IRC log for #brlcad on 20160629

00:12.59*** join/#brlcad tajjwsnbdblwvyaj (~armin@dslb-092-074-228-216.092.074.pools.vodafone-ip.de)
02:27.38bojHi all, I want to share an online webpage which lists the instructions on how to use some of mged commands, but it is a lit bit old, just for a quick reference.
02:27.42bojhttps://manned.org/browse/freebsd-9.2/brlcad/7.24.0
03:49.47*** join/#brlcad Zitar (~Zitar@31.203.113.156)
04:19.23Notify03BRL-CAD Wiki:Tandoorichick * 9773 /wiki/User:Tandoorichick/GSoC2016/Logs: /* Development Logs */
04:28.32*** join/#brlcad tandoorichick (~rakshika@117.222.157.133)
06:04.35*** join/#brlcad Zitar (~Zitar@31.203.113.156)
06:26.31*** join/#brlcad Zitar (~Zitar@31.203.113.156)
06:41.55*** join/#brlcad amarjeet (~amarjeet@202.164.53.117)
06:43.50*** join/#brlcad tandoorichick (~rakshika@117.230.25.46)
06:50.30*** join/#brlcad teepee_ (~teepee@unaffiliated/teepee)
06:52.10*** join/#brlcad tandoorichick_ (~rakshika@117.242.93.170)
06:54.04*** join/#brlcad ries (~ries@D979C7EF.cm-3-2d.dynamic.ziggo.nl)
08:16.31*** join/#brlcad sniok (~sniok@89.252.29.238)
08:59.34*** join/#brlcad amarjeet (~amarjeet@202.164.53.117)
09:51.23*** join/#brlcad Zitar (~Zitar@31.203.113.156)
09:58.28*** join/#brlcad tandoorichick_ (~rakshika@117.242.93.170)
11:57.32*** join/#brlcad teepee] (bc5c2133@gateway/web/freenode/ip.188.92.33.51)
12:08.52*** join/#brlcad asad_ (~asad00@host10-2.natpool.mwn.de)
13:21.46*** join/#brlcad tandoorichick_ (~rakshika@117.222.157.133)
13:29.58*** join/#brlcad yorik (~yorik@179.111.189.40)
13:38.58*** join/#brlcad Mathnerd314 (~quassel@supertux/Mathnerd314)
14:20.00starseekerbrlcad: I was looking at the OpenSCAD user manual and noticed that they appear to have some form of integrated revision control:  https://en.wikibooks.org/wiki/OpenSCAD_User_Manual
14:21.18starseekertheir log looks a lot like our edit log - is this a feature we could turn on on our websites to communicate their status as far as pending changes?
14:46.17teepee]starseeker: standard wikimedia :)
14:52.03*** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
15:24.38*** mode/#brlcad [+o brlcad] by ChanServ
15:24.43*** join/#brlcad amarjeet (~amarjeet@101.211.212.173)
15:25.09brlcadstarseeker: yep, go for it ;)
15:26.09brlcadboj: thanks for sharing
15:26.30brlcadhadn't seen that particular hosting, decent manpage stylesheet
15:26.57brlcadasad_: I'm available today if you are as well to talk
15:28.41starseekerO.O https://github.com/Microsoft/Mesh-processing-library
15:29.11starseekerthat's by Hoppe, if I'm reading this right
15:29.36brlcad``Erik: video: http://www.dailymail.co.uk/video/sciencetech/video-1133188/Electric-car-breaks-record-accelerating-100km-h-1-8secs.html
15:29.43starseekerand MIT licensed
15:30.12starseekerC++11, but even so quite possibly a very useful resource :-)
15:33.13starseekerthis page maps papers to code: https://rawgit.com/Microsoft/Mesh-processing-library/master/README.html
15:36.17starseekerpity the bspline surface reconstruction paper isn't there
15:41.28bojHi brlcad, now I want to do something related to tessellation of NURBS, do you have some suggestions?
15:42.09bojbecause I find that there is an old implementation, do I need to follow the old code and re-implement it in new brep or..?
15:43.14starseekerboj: did you see my email?
