#emc | Logs for 2008-07-12

[02:21:43] <john_f> ?
[02:22:14] <archivist> ? what
[02:51:27] <Ito-Brazil> Hello ALL: Has EMC2 some milling time ?
[02:52:10] <archivist> its in hard use by many
[02:53:19] <Ito-Brazil> sorry i didn´t understood
[02:55:04] <Ito-Brazil> i need just emc says how time elapsed to do a job
[02:56:10] <seb_kuzminsky> you want emc to tell you how long it would take to run a gcode program?
[03:04:08] <cradek> Ito-Brazil: file/properties
[03:04:17] <cradek> in AXIS gui only
[03:08:32] <Ito-Brazil> hm, is that dificult to put it on MINI ?
[03:09:07] <cradek> it would not be simple because MINI does not know how to parse and evaluate the gcode program.
[03:09:32] <Ito-Brazil> is estimated or real time elapsed ? I mead, EMC calculates before or say the time after finishes ?
[03:09:47] <cradek> AXIS calculates before
[03:10:26] <Ito-Brazil> hm.. I´m sayin a simply clock that starts on <RUN> button and stops qhen g-code finishes
[03:10:30] <Ito-Brazil> just like that
[03:10:43] <cradek> then no, none of the guis have that
[03:11:30] <Ito-Brazil> cradek how can i do it? i mean, simple way
[03:11:46] <Ito-Brazil> may i program on PY, or something ?
[03:11:50] <cradek> simplest way would be to use your wristwatch
[03:11:54] <jmkasunich2_> heh
[03:11:59] <jmkasunich2_> I've wanted the same feature
[03:12:10] <cradek> really
[03:12:11] <jmkasunich2_> I either forget to note the start time, or forget to note the end time
[03:12:35] <jmkasunich2_> I missed the start of the conversation, is the request for a clock that simply times the execution of the program?
[03:12:39] <Ito-Brazil> it´s simple but helps a lot to get the entire elapsed time when you mill lots of parts and need to say your price
[03:12:58] <cradek> yes that's my understanding
[03:13:43] <jmkasunich2_> when I had 18 parts to do, I tweaked the program after each run for the first half-dozen or so - reducing air cut time, increasing feeds, etc - trying to speed it up
[03:13:49] <jmkasunich2_> the ability to measure that result would be nice
[03:13:53] <Ito-Brazil> jmkasunich2: yes simply says "Finished in xx:yy "
[03:14:36] <Ito-Brazil> It´s a good feature, even TurboCNC has it
[03:14:57] <jmkasunich2_> I tried to add it to axis once, but got lost in the python jungle
[03:15:10] <cradek> unfortunately each gui would need its own implementation of something like this.
[03:15:23] <jmkasunich2_> I think of it as a gui feature
[03:15:38] <jmkasunich2_> and I only use one gui ;-)
[03:16:31] <Ito-Brazil> i see
[03:16:58] <Ito-Brazil> well i hope to see it on to do´s skedule sometime...
[03:17:06] <cradek> Ito-Brazil: the best ways to get it incorporated would be to write it and submit a patch, or file a feature request on sourceforge's tracker
[03:17:24] <jmkasunich2_> I started the former, and gave up, then considered the latter but forgot
[03:17:35] <jmkasunich2_> * jmkasunich2_ <-- not very effective
[03:17:48] <cradek> a third option would be to bother jmkasunich2_ about it until he does something
[03:18:09] <cradek> my list is in decreasing order of effectiveness
[03:18:14] <archivist> * archivist contemplate having a looksee to do it, or just nag jmkasunich as well
[03:18:23] <jmkasunich2_> "does something" would probably consist of begging you or jepler for help, probably resulting in jepler doing it
[03:18:25] <Ito-Brazil> I gess it´s not hard, i mean: take the end time and subtract the end sime
[03:18:44] <Ito-Brazil> implemented on RUN button
[03:18:50] <jmkasunich2_> the hard part is the display parts - gui stuff
[03:19:01] <cradek> a little harder if you want to handle pause, step, optional stop, manual tool change, etc properly (??)
[03:19:02] <archivist> a few lines plus some display foo
[03:19:04] <jmkasunich2_> eww, I got gui goo on me
[03:19:27] <jmkasunich2_> define properly - for me, I'm interested in end-to-end clock time
[03:19:37] <cradek> yes exactly - define properly
[03:19:47] <Ito-Brazil> me too, no matter if i paused or something
[03:19:56] <jmkasunich2_> end-to-end clock time is rather simple to calculate
[03:20:01] <cradek> but if you fall asleep and don't change a tool when it asks ...
