what kind of noise does a unix pirate make?
fjungclaus_ is now known as fjungclaus-away
03cradek 07v2_1_branch * 10emc2/tcl/bin/pickconfig.tcl: wording
03cradek 07TRUNK * 10emc2/tcl/bin/pickconfig.tcl: wording
03jepler 07v2_1_branch * 10emc2/tcl/bin/tkbackplot.tcl: backplot should expand when window is resized
03jepler 07TRUNK * 10emc2/tcl/bin/tkbackplot.tcl: merge rev 220.127.116.11: backplot should expand when window is resized
03jepler 07v2_1_branch * 10emc2/tcl/bin/emccalib.tcl: fix the case where the ini value is used to set a pin, not a parameter
03jepler 07TRUNK * 10emc2/tcl/bin/emccalib.tcl: merge rev 18.104.22.168: fix the case where the ini value is used to set a pin, not a parameter
03jepler 07TRUNK * 10emc2/src/hal/utils/halcmd.c: make gets work on pins as well, for symmetry with getp
03jepler 07TRUNK * 10emc2/src/hal/utils/halcmd.c: restore line that was accidentally deleted
03cradek 07v2_1_branch * 10emc2/tcl/bin/emccalib.tcl: remove debug statement
03cradek 07TRUNK * 10emc2/tcl/bin/emccalib.tcl: remove debug statement
03jepler 07TRUNK * 10emc2/tcl/bin/emccalib.tcl: "halcmd getp" now gets pin values so use it directly instead of parsing "show" output
cradek: thanks for the reminder - I saw a debug print from halscope, gonna fix that before I forget again
I do love to leave little turds everywhere
printf this and puts that
it's a terrible habit
good night guys
I think I just removed another jepler dropping
scope_vert.c: printf("change_page: %d\n", page_num);
03jmkasunich 07v2_1_branch * 10emc2/docs/man/man9/sim_encoder.9: fix typo
03jmkasunich 07v2_1_branch * 10emc2/src/hal/components/siggen.c: siggen control pins are inputs, not i/o
03jmkasunich 07TRUNK * 10emc2/docs/man/man9/sim_encoder.9: fix typo
03jmkasunich 07TRUNK * 10emc2/src/hal/components/siggen.c: siggen control pins are inputs, not i/o
03jmkasunich 07TRUNK * 10emc2/src/hal/utils/scope_vert.c: remove debugging print
03jmkasunich 07TRUNK * 10emc2/docs/man/man9/encoder.9: implement improved velocity output (much lower jitter, especialy at low speeds), also ensured that data transfer from the fast update() function to the slow capture() one is atomic
03jmkasunich 07TRUNK * 10emc2/src/hal/components/encoder.c: implement improved velocity output (much lower jitter, especialy at low speeds), also ensured that data transfer from the fast update() function to the slow capture() one is atomic
03cradek 07v2_1_branch * 10emc2/docs/man/man9/ (9 files): many spelling/grammar/typo fixes
copy those fixes to TRUNK.....
working on it
nag nag nag
I'll thank you later for the nagging
03cradek 07TRUNK * 10emc2/docs/man/man9/ (9 files): spelling/grammar/typo fixes
that was fun - I used cvs diff -D"hour ago" in the branch
I committed stuff less than an hour ago
stuff that was already in TRUNK
I don't think I screwed anything up - I checked the diff before I committed
manpage blindness setting in - goodnight
wheeeeeeee wrote my first non Hello World Palm program!
03cradek 07v2_1_branch * 10emc2/src/hal/components/ (comp.comp hypot.comp): spelling fixes
03cradek 07TRUNK * 10emc2/src/hal/components/ (comp.comp hypot.comp): spelling fixes
Jymmmm is now known as Jymmm
SWPadnos: $450 http://www.pixelmagicsystems.com/products/media_players/hd_mediabox.htm
Hey, they're in SJ!
lol 192khz dac and crystal oven
in case you need to broadcast packet radio from your media player?
thru the audio port
well streaming, yeah.
"We are making the transition to the GPL license. " <- is this still true?
can i change it to say "EMC is entirely GPL and LGPL code." ?
From a legal perspective I'd say you couldn't.
