igneum-network/igneum
1
0
Fork 0
Code Issues Pull requests Releases Activity
2395 commits 395 branches 1 tag 2.7 GiB
Commit graph

1551 commits

Author SHA1 Message Date
igneum-labs
08010e76a4 Class v6 family gate, 6.9 closed: the best-mix table's exact or = 0 attempts rows at both widths on build-4 (r 0.29 and 0.16, 0 exhausted, the floors at the width's cost), the owed line closed; the harness ported onto class-v6 a7c2c082f for the layer-8-off acceptance 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 17:20:45 +00:00
igneum-labs
8ce0956b48 Class v6 family gate, amendment 6.9: the best-mix genesis table's acceptance for D1 (PASS: r 0.33 and 0.19 at widths 4 and 1, 0 exhausted, the floors at the width's cost, attack-f8 at the exact table 0 hot sets and 0 over 1.2x at 2^20; the exact or = 0 attempts rows owed with build-3 unreachable); the Igneum 2.0 pause recorded 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 16:54:11 +00:00
igneum-labs
4eca18c34e Class v6 family gate, the full report: point B's row (0 hot sets at the 64 shape, 5 over 1.2x, the floor refusing 1 of 64) and the 18:00 amendment line (the best-mix genesis table under test) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 14:40:03 +00:00
igneum-labs
72b58cec74 Class v6 family gate, the full report (16:30 BST): 38,000 drawn eras, the lossy-share curve per shape (the exhaustion a 256-shape interaction, about 6x the independent estimate), the pre-floor spread per width and the width-4 floor decision (0.995 kept at its measured cost), the two live-dataset censuses (128 band epochs: 2 hot sets at the 256 shape, both refused by the class v5 floor), the seven known-failed seeds through the sigma form (one value-level test, the bucket bound retired into the bit read; p212 and p225 at bit R), the x4, x8 and x16 verifier rows (x16 outside the band on the 10 ms gate), the band stratum's gate records as JSON lines; every row's log under logs/ 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 14:31:37 +00:00
igneum-labs
a39dbeca29 Class v6 family gate, the 17:00 cut: the harness as a plain diff at 3dc3117c (the founder-strings check refuses the patch form's author lines) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 11:35:25 +00:00
igneum-labs
debcb01092 Class v6 family gate, the 17:00 cut: the last-resort rows corrected for class v5's verified scan (the finding at the corner is the correlation of one era's attempts, 1,000x the independent estimate, not an unchecked program); the harness patch series at 236ef3a5 (the lossy-share curve and the scan's verdict column) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 11:27:11 +00:00
igneum-labs
d4cd6a3ce2 Class v6 family gate, the 17:00 cut: the measured coverage (24,000 drawn eras through the ring-B tests on build-1 and build-3, 54 core-hours), per stratum and per axis, with the bound arithmetic on the counts; four measured rows (the lossy corner of B = 4 exhausts the 256 cap in 1.2 percent of eras, the band capped at the lossy families; the (c''') floor's per-width calibration; the shape axis's draw cost; the era-stride bias at bit R in half the eras of every stratum); the census rows, logs, scripts and the harness patch series under logs/ 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 11:25:26 +00:00
igneum-labs
3ee2030f14 Merge build/master e096e381 into class-v6-family-gate (export-exclude.txt: both research paths kept) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:32:44 +00:00
igneum-labs
836e97d061 Merge ca3-coord-record c4510485 into master (gate: green on c4510485, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 10:31:47 +00:00
igneum-labs
95c3a37483 Merge class-v6-history 49d7ff46 into master (gate: green on 49d7ff46, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 10:26:13 +00:00
igneum-labs
