| 06:00.57 | *** join/#brlcad Twingy (n=justin@74.92.144.217) |
| 08:10.35 | *** join/#brlcad Z80-Boy (i=clock@77-56-77-128.dclient.hispeed.ch) |
| 09:23.51 | *** join/#brlcad alex_joni (n=juve@emc/board-of-directors/alexjoni) |
| 10:14.27 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) |
| 10:15.48 | Axman6 | hey brlcad, i was wondering if you'd found a fix for my circular dependancy problem? |
| 10:22.14 | Axman6 | i'll be back, testing things and need to reboot |
| 10:23.10 | Z80-Boy | I'll be back... --Arnold Schwarzenegger |
| 10:26.41 | Axman6 | heh :) |
| 11:04.44 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) |
| 11:48.13 | *** join/#brlcad Elperion (n=Bary@p5487683A.dip.t-dialin.net) |
| 14:24.36 | Z80-Boy | brlcad: does brlcad also calculate polarization, Brewster angle etc.? |
| 15:19.36 | CIA-28 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc8/libtcl/libtcl.vcproj: Added a call to build the missing include/conf files. |
| 17:02.07 | *** join/#brlcad Z80-Boy (i=clock@77-56-77-128.dclient.hispeed.ch) |
| 17:02.50 | Axman6 | brlcad: you around? |
| 18:27.53 | CIA-28 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc8/tclsh/library/installTree.tcl: No longer need to create the missing include/conf files. Added code to create tclIndex files. Added the shareDir variable to simplify things a bit. |
| 18:30.42 | CIA-28 | BRL-CAD: 03bob1961 * 10brlcad/misc/win32-msvc8/ (4 files in 4 dirs): Mods for getting the installTree script to create tclIndex files. |
| 18:37.01 | brlcad | Axman6: yeah, sorry .. was busy coding ;) |
| 18:37.10 | Axman6 | heh :) |
| 18:37.15 | Axman6 | so i see |
| 18:37.35 | brlcad | that's bob, different dev :) |
| 18:37.40 | brlcad | not committed my stuff yet |
| 18:37.42 | Axman6 | i'm lucky i'm still awake at 5:30AM then i guess :P |
| 18:37.43 | CIA-28 | BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/ami.tcl: Created the make_tclIndex proc. |
| 18:37.49 | brlcad | wow |
| 18:37.55 | Axman6 | ah right, didn't check the dev |
| 18:38.15 | Axman6 | yeah, woke up at 3:30 yesterday, so hard to sleep |
| 18:38.25 | Axman6 | the sun's already up -_- |
| 18:38.33 | brlcad | :) |
| 18:39.12 | brlcad | so one thing to try, just to see if it helps with the problem is to move the libpng out of the way that is in /usr/X11/lib |
| 18:39.35 | brlcad | if that works, then it can be put back and can work on coercing the search path ordering |
| 18:39.51 | Axman6 | YEAH OK |
| 18:40.15 | Axman6 | whoops |
| 18:43.23 | Axman6 | http://pastie.textmate.org/124117 |
| 18:44.35 | brlcad | hmm |
| 18:46.58 | brlcad | here's an idea, might be enough to just avoid the whole problem by disabling tcl's core foundation interface |
| 18:47.13 | brlcad | try adding --disable-corefoundation to configure |
| 18:47.50 | brlcad | pastebin the entire configure output if it'll fit |
| 18:47.50 | Axman6 | ok |
| 18:54.49 | Axman6 | http://pastie.textmate.org/124120 |
| 18:54.59 | Axman6 | hopefully |
| 18:55.25 | Axman6 | yeah that got it all |
| 19:00.57 | brlcad | heh |
| 19:01.06 | Axman6 | hmm, running make 0j2 really does make a difference |
| 19:01.18 | Axman6 | -j2* |
| 19:01.18 | brlcad | fwiw, --with-all doesn't do anything |
| 19:01.33 | Axman6 | hmm, maybe that was something else |
| 19:01.58 | Axman6 | it was suggsted to me somewhere. anyway |
| 19:02.02 | brlcad | there is a --enable-all option |
| 19:02.15 | Axman6 | probably bzflag heh |
| 19:02.18 | brlcad | but your results are fine, dont' really need it |
| 19:02.38 | brlcad | bz doesn't have a --with-all either ;) |
| 19:03.26 | brlcad | our configure script actually is a decent benchmark of overall I/O performance :) |
| 19:03.27 | Axman6 | bzflag the game? |
| 19:03.30 | brlcad | yeah |
