[ODE] Incorrect depth on trimesh collisions

Joost van Dongen tsgoo at hotmail.com
Sun Aug 5 02:31:51 MST 2007


Hello Jon,
 
Thanks for the quick reply! I am not setting the previous transform, so that is something I will try now. However, I previously already tried moving to ODE 0.8 and I got linker errors when doing that. Compilation goes okay, but the linker gives me a number of errors that I do not know how to solve. I am using Visual Studio 7.1 and simply downloaded the win32 distributable from the ODE-site. Linking goes okay in 0.5 and 0.6, but this is what 0.8 gives me:
 
http://ronimo.hku.nl/Joost/ODE-0.8-Linker-Errors.txt
 
What am I going wrong here?
 
Thanks in advance,
Joost "Oogst" van Dongen
 
 
> Date: Sun, 5 Aug 2007 01:45:28 -0700> From: hplus-ode at mindcontrol.org> To: tsgoo at hotmail.com> CC: ode at ode.org> Subject: Re: [ODE] Incorrect depth on trimesh collisions> > First, you might want to upgrade to versio 0.8, which is the lastest > release.> > Second, are you setting the "previous transform" of the trimesh object > (not the data object) each frame? That's necessary to allow the trimesh > collider to give less arbitrary penetrations.> > Cheers,> > / h+> > > Joost van Dongen wrote:> > Hello all,> > > > I am working on a game with some physicsed trimesh objects, but they > > bounce through the level at extreme speeds. After a lot of research > > into what was happening, I found out that this happened because the > > depth that was reported by the collision detection is way too large, > > resulting in values like 6.1 when the real value was more like 0.05. I > > checked this by drawing the collisions in my game and they are indeed > > way too deep. I am using ODE 0.5 (the one destributed with the > > OGRE-engine) and am using the functions> > dGeomTriMeshDataBuildSingle1() and dCreateTriMesh() to create the > > collision detection objects. The collision positions seem to be > > totally okay, only the depths are broken. I think the problem only > > exists with trimesh-to-trimesh collisions and not with > > trimesh-to-sphere collisions.> > > > So my question is: what might be causing this? I have no clue > > whatsoever what I might be doing wrong, especially as the collision > > positions are not broken.> > > > Thanks in advance!> > > > Greetz,> > Joost "Oogst" van Dongen> >> > ------------------------------------------------------------------------> > Play free games, earn tickets, get cool prizes! Join Live Search > > Club. Join Live Search Club! > > <http://club.live.com/home.aspx?icid=CLUB_wlmailtextlink>> > ------------------------------------------------------------------------> >> > _______________________________________________> > ODE mailing list> > ODE at ode.org> > http://ode.org/mailman/listinfo/ode
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://ode.org/pipermail/ode/attachments/20070805/4ef5f72f/attachment.htm


More information about the ODE mailing list