Thank you for the replies. I appreciate it.
We play pop song covers pretty much exactly like the original recordings, so we're making loops when and where we can to augment the songs. For example, we do the old 80's song "Jenny 867-5309", I record the intro riff once, then play it back throughout the song when it fits. At another point, I'll record a verse chorus chord progression so I can play it back under the guitar solo. Same with vocal parts so that by the end of the song, my partner and I are singing with multi part harmonies.
So I'm not building full loops and playing over them, just copying bits and pieces here and there to play back at different times.
I determined that this was more 'legit' than using complete backing tracks, however our drums still are at this point canned.
The method leoncremonini described sounds interesting to me.
I use an X18 digital mixer, so I assume I could route the inputs I wanted to loop to my laptop from the X18 via USB, silently (in ALK2) record the parts I want to record and send the play loops back to a couple of new channels on the X18. This seems to be the best way to avoid latency in monitoring while performing, but I feel like I'd lose a lot of other functionality ALK2 provides (like easy automation of the whole mix and effects).
I've tried a few loops on my laptop with my Focusrite interface, and the latency was more than I could deal with live - but I didn't mess with the buffer size from the default. Have people had success getting the latency down to minimal levels going with buffer sizes down to 128 while still maintaining stability and quality?
The full mix version of this setup would be to use my 4 in 4 out USB interface, run all four inputs into ALK2, the same four back out to my mixer after they've passed through the record/monitor tracks. This introduces some other issues as well - like the vocal mics would be dead between songs.
Thanks again. Lots to think about.