| 00:05.12 | starseeker | Ralith: Question. IIRC, Stellarium uses overriding of the Qt drawing and has the widget draw opengl instead - would it be possible to do a similar override and have the widget output Ogre calls, (i.e. speak the same language the original in-Ogre toolkit did?) |
| 00:15.12 | starseeker | Ralith: I'm going to take a stab at building g3d - are there recent build instructions somewhere? |
| 00:16.34 | starseeker | hunts down nvidia-cg |
| 00:21.23 | starseeker | eyes src/other/uuid as it fails to build due to missing .h file... |
| 00:32.54 | Ralith | starseeker: I'm not sure how practical it is to instruct Ogre at such a low levelâI think the links I mentioned last night might bear investigation before that, especially considering the possibility of them integrating cleanly with the existing Qt hold. |
| 00:32.59 | Ralith | er |
| 00:33.00 | Ralith | code |
| 00:33.26 | starseeker | how do I build this sucker? Do the INSTALL instructions still hold? |
| 00:34.14 | Ralith | INSTALL instructions? |
| 00:34.24 | Ralith | old g3d is actually going to be broken now |
| 00:34.34 | starseeker | OK, how do I build your code? |
| 00:34.34 | Ralith | new g3d is a matter of cmake . && make ogretest |
| 00:34.43 | starseeker | will that build Ogre too? |
| 00:34.47 | Ralith | no. |
| 00:34.55 | starseeker | ... |
| 00:34.56 | Ralith | your OS should have a Ogre package |
| 00:35.10 | starseeker | I thought you stuck it in src other |
| 00:35.18 | Ralith | I did |
| 00:35.21 | Ralith | haven't done any buildsystem integration yet |
| 00:35.26 | starseeker | ah |
| 00:35.45 | Ralith | also, it may be advisable to simply use latest ogre trunk instead |
| 00:35.55 | Ralith | if you're going to be building it anyway. |
| 00:35.58 | starseeker | do you need ois? |
| 00:36.00 | Ralith | no. |
| 00:36.05 | starseeker | k |
| 00:36.05 | Ralith | hm |
| 00:36.11 | Ralith | cmake won't let you go on without it though |
| 00:36.16 | Ralith | I should probably scrap old g3d's buildsystem |
| 00:36.21 | Ralith | since I've broken it anyway at this point |
| 00:36.26 | Ralith | lemme do that |
| 00:36.29 | starseeker | nods |
| 00:36.45 | starseeker | starts Ogre building - on this machine, that'll be hours |
| 00:37.20 | Ralith | almost suggests binaries, but recalls that we depend on several tweaks not extant in the official builds |
| 00:38.34 | starseeker | this is gentoo - I do source :-P |
| 00:39.04 | Ralith | s/source/hours-long periods of impatience for every major install/ |
| 00:39.05 | Ralith | :D |
| 00:39.17 | starseeker | yeah, that too |
| 00:39.23 | starseeker | but, it's a good time to hit the gym |
| 00:39.29 | Ralith | seeya! |
| 00:39.38 | Ralith | will have cmake reworked before you're back. |
| 00:39.57 | starseeker | downloads the BZFlag FLOSS broadcast |
| 00:40.18 | starseeker | "Christopher "Sean" Morrison" for the free and open source tank game, BZFlag." |
| 00:40.25 | starseeker | http://twit.tv/floss78 |
| 00:40.31 | Ralith | no mention of BRL-CAD? |
| 00:41.05 | starseeker | we'll see :-) |
| 00:41.15 | starseeker | presumably bzflag has the wider audience right now |
| 00:41.24 | Ralith | regrettably true. |
| 00:43.47 | starseeker | notes Ogre uses FreeImage - which is GPL and FIPL... |
| 00:43.52 | starseeker | http://freeimage.sourceforge.net/license.html |
| 00:43.54 | starseeker | hrm |
| 00:44.37 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-27.sbndin.btas.verizon.net) | |
| 00:45.06 | Ralith | FIPL? |
| 00:45.33 | Ralith | if we do happen to decide to pass on ogre after all, that would certainly make the OpenGL/Qt issue easier. |
