#emc | Logs for 2007-01-17

[00:31:47] <anonimasu> night
[00:48:27] <A-L-P-H-A> SWPadnos, are you sending me usps spam? :D
[00:48:37] <SWPadnos> I don't think so :)
[00:48:46] <A-L-P-H-A> :) I'm just kidding. :)
[00:48:57] <SWPadnos> unfortunately, it took me like 45 minutes to scrape all the ice off my van, so the package didn't get to the PS today :(
[00:49:00] <A-L-P-H-A> you supplied them my email addy...
[00:49:16] <A-L-P-H-A> it took me 25 minutes...
[00:49:24] <A-L-P-H-A> my friend was not happy with me being late.
[00:49:26] <SWPadnos> heh
[00:49:32] <A-L-P-H-A> she kinda didn't say anything... but she understood.
[00:49:44] <A-L-P-H-A> then felt bad cause I drove in slush, and cold just to spend time with her.
[00:50:03] <A-L-P-H-A> BTW, she's married, and no I'm not tying to get into her pants... though we did have a talk about how flirty she is with me...
[00:50:03] <SWPadnos> there had been snow, then ice for a while, then more snow. I left the first layer of snow to make it easier to clear the ice off, but that didn't work on the vertical surfaces (about 100 ft^2 of windows)
[00:50:19] <SWPadnos> oh good - get any answers?
[00:50:47] <A-L-P-H-A> SWPadnos, she doesn't realize what's she's doing, and just feels TOO comfortable with me... I knew that, but just wanted her to be more aware in the future...
[00:51:01] <SWPadnos> ah. good to get it "out there" at least
[00:51:40] <A-L-P-H-A> 1.) she pinched my ass. 2.) she's straddled me while i was sitting on the couch. 3.) she's complete pressed up against me to touch my face. 4.) she plays footsies with me and my outter thigh. hmmmm.... think that's about it.
[00:53:00] <SWPadnos> oh, well if that's it - no problem :)
[00:54:59] <A-L-P-H-A> yeah... she hasn't french kissed me, or grabbed my member yet... I think I would be super uncomfortable with that... only way I would be comfortable would have been in the past, when she WASN'T MARRIED.
[00:55:22] <A-L-P-H-A> she's just way to friendly... but anyways. :)
[01:21:57] <CIA-8> 03jepler 07TRUNK * 10emc2/configs/common/emc.nml: increase nml buffer size for 64-bit machines
[01:25:38] <A-L-P-H-A> hey Jymmmm
[01:39:49] <jepler> I got gcam to run on dapper, but I don't think its gcode output is right
[01:40:00] <jepler> it uses ; for comments instead of (..) and doesn't put m2 at the end, but that's the least of the problems
[01:40:21] <jepler> it seems to have spurious cutting moves through what should be the remaining material
[01:41:45] <jepler> http://emergent.unpy.net/index.cgi-files/sandbox/gcam-vs-axis.png
[02:03:47] <DanielFalck> jepler: let Twingy know
[02:12:40] <Twingy> hrm?
[02:14:23] <jepler> Twingy: I hear I should talk to you
[02:14:29] <DanielFalck> Twingy, did you see jepler's comments above?
[02:14:29] <Twingy> ok
[02:14:36] <Twingy> just saw the link
[02:15:03] <Twingy> 2007.01.09 has a few bugs, looks like renesis posted them on bugzilla for me to take a look at this weekend
[02:15:05] <jepler> Twingy: I made a few changes to gcam-2007.01.09 to make it compile on dapper, ran the Continuity program, changed it so that emc2 will load it, and got the results shown in the screenshot
[02:15:17] <Twingy> kk
[02:15:36] <Twingy> some one on the discussion forum said they made some kind of modifications to make it compatible with emc
[02:15:39] <jepler> the changes I made are here: http://emergent.unpy.net/index.cgi-files/sandbox/gcam-2007.01.09+dapper.patch
[02:15:43] <Twingy> but they never followed up after my reply
[02:15:58] <Twingy> file not found?
[02:16:13] <jepler> hm let me find the right URL
[02:16:28] <Twingy> you probably won't see a "stable" release for another 2-3 months
[02:16:42] <Twingy> the last release was pretty close to stable though
[02:16:49] <Twingy> the december one
[02:17:27] <jepler> try that URL again
[02:17:42] <jepler> if it's probably just transitory brokenness I won't pursue it any further
[02:17:50] <jepler> it is too bad that gcam won't compile on dapper OOTB though
[02:18:22] <jepler> yuck that diff looks bigger than it really is due to all the differences in the configure script
[02:18:32] <Twingy> um
[02:18:37] <Twingy> other than remove the gtk 2.10 stuff
[02:18:39] <Twingy> what did you do?
[02:18:48] <jepler> - export CFLAGS="-O3 -std=c99 -Wmissing-prototypes -pedantic -ffast-math -fomit-frame-pointer"
[02:18:48] <Twingy> cause 90% of that is autoconf junk
[02:18:52] <jepler> + export CFLAGS="-O3 -std=gnu99 -Wmissing-prototypes -pedantic -ffast-math -fomit-frame-pointer"
[02:18:58] <Twingy> so not much point in a patch
[02:19:13] <Twingy> are you not using gcc?
[02:19:25] <jepler> under -std=c99, <math.h> does not define M_PI
[02:19:39] <jepler> on dapper and another system using gcc4.1.
[02:19:49] <Twingy> right, I actually fixed that on the windows release the day after
[02:20:09] <Twingy> gcam wasn't supposed to use M_PI, so fixed that
[02:20:11] <jepler> -std=gnu99 makes it be defined
[02:20:28] <Twingy> um
[02:20:35] <Twingy> what are the changes to make it "work with emc" ?
[02:20:42] <jepler> oh those I did by editing the output
[02:20:46] <Twingy> * Twingy thought he was using fairly standard g-codes
[02:20:48] <jepler> change ; ... to ( ... )
[02:20:54] <jepler> add M2 to the end
[02:21:09] <Twingy> ah, emc doesn't grok semicolon?
[02:21:16] <jepler> no, it only uses (...) for comments
[02:21:31] <Twingy> oh ok, anything other than that?
[02:21:35] <jepler> that was it
[02:21:38] <Twingy> kk
[02:21:51] <Twingy> I will make Export G-Code smarter
[02:22:00] <Twingy> and let you choose profiles then
[02:22:30] <jepler> looks like the assumption that ";" is a comment character is spread all throughout libgcode
[02:22:43] <jepler> * jepler takes a look at the source
[02:22:56] <Twingy> yep, but a quick pass with find / sed and fixed
[02:23:47] <Twingy> I might do something more elaborate though
[02:24:30] <Twingy> perhaps this sunday I will have another release which fixes most of the bugs in 2007.01.09
[02:25:00] <Twingy> you should install gtk 2.10 though
[02:25:07] <Twingy> it makes the treeview so much better
[02:25:16] <Twingy> (gtk 2.10 is 6 months old btw)
[02:25:25] <Twingy> not bleeding edge stuff
[02:27:25] <jepler> I appreciate that you feel that way, but dapper is at 2.8 (not 2.6 as I put in the patch) and there doesn't seem to be an easily installable 2.10.
[02:27:58] <Twingy> ah, I just went into freebsd ports and did "make install" and it just worked *shrug*
[02:28:40] <jepler> I suspect "edgy" is 2.10 as well, but the emc people made a decision to skip it and skip with the Ubuntu LTS release since it has a longer shelf life
[02:29:16] <Twingy> well sooner or later it'll catch up
[02:29:23] <Twingy> probly within a few months I guess
[02:29:53] <jepler> no, I doubt they will ever change the gtk version for dapper
[02:30:32] <Twingy> k...
[02:31:05] <SWPadnos> it's not a retro thing - the idea is maximum stability for 5 years ...
[02:31:22] <Twingy> k
[02:56:50] <owad> Is 125 in-lb a reasonable amount of torque for a rotary table on a small mill?
[02:58:45] <owad> There are some rotary tables on eBay that look like a good deal: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=160072564167
[02:59:45] <jepler> it's certainly overkill if you're using the rotary only for indexing
[02:59:50] <jepler> 4.5RPM is sloooooooowwwwwwwww
[03:00:36] <jepler> I don't see that it's dual shaft so adding an encoder may be tricky too
[03:01:17] <jepler> nevermind about that last bit -- I looked at the additional photos
[03:02:04] <jepler> no, I say it again -- not dual shaft .. http://www.in-front.com/ebay/pics/rotary_ml.JPG
[03:02:08] <SWPadnos> there are people using two of those to make a B/C rotary head
[03:02:35] <owad> I wonder if it'd be worthwhile, even if one were to replace the motor. There aren't many manual rotary tables you can get for $50, either.
[03:02:58] <jepler> I wonder how the gear's backlash is
[03:03:05] <SWPadnos> it's a satellite motor ...
[03:03:14] <SWPadnos> satellite antenna positioner
[03:03:19] <skullworks> doesn't look complete
[03:03:31] <SWPadnos> s/antenna/dish/ (need more coffee, I guess)
[03:04:11] <skullworks> looks like its up to you to build the mounting chassis
[03:04:17] <owad> I guess that explains the speed
[03:16:10] <owad> Can it be any worse than retrofitting one of those little harbor freight rotary tables?
[03:17:39] <SWPadnos> check the archives for the Yahoo group "CNC_Toolkit"
[03:17:53] <SWPadnos> I don't remember what people say about backlash or what they had to do to them
[03:18:08] <owad> thanks
[03:18:13] <SWPadnos> I'm prettywsure there was a discussion about them there
[03:18:20] <SWPadnos> ... pretty sure ...
[03:18:50] <owad> about the harbor freight tables or the satellite dish parts?
[03:19:06] <SWPadnos> satellite parts
[03:19:28] <SWPadnos> available here: http://www.in-front.com/
[03:20:01] <SWPadnos> one guy did mention that the top speed is fairly limiting
[03:21:42] <SWPadnos> there are two discussions I found: one started on 8/2/2005, title "[CNC_Toolkit] 5 axis Gantry router - satellite part alternative??"
[03:22:03] <SWPadnos> one started 5/25/06, title "[CNC_Toolkit] Satellite rotary unit"
[03:22:12] <owad> this looks like another: http://groups.yahoo.com/group/CNC_Toolkit/message/1363
[03:22:22] <owad> oh wait, that's ancient
[03:23:21] <owad> same seller, though
[03:23:29] <owad> I guess he's been selling these for years?
[03:23:48] <SWPadnos> looks like it wasn't really plug-n-play for people - they made changes
[03:23:54] <SWPadnos> seems that way
[03:24:09] <SWPadnos> apparently the guy claims that the mechanism has zero backlash
[03:29:09] <owad> He's got them for $46 shipped over here: http://www.in-front.com/
[03:29:35] <owad> oh
[03:29:35] <owad> ha
[03:29:44] <owad> I need to read closer. :-)
[03:32:52] <owad> Will emc allow me to control my XYZ-axes with a 5i20 and my a-axis through the parallel port?
[03:34:51] <jepler> owad: yes, you can mix and match different hardware drivers without problem
[03:37:57] <owad> thanks
[03:49:26] <CIA-8> 03jmkasunich 07TRUNK * 10emc2/src/hal/utils/halcmd.c: fix handling of loadusr -W when component becomes ready and then immediately exits
[04:24:12] <maddash> I just discovered `G6.2` -- "nurbs curve" (http://www.jssmdx.com/Reference/GCodeMill.htm) -- does emc support this?
[04:24:28] <SWPLinux> no
[04:24:33] <maddash> it would be so neat to approximate my splines with nurbs instead of arcs
[04:25:26] <maddash> SWPadnos: what algorithm do you use to approximate complex curves (ie splines)?
[04:25:35] <SWPLinux> I don't do that ;)
[04:26:52] <tomp> jepler: know of a list of the click events available to tkinter? i've used up <Double-1> thru 2 and want 1 more...
