#jogamp @ irc.freenode.net - 20130702 05:05:23 (UTC)

20130702 05:05:23 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20130701181922.html
20130702 05:05:23 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20130702050523.html
20130702 05:17:12 <olamedia> >_< caps.setBackgroundOpaque(false); is working too..
20130702 06:45:39 * monsieur_max (~maxime@anon) has joined #jogamp
20130702 07:12:44 * [Mike] (~Mike]@anon) Quit (Read error: Connection reset by peer)
20130702 07:43:40 <sgothel> Good morning .. hopefully final version <http://jogamp.org/doc/siggraph2013/t-shirt/JogAmpBOFSiggraph2013TShirtBack-path.svg>
20130702 07:44:09 <sgothel> Xerxes hint converting text to vectors .. was essential.
20130702 07:44:32 <sgothel> @Mark: Maven and jspatial added as well
20130702 13:06:56 <rmk0> hm... can't seem to open a 'core' 3.1 context on mesa 9.1.3 on AMD
20130702 13:07:05 <rmk0> sgothel: eyeing it now
20130702 13:12:01 <sgothel> https://jogamp.org/git/?p=jogl.git;a=commit;h=205a17de1abec614d3d06386f873170fc1691e86 ... another 'orgy' of X11 EWMH, fullscreen and mode selection .. duh
20130702 13:12:37 <sgothel> @Mark: Disabled by quirk .. see w/ jogl.debug.GLContext ..
20130702 13:12:43 <sgothel> -Djogl.debug.GLContext
20130702 13:13:13 <sgothel> https://jogamp.org/git/?p=jogl.git;a=commit;h=a053732720628d57d38bda401053833259bf1657
20130702 13:13:29 <sgothel> https://jogamp.org/git/?p=jogl.git;a=blobdiff;f=src/jogl/classes/jogamp/opengl/GLContextImpl.java;h=d6f97662eb679109a3ebec0d8fd1f6c760df0df6;hp=6254b6f44bc4ee853bbeaef49bc26b9479a91431;hb=a053732720628d57d38bda401053833259bf1657;hpb=4b7833fab73166338a3a2f27920a7b28a3e5e30f
20130702 13:13:37 * rmk0 eyes it
20130702 13:13:52 <sgothel> if you can confirm .. it would work for your driver version .. we can confine the quirk range ..
20130702 13:14:07 <sgothel> import: is the 'driver vendor version' visible ?
20130702 13:14:27 <rmk0> is there some definitive way i can confirm it?
20130702 13:14:48 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=759
20130702 13:15:03 <rmk0> got it
20130702 13:15:06 <sgothel> "the main issue is that i dont get to render anything on screen. Not from my rendering routines and not from any of jogl routines (like text rendering)."
20130702 13:15:17 <sgothel> so .. if it works for you .. rendering anything .. well
20130702 13:16:04 <sgothel> currently we really don't know what criteria for this bug are valid .. driver version, gl ctx, even the hardware ?
20130702 13:19:37 <rmk0> is there a way to override the quirk without patching the jogl source?
20130702 13:19:57 <rmk0> i could write something quickly to test rendering if i could force it to open the context
20130702 13:20:56 <rmk0> otherwise it'll be something done in C with glfw
20130702 13:22:49 <sgothel> can't compile jogl ?
20130702 13:23:05 <sgothel> but .. yes .. good idea in general to retest!
20130702 13:23:39 <rmk0> i can compile jogl, it's just... rather long winded for such a simple thing
20130702 13:23:42 <sgothel> you may have noticed .. it widens criteria of the quirk disabling using intel/mesa 3.1 .. we did that once
20130702 13:23:49 <rmk0> yep
20130702 13:24:06 <sgothel> I can add this for the future .. sure .. won't help you right now though :)
20130702 13:24:12 <rmk0> hehe
20130702 13:24:22 <sgothel> come on .. delete a few lines and kick it off :)
20130702 13:24:34 <rmk0> /o\
20130702 13:24:51 <sgothel> so how is the new t-shirt ?
20130702 13:24:59 <rmk0> hectic!
20130702 13:25:02 <sgothel> I played w/ inkscape a lot :)
20130702 13:25:05 <sgothel> oh ..
20130702 13:25:10 <sgothel> too noisy ?
20130702 13:25:10 <rmk0> i like it
20130702 13:25:14 <sgothel> ah :)
20130702 13:25:37 <rmk0> certainly gets the point across
20130702 13:25:39 <sgothel> had to find idea for the java env .. and OS .. so sun and water/sea-gal ..
20130702 13:28:11 <sgothel> now media comes out w/ all the google/nsa glass reports .. hehe .. I would say 'GREAT timing' :)
20130702 13:28:21 <sgothel> reports .. = reviews
20130702 14:34:51 <rmk0> sgothel: more info added to 759
20130702 14:47:03 <sgothel> Mark .. your comment assumes the discussion above didn't happen :)
20130702 14:47:31 <rmk0> jogl build is in progress
20130702 14:47:38 <sgothel> what I would like to have is our dump of the GLContext versions ..
20130702 14:48:14 <rmk0> was mainly trying to answer the question "is AMD's 3.1 context actually capable of output at all" first of all
20130702 14:48:22 <rmk0> seems it is... JOGL comes next
20130702 15:00:55 <rmk0> ... wow
20130702 15:01:34 <rmk0> i just ran the unit tests
20130702 15:01:45 <rmk0> i did see some "3.1 (Core Profile) Mesa 9.1.3" go past
20130702 15:01:50 <rmk0> glxgears is ... screwed
20130702 15:01:59 <rmk0> in some contexts, it's fine
20130702 15:02:05 <rmk0> will try to get some screenshots from individual tests
20130702 15:02:10 <rmk0> waiting for eclipse to import and build
20130702 15:15:47 <sgothel> ah .. so the quirk is still valid ..
