ECO-05 batch 02 on the RX 7600's class v6 cell (15.79 MH/s measured, the knee 5.4 modelled; the metered knee is batch 03) and the B580's class v6 cell (11.0 measured, watts estimated): one more sustainable world (10R at 0.10, by 1.6 percent on the 7600's entry), 0 of 108 cells pass the envelope, RUN FAIL unchanged; the driver's override hook carried; the cubes and the registry rows recorded at manifest efd03e2c
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
igneum-labs
parent
efd03e2cfe
commit
c2fc307c5d
8 changed files with 5426 additions and 13 deletions
|
|
@ -96,6 +96,34 @@ is 10x a public GPU-mined chain's revenue at the cheapest tariff on the axis; th
|
||||||
(section 5, "sustainable expected" from R upward at 0.03 to 0.25) is contradicted by the driver at both values of R
|
(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.
|
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)
|
## 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 |
|
| Revenue | Tariff | Demand | Sustainable | r (micro-USD per MH/s-h) | GPU hash (TH/s) | Classes with positive entry | Vendors | Cohort above water |
|
||||||
|
|
|
||||||
|
|
@ -37,7 +37,7 @@ instrument behind it (the Arc B580's watts).
|
||||||
| 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 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 |
|
| 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 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 | 6.19 / 8.14 | 13.9 | 6.19 | 269 / 270 / 200 | 40 percent | 472 / 859 / 1697 | 1118 / 1505 / 2343 | 1122 / 1509 / 2347 | 49 pct | 3597 | stock measured 13.88 MH/s at 113 W (PC 1, 15:46 UK, class v4 sub-version 3; class v5 within 2 pct); knee estimated by the 9070 XT's ADLX grid | 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 |
|
| 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 |
|
| 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 |
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -1674,16 +1674,16 @@
|
||||||
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
||||||
"run_status": "RUNNING",
|
"run_status": "RUNNING",
|
||||||
"evidence_path": "sim/economy/coexist/README.md",
|
"evidence_path": "sim/economy/coexist/README.md",
|
||||||
"run_id": "eco05-20261008-01b",
|
"run_id": "eco05-20261008-02",
|
||||||
"updated": "2026-10-08T19:55:03.415Z",
|
"updated": "2026-10-08T19:59:53.932Z",
|
||||||
"evidence_record": {
|
"evidence_record": {
|
||||||
"cell": "model:coexist-fixtures",
|
"cell": "model:coexist-fixtures",
|
||||||
"manifest_sha": "24883757",
|
"manifest_sha": "3f2050ab",
|
||||||
"coverage": {
|
"coverage": {
|
||||||
"ECO-01": "partial: the hand-worked fixtures and the second implementation; the independent reconciliation of inputs and units remains",
|
"ECO-01": "partial: the hand-worked fixtures and the second implementation; the independent reconciliation of inputs and units remains",
|
||||||
"ECO-08": "partial: the package runs from the served files (README, inputs with sources and labels, fixtures); the independent report, the rerun and the perturbations remain"
|
