| 01:24.43 | *** join/#brlcad ChanServ (ChanServ@services.) |
| 01:24.43 | *** mode/#brlcad [+o ChanServ] by irc.freenode.net |
| 05:00.41 | *** part/#brlcad thing0 (n=ric@203-59-138-206.perm.iinet.net.au) |
| 06:35.32 | brlcad | Z80-Boy: you ever get the two new profiles up with the -E mcount -E _mcount ? |
| 07:04.19 | *** join/#brlcad louipc (n=louipc@bas8-toronto63-1096667620.dsl.bell.ca) |
| 07:10.46 | Z80-Boy | um... |
| 07:10.49 | Z80-Boy | I can do it |
| 07:11.45 | Z80-Boy | but can yo ureproduce the 210x slowdown in your installation? |
| 07:13.11 | Z80-Boy | Now I get a weird behaviour of the framebuffer window |
| 07:13.27 | Z80-Boy | If I put mouse over it, the whole screen turns black, except that the mouse pointer is visible. |
| 07:13.44 | Z80-Boy | I produced the framebuffer with the same command as before, and it didn't happen. |
| 07:14.01 | Z80-Boy | If I click on the framebuffer after the screen turns black, it goes away. |
| 07:15.24 | Z80-Boy | Interesting is what happens if I put a video play on the desktop background using mplayer -rootwin -fs |
| 07:15.28 | brlcad | focus man |
| 07:15.34 | brlcad | looking to get the profile logs :) |
| 07:15.54 | brlcad | that is the critical next step |
| 07:16.00 | Z80-Boy | When I go with the mouse over the framebuffer, I see video over the whole screen with the two xterm's and the framebuffer cut out like black masks - black rectangles |
| 07:16.10 | Z80-Boy | But why a single command behaves in 3 different ways? |
| 07:16.25 | Z80-Boy | Once raytraces, one crashes, once produces a blackhole framebuffer? |
| 07:16.53 | brlcad | that's two separate bugs |
| 07:17.08 | brlcad | a bug in X11 and a bug in the OpenGL interface in BRL-CAD |
| 07:17.14 | Z80-Boy | I'll upload the gprof logs to the 210x bugreport, right? |
| 07:17.26 | Z80-Boy | Does it mean you have already seen this bug? |
| 07:17.26 | brlcad | sure |
| 07:17.30 | brlcad | yes |
| 07:17.34 | Z80-Boy | hehe :) |
| 07:17.40 | brlcad | you can get around it with --disable-opengl |
| 07:17.45 | brlcad | er, --without-opengl |
| 07:17.55 | Z80-Boy | do you think opengl is a piece of crap? |
| 07:18.02 | brlcad | not in the least |
| 07:18.11 | Z80-Boy | is it designed well? |
| 07:18.14 | brlcad | that just disables the libfb and libdm interfaces that use opengl |
| 07:18.31 | brlcad | doesn't remove *any* functionality in brl-cad, just changes the language it's speaking |
| 07:18.49 | Z80-Boy | Oh a little problem |
| 07:19.08 | Z80-Boy | I reinstalled the systemwide brl-cad with a non-gprof version since I wanted to leave Ronja raytracing overnight |
| 07:19.13 | brlcad | e.g. run rt -F/dev/Xl .. it should not exhibit the black screen |
| 07:19.15 | Z80-Boy | and the gprof version crashes and is slower |
| 07:19.21 | Z80-Boy | now I have to recompile again omg |
| 07:19.34 | Z80-Boy | But - you said yesterday I should try something with make clean ./configure make |
| 07:19.40 | Z80-Boy | I did it and the make was successful |
| 07:19.47 | Z80-Boy | what should have I figured out after? |
| 07:20.02 | Z80-Boy | You said I should look if there's a -lc flag but I don't know where exactly I should look. |
| 07:20.17 | brlcad | ah, yes -- on the linker line for bwish |
| 07:20.32 | Z80-Boy | now I guess I do make clean in the src/bwish right to reinvoke the linker? |
| 07:20.41 | brlcad | make noprod |
| 07:20.48 | brlcad | that removes just the binary |
| 07:20.51 | brlcad | so it will relink |
| 07:21.17 | Z80-Boy | the linker line for bwish begins with libtool is it what you want? |
| 07:22.06 | Z80-Boy | If I run it without --silent, I see there is no -lc passed to the gc |
| 07:22.07 | Z80-Boy | gcc |
| 07:22.15 | Z80-Boy | ~pastebin |
| 07:22.18 | ibot | somebody said pastebin was a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste, or http://rafb.net/paste/, or http://pastebin.com is usually painfully too slow and unresponsive to use, use one of the other pastebin sites, or dpaste.com is a very nice pastebin as well |
