You're quite right about using Pitch 'n Time before signal processing generally giving better results. To me the likely reason would be that there is almost always an associated phase shift of the affected frequencies. With complex frequency content over 2 or more channels, we end up with complex phase relationships that are constantly changing as the effect unit responds to the makeup of the input.
Pitch 'n Time doesn't change the phase of the audio at all (except in multi-channel mode). So while the phase is the same in the output, the time scale is changed, and chances are the phase relationships won't exactly match what would result if we had applied our effects to the TC/E'd audio.
Then again if you're EQing to fix phase problems, you might be better off doing that prior to Pitch 'n Time, so as usual it's best to experiment to find what suits the situation.
In the case of Pitch 'n Time mono and multi aren't describing the material they're best suited for, but rather the way in which it carries out it's processing. That is, in mono mode each track is processed independently whereas multi channel mode alters them all together forcing them in phase with each other.
Multi channel mode is only useful when you need to force multiple signals into phase. It's a drastic change which really won't suit anything except those specific cases where it's called for, such as multiple tracks of the same instrument with wildly different effects applied, playing back in sync. Phase cancellation might be messing up the result of pitch shifting them all down an octave, so pulling them into phase is necessary.
you can grab a PDF of the manual from the downloads page -
www.serato.com
