#linuxcnc-devel | Logs for 2014-11-16

[00:56:29] <zeeshan-MILL> hm
[00:56:37] <zeeshan-MILL> im geting about 18000 ns
[00:56:51] <zeeshan-MILL> after couple hours of running this thing
[00:57:01] <zeeshan-MILL> whoops
[09:49:10] <jepler> so what do we do about the guy who wants to improve sampler, but says he can't generate a patch, so he can't give a signed-off-by to his changes?
[10:11:23] <cradek> you can SOB the changes if you feel it's appropriate
[10:19:01] <jepler> this rebase of liblinuxcnc-ui onto master is harder going than I expected
[10:25:39] <jepler> huh and my gitk's broken (shows terminal escape codes in the diff window)
[10:56:06] <seb_kuzm1nsky> the bug is in the if statement on line 3436 of task's main
[10:56:10] <seb_kuzm1nsky> http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=blob;f=src/emc/task/emctaskmain.cc;h=6c1175e472f506d2fdefd876337a19d16b1d6ce7;hb=refs/heads/2.6#l3436
[10:56:23] <seb_kuzm1nsky> it's forgetting to check mdi_execute_queue.len()
[10:56:53] <seb_kuzm1nsky> task can get itself into a situation where there's nothing on the interp_list, nothing in emcTaskCommand, but queued mdi commands in mdi_execute_queue
[10:57:19] <seb_kuzm1nsky> and in that situation, status->status blinks Done for one task-cycle
[11:13:09] <jepler> that sure could be it
[11:13:12] <jepler> seb_kuzm1nsky: good sleuthing
[11:13:39] <jepler> (are status and task.status ever set to different values?)
[12:05:15] <seb_kuzm1nsky> i dont think so
[13:12:26] <cradek> what are the chances of getting the 5i25 to run my C axis when I don't actually know what pins it's wired to?
[13:16:56] <Tom_itx> good
[14:03:10] <cradek> hm, I think it's hooked to pins 16 and 17
[14:05:03] <cradek> linksp Cstep => parport.0.pin-17-out
[14:05:03] <cradek> linksp Cdir => parport.0.pin-16-out
[14:05:04] <cradek> yep
[14:05:09] <cradek> (hey why not look at the old config)
[14:15:02] <cradek> well that bites
[14:24:08] <kwallace> I'm using feed rate compensation on arcs in my milling g-code generators. I calculate an adjusted feed from the ratio between the tool control point radius and the arc surface radius. This gives me the same rate on the arc surface as called out for a straight cut. In giving it more thought, the tool is usually cutting most of the tool diameter width, so with wide tools on small radii, the feed rate can be very different across the tool width.
[14:26:03] <cradek> yeah it's really not clear what's best to do
[14:33:02] <kwallace> I guess I need to look at each set of values for tool and path, then change the calculation to fit if needed. I kind of wish I had thought of this earlier.
[14:33:40] <cradek> "just don't do that" might be a decent solution
[14:34:09] <cradek> or only slow down if your part is on the long side, don't speed up if it's on the short
[14:34:31] <cradek> I have never compensated arc feeds in that way and it hasn't bitten me yet
[14:35:45] <kwallace> Indeed. I wonder what G41/42 does.
[14:35:57] <Tom_itx> i can say it's not something production shops worry with
[14:36:23] <cradek> g41/g42 mode theoretically has enough knowledge to do this, but it doesn't do it
[14:36:50] <cradek> I bet it causes more problems than it solves, unless the tools are very big
[14:37:55] <Tom_itx> generally speaking larger tools are used for roughing passes where it matters little
[14:38:04] <Tom_itx> it's the finish pass that pays the money
[14:38:15] <cradek> did you go down this path for practical reasons - to solve a problem, or just because it seemed right to do?
[14:39:31] <Tom_itx> climb vs conventional will cause you just as much grief
[14:39:41] <kwallace> It seemed right, plus some of my cuts sounded better, but it may not apply everywhere.
[14:42:18] <kwallace> I seem to recall being influenced by this: http://www.sandvik.coromant.com/en-gb/knowledge/milling/application_overview/holes_and_cavities/circular_external_milling_ramping/Pages/default.aspx
[14:42:34] <Tom_itx> i think my cam has a setting for arcs but i'm not sure i use it
[14:43:35] <kwallace> Fortunately, my CNC machines are pretty tight so I always climb.
[14:43:44] <Tom_itx> yeah
[14:43:50] <Tom_itx> you're better off if you can
[14:46:56] <kwallace> I think I'll look at calculating the max and min effective feed rate, then apply compensation if it seems reasonable, but otherwise turn it off.
[14:49:26] <cradek> I'd think the only-slow-down algorithm might be the "reasonable" test
[14:54:10] <jepler> > Subject: [UNNECESSARY PATCH 00/16] signal: coding style wankery...
[14:55:27] <kwallace> I think stainless, titanium or inconel might have a problem with a slow down. :)
[14:58:24] <cradek> grr, if the motor on C was the same, I could just use Y's amp
[15:01:21] <Tom_itx> kwallace is it that noticeable a change?
[15:01:48] <Tom_itx> true they'll work harden but constant feed is what you're after there
[15:03:53] <kwallace> http://www.cnczone.com/forums/tormach-personal-cnc-mill/206418-machining-304-ss-first-time-tormach-steel.html
[15:05:53] <Tom_itx> FSWizzard is a good tool to follow
[15:06:00] <cradek> so, looks like custom 5i25 firmware, or go back to a plain parallel port
[15:06:04] <cradek> I don't like either option
[15:06:24] <Tom_itx> why not?
[15:06:32] <Tom_itx> it's not that difficult to mod a bit file
[15:06:52] <cradek> if I have one that nobody else is using, it won't get bugfixes
[15:07:11] <Tom_itx> oh
[15:07:18] <Tom_itx> early test model?
[15:07:43] <cradek> all software is an early test model :-)
[15:16:32] <kwallace> Recently, I wanted to play with a FET bridge using a 20kHz PWM with decent resolution. I was surprised to find my Arduino could do it at 20kHz and 10 bit resolution with a few lines of code. I didn't need to resort to an FPGA.
[15:18:12] <Tom_itx> yep
[15:21:17] <cradek> pcw_home: would you build me a custom 5i25 firmware please?
[15:23:40] <cradek> I think I am currently using 5i25_dmmbob1x2.pin but I need an additional stepgen on the external connector, dir=16 step=17
[15:24:02] <cradek> I don't use the PWMs
[15:24:13] <Tom_itx> is your current bit file custom?
[15:24:42] <cradek> no
[15:25:32] <Tom_itx> just one extra stepgen on pin 16 17?
[15:27:13] <Tom_itx> i don't see the file in the zip
[15:29:33] <cradek> that's in my 5i25.zip
[15:29:53] <cradek> I think it's the one that matches what I'm getting from mesaflash --readhmid
[15:30:48] <Tom_itx> i just grabbed a fresh one...
[15:35:30] <Tom_itx> there it is
[15:56:07] <Tom_itx> not finding a .vhd file for it
[15:56:58] <Tom_itx> maybe the 5i25 is configured differently..
[15:57:05] <Tom_itx> i'm not sure since i don't have one
[16:10:42] <Tom_itx> cradek if i could find it i'd look at it...
[16:15:12] <cradek> Tom_itx: thanks, I can wait for pcw to help me tomorrow
[16:15:24] <cradek> I'm so far down the rabbit hole I bet I wouldn't get anything cut today anyhow
[16:15:41] <Tom_itx> i see the xml and pin but no vhd
[16:15:57] <Tom_itx> not in the source dir
[16:16:51] <cradek> chris@emc:~/Downloads/5i25/configs/hostmot2/source$ unzip -v hostmot2.zip |grep BOB
[16:16:54] <cradek> 10015 Defl:N 2329 77% 2014-01-14 16:19 10ce2727 PIN_DMMBOB1x2_34.vhd
[16:17:04] <cradek> that's surely it
[16:17:08] <skunkworks_> cradek: whatcha makin?
