GPU model | GA104 | GV100 |
---|---|---|
GPU variant | GA104-200-A1 | GV100-400-A1 |
Architecture | Ampere | Volta |
Foundry | Samsung | TSMC |
Manufacturing process | 8 nm | 12 nm |
Number of transistors | 17,400 million | 21,100 million |
Die size | 392 mm² | 815 mm² |
Release date | Dec 1st, 2020 | Dec 7th, 2017 |
---|---|---|
gpu.details.availability | Dec 2nd, 2020 | — |
Generation | GeForce 30 | GeForce 10 |
Predecessor | GeForce 20 | GeForce 900 |
Production | Active | End-of-life |
Launch price | 399 USD | 2,999 USD |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Reviews | 64 in our database | 4 in our database |
Successor | — | GeForce 20 |
Base clock | 1410 MHz | 1200 MHz |
---|---|---|
Boost clock | 1665 MHz | 1455 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 848 MHz 1696 Mbps effective |
Memory size | 8 GB | 12 GB |
---|---|---|
Memory type | GDDR6 | HBM2 |
Memory bus width | 256 bit | 3072 bit |
Bandwidth | 448.0 GB/s | 651.3 GB/s |
Shading units | 4864 | 5120 |
---|---|---|
TMUs | 152 | 320 |
ROPs | 80 | 96 |
SM count | 38 | 80 |
Tensor cores | 152 | 640 |
RT cores | 38 | — |
Cache L1 | 128 KB (per SM) | 96 KB (per SM) |
Cache L2 | 4 MB | 4.5 MB |
Pixel rate | 133.2 GPixel/s | 139.7 GPixel/s |
---|---|---|
Texture rate | 253.1 GTexel/s | 465.6 GTexel/s |
FP16 (half) performance | 16.20 TFLOPS (1:1) | 29.80 TFLOPS (2:1) |
FP32 (float) performance | 16.20 TFLOPS | 14.90 TFLOPS |
FP64 (double) performance | 253.1 GFLOPS (1:64) | 7.450 TFLOPS (1:2) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 242 mm 9.5 inches | 267 mm 10.5 inches |
Width | 112 mm 4.4 inches | 112 mm 4.4 inches |
TDP | 200 W | 250 W |
Suggested PSU | 550 W | 600 W |
Outputs | 1x HDMI3x DisplayPort | 1x HDMI3x DisplayPort |
Power connectors | 1x 12-pin | 1x 6-pin + 1x 8-pin |
Board number | PG142 SKU 20 | PG500 SKU 0 |
Height | — | 40 mm 1.6 inches |
DirectX | 12 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 8.6 | 7.0 |
Shader model | 6.6 | 6.6 |