As many are aware, Microsoft released their C++ Accelerated Massive Parallelism specification. I was wondering if there was anyone on the dev team (or community) who would be willing to look into the possibility of Handbrake being written therein.
I know you guys were sour on using OpenCL and CUDA in Handbrake, but given that no GPU-specific code needs to be written, I thought it might be more palatable. I, and many others, have long wished to utilize our graphics cards for encoding.
C++ AMP for the Windows Branch
Forum rules
*******************************
Please be aware we are now using GitHub for issue tracking and feature requests.
- This section of the forum is now closed to new topics.
*******************************
*******************************
Please be aware we are now using GitHub for issue tracking and feature requests.
- This section of the forum is now closed to new topics.
*******************************
-
Carborundum
- Posts: 22
- Joined: Tue Jan 18, 2011 11:29 pm
Re: C++ AMP for the Windows Branch
Handbrake is a frontend for other programs (x264, ffmpeg, etc) which do the actual encoding. You will need to ask the devs of those programs, although chances are no-one will be interesting in writing anything Windows specific.
In any case, C++ AMP is intended for GPGPU applications, same as OpenCL or CUDA. As I understand it, the reason video encoding software isn't being written in these languages is not that it would be too much work, but rather that it wouldn't yield any increase in performance.
In any case, C++ AMP is intended for GPGPU applications, same as OpenCL or CUDA. As I understand it, the reason video encoding software isn't being written in these languages is not that it would be too much work, but rather that it wouldn't yield any increase in performance.