[ODE] LCP internal error, s <= 0

[Marco Grubert]

Never mind. I had changed the culling algorithm and now created two joints 
for each contact. One for body A hitting body B and one for B against A. 
Guess ODE doesn't like that..

- Marco

----- Original Message ----- 
From: "Bram Stolk" <bram at sara.nl>
To: "Marco Grubert" <mgrubert at conitec.net>
Cc: <ode at q12.org>
Sent: Monday, December 05, 2005 01:48
Subject: Re: [ODE] LCP internal error, s <= 0


> Marco Grubert wrote:
>> Hello,
>>
>> Despite not having made any major code changes in the past 6 months I am 
>> now
>> suddenly encountering lots of s<=0 error messages that have never occured 
>> in
>> the past years.
>> The LCP solver algorithm is beyond me and I am not sure what is causing 
>> this
>> issue. Could anyone suggest some good points to check so that I can pin 
>> down
>> the problem ?
>
> With what version of ODE?
> CVS, or CVS-Unstable?
>
> When did it first occur, after a cvs update?
> At what date?
>
> Can you run all the examples that come with ODE, to see if they
> all work?
>
>  Bram
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode