#jogamp @ irc.freenode.net - 20130429 05:05:56 (UTC)


20130429 05:05:56 -CatOut- Previous @ http://jogamp.org/log/irc/jogamp_20130428050642.html
20130429 05:05:56 -CatOut- This channel is logged @ http://jogamp.org/log/irc/jogamp_20130429050556.html
20130429 06:04:58 * DemoscenePassiv (~Lutsche@anon) has joined #jogamp
20130429 06:48:12 * DemoscenePassiv (~Lutsche@anon) Quit (Ping timeout: 264 seconds)
20130429 08:38:15 * odin__ (~Odin@anon) has joined #jogamp
20130429 10:04:14 * gouessej (546051e4@anon) has joined #jogamp
20130429 10:04:53 <gouessej> hi
20130429 10:05:22 <gouessej> there is still a problem, I have reopened the bug 723
20130429 10:06:06 <gouessej> the arrows are not working under Windows
20130429 10:06:37 <gouessej> Windows 7
20130429 10:06:49 <gouessej> It seems to be worse under XP
20130429 10:09:31 <gouessej> I will perform some tests under Windows 7 tonight
20130429 10:22:21 <gouessej> rmk0 do you use Ada too??
20130429 10:27:38 <gouessej> it looks good :) http://fossil.io7m.com/repo.cgi/io7m-jspatial/index
20130429 10:58:34 <sgothel> Hi Julien. Before opening or re-opening a bug, I would prefer a more solid investigation 1st than the other way around, since a bug claim shall lead to an investigation, i.e. is serious. However, since you do that now .. ok'ish .. but IMHO doesn't look too good w/o proper data and proof.
20130429 10:59:11 <sgothel> 'seems to be worse' .. etc .. is not really a technical description we can deal w/
20130429 10:59:39 <sgothel> sorry for nagging .. it's my job :)
20130429 11:00:14 <gouessej> I checked by myself once
20130429 11:00:25 <gouessej> at home under Windows 7
20130429 11:00:46 <gouessej> I recompiled the whole project and Ardor3D with the latest aggregated build
20130429 11:00:46 <sgothel> maybe, but there is no technical description what does not work etc
20130429 11:01:06 <sgothel> just 'still a problem'
20130429 11:01:26 <gouessej> When I press an arrow key, the wrong code is returned and Ardor3D doesn't recognize it as an arrow key
20130429 11:01:38 <sgothel> what that problem is .. i.e. phenomenon is not being told - oh
20130429 11:01:47 <gouessej> Tonight I will provide the exact returned values
20130429 11:01:48 <sgothel> Windows 'eh ?
20130429 11:01:53 <gouessej> Windows 7
20130429 11:02:16 <sgothel> ok I test on German keyboard w/ Win7 and WinXP now .. and will post results, if it works here I close it. And IMHO .. it works
20130429 11:02:28 <sgothel> what is AZERTY ?
20130429 11:02:35 <sgothel> French ?
20130429 11:02:36 <gouessej> French keyboards
20130429 11:02:43 <sgothel> good .. will include that
20130429 11:02:58 <sgothel> I will simply run NEWT ok ?
20130429 11:03:04 <sgothel> i.e. TestGearsES2NEWT
20130429 11:03:15 <gouessej> QWERTZ and QWERTY are very close, you might not reproduce my bug
20130429 11:03:17 <sgothel> or what is your other test case ?
20130429 11:03:30 <sgothel> a native NEWT only ?
20130429 11:04:02 <gouessej> I used a testcase from Ardor3D but I can use something using only NEWT if you prefer
20130429 11:04:38 <sgothel> ofc .. only internal unit test are valid, and if thats not enough - we need to create a new one
20130429 11:04:59 <gouessej> I can use TestGearsES2NEWT
20130429 11:05:16 <gouessej> if it works with a crappy Intel chipset
20130429 11:05:20 <sgothel> Same goes w/ your JOAL patch .. which is still pending .. I will pick it up later when working on a/v again
20130429 11:05:59 <gouessej> I don't plan not to provide more data
20130429 11:06:26 <gouessej> I'm aware of the lack of information in my bug report
20130429 11:06:28 <sgothel> hu? that translate to 'you plan to provide more data' ?
20130429 11:06:34 <sgothel> ah :)
20130429 11:06:56 <sgothel> double negation in english is tricky :)
20130429 11:07:03 <gouessej> the first time, I gave the returned codes
20130429 11:07:10 <gouessej> I will do the same
20130429 11:07:19 <gouessej> with keySym and keyCode
20130429 11:07:47 <sgothel> the simple dump of TestGearsESNewt w/ -Dnewt.debug.Window.KeyEvent property would suffice
20130429 11:07:56 <sgothel> that is most readable
20130429 11:08:01 <gouessej> but I have no access to the machine under XP yet :s
20130429 11:08:20 <sgothel> I do the tests today on Win7 and WinXP (virtual box)
20130429 11:08:27 <sgothel> w/ french layout .. np
20130429 11:09:13 <gouessej> the numpad is working, the problem comes only from the arrows
20130429 11:09:19 <sgothel> multi monitor will be finished today, yes .. it results in API change ofc
20130429 11:09:38 <gouessej> cool :) which changes?
20130429 11:10:04 <sgothel> now I realize how ignorant I was regarding the multi monitor thingy .. must be b/c of just reading XRandR 1.0 :)
20130429 11:10:45 <gouessej> I will have to update Ardor3D too
20130429 11:10:58 <sgothel> http://jogamp.org/log/irc/jogamp_20130427050628.html#l8
20130429 11:11:14 <sgothel> yes .. ofc .. and KeyTypes will be gone as well .. pretty soon!
