AMD Radeon Pro Vega 20 vs NVIDIA Tesla K8
What is the difference between AMD Radeon Pro Vega 20 and NVIDIA Tesla K8. Find out which graphics card has better performance.
Graphics Processor (GPU)
| Vega 12 | GPU Name | GK104 |
| GCN 5.0 | Architecture | Kepler |
| GlobalFoundries | Foundry | TSMC |
| 14 nm | Process Size | 28 nm |
| unknown | Transistors | 3,540 million |
| unknown | Die Size | 294 mm² |
Graphics Card
| Nov 14th, 2018 | Release Date | Sep 16th, 2014 |
| Radeon Pro Mac (Vega Series) | Family | Tesla (Kxx) |
| Active | Production | End-of-life |
| PCIe 3.0 x16 | Bus Interface | PCIe 2.0 x16 |
Memory
| 4 GB | Memory Size | 8 GB |
| HBM2 | Memory Type | GDDR5 |
| 1024 bit | Memory Bus | 256 bit |
| 189.4 GB/s | Bandwidth | 160.0 GB/s |
Performance
| 41.06 GPixel/s | Pixel fillrate | 25.95 GPixel/s |
| 102.6 GTexel/s | Texture fillrate | 103.8 GTexel/s |
| 3.284 TFLOPS | FP32 (float) performance | 2.491 TFLOPS |
| 205.3 GFLOPS (1:16) | FP64 (double) performance | 103.8 GFLOPS (1:24) |
Clock Speeds
| 815 MHz | Base Clock | 693 MHz |
| 1283 MHz | Boost Clock | 811 MHz |
| 740 MHz 1480 Mbps effective | Memory Clock | 1250 MHz 5 Gbps effective |
Render Config
| 1280 | Shading Units | 1536 |
| 80 | TMUs | 128 |
| 32 | ROPs | 32 |
| 16 KB (per CU) | L1 Cache | 16 KB (per SMX) |
| 1024 KB | L2 Cache | 512 KB |
Board Design
| IGP | Slot Width | Single-slot |
| 100 W | Thermal design power (TDP) | 100 W |
| No outputs | Display Connectors | No outputs |
API support
| 12 (12_1) | DirectX | 12 (11_0) |
| 4.6 | OpenGL | 4.6 |
| 2.1 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.1 |
| 6.3 | Shader Model | 5.1 |