"ECO-08": "partial: the package runs from the served files (README, inputs with sources and labels, fixtures); the independent report, the rerun and the perturbations remain"
|
||||||
},
|
},
|
||||||
"at": "2026-10-08T19:55:03.415Z"
|
"at": "2026-10-08T19:59:53.932Z"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
|
|
@ -1859,15 +1859,15 @@
|
||||||
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
||||||
"run_status": "FAIL",
|
"run_status": "FAIL",
|
||||||
"evidence_path": "docs/analysis/class-v6/eco-05-results.md",
|
"evidence_path": "docs/analysis/class-v6/eco-05-results.md",
|
||||||
"run_id": "eco05-20261008-01b",
|
"run_id": "eco05-20261008-02",
|
||||||
"updated": "2026-10-08T19:55:03.415Z",
|
"updated": "2026-10-08T19:59:53.932Z",
|
||||||
"evidence_record": {
|
"evidence_record": {
|
||||||
"cell": "model:eco05",
|
"cell": "model:eco05",
|
||||||
"manifest_sha": "24883757",
|
"manifest_sha": "3f2050ab",
|
||||||
"coverage": {
|
"coverage": {
|
||||||
"ECO-05": "full on the frozen register's grid and the eight adversarial rows at the team's reconciled specialist rows; the independent reviewer's own adversarial rows (ECO-08) remain"
|
"ECO-05": "full on the frozen register's grid and the eight adversarial rows at the team's reconciled specialist rows; the independent reviewer's own adversarial rows (ECO-08) remain"
|
||||||
},
|
},
|
||||||
"at": "2026-10-08T19:55:03.415Z"
|
"at": "2026-10-08T19:59:53.932Z"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
|
|
@ -1995,16 +1995,16 @@
|
||||||
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
"owner_lane": "research lane (ad6a2bd47d4a46105)",
|
||||||
"run_status": "RUNNING",
|
"run_status": "RUNNING",
|
||||||
"evidence_path": "sim/economy/coexist/README.md",
|
"evidence_path": "sim/economy/coexist/README.md",
|
||||||
"run_id": "eco05-20261008-01b",
|
"run_id": "eco05-20261008-02",
|
||||||
"updated": "2026-10-08T19:55:03.415Z",
|
"updated": "2026-10-08T19:59:53.932Z",
|
||||||
"evidence_record": {
|
"evidence_record": {
|
||||||
"cell": "model:coexist-fixtures",
|
"cell": "model:coexist-fixtures",
|
||||||
"manifest_sha": "24883757",
|
"manifest_sha": "3f2050ab",
|
||||||
"coverage": {
|
"coverage": {
|
||||||
"ECO-01": "partial: the hand-worked fixtures and the second implementation; the independent reconciliation of inputs and units remains",
|
"ECO-01": "partial: the hand-worked fixtures and the second implementation; the independent reconciliation of inputs and units remains",
|
||||||
"ECO-08": "partial: the package runs from the served files (README, inputs with sources and labels, fixtures); the independent report, the rerun and the perturbations remain"
|
"ECO-08": "partial: the package runs from the served files (README, inputs with sources and labels, fixtures); the independent report, the rerun and the perturbations remain"
|
||||||
},
|
},
|
||||||
"at": "2026-10-08T19:55:03.415Z"
|
"at": "2026-10-08T19:59:53.932Z"
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
]
|
]
|
||||||
|
|
|
||||||
|
|
@ -37,6 +37,12 @@ POP = [
|
||||||
("H100 (hosted)", 80, 2.00, 90.0, 25000, 25000, 18000, 0.50, 3.0, 2000, "NVIDIA", False),
|
