#emc | Logs for 2005-09-16

[00:05:35] <dave-e> Hi paul I'm back up on the new monitor at 60 Hz. Tnx
[00:06:00] <dave-e> synergy dwgs look really nice.
[00:10:11] <paul_c> If you ever need to know the modeline for a res & sync.... gtf
[00:10:58] <dave-e> OK...next problem...mounting the floppy under 4.20
[00:12:01] <paul_c> symptoms ?
[00:13:25] <dave-e> well for starters what is the proper device name
[00:14:07] <dave-e> gtf looks cute
[00:15:55] <paul_c> dev/fd0
[00:16:01] <paul_c> /dev/fd0
[00:16:07] <dave-e> mount: /dev/fd0 is not a valid block device
[00:16:31] <paul_c> Is floppy enebled in the BIOS ?
[00:16:59] <dave-e> hmmm...I can't guarantee that...be back in awhile
[00:28:37] <dave-e> hey paul...one more problem bites the dust! Floppy works. :-)
[00:28:57] <paul_c> operator error.
[00:36:05] <paul_c> Hi Ray.
[00:36:17] <rayh> Hi Paul.
[00:37:26] <rayh> Sent off the revised INI but have not heard any report.
[00:37:43] <paul_c> Got a script that will update from old to current...
[00:37:59] <paul_c> Requires some input from the usr.
[00:38:33] <rayh> That.s okay. I can write a page that describes the process.
[00:39:15] <rayh> Mesa still does NOT have the PC104 boards. 2 weeks more they say!!
[00:39:34] <paul_c> I can wait...
[00:40:14] <rayh> They said they might ship a 200k chip rather than the 100 that is standard.
[00:40:34] <rayh> Wouldn't think that would cause problems.
[00:40:42] <paul_c> I would prefer a 200K chip
[00:42:29] <rayh> Okay. Do you want an analog board as well or the 100 watt motor card?
[00:44:47] <rayh> I guessed that you wanted the 50 pin ribbon version rather than the odd connectors.
[00:46:58] <paul_c> Yes, the IDC50 connectors mate with the cables I already use.
[00:47:19] <rayh> Figured as much.
[00:48:13] <paul_c> The 7137 would be nice, but not essential.
[00:48:44] <paul_c> The 7130 & 7132, not much use for...
[00:51:45] <rayh> That is the board Mesa built to our specs. I ordered one once and they said it wasn't ready.
[00:52:48] <rayh> I think that Todd got one but I've not seen it. I'll try to get you one.
[00:53:01] <dave-e> paul...responding to your comment on operator error....more like operator duh!
[00:53:27] <rayh> Hi Dave. Didn't see you.
[00:53:46] <rayh> my alexia is showing!
[00:53:47] <paul_c> dave-e: You wouldn't be the first or last to do that..
[00:54:40] <dave-e> hi ray
[00:54:49] <dave-e> been hiding, just like paul does
[00:56:51] <bosone> hello :)
[00:57:54] <rayh> How you doing today.
[00:58:18] <bosone> today i disassaembled an hp2500c broken printer
[00:58:34] <dave-e> any good parts/
[00:59:19] <bosone> all hp inkjet printer have dc motors with optical encoders!
[00:59:56] <dave-e> enough counts on the encoder to be useful?
[01:00:59] <bosone> there are a rotary 1440 count encoders
[01:01:21] <dave-e> good...even useful
[01:01:51] <bosone> and a linear encoder of unknow counts
[01:02:19] <dave-e> maybe correlates with character spacing/
[01:03:05] <bosone> yes i think 200 or 300 dpi
[01:03:11] <rayh> Are all characters equally spaced on these printers?
[01:04:00] <bosone> this is an inkjet
[01:05:15] <bosone> in the board there are 2 mistery motorola chips xc79203fn and sc79202fb
[01:05:37] <bosone> i can't find the datasheet!
[01:06:01] <bosone> i think they are custom for hp
[01:07:42] <AchiestDragon> probablay
[01:08:26] <bosone> which is the easy cheap way to connect dual or triple dac to emc via lpt?
[01:09:09] <bosone> 8 bit are good?
[01:09:16] <AchiestDragon> http://www.cpcares.com/comp/MS/XC79203FN.html
[01:10:12] <AchiestDragon> sc79202fb canot find data on
[01:11:42] <bosone> thanks but http://www.cpcares.com/comp/MS/XC79203FN.html have broken link!
[01:12:31] <AchiestDragon> stupid site says it has it
[01:12:36] <AchiestDragon> :(
[01:13:39] <AchiestDragon> btw google using the part number , theres a few but most are just selling the parts
[01:14:08] <dave-e> lots of claims but not much solid info
[01:14:38] <AchiestDragon> if you google with the part and add data to the end then you should get the ones with links to the data sheets
[01:15:37] <AchiestDragon> http://www.eserviceinfo.com/ may be helpfull also
[01:17:32] <bosone> ok thanks, but i think they are custom private documentations of hp
[01:18:02] <bosone> i have a stupid idea...
[01:19:49] <robin_sz> evening ...
[01:19:55] <AchiestDragon> hi
[01:19:56] <bosone> can i connect to a secondary pci vga card the r-g-b output to x-y-z driver?
[01:20:15] <bosone> isn't inpossible
[01:20:38] <AchiestDragon> well vidieo is analog so not easy
[01:21:23] <bosone> for dc motor, not for stepper
[01:22:15] <robin_sz> hmmm, I suspect it will insist on having line blanking and field blanking
[01:22:22] <AchiestDragon> well in a 24bit colour mode you would get 256 levels on each
[01:22:50] <AchiestDragon> you may be able to frig it in software to omit the blanking
[01:22:56] <robin_sz> maybe ...
[01:23:32] <robin_sz> trouble is .. its all memory mapped output too
[01:23:40] <bosone> or in hardware, sampling over the sync
[01:24:01] <robin_sz> well, there are easier ways :)
[01:24:03] <AchiestDragon> but not easy , would take more electronics on the interface to it that it would be easyer to build a pci programable i/o card
[01:24:25] <AchiestDragon> with motor dive on it
[01:24:54] <bosone> are sure?
[01:25:28] <robin_sz> so, anyone knpw about Linux printing?
[01:25:34] <bosone> wy trash old pci video cards?
[01:26:27] <bosone> robin_sz: not much
[01:28:27] <AchiestDragon> robin_sz: it used to be a sod to configure way back in 1994-8 but since then not had a problem since i swiched to suse and it seems to auto configure ok
[01:28:51] <bosone> emc2 run on 2.6 kernel?
[01:28:56] <robin_sz> im trying to work out how to add a postscript background image to all the printed output ..
[01:29:01] <robin_sz> like a letterhead ...
