IRC log for #brlcad on 20110914

00:04.01 *** join/#brlcad brlcad (~sean@BZ.BZFLAG.BZ)
00:04.49 brlcad mm e-mail troll
00:35.44 *** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
00:51.23 abhi2011 brlcad , is there a way to change the rt resolution from the default 512 by 512
00:56.24 abhi2011 ok got it
00:56.28 abhi2011 the -s option
01:00.14 abhi2011 hmm trying with -s1024, seems like have to leave the rendering job overnight :P
01:10.42 CIA-133 BRL-CAD: 03abhi2011 * r46679 10/brlcad/trunk/src/libged/simulate/ (simphysics.cpp simulate.c simulate.h): Added code to simulate command for showing bounding boxes and rigid body activation state
01:24.01 CIA-133 BRL-CAD: 03starseeker * r46680 10/brlcad/trunk/ (6 files in 4 dirs): More distcheck progress, although there's still at least one bug to iron out.
03:17.59 brlcad abhi2011: the number of pixels increases quadratically as you increase the image size
03:18.27 brlcad so going from 512x512 to 1024x1024 is a 4x increase in the number of pixels being rendered
03:21.08 brlcad usually better to get your scene set up exactly how you want it at 256x256 or similar without materials for a quick preview, then render high at some much higher resolution (I tend to like 4096x4096 with jitter and hypersampling) with materials and custom light sources
03:21.37 brlcad but then those usually beg for some quick distributed rendering with lots of computers :)
03:27.44 *** join/#brlcad louipc (~louipc@archlinux/trusteduser/louipc)
05:18.45 *** join/#brlcad ibot (~ibot@rikers.org)
05:18.45 *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.20.2 is posted (20110701) || BRL-CAD is participating in the ESA Summer of Code in Space!
11:38.52 CIA-133 BRL-CAD: 03starseeker * r46681 10/brlcad/trunk/src/libged/simulate/simulate.c: comment out rv until it's actually used, breaks strict building as the code currently stands.
12:00.34 CIA-133 BRL-CAD: 03starseeker * r46682 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/BRLCAD_Util.cmake): fix paths that are appended to cmakefiles.cmake and cmakedirs.cmake
12:25.03 *** join/#brlcad abhi2011 (~chatzilla@ip170-79-211-87.adsl2.static.versatel.nl)
12:56.07 CIA-133 BRL-CAD: 03tbrowder2 * r46683 10/brlcad/trunk/doc/docbook/Makefile.am: correct XSLTPROC options; add new document to tree
13:17.26 CIA-133 BRL-CAD: 03starseeker * r46684 10/brlcad/trunk/ (28 files in 5 dirs): (log message trimmed)
13:17.26 CIA-133 BRL-CAD: Variety of distcheck fixes, including a nifty problem where a directory named
13:17.26 CIA-133 BRL-CAD: 'off' was causing a logic failure in the path building logic. dist files are
13:17.26 CIA-133 BRL-CAD: now actual cmake files that are included - this allows CMake to re-run after one
13:17.26 CIA-133 BRL-CAD: is edited, which is the correct behavior. directories that are specified (as
13:17.27 CIA-133 BRL-CAD: opposed to recognized as parents of files specfied) use svn info to build the
13:17.28 CIA-133 BRL-CAD: list of files below the ignored directory, which means temporary files should be
13:41.14 CIA-133 BRL-CAD: 03tbrowder2 * r46685 10/brlcad/trunk/doc/docbook/presentations/ (13 files in 3 dirs): new document added to tree
14:08.24 CIA-133 BRL-CAD: 03brlcad * r46686 10/brlcad/trunk/doc/docbook/Makefile.am: fully sort the list
14:08.56 CIA-133 BRL-CAD: 03Tbrowder 07http://brlcad.org * r3158 10/wiki/FAQ: add a new bit of trivia from the old-timers
14:17.39 CIA-133 BRL-CAD: 03starseeker * r46687 10/brlcad/trunk/misc/CMake/distcheck_buildsys.cmake.in:
14:17.39 CIA-133 BRL-CAD: OK. First rule if building a dist tarball from a tarball or other non-svn
14:17.39 CIA-133 BRL-CAD: checkout source, don't. CPack has no way of knowing what to exclude in the
14:17.39 CIA-133 BRL-CAD: fully ignored sub-directories. If you must, distcheck will do what it can where
14:17.39 CIA-133 BRL-CAD: it can to spot files that don't belong, and warn you of your peril.
