IRC log for #brlcad on 20070605

00:15.34 *** join/#brlcad docelic (n=docelic@212.15.174.26)
01:49.26 *** join/#brlcad louipc_ (n=louipc@bas8-toronto63-1177706227.dsl.bell.ca)
02:51.55 *** join/#brlcad digitalfredy (n=digitalf@200.71.62.161)
04:05.01 *** join/#brlcad RodGallowGlass (n=mario_du@bas2-sudbury98-1177871953.dsl.bell.ca)
04:06.04 RodGallowGlass http://irix32.spaces.live.com/photos <----- why can't I get 7-10 to do this
04:06.28 RodGallowGlass windowsside album
04:07.27 RodGallowGlass framebuffer came up sweetly, had four open at once
04:10.40 RodGallowGlass hah faceplate gui now works, i wont post another shot for now
04:11.24 RodGallowGlass may I change the version number on my copy of 7.8.4 or is that prohibited?
04:16.13 brlcad it's open source, you can do wahtever you want with it... just what's the point?
04:16.46 brlcad it doesn't provide any benefit
04:17.54 RodGallowGlass to distinguish what im doing from whats on sourceforge, or does the stuff on sourceforge dissapear when a new release comes out
04:19.44 RodGallowGlass btw what are you doing still up :)
04:21.20 RodGallowGlass you worry that I'll spread it around?
04:21.56 RodGallowGlass it's not going anyfurther than me and you unless *you spread it around
04:22.26 RodGallowGlass but im trying to convert to 7.10
04:30.18 yukonbob brlcad: 1 quick q -- when rendering large DSPs, (like that pac. north west stuff we were talking about) are gob & gobs of RAM typically required? I've got 1G RAM, but am dying w/ insufficient memory... can I tune rt? (I've even tried "-j 10,10,50,50", but still run out of memory) btw, I've got 4G swap...
04:32.13 brlcad it is an in-core algorithm, so it has to be able to represent a 3D version of the data in memory (which is some multiplier of the input datasize that escapes me at the moment, but is larger of course)
04:32.47 brlcad the acceleration structure it uses to progress fast through the dsp also takes up some memory space
04:33.22 brlcad how big is the data set?
04:33.27 yukonbob out of curiousity, what are the specs of the machine you used to build pugeot sound?
04:33.50 yukonbob I'm using the smallest of that Pacific Northwest set...
04:34.10 brlcad the size of the image being rendered doesn't matter much at all, it's the dsp data itself that will likely dominate
04:34.17 yukonbob 37G
04:34.32 yukonbob Megs, rather :)
04:34.42 brlcad hehe, i was going to say...
04:36.09 brlcad you mean the puget sound dataset?
04:37.50 yukonbob oh no -- I'm talking about the whole region -- the specfic name I'm using is...disappointment
04:38.19 brlcad i mean this is data you have then from some other source?
04:39.07 brlcad more importantly, when you create the dsp, does it show the wireframe in mged?
04:39.08 yukonbob "disappoint" rather -- for Cape Disappointment -- it's from http://duff.geology.washington.edu/data/raster/tenmeter/onebytwo10/
04:39.20 yukonbob wireframe looks like I imagine it should...
04:40.19 yukonbob 1542x12018 data points
04:40.25 brlcad can you post up the .g somewhere so I can look at it? (can upload via anon ftp to ftp.brlcad.org/incoming)
04:41.03 brlcad hm, that is indeed just 37MB
04:45.45 yukonbob brlcad: can you unzip bzip2 compressed-files?
04:47.29 yukonbob "disappointment.dsp.bz2" incoming...
04:52.00 PrezKennedy i want FiOS!!
04:52.17 *** join/#brlcad cadguy (n=cadguy@trapezewireless121.insecure.utah.edu)
04:56.03 yukonbob brlcad: disappointment.dsp.bz2 is uploaded --- I renamed the extention .dsp for my own identification, but w/i the .zip archive, it was the .bil file...
