Do you really need a i386 based Mac to build a universal binary??  I was under the impression that the intel cross-compiler was included with the latest XCode 2.2.  Am I mistaken?

anybody working on the JOGL JNI lib for OSX/x86?

JOGL has been working great as a cross-platform way of generating OpenGL contexts for handing off to other native modules.

However Jogl (1.1b07) seems to be leaking about 0.5 or 1.5 MB of native memory for each GLCanvas that is created, used, and destroyed on Win32 and Mac respectively.  Progress can be tracked using "ps" on the Mac and the task manager process list on Windows.  Eventually the process runs out of memory and dies.  The Linux implementation doesn't has this problem...memory usage is constant and the program runs indefinitely.

I've built JOGL on Win32 and activated the DEBUG code to confirm that the destroy methods are being called (destroy & destroyImpl) as expected when the GLCanvases are removed from their containers.  Runtime.getRuntime().totalMemory() and Runtime.getRuntime().freeMemory() also don't grow, hence the basis for my suspicion that itis native memory that's leaking somewhere in the platform-specific implementations...

Before I go through the trouble of boiling this down to a simple test case for a bug report, is anyone already familar with this issue and know of a solution?  Or, is this the expected behavior with no workaround likely?

Sure thing!  I'll put it up in a bit.  In the mean time, gziemski, have you tried running my JOGL version?  What was your result?

Edit:  I am assuming you are running on a Mac, of course!

I decided to go the tedious route and write a C implemetation using GLUT and sure enough, I get only a white square.  It looks like the Mac doesn't properly implement the latest shader extensions (Even though both or my OS X cards claim 1.5+ compliance and list all of the required extension strings!).

Can any of you other Mac users out there run the above code and post your results, just so I can get a few more data points to further justify my results?

Hi Corwin,

This is a thread from the lwjgl forum that might be of a little help.
It looks like they aren't supported on 10.2 and in the thread, they thought they might be in 10.3, but maybe they aren't there yet?

No, they're supported on the OS, and I can write C that demonstrates that.

Also, Render-To-Texture doesn't work either. Has *anybody* gotten these to work?

I note that in the documentation it is mentioned that GLJPanel doesn't work- can anybody tell me what the problem with it is?

Also, how will the performance with this be? I would imagine it would be terrible, but maybe my understanding is incorrect. There's no way to get an actual GL context on the screen with a GLJPanel (since it can't draw in order with all the other Swing components) so it has to do a glReadPixels and then display that image through Swing. Is there something I'm missing, or would this just be excruciatingly slow?

Thanks!

Corwin

Thank you very much for your replies!

From what you mentioned, i recon that there should actually no comparing Java3D to JOGL, cause they both suit different needs on different levels.
So far, i hope you don't mind if i ask some questions:

.
.
.

5. So far, Java3D is (and i guess it will, cause it is not supported any longerr by sun, or is it again?) only supported on Windows and Solaris. But JOGL is supported on much more platforms. That seems to me like a very big advantage.

ohhhhh, thanks for the update Gerard!

So, I suppose the developer update will be made available from the 28th onwards?  

In general, how are things going with the implementation of hardware accelerated Java2D commands?  What else are you working on at the moment or aren't I allowed to ask.  

Cheers,

ribot.