("H100 (hosted)", 80, 2.00, 90.0, 25000, 25000, 18000, 0.50, 3.0, 2000, "NVIDIA", False),
|
||||||
("A100 (used)", 80, 2.89, 60.0, 10000, 10000, 6000, 0.35, 5.0, 2000, "NVIDIA", False),
|
("A100 (used)", 80, 2.89, 60.0, 10000, 10000, 6000, 0.35, 5.0, 2000, "NVIDIA", False),
|
||||||
]
|
]
|
||||||
|
# per-class overrides for a re-run on measured rows: ECO05_OVERRIDES='{"RX 7600 8 GB": {"uJ": 5.4, "mhs": 15.79, "label": "..."}}'
|
||||||
|
import os as _os
|
||||||
|
_ov = json.loads(_os.environ.get("ECO05_OVERRIDES", "{}"))
|
||||||
|
if _ov:
|
||||||
|
POP = [tuple([c[0], c[1], _ov.get(c[0], {}).get("uJ", c[2]), _ov.get(c[0], {}).get("mhs", c[3])] + list(c[4:])) for c in POP]
|
||||||
|
print("# overrides:", json.dumps(_ov))
|
||||||
E_LOCK = 2.33
|
E_LOCK = 2.33
|
||||||
SPEC = {
|
SPEC = {
|
||||||
"die": dict(uj=E_LOCK / 2.3, uj_ahead=E_LOCK / 3.1, usd=1.0, lo=0.5, hi=1.6, label="N2 SRAM die, reconciled USD 1.0 (0.5 to 1.6), 2.3x / 3.1x"),
|
"die": dict(uj=E_LOCK / 2.3, uj_ahead=E_LOCK / 3.1, usd=1.0, lo=0.5, hi=1.6, label="N2 SRAM die, reconciled USD 1.0 (0.5 to 1.6), 2.3x / 3.1x"),
|
||||||
|
|
|
||||||
File diff suppressed because it is too large
Load diff
File diff suppressed because it is too large
Load diff
|
|
@ -0,0 +1,182 @@
|
||||||
|
# overrides: {"RX 7600 8 GB": {"uJ": 5.4, "mhs": 15.79}, "Intel Arc B580": {"uJ": 10.0, "mhs": 11.0}}
|
||||||
|
SUSTAIN revenue tariff demand sustainable r_uUSD gpu_TH n_entry n_vendors op_ok_pct
|
||||||
|
SUSTAIN 0.25R 0.03 none 0 281 3.21 7 1 95
|
||||||
|
SUSTAIN 0.25R 0.03 normal 0 281 3.21 9 1 95
|
||||||
|
SUSTAIN 0.25R 0.03 spiking 0 281 3.21 13 1 95
|
||||||
|
SUSTAIN 0.25R 0.1 none 0 418 2.16 7 1 45
|
||||||
|
SUSTAIN 0.25R 0.1 normal 0 418 2.16 9 1 91
|
||||||
|
SUSTAIN 0.25R 0.1 spiking 0 418 2.16 13 1 91
|
||||||
|
SUSTAIN 0.25R 0.25 none 0 660 1.37 6 1 39
|
||||||
|
SUSTAIN 0.25R 0.25 normal 0 660 1.37 9 1 45
|
||||||
|
SUSTAIN 0.25R 0.25 spiking 0 660 1.37 13 1 91
|
||||||
|
SUSTAIN 0.25R 0.4 none 0 899 1.00 4 1 36
|
||||||
|
SUSTAIN 0.25R 0.4 normal 0 899 1.00 7 1 41
|
||||||
|
SUSTAIN 0.25R 0.4 spiking 0 899 1.00 13 1 91
|
||||||
|
SUSTAIN R 0.03 none 0 401 9.00 8 1 95
|
||||||
|
SUSTAIN R 0.03 normal 0 401 9.00 10 1 95
|
||||||
|
SUSTAIN R 0.03 spiking 0 401 9.00 15 1 95
|
||||||
|
SUSTAIN R 0.1 none 0 535 6.75 8 1 68
|
||||||
|
SUSTAIN R 0.1 normal 0 535 6.75 10 1 91
|
||||||
|
SUSTAIN R 0.1 spiking 0 535 6.75 15 1 91
|
||||||
|
SUSTAIN R 0.25 none 0 749 5.10 8 1 45
|
||||||
|
SUSTAIN R 0.25 normal 0 749 5.10 10 1 68
|
||||||
|
SUSTAIN R 0.25 spiking 0 749 5.10 15 1 91
|
||||||
|
SUSTAIN R 0.4 none 0 1014 4.70 8 1 45
|
||||||
|
SUSTAIN R 0.4 normal 0 1014 4.70 10 1 45
|
||||||
|
SUSTAIN R 0.4 spiking 0 1014 4.70 15 1 91
|
||||||
|
SUSTAIN 4R 0.03 none 0 684 21.11 12 1 99
|
||||||
|
SUSTAIN 4R 0.03 normal 0 684 21.11 15 1 99
|
||||||
|
SUSTAIN 4R 0.03 spiking 0 684 21.11 15 1 99
|
||||||
|
SUSTAIN 4R 0.1 none 0 820 17.61 11 1 95
|
||||||
|
SUSTAIN 4R 0.1 normal 0 820 17.61 14 1 95
|
||||||
|
SUSTAIN 4R 0.1 spiking 0 820 17.61 15 1 95
|
||||||
|
SUSTAIN 4R 0.25 none 0 1049 13.76 11 1 57
|
||||||
|
SUSTAIN 4R 0.25 normal 0 1049 13.76 12 1 68
