#emc | Logs for 2008-02-15

Back
[00:11:15] <gezar> howdy guys
[00:12:06] <gezar> sorry about the abrupt leaving last night, I was busy working on math, and I need the emc box to be in windows so I rebooted it at console without thinking
[00:12:45] <fenn> i dont think anyone noticed
[00:14:00] <gezar> that makes me sad
[00:14:37] <gezar> but its understandable
[00:16:49] <jlmjvm> has anyone in here used the index homing?
[00:17:15] <jmkasunich> on the mazak, months ago
[00:17:37] <jmkasunich> (but I also wrote much of the code for that, so I know a bit or two about it)
[00:18:06] <eric_U> I finally ordered taps so I could install my prox switches for homing
[00:18:15] <jlmjvm> for some reason the x and y axis make a strange move after they home
[00:18:36] <eric_U> they leave home, as it were
[00:19:00] <jlmjvm> x moves about 2.5 and y about 2 inches,but those numbers arent in the ini file
[00:19:04] <jmkasunich> if [AXIS_n]HOME != [AXIS]HOME_OFFSET. there will be a move
[00:19:18] <jmkasunich> HOME_OFFSET says where the switch is in machine coords
[00:19:19] <jlmjvm> the numbers are zero
[00:19:26] <jmkasunich> HOME says where you want to be in machine coords
[00:19:56] <jlmjvm> agreed
[00:20:11] <jmkasunich> so both are zero in your file?
[00:20:28] <jlmjvm> yes
[00:20:28] <jmkasunich> ok, what encoder card are you using? or are you counting encoder pulses in software?
[00:20:52] <jlmjvm> usc with a gecko interface card
[00:21:02] <jlmjvm> z works properly
[00:21:17] <jmkasunich> all axes are configured the same?
[00:21:35] <jlmjvm> yep
[00:22:05] <jlmjvm> should i try to update to 2.2.3,i think its 2.1.7
[00:22:11] <jmkasunich> my knee jerk reaction is to say the software is identical for all axes, if there is a software issue it should affect all the same
[00:22:41] <jmkasunich> if one works and another doesn't, either the configs aren't truly the same, or there is a problem that is specific to one axis, in wiring, hardware, etc
[00:23:10] <jmkasunich> usc indexing had some bugs, I think we squashed a few back at the workshop in June 2007
[00:24:01] <jmkasunich> I'm pressed for time right now so I can't go digging, but if you go to http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/ you can drill down to emc2/src/hal/drivers/ppmc.c (I think thats the filename)
[00:24:11] <jmkasunich> look at the revision history and see what is different
[00:24:29] <jlmjvm> i can turn off the index and it homes just fine,turn it on and i get the strange move
[00:24:44] <jmkasunich> I decided to dig - http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/src/hal/drivers/hal_ppmc.c?graph=1
[00:25:24] <jmkasunich> there has been significant index work on both the 2.1 branch and the trunk (which is also the 2.2 branch) since the point where they split
[00:25:50] <jmkasunich> it would appear the jmelson put the same fixes on 2.1 and trunk (he committed to both on 6/14)
[00:26:01] <jmkasunich> but I really can't keep track of the status of his driver
[00:26:31] <jmkasunich> number 1 rule for index homing - the driver MUST comply with the canonical encoder spec as to what it does with index-enable
[00:26:56] <jmkasunich> if you can get a halscope tracing that shows the driver doing the right thing and EMC doing the wrong thing, its an EMC bug
[00:27:05] <jmkasunich> more likely, you'll see the driver doing the wrong thing
[00:27:19] <jmkasunich> now I really do have to go do something else
[00:27:20] <jlmjvm> i emailed jon my files and the dmesg to see if its rev3
[00:27:31] <jlmjvm> thanks john
[00:32:58] <eric_U> anyone else notice that when BigJohnT enters the room it goes quiet?
[00:33:27] <eric_U> that's because we're all afraid that he'll do to us what he did to GammaX
[00:33:37] <BigJohnT> I shot him
[00:33:44] <eric_U> lol
[00:33:55] <BigJohnT> wanna see the picture again?
[00:34:04] <eric_U> I never typed lol before in my life, I just did it in his memory
[00:35:07] <gezar> what did happen to gammax odd, cause I was thinking about him ?
[00:35:25] <BigJohnT> I was wondering too...
[00:35:51] <gezar> I wonder if he got his machine going, or if he has a massive pile of spent money
[00:36:49] <BigJohnT> you guys read the emc users digest tonight?
[00:37:05] <eric_U> I saw this comic and thought of gezar: http://www.phdcomics.com/comics/archive.php?comicid=974
[00:37:15] <eric_U> doomed, I say
[00:37:35] <eric_U> BigJohnT, couldn't say if I'm up to date, which one?
[00:37:48] <gezar> hahhahahaha
[00:38:36] <BigJohnT> the history of emc by Ray Henry
[00:38:45] <gezar> na the teaher asks "you understand this stuff?" I nod my head yes then I fail
[00:38:52] <gezar> ray's email was great
[00:38:57] <eric_U> I like the one where she staples the taco bell application to the papers
[00:39:30] <BigJohnT> yep
[00:39:43] <gezar> eric_U: you do realize im a journeyman machinist right?
[00:40:01] <eric_U> and you work at taco bell?
[00:40:08] <gezar> cant afford to work there
[00:40:10] <eric_U> are you still working?
[00:40:20] <gezar> no, im done with that stuff in this state
[00:40:56] <gezar> so sick and tired of working for nothing
[00:41:16] <gezar> the whole wake up and go make my boss rich scheme wasnt paying off
[00:41:57] <BigJohnT> I like that plan myself... but wait I'm the boss AND I have to do the work!
[00:42:14] <eric_U> tell me how to avoid it
[00:42:17] <BigJohnT> well at least we take turns being the boss
[00:42:19] <gezar> but to be honest, it wasnt about the work, it was about feeling like I was his properity
[00:42:32] <eric_U> that can be a problem
[00:43:00] <BigJohnT> it looks like we may have to start having employees and that scares me
[00:43:13] <gezar> its one thing to enjoy work, its another thing when your talked to as if your owned
[00:43:57] <gezar> its hard to explain, but it wasnt about being asked to do stuff, it was about being compared to specific others, it was really dificult
[00:44:42] <gezar> BigJohnT: whatever you do once you have workers, dont compare the smart ones to the guys who sweep the floor
[00:45:02] <BigJohnT> Ok, I sweep the floor...
[00:45:30] <eric_U> my biggest complaint about my bosses is that they treat me like I expect they would treat a homeless beggar back in India. That's why they can't keep anyone
[00:45:30] <BigJohnT> you mean like compare this one to that one cause they can do something different
[00:46:05] <gezar> well, would you compare your die maker to a welder?