34e41d3518 Class v6 family gate (lane D), first cut: the parameter table with bands and margins, the sampling bound (rule of three, the union bound over the per-era tests' measured miss rates, the stratified corners), the three rings of per-era tests with known-failed shapes and clean-seed costs, the board over the testnet period and the proof-of-testing record format, the honest line; the harness running on the boxes and its first 16-era reading (the era-stride class at 130 to 511 sigma on one address bit in 7 of 16 drawn eras); docs/analysis/class-v6 excluded from the public export 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:25:17 +00:00
igneum-labs
983d766481 Merge class-v6-hardware ffc3a54b into master (gate: green on ffc3a54b, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 10:24:52 +00:00
igneum-labs
c45104856b Counter ASIC 3.0 status: main's word on the two reds (the detach in 0.3.25 if green by 12:15), the day-key scan to 2^28 PASS, the sweep reruns read as none, the attack-pass branch merged and pushed, the family gate's first coverage (16 eras; the index-bit bias firing at address bit R on 7 of 16 drawn eras, a class v6 design row: load_index's form, not a floor; the bucket bound in sigma), the ota test literal fixed (36f7fefa, the detach rides 0.3.25) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:24:10 +00:00
igneum-labs
49d7ff462b Class v6 history lane, first cut: every ASIC-resistant proof of work and its chips by mechanism, the four-layer mapping with a number per chip class, and the three lessons that bind on v6 
docs/analysis/class-v6/history.md: 21 chip rows by mechanism (what each chip specialised, what the design missed, the timeline, the v6 layer, closed or not, the per-joule number); in-depth sections for Ethash (the E3's DDR3, Linzhi's 4.4 GB beside 64 compute units, the Jasminer X4's DRAM hybrid-bonded onto a 40/45 nm die, the E9), ProgPoW (the Least Authority and Rao audits read from the PDFs, Kik, Linzhi's claim, the adopters), RandomX (the X5's RISC-V board at 1.46x, the X9 announced December 2025 and withdrawn May 2026 with zero units, RandomX v2 released 25 March 2026 with activation pending, the Qubic episode as the detector's lesson), Cuckoo (the GRN1's 512 MiB SRAM die, the G32, the iPollo G1, the 2025 bounty), Equihash (the Z9's single die, the Fudan parameter-following design), Scrypt and Argon2 (the BM1485's on-die scratchpads, Percival's lookup gap, MTP), the no-chip hashes, kHeavyHash, CryptoNight, X16R, Lyra2REv2, the compute rows. The four layers against the history layer by layer, with the E3's death dates, Autolykos's growth rule, the design-cycle and FPGA-compile clocks, and the honest verdict that the stored-dataset chip (class C) is not closed by any layer: 5.1x per joule at zero premium, 2.1x with the shadow at k = 1. Two corrections to the 5 October file (the X9, RandomX v2) and one claim marked unverified (the 5x fork-surviving chip). First cut for 15:00 UK; the full report by 09:00 UK on 9 October.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:19:10 +00:00
igneum-labs
ffc3a54bdc Class v6 research lane B: the hardware future, five years out (first cut and the full report in one file) 
docs/analysis/class-v6/hardware-future.md: one table over PIM and PNM (HBM-PIM, AiM, LPDDR5X-PIM, UPMEM), HBM3E, HBM4 and the custom HBM4E base die, fine-grained DRAM on logic, LPDDR6, 3D DRAM, RLDRAM 3 and Folded Banks, FPGA with HBM2e, wafer-scale, CXL, optical I/O, chiplets and UCIe, and SRAM at N2: energy per dependent read, latency, cost per GB, availability to a non-hyperscaler, the k_read and shadow k bands, the edge per joule at zero shadow against the 5090 and the M5 Max, and the one layer of the four that blunts each with a number. The three findings: a 2 GiB SRAM full store on one N2 reticle is the strongest five-year chip (13x to 17x per joule, about USD 0.3 per MH/s) and no layer reaches it (layer 2 moves its capex, not its joules; the shadow at k = 0.5 holds 4.8x); per-bank PIM is blind (1.6 percent of reads in-bank at 2 GiB) and the custom HBM4E base die is the f = 1 chip with its controller in the stack (6.5x to 14x), untouched by layers 1 to 4; the denominator is the wrong card (the M5 Max at 0.78 microjoules halves to thirds every chip edge). Section 5 carries the k bands for the research lane's chip rows; section 7 the four decisions with defaults and deadlines; section 9 the sources with URLs and dates. tools/ci/export-exclude.txt: docs/analysis/class-v6 is research, not public export.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:17:50 +00:00
