#emc | Logs for 2006-05-08

[00:15:24] <asdfqwega> yay for me - becoming a more proficient linux user
[00:16:00] <asdfqwega> setting up apt-cacher, among other things
[00:18:22] <asdfqwega> now I just need to figure out how to do multiple terminals over ssh like I do tabbed x-terminal
[00:19:15] <alex_joni> run more than one ssh?
[00:19:31] <alex_joni> each inside a tabbed x-terminal?
[00:20:31] <CIA-4> 03cradek 07v2_0_branch * 10emc2/VERSION:
[00:20:31] <CIA-4> since there will probably be changes, we need to identify the version
[00:20:31] <CIA-4> as between 2.0.0 and 2.0.1 - if you want 2.0.0 exactly, use the tag
[00:20:31] <CIA-4> RELEASE_2_0_0
[00:33:58] <Jymmmm> Jymmmm is now known as Red70sShow
[00:33:58] <Red70sShow> Red70sShow is now known as Jymmmm
[01:04:25] <CIA-4> 03cradek 07maint1_3 * 10axis/tcl/axis.tcl: remove debug message
[01:30:35] <rayh> rayh is making virtual swarf with v2
[01:30:54] <rayh> Looks good guys, Thanks for all the sweat on this.
[01:31:07] <jepler> you guys made 2.0.0? hooray.
[01:32:23] <cradek> yay whee!
[01:32:40] <jepler> debs yet?
[01:32:44] <cradek> yep
[01:33:27] <SWPadnos> 2.0.0 huh - woohoo!
[01:35:09] <A-L-P-H-A> :)
[01:35:21] <A-L-P-H-A> oh
[01:35:29] <A-L-P-H-A> don't forget to be trendy... name it a beta as well
[01:36:54] <jepler> 2.0.0 minus epsilon pre-beta sqrt(-k)
[01:41:35] <cradek> I hope this is a good one... seems like it will be
[01:45:16] <robin_sz> kewl
[01:45:43] <SWPadnos> damn - I just installed on my (non-dapper :( ) ubuntu machine
[01:45:58] <robin_sz> is that a good thing?
[01:46:18] <SWPadnos> well, I have DSL, so the upgrade isn't too bad
[01:53:01] <SWPadnos> by the way - I did some base-thread benchmarking on my Athlon 2200 machine
[01:54:43] <jmkasunich> what did you find
[01:54:52] <SWPadnos> one sec
[01:57:12] <rayh> took that long eh?
[01:57:19] <SWPadnos> heh
[01:57:34] <SWPadnos> the test method was to run emc, then use a watch loop at 0.1 second intervals to get the baser thread execution time
[01:57:49] <SWPadnos> highest: 6905 cycles
[01:57:56] <SWPadnos> lowest: 5070 cycles
[01:58:06] <SWPadnos> average: 6455.56
[01:58:12] <SWPadnos> Std. Dev: 77.32
[01:59:14] <SWPadnos> the graph looks like a fuzzy line, with 8 large downward spikes, and a couple dozen small upward spikes
[01:59:32] <cradek> what's 6905 cycles in seconds?
[01:59:34] <SWPadnos> the clock is 1668 MHz
[01:59:40] <SWPadnos> roughly 4.something
[01:59:44] <jmkasunich> you only capture about 1 of 2000 samples
[01:59:46] <rayh> Were you running a program with emc?
[02:00:00] <SWPadnos> no program, but that shouldn't affect the base thread exec time
[02:00:05] <jmkasunich> did the max captured by hal agree with your max?
[02:00:12] <jmkasunich> it does
[02:00:19] <SWPadnos> actually, the period was 15000 ns, so it was even fewer samples
[02:00:24] <jmkasunich> oh, you mean gcode program?
[02:00:30] <SWPadnos> the total sample length was ~7250 samples
[02:00:39] <rayh> That is what I was thinking.
[02:00:43] <jmkasunich> any user space program will affect base thread execution, because it causes cache rot
[02:00:53] <SWPadnos> no, there had been a previous long max. I didn't clear it before the test
[02:01:44] <SWPadnos> the script is a one-liner
[02:06:56] <SWPadnos> well. that was a bummer. the machine spontaneously rebooted
[02:07:17] <SWPadnos> I ran emc, and hit alt-F before the splash screen had gone away
[02:07:52] <jmkasunich> alt-F?
[02:08:07] <SWPadnos> for the File menu in tkemc
[02:08:24] <SWPadnos> tkemc had already popped up behind the popimage
[02:10:24] <rayh> shorten the time in the ini
[02:11:10] <SWPadnos> I can do that
[02:11:37] <cradek> just checking, you know that doesn't have anything to do with the reboot problem, right?
[02:11:43] <jmkasunich> the disturbinh thing is yhsy linux isn
[02:11:52] <SWPadnos> which reboot problem?
[02:11:53] <jmkasunich> isn't supposed to spontaneously reboot
[02:11:56] <SWPadnos> heh
[02:12:03] <SWPadnos> too much to drink?
[02:12:05] <SWPadnos> ;)
[02:12:26] <SWPadnos> oh - of course. (finally understood the question)
[02:12:39] <cradek> just checking
[02:12:57] <cradek> did you check your logs?
[02:12:59] <SWPadnos> you can't reset the thread tmax
[02:13:04] <SWPadnos> it's not a param
[02:13:08] <SWPadnos> nope
[02:13:21] <SWPadnos> I'm not sure there was time for anything to have been physically written
[02:13:28] <cradek> yeah, usually not
[02:13:52] <jmkasunich> any console oops was hidden behind X
[02:14:00] <SWPadnos> right
[02:14:14] <SWPadnos> ah - reset each function tmax to 0
[02:14:26] <SWPadnos> (one place where multiple arguments would be nice)
[02:14:52] <CIA-4> 03jepler 07HEAD * 10emc2/src/Makefile:
[02:14:52] <CIA-4> add a PC speaker driver. This driver outputs a click every time one of the
[02:14:52] <CIA-4> 8 bits connected to it changes. When these bits are outputs from stepgen,
[02:14:52] <CIA-4> it sounds like running a stepper machine. The new configs/stepper/sim_inch.ini
[02:14:52] <CIA-4> uses this configuration to permit "ear testing" of the trajectory.
[02:14:53] <CIA-4> 03jepler 07HEAD * 10emc2/configs/stepper/ (sim_inch.ini sim_pinout.hal):
[02:14:56] <CIA-4> add a PC speaker driver. This driver outputs a click every time one of the
[02:14:58] <CIA-4> 8 bits connected to it changes. When these bits are outputs from stepgen,
[02:15:00] <CIA-4> it sounds like running a stepper machine. The new configs/stepper/sim_inch.ini
[02:15:02] <CIA-4> uses this configuration to permit "ear testing" of the trajectory.
[02:15:04] <SWPadnos> nope - that doesn't work :(
[02:15:05] <CIA-4> 03jepler 07HEAD * 10emc2/src/hal/drivers/hal_speaker.c:
[02:15:07] <CIA-4> add a PC speaker driver. This driver outputs a click every time one of the
[02:15:08] <CIA-4> 8 bits connected to it changes. When these bits are outputs from stepgen,
[02:15:10] <CIA-4> it sounds like running a stepper machine. The new configs/stepper/sim_inch.ini
[02:15:12] <CIA-4> uses this configuration to permit "ear testing" of the trajectory.
[02:15:14] <CIA-4> 03jepler 07HEAD * 10emc2/src/module_helper/module_helper.c:
[02:15:16] <CIA-4> add a PC speaker driver. This driver outputs a click every time one of the
[02:15:18] <CIA-4> 8 bits connected to it changes. When these bits are outputs from stepgen,
[02:15:20] <jepler> ouch that's a lot of lines from cia!
