IRC log for #brlcad on 20091110

00:13.59*** join/#brlcad Ralith (n=ralith@69.90.49.189)
03:24.08*** join/#brlcad brlcad (n=sean@BZ.BZFLAG.BZ) [NETSPLIT VICTIM]
05:11.53*** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6)
07:36.47*** join/#brlcad Don__ (n=Don@c-71-238-51-148.hsd1.mi.comcast.net)
08:28.34*** join/#brlcad Ralith (n=ralith@69.90.49.189)
11:49.29d-lo_mernin all!
11:55.17``Erikfarts in d-lo's general direction
12:03.09CIA-61BRL-CAD: 03indianlarry * r36453 10/brlcad/trunk/src/librt/opennurbs_ext.cpp: Fixed bug where 'min_dist' wasn't being set in SurfaceTree::getSurfacePoint()
12:32.13CIA-61BRL-CAD: 03indianlarry * r36454 10/brlcad/trunk/src/conv/step/ (5 files): Added openNURBS brep generation code for Toroidal and Spherical surfacs.
12:35.59CIA-61BRL-CAD: 03indianlarry * r36455 10/brlcad/trunk/src/conv/step/PullbackCurve.cpp:
12:36.00CIA-61BRL-CAD: Added checks around curve sampling counts when trying to resolve seam issues.
12:36.02CIA-61BRL-CAD: Also cleaned up some debug code.
13:25.13*** join/#brlcad R0b0t1 (n=Enigma@unaffiliated/r0b0t1)
13:55.03CIA-61BRL-CAD: 03brlcad * r36456 10/brlcad/trunk/src/ (archer/archer.bat mged/mged.bat util/rtwizard.bat): the batch file numbers didn't get bumped, bump to 7.16.2 for release
13:59.43CIA-61BRL-CAD: 03brlcad * r36457 10/brlcad/branches/STABLE/ (14 files in 6 dirs): merge trunk to STABLE from r36450 to HEAD r36456
14:05.18CIA-61BRL-CAD: 03brlcad * r36458 10/brlcad/trunk/ (6 files in 5 dirs): bump to 7.16.3 in anticipation of 7.16.4 being the next release iteration.
14:19.18*** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net)
14:23.54``Erikhm Benchmark results indicate an approximate VGR performance metric of 26678
14:30.32brlcadnot bad, on par with an oct-core mac
14:32.26``Erikoct-core xeon 3ghz, 64g ram, running rhel5
14:32.35``Erikwe have 3 now
14:32.58brlcadah, okay, so basically same config
14:33.08brlcadas the macs, just catching up
14:33.12``Erik<-- points at the ram
14:33.16``Erikslight advantage :)
14:33.43``Erikless syscall overhead, but more suckage in other aspects *shrug* just another resource
14:33.44brlcadI don't think we've pushed the 16GB barrier yet :)
14:34.32``Erikthinks the marching cubes thing will generate some very ugly meshes pre-decimation, might do the trick
14:37.55brlcadhow long does a ./configure --enable-all default compilation take?
14:45.47``Erikhm, I was building over nfs
14:59.24CIA-61BRL-CAD: 03brlcad * r36459 10/brlcad/trunk/bench/run.sh:
14:59.26CIA-61BRL-CAD: if any of the benchmark tests fail, blather loudly that the testing failed with
14:59.28CIA-61BRL-CAD: instructions to the user. this should cause the benchmark to halt compilation
14:59.30CIA-61BRL-CAD: testing (via make check or make bench) and avoids writing out 0 values into the
14:59.32CIA-61BRL-CAD: summary file.
15:08.28``Erikconfigure --enable-all, source on nfs, build on local disk: real 3m7.083s | user 12m45.142s | sys 7m25.372s
15:08.51``Eriktime ( $HOME/src/brlcad/configure --enable-all && make -sj13)
15:09.21brlcadhm, that's actually not too shabby
15:09.25``Erikwoops
15:09.32``Erik16 core machine, I only did 13
15:09.45``Eriknote that the source was all cached up from a previous build, too
15:09.49brlcada 16 core gave  27k vgr?