14:20.59 CIA-133 BRL-CAD: 03Tbrowder 07http://brlcad.org * r3159 10/wiki/SVN: add link to some more SVN info
14:27.42 CIA-133 BRL-CAD: 03Tbrowder 07http://brlcad.org * r3160 10/wiki/SVN:
14:29.52 CIA-133 BRL-CAD: 03Tbrowder 07http://brlcad.org * r3161 10/wiki/SVN: add link to more svn info
14:50.24 CIA-133 BRL-CAD: 03starseeker * r46688 10/brlcad/trunk/misc/CMake/distcheck_buildsys.cmake.in:
14:50.24 CIA-133 BRL-CAD: Be nice to the users, if the ignored file count gets too large refer them to a
14:50.24 CIA-133 BRL-CAD: file instead of dumping thousands of lines to the terminal. Need to do
14:50.24 CIA-133 BRL-CAD: something more intelligent for in-src distcheck building, which is what prompted
14:50.24 CIA-133 BRL-CAD: the change, but might as well handle this for cases with lots of stray files
14:50.25 CIA-133 BRL-CAD: too.
14:56.06 *** join/#brlcad n_reed (~molto_cre@BZ.BZFLAG.BZ)
15:07.45 CIA-133 BRL-CAD: 03Abhi2011 07http://brlcad.org * r3162 10/wiki/User:Abhijit: /* Log */
15:16.21 CIA-133 BRL-CAD: 03Abhi2011 07http://brlcad.org * r3163 10/wiki/User:Abhijit: /* Development timeline */
15:32.20 CIA-133 BRL-CAD: 03tbrowder2 * r46689 10/brlcad/trunk/doc/docbook/Makefile.am: more file name sorting
16:16.37 brlcad interesting, http://www.cs.washington.edu/research/constraints/cassowary/
16:32.02 brlcad secondary lead if spirit work ever hits a roadblock
16:51.07 CIA-133 BRL-CAD: 03tbrowder2 * r46690 10/brlcad/trunk/doc/docbook/README: added blub on new presentations directory
16:52.50 CIA-133 BRL-CAD: 03tbrowder2 * r46691 10/brlcad/trunk/doc/docbook/ (Makefile.am system/man1/en/Makefile.am): renamed build variable to better track file tree name
16:53.13 *** join/#brlcad Stattrav (~Stattrav@61.12.114.82)
16:53.13 *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
16:55.36 starseeker brlcad: does it do floating point?
16:56.45 starseeker ah, I think I've run across that before
16:57.29 starseeker that also needs GTL: http://www.fim.uni-passau.de/en/fim/faculty/chairs/theoretische-informatik/projects.html
16:58.53 starseeker O.o am I seeing that right, Apple is using it?
17:00.48 starseeker perhaps I underestimated it
17:05.19 starseeker looks like it also wants guile
17:29.58 brlcad nods
17:30.29 brlcad it's also dead/unsupported along with GTL, but an interesting lib nonetheless
17:31.24 brlcad i'd be a little surprised if guile was actually required
17:31.30 brlcad there's lots of bindings to various languages, I suspect that's where it's used
17:35.04 starseeker nods - yeah, most of it looks like demos
17:35.21 starseeker probably not too hard to get the interesting subset building
17:35.57 starseeker nifty: http://www.badros.com/greg/cassowary/js/quaddemo.html
17:36.55 starseeker is quite intrigued by Apple's use
17:43.53 starseeker doesn't see any code from Apple on the cassowary site... wonder if they just used it as-is?
17:44.02 starseeker would be surprising for code that old
17:46.34 starseeker brlcad: was that the academic library you were thinking of?
