Volthorne, on 05 June 2013 - 10:20 PM, said:
It seems to me you've got some things backwards... namely the stride length and weight multipliers. But I'm sure those were just typos, right?
No. The thing about mass is debatable, since more mass means more inertia, so less shaking, but also heavier impulses from steps, so more.. but longer strides mean more momentum, more momentum at impact means more impulse, so more shaking. The shaking frequency should of course decrease with longer strides, but the amplitude increases.
Optimum would be of course if the engine could actually handle the physics ( which are not that complicated in this case, classical mechanics with only 2 possible points where the force acts on the body, but still) and apply the impulses from steps correctley to a good ragdoll of your mech. Since the last part is pretty complex, aproximating your mech as cylindrical with an apropriately set center of mass should be enough.
Although I logn for the full version, where mounting an AC20 on one arm and nothing on the other would make you bounce more to one side thennthe other, and loosing an arm would move your center of mass correctley