#jogamp @ irc.freenode.net - 20130813 05:06:07 (UTC)

20130813 05:06:07 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20130812050607.html
20130813 05:06:07 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20130813050607.html
20130813 05:15:42 * xranby (~xranby@anon) Quit (Ping timeout: 276 seconds)
20130813 05:17:44 * xranby (~xranby@anon) has joined #jogamp
20130813 07:19:26 * xranby (~xranby@anon) Quit (Ping timeout: 240 seconds)
20130813 07:21:22 * xranby (~xranby@anon) has joined #jogamp
20130813 08:29:24 * xranby1 (~familjen@anon) has joined #jogamp
20130813 10:39:39 * kermyt_ (~kermyt@anon) has joined #jogamp
20130813 10:40:34 * kermyt (~kermyt@anon) Quit (*.net *.split)
20130813 10:40:36 * kermyt_ is now known as kermyt
20130813 12:02:26 * vinz (~vinz@anon) has joined #jogamp
20130813 12:03:07 <vinz> Hy there :)
20130813 12:16:57 <vinz> Does anyone have a working example of jogl working with GL2ES2 and VOB ?
20130813 12:20:39 <xranby> vinz: all the applet demos on the jogl website use the GL2ES2 profile. What kind of VBO example do you need? graph and gears uses VBO AFAIK
20130813 12:20:46 <xranby> https://jogamp.org/deployment/jogamp-current/jogl-test-applets.html
20130813 12:20:57 <xranby> the sourcecode for these applets is inside the jogl project
20130813 12:21:40 <xranby> https://jogamp.org/git/?p=jogl.git;a=tree;f=src/test/com/jogamp/opengl/test/junit;h=ae2636b8b866878eac0799cec0d10f5142eca6ac;hb=HEAD
20130813 12:24:56 <vinz> I wanted to get started with WebGL but first wanted to properly learn it from a language im familliar with
20130813 12:25:15 <vinz> so i started the NeHe tutorial
20130813 12:25:48 <xranby> is the nehe tutorial updated for jogl 2.0.2 ?
20130813 12:25:59 <vinz> but i read that was quiet "old style open gl" then i read about VOB and tryed to basicly implement it but every time i run something i get an
20130813 12:26:29 <vinz> javax.media.opengl.GLException: element vertex_buffer_object must be enabled to call this method
20130813 12:27:15 <vinz> so wanted to compare it with something really basic which is working to understand what i'm doing wrong
20130813 12:27:42 <vinz> btw thank's for the links
20130813 12:29:10 <xranby> vinz: i can give you a good guide to modern opengl
20130813 12:29:27 <xranby> vinz: http://www.arcsynthesis.org/gltut/
20130813 12:29:29 <vinz> oh would be really interested in :)
20130813 12:29:45 <xranby> this book only uses the programmable pipeline
20130813 12:30:29 <xranby> all sourcecode is available here: https://bitbucket.org/alfonse/gltut/wiki/Home
20130813 12:30:50 <xranby> there is one community member who is currently porting this book to use jogl let me see if i can find the link
20130813 12:32:08 <vinz> that's ever more interessting ;)
20130813 12:49:23 <vinz> Unfortunately i have to leave, thank you for your help and i will carefully read the online book you provided me seems really interessting, anyway will be back here :)
20130813 12:49:31 <vinz> Thank's again！
20130813 12:49:51 * vinz (~vinz@anon) Quit (Quit: Leaving)
20130813 14:09:10 <xranby> JOGL port of gltut examples by Giuseppe Barbieri https://github.com/elect86/modern-jogl-examples
20130813 14:13:00 <sgothel> .. right, I chatted w/ Giuseppe .. and looks at the code yesterday. nice stuff ..
20130813 14:14:01 <sgothel> one thing though .. the ported code uses Vec* and Matrix class encapsulation to do all similar stuff as we do (PMVMatrix, .. etc) - and hence I advise not to use this method due to efficiency.
20130813 14:14:54 <sgothel> *low level nagging* :)
20130813 14:15:37 <sgothel> @Xerxes: Did you know that the PTS of audio/video is independent ? duh!
20130813 14:15:54 <sgothel> (at least for those mp4 test cases ..)
