Intel UHD Graphics P630 vs NVIDIA GRID A100A
Graphics Processor
| GPU model | Coffee Lake GT2 | GA100 |
|---|---|---|
| Architecture | Generation 9.5 | Ampere |
| Foundry | Intel | TSMC |
| Manufacturing process | 14 nm+++ | 7 nm |
| Number of transistors | unknown | 54,200 million |
| Die size | unknown | 826 mm² |
Integrated Graphics
| Release date | May 24th, 2018 | — |
|---|---|---|
| Generation | HD Graphics-W (Coffee Lake) | — |
| Production | Active | — |
| Interface | Ring Bus | — |
| Reviews | 1 in our database | — |
Clocks
| Base clock | 350 MHz | 900 MHz |
|---|---|---|
| Boost clock | 1100 MHz | 1005 MHz |
| Memory clock | System Shared | 1215 MHz 2.4 Gbps effective |
Memory Configuration
| Memory size | System Shared | 48 GB |
|---|---|---|
| Memory type | System Shared | HBM2E |
| Memory bus width | System Shared | 6144 bit |
| Bandwidth | System Dependent | 1,866 GB/s |
Render Configuration
| Shading units | 192 | 6912 |
|---|---|---|
| TMUs | 24 | 432 |
| ROPs | 3 | 192 |
| Execution units | 24 | — |
| SM count | — | 108 |
| Tensor cores | — | 432 |
| Cache L1 | — | 192 KB (per SM) |
| Cache L2 | — | 48 MB |
Performance
| Pixel rate | 3.300 GPixel/s | 193.0 GPixel/s |
|---|---|---|
| Texture rate | 26.40 GTexel/s | 434.2 GTexel/s |
| FP16 (half) performance | 844.8 GFLOPS (2:1) | 55.57 TFLOPS (4:1) |
| FP32 (float) performance | 422.4 GFLOPS | 13.89 TFLOPS |
| FP64 (double) performance | 105.6 GFLOPS (1:4) | 6.947 TFLOPS (1:2) |
Dimensions & Outputs
| Slot width | IGP | IGP |
|---|---|---|
| TDP | 15 W | 400 W |
| Outputs | No outputs | No outputs |
| Suggested PSU | — | 800 W |
| Power connectors | — | None |
API Support & Features
| DirectX | 12 (12_1) | — |
|---|---|---|
| OpenGL | 4.6 | — |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | — |
| Shader model | 6.4 | — |
| CUDA | — | 8.0 |
Other Features
Graphics Card
| Release date | — | May 14th, 2020 |
|---|---|---|
| Generation | — | GRID |
| Production | — | Active |
| Interface | — | PCIe 4.0 x16 |