There was/are questionable items that didnt' have the full "release"
that is a rumor that i'm trying to dispel, which is why i'm asking
It's not a rumor.
then provide something specific
The was code submitted, but no license/perms were "official" given, that I know of.
if it's not gpl or lgpl it doesnt belong in cvs IMHO
I don't have to, only a full code audit would do that.
you are parroting the exact thing that's bugging me
Well, talk to the board members
Not trying to be difficult, just would rather take the safe route
if it's causing would-be developers to shy away from the project and make their own cnc control from scratch, it's a big problem
how many have questioned this?
i didnt know it was an issue at all until erdizz mentioned it (or whatever his name was)
when i asked him why he wrote mync
fenn check the irc logs for the last couple of years.
it's been a questionaable issue that long
people still believe the earth is flat, too
why are you aiming your frustrations at the messenger?
you arent saying anything helpful
Ok, fine. go ahead and make it say anything you want it to.
sigh.. guess i'm going to have to sign up to the mailing list again
Just this once .. here's the log: http://www.linuxcnc.org/irc/irc.freenode.net:6667/emc/2007-01-27.txt
fenn: you there?
Good morning (mine) alex_joni.
replying to your email
I just read fenn's email and as you saw, responded. I counted the number of files in emc2 -- about 3000.
lerman: I've been over them a few times
and the other board members too
A bunch of them have no license info -- eg.*.ini
lerman: I start to think that removing the license info from all is better, and keep a generic license on the whole package
The only way to put it to bed is to have a *formal* auditing process. -- That means with a document.
most the OSS software I've seen does it like that
lerman: that is ok, but will probably be based on TRUNK
I don't think configuration files need licenses
The problem with the generic license is the history. Someone can always pop up and say: Joe Blow wrote that piece of software and I don't remember the license being there.
I didn't quite understand that, can you expand?
Why not? If you copy a file, you need a license to copy it.
why would you?
expanding... There may be historical pieces that didn't have licenses with them. Some of that code is very old.
the code produced by NIST was released as PD
People copy the .ini files in the process of downloading them. The act of creating a file creates a copyright (under modern law).
that means anyone can take EMC1 (possibly minus 10% that was added later) and do whatever they want with it
To dispell the myth (if that's what it is), we should be able to define the status of each file.
I have no problem with dispelling the myth, if there even is one
as far as I'm concearned all the licensing problems have been taken care of
Yes. If it comes from NIST. So, each of those files should have a header saying that it was originally PD and the changes are GPL.
lerman: that is valid for emc1 only
But there is a myth... (erDIZz seems to believe it).
I am not going to track down all authors that commited code on emc1! and ask tehm under what license they implied their code/additions to be released
emc2 on the other hand started from a clear premise: GPL or LGPL code only
EMC2 has a variety of authors. We should have a firm policy and enforce it that any author who is given the right to commit to emc "signs" a document GPLing anything committed.
well there is one, it's just not formal
If it isn't in writing... it isn't.
ok, now taking care of the myth
fenn: hope you're reading too :)
I know of 2 licensing problems in the last year
Perhaps the board should send out a letter to each of the authors who ever commited asking them to reaffirm that everything commited is GPL.
1. Paul made some comments about copyright infringements (referring to: a. 5i20 driver/work, b. genhexkins.* c. debian/*
these issues have beeen analyzed and confirmed as non-issues
I recall there was something about the mesa driver (or microcode) that was later resolved.
the Mesa work was GPL, and the rest of the work which is licensed by Pete Vavaroutsos (is also GPL)
b. the genhexkins were in fact written by Brian Register, but he was an intern at NIST when he wrote the stuff so it falls under PD, and moved to emc2 as GPL with the necessary changes
lerman, I just beleive that I don't have enough information
c. the fact that Paul's name didn't appear on the debian/* files was an overlooking on my part, and it has been resolved
the second even was more recent when Brett Smith (FSF Licensing Compliance Engineer) mentioned some licensing is problematic
I'm convinced. But if I were a manager at a large company wanting to integrate EMC2, I might be a little concerned by the lack of documentation.
What did Smith have to say?
we had some license info stating GPL v2.1
there is no such thing, and it was simply a bad paste
Is that really a title Licensing Compliance Engineer? Is that really an engineer?
lerman: he signed the email like that
Licensing Compliance Engineer, Free Software Foundation
we took care of the issue, notified him, and didn't hear anything else about other problems
lerman: to summarize: 1. I agree we can make an audit (it will probably take time, and the result is clear to me), but I agree that we should do it if people think it helps.