|
||||||
|
SUSTAIN 4R 0.25 spiking 0 1049 13.76 15 1 91
|
||||||
|
SUSTAIN 4R 0.4 none 0 1315 10.99 11 1 46
|
||||||
|
SUSTAIN 4R 0.4 normal 0 1315 10.99 12 1 58
|
||||||
|
SUSTAIN 4R 0.4 spiking 0 1315 10.99 15 1 92
|
||||||
|
SUSTAIN 10R 0.03 none 1 948 38.56 16 2 100
|
||||||
|
SUSTAIN 10R 0.03 normal 1 948 38.56 16 2 100
|
||||||
|
SUSTAIN 10R 0.03 spiking 1 948 38.56 16 2 100
|
||||||
|
SUSTAIN 10R 0.1 none 1 1225 29.47 14 2 100
|
||||||
|
SUSTAIN 10R 0.1 normal 1 1225 29.47 16 2 100
|
||||||
|
SUSTAIN 10R 0.1 spiking 1 1225 29.47 16 2 100
|
||||||
|
SUSTAIN 10R 0.25 none 0 1599 23.48 13 1 97
|
||||||
|
SUSTAIN 10R 0.25 normal 0 1599 23.48 14 1 97
|
||||||
|
SUSTAIN 10R 0.25 spiking 0 1599 23.48 15 1 97
|
||||||
|
SUSTAIN 10R 0.4 none 0 1982 18.21 12 1 69
|
||||||
|
SUSTAIN 10R 0.4 normal 0 1982 18.21 13 1 69
|
||||||
|
SUSTAIN 10R 0.4 spiking 0 1982 18.21 15 1 92
|
||||||
|
SUMMARY spec sustainable_worlds pass_all pass_median pass_p90 pass_core median_min median_max core_max_max
|
||||||
|
SUMMARY die 108 0 12 0 9 0.95 11.74 23.59
|
||||||
|
SUMMARY hybrid 108 0 24 0 12 0.65 6.75 13.55
|
||||||
|
SUMMARY board 108 0 39 0 24 0.48 4.41 8.86
|
||||||
|
BOUNDARY spec revenue tariff_0.03 tariff_0.10 tariff_0.25 tariff_0.40 (median ratio; life 3, dev zero, private, normal demand; * = sustainable)
|
||||||
|
BOUNDARY die 0.25R 9.80 7.34 5.80 5.26 | consumer-16 median 8.80 / 7.10 / 5.55 / 4.88 | best core 5.14 / 3.87 / 2.98 / 2.67
|
||||||
|
BOUNDARY die R 9.80 7.34 5.80 5.26 | consumer-16 median 8.80 / 7.10 / 5.55 / 4.88 | best core 5.14 / 3.87 / 2.98 / 2.67
|
||||||
|
BOUNDARY die 4R 9.80 7.34 5.80 5.26 | consumer-16 median 8.80 / 7.10 / 5.55 / 4.88 | best core 5.14 / 3.87 / 2.98 / 2.67
|
||||||
|
BOUNDARY die 10R 9.80* 7.34* 5.80 5.26 | consumer-16 median 8.80 / 7.10 / 5.55 / 4.88 | best core 5.14 / 3.87 / 2.98 / 2.67
|
||||||
|
BOUNDARY hybrid 0.25R 5.12 4.59 4.13 3.94 | consumer-16 median 4.60 / 4.43 / 3.95 / 3.66 | best core 2.68 / 2.42 / 2.12 / 2.00
|
||||||
|
BOUNDARY hybrid R 5.12 4.59 4.13 3.94 | consumer-16 median 4.60 / 4.43 / 3.95 / 3.66 | best core 2.68 / 2.42 / 2.12 / 2.00
|
||||||
|
BOUNDARY hybrid 4R 5.12 4.59 4.13 3.94 | consumer-16 median 4.60 / 4.43 / 3.95 / 3.66 | best core 2.68 / 2.42 / 2.12 / 2.00
|
||||||
|
BOUNDARY hybrid 10R 5.12* 4.59* 4.13 3.94 | consumer-16 median 4.60 / 4.43 / 3.95 / 3.66 | best core 2.68 / 2.42 / 2.12 / 2.00
|
||||||
|
BOUNDARY board 0.25R 3.24 3.09 2.95 2.88 | consumer-16 median 2.91 / 2.99 / 2.82 / 2.67 | best core 1.70 / 1.63 / 1.51 / 1.46
|
||||||
|
BOUNDARY board R 3.24 3.09 2.95 2.88 | consumer-16 median 2.91 / 2.99 / 2.82 / 2.67 | best core 1.70 / 1.63 / 1.51 / 1.46
|
||||||
|
BOUNDARY board 4R 3.24 3.09 2.95 2.88 | consumer-16 median 2.91 / 2.99 / 2.82 / 2.67 | best core 1.70 / 1.63 / 1.51 / 1.46
|
||||||
|
BOUNDARY board 10R 3.24* 3.09* 2.95 2.88 | consumer-16 median 2.91 / 2.99 / 2.82 / 2.67 | best core 1.70 / 1.63 / 1.51 / 1.46
|
||||||
|
HETERO spec revenue gpu_tariff median p90 core_max (specialist at 0.03; life 3, dev zero, private, normal)
|
||||||
|
HETERO die 0.25R 0.03 9.80 49.65 19.69
|
||||||
