#jogamp @ irc.freenode.net - 20141020 05:06:24 (UTC)
20141020 05:06:24 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20141019050623.html
20141020 05:06:24 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20141020050624.html
20141020 05:51:26 * xranby (~xranby@anon) Quit (Ping timeout: 272 seconds)
20141020 06:04:51 * xranby (~xranby@anon) has joined #jogamp
20141020 06:32:51 * eclesia (~husky@anon) has joined #jogamp
20141020 06:32:56 <eclesia> good morning
20141020 07:04:52 * monsieur_max (~maxime@anon) has joined #jogamp
20141020 07:23:31 * masterzen_ (~masterzen@anon) Quit (Quit: Au revoir!)
20141020 07:40:48 * xranby (~xranby@anon) Quit (Ping timeout: 260 seconds)
20141020 07:55:28 * xranby (~xranby@anon) has joined #jogamp
20141020 08:33:18 <zubzub> How am I supposed to use a VBO?
20141020 08:33:27 <zubzub> can I reuse it every time?
20141020 08:33:37 <zubzub> ie before I call drawTriangles
20141020 08:33:48 <zubzub> or is it ok to create a new one each time?
20141020 08:34:05 <zubzub> (my guess is to reuse it?)
20141020 08:40:57 <eclesia> yes reuse them
20141020 08:43:08 <xranby> zubzub: the main idea with VBO is to make OpenGL more explicit, you as a programmer knows best if it is good to reuse or re create them.
20141020 08:43:27 <xranby> zubzub: it makes new opengl driver easier to create, because the driver do not have to guess
20141020 08:47:08 <zubzub> well in my case I can reuse them just fine
20141020 08:47:18 <zubzub> which I should do for better performance I guess
20141020 08:56:37 <xranby> yes, reusing is the key to better performance
20141020 09:00:43 * doev (~doev@anon) has joined #jogamp
20141020 09:02:58 * zzuegg (~zzuegg@anon) has joined #jogamp
20141020 09:39:14 <sgothel> @zuzub: look at instancing VBO usage, where you can pass an index to the shader, where you could apply extra 'stuff' like tranformation for each vertex
20141020 10:37:54 * xranby (~xranby@anon) Quit (Quit: Leaving.)
20141020 10:37:58 * xranby1 (~xranby@anon) has joined #jogamp
20141020 11:29:42 <sgothel> @Mark/All: http://forum.jogamp.org/Forum-improvement-tp4033326p4033414.html <- 'Discourse' [short] review ..
20141020 11:30:08 <sgothel> https://github.com/discourse/discourse
20141020 11:30:31 <sgothel> Ruby .. it seems
20141020 11:35:06 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=1095#c3 .. updated
20141020 11:38:28 * jvanek (jvanek@anon) has joined #jogamp
20141020 11:42:13 * xranby1 (~xranby@anon) Quit (Ping timeout: 258 seconds)
20141020 11:55:25 * xranby (~xranby@anon) has joined #jogamp
20141020 12:13:25 <eclesia> ruby +1, lots of project in public domain in this language ^^
20141020 12:15:30 * gouessej (5ee4b442@anon) has joined #jogamp
20141020 12:15:38 <gouessej> Hi
20141020 12:15:42 <sgothel> hi
20141020 12:15:47 <gouessej> Sven, I need your help
20141020 12:15:47 <eclesia> hi
20141020 12:16:04 <gouessej> I get this under Windows 7, probably ATI: javax.media.opengl.GLException: AWT-EventQueue-0: WindowsWGLContex.createContextImpl ctx !ARB, profile > GL2 requested (OpenGL >= 3.0.1). Requested: GLProfile[GL3bc/GL3bc.hw], current: 2.1 (Compat profile, FBO, hardware) - 2.1.8787
20141020 12:16:21 <gouessej> It might have something to core with the bug 1038
20141020 12:16:29 <gouessej> "to do" sorry
20141020 12:16:55 <gouessej> Should I use jogl.disable.openglcore?
20141020 12:17:54 <gouessej> sgothel: https://github.com/sgothel/jogl/blob/master/src/jogl/classes/jogamp/opengl/windows/wgl/WindowsWGLContext.java#L371
20141020 12:18:03 * doev (~doev@anon) Quit (Quit: Verlassend)
20141020 12:18:17 <sgothel> autch ..
20141020 12:18:42 <sgothel> so that means no ARB extension exists .. ?
20141020 12:20:04 <sgothel> jogl.disable.openglcore, maybe
20141020 12:20:31 <gouessej> :s
20141020 12:22:16 <gouessej> Why does it create a GL3 context whereas I use getMaxFixedFunc(true)?
20141020 12:22:27 <sgothel> it shall not
20141020 12:23:01 <sgothel> GL_PROFILE_LIST_MAX_FIXEDFUNC = new String[] { GL4bc, GL3bc, GL2, GLES1 }; <- it uses this set
20141020 12:23:40 <sgothel> so something must have went wrong, check the GLPRofile context mapping .. e.g. simply via GLWindow.main - or better the whole GLProfile unit test
20141020 12:24:07 <sgothel> TestGLProfile01NEWT <- this one
20141020 12:24:11 <gouessej> ok I will ask the customer to use the applet to get it
20141020 12:26:37 <gouessej> ATI Radeon 3100
20141020 12:32:21 <rmk0> .o.
