#emc-devel | Logs for 2009-01-11

[00:13:35] <BigJohnT> jepler: do you know the syntax for the pyVCP tabs widget?
[00:30:29] <SWPadnos> the properties probably get passed through to an underlying tk widget
[00:31:45] <BigJohnT> looks like it uses the notebook widget
[00:32:11] <BigJohnT> * BigJohnT don't know tk :/
[00:32:51] <SWPadnos> me either, which is why I couldn't be more specific ;)
[00:32:58] <BigJohnT> :)
[00:33:13] <BigJohnT> * BigJohnT heads upstairs to relax
[00:53:55] <CIA-1> EMC: 03tissf 07TRUNK * 10emc2/docs/src/ (docs.xml index.tmpl index_fr.tmpl): french translation update
[01:24:04] <CIA-1> EMC: 03cmorley 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/ (stepconf.py stepconf.glade): add a checkbox for creation of desktop shorcut (symlink) to config files. Defaults to always creating them...maybe too intrusive for experienced user btu novice woud be happy
[01:32:32] <cradek> what became of diameter mode? I thought it was in cvs, but I don't see it.
[01:34:22] <jepler> I thought we argued too hard about it
[01:34:45] <jepler> bigjohnt: no clue about the tab widget
[01:36:17] <cradek> jepler: hm, I thought it was nearly argued out
[02:21:45] <jepler> cradek: you're a terrible programmer
[02:21:57] <jepler> I can't believe how you failed to include cstdlib
[02:38:42] <SWPadnos> I include a little cstdlib with every meal
[02:39:12] <jmkasunich> adds flavor
[02:39:23] <SWPadnos> and it's good for bad breath
[03:13:56] <jepler> I can't imagine the glee he feels when he finds something to embarass us about
[03:14:00] <jepler> "oh I've got them this time"
[03:14:03] <jepler> cackles with glee
[04:08:42] <cradek> jepler: I have no idea what cstdlib is
[04:09:10] <SWPadnos> I think it's the C standard library stuff for c++, but that's a guess
[04:09:14] <cradek> I don't get any warnings - usually that's what tells me I have to add a header file :-)
[04:09:32] <SWPadnos> gcc 4.3 is a bit more pedantic, it seems
[04:11:07] <cradek> "will at least compile" haha
[04:12:14] <SWPadnos> in a theoretical way, it would be nice to have more compile farm slots/buildbot slaves to test things on other systems
[04:12:38] <SWPadnos> like 8.10, soon 9.04, maybe other Linux versions, and also later compilers
[04:12:52] <cradek> seb tried to work with paul to get that, but paul wouldn't cooperate
[04:13:00] <SWPadnos> not for long anyway
[04:13:38] <cradek> I guess I could set up a newer machine and try to build sim - but so boring
[04:13:53] <cradek> if he'd send a patch and it didn't break the build here, I'd apply it
[04:14:08] <SWPadnos> sure
[04:14:24] <cradek> I hate to blindly add stuff with no way of testing except to wait for a few days and see if paul says something else snippy
[04:14:49] <SWPadnos> we'll be in the same boat as before soon - our "LTS" package/OS will be too out of date for some people, and there will be pressure to get things working and tested on a newer OS version
[04:14:59] <SWPadnos> yeah, that's no fun
[04:19:46] <cradek> ok, I answered him
[04:20:26] <SWPadnos> uh oh. dare I look? ;)
[04:20:55] <cradek> I'm always a gentleman
[04:21:10] <SWPadnos> um. yeah
[04:21:12] <SWPadnos> that
[04:21:14] <SWPadnos> that
[04:21:17] <SWPadnos> argh
[04:21:22] <SWPadnos> that's what I was going to say
[04:26:05] <cradek> my strange story for the day: some guy has been trying to call me for a few days. today I talked to him. he was calling about something I posted on my website (a font I made). he prefers the phone to email, he says.
[04:26:15] <cradek> he was nice, but the whole thing was kind of surreal
[04:26:15] <SWPadnos> odd
[04:26:44] <cradek> I was sure it would be a scam, but no - he just wanted to talk about fonts
[04:27:04] <SWPadnos> did you whistle them to him over the phone? :)
[04:27:08] <cradek> he never told me his name or anything.
[04:27:30] <cradek> funny - I used a modem today. they sure are slow.
[04:27:37] <SWPadnos> heh
[04:27:45] <SWPadnos> sqrt(acceptable_speed)
[13:09:15] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/src/hal/user_comps/gs2_vfd.c: add is-stopped pin
[16:04:41] <jepler> cradek: for the headers specified in the C standard <foo.h>, C++ specifies a header <cfoo>. I am not sure why.
