Product Compare
AMD Radeon R9 M470 vs NVIDIA Tesla M40
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | May 2016 | Nov 2015 |
| Generation | Gem System | Tesla Maxwell |
| Type | Mobile | Professional |
| Bus Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Base Clock | 900 MHz | 948 MHz |
| Boost Clock | 1000 MHz | 1112 MHz |
| Memory Clock | 1375 MHz | 1502 MHz |
| Memory | ||
| Memory Size | 4GB | 12GB |
| Memory Type | GDDR5 | GDDR5 |
| Memory Bus | 128bit | 384bit |
| Bandwidth | 88.00GB/s | 288.4GB/s |
| Render Config | ||
| Compute Units | 12 | |
| Shading Units | 768 | 3072 |
| TMUs | 48 | 192 |
| ROPs | 16 | 96 |
| L1 Cache | 16 KB (per CU) | 48 KB (per SMM) |
| L2 Cache | 256 KB | 3 MB |
| Theoretical Performance | ||
| Pixel Rate | 16.00 GPixel/s | 106.8 GPixel/s |
| Texture Rate | 48.00 GTexel/s | 213.5 GTexel/s |
| FP32 (float) | 1.536 TFLOPS | 6.832 TFLOPS |
| FP64 (double) | 96.00 GFLOPS | 213.5 GFLOPS |
| Graphics Processor | ||
| GPU Name | Emerald | GM200 |
| GPU Variant | Emerald PRO | GM200-895-A1 |
| Architecture | GCN 2.0 | Maxwell 2.0 |
| Foundry | TSMC | TSMC |
| Process Size | 28 nm | 28 nm |
| Transistors | 2.08 billion | 8 billion |
| Die Size | 160 mm² | 601 mm² |
| Board Design | ||
| TDP | Unknown | 250W |
| Outputs | Portable Device Dependent | No outputs |
| Suggested PSU | 600 W | |
| Power Connectors | 8-pin EPS | |
| Graphics Features | ||
| DirectX | 12 (12_0) | 12 (12_1) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2.170 | 1.3 |
| Shader Model | 6.5 | 6.7 |
| CUDA | 5.2 | |