NVIDIA P106-100 vs NVIDIA Tesla P100 PCIe 16 GB
Processeur graphique
| Nom du GPU | GP106 | GP100 |
|---|---|---|
| GPU variant | GP106-100-A1 | GP100-893-A1 |
| Architecture | Pascal | Pascal |
| Fonderie | TSMC | TSMC |
| Taille du processus | 16 nm | 16 nm |
| Transistors | 4,400 million | 15,300 million |
| Taille des matrices | 200 mm² | 610 mm² |
Carte graphique
| Date de publication | Jun 19th, 2017 | Jun 20th, 2016 |
|---|---|---|
| Génération | Mining GPUs | Tesla |
| Production | End-of-life | End-of-life |
| Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
| Prix de lancement | — | 5,699 USD |
Vitesse clock
| Base clock | 1506 MHz | 1190 MHz |
|---|---|---|
| Boost clock | 1709 MHz | 1329 MHz |
| Mémoire clock | 2002 MHz 8 Gbps effective | 715 MHz 1430 Mbps effective |
Mémoire
| Taille de la mémoire | 6 GB | 16 GB |
|---|---|---|
| Type de mémoire | GDDR5 | HBM2 |
| Bus mémoire | 192 bit | 4096 bit |
| Bande passante | 192.2 GB/s | 732.2 GB/s |
Render Config
| Shading units | 1280 | 3584 |
|---|---|---|
| TMUs | 80 | 224 |
| ROPs | 48 | 96 |
| Comptage SM | 10 | 56 |
| Cache L1 | 48 KB (per SM) | 24 KB (per SM) |
| Cache L2 | 1536 KB | 4 MB |
Performance théorique
| Taux de pixel | 82.03 GPixel/s | 127.6 GPixel/s |
|---|---|---|
| Taux de texture | 136.7 GTexel/s | 297.7 GTexel/s |
| FP16 (half) performance | 68.36 GFLOPS (1:64) | 19.05 TFLOPS (2:1) |
| FP32 (float) performance | 4.375 TFLOPS | 9.526 TFLOPS |
| FP64 (double) performance | 136.7 GFLOPS (1:32) | 4.763 TFLOPS (1:2) |
Board Design
| Taille Slot | Dual-slot | Dual-slot |
|---|---|---|
| Longueur | 250 mm 9.8 inches | 267 mm 10.5 inches |
| TDP | 120 W | 250 W |
| Suggestions pour le PSU | 300 W | 600 W |
| Sorties | No outputs | No outputs |
| Connecteurs de puissance | 1x 6-pin | 1x 8-pin |
| Board number | PG410 SKU 100 | — |
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 |
