[ODE] Trimesh collision: not all triangles are colliding

Sazzad Karim remon316 at hotmail.com
Wed Jan 14 11:26:08 MST 2004

  We are working 'only' on windows platform. We have managed to solve our 
problem using the SimpleTrimesh assignment function, but the normal one that 
takes vertices, indices and strides didn't work (we are sure about it). 
Maybe some trimesh functions have been unimplemented or have errors. But we 
are happy that it is woking now :) Hey you can try out the Simple function 
too and see if it works!! But i hope to see a new release from the ODE team 
fixing the trimesh system soon.
P.S. BTW compling the ODE library as a dll was quite a drag. My teammate 
Zahid compiled it by changing/adding many config settings and parameters

>From: Frederic Marmond <fmarmond at eprocess.fr>
>Reply-To: fmarmond at eprocess.fr
>To: Sazzad Karim <remon316 at hotmail.com>
>CC: ODE at q12.org
>Subject: Re: [ODE] Trimesh collision: not all triangles are colliding
>Date: Tue, 13 Jan 2004 09:58:14 +0100
>MIME-Version: 1.0
>Received: from intranet.eprocess.fr ([]) by 
>mc11-f14.hotmail.com with Microsoft SMTPSVC(5.0.2195.6824); Tue, 13 Jan 
>2004 00:54:41 -0800
>Received: from eprocess.fr (fmarmond.mtp.eprocess.fr [])by 
>intranet.eprocess.fr (8.9.3/8.9.3) with ESMTP id JAA26222;Tue, 13 Jan 2004 
>09:54:39 +0100
>X-Message-Info: JGTYoYF78jFjqsc4zJt9bY/k/jttSfpw
>Message-ID: <4003B326.9020700 at eprocess.fr>
>User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3.1) 
>Gecko/20030618 Debian/1.3.1-3
>X-Accept-Language: en
>References: <Law14-F64q89MjwPQBy00003a0a at hotmail.com>
>In-Reply-To: <Law14-F64q89MjwPQBy00003a0a at hotmail.com>
>Return-Path: fmarmond at eprocess.fr
>X-OriginalArrivalTime: 13 Jan 2004 08:54:41.0472 (UTC) 
>On which platform are you working?
>If yes, I found a bug in the OPCODE library, that forced me to compile the 
>collision part without optimisation for a file (-O0), on gcc-3.x 
>compilator. Worked fine (-O3 for all files) with a gcc-2.95 one. The result 
>was as you said, some triangles collides, others don't as they would have 
>been colliding.
>If you are in this situation, ask me, and i'll try to find in my archives 
>precisaly which caused the problem, and a test-case that reproduce the 
>Sazzad Karim wrote:
>>My team is testing with ODE for collision in our game; we have used OBB 
>>bounding boxes for all static objects and terrain mesh for the terrain. 
>>The OBB player to other OBBs work fine, but just as we integrated the 
>>trimesh part it is not working, some triangles in the mesh collide but 
>>most don't! Are we doing something wrong or havent initialized something 
>>or ODE have some unimplemeted function on trimesh? As far as i'm concerned 
>>correct data is passed to the trimesh of ODE. If you've encountered such a 
>>problem please help.
>>M Sazzad Karim
>>Game Programmer
>>MSN 8 with e-mail virus protection service: 2 months FREE* 
>>ODE mailing list
>>ODE at q12.org

Protect your PC - get McAfee.com VirusScan Online 

More information about the ODE mailing list