#jogamp @ irc.freenode.net - 20150814 05:05:29 (UTC)

20150814 05:05:29 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20150813050529.html
20150814 05:05:29 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20150814050529.html
20150814 05:21:44 * a11md11 (c0e2cd65@anon) has joined #jogamp
20150814 05:24:22 * a11md11 (c0e2cd65@anon) Quit (Client Quit)
20150814 05:43:21 <xranby_> badshah400: welcome
20150814 05:43:58 <xranby_> Marcus Lagergren at JVMLS 2015 - Nashorn for Java 9 https://youtu.be/aROpSjXr4TU the only talk that did not need a safe harbor statement #freedom
20150814 05:44:11 <badshah400> Hello xranby_! How do you do?
20150814 05:44:58 <xranby_> badshah400: looking better, the lists of tasks to solve before teh 2.3.2 release shrink
20150814 05:45:18 <xranby_> http://jogamp.org/wiki/index.php/SW_Tracking_Report_Objectives_for_the_release_2.3.2
20150814 05:46:16 <badshah400> xranby_: cool! Is there a planned release date for 2.3.2?
20150814 05:47:06 <xranby_> badshah400: no planned date, we release when we have bugs fixed and implemented enough enhancements, we do have a test build a "release candidate"
20150814 05:47:28 <xranby_> http://forum.jogamp.org/RC-Test-Build-2-3-2-rc-20150812-td4035064.html
20150814 05:47:47 <xranby_> thus you can test all changes listed using this build
20150814 05:48:36 <xranby_> like there is an single jar that you may use that include all jogamp modules that work on all the platforms we support
20150814 05:49:00 <xranby_> badshah400: we hope to have the release done within 2 weeks
20150814 05:49:17 <badshah400> xranby_: i will try to do my usual testing procedure this weekend -- build rpm for openSUSE, install rpm, test new features
20150814 05:50:02 <badshah400> xranby_: thanks for all the info, really helpful
20150814 05:50:06 <xranby_> we need one week to add the new opengl es 3.2 , specifications landed only a day ago
20150814 05:50:37 <xranby_> but since we have the gluegen tool that generates the binding using the khronos opengl es headers
20150814 05:50:46 <xranby_> this addition should go fast
20150814 05:51:27 <xranby_> sgothel simply need to go through the list of changes in the specification and see that we map the specification to the right places
20150814 05:52:37 <badshah400> xranby_: great! looking fwd to test the new version.
20150814 05:53:49 <xranby_> badshah400: which version of jogamp is currently packaged for opensuse?
20150814 05:56:02 <badshah400> xranby_: I have e.g.: jogl-v2.3.1, gluegen-v2.3.1 and so on
20150814 05:56:12 <xranby_> great
20150814 05:56:13 <badshah400> so pretty much up to date
20150814 05:56:21 <xranby_> yes its the current release
20150814 05:56:49 <badshah400> xranby_: I try to keep up with the latest upstream for most packages I maintain
20150814 05:57:15 <xranby_> thank you
20150814 05:59:13 <badshah400> xranby_: for that I am very grateful to you guys because I have received a lot of help in this channel when I ran into any trouble
20150814 06:00:02 <badshah400> usually no trouble with jogmap per se, but e.g. when some pkgs break due to bkward incompatible changes and so on
20150814 06:03:24 <xranby_> badshah400: sgothel have gone great lenghts to document all changes in the public, for example if there is an api change it is also reflected in the jogamp junit tests. our bugzilla bug logs include a list of all the git changes that the bug touches so that you can find all commits
20150814 06:03:43 <xranby_> the jogamp bugzilla gus is our work logs
20150814 06:03:46 <xranby_> bugs
20150814 06:04:05 <xranby_> thus every api change have a bug to track it
20150814 06:05:07 <xranby_> i have seen other projects where they simply say som things are fixed without having a list of changes that you can study
20150814 06:05:55 <badshah400> xranby_: yes, and I have found that very helpful in the past; that's how I fix issues for pkgs that run into problems with a jogamp update
20150814 06:06:03 <xranby_> great
20150814 06:06:23 <xranby_> its good to know that the effort put into documenting it this way work for you
20150814 06:06:32 <badshah400> I think the documentation of changes is brilliant. So thanks a lot to you guys.
