Product Compare
NVIDIA GRID A100A vs AMD Radeon Pro W6900X
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | May 2020 | Aug 2021 |
| Generation | GRID | Radeon Pro Mac |
| Type | Professional | Desktop |
| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Clock Speeds | ||
| Base Clock | 1095 MHz | 1825 MHz |
| Boost Clock | 1410 MHz | 2150 MHz |
| Memory Clock | 1215 MHz | 2000 MHz |
| Memory | ||
| Memory Size | 32GB | 32GB |
| Memory Type | HBM2e | GDDR6 |
| Memory Bus | 6144bit | 256bit |
| Bandwidth | 1.87TB/s | 512.0GB/s |
| Render Config | ||
| SM Count | 108 | |
| Shading Units | 6912 | 5120 |
| TMUs | 432 | 320 |
| ROPs | 192 | 128 |
| Tensor Cores | 432 | |
| L1 Cache | 192 KB (per SM) | 128 KB per Array |
| L2 Cache | 32 MB | 4 MB |
| Compute Units | 80 | |
| RT Cores | 80 | |
| L3 Cache | 128 MB | |
| Theoretical Performance | ||
| Pixel Rate | 270.7 GPixel/s | 275.2 GPixel/s |
| Texture Rate | 609.1 GTexel/s | 688.0 GTexel/s |
| FP16 (half) | 77.97 TFLOPS | 44.03 TFLOPS |
| FP32 (float) | 19.49 TFLOPS | 22.02 TFLOPS |
| FP64 (double) | 9.746 TFLOPS | 1376 GFLOPS |
| Graphics Processor | ||
| GPU Name | GA100 | Navi 21 |
| Architecture | Ampere | RDNA 2.0 |
| Foundry | TSMC | TSMC |
| Process Size | 7 nm | 7 nm |
| Transistors | 54.2 billion | 26.8 billion |
| Die Size | 826 mm² | 520 mm² |
| GPU Variant | Navi 21 Pro-XTA | |
| Board Design | ||
| TDP | 400W | 300W |
| Suggested PSU | 800 W | 700 W |
| Outputs | No outputs | 1x HDMI 2.14x Thunderbolt |
| Power Connectors | None | None |
| Graphics Features | ||
| DirectX | N/A | 12 Ultimate (12_2) |
| OpenGL | N/A | 4.6 |
| OpenCL | 3.0 | 2.1 |
| Vulkan | N/A | 1.3 |
| CUDA | 8.0 | |
| Shader Model | N/A | 6.5 |