at the table's fraction of MSRP, with 3 percent a year of failures; power as the owner.
Three electricity prices (USD 0.06, 0.12, 0.25 per kWh); hosting 0 at home. The energy columns: the floor is the knee
with the card's knobs (the lock on Blackwell and Ada, the cap on Ampere, the ADLX offsets on AMD; nothing on Apple or
Intel), the stock row unlocked at 100 percent; "measured" names the instrument's row (PC 1, PC 2, a rented host, lane
4's sweep), "modelled" a knee read from a measured stock row by the card's architecture, "estimated" a figure with no
instrument behind it (the Arc B580's watts).
## 2. The plan's population (block 1)
| Class | Memory | Microjoules per hash, floor / stock | MH/s at the floor | Wh per MH/s-hour | MSRP / street / used (USD) | Resale after 2 y | Existing owner at 0.06 / 0.12 / 0.25 | New entrant at MSRP, 0.06 / 0.12 / 0.25 | New entrant at street, 0.06 / 0.12 / 0.25 | Hardware share of the street entrant at 0.12 | Alternative use (rental yield), per MH/s-hour | Energy label | Rate label |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTX 5090 | 32 GB | 2.33 / 3.48 | 134.8 | 2.33 | 1999 / 2600 / 2200 | 55 percent | 243 / 388 / 704 | 566 / 711 / 1027 | 846 / 992 / 1307 | 71 pct | 2967 | measured (PC 1, tiers file) | measured |
| RTX 5080 | 16 GB | 2.06 / 3.48 | 71.2 | 2.06 | 999 / 1100 / 900 | 50 percent | 204 / 332 / 611 | 570 / 699 / 978 | 660 / 788 / 1067 | 67 pct | 2809 | measured (rented, lane 4) | measured |
| RTX 4090 | 24 GB | 3.58 / 5.00 | 58.0 | 3.58 | 1599 / 1700 / 1300 | 45 percent | 357 / 580 / 1065 | 1178 / 1402 / 1887 | 1288 / 1512 / 1996 | 70 pct | 6034 | stock measured, knee modelled (lane 4) | measured |
| RTX 3090 (used) | 24 GB | 4.51 / 5.03 | 37.8 | 4.51 | 1499 / 900 / 650 | 30 percent | 384 / 666 / 1276 | 2030 / 2311 / 2922 | 1032 / 1314 / 1924 | 57 pct | 5291 | stock measured, cap modelled (lane 4) | measured |
| RX 9070 XT | 16 GB | 7.90 / 10.70 | 18.9 | 7.90 | 599 / 650 / 520 | 45 percent | 657 / 1151 / 2220 | 1591 / 2085 / 3154 | 1761 / 2255 / 3324 | 56 pct | 7937 | measured (PC 1, ADLX -500 MHz -30 pct) | measured |
| RX 7600 8 GB | 8 GB | 5.4 (modelled knee) / 7.2 | 15.8 | 5.40 | 269 / 270 / 200 | 40 percent | 413 / 750 / 1,481 | 981 / 1,318 / 2,049 | 985 / 1,322 / 2,053 | 49 pct | 3,167 | class v6: rate MEASURED 15.79 MH/s (PC 1, the AMD-and-Intel lane); stock 113 W measured, 7.2 microjoules; the knee 5.4 MODELLED by the 9070 XT's ADLX grid (24 percent of the draw for no rate), the metered knee the morning's batch 03; the earlier row (13.9 MH/s, 6.2 / 8.1) was the class v4 sub-version 3 read | measured rate and stock watts, modelled knee |
| Intel Arc B580 | 12 GB | 10.40 / 10.40 | 10.7 | 10.40 | 249 / 260 / 200 | 40 percent | 761 / 1411 / 2819 | 1529 / 2179 / 3587 | 1594 / 2243 / 3651 | 42 pct | 4673 | rate measured 10.4 to 11 MH/s (PC 2); watts unread, about 110 W by the board's class (estimated); no lever | measured rate, estimated watts |
| Apple M5 Max (reported, not headlined) | 36 GB | 1.40 / 1.40 | 27.1 | 1.40 | 3999 / 4000 / 3200 | 55 percent | 789 / 876 / 1066 | 4268 / 4356 / 4545 | 4271 / 4358 / 4548 | 96 pct | 0 (a workstation; an owner only) | measured (GPU and DRAM channels, 6 October) | measured |
Two more Apple rows (added 18:2x UK; the founder's own powermetrics reads, 8 October 18:1x UK, v2.0.0 class v5 on
the 2.0 devnet, TEAM-REPORTED, GPU power only and not the wall, which is why they sit beside and not in place of the
metered 1.40 row above; reported, not headlined; the Apple tier mines and does not prove; a Mac is an owner only, the
entrant rows are the arithmetic for completeness):
| Class | Memory | Microjoules per hash (GPU power only) | MH/s | Wh per MH/s-hour | MSRP / street / used (USD) | Resale after 2 y | Existing owner at 0.06 / 0.12 / 0.25 | New entrant at MSRP, 0.06 / 0.12 / 0.25 | New entrant at street | Hardware share of the entrant at 0.12 | Energy label | Rate label |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MacBook Pro M5 Max, 40 GPU cores, 64 GB | 64 GB unified | 0.73 | 22.67 | 0.73 | 3,999 / 4,000 / 3,200 | 55 percent | 884 / 930 / 1,028 | 5,043 / 5,089 / 5,188 | 5,046 / 5,092 / 5,191 | 98 percent | team-reported (powermetrics, GPU rail only, 16.65 W) | team-reported |
| Mac mini M6, 12 GPU cores, 16 GB | 16 GB unified | 0.73 | 11.0 | 0.73 | 999 / 1,000 / 800 | 55 percent | 478 / 523 / 622 | 2,619 / 2,664 / 2,763 | 2,624 / 2,670 / 2,769 | 97 percent | team-reported (powermetrics, GPU rail only, 8.03 W) | team-reported |
Reading: at the GPU rail alone both Macs are the most efficient honest joule in the table (0.73 against the 5070 Ti's
1.70), and still the dearest honest owners per accepted unit after the datacentre parts (the wear allowance on a USD
3,200 to 800 machine at 11 to 23 MH/s), because the Mac's cost is the machine, not the power; the M6 mini's owner
(478 at 0.06) sits between the 3060 and the 9070 XT, the M5 Max laptop's (884) above every consumer card. The wall
and the SoC's other rails are not in the 0.73; the metered M5 Max row (1.40 at the GPU and DRAM channels) is the
comparable figure for the coexistence model.
## 3. The rest of the measured classes (block 2)
| Class | Memory | Microjoules per hash, floor / stock | MH/s at the floor | Wh per MH/s-hour | MSRP / street / used (USD) | Resale after 2 y | Existing owner at 0.06 / 0.12 / 0.25 | New entrant at MSRP, 0.06 / 0.12 / 0.25 | New entrant at street, 0.06 / 0.12 / 0.25 | Hardware share of the street entrant at 0.12 | Alternative use (rental yield), per MH/s-hour | Energy label | Rate label |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTX 5070 Ti | 16 GB | 1.70 / 2.84 | 77.0 | 1.70 | 749 / 800 / 650 | 50 percent | 156 / 263 / 493 | 412 / 519 / 749 | 454 / 560 / 790 | 62 pct | 1948 | stock measured (rented), knee modelled | measured |
| RTX 5070 | 12 GB | 1.75 / 2.99 | 41.0 | 1.75 | 549 / 560 / 450 | 50 percent | 175 / 284 / 521 | 531 / 640 / 877 | 548 / 657 / 894 | 67 pct | 2439 | stock measured, knee modelled | measured |
| RTX 5060 Ti 16 GB | 16 GB | 2.36 / 4.02 | 19.0 | 2.36 | 429 / 450 / 360 | 45 percent | 260 / 407 / 727 | 929 / 1077 / 1396 | 999 / 1146 / 1466 | 74 pct | 4211 | stock measured, knee modelled | measured |
| RTX 5060 | 8 GB | 2.21 / 3.76 | 17.0 | 2.21 | 299 / 310 / 250 | 45 percent | 225 / 364 / 663 | 747 / 885 / 1184 | 788 / 926 / 1225 | 70 pct | 3529 | stock measured, knee modelled | measured |
| RTX 4080 | 16 GB | 3.51 / 4.92 | 45.0 | 3.51 | 999 / 1000 / 700 | 40 percent | 312 / 531 / 1006 | 1058 / 1277 / 1752 | 1059 / 1279 / 1754 | 66 pct | 4444 | stock measured, knee modelled | measured |
| RTX 4070 | 12 GB | 3.58 / 5.82 | 31.1 | 3.58 | 549 / 550 / 400 | 40 percent | 300 / 524 / 1008 | 891 / 1114 / 1599 | 893 / 1116 / 1601 | 60 pct | 3215 | measured (PC, 1,860 MHz + 50 pct cap) | measured |
| RTX 4060 Ti 16 GB | 16 GB | 3.81 / 5.32 | 17.6 | 3.81 | 499 / 420 / 290 | 35 percent | 336 / 574 / 1090 | 1398 / 1636 / 2152 | 1116 / 1354 / 1870 | 65 pct | 4545 | stock measured, knee modelled | measured |
| RTX 3080 (used) | 10 GB | 4.20 / 4.54 | 40.8 | 4.20 | 699 / 450 / 330 | 25 percent | 310 / 573 / 1141 | 1071 / 1334 / 1902 | 687 / 950 / 1518 | 45 pct | 2941 | modelled (both rented hosts capped) | measured |
| RTX 3060 (used) | 12 GB | 5.77 / 6.40 | 23.8 | 5.77 | 329 / 260 / 190 | 25 percent | 408 / 768 / 1550 | 1013 / 1374 / 2155 | 831 / 1191 / 1972 | 39 pct | 2521 | stock measured, cap modelled | measured |
| H100 (hosted) | 80 GB | 2.00 / 2.58 | 90.0 | 2.00 | 25000 / 25000 / 18000 | 50 percent | 1313 / 1438 / 1709 | 8869 / 8994 / 9265 | 8869 / 8994 / 9265 | 97 pct | 27778 | stock measured (rented), lock modelled | measured |
| A100 (used) | 80 GB | 2.89 / 2.99 | 60.0 | 2.89 | 10000 / 10000 / 6000 | 35 percent | 775 / 955 / 1346 | 7001 / 7182 / 7573 | 7001 / 7182 / 7573 | 95 pct | 16667 | stock measured (rented) | measured |
## 4. The chip rows beside them, at the adversary lane's reconciled tickets (development sunk, farm hosting USD 0.02 per kWh on top)
| Chip (reconciled ticket) | Microjoules per hash | USD per MH/s | Life | All-in at 0.06 / 0.12 / 0.25 | Hardware / power at 0.06 | Label |
|---|---|---|---|---|---|---|
| GDDR7 machine (adversary lane, reconciled) | 1.55 | 4.84 | 1 y | 722 / 819 / 1029 | 586 / 128 | modelled: 1.5x per joule vs the 5090 lock, USD 4.84 per MH/s |
| GDDR7 machine (adversary lane, reconciled) | 1.55 | 4.84 | 3 y | 338 / 435 / 646 | 207 / 128 | modelled: 1.5x per joule vs the 5090 lock, USD 4.84 per MH/s |
| GDDR7 machine (adversary lane, reconciled) | 1.55 | 4.84 | 5 y | 262 / 359 / 569 | 131 / 128 | modelled: 1.5x per joule vs the 5090 lock, USD 4.84 per MH/s |
| Stored-half hybrid (reconciled) | 1.21 | 2.80 | 1 y | 443 / 519 / 682 | 339 / 100 | modelled: 1.93x, USD 2.80 |
| Stored-half hybrid (reconciled) | 1.21 | 2.80 | 3 y | 221 / 297 / 460 | 120 / 100 | modelled: 1.93x, USD 2.80 |
| Stored-half hybrid (reconciled) | 1.21 | 2.80 | 5 y | 177 / 253 / 416 | 76 / 100 | modelled: 1.93x, USD 2.80 |
| N2 SRAM die (reconciled, 1 to 1.5 kW machine) | 1.01 | 1.00 | 1 y | 207 / 270 / 407 | 121 / 84 | modelled: 2.3x, USD 1.0 (0.5 to 1.6) |
| N2 SRAM die (reconciled, 1 to 1.5 kW machine) | 1.01 | 1.00 | 3 y | 128 / 191 / 328 | 43 / 84 | modelled: 2.3x, USD 1.0 (0.5 to 1.6) |
| N2 SRAM die (reconciled, 1 to 1.5 kW machine) | 1.01 | 1.00 | 5 y | 112 / 175 / 312 | 27 / 84 | modelled: 2.3x, USD 1.0 (0.5 to 1.6) |
## 5. Reading the table
1. On every consumer card the owner's cost is 60 to 70 percent electricity, so the grid price is the owner's whole
variable; the entrant's cost is 60 to 75 percent hardware at street prices, so the card's price and its resale are
the entrant's. The cheapest honest owner is a 5070 Ti at its knee (156 micro-USD at 0.06, 263 at 0.12); the cheapest
entrant the same card at MSRP (412 at 0.06, 519 at 0.12).
2. The 8 GB tier (the RX 7600 at 13.9 MH/s and 6.2 microjoules at its estimated knee, 8.1 measured at stock) costs
an owner 2.6x the 5070 Ti per accepted unit and an entrant 1.4x at street; the Arc B580 (10.7 MH/s measured, watts
estimated) 4x and 1.5x. These are the classes the dataset floor retires first, and their owners mine only below
about 8 cents per kWh at the GPU equilibrium of the coexistence model.
3. Datacentre parts and the Mac never enter as entrants (hardware 95 percent or more of the cost) and mine only as
owners with nothing better to do, which their rental yields (USD 1 to 3 an hour, 10x to 30x their mining cost per
MH/s-hour) say they always have.
4. Against the chip rows at the reconciled tickets and a 3-year life, the GDDR7 machine (338 micro-USD at 0.06) sits
above every Blackwell owner and between the Blackwell entrant at MSRP and at street; the hybrid (221) and the die (128) sit below every owner at 0.12. The coexistence model carries what that means; this table carries the cells.
## 6. Unverified and owed
- The RX 7600's knee is estimated by the 9070 XT's ADLX grid; its own 24-point grid runs on PC 1 and replaces the
floor cell. Its dataset rows landed (18:2x UK, PC 1, rate only, fingerprints equal to the 5090's): 15.75 / 15.46 /
15.34 / 15.35 MH/s at 1 / 2 / 4 / 5.5 GiB, the 5.5 GiB dataset fitting at 6,732 MiB of 8,176; the table's 13.9
MH/s is the class v5 row at stock from the card-in job, within the day's spread (13.4 to 15.8 across jobs and
classes); the 8 GB tier pays under 3 percent of rate for the 5.5 GiB step. The Arc B580's watts are unread (about 110 W by the board's class); the Intel lane's
meter row replaces them. The Ada and Ampere knees are modelled (no rented host allows the lock).
- MSRP and street prices are October 2026 approximations; the 5090's street has been 2x MSRP this year; used and
resale fractions are approximate.
- The rental yields are approximate public-market figures; the accepted-work factor, the wear allowance and the
failure rate are assumptions.
- Nothing was run on the Mac; the script ran on build-4.

12
docs/analysis/class-v6/rows/fleet-f05-20261008.md Normal file
View file

@ -0,0 +1,12 @@
# Review B's F05: the reg64 address mix in its closed prefix form, measured (8 October 2026)
The fleet lane's rows on the generator-6 worker (igneum-worker-cuda-gen6, sha256 804a6f7f...), stock, 250 batches of 2^24 nonces, block-warps 1, nvidia-smi at 1 Hz, W = mean busy board power; both packs the same program (fingerprint e8f4f3289c6ee1fc on both packs on both cards, the gate); the packs hl-v6-all.tgz (the fold form, the object) and hl-v6-all-prefix.tgz (the closed prefix form, `igneum-pow export --reg64-prefix`) at program id 0x4de7b836cc40a4ea. Evidence on build-1: /srv/artefacts/tas/54900078/ (the 5090) and /srv/artefacts/tas/si1xc4yhpakk1v/ (the 4090): the rows, the four bench logs, the four 1 Hz power samples.
| Card | Text | MH/s | W | nJ per hash | registers | blocks per SM | spill | fingerprint | check |
|---|---|---|---|---|---|---|---|---|---|
| RTX 5090 (sm_120, driver 595.58) | fold | 70.647 | 437.8 | 6.20 | 100 | 16 | 0 | e8f4f3289c6ee1fc | PASS |
| RTX 5090 | prefix | 70.166 | 500.0 | 7.13 | 248 | 8 | 0 | e8f4f3289c6ee1fc | PASS |
| RTX 4090 (sm_89, driver 580.159) | fold | 31.313 | 238.1 | 7.60 | 93 | 20 | 0 | e8f4f3289c6ee1fc | PASS |
| RTX 4090 | prefix | 31.288 | 231.7 | 7.41 | 154 | 12 | 0 | e8f4f3289c6ee1fc | PASS |
Reading: the closed form is an exact equivalent (verify::reg64_address_source, proven on every source register in the crate's tests) and on a GPU it buys nothing: the card is memory-bound at the dataset on both texts; the prefix's running total costs 61 registers on sm_89 and 148 on sm_120, halving the 5090's resident blocks and costing it 14 percent more energy per hash; the 4090 reads a wash inside the sampler's noise. The verifier keeps the fold as the definition; the chip's gain from a prefix tree is the adversary lane's pricing.

3
docs/analysis/review-2026-10-08-b/f10/3090-full-r1.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.710 prod_wall_s=58.497 prod_mhs=9.18 prod_mean_w=164.9 prod_j_per_batch=301.4 prod_mh_per_w=0.0557 samples=234 steady_wall_s=53.787 steady_mhs=9.98 steady_mean_w=168.3 steady_j_per_batch=282.9 steady_mh_per_w=0.0593 samples=215 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 1630.34 ms, select 0.000 ms, read-back 24.723 ms, scan 24.119 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/3090-full-r2.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.231 prod_wall_s=90.096 prod_mhs=5.96 prod_mean_w=134.4 prod_j_per_batch=378.4 prod_mh_per_w=0.0443 samples=360 steady_wall_s=85.865 steady_mhs=6.25 steady_mean_w=136.9 steady_j_per_batch=367.3 steady_mh_per_w=0.0457 samples=343 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 2637.86 ms, select 0.000 ms, read-back 22.784 ms, scan 20.625 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/3090-full-r3.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.110 prod_wall_s=23.537 prod_mhs=22.81 prod_mean_w=282.4 prod_j_per_batch=207.7 prod_mh_per_w=0.0808 samples=94 steady_wall_s=19.428 steady_mhs=27.63 steady_mean_w=325.9 steady_j_per_batch=197.9 steady_mh_per_w=0.0848 samples=77 worker_cpu_s=5.61 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 557.15 ms, select 0.000 ms, read-back 24.225 ms, scan 23.753 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/3090-full-r4.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.397 prod_wall_s=90.051 prod_mhs=5.96 prod_mean_w=134.3 prod_j_per_batch=377.9 prod_mh_per_w=0.0444 samples=360 steady_wall_s=85.654 steady_mhs=6.27 steady_mean_w=136.7 steady_j_per_batch=365.9 steady_mh_per_w=0.0459 samples=342 worker_cpu_s=5.81 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 2624.25 ms, select 0.000 ms, read-back 27.536 ms, scan 22.950 ms; 32 jobs 32 chunks

4
docs/analysis/review-2026-10-08-b/f10/3090-kernel-bench.log Normal file
View file

@ -0,0 +1,4 @@
info igneum-worker-cuda 1.0 (4 October 2026): device 0 NVIDIA_GeForce_RTX_3090 (sm_86, 82 SMs), driver 13.2 from libcuda.so.1, NVRTC 12.8 from libnvrtc.so.12, target sm_86 (the device's architecture, listed by NVRTC)
pack /root/p02/packs/hl-v6-all on NVIDIA_GeForce_RTX_3090: nvrtc 2558 cache 3 dataset 37 hot 0 check 1190 race 0 ms variant base class v5 (state leaves 93, uploaded for the build and freed); self-test PASS (cache head, last line and FNV-1a 64 448274a57f508cbc; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes)
warm-up dispatch (base 0): 554.01 ms; 5 timed dispatches of 16777216 nonces: mean 557.06 ms
RESULT pack=/root/p02/packs/hl-v6-all class=mx8-erad810f22d+sh256x27+state+reg64c+fold+rw device=NVIDIA_GeForce_RTX_3090 arch=sm_86 regs=87 blocks_per_sm=16 warps=0 resident=1312 arena_mib=0 hot_mib=0 hot_slots=0 hot_fill_ms=0.00 nonces=16777216 batches=5 check=PASS fingerprint=59e6708e46f1e87c mhs=30.117 loads=128 bytes=512 scratch_ops=0 time=wall

