Hi Paul,
There is indeed such a thing as mic shadowing, whereby the mic's "field of view" can be partially obstructed, but this can only yield such an audible effect if the obstructing object is within few centimeters from the mic. More importantly, there would need to be a transition between affected and non-affected keys, just as a shadow has blurry edges. I would venture a guess and say that the close1 mics where not even that close to the strings because I don't hear a proximity effect (as I do on the tube mic), so they must have been at least 40-50 cm from the soundboard.
Whatever the cause, whether shadowing, phase cancellation or anything else, the result is simply not desirable, so why not fix it?
If the entire piano is sampled all at once, there is indeed nothing that can be done per key, neither volume nor panning, but that's not the case here.
Thank you for fixing mid1 and mid2, but here too you've initially suggested that these have not been manipulated. In this case mic shadowing could not have been the cause because of the bigger distance to the braces. Maybe whatever caused the mid1 and mid2 panning issues also affected close1?
As a side note, I listen to a lot of piano music, jazz and classical, old and modern, and I don't remember ever hearing such panning side effects where, for example, A2 is farther to the left than A1, as it's the case with close1. Maybe you have a few examples?
Thank you