[04:27:14] <tomp> i got adjustable scaling up by 10 with <Double-1> and down by 10 with <Double-3> and restore original scale with <Double-2>, and wanted restore original main value with maybe <Shift-Double-1> but cant find the magic words
[04:27:24] <SWPLinux> tomp: jepler went to bed. you may get an answer tomorrow ...
[04:27:32] <tomp> thanks
[04:28:11] <maddash> jepler sleeps?!
[04:28:33] <maddash> j/k. heh. I lose control of my humor when it gets late.
[04:28:46] <SWPLinux> that's what he says anyway - I don't know that it's true
[04:34:18] <cradek> Q: What did Hannibal say when he saw 1,000 elephants coming over the hill?
[04:34:37] <cradek> A: "Look, there's 1,000 elephants coming over the hill."
[04:34:48] <SWPadnos> heh
[04:35:02] <cradek> I didn't know "elephant jokes" existed, but I guess they do
[04:35:12] <SWPadnos> oh yeah
[04:35:23] <SWPadnos> I'm just waiting for the "blonde elephant" jokes
[04:35:56] <maddash> isn't hannibal stating the obvious?
[04:36:20] <cradek> Q: What did he say when he saw 1,000 elephants with sunglasses on, coming over the hill?
[04:36:32] <SWPadnos> ?
[04:36:33] <cradek> A: Nothing; he didn't recognize them.
[04:36:50] <cradek> ok I'll stop
[04:37:03] <maddash> arghhhh gimme that full-text!
[04:37:28] <cradek> http://paul.merton.ox.ac.uk/misc/elephants.html
[04:37:40] <cradek> they're not all good - those are my favorites I think
[04:38:07] <maddash> looool I was referring to this IEEE article I was trying to dig up...
[04:38:24] <cradek> oh haha
[04:38:32] <cradek> IEEE can bite me
[04:38:47] <maddash> * maddash bites cradek for the IE^3.
[04:39:27] <cradek> any organization that keeps scholarly work secret (or tries to extract money from people for it) isn't on my good side
[04:39:28] <maddash> aeeesh. if only REALIZE handled splines...
[04:39:45] <jmk-wvm> ok, now all I have to do is figure out how to lurk in this channel from a network where port 6667 is blocked...
[04:39:48] <cradek> gcode doesn't even handle splines
[04:39:47] <maddash> well, keep in mind that the publishers need to eat...
[04:40:18] <SWPadnos> make a little port redirector (if you can't find one) that connects to your home IP and gets reflected here
[04:40:23] <maddash> I said "REALIZE," not "EMC" or "G-code."
[04:40:47] <SWPadnos> since REALIZE generates G-code, there's a logical connection there ...
[04:40:49] <cradek> ?
[04:41:03] <cradek> yeah, my reading comprehension is generally pretty good
[04:41:12] <maddash> rofl
[04:41:30] <cradek> is there a spline representation in dxf/autolisp? just add it if so
[04:41:50] <cradek> there is a well-known recursive algorithm for approximating a spline to a certain error with segments
[04:42:07] <maddash> yeah, but I still have to program in the messy code that approximates splines w/arcs.
[04:42:28] <SWPadnos> either that or add the NURBS G-code to the interpreter
[04:42:37] <SWPadnos> not that I'm suggesting that
[04:42:49] <jmk-wvm> I think I might do something really novel tonight
[04:42:54] <maddash> G6.2 is so cool.
[04:42:58] <jmk-wvm> go to bed by midnight
[04:43:01] <SWPadnos> hahahahahaha
[04:43:10] <SWPadnos> sorry
[04:43:48] <maddash> if only there was a bittorrent for academic journals
[04:43:56] <cradek> I admit it's a little tempting to add spline support to emc.
[04:44:16] <SWPadnos> hard to type in the codes though
[04:44:19] <cradek> jepler has done most of the math research
[04:44:24] <cradek> yeah it's useless for hand generated stuff
[04:44:26] <SWPadnos> describing an arc is hard enough
[04:44:29] <maddash> well, it's probably called "G6.2" and not "G<integer>" for a reason...
[04:44:43] <SWPadnos> no - there are lots of Gxx.x or Mxx.x codes
[04:45:23] <maddash> and even if emc supported G6.2, it's the hardware support for it that makes the final call
[04:45:41] <cradek> what do you mean hardware?
[04:46:02] <jmkasunich> we don't need no steenkin' hardware
[04:46:10] <SWPadnos> muahahahahahah!
[04:46:15] <cradek> says jmk who's not in bed yet
[04:46:26] <maddash> wrong jmk...?
[04:46:28] <SWPadnos> he's got 13-delta minutes left
[04:46:33] <jmkasunich> hey, jmk-wvm is fast asleep
[04:46:38] <SWPadnos> heh
[04:46:51] <maddash> "hardware" - the piece of machinery that drinks the G-code...
[04:47:16] <maddash> oh god. what am I talking about?
[04:47:17] <SWPadnos> that's the computer running emc2, around here
[04:47:32] <cradek> * cradek checks the channel name
[04:48:18] <maddash> ugh, elsevier's down
[04:48:24] <cradek> g'night all
[04:48:30] <maddash> say, what math work did jepler do?
[04:48:30] <jmkasunich> goodnight
[04:48:32] <SWPadnos> see you
[04:50:04] <SWPadnos> maddash, dunno. I know there was talk of more complex curves at some point maybe a year ago
[04:50:20] <SWPadnos> but that was in relation to the trajectory planner, not new input G-codes
[04:52:00] <maddash> what timezone are you in?
[04:52:08] <SWPadnos> EST5EDT
[04:52:15] <maddash> you sound unusually wide-awake
[04:52:24] <SWPadnos> I just ate Doritos
[04:52:42] <SWPadnos> I tend to be a night owl
[04:52:51] <jmkasunich> SWPadnos: stays up later than I do
[04:53:02] <maddash> EST? are you planning an all-nighter?
[04:53:14] <SWPadnos> nope. it's only midnight (almost) here
[04:54:00] <maddash> I'm in EST as well, and the only reason I can manage to type this out is b/c I'm in "bad-ass" mode.
[04:54:16] <SWPadnos> maybe the pot of coffee per day helps me
[04:54:40] <maddash> actually, tabasco sauce works better
[04:54:49] <SWPadnos> that gives me bad ass
[04:56:20] <maddash> well, gotta pass out. cya.
[04:56:26] <SWPadnos> have fun
[05:03:32] <jmkasunich> I wonder what an Xilinx XC73108 is?
[05:03:38] <SWPadnos> hmmm
[05:03:51] <SWPadnos> a CPLD
[05:04:18] <SWPadnos> the ISE doesn't support that, right?
[05:04:28] <jmkasunich> no clue
[05:04:28] <SWPadnos> err - webpack, that is
[05:04:35] <SWPadnos> ah. looking at a board?
[05:04:38] <jmkasunich> I just found 18 of them in a box of crap
[05:04:41] <SWPadnos> heh
[05:04:54] <jmkasunich> not that I'd ever be able to solder a 225 BGA
[05:04:59] <SWPadnos> http://www.mhl.tuc.gr/data_books/Xilinx/73108.pdf
[05:05:38] <jmkasunich> google found a datasheet
[05:05:45] <jmkasunich> I can't get excited about cplds tho
[05:06:48] <SWPadnos> they changed the process for that chip from 1.0 um to 0.85 um around 8/1995 ...
[05:06:54] <SWPadnos> it's an old CPLD to boot :)
[05:07:50] <jmkasunich> it was in a box of old stuff
[05:08:12] <SWPadnos> that's appropriate then
[05:08:22] <jmkasunich> I just found what appears to be a 24 channel opto-22 board
[05:08:33] <SWPadnos> cool!
[05:08:48] <jmkasunich> no optos on it of course
[05:09:05] <SWPadnos> I have several that are too big for the USC ...
[05:09:19] <SWPadnos> ODC5 and OAC5 type
[05:09:36] <jmkasunich> but it has 50 pin connector, twelve 4 position screw terminal blocks, 24 leds... pretty nice
[05:10:02] <SWPadnos> that may match the 5i20 connector ...
[05:10:17] <SWPadnos> they do mention relay rack compatibility
[05:10:35] <jmkasunich> I've got a mix of the modules, most of which don't fit the USC
[05:10:43] <jmkasunich> looks like at least some do fit this board tho
[05:10:46] <SWPadnos> heh - a common problem
[05:12:26] <jmkasunich> actually, _most_ of them fit this board
[05:12:46] <SWPadnos> I bet all but the input or multi-channel output modules fit
[05:13:28] <jmkasunich> I have several smaller boards (4 channel) and none of my modules fit them
[05:13:37] <jmkasunich> I thought I had bastard modules
[05:13:44] <jmkasunich> but I think they were bastard boards
[05:13:51] <jmkasunich> they take a longer, skinner module
[05:13:57] <SWPadnos> odd. is it pin spacing, number of pins, module size?
[05:13:58] <SWPadnos> ah
[05:14:18] <jmkasunich> yes
[05:14:21] <SWPadnos> heh
[05:14:45] <SWPadnos> there are two or three sizes (of the board mount ones)
[05:15:11] <jmkasunich> I have 8 OAC5, 3 ODC5, 18 IDC5, and 13 IAC5
[05:15:45] <SWPadnos> does the board have the sockets for the input modules?
[05:15:48] <SWPadnos> (extra pin)
[05:15:53] <jmkasunich> yes
[05:15:56] <SWPadnos> cool
[05:16:20] <jmkasunich> the one 4 channel board that I had that fit my modules only had 4 pins
[05:16:40] <jmkasunich> another reason I was kinda bummed about my collection
[05:17:01] <jmkasunich> I have 4 four channel boards with 5 pins, for the long skinny modules
[05:17:14] <jmkasunich> and only one long skinny moduke
[05:17:28] <SWPadnos> I have one USC for 8 Waldom D2W series input modules only ...
[05:17:55] <SWPadnos> Hmmm - did you give me one of the small boards at CNC workshop last year?
[05:18:06] <SWPadnos> I think you had a couple
[05:19:16] <jmkasunich> maybe
[05:19:32] <jmkasunich> since I didn't have anything that would plug into it...
[05:19:58] <SWPadnos> yeah. though I have a vague memory (possibly invented) of trying my modules, and having them not fit
[05:23:10] <jmkasunich> well, I'm off
[05:23:13] <jmkasunich> goodnight
[05:23:19] <SWPadnos> see you later
[05:32:19] <A-L-P-H-A> SWPadnos... what cha doing?
[05:32:34] <SWPadnos> thinking about how to test this lens I'm supposed to test
[05:32:43] <A-L-P-H-A> "lens"?
[05:32:46] <A-L-P-H-A> optics?
[05:32:52] <SWPadnos> yes, lens
[05:32:55] <SWPadnos> fora camera
[05:32:59] <A-L-P-H-A> what do you do for a living??
[05:33:06] <SWPadnos> electrical engineering / programming
[05:33:15] <A-L-P-H-A> oh
[05:33:53] <A-L-P-H-A> SWPadnos, make it do test shots of a given test image... dpreview.com has those test images... but I think you're supposed to buy those suckers for like $100 / sheet or something stupid like that.
[05:34:22] <SWPadnos> macbeth or SMPTE chart - that's not the kind of test I'm talking about
[05:35:30] <SWPadnos> I need to see how repeatable the electronic aperture control is on this lens, so I can tell if it's possible to use it in an application for which it wasn't really designed
[05:40:57] <A-L-P-H-A> oh...
[05:41:30] <A-L-P-H-A> I've got zero clue... I'm not an choochoo train driver.
[05:41:56] <SWPadnos> I see ... :)
[05:42:36] <A-L-P-H-A> I was talking to this girl... [yes another one]... trying to tell her about a zoom-crop feature of my gallery...
[05:43:01] <A-L-P-H-A> but she didn't know what aspect ratio was... but I didn't know how to really define it, as a ratio of width and height... she didn't really get it.
[05:43:01] <SWPadnos> did she slap you?
[05:43:16] <SWPadnos> once you say "ratio", you lose a lot of people
[05:43:23] <A-L-P-H-A> :(
[05:43:33] <A-L-P-H-A> people need to learn math...