igneum-labs
c22a541d12 Counter ASIC 3.0 status: the founder's load order on the four boxes and its placement (lane D's family sampling, the attack families at scale, lane C's packs, the pin matrix, the cross builds), the drift class settled by a fresh-walk node, the pin candidate e0644958 confirmed, the Windows installer class and the detach fix, the tiers' class-flip pair gated (081b3ba7), the rpc up, two reds on the release line from the first ten minutes of the load order (the missing 60x field; the detach's test literal) with their fixes and defaults 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:14:50 +00:00
igneum-labs
0c548b60c7 Merge ship-docs-0321 2a4a828b into master (gate: green on 2a4a828b, recorded by tools/ci/pre-push.sh; the full gate runs in CI on this merge) 2026-10-08 10:12:42 +00:00
igneum-labs
2a4a828be0 release record section 24: the first lock on cc902690 (checkpoint 1931, 10:06:32 BST), the move's count and the three puller faults closed, 0.3.24 Windows take 2 (Inno stopped the app and copied nothing; the /IGNOTA=1 rule), 0.3.25 opened (tip 87d8862d, the ceiling cut e0644958, digest 1b37cb9d, the DMG 501ba293 staged, the clock), the tiers at the class flip, the signed apps pipeline (Igneum Labs LLC) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:03:40 +00:00
igneum-labs
e62adc83cf Counter ASIC 3.0 status: the 0.3.25 pairs and app tip (87d8862d on the ceiling cut e0644958, its fast-time PASS at 10:54), the tiers' class-flip UI half (d949e274), the four builder lanes and the class v6 declaration with its four research lanes spawned and the outline on the mirror, the Arc re-read and v5lock jobs live, the held-push cause (the branch's own hook predating the per-process fix) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 10:01:24 +00:00
igneum-labs
6443f921e9 Counter ASIC 3.0 status: the 0.3.25 pairs on 7bd2940f and the re-pointed app tip c68aea33 (every input green at 10:33 bar the pin's gates and the crossing), main's sweep finding (class v5 17 percent under v4 on a power-capped A4000; a knee-locked card may lose a few percent at the crossing until Ember re-tunes) and its two orders with their defaults 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:48:45 +00:00
igneum-labs
daba2e69d2 Counter ASIC 3.0 status: the 10:05 move's read-back (36 of 36 on the pin by 10:12, the first lock 10:06:32, carrying resumed; three fault classes named), the tiers gated (d3d0704a), the knob's tip e2962b89, the 0.3.25 app line assembled at 10:29 (e0d4425f), the node line's tip 7bd2940f with the one-shot gossip fix and the drift item reclassified 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:31:37 +00:00
igneum-labs
01a160456e Counter ASIC 3.0 status: the 0.3.25 plan (the shipper's), main's word (the tiers both halves, gated) and the founder's word (0.3.25 today: the app line by 12:30, the pin by 13:00, the move after the 68,400 crossing reads clean, about 13:30), card-in ready, the tiers' UI half merged, the second deploy (a90fde64), the node line placeable ahead of its minute (c6629572), the knob's tip e2962b89, the Hetzner boxes on the founder's word 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:20:01 +00:00
igneum-labs
