[ODE] Contacts not constraining user forces ?

[Marco Grubert]

Thank you for your tips. It looks like I was wrong and the behavior is not
related to user forces vs. gravity as I can reproduce it with certain
gravity values and no other forces as well. Even though contact points are
generated and the penetration depth looks about right, in some cases the
constraint force is not large enough to push the body out of the (planar)
mesh. Smaller timesteps obviously help because then penetration won't be as
deep but I was trying to reduce the physics frequency for better
performance. Next I will check how CFM influences things.

Thanks,
  Marco

----- Original Message ----- 
From: "Erin Catto" <erincatto at sbcglobal.net>
To: "'Marco Grubert'" <mgrubert at conitec.net>; <ode at q12.org>
Sent: Wednesday, April 20, 2005 23:13
Subject: RE: [ODE] Contacts not constraining user forces ?


>A few things to try:
>
> 1. Set the contact CFM and global CFM to zero.
> 2. Depending on your masses, that force may be very large compared to
> gravity. Try reducing the force by factors of 10.
> 3. Is a contact point still being generated in the second version? If so,
> is
> the normal and penetration the same?
> 4. Try the same setup with a plane or box instead of the mesh.
>
> Erin