Yes I have experimented with this. The clips in question have two separate audio channels (i.e. they are two separate mono channels and not stereo). This is how CatDV sees them as it references them correctly in the metadata and I get the option to choose a track to hear in the playback section.
However, it seems impossible to generate a proxy through CatDV that will have the same audio structure. When I try to do it an uncompressed as you have suggested it turns it into stereo, which is effectively a mixture of both tracks that cannot be separated, i.e. it has taken both channels and mixed them into stereo, so its a mash up of the two sources, i.e. I can’t get the clean channels.
Lets imagine you have two wireless mics recording a documentary shoot into one camera, I want to be able to separate them at the proxy level for transcription, so effectively I still need a separate channel 1 and channel 2. I cannot get this through CatDV when creative proxies.
If I preselect mono in the audio encoding options I get just one channel.
I am surprised there is not a pass through option as it seems that its something quicktime can handle through compressor and other software. If I do it through compressor I can pass through the audio and have compressed low quality video but with the audio clear and still as separated tracks.
Maybe I am missing something?