[01:30:13] <AchiestDragon> no ,cant help on that , only print pictures or pdf's and the occasional manual and plain text
[01:30:31] <robin_sz> yeah
[01:31:12] <robin_sz> I use Sage for accounts ... you have to use special Sage invoices, that cost a fscking fortune .. I figured on scanning one and having it overlayed automagically on a special print device ...
[01:33:00] <AchiestDragon> print it in two stages , print out the blank form with the leter head on some sheets , put them back in the printer ( correct way arround) and treat them as the preprinted ones
[01:33:19] <robin_sz> errm, lets think about that .. errm, no.
[01:35:23] <AchiestDragon> it works ok
[01:35:28] <robin_sz> impractical ..
[01:35:39] <robin_sz> too much fsckign about ...
[01:36:01] <robin_sz> there will be a solution, I just havent found it yet ...
[01:36:24] <AchiestDragon> print out one , go to local printers and get a couple of reams of them pre printed
[01:36:42] <robin_sz> also impractical
[01:36:53] <jepler> robin_sz: programs like "mpage" can do things to compliant postscript input, like 4-up printing. What you want should be possible too, but it might just require a postscript guru
[01:37:01] <robin_sz> wanna see how many Sage invoice we waste by accidentally printing letter on them?
[01:37:46] <robin_sz> jepler: yeah, I figure it might be possible with some magical GS options in the lpr filters
[01:38:24] <robin_sz> * robin_sz installs mpage
[01:38:30] <robin_sz> anyway .. back to CNC :)
[01:38:33] <jepler> cups has some kind of watermarking option..
[01:38:42] <jepler> $ lp -d foo1 -o even:Watermark=on -o odd:ColorMode=Gray file
[01:38:43] <AchiestDragon> but you got to keep enableing and disabling it or if you print you will get the background on all printouts
[01:38:46] <jepler> http://www.linuxprinting.org/cups-doc.html
[01:38:54] <robin_sz> it does .. but the documentation seems a bit transparent that
[01:39:18] <robin_sz> found that bit .. I *think* it just turns on preset watermarking options in the printer ...
[01:39:56] <jepler> oh
[01:40:05] <robin_sz> AchiestDragon: no, I'll just set up several virtual printers, that are one physical printer, but print either plain, a letterhead, a invoice or a statement
[01:40:13] <AchiestDragon> k
[01:40:45] <robin_sz> ooh, I lurv debian
[01:42:08] <robin_sz> so .. that servo amp off ebay turned up ...
[01:42:41] <robin_sz> seems a bit complicated .. has lots of tweaks for Kv, accel etc
[01:45:14] <bosone> sorry, i have another stupid idea about vga cards....
[01:46:44] <bosone> can be used as triple pwm (with custom patterns) without cpu load
[01:47:32] <bosone> so we have virtually a lot of bits
[01:51:06] <jepler> bosone: Why limit yourself to PWM? They're very fast DACs, after all, though not terribly accurate ones.
[01:51:51] <jepler> bosone: also you're typically limited to at most a few thousand pixels per line at most
[01:53:37] <jepler> I've used VGA for a very fast DAC before. http://emergent.unpy.net/index.cgi/01115494353 http://emergent.unpy.net/index.cgi/01114979771 http://emergent.unpy.net/index.cgi/01114971365
[01:54:00] <paul_c> rayh: User intervention required at four points.
[01:54:57] <jepler> I haven't done anything more with it for a few months, because I'm not sure how to resolve the remaining issues (what to do about the way changing Z (brightness) makes the other channels all shift voltages, basically)
[01:55:01] <bosone> very great!!
[01:55:50] <jepler> maybe you recognize some of the games
[01:56:26] <bosone> which card you used?
[01:56:42] <jepler> The built-in video of my laptop (ATI radeon family)
[01:57:18] <bosone> linux framebuffer?
[01:57:44] <bosone> or direct driving?
[01:57:49] <jepler> No, XFree86 in standard video modes. SDL for the terminal application, xmame's own framebuffer library for the video games
[01:58:19] <jepler> I experimented with setting weird display resolutions in X (e.g., 4096x256) but I must have been doing something wrong as it never worked.
[02:04:01] <bosone> ok, now i'm going to bed. G night all :)
[02:04:15] <jepler> nige
[02:04:16] <jepler> te
[02:04:18] <jepler> nite
[02:05:10] <AchiestDragon> been playing at exporting blender stuff to dfx then g code today
[02:06:20] <jepler> why not cut out the middle step? Isn't blender scriptable with python?
[02:07:37] <AchiestDragon> got one big bug , have to edit the gcode, it starts by milling out from the center of the work to a corner of it at full depth before starting to do it properly , then at full depth cuts a cross out just before it finishes
[02:09:07] <AchiestDragon> i also check the scale when its in autocad , seem to have better control of that there
[02:09:57] <jepler> yeah, programs that are for art rather than cad often don't care too much about the actual size of anything
[02:10:45] <rayh> Hi paul_c. What are the four points?
[02:10:50] <robin_sz> night ...
[02:11:25] <AchiestDragon> found it good to draw a template of roughly the right size in autocad and inport it into blender first ( get it close that way )
[02:11:59] <paul_c> 1) apt-cdrom prompts for the CD
[02:12:25] <paul_c> 2) Need to OK wordlist.
[02:12:54] <paul_c> 3) Answer "yes" when asked about removing the running kernel.
[02:13:16] <paul_c> 4) answer "y" to upgrading libc
[02:13:34] <paul_c> 5) answer "y" to restarting services.
[02:13:52] <rayh> That should not present serious issues for their customers.
[02:16:10] <paul_c> Need to put in a final "enter" to reboot prompt..
[02:16:33] <rayh> That sounds great, Paul.
[02:17:04] <rayh> Should be no problem with pmt.
[02:18:03] <paul_c> Not sure if running it from the CD would cause problems...
[02:18:35] <paul_c> apt-cdrom unmounts the CD - But there is an option to prevent that.
[02:18:48] <rayh> Can we run it from the CD?
[02:19:09] <paul_c> would need to burn & test to be sure.
[02:19:57] <rayh> You mean click on the script on the cd rather than copying it to hd?
[02:20:28] <paul_c> yes.
[02:21:03] <jepler> what you need is a special ps/2 dongle that plugs into the keyboard and types return return y y return at the right intervals
[02:21:29] <paul_c> jepler: It's called a user. ;)
[02:21:42] <rayh> You supply those in quantity?
[02:25:32] <paul_c> Use the -m flag to apt-cdrom should drop point 1.
[02:26:35] <Jacky^> Gnight
[02:26:46] <jepler> you could use a camera and a dongle to make it a closed-loop type system
[02:54:18] <mrallen> how does G40-G42 (tool offset) and G43,G49 (tool length offset) work?