17:47.26 starseeker vaguely remembers getting as far as determining it needed GTL and that GTL still exists before, but I didn't follow up since gecode looked more modern and maintained
17:48.20 starseeker wasn't fully aware of gecode's lack of floating point solver support at the time
18:00.26 brlcad starseeker: it might have been, don't remember for certain
18:01.14 starseeker will play with it later - struggling to convince CPack not to include build files in its tarballs
18:01.18 brlcad still, greener pastures and all -- it's just a good thought to keep in mind's back pocket, maybe wire up to the libpc test cases to see how it does
18:01.51 starseeker may have to at least require a subdirectory within the source tree - the source=build case is a nightmare
18:02.12 brlcad nightmare in what sense?
18:02.37 starseeker I have to tell CPack to exclude files, but the simple act of RUNNING cpack within the build tree creates more files
18:03.07 starseeker which I can't list for CPack because they don't exist ahead of time
18:03.22 brlcad you mean for distcheck?
18:03.24 starseeker yeah
18:04.05 brlcad if it's *only* for distchecking, not a big deal -- or even let cpack pack them, it'll just warn on them down the line
18:04.33 starseeker as long as it's smart enough to not be recursive, I guess...
18:04.34 brlcad in practice for proper dist building, it can be out of dir no worries
18:05.11 brlcad in source build support is really only needed to support simple compilation/install
18:05.30 brlcad distchecking is "in-house stuff, in-house rules"
18:05.54 starseeker I thought one of the points though was to verify the tarball had everything we wanted it to
18:06.01 brlcad sure
18:06.13 starseeker (admittedly that works a little different now, with that verification happening up front, sorta...)
18:06.35 brlcad so it has too much if someone built in-dir and ran distcheck, but it could warn and still run all the tests
18:07.03 starseeker nods... true
18:07.28 brlcad checking the tarball is only half the picture, and the lesser-utilized half at that
18:07.41 brlcad the other half of the distcheck is all the testing
18:07.54 brlcad that is desirable to run and re-run as often as possible
18:08.16 brlcad so is distcheck ready to go now?
18:08.26 brlcad I'm going through commits now to see if everything is documented
18:08.43 starseeker close, but I wouldn't consider it properly hooked in yet
18:10.31 CIA-133 BRL-CAD: 03brlcad * r46692 10/brlcad/trunk/NEWS: tom converted traNese's intro to Tcl/Tk presentation into the docbook format. that means it'll get installed in html form (and potentially as pdf).
18:10.42 brlcad close enough for me to distcheck a source release tarball? :)
18:10.55 starseeker urm. from out of dir build, maybe
18:11.13 brlcad was hoping for better than maybe :D
18:11.32 starseeker brlcad: I've totally rewritten it twice in the last two days :-P
18:11.36 *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol)
18:11.38 starseeker it's hard to be definitive just yet
18:11.41 brlcad I know, that's why I'm asking
18:12.47 brlcad abhi2011: how's that render coming along?
18:13.06 brlcad wants to set up his own scene here real soon now...
18:24.29 jordisayol brlcad: I see that "sh/make-deb.sh" was modified by tbrowder2
18:25.28 jordisayol brlcad: is there some problem with my maintenance?
18:25.44 piksi are there screenshots or feature lists of the current archer ui?
18:26.29 CIA-133 BRL-CAD: 03brlcad * r46693 10/brlcad/trunk/src/librt/primitives/nmg/nmg_tri.c: if the code added to quiet compiler warnings has to also be quieted (with #if 0), then it should just be removed...
18:30.18 starseeker piksi: this is fairly current: http://bzflag.bz/~starseeker/archer_win32_native_built_docs.png
18:30.58 starseeker this is still fairly close appearance wise: http://bzflag.bz/~starseeker/archer_latest.png
18:31.41 brlcad jordisayol: not at all, at least not that I'm aware of -- were the changes significant/controversial?
18:31.55 brlcad I think he was just trying to help
18:32.29 brlcad tom's a relatively new committer, communicates via mailing list
18:32.49 jordisayol brlcad: I'm just checking, it appear that only added a "test dependency only" ability
18:33.55 jordisayol brlcad: is just to know if I have to continue to build new deb/rpm packages
18:33.59 piksi starseeker: thanks. last time i tried modeling something with MGED i found it extremely cumbersome compared to e.g. current versions of autocad or autodesk revit. archer ui looks pretty much like autocad circa 2004. what kind of an ui philosophy is being used with archer?
