GPU model | GV100 | GV100 |
---|---|---|
Architecture | Volta | Volta |
Foundry | TSMC | TSMC |
Manufacturing process | 12 nm | 12 nm |
Number of transistors | 21,100 million | 21,100 million |
Die size | 815 mm² | 815 mm² |
GPU variant | — | GV100-400-A1 |
Release date | Jun 21st, 2017 | Dec 7th, 2017 |
---|---|---|
Generation | Tesla | GeForce 10 |
Production | End-of-life | End-of-life |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Predecessor | — | GeForce 900 |
Successor | — | GeForce 20 |
Launch price | — | 2,999 USD |
Reviews | — | 4 in our database |
Base clock | 1245 MHz | 1200 MHz |
---|---|---|
Boost clock | 1380 MHz | 1455 MHz |
Memory clock | 876 MHz 1752 Mbps effective | 848 MHz 1696 Mbps effective |
Memory size | 16 GB | 12 GB |
---|---|---|
Memory type | HBM2 | HBM2 |
Memory bus width | 4096 bit | 3072 bit |
Bandwidth | 897.0 GB/s | 651.3 GB/s |
Shading units | 5120 | 5120 |
---|---|---|
TMUs | 320 | 320 |
ROPs | 128 | 96 |
SM count | 80 | 80 |
Tensor cores | 640 | 640 |
Cache L1 | 128 KB (per SM) | 96 KB (per SM) |
Cache L2 | 6 MB | 4.5 MB |
Pixel rate | 176.6 GPixel/s | 139.7 GPixel/s |
---|---|---|
Texture rate | 441.6 GTexel/s | 465.6 GTexel/s |
FP16 (half) performance | 28.26 TFLOPS (2:1) | 29.80 TFLOPS (2:1) |
FP32 (float) performance | 14.13 TFLOPS | 14.90 TFLOPS |
FP64 (double) performance | 7.066 TFLOPS (1:2) | 7.450 TFLOPS (1:2) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
TDP | 300 W | 250 W |
Suggested PSU | 700 W | 600 W |
Outputs | No outputs | 1x HDMI3x DisplayPort |
Power connectors | 2x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | PG500 SKU 200 | PG500 SKU 0 |
Length | — | 267 mm 10.5 inches |
Width | — | 112 mm 4.4 inches |
Height | — | 40 mm 1.6 inches |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.0 | 7.0 |
Shader model | 6.6 | 6.6 |