[03:02:52] <mrallen> i found the g-code reference on linuxcnc.org (and a wiki version of it), but both refer to further info on 'separate pages' that i can't find links to
[03:14:55] <cradek> http://www.isd.mel.nist.gov/personnel/kramer/pubs/RS274NGC_3.web/RS274NGC_33a.html#1013772
[03:28:52] <mrallen> thanks
[06:42:58] <Jymmmm> Jymmmm is now known as Jymmmmm
[06:43:03] <Jymmmmm> Jymmmmm is now known as Jymmm
[07:35:48] <Jymmm> Ok, I've modified /usr/local/emc/tcl/lib/tkemc/tkemcSetKetBings.tcl to include the numpad, but no changes are being accepted.
[07:37:30] <anonimasu> *yawns*
[07:37:58] <Jymmm> * Jymmm hands anonimasu a breath mint
[07:38:38] <Jymmm> anonimasu any thughts?
[07:39:49] <anonimasu> thanks
[07:39:51] <anonimasu> not really
[07:39:53] <anonimasu> I just woke up
[07:39:58] <Jymmm> ah, ok
[07:40:00] <anonimasu> I am ordering pistons for my car
[07:40:32] <Jymmm> ah, ok.
[07:40:57] <anonimasu> forged ones..
[07:40:57] <anonimasu> :)
[07:50:34] <Jymmm> I wish I knew if I was editing the right file, or just doing somethign wrong =(
[08:08:47] <Jymmm> G'Night Folks!
[10:30:17] <_AchiestDragon> _AchiestDragon is now known as AchiestDragon
[13:49:00] <anonimasu> hello
[14:05:33] <anonimasu> * anonimasu yuawns
[14:06:11] <bosone> hello
[14:23:23] <Jacky^> hi :P
[14:23:33] <anonimasu> what's up?
[14:24:14] <Jacky^> tryng to get singleface pcb fo optos circuit
[14:24:28] <Jacky^> seem to be hard..
[14:25:52] <Jacky^> * Jacky^ listen KC & The Sunshine Band
[14:25:55] <Jacky^> :D
[14:28:21] <anonimasu> hm ok
[14:29:13] <Jacky^> babyyy I love u sooo :P
[14:29:33] <Jacky^> please don't gooo
[14:29:54] <Jacky^> i like the second track
[14:30:03] <Jacky^> coool :D
[14:30:44] <anonimasu> hehe
[14:30:47] <anonimasu> I am slacking
[14:30:56] <anonimasu> thinking of machining projects
[14:31:10] <Jacky^> O_O
[14:31:14] <Jacky^> :-)
[14:31:42] <Jacky^> ^O-O^
[14:37:43] <mrallen> anonimasu: i hand-rolled a visualmill post for sherline
[14:37:57] <anonimasu> mrallen: nice
[14:39:52] <anonimasu> I am looking for needle roller bearings
[14:39:58] <anonimasu> err linear bearings
[14:41:51] <mrallen> type in 'needle bearings' on mcmaster.com
[14:42:15] <anonimasu> cant shop there I am in sweden
[14:44:35] <anonimasu> and I need linear ones
[14:47:34] <anonimasu> :/
[14:48:18] <anonimasu> they look like a strip with a few needle rollers on them
[14:49:00] <anonimasu> http://www.formsprag.com/images/RW.jpg
[14:49:03] <anonimasu> but tiny
[14:51:06] <Jacky^> hello Jymmm :P
[14:51:20] <Jacky^> good morning america :D
[14:51:22] <anonimasu> got any clue on where to shop for that?
[14:52:21] <Jymmm> http://www.stieber.de/lang_sc/index.php
[14:53:00] <Jymmm> hi Jacky^
[14:53:40] <Jacky^> http://www.usedpchub.com/uph/
[14:53:46] <Jacky^> hi Jymmm
[14:54:49] <Jymmm> Jacky^ you just said that
[14:55:05] <Jacky^> :P
[14:55:56] <Jymmm> I'm not even thru my first cup of coffee and Jacky^ is already mucking with me!
[14:56:01] <anonimasu> perhaps..
[14:56:07] <Jacky^> hahaha
[14:56:10] <Jacky^> :D
[14:56:53] <Jacky^> Jymmm: you know, i'm happy whan i see you entering :P
[14:58:17] <Jacky^> Jymmm: good weather here, sun, nice temperature and hard working :P
[14:58:56] <anonimasu> Jymmm: nice :)
[14:59:00] <anonimasu> I should grab a cup of tea
[14:59:12] <Jacky^> i decided to try my first doubleside pcb today..
[14:59:22] <Jacky^> working on eagle :)
[15:00:32] <anonimasu> I am debating if I should throw the PLC inside of the fridge to try how well it handles cold
[15:01:31] <anonimasu> or well one
[15:01:33] <anonimasu> of the plc's ;)
[15:03:45] <anonimasu> I think I will.
[15:19:12] <bosone> my stupid idea for emc http://www.webalice.it/bosone77/index.html what do you think?
[15:23:13] <bosone> i think also to use vga as pwm generator using standard linux framebuffer devices
[15:23:37] <Jacky^> bosone: interesting
[15:24:55] <Jacky^> looks like youre searching an alternative emc way
[15:25:16] <Jacky^> an intensive hacking is needed :P
[15:25:27] <Jacky^> but its nice
[15:26:57] <bosone> the only thing is if writing to framebuffer (/dev/fb1) is instant refresh or frame refresh, i dont know
[15:27:31] <Jacky^> uhm
[15:27:36] <anonimasu> bosone: you could write a driver that would write the data in realtime to th address..
[15:27:52] <bosone> in case of frame refresh it's to slow
[15:28:48] <bosone> i'd like to use framebuffer for it's hardware abstraction
[15:29:48] <bosone> so everyone can put a secondary pci card and have dac or pwm output
[15:29:55] <Jacky^> bosone: this is kernel hacking..
[15:30:10] <bosone> no!
[15:30:14] <anonimasu> bosone: write a custom driver for it, that does what you would like to do..
[15:30:21] <anonimasu> just my 5c
[15:31:16] <anonimasu> bosone: I am not sure that the default framebuffer driver will do what you would like to do..
[15:31:55] <bosone> a custom driver for emc that write patterns to framebuffer devices
[15:32:19] <bosone> I am not sure too
[15:32:38] <anonimasu> um, that would be the way it would end up..
[15:32:49] <anonimasu> but, can you really use the linux framebuffer stuff to do it.
[15:39:56] <Jacky^> bosone: never seen the cathedral and the bazaar ?
[15:40:35] <Jacky^> http://www.apogeonline.com/openpress/doc/cathedral.html
[15:40:49] <Jacky^> is what you need ..
[15:41:33] <Jacky^> and.. the secret of linux !
