I am designing the xith3d engine to be agnostic to both LWJGL and JOGL and am attempting to design the interaction between my lowest level render peers and the higher level remderer logic.
I started looking over JOGL in more depth and had some questions:
-
Where is the documentation located? I pulled what is in CVS but there isn’t really much there. The userguide.html is just a one page overview. The two demo programs also don’t give you too many clues as to how it is to be used.
-
What exactly should we be doing in the init method? In the gear demo they are drawing geometry, in other places I have seen what looks like frame setup stuff (projection matrix, clearing background). But I have also see the GL and GLU handles being fetched from the drawable interface.
-
Where is the documentation for GL and GLU? In particular I would like to see the options for passing geometry to the interface (NIO Buffers, Arrays, etc). I have plenty of docs on the opengl interface in general, what I need is the specifics (preferablly javadoc) for the JOGL interface to those methods. For example LWJGL 0.6 uses system addresses against direct NIO buffers for passing in buffers, which is extremely fast.
-
What drove the decision to force AWT to underly the canvas implementations? Is it possible to somehow abstract from that so we can have a third drawable (heavyweight componant, lightweight componant, OS level window).
-
How have you found the performance of gathering mouse events through the AWT event mechanism? I have found that in general WLToolkit under windows just sucks the life out applications, which is why the polling mechanism of LWJGL is attractive to many people.
-
How stable is the current release of JOGL right now under Lunix, Windows and MacOS?
Thanks!