10
docs/analysis/review-2026-10-08-b/f10/3090-run-more.log Normal file
View file

@ -0,0 +1,10 @@
== 3090-full-r3 2026-10-08T19:44:29Z sm 210 MHz, Not Active
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.110 prod_wall_s=23.537 prod_mhs=22.81 prod_mean_w=282.4 prod_j_per_batch=207.7 prod_mh_per_w=0.0808 samples=94 steady_wall_s=19.428 steady_mhs=27.63 steady_mean_w=325.9 steady_j_per_batch=197.9 steady_mh_per_w=0.0848 samples=77 worker_cpu_s=5.61 worker_sha=9b36b523d0812c3c
3090-full-r3 sm MHz mean 1769 min 210 max 1965 over 103 samples
== 3090-select-r3 2026-10-08T19:44:55Z sm 1950 MHz, Not Active
card=NVIDIA_GeForce_RTX_3090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.297 prod_wall_s=83.427 prod_mhs=6.44 prod_mean_w=138.7 prod_j_per_batch=361.6 prod_mh_per_w=0.0464 samples=334 steady_wall_s=79.130 steady_mhs=6.78 steady_mean_w=140.1 steady_j_per_batch=346.4 steady_mh_per_w=0.0484 samples=316 worker_cpu_s=4.23 worker_sha=9b36b523d0812c3c
3090-select-r3 sm MHz mean 318 min 210 max 1950 over 343 samples
== 3090-full-r4 2026-10-08T19:46:24Z sm 210 MHz, Active
card=NVIDIA_GeForce_RTX_3090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.397 prod_wall_s=90.051 prod_mhs=5.96 prod_mean_w=134.3 prod_j_per_batch=377.9 prod_mh_per_w=0.0444 samples=360 steady_wall_s=85.654 steady_mhs=6.27 steady_mean_w=136.7 steady_j_per_batch=365.9 steady_mh_per_w=0.0459 samples=342 worker_cpu_s=5.81 worker_sha=9b36b523d0812c3c
3090-full-r4 sm MHz mean 210 min 210 max 210 over 369 samples
== 3090-select-r4 2026-10-08T19:47:59Z sm 210 MHz, Active

3
docs/analysis/review-2026-10-08-b/f10/3090-select-r1.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=7.093 prod_wall_s=91.584 prod_mhs=5.86 prod_mean_w=131.7 prod_j_per_batch=376.9 prod_mh_per_w=0.0445 samples=366 steady_wall_s=84.491 steady_mhs=6.35 steady_mean_w=136.9 steady_j_per_batch=361.5 steady_mh_per_w=0.0464 samples=337 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 2637.57 ms, select 0.532 ms, read-back 0.296 ms, scan 0.019 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/3090-select-r2.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.138 prod_wall_s=21.996 prod_mhs=24.41 prod_mean_w=286.1 prod_j_per_batch=196.7 prod_mh_per_w=0.0853 samples=88 steady_wall_s=17.858 steady_mhs=30.06 steady_mean_w=330.7 steady_j_per_batch=184.6 steady_mh_per_w=0.0909 samples=71 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 555.70 ms, select 0.268 ms, read-back 0.217 ms, scan 0.013 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/3090-select-r3.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_3090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=4.297 prod_wall_s=83.427 prod_mhs=6.44 prod_mean_w=138.7 prod_j_per_batch=361.6 prod_mh_per_w=0.0464 samples=334 steady_wall_s=79.130 steady_mhs=6.78 steady_mean_w=140.1 steady_j_per_batch=346.4 steady_mh_per_w=0.0484 samples=316 worker_cpu_s=4.23 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_3090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 87 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_86 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 2470.39 ms, select 0.503 ms, read-back 0.206 ms, scan 0.017 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-full-r1.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.275 prod_wall_s=10.715 prod_mhs=50.10 prod_mean_w=278.0 prod_j_per_batch=93.1 prod_mh_per_w=0.1802 samples=43 steady_wall_s=8.440 steady_mhs=63.61 steady_mean_w=328.3 steady_j_per_batch=86.6 steady_mh_per_w=0.1938 samples=34 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 236.64 ms, select 0.000 ms, read-back 12.794 ms, scan 13.245 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-full-r2.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.373 prod_wall_s=11.405 prod_mhs=47.07 prod_mean_w=270.7 prod_j_per_batch=96.5 prod_mh_per_w=0.1739 samples=46 steady_wall_s=9.031 steady_mhs=59.45 steady_mean_w=326.9 steady_j_per_batch=92.3 steady_mh_per_w=0.1819 samples=36 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 236.65 ms, select 0.000 ms, read-back 25.097 ms, scan 19.536 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-full-r3.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.400 prod_wall_s=10.582 prod_mhs=50.73 prod_mean_w=267.7 prod_j_per_batch=88.5 prod_mh_per_w=0.1895 samples=43 steady_wall_s=8.182 steady_mhs=65.62 steady_mean_w=335.4 steady_j_per_batch=85.8 steady_mh_per_w=0.1956 samples=33 worker_cpu_s=2.97 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 236.73 ms, select 0.000 ms, read-back 7.679 ms, scan 10.341 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-full-r4.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.334 prod_wall_s=10.566 prod_mhs=50.81 prod_mean_w=285.8 prod_j_per_batch=94.4 prod_mh_per_w=0.1778 samples=43 steady_wall_s=8.233 steady_mhs=65.21 steady_mean_w=351.5 steady_j_per_batch=90.4 steady_mh_per_w=0.1855 samples=33 worker_cpu_s=2.92 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 237.01 ms, select 0.000 ms, read-back 8.639 ms, scan 10.488 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-full-r5.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.463 prod_wall_s=10.767 prod_mhs=49.86 prod_mean_w=276.8 prod_j_per_batch=93.1 prod_mh_per_w=0.1801 samples=43 steady_wall_s=8.304 steady_mhs=64.65 steady_mean_w=340.3 steady_j_per_batch=88.3 steady_mh_per_w=0.1900 samples=33 worker_cpu_s=3.13 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback full worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 134217728 B (full, 0 full fallbacks); mean per chunk: kernel+sync 236.70 ms, select 0.000 ms, read-back 9.028 ms, scan 12.796 ms; 32 jobs 32 chunks

4
docs/analysis/review-2026-10-08-b/f10/5090-kernel-bench.log Normal file
View file

@ -0,0 +1,4 @@
info igneum-worker-cuda 1.0 (4 October 2026): device 0 NVIDIA_GeForce_RTX_5090 (sm_120, 170 SMs), driver 13.2 from libcuda.so.1, NVRTC 12.8 from libnvrtc.so.12, target sm_120 (the device's architecture, listed by NVRTC)
pack /root/p02/packs/hl-v6-all on NVIDIA_GeForce_RTX_5090: nvrtc 1476 cache 1 dataset 13 hot 0 check 532 race 0 ms variant base class v5 (state leaves 93, uploaded for the build and freed); self-test PASS (cache head, last line and FNV-1a 64 448274a57f508cbc; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes)
warm-up dispatch (base 0): 236.80 ms; 5 timed dispatches of 16777216 nonces: mean 236.68 ms
RESULT pack=/root/p02/packs/hl-v6-all class=mx8-erad810f22d+sh256x27+state+reg64c+fold+rw device=NVIDIA_GeForce_RTX_5090 arch=sm_120 regs=92 blocks_per_sm=20 warps=0 resident=3400 arena_mib=0 hot_mib=0 hot_slots=0 hot_fill_ms=0.00 nonces=16777216 batches=5 check=PASS fingerprint=59e6708e46f1e87c mhs=70.885 loads=128 bytes=512 scratch_ops=0 time=wall

19
docs/analysis/review-2026-10-08-b/f10/5090-run-more.log Normal file
View file

@ -0,0 +1,19 @@
== 5090-full-r3 2026-10-08T19:44:33Z sm 180 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.400 prod_wall_s=10.582 prod_mhs=50.73 prod_mean_w=267.7 prod_j_per_batch=88.5 prod_mh_per_w=0.1895 samples=43 steady_wall_s=8.182 steady_mhs=65.62 steady_mean_w=335.4 steady_j_per_batch=85.8 steady_mh_per_w=0.1956 samples=33 worker_cpu_s=2.97 worker_sha=9b36b523d0812c3c
5090-full-r3 sm MHz mean 2387 min 180 max 2895 over 51 samples
== 5090-select-r3 2026-10-08T19:44:46Z sm 2895 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.327 prod_wall_s=9.951 prod_mhs=53.95 prod_mean_w=290.5 prod_j_per_batch=90.3 prod_mh_per_w=0.1857 samples=40 steady_wall_s=7.624 steady_mhs=70.42 steady_mean_w=365.4 steady_j_per_batch=87.1 steady_mh_per_w=0.1927 samples=30 worker_cpu_s=2.33 worker_sha=9b36b523d0812c3c
5090-select-r3 sm MHz mean 2636 min 472 max 2895 over 49 samples
== 5090-full-r4 2026-10-08T19:44:58Z sm 2895 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.334 prod_wall_s=10.566 prod_mhs=50.81 prod_mean_w=285.8 prod_j_per_batch=94.4 prod_mh_per_w=0.1778 samples=43 steady_wall_s=8.233 steady_mhs=65.21 steady_mean_w=351.5 steady_j_per_batch=90.4 steady_mh_per_w=0.1855 samples=33 worker_cpu_s=2.92 worker_sha=9b36b523d0812c3c
5090-full-r4 sm MHz mean 2666 min 457 max 2895 over 51 samples
== 5090-select-r4 2026-10-08T19:45:12Z sm 2895 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.590 prod_wall_s=10.216 prod_mhs=52.55 prod_mean_w=285.4 prod_j_per_batch=91.1 prod_mh_per_w=0.1841 samples=41 steady_wall_s=7.626 steady_mhs=70.40 steady_mean_w=355.9 steady_j_per_batch=84.8 steady_mh_per_w=0.1978 samples=31 worker_cpu_s=2.55 worker_sha=9b36b523d0812c3c
5090-select-r4 sm MHz mean 2608 min 555 max 2895 over 50 samples
== 5090-full-r5 2026-10-08T19:45:24Z sm 2880 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=full jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.463 prod_wall_s=10.767 prod_mhs=49.86 prod_mean_w=276.8 prod_j_per_batch=93.1 prod_mh_per_w=0.1801 samples=43 steady_wall_s=8.304 steady_mhs=64.65 steady_mean_w=340.3 steady_j_per_batch=88.3 steady_mh_per_w=0.1900 samples=33 worker_cpu_s=3.13 worker_sha=9b36b523d0812c3c
5090-full-r5 sm MHz mean 2679 min 795 max 2880 over 52 samples
== 5090-select-r5 2026-10-08T19:45:38Z sm 2880 MHz, Not Active
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.393 prod_wall_s=10.019 prod_mhs=53.59 prod_mean_w=287.5 prod_j_per_batch=90.0 prod_mh_per_w=0.1864 samples=40 steady_wall_s=7.626 steady_mhs=70.40 steady_mean_w=360.8 steady_j_per_batch=86.0 steady_mh_per_w=0.1951 samples=30 worker_cpu_s=2.39 worker_sha=9b36b523d0812c3c
5090-select-r5 sm MHz mean 2771 min 2212 max 2880 over 49 samples
== end 2026-10-08T19:45:50Z

3
docs/analysis/review-2026-10-08-b/f10/5090-select-r1.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.447 prod_wall_s=10.070 prod_mhs=53.31 prod_mean_w=290.5 prod_j_per_batch=91.4 prod_mh_per_w=0.1835 samples=41 steady_wall_s=7.624 steady_mhs=70.42 steady_mean_w=362.5 steady_j_per_batch=86.4 steady_mh_per_w=0.1943 samples=31 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 236.71 ms, select 0.234 ms, read-back 0.178 ms, scan 0.005 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-select-r2.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.408 prod_wall_s=10.032 prod_mhs=53.52 prod_mean_w=289.1 prod_j_per_batch=90.6 prod_mh_per_w=0.1851 samples=40 steady_wall_s=7.624 steady_mhs=70.42 steady_mean_w=363.0 steady_j_per_batch=86.5 steady_mh_per_w=0.1940 samples=30 worker_cpu_s=0.00 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 236.69 ms, select 0.245 ms, read-back 0.270 ms, scan 0.007 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-select-r3.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.327 prod_wall_s=9.951 prod_mhs=53.95 prod_mean_w=290.5 prod_j_per_batch=90.3 prod_mh_per_w=0.1857 samples=40 steady_wall_s=7.624 steady_mhs=70.42 steady_mean_w=365.4 steady_j_per_batch=87.1 steady_mh_per_w=0.1927 samples=30 worker_cpu_s=2.33 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 236.71 ms, select 0.262 ms, read-back 0.353 ms, scan 0.006 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-select-r4.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.590 prod_wall_s=10.216 prod_mhs=52.55 prod_mean_w=285.4 prod_j_per_batch=91.1 prod_mh_per_w=0.1841 samples=41 steady_wall_s=7.626 steady_mhs=70.40 steady_mean_w=355.9 steady_j_per_batch=84.8 steady_mh_per_w=0.1978 samples=31 worker_cpu_s=2.55 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 236.63 ms, select 0.239 ms, read-back 0.320 ms, scan 0.010 ms; 32 jobs 32 chunks

3
docs/analysis/review-2026-10-08-b/f10/5090-select-r5.row Normal file
View file

@ -0,0 +1,3 @@
card=NVIDIA_GeForce_RTX_5090 mode=select jobs=32 batch_log2=24 hashes=536870912 found=519 setup_s=2.393 prod_wall_s=10.019 prod_mhs=53.59 prod_mean_w=287.5 prod_j_per_batch=90.0 prod_mh_per_w=0.1864 samples=40 steady_wall_s=7.626 steady_mhs=70.40 steady_mean_w=360.8 steady_j_per_batch=86.0 steady_mh_per_w=0.1951 samples=30 worker_cpu_s=2.39 worker_sha=9b36b523d0812c3c
ready: ready cuda NVIDIA_GeForce_RTX_5090 pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 16777216 regs 92 prepare 1 path nvrtc 12.8 driver 13.2 arch sm_120 variant base race off readback select worker 1.0 (4 October 2026)
transfers: transfers per chunk: down 536 B (select, 0 full fallbacks); mean per chunk: kernel+sync 236.78 ms, select 0.296 ms, read-back 0.244 ms, scan 0.007 ms; 32 jobs 32 chunks

