Product Compare
AMD Radeon Instinct MI300X vs NVIDIA GeForce RTX 4070 Max-Q
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Dec 2023 | Jan 2023 |
| Generation | Radeon Instinct | GeForce 40 Mobile |
| Type | Professional | Mobile |
| Bus Interface | PCIe 5.0 x16 | PCIe 4.0 x8 |
| Clock Speeds | ||
| Base Clock | 1000 MHz | 735 MHz |
| Boost Clock | 2100 MHz | 1230 MHz |
| Memory Clock | 1300 MHz | 2000 MHz |
| Memory | ||
| Memory Size | 192GB | 8GB |
| Memory Type | HBM3 | GDDR6 |
| Memory Bus | 8192bit | 128bit |
| Bandwidth | 5.32TB/s | 256.0GB/s |
| Render Config | ||
| Compute Units | 304 | |
| Shading Units | 19456 | 4608 |
| TMUs | 1216 | 144 |
| ROPs | 0 | 48 |
| Tensor Cores | 1216 | 144 |
| L1 Cache | 16 KB (per CU) | 128 KB (per SM) |
| L2 Cache | 16 MB | 32 MB |
| L3 Cache | 256 MB | |
| SM Count | 36 | |
| RT Cores | 36 | |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 59.04 GPixel/s |
| Texture Rate | 2553.6 GTexel/s | 177.1 GTexel/s |
| FP16 (half) | 81.72 TFLOPS | 11.34 TFLOPS |
| FP32 (float) | 81.72 TFLOPS | 11.34 TFLOPS |
| FP64 (double) | 40.86 TFLOPS | 177.1 GFLOPS |
| Graphics Processor | ||
| GPU Name | Aqua Vanjaram | AD106 |
| Architecture | CDNA 3.0 | Ada Lovelace |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 5 nm |
| Transistors | 153 billion | 22.9 billion |
| Die Size | 1017 mm² | 188 mm² |
| GPU Variant | GN21-X6 | |
| Board Design | ||
| TDP | 750W | 35W |
| Suggested PSU | 1150 W | |
| Outputs | No outputs | Portable Device Dependent |
| Power Connectors | None | None |
| 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 | 6.8 |
| CUDA | 8.9 | |