[02:15:25] <CIA-4> it sounds like running a stepper machine. The new configs/stepper/sim_inch.ini
[02:15:25] <CIA-4> uses this configuration to permit "ear testing" of the trajectory.
[02:15:35] <cradek> *cool*
[02:15:46] <jepler> wow, with stars and everything?
[02:15:57] <cradek> yeah that'll be really neat
[02:15:58] <SWPadnos> *_ cool _*
[02:16:03] <SWPadnos> *_cool_*
[02:16:06] <SWPadnos> hmm
[02:16:08] <cradek> in fact that's one to try right now
[02:16:14] <cradek> wonder if I have a speaker...
[02:16:15] <jepler> it's really bugging Ingrid, I think.
[02:16:19] <cradek> haha
[02:16:25] <jepler> I'm running 'tort.ngc'
[02:16:38] <cradek> wooooOOOOOweeeeooooooOOoooeEEE
[02:16:52] <jepler> I was really surprised exactly how much like your mill this sounds
[02:16:55] <jmkasunich> who is ingrid?
[02:16:59] <jepler> jmkasunich: my girlfriend
[02:17:22] <jmkasunich> don't bug her too much
[02:17:29] <cradek> so this speaker driver is realtime?
[02:17:32] <jepler> cradek: yes
[02:18:08] <SWPadnos> you write to port 61 or some such, right?
[02:18:11] <jepler> you connect up to 8 bits to it, and run it from base_thread. if any of the bits change, you get a click.
[02:18:14] <jepler> SWPadnos: yes
[02:18:34] <cradek> hahaha
[02:18:36] <cradek> that's so cool
[02:18:40] <Jymmmm> LOL.... make it play differetnt songs depending on what it does...
[02:19:05] <SWPadnos> stereo - listen to two axes at once ;)
[02:20:01] <cradek> that does sound surprisingly like steppers
[02:20:49] <rayh> see you guys. great job on the v2 release.
[02:20:56] <cradek> thanks, goodnight
[02:21:01] <SWPadnos> hey - no updater message
[02:21:05] <jmkasunich> goondight ray
[02:21:12] <SWPadnos> I feel cheated
[02:21:16] <SWPadnos> see you Ray
[02:21:46] <cradek> jepler: wow, that's irritating
[02:22:17] <jepler> cradek: yeah it is
[02:22:34] <cradek> jepler: you need to fix up your comments in hal_speaker.c
[02:22:43] <cradek> This driver supports only for demonstration how to write a byte (char)
[02:22:44] <cradek> to a hardware adress, here we use the parallel port (0x378).
[02:23:02] <cradek> even aside from "adress" I mean
[02:23:10] <SWPadnos> yoda-speak comments
[02:23:22] <jepler> yeah I should
[02:23:23] <jmkasunich> that was a really quick coding job I think
[02:23:26] <jepler> I just copied hal_skeleton
[02:23:33] <SWPadnos> and even faster commenting ;)
[02:23:44] <jmkasunich> that crap in skeleton about non-rt should go away
[02:23:49] <cradek> comments are longer than the code
[02:23:55] <jmkasunich> non-rt drivers are wrong in so many ways
[02:24:28] <cradek> jmkasunich: I've been meaning to say for a while: returning -1 all willy-nilly gives wrong errors to the user
[02:24:42] <jmkasunich> what can I say other than I didn
[02:24:46] <jmkasunich> didn't know better
[02:24:47] <jmkasunich> phone
[02:24:50] <cradek> ah ok
[02:25:01] <CIA-4> 03rayh 07HEAD * 10documents/lyx/emc2/hal_show.lyx: first part of halshow text
[02:25:07] <cradek> I guess someone *ahem* should fix them all
[02:25:25] <cradek> I've done a few but it's so boring and it's hard to pick the right error to return
[02:25:38] <SWPadnos> I guess I need to get a copy of my keys on this machine
[02:25:57] <cradek> alternatively, I can add another key
[02:26:00] <SWPadnos> I gave up on Dapper - the change to XOrg 7 seems to make it not work on my ATI 8500DV
[02:26:07] <SWPadnos> that could work too
[02:27:40] <SWPadnos> funny. I now have 4 computers on in my office
[02:27:45] <SWPadnos> it's getting warm
[02:28:52] <cradek> a while back I left an HP 385 that I was working on running overnight in my office at work, and it was quite warn the next morning, I was really surprised
[02:28:56] <CIA-4> 03rayh 07HEAD * 10documents/images/HAL-Show1.png: image for halshow.lyx
[02:29:24] <SWPadnos> at least the Athlon system is pretty responsive at 10 uS bae period
[02:29:28] <SWPadnos> base
[02:29:36] <cradek> one of those where every component has its own power supply on its own power cord about the diameter of your thumb
[02:29:41] <SWPadnos> heh
[02:30:02] <cradek> are you pretty sure you never get overruns at 10us?
[02:30:12] <CIA-4> 03jepler 07HEAD * 10emc2/src/hal/drivers/hal_speaker.c: clean up the file
[02:30:26] <SWPadnos> I suspect I get some, but it's really hard to tell
[02:30:47] <SWPadnos> though the worst recorded thread tMax was ~4.2 uS
[02:30:54] <CIA-4> 03rayh 07HEAD * 10documents/images/HAL-Show2.png: image for halshow.lyx
[02:31:03] <cradek> 10 seems pretty safe then
[02:31:10] <SWPadnos> I wasn't able to test latency, by the way. the test suite complained about a /dev file being missing
[02:31:15] <cradek> I wish we could detect overruns again
[02:31:19] <CIA-4> 03rayh 07HEAD * 10documents/images/HAL-Show3.png: image for halshow.lyx
[02:31:30] <cradek> mkdir /dev/rtf; mknod /dev/rtf/3 c 150 3
[02:32:00] <cradek> and then slap udev a couple times
[02:32:11] <SWPadnos> heh
[02:32:31] <SWPadnos> does slap require root privileges? ;)
[02:32:41] <cradek> probably
[02:32:56] <CIA-4> 03rayh 07HEAD * 10documents/lyx/emc2/hal_show.lyx: edit graphic name
[02:33:23] <CIA-4> 03jepler 07HEAD * 10emc2/configs/stepper/sim_inch.ini: increase DEFAULT_VELOCITY so jogging works
[02:33:52] <jepler> (silly AXIS)
[02:33:59] <SWPadnos> do you know which of the tests is the most "useful" in the context of emc?
[02:34:21] <cradek> I think kern latency is the one you use to see if your machine works right
[02:34:40] <SWPadnos> hmmm - it gives me a negative average
[02:34:54] <cradek> for instance that one goes wrong when I run opengl stuff on my nvidia
[02:35:13] <SWPadnos> ok. is glxgears enough to trigger the spikes?
