#emc-devel | Logs for 2010-09-19

[02:01:06] <SteveStallings> SteveStallings is now known as steves_logging
[10:17:35] <Dave911_> Dave911_ is now known as Dave911
[17:23:01] <psha> Good evening (at least here is evening).
[17:23:33] <psha> Can anybody tell me if there is some serious limitation preventing halui to execute many commands in MDI_COMMAND statement
[17:23:45] <psha> Except that it's not implemented :)
[17:25:43] <micges> limit is 64 by implementation
[17:26:14] <psha> No, not many MDI_STATMENTS
[17:26:31] <psha> But many G-code statments (e.g G0 Z1; G0 X0 Y0; G0 Z0)
[17:26:37] <psha> in one MDI_COMMAND
[17:26:40] <micges> buoh
[17:26:56] <micges> it's limit by gcode syntax
[17:27:21] <micges> so only way is to use named O subs
[17:27:27] <psha> That's right. But is there some reasons why not to split string on ;?
[17:28:28] <psha> Right now triggering mdi-command pin results into submitting mdi command to emc without waiting for it result
[17:28:36] <micges> emc doesn't support multilines MDI
[17:29:17] <micges> that's true
[17:29:21] <psha> I've patched halui so it supports ';' delimetered G-codes.
[17:29:34] <psha> My question is if it is 'wrong by design' or not
[17:30:24] <psha> It is possible to run many g-codes in sequence with triggering motion.digital-out-XX and oneshot's
[17:30:30] <psha> But overcomplicated
[17:31:12] <psha> So I'm wondering why MDI_COMMAND was not changed to support many commands.
[17:31:28] <psha> I'm not common with CNC so I may miss some trivial things
[17:32:48] <micges> how is emc behave after you issue mdi command with 'G0 Z1; G0 X0 Y0; G0 Z0' ?
[17:34:27] <psha> It's executing them in order
[17:35:16] <psha> But maybe there are some difficulties with use interaction while doing this...
[17:35:43] <psha> 's/use/user/'
[17:36:16] <micges> I think there is no such functionality in emc becuse nobody needs it much
[17:37:15] <micges> can you make a patch? and paste at pastebin.com
[17:37:29] <psha> Right. So maybe you may point me to correct solution for my task.
[17:37:49] <psha> Consider "return to home" algorithm.
[17:38:06] <psha> First we need to raise on Z coordinate to the "safe height"
[17:38:09] <psha> and only then move home
[17:38:31] <micges> right
[17:38:45] <psha> Otherwise we may damage mill if we are staying on current Z
[17:38:55] <psha> So we need to perform at least 2 tasks in order
[17:39:02] <psha> What is correct way to do it?
[17:39:42] <micges> let me see
[17:41:25] <psha> I've found solution with oneshot's and motion.digital-out-XX on wiki
[17:41:34] <psha> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Adding_More_Controls_To_Simple_Remote_Pendant
[17:41:41] <psha> But it's horribly complicated
[17:43:25] <micges> let say we have zeroes in 5161-5166 parameters in *.var file and have Z max at 200mm
[17:44:14] <micges> then G28 Z200 will first move z to 200 and then move XYZ to 0
[17:44:50] <micges> hmm so you must have parameter 5263=200 and then it will work
[17:44:56] <micges> http://www.linuxcnc.org/docview/html//gcode_main.html#sub:G28,-G30:-Return
[17:45:39] <psha> Oh, great!
[17:49:23] <psha> So my root problem (how to return home) is solved.
[17:49:53] <psha> If it's not difficult may you help me a bit more.
[17:50:48] <micges> psha: join #emc
[17:51:26] <psha> Sorry for incorrect questions for -dev channel )
[19:56:02] <ries_> ries_ is now known as ries
[20:28:55] <psha> micges: any chance for review today?
[20:29:49] <micges> not by me, wait for others developers
[20:31:27] <psha> maybe mailing list will be more suitable for such questions?
[20:31:57] <psha> it seem to be pretty large time offset between me and most of devs
[20:32:16] <micges> it's sunday
[20:32:23] <psha> hm, really )
[20:32:57] <micges> on sundays there is quiet here
[20:33:17] <psha> so I'll try it again on monday
[20:33:39] <psha> thanks for help and great software
[20:33:43] <micges> send mail to mailing list
[20:33:59] <psha> tomorrow ) it's deep night here...
[20:34:11] <micges> here is 22:30
[20:34:24] <psha> 00:30
[20:34:28] <micges> in moscow I think is about 1 am
[20:36:45] <psha> 2 hour is small offset... it was very hard to push patches upstream where only dev was living in a EST timezone...
[20:37:39] <micges> I'm from Poland
[20:38:23] <psha> pretty close
[20:39:04] <psha> and other devs? mostly europe or worldwide?
[20:39:33] <micges> in europe just few, rest from USA
[20:40:19] <psha> thanks again for your help :) see you later
[20:40:27] <alex_joni> 23:30 here ;)
[20:40:37] <psha> even closer )
[20:40:57] <alex_joni> but that's about it for europe
[20:41:03] <alex_joni> the rest are across the pond
[20:41:35] <psha> is not it hard to communicate?
[20:42:14] <psha> from my experience it is not so hard but very slow...
[20:42:26] <psha> even trivial questions may hang for a long time
[20:43:13] <alex_joni> I didn't see problems like that
[20:43:34] <alex_joni> I used to develop pretty late, so timedifference was not much of an issue
[20:43:50] <alex_joni> lately things have changed, but that's only on the personal side
[20:47:17] <psha> at least leaving :) good night :)