Re: [osg-users] 3DS loader broken for models that use pivot points?
I, Just an idea but if you want to have a pivot point, add a Helper at the pivot point of your model, and put it as parent of all the model... It is not exactly what you mean, but it can do what you want. Regards, Vincent. 2008/8/6 [EMAIL PROTECTED] I have some 3DS models that I am trying to load that seem to be getting a bad matrix transformation that end up flattening the geometry. According to a comment in the source, the 'new' method of converting 3DS files should be better, but have problems when pivot points are present in the model. I have attached an example file that exhibits this problem. Has anyone else seen this? I am using 2.4.0 Thanks Chuck Seberino ___ 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] moving models...
Hi Shayne, Tueller, Shayne R Civ USAF AFMC 519 SMXS/MXDEC wrote: Ok, I updated the transform node as given below for position changes and this seems to work. However, when I add an orientation update (i.e. multiply the translation by a rotation) my model disappears completely. Any idea on what's going on with that? Two things: - check the order of Matrix multiplicationsi. Are you maybe rotating around global instead of local? The way you describe your mults indicates it may be the former. - is the node you're rotating modelled in the origin? Cheers, /ulrich ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] using shaders to find fthe maximum value in a
tx for u response Art ,
if i understand u correctly my code will be something like this ,i am using
glew .
(suppose my input texture where i want to find my maximum value is
(width,height) and i have a FrameBuffer class and a GL_Texture class, and a
shader.)
inputtexture=new GL_Texture(...width,height...)
n=2;
fbo=new FrameBuffer(...)
loop
{
outputexture=new GL_Texture(..width/n,height/n)
//activate fbo to render to texture
fbo-Bind();
//attach a outputexturetexture to render to it
fbo-AttachTexture(outputexture);
//setup viewport to size of outputetexture
glViewport(0, 0,width/n,height/n);
glMatrixMode (GL_PROJECTION);
glLoadIdentity ();
glOrtho (0, (GLdouble) ,width/n, 0,height/n, -30, 30);
glClearColor(0,0,0,0);
glClear (GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
glMatrixMode(GL_MODELVIEW);
glLoadIdentity();
//activate inputtexture and run shader
inputtexture-Activate(GL_TEXTURE0);
shader-begin();
//pass inputtexture to shader
shader-setUniform1i(texture0,0);
//Here i render a Rectangle at width/n and height/n
//disable shader and fbo
shader-end();
fbo-Disable();
//go to next step of reducing inputtexture
n*=2;
//copy outputetexture to inputtexture for next iteration :
inputtexture=outputexture
inputtexture=GetCopy(outputexture);
delete outputexture;
}//end loop
and in the fragment shader all i have to do is : each pixel shader will
calulate a maximum for 4 neighberhood texels of the input texture ?
is it correct ?
thx in advance
Message: 5
Date: Tue, 5 Aug 2008 17:09:45 + (GMT)
From: Art Tevs [EMAIL PROTECTED]
Subject: Re: [osg-users] using shaders to find fthe maximum value in a
texture
To: OpenSceneGraph Users osg-users@lists.openscenegraph.org
Message-ID: [EMAIL PROTECTED]
Content-Type: text/plain; charset=utf-8
Do you need the maximum value over all texture elements?
If yes, then you just have to compute a mipmap with max as combining
operation. In the last level, where only one pixel left, you will get the
maximum value.
It is pretty straight forward to implement such an algorithm.
Either any GPGPU approach can help you, or if you use osg, then take a look
into osgPPU's hdr example.
Best regards,
Art
--- Abdallah Deeb [EMAIL PROTECTED] schrieb am Di, 5.8.2008:
Von: Abdallah Deeb [EMAIL PROTECTED]
Betreff: [osg-users] using shaders to find fthe maximum value in a texture
An: osg-users@lists.openscenegraph.org
Datum: Dienstag, 5. August 2008, 16:56
Hello,
i am new to glsl , i have a probleme ( it seems that it is
fondamental problem in shaders)
in order to calculate the maximum/minumum value in a
texture passed to fragment shader ,
it seems that there is as special architecture to implement
to resolve this probleme .
i passed a days searching on the net to solve this problem
but no one discuss this probleme and how to calculate a
maximum or minimum in a texture or vector norm ... i heard
something called vector reduce in shaders.. but i have no
clear idea about it .
so in resume my question is how can i create a fragment
shader that can calculate (a maximum value for ex) in a
texture passed to it ,
if anyone can help me plz , i really need an
implementation or a reference docs..
thx in advance .
_
Explore the seven wonders of the world
http://search.msn.com/results.aspx?q=7+wonders+worldmkt=en-USform=QBRE___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenSceneGraph-2.6.0 release is available!
Hi Mario, On Wed, Aug 6, 2008 at 4:40 AM, Mario Valle [EMAIL PROTECTED] wrote: Only a minor glitch. www.openscenegraph.com redirects you to www.openscenegraph.org, the main OSG site. There is no way to know what Professional Services offer... Is it intentional? This is a problem rather unrelated to 2.6... A year and half ago the openscenegraph.com website was hastily repurposed to become openscenegraph.org when we lost access to the old server for .org. We now have a new server for .org and it's hosted without further problems, but... I've been so rushed off my feet I haven't had a chance to put up a new website for openscenegraph.com. It's on my todo list to put up a new website for my company, just waiting for a quiet period so that I can sit down plan, put it together etc. Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] Light attenuation do nothing
Hi all, I've a scene with 2 lights : The first one (n°0) is the one for the whole scene, with ambiant component, diffuse, specular, ... The second one is attached on a node, and have to light only this node and not all the scene. To do that, I put some attenuation on the second light to avoid it light the reste of the scene... but it does nothing ! light1-setAmbient(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setDiffuse(osg::Vec4(intensity, 0, 0, 1)); light1-setSpecular(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setPosition(osg::Vec4(p1, 0)); light1-setDirection(-p1); light1-setSpotCutoff(0.5f); light1-setConstantAttenuation(0.95f); light1-setLinearAttenuation(0.001f); light1-setQuadraticAttenuation(0.0002f); osg::LightSource* satLightSource1 = new osg::LightSource; satLightSource1-setLight(light1.get()); satLightSource1-setLocalStateSetModes( osg::StateAttribute::ON ); satLightSource1-getOrCreateStateSet()-setMode(GL_LIGHTING, osg::StateAttribute::ON); osg::LightModel* lightModel1 = new osg::LightModel; lightModel1-setAmbientIntensity(osg::Vec4(0.0f,0.0f,0.0f,1.0f)); //0,0,0,1 satLightSource1-getOrCreateStateSet()-setAttribute(lightModel1); group-addChild(satLightSource1); _root-getOrCreateStateSet()-setMode(GL_LIGHT1, osg::StateAttribute::ON); When I modify quadratic attenuation and linear attenuation, nothing appear to be changed in the scene, and a lot of nodes are lighted in red... Some idea why it doesn't works ? Thanks, Regards, Vincent. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Light attenuation do nothing