[02:35:35] <cradek> with the nvidia closed source driver, yes
[02:36:04] <SWPadnos> it doesn't seem to matter with "ati". I haven't tried fglrx yet
[02:36:30] <SWPadnos> I still have negative numbers for the average
[02:36:32] <cradek> as far as I've seen, only the nvidia closed-source driver breaks rtai
[02:37:00] <cradek> I don't think the negatives mean anything is wrong. You can calibrate (somehow) so the average is around 0 I think
[02:37:31] <SWPadnos> true - if one is late, and the next is "on time", then they'll cancel out
[02:37:36] <SWPadnos> the errors, that is
[02:37:53] <cradek> RTH| lat min| ovl min| lat avg| lat max| ovl max| overruns
[02:37:53] <cradek> RTD| -2094| -2208| 663| 9513| 22156| 0
[02:37:53] <cradek> RTD| -2064| -2208| 628| 39856| 39856| 0
[02:37:53] <cradek> RTD| -1999| -2208| 687| 190990| 190990| 6
[02:37:53] <cradek> RTD| -1945| -2208| 567| 161496| 190990| 7
[02:37:59] <cradek> here's me starting glxgears
[02:38:21] <CIA-4> 03jepler 07HEAD * 10documents/lyx/emc2/hal_show.lyx: fix filename to be relative
[02:38:42] <jepler> that mistake must be really easy to make; I think it happens automatically if you use the file browser.
[02:39:13] <jmkasunich> auto-bug!
[02:39:17] <cradek> yeah I think each doc writer has done it now
[02:39:46] <jepler> I know I did it once
[02:39:57] <jepler> that's why it jumped right out at me when I read through the commit messages
[02:41:57] <SWPadnos> I wonder if you can stop that by specifying ./ as the PATH prefix?
[02:48:44] <jepler> I wonder why lyx doesn't have the "duh, all the filenames should be relative" flag
[02:49:29] <jmkasunich> it was designed with a single user in mind
[02:50:23] <SWPadnos> according to one page I found, the paths should be relative if they're under the document dir
[02:50:33] <SWPadnos> but if you do ../images/, you're hosed
[02:53:21] <jepler> CVS can't store symlinks, can it?
[02:53:28] <cradek> no
[02:53:32] <jepler> lyx/emc2/images -> ../../images
[02:54:01] <jepler> since they're "just" images one doesn't lose a whole lot of useful history by simply "cvs rm" and "cvs add" to move them
[02:58:26] <SWPadnos> that speaker driver is really cool!
[02:58:44] <SWPadnos> what's the glitch at every segment transition? ;)
[03:00:42] <jepler> maybe a bug in the speaker driver
[03:00:46] <jepler> goodnight all
[03:00:55] <jmkasunich> goodnight
[03:01:02] <SWPadnos> I was thinking it had exposed a TP troblem ... :)
[03:01:04] <SWPadnos> night
[03:03:04] <SWPadnos> hmmm. you know - it might be a good thing to have a halcmd command line in Axis, like the MDI buffer
[03:03:15] <SWPadnos> type in a single command, run it in halcmd
[03:03:19] <jmkasunich> why?
[03:03:32] <jmkasunich> general users should _not_ be messing with hal
[03:03:41] <SWPadnos> well, I was just thinking of putting an and2 in this sim config, so I could setp sound off ;)
[03:04:26] <jmkasunich> I like the speaker component, but I'm not so sure it should have gone into a stock config
[03:04:32] <SWPadnos> it actually does have applications for "special" machines as well
[03:04:47] <SWPadnos> it's a new ini, so it's not quite "stock"
[03:04:57] <SWPadnos> it should probably be called "sound" though
[03:05:00] <jmkasunich> oh, ok
[03:05:28] <SWPadnos> sim_inch.ini, in stepper/
[05:11:01] <giacus> morning !
[05:11:18] <giacus> happy Emc2 to all :P
[05:42:49] <Jymmmm> you make it sound like Christmas
[05:45:02] <giacus> hehe
[05:45:14] <giacus> it seems so
[05:45:16] <giacus> :D
[07:41:00] <alex_joni> morning
[07:44:33] <Jymmmm> hi
[07:52:43] <A-L-P-H-A> ohla
[07:54:15] <alex_joni> Jymmmm: you should carve some of these: http://mightyillusions.blogspot.com/2006/05/cylinder-mirror-optical-illusion.html
[07:58:43] <Jymmmm> alex_joni oh man... you know how logn they would take to carve?!
[07:58:57] <Jymmmm> But....
[07:59:26] <Jymmmm> It does give me a couple of ideas though =)
[08:02:08] <A-L-P-H-A> oh wow, now that's COOL.
[08:15:16] <Jymmmm> A-L-P-H-A how you doin?
[08:16:42] <A-L-P-H-A> good
[08:16:45] <A-L-P-H-A> did you get my pm?
[08:16:55] <Jymmmm> and replied back too
[08:17:00] <A-L-P-H-A> I didn't see the reply
[08:17:04] <A-L-P-H-A> could you send it again?
[08:17:05] <A-L-P-H-A> strange.
[08:17:09] <Jymmmm> oh hell, hang on...
[08:17:12] <A-L-P-H-A> or if you want to discuss it here...
[08:17:35] <A-L-P-H-A> still nothing
[08:17:56] <A-L-P-H-A> strnage that I'm not getting your pm
[08:19:01] <Jymmmm> it's a server issue.
[08:19:46] <Jymmmm> Not bad, having issues with stalling when the job runs a couple hours still, trying to narrow it down.
[08:19:59] <Jymmmm> Working on a new clamoing method for thin stock
[08:20:17] <A-L-P-H-A> oh
[08:20:18] <A-L-P-H-A> I see.
[08:20:20] <A-L-P-H-A> ignore that invite
[08:20:25] <Jymmmm> okey dokey
[08:21:13] <A-L-P-H-A> something funky is going on with my dvd recorder.
[08:21:19] <A-L-P-H-A> can't do backups right now. :(
[08:21:25] <Jymmmm> computer or living room?
[08:21:31] <A-L-P-H-A> computer
[08:21:36] <A-L-P-H-A> nero's flaking up
[08:22:13] <Jymmmm> I only have issue with nero when it's it's time of the month
[08:22:30] <A-L-P-H-A> I think it was folding at home that was causing it to puke
[08:22:45] <Jymmmm> folding at home?
[08:23:29] <A-L-P-H-A> protein folding...
[08:23:47] <A-L-P-H-A> spare CPU cycles are used to computer protein structures for drugs, and cures for deseases.
[08:23:57] <Jymmmm> oh, duh!
[08:24:18] <Jymmmm> that would do it, so would virus scanner and anything else running in the background
[08:24:48] <Jymmmm> or popups too
[08:25:19] <A-L-P-H-A> sure.
[08:25:28] <A-L-P-H-A> well, they're disabled now
[08:25:38] <A-L-P-H-A> and I'm starting to feel the effects of 4:26am
[08:26:29] <Jymmmm> I bet
[08:27:28] <A-L-P-H-A> know what I want to make?
[08:27:31] <A-L-P-H-A> a yoyo on the lathe.
[08:27:38] <Jymmmm> really?
[08:27:41] <A-L-P-H-A> something with a bearing inside or something.
[08:27:49] <A-L-P-H-A> yeah, why not? a cool alu. one.
[08:27:54] <Jymmmm> yoyo's have bearings?
[08:28:00] <A-L-P-H-A> but I don't know what kinda bearings I should put in.
[08:28:00] <Jymmmm> oh, ok.
[08:28:22] <A-L-P-H-A> competative yoyo's go for stupid amounts.
[08:28:32] <Jymmmm> ah, didnt know that.
[08:28:40] <A-L-P-H-A> like $100 bucks.
[08:28:43] <A-L-P-H-A> each
[08:28:57] <A-L-P-H-A> and they have a silly bearing inside... which I can't figure out either.
[08:29:16] <A-L-P-H-A> I'll check on wiki.
[08:31:13] <Jymmmm> do you already have these bearings you want to use?
[08:34:18] <A-L-P-H-A> nope
[08:34:22] <A-L-P-H-A> I probably have something.
