Product Compare
NVIDIA A800 SXM4 80 GB vs AMD Radeon Instinct MI325X
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Aug 2022 | Oct 2024 |
| Generation | Ampere | Radeon Instinct |
| Type | AI GPU | Professional |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1155 MHz | 1000 MHz |
| Boost Clock | 1410 MHz | 2100 MHz |
| Memory Clock | 1593 MHz | 1500 MHz |
| Memory | ||
| Memory Size | 80GB | 256GB |
| Memory Type | HBM2e | HBM3e |
| Memory Bus | 5120bit | 8192bit |
| Bandwidth | 2039GB/s | 6.14TB/s |
| Render Config | ||
| SM Count | 108 | |
| Shading Units | 6912 | 19456 |
| TMUs | 432 | 1216 |
| ROPs | 160 | 0 |
| Tensor Cores | 432 | 1216 |
| L1 Cache | 192 KB (per SM) | 16 KB (per CU) |
| L2 Cache | 40 MB | 16 MB |
| Compute Units | 304 | |
| L3 Cache | 256 MB | |
| Theoretical Performance | ||
| Pixel Rate | 225.6 GPixel/s | 0 MPixel/s |
| Texture Rate | 609.1 GTexel/s | 2553.6 GTexel/s |
| FP16 (half) | 77.97 TFLOPS | 81.72 TFLOPS |
| FP32 (float) | 19.49 TFLOPS | 81.72 TFLOPS |
| FP64 (double) | 9.746 TFLOPS | 40.86 TFLOPS |
| Graphics Processor | ||
| GPU Name | GA100 | Aqua Vanjaram |
| Architecture | Ampere | CDNA 3.0 |
| Foundry | TSMC | TSMC |
| Process Size | 7 nm | 5 nm |
| Transistors | 54.2 billion | 153 billion |
| Die Size | 826 mm² | 1017 mm² |
| Board Design | ||
| TDP | 400W | 1000W |
| Suggested PSU | 800 W | 1400 W |
| Outputs | No outputs | No outputs |
| Power Connectors | None | None |
| Graphics Features | ||
| DirectX | N/A | N/A |
| OpenGL | N/A | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | N/A |
| CUDA | 8.0 | |
| Shader Model | N/A | N/A |