NVIDIA Tesla K20s vs AMD Radeon R9 M395X Mac Edition
What is the difference between NVIDIA Tesla K20s and AMD Radeon R9 M395X Mac Edition. Find out which graphics card has better performance.
Graphics Processor (GPU)
| GK110 | GPU Name | Amethyst |
| Kepler | Architecture | GCN 3.0 |
| TSMC | Foundry | TSMC |
| 28 nm | Process Size | 28 nm |
| 7,080 million | Transistors | 5,000 million |
| 561 mm² | Die Size | 366 mm² |
Graphics Card
| Feb 18th, 2013 | Release Date | May 5th, 2015 |
| Tesla (Kxx) | Family | Crystal System (R9 M300) |
| End-of-life | Production | Active |
| PCIe 2.0 x16 | Bus Interface | MXM-B (3.0) |
Memory
| 5 GB | Memory Size | 4 GB |
| GDDR5 | Memory Type | GDDR5 |
| 320 bit | Memory Bus | 256 bit |
| 208.0 GB/s | Bandwidth | 174.7 GB/s |
Performance
| 29.90 GPixel/s | Pixel fillrate | 29.09 GPixel/s |
| 119.6 GTexel/s | Texture fillrate | 116.4 GTexel/s |
| 2.870 TFLOPS | FP32 (float) performance | 3.723 TFLOPS |
| 956.8 GFLOPS (1:3) | FP64 (double) performance | 232.7 GFLOPS (1:16) |
Clock Speeds
| 1300 MHz 5.2 Gbps effective | Memory Clock | 1365 MHz 5.5 Gbps effective |
| 575 MHz | GPU Clock | 909 MHz |
Render Config
| 2496 | Shading Units | 2048 |
| 208 | TMUs | 128 |
| 40 | ROPs | 32 |
| 16 KB (per SMX) | L1 Cache | 16 KB (per CU) |
| 1280 KB | L2 Cache | 512 KB |
Board Design
| Dual-slot | Slot Width | MXM Module |
| 225 W | Thermal design power (TDP) | 250 W |
| No outputs | Display Connectors | No outputs |
API support
| 12 (11_0) | DirectX | 12 (12_0) |
| 4.6 | OpenGL | 4.6 |
| 3.0 | OpenCL | 2.0 |
| 1.1 | Vulkan | 1.2 |
| 5.1 | Shader Model | 6.3 |