Hi Simon,
Simon Haegler wrote:
>
> thanks for the hint. first the buildings, then the vegetation ;-)
Why do you have to start with the hard stuff? ;)
> a related question: what is the fastest way to make a subgraph transparent?
>
> i tried MakeTransparentGraphOp, but the framerate dro
Hi Carsten,
Carsten Neumann wrote:
>
> yes, but if scons sets _DEBUG on all platforms for debug builds (does
> it?) we could put this in OSGConfig.h:
>
> #ifdef _DEBUG
> #define OSG_DEBUG
> #endif
>
> and maybe also put it into OSGConfigured.h ?
Does scons always do that? In that case
Hi Allen,
Allen Bierbaum wrote:
>
> How would I go through the list and find the ones I want? For example
> how would I know which attachment container is "MyEditorStuff" if there
> are possibly other attachment containers off the root of the scene?
Even after removing the group id yo
Dirk Reiners wrote:
> Hi Allen,
>
>Allen Bierbaum wrote:
>
>
>>- First, it is possible to have objects (field containers) that are part
>>of the concept of a scene, but are not actually in the scene graph.
>>
>>
>
>Personally I disagree, but that's a different story.
>
>...
>
>
>>
Hi Allen,
Allen Bierbaum wrote:
> One idea I have been toying with is to just use AttachmentContainers.
> For example if I wanted to store materials and FBO objects.
>
> rootNode
>- AttachmentContainer:"MyEditorStuff"
> - AttachmentContainer: "Materials"
> - Material:
Hi Allen,
Allen Bierbaum wrote:
>
> - First, it is possible to have objects (field containers) that are part
> of the concept of a scene, but are not actually in the scene graph.
Personally I disagree, but that's a different story.
...
> I think it would be better to support holding
With the help of Infiscape we now have a Buildbot master in place and now need some slave machines to volunteer some CPU cycles for daily testing of OpenSG 2.More instructions can be found at:
http://opensg.vrsource.org/trac/wiki/DailyBuildsMachines we already have covered are:Fedora Core 5 - AMD6
Allen Bierbaum wrote:
[snip]
> Anyway, does anyone have a good idea for how to deal with this in my
> application? Is there any other class that I should be looking at that
> is new in OpenSG2?
In our app we have a big registry (called DataSpace) which is a map from
string to object (or actual
Antonio Bleile wrote:
> Scrive Carsten Neumann <[EMAIL PROTECTED]>:
>
>> Antonio Bleile wrote:
>>> Here's the next issue, sorry cut&paste from a dos box ;(:
>>> Any ideas?
>> Hm, any chance the compiler will respect setting LANG=en or do you know
>> another way of getting it to produce english mes
Scrive Carsten Neumann <[EMAIL PROTECTED]>:
> Antonio Bleile wrote:
> > Here's the next issue, sorry cut&paste from a dos box ;(:
> > Any ideas?
>
> Hm, any chance the compiler will respect setting LANG=en or do you know
> another way of getting it to produce english messages ?
> I'm afraid it's d
Antonio Bleile wrote:
> Here's the next issue, sorry cut&paste from a dos box ;(:
> Any ideas?
Hm, any chance the compiler will respect setting LANG=en or do you know
another way of getting it to produce english messages ?
I'm afraid it's difficult for me to decipher the error messages - I can
get
As part of the work I am doing on OpenSG2 and pyOpenSG I am working on a
scene editor. It is not going to be anything as full featured as VRED,
but it should allow editing of basic scenes and debugging/development of
shaders and multi-pass rendering with OpenSG2.
Are part of this effort I have
> Another way that's useful for trees & foliage, is to use a randomized
> alpha map and blend via alpha-test. A bit ditherish, but foliage is
> pretty ditherish anyway. It depends on your app if it will look good or
> not though.
thanks for the hint. first the buildings, then the vegetation ;-)
Here's the next issue, sorry cut&paste from a dos box ;(:
Any ideas?
scons: Building targets ...
cl /nologo /O2 /W1 /MD /EHsc /GR /TP /Ic:\programmi\boost\boost_1_33_1
/DWIN32 /
D_WINDOWS /DNDEBUG /DOSG_COMPILESYSTEMLIB
/Ibuild.win32.7.1\instlinks\include /I
build.win32.7.1\instlinks\include\Ope
Hi Carsten,
> Did you tell scons where to find boost ? If you installed it to
> D:\Files\boost-1.33 (below this there should be a directory called
> "boost" with headers), call it with
>
> scons BoostBaseDir="D:\Files\boost-1.33"
>
> Hope it helps,
> Carsten
>
> PS: Big cave
Hi Dirk and Allen,
Dirk Reiners wrote:
> Allen Bierbaum wrote:
>> Carsten Neumann wrote:
>>> I can not find the place where OSG_DEBUG is defined, so I assume it
>>> should be on the compiler command line ?
