Product Compare
NVIDIA RTX A4000 vs NVIDIA Tesla K20X
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Apr 2021 | Nov 2012 |
| Generation | Quadro Ampere | Tesla |
| Type | Professional | Professional |
| Bus Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Base Clock | 735 MHz | |
| Boost Clock | 1560 MHz | |
| Memory Clock | 1750 MHz | 1300 MHz |
| Memory | ||
| Memory Size | 16GB | 6GB |
| Memory Type | GDDR6 | GDDR5 |
| Memory Bus | 256bit | 384bit |
| Bandwidth | 448.0GB/s | 249.6GB/s |
| Render Config | ||
| SM Count | 48 | |
| Shading Units | 6144 | 2688 |
| TMUs | 192 | 224 |
| ROPs | 96 | 48 |
| Tensor Cores | 192 | |
| RT Cores | 48 | |
| L1 Cache | 128 KB (per SM) | 16 KB (per SMX) |
| L2 Cache | 4 MB | 1536 KB |
| Theoretical Performance | ||
| Pixel Rate | 149.8 GPixel/s | 40.99 GPixel/s |
| Texture Rate | 299.5 GTexel/s | 164.0 GTexel/s |
| FP16 (half) | 19.17 TFLOPS | |
| FP32 (float) | 19.17 TFLOPS | 3.935 TFLOPS |
| FP64 (double) | 299.5 GFLOPS | 1312 GFLOPS |
| Graphics Processor | ||
| GPU Name | GA104 | GK110 |
| GPU Variant | GA104-875-A1 | |
| Architecture | Ampere | Kepler |
| Foundry | Samsung | TSMC |
| Process Size | 8 nm | 28 nm |
| Transistors | 17.4 billion | 7.08 billion |
| Die Size | 392 mm² | 561 mm² |
| Board Design | ||
| TDP | 140W | 235W |
| Suggested PSU | 300 W | 550 W |
| Outputs | 4x DisplayPort 1.4a | No outputs |
| Power Connectors | 1x 6-pin | 1x 6-pin + 1x 8-pin |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | 12 (11_0) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | 1.1 |
| CUDA | 8.6 | 3.5 |
| Shader Model | 6.7 | 5.1 |