I think that is the easiest way to handle it. Obviously doing that breaks the latency compensation that the system is known for however.
I will say that we use our Main L/R out as a “band group”. In other words, I send band inputs to various groups and then to my band group (really main l/r) and then out to matrixes (main l/r, subs and fills for example). For vocals, I send those inputs to a vocal group and then direct to the matrixes. This give me an “extra layer” of processing my instrument/band channels before everything (including vocals) is combined for the outputs at the matrix level. It also gives me an easy way to keep my vocals out of the subs – I simply don’t assign the vocal group to the sub matrix. I do have to use a DCA with all of my matrix assigned to it as my “master buss” because the normal l/r fader only controls my band group.
Now I realize that is a very narrow solution that isn’t as flexible as sending groups -> inputs -> back into groups. It might not be sufficient for what you are trying to accomplish. But if it does work for you, it won’t break the latency compensation of the system.
.