20130702 15:16:00 <rmk0> i think quirk may not be a strong enough word for it
20130702 15:16:01 <sgothel> thx for confirmation .. pls state this fact in that bug ..
20130702 15:16:07 <sgothel> :)
20130702 15:16:14 <rmk0> yep, preparing something
20130702 15:16:15 <sgothel> I used the word from driver development
20130702 15:16:26 <sgothel> Linux kernel .. uses that word as well
20130702 15:16:37 <sgothel> i.e. our quirk .. to bugs
20130702 15:17:01 <rmk0> yep
20130702 15:17:07 <rmk0> http://waste.io7m.com/2013/07/02/TestGearsES2NEWT.test03_GL3___-n0000-GL3___-hw-onscreen-Bdbl-Frgb__Irgba8880-D24-St00-Sa00_default-0640x0480.png
20130702 15:17:13 <rmk0> looks ok
20130702 15:17:15 <rmk0> as does...
20130702 15:17:19 <rmk0> http://waste.io7m.com/2013/07/02/TestGearsES2NEWT.test02_GLES2_-n0000-GLES2_-hw-onscreen-Bdbl-Frgb__Irgba8880-D16-St00-Sa00_default-0640x0480.png
20130702 15:17:23 <rmk0> and then suddenly:
20130702 15:17:27 <rmk0> http://waste.io7m.com/2013/07/02/TestGearsES2NEWT.test01_GL2ES2-n0000-GL3bc_-hw-onscreen-Bdbl-Frgb__Irgba8880-D24-St00-Sa00_default-0640x0480.png
20130702 15:18:11 <sgothel> so only the non-core 3.1 is buggy for AMD ? great!
20130702 15:18:46 <rmk0> i'm trying to determine which context does that, actually... trying to decipher the test output
20130702 15:19:08 <rmk0> it's almost certainly the 3.1 core, but it may be the 3.1 core configured in a particular way
20130702 15:19:23 <sgothel> thats GREAT, maybe we can enable GL3 core .. i.e. confine criteria for AMD .. awesome, pls be verbose in bug report :)
20130702 15:19:42 <sgothel> we distinguish core / bc in our ctx creation!
20130702 15:19:53 <sgothel> i.e. do not alias ..
20130702 15:20:02 <rmk0> http://waste.io7m.com/2013/07/02/gl3.txt
20130702 15:20:18 <rmk0> does that make any sense to you? that's TestGearsES2NEWT.test01_GL2ES2 run with -Djogl.debug.GLContext
20130702 15:21:28 <rmk0> http://waste.io7m.com/2013/07/02/gl3_2.txt
20130702 15:21:55 <rmk0> and that's the same test but presumably with the GL3 profile as opposed to GL2ES2
20130702 15:22:57 <rmk0> test01_GL2ES2 shows broken output, test03_GL3 shows good output, to clarify
20130702 15:23:50 <rmk0> ... ah shit, i don't think this source is current
20130702 15:23:52 * rmk0 kicks git
20130702 15:23:59 <rmk0> ugh
20130702 15:24:16 <rmk0> yes, sorry... i need to pull again. i thought i was running the current source, but i'm apparently not
20130702 15:24:28 <rmk0> disregard the above...
20130702 15:24:28 <sgothel> np
20130702 15:25:52 <sgothel> I also expect to see ", vendor <venodor-version>" .. somewhere ..
20130702 15:26:12 <sgothel> JOGLVersion.getGLStrings(..)
20130702 15:26:38 <sgothel> .. or GLContext.toString() ..
20130702 15:26:45 <sgothel> *vendor-version*
20130702 15:27:50 <sgothel> so you say we may can do:
20130702 15:27:54 <sgothel> - ( (glRenderer.contains( MesaRendererIntelsp ) && compatCtx) || glRenderer.contains( MesaRendererAMDsp ) )
20130702 15:28:20 <sgothel> + ( ( glRenderer.contains( MesaRendererIntelsp ) || glRenderer.contains( MesaRendererAMDsp ) && compatCtx) )
20130702 15:28:50 <sgothel> I was asking Juan to test this AFAIK .. but he mentioned it doesn't work .. hmm
20130702 15:28:59 <sgothel> event though I doubt he tested a core profile :)
20130702 15:29:35 <sgothel> see his description ..
20130702 15:29:51 <sgothel> " if( compatCtx && ( major > 3 || major == 3 && minor >= 1 ) ) " <- he said it fails .. hmm
20130702 15:30:24 <rmk0> i can't speak for the major > 3
20130702 15:30:28 <rmk0> this hardware is too old for GL4
20130702 15:30:36 <sgothel> lol
20130702 15:30:47 <sgothel> we are talking Mesa right :)
20130702 15:30:53 <rmk0> yep
20130702 15:31:04 <rmk0> it does seem from running the pre-quirk source that the 3.1 context does work in some cases
20130702 15:31:18 <rmk0> the GL2ES2 profile upset it
20130702 15:31:21 <sgothel> so he said .. ' if( compatCtx && ( major == 3 && minor >= 1 ) ) ' doesn't work ..
20130702 15:31:47 <rmk0> just waiting for jogl to finish building
20130702 15:31:48 <sgothel> yes .. you would need to request GL3
20130702 15:32:06 <sgothel> i.e. a core only profile
20130702 15:32:20 <rmk0> just want to make sure we're not misunderstanding each other
20130702 15:32:22 <sgothel> I guess we have a nice get-method for this
20130702 15:32:33 <rmk0> those tests i ran from the older source... in all cases, it seemed to be using the 3.1 core profile, yes?
20130702 15:32:41 <sgothel> the question here is: can we allow AMD+Mesa+GL>=3.1 ?
