IRC log for #brlcad on 20091001

00:17.22CIA-28BRL-CAD: 03brlcad * r36119 10/brlcad/trunk/src/librt/primitives/ (14 files in 14 dirs): opennurbs provides a static factory allocator so that objects are instantiated within the DLL on Windows. be a good citizen and call that instead of calling new directly.
00:27.48brlcadis intrigued by the Loft() routines in opennurbs .. another potential means to implement a generalized sweep primitive coupled with the new brep() funcs
00:42.12*** join/#brlcad SWPadnos (n=Me@dsl107.esjtvtli.sover.net)
00:45.22CIA-28BRL-CAD: 03brlcad * r36120 10/brlcad/trunk/src/librt/primitives/ (4 files in 4 dirs): more static factory allocation via New() instead of new
01:35.26CIA-28BRL-CAD: 03brlcad * r36121 10/brlcad/trunk/include/ (nurb.h raytrace.h): move all of the old nurbs functions out of raytrace.h into the nurbs.h header. removed duplication and fixed errors (non-existent functions) in the process.
02:04.21*** join/#brlcad SWPadnos (n=Me@dsl107.esjtvtli.sover.net)
02:36.00*** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-102.sbndin.btas.verizon.net)
03:43.09CIA-28BRL-CAD: 03brlcad * r36122 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: initial stubbing of the surfaces into the container. not quite.
03:58.08CIA-28BRL-CAD: 03brlcad * r36123 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: u/v were flipped, now to fix the knots
04:29.11brlcadhah, we have liftoff
04:33.37CIA-28BRL-CAD: 03brlcad * r36124 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: progress. just got first old NURBS surface to render as a new NURBS surface
04:47.50brlcadoh my god
04:48.02brlcad(!!!)
04:48.18brlcadit fraskingasd worked on the frag the first attempt
04:48.21brlcadand it's blazing fast
04:49.41brlcaddoesn't believe it, goes to triple check things
04:52.44brlcadsure enough, prepped 903 surfaces in less than 2 sec, rendered a 512x512 in 19sec .. now to revert
04:54.28CIA-28BRL-CAD: 03brlcad * r36125 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: this is freaking cool. but gotta turn off debugging else run into the old mged-lockup bug flooding the output channel
04:55.53CIA-28BRL-CAD: 03brlcad * r36126 10/brlcad/trunk/src/librt/primitives/bspline/bspline.cpp: turn off the old wireframe. stay's code definitely has a leg up there showing a detailed surface instead of just the edges, but for something as complex as the frag, it works out even better with just edges
04:57.00CIA-28BRL-CAD: 03brlcad * r36127 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: kill the initial bounding volume code now that things seem to be working.
04:58.45brlcadhah .. now for comparison.  old nurbs prep took about 0 secs, rendered a 512x512 in ...
05:00.08brlcad(still rendering)
05:03.28brlcad(halfway)
05:11.21brlcad702 sec wallclock, 124 sec cpu  (so even if the malloc problem were fixed, it's still nearly 10x slower)
05:22.40brlcadhttp://brlcad.org/tmp/nurbs2brep/
05:22.55brlcadpics ftw
05:23.32brlcadnow the real beast test .. glass t72
05:45.28brlcadlooks like there are over 1000 nurbs surfaces mixed across 6 old bspline primitives, integrated with other objects
06:15.37CIA-28BRL-CAD: 03brlcad * r36128 10/brlcad/trunk/src/librt/primitives/bspline/bspline.cpp: add a couple #define handles to control the wireframe methods
06:24.39brlcadYoshi47: http://brlcad.org/tmp/impeller/impeller2.png
07:42.08CIA-28BRL-CAD: 03brlcad * r36129 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: add a little more diagnostic information about whether there's just one surface, valid object, valid topology, valid geometry, manifold
09:39.20CIA-28BRL-CAD: 03d_rossberg * r36130 10/rt^3/tags/rel-7-16-0/: tag to assign this core interface version to the corresponding BRL-CAD version
09:43.03CIA-28BRL-CAD: 03d_rossberg * r36131 10/brlcad/trunk/src/librt/CMakeLists.txt: typo in file name suffix
10:53.12Yoshi47brlcad, you manage to get the script going? is that the new one? or old one?
