Re: [osg-users] Skybox example problem

[Großer Martin]

Hello,

ok, it seems that nobody has an idea. Maybe is the problem the range of
values?

Cheers, Martin

Am Montag, den 06.04.2009, 10:30 +0200 schrieb "Martin Großer":
> Hello all,
> 
> I compiled the skybox example. And I try to load a big scene. Now I have 
> artefacts in the skybox. I can see the environment sphere. I try a smaller 
> scene and zoom out. I get the same result. When I change the following line 
> from ...
> 
> clearNode->setRequiresClear(false);
> 
> ... in ...
> 
> clearNode->setRequiresClear(true);
> 
> ... then the environment sphere go missing and I see the clear color.
> 
> Is the problem the distance of the far clipping plane? And I think the 
> environment sphere is translate with the eye position?
> 
> 
> Cheers, Martin

___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] OpenFlight-Plugin - How to store individual object in seperate geode?

[Bulkhead]

Hi Brede,

Thanks very much for the info. Yup, with osgconv, I managed to compare the 
source flt database and the coverted osg format. Seems like the objects are 
already merged in the flt database. So I guess I have to find a way to modify 
the source before loading it to osg  [Wink] 

Once again, thanks for your help.

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9818#9818





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] VPB texture disappear

[Paul Martz]

Thanks. I'll give that a try.
 
If NPOT is the issue, then apparently VPB requires NPOT support at DB build
time, not just at run time. I found that creating the same DB on my Windows
laptop with Quadro 1500M works great.
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 

  _  

From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Jason
Beverage
Sent: Monday, April 06, 2009 7:41 PM
To: OpenSceneGraph Users
Subject: Re: [osg-users] VPB texture disappear


Hey Paul,

I've seen this issue before with non power of two textures not being
supported in some situations.  VPB tries to use npot textures at the lowest
level of detail to most closely match the resolution of the source imagery.
Try generating your DB using --pot and see what happens, I bet it works.

Thanks,

Jason


On Mon, Apr 6, 2009 at 9:37 PM, Paul Martz  wrote:


Same issue viewing the DB on a Windows box, so apparently not a driver
issue.
 
I'll try building the DB on Windows. This might take a while; it's not a hot
box like my MacPro...
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 

  _  

From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Paul Martz
Sent: Monday, April 06, 2009 7:30 PM
To: osg-users@lists.openscenegraph.org
Subject: [osg-users] VPB texture disappear


Hi Robert -- I'm seeing an issue where, as the viewpoint approaches the
terrain, textures surrounding a high-res inset disappear. This is with very
current VPB and OSG running on Mac OS X with a GeForce 8800 card.
 
ss0.jpg shows the scene just before the issue occurs. As I move the
viewpoint forward (closer to the terrain), I see the image in ss1.jpg. I
also get several instances of the following error in the console:
  Warning: detected OpenGL error 'invalid enumerant' after
RenderBin::draw(,)
There is strong correlation between this error and the appearance of
segments of the database rendered with a missing texture image.
 
I've attached the mkdb.sh script I used to generate the database.
 
Do you have any hints on what might be going on here? This is not the first
time I've seen this issue, so I don't think it's new.
 
I've tried changing the value of OSG_DATABASE_PAGER_DRAWABLE but this has no
effect on the issue.
 
I'll try rendering the DB on a Windows box and see if I get the same
results.
 
Thanks for any help.
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 

___
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] VPB texture disappear

[Jason Beverage]

Hey Paul,

I've seen this issue before with non power of two textures not being
supported in some situations.  VPB tries to use npot textures at the lowest
level of detail to most closely match the resolution of the source imagery.
Try generating your DB using --pot and see what happens, I bet it works.

Thanks,

Jason

On Mon, Apr 6, 2009 at 9:37 PM, Paul Martz  wrote:

>  Same issue viewing the DB on a Windows box, so apparently not a driver
> issue.
>
> I'll try building the DB on Windows. This might take a while; it's not a
> hot box like my MacPro...
>
> Paul Martz
> *Skew Matrix Software LLC*
> http://www.skew-matrix.com
> +1 303 859 9466
>
>
>  --
> *From:* osg-users-boun...@lists.openscenegraph.org [mailto:
> osg-users-boun...@lists.openscenegraph.org] *On Behalf Of *Paul Martz
> *Sent:* Monday, April 06, 2009 7:30 PM
> *To:* osg-users@lists.openscenegraph.org
> *Subject:* [osg-users] VPB texture disappear
>
>  Hi Robert -- I'm seeing an issue where, as the viewpoint approaches the
> terrain, textures surrounding a high-res inset disappear. This is with very
> current VPB and OSG running on Mac OS X with a GeForce 8800 card.
>
> ss0.jpg shows the scene just before the issue occurs. As I move the
> viewpoint forward (closer to the terrain), I see the image in ss1.jpg. I
> also get several instances of the following error in the console:
>   Warning: detected OpenGL error 'invalid enumerant' after
> RenderBin::draw(,)
> There is strong correlation between this error and the appearance of
> segments of the database rendered with a missing texture image.
>
> I've attached the mkdb.sh script I used to generate the database.
>
> Do you have any hints on what might be going on here? This is not the first
> time I've seen this issue, so I don't think it's new.
>
> I've tried changing the value of OSG_DATABASE_PAGER_DRAWABLE but this has
> no effect on the issue.
>
> I'll try rendering the DB on a Windows box and see if I get the same
> results.
>
> Thanks for any help.
>
> Paul Martz
> *Skew Matrix Software LLC*
> http://www.skew-matrix.com
> +1 303 859 9466
>
>
> ___
> 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] VPB texture disappear

[Paul Martz]

Same issue viewing the DB on a Windows box, so apparently not a driver
issue.
 
I'll try building the DB on Windows. This might take a while; it's not a hot
box like my MacPro...
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 

  _  

From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Paul Martz
Sent: Monday, April 06, 2009 7:30 PM
To: osg-users@lists.openscenegraph.org
Subject: [osg-users] VPB texture disappear


Hi Robert -- I'm seeing an issue where, as the viewpoint approaches the
terrain, textures surrounding a high-res inset disappear. This is with very
current VPB and OSG running on Mac OS X with a GeForce 8800 card.
 
ss0.jpg shows the scene just before the issue occurs. As I move the
viewpoint forward (closer to the terrain), I see the image in ss1.jpg. I
also get several instances of the following error in the console:
  Warning: detected OpenGL error 'invalid enumerant' after
RenderBin::draw(,)
There is strong correlation between this error and the appearance of
segments of the database rendered with a missing texture image.
 
I've attached the mkdb.sh script I used to generate the database.
 
Do you have any hints on what might be going on here? This is not the first
time I've seen this issue, so I don't think it's new.
 
I've tried changing the value of OSG_DATABASE_PAGER_DRAWABLE but this has no
effect on the issue.
 
I'll try rendering the DB on a Windows box and see if I get the same
results.
 
Thanks for any help.
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

[osg-users] VPB texture disappear

[Paul Martz]

Hi Robert -- I'm seeing an issue where, as the viewpoint approaches the
terrain, textures surrounding a high-res inset disappear. This is with very
current VPB and OSG running on Mac OS X with a GeForce 8800 card.
 
ss0.jpg shows the scene just before the issue occurs. As I move the
viewpoint forward (closer to the terrain), I see the image in ss1.jpg. I
also get several instances of the following error in the console:
  Warning: detected OpenGL error 'invalid enumerant' after
RenderBin::draw(,)
There is strong correlation between this error and the appearance of
segments of the database rendered with a missing texture image.
 
I've attached the mkdb.sh script I used to generate the database.
 
Do you have any hints on what might be going on here? This is not the first
time I've seen this issue, so I don't think it's new.
 
I've tried changing the value of OSG_DATABASE_PAGER_DRAWABLE but this has no
effect on the issue.
 
I'll try rendering the DB on a Windows box and see if I get the same
results.
 
Thanks for any help.
 
Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com  
+1 303 859 9466
 
<><>#!/bin/sh
osgdem -v 10 --geocentric --terrain -d ../source-elev -t ../source-img --so 
globe.vpb -o globe.ive
vpbcache --cache globe.cache -s globe.vpb --add
vpbmaster -s globe.vpb --cache globe.cache
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

[osg-users] searching for old friend

[grafXone]

howdy


im looking for an old friend
>>> aweson / ex illusion aka marc balzig <<<

ive seen a post on this list but its from 2006..

so if someone knows him please tell him im looking for him.
maybe hes still registered in the mailing list. :))

regards


gentleman

Pt! Schon vom neuen WEB.DE MultiMessenger gehört? 
Der kann`s mit allen: http://www.produkte.web.de/messenger/?did=3123

___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] problem in picking operation

[Martin Beckett]

I was thinking of type, number of vertices, normals, colour - address of parent 
and any children.
I agree it's hard to make it part of the standard build because every user 
would want something different. I just wanted to make it easier for a user (me) 
to understand what is happening in a scenegraph without drilling down in the 
debugger.

A print visitor is more OSG'ish in style and is easier to overload to provide 
more detail if the user needs it. 

The print visitor takes an ostream directly but I could use getNotifyLevel() 
inside the apply() and output more detail (such as every vertex) at high levels.

For the archives:

Code:

#include 

PrintVisitor pv(osg::notify(osg::NOTICE));
node->accept(pv);



[/code]

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9810#9810





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] start/stop viewer

[Bob Youmans]

Thanks for the very helpful guidance you provide.

 

In the MFC example, I think the .run() method is already replaced by this:

 

//viewer->run();

while(!viewer->done())

{

osg->PreFrameUpdate();

viewer->frame();

osg->PostFrameUpdate();

}

I understand your suggestion about the details in the frame() call.  Here's
more detail about my intended use:

 

For example, I want the user to be able to spin the model and look at it,
but then click a start button and then (ignoring the user inputs)
systematically traverse an axis between some values, and/or viewing angles,
and then when finished, let the user look at it again (with the rendering
details changed (color transparency, other textures, etc.).

 

I was thinking I could stop the normal viewer threading that responds to
user input and then drive it myself from another function.When finished I
would restart the viewer to respond to user input.  There's probably a
better way.

 

Would I be better off using update traversals, animation stuff, or some
other osg design I haven't learned yet?  

 

How about osgPPU? What do you think of it?

 

Thanks,

bob

 

From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Robert
Osfield
Sent: Monday, April 06, 2009 2:37 PM
To: OpenSceneGraph Users
Subject: Re: [osg-users] start/stop viewer

 

Hi Bob,

The viewer only renders a frame when you call viewer.frame().  If you are
calling the convenience function viewer.run() then just replace this run
call with the constuent parts of Viewer::run() i.e.

while(!viewer.done())
{
   viewer.advance();
   viewer.eventTraversal();
   viewer.updateTraversal();
   viewer.renderingTraversals();
}

Robert.

On Mon, Apr 6, 2009 at 6:56 PM, Bob Youmans  wrote:

Hello osg-list,

 

I'd like to find the best way to adapt osg::viewer to control its rendering:
in one case I want the usual case where viewer->run is doing its thing at 60
fps.  But, then, I have other GUI windows (MFC-based) and I want to stop the
viewer, and then render a single frame, all using the same geometry.  What's
the best approach for such a design?  Can I (using the examples for MFC)
call viewer->stopThreads and then "manually" render a single frame, and then
call  run, startThreads, etc. subsequently, and repeat this as often as I
like?

 

Thanks,

Bob


___
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] problem in picking operation

[Robert Osfield]

Hi Martin,

On Mon, Apr 6, 2009 at 8:42 PM, Martin Beckett  wrote:

> Defining << for
> osg::Node* node = osgDB::readNodeFile("somefile.osg");
> std::cout << *node << std::endl;
> Isn't such a daft idea - it would certainly help with debugging - I think I
> will take a look at doing that unless it's obviously stupid?
>

Um so what exactly would you output if you did implement a ostream <<
operator for osg::Node? How would you format it.

Too many different variables at play to even begin to implement such an
operator.

In svn/trunk there is a osgUtil::PrintVisitor that does so very basic
printing of class name and nesting, perhaps this is sufficient for your
purposes.

Robert,
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] problem in picking operation

[Jean-Sébastien Guay]

Hi Martin,

Defining << for 
osg::Node* node = osgDB::readNodeFile("somefile.osg"); 
std::cout << *node << std::endl;

Isn't such a daft idea - it would certainly help with debugging - I think I 
will take a look at doing that unless it's obviously stupid?


No, I don't think it's stupid, I guess everyone would have slightly 
different expectations as to what it should output but a suitable 
default version could be integrated into OSG, sure.


I guess you could print out the address and, if name is not empty, the name.

If you really have time on your hands, you could do the same for Group 
(which would go through its children and print them out), Drawable, 
Geode (which would go through its drawables), MatrixTransform (which 
would print out the matrix), etc. :-)


Have fun,

J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] problem in picking operation

[Martin Beckett]

Defining << for 
osg::Node* node = osgDB::readNodeFile("somefile.osg"); 
std::cout << *node << std::endl;
Isn't such a daft idea - it would certainly help with debugging - I think I 
will take a look at doing that unless it's obviously stupid?

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9806#9806





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] start/stop viewer

[Robert Osfield]

Hi Bob,

The viewer only renders a frame when you call viewer.frame().  If you are
calling the convenience function viewer.run() then just replace this run
call with the constuent parts of Viewer::run() i.e.

while(!viewer.done())
{
   viewer.advance();
   viewer.eventTraversal();
   viewer.updateTraversal();
   viewer.renderingTraversals();
}

Robert.

On Mon, Apr 6, 2009 at 6:56 PM, Bob Youmans  wrote:

>  Hello osg-list,
>
>
>
> I’d like to find the best way to adapt osg::viewer to control its
> rendering:  in one case I want the usual case where viewer->run is doing its
> thing at 60 fps.  But, then, I have other GUI windows (MFC-based) and I want
> to stop the viewer, and then render a single frame, all using the same
> geometry.  What’s the best approach for such a design?  Can I (using the
> examples for MFC) call viewer->stopThreads and then “manually” render a
> single frame, and then call  run, startThreads, etc. subsequently, and
> repeat this as often as I like?
>
>
>
> Thanks,
>
> Bob
>
> ___
> 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] How to get heightfield from ReadFileCallback VPB Terrain

[Ralf Stokholm]

Hi Robert

"Lower levels of detail aren't unreferenced but kept around in any paged
database, this includes a osgTerrain::TerrainTile based paged database."

I was expecting this.

With regards to active tiles, my problem is that going this approach I
cant figure out how to extract the state information from the
scenegraph. From what I understand the TerrainTiles that I process
during my read file callback will not provide any information on when
they are invalidated by the loading of higher resolution tiles, right?

I understand that the physics part is specific to my application, but
an interface or "pattern" for syncronising the active scenegraph
generated when running an osgTerrain::TerrainTile based paged database
with a given application is usefull in other places.

As I said so far we have been using a cullvisitor that runs through
the entire scenegraph looking for "terrain" nodes. It then checks if
these nodes are already in our set of nodes. If its not a new physics
object is generated by pulling the geometry using a trianglevisitor.
This method is both slow, repeats bookkeping already performed in the
Terrain/Scene graph machinery and uses way more memmory than is
neccecery. Basically I end up generating 3 unique vertices and 3
indecies for each triangle rendered. This will consime roughly 28
times the amount of memmory that would be needed if I could use the
heightfield data directly.

