| 03:19.34 | starseeker | hah - someone did some work on qconsole - even have a Qt tcl console example |
| 03:19.47 | starseeker | http://sourceforge.net/projects/qconsole/ |
| 03:27.09 | starseeker | builds too (well, with one tweak) |
| 03:29.11 | starseeker | ah, crud - licensing is still messed up |
| 03:31.04 | CIA-69 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/block: blocked [[User:LewisxuekdeowyvSchwertner]] with an expiry time of infinite (account creation disabled, e-mail blocked): Inserting nonsense/gibberish into pages |
| 03:31.07 | CIA-69 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/delete: deleted "[[?????????????????? ???????????????????????? ??????????????? ??????????????????????????? ???????????????????????? ?????????????????????????]]" |
| 04:06.30 | CIA-69 | BRL-CAD: 03Elf11 07http://brlcad.org * r4441 10/wiki/Mged_simulation: |
| 04:09.55 | CIA-69 | BRL-CAD: 03Elf11 07http://brlcad.org * r4442 10/wiki/User:Elf11: /* Log */ |
| 10:37.34 | *** join/#brlcad stas (~stas@82.208.133.12) | |
| 12:54.42 | CIA-69 | BRL-CAD: 03n_reed * r52457 10/brlcad/trunk/src/libbu/tests/bu_redblack.c: check scanf return value |
| 13:18.53 | CIA-69 | BRL-CAD: 03brlcad * r52458 10/brlcad/trunk/BUGS: ogl display manager is creating windows with a transparent background. first reported by elf_ and todayman via IRC. |
| 13:54.13 | CIA-69 | BRL-CAD: 03brlcad * r52459 10/brlcad/trunk/src/libnurbs/opennurbs_ext.cpp: underscores are reserved for compiler use. fixes compilation failure with gcc 4.1.3 |
| 13:55.34 | CIA-69 | BRL-CAD: 03brlcad * r52460 10/brlcad/trunk/src/libnurbs/opennurbs_ext.cpp: init var |
| 14:14.45 | CIA-69 | BRL-CAD: 03brlcad * r52461 10/brlcad/trunk/src/librt/binunif/db5_bin.c: isdigit takes an int, upcast |
| 14:37.44 | CIA-69 | BRL-CAD: 03brlcad * r52462 10/brlcad/trunk/src/librt/primitives/brep/brep.cpp: looks like we're intentionally casting the double to int as a quickly sortable 't' parameter, so cast accordingly |
| 14:50.32 | CIA-69 | BRL-CAD: 03brlcad * r52464 10/brlcad/trunk/CMakeLists.txt: one of two steps needed to support 10.5+ compilation by default |
| 14:53.48 | CIA-69 | BRL-CAD: 03brlcad * r52465 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/CompilerFlags.cmake): move it over to the other compilation flags, wrap to vendor |
| 15:08.58 | starseeker | woot - qconsole author did go LGPLv2 on his licensing - he just hadn't updated his code headers. Fixed it last night |
| 15:23.44 | elf_ | brlcad, are you around? :) |
| 15:27.34 | brlcad | }}/names |
| 15:27.55 | brlcad | elf_: sometimes |
| 15:28.22 | elf_ | Haha okay :) was going to ask if you got a chance to look at that wiki page and if it looks okay and if so what next? |
| 15:43.54 | brlcad | elf_: you tell me |
| 15:44.21 | elf_ | I tell you about what? if the page looks good or what next? |
| 15:45.19 | brlcad | yes |
| 15:46.21 | elf_ | Okay, the page looks good to me, but I wanted a second opinion... |
| 15:46.40 | *** join/#brlcad CIA-68 (cia@cia.vc) | |
| 15:47.04 | brlcad | elf_: how would you improve it? |
| 15:47.18 | elf_ | I think about what's next I can say that collision detection is not working, or not working properly at least, from the simulation the cube stops but not before entering the plane |
| 15:49.05 | elf_ | Hmm how would I improve it, hard question, probably I would try and make the simulation a little bit more complicated, instead of just one cube falling there would be more, stashing them on top of each other and then even tumbling over |
| 15:50.18 | elf_ | the explanation so far seems okay, made one of my friends to follow it and he did get the cube falling to the ground |
| 15:52.46 | brlcad | elf_: that last bit made me smile |
| 15:52.58 | brlcad | even better third-party validation |
| 15:53.04 | brlcad | so did you friend have any questions? |
| 15:53.15 | brlcad | anything you explained to them while they were following it? |
| 15:53.45 | elf_ | He was more interested about the brlcad program he never heard of it, so he wanted to know more about it and if it was anything like CAD |
| 15:54.26 | brlcad | so the only thing I notice is the simulation wart |
| 15:54.56 | elf_ | the simulation what? sorry not familiar with that word :) |
| 15:55.13 | brlcad | the cube falls to the ground and then it looks like a boat tipping over |
| 15:55.18 | brlcad | because of the collision issue |
| 15:55.24 | brlcad | your tutorial doesn't say anything about that |
| 15:56.21 | brlcad | you should mention that somewhere in a paragraph or box by itself just so it's clear that it's only using the center point for collision |
| 15:56.52 | brlcad | I'm curious if you see that same behavior if you scale everything up 10000 units |
| 15:57.28 | elf_ | I can try that |
| 15:57.36 | brlcad | basically, before your first 'in' command, run "units m" |
| 15:58.09 | elf_ | yeah to set it to meters :) |
| 15:58.16 | brlcad | right |
| 15:58.34 | brlcad | as it is, you made a 1mm^3 tiny box |
| 15:58.57 | brlcad | if abhijit had any sort of minimum tolerance, it could be size-sensitive |
