Fullscreen on secondary monitor

I don't know why your key listener doesn't work and the very first call of addReference() can take a long time.

I don't think that a more "modern" forum would noticeably increase JogAmp's success but it might discourage JogAmp users to post the same questions on StackOverflow or to post exclusively on StackOverflow... whereas at the end you know who answers most of the JOGL specific questions on StackOverflow: me.

gouessej wrote

I don't know why your key listener doesn't work and the very first call of addReference() can take a long time.

I don't think that a more "modern" forum would noticeably increase JogAmp's success but it might encourage JogAmp users to post the same questions on StackOverflow or to post exclusively on StackOverflow...

Take in account a new user looking to choose what to use among jogl and lwjgl may also evaluate it based on the corresponding websites and forums.. So I wouldnt underestimate it :)

gouessej wrote

 whereas at the end you know who answers most of the JOGL specific questions on StackOverflow: me.

That's true :D, but sometimes I also see other people answering jogl related stuff on stackOverflow

elect wrote

Take in account a new user looking to choose what to use among jogl and ***** may also evaluate it based on the corresponding websites and forums.. So I wouldnt underestimate it :)

Personally, if a developer prefers using another API because its forum is more beautiful, I will just ignore him. I don't mind. This isn't the kind of developer I'd like to attract and I'm probably not alone to think that the JogAmp contributors have more important tasks to do than switching to a more "modern" forum. I don't see the interest of attracting very superficial developers who might switch to another API every month just because it's "cool". In my humble opinion, improving the wiki and writing a lot more tutorials is a lot more important than improving the forum. If you really think that talking about our main competitor is the best way of obtaining anything, you will quickly learn that you're completely wrong. Please avoid mentioning it on this forum except when it is absolutely necessary, this forum isn't a big billboard for it.

Sven, Xerxes, Rami, Harvey, Wade and the others make a great job with JogAmp APIs and Java3D. If you have some time to spare on improving the forum, why not?

elect wrote

That's true :D, but sometimes I also see other people answering jogl related stuff on stackOverflow

I don't deny that, that's why I wrote "most", I didn't write that I answer all questions but anyway, sorry to be harsh but other SO users who have answered to JOGL specific (not just vaguely JOGL related) questions have almost never given a reply that I wouldn't have been able to provide (except other JogAmp contributors and a very few JogAmp users of course). It's even worse, some of them used JOGL specific questions to advertise their crappy APIs, some gave very bad advises (for example putting all JARs into /lib/ext, using obsolete versions of JOGL, ...), some suggested overcomplicated solutions, ... In other terms, using SO for JogAmp is most of the time a bad idea and there are better places to talk about OpenGL specific questions.

gouessej wrote

Personally, if a developer prefers using another API because its forum is more beautiful, I will just ignore him. I don't mind. This isn't the kind of developer I'd like to attract and I'm probably not alone to think that the JogAmp contributors have more important tasks to do than switching to a more "modern" forum. I don't see the interest of attracting very superficial developers who might switch to another API every month just because it's "cool". In my humble opinion, improving the wiki and writing a lot more tutorials is a lot more important than improving the forum. If you really think that talking about our main competitor is the best way of obtaining anything, you will quickly learn that you're completely wrong. Please avoid mentioning it on this forum except when it is absolutely necessary, this forum isn't a big billboard for it.

 

I agree with everything, but the world we deal with is different from what an ideal world should be.

gouessej wrote

Sven, Xerxes, Rami, Harvey, Wade and the others make a great job with JogAmp APIs and Java3D. If you have some time to spare on improving the forum, why not?

Sure, this was already implicit from my side :)

I will open a new thread in the general section

elect wrote

I agree with everything, but the world we deal with is different from what an ideal world should be.

Actually, we disagree. Let me rephrase that: In my humble opinion, spending tons of time in improving the forum shouldn't be done only with the hope of attracting new users because the kind of newcomers it would attract wouldn't deserve such efforts. In the real world, there are tons of people attracted by eye candy things but we aren't computer graphics artists and this forum targets the developers. Nevertheless, I'll look at your suggestions because I assume that some efforts could be done but for better reasons.

On 10/13/2014 10:23 AM, elect [via jogamp] wrote:
>     Sven Gothel wrote
>     we surely speak a different language, cannot grok your statement.
>
>     maybe you _read_ the demo then :)
>
> Anyway, since I couldnt run them, I did read them, that's why I found the
> addReference() missing..

please don't also miss the removeReference()
.. or use createNative() (sic - i.e. might be spelled different, see Screen
API - compare unit tests).

>
> I can have the fullscreen on the i-th monitor I want :)

great!

>
> Just few considerations:
>
> - the keyListener doesnt work anymore now at fullscreen, but I will
> investigate something more on this, maybe it is my fault

dunno about that, unit tests work in this regards.

>
> - addReference() takes a lot of time, more than 10s, is it normal?

no, our unit tests are much faster.

also, these 10s would appear also if you natively realize a NEWT GLWindow
or the like via setVisible(..).

~Sven

Sven Gothel wrote

please don't also miss the removeReference()
.. or use createNative() (sic - i.e. might be spelled different, see Screen
API - compare unit tests).

What is the point of doing add and removeReference()?

Sven Gothel wrote

no, our unit tests are much faster.

also, these 10s would appear also if you natively realize a NEWT GLWindow
or the like via setVisible(..).

~Sven

I dont get what you mean, at the moment I just create a jFrame and add the newtCanvasAWT to it

I have looked at WindowImpl (setFullscreenImpl() + FullscreenAction) and GLWindow. I have found nothing obvious. Please provide a tiny test case or adapt an existing unit test to help us to reproduce your bug. Do you confirm you don't loose your listener when you call setFullscreen(true)?

Edit.: Do you use an AWT key listener or a NEWT key listener with your NewtCanvasAWT?

gouessej wrote

I have looked at WindowImpl (setFullscreenImpl() + FullscreenAction) and GLWindow. I have found nothing obvious. Please provide a tiny test case or adapt an existing unit test to help us to reproduce your bug. Do you confirm you don't loose your listener when you call setFullscreen(true)?

Edit.: Do you use an AWT key listener or a NEWT key listener with your NewtCanvasAWT?

No wait, gouessej, I still have to look at it, it is highly probable I am doing something wrong there, I was asking quick if it was a sort of known bug or something similar. Anyway newt key listener

Maybe you just lose the focus.

gouessej wrote

Maybe you just lose the focus.

Yeah, something like that.

Anyway Sven, have you got the dk2? Because I just discovered the 10s latency at the start is affecting *all* my jogl projects. If I turn the dk2 off, I get no problem. When dk2 is on in those 10s the pc becames very very slow, even simple operation are executed slowly..

I tried to isolate the problem. It comes when I add the newtCanvasAWT to my jPanel.