To answer the question about vibration I would say that the DIMU feels about the same the AIMU in flight and has the same sensitivity to vibrations. I dont think its overall as bad as many people think.
But totally unscientific i can say that I had one copter that had the Z-acc saturation with AIMU. With DIMU it was still there, but not as pronounced and instead of gaining altitude rapidly it would make short bursts and recover. Still bad, but different...On that copter the cause was EPP0845 props that was grossly out of balance. Once replaced with HQP0850 problem went away completely for both IMU´s
DIMU offer options to explore the internal lowpass filtering, but as Norbert said there is only a handful of handbuilt DIMUs in existance at the moment and even team members are waiting for them, so focus has been on getting the basic stuff working and confirmed first. We have plans to test different settings and there is focus on tuning the UKF to handle vibrations and prevent runaways better, as soon as more people actually have them and we can do some broader testing.
But in the end the socalled "vibration sensitivity" in AQ comes down to the fact that at some certain amplitudes and frequencies the ACC data becomes corrupted by saturation and even though the UKF is very powerful, it just cant recreate data that is getting lost by saturation.
With a selfcorrecting filter, you simply need the input to be as clean as possible to acheive a solid estimation. This is absolutely no different than any other system out there - if data gets lost, the attitude and velocity estimations suffer and funny stuff begins to happen.
I will offer some fully assembled boards from Viacopter when the DIMU becomes available. We have done header assembly for custumers before and I will do it again if asked...