GPU model | Vega 10 | GV100 |
---|---|---|
GPU variant | Vega 10 XL (215-0894216) | — |
Architecture | GCN 5.0 | Volta |
Foundry | GlobalFoundries | TSMC |
Manufacturing process | 14 nm | 12 nm |
Number of transistors | 12,500 million | 21,100 million |
Die size | 495 mm² | 815 mm² |
Release date | Aug 14th, 2017 | — |
---|---|---|
Generation | Radeon Pro Mac (Vega Series) | — |
Production | Active | — |
Interface | PCIe 3.0 x16 | — |
Base clock | 1138 MHz | 1245 MHz |
---|---|---|
Boost clock | 1250 MHz | 1380 MHz |
Memory clock | 786 MHz 1572 Mbps effective | 876 MHz 1752 Mbps effective |
Memory size | 8 GB | 16 GB |
---|---|---|
Memory type | HBM2 | HBM2 |
Memory bus width | 2048 bit | 4096 bit |
Bandwidth | 402.4 GB/s | 897.0 GB/s |
Shading units | 3584 | 5120 |
---|---|---|
TMUs | 224 | 320 |
ROPs | 64 | 128 |
Compute units | 56 | — |
Cache L1 | 16 KB (per CU) | 128 KB (per SM) |
Cache L2 | 4 MB | 6 MB |
SM count | — | 80 |
Tensor cores | — | 640 |
Pixel rate | 80.00 GPixel/s | 176.6 GPixel/s |
---|---|---|
Texture rate | 280.0 GTexel/s | 441.6 GTexel/s |
FP16 (half) performance | 17.92 TFLOPS (2:1) | 28.26 TFLOPS (2:1) |
FP32 (float) performance | 8.960 TFLOPS | 14.13 TFLOPS |
FP64 (double) performance | 560.0 GFLOPS (1:16) | 7.066 TFLOPS (1:2) |
Slot width | IGP | Dual-slot |
---|---|---|
TDP | 210 W | 300 W |
Outputs | 1x HDMI3x DisplayPort | No outputs |
Power connectors | None | 2x 8-pin |
Suggested PSU | — | 700 W |
Board number | — | PG500 SKU 200 |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 7.0 |
Release date | — | Jun 21st, 2017 |
---|---|---|
Generation | — | Tesla |
Production | — | End-of-life |
Interface | — | PCIe 3.0 x16 |