[ODE] Simulation grinding to a halt

[Warrick Buchanan]

Yup I do as in the boxstack test.  Interestingly I upgraded to .35 and the
problem has disappeared?  Also before I did this I notcied the problem only
occured if I did not define a plane for the top of the viewing frustum (but
did for every other frustum plane) which is wierd.  But I guess as its gone
away with new release maybe it has been bug fixed?  The only problem I do
have now is that objects occasionally get stuck against the side frustum
planes but do eventually fall down - not sure why that happens?  I can check
to see if they've got lodged and give them a small kick but it seems a bit
strange as they are only lodged against the side planes - my simulation is
of a reasonable scale.

Thanks,

Warrick.

-----Original Message-----
From: Shaul Kedem [mailto:shaul_kedem@yahoo.com]
Sent: Sunday, February 09, 2003 12:22 AM
To: Warrick Buchanan; ode@q12.org
Subject: Re: [ODE] Simulation grinding to a halt


Hi,

 Do you call dJointGroupEmpty (dJointGroupID) ?

 ODE will slow down if too many contacts are created.
It will overflow the stack (on windows) if too many
contacts are created (but that has a solution which is
to use a bigger stack when linking).

 If you need more info I suggest reading past posts
about stack overflow and simulation slow-down,

 Which reminds me... what happened to the search
engine? :-)

Shaul



__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com