Re: [osg-users] current View Point
i have two viewers and put the scene in groups one for the whole scene and one
for the pointer i want to move:
osg::Group* root = new osg::Group; // groupp have all other objects
osg::Group* Mainroot = new osg::Group; // the whole scene
Mainroot-addChild(root);
osgViewer::Viewer viewer;
osgViewer::Viewer viewer1;
//set left camera for left view
osg::ref_ptrosg::Camera cameraL = new osg::Camera;
...
// add the matrix transformation
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix( cameraL-getViewMatrix() );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
viewer.setCameraManipulator( keyswitchManipulatorL.get() );
//set right camera for the right view
osg::ref_ptrosg::Camera cameraR = new osg::Camera;
viewer1.setCameraManipulator( keyswitchManipulatorR.get() );
//load data for each viewer
viewer.setSceneData( root);
viewer1.setSceneData( Mainroot);
//run the viewer
while( !viewer.done() )
{
// fire off the cull and draw traversals of the scene.
viewer.frame();
viewer1.frame();
}
now the object i used did not move what is the problem ??
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] current View Point
To: olfat_ibra...@yahoo.com, OpenSceneGraph Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 2:46 AM
Hi,
Please, make replies to your posts ! do not make a new
thread each time ...
If you reply to your post, we will have all the mailing
history of this
question, and we can see the evolution of the problem.
Next, just with this code, we cannot see what the problem
is.
Tell us what is good in your code, what you want to do with
this code...
For your problem : maybe it works, but you need a callback
or a loop to see
the matrix changing in real-time... Not easy to understand
this little peace
of code without more explanations.
Regards,
Vincent.
2008/12/16 olfat ibrahim olfat_ibra...@yahoo.com
hi :
iam creating a program that have two windows one i
make a drive in it and
the other is like a map i want to put indecator in the
2nd window that
indicate my current position in the first one
i used the following code but nothing happend:
// Projection and ModelView matrices
osg::Matrixd proj;
osg::Matrixd mv;
...
proj = cameraL-getProjectionMatrix();
mv = cameraL-getViewMatrix();
// Create parent MatrixTransform to
transform the view
volume by
// the inverse ModelView matrix.
osg::MatrixTransform* mt = new
osg::MatrixTransform;
mt-setMatrix(
osg::Matrixd::inverse( mv ) );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
...
can some one tell me where iam going wrong ??
tahnks
___
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
Re: [osg-users] current View Point
Each frame, you have to update your matrix. For the moment you set it, next
you frame() the viewer, nothing more. Matrix will no update itself.
To update it, do it in the viewer loop (after/before frame()), or create a
new nodeCallback. Have a look on nodeCallback to see how to use it.
Regards,
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
i have two viewers and put the scene in groups one for the whole scene and
one for the pointer i want to move:
osg::Group* root = new osg::Group; // groupp have all other objects
osg::Group* Mainroot = new osg::Group; // the whole scene
Mainroot-addChild(root);
osgViewer::Viewer viewer;
osgViewer::Viewer viewer1;
//set left camera for left view
osg::ref_ptrosg::Camera cameraL = new osg::Camera;
...
// add the matrix transformation
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix( cameraL-getViewMatrix() );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
viewer.setCameraManipulator( keyswitchManipulatorL.get() );
//set right camera for the right view
osg::ref_ptrosg::Camera cameraR = new osg::Camera;
viewer1.setCameraManipulator( keyswitchManipulatorR.get() );
//load data for each viewer
viewer.setSceneData( root);
viewer1.setSceneData( Mainroot);
//run the viewer
while( !viewer.done() )
{
// fire off the cull and draw traversals of the scene.
viewer.frame();
viewer1.frame();
}
now the object i used did not move what is the problem ??
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] current View Point
To: olfat_ibra...@yahoo.com, OpenSceneGraph Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 2:46 AM
Hi,
Please, make replies to your posts ! do not make a new
thread each time ...
If you reply to your post, we will have all the mailing
history of this
question, and we can see the evolution of the problem.
Next, just with this code, we cannot see what the problem
is.
Tell us what is good in your code, what you want to do with
this code...
For your problem : maybe it works, but you need a callback
or a loop to see
the matrix changing in real-time... Not easy to understand
this little peace
of code without more explanations.
Regards,
Vincent.
2008/12/16 olfat ibrahim olfat_ibra...@yahoo.com
hi :
iam creating a program that have two windows one i
make a drive in it and
the other is like a map i want to put indecator in the
2nd window that
indicate my current position in the first one
i used the following code but nothing happend:
// Projection and ModelView matrices
osg::Matrixd proj;
osg::Matrixd mv;
...
proj = cameraL-getProjectionMatrix();
mv = cameraL-getViewMatrix();
// Create parent MatrixTransform to
transform the view
volume by
// the inverse ModelView matrix.
osg::MatrixTransform* mt = new
osg::MatrixTransform;
mt-setMatrix(
osg::Matrixd::inverse( mv ) );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
...
can some one tell me where iam going wrong ??
tahnks
___
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
Re: [osg-users] Removing the tgz, osgtgz and zip plugins
I doubt very seriously that it would take much effort to port these to a less hackish method. If I get some free time, i'll look into doing it. It is only necessary to make war with five things: with the maladies of the body, with the ignorances of the mind, with the passions of the body, with the seditions of the city, with the discords of families. - Tacitus The desire for safety stands against every great and noble enterprise. - Tacitus Those who would give up essential liberty to purchase a little temporary safety deserve neither liberty nor safety. - Benjamin Franklin Numquam ponenda est pluralitas sine necessitate. - William of Ockham On Wed, Dec 17, 2008 at 1:42 AM, Gordon Tomlinson gor...@gordon-tomlinson.com wrote: I use them and would not want to lose them, IVE, OSGA is not a suitable replacement for me I don't see why they need to removed they work( maybe not to everyone's liking but they work ), they're useful to some As to re-implementation , who's going to pay my company for the time and expenses to re-rite these go through all r Testing ands sign off procedures ? Why change something that is not broken, might not be the best solution but they work Capture the magic of Christmas this year see http://www.capturethemagic.com __ Gordon Tomlinson gor...@gordontomlinson.com IM: gordon3db...@3dscenegraph.com www.vis-sim.com www.gordontomlinson.com __ -Original Message- From: osg-users-boun...@lists.openscenegraph.org [mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Jason Daly Sent: Tuesday, December 16, 2008 11:56 PM To: OpenSceneGraph Users Subject: Re: [osg-users] Removing the tgz, osgtgz and zip plugins Ulrich Hertlein wrote: I agree that the implementation is hacky but the functionality is definitively useful (speaking for zip files only): - they are a common way to distribute a complete model (non-OSG objects, textures, etc.) - they can be easily created by the artists, no need to have OSG installed - usable by other packages; .osg, .osga, and .ive are leaf file formats if you will That, and couldn't they be re-implemented using zlib (which is already necessary for the png plugin)? --J ___ 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
Re: [osg-users] [osgAL] Link error in release
Hi Anna, It was so obvious that I didn't see it! Yes, adding the lib into the project fixed it. I was so sure that could not be so simple that I didn't even checked it. Thank you very much. Sukender PVLE - Lightweight cross-platform game engine - http://pvle.sourceforge.net/ Le Wed, 17 Dec 2008 00:28:42 +0100, Anna Sokol annaso...@gmail.com a écrit: Hi Sukender, unresolved external symbol unsigned int __thiscall OpenThreads::Atomic::operator++() referenced in function osg::ref_ptrosgDB::DotOsgWrapper osg::ref_ptrclass osgDB::DotOsgWrapper::operator=(osgDB::DotOsgWrapper *) The above error occurs because OpenThreads is not included in the library dependencies for osgAL. This didn't cause any errors with osg2.4 but did cause compile errors with osg2.6. and osg2.6.1 . By including OpenThreads library dependency the compile errors went away with osg2.6 and osg2.6.1 . I haven't tried to compile with osg2.7.x or the osg SVN. There might be additional incompatibilities there. I used Microsoft Visual Studio 2005 to compile. Hope this helps. - Anna ___ 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
Re: [osg-users] Help Rendering and Reading Floating Point Textures
Brian, JP, The osgmultiplerendertargets example uses (by default) GL_FLOAT_RGBA32_NV, rather than either GL_RGBA32F_ARB, or even GL_RGBA16F_ARB. Could this be a card issue? I would be interested to know whether this example works for you if you change to GL_RGBA32F_ARB or GL_RGBA16F_ARB. I admit, I haven't tried it myself yet - and now intend to - but certainly in my own app (on ATI), I have had issues with GL_RGBA16F_ARB and GL_FLOAT in the past. Things might be better now. David ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg firefox plugin alpha
