AMD Radeon Instinct MI50 vs NVIDIA GRID A100A
What is the difference between AMD Radeon Instinct MI50 and NVIDIA GRID A100A. Find out which graphics card has better performance.
Graphics Processor (GPU)
| Vega 20 | GPU Name | GA100 |
| GCN 5.1 | Architecture | Ampere |
| TSMC | Foundry | TSMC |
| 7 nm | Process Size | 7 nm |
| 13,230 million | Transistors | 54,200 million |
| 331 mm² | Die Size | 826 mm² |
Graphics Card
| Nov 18th, 2018 | Release Date | May 14th, 2020 |
| Radeon Instinct (MIx) | Family | GRID (Ax) |
| Active | Production | Active |
| PCIe 4.0 x16 | Bus Interface | PCIe 4.0 x16 |
Memory
| 16 GB | Memory Size | 48 GB |
| HBM2 | Memory Type | HBM2e |
| 4096 bit | Memory Bus | 6144 bit |
| 1,024 GB/s | Bandwidth | 1,866 GB/s |
Performance
| 111.7 GPixel/s | Pixel fillrate | 193.0 GPixel/s |
| 419.0 GTexel/s | Texture fillrate | 434.2 GTexel/s |
| 26.82 TFLOPS (2:1) | FP16 (half) performance | 55.57 TFLOPS (4:1) |
| 13.41 TFLOPS | FP32 (float) performance | 13.89 TFLOPS |
| 6.705 TFLOPS (1:2) | FP64 (double) performance | 6.947 TFLOPS (1:2) |
Clock Speeds
| 1200 MHz | Base Clock | 900 MHz |
| 1746 MHz | Boost Clock | 1005 MHz |
| 1000 MHz 2 Gbps effective | Memory Clock | 1215 MHz 2.4 Gbps effective |
Render Config
| 3840 | Shading Units | 6912 |
| 240 | TMUs | 432 |
| 64 | ROPs | 192 |
| 16 KB (per CU) | L1 Cache | 192 KB (per SM) |
| 4 MB | L2 Cache | 48 MB |
Board Design
| Dual-slot | Slot Width | IGP |
| 300 W | Thermal design power (TDP) | 400 W |
| 700 W | Suggested PSU | 800 W |
| 1x mini-DisplayPort | Display Connectors | No outputs |
| 1x 6-pin + 1x 8-pin | Power Connectors | None |
API support
| 12 (12_1) | DirectX | N/A |
| 4.6 | OpenGL | N/A |
| 2.1 | OpenCL | 3.0 |
| 1.2 | Vulkan | N/A |
| 6.4 | Shader Model | N/A |