On Wed, Sep 11, 2024 at 12:20 PM Niklas Haas <ffm...@haasn.xyz> wrote: > > On Wed, 11 Sep 2024 02:45:59 +0300 Andrew Randrianasulu > <randrianas...@gmail.com> wrote: > > I can't find entry about it in ffmpeg/MAINTAINERS > > > > bugs meanwhile exist > > > > https://trac.ffmpeg.org/ticket/3345 > > https://trac.ffmpeg.org/ticket/7978 > > There is on ongoing effort to redesign swscale, as well as fixing all > currently > open bugs, throughout the rest of 2024. Stay tuned for this.
I'm looking into your tree on github. But I think something still off at the input side of ffmpeg git? https://trac.ffmpeg.org/ticket/11182#comment:32 if I patched correct function it still calls limited to full range conversion while y4m input was already in full range (as far as I understand color range propagates in this case from demuxer, not codec) I updated my ffmpeg to 3f84d1d1fb75e4192c35664b4dd4741e8381b667 and it still behaves the same I also looked at internals of swscale, there is funny line: https://git.ffmpeg.org/gitweb/ffmpeg.git/blob/HEAD:/libswscale/utils.c#l876 dstRange = 0; //FIXME range = 1 is handled elsewhere does this mean that dst *range* is ignored as far as matrix coefficient generation in fill_rgb2yuv_table is concerned? I do not think our code [0], uses it directly, and commenting out this line does not change outcome for 11182 I also hope new API for avscale will be not very hard to use by not-quite programmers :) like me. as for signal theory and filter configurability .. I forced to read on it because I want to understand how it works both conceptually and in real-world, so I can have at least SOME idea where bugs might be, I tried to add some sws_flags ( int flags= SWS_BITEXACT | SWS_ACCURATE_RND) outside of SWS_POINT to our code, but this does not improve outcome for our bug - https://www.cinelerra-gg.org/bugtracker/view.php?id=665 Avoiding whole call to sws_setColorspaceDetails for yuv8->yuv8 "solved" problem but then we need to add some filters and properties on enc manually? [0] https://git.cinelerra-gg.org/git/?p=goodguy/cinelerra.git;a=blob;f=cinelerra-5.1/cinelerra/ffmpeg.C;h=9b8832dd718c94dd8b815edc7eb5c9d732953dd6;hb=HEAD#l1682 https://git.cinelerra-gg.org/git/?p=goodguy/cinelerra.git;a=blob;f=cinelerra-5.1/cinelerra/ffmpeg.C;h=9b8832dd718c94dd8b815edc7eb5c9d732953dd6;hb=HEAD#l1812 > > > > > I often saw suggestion to use -vf zscale but this is not default, and z.img > > upstream seems to be in suspended animation: > > > > https://github.com/sekrit-twc/zimg/issues/207 > > _______________________________________________ > > 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". _______________________________________________ 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".
