[osg-users] Unwanted culling in 3.6.4 vs 3.5.1
Anders Backman
andersb at cs.umu.se
Thu Jan 9 06:06:26 PST 2020
Problem solved:
obj->setAutoScaleTransitionWidthRatio(0.001f); // Was 0 earlier.
This seems to be something that changed between the two versions!
Now it works.
Thanks for the feedback!
/Anders
On Thu, Jan 9, 2020 at 3:02 PM Anders Backman <andersb at cs.umu.se> wrote:
> [image: image.png]
> Actually, I managed to reproduce it, sort of in a sample application.
>
> When running this exact code in 3.6.4 I get this:
>
> A quick blink (of a white box), then nothing and lots of culling warnings:
> CullVisitor::apply(Geode&) detected NaN,
> depth=nan, center=(0 0 7.125),
> matrix={
> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
> -0.5 -7.125 -45.6982 1
>
>
>
> But with 3.5.1 I get:
>
> [image: image.png]
> Which is what I would expect.
> Anyone that can reproduce this? No models needed, just compile the code
> and run.
>
> Cheers,
> Anders
>
>
>
> On Thu, Jan 9, 2020 at 2:56 PM Anders Backman <andersb at cs.umu.se> wrote:
>
>> That is so strange..!
>>
>> > osgversion
>> OpenSceneGraph Library 3.6.4
>>
>> Captured a short video of the issue:
>> osgviewer --window 0 0 1024 768 osg_3.6.4_culled.osgt
>>
>> https://gofile.io/?c=A6LCw6
>>
>> What on earth could be the difference here? Have tested on several
>> computers (only Windows though).
>> Windows 10, NVida GeForce card.
>>
>> Also, how it works in 3.5.1:
>>
>> ">"osgversion
>> OpenSceneGraph Library 3.5.1
>>
>> osgviewer --window 0 0 1024 768 osg_3.5.4_not_culled.osgt
>> https://gofile.io/?c=i2Ssx0
>>
>> On Thu, Jan 9, 2020 at 2:00 PM Voerman, L. <l.voerman at rug.nl> wrote:
>>
>>> Hi Anders,
>>> I did have a quick look, but I can't see any unwanted culling, nor do I
>>> get the warnings you write about.
>>> tried versions 3.6.3 3.6.4 and 3.6.5 on windows with
>>> OSG_NOTIFY_LEVEL=INFO
>>> osgviewer osg_3.6.4_culled.osgt
>>>
>>> Looking in the file I see a lot of empty osg::Text nodes - I guess that
>>> causes an empty bounding sphere in the specific version of osg you have.
>>> Laurens.
>>>
>>> On Thu, Jan 9, 2020 at 1:19 PM Anders Backman <andersb at cs.umu.se> wrote:
>>>
>>>> Another issue I discovered with 3.6.4 is that we now suddenly get
>>>> unwanted culling.
>>>> At first it looks like a small feature culling thing (which we disable
>>>> at global level with:
>>>>
>>>> // Don't do small feature culling
>>>> osg::CullStack::CullingMode cullingMode =
>>>> m_viewer->getCamera()->getCullingMode();
>>>> cullingMode &= ~(osg::CullStack::SMALL_FEATURE_CULLING);
>>>> m_viewer->getCamera()->setCullingMode(cullingMode);
>>>>
>>>>
>>>> However, this does not look like a small feature thing to me at all.
>>>> Actually, it culls even when you get close to the objects.
>>>> Attached are two osgt-files.
>>>>
>>>> One is saved from OSG 3.5.1 (works as intended).
>>>>
>>>> Second one is saved from OSG 3.6.4 (where we get the unwanted culling).
>>>>
>>>>
>>>> If you use a later version of OSG (3.6.4) the one from 3.6.4 generates
>>>> lots of warnings:
>>>>
>>>> CullVisitor::apply(Geode&) detected NaN,
>>>> depth=nan, center=(0 0 7.125),
>>>> matrix={
>>>> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
>>>> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
>>>> -nan(ind) -nan(ind) -nan(ind) -nan(ind)
>>>> 0.187249 -0.470484 -6.20285 1
>>>>
>>>> whereas the one from 3.5.1 does not.
>>>>
>>>> Anyone give me a hand on this? I am really stuck.
>>>> I was first totally into small feature culling, trying to dig up old
>>>> code where people was trying to disable small feature culling on subgraphs
>>>> etc. But that is not the issue here.
>>>>
>>>> Any suggestion would help. I tried to make the scene as small as
>>>> possible.
>>>>
>>>> I tried to attach the files, but it made the message too big.
>>>> Instead I shared the files using gofile: https://gofile.io/?c=M5xPmU
>>>> There are two files osg_3.5.1_not_culled.osgt
>>>> and osg_3.6.4_culled.osgt
>>>>
>>>> Thanks,
>>>> Anders
>>>>
>>>>
>>>> --
>>>> __________________________________________
>>>> Anders Backman, HPC2N
>>>> 90187 Umeå University, Sweden
>>>> anders at cs.umu.se http://www.hpc2n.umu.se
>>>> Cell: +46-70-392 64 67
>>>> _______________________________________________
>>>> osg-users mailing list
>>>> osg-users at lists.openscenegraph.org
>>>>
>>>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>>>>
>>> _______________________________________________
>>> osg-users mailing list
>>> osg-users at lists.openscenegraph.org
>>> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>>>
>>
>>
>> --
>> __________________________________________
>> Anders Backman, HPC2N
>> 90187 Umeå University, Sweden
>> anders at cs.umu.se http://www.hpc2n.umu.se
>> Cell: +46-70-392 64 67
>>
>
>
> --
> __________________________________________
> Anders Backman, HPC2N
> 90187 Umeå University, Sweden
> anders at cs.umu.se http://www.hpc2n.umu.se
> Cell: +46-70-392 64 67
>
--
__________________________________________
Anders Backman, HPC2N
90187 Umeå University, Sweden
anders at cs.umu.se http://www.hpc2n.umu.se
Cell: +46-70-392 64 67
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20200109/01c3ee2a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 23601 bytes
Desc: not available
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20200109/01c3ee2a/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 24890 bytes
Desc: not available
URL: <http://lists.openscenegraph.org/pipermail/osg-users-openscenegraph.org/attachments/20200109/01c3ee2a/attachment-0001.png>
More information about the osg-users
mailing list