#2. there are no issues regarding licensing in emc2 that I (or any other board member) know of at the moment
I just did a quick scan of files looking for GPL. Of 92 *.cc; 86 have the string "GPL" in them.
Of 160 *.c; 96 have GPL
Of 114 *.h; 56 have GPL
I'm not sure if that proves anything :-)
$ grep GPL /usr/share/common-licenses/GPL || echo "not found"
i'd say it doesn't prove anything
scanning /usr/include in the same fashion shows that it is not quite common to make (L)GPL statements in .h files
erDiZz: that is correct
and if you look at the linux source tree, bet you'll find files with no license on them
* erDiZz experiences strange disconnects
And I suppose there could be some files containing the string: "This is not GPL"
lerman: not in emc2
erDiZz: may I ask you why you think licensing is not clear with emc2 ?
alex_joni, I don't "think", I've just read a statement by something else. I gave you a link to that place
not me :) you talked to fenn I think..
[14:33:37] <erDiZz> http://www.tuxcnc.org/pivot/entry.php?id=5#body
[14:33:41] <erDiZz> http://www.tuxcnc.org/pivot/entry.php?id=7#body
right at the bottom of both pages
erDiZz: I am sorry that site confused you
it didn't have any impact on me :)
I am sorry to admit that is an previous core developer of emc2 who forked the project, and it seems he didn't get quite well over the break
imo (as a board member) that is just FUD
there doesn't seem to be any traces of that fork except that site
and yes, that guy isn't accurate event a bit
erDiZz: he doesn't seem to want to be linked to emc2
I see that from what he'd written about mync
survival of the fittest I guess... only time will tell if the fork gets anywhere
actually the stuff he "talks about" at tuxcnc is not the fork of emc/emc2 I'm talking about
erDiZz: he has another project called BDI (brain dead install), which was a full OS with rt patches, and emc on it
he now has the same thing using some emc1/emc2 crossversion (libnml as in emc2, the rest of the code is based on emc1), compiled on 2.6 sarge
I'm still using a BDI.
lerman: and I'm not blaming you
that's not a bad idea to have such a thing
you're free to do whatever you like
erDiZz: actually it is
erDiZz: I see no reason to distribute/support a full OS
I'll convert to EMC2 sometime -- when I have to.
we (the EMC/emc2 community) distribute debian packages compiled and tested for Ubuntu
How does the BDI differ in principle from Ubuntu.
there is also a Live-CD with emc2 & RT kenrel and all you need on it
alex_joni, that's quite limiting
the difference is that we do not support the distro itself, but only our packages on it
if one has to have a ubuntu install than he's restricted by its general-purposeness
erDiZz: limiting? how so_
erDiZz: I don't see why that is more limiting than a Sarge install?
it's about embedded software... general-purpose desktop distro doesn't fit into the picture very well
erDiZz: you are free to take emc2 and compile it on another system
we try to make it work on as many systems as possible
but if BDI is about somthing based on Sarge then, well, you're right
To me, the gotcha is the RT kernel patches.
emc2 has a subsystem called RTAPI (which is a layer of abstraction over RTAI and RTLinux, so you can use both)
easily one could add Xenomai if needed, or you could add your own kernel hacks :D
erDiZz: the idea is that we keep emc2 as generic as possible, but we chose one platform we like to offer support on
I like my kernel hacks. but I don't think others will like to compile the kernel with them
You can't just "install" emc. You need to do some kernel stuff; and not everyone is comfortable building a kernel.
that's why I'm about to provide somthing that is ready-to-boot
lerman: you can just install emc on Ubuntu
apt-get install emc2 is all you need
Is that because Ubuntu already has the RT kernel stuff?
kernel packages will be installed automatically for you
no, this is because we have kernel packages in the same repository as the emc2 package
they are simply installed as dependencies
users don't need to compile anything
Do other common distros besides Ubuntu support kernel packages?
we had to select some linux distribution as the one for which we would create and distribute kernel images -- we chose ubuntu
wth is a sloccount?