[16:06:41] <jepler> "The C International Standard specifies 18 headers which must be provided by a conforming hosted implementation. The name of each of these headers is of the form name.h. The C++ Standard Library includes the C Standard Library and, hence, includes these 18 headers. Additionally, for each of the 18 headers specified by the C International Standard, the C++ standard specifies a corresponding header that is functionally equivalent to its C library cou
[16:13:39] <alex_joni> and there's printf() in that file which isn't defined
[16:27:50] <jepler> well slap me on the ass and call me sally
[16:34:38] <SWPadnos> pass
[16:43:37] <steves_logging> steves_logging is now known as steve_stallings
[18:04:04] <cradek> oh it's about printf? I could take those out (or #if 0 instead of if(0))
[18:04:15] <alex_joni> probably printf()
[18:04:22] <alex_joni> * alex_joni is guessing
[18:04:25] <cradek> it's a bad habit of mine to leave those droppings everywhere
[18:16:50] <steve_stallings> EMC at Cabin Fever - Since cloning the web site is not workable, I would like to have a printout of the latest EMC documentation for visitors to browse.
[18:17:16] <steve_stallings> Is the 2.3 documentation OK to use, or should I stick with the 2.2 stuff?
[18:17:59] <steve_stallings> What are the current goals for 2.3 in terms of feature additions and release date?
[18:19:45] <steve_stallings> It is looking like Chris Helgesen will manage to put together a slide show about his Hurco retrofit. We may also have two setups running EMC.
[18:23:12] <steve_stallings> There are several nice videos of EMC on Youtube.com but I don't think there is any way of capturing them for off-line display and net access at the show is limited.
[18:27:39] <jepler> steve_stallings: bigjohnt is probably the best one to ask about the state of the documentation
[18:27:59] <jepler> for reading, the 2.3 docs are probably fine ..
[18:28:09] <jepler> I think stuff mostly gets documented after the developer thinks it's finished
[18:30:10] <KimK> steve_stallings: Sometimes videos can be captured in Firefox by playing them through, then rummaging in Firefox's cache in reverse chronological order for the most likely suspects.
[18:34:37] <KimK> Did you have those little Panasonic servos and drives at the 2008 Workshop? Still have any?
[18:35:33] <SWPadnos> I think that was Scott (something) - the blond guy who had the Puma robot
[18:35:43] <SWPadnos> blonde
[18:35:45] <SWPadnos> whatever
[18:36:41] <KimK> Oh, OK, I'm mistaken then. Nevermind. Thanks for straightening me out.
[18:37:06] <SWPadnos> if it's the same set of small panasonic servos I
[18:37:10] <SWPadnos> I'm thinking of
[18:37:47] <KimK> He had four (eight?) of them mounted in a portable demo cube/frame, that one?
[18:37:54] <SWPadnos> yep
[18:38:07] <SWPadnos> the plexiglas demo thingie
[18:38:10] <SWPadnos> +s
[18:38:54] <KimK> I've recently realized that a set would be handy to have around.
[18:39:02] <steve_stallings> Yep, that was Scott. He had motors and drivers, but more motors than drivers if I remember correctly.
[18:42:33] <steve_stallings> Any idea where the Firefox cache files are on a widows box? I cannot seem to find anything. Perhaps the Flash applet that Youtube uses does not leave droppings.
[18:42:42] <KimK> OK, thanks. I'm sure I have his contact info, just have to find it.
[18:43:45] <KimK> Should be under the "salt" folder under Firefox/Mozilla under Docs & Settings, I would guess?
[18:44:58] <KimK> Docs & Settings / Applications / Firefox (Mozilla?) / "salt" / cache / You are here
[18:45:08] <KimK> Maybe?
[18:46:27] <KimK> "salt" is a randomly-named folder to thwart pre-written viruses, scripts, etc.
[18:48:40] <SWPadnos> c:\documents and settings\<username>\application data\mozilla\firefox\profiles\<random directory name>
[18:48:58] <SWPadnos> under that is a cache dir, as well as other things
[18:49:13] <KimK> Thanks, SWP
[18:50:41] <SWPadnos> hmmm. actually, it looks like the cache isn't there
[18:50:49] <SWPadnos> but you should be able to see where it is in about:config
[18:51:03] <KimK> I'm sure you're close, though
[18:53:31] <SWPadnos> this may help: http://www.walkernews.net/2007/11/19/how-to-save-flash-video-from-firefox-cache-folder/
[18:56:19] <dgarr> ´╗┐steve_stallings: ubuntu has a package: clive that will download youtube videos to a playable file
[18:59:44] <steve_stallings> Thanks. At the moment I am wainting for the cache to fill on the video of 5 axis demo from CNC workshop so I can try to look at it offline.
[19:00:13] <KimK> I hope it works for you
[19:05:36] <steve_stallings> Well it captured a 20MB file. Now I just have to figure out how to do something with it.... AH! The VLC player for Windows handles it OK. Brightness, contrast etc. are way off base, but can be adjusted.