[16:17:26] <Tom_itx> lemme look again
[16:17:29] <cradek> skunkworks_: max needs to be 5 axis again, for cutting some lenses
[16:17:45] <skunkworks_> neat@
[16:18:46] <Tom_itx> ok i see that
[16:22:06] <Tom_itx> what IO are they?
[16:22:13] <Tom_itx> 5 & 7?
[16:22:34] <Tom_itx> those are pin 16 & 17 but not sure which connector
[16:22:55] <cradek> pretty sure 5/7 is right
[16:23:12] <cradek> you could add a stepgen to both connectors
[16:31:17] <Tom_itx> heh. i forgot i swapped hdd. installinx webpack...
[16:31:55] <cradek> you're brave, thanks for doing it
[16:32:09] <Tom_itx> i did a custom one for mine
[16:32:22] <Tom_itx> i just haven't done a 5i25
[16:32:26] <Tom_itx> looks quite similar
[16:32:47] <Tom_itx> i needed to get this set up again anyway
[17:11:23] <Tom_itx> cradek are you using all 8 stepgens now?
[17:12:01] <cradek> no, I only need the 5 of them on the external connector
[17:16:48] <Tom_itx> are you using the internal connector?
[17:18:00] <Tom_itx> i have 5 on the external one now
[17:18:16] <cradek> no, I don't care about the internal connector
[17:18:34] <Tom_itx> dir pin 16 step pin 17
[17:19:11] <cradek> yay, that's right
[17:23:35] <Tom_itx> ok we'll see what happens
[17:29:52] <seb_kuzminsky> i've got a test here, written in C using lui, that triggers the dropped-mdi bug within about an hour of repeated running
[17:30:20] <seb_kuzminsky> i'm confident the bug is present in task in 2.6
[17:30:38] <seb_kuzminsky> i tried to write a test using the "linuxcnc" python module in 2.6 to trigger the bug, but it won't bite
[17:30:44] <cradek> ugh, an hour
[17:30:56] <Tom_itx> generating bit file
[17:30:57] <seb_kuzminsky> the bug is highly sensitive to timing, and i'm guessing that python's too slow to trigger it
[17:31:24] <seb_kuzminsky> i think i'm going to make sure it's fixed in my lui branch, then cherry-pick the task fix back to 2.6 and merge it up
[17:31:40] <seb_kuzminsky> because i'm so sick of this thing that i dont want to write a non-lui C program to test for the bug
[17:31:56] <cradek> yeah just fix it and move on, jeez
[17:32:05] <seb_kuzminsky> cradek: i really want to see a video of max grinding lenses
[17:32:06] <seb_kuzminsky> :-)
[17:32:20] <cradek> a test that triggers a bug once an hour isn't something we want the buildbot to run anyway
[17:32:34] <seb_kuzminsky> heh yeah
[17:32:58] <seb_kuzminsky> it's blind luck that the little lui test happened to trigger the bug randomly one time
[17:33:03] <seb_kuzminsky> for some definition of luck
[17:33:41] <seb_kuzminsky> <+cradek> yeah just fix it and move on, jeez
[17:33:48] <seb_kuzminsky> i get that a lot from my $DAYJOB boss too...
[17:33:55] <cradek> heh
[17:41:23] <Tom_itx> cradek, see what this does: http://tom-itx.ddns01.com:81/~webpage/cnc/C-Radek/
[17:41:41] <Tom_itx> added the c suffix
[17:41:51] <cradek> Tom_itx: could you also put the vhd and pin there?
