Hi Dan, SceneView is part of the original beginnings of the OSG, when it was single threaded, single context, and is now scheduled for deprecation. It will eventually be replaced entirely by functionality in osgViewer::Renderer, but for now I've simply reused SceneView.
With any project that is a decade in the making you'll find a lot of history of it's evolution still in it, it's partly down to reusing code that does the job, and partly about maintaining backwards compatibility during transitions. So if you want to learn about the scene graph design and implementation keep this it mind that is living project, it's not only evolved alot through it's history, but it's still evolving a good old rate. As for creating a new scene graph in another language, I guess the OSG isn't too bad a place to start, but if you want to develop a really successful project you'll need under stand about wider project management as well as just the technical aspects of design and implementation. It also take a lot of work to write a full featured scene graph, hundreds of thousands of lines of code, and many man hours. Robert. On Sat, Nov 7, 2009 at 12:46 PM, Dan R <[email protected]> wrote: > Hello, I'm new to OSG and am just looking over the code structure to > better learn how a proper scene graph library is laid out. I'm > interested in using (porting a stripped down version) of osg for a > project I'm working on in another language. I pretty much understand > the flow of the graph, however, when it comes down to the actual > rendering traversal I get a bit lost. So I have these two questions, > but maybe being directed to some older posts would be good answers... > > 1) The osgViewer::Renderer class appears frequently throughout the > code, however, it seems to be based around the osgUtils::SceneView > class. The SceneView class says that it's deprecated. Does this also > mean that the Renderer class is deprecated, or simply that any calls > to SceneView are no longer used in the normal work flow? > > 2) In a SingleThreaded case, where are the operations that result in > rendering the Drawables get added? I see that in the > renderingTraversals method, it first traverses the graph with a > getBound() call, then goes through each camera and does a cull() (this > uses a Renderer object from my first question), then goes through each > GraphicsContext and calls runOperations. Where I get lost is, where > are the operations for the GraphicsContext defined and added, that > actually end up doing the work of rendering the drawables? > > Any direction in this case would clear a lot up for me, thanks, > > Dan > _______________________________________________ > osg-users mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