Hi all, the LD_LIBRARY_PATH issue for the linux plugin is still annoying. I tried to use a proxy shared lib, which sets up an appriopriate path using setenv() and then calls dlopen() on the plugin lib. But it fails because LD_LIBRARY_PATH is only evaluated once during startup of the firefox program. It cannot be manipulated later. any more ideas? regards Ralph Ralph Kern schrieb: Hi Andreas, I'm making some progress in the linux port. I already can see the models on your page, but only if I start firefox with LD_LIBRARY_PATH set to the plugin directory. Now I need to search for some hack to port your library search path extensions to linux. This is not easy, because the plugin is not lazily loaded like in Windows. And there's no SetDllDirectory like in Windows. So I need to find a way to patch the search path before the FF plugin is loaded, otherwise the load fails because it cannot locate libosgdb.so and libosgViewer.so. I think the right path is to have a proxy plugin, which manipulates the path and then loads the correct plugin using dlopen(). If someone knows a better way, let me know! Regards Ralph Andreas Goebel schrieb: Hi, I have changed a little bit in the xpi and in the plugin itself. It now not only adds the plugin-directory to the osg-file-path, but to the windows-dll-search-path. This worked at least on my machine to enable .ive-support without having the osg-libs in your path. Please: Those people who didn´t see the lower three models, uninstall the old plugin, install the new one and try. Note that that function (named SetDllDirectory ) is only available from WinXP SP1 onwards, so this plugin won´t work with older windows versions. I think that only very few people are using older windows now. Hope this works ... Regards, Andreas ___ 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
Re: [osg-users] Build error on MAc OSX 10.4 and OSG 2.7.4
Hi Paul, I believe this one has already been addressed in the 2.7.x dev series, could you try 2.7.7 or svn/trunk. Robert. On Wed, Dec 17, 2008 at 4:11 PM, Paul Fotheringham osg_u...@yahoo.co.uk wrote: Hi, I think I know what the problem is. I'm getting the same unresolved symbols building on MacOS 10.4 and OSG 2.6.1 using cmake. Your link line has /usr/lib/libm.dylib rather than your SDK version of the library. I don't know much about these things myself as I've only been using a Mac for about 1 day now but this blog entry details the issue http://www.red-sweater.com/blog/246/missing-link As far as cmake goes, and I've only been looking at _that_ for about 2 days, it seems the problem is in the top-level CMakeLists.txt file where it has the line FIND_LIBRARY(MATH_LIBRARY m) wrapped in an IF(UNIX) clause. I think this particular line should also be wrappend in an IF(NOT APPLE) clause. Doing this fixes up the link lines for me and the problem goes away. As I say I'm certainly no Mac expert so maybe this isn't quite right. I'm certainly not confident enough to post it to osg-submissions :) I haven't looked at the Xcode project at all but I imagine something similar must be going on there? Paul Fotheringham ___ 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] An observation on the 2.7.7 dae plugin build
I don't use the osg third party sources, preferring to keep my own copies of dependant libary sources. So I switch on the advanced mode of CMake and fill in my own include and library locations. I filled in (I think they may actually have been filled in automatically, but I am not going back to check) COLLADA_DYNAMIC_LIBRARY, COLLADA_DYNAMIC_LIBRARY_DEBUG, COLLLADA_STATIC_LIBRAY, COLLADA_STATIC_LIBRARY_DEBUG and I set COLLADA_USE_STATIC to OFF. However I still got a static build, which gave me errors as it could not find my libxml2 library (another problem!). The only way I could get a dynamic build was to set the entries for the the static libraries to empty. I find this a bit confusing as the help for COLLADA_USE_STATIC says "Set to ON to build OpenSceneGraph with static Collada support". Enjoy, Roger ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] Bypassing the use of the CURL plugin for all http requests
Hi Robert, Is there a way to bypass the use of the CURL plugin for all http requests? For most cases this is nice, but I'm working on a few plugins where it would be very beneficial to simply have the full URL passed in so that I can deal with it myself rather than having CURL try to download the file. The workaround I've been using currently is sticking a prefix on the address and having my plugin strip it off before it uses it. So, something like osgviewer server:http://www.server.com/myfile.extension. This works well so far, I'm just curious if there is a better way. Maybe an option in ReaderWriterOptions? Thanks! Jason ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgPPU v0.3 cycles problem (Art Tevs)
It does work fine. Thanks Hi Alexandre, The problem is that from somewhere in the osg there is a node visitor running over the osgPPU pipeline which hasn't specified type. I would expect here to run the update visitor first, but there is some other visitor running, and I do not know which one ;(, because the visitor has no specified type information (getVisitorType() returns just NODE_VISITOR). I have updated the trunk version of osgPPU. It seems the problems with the cycles are gone now. Test it out. If you would like to stay on version 0.3 of osgPPU, then you have to rewrite the ::traverse method of Processor.cpp from the trunk in your copy of osgPPU. Cheers, art ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgVolume development report?
Hi Aitor, The Joint Medical Visualization Project that funds my work on osgVolume has had me address the multi-media side of the 3D educational lectures that project is about, this has meant working on Pdf readers, Vnc clients and Browers in 3D, rather than push on with osgVolume. osgVolume isn't forgotten though, just a bit delayed by the shuffle of priorities. This week I'll be spending a day porting the volume rendering functionality from the osgvolume example into osgVolume, this will serve as the basis for osgVolume for the OSG-2.8 release. It won't be too sophisticated, just basic shader based ray tracing and iso surface rendering. A big chunk of osgVolume is the Dicom loader, and this is already available in svn/trunk, although used in conjunction with the osgvolume example right now rather than generate osgVolume nodes, although for 2.8 this should be available. Robert. On Wed, Dec 17, 2008 at 3:30 PM, Aitor Moreno aitormor...@gmail.com wrote: Hi all, I read in previous mails that the osgVolume is currently being implemented with a bunch of new functionality that would be great. In the SVN, I can see that there is some uploaded source code (http://www.openscenegraph.org/projects/osg/browser/OpenSceneGraph/trunk/src/osgVolume), but it seems to be the very first steps of the development (no commits in the last 3 months). Can any involved developer write a few lines about the state of the osgVolume? Thanks in advance. -- Aitor Moreno aitormoreno [...@] gmail.com ___ 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
Re: [osg-users] Bypassing the use of the CURL plugin for all http requests
Hi Jason, The only way I can think of right now to bpass the http/curl mechanism in the Registry would to have Registry::ReadFileCallback that catches the reads and then passes it to your custom code on demand. This behaviour is something that really should be configurable in a so that one can control how the protocol prefixes are handled. This would require mods to Registry. Perhaps via a callback that allows you to override the default behaviour. The other route would be to have the option of simply disabling the use of the curl plugin in case of http prefix, such as via ReaderWriter::Options. Right now I'm very busy with other work so can't go diving into code, you're welcome to come up with a Registry revision. The second of the above suggestions feels like the most reusable solution to me. Robert. On Wed, Dec 17, 2008 at 6:01 PM, Jason Beverage jasonbever...@gmail.com wrote: Hi Robert, Is there a way to bypass the use of the CURL plugin for all http requests? For most cases this is nice, but I'm working on a few plugins where it would be very beneficial to simply have the full URL passed in so that I can deal with it myself rather than having CURL try to download the file. The workaround I've been using currently is sticking a prefix on the address and having my plugin strip it off before it uses it. So, something like osgviewer server:http://www.server.com/myfile.extension. This works well so far, I'm just curious if there is a better way. Maybe an option in ReaderWriterOptions? Thanks! Jason ___ 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] Frame buffer object in cull traversal
Hi, Is is possible to use frame buffer objects of opengl without creation of a secondary camera in openscenegraph? In the single cull traversal phase of the scene graph, I would like to draw one specific geode to a FBO and all other geodes to the frame buffer.To do that, in the cull traversal of geode (e.g. in user defined cull callback of geode) with gl commands, would it be stable to change the render target to a FBO with gl command syntax? Thanks in advance ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Bypassing the use of the CURL plugin for all http requests
Thanks Robert, I'll take a look at it when I get a chance, the prefix route is working out well right now. Thanks! Jason On Wed, Dec 17, 2008 at 2:05 PM, Robert Osfield robert.osfi...@gmail.comwrote: Hi Jason, The only way I can think of right now to bpass the http/curl mechanism in the Registry would to have Registry::ReadFileCallback that catches the reads and then passes it to your custom code on demand. This behaviour is something that really should be configurable in a so that one can control how the protocol prefixes are handled. This would require mods to Registry. Perhaps via a callback that allows you to override the default behaviour. The other route would be to have the option of simply disabling the use of the curl plugin in case of http prefix, such as via ReaderWriter::Options. Right now I'm very busy with other work so can't go diving into code, you're welcome to come up with a Registry revision. The second of the above suggestions feels like the most reusable solution to me. Robert. On Wed, Dec 17, 2008 at 6:01 PM, Jason Beverage jasonbever...@gmail.com wrote: Hi Robert, Is there a way to bypass the use of the CURL plugin for all http requests? For most cases this is nice, but I'm working on a few plugins where it would be very beneficial to simply have the full URL passed in so that I can deal with it myself rather than having CURL try to download the file. The workaround I've been using currently is sticking a prefix on the address and having my plugin strip it off before it uses it. So, something like osgviewer server: http://www.server.com/myfile.extension. This works well so far, I'm just curious if there is a better way. Maybe an option in ReaderWriterOptions? Thanks! Jason ___ 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] osgVolume development report?
