GPU model | DG1 | GP10B |
---|---|---|
Architecture | Generation 12.1 | Pascal |
Foundry | Intel | TSMC |
Manufacturing process | 10 nm | 16 nm |
Number of transistors | unknown | unknown |
Die size | 95 mm² | unknown |
GPU variant | — | Tegra X2 |
Release date | Oct 31st, 2020 | — |
---|---|---|
Generation | Iris Xe Graphics-M (Tiger Lake) | — |
Production | Active | — |
Interface | PCIe 4.0 x4 | — |
Reviews | 1 in our database | — |
Base clock | 300 MHz | 854 MHz |
---|---|---|
Boost clock | 1650 MHz | 1465 MHz |
Memory clock | 2133 MHz 4.3 Gbps effective | System Shared |
Memory size | 4 GB | System Shared |
---|---|---|
Memory type | LPDDR4X | System Shared |
Memory bus width | 128 bit | System Shared |
Bandwidth | 68.26 GB/s | System Dependent |
Shading units | 768 | 256 |
---|---|---|
TMUs | 48 | 16 |
ROPs | 24 | 16 |
Execution units | 96 | — |
Cache L2 | 1024 KB | 512 KB |
gpu.details.l3-cache | 16 MB | — |
SM count | — | 2 |
Cache L1 | — | 48 KB (per SM) |
Pixel rate | 39.60 GPixel/s | 23.44 GPixel/s |
---|---|---|
Texture rate | 79.20 GTexel/s | 23.44 GTexel/s |
FP16 (half) performance | 5.069 TFLOPS (2:1) | 1.500 TFLOPS (2:1) |
FP32 (float) performance | 2.534 TFLOPS | 750.1 GFLOPS |
FP64 (double) performance | 633.6 GFLOPS (1:4) | 23.44 GFLOPS (1:32) |
Slot width | IGP | IGP |
---|---|---|
TDP | 25 W | 15 W |
Outputs | No outputs | No outputs |
Length | — | 87 mm 3.4 inches |
Width | — | 50 mm 2 inches |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 1.2 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.4 |
CUDA | — | 6.2 |
Release date | — | 2016 |
---|---|---|
Generation | — | Tegra |
Production | — | Active |
Launch price | — | 249 USD |
Interface | — | IGP |