My initial thoughts are that the smoother, less peaky frequency response of the CAs is likely to be easier to EQ into a response I'd be satisfied with. Not saying the ATs couldn't be EQ'd to satisfaction, only that higher 'Q' resonant peaks like I'm hearing in the AT sample here can sometimes be more of a challenge to deal with than starting with a flatter "low Q" response. Sometimes I can correct those resonant peaks relatively easily with targeted EQ, sometimes not. On initial listen, the CAs seem like a potentially more fruitful starting point, an easier base-line raw response to work with.
I would love a further explanation of this, because it seems important and yet I only understand about half of it. In particular, when you say "higher 'Q' resonant peaks," what are you talking about here?
Basically 'Q' is a measure of how broad a peak or dip is. A peak / dip in frequency response response can be described by three primary characteristics: it's center frequency (the main frequency of interest), it's amplitude (the maximum level of the peak / dip), and it's Q (the width of the peak / dip). The higher the Q, the narrower the peak, the 'sharper' it appears in a visual display and the more resonant it is- that is, it takes longer for the energy stored in that oscillation to dissipate.
Low Q "flatter looking" curves are easy to work with and useful for general tone shaping as broad EQ boost / cut adjustments. With the application of low Q filters even very small amplitude changes are relatively easy to hear and those adjustments usually sounds relatively natural. An exact center frequency is less critical with a low Q filter because such a large range of frequencies are being effected.
High Q "sharp looking" curves are trickier to work with, are mostly useful for counteracting specific peaks or resonances in the frequency response. The overall frequency range they effect is narrow, and because of that changes are harder to hear even if they may be of relatively high amplitude. To successfully reduce an unwanted peaky resonance in the response by applying a matching inverse curve filter, the center frequency needs to match, the Q needs to match closely enough and the amplitude needs to be dialed in appropriately. Doing all that is tricker and takes more skill than applying broad, low 'Q' filters, which act more like "bass", "mid" and "trebble" tone controls. Countering a peak with an inverse filter usually entails finding the center frequency of the offending resonance, playing with the Q of the filter to find a setting which is about the same frequency width so only the resonance is affected by the filter and not too much or too little of the immediate frequency ranges around it, and playing with the amplitude of the filter to reduce the offending resonance enough without causing more problems than are being corrected.
It's easier to start with a relatively flat and even response and reshape that into to whatever response you want to hear. It's more difficult to carefully try an counteract peaky resonances before doing the overall response shaping.
A pitfall is that significant resonant peaks in the response can sound like broad tonal effects, especially to someone not yet skilled in hearing and identifying them as such- they can make the overall sound seem tonally 'brighter' or 'bassier' or whatever. A broad, low Q cut of the high frequency range will reduce the amplitude of a high Q high frequency peak, but also reduces the amplitude of all the other frequencies in close proximity of the peak. The problematic peak is still there, but the overall sound changes tonally. The original problem hasn't really been fixed only reduced in level a bit along with everything else around it, so the fix has produced a new problem of making the sound 'dull'. This is what makes EQ tricky, and the problem many tapers have in learning to hear the problem and fine tune the appropriate fix. It's also a primary contributor to the "sound" of different microphones, and how easily that can be corrected and adjusted or not.
A recording with a non-peaky response is "easier to EQ" even if it may not start out sounding exactly like what you want. A recording with peaky resonances may at first listen sound closer to the sound you want tonally, but can be much more difficult to work with in getting it just right.
The response of the microphone is a big contributor to this, but it's not the only contributor. The sound sources themselves, the PA, and the room acoustics, mechanical vibrations, and other stuff all contribute different resonant responses in the resulting recording. Yet this is a big part of why the output of high quality microphones often starts off sounding closer to what we want, and is easier to manipulate even when it is not close to exactly what we want, in comparison to mics which may at first listen sound tonally closer to what we want, but are far more difficult to manipulate with EQ.
Irregardless of audience chatter, that amount of height difference can be quite significant with regards to tonal EQ balance from some PAs and some rooms. I'd certainly not dismiss this variable outright, it may be quite significant here.
Next time I'll be sure to set them right next to each other. This was a last-second decision — oh, hey, I have these other mics I'm not using, I can run a comparison — and I didn't have time to mount the CA-14s on the stand right next to the AT853s.
I totally understand that. Glad you found enough time in the heat of the moment to run both, it's the best way to compare and figure out what works best in the real world.