05:03.28 brlcad looking
05:06.41 yukonbob ...then I just open new db, and "in testmap dsp" file, [filename], 1542, 12018, 0, ad, 1, 1
05:07.31 yukonbob sometimes I'll even get a malloc issue at this point, but if I restart gmed, the 'testmap' is there, and seems usable.
05:08.53 brlcad k
05:09.05 yukonbob wireframe looks "reasonable" and is zoom-able/rotate-able, etc., but a simple "rt -o map.pix" chokes for me:
05:09.30 yukonbob Additional mem=8192., #malloc=34, #free=2, #realloc=1 (32 retained) 0 malloc1738539264 dsp_bb array
05:09.33 yukonbob bu_malloc: Insufficient memory available, sbrk(0)=xc745000
05:09.36 yukonbob bu_malloc: malloc failure
05:09.38 yukonbob Raytrace complete.
05:09.58 brlcad trying to allocate 1.7GB
05:10.50 yukonbob ... I guess I'm just shy, and that's that for that, then :P ?
05:12.51 yukonbob otherwise, I can strip data --- ie: remove every other column in odd rows, and skip even rows to "take strips" out of data and interpolate down...
05:14.20 brlcad i'm not convinced there's not something else going on, a bug that was introduced perhaps
05:14.57 yukonbob I'm running 7.8, btw
05:15.55 yukonbob out of curiousity, what are the specs of the machine that you rendered Pugeot Sound with?
05:15.56 brlcad have to rebuild here, so it'll be a little while before I can test
05:16.00 yukonbob np
05:18.24 brlcad iirc, I rendered that on a dual G4 a couple years ago, 2GB ram
05:18.57 yukonbob ok -- and that dataset is substantially larger than "disappointment"...
05:19.39 brlcad yeah, using the 16kx16k dataset .. about 500MB raw data
05:22.59 *** join/#brlcad bjorkBSD (n=bjork@ip70-178-214-102.ks.ks.cox.net)
05:23.24 *** join/#brlcad digitalfredy (n=digitalf@200.71.62.161)
05:39.52 yukonbob brlcad: fyi -- I downsampled the data (stipped every second row, every second column) to make dataset 0.25 of previous size, and still fails w/ malloc...
05:42.08 yukonbob ...and occasionally fails w/ malloc on creating the dsp, and crashes... but enough bug reports for tonight -- good evening :)
05:50.06 brlcad yukonbob: thanks for the reports.. i'll have to call it a night as well and pick up on this tomorrow -- it's highly suspect that you are just up against some bug
05:52.00 brlcad i'm curious how you parsed dem data so quickly too... folks have begged for a dem to dsp converter for quite a while
06:45.19 *** join/#brlcad Elperion (n=Bary@p54876DBE.dip.t-dialin.net)
06:46.32 *** join/#brlcad akreal (n=ak@ll-81-222-164-251.awanti.ru)
06:50.55 *** join/#brlcad elite01 (n=elite01@dslc-082-082-066-123.pools.arcor-ip.net)
07:19.24 *** join/#brlcad Laniakea (n=clock@zux221-122-143.adsl.green.ch)
07:58.37 brlcad yukonbob: well the good news is that I can reproduce the problem with that data set, it goes into a frenzy getting stuck in an allocation loop of some sort during prep
07:58.58 brlcad shouldn't be too hard to pin down since I can reproduce it .. let you know what I find out
08:01.47 CIA-4 BRL-CAD: 03brlcad * 10brlcad/BUGS: dsp primitive is going into what seems to be an unbounded memory allocation loop inside dsp_layers during prep, reported by yukonbob (thx)
10:31.34 *** join/#brlcad ak__ (n=ak@ll-81-222-164-251.awanti.ru)
11:04.53 *** join/#brlcad docelic (n=docelic@212.15.179.250)
11:59.53 *** join/#brlcad deltazap (n=deltazap@pool-71-251-104-132.tampfl.fios.verizon.net)
11:59.59 *** join/#brlcad bjorkBSD (n=bjork@ip70-178-214-102.ks.ks.cox.net)
12:01.10 *** join/#brlcad docelic (n=docelic@212.15.179.250)
12:07.06 tarzeau brlcad: what is the status on modification and redistribution of the Documentation at http://brlcad.com/ ?
