I resolved the problem by updating to HEAD. Thanks all!
On Tue, Oct 8, 2013 at 7:24 AM, Jason MacDonald jnm...@gmail.com wrote:
Hi,
That is the exact build that changed things up. Revision 13710 fixes the
uninitialized epsilon, but revision 13753 changed things up again.
Cheers,
Jason
Hi,
That is the exact build that changed things up. Revision 13710 fixes the
uninitialized epsilon, but revision 13753 changed things up again.
Cheers,
Jason
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=56671#56671
Hi,
I'm not sure of the details of your problem, but I recently had a problem with
intersections as well.
The epsilon value was changed in LineSegmentIntersector.cpp line 410.
I changed back to 1e-4 from 1e-13 and it solved my problems.
Cheers,
Jason
--
Read this topic online
Hi Jason,
In svn/trunk and OSG-3.2 branch I've changed the epsilon to 1e-6 to handle
axis align planar bounding boxes.
Robert.
On 7 October 2013 19:26, Jason MacDonald jnm...@gmail.com wrote:
Hi,
I'm not sure of the details of your problem, but I recently had a problem
with intersections
Thanks for the tip. I'm see an epsilon of
-6.2774385622041925e+066
which looks uninitialized to me. I'm going to dig further. I last built
revision 13699. Maybe I'll just update and rebuild everything and see if
the problem is already resolved.
On Mon, Oct 7, 2013 at 1:49 PM, Robert Osfield
Some working code of mine stopped working when I upgraded to OSG 3.2 from
a previous 3.x dev release.
I'm using
bool computeIntersections(float x,float y, const osg::NodePath nodePath,
osgUtil::LineSegmentIntersector::Intersections
intersections,osg::Node::NodeMask traversalMask =
Hi Chris,
The problem with the computeIntersections() method is that it's ambiguous
w.r.t which window and which camera the x,y coordinates are associated
with, the function has a stab at guessing what you might mean. If you are
seeing a regression then it sounds like this guess might be what
Ok. I switched to the non-deprecated method I believed was correct, and I'm
still seeing a failure, so I need to dig through and see what's going on.
On Fri, Oct 4, 2013 at 2:33 PM, Robert Osfield robert.osfi...@gmail.comwrote:
Hi Chris,
The problem with the computeIntersections() method is
8 matches
Mail list logo
