Re: [osg-users] Potential fix for osgDB MSVC linkage issues (FORCE:MULTIPLE needed)

[Nickolai Medvedev]

Hi, guys!

Let's remember, what exactly it std::fstream:

typedef basic_ifstream< char, char_traits< char > > ifstream;
typedef basic_ofstream< char, char_traits< char > > ofstream;
typedef basic_fstream< char, char_traits< char > > fstream;

So, if problem is names, maybe do something like:

typedef basic_ifstream< char, char_traits< char > > osg_ifstream;

or to inherit osgDB:: fstream from std::basic_fstream< char, std::char_traits< 
char > >.

Sorry, i have no time that to check it.

Cheers,
Nickolai

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=68432#68432





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

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[Jannik Heller]

Hi Robert,

I can confirm that using dirtyGLObjects() does update the geometry.

I found a new bug, there seems to be a state leak somewhere in the interaction 
of osgUtil::IncrementalCompileOperation, VAO and dynamic geometry. It's a weird 
issue that comes and goes, some static meshes either flickering in/out of 
existance or having their vertices garbled. I am still trying to nail down the 
conditions for this elusive bug, and trying to create a reproducible example. 
Will update you with new findings.

Cheers,
Jannik

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=68431#68431





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

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[Remo Eichenberger]

Hi Julien

Thanks for your hint :) I've tried it with GL3 core profile on MacOSX 10.11.6 
with "vertex_array_object" branch. I tried to load a stock OSG sample model: 
glider.osg (and also glsl_simple.osgt). But i can't see the model. Here is my 
debug output:


Code:

Duplicate enum value 10 with old string: GL_LINES_ADJACENCY_EXT and new string: 
GL_LINES_ADJACENCY
Duplicate enum value 11 with old string: GL_LINE_STRIP_ADJACENCY_EXT and new 
string: GL_LINE_STRIP_ADJACENCY
Duplicate enum value 12 with old string: GL_TRIANGLES_ADJACENCY_EXT and new 
string: GL_TRIANGLES_ADJACENCY
Duplicate enum value 13 with old string: GL_TRIANGLE_STRIP_ADJACENCY_EXT and 
new string: GL_TRIANGLE_STRIP_ADJACENCY
DatabasePager::addDatabaseThread() HANDLE_NON_HTTP
DatabasePager::addDatabaseThread() HANDLE_ONLY_HTTP
Opened DynamicLibrary osgPlugins-3.5.4/osgdb_osgd.so
CullSettings::readEnvironmentalVariables()
CullSettings::readEnvironmentalVariables()
Constructing PixelBufferObject for image=0x10300a8c0
Opened DynamicLibrary osgPlugins-3.5.4/osgdb_deprecated_osgd.so
OSGReaderWriter wrappers loaded OK

Stats before:
 Object TypeUnique Instanced
 ----- -
 StateSet3 6
 Group   0 0
 Transform   0 0
 LOD 0 0
 Switch  0 0
 Geode   1 1
 Drawable6 6
 Geometry6 6
 Fast geom.  6 6
 Vertices  676   676
 Primitives640   640
Optimizer::optimize() doing TESSELLATE_GEOMETRY
Optimizer::optimize() doing REMOVE_LOADED_PROXY_NODES
Optimizer::optimize() doing COMBINE_ADJACENT_LODS
Optimizer::optimize() doing OPTIMIZE_TEXTURE_SETTINGS
Optimizer::optimize() doing SHARE_DUPLICATE_STATE
Num of StateSet=3
state attribute list
0x103329380  Material
0x103329f60  Material
searching for duplicate attributes
searching for duplicate attributes
Optimizer::optimize() doing TEXTURE_ATLAS_BUILDER
Num of StateSet=3
state attribute list
0x103329380  Material
0x103329f60  Material
searching for duplicate attributes
searching for duplicate attributes
Optimizer::optimize() doing COPY_SHARED_NODES
Shared node 0
Optimizer::optimize() doing 
FLATTEN_STATIC_TRANSFORMS_DUPLICATING_SHARED_SUBGRAPHS
Optimizer::optimize() doing MERGE_GEODES
MERGE_GEODES took 4e-06
Optimizer::optimize() doing CHECK_GEOMETRY
Optimizer::optimize() doing MAKE_FAST_GEOMETRY
Optimizer::optimize() doing MERGE_GEOMETRY
MERGE_GEOMETRY took 4.4e-05
Optimizer::optimize() doing TRISTRIP_GEOMETRY
TriStripVisitor::stripify(Geometry&): Number of indices546 numUnique294
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.85714
TriStripVisitor::stripify(Geometry&): doing tri strip
TriStripVisitor::stripify(Geometry&): Number of indices26 numUnique24
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.08333
TriStripVisitor::stripify(Geometry&): doing tri strip
TriStripVisitor::stripify(Geometry&): Number of indices26 numUnique24
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.08333
TriStripVisitor::stripify(Geometry&): doing tri strip
TriStripVisitor::stripify(Geometry&): Number of indices26 numUnique24
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.08333
TriStripVisitor::stripify(Geometry&): doing tri strip
TriStripVisitor::stripify(Geometry&): Number of indices26 numUnique24
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.08333
TriStripVisitor::stripify(Geometry&): doing tri strip
TriStripVisitor::stripify(Geometry&): Number of indices26 numUnique24
TriStripVisitor::stripify(Geometry&): ratio indices/numUnique1.08333
TriStripVisitor::stripify(Geometry&): doing tri strip
Optimizer::optimize() doing REMOVE_REDUNDANT_NODES
Optimizer::optimize() doing SPATIALIZE_GROUPS
Optimizer::optimize() doing BUFFER_OBJECT_SETTINGS
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)
geometry.setUseVertexBufferObjects(1)
geometry.setUseDisplayList(0)

