Thanks for your response, Squids. Yeah -- it seemed like that would be the only way to do it given CPU limitations -- but just wondered if I was missing something -- because I always prefer not rendering/freezing until the end.
I wasn't sure if you were overthinking it or if I just wasn't catching what you meant. But, now I have a better idea and also I think it helps to think of the application of its use because the plug-in can be used in so many ways.
So, if you wanted to do something like you're saying here where you'd simulate the sound of putting an instrument in a room at Sunset Sound (which, to be fair, is never going to be exactly the same no matter what plug-in you used for room simulation but knowing this the idea is to get close and have fun with it in my opinion... in most cases most people wouldn't even be able to tell that you DIDN'T actually record in the studio). To do that, I would put SSSR as an insert effect on that instrument (or on a buss insert if it's a multi-miked instrument) and then into EQ, compression and anything else I'd run a sound recorded in the room into on the console and outboard gear if I was at that studio (or even in my studio which is part DAW and part analog). But, as I said before, if CPU power was a concern then I'd just commit/freeze the track and print to audio the desired room sound (and keep the unprocessed original dry sound around as an inactive track in case I change my mind). Nothing wrong with printing it (ie. committing it to audio without the plug-in anymore - baked in as you said) because, after all, if you DID record the instrument in the actual room at the studio it would be recorded with that sound anyway. Now you're just using the plug-in to make it sound like you recorded it in that room. That should work quite nicely in a lot of cases!
Hope that helps. I like the way you're thinking and thanks for helping to take the discussion in new areas as I'm sure it's giving other people ideas on how they could use it too. You see? This is how a thread like this can be GOOD! People being friendly with each other and sharing ideas. That's great. I enjoy being around for that!
For that reason -- I find the CPU usage of IK Tapes, for example, causes me to avoid it other than one instance on the 2-bus. (Even though you could apply the same "baked-in" logic to rendering early with that plugin too -- to simulate multi-track recording.)
At least w/a tape emulation you can simulate mixing down from digital to a 2-track tape master by using just one instance.
But you're absolutely right that -- in theory -- if I were actually recording in the room, the SS sound would be irretrievably baked in anyway -- just like if I were actually recording to multi-track tape, the tape sound would be baked-in -- which justifies early rendering in either case.
Workflow is a consideration. If SSSR weren't (relative to my budget level) so expensive, I'd be more tempted to consider a workflow like this or conversely if there was a workflow to accomplish this that was more flexible (or a really low CPU usage to use many instances), I'd be more tempted to exceed my budget.
Thanks again for the discussion. I'm curious to hear user reaction to SSSR when it's finally released.