#linuxcnc-devel | Logs for 2016-03-31

Back
[00:34:35] <linuxcnc-build> build #1137 of 4015.deb-wheezy-rtpreempt-amd64 is complete: Failure [4failed shell_3] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4015.deb-wheezy-rtpreempt-amd64/builds/1137 blamelist: Jeff Epler <jepler@unpythonic.net>, Chris Morley <chrisinnanaimo@hotmail.com>, Norbert Schechner <nieson@web.de>, Jan Mr?zek
[00:34:36] <linuxcnc-build> <email@honzamrazek.cz>
[00:38:30] <alfro_> I am novice so please forgive me for dumb query but what is the diff between linuxCNC and machinekit?
[01:06:54] <linuxcnc-build> build #214 of 4017.5.deb-wheezy-armhf is complete: Failure [4failed shell_3] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4017.5.deb-wheezy-armhf/builds/214 blamelist: Norbert Schechner <nieson@web.de>
[02:28:13] <linuxcnc-build> build #1795 of 4017.deb-wheezy-amd64 is complete: Failure [4failed shell_1] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4017.deb-wheezy-amd64/builds/1795 blamelist: Norbert Schechner <nieson@web.de>
[11:37:35] <seb_kuzminsky> win 4
[11:37:38] <seb_kuzminsky> whoops
[11:41:46] * jepler wonders what's so much better about window 4 than this window
[11:44:05] * cradek wonders what's wrong with seb's alt key
[12:39:21] <seb_kuzminsky> window 4 has #yosys in it
[12:39:58] <seb_kuzminsky> also, the alt key on my mac keyboard doesnt pass through ssh from work to the screen session on my home machine for some reason
[12:40:14] <seb_kuzminsky> i'm a mess, i know it
[13:02:44] <cradek> sorry, I can't parse "mac ... work"
[13:03:40] <jepler> *whispers* ESC is meta
[13:04:06] <cradek> do they have escape keys?
[13:04:16] <jepler> oh for all I know, no
[13:04:33] <cradek> can't you just plug your Model M into it?
[13:05:38] <jepler> now we're just being mean to him
[13:08:32] <seb_kuzminsky> i have no pride left
[13:55:43] <PCW> hmm are jog inputs to motion bounded by accel limits?
[13:57:37] <seb_kuzminsky> yeah, should be
[13:57:53] <seb_kuzminsky> but only if you jog using Model M keyboards
[13:59:09] <cradek> you mean the wheel?
[13:59:50] <cradek> ... yes wheel jogs obey all the constraints
[14:04:19] <PCW> hmm trying to figure out why a person on the forum gets stepgen following errors when jogging
[14:05:32] <cradek> what version?
[14:10:57] <PCW> I think 2.7.4
[14:11:39] <cradek> hm ok. (earlier versions often used lower than full acceleration and could hide marginal problems)
[14:11:48] <PCW> looks ok (just tried 1 inch jog with halcmd)
[14:12:12] <cradek> is he in stepgen position or velocity mode?
[14:12:28] <PCW> velocity with PID
[14:12:55] <cradek> you think the tuning is right?
[14:13:33] <PCW> Yeah P=1000 (for 1 KHz servo thread) and FF1 = 1
[14:14:44] <cradek> do you have a plot of what happens?
[14:14:52] <PCW> maybe something funky with function order or clocks
[14:15:06] <PCW> No that's the next step
[14:15:21] <cradek> I do realize I'm asking obvious questions
[14:15:46] <seb_kuzminsky> it'd be cool if something kept a record of the past 10 seconds or so of motion inputs & output, and wrote it to a file on estop/ferror
[14:16:06] <PCW> i can get < 2 uInch FE on the PID stepgen with 1G accell and 2400IPM
[14:16:34] <PCW> so something is definately funky
[14:17:42] <PCW> yeah a record, also FE limit for warning and separate FE limit for fault would be nice
[14:18:18] <PCW> even just recording the peak FE would be helpful
[16:01:15] <jepler> so what do we want in the forum -- multi-minute post times, or e-mails (particularly sign-up and password reset messages) that don't make it to people on widely-used e-mail providers such as *cough* microsoft
[16:03:01] <seb_kuzminsky> i'd prefer correctness (email to everyone) over performance in this case
[16:03:21] <cradek> I'm afraid I disagree
[16:03:34] <cradek> multi-minute pauses when posting make the forum unusable
[16:04:05] <cradek> it's better if it's usable for people who have successfully signed up, but X% of people have some trouble signing up
[16:05:40] <cradek> is there some kind of alternate routine that X% of people can use to sign up?