Stats after:
 Object TypeUnique Instanced
 ----- -
 StateSet3 6
 Group   0 0
 Transform   0 0
 LOD 0 0
 Switch  0 0
 Geode   1 1
 Drawable6 6
 Geometry6 6
 Fast geom.  6 6
 Vertices  414   414
 Primitives640   640
CocoaWindowingSystemInterface::initAsStandaloneApplication 
ContextData::registerGraphicsContext 0x103008a50
ShaderComposer::ShaderComposer() 0x103010d60
ContextData::ContextData()0x103013e70

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[Julien Valentin]

Hi Remoe
You haven't search deep enough for OSG_VERTEX_BUFFER_HINT environment variable 
https://github.com/openscenegraph/OpenSceneGraph/compare/vertex_array_object
src/osg/DisplaySettings.cpp 684
src/osg/state.cpp l179

And under MacOS I think you should use osgsimplegl3 example as base code for 
testing. AFAIK it's the simpliest way to work with a core profile.



remoe wrote:
> Hi Robert
> 
> I've tried to test "vertex_array_object" branch with MacOSX to run 
> osgsimplegl3. Does anyone has a "running" model without any shader errors to 
> test this?
> 
> btw: I don't see any OSG_VERTEX_BUFFER_HINT environment variable check on the 
> current source code. How can i enable the VAO support of OSG?
> 
> Cheers,
> Remo


--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=68429#68429





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

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[wang dexing]

Hi Robert,

It is my fault.I thought I didn't express my idea clearly,so I explained many 
times.now I get it.


... 

Thank you!

Cheers,
wang

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=68428#68428





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

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[Remo Eichenberger]

Hi Robert

I've tried to test "vertex_array_object" branch with MacOSX to run 
osgsimplegl3. Does anyone has a "running" model without any shader errors to 
test this?

btw: I don't see any OSG_VERTEX_BUFFER_HINT environment variable check on the 
current source code. How can i enable the VAO support of OSG?

Cheers,
Remo

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=68427#68427





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

Re: [osg-users] [About VertexArrayObject Branch] FeedBack and Insight

[Robert Osfield]

HI Wang,

On 19 August 2016 at 02:35, wang dexing  wrote:
>> I have outlined why mixing scene graph state with the rendering back-end is 
>> a bad idea.
>
> Could you tell me where I can read it?

My earlier reply to you.

> By the way,I don't mix scene graph state with the rendering back-end,but make 
> it possible,since I only add an extension layer above Geometry to manage 
> 'vao'.You can do other things except that,for example,choose 'vbos' that you 
> want from different 'Geometrys' to form a new drawing object(or sharing 
> 'vbos'  in other words).
> I know you best job in latest incarnation,and I'm not here to let you make 
> any changes.I only want to discuss its possibility.Maybe we can discuss in a 
> pleasant way,not so serious.javascriptemoticon(':D')

I'm sorry but I'm not the business of endlessly discussing options, I
have already looked what you have written and feel it's a bad idea,
there is little to gain from endlessly going over it.  I don't owe you
a detailed essay on scene graphs to explain why.

I have poured a great deal of my time, unpaid, into the VAO
implementation, my focus is on rapping up this implementation and
getting OSG-3.6 out the door.  Anything that is distraction from this
task is something I have to minimise and why I'm rather fed up having
to reply to a dead-end thread that you are trying to keep alive.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org