Nom du GPU | G73 | GP104 |
---|---|---|
GPU variant | G73-N-B1 | GP104-150-KA-A1 |
Architecture | Curie | Pascal |
Fonderie | TSMC | TSMC |
Taille du processus | 90 nm | 16 nm |
Transistors | 177 million | 7,200 million |
Taille des matrices | 125 mm² | 314 mm² |
Date de publication | Mar 22nd, 2006 | Oct 18th, 2018 |
---|---|---|
Génération | GeForce 7 PCIe | GeForce 10 |
Production | End-of-life | End-of-life |
Bus interface | PCIe 1.0 x16 | PCIe 3.0 x16 |
Critiques | 83 in our database | 117 in our database |
Prédécesseur | GeForce 6 PCIe | GeForce 900 |
Successeur | GeForce 8 | GeForce 20 |
GPU clock | 400 MHz | — |
---|---|---|
Mémoire clock | 400 MHz 800 Mbps effective | 1001 MHz 8 Gbps effective |
Base clock | — | 1506 MHz |
Boost clock | — | 1709 MHz |
Taille de la mémoire | 256 MB | 6 GB |
---|---|---|
Type de mémoire | DDR2 | GDDR5X |
Bus mémoire | 128 bit | 192 bit |
Bande passante | 12.80 GB/s | 192.2 GB/s |
Pixel shaders | 12 | — |
---|---|---|
Vertex shaders | 5 | — |
TMUs | 12 | 80 |
ROPs | 8 | 48 |
Shading units | — | 1280 |
Comptage SM | — | 10 |
Cache L1 | — | 48 KB (per SM) |
Cache L2 | — | 1536 KB |
Taux de pixel | 3.200 GPixel/s | 82.03 GPixel/s |
---|---|---|
Vertex rate | 500.0 MVertices/s | — |
Taux de texture | 4.800 GTexel/s | 136.7 GTexel/s |
FP16 (half) performance | — | 68.36 GFLOPS (1:64) |
FP32 (float) performance | — | 4.375 TFLOPS |
FP64 (double) performance | — | 136.7 GFLOPS (1:32) |
Taille Slot | Single-slot | Dual-slot |
---|---|---|
TDP | 27 W | 120 W |
Suggestions pour le PSU | 200 W | 300 W |
Sorties | 1x DVI1x VGA1x S-Video | 1x DVI1x HDMI3x DisplayPort |
Connecteurs de puissance | None | 1x 6-pin |
Board number | P501, P900 | — |
Longueur | — | 250 mm 9.8 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 |