AMD Radeon Instinct MI50 vs NVIDIA GRID RTX T10-16
What is the difference between AMD Radeon Instinct MI50 and NVIDIA GRID RTX T10-16. Find out which graphics card has better performance.
Graphics Processor (GPU)
| Vega 20 | GPU Name | TU102 |
| Vega 20 GLXT | Model number | TU102-875-A1 |
| GCN 5.1 | Architecture | Turing |
| TSMC | Foundry | TSMC |
| 7 nm | Process Size | 12 nm |
| 13,230 million | Transistors | 18,600 million |
| 331 mm² | Die Size | 754 mm² |
Graphics Card
| Nov 18th, 2018 | Release Date | Unknown |
| Radeon Instinct (MIx) | Family | GRID (Tx) |
| Active | Production | Active |
| PCIe 4.0 x16 | Bus Interface | PCIe 3.0 x16 |
Memory
| 16 GB | Memory Size | 16 GB |
| HBM2 | Memory Type | GDDR6 |
| 4096 bit | Memory Bus | 384 bit |
| 1,024 GB/s | Bandwidth | 604.8 GB/s |
Performance
| 111.7 GPixel/s | Pixel fillrate | 133.9 GPixel/s |
| 419.0 GTexel/s | Texture fillrate | 401.8 GTexel/s |
| 26.82 TFLOPS (2:1) | FP16 (half) performance | 25.71 TFLOPS (2:1) |
| 13.41 TFLOPS | FP32 (float) performance | 12.86 TFLOPS |
| 6.705 TFLOPS (1:2) | FP64 (double) performance | 401.8 GFLOPS (1:32) |
Clock Speeds
| 1200 MHz | Base Clock | 1065 MHz |
| 1746 MHz | Boost Clock | 1395 MHz |
| 1000 MHz 2 Gbps effective | Memory Clock | 1575 MHz 12.6 Gbps effective |
Render Config
| 3840 | Shading Units | 4608 |
| 240 | TMUs | 288 |
| 64 | ROPs | 96 |
| 16 KB (per CU) | L1 Cache | 64 KB (per SM) |
| 4 MB | L2 Cache | 6 MB |
Board Design
| Dual-slot | Slot Width | Dual-slot |
| 267 mm 10.5 inches | Length | 267 mm 10.5 inches |
| 300 W | Thermal design power (TDP) | 260 W |
| 700 W | Suggested PSU | 600 W |
| 1x mini-DisplayPort | Display Connectors | No outputs |
| 1x 6-pin + 1x 8-pin | Power Connectors | 1x 6-pin + 1x 8-pin |
API support
| 12 (12_1) | DirectX | 12 Ultimate (12_2) |
| 4.6 | OpenGL | 4.6 |
| 2.1 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.2 |
| 6.4 | Shader Model | 6.6 |