#emc | Logs for 2005-07-30

[00:04:48] <Jymmm> anyone know of a *live* PS viewer?
[00:06:56] <robin_sz> live?
[00:07:00] <robin_sz> in what way?
[00:07:07] <Jacky^> Jymmm: take a look, that nice light: http://shakelight.notanumberinc.com/flashlight/ :)
[00:07:12] <Jacky^> hi robin_sz
[00:07:16] <Jymmm> edit the PS in a text eidtor, and hit refresh in the viewer
[00:07:59] <Jymmm> Jacky^: I have one of those... you HAVE to keep ot at least 24" away from CRT's =)
[00:08:08] <robin_sz> writing the ps directly? ick
[00:08:20] <Jacky^> Jymmm: i bought it some day ago..
[00:08:29] <Jymmm> Writing the PS is easy, viewing it live is an issue
[00:08:36] <robin_sz> Jymmm: ghostscript?
[00:08:45] <Jymmm> robin_sz unfortanantly not live
[00:09:55] <robin_sz> well ...
[00:10:22] <robin_sz> dont you just type 'foo.ps run' on the command line
[00:11:00] <Jymmm> No clue in nix, on M$ I just convert to PDF live, but awefully slow
[00:12:51] <robin_sz> yeah, its (foo.ps) run on the GS commandline
[00:13:15] <robin_sz> I presume that will work in GS for 'doze too
[00:13:51] <robin_sz> converting to ps to pdf on doze? I didnt reallise you could do that on doze
[00:13:58] <Jymmm> let me try, GS wasn't processing the PS the last time I tried it.
[00:14:25] <Jymmm> oh yeah, actually I like to covert PS|PDF to SWF
[00:14:39] <robin_sz> swf?
[00:14:43] <Jymmm> swf == flash
[00:14:52] <robin_sz> ahh, dont think I have that
[00:15:09] <Jymmm> I LOVE the SWF format!
[00:15:25] <Jymmm> esp after they released the spec a couple years ago
[00:15:30] <Jacky^> I think it doesnt exist in linux
[00:15:38] <robin_sz> ive seen it used on a few websites
[00:15:46] <Jymmm> Jacky^ Flash player? YEah, but iffy
[00:16:02] <Jacky^> yeah, just the player
[00:16:03] <robin_sz> I try to avoid them really
[00:16:17] <Jymmm> robin_sz : Most see Flash the first time as fucking banner ads, but SWF format itself is awesome!
[00:16:42] <robin_sz> I most often see it as an 'intro' page on websites
[00:16:47] <robin_sz> which is VERY useful
[00:16:51] <Jymmm> ps, pdf, text, imges, video, audio, all build in
[00:17:24] <Jymmm> embedded fonts, etc
[00:17:40] <robin_sz> as soon as I see an SW Flash intro .. I just know the site will be a pile of wank, so I can hit the back button without actually having to go to the trouble of entering ..
[00:18:44] <robin_sz> quite a timesaver really
[00:19:47] <Jacky^> it does not respect gnu guidelinees :)
[00:19:58] <Jymmm> you cna even embedd 3D CAD files =)
[00:20:21] <Jymmm> quit being a cheap basterd Jacky^! =)
[00:20:29] <robin_sz> it can be good in some ways though ...
[00:20:46] <robin_sz> www.badgerbadgerbadger.com for example
[00:21:53] <Jymmm> robin_sz : This is much better --> http://www.freedomunderground.org/memoryhole/pentagon244.swf
[00:22:59] <Jacky^> et the end, however, flash it has not had the happened one that macromedium one expected, are many years that exist by now..
[00:23:48] <Jacky^> they said 5 years ago that it would have replaced html
[00:23:55] <Jacky^> bah..
[00:25:01] <Jacky^> I prefer java
[00:27:51] <Jymmm> too many java viruses
[00:29:22] <robin_sz> java is a fully-feaured programming langauge, swf is not. apples and oranges
[00:29:37] <Jacky^> can run everywhere..
[00:29:42] <robin_sz> ummm
[00:29:44] <robin_sz> no
[00:29:55] <robin_sz> write once,
[00:29:56] <Jacky^> palm pc
[00:29:59] <Jacky^> cell phone
[00:30:01] <robin_sz> debug everywhere
[00:30:02] <Jacky^> no ?