| 07:22.20 | brlcad | make LIBTOOL="sh ../../libtool" LIBTOOLFLAGS="" |
| 07:22.27 | Z80-Boy | ~pastebin.bz |
| 07:22.57 | brlcad | (which basically just removes the --silent) |
| 07:23.02 | Z80-Boy | http://pastebin.bzflag.bz/m6f867799 |
| 07:24.02 | brlcad | okay, that's good |
| 07:24.09 | Z80-Boy | Do you need more information? |
| 07:24.36 | brlcad | curious, yes |
| 07:25.32 | brlcad | grep '\$' ../../config.log | head -1 |
| 07:26.03 | Z80-Boy | <PROTECTED> |
| 07:26.29 | brlcad | have you modified your configure.ac file? |
| 07:26.35 | brlcad | or did you set CFLAGS/CPPFLAGS? |
| 07:26.47 | Z80-Boy | clock@kestrel:~/brlcad-vanilla/src/bwish$ echo $CFLAGS |
| 07:26.47 | Z80-Boy | -O3 -fomit-frame-pointer -march=pentium3 -pipe |
| 07:26.54 | Z80-Boy | CPPFLAGS are empty |
| 07:27.14 | Z80-Boy | I didn't change configure.ac myself. I maybe ran autogen.sh which I don't know if it changes configure.ac |
| 07:27.17 | brlcad | yeah, fyi that's no good for doing profile builds -- it can override |
| 07:27.24 | Z80-Boy | aha! |
| 07:27.39 | Z80-Boy | does unset work like export? |
| 07:27.45 | brlcad | but thats also not likely the -lc problem |
| 07:27.49 | Z80-Boy | Or does it unset just for one level of the shells? |
| 07:28.21 | Z80-Boy | I'll try it without the CPPFLAGS |
| 07:29.14 | Z80-Boy | Can you try the two raytraces on your installation? |
| 07:29.19 | Z80-Boy | Do you also get 1:210 difference? |
| 07:29.39 | brlcad | no, I really cannot try them right now |
| 07:29.59 | brlcad | that's why I was asking you for the profiles .. you know, to save some time .. |
| 07:30.36 | brlcad | because that's all I'll be doing too anyways, I really don't doubt that it's slower |
| 07:30.50 | brlcad | the question is just why and what if anything can be done |
| 07:30.55 | Z80-Boy | OK doing export CFLAGS="" |
| 07:31.14 | brlcad | what problem are you working on? |
| 07:31.21 | brlcad | or what do you expect that to change? ... |
| 07:31.36 | Z80-Boy | now I am working on the logs you said |
| 07:31.46 | brlcad | so not the -lc issue |
| 07:31.52 | Z80-Boy | no |
| 07:31.58 | brlcad | okay! |
| 07:32.16 | brlcad | you've got a lot of ADD tendancies sometimes.. :) |
| 07:32.19 | Z80-Boy | and now ./configure --enable-profiling --enable-debug --disable-optimized |
| 07:32.30 | Z80-Boy | Attention Deficiency Disorder? |
| 07:32.44 | brlcad | deficit |
| 07:32.59 | brlcad | or AD/HD |
| 07:34.07 | brlcad | i.e. focus on the task at hand ;) |
| 07:36.18 | Z80-Boy | We can see that in a normal adult without ADHD, the glucose metabolism in the brain is so high that significant amounts of glucose are metabolised as far as 20 centimeters outside the head: http://upload.wikimedia.org/wikipedia/en/5/5a/Adhdbrain.gif |
| 07:36.20 | brlcad | anyways, so the next step, run that configure line, run make, install, run gprof with the -I options, capture output to file |
| 07:37.01 | Z80-Boy | Oh they have a picture of the Hippocrites |
| 07:37.29 | Z80-Boy | "A marked decrease in academic skills such as reading, spelling, or math is common with children who have ADHD." |
| 07:37.34 | Z80-Boy | I had a marked increase |
| 07:37.48 | Z80-Boy | configure: error: `CFLAGS' has changed since the previous run: |
| 07:37.56 | Z80-Boy | now I should run make distclen? |
| 07:38.12 | brlcad | sure, but then you'll need to run autogen.sh again too |
| 07:38.20 | Z80-Boy | they say also rm ../../../../config.cache.openbsd4.0.kestrel.barix.local |
| 07:38.24 | Z80-Boy | but that doesn't work |
| 07:41.54 | Z80-Boy | As with many conditions in the field of psychiatry, ADHD can be explained as a social construct |
| 07:42.03 | Z80-Boy | I don't have a disorder. They have a social construct ;-) |