[08:34:37] <A-L-P-H-A> wiki said SKF made it as an market ploy... but it caught on
[08:34:56] <Jymmmm> I got some abec7 bearings from vxb, not bad pricing
[08:35:53] <A-L-P-H-A> yeah, I've seen VXB on ebay
[08:35:56] <Jymmmm> they do have some tiny shits
[08:36:13] <Jymmmm> they will add you to their mailing list though.
[08:41:55] <A-L-P-H-A> http://www.henrys-online.de/En/E-YoZub.htm hmm... so that's what it looks like
[08:42:13] <A-L-P-H-A> no biggie.
[08:42:49] <Jymmmm> looking....
[08:47:54] <A-L-P-H-A> http://www.sciencenews.org/articles/20040417/bob9.asp heh
[08:49:23] <Jymmmm> Hey, turn one and put razor sharp spikes in it =)
[08:49:37] <A-L-P-H-A> I'll get right on that
[08:50:21] <Jymmmm> It might actually sell
[08:50:35] <Jymmmm> have to be perfectly balanced though
[08:54:27] <A-L-P-H-A> I'm off to sleep
[08:54:28] <A-L-P-H-A> later
[08:54:34] <Jymmmm> Me too
[08:54:37] <Jymmmm> laters
[12:07:45] <CIA-4> 03flo-h 07HEAD * 10emc2/tcl/bin/emccalib.tcl: some more i18n
[12:14:11] <CIA-4> 03flo-h 07v2_0_branch * 10emc2/tcl/bin/emccalib.tcl: merge some more i18n
[14:12:23] <giacus> ls
[14:12:25] <giacus> ops
[14:12:40] <fenn> * fenn whaps giacus upside the head with a thick UNIX manual
[14:12:50] <giacus> not found :)
[14:15:22] <rayh> he meant les;)
[14:20:57] <SWPadnos> ok. since I foolishly destroyed the source data, here are the stats from a n overnight run of the "base thread duration checker":
[14:21:10] <SWPadnos> total samples 332980
[14:21:25] <SWPadnos> average cycles: 6448.53
[14:21:47] <SWPadnos> min, max cycles: 5077 7575
[14:22:27] <SWPadnos> this machine is an Athlon 2200, with a 1668 MHz CPU clock
[14:22:52] <giacus> did you stressed it enough ? :P
[14:23:12] <SWPadnos> maybe ;)
[14:23:27] <giacus> cool
[14:23:28] <SWPadnos> there wasn;t a G-code program running, but the base thread doesn't care about that
[14:24:07] <SWPadnos> I did a similar test with glxgears running and also moving windows (and desktops) around, and there wasn't any difference
[14:24:35] <giacus> nice
[14:25:36] <giacus> there's an interesting post on the ML about threads
[14:25:53] <giacus> of today
[14:26:27] <SWPadnos> the stuff about "external control"?
[14:26:46] <giacus> it seems so
[14:26:48] <SWPadnos> sorry - "motion extension"
[14:27:58] <giacus> yeah
[14:30:12] <giacus> latest night, instead, I tried the lg3d-live
[14:30:24] <SWPadnos> lg3d?
[14:30:45] <giacus> loaded all in ram, work pretty nice on my P4 3ghz
[14:31:00] <giacus> yeah.. https://lg3d-livecd.dev.java.net/
[14:31:06] <giacus> hard distro ..
[14:31:13] <SWPadnos> ah - looking glass
[14:31:17] <giacus> yeah
[14:31:31] <SWPadnos> kind of the precursor to xgl (only more useful)
[14:31:34] <giacus> I played a bit with the testing
[14:32:37] <SWPadnos> I should try that on my dual Opteron / nvidia 7800GT machine
[14:32:53] <giacus> :D
[14:33:27] <SWPadnos> the kororaa livecd wouldn't boot - couldn't find root on the SATA CD-ROM drive
[14:34:15] <giacus> maybe could boot by a pendrive, if the bios have the support
[14:34:22] <giacus> not ?
[14:35:05] <giacus> I've an old 128 mb, but its time to buy 1gb pendrive
[14:35:31] <giacus> they're almost cheap here now
[14:35:48] <SWPadnos> I havea 1GB usb drive, but I'm not sure I want to wipe it for that test ;)
[14:35:57] <giacus> hehe
[14:36:29] <SWPadnos> of course, it's mostly full of images of streaks of light (rather than the supermodel :( )
[14:36:38] <SWPadnos> which I can probably lose at this point
[14:49:51] <fenn> i have to say i'm pretty disappointed with codeg
[14:50:08] <fenn> i thought it would at least be able to do multiple outlines, one inside the other
[14:59:18] <CIA-4> 03cradek 07HEAD * 10axis/scripts/axis.py: basic lathe support (very unfinished)
[15:00:04] <giacus> could be correct to call the EMC2 release EMC2 2.0.0 ?
[15:00:29] <giacus> or just EMC2.0.0
[15:01:29] <fenn> hmm i spoke too soon, looks like the latest version of codeg does what i want
[15:01:43] <giacus> :D
[15:02:31] <giacus> fenn: are you sad for that ? :))
[15:03:04] <giacus> anyone here aorund remember the old Dbase ?
[15:03:18] <giacus> I recall a version called DBIII Plus
[15:03:31] <SWPadnos> there was a version 1, you know ;)
[15:03:45] <giacus> when DBII was never released ..
[15:03:50] <giacus> right ?
[15:03:58] <SWPadnos> DB2 and DBase 2 are different things
[15:04:00] <giacus> I - III
[15:04:51] <giacus> weird
[15:05:08] <SWPadnos> hmmm - maybe I was wrong - there was n oversion 1, they started with II ;)
[15:05:15] <SWPadnos> http://www.pcmag.com/encyclopedia_term/0,2542,t=dBASE+versions&i=40945,00.asp
[15:05:49] <SWPadnos> at least, DBase II was the first version for the PC, rather than CP/M
[15:05:52] <giacus> yup
[15:06:30] <SWPadnos> I still have Paradox (which is one of those programs that's about good enough to keep Windows around for)
[15:07:11] <giacus> yeah, I recall it too
[15:07:25] <giacus> I played for some year with Borland Delphi II
[15:07:40] <giacus> my first RAD
[15:08:23] <giacus> until 1999 or so ..
[15:08:29] <SWPadnos> I never quite got the hang of Delphi, though I used to use Borland Pascal and I still use BC++ Builder
[15:08:40] <SWPadnos> (and turbo p[ascal / C as well, of course)
[15:08:44] <SWPadnos> -[
[15:09:02] <SWPadnos> woohoo - the stupid LabView mass compile finally finished without any errors
[15:09:13] <SWPadnos> what a nightmare LabView is
[15:09:26] <SWPadnos> bbiab
[15:09:32] <giacus> K :)
[15:19:49] <SkunkWorks> morning
[15:20:00] <giacus> hello SkunkWorks
[15:20:07] <SkunkWorks> anyone know the command to reconfigure xwindows?
[15:20:19] <giacus> wich distro ?
[15:20:19] <SkunkWorks> Hi jackey
[15:20:25] <SkunkWorks> ubuntu gnome
[15:20:53] <giacus> dpkg-reconfigure should work ..
[15:21:09] <giacus> x-window-system if I recall ..
[15:21:20] <giacus> latest time I did it was 5 years ago :/
[15:21:40] <giacus> check the name of the package
[15:22:57] <giacus> maybe you should use sudo dpkg-reconfigure x-window-system
[15:23:10] <SkunkWorks> yes - that is it - I just found the directions
[15:23:15] <SkunkWorks> Thanks
[15:23:20] <giacus> :)
[15:25:47] <giacus> sudo dpkg-reconfigure xserver-xorg could be ok too ..