18:34.12 brlcad jordisayol: oh, please! :)
18:34.46 brlcad you don't "have to" but it's greatly helpful and appreciated (as seen by the loads and loads of downloads)
18:35.13 starseeker piksi: um. "make it not suck?" not sure what you mean by a ui philosophy
18:35.30 piksi jordisayol is responsible for rpm packaging?
18:35.40 jordisayol brlcad: no no, it's ok. i just want to know
18:35.45 piksi then i have to thank you, i was pleasantly surprised to find a fedora package for my system :-)
18:36.12 jordisayol piksi: i do my best, and yes i buils rpm packages
18:36.18 jordisayol build*
18:36.45 jordisayol piksi: is there some problem with them?
18:36.54 brlcad piksi: my goal is towards a completely modeless redesigned ui but we're not going to get there in one jump
18:37.18 piksi jordisayol: no, i was just very happy that there was an rpm and i didn't have to compile :-)
18:37.33 jordisayol piksi: ;-)
18:37.36 piksi starseeker: ui philosophy, i.e. all tools working in a consistent manner, designing the tools from the viewpoint of a designer (for example having good 3d snapping capabilities, versatile extruding/lofting/revolving tools etc)
18:37.50 brlcad archer is already a huge stepping stone as it is, even if it just brings us into the 90's ;) .. to do more would increase risk of delays (or outright failure)
18:39.26 piksi brlcad: if by mode you mean the command line type interface of autocad, i find it extremely fast compared to many of the recent autodesk ui changes that move towards clicking (without being "in a command state")
18:39.43 brlcad a lot of the nifty UI features you mention require extensive backend support, too -- which has little if anything to do with the overarching ui design
18:39.57 jordisayol brlcad: as a main developer, can You tell tbrowder2 to communicate me just ins deb/rpm modifications? just for a better building result
18:40.22 brlcad piksi: that is not what is meant by mode, http://en.wikipedia.org/wiki/Mode_(computer_interface)
18:40.25 piksi brlcad: yes, i understand that the ui features i'm longing for require an *extensive* framework which is not easy to do at all
18:41.11 piksi brlcad: ah now i understood. yes, that is a good thing to avoid
18:41.17 brlcad jordisayol: hm? didn't understand the "just ins"
18:42.02 brlcad jordisayol: if you see something undesirable, you can revert it from the deb/rpm files -- that merely begs a dev-dev discussion
18:42.04 jordisayol brlcad: sorry, is a typo. just for files directly involved in deb/rpm building packages
18:42.37 piksi brlcad: is any kind of an input/suggestions/feature requests for the upcoming ui features needed or accepted from users? my background is in architecture but i've used brl from time to time to model csg-stuff instead of autocad
18:43.01 jordisayol well, I've modified "sh/make_debsh" to disable source build, and now i find this modification, that's all
18:43.05 brlcad fwiw, a commit effectively is a means of communication -- and one that's guaranteed to reach everyone
18:43.22 brlcad otherwise it's a bit tricky with him on the mailing list and you on irc -- the discussions shouldn't be in private
18:44.23 starseeker brlcad: yep, though so - new docbook stuff sets off cmake distcheck
18:44.26 starseeker one sec...
18:45.28 jordisayol brlcad: ok
18:47.00 brlcad piksi: input/suggestions/feature requests are always welcome but with heeded caution that we've probably heard/thought of it too (we have a TODO list that probably represents a 1000-staff years of effort and that still wouldn't get us all the features of the "big five")
18:47.11 piksi :-)
18:47.44 piksi brlcad: yeah i wasn't expecting to introduce anything groundbreaking in terms of design. is the todo list public?