[00:46:16] <BigJohnT> I used to be a supervisor and I tried very hard to support the people doing work for me
[00:46:20] <eric_U> welders play more dangerous pranks
[00:46:49] <BigJohnT> can a die maker weld as good as a welder and so on
[00:47:03] <gezar> im not talking about a "welder" with a tig torch who has honed their skills to a fine blade, im talking about a rough fabricating fitter
[00:47:26] <tomp2> toolmaker motto: my toolbox has a handle on it ( i can leave anytime )
[00:47:38] <jmkasunich> gezar: sounds like you want a boss who will treat some trades/people better than others
[00:47:38] <BigJohnT> IMO each job is important to the finished product
[00:47:41] <eric_U> I should start my own business, but I'd probably end up doing mostly government contract work
[00:47:49] <jmkasunich> I want a boss that treats everyone fairly
[00:48:17] <gezar> jmkasunich: no, it wasnt that either, it was more like this, "my guys do better work then you, and you dont do anything right"
[00:48:28] <eric_U> I want a boss that brings in money and can manage
[00:48:45] <gezar> speeking to the machinists in the shop, when referencing the fabricators
[00:49:22] <gezar> it was just really dishearting is all
[00:49:33] <BigJohnT> I had lots of guys and gals that couldn't do anything right, but will an open ear and a closed mouth you can learn a lot and give them the tools they need to do well
[00:49:39] <BigJohnT> if they want to...
[00:50:10] <eric_U> dealing with people is difficult
[00:50:22] <gezar> well, this guy felt he had to exert his position of dominance
[00:51:10] <gezar> do you guys know the type im talking about?
[00:51:11] <eric_U> I understand that to some degree
[00:51:18] <BigJohnT> the most difficult thing is communications... what is perfectly clear to me is perfectly clear to me... the challenge is getting someone else to understand your thoughts
[00:51:42] <eric_U> some people can't function if their boss treats them like a human
[00:51:47] <BigJohnT> the puffy chest know it alls....
[00:53:27] <gezar> this guy was the owners son, and had just been promoted to president of the company, and he wanted to make what was already clear, perfectly clear, in a way that instead of lifing people up, put them down
[00:53:38] <eric_U> that's always a mistake
[00:54:10] <gezar> I call it little man syndronme
[00:54:20] <eric_U> my boss is brutal to everyone almost all the time, and so people don't do much work
[00:54:29] <eric_U> they get the same treatment either way
[00:54:48] <maddash> SWPadnos, jepler, cradek: have you reached a consensus about the unhoming?
[00:55:41] <gezar> brutal is one thing, this guy just had it all, as an example, my direct super, went and got a new truck, so this guy, went and got a uber king ranch ford 250 i think, the same day he told us no bonuses
[00:56:40] <gezar> oh well
[00:56:42] <BigJohnT> real sensitive guy...
[00:56:55] <gezar> he just coudlnt stand to be outdone
[00:57:16] <BigJohnT> he needs to be alone...
[00:58:13] <gezar> who knows, after I left, 3 more people loaded their boxes up as well
[00:59:35] <gezar> and skilled labor can only take so much, just like teachers, they are supposed to be professional, they want to be treated as such, and paid as such, yet a good hunk of them are no where near the part, which makes it harder for the good ones to stay in the field
[01:01:51] <gezar> I say that because my wife, just recently got told that she will not be offered a contract for next year, because the school super and board last year, spent too much money, and the new super who is faced with fixing this has to fire teachers.....she had been trying to get on at this district for 7 years, and finally got on, but because she is new to the district, she wasnt given enough points to make the stay point. but other teachers who are worthless but have
[01:02:41] <gezar> end of rant, thanks for listening
[01:03:27] <BigJohnT> sounds like union to me...
[01:06:27] <gezar> any of you into baseball?
[01:07:04] <BigJohnT> fenn: you around?
[01:08:16] <BigJohnT> I use to go watch the Zephers
[01:08:25] <eric_U> so I finally order taps so I can install my prox switches for homing, and now I realize that the way covers will probably trigger them
[01:08:38] <eric_U> we have a local pro team, it's a hoot
[01:08:46] <BigJohnT> are they sheilded?
[01:08:56] <dmess> i like the rant.... teachers have it RUF....
[01:09:18] <fenn> BigJohnT: ya
[01:09:21] <dmess> kids these days... which 1 has the gun...
[01:09:40] <BigJohnT> where do I put the diff file?
[01:09:58] <eric_U> my prox switches only seem to trigger off of one end if that's what you mean by shielded
[01:10:21] <dmess> E.. oops
[01:10:34] <BigJohnT> eric shielded have the metal casing all the way to the end and the range is shorter
[01:10:43] <eric_U> yes, that's them
[01:10:52] <BigJohnT> and can be closer to metal from the sides
[01:11:04] <eric_U> they can go in metal, I checked that
[01:11:10] <eric_U> before I ordered taps
[01:11:31] <eric_U> but I'm not sure what the range is, hopefully short
[01:11:46] <BigJohnT> you can test it with a target
[01:11:54] <fenn> here's what i'm playing with: http://fennetic.net/pub/irc/cxf2g.png
[01:12:20] <BigJohnT> ok
[01:12:28] <eric_U> just don't try that at your prison workshop
[01:12:36] <eric_U> nvm
[01:12:52] <BigJohnT> if the man page is generated by "comp" whatever that is how do you add content to it?
[01:13:19] <fenn> BigJohnT: developers mailing list? comp files have the man page embedded at the top of the source code
[01:14:08] <tomp2> in the src code for the comp
[01:14:09] <BigJohnT> I was looking for and2 just to see but couldn't find it after installing CVS and downloading the trunk
[01:15:30] <tomp2> donuts? 'pack my box five dozen' lyrics?
[01:16:29] <BigJohnT> so the man page is generated from and2.comp?
[01:16:37] <tomp2> yep
[01:17:07] <fenn> tomp2: Pack my box with five dozen liquor jugs
[01:18:05] <tomp2> BigJohnT: >>only<< if author wrote the docs there, and the and2.comp file has squat
[01:18:18] <tomp2> fenn: cool
[01:18:19] <BigJohnT> If you want to have a bit more info you copy the weighted_sum.9 file format and place it in the docs/man/man9 folder
[01:18:45] <BigJohnT> ?
[01:19:34] <BigJohnT> or add content to the and2.comp file?
[01:20:10] <BigJohnT> did everyone get their sweetheart some flowers or candy?
[01:20:17] <tomp2> weighted_sum is not a 'comp' type 'comp' :) this means it's a C src file and others are a macro like language unfortunately called 'comp'...
[01:20:35] <BigJohnT> mine got a case of 20 gauge shotgun shells
[01:20:45] <BigJohnT> ok, I can see that this will be fun
[01:20:54] <tomp2> so the C src files need man pages written like man pages, but 'comp' script files are 'self-documenting'
[01:21:13] <fenn> the stuff between """ is the man page data
[01:21:28] <BigJohnT> ok, I see that
[01:21:34] <BigJohnT> like in python
[01:22:07] <fenn> and2 looks kinda strange to me as far as line breaks and semicolons go
[01:22:55] <BigJohnT> what does the \\f do?
[01:23:15] <fenn> \ escapes the \ so you end up with \f
[01:23:25] <BigJohnT> ok
[01:23:36] <fenn> it's the format command for man pages (groff-man i think)
[01:27:02] <BigJohnT> ok, I have to go... thanks
[01:29:17] <tomp2> where do cxf files come from?
[01:32:23] <fenn> qcad - i guess they made them from scratch?
