#emc-devel | Logs for 2009-04-29

[00:02:56] <dgarr> mandelbrot set -- xaos->gimp->potrace (xaos is a very cool app (available with synaptic))
[00:03:33] <jepler> yeah I've run it before
[00:04:12] <jepler> back when it was a real "wow" to be doing 'realtime' fractal zooming
[00:04:30] <jepler> at some crazy resolution like 640x480 no doubt
[00:04:50] <dgarr> it's still pretty good on a six year old laptop
[00:07:56] <jepler> it turns out my own expectations increase at a rate higher than moore's law
[02:17:36] <jepler> ooh I just ran across a powerful rage pro turbo pci card
[02:17:47] <jepler> any takers before I sell this baby on ebay?
[02:19:18] <SWPadnos> check latency, check latency
[11:59:40] <jepler> no good deed goes unpunished
[12:08:33] <alex_joni> jepler: ?
[12:14:27] <jepler> alex_joni: cmorley's stepconf additions make it not fit on 800x600 anymore
[12:16:22] <alex_joni> oh, bummer
[13:49:08] <BJT-Work> I just booted up a computer on the LiveCD and I get "BusyBox v1.1.3..."
[13:49:18] <BJT-Work> and a prompt (initramfs)
[13:50:41] <skunkworks> try adding irqpoll to the bootparams
[13:50:57] <cradek> I think that means it couldn't find the cd after booting
[13:51:21] <skunkworks> I always get those 2 errors mixed up.
[13:51:35] <BJT-Work> ok, thanks
[13:52:18] <skunkworks> Good morning
[13:52:55] <BJT-Work> good morning
[13:57:01] <skunkworks> alex_joni: does the error stop the update?
[13:57:18] <skunkworks> - yes it does.
[14:05:50] <alex_joni> skunkworks: bummer.. so I guess aj13 is next (later tonight)
[14:06:05] <alex_joni> bbl..
[14:06:14] <cradek> hey, these that seb linked to are neat: http://www.linengineering.com/LinE/contents/stepmotors/SilverPak_17D.aspx
[14:16:36] <skunkworks> Those are neat.
[14:20:04] <jepler> I don't see prices
[14:20:26] <cradek> hmmm.
[14:22:17] <jepler> looks like that's the same brand of motor that's on zenbot, but mine's not got the integrated driver
[14:25:31] <cradek> too bad they don't have pricing and a web order form, I might get some to play with. seems like a neat idea.
[14:26:18] <SWPadnos> I wonder how they are relative to noise
[14:26:25] <SWPadnos> since you have longer step/dir wires
[14:28:49] <cradek> good question.
[14:29:24] <SWPadnos> they're pretty low power (85 oz-in), so the noise they emit may not be too bad
[14:30:51] <cradek> jepler: is adding shift-jog really so easy?
[14:31:11] <jepler> I haven't looked at his patch
[14:31:14] <cradek> are there not separate shift-key-release?
[14:31:18] <SWPadnos> if (shift) jog_faster :)
[14:31:35] <jepler> tk has complicated rules
[14:32:02] <cradek> that doesn't surprise me I guess.
[14:32:17] <jepler> nfw am I changing it in 2.3 except to fix bugs, but I'd be happy to see it changed in 2.4 if there are no obvious problems
[14:32:32] <cradek> me too
[14:33:02] <jepler> that reminds me, you promised to test some change that I should also have the intelligence not to put in 2.3.
[14:33:10] <jepler> but I forgot what it was, so I guess you're off the hook
[14:33:19] <cradek> touch off units
[14:33:36] <jepler> oh right
[14:33:38] <cradek> it works ecept for that cosmetic problem I showd you
[14:33:41] <cradek> eek
[14:33:51] <cradek> sorry, thumbs
[14:51:26] <steves_logging> steves_logging is now known as steve_stallings
[14:51:50] <skunkworks> steve_stallings!
[14:52:39] <steve_stallings> I am trying to get an module to load and work on 8.04-RTAI and need help. Suggestions for forum or channel?
[14:53:20] <steve_stallings> Lenovo T400 laptop, 8.04LTS has a back ported driver that works, but it is not in the EMC distribution.
[14:53:24] <SWPadnos> user channel, most likely
[14:53:34] <SWPadnos> oh, not an EMC module
[14:53:47] <steve_stallings> Got driver compiled and it does load, but udev does not create eth0
[14:53:58] <SWPadnos> no clue
[14:54:26] <steve_stallings> figured it would be some non-EMC forum, but where?