[15:47:27] <anonimasu> :)
[16:06:45] <anonimasu> * anonimasu yawns
[16:32:07] <Jymmm> paul_c you awake?
[16:37:14] <Jymmm> paul_c you alive?
[16:47:40] <cradek> it's a bit early for him I think
[16:51:51] <Jymmm> ah
[16:52:25] <Jymmm> I'm just trying to binfd the right file to edit for key bindings
[16:52:36] <Jymmm> find
[16:52:59] <Jymmm> I file I edited makes no change.
[16:53:39] <Jymmm> tkemcSetKeyBings.tcl
[16:53:44] <Jacky^> ta -daaa !
[16:53:47] <Jymmm> tkemcSetKeyBindings.tcl
[16:53:52] <Jacky^> pcb finished :P
[16:54:36] <Jymmm> Jacky^ now for you to circuit test it
[16:54:54] <Jymmm> ohm out each and every trace!!!
[16:54:55] <Jacky^> ;)
[16:55:00] <jepler> finished with the design or with the etching, drilling, and assembly?
[16:55:10] <Jacky^> nahh
[16:55:32] <Jacky^> just the pcb doubleside draw
[16:56:27] <Jymmm> jepler you have any idea the file I need to edit ro remap the key bindings by chance?
[16:56:34] <Jymmm> s/ro/to/
[16:57:06] <jepler> Jymmm: not in tkemc or mini, no clue
[16:57:30] <cradek> do people still use those other substandard guis?
[16:57:42] <cradek> oops, inside thought
[17:02:46] <Jymmm> axis segfaults on me
[17:04:14] <Jymmm> jepler what about in axis ?
[17:04:39] <cradek> axis.py
[17:05:41] <Jymmm> doens't exist.
[17:05:57] <cradek> plat/.../bin/axis
[17:09:18] <Jymmm> root_window.bind ?
[17:10:16] <jepler> yes
[17:10:36] <cradek> if you're working on jogging, you probably will want to use bind_axis
[17:10:43] <jepler> If you want to bind keys to move an axis, use the 'bind_axis' convenience function. around line 1187 in my copy of axis.
[17:10:47] <jepler> er, what cradek said.
[17:14:26] <Jymmm> what is def bind_axis(a,b,d) ????????
[17:14:43] <Jymmm> specifically the a/b/d
[17:14:49] <Jymmm> a,b,d
[17:15:05] <cradek> the arguments
[17:15:23] <Jymmm> not the axis themselves (xyzabc) ?
[17:16:34] <Jymmm> btw this is rev 1.55
[17:28:15] <jepler> Jymmm: look at where it's used. THe arguments are the names of the keys and the number of the axis.
[17:29:26] <jepler> the first use of bind_axis binds Left and Right so that they move axis 0 (typically the X axis)
[17:29:47] <Jymmm> jepler I tried duplicating the lines and appending KP_%s but it didn;t like the bracketleft/right
[17:29:59] <Jymmm> =)
[17:30:38] <anonimasu> hm
[17:30:59] <jepler> You mean 'bind_axis("KP_bracketleft", ...)'? Yeah, there is no key KP_bracketleft
[17:31:58] <Jymmm> no, I tried duplicating the lines that have lambda in them
[17:32:21] <Jymmm> should I try just KP_Left instead?
[17:32:40] <jepler> you want to add another *call* to bind_axis, not change the definition of bind_axis
[17:34:27] <jepler> bind_axis("KP_Left", "KP_Right", 0)
[17:35:03] <Jymmm> ok, that what I just finished doing, trying now =)
[17:35:06] <anonimasu> *makes mental note to remember to do it*
[17:35:12] <anonimasu> Jymmm: have the machine assembled yet?
[17:35:18] <anonimasu> Jymmm: parts parts parts!
[17:35:19] <anonimasu> :D
[17:35:36] <Jymmm> anonimasu dont be an ass.
[17:35:46] <Jymmm> fuck! axis segfauleted agian
[17:36:16] <jepler> and similarly for axis 1 and 2. If you have an axis 3 then you'll have to use something else like KP_Divide and KP_Multiply
[17:36:21] <anonimasu> Jymmm: I am not being an ass.
[17:37:45] <jepler> I wish I had some idea what was causing your axis to segfault. axis never segfaults for me or for cradek..
[17:38:01] <jepler> except when we tried to run it on one of the bdi4 releases once
[17:38:11] <jepler> but that seemed to be general instability when running opengl software
[17:38:50] <Jymmm> jepler I can't get paste axis segfaulting. generic.run line 616: 13708 seg fault plat/nonrealtime/bin/$display -ini $INIFILE
[17:39:17] <cradek> Jymmm: does other opengl software work for you?
[17:39:27] <cradek> Jymmm: what do you do right before it crashes?
[17:39:45] <Jymmm> cradek power on, and try to jog then POOF!
[17:40:06] <Jymmm> cradek no clue on other opengl
[17:40:52] <jepler> is there a "core" file created when it segfaults?
[17:41:06] <jepler> if so, do you know how to use gdb to get a traceback?
[17:41:16] <Jymmm> no clue, and no
[17:41:30] <Jymmm> anonimasu Ah, I thought you were having a robin moment there.
[17:42:20] <anonimasu> Jymmm: I am mostly excited to see how it goes
[17:42:40] <anonimasu> Jymmm: and well excited to hear, how making your first parts feels :)
[17:44:41] <jepler> cradek: on my emc1 (simulator), stat.origin does not change when I hit shift-home, until I load a file.
[17:45:40] <jepler> so it looks like an emc1 simulator-only bug, maybe
[17:45:57] <cradek> oh, good (I guess)
[17:46:11] <Jymmm> anonimasu STILL waiting for parts.
[17:46:12] <cradek> I meant to try sim at home last night, but I forgot
[17:46:40] <bosone> i found a nice old article about framebuffer devices, with a small code example: http://www.linux-magazine.com/issue/03/EmbeddedGraphics.pdf
[17:47:09] <cradek> was it mrallen we were helping yesterday? I don't remember him saying he was running sim
[17:47:20] <cradek> I don't even know if bdi4 has sim
[17:47:51] <Jymmm> If it's GL based, it just might be this machine too.
[17:47:58] <jepler> bosone: you are talking about doing pwm, so basically if you want to do a 50% duty cycle you'll store a bright value in the channel for 50% of the scanline and a dark value for the rest?
[17:48:04] <jepler> Jymmm: yes, axis is opengl based.
[17:48:33] <cradek> Jymmm: see if you can run a few glxgears for a while
[17:48:51] <Jymmm> cradek ?
[17:49:02] <cradek> Jymmm: glxgears
[17:49:04] <jepler> glxgears is a standard opengl test program
[17:49:04] <cradek> Jymmm: run it
[17:49:11] <Jymmm> running now...