| 00:50.31 | CIA-38 | BRL-CAD: 03ralith * r35249 10/rt^3/trunk/src/g3d/ (CMakeLists.txt main.cxx main.cxx ogretest.cxx): Retargeted build system at new G3D exclusively. Mocha, RBGUI, and OIS are no longer necessary for build. |
| 00:57.09 | CIA-38 | BRL-CAD: 03ralith * r35250 10/rt^3/trunk/src/g3d/INSTALL: Added a simple, minimal INSTALL file. |
| 01:04.57 | Ralith | hm, that's odd |
| 01:05.21 | Ralith | I just checked my soc account balance for the first time, and I've got $2,005 rather than $2,500 O.o |
| 01:09.46 | brlcad | actually, bzflag has a very similarly sized community, code base, devs, and google counts .. but is just a lot more well known :) |
| 01:10.21 | brlcad | starseeker: uuid might be my fault |
| 01:10.43 | Ralith | BRL-CAD has a community outside of its developers? |
| 01:10.47 | Ralith | I didn't know that >_> |
| 01:11.34 | Ralith | it occurs to me that, had we manpower to moderate and maintain, a web forum might help. |
| 01:11.51 | brlcad | there is a web forum :) |
| 01:11.55 | brlcad | it just sucks |
| 01:12.04 | *** join/#brlcad LarsG (n=lars@nusnet-201-104.dynip.nus.edu.sg) | |
| 01:12.09 | Ralith | the sourceforge forum doesn't count :P |
| 01:12.11 | brlcad | there's also a new one, but waiting for sf.net to work on data migration first |
| 01:12.14 | *** part/#brlcad LarsG (n=lars@nusnet-201-104.dynip.nus.edu.sg) | |
| 01:12.19 | Ralith | oh, cool |
| 01:13.26 | Ralith | that's odd; the 'Geometry' tracker is listed as having 3 items, but when I go to it it's shown as empty |
| 01:14.26 | brlcad | check your filters |
| 01:14.45 | Ralith | oh, those things are cached O.o |
| 01:14.48 | Ralith | that's... odd behavior |
| 01:15.42 | Ralith | goes looking for someone to complain to about the money thing. |
| 01:16.50 | Ralith | aaand the appspot site is down. |
| 01:27.32 | brlcad | heh |
| 01:27.34 | brlcad | it's a conspiracy! |
| 01:28.02 | Ralith | indeed. >_> |
| 01:29.12 | Ralith | oh cool! |
| 01:29.30 | Ralith | sourceforge modified itself such that you can directly wget download links and be correctly redirected! |
| 01:33.27 | CIA-38 | BRL-CAD: 03brlcad * r35251 10/rt^3/trunk/include/uuid/ (uuid_vers.h uuidcpp.h): these probably belong over in src/other/uuid with the rest of their sources. revert the bad changes from r35247 that caught these in the mix. |
| 01:54.35 | *** join/#brlcad louipc (n=louipc@archlinux/trusteduser/louipc) | |
| 01:57.28 | *** join/#brlcad louipc (n=louipc@archlinux/trusteduser/louipc) | |
| 02:13.35 | brlcad | starseeker: yeah, that will be a problem down the road that we'll have to address before distributing |
| 02:13.59 | brlcad | freeimage can be disabled, though -- but you have to supply an alternate image loader |
| 02:14.20 | Ralith | can it use DeviL? |
| 02:14.22 | brlcad | --disable-freeimage --enable-openexr is probably the way to go |
| 02:14.45 | Ralith | recalls openexr having issues on FreeBSD |
| 02:15.19 | brlcad | it already uses devil too |
| 02:15.24 | brlcad | or can at least |
| 02:15.55 | Ralith | then I imagine it should be easy to swap in for freeimage |
| 02:16.02 | starseeker | considers disabling freeimage now - it's not working out of box... |
| 02:16.10 | Ralith | I *think* DeviL is more helpfully licensed. |
| 02:16.23 | starseeker | brlcad: are those arguments to cmake? |
| 02:16.27 | brlcad | its lgpl |
| 02:16.37 | brlcad | those are ogre build options |
| 02:16.43 | Ralith | starseeker: I don't think Ogre's cmake usage is stable yet |
| 02:16.48 | Ralith | use ./configure etc. |
| 02:16.54 | starseeker | oh |