18:47.55 brlcad piksi: more constructive is the suggestions that have real productive effort tied to them -- like coming up with a detailed use case or a real prototype mock up
18:48.07 piksi yeah i always try to provide mockups
18:49.25 brlcad even better is to relate it to the current state of the software, like looking at MGED or Archer's horrid menu hierarchy and detailing exactly what a new/better hierarchy would look like, etc
18:50.09 brlcad as for the 3rd generation interface, a good bit of mock-up design effort has already happened (in a non-CAD/agnostic way to help with the underlying usability concepts)
18:51.58 brlcad one prototype video is here: http://brlcad.org/design/gui/
18:52.17 CIA-133 BRL-CAD: 03starseeker * r46694 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/distcheck_buildsys.cmake.in):
18:52.17 CIA-133 BRL-CAD: Deduce the in-source build directory rather than hardcode one devs typical
18:52.17 CIA-133 BRL-CAD: defaults. Have the distcheck routine use the variable as well to clean up its
18:52.17 CIA-133 BRL-CAD: output. Looks like this is helpful, but not clear yet if it is a full solution.
18:53.03 brlcad jordisayol: to reiterate, if he modified one of the scripts your using and it affects your ability to prepare a release, there is ABSOLUTELY no problem reverting that change
18:53.30 brlcad if it doesn't affect what you're doing or can be accommodated, then great .. just ignore it or accommodate or ask him to make it optional or whatever works ;)
18:53.52 brlcad if you reply to the commit e-mail, it'll go to the brlcad-devel mailing list and reach him
18:54.26 jordisayol brlcad: well,the only two files out of /misc/debian dir that I modified are make_deb.sh and make_rpm.sh
18:54.42 jordisayol brlcad: and he modified make_deb.sh
18:55.01 starseeker jordisayol: the changes conflict?
18:55.33 jordisayol is not a bit modification, just added a new option to just test is dependencies are present in system and exit
18:55.51 jordisayol is => if
18:56.14 brlcad modifications are to be expected anywhere in the source tree, part of collaborative development
18:56.42 brlcad what's really cool is having three devs all working in the same file at the same time, frequently updating, and just seeing the code flow :)
18:57.00 brlcad i've only seen it a couple times in my career, but it's a thing of beauty :D
18:57.29 brlcad hyperproductivity!
18:58.19 CIA-133 BRL-CAD: 03starseeker * r46695 10/brlcad/trunk/doc/docbook/ (3 files in 3 dirs): Fix up the presentations CMakeLists.txt logic
18:59.04 jordisayol brlcad: I apologize, I was surprised for this modification, that's all
18:59.28 starseeker jordisayol: generally, that's a good thing - it means people are taking an interest in your work :-)
19:00.26 jordisayol brlcad: yes, shure, but it can means to that i'm not doing it so good too :-)
19:01.54 brlcad it didn't even occur to me to think that it meant you were doing something so good :)
19:02.59 brlcad what starseeker said, just someone else taking an interest in the scripts, in what you'd done .. that's usually a "really good" thing
19:03.57 jordisayol brlcad: You and starseeker have a couple of beers payed ;-)
19:05.37 starseeker brlcad actually went to some trouble to remove individual authorship entries from the various source files and instead record them in the AUTHORS file - we don't want to wind up with a situation where people don't want to touch code because "it's somebody else's"
19:06.53 starseeker so no worries - we're all trying to Make Things Better :-)
19:08.46 CIA-133 BRL-CAD: 03brlcad * r46696 10/brlcad/trunk/NEWS: bob has initial support in for editing pipe primitives within archer. support for appending, prepending, deleting, moving, selecting, and scaling.
19:13.23 CIA-133 BRL-CAD: 03starseeker * r46697 10/brlcad/trunk/CMakeLists.txt:
19:13.23 CIA-133 BRL-CAD: Be sure we don't do anything in the case where the build and the source
19:13.23 CIA-133 BRL-CAD: directories are the same directory - unless I'm missing some CPack option or
19:13.23 CIA-133 BRL-CAD: feature (possible) that case is just not going to mix well with CPack's approach
19:13.23 CIA-133 BRL-CAD: to archive creation.