20130429 11:11:18 <sgothel> KeyTyped!
20130429 11:11:25 <gouessej> sounds good
20130429 11:11:36 <sgothel> so best to not use it anymore already
20130429 11:11:49 <gouessej> and PointerType?
20130429 11:11:54 <sgothel> hu ?
20130429 11:12:13 <sgothel> whats wrong w/ that (which is unrelated to Newt event keyTyped)
20130429 11:12:34 <gouessej> sorry you wrote 'KeyTypes '
20130429 11:12:41 <gouessej> now I understand
20130429 11:12:44 <gouessej> I'm silly
20130429 11:13:16 <sgothel> sorry .. the use case Newt w/ AWT events is already broken, i.e. GLJPanel .. no keyTyped event already
20130429 11:13:46 <sgothel> so instead of adding it to the AWT-NEWT event factory, I guess it's best to fix that bug 688
20130429 11:14:36 <gouessej> ok
20130429 11:15:39 <gouessej> How do you want me to test without this use case?
20130429 11:15:54 <sgothel> pardon me ?
20130429 11:16:18 <sgothel> you mean test 723 beyond this use case ?
20130429 11:17:36 <gouessej> yes
20130429 11:17:52 <gouessej> You want something from NEWT, without Ardor3D
20130429 11:17:53 <sgothel> lets say we both test 723 w/ TestGearsES2NEWT w/ debug flag . and if OK .. good, or find real culprit. but lets test properly 1st
20130429 11:18:12 <sgothel> if NEWT delivers proper events .. it is good
20130429 11:18:44 <sgothel> let produce results 1st .. then decide
20130429 11:18:52 <gouessej> the behaviour is not the same under GNU Linux. If there is a problem, it comes from the native code
20130429 11:19:04 <sgothel> all possible ofc
20130429 11:21:09 <gouessej> I will probably add an automatic testing system into my game in order to be able to run tests within TUER on machines in which I cannot install a full development environment
20130429 11:22:07 <sgothel> how about: you add an Java Webstart test environment of selected unit tests of ours .. ?
20130429 11:22:36 <gouessej> On the long term it would be better, you're right
20130429 11:22:46 <sgothel> we discussed this w/ Xerxes a while ago .. but nobody picked it up yes
20130429 11:22:54 <sgothel> *yet
20130429 11:23:32 <sgothel> it's not too much work, i.e. add junit/ant jars .. and allow selection of unit tests, then package results somewhat to deliver it ..
20130429 11:23:48 <sgothel> like group selection of NEWT only, NEWT/AWT .. etc tests
20130429 11:24:23 <gouessej> I can look at how other demos are packaged
20130429 11:24:24 <sgothel> so if folks have problems on their special machine .. they could perform those tests w/o being a magician :)
20130429 11:25:04 <sgothel> it would 'just' require to write that encapsulated test launcher thingy etc
20130429 11:25:34 <sgothel> we would include that to GlueGen .. so JOAL, JOGL, .. could benefit from it
20130429 11:26:35 <gouessej> Ardor3D has an example runner, I could do the same thing
20130429 11:26:48 <gouessej> a tree with all demos
20130429 11:26:52 <gouessej> by category
20130429 11:28:05 <gouessej> Ok I have to leave. I will add some data into the bug report 723 in about 7 hours
20130429 11:28:24 <gouessej> Bye
20130429 11:28:31 * gouessej (546051e4@anon) Quit (Quit: Page closed)
20130429 11:29:36 <xranby> sgothel: i focus on the low hanging fruit to start with, right now i am reviewing the seek code. the current issue is that the video sync code delays 10s if you seek 10s forward, seeking backward is instantanious since the video sync code then dont add the delay
20130429 11:30:21 <xranby> it all boils down to that the returned frame pkt_pts value returned by libav stays at 0
20130429 11:31:18 <sgothel> :)
20130429 11:31:58 <sgothel> sounds great!
20130429 15:07:56 * odin__ (~Odin@anon) Quit (Quit: Leaving)
20130429 15:08:39 * odin_ (~Odin@anon) has joined #jogamp
20130429 16:51:41 * [Mike] (~Mike]@anon) has joined #jogamp
20130429 17:18:34 * DemoscenePassiv (~Lutsche@anon) has joined #jogamp
20130429 17:18:54 * xranby_ac100 (~xranby@anon) has joined #jogamp
20130429 19:15:07 * DemoscenePassiv (~Lutsche@anon) Quit ()
20130429 20:28:33 * xranby_ac100 (~xranby@anon) Quit (Ping timeout: 240 seconds)
20130429 21:11:26 * void256 (~void@anon) has joined #jogamp
20130429 21:48:51 * gouessej (5279c7b1@anon) has joined #jogamp
20130429 21:48:59 <gouessej> Hi
20130429 21:49:56 <gouessej> Why is the surface permanently changed in NEWT under Windows 7? I always get logs with this content: "WindowsWindow: surface change 0x....... -> 0x........."
20130429 22:06:04 * gouessej (5279c7b1@anon) Quit (Quit: Page closed)
20130429 23:14:21 * void256 (~void@anon) Quit (Quit: ChatZilla 0.9.90 [Firefox 20.0/20130326150557])
20130430 01:14:36 * odin_ (~Odin@anon) Quit (Ping timeout: 264 seconds)
20130430 01:41:52 * odin_ (~Odin@anon) has joined #jogamp
20130430 04:52:16 * [Mike] (~Mike]@anon) Quit ()
20130430 05:06:09 -CatOut- Continue @ http://jogamp.org/log/irc/jogamp_20130430050609.html