| 19:03.40 | Axman6 | hmm, wonder what it was then... |
| 19:03.45 | Axman6 | oh well, not important ; |
| 19:04.05 | Axman6 | haha, awesome |
| 19:04.28 | Axman6 | brl-cad and bzflag were the two things i was trying to get compiled yesterday :P |
| 19:04.35 | brlcad | cad and bz are the two I spend most of my time on :) |
| 19:05.36 | brlcad | as witnessed by the ohloh metrics .. :) |
| 19:05.56 | Axman6 | well while i'm at it, when i compiled bzflag, it would die after about a minute. but i believe thT's an SDL problem |
| 19:06.02 | brlcad | so have you tried the build with the --disable-corefoundation flag? |
| 19:06.16 | Axman6 | yeah, it's building now |
| 19:06.18 | brlcad | how'd you build sdl? |
| 19:06.27 | Axman6 | using macports |
| 19:06.32 | brlcad | what version? |
| 19:06.41 | Axman6 | 1.2.12? that sound right? |
| 19:06.45 | brlcad | yeah |
| 19:06.50 | brlcad | that's nfg :) |
| 19:06.54 | Axman6 | apparently 1.2.11 works fine |
| 19:06.56 | brlcad | needs to be .11 |
| 19:07.01 | Axman6 | heh, yeah |
| 19:07.08 | brlcad | or you have to update the SDLmain.m in BZ |
| 19:07.13 | brlcad | to the new one for .12 |
| 19:09.00 | Axman6 | don't happen to know what needs changing off the top of your head do you? |
| 19:10.46 | Axman6 | :P |
| 19:10.58 | Axman6 | i should check it out of svn and build it |
| 19:11.31 | brlcad | we just posted a new release |
| 19:11.42 | brlcad | svn head is the next (incompatible) dev version |
| 19:11.44 | Axman6 | .10 one? |
| 19:11.48 | brlcad | yeah |
| 19:11.50 | brlcad | that's the latest |
| 19:11.59 | Axman6 | yeah that didn't build for me |
| 19:12.09 | Axman6 | well it did, but had the sdl problem |
| 19:12.42 | brlcad | yeah, easiest solution is to just install the .11 SDL framework into /Library/Frameworks and run xcodebuild |
| 19:14.25 | brlcad | otherwise, you have to replace the SDLmain files in src/platform/MacOS (it's usually bundled in a dev download from SDL folks) |
| 19:14.58 | Axman6 | righto |
| 19:15.14 | Axman6 | well brl-cad first i think, bzflag later ;) |
| 19:15.27 | Axman6 | i'll probably find brl-cad more fun anyway |
| 19:16.15 | Axman6 | downloaded someone's unofficial intel build of bz... died instantly every time :( |
| 19:17.24 | Axman6 | "-DPNG_NO_MMX_CODE" what's that do exactly? |
| 19:18.17 | Axman6 | well really, what do the -D flags to gcc do? |
| 19:21.27 | Axman6 | this gets the fans giong quite effectively :| |
| 19:26.49 | brlcad | -D flags set preprocessor defines |
| 19:29.39 | Axman6 | bleh, i never got that far into my C++ book ;) |
| 19:29.50 | brlcad | which are generally symbols that can modify compilation behavior, turn things on/off but are completely driven by the code |
| 19:30.29 | Axman6 | bleh, same error again |
| 19:30.30 | brlcad | -DPNG_NO_MMX_CODE is as if you added "#define PNG_NO_MMX_CODE 1" to the top of every file being compiled |
| 19:31.50 | brlcad | exact same error? |
| 19:32.02 | Axman6 | yeah |
| 19:32.08 | Axman6 | afaict |
| 19:33.26 | Axman6 | http://tinypic.com/view.php?pic=6xkfj3p&s=1 <-- eases compilation pain |
| 19:34.17 | brlcad | your cat? :) |
| 19:34.28 | brlcad | or paris' top falling off |
| 19:35.37 | Axman6 | heh, yeah my cat |
| 19:37.28 | Z80-Boy | brlcad: first piece of optic which was prototyped using BRL-CAD |
| 19:37.59 | brlcad | for what it's worth, it'd be willing to interactively debug this problem over ssh if you're willing to set up a temp account -- otherwise, next thing is to edit the top-level configure.ac file and comment out the corefoundation block |
| 19:38.46 | brlcad | Z80-Boy: to answer your question, there's no polarization in liboptical -- maybe in libmultispectral, but I don't believe so |
| 19:38.47 | Z80-Boy | brlcad: http://ronja.twibright.com/3d/prismascope.g render in rear view |
