Crazy,
That kind of chimes with my own experience of using a 3 axis giro (not the AS3X) on a 3D model. It didn't help much with wing rock and in fact might actually make things worse in certain circumstances. I ended up snapping out of a harrier and crashing when using a giro.
I also found that the giro I was using (the FY-30A) would not cope with high energy tumbling type manoeuvres, i think the AS3X might be the same because by design a giro will fight any manoeuvre that's not commended by control surfaces. If you for instance try to do a snap roll using elevator and rudder the giro will try to fight the roll because neither rudder nor elevator is a roll command, as far as the giro is concerned the plane shouldn't be rolling!
I also found normal turns needed a very different approach with much more rudder input than normal. This is for the same reason, as far as the giro is concerned aileron is a roll commend not a 'turn command'.. So the giro applies opposite rudder and tries to keep the fuselage pointing in the original direction of travel! This makes for some really ugly turns until you get the hang of exaggerated rudder co-ordination.
How much you notice these effects depends entirely on how much gain you set, but if you want to get rid of them altogether you would have to set the gain so low that the giro would basically not be working. Where the giro worked great was in smoothing out general flight and in pattern type aerobatics, it worked wonders for that stuff. You do also have to be careful with gain settings, if you use optimum gain for slow/medium speed flight you run risk of getting violent flutter at higher speed, first time this happened to me i thought I would lose the ailerons.
we plan to do both to show the difference in the video
