#jogamp @ irc.freenode.net - 20140110 05:05:38 (UTC)

20140110 05:05:38 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20140109050537.html
20140110 05:05:38 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20140110050538.html
20140110 05:30:14 * xranby (~xranby@anon) Quit (Ping timeout: 264 seconds)
20140110 05:45:50 * xranby (~xranby@anon) has joined #jogamp
20140110 07:33:13 * [Mike] (~Mike]@anon) Quit ()
20140110 08:12:45 * xranby (~xranby@anon) Quit (Read error: Operation timed out)
20140110 08:26:55 * xranby (~xranby@anon) has joined #jogamp
20140110 08:42:02 * D3zmodos (~D3zmodos@anon) Quit (Read error: Connection reset by peer)
20140110 08:43:55 * D3zmodos (~D3zmodos@anon) has joined #jogamp
20140110 09:41:41 * monsieur_max (~maxime@anon) has joined #jogamp
20140110 11:37:41 * D3zmodos (~D3zmodos@anon) Quit (Read error: Connection reset by peer)
20140110 11:38:39 * D3zmodos (~D3zmodos@anon) has joined #jogamp
20140110 11:47:07 * gouessej (546051e4@anon) has joined #jogamp
20140110 11:47:14 <gouessej> Hi
20140110 11:48:51 <gouessej> Ardor3D JOGL NEWT backend supports the change of cursor now: https://github.com/Renanse/Ardor3D/pull/68
20140110 11:49:13 <gouessej> I'll implement the same feature in JMonkeyEngine and then in LibGDX
20140110 11:50:36 <gouessej> Xerxes and Sven, perhaps I'll need some help to solve my problems with Git, I tried to update LibGDX code and push it
20140110 11:56:12 <xranby> gouessej: i can assist with git
20140110 11:56:17 * xranby (~xranby@anon) Quit (Read error: Operation timed out)
20140110 11:56:45 <gouessej> thanks Xerxes
20140110 11:57:00 * xranby (~xranby@anon) has joined #jogamp
20140110 11:57:11 <xranby> gouessej: i can assist with git
20140110 11:57:57 <gouessej> Git complained about lost changes if I did a merge :s
20140110 11:58:28 <gouessej> I'll give you the exact "complaint" tonight when I'm back home
20140110 11:58:49 <xranby> gouessej: do like this.. 1. git branch fixed-version
20140110 11:59:07 <xranby> it will create a branch named fixed-version at your current "head"
20140110 11:59:35 <xranby> git checkout fixed-version
20140110 11:59:42 <xranby> switches to your new branch
20140110 11:59:45 <gouessej> Actually, I make a pull to get your changes and another pull to get LibGDX changes
20140110 12:00:02 <xranby> in that case
20140110 12:00:12 <xranby> thinks
20140110 12:00:47 <xranby> you may start on my branch
20140110 12:01:40 <xranby> then git merge the master libgdx head
20140110 12:01:57 <gouessej> In that case, please can you at least get the latest updates of LibGDX even though it won't compile? Then I'll get your code, fix it, push it and you'll get my changes
20140110 12:03:35 <xranby> gouessej: yes i can do that
20140110 12:04:15 <gouessej> thank you very much
20140110 12:05:50 <gouessej> I will delete my local craps and make a git checkout on my repository. Then I will do "git pull xranby master" and "git push origin master".
20140110 12:06:37 <gouessej> After that, I will fix the compile error and do "git commit -a -m "Fixes"" and "git push origin master"
20140110 12:06:37 <xranby> ok i tell you when i am ready
20140110 12:06:51 <gouessej> thanks, your help is really appreciated
20140110 12:08:47 <gouessej> I have to work a bit on my model converter, NEWT Input API and JavaFX/OpenJFX. I don't want to spend a lot of time on LibGDX and it is not difficult to maintain, it's a thin library
20140110 12:09:57 <xranby> gouessej: you rock!