| 07:43.15 | brlcad | you just spent 10 minutes on a diversion, interleaved with two different sorts of code comments, .. and yet still claim to not have issues with getting distracted? :) |
| 07:43.35 | Z80-Boy | I have problems concentrating both in work and on my personal projects |
| 07:44.05 | Z80-Boy | The only moment when I don't have problems concentrating is if I go to a skatepark, snowboarding or surfing |
| 07:44.45 | Z80-Boy | Although these activities are inherently very boring (just a repetitive motion), I don't have a problem there ;-) |
| 07:45.35 | brlcad | back in a while... look forward to seeing the reports if you can generate them ;) |
| 07:46.02 | brlcad | if anything even just seeing the profile of the slow one might help by itself at this point |
| 07:46.10 | Z80-Boy | now th econfigure passed because I deleted the cache |
| 07:46.16 | brlcad | as just mentioning calloc is very suspicious |
| 07:46.31 | Z80-Boy | did you see my comments about calloc possible to optimize? |
| 07:46.46 | Z80-Boy | In one place you call calloc and then overwrite 99% of the buffer with data |
| 07:46.49 | brlcad | yeah, but they're pointless |
| 07:46.58 | Z80-Boy | so the calloc is used as a brute force method to add a terminator |
| 07:47.03 | brlcad | need a profile that indicates that that is a calloc that matters in the least |
| 07:47.04 | Z80-Boy | Is it a code that is not executed often? |
| 07:47.48 | brlcad | there's no need to guess whether it's executed often, no matter how educated you feel it may be -- that is exactly what a profilers is for |
| 07:48.09 | brlcad | in fact guessing is *usually* wrong and/or misleading |
| 07:50.15 | brlcad | it very well could be the calloc in question, or it could be from a completely different place in the code (more likely but not impossible that it's not) |
| 07:51.37 | Z80-Boy | Hmm looks like I have ADHD |
| 07:51.58 | brlcad | the profiler will conveniently tell us exactly where and how bad the code is called .. good stuff if leveraged |
| 07:53.42 | Z80-Boy | we just add more -E flags? |
| 08:00.02 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-029-074.pools.arcor-ip.net) |
| 08:12.05 | Z80-Boy | Hehe when I rename the brlcad directory to brlcad-profiling during make install then things go astray miserably ;-) |
| 08:23.53 | Z80-Boy | Now it says automake-1.9: ## Internal Error ## |
| 08:24.11 | Z80-Boy | Use of uninitialized value in string eq at /usr/local/bin/automake-1.9 line 4673, <GEN0> line 1. |
| 08:24.33 | Z80-Boy | How do I cvs checkout fresh brl-cad? |
| 08:24.55 | Z80-Boy | Oh it's documented I didn't really expect that |
| 08:32.26 | louipc | man I can't find detail specs on collets :( I wanted to draw some up |
| 09:19.30 | *** join/#brlcad Elperion (n=Bary@p5487475E.dip.t-dialin.net) |
| 09:30.40 | Z80-Boy | brlcad: just got it |
| 09:32.18 | Z80-Boy | sourceforget is slow |
| 09:36.01 | Z80-Boy | Firefox just hangs in the middle of scrolling and prints "loading data from <some damned ad site>" |
| 09:38.23 | minute-ssh | Sourceforge adverts are incredibly annoying, especially given the slow speed the site is at in general. |
| 09:40.34 | Z80-Boy | How can I make my OS return 127.0.0.1 to www.google-analytics.com? |
| 09:41.31 | Z80-Boy | I have lookup file bind still it ignores an entry in /etc/hosts |
| 09:43.38 | minute-ssh | If it is really important you could set up your own DNS system. |
| 09:44.01 | Z80-Boy | I don't want to do that |
| 09:44.07 | Z80-Boy | I just want to know why OpenBSD is a piece of shit |
| 09:44.24 | Z80-Boy | cause man resolv.conf says it should work with lookup file bind and I have this line in /etc/resolv.conf |
| 09:44.24 | minute-ssh | Hosts should work, but you might need to restart\reload something. |
| 09:44.39 | minute-ssh | Google Analytics is slow indeed. |
| 09:45.50 | Z80-Boy | OK I'll reload the system |
| 09:53.34 | *** join/#brlcad Z80-Boy (i=clock@77-56-77-7.dclient.hispeed.ch) |