[17:49:26] <Jymmm> 215f in 5s = 43fps
[17:49:39] <Jymmm> 218f in 5 s = 49.6fps
[17:49:39] <cradek> 9719 frames in 5.0 seconds = 1943.800 FPS
[17:50:22] <cradek> 11178 frames in 5.0 seconds = 2235.600 FPS
[17:50:29] <cradek> yay nvidia
[17:50:37] <Jymmm> ok so it's avg 49 fps
[17:50:39] <cradek> sure wish it worked with realtime
[17:50:53] <cradek> Jymmm: resize it a few times - try to get it to crash
[17:50:59] <jepler> 56186 frames in 5.0 seconds = 11237.200 FPS
[17:51:04] <jepler> do I win something?
[17:51:47] <cradek> is that the default window size?
[17:51:54] <jepler> er, no
[17:52:04] <Jymmm> 50% 12fps, 75% 6fps
[17:52:24] <Jymmm> fullscreen 8fps
[17:52:35] <Jymmm> 1024*768*24
[17:52:57] <cradek> I get 140.8FPS fullscreen (1600x1200)
[17:53:05] <jepler> but this isn't about how many fps you get
[17:53:09] <jepler> this is about whether it crashes
[17:53:10] <cradek> right
[17:53:15] <cradek> or draws wrong
[17:53:18] <Jymmm> no crashing
[17:53:42] <cradek> does it look right (no holes in the gears)?
[17:53:58] <cradek> other than the center hole, of course
[17:53:58] <Jymmm> other than the center holes, no.
[17:54:10] <jepler> I was gonna ask...
[17:54:11] <Jymmm> 172fps now
[17:55:14] <Jymmm> resizing, graging the window around... no crash
[17:55:18] <Jymmm> dragging
[17:55:23] <cradek> huh
[17:55:27] <cradek> wonder what's wrong with your axis then
[17:55:33] <cradek> is this a bdi system?
[17:55:45] <Jymmm> 4.25 or .26
[17:55:58] <cradek> it would be nice if you could do the update and retest
[17:56:15] <cradek> I've lost track of all the bdi4 releases
[17:57:02] <jepler> looks like 4.23 is the newest one I have in iso, but I haven't tried to install it anywhere
[17:57:11] <cradek> I have 4.25 here
[17:57:18] <jepler> spare machine?
[17:57:27] <Jymmm> heh, how do I tell WHICH ver I actually have?
[17:57:34] <cradek> I have most of a spare machine here
[17:57:38] <cradek> Jymmm: no clue
[17:57:57] <Jymmm> * Jymmm kicks paul_c out of bed!
[17:58:39] <Jymmm> well, before it crashed, the numpad jogging worked.
[17:59:31] <jepler> that's not so useful, if it crashes whenever you jog
[17:59:41] <Jymmm> Um whats the diff between linux_rtai and nonrealtime ?
[18:00:30] <Jymmm> oh, nm. didn't relaize it's a symlink
[18:00:59] <anonimasu> :)
[18:04:17] <Jymmm> but is that just for a simulator?
[18:05:05] <anonimasu> hm
[18:05:15] <anonimasu> [emc-developers] how to crash a mill
[18:05:31] <anonimasu> :/
[18:06:34] <Jymmm> Step 1) Get yourself 6 cases of Jose Quervo Gold Tequilia (half gallon platic containers)
[18:06:39] <anonimasu> ye
[18:06:40] <anonimasu> ya
[18:06:42] <anonimasu> heh
[18:07:18] <Jymmm> Step 2) Get an IV Kit (no regulator required)
[18:07:36] <anonimasu> lol
[18:09:45] <Jymmm> Anyone know the difference between 'bind AutoBindings' and 'bind ManualBindings' by chance?
[18:09:55] <anonimasu> no.
[18:10:03] <anonimasu> :D
[18:11:09] <anonimasu> hm.. maybe I should draw this stuf.f.
[18:24:21] <anonimasu> and stop pondering on how to
[18:47:54] <Jymmm> anonimasu: DONT DO THAT! Anything but that!!!
[18:48:38] <Jymmm> anyone know if you can add images to the wiki?
[19:23:09] <anonimasu> Jymmm: other stuff ;)
[19:28:19] <anonimasu> Jymmm: slide throttle body for the car
[19:43:31] <cradek> well jepler and I installed bdi 4.25 and the software GL is definitely very broken
[19:47:01] <Jymmm> what is broken?
[19:49:29] <anonimasu> :
[19:56:04] <cradek> something about GL
[19:56:11] <cradek> axis displays wrong and then glxgears displays wrong
[19:56:19] <cradek> we haven't seen either crash yet, though
[19:56:27] <cradek> we're going to try updating various things.
[19:57:05] <Jymmm> install it on a shitty video card
[20:38:37] <Jacky^> opengl ?
[20:58:26] <alex_joni> greetings gents
[20:58:58] <anonimasu> hello alex
[20:59:04] <alex_joni> hey anders
[20:59:08] <alex_joni> what's up?
[20:59:58] <anonimasu> making parts :9
[21:00:00] <anonimasu> :)
[21:00:02] <alex_joni> cool
[21:00:07] <alex_joni> * alex_joni is in d�sseldorf
[21:00:23] <anonimasu> nice
[21:00:31] <anonimasu> I'll be going there the weekend after the next
[21:00:35] <alex_joni> heading home on monday
[21:00:48] <alex_joni> what's "there" ?
[21:01:34] <anonimasu> a racing event
[21:02:36] <anonimasu> got a customer I've helped build a plc system for a racecar
[21:02:45] <anonimasu> going along as tech :)
[21:03:17] <alex_joni> nice
[21:03:33] <alex_joni> been kinda busy moving around lately
[21:03:36] <alex_joni> :(
[21:03:45] <alex_joni> I'll be home on monday night
[21:03:50] <alex_joni> going away on tuesday
[21:04:13] <alex_joni> 1 week to some training
[21:04:18] <alex_joni> the next week to a customer
[21:04:28] <anonimasu> :/
[21:04:29] <alex_joni> then "maybe" a week I'll be home
[21:04:43] <alex_joni> the week after that I'll probably be in Lund
[21:04:44] <alex_joni> ;)
[21:05:32] <anonimasu> :
[21:05:33] <anonimasu> hehe
[21:05:53] <alex_joni> yeah
[21:06:04] <alex_joni> was away this week, and 2 weeks before too
[21:06:13] <anonimasu> what does the lady say about it?
