Re[1]: [ODE] Negative penetration depths

[Thomas Harte]

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--=_NextPart_Caramail_0109651031181586_ID
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

Russ Smith :
> ODE's contact constraint says:
> "these two points on these two bodies must have a positive separation
> velocity as measured along the contact normal, and the force required to
> separate them must also be positive".
...
> ODE's notion of 'push apart' and 'suck together'
> depend on the contact normal and the penetration depth. so a negative
> penetration depth will suck bodies together, as will a flipped normal
> vector. many instabilities can be traced to bad contact generation that
> gets these signs wrong.

So adding a contact constraint at any time means that the object will not be able to move 
further along the normal vector, even if penetration depth is negative? Or will the 
pushing occur anyway as a result of the thing that isn't a penalty method in anything 
other than natural language? And what is the correct term for the maths you are using to 
reduce penetration if it is not a 'penalty method'?

-Thomas

P.s. sorry again for the other message appearing so many times.
______________________________________________________
Check out all the latest outrageous email attachments on the Outrageous Email Chart! - http://viral.lycos.co.uk	


--=_NextPart_Caramail_0109651031181586_ID--