20130702 15:32:41 <rmk0> as in, before the quirk was put in place
20130702 15:33:14 <sgothel> yes
20130702 15:33:35 <sgothel> rephrase: can we allow AMD+Mesa+GL>=3.1-core ?
20130702 15:33:37 <rmk0> but the test that asked for a GL2ES2 profile produced broken output (if that is what the test is doing, i don't fully understand it yet)
20130702 15:33:57 <rmk0> i'm still trying to determine that... waiting for jogl to finish building
20130702 15:34:03 <rmk0> i think it's allowable in ... some cases
20130702 15:34:04 <sgothel> if GL3.1-bc is avail, it's used if requesting GL2ES2
20130702 15:34:27 <sgothel> if not, GL2 ..( GL 3.0 e.g.)
20130702 15:34:40 <sgothel> yes, Mesa has a new software path for >= 3.1 !
20130702 15:34:43 <rmk0> i'm confused...
20130702 15:34:49 <rmk0> ++++ UITestCase.setUp: com.jogamp.opengl.test.junit.jogl.demos.es2.newt.TestGearsES2NEWT - test01_GL2ES2
20130702 15:35:07 <rmk0> if you look at what that test is actually given, it receives:
20130702 15:35:08 <rmk0> GL_VENDOR: X.Org
20130702 15:35:08 <rmk0> GL_RENDERER: Gallium 0.4 on AMD RV730
20130702 15:35:08 <rmk0> GL_VERSION: 3.1 (Core Profile) Mesa 9.1.3
20130702 15:35:56 <sgothel> what is jogl saying ? .. i.e. one of our print version (of GLContext.toString())
20130702 15:37:35 <sgothel> System.err.println("GL:" + gl + ", " + gl.getContext().getGLVersion());
20130702 15:37:43 <sgothel> that output line of GearsES2 .. pls :)
20130702 15:37:44 <rmk0> one sec... every time i try to do something, eclipse decides to rebuild both gluegen and jogl
20130702 15:37:56 <sgothel> I never use eclipse to build ..
20130702 15:38:12 <sgothel> too slow .. etc
20130702 15:38:48 <rmk0> oh dear
20130702 15:40:26 <rmk0> ok, let's see what we have here...
20130702 15:43:08 <rmk0> sigh
20130702 15:43:14 * rmk0 waits for eclipse to start responding again
20130702 15:43:42 <sgothel> use cmd-line scripts to build .. (and test) .. guess why I have 'em there :)
20130702 15:44:38 <rmk0> yeah, this is enormously painful
20130702 15:44:44 <rmk0> let me... try something else
20130702 15:45:29 <rmk0> ok, right... results are as expected
20130702 15:45:41 <rmk0> the GLES2, GL2ES2 tests both get given a non-core context
20130702 15:45:43 <sgothel> order GL2ES2: GL4bc, GL3, GL3bc, GL3, GL2, GLES2 ...
20130702 15:45:49 <rmk0> and the GL3 test doesn't run, obviously
20130702 15:46:00 <rmk0> so now i need to try undoing the quirk slightly, and running again
20130702 15:46:35 <rmk0> the tests that do run give correct output
20130702 15:46:36 <sgothel> btw: you agree that enter, backspace, tab are printable keys ?
20130702 15:47:00 <rmk0> hm... not sure about backspace
20130702 15:47:03 <sgothel> great .. can you dump the lines here .. only the last one
20130702 15:47:19 <sgothel> GearsES2: System.err.println("GL:" + gl + ", " + gl.getContext().getGLVersion());
20130702 15:47:24 <sgothel> that line pls :)
20130702 15:47:40 <rmk0> for the two tests that do execute?
20130702 15:47:45 <sgothel> yup
20130702 15:50:45 <rmk0> http://waste.io7m.com/2013/07/02/current_gl2es2.txt
20130702 15:50:48 <rmk0> http://waste.io7m.com/2013/07/02/current_gles2.txt
20130702 15:52:21 <sgothel> so what is your summary ? core works and compatible not (i.e. disabling by quirk is ok here) ?
20130702 15:52:54 <rmk0> well, i've not made any changes yet... i'm going to try disabling the quirk partly and running the GL3 test to see what happens
20130702 15:53:09 <rmk0> that does sound right though... may be that their 3.1 compability context is broken
20130702 15:53:12 <sgothel> note: OpenGL ES 3.0 Mesa 9.1.3 :)
20130702 15:53:16 <rmk0> *compatibility
20130702 15:53:32 <rmk0> i suspect core works and compatibility is buggy
20130702 15:53:32 <sgothel> pls test w/ forceGL3 !
20130702 15:53:41 <rmk0> what is forceGL3?
20130702 15:53:57 <sgothel> the test ..see rgs
20130702 15:54:06 <sgothel> see args '-es2' '-es3'
20130702 15:54:15 <sgothel> sorry .. '-gl3'
20130702 15:54:15 <rmk0> ah, i was running them as junit
20130702 15:54:18 <rmk0> right
20130702 15:54:45 <sgothel> so having test default 'gl2es2' (highes profile) .. and -es2 and -gl3 (core only)
20130702 15:54:51 <rmk0> you'll have to bear with me, i'm not too familiar with the codebase
20130702 15:55:12 <sgothel> np .. can't expect anybody understanding my complicated crap :)
20130702 15:55:17 <rmk0> hehe
20130702 15:55:38 <sgothel> .. which .. OFC .. makes it easy in the end .. haha
20130702 15:56:10 <rmk0> i assume you were expecting to see this:
20130702 15:56:11 <rmk0> http://waste.io7m.com/2013/07/02/gl3_nope.txt
20130702 15:56:17 <rmk0> that's with current source with no changes
20130702 15:56:19 <rmk0> just to be sure...
