GPU model | Navi 23 | GP104 |
---|---|---|
Architecture | RDNA 2.0 | Pascal |
Foundry | TSMC | TSMC |
Manufacturing process | 7 nm | 16 nm |
Number of transistors | 11,060 million | 7,200 million |
Die size | 237 mm² | 314 mm² |
GPU variant | — | GP104-400-A1 |
Release date | Jun 8th, 2021 | May 27th, 2016 |
---|---|---|
gpu.details.availability | Jul, 2021 | — |
Generation | Radeon Pro | GeForce 10 |
Production | Active | End-of-life |
Launch price | 649 USD | 599 USD |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Predecessor | — | GeForce 900 |
Successor | — | GeForce 20 |
Reviews | — | 183 in our database |
Base clock | 2331 MHz | 1607 MHz |
---|---|---|
Boost clock | 2903 MHz | 1733 MHz |
Memory clock | 1750 MHz 14 Gbps effective | 1251 MHz 10 Gbps effective |
Memory size | 8 GB | 8 GB |
---|---|---|
Memory type | GDDR6 | GDDR5X |
Memory bus width | 128 bit | 256 bit |
Bandwidth | 224.0 GB/s | 320.3 GB/s |
Shading units | 1792 | 2560 |
---|---|---|
TMUs | 112 | 160 |
ROPs | 64 | 64 |
Compute units | 28 | — |
RT cores | 28 | — |
gpu.details.l0-cache | 32 KB per WGP | — |
Cache L1 | 128 KB per Array | 48 KB (per SM) |
Cache L2 | 2 MB | 2 MB |
gpu.details.l3-cache | 32 MB | — |
SM count | — | 20 |
Pixel rate | 185.8 GPixel/s | 110.9 GPixel/s |
---|---|---|
Texture rate | 325.1 GTexel/s | 277.3 GTexel/s |
FP16 (half) performance | 20.81 TFLOPS (2:1) | 138.6 GFLOPS (1:64) |
FP32 (float) performance | 10.40 TFLOPS | 8.873 TFLOPS |
FP64 (double) performance | 650.3 GFLOPS (1:16) | 277.3 GFLOPS (1:32) |
Slot width | Single-slot | Dual-slot |
---|---|---|
Length | 241 mm 9.5 inches | 267 mm 10.5 inches |
TDP | 100 W | 180 W |
Suggested PSU | 300 W | 450 W |
Outputs | 4x DisplayPort | 1x DVI1x HDMI3x DisplayPort |
Power connectors | 1x 6-pin | 1x 8-pin |
Width | — | 112 mm 4.4 inches |
Height | — | 40 mm 1.6 inches |
Board number | — | PG413 SKU 0 |
DirectX | 12.0 Ultimate (12_2) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.5 | 6.4 |
CUDA | — | 6.1 |