[Doom9]

I have a conceptual question, for bond, and everybody else:

Does it make sense that you can configure audio streams to be muxed in the main GUI or should this only be possible in the muxing window?

Right now, you can select .aac/.mp4 audio tracks in the main GUI, then when you press AutoEncode, you configure the final output name, split size, etc. And if you check "Add subtitles & chapters" before pressing GO, you can configure the name of the audio tracks you'll get (be it tracks that will be muxed, or tracks that will be encoded and then muxed), and add additional streams.

To permit configuration of both audio tracks that will be encoded and those that will only be muxed in the main window, I had to go through considerable hoops. Now with the addition of AVI output in auto encoding mode, it might get worse, and conceptually I'm not sure it makes a lot of sense to configure audio tracks that are to be muxed, not encoded, in the main GUI which is clearly meant for setting up jobs that are to be encoded.

Thus, what would you think about the following:

In the main GUI in the audio section, you configure audio jobs that are to be encoded. Then if you go into auto-mode, the checkbox you can check to give audio tracks a language, as well as adding subtitle tracks and chapters, will lead you to the mux configuration window where you can not only add a language descriptor to existing audio tracks (those that you set up for encoding in the main GUI), but you'll also be able to add additional existing audio tracks (.aac/.mp4 files for mp4 output).

And in case of AVI output, the additional window will just allow you to specify pre-existing MP3/AC3 audio tracks.

Or simply put: everything that you're going to encode is set up in the main window, everything that you're going to mux is set up in an additional window from within the auto-encode window (and of course the muxing functionality will also be available separately from the tools menu.. the mp4 muxer is already available from that menu today).

I think that way would be cleaner insofar that encoding and muxing are clearly separated. Considering all the work already put into this, it might not really make my life easier at this point, but from a coding point of view at least it would still be a more easy to understand workflow.