#emc-devel | Logs for 2009-01-30

[00:14:48] <BigJohnT> what is the jmk-robot
[00:15:25] <skunkworks> heh - It is his robot he is making for a engineering competition.
[00:15:59] <skunkworks> I hope it doesn't become self aware...
[00:16:10] <BigJohnT> it's alive?
[00:19:11] <skunkworks> If some could do it - it would be jmk :)
[00:19:18] <skunkworks> *someone
[00:20:23] <BigJohnT> that's what I was thinking
[00:20:44] <jmk-robot> igor, steal me a brain
[00:21:13] <BigJohnT> it is alive!
[00:21:41] <BigJohnT> do you want the one marked abby normal
[00:21:52] <skunkworks> heh
[00:45:16] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/html/gcode_fr.html: french translation - try to follow John
[00:45:16] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/hal/ (pyvcp.lyx pyvcp_fr.lyx): french translation - try to follow John
[00:45:17] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/gcode/main_fr.lyx: french translation - try to follow John
[00:45:17] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/src/po/ (fr.po fr_axis.po fr_rs274_err.po): french translation - try to follow John
[01:13:41] <SWPadnos> hey jmk-robot, could you paste the output of lsusb somewhere?
[01:14:16] <jmkasunich> with or without the camera plugged in?
[01:14:32] <SWPadnos> either, I'm more curious about the root hubs
[01:14:45] <SWPadnos> the specs say there are 8 USB2 ports. that would be useful
[01:15:27] <jmk-robot> very non-informative
[01:15:42] <jmk-robot> ah, -v might help
[01:16:12] <SWPadnos> EHCI vs hmmm. no - wait a sec
[01:16:49] <jmk-robot> http://pastebin.ca/1322345
[01:17:26] <SWPadnos> maybe it was usbview that gave a nice listing (in a tree)
[01:17:34] <SWPadnos> but it doesn't work without effort on modern systems
[01:17:37] <SWPadnos> thanks
[01:17:41] <jmkasunich> np
[01:25:16] <SWPadnos> hmmm. could you repeat with sudo?
[01:26:00] <jepler> any volunteers to look at MichaƂ Geszkiewicz's patches?
[01:27:21] <SWPadnos> I looked at the first one briefly
[01:27:29] <jmk-robot> http://pastebin.ca/1322349
[01:27:55] <SWPadnos> it seemed like it would do what he said, but I didn't look closely enough to see if he's wrong or think about whether it's a good idea
[01:27:57] <SWPadnos> thanks
[01:28:06] <jepler> it looks OK to me but I haven't tried it
[01:28:20] <jmkasunich> that balancing thing is evil
[01:28:39] <cradek> luckily I didn't try that
[01:29:12] <jmkasunich> on 29 of 40
[01:29:26] <SWPadnos> well isn't that interesting. it looks like only one of those hubs is high speed
[01:29:31] <SWPadnos> the others are 12 Mbit
[01:29:42] <SWPadnos> (if I'm reading that output correctly)
[01:31:16] <jmkasunich> heh - got the all triangle one
[01:31:53] <SWPadnos> yeah, 29 is hard
[01:32:11] <jmkasunich> I got lucky on 31
[01:32:20] <jmkasunich> it was falling apart, and I just threw something in
[01:32:32] <jmkasunich> it stopped, and stayed together as I piled everything on top
[01:32:42] <SWPadnos> yep. had a couple of thse :)
[01:32:42] <jmkasunich> couldn't have made that config on purpose
[01:32:45] <SWPadnos> those
[01:37:56] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/hal/classicladder/protocol_modbus_master.c: don't assume signed char
[01:38:28] <jmkasunich> its odd how one is a pain, and the next is trivial
[01:38:34] <SWPadnos> yep
[01:38:53] <SWPadnos> after 29, I think I cruised through the rest in no time (except maybe one)
[01:39:06] <SWPadnos> then the inferno level is annoying - the physics are slightly different
[01:39:24] <SWPadnos> it seems that there's a slight list one way or the other on each one
[01:39:52] <jmkasunich> I think I'll skip that level, I want to accomplish something this evening
[01:40:09] <SWPadnos> good idea
[01:40:25] <SWPadnos> is your camera USB1 or USB2?