| 19:39.15 | Z80-Boy | brlcad: it's a telescope which magnifies 4 times and consists only from 4 prisms, no curved optical interfaces ;-) |
| 19:40.33 | Axman6 | brlcad: yeah i would let you ssh in, but nat would make it a problem. which is a pitty, because leopard comes with a nice temp account feature. erases all files when you log out |
| 19:41.23 | brlcad | nat isn't forwarding port 22? |
| 19:41.33 | Axman6 | well not to this machine |
| 19:41.48 | brlcad | could run ssh manually on another port being forwarded for single session |
| 19:41.51 | Axman6 | to the G5 |
| 19:42.58 | brlcad | if you want to keep plugging away at it, edit configure.ac and look for the block that starts with this line: |
| 19:43.01 | brlcad | AC_MSG_CHECKING([whether -framework CoreFoundation is available]) |
| 19:43.04 | Z80-Boy | brlcad: ehm forgot the name of the toplevel to render... "scene" |
| 19:43.07 | Axman6 | i'll try the commenting out of core foundation stuff first. |
| 19:43.17 | brlcad | and comment it all out (prefix each line with #) |
| 19:43.33 | brlcad | Z80-Boy: already seen it :) |
| 19:44.07 | brlcad | http://bzflag.bz/~sean/tmp/zoom.png |
| 19:44.45 | Z80-Boy | brlcad: you know how it works? |
| 19:44.46 | Axman6 | Z80-Boy: how's that work? |
| 19:44.53 | Axman6 | heh |
| 19:45.51 | Z80-Boy | Axman6: I just phoned God and asked him if he could make prism magnify from now |
| 19:46.04 | Z80-Boy | because prisms are easier to make than lenses |
| 19:46.12 | Axman6 | ah right, yeah he's good like that sometimes |
| 19:46.29 | louipc-bot | especially big lenses eh? |
| 19:46.46 | Z80-Boy | Axman6: if you draw the prism picture from the schoolbook you can see the beam inside the prism is squished |
| 19:47.00 | Z80-Boy | so if you cut the prism in half into a smaller prism the squished beam comes out |
| 19:47.13 | Z80-Boy | and there is some obscure theorem meaning that if you squish the beam it magnifies |
| 19:47.23 | Axman6 | heh |
| 19:47.26 | Z80-Boy | It also does a spectra decomposition so you need to use a pair to cancel that out |
| 19:47.43 | Z80-Boy | and another pair of prisms to do the magnification in the other dimension because the picture is 2D |
| 19:47.50 | Z80-Boy | so with 4 prisms you can have a telescope |
| 19:47.54 | louipc-bot | o.O |
| 19:47.57 | Axman6 | i'll have to check your file when i get brl-cad compiled ; |
| 19:48.09 | Z80-Boy | No need for complicated grinding of lenses - just rub two pieces of glass against each other with some grit |
| 19:48.20 | Z80-Boy | or make like aquarium from window pane float glass ;-) |
| 19:48.28 | brlcad | Z80-Boy: yes, I know how prisms work.. and I saw how you oriented them in the model -- pretty creative really |
| 19:49.03 | Z80-Boy | brlcad: I actually saw the pair as an anamorphotic pair to fix fabry perror laser output beam shape |
| 19:49.11 | Z80-Boy | perrot |
| 19:49.27 | louipc-bot | actually don't the most powerful telescopes use curved mirrors? |
| 19:49.35 | Z80-Boy | yes |
| 19:49.49 | Z80-Boy | but you need exact parabolic shape |
| 19:49.53 | Axman6 | cassigrain telescopes |
| 19:49.56 | Z80-Boy | that's more complicated than an exact flat shape |
| 19:50.04 | louipc-bot | yeah |
| 19:50.07 | Axman6 | as demonstrated by the hubble |
| 19:50.17 | Z80-Boy | but if you tilt my model the picture gets distorted |
| 19:50.19 | Axman6 | which they screwed up originally |
| 19:50.24 | Z80-Boy | it seems to work only for narrow field of view |
| 19:50.29 | Z80-Boy | But for laser beams it's great |
| 19:50.36 | louipc-bot | :D |
| 19:50.48 | Z80-Boy | or for burning stuff with the sun... |
| 19:50.53 | Z80-Boy | that's also narrow field of view |
| 19:50.58 | Z80-Boy | but a fresnel lens is lighter |