[21:06:23] <alex_joni> well... the 2 weeks before she was with me
[21:06:34] <alex_joni> the next week (the training) she'll be there too ;)
[21:07:15] <anonimasu> :)
[21:07:17] <anonimasu> nice
[21:07:41] <alex_joni> yeah
[21:07:51] <alex_joni> and for the rest I need to compensate with gifts
[21:07:52] <alex_joni> :D
[21:10:21] <anonimasu> haha
[21:10:23] <anonimasu> yeah
[21:10:29] <alex_joni> you know how it goes :D
[21:10:41] <alex_joni> anything interesting with emc lately?
[21:11:27] <paul_c> 2.6.12 kernel coming up.
[21:11:36] <alex_joni> oh.. greetings from germany paul
[21:12:15] <paul_c> where's the postcards ??
[21:16:52] <alex_joni> where's the postal addy ?
[21:41:41] <alex_joni> * alex_joni is gone in 5 mins
[21:41:59] <cradek> paul_c: in your rtai build are there any floating point options?
[21:42:17] <zwelch> is there any reason why EMC2 requires RT, even if all one wants to do is try it out (e.g. in backplot mode)? (yes, i grok why one needs RT for controlling actual hardware, but i don't have any yet)
[21:42:39] <cradek> zwelch: because nobody has made the non-rt simulator work yet
[21:42:49] <cradek> zwelch: emc1 does not require realtime unless you are controlling a machine
[21:42:59] <paul_c> cradek: There is floating point math used in realtime (by all MC variants)
[21:43:24] <zwelch> paul_c: but that shouldn't matter in a sim, yah?
[21:43:27] <cradek> paul_c: I know, but we've found a problem with fp when rt is running on your bdi
[21:43:42] <cradek> paul_c: so I'm asking if you can confirm for sure that the rtai fp options are set correctly
[21:44:57] <cradek> zwelch: you can build emc1/sim on nearly any linux.
[21:45:48] <zwelch> i'm a software developer, so i was trying to "get with the program" by trying emc2
[21:46:09] <alex_joni> bye guys
[21:46:13] <cradek> zwelch: cool, I understand
[21:46:23] <paul_c> cradek: Checked - servo loop uses fp.
[21:46:24] <zwelch> i looked at the configure.in script and (briefly) thought about trying to rewrite it without the RT requirements
[21:46:42] <cradek> paul_c: I know emc uses fp, let me explain my question better
[21:47:25] <paul_c> Are the fp flags being saved by RTAI when it switches context.
[21:47:28] <cradek> paul_c: the problem is that the rt is messing up the non-rt fp stack. I would like you to confirm that you built rtai with fp support, becuase maybe, this tells it to save and restore the processor's fp stack.
[21:47:38] <cradek> no, not just flags, the stack
[21:48:09] <cradek> if it *isn't*, emc will still work!! but any non-rt fp (like axis) will be porked.
[21:48:21] <zwelch> unfortunately, the fact that the selected RT system provides a `rt-config` program that provides the build options makes that a little tricky to do right; unless someone can suggest a better idea, i'll take a look at the EMC1 base and see how it handles gracefully degrading in the case of no-RT
[21:51:51] <paul_c> If you installed rtai-dev, "rtai-config --config" will point you to the options used.
[21:52:18] <cradek> cool
[21:52:20] <paul_c> CONFIG_RTAI_MATH & CONFIG_RTAI_FPU_SUPPORT both enabled.
[21:52:26] <cradek> yes it is turned on.
[21:52:44] <cradek> then, rtai itself is broken.
[21:53:12] <cradek> this is why axis only works for me on my rtlinux machine.
[21:53:53] <cradek> jeff and I will write a bug report and get it to the rtai guys
[21:54:37] <jepler> we've only caused the problem under emc so far, though.
[21:54:37] <paul_c> Before you do. Test with the magma branch.
[21:54:48] <jepler> in fact it only happens when we take the machine out of estop and turn it on
[21:54:56] <jepler> that's not a very good bug report to submit to the rtai guys
[21:55:12] <cradek> true, we need a very simple rt module
[21:55:14] <paul_c> I doubt Paolo will that interested in Vesuvio bug reports..
[21:55:44] <cradek> is one unstable and the other stable?
[21:56:02] <paul_c> kernel & rtai-dev packages available from theapt mirrors
[21:56:58] <paul_c> vulcano is the curren "stable" branch.
[21:57:21] <jepler> volcano?
[21:57:46] <paul_c> Vulcano - a.k.a. rtai-3.2
[21:58:00] <cradek> vulcano [sic]
[21:58:29] <paul_c> that's the spelling RTAI uses.
[21:58:52] <zwelch> will emc work with the new rtai fusion branch?
[21:59:13] <paul_c> emc1 or 2 ?
[21:59:17] <zwelch> * zwelch shrugs
[21:59:30] <paul_c> fat chance to both.
[21:59:42] <cradek> zwelch: today, I heartily recommend rtlinux-3.2
[21:59:57] <cradek> zwelch: it will run both emc1 and emc2 (and correctly)
[22:00:45] <mrallen> end-mill question: 2-flute vs 4-flute for cutting wood / other materials. what should I look for and what's best for what?
[22:01:11] <cradek> mrallen: 2 flute will let your (wood or Al) chips out better.
[22:01:39] <paul_c> three & four flute are much more rigid
[22:01:54] <mrallen> more flutes == harder material, in other words
[22:01:56] <paul_c> better suited to hig(er) speed machinging
[22:02:25] <paul_c> darn this keyboard..
[22:02:26] <anonimasu> well, apparently the 3 flute cutters are most rigid..
[22:02:29] <anonimasu> or well machines the best..
[22:02:52] <cradek> paul_c: I think mrallen has a sherline... not much high speed anything there (nothing derogatory meant - I also have sherline stuff)
[22:03:11] <anonimasu> 3 flues is optimal for alu atleast..
[22:03:14] <paul_c> * paul_c also has a Sherline
[22:03:15] <mrallen> sherline is limited to 2800rpm unless you get a 10k rpm mod (available from sherline also)
[22:03:29] <anonimasu> it has to do with the flute gap..
[22:03:44] <anonimasu> and at higher speeds with the frequency which the edges hit the material..
[22:03:46] <mrallen> i'm cuttind wood at the moment, but plan to do some 6061 after i'm comfortable with the machine
[22:03:50] <paul_c> mrallen: If you go for the 10K spindle, use the ER16 collets.
[22:04:15] <mrallen> paul_c: ER16 available from sherline? i'm still newish to this.
[22:04:26] <mrallen> paul_c: where do you get endmills?
[22:04:32] <paul_c> for gummy materials, coolant is a must...
[22:04:50] <anonimasu> you can dry mill alu nicely, at the right feeds/speeds
[22:04:51] <paul_c> * paul_c uses flood coolant with aluminium.
[22:05:12] <anonimasu> but coolant makes it cut nicely
[22:05:20] <mrallen> paul_c: what do you do for an enclosure? is the coolant controlled from EMC?
