GPU model | TU102 | GV100 |
---|---|---|
GPU variant | TU102-875-A1 | — |
Architecture | Turing | Volta |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 12 nm |
Number of transistors | 18,600 million | 21,100 million |
Die size | 754 mm² | 815 mm² |
Release date | Aug 13th, 2018 | Mar 27th, 2018 |
---|---|---|
Generation | Quadro | Tesla |
Production | Active | End-of-life |
Launch price | 6,299 USD | — |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Base clock | 1440 MHz | 1290 MHz |
---|---|---|
Boost clock | 1770 MHz | 1530 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 876 MHz 1752 Mbps effective |
Memory size | 24 GB | 32 GB |
---|---|---|
Memory type | GDDR6 | HBM2 |
Memory bus width | 384 bit | 4096 bit |
Bandwidth | 672.0 GB/s | 897.0 GB/s |
Shading units | 4608 | 5120 |
---|---|---|
TMUs | 288 | 320 |
ROPs | 96 | 128 |
SM count | 72 | 80 |
Tensor cores | 576 | 640 |
RT cores | 72 | — |
Cache L1 | 64 KB (per SM) | 128 KB (per SM) |
Cache L2 | 6 MB | 6 MB |
Pixel rate | 169.9 GPixel/s | 195.8 GPixel/s |
---|---|---|
Texture rate | 509.8 GTexel/s | 489.6 GTexel/s |
FP16 (half) performance | 32.62 TFLOPS (2:1) | 31.33 TFLOPS (2:1) |
FP32 (float) performance | 16.31 TFLOPS | 15.67 TFLOPS |
FP64 (double) performance | 509.8 GFLOPS (1:32) | 7.834 TFLOPS (1:2) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | — |
Width | 111 mm 4.4 inches | — |
TDP | 260 W | 250 W |
Suggested PSU | 600 W | 600 W |
Outputs | 4x DisplayPort1x USB Type-C | No outputs |
Power connectors | 1x 6-pin + 1x 8-pin | None |
Board number | PG150 SKU 510 | — |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 7.0 |
Shader model | 6.6 | 6.6 |