JogAmp IRC Chat Logs

#jogamp @ irc.freenode.net - 20130623 05:05:27 (UTC)

20130623 05:05:27 -CatOut- Previous @ http://jogamp.org/log/irc/jogamp_20130622050527.html
20130623 05:05:27 -CatOut- This channel is logged @ http://jogamp.org/log/irc/jogamp_20130623050527.html
20130623 05:38:11 * ___m___ is now known as [Mike]
20130623 07:49:52 * monsieur_max (~maxime@anon) has joined #jogamp
20130623 08:57:12 * [Mike] (~Mike]@anon) Quit (Read error: Connection reset by peer)
20130623 09:13:25 <sgothel> building again .. brrrr .. scary :)
20130623 09:13:55 <monsieur_max> :) sunday morning is the best time for this
20130623 09:14:09 <sgothel> yup .. probably
20130623 09:14:56 <monsieur_max> ready to ruin the whole day and the incoming week !
20130623 09:15:28 <sgothel> not that bad .. I have tested the java7 setup and signed build on linux, windows and android already :)
20130623 09:15:55 <monsieur_max> great :)
20130623 09:16:16 <sgothel> .. and took good sleep between the duty cycles .. very important to stay relaxed :)
20130623 09:17:16 <monsieur_max> i could not agree more
20130623 09:18:03 <sgothel> too bad java7 on osx doesn't support x32 anymore .. hmm
20130623 09:18:44 <sgothel> what are 32bit OSX machines anyway ? OSX 10.5.8 - 10.6.. ?
20130623 09:26:45 <sgothel> Message: FATAL: sorry, too many clients already RequestURI: /general-f784504.html Server: n3.nabble.com
20130623 09:26:49 <sgothel> oh oh ..
20130623 10:01:29 <sgothel> @Xerxes: I just realized :)
20130623 10:01:57 * xranby1 (~familjen@anon) has joined #jogamp
20130623 10:02:14 <xranby1> https://jogamp.org/bugzilla/show_bug.cgi?id=762 - Too many users on the JogAmp Nabble forum; forum needs replacement
20130623 10:02:16 <sgothel> could not even do a backup
20130623 10:02:21 <sgothel> jaja :)
20130623 10:02:37 <sgothel> I use email .. hence I didn't see it yesterday
20130623 10:02:47 <sgothel> Mark was mumbling something
20130623 10:03:15 <xranby1> ideally we want to mirror the nabble forum so that all the existing forum.jogamp.org/* url's still work after a forum switch
20130623 10:03:35 <sgothel> it's not open source AFAIK ..
20130623 10:03:50 <sgothel> yes - I would like to have the forum 'home' as well
20130623 10:04:05 <sgothel> if shit hits the fan .. after siggraph we setup good ole mailman :)
20130623 10:04:36 <xranby1> hmm.. well what i do like about the nabble thing is the ability to do web edits and re-edits
20130623 10:04:45 <sgothel> https://jogamp.org/ -> https://jogamp.org/wiki/index.php/JogAmp_BOF_SIGGRAPH_2013_PLANNING + https://jogamp.org/wiki/index.php/JogAmp_SIGGRAPH_2012_-_2013_Improvements
20130623 10:04:53 <sgothel> ( I don't see re-edits .. email)
20130623 10:05:06 <sgothel> so .. I polished pages .. w/ content .. and a story line
20130623 10:05:31 <xranby1> google see the re-edits
20130623 10:05:32 <xranby1> :)
20130623 10:05:43 <sgothel> haha, yes
20130623 10:05:49 <xranby1> thus if people search for jogamp stuff they find the forum posts
20130623 10:06:09 <sgothel> all great - so whats our alternative ? me all open
20130623 10:06:26 <sgothel> (get a nabble compatible software ..)
20130623 10:06:28 <xranby1> need to scout,, i have no idea
20130623 10:06:39 <sgothel> but .. back to important things :)
20130623 10:06:53 <sgothel> 761 done
20130623 10:06:57 <sgothel> building now ..
20130623 10:07:03 <xranby1> i can try talk to libgdx author for suggestions on good slides/demos
20130623 10:07:12 <sgothel> awesome
20130623 10:07:20 <sgothel> maybe they have a few
20130623 10:08:06 <sgothel> Whats w/ Julien .. hmm .. have to contact him: jme3 .. ardor3d .. (and the maintainer)
20130623 10:08:17 <sgothel> Wade is like missing in action
20130623 10:08:35 <sgothel> you are ok w/ showing WorldWind ?
20130623 10:09:25 <xranby1> i do not have a good machine to demo world wind on
20130623 10:09:32 <sgothel> we have ..
20130623 10:09:38 <xranby1> world wind works best on traditional desktop opengl hardware
20130623 10:09:43 <sgothel> ok
20130623 10:09:52 <xranby1> we may demo their 2.0 builds
20130623 10:09:58 <sgothel> duh .. ofc :)
20130623 10:10:10 <sgothel> i.e. using the new jogl ..
20130623 10:10:19 <xranby1> yup... that is a milestone
20130623 10:10:34 <sgothel> also updated jogl_joal as napplet - works good
20130623 10:10:48 <xranby1> thank you
20130623 10:10:52 <xranby1> awesome
20130623 10:11:05 <sgothel> I tested the demo applet from jausoft loading jogamp jars (w/ the new security stuff) - works good
20130623 10:11:28 <xranby1> did you test icedtea-web as well?
20130623 10:11:35 <sgothel> so today .. I will make it RC12 due to security and remove all the old signed builds
20130623 10:11:45 <sgothel> yes - same problems w/ mixed code
20130623 10:12:04 <xranby1> ok, not sure how to convince the redhat team hot to make it work out of the box
20130623 10:12:12 <sgothel> also tested w/ java6 .. still works :)
20130623 10:12:27 <sgothel> well, we have to tell folks - to sign it all then ..
20130623 10:12:37 <sgothel> if desired to work w/ icedtea
20130623 10:12:40 <xranby1> since they see mixed mode as a security risk and may not want to support it
20130623 10:12:53 <sgothel> yeah well :)
20130623 10:13:09 <sgothel> when I read the html api-doc insecurity .. I went nuts :)
20130623 10:13:33 <sgothel> hence the whole overhaul of build system .. no more java6 anyways
20130623 10:13:48 <sgothel> good that we had the source/target 1.6 stuff in place
20130623 10:14:01 <sgothel> (produce java6 bytecode w/ java7)
20130623 10:14:12 <xranby1> actually there is some people that have sucessfully built icedtea 6 / openjdk 6 for windows
20130623 10:14:31 <sgothel> 6 -> 7: then GREAT
20130623 10:14:40 <sgothel> yes, would be a leap fwd
20130623 10:15:14 <xranby1> and on x86 the opensource hotspot do work as intended
20130623 10:15:20 <xranby1> fast etc
20130623 10:15:30 <sgothel> so you like idea: start w/ an hour (2/3) of demos / showcases how to use jogamp ? then jogamp itself - Q&A - done
20130623 10:15:51 <sgothel> it's the same hotspot - I know
20130623 10:16:08 <sgothel> (well, w/o Oracles other beast :)
20130623 10:17:16 <sgothel> looks like a few minor regressions on build 1012 .. not too bad
20130623 10:20:04 <sgothel> debug ctx - yup, is related to changes
20130623 10:29:59 <rmk0> lo
20130623 10:30:00 <xranby1> Great, i will demo this libgdx game.. http://www.ludumdare.com/compo/ludum-dare-26/?action=preview&uid=15341
20130623 10:30:10 <xranby1> porting is trivial
20130623 10:30:30 <xranby1> got the suggestion from Mario/libgdx
20130623 10:31:05 <sgothel> hmm .. still time to finish ffmpeg/audio ? .. IMHO more important that porting another game :)
20130623 10:31:16 <sgothel> arent' there many avail already ?
20130623 10:31:35 <sgothel> I mean libGDX + JOGL out of the box ?
