[ODE] Re: Box/trimesh collision problem

[Patrick Enoch]

Hello,

I have tried remembering not only the best, but also the second-best  
axis in dCollideBTL. But with no success,
the escape-normals arent getting better this way.
Has anyone else worked on this file, or any hints how to make the  
escape normals better? Or generate more escapes?
The escape computed is "correct", but not sufficient to prevent  
interpenetration.
The triangles returned by the collider are correct. The problem seems  
to lie in the interpretation.

Patrick



On 20. Jun 2005, at 10:35 Uhr, Maurizio de Pascale wrote:

> Hi Pierre,
> I was just talking more in general saying novodex support for  
> meshes (i.e. mesh-mesh) is not that good.
>
> bye,
> Maurizio de Pascale
> mdepascale at dii.unisi.it
>
> Pierre Terdiman wrote:
>> Yeah, pmaps are limited. But you don't need them for box-mesh....
>>
>> ----- Original Message -----
>> From: Maurizio de Pascale
>> Cc: ode at q12.org
>> Sent: Sunday, June 19, 2005 10:54 AM
>> Subject: Re: [ODE] Re: Box/trimesh collision problem
>>
>> Hi Jonathan,
>> I've never tested ode trimesh support but I can say the novodex's  
>> one isn't that good.
>> Even when using big pmaps you get a lots of interpenetrations.
>> And just to be clear, yes, I'm currently using novodex.
>>
>> bye,
>> Maurizio de Pascale
>> mdepascale at dii.unisi.it
>>
>> j wrote:
>>> Hi all Just as a comment: I've also been trying to get the box- 
>>> trimesh to work, and fiddled with anything short of tracing into  
>>> OPCODE but to no avail. I'd really like to learn about this  
>>> problem because I'm even considering switching to novodex because  
>>> of it and I'd hate to do that. Reading the posts on the subject  
>>> it seems to be a problem with the collision- detection. I believe  
>>> Pierre is on this list as well, perhaps he has a way ? regards,  
>>> Jonathan _______________________________________________ ODE  
>>> mailing list ODE at q12.org http://q12.org/mailman/listinfo/ode
>>
>> _______________________________________________
>> ODE mailing list
>> ODE at q12.org
>> http://q12.org/mailman/listinfo/ode
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>