| 09:54.19 | Z80-Boy | Hehe it works like charm |
| 09:54.30 | Z80-Boy | /etc/hosts: |
| 09:54.32 | Z80-Boy | 0.0.0.0 google-analytics.com |
| 09:54.32 | Z80-Boy | 0.0.0.0 ad.doubleclick.net |
| 09:54.35 | Z80-Boy | /etc/resolv.conf: |
| 09:54.42 | Z80-Boy | lookup file bind |
| 09:54.51 | Z80-Boy | brlcad: put it into your system, you'll save truckloads of time! |
| 09:55.01 | Z80-Boy | and tell all friends! |
| 10:18.45 | *** join/#brlcad minute-ssh (n=MinuteEl@bz.bzflag.bz) |
| 12:20.35 | *** join/#brlcad thing0 (n=ric@203-59-198-159.dyn.iinet.net.au) |
| 12:26.21 | *** join/#brlcad thing1 (n=ric@203-59-198-159.dyn.iinet.net.au) |
| 12:50.54 | *** part/#brlcad thing1 (n=ric@203-59-198-159.dyn.iinet.net.au) |
| 13:14.56 | *** join/#brlcad Z80-Boy (i=clock@77-56-77-7.dclient.hispeed.ch) |
| 17:08.52 | *** join/#brlcad thing0 (n=ric@203-59-180-167.dyn.iinet.net.au) |
| 17:42.12 | *** join/#brlcad nollan (i=nollan@bnc.kostar.5kr.hos.egn.se) |
| 17:43.22 | nollan | hey folks. i just discovered brlcad. i added it from freebsd ports w/o any problems, however when i run mged.sh it initialized and then detaches. but i get no windows.. |
| 17:44.44 | nollan | any suggestions what i should do? |
| 17:46.19 | yukonbob | nollan: what does the mged.sh file contain? I just run an mged binary.... |
| 17:46.43 | louipc | yeah I don't have an mged.sh on my system |
| 17:46.59 | louipc | but I'm in linux and I built BRLCAD myself |
| 17:48.12 | nollan | yukonbob, it sets variables like path and brlcad home path.. then it runs mged binary |
| 17:52.37 | louipc | try inputting the commands without the script and see what happens |
| 17:52.48 | nollan | how? |
| 17:52.54 | nollan | [frodo]-[/home/nollan]-$ /usr/local/brlcad/bin/mged |
| 17:52.54 | nollan | Initializing and backgrounding, please wait...Detached |
| 17:52.54 | thing0 | magic |
| 17:53.21 | nollan | where should i put the commands? |
| 17:53.37 | nollan | there is no window nor dialog boxes, nothing |
| 17:53.44 | louipc | yeah mged is supposed to detach |
| 17:53.53 | yukonbob | nollan: try running a [s|k|p]trace on the binary and see what it shows... |
| 17:54.31 | louipc | nollan: try to give it a model on the command line maybe... |
| 17:54.44 | nollan | to open you mean |
| 17:54.47 | louipc | yea |
| 17:55.10 | louipc | there's some in brlcad/share/db |
| 17:55.23 | louipc | i'm not sure what the exact path would be on your system ... |
| 17:55.32 | nollan | ok i'll try |
| 17:55.39 | nollan | ktracing now |
| 17:57.01 | nollan | hmm how do i check the ktrace.out file for errors? |
| 17:57.10 | yukonbob | kdump | less |
| 17:57.44 | nollan | ok |
| 17:59.55 | nollan | it says something about /etc/libmap.conf |
| 18:00.30 | yukonbob | nollan: maybe "kdump > mytmpfile" and post the results to a pastebin... |
| 18:01.12 | nollan | ok, but it's huge |
| 18:01.35 | yukonbob | ...then just that last relevant bits? |
| 18:02.29 | yukonbob | you know what I mean when I say "pastebin" right? Don't flood this channel w/ a cut/paste... |
| 18:03.03 | yukonbob | ~pastebin |
| 18:03.04 | ibot | from memory, pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca, or http://channels.debian.net/paste, or http://rafb.net/paste/, or http://pastebin.com is usually painfully too slow and unresponsive to use, use one of the other pastebin sites, or dpaste.com is a very nice pastebin as well |
| 18:03.15 | nollan | http://pastebin.ca/735505 |
| 18:03.32 | nollan | sure i've used it before. lots of times.. |
| 18:03.46 | nollan | mostrly for reading thou |
| 18:05.40 | yukonbob | ? I can't see any signs that mged quit |
| 18:06.07 | yukonbob | ps aux | grep mged |
| 18:06.18 | nollan | i'm browsing throu it now and theres a lot of libraries not found... |
| 18:06.27 | nollan | i guess it's a dependency issue |
| 18:06.53 | nollan | yeah it starts, but gives me no windows or dialogboxes |
