Hi, this submission resolves an issue regarding the setup of a slave camera with an (incorrect) coding style. Sharing the same instance of osg::Viewport with a camera and a slave camera causes incorrect rescaling in GraphicsContext::resizedImplementation, due to the viewport being rescaled twice (once per camera). Though viewports sharing is not intentional, it can be done with the current version of OSG and be a potential pitfall for anyone.
As Robert pointed out, i opted for modifying the resize code where the break is to minimize code changes, avoiding the duplicate resize of the viewport with the use of a vector and a search for duplicates. Not very elegant (avoiding an effect of a cause), another approach could be ripping out the method osg::Camera::setViewport(osg::Viewport*) which is more inline with Roberts rationale behind not to share viewports between cameras and left only its overloaded method setViewport(x,y,width,height). But this approach need some refactoring due to the intense use of the method. Notice also that the resize works well without this change if no sharing occurs, and the user of the method can opt for always call setViewport with a new instance. The modification are for OSG trunk version (revision 10260). Best regards, Himar. > Hi Himar, > > The assumption that I made with osg::Camera is that each camera would > have it's own Viewport, sharing a Viewport between Camera's is not > something I've considered so code like the GraphicsContext code for > passing up updates to resizing of the window will probably be resizing > the viewport onces for each Camera. > > Solutions would be to clone the viewport for your slave camera, or > modify the resize code so it doesn't resize viewports that it's > already resized. The later is really the best solution as it would > remove the assumption of one Viewport per Camera. I'm currently got > my head down on OpenGL ES work so if you could dive into the > appropriate code and provide a solution it'd appreciated :-) > > Cheers, > Robert. > > > On Mon, Oct 26, 2009 at 1:45 PM, Himar Carmona <[email protected]> > wrote: > > Hi, > > > > i noticed a strange (buggy?) behavior with two cameras sharing the > > same Viewport instance with version 2.9.5 (Windows XP). > > > > Situation: Main camera, adding a slave camera (with addSlave) and > > setting the slave camera's viewport with this line of code: > > > > slaveCamera->setViewport( viewer.getCamera()->getViewport()); > > > > The objective is to have two cameras rendering to the whole window > > > > Problem: If the viewer is in window mode (setUpViewInWindow) and > > the window is resized, the viewport updates incorrectly. In my case, > > it doesn't fill the window. > > > > I suspect this behaviour is due to > > GraphicsContext::resizedImplementation having updated the viewport > > twice, since it is share between the two cameras. > > > > Workaround: Don't share the viewport between two cameras :) > > > > slaveCamera->setViewport(new > > osg::Viewport(*(viewer.getCamera()->getViewport()))); > > > > or > > > > osg::Viewport* vp = viewer.getCamera()->getViewport(); > > slaveCamera->setViewport(vp->x(), vp->y(), vp->width(), > > vp->height()); > > > > If this use isn't as expected (i use it wrong) or if > > resizedImplementation need to be fixed, then i could try do the job > > and submit a patch for this odd behaviour. At least i could modify > > Camera and add some comments regarding this behavior. But first i > > wanted to know experts advise about it. > > > > > > Best regards, > > Himar. >
GraphicsContext.cpp
Description: Binary data
_______________________________________________ osg-submissions mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org