[00:30:27] <robin_sz> too many snicks and tricks for different platforms
[00:31:01] <robin_sz> its OK if you stick to simple core code
[00:32:05] <robin_sz> ive written a few server-side apps in it now .. fairly nice
[00:33:10] <Jacky^> I was looking for some java-code to control parport via web
[00:33:17] <Jacky^> too difficult for me
[00:33:43] <robin_sz> so ...
[00:33:56] <robin_sz> you just want to turn pins on/off from a web page?
[00:34:11] <Jacky^> nope..
[00:34:26] <Jacky^> Ive seen the small cgi in epanorama for this
[00:34:46] <Jacky^> working with a simple web page
[00:35:17] <Jacky^> i was looking for something to move a webcam from remote pc
[00:35:26] <Jacky^> then turn on off a light
[00:35:42] <Jacky^> moving the cam using a stepper of course..
[00:36:10] <robin_sz> isnt there some addons for 'motion' to do that?
[00:36:20] <Jacky^> i've no clue
[00:36:28] <Jacky^> some javascript maybe..
[00:36:39] <robin_sz> no
[00:36:43] <Jacky^> seem not simple to control parport from java
[00:36:52] <robin_sz> * robin_sz nods
[00:37:00] <Jacky^> :)
[00:37:01] <robin_sz> do you mean Java in a web browser?
[00:37:07] <Jacky^> yeah
[00:37:11] <robin_sz> then no
[00:37:17] <Jacky^> a mean using java language
[00:37:26] <Jacky^> I*
[00:37:28] <robin_sz> using Java its easy
[00:37:47] <robin_sz> but not as an applet, for obvious reasons
[00:38:30] <Jacky^> I've not found a clean example of this on the net
[00:38:39] <robin_sz> obviously
[00:38:53] <robin_sz> applets are not supposed to be able to touch hardware ...
[00:38:58] <robin_sz> security issues and all that
[00:39:03] <Jacky^> right
[00:39:14] <robin_sz> a signed applet might
[00:39:29] <Jacky^> give access to parport should be dangerous, I read ..
[00:39:36] <robin_sz> exactly
[00:39:38] <Jacky^> as non root
[00:39:54] <robin_sz> but its only applets that are restricted in that way
[00:40:04] <robin_sz> as a java program, its easy
[00:40:25] <robin_sz> java MyProgram etc
[00:40:31] <robin_sz> fronm the command line
[00:40:50] <Jacky^> :)
[00:41:07] <Jacky^> I've to do yet my first java cup :\
[00:41:31] <Jacky^> another solution maybe was here: http://www.drivemeinsane.com/
[00:41:53] <Jacky^> I tried to compile theyr software, but wont compile ..
[00:42:23] <Jacky^> that should be pretty sicure for remote access
[00:42:47] <Jymmm> just get a camera with built in web server
[00:43:10] <Jacky^> Jymmm: I know.. euro 250
[00:43:39] <Jacky^> for 250 I bought 3 webcam and 3 tv card
[00:44:10] <Jacky^> at the end, commercial webcam are using linux
[00:44:15] <Jacky^> with boa server
[00:44:17] <Jacky^> :)
[00:45:01] <Jymmm> http://shop2.outpost.com/%7BQurCc0oce2dd31z5LNj2ssoHY2qSeiI1f52QjJU42xelpU1V9PXv|2180629853395411844/184095267/6/7001/7001/7002/7002/7001/-1|4396956982378696067/184095266/6/7001/7001/7002/7002/7001/-1|1122681459318%7D/product/3950308?site=sr:SEARCH:MAIN_RSLT_PG&WebLogicSession=QurCc0oce2dd31z5LNj2ssoHY2qSeiI1f52QjJU42xelpU1V9PXv|2180629853395411844/184095267/6/7001/7001/7002/7002/7001/-1|4396956982378696067/184095266/6/7001/7001/7002/7002/7001/-1
[00:45:53] <Jacky^> Jymmm: nice
[00:46:19] <Jacky^> I friend of mine has bought one for 250
[00:46:48] <Jacky^> ah !
[00:46:59] <Jacky^> without motion control ..
[00:47:36] <Jacky^> it can send on lan the streaming video to record from a remote pc
[00:50:33] <Jymmm> http://www.x10.com/cameras/index.html
[00:51:46] <Jacky^> nice prices, there
[00:54:32] <Jacky^> take a look here: http://www.intellisystem.it/LISTINO/camereaxis.htm
[00:55:09] <Jacky^> :\
[01:01:50] <anonimasu> thoose are good prices for thoose cams..