A-L-P-H-A: counting sourcecode lines
Having packages in the main repository is a sign of maturity. I always appreciate that when deciding what software to use.
lerman: there are other distros that use kernel packages, but imo the Debian system is the most mature/easy to work with
we have some contributed kernel+RTAI RPM's for FC3 I think
but afaik it's a pain to properly ensure dependencies on FC for something like this
I do have a Ubuntu system running simulation only so that I can make interp changes.
switching from that to RT is a matter of 10 minutes
just out of curiosity
did you try my livecd?
erDiZz: nope :)
oh, well. the point is that I'm not sure about its bootability
I've tested it on only five quite similar boxes or so
can you give me a link?
erDiZz: from looking at mync.sf.. I like your editor
it would be nice if you could distribute that as a separate app :)
it may look fine, but it's not that functional
well - fix it! :)
and my g-code dialect is quite limited, so I don't want to put too much work into my interpreter
I thought about reusing the interpreter that you use in EMC
erDiZz: I see no issues with doing that
have you ever seen/heard about Mach2/3 ?
alex_joni, mee too. I'll hack on it when time comes
Mach2/3? is it about a microkernel?
no, it's a commercial NC control based on windows
That's a different MACH.
erDiZz: it seems to be quite popular on the DIY scene
sometimes people trust a software which costs 150$ more than one that's free :)
and how do they do real-time in windows? what do they use for that?
not only is Mach based on the EMC interp, so is DeskCNC, and the interpreter that was put in the G-Rex
anyways, it started with the interpreter from EMC, and adapted that to fit their own needs
he has a driver that takes over the timer interrupt for step generation
erDiZz: some custom kernel driver which makes it work most of the time
it plays havoc with other stuff, in my experience (like winamp and all system timing delays)
it seems impossible to achieve finite response time without hacking the kernel
it works OK, since Windows isn't as picky about what a driver can do
erDiZz: it sometimes happens that you get a handfull of pulses in a bunch with Mach
but they claim it's quite seldom :)
it all averages out, I'm sure :)
yeah, the total count sums up :D
To us, the question shouldn't be what's bad in Mach -- it should be what is good. We should find out what is good and then incorporate those aspects into EMC (where it makes sense).
That's why I'm working on my GWiz subproject.
uh-oh. I hope Robin isn't looking ;)
(but I agree with the principle)
lerman: I tend to disagree :D
we need to keep improving emc2 untill they start looking at what they can incorporate in Mach :D
In general? or is there something specific?
(which I'm sure they already do :P)
API docs on EMC or something like that?
Yes. That too.
erDiZz: depends what you need
alex_joni, say, the G-code interpreter
I look at the question as "why does anyone choose "brand X" over EMC"?
erDiZz: Use the source Luke!
if there's actually something useful, then it should be part of emc as well
erDiZz: joke aside, you can generate some doxygen info from the source tree
and that might be somehow helpfull..
but the interp. is basicly in a folder on it's own
and there are a few members to a class you need to mess with
oh, well. That's why the thing I'm doing at the moment is API docs. People never get the idea from the sole source
the interpreter calls some canonical functions based on the interpreted things
erDiZz: no, but you can always ask in here
the canon calls are in the RS274NGC documentation
ah, I have that paper. just forgot about it
The source of emccanon.cc is pretty readible.
03jepler 07TRUNK * 10emc2/src/hal/ (hal.h hal_lib.c hal_priv.h):
make pin and param direction enum values distinct, and check direction of pins
and params when creating them. the enum change requires a HAL version bump
hi samco :)
good morning alex
or evening : )
hi there (evening here)
I just purchased another lathe for work
oh, sounds nice
Mori Seiki ZL25
2 turrets that communicate to each other through 2 processes
might be 2 controls
[16:36:06] <alex_joni> http://www.emachinetool.com/usedcnc/catalog/vertical.cfm?ProductID=12848
quite a beast :D
yes, like that
interesting how the machining world calls it 4th axis
kind of confusing if you've gotten used to 4th axis on a mill
25 HP spindle motor..nice
It uses a scheme that hands control back and forth to the opposite turret via M codes
I really think that it has 2 seperate Fanuc controls built in
used or new ??
awallin: you don't buy machines like that new :D
I find it interesting how you sync the controls
400k$ I think :)
pretty pricey new
yeah, I would just be hesitant to buy something used... how do you check the ballscrews are ok and the ways etc
I flew out to the shop and ran it
I know that there's going to be a lot of work involved getting it up to our standards
but even if we put $20k into it we're doing ok
It ran fine
but, back to the control...
it syncs up by handing a m100 to the other control
which acknowledges by having an m100 in it's program
and then has a m101 in it's program that hands a signal back to the first control
heh.. so simple 2 bit sync
I should say the controls don't wait either
DanielFalck: are you going to do an emc2 retrofit on this??
no, just thought you might find it interesting how it does two turrets
DanielFalck: how does one control wait on the handover?
just a sec
wife is talking to me...
that happens :)
can do more g code
oh, so the M100 is blocking ?