Hi Vincent, You'll need to set the light number of the light via light-setLightNum(num); The first light should be 0, and your second one 1. Robert. On Wed, Aug 6, 2008 at 10:53 AM, Vincent Bourdier [EMAIL PROTECTED] wrote: Hi all, I've a scene with 2 lights : The first one (n°0) is the one for the whole scene, with ambiant component, diffuse, specular, ... The second one is attached on a node, and have to light only this node and not all the scene. To do that, I put some attenuation on the second light to avoid it light the reste of the scene... but it does nothing ! light1-setAmbient(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setDiffuse(osg::Vec4(intensity, 0, 0, 1)); light1-setSpecular(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setPosition(osg::Vec4(p1, 0)); light1-setDirection(-p1); light1-setSpotCutoff(0.5f); light1-setConstantAttenuation(0.95f); light1-setLinearAttenuation(0.001f); light1-setQuadraticAttenuation(0.0002f); osg::LightSource* satLightSource1 = new osg::LightSource; satLightSource1-setLight(light1.get()); satLightSource1-setLocalStateSetModes( osg::StateAttribute::ON ); satLightSource1-getOrCreateStateSet()-setMode(GL_LIGHTING, osg::StateAttribute::ON); osg::LightModel* lightModel1 = new osg::LightModel; lightModel1-setAmbientIntensity(osg::Vec4(0.0f,0.0f,0.0f,1.0f)); //0,0,0,1 satLightSource1-getOrCreateStateSet()-setAttribute(lightModel1); group-addChild(satLightSource1); _root-getOrCreateStateSet()-setMode(GL_LIGHT1, osg::StateAttribute::ON); When I modify quadratic attenuation and linear attenuation, nothing appear to be changed in the scene, and a lot of nodes are lighted in red... Some idea why it doesn't works ? Thanks, Regards, Vincent. ___ 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] Light attenuation do nothing
Hi Robert, Yes I know, i forgot to paste the code : osg::ref_ptrosg::Light light1 = new osg::Light; light1-setLightNum(1); Sorry Any idea ? because the red light is visible, but no attenuation modify it... and this is the real problem. Thanks, Vincent. 2008/8/6 Robert Osfield [EMAIL PROTECTED] Hi Vincent, You'll need to set the light number of the light via light-setLightNum(num); The first light should be 0, and your second one 1. Robert. On Wed, Aug 6, 2008 at 10:53 AM, Vincent Bourdier [EMAIL PROTECTED] wrote: Hi all, I've a scene with 2 lights : The first one (n°0) is the one for the whole scene, with ambiant component, diffuse, specular, ... The second one is attached on a node, and have to light only this node and not all the scene. To do that, I put some attenuation on the second light to avoid it light the reste of the scene... but it does nothing ! light1-setAmbient(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setDiffuse(osg::Vec4(intensity, 0, 0, 1)); light1-setSpecular(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setPosition(osg::Vec4(p1, 0)); light1-setDirection(-p1); light1-setSpotCutoff(0.5f); light1-setConstantAttenuation(0.95f); light1-setLinearAttenuation(0.001f); light1-setQuadraticAttenuation(0.0002f); osg::LightSource* satLightSource1 = new osg::LightSource; satLightSource1-setLight(light1.get()); satLightSource1-setLocalStateSetModes( osg::StateAttribute::ON ); satLightSource1-getOrCreateStateSet()-setMode(GL_LIGHTING, osg::StateAttribute::ON); osg::LightModel* lightModel1 = new osg::LightModel; lightModel1-setAmbientIntensity(osg::Vec4(0.0f,0.0f,0.0f,1.0f)); //0,0,0,1 satLightSource1-getOrCreateStateSet()-setAttribute(lightModel1); group-addChild(satLightSource1); _root-getOrCreateStateSet()-setMode(GL_LIGHT1, osg::StateAttribute::ON); When I modify quadratic attenuation and linear attenuation, nothing appear to be changed in the scene, and a lot of nodes are lighted in red... Some idea why it doesn't works ? Thanks, Regards, Vincent. ___ 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] vpb build_master.source --radius-equator
Hi J.P, Currently the radius settings are stored in the BuildOptions's EllipsoidModel, it should staight forward to add the serialization of its parameters. I'm working on VPB so I'll have a look at doing this item. Robert. On Tue, Aug 5, 2008 at 3:54 PM, J.P. Delport [EMAIL PROTECTED] wrote: Hi, when setting the values --radius-equator and --radius-polar, the values in the ellipsoidmodel are set in vpb, but they are not written to the build_master.source file. So, when doing a distributed build, these options get discarded when osgdem reads the .source file. I am still trying to get my head around how the parsing of the .source file works. Would a fix be as easy as adding the options to the BuildOptions and BuildOptionsIO files? I need to build a DB with specific radii. Any pointers welcome. rgds jp -- 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 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] vpb build_master.source --radius-equator
Thanks Robert, for now I've just hacked the BuildOptions code to always use my ellipsoid, but will use and test the general fix when you are done. rgds jp Robert Osfield wrote: Hi J.P, Currently the radius settings are stored in the BuildOptions's EllipsoidModel, it should staight forward to add the serialization of its parameters. I'm working on VPB so I'll have a look at doing this item. Robert. On Tue, Aug 5, 2008 at 3:54 PM, J.P. Delport [EMAIL PROTECTED] wrote: Hi, when setting the values --radius-equator and --radius-polar, the values in the ellipsoidmodel are set in vpb, but they are not written to the build_master.source file. So, when doing a distributed build, these options get discarded when osgdem reads the .source file. I am still trying to get my head around how the parsing of the .source file works. Would a fix be as easy as adding the options to the BuildOptions and BuildOptionsIO files? I need to build a DB with specific radii. Any pointers welcome. rgds jp -- 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] OpenSceneGraph-2.6.0 release is available!
