Product Compare
NVIDIA Tesla K20s vs NVIDIA RTX 4000 SFF Ada Generation
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Feb 2013 | Nov 2018 |
| Generation | Tesla | Radeon Instinct |
| Type | Professional | Professional |
| Bus Interface | PCIe 2.0 x16 | PCIe 4.0 x16 |
| Clock Speeds | ||
| Memory Clock | 1300 MHz | 1000 MHz |
| Base Clock | 1200 MHz | |
| Boost Clock | 1746 MHz | |
| Memory | ||
| Memory Size | 5GB | 16GB |
| Memory Type | GDDR5 | HBM2 |
| Memory Bus | 320bit | 4096bit |
| Bandwidth | 208.0GB/s | 1.02TB/s |
| Render Config | ||
| Shading Units | 2496 | 3840 |
| TMUs | 208 | 240 |
| ROPs | 40 | 64 |
| L1 Cache | 16 KB (per SMX) | 16 KB (per CU) |
| L2 Cache | 1280 KB | 4 MB |
| SM Count | 48 | |
| Compute Units | 60 | |
| Tensor Cores | 192 | |
| RT Cores | 48 | |
| Theoretical Performance | ||
| Pixel Rate | 29.90 GPixel/s | 111.7 GPixel/s |
| Texture Rate | 119.6 GTexel/s | 419.0 GTexel/s |
| FP32 (float) | 2.870 TFLOPS | 13.41 TFLOPS |
| FP64 (double) | 956.8 GFLOPS | 6.705 TFLOPS |
| FP16 (half) | 26.82 TFLOPS | |
| Graphics Processor | ||
| GPU Name | GK110 | Vega 20 |
| Architecture | Kepler | GCN 5.1 |
| Foundry | TSMC | TSMC |
| Process Size | 28 nm | 7 nm |
| Transistors | 7.08 billion | 13.23 billion |
| Die Size | 561 mm² | 331 mm² |
| GPU Variant | Vega 20 GLXT | |
| Board Design | ||
| TDP | 225W | 300W |
| Suggested PSU | 550 W | 700 W |
| Outputs | No outputs | 1x mini-DisplayPort 1.4a |
| Power Connectors | 2x 8-pin | |
| Graphics Features | ||
| DirectX | 12 (11_0) | 12 (12_1) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 2.1 |
| Vulkan | 1.1 | 1.3 |
| CUDA | 3.5 | 8.9 |
| Shader Model | 5.1 | 6.7 |