| 19:51.15 | brlcad | you can create rather exact parabolic shapes using the epa if you wanted to, but have fun grinding it ;) |
| 19:51.16 | Z80-Boy | It just demonstrates the people in the middle ages were stupid, they could make themselves crude telescopes from grinded gems |
| 19:51.52 | Z80-Boy | they didn't need a curved surface for that |
| 19:53.31 | Z80-Boy | Or they could stick 4 prisms in front of the hubble to double the input aperture for free |
| 19:53.59 | Z80-Boy | just get some window panes into the sky and then fill with water |
| 19:54.31 | Axman6 | but that would severly reduce the amount of light getting through to the sensors |
| 19:54.32 | Z80-Boy | this way they could build an integralactic megagigatelescope to concentrate significant amount of solar power and then they could zap whole planets in the sky |
| 19:54.33 | brlcad | http://bzflag.bz/~sean/tmp/prisms.png |
| 19:55.02 | Z80-Boy | Axman6: they can coat with anti reflex |
| 19:55.11 | Axman6 | hdh |
| 19:55.15 | Axman6 | heh* |
| 19:55.59 | Axman6 | well water would be very bad. the reason they have the hubble... in space.. is because water absorbs infrared light |
| 19:56.16 | Z80-Boy | aha |
| 19:56.28 | Axman6 | brlcad: so you do your dev work on a mac? |
| 19:56.38 | brlcad | most of the time |
| 19:56.43 | brlcad | depends what I"m doing |
| 19:57.17 | Axman6 | argh! same error again! |
| 19:57.21 | brlcad | it's one of the easiest platforms for general work, less fuss, great set of dev tools (profiler, memory tests, etc) |
| 19:57.48 | brlcad | Axman6: did you clean out the previous build? |
| 19:57.52 | Axman6 | yeah |
| 19:57.59 | brlcad | huh |
| 19:58.10 | Axman6 | but you can neverr be sure, so i'll trash it and start again |
| 19:58.10 | brlcad | otool -L src/bwish/.libs/btclsh |
| 19:58.49 | brlcad | and otool -L src/other/tcl/unix/libtcl.dylib |
| 19:59.15 | brlcad | needed both the section commented out and the --disable-corefoundation flag |
| 19:59.32 | Z80-Boy | brlcad: On Mac, not only the background should be white, but the rpps should also have rounded corners! |
| 19:59.53 | Axman6 | http://pastie.textmate.org/124129 |
| 20:00.22 | Axman6 | and otool: can't open file: src/other/tcl/unix/libtcl.dylib (No such file or directory) |
| 20:01.06 | brlcad | ooh, Carbon |
| 20:02.09 | Axman6 | nawww, cat's asleep and twitching on me |
| 20:04.14 | Axman6 | anyway. is carbon necessary for this? |
| 20:04.48 | brlcad | no, it's not |
| 20:05.16 | brlcad | looking for the call to leave out |
| 20:05.36 | Axman6 | 'dnl check whether the Carbon framework is available (used for X11 focus)' block? |
| 20:06.57 | brlcad | yeah, the next block |
| 20:07.20 | brlcad | and then it should fail somewhere in the code where it's using that (which is what I'm hunting for) |
| 20:08.16 | Axman6 | AC_CHECK_HEADERS( \ Carbon/Carbon.h \ |
| 20:08.37 | Axman6 | comment that out? (the carbon bit) |
| 20:08.41 | brlcad | yeah |
| 20:08.46 | brlcad | that was it |
| 20:08.57 | brlcad | that triggers the code in libdm |
| 20:09.18 | brlcad | you'll remove that line, can't just prefix it |
| 20:09.31 | Axman6 | yeah, realised that |
| 20:10.16 | Axman6 | wow, this bindi irwin show is odd |
| 20:14.59 | *** join/#brlcad yukonbob (n=bch@whthyt237-51.northwestel.net) |
| 20:15.59 | yukonbob | hello, cadheads |
| 20:16.12 | Axman6 | 'lo again |
| 20:21.54 | CIA-28 | BRL-CAD: 03brlcad * 10brlcad/src/ (lgt/error.c lgt/prnt.c libfb/fb_log.c): mass simplification since we can assume stdarg is available, it's c89. remove the alternatives. |
| 20:22.02 | Axman6 | urgh, same error :| |
| 20:22.56 | brlcad | otool shows what now? |
| 20:23.31 | brlcad | because we have to get rid of all of them before it goes away |
