Michael R. Hines <michael@...> writes: > the problem was that I was not using a *actual* > container to encapsulate my audio/video streams. > For example, if you make a call like this, for example: > avformat_alloc_output_context2(&ctx->outAudioFormatCtx, NULL, "ac3", NULL) > or avformat_alloc_output_context2(&ctx->outAudioFormatCtx, NULL, "h264", > NULL)
Please understand that both "ac3" and "h264" are "actual" containers in the sense that they are defined by public standards. You can test those containers with ffmpeg (the application): $ ffmpeg -i input -f ac3 out1 $ ffmpeg -i input -f h264 out2 (The second example needs a ffmpeg executable with x264 support.) Both output files should decode fine with FFmpeg (and out2 for example with the H264 reference decoder). Both "ac3" and "h264" are of course raw containers, but that does not make them less "actual" imo (especially given that both are defined in "actual" codec standards). Please note that I suspect the h264 format does support timestamps, FFmpeg support for these timestamps is unfortunately limited, several tickets are open. I am not sure how timestamps for encoded audio would help, afaik when decoding ac3 streams the decoding speed is defined. Carl Eugen _______________________________________________ Libav-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/libav-user
