Re: [osg-users] Help: can the location of the fragments of the image rendered by OpenGL pipeline be changed arbitrarily?

Hi, Like Chris Hanson said, you will need to use the new image write features and that would require writing to a different texture than the one that is currently bound to the camera. You could consider using a compute shader for this instead. Have a look at the osgcompute example. You can see

[osg-users] The benefits of the multithreaded modes or the lack thereof

Hi, There is this thing with the otherwise fantastic OpenSceneGraph that has been bothering me for some time and that is the relative ease with which I seem to lose the benefit of multithreaded rendering and how hard it is to fix. Our application runs with the default DrawThreadPerContext

[osg-users] Bug in initial renderer compile?

Hi, During development of a new shader-based point-rendering system I started getting INVALID_OPERATION at start of State::apply(StateSet*) for the first rendered frame (subsequent frames are OK). The scene consists of a geode, shapedrawable box (no stateset/fixed function pipeline) and the

Re: [osg-users] Bug in initial renderer compile?

Hi Robert and Nick, It does very much look like a missing cleanup after compilation. The following things need to be present: - a geometry shader - a uniform - rendering the node using the above after an object that is not compatible with the shader program. I have attached a modified version

Re: [osg-users] Bug in initial renderer compile?

Hi again, I should mention that I am currently on OpenSceneGraph 3.1.4. :-) Michael -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=57962#57962 ___ osg-users mailing list

[osg-users] Using integer format textures in an FBO attachment

Hi, I'm having trouble attaching integer format textures to an FBO camera. The following code: Code: #include stdafx.h #include osgViewer/Viewer #include osg/Texture2D int main(int argc, char* argv[]) { osg::ref_ptrosg::Texture2D targetTex = new osg::Texture2D;

Re: [osg-users] Reading from a texture in a compute shader

Hi, Robert Thank you! I will give it a try. Cheers, Michael -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=53521#53521 ___ osg-users mailing list osg-users@lists.openscenegraph.org

Re: [osg-users] Reading from a texture in a compute shader

Hi again, Robert That did the trick! I moved applyModeList and applyAttributeList down right before if(_shaderCompositionEnabled). I don't know if it produces any side effects. I tried running my application on top of the change and didn't notice anything. It uses osgEarth and Wang Rui's

Re: [osg-users] Reading from a texture in a compute shader

I have digged into this a little further and using glIntercept I learned that glDispatchCompute ends up being executed before the flowers texture has been bound. Adding a second quad to the scene with the flowers texture on texture unit 1 before the compute node fixes the problem, but is

[osg-users] Reading from a texture in a compute shader

Hi, I am trying to implement a histogram using the new compute shader support, but I don't seem to be able to read from a texture. I am new to compute shaders, so I might be missing something obvious. I have attached a modified version of the osgcomputeshaders.cpp file in which I simply load

Re: [osg-users] [3rdparty] SpeedTree 6.0 integration

Hi, hybr Thank you! I hadn't thought of the view and projection matrices - I was focused on still active shader programs (which I later ruled out, though). Also, I will try gDebugger to see if that can shed some light on the issue. Cheers, Michael -- Read this topic online

[osg-users] [3rdparty] SpeedTree 6.0 integration

Hi, everyone! Like others before us, we are trying to integrate SpeedTree into our OpenSceneGraph-based application and are having some trouble. We are using the custum-drawable approach and the trees are (finally) rendering fine. The problem is that, although simple objects like cow.osg

Re: [osg-users] Databasepager + multiple views + different camera positions = memory leak?

Hi, Robert! You are free to use my code as you wish :-) Do I need to download anything extra wrt the earth.ive you linked to? It does not contain any PagedLOD nodes (at least the statistics view does not report any), and so will not have any issues. Thanks for looking into the issue and

Re: [osg-users] Databasepager + multiple views + different camera positions = memory leak?

Hi everyone! This is a tough one. I'm now certain that there is more than one bug at large here! At least one in the pager (the one mentioned in the PagedLOD experts thread) and at least one in the txp plugin. The one I have been chasing for the past 2 days is within the txp plugin and looks

Re: [osg-users] Databasepager + multiple views + different camera positions = memory leak?

Hello everyone, We are having the exact same problem at our company for quite some time (currently on osg 2.8.0 via Delta3D and using TerraPage terrains). I have verified that viewing the terrain in the osgViewer works fine, but in our app, which also uses multiple cameras, the

Re: [osg-users] Particle system setInitialRotationalSpeedRange()

he has married. Then he's finished. 2008/12/5 Michael Bach Jensen [EMAIL PROTECTED]: Hi Thom, Thanks for the detective work. The change you've suggested as causing the regression is r7961. http://www.openscenegraph.org/projects/osg/changeset/7961 It's six months since I code this so

[osg-users] Particle system setInitialRotationalSpeedRange()

Hi all! Does anyone know if the bug mentioned in the following post will be fixed or if there is a workaround? http://www.mail-archive.com/osg-users@lists.openscenegraph.org/msg18069.html We use it, like Erik Johnson mentioned, to create a nice motion to smoke effects. /Michael *** Steve

[osg-users] Dramatic increase in event traversal time during mouse movement

Hello, group! I was looking at the nice statshandler graphics in osgViewer.exe, and noticed a dramatic increase in the time spent in the event traversal stage when the mouse was moved (goes from 0.02ms to around 8ms). Strangely enough, rotating the view by dragging with the mouse, almost does not