15:43.37bojsorry, let me check my email now.
15:44.49bojstarseeker, I have not received your email.
15:46.03starseekerboj: I sent it to the BRL-CAD list... one second...
15:46.30starseekerhttps://sourceforge.net/p/brlcad/mailman/message/35190445/
15:47.09bojOK, but in the lastest volume, I only see my post, but without reply..weird..
15:47.40starseekeroh, if you're doing volumes rather than individual emails you would need to wait for the next collection of emails
15:48.05starseekerno matter
15:48.17bojI see, thanks, starseeker, I will check it now. :-)
15:50.39bojstarseeker, as for plate mode NURBS ray tracing, I have write all my test on my blog, could you help me check it?
15:52.31bojI just want to know do I miss something and I need to make further refinement.
15:53.42starseekerboj: the check will be the mathematical accuracy of the raytracing results
15:54.21bojI jsut saw your reply, but not really understand..
15:54.48starseekeryou have a "sphere shell" which you can calculate a volume for using the traditional volume formula: Volume(sph_outer) - Volume(sph_inner)
15:55.34starseekerwhen you use the tool "gqa" - a BRL-CAD tool that will calculate the volume of objects using raytracing - you should get an answer that very closely matches the pure math answer
15:56.12starseekerso the "outer" and "inner" sphere radii will be determined by the thickness you choose to assign to the plate mode NURBS sphere
15:57.19starseekeryou can adjust the thickness of your plate mode NURBS sphere to be thinner or thicker
15:57.36starseekeri.e. lesser or greater volume
15:57.51starseekerremember that a plate mode solid, although thin, does have a non-zero volume
15:58.09bojyes, I understand.
15:58.13starseekerthe raytracer and a pure mathematical volume calculation should closely agree what that volume is
15:58.36bojis there any threshold?
15:58.44starseekerboj: in MGED, if you type "man gqa" you should get a manual page for the gqa tool
15:58.47starseekerboj: threshold?
15:59.19boja threshold to determine the consistency of 'calculated volume' and 'raytracing volume'?
15:59.48starseekerfor the closed sphere, they should be the same
15:59.59starseekerbecause there are no unmated NURBS edges
16:00.45bojyes, but for closed sphere, the hit point will be one in hit and one out hit?
16:01.03starseekernot in plate mode - you should get two in hits and two out hits
16:01.24starseekerif I understand the methodology correctly, at least
16:02.08starseekeryou can test with a facetized sphere - it should report two hits in solid mode and four in plate mode if I'm right
16:03.09bojyes, should be right.
16:03.48bojI will try it later.
16:04.06bojbut why the first step is to check the volume consistency?
16:04.13starseekerfor NURBS surfaces with unmated edges (like your 1/6th of a Cobb sphere example) the accuracy of plate move volume estimates should get lower as you increase the thickness, as you'll "miss" grazing hits
16:05.06starseekerboj: because we want to make sure the methodology (plate mode NURBS raytracing) produces physically reasonable results
16:05.51starseekersolid raytracing (as opposed to first hit raytracing) provides volumetric information, so it's a good test that the ray intersections are correct
16:06.22bojtill now, I only test the hit point, and it seems to be correct.
16:06.39starseekernods - to produce a thickness though, we need more than one hit point
16:06.55starseekerlike the BoTs - you need to use that hit point to "deduce" the plate mode hit points
16:07.12bojyes, in my current implementation, I can get two in/out hit for each hit in plate mode.
16:07.32starseekerso there should be some distance between those two points
16:07.44bojI do the same thing for brep now..using hit distance to deduce the hit point.
16:08.23bojyes, the distance now is just the thickness value, what I do is to center the hit point and offset it back and forth.
16:08.31starseekerboj: another basic test is to create a plate mode NURBS surface (I think Ayam can do this if you save as an openNURBS 3dm file and use our 3dm-g importer) and compare the hit points on that surface to the same hit points on an equalivent BoT
16:09.05starseekerboj: right - so that offset and those two hit points give you a line segment - our tools take those line segements and use them to deduce volume
16:10.22bojAyam..never used before..