Hello all, Congrats to all, and thanks to everyone who helped with submissions, testing, suggestions, discussion and comments. Great release! J-S -- __ Jean-Sebastien Guay[EMAIL PROTECTED] 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
[osg-users] About sort geometry
Hi all, I'm now working on a volume rendering project. What I need is to improve performance. I had to split one volume bounding box into numbers of boxes. The very important work to do is to sort the geometry. Boxes must be rendered from nearest position to farest to get right visual. So, does OSG has some functions to perform this task? Can the setRenderingHint(osg::StateSet::TRANSPARENT_BIN) really help me? Thanks! Hesicong ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] About sort geometry
Hi, Robert, Is there a way to visually see the render order of geometry or the console output of current rendering geometry? The draw callback runs once and I can't use it to determine the render order. Can you give me some advice? Robert Osfield wrote: On Wed, Aug 6, 2008 at 3:09 PM, hesicong2006 [EMAIL PROTECTED] wrote: Hi, robert, I just noticed this line: myvolumestateset-setRenderBinDetails(9,"FrontToBack"); Why you set details to 9 instead of other values? It's just the bin number, 9 will draw before the standard transparent bin that defaults to 10, 0 is the default opaque bin. Choose what ever value you want to control the high level ordering of bins. Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org __ Information from ESET NOD32 Antivirus, version of virus signature database 3329 (20080805) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] About sort geometry
Hi, On Wed, Aug 6, 2008 at 4:48 PM, Robert Osfield [EMAIL PROTECTED] wrote: On Wed, Aug 6, 2008 at 3:40 PM, hesicong2006 [EMAIL PROTECTED] wrote: Hi, Robert, Is there a way to visually see the render order of geometry or the console output of current rendering geometry? Give gDEBugger a try (there is a free trial of 30 days). You can break during rendering and step through OpenGL calls just like you can step in your program with a debugger. You will then be able to see the primitives rendered one by one. Regards Thibault No. The draw callback runs once and I can't use it to determine the render order. Can you give me some advice? Not when I'm this busy. ___ 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] About sort geometry
Thanks Thibault, I use gDebugger and it should be a way to do so, but it's quite boring to see the OpenGL commands. I must find a way to see the rendering detail in order to debug my program. If there's really no simple way to do this, it will be a pity. Thibault Genessay wrote: Hi, On Wed, Aug 6, 2008 at 4:48 PM, Robert Osfield [EMAIL PROTECTED] wrote: On Wed, Aug 6, 2008 at 3:40 PM, hesicong2006 [EMAIL PROTECTED] wrote: Hi, Robert, Is there a way to visually see the render order of geometry or the console output of current rendering geometry? Give gDEBugger a try (there is a free trial of 30 days). You can break during rendering and step through OpenGL calls just like you can step in your program with a debugger. You will then be able to see the primitives rendered one by one. Regards Thibault No. The draw callback runs once and I can't use it to determine the render order. Can you give me some advice? Not when I'm this busy. ___ 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 __ Information from ESET NOD32 Antivirus, version of virus signature database 3329 (20080805) __ The message was checked by ESET NOD32 Antivirus. http://www.eset.com ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Light attenuation do nothing
Hi, no answers... So i've some questions to help me : * Does light direction and spotcuttoff incompatibles with attenuations ? * how to compute good attenuation values ? * why all the scene seems to be lighted by a directional spotlight, including node behind spot, in the good direction ? * how to light a specified node, without (if possible) lightning the other one ? Thanks, Regards, Vincent. 2008/8/6 Vincent Bourdier [EMAIL PROTECTED] Hi Robert, Yes I know, i forgot to paste the code : osg::ref_ptrosg::Light light1 = new osg::Light; light1-setLightNum(1); Sorry Any idea ? because the red light is visible, but no attenuation modify it... and this is the real problem. Thanks, Vincent. 2008/8/6 Robert Osfield [EMAIL PROTECTED] Hi Vincent, You'll need to set the light number of the light via light-setLightNum(num); The first light should be 0, and your second one 1. Robert. On Wed, Aug 6, 2008 at 10:53 AM, Vincent Bourdier [EMAIL PROTECTED] wrote: Hi all, I've a scene with 2 lights : The first one (n°0) is the one for the whole scene, with ambiant component, diffuse, specular, ... The second one is attached on a node, and have to light only this node and not all the scene. To do that, I put some attenuation on the second light to avoid it light the reste of the scene... but it does nothing ! light1-setAmbient(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setDiffuse(osg::Vec4(intensity, 0, 0, 1)); light1-setSpecular(osg::Vec4(0.0f, 0.0f, 0.0f, 1)); light1-setPosition(osg::Vec4(p1, 0)); light1-setDirection(-p1); light1-setSpotCutoff(0.5f); light1-setConstantAttenuation(0.95f); light1-setLinearAttenuation(0.001f); light1-setQuadraticAttenuation(0.0002f); osg::LightSource* satLightSource1 = new osg::LightSource; satLightSource1-setLight(light1.get()); satLightSource1-setLocalStateSetModes( osg::StateAttribute::ON ); satLightSource1-getOrCreateStateSet()-setMode(GL_LIGHTING, osg::StateAttribute::ON); osg::LightModel* lightModel1 = new osg::LightModel; lightModel1-setAmbientIntensity(osg::Vec4(0.0f,0.0f,0.0f,1.0f)); //0,0,0,1 satLightSource1-getOrCreateStateSet()-setAttribute(lightModel1); group-addChild(satLightSource1); _root-getOrCreateStateSet()-setMode(GL_LIGHT1, osg::StateAttribute::ON); When I modify quadratic attenuation and linear attenuation, nothing appear to be changed in the scene, and a lot of nodes are lighted in red... Some idea why it doesn't works ? Thanks, Regards, Vincent. ___ 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] Polygon Mode = Wire and some Depth culling?
Hello, the Cartoon renderer created cel shading, that was not exactly what I wanted since I cannot controll the color of the shaed of the model itself. Now I am trying to get a grip onto oscscribe, I belive that using that technique and rendering the model as background color without shading may achieve the result I need. Thanks :) -- From: Robert Osfield [EMAIL PROTECTED] Sent: Wednesday, August 06, 2008 11:22 AM To: OpenSceneGraph Users osg-users@lists.openscenegraph.org Subject: Re: [osg-users] Polygon Mode = Wire and some Depth culling? Hi Sam, Hidden line removal is typically using a multi-pass technique, where you render the subgraph twice, once with FILL PolygonMode, depth write on, but colour write off, then a second time with LINE PolygonMode and PolygonOffset to move it forward and depth write and colour turned back on (although you could get awy with depth write off in this pass). The closest example of this would be the osgscribe example as it plays with two passes and modifying the PolygonMode. Robert. On Wed, Aug 6, 2008 at 10:17 AM, Sam [EMAIL PROTECTED] wrote: Hello, I think hidden line removal could be it (was missing this term), but I am actually short on time so I only can check that later. An example of what I mean: If I see a wired cube I usually see the lines that are at the cube's back. Quick google link - http://www.cs.bham.ac.uk/~axs/fig/cube-only.png What I want to render, is only the left, front and top side of the cube but still remain in wired mode (like as the cube would be filled white). So I think yes, this seems to be hidden line removal (correct me if I am wrong) Thanks for the term so far :) Sam -- From: Robert Osfield [EMAIL PROTECTED] Sent: Wednesday, August 06, 2008 10:51 AM To: OpenSceneGraph Users osg-users@lists.openscenegraph.org Subject: Re: [osg-users] Polygon Mode = Wire and some Depth culling? Hi Sam, I can't work out form your email the exact effect you are after. Is that you want wireframe with hidden line removal? Robert. On Wed, Aug 6, 2008 at 9:27 AM, Sam [EMAIL PROTECTED] wrote: Hello, I am drawing a model as a wire-model. So I see only lines. Now I am looking for a way to cull all the lines in this case which would be otherwise invisible because of the depth check. I am not talking about backface culing. The aim is here, to see only all edges of a model, that would be visible if the polygon mode would be not in wire mode and all edges that would be filtered by depth check, even if they are frontfaces should not be drawn. Is this possible? GL_DEPTH_TEST is enabled and I checked osg::Depth but haven't found anything helpful yet. Thanks Sam ___ 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 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Light attenuation do nothing
On Wed, Aug 6, 2008 at 5:09 PM, Vincent Bourdier [EMAIL PROTECTED] wrote: Hi, no answers... So i've some questions to help me : * Does light direction and spotcuttoff incompatibles with attenuations ? * how to compute good attenuation values ? * why all the scene seems to be lighted by a directional spotlight, including node behind spot, in the good direction ? * how to light a specified node, without (if possible) lightning the other one ? OSG lighting is just standard OpenGL lighting, so pick up an book with a chapter on OpenGL lighting you'll understand what the OSG does. Robert. ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] OSG 2.6: No OpenGL Context?