[14:54:47] <jepler> did you look in dmesg for information?
[14:55:08] <jepler> did you use 'ifconfig -a'? unconfigured interfaces are not shown by default.
[14:55:27] <jepler> if a firmware file is necessary, did you make sure it was installed in the right place?
[14:55:32] <steve_stallings> still slightly above my pay grade, I am learning rapidly but don't yet know lots of stuff
[14:57:00] <steve_stallings> .... moment... rebooting T400 so I can check stuff, currently running std distribution to verify it still works
[15:02:50] <steve_stallings> OK.. ifconfig -a verifies that lo loopback device is only one present, lsmod does show the e1000e module that I created is being automatically loaded, firmware was not mentioned by Intel or any other stuff I found
[15:03:48] <cradek> seems like there would be an error in dmesg if the card is not detected, or firmware is missing, etc
[15:05:09] <steve_stallings> I am puzzled that even when running standard distribution with eth0 working, the /etc/network/interfaces file still only shows the lo device, udev is funny
[15:05:42] <SWPadnos> isn't /etc/networking/interfaces a configuration file?
[15:05:44] <cradek> the newish gnome network manager goop now bypasses /etc/network/interfaces ... somehow
[15:05:52] <SWPadnos> ie, it has nothing to do with what's loaded or detected
[15:06:09] <cradek> SWPadnos: used to be, you'd list the interfaces in there that you want it to do something about.
[15:06:18] <SWPadnos> right
[15:06:24] <cradek> now the gnome goop does it automagically after you log in
[15:06:33] <cradek> (which is an abomination)
[15:06:33] <SWPadnos> ie, it's an input to the metworking system, not an output/status from it
[15:06:39] <SWPadnos> networking
[15:06:40] <cradek> yes
[15:07:36] <steve_stallings> ok, looking at dmesg output, newbie gotta learn redirect, grep, etc next I guess
[15:08:23] <cradek> dmesg -c; insmod whatever; dmesg
[15:13:51] <steve_stallings> brute force reading dmesg, e1000e loads without error, prints copyright, then "e1000e: probe of 0000:00:19.0 failed with error -5"
[15:14:23] <cradek> well that's a lovely error message...
[15:14:33] <steve_stallings> Google is my friend...
[15:15:01] <cradek> yeah I was just doing that too
[15:15:23] <steve_stallings> yikes... lots of hits
[15:35:10] <steve_stallings> fyi - I had read about the NVM corruption problems with beta drivers and had backed up my NVM before starting, I am NOT getting the NVM checksum error mentioned by many others
[15:41:19] <steve_stallings> enough headaches for now, thanks for the help
[15:41:33] <steve_stallings> steve_stallings is now known as steves_logging
[15:43:16] <cradek> hope you get it soon.
[15:51:44] <cradek> jepler: this works great except the shifted version of KeyPress-[ is not Shift-KeyPress-[ so it only works as-written on x,y,z.
[15:52:20] <cradek> and I assume shift-[ => { is not universal for all keyboards
[15:55:37] <jepler> no, it's not
[15:56:35] <cradek> how do you feel about shift-jog for xyz only?
[16:00:59] <jepler> it's better that what we have now
[16:01:10] <jepler> I would like the shift-jog feature
[16:01:23] <cradek> it seems nice
[16:02:20] <CIA-1> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: shift-jog for XYZ jogs at full speed.
[16:04:10] <jepler> yay!
[16:06:25] <BJT-Work> cradek: does shift jog rapid?
[16:06:48] <cradek> yes but for the xyz keys only.
[16:06:53] <BJT-Work> ok
[16:07:00] <cradek> in trunk, not in 2.3
[16:07:07] <BJT-Work> :)
[16:07:32] <BJT-Work> what was the new comment for g code I saw it but forgot to save it
[16:07:41] <BJT-Work> ; or something?
[16:07:48] <cradek> yes ; is a comment now
[16:07:55] <BJT-Work> only need one?
[16:07:59] <cradek> right
[16:08:05] <BJT-Work> cool, thanks
[16:08:05] <cradek> everything from ; to the end of the line is a comment
[16:08:17] <BJT-Work> that is much easier :)
[16:08:26] <cradek> except if ; is in a comment!
[16:08:39] <BJT-Work> ok
[16:08:41] <cradek> g0 (this sucks; but whatever) x1
[16:09:11] <BJT-Work> was that in trunk only?