[05:43:53] <A-L-P-H-A> I was going to start using widescreen version normal tv... but that probably wouldn't have helped either for her.
[05:44:28] <SWPadnos> fat people have a wider aspect ratio ...
[05:45:12] <A-L-P-H-A> hahaha
[05:45:22] <A-L-P-H-A> she's normal sized... with a big chest. :)
[05:45:28] <A-L-P-H-A> yumm.
[05:45:31] <SWPadnos> heh
[05:45:36] <A-L-P-H-A> I mean. nm...
[05:46:01] <SWPadnos> funny thing about aspect ratio (WRT TV)
[05:46:28] <SWPadnos> I was at the BBC for a day, and we were sitting around watching the 10 or so plasma screens they have in one lobby/break area
[05:46:48] <SWPadnos> half of them didn't have the aspect ratio set correctly for thesource, so peoples faces were squashed
[05:46:48] <A-L-P-H-A> were you visiting robin_sz?
[05:46:51] <SWPadnos> nope
[05:47:02] <A-L-P-H-A> oh... he worked for the bbc. :)
[05:47:08] <SWPadnos> same thing at the local First Night thing, but this time with a cable van showing you how great digital cable is :)
[05:47:11] <SWPadnos> yep - long ago
[05:47:24] <A-L-P-H-A> I want to see more topgear!
[05:47:31] <SWPadnos> heh
[05:47:45] <SWPadnos> I think the husband of a friend of mine directed a few of those
[05:47:51] <SWPadnos> err - ex-husband
[05:48:31] <A-L-P-H-A> damn it!
[05:48:36] <A-L-P-H-A> I was gonna say... get me some tickets... :D
[05:48:47] <SWPadnos> heh
[05:48:54] <SWPadnos> I guess they do some studio stuff
[05:48:55] <A-L-P-H-A> robin_sz, used to produce it.
[05:49:11] <SWPadnos> have you seen the episode where they try to destroy a Toyota HiLux pickup truck?
[05:49:15] <A-L-P-H-A> I'd go there, all decked out in Canuck gear. :)
[05:49:21] <A-L-P-H-A> yup
[05:49:23] <SWPadnos> hockey stuff?
[05:49:29] <SWPadnos> I'd love to buy a HiLux
[05:49:32] <A-L-P-H-A> they put it on top of an imploding building.
[05:49:35] <SWPadnos> yep
[05:49:39] <SWPadnos> among other things
[05:49:45] <A-L-P-H-A> yeah... but that was the last of it.
[05:49:48] <SWPadnos> (actually, that's the only episode I've watched)
[05:49:49] <A-L-P-H-A> did it start? I don't remember.
[05:49:59] <SWPadnos> it did start, they drove it into the studio afterwards
[05:49:59] <A-L-P-H-A> SWPadnos... sacralege.
[05:50:07] <SWPadnos> all flat and stuff
[05:50:09] <A-L-P-H-A> I thought they towed it ot the studio.
[05:50:19] <SWPadnos> nope
[05:50:23] <A-L-P-H-A> :|
[05:50:27] <A-L-P-H-A> I gotta rewatch that ep.
[05:50:40] <SWPadnos> they needed to replace some "soft stuff" I think - maybe a hose or two, plus add gas
[05:51:58] <A-L-P-H-A> SWPadnos, ifyou program, do you use a nice friendly programming font?
[05:52:21] <SWPadnos> nope. usually courier or something
[05:52:27] <SWPadnos> depends on the OS also
[05:52:35] <A-L-P-H-A> well... they all can use TTF.
[05:52:50] <A-L-P-H-A> http://www.proggyfonts.com/index.php?menu=download
[05:52:51] <SWPadnos> proportional spacing is detrimental to programming, for me
[05:53:03] <SWPadnos> ah - monospace.
[05:53:06] <A-L-P-H-A> of course!
[05:53:21] <SWPadnos> I use the one (or two) that comes with Windows, or courier
[05:53:44] <A-L-P-H-A> I use one of those... it's nice... with the bold stuff... makdes reading things on the page so much easier.
[06:07:29] <A-L-P-H-A> banana!
[06:39:03] <K`zan> http://wrlabs.shacknet.nu/~vw/MyMachineShop/Tools/uMillBeltDrive/
[06:43:57] <A-L-P-H-A> K`zan, and how much did that cost?
[06:46:49] <K`zan> $129 plus ?$7 for an extra belt.
[06:46:59] <A-L-P-H-A> you couldn't hav emade it yourself???
[06:47:25] <A-L-P-H-A> that's one gecko right there! http://geckodrive.com/product.cfm?pid=9
[06:47:43] <K`zan> Probably, but when I ordered it I wasn't real confident. Now I might tackle it.
[06:48:18] <K`zan> I thought the 201 wasn't recommended?!?
[06:48:30] <K`zan> Chow, bbiaf
[06:48:41] <A-L-P-H-A> it's perfectly fine
[06:57:48] <A-L-P-H-A> K`zan, your hero... http://www.tpmmuckraker.com/archives/002349.php :)
[07:05:47] <K`zan> A-L-P-H-A: feinstein? She has all the integrity of a syphlitic rabid rat. Heard nothing on this *anywhere* else...
[07:15:14] <A-L-P-H-A> K`zan, umm... here's like 97 other articles. http://news.google.com/news?ie=UTF-8&oe=UTF-8&rls=org.mozilla:en-US:official&client=firefox-a&tab=wn&ncl=1112765726&hl=en&filter=0
[07:21:39] <ds3> Hmmm
[08:41:10] <anonimasu> 5morning
[08:41:27] <anonimasu> K`zan: any gecko will work.
[08:45:01] <A-L-P-H-A> anonimasu... she's stuborn... just let her enjoy her xylotex drivers.
[08:45:18] <anonimasu> haha
[08:45:36] <anonimasu> yeah, with the great midband resonance compensation
[08:46:00] <A-L-P-H-A> :)
[08:46:09] <A-L-P-H-A> I dunno... never had something like that. :)
[08:46:11] <anonimasu> haha
[08:46:12] <anonimasu> xylotex?
[08:46:17] <A-L-P-H-A> I got geckos.
[08:46:21] <anonimasu> me neither :)
[08:47:36] <anonimasu> but it'd suck to have
[09:10:00] <alex_joni> *yawn*
[09:10:33] <anonimasu> agreed..
[09:10:35] <anonimasu> *yawns too*
[09:14:57] <alex_joni> it's nice to get to work after 10 :)
[09:15:12] <anonimasu> haha
[09:15:15] <alex_joni> until I need to get really busy that is :)
[09:15:16] <anonimasu> yes
[09:21:40] <A-L-P-H-A> http://www.omax.com/
[10:41:33] <alex_joni> has anyone of you used IP security cameras?
[10:42:56] <alex_joni> http://www.planet.com.tw/product/product_dm.php?product_id=20&menu_id=24
[10:43:00] <alex_joni> something like that
[11:16:39] <rafa> hello frineds
[11:17:03] <rafa> i make one program and i want to use it for a Mcod
[11:17:15] <anonimasu> mcod?
[11:17:17] <anonimasu> mcode?
[11:17:25] <rafa> mcode
[11:17:38] <rafa> i named my programa m150
[11:17:50] <anonimasu> ah, you just place it in a special directory
[11:18:03] <anonimasu> let me look at the wiki
[11:18:05] <rafa> and i copy it for directory /usr/share/emc?
[11:18:14] <anonimasu> a subdirectory for emc
[11:18:26] <rafa> watt?
[11:18:39] <rafa> please
[11:18:48] <anonimasu> I'm looking chill.
[11:19:22] <rafa> ok
[11:19:33] <anonimasu> are you running emc1?
[11:19:43] <rafa> emcl?
[11:19:58] <anonimasu> emc or emc2?
[11:20:03] <rafa> emc2
[11:20:16] <anonimasu> ok
[11:21:04] <rafa> and ..?
[11:21:27] <anonimasu> I'm looking..
[11:21:34] <anonimasu> I dont have a box with emc here right now
[11:21:38] <rafa> ok sorry
[11:22:16] <anonimasu> I'll find it in a sec..
[11:24:34] <anonimasu> it's not in the manual from what I can see
[11:27:37] <anonimasu> and I cant find the cvs :) either
[11:28:49] <anonimasu> ah now..
[11:28:49] <anonimasu> :9
[11:29:23] <alex_joni> you need to put it in the config dir
[11:29:28] <alex_joni> and make it executable
[11:29:33] <alex_joni> or was it nc_files dir?
[11:29:38] <alex_joni> one of those two
[11:29:44] <anonimasu> alex_joni: you are probably right :)
[11:29:53] <alex_joni> nc_files dir
[11:30:00] <alex_joni> make it executable and put it there
[11:30:04] <anonimasu> yeah it's nc_files
[11:30:12] <anonimasu> if you look there there's a M101 command
[11:30:17] <alex_joni> you can find the location of the nc_files dir by reading your ini file
[11:30:28] <alex_joni> FSCK.. found a worm :(
[11:32:24] <anonimasu> a:(
[11:33:07] <rafa> thanks for all
[11:33:11] <rafa> but ..
[11:33:25] <rafa> im my sistem i have one folder
[11:33:40] <rafa> the name is /usr/share/emc/ncfiles
[11:34:08] <rafa> in the directory have one source in C
[11:34:13] <rafa> M102.C
[11:34:26] <rafa> source says:
[11:35:12] <rafa> compile M102.C gccM102.c -o M102 and copy M102 executable from directory emc/program
[11:35:39] <rafa> where diterctory emc/programa?
[11:35:43] <rafa> where diterctory emc/program?
[11:37:33] <anonimasu> you dont have to do that
[11:37:39] <anonimasu> you are in teh right directory already
[11:37:41] <anonimasu> try M102
[11:37:48] <rafa> yes
[11:37:51] <rafa> i tri it
[11:37:55] <rafa> but
[11:38:10] <rafa> in /usr/share/emc/ncfiles?
[11:38:55] <alex_joni> I think so
[11:39:02] <alex_joni> you probably need sudo to copy there
[11:39:27] <anonimasu> yeah
[11:39:53] <rafa> but i use sudo
[11:40:01] <rafa> for it
[11:40:01] <anonimasu> chmod +x
[11:40:03] <anonimasu> chmod +x M102
[11:40:11] <anonimasu> alex_joni: or dosent it care?
[11:40:30] <rafa> please one moment
[11:41:07] <alex_joni> anonimasu: chmod +x is needed
[11:41:19] <rafa> mi executable must be where?
[11:41:32] <alex_joni> in /usr/share/emc/ncfiles
[11:41:46] <alex_joni> named M1xx (M101, M102, M103, etc)
[11:41:49] <alex_joni> and executable
[11:42:17] <alex_joni> m150 is not correct
[11:42:20] <alex_joni> M150 is correct
[11:43:08] <rafa> ok
[11:43:23] <rafa> i am in /usr/share/emc/ncfiles
[11:43:25] <rafa> ok?
[11:43:31] <alex_joni> yes
[11:43:42] <rafa> and i compile M102.c
[11:43:51] <alex_joni> ok..
[11:44:03] <rafa> and make chmod +x M102
[11:44:06] <rafa> but
[11:44:09] <rafa> in axis
[11:44:16] <rafa> returns:
[11:44:17] <alex_joni> you need to restart emc2
[11:44:31] <rafa> how?
[11:44:38] <alex_joni> close AXIS, start emc2 again
[11:44:50] <rafa> i make it
[11:45:13] <rafa> did not give certain
[11:46:17] <rafa> im AXIS returns "Unknow m code used"
[11:48:09] <alex_joni> start emc from a terminal
[11:48:24] <alex_joni> you should see the message "added custom m-code"
[11:48:40] <alex_joni> make sure your ini file has NC_FILES = /usr/share/emc/ncfiles
[11:51:19] <rafa> ok i tri it
[11:51:34] <alex_joni> sorry.. I mean PROGRAM_PREFIX = /usr/share/emc/ncfiles
[11:51:36] <rafa> NC_FILES not in ini file
[11:51:45] <rafa> i added it?
