pyopengl-users

[PyOpenGL-Users] OpenGLContext 2.0 on the way, status update, requests?

[Mike C. F. <mcf...@ro...>]

I've spent the last few weeks working on OpenGLContext, and am getting 
to the point where an early 2.0 release is becoming visible somewhere on 
the horizon.  This message will serve as a heads-up regarding the new 
development work.  I'm also wondering if there are particular things 
that people would like to see in the package's next iteration.  I've 
been working primarily on the scene graph package (and a few new 
supporting packages)

Here's a quick summary of what's changed...

    * have moved to 2.2.x+ (the new stable Python version)
          o has significantly simplified the code base
          o have moved away from the mcf.vrml code base for providing
            VRML functionality, am using a much smaller custom-written
            vrml package (including a more robust parser).
          o nodes are now new-style Python objects with fields being
            property sub-classes
          o means that anyone using the package needs to upgrade to
            2.2.x (preferably 2.2.2), though the old version (1.0) will
            continue to be available for earlier Python versions.
    * have moved to a visitor pattern for all rendering traversals
          o means that the rendering algorithms tend to be localized in
            a single file, with only small bits of functionality added
            to the individual node-classes
    * have just completed the base work for a stencil-shadow-rendering
      pass-set and supporting objects
          o is basically functional, though there are a number of fairly
            common cases which need to be dealt with before it can be
            considered finished
    * have added support for more of the VRML97 background node
    * threaded image loading is working
    * the new scene graph implementation includes generic caching
      facilities (based on field-watchers, which are in turn based on
      the new field classes and the dispatcher module)
    * lights are now properly positioned within the scene graphs
    * switches are implemented
    * integrated node-path objects provide both forward and backward
      manual transformation-matrix calculations
    * there are display list and texture helper objects for managing
      resource life-cycles
    * there are a few new debugging tools
          o save the depth buffer or the stencil buffer to a PIL image
          o log OpenGL.GL commands to sys.stderr
          o helpers to track changes to OpenGL state

I'm probably still a few weeks away from a first 2.0 alpha release, so 
there is still time for feature requests.  Let me know what you want to 
see (within reason, I'm not looking to double the length of the project 
or anything).

Enjoy yourselves,
Mike

_______________________________________
  Mike C. Fletcher
  Designer, VR Plumber, Coder
  http://members.rogers.com/mcfletch/

Re: [PyOpenGL-Users] OpenGLContext 2.0 on the way, status update, requests?

[<il...@ya...>]

Excellent!

Sounds really cool.  I hope you can get the shadow
volumes are in the release?

The logging of OpenGL.GL commands sounds really useful
too :)

What is a visitor pattern?



 --- "Mike C. Fletcher" <mcf...@ro...> wrote: >
I've spent the last few weeks working on
> OpenGLContext, and am getting 
> to the point where an early 2.0 release is becoming
> visible somewhere on 
> the horizon.  This message will serve as a heads-up
> regarding the new 
> development work.  I'm also wondering if there are
> particular things 
> that people would like to see in the package's next
> iteration.  I've 
> been working primarily on the scene graph package
> (and a few new 
> supporting packages)
> 
> Here's a quick summary of what's changed...
> 
>     * have moved to 2.2.x+ (the new stable Python
> version)
>           o has significantly simplified the code
> base
>           o have moved away from the mcf.vrml code
> base for providing
>             VRML functionality, am using a much
> smaller custom-written
>             vrml package (including a more robust
> parser).
>           o nodes are now new-style Python objects
> with fields being
>             property sub-classes
>           o means that anyone using the package
> needs to upgrade to
>             2.2.x (preferably 2.2.2), though the old
> version (1.0) will
>             continue to be available for earlier
> Python versions.
>     * have moved to a visitor pattern for all
> rendering traversals
>           o means that the rendering algorithms tend
> to be localized in
>             a single file, with only small bits of
> functionality added
>             to the individual node-classes
>     * have just completed the base work for a
> stencil-shadow-rendering
>       pass-set and supporting objects
>           o is basically functional, though there
> are a number of fairly
>             common cases which need to be dealt with
> before it can be
>             considered finished
>     * have added support for more of the VRML97
> background node
>     * threaded image loading is working
>     * the new scene graph implementation includes
> generic caching
>       facilities (based on field-watchers, which are
> in turn based on
>       the new field classes and the dispatcher
> module)
>     * lights are now properly positioned within the
> scene graphs
>     * switches are implemented
>     * integrated node-path objects provide both
> forward and backward
>       manual transformation-matrix calculations
>     * there are display list and texture helper
> objects for managing
>       resource life-cycles
>     * there are a few new debugging tools
>           o save the depth buffer or the stencil
> buffer to a PIL image
>           o log OpenGL.GL commands to sys.stderr
>           o helpers to track changes to OpenGL state
> 
> I'm probably still a few weeks away from a first 2.0
> alpha release, so 
> there is still time for feature requests.  Let me
> know what you want to 
> see (within reason, I'm not looking to double the
> length of the project 
> or anything).
> 
> Enjoy yourselves,
> Mike
> 
> _______________________________________
>   Mike C. Fletcher
>   Designer, VR Plumber, Coder
>   http://members.rogers.com/mcfletch/
> 
> 
> 
> 
> 
>
-------------------------------------------------------
> This sf.net email is sponsored by:
> With Great Power, Comes Great Responsibility 
> Learn to use your power at OSDN's High Performance
> Computing Channel
> http://hpc.devchannel.org/
> _______________________________________________
> PyOpenGL Homepage
> http://pyopengl.sourceforge.net
> _______________________________________________
> PyOpenGL-Users mailing list
> PyO...@li...
>
https://lists.sourceforge.net/lists/listinfo/pyopengl-users 

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com

Re: [PyOpenGL-Users] OpenGLContext 2.0 on the way, status update, requests?

[Mike C. F. <mcf...@ro...>]

Shadow volumes should be in the release w/out a problem, they seem to be 
working as reliably as anything within the system.  (Looks cool loading 
a complex VRML scene and seeing it shadowed in real-time, but it's still 
slow (I haven't done any optimisations in the scenegraph, either for the 
basic or the shadowed passes)).

Visitor pattern (in the strictest sense) is an external object which 
traverses a node-graph using a protocol which has the nodes dispatch the 
visitor to their children.  I'm using something similar, save that the 
only protocol the nodes follow is "get children of particular types", 
and the visitor does the actual broadcasting (which lets the visitor 
decide which children it wants to visit).  In the particular case, the 
rendering passes traverse the scenegraph from the context down, 
performing appropriate actions for the particular node when the node is 
reached.

Enjoy,
Mike

Rene Dudfield wrote:

>Excellent!
>
>Sounds really cool.  I hope you can get the shadow
>volumes are in the release?
>
>The logging of OpenGL.GL commands sounds really useful
>too :)
>
>What is a visitor pattern?
>
>
>
> --- "Mike C. Fletcher" <mcf...@ro...> wrote: >
>I've spent the last few weeks working on
>  
>
...

>>    * have moved to a visitor pattern for all
>>rendering traversals
>>          o means that the rendering algorithms tend
>>to be localized in
>>            a single file, with only small bits of
>>functionality added
>>            to the individual node-classes
>>    * have just completed the base work for a
>>stencil-shadow-rendering
>>      pass-set and supporting objects
>>          o is basically functional, though there
>>are a number of fairly
>>            common cases which need to be dealt with
>>before it can be
>>            considered finished
>>    
>>
...

_______________________________________
  Mike C. Fletcher
  Designer, VR Plumber, Coder
  http://members.rogers.com/mcfletch/