On Mon, Aug 16, 2010 at 9:24 PM, Jolley, Thomas P < thomas.p.jol...@boeing.com> wrote:
> Hi Serge, > > You may want to research the following posts and see if they are relevent > to your particle problems. > > * > http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/2009-April/027011.html > * > > http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/2009-May/028435.html > > The above talks about a floating point precision bug that was introduced > due to a change in 2008. The bug hasn't been resolved yet. I suspect it > will wait until the osgParticle is refactored since not many see the > problem. If the above describes your problem you will need to undo the 2008 > change. > > > http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/2009-October/033416.html > > > I'm willing to reconsider the change in question, revision 8881. Thomas Jolley and I came up with different approaches to solve the problem of a model leaving particles behind in a geocentric coordinate system. My problem was in the FlightGear flight simulator; given Thomas' mail address, I imagine his application isn't too different. Anyway, his solution didn't require any changes to OSG; mine does, and the changes break his solution. I now see a couple of additional problems that make my solution -- rooting the particle system in the world frame -- less than ideal. One is that ModularEmitter::emitParticles() has to make calls to getWorldMatrices, a potentially expensive function. The other is that making the particle system sit under an independent node in the scene graph makes it more complicated to clean up when the model + particle system is removed from the scene graph and deleted. So, I suggest that we revert that change and document Thomas' suggestion on the wiki as the preferred way to approach this kind of particle system. It doesn't seem like many other people were affected one way or another by the change. Tim
_______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
