Nom du GPU | M82 | GF108 |
---|---|---|
Architecture | TeraScale | Fermi |
Fonderie | TSMC | TSMC |
Taille du processus | 55 nm | 40 nm |
Transistors | 181 million | 585 million |
Taille des matrices | 67 mm² | 116 mm² |
Date de publication | Jan 7th, 2008 | May 4th, 2012 |
---|---|---|
Génération | M8x (Mobility HD 3400) | GeForce 600M |
Production | End-of-life | End-of-life |
Bus interface | MXM-II | PCIe 2.0 x16 |
Critiques | 2 in our database | 2 in our database |
Prédécesseur | M7x | GeForce 500M |
Successeur | M9x | GeForce 700M |
GPU clock | 680 MHz | 753 MHz |
---|---|---|
Mémoire clock | 400 MHz 800 Mbps effective | 785 MHz 3.1 Gbps effective |
Shader clock | — | 1505 MHz |
Taille de la mémoire | 256 MB | 2 GB |
---|---|---|
Type de mémoire | DDR2 | GDDR5 |
Bus mémoire | 64 bit | 128 bit |
Bande passante | 6.400 GB/s | 50.24 GB/s |
Shading units | 40 | 96 |
---|---|---|
TMUs | 4 | 16 |
ROPs | 4 | 4 |
Compute units | 2 | — |
Cache L2 | 64 KB | 256 KB |
Comptage SM | — | 2 |
Cache L1 | — | 64 KB (per SM) |
Taux de pixel | 2.720 GPixel/s | 3.012 GPixel/s |
---|---|---|
Taux de texture | 2.720 GTexel/s | 12.05 GTexel/s |
FP32 (float) performance | 54.40 GFLOPS | 289.0 GFLOPS |
FP64 (double) performance | — | 24.08 GFLOPS (1:12) |
Taille Slot | MXM Module | — |
---|---|---|
TDP | unknown | 32 W |
Sorties | No outputs | No outputs |
DirectX | 10.1 | 12 (11_0) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | — | 1.1 |
Vulkan | — | — |
Shader model | 4.1 | 5.1 |
CUDA | — | 2.1 |