GPU model | Navi 22 | TU104 |
---|---|---|
GPU variant | Navi 22 | N19E-Q3-A1 |
Architecture | RDNA 2.0 | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 7 nm | 12 nm |
Number of transistors | 17,200 million | 13,600 million |
Die size | 335 mm² | 545 mm² |
Release date | Unknown | May 27th, 2019 |
---|---|---|
Generation | Mobility Radeon (Navi II) | Quadro Mobile (Tx000) |
Production | Active | Active |
Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
Base clock | 1489 MHz | 1110 MHz |
---|---|---|
Boost clock | 2250 MHz | 1560 MHz |
Memory clock | 2000 MHz 16 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 10 GB | 8 GB |
---|---|---|
Memory type | GDDR6 | GDDR6 |
Memory bus width | 160 bit | 256 bit |
Bandwidth | 320.0 GB/s | 448.0 GB/s |
Shading units | 2304 | 2560 |
---|---|---|
TMUs | 144 | 160 |
ROPs | 64 | 64 |
Compute units | 36 | — |
RT cores | 36 | 40 |
gpu.details.l0-cache | 32 KB per WGP | — |
Cache L1 | 128 KB per Array | 64 KB (per SM) |
Cache L2 | 3 MB | 4 MB |
gpu.details.l3-cache | 96 MB | — |
SM count | — | 40 |
Tensor cores | — | 320 |
Pixel rate | 144.0 GPixel/s | 99.84 GPixel/s |
---|---|---|
Texture rate | 324.0 GTexel/s | 249.6 GTexel/s |
FP16 (half) performance | 20.74 TFLOPS (2:1) | 15.97 TFLOPS (2:1) |
FP32 (float) performance | 10.37 TFLOPS | 7.987 TFLOPS |
FP64 (double) performance | 648.0 GFLOPS (1:16) | 249.6 GFLOPS (1:32) |
Slot width | IGP | IGP |
---|---|---|
TDP | 180 W | 110 W |
Outputs | No outputs | No outputs |
Power connectors | None | None |
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 | 6.5 | 6.6 |
CUDA | — | 7.5 |