Hi,
I'm currently in the process of trying to make shadows work in OpenMW and am
therefore trying to learn about osgShadow and also convince it to do a bunch of
things.
I have several questions, so I'll try and organise them nicely with bullet
points and subheadings etc. In case it affects
I'm having a similar issue but under slightly different circumstances. Unlike
the original author of this thread, my leftmost monitor shows as having the
lower index.
Is there anything I can do about this?
--
Read this topic online here:
kornerr wrote:
> Hi, Chris,
>
> just create EditorConfig and share it. Those who are interested will
> follow you.Don't wait for us to want your thing. Just do the thing,
> show it and explain its benefits.
> You're not alone, we're here, but you lead EditorConfig initiative, not us.
>
>
All out of the 3.6 branch, 3.4.1 and OpenMW's 3.4.1 fork (which doesn't make
any changes to osgText, and we're trying to abandon, but it's much faster than
the official 3.4.1 release). If everything's working as we think it does,
OpenMW should work with any OSG release after 3.4.
The issue of
The problem is specifically because it's static. Unless I'm manually keeping
track of when the last text object is removed from the scene, I can't release
the program without it potentially still being in use. If I never release it,
references to it can persist beyond the lifetime of the
Hi,
It's been a couple of weeks, and I've not had any response about any official
policy on indentation in the case of existing files that already break the
global rule. Should I just go for the stance of "it's supposed to be spaces for
everything except for files that are already entirely
I'm fine making the pull request. I'd just like confirmation of exactly what
the whitespace conventions are - I wouldn't want to set it to use four spaces
for indentation for everything and then find out later that the CMake files use
tabs or anything like that.
The main rules are listed here:
I just grepped the repo for tabs just in case, and found that that's not
entirely accurate. There are a few files where tabs have snuck into otherwise
space-indented files (which is fine - adding an editorconfig file won't meddle
with existing stuff until someone edits that line), but it looks
The problem was definitely user error - parts of the scene graph were being
removed before OSG had had a chance to clean up.
Cheers,
Chris
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=75723#75723
Hi,
I know that when an OpenGL context is destroyed, the associated viewer can only
call releaseGLObjects(osg::State *) for nodes still attached to the scene
graph, so if you're removing things before that, you're manually supposed to
call releaseGLObjects on the subgraph.
However, as I see
I've been informed that I was wrong about this. While I am seeing issues with
my application, the context ID is actually being set, just it's being set to
the same value as the previous context as OSG reuses IDs for dead contexts.
What I know about the issue so far is that the per-context
Hi,
For context, this discussion is about the OpenMW CS tool, the source of which
is available here: https://gitlab.com/OpenMW/openmw/tree/master/apps/opencs.
I've kind of inherited maintenance responsibilities for the graphical parts of
this without necessarily knowing much of how it works.
Hi,
It's been a couple of weeks since the last reply to this. I'm guessing this has
fallen off the radar, but I still need to fix the issue, so would appreciate if
another look could be taken at this.
Thank you!
Cheers,
Chris
--
Read this topic online here:
Hi,
Here's a list by file extension of which indentation styles are used.
Annoyingly, a lot of the file extensions that always have tabs somewhere often
have spaces, too, so it's not quite as simple as declaring all .java files use
tabs (although I'm assuming they all used tabs when they were
at GitHub, I would
> recommend re-raising this issue and creating a pull request.
>
> Regards,
> Michael.
>
> On Fri, 12 Apr 2019 at 21:27, Chris Djali <> wrote:
>
> >
> > Hi,
> >
> > Here's a list by file extension of which indentation style
Hi,
The following basically simulates the use case that's causing problems.
Code:
#include
#include
#include
#include
#include
#include
#include
/** A representation of something that gets edited.
* Pretend it's actually more complicated than this so that reference counting
the
Hi,
I just noticed (when looking at a local modification and realising my IDE had
added tabs to a file using spaces) that the OSG repo doesn't contain an
EditorConfig file. This is basically a file that instructs IDEs and editors of
a project's whitespace style, overriding the user's settings
Thanks Voerman. I'd looked through the examples for other things, but not for
multiple viewports, so missed that. It seems to be a good starting point once I
switched the GLSL extension from the NV variant to the ARB one. Despite not
being core until OpenGL 4.1, it looks like multiple viewports
Hi,
It looks like this has fallen off the radar again as it's been a month. I'd
still rather fix this in a robust way rather than making a guess as to the most
sensible approach and creating maintenance problems down the line.
Thank you!
Cheers,
Chris
--
Read this topic
When the widget is popped out, it causes the creation of a new OpenGL context.