283
docs/analysis/review-2026-10-08-b/f10/emu-test.log Normal file
View file

@ -0,0 +1,283 @@
igneum-pow export /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-b
seed "igneum-epoch/5e5d0c3b2a19f8e7d6c5b4a3928170f6e5d4c3b2a1908f7e6d5c4b3a2918f7e6/day/69676e65756d2d6461792ffa51000000000000", day "bytes:69676e65756d2d6461792ffa51000000000000", dataset 2^28 words (memory-hard), generator v2 attempt 0 program id 36e7763af4220724, loads/hash 128; epoch built in 376.1 ms
/home/build/wt/worker-select-pass/proto-cuda/nvrtc/worker.cpp:131:14: warning: ‘void* libSym(void*, const char*)’ defined but not used [-Wunused-function]
131 | static void* libSym(void* lib, const char* name) {
| ^~~~~~
/home/build/wt/worker-select-pass/proto-cuda/nvrtc/worker.cpp:124:14: warning: ‘void* libOpen(const std::string&)’ defined but not used [-Wunused-function]
124 | static void* libOpen(const std::string& name) {
| ^~~~~~~
/home/build/wt/worker-select-pass/proto-cuda/nvrtc/worker.cpp:109:20: warning: ‘std::string exeDir()’ defined but not used [-Wunused-function]
109 | static std::string exeDir() {
| ^~~~~~
built /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/igneum-worker-cuda-emu (CPU emulation, not a GPU build)
== --check pack A
info igneum-worker-cuda 1.0 (4 October 2026): device 0 CPU_emulation_shim_(not_a_GPU) (sm_120, 0 SMs), driver 12.8 from emulation (host threads, no GPU), NVRTC 12.8 from emulation (source recorded and checked, nothing compiled), target sm_120 (the device's architecture, listed by NVRTC)
info emu-nvrtc: kernel.cu source check PASS for pack 1: 7385 bytes handed over = the pack file's first 7385 of 8893 bytes; dropped tail = host launch wrappers only (1508 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info emu-nvrtc: kernel_bound.cu source check PASS for pack 1: 6003 bytes handed over = the pack file's first 6003 of 6887 bytes; dropped tail = host launch wrappers only (884 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
check PASS /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-a in 2679 ms: nvrtc 0 cache 16 dataset 2193 hot 0 check 470 race 0 ms variant base class v2; self-test PASS (cache head, last line and FNV-1a 64 448274a57f508cbc; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes)
epoch edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07 day 69676e65756d2d6461792ffa50000000000000, dataset 2^28 words, cache 2^26 words in 65536 segments, 0 registers, 0 blocks/SM at 1 warp(s)/block, target sm_120
== the annotation rule: without -default-device the stand-in must reject program.h's igneum_launch_* declarations as host code (what the RTX 5090 reported on 4 October 2026)
flagged as NVRTC does: program.h(46): cudaError_t igneum_launch_cache_fill(uint32_t* cache, uint32_t nSegments);
== --serve: two jobs on A, prepare B, a job on B (swap), a job on A after the swap (error), quit
info igneum-worker-cuda 1.0 (4 October 2026): device 0 CPU_emulation_shim_(not_a_GPU) (sm_120, 0 SMs), driver 12.8 from emulation (host threads, no GPU), NVRTC 12.8 from emulation (source recorded and checked, nothing compiled), target sm_120 (the device's architecture, listed by NVRTC)
info emu-nvrtc: kernel.cu source check PASS for pack 1: 7385 bytes handed over = the pack file's first 7385 of 8893 bytes; dropped tail = host launch wrappers only (1508 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info emu-nvrtc: kernel_bound.cu source check PASS for pack 1: 6003 bytes handed over = the pack file's first 6003 of 6887 bytes; dropped tail = host launch wrappers only (884 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info emu-nvrtc: igneum_select.cu source check PASS: 763 bytes, the select pass runs as a host loop
ready cuda CPU_emulation_shim_(not_a_GPU) pack igneum-epoch/edc4fa844da9dc98d37e965176f6558a31560e40502ab3ae5491b21aaaabfb07/day/69676e65756d2d6461792ffa50000000000000 dataset-log2 28 batch 8192 regs 0 prepare 1 path nvrtc 12.8 driver 12.8 arch sm_120 variant base race on readback select worker 1.0 (4 October 2026)
info readback select (per dispatch of 8192 nonces: the hits and 34 sentinel words come back; the GPU scans)
info first pack /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-a: nvrtc 0 cache 15 dataset 2686 hot 0 check 472 race 0 ms variant base class v2; self-test PASS (cache head, last line and FNV-1a 64 448274a57f508cbc; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes)
race edc4fa844da9dc98 device CPU_emulation_shim_(not_a_GPU) variants 1 base only, no race (emulation)
found 1 0 85f8352a2701deaa edc4fa844da9dc98
found 1 1 287bddfe85a71a6c edc4fa844da9dc98
found 1 2 fd65144d28e0e2b1 edc4fa844da9dc98
found 1 3 4c451942cdccec1a edc4fa844da9dc98
found 1 4 b974b6a86b1cf0ed edc4fa844da9dc98
found 1 5 c470d05d126a1a61 edc4fa844da9dc98
found 1 6 824f53a0342e593d edc4fa844da9dc98
found 1 7 ab83476b2d6aa8d0 edc4fa844da9dc98
found 1 8 bb1a171d76ebe772 edc4fa844da9dc98
found 1 9 952788c85aa6ba6a edc4fa844da9dc98
found 1 10 eab425f84eeb2a81 edc4fa844da9dc98
found 1 11 6279385502e9b189 edc4fa844da9dc98
found 1 12 92bb6782d9337a70 edc4fa844da9dc98
found 1 13 bf69940f96c7607b edc4fa844da9dc98
found 1 14 4ec115ccd6700418 edc4fa844da9dc98
found 1 15 aee365ce8b71d4fd edc4fa844da9dc98
found 1 16 3b1074a344f48d75 edc4fa844da9dc98
found 1 17 7bbba30cdeee2318 edc4fa844da9dc98
found 1 18 85917afa2fb98002 edc4fa844da9dc98
found 1 19 a57d924a33c9eb47 edc4fa844da9dc98
found 1 20 ddf5d02d5440bed5 edc4fa844da9dc98
found 1 21 e21722ea35922f3d edc4fa844da9dc98
found 1 22 7c71ff55e0202ea0 edc4fa844da9dc98
found 1 23 2e264b7a3db0424e edc4fa844da9dc98
found 1 24 737435f635d5cd13 edc4fa844da9dc98
found 1 25 26a94c56d64fa788 edc4fa844da9dc98
found 1 26 c5a88d468a03eebb edc4fa844da9dc98
found 1 27 31ca283df921550d edc4fa844da9dc98
found 1 28 06d3199ce7b1adf4 edc4fa844da9dc98
found 1 29 0c8f5f7605deecd1 edc4fa844da9dc98
found 1 30 c43156d1e4a5cbec edc4fa844da9dc98
found 1 31 78ba75b0e4c908d5 edc4fa844da9dc98
found 1 32 e40b3c4ce16e0674 edc4fa844da9dc98
found 1 33 9a6b697360a70406 edc4fa844da9dc98
found 1 34 e28eb3b8c845ef22 edc4fa844da9dc98
found 1 35 a8db66a07fa26a94 edc4fa844da9dc98
found 1 36 81d256fa67d8bed3 edc4fa844da9dc98
found 1 37 f681ad4e0426b0e5 edc4fa844da9dc98
found 1 38 ad49770893af9688 edc4fa844da9dc98
found 1 39 e830f95c3e013a41 edc4fa844da9dc98
found 1 40 7f050e695bd46847 edc4fa844da9dc98
found 1 41 b6c2ae03f8ddf3ae edc4fa844da9dc98
found 1 42 772509fa6f742f94 edc4fa844da9dc98
found 1 43 3d0317e357e2381d edc4fa844da9dc98
found 1 44 94f2baf0d5af5a6b edc4fa844da9dc98
found 1 45 3e2ba39651256a56 edc4fa844da9dc98
found 1 46 588295beedbc0aff edc4fa844da9dc98
found 1 47 edc862c227a7f27d edc4fa844da9dc98
found 1 48 a540ec001fd3fe33 edc4fa844da9dc98
found 1 49 3f3a2aa56b364df1 edc4fa844da9dc98
found 1 50 b395503049cfefa2 edc4fa844da9dc98
found 1 51 bf7a4473bf7be326 edc4fa844da9dc98
found 1 52 8af17d040c654b2a edc4fa844da9dc98
found 1 53 d2670d56b9848c44 edc4fa844da9dc98
found 1 54 ac5f448acfb856d5 edc4fa844da9dc98
found 1 55 89d2248ea07b109a edc4fa844da9dc98
found 1 56 dd4c5e4c6bb7c2e6 edc4fa844da9dc98
found 1 57 2529ceca58759d77 edc4fa844da9dc98
found 1 58 0bdff5ded9bc65fb edc4fa844da9dc98
found 1 59 6ea20ccedaf3e18c edc4fa844da9dc98
found 1 60 bedd6f14e5fa465d edc4fa844da9dc98
found 1 61 08d7d815f5786ced edc4fa844da9dc98
found 1 62 7d289bf14d911c5f edc4fa844da9dc98
found 1 63 e2f4237e31cbd041 edc4fa844da9dc98
done 1 64 9.66
found 2 4294967264 7b6273e5bd32d0b5 edc4fa844da9dc98
found 2 4294967265 38ca4cce09bfb623 edc4fa844da9dc98
found 2 4294967266 d8b0ed94f45cdbf7 edc4fa844da9dc98
found 2 4294967267 9d8faff62d0efa15 edc4fa844da9dc98
found 2 4294967268 1df806f32926defe edc4fa844da9dc98
found 2 4294967269 e3654b2d705786bb edc4fa844da9dc98
found 2 4294967270 ab6741286c7c88dd edc4fa844da9dc98
found 2 4294967271 a15815bf414e9320 edc4fa844da9dc98
found 2 4294967272 6f1cc5a7a2fcc7c0 edc4fa844da9dc98
found 2 4294967273 6e98c101317928c6 edc4fa844da9dc98
found 2 4294967274 d87e3fefd2a7b381 edc4fa844da9dc98
found 2 4294967275 f7f30848fd476fb5 edc4fa844da9dc98
found 2 4294967276 257fc4d2e3603beb edc4fa844da9dc98
found 2 4294967277 1aab6f77c1365d4c edc4fa844da9dc98
found 2 4294967278 1c9d005f8f4ac463 edc4fa844da9dc98
found 2 4294967279 161427ad7f81f2f9 edc4fa844da9dc98
found 2 4294967280 e8611bc03fb827c1 edc4fa844da9dc98
found 2 4294967281 42454a79caee8c69 edc4fa844da9dc98
found 2 4294967282 190aa2b3529b589f edc4fa844da9dc98
found 2 4294967283 83d56d283df7f1e0 edc4fa844da9dc98
found 2 4294967284 fb355b9c51ca8988 edc4fa844da9dc98
found 2 4294967285 f5b626e35d01424a edc4fa844da9dc98
found 2 4294967286 8a0b81e4053fe9b0 edc4fa844da9dc98
found 2 4294967287 c1aa110e3db24ed7 edc4fa844da9dc98
found 2 4294967288 fc53be92ceb49a8e edc4fa844da9dc98
found 2 4294967289 c3296889fb569a0a edc4fa844da9dc98
found 2 4294967290 e38850154c2dea6c edc4fa844da9dc98
found 2 4294967291 d99fda96ee76c555 edc4fa844da9dc98
found 2 4294967292 fac8684a83d19072 edc4fa844da9dc98
found 2 4294967293 51085b8a57f3720d edc4fa844da9dc98
found 2 4294967294 7311e9864d96be76 edc4fa844da9dc98
found 2 4294967295 2c52ef30b1f52cbf edc4fa844da9dc98
found 2 4294967296 aaee181f46a06d70 edc4fa844da9dc98
found 2 4294967297 aa45e128a7ab498b edc4fa844da9dc98
found 2 4294967298 eb0f5f050141457c edc4fa844da9dc98
found 2 4294967299 7be5e2d3e944d761 edc4fa844da9dc98
found 2 4294967300 804964e4293faa67 edc4fa844da9dc98
found 2 4294967301 7567f837977d77da edc4fa844da9dc98
found 2 4294967302 1ff1311a643d3b43 edc4fa844da9dc98
found 2 4294967303 60cda5dadaaa941c edc4fa844da9dc98
found 2 4294967304 63542e063db3fb34 edc4fa844da9dc98
found 2 4294967305 e0ac767c5fab1ab4 edc4fa844da9dc98
found 2 4294967306 e2423cd93bfb5aa0 edc4fa844da9dc98
found 2 4294967307 c3210435edd81d5a edc4fa844da9dc98
found 2 4294967308 a0226e9f79d09099 edc4fa844da9dc98
found 2 4294967309 fc143bbd8555fdb0 edc4fa844da9dc98
found 2 4294967310 cc9008673f527726 edc4fa844da9dc98
found 2 4294967311 bb329b9bcf8e66fd edc4fa844da9dc98
found 2 4294967312 164442e9130c7b7a edc4fa844da9dc98
found 2 4294967313 48e43b690541132d edc4fa844da9dc98
found 2 4294967314 f4ba8f420da3337e edc4fa844da9dc98
found 2 4294967315 050dff2f1846ab61 edc4fa844da9dc98
found 2 4294967316 74972ac9ab8fb364 edc4fa844da9dc98
found 2 4294967317 b7ef1e3155e9a7f3 edc4fa844da9dc98
found 2 4294967318 3ecb2b74489fc8a4 edc4fa844da9dc98
found 2 4294967319 64fa8c8296a3d662 edc4fa844da9dc98
found 2 4294967320 6c4411933e7f690c edc4fa844da9dc98
found 2 4294967321 23234ffc83690177 edc4fa844da9dc98
found 2 4294967322 01dd1d050f87e26e edc4fa844da9dc98
found 2 4294967323 e3a05b34dde32af0 edc4fa844da9dc98
found 2 4294967324 0bcf244d4c73562d edc4fa844da9dc98
found 2 4294967325 375800d4039b5c1f edc4fa844da9dc98
found 2 4294967326 42b9a4c28e42c996 edc4fa844da9dc98
found 2 4294967327 f56d052d60a7338e edc4fa844da9dc98
done 2 64 10.89
info prepare started for epoch 5e5d0c3b2a19f8e7 day 69676e65756d2d6461792ffa51000000000000 from /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-b (NVRTC sm_120 in the background)
info emu-nvrtc: kernel.cu source check PASS for pack 2: 7206 bytes handed over = the pack file's first 7206 of 8714 bytes; dropped tail = host launch wrappers only (1508 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info emu-nvrtc: kernel_bound.cu source check PASS for pack 2: 5824 bytes handed over = the pack file's first 5824 of 6708 bytes; dropped tail = host launch wrappers only (884 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
race 5e5d0c3b2a19f8e7 device CPU_emulation_shim_(not_a_GPU) variants 1 base only, no race (emulation)
prepared 5e5d0c3b2a19f8e7d6c5b4a3928170f6e5d4c3b2a1908f7e6d5c4b3a2918f7e6 69676e65756d2d6461792ffa51000000000000 36806.6 nvrtc 0 cache 15 dataset 1535 hot 0 check 462 race 0 ms variant base class v2; self-test PASS (cache head, last line and FNV-1a 64 b5b0660c9db7ce59; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes) resident 2 programs 2 datasets
found 3 128 72b866f3367df254 edc4fa844da9dc98
found 3 129 a9772a8b6a793277 edc4fa844da9dc98
found 3 130 d07426082b75402d edc4fa844da9dc98
found 3 131 408083ae76ee518b edc4fa844da9dc98
found 3 132 c45e395f38ad22d8 edc4fa844da9dc98
found 3 133 3dbee02765058ade edc4fa844da9dc98
found 3 134 cd5f25857807cfa9 edc4fa844da9dc98
found 3 135 879e1e3fe942214a edc4fa844da9dc98
found 3 136 d2133aca0cc47d7a edc4fa844da9dc98
found 3 137 b143a77f8b19741c edc4fa844da9dc98
found 3 138 d3ed2304cc49a6c7 edc4fa844da9dc98
found 3 139 75cf3d6584e42e65 edc4fa844da9dc98
found 3 140 86db0130d062a7b7 edc4fa844da9dc98
found 3 141 05f62574da3354c5 edc4fa844da9dc98
found 3 142 7c961bec00e0cc4b edc4fa844da9dc98
found 3 143 db8c0f4e55c387ad edc4fa844da9dc98
found 3 144 7f1454c90435afbd edc4fa844da9dc98
found 3 145 002b3ca8c7bfd154 edc4fa844da9dc98
found 3 146 edae0b061a210205 edc4fa844da9dc98
found 3 147 e478cf93a9878f51 edc4fa844da9dc98
found 3 148 32c7dfd0fe55a026 edc4fa844da9dc98
found 3 149 06b93d72fb16404c edc4fa844da9dc98
found 3 150 4b6cace41d98e7fd edc4fa844da9dc98
found 3 151 41e2649e041acc06 edc4fa844da9dc98
found 3 152 0453f57fae7f514a edc4fa844da9dc98
found 3 153 ff01e0015a3f39e9 edc4fa844da9dc98
found 3 154 491ae49020261797 edc4fa844da9dc98
found 3 155 ef1bd996a361128a edc4fa844da9dc98
found 3 156 794082f3f114767d edc4fa844da9dc98
found 3 157 5822e90044ea782a edc4fa844da9dc98
found 3 158 259700c74b5cc149 edc4fa844da9dc98
found 3 159 235a57ae0bb5a2ed edc4fa844da9dc98
done 3 32 9.91
info switched to the prepared pair epoch 5e5d0c3b2a19f8e7 day 69676e65756d2d6461792ffa51000000000000 (class v2) in 0.00 ms
found 4 0 992153a3ef2b07fc 5e5d0c3b2a19f8e7
found 4 1 55e25d6fb49f4351 5e5d0c3b2a19f8e7
found 4 2 54a9131bc3066ee4 5e5d0c3b2a19f8e7
found 4 3 1c027229270c7d33 5e5d0c3b2a19f8e7
found 4 4 3806eac5cad07581 5e5d0c3b2a19f8e7
found 4 5 6a87c5563912348c 5e5d0c3b2a19f8e7
found 4 6 0b802423721bc6af 5e5d0c3b2a19f8e7
found 4 7 27ec87f124a88548 5e5d0c3b2a19f8e7
found 4 8 474bf5241c3a8ec3 5e5d0c3b2a19f8e7
found 4 9 532dbcf0caa93f96 5e5d0c3b2a19f8e7
found 4 10 259186fa7df9be63 5e5d0c3b2a19f8e7
found 4 11 f3b8c0df8787ff76 5e5d0c3b2a19f8e7
found 4 12 0e25eb13e9ddb276 5e5d0c3b2a19f8e7
found 4 13 a28c32bc4aeabf61 5e5d0c3b2a19f8e7
found 4 14 1e9c1a2be3f3f110 5e5d0c3b2a19f8e7
found 4 15 9d4a1cfde84d3aba 5e5d0c3b2a19f8e7
found 4 16 095727897ed26f5a 5e5d0c3b2a19f8e7
found 4 17 fdcbce6e7d66f667 5e5d0c3b2a19f8e7
found 4 18 aefcdcf46d9e6a4a 5e5d0c3b2a19f8e7
found 4 19 c6ec5a44a97ca398 5e5d0c3b2a19f8e7
found 4 20 50fce9ed2020b450 5e5d0c3b2a19f8e7
found 4 21 8c2fe5af5ee1e9f3 5e5d0c3b2a19f8e7
found 4 22 500ad998c15f42b6 5e5d0c3b2a19f8e7
found 4 23 95965bedaeaa462d 5e5d0c3b2a19f8e7
found 4 24 df645bcb6d1cc3ad 5e5d0c3b2a19f8e7
found 4 25 716b7f5dcc5c30dc 5e5d0c3b2a19f8e7
found 4 26 84fcbe9758749444 5e5d0c3b2a19f8e7
found 4 27 29511da3232f2dff 5e5d0c3b2a19f8e7
found 4 28 b976fd6a159db063 5e5d0c3b2a19f8e7
found 4 29 69889d22e3cb6709 5e5d0c3b2a19f8e7
found 4 30 151cc4a784c80add 5e5d0c3b2a19f8e7
found 4 31 27b10106d08793a5 5e5d0c3b2a19f8e7
done 4 32 8.22
info dropped the previous pair (its program, cache and dataset)
info job 5 is for epoch edc4fa844da9dc98 day 69676e65756d2d6461792ffa50000000000000, which is not resident; building its pack /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-a now (foreground)
info emu-nvrtc: kernel.cu source check PASS for pack 1: 7385 bytes handed over = the pack file's first 7385 of 8893 bytes; dropped tail = host launch wrappers only (1508 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info emu-nvrtc: kernel_bound.cu source check PASS for pack 1: 6003 bytes handed over = the pack file's first 6003 of 6887 bytes; dropped tail = host launch wrappers only (884 bytes); program.h (3024 bytes) and memhard.h (5848 bytes) byte-identical
info built /home/build/wt/worker-select-pass/proto-cuda/nvrtc/emu/build/pack-a: nvrtc 0 cache 15 dataset 1295 hot 0 check 481 race 0 ms variant base class v2; self-test PASS (cache head, last line and FNV-1a 64 448274a57f508cbc; dataset head, word [268435455] and 64 samples; 96 of 96 vector lanes)
race edc4fa844da9dc98 device CPU_emulation_shim_(not_a_GPU) variants 1 base only, no race (emulation)
info switched to the prepared pair epoch edc4fa844da9dc98 day 69676e65756d2d6461792ffa50000000000000 (class v2) in 1791.36 ms
found 5 96 cb13242c3a1a4ffa edc4fa844da9dc98
found 5 97 f3071494c5382b79 edc4fa844da9dc98
found 5 98 27d7c3cf7cec0c32 edc4fa844da9dc98
found 5 99 4137aaf4e3ec29e1 edc4fa844da9dc98
found 5 100 3e18fe3b1190d59f edc4fa844da9dc98
found 5 101 c1d940643197c68e edc4fa844da9dc98
found 5 102 005901b495cc73be edc4fa844da9dc98
found 5 103 2d18e57bfa59ce02 edc4fa844da9dc98
found 5 104 326bc2b1e5eec1d3 edc4fa844da9dc98
found 5 105 84b754dc15b156b0 edc4fa844da9dc98
found 5 106 bdb2c9a6d88c78ce edc4fa844da9dc98
found 5 107 89d57198d7f17874 edc4fa844da9dc98
found 5 108 3c96f431799d9d97 edc4fa844da9dc98
found 5 109 2d98f14c5b8153a4 edc4fa844da9dc98
found 5 110 a6d1bf98ddda41fb edc4fa844da9dc98
found 5 111 cd14d9e675dd60c0 edc4fa844da9dc98
found 5 112 349128ba141dd961 edc4fa844da9dc98
found 5 113 af2496d2e09bce14 edc4fa844da9dc98
found 5 114 59ca36336962df25 edc4fa844da9dc98
found 5 115 f44a3eab722df34b edc4fa844da9dc98
found 5 116 f7c766cb70017bb1 edc4fa844da9dc98
found 5 117 4fcabac398cca930 edc4fa844da9dc98
found 5 118 f2e1cb0a1478ed76 edc4fa844da9dc98
found 5 119 5fc703e08569cf0d edc4fa844da9dc98
found 5 120 4aa1b3c1b264ff4e edc4fa844da9dc98
found 5 121 008922eb5f1b0558 edc4fa844da9dc98
found 5 122 7a00d435fa182ea4 edc4fa844da9dc98
found 5 123 347fd45b06777e05 edc4fa844da9dc98
found 5 124 86e87b980be70737 edc4fa844da9dc98
found 5 125 7cdbe6e309e01af7 edc4fa844da9dc98
found 5 126 22de307670d9ee71 edc4fa844da9dc98
found 5 127 d582b681bf406666 edc4fa844da9dc98
done 5 32 1800.01
info dropped the previous pair (its program, cache and dataset)
info transfers per chunk: down 873 B (select, 0 full fallbacks); mean per chunk: kernel+sync 7.87 ms, select 0.001 ms, read-back 0.001 ms, scan 0.008 ms; 5 jobs 6 chunks
== verdict
PASS: ready + prepare 1; 64 + 64 + 32 found on pack A, prepared B with self-test PASS, swapped, 32 found on B, job 5 served after the worker rebuilt pack A by itself (self-heal); 17 sampled hashes (both packs, both sides of the 32-bit nonce boundary) equal igneum-pow hash-bound
NVRTC source check PASS for 7 files in --serve, 2 in --check
== the select pass (8 October 2026): --readback select equals the full read-back, known-failed first (emu/select-check.sh)
== select_fault_detected (known-failed first: the counter is not reset from the second chunk on)
flagged: 84 stale or extra lines in the fault run
== found_set_equals_full_* (every case, line for line)
PASS: select = full over 2255 found lines and 10 jobs (zero hits, partial, all-hit overflow with 2 full fallbacks, a 96-nonce tail, the high-32 rollover, two queued jobs, hits then zero, a mismatch error then a job); the fault run was flagged first
transfers: transfers per chunk: down 1234 B (select, 2 full fallbacks); mean per chunk: kernel+sync 149.78 ms, select 0.003 ms, read-back 0.001 ms, scan 0.052 ms; 10 jobs 20 chunks
transfers: transfers per chunk: down 7411 B (full, 0 full fallbacks); mean per chunk: kernel+sync 142.04 ms, select 0.000 ms, read-back 0.002 ms, scan 0.050 ms; 10 jobs 20 chunks

31
docs/analysis/review-2026-10-08-b/f10/geometry-check-mini.log Normal file
View file

@ -0,0 +1,31 @@
ok geometry ds55: loads as 1476395008 words, 92274688 items, mulshift 1, 5905580032 bytes
ok geometry ds85: loads as 2281701376 words, 142606336 items, mulshift 1, 9126805504 bytes
ok geometry ds115: loads as 3087007744 words, 192937984 items, mulshift 1, 12348030976 bytes
ok geometry legacy-2^28: loads as 268435456 words, 16777216 items, mulshift 0, 1073741824 bytes
ok geometry legacy-2^30: loads as 1073741824 words, 67108864 items, mulshift 0, 4294967296 bytes
ok geometry legacy-2^31-explicit-count: loads as 2147483648 words, 134217728 items, mulshift 0, 8589934592 bytes
ok geometry wrong-count-not-a-65536-multiple: refused with "multiple of 65,536"
ok geometry wrong-log2-for-the-count: refused with "IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count"
ok geometry wrong-log2-for-a-power-of-two-count: refused with "IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count"
ok geometry wrong-items: refused with "IGNEUM_DATASET_ITEMS is not IGNEUM_DATASET_WORDS / 16"
ok geometry wrong-mulshift-on-a-non-power-of-two: refused with "IGNEUM_DATASET_MULSHIFT does not match the word count"
ok geometry log2-32: refused with "IGNEUM_DATASET_LOG2 above 31"
ok geometry overflowing-count-2^32: refused with "IGNEUM_DATASET_WORDS above 2^32 - 1"
ok geometry overflowing-count-2^40: refused with "IGNEUM_DATASET_WORDS above 2^32 - 1"
ok geometry count-below-the-minimum: refused with "multiple of 65,536"
ok geometry log2-below-20: refused with "sizes out of range"
ok geometry missing-log2: refused with "has no IGNEUM_DATASET_LOG2"
ok geometry missing-leaves-on-a-class-v5-pack: refused with "without IGNEUM_STATE_LEAVES"
ok geometry leaves-on-a-class-v2-pack: refused with "state leaves belong to class v5"
ok the vectors file carries every case (18 or more): 19
ok pack /Users/igneum/Projects/igneum/proto-cuda/packs/igneum-devnet-v4-epoch0: 268435456 words (2^28), 16777216 items, 1073741824 bytes
ok pack /Users/igneum/Projects/igneum/proto-cuda/packs-ca3-v5/v5-dn3-epoch0: 268435456 words (2^28), 16777216 items, 1073741824 bytes
geometry-check: 19 vectors, 2 packs, 0 failure(s)
== Metal serve smoke, class v5 pack (packs-ca3-v5/v5-dn3-epoch0), the exact-geometry pack path
ready metal Apple_M6 dataset-log2 28 batch 4194304 prepare 1 race off
prepared 4020cb4382e3fe4b281c817c02582e147d8f851f566ae9172b28912b8e68b925 69676e65756d2d6461792ffd50000000000000 241.1 program 81.3 dataset 159.8 race 0.0 variant base class v5 loads/hash 128 cache-fill 4.0 build 95.8 resident 1 programs 1 datasets
found 1 0 c6e114e3d317ee3e 4020cb4382e3fe4b
found 1 63 77fe95d9ed898030 4020cb4382e3fe4b
done 1 64 54.50
== Metal serve smoke, the v4 pack (igneum-devnet-v4-epoch0, class v2 generator path): ready metal Apple_M6 ...; found 1 0 65413452d4d05a3c edc4fa844da9dc98; found 1 63 4b8c6966db6ac21f edc4fa844da9dc98; done 1 64 61.69
== build: swiftc -O main.swift on the mini (Apple M6, CLT swift 6.4), Thu Oct 8 20:51:52 BST 2026

