I agree with Chris. The only way to make this kind of comparison is to remove all variables except the capsules, ALL VARIABLES. I have made close "stack" recordings of bluegrass bands in places like the Carolina Theater and the CVPAC down in Hickory, and found that even when running XY, the centimeter or so separating the caps on the vertical axis is enough to result in very slight harmonic differences between the channels, simply because this tiny spacing still places the capsules in different proximities to the individual drivers in the cabinet.
I suggest you do the living room test as Chris suggested, only swapping the caps on the same body left in fixed postion for each one, recording off one single speaker. I predict that you will find little difference between the caps. Also, you should take a look at the QA spectrogram sheets that are included with every capsule AKG sells with the capsule's serial number printed on it, to show that it met published specs before leaving the factory in Austria. I have looked at mine, and there are a couple of very tiny differences in amplitudes, but they remain with the +/- DB range specified on the data sheets.
Good Luck!