I see that you've told Qt to share things between the old and new context,
which means you'll still be able to use some things in the new one, but I'd
look into what can and can't be shared as it isn't everything.
Hi,
I'm investigating using a geometry shader to render multiple shadow map
cascades in one pass in OpenMW. While I've heard conflicting (but mostly
negative) accounts of how much additional performance this can bring, I reckon
it's likely to help, as OpenMW uses a ridiculous quantity of tiny
Hi,
Most of my testing was with a ShapeDrawable with a sphere, and that uses quad
strips. However, I've also tried it with some of OSG's test models, such as
spaceship.osgt, and that specific example only uses triangles and is broken,
too.
Cheers,
Chris
--
Read this topic
Hi,
I've been playing around with this, and have an annoying issue. On my machine,
which runs Windows and has a Radeon Vega 56, using three viewports with
geometry shader instancing makes some triangles go missing in the second and
third viewports. This happens in my own application and in a
Hi,
You should be able to dynamic_cast or asCullVisitor the arguments to a regular
node callback to get a cull visitor pointer, and then from that, you can access
the camera. You don't need to have a version with a boolean return value,
either. Just don't traverse the node's children from the
Hi,
I think they're talking about the C preprocessor definition, FLT_MAX rather
than anything specific to anything used here:
https://www.tutorialspoint.com/c_standard_library/float_h.htm
It's the maximum value of an IEEE 32-bit float.
Cheers,
Chris
--
Read this topic online
Hi,
Apparently, that approach won't work here - in my actual application, the
multiple views are managed with a composite viewer, and that only calls its
cleanup operation in its destructor and not when views are removed. I'll still
be able to make something work, but it would have been nice
Hi,
When the text node's releaseGLObjects is called, it calls the same method of:
The
GlyphQuads
The
Font (which is potentially problematic when it doesn't get skipped).
The vertex attributes (
_coords, _normals, _colorCoords, _texcoords)
The primitives in
Hi,
Turns out that it's not a font in the object cache that's causing me grief
after all, but instead the static default font here:
https://github.com/openscenegraph/OpenSceneGraph/blob/OpenSceneGraph-3.6/src/osgText/Font.cpp#L40.
This is initialised from a GLubyte array instead of a file, so
Hi,
I don't want to be the guy that says "Go away and Google your problem", but I
was curious and searched for OpenGL on headless Windows and saw a few results
where people had managed to do it. Even if you don't find something compatible
with your situation, you're likely to find more and
Hi,
This surprises me, but apparently clearing the object cache is enough to stop
anything from leaking (otherwise, in the case where the second text node was
removed, too, without releasing its GL objects, CodeXL was reporting two leaked
shaders and three leaked program pipelines). I was
Hi,
I've been putting off responding to the last post as I wanted to be sure I
wasn't misunderstanding you, but after several rereadings, I still think you've
missed my point. I'll try explaining the issue again, but I'm not sure which
part isn't clear, so I don't know how much it'll help.
In
Hi,
I've only tested that example against the 3.6 branch (as of commit 4b6d9287,
which was the latest yesterday), and while it doesn't actually crash or show
errors, putting tracepoints in shows that the shader program is recompiled and
the glyph texture is added to the orphaned list once the
Hi,
Hopefully this example illustrates the underlying problem with
osgText::Text::releaseGLObjects without the multiple viewer schenanigans in my
previous example:
Code:
#include
#include
#include
#include
int main()
{
osgViewer::Viewer viewer;
// Single-threaded mode so we
Hi,
ShaderGen from 3D Labs is an old tool that can generate equivalent GLSL 1.2
shaders for a fairly large subset of FFP state. I've used it a fair amount to
sanity-check things. If you just want a basic shader as a start point to drop
into an FFP application, it's pretty good, but there are
Hi,
Fixed-function materials can track either the material colour or the vertex
colour depending on if glColorMaterial is used, so you can probably get the
behaviour you want by setting a material on the stateset as before, but
changing its mode so that (for example) the diffuse and ambient
Hi,
Did my last message disappear because of the forum issues? It's been a while
and there's no reply yet...
Cheers,
Chris
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=76596#76596
___
osg-users
Here it is again:
AnyOldName3 wrote:
> Hi,
>
> Hopefully, after that two-week hiatus, we can get this sorted.
>
>
> > You can always subclass from Camera and override the destruct if you want
> > to add this "fix" if it suits your usage case.
>
>
> I'm already going to have to include a
Hi,
It's been a week, so I'm just checking this hasn't been forgotten.