20150814 06:10:39 * monsieur_max (~maxime@anon) has joined #jogamp
20150814 06:17:59 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150814 06:45:16 * elect_ (~elect@anon) has joined #jogamp
20150814 06:46:26 <elect_> hi
20150814 06:59:13 <elect_> http://www.saschawillems.de/?p=1886
20150814 07:03:54 <xranby_> elect_: nice
20150814 07:04:35 <elect_> as I thought, the big advantage of vulkan over gl is mainly the true multi-thread support
20150814 07:05:11 <elect_> if you remain mainly single-thread, you wont experience such a big difference between modern GL and Vulkan
20150814 07:06:09 <xranby_> but since single threaded CPU's cant go faster
20150814 07:06:22 <xranby_> you have to do multithreaded CPU work
20150814 07:06:28 <xranby_> to speed up your applications
20150814 07:06:43 <elect_> if you want to squeeze up to the last drop, yes
20150814 07:07:06 <elect_> but this comes at the price of memory management..
20150814 07:07:19 <elect_> and learning a completely new api
20150814 07:07:35 <elect_> I guess it doesnt make sense for me
20150814 07:07:42 <elect_> as a small devb
20150814 07:07:45 <xranby_> if screen sizes continue to increase on devices then you may need to be required to do it the right way
20150814 07:08:01 <xranby_> lets see
20150814 07:13:25 * monsieur_max (~maxime@anon) has joined #jogamp
20150814 07:39:59 * eclesia (~husky@anon) has joined #jogamp
20150814 07:40:04 <eclesia> good morning
20150814 07:44:22 <xranby_> eclesia: hi!
20150814 07:45:08 <eclesia> xranby_: hi
20150814 07:51:16 <xranby_> eclesia: have you had chance to test the new jogamp release candidate ?
20150814 07:51:42 <xranby_> observed anything odd etc
20150814 07:52:25 <eclesia> nope sorry, was stuck on something else
20150814 07:53:55 <eclesia> xranby_: same question : and your trip in unlicense-lib, found odd stuff ?
20150814 07:54:19 <xranby_> eclesia: yes, always something preventing my mvn build to compile it
20150814 07:54:31 <xranby_> one second i will give you the build error
20150814 07:55:05 <eclesia> I should install a jenkins somewhere ...
20150814 07:56:22 <xranby_> [INFO] API - geometries .................................. FAILURE [1.263s]
20150814 07:56:56 <xranby_> Results :
20150814 07:56:56 <xranby_> Tests in error:
20150814 07:56:56 <xranby_> triangleTest(un.impl.geometry.triangulate.EarClippingTest): un.api.geometry.DefaultTupleBuffer1D.<init>(III)V
20150814 07:56:56 <xranby_> quadTest(un.impl.geometry.triangulate.EarClippingTest): un.api.geometry.DefaultTupleBuffer1D.<init>(III)V
20150814 07:56:56 <xranby_> vTest(un.impl.geometry.triangulate.EarClippingTest): un.api.geometry.DefaultTupleBuffer1D.<init>(III)V
20150814 07:56:56 <xranby_> Tests run: 951, Failures: 0, Errors: 3, Skipped: 4
20150814 07:57:13 <xranby_> so there was not a build error per see, but the test errors prevent the build to continue
20150814 07:57:32 * badshah400 (~badshah40@anon) Quit (Read error: Connection reset by peer)
20150814 07:57:44 <eclesia> strange. I'll check this evening
20150814 07:59:41 <xranby_> java.lang.NoSuchMethodError: un.api.geometry.DefaultTupleBuffer1D.<init>(III)V
20150814 07:59:56 <xranby_> at un.impl.geometry.Geometries.toTupleBuffer(Geometries.java:409)
20150814 08:01:24 <xranby_> unlicense$ mvn --version
20150814 08:01:24 <xranby_> Apache Maven 3.0.5
20150814 08:01:24 <xranby_> Maven home: /usr/share/maven
20150814 08:01:24 <xranby_> Java version: 1.8.0_45, vendor: Azul Systems, Inc.
20150814 08:01:24 <xranby_> Java home: /usr/lib/jvm/zulu-8-amd64/jre
20150814 08:01:40 <xranby_> i use mvn install to build it
20150814 08:02:00 <eclesia> are you up to date ?
20150814 08:02:49 <xranby_> yes i am at 2401:0000b023aed5
20150814 08:03:23 <monsieur_max> xranby : the rc is on maven ?
