Product Compare
NVIDIA B200 vs NVIDIA CMP 40HX
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2024 | Feb 2021 |
| Generation | Server Blackwell | Mining GPUs |
| Type | Desktop | Desktop |
| Bus Interface | PCIe 5.0 x16 | PCIe 1.0 x4 |
| Clock Speeds | ||
| Base Clock | 700 MHz | 1470 MHz |
| Boost Clock | 1965 MHz | 1650 MHz |
| Memory Clock | 2000 MHz | 1750 MHz |
| Memory | ||
| Memory Size | 90GB | 8GB |
| Memory Type | HBM3e | GDDR6 |
| Memory Bus | 4096bit | 256bit |
| Bandwidth | 4.10TB/s | 448.0GB/s |
| Render Config | ||
| SM Count | 148 | 36 |
| Shading Units | 18944 | 2304 |
| TMUs | 592 | 144 |
| ROPs | 24 | 64 |
| Tensor Cores | 592 | 288 |
| L1 Cache | 256 KB (per SM) | 64 KB (per SM) |
| L2 Cache | 50 MB | 4 MB |
| RT Cores | 36 | |
| Theoretical Performance | ||
| Pixel Rate | 47.16 GPixel/s | 105.6 GPixel/s |
| Texture Rate | 1163.3 GTexel/s | 237.6 GTexel/s |
| FP16 (half) | 1191.2 TFLOPS | 15.21 TFLOPS |
| FP32 (float) | 74.45 TFLOPS | 7.603 TFLOPS |
| FP64 (double) | 37.22 TFLOPS | 237.6 GFLOPS |
| Graphics Processor | ||
| GPU Name | GB100 | TU106 |
| Architecture | Blackwell | Turing |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 12 nm |
| Transistors | 104 billion | 10.8 billion |
| Die Size | Unknown | 445 mm² |
| GPU Variant | TU106-100-A1 | |
| Board Design | ||
| TDP | 1000W | 185W |
| Suggested PSU | 1400 W | 450 W |
| Outputs | No outputs | No outputs |
| Power Connectors | 1x 8-pin | |
| Graphics Features | ||
| DirectX | N/A | 12 Ultimate (12_2) |
| OpenGL | N/A | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | N/A | 1.3 |
| CUDA | 10.0 | 7.5 |
| Shader Model | N/A | 6.8 |