GPU model | GV100 | TU102 |
---|---|---|
GPU variant | GV100-875-A1 | TU102-400-A1 |
Architecture | Volta | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 12 nm |
Number of transistors | 21,100 million | 18,600 million |
Die size | 815 mm² | 754 mm² |
Release date | Mar 27th, 2018 | Dec 18th, 2018 |
---|---|---|
Generation | Quadro | GeForce 20 |
Production | End-of-life | Active |
Launch price | 8,999 USD | 2,499 USD |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Predecessor | — | GeForce 10 |
Reviews | — | 6 in our database |
Successor | — | GeForce 30 |
Base clock | 1132 MHz | 1350 MHz |
---|---|---|
Boost clock | 1627 MHz | 1770 MHz |
Memory clock | 848 MHz 1696 Mbps effective | 1750 MHz 14 Gbps effective |
Memory size | 32 GB | 24 GB |
---|---|---|
Memory type | HBM2 | GDDR6 |
Memory bus width | 4096 bit | 384 bit |
Bandwidth | 868.4 GB/s | 672.0 GB/s |
Shading units | 5120 | 4608 |
---|---|---|
TMUs | 320 | 288 |
ROPs | 128 | 96 |
SM count | 80 | 72 |
Tensor cores | 640 | 576 |
Cache L1 | 128 KB (per SM) | 64 KB (per SM) |
Cache L2 | 6 MB | 6 MB |
RT cores | — | 72 |
Pixel rate | 208.3 GPixel/s | 169.9 GPixel/s |
---|---|---|
Texture rate | 520.6 GTexel/s | 509.8 GTexel/s |
FP16 (half) performance | 33.32 TFLOPS (2:1) | 32.62 TFLOPS (2:1) |
FP32 (float) performance | 16.66 TFLOPS | 16.31 TFLOPS |
FP64 (double) performance | 8.330 TFLOPS (1:2) | 509.8 GFLOPS (1:32) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
Width | 111 mm 4.4 inches | 116 mm 4.6 inches |
TDP | 250 W | 280 W |
Suggested PSU | 600 W | 600 W |
Outputs | 4x DisplayPort | 1x HDMI3x DisplayPort1x USB Type-C |
Power connectors | 8-pin EPS | 2x 8-pin |
Board number | PG500 SKU 500 | PG150 SKU 0 |
Height | — | 35 mm 1.4 inches |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.0 | 7.5 |
Shader model | 6.6 | 6.6 |