GPU model | Navi 22 | GV100 |
---|---|---|
GPU variant | Navi 22 XTL | GV100-400-A1 |
Architecture | RDNA 2.0 | Volta |
Foundry | TSMC | TSMC |
Manufacturing process | 7 nm | 12 nm |
Number of transistors | 17,200 million | 21,100 million |
Die size | 335 mm² | 815 mm² |
Release date | 2021 | Dec 7th, 2017 |
---|---|---|
Generation | Navi II | GeForce 10 |
Predecessor | Navi | GeForce 900 |
Production | Active | End-of-life |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Successor | Navi III | GeForce 20 |
Launch price | — | 2,999 USD |
Reviews | — | 4 in our database |
Base clock | 2200 MHz | 1200 MHz |
---|---|---|
Boost clock | 2500 MHz | 1455 MHz |
Memory clock | 2000 MHz 16 Gbps effective | 848 MHz 1696 Mbps effective |
Memory size | 6 GB | 12 GB |
---|---|---|
Memory type | GDDR6 | HBM2 |
Memory bus width | 192 bit | 3072 bit |
Bandwidth | 384.0 GB/s | 651.3 GB/s |
Shading units | 2304 | 5120 |
---|---|---|
TMUs | 144 | 320 |
ROPs | 64 | 96 |
Compute units | 36 | — |
RT cores | 36 | — |
gpu.details.l0-cache | 32 KB per WGP | — |
Cache L1 | 128 KB per Array | 96 KB (per SM) |
Cache L2 | 3 MB | 4.5 MB |
gpu.details.l3-cache | 96 MB | — |
SM count | — | 80 |
Tensor cores | — | 640 |
Pixel rate | 160.0 GPixel/s | 139.7 GPixel/s |
---|---|---|
Texture rate | 360.0 GTexel/s | 465.6 GTexel/s |
FP16 (half) performance | 23.04 TFLOPS (2:1) | 29.80 TFLOPS (2:1) |
FP32 (float) performance | 11.52 TFLOPS | 14.90 TFLOPS |
FP64 (double) performance | 720.0 GFLOPS (1:16) | 7.450 TFLOPS (1:2) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
Width | 110 mm 4.3 inches | 112 mm 4.4 inches |
Height | 40 mm 1.6 inches | 40 mm 1.6 inches |
TDP | 180 W | 250 W |
Suggested PSU | 450 W | 600 W |
Outputs | 1x HDMI2x DisplayPort1x USB Type-C | 1x HDMI3x DisplayPort |
Power connectors | 1x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | — | PG500 SKU 0 |
DirectX | 12 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 | 6.6 |
CUDA | — | 7.0 |