Is JOGL abandoned / dead?

I have been running into problems with JOGL and JDK11 and it seems that the JOGL project is not being maintained very much any more. Is this correct? No new releases in several years, and hardly any commits to the repo. I'm not complaining, just trying to understand the state of the project.

If JOGL is no longer being maintained, are there any alternative libraries that can render inside a JPanel? LWJGL does not seem to support that (unless you manually shuffle pixels around.)

Thanks,
-Farrell

Hi Farrell,

I can't speak to the state of the official releases, but I have recently fixed up JOGL to work better with Java 11. You can either compile my java-11-fixes branches (https://github.com/WadeWalker/gluegen/tree/java-11-fixes and https://github.com/WadeWalker/jogl/tree/java-11-fixes), or get the built JARs from my project at https://github.com/WadeWalker/com.jogamp.jogl.

Thanks, I'll check that out. Do you know if your fixes might have solved the high-dpi problems? For example, with Windows 10 display scaling set to 200%, my GLCanvas should fill the entire JPanel, but instead occupies only a quarter of it. (With JDK8, the GLCanvas occupies all of the JPanel, as it should.) This category of problem has been discussed in a few threads, including:

http://forum.jogamp.org/JOGL-broken-with-JRE-gt-8-and-Windows-window-scaling-td4039122.html
http://forum.jogamp.org/Jogl-Jogamp-on-Java-9-td4038012i100.html

Thanks,
-Farrell

Not sure about the high-dpi problems. There was one DPI-related API that I had to change to get rid of some Java 9+ warnings (see https://github.com/WadeWalker/jogl/commit/7aed16a5fbe316ab40f0ba371269c10e4406feef), but I'm not sure if it'll affect JPanel, you'd have to try it and see.

Hello

Please don't mention any competitors here, our forum doesn't have to become a billboard for them. Do you really think that this forum is the right place to claim that our project is dead?

JOGL 2.3.3 is still being prepared, you talked about a single repo whereas there are several contributors. We just try to avoid bothering Sven uselessly as he's very busy. Jenkins has been up for some months now, things are going better as time goes by.

Moreover, I remind you that I provided a patched version just to help a few individuals and I really appreciate Wade's efforts.

Seriously, there are some real bugs to fix but some developers shouldn't be worried about a few warnings.

Wade, if I find some time to get rid of the bug affecting Mesa (smartly this time, which is a challenge), do you think that we could plan to provide a release candidate?

I've only been working on Java 11-based patches -- would that be allowed in a release candidate? I'm not sure what the plan is there.

At first, I have to ensure that your changes don't break Java < 1.9 and to suggest some changes otherwise.

Then, you'll have to make a pull request against Sven's repository so that your changes will go into the "official" JOGL.

After that, I (or we?) will have to look at the known fixes and include some of them into some other pull requests and I'll have to provide a smarter fix for the bug affecting Mesa.

It's still possible to set Java 1.8 compliance when compiling with Java 11, but I'm not sure if the resulting build will actually work with Java 1.8, because to compile with Java 11 I had to remove -bootclasspath from our build.xml, so Java 1.8 may not be able to find these classes anymore.

There are also some unit test failures when running on modern OSes that would need to be worked through. There's the issue of GTK 3 compatibility on Linux, and there may be other problems with macOS 10.14 that I don't know about (my laptop only supports up to 10.13; I need to buy a new Mac :)). I also don't have any AMD GPUs, so any testing I do on Windows/Linux/Mac would be Nvidia-only. What I'm trying to say is that doing a high-quality new release would take a good bit of work, and would require hardware that I don't currently have available :)

Ok. Then, I'll try to build with Java 1.8. A nightly build with some fixes would be a more realist aim.

Hi Julien,
I really appreciate your efforts maintaining this and Java3D. We have a very important app that depends on it.

Have the issues with pixel and window size been fixed in the latest release? The last time I looked several years ago they had not, and we have stuck with Java 8.1 because of this issue.

Thanks,
Scott

sstevenson638 wrote

Hi Julien,
I really appreciate your efforts maintaining this and Java3D. We have a very important app that depends on it.

Have the issues with pixel and window size been fixed in the latest release? The last time I looked several years ago they had not, and we have stuck with Java 8.1 because of this issue.

Thanks,
Scott

Hi Scott,

I am currently about to resurrect the beast ;-)
A few pointers / threads
- 2.4.0 Release
  https://forum.jogamp.org/Release-2-4-0-Progress-tp4042061p4042141.html
- Management, call for funding
  https://forum.jogamp.org/JogAmp-Management-Kick-Off-Meeting-tp4042042p4042142.html
- PixelScale (window-units etc)
  https://forum.jogamp.org/GLcanvas-vs-NEWT-on-Hi-DPI-Screens-tp4041191p4042145.html
- Our git repos
  https://jogamp.org/cgit/?s=idle
- Our CI
  https://jogamp.org/chuck/

In total (once again), all can be resolved well.
'Only' issue left is a response from companies funding the work .. 'only' :)
Needless to say, I will try my best these days, but to continue the project properly
and to make work sustainable this must be resolved.

If you or anybody can help here with communication or perhaps even
being in a company capable of funding .. please do so.

Thank you.

~Sven

I like your sense of humor. Seriously, work must pay.