NVIDIA Quadro P1000 vs NVIDIA Tesla P100 DGXS
Processeur graphique
| Nom du GPU | GP107 | GP100 |
|---|---|---|
| GPU variant | GP107-860-A1 | — |
| Architecture | Pascal | Pascal |
| Fonderie | Samsung | TSMC |
| Taille du processus | 14 nm | 16 nm |
| Transistors | 3,300 million | 15,300 million |
| Taille des matrices | 132 mm² | 610 mm² |
Carte graphique
| Date de publication | Feb 7th, 2017 | Apr 5th, 2016 |
|---|---|---|
| Génération | Quadro | Tesla |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Vitesse clock
| Base clock | 1354 MHz | 1328 MHz |
|---|---|---|
| Boost clock | 1392 MHz | 1480 MHz |
| Mémoire clock | 1502 MHz 6 Gbps effective | 715 MHz 1430 Mbps effective |
Mémoire
| Taille de la mémoire | 4 GB | 16 GB |
|---|---|---|
| Type de mémoire | GDDR5 | HBM2 |
| Bus mémoire | 128 bit | 4096 bit |
| Bande passante | 96.13 GB/s | 732.2 GB/s |
Render Config
| Shading units | 512 | 3584 |
|---|---|---|
| TMUs | 32 | 224 |
| ROPs | 16 | 96 |
| Comptage SM | 4 | 56 |
| Cache L1 | 48 KB (per SM) | 24 KB (per SM) |
| Cache L2 | 1024 KB | 4 MB |
Performance théorique
| Taux de pixel | 22.27 GPixel/s | 142.1 GPixel/s |
|---|---|---|
| Taux de texture | 44.54 GTexel/s | 331.5 GTexel/s |
| FP16 (half) performance | 22.27 GFLOPS (1:64) | 21.22 TFLOPS (2:1) |
| FP32 (float) performance | 1,425 GFLOPS | 10.61 TFLOPS |
| FP64 (double) performance | 44.54 GFLOPS (1:32) | 5.304 TFLOPS (1:2) |
Board Design
| Taille Slot | Single-slot | — |
|---|---|---|
| Longueur | 150 mm 5.9 inches | — |
| Taille | 69 mm 2.7 inches | — |
| TDP | 47 W | 300 W |
| Suggestions pour le PSU | 200 W | 700 W |
| Sorties | 4x mini-DisplayPort | No outputs |
| Connecteurs de puissance | None | None |
| Board number | PG212 SKU 502 | — |
Caractéristiques graphiques
| 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 | 6.0 |
| Shader model | 6.4 | 6.4 |
