I've already done my research on the Droid X, the problem doesn't exist - at least not to the extent that it does on the Apple iPhone and exposed impedance areas. Holding the iPhone 4 'naturally', I could generate the problem very easily and repeatedly. I could not duplicate it on the Droid, Droid X, Incredible, or EVO. But then again, its hard to test because I couldn't run the Droid X on an inferior network like AT&T to get raw data off a network that barely has connectivity to begin with.
I think the best 'proof' of the problem was demonstrated by Brian Klug with real raw numbers, not bars. Maybe Motorola should "fix" the problem by changing the way it displays bars.