GPU model | GA102 | GA102 |
---|---|---|
GPU variant | GA102-220-A1 | GA102-350-A1 |
Architecture | Ampere | Ampere |
Foundry | Samsung | Samsung |
Manufacturing process | 8 nm | 8 nm |
Number of transistors | 28,300 million | 28,300 million |
Die size | 628 mm² | 628 mm² |
Release date | Jan 11th, 2022 | Unknown |
---|---|---|
gpu.details.availability | Jan, 2022 | 2022 |
Generation | GeForce 30 | GeForce 30 |
Predecessor | GeForce 20 | GeForce 20 |
Production | Active | Unreleased |
Launch price | 799 USD | 1,699 USD |
Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
Reviews | — | 55 in our database |
Base clock | 1260 MHz | 1560 MHz |
---|---|---|
Boost clock | 1710 MHz | 1860 MHz |
Memory clock | 1188 MHz 19 Gbps effective | 1325 MHz 21.2 Gbps effective |
Memory size | 12 GB | 24 GB |
---|---|---|
Memory type | GDDR6X | GDDR6X |
Memory bus width | 384 bit | 384 bit |
Bandwidth | 912.4 GB/s | 1,018 GB/s |
Shading units | 8960 | 10752 |
---|---|---|
TMUs | 280 | 336 |
ROPs | 112 | 112 |
SM count | 70 | 84 |
Tensor cores | 280 | 336 |
RT cores | 70 | 84 |
Cache L1 | 128 KB (per SM) | 128 KB (per SM) |
Cache L2 | 5 MB | 6 MB |
Pixel rate | 191.5 GPixel/s | 208.3 GPixel/s |
---|---|---|
Texture rate | 478.8 GTexel/s | 625.0 GTexel/s |
FP16 (half) performance | 30.64 TFLOPS (1:1) | 40.00 TFLOPS (1:1) |
FP32 (float) performance | 30.64 TFLOPS | 40.00 TFLOPS |
FP64 (double) performance | 478.8 GFLOPS (1:64) | 625.0 GFLOPS (1:64) |
Slot width | Dual-slot | Triple-slot |
---|---|---|
Length | 285 mm 11.2 inches | 336 mm 13.2 inches |
Width | 112 mm 4.4 inches | 140 mm 5.5 inches |
Height | 40 mm 1.6 inches | 61 mm 2.4 inches |
TDP | 350 W | 450 W |
Suggested PSU | 750 W | 850 W |
Outputs | 1x HDMI3x DisplayPort | 1x HDMI3x DisplayPort |
Power connectors | 1x 12-pin | 1x 12-pin |
Board number | — | PG132 SKU |
DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 8.6 | 8.6 |
Shader model | 6.6 | 6.6 |