In upgrading my app today to OSG 2.6, I couldn't help but noticed that it
crashed when I attempted to retrieve the OpenGL Vendor name. It wasn't a great
deal of work to determine that there was no valid OpenGL context when it was
called. A little further work showed that where I called
this-_view-setThreadingModel(osgViewer::Viewer::SingleThreaded); seemed to
cause the problem. When I commented this out, i had a valid OpenGL Context,
but OSG crashed later the first time it went to render the scene. The app
works fine with OSG 2.4, so I assume I must have just been doing something
wrong this whole time and it's finally catching up with me! I have tried to
cut the class back as much as possible for an example (below). The app creates
an instance of this class, then calls graphicsInit(), then, finally,
graphicsLoop(). I have a feeling the error lies in the constructor itself, but
I don't know what I'm doing wrong.
(Just as a bit of background, this class is supposed to support Linux and
Windows. Under Windows, it supports rendering to its own OSG-created window or
rendering to an OSG window inside a windows control.)
Any ideas would be greatly appreciated!
Thanks!
OSGExample.cpp
#include iostream
#include osg/Group
#include osg/Geode
#include osg/Shader
#include osg/MatrixTransform
#include osg/Projection
#include osg/DisplaySettings
#include osg/Texture2D
#include osg/Drawable
#include osg/CopyOp
#include osg/TexGen
#include osg/Matrix
#include osg/FrontFace
#include osg/StateSet
#include osg/Notify
#include osgDB/ReadFile
#include osgDB/FileUtils
#include osgUtil/Optimizer
#include osgViewer/View
#include osgViewer/Viewer
#include osgViewer/CompositeViewer
#include osgViewer/ViewerEventHandlers
#ifdef WIN32
#include osgViewer/api/Win32/GraphicsWindowWin32
#endif
using std::cerr;
class OSG
{
public:
#ifdef WIN32
OSG(unsigned int multisamples = 1, HWND hostWindowHWND
= NULL);
#else
OSG(unsigned int multisamples = 1);
#endif
virtual ~OSG();
virtual void graphicsInit();
virtual void graphicsLoop();
private:
std::string _vendor;
osgGA::MatrixManipulator* _viewerMatrix;
osg::State* _state;
osg::ref_ptrosg::Group _sceneGroup;
osg::ref_ptrosg::GraphicsContext _graphicsContext;
osg::ref_ptrosg::GraphicsContext::Traits _traits;
osg::ref_ptrosg::DisplaySettings _displaySettings;
osg::ref_ptrosg::GraphicsContext::WindowingSystemInterface
_wsi;
osgViewer::Viewer* _view;
osgViewer::Viewer::Windows _viewWindows;
// EventHandler* _eventHandler;
osg::ref_ptrosg::Node _hiveNode;
unsigned int _uniqueGraphicsStateID;
unsigned int _multisamples;
std::string _windowTitle;
#ifdef WIN32
HWND hostHWND;
HWND _hwnd;
HDC _hdc;
HGLRC _wglContext;
#endif
};
#ifdef WIN32
OSG::OSG(unsigned int multisamples, HWND hostWindowHWND)
#else
OSG::OSG(unsigned int multisamples)
#endif
{
osg::setNotifyLevel(osg::ALWAYS);
this-_view = NULL;
this-_uniqueGraphicsStateID = 0;
this-_multisamples = 0;
// this-_eventHandler = NULL;
this-_viewerMatrix = NULL;
this-_state = NULL;
this-_vendor = (null);
#ifdef WIN32
this-hostHWND = NULL;
this-_hwnd = NULL;
this-_hdc = NULL;
this-_wglContext = NULL;
#endif
this-_multisamples = multisamples;
// Create an event handler
// this-_eventHandler = new EventHandler();
// Setup the Windowing System Interface.
this-_wsi = osg::GraphicsContext::getWindowingSystemInterface();
// Setup the graphics context traits.
{
this-_traits = new osg::GraphicsContext::Traits;
this-_traits-doubleBuffer = true;
this-_traits-x = 0;
this-_traits-y = 0;
this-_traits-width = 256;
this-_traits-height = 256;
this-_traits-supportsResize = true;
this-_traits-windowDecoration = true;
this-_traits-windowName = myOSG;
// Only used if we inherit from a more complex Windows Control
#ifdef WIN32
if(hostWindowHWND != NULL)
{
this-hostHWND = hostWindowHWND;
// Get the current window size
RECT rect;
[osg-users] problem with addEventHandler in the osgviewerWX example
I have the osgviewerWX example running with OSG v2.2 and WxWidgets v2.8.4 and everything seems to be working correctly. But I did notice that the Stats display does not come up, although in the example the StatsHandler was added to the viewer EventHandler. viewer-addEventHandler(new osgViewer::StatsHandler); This makes sense because I would assume that all events are now being handled by Wx EventHandlers. My question: Is their a way to get the StatsHandler and other similar Handlers to work when running with WxWidgets? Thank you ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OSG 2.6: No OpenGL Context?
Hi bbblurururughghghghghhbbb? Sorry no signature on you email so this
guess was as good as any.
In OSG 2.6 the viewer.frame() by default releases the graphics context
rather than keeping it current, this change allows the viewer to be
interleaved with other viewers within a single frame loop. You can
turn off this release of the graphics context via the
ViewerBase::setReleaseContextAtEndOfFrameHint method:
/** Hint to tell the renderingTraversals() method whether to
call relaseContext() on the last
* context that was made current by the thread calling
renderingTraverals(). Note, when
* running multi-threaded viewer no threads will be made
current or release current.
* Setting this hint to false can enable the frame loop to be
lazy about calling makeCurrent
* and releaseContext on each new frame, helping performance.
However, if you frame loop
* is managing multiple graphics context all from the main
frame thread then this hint must
* be left on, otherwise the wrong context could be left
active, introducing errors in rendering.*/
void setReleaseContextAtEndOfFrameHint(bool hint) {
_releaseContextAtEndOfFrameHint = hint; }
However, I'd strongly recommend that you don't use this, and instead
just fix your application so you aren't doing OpenGL calls in the
frame loop as such usage is makes a really bad assumption that that
thread will have the graphics context current. Why is so bad?