20130623 10:32:04 <sgothel> made with libgdx, arte.. ah
20130623 10:32:07 <xranby1> the libgdx demo repository only contains a handfull examples
20130623 10:32:29 <xranby1> thus i need to scout outside the main libgdx repository for good demos
20130623 10:33:06 <rmk0> sgothel: i was actually eyeing nabble because i was looking for something for io7m.com... don't want to run a forum/pile of mailing lists myself
20130623 10:33:08 <xranby1> sure ffmpeg/audio is more important
20130623 10:33:10 <sgothel> i.e. could we ask the authors to have such a demo on desktop + mobile / android .. hmm, but ofc GREAT if you can do that
20130623 10:33:16 <rmk0> had no idea that the jogamp forum had crashed
20130623 10:33:34 <sgothel> well .. let's say DOS
20130623 10:33:37 <xranby1> rmk0: the forum got too many users
20130623 10:33:42 <sgothel> haha
20130623 10:33:44 <xranby1> we are too popular :)
20130623 10:33:46 <rmk0> yep!
20130623 10:34:46 <rmk0> it's nice that it provides a mailing list interface and a forum combined
20130623 10:34:55 <rmk0> not aware of anything else that does that
20130623 10:35:02 <sgothel> yes .. like to have that on jogamp ..
20130623 10:35:14 <sgothel> at least .. then its our fault :)
20130623 10:35:37 <sgothel> note: you all can launch such a beast, since you are jogamp_web group members
20130623 10:36:07 <sgothel> will fiddle the mysql db .. then
20130623 10:37:26 <sgothel> oh dear .. now the tests start to fail .. windows and solaris
20130623 10:37:59 <sgothel> UITestCase.initSingletonInstance .. port locking fails .. hmm
20130623 10:39:19 <sgothel> SWT issues w/ Java7 ?! hmm
20130623 10:45:34 <rmk0> friend of mine who i trust to be extremely picky when it comes to software is using fluxbb
20130623 10:45:37 <rmk0> going to take a look at it
20130623 10:48:12 <xranby1> Days of wonders Android JOGL Ticket to Ride have got a really good review: http://androidspin.com/2013/06/22/game-review-ticket-to-ride-a-ride-worth-taking/
20130623 10:49:13 <sgothel> lunch .. laters
20130623 10:50:04 <monsieur_max> xranby1: nice, thanks for the link
20130623 10:52:30 * sgothel (~sven@anon) Quit (Quit: Leaving.)
20130623 10:52:50 * sgothel (~sven@anon) has joined #jogamp
20130623 10:52:50 * ChanServ sets mode +v sgothel
20130623 11:04:50 * olamedia (~olamedia@anon) has joined #jogamp
20130623 11:04:53 <xranby1> olamedia: welcome back, we would like to have you running a demo of your minecraft game at the JogAmp Siggraph 2013 bof! will you be able to attend?
20130623 11:05:35 <xranby1> olamedia: http://forum.jogamp.org/Roadmap-2-0-2-SIGGRAPH-2013-Update-td4029433.html
20130623 11:08:30 <olamedia> i'll be able to send latest code to run, but not to attend
20130623 11:12:51 <xranby1> olamedia: ok that would be most a most welcome addition
20130623 11:55:54 <xranby1> monsieur_max: This link may be of interest as well: http://penny-arcade.com/report/article/ticket-to-ride-is-here-for-android-why-days-of-wonder-changed-their-apple-o - "Ticket to Ride is here for Android! Why Days of Wonder changed their “Apple only” stance"
20130623 11:56:35 <monsieur_max> great thanks
20130623 11:57:11 <xranby1> be sure to read Eric @ DoW comment in the comment section
20130623 11:57:20 <monsieur_max> if things goes weel, maybe y game could be part of the siggraph 2014 showcase :)
20130623 11:57:35 <xranby1> they basically use the same codebase for both the PC steam version and the Android version
20130623 11:57:46 <xranby1> monsieur_max: that would be really nice
20130623 12:35:40 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20130623 12:46:27 <olamedia> looks like it's working, will continue later https://github.com/olamedia/modloader
20130623 12:57:28 <sgothel> we may get a few slides from Erik/DoW
20130623 12:57:58 <xranby1> sgothel: great
20130623 12:58:28 <sgothel> however I can talk about a few detail .. i.e. the Android hardening .. preserved GL ctx .. etc
20130623 12:58:38 <xranby1> sgothel: during jogl builds on my machine i see silent gluegen errors, not stopping the build, in the logs http://paste.ubuntu.com/5792429/ should i investigate further?