[22:05:30] <paul_c> mrallen: Sherline do an ER16 spindle for the industrial range.
[22:06:11] <paul_c> End mills, I get from J&L, Cromwell, & other local suppliers.
[22:06:19] <mrallen> paul_c: please explain industrial. is this something I can easily use with my 2000?
[22:06:57] <paul_c> Sherline do a range of slides & spindles aimed at custom industrial applications.
[22:07:19] <paul_c> Much of it is interchangable with the standard mills & lathes.
[22:07:28] <cradek> I like "Alumicut" for low-speed machining of Al
[22:07:40] <cradek> I think it's kerosene and oil
[22:07:55] <mrallen> no mention of ER16 on sherline's web site
[22:08:17] <paul_c> You'd need to ring Sherline & talk to Craig.
[22:08:48] <paul_c> I have an ER16 spindle here.
[22:11:21] <mrallen> do you have an enclosure for your sherline?
[22:11:31] <mrallen> and waht do you use for flooding coolant?
[22:12:00] <paul_c> mrallen: I think there is also an ER16 adaptor that screws on to a standard spindle..
[22:13:48] <mrallen> i haven't even gotten around to propery calibrating this thing (backlast, etc). the first couple of weeks has been wood play. i finally broke down and got visuallmill to create tool paths.
[22:14:08] <anonimasu> mrallen: if you want help with it, ask :)
[22:14:46] <zwelch> is the NIST rcslib that i found in the emc cvs is the appropriate version to use for building v1 and v2?
[22:14:52] <mrallen> i will need help when it comes to calibration. i need to first get a dial indicator. then i will need to figure out how to test/fix backlash.
[22:15:05] <mrallen> paul
[22:15:16] <mrallen> paul_c: does your headstock allow spindle control from EMC?
[22:15:56] <paul_c> Not yet - I had been waiting for someone to send me some hardware to test.
[22:16:34] <mrallen> now that i've started playing with this thing, i'm finding an endless need for new parts -- like a 4th axis
[22:17:05] <paul_c> rotary table with angle plate
[22:17:12] <paul_c> tilting table
[22:17:16] <mrallen> what do i need to cut threaded parts on the 2000? or is that not possible and i need a lathe instead?
[22:17:29] <paul_c> You can do thread milling
[22:17:46] <mrallen> rotary table is required, i presume ??
[22:17:49] <anonimasu> lol
[22:17:50] <anonimasu> no
[22:18:04] <anonimasu> just 3 axis interpolation
[22:18:37] <mrallen> what type of end-mill do you use?
[22:19:14] <anonimasu> a threadmill
[22:19:29] <anonimasu> but they are super expensive
[22:19:46] <anonimasu> you might be better off with a tapping head
[22:19:59] <cradek> often, all you need is a tap and die set (or a lathe)
[22:20:29] <cradek> depends what you're making, but don't forget taps are cheap, easy to use, and effective
[22:20:38] <mrallen> i was thinking of making a large wooden screw
[22:20:52] <paul_c> mrallen: You want to get Jerry G to make a house call.
[22:21:00] <mrallen> who is that
[22:21:08] <cradek> a rotary axis (and tailstock) and a 60 degree V tool will do that for you
[22:21:18] <paul_c> He's on the SherlineCNC list (I think)
[22:22:14] <paul_c> on second thought... Just the sherlien list.
[22:22:21] <zwelch> so, i'm not sure if i believe it's possible to build emc without rt{ai,linux}. emc1 wants the nist rcslib, whose configure.in seems to have been the source of the same parts for emc2's
[22:22:36] <cradek> zwelch: that's exactly right
[22:22:56] <paul_c> The other alternative. Ask on the lists to see who is in your area and see if they want to meet for a beer.
[22:23:02] <mrallen> righ now i need to make an enclosure for the sherline ... cutting wood is spitting out a lot of fine dust. my lab is in the basement, but i have computers there, too
[22:23:21] <mrallen> paul_c: yep, need to find a few people in the DC, NoVA area
[22:23:35] <paul_c> zwelch: EMC1/rcslib can be built without RT
[22:23:42] <zwelch> cradek: i guess i read too much into your earlier comment (<cradek> zwelch: you can build emc1/sim on nearly any linux.)
[22:23:54] <paul_c> mrallen: Matt Shaver, Fred Proctor.
[22:24:02] <zwelch> paul_c: how does one get around the "checking for RT dir.." configure check?
[22:24:45] <zwelch> or was this where i was suppose to find the rtai-dev bits?
[22:25:06] <zwelch> * zwelch is on Gentoo, so that's not as easy as I assume it would be in Debian
[22:25:24] <paul_c> the check for RT dir is needed for some RTAI/RTLinux headers that are required.
[22:26:03] <paul_c> the RT check is a simple search for rtai/rtlinux signature files.
[22:29:58] <paul_c> mrallen: If you are anywhere near the Gaithersburg/Baltimore area...
[22:32:32] <mrallen> paul_c: 30 miin
[22:35:37] <paul_c> then there is Springfield, VA....
[22:36:07] <paul_c> steves_logging: That's your cue...
[22:38:10] <paul_c> mind you... Dulles & Baltimore are only eight hours or so from here.
[22:39:05] <cradek> zwelch: you can build the emc1 sim, which requires rcslib, without realtime
[22:39:19] <cradek> zwelch: sorry if I was not being clear
[22:39:53] <zwelch> i'm not sure if it counts as "a not-RT" if i'm going to have to create a patched kernel just to get the rt userland package to build (even if i won't have to reboot into the rt kernel), and I can't see how to make rcslib build without a built userland rt-config script
[22:39:59] <paul_c> zwelch: and you can build emc1/rcslib sim on a 2.6 kernel.
[22:40:16] <zwelch> yah, i'm only interested in 2.6
[22:40:35] <paul_c> How are you trying to build rcslib ?
[22:41:12] <zwelch> well, i've tried the etc/configure script, which does the RT dir check straight off
[22:41:25] <zwelch> (oh, and i got the source from the emc repo, as i indicated somewhere above)
[22:41:35] <cradek> oh, I forgot about that mess
[22:41:47] <cradek> try something like cd rcslib; ./buildrcs linux_2_4_20
[22:41:51] <paul_c> OK... cd rcslib/src
[22:41:58] <paul_c> make PLAT=linux
[22:42:25] <cradek> ok, too many cooks spoil the soup, I'll butt out
[22:42:29] <zwelch> heh
[22:42:32] <cradek> there are too many ways to do each thing
[22:42:34] <jepler> cradek: maybe I still have that fc4+rtai installation at home.
[22:42:43] <cradek> jepler: that would be great
[22:43:14] <paul_c> PLAT=linux is what the compile farm is using on a 2.6 kernel.
