I was just wondering if there are plans to improve the Cuda GPU acceleration codecs and if they'll be implemented?
I have the GTX460, and depending on the movie I'm compressing the most GPU usage I ever see is maybe 20% , the average GPU usage is usually less than 15%. I know that some people are working on taking full advantage of the new Fermi architecture.
For me Cuda is definitely the quickest option. I've tried Cuda vs Lightening, coreAVC, or just Software. It's the quickest when it's CUDA/CUDA for encode/deencode. I know theres a lot of room for improvement because of the small amount of GPU usage I'm seeing. I think right now, the software is not taking full advantage of the potential of the new Cuda Fermi processors.
I have the GTX460, and depending on the movie I'm compressing the most GPU usage I ever see is maybe 20% , the average GPU usage is usually less than 15%. I know that some people are working on taking full advantage of the new Fermi architecture.
For me Cuda is definitely the quickest option. I've tried Cuda vs Lightening, coreAVC, or just Software. It's the quickest when it's CUDA/CUDA for encode/deencode. I know theres a lot of room for improvement because of the small amount of GPU usage I'm seeing. I think right now, the software is not taking full advantage of the potential of the new Cuda Fermi processors.
Comment