20130623 12:59:14 <sgothel> thats an old one, no regression - well, yeah - should get fixed eventually
20130623 12:59:25 <sgothel> not a prio now, need to fix regressions now
20130623 13:01:39 <xranby1> sgothel: ok i will continue smoketest
20130623 13:02:11 <xranby1> reporting all i can find big and little
20130623 13:02:25 <xranby1> ... and nothing to report yet.. that is good
20130623 13:03:58 <sgothel> thank you Xerxes - tests are appreciated .. later we need to test the jenkins builds when reliable (linux/android arm*)
20130623 13:05:21 <xranby1> some visual artifacts seen running graph using the fglrx drivers
20130623 13:06:02 <sgothel> usually b/c test is too fast :) - and msaa1 looks really bad there
20130623 13:08:36 <xranby1> have you seen ramis latest blogposts reguarding precision qualifiers? what is interesting is that people have invented ways to find out the floating point precision used by the drivers
20130623 13:09:45 <sgothel> yup .. well, spec says (and always emphasized .. even on latest siggraph) .. it's not reliable at all. so you would need to make calculations and read it back (impossible w/ ES2)
20130623 13:10:01 <sgothel> ok .. the extention macro def is there .. well
20130623 13:10:23 <sgothel> medium is usually workin
20130623 13:10:46 <sgothel> defaults: highp vertex, mediump fragment
20130623 13:11:20 <sgothel> and on desktop, its ignored anyways
20130623 13:11:46 <xranby1> sgothel: rami discovered the hard way that the mediump was not sufficient for his usecase when using the Mali-T604 drivers :/
20130623 13:12:09 <xranby1> still within spec
20130623 13:13:31 <sgothel> so they are 1st to respect that value, and mali 400 used higher mediump 'defaults' .. welcome to quirk wonderland :)
20130623 13:14:08 <sgothel> hope ES3 has the precision query .. so all this is obsolete
20130623 13:14:53 <xranby1> http://youilabs.com/blog/mobile-gpu-floating-point-accuracy-variances/ neat benchmark to test the floating point precision on any shader GPU
20130623 13:27:44 <xranby1> sgothel: http://paste.ubuntu.com/5792515/
20130623 13:28:11 <xranby1> assertion triggered
20130623 13:43:07 <sgothel> yup .. some test cases stressing some features do fail on fglrx :( .. usually ctx/drawable switch .. etc - nothing really we can do here - this one, I am not so sure about, but I have seen it before.
20130623 13:43:38 <sgothel> only hope is - Mesa9* AMD becomes more usable :)
20130623 13:44:41 <sgothel> so I was quite happy about the report of Bug 759
20130623 13:45:12 <sgothel> which fglrx driver / GPU combination ?
20130623 13:45:38 <sgothel> so have you tested Bug 761 ?
20130623 13:46:03 <sgothel> i.e. w/ your test case - mine were OK w/ latest fglrx driver and a HD5xxx
20130623 13:48:47 <xranby1> am still running the junit.run tests using the fixes from 761.. trying to stress the drivers
20130623 13:50:54 <sgothel> https://jogamp.org/chuck/job/jogl/1012/label=linux-x86_64-amd/testReport/
20130623 13:51:08 <sgothel> https://jogamp.org/chuck/job/jogl/1011/label=linux-x86_64-amd/testReport/
20130623 13:51:19 <sgothel> .. etc .. you see some fluctuations / threading issues
20130623 13:51:40 <sgothel> usually < 5 errors
20130623 14:01:21 <xranby1> ok 33min later ... all tests -1 passed thus all in all it looks stable
20130623 14:01:31 <xranby1> or more stable than before
20130623 14:02:15 <sgothel> sweet
20130623 14:03:43 * monsieur_max (~maxime@anon) has joined #jogamp
20130623 14:08:38 <xranby1> OpenGL vendor string: Advanced Micro Devices, Inc.
20130623 14:08:38 <xranby1> OpenGL renderer string: ATI Radeon HD 4800 Series
20130623 14:08:38 <xranby1> OpenGL version string: 3.3.11627 Compatibility Profile Context
20130623 14:08:57 <xranby1> sgothel: is ther a way to get a better version string for this driver?
20130623 14:11:20 <sgothel> ofc ..
20130623 14:11:50 <sgothel> lately we provide the vendor version .. if you run TestGearsES2.*NEWT ..