20130702 15:56:33 <olamedia> ohayou
20130702 15:56:42 <sgothel> yes
20130702 15:56:46 <rmk0> right
20130702 15:56:49 * rmk0 eyes quirks
20130702 15:57:13 <sgothel> so restricting the quirk .. to compatible only .. should enable test .. and we see whether it works
20130702 15:57:45 <sgothel> that would bring intel & amd quirk up to par .. i.e. hinting that it is independent of the actual driver :)
20130702 15:57:56 <olamedia> what is a correct way to avoid flickering when inserting NEWT into JFrame? caps.setBackgroundOpaque(false); is working, but i don't understand why
20130702 15:57:59 <sgothel> then we could remove intel and amd alltogether and blame it on mesa core
20130702 15:58:05 <rmk0> i suspect they don't like working on the compatibility context at all
20130702 15:58:06 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20130702 15:58:11 <rmk0> lot of cruft in there, obviously
20130702 15:58:17 <sgothel> yup
20130702 15:58:20 <rmk0> get the feeling they badly want everyone to move to core
20130702 15:58:38 <sgothel> well .. nobody there to work on ffp ..
20130702 15:58:45 <rmk0> nope
20130702 15:59:22 <sgothel> @Ola: I assume you talk about NewtCanvasAWT .. well, I don't see those artifacts in our unit tests
20130702 15:59:43 <olamedia> probably because of caps.setBackgroundOpaque(false);
20130702 15:59:56 <sgothel> you may file a report (best forum 1st) .. w/ proper details, platform, java -> Wiki/FAQ
20130702 16:00:00 <olamedia> i saw this in demo
20130702 16:00:21 <olamedia> i mean, i did found caps.setBackgroundOpaque(false); in demo and tried
20130702 16:00:22 <sgothel> I don't think we use this in all our unit tests
20130702 16:00:36 <sgothel> only in one transparency demo
20130702 16:00:53 <olamedia> that was a demo with NEWTCnvas+jfarme
20130702 16:01:08 <sgothel> precise .. URL of unit test . git scm ..
20130702 16:07:49 <olamedia> https://github.com/olamedia/modloader - Launcher https://github.com/olamedia/graphics-core as jar into modloader /coremods folder, jogl into /lib
20130702 16:07:50 <rmk0> sgothel: http://waste.io7m.com/2013/07/02/current_reduced_quirk_gl3.txt
20130702 16:08:06 <rmk0> forcing GL3, gives correct output with a 3.1 core context \o/
20130702 16:08:19 <rmk0> don't worry, i'm writing up all of this stuff for the bug report
20130702 16:09:01 <sgothel> sweet ..
20130702 16:09:11 <olamedia> GraphicsCoreMod.java:80 caps.setBackgroundOpaque(false); when set to true, i can see flickering on ubuntu + ati 13.1 legacy drivers
20130702 16:09:27 <sgothel> in this case .. I vote to simplify quirk, i.e. Mesa GL >= 3.1 only for core - at all !
20130702 16:09:39 <sgothel> (sine we know they don't work on compat ..)
20130702 16:09:43 <sgothel> *since*
20130702 16:10:02 <rmk0> http://waste.io7m.com/2013/07/02/diff.txt
20130702 16:10:07 <rmk0> that's my current diff, not sure if you want to adjust that
20130702 16:10:25 <sgothel> see above .. i.e.
20130702 16:10:44 <sgothel> if ( compatCtx && ( major > 3 || major == 3 && minor >= 1 ) ) { ..}
20130702 16:11:04 <olamedia> Better to watch with GraphicsCoreMod.java:122 animator = new FPSAnimator(60); set to 1 frame per second
20130702 16:11:06 <sgothel> .. we are already in the Mesa branch :)
20130702 16:11:13 <rmk0> right
20130702 16:11:50 <sgothel> maybe Juan .. didn't test right .. since he claimed it doesn't work - so I take your word / test - makes sense!
20130702 16:12:27 <olamedia> With this settings I can see a race to draw background between opengl calls and jframe
20130702 16:12:51 <olamedia> And jframe was winning :/
20130702 16:12:53 <sgothel> @Ola: pls send an email to forum w/ proper data .. Wiki/FAQ we need all infos .. incl. test case / machine / etc
20130702 16:13:29 <olamedia> >_> Not enought info i have provided already?
20130702 16:13:47 <sgothel> if you read the FAQ - you know that you did not!
20130702 16:14:11 <sgothel> plus .. above reference is a none ref .. class is not part of jogl .. I asked for an URL to git repo
20130702 16:14:41 <olamedia> above ^ i send 2 url to 2 git repos
20130702 16:14:49 <olamedia> *sent
20130702 16:14:54 <olamedia> *urls
20130702 16:16:41 <sgothel> that is not a unit test of ours sir
20130702 16:17:04 <sgothel> also not a complete URL to the actual file / line .. tsts
20130702 16:17:43 <olamedia> https://github.com/olamedia/graphics-core/blob/master/src/ru/olamedia/graphics/GraphicsCoreMod.java#L80
20130702 16:18:23 <sgothel> and what does this proof Sir ? You mentioned some official code where we do set this ..
20130702 16:19:34 <sgothel> however .. pls write a small _PROPER_ unit tests, maybe based of one of ours .. and add some description w/ full bug report data please. otherwise this info is lost.
20130702 16:20:06 <sgothel> such behavior is based on many side effects .. sloppiness won't help us here - must have accurate data
20130702 16:21:17 <olamedia> >_> oh... and u saying i'm lazy
20130702 16:21:21 <olamedia> u can't read 20 lines of code?