75
docs/analysis/review-2026-10-08-b/f10/packfile-test.log Normal file
View file

@ -0,0 +1,75 @@
ok the fixture seeds unhex (32 epoch bytes, 19 day bytes)
ok attempt 0 of epoch 34 = the bare seed words (06af2a61 4d67274e ...)
ok attempt 1 of epoch 34 = words of seed || 01000000 (0dcff56b 6b1beb0d ...)
ok the two attempts differ
ok known-good: the attempt-1 pack loads
ok known-good: attempt, words and epoch hex read back
ok known-good: the checked-in attempt-0 pack loads from program.h's own bytes
ok the checked-in pack is attempt 0
ok known-mismatched: bare words under attempt 1 are refused
ok the refusal names the attempt and the epoch in plain words
(program pack and its seeds disagree: IGNEUM_SEEDW_INIT is not attempt 1 of the epoch seed 009858237e118f69 (attempt 1 gives 0dcff56b 6b1beb0d ..., the pack has 06af2a61 4d67274e ...); run igneum-miner export-pack again)
ok known-mismatched: seeds.txt of epoch 33 with program.h of epoch 34 is refused
ok the refusal says the files disagree
ok no attempt line reads as attempt 0 and the bare words load
ok a generator 2 pack without a class line is class v2 with no era
ok known-failed: a pack with IGNEUM_DATASET_WORDS is sized by the word count (1476395008 words, 92274688 items, multiply-shift), not by 1 << 30
ok the count is not the floor's power of two and the items are words / 16
ok known-good: a pack without the line keeps 1 << IGNEUM_DATASET_LOG2 (2^28 words, 2^24 items, the mask path)
ok known-good: the checked-in pack is sized by its power of two
ok generator 9 is refused in plain words
ok a generator 3 pack loads as class v3 with its era seed
ok the v3 pack matches a job naming class v3 and its era
ok a job naming no class accepts the pack
ok a job naming class v2 refuses the v3 pack
ok a job naming another era refuses the v3 pack
ok a job naming class v3 refuses a v2 pack
ok an era named on a v2 job is ignored
ok a generator 4 pack loads as class v4 with its era seed
ok a generator 3 pack with a shadow block (a v4 program stamped v3) is refused
ok a generator 4 pack without the shadow block is refused
ok the v4 pack with its block loads again
ok the v4 pack matches a job naming class v4 and its era
ok a job naming class v3 refuses the v4 pack
ok a job naming another era refuses the v4 pack
ok a job naming class v4 refuses a v3 pack
ok the class=v4 token is read
ok a v3 class line on a generator 4 pack is refused
ok a class line that contradicts the generator is refused
ok a pack with no generator line (generator 1) is refused
ok known-failed: a generator 5 pack without the leaf count is refused in plain words
ok a generator 5 pack with the leaf count loads as class v5 (count, FNV and file name read back)
ok the v5 pack matches a job naming class v5 and its era
ok a job naming class v4 refuses the v5 pack
ok known-failed: the leaves file missing refuses the build in plain words
ok known-failed: a short leaves file is refused with both sizes named
ok known-failed: leaves of another root (one bit) are refused by the FNV
ok known-good: the right leaves load as 2 x 16 words
ok known-failed: a generator 4 pack carrying IGNEUM_STATE_LEAVES is refused
ok a class v4 pack has no leaves and the leaf loader hands back none
ok known-good: the checked-in class v5 pack loads with its leaf count and vectors
ok known-good: the checked-in pack's leaves.bin loads under the pack's FNV-1a 64
v5 pack: 11 leaves, FNV-1a 64 f141bfee2a8b11b0, state root 7e37a9fb19b154d32daf5bf30a50d339a75029fbc9eec9ea20e95439dba5a311
ok the geometry vectors file opens
ok geometry ds55: loads as 1476395008 words, 92274688 items, mulshift 1, 5905580032 bytes
ok geometry ds85: loads as 2281701376 words, 142606336 items, mulshift 1, 9126805504 bytes
ok geometry ds115: loads as 3087007744 words, 192937984 items, mulshift 1, 12348030976 bytes
ok geometry legacy-2^28: loads as 268435456 words, 16777216 items, mulshift 0, 1073741824 bytes
ok geometry legacy-2^30: loads as 1073741824 words, 67108864 items, mulshift 0, 4294967296 bytes
ok geometry legacy-2^31-explicit-count: loads as 2147483648 words, 134217728 items, mulshift 0, 8589934592 bytes
ok geometry wrong-count-not-a-65536-multiple: refused with "multiple of 65,536"
ok geometry wrong-log2-for-the-count: refused with "IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count"
ok geometry wrong-log2-for-a-power-of-two-count: refused with "IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count"
ok geometry wrong-items: refused with "IGNEUM_DATASET_ITEMS is not IGNEUM_DATASET_WORDS / 16"
ok geometry wrong-mulshift-on-a-non-power-of-two: refused with "IGNEUM_DATASET_MULSHIFT does not match the word count"
ok geometry log2-32: refused with "IGNEUM_DATASET_LOG2 above 31"
ok geometry overflowing-count-2^32: refused with "IGNEUM_DATASET_WORDS above 2^32 - 1"
ok geometry overflowing-count-2^40: refused with "IGNEUM_DATASET_WORDS above 2^32 - 1"
ok geometry count-below-the-minimum: refused with "multiple of 65,536"
ok geometry log2-below-20: refused with "sizes out of range"
ok geometry missing-log2: refused with "has no IGNEUM_DATASET_LOG2"
ok geometry missing-leaves-on-a-class-v5-pack: refused with "without IGNEUM_STATE_LEAVES"
ok geometry leaves-on-a-class-v2-pack: refused with "state leaves belong to class v5"
ok the vectors file carries every case (18 or more)
19 geometry vectors
/tmp/igneum-packfile-test/packfile-test: 0 failure(s)

109
docs/analysis/review-2026-10-08-b/f10/rows.md Normal file
View file

@ -0,0 +1,109 @@
# F10 worker lane: the CUDA select pass, measured (8 October 2026)
Review B finding F10 (docs/analysis/review-2026-10-08-b/findings.json): the CUDA serving loop read 8 bytes per nonce back
over the bus (128 MiB per 2^24-nonce dispatch) and scanned them on the host under the GPU mutex. The select pass the
OpenCL worker has carried since 5 October 2026 is now on CUDA (proto-cuda/nvrtc/worker.cpp, landed on master as
964cdf6c, commits 12cb65cc and 4b0de631): a tiny `igneum_select` kernel built once by NVRTC writes the hits behind an
atomic counter plus 34 sentinel words, the host reads back a few hundred bytes, a chunk with more than 256 hits falls
back to the full read, the found lines come in nonce order from the sorted hits and carry the kernel's own hash word.
Byte-preserving: the bound kernel and its result layout are untouched; no valid hash changes.
Two numbers per card, never interchangeable: the **kernel timer** (`--bench`, the dispatch alone) and the
**serving-mode numbers** (`proto-cuda/nvrtc/serve-bench.sh`: the worker in `--serve` driven by a stub pool exactly as
the miner drives it, one job queued behind the one in flight, every found line read at the miner's submit point). The
**production** number runs from the worker process start (NVRTC compile, cache, dataset, self-test included) to the last
done line; the **steady** number from the ready line. Power is nvidia-smi's board reading at 4 Hz over the same window
(a RunPod container exposes no host rail; the worker's own CPU seconds stand in for the host-side cost of the read-back
and scan). Target 00000fffffffffff (one hit per 2^20 nonces, 16 found lines per job), 32 jobs of 2^24 nonces, `--race off`,
variant base, pack hl-v6-all (class v6 kit packs-class-v6-20261008T162324Z, class string
mx8-erad810f22d+sh256x27+state+reg64c+fold+rw). Worker binary sha256 9b36b523d0812c3c (Ubuntu 24.04 build of 12cb65cc).
The worker has no stale-job cancellation (a job runs to its done line; the next line waits), so the stub's queued job is
the stale case the miner produces; nothing is cancelled mid-dispatch on any backend today.
## Equivalence (build-7, CPU emulation, proto-cuda/nvrtc/emu/test.sh, green 20:20 and again 20:55 UK)
`emu/select-check.sh`, known-failed first: the fault run (IGNEUM_READBACK_FAULT_TEST skips the counter reset from the
second chunk on) is flagged with 84 stale lines; then `--readback select` equals the full read line for line over 2255
found lines and 10 jobs: zero hits, partial hits, the all-hit overflow (every chunk over 256 hits, 2 full fallbacks), a
96-nonce tail with a 32-lane launch (POW-01 in the record), the high-32 rollover (ROT-01 in the record), two queued jobs,
hits then a zero chunk, a mismatch error then a job, and every found line's fifth field equal to the pack's epoch (I05).
`emu/serve-check.sh`: 17 sampled hashes on both packs and both sides of the 32-bit boundary equal igneum-pow hash-bound;
the swap case's found lines carry pack B's epoch. Logs: emu-test.log beside this file.
## Kernel timer
| card | arch | ptxas (nvcc 12.8, the pack's kernel_bound.cu) | --bench 5 x 2^24 | kernel MH/s |
|---|---|---|---|---|
| RTX 5090 (pod 4jpym0xwr1jnf3, driver 595.91.07) | sm_120 | 92 registers, 0 bytes spill stores, 0 bytes spill loads, 0 bytes stack; worker: 20 blocks/SM, 3400 resident warps | 236.75 ms | 70.86 |
| RTX 3090 (pod xq411hlgu4qzri, driver 595.71.05) | sm_86 | 87 registers, 0 bytes spill stores, 0 bytes spill loads, 0 bytes stack, 408 bytes cmem; worker: 16 blocks/SM, 1312 resident warps | warm-up 554.73 ms, timed mean 2590.06 ms (throttled, see below) | 30.2 at clock, 6.5 throttled |
## Serving mode
The pod rows (the .row files beside this file carry every field; the clocks are nvidia-smi clocks.sm means over each run):
### 5090
Mean SM clock 2387 to 2823 MHz over the runs (2895 max); no throttle reason active.
| run | mode | setup s | production MH/s | production J per batch | steady MH/s | steady W | steady J per batch | steady MH per W | worker CPU s | per chunk: kernel+sync, select, read-back, scan ms |
|---|---|---|---|---|---|---|---|---|---|---|
| 5090-full-r1 | full | 2.275 | 50.10 | 93.1 | 63.61 | 328.3 | 86.6 | 0.1938 | (subshell pid, not read) | 236.64, 0.000, 12.794, 13.245 |
| 5090-full-r2 | full | 2.373 | 47.07 | 96.5 | 59.45 | 326.9 | 92.3 | 0.1819 | (subshell pid, not read) | 236.65, 0.000, 25.097, 19.536 |
| 5090-full-r3 | full | 2.400 | 50.73 | 88.5 | 65.62 | 335.4 | 85.8 | 0.1956 | 2.97 | 236.73, 0.000, 7.679, 10.341 |
| 5090-full-r4 | full | 2.334 | 50.81 | 94.4 | 65.21 | 351.5 | 90.4 | 0.1855 | 2.92 | 237.01, 0.000, 8.639, 10.488 |
| 5090-full-r5 | full | 2.463 | 49.86 | 93.1 | 64.65 | 340.3 | 88.3 | 0.1900 | 3.13 | 236.70, 0.000, 9.028, 12.796 |
| 5090-select-r1 | select | 2.447 | 53.31 | 91.4 | 70.42 | 362.5 | 86.4 | 0.1943 | (subshell pid, not read) | 236.71, 0.234, 0.178, 0.005 |
| 5090-select-r2 | select | 2.408 | 53.52 | 90.6 | 70.42 | 363.0 | 86.5 | 0.1940 | (subshell pid, not read) | 236.69, 0.245, 0.270, 0.007 |
| 5090-select-r3 | select | 2.327 | 53.95 | 90.3 | 70.42 | 365.4 | 87.1 | 0.1927 | 2.33 | 236.71, 0.262, 0.353, 0.006 |
| 5090-select-r4 | select | 2.590 | 52.55 | 91.1 | 70.40 | 355.9 | 84.8 | 0.1978 | 2.55 | 236.63, 0.239, 0.320, 0.010 |
| 5090-select-r5 | select | 2.393 | 53.59 | 90.0 | 70.40 | 360.8 | 86.0 | 0.1951 | 2.39 | 236.78, 0.296, 0.244, 0.007 |
5090 reading (five pairs): steady 70.41 MH/s select against 63.71 full (+10.5 percent; select sits at the kernel timer's
70.86, the read-back cost is gone); production 53.38 against 49.71 (+7.4 percent; the rest of the gap to the kernel is
the 2.3 to 2.6 s setup of each 7.6 s run); energy per 2^24 batch 86.2 J select against 88.7 J full (-2.8 percent; the card
draws more while it is never idle, 361 W against 337 W, and finishes sooner); 0.195 against 0.188 MH per W; worker CPU
2.3 to 2.6 s against 2.9 to 3.1 s per 32 jobs. The full path pays 7.7 to 25.1 ms of read-back plus 10.3 to 19.5 ms of host
scan per 236.7 ms chunk (8 to 19 percent); the select pass costs 0.23 to 0.30 ms plus 0.18 to 0.35 ms of read-back.
### 3090
The pod's host holds the card at 210 MHz under `clocks_event_reasons.sw_power_cap` for whole runs (Active at 21 W idle;
`-lgc` and `-pl` refused in the container), so the rows split by the mean SM clock of each run.
| run | mode | mean SM MHz | setup s | production MH/s | production J per batch | steady MH/s | steady W | steady J per batch | steady MH per W | worker CPU s | per chunk: kernel+sync, select, read-back, scan ms |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 3090-full-r3 | full | 1769 | 4.110 | 22.81 | 207.7 | 27.63 | 325.9 | 197.9 | 0.0848 | 5.61 | 557.15, 0.000, 24.225, 23.753 |
| 3090-select-r2 | select | 1733 | 4.138 | 24.41 | 196.7 | 30.06 | 330.7 | 184.6 | 0.0909 | (subshell pid, not read) | 555.70, 0.268, 0.217, 0.013 |
| 3090-full-r1 | full | 632 | 4.710 | 9.18 | 301.4 | 9.98 | 168.3 | 282.9 | 0.0593 | (subshell pid, not read) | 1630.34, 0.000, 24.723, 24.119 |
| 3090-full-r2 | full | 210 | 4.231 | 5.96 | 378.4 | 6.25 | 136.9 | 367.3 | 0.0457 | (subshell pid, not read) | 2637.86, 0.000, 22.784, 20.625 |
| 3090-full-r4 | full | 210 | 4.397 | 5.96 | 377.9 | 6.27 | 136.7 | 365.9 | 0.0459 | 5.81 | 2624.25, 0.000, 27.536, 22.950 |
| 3090-select-r1 | select | 210 | 7.093 | 5.86 | 376.9 | 6.35 | 136.9 | 361.5 | 0.0464 | (subshell pid, not read) | 2637.57, 0.532, 0.296, 0.019 |
| 3090-select-r3 | select | 318 | 4.297 | 6.44 | 361.6 | 6.78 | 140.1 | 346.4 | 0.0484 | 4.23 | 2470.39, 0.503, 0.206, 0.017 |
3090 reading (the one pair at the card's clock): steady 30.06 MH/s select against 27.63 full (+8.8 percent); production
24.41 against 22.81 (+7.0 percent); 184.6 J per batch against 197.9 (-6.7 percent); the full path's read-back plus scan is
48 ms of a 557 ms chunk (8.6 percent). The 6.5 MH/s kernel read of the evening is the 210 MHz pod state, not the class:
ptxas on sm_86 shows 87 registers and no spill or stack bytes (sm_120: 92 registers, none); the worker's occupancy is 16
blocks of one warp per SM on sm_86 against 20 on sm_120, register-limited on both. At clock the 5090 to 3090 gap is 2.34x
(70.4 to 30.1), beside the memory gap (1,792 to 936 GB/s, 96 MiB to 6 MiB of L2); the 10.9x read was the power cap. A
consumer-tier number for the site needs a 3090 whose clock the host does not cap; the fleet lane holds the pod id.
## Record
Harness cells (tools/ci/test-map.json): `harness:cuda-select-check` (ROT-01 partial: the high-32 rollover case; POW-01
partial: the select pass equals the full read on every case named above, the hash sample against igneum-pow) and
`harness:pack-geometry` (POW-02 partial: the geometry rule on 19 conformance vectors in C on build-7 and in Swift on the
mini, and the two checked-in packs). Batch tools/ci/batches/f10-worker-20261008.json.
## I06 (master review, gate INT-10) and I05, landed with the rows
One normative geometry rule: packfile.h pf_load (CUDA and the OpenCL generic worker) gained the 64-bit count read with
the 2^32 refusal and the floor-log2 rule; proto-metal/main.swift gained `packGeometry` (the same rules, the same refusal
words) and sizes the day's dataset from the exact word count, refusing before any Metal work; `--geometry-check
<vectors> [pack dir...]` runs the vectors. proto-cuda/nvrtc/emu/geometry-vectors.txt is the shared file (ds55, ds85,
ds115, the power-of-two sizes, wrong count or log2, overflowing counts, missing leaves, leaves on a v2 pack), read by
emu/packfile-test.c on build-7 (19 vectors, 0 failures) and by the Metal host on the mini (19 vectors, 2 packs, 0
failures; geometry-check-mini.log). I05: every worker's found line carries a fifth field, the epoch the hash was computed
under (16 hex): CUDA and OpenCL from the pair, Metal from the job (metal-serve-smoke.log on the mini: the v4 pack
through the Swift path, the class v5 pack through the exact-geometry pack path). The miner reads a found line with four or
more fields (igneum/miner/src/main.rs:778).

112
docs/design/class-v6-rotating-family.md
View file

File diff suppressed because one or more lines are too long

4
docs/design/proving-payment.md
View file

@ -19,6 +19,10 @@ Why the 10 percent burn on the proving payment: the burn of the proving base fee
In a capacity-limited world (1 percent of the cards proving at the measured 5.5 percent proving efficiency) a lasting 1,000x proving demand spike is not restored in 150 periods when the internal pool is a fixed sum (the provers go to the external fee market and the internal backlog grows without bound) and is restored in 35 periods with this congestion-priced internal proving fee (peak 2x). The fixed pool is a subsidy, not a price; internal proving needs the user-funded, congestion-priced fee, which this decision gives it (`docs/analysis/class-v6/operator-simulation.md`).
## What the chain pays today, as served (the coordinator, 17:4x UK: no number on the page the chain does not pay)
The 2.0 devnet's own read-back on block one: the emission 3.168808781 IGN per block, split 2.5350470248 to the miner and 0.6337617562 to the proving pool (80 / 20, exact). The served fee and emission table states that split as the chain pays it today, the proving pool as a fixed share of emission, in the code; beside it this document's finding that a fixed pool is a subsidy, not a price, and that the congestion-priced user-funded proving payment is the design behind `proving_payment_activation_daa` (designed, in the code behind the constant, the guest's mirror owed before any height; not paid today). The tip-share prose agrees with the table: the tip 80 percent to the block's miner and 20 percent to the developer registrations, as the code pays it, nothing to the provers.
## What a prover earns, in one line
Per block: its shards' part of 20 percent of the block subsidy (2.5, 5.3) plus its shards' part of 90 percent of the block's proving payment (`pgas used x f_p`); per job: 90 percent of the external job fee (5.4). Nothing from the tip.