[16:09:19] <cradek> no, the ; comemnt is in 2.3
[16:09:30] <BJT-Work> both ok
[16:09:59] <BJT-Work> I meant to save the commit but forgot... must be sometimers
[16:10:34] <cradek> * cradek gently points BJT-Work to the changelog at http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/debian/changelog?rev=;only_with_tag=RELEASE_2_3_0
[16:12:33] <BJT-Work> LOL, I was going to look for that this morning but ran out of time
[16:13:12] <cradek> :-)
[16:13:20] <cradek> there might be a lot of undocumented stuff in there - not sure.
[16:13:44] <BJT-Work> don't see anything on the comment
[16:14:56] <BJT-Work> now I see it
[16:15:19] <BJT-Work> I need a shortcut to that on my desktop :)
[16:15:33] <CIA-1> EMC: 03cradek 07TRUNK * 10emc2/debian/changelog: shift+jog for traverse speed, thanks Michael Buesch
[16:44:42] <jepler> hm, axis needs to know whether mdi before home is permitted or not, and enable/disable "touch off" "zero coordinate system" and the mdi tab based on it
[16:45:07] <cradek> ick
[16:45:12] <jepler> or maybe the surprised user is better served by the error than by disabled controls..
[16:45:35] <jepler> oh, and the run / step buttons
[16:46:12] <cradek> I've always thought that the gui should be able to say why a thing is disabled. remember the win95 question mark button?
[16:46:19] <jepler> yeah that'd be neat
[18:16:28] <alex_joni> cradek: I think you missed checking in axis.tcl changes
[18:16:44] <alex_joni> (by reading michael buesch's patch from his second email (2/2))
[18:18:20] <cradek> I didn't do the gui part, only the keyboard part.
[18:21:48] <alex_joni> ah, ok.. was just slightly reading the messages I got
[18:21:55] <alex_joni> * alex_joni isn't really completely awake yet
[18:22:24] <cradek> I'm positive I want the keyboard change, not so sure about the gui change.
[18:32:36] <cradek> BJT-Work: I think [DISPLAY] OPEN_FILE is not in the docs (it lets you open a custom program when AXIS starts, instead of the splash screen)
[18:37:08] <jepler> http://linuxcnc.org/docs/2.3/html/config_ini_config.html#sub:[DISPLAY]-section
[18:37:13] <jepler> it's there, section
[18:37:42] <jepler> it's not in axis.lyx, though, while at least one inifile item *is* (lathe mode)
[18:38:03] <cradek> ok, I was looking in axis.lyx.
[19:06:59] <BJT-Work> ok thanks cradek
[19:11:03] <cradek> maybe a cross reference from axis.lyx to the section jepler found?
[19:11:07] <BJT-Work> cradek: http://www.linuxcnc.org/docview/html//config_ini_config.html#sub:[DISPLAY]-section
[19:11:39] <cradek> yeah, I was wrong that it didn't exist - I was looking in the axis specific docs.
[19:11:55] <alex_joni> it's probably axis specific though
[19:12:01] <cradek> yes it is
[19:12:25] <BJT-Work> yes, it is in the AXIS section of DISPLAY
[19:12:56] <BJT-Work> * BJT-Work wonders if it needs to be in the axis.lyx as well...
[19:13:01] <jepler> duplication bad
[19:13:02] <alex_joni> maybe we should have a link for AXIS specific ini things, from the axis.lyx
[19:13:03] <jepler> references good
[19:13:15] <jepler> "for inifile settings that change how axis works, see ..."
[19:13:22] <BJT-Work> yep
[19:27:38] <skunkworks> alex_joni: the mesa firmware 'is' on the latest livecd ;) http://imagebin.ca/img/z0vrLSPy.jpg
[19:29:33] <alex_joni> skunkworks: I know it is :)
[19:29:50] <alex_joni> thanks for checking though
[19:29:51] <jepler> I don't understand why you put quotes around 'is'
[19:30:34] <alex_joni> maybe it's a philosophical thingie
[19:30:58] <alex_joni> as it's not palpable, it might not even exist
[19:31:07] <skunkworks> I thought the last livecd had missing firmware
[19:31:30] <alex_joni> that was aj10, aj12 has the fw
[19:31:46] <skunkworks> Also - I don't know how to underline..