Brgs. Ralf

2009/4/6 Robert Osfield :
> Hi Ralf,
>
> Lower levels of detail aren't unreferenced but kept around in any paged
> database, this includes a osgTerrain::TerrainTile based paged database.
>
> As how to manage the mapping of which physics tiles should be active... well
> I don't know what to recommend, there a lots of different ways you could
> probably tackle it.  My job is the scene graph, I'll leave application dev
> issues to members of the community.
>
> Robert.
>
> On Mon, Apr 6, 2009 at 3:12 PM, Ralf Stokholm 
> wrote:
>>
>> Hi Robert
>>
>> I have gotten a bit further with this, and there are some indications
>> it might be a viable solution.
>>
>> But I have run into a problem, it seams that Im not removing patches
>> from my physics engine when higher resolutions patches replaces them.
>>
>> Can You confirm that If a given osgTerrain::TerrainTile is replaced by
>> its four higher resolution children the acociated heightfields are not
>> deleted, and as a result the observer callback is not called?
>>
>> And if this is the case do you have any good ideas on how to figure
>> out when to disable these patches in the physics engine, so that I
>> dont have multiple patches loaded for the same area?
>>
>> Brgs.
>>
>> Ralf Stokholm
>>
>> 2009/4/3 Robert Osfield :
>> > Hi Ralf,
>> >
>> > Use the osg::Observer mechanism for watching when objects get deleted.
>> > Just
>> > register your observer with the TerrainTile/HeightField you've set up
>> > your
>> > physics for then you'll be able to get your observer called when it goes
>> > out
>> > of scope.   osg::observert_ptr<> uses osg::Observer, as does osg::State
>> > so
>> > have a look at these for usage of Observer.
>> >
>> > Robert.
>> >
>> > On Fri, Apr 3, 2009 at 11:44 AM, Ralf Stokholm 
>> > wrote:
>> >>
>> >> Hi Robert
>> >>
>> >> Its not that I need to keep the memmory around.
>> >>
>> >> I need to know when a terrain tile is unloaded from the scene and use
>> >> that event to unload the physics heightfield as well.
>> >>
>> >> Im guessing the life of a TerrainTile is similar to this:
>> >>
>> >> 1. Loaded/created by the database pager
>> >> 2. Added to the scenegraph by the databasepager.
>> >> 3. Disabled/hidden by the CullVisitor.
>> >> 4. Enabled/Shown by the CullVisitor. (possibly if it is only culled
>> >> out for a short time OSG 2.6.1)
>> >> 5. Unloaded/Destroyed.
>> >>
>> >> I need a way to catch at least event 1 and 5. Could this be done using
>> >> reference counting or is there a better way?
>> >>
>> >> Brgs.
>> >>
>> >> Ralf Stokholm
>> >>
>> >> 2009/4/3 Robert Osfield :
>> >> > Hi Ralf,
>> >> >
>> >> > If you need to keep the memory around then use a ref_ptr<> to the
>> >> > HeightField, but... this will keep the memory around so if you do no
>> >> > longer
>> >> > need it you might want to implement a scheme to free it, such as
>> >> > watching
>> >> > when the ref count goes to one.
>> >> >
>> >> > The other approach is to use an observer_ptr<> that doesn't change
>> >> > the
>> >> > ref
>> >> > count of the HeightFiled, and gets reset to 0 when the HeightField
>> >> > get
>> >> > deleted.  You can also use a custom osg::Observer to watch for object
>> >> > deletions as well.
>> >> >
>> >> > Robert
>> >> >
>> >> > On Fri, Apr 3, 2009 at 11:22 AM, Ralf Stokholm
>> >> > 
>> >> > wrote:
>> >> >>
>> >> >> Hi Robert (list)
>> >> >>
>> >> >> Im trying to use the ReadFileCallback to load the heightfield int my
>> >> >> physics engine.
>> >> >>
>> >> >> So far I have been using a CullVisitor to grap the triangles from
>> >> >> the
>> >> >

[osg-users] start/stop viewer

[Bob Youmans]

Hello osg-list,

 

I'd like to find the best way to adapt osg::viewer to control its rendering:
in one case I want the usual case where viewer->run is doing its thing at 60
fps.  But, then, I have other GUI windows (MFC-based) and I want to stop the
viewer, and then render a single frame, all using the same geometry.  What's
the best approach for such a design?  Can I (using the examples for MFC)
call viewer->stopThreads and then "manually" render a single frame, and then
call  run, startThreads, etc. subsequently, and repeat this as often as I
like?

 

Thanks,

Bob

___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] problem in picking operation

[Jean-Sébastien Guay]

Hello Msrsas? (please sign posts using a real name, it makes exchanges 
much more friendly)



1. If I use "readNodeFile" and try to print out using std, it gives the error as error 
LNK2019: unresolved external symbol "__declspec(dllimport) public: class 
std::basic_ostream

You're not being very precise (including actual code would have helped), 
but this sounds like you're trying to do something like this:


osg::Node* node = osgDB::readNodeFile("somefile.osg");
std::cout << *node << std::endl;

Now, unless I'm mistaken, this cannot work because nowhere in OSG is an 
output operator (operator<<) defined which takes an osg::Node. That's 
why you'd be getting an error about an unresolved external symbol, 
because the linker is trying to find this output operator but it isn't 
defined anywhere. You could make one if you want.



2. QSG Picking example gives the output but if i click on the cow object it gives error 
as "Unhandled exception at 0x7c901230..." and ask for user breakpoint in 
dbgheap.c file.
I have tried intersection using line intersector in other program but the 
same error occurs.


Seems to me like you're trying to dereference a null pointer? Again, 
relevant code would help and perhaps also a stack trace... "Unhandled 
exception" is just a generic error message meaning "something went 
wrong", without more info no one could do more than guess at what 
exactly might be going wrong.


Hope this helps,

J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] problem in picking operation

[Martin Beckett]

The ostream error isn't OSG. It sounds like somehow you don't have the standard 
library working. Can you write just a simple c++ program to print on stdout?
For console output in osg it's safer to use the osg::Notify stream.

For the second part - If you haven't changed the example code to introduce a 
bug, I suspect you have a mix of OSG compiled with a different compiler. Can 
you build osg with the VS 2003?

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9801#9801





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] RTT and viewport offset

[Paul Martz]

Hi Rick -- That's a piece of code I've looked at, and scratched my head a
little, but I think it's right.

Adding the xy to the width and height makes sense. OSG is trying to compute
the total space required for rendering. If your viewport is
(200,50,800,600), then the total number of pixels you need in that buffer
is: ( (200+800) * (50+600) ) pixels. If you fail to add in the xy values,
the resulting buffer will not be big enough.

I'm curious if you have small reproducer code that shows this is really an
issue? Because what I hear you saying is:
 * You render to a texture (presumable with the default viewport)
 * Then you render that texture in a scene with xy viewport offsets.

So in that case, there would be two render stages:
 * In the first stage, all the buffer attachments would be the size of your
destination texture, which presumably uses the default viewport with (0,0)
xy offsets.
 * Only the second stage would need buffer attachments whose size
incorporates the non-zero xy viewport offsets.

The two sets of buffers should be consistent within each render stage and
you should not get an OpenGL error.

Paul Martz
Skew Matrix Software LLC
http://www.skew-matrix.com
+1 303 859 9466

-Original Message-
From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Rick
Appleton
Sent: Monday, April 06, 2009 8:39 AM
To: osg-users@lists.openscenegraph.org
Subject: [osg-users] RTT and viewport offset

Hi,

I've encountered an issue with rendering to texture and then displaying that
texture with a viewport that doesn't start at 0,0. I'm using osgPPU for
this, but I believe the issue is in OSG.

In osgUtil/RenderStage.cpp, in RenderStage::runCameraSetUp the width and
height are calculated as:

int width = static_cast(_viewport->x() + _viewport->width());
int height = static_cast(_viewport->y() + _viewport->height());

This width and height is then used to create RenderBuffers for the missing
attachments. If the viewport starts at non-zero values, this means the
created RenderBuffers will not have dimensions (vp->width, vp->height)
whereas the attached texture likely will have those dimenstions. Thus the
FBO will then generate FRAMEBUFFER_INCOMPLETE_DIMENSIONS_EXT error.

Removing the additions of _viewport->x()/y() fixes my issue. I was wondering
if there was a good reason for doing it the way it's done.

Kind regards,
Rick

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9797#9797





___
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] How to get heightfield from ReadFileCallback VPB Terrain

[Robert Osfield]

Hi Ralf,

Lower levels of detail aren't unreferenced but kept around in any paged
database, this includes a osgTerrain::TerrainTile based paged database.

As how to manage the mapping of which physics tiles should be active... well
I don't know what to recommend, there a lots of different ways you could
probably tackle it.  My job is the scene graph, I'll leave application dev
issues to members of the community.

Robert.

