NVIDIA Quadro P2000 vs NVIDIA Tesla V100S PCIe 32 GB
Grafikprozessor
| GPU Name | GP106 | GV100 |
|---|---|---|
| GPU Variante | GP106-875-A1 | — |
| Architektur | Pascal | Volta |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 16 nm | 12 nm |
| Transistoren | 4,400 million | 21,100 million |
| Chipgröße | 200 mm² | 815 mm² |
Grafikkarte
| Veröffentlichungsdatum | Feb 6th, 2017 | Nov 26th, 2019 |
|---|---|---|
| Generation | Quadro | Tesla |
| Produktion | End-of-life | End-of-life |
| Businterface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 1076 MHz | 1245 MHz |
|---|---|---|
| Boost-Takt | 1480 MHz | 1597 MHz |
| Speicher-Taktung | 1752 MHz 7 Gbps effective | 1106 MHz 2.2 Gbps effective |
Speicher
| Speichergröße | 5 GB | 32 GB |
|---|---|---|
| Speichertyp | GDDR5 | HBM2 |
| Speicherbus | 160 bit | 4096 bit |
| Bandbreite | 140.2 GB/s | 1,133 GB/s |
Render-Konfiguration
| Shading-Einheiten | 1024 | 5120 |
|---|---|---|
| TMUs | 64 | 320 |
| ROPs | 40 | 128 |
| Anzahl SM | 8 | 80 |
| Cache L1 | 48 KB (per SM) | 128 KB (per SM) |
| Cache L2 | 1280 KB | 6 MB |
| Tensor Cores | — | 640 |
Theoretische Leistung
| Pixelrate | 59.20 GPixel/s | 204.4 GPixel/s |
|---|---|---|
| Texturrate | 94.72 GTexel/s | 511.0 GTexel/s |
| FP16 (half) Leistung | 47.36 GFLOPS (1:64) | 32.71 TFLOPS (2:1) |
| FP32 (float) Leistung | 3.031 TFLOPS | 16.35 TFLOPS |
| FP64 (double) Leistung | 94.72 GFLOPS (1:32) | 8.177 TFLOPS (1:2) |
Boarddesign
| Steckplatzbreite | Single-slot | Dual-slot |
|---|---|---|
| Länge | 196 mm 7.7 inches | — |
| Breite | 111 mm 4.4 inches | — |
| TDP | 75 W | 250 W |
| Vorgeschlagenes Netzteil | 250 W | 600 W |
| Ausgabe | 4x DisplayPort | No outputs |
| Stromverbindung | None | 2x 8-pin |
| Board-Nummer | PG410 SKU 502 | PG500 |
Grafik-Features
| DirectX | 12 (12_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| CUDA | 6.1 | 7.0 |
| Shader-Modell | 6.4 | 6.6 |