[01:02:24] <Jacky^> anonimasu: :)
[01:02:38] <Jymmm> too bad the quality of x10 cameras is shit
[01:02:49] <anonimasu> axis should be good stuff..
[01:02:54] <anonimasu> for surveilance and so..
[01:03:08] <anonimasu> but the surveilance software is 1000eur
[01:03:10] <anonimasu> ;)
[01:03:28] <Jacky^> I no need long distance..
[01:03:49] <Jacky^> so I can run it near the PC
[01:04:07] <Jacky^> just connecting to usb ports
[01:04:33] <Jacky^> with normal webcams
[01:05:36] <Jacky^> and one TV card for any webcam..
[01:16:13] <Jacky^> GoodNight
[01:16:21] <Jacky^> *_*
[01:16:28] <Jacky^> see tomorrow
[02:20:10] <SWP_Away> SWP_Away is now known as SWPadnos
[02:27:06] <SWPadnos> SWPadnos is now known as SWP_Away
[16:37:30] <SWPadnos> gah - stupid Chatzilla ate my line
[16:37:52] <alex_joni> started with / ?
[16:38:15] <SWPadnos> nope - there's a bug (or a feature I don't know about)
[16:38:36] <alex_joni> documented bugs are features
[16:38:37] <alex_joni> :D
[16:38:38] <SWPadnos> if you hit <ctrl>-<down arrow>, the line you're editing disappears
[16:39:02] <alex_joni> undocumented features are bugs :)
[16:39:12] <SWPadnos> it's easy to do accidentally when you're trying to mark words with <strl>-<shift><arrow>
[16:39:18] <SWPadnos> <ctrl>, that is
[16:57:49] <alex_joni> well.. going away for a while
[16:57:53] <alex_joni> have some stuff to do
[16:57:54] <alex_joni> later
[16:57:59] <SWPadnos> see ya
[17:14:49] <SWPadnos> SWPadnos is now known as SWP_Away
[17:30:59] <les> ok the tp site is up. prob has mistakes. anyone care to proofread?
[17:31:04] <les> http://www.lmwatts.com/traj/traptp.html
[17:37:01] <les> site is up.
[17:37:07] <SWP_Away> looking
[17:37:14] <les> thanks
[17:37:39] <les> the pseudocode is sloppy...I really need to redo it
[17:37:53] <SWP_Away> heh - either that or I haven't had enough coffee :)
[17:39:27] <anonimasu> looks good
[17:40:39] <SWP_Away> so - this is what's supposed to happen, not necessarily what tp.c actually does, right?
[17:40:53] <SWP_Away> (ie, compare this to tp.c to find the error)
[17:40:59] <les> yeah
[17:41:01] <anonimasu> yeah..
[17:41:11] <SWP_Away> ok.
[17:41:14] <anonimasu> I wonder if we could re-write the tc with the help of that..
[17:41:25] <les> this is just a simple intro...I did not get into clamping etc
[17:41:26] <anonimasu> les: the tp does the velocity vector calcs right?
[17:41:39] <les> but clamping is not complicated either
[17:41:54] <les> You see how remarkably simple it is?
[17:42:01] <SWP_Away> It looks like there's a typo in the last part: acceleration = D^2x... - that's supposed to be a derivative 'd', right?
[17:42:30] <anonimasu> and the tc just calcs the velocity.. from-to/cycletime
[17:42:34] <les> yeah second derivative...should be lower case...thanks
[17:43:36] <les> I am looking at mathtype...seems like a simple math editor with no learning curve
[17:43:44] <les> about a hundred bucks
[17:43:47] <les> may get it
[17:43:56] <anonimasu> les: the logic there is to calc with negative and positive velocoties..
[17:43:58] <anonimasu> as far as I can tell..
[17:44:00] <anonimasu> most of it anyway
[17:44:03] <SWP_Away> OOO works - try it
[17:44:38] <les> OOO looks good but I need to do ,gif s etc
[17:44:49] <Jymmm> les you mean for writing formulas?
[17:44:54] <SWP_Away> just remember - use curly braces to make groups (like sqrt(2x) should be sqrt{2x}), and the pound sign separates elements of a matrix or vector
[17:44:55] <les> yeah.
[17:45:20] <SWP_Away> just do a screen capture
[17:45:30] <les> I am missing some parens in the pseudocode
[17:45:34] <anonimasu> les: to what did you just way "yeah"
[17:45:34] <anonimasu> ;)
[17:45:43] <anonimasu> les: did you read my rant?
