Product Compare
NVIDIA GeForce RTX 3070 Ti Max Q vs NVIDIA H800 SXM5
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2022 | Mar 2023 |
| Generation | GeForce 30 Mobile | Server Hopper |
| Type | Mobile | AI GPU |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 510 MHz | 1095 MHz |
| Boost Clock | 1035 MHz | 1755 MHz |
| Memory Clock | 1500 MHz | 1313 MHz |
| Memory | ||
| Memory Size | 8GB | 80GB |
| Memory Type | GDDR6 | HBM3 |
| Memory Bus | 256bit | 5120bit |
| Bandwidth | 384.0GB/s | 3.36TB/s |
| Render Config | ||
| SM Count | 46 | 132 |
| Shading Units | 5888 | 16896 |
| TMUs | 184 | 528 |
| ROPs | 96 | 24 |
| Tensor Cores | 184 | 528 |
| RT Cores | 46 | |
| L1 Cache | 128 KB (per SM) | 256 KB (per SM) |
| L2 Cache | 4 MB | 50 MB |
| Theoretical Performance | ||
| Pixel Rate | 99.36 GPixel/s | 42.12 GPixel/s |
| Texture Rate | 190.4 GTexel/s | 926.6 GTexel/s |
| FP16 (half) | 12.19 TFLOPS | 237.2 TFLOPS |
| FP32 (float) | 12.19 TFLOPS | 59.30 TFLOPS |
| FP64 (double) | 190.4 GFLOPS | 29.65 TFLOPS |
| Graphics Processor | ||
| GPU Name | GA104 | GH100 |
| Architecture | Ampere | Hopper |
| Foundry | Samsung | TSMC |
| Process Size | 8 nm | 5 nm |
| Transistors | 17.4 billion | 80 billion |
| Die Size | 392 mm² | 814 mm² |
| Board Design | ||
| TDP | 80W | 700W |
| Outputs | No outputs | No outputs |
| Power Connectors | None | 8-pin EPS |
| Suggested PSU | 1100 W | |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | N/A |
| CUDA | 8.6 | 9.0 |
| Shader Model | 6.6 | N/A |