M101 to get back to head 1
something like that
so just the basic idea of sending sync signals between controls
is interesting to me
you could possibly create a 12 axis control with emc
yeah, that's easy
maybe not coordinated motion between all axis
the problme is with the G-Code
but it could be interesting
when you're at the control of one of these machines, you flip between programs by switching the head1 head2 switch
you have to upload 2 seperate programs
or type in 2 seperate programs
strange hack :D
bet it could have been hidden better from the user :)
but that's how it's done by Fanuc and some others
2 seperate machines sharing one spindle
* Vq^ waves at alex_joni
* alex_joni waves back
* Vq^ is trying to figure out how to add a custom M-code in emc2
talk to you later. gotta go work in my shop....
Vq^: make a script/program whatever, make it executable, name it M1xx and put it in the nc_files dir
DanielFalck: thanks for the info
Vq^: the entry in the ini is called "PROGRAM_PREFIX"
anyone know how to install .egg files in python (winXP) ?
[17:18:35] <tomp> http://peak.telecommunity.com/DevCenter/PythonEggs
tomp: yes, but I get: zipimport.ZipImportError: not a Zip file: 'PyOpenGL-3.0.0a5-py2.4.egg'
now it worked, it seems I have OpenGL now, but still need GLUT
how can this be so hard??
egg for PyOpenGL has been reported to do that http://blog.vrplumber.com/1710
"it was hard to write this. it should be hard to use it"
tomp: yeah, I got over that hurdle by downloading the .zip installation package instead
ERROR: PyOpenGL not installed properly. god dammit
awallin: not nice to cuss :)
neat -- http://www.flickr.com/photo_zoom.gne?id=361611097&size=l
jepler: very neat.. what is it?
jepler: what lighting mode in opengl gives the 'flat' shading that's used in AXIS?
alex_joni: high voltage spark with an LED in the middle
awallin: most things are drawn with lighting disabled -- glDisable(GL_LIGHTING). the cone is drawn with one directional light, I think. see the function 'basic_lighting()' in the main axis script.
jepler: I figured that much, but wondered what the purpose of that might be
alex_joni: no purpose beyond looking pretty
Any thoughts on converting an old 1990 HAAS VF1 to EMC control? Top right corner: http://www.quakercityauction.com/Auction6.html
sure, why not, if you have the time, money and patience
It's got a 20-tool ATC, which, IIRC, wouldn't work too well with EMC.
what control signals do the motors take?
ATCs are fine with EMC2
owad: there's nothing by default in emc2 for controlling an atc, but it's easy enough to program something yourself
no idea. It has a broken HAAS CNC control, but doesn't say which model. I'm just now looking at the brochure that came today.
there's a 24 or 25-position tool changer on the Mazak (at CNC Wrokshop)
I think I may get this Benchman working, sell that, then buy something bigger like the HAAS.
owad: given enough time and patience you can get any machine working with emc2 :D
maybe you should do it like the CNC workshop
organize the thing, and let others pay you to fix your machine </joke>
I'm thinking about coming out to that.
thats a good idea :)
Has emc been used with any addtiive-type machines? like the reprap, or 3d printers
dunno.. maybe :)
I've got a pair of robotic arms I'd like to do something with, and I was wondering about hooking a syringe up to them like the fab@home guys did with their device.
you're asking how many axis, right?