[19:10:59] <dgarr> totem (standard on ubuntu i think) works for me
[19:11:35] <dgarr> clive is a neat program, the --throttle option is useful for downloading sometimes
[19:23:50] <steve_stallings> thanks guys for the video help
[19:24:06] <steve_stallings> steve_stallings is now known as steves_logging
[19:31:16] <steves_logging> steves_logging is now known as steve_stallings
[19:31:51] <steve_stallings> So, John thought it was safe to come back.... Oh, hi John. I had a question about documentation.
[19:32:21] <BigJohnT> hi Steve
[19:32:40] <steve_stallings> I am attending the Cabin Fever show next weekend and wanted to have some EMC stuff to show. Net access is limited, so I planned to print out documentation to browse.
[19:33:14] <steve_stallings> Is the 2.3 stuff suitable? Would it be better to wait until, say Thursday, to print it?
[19:34:38] <BigJohnT> AFAIK I don't have any major changes in the next few days. And the 2.3 docs are the latest and most up to date as I stopped adding to the 2.3 in December or so
[19:35:07] <BigJohnT> You printing the PDF's
[19:35:07] <steve_stallings> OK, thanks.
[19:35:13] <steve_stallings> yes
[19:35:26] <BigJohnT> one second
[19:36:52] <BigJohnT> are you printing them to reference or to hand out?
[19:38:17] <BigJohnT> The getting started guide is pretty short it is only 35 pages
[19:39:05] <BigJohnT> I meant I stopped adding to 2.2 in December or so... just got up from a nap
[19:39:38] <steve_stallings> Making a book to browse at my booth where there will be a few other EMC related things nearby.
[19:39:40] <SWPadnos> this has a (probably incomplete) list of changes: http://cvs.linuxcnc.org/cvs/emc2/debian/changelog?rev=1.46
[19:40:28] <SWPadnos> cradek's concave cutter comp is missing, for instance (and that's a big one)
[19:40:50] <steve_stallings> will that be in the 2.3 release?
[19:40:57] <BigJohnT> ok, I see one little thing I need to fix on the cover of the integrator and cradek's cutter comp is missing
[19:41:04] <SWPadnos> it's in TRUNK, so unless someone complains, yes it will
[19:41:45] <steve_stallings> good to know, the MACH crowd has been making big noise about their updated cutter comp of late
[19:43:48] <BigJohnT> hmmm, I'm working with Peter Wallace on a THC for the FPGA cards but we are just getting started
[19:45:01] <steve_stallings> hardware or HAL based so it doesn't need support in the motion planner?
[19:45:40] <BigJohnT> harware
[19:45:44] <BigJohnT> hardware
[19:46:35] <steve_stallings> some of the magic is in compensating the gap voltage when motion slows down in a corner, is that going to be possible?
[19:47:00] <steve_stallings> this is called "anti-dive"
[19:47:24] <SWPadnos> actual velocity is available in HAL, it's the "final programmed value" that isn't so easy to get
[19:47:41] <BigJohnT> from my research blocking the compensation when your not "at speed" works well
[19:47:50] <SWPadnos> (ie, F word modified by GUI FO, and limited by accel/vel parameters)
[19:48:30] <BigJohnT> It will have to be turned on and off by the g code I think
[19:50:07] <BigJohnT> I did quite a bit of work on it last year up to the point of needing the A-D into EMC...
[19:50:25] <steve_stallings> should the corner comp stuff be listed in the changelog?
[19:50:58] <steve_stallings> most of the THC's used with Mach only use a window comparator
[19:51:16] <SWPadnos> I have an A/D + D/A card that plugs into a Mesa connector
[19:51:26] <SWPadnos> but it's expensive as hell :)
[19:52:17] <BigJohnT> we don't need much just one channel and Peter has an inexpensive solution in mind for me to try out
[19:52:43] <SWPadnos> heh. imne's 200 kHz, 16 bit, 6 input and 8 output :)
[19:52:46] <SWPadnos> mine's
[19:52:58] <BigJohnT> by window comparator do you mean a rolling average of the arc voltage?
[19:53:20] <steve_stallings> PMDX has been wondering if we can stand the product support load if we introduce a THC.
[19:53:28] <SWPadnos> no - above this = go up / below that go down / inside the window do nothing
[19:53:51] <SWPadnos> (or the other way around, but stilll :) )
[19:53:58] <steve_stallings> No, they just had a manual voltage setpoint and the window was a nominal % above and below the set point.
[19:55:32] <steve_stallings> The newer fancy CandCNC stuff uses serial communications to a micro in the THC that allows the computer (and therefore the GUI) to control the setpoint.
