Nom du GPU | G70 | GP107 |
---|---|---|
Architecture | Curie | Pascal |
Fonderie | TSMC | Samsung |
Taille du processus | 110 nm | 14 nm |
Transistors | 302 million | 3,300 million |
Taille des matrices | 333 mm² | 132 mm² |
GPU variant | QDFX-4500-N-A2 | — |
Date de publication | Jul 28th, 2005 | Feb 7th, 2017 |
---|---|---|
Génération | Quadro FX | Quadro |
Production | End-of-life | End-of-life |
Prix de lancement | 2,499 USD | — |
Bus interface | PCIe 1.0 x16 | PCIe 3.0 x16 |
Critiques | 26 in our database | — |
GPU clock | 430 MHz | — |
---|---|---|
Mémoire clock | 525 MHz 1050 Mbps effective | 1002 MHz 4 Gbps effective |
Base clock | — | 1228 MHz |
Boost clock | — | 1252 MHz |
Taille de la mémoire | 512 MB | 2 GB |
---|---|---|
Type de mémoire | GDDR3 | GDDR5 |
Bus mémoire | 256 bit | 64 bit |
Bande passante | 33.60 GB/s | 32.06 GB/s |
Pixel shaders | 24 | — |
---|---|---|
Vertex shaders | 8 | — |
TMUs | 24 | 16 |
ROPs | 16 | 16 |
Shading units | — | 256 |
Comptage SM | — | 2 |
Cache L1 | — | 48 KB (per SM) |
Cache L2 | — | 512 KB |
Taux de pixel | 6.880 GPixel/s | 20.03 GPixel/s |
---|---|---|
Vertex rate | 860.0 MVertices/s | — |
Taux de texture | 10.32 GTexel/s | 20.03 GTexel/s |
FP16 (half) performance | — | 10.02 GFLOPS (1:64) |
FP32 (float) performance | — | 641.0 GFLOPS |
FP64 (double) performance | — | 20.03 GFLOPS (1:32) |
Taille Slot | Dual-slot | Single-slot |
---|---|---|
Longueur | 230 mm 9.1 inches | 150 mm 5.9 inches |
Taille | 111 mm 4.4 inches | 69 mm 2.7 inches |
TDP | 109 W | 30 W |
Suggestions pour le PSU | 300 W | 200 W |
Sorties | 2x DVI1x S-Video | 3x mini-DisplayPort |
Connecteurs de puissance | 1x 6-pin | None |
Board number | P347, P348 | PG212 SKU 500 |
Taille | 38 mm 1.5 inches | — |
DirectX | 9.0c (9_3) | 12 (12_1) |
---|---|---|
OpenGL | 2.1 | 4.6 |
OpenCL | — | 3.0 |
Vulkan | — | 1.2 |
Shader model | 3.0 | 6.4 |
CUDA | — | 6.1 |