plenty for this :)
DOF= degrees of freedom
It's a Microbot Alpha II+
This one here: http://www.questechzone.com/microbot/alpha_specs.htm
but that link only works in IE, unless you stop the page just before it's finished loading
I just noticed that :D
strong enough to lift a syringe, 18.4" reach, .015" repeatability
certainly good enough for a toy, I figure
it takes BASIC-like commands
alex_joni, have you seen these? http://www.baumerelectric.com/en/shop/catalog/Winkel_und_Positionsmesssysteme/Drehgeber_absolut_Multiturn_BOMH_SSI_Dignalizer/
03cradek 07v2_1_branch * 10emc2/docs/html/gcode.html: g76 and typo fix
on the subject of robot DOFs (and measuring thereof :) )
The page cannot be found
they're 18-bit resolution absolute encoders
and they have multiturn versions (I think 14 bits for the number of revs)
seems the Open Link feature decide to strip the trailing /
they're expensive as hell (in the $1000 range, I think), but very accurate
nice, but too expensive
like most things :(
at least for robots (not sure about other areas) the current way to go is motors with included resolvers
I haven't seen an encoder in years
only on older systems (~1990)
and an external battery-backed revolution counter?
it's all in the servo these days
servo has some battery buffer, and saves position on shutdown
so if the power goes out and the joint is moved, you will still be able to get absolute position on power-up, without moving the joint?
what about motion when off?
if the joint gets moved while it's switched off, you need to resync it
but the brake should prevent that :D
the motors also have a brake inside
ok, so maybe I'll have to think of something that hasn't been done yet :)
active without power
there are a lot of new names today
cradek: plattschnauze is the guy from germany using the puma kins to control his bot
03jepler 07v2_1_branch * 10emc2/docs/src/config/ini_config.lyx: improve discussion of INPUT_SCALE since there is no longer a second number. Remove paragraph that seemed out-of-place.
just in case you missed it from last night ...
what noise does a unix pirate make?
huh eh? lol
any HAL experts lurking? I got a question about linking "halui.machine.is-on" to an out pin on my parport
Ziegler: go ahead ans ask, I'm sure someone will answer
well... Im trying to link the gui machine on button to send a signal to my parallel port
so to hit a really and turn on my machine
cant seem to get the files right
SWPadnos: baumer is nice..
SWPadnos: though I have no idea about pricing
newsig PowerSwitch bit
linkps halui.machine.is-on PowerSwitch
in standard pinout.hal:
linkps parport.0.pin-05-out PowerSwitch
Ziegler: that sounds correct. do you get an error?
seems not to work as I cant load the gui unless I comment those lines out
Ziegler: do you know how to launch emc from the terminal? That way you can see the error message, if one is printed.
oh sure let me try that out.. one menent
did you activate support for halui by adding the lne 'HALUI = halui' in the [HAL] section of your inifile?
let me check
that goes in my stepper_inch.ini file?
let me add it!
ok... errors changed.. but new error: HAL: ERROR: pin 'halui.machine.is-on' not found
I am using EMC2 - 2.0.4
Ziegler: there is no halui.machine.is-on in 2.0.4
you probably want to use axis.N.amp-enable-out
but there is a axis.0.amp-enable-out
alex_joni: are you sure? I see it in this version, which CVS says was the one in all the 2.0.x versions. http://cvs.linuxcnc.org/cgi-bin/cvsweb.cgi/emc2/src/emc/usr_intf/halui.cc?rev=1.6;content-type=text%2Fx-cvsweb-markup;only_with_tag=v2_0_branch
that is what you should use to enable your motor drivers, etc
so.... linkps axis.0.amp-enable-out PowerSwitch
jepler: not so sure..
but it sure isn't started at that time
oh.... let me try it and see what happens
jepler: halui got started way later in 2.0.x
I think way after the HAL files were run
so what is the latest stable of EMC2 now?
2.0.5 right now
(that worked cradek)
but this will change in a few days
I verified that in 2.0.5, halui.machine.is-on exists, dunno about the order it is started though
I love this software
Ziegler: good to know that :)
I have much to learn...
but I am enjoying it
03alex_joni 07v2_1_branch * 10emc2/debian/changelog: use 2.0.x changelog along with the newer features
hmmm now I am getting a joint overide error
when I try to jog
joint following error?
are these steppers?
joint 0 following error
yes they are
Ziegler: you are trying to get a velocity/accel which is too great
no, I bet you disabled stepgen.0 by linking axis.0.amp-enable-out to something else
oh.. did it work before?
that was already hooked to a signal "Xen"
yeah it worked a miunte ago
(excuse the typing and spelling)
ok, then it's as cradek says
you want 'linksp Xen parport.0.whatever'
you can see the Xen stuff in core_stepper.hal
what is Xen?