[01:36:10] <tomp2> cadastral exchange format, made for maps, and python tools to fiddle the italian variant... http://initd.org/pub/software/icxf/
[01:36:25] <tomp2> internet is great
[01:36:32] <tomp2> a great time waster ;)
[01:38:35] <tomp2> http://en.wikipedia.org/wiki/Cadastre cadastral included
[01:40:06] <fenn> hmm.. well i coulda used that instead eh
[01:40:58] <fenn> oh well
[01:43:19] <fenn> i think its totally unrelated
[01:53:13] <tomp2> sorry i saw cxf2dxf and cxf2svf and fuiggered it might be of use
[01:53:22] <tomp2> cxf2svg
[01:53:41] <tomp2> http://www.initd.org/svn/initd/icxf/trunk/README
[02:13:08] <SkullWorks_PGAB> Tonites random thought/question
[02:13:59] <SkullWorks_PGAB> Is there an easy way to switch between inch/mm settings - or do you need two complete seperate profiles
[02:14:56] <fenn> g20/g21?
[02:15:20] <SkullWorks_PGAB> isn't that just a units check
[02:15:32] <fenn> i dont know what you're asking then
[02:15:32] <cradek> what do you mean settings
[02:16:04] <SkullWorks_PGAB> well sometimes I run all meteric programs
[02:16:13] <SkullWorks_PGAB> other times all inch
[02:16:21] <SkullWorks_PGAB> depends on the source
[02:16:24] <cradek> put g20 or g21 at the beginning of each program as appropriate
[02:17:01] <fenn> you should always have either g20 or g21, otherwise the program doesn't contain enough information to duplicate the part on a different machine
[02:17:17] <cradek> agreed
[02:17:21] <SkullWorks_PGAB> so it won't require different valuse for say soft limits or such?
[02:17:28] <fenn> and may crash your machine if you are using mm code on an inch machine
[02:17:30] <cradek> nope it will work properly
[02:18:10] <cradek> SkullWorks_PGAB: nope it will work properly
[02:18:22] <cradek> g20/g21 changes the units of the gcode, nothing else
[02:18:22] <SkullWorks_PGAB> will I need to clear out the tool table and reset the tool valuse?
[02:18:29] <cradek> nope
[02:18:37] <cradek> that would suck
[02:18:43] <SWPadnos> tool values are in machine units, aren't they?
[02:18:46] <cradek> yes
[02:18:58] <SkullWorks_PGAB> so they would have to be chNGED
[02:18:59] <cradek> you can change the units you see in the gui - that's yet another separate issue
[02:19:00] <fenn> hmm what happens if you use mixed mm/inch in your .ini file? >:)
[02:19:03] <SWPadnos> ok, so G20/G21 make no difference, but if you used separate mm/inch setups you would need different tables :)
[02:19:08] <cradek> SkullWorks_PGAB: no
[02:19:27] <cradek> SkullWorks_PGAB: machine units (the way your leadscrews, tools, and home position are measured) do NOT change
[02:20:08] <cradek> the units you see in the gui can change, so as to match the micrometer in your hand. that's on the view menu
[02:20:19] <cradek> the units of the gcode can change to match the part program, using g20/g21
[02:20:25] <cradek> these are all separate issues
[02:22:20] <SkullWorks_PGAB> OK cause on the commercial machines - it just bumps the decimal in the existing tool offsets from a ####.### to ###.####
[02:22:37] <cradek> so your whole tool table is screwed?
[02:22:43] <SkullWorks_PGAB> and the numbers are way off
[02:22:47] <cradek> well I think that's stupid
[02:22:50] <SkullWorks_PGAB> yep
[02:22:51] <cradek> emc is smarter
[02:22:58] <SkullWorks_PGAB> great
[02:23:08] <cradek> (emc used to do that - I fixed it)
[02:23:36] <cradek> your tools don't grow by a factor of 10 or 25.4 just because you drop your ruler and pick up a different one
[02:24:13] <cradek> so with one config you can make parts in either units just as easily
[02:24:53] <cradek> heck imagine an adapter with inch thread on one end and metric on the other - you could program it the natural way (switching units) with emc
[02:25:15] <SkullWorks_PGAB> so the only changes that might be handy are gui display - not core emc/hal
[02:27:04] <cradek> if your calipers are inch you might want the DRO in inch, even though the part program is metric - you can do that (if using AXIS)
[02:27:36] <cradek> brb
[02:28:08] <SkullWorks_PGAB> I work wknds sometimes at a shop that only works in inch - my normal weekly job is all ISO 9002 metric - I tend to work in whatever units are on the print I have.
[02:29:28] <tomp2> if theres a way to tell if the machine is in G20 or G21, a parametric program can scale by 1 or 25.4 ( gotta do that a lot on Heid macros run in inch or mm )
[02:30:46] <SWPadnos> hmmm. it should be possible (but I'm not saying easy) to make the mode and some scale factor available in some #parameters
[02:31:34] <gezar> from what i remember about g20/21 it switches all inch to metric or vise versa, even tool tables
[02:31:35] <SWPadnos> but that only tells you what mode the machine is in - it doesn't telll you what the code was supposed to be in (the code would need to know whether to multiply or divide by that factor)
[02:31:51] <cradek> I've talked before about a state stack. you would push, set whatever is appropriate for your macro, pop and return
[02:32:07] <cradek> macro/subroutine/whatever you call it
[02:32:18] <SWPadnos> yes, that would be very useful for reversing programs as well
[02:32:31] <cradek> if we want tool change programmed in gcode, for instance, I think we'd want something like that
[02:32:54] <SWPadnos> yes, you'd need a "local" offset coordinate system, for instance
[02:32:57] <cradek> but I'm a programmer and what makes sense to me may not make sense
[02:33:01] <SWPadnos> (or always run in absoute)
[02:33:04] <SWPadnos> heh
[02:33:16] <cradek> well I play a programmer on TV anyway
[02:33:26] <SWPadnos> I stayed in a Holiday Inn Express!
[02:33:29] <tomp2> i write all my macros in mm, and according to (soemthing similar to a # parm ) i read data tables ( kinda like tool tables ) and run program in whatever masure the user likes ( same macro )
[02:33:55] <cradek> bbl
[02:35:04] <tomp2> 'i play a programmer on tv' classic :-)
[02:52:19] <maddash> HEY! do we want the EMC_UNHOME NML msg or not?
[02:52:48] <SWPadnos> the answer is yes, but I'm not sure we ever decided on what it should do
[02:53:16] <maddash> er, set emcstatus->motion.axis[*].homed to 0?
[02:53:39] <SWPadnos> well, sure, that's the easy part. it's when it gets sent that's in question ;)
[02:54:06] <maddash> huh? I'm not understanding you
[02:54:09] <maddash> look,
[02:54:50] <maddash> when EMC_UNHOME is received by emctaskmain, emcUnhome(int axis) is called, which then sets the axis' home flag to zero, right?
[02:57:06] <maddash> what exactly is the hard part?
[02:57:49] <SWPadnos> what sends the message, and when?
[02:58:11] <SWPadnos> adding the message is a no-brainer, so go ahead and do it if you like :)
[02:58:27] <maddash> the gui?
[02:59:08] <maddash> parsing the [TRAJ]HOME_TAINTED={estop,never,etc.}
[03:01:40] <SWPadnos> well, you see that's the interesting part
[03:01:59] <SWPadnos> adding a message does nothing unless the message gets sent from somewhere
[03:02:35] <maddash> so send it from the gui
[03:03:23] <SWPadnos> so you want the user to click a button so the GUI removes the symbol that tells the axes are homed?
[03:03:49] <cradek> I want that ability
[03:04:22] <cradek> like HOME, UNHOME should take an axis number, or -1 for all
[03:04:28] <SWPadnos> sure, adding the message is a good first step, the automatic message sending on stop/off can be added later
[03:04:45] <cradek> yes
[03:05:43] <maddash> ...so, shall I release the coding beast from within?
[03:06:12] <cradek> haha
[03:06:15] <SWPadnos> sure, the timer has been running for hours ;)
[03:06:52] <cradek> maddash: make sure you are working with an up-to-date trunk checkout so your patch is useful to a committer
[03:09:13] <maddash> cradek: yep. I spent thirty whole minutes to download the ginormous head tar.gz from the cvs
[03:09:25] <SWPadnos> why not just update?
[03:09:39] <cradek> you should be using cvs, so you can use cvs diff
[03:09:43] <maddash> because `dpkg -l|grep cvs` returns nothing?
[03:09:53] <SWPadnos> what does sudo apt-get cvs do?
[03:10:07] <maddash> "bash: sudo: command not found"
[03:10:34] <SWPadnos> I'm getting suspicious that you won't be able to even compile-test this before submitting a patch ...
[03:10:40] <cradek> maddash: I'm sure you can figure out how to install and use cvs - that's necessary for you to contribute to emc
[03:10:52] <maddash> don't worry, I've got the HEAD code, so unless one of you guys commit in the next few minutes just to screw me over, my patch should be fine
[03:10:54] <SWPadnos> if you have a compiler then you should be able to install CVS
[03:11:11] <SWPadnos> (ie, you installed the comipler ...)
[03:11:14] <maddash> SWPadnos: --enable-simulator
[03:11:15] <SWPadnos> compiler too
[03:11:33] <maddash> bbl.
[03:11:43] <SWPadnos> so what - if you can compile, you have a ciompiler, which means the compiler was installed, which means CVS can be installed using the same techniwue
[03:11:51] <maddash> btw, do I submit informally via pastebin, or what?
[03:11:52] <SWPadnos> - spelling errors
[03:13:17] <SWPadnos> you could email to the dev list if it's not too big
[03:13:39] <SWPadnos> if it is too big, then putting it somewhere online and emailing the dev list with the URL should be fine
[03:16:59] <maddash> crap -- should I update the gui codebase as well?
[03:17:17] <maddash> i'm asking b/c of the appreciable number of guis in emc
[03:18:05] <SWPadnos> add the NML message as a separate patch
[03:18:17] <SWPadnos> then the GUIs and other modules can be modified to use it
[03:18:36] <SWPadnos> then the ini feature we discussed can be added in a separate patch
[03:35:47] <jlmjvm> i think Jon found my index homing problem
[03:35:51] <jlmjvm> http://pastebin.ca/904655
[03:39:08] <toastydeath> !features
[03:53:30] <fenn> arooo
[03:58:58] <fenn> the chinese characters in the unicode stick font look pretty ugly
[04:58:02] <toastydeath> ffff
[10:17:03] <micges> hi all
[10:18:15] <alex_joni> hi
[10:19:39] <micges> yesterday and today we installing emc software on old ELCEDE laser
[10:19:49] <micges> ist very fun job
[10:19:57] <micges> and interesting :)
[10:20:24] <micges> it XY table works great
[10:29:57] <archivist> interesting machine I saw yesterday at a show http://www.cybamantech.co.uk/?page=cybaman
[12:22:36] <cradek_> cradek_ is now known as cradek
[13:02:42] <jlmjvm> jepler:did you see the pastebin i did last night about the index homing?
[13:05:37] <jepler> jlmjvm: I am looking at it now
[13:05:47] <jlmjvm> k
[13:05:52] <jepler> I can tell the top lines are incorrect and the bottom lines may be correct
[13:05:57] <jepler> did you copy those from some example file?
[13:09:21] <jlmjvm> steve helped me with the net syntax in general late one night,im probably the "typo" culprit for the other 2 lines
[13:10:05] <jlmjvm> i had 1 motor he and he got it to work
[13:10:43] <jlmjvm> looks like a no brainer to get this feature working now
[13:12:09] <jlmjvm> dont want to start the victory dance quite yet,but it looks like that was the problem
[13:13:27] <jlmjvm> will be going to the school today to give it a shot
[13:15:43] <alex_joni> wheee http://www.chrishodgephotos.co.uk/funpics/pixww/fun07/07004.jpg
[13:16:01] <alex_joni> http://www.chrishodgephotos.co.uk/funpics/pixww/fun07/07005.jpg <- scary
[13:19:49] <Vq^> neat!
[13:20:30] <jepler> jlmjvm: yes I bet it was -- if X worked and the others didn't, those lines sure explain it
[13:21:36] <jlmjvm> actually only z worked correctly,that kinda confused me,but its obvious they need to be changed
[13:25:12] <jlmjvm> gonna make the changes and report to behavior back to jon and you guys
[13:25:26] <jlmjvm> the behavior
[13:25:39] <jlmjvm> need more coffee
[13:38:53] <archivist> alex_joni, I want one!
[13:39:16] <archivist> live cnc gear cutting on webcam
[13:40:23] <archivist> www.archivist.info/cnc/cnccam.html
[13:47:33] <alex_joni> archivist: whee.. nice
[13:47:51] <alex_joni> webcam speed: 1tooth/update
[13:48:10] <archivist> no webcam on 6 sec update
[13:48:14] <archivist> 5 sec
[13:49:14] <archivist> hmm torch dying
[13:49:26] <tomp> archivist: great stuff
[13:49:34] <archivist> and finished cut /me measures dia
[13:50:10] <tomp> led ring around lens, what i see commercial inspection cameras using
[13:50:26] <archivist> slightly under /near enough
[13:50:48] <archivist> tomp yes went too exhibition yesterday to get ideas
[13:51:39] <tomp> and the cybaman is great , posted it a year or 3 ago so the devs could see kins, found it with searches for paraglides
[13:52:06] <alex_joni> cybaman?
[13:52:10] <tomp> graham stabler turned me onto it
[13:52:28] <tomp> http://www.cybamantech.co.uk/?page=cybaman
[13:53:22] <archivist> was interesting watching it run
[13:53:55] <archivist> the salesman denied any tool spring!!
[13:53:58] <alex_joni> ah ok.. remember it from a couple years ago
[13:54:01] <tomp> didnt see them at imts, what show did you goto? emo?
[13:54:29] <archivist> machine building at NEC birmingham
[13:54:49] <archivist> £75K
[13:56:31] <tomp> venditori ( pirata ! )
[13:57:41] <tomp> i spent most of my machine tool life making machines do what salesmen said the machines could do ( lies! )
[13:58:31] <alex_joni> tomp: that's a nice quote
[13:58:45] <archivist> hehe
[13:59:16] <archivist> I could see a mark left where it pulled out
[14:04:13] <alex_joni> would someone in here happen to have a book called "Wireless intelligent networking" by Gerald Christensen ?