9
docs/evidence.md
View file

File diff suppressed because one or more lines are too long

6
docs/fud-ledger-2.0.md
View file

@ -241,7 +241,7 @@ Pin: Architecture and product, fifth box; Leadership tests, third box.
Status: Decided (8 October 2026): the boundary is stated wherever the proof architecture is explained.
Answer: Proven execution is not finality. A proof shows a block's execution was computed correctly; finality is a property of the chain's own consensus and checkpoints.
Answer: Proven execution is not automatically finality. A proof shows a block's execution was computed correctly; finality is a property of the chain's own consensus and checkpoints.
Evidence: `docs/plans/igneum-2.0.md` (the objective: the three boundaries).
@ -263,7 +263,7 @@ Pin: The objective, the three boundaries; Architecture and product, sixth box.
Status: Decided (8 October 2026): the boundary is stated wherever the proof architecture is explained.
Answer: ZK is not privacy. The proofs here verify execution; transaction data is public, and privacy would need additional application or protocol design.
Answer: ZK technology does not automatically make transactions private. The proofs here verify execution; transaction data is public, and privacy would need additional application or protocol design.
Evidence: `docs/plans/igneum-2.0.md` (the objective: the three boundaries).
@ -298,7 +298,7 @@ Pin: Leadership tests, first box (Ergo).
Status: Conceded (8 October 2026): no "number one" claim before comparative results and adoption exist.
Answer: The ceiling of the served ranking language is serious contention for the top of the GPU-mining space on engineering and operator proposition. The tests that would earn more are miners staying through hard conditions, customers repeatedly paying for proofs, and the network running without the founding team (D5); none is met yet.
Answer: No present-tense rank is served. Published claims name the evaluated release, evidence and boundary conditions. An independently substantiated pass of the acceptance standard would support a serious case that Igneum is in contention for leadership among GPU-first networks and would not award a numerical rank; it would not guarantee adoption, perpetual GPU profitability, defeat of all future chips or a numerical number-one ranking. The tests that would earn it are miners staying through hard conditions, customers repeatedly paying for proofs, and the network running without the founding team (D5); none is met yet.
Evidence: `docs/plans/igneum-2.0.md` (Leadership tests, second to fifth boxes).

15
docs/plans/counter-asic-3-status.md
View file

File diff suppressed because one or more lines are too long

61
docs/plans/igneum-2.0-test-harness-map.md
View file

@ -260,12 +260,53 @@ Rule: a case maps to a cell only where the cell's tests visibly answer it; cover
### suite:pool
- Command: `tools/build-remote.sh --box 2 -- test --release -p igneum-pool (from pool/, the pool-2.0 branch against the fork's successor-2.0.1)`
- Box class: suite (bounded; box 2)
- Command: `tools/build-remote.sh --box 6 -- test --release -p igneum-pool (from pool/, the pool-review-b branch off pool-2.0 against the fork's successor-2.0.1 515ba674; cargo check against the 2.0.1 manifest's node 7cfa422a)`
- Box class: suite (bounded; box 6)
- Fixtures: F0, F1
- Cases:
- UX-05 Keep voting keys with the miner through pooling: partial: the vote-key commitment on the wire (verify::tests::a_share_under_another_key_is_refused_before_the_hash: known-pass the member's key, known-fail another key on the share and a job naming another key, refused with code vote_key before the hash), the open pool's sidechain::check_key (the claimed key, the header's key and the reveal are one key), the same-nonce-other-template wrong_hash test, the TLS binding test; the malicious-pool run and the leave-and-retain run on the devnet-4 pair are the UX-05 batch's other evidence
- UX-04 Pay small operators without hidden custody: partial: PPLNS distribution tests (fee first, never overpays, late joiner), the payout key file round trip, the signed transfer decode; the live payout path on the devnet-4 pair is the UX-04 batch's evidence
- UX-05 Keep voting keys with the miner through pooling: partial: the vote-key commitment on the wire (verify::tests::a_share_under_another_key_is_refused_before_the_hash: known-pass the member's key, known-fail another key on the share and a job naming another key, refused with code vote_key before the hash), the open pool's sidechain::check_key (the claimed key, the header's key and the reveal are one key), the same-nonce-other-template wrong_hash test, the TLS binding test; the malicious-pool run and the leave-and-retain run on the devnet-4 pair are the UX-05 batch's other evidence; review B INT-15 (8 October 2026): server::authorise_tests::a_replayed_proof_of_possession_is_not_an_authorisation (a public PoP replayed into another session, payout, pool or chain, or under another key, refused; only the challenge-bound binding v2 admits on a public network) and the_binding_policy_follows_the_network
- UX-04 Pay small operators without hidden custody: partial: PPLNS distribution tests (fee first, never overpays, late joiner), the payout key file round trip, the signed transfer decode; the live payout path on the devnet-4 pair is the UX-04 batch's evidence; review B F12 and INT-03/INT-04 (8 October 2026): payout::intent_tests (a restart from the pre-broadcast snapshot, a lost send response, a crash on either side of the broadcast, a stuck transaction replaced under the same intent, a reorged receipt and a finality pause, a failed transaction, a ledger that cannot be written: never a duplicate or a lost obligation, finalised only under the chain's final lock) and state::ledger_tests (an unreadable ledger is refused or restored, never emptied); F13/INT-16: frame, admission and pool::share_tests (the frame bound while reading, the bounded outgoing queue, one membership per session, the nonce and in-flight bounds, the share and connection budgets)
### harness:economics-model
- Command: `python3 refpop.py && python3 coexist2.py && python3 hybrid.py && python3 market.py && python3 surface.py && python3 opsim.py 0.5 (from sim/economy/coexist/; no packages; a box or any Python 3.10)`
- Box class: harness (research; build-4 today; any box)
- Fixtures: F0
- Cases:
- ECO-02 Separate existing-owner and new-entrant viability: partial: the existing-owner and new-entrant tests per class at three tariffs, MSRP and street, with resale (reference-population.md); resale at zero, failures and financing not yet stressed
- ECO-03 Let the specialist keep its sunk development: partial: the sunk-development case first with low and high ranges, private mining, hardware sales and the hybrid, GPU re-entry after the fleet (coexistence-model.md sections 2, 8a, 12 to 12c; the surface in floor/sram-and-floor.md 4.4); the source thresholds retired to coexistence-workbook.md
- ECO-04 Model entry, exit and difficulty response: partial: a per-class supply curve with the installed base as a cap and re-entry (coexistence-model.md), the market-structure axis (8a), the operator simulation (operator-simulation.md); dynamic agent policies beyond the single-rule reaction are the simulation second cut
- ECO-06 Fund security and proving as issuance falls: partial: the recipients and burns reconciled on the chain block-one split (docs/design/proving-payment.md), proving demand as an income axis at none, launch and spike (coexistence-model.md); the full-horizon issuance ledger owed
- ECO-07 Price memory growth and honest-card displacement: partial: each dataset size scored by adversary burden against commodity burden on measured card rows and modelled chip tickets (docs/design/class-v6-rotating-family.md 10.0u); the 8.5 and 11.5 GiB knee costs expected, not measured
- INC-02 Keep revenue streams and claims separate: partial: the income streams kept separate in the model (mining subsidy, internal proving, external jobs; burns never counted as payment); the node-record and Ember reconciliation are the node and app lanes
- INC-03 Let a modified client choose the most profitable task: partial: the profit-maximising operator simulation first run (mine, internal prove, external prove, off); per-card agents and switching costs are the second cut
- INC-04 Survive external-demand spikes and token declines: partial: the four shocks in the operator simulation (the demand spike, the price fall, the largest provers leaving, the entrants), restored or not and in how many periods
- INC-08 Quantify provider and supplier failure concentration: partial: the supplier and operator dependence table (coexistence-model.md 8a); the signing, proving and aggregation roles are the node lane
### harness:cuda-select-check
- Command: `proto-cuda/nvrtc/emu/test.sh on a box (the CUDA worker under the CPU emulation backend): emu/select-check.sh (--readback select against the full read, known-failed first) and emu/serve-check.sh (the hash sample against igneum-pow hash-bound, both sides of the 32-bit boundary, the swap)`
- Box class: suite (CPU emulation, any build box)
- Fixtures: F0, F1
- Cases:
- ROT-01 Agree across every hourly boundary: partial: the worker's high-32 nonce rollover inside a job (a job from 2^32 - 32) finds the same set on both read-back paths and the boundary nonces hash as igneum-pow; the hourly program boundary itself is the fast-time harness
- POW-01 Match independent execution across every backend: partial: the CUDA select pass equals the full read line for line (zero hits, partial, the all-hit overflow fallback, a 96-nonce tail with a 32-lane launch, queued jobs, hits then zero, the epoch tag) and 17 sampled hashes equal igneum-pow; the million-vector campaign is harness:p01-vectors
### harness:pack-geometry
- Command: `proto-cuda/nvrtc/emu/packfile-test.sh on a box (packfile.h pf_load over emu/geometry-vectors.txt and the checked-in packs) and igneum-bench --geometry-check proto-cuda/nvrtc/emu/geometry-vectors.txt <pack dirs> on the Mac mini (proto-metal/main.swift packGeometry over the same file)`
- Box class: suite (a build box for C; the mini for Swift)
- Fixtures: F0
- Cases:
- POW-02 Validate generated programs and index folding: partial: the dataset geometry rule (exact word count, items, multiply-shift, 64-bit bytes, the floor-log2 and 2^32 refusals, missing leaves) agrees across the CUDA and OpenCL loader and the Metal host on 19 conformance vectors and the two checked-in packs (master review I06, gate INT-10); the rest of POW-02 is suite:pow
### site:manifest
- Command: `the site build's manifest regeneration from the dl index and the node's start line, read back at /release.json (site/release-manifest.json)`
- Box class: none
- Fixtures: F0
- Cases:
- OPS-03 Separate update distribution from consensus authority: partial: the manifest's versions and network blocks regenerate from their sources; the separation itself is the node suites' (suite:exec, suite:p2p-flows)
### experiment:connected-state
@ -307,11 +348,6 @@ Rule: a case maps to a cell only where the cell's tests visibly answer it; cover
- ADV-07 Evaluate lifetime without forced obsolescence: lifetime rows are the adversary lanes' models
- ROT-03 Test miner-voted bring-forward governance: miner-voted bring-forward needs a vote harness on the fault network F4
- ROT-04 Resist seed selection and faster evaluators: seed-selection resistance is the census harness (the class v6 invention lane), not yet in the matrix
- ECO-02 Separate existing-owner and new-entrant viability: F7
- ECO-03 Let the specialist keep its sunk development: F7
- ECO-04 Model entry, exit and difficulty response: F7
- ECO-06 Fund security and proving as issuance falls: F7
- ECO-07 Price memory growth and honest-card displacement: F7
- EVM-01 Match the selected EVM semantics: no EVM conformance-vector harness is mapped tonight; the exec suite does not run the reference test vectors
- EVM-07 Handle execution denial-of-service workloads: execution DoS workloads need the workload catalogue F3
- CAP-01 Reproduce the historical consumer-shard result: the consumer-shard reproduction is the fleet lane's pods
@ -322,18 +358,13 @@ Rule: a case maps to a cell only where the cell's tests visibly answer it; cover
- CAP-06 Calibrate assignment windows to paid completion: assignment windows are the proving fleet's
- CAP-07 Reassign work when inputs or providers disappear: reassignment is the proving fleet's
- CAP-08 Deliver customer-verifiable output at scale: customer-verifiable output is the reference apps plus the fleet
- INC-02 Keep revenue streams and claims separate: revenue separation is the economics lane's model
- INC-03 Let a modified client choose the most profitable task: the modified-client task choice needs an adversarial client harness
- INC-04 Survive external-demand spikes and token declines: demand spikes are the economic model
- INC-05 Contain job reservation and identity-splitting abuse: reservation abuse needs the capacity harness
- INC-06 Test difficulty and timestamp manipulation: difficulty and timestamp manipulation needs the fault network F4
- INC-07 Resist self-dealing fees and fake proving demand: self-dealing fees need the economic model and a live window
- INC-08 Quantify provider and supplier failure concentration: failure concentration is a live-window measurement
- FIN-06 Analyse old-key compromise and long-range histories: old-key compromise analysis is the finality lane's long-range scenario, not yet in the sim
- VER-07 Protect wallet keys, signing and recovery: wallet key protection is the wallet lane's security row
- OPS-01 Run the complete no-founder exercise: the no-founder exercise is an operations run, not a suite
- OPS-02 Diversify bootstrap and resist peer isolation: bootstrap diversity needs the fault network F4
- OPS-03 Separate update distribution from consensus authority: update distribution separation is the update-return lane's row
- OPS-05 Isolate untrusted proving workloads: proving workload isolation is the fleet's pod row
- OPS-07 Detect failures with usable evidence and runbooks: runbook detection is an operations run
- OPS-08 Repeat independent operation across releases: repeat independent operation is a cross-release observation
@ -460,4 +491,4 @@ Rule: a case maps to a cell only where the cell's tests visibly answer it; cover
## Count
171 automated cases: 59 mapped to a cell, 162 NOT RUN with a reason.
171 automated cases: 69 mapped to a cell, 152 NOT RUN with a reason.

1506
docs/plans/igneum-2.0-test-registry.json
View file

File diff suppressed because it is too large Load diff

16
docs/plans/pool.md
View file

@ -238,3 +238,19 @@ the fork worktree it points at (`vendor/igneum-node-pr`) as a whole repository,
the box by hand (`ln -s igneum-node-pr /srv/builds/<worktree>/vendor/igneum-node`), the one step the scripts do not do.
**Rule (7 October 2026, the Devnet 3 pair):** a pool daemon is built from the same repo tree as the chain's node, never a release behind; the 0.3.21-tree daemon hashed class v4 sub-version 2 against Devnet 3's re-pinned sub-version 3 and refused every share as WRONG HASH. The Devnet 3 pair runs the 0.3.22-tree daemon (c15b39b0, fork 69d1b56e).
## 11. 8 October 2026: review B F12, F13, INT-04 and INT-15 closed by construction (branch pool-review-b, landed from pool-review-b-land)
The second external review (docs/analysis/review-2026-10-08-b) and the full-system review's I08 read the pool crate statically. What was true on 8 October morning and what the crate does now:
| Finding | Was | Now | The regression that pins it (pool crate, release suite on build-6) |
|---|---|---|---|
| F12 payouts: broadcast before a durable debit | `send`, then the in-memory debit, then a periodic snapshot; a crash in between paid again from the stale balance with the next nonce; a receipt was `confirmed` the moment it appeared | `state::PaymentIntent`: the signed transaction and its hash on disk with the balance reserved in one fsync'd write BEFORE the broadcast; states prepared, broadcast, mined, finalised, failed, with reorged and replaced as events; every round and every restart reconciles by the intent's hashes (the same bytes again, a fee replacement under the same nonce after `replace_after_s`, a spent nonce re-signed under the same intent); finalised only when `igneum_getTransactionStatus` says locked and `igneum_getFinalityCheckpoints` says finality active with `latestLockKind` final (the node lane's F04 field; recovery or a pause keeps the obligation pending); a receipt that goes away is a reorg back to broadcast | `payout::intent_tests`: a_restart_from_the_pre_broadcast_snapshot_never_pays_twice, a_lost_send_response_is_reconciled_by_the_intent_and_never_paid_again, a_stuck_transaction_is_replaced_under_the_same_intent_not_paid_again, a_reorged_receipt_and_a_finality_pause_leave_the_obligation_pending, a_crash_on_either_side_of_the_broadcast_pays_once, a_failed_transaction_re_credits_once (a mock eth_ and igneum_ JSON-RPC on 127.0.0.1, `mock_rpc.rs`: the real client, signing and 2718 encoding; nothing leaves the process) |
| F13 admission: `MAX_LINE` checked after the line, an unbounded writer queue, the nonce in `seen` before its check, a repeated `authorize` adding members and the disconnect removing the last | as the review read it | `frame::read_frame` refuses a frame the moment it crosses the bound (memory: the bound plus one 64 KiB buffer; time: the read budget); `admission::Outgoing` is bounded (`--out-queue`) and a full queue disconnects the slow reader; `admission::Members` is keyed by session (a re-authorise replaces atomically, the disconnect removes that session's member only, `--max-members`); a nonce is in flight while checked and remembered only when accepted, `seen` bounded per job; the vote key and the claimed hash against the share target are prefilters before the verifier; a share budget per member (`--shares-per-s`, `--share-burst`), an in-flight bound (`--in-flight-checks`), connection budgets (`--max-connections`, `--max-connections-per-ip`), an authorise deadline (`--authorise-within-s`). The member's vote key stays committed in the template header (node.rs `issue_job`), unchanged | `frame::tests::an_oversized_unterminated_frame_is_refused_within_the_memory_and_time_budget`, `admission::tests::repeated_authorisation_on_one_session_leaves_one_member_and_the_disconnect_removes_it`, `the_member_bound_holds_and_a_replacement_does_not_count_twice`, `a_slow_reader_cannot_grow_the_outgoing_queue`, `the_share_budget_is_a_bucket`, `connections_are_bounded_in_total_and_per_address`, `pool::share_tests::an_invalid_share_flood_leaves_no_nonce_state`, `a_share_flood_over_the_budget_is_refused_before_the_check` |
| INT-04 the ledger: an unreadable `state.json` started the pool empty | "unreadable; starting empty" | every save keeps the previous snapshot and a SHA-256 sidecar; a present ledger that cannot be read or fails its digest restores the previous snapshot or refuses to start (exit 4, the restore line; `--start-empty` only knowingly); a save that fails broadcasts nothing | `state::ledger_tests::an_unreadable_ledger_is_refused_or_restored_never_emptied`, `sent_payments_of_the_earlier_loop_become_intents`, `payout::intent_tests::a_ledger_that_cannot_be_written_broadcasts_nothing` |
| INT-15 (I08): a public proof of possession was the authorisation over plain TCP; over TLS the exporter bound the session but not the payout | as read | the challenge-bound binding v2 (spec 9.3): a signature over this session's challenge, the exporter, the payout and the pool address; replay into another session, payout, pool, chain or under another key is refused (`server::authorise_check`); the public networks take v2 only, devnet admits v1 and the unbound authorise until the 2.0.2 miner carries v2 (the node lane, by 23:30 on 8 October) | `server::authorise_tests::a_replayed_proof_of_possession_is_not_an_authorisation`, `the_binding_policy_follows_the_network` |
| F03 seam: `EpochSeeds` without `shadow_reps`, `KeyReveal` without `sig_scheme` on master's copy | master's pool predated the pool-2.0 line | the pool-2.0 tree landed on master; `cargo check` green against the 2.0.1 manifest's node sha 7cfa422a on build-6, the suite green against successor-2.0.1 515ba674 | the rule 24 crate gate at every landing |
Known-failed first: commit 75383929 on pool-review-b carried the tests over the 8 October semantics, and the suite read seven red on build-6 (the three payout tests paid 200 of 100 due or never settled, the frame reader blocked on a newline that never came, the registry kept two members, the queue took 10,000 lines, the nonce set held 100,000 refused nonces); 065148a2 and 4a8dcbc0 turned them green, 51 of 51.
Consequences per tier. A pool member on any card: a payout cannot be paid twice or lost across a pool crash, a lost node answer, a reorg or a stuck transaction; a payment shows `mined` until the chain's own finality covers it, so the row a member reads as final is final. A pool operator: the ledger cannot start empty over a corrupt file, the previous snapshot and its digest are always there to restore from, and one bad member cannot hold the daemon's memory or its verifier (every bound has a flag and a default). The network: a captured `authorize` cannot be replayed to redirect a member's pay or to vote under its key from another session, and the vote key stays the member's in every template (the pin "vote keys stay with the miner", docs/plans/igneum-2.0.md).

2
docs/plans/release-rules.md
View file

@ -19,3 +19,5 @@ Every cut of the Igneum Miner app and its node runs under these. The dated plan
12. **The Discord card only when every platform is live.** Live manifest changes beyond the binaries (a moved consensus floor) go out only on the founder's explicit word, staged beside the release with their digest and a one-line diff.
13. **Ship on green:** no calendar waits; when the gates are green, publish and state the clock time (UK). Checkpoints are for slips, not for waiting.
15. **The version bump is the release branch's first commit (7 October 2026, after the 0.3.23 miss).** When a release-0.3.N branch opens, its first commit moves the six version places (app/igneum-app/Cargo.toml and Cargo.lock, app/windows/version.h, app/igneum-app/resources/igneum-app.rc's four fields, packaging/mac/app/Info.plist, the installer's AppVersion), never left to the cut: release-0.3.23 opened at 4cdcab31 and carried 0.3.22 in every place until 21:26 BST, so the app exes and the window host crossed from its first closed tip read 0.3.22 and were void. Gate check: on a push to release-0.3.N the pre-push gate (tools/ci/release-version-check.sh, self-test on tonight's shape first) reads all six places against the branch name and goes red on any mismatch; the .rc was the second layer of the same miss (the box cross failed in build.rs at 21:30 BST).
16. **The public miner carries no remote execution; the fleet runs the lab build (the founder's ruling on review B F14, 8 October 2026, 19:57 UK).** From 2.0.2 the public build has no run-script, fetch, collect, restart or update-now jobs, no jobs feed, no wake listener and no relay client (cargo feature `lab`, off by default), and documents its least-privilege boundary; "automatic updates off" stays off for an urgent manifest (pause and notify only). Our fleet, PC 1, PC 2 and the mini run the lab build only (product "Igneum Miner Lab", channel `igneum-2.0-devnet-lab`; the kits Igneum-Miner-Lab-Setup-<v>.exe, the lab DMG, the lab hive; the per-PC combined exes wrap the lab Setup), which verifies its manifest and jobs feed against the lab root (~/.config/igneum/lab-signing on the Mac, never in the repo), separate from the OTA key; publish-manifest.sh and publish-jobs.sh sign the lab channel with the lab key and a public channel with the public key, never cross (`--lab` / `--channel lab`). The relay agent refuses to run anything beside a public engine, and the jobs publisher refuses a target whose last upload is not a lab build.