[01:40:31] <jmkasunich> dunno
[01:40:43] <skunkworks> I gave up at 29...
[01:40:48] <skunkworks> (level not age)
[01:41:00] <SWPadnos> I think it's USB1
[01:42:12] <SWPadnos> yep. bcdUSB 1.00 for the camera
[01:42:22] <SWPadnos> so plugging it into the USB2 port won't help :)
[01:43:52] <jmkasunich> done - bookmarked so I can play inferno some other day
[01:44:33] <SWPadnos> it's the same levels (as far as I could remember after 40 intervening), with vastly different graphics and somewhat different physics
[01:56:18] <SWPadnos> yep. bastids. the ICH7 chip has one EHCI (Enhanced) controller and 4 UHCI (Universal) USB controllers
[01:56:48] <jmkasunich> what does that mean to a normal person?
[01:57:01] <jepler> jmkasunich: if you don't plug your usb hard drive into the right port, it goes slow
[01:57:05] <SWPadnos> it has one 480 mbit (USB2) port and several slow ones internally
[01:57:23] <jepler> but externally all ports look identical, which helps
[01:57:36] <jmkasunich> it won't matter which port I plug my webcam into will it? ISTR you said the cam is slow anyway?
[01:57:51] <SWPadnos> yes
[01:58:02] <jmkasunich> good enough ;-)
[01:58:12] <SWPadnos> I'm thinking for my application, where I want to connect several high resolution cameras and download images from all of them
[01:58:38] <jmkasunich> I wonder which connector(s) are the fast one(s)
[01:59:06] <SWPadnos> you can find out by plugging the camera into each and looking at lsusb output each time you move it
[01:59:22] <jmkasunich> I guess I don't really care
[01:59:27] <jmkasunich> if you do I'll do the test
[01:59:31] <SWPadnos> the fast one is #5
[01:59:35] <jmkasunich> there are 4 on the back panel, and 4 on headers
[01:59:55] <SWPadnos> it shouldn't take long - you don't need -v, just look at the Bus number for the camera in the short form list
[02:00:07] <SWPadnos> (it was 001 in the first pastebin)
[02:00:27] <SWPadnos> if you ever see it connected to Bus #5, it's on the high speed hub
[02:01:06] <jmk-robot> bus 1 dev 2 and bus 1 dev 3 are back panel, by themselves
[02:01:21] <jepler> I think the device number increments each time you plug something in
[02:01:33] <SWPadnos> (which is also somewhat crippled because it's a single Transaction Trasnlator model, not multiple trasnlator, which means that multiple slow devices don't get multiplexed into the high speed stream very well)
[02:01:50] <SWPadnos> yes, dev number does go up as you unplug and re-plug
[02:01:50] <jmk-robot> bus 2 dev 3 and bus 2 dev 4 are back panel under the eth jack
[02:02:44] <jmk-robot> so does that mean each pair of back panel jacks is one bus?
[02:02:57] <jmk-robot> seems like I'm gonna run out of connectors way before I run out of busses
[02:03:19] <SWPadnos> 4 sets of connectors, 5 possible buses
[02:03:29] <jmk-robot> I've tested all the regular USB jacks
[02:03:30] <jmk-robot> don't have a header to USB cable handy
[02:03:35] <SWPadnos> ok. thanks
[02:03:44] <SWPadnos> the one that's USB2 is probably labeled actually
[02:04:26] <jmk-robot> J1BU and J4BU
[02:06:17] <jmk-robot> other fine print: "usb flash card" "usb fp hdr"
[02:06:39] <jepler> fp "front panel"
[02:06:42] <SWPadnos> and a yellow audio header :)
[02:06:59] <jmk-robot> yeah, I didn't bother reading around the audio
[02:07:01] <jepler> "flash card" might be the high speed one
[02:07:10] <SWPadnos> the manual says they're both USB2.0 Hi Speed
[02:07:41] <jmk-robot> what a novel concept (manuals)
[02:07:48] <SWPadnos> heh
[02:08:08] <SWPadnos> I don't see the one labeled for cf in the NewEgg photos
[02:08:20] <SWPadnos> I see the FP hdr, but it doesn't say USB
[02:08:27] <SWPadnos> (that's for power/reset buttons, lights ...()
[02:08:37] <jmkasunich> url for newegg pic?