20130623 14:12:44 <sgothel> disclaimer: depends on driver ofc
20130623 14:13:36 <sgothel> getGLVendorVersionNumber()
20130623 14:13:47 <sgothel> GLContext.getGLVendorVersionNumber()
20130623 14:14:20 <sgothel> new drivers usually have it at the end .. of GL_VERSION, and we parse that one - was required for bug ..
20130623 14:14:53 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=706#c46
20130623 14:15:04 <sgothel> GL Version 4.0 (hardware) - 4.0.9910 Compatibility Profile Context [GL 4.0.0, vendor 0.0.0 (n/a)
20130623 14:15:15 <sgothel> GL Version 4.2 (hardware) - 4.2.12198 Compatibility Profile Context 12.102.3.0 [GL 4.2.0, vendor 12.102.3 (12.102.3.0)]
20130623 14:17:06 <xranby1> on this machine SharedContext: X11GLXContext [Version 3.3 (hardware) - 3.3.11627 Compatibility Profile Context [GL 3.3.0, vendor 0.0.0 (n/a)]
20130623 14:17:42 <xranby1> thus not supported for this driver i guess
20130623 14:19:12 <sgothel> yup
20130623 14:19:26 <sgothel> i.e. driver simply doesn't add it's version to GL_VERSION
20130623 14:19:30 <xranby1> com.jogamp.opengl.test.junit.jogl.acore.TestBug669RecursiveGLContext02NEWT can reproduce the issue i face
20130623 14:19:47 <sgothel> 'good'
20130623 14:21:10 <xranby1> hmm . the test can also pass sometimes
20130623 14:21:59 <sgothel> threading issue .. fglrx is quite sensitive .. and I could not find a leak in our locking [yet]
20130623 14:22:17 <sgothel> actually .. we don't really lock anymore .. i.e. non blocking impl.
20130623 14:22:53 <sgothel> so I assume fglrx is simply using aliases of the display connection - and that doesn't work properly (-> their XCB handler, used in glX functions)
20130623 14:23:05 <sgothel> in short: ignore it for now
20130623 14:23:55 <sgothel> fglrx: good shader performance / buggy X11 part; Mesa/AMD: medium shader perf. / great X11 part :)
20130623 14:24:32 <xranby1> :)
20130623 14:24:58 <xranby1> i hope the Mesa/AMD will win over time
20130623 14:25:38 <sgothel> guess what is Lockheed Martin and L3 Communication using ? The open source variation ..
20130623 14:26:11 <sgothel> so there was this one great line in Sun's ole [open] source stuff .. no nuclear ..
20130623 14:26:31 <xranby1> can it be removed
20130623 14:26:32 <xranby1> ?
20130623 14:26:46 <xranby1> is it a suggestion or a legally binding line
20130623 14:26:46 <sgothel> hu ?
20130623 14:27:00 <sgothel> it was binding if using a pre-compiled build
20130623 14:27:12 <sgothel> you like to build nuclear weapons now ?
20130623 14:27:28 <xranby1> i want the 4 freedoms of software
20130623 14:27:31 <sgothel> seriously .. Siemens asked about it :)
20130623 14:27:40 <sgothel> the 4 freedoms of death
20130623 14:27:44 <xranby1> jaja
20130623 14:28:01 <sgothel> in short: nope, not binding
20130623 14:28:09 <xranby1> ok then i suggest we remove it
20130623 14:28:16 <sgothel> we still have it ?
20130623 14:28:18 <xranby1> yes
20130623 14:28:45 <sgothel> fascinating .. I was thinking .. we should allow nuclear fusion power plants to use JOGL
20130623 14:30:02 <sgothel> * You acknowledge that this software is not designed or intended for use
20130623 14:30:02 <sgothel> * in the design, construction, operation or maintenance of any nuclear
20130623 14:30:02 <sgothel> * facility.
20130623 14:30:30 <sgothel> This is simply a disclaimer .. like those 'don't sue us' ..
20130623 14:30:48 <sgothel> and it's true - I have not design the new parts either for such usage :)
20130623 14:30:55 <sgothel> but can't help it if they do
20130623 14:31:36 <xranby1> have there been a case where free software have been sued for not having such an disclaimer in place?