10:53.23Yoshi47nice rendering though
10:53.27Yoshi47thanks
10:56.05Yoshi47brlcad, your last 12 lines, did you do the test with the impeller?
10:57.33Yoshi47that reminds me i have to clean up that tutorial i wrote, maybe when things settle down, My daughter is in the hospital again, she not eating or drinking much, so she needs a bunch of tests, but don't worry she happy and pretty healthy otherwise, they just plugged her in to the IV! She got a port under the skin, kinda cool actually.
11:12.25brlcadYoshi47: all that chatter before the impeller2 url had nothing to do with the impeller :)
11:12.32Yoshi47oh ok
11:13.04Yoshi47so was it the new one?
11:14.23brlcadI still don't know new/old, just the one I have :)
11:14.42brlcadwhich I have to imagine is new, but before you wrote the script
11:14.53brlcadbecause otherwise you wouldn't have needed to recreate it
11:15.50Yoshi47ah does it have a key way? when you have time to check, no rush, no key way means its the old one, the new one for the tutorial i added a keyway
11:33.16*** join/#brlcad roberthl (n=robert@rhl.me.uk)
11:33.47brlcadheh, actually I seemed to have just lost the .g file....
11:33.58brlcadsearching the filesystem
11:36.37Yoshi47lol
11:37.27Yoshi47well you can paste each of these lines into mged to make the new one, http://ej.nijenhuis.pixi.me/d/2452-1/impeller.txt?g2_GALLERYSID=547cde1a5f437fb485373a2a612d12ce
11:37.33Yoshi47if you want,
11:38.47brlcadhow does that help answer the question? :)
11:39.00brlcadif you can't tell from the pictures, I certainly can't :)
11:39.36Yoshi47one sec then, i'll render it quickly.
11:39.43Yoshi47opening up another instance of mged
11:41.23Yoshi47have to copy file from home
11:49.15Yoshi47thats the old one, the new one only has 11 fins, but i do have to admit the old one looks better with 12 fins, i'll change the script to include 12 fins
11:51.13Yoshi47im going to redo it with the new command and see what it get then i'll reupload the new commands
11:51.37brlcadk
11:57.47CIA-28BRL-CAD: 03brlcad * r36132 10/brlcad/trunk/src/librt/primitives/bspline/nurb_brep.cpp: comment cleanup, print surface count
12:01.48Yoshi47on that render you just did, what the little white thing at the top closer to the left side?
12:05.45Yoshi47here is the commands to create the new tutorial Impeller, which has 12 fins like the old one but has an added keyway, if you do get around to a render can you get the keyway visible, not straight on but to the side a little bit, http://ej.nijenhuis.pixi.me/d/2456-1/impeller.txt?g2_GALLERYSID=547cde1a5f437fb485373a2a612d12ce
12:13.04brlcadthat's a custom light source
12:13.11CIA-28BRL-CAD: 03brlcad * r36133 10/brlcad/trunk/src/librt/opennurbs_ext.cpp: (c-set-offset 'innamespace 0) .. undo drain bramage indent within namespace, cleanup
12:40.04``Erikdang emacs specialness :D
12:57.21*** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-102.sbndin.btas.verizon.net)
12:57.48Yoshi47brlcad, thought so
13:21.25*** join/#brlcad LarsG (n=lars@nusnet-223-10.dynip.nus.edu.sg)
13:21.38*** part/#brlcad LarsG (n=lars@nusnet-223-10.dynip.nus.edu.sg)
13:28.14*** join/#brlcad roberthl (n=robert@silentflame/member/roberthl) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad starseeker (n=starseek@63.246.136.16) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad brlcad (n=sean@63.246.136.16) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad archivist_emc (n=archivis@host81-149-119-172.in-addr.btopenworld.com) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad akafubu (n=akafubu@unaffiliated/akafubu) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad archivist (n=archivis@host81-149-119-172.in-addr.btopenworld.com) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad PrezKennedyII (i=Matthew@whitecalf.net) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad yukonbob (i=1000@s142-179-54-198.bc.hsia.telus.net) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad b0ef (n=b0ef@157.26.202.84.customer.cdi.no) [NETSPLIT VICTIM]
13:28.14*** join/#brlcad louipc (n=louipc@archlinux/trusteduser/louipc) [NETSPLIT VICTIM]
14:11.12*** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-102.sbndin.btas.verizon.net)
14:22.53``Erikheh, wow, my brain parsed "DirectX 11" as "direct X11" O.o I must be a *nix geek.