[11:51:51] <alex_joni> no..
[11:51:58] <alex_joni> it is PROGRAM_PREFIX
[11:52:05] <rafa> ok
[11:52:12] <rafa> it yes
[11:54:22] <rafa> ok
[11:54:28] <rafa> its ok
[11:56:57] <anonimasu> ^_^
[11:59:01] <rafa> thanks for all
[12:03:26] <slundell> Anyone knows where to find example STL files? I just want to try out some programs, but do not have any stl files here.
[12:08:34] <anonimasu> google..
[12:12:44] <anonimasu> I'd get you one but im busy :/
[12:17:42] <danex> Hello Alex_joni
[12:18:12] <alex_joni> hi danex
[12:18:24] <danex> How have you bee?
[12:18:29] <danex> been?
[12:18:34] <alex_joni> better than 2 hours ago :)
[12:18:50] <alex_joni> finished cleaning up a virus
[12:18:53] <danex> What happened two hours ago?
[12:19:01] <danex> Oh
[12:19:25] <danex> On a windows Box?
[12:19:31] <alex_joni> on a windows network :D
[12:19:53] <danex> That is a bad day!
[12:20:04] <alex_joni> seems to be better now :)
[12:20:15] <danex> Great
[12:20:50] <danex> I have been making the doctors rich for the past couple months
[12:21:19] <danex> Have not had much time to work with EMC
[12:23:33] <danex> Has anyone run emc on Ubuntu 6.10?
[12:27:00] <danex> Time to go to work, Hope you have no more virus to deal with for a while
[12:27:18] <alex_joni> danex: no, we decided to stick with 6.06
[12:27:29] <alex_joni> it's a LTS (long term support), and works just great
[12:27:41] <danex> good
[12:28:06] <alex_joni> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#Why_aren_t_there_packages_for_Ubuntu_6_10_Edgy_Eft
[12:29:07] <danex> Has Halui been adopted in the latest release?
[12:29:23] <alex_joni> it will be there
[12:29:24] <alex_joni> :)
[12:29:29] <alex_joni> 2.1.0 is in prerelease state
[12:29:30] <alex_joni> alpha0 :D
[12:29:55] <alex_joni> closer to beta actually.. only minor things to fix I think
[12:30:30] <danex> I have to go now, do not want to be late on the first day back :(
[12:30:41] <danex> Have a Great Day
[12:31:17] <alex_joni> thx.. same to you
[12:32:13] <tomp> pyvcp knobs pix http://imagebin.org/6945
[12:32:16] <tomp> pyvcp_widgets.py http://pastebin.ca/319271
[12:32:32] <tomp> text xml for knobs http://pastebin.ca/319274
[12:33:17] <tomp> gota run, bb2nite
[12:34:01] <tomp> logging
[12:39:42] <awallin> tomp: nice work, I'll try the new dial widget when I have time.
[12:49:54] <rafa> hello frineds
[12:50:38] <rafa> i make my progrma
[12:50:45] <rafa> and in axis ok
[12:51:13] <rafa> the idea of my programa its on one motor, for exemple
[12:51:31] <rafa> my program named M105
[12:51:33] <rafa> and
[12:51:53] <rafa> where in directory /usr/share/emc/ncfiles
[12:52:11] <rafa> the source of my programs is:
[12:52:40] <rafa> #include<stdio.h>
[12:53:20] <rafa> #define DATA_END 0x378
[12:53:28] <rafa> int main()
[12:53:30] <rafa> {
[12:53:40] <alex_joni> you cannot write to the parport from a program while emc is running
[12:54:40] <alex_joni> do you want to turn on a pin on the parport?
[12:55:24] <alex_joni> you can do that by using M64 P1, and linking motion.digital-out-01 to parport.0.pin-15-out
[12:55:33] <alex_joni> with M65P1 you can turn the pin off
[12:55:56] <alex_joni> the same works for P1,P2,P3,P4 (motion.digital-out-01, ...-02, ...-03, ...-04)
[13:00:14] <anonimasu> alex_joni: can you delay your program?
[13:00:35] <anonimasu> err does emc wait for the M code to return before continuing?
[13:01:01] <alex_joni> anonimasu: no it doesn't
[13:01:10] <alex_joni> but you can use G4 dwell
[13:01:16] <anonimasu> hm..
[13:01:19] <anonimasu> that sucks..:D
[13:01:32] <anonimasu> alex_joni: can you call python programs as M commands?
[13:03:26] <alex_joni> anonimasu: anything executable
[13:05:16] <alex_joni> even fortran programs if you like
[13:06:31] <rafa> hello frineds
[13:06:45] <rafa> come bak
[13:07:08] <anonimasu> alex_joni: neat, then you can pause until your whatever operation's done
[13:07:36] <alex_joni> anonimasu: you can use the emc module in your program to generate pause/resume messages to emc
[13:07:55] <rafa> how to make a M code for turn on one LED or motor, if no write parport in emc runing?
[13:08:11] <anonimasu> rafa: use python, and connect it to a hal pin
[13:08:11] <alex_joni> rafa: I explained before you left
[13:08:11] <anonimasu> )
[13:08:13] <anonimasu> :)
[13:08:19] <alex_joni> anonimasu: don't make it hard for him
[13:08:20] <anonimasu> or use C and connect to a hal pin
[13:08:20] <anonimasu> ;)
[13:08:29] <rafa> ?
[13:08:34] <rafa> left?
[13:08:36] <anonimasu> logger_emc: bookmark
[13:08:37] <anonimasu> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-01-17.txt
[13:08:53] <anonimasu> 13:56 < alex_joni> you can do that by using M64 P1, and linking motion.digital-out-01 to parport.0.pin-15-out
[13:08:53] <alex_joni> rafa: use G64 P1 to turn motion.digital-out-01 on, and G65 P1 to turn motion.digital-out-01 off
[13:08:56] <anonimasu> 13:56 < alex_joni> with M65P1 you can turn the pin off
[13:08:59] <anonimasu> 13:56 < alex_joni> the same works for P1,P2,P3,P4 (motion.digital-out-01, ...-02, ...-03, ...-04)
[13:09:00] <alex_joni> err.. M64 :)
[13:09:40] <rafa> i dont under stand
[13:10:44] <rafa> write thiis in standatr pinoult.hal?
[13:10:51] <alex_joni> do you know what G-code is?
[13:10:59] <alex_joni> M64P1 is a g-code command.. ok?
[13:11:06] <rafa> ok
[13:11:22] <alex_joni> when you write that, emc2 turns the pin 'motion.digital-out-01' TRUE
[13:11:46] <rafa> in standatr pinoult.hal?
[13:12:01] <rafa> or no?
[13:13:51] <rafa> ?
[13:14:19] <rafa> write this ins standart_pinout.hal?
[13:14:23] <rafa> please
[13:17:56] <rafa> joni:
[13:35:46] <paragon36> Hello All .... What method could I use to multiply step pulses from the paraport into a l297 / l298 driver?
[13:47:35] <jepler> paragon36: good question. I have looked around a bit and not found a discrete solution or an open source microcontroller firmware to do this. This stepper driver board has a step multipler, which I believe I was told is a PIC microcontroller: http://www.pmdx.com/PMDX-150/index.html
[13:47:48] <jepler> paragon36: (see page 9)
[13:49:56] <jepler> "The step multiplier operates like a phase-locked loop (PLL). It predicts the interval between the most recent step input and the next step input. It then outputs four equally spaced step pulses during that time interval."
[13:54:55] <paragon36> Thanks Jepler Ill take a look!
[13:59:07] <skunkworks> logger_emc: bookmark
[13:59:07] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-01-17.txt
[14:04:20] <paragon36> When using half steps on an L297 does one step pulse from emc equate to 2 half step pulses on the output of the L297 or does one emc step = one half step?
[14:04:47] <jepler> paragon36: the latter
[14:06:26] <paragon36> one for one .... so one rev on a 200 full step motor = 400 EMC steps per motor rev
[14:06:34] <jepler> right
[14:07:06] <jepler> so if you've got a 1mm/rev leadscrew, that's an INPUT_SCALE of 400.
[14:07:24] <jepler> (assuming direct drive)
[14:07:24] <paragon36> Which means more overhead on EMC2! How does one equate that to the base_period
[14:07:26] <paragon36> ?
[14:07:46] <paragon36> ? = appending to my question not your answer ;-)
[14:07:48] <jepler> how many mm/sec do you want to move?
[14:08:27] <paragon36> Not sure as I would have to check my original .ini file which I do not have access to at the moment.
[14:09:00] <jepler> ballpark?
[14:09:07] <jepler> 25mm/s? 100mm/s?
[14:10:08] <paragon36> This is on an old p200 pro which was running EMC1 ...! No idea Jepler it was a while ago since I set it up (3 Years plus) The machine is a Denford Starmill
[14:10:16] <jepler> once you find your maximum desired step rate in steps per second, multiply by 2 to find the absolute minimum number of BASE_PERIODS per second.
[14:11:14] <jepler> 25 mm/s * 400 steps per mm = 10000 steps per second. 1/20000 second = 50000ns, the default BASE_PERIOD
[14:12:23] <jepler> almost all machines will do a BASE_PERIOD of 20000, which would get you to 62.5mm/second
[14:12:55] <paragon36> One thing I found with EMC1 and I probably mentioned it before was that the stepper ran rough and if the speed was too fast missed steps. But when running via the denford DOS program and Denfords internal controller the steppers run silk smooth and a lot faster without missing steps what could cause this as I never got to the bottom of it?
[14:13:43] <paragon36> EMC1 also caused a lot more vibration!
[14:14:21] <jepler> it's hard to say
[14:14:46] <jepler> bbl, coffee time here at the office
[14:15:35] <paragon36> One of the things I am going to try is to stick an ocilloscope on the inputs to the L297 / L298 denford driver boards and compare the pulse streams from EMC2 and Denford Controller board.
[14:17:25] <paragon36> Once suspect is that I am running the L297 with full steps via EMC while the denford controller maybe running with half steps.
[14:17:34] <paragon36> s/Once/one/
[14:22:03] <anonimasu> paragon36: that would explain ti
[14:44:18] <jepler> interesting -- I just noticed in the emc1 source that there was some kind of option to toggle the speaker port -- the comments indicate this worked as a watchdog ?
[14:44:33] <jepler> if (wdEnabled && wdCount-- <= 0) {
[14:44:37] <jepler> outb(soundByte, SOUND_PORT);
[14:44:36] <jepler> wdCount = wdWait;
[14:44:52] <cradek> strange
[14:45:37] <cradek> I think I remember hearing people talk about something to do with a soundcard
[14:45:51] <cradek> are you sure that's the pc speaker?
[14:45:59] <jepler> no, not sure
[14:46:33] <jepler> #define SOUND_PORT 0x61, the same port I used in hal_speaker
[14:48:11] <cradek> paragon36: halfstep/fullstep is selected by jumper on the L297 so I don't think that's it
[14:49:08] <cradek> paragon36: were you using steppermod or freqmod with emc1?
[14:49:18] <jepler> cradek: am I right in thinking that emc2's stepping output is a lot better than emc1 was?
[14:50:01] <jepler> I'm looking at the emc1 source code (emcmot.c), and I think that the number of BASE_PERIODS per step was constant through a whole servo cycle, which is not true in emc2.
[14:50:01] <cradek> I think it's slightly better than freqmod, and MUCH better than steppermod
[14:50:21] <paragon36> Sorry Guys! Was distracted with work ... dred the thought .... ;-) From memory it was steppermod.
[14:50:44] <cradek> paragon36: ok I think that explains it. steppermod was very primitive.
[14:51:25] <paragon36> What is it called in emc2 freqmod ?
[14:51:39] <jepler> the emc2 step generator you want is 'stepgen'
[14:51:51] <cradek> paragon36: the stepper driver is called stepgen - it's a new approach that's not really like the emc1 way
[14:52:40] <cradek> (emc1 pretended the steppers were servos)
[14:52:52] <paragon36> Cool hopefully this will run better. I'll try it out. Thanks for the info.
