Hi Max,

I dont thing this is a bug since I have been using this since years ago, and
I am pretty sure it works ok. I will have a look at this after the holidays.
The thing I noticed was that your light point didnt look like light point I
am used to see in osg.

Glad it works though
Nick

http://www.linkedin.com/in/tnick


On Wed, Dec 30, 2009 at 11:30 PM, Max Bandazian <[email protected]>wrote:

> Thanks, Nick. I tried this out and it does work around the issue, and
> so far I haven't seen a significant performance hit.
>
> This is clearly a bug, though. The lights in the .osg file are set to
> have a minimum 3 pixel size (I have no idea if this means 3 pixel
> radius, 3 pixel diameter, or what), but even if I set the small
> feature culling size to < 1 pixel they get culled out. That strongly
> indicates that the culling algorithm is using the bounding sphere to
> decide how many pixels the geometry will have, and in this case that's
> just wrong.
>
> I'd like to suggest a fix, but I'm just not familiar enough with the
> culling code to do that. I hope someone more experienced can take a
> look at it.
>
> Max
>
> On Mon, Dec 28, 2009 at 4:33 PM, Trajce Nikolov
> <[email protected]> wrote:
> > Hi,
> > I did quick test here. You should disable small feature culling. Have a
> look
> > at
> > viewer.getCamera()->setCullingMode(osg::CullSettings::
> > Nick
> >
> > http://www.linkedin.com/in/tnick
> > Sent from Ünalan, İstanbul, Turkey
> >
> > On Mon, Dec 28, 2009 at 11:08 PM, Max Bandazian <[email protected]>
> > wrote:
> >>
> >> Hi,
> >>
> >> I'm having some trouble with the culling of an osgSim::LightPointNode.
> >> They seem to be getting culled out at some arbitrary distance,
> >> although I could also be misinterpreting what the class is supposed to
> >> do (the header is almost entirely uncommented, so...)
> >>
> >>  The attached .osg file has one triangle and one light point. I'd
> >> expect that the lightpoint stays visible until it's either behind the
> >> far clipping plane or the eyepoint is more than
> >> sqrt(maxVisibleDistance2) away, but in reality it stops getting drawn
> >> when the eyepoint is ~1500 away. I'm testing in osgviewer, with
> >> version 2.8.2. Is this somehow the correct behavior, or is there a bug
> >> in the culling?
> >>
> >> Thanks,
> >> Max Bandazian
> >>
> >> _______________________________________________
> >> osg-users mailing list
> >> [email protected]
> >>
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
> >>
> >
> >
> > _______________________________________________
> > osg-users mailing list
> > [email protected]
> >
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
> >
> >
> _______________________________________________
> osg-users mailing list
> [email protected]
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>
_______________________________________________
osg-users mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Reply via email to