Product Compare
NVIDIA RTX 4000 SFF Ada Generation vs NVIDIA Tesla K40st
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Nov 2018 | Nov 2013 |
| Generation | Radeon Instinct | Tesla |
| Type | Professional | Professional |
| Bus Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Base Clock | 1200 MHz | |
| Boost Clock | 1746 MHz | |
| Memory Clock | 1000 MHz | 1502 MHz |
| Memory | ||
| Memory Size | 16GB | 12GB |
| Memory Type | HBM2 | GDDR5 |
| Memory Bus | 4096bit | 384bit |
| Bandwidth | 1.02TB/s | 288.4GB/s |
| Render Config | ||
| SM Count | 48 | |
| Compute Units | 60 | |
| Shading Units | 3840 | 2880 |
| TMUs | 240 | 240 |
| ROPs | 64 | 48 |
| Tensor Cores | 192 | |
| RT Cores | 48 | |
| L1 Cache | 16 KB (per CU) | 16 KB (per SMX) |
| L2 Cache | 4 MB | 1536 KB |
| Theoretical Performance | ||
| Pixel Rate | 111.7 GPixel/s | 34.50 GPixel/s |
| Texture Rate | 419.0 GTexel/s | 138.0 GTexel/s |
| FP16 (half) | 26.82 TFLOPS | |
| FP32 (float) | 13.41 TFLOPS | 3.312 TFLOPS |
| FP64 (double) | 6.705 TFLOPS | 1104 GFLOPS |
| Graphics Processor | ||
| GPU Name | Vega 20 | GK110B |
| GPU Variant | Vega 20 GLXT | |
| Architecture | GCN 5.1 | Kepler |
| Foundry | TSMC | TSMC |
| Process Size | 7 nm | 28 nm |
| Transistors | 13.23 billion | 7.08 billion |
| Die Size | 331 mm² | 561 mm² |
| Board Design | ||
| TDP | 300W | 245W |
| Suggested PSU | 700 W | 550 W |
| Outputs | 1x mini-DisplayPort 1.4a | No outputs |
| Power Connectors | 2x 8-pin | |
| Graphics Features | ||
| DirectX | 12 (12_1) | 12 (11_1) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.3 | 1.1 |
| CUDA | 8.9 | 3.5 |
| Shader Model | 6.7 | 5.1 |