| 02:17.20 | Ralith | imma go work on foods. |
| 02:17.22 | Ralith | good luck with ogre. |
| 02:17.34 | starseeker | hmm - did we not import the configure.ac file? |
| 02:17.46 | Ralith | possible, though I'd be surprised |
| 02:17.46 | starseeker | pulls trunk |
| 02:17.58 | Ralith | brlcad: I do wonder if Ogre is really worth all the trouble. |
| 02:18.26 | Ralith | especially considering that its intended targetâgamesâis not what we're doing. |
| 02:21.43 | brlcad | it's intended target is a render engine, no more no less -- just happens to be the gaming that is attracted to it the most |
| 02:22.02 | brlcad | we'd have the same if not more problems with pretty much any of the fully featured graphics engines |
| 02:22.32 | brlcad | ogre has the advantage of having a ton of momentum, an active community, great project leadership |
| 02:28.12 | Ralith | but do we *need* a fully featured graphics engine? |
| 02:28.44 | Ralith | the latest-and-greatest graphical effects are generally orthogonal to the problem of modeling, imo. |
| 02:29.07 | brlcad | we're not using it for effects |
| 02:29.13 | louipc | we need it for the simulations in the future :D |
| 02:29.42 | Ralith | brlcad: that's my point. Ogre's goals and our requirements don't seem terribly well aligned. |
| 02:29.44 | brlcad | at least it's certainly not a primary benefit |
| 02:29.58 | brlcad | their goals aren't just "effects" either .. |
| 02:30.52 | Ralith | I know, but they're certainly much more interested in e.g. support for highly realistic rendering than is necessary for our use. |
| 02:31.06 | brlcad | because they already do what we need them to do |
| 02:31.18 | starseeker | actually, highly realistic rendering can be nice for model visualization (e.g. product design) |
| 02:31.20 | brlcad | all that's left is "make it more pretty" |
| 02:31.26 | brlcad | you seem to be forgetting a core feature of a graphics engine .. |
| 02:31.33 | brlcad | good scene graph management |
| 02:31.38 | brlcad | they have that down |
| 02:31.46 | Ralith | I guess I'll take your word for that |
| 02:32.15 | Ralith | hopefully it will pay off once g3d starts displaying more than test spheres. |
| 02:32.22 | starseeker | it will |
| 02:32.27 | brlcad | it doesn't matter when you're displaying boxes and simple objects, but really starts to matter immensely with complex real-world geometries |
| 02:33.12 | brlcad | mged gets away with it "okay" simply because it only draws the wireframes, but even chokes out on that at times because it doesn't have scene graph management |
| 02:33.36 | brlcad | and makes "making it look good" nearly impossible without major restructuring (which amounts to implementing a graphics engine) |
| 02:33.48 | louipc | hmmm |
| 02:33.58 | Ralith | I'm also a bit put off by their emphasis on windows support, but I guess it's Good Enough on unixen. |
| 02:35.18 | starseeker | erm. Latest trunk checkout of ogre has neither configure.ac nor bootstrap in the top level... |
| 02:35.27 | Ralith | >_> |
| 02:35.27 | starseeker | wonders if they're moving over for real |
| 02:35.34 | Ralith | starseeker: latest trunk might be. |
| 02:35.45 | Ralith | last I checked was the same checkout that I committed |
| 02:35.50 | Ralith | could be they've got it stable now. |
| 02:36.04 | starseeker | looks for how to enable openexr and disable freeimage... |
| 02:36.21 | Ralith | starseeker: ccmake .? |
| 02:36.30 | starseeker | ah |
| 02:36.40 | starseeker | hasn't played with cmake in a while |
| 02:37.29 | Ralith | wonders how OSG compares |
| 02:38.01 | Ralith | eats. |