20130702 16:21:39 <sgothel> I guess you have the wrong perception of what it takes to manage this code base
20130702 16:22:05 <sgothel> I am very glad that most follow bug report recommendations - I can't thank these participants enough
20130702 16:22:35 <olamedia> create NewtCanvasAWT with caps.setBackgroundOpaque(true); JFrame frame.add(canvas); frame.pack(); than start animator
20130702 16:22:49 <sgothel> also - me and all of us being here at service .. is a great services - bu we cannot just simply pull everything out of ones nose :)
20130702 16:22:50 <olamedia> *then
20130702 16:22:54 <olamedia> that's all
20130702 16:23:36 <sgothel> what you ask here is outrageous, i.e. me asking to follow your desired workflow - instead of taking our proven advise and follow our workflow - which .. yes, works :)
20130702 16:23:39 <olamedia> and with legacy 13.1 ati drivers on ubuntu radeon 3000 u will see flickering
20130702 16:23:56 <sgothel> @Ola: File a freaking bugreport _PLEASE_ thank you
20130702 16:24:10 <sgothel> @Mark: Pls help me somebody here :)
20130702 16:24:11 <olamedia> i DO NOT have a bug already
20130702 16:24:32 <sgothel> flickering is a bug already
20130702 16:25:03 <olamedia> create NewtCanvasAWT with caps.setBackgroundOpaque(true); JFrame frame.add(canvas); frame.pack(); than start animator and with legacy 13.1 ati drivers on ubuntu radeon 3000 u will see flickering
20130702 16:25:11 <olamedia> taht's all
20130702 16:26:37 <olamedia> *Until first resize of jframe
20130702 16:27:19 <olamedia> When playing minecraft i have to resize window before i start game because of flickering
20130702 16:29:52 <olamedia> latest 13.6 ati drivers supports radeon starting from 5000
20130702 16:29:58 <olamedia> only :/
20130702 16:30:45 <sgothel> @Mark: You provide a git pull req ?
20130702 16:31:08 <olamedia> 13.1 fixed Source engine games, but i din't seen flickering on old drivers
20130702 16:32:27 <olamedia> System.setProperty("sun.java2d.opengl", "True") can fix this too
20130702 16:32:30 <rmk0> sgothel: er, added a diff as an attachment... d'you want me to fork and patch instead?
20130702 16:32:50 <sgothel> you have a fork / branch already - all is a branch under git :)
20130702 16:33:01 <rmk0> hehe, yes was just typing "i'd actually forgotten i had my own fork there..."
20130702 16:33:05 <sgothel> thought it's nice to have you in the history :)
20130702 16:33:06 * rmk0 updates
20130702 16:33:54 <sgothel> -- if(glRenderer.contains( MesaRendererIntelsp ) || (glRenderer.contains( MesaRendererAMDsp )))
20130702 16:34:06 <sgothel> only: if (compatCtx && (major > 3 || (major == 3 && minor >= 1))) {
20130702 16:34:39 <sgothel> important that you test this .. me not having a machine set up .. but will soon .. (1 week or so)
20130702 16:35:00 <rmk0> yep
20130702 16:35:02 <sgothel> i.e. quirk GPU vendor agnostic .. just Mesa >= 3.1 && compat
20130702 16:35:11 <rmk0> ah, right
20130702 16:35:17 <rmk0> so... widen the scope further still?
20130702 16:35:33 <sgothel> yes. .. we claim: don't trust Mesa >= 3.1 compatibility
20130702 16:36:02 <sgothel> you can also say 'why' in the actual quirk description .. (i.e. hey do not develop FFP .. anymore)
20130702 16:36:24 <sgothel> GLRendererQuirks.GLNonCompliant <- in source code
20130702 16:37:00 * rmk0 eyes it
20130702 16:37:15 <sgothel> all quirks are sort of documented there ..
20130702 16:37:27 <sgothel> and easy to track if code branches due to a quirk .. etc
20130702 16:39:21 <rmk0> http://waste.io7m.com/2013/07/02/doc.txt
20130702 16:39:24 <rmk0> is that sufficient?
20130702 16:39:27 <rmk0> will test on intel shortly
20130702 16:44:33 <sgothel> pefect ..
20130702 16:44:49 <rmk0> running test suites
20130702 16:44:55 <sgothel> ENTER key .. is not printable and hence a control key ?
20130702 16:45:06 <sgothel> have to move NEWT ENTER 0xa -> 0xd ..
20130702 16:45:15 <sgothel> 0xa is line_feed .. ooops
20130702 16:45:16 <rmk0> .. tricky
20130702 16:45:30 <rmk0> possibly silly question, but what makes something "printable"?
20130702 16:45:31 <sgothel> line_feed .. is n/a on keyboards
20130702 16:45:36 <sgothel> right
20130702 16:45:46 <sgothel> BS and TAB .. I get it ..
20130702 16:46:01 <sgothel> i.e. sort of visible if seen from editor perspective ..
20130702 16:46:11 <sgothel> but not from a font definition perspective :)
20130702 16:46:20 <sgothel> i.e. empty
20130702 16:47:11 <sgothel> Bug 767 .. maybe a compromise ? i.e. incl BS + TAB, but still exclude ENTER .. while fixing ENTER to 0xd
20130702 16:47:20 <sgothel> sounds reasonable .. I guess
20130702 16:47:33 <sgothel> SPACE is printable as well .. sure
20130702 16:48:31 <olamedia> enter should be 13 ^_^
20130702 16:48:52 <sgothel> yup
20130702 16:49:13 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=767#c2
20130702 17:00:12 * hharrison (~chatzilla@anon) has joined #jogamp
20130702 17:03:26 <hharrison> sgothel: nice job on the shirt design
20130702 17:05:55 <sgothel> Hi Harvey .. thx to your suggestions ofc
20130702 17:07:06 <sgothel> so the *-path.svg file should go to the printer .. obviously .. great hint from Xerxes
20130702 17:07:31 <sgothel> .. and the front svg .. sure (not text in there)
20130702 17:07:48 <sgothel> just stating the obvious .. :)
20130702 17:08:08 <hharrison> I'm just going to point them at https://jogamp.org/doc/siggraph2013/t-shirt/
20130702 17:08:28 <sgothel> but pls tell them the files they should use ..