Hi all, I read in previous mails that the osgVolume is currently being implemented with a bunch of new functionality that would be great. In the SVN, I can see that there is some uploaded source code ( http://www.openscenegraph.org/projects/osg/browser/OpenSceneGraph/trunk/src/osgVolume), but it seems to be the very first steps of the development (no commits in the last 3 months). Can any involved developer write a few lines about the state of the osgVolume? Thanks in advance. -- Aitor Moreno aitormoreno [...@] gmail.com ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Removing the tgz, osgtgz and zip plugins
Gordon Tomlinson wrote: As to re-implementation , who's going to pay my company for the time and expenses to re-rite these go through all r Testing ands sign off procedures ? I don't understand this point. If you get the same functionality, just with a real library implementation and not a hacky system() call, why would you or your company care? Why change something that is not broken, might not be the best solution but they work To me, because it's not the best solution is a good enough reason to change them (to reimplement them in a non-hacky way). I agree that they should not be removed, though. --J ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg firefox plugin alpha
Hi Andreas,
your proposal is what I have already done:
The current shutdown is:
void nsPluginInstance::shut()
{
#ifdef _WIN32
subclass it back
SubclassWindow(mhWnd, lpOldProc);
#endif
mThread-cancel();
while (mThread-isRunning())
sleep(1);
delete mThread;
viewer = NULL; viewer is destructed here!
mhWnd = NULL;
mInitialized = FALSE;
}
regards Ralph
Andreas Goebel schrieb:
Hi Ralph,
thanks for porting the plugin. To me the stack-trace sounds as if
osgViewer tried to make the gl-context current that is no longer
existant. Maybe Firefox first deletes the window-context and then calls
the plugins destructor.
Maybe you could destroy the GL context actively here:
void nsPluginInstance::shut()
{
// We add DisableOpenGL
DisableOpenGL( mhWnd, hDC, hRC );
// subclass it back
SubclassWindow(mhWnd, lpOldProc);
mhWnd = NULL;
mInitialized = FALSE;
}
instead of waiting for the Viewers destructor to do it.
Regards,
Andreas
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] current View Point
Have a look on the Third person view example... You get your code from it,
and this example works well, so you should use it a base code, and modify it
to have what you are looking for.
You will be able to reply yourself I think.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
sorry for much truble but i looked and found two ways can not understand
how they work
the first way is to use call back :
so i added to the code :
class UpdateCallback : public osg::NodeCallback
{
virtual void operator()(osg::Node* node, osg::NodeVisitor* nv)
{
std::coutupdate callback - pre traversenodestd::endl;
traverse(node,nv);
std::coutupdate callback - post traversenodestd::endl;
}
};
and to the main loop :
mt-setUpdateCallback(new UpdateCallback());
but it made no changes.
the second way :
i added to the code :
// Given a Camera, create a wireframe representation of its
// view frustum. Create a default representation if camera==NULL.
osg::Node*
makeFrustumFromCamera( osg::Camera* camera )
{
osg::Node* loadedMode35 ;
loadedMode35 = osgDB::readNodeFile(red.3DS);
// Projection and ModelView matrices
osg::Matrixd proj;
osg::Matrixd mv;
if (camera)
{
proj = camera-getProjectionMatrix();
mv = camera-getViewMatrix();
}
else
{
// Create some kind of reasonable default Projection matrix.
proj.makePerspective( 30., 1., 1., 10. );
// leave mv as identity
}
// Get near and far from the Projection matrix.
const double near = proj(3,2) / (proj(2,2)-1.0);
const double far = proj(3,2) / (1.0+proj(2,2));
// Get the sides of the near plane.
const double nLeft = near * (proj(2,0)-1.0) / proj(0,0);
const double nRight = near * (1.0+proj(2,0)) / proj(0,0);
const double nTop = near * (1.0+proj(2,1)) / proj(1,1);
const double nBottom = near * (proj(2,1)-1.0) / proj(1,1);
// Get the sides of the far plane.
const double fLeft = far * (proj(2,0)-1.0) / proj(0,0);
const double fRight = far * (1.0+proj(2,0)) / proj(0,0);
const double fTop = far * (1.0+proj(2,1)) / proj(1,1);
const double fBottom = far * (proj(2,1)-1.0) / proj(1,1);
// Create parent MatrixTransform to transform the view volume by
// the inverse ModelView matrix.
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix( osg::Matrixd::inverse( mv ) );
mt-addChild( loadedMode35 );
return mt;
}
and i was suppose to change the loop to :
while (!viewer.done())
{
// Update the wireframe frustum
root-removeChild( 0, 1 );
root-insertChild( 0, makeFrustumFromCamera(
viewer.getView( 0 )-getCamera() ) );
viewer.frame();
}
but i was not sure what is this suppose to do ??
also i did not find viewer.getView function or equivelant to it
is there some way to help me fix or understand one of the two previous ways
or is there a way to directly update the node positon from the camira
position in the loop like using :
osg::NodeCallback* nc = new osg::AnimationPathCallback(
),osg::Vec3(0.0f,0.0f,1.0f),osg::inDegrees());
transform-setUpdateCallback(nc);
or transform(x,y,z)
rotate (h,p,r);
and get it directly from the current camera position matrix ?
please help
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] current View Point
To: olfat_ibra...@yahoo.com, OpenSceneGraph Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 3:09 AM
Each frame, you have to update your matrix. For the moment
you set it, next
you frame() the viewer, nothing more. Matrix will no update
itself.
To update it, do it in the viewer loop (after/before
frame()), or create a
new nodeCallback. Have a look on nodeCallback to see how to
use it.
Regards,
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
i have two viewers and put the scene in groups one for
the whole scene and
one for the pointer i want to move:
osg::Group* root = new osg::Group; // groupp have all
other objects
osg::Group* Mainroot = new osg::Group; // the
whole scene
Mainroot-addChild(root);
osgViewer::Viewer viewer;
osgViewer::Viewer viewer1;
//set left camera for left view
osg::ref_ptrosg::Camera cameraL = new
osg::Camera;
...
// add the matrix transformation
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix(
cameraL-getViewMatrix() );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
viewer.setCameraManipulator(
keyswitchManipulatorL.get() );
//set right camera for the right view
osg::ref_ptrosg::Camera cameraR = new
osg::Camera;
viewer1.setCameraManipulator(
Re: [osg-users] osg firefox plugin alpha
Hi Ralf, I don't have any fresh ideas for solving the path issue, save perhaps for installing the OSG in standard paths. The alternative to playing with paths is to try a static build of the OSG and your plugin. Robert. On Wed, Dec 17, 2008 at 9:35 AM, Ralph Kern usene...@rk-se.de wrote: Hi all, the LD_LIBRARY_PATH issue for the linux plugin is still annoying. I tried to use a proxy shared lib, which sets up an appriopriate path using setenv() and then calls dlopen() on the plugin lib. But it fails because LD_LIBRARY_PATH is only evaluated once during startup of the firefox program. It cannot be manipulated later. any more ideas? regards Ralph Ralph Kern schrieb: Hi Andreas, I'm making some progress in the linux port. I already can see the models on your page, but only if I start firefox with LD_LIBRARY_PATH set to the plugin directory. Now I need to search for some hack to port your library search path extensions to linux. This is not easy, because the plugin is not lazily loaded like in Windows. And there's no SetDllDirectory like in Windows. So I need to find a way to patch the search path before the FF plugin is loaded, otherwise the load fails because it cannot locate libosgdb.so and libosgViewer.so. I think the right path is to have a proxy plugin, which manipulates the path and then loads the correct plugin using dlopen(). If someone knows a better way, let me know! Regards Ralph Andreas Goebel schrieb: Hi, I have changed a little bit in the xpi and in the plugin itself. It now not only adds the plugin-directory to the osg-file-path, but to the windows-dll-search-path. This worked at least on my machine to enable .ive-support without having the osg-libs in your path. Please: Those people who didn´t see the lower three models, uninstall the old plugin, install the new one and try. Note that that function (named SetDllDirectory ) is only available from WinXP SP1 onwards, so this plugin won´t work with older windows versions. I think that only very few people are using older windows now. Hope this works ... Regards, Andreas ___ 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
Re: [osg-users] current View Point
sorry for much truble but i looked and found two ways can not understand how
they work
the first way is to use call back :
so i added to the code :
class UpdateCallback : public osg::NodeCallback
{
virtual void operator()(osg::Node* node, osg::NodeVisitor* nv)
{
std::coutupdate callback - pre traversenodestd::endl;
traverse(node,nv);
std::coutupdate callback - post traversenodestd::endl;
}
};
and to the main loop :
mt-setUpdateCallback(new UpdateCallback());
but it made no changes.
