| 00:08.12 | brlcad | not really |
| 00:08.58 | louipc | is it possible to make a helix in BRLCAD? |
| 00:09.03 | brlcad | wasn't a serious bug, only rears its head when you write *other* code that is already wrong .. that was just a sanity check that should always be true |
| 00:09.46 | brlcad | louipc: ahhh, you're not the first to ask that ;) |
| 00:09.51 | brlcad | depends what kind, some yes |
| 00:10.03 | brlcad | others, not so easy |
| 00:11.09 | louipc | something like a sph traversing along the length of a rcc and subtracting it's path |
| 00:11.31 | brlcad | you can generally make one in one of a couple ways, using half-torii patterned up, or using the pipe primitive -- which of course limits the shape of the helix |
| 00:11.54 | brlcad | yeah, either torus or pipe effectively will give you that shape |
| 00:12.23 | louipc | oh patterned up I see |
| 00:12.24 | brlcad | a torus is representable as a swept sphere |
| 00:12.30 | louipc | I'll have to experiment |
| 00:17.39 | louipc | do you think it would be possible to draw something like this in BRLCAD? http://www.friotherm.com/impeller.jpg |
| 00:21.37 | brlcad | yeah, it's possible .. quite a hard part to model regardless, but possible |
| 00:22.30 | brlcad | the entire base structure is pretty trivial, something like an rcc, with an inner rcc and a torus subtracted |
| 00:22.46 | brlcad | the hard part is one of the blades |
| 00:22.51 | louipc | yeah it's the blades with their curves which are tricky |
| 00:23.34 | brlcad | if we had a birail primitive (on the wish list), it'd be easier, define two curves, interpolates the path |
| 00:24.33 | brlcad | without it, the shape looks like the edge of two parabolic ellipsoids attached together where the direct changes |
| 00:26.37 | brlcad | would be seriously tricky csg, probably about 10 primitives for each of the two blade sizes, then replicated in a radial pattern |
| 00:29.10 | brlcad | no easy way you'd get those blends along the base, though, that's for sure -- curvature on curvature is a weakness |
| 00:31.42 | louipc | yeah it's quite complex. hopefully I could make one of those eventually |
| 00:32.46 | brlcad | definitely eventually |
| 00:33.15 | louipc | but on a CNC :D |
| 00:33.33 | brlcad | in the wish list big picture, birail is a relatively high-priority item |
| 00:34.09 | brlcad | yeah.. i don't know how you'd even begin that on a CNC other than cutting out the raw shape |
| 00:34.41 | brlcad | that thing has plates that have been pressed to form, then power welded to the base if I had to guess from the look |
| 01:29.28 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/libbu/log.c: |
| 01:29.28 | CIA-4 | BRL-CAD: put comments on their own line. was going to add code to bu_log() to try and |
| 01:29.28 | CIA-4 | BRL-CAD: detect when there are no ... args, but it does not seem to work (at least on mac |
| 01:29.28 | CIA-4 | BRL-CAD: os x) as va_arg keeps scanning past the end giving the wrong count! |
| 01:31.23 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/include/bu.h: |
| 01:31.23 | CIA-4 | BRL-CAD: add GCC attribute declarations (when we're compiling with GCC of course) that |
| 01:31.23 | CIA-4 | BRL-CAD: help the compiler know that a particular function has printf-style arguments or |
| 01:31.23 | CIA-4 | BRL-CAD: is a function that never returns. add the attributes to the likes of bu_log, |
| 01:31.23 | CIA-4 | BRL-CAD: bu_flog, and bu_bomb. code inspired from bz, thx. |
| 01:37.04 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/ (librt/dg_obj.c librt/wdb_obj.c mged/cmd.c): |
| 01:37.04 | CIA-4 | BRL-CAD: change wdb_tree_cmd() to dgo_tree_cmd() so that we can have access to the |
| 01:37.04 | CIA-4 | BRL-CAD: display geometry (particularly, to get access to the 'who' command). modify the |
| 01:37.05 | CIA-4 | BRL-CAD: behavior of the mged 'tree' command so that if you don't provide it any options, |
| 01:37.05 | CIA-4 | BRL-CAD: it will display the tree for all objects displayed. this is in response and |
| 01:37.07 | CIA-4 | BRL-CAD: inspired by a tcl snippet from david loman that needed this functionality to aid |
