GPU model | Vega 10 | DG1 |
---|---|---|
GPU variant | Vega 10 XL (215-0894216) | — |
Architecture | GCN 5.0 | Generation 12.1 |
Foundry | GlobalFoundries | Intel |
Manufacturing process | 14 nm | 10 nm |
Number of transistors | 12,500 million | unknown |
Die size | 495 mm² | 95 mm² |
Release date | Jun 1st, 2018 | Oct 31st, 2020 |
---|---|---|
Generation | Mobility Radeon (Vega) | Iris Xe Graphics-M (Tiger Lake) |
Production | Active | Active |
Interface | PCIe 3.0 x16 | PCIe 4.0 x4 |
Reviews | 23 in our database | 1 in our database |
Predecessor | Crystal System | — |
Base clock | 1138 MHz | 300 MHz |
---|---|---|
Boost clock | 1301 MHz | 1650 MHz |
Memory clock | 800 MHz 1600 Mbps effective | 2133 MHz 4.3 Gbps effective |
Memory size | 8 GB | 4 GB |
---|---|---|
Memory type | HBM2 | LPDDR4X |
Memory bus width | 2048 bit | 128 bit |
Bandwidth | 409.6 GB/s | 68.26 GB/s |
Shading units | 3584 | 768 |
---|---|---|
TMUs | 224 | 48 |
ROPs | 64 | 24 |
Compute units | 56 | — |
Cache L1 | 16 KB (per CU) | — |
Cache L2 | 4 MB | 1024 KB |
Execution units | — | 96 |
gpu.details.l3-cache | — | 16 MB |
Pixel rate | 83.26 GPixel/s | 39.60 GPixel/s |
---|---|---|
Texture rate | 291.4 GTexel/s | 79.20 GTexel/s |
FP16 (half) performance | 18.65 TFLOPS (2:1) | 5.069 TFLOPS (2:1) |
FP32 (float) performance | 9.326 TFLOPS | 2.534 TFLOPS |
FP64 (double) performance | 582.8 GFLOPS (1:16) | 633.6 GFLOPS (1:4) |
Slot width | MXM Module | IGP |
---|---|---|
Length | 105 mm 4.1 inches | — |
TDP | 120 W | 25 W |
Outputs | 1x HDMI3x DisplayPort | No outputs |
Power connectors | None | — |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 2.1 | 3.0 |
Vulkan | 1.2 | 1.2 |
Shader model | 6.4 | 6.4 |