quick question are there any techincal resources that explain why a coincident pair works better than near coincident with out rigger omnis? I'm not sure If I understand why that is true, or at least why people think that is true.
I'll take a stab in the dark with a guess/theory based on my listening/experiece:
There will already be an exaggerated difference between between the L and R of the split omni pair (vis-a-vis coincident or near-coincident) in the time the sound reaches each mic when the source is not 'centered.' When summing a stereo mix from a split omni pair with a center pair, it seems logical that a coincident pair in the center will minimize any additional 'soupiness' or lack of detail/definition in the stereo image upon mixdown. Rather than the ears/brain processing the overlap of two distinct stereo images when the center pair is near-coincident (left-ctr left-ctr right-right), a coincident center pair makes it one large stereo image with a crisp/defined center as well as distinct left and right (left-ctr pair-right).
If a loud sound comes from one side of the stage/PA stack, you will have 3 differing lengths of time between when the sound reaches the capsules of the 4 mics (fig. 1), with the sound reaching the capsules at the same time for the center pair. With a near-coincident center pair, there will be 4 different lengths of time (fig. 2) with the center pair approximating the regular difference perceived by the a set of human ears from one location, and the omni outriggers as an extended/exagerrated difference super-imposed on top of that.
I think of the differences in sound resulting from the varying differences to be similar to separate, closely matching but not exact, 2 dimensional outlines of an image of a 3 dimensional object...like the object's silhouette and it's shadow(s). In the case of of a coincident center pair, to me it is like the omni outriggers make up the primary silhouette and shadow, which can become a blurry double image by themselves if not precise. A coincident center pair in this context would be analagous to giving the shadow (or spaces between the outlines) an extra clear line/limit (fig. 3), thereby adding some crispness and definition to the image. A near-coincident pair, other other hand, would be like having a couple similar, but slightly different, lines between the silhouette and shadow, like shading (fig. 4).
I believe it boils down to phasing issues that are more likely to occur with an exagerrated delay between the outriggers once summed with a similar, but slightly different, delay between the center mics as well. With a coincident center pair, this additional variable is minimized and easier to for the ears/brain to ignore or treat as one large image instead of two images superimposed upon one another.
(fig. 1 - coincident center pair: m/s, XY, Blumlein)
PA-STAGESTAGESTAGESTAGE-PA
----------------
----------------
O MS O
(fig. 2 - near-coincident center pair: DIN, DINa, ORTF, NOS, etc.)
PA-STAGESTAGESTAGESTAGE-PA
----------------
----------------
O OR_TF O
(fig 3. - coincident center pair)
__
\\ \
\\ \
\\ \
\\ \ (fig 4. - near coincident center pair)
___
\\`\
\\`\
\\`\
\\`\