GPU model | GA104 | TU102 |
---|---|---|
GPU variant | GA104-200-A1 | TU102-400-A1 |
Architecture | Ampere | Turing |
Foundry | Samsung | TSMC |
Manufacturing process | 8 nm | 12 nm |
Number of transistors | 17,400 million | 18,600 million |
Die size | 392 mm² | 754 mm² |
Release date | Dec 1st, 2020 | Dec 18th, 2018 |
---|---|---|
gpu.details.availability | Dec 2nd, 2020 | — |
Generation | GeForce 30 | GeForce 20 |
Predecessor | GeForce 20 | GeForce 10 |
Production | Active | Active |
Launch price | 399 USD | 2,499 USD |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Reviews | 64 in our database | 6 in our database |
Successor | — | GeForce 30 |
Base clock | 1410 MHz | 1350 MHz |
---|---|---|
Boost clock | 1665 MHz | 1770 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 8 GB | 24 GB |
---|---|---|
Memory type | GDDR6 | GDDR6 |
Memory bus width | 256 bit | 384 bit |
Bandwidth | 448.0 GB/s | 672.0 GB/s |
Shading units | 4864 | 4608 |
---|---|---|
TMUs | 152 | 288 |
ROPs | 80 | 96 |
SM count | 38 | 72 |
Tensor cores | 152 | 576 |
RT cores | 38 | 72 |
Cache L1 | 128 KB (per SM) | 64 KB (per SM) |
Cache L2 | 4 MB | 6 MB |
Pixel rate | 133.2 GPixel/s | 169.9 GPixel/s |
---|---|---|
Texture rate | 253.1 GTexel/s | 509.8 GTexel/s |
FP16 (half) performance | 16.20 TFLOPS (1:1) | 32.62 TFLOPS (2:1) |
FP32 (float) performance | 16.20 TFLOPS | 16.31 TFLOPS |
FP64 (double) performance | 253.1 GFLOPS (1:64) | 509.8 GFLOPS (1:32) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 242 mm 9.5 inches | 267 mm 10.5 inches |
Width | 112 mm 4.4 inches | 116 mm 4.6 inches |
TDP | 200 W | 280 W |
Suggested PSU | 550 W | 600 W |
Outputs | 1x HDMI3x DisplayPort | 1x HDMI3x DisplayPort1x USB Type-C |
Power connectors | 1x 12-pin | 2x 8-pin |
Board number | PG142 SKU 20 | PG150 SKU 0 |
Height | — | 35 mm 1.4 inches |
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 | 7.5 |
Shader model | 6.6 | 6.6 |