15:09.54``Erikyeh
15:09.59``Erikit is linux, after all
15:10.01brlcadthat don't sound right
15:10.11``Erikremakes -j17
15:10.19brlcadsure it's not 8-core with hyperthreading?
15:10.20``Erikoh, hah, I don't think I did --enable-optimized for the benchmark
15:10.25brlcadahh
15:10.29``Erikxeon cpu's
15:10.31brlcadthat'd be about right then
15:10.45``Erikkicks the tires some more
15:10.46brlcadwould expect 40-50k
15:11.47``Erikdang step and opennurbs, jacking the build time O.o :D
15:11.59brlcadand the docs ;)
15:12.12``Erikyeah heh
15:12.19brlcadthat would probably be a 1min build without the trio
15:12.27brlcadmaybe 2min
15:12.53``Erikimagine shifting to a single dir build
15:13.16*** join/#brlcad mafm (n=mafm@83.38.34.217)
15:13.34``Erik-sj31: real 3m27.715s | user 13m49.803s | sys 10m25.144s
15:14.08brlcadyeah, looks like sys is saturated around 13 :)
15:14.13``Erikcache spaz, I imagine
15:14.33brlcador is that enable-optimized?
15:14.55brlcadusually test compile time unoptimized, benchmark optimized
15:15.17brlcadoptimized compile time jacks things up
15:15.26``Erikbuild was unoptimized, am building optimized now for bench
15:16.47CIA-61BRL-CAD: 03brlcad * r36460 10/brlcad/trunk/bench/run.sh: let the user know about 'benckmark clean' to remove the plethora of pix files that get generated. make clean mention clobber to remove the log files.
15:18.07CIA-61BRL-CAD: 03brlcad * r36461 10/brlcad/trunk/HACKING: clean up stable sync instructions since it's always safe to grab CURR, and might as well export PREV so the rest can be copy-pasted without change.
15:25.13``Erik'double t diner' kinda confuses me... it's not... really... a diner... more a normal restaurant, half attempting to be upscale, with a couple diner hints to the decor O.o
15:25.32``ErikI mean, c'mon, a diner with a bar? O.O
15:27.48brlcadat least they didn't name it the "Double D Diner" .. that would have probably attracted a totally different clientelle
15:28.05``Erikindeed, I would've gone there much sooner
15:29.37*** join/#brlcad samrose (n=samrose@adsl-71-156-15-164.dsl.sfldmi.sbcglobal.net)
15:40.25*** join/#brlcad Elrohir (n=kvirc@p5B14EFD2.dip.t-dialin.net)
15:43.59*** join/#brlcad KingofCSU (n=king@222.247.93.92)
16:37.04*** join/#brlcad Elrohir (n=kvirc@p5B14EFD2.dip.t-dialin.net)
16:52.04*** join/#brlcad SWPadnos (n=Me@dsl107.esjtvtli.sover.net)
19:00.54*** join/#brlcad samrose (n=samrose@adsl-71-156-15-164.dsl.sfldmi.sbcglobal.net)
19:09.31yukonbobIt's not a diner unless it's got a rotating showcase of lemon marangue pies
19:12.11archivistluvs lemon marangue pie
19:47.41*** join/#brlcad PrezKennedy (n=Matthew@whitecalf.net)
19:59.27starseeker``Erik: I keep telling ya, disable the doc build if it bugs you ;-)
20:06.35*** join/#brlcad b0ef (n=b0ef@157.26.202.84.customer.cdi.no)
20:31.49CIA-61BRL-CAD: 03indianlarry * r36462 10/brlcad/trunk/src/librt/primitives/brep/brep_debug.cpp: additional face details added to "brep brep.s info F 0" command
20:49.02CIA-61BRL-CAD: 03indianlarry * r36463 10/brlcad/trunk/src/conv/step/OpenNurbsInterfaces.cpp: use 3d curve bounding boxes to ensure linear extrusion sized properly, also checks that both the origin and face bounding box are used to determine extents of cylindrical surface
20:56.25``Erikhey, a patch bump, nifty
20:57.41starseekeryeah, should be just about ready to roll
21:18.12``Erikyup, one of them stepped on my laptop again. *sigh*.