[19:31:58] <skunkworks> ;)
[19:32:11] <cradek> _underline_
[19:32:19] <alex_joni> _underline_
[19:33:59] <jepler> underline ?
[19:34:34] <skunkworks> <u>hello</u> ? ;)
[19:34:45] <alex_joni> skunkworks: use _
[19:34:53] <alex_joni> before and after the word
[19:35:10] <skunkworks> doen't seem to work in this client _test_
[19:35:19] <alex_joni> sure looks ok here :D
[19:35:32] <alex_joni> get a less crappy client
[19:35:36] <skunkworks> huh - only jeplers is underlined in my window
[19:38:22] <skunkworks> http://imagebin.ca/img/XOvOmy.png
[19:38:30] <cradek> I would type that as: firmware IS on the...
[19:38:52] <cradek> I can communicate in 7-bit ascii
[19:39:21] <skunkworks> yes - but you have a better grasp of grammar
[19:39:22] <alex_joni> ok, the IRC client is actually not THAT crappy, the browser tops it
[19:39:30] <skunkworks> heh
[19:39:35] <cradek> in baudot you'd type: FIRMWARE /IS/ ON THE
[19:40:34] <alex_joni> how about Murray ?
[19:41:29] <alex_joni> heh.. RYRYRYRYRY
[19:42:01] <jepler> murray?
[19:42:23] <alex_joni> http://en.wikipedia.org/wiki/Baudot_code#Murray_code
[19:42:48] <cradek> Note further that the "control" characters were chosen so that they were either symmetric or in useful pairs so that inserting a tape "upside down" did not result in problems for the equipment and the resulting printout could be deciphered. Thus FIGS (11011), LTRS (11111) and space (00100) are invariant, while CR (01000) and LF (00010), generally used as a pair, result in the same output when the tape is reversed.
[19:42:55] <cradek> that's cool...
[19:43:17] <alex_joni> yup
[19:43:35] <alex_joni> back when people were taking a lot of time with the specs before pushing them out :D
[19:46:00] <alex_joni> * alex_joni whines for faster upload
[19:46:06] <jepler> The control codes felt essential for data transmission were the start of message (SOM), end of address (EOA), end of message (EOM), end of transmission (EOT), "who are you?" (WRU), "are you?" (RU), a reserved device control (DC0), synchronous idle (SYNC), and acknowledge (ACK). These were positioned to maximize the Hamming distance between their bit patterns.[19]
[19:49:27] <alex_joni> pushing an ISO at 1Mbit/sec is really slow :)
[19:50:45] <skunkworks> are we there yet? ;)
[19:56:10] <alex_joni> 1:25:14 ETA
[20:05:31] <alex_joni> interesting emc _interpreter_ bug
[20:05:47] <SWPadnos> heh
[20:06:25] <alex_joni> at least he signed in
[20:07:07] <alex_joni> hmm.. cradek was faster :)
[20:08:37] <skunkworks> I think jepler or someone changed it so you couldn't annonomously add bug tickets.
[20:08:43] <skunkworks> iirc
[20:15:42] <SWPadnos> heh
[20:15:46] <SWPadnos> strike two
[20:19:53] <cradek> he did it again!!??
[20:20:59] <SWPadnos> yeah, now it's a feature request
[20:23:21] <cradek> dangit I hate cleaning up after people
[20:33:55] <cradek> I can't fathom why the sf trackers now default to showing all states, instead of just open
[21:13:49] <alex_joni> wb skunkworks
[21:13:57] <alex_joni> still at 88% only :)
[21:55:20] <alex_joni> skunkworks: it's up on my mirror.. uploading from there to linuxcnc.org now
[22:40:36] <alex_joni> upload finished
[22:40:53] <alex_joni> I'll test aj13 later today, and replace the links if it's ok
[22:51:00] <alex_joni> good night all
[22:58:19] <SWPadnos> see you
[23:23:15] <CIA-1> EMC: 03bigjohnt 07v2_3_branch * 10emc2/docs/src/gui/axis.lyx: add reference to axis specific ini settings and some minor clean up
[23:26:02] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gui/axis.lyx: add reference to axis specific ini settings and some minor clean u p
[23:38:58] <CIA-1> EMC: 03bigjohnt 07v2_3_branch * 10emc2/docs/src/gcode/overview.lyx: add info on comments
[23:39:30] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gcode/overview.lyx: add info on comments
[23:48:17] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gui/axis.lyx: add info on keyboard rapid moves for xyz