| 18:07.04 | yukonbob | ports should handle all that -- and mged would complain on the cmdline if it couldn't find what it wanted... |
| 18:07.23 | nollan | k |
| 18:07.36 | yukonbob | the 'not found' issues might be first (or second, etc) attempts to find what it's looking for, until it finds them later... |
| 18:07.49 | nollan | yeah. i guess |
| 18:08.09 | yukonbob | what window manager are you running? |
| 18:08.11 | nollan | i didn't compile it thou -i pkg_add -r it |
| 18:08.14 | nollan | fluxbox |
| 18:09.51 | yukonbob | nollan: can you run bwish? |
| 18:10.16 | nollan | dunno is it included in brlcad? |
| 18:10.22 | yukonbob | try it and see ;) |
| 18:10.35 | nollan | i got a dialog |
| 18:10.36 | yukonbob | (yes, it is part of brlcad) |
| 18:10.42 | nollan | and a window |
| 18:10.53 | nollan | and a bwish> - prompt |
| 18:10.54 | yukonbob | are you running Xinerama? |
| 18:11.04 | yukonbob | ( or some dual-head monitor setup?) |
| 18:11.09 | nollan | nope, never heard of it |
| 18:11.15 | nollan | nope no dualhead |
| 18:12.10 | nollan | w8 i'll try a thing brb |
| 18:20.30 | nollan | it works now. i had to remove some options form xorg.conf |
| 18:20.34 | nollan | thanks |
| 18:20.49 | yukonbob | nice! |
| 18:20.51 | yukonbob | have fun :) |
| 18:21.08 | nollan | thanks! i'm totally new.. |
| 18:21.10 | nollan | :) |
| 18:21.29 | louipc | odd what options were those? |
| 18:21.45 | yukonbob | DontShowMged = YES |
| 18:23.56 | nollan | haha ahh, experimental stuff. |
| 18:24.40 | yukonbob | paste them here -- might come up again for somebody else... |
| 18:24.49 | yukonbob | "here" == pastebin ;) |
| 18:24.54 | nollan | pageflips agpmode depthmoves renderaccel |
| 18:25.03 | yukonbob | that's it? |
| 18:25.05 | nollan | ati driver |
| 18:25.07 | nollan | yeah |
| 18:25.10 | yukonbob | hrmm. |
| 18:35.42 | nollan | however when i load examples - they are all empty |
| 18:36.25 | yukonbob | 'ls' to see lists of 'objects, regions, groups' ,etc, and 'e [somename]' to show it... |
| 18:38.07 | yukonbob | there are a few .pdfs you'll be interested in -- a 3-page cheat for mged, and a Tutorial "Principles of Effective Modelling", esp... |
| 18:38.09 | louipc | nollan: if you load the file via command line then it should display automatically |
| 18:40.02 | yukonbob | nollan: ftp.brlcad.org/pub/BRL-CAD/Documentation |
| 18:40.05 | louipc | nollan: type tops in mged to see the top level parts of the model |
| 18:40.35 | *** join/#brlcad Z80-Boy (i=clock@217-162-108-140.dclient.hispeed.ch) |
| 18:41.25 | nollan | mged> load /usr/local/brlcad/share/brlcad/7.8.2/db/demo.g |
| 18:41.25 | nollan | Error: couldn't load file "/usr/local/brlcad/share/brlcad/7.8.2/db/demo.g": /usr/local/brlcad/share/brlcad/7.8.2/db/demo.g: invalid file format |
| 18:41.31 | louipc | nollan: use draw or e in mged to display the objects |
| 18:41.35 | louipc | !! |
| 18:42.25 | nollan | i guess i'm off to the docs |
| 18:42.26 | nollan | :) |
| 18:42.52 | louipc | you're using a really old version too :P |
| 18:43.06 | nollan | ok. that was the one in ports |
| 18:43.17 | yukonbob | nollan: what version is it? |
| 18:43.28 | yukonbob | oh -- 7.8.2... |
| 18:43.50 | nollan | pkg_info | grep brl |
| 18:43.51 | nollan | brlcad-7.8.2 CSG modelling system from the US Ball |
| 18:46.40 | Z80-Boy | oh yeah US Ball |
| 18:46.52 | Z80-Boy | let's play US Ball! |
| 18:47.09 | yukonbob | US Ballroom Dancing Laboratory |
| 19:05.06 | *** join/#brlcad yukonbob (n=yukonbob@198.235.198.234) |
| 19:35.08 | brlcad | heh, ball |
| 19:35.31 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/conv/Makefile.am: include the intaval-g.py script in the dist |
| 19:35.43 | brlcad | Z80-Boy: hm, dunno what you changed, but those profiles are useless |
| 19:35.51 | brlcad | they didn't have any numbers |
| 19:37.20 | Z80-Boy | brlcad: did you try to reproduce the rendering times? |