the second way :
i added to the code :
// Given a Camera, create a wireframe representation of its
// view frustum. Create a default representation if camera==NULL.
osg::Node*
makeFrustumFromCamera( osg::Camera* camera )
{
osg::Node* loadedMode35 ;
loadedMode35 = osgDB::readNodeFile(red.3DS);
// Projection and ModelView matrices
osg::Matrixd proj;
osg::Matrixd mv;
if (camera)
{
proj = camera-getProjectionMatrix();
mv = camera-getViewMatrix();
}
else
{
// Create some kind of reasonable default Projection matrix.
proj.makePerspective( 30., 1., 1., 10. );
// leave mv as identity
}
// Get near and far from the Projection matrix.
const double near = proj(3,2) / (proj(2,2)-1.0);
const double far = proj(3,2) / (1.0+proj(2,2));
// Get the sides of the near plane.
const double nLeft = near * (proj(2,0)-1.0) / proj(0,0);
const double nRight = near * (1.0+proj(2,0)) / proj(0,0);
const double nTop = near * (1.0+proj(2,1)) / proj(1,1);
const double nBottom = near * (proj(2,1)-1.0) / proj(1,1);
// Get the sides of the far plane.
const double fLeft = far * (proj(2,0)-1.0) / proj(0,0);
const double fRight = far * (1.0+proj(2,0)) / proj(0,0);
const double fTop = far * (1.0+proj(2,1)) / proj(1,1);
const double fBottom = far * (proj(2,1)-1.0) / proj(1,1);
// Create parent MatrixTransform to transform the view volume by
// the inverse ModelView matrix.
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix( osg::Matrixd::inverse( mv ) );
mt-addChild( loadedMode35 );
return mt;
}
and i was suppose to change the loop to :
while (!viewer.done())
{
// Update the wireframe frustum
root-removeChild( 0, 1 );
root-insertChild( 0, makeFrustumFromCamera(
viewer.getView( 0 )-getCamera() ) );
viewer.frame();
}
but i was not sure what is this suppose to do ??
also i did not find viewer.getView function or equivelant to it
is there some way to help me fix or understand one of the two previous ways or
is there a way to directly update the node positon from the camira position in
the loop like using :
osg::NodeCallback* nc = new osg::AnimationPathCallback(
),osg::Vec3(0.0f,0.0f,1.0f),osg::inDegrees());
transform-setUpdateCallback(nc);
or transform(x,y,z)
rotate (h,p,r);
and get it directly from the current camera position matrix ?
please help
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] current View Point
To: olfat_ibra...@yahoo.com, OpenSceneGraph Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 3:09 AM
Each frame, you have to update your matrix. For the moment
you set it, next
you frame() the viewer, nothing more. Matrix will no update
itself.
To update it, do it in the viewer loop (after/before
frame()), or create a
new nodeCallback. Have a look on nodeCallback to see how to
use it.
Regards,
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
i have two viewers and put the scene in groups one for
the whole scene and
one for the pointer i want to move:
osg::Group* root = new osg::Group; // groupp have all
other objects
osg::Group* Mainroot = new osg::Group; // the
whole scene
Mainroot-addChild(root);
osgViewer::Viewer viewer;
osgViewer::Viewer viewer1;
//set left camera for left view
osg::ref_ptrosg::Camera cameraL = new
osg::Camera;
...
// add the matrix transformation
osg::MatrixTransform* mt = new osg::MatrixTransform;
mt-setMatrix(
cameraL-getViewMatrix() );
mt-addChild( loadedMode35 );
Mainroot-addChild(mt);
viewer.setCameraManipulator(
keyswitchManipulatorL.get() );
//set right camera for the right view
osg::ref_ptrosg::Camera cameraR = new
osg::Camera;
viewer1.setCameraManipulator(
keyswitchManipulatorR.get() );
//load data for each viewer
viewer.setSceneData( root);
viewer1.setSceneData( Mainroot);
//run the viewer
while( !viewer.done() )
{
// fire off the cull and draw traversals of the
scene.
viewer.frame();
viewer1.frame();
}
now the object i used did not move what is the
Re: [osg-users] osgPPU v0.3 cycles problem
Hi Alexandre, The problem is that from somewhere in the osg there is a node visitor running over the osgPPU pipeline which hasn't specified type. I would expect here to run the update visitor first, but there is some other visitor running, and I do not know which one ;(, because the visitor has no specified type information (getVisitorType() returns just NODE_VISITOR). I have updated the trunk version of osgPPU. It seems the problems with the cycles are gone now. Test it out. If you would like to stay on version 0.3 of osgPPU, then you have to rewrite the ::traverse method of Processor.cpp from the trunk in your copy of osgPPU. Cheers, art --- alexandre amyot murray alex.amyotmur...@gmail.com schrieb am Di, 16.12.2008: Von: alexandre amyot murray alex.amyotmur...@gmail.com Betreff: [osg-users] osgPPU v0.3 cycles problem An: osg-users@lists.openscenegraph.org Datum: Dienstag, 16. Dezember 2008, 22:19 Hi (Art :) ) I just upgrade osgPPU to version 0.3. After I changed every osgPPU::Shader to osgPPU::ShaderAttribute, my application still crashes in an unexpected way. But after a couple of hours of investigation (I almost went back to version 0.2), I finally found what was the bug. I have a lot of UnitInOut in my application that add themself to create a specific effect. Exemple : osgPPU::UnitInOut * aUnit = new osgPPU::UnitInOut(); osgPPU::ShaderAttribute* aShaderAttribute = new osgPPU::ShaderAttribute(); aShaderAttribute-addShader(someShader); aShaderAttribute-add(tex0, osg::Uniform::SAMPLER_2D); aShaderAttribute-set(tex0, 0); aShaderAttribute-add(tex1, osg::Uniform::SAMPLER_2D); aShaderAttribute-set(tex1, 1); aUnit-getOrCreateStateSet()-setAttributeAndModes(aShaderAttribute); previousUnit-addChild( aUnit ); aUnit-addChild( aUnit ); // this line seems to not be valid anymore It seems like this is not a good practice anymore and I didn't find a way to replicate it. Hope you can help me Thanks Alex ___ 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
Re: [osg-users] Precipitation effect coming too fast, and hiding on some areas
Hi Robert, On Mon, Dec 15, 2008 at 7:55 AM, Robert Osfield robert.osfi...@gmail.comwrote: Hi Fred Jefferson, On Sun, Dec 14, 2008 at 10:49 AM, Frederic Bouvier fredlis...@free.fr wrote: I have a patch to PrecipitationEffect that seems to work. I'll submit it to osg-submissions I've just done a review of these changes and don't feel that there are mostr straightforward and effecient way to implement what is effectively near plan clipping. The code will make the vertex program slower, and not cut down on rasteristation and fill load. I feel a better route would be to either: 1) use two Camera's to depth partition the scene as suggested earlier in this thread, as this would give you control over the worlds depth range where the precipitation exists This one works! 2) Use an osg::ClipPlane to clip the precipitation effect. Using a clip plan would also allow you to control angle of the clipping. I don't know how to use that (I'm a beginner). Fred, I would be grateful if you could share your code, if you do it. Neither the above routes require any changes to the core OSG, and both would not affect the vertex program load, and would clip out fragment so would cut down the fragment program load so would be more efficient than the submitted fix. Could you test out the above suggestions? Thanks, Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org -- Jefferson Pinheiro (51) 9192 3535 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg firefox plugin alpha
Hi Robert, I tried now to use the standard installed OpenSceneGraph (Ubuntu 8.10 has osg 2.4.0-1.1). But I find the appropriate include files in the Ubuntu packages. I'll try otherway round to install the svn version to see if it works with standard paths. regards Ralph Robert Osfield schrieb: Hi Ralf, I don't have any fresh ideas for solving the path issue, save perhaps for installing the OSG in standard paths. The alternative to playing with paths is to try a static build of the OSG and your plugin. Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Unknown visitor running first and general visitor types
Sorry, I have send the message before completing it:
unsigned int osg::computeVisitorType(const std::string type_id)
{
return some_hash(type_id);
}
This would enable the possibility to check out which type of visitor is
currently traversing the node. Also in future releases or in external
applications with just different visitors there would be a possibility to
detect visitor types in a simple way.
