[Rich Breen]

If you're on a native rig, if your session is at 48k, if all your audio sources are within Pro Tools, then it's fairly easy. Anytime any of those things is not true things get more complicated.

Personally, I have an HDX rig which feeds a CoreAudio interface (Metric Halo ULN8) digitally for monitor control and monitoring non-Pro Tools sources (itunes, talkback mic, system audio, etc).

At 48k all of this can feed Zoom no problem, at other sample rates I need to feed the ULN8 from PT analog so that system sample rate can stay at 48k for Zoom.

One caveat here; when many CoreAudio apps (like Zoom) are presented with *more* than two channels (for example, the ULN8 presents 32 outputs to CoreAudio), they just sum everything down to mono - stereo in to Zoom allows far better quality sound (assuming "original sound" is turned on in Zoom), so I need to use Loopback to to create an audio device that picks off two and only two channels from my Metric Halo CoreAudio feeds and that's what Zoom sees as its audio input. There's also a few different locations you have to go to to turn on stereo sound in Zoom: both in the app's audio prefs, but also in the meeting prefs in your account page on Zoom's website (they really make this convoluted).

All of this works, and sounds good, but is admittedly moderately complex. The aggregate device method suggested in the above linked paper would probably also work well with mixed sample rates, but is only for native systems.

best,
rich