[ODE] non static deltaTime in the step (worldStep(world,deltaTime) instead of worldStep(world,0.05)) => problem?

[Frédéric Marmond]

hum...
As you can see in the little exemple I sent with my mail, it's all at the
same frequency!
I do a collision detection and just after the step=> if 100 collisions 
detection per second, 100 physics per second!

In this example, I run a very basic collision detection, but NOT THE ONE 
INCLUDED IN ODE.

?
Heu, I don't understand... Do you mean that you have a way to solve it, or do 
you have the same problem?
(I'm running it at office (shuuuut...) on a P3 500Mhz, and at home on a 
Athlon1800+ (1533Mhz))

Fred

Le Vendredi 11 Janvier 2002 15:44, vous avez écrit :
> Frédéric Marmond wrote:
> > The problem is that when deltaTime is too small, the ODE is very
> > instable.
> >
> > So, I introduced a little loop to slow down the prog.
> > deltaTime is bigger, and the ODE is ok.
> >
> > When i say that ODE is instable:
> > I use only one body, without force.
> > It start at Z>0.0 => it falls.
> > I only detect collision when it touch the Z=0 plane (ground)
> > After a bounce, it stabilize at this position (touch the plane Z=0)
> >
> > But, after a while, it moves again, and make a great jump (unless its
> > mass is 800 and the gravitation is -9.81!!!)
>
> I had a similar problem for a while. I traced the problem down to
> the fact that I was running ODE at 500Hz and my collision detection
> at the frame rate (roughly 100Hz).  When I ran the collision
> detection at the same rate as the physics, the problem went away.
>
> Since the subject has come up - is there a way to run the collision
> detection at a lower frequency than the physics? Should my split
> frequency example have worked?
>
> Jason
> 379

-------------------------------------------------------