00:09 <cone-528> ffmpeg Gyan Doshi release/7.1:68b5db246407: avcodec/libsvtav1: unbreak build with latest svtav1

00:14 <fflogger> [newticket] spixi: Ticket #11492 ([ffmpeg] Support for QuickTime skin file) created https://trac.ffmpeg.org/ticket/11492

00:21 <fflogger> [editedticket] spixi: Ticket #11492 ([ffmpeg] Support for QuickTime skin file) updated https://trac.ffmpeg.org/ticket/11492#comment:1

04:41 <fflogger> [editedticket] Gyan: Ticket #11492 ([ffmpeg] Support for QuickTime skin file) updated https://trac.ffmpeg.org/ticket/11492#comment:2

10:38 <fflogger> [editedticket] ePirat: Ticket #11492 ([ffmpeg] Support for QuickTime skin file) updated https://trac.ffmpeg.org/ticket/11492#comment:3

10:58 <fflogger> [newticket] agH94PeodEE9: Ticket #11493 ([undetermined] ffmpeg causes mkv corruption when remuxing existing MKV with WEBVTT (.vtt) subtitles that have empty cue blocks) created https://trac.ffmpeg.org/ticket/11493

13:04 <fflogger> [editedticket] Marc: Ticket #4298 ([avfilter] zoompan filter creates shaky image) updated https://trac.ffmpeg.org/ticket/4298#comment:17

13:25 <cone-409> ffmpeg Brad Smith master:30a8641465f7: lsws/ppc/yuv2rgb_altivec: Fix build in non-VSX environments with Clang

13:38 <MihirGore> Hello everyone! I'm Mihir Gore, a junior electronics undergrad at VJTI Mumbai. I was a GSoC '24 contributor with JdeRobot, where I worked on optimizing ML models for computer vision and robotics.

13:38 <MihirGore> I've recently submitted a patch adding VPX range encoder support to FFmpeg, extending the range coder structure for both encoding and decoding. Currently, I'm interested in working on the VP6 encoder project for GSoC 2025 and would love to discuss ideas or get any feedback on my patch. Looking forward to contributing and learning from you all!

14:40 <cone-409> ffmpeg Dmitrii Ovchinnikov master:c7ff7345eb9a: Enable 10-bit output for RGB input in AMF HEVC and AV1

14:40 <cone-409> ffmpeg Dmitrii Ovchinnikov master:7225e307be6a: Added support for direct RGB input to AMF encoder

14:42 <haasn> JEEB: can you explain the different YCgCo variants to me and how to handle them?

14:44 <MihirGore26> Hey everyone, I'm Mihir Gore, a junior electronics undergrad at VJTI Mumbai. I was a GSoC '24 contributor with JdeRobot. I've submitted a patch adding VPX range encoder support to FFmpeg and am interested in the VP6 encoder project for GSoC 2025. Looking forward to contributing and learning from you all!

14:49 <JEEB> haasn: so the difference is that there's the reversible and the non-reversible variant. the old H.273 variant basically was dependant on the bit depth (if you look at H.264 etc from which it was copied to H.273), where if chroma had +1 bit compared to luma (if I recall correctly), then it was the reversible variant. otherwise non-reversible. then they removed differing bit depth support in VVC :D

14:49 <JEEB> (or well, it was in general removed in H.273 I think)

14:51 <JEEB> thus you now have the new variants for reversible specifically. one is for even and one is for odd amount of bits

15:22 <haasn> JEEB: do you have formulas somewhere?

15:23 <JEEB> https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-H.273-202407-I!!PDF-E&type=items , (58) ->

15:24 <JEEB> > Otherwise, if MatrixCoefficients is equal to 8 (with BitDepthC not equal to BitDepthY), 16, or 17, equations 58 to 61 apply

16:17 <JEEB> ohai

16:17 <Sean_McG> BTW not sure who to thank, but I like the new stylesheet for FATE.

16:18 <Sean_McG> JEEB :)

16:58 <fflogger> [editedticket] SpaceAgeHero: Ticket #11376 ([undetermined] undefined symbol: vaMapBuffer2) updated https://trac.ffmpeg.org/ticket/11376#comment:1

18:00 <haasn> 312 FATE tests changed by new swscale code :')

18:00 <haasn> because it's not bit exact with the previous..

18:00 <haasn> sad

18:01 <haasn> Maybe I could try recreating the existing dither pattern exactly but it's not correctly rounding in all cases and also smaller

18:01 <haasn> and I will switch to blue noise dither anyway

19:09 <fflogger> [editedticket] Marc: Ticket #4298 ([avfilter] zoompan filter creates shaky image) updated https://trac.ffmpeg.org/ticket/4298#comment:18