[17:41:59] <Tom_itx> vhd yes
[17:42:01] <Tom_itx> pin maybe
[17:42:04] <cradek> heh
[17:44:09] <Tom_itx> i don't think it generates a pin file
[17:44:12] <Tom_itx> i've never used one
[17:45:34] <Tom_itx> you can edit your other one to reflect pin 16 17 stepgen
[17:47:41] <Tom_itx> i should clean up the vhd file
[17:47:51] <Tom_itx> i left the original lines commented out...
[17:55:45] <Tom_itx> that give you 9 stepgens
[17:55:50] <Tom_itx> instead of 8
[18:02:59] <cradek> Tom_itx: thanks, I will try it after dinner
[18:03:31] <Tom_itx> if for some reason that's too many stepgens i can get rid of one off the other connector
[18:03:49] <Tom_itx> i ran into one once where i had too many functions defined
[18:03:55] <Tom_itx> but this is just instances
[18:14:58] <Tom_itx> i noticed i mislabelled them A step dir... there's already a set for A.. no biggie
[18:15:41] <Tom_itx> maybe B Step B dir would be better
[18:16:50] <Tom_itx> fixed..
[18:27:14] <pcw_home> cradek: looks like Tom_itx has it covered, if you have any troubles, I can make one tommorow
[18:29:21] <pcw_home> but beware, untested bitfiles on a 5i25 can brick a 5i25 since the PCI stuff needs to work to re-flash
[18:29:22] <pcw_home> (well only a temporary brick if you have a JTAG cable)
[18:30:49] <pcw_home> most common mistake: no .ucf file associated so pinout is random (this seldom works well)
[18:42:07] <Tom_itx> pcw_home did you look at the file?
[18:44:38] <Tom_itx> pcw_home, also what generates the .pin file?
[18:44:51] <Tom_itx> i gather it's just an information file...
[18:45:30] <pcw_home> the pinout file looks OK
[18:45:31] <pcw_home> mesaflash --readhmid generates pinout files for example
[18:45:59] <pcw_home> Jeplers build scripts do also
[18:46:24] <pcw_home> or maybe that just .xml files, I forget
[18:48:14] <pcw_home> the main danger in 5i25 file is making sure the proper ucf file is assosciated
[18:48:15] <pcw_home> (ISE cheerfully makes random pinouts without complaint if no .ucf file is assoscited with the top level file)
[18:48:17] <Tom_itx> it seemed to build ok
[18:50:58] <pcw_home> if there are conflicts in the pinout file you will get build errors but of course
[18:50:59] <pcw_home> illogical things like all dir and no step pins are legal so not checked
[18:51:25] <Tom_itx> yeah
[18:52:28] <Tom_itx> 5i25.ucf is associated with tht TopPCIHostmot2.vhd
[18:52:41] <pcw_home> you can verify the created bitfiles pinout with the pinout report
[18:52:41] <Tom_itx> just double checked
[18:53:07] <pcw_home> So its probably OK
[18:53:46] <Tom_itx> my only real concern was having 9 stepgens
[18:54:00] <Tom_itx> he's not using them all...
[18:55:35] <pcw_home> 15 or so should be OK
[18:56:05] <pcw_home> (eventually you will run out of space, that the only real limit)
[18:57:44] <pcw_home> plus thay add compile time (the more stuff in the FPGA _and_ the closer you are to being full. the slower the compilation will be)
[19:08:49] <Tom_itx> pcw_home where do i find the pinout report?
[19:09:23] <pcw_home> its one of the things map does
[19:10:04] <pcw_home> probably have to root through the menus to find it
[19:12:20] <Tom_itx> timing analizer -> post map?
[19:12:29] <cradek> pcw_home: do you think it's safe to try?
