Thanks for posting these samples.
I very much like what the shuffling is doing in these examples, which to my perception increases the sense of dimension, openness, and naturalness, particularly in the low frequency range where it is being specifically targeted. The effective increase of the ratio of difference (Side) signal to sum (Mid) signal at low frequencies also serves to bring out more natural reverberation of the room that was already present in the recording but just not as apparent as with the shuffling applied.
Comparing the two mic configs, I also greatly prefer the 22cm AB omnis to the Gerzon cardioids. The AB omnis with the shuffling are just lovely. The Gerzon pair seems to act too much like section mics overly highlighting the violin section on the left and the cello and bass sections on the right, with too much separation between them and not enough of all the other instrumentation to properly glue those two halves together. Still, the shuffling improves the Gerzon configuration as well and in a similar way.
What the shuffling is doing here is exactly what I like to hear in my recordings, and why I gravitate to using larger omni spacings which in part naturally achieve what the shuffling is doing here. More specifically, that is achieving sufficiently minimal diffuse-field correlation or DFC at low frequencies. Shuffling decreases the inevitable low frequency correlation of diffuse field pickup inherent to a narrow mic spacing, while a wider mic spacing produces less diffuse-field low frequency correlation as a result of the additional spacing.
That the shuffling is working equally well for both of the configurations in Voltroni's examples reinforces my conception of shuffling being a potentially useful tool for most coincident and near-spaced mic configs, as it serves to "perceptually correct" for the overly high DFC of a near-spaced pair as frequency decreases, and for the lack of phase differences in a coincident pair.
The cavat in my using larger spacings is that is I then end up mixing in additional near-spaced or coincident channels to essentially "correct for" other less desirable stuff that is a result of using those wider spacings by themselves. The application of shuffling to recordings made using narrower spacings such as these achieves something similar by instead "correcting" the less desirable low frequency aspects that naturally result from a narrower spacing. Sort of opposite approaches which produce similar results in regards to the low frequency difference information in the recording.
If up for playing around with stuff, you might try adding a bit of shuffling at high frequencies and see if you like what that does in opening up the sense of "air" in the recording, while leaving the midrange alone to keep it dry, tight and perceptually closer.
