[ODE] Confused on the results from dJointGetFeedback on a slider.

[Sten Anders Karlsson]

Hello!

I use a slider that has  dParamVel = 0 and dParamFMax = <some large value>
to make the joint not moving. 

I now use the dJointGetFeedback to see what forces the slider applies
to the joined bodies.

The odd thing is that I get different results from the dJointGetFeedback
depending what axis orientation I have on the slider.  This would be OK if
the slider allowed movement, but it does not move.  I would instead expect
that a non moving joint would affect the joined bodies with forces the
same way as a fixed joint.

Or, now as I write this I come think of a possibility, is it maybe so that
a non moving slider still has a degree of freedom, like a hinge, the
bodies can rotate around the slider axis?  

And just an other question - that unfortunately reveals my lacking
geometry abilities - but wouldn't it be correct to use the x,y,z forces
from the feedback struct and just multiply (dot product) the with the
slider axis vector x,y,z to get the forces that acts in the sliders axis
direction, asuming the slider axis is pointing in the direction from body1
to body2 (all torques are perpendicular to the slider)?

Or, what I realy want to do is to construct a spring-damper and start of
by finding out what the initial load is on the spring, before I allow
anything to move around, so that I can preload the spring with a good
starting value.

Any feedback (pun intnd.) is appreciated. 

	-- Anders