Product Compare
Intel Data Center GPU Max Subsystem vs NVIDIA CMP 170HX 8 GB
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2023 | Sep 2021 |
| Generation | Data Center GPU | Mining GPUs |
| Type | Professional | Desktop |
| Bus Interface | PCIe 5.0 x16 | PCIe 1.0 x4 |
| Clock Speeds | ||
| Base Clock | 900 MHz | 1140 MHz |
| Boost Clock | 1600 MHz | 1410 MHz |
| Memory Clock | 1565 MHz | 1458 MHz |
| Memory | ||
| Memory Size | 128GB | 8GB |
| Memory Type | HBM2e | HBM2e |
| Memory Bus | 8192bit | 4096bit |
| Bandwidth | 3205GB/s | 1.49TB/s |
| Render Config | ||
| Shading Units | 16384 | 4480 |
| TMUs | 1024 | 280 |
| ROPs | 0 | 128 |
| Tensor Cores | 1024 | 280 |
| RT Cores | 128 | |
| L1 Cache | 64 KB (per EU) | 192 KB (per SM) |
| L2 Cache | 408 MB | 8 MB |
| SM Count | 70 | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 180.5 GPixel/s |
| Texture Rate | 1638 GTexel/s | 394.8 GTexel/s |
| FP16 (half) | 52.43 TFLOPS | 50.53 TFLOPS |
| FP32 (float) | 52.43 TFLOPS | 12.63 TFLOPS |
| FP64 (double) | 52.43 TFLOPS | 6.317 TFLOPS |
| Graphics Processor | ||
| GPU Name | Ponte Vecchio | GA100 |
| Architecture | Generation 12.5 | Ampere |
| Foundry | Intel | TSMC |
| Process Size | 10 nm | 7 nm |
| Transistors | 100 billion | 54.2 billion |
| Die Size | 1280 mm² | 826 mm² |
| GPU Variant | GA100-105F-A1 | |
| Board Design | ||
| TDP | 2400W | 250W |
| Suggested PSU | 2800 W | 600 W |
| Outputs | No outputs | No outputs |
| Power Connectors | 1x 16-pin | 2x 8-pin |
| Graphics Features | ||
| DirectX | 12 (12_1) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | N/A |
| Shader Model | 6.6 | N/A |
| CUDA | 8.0 | |