On 12/17/18, Michael Niedermayer <mich...@niedermayer.cc> wrote: > On Mon, Dec 17, 2018 at 11:24:43AM +0100, Paul B Mahol wrote: >> On 12/14/18, Michael Niedermayer <mich...@niedermayer.cc> wrote: >> > Hi >> > >> > replying to a comment from IRC here because "durandal_1707 has quit" so >> > it >> > wouldnt be seen, and might be interresting for others so not mailing >> > private >> > >> > <durandal_1707> michaelni: nut does not take into consideration duration >> > of >> > last frame to return real duration >> > <durandal_1707> michaelni: so duration is always smaller by small >> > amount >> > <durandal_1707> michaelni: looks like design failure >> > >> > Its a bit ago that i worked on nut but the last frame duration was >> > considered, >> > i do remember there was discussion about it. >> > >> > looking at nut.txt, it says this: >> > All streams SHOULD end with EOR, where the pts of the EOR indicates >> > the >> > end presentation time of the final frame. >> > >> > That stores and gives the last duration. Our muxer seems to not >> > implement >> > this >> > so its a implementation issue not a design one >> >> If one need to parse all packets to get last duration that is flawed >> design. > > That should not be neccessary > you can find the duration from the next packet as with any other packet > thats not > the last. And there is a index from which it should be possible to find the > last > too
Nope, pkt->duration is nowhere actually stored. Except this EOR thing not implemented by you. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