21:18.13*** join/#brlcad IriX64 (n=IriX64@bas2-sudbury98-1128565729.dsl.bell.ca)
21:18.27starseekerThe Cats Strike?
21:18.54``Erikwonders if changing the ground to something with a high potential would fuck up the circuitry
21:19.07``Erikprobably would on discharge into the cat... :/
21:19.17``Erikmotion sensor spray-bottles I guess
21:19.21starseekerjust put the laptop in a box or something
21:19.42``Erikheh
21:19.49``Erikunless they knocked the power cord loose... O.o
21:19.55starseekeraaah
21:19.56``Erikwill have to discover the damage when he gets home
21:20.09starseekerso, how do you like having cats? ;-P
21:20.23``Erikdepends on if they're being turds or not
21:20.36``Erikwhat's the phrase? "I like kids... other peoples kids..."?
21:22.45``Erik"we can hear you having sex" is such an excellent ssid
21:23.07``Erikhttp://failblog.files.wordpress.com/2009/10/epic-fail-wireless-network-fail.jpg
21:24.44starseekerheh:  "First Law of Bicycling: No matter which way you ride, it's uphill and against the wind."
21:26.24starseekerhmm... multithreaded kpathsea library in TeXLive
21:26.46starseekerand the TeXLive build uses automake
21:26.53starseekermust investigate this 2009 release
21:27.05``Erikisn't kpathsea mostly io bound? wouldn't multithreading just screw the pooch for disk read sequencing?
21:27.24starseekerprobably depends on the system
21:27.39starseekerhas a small, slow, crappy system by modern standards
21:27.53``Erikpats his pIII 650mhz with 128m ram
21:28.02``Erikand a massive 20g hdd
21:29.49starseekersights fondly - I remember when 1 Gig was huge and they were talking like 100 Terabytes to back up the entire internet
21:30.19``ErikI was just remarking earlier that I was using a 64gb ram machine and started with a 64kb ram machine...
21:32.24starseekerwonders if 3D printing will be the next big game changer for personal computers
21:32.24*** join/#brlcad Elrohir (n=kvirc@p5B14EFD2.dip.t-dialin.net)
21:32.49starseeker(real, high quality 3D printing that is)
21:33.59starseekerdoes NOT remember fondly the days trying to fit powerpoint presentations for a class onto 1.4 meg floppies
21:34.14starseekereven zip disks were SO welcome
21:47.35``Erikheh
21:48.39``Erikstarted with 4 track cassettes, then upgraded to 5.25" disks... the 3.5" was insane, you could partition it and shove BUNCHES of 5 1/4 inchers on a single disk, it was almost like a hard drive!
21:49.24yukonbob``Erik: !!650Mhz
21:49.29yukonbobmodel name      : Pentium Pro
21:49.29yukonbobstepping        : 9
21:49.29yukonbobcpu MHz         : 199.313
21:49.45yukonbob<napolean dynamite>lucky</napolean dynamite>
21:50.35``Erikheh, I still have a 120mhz cyrix 5x86 with 48m ram and a 1.6gb disk (with debian on it), does that count? :0
21:50.38``Erik:)
21:50.57``Erikhttp://thegreatgeekmanual.com/images/geekhistory/june/coleco-adam-computer.jpg   <-- was his first 'real' computer
21:51.46``Erikand a book called "games apples play", which meant porting apple basic to the cp/m basic the coleco used
21:52.09``ErikI think that was the name of the book, been a while
21:52.33``Erikhad spiral binding and iirc, drawing of goofy robots
21:52.55``Erikshakes his walker at the whippersnappers
21:54.51yukonbobw00t!! Adam with the printer that was a critical part of the system...
21:54.59yukonbobhad one of those...