[15:41:58] <SWPadnos> hmm. I wonder how well Windows networking will deal with copying ~5GB of data in ~34000 files (and 3714 dirs) in one drag-n-drop operation ;)
[15:42:24] <SkunkWorks> as long as there are not any locked files it should work.
[15:42:35] <SWPadnos> as long as there are no problems, it should work ;)
[15:42:54] <SWPadnos> Windows 2000 has this habit of keeping files locked for no apparent reason
[15:43:40] <SkunkWorks> Yes :)
[15:46:20] <SkunkWorks> I copy 67k files (26gb) pretty regualarly
[15:46:36] <SWPadnos> that's cool.
[15:46:44] <SWPadnos> do they retain their timestamps and stuff?
[15:46:48] <SkunkWorks> But I use xcopy most of the time. and have it skip files it can't copy
[15:47:07] <SkunkWorks> yes
[15:47:36] <SWPadnos> yes. if I could use the Take Command copy command, that would be the best
[15:48:27] <fenn> i take it back, codeg blows donkey balls
[15:50:32] <giacus> anyone familiar with shell scripts here ?
[15:50:51] <bill203> barely
[15:50:58] <giacus> on what depend a command in a script if executed as root or as user ?
[15:51:02] <SWPadnos> yeah - they're used to script things in shells
[15:51:20] <SWPadnos> that made no sense - can you rephrase the question?
[15:51:29] <giacus> ok..
[15:51:34] <SWPadnos> heh - thanks ;)
[15:51:48] <bill203> I thihk you can get the current username out of the environment
[15:52:02] <fenn> codeg sucking big-time: http://fenn.dyndns.org/pub/irc/random-screenshot3.png
[15:52:02] <giacus> I've irssi running, I want to create a script to autorun it in case of blackout
[15:52:24] <giacus> I know, I could use rcconf in debian for that
[15:52:32] <giacus> but it will run as ROOT
[15:52:37] <giacus> right ?
[15:52:48] <bill203> yeah
[15:52:56] <cradek> fenn: is it just cutting those in a peculiar order?
[15:52:59] <giacus> well, I want to execute it as user instead
[15:53:03] <bill203> you could do a cron job that checks to see if it's running, and launches it if it's not.
[15:53:17] <bill203> but I dont think there's any way to do per-user jobs at boot time.
[15:53:21] <giacus> or choose wich command to run as root or normal user
[15:53:36] <giacus> yeah.. at boot time
[15:53:46] <SWPadnos> you can use su for that, I think
[15:54:03] <jepler> giacus: if you want to run a command as a different user, then you have to use something like 'sudo', 'su', 'runuser' (the last may be a fedora-ism, I'm not sure)
[15:54:05] <SWPadnos> su giacus irssi
[15:54:21] <cradek> fenn: I like how it generates gcode in ... picoinches?
[15:54:22] <fenn> cradek: it cuts everything in a peculiar order.. and outputs everything as little points
[15:54:31] <giacus> K, thanks, now tryng
[15:54:43] <jepler> on my fedora system I allowed several users to run commands as themselves when the system restarts with a series of commands like: sudo -u jepler sh ~jepler/.atboot &
[15:54:47] <jepler> sudo -u jepler sh ~jepler/.atboot &
[15:54:47] <fenn> and as far as i can tell there's no support for lifting the head between cuts??
[15:54:49] <jepler> oops
[15:54:53] <SkunkWorks> fenn: some type of cad with a post proccessor?
[15:54:59] <rayh> bash: runuser: command not found -- ubuntu
[15:55:01] <fenn> qcad and codeg
[15:55:22] <cradek> fenn: ha, that's .000254 angstroms
[15:55:51] <jepler> what's angstroms?
[15:56:31] <fenn> about the size of a proton
[15:56:43] <SWPadnos> 1e-10 meters, I think
[15:56:50] <jepler> angstrom 1e-10 m # Convenient for describing molecular sizes
[15:57:01] <jepler> but what caused the mention of angstroms?
[15:57:35] <fenn> codeg outputs points with 12 decimal places
[15:57:37] <jepler> 10:54:25 <cradek> fenn: I like how it generates gcode in ... picoinches?
[15:57:40] <jepler> oh now I get it
[15:57:42] <cradek> N00005 G01 X-2.112420673858 Y0.01830701120292
[15:58:12] <fenn> that'st he beginning of a circle btw
[15:58:26] <cradek> of course, a circle
[15:58:39] <fenn> g02 would be too obvious
[15:58:42] <SWPadnos> too bad he hasn't put anything into cvs
[15:58:50] <cradek> REALIZE would output this arc/line stuff properly
[15:58:56] <cradek> ... if you have autocad
[15:59:14] <fenn> why do i need autocad for realize?
[15:59:18] <SWPadnos> Catia would probably do it as well
[15:59:22] <SWPadnos> .... if you have Catia
[15:59:26] <cradek> because it's autolisp
[15:59:40] <fenn> i'm sure any cad program out there would be better than what i'm using
[15:59:56] <cradek> doesn't look like a high bar
[16:00:03] <fenn> i'm a masochist i guess
[16:00:14] <fenn> so i guess i'll write these by hand
[16:00:19] <fenn> * fenn sighs
[16:03:08] <cradek> when cutting threads on a lathe manually, I know you cut the left side of the thread. Is this because it keeps the compound slide from running into the spindle (it sticks out to the right) or is there some benefit? I just realized my threading code was cutting on the right side.
[16:03:17] <giacus> http://www.wings3d.com/
[16:03:31] <giacus> could be a good point to start
[16:03:57] <fenn> the force is directed to the headstock bearing, which is live (roller bearings) instead of the tailstock which is traditionally a dead center
[16:04:38] <cradek> good point, you generally want to push in on the headstock
[16:04:49] <fenn> thats also why most cutters are right handed
[16:05:38] <cradek> also backlash in the Z leadscrew will mess you up if you're cutting on the right but feeding left
[16:14:22] <CIA-4> 03rayh 07HEAD * 10documents/lyx/emc2/hal_show.lyx: draft of chapter
[16:15:12] <bill203> * bill203 broke down and bought more rails.
[16:22:02] <CIA-4> 03rayh 07HEAD * 10documents/images/HAL-Show5.png: watch page
[16:25:06] <SWPadnos> rayh, if you see this, can you change the image fiole reference to a relative one instead of /home/rayh/emc ...
[16:26:55] <CIA-4> 03cradek 07HEAD * 10emc2/nc_files/threading.ngc: cut on the left side, not the right; add a couple spring passes at the end
[16:27:39] <fenn> uh, emc2 doesnt seem to like a percent sign at the end of a program??
[16:28:02] <cradek> I think if you have % signs they must be at both the beginning and end
[16:29:42] <fenn> i didnt knowthat
[16:30:05] <fenn> "somewhere around line 7 bad character" doesnt really help elucidate the nature of the problem
[16:31:06] <SkunkWorks> I like those errors - somewhere around. :) takes me a bit to figure out what I did wrong
[16:31:30] <giacus> :D
[16:32:06] <rayh> I did. I think.
[16:32:24] <giacus> SkunkWorks: I agree with you, the time isnt money :P
[16:32:32] <giacus> maybe..
[16:32:46] <rayh> In the latest commit.
[16:32:51] <SkunkWorks> mostly tinker time. so it isn't a big deal
[16:33:34] <jepler> fenn: yeah, either I'm doing something wrong in axis or the line number is pretty unreliable. That's why it says "near", not "on".
