This will allow for Direct Play in most cases IF the front end streaming app is smart enough to select AAC stereo ONLY when necessary.
It's hard for me to confirm that this is the case without control of the front end streaming app, but this might be a worthwhile bet to help eliminate audio transcoding.
This would help avoid duplicating video for stereo aac scenarios.
As far as alternative language audio, this would be an even more complex assumption to make by the front end streaming app. Perhaps there is a way such an app can always default select 'best quality audio stream AND correct language', but this opens another can of worms that might be best to leave for separate aliases just now (I would also say the same for subtitles).
IMO I would actually need full control over the front end streaming app (code wise) so to make it MAKE such correct decisions every time. But that luxury is not available yet.
At the end of the day, none of this should effect vfo development, and instead only effects the ffmpeg commands we use.
This will likely require further development into 'audio scenarios' but that was already in the plans anyway.
Might be a good idea to allow for down mixing to AAC stereo WHEN APPLICABLE.
stream 0: Video stream 1: Audio (Highest Quality English) stream 2: Audio (AAC stereo)
if stream 1 is already AAC stereo, do not run.
This will allow for Direct Play in most cases IF the front end streaming app is smart enough to select AAC stereo ONLY when necessary.
It's hard for me to confirm that this is the case without control of the front end streaming app, but this might be a worthwhile bet to help eliminate audio transcoding.
This would help avoid duplicating video for stereo aac scenarios.
As far as alternative language audio, this would be an even more complex assumption to make by the front end streaming app. Perhaps there is a way such an app can always default select 'best quality audio stream AND correct language', but this opens another can of worms that might be best to leave for separate aliases just now (I would also say the same for subtitles).
IMO I would actually need full control over the front end streaming app (code wise) so to make it MAKE such correct decisions every time. But that luxury is not available yet.
At the end of the day, none of this should effect vfo development, and instead only effects the ffmpeg commands we use. This will likely require further development into 'audio scenarios' but that was already in the plans anyway.