Yes, it is related to the potential problem. The problem may be here: https://github.com/FFmpeg/FFmpeg/blob/40dda881d6ad761b4589c3078b925e0d849546b3/libavcodec/jpeg2000dec.c#L2139
Shifting down by 16 bits here is too much because it will cancel the value of "1" (=0.5) below the binary point. > On Aug 23, 2024, at 1:36, Pierre-Anthony Lemieux <p...@sandflow.com> wrote: > > On Thu, Aug 22, 2024 at 2:24?AM WATANABE Osamu > <owata...@es.takushoku-u.ac.jp> wrote: >> >> Hi Pierre, Tomas, >> >> For lossy codestreams, I found a potential problem on the current 9x7 >> integer IDWT (which will be activated with the +bitexact flag.) > > Probably related? > > https://trac.ffmpeg.org/ticket/10123 > >> >> I have confirmed that FFMPEG passes all the test cases without the flag. >> >> The problem has been locally solved with the updated code, however, the >> patch set is not yet ready. >> (It won't take long) >> >> Best, >> Osamu >> >> >>> 2024/08/22 1:01、Pierre-Anthony Lemieux <p...@sandflow.com>のメール: >>> >>> ?On Wed, Aug 21, 2024 at 8:54?AM Tomas Hardin <g...@haerdin.se> wrote: >>>> >>>> man 2024-08-19 klockan 23:03 -0700 skrev p...@sandflow.com: >>>>> From: Pierre-Anthony Lemieux <p...@palemieux.com> >>>>> >>>>> --- >>>>> tests/fate/jpeg2000.mak | 129 >>>> >>>> I presume you've tested that the frames match the references as >>>> required. If so then it looks OK to me >>> >>> @osamu you checked that all the codestreams decoded within acceptable >>> limits, right? If so, I will add you as co-author. >>> >>> (I have only checked a few of them for accuracy per the spec. I have >>> checked all of them for memory leaks/crashes.) >>> >>>> >>>> /Tomas >>>> _______________________________________________ >>>> 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".
