Product Compare
NVIDIA GeForce RTX 4070 Max-Q vs AMD Radeon Instinct MI250X
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2023 | Nov 2021 |
| Generation | GeForce 40 Mobile | Radeon Instinct |
| Type | Mobile | Professional |
| Bus Interface | PCIe 4.0 x8 | PCIe 4.0 x16 |
| Clock Speeds | ||
| Base Clock | 735 MHz | 1000 MHz |
| Boost Clock | 1230 MHz | 1700 MHz |
| Memory Clock | 2000 MHz | 1600 MHz |
| Memory | ||
| Memory Size | 8GB | 128GB |
| Memory Type | GDDR6 | HBM2e |
| Memory Bus | 128bit | 8192bit |
| Bandwidth | 256.0GB/s | 3277GB/s |
| Render Config | ||
| SM Count | 36 | |
| Shading Units | 4608 | 14080 |
| TMUs | 144 | 880 |
| ROPs | 48 | 0 |
| Tensor Cores | 144 | |
| RT Cores | 36 | |
| L1 Cache | 128 KB (per SM) | 16 KB (per CU) |
| L2 Cache | 32 MB | 16 MB |
| Compute Units | 220 | |
| Theoretical Performance | ||
| Pixel Rate | 59.04 GPixel/s | 0 MPixel/s |
| Texture Rate | 177.1 GTexel/s | 1496 GTexel/s |
| FP16 (half) | 11.34 TFLOPS | 383.0 TFLOPS |
| FP32 (float) | 11.34 TFLOPS | 47.87 TFLOPS |
| FP64 (double) | 177.1 GFLOPS | 47.87 TFLOPS |
| Graphics Processor | ||
| GPU Name | AD106 | Aldebaran |
| GPU Variant | GN21-X6 | Aldebaran XT |
| Architecture | Ada Lovelace | CDNA 2.0 |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 6 nm |
| Transistors | 22.9 billion | 58.2 billion |
| Die Size | 188 mm² | Unknown |
| Board Design | ||
| TDP | 35W | 500W |
| Outputs | Portable Device Dependent | No outputs |
| Power Connectors | None | 2x 8-pin |
| Suggested PSU | 900 W | |
| 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 | 8.9 | |
| Shader Model | 6.8 | N/A |