Hi Sonya >but what happens if mouse is close enough and supposed >to be snapped to vertex but mouse is not on object.?
I can think of two things - set bounding sphere to be larger so that you intersect object when you click near object ( see http://vtp.googlecode.com/svn-history/r7601/trunk/TerrainSDK/vtlib/vtosg/NodeOSG.cpp as an example), then create custom intersector to suite your needs see custom intersectors here http://www.packtpub.com/sites/default/files/downloads/6884_Chapter10.pdf Regards Sergey On Sat, Jul 26, 2014 at 11:11 PM, Sonya Blade <sonyablade2...@hotmail.com> wrote: > Hi Sergey, > > Thanks for the tutorial, but I believe that the problem is: Casted ray > with world_mouse = screen_mouse*inverse(MVPW) > matrix manipulation is unware of whether it crossed an object or not, it > simply returns near/far points of frustum. > On the other hand using, the" LineSegmentIntersector >> > getWorldIntersectionPoint() " and example in *OSG cookbook* > *Ch3 Selecting a point on the model** solves* the issue of what is behind > the mouse but I believe that it has drawbacks. > Because this tecnique requires the mouse to be hoovered on object, but > what happens if mouse is close enough and supposed > to be snapped to vertex but mouse is not on object.? > > The above is my intuitive reasoning , I tried to put it in succinct manner > and hope that I well versed it. If it is not clear I'll try to > ealaborate it with images. > > Regards, > ------------------------------ > Date: Sat, 26 Jul 2014 14:11:38 +0400 > From: sergey.fo...@gmail.com > To: osg-users@lists.openscenegraph.org > Subject: Re: [osg-users] Mouse World Position in Perspective View > > > Hi Sonya, > > I think you could solve your problem if you understand theory what goes > during the whole process and test if anything works from simplest examples, > then after tests you'll uncover a problem > > for theory I think following to links are useful, but see not only first > link, but also second > > http://www.mvps.org/directx/articles/rayproj.htm > http://www.mvps.org/directx/articles/improved_ray_picking.htm > > > Regards > Sergey > > > On Sat, Jul 26, 2014 at 1:53 AM, Sonya Blade <sonyablade2...@hotmail.com> > wrote: > > Dear All, > > I'm having a hard time getting the correct results for the mouse position > in world coordinates. > The problem is: retrived world mouse position is not even close to those > when I hoover over an > object so I couldn't find any correlation between to which vertex of > object is mouse close to. > > To obtain the world mouse position I use world_mouse = > screen_mouse*inverse(MVPW), for obtaining > the vertex position I use LineSegmentIntersector >> > getWorldIntersectionPoint() and there is absolutey > not any coherent value that I can find any relation. My aim was to snap > the mouse to the closest vertex > in object, but produced results are as follow. > > world coords vertex(6.97538 20.55 18.7482" > local coords vertex(0.876922 -3.74963 4" > Mouse Position -101.794 44.6545 -130.128" > > Your help will be appreciated, > > _______________________________________________ > osg-users mailing list > osg-users@lists.openscenegraph.org > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > > > _______________________________________________ osg-users mailing list > osg-users@lists.openscenegraph.org > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > _______________________________________________ > osg-users mailing list > osg-users@lists.openscenegraph.org > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > >
_______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
