AMD Radeon Pro WX Vega M GL vs NVIDIA Quadro P1000 Mobile
Grafikprozessor
| GPU Name | Polaris 22 | GP107 |
|---|---|---|
| GPU Variante | VegaM MGL XL | — |
| Architektur | GCN 4.0 | Pascal |
| Hersteller | GlobalFoundries | Samsung |
| Prozessgröße | 14 nm | 14 nm |
| Transistoren | 5,000 million | 3,300 million |
| Chipgröße | 208 mm² | 132 mm² |
Mobile Grafik
| Veröffentlichungsdatum | Apr 24th, 2018 | Feb 7th, 2017 |
|---|---|---|
| Generation | Vega (Vega M) | Quadro Mobile (Px000) |
| Produktion | Active | End-of-life |
| Businterface | IGP | PCIe 3.0 x16 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 931 MHz | 1303 MHz |
|---|---|---|
| Boost-Takt | 1011 MHz | 1519 MHz |
| Speicher-Taktung | 700 MHz 1400 Mbps effective | 1502 MHz 6 Gbps effective |
Speicher
| Speichergröße | 4 GB | 4 GB |
|---|---|---|
| Speichertyp | HBM2 | GDDR5 |
| Speicherbus | 1024 bit | 128 bit |
| Bandbreite | 179.2 GB/s | 96.13 GB/s |
Render-Konfiguration
| Shading-Einheiten | 1280 | 512 |
|---|---|---|
| TMUs | 80 | 32 |
| ROPs | 32 | 16 |
| Compute Units | 20 | — |
| Cache L1 | 16 KB (per CU) | 48 KB (per SM) |
| Cache L2 | 1024 KB | 1024 KB |
| Anzahl SM | — | 4 |
Theoretische Leistung
| Pixelrate | 32.35 GPixel/s | 24.30 GPixel/s |
|---|---|---|
| Texturrate | 80.88 GTexel/s | 48.61 GTexel/s |
| FP16 (half) Leistung | 2.588 TFLOPS (1:1) | 24.30 GFLOPS (1:64) |
| FP32 (float) Leistung | 2.588 TFLOPS | 1.555 TFLOPS |
| FP64 (double) Leistung | 161.8 GFLOPS (1:16) | 48.61 GFLOPS (1:32) |
Boarddesign
| Steckplatzbreite | IGP | — |
|---|---|---|
| TDP | 65 W | 40 W |
| Ausgabe | No outputs | No outputs |
| Board-Nummer | D136 | E2904 SKU 1 |
| Stromverbindung | — | None |
Grafik-Features
| DirectX | 12 (12_0) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader-Modell | 6.4 | 6.4 |
| CUDA | — | 6.1 |
