GPU model | DG1 | TU117 |
---|---|---|
Architecture | Generation 12.1 | Turing |
Foundry | Intel | TSMC |
Manufacturing process | 10 nm | 12 nm |
Number of transistors | unknown | 4,700 million |
Die size | 95 mm² | 200 mm² |
Release date | Oct 31st, 2020 | Apr 12th, 2021 |
---|---|---|
Generation | Iris Xe Graphics-M (Tiger Lake) | Quadro Mobile (Tx000) |
Production | Active | Active |
Interface | PCIe 4.0 x4 | PCIe 3.0 x16 |
Reviews | 1 in our database | — |
Base clock | 300 MHz | 855 MHz |
---|---|---|
Boost clock | 1650 MHz | 1425 MHz |
Memory clock | 2133 MHz 4.3 Gbps effective | 1250 MHz 10 Gbps effective |
Memory size | 4 GB | 4 GB |
---|---|---|
Memory type | LPDDR4X | GDDR6 |
Memory bus width | 128 bit | 128 bit |
Bandwidth | 68.26 GB/s | 160.0 GB/s |
Shading units | 768 | 1024 |
---|---|---|
TMUs | 48 | 64 |
ROPs | 24 | 32 |
Execution units | 96 | — |
Cache L2 | 1024 KB | 1024 KB |
gpu.details.l3-cache | 16 MB | — |
SM count | — | 16 |
Cache L1 | — | 64 KB (per SM) |
Pixel rate | 39.60 GPixel/s | 45.60 GPixel/s |
---|---|---|
Texture rate | 79.20 GTexel/s | 91.20 GTexel/s |
FP16 (half) performance | 5.069 TFLOPS (2:1) | 5.837 TFLOPS (2:1) |
FP32 (float) performance | 2.534 TFLOPS | 2.918 TFLOPS |
FP64 (double) performance | 633.6 GFLOPS (1:4) | 91.20 GFLOPS (1:32) |
Slot width | IGP | IGP |
---|---|---|
TDP | 25 W | 18 W |
Outputs | No outputs | No outputs |
Power connectors | — | None |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.6 |
CUDA | — | 7.5 |