On Mon, Apr 6, 2009 at 3:12 PM, Ralf Stokholm wrote:

> Hi Robert
>
> I have gotten a bit further with this, and there are some indications
> it might be a viable solution.
>
> But I have run into a problem, it seams that Im not removing patches
> from my physics engine when higher resolutions patches replaces them.
>
> Can You confirm that If a given osgTerrain::TerrainTile is replaced by
> its four higher resolution children the acociated heightfields are not
> deleted, and as a result the observer callback is not called?
>
> And if this is the case do you have any good ideas on how to figure
> out when to disable these patches in the physics engine, so that I
> dont have multiple patches loaded for the same area?
>
> Brgs.
>
> Ralf Stokholm
>
> 2009/4/3 Robert Osfield :
> > Hi Ralf,
> >
> > Use the osg::Observer mechanism for watching when objects get deleted.
> Just
> > register your observer with the TerrainTile/HeightField you've set up
> your
> > physics for then you'll be able to get your observer called when it goes
> out
> > of scope.   osg::observert_ptr<> uses osg::Observer, as does osg::State
> so
> > have a look at these for usage of Observer.
> >
> > Robert.
> >
> > On Fri, Apr 3, 2009 at 11:44 AM, Ralf Stokholm 
> > wrote:
> >>
> >> Hi Robert
> >>
> >> Its not that I need to keep the memmory around.
> >>
> >> I need to know when a terrain tile is unloaded from the scene and use
> >> that event to unload the physics heightfield as well.
> >>
> >> Im guessing the life of a TerrainTile is similar to this:
> >>
> >> 1. Loaded/created by the database pager
> >> 2. Added to the scenegraph by the databasepager.
> >> 3. Disabled/hidden by the CullVisitor.
> >> 4. Enabled/Shown by the CullVisitor. (possibly if it is only culled
> >> out for a short time OSG 2.6.1)
> >> 5. Unloaded/Destroyed.
> >>
> >> I need a way to catch at least event 1 and 5. Could this be done using
> >> reference counting or is there a better way?
> >>
> >> Brgs.
> >>
> >> Ralf Stokholm
> >>
> >> 2009/4/3 Robert Osfield :
> >> > Hi Ralf,
> >> >
> >> > If you need to keep the memory around then use a ref_ptr<> to the
> >> > HeightField, but... this will keep the memory around so if you do no
> >> > longer
> >> > need it you might want to implement a scheme to free it, such as
> >> > watching
> >> > when the ref count goes to one.
> >> >
> >> > The other approach is to use an observer_ptr<> that doesn't change the
> >> > ref
> >> > count of the HeightFiled, and gets reset to 0 when the HeightField get
> >> > deleted.  You can also use a custom osg::Observer to watch for object
> >> > deletions as well.
> >> >
> >> > Robert
> >> >
> >> > On Fri, Apr 3, 2009 at 11:22 AM, Ralf Stokholm <
> alfma...@arenalogic.com>
> >> > wrote:
> >> >>
> >> >> Hi Robert (list)
> >> >>
> >> >> Im trying to use the ReadFileCallback to load the heightfield int my
> >> >> physics engine.
> >> >>
> >> >> So far I have been using a CullVisitor to grap the triangles from the
> >> >> finished scene, but it has major drawback in that it forces me to
> load
> >> >> the data as meshes which is terrible for memmory and takes a long
> time
> >> >> to load.
> >> >>
> >> >> It seams im able to load the data as heightfields from the
> >> >> ReadFileCallback, but this leads to my next problem. I need a way to
> >> >> reference the data later, to be able to remove it when tiles are
> >> >> deletet and disable it when tiles are disabled by the scene
> >> >> cullvisitor.
> >> >>
> >> >> Is there a way to relate the nodes i visit in my ReadFileCallback to
> >> >> nodes i visit in a special cullvisitor for the final scenegraph?
> >> >>
> >> >> Basically do you have any hints on how to syncronise the state of my
> >> >> physics engine with the state of my Paged VPB terrain?
> >> >>
> >> >> Brgs.
> >> >>
> >> >> Ralf Stokholm
> >> >>
> >> >>
> >> >> 2009/4/2 Ralf Stokholm :
> >> >> > Hi Robert
> >> >> >
> >> >> > Thanks a lot, was starting to get at this but using a more
> primitive
> >> >> > approach of running through all the children of the node that I did
> >> >> > get.
> >> >> >
> >> >> > Will try to implement your method.
> >> >> >
> >> >> > Brgs.
> >> >> >
> >> >> > Ralf Stokholm
> >> >> >
> >> >> > 2009/4/2 Robert Osfield :
> >> >> >> Hi Ralf,
> >> >> >>
> >> >> >> You'll need to write a custom NodeVisitor that traverses through
> the
> >> >> >> loaded
> >> >> >> scene graph and picks out the TerrainTile using dyanmic cast.   It
> >> >> >> just
> >> >> >> so
> >> >> >> happens that I'm actually working on a ReadFileCallback/custom
> >> >> >> Visi

[osg-users] RTT and viewport offset

[Rick Appleton]

Hi,

I've encountered an issue with rendering to texture and then displaying that 
texture with a viewport that doesn't start at 0,0. I'm using osgPPU for this, 
but I believe the issue is in OSG.

In osgUtil/RenderStage.cpp, in RenderStage::runCameraSetUp the width and height 
are calculated as:

int width = static_cast(_viewport->x() + _viewport->width());
int height = static_cast(_viewport->y() + _viewport->height());

This width and height is then used to create RenderBuffers for the missing 
attachments. If the viewport starts at non-zero values, this means the created 
RenderBuffers will not have dimensions (vp->width, vp->height) whereas the 
attached texture likely will have those dimenstions. Thus the FBO will then 
generate FRAMEBUFFER_INCOMPLETE_DIMENSIONS_EXT error.

Removing the additions of _viewport->x()/y() fixes my issue. I was wondering if 
there was a good reason for doing it the way it's done.

Kind regards,
Rick

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9797#9797





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] How to get heightfield from ReadFileCallback VPB Terrain

[Ralf Stokholm]

Hi Robert

I have gotten a bit further with this, and there are some indications
it might be a viable solution.

But I have run into a problem, it seams that Im not removing patches
from my physics engine when higher resolutions patches replaces them.

Can You confirm that If a given osgTerrain::TerrainTile is replaced by
its four higher resolution children the acociated heightfields are not
deleted, and as a result the observer callback is not called?

And if this is the case do you have any good ideas on how to figure
out when to disable these patches in the physics engine, so that I
dont have multiple patches loaded for the same area?

Brgs.