[16:33:35] <cradek> fenn: http://www.isd.mel.nist.gov/personnel/kramer/pubs/RS274NGC_3.web/RS274NGC_33a.html#1001702
[16:33:39] <Jymmmm> SWPadnos Have you played with SMT LEDs?
[16:33:39] <SWPadnos> the commit message has the full path
[16:33:42] <SWPadnos> yes
[16:33:56] <SWPadnos> nowhere near as bright as the big through hole ones
[16:34:26] <SWPadnos> well - bright, but nowhere near as much luminous flux
[16:34:31] <Jymmmm> SWPadnos: Do you think it's possible to solder a piece of 30ga wirewrap wire across a SMT LED, then clip it in the middle?
[16:34:49] <bill203> * bill203 trys to imagine why you'd want to do that.
[16:34:55] <SWPadnos> Jymmmm, yes, if you have fine enough cutters
[16:35:09] <Jymmmm> bill203 no PCB
[16:35:12] <SWPadnos> Lindstrom #8149 should work
[16:35:24] <Jymmmm> * Jymmmm googles
[16:36:15] <Jymmmm> Oh, I got some like that.
[16:36:34] <SWPadnos> you aren't looking at the photo on hmcelectronics.com are you?
[16:36:42] <Jymmmm> yeah
[16:36:49] <SWPadnos> it's not the right photo
[16:37:04] <SWPadnos> http://www.techtron.net.au/xcart/product.php?productid=19500&cat=0&page=1
[16:37:07] <SWPadnos> small, but correct
[16:38:47] <SWPadnos> http://www.chiphua.com/Images/rxtip.jpg
[16:39:14] <Jymmmm> Heh... might be easier with finger nail clippers
[16:40:06] <SWPadnos> don't bet on it
[16:40:44] <SWPadnos> the 8149 is small enough that you can cut an individual pin on an SOIC package, without touching the adjacent pins
[16:41:02] <SWPadnos> you can cut smaller pitch TQFP as well
[16:41:38] <Jymmmm> http://cgi.ebay.com/Surface-Mount-CHIP-LEDS-High-Efficiency-RED-0805-50_W0QQitemZ7617569327QQcategoryZ26207QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[16:42:33] <SWPadnos> you want this for indicator lights, not illimunation, right?
[16:44:45] <giacus> uhu? giacus isn't a sad person either, smiling 14.2% of the time, giacus is also a sad person, crying 1.5% of the time.
[16:44:53] <giacus> at the same time ? :D lol
[16:45:18] <giacus> * giacus wonder if giacus is a crazy person or pisg is weird ..
[16:45:25] <SWPadnos> http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?Ref=171086&Row=26865&Site=US
[16:46:09] <jepler> giacus: ummm .. both ?
[16:46:21] <giacus> yup :()
[16:46:24] <giacus> haha
[16:48:22] <giacus> SkunkWorks: SkunkWorks couldn't decide whether to stay or go. 2 joins during this reporting period!
[16:48:29] <giacus> rotfl
[16:49:33] <SkunkWorks> :) my connection to irc sucks here for some reason. :)
[16:54:27] <Jymmmm> SWPadnos: illimunation
[16:54:41] <SWPadnos> then give up on surface mount ;)
[16:55:09] <Jymmmm> * Jymmmm wipes that off the drawing board
[16:55:24] <SWPadnos> note that the brightness of the LED you linked to is 1/27000 that of the big white ones
[16:55:36] <SWPadnos> that was 5 mcd, not 5 cd
[16:56:51] <Jymmmm> SWPadnos: I thinkif I could find 5mm 80,000mcd with a wide viewing angle, I'd be ok.
[16:57:10] <SWPadnos> you and the rest of the world ;)
[16:57:40] <Jymmmm> oh, one of those kind things huh?
[16:57:43] <SWPadnos> the problem is the actual luminous flux density
[16:57:55] <SWPadnos> you have atiny chip - say 1mm x 1mm
[16:58:15] <SWPadnos> this thing puts out some number of photons, regardles sof the lens / diffusion
[16:58:27] <SWPadnos> smaller spot size = brighter LED (higher mcd rating)
[16:58:48] <SWPadnos> bigger spot / more diffusion = less intensity, and a lower mcd rating
[16:58:55] <SWPadnos> same total light output
[16:59:01] <Jymmmm> ah
[16:59:03] <SWPadnos> (or illuminance)
[16:59:38] <SWPadnos> that's one reason why the LuxeonLEDs are so cool - they actually put out a lot of light for their size/power
[17:00:01] <SWPadnos> the bright ones I got also seem pretty good, though I haven't done any real testing
[17:00:16] <Jymmmm> SWPadnos angle on them?
[17:00:37] <SWPadnos> 20 degrees, I think
[17:00:43] <SWPadnos> it's a pretty big spot size
[17:02:25] <Jymmmm> I'm thinking more like 30 or 45 degrees
[17:02:55] <SWPadnos> http://www.besthongkong.com/images/eBay/20_40degree_400x300.jpg
[17:03:09] <SWPadnos> that's what mine are
[17:04:17] <SWPadnos> oops - that doesn't really say. theyre listed as 12 degrees
[17:05:45] <Jymmmm> 135cd ?
[17:08:31] <bill203> dont the luxeons have external lenses?
[17:08:39] <bill203> the one I got a while ago did.
[17:14:08] <Jymmmm> dont they have heat sinks?
[17:15:22] <bill203> yeah.
[17:15:32] <bill203> well, they're on a pcb, you should sink it more.
[17:22:01] <Jymmmm> I wonder how well these perform... http://cgi.ebay.com/ebaymotors/12V-MR16-Halogen-Marine-Boat-Solar-LED-Light-Bulb-RGB_W0QQcmdZViewItemQQcategoryZ50437QQitemZ4637641927QQrdZ1
[17:22:16] <anonimasu> hm
[17:22:17] <anonimasu> *yawn*
[17:22:30] <Jymmmm> hey anonimasu
[17:22:37] <anonimasu> what's up?
[17:23:02] <Jymmmm> nada, just mucking around with LEDs
[17:24:02] <anonimasu> ok
[17:24:10] <anonimasu> I'm doing a server upgrade at work
[17:24:10] <anonimasu> :)
[17:24:28] <Jymmmm> lucky you.... not!
[17:24:29] <anonimasu> os upgrade
[17:24:45] <anonimasu> yeah, tomorrow will be hell :D
[17:24:53] <Jymmmm> nix or m$
[17:24:58] <anonimasu> m$
[17:25:07] <anonimasu> 2k3r2
[17:25:14] <Jymmmm> are you upgrading, or doing a clean install?
[17:25:19] <anonimasu> I did a clean install
[17:25:32] <Jymmmm> ok, you'll be good
[17:25:37] <anonimasu> I fear upgrades..
[17:26:09] <anonimasu> I'll have to re-add everything in the domain but it's a reasonable ammount of work
[17:26:18] <Jymmmm> upgrades will always bite you in the ass
[17:26:28] <anonimasu> yep
[17:26:48] <Jymmmm> just make it a dc and let it replicate on it's own
[17:27:09] <anonimasu> ah, I've got one server..
[17:27:21] <Jymmmm> oh, your screwed =)
[17:27:27] <anonimasu> ya
[17:27:38] <anonimasu> it's < 10 computers so it's not a big deal..
[17:27:50] <Jymmmm> even if you had made a tempoary dc, would have saved you some effort
[17:27:51] <SkunkWorks> ok everybody - reset your passwords
[17:27:57] <anonimasu> lol
[17:28:01] <anonimasu> they dont use passwords..
