Eric,

I believe that you have hit the nail on the head. My misuse of
OSG::MatrixFrustum() caused me some headaches a couple of weeks ago, but I
thought I had it fixed. After doing more careful tracing through the code
path for OSG::RenderTraversalAction(Base), I think that my use of
OSG::MatrixCameraDecorator (and consequently OSG::MatrixFrustum()) appears
to be biting me. I may have to go back to the design drawing board to get
this sorted out. Thanks for the tip.

 -Patrick

Eric Maslowski wrote:
> Hi Patrick,
>   I may be missing the overall question, but we've run into some frustum
> problems (see below) and have tracked down it's assignment to the camera
> decorators. More specifically, MatrixFrustum(). I can't say whether this is
> the same place the frustum is set for intersection tests, but hope it helps
> with your search.
> 
> Previous frustum problems:
> http://www.mail-archive.com/[email protected]/msg08487.html
> 
> Cheers
> 
> E.
>  
> ---
> Eric Maslowski
> Research Computer Specialist
> University of Michigan 3D Lab
> 
> Autodesk 3D Studio Max Certified Trainer
> 
> email:  [EMAIL PROTECTED]
> office: 734-615-9699
> mobile: 734-730-9904
> 
> 
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Patrick
> Hartling
> Sent: Thursday, October 18, 2007 19:03
> To: users
> Subject: [Opensg-users] Frustum disparity between OSG::RenderTraversalAction
> instances in cluster scenario
> 
> I have been digging more into the problem that I described in a message that
> I posted last week (subject "OSG::Group core and clustering oddity"), and I
> have ruled out OSG::Group and its subclasses being a factor. I don't think
> there is anything about PyOpenSG that is causing the problems, either. My
> current suspicion is that I need to do some more work in the C++ render
> server application, and the lead I am following now has to do with
> OSG::RenderTraversalAction and how its frustum is set up.
> 
> In my client GUI, I have a valid frustum in the OSG::RenderTraversalAction
> instance, and everything renders well. In the render servers, however, the
> OSG::RenderTraversalAction frustum has near and far planes that seem totally
> unrelated to what is being set up on the client side. In particular, the far
> plane appears to be invalid with a normal of the form [nan, nan, nan] and a
> distance from the origin of -inf.
> 
> I understand that OSG::RenderTraversalAction is not a field container, and I
> am creating the instances manually on both ends of the communication
> channel. However, I am not doing anything directly that sets the frustum for
> an instance of OSG::RenderTraversalAction, and I don't know how it gets set.
> Not knowing this is making it hard for me to figure out why the render
> servers have what seems to be a completely bogus frustum. How is the frustum
> supposed to get set? I have been tracing through the code when the traversal
> action is applied to the scene graph, and I see where the frustum is used
> for intersection testing. What I have not yet seen is where, how, or when
> that frustum is set. Is this something that I should be doing in my render
> server code based on information that is received from the client GUI via
> the scene graph sharing?
> 
>  -Patrick
> 
> 
> --
> Patrick L. Hartling
> Senior Software Engineer, Priority 5
> http://www.priority5.com/


-- 
Patrick L. Hartling
Senior Software Engineer, Priority 5
http://www.priority5.com/

Attachment: signature.asc
Description: OpenPGP digital signature

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users

Reply via email to