Product Compare
NVIDIA B200 vs AMD Radeon HD 6870
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2024 | Oct 2010 |
| Generation | Server Blackwell | Northern Islands |
| Type | Desktop | Desktop |
| Bus Interface | PCIe 5.0 x16 | PCIe 2.0 x16 |
| Clock Speeds | ||
| Base Clock | 700 MHz | |
| Boost Clock | 1965 MHz | |
| Memory Clock | 2000 MHz | 1050 MHz |
| Memory | ||
| Memory Size | 90GB | 1024MB |
| Memory Type | HBM3e | GDDR5 |
| Memory Bus | 4096bit | 256bit |
| Bandwidth | 4.10TB/s | 134.4GB/s |
| Render Config | ||
| SM Count | 148 | |
| Shading Units | 18944 | 1120 |
| TMUs | 592 | 56 |
| ROPs | 24 | 32 |
| Tensor Cores | 592 | |
| L1 Cache | 256 KB (per SM) | 8 KB (per CU) |
| L2 Cache | 50 MB | 512 KB |
| Compute Units | 14 | |
| Theoretical Performance | ||
| Pixel Rate | 47.16 GPixel/s | 28.80 GPixel/s |
| Texture Rate | 1163.3 GTexel/s | 50.40 GTexel/s |
| FP16 (half) | 1191.2 TFLOPS | |
| FP32 (float) | 74.45 TFLOPS | 2.016 TFLOPS |
| FP64 (double) | 37.22 TFLOPS | |
| Graphics Processor | ||
| GPU Name | GB100 | Barts |
| Architecture | Blackwell | TeraScale 2 |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 40 nm |
| Transistors | 104 billion | 1.7 billion |
| Die Size | Unknown | 255 mm² |
| GPU Variant | Barts XT(215-0798000) | |
| Board Design | ||
| TDP | 1000W | 151W |
| Suggested PSU | 1400 W | 450 W |
| Outputs | No outputs | 2x DVI1x HDMI 1.3a2x mini-DisplayPort 1.1 |
| Power Connectors | 2x 6-pin | |
| Graphics Features | ||
| DirectX | N/A | 11.2 (11_0) |
| OpenGL | N/A | 4.4 |
| OpenCL | 3.0 | 1.2 |
| Vulkan | N/A | N/A |
| CUDA | 10.0 | |
| Shader Model | N/A | 5.0 |