[02:08:44] <SWPadnos> http://www.newegg.com/Product/ShowImage.aspx?CurImage=13-121-359-S02&ISList=13-121-359-S01%2c13-121-359-S02%2c13-121-359-S03%2c13-121-359-S04%2c13-121-359-S05&S7ImageFlag=1&Item=N82E16813121359&Depa=0&WaterMark=1&Description=Intel%20BOXD945GCLF2%20Atom%20330%20Mini%20ITX%20Motherboard%2fCPU%20Combo%20-%20Retail
[02:08:53] <SWPadnos> that's a heckofa URL
[02:09:20] <jmkasunich> in the top view
[02:09:29] <jmkasunich> front panel header is to the right of the ram slot
[02:09:39] <SWPadnos> yes, near a mounting hole
[02:09:43] <SWPadnos> top right
[02:09:48] <jmkasunich> usb header labeled "usb fp hdr" is near the middle of the PCI slot
[02:10:09] <SWPadnos> ah, I see the text now
[02:10:09] <jmkasunich> usb header labeled "usb flash hdr" is near just above the audio header
[02:10:11] <SWPadnos> next to the cap
[02:10:19] <jmkasunich> yes
[02:10:35] <jmkasunich> the header to the left of the audio is S/PDIF (whatever that is)
[02:10:49] <SWPadnos> digital audio
[02:11:22] <SWPadnos> Sony/Philips Digital Format (I may be part of DIgital, or it may be Interchange or something)
[02:11:23] <jmkasunich> there's also s-video (didn't even notice that)
[02:11:37] <SWPadnos> interface maybe
[02:17:37] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/tests/interp/inside-corners/ (README expected test.ngc test.sh test.tbl test.var): new test of cutter compensation
[02:22:22] <jmkasunich> bah, humbug
[02:22:34] <SWPadnos> ?
[02:22:47] <jmkasunich> apparently "stderr" isn't a constant
[02:23:03] <SWPadnos> hmmm
[02:23:09] <SWPadnos> shuold be 2 just about everywhere
[02:23:13] <SWPadnos> should
[02:23:20] <jmkasunich> I want to point a FILE * at stderr, then later optionally open a log file
[02:24:13] <SWPadnos> hmmm
[02:24:36] <SWPadnos> man stderr shows that it's declared as extern FILE *stderr;
[02:24:45] <jmkasunich> yeah
[02:24:50] <SWPadnos> so you should be able to say myfile=stderr
[02:25:10] <jmkasunich> I was trying to say "static FILE *myfile = stderr;"
[02:25:14] <SWPadnos> ah
[02:25:21] <jmkasunich> to ensure that it was set up before anything tried to print to it
[02:25:35] <SWPadnos> you have a main?