| 19:44.10 | brlcad | nollan: load isn't the right command, that's the command to load tcl/tk extensions |
| 19:44.16 | brlcad | opendb would be the command |
| 19:45.23 | brlcad | and brl-cad .g files are "multi-geometry" files with potentially many models per file, so there's no "default" object that will display when you open a .g file -- you can either use the geometry browser or run the "tops" command to see the top-level objects that you can display |
| 19:46.30 | nollan | ah. right |
| 20:04.16 | *** join/#brlcad luckyshot (n=lucky@CABLE-72-53-76-204.cia.com) |
| 20:04.24 | luckyshot | Hey guys, |
| 20:04.32 | luckyshot | Does anyone in here use UGS NX series? |
| 20:25.38 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/ (NEWS src/fb/pix-fb.c): accept karel's sf patch 1802016, which provides a -p option to pix-fb causing the application to pause the specified number of seconds before exiting and closing the framebuffer. |
| 20:27.25 | brlcad | luckyshot: i've used it before |
| 20:27.39 | brlcad | this is certainly not an nx support channel though ;) |
| 20:28.10 | brlcad | if you want to fund the same level that you give to UGS to BRL-CAD, then I'd be glad to help ya though ;) |
| 20:56.09 | *** join/#brlcad Elperion (n=Bary@p5487475E.dip.t-dialin.net) |
| 21:01.08 | luckyshot | lol |
| 21:01.26 | luckyshot | bclcad, just wanted to know why nx5 files wont open in nx4 |
| 21:01.29 | luckyshot | its very frustrating |
| 21:36.37 | *** join/#brlcad Bariton (n=Bary@p548775CF.dip.t-dialin.net) |
| 21:39.06 | ``Erik | erm, cuz the format changed? |
| 21:39.16 | ``Erik | they, uh, added shit? |
| 21:45.34 | luckyshot | file format is the same |
| 21:46.04 | ``Erik | then why is it 5 instead of 4.1? :D *duck* |
| 21:46.30 | ``Erik | this ain't a unigraphics help channel, I've never used ug... tried installing it once, that's it:) |
| 21:58.09 | brlcad | if the format is the "same" then it's not an "nx5 file", it's something else .. they break compatibility just like most other major apps when there's a new major version |
| 22:01.11 | luckyshot | i guess |
| 22:01.25 | luckyshot | ms word still is backwrds compatible with older versions |
| 22:06.24 | brlcad | only to an extent |
| 22:06.31 | brlcad | you can create new versions that are not |
| 22:06.42 | brlcad | word simply hasn't *had* a new release in years |
| 22:07.25 | brlcad | they used to break compatibility all the time, 95 not working in 97, 2000 not working on either, etc |
| 22:07.40 | luckyshot | oh okay |
| 22:08.00 | luckyshot | is brl cad easy to use? |
| 22:08.06 | brlcad | not really |
| 22:08.17 | brlcad | at least not until you've gone up the learning curve |
| 22:08.22 | luckyshot | nx4 and 5 are extrmely simple |
| 22:08.27 | brlcad | it's easy to use, but not easy to learn |
| 22:08.29 | luckyshot | are they of the same level of difficulty? |
| 22:09.04 | brlcad | give it a try, some find it exceptionally familiar |
| 22:09.17 | brlcad | the tutorial book on the website might give you an idea of what to expect |
| 22:09.20 | luckyshot | definitely |
| 22:09.22 | luckyshot | does it work in vista? |
| 22:09.29 | brlcad | http://ftp.brlcad.org/VolumeII-Introduction_to_MGED.pdf |
| 22:09.53 | brlcad | only one of brl-cad's 400+ tools, but one of the most prominent and with a gui |
| 22:10.21 | brlcad | I don't know that anyone has even tried under vista |
| 22:10.33 | luckyshot | ok |
| 22:12.25 | brlcad | howdy bob |
| 22:13.18 | yukonbob | brlcad: hey -- what's shaking? |
| 22:47.05 | brlcad | howdy yukonbob -- most of the cfd/fea tools are indirect or generic |
| 22:47.27 | brlcad | like being able to extract a mesh of given geometry, or the Cubit exporter, or voxelization of geometry, etc |
| 23:00.28 | yukonbob | note: anybody who posts links to images of havoc is subject to being banned from #brlcad... |
| 23:00.46 | yukonbob | ;) |
