Can someone explain me how it works in Audacity??? I mark “Use custom mix” then when i export in different formats it gives me another windows with so many channels 
It’s not something that I do myself, but the manual page is here: http://manual.audacityteam.org/o/man/advanced_mixing_options.html
The idea is that you can create a multi-channel audio file by “mapping” (connecting) the tracks in the Audacity project to the required channel numbers in the file that you are creating.
See here for some information about channel mapping for surround sound: https://en.wikipedia.org/wiki/Surround_sound
I don’t see how you could possibly do surround in Audacity as there is no way to listen to your mix…
What if you import a multi-channel (surround sound) file for editing? You can use Sync-Lock to keep the tracks synchronised during editing (http://manual.audacityteam.org/o/man/sync_locked_track_groups.html), but then you need some way to export the edited tracks as a multi-channel sound file.
As in: edit, export, listen?
Won’t work. You need “live” output for panning, fi.
The way surround is created, it starts with 4 recorded channels*. All the rest is math in a plugin. But you need to hear the results and then 4 channel output is the bare minimum. You can’t use headphones, not even the so-called 5.1 stuff. 8 channels would be plenty, even if we have done 32 channel surround. But that was a showoff for an exhibition.
You can’t simply “mix” 4 independently created channels, as these will lack the necessary time/phase differences between the channels. Even pseudo stereo fails.
*Unless it’s synthetic sound, as used in games. I don’t even see people who create game audio work this way. I’m not really familiar with the sector, but I think there’s a fair chance that even more math is involved. And sound following an in-game character, fi, can only be judged by listening, afaik.
As in:
[surround sound input file]
----talk —blah —blah —SONG—more talk—blah—blah—
to:
[surround sound output file]
-SONG-
In this case, all you need to do is to trim the same amount from the start of each track, and the same amount from the end of each track, then export as a multi-channel audio file. If the exported file is the same as the imported file, then the channel mapping will be:
track 1 ->channel 1
track 2 → channel 2
…
If the export format is different, then different channel mapping may be required.
If you want to change the number of channels, then other software may be more appropriate.
Yes, Steve, that limited use case would work…
Thanks