12:07.21 tarzeau brlcad: is it possible to allow for change/redistribution in the documents? and the sources?
12:13.56 *** join/#brlcad elite01_ (n=elite01@dslb-088-070-015-175.pools.arcor-ip.net)
12:32.39 *** join/#brlcad Elperion (n=Bary@p54876DBE.dip.t-dialin.net)
12:36.41 *** part/#brlcad digitalfredy (n=digitalf@200.71.62.161)
12:37.39 Laniakea tarzeau: hi
12:37.48 tarzeau hi Laniakea
12:37.57 tarzeau karel!
12:38.09 tarzeau Laniakea: how are you able to build brlcad on linux?
12:38.29 tarzeau Laniakea: and do you know www.sauerbraten.org ?
12:38.43 Laniakea tarzeau: just compile from the sources
12:39.07 Laniakea tarzeau: no I didn't know sauerbraten
12:39.10 Laniakea is it free software?
12:39.17 tarzeau yes it is
12:39.40 tarzeau i get this error when i compile from the source: .libs/itclStubLib.o: In function `Itcl_InitStubs':
12:39.43 tarzeau /var/www/debian/brl-cad/brlcad-0+20070603/src/other/incrTcl/itcl/generic/itclStubLib.c:71: undefined reference to `tclStubsPtr'
12:40.41 Laniakea oh freeware, sucks
12:40.42 Laniakea tarzeau: actually the latest brl-cad doesn't compile on my openbsd
12:40.49 Laniakea so I use the one before :)
12:40.56 tarzeau which version is that?
12:41.03 tarzeau why does it not compile for you, same reason?
12:41.18 tarzeau do you use special configure parameters?
12:41.53 Laniakea tarzeau: the working one is 7.8.4
12:42.02 Laniakea I don't remember the nonworking one and don't remember the error message
12:44.51 tarzeau Laniakea: k, i try that
12:45.08 tarzeau Laniakea: any links2 news? and will you play or model stuff in sauerbraten?
12:45.25 Laniakea tarzeau: no, since sauerbraten is not free software
12:45.30 Laniakea we fixed some serious bugs in links
12:50.12 elite01_ Laniakea, the engine is free
12:50.53 tarzeau elite01_: i'm working on free maps,textures,skyboxes
12:51.01 tarzeau elite01_: gonna find someone to make free music/sound effects
12:51.24 tarzeau Laniakea: wonder what those serious bugs are...
12:51.39 tarzeau Laniakea: any plan to stop development of version 0.x/1.x of links?
12:56.45 Laniakea I don't know about any plan
13:00.30 ``Erik what's the error with 7.10.0 on obsd?
13:00.52 tarzeau i might've been missing libraries, retrying
13:01.21 ``Erik heh, damn, I looked at sauerbraten and went "wow, that looks like cube, did cube change its name?"
13:01.33 ``Erik is lee (eihrul/phinode) still working on it?
13:01.33 tarzeau ``Erik: yes :)
13:01.44 tarzeau ``Erik: it's cube 2, yeah they do, on quakenet #sauerbraten
13:02.18 ``Erik when he was going to cmu
13:02.22 tarzeau heh
13:02.26 ``Erik small world
13:02.30 tarzeau indeed
13:02.57 tarzeau did cube also head the built-in editor?
13:03.13 ``Erik huh?
13:03.25 tarzeau sauerbraten has a built-in editor
13:03.30 tarzeau i see
13:04.28 ``Erik um, my interests were more science/engineering visualization, internals (working on ogl drivers, etc) and I flirted a bit with making a space combat game similar to the old xwing and wing commander games
13:04.38 ``Erik "another fps" wasn't really on my radar
13:04.44 tarzeau i see
13:04.50 tarzeau it's also an RPG
13:04.54 tarzeau www.eisenstern.com :)
13:05.13 ``Erik heh, all my free time is blown on WoW these days :( *cry*
13:45.23 *** join/#brlcad elite01 (n=elite01@dslb-088-070-015-175.pools.arcor-ip.net)
14:50.00 *** join/#brlcad louipc_ (n=louipc@bas8-toronto63-1096667846.dsl.bell.ca)
15:54.23 *** join/#brlcad thing0 (i=thing0@203-59-179-212.dyn.iinet.net.au)
15:54.42 thing0 hey everyone
15:54.42 thing0 what's up?