| 20:24.00 | brlcad | to btclsh and libtcl -- that last edit should have removed from btclsh, but there might still be one on libtcl |
| 20:24.01 | Axman6 | exact same thing |
| 20:24.38 | Axman6 | i'll try trashing it, and going again from there |
| 20:25.02 | brlcad | cd src/bwish && make noprod && make LIBTOOL="sh ../../libtool" |
| 20:25.15 | brlcad | pastebin the output of that please |
| 20:25.40 | Axman6 | make: *** No rule to make target `noprod'. Stop. |
| 20:26.02 | Axman6 | typo? |
| 20:26.09 | brlcad | er, are you building out of tree or something? |
| 20:26.25 | Axman6 | wait, probably need to configure first ;) |
| 20:26.40 | brlcad | you just built, no? |
| 20:26.56 | brlcad | you should only need to configure again if you did a distclean |
| 20:27.10 | brlcad | or if you manually deleted the Makefile |
| 20:27.16 | Axman6 | yeah, but i trashed it. haven't empried trash though |
| 20:28.16 | Axman6 | http://pastie.textmate.org/124134 |
| 20:29.13 | brlcad | two problems there |
| 20:29.28 | brlcad | it shouldn't have all those -framework CoreFoundation |
| 20:29.38 | brlcad | OOOooh.. sorry |
| 20:29.46 | brlcad | you have to run ./autogen.sh after editing configure.ac |
| 20:30.06 | brlcad | you're running an out-of-date configure |
| 20:31.51 | Axman6 | ahh |
| 20:33.16 | Axman6 | 'sudo ln -s /opt/local/bin/glibtoolize /opt/local/bin/libtoolize' um, ok, don't see how that's good advice |
| 20:33.29 | Axman6 | oh wait, missed the g |
| 20:36.01 | Axman6 | ok, compiling |
| 20:48.06 | Axman6 | well, it's been going a good long time, looking food :) |
| 20:49.59 | Axman6 | why does brl-cad install in /usr btw? why not /usr/local? |
| 20:52.18 | Axman6 | argh! same error! |
| 20:52.38 | Axman6 | but... on a different part of the colpile |
| 20:53.12 | Axman6 | make[2]: *** [pkgIndex.tcl] Trace/BPT trap |
| 20:53.39 | Axman6 | which i don't think was the same place |
| 20:57.05 | Axman6 | wait, not the same error |
| 20:57.21 | Axman6 | oh, no it is :( |
| 21:03.31 | Axman6 | brlcad: anything i should show you? |
| 21:07.32 | Axman6 | heh :) |
| 21:07.37 | Axman6 | i should get some breakfast |
| 21:07.47 | brlcad | Axman6: still focusing on the link line for btclsh and the otool libs |
| 21:08.51 | Axman6 | http://pastie.textmate.org/124147 |
| 21:08.59 | brlcad | if it still lists CoreFoundation during cd src/bwish && make noprod && make LIBTOOL="sh ../../libtool" |
| 21:09.05 | Axman6 | no error there, the link one |
| 21:09.26 | brlcad | okay, that's better |
| 21:09.36 | brlcad | otool -L .libs/btclsh |
| 21:10.26 | Axman6 | http://pastie.textmate.org/124148 |
| 21:11.04 | brlcad | much better |
| 21:11.23 | brlcad | if you run ./btclsh I presume it gives the dyld error? |
| 21:12.08 | Axman6 | so is it yep |
| 21:12.23 | Axman6 | yep even |
| 21:12.43 | Axman6 | i was going to say so is it the dyld: Symbol not found: __cg_png_create_info_struct |
| 21:12.50 | Axman6 | bit that needs fixing? |
| 21:14.15 | brlcad | yeah, cd ../.. && otool -L src/other/tcl/unix/libtcl8.5.dylib |
| 21:14.56 | Axman6 | three lines, want them here? |
| 21:15.02 | brlcad | sure |
| 21:15.25 | Axman6 | /usr/brlcad/lib/libtcl8.5.dylib (compatibility version 8.5.0, current version 8.5.0) |
| 21:15.29 | Axman6 | /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version 1.0.0) |
| 21:15.31 | Axman6 | /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 111.0.0) |
| 21:15.53 | brlcad | that's it? |
| 21:16.11 | Axman6 | yeah :\ |
| 21:16.14 | brlcad | well damnits |
| 21:16.57 | Axman6 | indeed, it's not making much sense now |
| 21:18.09 | brlcad | can you run: src/other/tcl/unix/tclsh ? |
| 21:18.51 | Axman6 | no |
