Product Compare
AMD Radeon HD 6870 vs NVIDIA Tesla K40m
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Oct 2010 | Nov 2013 |
| Generation | Northern Islands | Tesla |
| Type | Desktop | Professional |
| Bus Interface | PCIe 2.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Memory Clock | 1050 MHz | 1502 MHz |
| Base Clock | 745 MHz | |
| Boost Clock | 876 MHz | |
| Memory | ||
| Memory Size | 1024MB | 12GB |
| Memory Type | GDDR5 | GDDR5 |
| Memory Bus | 256bit | 384bit |
| Bandwidth | 134.4GB/s | 288.4GB/s |
| Render Config | ||
| Compute Units | 14 | |
| Shading Units | 1120 | 2880 |
| TMUs | 56 | 240 |
| ROPs | 32 | 48 |
| L1 Cache | 8 KB (per CU) | 16 KB (per SMX) |
| L2 Cache | 512 KB | 1536 KB |
| Theoretical Performance | ||
| Pixel Rate | 28.80 GPixel/s | 52.56 GPixel/s |
| Texture Rate | 50.40 GTexel/s | 210.2 GTexel/s |
| FP32 (float) | 2.016 TFLOPS | 5.046 TFLOPS |
| FP64 (double) | 1.682 TFLOPS | |
| Graphics Processor | ||
| GPU Name | Barts | GK110B |
| GPU Variant | Barts XT(215-0798000) | |
| Architecture | TeraScale 2 | Kepler |
| Foundry | TSMC | TSMC |
| Process Size | 40 nm | 28 nm |
| Transistors | 1.7 billion | 7.08 billion |
| Die Size | 255 mm² | 561 mm² |
| Board Design | ||
| TDP | 151W | 245W |
| Suggested PSU | 450 W | 550 W |
| Outputs | 2x DVI1x HDMI 1.3a2x mini-DisplayPort 1.1 | No outputs |
| Power Connectors | 2x 6-pin | |
| Graphics Features | ||
| DirectX | 11.2 (11_0) | 12 (11_1) |
| OpenGL | 4.4 | 4.6 |
| OpenCL | 1.2 | 3.0 |
| Vulkan | N/A | 1.1 |
| Shader Model | 5.0 | 5.1 |
| CUDA | 3.5 | |