20150814 08:03:44 <xranby_> monsieur_max: rc's are only put into the jogamp test maven
20150814 08:03:50 <xranby_> on the jogamp.org server
20150814 08:05:45 <xranby_> monsieur_max: hmm unfortunally only the 2.3.2-rc-20150716 is published on the test maven
20150814 08:05:55 <monsieur_max> ok
20150814 08:06:14 <xranby_> http://jogamp.org/wiki/index.php/Maven#The_jogamp.org_test_repository_.28optional.29
20150814 08:07:36 <eclesia> xranby_: there is something wrong, there is only 2400 commits not 2401
20150814 08:07:58 <eclesia> try : hg revert --all --no-backup
20150814 08:08:01 <eclesia> hg pul
20150814 08:08:03 <eclesia> hg update
20150814 08:08:06 <eclesia> mvn clean install
20150814 08:09:07 <xranby_> ändring: 2401:0000b023aed5
20150814 08:09:07 <xranby_> märke: tip
20150814 08:09:07 <xranby_> användare: Johann Sorel<sorel.johann@inbox.com>
20150814 08:09:07 <xranby_> datum: Thu Aug 13 23:36:02 2015 +0200
20150814 08:09:20 <monsieur_max> xranby_: mmmh i'm using gradle on y projects, not sure how to use the test repo
20150814 08:10:49 <xranby_> monsieur_max: in the build.gradle file
20150814 08:11:46 <xranby_> repositories { maven { url "http://www.jogamp.org/deployment/maven/" } }
20150814 08:11:51 <xranby_> should be enough
20150814 08:12:54 <xranby_> thats the way to add maven repositorys in gradle
20150814 08:13:07 <xranby_> you can also add you local maven repository using
20150814 08:13:25 <monsieur_max> ok, would not expect to be as simple as that
20150814 08:13:35 <xranby_> repositories { mavenLocal() }
20150814 08:14:09 * badshah400 (~badshah40@anon) has joined #jogamp
20150814 08:14:17 <xranby_> in case you build something on your own macine using mvn install
20150814 08:14:41 <monsieur_max> ok, so, as the last rc is not available, not sure it's worth the try
20150814 08:15:12 <xranby_> the 2.3.2-rc-20150716 at least contain some work
20150814 08:16:28 <xranby_> the diff in changes between the 2.3.2-rc-20150716 and 2.3.2-rc-20150812 is raspberry pi fixes and sven's reworked bitfield's and gouessejs image parsers
20150814 08:17:47 * gouessej (5ee4b442@anon) has joined #jogamp
20150814 08:17:56 <gouessej> Hi
20150814 08:18:42 <xranby_> gouessej: hi!
20150814 08:18:46 <gouessej> rmk0: Please can you push the latest RC on Maven? I'd like to use it with JogAmp's Ardor3D Continuation, I don't need to use my image format detection twice :s
20150814 08:19:24 <xranby_> which deployment sctipt is used to push to the jogamp test maven? http://www.jogamp.org/deployment/maven/
20150814 08:19:37 <gouessej> xranby_: Hi. We'll probably get numerous sweet pots of Vulkan in OpenGL, I understand elect's position
20150814 08:21:23 <gouessej> monsieur_max: I probably have an example about adding the development repository into a Gradle script...
20150814 08:21:53 <xranby_> gouressej: build.grade add repositories { maven { url "http://www.jogamp.org/deployment/maven/" } }
20150814 08:22:01 <gouessej> monsieur_max: https://github.com/gouessej/Ardor3D/blob/master/build.gradle#L24
20150814 08:22:10 <monsieur_max> gouessej: well, it's ok for me from what xranby told me
20150814 08:22:29 <xranby_> all good
20150814 08:22:36 <xranby_> everyone uses the same syntax
20150814 08:23:33 <monsieur_max> mmmh still, in my current build.gradle, i specify the version ( ex : compile group: 'org.jogamp.gluegen', name: 'gluegen-rt-main', version: '2.3.1' )
20150814 08:23:43 <monsieur_max> how does it work with this test repo .
20150814 08:23:44 <monsieur_max> ?
20150814 08:24:32 <gouessej> monsieur_max: it's the same
20150814 08:24:49 <monsieur_max> ok, so just specifying 2.3.2, i suppose
20150814 08:24:51 <gouessej> monsieur_max: https://github.com/gouessej/Ardor3D/blob/master/ardor3d-jogl/build.gradle#L6
20150814 08:25:01 <gouessej> monsieur_max: not exactly
20150814 08:25:09 <monsieur_max> ok saw it
20150814 08:25:49 <gouessej> monsieur_max: there is a more recent build but it isn't on the Maven repository yet
20150814 08:26:06 <monsieur_max> would something like version:'+' would work with the 'rc'
20150814 08:26:38 <elect_> salut gouessej
20150814 08:26:50 <xranby_> faik, one of use should be able to login into the jogamp.org server and simply run the right script to promote the last test build to the test maven but I dont know which script :/
20150814 08:27:24 <xranby_> rmk0 and sgothel usually do the maven promotions
20150814 08:27:45 <gouessej> elect_: Buongiorno
20150814 08:28:35 <gouessej> elect_: "pronto" would be better as an equivalent of "salut"
20150814 08:29:11 <gouessej> elect_: "Buongiorno" means "Good morning" in italian
20150814 08:30:12 <gouessej> xranby_: There is no hurry, they can do it Sunday or Monday, can't they?