[14:53:24] <cradek> how long ago did you try emc1?
[14:53:59] <paragon36> On another note have you found a nice way to interface with the pluto-p board ie nice connectors? (emc1 = two years agao maybe more)
[14:54:28] <cradek> not really. just the usual IDP connectors on ribbon cable.
[14:56:05] <anonimasu> paragon36: you can count on emc2 being better :
[14:56:07] <anonimasu> :)
[14:57:43] <paragon36> IDP is that the same as the connectors used for IDE drives?
[14:57:52] <paragon36> Cheers Anon :-)
[15:01:03] <skunkworks> I found cables ends for all but the 12 pin connector on the right side of the pluto. THen just made my ribbon cables. (I made a 12 pin plug by grinding down a 14 pin one ;)
[15:02:03] <skunkworks> crappy picture
[15:02:04] <skunkworks> http://www.electronicsam.com/images/KandT/servostart/pluto.jpg
[15:03:35] <paragon36> Thanks skunkworks!
[15:05:15] <skunkworks> but I have a very large selection of old motherboard ribbon cables for things like serial/game/printer ports.
[15:07:30] <jepler> who doesn't? :-P
[15:13:55] <skunkworks> :)
[15:18:44] <skunkworks> alex_joni: Are you around?
[15:19:55] <alex_joni> skunkworks: yeah, but very busy :(
[15:20:02] <skunkworks> ah - those are debs - not iso.. Duh
[15:20:30] <skunkworks> thats ok - just wondering about the alpha 2.1.0
[15:22:17] <rafa> friend alex_joni
[15:22:53] <rafa> please teel-me how a make one M Code for turn on one Led or motor..
[15:23:02] <jepler> skunkworks: what are you wondering about it?
[15:23:29] <rafa> see for i not write parport in emc is runing
[15:24:39] <skunkworks> jepler: How I would test it.
[15:24:53] <rafa> ?
[15:25:03] <alex_joni> rafa: use M64P1 in the code to turn it on
[15:25:07] <alex_joni> M65P1 to turn it off
[15:25:15] <alex_joni> then in you standard_pinout.hal put this:
[15:25:20] <alex_joni> newsig led bit
[15:25:47] <alex_joni> linksp led motion.digital-out-01
[15:25:56] <alex_joni> linksp led parport.0.pin-15-out
[15:26:47] <jepler> skunkworks: download the .deb files, and at the terminal run 'dpkg -i whatever.deb'. If it complains that some package is not available, use apt-get to install it (iirc there are a few new dependencies in 2.1).
[15:27:39] <alex_joni> skunkworks: do you know where the packages are?
[15:28:22] <jepler> alex_joni: what changes are in the rtai-modules deb?
[15:32:02] <skunkworks> http://dsplabs.cs.utt.ro/~juve/tempdebs/
[15:33:26] <skunkworks> Ok - I will give it a try - thaniks
[15:33:27] <skunkworks> thanks
[15:34:15] <alex_joni> jepler: the SMI change I made
[15:34:25] <jepler> alex_joni: ah
[15:34:35] <paragon36> Just been searching for connectors to use on the pluto-p here in the UK but all I can find are IDC and IDP what is the difference between them please?
[15:34:35] <alex_joni> basicly a patch to the rtai_smi module to allow specifying what should be enabled at insmod time
[15:35:04] <paragon36> s/IDC and IDP/IDC and NOT IDP/
[15:35:23] <cradek> I think that means the same thing
[15:35:54] <jepler> I'm more familiar with IDC, for Insulation Displacement Connector.
[15:36:06] <paragon36> It's the same as the IDE connector within a PC or it appears to be!
[15:36:37] <eholmgren> http://www.linuxdevices.com/news/NS2805234558.html
[15:36:54] <jepler> paragon36: yes, the IDE cable has the same spacing as the pins on the pluto
[15:36:56] <eholmgren> apparently the skycams do use rtlinux
[15:37:09] <paragon36> Thanks Jepler!
[15:37:20] <fenn> paragon36: it's IDC, not IDP
[15:37:37] <paragon36> Thanks fenn.
[15:37:54] <cradek> sorry I used the wrong letters
[15:38:02] <fenn> IDC == insulation displacement connector, because it stabs the wires
[15:38:28] <fenn> weird that there is no open source microcontroller step multiplier
[15:38:29] <eholmgren> same sort of setup as the emc controlled tripod ... except with 20HP motors
[15:38:49] <fenn> that should be pretty easy to code up
[15:39:05] <paragon36> No problem Cradek your've been a great help :-)
[15:39:19] <skunkworks> I have had good luck disasembling idc's and clamping new ribon back in.
[15:39:58] <fenn> you break the snap tabs on the sides sometimes though
[15:40:00] <jepler> I invariably break them if I try something like that
[15:40:09] <jepler> skunkworks: what for, to make the cable longer or shorter?
[15:40:15] <fenn> with the blue ones.. they have glass in them or something
[15:41:44] <skunkworks> jepler: I made double ended cables for the pluto so I can plug one end of the ribbon into the pluto and the other into the breakout board. (first time I have had idc appart. pretty cool)
[15:43:56] <skunkworks> I don't have any cables that have the same plug on each end other than ide/scsi. The smaller pin idc's have serial/game/whatever on the other end.
[15:45:59] <fenn> or you can just buy the connectors from <electronics supplier of choice>
[15:46:39] <fenn> looks like the pluto board is just a little too small; do the connectors interfere with each other?
[15:47:08] <paragon36> Are the IDE conectors the same pitch as the pluto-p ie will they fit if shortend to 26 pin?
[15:47:36] <fenn> yes
[15:47:55] <paragon36> Cool IDE = 40 pin right?
[15:48:04] <skunkworks> I could also do a 10 - 26 - 12 idc to-> 50 scsi end - then use one of the breakout board they found for the mesa project.
[15:48:06] <fenn> i think
[15:48:24] <skunkworks> boards
[15:48:57] <fenn> paragon36: newer ide cables have extra signal ground lines, so there are like 80 wires
[15:49:08] <fenn> same connector though
[15:50:06] <paragon36> I have plenty of ide cables but would need to get the pcb mounted male ide conectors to solder to the pluto-p .... looking at rs electronics as we speak.
[15:51:06] <cradek> oh I forgot it didn't come with those
[15:51:24] <cradek> (I keep that stuff around too)
[15:52:04] <paragon36> The board just comes with holes for the connectors
[15:52:06] <cradek> get a few long strips of the single row and the double row - then you can snap them to the lengths you need
[15:53:41] <paragon36> Just remebered I have some ic mounts that I could snap in half and to lenth that should work and a lot cheaper than the pcb icd's (£6.00 here in the UK)
[15:54:04] <jepler> not sure if that will work -- don't you still end up with a receptacle?
[15:55:20] <paragon36> Doh ... your right.. The legs may be long enough to come from the underside but what a pain!
[15:55:44] <fenn> RS is a ripoff
[15:56:08] <fenn> shouldn't be more than 20p
[15:56:10] <paragon36> Sure is ... Farnell not much better with there min 20UKP order.
[15:56:48] <paragon36> Not many indepenant suppliers these days :-(
[15:57:27] <fenn> well, order 10 of each thing in the catalog from futurlec.com :)
[15:58:06] <fenn> oh, they might not even ship to uk because of rohs
[15:58:28] <fenn> no, they have a london office
[15:59:33] <skunkworks> I take it digikey doesn't ship to the uk?
[16:00:35] <fenn> digikey can be frustrating when all you want is a 26 pin header
[16:03:16] <skunkworks> * skunkworks ordered his headers for digikey. using the part number from SWPLinux.
[16:03:59] <alex_joni> skunkworks: managed with those packages?
[16:04:57] <alex_joni> you can easyly set up a local repo, then you automatically get all updates
[16:05:09] <skunkworks> http://catalog.digikey.com/scripts/partsearch.dll?Detail?name=S2011E-36-ND
[16:05:14] <alex_joni> 1. download the debs put them in /home/skunks/repo/binary
[16:05:31] <alex_joni> 2. go to /home/skunks/repo && issue: dpkg-scanpackages binary /dev/null | gzip -9c > binary/Packages.gz
[16:05:58] <alex_joni> 3. add 'deb file:/home/skunks/repo binary/' to your /etc/apt/sources.list
[16:06:07] <alex_joni> 4. sudo apt-get update && upgrade
[16:06:30] <alex_joni> maybe I should write a wiki with that info :)
[16:06:33] <skunkworks> alex_joni: I could have figured that all out my self ;)
[16:06:37] <skunkworks> Thanks
[16:08:00] <alex_joni> I'm sure you would have
[16:13:48] <owad> Has anybody tried running emc2 on an intel Mac yet? I just attempted to boot my Mac mini off the live cd, and it looks like it's frozen at trying to load the desktop.
[16:14:39] <cradek> owad: not that I've heard of
[16:15:05] <paragon36> Fen ... fururlec offer domestic postage rates :-) Im trying to find out what methods of payments they accept... ie CC debit, debit card, Paypal would be nice ;-)
[16:17:32] <alex_joni> owad: there was someone on the mailing lists trying to do that, but I don't think he succeeded
[16:18:17] <owad> It did seem to run ok in Parallels, but it was a bit unresponsive.
[16:18:49] <owad> heh - the screensaver just kicked on
[16:19:06] <owad> but it still won't load the desktop
[16:22:13] <skunkworks> owad: stupid question but - how much memory is in it?
[16:22:30] <owad> 512 MB
[16:28:43] <jepler> hi ray!
[16:29:06] <alex_joni> bbl
[16:35:48] <rayh> hi guys
[16:35:52] <cradek> hi
[16:36:23] <rayh> darn phone brb
[16:48:00] <rayh> back -- a customer running bdi found a problem with offsets on the 4th
[16:48:15] <SWPadnos> is it fixed yet?
[16:48:22] <SWPadnos> :)
[16:48:27] <cradek> in emc2?
[16:48:51] <rayh> What I was wondering was if anyone had visited that code for emc2
[16:49:23] <cradek> what's the problem and in what version of emc?
[16:49:54] <rayh> The problem showed up when an angular offset had been set in g55 and the interpreter encountered a g2/g3 with ijk locus.
[16:51:25] <cradek> I think that's fixed in emc2 and emc1.2 both
[16:51:32] <cradek> is he running a really old emc1?
[16:51:51] <rayh> Seems like BDI 4.20 or so.
[16:52:23] <fenn> paragon36: make sure you tell them to ship it anyway even if something is out of stock
[16:52:31] <fenn> paragon36: otherwise it may take like 3 months to get your package
[16:52:38] <cradek> what happens? does the rotary axis turn during the arc, even though there is no A word in the arc?
[16:53:03] <rayh> Right.
[16:53:48] <cradek> yeah I remember this problem and I'm pretty sure it's fixed in emc2 and emc1.2 both
[16:54:07] <cradek> I don't know exactly what he would be using with bdi4.20
[16:54:24] <rayh> Nor do I remember what age it was.
[16:54:39] <rayh> I'll run a few tests here and see what shakes out.
[16:54:41] <rayh> Thanks.
[16:55:06] <cradek> is he using tool length comp?
[16:56:03] <cradek> wait that was a separate problem (arcs in non-XY planes)
[16:56:10] <cradek> he probably has that problem too :-)
[16:57:21] <cradek> your bug was fixed by me in emc2 in Dec '05
[16:57:28] <cradek> http://cvs.linuxcnc.org/cvs/emc2/src/emc/task/emccanon.cc.diff?r1=1.23;r2=1.24
[16:58:37] <cradek> I think someone (me or paul) put that in emc1 around that time too. I would think it made it into bdi4emc also, but we can't say for sure
[16:59:32] <rayh> Okay. Eventually we will have to upgrade his machine but as long as it's fixed I'm good to go here.
