Product Compare
AMD Radeon Instinct MI308X vs NVIDIA P102 101
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Dec 2023 | Jan 2018 |
| Generation | Radeon Instinct | Mining GPUs |
| Type | Professional | Desktop |
| Bus Interface | PCIe 5.0 x16 | PCIe 3.0 x4 |
| Clock Speeds | ||
| Base Clock | 1000 MHz | 1557 MHz |
| Boost Clock | 2100 MHz | 1670 MHz |
| Memory Clock | 1300 MHz | 2002 MHz |
| Memory | ||
| Memory Size | 192GB | 10GB |
| Memory Type | HBM3 | GDDR5 |
| Memory Bus | 8192bit | 320bit |
| Bandwidth | 5.32TB/s | 320.3GB/s |
| Render Config | ||
| Compute Units | 304 | |
| Shading Units | 19456 | 3200 |
| TMUs | 1216 | 200 |
| ROPs | 0 | 80 |
| Tensor Cores | 1216 | |
| L1 Cache | 16 KB (per CU) | 48 KB (per SM) |
| L2 Cache | 16 MB | 0 MB |
| L3 Cache | 256 MB | |
| SM Count | 25 | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 133.6 GPixel/s |
| Texture Rate | 2553.6 GTexel/s | 334.0 GTexel/s |
| FP16 (half) | 81.72 TFLOPS | 167.0 GFLOPS |
| FP32 (float) | 81.72 TFLOPS | 10.69 TFLOPS |
| FP64 (double) | 40.86 TFLOPS | 334.0 GFLOPS |
| Graphics Processor | ||
| GPU Name | Aqua Vanjaram | GP102 |
| Architecture | CDNA 3.0 | Pascal |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 16 nm |
| Transistors | 153 billion | 11.8 billion |
| Die Size | 1017 mm² | 471 mm² |
| GPU Variant | GP102-101-A1 | |
| Board Design | ||
| TDP | 750W | 250W |
| Suggested PSU | 1150 W | 600 W |
| Outputs | No outputs | No outputs |
| Power Connectors | None | 2x 8-pin |
| Graphics Features | ||
| DirectX | N/A | 12 (12_1) |
| OpenGL | N/A | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | 1.3 |
| Shader Model | N/A | 6.4 |
| CUDA | 6.1 | |