Nom du GPU | G92 | GF108 |
---|---|---|
GPU variant | NB8E-GT-A2 | N11P-GE-A1 |
Architecture | Tesla | Fermi |
Fonderie | TSMC | TSMC |
Taille du processus | 65 nm | 40 nm |
Transistors | 754 million | 585 million |
Taille des matrices | 324 mm² | 116 mm² |
Date de publication | Nov 1st, 2007 | Sep 3rd, 2010 |
---|---|---|
Génération | GeForce 8M (8000M) | GeForce 400M |
Production | End-of-life | End-of-life |
Bus interface | MXM-HE | PCIe 2.0 x16 |
Prédécesseur | GeForce Go 7 | GeForce 300M |
Successeur | GeForce 9M | GeForce 500M |
GPU clock | 500 MHz | 500 MHz |
---|---|---|
Shader clock | 1250 MHz | 1000 MHz |
Mémoire clock | 800 MHz 1600 Mbps effective | 800 MHz 1600 Mbps effective |
Taille de la mémoire | 512 MB | 512 MB |
---|---|---|
Type de mémoire | GDDR3 | DDR3 |
Bus mémoire | 256 bit | 128 bit |
Bande passante | 51.20 GB/s | 25.60 GB/s |
Shading units | 64 | 96 |
---|---|---|
TMUs | 32 | 16 |
ROPs | 16 | 4 |
Comptage SM | 8 | 2 |
Cache L2 | 64 KB | 256 KB |
Cache L1 | — | 64 KB (per SM) |
Taux de pixel | 8.000 GPixel/s | 2.000 GPixel/s |
---|---|---|
Taux de texture | 16.00 GTexel/s | 8.000 GTexel/s |
FP32 (float) performance | 160.0 GFLOPS | 192.0 GFLOPS |
FP64 (double) performance | — | 16.00 GFLOPS (1:12) |
Taille Slot | MXM Module | — |
---|---|---|
TDP | 50 W | 23 W |
Sorties | No outputs | No outputs |
DirectX | 11.1 (10_0) | 12 (11_0) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 | 1.1 |
Vulkan | — | — |
CUDA | 1.1 | 2.1 |
Shader model | 4.0 | 5.1 |