Ralf Stokholm

2009/4/3 Robert Osfield :
> Hi Ralf,
>
> Use the osg::Observer mechanism for watching when objects get deleted.  Just
> register your observer with the TerrainTile/HeightField you've set up your
> physics for then you'll be able to get your observer called when it goes out
> of scope.   osg::observert_ptr<> uses osg::Observer, as does osg::State so
> have a look at these for usage of Observer.
>
> Robert.
>
> On Fri, Apr 3, 2009 at 11:44 AM, Ralf Stokholm 
> wrote:
>>
>> Hi Robert
>>
>> Its not that I need to keep the memmory around.
>>
>> I need to know when a terrain tile is unloaded from the scene and use
>> that event to unload the physics heightfield as well.
>>
>> Im guessing the life of a TerrainTile is similar to this:
>>
>> 1. Loaded/created by the database pager
>> 2. Added to the scenegraph by the databasepager.
>> 3. Disabled/hidden by the CullVisitor.
>> 4. Enabled/Shown by the CullVisitor. (possibly if it is only culled
>> out for a short time OSG 2.6.1)
>> 5. Unloaded/Destroyed.
>>
>> I need a way to catch at least event 1 and 5. Could this be done using
>> reference counting or is there a better way?
>>
>> Brgs.
>>
>> Ralf Stokholm
>>
>> 2009/4/3 Robert Osfield :
>> > Hi Ralf,
>> >
>> > If you need to keep the memory around then use a ref_ptr<> to the
>> > HeightField, but... this will keep the memory around so if you do no
>> > longer
>> > need it you might want to implement a scheme to free it, such as
>> > watching
>> > when the ref count goes to one.
>> >
>> > The other approach is to use an observer_ptr<> that doesn't change the
>> > ref
>> > count of the HeightFiled, and gets reset to 0 when the HeightField get
>> > deleted.  You can also use a custom osg::Observer to watch for object
>> > deletions as well.
>> >
>> > Robert
>> >
>> > On Fri, Apr 3, 2009 at 11:22 AM, Ralf Stokholm 
>> > wrote:
>> >>
>> >> Hi Robert (list)
>> >>
>> >> Im trying to use the ReadFileCallback to load the heightfield int my
>> >> physics engine.
>> >>
>> >> So far I have been using a CullVisitor to grap the triangles from the
>> >> finished scene, but it has major drawback in that it forces me to load
>> >> the data as meshes which is terrible for memmory and takes a long time
>> >> to load.
>> >>
>> >> It seams im able to load the data as heightfields from the
>> >> ReadFileCallback, but this leads to my next problem. I need a way to
>> >> reference the data later, to be able to remove it when tiles are
>> >> deletet and disable it when tiles are disabled by the scene
>> >> cullvisitor.
>> >>
>> >> Is there a way to relate the nodes i visit in my ReadFileCallback to
>> >> nodes i visit in a special cullvisitor for the final scenegraph?
>> >>
>> >> Basically do you have any hints on how to syncronise the state of my
>> >> physics engine with the state of my Paged VPB terrain?
>> >>
>> >> Brgs.
>> >>
>> >> Ralf Stokholm
>> >>
>> >>
>> >> 2009/4/2 Ralf Stokholm :
>> >> > Hi Robert
>> >> >
>> >> > Thanks a lot, was starting to get at this but using a more primitive
>> >> > approach of running through all the children of the node that I did
>> >> > get.
>> >> >
>> >> > Will try to implement your method.
>> >> >
>> >> > Brgs.
>> >> >
>> >> > Ralf Stokholm
>> >> >
>> >> > 2009/4/2 Robert Osfield :
>> >> >> Hi Ralf,
>> >> >>
>> >> >> You'll need to write a custom NodeVisitor that traverses through the
>> >> >> loaded
>> >> >> scene graph and picks out the TerrainTile using dyanmic cast.   It
>> >> >> just
>> >> >> so
>> >> >> happens that I'm actually working on a ReadFileCallback/custom
>> >> >> Visitor
>> >> >> that
>> >> >> does pick out all the TerrainTile's in a subgraph (there will
>> >> >> typically
>> >> >> be
>> >> >> one, two or four due to the quad tree structure generated by VPB.)
>> >> >>
>> >> >> The visitor looks like:
>> >> >>
>> >> >> class FindTerrainTileVisitor : public osg::NodeVisitor
>> >> >> {
>> >> >> public:
>> >> >>     FindTerrainTileVisitor():
>> >> >>     osg::NodeVisitor(osg::NodeVisitor::TRAVERSE_ALL_CHILDREN)
>> >> >>     {
>> >> >>     }
>> >> >>
>> >> >>     void reset()
>> >> >>     {
>> >> >>     _terrainTiles.clear();
>> >> >>     }
>> >> >>
>> >> >>     void apply(osg::Group& group)
>> >> >>     {
>> >> >>     osgTerrain::TerrainTile* terrainTile =
>> >> >> dynamic_cast(&group);
>> >> >>     if (terrainTi

Re: [osg-users] Problem with osg 2.8.0 and Windows XP

[Philipp J.]

Hi Robert,

thanks for your answer! I am currently working to get the stack trace myself. 
This might help even me to understand the problem. 
I hoped someone familiar with that environment might see the problem 
immediately after having a short look at it, maybe that's not so realistic as 
you just wrote. 

By the way: MinGW is the one from inside the Qt 4.5 SDK. MSYS is of version 
1.0.10.

Thanks again!

Philipp

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9795#9795





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Problem with osg 2.8.0 and Windows XP

[Robert Osfield]

Hi Philipp,

I can't see how others can help you unless they've used your specific
version of windows + Mingw and have come across the exact same error.   I
have heard of others using Mingw so it should be possible to get things
working, but it's a pretty uncommon platform for developers to use so it's
not widely tested like the mainstream platforms combinations.

Information to help others help you, that you've so far not provided are:

   Mingw version.
   Stack trace of the crash.
   An explanation of what leads up to the crash/what you app does that is
different from osgviewer.

Without this information bobody has anything to go on so won't be able to
help.

Robert.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Problem with osg 2.8.0 and Windows XP

[Philipp J.]

Maybe this helps: osgsimulation crashes as well [Crying or Very sad] :


Code:

[...]
GL_NV_texture_shader2
GL_NV_texture_shader3
GL_NV_transform_feedback
GL_NV_vertex_array_range
GL_NV_vertex_array_range2
GL_NV_vertex_program
GL_NV_vertex_program1_1
GL_NV_vertex_program2
GL_NV_vertex_program2_option
GL_NV_vertex_program3
GL_S3_s3tc
GL_SGIS_generate_mipmap
GL_SGIS_texture_lod
GL_SGIX_depth_texture
GL_SGIX_shadow
GL_SUN_slice_accum
GL_WIN_swap_hint
WGL_ARB_buffer_region
WGL_ARB_create_context
WGL_ARB_extensions_string
WGL_ARB_make_current_read
WGL_ARB_multisample
WGL_ARB_pbuffer
WGL_ARB_pixel_format
WGL_ARB_pixel_format_float
WGL_ARB_render_texture
WGL_ATI_pixel_format_float
WGL_EXT_extensions_string
WGL_EXT_framebuffer_sRGB
WGL_EXT_pixel_format_packed_float
WGL_EXT_swap_control
WGL_NV_float_buffer
WGL_NV_multisample_coverage
WGL_NV_render_depth_texture
WGL_NV_render_texture_rectangle
OpenGL extension 'GL_ARB_vertex_program' is supported.
OpenGL extension 'GL_EXT_secondary_color' is supported.
OpenGL extension 'GL_EXT_fog_coord' is supported.
OpenGL extension 'GL_ARB_multitexture' is supported.
OpenGL extension 'GL_NV_occlusion_query' is supported.
OpenGL extension 'GL_ARB_occlusion_query' is supported.
OpenGL extension 'GL_EXT_timer_query' is supported.
Done DatabasePager::addLoadedDataToSceneGraph0.00180212ms,  0.000805152ms  
objects0

OpenGL extension 'GL_ARB_vertex_program' is supported.
Setting up osg::Camera::FRAME_BUFFER_OBJECT
OpenGL extension 'GL_ARB_multitexture' is supported.
OpenGL extension 'GL_EXT_texture_filter_anisotropic' is supported.
OpenGL extension 'GL_ARB_texture_compression' is supported.
OpenGL extension 'GL_EXT_texture_compression_s3tc' is supported.
OpenGL extension 'GL_IBM_texture_mirrored_repeat' is supported.
OpenGL extension 'GL_EXT_texture_edge_clamp' is supported.
OpenGL extension 'GL_ARB_texture_border_clamp' is supported.
OpenGL extension 'GL_SGIS_generate_mipmap' is supported.
OpenGL extension 'GL_ARB_shadow' is supported.
OpenGL extension 'GL_ARB_shadow_ambient' is not supported.
OpenGL extension 'GL_APPLE_client_storage' is not supported.
OpenGL extension 'GL_ARB_texture_non_power_of_two' is supported.
OpenGL extension 'GL_EXT_texture_integer' is supported.



--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9793#9793





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] Warning : Could not find plug-in forreading objects from file .osga?

[Tomlinson, Gordon]

Make sure that the OSG DLL's and plugin DLL's are in your PATH 


Gordon
Product Manager 3d
__
Gordon Tomlinson
Email  : gtomlinson @ overwatch.textron.com
__


-Original Message-
From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Akilan
Sent: Monday, April 06, 2009 6:50 AM
To: osg-users@lists.openscenegraph.org
Subject: [osg-users] [osgPlugins] Warning : Could not find plug-in
forreading objects from file .osga?

Hi,

I am using osg2.2(from binary executable) on visual studio 2005. 
My application loads (readNodeFile) osga files from specified folder.
It happens properly when the folder contains less number of osga files. 
When the number of files exceed certain number (say some 500 or more),
it throws  "Warning : Could not find plug-in for reading objects from
file xyz.osga" and readNodeFile fails.

I saw  the post for similar kind of problem while reading png /jpeg
files.
Let me know what needs to be done for osga file reading error?
... 

