Product Compare
AMD Radeon Pro W6800 vs NVIDIA H100 NVL 94 GB
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jun 2021 | Mar 2023 |
| Generation | Radeon Pro | Server Hopper |
| Type | Desktop | AI GPU |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 2075 MHz | 1080 MHz |
| Boost Clock | 2320 MHz | 1785 MHz |
| Memory Clock | 2000 MHz | 1313 MHz |
| Memory | ||
| Memory Size | 32GB | 94GB |
| Memory Type | GDDR6 | HBM3 |
| Memory Bus | 256bit | 6016bit |
| Bandwidth | 512.0GB/s | 3.95TB/s |
| Render Config | ||
| Compute Units | 60 | |
| Shading Units | 3840 | 16896 |
| TMUs | 240 | 528 |
| ROPs | 96 | 24 |
| RT Cores | 60 | |
| L1 Cache | 128 KB per Array | 256 KB (per SM) |
| L2 Cache | 4 MB | 50 MB |
| L3 Cache | 128 MB | |
| SM Count | 132 | |
| Tensor Cores | 528 | |
| Theoretical Performance | ||
| Pixel Rate | 222.7 GPixel/s | 42.84 GPixel/s |
| Texture Rate | 556.8 GTexel/s | 942.5 GTexel/s |
| FP16 (half) | 35.64 TFLOPS | 241.3 TFLOPS |
| FP32 (float) | 17.82 TFLOPS | 60.32 TFLOPS |
| FP64 (double) | 1114 GFLOPS | 30.16 TFLOPS |
| Graphics Processor | ||
| GPU Name | Navi 21 | GH100 |
| Architecture | RDNA 2.0 | Hopper |
| Foundry | TSMC | TSMC |
| Process Size | 7 nm | 5 nm |
| Transistors | 26.8 billion | 80 billion |
| Die Size | 520 mm² | 814 mm² |
| Board Design | ||
| TDP | 250W | 400W |
| Suggested PSU | 600 W | 800 W |
| Outputs | 6x mini-DisplayPort 1.4a | No outputs |
| Power Connectors | 1x 6-pin + 1x 8-pin | 8-pin EPS |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.3 | N/A |
| Shader Model | 6.5 | N/A |
| CUDA | 9.0 | |