[17:45:44] <les> yeah could do screen capture
[17:45:45] <SWP_Away> to Jymmm , I'd imagine
[17:46:16] <les> which rant heh
[17:46:59] <anonimasu> after "how remarkably simple it is"
[17:47:25] <SWP_Away> SWP_Away is now known as SWPadnos
[17:47:36] <les> You know, I am getting paid a lot of money to write tech articles these days...I can afford even a few hundred for a good quick math editor
[17:47:49] <les> oh let me check...
[17:48:52] <les> I see no rant....heh
[17:48:58] <anonimasu> hm, what do you see?
[17:48:58] <anonimasu> crap?
[17:49:02] <anonimasu> :D
[17:49:04] <les> haha
[17:49:17] <SWPadnos> (it looked like a couple of comments to me :) )
[17:49:43] <les> I will rant.....How the hell cap tp/tc end up so complicated...even if it is vector?
[17:49:45] <les> heh
[17:49:59] <anonimasu> heh
[17:50:08] <SWPadnos> theory != practice
[17:50:21] <les> i guess.
[17:50:33] <les> theory to malpractice.
[17:50:40] <SWPadnos> heh
[17:50:41] <anonimasu> yeah
[17:51:03] <anonimasu> les: but as I said, I wonder if the stuff you wrote could be plugged right in..
[17:51:12] <anonimasu> I think all variables are there..
[17:51:15] <SWPadnos> les: you can save OOO Math files as MathML, which will render correctly in most browsers (possibly with a plug-in)
[17:51:23] <anonimasu> except for clamp & scaling
[17:51:38] <les> I couldn't...since I can't figure out what's going on in there
[17:51:52] <les> clamping is just time adapting
[17:52:56] <les> Oh the begginning and end segments have some determinant terms that I didn't put in there...looked too nasty in pseudocode
[17:53:09] <les> but not complicated math
[17:53:18] <les> just quadratic formula
[17:53:35] <anonimasu> as long as it's formulas with clearly defined variables I have no problem following them..
[17:53:52] <anonimasu> :)
[17:54:01] <SWPadnos> the variable tinb - is that "time in blend"?
[17:54:48] <les> yeah...right out of the book. Just consider it an intermediate calc
[17:55:11] <anonimasu> maybe we should call prolog from the tp..
[17:55:12] <SWPadnos> can you accept DCC?
[17:55:17] <les> should have changed the name to protect the innocent
[17:55:25] <les> yes I can DCC
[17:55:40] <anonimasu> and just use the real formula ;)
[17:55:43] <anonimasu> *grins*
[17:57:08] <Jymmm> les : I have a utility for writing out formulas if you're interested?
[17:57:20] <SWPadnos> dcc les
[17:57:23] <SWPadnos> oops
[17:57:28] <les> that makes graphics?
[17:57:36] <les> waiting...
[17:57:39] <Jymmm> les : sure
[17:58:00] <Jymmm> les it makes those funky text book ones
[17:58:22] <SWPadnos> les, tell me if that looks good to you (took about 6 seconds to write)
[17:59:00] <les> yeah looks nice
[17:59:02] <SWPadnos> and here's the source code: t sub b = t-(t sub j over 2)
[17:59:15] <les> heh
[17:59:30] <SWPadnos> kind of like reading an equation out loud, huh? :)
[17:59:46] <les> yeah
[18:00:21] <les> so that must be OOO...what do you have jymmm
[18:00:30] <SWPadnos> yep - that was OOO
[18:01:10] <Jymmm> les : OOO ?
[18:01:21] <SWPadnos> OpenOffice.Org
[18:01:29] <les> got it
[18:01:32] <mshaver_away> mshaver_away is now known as mshaver
[18:01:34] <SWPadnos> http://www.openoffice.org/
[18:01:35] <mshaver> les?
[18:01:40] <les> I will check this out.
[18:01:43] <les> hi matt
[18:01:45] <Jymmm> nah, this isn't OO, it's a website
[18:01:50] <mshaver> Hi
[18:01:55] <Jymmm> all live online
[18:02:02] <les> oh...cool
[18:02:04] <mshaver> I'm trying to learn how logging works
[18:02:19] <SWPadnos> well - programs write out data to a file, so you can look at it later :)
[18:02:44] <les> I have done it a little but not much...the old version I had was screwed up...now it is prob ok
[18:02:45] <Jymmm> les gimme a few to find it.