20130813 14:16:25 <xranby> sgothel: yes the mp4 testcases are packed funny
20130813 14:16:30 <sgothel> I was looking in avplay .. to find the SCR (system clock) correlation .. since I thought the pts is simply wrong .. :)
20130813 14:17:17 <xranby> i have one ogg file were the audio is played back at double speed using the impl of GLMediaPlayer from 2 days ago
20130813 14:17:26 <xranby> let me see if it is online
20130813 14:17:36 <sgothel> for now, I leave the sync-code in GLMediaPlayerImpl* .. only sleep a bit in case we are too fast .. can be override .. etc and refined later.
20130813 14:18:12 <xranby> do you sleep in the render thread? or a separate audio thread?
20130813 14:18:16 <sgothel> final testing .. threaded decoding (>= 4 textures), so far seems to be fine.
20130813 14:18:35 <sgothel> all sync must be done in render thread, all pts are pushed to A/V-Frame data
20130813 14:18:51 <xranby> there was some issues witht he omx .c and .h files after the WIP checking
20130813 14:18:53 <sgothel> since no async is performed now, assuming a consistent audio stream ..
20130813 14:19:07 <sgothel> yup .. will check ..
20130813 14:19:41 <sgothel> one audio issue was the lack of negotiation of the sample-rate .. etc between ffmpeg/openal, fixed.
20130813 14:20:07 <sgothel> thx to ffmpeg's ac3-decoder, downsampling (stereo( works out of the box
20130813 14:20:36 <xranby> btw.. i found myself to be one of the handful people to have managed to build openjfx using openjdk yesterday https://twitter.com/xranby/status/366925956734144512/photo/1
20130813 14:21:08 <xranby> it was working but.. hmm slow
20130813 14:21:34 <xranby> and no media support
20130813 14:22:08 <sgothel> btw .. I test manually w/ Ji-Gong, aka openjdk8-compat1 :)
20130813 14:22:08 <xranby> and no runtime platform autodetection
20130813 14:22:35 <xranby> they really only use compiletime platform selection
20130813 14:22:54 <sgothel> forget openjfx .. it's a dead horse already
20130813 14:23:53 <xranby> (16:23:41) sgothel: btw .. I test manually w/ Ji-Gong, aka openjdk8-compat1 :) <-- great
20130813 14:24:01 <xranby> btw.. i taked with acreo today
20130813 14:24:09 <xranby> and introduced you and the ji gong project
20130813 14:24:30 <sgothel> whats 'acreo' ?
20130813 14:24:31 <xranby> they will get back with a list of european fuding options
20130813 14:24:38 <sgothel> wow .. sweet!
20130813 14:24:56 <xranby> Acreo is the swedish gowernment run research company who also manages swedish research grants
20130813 14:25:16 <sgothel> then we can share it .. put some horsepower to the running one :)
20130813 14:25:41 <xranby> basically there will be european funding available with focus on preventing the financial crisis from spreading
20130813 14:25:52 <sgothel> haha :)
20130813 14:26:17 <xranby> thus the key is to put into the applications how the funding will create spinoff jobs in europe
20130813 14:26:24 <sgothel> hauling w/ the wolves .. playing the game .. oh well, guess we are forced to do so ..
20130813 14:26:46 <sgothel> oh, we can list a lot of companies using our stuff .. so we have good background
20130813 14:27:34 <xranby> they liked the idea to have a license free foundation that can be used by european car manufacturers
20130813 14:27:57 <sgothel> had a chat w/ Giuseppe's employer and him (reknow.de) .. they have a nice application (viewports .. of machines etc) .. and all the other [small and bigger] companies.
20130813 14:28:14 <sgothel> yup, thats what the dude said at siggraph as well, 'exploitable' stuff ..
20130813 14:28:26 <sgothel> so .. let me call this lazy notary again .. dih!
20130813 14:28:50 <sgothel> (to move all assets to foundation)
20130813 14:29:19 <xranby> i dont think that is key to get the funding
20130813 14:29:25 <xranby> maybe maybe not
20130813 14:29:47 <xranby> funding can go directly to jausoft etc
20130813 14:29:57 <xranby> since there is a project plan
20130813 14:30:08 <xranby> with archivable goals
20130813 14:31:38 <xranby> in any case, i will report back when i get more information from Acreo
20130813 14:31:54 <sgothel> yes .. and this can be orthogonal, a foundation emphasizes the 'exploit-ability' -> i.e. availability of project beyond lifecycle!
