The question about m_normalWorldOnB

Xuhai.Tang
Posts: 16
Joined: Wed Feb 23, 2011 4:54 pm

The question about m_normalWorldOnB

Post by Xuhai.Tang »

Hi,

I am trying to compute the force (F) at the contact point location. I define F=I*60, where I is the impulse. For the direction, I assign the normal to the master object (N):
N = m_normalWorldOnB
However, N is not always outward pointing. Does this make sense? Is there a better way of characterizing the force at the contact location?

Thanks in advance,
Xuhai
User avatar
Erwin Coumans
Site Admin
Posts: 4221
Joined: Sun Jun 26, 2005 6:43 pm
Location: California, USA

Re: The question about m_normalWorldOnB

Post by Erwin Coumans »

The normal is pointing from object B towards object A.

If this is not the case, please provide a reproduction case.
Thanks,
Erwin
Xuhai.Tang
Posts: 16
Joined: Wed Feb 23, 2011 4:54 pm

Re: The question about m_normalWorldOnB

Post by Xuhai.Tang »

Erwin Coumans wrote:The normal is pointing from object B towards object A.

If this is not the case, please provide a reproduction case.
Thanks,
Erwin
Thank you.

I have found the reason.
(1) I changed ‘#define CONVEX_DISTANCE_MARGIN btScalar(0.04)’ to ‘#define CONVEX_DISTANCE_MARGIN btScalar(0.00)’.
(2) after the above change, the nornal is wrong when the distance between getPositionWorldOnA and getPositionWorldOnB is smaller than 10e-6.

By the way, what is differences between CONVEX_DISTANCE_MARGIN and m_collisionMargin in the btCompoundShape.h?

Thank you in advance
Xuhai Tang