Product Compare
Moore Threads MTT S4000 vs NVIDIA Tesla K40s
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Unknown | Nov 2013 |
| Generation | MUSA-Chunxiao | Tesla Kepler |
| Type | Desktop | Professional |
| Bus Interface | PCIe 5.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Base Clock | 1500 MHz | 706 MHz |
| Memory Clock | 2000 MHz | 1502 MHz |
| Boost Clock | 706 MHz | |
| Memory | ||
| Memory Size | 48GB | 12GB |
| Memory Type | GDDR6 | GDDR5 |
| Memory Bus | 384bit | 384bit |
| Bandwidth | 768.0GB/s | 288.4GB/s |
| Render Config | ||
| Compute Units | 64 | |
| Shading Units | 8192 | 2880 |
| TMUs | 512 | 240 |
| ROPs | 512 | 48 |
| Tensor Cores | 128 | |
| L2 Cache | 4 MB | 1536 KB |
| L1 Cache | 16 KB (per SMX) | |
| Theoretical Performance | ||
| Pixel Rate | 768.0 GPixel/s | 42.36 GPixel/s |
| Texture Rate | 768.0 GTexel/s | 169.4 GTexel/s |
| FP16 (half) | 49.15 TFLOPS | |
| FP32 (float) | 24.58 TFLOPS | 4.067 TFLOPS |
| FP64 (double) | 384.0 GFLOPS | 1355.5 GFLOPS |
| Graphics Processor | ||
| GPU Name | MTT Gen 2 | GK110B |
| Architecture | MUSA-Chunxiao | Kepler |
| Foundry | TSMC | TSMC |
| Process Size | 12 nm | 28 nm |
| Transistors | 22 billion | 7.08 billion |
| Die Size | Unknown | 561 mm² |
| Board Design | ||
| TDP | 450W | 245W |
| Suggested PSU | 850 W | 550 W |
| Outputs | 4x DisplayPort 1.4a | No outputs |
| Power Connectors | 2x 8-pin | |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | 12 (11_1) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.4 | 1.2.175 |
| Shader Model | 6.5 | 6.5 (5.1) |
| CUDA | 3.5 | |