It's bad because the OSG allows you to run with various graphics
window setups and various threading models, almost all of which don't
ever a single graphis context that is current in the main thread,
amoung all this various ways of using the viewers one one case can
ever have the context current in the main thread and that's single
context, single threaded, but even slightly reconfigure the viewer and
BANG your applications breaks.
The right way to add OpenGL calls to the frame loop is the use a
Camera pre and post draw callback. This applies to all the OSG 2.x
series.
Robert.
On Wed, Aug 6, 2008 at 7:00 PM, [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
In upgrading my app today to OSG 2.6, I couldn't help but noticed that it
crashed when I attempted to retrieve the OpenGL Vendor name. It wasn't a
great deal of work to determine that there was no valid OpenGL context when
it was called. A little further work showed that where I called
this-_view-setThreadingModel(osgViewer::Viewer::SingleThreaded); seemed
to cause the problem. When I commented this out, i had a valid OpenGL
Context, but OSG crashed later the first time it went to render the scene.
The app works fine with OSG 2.4, so I assume I must have just been doing
something wrong this whole time and it's finally catching up with me! I
have tried to cut the class back as much as possible for an example
(below). The app creates an instance of this class, then calls
graphicsInit(), then, finally, graphicsLoop(). I have a feeling the error
lies in the constructor itself, but I don't know what I'm doing wrong.
(Just as a bit of background, this class is supposed to support Linux and
Windows. Under Windows, it supports rendering to its own OSG-created window
or rendering to an OSG window inside a windows control.)
Any ideas would be greatly appreciated!
Thanks!
OSGExample.cpp
#include iostream
#include osg/Group
#include osg/Geode
#include osg/Shader
#include osg/MatrixTransform
#include osg/Projection
#include osg/DisplaySettings
#include osg/Texture2D
#include osg/Drawable
#include osg/CopyOp
#include osg/TexGen
#include osg/Matrix
#include osg/FrontFace
#include osg/StateSet
#include osg/Notify
#include osgDB/ReadFile
#include osgDB/FileUtils
#include osgUtil/Optimizer
#include osgViewer/View
#include osgViewer/Viewer
#include osgViewer/CompositeViewer
#include osgViewer/ViewerEventHandlers
#ifdef WIN32
#include osgViewer/api/Win32/GraphicsWindowWin32
#endif
using std::cerr;
class OSG
{
public:
#ifdef WIN32
OSG(unsigned int multisamples = 1, HWND
hostWindowHWND = NULL);
#else
OSG(unsigned int multisamples = 1);
#endif
virtual ~OSG();
virtual void graphicsInit();
virtual void graphicsLoop();
private:
std::string _vendor;
osgGA::MatrixManipulator* _viewerMatrix;
osg::State* _state;
osg::ref_ptrosg::Group _sceneGroup;
osg::ref_ptrosg::GraphicsContext _graphicsContext;
osg::ref_ptrosg::GraphicsContext::Traits _traits;
osg::ref_ptrosg::DisplaySettings _displaySettings;
osg::ref_ptrosg::GraphicsContext::WindowingSystemInterface
_wsi;
osgViewer::Viewer* _view;
Re: [osg-users] OSG 2.6: No OpenGL Context?
Hi
i am not sure where the bug is coming from, but we had in the past (one
week) two threads on this topic. one was the problem of having a valid
OpenGL context for GL_Vendor,... request and the other was on Grapics
contex.
Is there a need to support such information in the osg core? if yes, we can
add a draw callback to retrieve the information at the end of the first
drawing pass, if this would be too late, because the application
needs such information at the really beginning may we have to retrieve this
information at the begining.
what is the community thining or the needs?
may you try to retrieve your vendor name in a draw callback. But your code
is not at all openscenegraph standard :-)
adrian
2008/8/6 [EMAIL PROTECTED] [EMAIL PROTECTED]
In upgrading my app today to OSG 2.6, I couldn't help but noticed that it
crashed when I attempted to retrieve the OpenGL Vendor name. It wasn't a
great deal of work to determine that there was no valid OpenGL context when
it was called. A little further work showed that where I called
this-_view-setThreadingModel(osgViewer::Viewer::SingleThreaded); seemed
to cause the problem. When I commented this out, i had a valid OpenGL
Context, but OSG crashed later the first time it went to render the scene.
The app works fine with OSG 2.4, so I assume I must have just been doing
something wrong this whole time and it's finally catching up with me! I
have tried to cut the class back as much as possible for an example
(below). The app creates an instance of this class, then calls
graphicsInit(), then, finally, graphicsLoop(). I have a feeling the error
lies in the constructor itself, but I don't know what I'm doing wrong.
(Just as a bit of background, this class is supposed to support Linux and
Windows. Under Windows, it supports rendering to its own OSG-created window
or rendering to an OSG window inside a windows control.)
Any ideas would be greatly appreciated!
Thanks!
OSGExample.cpp
#include iostream
#include osg/Group
#include osg/Geode
#include osg/Shader
#include osg/MatrixTransform
#include osg/Projection
#include osg/DisplaySettings
#include osg/Texture2D
#include osg/Drawable
#include osg/CopyOp
#include osg/TexGen
#include osg/Matrix
#include osg/FrontFace
#include osg/StateSet
#include osg/Notify
#include osgDB/ReadFile
#include osgDB/FileUtils
#include osgUtil/Optimizer
#include osgViewer/View
#include osgViewer/Viewer
#include osgViewer/CompositeViewer
#include osgViewer/ViewerEventHandlers
#ifdef WIN32
#include osgViewer/api/Win32/GraphicsWindowWin32
#endif
using std::cerr;
class OSG
{
public:
#ifdef WIN32
OSG(unsigned int multisamples = 1, HWND
hostWindowHWND = NULL);
#else
OSG(unsigned int multisamples = 1);
#endif
virtual ~OSG();
virtual void graphicsInit();
virtual void graphicsLoop();
private:
std::string _vendor;
osgGA::MatrixManipulator* _viewerMatrix;
osg::State* _state;
osg::ref_ptrosg::Group _sceneGroup;
osg::ref_ptrosg::GraphicsContext _graphicsContext;
osg::ref_ptrosg::GraphicsContext::Traits _traits;
osg::ref_ptrosg::DisplaySettings _displaySettings;
osg::ref_ptrosg::GraphicsContext::WindowingSystemInterface _wsi;
osgViewer::Viewer* _view;
osgViewer::Viewer::Windows _viewWindows;
// EventHandler* _eventHandler;
osg::ref_ptrosg::Node _hiveNode;
unsigned int _uniqueGraphicsStateID;
unsigned int _multisamples;
std::string _windowTitle;
#ifdef WIN32
HWND hostHWND;
HWND _hwnd;
HDC _hdc;
HGLRC _wglContext;
#endif
};
#ifdef WIN32
OSG::OSG(unsigned int multisamples, HWND hostWindowHWND)
#else
OSG::OSG(unsigned int multisamples)
#endif
{
osg::setNotifyLevel(osg::ALWAYS);
this-_view = NULL;
this-_uniqueGraphicsStateID = 0;
this-_multisamples = 0;
// this-_eventHandler = NULL;
this-_viewerMatrix = NULL;
this-_state = NULL;
this-_vendor = (null);
#ifdef WIN32
this-hostHWND = NULL;
this-_hwnd = NULL;
this-_hdc = NULL;
this-_wglContext = NULL;
#endif
this-_multisamples = multisamples;
// Create an event handler
// this-_eventHandler = new EventHandler();
// Setup the Windowing System Interface.