| 21:19.00 | Axman6 | dyld: Library not loaded: /usr/brlcad/lib/libtcl8.5.dylib Referenced from: /Users/Axman/compiled-projects/brlcad-7.10.2/src/other/tcl/unix/tclsh Reason: image not found |
| 21:19.01 | brlcad | same cg error? |
| 21:19.03 | Axman6 | Trace/BPT trap |
| 21:19.11 | brlcad | ah |
| 21:19.18 | brlcad | cd src/other && make install |
| 21:20.53 | Axman6 | ok i can run tclsh now |
| 21:24.43 | brlcad | well that's good .. |
| 21:25.07 | brlcad | could change the script to use tclsh instead, but btclsh is not really much different .. |
| 21:25.09 | Axman6 | same compile error though |
| 21:25.23 | brlcad | right, that compile error is it actually trying to run btclsh |
| 21:25.41 | Axman6 | edit the makefile? |
| 21:26.37 | brlcad | no |
| 21:26.40 | brlcad | at least not yet.. |
| 21:29.17 | Axman6 | s/$(top_builddir)\/src\/bwish\/btclsh/tclsh ? |
| 21:29.30 | Axman6 | (textmate style though) |
| 21:30.04 | brlcad | sure, you can try that -- have to edit all the Makefiles in src/tclscripts |
| 21:30.23 | Axman6 | yeah, textmate makes that easy |
| 21:30.31 | brlcad | and have to edit ami.tcl to have a "package require Itcl" in it |
| 21:31.20 | brlcad | but if btclsh won't run, then mged won't either very likely as they have nearly the same dependencies |
| 21:32.22 | Axman6 | well, passed the usual error stage |
| 21:34.38 | brlcad | alas all of this is just attempting to work around the problem instead of fixing/diagnosing the actual problem in the ImageIO framework |
| 21:35.07 | Axman6 | Elapsed compilation time: |
| 21:35.07 | Axman6 | 16 hours, 3 minutes, 33 seconds |
| 21:35.08 | Axman6 | haha |
| 21:35.26 | brlcad | :) |
| 21:35.42 | Axman6 | is there a make check or make install option? |
| 21:35.48 | Axman6 | uh, make test |
| 21:37.03 | Axman6 | yeah, mged dies |
| 21:37.19 | brlcad | run make install |
| 21:37.34 | brlcad | maybe different dyld behavior after installed |
| 21:37.43 | brlcad | since we do at least embed the right rpaths for what we use |
| 21:37.51 | Axman6 | yeah |
| 21:39.02 | Axman6 | well, at least this is all working towards possible leopard compatability |
| 21:40.00 | Axman6 | hey does launchd handling X11 launching affect you at all with brlcad? |
| 21:40.24 | Axman6 | and installed now. run make test? |
| 21:41.10 | brlcad | no, run "benchmark" in some directory where it's okay to output a lot of files |
| 21:41.34 | brlcad | or "/usr/brlcad/bin/benchmark" if you haven't added /usr/brlcad/bin to your path |
| 21:42.38 | brlcad | it'll take about 10 minutes and evaluate your system's performance |
| 21:42.46 | Axman6 | seems to be working well |
| 21:43.10 | brlcad | that's good, part to be expected since it's not gui-driven |
| 21:43.23 | brlcad | next would be to run mged -c |
| 21:47.37 | Axman6 | do you use threading a lot for brl-cad? |
| 21:47.53 | CIA-28 | BRL-CAD: 03brlcad * 10brlcad/src/ (35 files in 10 dirs): more flawfinder level 4 quellage, mostly str*() buffer boundary limits. |
| 21:49.10 | Axman6 | Total testing time elapsed: 6 minutes, 30 seconds |
| 21:49.51 | brlcad | the ray-tracer is extensively multithreaded and designed for parallel computation |
| 21:50.00 | CIA-28 | BRL-CAD: 03brlcad * 10brlcad/src/libfft/ (fftc.c ifftc.c): don't use bu_exit() so we can still avoid the libbu dependency |
| 21:50.00 | Axman6 | excellent |
| 21:50.09 | brlcad | will use however many cpus you have available by default |
| 21:51.04 | Axman6 | well, mged -c seems to run ok... |
| 21:52.26 | Axman6 | anything i can do to test it a bit more? |
| 21:52.48 | brlcad | when mged runs, run the "gui" command |
| 21:54.08 | Axman6 | X Error of failed request: BadMatch (invalid parameter attributes) \ Major opcode of failed request: 70 (X_PolyFillRectangle) \ Serial number of failed request: 1738 \ Current serial number in output stream: 1782 |