All osg visitors might be named with osg_ prefix and hence make them more or
less unique (of course in the limitation of the hash function used).
What about that idea?
art
It seems that in the current version of osg, there is some
unknown node visitor running over all of my nodes. The node
type does return just NODE_VISITOR=0. Hence I do not see any
way to find out, which type of visitor is traversing there.
Currently there exists so many different node visitors in
the core osg, that it is somehow very strange to have the
VisitorType enum only describing couple of them, which are
even come from osgUtil and not from osg ;). I would propose
to have here some other kind of identification of a visitor.
What about being the visitor type not an enumeration but an
integer and create new visitor types by something like a
hash function over a string or so. Maybe a global method in
the core osg like:
unsigned int osg::computeVisitorType(const std::string
type_id)
{
}
___
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
Re: [osg-users] Unknown visitor running first and general visitor types
Hi
yes we could also add some kind of getName() which returns a string. However
using a hash based approach would have more performance, because only integer
comparison is needed (if you want to check for the visitor type during a
traversal call for example). It will also remove the VisitorType enum which
isn't really covers the whole bunch of visitor types.
cheers,
art
--- Robert Osfield robert.osfi...@gmail.com schrieb am Mi, 17.12.2008:
Von: Robert Osfield robert.osfi...@gmail.com
Betreff: Re: [osg-users] Unknown visitor running first and general visitor
types
An: stud_in...@yahoo.de, OpenSceneGraph Users
osg-users@lists.openscenegraph.org
Datum: Mittwoch, 17. Dezember 2008, 12:17
Hi Art,
All osg::Object's have a className() method.
NodeVisitor is just
subclass from osg::Referenced so doesn't have this, but
we could
certainly add it, it'd be less effort and more
informative than the
coming up with some hashmap scheme.
Robert.
On Wed, Dec 17, 2008 at 10:30 AM, Art Tevs
stud_in...@yahoo.de wrote:
Hi folks,
It seems that in the current version of osg, there is
some unknown node visitor running over all of my nodes. The
node type does return just NODE_VISITOR=0. Hence I do not
see any way to find out, which type of visitor is traversing
there.
Currently there exists so many different node visitors
in the core osg, that it is somehow very strange to have the
VisitorType enum only describing couple of them, which are
even come from osgUtil and not from osg ;). I would propose
to have here some other kind of identification of a visitor.
What about being the visitor type not an enumeration
but an integer and create new visitor types by something
like a hash function over a string or so. Maybe a global
method in the core osg like:
unsigned int osg::computeVisitorType(const
std::string type_id)
{
}
___
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
Re: [osg-users] osg firefox plugin alpha
Ralph Kern schrieb:
Hi Robert and others,
I now made an install of the OpenSceneGraph sources.
The FireFox plugin now works insofar it can show Andreas' website examples.
But I have two issues:
- there's some BadMatch XError Event generated (which I just ignore for now)
- when leaving the page, FireFox crashes with an segmentation violation,
and that's a big problem
Stack trace says:
#0 0xaff0ace0 in ?? ()
#1 0xb0734a60 in glXMakeCurrent () from /usr/lib/libGL.so.1
#2 0xb08d8324 in osgViewer::GraphicsWindowX11::makeCurrentImplementation ()
from /usr/local/lib/libosgViewer.so.48
#3 0xb0b9fd34 in osg::GraphicsContext::makeCurrent ()
from /usr/local/lib/libosg.so.48
#4 0xb0ba1147 in osg::GraphicsContext::close ()
from /usr/local/lib/libosg.so.48
#5 0xb08c1f59 in osgViewer::Viewer::~Viewer ()
from /usr/local/lib/libosgViewer.so.48
#6 0xb1e775d0 in osg::Referenced::unref (this=0xa400134)
at ../OpenSceneGraph/include/osg/Referenced:184
#7 0xb1e7763f in osg::ref_ptrosgViewer::Viewer::operator=
(this=0x9afe590,
ptr=0x0) at ../OpenSceneGraph/include/osg/ref_ptr:54
#8 0xb1e75e46 in nsPluginInstance::shut (this=0x9afe580) at plugin.cpp:320
#9 0xb1e78ff3 in NPP_Destroy (instance=0xa1acd48, save=0xbffef4c4)
at npp_gate.cpp:86
#10 0xb78c054d in ?? () from /usr/lib/xulrunner-1.9.0.4/libxul.so
#11 0xb747e092 in ?? () from /usr/lib/xulrunner-1.9.0.4/libxul.so
#12 0xb747e174 in ?? () from /usr/lib/xulrunner-1.9.0.4/libxul.so
[...]
I already switched to SingleThreaded, but with no avail.
Hi Ralph,
thanks for porting the plugin. To me the stack-trace sounds as if
osgViewer tried to make the gl-context current that is no longer
existant. Maybe Firefox first deletes the window-context and then calls
the plugins destructor.
Maybe you could destroy the GL context actively here:
void nsPluginInstance::shut()
{
// We add DisableOpenGL
DisableOpenGL( mhWnd, hDC, hRC );
// subclass it back
SubclassWindow(mhWnd, lpOldProc);
mhWnd = NULL;
mInitialized = FALSE;
}
instead of waiting for the Viewers destructor to do it.
Regards,
Andreas
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Unknown visitor running first and general visitor types
Hi Art,
All osg::Object's have a className() method. NodeVisitor is just
subclass from osg::Referenced so doesn't have this, but we could
certainly add it, it'd be less effort and more informative than the
coming up with some hashmap scheme.
Robert.
On Wed, Dec 17, 2008 at 10:30 AM, Art Tevs stud_in...@yahoo.de wrote:
Hi folks,
It seems that in the current version of osg, there is some unknown node
visitor running over all of my nodes. The node type does return just
NODE_VISITOR=0. Hence I do not see any way to find out, which type of visitor
is traversing there.
Currently there exists so many different node visitors in the core osg, that
it is somehow very strange to have the VisitorType enum only describing
couple of them, which are even come from osgUtil and not from osg ;). I would
propose to have here some other kind of identification of a visitor.
What about being the visitor type not an enumeration but an integer and
create new visitor types by something like a hash function over a string or
so. Maybe a global method in the core osg like:
unsigned int osg::computeVisitorType(const std::string type_id)
{
}
___
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] Unknown visitor running first and general visitor types
Hi folks,
It seems that in the current version of osg, there is some unknown node visitor
running over all of my nodes. The node type does return just NODE_VISITOR=0.
Hence I do not see any way to find out, which type of visitor is traversing
there.
Currently there exists so many different node visitors in the core osg, that it
is somehow very strange to have the VisitorType enum only describing couple of
them, which are even come from osgUtil and not from osg ;). I would propose to
have here some other kind of identification of a visitor.
What about being the visitor type not an enumeration but an integer and create
new visitor types by something like a hash function over a string or so. Maybe
a global method in the core osg like:
unsigned int osg::computeVisitorType(const std::string type_id)
{
}
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] my current position
--- On Wed, 12/17/08, olfat ibrahim olfat_ibra...@yahoo.com wrote: From: olfat ibrahim olfat_ibra...@yahoo.com Subject: my current position To: OSG osg-users@lists.openscenegraph.org Date: Wednesday, December 17, 2008, 9:01 AM i am using keyswitchManipulator : i need to print my current location while iam driving in the scene : i tried the functions : keyswitchManipulatorL-getMatrix() viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up) cameraL-getViewMatrixAsLookAt(eye, center, up) all of those does not get me my moving statuse . i need my x,y,z and h,p,r current position ; is there a function should i call to get me what i need ? thanks ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] my current position
Hi, 2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com --- On Wed, 12/17/08, olfat ibrahim olfat_ibra...@yahoo.com wrote: From: olfat ibrahim olfat_ibra...@yahoo.com Subject: my current position To: OSG osg-users@lists.openscenegraph.org Date: Wednesday, December 17, 2008, 9:01 AM i am using keyswitchManipulator : i need to print my current location while iam driving in the scene : i tried the functions : keyswitchManipulatorL-getMatrix() viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up) eye is camera position. cameraL-getViewMatrixAsLookAt(eye, center, up) all of those does not get me my moving statuse . i need my x,y,z and h,p,r current position ; Are you sure you get the position each frame ? because theses functions will give you vector copy, so if you don't get them each frame, they will not change. What do you meen by h,r,p ? second camera position ? Vincent. is there a function should i call to get me what i need ? thanks ___ 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
Re: [osg-users] Build error in osgdb_osgAnimation (Windows, SVN)
Thanks JS, fix merged and submitted to SVN.
