GeForce RTX 2080 Ti
Status and notes
Turing is a separate module, not a build of the GA10x/AD10x one: the booter payload constants, code addresses and teardown differ per architecture. Measured on a 22 GB board with driver 575.51.03, where FP16 accumulate went from 88-98% of peak to 120-161% of peak depending on matrix size. Same-card rows are on the benchmarks page.
Measured throughput
| Mode | Stock | Unlocked | Gain |
|---|---|---|---|
f16/f32 | 50.30 TFLOP/s | 68.30 TFLOP/s | 1.36× |
f16/f32 | 55.70 TFLOP/s | 91.40 TFLOP/s | 1.64× |
f16/f32 | 53.70 TFLOP/s | 81.70 TFLOP/s | 1.52× |
f16/f32 | 52.80 TFLOP/s | 77.90 TFLOP/s | 1.48× |