Nom du GPU | TU116 | GP102 |
---|---|---|
GPU variant | TU116-300-A1 | GP102-400-A1 |
Architecture | Turing | Pascal |
Fonderie | TSMC | TSMC |
Taille du processus | 12 nm | 16 nm |
Transistors | 6,600 million | 11,800 million |
Taille des matrices | 284 mm² | 471 mm² |
Date de publication | Oct 29th, 2019 | Aug 2nd, 2016 |
---|---|---|
Génération | GeForce 16 | GeForce 10 |
Prédécesseur | GeForce 10 | GeForce 900 |
Successeur | GeForce 20 | GeForce 20 |
Production | Active | End-of-life |
Prix de lancement | 229 USD | 1,199 USD |
Bus interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Critiques | 42 in our database | 15 in our database |
Base clock | 1530 MHz | 1417 MHz |
---|---|---|
Boost clock | 1785 MHz | 1531 MHz |
Mémoire clock | 1750 MHz 14 Gbps effective | 1251 MHz 10 Gbps effective |
Taille de la mémoire | 6 GB | 12 GB |
---|---|---|
Type de mémoire | GDDR6 | GDDR5X |
Bus mémoire | 192 bit | 384 bit |
Bande passante | 336.0 GB/s | 480.4 GB/s |
Shading units | 1408 | 3584 |
---|---|---|
TMUs | 88 | 224 |
ROPs | 48 | 96 |
Comptage SM | 22 | 28 |
Cache L1 | 64 KB (per SM) | 48 KB (per SM) |
Cache L2 | 1536 KB | 3 MB |
Taux de pixel | 85.68 GPixel/s | 147.0 GPixel/s |
---|---|---|
Taux de texture | 157.1 GTexel/s | 342.9 GTexel/s |
FP16 (half) performance | 10.05 TFLOPS (2:1) | 171.5 GFLOPS (1:64) |
FP32 (float) performance | 5.027 TFLOPS | 10.97 TFLOPS |
FP64 (double) performance | 157.1 GFLOPS (1:32) | 342.9 GFLOPS (1:32) |
Taille Slot | Dual-slot | Dual-slot |
---|---|---|
Longueur | 229 mm 9 inches | 267 mm 10.5 inches |
Taille | 111 mm 4.4 inches | 112 mm 4.4 inches |
Taille | 35 mm 1.4 inches | 40 mm 1.6 inches |
TDP | 125 W | 250 W |
Suggestions pour le PSU | 300 W | 600 W |
Sorties | 1x DVI1x HDMI1x DisplayPort | 1x DVI1x HDMI3x DisplayPort |
Connecteurs de puissance | 1x 8-pin | 1x 6-pin + 1x 8-pin |
Board number | PG161 SKU 26 | PG611 SKU 0 |
DirectX | 12 (12_1) | 12 (12_1) |
---|---|---|
OpenGL | 4.6 | 4.6 |
OpenCL | 3.0 | 3.0 |
Vulkan | 1.2 | 1.2 |
CUDA | 7.5 | 6.1 |
Shader model | 6.6 | 6.4 |