| 21:54.39 | brlcad | is X11 running? |
| 21:57.01 | Axman6 | it is, though that doesn't matter in leopard. launchd launches it on demand |
| 21:57.21 | brlcad | I know, just checking |
| 21:58.10 | brlcad | that's actually pretty stumping now |
| 21:58.24 | Axman6 | woah, running without -c gives a pretty funky error |
| 21:58.32 | brlcad | http://www.nabble.com/X11-trouble-on-OSX-10.5-t4783777.html |
| 21:58.39 | brlcad | that's prety much the same problem |
| 21:59.38 | Axman6 | The process has forked and you cannot use this CoreFoundation functionality safely. You MUST exec(). |
| 21:59.41 | Axman6 | Break on __THE_PROCESS_HAS_FORKED_AND_YOU_CANNOT_USE_THIS_COREFOUNDATION_FUNCTIONALITY___YOU_MUST_EXEC__() to debug. |
| 22:00.01 | Axman6 | about 30 times |
| 22:01.39 | brlcad | that one i've heard about |
| 22:01.46 | brlcad | add the -f option to mged |
| 22:02.13 | brlcad | looks like Tcl still ended up using corefoundation even though you added --disable-corefoundation |
| 22:02.18 | Axman6 | same X error as before |
| 22:02.41 | brlcad | well, so good news and bad news :) |
| 22:03.04 | brlcad | you do have mged at your disposal but only in console mode |
| 22:03.18 | Axman6 | heh, dang |
| 22:03.37 | brlcad | which is fully functional, other than not showing you the gui |
| 22:04.06 | Axman6 | yeah |
| 22:04.27 | Axman6 | yeah i often found it easier to use the cli for a lot of things than to use the gui |
| 22:05.07 | Axman6 | but it's been a long time, and i don't remember anything about how to use brl-cad :\ |
| 22:07.59 | brlcad | the tutorials on the website do work for console mode too, just have to use the mged quick reference to know/call the command(s) that relate to gui actions |
| 22:08.23 | brlcad | 2/3rds the tutorials are on the command console anyways |
| 22:08.43 | Axman6 | yeah |
| 22:08.56 | brlcad | I'm not sure there's much more I can do without getting my hands on 10.5 so I can interactively debug the X calls |
| 22:09.18 | Axman6 | i could work on getting you ssh access to this machine if you like |
| 22:11.10 | brlcad | that'll help get the default compile working, but the X issue might still persist -- I'll give it a try, though |
| 22:12.47 | Axman6 | ok. would tomorrow be better for you? give you a break, and time to think maybe. i'll be going out kind of soon |
| 22:15.33 | Axman6 | actually, that might give you a good chance to work on it. i could log the guess account in, or create a temporary one (guest might not have ssh access). and i'll add a .bash_profile that's the same as mine so you have the same paths. might even e able to get you vnc access so you can see if the gui's working properly |
| 22:20.25 | brlcad | I can tunnel X over ssh to one of my systems to at least make sure that works |
| 22:20.51 | Axman6 | oh yeah, heh |
| 22:20.54 | brlcad | but yeah, tomorrow would be better |
| 22:21.58 | Axman6 | well how about i set up an account for you now, and you just try ssh'ing i, and forwarding X. get those hopefully easy problems out of the way |
| 22:22.13 | brlcad | sure |
| 22:22.39 | Axman6 | alright. any particular account name? :P |
| 22:23.14 | CIA-28 | BRL-CAD: 03brlcad * 10brlcad/src/librtserver/ (rtserver.c rtserverTest.c): more finder quelling |
| 22:23.27 | PrezKennedy | Vista infuriates me!! argh |
| 22:23.33 | PrezKennedy | its so bad id almost try a mac |
| 22:23.39 | PrezKennedy | :) |
| 22:23.40 | brlcad | Axman6: "sean" ? :) |
| 22:24.07 | brlcad | PrezKennedy: you should |
| 22:24.14 | Axman6 | and password? i'll assume PrezKennedy is on windows and doesn't have ssh, so it'll be pretty safe :P |
| 22:24.41 | brlcad | Axman6: you can pm me whatever you like |
| 22:25.00 | brlcad | this channel is fully logged, i'm sure others could :) |