| 23:03.40 | CIA-4 | BRL-CAD: 03starseeker * 10brlcad/doc/book/VolumeII.xml: Markup through Lesson 11 |
| 23:03.48 | poolio_ | Woah, has Irix disappeared? |
| 23:04.55 | starseeker | Wouldn't that be an awesome tag line - BRL-CAD - able to simulate moon shots! |
| 23:05.51 | yukonbob | poolio_: :) ... one day some time ago I threatened to reach through the internet and take his modem away if he posted another havoc image, and haven't seen another since... |
| 23:07.11 | brlcad | starseeker: heh, is their .g support supposed to be for BRL-CAD or something else? |
| 23:07.39 | brlcad | very possible they might have support for v4 .g format from interactions over a decade ago |
| 23:07.39 | starseeker | I'm sure it's for something else - I just noticed it in their list of "supported" file extensions |
| 23:07.48 | starseeker | oh, really? |
| 23:08.01 | starseeker | huh - is there a v4 .g file around somewhere? |
| 23:08.39 | brlcad | run dbupgrade -r on a v5 |
| 23:09.09 | starseeker | righto... |
| 23:10.54 | starseeker | Interesting... |
| 23:11.09 | starseeker | It does add it as an object, but doesn't display anything |
| 23:11.37 | starseeker | Says it is a polygonal mesh with zero cells and zero points |
| 23:17.34 | Z80-Boy | starseeker: I guess they had, but they had to use a boolean operator and that made it 210x slower so they couldn't make it into the launch so they dropped it |
| 23:18.21 | Z80-Boy | Or they realized that when they mirror the foam, it gets 3 times as big and changes shape ;-) |
| 23:18.23 | brlcad | sounds like it's a busted parser |
| 23:18.34 | Z80-Boy | brlcad: sorry, just joking :) |
| 23:18.41 | starseeker | in paraview? |
| 23:19.02 | brlcad | Z80-Boy: I have the profiles done here now |
| 23:19.08 | Z80-Boy | brlcad: and the ratio? |
| 23:19.10 | brlcad | *with* useful results :) |
| 23:19.12 | Z80-Boy | 1:210? |
| 23:19.18 | brlcad | dude, get over the ratio |
| 23:19.24 | Z80-Boy | more than 210? |
| 23:19.38 | brlcad | i already told you i didn't doubt that one is slower |
| 23:19.58 | Z80-Boy | man you'll become a hero for speeding up the U.S. Army development 200 times |
| 23:20.05 | brlcad | and it would be expected that it's a little slower, but not necessarily drasticly |
| 23:20.21 | Z80-Boy | That's good |
| 23:20.25 | brlcad | two orders implies something else is wrong |
| 23:20.30 | Z80-Boy | because now you can poke at it and don't have to ask me |
| 23:20.51 | Z80-Boy | which function does consume it? |
| 23:20.59 | brlcad | it's not good, it means i'm not working on other things that are really begging for attention as well :P |
| 23:21.28 | brlcad | so no time was saved in the end |
| 23:21.38 | Z80-Boy | which function did gprof say consumes the most? |
| 23:21.38 | brlcad | it doesn't help anyone for everything to be in my hands :) |
| 23:22.02 | Z80-Boy | is it calloc? |
| 23:22.24 | brlcad | I'd suggest working more with gprof yourself to get the answer to that, so next time you can generate the profile :) somthing went wrong in those steps .. maybe an openbsd gprof problem |
| 23:23.36 | Z80-Boy | which function is it? |
| 23:24.43 | Z80-Boy | I have to go sleeping if I don't want to be zombie tomorrow |
| 23:24.44 | Z80-Boy | good night |
| 23:24.47 | brlcad | cya |
| 23:25.08 | brlcad | he can be a selfish lil punk sometimes |
| 23:25.23 | brlcad | good kid, lots of good efforts, but impatient to no end |
| 23:27.12 | brlcad | no |
| 23:27.14 | brlcad | hell no :) |
| 23:27.17 | brlcad | not yet at least |
| 23:27.37 | yukonbob | (that you know of) |
| 23:27.45 | brlcad | they're not very productive to my coding and lifestyle habits right now, maybe in a few years :) |
| 23:27.53 | starseeker | "who needs kids when one has Z80-Boy on IRC" |
| 23:36.28 | brlcad | mmm.. lisp experience could certainly come in handy ;-) |
