Nom du GPU | GP102 | TU102 |
---|---|---|
GPU variant | GP102-875-A1 | TU102-875-A1 |
Architecture | Pascal | Turing |
Fonderie | TSMC | TSMC |
Taille du processus | 16 nm | 12 nm |
Transistors | 11,800 million | 18,600 million |
Taille des matrices | 471 mm² | 754 mm² |
Date de publication | Oct 1st, 2016 | Aug 13th, 2018 |
---|---|---|
Génération | Quadro | Quadro |
Production | End-of-life | Active |
Prix de lancement | 5,999 USD | 6,299 USD |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Base clock | 1506 MHz | 1305 MHz |
---|---|---|
Boost clock | 1645 MHz | 1560 MHz |
Mémoire clock | 1127 MHz 9 Gbps effective | 1750 MHz 14 Gbps effective |
Taille de la mémoire | 24 GB | 24 GB |
---|---|---|
Type de mémoire | GDDR5X | GDDR6 |
Bus mémoire | 384 bit | 384 bit |
Bande passante | 432.8 GB/s | 672.0 GB/s |
Shading units | 3840 | 4608 |
---|---|---|
TMUs | 240 | 288 |
ROPs | 96 | 96 |
Comptage SM | 30 | 72 |
Cache L1 | 48 KB (per SM) | 64 KB (per SM) |
Cache L2 | 3 MB | 6 MB |
Tensor cores | — | 576 |
RT cores | — | 72 |
Taux de pixel | 157.9 GPixel/s | 149.8 GPixel/s |
---|---|---|
Taux de texture | 394.8 GTexel/s | 449.3 GTexel/s |
FP16 (half) performance | 197.4 GFLOPS (1:64) | 28.75 TFLOPS (2:1) |
FP32 (float) performance | 12.63 TFLOPS | 14.38 TFLOPS |
FP64 (double) performance | 394.8 GFLOPS (1:32) | 449.3 GFLOPS (1:32) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 267 mm 10.5 inches | 267 mm 10.5 inches |
Taille | 111 mm 4.4 inches | 111 mm 4.4 inches |
TDP | 250 W | 260 W |
Suggestions pour le PSU | 600 W | 600 W |
Sorties | 1x DVI4x DisplayPort | No outputs |
Connecteurs de puissance | 1x 6-pin + 1x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | PG611 SKU 500 | PG150 SKU 510 |
DirectX | 12 (12_1) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 6.1 | 7.5 |
Shader model | 6.4 | 6.6 |