20130813 14:32:14 <xranby> good point
20130813 14:32:21 <sgothel> ok, tonight or thursday - phonecall w/ notary
20130813 14:32:41 <sgothel> so min. goal of foundation is that all of us, 'own' all the equipment etc!
20130813 14:33:01 <sgothel> i.e. call to service provider .. etc etc
20130813 14:33:16 <sgothel> more fine grained stuff can be solved as we go
20130813 14:33:40 <sgothel> right .. the ssl certs can be signed by foundation as well :)
20130813 14:33:45 <xranby> rule of thumb.. projects usually fund future expenses
20130813 14:34:18 <sgothel> yes, that is already the truth .. i.e. some of our clients work finances the future .. sure.
20130813 14:34:26 <xranby> some ties you can get rejected funding if you try to add already purcased equipment before the project start
20130813 14:35:00 <sgothel> we don't care about stuff we have now .. and we focus on the expensive stuff: people / contracting
20130813 14:35:05 <xranby> good
20130813 14:35:31 <sgothel> stuff we have goes to foundation (server, test machines .. etc)
20130813 14:36:16 <sgothel> but as my accountant always says: software business is zero costs - all work hours only :)
20130813 16:12:14 * void256 (~chatzilla@anon) has joined #jogamp
20130813 16:12:47 <sgothel> Hi Jens .. how is it going ?
20130813 16:13:06 <sgothel> didn't chat w/ you since siggraph .. hope presentation was OK
20130813 16:13:17 <sgothel> btw .. need address for t-shirt
20130813 16:13:43 <sgothel> @all (Mark, Julien, Dominik, ..): I need an address for the t-shirts !
20130813 16:15:23 * xranby (~xranby@anon) Quit (Ping timeout: 240 seconds)
20130813 16:16:41 <void256> hi sven - I should ask you how the presentation was :) I kinda was missing a live blog / tweet / report from you guys but I figured that it probably was quite a bit stressful ;)
20130813 16:16:56 <void256> so thats ok
20130813 16:18:03 <void256> I need some help with newt keyboard events
20130813 16:18:25 <void256> with awt there was KeyEvent.KEY_LOCATION_RIGHT and so
20130813 16:18:54 <void256> I'm missing that with the Newt KeyEvent - how is that supposed to be converted?
20130813 16:21:13 * xranby (~xranby@anon) has joined #jogamp
20130813 16:21:51 <void256> xranby: can you check the last view lines in the chat log eventually - I think you're the one that knows the answer :)
20130813 16:24:53 <void256> I'm adding support to the JoglInputsystem to request input events that have not been handled by nifty - this is required by people that want to display nifty above their own application and are interested in events that nifty did not handle. thats no problem since I've already solved that but support in the JoglInputSystem was missing
20130813 16:26:38 <void256> and as I was adding that I noticed that the AwtToNiftyKeyCodeConverter still uses awt imports instead of newt imports as the JoglInputsystem does ... and long story short: as I changed that to newt it complains that some constants like KeyEvent.KEY_LOCATION_RIGHT don't exist anymore and now I wonder how I should translate that
20130813 16:26:45 <void256> whew (long story ^^)
20130813 16:27:17 -mist- [Global Notice] Please note that emails from services (nickserv etc..) are currently unavailable. ETA to their return is approx 1-2 hours. During this downtime, new registration emails will fail, and password resets cannot be issued.
20130813 16:31:49 <sgothel> oh ho ..
20130813 16:32:01 <sgothel> we have a *RIGHT* .. let me see
20130813 16:33:19 <sgothel> we only have L and XL left .. your choice ..
20130813 16:33:45 <sgothel> KeyEvent.VK_RIGHT
20130813 16:33:47 <void256> case KeyEvent.VK_ALT: // fall through
20130813 16:33:48 <void256> if (location == KeyEvent.KEY_LOCATION_RIGHT)
20130813 16:33:50 <void256> return KeyboardInputEvent.KEY_RMENU;
20130813 16:33:51 <void256> else
20130813 16:33:53 <void256> return KeyboardInputEvent.KEY_LMENU;
20130813 16:33:55 <void256> I have to translate that for example
20130813 16:34:21 <void256> is there a VK_ALT_RIGHT?
