[Tom McCreadie]

"Gary Eickmeier wrote:
"Scott Dorsey" wrote in message
...

Decoded MS and coincident cardioids are mathematically identical and they
are very close to interchangeable with top-grade microphones. As you get
to cheaper microphones that are less flat off-axis, the X-Y configuration
becomes more of an issue since the center of the soundstage is off-axis on
both mikes.

In other words, coincident XY is the same as MS with the M and S about the
same gain, and also the XY can be converted back to an exactly equivalent MS
pair that can be manipulated just as if the original were MS (?).

No, the gain doesn't come into it. You are reversibly and losslessly
"translating" all the stereo information data from its listenable X/Y format
into a not-directly-listenable M/S format (All the information in the MS format
is available to re-create the XY.)

The ratio of M to S that happens to result from the XY - MS operation will
depend on (a) the polar pattern of the XY mics and (b) their angular splay. And
as a corollary, the pattern and splay of a (virtual) XY pair is driven by the
(a) the chosen M mic polar pattern and (b) the M to S ratio. This "about the
same gain" is irrelevant.

Think of XY - MS as a sort of translation process. Take a word "tree";
translate that into French using your an En-Fr dictionary; that may be
unrecognizable for your kids, but once they consult a Fr-En dictionary it all
becomes perfectly clear again.

Many folks have a persistent notion that only MS allows width adjustment (via
gain ratio changes.) Any coincident XY array can do that...just that it's a mite
less convenient, entailing an extra step:
MS - XY
MS - ratio change - M'S' - X'Y'
XY - M/S - ratio change - M'S' - X'Y'