40
docs/spec/09-pool-protocol.md
View file

@ -1,6 +1,6 @@
# Igneum protocol specification, section 9: the pool protocol
Spec version 0.1, 3 October 2026. Status of this section: Designed. Nothing here is implemented. The only miner-to-worker protocol that exists is the devnet worker protocol (`igneum-miner --worker`, `vendor/igneum-node/igneum/miner/src/main.rs`; `--serve` in `proto-cuda/host.cu`, `proto-opencl/host.c`, `proto-metal/main.swift`), which runs between two processes on one machine and is cited here where this section reuses it. The review of 3 October 2026 found that no pool protocol existed and that "shares on a 64-bit lane hash with vardiff are unspecified" (`docs/review/round-3-2026-10-03.md`, farm operator, attack 3, R3.15). This section is the answer.
Spec version 0.2, 7 October 2026 (0.1 of 3 October 2026). Status of this section: Implemented in part. The reference pool (`pool/`, docs/plans/pool.md) implements 9.3 (TLS 1.3 and the binding, 7 October 2026), 9.4 mode A, 9.5, 9.6, 9.8 and 9.9; 9.7 (vote relay) and modes B and C are designed only; 9.12 (the open pool, the share sidechain with no operator) is implemented and gated on a fast-time network (pool.md section 10). The original note stands for what is still only designed. The only miner-to-worker protocol that exists is the devnet worker protocol (`igneum-miner --worker`, `vendor/igneum-node/igneum/miner/src/main.rs`; `--serve` in `proto-cuda/host.cu`, `proto-opencl/host.c`, `proto-metal/main.swift`), which runs between two processes on one machine and is cited here where this section reuses it. The review of 3 October 2026 found that no pool protocol existed and that "shares on a 64-bit lane hash with vardiff are unspecified" (`docs/review/round-3-2026-10-03.md`, farm operator, attack 3, R3.15). This section is the answer.
Lineage, credited. The shape is Stratum V2's (Braiins and the Stratum V2 working group, stratumprotocol.org), from memory and approximate: a binary, encrypted protocol with a Mining Protocol for shares, a Job Declaration Protocol under which a miner builds its own block template from its own node and declares it to the pool, which may accept it and must then pay shares found on it, and a Template Distribution Protocol between the miner and its node. Igneum keeps the three ideas (encrypted transport, miner-built templates under the pool's payout, pool pays by shares) and changes what the header carries: on Igneum the header names a vote key (section 2.4) and the vote key decides finality (section 3), so this section also fixes whose key a pooled block names and who signs the checkpoint votes. The ledger entries F10 and G6 (pools hold the votes; Stratum V2 job declaration is optional) are the two criticisms this section is written against.
@ -41,12 +41,14 @@ Why. The pool protocol crosses operators and languages: pools, mining operating
| Rule | Value | Label |
|---|---|---|
| Transport | TLS 1.3, server-authenticated; the pool's certificate chain or a pinned public key the member obtained with the pool's address | Designed |
| Framing | One JSON object per line, `\n` terminated, no object over 4 MiB (a template with its transactions) | Designed |
| Framing | One JSON object per line, `\n` terminated, no object over 4 MiB (a template with its transactions). The bound is enforced while the bytes arrive (review B F13, 8 October 2026): a frame is refused with `bye` the moment it crosses 4 MiB, never collected first; the pool also bounds its outgoing queue per connection (a reader that does not drain is disconnected), its connections in total and per address, its members, each member's shares a second and checks in flight, and the time from connect to `authorize` | Implemented (8 October 2026, `pool/src/frame.rs`, `admission.rs`) |
| Field order | Irrelevant; unknown fields MUST be ignored | Designed |
| Integers | JSON numbers up to 2^53; anything that can exceed it (nonces, targets, hashes, weights in base units) is a hex string | Designed |
| Identifiers | `id` on every request, echoed on the reply; notifications carry no `id` | Designed |
| Port | 4461 (the mainnet chain id, section 7.4), 4462 testnet, 4463 devnet; a pool MAY use another and say so in its address | Designed |
| Member authentication | The `authorize` message carries the member's BLS public key and a proof of possession (the `KeyReveal` of section 3.10, 144 bytes), plus a signature over the session's TLS exporter value under the same key, so the key proves it is live and the connection is bound to it | Designed |
| Member authentication | The `authorize` message carries the member's BLS public key and a proof of possession (the `KeyReveal` of section 3.10, 144 bytes), plus `binding`: a signature over the session's TLS exporter value under the same key, so the key proves it is live and the connection is bound to it | Implemented (7 October 2026, O-9.7 closed below) |
| The binding's bytes (O-9.7) | Exporter: TLS 1.3 `export_keying_material` with label `EXPORTER-igneum-pool-binding`, context = the chain id as 8 little-endian bytes, 32 bytes out, on both sides. Message: `igneum-pool-binding-v1/` \|\| chain id (8 bytes LE) \|\| exporter. Signature: BLS12-381 G2 under the tag `IGNEUM_POOL_BINDING_V1_BLS12381G2_XMD:SHA-256_SSWU_RO_NUL_` (never the vote or the PoP tag; a binding is not a vote). The pool refuses an `authorize` whose binding does not verify for THIS connection's exporter with `bye`; a replay on another connection carries another exporter and fails. Over plain TCP the field is empty and ignored (devnet only; testnet and mainnet pools refuse plain without `--allow-plain`) **Binding v2 (review B INT-15, 8 October 2026):** `welcome` carries `challenge` (32 bytes hex, fresh per connection) and `binding` (`v2`, `v1` or `none`, the pool's policy); the member signs, with `sign_binding` and the same tag, `session_binding_v2` = SHA-256 of `igneum-pool-session-v2/` \|\| chain id (8 bytes LE) \|\| challenge (32) \|\| the TLS exporter (32; all zero over plain TCP) \|\| the payout address (20) \|\| the pool's coinbase address (20, `welcome.pool_address`), so a public proof of possession replayed into another session, another payout, another pool or another chain is refused; the challenge is spent by the authorise it admits and `authorized` carries the next. Testnet and mainnet pools admit v2 only; a devnet pool admits the exporter-only v1 binding over TLS and, over plain TCP, an unbound authorise (both logged) unless `--require-binding-v2` | Implemented (`consensus/core/src/finality.rs binding_message, sign_binding, verify_binding; pool `tls.rs`; miner `pool.rs`) |
| Server identity | A chain from a root the member's system trusts (`--pool-tls`), or a pinned certificate (`--pool-pin <hex>`: BLAKE2b of `igneum-pool-cert-pin-v1` \|\| the DER certificate, the pin a self-signed pool prints at start and shows on its page) | Implemented |
## 9.4 Templates
@ -87,16 +89,16 @@ Direction P = pool, M = member. Every message is one object with `"t"` naming th
| Type | Direction | Fields | Meaning |
|---|---|---|---|
| `hello` | M to P | `versions` (list), `chain_id`, `client`, `modes` supported | Opens the session |
| `welcome` | P to M | `version`, `chain_id`, `pool_address`, `modes`, `vote_mode` (`member` or `pool`), `declaration` {`tag`, `max_extra_bytes`, `max_declarations_per_s`, `declare_timeout_ms`}, `share_scheme` (`pplns` or `pps` with its parameters), `min_shift`, `max_shift` | The pool's terms; the member shows `vote_mode` and `share_scheme` to its operator |
| `authorize` | M to P | `pubkey` (48 bytes hex), `pop` (96 bytes hex), `binding` (signature over the TLS exporter), `label` | Names the member by its vote key |
| `authorized` | P to M | `member_id`, `revealed` (true when the key has been revealed on chain, per the pool's node) | |
| `seeds` | P to M | `epoch_seed`, `day_seed`, `seed_source`, `vdf_proof` (516 bytes hex, section 4.4, when the member asks), `next_epoch_seed`, `next_day_seed`, `next_at_daa` | Current and next seed pair so the member can `prepare` its workers ahead (the worker protocol's `prepare`) |
| `welcome` | P to M | `version`, `chain_id`, `pool_address`, `modes`, `vote_mode` (`member` or `pool`), `declaration` {`tag`, `max_extra_bytes`, `max_declarations_per_s`, `declare_timeout_ms`}, `challenge` (32 bytes hex, this session's; 9.3 binding v2), `binding` (`v2`, `v1` or `none`: what the pool admits), `share_scheme` (`pplns`, `pps` or `pplns-sidechain` with its parameters, and `software_dev_fee_percent`: the software fee a member pays in this mode, published beside the pool fee), `min_shift`, `max_shift`, `pool_mode` (`operator` or `open`, 9.12), `tls` | The pool's terms; the member shows `vote_mode` and `share_scheme` to its operator |
| `authorize` | M to P | `pubkey` (48 bytes hex), `pop` (96 bytes hex), `binding` (96 bytes hex: binding v2 over this session's challenge, exporter, payout and pool, 9.3; the exporter-only v1 on devnet), `label`, `payout` (the v0 EVM payout address) | Names the member by its vote key; a public PoP alone is not an authorisation (review B INT-15) |
| `authorized` | P to M | `member_id`, `revealed` (true when the key has been revealed on chain, per the pool's node), `key_hash`, `challenge` (the next 32-byte challenge of this session, for a re-authorise) | |
| `seeds` | P to M | `epoch_seed`, `day_seed`, `seed_source`, `vdf_proof` (516 bytes hex, section 4.4, when the member asks), `next_epoch_seed`, `next_day_seed`, `next_at_daa`, `program_class`, `next_program_class`, `era_seed`, `shadow_reps`, `next_shadow_reps` (the program class, era seed and latency-ladder rung of the current and the next epoch: a member refuses a job whose program it cannot name) | Current and next seed pair so the member can `prepare` its workers ahead (the worker protocol's `prepare`) |
| `set_target` | P to M | `shift` s (9.8) | Vardiff |
| `template` | P to M | `template_id`, `mode` (A or B), `header` (every field but nonce, hex), `coinbase`, `transactions` (A) or `merkle_path` plus `tx_count` (B), `votes_carried` (indices of the member's votes in this template) | A new template; supersedes the previous one for new jobs |
| `declare_template` | M to P | `header`, `coinbase`, `transactions` | Mode C |
| `template_ack` | P to M | `template_id` | |
| `template_refused` | P to M | `code` in {`invalid`, `payout`, `tag`, `oversize`, `rate`, `capacity`, `other`}, `detail` | |
| `job` | P to M | `job_id`, `template_id`, `prehash` (64 hex), `vote_key_hash` (64 hex, the key the template's header names; Igneum 2.0), `target64` (16 hex), `share_target64` (16 hex), `nonce_start` (hex, a multiple of 32), `nonce_count` (hex, a multiple of 32), `clean` (bool) | Work. `clean` true means abandon earlier jobs. The member forwards it to its workers as a worker-protocol `job` line with `share_target64` in the target field |
| `job` | P to M | `job_id`, `template_id`, `prehash` (64 hex), `vote_key_hash` (64 hex, the key the template's header names; Igneum 2.0), `target64` (16 hex), `share_target64` (16 hex), `nonce_start` (hex, a multiple of 32), `nonce_count` (hex, a multiple of 32), `epoch_seed`, `day`, `program_class`, `era_seed`, `shadow_reps`, `clean` (bool) | Work. `clean` true means abandon earlier jobs. The member forwards it to its workers as a worker-protocol `job` line with `share_target64` in the target field; its CPU re-check hashes exactly the program the job names |
| `job_refused` | M to P | `job_id`, `code` in {`vote_key`, `payout`, `reveal`, `votes`, `header`, `merkle`, `invalid`, `seeds`}, `detail` | The member will not hash this job and says why |
| `share` | M to P | `job_id`, `nonce` (16 hex), `hash` (16 hex), `vote_key_hash` (64 hex, the member's key; Igneum 2.0) | A lane hash at or below `share_target64`. The pool MUST refuse a share whose `vote_key_hash` is not the session's authorised key or not the key of the job's template (`share_result` code `vote_key`), so a share is evidence on its own of whose key the work was done under |
| `share_result` | P to M | `job_id`, `nonce`, `accepted` (bool), `code` in {`ok`, `stale`, `duplicate`, `above_target`, `wrong_hash`, `unknown_job`, `vote_key`}, `weight` (the share's weight, 9.8) | |
@ -152,11 +154,27 @@ The program changes every epoch and the dataset every day (section 1.12), and an
2. The member MUST check `epoch_seed` and `next_epoch_seed` against its verifier (the VDF output for the epoch, section 4.3; on the devnet, the hash of the last selected-chain block below `3,600 e - 600`) and refuse jobs on a seed its verifier does not confirm (`job_refused`, code `seeds`). A pool that could choose the seed could choose the program, which is the grinding section 4 exists to prevent.
3. A member whose verifier is the light client of section 10 verifies the VDF proof (516 bytes, 4.5 ms, Measured, section 4.4) that the pool forwards in `seeds`, against the `seed_source` its checkpoint chain confirms.
## 9.12 The open pool: a share sidechain with no operator
Added 7 October 2026 (mission item 11; docs/plans/pool.md section 10). Lineage: Monero P2Pool (SChernykh, 2021: a 10-second share sidechain merge-mined with the main chain, PPLNS over the window, payouts as coinbase outputs, no operator; from its README, approximate), which this section rebuilds on three objects Igneum already had: the vote key in the header (9.6), the coinbase extra data (section 3.10's key reveal and the devnet's `IGNA` address), and an execution layer that pays by rule from consensus data (the 20 percent proving share). Nothing here changes block validity; the one consensus-visible rule is the executor's split (item 5).
1. A member of the open pool runs its own node and the open daemon beside it (`igneum-pool --open`). The daemon serves this section's member protocol to the member's own miners, builds every template from the member's node with the member's key in the header and the member's own payout address in the coinbase (`IGNA`), and holds no key and no balance.
2. A share is a template at the share chain's target: a lane hash at or below `share_target64` of the chain, with the coinbase extra data carrying, beside the reveal and `IGNA`, the share it extends (`IGNH` \|\| 64 hex: the parent's block hash, the zero hash for a genesis share) and the window's payout split (`IGNW` \|\| 4 hex count \|\| count x (40 hex address \|\| 8 hex weight), `kaspa_consensus_core::evm`). A share's identity is its block hash. A share whose lane hash is also at or below the block target is a block, submitted to the member's node as any block.
3. The chain. One parent per share; the heaviest cumulative work (the sum of `2^64 / target` over the ancestry) is the tip; a share off the tip's ancestry is stale and pays nothing; a share extending more than 256 heights below the tip is refused. Target: `chain_share_s` seconds per share (10 by default), held by the Kaspa DAA's shape over the last 30 shares (the window's mean target times its actual span over its expected span, header timestamps), clamped to a factor of 4 against the parent's target, never above 2^62 nor ten doublings above the chain's first target; the first target is a chain constant (`--chain-genesis-target64`, by default the network's genesis block target eight times easier; never a node's current template, which differs between members). Members gossip shares (`p2p.rs`: `hello`, `share`, `get_shares`, newline JSON); every share is checked by every member: the parent known, the height, the target the chain fixes after the parent, the block hash and the body's merkle root, the reveal's key in the header, `IGNA` the share's address, `IGNH` the parent, `IGNW` byte-equal to the split the chain computes for that parent and that address, the seeds either those the member's own node reported for the share's epoch or (another node's view of the same epoch, which a DAG allows while the chain below the lead settles) an epoch seed that is a block the member's node holds at that epoch's seed depth with the epoch's class, era and rung, and the PoW under those seeds. Nodes never see a share.
4. The split. For a share by address A at target T extending parent P: the last `window_shares - 1` shares ending at P plus the share itself, weighed by `2^64 / target`, summed per address; the software dev fee as one more entry at `open_dev_fee_percent` of the whole (the same 1 percent the solo miner pays, so solo, pool-0 and the open pool are fee-neutral); scaled to u32 weights of 2^32; descending by weight then ascending by address; at most 256 entries (the smallest dropped past it). Every member of a chain holds the same `window_shares`, `chain_share_s`, fee and dev address, or its shares mismatch on the split and are refused: these four are the chain's constants, named by its `--open-chain` string.
5. Payment (the consensus-visible rule, `igneum/exec/src/executor.rs`): from `pool_split_activation_daa` on (a params field, in the digest once set, never by default; the fast-time gate ran it at 0), a blue block whose coinbase carries a well-formed `IGNW` has its producer share paid by the split: each entry `floor(share x weight / sum)`, the rounding dust to the block's `IGNA` address; a block without a split, below the switch, or with a malformed split pays its `IGNA` address alone, as every block does today. The 20 percent proving share is untouched. So a block found on the share chain pays the window from its own coinbase, every node computes the same credits from the block alone, and nobody holds anything for anybody.
6. What a withheld share earns: nothing. A share kept from the gossip is in no other member's chain, so no block but the withholder's own names it in a split, and the withholder's own blocks pay the others' shares by the same rule. Dropping another member's shares is the same act and costs the dropper the same.
7. The drop proof. Every share a member finds is written to its own `shares.log` as one self-certifying line (the seeds, the header with its nonce, the coinbase): `igneum-pool verify-share <line> [--block <raw block>]` re-hashes it, and against a block says whether that block's split pays the share's address (exit 3, `BLOCK DROPS`, when it does not).
8. What is not here yet (pool.md section 10.6): uncles (a fork's loser is stale; the gate measures the stale rate), share bodies as transaction ids instead of the full template, a late joiner's verification of shares from epochs its node no longer reports, and the chain's share rate against the network's: at 1 block a second a 10-second share chain is sub-block work only for a pool under a tenth of the network, so the gate ran the chain at 0.1 s per share and a public open chain will set its rate from its hashrate (the mini and nano chains of Monero P2Pool are the same answer).
## 9.10 Parameters in this section
| Parameter | Value | Label |
|---|---|---|
| Transport | TLS 1.3, newline-delimited JSON | Designed |
| Transport | TLS 1.3, newline-delimited JSON | Implemented (7 October 2026); plain TCP allowed on the devnet only |
| Binding | `igneum-pool-binding-v1/` \|\| chain id LE \|\| 32-byte exporter (`EXPORTER-igneum-pool-binding`, context chain id LE), tag `IGNEUM_POOL_BINDING_V1_...` | Implemented |
| Open pool: `chain_share_s`, `window_shares`, `open_dev_fee_percent`, split entries, first target | 10 s, 2,160, 1 percent, at most 256, `target64(genesis bits) << 3` | Implemented (9.12); the public chain's rate is open, 9.12 item 8 |
| Open pool: `pool_split_activation_daa` | never by default; 0 on a new chain; in the digest once set | Implemented (params, executor) |
| Ports | 4461 / 4462 / 4463 | Designed |
| Required modes | A (full template) and C (declared); B optional | Designed |
| `declare_timeout_ms` | 500 | Designed, Open O-9.5 |
@ -184,6 +202,8 @@ The program changes every epoch and the dataset every day (section 1.12), and an
| O-9.4 | Mode A item 5 needs a node RPC that validates a block as a template without submitting it; none is named | Add the RPC to the fork (candidate: `validateBlockTemplate`), with the cost per call measured at the design document's gas budgets | 2 |
| O-9.5 | Every timing here is Designed: template size and bandwidth per member at 1 BPS, the member's check cost (9.4.1), `declare_timeout_ms`, `share_interval_s`, the pool's verification throughput on a server core | A reference pool with 100 members on the phase 4 devnet: record template bytes per second per member, check time per template on a 2019-class laptop core, declared-template acceptance latency, shares verified per second per core on a server CPU; set the four parameters from the distributions | 4 |
| O-9.6 | The carriage ratio of 9.7 item 6 has no threshold, and its 60-block window is a guess | On the same devnet, one pool conforming and one dropping every vote: record both ratios per key over a day; set the warning threshold at the point that separates them with no false warnings on the conforming pool | 4 |
| O-9.7 | The TLS exporter binding in `authorize` is named and not specified (which exporter label, which bytes are signed) | Write the exact bytes at the first implementation; test that a replayed `authorize` on a second connection is refused | 4 |
| O-9.7 | CLOSED 7 October 2026: the exact bytes are in 9.3 (label, context, message, tag); `pool/src/tls.rs` tests a replayed `authorize` on a second connection refused, and `finality.rs` tests the signature bound to one exporter and one chain id | Written at the first implementation, as asked | 4 |
| O-9.10 | The open pool's share rate against the network's (9.12 item 8): a 10-second chain is sub-block work only for a pool under a tenth of a 1-block-a-second network; the gate ran at 0.1 s per share | A public open chain names its rate from its hashrate (one chain per decade of hashrate, the Monero P2Pool answer) or carries uncles; decide at the first public open chain | 4 |
| O-9.11 | A late joiner of the open pool cannot verify shares of epochs its node no longer reports (`Open::seeds_for`), so its sync is bounded to those epochs | A node RPC that answers the seeds of any past epoch, or the epoch seeds carried in the share chain's own headers, before a chain older than the node's epoch window is public | 4 |
| O-9.8 | A member with only the light client of section 10 as its verifier checks headers against a checkpoint chain, not against a full node's tip, so item 4 of 9.4.1 (parents are known tips) is weaker for it: it can confirm the parents descend from the last certified checkpoint and no more | Decide at gate 4 whether a light-client member may vote (9.7 item 2 says yes) after the eclipse test of O-3.7 is re-run with light-client members in the model | 3, with 4 |
| O-9.9 | HiveOS and the rental markets need this protocol to list the algorithm (ledger entry on rental, "cannot list an algorithm whose kernel changes hourly without a stratum for it") | The reference member runs under HiveOS against the reference pool through 24 epoch changes with in-worker compilation (R3.15's acceptance test: outage under 2 s per change) | 4 |