16:10.23starseekerfor a planar NURBS surface in plate mode and a planar BoT in plate mode, the answers should match as long as they describe the same plate
16:10.40bojI understand.
16:10.40starseekerboj: http://ayam.sourceforge.net/
16:10.58bojthanks, starseeker
16:11.06starseekeronce you work with those basic examples, there's also a nasty test surface in our repo
16:11.47bojcould you give me some examples or is there a manual for me to refer to on how to test it?
16:11.49starseekersrc/librt/tests/extreme_ssi_test.g
16:12.26starseekerboj: you can open that .g file, turn that NURBS surface into a plate mode surface, and see what it does
16:12.39starseekerin principle it should work, but you never know
16:13.24bojnods, I do need to run more stress test examples
16:14.03starseekerthe plate and sphere tests are good starting points
16:14.29bojit should be, because they are simple and easy to create and test.:)
16:15.36bojI will do it recently, and post my result on my blog, thanks starseeker
16:16.37starseekermy pleasure - good luck!
16:16.52boj:)
16:29.16brlcadboj: hi, I just responded to your e-mail
16:30.12brlcadI agree with cliff that if you think it's working well, the next immediate step is to validate the implementation numerically
16:30.29brlcadI'd also like to see a little more variety in your manual testing
16:30.53bojthanks brlcad, and where I can see your reply?
16:31.14bojyes, I will do it with more test case and try to verify the result.
16:31.20brlcadboj: um, your e-mail client? :)
16:31.29brlcadyou sent a mail to the mailing list, I replied to the mailing list
16:32.04bojOK, I see it now. :-)
16:33.12brlcadboj, have you tried viewing your plate-mode surfaces in shaded mode?
16:33.52bojwhat I mean 'rough' is I do not really if it is right for all examples, but after this I will check it.
16:34.03*** join/#brlcad tandoorichick (~Thunderbi@117.222.157.133)
16:34.18bojbrlcad, what do you mean shaded mode? how to activate it?
16:35.34*** join/#brlcad tandoorichick (~Thunderbi@117.222.157.133)
16:36.07bojby the way, how to create more various models? like nurbs with trimmed curves?
16:39.36*** join/#brlcad tandoorichick (~Thunderbi@117.222.157.133)
16:49.30bojOK, I just saw that, maybe I can use Ayam(recommended by starseeker above) to make more nurbs examples and test it later..
17:19.05Notify03BRL-CAD Wiki:Adreabroline * 0 /wiki/User:Adreabroline:
17:21.48Notify03BRL-CAD Wiki:Adreabroline * 9774 /wiki/Quickbooks_p.h.o.n.e_n.u.m.b.e.r(1.8.0.0.7.2.8.7.3.5.6))_Quickbooks_s.u.p.p.o.r.t_p.h.o.n.e_n.u.m.b.e.r_Quickbooks_h.e.l.p.l.i.n.e_p.h.o.n.e_n.u.m.b.e.r: Created page with "Phone N.um.b.e.r 1.8.0.0.7.2.8.7..3.5.6 Q.u.i.c.k.b.o.o.k.s su.p.p.o.r.t p.h.o.n.e n.u.m.b.e.r Q.u.i.c.k.b.o.o.k.s p.r.o su.p.p.o.r.t p.h.o.n.e n.u.m.b.e.r Helpline “”..."
17:40.35*** join/#brlcad sniok (~sniok@89.252.2.135)
18:07.28*** join/#brlcad sniok (~sniok@89.252.2.135)
18:11.57*** join/#brlcad sniok (~sniok@89.252.2.135)
18:23.26brlcadboj: draw -m1
18:23.37brlcador in archer, it's one of the right-click draw modes
18:26.07*** part/#brlcad sniok (~sniok@89.252.2.135)
18:31.33*** join/#brlcad sniok (~sniok@89.252.2.135)
19:14.37Notify03BRL-CAD:ejno * 68197 (brlcad/trunk/include/brep/bbnode.h brlcad/trunk/include/brep/brnode.h and 8 others): use 'const'
19:15.14brlcadhere's a project I'd forgotten about...