[02:25:48] <jmkasunich> I could do it in there, but this code is somewhat of a library
[02:25:55] <jmkasunich> (handles logging)
[02:26:38] <jmkasunich> I have a plan - set the var to NULL, and put if (myfile == NULL ) myfile = stderr; before the two or three places where I actually do writes
[02:26:44] <SWPadnos> if (file==NULL) {dont_print}
[02:27:04] <SWPadnos> heh
[02:29:57] <SWPadnos> cool. the GeForce 8200 is CUDA enabled
[02:37:48] <CIA-2> EMC: 03jepler 07TRUNK * 10emc2/src/Makefile:
[02:37:48] <CIA-2> EMC: this fixes a weird failure mode on chris's ppc
[02:37:48] <CIA-2> EMC: the linker on ppc behaves a little differently when two components define the
[02:37:48] <CIA-2> EMC: same data with extern linkage. -Bsymbolic fixes this, and doesn't seem to
[02:37:48] <CIA-2> EMC: cause any harm on my intel system
[11:59:40] <CIA-2> EMC: 03alex_joni 07TRUNK * 10emc2/src/emc/rs274ngc/interp_convert.cc: fix condition check for M66 E1 Q0 L0 - which should be valid
[12:20:40] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/hal/pyvcp.lyx: minor edit and spello's
[12:26:57] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/Master_Integrator.lyx: remove deprecated vcp
[12:33:23] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/install/compiling_emc2.lyx: add info on live cd
[15:47:47] <tomaw> [Global Notice] Hi all. One of a client servers appears to be having connectivity problems which we're currently investigating. On our more positive note freenode turned 15 yesterday! Happy birthday everyone and thanks for using freenode!
[16:41:15] <skunkworks> logger_dev: bookmark
[16:41:15] <skunkworks> Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emcdevel/2009-01-30.txt
[17:42:47] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/ (Submakefile docs.xml index.tmpl index_fr.tmpl): french translation - adding recent mods
[17:42:49] <CIA-2> EMC: 03tissf 07TRUNK * 10emc2/docs/src/examples/ (pci_parallel_port_fr.lyx misc_fr.lyx mpg_fr.lyx): french translation - adding recent mods
[17:44:31] <tissf> hi all
[17:44:45] <SWPadnos> hi
[17:45:04] <jepler> net xstep => parport.0.pin-02-out
[17:45:04] <jepler> setp parport.0.pin-02-out-reset 1
[17:45:04] <jepler> net xdir => parport.0.pin-03-out
[17:45:09] <jepler> net xstep => parport.0.pin-08-out
[17:45:09] <jepler> setp parport.0.pin-08-out-reset 1
[17:45:09] <jepler> net xdir => parport.0.pin-09-out
[17:45:15] <jepler> this looks plausible to me
[17:45:40] <SWPadnos> yep, that looks good to me
[17:45:48] <SWPadnos> cross off one hunch :)
[17:46:07] <cradek> same version?
[17:46:20] <jepler> 2.2.something here
[17:46:24] <SWPadnos> good question
[20:37:17] <cradek> jepler: could you look at the initial maxvelocity zero thing again? (sorry)
[20:42:27] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gcode/main.lyx: minor edits
[20:42:48] <jepler> cradek: nooooooooo!
[20:44:05] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/User_Concepts.lyx: add info on feed rate
[20:48:41] <CIA-2> EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/common/user_intro.lyx: fix floating figure and minor things
[20:52:07] <BigJohnT> * BigJohnT runs and hides before cradek asks me to look at it :)
[20:57:36] <jepler> cradek: try this little change:
[20:57:36] <jepler> proc setval {vel max_speed} {
[20:57:37] <jepler> - if {$vel == $max_speed} { return 1 }
[20:57:37] <jepler> + if {$vel >= $max_speed} { return 1 }
[21:24:08] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/share/axis/tcl/axis.tcl: another fix for the maxvel slider sometimes starting at zero
[21:27:08] <cradek> * cradek checks the calendar carefully
[21:27:10] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/emctop.py: get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:27:12] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/extensions/emcmodule.cc: get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:27:13] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/task/taskintf.cc: get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:27:16] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/nml_intf/emc_nml.hh: get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:27:17] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/motion/motion.h: get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:27:21] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/kinematics/ (tc.c tc.h tp.c): get per-axis dtg into the stat buffer. presenting it sanely is left as an exercise for the student.
[21:32:05] <alex_joni> whee
[21:32:14] <alex_joni> it seems deadlines are really good :)
[21:32:45] <cradek> I don't know how to show this information without it looking bad
[21:33:03] <alex_joni> per axis dtg?