25
packaging/ota/publish-jobs.sh
View file

@ -46,6 +46,12 @@ ROOT="$(cd "$HERE/../.." && pwd)"
export PATH="$HOME/.cargo/bin:$PATH"
KEY="$HOME/.config/igneum/ota-signing-key"
PUB="$HOME/.config/igneum/ota-signing-key.pub"
# F14 (8 October 2026): the jobs feed is a LAB thing (the public build has no jobrun); a jobs file is signed with the lab root
# when --channel lab is given (the 2.0.2 lab builds verify against it), with the OTA key for the 0.3.x and 2.0.0/2.0.1 engines
# that still carry the jobs feed; never both roots on one file, never the lab key on a public channel
LAB_KEY="$HOME/.config/igneum/lab-signing/lab-signing-key"
LAB_PUB="$HOME/.config/igneum/lab-signing/lab-signing-key.pub"
JOBS_CHANNEL="public"
TOKEN_FILE="$HOME/.config/igneum/dl-token"
SIGNER="$ROOT/app/igneum-app/target/release/igneum-ota-sign"
@ -96,6 +102,7 @@ while [ $# -gt 0 ]; do
--glob) GLOBS+=("$2"); shift 2 ;;
--command) COMMAND="$2"; shift 2 ;;
--what) WHAT="$2"; shift 2 ;;
--channel) JOBS_CHANNEL="$2"; shift 2 ;;
--allow-app-restart) ALLOW_APP_RESTART=1; shift ;;
--zip) ZIP="$2"; shift 2 ;;
--fixtures) FIXTURES="$2"; shift 2 ;;
@ -128,6 +135,24 @@ if [ "$CMD" = add ] && [ "${KIND:-}" = restart ] && [ "${WHAT:-app}" = app ] &&
fi
case "$CMD" in add|list|remove|sign|verify) ;; *) sed -n '2,35p' "$0" | sed 's/^# \{0,1\}//'; exit 2 ;; esac
# F14: the intake header of a 2.0.2+ engine reads `edition=lab|public` (the window lane's reviewb-202); a target whose last
# upload says edition=public is refused on every add (the public miner has no jobs feed; a job for it is a mistake or worse).
# Pre-2.0.2 headers carry no edition and pass. Reads tools/logs.mjs --rotation's header parse when the Mac has DATABASE_URL.
target_edition() { # <machine id8> -> lab | public | '' (unknown or no DATABASE_URL)
[ -f "$HOME/.config/igneum/env" ] || { echo ""; return; }
{ node "$ROOT/tools/logs.mjs" --header "$1" 2>/dev/null || true; } | sed -n 's/.*edition=\([a-z]*\).*/\1/p' | head -1 || true
}
if [ "$CMD" = add ] && [ -n "${TARGET:-}" ] && [ "$TARGET" != all ]; then
for t in ${TARGET//,/ }; do
ed="$(target_edition "$t")"
if [ "$ed" = public ]; then echo "refused: $t last uploaded as edition=public (the public miner carries no remote execution, F14); only a lab build takes jobs" >&2; exit 2; fi
done
fi
case "$JOBS_CHANNEL" in
lab) KEY="$LAB_KEY"; PUB="$LAB_PUB"; [ -f "$LAB_KEY" ] || { echo "no $LAB_KEY: the lab channel signs with the lab root only" >&2; exit 1; } ;;
public) ;;
*) echo "refused: --channel must be lab or public (got '$JOBS_CHANNEL'); a cross of root and channel never signs" >&2; exit 2 ;;
esac
[ -f "$KEY" ] || { echo "no $KEY (see packaging/ota/README.md, Keys)" >&2; exit 1; }
[ -f "$PUB" ] || { echo "no $PUB" >&2; exit 1; }
[ -f "$TOKEN_FILE" ] || { echo "no $TOKEN_FILE" >&2; exit 1; }

23
packaging/ota/publish-manifest.sh
View file

@ -42,6 +42,13 @@ ROOT="$(cd "$HERE/../.." && pwd)"
export PATH="$HOME/.cargo/bin:$PATH"
KEY="$HOME/.config/igneum/ota-signing-key"
PUB="$HOME/.config/igneum/ota-signing-key.pub"
# F14 (the founder's ruling, 8 October 2026, 19:57 UK): the lab build (product "Igneum Miner Lab", our fleet, PC 1, PC 2, the
# mini) verifies against its OWN root; the public build carries only the public root. The key follows the channel: the lab
# channel igneum-2.0-devnet-lab signs with ~/.config/igneum/lab-signing/lab-signing-key, every other channel with the OTA key,
# and a cross (--lab-key on a public channel, the OTA key on the lab channel) is refused before anything is written.
LAB_CHANNEL="igneum-2.0-devnet-lab"
LAB_KEY="$HOME/.config/igneum/lab-signing/lab-signing-key"
LAB_PUB="$HOME/.config/igneum/lab-signing/lab-signing-key.pub"
TOKEN_FILE="$HOME/.config/igneum/dl-token"
SIGNER="$ROOT/app/igneum-app/target/release/igneum-ota-sign"
PRODUCT="app"
@ -60,6 +67,8 @@ while [ $# -gt 0 ]; do
--override) OVERRIDE="$2"; shift 2 ;; # consensus.override: the exact JSON object every app writes to its override.json (all height switches, not just the new one)
--min-supported) MIN_SUPPORTED="$2"; shift 2 ;;
--channel) CHANNEL="$2"; shift 2 ;;
--no-nudge) NUDGE=0; shift ;;
--lab) CHANNEL="$LAB_CHANNEL"; shift ;;
--product) PRODUCT="$2"; shift 2 ;;
--tuning) TUNING_FILE="$2"; shift 2 ;;
--no-tuning) NO_TUNING=1; shift ;;
@ -115,6 +124,10 @@ MF="igneum-$PRODUCT-latest.json"
[ -n "$VERSION" ] || [ "$VERIFY_ONLY" = 1 ] || { echo "--version is required" >&2; exit 2; }
[ -n "$VERSION" ] || VERSION="(the folder's)"
[ "$VERIFY_ONLY" = 1 ] || case "$VERSION" in [0-9]*.[0-9]*.[0-9]*) ;; *) echo "--version must be major.minor.patch" >&2; exit 2 ;; esac
case "$CHANNEL" in
"$LAB_CHANNEL") KEY="$LAB_KEY"; PUB="$LAB_PUB"; [ -f "$LAB_KEY" ] || { echo "no $LAB_KEY: the lab channel signs with the lab root only (igneum-ota-sign keygen $LAB_KEY $LAB_PUB once, on this Mac)" >&2; exit 1; } ;;
*lab*) echo "refused: channel '$CHANNEL' names a lab but is not '$LAB_CHANNEL'; a public channel never signs with the lab root and a lab channel never with the OTA key" >&2; exit 2 ;;
esac
[ -f "$KEY" ] || { echo "no $KEY: run $SIGNER keygen $KEY $PUB once (the public key then goes into src/manifest.rs)" >&2; exit 1; }
[ -f "$PUB" ] || { echo "no $PUB" >&2; exit 1; }
[ -f "$TOKEN_FILE" ] || { echo "no $TOKEN_FILE" >&2; exit 1; }
@ -269,6 +282,16 @@ if [ "$DEPLOY" = 1 ]; then
if [ "$PUBLIC" = 1 ]; then "$HERE/publish-public.sh" --verify --no-prune || exit 1; fi
# the console's Builds tab (relay/): one build event and a fresh downloads listing; never fatal
node "$ROOT/tools/console.mjs" post --kind build --title "OTA manifest $PRODUCT $VERSION ($CHANNEL) live" --body "$NOTES" >/dev/null 2>&1 || true
# the nudge (2.0.2, 8 October 2026): every machine with a relay secret is asked once, through the relay, to run
# packaging/ota/update-check.ps1 (the engine's own api/update/check), so a published entry stages within a minute instead
# of at the engine's next hourly check; --no-nudge skips it; a machine the relay cannot reach is one line, never a failure
if [ "${NUDGE:-1}" = 1 ] && [ -d "$HOME/.config/igneum/relay-machines" ]; then
for sf in "$HOME/.config/igneum/relay-machines"/*; do
[ -f "$sf" ] || continue
m="$(basename "$sf")"
if node "$ROOT/tools/relay.mjs" run "$m" "update check after the $VERSION ($CHANNEL) publish: the engine's own api/update/check, then what it staged" "$ROOT/packaging/ota/update-check.ps1" >/dev/null 2>&1; then echo "nudged $m (api/update/check through the relay)"; else echo "could not nudge $m through the relay (it checks on its own clock)"; fi
done
fi
node "$ROOT/tools/console.mjs" sync-dl >/dev/null 2>&1 || true
else
if [ -n "$DLSITE" ]; then

14
packaging/ota/update-check.ps1 Normal file
View file

@ -0,0 +1,14 @@
# The nudge after a manifest publish (2.0.2, the shipper's order of 8 October 2026): the engine's own manifest check is hourly
# today (ten minutes from 2.0.2), so a published entry would wait up to an hour; this run, posted by publish-manifest.sh
# through the relay to every machine with a secret, asks the local engine once through its own API (api/update/check,
# the token in app.url) and reads what it staged. Reads only beyond that one POST; nothing installed here: the engine's
# own path installs at its next safe moment and the MINING-ON relay run reads the apply.
$ErrorActionPreference = 'Continue'
$urlFile = Join-Path $env:LOCALAPPDATA 'igneum\app\app.url'
if (-not (Test-Path -LiteralPath $urlFile)) { Write-Output 'RESULT UPDATE-CHECK no app.url (the app is not running)'; exit 1 }
$u = (Get-Content -LiteralPath $urlFile -Raw).Trim()
try { $before = Invoke-RestMethod -Uri ($u + 'api/state') -TimeoutSec 6 -UseBasicParsing } catch { Write-Output ("RESULT UPDATE-CHECK no engine answers: " + $_.Exception.Message); exit 1 }
try { Invoke-RestMethod -Method Post -Uri ($u + 'api/update/check') -TimeoutSec 20 -UseBasicParsing | Out-Null } catch { Write-Output ("RESULT UPDATE-CHECK api/update/check failed: " + $_.Exception.Message); exit 1 }
Start-Sleep -Seconds 45
try { $after = Invoke-RestMethod -Uri ($u + 'api/state') -TimeoutSec 6 -UseBasicParsing } catch { $after = $before }
Write-Output ('RESULT UPDATE-CHECK asked version=' + $after.version + ' manifest=' + $after.update.version + ' state=' + $after.update.state + ' note="' + $after.update.note + '" at ' + (Get-Date).ToUniversalTime().ToString('HH:mm:ss') + 'Z')

124
proto-cuda/nvrtc/emu/geometry-vectors.txt Normal file
View file

@ -0,0 +1,124 @@
# Pack geometry conformance vectors (8 October 2026, master review I06, gate INT-10): one normative rule set for the
# dataset geometry of a program pack, read by every host the same way: packfile.h pf_load (the CUDA worker and the
# OpenCL generic worker; checked by emu/packfile-test.c) and proto-metal/main.swift packGeometry (checked by
# igneum-bench --geometry-check). A host allocates, builds and hashes from the exact word count, never from the log2
# alone, with 64-bit byte sizes, and refuses an unsupported pack before any GPU launch.
#
# Each case: `case <name>`, then `define <NAME> <value>` lines appended to a base program.h that carries everything
# but the geometry (seeds, generator 2, attempt, the init words, the cache sizes, IGNEUM_DATASET_MODE 1), then one
# `expect ok words=<n> items=<n> mulshift=<0|1> bytes=<n>` or `expect refuse <a substring of the refusal>`.
# A `define IGNEUM_GENERATOR` or `define IGNEUM_STATE_LEAVES` line overrides the base's class fields.
case ds55
define IGNEUM_DATASET_LOG2 30
define IGNEUM_DATASET_WORDS 1476395008u
define IGNEUM_DATASET_ITEMS 92274688u
define IGNEUM_DATASET_BYTES 5905580032ull
define IGNEUM_DATASET_MULSHIFT 1
expect ok words=1476395008 items=92274688 mulshift=1 bytes=5905580032
case ds85
define IGNEUM_DATASET_LOG2 31
define IGNEUM_DATASET_WORDS 2281701376u
define IGNEUM_DATASET_ITEMS 142606336u
define IGNEUM_DATASET_MULSHIFT 1
expect ok words=2281701376 items=142606336 mulshift=1 bytes=9126805504
case ds115
define IGNEUM_DATASET_LOG2 31
define IGNEUM_DATASET_WORDS 3087007744u
define IGNEUM_DATASET_ITEMS 192937984u
define IGNEUM_DATASET_MULSHIFT 1
expect ok words=3087007744 items=192937984 mulshift=1 bytes=12348030976
case legacy-2^28
define IGNEUM_DATASET_LOG2 28
expect ok words=268435456 items=16777216 mulshift=0 bytes=1073741824
case legacy-2^30
define IGNEUM_DATASET_LOG2 30
expect ok words=1073741824 items=67108864 mulshift=0 bytes=4294967296
case legacy-2^31-explicit-count
define IGNEUM_DATASET_LOG2 31
define IGNEUM_DATASET_WORDS 2147483648u
define IGNEUM_DATASET_ITEMS 134217728u
define IGNEUM_DATASET_MULSHIFT 0
expect ok words=2147483648 items=134217728 mulshift=0 bytes=8589934592
case wrong-count-not-a-65536-multiple
define IGNEUM_DATASET_LOG2 30
define IGNEUM_DATASET_WORDS 1476395009u
define IGNEUM_DATASET_ITEMS 92274688u
define IGNEUM_DATASET_MULSHIFT 1
expect refuse multiple of 65,536
case wrong-log2-for-the-count
define IGNEUM_DATASET_LOG2 29
define IGNEUM_DATASET_WORDS 1476395008u
define IGNEUM_DATASET_ITEMS 92274688u
define IGNEUM_DATASET_MULSHIFT 1
expect refuse IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count
case wrong-log2-for-a-power-of-two-count
define IGNEUM_DATASET_LOG2 28
define IGNEUM_DATASET_WORDS 1073741824u
define IGNEUM_DATASET_ITEMS 67108864u
define IGNEUM_DATASET_MULSHIFT 0
expect refuse IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count
case wrong-items
define IGNEUM_DATASET_LOG2 30
define IGNEUM_DATASET_WORDS 1476395008u
define IGNEUM_DATASET_ITEMS 92274689u
define IGNEUM_DATASET_MULSHIFT 1
expect refuse IGNEUM_DATASET_ITEMS is not IGNEUM_DATASET_WORDS / 16
case wrong-mulshift-on-a-non-power-of-two
define IGNEUM_DATASET_LOG2 30
define IGNEUM_DATASET_WORDS 1476395008u
define IGNEUM_DATASET_ITEMS 92274688u
define IGNEUM_DATASET_MULSHIFT 0
expect refuse IGNEUM_DATASET_MULSHIFT does not match the word count
case log2-32
define IGNEUM_DATASET_LOG2 32
expect refuse IGNEUM_DATASET_LOG2 above 31
case overflowing-count-2^32
define IGNEUM_DATASET_LOG2 31
define IGNEUM_DATASET_WORDS 4294967296ull
define IGNEUM_DATASET_MULSHIFT 0
expect refuse IGNEUM_DATASET_WORDS above 2^32 - 1
case overflowing-count-2^40
define IGNEUM_DATASET_LOG2 31
define IGNEUM_DATASET_WORDS 1099511627776ull
expect refuse IGNEUM_DATASET_WORDS above 2^32 - 1
case count-below-the-minimum
define IGNEUM_DATASET_LOG2 15
define IGNEUM_DATASET_WORDS 32768u
expect refuse multiple of 65,536
case log2-below-20
define IGNEUM_DATASET_LOG2 19
expect refuse sizes out of range
case missing-log2
define IGNEUM_DATASET_WORDS 1476395008u
expect refuse has no IGNEUM_DATASET_LOG2
case missing-leaves-on-a-class-v5-pack
define IGNEUM_GENERATOR 5
define IGNEUM_PROGRAM_CLASS "v5"
define IGNEUM_ERA_SEED_HEX "bed7ab62cbece66cf791485336d81d90fa1452ffed28ecd8a7416960ef64164c"
define IGNEUM_SHADOW_INSTRS 256
define IGNEUM_SHADOW_REPS 27
define IGNEUM_DATASET_LOG2 28
expect refuse without IGNEUM_STATE_LEAVES
case leaves-on-a-class-v2-pack
define IGNEUM_STATE_LEAVES 93
define IGNEUM_DATASET_LOG2 28
expect refuse state leaves belong to class v5

50
proto-cuda/nvrtc/emu/packfile-test.c
View file

@ -269,6 +269,56 @@ int main(int argc, char** argv) {
}
}
// 6. The geometry conformance vectors (8 October 2026, master review I06, gate INT-10): emu/geometry-vectors.txt, one
// normative rule set shared with the Metal host (igneum-bench --geometry-check reads the same file). Each case's defines
// are appended to a base program.h with everything but the geometry; pf_load must size the pack exactly as the vector
// says (words, items, mulshift, 64-bit bytes) or refuse it with the named words. Known-failed first: the vectors file
// itself carries the refusals (an overflowing count, a log2 that is not the floor, a wrong item count).
if (argc > 3) {
FILE* vf = fopen(argv[3], "rb");
char line[512], name[128] = {0}, defs[4096] = {0}, text[8192], sw[200], kw[200];
int cases = 0, inCase = 0;
words_hex(att1, sw); words_hex(keyw, kw);
CHECK(vf != NULL, "the geometry vectors file opens");
while (vf && fgets(line, sizeof(line), vf)) {
line[strcspn(line, "\r\n")] = 0;
if (line[0] == '#' || line[0] == 0) continue;
if (strncmp(line, "case ", 5) == 0) { snprintf(name, sizeof(name), "%s", line + 5); defs[0] = 0; inCase = 1; continue; }
if (!inCase) continue;
if (strncmp(line, "define ", 7) == 0) { snprintf(defs + strlen(defs), sizeof(defs) - strlen(defs), "#define %s\n", line + 7); continue; }
if (strncmp(line, "expect ", 7) == 0) {
const char* ex = line + 7;
int hasGen = strstr(defs, "#define IGNEUM_GENERATOR ") != NULL;
char what[400];
snprintf(text, sizeof(text),
"#define IGNEUM_SEED_BYTES_HEX \"%s\"\n#define IGNEUM_DAY_BYTES_HEX \"%s\"\n%s#define IGNEUM_PROGRAM_ATTEMPT 1\n#define IGNEUM_DATASET_MODE 1\n"
"#define IGNEUM_SEEDW_INIT { %s }\n#define IGNEUM_KEY_INIT { %s }\n#define IGNEUM_CACHE_LOG2_WORDS 26\n#define IGNEUM_CACHE_SEGMENTS 4096u\n%s",
EPOCH_34, DAY_20731, hasGen ? "" : "#define IGNEUM_GENERATOR 2\n", sw, kw, defs);
write_file(dir, "program.h", text);
write_file(dir, "seeds.txt", "epoch_seed_hex " EPOCH_34 "\nday_seed_hex " DAY_20731 "\n");
err[0] = 0;
++cases;
if (strncmp(ex, "ok ", 3) == 0) {
unsigned long long w = 0, it = 0, ms = 0, by = 0;
int ok = pf_load(dir, &pk, err, sizeof(err));
sscanf(ex, "ok words=%llu items=%llu mulshift=%llu bytes=%llu", &w, &it, &ms, &by);
snprintf(what, sizeof(what), "geometry %s: loads as %llu words, %llu items, mulshift %llu, %llu bytes", name, w, it, ms, by);
CHECK(ok == 1 && pk.datasetWords == (uint32_t)w && pk.datasetItems == (uint32_t)it && pk.datasetMulshift == (uint32_t)ms && (uint64_t)pk.datasetWords * 4ull == by, what);
if (!ok) printf(" (%s)\n", err);
} else if (strncmp(ex, "refuse ", 7) == 0) {
int ok = pf_load(dir, &pk, err, sizeof(err));
snprintf(what, sizeof(what), "geometry %s: refused with \"%s\"", name, ex + 7);
CHECK(ok == 0 && strstr(err, ex + 7) != NULL, what);
if (ok) printf(" (loaded: %u words)\n", (unsigned)pk.datasetWords); else if (strstr(err, ex + 7) == NULL) printf(" (said: %s)\n", err);
} else { CHECK(0, "a vector's expect line is ok or refuse"); }
inCase = 0;
}
}
if (vf) fclose(vf);
CHECK(cases >= 18, "the vectors file carries every case (18 or more)");
printf(" %d geometry vectors\n", cases);
}
printf("%s: %d failure(s)\n", argv[0], failures);
return failures ? 1 : 0;
}