On Wed, Dec 17, 2008 at 3:04 AM, Jean-Sébastien Guay
jean-sebastien.g...@cm-labs.com wrote:
Hi again,
Perhaps a missing EXPORT or something like that?
Yep. Fix attached.
J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
http://www.cm-labs.com/
http://whitestar02.webhop.org/
/* -*-c++-*-
* Copyright (C) 2008 Cedric Pinson morni...@plopbyte.net
*
* This library is open source and may be redistributed and/or modified
under
* the terms of the OpenSceneGraph Public License (OSGPL) version 0.0 or
* (at your option) any later version. The full license is in LICENSE file
* included with this distribution, and on the openscenegraph.org website.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* OpenSceneGraph Public License for more details.
*/
#ifndef OSGANIMATION_SKELETON_H
#define OSGANIMATION_SKELETON_H
#include osg/MatrixTransform
#include osgAnimation/Bone
#include osgAnimation/Export
namespace osgAnimation
{
class OSGANIMATION_EXPORT Skeleton : public Bone
{
public:
META_Node(osgAnimation, Skeleton);
struct OSGANIMATION_EXPORT UpdateSkeleton : public osg::NodeCallback
{
META_Object(osgAnimation, UpdateSkeleton);
UpdateSkeleton() {}
UpdateSkeleton(const UpdateSkeleton us, const osg::CopyOp
copyop= osg::CopyOp::SHALLOW_COPY) : osg::NodeCallback(us, copyop) {}
virtual void operator()(osg::Node* node, osg::NodeVisitor* nv);
};
Skeleton(const Skeleton b, const osg::CopyOp copyop=
osg::CopyOp::SHALLOW_COPY) : Bone(b,copyop) {}
Skeleton();
void setDefaultUpdateCallback(void);
void computeBindMatrix() { _invBindInSkeletonSpace =
osg::Matrix::inverse(_bindInBoneSpace); }
};
}
#endif
___
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
Re: [osg-users] out-of-source builds don't work reliably?
Hi J-S, We've never tried multiple platforms with the same in-source tree, and I wouldn't expect that it should work. However, I would expect that we'd be able to build debug and release versions for a single platform in the same in-source tree. Has this been working reliably for you? We've been trying to set up all builds as out of source with exceptions for Windows (because of our network more than any other reason), and now Mac OS X. These platforms each have their own source trees. Thanks again for your help and experience with this. On your advice I'll stick with out-for-source for 32 and 64 bit Linux. -Don Jean-Sébastien Guay wrote: Hi Don, Because of way our network is set up we build Windows on a seperate drive, so it's always been built in-source. We have several flavors of Linux, Mac, and maybe even Solaris (if we get around to it) that I've been trying to build off the same source directory. Well, building multiple platforms in-source in the same source tree (or in the same build/ directory, or installing to the same directory) isn't a good idea in general. Apart from the fact that the Config file (for both OpenThreads and OSG) will be (and must be) different on different platforms, it can introduce subtle problems, especially for similar platform like unices (which will typically use the same filenames for shared objects, for example). I think you may be making things difficult for yourself :-) My advice: build out-of-source, and use the install target. I find it's easier to have control over what you get that way. J-S ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Removing the tgz, osgtgz and zip plugins
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Gordon Tomlinson wrote: I use them and would not want to lose them, IVE, OSGA is not a suitable replacement for me I don't see why they need to removed they work( maybe not to everyone's liking but they work ), they're useful to some As to re-implementation , who's going to pay my company for the time and expenses to re-rite these go through all r Testing ands sign off procedures ? Why change something that is not broken, might not be the best solution but they work Completely seconded. Maybe they should be reimplemented using a better method, but to throw out support for zip files would be a mistake - it is infinitely easier to move models around in a zip file than to explain to a designer how to make an osga file or ive file, requiring extra tools and so on. It is also easier to archive such models - everything is in one place and I can always open them. IVE or osga formats can change over time and I cannot load them back into a modeling tool. I think these should be kept and perhaps fixed. Regards, Jan -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mandriva - http://enigmail.mozdev.org iD8DBQFJSOYEn11XseNj94gRAlLOAKCH0CsenXYQXdabzb4Txlxc15p/qgCgrNTQ HrcqqLag/v1RDl7nFpxIy0M= =5Kdv -END PGP SIGNATURE- ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Help Rendering and Reading Floating Point Textures
Hi David, David Spilling wrote: Brian, JP, The osgmultiplerendertargets example uses (by default) GL_FLOAT_RGBA32_NV, rather than either GL_RGBA32F_ARB, or even GL_RGBA16F_ARB. Could this be a card issue? Yes, if I remember correctly, at the time I made the example I was fiddling with various settings. On the card I tested, the GL_FLOAT_RGBA32_NV was faster than GL_RGBA32F_ARB. There is no real reason to not use the ARB formats. I'll submit a modified example with the NV options commented out and documented. I would be interested to know whether this example works for you if you change to GL_RGBA32F_ARB or GL_RGBA16F_ARB. I admit, I haven't tried it myself yet - and now intend to - but certainly in my own app (on ATI), I have had issues with GL_RGBA16F_ARB and GL_FLOAT in the past. Things might be better now. I know GL_RGBA32F_ARB and GL_FLOAT work fine, we use it all the time on NVidia 8-series and up. I've just tested GL_RGBA16F_ARB with GL_FLOAT and it works fine after I replaced all e-12 and e12 with e-3 and e3. This is on GeForce 7400Go. jp David ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org -- This message is subject to the CSIR's copyright terms and conditions, e-mail legal notice, and implemented Open Document Format (ODF) standard. The full disclaimer details can be found at http://www.csir.co.za/disclaimer.html. This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. MailScanner thanks Transtec Computers for their support. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osg firefox plugin alpha
Hi Andreas,
just browsed along through the sources, it seems the right time to
handle the viewer destruction is in nsPluginInstance::SetWindow() when
called with NULL:
see
http://devedge-temp.mozilla.org/library/manuals/2002/plugin/1.0/npp_api12.html
and
npp_gate.cpp:
// during this call we know when the plugin window is ready or
// is about to be destroyed so we can do some gui specific
// initialization and shutdown
NPError NPP_SetWindow (NPP instance, NPWindow* pNPWindow)
{
if(instance == NULL)
return NPERR_INVALID_INSTANCE_ERROR;
NPError rv = NPERR_NO_ERROR;
if(pNPWindow == NULL)
return NPERR_GENERIC_ERROR;
nsPluginInstanceBase * plugin = (nsPluginInstanceBase *)instance-pdata;
if(plugin == NULL)
return NPERR_GENERIC_ERROR;
// window just created
if(!plugin-isInitialized() (pNPWindow-window != NULL)) {
if(!plugin-init(pNPWindow)) {
NS_DestroyPluginInstance(plugin);
return NPERR_MODULE_LOAD_FAILED_ERROR;
}
}
// window goes away
if((pNPWindow-window == NULL) plugin-isInitialized())
return plugin-SetWindow(pNPWindow);
// window resized?
if(plugin-isInitialized() (pNPWindow-window != NULL))
return plugin-SetWindow(pNPWindow);
// this should not happen, nothing to do
if((pNPWindow-window == NULL) !plugin-isInitialized())
return plugin-SetWindow(pNPWindow);
return rv;
}
regards Ralph
Ralph Kern schrieb:
Hi Andreas,
your proposal is what I have already done:
The current shutdown is:
void nsPluginInstance::shut()
{
#ifdef _WIN32
subclass it back
SubclassWindow(mhWnd, lpOldProc);
#endif
mThread-cancel();
while (mThread-isRunning())
sleep(1);
delete mThread;
viewer = NULL; viewer is destructed here!
mhWnd = NULL;
mInitialized = FALSE;
}
regards Ralph
Andreas Goebel schrieb:
Hi Ralph,
thanks for porting the plugin. To me the stack-trace sounds as if
osgViewer tried to make the gl-context current that is no longer
existant. Maybe Firefox first deletes the window-context and then calls
the plugins destructor.
Maybe you could destroy the GL context actively here:
void nsPluginInstance::shut()
{
// We add DisableOpenGL
DisableOpenGL( mhWnd, hDC, hRC );
// subclass it back
SubclassWindow(mhWnd, lpOldProc);
mhWnd = NULL;
mInitialized = FALSE;
}
instead of waiting for the Viewers destructor to do it.
