Nom du GPU | GM108 | NV37 |
---|---|---|
GPU variant | N16S-GMR | NV37 GL |
Architecture | Maxwell | Rankine |
Fonderie | TSMC | TSMC |
Taille du processus | 28 nm | 150 nm |
Transistors | unknown | 45 million |
Taille des matrices | unknown | 91 mm² |
Date de publication | Mar 1st, 2016 | — |
---|---|---|
Génération | GeForce 900M | — |
Production | End-of-life | — |
Bus interface | PCIe 3.0 x8 | — |
Prédécesseur | GeForce 800M | — |
Successeur | GeForce 10 Mobile | — |
Base clock | 952 MHz | — |
---|---|---|
Boost clock | 1020 MHz | — |
Mémoire clock | 900 MHz 1800 Mbps effective | 250 MHz 500 Mbps effective |
GPU clock | — | 275 MHz |
Taille de la mémoire | 2 GB | 64 MB |
---|---|---|
Type de mémoire | DDR3 | DDR |
Bus mémoire | 64 bit | 128 bit |
Bande passante | 14.40 GB/s | 8.000 GB/s |
Shading units | 384 | — |
---|---|---|
TMUs | 24 | 4 |
ROPs | 8 | 4 |
Comptage SMM | 3 | — |
Cache L1 | 64 KB (per SMM) | — |
Cache L2 | 1024 KB | — |
Pixel shaders | — | 4 |
Vertex shaders | — | 2 |
Taux de pixel | 8.160 GPixel/s | 1.100 GPixel/s |
---|---|---|
Taux de texture | 24.48 GTexel/s | 1.100 GTexel/s |
FP32 (float) performance | 783.4 GFLOPS | — |
FP64 (double) performance | 24.48 GFLOPS (1:32) | — |
Vertex rate | — | 137.5 MVertices/s |
Taille Slot | MXM Module | Single-slot |
---|---|---|
TDP | 17 W | 13 W |
Sorties | No outputs | 1x DMS-59 |
Connecteurs de puissance | None | None |
Longueur | — | 168 mm 6.6 inches |
Suggestions pour le PSU | — | 200 W |
DirectX | 12 (11_0) | 9.0a |
---|---|---|
OpenGL | 4.6 | 1.5 (2.1) |
OpenCL | 3.0 | — |
Vulkan | 1.1 | — |
CUDA | 5.0 | — |
Shader model | 5.1 | — |
Pixel shader | — | 2.0a |
Vertex shader | — | 2.0a |
Date de publication | — | May 25th, 2004 |
---|---|---|
Génération | — | Quadro NVS |
Production | — | End-of-life |
Bus interface | — | PCIe 1.0 x16 |
Critiques | — | 26 in our database |