Product Compare
AMD Radeon Instinct MI250 vs NVIDIA GeForce RTX 4080
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Nov 2021 | Sep 2022 |
| Generation | Radeon Instinct | GeForce 40 |
| Type | Professional | Desktop |
| Bus Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| Clock Speeds | ||
| Base Clock | 1000 MHz | 2205 MHz |
| Boost Clock | 1700 MHz | 2505 MHz |
| Memory Clock | 1600 MHz | 1400 MHz |
| Memory | ||
| Memory Size | 128GB | 16GB |
| Memory Type | HBM2e | GDDR6X |
| Memory Bus | 8192bit | 256bit |
| Bandwidth | 3277GB/s | 716.8GB/s |
| Render Config | ||
| Compute Units | 208 | |
| Shading Units | 13312 | 9728 |
| TMUs | 832 | 304 |
| ROPs | 0 | 112 |
| L1 Cache | 16 KB (per CU) | 128 KB (per SM) |
| L2 Cache | 16 MB | 64 MB |
| SM Count | 76 | |
| Tensor Cores | 304 | |
| RT Cores | 76 | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 280.6 GPixel/s |
| Texture Rate | 1414 GTexel/s | 761.5 GTexel/s |
| FP16 (half) | 362.1 TFLOPS | 48.74 TFLOPS |
| FP32 (float) | 45.26 TFLOPS | 48.74 TFLOPS |
| FP64 (double) | 45.26 TFLOPS | 761.5 GFLOPS |
| Graphics Processor | ||
| GPU Name | Aldebaran | AD103 |
| GPU Variant | Aldebaran | AD103-300-A1 |
| Architecture | CDNA 2.0 | Ada Lovelace |
| Foundry | TSMC | TSMC |
| Process Size | 6 nm | 5 nm |
| Transistors | 58.2 billion | 45.9 billion |
| Die Size | Unknown | 379 mm² |
| Board Design | ||
| TDP | 500W | 320W |
| Suggested PSU | 900 W | 700 W |
| Outputs | No outputs | 1x HDMI 2.13x DisplayPort 1.4a |
| Power Connectors | 2x 8-pin | 1x 16-pin |
| Graphics Features | ||
| DirectX | N/A | 12 Ultimate (12_2) |
| OpenGL | N/A | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | 1.3 |
| Shader Model | N/A | 6.7 |
| CUDA | 8.9 | |