|
HETERO die 0.25R 0.1 13.00 50.94 23.12
|
||||||
|
HETERO die 0.25R 0.25 19.84 53.71 36.06
|
||||||
|
HETERO die 0.25R 0.4 26.68 58.61 52.35
|
||||||
|
HETERO die R 0.03 9.80 49.65 19.69
|
||||||
|
HETERO die R 0.1 13.00 50.94 23.12
|
||||||
|
HETERO die R 0.25 19.84 53.71 36.06
|
||||||
|
HETERO die R 0.4 26.68 58.61 52.35
|
||||||
|
HETERO die 4R 0.03 9.80 49.65 19.69
|
||||||
|
HETERO die 4R 0.1 13.00 50.94 23.12
|
||||||
|
HETERO die 4R 0.25 19.84 53.71 36.06
|
||||||
|
HETERO die 4R 0.4 26.68 58.61 52.35
|
||||||
|
HETERO die 10R 0.03 9.80 49.65 19.69
|
||||||
|
HETERO die 10R 0.1 13.00 50.94 23.12
|
||||||
|
HETERO die 10R 0.25 19.84 53.71 36.06
|
||||||
|
HETERO die 10R 0.4 26.68 58.61 52.35
|
||||||
|
HETERO hybrid 0.25R 0.03 5.12 25.92 10.28
|
||||||
|
HETERO hybrid 0.25R 0.1 6.78 26.59 12.07
|
||||||
|
HETERO hybrid 0.25R 0.25 10.36 28.04 18.83
|
||||||
|
HETERO hybrid 0.25R 0.4 13.93 30.60 27.33
|
||||||
|
HETERO hybrid R 0.03 5.12 25.92 10.28
|
||||||
|
HETERO hybrid R 0.1 6.78 26.59 12.07
|
||||||
|
HETERO hybrid R 0.25 10.36 28.04 18.83
|
||||||
|
HETERO hybrid R 0.4 13.93 30.60 27.33
|
||||||
|
HETERO hybrid 4R 0.03 5.12 25.92 10.28
|
||||||
|
HETERO hybrid 4R 0.1 6.78 26.59 12.07
|
||||||
|
HETERO hybrid 4R 0.25 10.36 28.04 18.83
|
||||||
|
HETERO hybrid 4R 0.4 13.93 30.60 27.33
|
||||||
|
HETERO hybrid 10R 0.03 5.12 25.92 10.28
|
||||||
|
HETERO hybrid 10R 0.1 6.78 26.59 12.07
|
||||||
|
HETERO hybrid 10R 0.25 10.36 28.04 18.83
|
||||||
|
HETERO hybrid 10R 0.4 13.93 30.60 27.33
|
||||||
|
HETERO board 0.25R 0.03 3.24 16.43 6.52
|
||||||
|
HETERO board 0.25R 0.1 4.30 16.86 7.65
|
||||||
|
HETERO board 0.25R 0.25 6.57 17.78 11.93
|
||||||
|
HETERO board 0.25R 0.4 8.83 19.40 17.33
|
||||||
|
HETERO board R 0.03 3.24 16.43 6.52
|
||||||
|
HETERO board R 0.1 4.30 16.86 7.65
|
||||||
|
HETERO board R 0.25 6.57 17.78 11.93
|
||||||
|
HETERO board R 0.4 8.83 19.40 17.33
|
||||||
|
HETERO board 4R 0.03 3.24 16.43 6.52
|
||||||
|
HETERO board 4R 0.1 4.30 16.86 7.65
|
||||||
|
HETERO board 4R 0.25 6.57 17.78 11.93
|
||||||
|
HETERO board 4R 0.4 8.83 19.40 17.33
|
||||||
|
HETERO board 10R 0.03 3.24 16.43 6.52
|
||||||
|
HETERO board 10R 0.1 4.30 16.86 7.65
|
||||||
|
HETERO board 10R 0.25 6.57 17.78 11.93
|
||||||
|
HETERO board 10R 0.4 8.83 19.40 17.33
|
||||||
|
ADV combination spec revenue median p90 core_max sustainable
|
||||||
|
ADV 1 die at USD 0.5, a node ahead, dev zero, private, GPU 0.40 vs specialist 0.03 die R 42.14 92.58 82.69 0
|
||||||
|
ADV 2 die taking the whole chain (dev USD 75 M over 100 pct), private, 0.10 die R 2.05 8.02 3.64 0
|
||||||
|
ADV 3 hardware sales at zero development, die, 0.10 die R 5.85 22.95 10.41 0
|
||||||
|
ADV 4 GPU entrant at street, die at 1.0, 0.10 die R 7.10 28.79 12.31 0
|
||||||
|
ADV 5 GPU generation 1.5x vs die a node ahead, 0.10 die R 7.90 35.21 15.03 0
|
||||||
|
ADV 6 spiking demand at the measured 5.5 pct proving efficiency, die, 0.10 die R (run with argv[1] = 0.055)
|
||||||
|
ADV 7 die at USD 1.6 (the dear end), 0.10 die R 6.37 24.97 11.33 0
|
||||||
|
ADV 8 board at street GPU prices, 0.25 board R 2.90 7.98 5.45 0