Regards,
Andreas
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Unknown visitor running first and general visitor types
Hi Art, On Wed, Dec 17, 2008 at 11:33 AM, Art Tevs stud_in...@yahoo.de wrote: yes we could also add some kind of getName() which returns a string. However using a hash based approach would have more performance, because only integer comparison is needed (if you want to check for the visitor type during a traversal call for example). It will also remove the VisitorType enum which isn't really covers the whole bunch of visitor types. Um... I'm absolutely not about to go breaking the API so VisitorType stays. The VisitorType is very useful for detecting specific types of visitors such as event visitors, cull visitors etc as these have a special role in the OSG as they are a fundamental part of each frame. The rest of the NodeVisitor provided specialist functions. className() is not getName(), one uses getName() for names of objects, className() in the OSG is used exclusively for reporting the name of the class. This is both useful for serialisation and debugging purposes. I'm afraid I don't get the value of hash map in this context, the performance aspect is trivial as this type of operation is done so rarely per frame. I have begun adding className and libraryName and a META_NodeVisitor macro definition of these to all the core OSG node visitor, you are welcome use this or ignore them. I'm certainly not about to drop existing API's to appease you preferences. Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] please test... SSAO (v1.0)
Hi, Some ref_ptr error in code... Sorry, but on my old PC shader/OpenGl problem... I will test on newer PC if I have time. Regards, Vincent. 2008/12/17 Adrian Egli OpenSceneGraph (3D) 3dh...@gmail.com Hi all , please test this first version of SSAO Effect (Node) There is still a problem (512 x 512) texture, gaussian filter purly implemented on GPU finally i have 10-15 FPS /adrian -- Adrian Egli ___ 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] Open Inventor like Sensors in OSG?
Hi, In Open Inventor, sensors are special classes that respond when an object state is changed and call a callback. They were IMHO not very clean, designed before Design Patterns, but I think they are like observers or signal/slot objects. Is there anything similar in Open Scenegraph? Regards, -- A. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] my current position
thanks again for your help :
but i have another problem :
while( !viewer.done() || !viewer1.done() )
{
viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up);
cout eye2 eye.x() eye.y() eye.z()
endl;
cout center center.x() center.y()
center.z() endl;
cout up up.x() up.y() up.z()
endl;
osg::Matrixd m(viewer.getCamera()-getViewMatrix());
mt-setMatrix(osg::Matrix::translate(eye.x(),eye.y(),eye.z())*
osg::Matrix::rotate(m.getRotate().inverse()));
i tried m.getRotate().inverse() and m.getRotate()
they rotate the object but i the wrong way they flip it
does any one have recomendations ??
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] my current position
To: olfat_ibra...@yahoo.com
Date: Wednesday, December 17, 2008, 9:58 AM
You can do it using different things:
* You get eye, center, up . With theses vector, you can
compute the front,
up, side vectors of the camera, an so compute the rotation.
(eye is camera position, center is the point the camera
look at)
* You can get the matrix from the manipulator, and
getRotation from it. You
will have the quaternion representing the rotation. (or
rotation inverse, I
don't remember)
* you can get the matrix from manipulator, and use
getFrontVector(...),
getSidevector() and getUpVector() from it. After that, just
compute the
rotation.
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
you were right thank you very much
now i need to get the H P R
by H P R i mean the rotation arround x ,y , z axis .
i tried the three functions in the following mannare :
osg::Vec3f eye;
osg::Vec3f center;
osg::Vec3f up;
osg::Vec3f eye1;
osg::Vec3f center1;
osg::Vec3f up1;
osg::Vec3f position;
osg::Matrixd
m(keyswitchManipulatorL-getCoordinateFrame(position));
while( !viewer.done() || !viewer1.done() )
{
cameraL-getViewMatrixAsLookAt(eye1,
center1, up1);
viewer.getCamera()-getViewMatrixAsLookAt(eye, center,
up);
osg::Matrixd
m(keyswitchManipulatorL-getCoordinateFrame(position));
// cout position.x
position.y position.z endl;
cout eye
position.x() position.y()
position.z() endl;
cout eye
eye.x() eye.y()
eye.z() endl;
cout center
center.x()
center.y()
center.z() endl;
cout up
up.x() up.y()
up.z()
endl;
cout eye
eye1.x() eye1.y()
eye1.z() endl;
cout center
center1.x()
center1.y()
center1.z() endl;
cout up
up1.x() up1.y()
up1.z()
endl;
// fire off the cull and draw traversals of the
scene.
viewer.frame();
}
can any ong give me ideas ??
--- On Wed, 12/17/08, Vincent Bourdier
vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier
vincent.bourd...@gmail.com
Subject: Re: [osg-users] my current position
To: olfat_ibra...@yahoo.com, OpenSceneGraph
Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 9:37 AM
Hi,
2008/12/17 olfat ibrahim
olfat_ibra...@yahoo.com
--- On Wed, 12/17/08, olfat ibrahim
olfat_ibra...@yahoo.com wrote:
From: olfat ibrahim
olfat_ibra...@yahoo.com
Subject: my current position
To: OSG
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008,
9:01 AM
i am using keyswitchManipulator :
i need to print my current location
while iam
driving in
the scene :
i tried the functions :
keyswitchManipulatorL-getMatrix()
viewer.getCamera()-getViewMatrixAsLookAt(eye,
center,
up)
eye is camera position.
cameraL-getViewMatrixAsLookAt(eye,
center,
up)
all of those does not get me my moving
statuse .
i need my x,y,z and h,p,r current
position ;
Are you sure you get the position each frame ?
because
theses functions will
give you
vector copy, so if you don't get them each
frame, they
will not change.
What do you meen by h,r,p ? second camera
position ?
Vincent.
is there a function should i call to
get me what
i need ?
thanks
___
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
Re: [osg-users] principle of Light and LightSource
So, when I want to use more lights I see only the first light. Thats the graph with two lights: To render with multiple lights, arrange the LightSource group nodes vertically: root | V lightSource 0 | V lightSource 1 | V geodes ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgLua?
Hi Jose-Luis, be assured I seen these messages before and also did the whole patches/linker flags etc. pp. that were suggested there. Still, the RTTI is stripped - so *nix is not going to work. Admittetly, I updated the osgLua sources in an intree version and added my own copy of Lua 5.1.4 into the buildsys (with the patch). Again, I checked that in the whole chain mentioned libs are being compiled with these flags. Its a pitty that this is been ignored by the Lua developers - I can't see why it is so outragous difficult to add this options. I will try to investigate in more depth - after all this is a very specific use-case. Are you still working on osgLua? I can flick you a few patches that might fix a couple of problems. Cheers, Hartmut Jose Luis Hidalgo wrote: Hi Hartmut, The problem is how lua loads libraries, is an issue that comes from time to time, and never is important enough to be solved: http://lua-users.org/lists/lua-l/2007-10/msg00098.html the answer from Luiz was: http://lua-users.org/lists/lua-l/2007-10/msg00120.html Basically that means you can not use osgLua properly from a vanilla lua interpreter+libraries. I tried myself to ask for it on the list: http://lua-users.org/lists/lua-l/2006-11/msg00174.html But... nothing :( Hope it helps. Jose-Luis. On Mon, Dec 15, 2008 at 9:34 PM, Hartmut Seichter li...@technotecture.com wrote: No, not from Lua (or I guess a lua interpreter) itself - I have a wxWidgets application which loads various plugins and one of them is a based around osgLua - thus, lua_open etc pp should happen in that context. On Windows and Mac this whole mechanism works fine - on Linux the RTTI seems to be the showstopper. I added the Linkerflags mentioned somewhere to the whole chain of dependencies: App Plugin Lua Lib Any other ideas? H Jose Luis Hidalgo wrote: Hi Hartmut, How do you try to load osgLua? from lua itself? Cheers, Jose-Lus. On Sun, Dec 14, 2008 at 10:25 PM, Hartmut Seichter li...@technotecture.com wrote: Hi there, I am trying to combine the power of osgLua and osgSWIG in a new project (no I am not mad :)) ... However, I couldn't get osgLua play nicely on Linux - I did all the patching mentioned but no avail. I am on 64bit Ubuntu ... any idea where things go haywire ... seems that RTTI get still stripped - print(osg.Node) returns some mangled reference to osgObject. Cheers, Hartmut -- Hartmut Seichter, PhD (HKU), Dipl-Ing.(BUW), Postdoctoral Fellow, HITLabNZ ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org -- Hartmut Seichter, PhD (HKU), Dipl-Ing.(BUW), Postdoctoral Fellow, HITLabNZ ___ 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] Light Position Behavior
I understand that positioning the camera SHOULD fix my problem, but it isn't. I have the light positioned at (0,0,100,1) and even after i set my camera with setViewMatrixAsLookAt it is still trying to view the entire bounding box of the scene. This is the fix i was looking for when i started the post =) Looking forward to some guidance! Thanks! ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] synchronize two view windows
hello :
i need to synchronize two view windows .
i created two views and one key switch Manipulator
i want as i walk in one window my view synchronized in the other window with
solaly differ view.
this is my try but it is drive in the first view but no moves happened in the
second view :
while( !viewer.done() || !viewer1.done() )
{
viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up);
viewer1.getCamera()-setViewMatrixAsLookAt(eye, center, up);
viewer.frame();
viewer1.frame();
}
what iam doing wrong ?
