Nom du GPU | NV40 | GF108 |
---|---|---|
Architecture | Curie | Fermi |
Fonderie | TSMC | TSMC |
Taille du processus | 130 nm | 40 nm |
Transistors | 222 million | 585 million |
Taille des matrices | 287 mm² | 116 mm² |
GPU variant | — | GF108-400-A1 |
Date de publication | Apr 14th, 2004 | Jun 18th, 2014 |
---|---|---|
Génération | GeForce 6 AGP | GeForce 700 |
Production | End-of-life | End-of-life |
Bus interface | AGP 8x | PCIe 2.0 x16 |
Critiques | 115 in our database | — |
Prédécesseur | GeForce FX | GeForce 600 |
Successeur | GeForce 7 AGP | GeForce 900 |
GPU clock | 325 MHz | 700 MHz |
---|---|---|
Mémoire clock | 350 MHz 700 Mbps effective | 800 MHz 1600 Mbps effective |
Shader clock | — | 1400 MHz |
Taille de la mémoire | 128 MB | 2 GB |
---|---|---|
Type de mémoire | DDR | DDR3 |
Bus mémoire | 256 bit | 128 bit |
Bande passante | 22.40 GB/s | 25.60 GB/s |
Pixel shaders | 12 | — |
---|---|---|
Vertex shaders | 5 | — |
TMUs | 12 | 16 |
ROPs | 12 | 4 |
Shading units | — | 96 |
Comptage SM | — | 2 |
Cache L1 | — | 64 KB (per SM) |
Cache L2 | — | 256 KB |
Taux de pixel | 3.900 GPixel/s | 2.800 GPixel/s |
---|---|---|
Vertex rate | 406.3 MVertices/s | — |
Taux de texture | 3.900 GTexel/s | 11.20 GTexel/s |
FP32 (float) performance | — | 268.8 GFLOPS |
FP64 (double) performance | — | 22.40 GFLOPS (1:12) |
Taille Slot | Single-slot | Single-slot |
---|---|---|
TDP | unknown | 49 W |
Suggestions pour le PSU | 200 W | 200 W |
Sorties | 1x DVI1x VGA1x S-Video | 1x DVI1x HDMI1x VGA |
Connecteurs de puissance | 1x Molex | None |
Board number | P201, P212 | P1070, P1071 |
Longueur | — | 145 mm 5.7 inches |
DirectX | 9.0c (9_3) | 12 (11_0) |
---|---|---|
OpenGL | 2.1 | 4.6 |
OpenCL | — | 1.1 |
Vulkan | — | — |
Shader model | 3.0 | 5.1 |
CUDA | — | 2.1 |