15:56.36 elite01 thing0, the ceiling *hahahalolololroflmao*
16:04.28 thing0 lol
16:04.36 thing0 funny elite01 funny...
16:04.39 thing0 ;)
16:04.58 elite01 absolutely
16:05.34 elite01 normally, i'm not like that, but this is the low mark of the week
16:25.49 ``Erik uh huh, rrriiiigggghhhhhttttt
16:25.50 elite01 o rly!
16:25.50 ``Erik damnit
16:25.50 ``Erik just hAD to remind me
16:27.36 ``Erik http://icanhascheezburger.files.wordpress.com/2007/06/lolcode_r.jpg
16:27.37 ``Erik awesome
16:58.08 *** join/#brlcad docelic (n=docelic@212.91.117.113)
17:56.32 brlcad tarzeau: yes, change/redistribution is allowed -- some are public domain, some are BSD licensed
17:57.03 tarzeau brlcad: is that written somewhere?
17:57.12 tarzeau i can't find any info it's public domain or bsd licensed
17:57.21 tarzeau hey docelic
17:57.57 brlcad tarzeau: alas, the docs aren't doc'd very well :)
17:58.11 tarzeau http://bp1.blogger.com/_MqThGo7hFdU/RldMKz1NEJI/AAAAAAAAAEA/qCFzfluWH94/s1600-h/ohalp.jpg
17:58.18 tarzeau ``Erik: for ya
17:58.30 brlcad i can tell you as needed, it's generally a matter of whether the document was created before open sourcing or whether it's a signficant derivative
17:58.42 docelic hey hey tarzeau
17:58.52 tarzeau docelic: since where are you using brlcad? and what for?
17:58.54 brlcad pre-open source documents, like the tutorial series, are all public domain
17:59.02 docelic tarzeau: not using, just interested
17:59.22 brlcad those that are "part of the sources" and mostly those in CVS are all BDL (BSD doc license)
17:59.29 brlcad and that's stated in the COPYING file
17:59.56 tarzeau ok thanks
18:00.03 tarzeau first need to get the debian package properly done
18:00.08 tarzeau docelic: i see :)
18:01.46 jack- tarzeau: how are you handling the structure?
18:01.57 tarzeau jack-: what structure?
18:02.05 jack- like, you certainly dont want to stuff all the binaries into $prefix/bin
18:02.15 tarzeau jack-: all binaries go into /usr/bin
18:02.28 tarzeau if they are not supposed to be run from users they go to /usr/lib/brlcad/
18:02.30 jack- i let brlcad completely reside in $prefix/lib/brlcad
18:02.45 tarzeau jack-: and you got debian (or ubuntu)?
18:02.50 jack- (for fink/macosx that is..pretty debianish)
18:02.51 tarzeau jack-: currently i'm not able to build it
18:03.00 jack- yes, ubuntu, but it doesnt matter for brlcad ;)
18:03.01 tarzeau ah yeah, i looked into fink lately
18:03.05 jack- i'm a mac guy
18:03.08 tarzeau can you build it on ubuntu?
18:03.13 jack- never tried
18:03.19 tarzeau jack-: cool, me too (indirectly, nextstep/openstep). you into cocoa?
18:03.22 jack- does it break at sdl stuff?
18:03.26 jack- not at all :)
18:03.43 jack- ..if yes, disable adrt and leave it out
18:03.50 jack- not mature enough to be useful
18:03.55 jack- and doesnt like to get built
18:04.00 tarzeau no, this is the problem:
18:04.13 tarzeau .libs/itclStubLib.o: In function `Itcl_InitStubs':
18:04.13 tarzeau /var/www/debian/brl-cad/brlcad-0+20070603/src/other/incrTcl/itcl/generic/itclStucollect2: ld returned 1 exit status
18:04.33 jack- hrm
18:04.37 tarzeau .libs/itclStubLib.o: In function `Itcl_InitStubs':
18:04.37 tarzeau /var/www/debian/brl-cad/brlcad-0+20070603/src/other/incrTcl/itcl/generic/itclStubLib.c:71: undefined reference to `tclStubsPtr'
18:04.56 jack- and if you get it to build its own libtcl/libtk?