20130702 17:08:48 <sgothel> the *-text.svg .. would not produce 1:1 results .. dunno how professional they are
20130702 17:08:55 <sgothel> usually less info is better :)
20130702 17:09:15 <hharrison> Yeah, I'll maybe point them directly at the two svgs
20130702 17:09:22 <sgothel> :]
20130702 17:09:41 <sgothel> w/ all those bugs .. we worked one .. we know .. anything can happen :)
20130702 17:10:04 <sgothel> and the PNG files .. just for verification .. like my email ..
20130702 17:10:10 <sgothel> so they know ..
20130702 17:10:28 <hharrison> So frustrated with the Canvas3d position bug, going to let that go dormant until after -u40 comes out
20130702 17:10:35 <sgothel> maybe CCT will pay for shirts .. will see
20130702 17:10:56 <sgothel> I will work on this probably today / tomorrow .. w/ the beta ..
20130702 17:11:07 <sgothel> will make a branch .. regarding JVM version
20130702 17:11:20 <sgothel> Xerxes pointed me to the tests of one guy ..
20130702 17:11:35 <sgothel> have you attached anything to a related bug ?
20130702 17:11:56 <sgothel> AFAIK .. this bug is sadly not yet filed (again) ?
20130702 17:11:57 <hharrison> Not at the moment, been doing remote debugging with an OSX user...debug by println
20130702 17:12:34 <sgothel> my advise: don't waste time on this now - lets re-test when I have a patch proposal when using the new beta40 ..
20130702 17:12:52 <sgothel> as we know .. current workaround doesn't work reliable
20130702 17:13:04 <sgothel> plus: workaround must be removed w/ beta40 ..
20130702 17:13:05 <hharrison> Wait a minute, did I get confused then, I thought this was what we were waiting for in -u40
20130702 17:13:07 <hharrison> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=2229714
20130702 17:13:35 <sgothel> yes .. but we have a workaround for it ..
20130702 17:13:59 <sgothel> so when testing w/ the beta form Oracle (which incl. this fix) .. it won't work either
20130702 17:14:19 <sgothel> hence I have to test w/ their beta Java version .. and disable workaround
20130702 17:14:27 <hharrison> Ahhh, fudge
20130702 17:14:32 <sgothel> related to: NEWT and NewtCanvasAWT
20130702 17:14:59 <sgothel> AWT .. errr .. we have no workaround ..
20130702 17:15:06 <sgothel> i.e. pure AWT
20130702 17:15:16 <hharrison> Yeah, I imagine that part is on me to figure out
20130702 17:15:37 <sgothel> but it will be out after siggraph .. so just ignore it :)
20130702 17:15:39 <hharrison> But I'm going to do it once at u40 time, and forget about anythign earlier on osx
20130702 17:15:47 <sgothel> yup
20130702 17:15:59 <sgothel> w/ u40 it's still buggy ?
20130702 17:16:26 <sgothel> http://forum.jogamp.org/Mac-OS-X-10-7-Oracle-JRE-7-Swing-integration-issues-td4027780.html#a4029292
20130702 17:16:54 <sgothel> so I do my 7u40 tests until tomorrow .. as described above
20130702 17:17:22 <sgothel> sadly .. he didn't file a bug report .. oh well, I will do tomorrow then
20130702 17:19:41 <sgothel> So they will give us a quote today / tomorrow ? If <= 1700 CAD .. we say yes. If 'much' lower, take more 10-20 t-shirts :)
20130702 17:19:57 <sgothel> Qun is concerned about your luggage, i.e. it may be heavy ?
20130702 17:20:09 <sgothel> 110.. hmm
20130702 17:20:24 <sgothel> dunno whats the weight, but probably .. not little
20130702 17:21:39 <sgothel> We can pick you up at the airport .. if you like.
20130702 17:29:14 <sgothel> dinner break .. laters
20130702 17:30:36 * monsieur_max (~maxime@anon) has joined #jogamp
20130702 17:35:32 * [Mike] (~Mike]@anon) has joined #jogamp
20130702 17:38:21 * Eclesia (~eclesia@anon) has joined #jogamp
20130702 17:38:30 <Eclesia> hi
20130702 17:43:51 <hharrison> sorry, got pulled into a meeting, should have a firm quote tonight/tomorrow morning
20130702 17:44:53 <hharrison> hello
20130702 17:46:29 * olamedia_ (~olamedia@anon) has joined #jogamp
20130702 17:49:43 * olamedia (~olamedia@anon) Quit (Ping timeout: 264 seconds)
20130702 18:00:33 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20130702 18:11:59 * olamedia_ (~olamedia@anon) has left #jogamp
20130702 18:12:36 * olamedia (~olamedia@anon) has joined #jogamp
20130702 18:12:48 * olamedia (~olamedia@anon) Quit (Client Quit)
20130702 18:16:13 * olamedia (~olamedia@anon) has joined #jogamp
20130702 18:17:17 * olamedia (~olamedia@anon) Quit (Client Quit)
20130702 18:17:28 * olamedia (~olamedia@anon) has joined #jogamp
20130702 18:22:22 <sgothel> hello .. dinner not quite ready, but I got a haircut :)
20130702 18:22:47 <sgothel> @Harvey: Thx .. maybe consider t-shirt weight .. :)
20130702 18:24:08 * olamedia (~olamedia@anon) Quit (Quit: Ухожу я от вас (xchat 2.4.5 или старше))
20130702 18:24:20 * olamedia (~olamedia@anon) has joined #jogamp
20130702 18:24:47 * olamedia (~olamedia@anon) Quit (Client Quit)
20130702 18:25:14 * olamedia (~olamedia@anon) has joined #jogamp
20130702 18:25:33 <hharrison> We'll see what 80-110 t-shirts weighs
20130702 18:26:49 <sgothel> if too much, lets consider snail mail .. hmm
20130702 18:27:07 <sgothel> (to anaheim ofc) .. you can bring only one bag right ?