20141020 12:32:32 <rmk0> i think you'd have to be fucking insane to run critical infrastructure on ruby
20141020 12:32:44 <rmk0> might as well write the whole thing in javascript and then throw it away
20141020 12:32:46 <rmk0> HELLO
20141020 12:33:54 <sgothel> :)
20141020 12:34:52 <sgothel> pls reply to bug report
20141020 12:35:32 <rmk0> i'm not sure me ranting about the sorry state of all "web development" would be all that constructive
20141020 12:35:34 <gouessej> 1038?
20141020 12:37:30 <xranby> gouessej: ATI's latest "catalyst" drivers only support Radeon R9 200, R7 200, HD 7000, HD 6000 and HD 5000 Series
20141020 12:38:07 <xranby> gouessej: i think ATi/AMD has dropped support for Radeon 3100, if your user has this card then only linux + mesa is "supported"
20141020 12:41:30 <gouessej> xranby: it's under Windows 7 :(
20141020 12:42:19 <xranby> ait/amd do not support this card, even if you use windows 7 http://support.amd.com/en-us/download
20141020 12:44:12 <xranby> mesa can be built for windows,
20141020 12:44:50 <gouessej> xranby: ok, it is a possibility if I can't work around the driver bug
20141020 12:52:06 <xranby> gouessej: the cheapest solution for the user is to upgrade the gpu,, time and cost wise
20141020 12:52:26 <xranby> its clear that hardware do not simply work on windows any more
20141020 12:52:40 <xranby> because the gpu manufacturers decide to drop support for old versions
20141020 12:53:52 <rmk0> does mesa give hardware acceleration on windows? assumed it was tied to the drm/dri infrastructure
20141020 12:54:07 <rmk0> it does have that llvm-based software backend that's supposed to be quite fast
20141020 12:57:54 <gouessej> xranby: this is what I advise when I can, especially under Windows
20141020 13:16:45 * bbbruce (~bx@anon) has joined #jogamp
20141020 13:16:45 * bbbruce (~bx@anon) Quit (Client Quit)
20141020 14:18:14 <gouessej> It's not very motivating. I just hope that most drivers of low end graphics cards under Windows won't have similar problems.
20141020 14:45:49 * gouessej (5ee4b442@anon) Quit (Quit: Page closed)
20141020 15:06:20 * eclesia (~husky@anon) has left #jogamp
20141020 15:45:38 * gouessej (5ee4b442@anon) has joined #jogamp
20141020 15:45:55 <gouessej> sgothel: I have a question for you about your bug fix (1038)
20141020 15:46:13 <gouessej> sgothel: https://github.com/sgothel/jogl/blob/master/src/jogl/classes/jogamp/opengl/windows/wgl/WindowsWGLContext.java#L366
20141020 15:46:37 <gouessej> sgothel: why do you use isGL3()? Shouldn't we use isGL3core() here?
20141020 15:48:17 <gouessej> sgothel: Your error message talks about OpenGL >= 3.0.1
20141020 15:48:32 <gouessej> sgothel: the chosen capabilities object uses GL3bc
20141020 15:48:49 <gouessej> sgothel: I don't want to use the core profile
20141020 15:49:17 <gouessej> sgothel: ATI Radeon 3100 supports only OpenGL 2.1 by the way
20141020 16:04:53 * gouessej (5ee4b442@anon) Quit (Quit: Page closed)
20141020 16:14:08 * jvanek (jvanek@anon) Quit (Quit: Leaving)
20141020 16:35:52 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20141020 17:10:43 * monsieur_max (~maxime@anon) has joined #jogamp
20141020 21:27:09 <sgothel> http://blog.lusis.org/blog/2014/10/19/software-empathy/ .. quotes:
20141020 21:27:21 <sgothel> "Remember: if a new user has a bad time, it’s a bug in logstash."
20141020 21:27:36 <sgothel> "You don’t have to accept that the user is always right but you have to accept that there is a user." :)
20141020 21:39:58 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20141020 22:00:47 * gouessej (5279497b@anon) has joined #jogamp
20141020 22:01:50 <gouessej> sgothel: https://jogamp.org/bugzilla/show_bug.cgi?id=1038#c6
20141020 22:03:08 <gouessej> sgothel: If you could reply to those 2 questions, it would help me to investigate further in the meantime (I will get some information in a few days)
20141020 22:04:34 <gouessej> sgothel: Good night
20141020 22:04:40 * gouessej (5279497b@anon) Quit (Client Quit)
20141020 22:56:12 * hharrison (~chatzilla@anon) has joined #jogamp
20141020 22:57:38 <hharrison> Well, that was a frustrating day, one model, for one case in our testsuite produces different results when run on a Java6 VM and a Java7 vm
20141020 22:57:51 <hharrison> just great
20141020 22:58:34 <hharrison> Anyone recall any java bugs related to double-precision?
20141020 23:57:06 <hharrison> yay: http://bugs.java.com/bugdatabase/view_bug.do?bug_id=7021568
20141021 01:47:17 * xranby (~xranby@anon) Quit (Ping timeout: 255 seconds)
20141021 02:01:47 * xranby (~xranby@anon) has joined #jogamp
20141021 04:52:51 * xranby (~xranby@anon) Quit (Ping timeout: 246 seconds)
20141021 05:05:31 * xranby (~xranby@anon) has joined #jogamp
20141021 05:06:24 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20141021050624.html