I’ve been working to get MG3 set up as a viable input device for MuseScore, and I’m looking for a little clarity on a couple of things. To wit:
Placement of MIDI Output module. Is there any reason I should deliberately place the MIDI Output module within a channel chain, or could I also put it on the Master channel? It appears to me on quickie testing that either is okay, as long as you don’t need to get audio passthru as well. Which actually raises a related question:
MG3’s interruption of audio signal. This is just the nerd in me wanting to understand how MG3 “thinks” and behaves. If I’m interpreting things correctly by observation, it looks to me like MG3 takes audio input and routes it to all three channel chains. This audio may or may not be processed by audio effects, but an audio signal continues to go down its channel chain, and through to and down the Master chain, remaining as an audio signal unless and until it encounters something that uses the pitch-to-MIDI conversion (where that “something” would most obviously be an Instrument or the MIDI Out module). At that point, the audio signal is effectively “turned off” and the corresponding MIDI notes are used to generate either new, replacement audio (via an Instrument) or simply MIDI output with no audio output (MIDI Out module). Is this description correct?
If I have that right, then theoretically I could have up to three different MIDI channel chains (with, say, different Transposers on them), that could all feed a single MIDI Out module on the Master channel. This would get me around the problem of MuseScore only listening to one MIDI input device at a time, which needs to be MG3 because that’s where the notes are coming from, but since each MIDI Out module represents itself as a different device to MuseScore (e.g., MIDI Guitar Out 1, MIDI Guitar Out 23, etc.), I can’t put one on each channel and be able to switch channels to get different configurations of Dynamics, Transposers, Vibrato, and Sustainer. (I do find myself running out of blocks in a given chain, otherwise I’d just switch between 3-4 Transposers in a single chain. 
)
And if I still have all this right, this begs one more question.
When does pitch-to-MIDI and MPE data get generated? I think I know the answer to this, but given all the above (and my usual problem of having the thought/revelation at a time when I’m not with a guitar to test it out) I’d like to be really clear on it: am I correct in presuming that the pitch-to-MIDI and MPE data is generated exactly once, right at the beginning, before passing both audio and this MIDI note and MPE data to each chain? I think this is right because I can see the little pitch rays behind the guitar lighting up even if I remove all MIDI modules and Instruments from all chains, leaving only audio to pass through.
Thanks for any clarity on this!