this-_wsi = osg::GraphicsContext::getWindowingSystemInterface();
// Setup the graphics context
Re: [osg-users] problem with addEventHandler in the osgviewerWX example
Please upgrade to 2.6, the osgviewerWX example now has stats working just fine in the viewer. On Wed, Aug 6, 2008 at 7:16 PM, Cysneros, Nelson SPAWAR [EMAIL PROTECTED] wrote: I have the osgviewerWX example running with OSG v2.2 and WxWidgets v2.8.4 and everything seems to be working correctly. But I did notice that the Stats display does not come up, although in the example the StatsHandler was added to the viewer EventHandler. viewer-addEventHandler(new osgViewer::StatsHandler); This makes sense because I would assume that all events are now being handled by Wx EventHandlers. My question: Is their a way to get the StatsHandler and other similar Handlers to work when running with WxWidgets? Thank you ___ 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 2.6: No OpenGL Context?
HI Adrian
I know for our apps we need to do some investigation of what OGL features we
have or don't have , and were doing this in the first draw, which is not
ideal as in many cases asking Opengl if it supports feature A is problematic
as drivers lie and say they say they support feature A but actually only
support it through a slow path or even s/w rendering ..
What we want to do at some point( when resources permit) is to have a
generic lib/app that say creates a config file that lists all the OGL
features, and confirm them with a suite of stress tests to verify if they
are supported and give an indication of the machines ability to cope etc.
then consume config at start up adjust the feature we can use etc.. at some
point in the future
_
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Adrian Egli
OpenSceneGraph (3D)
Sent: Wednesday, August 06, 2008 2:39 PM
To: OpenSceneGraph Users
Subject: Re: [osg-users] OSG 2.6: No OpenGL Context?
Hi
i am not sure where the bug is coming from, but we had in the past (one
week) two threads on this topic. one was the problem of having a valid
OpenGL context for GL_Vendor,... request and the other was on Grapics
contex.
Is there a need to support such information in the osg core? if yes, we can
add a draw callback to retrieve the information at the end of the first
drawing pass, if this would be too late, because the application
needs such information at the really beginning may we have to retrieve this
information at the begining.
what is the community thining or the needs?
may you try to retrieve your vendor name in a draw callback. But your code
is not at all openscenegraph standard :-)
adrian
2008/8/6 [EMAIL PROTECTED] [EMAIL PROTECTED]
In upgrading my app today to OSG 2.6, I couldn't help but noticed that it
crashed when I attempted to retrieve the OpenGL Vendor name. It wasn't a
great deal of work to determine that there was no valid OpenGL context when
it was called. A little further work showed that where I called
this-_view-setThreadingModel(osgViewer::Viewer::SingleThreaded); seemed
to cause the problem. When I commented this out, i had a valid OpenGL
Context, but OSG crashed later the first time it went to render the scene.
The app works fine with OSG 2.4, so I assume I must have just been doing
something wrong this whole time and it's finally catching up with me! I
have tried to cut the class back as much as possible for an example (below).
The app creates an instance of this class, then calls graphicsInit(), then,
finally, graphicsLoop(). I have a feeling the error lies in the constructor
itself, but I don't know what I'm doing wrong.
(Just as a bit of background, this class is supposed to support Linux and
Windows. Under Windows, it supports rendering to its own OSG-created window
or rendering to an OSG window inside a windows control.)
Any ideas would be greatly appreciated!
Thanks!
OSGExample.cpp
#include iostream
#include osg/Group
#include osg/Geode
#include osg/Shader
#include osg/MatrixTransform
#include osg/Projection
#include osg/DisplaySettings
#include osg/Texture2D
#include osg/Drawable
#include osg/CopyOp
#include osg/TexGen
#include osg/Matrix
#include osg/FrontFace
#include osg/StateSet
#include osg/Notify
#include osgDB/ReadFile
#include osgDB/FileUtils
#include osgUtil/Optimizer
#include osgViewer/View
#include osgViewer/Viewer
#include osgViewer/CompositeViewer
#include osgViewer/ViewerEventHandlers
#ifdef WIN32
#include osgViewer/api/Win32/GraphicsWindowWin32
#endif
using std::cerr;
class OSG
{
public:
#ifdef WIN32
OSG(unsigned int multisamples = 1, HWND
hostWindowHWND = NULL);
#else
OSG(unsigned int multisamples = 1);
#endif
virtual ~OSG();
virtual void graphicsInit();
virtual void graphicsLoop();
private:
std::string _vendor;
osgGA::MatrixManipulator* _viewerMatrix;
osg::State* _state;
osg::ref_ptrosg::Group _sceneGroup;
osg::ref_ptrosg::GraphicsContext _graphicsContext;
osg::ref_ptrosg::GraphicsContext::Traits _traits;
osg::ref_ptrosg::DisplaySettings _displaySettings;
osg::ref_ptrosg::GraphicsContext::WindowingSystemInterface
_wsi;
osgViewer::Viewer* _view;
osgViewer::Viewer::Windows _viewWindows;
// EventHandler* _eventHandler;
osg::ref_ptrosg::Node _hiveNode;
unsigned int _uniqueGraphicsStateID;
unsigned int _multisamples;
std::string _windowTitle;
#ifdef WIN32
HWND hostHWND;
HWND _hwnd;
HDC _hdc;
[osg-users] IntersectionVisitor and Billboards.
I am trying to use the IntersectionVisitor to intersect with billboards and think I am having the same problem as what I am quoting here. I was wondering if this was every resolved or not? I really only need to perform LineSegmentIntersections, but I cannot figure out how to get this to work correctly. If anyone knows of a solution I would be very appreciative of it. Thanks Judd Hi Shuxing Xiao, I just checked the code, and indeed the new IntersectionVisitor isn't computing the matrix required to position and rotate the drawables toward the intersector. Clearly this bit of work hasn't been completed. This leaves the question of what the implementation should look like, orientating a billboard towards a single line segment is straight forward, but what about a polytope, or a plane or a group of intersections? I'm not sure about these later two. The IntersectionVisitor is by design decoupled from the actual intersection types so doesn't directly know about the orientation. This suggests we need to add some orientation method into the Intersector base class. I'm very busy with other tasks right now so can't tackle this, feel free to have bash at implementing the orientation code. Robert. On 4/16/07, Shuxing Xiao [EMAIL PROTECTED] wrote: I'm using osgUtil::LinesegmentIntersector to pick objects from my scene. It works all fine except the Billboard nodes, all the objects can only be picked at the origin point. Since my drawables are added with a specified position, so I guess this specified position didn't contribute to the view matrix. I dig into the source code, and found that IntersectionVisitor treat Billboard the same as Geode. And I found that the old intersectvisitor considered the differences. I wonder how to deal with this problem? Thanks. ___ osg-users mailing list [EMAIL PROTECTED] http://openscenegraph.net/mailman/listinfo/osg-users http://www.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 2.6: No OpenGL Context?