20130623 14:32:18 <sgothel> don't know - but while rendering services .. I emphasize that the BSD license is valid, and I cannot go beyond that.
20130623 14:32:27 <sgothel> the license does say .. no reliability
20130623 14:32:54 <sgothel> can't remember the legal term .. indemification - yes
20130623 14:33:15 <sgothel> indemnification
20130623 14:33:38 <sgothel> they love it - i.e. I save the customers ass when anything happens - sure, can't do :)
20130623 14:34:08 <xranby1> i think "This software is provided "AS IS," without a warranty of any kind." should be enough
20130623 14:34:40 <sgothel> what we can say, is in the realms of our state of the art validation / unit tests .. you may feel safe and know what you are getting . but no indemnification.
20130623 14:35:33 <sgothel> hence the license audit .. etc, they have to make their own risk assesment
20130623 14:36:18 <sgothel> that (again) is the reason we have to put a 'name tag' to the commits, so in case lets say Safena claims a copyright .. we have to remove your commits
20130623 14:36:38 <xranby1> sure
20130623 14:36:46 <sgothel> Ola .. and I had a few discussions in this regard :)
20130623 14:37:26 <sgothel> IMHO not required to check passports :) - but a name tag to identify the source of a commit, thats all
20130623 14:38:11 <sgothel> that was one reason why LWJGL was not favored in one project, 'source unknown', risk too high
20130623 14:42:11 <xranby1> sgothel: good work with the summary of the last years improvements: http://jogamp.org/wiki/index.php/JogAmp_SIGGRAPH_2012_-_2013_Improvements its an impressive list
20130623 14:42:33 <xranby1> sgothel: is ES3 all done?
20130623 14:43:32 <sgothel> not yet :)
20130623 14:44:11 <xranby1> ok then i should quit distracting you on the progress towards that one..
20130623 14:44:28 <sgothel> hehe
20130623 14:56:55 <olamedia> can someone try if the modloader is actually working on other pc?
20130623 14:58:20 <olamedia> http://rghost.ru/users/ola3/releases/gc - loader.jar - is a main app, graphicsCore.jar to ./coremods, jogl & gluegen with natives into ./lib
20130623 14:59:02 <olamedia> require one of latest jogl builds
20130623 15:00:05 <xranby1> olamedia: what is the purpose of modloader?
20130623 15:00:53 <olamedia> my code got very complicated to continue.. want to split into different jars and make an api
20130623 15:01:36 <xranby1> i may test https://github.com/olamedia/modloader if i can compile it
20130623 15:01:57 <olamedia> u can, but there's no mods :/
20130623 15:02:11 <xranby1> where is the sourcecode to the mods?
20130623 15:02:54 <sgothel> yadt ? yet-another-deployment-tool ? OMG
20130623 15:03:32 <olamedia> https://github.com/olamedia/graphics-core
20130623 15:04:00 <olamedia> just a test frame filled with red color )
20130623 15:05:58 <olamedia> the last will require modloader in project path to compile
20130623 15:08:41 <olamedia> ...and jogl..
20130623 15:13:36 <sgothel> @Xerxes: Could you use the JOGL/JOAL dependency ? Branch merge in progress - Sorry for being an ass here, just trying to focus :)
20130623 15:28:38 <olamedia> btw was issue with keycodes fixed in latest jar autobuild (gluegen_686-joal_448-jogl_1011-jocl_810)?
20130623 15:28:44 <xranby1> sgothel: the jogl/joal dependency work as intended yes
20130623 15:29:28 <sgothel> thanks Xerxes .. great
20130623 15:31:08 <xranby1> olamedia: yes if you refer to https://jogamp.org/bugzilla/show_bug.cgi?id=641
20130623 15:33:38 <xranby1> olamedia: as long as you use NEWT it should work
20130623 15:43:37 <olamedia> what about modloader? there's probably GL2 code left (copy-pasted) in mod
20130623 15:44:17 <olamedia> :/ forgot that people here often use ES2 :/
20130623 15:44:43 <olamedia> 1-2 lines to change, i think
20130623 15:45:39 <sgothel> @Ola: Have you considered writing a unit test for features once added on your request. Harvey may do it .. but IMHO it's really not good spirit disregarding these things.