20150814 08:31:02 <elect_> pronto is like ready
20150814 08:31:40 <elect_> gouessej, at what point are you with tuer?
20150814 08:31:57 <xranby_> gouessej: *nods* however if would be good if more people knew the manual steps required to do publish jogamp maven deployments
20150814 08:32:30 * rmk0 appears
20150814 08:32:47 <gouessej> xranby_: Yes, you're right and Sven already told me where it is but I have forgotten this ... detail :s
20150814 08:32:57 <rmk0> is there a jogamp-all-platforms.7z available for the rc?
20150814 08:33:02 <xranby_> rmk0: yes
20150814 08:33:03 <rmk0> the scripts only know how to push those
20150814 08:33:20 <xranby_> http://forum.jogamp.org/RC-Test-Build-2-3-2-rc-20150812-td4035064.html
20150814 08:33:26 <gouessej> elect_: It is a bit stagnating because of a lack of spare time but TUER isn't dead for sure :)
20150814 08:33:29 <xranby_> http://jogamp.org/deployment/archive/master/gluegen_882-joal_608-jogl_1422-jocl_1071-signed/archive/jogamp-all-platforms.7z
20150814 08:33:51 <rmk0> right
20150814 08:33:54 <xranby_> rmk0: do you prefer the signed or the unsigned one?
20150814 08:34:08 <rmk0> no preference
20150814 08:34:21 <rmk0> i think if we've got signed, we should probably push signed
20150814 08:34:33 <elect_> did you write something about its specs/perf? For example how many triangles, if you have perf issues, if you do some frustum culling, etc
20150814 08:35:08 <gouessej> elect_: It's written in several articles and on JGO
20150814 08:35:15 <gouessej> elect_: frustum culling yes
20150814 08:35:28 <gouessej> elect_: contribution culling, not yet
20150814 08:35:55 <elect_> do I find the articles on your blog?
20150814 08:36:17 <gouessej> elect_: Yes but some of them are very long
20150814 08:36:19 <elect_> http://www.java-gaming.org/topics/tuer-truly-unusual-experience-of-revolution-fps-using-jogl/16094/1140/view.html?
20150814 08:36:43 <gouessej> elect_: The frustum culling is in JogAmp's Ardor3D Continuation
20150814 08:37:34 <gouessej> elect_: TUER has used several engines: http://tuer.sourceforge.net/en/project/#presentation
20150814 08:37:46 <gouessej> elect_: each one is mentioned above
20150814 08:38:39 <gouessej> elect_: the bug reports are here: http://sourceforge.net/p/tuer/tickets/
20150814 08:38:51 <gouessej> elect_: portal culling isn't fully implemented yet
20150814 08:39:00 <gouessej> elect_: it's a promising feature
20150814 08:39:10 * badshah400 (~badshah40@anon) Quit (Ping timeout: 245 seconds)
20150814 08:39:12 <elect_> you are doing raycasting for the bullets?
20150814 08:39:30 <gouessej> elect_: Yes but it is temporary
20150814 08:39:54 <elect_> I guess the AAA games use acceleration structures
20150814 08:39:59 <elect_> dont they?
20150814 08:40:37 <gouessej> elect_: Yes but I can't make a AAA game alone :) I think they use physics engines
20150814 08:40:52 <gouessej> elect_: it depends on the kind of projectile
20150814 08:41:15 <elect_> clear
20150814 08:41:16 <gouessej> elect_: a ray is enough to simulate a laser ray, it's logical :)
20150814 08:41:27 <gouessej> elect_: it's not good for a rocket
20150814 08:41:38 <elect_> because of their mass?
20150814 08:42:04 <gouessej> elect_: the rocket is slow, it's not realistic with a ray
20150814 08:42:30 <gouessej> elect_: when a projectile is very slow, the most basic collision checks are enough
20150814 08:43:03 <elect_> did you think about jbullet?
20150814 08:43:24 <gouessej> elect_: JBullet hasn't been updated since October 2010 as far as I know
20150814 08:43:51 <gouessej> elect_: If things go too much complicated, a Bullet binding will be a possible option
20150814 08:44:42 <gouessej> elect_: It sums up what I wrote about collision detection: http://sourceforge.net/p/tuer/tickets/31/
20150814 08:45:41 <gouessej> elect_: there are a few keywords in the ticket above, it helps to go further
20150814 08:45:55 <gouessej> elect_: "a posteriori collision detection"
20150814 08:46:19 <gouessej> elect_: "multisampling"
20150814 08:47:54 <gouessej> elect_: https://en.wikipedia.org/wiki/Collision_detection#A_posteriori_%28discrete%29_versus_a_priori_%28continuous%29
20150814 08:48:38 <rmk0> anyone know which pgp key is 0x961D96D5?