20130702 18:27:48 <sgothel> we will see .. you have my hotel address in Anaheim ?
20130702 18:28:02 <hharrison> check one bag, carry on one bag
20130702 18:28:11 <hharrison> Yeah, I have the address
20130702 18:28:20 <sgothel> so lets see ..
20130702 18:29:23 <sgothel> 110 - 120 :) .. getting more requests every day :)
20130702 18:29:43 <olamedia> olamedia
20130702 18:29:48 <olamedia> hmm
20130702 18:30:27 <sgothel> sure .. will mail one from Germany to Mark, Ola .. etc
20130702 18:31:15 <olamedia> test
20130702 18:31:36 <olamedia> ok.. was tuning xchat to autojoin channel and beep on messages
20130702 18:33:33 <olamedia> What is the best way to compile one project with specified jdk and then run another project using eclipse?
20130702 18:34:11 <hharrison> properties->java compiler->building->enable project specific settings
20130702 18:34:11 <olamedia> atm ant failing specify jdk
20130702 18:34:57 <olamedia> >_> and? i want to build & run another project pressing a run button
20130702 18:35:30 <olamedia> project specific settings do not apply for Ant
20130702 18:35:37 <hharrison> set up a new run configuration that specifies a particuler jre
20130702 18:36:08 <olamedia> @hharrison u missed the second part
20130702 18:36:32 <olamedia> i need to build jar to some folder & run another project
20130702 18:37:12 <olamedia> mb maven can do this?
20130702 18:37:27 <olamedia> i'm trying ant atm
20130702 18:39:29 <hharrison> Why not just make one project a dependnecy of the other in eclipse?
20130702 18:39:54 <olamedia> and?
20130702 18:40:14 <hharrison> and if the project needs to be built, it will be before running the other
20130702 18:40:40 <olamedia> well, i need to compile only changed project and i need to build jar
20130702 18:40:54 <hharrison> Well, if you are directly using the jar, eclipse will not help you
20130702 18:41:00 <olamedia> i don't want to build all dependencies each time
20130702 18:41:01 <hharrison> so...don't do that?
20130702 18:41:14 <hharrison> It will only build things that have changed
20130702 18:42:13 <olamedia> but what about jar? i have some modloader and need to place dependencies (compiled jars) into mod folder under modloader project
20130702 18:42:50 <hharrison> Hmm, doubt eclipse will help you there
20130702 18:43:54 <sgothel> @Harvey: You didn't tell me we missed Aussenco's JamSim .. duh
20130702 18:44:09 <hharrison> hah!
20130702 18:44:15 <sgothel> indeed sir :)
20130702 18:44:22 <sgothel> me adding 'eh ?
20130702 18:44:43 <sgothel> *plus the little link from jogl/www ..
20130702 18:44:44 <hharrison> Of course I'd appreciate it, but obvisouly I'm a terrible promoter
20130702 18:44:54 <sgothel> ok .. doing now
20130702 18:45:25 <hharrison> Oh, I have that text 1nd 160x160 png, was waiting on a final slide to send you the whole package
20130702 18:45:35 <sgothel> sweet
20130702 18:46:01 <hharrison> A good java3d demo fell in my lap last week
20130702 18:46:19 <hharrison> So I'll use that to show that it's still alive
20130702 18:46:49 <sgothel> '+Ausenco's JaamSim' ok ?
20130702 18:47:36 <hharrison> I'd prefer just JaamSim
20130702 18:48:13 <sgothel> ok .. if you don't like promoting :) .. other names are in though
20130702 18:48:13 <hharrison> We're trying to make it clear one is an opensource project and there's a commercial offering built on top, but separate
20130702 18:48:32 <sgothel> Sir - 'open source' does not mean 'no company' and not mean 'no profit' !
20130702 18:48:56 <sgothel> Qun, my wife & boss would be angry otherwise :)
20130702 18:49:00 <hharrison> If they were paying for my trip, I'd put the name on it, as it's my dime, I'm promoting the opensource only
20130702 18:49:20 <sgothel> oops .. they don't ? hmm .. jaja, ok
20130702 18:49:29 <sgothel> smaller .. better :)
20130702 18:50:01 <sgothel> around WorldWind right ? (timewise ..)
20130702 18:50:49 <sgothel> you will get at least 1-2 dinners from us then :)
20130702 18:52:57 <sgothel> above WorldWind then .. ok
20130702 19:04:55 <sgothel> @Harvey: Updated http://jogamp.org/doc/siggraph2013/t-shirt/
20130702 19:13:18 <olamedia> nice.. <javac target="1.6" listfiles="yes" fork="yes" executable="/usr/lib/jvm/java-6-openjdk-i386/bin/javac" includeantruntime="false" destdir="build/class" sourcepathref="dependency-sources" classpathref="classpath">
20130702 19:27:01 <olamedia> plus <target name="run" depends="compile-copy-jar">
20130702 19:27:02 <olamedia> <exec dir="/home/olamedia/Документы/__OlaCraft/workspace/ola-modloader/" command="java -jar /home/olamedia/Документы/__OlaCraft/workspace/ola-modloader/modloader.jar">
20130702 19:27:02 <olamedia> </exec>
20130702 19:27:02 <olamedia> </target>
20130702 19:27:23 <olamedia> so, i need to press only build button
20130702 19:27:32 <olamedia> nice
20130702 20:22:40 <olamedia> someone tried nifty-gui? i can't download manual from sourceforge :/
20130702 20:22:50 <olamedia> (no reply)
20130702 20:53:53 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20130702 20:59:23 * void256 (~chatzilla@anon) has joined #jogamp
20130702 21:04:48 * rmk0 reappears
20130702 21:05:30 <rmk0> run into a bizarre problem
20130702 21:05:45 <rmk0> the test suite hangs... i don't think it's anything i've done
20130702 21:06:28 <rmk0> i was trying to run the test suite to check my quirk patch, but i keep running into:
20130702 21:06:31 <rmk0> http://waste.io7m.com/2013/07/02/hang.txt
20130702 21:07:40 * rmk0-alt (~rmk0-alt@anon) has joined #jogamp
20130702 21:07:40 * rmk0-alt (~rmk0-alt@anon) Quit (Changing host)
20130702 21:07:40 * rmk0-alt (~rmk0-alt@anon) has joined #jogamp
20130702 21:08:33 <rmk0-alt> am running the suite again but writing a log this time...
