Product Compare
NVIDIA GRID K200 vs NVIDIA CMP 50HX
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jun 2013 | Jun 2021 |
| Generation | GRID | Mining GPUs |
| Type | Professional | Desktop |
| Bus Interface | PCIe 3.0 x16 | PCIe 1.0 x4 |
| Clock Speeds | ||
| Memory Clock | 1250 MHz | 1750 MHz |
| Base Clock | 1350 MHz | |
| Boost Clock | 1545 MHz | |
| Memory | ||
| Memory Size | 256MB | 10GB |
| Memory Type | GDDR5 | GDDR6 |
| Memory Bus | 256bit | 320bit |
| Bandwidth | 160.0GB/s | 560.0GB/s |
| Render Config | ||
| Shading Units | 1536 | 3584 |
| TMUs | 128 | 192 |
| ROPs | 32 | 80 |
| L1 Cache | 16 KB (per SMX) | 64 KB (per SM) |
| L2 Cache | 512 KB | 5 MB |
| SM Count | 56 | |
| Tensor Cores | 448 | |
| RT Cores | 56 | |
| Theoretical Performance | ||
| Pixel Rate | 23.84 GPixel/s | 123.6 GPixel/s |
| Texture Rate | 95.36 GTexel/s | 296.6 GTexel/s |
| FP32 (float) | 2.289 TFLOPS | 11.07 TFLOPS |
| FP64 (double) | 95.36 GFLOPS | 346.1 GFLOPS |
| FP16 (half) | 22.15 TFLOPS | |
| Graphics Processor | ||
| GPU Name | GK104 | TU102 |
| Architecture | Kepler | Turing |
| Foundry | TSMC | TSMC |
| Process Size | 28 nm | 12 nm |
| Transistors | 3.54 billion | 18.6 billion |
| Die Size | 294 mm² | 754 mm² |
| GPU Variant | TU102-100-A1 | |
| Board Design | ||
| TDP | 225W | 250W |
| Suggested PSU | 550 W | 600 W |
| Outputs | No outputs | No outputs |
| Power Connectors | 2x 8-pin | |
| Graphics Features | ||
| DirectX | 12 (11_0) | 12 Ultimate (12_2) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.1 | 1.3 |
| CUDA | 3.0 | 7.5 |
| Shader Model | 5.1 | 6.8 |