Thank you.

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9786#9786





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.or
g
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Problem with osg 2.8.0 and Windows XP

[Philipp J.]

Sorry, I forgot to mention the environment: 
I am using the MinGW compiler within the MSYS environment together with the 
current Qt4-SDK.
cheers,

Philipp

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9791#9791





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Loading of osga models dynamically?

[Robert Osfield]

Hi Aklin,

You post just confuses me, there really isn't enough context to what you are
doing to really know what is going on, and the the code itself just looks
odd, for instance the self reference line:

pLOD->setRange(child_no,pLOD->
>
> getMinRange(child_no),pLOD->getMaxRange(child_no));


Really don't know what you expect to be happening here.  I kinda get the
feeling that you need learn a bit more about the concept LODs as much as the
OSG API.

Robert.


On Mon, Apr 6, 2009 at 12:41 PM, Akilan  wrote:

> Hi
>
> I got some idea from the previous reply and did something in my
> application. But I am not sure that the my following code is doing dynamic
> loading,
>
> //On reading each node(osga file)
> osg::Node *node=osgDB::readNodeFile(argv[i]);
> osg::PagedLOD *pLOD=findPLOD(node);
> int child_no=pLOD->getChildIndex();
>
> pLOD->setRange(child_no,pLOD->getMinRange(child_no),pLOD->getMaxRange(child_no));
> group->addChild(node);
> ...
> ...
> viewer.setSceneData(group.get());
> viewer.run();
>
> If ur answer is redundant, pls execuse and get me cleared.
>
> --
> Read this topic online here:
> http://forum.openscenegraph.org/viewtopic.php?p=9788#9788
>
>
>
>
>
> ___
> 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] Loading of osga models dynamically?

[Akilan]

Hi

I got some idea from the previous reply and did something in my application. 
But I am not sure that the my following code is doing dynamic loading, 

//On reading each node(osga file)
osg::Node *node=osgDB::readNodeFile(argv[i]);
osg::PagedLOD *pLOD=findPLOD(node);
int child_no=pLOD->getChildIndex();
pLOD->setRange(child_no,pLOD->getMinRange(child_no),pLOD->getMaxRange(child_no));
group->addChild(node);
...
...
viewer.setSceneData(group.get());
viewer.run();

If ur answer is redundant, pls execuse and get me cleared.

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9788#9788





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] OpenFlight-Plugin - How to store individual object in seperate geode?

[Brede Johansen]

Bulkhead,

If you use osgconv to compare your OpenFlight structure with the OSG
structure (.osg) don't forget to set the environment variable
OSG_OPTIMIZER=OFF.  With the preserve??? reader options they should be
almost identical.

The reason OpenFlight faces becomes Geode+Drawable is because a drawable
can't store flight comments and billboard information.  If I remember
correct a long time ago the drawables couldn't even store the face id
because the drawables didn't have the name attribute.


Brede

On Mon, Apr 6, 2009 at 11:59 AM, Bulkhead  wrote:

> Hi Brede,
>
> Just some updates. I have glanced through the codes for the openflight
> plugin, and not sure if this is correct.
>
> Using the "preserveObject" option, the openflight Object record will be
> stored in a osg::Group node. All the faces will be stored as osg::Geode
> under the respective group node. Is this correct?
>
> By iterating through the groups, and collecting the geodes in each group, I
> managed to retrieve the land from my openflight terrain file. However, the
> buildings are still being grouped based on the texture (i.e. building with
> same texture are stored under the same group node). Is this how the
> openflight structure organize the record? Is it feasible to seperate these
> buildings?
>
> Thanks in advanced.
>
> --
> Read this topic online here:
> http://forum.openscenegraph.org/viewtopic.php?p=9778#9778
>
>
>
>
>
> ___
> 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] [osgPlugins] Warning : Could not find plug-in for reading objects from file .osga?

[Akilan]

Hi,

I am using osg2.2(from binary executable) on visual studio 2005. 
My application loads (readNodeFile) osga files from specified folder.
It happens properly when the folder contains less number of osga files. 
When the number of files exceed certain number (say some 500 or more), it 
throws  "Warning : Could not find plug-in for reading objects from file 
xyz.osga" and readNodeFile fails.

I saw  the post for similar kind of problem while reading png /jpeg files.
Let me know what needs to be done for osga file reading error?
... 

Thank you.

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9786#9786





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Windows users - VS2008 comments

[Gordon Tomlinson]

Something we have been told on 2VS2008 is the trick with upgrading projects
is not to led VS2008 doit , simply open the proje files in a tex editor and
manually change the version number from 2005 to 2008 version

 


__

Gordon Tomlinson 

  gor...@gordontomlinson.com
IM:   gordon3db...@3dscenegraph.com
  www.vis-sim.com
 www.gordontomlinson.com 


__

 

From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Donald
Cipperly
Sent: Monday, April 06, 2009 6:26 AM
To: OpenSceneGraph Users
Subject: Re: [osg-users] Windows users - VS2008 comments

 

We upgraded from VS2003 to VS2008 a few months ago and the only issue we ran
into was in the conversion of existing projects from VS2003 to VS2008.  It
seems there are a few issues with optimization settings from converted
projects.  Some of our projects that were showing as having /O2
optimization, where actually being built with no optimization.  We just
decided to re-create the projects in VS2008 and have not had any issues
since then.

- Donny



On Sat, Apr 4, 2009 at 11:30 AM, sherman wilcox 
wrote:

Slightly off-topic for this mailing list but here goes. For the
Windows users out there, those of you that are using or know of
developers using VS2008 - any comments? Problems, benefits, pros/cons?

My project is using VS2005 and has been for a while now but I'm
considering switching to VS2008. In my case there's only one reason -
the /MP benefit. /MP enables parallel compilation. For multiprocessor
or multicore systems it's a real performance increase. I'm not talking
about parallel builds of multiple projects but compiling multiple
modules in the same project(s) in parallel. On VS2005 /MP is
undocumented but works, mostly. It's a tremendous performance boost.
However, I've ran into a PDB corruption issue here as of late. Error
C2471. Not sure why it's suddenly happening as I've been using /MP for
a long while now. I believe Gordon warned me this could happen and it
has. I phoned Microsoft about this and they said that /MP is
unsupported on VS2005 and I have no recourse but to either not use
that feature (no way - too fast) or upgrade to VS2008. So, I may be
upgrading. However, there are numerous developers on my project and I
want to have a clear understanding of any potential issues before
moving forward. I'll be purchasing a copy to test but I'd like to hear
from the list.
___
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] Windows users - VS2008 comments

[Donald Cipperly]

We upgraded from VS2003 to VS2008 a few months ago and the only issue we ran
into was in the conversion of existing projects from VS2003 to VS2008.  It
seems there are a few issues with optimization settings from converted
projects.  Some of our projects that were showing as having /O2
optimization, where actually being built with no optimization.  We just
decided to re-create the projects in VS2008 and have not had any issues
since then.

- Donny


On Sat, Apr 4, 2009 at 11:30 AM, sherman wilcox wrote:

> Slightly off-topic for this mailing list but here goes. For the
> Windows users out there, those of you that are using or know of
> developers using VS2008 - any comments? Problems, benefits, pros/cons?
>
> My project is using VS2005 and has been for a while now but I'm
> considering switching to VS2008. In my case there's only one reason -
> the /MP benefit. /MP enables parallel compilation. For multiprocessor
> or multicore systems it's a real performance increase. I'm not talking
> about parallel builds of multiple projects but compiling multiple
> modules in the same project(s) in parallel. On VS2005 /MP is
> undocumented but works, mostly. It's a tremendous performance boost.
> However, I've ran into a PDB corruption issue here as of late. Error
> C2471. Not sure why it's suddenly happening as I've been using /MP for
> a long while now. I believe Gordon warned me this could happen and it
> has. I phoned Microsoft about this and they said that /MP is
> unsupported on VS2005 and I have no recourse but to either not use
> that feature (no way - too fast) or upgrade to VS2008. So, I may be
> upgrading. However, there are numerous developers on my project and I
> want to have a clear understanding of any potential issues before
> moving forward. I'll be purchasing a copy to test but I'd like to hear
> from the list.
> ___
> 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] Problem with osg 2.8.0 and Windows XP

[Philipp J.]