20130813 16:34:28 <sgothel> if e.isShiftDown() / isAltDown() ..
20130813 16:34:52 <sgothel> or getModifiers() && ALT_MASK ..
20130813 16:35:09 <void256> yeah we have that too ... odd that the original author used KEY_LOCATION_RIGHT then
20130813 16:35:40 <sgothel> sometimes the VK_* has valuable API doc included ..
20130813 16:36:22 <sgothel> btw .. we already have an AWT -> NEWT event converter !
20130813 16:36:56 <void256> https://github.com/void256/nifty-gui/blob/1.3/nifty-renderer-jogl2/src/main/java/de/lessvoid/nifty/renderer/jogl/input/AwtToNiftyKeyCodeConverter.java
20130813 16:37:17 <void256> that one originally converts AWT-Events to Nifty-Events
20130813 16:37:54 <void256> so if there is a NEWT -> AWT converter I could probably get away with that right before I call the existing class, right? =)
20130813 16:38:00 <sgothel> jogamp.newt.awt.event.AWTNewtEventFactory <- used and tested w/ NewtCanvasAWT (OSX, CALayer) .. and other test cases
20130813 16:38:42 <sgothel> yeah .. we don't have that one yet, since we try to motivate folks to move to an AWT agnostic world :)
20130813 16:38:53 <void256> ah crap xD I mean, ok
20130813 16:38:56 <void256> ;)
20130813 16:46:05 <void256> sorry to keep bothering ^^ but how would I distinguish between left and right shift or left and right control?
20130813 16:46:39 <sgothel> we don't have that .. I am afraid ..
20130813 16:46:48 <sgothel> i.e. n/a on all platforms :(
20130813 16:47:18 <sgothel> if native OS does distinguish it, we merge it .. to the one token
20130813 16:48:36 <void256> alright. I'll map it to left then .. I'm not aware of an nifty control that makes that difference. I just check and it seems that internally we check both - so should be ok, I guess :>
20130813 16:48:45 <void256> *a
20130813 16:48:53 <void256> *I've just checked
20130813 16:49:11 <void256> englisch leistungskurs ;)
20130813 16:49:23 <void256> but I sucked :>
20130813 16:49:34 <void256> (obviously)
20130813 16:50:08 <sgothel> all great!
20130813 16:52:40 <void256> seems to work too :D
20130813 17:23:20 * xranby (~xranby@anon) Quit (Ping timeout: 245 seconds)
20130813 17:29:14 * xranby (~xranby@anon) has joined #jogamp
20130813 18:22:57 -mist- [Global Notice] Services emails should now be working as normal, thanks.
20130813 19:48:27 * hharrison (~chatzilla@anon) has joined #jogamp
20130813 21:59:16 * vinz (~vinz@anon) has joined #jogamp
20130813 21:59:21 * vinz (~vinz@anon) has left #jogamp
20130813 21:59:30 * vinz (~vinz@anon) has joined #jogamp
20130813 22:41:23 * xranby (~xranby@anon) Quit (Ping timeout: 268 seconds)
20130813 22:45:21 * xranby (~xranby@anon) has joined #jogamp
20130813 23:01:31 * xranby1 (~familjen@anon) Quit (Quit: Leaving.)
20130813 23:31:30 * hharrison (~chatzilla@anon) Quit (Remote host closed the connection)
20130814 01:02:57 * void256 (~chatzilla@anon) Quit (Quit: ChatZilla 0.9.90.1 [Firefox 23.0/20130730113002])
20130814 02:40:08 <vinz> GL2ES2 dosen't have GL_GEOMETRY_SHADER ?
20130814 02:40:29 <sgothel> nope -> GLES3 / GL3ES3
20130814 02:40:49 <vinz> Ok thx :)
20130814 02:49:52 * vinz (~vinz@anon) Quit (Quit: Leaving)
20130814 05:06:07 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20130814050607.html