5
proto-cuda/nvrtc/emu/packfile-test.sh
View file

@ -1,6 +1,7 @@
#!/usr/bin/env bash
# The pack loader's seed rule (packfile.h) on a known-good and a known-mismatched pack, and the class v5 leaf rule on the
# checked-in v5 pack (7 October 2026): emu/packfile-test.c, C99, no GPU. Runs on the Mac in a second and in CI. Usage: emu/packfile-test.sh
# checked-in v5 pack (7 October 2026), and the geometry conformance vectors (emu/geometry-vectors.txt, 8 October 2026, shared with
# the Metal host): emu/packfile-test.c, C99, no GPU. Runs on a box in a second and in CI. Usage: emu/packfile-test.sh
set -euo pipefail
HERE="$(cd "$(dirname "$0")" && pwd)"
ROOT="$(cd "$HERE/../../.." && pwd)"
@ -8,4 +9,4 @@ OUT="${TMPDIR:-/tmp}/igneum-packfile-test"
mkdir -p "$OUT"
CC="${CC:-cc}"
"$CC" -std=c99 -Wall -Wextra -Wno-unused-function -O1 -o "$OUT/packfile-test" "$HERE/packfile-test.c"
"$OUT/packfile-test" "$ROOT/proto-cuda/packs/igneum-devnet-v4-epoch0" "$ROOT/proto-cuda/packs-ca3-v5/v5-dn3-epoch0"
"$OUT/packfile-test" "$ROOT/proto-cuda/packs/igneum-devnet-v4-epoch0" "$ROOT/proto-cuda/packs-ca3-v5/v5-dn3-epoch0" "$HERE/geometry-vectors.txt"

2
proto-cuda/nvrtc/emu/select-check.sh
View file

@ -13,6 +13,7 @@
# found_set_equals_full_stale_job_queued s1, s2: two job lines queued at once, the second served after the first
# buffer_reuse_hits_then_zero b1 then b2: a hit chunk followed by a zero-target chunk prints nothing stale
# mismatch_error_then_job m1 (no pair for its seeds) errors the same way, c1 after it is served
# epoch_tag every found line carries the pack's epoch (16 hex) as its fifth field, both modes (I05)
# Usage: select-check.sh <pack A dir> <out dir> <worker command...> (the command carries --serve --pack <pack A> etc.)
set -euo pipefail
PACK_A="$1"; OUT="$2"; shift 2
@ -74,6 +75,7 @@ grep -q '^done r1 2048' "$OUT/rb-select.norm" || { echo "FAIL: high32_rollover:
awk '$1 == "found" && $2 == "r1" && $3 + 0 >= 4294967296 { n++ } END { exit n > 0 ? 0 : 1 }' "$OUT/rb-select.norm" || { echo "FAIL: high32_rollover: no found nonce at or past 2^32"; exit 1; }
grep -q '^done s1 2048' "$OUT/rb-select.norm" && grep -q '^done s2 2048' "$OUT/rb-select.norm" || { echo "FAIL: stale_job_queued: s1 or s2 did not finish"; exit 1; }
[ "$(count b1 "$OUT/rb-select.norm")" -gt 0 ] && [ "$(count b2 "$OUT/rb-select.norm")" = 0 ] || { echo "FAIL: buffer_reuse_hits_then_zero: b1 $(count b1 "$OUT/rb-select.norm") found, b2 $(count b2 "$OUT/rb-select.norm")"; exit 1; }
awk -v e="${A_EPOCH:0:16}" '$1 == "found" && $5 != e { bad++ } END { exit bad > 0 ? 1 : 0 }' "$OUT/rb-select.norm" && awk -v e="${A_EPOCH:0:16}" '$1 == "found" && $5 != e { bad++ } END { exit bad > 0 ? 1 : 0 }' "$OUT/rb-full.norm" || { echo "FAIL: epoch_tag: a found line lacks the pack's epoch ${A_EPOCH:0:16} as its fifth field"; exit 1; }
grep -q '^error m1 ' "$OUT/rb-select.norm" && grep -q '^done c1 1024' "$OUT/rb-select.norm" || { echo "FAIL: mismatch_error_then_job: m1 did not error or c1 was not served"; exit 1; }
grep -q 'full fallbacks' "$OUT/rb-select.log" || { echo "FAIL: no transfers line in the select run"; exit 1; }
fb="$(sed -n 's/.*(select, \([0-9]*\) full fallbacks).*/\1/p' "$OUT/rb-select.log" | tail -1)"

2
proto-cuda/nvrtc/emu/serve-check.sh
View file

@ -53,6 +53,8 @@ elif grep -q '^info job 5 is for epoch .* building its pack' "$LOG" && [ "$(grep
else
echo "FAIL: job 5 was neither refused nor self-healed"; exit 1
fi
[ "$(awk -v e="${A_EPOCH:0:16}" '$1 == "found" && ($2 == 1 || $2 == 2 || $2 == 3) && $5 != e' "$LOG" | wc -l | tr -d ' ')" = 0 ] || { echo "FAIL: a found line of a pack A job does not carry epoch ${A_EPOCH:0:16} as its fifth field"; exit 1; }
[ "$(awk -v e="${B_EPOCH:0:16}" '$1 == "found" && $2 == 4 && $5 != e' "$LOG" | wc -l | tr -d ' ')" = 0 ] || { echo "FAIL: a found line of job 4 (pack B after the swap) does not carry epoch ${B_EPOCH:0:16} as its fifth field"; exit 1; }
grep -q '^found 2 4294967295 ' "$LOG" && grep -q '^found 2 4294967296 ' "$LOG" || { echo "FAIL: the 32-bit boundary nonces are missing"; exit 1; }
for x in 0 1 2 31 32 63; do check 1 $x "$A_EPOCH" "$A_DAY"; done
for x in 4294967264 4294967295 4294967296 4294967327; do check 2 $x "$A_EPOCH" "$A_DAY"; done

17
proto-cuda/nvrtc/packfile.h
View file

@ -302,8 +302,23 @@ static int pf_load(const char* dir, PfPack* pk, char* err, size_t cap) {
if (!pf_define_u32(prog, "IGNEUM_DATASET_LOG2", &pk->datasetLog2)) { free(prog); return pf_fail(err, cap, "program.h has no IGNEUM_DATASET_LOG2"); }
if (!pf_define_u32(prog, "IGNEUM_DATASET_MODE", &pk->datasetMode)) pk->datasetMode = 0;
if (pk->datasetLog2 > 31) { free(prog); return pf_fail(err, cap, "program.h IGNEUM_DATASET_LOG2 above 31: the word index is 32-bit"); }
if (!pf_define_u32(prog, "IGNEUM_DATASET_WORDS", &pk->datasetWords)) pk->datasetWords = 1u << pk->datasetLog2;
{
// the word count is parsed at 64 bits first: a count at or above 2^32 (ds115 is 3,087,007,744 words, under it) would wrap
// in the 32-bit field and size the dataset at a fraction of the pack (the geometry vectors, emu/geometry-vectors.txt, 8 October 2026)
uint64_t w64 = 0;
if (pf_define_u64(prog, "IGNEUM_DATASET_WORDS", &w64)) {
if (w64 > 0xffffffffull) { free(prog); return pf_fail(err, cap, "program.h IGNEUM_DATASET_WORDS above 2^32 - 1: the word index is 32-bit"); }
pk->datasetWords = (uint32_t)w64;
} else pk->datasetWords = 1u << pk->datasetLog2;
}
if (pk->datasetWords < (1u << 16) || (pk->datasetWords & 0xffffu) != 0) { free(prog); return pf_fail(err, cap, "program.h IGNEUM_DATASET_WORDS is not a multiple of 65,536 words"); }
{
// IGNEUM_DATASET_LOG2 is the floor of log2 of the count (the exporter writes it so; a reader that sizes by the log2 alone
// is then at most a factor of two short, never wrong by a different pack's geometry): a count that disagrees is refused
uint32_t fl = 0, w = pk->datasetWords;
while (w > 1u) { w >>= 1; ++fl; }
if (fl != pk->datasetLog2) { free(prog); return pf_fail(err, cap, "program.h IGNEUM_DATASET_LOG2 is not the floor of log2 of the word count"); }
}
if (!pf_define_u32(prog, "IGNEUM_DATASET_ITEMS", &pk->datasetItems)) pk->datasetItems = pk->datasetWords / 16u;
if (pk->datasetItems != pk->datasetWords / 16u) { free(prog); return pf_fail(err, cap, "program.h IGNEUM_DATASET_ITEMS is not IGNEUM_DATASET_WORDS / 16"); }
if (!pf_define_u32(prog, "IGNEUM_DATASET_MULSHIFT", &pk->datasetMulshift)) pk->datasetMulshift = 0;

10
proto-cuda/nvrtc/serve-bench.sh
View file

@ -21,9 +21,13 @@ LOG="$OUT/$RUN.worker.log"; TL="$OUT/$RUN.timeline"; SMI="$OUT/$RUN.smi"; ROW="$
now() { date +%s.%N; }
count=$((1 << B))
target="00000fffffffffff" # one hit per 2^20 nonces: 16 found lines per 2^24-nonce job, the serving shape (a block target is far tighter)
seedline() { sed -n "s/^$2 //p" "$1/seeds.txt"; }
# the pack's seeds: seeds.txt as igneum-miner export-pack writes it, else the byte seeds program.h carries (a kit pack)
seedline() { sed -n "s/^$2 //p" "$1/seeds.txt" 2>/dev/null || true; } # a missing file is not a failure under set -e
define() { sed -n "s/^#define $2 \"\(.*\)\"/\1/p" "$1/program.h" 2>/dev/null || true; }
EPOCH="$(seedline "$PACK" epoch_seed_hex)"; DAY="$(seedline "$PACK" day_seed_hex)"
[ -n "$EPOCH" ] && [ -n "$DAY" ] || { echo "no seeds.txt in $PACK" >&2; exit 2; }
[ -n "$EPOCH" ] || EPOCH="$(define "$PACK" IGNEUM_SEED_BYTES_HEX)"
[ -n "$DAY" ] || DAY="$(define "$PACK" IGNEUM_DAY_BYTES_HEX)"
[ -n "$EPOCH" ] && [ -n "$DAY" ] || { echo "no seeds.txt and no IGNEUM_SEED_BYTES_HEX / IGNEUM_DAY_BYTES_HEX in $PACK/program.h" >&2; exit 2; }
jobline() { printf 'job %d %064x %s %d %d %s %s\n' "$1" "$1" "$target" $(( $1 * count )) "$count" "$EPOCH" "$DAY"; }
# power sampler first (its own pid file)
@ -31,7 +35,7 @@ nvidia-smi -i "$DEV" --query-gpu=timestamp,power.draw,clocks.sm,temperature.gpu
echo $! > "$PIDDIR/$RUN.smi.pid"
sleep 1
t0=$(now)
coproc WK { IGNEUM_READBACK="$MODE" "$W" --serve --pack "$PACK" --device "$DEV" --batch-log2 "$B" --race off --readback "$MODE" 2>&1; }
coproc WK { exec env IGNEUM_READBACK="$MODE" "$W" --serve --pack "$PACK" --device "$DEV" --batch-log2 "$B" --race off --readback "$MODE" 2>&1; } # exec: the pid is the worker's, not a subshell's
wkpid=$WK_PID # bash clears WK_PID when the coproc ends; the copy outlives it
echo "$wkpid" > "$PIDDIR/$RUN.worker.pid"
sent=0; done_n=0; found=0; tready=""; tlast=""; cpu=""

9
proto-cuda/nvrtc/worker.cpp
View file

@ -18,8 +18,9 @@
// prepare <epoch_seed_hex 64> <day_seed_hex> <pack_dir> compile that pack in the background, build its cache
// and dataset, self-test it; a job on it then switches
// quit
// stdout: ready cuda <device> pack <seed string> dataset-log2 N batch B regs R prepare 1 path nvrtc ...
// found <job_id> <nonce u64> <hash_hex 16>
// stdout: ready cuda <device> pack <seed string> dataset-log2 N batch B regs R prepare 1 path nvrtc ... readback select|full ...
// found <job_id> <nonce u64> <hash_hex 16> <epoch_hex 16> the fifth field (8 October 2026, I05): the epoch of the pair
// the hash was computed under, so stale work is never read as current
// done <job_id> <hashes> <ms>
// error <job_id> <text>
// need <epoch_seed_hex> <day_seed_hex> before the mismatch error: the pair this worker lacks (the miner prepares it)
@ -1416,12 +1417,12 @@ static int runServe(Ctx& c, const Options& o, Pair* cur) {
}
for (uint32_t a = 0; a < nHits; ++a) {
uint64_t nonce = ((uint64_t)hi << 32) | (uint64_t)(uint32_t)(lo + (uint32_t)sel.hHits[2 * a]);
std::printf("found %s %llu %016llx\n", jobId.c_str(), (unsigned long long)nonce, (unsigned long long)sel.hHits[2 * a + 1]);
std::printf("found %s %llu %016llx %.16s\n", jobId.c_str(), (unsigned long long)nonce, (unsigned long long)sel.hHits[2 * a + 1], cur->epochHex.c_str());
}
} else {
for (uint32_t i = 0; i < chunk; ++i) if (hOut[i] <= target) {
uint64_t nonce = ((uint64_t)hi << 32) | (uint64_t)(uint32_t)(lo + i);
std::printf("found %s %llu %016llx\n", jobId.c_str(), (unsigned long long)nonce, (unsigned long long)hOut[i]);
std::printf("found %s %llu %016llx %.16s\n", jobId.c_str(), (unsigned long long)nonce, (unsigned long long)hOut[i], cur->epochHex.c_str());
}
}
scanMs = wallMs() - r0;

11
proto-opencl/host.c
View file

@ -1156,6 +1156,13 @@ typedef struct {
cl_uint stateLeaves; /* class v5: the state leaves the dataset was built from (uploaded for the build, released after); 0 otherwise */
} ServePair;
/* The found line's fifth field (8 October 2026, I05): the epoch the pair's program was built for, 16 hex, so stale work is never read
* as current; the compiled-in pack's epoch comes from its program.h. */
#ifndef IGNEUM_SEED_BYTES_HEX
#define IGNEUM_SEED_BYTES_HEX "0000000000000000"
#endif
static const char* pairEpochTag(const ServePair* p) { return p->epochHex[0] ? p->epochHex : IGNEUM_SEED_BYTES_HEX; }
static int hexEq(const char* a, const char* b) {
size_t i;
if (strlen(a) != strlen(b)) return 0;
@ -1929,12 +1936,12 @@ static int runServe(Device* dv, const DeviceInfo* di, const Options* o) {
for (a = 0; a < nHits; ++a) {
uint32_t idx = (uint32_t)hHits[2 * a];
unsigned long long nonce = ((unsigned long long)hi << 32) | (unsigned long long)(uint32_t)(lo + idx);
printf("found %s %llu %016llx\n", jobId, nonce, (unsigned long long)hHits[2 * a + 1]);
printf("found %s %llu %016llx %.16s\n", jobId, nonce, (unsigned long long)hHits[2 * a + 1], pairEpochTag(cur));
}
} else {
for (i = 0; i < chunk; ++i) if (hOut[i] <= target) {
unsigned long long nonce = ((unsigned long long)hi << 32) | (unsigned long long)(uint32_t)(lo + i);
printf("found %s %llu %016llx\n", jobId, nonce, (unsigned long long)hOut[i]);
printf("found %s %llu %016llx %.16s\n", jobId, nonce, (unsigned long long)hOut[i], pairEpochTag(cur));
}
}
scanMs = wallMs() - r0;

BIN
proving/igneum-prove/elf/prior/igneum-prove-aggregator.elf Normal file
View file

Binary file not shown.

BIN
proving/igneum-prove/elf/prior/igneum-prove-aggregator.vk Normal file
View file

Binary file not shown.

BIN
proving/igneum-prove/elf/prior/igneum-prove-program.elf Normal file
View file

Binary file not shown.

BIN
proving/igneum-prove/elf/prior/igneum-prove-program.vk Normal file
View file

Binary file not shown.

44
proving/igneum-prove/host/src/bin/pin.rs
View file

@ -70,10 +70,32 @@ fn main() -> Result<()> {
let out = arg("--out").map(PathBuf::from).unwrap_or_else(|| here.join("elf"));
std::fs::create_dir_all(&out)?;
let host = std::process::Command::new("uname").arg("-a").output().ok().map(|o| String::from_utf8_lossy(&o.stdout).trim().to_string()).unwrap_or_else(|| "unknown".into());
// Key succession (docs/design/key-succession.md, 8 October 2026): a pin over an existing elf/ moves the pair it
// holds to elf/prior/ and writes the manifest's `prior` block (the same per-program fields) and a `succession`
// block naming both pairs' ids; the top-level shard and aggregator stay the newest pair. At most two pairs live:
// a pin over an elf/ that already carries prior/ replaces prior/ with the current pair. The node reads H, W and
// the next ids from its object, never from the manifest; `succession` is for the served manifest and its check.
// `--no-prior` keeps today's shape (the bootstrap, or a pin that replaces a pair never served).
let keep_prior = !args.iter().any(|a| a == "--no-prior");
let prior = if keep_prior && out.join("manifest.json").exists() {
let current: serde_json::Value = serde_json::from_str(&std::fs::read_to_string(out.join("manifest.json"))?)?;
let prior_dir = out.join("prior");
std::fs::create_dir_all(&prior_dir)?;
for name in ["igneum-prove-program", "igneum-prove-aggregator"] {
for ext in ["elf", "vk"] {
let f = format!("{name}.{ext}");
std::fs::copy(out.join(&f), prior_dir.join(&f)).with_context(|| format!("move {f} to prior/"))?;
}
}
println!("RESULT pin: the pair pinned at {} moved to {} (prior)", current["pinned_at"].as_str().unwrap_or("?"), prior_dir.display());
Some(serde_json::json!({ "shard": current["shard"], "aggregator": current["aggregator"], "pinned_at": current["pinned_at"] }))
} else {
None
};
let light = LightProver::new();
let shard = pin_one(&light, &from, &out, "igneum-prove-program")?;
let aggregator = pin_one(&light, &from, &out, "igneum-prove-aggregator")?;
let manifest = serde_json::json!({
let mut manifest = serde_json::json!({
"format": "igneum-prove-elf-manifest-v1",
"sp1_crate_version": "6.8.1",
"sp1_circuit_version": sp1_sdk::SP1_CIRCUIT_VERSION,
@ -82,6 +104,26 @@ fn main() -> Result<()> {
"shard": shard,
"aggregator": aggregator,
});
if let Some(p) = prior {
manifest["succession"] = serde_json::json!({
"prior_shard_program_id": p["shard"]["program_id"],
"prior_aggregator_id": p["aggregator"]["program_id"],
"next_shard_program_id": manifest["shard"]["program_id"],
"next_aggregator_id": manifest["aggregator"]["program_id"],
});
manifest["prior"] = p;
} else {
// no pair to carry: prior/ is a copy of the current pair so the host's include_bytes compile, the manifest
// names no `prior` (the in-process verifier reads it as absent)
let prior_dir = out.join("prior");
std::fs::create_dir_all(&prior_dir)?;
for name in ["igneum-prove-program", "igneum-prove-aggregator"] {
for ext in ["elf", "vk"] {
let f = format!("{name}.{ext}");
std::fs::copy(out.join(&f), prior_dir.join(&f))?;
}
}
}
std::fs::write(out.join("manifest.json"), format!("{}\n", serde_json::to_string_pretty(&manifest)?))?;
println!("RESULT pin: manifest written to {}; rebuild the host so it embeds these files", out.join("manifest.json").display());
Ok(())

Some files were not shown because too many files have changed in this diff Show more