GPU model | Polaris 20 | TU104B |
---|---|---|
GPU variant | Polaris 20 | — |
Architecture | GCN 4.0 | Turing |
Foundry | GlobalFoundries | TSMC |
Manufacturing process | 14 nm | 12 nm |
Number of transistors | 5,700 million | 13,600 million |
Die size | 232 mm² | 545 mm² |
Release date | Mar 18th, 2019 | Jun 8th, 2020 |
---|---|---|
Generation | Radeon Pro Mac (500X Series) | Quadro Mobile (Tx000) |
Production | Active | Active |
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Base clock | 1000 MHz | 1035 MHz |
---|---|---|
Boost clock | 1105 MHz | 1545 MHz |
Memory clock | 1700 MHz 6.8 Gbps effective | 1750 MHz 14 Gbps effective |
Memory size | 4 GB | 16 GB |
---|---|---|
Memory type | GDDR5 | GDDR6 |
Memory bus width | 256 bit | 256 bit |
Bandwidth | 217.6 GB/s | 448.0 GB/s |
Shading units | 1792 | 3072 |
---|---|---|
TMUs | 112 | 192 |
ROPs | 32 | 64 |
Compute units | 28 | — |
Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
Cache L2 | 2 MB | 4 MB |
SM count | — | 48 |
Tensor cores | — | 384 |
RT cores | — | 48 |
Pixel rate | 35.36 GPixel/s | 98.88 GPixel/s |
---|---|---|
Texture rate | 123.8 GTexel/s | 296.6 GTexel/s |
FP16 (half) performance | 3.960 TFLOPS (1:1) | 18.98 TFLOPS (2:1) |
FP32 (float) performance | 3.960 TFLOPS | 9.492 TFLOPS |
FP64 (double) performance | 247.5 GFLOPS (1:16) | 296.6 GFLOPS (1:32) |
Slot width | IGP | IGP |
---|---|---|
TDP | 150 W | 110 W |
Outputs | No outputs | No outputs |
Power connectors | None | None |
DirectX | 12 (12_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 7.5 |