>>> Right now the scons build for 2 only puts a -D_DEBUG there, should
>>> OSGConfig
Antonio Bleile wrote:
> Now it stops compiling with a missing boost file:
>
> c:\Documents and
> Settings\bleile\Documenti\devel\OpenSG2\build.win32.7.1\instlin
> ks\include\OpenSG\OSGFieldForwards.h(45) : fatal error C1083: Impossibile
> aprire
> il file inclusione "boost/function.hpp": No such
> >>> Source\Base\Base\OSGSharedObjectHandler.cpp(502) : error C2065:
> >> '__func__':
> >>> iden
> >>> tificatore non dichiarato
> >>> scons: ***
> >>> [build.win32.8.0\type-optimized--arch-default\Source\Base\Base\OSGSha
> >>> redObjectHandler.obj] Error 2
> >>> scons: building terminated beca
Hi Allen,
Allen Bierbaum wrote:
> Carsten Neumann wrote:
>
>> Hi,
>>
>> I can not find the place where OSG_DEBUG is defined, so I assume it
>> should be on the compiler command line ?
>> Right now the scons build for 2 only puts a -D_DEBUG there, should
>> OSGConfig.h map this to OS
Hi Toni,
>>> Source\Base\Base\OSGSharedObjectHandler.cpp(502) : error C2065:
>> '__func__':
>>> iden
>>> tificatore non dichiarato
>>> scons: ***
>>> [build.win32.8.0\type-optimized--arch-default\Source\Base\Base\OSGSha
>>> redObjectHandler.obj] Error 2
>>> scons: building terminated becau
Hi Allen,
Allen Bierbaum wrote:
> I compiled everything last night hoping to avoid this type of problem.
> One other person seems to have run into the same issue. Is it possible
> that this is compiler specific? I am using gcc:
>
> g++ (GCC) 4.0.2 20051125 (Red Hat 4.0.2-8)
I have h
> Antonio Bleile wrote:
> > Hi Allen&Dirk,
> >
> > just updated the svn and calles scons, it still
> > uses the VC8 and fails compiling:
> >
> > But Allen did insert a command line option to scons
> > in order force usage of VS.NET 2003, right?
> > What is that option?
>
> MSVS_VERSION
>
> scon
Antonio Bleile wrote:
> Hi Allen&Dirk,
>
> just updated the svn and calles scons, it still
> uses the VC8 and fails compiling:
>
> But Allen did insert a command line option to scons
> in order force usage of VS.NET 2003, right?
> What is that option?
MSVS_VERSION
scons MSVS_VERSION=7.1
sho
Gerrit Voss wrote:
> On Tue, 2006-10-03 at 17:14 -0500, [EMAIL PROTECTED] wrote:
>> Author: allenb
>> Date: Tue Oct 3 17:14:39 2006
>> New Revision: 226
>> Trac changeset: http://opensg.vrsource.org/trac/changeset/226
>> ViewVC: http://realityforge.vrsource.org/viewvc/opensg?view=rev&revision=226
> Crossfading is not as simple as it sounds. SGI did it in Performer, but
> the only way to really make it work is using multisampling, as you don't
> want to fade the models to/from transparent, as that would allow you to
> see inside it, you want to fade the images of the actual models.
> Mul
> > Use two viewports (one for each camera), one in the back for further
> > geometry and the next one for closer. The nearest-viewport should have a
> > background that only erases the depth-buffer, not the color buffer.
> > (This is possible in several ways. I don't know if there is a out-of-the
Hi Praveen,
> Thanks for your answers Dirk and Marcus.
>
>
>>You could do that. It would be more efficient to either add a
>>new Action
>>for it or to generalize the existing IntersectAction. The
>>main thing to
>>do would be to remove the optimizations that speed up the
>>intersection
>>p
Dirk Reiners wrote:
> Hi Simon,
>
> Simon Haegler wrote:
>
>> i will start with adding an absolute fading range around each threshold, and
>> simply do a linear fade between the two adjacent lods.
>>
> Crossfading is not as simple as it sounds. SGI did it in Performer, but
> the onl
Hi Allen&Dirk,
just updated the svn and calles scons, it still
uses the VC8 and fails compiling:
But Allen did insert a command line option to scons
in order force usage of VS.NET 2003, right?
What is that option?
Source\Base\Base\OSGSharedObjectHandler.cpp(502) : error C2065: '__func__':
id
29 matches
Mail list logo