[14:13:12] <maddash> ok, boys and girls
[14:13:29] <maddash> I bring you ... the all-new, EMC_AXIS_UNHOME capability!
[14:14:49] <maddash> ... as soon as I can figure out how to subscribe to the -dev mailing list
[14:21:06] <alex_joni> http://photos1.blogger.com/blogger/7058/865/1600/citizen.jpg
[14:22:50] <archivist> what is it
[14:23:28] <ALS> what was rayh refering to his last web id got sold to a mega corp? in his "AWESOME" email
[14:24:57] <alex_joni> ALS: the company where he had his former email
[14:25:03] <alex_joni> archivist: a small robot
[14:25:27] <pelligiu> pelligiu is now known as Pipps
[14:25:30] <Pipps> Salve a tutti
[14:25:57] <Pipps> can you help me please?
[14:26:13] <alex_joni> we can try
[14:26:17] <tomp> hello, please ask
[14:26:30] <Pipps> i have downlad emc
[14:26:36] <Pipps> i use opensuse 10.3
[14:26:55] <Pipps> i have used tar, ./configure end make
[14:27:09] <Pipps> i start program
[14:27:40] <Pipps> but when i select the machine
[14:27:58] <Pipps> i have an error
[14:28:12] <Pipps> hal_lib.so
[14:28:17] <Pipps> No such file or directory
[14:28:45] <alex_joni> Pipps: did you use ./configure --enable-run-in-place ?
[14:28:53] <Pipps> yes
[14:29:01] <alex_joni> how do you run emc?
[14:29:06] <alex_joni> scripts/emc ?
[14:29:11] <Pipps> yes
[14:29:21] <alex_joni> maybe you have an installed emc around, and are running the wrong file?
[14:29:29] <alex_joni> did you configure for simulator?
[14:29:55] <Pipps> no are the first installation
[14:30:35] <Pipps> rtlib are empty..why?
[14:30:53] <Pipps> ahh excuse me for my bad english
[14:30:58] <Pipps> i am italian
[14:31:23] <tomp> maybe you could try the live cd first, it will not change your hard drive at all, then you can be sure it will work... http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#Installing_from_the_EMC2_Ubuntu_Live_CD_Preferred ( excuse my bad italian ;)
[14:31:57] <archivist> it has the needed realtime kernel
[14:32:35] <tomp> archivist: it = opensuse 10.3?
[14:32:57] <archivist> no I mead the live cd
[14:33:01] <archivist> mean
[14:33:03] <tomp> k
[14:33:27] <archivist> Im guessing thats missing/not installed
[14:34:33] <tomp> Pipps: could you try the live cd? would that help you?
[14:36:11] <BigJohnT> the live CD is the easy way to try EMC
[14:36:26] <Pipps> yes but i would install emc
[14:36:33] <Pipps> not play on live cd
[14:36:58] <tomp> you can, from the live cd, after you know it works
[14:37:23] <tomp> but, it would install ubuntu
[14:37:52] <tomp> in some empty partition OR over suse, as you wish
[14:38:13] <alex_joni> Pipps: you need to compile a realtime kernel in order for emc2 to work
[14:39:18] <Pipps> ah ok
[14:39:32] <Pipps> thx for help
[14:39:52] <tomp> Pipps: compiling a realtime kernel is not simple for most users, using the live cd is easy, installing to hard drive later is easy
[14:40:15] <tomp> np, you may not find compiling kernel difficult, most here do
[14:40:32] <tomp> i do ;)
[14:41:07] <Pipps> i download live cd
[14:41:08] <Pipps> thx
[14:41:09] <Pipps> bye
[14:41:34] <ALS> alex_joni: maybe gezer could use that small robot to load his sugar cubes on his palm mill :)
[14:41:43] <archivist> hehe
[14:42:04] <archivist> here have an atom of sugar
[14:46:39] <tomp> http://www.pinktentacle.com/2006/03/size-matters-for-citizens-eco-be/
[14:47:22] <tomp> driven by a small wristwatch motor
[14:49:27] <maddash> jepler: cradek: I emailed emc-developers@lists.sourceforge.net, hope that was the right thing to do
[14:58:24] <BigJohnT> Oh boy Solidworks is going to give me a 5% discount if I buy the subscription service
[15:00:44] <archivist> become a training company try and get educational discount
[15:26:42] <alex_joni> BigJohnT: I think there's a 500 off deal on alibre expert atm :)
[15:27:27] <BigJohnT> it won't import a Pro-E file and my biggest customer uses Pro-E
[15:27:43] <BigJohnT> :-(
[15:28:18] <BigJohnT> Pro-E 5k, Solidworks 4k, ...
[15:33:39] <alex_joni> BigJohnT: but it imports step ;)
[15:33:54] <alex_joni> and I think there are Pro-E importers out there
[15:34:10] <BigJohnT> Pro-E to step converters?
[15:35:08] <alex_joni> yeah
[15:35:16] <alex_joni> I know you lose the parametric stuff
[15:35:30] <alex_joni> so it's not a real solution if you need to change anything your customer sends you
[15:38:41] <BigJohnT> not normally but sometimes. Mostly I put their parts into my machine to make sure things fit well...
[15:40:30] <alex_joni> * alex_joni nods
[16:15:41] <Roguish> BigJohnT: you need to import solid models? STEP is probably the best format. most every program exports/imports STEP.
[16:16:17] <Roguish> ask your customer to export the
[16:16:27] <Roguish> STEP files for you.
[16:16:42] <Roguish> i do it all the time.
[16:39:33] <alex_joni> Roguish: yeah, but then you can't change features & such
[17:59:23] <fretless85> hey guys
[18:00:39] <fenn> good day to ya
[18:00:54] <fretless85> thanks fenn!
[18:01:29] <fretless85> how is it going?
[18:03:22] <fenn> progress has been stifled yet again
[18:04:27] <fretless85> ?
[18:05:19] <skunkworks> I hate that
[18:05:31] <skunkworks> usually caused by life in general
[18:05:46] <fenn> i'm sitting here at the root of my dependency tree watching the clouds go by
[18:07:12] <fretless85> http://mitglied.lycos.de/fretless85/00002.jpg
[18:07:20] <fretless85> thats all i got to say...
[18:07:54] <fretless85> now i got time...
[18:08:04] <fretless85> but cant do shit..
[18:09:20] <maddash> you 'got time' precisely because you 'cant do shit'
[18:09:28] <ALS> how'd that happen?
[18:09:45] <maddash> nice hands, btw. how old are you? 19-24?
[18:10:18] <fretless85> maddash, 23
[18:10:33] <fretless85> ALS, i put 7700lbs on it
[18:10:54] <ALS> I got that part but how?
[18:11:00] <skunkworks> did it look like a cartoon hand?
[18:11:04] <maddash> ALS: bencpressing
[18:11:17] <maddash> :P
[18:11:17] <fretless85> no
[18:11:42] <fretless85> skunkworks, ya actually it was a bit flat...
