Nom du GPU | G96 | GK107 |
---|---|---|
GPU variant | G96-259-A1 | — |
Architecture | Tesla | Kepler |
Fonderie | UMC | TSMC |
Taille du processus | 65 nm | 28 nm |
Transistors | 314 million | 1,270 million |
Taille des matrices | 144 mm² | 118 mm² |
Date de publication | Jul 29th, 2008 | Never Released |
---|---|---|
Génération | GeForce 9 | NVS |
Production | End-of-life | End-of-life |
Bus interface | PCIe 2.0 x16 | PCIe 3.0 x16 |
Critiques | 22 in our database | 26 in our database |
Prédécesseur | GeForce 8 | — |
Successeur | GeForce 200 | — |
GPU clock | 550 MHz | 797 MHz |
---|---|---|
Shader clock | 1375 MHz | — |
Mémoire clock | 504 MHz 1008 Mbps effective | 891 MHz 1782 Mbps effective |
Taille de la mémoire | 512 MB | 2 GB |
---|---|---|
Type de mémoire | DDR2 | DDR3 |
Bus mémoire | 128 bit | 128 bit |
Bande passante | 16.13 GB/s | 28.51 GB/s |
Shading units | 32 | 192 |
---|---|---|
TMUs | 16 | 16 |
ROPs | 8 | 16 |
Comptage SM | 4 | — |
Cache L2 | 32 KB | 256 KB |
Comptage SMX | — | 1 |
Cache L1 | — | 16 KB (per SMX) |
Taux de pixel | 4.400 GPixel/s | 3.188 GPixel/s |
---|---|---|
Taux de texture | 8.800 GTexel/s | 12.75 GTexel/s |
FP32 (float) performance | 88.00 GFLOPS | 306.0 GFLOPS |
FP64 (double) performance | — | 12.75 GFLOPS (1:24) |
Taille Slot | Single-slot | Single-slot |
---|---|---|
TDP | 40 W | 35 W |
Suggestions pour le PSU | 200 W | 200 W |
Sorties | 2x DVI1x S-Video | 4x mini-DisplayPort |
Connecteurs de puissance | None | — |
Longueur | — | 160 mm 6.3 inches |
DirectX | 11.1 (10_0) | 12 (11_0) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 | 3.0 |
Vulkan | — | 1.1 |
CUDA | 1.1 | 3.0 |
Shader model | 4.0 | 5.1 |