20140110 12:10:46 <gouessej> I hope you're right. I have spent a long time without working on the NEWT Input API and I'm almost sure I have lost the latest draft that I should have published :s
20140110 12:29:46 <xranby> gouessej: i have pushed a libgdx branch named jogamp-jogl-backend to github it contains our work merged on top of the current libgdx master
20140110 12:30:39 <gouessej> ok thanks
20140110 12:31:28 <xranby> i will let master branch match libgdx master
20140110 12:42:37 * Eclesia (~eclesia@anon) has joined #jogamp
20140110 12:43:12 <Eclesia> hi
20140110 12:49:12 <xranby> Eclesia: hi!
20140110 12:50:59 <gouessej> hi Eclesia
20140110 12:52:35 <Eclesia> any interesting news related to jogl during those holidays ? :)
20140110 12:53:50 <gouessej> baby steps as usual, nothing extraordinary
20140110 12:56:07 <gouessej> I don't know whether you feel concerned but I'd like to put Ardor3D JARs into http://jogamp.org/deployment/ardor3d
20140110 12:56:41 <gouessej> Nexus has been down for several months and some developers have complained about being unable to download the binary builds
20140110 12:57:42 <Eclesia> ardor3d has not been very active the last year, barely 1 or 2 commits per month :/
20140110 12:57:53 <gouessej> not this months
20140110 12:57:55 <gouessej> lol
20140110 12:58:04 <gouessej> about 10 commits
20140110 12:58:29 <Eclesia> http://www.ohloh.net/p/ardor3d
20140110 12:58:46 <gouessej> ohloh is a bit behind
20140110 12:59:02 <gouessej> https://github.com/Renanse/Ardor3D/commits/master
20140110 12:59:15 <gouessej> 9 commits since January, 1st
20140110 12:59:59 <Eclesia> project is still alive ok :)
20140110 13:00:11 <gouessej> of course it is
20140110 13:00:28 <gouessej> but some things are broken on the website
20140110 13:00:49 <gouessej> http://ardor3d.forumatic.com/viewtopic.php?f=15&t=12671
20140110 13:02:24 <Eclesia> I don't use ardor3d at home or at work so you can do whatever you want.
20140110 13:02:58 <Eclesia> (used it in the past, nice lib)
20140110 13:04:19 <gouessej> I remember you used it in the past, in an application of GIS, didn't you?
20140110 13:05:54 <Eclesia> yes but for our last project we decided to build our own engine above JOGL.
20140110 13:08:45 <xranby> Eclesia: please send in images + synopsis if you want the jogl based mapping software to appear at http://jogamp.org/jogl/www/
20140110 13:09:21 <gouessej> Why? There are already tons of mature engines and libraries with a JogAmp backend including Ardor3D, JMonkeyEngine, Java3D, Xith3D (?), LibGDX, jReality
20140110 13:09:42 <Eclesia> the engine is not available yet
20140110 13:09:47 <xranby> it would be good to list more mature applications as well
20140110 13:10:25 <gouessej> there are more and more Java3D applications that would deserve to be mentioned here
20140110 13:16:25 <Eclesia> why? because we think we can do better XD
20140110 13:16:37 * sgothel (~sgothel@anon) has joined #jogamp
20140110 13:16:42 * sgothel (~sgothel@anon) Quit (Changing host)
20140110 13:16:42 * sgothel (~sgothel@anon) has joined #jogamp
20140110 13:16:42 * ChanServ sets mode +v sgothel
20140110 13:17:09 <Eclesia> no, because we want to have a similar api for desktop and webgl
20140110 13:17:38 <Eclesia> hi sgothel
20140110 13:17:45 <sgothel> Hi All
20140110 13:17:51 <sgothel> Hi Eclesia
20140110 13:18:18 <sgothel> I went through the icon thingies .. and it turned out only XCURSOR allows one to push RGBA icons
20140110 13:18:31 <sgothel> hence no XPM etc need .. sorry for that
20140110 13:19:35 <gouessej> Eclesia, you can use OpenGL-ES with some of these engines
20140110 13:20:42 <sgothel> @Xerxes: I just updated RasperryPi to Debian-8 (Jessie) .. and I still had to move the Mesa libs 'out of the way' ..