[18:11:59] <fretless85> how it happend?
[18:12:05] <ALS> ya
[18:12:27] <fretless85> let me explain ;)
[18:13:08] <fretless85> if you got time *g*
[18:13:36] <fretless85> only one hand sucks,,,
[18:13:45] <ALS> not as much as you but don't let that stop ya
[18:13:59] <fretless85> lol
[18:14:25] <maddash> does that mean you can't masturbate?
[18:14:36] <maddash> * maddash ducks
[18:14:44] <fretless85> yea thats the worst part of it
[18:14:50] <fretless85> im left handed
[18:15:01] <ALS> maddash you give a hand
[18:15:08] <fretless85> haha
[18:15:45] <fretless85> ive had to clamp a big hook something like that http://www.stahlhammer.de/images/lasthaken/lamellenhaken.gif
[18:16:43] <fretless85> lifted it with a chain
[18:17:35] <fretless85> put it on the table on parallel ledges (right word?!)
[18:18:06] <fretless85> but i had to put a 0.5mm sheet under one side
[18:18:35] <fretless85> because it wasnt parallel in Z/X
[18:18:59] <fretless85> yea lifted it about 2" on one side
[18:19:12] <fretless85> put the sheet under it
[18:19:53] <fretless85> and in that moment the chain kind of slipped in a chain link
[18:20:15] <fretless85> the crane bumped
[18:20:39] <fretless85> and the hook gives me a "slap"
[18:21:10] <fretless85> ive screamed "FFFFFFFFFFFFUUUUUUUUUUUUUUUUUUCCCCCCCCCCKKKKKKKKK"
[18:21:23] <fretless85> pulled of my gloves
[18:21:50] <fretless85> and looked at my crocked flat smashed mushed fingers
[18:22:08] <maddash> that must have been traumatizing
[18:22:25] <fretless85> no
[18:22:40] <ALS> broken bones?
[18:22:57] <fretless85> yea brocken+mushed
[18:23:03] <fretless85> broken
[18:23:58] <ALS> I was gonna send ya back to work but you better sit this one out
[18:25:06] <fretless85> ive took some photos at the change of the bandages, interested? *g*
[18:25:38] <fretless85> i cant even grab the feedrate potentiometer
[18:26:10] <ALS> let um fly but with a warning like don't look
[18:26:17] <fretless85> but i can program with the right hand :P
[18:28:30] <fretless85> http://mitglied.lycos.de/fretless85/R0011340.JPG http://mitglied.lycos.de/fretless85/R0011341.JPG <<<<----do you a favor and DONT LOOK! moral of the story...keep you finger out of the way...
[18:29:03] <fretless85> it could been worse...
[18:29:37] <fretless85> i had got massif luck!
[18:30:28] <ALS> ya you still got em fingers that is
[18:30:57] <fretless85> you got it...
[18:31:33] <fretless85> i got no fking idea why it slipped...
[18:32:00] <fretless85> ive done this 1000 times...
[18:33:16] <fretless85> the problem is, 7700lbs is small for me and you get carelessly...so easily
[18:33:23] <ALS> you will have rethink the rigging you use
[18:33:33] <jlmjvm> jepler:that fixed the index homing,works fine now
[18:34:26] <fretless85> nothing bad with the rigging
[18:34:40] <fretless85> ive got no other way to lift it
[18:35:00] <ALS> must be something
[18:35:12] <fretless85> next time ill keep my fingers out of the way
[18:35:28] <ALS> unless the crane failed
[18:35:29] <fretless85> than tell me an other way?
[18:35:48] <ALS> cable or chain?
[18:36:01] <fretless85> i used a chain...
[18:36:16] <ALS> i mean the hoist
[18:36:48] <fretless85> cable
[18:37:32] <fretless85> 70500lbs crane
[18:37:40] <ALS> could the cable got over itself and sliped on the drum?
[18:37:55] <fretless85> no way!
[18:38:37] <fretless85> http://mitglied.lycos.de/fretless85/test/New%20Image7.JPG
[18:38:53] <fretless85> the one in the background
[18:39:34] <fretless85> (no thats not the part)
[19:01:04] <maddash> MUHAHAHA, I just won an ebay auction by outbidding in the last 7 seconds
[19:02:45] <fretless85> i allways do it like that...
[19:02:45] <skunkworks> sniper :)
[19:02:45] <maddash> * maddash feels *sniff* baaadasss...yeaahh
[19:03:09] <fretless85> can you do it on another way?!
[19:03:21] <maddash> fretless85: ?
[19:03:44] <fenn> smelly bad-ass
[19:04:02] <fretless85> i always bid in the last few seconds on ebay
[19:04:23] <fretless85> im not a price pusher ;)
[19:04:29] <maddash> yeah, but I a bash script to help me do it
[19:04:34] <fenn> i always do buy it now because otherwise either i forget about it for three weeks or i lose
[19:04:34] <maddash> I use*
[19:05:05] <fretless85> oh okay ;)
[19:05:31] <fretless85> shame on you :P
[19:05:32] <maddash> brb, gotta slow my heart down
[19:06:10] <fretless85> lol
[19:08:07] <maddash> well, back to windows land
[19:09:08] <archivist> I like putting a price to stop the snipers, then they dont get time to try again
[19:09:32] <fretless85> ;)
[19:10:13] <archivist> I got angle plates this week
[19:11:34] <fretless85> for?
[19:11:52] <fenn> book-ends
[19:12:04] <fenn> he collects books, you know
[19:12:05] <archivist> Ive got to raid the loose lenses at home to make a telecentric lens for my cnc machine vision
[19:12:27] <archivist> hehe a little ott for book ends
[19:12:43] <fretless85> fenn, lol
[19:13:19] <archivist> fretless85, its true /me does have a "few" books
[19:14:20] <fretless85> a few?
[19:14:50] <ALS> more than two
[19:15:28] <fretless85> really?
[19:16:13] <toastydeath> oh yeah i have three books
[19:16:15] <toastydeath> beat that
[19:16:45] <archivist> 3200 catalogued so far
[19:17:01] <fretless85> nice ;)
[19:17:05] <archivist> tis a bad habit
[19:17:26] <archivist> but nice when I want to know something
[19:17:47] <fretless85> try www.google.com
[19:18:10] <archivist> no intarweb at home
[19:18:24] <fretless85> just a bad joke :)
[19:20:22] <fretless85> everyone got a "bad habit"
[19:24:28] <archivist> yup
[19:25:39] <fretless85> for my self, i like to smash my fingers with big workpieces :D (just a joke ;) )
[19:27:44] <archivist> I had a broken leg at work 1976
[19:28:50] <fretless85> 2 weeks before that i fell off the machine lol...
[19:29:12] <archivist> one you will learn
[19:29:31] <archivist> one day you will learn
[19:29:49] <fretless85> yea i only learn the hard way
[19:31:42] <fretless85> http://a579.ac-images.myspacecdn.com/images01/12/l_b0690e681c334470583b19fdcc8d0582.jpg
[19:38:45] <toastydeath> archivist: i assume you shop used bookstores
[19:38:55] <toastydeath> do you find used engineering books frequently?