[16:05:55] <jepler> cradek: sure, use a different e-mail account
[16:06:00] <cradek> yeah even if it's "use gmail"
[16:06:17] <cradek> maybe if we can let them know that, we could call it fixed
[16:06:32] <seb_kuzminsky> i for one would not create a new email account in order to create a new forum account
[16:06:41] <jepler> the signup page did say something to that effect briefly, though I took it down when I made it start sending outgoing mails via gmail
[16:06:50] <cradek> sure but you also wouldn't use a shittacular service as your only email
[16:07:17] <cradek> I think people who use these free accounts tend to have several because of how bad they are
[16:07:43] <seb_kuzminsky> the people who signed up for email when hotmail was a reaonable choice probably often are not savvy enough to sign up for a newer one
[16:07:54] <jepler> heck I figured out how to sign up for gmail
[16:07:59] <jepler> I bet other people can too
[16:08:00] <seb_kuzminsky> (that's my hunch/prejudice)
[16:08:19] <cradek> breaking the forum for everyone to make it easier on those few people seems unacceptable to me
[16:08:41] <cradek> the tradeoff sucks, I know
[16:09:40] <cradek> there's a third option: some sisyphus among us could try to contact microsoft and tell them to fix it
[16:09:51] <jepler> ha hah hooohooohoo ha wheeee
[16:09:55] <seb_kuzminsky> \/disconnect
[16:10:23] <cradek> I hear they're responsive if you work at google
[16:11:02] <jepler> > Removal assistance fee is typically $350.00. Please see our email blacklist removal service page for more details.
[16:11:06] <jepler> you can just contact these people
[16:12:03] <cradek> /topic <jepler> ha hah hooohooohoo ha wheeee
[16:14:52] <jepler> I am also worried that if flo keeps bumping up against gmail's rate limiting, they will take some next step against what they regard as an ill-behaved system
[16:15:05] <cradek> that seems really likely
[16:15:16] <cradek> and it is one - gmail is not for that
[16:15:52] <cradek> it was an extremely clever hack, but it's just not going to work
[16:18:29] <jepler> .. I've changed it back and put back the warning on the sign-up page
[16:18:34] <jepler> "Some e-mail provders, notably hotmail / outlook / Microsoft-based services reject e-mails from forum.linuxcnc.org. If your normal e-mail address doesn't receive the e-mail necessary to complete the signup, use another e-mail provider such as gmail. As of 2016-03-28, gmail consistently delivers sign-up and password reset e-mails."
[16:18:59] <seb_kuzminsky> you rock
[16:19:30] <jepler> I'm doing my best at a task I don't want
[16:19:59] <seb_kuzminsky> you're doing way better at it than all the rest of us
[16:20:36] <jepler> Alert: The community is undergoing scheduled maintenance
[16:20:48] <seb_kuzminsky> and gauging by the amount of traffic on the forum, people appreciate it
[16:26:18] <skunkworks> https://github.com/synthetos/TinyG/wiki/Jerk-Controlled-Motion-Explained
[16:26:22] <jepler> I think that a good thing to do would be to get a proper spf record published for linuxcnc.org. that might cure some problems.
[16:29:02] <skunkworks> PCW: wonder if he has poor latency
[16:29:20] <jepler> >
[16:29:20] <jepler> Are you managing your IP and domain's sending reputation?
[16:29:29] <JT-Shop> PCW: is the hole spacing on the THCAD card 1.6" like the 7i73? (no drawing in the manual)
[16:31:26] <PCW> 1.550 Y 3.800 X
[16:32:00] <JT-Shop> thanks
[16:32:56] <skunkworks> I did have jog wheel issues with name on the k&5
[16:33:49] <skunkworks> K&t that I have not had a chance to figure out... fine with 2.7.4
[16:34:18] <skunkworks> Wow.. *ja
[17:44:02] <jepler> interesting -- rt-tests cyclictest added the ability to count SMIs
[17:44:10] <jepler> Daniel Bristot de Oliveira has added a hardware SMI count detection using
[17:44:14] <jepler> the MSR/SMI counter if available. Using the --smi option adds the count to
[17:44:19] <jepler> the end of the output, it looks like this (per thread)
[17:44:23] <jepler>
[17:44:27] <jepler> T: 0 ( 2285) P:80 I:1000 C: 9975 Min: 2 Act: 3 Avg: 4 Max: 20831 SMI: 2