20150814 08:48:52 <rmk0> that's what all the releases have been signed with this year, i'm guessing it's sgothels
20150814 08:52:39 * badshah400 (~badshah40@anon) has joined #jogamp
20150814 08:54:10 <rmk0> the server is suffering from low entropy
20150814 08:54:55 <xranby_> poor server.. there is an entropy hack to feel entropy from the audiocard
20150814 08:55:18 <rmk0> it actually hit 0 a moment ago... i've never seen that happen before
20150814 08:56:43 <xranby_> checking for entropy consumes enttropy as well hmm
20150814 08:57:41 <xranby_> cat /proc/sys/kernel/random/entropy_avail
20150814 09:02:10 <xranby_> do someone ahve good experience using the randomsound daemon to add entropy? https://packages.debian.org/stretch/admin/randomsound
20150814 09:05:16 <rmk0> ah, shit... need to get sgothel to set the correct permissions on the maven repository
20150814 09:05:19 * rmk0 nudges sgothel
20150814 09:05:32 <rmk0> it needs to be jogamp:maven otherwise the maven user can't deploy to it
20150814 09:05:44 <rmk0> there are a lot of directories here and there that are jogamp:jogamp
20150814 09:06:15 <rmk0> chown -R jogamp:maven /srv/www/jogamp.org/deployment/maven
20150814 09:06:26 <rmk0> chmod -R g+w /srv/www/jogamp.org/deployment/maven
20150814 09:06:28 <rmk0> would do it
20150814 09:06:45 <xranby_> one other option to add entropy -> https://packages.debian.org/stretch/rng-tools
20150814 09:24:47 <xranby_> rmk0: is this script suid? http://jogamp.org/deployment/maven/fix-uid-mod.sh
20150814 09:25:15 <xranby_> since it is on the server
20150814 09:25:24 <xranby_> may have a suid bit set
20150814 09:25:39 <xranby_> for you to use
20150814 09:26:48 <rmk0> heh
20150814 09:27:13 <rmk0> that's amusingly specific...
20150814 09:27:22 <rmk0> it's not setuid though, and setuid scripts are bad news
20150814 09:28:59 <xranby_> bad news for my reputation?
20150814 09:29:07 <xranby_> uhoh
20150814 09:29:44 * xranby_ hides
20150814 09:30:33 <rmk0> hehe
20150814 09:30:43 <rmk0> no, they're just unsafe. i don't think any modern OS allows them
20150814 09:33:37 * xranby_ used them all the time on the 90's to make the school computer club server dial back to my home.... ish.. i guess its prescribated that i borrowed the school phone to download debian
20150814 09:34:52 <xranby_> *tales from a time without broadband internet and dialup*
20150814 09:38:33 <xranby_> rmk0: yes setuid scripts that is parting using input are definately dangerous
20150814 09:38:58 <xranby_> since the user can set what to use as the command delimiter
20150814 09:39:13 <xranby_> parsing
20150814 09:39:38 <xranby_> IFS
20150814 09:40:34 <xranby_> internal field separator
20150814 09:41:02 <xranby_> This variable determines how Bash recognizes fields, or word boundaries, when it interprets character strings.
20150814 09:41:09 <xranby_> can be misused
20150814 09:43:56 <xranby_> why someone thought it be a good idea to let an user env variable change the syntax and function of a script interpreter beats me
20150814 10:01:22 * gouessej (5ee4b442@anon) Quit (Quit: Page closed)
20150814 10:45:21 * packet0 (~packet0@anon) has joined #jogamp
20150814 10:46:47 * gouessej (5ee4b442@anon) has joined #jogamp
20150814 10:47:15 <gouessej> xranby_: I tried to build JOGL yesterday, I probably missed something, it was looking for jcpp
20150814 11:08:18 <xranby_> gouessej: jcpp is included if you clone the gluegen git recursively
20150814 11:08:31 <xranby_> it is a git submodule
20150814 11:09:31 <xranby_> git clone --recurse-submodules git://jogamp.org/srv/scm/gluegen.git gluegen
20150814 11:09:41 <gouessej> xranby_: How can I do that?
20150814 11:09:45 <gouessej> ok thanks
20150814 11:14:29 <xranby_> how to use jogl with jruby https://gist.github.com/xranby/631dc8d025ea1b466d5b
20150814 11:15:48 <xranby_> found an user on twitter who wanted to file a jogl bugreport but refused to signup for a username on our bugzilla , the issue turned to be in the users code during VAO initialization
20150814 11:16:14 <xranby_> all in all.. the user at least provided a minimal jogl jruby app :)
20150814 11:16:46 <gouessej> xranby_: Is there anything about that in the wiki?