3f61c91d04 Counter ASIC 3.0 status: the site deployed (cbd29023 at 10:00 BST), row 17's floor sentence and the Arc row restored (85ad7421), the 0.3.25 node build's shape and main's word on the fee ceiling (no second object cut today), class-v5 8f481459 gated with the Intel fix, the per-process gh fix on master (1db6b438), the AMD knob for tonight's cut, the 0.3.24 move FIRED at 10:05:00 BST (the Mac entry and the hive live 10:08:35, the three nodes on the pin, class v5 at 68,400 about 12:54) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:11:58 +00:00
igneum-labs
798c74c4bb release rules 11 extended (never a kill pattern the shell carries; a kept datadir's locks); release 0.3.22 section 23: the move executed, 0.3.24 Mac and HiveOS LIVE 10:08:35 BST, the Windows chain on PC 2 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:11:06 +00:00
igneum-labs
a90fde6446 Merge ca3-coord-record 6ff93415 into master (gate: green on 6ff93415, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 09:07:55 +00:00
igneum-labs
6ff93415fc Counter ASIC 3.0 status: the gh switch owned by the attack-pass lane and the rule to every lane, the Intel fix on master (dabd0fea), the pairs on the pin, class-v5 a55fcc10 and the rebuilt kit (65b47211), the AMD grid prepared, the AP-F8-1 record line (59572d37) and the card-in job, the in-house pass's last landing (96262264), the attack-pass tail (2f1ce097), the signal-15 hook note 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 09:01:49 +00:00
igneum-labs
85ad74213c Merge spec-accept-23 a4efbc46 into master (gate: green on a4efbc46, recorded by tools/ci/pre-push.sh; landed on the box mirror) 2026-10-08 08:59:34 +00:00
igneum-labs
cbd290232b Merge ca3-coord-record 1cf233f6 into master (gate: green on 1cf233f6, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 08:58:40 +00:00
igneum-labs
2f1ce097ca Merge attack-pass 278e75e7 into master (gate: green on 278e75e7, recorded by tools/ci/pre-push.sh; the full gate runs in CI on this merge) 2026-10-08 08:55:29 +00:00
igneum-labs
962622640c Merge crypto-engage 48b5525a into master (gate: green on 48b5525a, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 08:54:42 +00:00
igneum-labs
59572d3796 Merge ca3-v4-amend ed85185a into master (gate: green on ed85185a, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 08:54:04 +00:00
igneum-labs
a4efbc4657 Merge build/master dabd0fea into spec-accept-23 (the Intel rotr worker fix taken; /miners rebuilt from the merged bench data) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:52:59 +00:00
igneum-labs
1cf233f66b Counter ASIC 3.0 status: the move's readings and the pin line on 5b673577, the founder's standing rule (defaults and deadlines; the shipper's clocks mirrored) and its distribution, the Arc self-test read (the shipped workers right, the kit lacking the Intel rotate fix), the AMD knob opened (a002732a), the F8 tail attributed on the Mac (a per-site bucket concentration), the fast-time PASS and the minute 10:05:00 BST, the gh exception on the Mac, adv-accept closed (BOUND, no BREAK) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:51:32 +00:00
igneum-labs
41a1440113 release 0.3.22: THE MINUTE 10:05 BST set in m5b67-1, the pairs and the hive on the pin, what moves at the minute 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:49:41 +00:00
igneum-labs
278e75e74f attack-pass: the status board, F8 row, gate line and re-gate paragraph read the tail as attributed 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:49:21 +00:00
igneum-labs
48b5525a1f in-house-pass.md: adv-accept closed (15.1 box-hours, 796,042 accepted programs, BOUND, no BREAK); the pass's totals final 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:48:22 +00:00
igneum-labs
4d193c31cb attack-pass: AP-F8-1 tail attributed (p4, p8, p10, p34 as per-site bucket concentration at a narrow-window site; the largest-bucket bound named for a next class) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:48:03 +00:00
igneum-labs
ed85185abf ledger AP-F8-1: the four-seed tail attributed (8 October 2026 Mac read at 2^24): one narrow-window site per seed, a quarter-bit bucket concentration from a mad, xor or sub writer, no lossy source, no chip consequence; the per-site bucket bound is a morning item 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:47:41 +00:00
