[ODE] Friction and Collision.

[Michael Anderson]

I figured out what I was doing wrong. I was not setting the angular velocity
on my avatar to zero, but I was clearing it's rotation.

On 3/5/07, Michael Anderson <christiancoder at gmail.com> wrote:
>
> I have a simulation with default ERP and CFM using a fixed time step of
> 0.01s. I am playing around with contact parameters using capsules. If I
> set the friction coefficient to dInfinity all works as I would expect, but
> if I set the friction coefficient to 0 I get undesired collision penetration
> and corrective forces. I have played with the CFM and ERP values finding
> that increasing CFM causes the capsules to pass through each other and
> increasing ERP causes stronger (undesired) corrective forces. Is this the
> expected behaviour and if it is: is there anything I can do to avoid the
> collision penetration (other than decreasing my time step and slowing down
> my objects)? Also does Bullet's continuous collision detection help avoid
> these types of problems?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mooshika.org/pipermail/ode/attachments/20070305/05d94b53/attachment.htm