[17:00:26] <cradek> yep I sure think it is
[17:00:38] <rayh> fantastic
[17:02:20] <cradek> if he uses length comp on that old version, also beware the G18 and G19 arcs, they are dangerous
[17:03:22] <rayh> Okay.
[17:04:24] <rayh> I'm not seeing any unexpected movement in A while I've got an offset working. Looks good here.
[17:04:39] <cradek> great
[17:05:13] <rayh> That test was using head from a couple days ago.
[17:05:59] <cradek> this bug was fixed in emc1 (the cradek_stable branch) in Aug '05, but I don't know if paul based bdi4emc on that branch or not
[17:06:21] <cradek> so before aug '05 all EMC versions had this bug
[17:06:36] <SWPadnos> that was around the time of the "great EMC schism"
[17:07:07] <cradek> yes it's definitely hard to say what code this user is running :-)
[17:07:41] <cradek> and it's fairly impossible to get him a simple patch for this bug
[17:20:30] <rayh> catch you all later.
[17:47:24] <tomp> awallin: could we connect widgets in the xml for pyvcp? ( do linksp linkps ? )
[17:48:15] <tomp> then we'd have have metawidgets (scriptable combos)
[18:03:14] <tomp> bbl
[18:29:05] <awallin> tomp: that's a possibility, but since using a pyvcp panel usually requires running a separate .hal file anyway, I don't see why hal commands should go into the XML
[18:47:10] <anonimasu> hi
[19:51:24] <alex_joni> hi all
[19:51:30] <SWPadnos> hi Alex
[19:51:54] <SWPadnos> I finally dropped your box off at FedEx. it'll probably arrive on Saturday or Monday
[19:52:07] <SWPadnos> (in CT, that is - not .ro :) )
[19:53:18] <alex_joni> yeah, I got the mail.. thanks
[20:01:15] <alex_joni> * alex_joni is burning a DVD at 16x
[20:01:48] <SWPadnos> whoah - slow down there
[20:01:50] <SWPadnos> the roads are icy
[20:01:56] <alex_joni> 22160 KB/s
[20:02:07] <alex_joni> SWPadnos: just testing the road here :)
[20:02:52] <alex_joni> the burner can do 18x fwiw
[20:03:54] <SWPadnos> what model?
[20:04:09] <alex_joni> LG super multi GSA-H22L
[20:04:35] <SWPadnos> ok. I haven't tried pushing my Plextor much (it's a 16x)
[20:04:50] <alex_joni> L stands for lightscribe.. still gotta test that :)
[20:05:10] <alex_joni> need some blanks for that
[20:07:28] <SWPadnos> you need Costco
[20:07:50] <SWPadnos> they had a deal 6 months ago orso - buy one stack of DVD+R (or -R) and get a second free
[20:08:09] <awallin> anyone sitting on a university campus with good access to e-journals? (better than I have at univ. helsinki that is...)
[20:08:19] <SWPadnos> they're stacks of 100 for $40 or so, TDK brand
[20:08:30] <SWPadnos> (but 2 stacks of 100 for $40 with the coupon :) )
[20:08:44] <alex_joni> SWPadnos: for lightscribe?
[20:09:17] <SWPadnos> actually, I think these won't work with LightScribe. they have a white printable top surface
[20:11:47] <awallin> guess not...
[20:19:20] <alex_joni> those work with printers
[20:19:35] <alex_joni> actually .. a small robot for multiplying disks with emc would be nice
[20:19:54] <alex_joni> one rotary axis, one vertical axis, one vacuum gripper
[20:20:12] <alex_joni> 3-4 spindles in a circle, one/two writers, one printer
[20:21:25] <paragon36> Hello All ... As discussed earlier the Denford Starmill runs a lot smother under EMC2 then it did under EMC1!
[20:21:37] <alex_joni> paragon36: nice to hear that
[20:21:53] <alex_joni> paragon36: it would be even better if you could post some pictures :)
[20:22:09] <paragon36> I will do Alex...
[20:22:18] <alex_joni> paragon36: great :)
[20:23:17] <paragon36> However in my haste while trying to debug the spindle drive a touche + with the GND of my scope and a big flash came out of the controller :-(
[20:24:15] <alex_joni> oops :(
[20:24:41] <paragon36> The spindle will not turn now! So I guess I am looking for a better way to drive a 50V + DC motor controlled via EMC
[20:24:51] <awalli1> -
[20:27:21] <alex_joni> +
[20:27:37] <jepler> paragon36: OUCH!
[20:28:25] <paragon36> The motor board appeared to take 0-12v the higher the voltage the faster the motor turned. I am hoping that the motor board is OK and it was just the denford controller which was made back in 1989 and I am planning to discard in favour of EMC
[20:28:57] <SWPadnos> paragon36, it's pretty common for speed controllers to take a 0-10V DC control input
[20:29:43] <jepler> paragon36: pwmgen to digital IO pin, RC filter to give 0-~5V, amplifier in non-inverting configuration to convert to 0-10V?
[20:29:55] <SWPadnos> if you have a battery around (like a 1.5V AA, C, or D), you can connectt hat to the control input and see if it turns the motor
[20:29:56] <paragon36> How does this get converted from digital input say from EMC2 to a 0 - 12v output?
[20:30:18] <SWPadnos> you need external hardware to do the voltage conversion and filtering
[20:30:47] <alex_joni> jepler: thought non-inverting amps are only subunitary in gain?
[20:30:51] <paragon36> Battery may not work due to its floating gnd (not common) ... I think
[20:31:04] <SWPadnos> no - that's a feature :)
[20:31:15] <alex_joni> guess you can put 2 inverting ones, but you still need negative supply voltage
[20:31:21] <SWPadnos> you connect the battery between the control ground and the speed input
[20:31:38] <jepler> alex_joni: not according to http://www.play-hookey.com/analog/non-inverting_amplifier.html
[20:32:06] <SWPadnos> yep - it depends on how you connect the op-amp </mr. obvious>
[20:32:43] <paragon36> When I was controlling the motor board directly I used a pot connected to the 3 input pins but I can't remember the pinout..
[20:33:25] <jepler> paragon36: you can create an "analog" signal with a simple RC filter and a digital input signal. The analog output voltage depends on what fraction of the time the input digital signal is high
[20:33:30] <SWPadnos> that'll work too: the speed controller probably has a supply output for manual control. connect the two ends of the pot to V+ and ground, and the wiper to the speed input
[20:33:29] <alex_joni> jepler: might have remembered something wrong then :(
[20:34:32] <jepler> paragon36: if the digital output is low all the time, the capacitor discharges and you get a 0V out. If it's high all the time, the capacitor charges and you get 5V out. If it's HIGH for 100uS and low for 100uS, then you get about 2.5V on the capacitor
[20:37:22] <jepler> (but you can't necessarily just connect your computer's GND to one end of this motor speed control pot)
[20:37:28] <paragon36> Jepler I think I understand you. but in this case I think I need to imatate a pot but digitaly or is what you are sugesting a way to do this? Sorry for ignorance regarding this matter!
[20:38:02] <SWPadnos> paragon36, the first thing you should do is make sure that your speed controller still works
[20:38:22] <SWPadnos> after that, you should start trying some of the methods jepler is pointing out
[20:38:32] <paragon36> Yeah agreed Ill dig out the pot that I used and connect it up!
[20:39:00] <jepler> paragon36: yeah, I'm talking about a way to generate an analog voltage from a digital output that is toggled on and off at a few kHz
[20:41:01] <paragon36> Thanks Guys.... Im going to test out the motor card and then go from there if all's well there then I can go forward with jeplers good idea..
[20:41:13] <paragon36> CU in a bit!
[20:42:32] <eholmgren> does an h-bridge disipate a significant amount of heat?
[20:43:09] <eholmgren> rectifying 20V @ ~5-10A maybe
[20:43:40] <alex_joni> eholmgren: depends on the Rdson of the transisters
[20:43:51] <alex_joni> if it's low Rds mosfets, then no
[20:44:01] <alex_joni> if it's BJT or Darlingtons then it might
[20:46:32] <eholmgren> wait ... I'm an idiot
[20:46:48] <eholmgren> a bridge rectifier ... not an h-bridge
[20:47:41] <SWPadnos> usually around 1V or so total drop, so at 10A that's 10 watts
[20:48:23] <alex_joni> 2 x 0.6V
[20:48:32] <alex_joni> 12W probably :)
[20:48:43] <SWPadnos> depends on the diodes ... :)
[20:48:48] <alex_joni> right
[20:48:50] <alex_joni> max 0.6V
[20:49:04] <cradek> paragon36: the little motor controller I worked on recently had a jumper for 0-10/12 or 0-5 input
[20:52:31] <cradek> so I set it for 0-5, discovered it had a suitable RC on the input already, and was isolated from the motor - so I hooked it right to the parport and used pwmgen to run it
[20:53:00] <jepler> yeah, don't forget about isolation. otherwise your next mistake might put 180V or worse through your PC.
[20:54:01] <cradek> somewhere I have a PCjr with a 7 bit parallel port because of a mistake like that
[20:54:25] <eholmgren> I've heard that magic smoke is the fountain of youth
[20:54:26] <jepler> hah
[20:54:44] <eholmgren> because you need to live at least long enough to fix the stuff you jsut blew up
[20:55:04] <cradek> and it was the cool sidecar parallel port (the jr didn't come with one) :-(
[20:59:44] <skunkworks> ooh - just got in a 2d barcoder reader with keyboard wedge... Fun
[21:00:18] <anonimasu> hm..
[21:00:26] <anonimasu> what are you going to do with it?
[21:00:46] <jepler> use them to get absolute feedback from a 2-axis mill
[21:00:47] <SWPadnos> read barcodes?
[21:00:54] <alex_joni> skunkworks: any luck with the 2.1.0 package?
[21:00:54] <jepler> by printing a special 1d code that is non-repeating
[21:01:12] <anonimasu> heh
[21:01:22] <skunkworks> alex_joni: I thought you would have noticed that I didn't download any yet ;)
[21:01:24] <anonimasu> jepler: well, can you really make them continous?
[21:01:45] <anonimasu> jepler: they convert the _code_ in hardware and spew it out the port as digits/stuff when done processing
[21:01:56] <alex_joni> heh: http://www.busboom.org/busboom/projects:cdrobot2
[21:01:58] <anonimasu> though you can set the repeat rate, but it's not instantaneous..
[21:02:04] <alex_joni> skunkworks: I don't monitor that site :)
[21:02:27] <jepler> anonimasu: yeah -- probably doesn't really work
[21:02:33] <jepler> but it was the first outlandish idea that popped into my head
[21:03:36] <anonimasu> jepler: sorry about breaking your idea :/
[21:04:57] <alex_joni> you can use it for homing though :)
[21:05:08] <anonimasu> yeah
[21:05:18] <alex_joni> works like a crappy microswitch :D
[21:06:17] <anonimasu> hehe
[21:06:18] <anonimasu> yeah
[21:09:10] <fenn> phydbleep made something sorta like what you are describing: http://fenn.dyndns.org/gingery_machines/index.php?encoders
[21:15:06] <danex> alex_joni: Is halui in the 2.0.5 package? or in the 2.1.0?
[21:16:24] <jepler> fenn: yeah that is something like what I was imagining -- only apparently it works
[21:17:38] <K`zan> Seattle PI: "...Snow days unfair to grown ups..." Oy...
[21:18:24] <alex_joni> danex: there is some halui in 2.0.x
[21:18:27] <alex_joni> but only limited
[21:18:37] <alex_joni> the rest is available in 2.1.0
[21:20:38] <danex> I am setting up EMC2 in my laptop for test purposes. I take it the method to install 2.10 is the same as for the "head"?
[21:21:14] <awalli1> laptops are usually bad at RT, so you might compile with --enable-simulator
[21:21:24] <awalli1> that way you don't have to use a RT kernel
[21:21:46] <fenn> and you can keep ACPI and stuff that's useful when you're on a laptop
[21:23:16] <paragon36> Looks like the motor board is hot! It has 240v ac suplied to it but not seeing 10-12v out to be honest not sure even if I should be ;-) What alternative do I have for driving 50-70v DC motor (again not sure from memory)
[21:23:27] <danex> What would be a good MotherBoard to use for EMC?