[21:33:09] <cradek> yes
[21:33:13] <alex_joni> expose it in halui
[21:33:17] <alex_joni> then pyvcp who needs it
[21:33:21] <cradek> no
[21:33:26] <cradek> that's terrible
[21:33:48] <jepler> cradek: did that maxvel change have any effect?
[21:33:55] <alex_joni> bbl
[21:33:57] <cradek> jepler: yes, I think it fixed it
[21:34:11] <cradek> I don't know how to be sure - it comes and goes
[21:35:29] <cradek> jepler: when I click the Preview/DRO tabs, my MDI field loses focus. It doesn't lose focus when I click other things. Can it be fixed?
[21:35:53] <jepler> I bet it was fuzzy thinking on my part
[21:38:30] <CIA-2> EMC: 03jepler 07TRUNK * 10emc2/share/axis/tcl/axis.tcl: don't lose focus on mdi due to switching between plot/dro tabs
[21:38:42] <cradek> yay, thanks
[21:44:59] <cradek> if I have a tuple t and a function f, how do I spell (mapcar 'f t) in python?
[21:46:45] <jepler> [f(i) for i in t] OR map(f, t)
[21:47:18] <jepler> >>> [x+1 for x in (1, 2, 3)]
[21:47:18] <jepler> [2, 3, 4]
[21:47:24] <cradek> thanks (I should know this)
[21:48:54] <jepler> np
[21:54:17] <SWPadnos> how do you spell "(mapcar 'f t)" in english?
[21:54:42] <cradek> make a new list by calling f() with each thing in the list t
[21:54:50] <SWPadnos> ah, ok
[21:54:58] <SWPadnos> I kind of got that from the python result
[21:55:01] <SWPadnos> that's lisp?
[21:55:05] <cradek> yeah
[21:58:13] <jepler> cradek: are you going to write an emc gui in lisp? you have over 30 hours left to do it, assuming january has 31 days.
[21:58:41] <cradek> um, nope
[21:58:47] <cradek> but I'm going to crap up AXIS some more
[21:58:49] <jepler> oh.
[21:58:50] <jepler> poo
[21:59:04] <SWPadnos> you could start it (with a GPL header only) and then count any changes as bugfixes
[22:08:22] <cradek> http://timeguy.com/cradek-files/emc/dtg.png
[22:09:38] <SWPadnos> hmmm. I think that got hit with the ugly stick
[22:09:43] <SWPadnos> or at least the "busy" stick
[22:10:34] <SWPadnos> otherwise, great job! :)
[22:10:38] <cradek> yeah, it's awful
[22:10:56] <cradek> and, what about lathes
[22:11:28] <SWPadnos> oh. ewww
[22:11:36] <SWPadnos> I was thinking it was a 9-axis config
[22:11:41] <cradek> yes, it is
[22:11:45] <SWPadnos> oh, good
[22:11:59] <SWPadnos> then lathes aren't as bad since they have fewer axes, and are therefore less busy
[22:11:59] <cradek> but, there are bunch of special cases for the dro to handle - one is lathes that show rad/dia
[22:12:10] <SWPadnos> LAEFTR
[22:12:19] <cradek> ?
[22:12:25] <SWPadnos> left as exercise for the reader
[22:12:31] <SWPadnos> :)
[22:18:05] <SWPadnos> wow. weird
[22:18:35] <SWPadnos> I just saw the bottom of an invoice on my desk, and couldn't for the life of me remember where I had just spent $19000+
[22:18:55] <SWPadnos> luckily, when I looked at it, it turned out to be stuff that I hadn't bought
[22:47:20] <CIA-2> EMC: 03cradek 07TRUNK * 10emc2/src/emc/usr_intf/axis/scripts/axis.py: add per-axis dtg to BIG DRO
[23:32:37] <alex_joni> guess Eric got a nice clear view of what it takes to build a rtai kernel