NVIDIA Quadro GV100 vs NVIDIA Tesla T10 Processor
Grafikprozessor
| GPU Name | GV100 | GT200B |
|---|---|---|
| GPU Variante | GV100-875-A1 | — |
| Architektur | Volta | Tesla 2.0 |
| Hersteller | TSMC | TSMC |
| Prozessgröße | 12 nm | 55 nm |
| Transistoren | 21,100 million | 1,400 million |
| Chipgröße | 815 mm² | 470 mm² |
Grafikkarte
| Veröffentlichungsdatum | Mar 27th, 2018 | Apr 9th, 2009 |
|---|---|---|
| Generation | Quadro | Tesla |
| Produktion | End-of-life | End-of-life |
| Originalpreis | 8,999 USD | — |
| Businterface | PCIe 3.0 x16 | PCIe 2.0 x16 |
Takt-Geschwindigkeiten
| Basis-Taktfrequenz | 1132 MHz | — |
|---|---|---|
| Boost-Takt | 1627 MHz | — |
| Speicher-Taktung | 848 MHz 1696 Mbps effective | 800 MHz 1600 Mbps effective |
| GPU-Taktung | — | 610 MHz |
| Shader-Taktung | — | 1296 MHz |
Speicher
| Speichergröße | 32 GB | 4 GB |
|---|---|---|
| Speichertyp | HBM2 | GDDR3 |
| Speicherbus | 4096 bit | 512 bit |
| Bandbreite | 868.4 GB/s | 102.4 GB/s |
Render-Konfiguration
| Shading-Einheiten | 5120 | 240 |
|---|---|---|
| TMUs | 320 | 80 |
| ROPs | 128 | 32 |
| Anzahl SM | 80 | 30 |
| Tensor Cores | 640 | — |
| Cache L1 | 128 KB (per SM) | — |
| Cache L2 | 6 MB | 256 KB |
Theoretische Leistung
| Pixelrate | 208.3 GPixel/s | 19.52 GPixel/s |
|---|---|---|
| Texturrate | 520.6 GTexel/s | 48.80 GTexel/s |
| FP16 (half) Leistung | 33.32 TFLOPS (2:1) | — |
| FP32 (float) Leistung | 16.66 TFLOPS | 622.1 GFLOPS |
| FP64 (double) Leistung | 8.330 TFLOPS (1:2) | 77.76 GFLOPS (1:8) |
Boarddesign
| Steckplatzbreite | Dual-slot | Dual-slot |
|---|---|---|
| Länge | 267 mm 10.5 inches | 267 mm 10.5 inches |
| Breite | 111 mm 4.4 inches | — |
| TDP | 250 W | 188 W |
| Vorgeschlagenes Netzteil | 600 W | 450 W |
| Ausgabe | 4x DisplayPort | No outputs |
| Stromverbindung | 8-pin EPS | — |
| Board-Nummer | PG500 SKU 500 | — |
Grafik-Features
| DirectX | 12 (12_1) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 3.0 | 1.1 |
| Vulkan | 1.2 | — |
| CUDA | 7.0 | 1.3 |
| Shader-Modell | 6.6 | 4.0 |