| 22:25.00 | Axman6 | PrezKennedy: heh, well i'm an op in #macosx, so i'd definitely say you should. ;) |
| 22:25.41 | Axman6 | good man |
| 22:37.20 | *** join/#brlcad Axman6_ (n=Axman6@61-69-24-60.netspeed.com.au) |
| 22:38.46 | Axman6 | ok, want to give ssh a go brlcad? |
| 22:50.20 | Axman6 | ok, seems to be working now rom here |
| 22:50.31 | Axman6 | from* |
| 22:52.03 | brlcad | ready |
| 22:52.56 | Axman6 | tried logging in? |
| 22:53.39 | Axman6 | oh and i guess i should copy over brlcad from my account ;) |
| 22:53.54 | brlcad | don't bother, I'll get a fresh copy so I can check in changes |
| 22:54.07 | brlcad | i'm in |
| 22:55.11 | Axman6 | heh, i'd prefer you didn't, my net's slow, and has a download limit :( |
| 22:55.18 | brlcad | downloading sources now -- I'll give the compile a go tomorrow |
| 22:55.24 | brlcad | ahh, ok |
| 22:55.32 | Axman6 | heh, nah it's ok, download it |
| 22:55.39 | brlcad | you sure? |
| 22:55.43 | brlcad | I can work with whatever |
| 22:55.57 | Axman6 | you do it your way ;) |
| 22:57.02 | brlcad | by the way, what was your benchmark summary? |
| 22:57.07 | brlcad | the vgr count |
| 22:57.16 | Axman6 | ok, you don't need this account logged in locally any more right? since you won't need vnc or anything to check if it's working |
| 22:57.22 | Axman6 | i think it was 6772 |
| 22:57.28 | Axman6 | i'll go check though |
| 22:57.46 | brlcad | cat summary will tell you |
| 22:57.52 | brlcad | and there's a log file |
| 22:58.36 | Axman6 | 6776 |
| 22:59.23 | brlcad | and you didn't compile optimized, so your system is probably somewhere around 13000 |
| 22:59.46 | brlcad | not too bad.. about a quad core ? |
| 22:59.51 | Axman6 | i think i did actually |
| 23:00.03 | Axman6 | dual core, so that makse sense |
| 23:00.12 | Axman6 | yeah i used --enable-optimized |
| 23:00.45 | Axman6 | 2x2.4GHz, 2GB RAM |
| 23:00.47 | brlcad | ah, okay |
| 23:01.31 | brlcad | scales pretty much linearly up to the oct-cores |
| 23:01.49 | brlcad | I see about 27000 there |
| 23:02.31 | Axman6 | very nice :) |
| 23:03.23 | Axman6 | does 64-bit give any speed benefits? |
| 23:03.59 | Axman6 | or just greater RAM access |
| 23:05.02 | Axman6 | oh and you'l probably need to use another prefix right? since you won't have sudo access |
| 23:05.22 | Axman6 | i guess i could give you rw access to /usr/brlcad |
| 23:06.48 | brlcad | 64bit hardly ever increases speed in any app -- just increases the size of the models it'll grok successfully |
| 23:07.53 | Axman6 | yeah. from the vids i saw rom WWDC 2006 i think only showed a speed increase when they were working with a file > 4GB |
| 23:08.11 | Axman6 | the improvment was pretty damn impressive though |
| 23:08.35 | Axman6 | something like going from 90s to 13s |
| 23:09.23 | brlcad | that's because the code ends up jumping through it's own hoops to go over 2GB/4GB |
| 23:09.43 | brlcad | most apps will just fail, abort, or crash at the boundary |
| 23:10.29 | brlcad | the demo was just a hat trick to show if you *did* have to deal with the limited memory and still load big models, what it would be to be able to not have to move stuff in and out of the address space |
| 23:11.46 | Axman6 | yeah, i understand it was a fairly rare situation, unless you work with things like genetics |
| 23:21.19 | Axman6 | brlcad: have you checked X forwarding? |
| 23:26.33 | Axman6 | ok, i'm heading out now. so i'll catch ya later. fi you decide to do anytning today, good luck :) |
| 23:52.27 | brlcad | not yet, just sorting out that base compilation problem is the first part of the problem |
| 23:52.37 | brlcad | will look into it later tonight and tomorrow |
| 23:59.54 | Axman6 | brlcad: i've also given you rw access to /usr/brlcad |