| 02:41.25 | starseeker | arrgh - they can disable freeimage but not enable anything else... |
| 02:42.10 | starseeker | gives it a whirl... |
| 02:43.59 | starseeker | Ralith: are the autotools build files in a lower directory? http://brlcad.svn.sourceforge.net/viewvc/brlcad/rt%5E3/trunk/src/other/ogre/ |
| 02:50.02 | brlcad | notes http://cvs.fedoraproject.org/viewvc//rpms/ogre/F-7/ogre.spec?view=markup which disables freeimage and cg because of the license |
| 02:50.35 | starseeker | has to agree with cg but regrets the implications |
| 02:52.46 | starseeker | brlcad: what are our usage implications without cg turned on? |
| 02:52.46 | brlcad | haven't read them in detail, but don't believe cg is as much an issue |
| 02:52.52 | starseeker | nods |
| 02:52.53 | brlcad | they're more problematic for gpl codes |
| 02:56.39 | Ralith | starseeker: weren't you using trunk now? |
| 02:56.52 | starseeker | I've tried both |
| 02:57.00 | starseeker | building trunk now, with freeimage off |
| 02:57.17 | starseeker | hoping it does something sane, since I didn't see any option to explicitly enable openexr or devil |
| 02:57.58 | Ralith | if it got past cmake, they're either embarassingly negligent or it's taken care of. |
| 02:59.08 | starseeker | if the cmake system is unstable, they may not have gotten around to the second tier stuff... |
| 02:59.43 | Ralith | if they scrapped GBS when cmake was unstable then they're more than negligent. |
| 03:00.20 | starseeker | Ralith: am I nuts or is there no autotools stuff in our branch? |
| 03:00.49 | Ralith | there appears to be no autotools stuff in our branch. |
| 03:00.53 | Ralith | I wonder how that happened O.o |
| 03:01.12 | Ralith | thought I recalled it still being in use there. |
| 03:01.16 | Ralith | guess I'm misremembering |
| 03:01.33 | starseeker | does your Qt work require FreeImage right now? |
| 03:02.11 | Ralith | uh, I have no idea |
| 03:02.23 | Ralith | I certainly don't use it directly |
| 03:02.30 | Ralith | and I can't imagine depending on any ogre functions that depend on it |
| 03:02.40 | starseeker | guess we'll find out :-) |
| 03:26.17 | starseeker | welllll.... |
| 03:36.00 | starseeker | grrrrr |
| 03:36.06 | starseeker | it can't find any of the libraries |
| 03:36.16 | Ralith | -_- |
| 03:36.29 | Ralith | it really seems to have bad support for things which aren't the official windowsy visual studio packages |
| 03:38.46 | starseeker | ogre built and installed, and so did g3d, but it doesn't want to run... |
| 03:39.07 | Ralith | yeah, g3d's install process isn't really debugged at all |
| 03:39.09 | Ralith | what error? |
| 03:39.17 | starseeker | manually symlinked libs into /usr/lib, but still gets ./RenderSystem_GL. System Error: ./RenderSystem_GL.so: undefined symbol: _ZN4Ogre15ResourceManager6unloadEm in DynLib::load at /home/cyapp/cadtoplevel/brlcad/ogre/OgreMain/src/OgreDynLib.cpp (line 90) |
| 03:39.29 | Ralith | ah, that. |
| 03:39.50 | Ralith | I guess they never fixed their cmake scripts for nonwindows. |
| 03:40.04 | Ralith | iirc it's that error which made me mark them off as unstable in the first place |
| 03:40.32 | starseeker | is there a workaround? |
| 03:40.33 | Ralith | starseeker: what're the libs named, and where are they? |
| 03:41.04 | Ralith | are you sure you don't have an old ogre lying around? |
| 03:41.20 | starseeker | usual suspects - librt.so.19, libged.so.19, libbu.so.19, etc... |
| 03:41.29 | starseeker | yeah, pretty sure |
| 03:41.59 | Ralith | er |
| 03:42.01 | Ralith | the Ogre libs, I mean |
| 03:42.49 | starseeker | I don't think I have an old ogre |
| 03:42.58 | Ralith | what're the libs named? |