[19:55:46] <BigJohnT> I plan on adding a rolling average if possible to smooth out the lumps but we will see
[19:56:12] <BigJohnT> I plan on using the pyVCP panel to set voltage and range
[19:56:38] <steve_stallings> HAL should make it possible to build a relative complex control
[19:56:59] <BigJohnT> complex but simple to run is my goal :)
[19:57:28] <steve_stallings> just don't tempt fate by going for idiot proof
[19:57:43] <BigJohnT> I usually stop at idiot resistant
[19:58:23] <BigJohnT> can HAL see a variable in a g code file?
[20:00:00] <SWPadnos> only if it's output to a digital/analog output pin on the motion controller
[20:00:08] <SWPadnos> with M6x or whatever those codes are
[20:00:17] <steve_stallings> if output by an M code
[20:00:25] <steve_stallings> dang I'm slow
[20:00:26] <SWPadnos> the #variables aren't accessible to HAL
[20:00:57] <BigJohnT> just thinking outloud that if I somehow put the voltage in the g code file the it could be set in the post processor...
[20:01:58] <SWPadnos> you could use S for that
[20:02:03] <steve_stallings> prehaps use the S word and pretend it was the spindle
[20:02:10] <SWPadnos> unless you need a spindle and a torch
[20:02:11] <BigJohnT> ahh, yes
[20:02:13] <SWPadnos> heh
[20:02:39] <steve_stallings> OK, I quit, too old or SWP has a negative time interface to IRC
[20:03:32] <BigJohnT> I think he is from the future
[20:03:54] <SWPadnos> no comment
[20:03:58] <steve_stallings> all I know is that he has the fasted fingers on the channel
[20:04:12] <SWPadnos> I'm thinking of thawing them soon
[20:04:22] <steve_stallings> and he must have a key preprogrammed with "hey"
[20:04:58] <BigJohnT> they came out with the heh key in 2010 I think
[20:05:03] <SWPadnos> a super-duper gaming keyboard with massive fragging macros would be useful
[20:05:57] <BigJohnT> is there any difference in the g code when using cradek's cutter comp stuff?
[20:06:24] <steve_stallings> ... and could any mention of it make it into the 2.3 manuals?
[20:06:45] <BigJohnT> as quick as I can understand what he did I can :)
[20:08:21] <steve_stallings> I didn't join in the praise earlier when documentation was mentioned, so let me say now that your work on it is greatly appreciated!!
[20:09:56] <steve_stallings> gotta go run some errands, see you later...
[20:10:05] <steve_stallings> steve_stallings is now known as steves_logging
[20:10:12] <BigJohnT> thanks
[20:13:00] <jtr_> jtr_ is now known as jtr
[20:14:33] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/ (4 files): changed copyright to 2009 and fix minor layout in integrator
[20:24:29] <micges> good evening all
[20:24:46] <BigJohnT> good evening
[20:25:23] <micges> I'm translating 274err file and I don't understand this line:
[20:25:39] <micges> Queue is not empty after tool change
[20:25:49] <micges> what queue ?
[20:27:59] <micges> On my machine I had another one error : Queue is not empty after probing
[20:28:37] <BigJohnT> that is above me :(
[20:28:39] <micges> I don't know until now what was the problem
[20:32:11] <micges> For me both messages aren't tell what caoused problem
[20:32:49] <micges> just my thinking.. I'm back to translating
[20:41:47] <alex_joni> micges: probably motion queue
[20:42:07] <alex_joni> the interpreter should basicly stop sending motion commands if it gets to a toolchange
[20:42:34] <alex_joni> if it sends them (doesn't see the toolchange), then after the toolchange an error is signalled if the motion is not empty
[20:42:39] <alex_joni> (just my guessing..)
[20:47:20] <micges> I understand
[20:55:08] <CIA-1> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gcode/main.lyx: add full circle for g2/3
[21:05:52] <CIA-1> EMC: 03tissf 07TRUNK * 10emc2/docs/src/ (5 files): french translation update - add basic_hal_fr and halui_example_fr
[21:05:53] <CIA-1> EMC: 03tissf 07TRUNK * 10emc2/docs/src/common/ (Linux_FAQ_fr.lyx Stepper_Diagnostics_fr.lyx): french translation update - add basic_hal_fr and halui_example_fr
[21:05:54] <CIA-1> EMC: 03tissf 07TRUNK * 10emc2/docs/src/hal/ (basic_hal_fr.lyx halui_examples_fr.lyx components_fr.lyx): french translation update - add basic_hal_fr and halui_example_fr
[21:12:02] <skunkworks> KimK: his name was steve..
[21:12:20] <skunkworks> I don't think I have his contact info here though
[21:34:27] <KimK> skunkworks: OK, thanks.
[21:53:27] <Guest432> Guest432 is now known as skunkworks