[ODE] Re: Contact Constraints: First Order Slip?

[Russ Smith]

> In other words, a sliding box will be slowed down until it reaches
> slip velocity, then it will continue to slide at that velocity
> forever. I guess this helps simplify constraint calculations.

ermmm ... it might look that way, but this is the wrong way to interpret
things. see the recently updated documentation.

> + it would be nice to have an accurate (conical) friction model

yes it would be nice - but it's rather hard to implement the way that
ODE is designed right now. but for most applications the friction
pyramid approximation model is visually indistinguishable from the real
thing.

> + some feedback from contact joints would be good.

this is not so hard to implement, there is already a dJointSetFeedback()
function that provides some information, but this is not exactly what
you want. it's on my to-do list to extend dJointSetFeedback() to provide
the computed force values *before* they are applied to bodies, so that
you can get contact normal force, motor force, etc.

russ.

-- 
Russell Smith
http://www.q12.org