Product Compare
AMD Radeon PRO W9070 vs NVIDIA H100 PCIe 96 GB
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Unknown | Mar 2023 |
| Generation | Mining GPUs | Tesla Hopper |
| Type | Desktop | AI GPU |
| Bus Interface | PCIe 3.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1120 MHz | 1665 MHz |
| Boost Clock | 1206 MHz | 1837 MHz |
| Memory Clock | 2100 MHz | 1313 MHz |
| Memory | ||
| Memory Size | 4GB | 96GB |
| Memory Type | GDDR5 | HBM3 |
| Memory Bus | 256bit | 5120bit |
| Bandwidth | 268.8GB/s | 1681GB/s |
| Render Config | ||
| Compute Units | 32 | |
| Shading Units | 2048 | 16896 |
| TMUs | 128 | 528 |
| ROPs | 32 | 24 |
| Tensor Cores | 128 | 528 |
| RT Cores | 64 | |
| L1 Cache | 16 KB (per CU) | 256 KB (per SM) |
| L2 Cache | 2 MB | 50 MB |
| L3 Cache | 64 MB | |
| SM Count | 132 | |
| Theoretical Performance | ||
| Pixel Rate | 38.59 GPixel/s | 44.09 GPixel/s |
| Texture Rate | 154.4 GTexel/s | 969.9 GTexel/s |
| FP16 (half) | 4.940 TFLOPS | 248.3 TFLOPS |
| FP32 (float) | 4.940 TFLOPS | 62.08 TFLOPS |
| FP64 (double) | 308.7 GFLOPS | 31.04 TFLOPS |
| Graphics Processor | ||
| GPU Name | Polaris 20 | GH100 |
| GPU Variant | Navi 48 XTW | |
| Architecture | GCN 4.0 | Hopper |
| Foundry | GlobalFoundries | TSMC |
| Process Size | 14 nm | 4 nm |
| Transistors | 5.7 billion | 80 billion |
| Die Size | 232 mm² | 814 mm² |
| Board Design | ||
| TDP | 180W | 700W |
| Suggested PSU | 450 W | 1100 W |
| Outputs | 1x HDMI 2.0b | No outputs |
| Power Connectors | 2x 8-pin | 8-pin EPS |
| Graphics Features | ||
| DirectX | 12 (12_0) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.3 | N/A |
| Shader Model | 6.7 | N/A |
| CUDA | 9.0 | |