Hello,
What I don’t understand is why don’t NVIDIA create a single method of GPU Acceleration to OS’s (Cross Platform)?
How can I use CUDA in Linux to speed up H264 1080p decoding in my NVIDIA GEFORCE 9600GT VRAM 512 GDDR3? Do I need to install it?
What’s the difference between CUDA and VDPAU (performance, decoding process, CPU usage, etc) and why don’t NVIDIA merge the two projects?
What’s the best GPU Acceleration method to use with LINUX in a NVIDIA graphical card? And with Windows?
Is there any way of mixing GPU Acceleration and Multi-Threaded FFMPEG (FFMPEG-MT) to decode H264 1080p videos together to improve performance?
PS: I’m using KUBUNTU 10.10 ; KDE 4.5.1; LINUX KERNEL 2.6.35-28-GENERIC-PAE ; NVIDIA 260.19.44 PROPRIETARY DRIVER (DOWNLOADED FROM NVIDIA.COM) IN A 32 BIT SYSTEM IN A NVIDIA GEFORCE 9600 GT VRAM GDDR3 512; INTEL QUAD CORE 2.66 GHZ Q6700; 4GB DDR2 800MHZ OF RAM
Thanks for your help and sorry about the tons of questions :) (I’m still a novice in this area),
André M.