| 15:59.15 | elf_ | yeah, we'll see if that happens with larger geometry too |
| 16:00.10 | brlcad | after checking that, it's time to get to code |
| 16:00.19 | brlcad | you asked what's next |
| 16:01.47 | brlcad | well, you're familiar with the basics of libged, you know how to use the simulate command, so you can make it better |
| 16:02.18 | brlcad | you have about six weeks remaining |
| 16:02.38 | brlcad | in that time, hopefully you can figure out the collision detection bug and expand it to arbitrary geometry |
| 16:03.43 | brlcad | I'd like to see the same simulation you just ran, but instead of a box, have it work with an ellipsoid .. get the ellipsoid to bounce and roll away |
| 16:03.51 | brlcad | drop the box on a corner and watch it tumble |
| 16:04.27 | elf_ | aha |
| 16:04.30 | brlcad | and HOPEFULLY get to the point where you could drop the m35 model (maybe tilted 20 degrees) and have it land on all four wheels |
| 16:05.18 | brlcad | this is the m35: http://brlcad.org/gallery/s/renderings/my_m35.png.html |
| 16:14.09 | elf_ | Abhi put in some code to draw AABBs to show the rays which are shot and I can start there to see why collision is not working |
| 16:15.34 | elf_ | Also I said awhile back about that standalone program in bullet, where a cube is lying over another static cube/groudn plane, I can check the contact points that bullet is generating and the normals and then replicate this inside mged, once this works then point generation and normals are correctly being generated |
| 16:22.48 | elf_ | Uhmm there seems to be another problem with the simulate command, I tried to run it for the geometry as the one in the tutorial with the only difference of now being in meters rather than mm and it gets aborted, 100 steps in mm != 100 steps in meters, which I find strange |
| 16:23.19 | elf_ | running the simulation for 100 steps in meters it looks like the cube is not moving at all |
| 16:25.00 | CIA-68 | BRL-CAD: 03brlcad * r52467 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: |
| 16:25.00 | CIA-68 | BRL-CAD: wasn't pulling the environment variable correctly, uses curlies. expand the |
| 16:25.00 | CIA-68 | BRL-CAD: other remaining piece, the ability to set the compiler/linker options for the |
| 16:25.00 | CIA-68 | BRL-CAD: various versions of the mac sdk. change default to 10.3 just because it seems |
| 16:25.02 | CIA-68 | BRL-CAD: to work, though I'd entirely expect X11 incompatibilities going that far back. |
| 16:26.57 | CIA-68 | BRL-CAD: 03brlcad * r52463 10/brlcad/trunk/src/librt/primitives/ell/ell.c: ensure initialized before use (compiler warned) |
| 16:27.40 | brlcad | elf_: when you check the code that shoots rays, make sure it's always shooting at least 4 rays (2x2) or more |
| 16:28.02 | brlcad | 3x3 would be fine for a min too, probably good compromise |
| 16:28.04 | *** join/#brlcad Al_Da_Best (~Al_Da_Bes@5e0e5085.bb.sky.com) | |
| 16:28.58 | elf_ | Okay |
| 16:29.23 | brlcad | elf_: so your entire simulation before moved the box 50mm .. across 100 simulation steps |
| 16:29.54 | *** join/#brlcad milamber (~devlin@d118-75-244-176.try.wideopenwest.com) | |
| 16:29.55 | brlcad | increasing everything by 1000, you might need to increase the simulation steps by that much to match |
| 16:30.10 | brlcad | simulate 100000 |
| 16:30.13 | elf_ | yeah the problem is that now I uhmm ran it for 10k steps and it aborted |
| 16:30.34 | elf_ | it might have something to do with my laptop or it might be something it |
| 16:30.42 | elf_ | I am trying again |
| 16:50.40 | elf_ | It gets aborted every time, I will run it for less than 10k more than one time and see what happens then |
| 17:09.46 | brlcad | so run in a debugger and figure out why ;) |
| 17:10.56 | brlcad | don't let the issues pile up even if it's not directly related to what you were looking into, just keep a note of what you were doing so if debugging the 10k crash takes a couple days, you can come back to what you were doing once it's fixed |
| 17:13.17 | elf_ | Okay, will run it in a debugger and see what's the problem :) |
| 17:31.52 | CIA-68 | BRL-CAD: 03brlcad * r52468 10/brlcad/trunk/TODO: regress failure, new command halting distcheck when adaptagrams is not available |
| 17:43.52 | *** join/#brlcad stas (~stas@86.122.32.234) | |
| 18:00.13 | CIA-68 | BRL-CAD: 03carlmoore * r52469 10/brlcad/trunk/src/librt/test_botpatches.cpp: remove trailing blanks/tabs |
| 19:28.02 | CIA-68 | BRL-CAD: 03bob1961 * r52470 10/brlcad/trunk/src/libdm/ (dm-X.c dm-ogl.c dm-rtgl.c): Modified the _choose_visual routines to return a visual that is on the screen of interest. Applied to dm-ogl.c, dm-X.c and dm-rtgl.c. Still need to look at dm-wgl.c |
| 20:28.25 | *** join/#brlcad yiyus (~124271242@je.je.je) | |
| 21:08.43 | brlcad | starseeker: you know what that commit from bob was about? |
| 22:14.57 | *** join/#brlcad merzo (~merzo@87-15-133-95.pool.ukrtel.net) | |
| 22:35.29 | *** join/#brlcad Yoshi47 (~jan@d24-204-236-81.home4.cgocable.net) | |
| 22:42.00 | *** join/#brlcad dtidrow (~dtidrow@c-68-84-167-135.hsd1.mi.comcast.net) | |
