Nom du GPU | GP106 | GV100 |
---|---|---|
GPU variant | GP106-875-A1 | — |
Architecture | Pascal | Volta |
Fonderie | TSMC | TSMC |
Taille du processus | 16 nm | 12 nm |
Transistors | 4,400 million | 21,100 million |
Taille des matrices | 200 mm² | 815 mm² |
Date de publication | Feb 6th, 2017 | Jun 21st, 2017 |
---|---|---|
Génération | Quadro | Tesla |
Production | End-of-life | End-of-life |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Base clock | 1076 MHz | 1245 MHz |
---|---|---|
Boost clock | 1480 MHz | 1380 MHz |
Mémoire clock | 1752 MHz 7 Gbps effective | 876 MHz 1752 Mbps effective |
Taille de la mémoire | 5 GB | 16 GB |
---|---|---|
Type de mémoire | GDDR5 | HBM2 |
Bus mémoire | 160 bit | 4096 bit |
Bande passante | 140.2 GB/s | 897.0 GB/s |
Shading units | 1024 | 5120 |
---|---|---|
TMUs | 64 | 320 |
ROPs | 40 | 128 |
Comptage SM | 8 | 80 |
Cache L1 | 48 KB (per SM) | 128 KB (per SM) |
Cache L2 | 1280 KB | 6 MB |
Tensor cores | — | 640 |
Taux de pixel | 59.20 GPixel/s | 176.6 GPixel/s |
---|---|---|
Taux de texture | 94.72 GTexel/s | 441.6 GTexel/s |
FP16 (half) performance | 47.36 GFLOPS (1:64) | 28.26 TFLOPS (2:1) |
FP32 (float) performance | 3.031 TFLOPS | 14.13 TFLOPS |
FP64 (double) performance | 94.72 GFLOPS (1:32) | 7.066 TFLOPS (1:2) |
Taille Slot | Single-slot | Dual-slot |
---|---|---|
Longueur | 196 mm 7.7 inches | — |
Taille | 111 mm 4.4 inches | — |
TDP | 75 W | 300 W |
Suggestions pour le PSU | 250 W | 700 W |
Sorties | 4x DisplayPort | No outputs |
Connecteurs de puissance | None | 2x 8-pin |
Board number | PG410 SKU 502 | PG500 SKU 200 |
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 model | 6.4 | 6.6 |