21:55.07yukonbobwith it's mock Apple II basic
21:56.41yukonbobhad a game that used the printer... it was a military game that send you communications on the ptr as if they were send "over the wire" from some remote headquarters.
21:56.48``Erikhm, didn't the apple][ use a modified ms-basic? (almost all the microcomputer basics were derived from micro-softs iirc)
21:57.03``Erikprinter, powersupply, whatever
21:57.06yukonbob<-- not actually sure...
21:57.37yukonbobis recalling the gfx via "hgr" and "hplot"
21:57.41yukonbob(iirc)
21:58.04yukonbobhgr ,and hgr2 (?)
21:58.16yukonbobhplot x1, y1 to x2, y2
21:59.09yukonbobapples I used were IIc and IIe
21:59.39``Erik<-- went from the adam to the c64, c128, c64c, then a very brief stint using dos/win31, then linux
22:00.09``Erikbrief like 6 months, after 2 I was telnetting to the isp's linux box, then wiping dos and installing linux...
22:00.17yukonbobno 128, but insert few Amigas in there...
22:00.52yukonbobmy intro to linux was similar, but from Win 95.
22:01.03``Erikreboot the 486 with the special doom2 boot disk (since the box only had 4 megs ram, couldn't have shit like himem loaded...)
22:01.05yukonbobthen compiling NCSA Mosaic to browse the interwebs.
22:01.20``Erikyehhhhh, my 6mo of win31 was in '96, I grew up po'
22:02.17yukonbobgives ``Erik a cookie.
22:02.39yukonbobnow look at you... all grown up and p0wning *BSD
22:02.55``Erikheh
22:03.05``Erikfbsd is hardly a pricey toy
22:03.18yukonbobtrue...
22:05.00``Erik<-- trying to decide if he wants to buy a sheeva openrd-client and a 1-1.5tb usb drive, or get a cheap dull box or something :/
22:06.11yukonbobsheeva == more smiles, I'm sure.
22:07.09``Erikyeh, mebbe I'll get both, but I want to replace the p3
22:12.24CIA-61BRL-CAD: 03brlcad * r36464 10/brlcad/trunk/src/libged/keep.c:
22:12.26CIA-61BRL-CAD: BAH! .. bu_optind has to be manually reset before bu_getopt(), otherwise we'll
22:12.28CIA-61BRL-CAD: get whatever previous bu_optind there was. this should fix the problem victor
22:12.30CIA-61BRL-CAD: noticed in sf bug 2890876 ( Keep command in mged is creating a file named keep
22:12.32CIA-61BRL-CAD: ).
22:15.41CIA-61BRL-CAD: 03brlcad * r36465 10/brlcad/branches/STABLE/src/libged/keep.c:
22:15.43CIA-61BRL-CAD: apply the initialization fix so 'keep' isn't broken. change should merge
22:15.45CIA-61BRL-CAD: cleanly with the previous trunk sync revision. slipping in to the release only
22:15.47CIA-61BRL-CAD: because it's not been tagged yet (and is a one-liner safe initialization).
22:17.24CIA-61BRL-CAD: 03brlcad * r36466 10/brlcad/tags/rel-7-16-2/: tagging release 7.16.2
23:33.06*** join/#brlcad SWPadnos (n=Me@dsl107.esjtvtli.sover.net)
23:37.57CIA-61BRL-CAD: 03brlcad * r36467 10/brlcad/trunk/src/libged/human.c: looks like human has the same problem as kill, need to init bu_getopt()
23:42.07CIA-61BRL-CAD: 03brlcad * r36468 10/brlcad/trunk/src/mged/ (cad_boundp.c cad_parea.c mged.c): few more bu_optind init-to-zeros just for sanity sake (as they're the first calls in their respective main() funcs)
23:57.17CIA-61BRL-CAD: 03brlcad * r36469 10/brlcad/trunk/HACKING: sf.net FRS changed slightly. path is different as our whole file release hierarchy is now traversable. should reorganize sometime soon to take advantage of this change.. until then, say what it is now.

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