[19:12:44] <cradek> pcw_home: I don't really have a jtag cable
[19:12:48] <cradek> (but how hard could it be)
[19:13:06] <pcw_home> a paralle port and a HC125
[19:13:42] <pcw_home> if Toms pinout report looks sane its probably OK
[19:14:20] <Tom_itx> i just replaced 2 null tags with an extra stepgen pair and renumbered them
[19:14:30] <Tom_itx> adding the stepgen count to 9 instead of 8
[19:14:38] <Tom_itx> at the top
[19:14:47] <Tom_itx> 'instances'
[19:15:09] <pcw_home> If mesaflash was a bit smarter, there would be safe way to test new bitfiles
[19:15:17] <Tom_itx> and added the 'c' suffix to the function name
[19:16:33] <pcw_home> Tom_itx: do you have recent source? The ICAP stuff is pretty recent (not required but nice)
[19:16:45] <Tom_itx> source for what?
[19:17:04] <pcw_home> hostmot2 source
[19:17:24] <Tom_itx> i just grabbed the 5i25 file from your site
[19:17:52] <Tom_itx> should be in there...
[19:18:04] <Tom_itx> my 7i43 files are a bit older
[19:18:25] <pcw_home> OK so that should be up-to-date, have to ask micges about a booting from fallback area option
[19:20:02] <Tom_itx> if he has concerns you can generate it for him tomorrow
[19:20:09] <pcw_home> (so you could test a new bitfile by writing it to the fallback area, and booting from there)
[19:20:10] <Tom_itx> i posted the source on my site
[19:20:10] <pcw_home> if your bit file was bad, a power cycle would be all thats needed to recover
[19:20:35] <Tom_itx> http://tom-itx.ddns01.com:81/~webpage/cnc/C-Radek/
[19:21:36] <Tom_itx> i don't wanna screw one up but i don't mind making one once in a while for ppl
[19:21:44] <pcw_home> I dont think theres anything you can do in a pinout file that will affect card access
[19:21:45] <pcw_home> really only bad ucf file or unassociated ucf file
[19:22:11] <pcw_home> so its very likely its OK
[20:17:35] <KGB-linuxcnc> 05dgarr/jog_ignore_fo 2b5c9ce 06linuxcnc 04. branch deleted * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=2b5c9ce
[20:17:49] <KGB-linuxcnc> 05dgarr/jog_no_fo 0d18f03 06linuxcnc 04. branch deleted * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=0d18f03
[20:42:14] <cradek> whee, burnt, rebooting
[20:42:49] <Tom_itx> ?
[20:44:57] <seb_kuzminsky> cradek: cool!
[20:45:19] <cradek> no brick!
[20:45:24] <seb_kuzminsky> it would be cool if we had a working hm2 buildbot, then you could add a config file for the firmware you want and get automatic builds & bugfixes for it
[20:45:30] <cradek> but the stepgens are in a weird order
[20:45:45] <Tom_itx> i found that changed on mine too
[20:45:48] <cradek> 2/3, 4/5, 16/17, 6/7, 8/9
[20:46:18] <cradek> seb_kuzminsky: yeah, that would have been cool, but probably not widely needed
[20:47:20] <Tom_itx> would you rather they be at the bottom?
[20:48:00] <cradek> yeah, but that wasn't given as a requirement originally :-)
[20:48:10] <cradek> it's fine, simple hal editing -- thanks again
[20:48:23] <Tom_itx> i made the instance order somewhat match the io
[20:48:32] <Tom_itx> that's probably why
[20:48:52] <cradek> yeah I was thinking more along the parport pin numbers
[20:51:02] <Tom_itx> i could move them to be stepgen 08
[20:51:05] <Tom_itx> if you like
[20:51:57] <cradek> I have jogs!
[20:51:59] <cradek> yay
[20:52:05] <Tom_itx> works?
[20:52:12] <cradek> yep
[20:52:14] <Tom_itx> cool
[20:52:20] <cradek> scale is wrong but it moves
[20:53:24] <cradek> actually the scale is right - jogs are just too slow
[20:53:27] <Tom_itx> keep that vhd file
[20:53:30] <cradek> so it's perfect
[20:53:35] <cradek> yeah it's right next to the .bit
[21:12:56] <Tom_itx> cradek, i'm reordering the stepgens... i'll post the files to the same link if you wanna use em
[21:13:24] <Tom_itx> suffix D instead of C
[21:14:46] <cradek> thanks, I'll download and keep them for next time around
[21:15:12] <Tom_itx> synthesizing right now..