| 03:43.33 | starseeker | libOgreMain.so.1 |
| 03:44.14 | starseeker | and we can't work with ogre-1.6.2, correct? |
| 03:44.29 | Ralith | correct. |
| 03:44.34 | Ralith | what about the libs it's trying to load? :P |
| 03:45.25 | starseeker | http://pastebin.bzflag.bz/m24552681 |
| 03:45.46 | Ralith | not those ones |
| 03:45.51 | Ralith | the ones Ogre's trying to load at runtime |
| 03:47.04 | Ralith | ls /usr/lib/OGRE or w/e |
| 03:48.02 | starseeker | It's installed in /usr/local/lib/OGRE/ |
| 03:48.14 | Ralith | okay, what's the contents of that dir? |
| 03:48.36 | starseeker | Plugin_BSPSceneManager.so Plugin_OctreeZone.so RenderSystem_GL.so |
| 03:48.36 | starseeker | Plugin_CgProgramManager.so Plugin_PCZSceneManager.so cmake |
| 03:48.36 | starseeker | Plugin_OctreeSceneManager.so Plugin_ParticleFX.so |
| 04:00.16 | Ralith | hm |
| 04:00.18 | Ralith | that seems about right |
| 04:00.29 | Ralith | identical to my system, even. |
| 04:00.35 | Ralith | lemme update my Ogre and see if I don't encounter the same issue |
| 04:35.28 | Ralith | starseeker: how did you disable freeimage? |
| 04:36.01 | Ralith | oh wait found it |
| 04:36.46 | Ralith | builds |
| 07:42.38 | *** join/#brlcad _clock_ (n=_sushi_@80-218-244-105.dclient.hispeed.ch) | |
| 07:48.47 | *** join/#brlcad hippieindamakin8 (n=hippiein@210.212.55.3) | |
| 10:02.09 | *** join/#brlcad Axman6 (n=Axman6@210.11.145.150) | |
| 10:07.29 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) | |
| 11:52.08 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-27.sbndin.btas.verizon.net) | |
| 13:24.38 | *** join/#brlcad samrose (n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net) | |
| 13:34.52 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) | |
| 14:05.08 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-27.sbndin.btas.verizon.net) | |
| 14:16.09 | *** join/#brlcad BigAToo1 (n=BigAToo@pool-96-230-124-27.sbndin.btas.verizon.net) | |
| 14:35.08 | *** join/#brlcad mafm (n=mafm@74.Red-83-42-152.dynamicIP.rima-tde.net) | |
| 15:28.02 | CIA-38 | BRL-CAD: 03starseeker * r35252 10/brlcad/trunk/src/librt/opennurbs_ext.cpp: Tweak flatness test comment. |
| 16:47.18 | *** join/#brlcad mafm (n=mafm@74.Red-83-42-152.dynamicIP.rima-tde.net) | |
| 16:48.52 | *** join/#brlcad mafm (n=mafm@74.Red-83-42-152.dynamicIP.rima-tde.net) | |
| 17:06.55 | *** join/#brlcad mafm (n=mafm@74.Red-83-42-152.dynamicIP.rima-tde.net) | |
| 18:03.26 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-27.sbndin.btas.verizon.net) | |
| 18:32.02 | *** join/#brlcad jdoliner (n=jdoliner@c-68-51-75-169.hsd1.il.comcast.net) | |
| 19:26.19 | *** join/#brlcad elena (n=elena@89.136.118.141) | |
| 19:39.56 | *** join/#brlcad hippieindamakin8 (n=hippiein@210.212.55.3) | |
| 20:04.26 | *** join/#brlcad samrose (n=samrose@adsl-76-252-64-240.dsl.lgtpmi.sbcglobal.net) | |
| 20:04.39 | CIA-38 | BRL-CAD: 03irpguardian * r35253 10/brlcad/trunk/src/proc-db/human.c: Reworked how bounding boxes are made, but are still inaccurate. |
| 20:17.07 | *** join/#brlcad mafm (n=mafm@74.Red-83-42-152.dynamicIP.rima-tde.net) | |
| 20:33.43 | *** join/#brlcad elena (n=elena@89.136.118.141) | |
| 21:27.49 | *** join/#brlcad roberthl (n=robert@rhl.me.uk) | |
| 23:10.48 | Ralith | so what's with all these "X will be initialized after Y" warnings g++ is giving me from ctor initialization lists? Why is that warning-worthy? |
| 23:11.44 | Ralith | starseeker: also, latest ogre trunk works fine here. |
| 23:11.53 | Ralith | with freeimage disabled, even. |
| 23:11.55 | Ralith | perhaps something went amiss in your build. |