[03:20:03] <Ito-Brazil> just how time i spent on that g-code
[03:20:06] <archivist> end to end mainly as stop time needs charging for
[03:20:14] <jmkasunich2_> then my hourly rate just dropped - so be it
[03:20:21] <cradek> heh
[03:20:44] <Ito-Brazil> lol
[03:20:46] <jmkasunich2_> I'm actually quite interested in measureing my tool change time, etc, as part of the total
[03:21:19] <archivist> a table of values and a total
[03:21:25] <Ito-Brazil> well the timer can have a milling time, change tool time, total time...
[03:21:35] <jmkasunich2_> heh, now its getting complicated
[03:21:43] <cradek> told you so
[03:22:07] <Ito-Brazil> hehe it can be complicated, but can start with a simple time elapsed
[03:22:12] <cradek> but really, the simplest possible implementation could be pretty simple
[03:22:13] <jmkasunich2_> I just want hh:mm:ss, that starts at zero when I hit "R", and stops when the program hits end-of-file (M2, whatever)
[03:22:26] <Ito-Brazil> Yeh
[03:22:40] <Ito-Brazil> me too
[03:22:59] <cradek> halui.program.is-running
[03:22:59] <cradek> status pin telling that a program is running
[03:23:01] <Ito-Brazil> and this stuff can evolute to the other clocks
[03:23:04] <archivist> and store in a last job text file at end
[03:23:07] <Ito-Brazil> Later hehe
[03:23:18] <jmkasunich2_> heh, do it in hal?
[03:23:28] <jmkasunich2_> we need a pyvcp clock widget
[03:23:32] <cradek> reset and start counting up when a signal goes high, stop counting when it goes low
[03:23:54] <Ito-Brazil> nice cradek!
[03:23:54] <cradek> just a thought. it's probably a stupid way to do it.
[03:24:06] <jmkasunich2_> it means I don't have to understand how axis works
[03:24:23] <jmkasunich2_> and it could be used to time other things
[03:24:43] <Ito-Brazil> i gess not. It´s just a time mesuring when halui.program.is-runing is ON
[03:25:36] <Ito-Brazil> how can i learn about this language?
[03:25:55] <Ito-Brazil> i mean what language is used to build those things on EMC2?
[03:26:06] <jmkasunich2_> depends on what "things" you are talking about
[03:26:17] <Ito-Brazil> the interfaces
[03:26:22] <jmkasunich2_> the GUI?
[03:26:28] <jmkasunich2_> there are several, written in several languages
[03:26:57] <Ito-Brazil> EMC2 MINI, TKEMC, AXIS ... how to put a clock there? its made on python? what python
[03:27:08] <cradek> I verified that halui.program.is-running works how you want
[03:27:15] <Ito-Brazil> wow
[03:27:18] <jmkasunich2_> you are asking questions that don't have simple answers
[03:27:30] <Ito-Brazil> lol sorry, really
[03:27:31] <jmkasunich2_> I'm gonna see if I can add a pyvcp clock widget
[03:27:45] <jmkasunich2_> if so, then adding the clock to axis (and only axis) will be very simple
[03:27:59] <jmkasunich2_> you could also add it using a standalone pycvp window if you are using one of the other GUIs
[03:28:10] <jmkasunich2_> either way, it becomes a config thing (HAL), not a programming thing
[03:30:04] <archivist> or just store events and timestamps report later
[03:31:11] <archivist> anyway 4:30 am time to booger off home
[03:31:18] <Ito-Brazil> this clock can just be configured on HAL features? is that right?
[03:32:24] <jmkasunich> it isn't invented yet
[03:32:28] <jmkasunich> but that is the idea
[03:32:35] <archivist> events and timestamps from HAL pins could be a useful debugging tool also
[03:33:21] <jmkasunich> timestamps would want to come from a RT component for accuracy
[03:33:48] <jmkasunich> halsampler is already something that works sort of like that
[03:34:57] <archivist> I was going to design a logic analyser that way some years ago, but never got round to it
[03:37:26] <Ito-Brazil> can i do it on Virtual Control Panels ?
[03:37:33] <Ito-Brazil> may i place a link ?