the Xenable signal
that did it
is it serving another purpose in the file?
it also enables the step generator
when the machine is switched off, or in estop the step generator is disabled
on servo machines that happens to pid and dac outputs
now... I dont need exact details... but I should be able to do this the other way around yes? such as open a physical switch and have the machine power in the software open?
you can control ESTOP from the outside
iocontrol.0.emc-estop-in is the pin iirc
ok cool... what about the play and pause gui buttons?
there should be a linkpp in your file
you mean as hal pins?
those will be available in 2.1.0
in a couple of days
just have to figure out how to dcc the dev team massive amounts of beer now ...
* Ziegler starts pouring into the obsolete floppy drive
send concentrate, we have lots of water available
anyone working with OpenGL: is there a limit to how far away the camera can be placed? I seem to experience such a limit...
awallin: when setting up a projection matric with gluPerspective, you specify 'near' and 'far' planes. items beyond those planes are not shown
jepler: yeah, that might be the problem. thanks.
yep! that was it
now I can read in an STL file, and I can start testing the z-slicing algorithm
it's desirable not to have zNear and zFar include too much 'empty space', because that decreases the accuracy of zbuffering
guess it can be set a bit farther than the bouding box of the objects
go dammit this SPE python editor. keeps crashing
anyone know a good book about UI design? how to organize a model-view-controller kind of program
even for simple things I find my ad hoc program is a mess and need serious re-structuring
03cradek 07v2_1_branch * 10emc2/debian/changelog: more changes documented
Some of the Smalltalk books did a good job of MVC description, I think.
alex_joni: we'll never know if we have everything :-)
right.. we need a real cattle prod
and start prodding people that improve emc2 and don't write it to docs/NEWS
I'm gonna make a motion on that
that's all of us
the cattle prod motion?
good luck finding a second
hmm.. how about a small buzzer in the computer chair?
I'll second it
thanks John :)
maybe Ray and Jon vote in favour too :D
[21:45:09] <cradek> http://www.youtube.com/watch?v=5blbv4WFriM
*nearly* funny in places
[21:47:43] <jmkasunich> http://www.youtube.com/watch?v=WofFb_eOxxA
honey, why is the water bill so high?
haha "be quiet, I wanna put it on the web"
[21:50:19] <alex_joni> http://www.youtube.com/watch?v=n4v9PJZSygM&NR
cradek: was that in the audio?
yeah at 01:54
[21:50:46] <robin_sz> http://www.youtube.com/watch?v=NPVlljVWqBg&mode=related&search=
hahaha.. they have mass youtube comments now
"please don't read this!
In 1945,a young girl named katu lata kulu came over to America in a grey boat from Africa. A mysterious man killed her by cutting the word "LATUALATUKA" into her back. now that you have read this measge she will come to your house on a full moon and steal your soul unless you follow these directions:
1. Retype this message as a comment for three other videos"
and the other two things?
you'll never know HAHAHAAHAHA
I played with the welder today :)
you shouldn't play with your welders
play with your wife instead :D
* robin_sz ponders that
for all you know, my wife might be a welder ;)
210cm 102kg, tattoos and piercings
so many possibilities for innuendo
Slow day on #emc today :).
I havent been around
Ok, does that mean that now things are really slow <gs&r> :-)
c or v?
it could be slower ...
xargs - build and execute command lines from standard input
xargs [-0prtx] [-E eof-str] [-e[eof-str]] [--eof[=eof-str]]
[--null] [-d delimiter] [--delimiter delimiter] [-I replace-str]
[-i[replace-str]] [--replace[=replace-str]] [-l[max-lines]] [-L
max-lines] [--max-lines[=max-lines]] [-n max-args]
[--max-args=max-args] [-s max-chars] [--max-chars=max-chars] [-P
max-procs] [--max-procs=max-procs] [--interactive] [--verbose]
[--exit] [--no-run-if-empty] [--arg-file=file] [--version]
[--help] [command [initial-arguments]]
like my 200GB transfer from an ata drive on a 40-pin IDE cable to a RAID-1 paired on the same chanel ...
* ejholmgren winces
ejholmgren; ah, yummy
fjungclaus_ is now known as fjungclaus