[21:23:42] <paragon36> s/hot/shot/
[21:24:04] <alex_joni> danex: that's hard to say
[21:24:22] <alex_joni> re: same method.. there are some experimental debs which mean you don't need to build emc2
[21:25:41] <danex> Where would the experimental debs be located?
[21:27:36] <danex> As for motherboards, I would like to know of one that has worked well for someone in the past. Just as a starting point.
[21:28:50] <cradek> danex: I think people generally have good luck with late PIII boards (800-1000 MHz) without onboard video
[21:29:17] <alex_joni> right
[21:29:38] <alex_joni> I have a system which also works VERY well.. Athlon 1400+ with MSI mobo
[21:29:38] <SWPadnos> also, Ray has appraently had good luck with a Via CN700-based mini ITX board (even though it has onboard video ...)
[21:29:41] <cradek> ideally with 512MB or more of ram
[21:29:59] <alex_joni> mine is 256, but I don't feel the lack of more
[21:33:51] <jepler> cradek: do you trust that 2-year-old athlon with VIA chipset?
[21:34:32] <danex> The onboard video problems I am aware of, had problems with my first machine on this
[21:34:51] <alex_joni> danex: there is a page somewhere with rtai latency test results
[21:35:36] <danex> This was resolved with help from all on this irc, Thanks
[21:35:41] <awalli1> Rugludallur: I'm online now
[21:39:58] <cradek> jepler: umm, it's always had an nvidia card, and has never worked right; I don't know how related those two things are.
[21:40:46] <jepler> I see
[21:41:23] <lerneaen_hydra> yo
[21:41:28] <lerneaen_hydra> err
[21:41:30] <lerneaen_hydra> lo
[21:42:11] <alex_joni> danex: so.. need directions for the new package?
[21:42:49] <alex_joni> cradek: what kind of via?
[21:43:18] <alex_joni> mine says 'VIA Technologies, Inc. VT8377 [KT400/KT600 AGP] Host Bridge'
[21:44:07] <cradek> Host bridge: VIA Technologies, Inc. K8M800 Host Bridge
[21:44:16] <alex_joni> oh, so fairly newer
[21:44:39] <cradek> if you say so
[21:44:41] <alex_joni> that one is for AMD64
[21:44:45] <lerneaen_hydra> hmm, isn't that for the K8, AMD64 arch?
[21:44:54] <cradek> AMD Sempron(tm) Processor 3300+
[21:44:56] <danex> alex_joni: If an expermintal deb is out there, I would like to try it,
[21:44:57] <cradek> (2000MHz)
[21:45:00] <lerneaen_hydra> oh, ok
[21:45:06] <lerneaen_hydra> socket 754 probably then
[21:45:07] <alex_joni> danex: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Emc2.1.0 be my guest :)
[21:45:08] <lerneaen_hydra> 32 bit
[21:46:09] <alex_joni> strange, via doesn't say anything beside amd64 /opteron for k8m800
[21:46:10] <alex_joni> http://www.via.com.tw/en/products/chipsets/k8-series/k8m800/
[21:46:36] <lerneaen_hydra> the sempron is an amd64 core with various bits stripped out IIRC
[21:47:08] <jepler> mostly a smaller cache
[21:47:23] <lerneaen_hydra> no 64 bit extensions either
[21:47:49] <jepler> some semprons do have 64 bit extensions
[21:47:53] <jepler> model name : AMD Sempron(tm) Processor 3000+
[21:48:04] <jepler> Linux bald.unpythonic.net 2.6.17-1.2141_FC4_jepler2 #1 Sun Sep 17 10:38:09 CDT 2006 x86_64 x86_64 x86_64 GNU/Linux
[21:48:08] <jepler> like this one
[21:48:32] <cradek> I don't think mine does, even though it has 300 bogomegahertz more than yours?
[21:49:02] <alex_joni> danex: http://issaris.homelinux.org/~takis/projects/rtai/livecd/list.php
[21:49:52] <jepler> I don't know
[21:50:41] <alex_joni> cradek: http://www.amdboard.com/sempron_062405.html
[21:50:42] <cradek> maybe I unwittingly got x64
[21:50:49] <alex_joni> there is a revision which has amd64
[21:50:52] <alex_joni> not x64 :)
[21:51:16] <alex_joni> x86-64
[21:51:55] <cradek> it says the stepping is "2" which is not at all like "BA" or "B0"
[21:52:42] <lerneaen_hydra> jepler: oh, didn't know that
[21:52:45] <lerneaen_hydra> cool
[21:52:46] <alex_joni> can you paste the full line?
[21:52:53] <alex_joni> family, model, stepping?
[21:53:04] <lerneaen_hydra> cradek: how's RT on it?
[21:53:09] <cradek> CPU0: AMD Sempron(tm) Processor 3300+ stepping 02
[21:53:23] <cradek> CPU: After all inits, caps: 078bfbff e3d3fbff 00000000 00000410 00000001 00000000 00000001
[21:54:19] <cradek> lerneaen_hydra: like I said earlier, it has always failed the latency test and has always had an nvidia graphics card. I don't know exactly how those two facts are related.
[21:54:32] <lerneaen_hydra> oh
[21:54:39] <lerneaen_hydra> at which base_freq?
[21:54:52] <lerneaen_hydra> using the opensource driver, right?
[21:54:53] <cradek> whatever the rtai test uses
[21:55:05] <cradek> yes I'm pretty sure I tried it that way
[21:55:15] <lerneaen_hydra> hmm, funky
[21:55:15] <cradek> (I don't use it for realtime usually)
[21:55:18] <jepler> my CPU is http://pastebin.ca/319909
[21:55:26] <danex> alex_joni: Thanks, I am downloading the Test Disk and will try it on a few machines. The debs I will try tommorow.
[21:56:04] <alex_joni> danex: I mostly linked that test disk for their database
[21:56:14] <cradek> jepler: except for MHz mine is the same
[21:56:49] <jepler> cradek: does it have this line in cpuinfo? address sizes : 40 bits physical, 48 bits virtual
[21:56:56] <cradek> well as far as family/model/stepping goes
[21:57:01] <alex_joni> jepler: what made you try the x86_64 ?
[21:57:10] <cradek> no it doesn't have that
[21:57:13] <alex_joni> the address sizes?
[21:57:14] <danex> alex_joni: the more information the better, true?
[21:57:19] <alex_joni> danex: sure :D
[21:57:22] <cradek> yours has a lot of extra lines
[21:57:25] <jepler> alex_joni: because it was the new technology?
[21:57:40] <alex_joni> danex: the emc2 live CD also has the rtai latency test on it
[21:57:47] <alex_joni> jepler: ok, so you knew it had amd64 ?
[21:57:51] <jepler> alex_joni: yes
[21:57:54] <danex> Have to go now, Thanks everyone
[21:57:55] <alex_joni> because the flags don't say that
[21:58:03] <alex_joni> danex: please report on the emc2 package
[21:58:14] <jepler> alex_joni: http://www.amdcompare.com/us-en/desktop/details.aspx?opn=SDA3000AIO2BX
[21:58:17] <cradek> I'll try to boot the 64 bit dapper on mine, I think I have a cd
[21:58:18] <alex_joni> either success, or what problems you encountered
[21:59:01] <alex_joni> jepler: I see no amd64 there :D
[21:59:01] <danex> alex_joni: I will report, again Thanks
[21:59:14] <alex_joni> bye :)
[21:59:27] <alex_joni> jepler: nm
[21:59:32] <alex_joni> Operating Mode32/64
[21:59:58] <alex_joni> cradek: guess you can check yours if you still know the P/N
[22:00:30] <cradek> grep P/N /proc/cpuinfo
[22:00:39] <cradek> * cradek shrugs
[22:00:50] <Rugludallur> hmm is someone trying to run on x86_64 ubuntu ?
[22:01:01] <jepler> I'm surprised nothing in the 'flags' says 64 bits
[22:01:11] <jepler> Rugludallur: I run emc2 sim on x86_64 ubuntu just fine
[22:01:26] <Rugludallur> jepler: really, I had problems compiling when I tried on a machine at work
[22:01:27] <alex_joni> cradek: grep the markings on the CPU :P
[22:01:29] <jepler> based on all I've read, realtime support for x86_64 is not ready yet
[22:01:42] <SWPadnos> Rugludallur, I also have a sim install on x86_64 SMP
[22:01:52] <cradek> alex_joni: umm, it's in the case under a heat sink and stuff
[22:01:56] <Rugludallur> hmmf, need to check that again
[22:02:00] <cradek> I guess I really don't care :-)
[22:02:08] <alex_joni> x86_64 (beta) (from www.rtai.org)
[22:02:29] <SWPadnos> yeah - only available through CVS (/ CVS tarballs), I Think
[22:02:58] <jepler> out of all the 3300s, one stepping is 32-bit only, the others are 32/64. http://www.amdcompare.com/us-en/desktop/SideBySide.aspx?opn=SDA3300AIO2BO&opn=SDA3300AIO2BA&opn=SDA3300AIO2BA&opn=SDA3300AIO2BX
[22:03:15] <SWPadnos> cradek probably got lucky ;)
[22:03:49] <alex_joni> cradek: you can heat it up to 70C
[22:04:00] <alex_joni> if it still works there's 50% chance it only works for 32
[22:04:11] <Rugludallur> lol
[22:04:13] <cradek> ha
[22:04:26] <jepler> confusingly SDA3300AIO2BA is both a 32-only and a 32/64 part !?
[22:04:31] <cradek> so according to that table you can't tell from the stepping.
[22:04:36] <cradek> nice.
[22:04:38] <SWPadnos> hmmm - use that as an EStop for motor overheating. mount CPU to motor, and when the motor gets too hot, the computer crashes
[22:04:50] <Rugludallur> just try booting a 64 bit livecd :D
[22:04:51] <alex_joni> ROFL
[22:04:59] <alex_joni> SWPadnos: together with a chargepump of course
[22:05:06] <SWPadnos> of course :)
[22:05:10] <cradek> _Remove_Item__Remove_Item__Remove_Item__Remove_Item__Remove_Item_
[22:05:20] <alex_joni> cradek lost it
[22:05:23] <SWPadnos> hey - this isn't the SQL channel
[22:05:31] <anonimasu> * anonimasu slaps cradek
[22:05:34] <alex_joni> 5 times he lost it :D
[22:05:43] <alex_joni> oh-oh..
[22:05:47] <alex_joni> * alex_joni fears for anonimasu
[22:05:50] <lerneaen_hydra> ... wth?
[22:06:11] <lerneaen_hydra> I leave for 5 minutes and just look what you're all up to
[22:06:26] <alex_joni> Remove ItemRemove ItemRemove ItemRemove Item
[22:06:31] <lerneaen_hydra> I swear, I really do have to keep an eye on you all the time
[22:06:32] <alex_joni> I only have 4 on that page :)
[22:06:36] <lerneaen_hydra> * lerneaen_hydra sigh
[22:11:39] <Rugludallur> just try booting a 64 bit livecd :D
[22:11:53] <Rugludallur> errr sorry guys, cought in a loop
[22:12:01] <alex_joni> heh
[22:13:58] <jepler> cradek: look for "lm" on the flags: line
[22:14:03] <jepler> cradek: that is "long mode"
[22:18:45] <alex_joni> #define X86_FEATURE_LM (1*32+29) /* Long Mode (x86-64)
[22:21:00] <jepler> 15:53:23 <cradek> CPU: After all inits, caps: 078bfbff e3d3fbff 00000000 00000410 00000001 00000000 00000001
[22:21:33] <jepler> I'm guessing yes, because the 29th bit in e3d3fbff is set.
[22:21:56] <alex_joni> why in e3d.. ?
[22:22:07] <alex_joni> * alex_joni is too tired to think :)
[22:22:09] <jepler> it's in the second word .. 1*32 bits in
[22:22:35] <alex_joni> so 078.. is the lowermost word?
