AMD Radeon Instinct MI25 vs NVIDIA Tesla P40
What is the difference between AMD Radeon Instinct MI25 and NVIDIA Tesla P40. Find out which graphics card has better performance.
Graphics Processor (GPU)
| Vega 10 | GPU Name | GP102 |
| GCN 5.0 | Architecture | Pascal |
| GlobalFoundries | Foundry | TSMC |
| 14 nm | Process Size | 16 nm |
| 12,500 million | Transistors | 11,800 million |
| 495 mm² | Die Size | 471 mm² |
Graphics Card
| Jun 27th, 2017 | Release Date | Sep 13th, 2016 |
| Radeon Instinct (MIx) | Family | Tesla (Pxx) |
| Active | Production | Active |
| PCIe 3.0 x16 | Bus Interface | PCIe 3.0 x16 |
Memory
| 16 GB | Memory Size | 24 GB |
| HBM2 | Memory Type | GDDR5X |
| 2048 bit | Memory Bus | 384 bit |
| 436.2 GB/s | Bandwidth | 694.3 GB/s |
Performance
| 96.00 GPixel/s | Pixel fillrate | 147.0 GPixel/s |
| 384.0 GTexel/s | Texture fillrate | 367.4 GTexel/s |
| 24.58 TFLOPS (2:1) | FP16 (half) performance | 183.7 GFLOPS (1:64) |
| 12.29 TFLOPS | FP32 (float) performance | 11.76 TFLOPS |
| 768.0 GFLOPS (1:16) | FP64 (double) performance | 367.4 GFLOPS (1:32) |
Clock Speeds
| 1400 MHz | Base Clock | 1303 MHz |
| 1500 MHz | Boost Clock | 1531 MHz |
| 852 MHz 1704 Mbps effective | Memory Clock | 1808 MHz 14.5 Gbps effective |
Render Config
| 4096 | Shading Units | 3840 |
| 256 | TMUs | 240 |
| 64 | ROPs | 96 |
| 16 KB (per CU) | L1 Cache | 48 KB (per SM) |
| 4 MB | L2 Cache | 3 MB |
Board Design
| Dual-slot | Slot Width | Dual-slot |
| 267 mm 10.5 inches | Length | 267 mm 10.5 inches |
| 111 mm 4.4 inches | Width | 111 mm 4.4 inches |
| 300 W | Thermal design power (TDP) | 250 W |
| 700 W | Suggested PSU | 600 W |
| No outputs | Display Connectors | No outputs |
| 2x 8-pin | Power Connectors | 1x 6-pin + 1x 8-pin |
| D051 | Board Number | PG610 SKU 200 |
API support
| 12 (12_1) | DirectX | 12 (12_1) |
| 4.6 | OpenGL | 4.6 |
| 2.1 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.2 |
| 6.4 | Shader Model | 6.4 |