20130623 15:47:10 <olamedia> @sgothel about keys? well.. at least i will modify my code and test on my machine.. don't know how to test key presses automatically
20130623 15:47:50 <sgothel> nope - keys are fine - once you let me beg you to complete a frustum unit test
20130623 15:48:27 <olamedia> oh.. that thing.. is so complicated >_<
20130623 15:48:58 <sgothel> lots of stuff is complicated .. and ofc this is BS since you use it in your code :)
20130623 15:49:12 <sgothel> its the attitude ..
20130623 15:49:30 <sgothel> which makes me grumpy sometimes .. but well
20130623 15:53:34 <rmk0> software without tests is broken
20130623 15:54:16 <sgothel> yes .. if we are lucky you (Mark) or Harvey will add a unit test (thanks to your reliability) .. otherwise it will get removed
20130623 15:55:10 <rmk0> i've yet to implement a frustum, so i don't know what's correct and what isn't yet
20130623 15:55:15 <rmk0> you'll definitely get one when i do
20130623 15:55:48 <rmk0> if harvey doesn't get there first
20130623 15:56:29 <sgothel> sweet - the implementation I did includes link to documentation/paper - however, I thought Ola would like to practice 'give and take' philosophy.
20130623 15:57:08 <sgothel> .. and as I my ole Basketball couch told me, 'can't do means don't want to do' :)
20130623 15:57:55 <sgothel> Thx to Petr and Harvey most of our math stuff got validated .. so there is still hope
20130623 15:58:41 <sgothel> Java7 AWT focus traversal .. 'works different now' .. back to the basics :)
20130623 15:59:38 <olamedia> may be it's better for me to wait others >_> i spent more than one month experimenting with matrices already
20130623 16:01:07 <olamedia> wish i properly learned matrices when i have studied them
20130623 16:02:26 <xranby1> olamedia: one way to construct a test is to construct a test that uses the same input and checks for the expected output you currently use in your application
20130623 16:03:17 <xranby1> olamedia: at least the test will report that your application potentially break if someone introduce an error into the frustum implementation
20130623 16:03:41 <xranby1> or correct the frustum implementatino
20130623 16:04:26 <xranby1> in case it gets corrected then you may look at how the test got updated in order to know how to correct your applicatino
20130623 16:04:47 <sgothel> TestPMVMatrix01NEWT <- Boilerplate
20130623 16:10:41 <xranby1> https://gist.github.com/olamedia/4067336 google found this, thanks olamedia
20130623 16:11:19 <sgothel> lol
20130623 16:21:35 <olamedia> lol
20130623 16:22:00 <olamedia> olamedia created this gist 7 months ago.
20130623 16:22:22 <olamedia> not sure if it works somehow
20130623 16:24:59 <olamedia> it tries to check if given FAR,LEFT, BOTTOM planes have a 0 distance to "flb" (FarLeftBottom) point
20130623 16:25:16 * xranby1 (~familjen@anon) Quit (Quit: Leaving.)
20130623 16:27:00 <olamedia> point is calculated using vectors (moving to far plane, then 1/2 far plane width to left & 1/2 far plane height to bottom)
20130623 16:27:40 <sgothel> why don't you document this in your code .. intention, and make a proper junit test out of it .. and sent a git pull request for us to review?
20130623 16:28:18 <sgothel> all of this here is wasted time otherwise - that is the point I try to teach you here for teamwork
20130623 16:28:33 <olamedia> >_> as i can remember, i made this code as sample for u to convert into real test
20130623 16:29:25 <sgothel> yes . and isn't it a bit outrageous actually ? i.e. after I properly (hope so ? :) impl. the frustum following the paper you provided ?
20130623 16:30:00 <sgothel> point taken - I realized that I had no time yet to deal w/ it
20130623 16:30:35 <sgothel> another point taken - 'shooting around', like pls test my stuff - then made me a bit unbalanced, due to the IMHO unbalanced 'give and take' sheet :)
20130623 16:31:03 <sgothel> so my apologies for my mood - but still, I think it would be great if you could do the 'teamwork' thingy :)
20130623 16:32:57 <olamedia> well, can't i ask someone from team to help me make correct unit test? i'm newbee in java & unittests are so far away for me
20130623 16:33:36 <sgothel> (06:04:47 PM) sgothel: TestPMVMatrix01NEWT <- Boilerplate <- java file in jogl repo
20130623 16:33:54 <sgothel> just copy paste .. easy to grasp - come on .. your are creating WORLDS :)
20130623 16:34:26 <olamedia> XD if it's just copy-paste why noone did this in 7 months?