| 01:37.09 | CIA-4 | BRL-CAD: in analyzing gqa output. |
| 01:44.26 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/NEWS: mged 'tree' command now shows e'd objects instead of returning a usage error, processing the geometry that is e'd up on the display. |
| 01:47.32 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/tclscripts/helplib.tcl: update the usage to reflect the change |
| 02:10.55 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/ (NEWS src/rt/opt.c): |
| 02:10.55 | CIA-4 | BRL-CAD: john noticed that the -g and -G options were not working on the rt apps anymore |
| 02:10.55 | CIA-4 | BRL-CAD: due to the change to the default aspect from 1.0 to 0.0; and sure enough, he was |
| 02:10.55 | CIA-4 | BRL-CAD: right. the aspect ratio change (made several years ago!) was for the -V view |
| 02:10.55 | CIA-4 | BRL-CAD: aspect option, but obviously had other users in other files that used the value. |
| 02:10.57 | CIA-4 | BRL-CAD: this changes the default back to 1.0 while also adding a sanity check into the |
| 02:10.59 | CIA-4 | BRL-CAD: argument processor to make sure the aspect ratio is always a sane value. |
| 02:13.37 | louipc | brlcad: actually impellers are milled on CNC machines |
| 02:16.17 | brlcad | ah, interesting |
| 02:16.25 | louipc | http://video.google.com/videoplay?docid=4984324514373898629&q=impeller+machining&total=3&start=0&num=10&so=0&type=search&plindex=0 |
| 02:16.27 | brlcad | would be interesting g-code |
| 02:16.34 | louipc | 5-axis |
| 02:16.45 | louipc | it's pretty damned sexy |
| 02:17.43 | starseeker | coool |
| 02:17.46 | louipc | damn I want one of those machines |
| 02:17.47 | brlcad | wow, that's pretty cool |
| 02:18.31 | brlcad | and it followed contours, so it had to have knowledge of the brep seams in the code |
| 02:18.54 | brlcad | pretty nifty/tricky |
| 02:19.22 | louipc | this one is even better http://video.google.com/videoplay?docid=9040285813914114865&q=impeller+machining&total=3&start=0&num=10&so=0&type=search&plindex=1 |
| 02:20.50 | brlcad | somehow, I'm betting there's custom "impeller" software so that it can optimally cut it given how much precision those blades need to have |
| 02:23.20 | louipc | yeah there's some kind of software for that operation |
| 04:07.45 | yukonbob | louipc: very cool |
| 05:27.53 | *** join/#brlcad IriX64 (n=mariodot@bas2-sudbury98-1178015622.dsl.bell.ca) |
| 06:19.25 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/include/opennurbs_ext.h: |
| 06:19.25 | CIA-4 | BRL-CAD: fix a truely bizarre VC8 compilation error (spawning from warning C4346) where |
| 06:19.25 | CIA-4 | BRL-CAD: we either have to get rid of the typdefs or add this typename qualifier. docs |
| 06:19.25 | CIA-4 | BRL-CAD: purport that this is actually compliance required, so go with it. |
| 07:18.21 | *** join/#brlcad Z80-Boy (n=clock@zux221-122-143.adsl.green.ch) |
| 07:44.46 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-002-234.pools.arcor-ip.net) |
| 07:54.58 | *** join/#brlcad starseeker (n=CY@ip72-218-16-62.hr.hr.cox.net) |
| 08:00.30 | CIA-4 | BRL-CAD: 03d_rossberg * 10brlcad/include/opennurbs_ext.h: |
| 08:00.31 | CIA-4 | BRL-CAD: the last change crashed the build with MSVC V6.0 compiler |
| 08:00.31 | CIA-4 | BRL-CAD: made an ifdef for this compiler |
| 08:15.59 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/include/opennurbs_ext.h: try just getting rid of the typedef -- does this work in vc6? |
| 08:33.31 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) |
| 08:49.22 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/include/conf/ (MINOR PATCH): we're already on 7.11 for head, next release after 7.10.4 release will be a minor revision release at 7.12.0. |
| 08:49.31 | brlcad | hey daniel -- did that last opennurbs_ext.h change work for you? |
| 08:49.40 | brlcad | if not, I'll revert it |
| 08:52.16 | d_rossberg | brlcad: I've already fixed it (today), I hape to get rid of the MSVC 6 next year |
| 08:53.09 | brlcad | i know you did, then I commited like two minutes after you :) |
| 08:53.33 | brlcad | got rid of the typedef altogether, if it works, it's even simpler (should work..) |