[22:23:06] <SWPadnos> depends on the print order -I was wondering that myself
[22:23:24] <alex_joni> 3d now is 1*32+30
[22:23:29] <alex_joni> 3d now is 1*32+31
[22:23:34] <alex_joni> 3d now ext is 1*32+30
[22:24:24] <SWPadnos> well - thatt'll tell you which word is which
[22:24:41] <alex_joni> jepler: you sure 29 is set?
[22:24:53] <alex_joni> 11100011110100111111101111111111
[22:25:08] <SWPadnos> yep - it's the third '1'
[22:25:19] <alex_joni> right, 31,30,29..
[22:25:44] <alex_joni> 1110.0011..1101.0011..1111.1011..1111.1111
[22:26:50] <alex_joni> cool :)
[22:26:50] <SWPadnos> so it either has no 3DNow and no 64-bit, or it hs both (I'm betting on both)
[22:27:12] <alex_joni> there are no recent amd's without 3dnow
[22:27:19] <SWPadnos> that's why I'd bet on both ;)
[22:27:23] <paragon36> Hello Chaps, I just noticed in TkEMC there is are buttons to select spindle speed < > when using stepper control .ini what type of output is seen at the parport PWM?
[22:27:25] <alex_joni> I checked the mp flag which isn't set
[22:27:46] <alex_joni> paragon36: in emc2 you can link the speed prescription (float) to anything you like
[22:27:53] <SWPadnos> paragon36, those buttons have nothing to do with the parallel port PWM - that kind of configuration is done in HAL
[22:27:56] <alex_joni> either a DAC or a freqgen outputting PWM
[22:28:42] <paragon36> Whats the most common DAC or PWM?
[22:28:42] <SWPadnos> don't the buttons actually set output bits (for hardware devices that have increase/decrease buttons)?
[22:28:55] <alex_joni> if you have a motion control board, then usually DAC
[22:29:08] <alex_joni> but with steppers/parport PWM is the most common
[22:29:21] <alex_joni> although people with spindle speed control usually go for the motion control board too :)
[22:29:28] <alex_joni> SWPadnos: no
[22:29:48] <alex_joni> SWPadnos: those don't exist anymore, the TkEMC buttons simply increase/decrease the current spindle speed
[22:30:00] <SWPadnos> ok - I was thinking of the (now missing) spindle-incr / decr pins
[22:30:03] <SWPadnos> ok
[22:30:13] <alex_joni> of course spindle_speed_override is multiplied with that, and the output is written to motion.spindle-speed-out
[22:31:10] <alex_joni> this is a _really_ usefull email response: http://www.freelists.org/archives/oracle-l/02-2005/msg00839.html
[22:31:29] <A-L-P-H-A> anyone know how to set the gamma in X with nvidia drivers?
[22:31:30] <paragon36> I think the DC spindle motor I have runs 0-190V DC this was from notes I found when taking measurments 3 years ago so I could be wrong. Does this seem possible? 190V!
[22:31:58] <alex_joni> 190V is not much for a biggish DC motor
[22:32:23] <paragon36> 190V DC ?
[22:32:28] <alex_joni> bet it was driven from 110AC regulated
[22:32:36] <alex_joni> paragon36: yeah
[22:32:47] <paragon36> Driven 240V ac
[22:33:02] <alex_joni> 110AC regulated only give 160V DC
[22:33:06] <alex_joni> so you might be right :)
[22:33:47] <alex_joni> the general rule: the higher the voltage, the smaller the current for the same rating
[22:34:00] <alex_joni> that means thinner wires & all, but better insulation
[22:34:22] <alex_joni> nowadays servomotors are at least 600V AC
[22:35:50] <alex_joni> I'm off to bed
[22:35:54] <alex_joni> good night all
[22:36:47] <A-L-P-H-A> night aj
[22:37:02] <SWPadnos> see you Alex
[22:38:45] <paragon36> The spindle board has like 30 + resisters and it has LM324N ic's which I think are high gain frequency compensated operation amplifiers the more I look at the board the more it bafffles me! It takes 240V AC which goes into a small 3KVA tranformer that steps down to 12V it has some really big diodes (P600K) and like 0.1uf+47ohm X2 RC Units other than that, thats it. Can't make head or tail...
[22:38:47] <paragon36> ...how it works?
[22:39:14] <paragon36> CU Alex
[22:39:35] <SWPadnos> it's probably a PWM supply - basically a switch-mode supply like those in a computer
[22:40:16] <SWPadnos> I'd be really surprised if the output of the 3 kVA transformer is intended to be 12V
[22:40:54] <paragon36> Ive worked with SMPS's and built some using H-Bridge pwm this doesnt appear to be one plus it was made in 1988-89
[22:41:04] <lerneaen_hydra> 3000VA @ 12V? O.o
[22:41:20] <lerneaen_hydra> that's like 250A
[22:41:21] <paragon36> Ooops 3va ... lol
[22:41:27] <lerneaen_hydra> oh, right
[22:41:32] <SWPadnos> SMPSs have been around for a lot longer than that
[22:41:33] <SWPadnos> heh
[22:41:35] <lerneaen_hydra> that's a bit better
[22:41:46] <paragon36> max load 1.5A per winding
[22:41:59] <SWPadnos> 30 VA then
[22:42:09] <SWPadnos> (total, -ish)
[22:43:27] <paragon36> Could be!
[22:44:04] <SWPadnos> that's the control transformer. the incoming line is likely rectified then used in PWM mode to drive the motor
[22:46:30] <paragon36> Think your right there that what the large diodes could be for. Rather than using a one piece diode as is common today
[22:46:52] <paragon36> One piece rectifer I mean.
[22:48:12] <paragon36> any idea what txn610 ic is?
[22:49:04] <SWPadnos> nope
[22:50:23] <paragon36> Me neither just google it and it comes up as a hard to get but not what it does :-(
[22:50:49] <SWPadnos> yeah - all those "obsolete parts"vendors are annoying in that you have to subscribe to get datasheets
[22:51:16] <paragon36> Information is money ;-)
[22:51:24] <paragon36> Well to them!
[22:51:59] <SWPadnos> it's an SCR
[22:52:09] <SWPadnos> http://www.datasheets.org.uk/specsheet.php?part=TXN610
[22:54:06] <paragon36> Thanks SWPadnos... the center legs are attached to 240V ac
[22:54:30] <SWPadnos> dunno the pinout - only that it's an SCR ;)
[23:24:21] <jmkasunich> paragon36: its probably a phase controlled bridge
[23:25:44] <robin_sz> meep?
[23:25:52] <jmkasunich> ?peem
[23:27:27] <robin_sz> are you well?
[23:27:58] <paragon36> Thanks JM ... I have found a MC7812 positive voltage regulator on the board and I suspect that this may be part of the issue as I am no longer getting 12Vdc at the desired pin but rather 2-v volts.
[23:28:05] <robin_sz> tidies out the basement yet?
[23:28:28] <robin_sz> paragon36, or something is fried and pullign it down
[23:28:47] <robin_sz> more likely imho
[23:29:06] <paragon36> Could be the LM324n's
[23:29:21] <robin_sz> could be almost anything
[23:29:40] <robin_sz> see what gets hot :)
[23:29:59] <jmkasunich> robin_sz: yes, lots of progress in the basement
[23:30:16] <robin_sz> paragon36, don't suppose you have a toneohm?
[23:30:23] <robin_sz> jmkasunich, more space?
[23:30:27] <jmkasunich> before: http://jmkasunich.dyndns.org/cgi-bin/blosxom/shoptask/mess-before.html
[23:30:38] <jmkasunich> after: http://jmkasunich.dyndns.org/cgi-bin/blosxom/shoptask/mess-12-16-06.html
[23:31:03] <jmkasunich> since then I've moved the machine and done some other work
[23:31:04] <A-L-P-H-A> anyone know how to manually mount a samba share of another linux box?
[23:31:17] <robin_sz> A-L-P-H-A, yeah
[23:31:17] <paragon36> This board take's 240Vac input would it cause any harm to run it at a safer ac voltage for testing as I posess a 110 variac transformer?
[23:31:18] <A-L-P-H-A> or should I use NFS?
[23:31:25] <robin_sz> nah, smbclient
[23:31:30] <SWPadnos> smb:// in the file browser
[23:31:36] <lerneaen_hydra> jmkasunich nice mess you've got there
[23:31:36] <robin_sz> that also
[23:31:48] <SWPadnos> or smbmount
[23:31:48] <jmkasunich> lerneaen_hydra: its better now
[23:31:55] <paragon36> smbmount
[23:31:55] <lerneaen_hydra> couldn't have been fun cleaning all that out
[23:31:58] <robin_sz> smbmount is what I meant
[23:31:59] <A-L-P-H-A> haha... I kept typing in smb://\\
[23:32:06] <A-L-P-H-A> that's why it wasn't working! :)
[23:32:09] <A-L-P-H-A> robin_sz: , thanks.
[23:33:46] <paragon36> robin_sz: toneohm > buzzer on multimeter for 0 resistance? Yes
[23:33:56] <A-L-P-H-A> SWP you too
[23:35:37] <robin_sz> paragon36, sorta, makes low buzz on high res, hi squeal on low res, you can trace the path to the low res component thats fried by ear ...
[23:35:43] <jepler> A-L-P-H-A: did you try the "xgamma" command? e.g., xgamma -gamma 0.7
[23:36:01] <robin_sz> ear can detect which track of a PCB it goes down as you follow it along
[23:37:05] <paragon36> In that case no I only have the High squeel on 0 resistance ...
[23:37:46] <robin_sz> oh, fair enough .. find the hot thing then
[23:38:01] <robin_sz> that current has to go somewhere to pull it to 2V
[23:38:20] <robin_sz> unless its folded back of course
[23:39:07] <paragon36> Another idea is I could hook up the 12v ac to another 12vac supply for testing rather then playing with the 240V.
[23:39:18] <robin_sz> good call
[23:39:31] <robin_sz> saves dying
[23:40:02] <paragon36> Yep beleive it or not that's quite important to me ... lol
[23:41:07] <paragon36> What other means could I choose to drive this 190v dc motor? pwm h-bridge?
[23:41:27] <anonimasu> yes, if you are good at power electronics
[23:41:39] <jmkasunich> how many amps?
[23:42:01] <anonimasu> www.rutex.com
[23:42:05] <anonimasu> has drives up to 200V
[23:42:09] <jmkasunich> thats not a current rating
[23:42:13] <jmkasunich> how many amps?
[23:42:22] <jmkasunich> ;-)
[23:42:23] <anonimasu> 200v 10A
[23:42:57] <paragon36> I built an induction heater using a PLL and irfp460's heated a half inch steel rod to white heat. ... (not sure of the amps as I cant see the motor as it is enclosed.
[23:43:26] <jmkasunich> guess?
[23:43:29] <paragon36> Oh the motor board say 8ams
[23:43:38] <jmkasunich> ok
[23:43:42] <jmkasunich> where are you located?
[23:44:11] <CIA-8> 03jepler 07TRUNK * 10emc2/src/ (configure configure.in): check for working groff -Thtml
[23:44:32] <paragon36> UK London ... Sorry I was wrong it does not say 8 at all god knows where I pulled that from!
[23:44:41] <jmkasunich> paragon36: the reason I ask is that I have some drives that might be suitable
[23:44:56] <jmkasunich> but its not worth the hassle of shipping one across the ocean
[23:45:07] <paragon36> :-)
[23:45:43] <A-L-P-H-A> I feel soo geek right now... 5 machines near me... all running linux... except one box dual boots into windows.
[23:46:00] <lerneaen_hydra> A-L-P-H-A: yay!
[23:46:24] <paragon36> :-)
[23:47:53] <lerneaen_hydra> A-L-P-H-A: seen rfad yet?
[23:48:07] <lerneaen_hydra> * lerneaen_hydra is feeling naggy today
[23:48:54] <paragon36> The board appears to be a Fenner Electronic Controls anyone heard of them ... Maybe out of business as it was made in 1988-89 ;-)