On Mon, May 27, 2019 at 7:52 AM Nicolas George <geo...@nsup.org> wrote:
> Jun Li (12019-05-27): > > I compared the perf between vflip and this patch before, they take almost > > the same time, around 9ms for a frame. > > Hflip is a little special, but surely I can update the patch to avoid > frame > > copy, and it will be as fast as hflip. > > > > But still this is duplicated feature with vflip/hflip, the only benefit I > > can think of is rotate180. > > Previously we need chain two filter (vflip + hflip) to do rotate180, here > > we only need one. > > Please note that there are two separate issues interacting here: having > separate filters, having separate code. > > Merging the features of a filter into a more generic one is, IMHO, a > good thing. In this instance, it makes the rest of the code simpler. > > But that does not mean you have to duplicate the code. You should almost > never duplicate the code. What you should do is make sure the two > filters are using the same code, i.e. calling the same functions to do > the actual work. > > Thanks for pointing that out. I updated the version and merged the existing flip operation into transpose, removed my previous implementation. https://patchwork.ffmpeg.org/patch/13311/ https://patchwork.ffmpeg.org/patch/13312/ Thanks a lot for your code review. Best Regard, Jun Regards, > > -- > Nicolas George > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe". _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
