Yes, the timing is crucial here. The background audio serves as a metronome in that sense, but moreover it also contains the harmonies etc that the VSTi realtime/live playing is supposed to integrate with.
All of my VSTi:s are registered. But the kind of limitations in other audio software that you mention is yet-to-be-discovered by me – I haven’t so far encountered any such program that would stop e.g. Soundforge from sampling at least the ”What U Hear”-audio data.
I know that the Renoise coders have built in a ”scan” for Virtual Audio Cables and similar software (Renoise won’t load unless such VACs are inactive). But I don’t see that point at all, since you always can generate a WAV-file in realtime by the procedure I just mentioned above.
The reason why I want this particular solution is because I use an early version of ACID (1.0 rules) as my recording backend. In that software it is very convenient to just hit the ”Record” button in ACID and record the audio from one of multiple possible sources into a new track. If the result is no good, I just delete it and take a new shot. I enjoy playing live, because that’s the kind of music and arrangements that I most frequently compose.
Granted, this kind of workflow has an obvious negative side to it: one cannot ”polish” individual notes within a recorded sequence (as you can in Renoise or in Cubase). But still, that is the way I prefer to work. More demanding programming, i.e. stuff that I can’t play live, I usually ”step-edit” in Reason or Finale.
Voxengo Recorder does the trick in a way. It filters out only the audio that is generated from the specified VSTi playback – so it doesn’t matter that other sounds are played simultaneously in the background – and renders it to a new WAV-file.
However, that procedure is still too time-consuming, I think. What I want is to route the VSTi audio directly into ACID’s recorder, while simultaneously playing the already existing tracks in ACID in the background.
While I can see that two computers certainly will do the trick, I cannot see how two soundcards will suffice for this particular problem. Does anyone here have a suggestion for a soundcard that can actually co-exist with my SB Audigy2 ZS card, in the same machine, and route the audio signals as demanded without some sort of advanced SPDIF loopback solution?