[17:28:11] <SkunkWorks> :)
[17:28:14] <Jymmmm> anonimasu they do now =)
[17:28:50] <anonimasu> haha
[17:29:06] <Jymmmm> ...and must be no less than 24 characters in length and must include upper/lower case letters, numbers, and puncuation
[17:29:21] <anonimasu> ^_^
[17:29:25] <anonimasu> *forces a new policy*
[17:29:32] <anonimasu> I love AD sometimes :p
[17:29:48] <SkunkWorks> we have 3 w2k servers 1 novell and now 1 ubuntu we are playing with.
[17:30:04] <anonimasu> w2k were ok.. but it was kind of buggy
[17:30:09] <Jymmmm> It's really not hard to have strong passwords w/o pos-it notes on the mointors
[17:30:17] <Jymmmm> post-it
[17:30:21] <anonimasu> I had a fight yesterday with the server
[17:30:27] <Jymmmm> who won?
[17:30:28] <anonimasu> it decided that it were
[17:30:30] <anonimasu> I won
[17:30:40] <anonimasu> it's in reality..
[17:30:48] <anonimasu> I removed a nic from it and it went ^
[17:30:51] <SkunkWorks> I think the one server we are using for accounting has been up for about 3 months - I can't remember if it has ever "crashed"
[17:31:09] <Jymmmm> 3 months? LOL
[17:31:09] <SkunkWorks> suprisingly
[17:31:13] <anonimasu> heh
[17:31:34] <SkunkWorks> that was because they cut the power to the office. had to shut down :)
[17:31:38] <Jymmmm> I've had a NT4 server up and running with 99.6% uptime over four years
[17:32:04] <Jymmmm> one was a power failure, other times was to reboot after SP
[17:32:10] <SkunkWorks> :)
[17:32:26] <Jymmmm> The TRICK is to NOT install any M$ software =)
[17:32:46] <Jymmmm> especially IE
[17:32:56] <SkunkWorks> funny - this server has win2k server and navision - that is it.
[17:33:19] <anonimasu> we run some ts apps on the server
[17:33:20] <anonimasu> that's all
[17:33:29] <anonimasu> and the fsecure administration stuff..
[17:33:30] <SkunkWorks> and navision has it own db engine
[17:35:32] <SkunkWorks> used to be that the novell server would run for years without and issue - but something borked in it I have not figured out yet.
[17:35:44] <SkunkWorks> has made it a mothly rebooter.
[17:36:32] <SkunkWorks> no abend log or anything. almost has to be a hardware issue of some kind.
[17:38:20] <anonimasu> yep
[17:38:26] <anonimasu> hm I like the new reobot stuff
[17:38:32] <anonimasu> "enter reason for reboot"
[17:38:48] <anonimasu> scheudled downtime, hardware upgrade/security errors
[17:39:00] <anonimasu> or something like that
[17:39:26] <SkunkWorks> "boss can't get email - thinks its a server issue"
[17:39:58] <Jymmmm> but in reality the admin deleted all his messages
[17:40:28] <SkunkWorks> "In reality he deleted them himself - just won't admit to it"
[17:41:50] <SkunkWorks> How do I unsend and email?
[17:41:56] <SkunkWorks> :)
[17:45:08] <Jymmmm> *sigh*
[17:46:18] <anonimasu> lol
[17:46:28] <anonimasu> delay it in the mailserver
[17:46:47] <anonimasu> and write a patch that gives a ticket no to each mail
[17:46:57] <anonimasu> and a program to snipe them from the queue ;)
[17:49:25] <Jymmmm> no need, just add to the spam filter =)
[18:23:11] <anonimasu> hehe
[18:27:47] <giacus> strange.. 425 Could not open data connection to port 32809: Connection timed out
[18:28:09] <giacus> 425 Could not open data connection to port 32811: Connection timed out
[18:28:18] <giacus> ftp won't work ..
[18:28:27] <giacus> gftp work ..
[18:28:35] <giacus> * giacus nods
[18:56:09] <giacus> stupid ..
[18:56:21] <giacus> I was not using passive mode
[18:57:18] <giacus> * giacus slaps himself
[18:59:12] <SWPadnos> I guess you're in passive aggressive mode now ;)
[19:00:21] <giacus> sometime I forget I'm using a FW .. :/
[19:08:22] <giacus> SWPadnos: you got the first place http://www.giacus.org/emcstats/emc.html
[19:09:17] <giacus> I've to tell pisg you're 'M' in order to prevent it write him/herself
[19:09:40] <giacus> also forgot some <br/> tag after the banner.. its seems
[19:10:01] <giacus> and update every 30 minuts should be ok , I guess
[19:10:16] <giacus> oh.. and time is UTC
[19:16:16] <SWPadnos> shit. nobody is foul-mouthed
[19:16:19] <SWPadnos> fuckers
[19:17:29] <fenn> ah stuff it in your cock hole, ass-face
[19:17:41] <giacus> haha
[19:17:52] <SWPadnos> cock-hole? who says I have a cock-hole?
[19:18:01] <jepler> belgium! belgium!
[19:18:09] <fenn> oh you got a prince albert or something?
[19:18:12] <SWPadnos> or, as my mother used to say: "goddamned motherfucking cocksucking sons of bitches"
[19:18:24] <SWPadnos> I'm not sure if she realized the implications
[19:18:46] <giacus> * giacus turn filter on
[19:18:48] <giacus> :D
[19:18:49] <SWPadnos> heh
[19:18:54] <fenn> not sure i ever heard my mother cuss
[19:19:06] <SWPadnos> we can abbreviate that as GDMFCSSOB
[19:19:16] <SWPadnos> as I've been doing afor a few decades
[19:21:00] <giacus> all damned males here around ? bleah *_*
[19:21:16] <fenn> * fenn shrugs
[19:21:40] <fenn> remember what happened last time you thought there was a girl here?
[19:22:27] <giacus> I have to check my ram .. can't forget :(
[19:24:46] <fenn> 2005-10-25 22:01
[19:25:12] <fenn> (romanian time)
[19:25:57] <giacus> giacus is a very aggressive person. He attacked others 1 times. ?
[19:26:02] <giacus> when ? O_O
[19:26:22] <fenn> * giacus slaps himself
[19:27:11] <alex_joni> hi all
[19:27:17] <SkunkWorks> Hi alex
[19:27:25] <alex_joni> * alex_joni just got home
[19:27:33] <giacus> fenn: right! I forgot it ..
[19:27:36] <giacus> hi alex_joni
[19:27:49] <alex_joni> giacus: just came home from 'Tosca'
[19:27:59] <giacus> Toscana ?
[19:28:21] <giacus> oh .. understood :)
[19:28:35] <giacus> this evening mi mind its a bit slow
[19:29:29] <giacus> ah.. hello K4ts
[19:29:31] <K4ts> hello
[19:30:12] <giacus> its ok ?
[19:30:13] <fenn> today i made my first functional part on the cnc foamcutter.. hm its only been what, 3 years now?
[19:30:23] <SWPadnos> woohoo!.
[19:30:35] <fenn> * fenn prepares for seppuku
[19:30:36] <SWPadnos> I've only got 1 year left, if I want to beat that record :)
[19:31:02] <alex_joni> giacus: very nice
[19:32:02] <giacus> :-)
[19:51:59] <bill203> congrats fenn
[19:55:25] <K4ts> giacus very nice what?
[19:55:49] <alex_joni> K4ts: was at 'tosca' tonight
[19:56:16] <K4ts> tosca'
[19:56:19] <K4ts> ?
[19:56:34] <alex_joni> si, c'e un opera
[19:56:39] <giacus> opera
[19:56:46] <K4ts> ye conosco
[19:56:50] <K4ts> yes
[19:57:03] <K4ts> teatro?