19:15.33 starseeker brlcad: cmake distcheck now succeeds
19:18.05 CIA-133 BRL-CAD: 03starseeker * r46698 10/brlcad/trunk/CMakeLists.txt: fix if location
19:25.59 jordisayol brlcad: there are some errors on tbrowder2 make_deb.sh modifications. He treat some packages names as executable files names, and this will cause the script failure
19:29.56 piksi brlcad: thanks, i'll look into it
19:31.31 brlcad piksi: the TODO file in any source distribution contains several hundred items (some huge, some tiny)
19:31.48 piksi :-)
19:32.30 brlcad the documentation section at the bottom is particularly ripe for getting involved in a useful way, but that extends to (unlisted) developer documents (aforementioned mockups etc) too
20:05.31 *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol)
20:05.31 *** join/#brlcad n_reed (~molto_cre@BZ.BZFLAG.BZ)
20:05.31 *** join/#brlcad yiyus (1242712427@je.je.je)
20:05.31 *** join/#brlcad kanzure (~kanzure@131.252.130.248)
20:05.31 *** join/#brlcad piksi (piksi@pi-xi.net)
20:05.31 *** join/#brlcad louipc (~louipc@archlinux/trusteduser/louipc)
20:05.31 *** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
20:05.31 *** join/#brlcad brlcad (~sean@BZ.BZFLAG.BZ)
20:05.31 *** join/#brlcad plaes (~plaes@gn237.zone.eu)
20:05.31 *** join/#brlcad ``Erik (Here@c-69-140-109-104.hsd1.md.comcast.net)
20:05.31 *** join/#brlcad kunigami1 (~kunigami@loco-gw.ic.unicamp.br)
20:05.31 *** join/#brlcad dtidrow (~dtidrow@c-68-84-167-135.hsd1.mi.comcast.net)
20:05.31 *** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net)
20:05.31 *** join/#brlcad starseeker (~starseeke@BZ.BZFLAG.BZ)
20:05.31 *** join/#brlcad CIA-133 (~CIA@cia.atheme.org)
20:19.05 *** join/#brlcad kanzure (~kanzure@131.252.130.248)
20:45.23 CIA-133 BRL-CAD: 03starseeker * r46699 10/brlcad/trunk/ (4 files in 2 dirs): Make a stab at consolidation of all distcheck logic into one file.
21:12.00 CIA-133 BRL-CAD: 03starseeker * r46700 10/brlcad/trunk/ (4 files in 2 dirs): Hmm. $(MAKE) isn't working as expected in EXECUTE_PROCESS - revert the one file solution.
21:14.18 starseeker oh, duh - right, $(MAKE) is specific to make files
21:21.17 CIA-133 BRL-CAD: 03starseeker * r46701 10/brlcad/trunk/CMakeLists.txt: Fix numbers on distcheck stages.
21:24.35 CIA-133 BRL-CAD: 03starseeker * r46702 10/brlcad/trunk/CMakeLists.txt: try searching for cpack rather than hardcoding it.
21:31.40 *** join/#brlcad abhi2011 (~chatzilla@ip170-79-211-87.adsl2.static.versatel.nl)
21:32.09 CIA-133 BRL-CAD: 03starseeker * r46703 10/brlcad/trunk/ (3 files in 2 dirs): completed distcheck does not necessarily imply distribution ready files, so don't assert that. No need for distcheck message file, and not using the old svncheck anymore.
22:20.42 *** part/#brlcad n_reed (~molto_cre@BZ.BZFLAG.BZ)
22:57.26 CIA-133 BRL-CAD: 03tbrowder2 * r46704 10/brlcad/trunk/doc/docbook/resources/docbook-5.0/ (36 files in 8 dirs): add DocBook schemas for validation
22:58.30 CIA-133 BRL-CAD: 03tbrowder2 * r46705 10/brlcad/trunk/doc/docbook/Makefile.am: add make target for DocBook validation
23:18.43 CIA-133 BRL-CAD: 03jordisayol * r46706 10/brlcad/trunk/sh/make_deb.sh: Correct some minor erros.
23:58.28 CIA-133 BRL-CAD: 03r_weiss * r46707 10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: Updated function nmg_isect_two_face2p_jra within file nmg_inter.c. Improved the logic for determining where edges should be cut.

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.