20150814 11:16:52 <xranby_> dont think so
20150814 11:22:26 <gouessej> I wondered what kettle was. Now, I have the answer: https://github.com/florf/kettle
20150814 11:23:53 <gouessej> xranby_: florf should use the JogAmp fat JAR to simplify the use of her/his library
20150814 11:24:05 <gouessej> https://github.com/florf/foxfire/blob/master/readme.md
20150814 11:24:07 <xranby_> it work as is with the fat jar
20150814 11:24:27 <xranby_> aha
20150814 11:24:50 <xranby_> his code's first line
20150814 11:24:53 <xranby_> adds all jars
20150814 11:24:55 <xranby_> Dir.glob('*.jar').each {|jar| require jar }
20150814 11:25:07 <xranby_> found inside the same folder as the ruby file
20150814 11:25:28 <xranby_> thus you simply need to place the jogamp-fat.jar in the same folder and it starts to work
20150814 11:28:24 * gouessej (5ee4b442@anon) Quit (Ping timeout: 246 seconds)
20150814 11:30:29 * gouessej (5ee4b442@anon) has joined #jogamp
20150814 11:32:31 <xranby_> the jruby world is as mysterious as the python world
20150814 11:32:52 <xranby_> many secret commands needed to install all dependencies
20150814 11:34:06 <gouessej> xranby_: (s)he should simplify the instructions, that's what I meant as they mention numerous JARs
20150814 11:34:21 <xranby_> i agree
20150814 11:34:28 <xranby_> no idea hot ot use his project
20150814 11:34:35 <xranby_> i did manage to get his one filer to work
20150814 11:34:45 <xranby_> the gist example
20150814 11:40:28 <xranby_> ... or hers
20150814 11:40:44 <xranby_> ... or she's
20150814 11:43:59 * eclesia starting to play with maven plugin + codehaus plexus compiler + unlicense class writer : all at maven build time : that's awesome !
20150814 11:52:39 <gouessej> xranby_: Have you ever used Jainja JVM?
20150814 11:52:51 <xranby_> gouessej: i have started it
20150814 11:52:53 <gouessej> http://bloggl.thenesis.org/2014/04/jainja-08-released.html
20150814 11:53:34 <xranby_> but i have not tested all ways you can use it
20150814 11:56:52 <xranby_> i have not tested to use jainja with jogamp
20150814 11:57:33 <xranby_> the question is if we can run it.. jainja only support java 1.5
20150814 11:58:49 * packet0 (~packet0@anon) Quit (Remote host closed the connection)
20150814 11:59:18 <gouessej> xranby_: It works under Android too :) It seems to use Avian under the hood
20150814 11:59:26 <xranby_> aha
20150814 11:59:47 <xranby_> it has soo many backends :)
20150814 12:00:59 * packet0 (~packet0@anon) has joined #jogamp
20150814 12:01:03 <gouessej> I should contact his author. I'd like to run OpenJDK under Android and it's possible to do it with Avian. He probably succeeded in doing something similar.
20150814 12:02:19 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150814 12:03:09 <xranby_> Guillaume Legris is very capable, he was the one who kind of managed to get android running on top of openjdk
20150814 12:03:19 <xranby_> by hacking the android stack
20150814 12:03:30 <xranby_> inside the icedrobot project
20150814 12:05:26 <eclesia> hmm... maybe I just contact him too to make a VM for me ^^
20150814 12:05:31 <eclesia> I should*
20150814 12:06:03 <gouessej> "J2ME-compatible virtual machine written in JavaScript": https://github.com/mozilla/pluotsorbet
20150814 12:08:12 <xranby_> i know that Guillaume has managed to get his jvm running on platform sucha s the nintendo wii
20150814 12:08:20 <xranby_> etc
20150814 12:09:00 <gouessej> xranby_: :)
20150814 12:21:11 * gouessej (5ee4b442@anon) Quit (Quit: Page closed)
20150814 12:27:27 * rmk0 prods sgothel with a broom
20150814 12:31:16 <xranby_> http://joshondesign.com/2015/08/01/needajob
20150814 12:37:46 <xranby_> rmk0: sgother is wearing a special broom resistant suit ... that expire on sunday
20150814 12:37:56 <xranby_> sgothel
20150814 12:38:07 <rmk0> hrm
20150814 13:46:44 <eclesia> broomproof ... that's a profesionnal
20150814 13:51:33 <zubzub> https://wiki.openjdk.java.net/display/Sumatra/Main
20150814 13:51:37 <zubzub> anyone heard from that?
20150814 13:53:30 <zubzub> anyone tried this:
20150814 13:53:32 <zubzub> http://www.oracle.com/technetwork/oracle-labs/program-languages/overview/index.html
20150814 13:54:15 <xranby_> zubzub: i try to avoid agreein with the OTN license
20150814 13:55:08 <zubzub> I usually lie when I agree with it
20150814 14:40:39 <eclesia> is it possible to display something using NEWT even when there is no opengl available ?