20140110 13:20:49 <sgothel> otherwise it works well
20140110 13:21:01 <Eclesia> sgothel: so what format does it use ? there are other formats which could be used, not necesarly RGBA.
20140110 13:21:27 <sgothel> XCURSOR uses a simple { width, height, pixel1, .. }
20140110 13:21:40 <Eclesia> black-white, 16bit palette could still be interesting
20140110 13:21:41 <sgothel> oh wait .. that was the window-icon ..
20140110 13:21:48 <sgothel> XCURSOR simple RGBA
20140110 13:21:53 <gouessej> Eclesia, someone had a similar problem with the R-Pi, we discussed about that on the forum a few months ago when trying to run JMonkeyEngine 3 on it
20140110 13:22:24 <xranby> sgothel: hmm.. i have mostly only tested the "rasbian" distribution where it work out of the box.. (no mesa installed)
20140110 13:22:30 <sgothel> the non-rgba .. no interest to have an extra path here - but we can convert all to RGBA
20140110 13:22:54 <xranby> + it is using hard float
20140110 13:22:55 <sgothel> @Xerxes: good .. my updated installed mesa ..
20140110 13:23:00 <sgothel> hf .. yup
20140110 13:23:12 <sgothel> so now have fun w/ cursor :)
20140110 13:23:22 <xranby> and multiple windows!
20140110 13:23:25 <sgothel> http://jogamp.org/deployment/archive/master/gluegen_760-joal_510-jogl_1198-jocl_902/
20140110 13:23:56 <xranby> wow.. must test that
20140110 13:23:59 <sgothel> fullscreen 'works' .. but it seems the BCM doesn't handle it well :)
20140110 13:24:19 <sgothel> i.e. 1080p is too much
20140110 13:24:32 <xranby> you can change the memory split
20140110 13:24:38 <xranby> if you havent done so
20140110 13:24:42 <sgothel> dunno :)
20140110 13:24:48 <xranby> sudo raspi-config
20140110 13:25:08 <xranby> there you can change how much memory is allocated to the GPU
20140110 13:25:17 * Eclesia searching cursor classes
20140110 13:25:47 <sgothel> @Eclesia: NEWT PointerIcon (Display + Window)
20140110 13:25:58 <xranby> i hope this tool is available in debian jessy
20140110 13:26:09 <sgothel> yup .. rebooting
20140110 13:26:32 <gouessej> GLWindow.setPointerIcon() .getPointerIcon()
20140110 13:26:39 <sgothel> Testing w/ TestGLWindows02NEWTAnimated
20140110 13:26:44 <gouessej> Display.createPointerIcon()
20140110 13:26:53 <Eclesia> nice, will try it soon :)
20140110 13:27:32 <sgothel> TestGLWindows02NEWTAnimated <- Raspberry many NEWT windows ..
20140110 13:28:06 <gouessej> I use this API here: https://github.com/Renanse/Ardor3D/blob/master/ardor3d-jogl/src/main/java/com/ardor3d/input/jogl/JoglNewtMouseManager.java#L47
20140110 13:28:51 <sgothel> @Xerxes: Nice w/ 256MB .. seems to work more stable
20140110 13:29:25 <gouessej> Lol everyone owns a Raspberry Pi
20140110 13:30:07 <sgothel> .. and fullscreen works w/ 1080p then :)
20140110 13:30:15 <sgothel> +panda +beagle +...
20140110 13:30:33 <sgothel> @Xerxes: Now add the scroll wheel :)
20140110 13:31:05 <Eclesia> you remember the issue : "not seeing the frame in the desktop bar" ?
