noguy right you are about the vibration issues.
Too much vibration of the right (wrong) frequency can actually break the little tuning fork in the fork-type gyros. I would bet the ring gyro is less prone to this kind of damage. But since both the gyro types actually work by sensing a type of vibration, it stands to reason that they should be isolated from as much external vibration as possible for the best performance.

And the eflite parts are really not too accurate. As a micromachinist used to working to 0.0001 inch tolerances, the slop, sorry, play in the whole head system seems incredible to me. But now I realize it has to be that way to work properly. But the vibration and run-out (the technical term for bendage in shafts) in brand new parts is getting to be intolerable.
Many people upgrade to the aluminum CNC parts for this reason alone.
And then they find out that the aluminum "CNC" parts are just as bad.
Just kidding, my bad experiences probably aren't typical, but this brand new MicroHeli swashplate has more play in the main bearing than the stock plastic one I took off that probably has 25 hours flight time on it.
And the brand new MH aluminum swash that I installed on my CX separated--as I was installing it. Better then than in flight, but I sure was po'd.
The center hub can make the shaft look bent when it isn't. I always try both orientations of this part on the shaft to find the best way around.
And the variation in height of the shear pin hole can cause collective pitch problems even if nothing else is changed.