| 08:55.22 | d_rossberg | brlcad: ah, I havm't seen, I'm not on the commits mailing list (to much "spam"), I'll have a look ... |
| 08:59.11 | brlcad | yeah, that's way too much for lots of folks |
| 08:59.12 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/misc/win32-msvc/Dll/brlcad.def: include the other symbols that are currently being used in ef windows module |
| 08:59.29 | brlcad | could get the digest, but I'm sure you see the change during a cvs up anyways |
| 09:07.39 | CIA-4 | BRL-CAD: 03d_rossberg * 10brlcad/include/opennurbs_ext.h: there was a segment -> BVSegment<BV> replacement missing |
| 09:08.06 | d_rossberg | brlcad: it should work now ;-) |
| 09:12.46 | brlcad | cool :) |
| 09:13.04 | brlcad | ah, wasn't sure if that was one of them |
| 09:13.18 | brlcad | if it was a constructor call or if there was some other routine called segment |
| 09:13.36 | brlcad | since it wasn't qualified as a BVNode<>:: |
| 09:21.00 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-002-234.pools.arcor-ip.net) |
| 09:33.16 | *** join/#brlcad starseeker (n=CY@ip72-218-16-62.hr.hr.cox.net) |
| 10:00.11 | *** join/#brlcad elite01 (n=elite01@195.37.106.60) |
| 10:15.05 | Z80-Boy | brlcad: did you already find an opportunity to find what is taking so much time in the pathological rendering case? |
| 10:15.25 | Z80-Boy | brlcad: it's something in the boolean evaluation - that's called weaver isn't it? |
| 13:20.38 | *** join/#brlcad SWPadnos (n=Me@dsl245.esjtvtli.sover.net) |
| 13:22.37 | ``Erik | blah |
| 15:16.19 | *** join/#brlcad MinuteElectron (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 15:31.28 | *** join/#brlcad Elperion (n=Bary@p548764C6.dip.t-dialin.net) |
| 16:26.12 | CIA-4 | BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/swidgets/scripts/tree.itk: Added an optional doDraw parameter to the insert method. |
| 16:30.03 | CIA-4 | BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/archer/images/Themes/ (6 files in 3 dirs): Minor update. |
| 16:38.09 | CIA-4 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc7/brlcad/brlcad.sln: Added rtg3 |
| 16:42.32 | *** join/#brlcad yukonbob (n=yukonbob@198.235.198.234) |
| 16:53.59 | ``Erik | blehhh |
| 16:59.03 | *** join/#brlcad IriX64_ (n=mariodot@bas2-sudbury98-1177879294.dsl.bell.ca) |
| 17:26.22 | CIA-4 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc7/libtcl/libtcl.vcproj: Updates to accomodate earlier reorganization. |
| 17:39.31 | CIA-4 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc7/libtk/libtk.vcproj: Updates to accomodate earlier reorganization. |
| 17:43.23 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/adrt/isst/master/compnet.c: Include adrt_common.h so ADRT_NAME_SIZE gets defined. Add RCS tag. |
| 17:59.11 | CIA-4 | BRL-CAD: 03brlcad 07STABLE * 10brlcad/src/tclscripts/lib/Dm.tcl: |
| 17:59.11 | CIA-4 | BRL-CAD: fix a nasty startup delay in a particular viewer app that uses the Mged |
| 17:59.11 | CIA-4 | BRL-CAD: megawidget (which in turn uses the Display megawidget, which uses Dm, which had |
| 17:59.11 | CIA-4 | BRL-CAD: a :0 hard-coded). this fixes the bug by making it use the DISPLAY environment |
| 17:59.11 | CIA-4 | BRL-CAD: variable. |
| 18:06.59 | *** join/#brlcad MinuteElectron (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 18:41.59 | *** join/#brlcad minute (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 19:06.21 | *** join/#brlcad minute (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 19:23.59 | *** join/#brlcad iMinute (n=MinuteEl@cpc4-flit1-0-0-cust283.lutn.cable.ntl.com) |
| 19:38.44 | *** join/#brlcad iMinute (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 20:37.20 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-102-239.pools.arcor-ip.net) |
| 20:58.44 | *** join/#brlcad Elperion (n=Bary@p548764C6.dip.t-dialin.net) |
| 21:01.50 | *** join/#brlcad yukonbob (n=yukonbob@whthyt237-51.northwestel.net) |
| 22:38.19 | *** join/#brlcad Z80-Boy (i=clock@77-56-69-92.dclient.hispeed.ch) |
| 22:39.45 | *** join/#brlcad Z80-Boy (i=clock@77-56-69-92.dclient.hispeed.ch) |
