Right, resonance dodging and deflection compensation aren't going to cut it here, not by a long shot. I still think a sufficiently good 5 axis control loop could damp acoustics: for one, the (fractional) millisecond timescale of the acoustic oscillations are geological by compute standards, and while I'd expect limitations from sensing, actuation, and power electronics, I don't see anything fundamentally prohibitive. I'm pretty sure this is an engineering problem, not a physics problem, and that eventually someone is going to crack this nut. It won't be the Polar Bear.