[03:39:01] <Ito-Brazil> http://www.linuxcnc.org/docs/devel/html/hal_vcp.html#r1_5_7
[03:45:18] <jmkasunich> Ito-Brazil: that is exactly what I'm talking about
[03:45:45] <jmkasunich> I'm trying to add a widget similar to the Number widget (section 1.5.5 on that page) that will display hh:mm:ss
[03:46:58] <Ito-Brazil> cool
[04:05:21] <tomp> google code has http://code.google.com/p/edmcontrol/
[04:05:27] <tomp> but I cant find any contact to author or googlecode,
[04:05:37] <tomp> and the non-member checkout it a tree of empty files,
[04:05:39] <tomp> while the member checkout seems to require some contact.
[04:05:54] <tomp> catch-22! any ideas?
[04:25:05] <Ito-Brazil> got to go, cheers
[20:54:32] <ChanServ> [#emc] "This is the #emc channel - talk related to the Enhanced Machine Controller and general machining. Website: http://www.linuxcnc.org/, wiki at http://wiki.linuxcnc.org/"
[21:28:04] <alex_joni> good night all
[21:29:04] <tomp> gnite alex
[22:17:29] <davidf> Hi, Can anyone suggest a good cheap or free dialup ISP that will work with Ubuntu?
[22:20:11] <cradek> I suggest looking for a local one in your city and calling them to ask.
[22:20:58] <dmess> where are you located??
[22:21:29] <davidf> cradek, thanks, I wanted to sign up today (Sat) online. I forgot to pay my cable bill and wanted a backup in case I get shut down before Monday, LOL.
[22:21:35] <dmess> many cities are WI-fi these days
[22:21:52] <davidf> dmess, thanks, not gots.
[22:22:24] <dmess> Toronto is on its way.... not fully yet
[22:22:34] <davidf> Just thought it would be good to have an alternate connection in case of outages, etc.
[22:23:09] <davidf> Any of you guys tried any free ones, or Net-Zero, etc?
[22:23:12] <cradek> I would use the local coffee shop for that kind of pinch
[22:23:20] <dmess> you could have mine but long distance charges WILL apply
[22:23:32] <davidf> :)
[22:24:29] <dmess> where are you i can see if i can get a regional # for you to link into
[22:25:22] <dmess> MY ips is a close friend.... and anything goes
[22:26:34] <davidf> oops, sorry for the caps! Hit it by accident.
[22:26:54] <dmess> i liked missouri when i visited for 10 days...
[22:27:53] <dmess> but WOW your ppl are big.... i'm 155 lbs and i was 1/2 scale to most
[22:28:33] <dmess> the food i was served for dinner would have fed the family of 5 here at home..
[22:29:20] <dmess> met 1 guy in the shop i was in... 701 lbs
[22:30:08] <davidf> Well thats not average size!
[22:31:35] <dmess> no... i understand... but he aint helpin' the average...LOL
[22:32:02] <dmess> he was an inspector...
[22:32:32] <dmess> but if he leaned on the inspection table the laser seen it
[22:32:58] <davidf> I imagine just walking around might have done that, LOL.
[22:33:08] <dmess> 3d laser inspection with hand held targets
[22:33:52] <dmess> he made out pretty quick.. and had bagged a few deer the prior spring.
[22:35:01] <davidf> ??
[22:35:14] <dmess> HE was JUST the biggest man i EVER seen... i got to talk and know him... just cool... STILL workin' too... was no sighn of gving up...
[22:36:02] <dmess> he got by with a little help from hes friends..
[22:37:52] <davidf> well, I gotta go look into dialup some more... Just was wondering if anyone knew if linux vs windows would be an issue for that...
[22:38:56] <davidf> Don't see why it should be as long as you can establish a connection via Gnome software or whatever?
[22:39:07] <dmess> KC was the ONLY place i was ever erly for a flight and it left 1.5 HRS early...... the pilot walked thru the longe and asked WHO was going to YYZ... GET ON NOW OR YOU wont be getting there..... that was the friday BEFORE sept 11/2001
[22:41:46] <davidf> Well, thanks anyway... Gotta go, see ya!
[22:42:30] <davidf> bye
[23:15:07] <chr0n1c> hello! i wanted to report something strange/kinda nice that happened. i was running this 22 mb g-code file doing a tiny engraving of a daisy into some lexan, and using the computer to browse the web and do some irc chatting while the code was running... well i pushed the computer too far apparently when i tried to view an image and everything locked up EXCEPT FOR EMC2! i mean the mouse pointer wouldn't even move but the g-code
[23:15:34] <chr0n1c> ... the question is, is there anyway to recover from something like this without stopping the program?
[23:20:24] <chr0n1c> * chr0n1c lesson learned: use my laptop for the internet when i am running huge intricate engraving gcode files on the cnc.
[23:22:22] <Guest16377> Guest16377 is now known as raven2