20150814 14:41:23 <zubzub> you mean hw accel gl?
20150814 14:41:27 <zubzub> or any gl?
20150814 14:41:33 <eclesia> any gl
20150814 14:41:45 <zubzub> doubtful as then you are at the mercy of your display system
20150814 14:42:53 <elect_> http://i.imgur.com/wpW8uC6.png
20150814 14:42:59 <eclesia> xranby_: you were talking about something about it with gouessej a few das ago if I remember correctly. somethign about displaying an image obtained from cpu rasterizer
20150814 14:44:17 <xranby_> eclesia: my optinion is that OS such as linux always have a software opengl driver
20150814 14:44:43 <xranby_> eclesia: thus i am not that worried, mesa3d can provide a fallback
20150814 14:44:48 <xranby_> for use on servers
20150814 14:45:06 <eclesia> xranby_: so worst case there is something like a GL1 available ?
20150814 14:45:27 <xranby_> mesa 3d implement gl 3 in software
20150814 14:45:40 <xranby_> and is on its way to have full gl 4 support
20150814 14:45:45 <zubzub> newt *could* provide a bytebuffer sw buffer to display stuff
20150814 14:45:56 <zubzub> and provide an implementation for X11 using shm extension
20150814 14:46:06 <zubzub> for wayland it exists build-in
20150814 14:46:12 <zubzub> and surely it also exists for mac & win
20150814 14:46:28 <xranby_> eclesia: i have one usecase where i would like to have a simple 2d CPU api
20150814 14:46:39 <zubzub> thing is you need a widely different implementation for all those cases
20150814 14:46:45 <xranby_> to clear a no hardware accelerated newt window
20150814 14:47:00 <xranby_> where i was thinking there may be classes inside your project to do it
20150814 14:47:37 <xranby_> for example NEWT can initialize a regular Window
20150814 14:47:58 <xranby_> and i use it as the underlay to capture user input for an overlay window
20150814 14:48:15 <xranby_> if the overlay is transparent i would like the underlay to be transparent as well
20150814 14:48:27 <xranby_> right now newt gives me a window filled with garbage
20150814 14:48:38 <xranby_> and an event from the os that thewindow needs to be repainted
20150814 14:48:43 <eclesia> there is a cpu rasterizer, didn't use it for several months, not sure if it still works correctly.
20150814 14:48:58 <xranby_> yes such a cpu rasterizer may be usefull
20150814 14:49:18 <xranby_> gouessej has the idea to show dialouge boxes
20150814 14:49:22 <eclesia> but it's made using my apis
20150814 14:49:34 <xranby_> that require a cpu rasterizer that can at least print some text
20150814 14:49:38 <eclesia> geometries, colors, images , buffers ...
20150814 14:50:41 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150814 14:52:20 <eclesia> xranby_: I'll try to make a demo for this usecase this week end
20150814 14:53:17 <xranby_> eclesia: thank you!
20150814 14:54:38 * elect_ (~elect@anon) Quit (Ping timeout: 260 seconds)
20150814 15:38:26 * monsieur_max (~maxime@anon) has joined #jogamp
20150814 16:04:33 * eclesia (~husky@anon) has left #jogamp
20150814 16:28:53 * packet0 (~packet0@anon) Quit ()
20150814 17:00:42 * packet0 (~packet0@anon) has joined #jogamp
20150814 18:16:54 * packet0 (~packet0@anon) Quit ()
20150814 18:17:06 * packet0 (~packet0@anon) has joined #jogamp
20150814 18:33:19 * packet0 (~packet0@anon) Quit ()
20150814 18:33:52 * packet0 (~packet0@anon) has joined #jogamp
20150814 19:16:17 * packet0 (~packet0@anon) Quit ()
20150814 19:47:37 * Eclesia (~eclesia@anon) has joined #jogamp
20150814 19:52:12 <Eclesia> xranby_: your compilation error is really odd. I made a clean clone, removed my .m2 ... no issues at all
20150814 19:53:31 <xranby_> ok i will try the same
20150814 19:53:42 <xranby_> pulled in your latest fixes
20150814 19:53:52 <xranby_> deleting my .m2
20150814 19:55:03 <xranby_> ... removing the whole thing and making a clean clone
20150814 20:03:19 * Eclesia writing the cpu rendering demo
20150814 20:31:04 * rmk0 (~rmk0@anon) Quit (Ping timeout: 246 seconds)
20150814 20:33:04 <Eclesia> :) looks like the cpu rasterizer still work fine
20150814 20:37:07 <Eclesia> http://i.imgur.com/mVB9JeU.png
20150814 20:38:13 <Eclesia> xranby_: here is the demo : https://bitbucket.org/Eclesia/unlicense/src/tip/demo/src/main/java/demo/image/CPURendering.java
20150814 20:38:21 * rmk0 (~rmk0@anon) has joined #jogamp
20150814 20:38:21 * rmk0 (~rmk0@anon) Quit (Changing host)
20150814 20:38:21 * rmk0 (~rmk0@anon) has joined #jogamp
20150814 20:38:38 <Eclesia> I believe the code is self explaining hehe
20150814 20:39:29 <xranby_> amazing really
20150814 20:40:32 <Eclesia> xranby_: font rendering are not as nice as the opengl rendering. no multisampling unfortunatly. or you could make the image twice the size and scale it down using an image operator. that would work
20150814 20:45:01 * Eclesia really has to finish C to Java transcoder to have the sweet font rasterizer from https://github.com/nothings/stb/blob/master/stb_truetype.h
20150814 20:46:54 <Eclesia> xranby_: did it compile this time ?
