AMD Radeon Pro Vega 20 vs NVIDIA Quadro P3200 Max-Q
Processeur graphique
| Nom du GPU | Vega 12 | GP104 |
|---|---|---|
| GPU variant | Vega 12 XTA | — |
| Architecture | GCN 5.0 | Pascal |
| Fonderie | GlobalFoundries | TSMC |
| Taille du processus | 14 nm | 16 nm |
| Transistors | unknown | 7,200 million |
| Taille des matrices | unknown | 314 mm² |
Graphiques mobiles
| Date de publication | Nov 14th, 2018 | Feb 21st, 2018 |
|---|---|---|
| Génération | Radeon Pro Mac (Vega Series) | Quadro Mobile (Px200) |
| Production | Active | End-of-life |
| Bus interface | PCIe 3.0 x16 | MXM-B (3.0) |
Vitesse clock
| Base clock | 815 MHz | 1139 MHz |
|---|---|---|
| Boost clock | 1283 MHz | 1404 MHz |
| Mémoire clock | 740 MHz 1480 Mbps effective | 1753 MHz 7 Gbps effective |
Mémoire
| Taille de la mémoire | 4 GB | 6 GB |
|---|---|---|
| Type de mémoire | HBM2 | GDDR5 |
| Bus mémoire | 1024 bit | 192 bit |
| Bande passante | 189.4 GB/s | 168.3 GB/s |
Render Config
| Shading units | 1280 | 1792 |
|---|---|---|
| TMUs | 80 | 112 |
| ROPs | 32 | 64 |
| Compute units | 20 | — |
| Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
| Cache L2 | 1024 KB | 1536 KB |
| Comptage SM | — | 14 |
Performance théorique
| Taux de pixel | 41.06 GPixel/s | 89.86 GPixel/s |
|---|---|---|
| Taux de texture | 102.6 GTexel/s | 157.2 GTexel/s |
| FP16 (half) performance | 6.569 TFLOPS (2:1) | 78.62 GFLOPS (1:64) |
| FP32 (float) performance | 3.284 TFLOPS | 5.032 TFLOPS |
| FP64 (double) performance | 205.3 GFLOPS (1:16) | 157.2 GFLOPS (1:32) |
Board Design
| Taille Slot | IGP | MXM Module |
|---|---|---|
| TDP | 100 W | 75 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | — | None |
Caractéristiques graphiques
| DirectX | 12 (12_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.3 | 6.4 |
| CUDA | — | 6.1 |