18:05.03 jack- can be done with configure options
18:05.09 tarzeau jack-: do i want that?
18:05.12 jack- maybe
18:05.33 jack- i'm lucky enough to have a brlcad-compatible libtcl api
18:05.39 tarzeau you shouldn't do that on debian systems.. but then
18:05.48 jack- but my libtik caused trouble too, so i just let brlcad do its own one
18:05.52 tarzeau yuck src/ is full of other projects
18:05.54 jack- problem solved, works 100%
18:05.59 jack- yes :P
18:06.00 tarzeau what's the configure options for you?
18:06.05 jack- complete UTAHRLE etc
18:06.24 jack- ./configure --prefix=$MYPREFIX --otheroptions
18:06.25 tarzeau maybe it's a start to get ugly packages, then work them down
18:06.35 tarzeau i want to know your --otheroptions :)
18:06.43 jack- just do ./configure --help and read that bible..
18:06.46 jack- ok hold on a sec
18:06.47 tarzeau ok
18:06.56 tarzeau finishing to build pixie first.. but then i'll check that
18:08.00 jack- pixie rocks :) great renderer
18:08.32 jack- arent you pretty jealous sometimes when you see those 200 extra repos and 20000 packages in ubuntu..
18:08.47 tarzeau not at all
18:08.52 tarzeau i hate unofficial repos
18:09.04 tarzeau they break stuff, they are of lowest quality
18:09.09 tarzeau i like to have stuff in debian main
18:09.11 jack- i can understand that, yeah
18:09.25 tarzeau ubuntu doesn't have half as much of the software debian sid+experimental has
18:09.30 tarzeau jack-: besides i'm a packaging whore:
18:09.33 jack- oh? cool :)
18:09.38 tarzeau jack-: http://qa.debian.org/developer.php?login=gurkan@linuks.mine.nu
18:09.55 tarzeau (doing this also at work, well got the job after i was doing many packages)
18:10.00 tarzeau pretty cool actually
18:10.32 jack- :)
18:10.53 jack- http://pdb.finkproject.org/pdb/maintainer.php?maintainer=jackfink
18:10.55 jack- = me
18:11.09 jack- not less of a packaging whore i guess
18:11.12 tarzeau jack-: hehe :) and i got like 33 ports in freebsd
18:11.27 jack- but fink is fun, i discovered pretty much all the stuff myself
18:11.28 tarzeau i got a mac os x box too at work, did we meet in #fink ?
18:11.34 tarzeau i tried to make some debian2fink tool
18:11.38 jack- some things never got packaged into any dist yet :)
18:11.46 jack- might well have been :)
18:11.50 tarzeau jack-: check http://gnu.ethz.ch/fink/debian2fink.txt
18:11.56 tarzeau jack-: which? which?
18:12.00 tarzeau i'll make debian packages :)
18:12.26 jack- google for sphinx project
18:12.26 jack- some .edu site
18:12.27 tarzeau is there stats of fink packages and its maintainers? i could add them to http://io.debian.net/~tar/bugstats/
18:12.47 jack- hmm, i'm not in the core team..i'll check
18:13.00 jack- think there's no real stats system at all, yet
18:13.27 tarzeau do you know condor wisconsin ?
18:13.39 jack- the name rings a bell
18:13.44 jack- but i have no idea atm
18:13.46 tarzeau job distribution
18:15.44 jack- cksfv, evolvotron, bwm-ng, iftop, throttled are probably the rarest ones on the list
18:15.54 jack- not sure if they ever got packaged anywhere else yet
18:16.25 jack- packaging stuff is sweet, somehow
18:16.33 tarzeau oh evolvotron looks interesting
18:16.38 jack- 100% legal and still some kind of piracy ;)
18:16.41 tarzeau jack-: yeah many people get to use software
18:16.42 jack- yeah it rocks!
