Product Compare
NVIDIA RTX A6000 vs AMD Instinct MI300X
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Oct 2020 | Dec 2023 |
| Generation | Quadro Ampere | Radeon Instinct |
| Type | Professional | Professional |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1410 MHz | 1000 MHz |
| Boost Clock | 1800 MHz | 2100 MHz |
| Memory Clock | 2000 MHz | 2525 MHz |
| Memory | ||
| Memory Size | 48GB | 192GB |
| Memory Type | GDDR6 | HBM3 |
| Memory Bus | 384bit | 8192bit |
| Bandwidth | 768.0GB/s | 10.3TB/s |
| Render Config | ||
| SM Count | 84 | |
| Shading Units | 10752 | 19456 |
| TMUs | 336 | 1216 |
| ROPs | 112 | 0 |
| Tensor Cores | 336 | |
| RT Cores | 84 | |
| L1 Cache | 128 KB (per SM) | 16 KB (per CU) |
| L2 Cache | 6 MB | 16 MB |
| Compute Units | 304 | |
| L3 Cache | 256 MB | |
| Theoretical Performance | ||
| Pixel Rate | 201.6 GPixel/s | 0 MPixel/s |
| Texture Rate | 604.8 GTexel/s | 2553.6 GTexel/s |
| FP16 (half) | 38.71 TFLOPS | 653.7 TFLOPS |
| FP32 (float) | 38.71 TFLOPS | 81.72 TFLOPS |
| FP64 (double) | 604.8 GFLOPS | 81.72 TFLOPS |
| Graphics Processor | ||
| GPU Name | GA102 | Aqua Vanjaram |
| Architecture | Ampere | CDNA 3.0 |
| Foundry | Samsung | TSMC |
| Process Size | 8 nm | 5 nm |
| Transistors | 28.3 billion | 153 billion |
| Die Size | 628 mm² | 1017 mm² |
| Board Design | ||
| TDP | 300W | 750W |
| Suggested PSU | 700 W | 1150 W |
| Outputs | 4x DisplayPort 1.4a | No outputs |
| Power Connectors | 8-pin EPS | None |
| 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 | |
| Shader Model | 6.7 | N/A |