20140110 13:31:06 <sgothel> .. also startup seems to be faster now
20140110 13:31:19 <sgothel> there was something yes
20140110 13:31:35 <Eclesia> you found a solution ?
20140110 13:31:42 <sgothel> pls add the NEWT enhancement request in bugzilla
20140110 13:32:18 <sgothel> I haven't looked at it .. but I thought last time we discussed it 'a solution' was mentioned here ?
20140110 13:32:43 <sgothel> pls put it all in the bugzilla entry .. but won't be done now, after fosdem.
20140110 13:32:46 <xranby> gouessej: yes the pi is the common peoples embedded linux device
20140110 13:32:54 <sgothel> (or do it yourself)
20140110 13:33:18 <xranby> it is quite versatile despite its limitations
20140110 13:33:48 <xranby> what i like about it is that it make application developers aware of the embedded constraints
20140110 13:34:13 <sgothel> well, despite the hype .. and it's low price tag - there are for sure better devices like odroid etc (better performance etc) .. but everybody their toy
20140110 13:34:38 <gouessej> Why is the proprietary Java installed by default on the Pi? http://www.raspberrypi.org/archives/4920
20140110 13:34:52 <sgothel> as Xerxes says .. it's good that it's widespread for the awareness thingy
20140110 13:36:16 <sgothel> here it's OpenJDK+JamVM
20140110 13:37:28 <gouessej> Now the Raspberry Pi is used to promote closed source softwares
20140110 13:37:33 <gouessej> really silly
20140110 13:37:47 <sgothel> agreed .. the hype
20140110 13:38:29 <sgothel> but also: the BCM graphics (videocore) is highly proprietary - no open source driver, but the userlib (only ioctl'ing to blob)
20140110 13:38:41 <sgothel> no documentation .. etc
20140110 13:39:39 <Eclesia> here : https://jogamp.org/bugzilla/show_bug.cgi?id=940
20140110 13:40:16 <sgothel> thx
20140110 13:40:40 <Eclesia> xranby: If I remember correctly you found a solution to this, somekind of argument to remove when calling the windows manager, do you remember ?
20140110 13:40:43 <sgothel> if you can find some hints (solution) we found last time, pls add them
20140110 13:41:48 <sgothel> for sure we have to see to security issues w/ this feature
20140110 13:42:02 <sgothel> i.e. no one-and-only window w/o taskbar entry
20140110 13:42:12 <sgothel> otherwise .. application could steam the desktop
20140110 13:42:15 <sgothel> *steal*
20140110 13:42:40 <sgothel> popup: no-decoration - hence is possible already, but w/ that taskbar entry
20140110 13:43:17 <Eclesia> they could be attached to a main windows. glWindow.createChild(boolean desktopVisible)
20140110 13:43:18 <sgothel> and if used as a NEWT child window, maybe the window-icon already gets grouped ?
20140110 13:43:20 <Eclesia> something like that
20140110 13:43:52 * gouessej (546051e4@anon) Quit (Quit: Page closed)
20140110 13:46:49 <Eclesia> ha, found the irc log : http://jogamp.org/log/irc/jogamp_20131220050534.html
20140110 13:47:06 <Eclesia> 20131220 15:04:05 <xranby> Window state:
20140110 13:47:06 <Eclesia> 20131220 15:04:05 <xranby> Skip Taskbar
20140110 13:47:06 <Eclesia> 20131220 15:04:26 <xranby> xwininfo -wm
20140110 13:47:06 <Eclesia> 20131220 15:05:17 <xranby> so if the newt window adds this Skip Taskbar state then it will not show up in the desktop bar
20140110 13:47:41 <sgothel> added this conversation to 940 ..
20140110 13:49:16 <sgothel> ^^ maybe we should leave this up to the WM, i.e. how to group/visible child windows .. have to see how current child window behaves on some WMs (windows, osx, kwin, ..)
