Product Compare
AMD Radeon HD 8870 OEM vs NVIDIA Tesla K40s
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2013 | Nov 2013 |
| Generation | Sea Islands | Tesla Kepler |
| Type | Desktop | Professional |
| Bus Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Memory Clock | 1200 MHz | 1502 MHz |
| Base Clock | 706 MHz | |
| Boost Clock | 706 MHz | |
| Memory | ||
| Memory Size | 2GB | 12GB |
| Memory Type | GDDR5 | GDDR5 |
| Memory Bus | 256bit | 384bit |
| Bandwidth | 153.6GB/s | 288.4GB/s |
| Render Config | ||
| Compute Units | 20 | |
| Shading Units | 1280 | 2880 |
| TMUs | 80 | 240 |
| ROPs | 32 | 48 |
| L1 Cache | 16 KB (per CU) | 16 KB (per SMX) |
| L2 Cache | 512 KB | 1536 KB |
| Theoretical Performance | ||
| Pixel Rate | 32.00 GPixel/s | 42.36 GPixel/s |
| Texture Rate | 80.00 GTexel/s | 169.4 GTexel/s |
| FP32 (float) | 2.560 TFLOPS | 4.067 TFLOPS |
| FP64 (double) | 160.0 GFLOPS | 1355.5 GFLOPS |
| Graphics Processor | ||
| GPU Name | Pitcairn | GK110B |
| GPU Variant | Pitcairn XT(215-0828047) | |
| Architecture | GCN 1.0 | Kepler |
| Foundry | TSMC | TSMC |
| Process Size | 28 nm | 28 nm |
| Transistors | 2.8 billion | 7.08 billion |
| Die Size | 212 mm² | 561 mm² |
| Board Design | ||
| TDP | 175W | 245W |
| Suggested PSU | 450 W | 550 W |
| Outputs | 1x DVI1x HDMI 1.4a2x mini-DisplayPort 1.2 | No outputs |
| Power Connectors | 2x 6-pin | |
| Graphics Features | ||
| DirectX | 12 (11_1) | 12 (11_1) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | 1.2 | 1.2.175 |
| Shader Model | 5.1 | 6.5 (5.1) |
| CUDA | 3.5 | |