Hi,
I'm still getting the same problem running the simple example on Windows 7, I have the feeling that it comes from a mistake I made but I can't find which one. The sources from the github repository don't exactly match the code, I have a breakpoint line 81 of javax.media.j3d.Group and it doesn't refer to a code line in sources : https://github.com/hharrison/java3d-core/blob/master/src/classes/share/javax/media/j3d/Group.java I think it would be usefull for many people to have sources that match with the code so if you can and want to generate them Harvey, maybe it would help ;) Thanks anyway to everyone for your support. Antoine |
Administrator
|
On 05/31/2012 10:30 AM, ylliac [via jogamp] wrote:
> > I think it would be usefull for many people to have sources that match with > the code so if you can and want to generate them Harvey, maybe it would help ;) 'Soon' we will have automated builds & tests by our Jenkins server, then we can provide matching builds and source code. ~Sven signature.asc (910 bytes) Download Attachment |
OK then I can wait until 'Soon' to go further with the new Java3D :)
Antoine |
In any event, I'll produce a matching binary and source jar when I get back to my main machine and let you know
when it is available...sorry for the inconvenience. |
No problem.
As I said before, it's probably a problem in my configuration. I don't NEED to make it work right now so don't worry if you can't generate the sources before the Jenkins server is set. Antoine |
I've gone ahead and redone the build of the prebuilt jars and some source zipfiles to make sure they match
this time, please let me know how well (or not) they work for you, uploaded to my github pages same as last time except there is a -src zip file for each jar file now. Cheers, Harvey |
Thank you for the update Harvey.
I can't download the jar for j3dutils but the others files helped debugging and I found that the problem is in GraphicsConfigTemplate3D.runMonitor(int action) : static void runMonitor(int action) { // user thread will locked the globalLock when Renderer // thread invoke this function so we can't use // the same lock. synchronized (monitorLock) { switch (action) { case J3dThread.WAIT: // Issue 279 - loop until ready while (threadWaiting) { try { monitorLock.wait(); <----- IT WAITS FOREVER :) } catch (InterruptedException e) { System.err.println(e); } } break; case J3dThread.NOTIFY: monitorLock.notify(); threadWaiting = false; break; } } } It seems that the monitorLock is never unlocked and I don't know why. Any idea ? I found that J3dThread.NOTIFY should be sent by the Renderer in the doWork() method, but if the program is blocked, the renderer is probably not launched ? Antoine |
Hmm, I'll try to find time to look at that near the end of the week.
I reuploaded the j3dutils jar..no idea what happened. Cheers, Harvey |
In reply to this post by ylliac
Hi Antoine, please let me know if you've had any better luck - I've had the same symptoms (but under Ubuntu 12.04 64-bit) - without much luck getting past the point where the app locks up.
|
Hi,
I still have the same problem because and I didn't go further in looking for the solution. I don't know if Harvey had time to look at this ?
Antoine 2012/7/18 elotter [via jogamp] <[hidden email]> Hi Antoine, please let me know if you've had any better luck - I've had the same symptoms (but under Ubuntu 12.04 64-bit) - without much luck getting past the point where the app locks up. |
OK, should be great if a solution presents itself - it seems like we are having very similar symptoms under different OSes.
My use case should be interesting if I can get it working - the 3D view will be embedded in a NetBeans Platform application - once it works, I'll certainly write a tutorial of sorts showing how it works (and this may give this Java3D fork more publicity). Regards, Ernest |
Antoine, you also on JDK7u6 64-bit?
|
@harvey : I'll be happy to test with a debug .jar if you want to get more details on what's happening at the point where things get stuck. If so, please specify which jogl version I should use with the set.
Regards, Ernest |
In reply to this post by elotter
I'm on jdk6_33 32bits but on a 64bits computer. Le 19 juil. 2012 06:24, "elotter [via jogamp]" <[hidden email]> a écrit :
Antoine, you also on JDK7u6 64-bit? |
Administrator
|
In reply to this post by elotter
JOGL 2.0 RC9 should be fine. You can enable Java and native debug, it would be good for us:
-Dnewt.debug=all -Dnativewindow.debug=all -Djogl.debug=all
Julien Gouesse | Personal blog | Website
|
No problem, will do so.
|
This post was updated on .
In reply to this post by gouessej
Here's the point where it hits:
3D [dev] 1.6.0-pre1-daily-experimental daily AWT-EventQueue-0 - Info: NativeWindowFactory. |
I tried it again, this time additionally with -Dsun.java2d.opengl=true on, and got (on 64-bit Ubuntu 12.04):
3D [dev] 1.6.0-pre1-daily-experimental daily AWT-EventQueue-1 - Info: NativeWindowFactory. |
Administrator
|
CGLSurfaceData is only present on Mac, it should not be used under GNU Linux:
http://jausoft.com/git/?p=jogl.git;a=patch;h=5b2803ebb3ac8957f7b33f9bfd74c085e9ab720a
Julien Gouesse | Personal blog | Website
|
That's pretty weird, as I have only the 64-bit Linux JRE (7u6) with linux-amd64 natives for JOGL - why would it make these references to OS X?
|
Free forum by Nabble | Edit this page |