Cheers,
Chris
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=76638#76638
___
osg-users mailing list
Hi,
I'm trying to build a stripped-down version of the same behaviour as my
application, but it's got extra issues because things that are handled by what
I'm stripping out aren't being handled when they're gone.
What's the correct process for adding and removing views to and from an
Hi,
The reason I couldn't add or remove views during the event traversal was that
it was invalidating iterators that iterated over the views. Instead, I'm adding
an update operation to add or remove the viewer.
I now have a fairly minimal example that replicates my use case and bug:
Code:
Hi,
> Statusswitches are always costly. How does OSG identify if a texture or a
> program is identical when
> checking for status changes? Is it by comparing the pointers? If yes, it
> is necessary to manage a pool
> and use the same pointers everywhere?
StateSet::compare takes an
Hi,
Is the plan still to have OSG itself release GL objects in the object cache
when the contexts they're associated with are destroyed, and if so, have you
had any further thoughts about how this might be accomplished?
The problem is still that there's nothing in osgViewer that has the same
Hi,
As we've discussed, that only occasionally helps. In the boring single-viewer,
single-view case, it works, but no one would notice there was a problem there
anyway as nothing would actually render incorrectly.
My application just uses the camera created for it by OSG when an
Hi,
You need to use shadow2D instead of texture2D to sample a sampler2DShadow in
older GLSL versions.
Cheers,
Chris
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=76668#76668
___
osg-users mailing
Hi,
As someone who's added shadows to an OSG application, I can confidently say
you're going to have a bad time if you try and use anything in osgShadow. I was
told this before I'd really got started, and I didn't believe it, so continued
anyway, and while I did end up with impressive final
Hi,
Have you tried playing with osg::Material::setColorMode? It controls
glColorMaterial settings, which is used to switch between having the material
colour be determined by the global colour or vertex colours. You probably had
the Material set to use its own colour, but if you re-add it and
Hi,
> So... would change the destructor to:
>
>
> Camera::~Camera()
> {
> setCameraThread(0);
>
> if (_graphicsContext.valid())
>
> {
>releaseGLObjects(_graphicsContext->getState());
>
>
> _graphicsContext->removeCamera(this);
> }
> else
>
Hi,
You can't release objects that belong to a destroyed context, so you definitely
need to call all the releaseGLObjects methods before the context is destroyed.
You shouldn't need to call it on the root node provided the root node is still
attached to the viewer when the viewer and context
Hi,
Now Robert seems to be back from his break, I'm just going to mention the
potential footguns I found here in case they need sorting out. I've got a
reasonable solution for the issue I was having, so that doesn't really need
discussing any more, but it seems sensible to me that OSG might
Hi,
Now Robert seems to be back from his break, I'm just going to mention the
potential footguns I found here in case they need sorting out. I've got a
reasonable solution for the issue I was having, so that doesn't really need
discussing any more, but it seems sensible to me that OSG might
Hi,
This is almost certainly due to one of the issues I've outlined here:
http://forum.openscenegraph.org/viewtopic.php?p=76471#76471
The standard viewer definitely doesn't clean everything up that it's supposed
to, as demonstrated by the examples I gave in that thread. Specifically, this
Hi,
Try using the GUI version of CMake - if you select the compiler you want to use
as the generator, it'll try looking for that compiler instead of defaulting to
NMake makefiles, which usually use Visual C++.
Cheers,
Chris
--
Read this topic online here:
Hi,
Annoyingly, while that branch fixes the issue in my examples, it isn't fixing
it in my actual application. The cache gets cleared when the application exits
(potentially after all OpenGL contexts have been closed), but not when views
are added to and removed from the composite viewer, so
Hi,
I have more information. The GraphicsContext is only referenced by the camera
when the camera is destroyed. The camera detaches itself from the graphics
context in its destructor, then the camera's ref_ptr to the graphics context is
destroyed, triggering the destruction of the graphics
Hi,
> Are the Camera's being removed from the Window before it gets closed? I have
> just checked the GraphicsContext::close() method and it calls all the
> Camera::releaseGLObjects(), which in turn will call the
> Renderer::releaseGLObjects(), which will now call the
>
Hi,
Is it intentional that using osg::CopyOp::DEEP_COPY_USERDATA copies the
UserDataContainer, but doesn't deep-copy the actual things in it unless
osg::CopyOp::DEEP_COPY_OBJECTS is also enabled? This means that things can be
added or removed, but changes to existing things are shared.
We
Hi,
Are you definitely creating an OpenGL context with either a version less than
or equal to 3.0, or as a compatibility profile? This sounds to me like you've
got a core profile, so don't have access to old deprecated stuff, but are
trying to use it anyway.