20140110 13:49:37 <sgothel> kwin does group them .. osx same
20140110 13:50:56 <Eclesia> it's not really 'grouping', a tooltip doesn't deserve to be visible even if it's grouped in the toolbar
20140110 13:51:53 <sgothel> so then it's allow-anonymity if child+skip_taskbar set
20140110 13:52:55 <sgothel> plus .. no fullscreen for such windows
20140110 13:53:29 <sgothel> or maybe .. simply no fullscreen for skip-taskbar windows
20140110 13:56:03 <Eclesia> what about multiple screens ? application might have a main windows on one screen, and secondary windows on other frames, those being in fullscreen mode
20140110 13:56:29 <sgothel> you can do this already sure - not considering the taskbar
20140110 13:57:39 <sgothel> don't forget: the taskbar and WM behavior should not be overriden by applications excessively, at least this is the best 'nice' behavior of an app
20140110 13:58:14 <Eclesia> sure, I'm just looking for use-cases
20140110 13:58:25 <Eclesia> ha, gimp.
20140110 13:58:26 <sgothel> hence out fullscreen mode does consider 'unfocus' i.e. remove the fullscreen mode while not being focused
20140110 13:58:47 <sgothel> in gimp, I see all toolboxes in WM taskbar
20140110 13:58:53 * rmk0 (~rmk0@anon) Quit (Quit: Lost terminal)
20140110 13:58:56 <sgothel> and that is good :)
20140110 13:59:10 <Eclesia> I see only one in the taskbar
20140110 13:59:19 <sgothel> I agree w/ popups though
20140110 13:59:34 <sgothel> hence it's the WM's setting
20140110 14:00:02 <sgothel> let me check a NEWT child window test ..
20140110 14:00:16 <Eclesia> (14:53:29) sgothel: or maybe .. simply no fullscreen for skip-taskbar windows : +1 for this solution
20140110 14:00:42 <sgothel> then we need to define what is 'no fullscreen' ?
20140110 14:00:52 <sgothel> i.e. max size 1/2 of screen ? :)
20140110 14:01:10 <sgothel> or simply not the special fullscreen mode .. should work
20140110 14:01:16 <sgothel> i.e. taskbar switching ..
20140110 14:01:42 <sgothel> ok .. I guess we can live w/ that .. user can still bring up the console and kill the beast
20140110 14:01:48 <Eclesia> not the special fullscreen mode is ok
20140110 14:02:39 <Eclesia> if we start to thnik in size we are screwed, some combobox dropdow can be very long (example postgres datatype combobox)
20140110 14:02:49 <sgothel> ay
20140110 14:03:03 <sgothel> copied that to 940
20140110 14:03:19 <Eclesia> so to be fait we would have to calculate the % of area, not just width/height
20140110 14:03:23 <Eclesia> fair*
20140110 15:02:09 * D3zmodos (~D3zmodos@anon) Quit (Read error: Connection reset by peer)
20140110 15:04:21 * D3zmodos (~D3zmodos@anon) has joined #jogamp
20140110 18:04:53 * [Mike] (~Mike]@anon) has joined #jogamp
20140110 18:12:24 * hharrison (~chatzilla@anon) has joined #jogamp
20140110 18:14:21 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20140110 18:15:50 <hharrison> sgothel: any feeling on requiring recent git versions to build gluegen/jogl etc?
20140110 18:17:53 <hharrison> specifically git 1.8.1 or newer (Jan 2013 release)
20140110 18:19:48 <hharrison> Oh, actually, I think I see a way that I don't need that, nevermind
20140110 18:20:21 <hharrison> (I want the -q option for git symbolic-ref to remove the branch-name finding in the ant scripts)
20140110 18:20:51 <sgothel> git versions incompatible ?
20140110 18:22:53 <sgothel> Nice talk about 'X Security', when you filter out the bashing and knowing that the: WM already runs in userspace, DRI/Network 'redirection' is properly done ..