[19:57:03] <alex_joni> K4ts: molto bonissima
[19:57:05] <K4ts> o tv?
[19:57:09] <alex_joni> teatro
[19:57:58] <K4ts> alex_joni: you go?
[19:58:17] <alex_joni> yes
[20:00:44] <giacus> we've seen this band playng on saturday http://www.giacus.org/photo/eventi/bsm/index.html
[20:01:15] <giacus> was not so nice.. an old band (60' years)
[20:01:16] <K4ts> ahah dotto!
[20:01:27] <giacus> I lost al cooper instead
[20:01:30] <giacus> * giacus sigh
[20:02:16] <K4ts> ma alex dice bonissima
[20:02:22] <K4ts> mica l opera รจ bona ahah
[20:02:39] <giacus> molto buona is like very much
[20:02:40] <K4ts> :-)
[20:02:51] <giacus> very good
[20:03:09] <K4ts> alex in italy bonissima is sexy
[20:03:15] <giacus> hehe
[20:03:19] <alex_joni> K4ts: I knew I know that from somewhere
[20:03:29] <alex_joni> I mostly used bonissima while in italy :D
[20:03:37] <giacus> yeah, really ..
[20:03:52] <alex_joni> kidding ;)
[20:04:06] <alex_joni> there are some nice (bonissima) girls in italy..
[20:04:15] <giacus> 'Buonissima' is like very good 'bonissima' instead could mean a very sex woman
[20:04:20] <K4ts> yes
[20:04:22] <giacus> :)
[20:04:23] <alex_joni> but we have more over here :P
[20:04:29] <giacus> haha
[20:05:00] <alex_joni> giacus: for real :P
[20:05:01] <giacus> notice the 'u' :)
[20:05:06] <K4ts> alex also otello is very nice
[20:05:07] <giacus> I know..
[20:05:23] <alex_joni> K4ts: right, aida is nice too (saw it last year)
[20:05:30] <K4ts> wow
[20:05:33] <K4ts> si
[21:00:32] <bigAl> lyx-qt
[21:01:27] <jepler> bash: lyx-qt: command not found
[21:04:53] <alex_joni> LyX Version 1.3.6 of Sat, Jul 16, 2005
[21:06:37] <jepler> lyx: cannot connect to X server
[21:06:48] <alex_joni> borked setup :)
[21:08:49] <bigAl> looking to see if there is any install notes for emc2
[21:09:43] <giacus> there's a wiki
[21:10:25] <giacus> are you installing from source ?
[21:10:44] <bigAl> trying to.
[21:10:55] <giacus> there's also some nice disribution ready to use
[21:11:06] <jepler> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2
[21:11:10] <bigAl> have rcslib and emc running now.
[21:17:11] <alex_joni> bigAl: any problems with emc2?
[21:18:27] <bigAl> If I want to diable compile of a module, I just change an 'm' to 'n' in Makefile.inc ?
[21:18:52] <jepler> yes
[21:19:02] <jepler> but usually that's not necessary
[21:20:19] <bigAl> CONFIG_HAL_SKELETON=n but hal_skeleton.o is still compiled
[21:20:21] <bigAl> and fails.
[21:38:08] <NickServ> This nickname is owned by someone else
[21:38:08] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[21:38:22] <jepler> the first thing I'd try is to replace it with a stub function that always returns 0 or something
[21:39:22] <bigAl> the test_set_bit functions are for kernel only.
[21:39:30] <jepler> we'll gladly incorporate any patches that help you build on powerpc as long as it doesn't interfere with building on x86 systems and doesn't make it too much uglier.
[21:42:01] <bigAl> rdtscll is a cpu cycle counter read ?
[21:42:25] <jepler> yes, the x86 instruction is "rdtsc" for "read timestamp counter"; it's 64 bits, so "ll".
[21:44:44] <jepler> I have to go now. Best of luck. If you don't find help here, please join our developer mailing list. https://lists.sourceforge.net/lists/listinfo/emc-developers
[21:44:54] <bigAl> why not use get_cycles() ?
[21:45:17] <jepler> bigAl: usually when we've done something dumb, it's because we weren't aware of the better alternative
[21:46:01] <jepler> sometimes it's because we wanted to retain compatability with even very old systems---one of the systems we test on is kernel 2.2.18-rtl3.0, for instance.
[21:46:33] <jepler> bigAl: see you later.
[21:49:26] <SWPadnos> get_cycles would probably work. do you know of a similar function that gets the time, *very* quickly?
[21:50:02] <SWPadnos> do_gettimeofday is too slow on some systems, so other than that one ;)
[21:56:42] <CIA-4> 03alex_joni 07HEAD * 10emc2/src/emc/usr_intf/halui.cc: some more work on program start, pause, resume, step
[22:04:43] <bigAl> I do not think you want to use do_gettimeofday()
[22:05:32] <alex_joni> bigAl: right, that's no good.. but maybe you know another one we missed?
[22:05:46] <bigAl> Run ntpdate, and time can jump backwards
[22:08:43] <bigAl> alex_joni, get_cycles() is portable and fast.
[22:09:33] <alex_joni> right, but converting to time is a pita
[22:09:57] <alex_joni> if you take into account processors that change cpu_freq & all
[22:10:30] <bigAl> * bigAl sigh
[22:12:01] <CIA-4> 03alex_joni 07HEAD * 10emc2/configs/halui_halvcp/ (8 files): added new example config
[22:12:48] <jmkasunich> jepler jmkasunich: was bdi-tng one of the BDIs that needed its /usr/bin/make upgraded before it could build emc2?
[22:12:52] <CIA-4> 03alex_joni 07HEAD * 10emc2/src/Makefile: added halui_halvcp to the list of configs
[22:12:56] <jmkasunich> jepler: yes, I think so
[22:14:14] <alex_joni> hi john
[22:14:38] <jmkasunich> hi alex
[22:15:01] <alex_joni> what's up? .. guees I am :D
[22:15:29] <alex_joni> halvcp & halui make a nice team
[22:24:41] <bigAl> 18 .o files in rtlib - Does that sound right ?
[22:24:57] <K4ts> night
[22:26:04] <alex_joni> bigAl: I have about 30
[22:29:33] <alex_joni> good night all
[22:29:42] <alex_joni> * alex_joni heads to bed
[22:31:13] <giacus> night alex_joni
[22:36:53] <bigAl> OK. 29 .o modules.
[22:51:32] <bigAl> what to do with all the bit test/set functions that someone used in user space.
[22:57:19] <giacus> what ?
[22:57:59] <bigAl> kernel functions in user space are so wrong.
[22:58:34] <giacus> it's an old topic ..
[23:01:12] <bigAl> not going to get emc2 working in a day
[23:02:19] <giacus> could be a good thing
[23:02:30] <bigAl> * bigAl sighs
[23:02:59] <giacus> Usability simply does not equal low learning curve.
[23:03:10] <giacus> a bit of time is needed
[23:04:44] <bigAl> getting it to compile would be a start.
[23:05:43] <bigAl> should not have to rewrite critical sections just because some fool decided to use non-portable functions.
[23:07:44] <SWPadnos> where other than rtapi.h do you see test_and_set?
[23:15:24] <bigAl> rtai_ulapi.c calls test_bit clear_bit and set_bit
[23:15:44] <SWPadnos> so it does
[23:17:10] <bigAl> In kernel space, they are not a problem.
[23:29:23] <bigAl> Guess stubbing those calls is not a good idea
[23:30:24] <bigAl> Replacing them with semaphores might work - if RTAI suppots them.