| 23:36.51 | brlcad | for the new modeler, lisp was on my list of languages we need to support near the beginning |
| 23:37.03 | brlcad | simply due to the autocad crowd |
| 23:38.00 | brlcad | probably lisp and python for non-interactive; and tcl and bash for interactive |
| 23:38.19 | yukonbob | tcl inter + non-inter |
| 23:38.27 | yukonbob | source "mycoolfile.tcl" |
| 23:38.28 | yukonbob | ;) |
| 23:38.39 | brlcad | yeah, tcl is one of the few that can probably do both |
| 23:39.12 | yukonbob | this talk of py/tcl/lisp* bring us back to the idea of SWIG |
| 23:39.15 | starseeker | Doing both is something I see more in research languages - they like the REPL prompts |
| 23:39.16 | brlcad | and not feel like ass in the process ;) |
| 23:40.01 | brlcad | yeah, but then the idea of an ml prompt make me shudder |
| 23:40.07 | starseeker | Lisp is really good at it, given a good compiler - there are some amusing stories about early web servers in lisp where the devs would get a support call, fix it on the phone in the running image, and then ask the customer to do a page refresh :-) |
| 23:40.26 | yukonbob | *given a good compiler |
| 23:40.42 | starseeker | There are a few. Allegro I think is pretty good, and sbcl on the free side |
| 23:41.03 | starseeker | Clisp is kinda dog slow, but has the advantage of running everywhere under the sun |
| 23:41.56 | starseeker | Scheme is regarded as a type of Lisp |
| 23:42.17 | starseeker | Scheme stays small and doesn't specify all that much - it lets extensions fight it out |
| 23:43.08 | starseeker | Common Lisp is the opposite - it's a huge spec |
| 23:43.11 | yukonbob | starseeker: right -- but if somebody is used to all the facilities of Common Lisp, scheme is not Common Lisp.. |
| 23:43.16 | yukonbob | :) |
| 23:43.21 | starseeker | True :-) |
| 23:43.28 | starseeker | but the core language principles are the same |
| 23:43.40 | yukonbob | (are (they (indeed))) |
| 23:43.50 | starseeker | Scheme is "a lisp", Fortran most definitely is not |
| 23:44.13 | yukonbob | could wait for Paul Graham to finish Arc. |
| 23:44.16 | brlcad | hm, so calloc didn't even make a blip on the profile |
| 23:44.30 | starseeker | yukonbob: Only if you can code from the grave |
| 23:44.37 | brlcad | looks like 98.9% of the extra time is spent doing the boolean evaluation, exactly like john said |
| 23:44.42 | yukonbob | ?Is he dead |
| 23:45.44 | brlcad | yukonbob: a brlcad major mode would be freaking awesome :) |
| 23:46.06 | starseeker | yukonbob: By the time Arc is done, you probably will be ;-) |
| 23:46.13 | yukonbob | starseeker: haha |
| 23:46.23 | brlcad | that'd be a great "summer of code" project :) |
| 23:47.00 | starseeker | Emacs has a lot of... erm... personality |
| 23:47.28 | yukonbob | brlcad: one of first things I do in mged is "array set env [list EDITOR xemacs]" |
| 23:48.11 | yukonbob | ... could skip that and just use xemacs -- that _would_ be very cool... |
| 23:48.12 | brlcad | hm, can't you just set env(EDITOR) xemacs ? |
| 23:48.59 | yukonbob | brlcad: could "env EDITOR=xemacs mged", or set EDITOR in my .profile, sure... |
| 23:52.13 | yukonbob | heh: http://homepage.ntlworld.com/ajwatson/cat-powered.gif |
| 23:53.45 | starseeker | now THAT'S hardcore ;-) |
| 23:54.10 | brlcad | yukonbob: no, I mean as tcl syntax inside mged |
| 23:54.30 | brlcad | is there any difference between "array set env [list EDITOR xemacs]" and "set env(EDITOR) xemacs"? |
| 23:54.40 | brlcad | i've never seen the array form |
| 23:55.25 | yukonbob | brlcad: you're right... |
| 23:56.00 | brlcad | wasn't trying to be right, it's informative to me too :) |
| 23:56.13 | yukonbob | you're still correct ;) |
| 23:56.24 | yukonbob | both work |
| 23:56.32 | brlcad | wonder if there is any technical difference |
| 23:56.56 | yukonbob | re: emacs -- are you thinking of completely working from w/i emacs, and have it interface w/ rt? |
| 23:58.04 | brlcad | sure |
