[ODE] movement force conversion

Piotr Obrzut piotr_obrzut at o2.pl
Tue Apr 12 15:23:36 MST 2005


PO> Without division by time forces are too low, but you have right that they will
PO> have to be multiplied by some constant value. I will have to change my
PO> priorities: before animation suposed to be right made and physics
PO> should only follow it; now I think that it will be better to store
PO> animations within physic model and this model will be responisble
PO> for executing animations and geting its bodies into right positions
PO> and physics state (so it would be animation_time independent and forces
PO> could be precomputed only once); Perhaps I will face problems with animation
PO> blending but it is not importand for me now.

PO> Once again : great thanks!

Oh and I forgot: I will still have to get desired position from
animation. So now questions are:

- How could I connect mentioned constant value with delta_time which
should be passed to animation system to get apropriate pose (so the outputs
from physics and animation will be as equal as possible without outside factors)?
(I hope that there is other way than just to experiment :) )

- When it will be best to add apropraite forces? Before each step?
(till now it was connected with animation time so it was called as
rather as possible).

 Piotr Obrzut

More information about the ODE mailing list