Hello everyone, I'm trying to encode video and audio directly from a capture board, 1 video stream and 8 single channel audio streams into a MXF file encoding it in Mpeg-2.
Everything is working fine except when I call av_write_trailer() at the end (and before freeing CodecContext as stated in the documentation). After using av_write_trailer, I end up with the whole video looking like this http://i.imgur.com/M6HkcpR.jpg . It seems like the decoder is misplacing the b-frames in the wrong position, but If i don't call this function I can play the media file, with the image as it should be, in VLC but the timestamp is completely wrong. Looking at mediainfo I can see that my file doesn't have the file duration (which later made sense when I looked inside the code.) After a few hours with the file mxfenc.c found that if I comment the next part inside the function mxf_write_footer(), the write trailer "kinda" works, (...) if (mxf->edit_unit_byte_count && s->oformat != &ff_mxf_opatom_muxer) { // no need to repeat index if ((err = mxf_write_partition(s, 0, 0, footer_partition_key, 0)) < 0) return err; } /*else { if ((err = mxf_write_partition(s, 0, 2, footer_partition_key, 0)) < 0) return err; mxf_write_klv_fill(s); mxf_write_index_table_segment(s); }*/ (...) The "kinda" works part, its because I can play the file correctly in VLC with the correct timestamps, and if I open it with Mediainfo everything looks good, but If I try to open it with Sorenson Squeeze (for example), I get a warning saying "Squeeze does not support playing this file". I carefully followed the documentation, samples, examples and I'm still stuck :S. I am using the ffmpeg 2.5.1 version and I also added a few changes that I saw in the master, and nothing changed. Additional info: Target Output file format is a 1080i59.94, XDCam 30Mbit MXF op1a. Regarding the 8 audio streams the codec options have the one provided by avcodec_get_context_defaults3 plus this changes: audio_codec_context->channels=1; audio_codec_context->sample_rate = 48000; audio_codec_context->bit_rate = 768000; audio_codec_context->sample_fmt =AV_SAMPLE_FMT_S16; audio_codec_context->flags= CODEC_FLAG_GLOBAL_HEADER; audio_codec_context->time_base->num = 48000; Regarding video I also do avcodec_get_context_defaults3 plus this: video_codec_context-> pix_fmt= AV_PIX_FMT_YUV422P; video_codec_context->height = 1080; video_codec_context->time_base->den = 30000; video_codec_context->time_base->num = 1001; video_codec_context->field_order= AVFieldOrder.AV_FIELD_TB; video_codec_context->width = 1920; video_codec_context->codec_id= AV_CODEC_ID_MPEG2VIDEO; video_codec_context->has_b_frames= 1; video_codec_context->max_b_frames = 2; video_codec_context->gop_size= 12; video_codec_context->global_quality= FF_QP2LAMBDA * 1; video_codec_context->rc_initial_buffer_occupancy= 10695476; video_codec_context->scenechange_threshold = 1000000000; video_codec_context->color_primaries= AVCOL_PRI_BT709; video_codec_context->color_trc= AVCOL_TRC_BT709; video_codec_context->qmin= 1; video_codec_context->flags |= CODEC_FLAG_GLOBAL_HEADER | CODEC_FLAG_INTERLACED_DCT | CODEC_FLAG_INTERLACED_ME; video_codec_context->bit_rate = 30000000; video_codec_context->rc_min_rate= 30000000; video_codec_context->rc_max_rate= 30000000; video_codec_context->rc_buffer_size= 10695476; I also had specified me_method and me_range, but nothing changed. Any idea what might be the problem? Thanks guys, Cheers _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
