A paid production workload on the reference fleet, measured on GeForce RTX 4080 and GeForce RTX 5090 on the same days. One workload, its own cells — never averaged with any other workload, because cross-workload aggregation is the substitution error this project exists to expose.
| DAY | GOODPUT GeForce RTX 4080 / GeForce RTX 5090 | ATTEMPTS | COMPLETIONS | CARDS |
|---|---|---|---|---|
| 2026-08-21 | 0.37× | 308 / 1312 | 274 / 1264 | 2 / 6 |
| 2026-08-22 | 0.33× | 205 / 1331 | 174 / 1269 | 2 / 6 |
| 2026-08-23 | 0.38× | 406 / 1693 | 380 / 1647 | 2 / 6 |
| 2026-08-24 | 0.59× | 76 / 294 | 69 / 281 | 2 / 6 |
| 2026-08-26 | 0.38× | 393 / 2652 | 375 / 2606 | 2 / 6 |
| 2026-08-27 | 0.36× | 389 / 3042 | 368 / 2969 | 2 / 6 |
| 2026-08-28 | 0.39× | 54 / 136 | 53 / 134 | 2 / 6 |
| 2026-08-29 | 0.40× | 360 / 1580 | 344 / 1559 | 2 / 6 |
| 2026-08-30 | 0.53× | 29 / 211 | 29 / 206 | 2 / 5 |
| 2026-08-31 | 0.34× | 99 / 1600 | 90 / 1578 | 2 / 6 |
| 2026-09-01 | 0.32× | 48 / 338 | 46 / 336 | 2 / 6 |
| 2026-09-02 | 0.41× | 23 / 137 | 21 / 133 | 2 / 6 |
| 2026-09-03 | 0.97× | 754 / 387 | 729 / 378 | 2 / 5 |
| 2026-09-04 | 0.56× | 101 / 136 | 96 / 133 | 2 / 5 |
The measured pair page for these deployments: side-by-side cells. Downloads: ratio cells, value board, the rental prices these were costed at.