20130623 16:43:51 <olamedia> that's not a boilerplate :/ there's some actual code
20130623 16:44:43 <olamedia> updated gist https://gist.github.com/olamedia/4067336 (to just copy-paste) but not sure if it's working
20130623 16:45:11 <olamedia> i don't have an environment to test
20130623 16:45:29 <olamedia> atm
20130623 16:46:36 <olamedia> and.. asserts should be replaced by epsilon version
20130623 16:51:01 <sgothel> will wait for your git pull request .. it's ok if it takes a little longer now - thank you
20130623 16:51:34 <sgothel> (or somebody else transforms it ..)
20130623 17:17:25 <rmk0> well... i'm impressed with fluxbb so far
20130623 17:17:36 <rmk0> install was extremely quick, the forum looks clean
20130623 17:17:41 <rmk0> http://forum.io7m.com
20130623 17:21:00 <rmk0> seems to have a decent security history, for a php program
20130623 17:30:50 <sgothel> Q: how to import nabble db ?
20130623 17:31:02 <olamedia> >_< most popular php programs have a very very long security history XD
20130623 17:31:04 <sgothel> thx for doing this
20130623 17:31:05 <rmk0> most likely impossible
20130623 17:31:38 <sgothel> and it's mailman compatible ..
20130623 17:32:22 <rmk0> will harass the fluxbb people
20130623 17:32:55 <olamedia> nabble db?
20130623 17:33:10 <sgothel> well .. we can download 'our' data
20130623 17:33:19 <sgothel> some sql tables
20130623 17:35:37 <rmk0> surprised they allow that
20130623 17:35:43 <rmk0> would expect lock-in
20130623 17:57:54 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20130623 18:08:59 <sgothel> downloading backup ..
20130623 18:09:23 <sgothel> Postgres SQL
20130623 18:10:10 <sgothel> hmm .. only 70 MB
20130623 18:16:18 <rmk0> aren't nabble user accounts shared across all ... things hosted by nabble?
20130623 18:16:44 <sgothel> dunno, but I would guess so
20130623 18:17:09 <rmk0> vaguely curious as to how they're represented in the sql dumps
20130623 18:17:23 <rmk0> as in... is there really enough information there to construct an equivalent user account in some other system
20130623 18:18:08 <rmk0> suspect fluxbb isn't a viable replacement for nabble
20130623 18:18:18 <rmk0> migration is going to mean writing a program, and there's no mail interface
20130623 18:18:20 <sgothel> site_762907_1040770263.zip @ jogamp_web home ..
20130623 18:18:26 * rmk0 eyes it
20130623 18:20:17 <sgothel> having a nap now .. good 'night' and good day
20130623 18:20:31 <rmk0> byeee!
20130623 18:56:23 * monsieur_max (~maxime@anon) has joined #jogamp
20130623 19:09:19 * [Mike] (~Mike]@anon) has joined #jogamp
20130623 21:23:55 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20130623 21:39:53 * rmk0 (~rmk0@anon) Quit (Ping timeout: 240 seconds)
20130623 21:41:20 * rmk0 (~rmk0@anon) has joined #jogamp
20130623 21:41:20 * rmk0 (~rmk0@anon) Quit (Changing host)
20130623 21:41:20 * rmk0 (~rmk0@anon) has joined #jogamp
20130623 23:08:06 * rmk0 (~rmk0@anon) Quit (Quit: Lost terminal)
20130623 23:08:41 * rmk0 (~rmk0@anon) has joined #jogamp
20130623 23:08:41 * rmk0 (~rmk0@anon) Quit (Changing host)
20130623 23:08:41 * rmk0 (~rmk0@anon) has joined #jogamp
20130624 02:04:47 * juank_prada (~juank_pra@anon) Quit (Quit: Leaving)
20130624 05:05:27 -CatOut- Continue @ http://jogamp.org/log/irc/jogamp_20130624050527.html