20130702 21:30:59 <rmk0-alt> sgothel: will a context with a core profile be using a, for example, GL3bc implementation?
20130702 21:31:06 <rmk0-alt> i feel like the answer should be no...
20130702 21:42:15 <rmk0-alt> yeah, HEAD is definitely bad here
20130702 21:42:31 <rmk0-alt> test suite hangs... waiting for log to finish writing and i'll upload
20130702 21:51:52 <olamedia> nifty's resource loader is failing to load own resources :/
20130702 21:55:07 <rmk0> http://waste.io7m.com/2013/07/02/jogl-2013-07-02-214115.log.gz
20130702 22:23:45 <olamedia> :/ noway i need to write my own gui library
20130702 22:25:02 <olamedia> nifty-gui looks not usable
20130702 22:58:59 <void256> what?
20130702 23:00:50 <void256> olamedia: please post an issue here: https://github.com/void256/nifty-gui/issues?state=open if you think that is a Nifty issue
20130702 23:01:21 <void256> I've been able to load resources using Nifty
20130702 23:01:55 <olamedia> my nifty jar is at /lib folder
20130702 23:02:32 <olamedia> so i was forced to extract and nxs files directly to base folder
20130702 23:02:39 <olamedia> *and put
20130702 23:03:01 <olamedia> anyway, i just got a black screen
20130702 23:03:28 <olamedia> (without any warnings
20130702 23:03:29 <void256> when you did what?
20130702 23:03:36 <void256> run an example?
20130702 23:03:39 <void256> your own code?
20130702 23:03:45 <olamedia> own
20130702 23:04:07 <void256> did you read the pdf manual or tried it directly?
20130702 23:04:16 <olamedia> read
20130702 23:04:24 <void256> good
20130702 23:04:40 <void256> xml or builder?
20130702 23:04:46 <olamedia> builder
20130702 23:05:05 <olamedia> niftyRenderDevice = new JoglRenderDevice();
20130702 23:05:05 <olamedia> niftyInputSystem = new JoglInputSystem();
20130702 23:05:05 <olamedia> niftyTimeProvider = new AccurateTimeProvider();
20130702 23:05:05 <olamedia> nifty = new Nifty(niftyRenderDevice, new NullSoundDevice(), niftyInputSystem, niftyTimeProvider);
20130702 23:05:05 <olamedia> nifty.getResourceLoader().addResourceLocation(new FileSystemLocation(new java.io.File("res")));
20130702 23:05:05 <olamedia> niftyScreen = new ScreenBuilder("start") {
20130702 23:05:08 <olamedia> {
20130702 23:05:10 <olamedia> layer(new LayerBuilder("background") {
20130702 23:05:12 <olamedia> {
20130702 23:05:14 <olamedia> backgroundColor(new Color(255f / 255f, 0f / 255f, 0f / 255f, 0.5f));
20130702 23:05:15 <olamedia> alignCenter();
20130702 23:05:18 <olamedia> childLayoutCenter();
20130702 23:05:20 <olamedia> }
20130702 23:05:22 <olamedia> });
20130702 23:05:24 <olamedia> }
20130702 23:05:26 <olamedia> }.build(nifty);
20130702 23:06:02 <void256> and you're calling nifty.render() and nifty.update() somewhere as well of course?
20130702 23:06:07 <olamedia> nifty.update();
20130702 23:06:07 <olamedia> nifty.render(false);
20130702 23:06:11 <olamedia> or true
20130702 23:06:37 <olamedia> at display method of GLEventListener
20130702 23:07:03 <void256> nifty.gotoScreen("start") missing?
20130702 23:07:29 <void256> can't remember if this happens automatic when using the builder tho
20130702 23:10:21 <void256> but worth a try
20130702 23:10:52 <olamedia> nothing
20130702 23:12:36 <void256> ok, one second
20130702 23:13:04 <void256> when you call nifty.render()
20130702 23:13:49 <void256> you've set up everything for 2d rendering right?
20130702 23:18:00 * void256_ (~void@anon) has joined #jogamp
20130702 23:18:15 <void256_> http://pastebin.com/sYf00dn6
20130702 23:18:41 <void256_> http://pastebin.com/YCDT6wz2
20130702 23:18:56 <void256_> it's a bit more than you need (especially the runner) but this works for me
20130702 23:20:48 <void256_> if you still have problems consider to create an issue on github
20130702 23:20:52 <void256_> I have to go to bed :)
20130702 23:20:56 <void256_> good luck!
20130702 23:21:03 * void256 (~chatzilla@anon) Quit (Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803])
20130702 23:21:09 * void256_ is now known as void256
20130702 23:21:31 * void256 (~void@anon) Quit (Client Quit)
20130702 23:24:15 <olamedia> should fix manual, yes, gotoScreen was a problem
20130703 03:23:15 * [Mike] (~Mike]@anon) Quit ()
20130703 03:52:10 * [Mike] (~Mike]@anon) has joined #jogamp
20130703 05:05:23 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20130703050523.html