Hi Adrian,
A good way to add this type of initial setup testing is to use a
custom viewer realize operation. The osgshaderterrain example has a
such as custom realize operation - although reviewing the code know
the actual operation addition is currently commented out with an
associated comment suggesting that it crashed... commenting it back in
now doesn't reveal any problems so I think this crash has been fixed
by other improvements to the osgViewer w.r.t initialization.
Robert.
On Wed, Aug 6, 2008 at 7:38 PM, Adrian Egli OpenSceneGraph (3D)
[EMAIL PROTECTED] wrote:
Hi
i am not sure where the bug is coming from, but we had in the past (one
week) two threads on this topic. one was the problem of having a valid
OpenGL context for GL_Vendor,... request and the other was on Grapics
contex.
Is there a need to support such information in the osg core? if yes, we can
add a draw callback to retrieve the information at the end of the first
drawing pass, if this would be too late, because the application
needs such information at the really beginning may we have to retrieve this
information at the begining.
what is the community thining or the needs?
may you try to retrieve your vendor name in a draw callback. But your code
is not at all openscenegraph standard :-)
adrian
2008/8/6 [EMAIL PROTECTED] [EMAIL PROTECTED]
In upgrading my app today to OSG 2.6, I couldn't help but noticed that it
crashed when I attempted to retrieve the OpenGL Vendor name. It wasn't a
great deal of work to determine that there was no valid OpenGL context when
it was called. A little further work showed that where I called
this-_view-setThreadingModel(osgViewer::Viewer::SingleThreaded); seemed
to cause the problem. When I commented this out, i had a valid OpenGL
Context, but OSG crashed later the first time it went to render the scene.
The app works fine with OSG 2.4, so I assume I must have just been doing
something wrong this whole time and it's finally catching up with me! I
have tried to cut the class back as much as possible for an example
(below). The app creates an instance of this class, then calls
graphicsInit(), then, finally, graphicsLoop(). I have a feeling the error
lies in the constructor itself, but I don't know what I'm doing wrong.
(Just as a bit of background, this class is supposed to support Linux and
Windows. Under Windows, it supports rendering to its own OSG-created window
or rendering to an OSG window inside a windows control.)
Any ideas would be greatly appreciated!
Thanks!
OSGExample.cpp
#include iostream
#include osg/Group
#include osg/Geode
#include osg/Shader
#include osg/MatrixTransform
#include osg/Projection
#include osg/DisplaySettings
#include osg/Texture2D
#include osg/Drawable
#include osg/CopyOp
#include osg/TexGen
#include osg/Matrix
#include osg/FrontFace
#include osg/StateSet
#include osg/Notify
#include osgDB/ReadFile
#include osgDB/FileUtils
#include osgUtil/Optimizer
#include osgViewer/View
#include osgViewer/Viewer
#include osgViewer/CompositeViewer
#include osgViewer/ViewerEventHandlers
#ifdef WIN32
#include osgViewer/api/Win32/GraphicsWindowWin32
#endif
using std::cerr;
class OSG
{
public:
#ifdef WIN32
OSG(unsigned int multisamples = 1, HWND
hostWindowHWND = NULL);
#else
OSG(unsigned int multisamples = 1);
#endif
virtual ~OSG();
virtual void graphicsInit();
virtual void graphicsLoop();
private:
std::string _vendor;
osgGA::MatrixManipulator* _viewerMatrix;
osg::State* _state;
osg::ref_ptrosg::Group _sceneGroup;
osg::ref_ptrosg::GraphicsContext _graphicsContext;
osg::ref_ptrosg::GraphicsContext::Traits _traits;
osg::ref_ptrosg::DisplaySettings _displaySettings;
osg::ref_ptrosg::GraphicsContext::WindowingSystemInterface _wsi;
osgViewer::Viewer* _view;
osgViewer::Viewer::Windows _viewWindows;
// EventHandler* _eventHandler;
osg::ref_ptrosg::Node _hiveNode;
unsigned int _uniqueGraphicsStateID;
unsigned int _multisamples;
std::string _windowTitle;
#ifdef WIN32
HWND hostHWND;
HWND _hwnd;
HDC _hdc;
HGLRC _wglContext;
#endif
};
#ifdef WIN32
OSG::OSG(unsigned int multisamples, HWND hostWindowHWND)
#else
OSG::OSG(unsigned int multisamples)
#endif
{
osg::setNotifyLevel(osg::ALWAYS);
this-_view = NULL;
this-_uniqueGraphicsStateID = 0;
this-_multisamples = 0;
// this-_eventHandler =
[osg-users] Mac / X-Code build problem
I'm encountering a bizarre problem with my X-Code build - with (so far) 2.5.5 and the 2.6.0-rc2 (will try 2.6.0-final in a moment). I'll say right now that I assume it must be a local config problem, I'm not trying to blame OSG for the problem (much), just need some help to get things working. Using the supplied X-Code projects, I can build all of OSG just fine. It so happens that I'm switching the SDK from 10.4 to 10.5, and I have to fix one path in project to build : the osgWidgets path references itself via a path that only works in checkout, not a snapshot. Both of these things seem pretty minor and unrelated to the problem I'm having. The problem is, having built **all** of OSG (in either debug or release, BTW), nothing works - all the OSG apps launch, but segfault within a few seconds - some immediately, some others sort of work (osgplanets, untextured) but then crash, others load but the display is broken (osgforrest, the tree billboard quads are random colors, then it crashes) and so on. The crashes are all very strange - deep inside the AGL / CGL code, pthread mutex or condvar internal segfaults (loads of these), or just in the guts of the GL driver. Most seem to be mangled pointers, but when I look at the OSG objects on the stack, they usually look okay - no obvious corruption, though I'm guessing what values are sane for the OSG internals. So, something is obviously very screwed up with my build (or my runtime environment). I've built and run countless other GL applications with the same tool chain, and I build C++ apps every day - I am sceptical that something is seriously wrong with my compiler / OpenGL framework / GL driver. Oh, yes, this happens on two different machines (both running 10.5 and X-Code 3.0 / 3.1), a MacPro and a MacBook Pro. S - what on earth can be going on? I'm looking forward to testing a binary 2.6.0 release, but I'm also mystified. The whole thing feels like memory corruption or some hellish threading problem, but I don't see why I would be getting those when no one else is. So, any bright ideas? I'd be happy to have my amazingly dumb mistake pointed out. I suppose I should try a Cmake build of 2.6.0 as well. James ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Mac / X-Code build problem