[21:16:30] <Tom_itx> should be the same pin order as you had with 16 & 17 tacked on the bottom
[21:25:36] <Tom_itx> posted.
[21:27:45] <cradek> saved, thanks
[21:28:07] <cradek> if I need the 5th axis I'll reflash so they're in order again, but for now it's fine
[21:28:40] <Tom_itx> you use mesaflash to upload the bit files?
[21:28:44] <cradek> yeah
[21:28:47] <Tom_itx> i've never done a 5i25 yes
[21:28:49] <Tom_itx> yet*
[21:28:58] <cradek> I just use the packaged one from the linuxcnc repository
[21:29:03] <Tom_itx> yeah
[21:29:26] <pcw_home> You should not need to reboot anymore...
[21:29:35] <cradek> it said I needed to
[21:30:12] <Tom_itx> pcw_home do the ether cards load the same way?
[21:30:16] <pcw_home> Ahh because your old bitfile didnt support ICAP
[21:31:10] <pcw_home> Yes, mesaflash works with the Ethernet and EPP cards also
[21:31:24] <pcw_home> maybe even serial
[21:32:39] <pcw_home> you can check if a PCI card supports ICAP by reading offset 0x78
[21:33:38] <pcw_home> sudo mesaflash --device 5i25 --rpo 0x78
[21:33:58] <Tom_itx> what exactly is ICAP?
[21:35:14] <jepler> Internal Configuration Access Port, allowing reconfiguration of the running FPGA
[21:36:44] <Tom_itx> do they store 2 configurations?
[21:36:50] <Tom_itx> a default and a runing one
[21:37:05] <pcw_home> The standard setup uses 2
[21:37:20] <pcw_home> a user config and a fallback config
[21:37:27] <Tom_itx> right
[21:37:34] <Tom_itx> in case of an oops
[21:37:56] <pcw_home> so for example if you pull the plug while writing a new config, thats recoverable
[21:40:45] <cradek> wow XC blending sucks
[21:45:04] <skunkworks_> cradek: what linuxcnc version?
[21:45:08] <cradek> 2.6.4
[21:45:49] <skunkworks_> I don't think master is much better./
[21:46:04] <cradek> heck, really?
[21:46:28] <skunkworks_> only xyz - all other blends fall back to parabolic
[21:47:14] <cradek> heh, figures he'd keep the broken parts
[21:57:05] <skunkworks_> what is the lense for?
[22:01:33] <cradek> eyeglasses
[22:02:54] <cradek> putting lenses in some cool vintage solid gold frames (ebay $19 hahaha)
[22:06:52] <skunkworks_> ah - ok.
[22:08:14] <cradek> oooops space heater + spindle + vacuum = nope
[22:08:26] <skunkworks_> well that sucks..
[22:08:50] <cradek> yep
[22:09:00] <cradek> it's very smart to have the lights on a different circuit
[22:09:28] <Tom_itx> heh
[22:09:58] <cradek> argh, my offsets are gone
[22:10:05] <cradek> siiigh
[22:10:30] <cradek> ok, maybe that's enough for tonight
[22:15:15] <cradek> oh hey, I just realized that with super glue I could make the 6" scale that I want: mm on one side, 10ths/100ths of inches on the other
[22:15:34] <cradek> as it is, I always grab the wrong one and there are those stupid fractions
[22:40:30] <seb_kuzminsky> cradek: i have that ruler! no superglue involved!
[22:40:49] <Tom_itx> not easy to find but i may have one
[22:41:05] <Tom_itx> got a bunch of 'general' rulers on a bankrupcy sale once
[22:45:43] <seb_kuzminsky> cradek: http://imgur.com/a/oCazB (whiskey for scale)
[22:46:50] <skunkworks_> I think cradek said - mm on one side - 10/100th inch on othe..
