[02:06:44] <jmkasunich> http://www.codinghorror.com/blog/images/the-only-valid-measurement-of-code-quality-wtfs-per-minute.png'
[02:06:59] <jmkasunich> http://www.codinghorror.com/blog/images/the-only-valid-measurement-of-code-quality-wtfs-per-minute.png
dunno where that ' came from
I can write about 1kwtf/day
thats pretty good
I can't type that fast
damn that's annoying - every day my UPS tests its battery (right about now)
and for the last two days it has decided that the battery needs a new one
so it beeps (loudly) for a minute or two
anybody seen this? http://lwn.net/Articles/319544/
todays's halscope testing: http://filebin.ca/njowyj/02mar_2.2.8.txt http://filebin.ca/tegzcp/02mar_2.2.6.txt http://filebin.ca/owdcbg/02mar_2.2.8-binary-update.txt
so the short version is that halscope (binary, compiled, any version whatsoever) is now totally borked on that machine?
that is disturbing
i'm convinced it has more to do with my machine/sw than with halscope though
unless i'm the only one running halscope on hardy
I only have one hardy machine here, but I did use halscope on it a bit
never saw anything funny
I use it on hardy, but sim
cradek uses it on hardy (his lathe)
rob needs to fix his internets
jmkasunich: that page is interesting, but so is this: http://sos-linux.svn.sourceforge.net/viewvc/sos-linux/offsched/trunk/Documentation/OFFSCHED.pdf
(linked from that page)
Guest865 is now known as skunkworks_
SWPadnos: how was the vacation?
record cold temperatures, but still better than here :)
how was yours?
I'm headed to Vegas tomorrow morning
70's and sunny
that's slightly better than most of the time in Hawaii :)
did you guys get out to the Hoover Dam?
yes - hoover dam - red rock canyon and the south rim of the grand canyon.. (we dropped the ball on the grand canyon - it was 45deg and we didn't dress for it. Lots of trips to the car ;)
did you take the "Hard Hat Tour"?
(at the dam)
no - we were with the reletives and not everyone wanted to.
the regular tour is pretty good too
I did do the 'big shot' or whatever it is called at the top of the stratosphire.
hmmm. don't know that one
I don't usually do much in Vegas itself, except go to trade shows
we went to a lot of the casinos to see what they had. (the last place we went was paris - that was probably the one we liked the most)
huh. never been in that one
Did you get to the Bellagio?
(plus watching the balasio(sp) water show from the ifle tower.
ah yes, that's a nice show
[17:46:55] <skunkworks_> http://imagebin.ca/img/zF4TSFr.jpg
[17:47:21] <skunkworks_> http://imagebin.ca/img/mngnnRy1.jpg
I have to stitch togather a few of the grand canyon pictures
that first one is 8 frames stiched together?
I don't know off the top of my head - but probably pretty close to 8
I think I saw 7 seams :)
but it could have been 8 seams
I was suprised at how well it worked - it is the software that came with the neighbors cannon
ok. I was going to ask what you used to do it
the hard part with that is getting the exposurethe same
took me a second to figure out the software couldn't automagically put them in order.
you have to check the settings for all the angles you want, then figure out what will work the best for all of them, then use manual settings and shoot them all the same way
SWPadnos: there are software stitiching packages out there that do that for you
I know the photoshop I use at work does
well, I like to shoot things "on film", since softwaer can't actually add any data
so if you need to do any editing to match the exposures/densities, you're probably losing information
so how do you barell distortion compensation?
using a formula ;)
and tiny scissors?
but you aren't adding any information, you're only moving information around so it looks better
in some cases, you duplicate information (take one source pixel value and put it in multiple destination pixels)
in others, you average pixel data, losing information (or at best keeping the same amount of information)
SWPadnos_ is now known as SWPadnos
heh - funny: http://portal.acm.org/citation.cfm?id=359060.359074
isn't that the name of your former/actual company?
yes, that's what's funny
I met a guy in Hawaii who mentioned it
I gave him my card (so we could get together later), and he said "Thoth was first used for a realtime OS ..."