19:15.14brlcadsulfredgecd@ornl.gov
19:15.17brlcadoops
19:15.31brlcadhttp://visac.ornl.gov
19:15.48brlcadthat, not the contact who I just contacted ;)
19:30.48Notify03BRL-CAD Wiki:Tandoorichick * 9775 /wiki/User:Tandoorichick/GSoC2016/Logs: /* Development Logs */
20:21.30*** join/#brlcad asad_ (~asad00@host10-2.natpool.mwn.de)
20:38.59Notify03BRL-CAD:ejno * 68198 (brlcad/trunk/include/brep/bbnode.h brlcad/trunk/include/brep/surfacetree.h and 3 others): add 'const'
20:40.49Notify03BRL-CAD:brlcad * 68199 brlcad/trunk/src/adrt/librender/render_util.c: convert to libbu memory management
20:54.40*** join/#brlcad teepee_ (~teepee@unaffiliated/teepee)
20:54.43Notify03BRL-CAD Wiki:Aprilatilley * 0 /wiki/User:Aprilatilley:
20:55.55Notify03BRL-CAD Wiki:Sean * 0 /wiki/User:Adreabroline:
20:56.05Notify03BRL-CAD Wiki:Sean * 0 /wiki/Quickbooks_p.h.o.n.e_n.u.m.b.e.r(1.8.0.0.7.2.8.7.3.5.6))_Quickbooks_s.u.p.p.o.r.t_p.h.o.n.e_n.u.m.b.e.r_Quickbooks_h.e.l.p.l.i.n.e_p.h.o.n.e_n.u.m.b.e.r: Spam: spam
21:04.01Notify03BRL-CAD:brlcad * 68200 brlcad/trunk/TODO: this needs a recheck to make sure it's correct changes, and a quick news entry if it's okay
21:07.46Notify03BRL-CAD:brlcad * 68201 brlcad/trunk/TODO: nick fixed archer flashing and improved window positioning
21:08.41Notify03BRL-CAD:brlcad * 68202 brlcad/trunk/TODO: search docs improved
21:26.16Notify03BRL-CAD:brlcad * 68203 brlcad/trunk/TODO: screwy change in 58189, needs another pair of eyes
21:45.01Notify03BRL-CAD:brlcad * 68204 brlcad/trunk/TODO: fixed bug but dont know which, if it's news-item worthy, can go on any release after 7.24.0 as appropriate
21:50.32Notify03BRL-CAD:brlcad * 68205 brlcad/trunk/TODO: tom added a ton of detail in 58325 on timestamping and overall formatting cleanup
21:54.07Notify03BRL-CAD:brlcad * 68206 brlcad/trunk/TODO: update astyle needed
21:54.41Notify03BRL-CAD:ejno * 68207 (brlcad/trunk/include/brep/bbnode.h brlcad/trunk/include/brep/brnode.h and 4 others): add 'const'; ws
22:00.47Notify03BRL-CAD:brlcad * 68208 brlcad/trunk/TODO: tom made changes in 58415 and a few other commits the propagated case insensitive attribute name handling. pretty major behavior change worthy of mention, but also a quick test to make sure it is indeed insensitive for any attributes or just ours
22:01.41Notify03BRL-CAD:brlcad * 68209 brlcad/trunk/TODO: gen-registered-attributes-html should get integrated with the doc work
22:07.27Notify03BRL-CAD:brlcad * 68210 brlcad/trunk/TODO: daniel fixed db_lookup bug, unquantified but real user-visible impact.
22:37.12Notify03BRL-CAD:brlcad * 68211 (brlcad/trunk/TODO brlcad/trunk/include/rt/comb.h): need API to return actual color-in-context so override colors are reported and handled correctly
22:46.02Notify03BRL-CAD:brlcad * 68212 brlcad/trunk/src/rt/sh_tcl.c: remove dead code
22:53.39Notify03BRL-CAD:brlcad * 68213 brlcad/trunk/TODO: nick says he probably tested it.
23:10.24*** join/#brlcad teepee_ (~teepee@unaffiliated/teepee)
23:38.32*** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)

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