[22:48:27] <seb_kuzminsky> i think that's what i showed
[22:48:53] <seb_kuzminsky> here's the 12" flexible version: http://www.use-enco.com/CGI/INSRIT?PMAKA=319-9574
[22:49:25] <seb_kuzminsky> and here's the 6": http://www.use-enco.com/CGI/INSRIT?PMAKA=319-9581
[22:50:33] <skunkworks_> how much whiskey have you been drinking?
[22:50:43] <seb_kuzminsky> err... some
[22:50:46] <seb_kuzminsky> what am i missing?
[22:50:57] <seb_kuzminsky> mm on one side, 10th/100ths on the other
[22:51:55] <seb_kuzminsky> i agree with cradek, the problem's not imperial units, the problem's fractions
[22:51:57] <cradek> by golly, that's exactly it
[22:51:59] <skunkworks_> the one you posted - http://imgur.com/a/oCazB is metric on both sides..
[22:52:17] <seb_kuzminsky> skunkworks_: how much whiskey have *you* been drinking? ;-)
[22:52:27] <cradek> well I know I haven't had enough
[22:52:39] <cradek> (none actually)
[22:52:45] <skunkworks_> heh - I must be totally missing something
[22:53:22] <seb_kuzminsky> cradek: it's my favourite
[22:53:34] <seb_kuzminsky> the 6" one lives in the little pocket on my shop apron
[22:53:35] <skunkworks_> oh - you mean - on the flip side...
[22:53:42] <seb_kuzminsky> haha
[22:53:53] * seb_kuzminsky points at skunkworks
[22:54:07] <skunkworks_> I have shingles - don't make fun..
[22:54:16] <seb_kuzminsky> and even cooler, it's got furlongs too
[22:54:22] <seb_kuzminsky> on the *in* side
[22:54:53] <cradek> I can have free shipping if I get 10 of them...
[22:55:22] <seb_kuzminsky> or if you wait 1-6 months for a free shipping deal
[22:55:48] <cradek> they could put it in an envelope with a 49c stamp, but noooo
[22:56:11] <seb_kuzminsky> well they'd also have to pay the guy/lady who goes to get it in the warehouse
[22:56:45] <seb_kuzminsky> i'm irrationally exhuberant to finally have a complete allen key set (again, and for a few months until i lose one)
[22:57:11] <seb_kuzminsky> http://www.use-enco.com/CGI/INSRIT?PMAKA=800-0177
[22:57:25] <seb_kuzminsky> (i'm trying to get cradek up to the $49 or whatever's needed for free shipping this week)
[22:57:39] <cradek> my latest spam says $99
[22:57:56] <cradek> they go in a special folder...
[22:58:24] <cradek> oh hey, I should get some 1/8" end mills because I'm on the last one
[22:59:33] <seb_kuzminsky> skunkworks_: how many shingles? in 100ths of an inch pls
[23:01:04] <skunkworks_> so far on my back.. only reports from my wife
[23:01:27] <seb_kuzminsky> stupid meat bodies
[23:01:39] <cradek> is that a virus?
[23:01:40] <seb_kuzminsky> when the revolution comes i'm going to have an awesome robot body
[23:10:19] <skunkworks_> yes
[23:14:51] <cradek> yuck, hope it's better soon
[23:15:45] <skunkworks_> thanks. trying to work on stress now :)
[23:20:00] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 60247a3 06linuxcnc 10src/emc/task/emctaskmain.cc task: fix the dropped-mdi bug * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=60247a3
[23:20:00] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 4b40d3a 06linuxcnc 10src/emc/task/emctaskmain.cc task: remove some dead code * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=4b40d3a
[23:20:57] <seb_kuzminsky> a 1-line patch is all i have to show for a week of debugging
[23:21:10] <seb_kuzminsky> goodnight, friends
[23:24:07] <skunkworks_> that fixes it?