Cheers,
Chris
--
Hi,
Somehow I missed the several parts about you intentionally using a core profile.
I guess the most reasonable thing to do is submit an issue report that
https://github.com/openscenegraph/OpenSceneGraph/blob/OpenSceneGraph-3.6/src/osg/ClampColor.cpp#L52-L53
should be fenced off so they don't
Hi,
Is there a way to copy something if all I have is a ref_ptr to Referenced? I
see that Object has a clone method, but Referenced doesn't. This poses a
potential problem with deep-copying user data, as
osg::DefaultUserDataContainer::_userData is a ref_ptr.
Cheers,
Chris
--
Hi,
I'm not seeing a particularly easy way to deep-copy the rest of the stuff in
osg::DefaultUserDataContainer, either. The description list is a vector of
strings, which are deep-copied, so that's fine, but the object list is harder
as I'd need to cast the const off the CopyOp if I were to
Hi,
Good, but would the new enum entry be for the current behaviour or the new
behaviour? If the current behaviour is a bug, then it makes sense to only keep
the old behaviour as the new enum value so everyone gets the fix, but
conceivably there could be applications that rely on the current
Hi,
How do you want backwards compatibility to be handled? The current behaviour
could just be dropped, or it could be available under a new flag name like
SHALLOW_COPY_USERDATA (potentially with the same value as the current deep copy
one), or the new behaviour could be under a new flag name
Hi,
Yeah, they're pretty obviously deprecated as they control fixed-function stuff,
and the whole FFP was deprecated ages ago.
Cheers,
Chris
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=76737#76737
Hi,
I still think you're completely missing my point because of a criticism of my
example code. For now, ignore the symptoms that brought me here. Just consider
your bog-standard single-viewer situation.
If a text node is added to the scene graph and stays attached for the whole
lifetime of
Hi,
It's been pointed out to me that the DefaultUserDataContainer object list can
be a little bit more deeply copied with this change:
Code:
--- a/src/osg/UserDataContainer.cpp
+++ b/src/osg/UserDataContainer.cpp
@@ -56,7 +56,10 @@ DefaultUserDataContainer::DefaultUserDataContainer(const
Hi,
It might not necessarily be fast, as you'd be changing state that would
normally stay the same between consecutive draw calls, but you could also
change the scissor rectangle to the region you want to keep, or you could draw
a rectangle with the same depth as the text, then set the depth
Hi,
The name texUnit0 comes from the shader, which you're providing. The texture
unit index (which my example snippet sets to 0) comes from your scene graph,
which you said came from osgOcean, and will almost certainly be 0 for a diffuse
map. If I was trying to check that, I'd run the
Hi,
I'm pretty sure you're not using your shaders at all. You're not setting
gl_Position in the vertex shader, so it's not going to compile, and OSG will
fall back to either its basic shader or fixed-function mode depending on how
you've set it up. You should be seeing errors printed to the
Hi,
OSGOcean is probably setting up the necessary fixed-function stuff to put its
texture in texture unit zero, but you'll need to tell the fragment shader that
its texUnit0 uniform refers to that texture unit. It should be enough just to
add a uniform to the same stateset as you're adding
Hi,
If you're not seeing any errors in the console, I don't think you've added your
shader to the scene properly. You have to assign something to gl_Position
otherwise the rasteriser has no idea where your vertex is, so doesn't know
which pixels any triangles cover, and nothing will get drawn.
Hi Robert,
OpenMW still experiences some regressions when built with OSG 3.6.x instead
of 3.4.1. It's completely possible they're because we're trying to coerce
OSG systems to do things they weren't intended to as it's a big project
created without much interaction with the OSG community.
The
On Wednesday, 18 December 2019 10:56:39 UTC, Robert Osfield wrote:
>
>
>
> On Tuesday, 17 December 2019 21:40:37 UTC, Chris Djali / AnyOldName3 wrote:
>>
>> Hi Robert,
>>
>> OpenMW still experiences some regressions when built with OSG 3.6.x
>> inst
Hi,
osg::Material pretty much encapsulates OpenGL's fixed-function material
state, so things like glMaterialfv (that controls material settings for
whole draw calls) and glColorMaterial (that controls what vertex colours
actually affect). If you're using a suitably old GLSL version, you can
simple solution shows up for anyone searching
for this in the future.
Cheers,
Chris
On Friday, 11 December 2020 at 19:38:26 UTC Chris Djali / AnyOldName3 wrote:
> Hi,
>
> I'm working on a node visitor that converts FFP state to shaders and am
> extending it so it can handle more
74 matches
Mail list logo
