KimK is now known as KimK_brb
jmkasunich: I hope you'll weigh in on the thread 'standard signal name in stepconf'
my reflexive reaction is to agree with you about droolers who can't write a net statement
but Chris probably has a bit of a point as well with standardization
except that it won't really make a difference if I'm trying to help a user, because I won't know anything about the stepconf "standard" signal names
if classicladder did the same thing with the same justification, you couldn't connect anything
you still have to know pin names for outputs if you need to invert outputs
I didn't read the original mail closely
and we very seldom change pin names, so who cares
is he naming the signals after the physical pin, or the function (like step, or limit, or whatever)?
after the physical pin, I think
"since you used stepconf from emc 2.3.x connect signal pp2digitalin2
to widget x"
oops excuse the terrible paste
I wish AXIS would remember the file loaded at shutdown and then reload it at startup
instead of the splash g-code?
or maybe a 'recently opened' thingy on the file menu
I'm torn because I wouldn't want to lose the venerable splash screen
it's the first thing many people cut, and it's pretty decent for that
I've been thinking a 'recent' menu would be nice
but not enough to code it
where is non-volatile info like a recent list usually stored?
I'm thinking of my case where I have lathe and mill configs - it would be nice if each config had its own recent list
AXIS already has droppings, see ~/.axis*
EMC: 03cmorley 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/stepconf.glade: Fix omission of out option in parport direction ( in/out ) combobox
yay, I think I fixed the remaining case (I knew of) that my concave cutter comp algorithm got wrong
[03:26:12] <cradek> http://timeguy.com/cradek-files/emc/trouble-case-fixed.png
it was this corner
"h" shape going right to left, line tangent to the arc
(the arc preceding is tangent to the "h" arc too)
EMC: 03cradek 07concave_comp2 * 10emc2/src/emc/rs274ngc/interp_convert.cc: fix corner case for tangent arc->line ("h" shape)
EMC: 03cradek 07concave_comp2 * 10emc2/src/emc/rs274ngc/interp_convert.cc: oops, debug print
this is a real "guess what the user wanted" case, yuck.
[03:42:13] <cradek> http://timeguy.com/cradek-files/emc/guess-what-the-user-wanted.png
jepler, note that the CVS server seems to be down
KimK_ is now known as KimK
just succeeded in getting my mesa 7i43 to talk to emc :D
whats up today in your part of the world
hmm.. actually looking at some emc2 code for a change :)
I must say, I'm a bit rusty here and there
oh well I'll just go back to the manuals perhaps it will come to me what is wrong later...
pastebin it, I might spot something
ok, it's pretty small driver do you want the whole thing?
[14:15:18] <BigJohnT> http://pastebin.ca/1299616
look at line 293 or so
ok.. what's the issue?
I get an at speed at the start I assume it is because the speed commanded is still 0 as is the actual speed at that point
hmm.. you set old_speed_cmd = * .. speed_command
hmm forgot to delete that try
hmm, I might need to just get a fresh copy and start over now that I understand more about what I didn't know about it :)
BigJohnT: wanna swap debugging with me?
I'm sure you don't :)
well at least mine half way works now :)
thanks for taking a look at it
sure .. a new set of eyes most always helps
cradek, jepler: any of you around?
jepler: that TxxM6 bug is evil
alex_joni: oh, the mdi one?
well... not your fault ;)
I went and reported it on sourceforge instead of shrugging and moving on
I got it to a point where it "sometimes' works right
ugh, that's the worst spot to be in
but why it works right "sometimes" sure beats me
it makes you never trust the solution that seems to work "all" the time, either
well.. the problem is that tool_prep has waiting_for_io as a precondition
but it should also have it as a post condition
hmm.. no, maybe that's not right
apparently it's a bad hosting day all 'round
07:40:07 up 3 min, 1 user, load average: 1.98, 0.90, 0.34
now it's up to 4.56
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/ (m5i20.lyx GS2.lyx): move to own file
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/hal/ (basic_hal.lyx drivers.lyx): moved m5i20 to seperate file
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/GS2.lyx: add drivers files to html and integrator manual
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/ (Master_Integrator.lyx Submakefile docs.xml index.tmpl): add drivers files to html and integrator manual
"The Perl Foundation has announced they are switching their version control systems to git."
ugh, I had no idea perl was in "perforce", a proprietary SCM
ah, their switch to perforce was part of activestate's efforts to build a business model around free software scripting languages (perl, python, tcl/tk)
before that, plain rcs? inconceivable
you keep using that word
I do not think it means what you think it means
holy crap, this is a beast: http://www.hitechglobal.com/Boards/V5_Multi_LX330.htm
I suspect that should hold enough IO's for any retrofit :D
[19:34:19] <cradek> http://pico-systems.com/images/vel48.png
(emc1: going around some cirles I think)
I remember there was a time (couple years ago), when people were very keen of posting gnuplot screenshots on the emc list :)
do you remember what symlink caused the hostmot2 hotplug on boot?
[Global Notice] Good Evening Ladies and Gentlemen -- it's time for a somewhat late christmas pantomime.. I need to do a spot of rehubbing, it shouldn't take too long and you're welcome to point and laugh if it all breaks! Apologies for the inconvenience and have a nice day!
n/m I found it
[19:53:06] <alex_joni> http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2008-09-11.txt
I thought it was already fixed in an rtai package update .. but I don't remember
I don't think so
* alex_joni looks
[20:04:17] <alex_joni> http://git.unpy.net/view?p=rtai.git;a=summary
I agree with jepler - I think it is already fixed
yeah, I had the problem, and (mumble) update fixed it
I still see this: ln -sf ../../../../usr/realtime-$(KERNEL_VER)/modules debian/tmp/lib/modules/$(KERNEL_VER)/rtai
in the latest source
maybe it's not the ln I think about ..
hmm.. are you on amd64?
there's a slightly different package for amd64
rtai_3.6.1-linuxcnc.2 vs. .3
hm, I don't see anything relevant in the changelog (for the ones I have installed)
I still see the link in the latest deb
it's /lib/modules/2.6.24-16-rtai/rtai -> ...
I know I talked about (asked for help with) this in irc
it affected my lathe
but I can't find the log
hmm.. I only found the log that we should remove it
maybe I just removed the symlink.
sorry for the derail. I don't remember.
[21:59:08] <alex_joni> http://git.unpy.net/view?p=rtai.git;a=commitdiff;h=aba15223f6890a3b353954e84f2e5dfd12cab50d
that's jeff's commit that enabled it
I'll rebuild and push a new i386 package
I do have that changelog entry
are you around a machine on which you can test the new one?
yes but I'm not having the problem now, for some reason I don't remember
I'm happy if the mesa config still runs
new rtai package pushed
can someone else check if all is still in order?
looks that way here..
Need to get 39.3MB of archives.
what do I do to test it? reboot? and look for what?
after update, it boots and runs
I'm not sure how to do a full test for you
boots & runs is enough for this
goodnight then :-)
the only thing that could have happened is for the build to cause broken modules (_really_ unlikely)
say goodnight Gracie
BigJohnT: goodnight Gracie
good night all
meh.. work again tomorrow :/
EMC: 03cmorley 07TRUNK * 10emc2/src/emc/usr_intf/stepconf/stepconf.py: Remove generation of signals for Parport 2 and 3. Fix the omission of writing parport 2 and 3 's output in the base thread. Fix address of parport 2 and 3 not being updated from basic page