Hey, this is great news and great work. Thank you so much.
I will grant you complete dev access to everything when I come back home. I love the idea of someone updating all that stuff while I don't have the time to do so.
I have also started to check every test (and benchmark and demo) from xith-tk and would like to release a new cooker (or beta?) when everything is working fine.
Oh yes, please do this. If you think, everything, that is currently supported works up to a certain level, do a beta release. If you think, there are certain things missing (featurewise), that have to be added, do a cooker release.
By the way, the link for installing xith on netbeans on the website is dead. I don't mind because I use Eclipse, but maybe someone has a copy of the blog's article to fix this so the information will be complete.
I will send Mathias a message, so that he can possibly forward you some data or an account or something. He's the original author of that article.
- ColoringAttributesTest: I see no difference when hitting the space key to change the projection mode, is this expected (maybe a newbie question on projection logic, sorry)?
I think, it was Yuri, who added this projection switch to many testcases, but it doesn't make sense for all of them. I removed it here and there. This may be a relic here and can be removed.
- Bleeding_Blending: this test is not exposed, is there a reason or was it missed?
I guess, I forgot to make this testcase extend Xith3DTest. So this is a bug.
- Xith3DLogoTest: for the moment I added SoundProcessor.getInstance().getSoundDriver().start() in the test to have sound on a second launch because the Xith3DEnvironment.destroy() also calls destroy on the sound driver; do you thing this should be placed in the Xith3DEnvironment.<init>?
I guess, I missed that one when I fixed the bug described in
this thread.
- PhongLighting: fire artefact or crash with LWJGL_AWT on my machine, other OpenGL layers run fine
LWJGL_AWT never worked as good as it should do. IIRC I never updated it to the latest LWJGL API, where the AWT rendering works slightly different to older versions of LWJGL.
This leads me to ask if it would not be the right time to set LWJGL as the default OpenGLLayer? ...
Yes! I wanted to do this for so long. Mathias objected once, because he intented to suggest Xith3D as a reference project on the JOGL homepage. And in this case it wouldn't be a good sign, if we made LWJGL the default layer. But since Mathias has left the project and won't deal with this anymore, we can and should definitely head this way now. I do agree, that LWJGL is by far the better binding.
Just make sure, that the testcases deal with the default in the correct way. So if a testcase explicitly selects LWJGL and a related testcase uses the default, while it is actually meant to explicitly use JOGL, it should be changed. I think, I have always done so, but who knows.
Marvin