[19:38:57] <toastydeath> or just fiction
[19:39:48] <archivist> nil fiction
[19:39:59] <toastydeath> ..?
[19:40:55] <archivist> secondhand book shops and a couple of other collections bought/scrounged
[19:41:25] <archivist> www.archivist.info/collection
[19:42:15] <toastydeath> you have no fiction, just engineering/nonfiction books?
[19:42:43] <toastydeath> that is the coolest thing i have ever heard
[19:42:47] <archivist> yes basicly
[19:43:31] <archivist> I did a bit of buying selling (thats where the fiction comes from)
[19:46:37] <jlmjvm> where is the g code list for emc?
[19:47:16] <fenn> http://linuxcnc.org/docs/html/gcode.html
[19:47:26] <jlmjvm> thanks,just found it
[19:49:21] <toastydeath> wow that is a nice page
[19:50:07] <ALS> hence the name archivist!
[19:51:03] <archivist_emc> er yes
[19:51:17] <archivist_emc> I also volunteer in archives
[19:52:28] <fenn> do they have public libraries in britain?
[19:53:22] <archivist_emc> yes
[19:53:42] <lerman> That's where they keep all the scrolls and other parchment items. :-)
[19:54:52] <fenn> i just mean, a place where you can find a wide variety of books, to borrow for $small
[19:55:59] <fretless85> you mean small£
[19:56:13] <lerman> archives comes from ar (the country code for Argentina) and chives (those onion like shoots).
[19:56:17] <fenn> yes but i've been traumatized by perl
[19:56:46] <acemi> http://www.linuxcnc.org/docs/html/config_stepconf.html in this page at part "1.6 Home and Limit Switch wiring options" it says that "when one switch is actuated, the value seen on INPUT goes from logic HIGH to LOW" but I think the opposite is true according the figure. somebody can check this?
[19:59:33] <lerman> fretless85: I seem to have trouble sending private messages.
[20:00:30] <fretless85> lerman, seams to...
[20:01:38] <fretless85> alex_joni talked with you about that heidenhain interpreter right?
[20:01:38] <acemi> fretless85: you are not a registeed user
[20:02:06] <acemi> only registered user can send private mesg
[20:02:08] <lerman> Correct. He did "talk" to me.
[20:02:15] <fretless85> i am but not logged in brb
[20:03:15] <fretless85> now i am?
[20:03:25] <acemi> no
[20:03:29] <lerman> I concluded that it is reasonable to do if and only if the canonical model for heidenhein is pretty close to that for rs274ngc.
[20:03:31] <acemi> /whois fretless85
[20:04:04] <fretless85> i am ;)
[20:04:27] <fenn> he's an impostor!
[20:05:00] <fenn> ah it's lerman who's not registered
[20:06:03] <acemi> to accept message from unregistered user, /msg nickserv set unfiltered on
[20:06:03] <lerman> This computer stuff is just too much trouble. I'm going back to paper and pencil. Correction: parchment and ink.
[20:06:07] <fretless85> canonical
[20:06:09] <fretless85> ?
[20:06:18] <fretless85> let me get my dictionary
[20:06:19] <fenn> i really prefer charcoal and wood
[20:06:21] <fretless85> ^^
[20:07:14] <fretless85> oh ok
[20:07:16] <lerman> Yes. rs274ngc (gcode) has an underlying model of the machine that provides a context for the commands.
[20:07:16] <fenn> fretless85: you wont find it in a dictionary. in this sense 'canonical' is the minimum set of simplest operations
[20:07:53] <fretless85> lerman, right
[20:07:54] <lerman> That model provides for three orthogonal (right handed) axes and three rotary axes.
[20:08:04] <fretless85> yea
[20:08:12] <fretless85> xyz abc
[20:08:15] <lerman> It provides for units of either mm or inch.
[20:08:25] <fretless85> and parallel uvw
[20:08:36] <lerman> There is a chapter in the spec concerning that model.
[20:09:01] <fretless85> the heidenhain can take mm or inch or "klartext" or gcode
[20:09:07] <fretless85> you can chose
[20:09:18] <lerman> Unless Hcode follows te same model as Gcode, it would not be a "simple" matter of redoing just the interpreter.
[20:09:52] <fretless85> i got you...
[20:10:23] <lerman> The canonical model provides for a fixed number of coordinate systems (nine, ten, I forget). If Hcode requires thirty coordinate systems that would be a problem.
[20:11:00] <fretless85> 9 on hdh
[20:11:11] <fretless85> xyz abc uvw
[20:11:17] <lerman> Assuming that the gcode canonical model is compatible, then it might be a reasonable interpreter to write.
[20:11:42] <fretless85> would be nice...got a bunch of cycles
[20:11:55] <fretless85> and i can do a lot more
[20:11:59] <lerman> Then it would be "simply" a model for someone with the development skill to "care enough to give his very best". :-)
[20:12:54] <lerman> He developer would need to know a lot about hdh. And about the guts of interpreters in general. And about the canonical model.
[20:13:21] <lerman> I'm good for about one of those three. (I've written some interpreters.)
[20:13:50] <fretless85> i know the hdh...i really do
[20:14:15] <fretless85> but i got no clue about the emc interpreter
[20:14:19] <lerman> How are your C programming skills?
[20:14:31] <fenn> lerman what do you think about parser-generators like antlr or spirit++?
[20:14:32] <fretless85> im an nc/plc guy
[20:15:25] <fretless85> C programing? ive read a book about it a few years ago :D
[20:15:55] <lerman> My parser generator skill are a generation before those. My personal view is that parser-generators seem to fall down when it comes to giving decent error messages and error handling in general.
[20:16:03] <fretless85> like i said im an nc/plc "freak"
[20:16:08] <lerman> But that was in the days of YACC.
[20:16:34] <lerman> I am a believer in automatic lexers, though (FLEX).
[20:17:06] <lerman> In the case of gcode, there isn't enough language to really matter.
[20:17:33] <lerman> Although, you might have noticed that I took the (very) lazy man's way out when it came to adding o-words.
[20:17:41] <toastydeath> i wish there was a modular machine controller
[20:18:06] <lerman> I thought that emc is a modular machine controller :-)
[20:18:13] <lerman> Can you be more specific?
[20:18:33] <toastydeath> like, have the machine control itself be one layer - the hard X/Y/Z, feed, speed, feedback loop
[20:18:35] <fenn> emc is not modular enough.. all that icky c++ code
[20:18:52] <toastydeath> then have the functions, like G0, g2, all the canned cycles
[20:18:57] <toastydeath> be a set of modules that sits on top of it
[20:19:01] <lerman> But HAL is modular.
[20:19:04] <toastydeath> and then have the interpreter a third layer
[20:19:29] <toastydeath> so if you wanted to run Hcode instead of Gcode you'd redefine the relationship between what the interpreter sees and what function module it used
[20:19:36] <lerman> Well, the interpreter DOES sit on top of canon.
[20:19:53] <toastydeath> but it's not a configuration file.
[20:20:18] <lerman> toastydeath: assuming that the canonical models are the same, you can do that.
[20:20:21] <toastydeath> there's no doodad saying "G0 = g0-default-rapid.py"
[20:20:37] <toastydeath> or what have you.