Hi there!

I hope anyone can help me with this problem:

I successfully compiled OSG 2.8.0 and used and linked it to my application 
without problems so far. But, when I want to run my application, it crashes 
after a few seconds with the typical MS crash dialog.
osgviewer.exe works fine with .ive and .png files which are also used by my 
application.
Setting the environment variable OSG_NOTIFY_LEVEL to DEBUG shows me this output 
up to the crash:


Code:

GraphicsContext::setWindowingSystemInterface() 0x3d9b48 0x707edab0
itr='c:\repository\software\3DVisualization\3DVisualization-devel\bin'
FindFileInPath() : trying 
C:\repository\software\3DVisualization\3DVisualization-devel\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll
 ...
itr='C:\WINDOWS\system32'
FindFileInPath() : trying 
C:\WINDOWS\system32\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='C:\WINDOWS\System'
FindFileInPath() : trying 
C:\WINDOWS\System\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='C:\WINDOWS'
FindFileInPath() : trying C:\WINDOWS\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='.'
FindFileInPath() : trying 
C:\repository\software\3DVisualization\3DVisualization-devel\osgPlugins-2.8.0\mingw_osgdb_ive.dll
 ...
itr='.'
FindFileInPath() : trying 
C:\repository\software\3DVisualization\3DVisualization-devel\osgPlugins-2.8.0\mingw_osgdb_ive.dll
 ...
itr='C:\Programme\msys\1.0\local\bin'
FindFileInPath() : trying 
C:\Programme\msys\1.0\local\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\programme\qt\2009.01\mingw\bin'
FindFileInPath() : trying 
C:\programme\qt\2009.01\mingw\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='C:\Programme\msys\1.0\bin'
FindFileInPath() : trying 
C:\Programme\msys\1.0\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\WINDOWS\system32'
FindFileInPath() : trying 
C:\WINDOWS\system32\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\WINDOWS'
FindFileInPath() : trying C:\WINDOWS\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\WINDOWS\System32\Wbem'
FindFileInPath() : trying 
C:\WINDOWS\System32\Wbem\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\programme\qt\2009.01\mingw\bin'
FindFileInPath() : trying 
C:\programme\qt\2009.01\mingw\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\Programme\Qt\2009.01\qt\bin'
FindFileInPath() : trying 
C:\Programme\Qt\2009.01\qt\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
itr='c:\Programme\OpenSceneGraph\bin'
FindFileInPath() : trying 
C:\Programme\OpenSceneGraph\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll ...
FindFileInPath() : USING 
C:\Programme\OpenSceneGraph\bin\osgPlugins-2.8.0\mingw_osgdb_ive.dll
Opened DynamicLibrary osgPlugins-2.8.0/mingw_osgdb_ive.dll
FindFileInPath(C:\repository\software\configuration/models/airports/transformed.ive):
 returning C:\repository\software\configuration/models/airports/transformed.ive
ive::DataInputStream.setLoadExternalReferenceFiles()=1
uncompressed ive stream
NodeTrackerManipulator::setTrackNode(Node*0x78ef978 ): Path set
setTrackNode()
  MatrixTransform ''
NodeTrackerManipulator::setTrackNode(Node*0x85bdfa0 ): Path set
setTrackNode()
  MatrixTransform ''
GraphicsContext::registerGraphicsContext 0x7001ff0
GraphicsContext::createNewContextID() creating contextID=0
Updating the MaxNumberOfGraphicsContexts to 1
CullSettings::readEnvironmentalVariables()
DatabasePager::addDatabaseThread() HANDLE_NON_HTTP
DatabasePager::addDatabaseThread() HANDLE_ONLY_HTTP
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
View::setSceneData() Reusing exisitng scene0x5ab9c48
NodeTrackerManipulator::setTrackNode(Node*0x78ef978 ): Path set
setTrackNode()
  MatrixTransform ''
NodeTrackerManipulator::setTrackNode(Node*0x85bdfa0 ): Path set
setTrackNode()
  MatrixTransform ''
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
Viewer::startThreading() - starting threading
Viewer::startThreading() - contexts.size()=1
Making scene thread safe
ViewerBase::startThreading() : Realizng window 0x7001ff0
Doing add
Doing add
Doing add
Doing add
Doing add
  gc->getGraphicsThread()->startThread() 0x700abe0
Set up threading
Doing run 0x700abe0 isRunning()=1
itr='c:\repository\software\3DVisualization\3DVisualization-devel\bin'
FindFileInPath() : trying 
C:\repository\software\3DVisualization\3DVisualization-devel\bin\osgPlugins-2.8.0\mingw_osgdb_png.dll
 ...
itr='C:\WINDOWS\system32'
FindFileInPath() : trying 
C:\WINDOWS\system32\osgPlugins-2.8.0\mingw_osgdb_png.dll ...
itr='C:\WINDOWS\System'
FindFileInPath() : trying 
C:\WINDOWS\System\osgP

Re: [osg-users] when to use VertexBufferObjects or DisplayLists

[Christian Sam]

thank you all for the performance hints, as far as i have understood you, i 
should try to move from a "geode->addDrawable(createSingleTreeSprite)" to a 
"geode->addDrawable(createMultiTreeGeometry(Cell)).

with that approach also different tree geometry (width/size) should be possible 
without performance loss? (although using different textures in one cell is 
impossible)

greetings,
christian

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9782#9782





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] animating textures

[Cedric Pinson]

Hi Roland,

Yes the idea should be to use a UpdateCallback that use easemotion. I 
will try to make new example with ease motion and stateset callback.
I guess that to fade in/out texture, the osgFX::MultitextureControl is 
ready to use.


Cheers,
Cedric

Roland Smeenk wrote:

jguo wrote:
  

I had a look in osgAnimation 
(http://www.openscenegraph.org/documentation/OpenSceneGraphReferenceDocs/a01556.html)
 but that namespace seems to be just related to geometry.




That's not entirely true. osgAnimation can be used to generate animated values. 
These values will have to be linked to some part of your scenegraph, which is 
done with an AnimationUpdateCallback. The currently available callbacks are 
indeed only for updating geometry, but a custom AnimationUpdateCallback could 
be written to update shader uniforms or the textureWeight in a 
MultiTextureControl.

--
Roland

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9768#9768





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
  


--
+33 (0) 6 63 20 03 56  Cedric Pinson mailto:morni...@plopbyte.net 
http://www.plopbyte.net


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] Windows users - VS2008 comments

[Tanguy Fautre]

Hi Sherman,

VS2008 has been slowly adopted by C++ developers, because when compared
to VS2005 it does not features anything new that is easily visible.

My experience is that if you get the choice between VS2005 and VS2008,
go for VS2008. It's more stable in general (the C++ compiler and the C++
standard library feature more bug fixes).

Also as you said, VS2008 has the /MP option, which I now use all the
time.


Tanguy


-Original Message-
From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of sherman
wilcox
Sent: Saturday 04 April 2009 17:30
To: OpenSceneGraph Users
Subject: [osg-users] Windows users - VS2008 comments

Slightly off-topic for this mailing list but here goes. For the
Windows users out there, those of you that are using or know of
developers using VS2008 - any comments? Problems, benefits, pros/cons?

My project is using VS2005 and has been for a while now but I'm
considering switching to VS2008. In my case there's only one reason -
the /MP benefit. /MP enables parallel compilation. For multiprocessor
or multicore systems it's a real performance increase. I'm not talking
about parallel builds of multiple projects but compiling multiple
modules in the same project(s) in parallel. On VS2005 /MP is
undocumented but works, mostly. It's a tremendous performance boost.
However, I've ran into a PDB corruption issue here as of late. Error
C2471. Not sure why it's suddenly happening as I've been using /MP for
a long while now. I believe Gordon warned me this could happen and it
has. I phoned Microsoft about this and they said that /MP is
unsupported on VS2005 and I have no recourse but to either not use
that feature (no way - too fast) or upgrade to VS2008. So, I may be
upgrading. However, there are numerous developers on my project and I
want to have a clear understanding of any potential issues before
moving forward. I'll be purchasing a copy to test but I'd like to hear
from the list.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.or
g
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] OpenFlight-Plugin - How to store individual object in seperate geode?

