I am aware of Martin Schneider's AES paper from 2001 and will probably revisit it, it's been a long time since I read it.
Reading the information available on the Schoeps website for the Polarflex system has been very informative. That is exactly the kind of tool needed to answer the questions I'm posing above in a empirical, non-theoretical way. Can you point me to any discussion of how users have found it to be most useful to them, or other 'application notes'?
Polarflex is pretty much an answer to the first of two questions I recently asked of the ambisonic research experts on the 'sursound' mailing list, concerning tools which could do similar things for 'virtual' microphones derived from ambisonic recordings. I asked:
Can anyone point me to B-format tools which can produce a virtual
microphone output which has the following specifications?
1) Specify a changing polar response that varies by frequency, morphing
smoothly between two specified first-order patterns set at two specific
frequencies.
2) Specify a changing frequency response that varies between two user
specified equalization curves: a direct on-axis curve and a 180-degree
off-axis curve, so that the frequency response of the virtual microphone
varies smoothly by the angle of incidence.
Reading about the signal processing technique used by the Schoeps CMIT microphones has also been enlightening and informative. I was aware of the CMIT 5 U, but only noticed the superCMIT 2 U after checking for more detail on how the system works. These seem to be 'shotgun mics' which might actually have more application for music recording from a distance (as so often done around here due to setup constraints) than traditional shotguns that traditionally have such badly compromised off-axis behavior. Any thoughts on that?
Thank you very much for all your help with this!