Hi Robert,
Everything works fine now. Thanks a lot for investigating this problem.
-- Nico
On Fri, Nov 27, 2009 at 12:49 PM, Robert Osfield
wrote:
> Hi Nico,
>
> I have just checked in the implementation of the GL object clean up,
> and all the debug messages are now quieten down/removed so th
Hi Nico,
I have just checked in the implementation of the GL object clean up,
and all the debug messages are now quieten down/removed so the code
should be good to go for real application work now.
Please let me know if you still see problems.
Robert.
_
Aargh. Yes, -r it was. I'll try this tonight or tomorrow.
Nico
On Thu, Nov 26, 2009 at 7:12 PM, Robert Osfield wrote:
> HI Nico,
>
> On Thu, Nov 26, 2009 at 5:16 PM, Nico Kruithof
> wrote:
> > Hi Robert,
> > Perfect! My application works exits normally again. The glider works as
> > well. Only
HI Nico,
On Thu, Nov 26, 2009 at 5:16 PM, Nico Kruithof wrote:
> Hi Robert,
> Perfect! My application works exits normally again. The glider works as
> well. Only the tetra file didn't. It crashes after the first viewer.
So you still have a crash with tetra.osg then? FYI, I've made a
couple of
Hi Robert,
Perfect! My application works exits normally again. The glider works as
well. Only the tetra file didn't. It crashes after the first viewer.
osgcamerad.exe -j 3 d:\tetra.osg:
Warning: Could not find plugin to read objects from file "3".
GLBufferObjectSet::GLBufferObjectSet _profile._s
Sure, I'll do that tonight. Thanks a lot for looking into this.
-- Nico
On Thu, Nov 26, 2009 at 1:36 PM, Robert Osfield wrote:
> Hi Nico,
>
> I've tracked down issues with both the clean up of the texture object
> pool and buffer object pool that has required introduction of some new
> methods f
Hi Nico,
I've tracked down issues with both the clean up of the texture object
pool and buffer object pool that has required introduction of some new
methods for cleanup. I have checked into svn/trunk the initial cut of
this code, it enables "osgcamera -r 5 tetra.osg" to work just fine, as
well a
Hi Nico,
osgcamera -r Tetra.osg is failing due to issue with clean up of the
buffer object pool. I have already tracked down a bug in the clean up
of the texture object pool, and found the same problem in the buffer
object pool and applied the same fix, but alas still the problem
persists so ther
Hi Robert,
Not too much good news I'm afraid. I tried my application and the osgcamera.
I have to stop now. I might have some time to test tomorrow night (it's a
private project for now).
I attached the model I'm using. It loads fine in my application, but crashes
when I close it. The osgcamera d
Hi Robert,
I realised that another difference is that I'm using the static viewer. I'll
try the texture patch, however the model I constructed didn't contain any
textures. Just vertices, normals and triangles.
-- Nico
On Wed, Nov 25, 2009 at 8:08 PM, Robert Osfield wrote:
> Hi Nico,
>
> I have
Hi Robert,
The first call (no --vbo) runs correctly. The second and third (with --vbo)
do not crash, but the second and third screen are wrong. I also tried the
fourth combination "osgcamera -r 3 --shared glider.osg". That one worked
fine.
-- Nico
This is the output:
osgcamera -r 3 glider.osg
+
Hi Nico,
I've now modified the osgcamera example so that it can be used as
tested bed for repeated construction of viewers and enabling of VBO's.
The example is now checked into svn/trunk, could you please try it
out on your system and see what errors you can pick up. Try the
following command l
Ok, thanks for looking into it. I'll make a simple compiling program that
causes the crash. I extracted the code snippet from my own program. I will
check the example under ubuntu as well.
Nico
On Wed, Nov 25, 2009 at 3:25 PM, Robert Osfield wrote:
> Hio Nico,
>
> I've attempted to reproduce the
Hio Nico,
I've attempted to reproduce the cash using a scene graph that I force
to use VBO's, and the run the viewer for a frame() or run() then
destruct the viewer, then the scene graph - along the lines of your
code fragment, but it runs reliable on my Kubuntu 9.04 64bit system
with svn/trunk.
Robert,
My response wasn't meant to emphasize that things got broken, just to give
you as detailed information as possible. I really do appreciate all the work
you put in OSG. If there is any way I could help / test let me know.
-- Nico
On Wed, Nov 25, 2009 at 11:21 AM, Robert Osfield
wrote:
>
Hi Nico,
On Wed, Nov 25, 2009 at 10:07 AM, Nico Kruithof wrote:
> I used windows vista 64 and 32. It happens with an up to date trunk and I
> didn't have the problem with a previous version. I don't know the revision I
> used to use, was probably before your openGL ES work.
Back in September I i
Hi Robert,
I used windows vista 64 and 32. It happens with an up to date trunk and I
didn't have the problem with a previous version. I don't know the revision I
used to use, was probably before your openGL ES work.
Thanks for looking into this.
-- Nico
On Wed, Nov 25, 2009 at 10:47 AM, Robert
HI Nico,
This is clearly a bug, as it certainly shouldn't crash with this type
of usage. I will be investigating various bugs today and will have a
look at this one as well.
Which platform are you working on?
Robert.
On Tue, Nov 24, 2009 at 8:18 PM, Nico Kruithof wrote:
> Hello,
> When a node
Hello,
When a node with VBO's enabled outlives the lifetime of it's renderer, I get
an access violation in GLBufferObjectSet::orphan when I try to destroy the
node. I tried to write down the scenario:
using namespace osg;
ref_ptr node(new Node);
ref_ptr geom(new osg::Geometry);
// ... fill geom
19 matches
Mail list logo