[20:20:47] <fenn> toastydeath: there is, but it's in icky c++
[20:20:56] <lerman> Interestingly, my plan is to add the ability to define canned cycles in pretty much that way.
[20:20:59] <fenn> and not.. modular
[20:21:28] <toastydeath> because if it was modular someone with a lot less knowledge could, for instance, write a lathe canned cycle for g70-g74 roughing/finishing
[20:21:56] <toastydeath> and drop it in the appropriate directory, and then change the g70-74 defintions
[20:22:17] <fenn> lerman while you have your hands covered in emc guts could you add an include function to g-code? and perhaps an 'execute script and insert the output here' special comment
[20:22:21] <lerman> Dammit. I don't have the energy to do that now. It really is only a few days work. But first I would have to get a virgin machine wit emc on it.
[20:22:41] <toastydeath> hey man i'm not asking for any features, i'm just saying it would be cool
[20:22:59] <toastydeath> the demand for stuff like that is probably low
[20:23:08] <fretless85> an hdh interpreter would be cool...
[20:23:29] <lerman> Well, there already is (sort of). If a function is called that is not in the current file, the interpreter goes out and finds one in another file and starts interpreting that.
[20:24:08] <toastydeath> that doesn't really do what i'm talking about though
[20:24:10] <lerman> hdh is only cool for those of us (well those of you) who are familiar with it.
[20:24:19] <toastydeath> because say I wrote a highspeed machining thing
[20:24:37] <toastydeath> i could replace the stock G1 command module with my own, or add another gcode
[20:24:45] <fretless85> lerman, its easyer to lern than gcode...
[20:25:11] <fretless85> and i will spread my knowledge
[20:25:34] <fretless85> learn
[20:25:42] <lerman> Spanish is easier to learn than English. But I already know English.
[20:26:11] <fenn> esperanto is easier to learn than spanish, but i already know spanish..
[20:26:18] <fretless85> a big plus is the lots of canned cycles
[20:26:33] <toastydeath> if you want canned cycles look at conversational programming
[20:26:45] <toastydeath> anybody's implementation of it.
[20:26:54] <toastydeath> "oh, you want a pocket? okay."
[20:26:59] <lerman> Or wait till lerman adds the ability to have user defined canned cycles.
[20:27:17] <lerman> Universal pocketing is a hard problem.
[20:27:26] <fenn> g-code is crap
[20:27:27] <toastydeath> it is, but the conversational guys do it
[20:27:38] <toastydeath> with islands and everything
[20:27:46] <fenn> it's always going to be hard to write part programs in g-code
[20:28:02] <fretless85> ;)
[20:28:08] <fenn> this is why we want a different interpreter
[20:28:20] <toastydeath> that's why i like the idea of being modular
[20:28:41] <fenn> yes, because g-code is still very useful as a data exchange format
[20:29:01] <fretless85> right
[20:29:01] <lerman> toastydeath: if you will write that universal pocketing program, I'll write and interface to let the interpreter invoke it.
[20:29:13] <toastydeath> i didn't say _I_ wrote it
[20:29:19] <fretless85> but its from the stone age of the cnc
[20:29:25] <toastydeath> i just sait the conversational controls have it
[20:29:27] <toastydeath> *said
[20:29:38] <fenn> hah. lerman i'm working on that universal pocketing program (slowly)
[20:29:51] <toastydeath> and even if i did have the interest, you'd get into holy wars about what toolpath to use
[20:30:02] <fretless85> and i got finished stuff on the hdh
[20:30:03] <fenn> http://fennetic.net/pub/irc/pocket.png
[20:30:10] <lerman> Does it have built in cutter radius compensation?
[20:30:32] <fenn> the idea is to cut a spiral pocket eventually, or at least a parallel offset pocket
[20:30:48] <lerman> I just looked at pocket.py LOOKING GOOD .
[20:30:48] <fenn> right now it just moves the cutter over a little bit
[20:31:01] <fenn> eh, yeah there's a lot of code
[20:31:37] <lerman> That program I've written in a previous life. It's called area fill (in computer graphics).
[20:32:01] <fenn> sure, raster scan is easy
[20:32:18] <lerman> Let's see how it does filling a bow tie shape.
[20:32:50] <fenn> the drawing is just cairo, you can select different fill rules
[20:33:05] <fenn> i was trying to do some sort of 'gouge detection' with the red fill
[20:33:56] <lerman> I'll be back in a few...
[20:43:57] <ALS> hey it was good to here from paul c. again
[20:44:32] <alex_joni> ALS: here?
[20:44:44] <ALS> allway a pleasure arrcive
[20:45:42] <ALS> I missed it the other day 1/13/08
[20:45:51] <alex_joni> ah
[20:46:10] <alex_joni> oh, he sent a couple of nice things lately
[20:46:22] <ALS> he's a real mach guy now
[20:47:03] <fenn> no he has his own fork to play with
[20:47:14] <toastydeath> i have forks!
[20:47:19] <toastydeath> i was told not to play with them though =(
[20:47:22] <alex_joni> fenn: you'd think that after 2 years
[20:47:36] <alex_joni> toastydeath: you were told to go stick them into your ..
[20:47:43] <toastydeath> ='(
[20:47:43] <alex_joni> er.. wait, that wasn't you :P
[20:47:44] <fenn> mouth
[20:47:50] <toastydeath> hahah.
[20:47:54] <fretless85> lol
[20:48:02] <alex_joni> sorry.. couldn't resit
[20:48:05] <alex_joni> resist even
[20:48:16] <toastydeath> you couldn't resit after you put a fork up your..
[20:48:29] <alex_joni> well.. I could one time
[20:48:34] <alex_joni> after that I never tried again
[20:49:22] <toastydeath> i'm not sure how to respond to that!
[20:49:32] <alex_joni> you'd better not
[20:49:47] <ALS> rayh said it all no f$*%way would I go back to emc1
[20:57:22] <alex_joni> well.. you haven't seen tuxcnc yet
[21:05:38] <ALS> have you? seen tux
[21:06:16] <alex_joni> nope.. no-one has :)
[21:06:32] <archivist> EMC2 with wings?
[21:06:54] <ALS> cumber bun
[21:07:44] <alex_joni> archivist: not related to emc2 ;)
[21:08:11] <alex_joni> bbl
[21:36:22] <Vq^> fenn: have you used spirit?
[21:36:33] <fenn> no, only looked at it a bit
[21:38:43] <Vq^> i was a bit curious, i have used parsec quite a bit (the library which inspired spirit)
[21:41:56] <Vq^> i have no idea how usefull it is compared to parsec thought
[23:44:56] <BigJohnT> anyone know how to run the truetype-tracer that comes with EMC?
[23:45:16] <BigJohnT> I see the executable in usr/bin/
[23:54:58] <eric_U> whazzit called?
[23:54:58] <alex_joni> BigJohnT: open a terminal
[23:55:13] <alex_joni> <executable> "string" > file.ngc
[23:57:02] <eric_U> that's cool
[23:58:28] <eric_U> that's very cool, I now have ngc to cut "joemama"
[23:59:16] <BigJohnT> ok