14:34.40d-lolol
15:03.22*** join/#brlcad archivist_emc (n=archivis@host81-149-119-172.in-addr.btopenworld.com)
15:08.17louipcawe no source tarball for 7.16.0?
15:10.21brlcadlouipc: not quite yet .. trying later today
15:10.38brlcadit is tagged, and the tarballs are sitting on my machine .. just didn't upload
15:10.45louipcaah cool
15:17.55``Eriksoooo, we can start breakin' shit again? :D
15:27.12brlcadyep
15:29.22brlcadYoshi47: http://brlcad.org/tmp/impeller has the .g I was using
15:38.15Yoshi47ah, i still have that one
16:03.57*** join/#brlcad mafm (n=mafm@2.Red-83-63-197.staticIP.rima-tde.net)
16:24.33*** join/#brlcad PrezKennedyIII (i=Matthew@whitecalf.net)
16:28.50*** join/#brlcad yukonbob (i=1000@s142-179-54-198.bc.hsia.telus.net) [NETSPLIT VICTIM]
16:29.49*** join/#brlcad brlcad (n=sean@BZ.BZFLAG.BZ) [NETSPLIT VICTIM]
16:31.53*** join/#brlcad starseeker (n=starseek@63.246.136.16) [NETSPLIT VICTIM]
16:48.48*** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net)
17:08.13*** join/#brlcad talcite (n=Matthew@134.117.254.248)
18:33.07*** join/#brlcad Ralith (n=ralith@69.90.49.189)
18:56.46``Erikhah http://news.ycombinator.com/item?id=855236
19:08.20*** join/#brlcad IriX64 (n=IriX64@bas2-sudbury98-1177871686.dsl.bell.ca)
19:37.57CIA-28BRL-CAD: 03erikgreenwald * r36134 10/brlcad/trunk/src/conv/obj-g.c: fixing some silly bugs.
19:51.46``Erikso much silliness.
19:51.59CIA-28BRL-CAD: 03erikgreenwald * r36135 10/brlcad/trunk/src/conv/obj-g.c: fix silly offset bug
19:59.05*** join/#brlcad PrezKennedyII (i=Matthew@whitecalf.net)
20:10.14``Erikhm.
20:32.31CIA-28BRL-CAD: 03erikgreenwald * r36136 10/brlcad/trunk/src/conv/obj-g.c: make region static (there's a reason). Prep the region in case no groups are defined in .obj file. Write last region.
20:38.33*** join/#brlcad cpc26 (n=cpc26@72.170.156.242)
20:46.02``Erikcurses some.
20:47.13*** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net)
20:52.08*** join/#brlcad ibot (i=ibot@rikers.org)
20:52.08*** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || Release 7.14.8 posted (20090511) || GSoC2009 Next Step: upload your code to google, wait for shirt ;) thanks everyone! || Logs at http://ibot.rikers.org/%23brlcad/
21:47.46CIA-28BRL-CAD: 03erikgreenwald * r36138 10/brlcad/trunk/src/conv/obj-g.c: improved command catching, start using the facemax... thingie... etc
22:35.32*** join/#brlcad mafm (n=mafm@2.Red-83-63-197.staticIP.rima-tde.net)
22:36.01brlcadyay, a curses isst inteface
22:41.39Ralithisst?
22:48.44``Erikwhu?
22:49.06``Erikheh
22:49.27``Erikyeahhhh, I think I'm gonna be busy with obj-g and playing with my balls for a bit, isst is kinda more back burner now
22:49.51``Erik<-- can't find anyone to take care of his balls for him :/
23:22.04brlcadRalith: isst is the "interactive shotline selection tool" .. a real-time ray-trace viewer that lets you spin a model around, look at split views, etc
23:25.07``Erikit exists in brlcad/src/adrt/ with the gui frontend in isst/
23:25.41``Erikhttp://brlcad.svn.sourceforge.net/viewvc/brlcad/isst/trunk/
23:30.56``ErikI HAVE HEARD THE CRY OF THE CARROTS!
23:36.34Ralithbrlcad: oh neat
23:39.20``Erikralith: do you have a recent BRL-CAD and GTK+ installed?

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