|
||||||
|
ADV 1 die at USD 0.5, a node ahead, dev zero, private, GPU 0.40 vs specialist 0.03 die 4R 42.14 92.58 82.69 0
|
||||||
|
ADV 2 die taking the whole chain (dev USD 75 M over 100 pct), private, 0.10 die 4R 3.68 14.44 6.55 0
|
||||||
|
ADV 3 hardware sales at zero development, die, 0.10 die 4R 5.85 22.95 10.41 0
|
||||||
|
ADV 4 GPU entrant at street, die at 1.0, 0.10 die 4R 7.10 28.79 12.31 0
|
||||||
|
ADV 5 GPU generation 1.5x vs die a node ahead, 0.10 die 4R 7.90 35.21 15.03 0
|
||||||
|
ADV 6 spiking demand at the measured 5.5 pct proving efficiency, die, 0.10 die 4R (run with argv[1] = 0.055)
|
||||||
|
ADV 7 die at USD 1.6 (the dear end), 0.10 die 4R 6.37 24.97 11.33 0
|
||||||
|
ADV 8 board at street GPU prices, 0.25 board 4R 2.90 7.98 5.45 0
|
||||||
|
EXPLAIN spec revenue tariff worst_core_class ratio gpu_entrant_uUSD spec_uUSD spec_hw spec_pw median_class median_gpu_uUSD best_core proving_income
|
||||||
|
EXPLAIN die 0.25R 0.03 RTX 3090 (used) 19.69 1889 96 43 52 RTX 3080 (used) 940 best core RTX 5090 5.14 (493) proving income best core 604 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 0.25R 0.1 RTX 3090 (used) 13.06 2217 170 43 125 RTX 3080 (used) 1246 best core RTX 5080 3.87 (656) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 0.25R 0.25 Intel Arc B580 10.54 3458 328 43 282 RTX 3080 (used) 1902 best core RTX 5080 2.98 (978) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 0.25R 0.4 Intel Arc B580 10.33 5020 486 43 439 RTX 3080 (used) 2558 best core RTX 5080 2.67 (1300) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN die R 0.03 RTX 3090 (used) 19.69 1889 96 43 52 RTX 3080 (used) 940 best core RTX 5090 5.14 (493) proving income best core 1965 uUSD/MH/s-h
|
||||||
|
EXPLAIN die R 0.1 RTX 3090 (used) 13.06 2217 170 43 125 RTX 3080 (used) 1246 best core RTX 5080 3.87 (656) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN die R 0.25 Intel Arc B580 10.54 3458 328 43 282 RTX 3080 (used) 1902 best core RTX 5080 2.98 (978) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN die R 0.4 Intel Arc B580 10.33 5020 486 43 439 RTX 3080 (used) 2558 best core RTX 5080 2.67 (1300) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 4R 0.03 RTX 3090 (used) 19.69 1889 96 43 52 RTX 3080 (used) 940 best core RTX 5090 5.14 (493) proving income best core 7406 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 4R 0.1 RTX 3090 (used) 13.06 2217 170 43 125 RTX 3080 (used) 1246 best core RTX 5080 3.87 (656) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 4R 0.25 Intel Arc B580 10.54 3458 328 43 282 RTX 3080 (used) 1902 best core RTX 5080 2.98 (978) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 4R 0.4 Intel Arc B580 10.33 5020 486 43 439 RTX 3080 (used) 2558 best core RTX 5080 2.67 (1300) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 10R 0.03 RTX 3090 (used) 19.69 1889 96 43 52 RTX 3080 (used) 940 best core RTX 5090 5.14 (493) proving income best core 18289 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 10R 0.1 RTX 3090 (used) 13.06 2217 170 43 125 RTX 3080 (used) 1246 best core RTX 5080 3.87 (656) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 10R 0.25 Intel Arc B580 10.54 3458 328 43 282 RTX 3080 (used) 1902 best core RTX 5080 2.98 (978) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN die 10R 0.4 Intel Arc B580 10.33 5020 486 43 439 RTX 3080 (used) 2558 best core RTX 5080 2.67 (1300) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 0.25R 0.03 RTX 3090 (used) 10.28 1889 184 120 62 RTX 3080 (used) 940 best core RTX 5090 2.68 (493) proving income best core 604 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 0.25R 0.1 RTX 3090 (used) 8.16 2217 272 120 149 RTX 3080 (used) 1246 best core RTX 5080 2.42 (656) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 0.25R 0.25 Intel Arc B580 7.51 3458 460 120 336 RTX 3080 (used) 1902 best core RTX 5080 2.12 (978) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 0.25R 0.4 Intel Arc B580 7.74 5020 649 120 523 RTX 3080 (used) 2558 best core RTX 5080 2.00 (1300) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid R 0.03 RTX 3090 (used) 10.28 1889 184 120 62 RTX 3080 (used) 940 best core RTX 5090 2.68 (493) proving income best core 1965 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid R 0.1 RTX 3090 (used) 8.16 2217 272 120 149 RTX 3080 (used) 1246 best core RTX 5080 2.42 (656) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid R 0.25 Intel Arc B580 7.51 3458 460 120 336 RTX 3080 (used) 1902 best core RTX 5080 2.12 (978) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid R 0.4 Intel Arc B580 7.74 5020 649 120 523 RTX 3080 (used) 2558 best core RTX 5080 2.00 (1300) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 4R 0.03 RTX 3090 (used) 10.28 1889 184 120 62 RTX 3080 (used) 940 best core RTX 5090 2.68 (493) proving income best core 7406 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 4R 0.1 RTX 3090 (used) 8.16 2217 272 120 149 RTX 3080 (used) 1246 best core RTX 5080 2.42 (656) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 4R 0.25 Intel Arc B580 7.51 3458 460 120 336 RTX 3080 (used) 1902 best core RTX 5080 2.12 (978) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 4R 0.4 Intel Arc B580 7.74 5020 649 120 523 RTX 3080 (used) 2558 best core RTX 5080 2.00 (1300) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 10R 0.03 RTX 3090 (used) 10.28 1889 184 120 62 RTX 3080 (used) 940 best core RTX 5090 2.68 (493) proving income best core 18289 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 10R 0.1 RTX 3090 (used) 8.16 2217 272 120 149 RTX 3080 (used) 1246 best core RTX 5080 2.42 (656) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 10R 0.25 Intel Arc B580 7.51 3458 460 120 336 RTX 3080 (used) 1902 best core RTX 5080 2.12 (978) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN hybrid 10R 0.4 Intel Arc B580 7.74 5020 649 120 523 RTX 3080 (used) 2558 best core RTX 5080 2.00 (1300) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 0.25R 0.03 RTX 3090 (used) 6.52 1889 290 207 80 RTX 3080 (used) 940 best core RTX 5090 1.70 (493) proving income best core 604 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 0.25R 0.1 RTX 3090 (used) 5.50 2217 403 207 192 RTX 3080 (used) 1246 best core RTX 5080 1.63 (656) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 0.25R 0.25 Intel Arc B580 5.36 3458 646 207 432 RTX 3080 (used) 1902 best core RTX 5080 1.51 (978) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 0.25R 0.4 Intel Arc B580 5.65 5020 888 207 673 RTX 3080 (used) 2558 best core RTX 5080 1.46 (1300) proving income best core 901 uUSD/MH/s-h