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Light Position Behavior
it seems that by using setUpViewerAsEmbeddedInwindow() it does not reposition based on the world's bounding sphere. interesting for those who didn't already know that. Anyway hope this helps someone else! -Rusty On Wed, Dec 17, 2008 at 5:01 PM, Ryan Morris russell.co...@gmail.comwrote: I understand that positioning the camera SHOULD fix my problem, but it isn't. I have the light positioned at (0,0,100,1) and even after i set my camera with setViewMatrixAsLookAt it is still trying to view the entire bounding box of the scene. This is the fix i was looking for when i started the post =) Looking forward to some guidance! Thanks! ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] Building the Collada plugin
OSG 2.6.1, VS 2008. So far I've downloaded COLLADA_1.4.1_DOM_1.3.0.zip, built it, and got a slew of errors trying to build the plugin. (Collada DOM 2.2, which looks like it might be downward compatible, uses a different library structure, and I've no idea which files are still needed.) Which version of Collada is known to work? Does the plugin work at all with VS 2008? Randolph ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] synchronize two view windows
Did you remove the camera manipulator for the second view?
It is only necessary to make war with five things: with the maladies
of the body, with the ignorances of the mind, with the passions of the
body, with the seditions of the city, with the discords of families.
- Tacitus
The desire for safety stands against every great and noble
enterprise. - Tacitus
Those who would give up essential liberty to purchase a little
temporary safety deserve neither liberty nor safety. - Benjamin
Franklin
Numquam ponenda est pluralitas sine necessitate. - William of Ockham
On Wed, Dec 17, 2008 at 8:25 PM, olfat ibrahim olfat_ibra...@yahoo.com wrote:
hello :
i need to synchronize two view windows .
i created two views and one key switch Manipulator
i want as i walk in one window my view synchronized in the other window with
solaly differ view.
this is my try but it is drive in the first view but no moves happened in the
second view :
while( !viewer.done() || !viewer1.done() )
{
viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up);
viewer1.getCamera()-setViewMatrixAsLookAt(eye, center, up);
viewer.frame();
viewer1.frame();
}
what iam doing wrong ?
___
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
Re: [osg-users] my current position
Hi,
Try manipulator-getMatrix() instead of camera view matrix...
Next, it is possible to have a rotation of the matrix along X of +/- PI/2...
don't know exactly when or why, but this can help...
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
thanks again for your help :
but i have another problem :
while( !viewer.done() || !viewer1.done() )
{
viewer.getCamera()-getViewMatrixAsLookAt(eye, center, up);
cout eye2 eye.x() eye.y()
eye.z() endl;
cout center center.x() center.y()
center.z() endl;
cout up up.x() up.y() up.z()
endl;
osg::Matrixd m(viewer.getCamera()-getViewMatrix());
mt-setMatrix(osg::Matrix::translate(eye.x(),eye.y(),eye.z())*
osg::Matrix::rotate(m.getRotate().inverse()));
i tried m.getRotate().inverse() and m.getRotate()
they rotate the object but i the wrong way they flip it
does any one have recomendations ??
--- On Wed, 12/17/08, Vincent Bourdier vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier vincent.bourd...@gmail.com
Subject: Re: [osg-users] my current position
To: olfat_ibra...@yahoo.com
Date: Wednesday, December 17, 2008, 9:58 AM
You can do it using different things:
* You get eye, center, up . With theses vector, you can
compute the front,
up, side vectors of the camera, an so compute the rotation.
(eye is camera position, center is the point the camera
look at)
* You can get the matrix from the manipulator, and
getRotation from it. You
will have the quaternion representing the rotation. (or
rotation inverse, I
don't remember)
* you can get the matrix from manipulator, and use
getFrontVector(...),
getSidevector() and getUpVector() from it. After that, just
compute the
rotation.
Vincent.
2008/12/17 olfat ibrahim olfat_ibra...@yahoo.com
you were right thank you very much
now i need to get the H P R
by H P R i mean the rotation arround x ,y , z axis .
i tried the three functions in the following mannare :
osg::Vec3f eye;
osg::Vec3f center;
osg::Vec3f up;
osg::Vec3f eye1;
osg::Vec3f center1;
osg::Vec3f up1;
osg::Vec3f position;
osg::Matrixd
m(keyswitchManipulatorL-getCoordinateFrame(position));
while( !viewer.done() || !viewer1.done() )
{
cameraL-getViewMatrixAsLookAt(eye1,
center1, up1);
viewer.getCamera()-getViewMatrixAsLookAt(eye, center,
up);
osg::Matrixd
m(keyswitchManipulatorL-getCoordinateFrame(position));
// cout position.x
position.y position.z endl;
cout eye
position.x() position.y()
position.z() endl;
cout eye
eye.x() eye.y()
eye.z() endl;
cout center
center.x()
center.y()
center.z() endl;
cout up
up.x() up.y()
up.z()
endl;
cout eye
eye1.x() eye1.y()
eye1.z() endl;
cout center
center1.x()
center1.y()
center1.z() endl;
cout up
up1.x() up1.y()
up1.z()
endl;
// fire off the cull and draw traversals of the
scene.
viewer.frame();
}
can any ong give me ideas ??
--- On Wed, 12/17/08, Vincent Bourdier
vincent.bourd...@gmail.com wrote:
From: Vincent Bourdier
vincent.bourd...@gmail.com
Subject: Re: [osg-users] my current position
To: olfat_ibra...@yahoo.com, OpenSceneGraph
Users
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008, 9:37 AM
Hi,
2008/12/17 olfat ibrahim
olfat_ibra...@yahoo.com
--- On Wed, 12/17/08, olfat ibrahim
olfat_ibra...@yahoo.com wrote:
From: olfat ibrahim
olfat_ibra...@yahoo.com
Subject: my current position
To: OSG
osg-users@lists.openscenegraph.org
Date: Wednesday, December 17, 2008,
9:01 AM
i am using keyswitchManipulator :
i need to print my current location
while iam
driving in
the scene :
i tried the functions :
keyswitchManipulatorL-getMatrix()
viewer.getCamera()-getViewMatrixAsLookAt(eye,
center,
up)
eye is camera position.
cameraL-getViewMatrixAsLookAt(eye,
center,
up)
all of those does not get me my moving
statuse .
i need my x,y,z and h,p,r current
position ;
Are you sure you get the position each frame ?
because
theses functions will
give you
vector copy, so if you don't get them each
frame, they
will not change.
What do you meen by h,r,p ? second camera
position ?
Vincent.
is there a function
Re: [osg-users] please test... SSAO (v1.0)
Works nicely, if a bit slow (~15fps) and the window can't be resized correctly. The code was missing three ref_ptr .get()s, so I guess you didn't compile the version you attached... http://paahdin.com/shared/images/ssao.png Lähettäjä: osg-users-boun...@lists.openscenegraph.org [mailto:osg-users-boun...@lists.openscenegraph.org] Puolesta Adrian Egli OpenSceneGraph (3D) Lähetetty: 17. joulukuuta 2008 18:24 Vastaanottaja: OpenSceneGraph Users Aihe: [osg-users] please test... SSAO (v1.0) Hi all , please test this first version of SSAO Effect (Node) There is still a problem (512 x 512) texture, gaussian filter purly implemented on GPU finally i have 10-15 FPS /adrian ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] glXBindTexImage
Hi all Has anyone attempted to use glXBindTexImage (and friends) to bind an X11 pixmap to a texture within OpenSceneGraph? You could then display the contents of an X window on a surface. I don't think it would be difficult to do, just wondering if it's already been done - best to avoid duplication of effort! Thanks Grahame ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] please test... SSAO (v1.0)
where are the bugs, my version compiled under VS 2005 (c++) latest SVN version /adrian 2008/12/18 Kimmo Kotajärvi kimmo.kotaja...@vianova.fi Works nicely, if a bit slow (~15fps) and the window can't be resized correctly. The code was missing three ref_ptr .get()s, so I guess you didn't compile the version you attached… http://paahdin.com/shared/images/ssao.png -- *Lähettäjä:* osg-users-boun...@lists.openscenegraph.org [mailto: osg-users-boun...@lists.openscenegraph.org] *Puolesta *Adrian Egli OpenSceneGraph (3D) *Lähetetty:* 17. joulukuuta 2008 18:24 *Vastaanottaja:* OpenSceneGraph Users *Aihe:* [osg-users] please test... SSAO (v1.0) Hi all , please test this first version of SSAO Effect (Node) There is still a problem (512 x 512) texture, gaussian filter purly implemented on GPU finally i have 10-15 FPS /adrian ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org -- Adrian Egli ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
