On Tue, Feb 24, 2009 at 11:06 PM, Mike Sullivan <[email protected]> wrote: > On Tue, Feb 24, 2009 at 6:45 PM, Ronald S. Bultje <[email protected]> wrote: >> Hi Mike, >> >> On Tue, Feb 24, 2009 at 8:17 PM, Mike Sullivan <[email protected]> wrote: >>> The problem is that since I used -r 2, I am inadvertently skipping >>> I-Frames. In other words I will get a frame with a timestamp of .5 >>> seconds, 1 second, 1.5 seconds, etc but the I frames don't line up >>> with this timing. >> >> I think this kind of stuff you'll want to handle with a new >> demuxer-specific option, so that the demuxer can choose which frames >> to skip. You can set the number of B-frames etc., but that will only >> sometimes work. >> >> HTH, >> Ronald >> _______________________________________________ >> libav-user mailing list >> [email protected] >> https://lists.mplayerhq.hu/mailman/listinfo/libav-user >> > > Gotcha, sounds good. So just leave out the -r and ignore the b-frames. > > Thanks for all the help, > > -Mike >
So now I have the format option all set up, but I just realized I can't (or at least I don't see how) I can get access to the AVCodecContext struct from inside the format_write_packet(AVFormatContext *, AVPacket *) function. I don't see anything in the AVFormatContext struct that lets points me to the AVCodecContext. Any ideas? I'll make sure to send out a patch once I'm done btw. Thanks again for all the help, -Mike _______________________________________________ libav-user mailing list [email protected] https://lists.mplayerhq.hu/mailman/listinfo/libav-user