20140110 18:22:57 <sgothel> http://media.ccc.de/browse/congress/2013/30C3_-_5499_-_en_-_saal_1_-_201312291830_-_x_security_-_ilja_van_sprundel.html
20140110 18:23:03 <sgothel> good overview
20140110 18:23:11 <hharrison> -q option was only added in 1.8.12
20140110 18:23:17 <hharrison> 1.8.1 I mean
20140110 18:23:24 <hharrison> (fat finger)
20140110 18:23:44 <sgothel> git help ???
20140110 18:23:53 <hharrison> git symbolic-ref
20140110 18:24:24 <hharrison> replaces the git branch --no-color + sed you currently have to find branch name when building
20140110 18:25:24 <sgothel> ah ..
20140110 18:25:39 <hharrison> git symbolic-ref --short -q HEAD
20140110 18:25:40 <sgothel> If you like .. earmark it and we can do it after 2.1.4
20140110 18:25:49 <hharrison> I'm just playing a bit with it
20140110 18:25:56 <sgothel> maybe a branch/patch on stash :)
20140110 18:26:16 <hharrison> We discussed a bit more common build helpers we could start using across the jogamp repos
20140110 18:26:24 <hharrison> This is me playing a bit with it
20140110 18:51:43 * zzuegg (~zzuegg@anon) Quit ()
20140110 18:53:43 * xranby (~xranby@anon) Quit (Read error: Operation timed out)
20140110 19:09:17 * xranby (~xranby@anon) has joined #jogamp
20140110 21:31:25 * xranby (~xranby@anon) Quit (Ping timeout: 246 seconds)
20140110 21:46:28 * xranby (~xranby@anon) has joined #jogamp
20140110 23:31:01 * xranby (~xranby@anon) Quit (Read error: Operation timed out)
20140110 23:33:43 * hharrison (~chatzilla@anon) Quit (Remote host closed the connection)
20140110 23:45:29 * xranby (~xranby@anon) has joined #jogamp
20140111 00:30:36 * kermyt (~kermyt@anon) Quit (Ping timeout: 276 seconds)
20140111 00:33:57 * kermyt (~kermyt@anon) has joined #jogamp
20140111 00:33:57 * kermyt (~kermyt@anon) Quit (Changing host)
20140111 00:33:57 * kermyt (~kermyt@anon) has joined #jogamp
20140111 00:35:49 * monsieur_max (~maxime@anon) has left #jogamp
20140111 01:33:24 * sgothel (~sgothel@anon) Quit (Quit: Leaving.)
20140111 01:47:56 * sgothel (~sgothel@anon) has joined #jogamp
20140111 01:47:56 * ChanServ sets mode +v sgothel
20140111 01:49:11 <sgothel> reminder: apt-get upgrade .. security bugs etc .. :)
20140111 02:16:02 <sgothel> testing amd-catalyst-13.11-betaV9.95-linux-x86.x86_64
20140111 02:32:36 * kermyt (~kermyt@anon) Quit (Quit: ZNC - http://znc.in)
20140111 02:33:29 * kermyt (~kermyt@anon) has joined #jogamp
20140111 02:38:22 * hharrison (~chatzilla@anon) has joined #jogamp
20140111 02:49:29 <sgothel> amd-catalyst-13.11-betaV9.95-linux-x86.x86_64 @ wheezy 64bit: 1 junit.run of JOGL: good
20140111 02:49:38 <sgothel> (no errors)
20140111 03:07:29 * xranby (~xranby@anon) Quit (Ping timeout: 252 seconds)
20140111 03:20:52 * xranby (~xranby@anon) has joined #jogamp
20140111 04:01:38 * [Mike] (~Mike]@anon) Quit ()
20140111 04:14:36 * hharrison (~chatzilla@anon) Quit (Quit: ChatZilla 0.9.90.1 [Firefox 26.0/20131209183133])
20140111 05:05:38 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20140111050538.html