20150814 20:47:44 <xranby_> Eclesia: no it did not.. someting is really odd
20150814 20:49:01 <xranby_> i will scrutinize my machine
20150814 20:49:38 <xranby_> [ERROR] /home/familjen/unlicense/unlicense/api/api-common/src/main/java/un/api/predicate/ClassMethod.java:[35,85] cannot find symbol
20150814 20:49:39 <xranby_> [ERROR] symbol: method getParameterCount()
20150814 20:49:57 <xranby_> i am at 2402:e57acd91a76c
20150814 20:50:11 <Eclesia> ? that's a completely different error
20150814 20:50:19 <xranby_> yes if failed earlier this time
20150814 20:50:40 <Eclesia> I'll make another try on my vm
20150814 20:52:49 <xranby_> Eclesia: i need to check my system.. since this error bailed out not finding java.lang.reflect.Method getParameterCount
20150814 20:53:25 <Eclesia> did you mess with your jdk somehow ?
20150814 20:53:51 <xranby_> yes i am using jdk 7 at tme moment maybe you need to use getParameterTypes().length
20150814 20:54:21 <xranby_> getParameterCount is jdk8 specific
20150814 20:54:45 <Eclesia> My bad,I didn't check
20150814 20:55:21 <Eclesia> xranby_ could you try on your machine ?
20150814 20:55:40 <Eclesia> too see if I missed anything else ^^
20150814 20:56:41 <xranby_> Eclesia: of course especially now that i think i have understood the issue
20150814 20:56:59 <Eclesia> build worked on my vm too. but it's also a jdk8
20150814 20:58:08 <xranby_> ok now i am back at the failing API - geometries
20150814 20:58:09 <xranby_> [ERROR] /home/familjen/unlicense/unlicense/api/api-geometry/src/main/java/un/api/geometry/index/BoundingAreaCheck.java:[319,16] cannot find symbol
20150814 20:58:09 <xranby_> symbol: method nextDown(double)
20150814 20:58:09 <xranby_> location: class java.lang.Math
20150814 20:58:24 <xranby_> have to see what jdk 7 people did
20150814 20:58:24 <Eclesia> erf
20150814 21:00:59 <Eclesia> I'm downloading a jdk6 an I will fix all this
20150814 21:01:02 <xranby_> nextAfter(d, Double.NEGATIVE_INFINITY); is equal to nextDown(double);
20150814 21:06:16 <xranby_> copiled it all using jdk 7 now BUILD SUCCESS
20150814 21:06:29 <Eclesia> sorry for the troubles
20150814 21:07:11 <xranby_> Eclesia: do you want a pull request?
20150814 21:07:21 <xranby_> or have you done jdk6 already?:)
20150814 21:08:32 <Eclesia> I'll do it, jdk6 will certainly need some other fixes too
20150814 21:22:15 <Eclesia> xranby_: fixed for jdk7. jdk6 will be a bit longer because of some generic issues.
20150814 21:43:52 <Eclesia> good night ++
20150814 21:44:00 * Eclesia (~eclesia@anon) Quit (Quit: Leaving.)
20150814 23:28:32 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150815 00:53:34 * a11mad11 (4c47eb44@anon) has joined #jogamp
20150815 01:14:24 * a11mad11 (4c47eb44@anon) Quit (Ping timeout: 246 seconds)
20150815 01:44:39 * a11mad11 (c0e2cd65@anon) has joined #jogamp
20150815 03:05:35 * packet0 (~packet0@anon) has joined #jogamp
20150815 03:31:32 * badshah400 (~badshah40@anon) has joined #jogamp
20150815 05:05:29 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20150815050529.html