GPU model | Navi 24 | GA102 |
---|---|---|
Architecture | RDNA 2.0 | Ampere |
Foundry | TSMC | Samsung |
Manufacturing process | 7 nm | 8 nm |
Number of transistors | unknown | 28,300 million |
Die size | unknown | 628 mm² |
GPU variant | — | GA102-200-KD-A1 |
Release date | Unknown | Sep 1st, 2020 |
---|---|---|
Generation | Navi II | GeForce 30 |
Predecessor | Navi | GeForce 20 |
Successor | Navi III | — |
Production | Unreleased | Active |
Interface | PCIe 4.0 x8 | PCIe 4.0 x16 |
Launch price | — | 699 USD |
gpu.details.availability | — | Sep 17th, 2020 |
Current price | — | Amazon / Newegg |
Reviews | — | 99 in our database |
Base clock | 2200 MHz | 1440 MHz |
---|---|---|
Boost clock | 2500 MHz | 1710 MHz |
Memory clock | 2000 MHz 16 Gbps effective | 1188 MHz 19 Gbps effective |
Memory size | 4 GB | 10 GB |
---|---|---|
Memory type | GDDR6 | GDDR6X |
Memory bus width | 64 bit | 320 bit |
Bandwidth | 128.0 GB/s | 760.3 GB/s |
Shading units | 1024 | 8704 |
---|---|---|
TMUs | 64 | 272 |
ROPs | 32 | 96 |
Compute units | 16 | — |
RT cores | 16 | 68 |
gpu.details.l0-cache | 32 KB per WGP | — |
Cache L1 | 128 KB per Array | 128 KB (per SM) |
Cache L2 | 1024 KB | 5 MB |
gpu.details.l3-cache | 16 MB | — |
SM count | — | 68 |
Tensor cores | — | 272 |
Pixel rate | 80.00 GPixel/s | 164.2 GPixel/s |
---|---|---|
Texture rate | 160.0 GTexel/s | 465.1 GTexel/s |
FP16 (half) performance | 10.24 TFLOPS (2:1) | 29.77 TFLOPS (1:1) |
FP32 (float) performance | 5.120 TFLOPS | 29.77 TFLOPS |
FP64 (double) performance | 320.0 GFLOPS (1:16) | 465.1 GFLOPS (1:64) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
TDP | 100 W | 320 W |
Suggested PSU | 300 W | 700 W |
Outputs | 1x HDMI2x DisplayPort | 1x HDMI3x DisplayPort |
Power connectors | 1x 6-pin | 1x 12-pin |
Length | — | 285 mm 11.2 inches |
Board number | — | PG132 SKU 30 |
Width | — | 112 mm 4.4 inches |
Height | — | 40 mm 1.6 inches |
DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.6 | 6.6 |
CUDA | — | 8.6 |