Re: [Opensg-users] vred and corrupt geometry

2005-07-01 Thread Allen Bierbaum
Dirk Reiners wrote: Hi Allen, On Fri, 2005-07-01 at 15:13 -0500, Allen Bierbaum wrote: Thanks. Is there any interest/plans in merging this into the VerifyGeoGraphOp in OpenSG? Interest: definitely. Do you wanna do it? ;) Plans: I didn't know Andreas had his own code in Vre

Re: [Opensg-users] vred and corrupt geometry

2005-07-01 Thread Dirk Reiners
Hi Allen, On Fri, 2005-07-01 at 15:13 -0500, Allen Bierbaum wrote: > Thanks. > > Is there any interest/plans in merging this into the VerifyGeoGraphOp in > OpenSG? Interest: definitely. Do you wanna do it? ;) Plans: I didn't know Andreas had his own code in Vred, so I didn't plan for

Re: [Opensg-users] vred and corrupt geometry

2005-07-01 Thread Allen Bierbaum
Thanks. Is there any interest/plans in merging this into the VerifyGeoGraphOp in OpenSG? -Allen Andreas Zieringer wrote: Hi Allen, I attached the code. Andreas This may be slightly off topic, but I use VRED to preview/review/check geometry that I dump out of OpenSG applications.

Re: [Opensg-users] what is OSG_SUPPORT_NO_GEO_INTERFACE?

2005-07-01 Thread Dirk Reiners
Hi Akos, On Thu, 2005-06-30 at 19:27 +0200, Akos Balazs wrote: > > Oh... That's so... reassuring. :-) I hope so, that was it's purpose. ;) > As a sidenode, there's at least one place in > common.i686-pc-linux-gnu-icc.in that explicitly check that the major > version number equals to

Re: [Opensg-users] vred and corrupt geometry

2005-07-01 Thread Andreas Zieringer
Hi Allen, I attached the code. Andreas This may be slightly off topic, but I use VRED to preview/review/check geometry that I dump out of OpenSG applications. In the latest version of VRED there is a feature where upon load it detects corrupted/invalid geometry and removes that geomet

[Opensg-users] vred and corrupt geometry

2005-07-01 Thread Allen Bierbaum
This may be slightly off topic, but I use VRED to preview/review/check geometry that I dump out of OpenSG applications. In the latest version of VRED there is a feature where upon load it detects corrupted/invalid geometry and removes that geometry. This is nice, but unfortunately it is

Re: [Opensg-users] glLight indices

2005-07-01 Thread Andreas Zieringer
Hi Terry, thanks for the hint it was a bug in the render action I fixed it. Andreas Yes. I was testing with testSHLLights, in which the headlight was already turned off. - Terry On 7/1/05, Andreas Zieringer <[EMAIL PROTECTED]> wrote: Hi Terry, thanks for the patch. Perhaps I should add su

Re: [Opensg-users] glLight indices

2005-07-01 Thread Terry Welsh
I believe each of the uniforms needs to be added with a line such as: myshlchunk->setUniform(OSGLight0Active, 0); So the user can add as many or as few as he wants to add, and the unused ones will not hurt performance. If this is all true, then adding all 32 shouldn't be damaging in any way. So

Re: [Opensg-users] glLight indices

2005-07-01 Thread Matthias Stiller
As promised in my yesterdays post, I can confirm that the Wildcat Realizm 800 supports 32 (!) active lights in hardware. Don't ask me about performance, we usually do not have scene with so many lights enabled at the same time though :) Now the question how to deal with this. Maybe an additiona

Re: [Opensg-users] How can I use OSGProjectionMatrix?

2005-07-01 Thread Andreas Zieringer
Hi Marcus, there is no advantage I prefer the glstate but someone complaint that the glstate didn't work on a ATI card ... Andreas What are the advantages of using osg-special parameters instead of those directly in the gl-state? /Marcus Andreas Zieringer wrote: Hi Estee, I fixed some

Re: [Opensg-users] Text patches

2005-07-01 Thread Dirk Reiners
Hi everybody, here are my $.02... On Thu, 2005-06-30 at 22:48 +0200, Patrick Dähne wrote: > Hello Allen, > > > The key changes were to: > > > > - Refactor the geometry data format used to not use any indexing. > > Dirk tells me that multi-indexing can perform poorly and it just > >

Re: [Opensg-users] How can I use OSGProjectionMatrix?

2005-07-01 Thread Marcus Lindblom
What are the advantages of using osg-special parameters instead of those directly in the gl-state? /Marcus Andreas Zieringer wrote: Hi Estee, I fixed some bugs should work now there is also a new example testCGShader2.cpp in OpenSG/Source/Contrib/CGChunk I renamed some parameters following

Re: [Opensg-users] How can I use OSGProjectionMatrix?

2005-07-01 Thread Andreas Zieringer
Hi Estee, I fixed some bugs should work now there is also a new example testCGShader2.cpp in OpenSG/Source/Contrib/CGChunk I renamed some parameters following are supported: OSGProjection OSGModelViewProjection OSGTexture OSGModelView OSGModelViewI OSGModelViewIT Andreas Hello all, I read

Re: [Opensg-users] glLight indices

2005-07-01 Thread Terry Welsh
Yes. I was testing with testSHLLights, in which the headlight was already turned off. - Terry On 7/1/05, Andreas Zieringer <[EMAIL PROTECTED]> wrote: > Hi Terry, > > thanks for the patch. Perhaps I should add support for arrays. Did you > disable the headlight? > > Andreas > > > Okay, I submit

Re: [Opensg-users] OpenGL Shading Language - Experimental/SHL working ?

2005-07-01 Thread Andreas Zieringer
Hi Tobias, it is already compiled into the system lib and it is stable. You can find some examples test programs in OpenSG/Source/Experimental/SHL and also a tutorial in OpenSG/Tutorials/22Shader.cpp Andreas Hi OpenSG users! i wonder if the SHL code under Source/Experimental is actually wo

Re: [Opensg-users] glLight indices

2005-07-01 Thread Andreas Zieringer
Hi Terry, thanks for the patch. Perhaps I should add support for arrays. Did you disable the headlight? Andreas Okay, I submitted a patch that adds OSGLight0Active through OSGLight7Active. There's a weird problem where RenderAction->getActiveLightMask() returns a 1 when I expect it to retur

Re: [Opensg-users] glLight indices

2005-07-01 Thread Marcus Lindblom
Regarding to compiling/linking, note that not all drivers support separate compilation and linking. Esp. some versions of nvidia drivers on windows (may also be the current) do not compile anything until linking, and then it is done by simply concatenating all sources into one file. Anyway, i