Mediacoder conversion stages and subtitles

[ChazZeromus]

Hey guys,

I'm converting a video for a mobile device and want to hardcode subtitles in mediacoder. And so far so good using mediacoder has worked for numerous amounts of videos. However the latest video I'm trying to convert now does not convert correctly with a mencoder, and avisynth as source codecs. The subtitles are separate .ass files which have successfully hardcoded in the past. I've been able to successfully convert videos using ffmpeg as the source codec, but subtitles do not render into the final video file. Avisynth produces a file but the video stream is much faster than the audio stream, but the converted video has subtitles.

I'm not familiar with the internals of the file formats, but as I understand the conversion process begins by feeding the file into a source codec that produces the raw video, which is then fed into an encoder (x264 in this case) and creates the converted video stream. The same happens with the audio in that a separate audio stream is created also. Then two streams are fed into a multiplexer to finally form the file that contains the streams.

If this is correct, then I want to know when the subtitles are finally hardcoded to the video. Are the subtitle options passed into the multiplexer? Or are they passed into the video encoder?

[MilesAhead]

I would give TEncoder a try. Hard subs are encoded into the video by the video encoder. According to the change log for TEncoder, 3.1 has fixed the problem where ffmpeg could not read .avs scripts. That is where you would use TextSub() filter to hard code the subs if they are text based.. such as .srt .ass .ssa. For image subs you use VobSub() filter.

If you use AvsPMod editor specialized to AviSynth then you can see a preview of the effect of the filters. As example, if the subtitles appear in the frame etc..