WiFiJeff, I don't wish to defend anyone's bad attitude, but by itself a rolloff in a microphone starting at 18.5 kHz is little cause for concern. Some of the ribbon mikes that some people think are so f'ing great roll off starting at 7 kHz, and you might be amazed to see the actual high-frequency response of some very highly regarded studio condenser microphones (particularly off-axis) as opposed to their published spec-sheet curves.
But as I said this is little cause for concern since at such high frequencies, there is rarely any significant sound energy even in "acoustic" music at normal recording distances--and if you record amplified music, forget it--there's essentially none at all.
You remarked on the fact that this person sells 24/96 kHz equipment. For the past five years or more it has become impossible to market any digital recording system that doesn't support at least 96 kHz. But the point of it was never to capture higher frequencies than 44.1 kHz can record; it was to capture the audible range more accurately. Any PCM audio system requires low-pass filters to exclude any frequencies at or above one-half the sampling rate. A sampling rate of 44.1 kHz, while objectively adequate for all frequencies audible to humans, leaves only the very small interval between 20 kHz and 22.05 kHz for this filtering to take effect.
The main justification for 96 kHz recording, such as it is, is to allow those filters to begin their effect higher in frequency (above 20 kHz) and to have gentler slopes. This allows for flat frequency response to 20 kHz plus better impulse response in the audible range, both of which can be measured and occasionally even heard by some people on some material when listening on electrostatic headphones--though not by most people on most material, and certainly not on any normal home loudspeaker systems.
Please pardon me if you already knew all this; I'm just trying to contribute a little perspective. I might have been interested in that system as well, except that having great versatility in processing the signals from crappy capsules ultimately isn't going to produce any results to get excited about. It was always the problem with the original Soundfield system (I'm thinking 20+ years ago) that its Calrec capsules just weren't really very good sounding, though I understand that the more recent ones are quite a bit better. Still, I have no use whatsoever for height information in a recording, and would rather use a mike with three really good capsules and correspondingly simpler processing.
--best regards