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 | 0.46× | 120 / 390 | 118 / 375 | 2 / 6 |
| 2026-08-22 | 0.44× | 177 / 518 | 169 / 503 | 2 / 6 |
| 2026-08-23 | 0.43× | 131 / 417 | 123 / 402 | 2 / 6 |
| 2026-08-24 | 0.44× | 144 / 58 | 133 / 55 | 2 / 6 |
| 2026-08-26 | 0.42× | 262 / 626 | 239 / 591 | 2 / 6 |
| 2026-08-27 | 0.44× | 330 / 795 | 298 / 750 | 2 / 6 |
| 2026-08-28 | 0.55× | 75 / 18 | 69 / 17 | 2 / 4 |
| 2026-08-29 | 0.40× | 236 / 141 | 215 / 135 | 2 / 6 |
| 2026-08-30 | 0.52× | 9 / 16 | 9 / 16 | 2 / 3 |
| 2026-08-31 | 0.35× | 119 / 33 | 105 / 33 | 2 / 3 |
| 2026-09-01 | 0.34× | 66 / 39 | 61 / 39 | 2 / 3 |
| 2026-09-02 | 0.49× | 18 / 29 | 14 / 27 | 2 / 6 |
| 2026-09-03 | 0.76× | 670 / 169 | 629 / 163 | 2 / 5 |
| 2026-09-04 | 0.51× | 88 / 38 | 86 / 38 | 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.