18:17.06 tarzeau jack-: like i found www.sauerbraten.org as a package, i got addicted to badly, since a week
18:17.07 tarzeau now to help improve the game i found misfit model 3d and packaged it right away :)
18:17.14 jack- cool :)
18:17.22 tarzeau then stumpled upon old wishlist bugreports like brlcad and pixie .. doing them as well
18:17.54 jack- once i know you better, i'll tell you what made me discover sphinx2/sphinx3 and like 5 other open source speech recognizin tools
18:18.01 jack- life rocks, somehow ;)
18:18.24 tarzeau heh k
18:18.38 tarzeau if you ever are bored check out my http://livecd.gnustep.org/
18:19.20 jack- nice :)
18:19.35 jack- like we have a guy in #fink who does an own slackintosh thing
18:19.38 jack- fun as well
18:20.02 jack- anyway, i need to jump out
18:20.05 jack- later :)
18:20.08 tarzeau jack-: you know what fink package is missing? opencubicplayer
18:20.10 tarzeau later
18:20.16 jack- opencubic?
18:20.20 tarzeau yeah music player
18:20.45 jack- better than amarok/xmms/juk/noatun/beep/etc/etc?
18:20.50 tarzeau http://cow.ethz.ch/ocp2.png
18:20.59 tarzeau http://cow.ethz.ch/ocp.png
18:21.07 tarzeau rly! kickass i swear.
18:21.24 jack- cute :)
18:21.32 jack- i'll check what it needs
18:21.47 tarzeau cu
20:12.07 *** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
20:16.43 *** join/#brlcad akreal (n=ak@87.249.56.198)
20:30.50 *** join/#brlcad ak__ (n=ak@87.249.56.199)
20:36.05 *** join/#brlcad akreal (n=ak@87.249.56.198)
20:36.05 CIA-4 BRL-CAD: 03jlowenz * 10brlcad/include/opennurbs_ext.h: debug to stdout instead of error (easier to log everything together)
20:39.15 CIA-4 BRL-CAD: 03jlowenz * 10brlcad/src/conv/iges/brlcad_brep.cpp: Add more debug statements... fix trim curve orientation issue (openNURBS trims must be properly oriented whether or not the model-space curve is) and fix isoparametric flag issue
20:41.43 CIA-4 BRL-CAD: 03jlowenz * 10brlcad/src/conv/iges/n_iges.cpp: Fix IGES string->double conversion bugs (duh)
20:41.54 *** join/#brlcad Laniakea (i=clock@217-162-231-209.dclient.hispeed.ch)
20:45.40 CIA-4 BRL-CAD: 03jlowenz * 10brlcad/src/librt/g_brep.cpp: Fix brep plot() problem with non-trivial faces (i.e. those coming from IGES), perform simplified edge rendering (still needs to be fleshed out to handle non-linear curves)
20:47.28 CIA-4 BRL-CAD: 03jlowenz * 10brlcad/src/librt/opennurbs_ext.cpp: Add some additional debug statements for testing pullback_curve behavior
20:48.14 *** join/#brlcad Elperion (n=Bary@p54876DBE.dip.t-dialin.net)
20:51.44 *** join/#brlcad Obscene_CNN (n=DiscoBan@owt-64-146-239-67.owt.com)
20:53.48 *** join/#brlcad ak__ (n=ak@87.249.56.199)
21:11.48 *** join/#brlcad akreal (n=ak@87.249.56.198)
21:16.22 *** join/#brlcad ak__ (n=ak@87.249.56.198)
21:25.34 *** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
21:25.46 *** join/#brlcad cad98 (n=96a3260a@bz.bzflag.bz)
21:25.52 *** join/#brlcad PrezKennedy (n=Matthew@c-69-255-112-97.hsd1.md.comcast.net)
21:57.32 *** join/#brlcad Obscene_CNN (n=DiscoBan@owt-64-146-239-67.owt.com)
22:53.36 *** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net)
23:33.19 *** join/#brlcad Twingy (n=justin@74.92.144.217)

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