Gary, I'm not really familiar with the details of your surround sound
shenanigans, but the following applies to straight two-channel stereo:
You will _always_ get an image lateral-inversion (your "channel reversal") of
certain rear-arriving sounds whenever you employ:
1. an XY array (except 180° back-to-back splayed) with mics of tighter pattern
than cardioid
or
2. an MS system that after decoding equates to the above array "1"
Why is this? The mic patterns in the above arrays always have a rear lobe of
negative polarity. And there will always exist some angular direction within the
rear right quadrant w.r.t. the array from which an arriving sound will deliver a
stronger mic signal via the negative lobe of the L-angled mic than via the
negative lobe of the R-angled sister mic.
Consider, for illustration, the classic ± 45° Blumlein. The axis of the rear
lobe of the L-angled mic (and the null of the R-angled mic) points to "4:30
o'clock". So a sound arriving from, say, 4:45 o'clock would deliver a strong
signal (albeit negative) into the L-channel, but a much wimpier signal (though
also negative) into the R-channel. Hence a sound source located at 4:45 o'clock
will image left of centre on two-speaker playback, i.e. as if it were located at
7:15 o'clock, i.e. lateral-inversion of image.
The actual angular width of this lateral-inversion rear sector varies with the
mic pattern, XY splay angle and M:S ratio....but there's never a
lateral-inversion when using omni in MS because an omni records all signals with
positive phase.
---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com