I mentioned that the ancient Egyptians may have preceded that by a little :)
* alex_joni likes seshat better
wonder if there's a non RT OS called Seshat .. that would be ironic
interesting. almost the same set of "skills" as Thoth
it's the female counter-person
yep - I guess Thoth took over later on
SWPadnos: did you end up kayaking(sp)
no, the tour we had signed up for was cancelled due to high winds
* skunkworks_ really needs to get a irc client installed that does spell checking
you had it right :)
wow - that is suprising
yeah. we asked them about it, since we have kayaks and are pretty good with them, but when they described the waves as "treacherous", we decided to do something else :)
good idea. :)
it was very windy the entire time we were there
it was pretty windy in vagas also.
of course this week is supposed to be nice and sunny :)
I was really happy with the weather
(beats ice and snow storms)
which we returned to
we had to circle a few times waiting for it to get clear enough to see the runway
when we got back to st paul - it was 0
oh, that's a little worse than here - it was 12 or so for us
we drove home from the airport - about 2+ hours and the roads where very good.
I guess I should go out and snow-blow the driveway and clear off the Jeep (so I don't have to do it at 6:00 tomorrow morning)
so I've been thinking about how to correlate a halscope trace with a real scope trace. I think it can be done if you have two spare digital output signals that you can update in the same HAL thread as you are doing your capture, and which you can capture on your scope. Signal 1 is a 2*PERIOD square wave, and signal 2 is a LFSR output with period long enough to correctly correlate with the right part of the halscope trace. A third signal would go
with the LFSR signal you can generally match up the two traces; with the square wave signal you can see the cycle-to-cycle jitter. leaving off the square wave, you could get the match and assume jitter is "low enough not to matter" when matching the traces.
you got cut off at "A third signal would go"
A third signal would go in one direction or the other to force halscope and the real scope to trigger at around the same time.
what's "waiting for s.axes" ?
axis prints it when task hasn't set up a plausible-looking status buffer in a timely fashion
* alex_joni wonders why
could be because of a real problem in task or it could be because task is just slow
I get it about 15 times
axis needs to know s.axes (essentially, which axis letters exist) before it can fully configure itself
on a certain config
ah, might be because of the VM
I get the same behaviour on stepper/stepper_mm.ini, so it's probably VMware
jepler: the 1:2.3.0~beta1 changelog mentions hm2 on 5i20 and 7i43, but not on the rest of the supported anyio boards
seb_kuzminsky: please feel free to retcon it
should i change the changelog for that version, or should i change it in 2.3.0-beta2?
I don't know, either way is a bit confusing
I think making a more accurate description of beta1 by changing its changelog after release is OK
ok will do
* seb_kuzminsky kicks CIA-2
EMC: 03seb 07TRUNK * 10emc2/debian/changelog: retroactively fix the hostmot2 blurb in the changelog
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/hostmot2.lyx: add info on hal pins and parameters
good night all
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/images/5i20-halpins.png: add image before buildbot catches me
good night alex_joni
that looks great BigJohnT
dang your quick seb_kuzminsky
hey seb_kuzminsky does mesa have a breakout board for the 50 pin cable?
i dont know...
i havent seen one
there are some 50-pin ribbon-to-screw-terminal boards listed on the wiki
ok, just wondering aloud how you might hook an encoder up....
on the emc wiki?
[23:26:26] <seb_kuzminsky> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?EMC2_Supported_Hardware
search for "Commercial breakout boards"
i'm using a 7i30 quad 100w servo driver
it's got a big fat 3x3 connector for each servo, with the encoder lines on it
dang terminal blocks must be costly...
they're quite expensive
for good ones, with detachable plugs, you've got to figure $1+ per contact
it's maybe $0.50 per contact for not very good, not-detachable ones
unless you're ordering 10k or more parts/year
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/gcode/main.lyx: add info to probe logging
oh crumb the image is too big
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/images/5i20-halpins.png: adjust image size
EMC: 03seb 07TRUNK * 10emc2/docs/man/man9/hostmot2.9: Added HAL object name examples to the manpage.
EMC: 03bigjohnt 07TRUNK * 10emc2/docs/src/drivers/hostmot2.lyx: that looks better