[18:02:52] <les> k
[18:03:06] <mshaver> I can get values of pos or vel to plot on the Y axis, but everything sits on Y zero
[18:03:22] <mshaver> which gives me a vertical line...
[18:03:56] <anonimasu> just press plot on the logging
[18:03:59] <les> so you have the file ok...just a gnuplot thing?
[18:04:00] <anonimasu> save then plot
[18:04:02] <anonimasu> that's the easiest way
[18:04:20] <les> I will have to defer on this one
[18:04:37] <mshaver> that's what i'm doing... what sets the X axis?
[18:04:57] <mshaver> i mean the graphing x axis
[18:08:04] <mshaver> I get this: http://users.erols.com/mshaver/snap1.png
[18:09:53] <anonimasu> that's strange
[18:14:40] <mshaver> ok, I'll dig into the tcl code later...
[18:14:49] <mshaver> mshaver is now known as mshaver_away
[18:34:02] <les> wow Matt that green on gray coordinates is great! It's so easy on my eyes! ;)
[18:34:19] <les> I'll have to change my display colors right away!
[18:37:38] <les> off to lunch...bbiaw
[18:41:03] <robin_sz> meep?
[19:09:26] <les> hi robin
[19:09:36] <les> was out cutting down another tree
[19:09:54] <robin_sz> heh
[19:10:21] <SWPadnos> hey Les - what's SGN?
[19:11:04] <les> SGN( arg) is just 1 times the sign of aeg
[19:11:12] <les> arg
[19:11:19] <SWPadnos> OK - I was wondering if it was sign
[19:11:42] <les> yeah...don't see that so much these days
[19:12:09] <SWPadnos> wouldn't that be -1, 1, or 0 though - ie, (arg / abs(arg))
[19:12:37] <SWPadnos> ah - times |desired acceleration| - I get it now
[19:12:43] <les> uh...yeah the zero thing I'm not sure.
[19:12:54] <les> heh +0 and -0.
[19:13:04] <SWPadnos> right - what's the sign of 0? (probably still in discussion in higher math circles :) )
[19:13:30] <les> limit calculus....
[19:13:41] <SWPadnos> My mother was in a course where they took half a semester to prove that there are no integers betweem 0 and 1
[19:13:45] <SWPadnos> between
[19:13:46] <les> l'hopital's rule...
[19:13:57] <les> hahaha
[19:14:03] <SWPadnos> right - plus if approached from the positive side...
[19:14:15] <anonimasu> that's a pretty stupid course
[19:14:16] <anonimasu> :
[19:14:18] <anonimasu> :)
[19:14:30] <Jymmm> so, what... there ARE intergers between 1 and 2 ?
[19:14:43] <SWPadnos> actually, it was "no integers between the additive identity and the multiplicative identity" - they just happen to be 0 and 1
[19:14:48] <les> breen
[19:14:57] <les> it was recently discovered.
[19:15:15] <les> old george carlin joke.
[19:15:24] <les> one, breen, two, ...
[19:15:28] <SWPadnos> the seven words you can never say on television?
[19:15:34] <les> hahaha
[19:16:04] <les> forcast for tonight...dark...with scattered light in the morning
[19:16:09] <les> heh
[19:16:16] <SWPadnos> heh
[19:16:28] <SWPadnos> on the moon, it would be non-scattered light
[19:17:07] <les> heh
[19:17:09] <Jymmm> SWPadnos Didn't you hear? There are clouds on the moon now!
[19:17:17] <SWPadnos> and ice lakes on Mars
[19:17:24] <Jymmm> ... ands the moon is made from green cheese too!
[19:17:41] <SWPadnos> have you looked at Google maps of the moon?
[19:17:47] <les> I was watching nasa tv...astronouts are space walking with putty knives and caulk guns. looks funny.
[19:17:54] <SWPadnos> zoom in too far and you get swiss cheese
[19:17:59] <les> yeah looked at googlemoon
[19:18:09] <les> yeah I saw that. funny
[19:18:25] <Jymmm> google moon?
[19:18:36] <Jymmm> where's the 'You are here' sign?
[19:18:36] <les> I think
[19:18:47] <SWPadnos> well - I'm not there, so ...
[19:18:56] <Jymmm> (questionable)
[19:19:02] <SWPadnos> watch it ...
[19:19:23] <Jymmm> SWPadnos : Hey, at least I didn't say you were on Pluto fucking goofy!