igneum-labs
049cf90901 release 0.3.22 section 22: the morning (main's word, the founder's rule, the fifth cut 5b673577, the move staged, the fleet's reading, the Arc answer) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:46:22 +00:00
igneum-labs
978ea619e6 Counter ASIC 3.0 status: main's morning words (the look landed, the floor sentence to row 17, CA4 parked, the Mac rule lifted for the F8 tail, route (A) at 10:45 BST), the build-server lane's honest state and the deploy and pairs plan, the fourth cut 5b673577 (floor 68,400), the Arc B580's self-test fault on both classes and the Intel kit held out of 0.3.24, the design pass on master (e59b127a) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:40:46 +00:00
igneum-labs
e8da833a16 Evidence row 17: the in-house pass's floor reading beside its sentence (nine of nine hot sets refused; the diffuse era-stride excess, bounded under 0.1 percent of a hash's reads per site, is not caught by the floor and is the next class's test; measured 8 October 2026, ledger AP-F8-1 and AP-F8-6); /evidence rebuilt 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 08:33:21 +00:00
igneum-labs
8e4eb37b2e Counter ASIC 3.0 status: the 9070 XT bench row on master (the AMD watts no longer owed), the hot-table ldcs rows (a dead lever; the hot family latency-bound), the PC 1 list main set closed on measured rows 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 07:13:01 +00:00
igneum-labs
4b603b28f5 Merge ca3-v4-amend 2fa54cc6 into master (gate: green on 2fa54cc6, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 07:11:38 +00:00
igneum-labs
2fa54cc664 bench-log: the hot-table packs on the RTX 5090, ldcs against base, unlocked and at the 1,300 MHz lock (36 rows, all PASS; ldcs dead, the hot packs lose 2 percent to the lock where mx8 loses 7.5) 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 07:05:19 +00:00
igneum-labs
41f90af1b6 Merge ca3-coord-record 14c453b4 into master (gate: green on 14c453b4, recorded by tools/ci/pre-push.sh; landed on the build mirror) 2026-10-08 06:27:36 +00:00
igneum-labs
489f4ff5bd Merge ship-docs-0321 b8e0772d into master (gate: green on b8e0772d, recorded by tools/ci/pre-push.sh; the full gate runs in CI on this merge) 2026-10-08 06:25:18 +00:00
igneum-labs
14c453b4ca Counter ASIC 3.0 status: the 5080 Ember tune (the lock worth +40 percent per watt over the stock climb; the knob's case), the knob on release-0.3.24 (8bb0e288), the 9070 XT tune (no AMD knob in 0.3.20; 18.9 MH/s at 202 W; the AMD core-clock knob a morning item), the hot-table job the last on PC 1 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 06:21:16 +00:00
igneum-labs
b8e0772d1d release 0.3.22 section 21: the core-clock knob cherry-picked onto release-0.3.24 (the union with the efficient-point ceiling), the measured Ember line from PC 1 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 06:19:49 +00:00
igneum-labs
544e4037b5 Chip model 5.11 and new-pow 5.1: the per-MAC figures corrected by the per-lane tile count (32 MACs per tile, not 1,024; the 4090's 0.056 pJ is 1.8 pJ, the 5090 reads 2.9 and 1.5 at the shadow's premium), the tensor-tile column withdrawn (a chip's k on tile work 0.03 to 0.3, below the ALU shadow's); Counter ASIC 3.0 status: the microbench's correction, the night's closing sentence on measured rows, the CA4 file's commits 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 06:08:07 +00:00
igneum-labs
89ef71e6c0 Counter ASIC 3.0 status: the microbench on the 5090 (20 probes at both states; shfl 55.8 pJ, mul 13.9, mulhi 39.6 against the ARX op's 11.3; int8 mma 1.36): the re-weight's re-opening condition not met on measured rows, the tile block's case in measured picojoules 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
 2026-10-08 06:00:29 +00:00