So you’re saying that it approximately tunes it 100% if you ask it to, but then upon doing a re-analysis, it finds fault with it’s last processing not being exact.
I also find that if you do things like this the audio starts getting messed up sounding (mangled) after repeated processing and bouncing of the same audio. Some note to note glitches start to appear upon repeated processing and flattening/bouncing. So it would be best to export/re-import I’m guessing. Though that takes a longer time to do, especially with the kind of numbers of times I have to do this with so many live tracks. Plus I have lots of bleeding since the mics are all open since it’s on stage and no direct boxes except for elec. bass and elec. guitar (2 out of 10 players of which some are doubling on various woodwinds)…