James Turner wrote: I'm encountering a bizarre problem with my X-Code build - with (so far) 2.5.5 and the 2.6.0-rc2 (will try 2.6.0-final in a moment). I'll say right now that I assume it must be a local config problem, I'm not trying to blame OSG for the problem (much), just need some help to get things working. Using the supplied X-Code projects, I can build all of OSG just fine. It so happens that I'm switching the SDK from 10.4 to 10.5, and I have to fix one path in project to build : the osgWidgets path references itself via a path that only works in checkout, not a snapshot. Both of these things seem pretty minor and unrelated to the problem I'm having. The problem is, having built **all** of OSG (in either debug or release, BTW), nothing works - all the OSG apps launch, but segfault within a few seconds - some immediately, some others sort of work (osgplanets, untextured) but then crash, others load but the display is broken (osgforrest, the tree billboard quads are random colors, then it crashes) and so on. The crashes are all very strange - deep inside the AGL / CGL code, pthread mutex or condvar internal segfaults (loads of these), or just in the guts of the GL driver. Most seem to be mangled pointers, but when I look at the OSG objects on the stack, they usually look okay - no obvious corruption, though I'm guessing what values are sane for the OSG internals. So, something is obviously very screwed up with my build (or my runtime environment). I've built and run countless other GL applications with the same tool chain, and I build C++ apps every day - I am sceptical that something is seriously wrong with my compiler / OpenGL framework / GL driver. Oh, yes, this happens on two different machines (both running 10.5 and X-Code 3.0 / 3.1), a MacPro and a MacBook Pro. S - what on earth can be going on? I'm looking forward to testing a binary 2.6.0 release, but I'm also mystified. The whole thing feels like memory corruption or some hellish threading problem, but I don't see why I would be getting those when no one else is. So, any bright ideas? I'd be happy to have my amazingly dumb mistake pointed out. I suppose I should try a Cmake build of 2.6.0 as well. James, I can't speak about the Xcode project that comes with OSG, but I believe it is no longer actively maintained. While Xcode support is still far from perfect within CMake, I have made many recent improvements to the CMake build system Xcode generator. Give it a try. CMake 2.6 is required, and there is a nice GUI version on the CMake downloads page. There are several caveats to building OSG on OSX that you should be aware of. Read the README.txt file in the base directory of the OSG 2.6 source code. Feedback is welcome. -Eric ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Mac / X-Code build problem
On 6 Aug 2008, at 21:57, Eric Sokolowsky wrote: I can't speak about the Xcode project that comes with OSG, but I believe it is no longer actively maintained. While Xcode support is still far from perfect within CMake, I have made many recent improvements to the CMake build system Xcode generator. Give it a try. CMake 2.6 is required, and there is a nice GUI version on the CMake downloads page. It's quite unusual (in an open source project) to discover that something isn't actively maintained when it apparently works fine :) Of course, it's not 'working fine', but equally, the projects are for X-Code 3.0, and give every appearance of building all the code and generally functioning. Anyway, thanks for pointing this out. I will go and try a CMake build now. There are several caveats to building OSG on OSX that you should be aware of. Read the README.txt file in the base directory of the OSG 2.6 source code. Feedback is welcome. I've read the README, it seems pretty clear and as far as I know I'm complying with its instructions - but I will double-check. Thanks, James ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] osgdem and UTM terrain tiles...
Using the GDAL tools, I've built a mosaic of terrain tiles that I'm stitching together using osgdem to create a terrain database. Each terrain tile uses a UTM mapping that is reprojected from a lat/long mapping. Unfortunately when I do the reprojection for each tile, the terrain tile gets slightly shifted so that when I stitch them all together using osgdem, I have discontinuities at each tile boundary in the finished terrain database. My question is, is there a way to avoid this nasty artifact using either GDAL or osgdem? Perhaps I need to resort to another tool to massage the data before handing it over to osgdem? Thanks for any input in advance. -Shayne smime.p7s Description: S/MIME cryptographic signature ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] osg and speedtree
Hi everyone, I know this question has been asked previously but I have not been able to locate a definitive answer so.. We have a need to implement speedtree into OSG, and am wondering if anyone has had prior success with this? From a cursory glance it does not appear that it will be as straight forward as the SPT documentation would have us believe. If so does anyone have any pointers, tips or perhaps even reference code they may be willing to share that may be of assistance??? I would be grateful for any help! cheers Jon Newell ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] 回复:
hi, Only set osg::Texture::REPEAT is not enough! -- 原始邮件 -- 发件人: Paul Melis[EMAIL PROTECTED]; 发送时间: 2008年7月30日(星期三) 下午04:57 收件人: OpenSceneGraph Usersosg-users@lists.openscenegraph.org; 主题: Re: [osg-users] monkey wrote: Thank for you all! osgExp is *REALLY NOT* surport Texture UV repeat! But I have make it run ok through change its code! Thank you very much! Looking into the current osgExp sources it seems that it already should have support for repeating textures. Whenever a texture coordinate is encountered that is outside the [0,1] range, the wrap setting for the exported Texture2D is set to osg::Texture::REPEAT. You said in an earlier mail that the UV coordinates in the exported file are incorrect (0..1 instead of 0..3), but do you see the REPEAT in the output file? Paul -- 原始邮件 -- *发件人:* GMD GammerMaxyandex.ru[EMAIL PROTECTED]; *发送时间:* 2008年7月29日(星期二) 凌晨02:59 *收件人:* osg-usersosg-users@lists.openscenegraph.org; *主题:* Re: [osg-users] help me! osgExp is not surport Texture UV repeat? Yes :( 28.07.08, 14:43, Alberto Luaces [EMAIL PROTECTED]: Hi, Sorry, I have no expertise with 3DSMax and/or osgExp, so I cannot help you further. Nevertheless, here is the osgExp code download page address: http://sourceforge.net/svn/?group_id=148454 I suppose there are compilation instructions within the source. El Viernes 25 Julio 2008ES 07:00:08 monkeymonkey escribió: hi!Thank you for your answer! I try it as you say, but not right also,i think because Texture Coord output of osgExp is not right,they are between 0 and 1,it should be 0 to 3. i change it , it is OK!, so it is the really reason. I must change osgExp scource code,and recompile it? how to do it? From: [EMAIL PROTECTED] To: osg-users@lists.openscenegraph.org Date: Wed, 23 Jul 2008 09:57:33 +0200 Subject: Re: [osg-users] help me! Hi, El Miércoles 23 Julio 2008ES 08:48:19 monkeymonkey escribió: hi! i use Texture UV repeat 3.0 in 3dsMax8,but when i output it to ive,then use osgViewer to look,texture is not right. it seem that repeat is not work! why?osgExp is not surport Texture UV repeat? how to make it? Convert your .ive to .osg with osgconv, then look for the wrap parameter in the file like wrap_s,t,... REPEAT If there is something other word than REPEAT, change it. Alberto ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org -- Много спама? Пользуйтесь Яндекс.Почтой http://mail.yandex.ru/nospam ___ 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
