GPU model | GA102 | GP104 |
---|---|---|
GPU variant | GA102-890-A1 | GP104-400-A1 |
Architecture | Ampere | Pascal |
Foundry | Samsung | TSMC |
Manufacturing process | 8 nm | 16 nm |
Number of transistors | 28,300 million | 7,200 million |
Die size | 628 mm² | 314 mm² |
Release date | Apr 12th, 2021 | May 27th, 2016 |
---|---|---|
gpu.details.availability | 2021 | — |
Generation | Tesla | GeForce 10 |
Production | Active | End-of-life |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Predecessor | — | GeForce 900 |
Successor | — | GeForce 20 |
Launch price | — | 599 USD |
Reviews | — | 183 in our database |
Base clock | 1320 MHz | 1607 MHz |
---|---|---|
Boost clock | 1710 MHz | 1733 MHz |
Memory clock | 1563 MHz 12.5 Gbps effective | 1251 MHz 10 Gbps effective |
Memory size | 12 GB | 8 GB |
---|---|---|
Memory type | GDDR6 | GDDR5X |
Memory bus width | 384 bit | 256 bit |
Bandwidth | 600.2 GB/s | 320.3 GB/s |
Shading units | 9216 | 2560 |
---|---|---|
TMUs | 288 | 160 |
ROPs | 96 | 64 |
SM count | 72 | 20 |
Tensor cores | 288 | — |
RT cores | 72 | — |
Cache L1 | 128 KB (per SM) | 48 KB (per SM) |
Cache L2 | 6 MB | 2 MB |
Pixel rate | 164.2 GPixel/s | 110.9 GPixel/s |
---|---|---|
Texture rate | 492.5 GTexel/s | 277.3 GTexel/s |
FP16 (half) performance | 31.52 TFLOPS (1:1) | 138.6 GFLOPS (1:64) |
FP32 (float) performance | 31.52 TFLOPS | 8.873 TFLOPS |
FP64 (double) performance | 985.0 GFLOPS (1:32) | 277.3 GFLOPS (1:32) |
Slot width | Single-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
Width | 112 mm 4.4 inches | 112 mm 4.4 inches |
TDP | 150 W | 180 W |
Suggested PSU | 450 W | 450 W |
Outputs | No outputs | 1x DVI1x HDMI3x DisplayPort |
Power connectors | 8-pin EPS | 1x 8-pin |
Board number | PG133 SKU 215 | PG413 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 | 6.1 |
Shader model | 6.6 | 6.4 |