|
||||||
|
EXPLAIN board R 0.03 RTX 3090 (used) 6.52 1889 290 207 80 RTX 3080 (used) 940 best core RTX 5090 1.70 (493) proving income best core 1965 uUSD/MH/s-h
|
||||||
|
EXPLAIN board R 0.1 RTX 3090 (used) 5.50 2217 403 207 192 RTX 3080 (used) 1246 best core RTX 5080 1.63 (656) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN board R 0.25 Intel Arc B580 5.36 3458 646 207 432 RTX 3080 (used) 1902 best core RTX 5080 1.51 (978) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN board R 0.4 Intel Arc B580 5.65 5020 888 207 673 RTX 3080 (used) 2558 best core RTX 5080 1.46 (1300) proving income best core 2929 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 4R 0.03 RTX 3090 (used) 6.52 1889 290 207 80 RTX 3080 (used) 940 best core RTX 5090 1.70 (493) proving income best core 7406 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 4R 0.1 RTX 3090 (used) 5.50 2217 403 207 192 RTX 3080 (used) 1246 best core RTX 5080 1.63 (656) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 4R 0.25 Intel Arc B580 5.36 3458 646 207 432 RTX 3080 (used) 1902 best core RTX 5080 1.51 (978) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 4R 0.4 Intel Arc B580 5.65 5020 888 207 673 RTX 3080 (used) 2558 best core RTX 5080 1.46 (1300) proving income best core 11042 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 10R 0.03 RTX 3090 (used) 6.52 1889 290 207 80 RTX 3080 (used) 940 best core RTX 5090 1.70 (493) proving income best core 18289 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 10R 0.1 RTX 3090 (used) 5.50 2217 403 207 192 RTX 3080 (used) 1246 best core RTX 5080 1.63 (656) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 10R 0.25 Intel Arc B580 5.36 3458 646 207 432 RTX 3080 (used) 1902 best core RTX 5080 1.51 (978) proving income best core 27268 uUSD/MH/s-h
|
||||||
|
EXPLAIN board 10R 0.4 Intel Arc B580 5.65 5020 888 207 673 RTX 3080 (used) 2558 best core RTX 5080 1.46 (1300) proving income best core 27268 uUSD/MH/s-h
|
||||||
11
tools/ci/batches/eco05-20261008-02.json
Normal file
11
tools/ci/batches/eco05-20261008-02.json
Normal file
|
|
@ -0,0 +1,11 @@
|
||||||
|
{
|
||||||
|
"run_id": "eco05-20261008-02",
|
||||||
|
"manifest_sha": "3f2050ab",
|
||||||
|
"cut_tip": "3f2050ab (master at the landing of batch 01b; the driver sim/economy/coexist/eco05.py with its per-class override hook travels in this batch)",
|
||||||
|
"evidence_dir": "docs/analysis/class-v6",
|
||||||
|
"boxes": ["build-4"],
|
||||||
|
"cells": [
|
||||||
|
{"cell": "model:eco05", "status": "FAIL", "evidence": "docs/analysis/class-v6/eco-05-results.md"},
|
||||||
|
{"cell": "model:coexist-fixtures", "status": "RUNNING", "evidence": "sim/economy/coexist/README.md"}
|
||||||
|
]
|
||||||
|
}
|
||||||
Loading…
Reference in a new issue