[Bulkhead]

Hi Brede,

Just some updates. I have glanced through the codes for the openflight plugin, 
and not sure if this is correct.

Using the "preserveObject" option, the openflight Object record will be stored 
in a osg::Group node. All the faces will be stored as osg::Geode under the 
respective group node. Is this correct?

By iterating through the groups, and collecting the geodes in each group, I 
managed to retrieve the land from my openflight terrain file. However, the 
buildings are still being grouped based on the texture (i.e. building with same 
texture are stored under the same group node). Is this how the openflight 
structure organize the record? Is it feasible to seperate these buildings?

Thanks in advanced.

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9778#9778





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

[osg-users] Skybox example problem

[Martin Großer]

Hello all,

I compiled the skybox example. And I try to load a big scene. Now I have 
artefacts in the skybox. I can see the environment sphere. I try a smaller 
scene and zoom out. I get the same result. When I change the following line 
from ...

clearNode->setRequiresClear(false);

... in ...

clearNode->setRequiresClear(true);

... then the environment sphere go missing and I see the clear color.

Is the problem the distance of the far clipping plane? And I think the 
environment sphere is translate with the eye position?


Cheers, Martin
-- 
Pt! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: 
http://www.gmx.net/de/go/multimessenger01
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] OpenFlight-Plugin - How to store individual object in seperate geode?

[Bulkhead]

Hi Brede,

Glad to hear from you again  :D 

I've written a geodeVisitor class to return all the loaded geodes. But before 
that, I would like to clarify something with you.

For experiment purposes, I've disabled totally the optimization (i.e. setting 
preserveObject and preserveFace) and commented out the optimizer block in my 
loader codes. Without any optimization at all, I noticed that each geode loaded 
using osgDB::readNodeFile() contains only 1 drawable.

Is my observation correct: Each geode loaded from openflight plugin contains 
only 1 drawable (for my case, it's a 2D triangle), and these geodes are merged 
to form an "object" during optimization (e.g. using MERGE_GEOMETRY)?

Or is this due to the openflight format? Or could it be the openflight file 
that I am using?

Please enlighten. I am pretty lost now, as I am new to 3D rendering and with 
very little knowledge about openflight format.


Thanks  [Wink]

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9775#9775





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [osgPlugins] OpenFlight-Plugin - How to store individual object in seperate geode?

[Brede Johansen]

Hi Bulkhead,

I don't know about a reference manual but you have the source.  The
Optimizer header file is a good reference.
Yes, you typically create a custom visitor and set the permitted
optimization for the objects (node,stateset, drawable etc.) in question.
For the option parameter I would probably start with 0 before fine tuning.


To Paul:
Next time ;-)
I'm trying to keep up with the development of OSG.  You never know when it
may come in handy.


Regards
Brede


On Mon, Apr 6, 2009 at 6:04 AM, Bulkhead  wrote:

> Hi Brede,
>
> Thanks for the reply.
>
> I did a search on the setPermissibleOptimizationsForObject() fucntion and
> read your posting on osg-users mail-archive, and yup, I think very high
> chance that this will solve my problem  [Wink]
>
> But unfortunately, I couldn't find any reference on how to use this
> function. The first parameter is an osg::Object pointer. Should I pass in
> every geode (using a geodeVisitor)? As for the second parameter, it will be
> all the optimization options, except MERGE_GEOMETRY?
>
> Also please advise if there's any reference manual that I can refer to? The
> one on osg wiki page doesn't give much info.
>
> Thanks in advanced.
>
> --
> Read this topic online here:
> http://forum.openscenegraph.org/viewtopic.php?p=9764#9764
>
>
>
>
>
> ___
> 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] animating textures

[Robert Osfield]

Hi Julia,

osgFX::MultiTextureControl would be able to do what you want.  It uses
register combiner extensions to do the blending, but you could use your own
fragment shader to control the blending precisely.

The other way to do it is use a 3D texture and then vary the r coord.  See
the osgtexture3D example for an example of this.

Robert.

On Mon, Apr 6, 2009 at 7:46 AM, Julia Guo  wrote:

> ah - I should have been more specific with what I meant by animation.
> I want to fade between two textures so that the previous texture slowly
> gets weaker while the new texture slowly gets stronger.
>
> Can shaders manipulate the texture opacity? (I expect so.)
> And do you think this is a better approach than what OSG provides with the
> MultiTextureControl?
>
> Julia
>
> --
> Read this topic online here:
> http://forum.openscenegraph.org/viewtopic.php?p=9766#9766
>
>
>
>
>
> ___
> 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] osgmultitexturecontrol example data base?

[Robert Osfield]

On Mon, Apr 6, 2009 at 1:41 AM, Julia Guo  wrote:

> Im also trying to run the example.
> Is the example data ready?
>

I haven't personally put at an publicly available example database together
yet.

Robert.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] VertexArray from a loaded .osg file.

[Robert Osfield]

On Mon, Apr 6, 2009 at 12:18 AM, Christian Sam  wrote:

> is it possible to use the VertexArray from the loaded scenefile in a new
> created Drawable (which will display my analyzing results), AND add
> additional vertices, based on the analyzing process to it?
> Or will i have to allocate a new VertexArray for my Drawable and copy the
> loaded file's VertexArray by Value into it?
>

Yes, ref counting means you can share objects such as osg::Vec3Arrays.  Note
that OSG itself has a .shp plugin that loads the shape files as Vec3dArray
which while very slow for rendering are very good to accuracy.

Robert.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] when to use VertexBufferObjects or DisplayLists

[Robert Osfield]

Hi Cristian,

For best performance clumping geometry together into a smaller set of large
tree clusters would be best.  As Roland says the osgforest example does
flesh out all the different approach that one can use, and misses some that
provided the best performance such as clustering lots of trees together into
the same osg::Geometry object.

Robert.

On Sun, Apr 5, 2009 at 6:13 PM, Christian Sam  wrote:

> hi,
>
> refering to osgforest as example, is a method like createTransformGraph,
> where the different looking trees are based on the same (scaled) tree model
> better from a performance point of view? i mean better than the two other
> fixed pipeline methods in this example
>
> thanks in advance,
> christian
>
> --
> Read this topic online here:
> http://forum.openscenegraph.org/viewtopic.php?p=9760#9760
>
>
>
>
>
> ___
> 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] when to use VertexBufferObjects or DisplayLists

[Roland Smeenk]

> VBO's are better for dynamic geometry, but do have a higher CPU overhead than 
> display lists. 


Remember that display lists are deprecated in OpenGL 3.0 and moved into an 
extension in OpenGL 3.1. The advised path is using VBO's...


> refering to osgforest as example ... 


I am not too happy with osgforest as an example, because it tries too keep the 
geometry sorted back to front for solving transparency issues. This will result 
in many drawables which is terrible for rendering performance. Note that this 
example is NOT representing how games are typically rendering forests. 

--
Roland

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9769#9769





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] animating textures

[Roland Smeenk]

jguo wrote:
> 
> I had a look in osgAnimation 
> (http://www.openscenegraph.org/documentation/OpenSceneGraphReferenceDocs/a01556.html)
>  but that namespace seems to be just related to geometry.


That's not entirely true. osgAnimation can be used to generate animated values. 
These values will have to be linked to some part of your scenegraph, which is 
done with an AnimationUpdateCallback. The currently available callbacks are 
indeed only for updating geometry, but a custom AnimationUpdateCallback could 
be written to update shader uniforms or the textureWeight in a 
MultiTextureControl.

--
Roland

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9768#9768





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

Re: [osg-users] [build] Unreckognized Commands

[Roland Smeenk]

arwise wrote:
> I recently downloaded Openscenegraph 2.9.2. But there seemed to be an issue 
> with the Collada-dom 2.2:
> 
> ..\..\..\..\collada-dom\dom\build\vc9-1.5\libcollada15dom21.dll : fatal 
> error LNK1107: invalid or corrupt file: cannot read at 0x2F8
> 
> It doesn't seem to like the dynamic library for some reason. Any suggestions?


The osg collada plugin currently only works with Collada 1.4 and not 1.5.

See 
http://www.openscenegraph.org/projects/osg/wiki/Support/KnowledgeBase/Collada

--
Roland

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=9767#9767





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org