Product Compare
Intel Data Center GPU Max 1350 vs NVIDIA Tesla K20s
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2023 | Feb 2013 |
| Generation | Data Center GPU | Tesla |
| Type | Professional | Professional |
| Bus Interface | PCIe 5.0 x16 | PCIe 2.0 x16 |
| Clock Speeds | ||
| Base Clock | 750 MHz | |
| Boost Clock | 1550 MHz | |
| Memory Clock | 1200 MHz | 1300 MHz |
| Memory | ||
| Memory Size | 96GB | 5GB |
| Memory Type | HBM2e | GDDR5 |
| Memory Bus | 8192bit | 320bit |
| Bandwidth | 2458GB/s | 208.0GB/s |
| Render Config | ||
| Shading Units | 14336 | 2496 |
| TMUs | 896 | 208 |
| ROPs | 0 | 40 |
| Tensor Cores | 896 | |
| RT Cores | 112 | |
| L1 Cache | 64 KB (per EU) | 16 KB (per SMX) |
| L2 Cache | 408 MB | 1280 KB |
| Theoretical Performance | ||
| Pixel Rate | 0 MPixel/s | 29.90 GPixel/s |
| Texture Rate | 1389 GTexel/s | 119.6 GTexel/s |
| FP16 (half) | 44.44 TFLOPS | |
| FP32 (float) | 44.44 TFLOPS | 2.870 TFLOPS |
| FP64 (double) | 44.44 TFLOPS | 956.8 GFLOPS |
| Graphics Processor | ||
| GPU Name | Ponte Vecchio | GK110 |
| Architecture | Generation 12.5 | Kepler |
| Foundry | Intel | TSMC |
| Process Size | 10 nm | 28 nm |
| Transistors | 100 billion | 7.08 billion |
| Die Size | 1280 mm² | 561 mm² |
| Board Design | ||
| TDP | 450W | 225W |
| Suggested PSU | 850 W | 550 W |
| Outputs | No outputs | No outputs |
| Graphics Features | ||
| DirectX | 12 (12_1) | 12 (11_0) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | 1.1 |
| Shader Model | 6.6 | 5.1 |
| CUDA | 3.5 | |