[22:46:15] <zwelch> paul_c: that trick has my computer busy compiling, but i did notice some icky insmod warnings ("-a: file not found!")
[22:48:40] <paul_c> the insmod warnings can be ignored..
[22:49:30] <paul_c> ksyms: QM_MODULES: Function not implemented
[22:50:09] <zwelch> also, it looks like the emc configure script needs a check to see if the tclx(-dev) package(s) exist; otherwise, the build fails half-way through with a missing header file
[22:51:00] <paul_c> rcslib/etc/configure is only needed for 2.4.xx-rtai-3.x builds
[22:51:56] <paul_c> could probably modify it for non-RT 2.6 builds, but it really isn't worth the effort (in my opinion)
[22:52:16] <zwelch> anything that gets you closer to "./configure; make; make install" is worth it
[22:52:39] <paul_c> not for the rcslib/emc1 branch
[22:53:05] <zwelch> * zwelch shrugs
[22:53:37] <paul_c> you get it working for 2.2 & 2.4 builds
[22:53:50] <paul_c> the usrs will want 2.6 support
[22:54:11] <paul_c> then they have little or no incentive to move to the emc2 tree
[22:54:23] <zwelch> heh
[22:54:40] <paul_c> in the meantime, valuable developer resources are being wasted on a dead end tree.
[22:55:05] <zwelch> i tend to agree
[22:56:00] <zwelch> of course, i'm a developer and see it very much from the same point of view
[22:56:33] <paul_c> A decision was made at the codeFest & later ratified by the board that rcslib/emc1 would only recieve bug fixes and any further development would take place in the emc2 tree.
[22:57:18] <zwelch> the key ingredient there is how you choose to define "bug fixes"
[22:57:55] <zwelch> but i'm just playing devil's advocate there; i think that's a fine policy
[22:58:21] <paul_c> "bugs" are anything that produces unexpected results during operation
[22:58:37] <paul_c> A screwball build system is not a bug.
[22:58:51] <zwelch> "it unexpectedly failed to install" is not a bug? :)
[22:58:54] <paul_c> It's just a pain in the arse.
[22:58:56] <zwelch> hehehe
[22:58:58] <zwelch> * zwelch agrees
[22:59:56] <paul_c> * paul_c doesn't see zwelch listed @ sf.
[23:00:51] <zwelch> * zwelch is... "evaluating"... emc for future cnc-like projects
[23:00:57] <anonimasu> :)
[23:01:58] <zwelch> i use quotes because i don't want to imply anything more than a keen desire to experiment with g-code programming and an innate belief that the investment could be made to pay off ;)
[23:03:08] <anonimasu> that depends soley on what scale you wish for it to pay off..
[23:03:28] <zwelch> as an exclusively open source developer, i thusly have a strong interest in the underlying tools required to sate my eclectic passions
[23:04:45] <zwelch> anonimasu: for perspective, my 50 year plan is "conquor the world with self-designed and fabricated robot army". Fill in the intervening goals for yourself. ;)
[23:06:57] <paul_c> zwelch: You up in Oregan ?
[23:07:07] <zwelch> more immediately, i have almost 15 years of professional development experience on Unix-like systems; my Linux experience started by adapting the ARM port to the ruggedized handheld for which i was firmware and system software architect
[23:07:23] <zwelch> paul_c: s/gan/gon/ and yes ;)
[23:07:49] <paul_c> picky bloody spell checkers... ;)
[23:07:51] <anonimasu> hm, that kind of rings a bell back in my head
[23:08:34] <paul_c> There is a hardcore EMC user group in the Portland area.
[23:08:53] <zwelch> i'm a few blocks away from Oregon State University (where they opened the OSU Open Source Lab lsat year) in Corvallis
[23:09:19] <zwelch> paul_c: i figure there must be a few around here, given the university and high tech sector concentration
[23:09:58] <paul_c> * paul_c occasionally drives up the west coast.
[23:10:39] <zwelch> there are CNC classes at the local community college starting a couple of weeks, and I've thought about going to meet the instructors to see if I might benefit
[23:10:57] <anonimasu> zwelch: a tip is to learn about machining before you start with cnc..
[23:11:03] <anonimasu> even though cnc:ing is much more fun..
[23:11:10] <zwelch> i've done some woodworking
[23:11:18] <anonimasu> having a perspective on feeds/spees/general machining knowhow helps
[23:11:39] <zwelch> well, i am a bit cross-trained
[23:11:53] <anonimasu> ah well, it's just a tip..
[23:12:01] <zwelch> while working on the rugged handheld, i had the chance to learn about the EE and ME sides of things
[23:12:16] <anonimasu> experienced operators can often teach you how to do stuff you never had a idea on how to do
[23:12:46] <zwelch> well, that's the most important thing i know - ask others if you don't know the answers yourself ;)
[23:13:28] <anonimasu> you would probably benefit from such a class..
[23:13:30] <anonimasu> I know I would :)
[23:14:04] <zwelch> yeah, the big question is "would it be an order of magnitude easier if i had some basic practical training first?"
[23:14:54] <zwelch> having worked in the field before going back to school and finishing my undergrad degree, my Comp Sci classes were really a breeze
[23:15:42] <zwelch> * zwelch wonders if anyone here plays a musical instrument and would like to attempt to draw a parallel between that learning process and machining/cnc :)
[23:16:19] <anonimasu> I do, but I cant draw a paralell between it..
[23:16:20] <anonimasu> :)
[23:19:32] <zwelch> well, if i had to, i'd say: "there is no 'wrong way' to learn to become a virtuoso on an instrument - it just takes a lot of practice"; i would imagine the same would true of any path to becoming a CNC guru, with the exception of those routes that would cause personal bodily harm ;)
[23:20:10] <zwelch> * zwelch has never heard of someone loosing a finger while learning to use a violin
[23:20:13] <anonimasu> zwelch: yeah that's true..
[23:22:54] <anonimasu> hm, I think I need to grab the other laptop to draw this part :/
[23:23:15] <anonimasu> this drawing will end up in teh scrap bin otherwise
[23:24:43] <anonimasu> I think that will be the best way..
[23:26:46] <Jacky^> * Jacky^ yawns*
[23:26:52] <Jacky^> :)
[23:26:58] <Jacky^> hello anonimasu
[23:27:21] <zwelch> woot. i've got emc1 built and the sim running
[23:27:51] <Jacky^> * Jacky^ playng with some girls around ...
[23:33:04] <robin_sz> meep?
[23:33:09] <anonimasu> https://www.lionsracing.de/www/cars/LR03/#
[23:33:11] <anonimasu> cute.
[23:33:13] <anonimasu> hey robin
[23:33:18] <robin_sz> dude!
[23:33:24] <anonimasu> ?