[19:19:26] <les> Hey Robin..aren't you gonna bitch about the code in my tp page? I did it with sitespinner again
[19:19:34] <les> horrible html I'm sure
[19:19:38] <SWPadnos> I wish they had a scale on their maps
[19:19:56] <robin_sz> les: only if you specifically ask me to comment
[19:19:57] <Jymmm> SWPadnos they do, it's right next to the 'you are here' sign =)
[19:20:01] <les> hahaha
[19:20:04] <SWPadnos> no wonder
[19:20:13] <SWPadnos> I could see the scale if I were in the picture
[19:20:25] <Jymmm> fuck this is taking forever!
[19:20:28] <robin_sz> les: but the content is more important than the form ... URI?
[19:20:38] <les> um...
[19:20:43] <SWPadnos> http://www.lmwatts.com/traj/traptp.html
[19:20:53] <les> ty
[19:20:58] <les> beat me to it
[19:21:03] <SWPadnos> sure - it's still open in my browser
[19:21:11] <les> haha
[19:21:20] <SWPadnos> I'm formula-izing the equations
[19:21:29] <les> oh neat
[19:22:27] <Jymmm> I swear it's gonna take two more days to finich copying these files!
[19:23:30] <robin_sz> a bit light on the maths so far huh?
[19:23:42] <les> I kept is simple
[19:23:46] <robin_sz> yeah
[19:24:00] <les> no first and last point...they are quadratics
[19:24:06] <les> no clamping either
[19:24:16] <SWPadnos> it seems that the "real world" limitations aren't taken into account - like figuring out whether a segment is too short to allow adequate acceleration
[19:24:24] <les> just wanted to explain the principle
[19:24:25] <robin_sz> after all, you are writing it to be read by anonimasu, needs to be simple ... ;)
[19:24:34] <robin_sz> * robin_sz ducks
[19:24:39] <les> hahaha
[19:25:16] <les> I was going to do the velocity adaptation part...but the pseudocode looked boilerplatey
[19:25:26] <robin_sz> hard to see how something so simple ends up as so much code
[19:25:55] <SWPadnos> corner cases (no pun intended) always lengthen code
[19:26:01] <SWPadnos> significantly
[19:26:24] <les> just take the upper velocity chart...pull on it and stretch it out....but velocity is t^-1 so it will get thinner in that dircetion
[19:26:40] <les> like poisson's ratio .5
[19:27:27] <les> just slow down till motions are realizable...by some heuristic
[19:27:47] <les> don't want to do it dynamically with each segment...too choppy
[19:28:19] <les> btw velocity adaptation in tp/tc does not work properly
[19:28:36] <les> may be related to our glitch!
[19:29:21] <SWPadnos> hmmm - I really need to get some other work done. Les - do you want the latest formulas? I've done most of the second section of the page - it's a good template for you
[19:29:56] <les> to be fair, tp/tc is prob operating on vectors in task space
[19:30:02] <les> yes ty
[19:30:52] <SWPadnos> ok - one sec
[19:30:57] <les> k
[19:31:01] <SWPadnos> did you download OpenOIffice?
[19:31:08] <SWPadnos> -I
[19:31:15] <les> not yet..I was cutting down trees
[19:31:18] <les> but I will
[19:31:26] <SWPadnos> heh
[19:32:25] <SWPadnos> that's the OOO format and a PDF
[19:33:07] <les> funny...default IE ignored it. Gee I wonder why...
[19:33:13] <SWPadnos> heh
[19:33:28] <SWPadnos> shall I DCC again?
[19:33:50] <les> DCCis active, but no transfer
[19:34:04] <SWPadnos> hmmm - did you accept the invitation?
[19:34:11] <les> yeah
[19:34:17] <les> try again please
[19:34:19] <SWPadnos> odd - I'll cancel and do it again
[19:35:15] <les> what is the extension?
[19:35:22] <SWPadnos> odf and pdf
[19:35:47] <les> ok the odf is a problem. once more and I will fix that
[19:35:53] <SWPadnos> ok
[19:36:51] <SWPadnos> I can just email them if you like
[19:37:18] <les> i put .odf in the accept list but then the dcc dialog box vanished.
[19:37:25] <les> one more then give up
[19:37:27] <SWPadnos> odd
[19:38:03] <SWPadnos> you may need to restart (the program) to get the changes to work
[19:38:33] <les> it is accepting but no transfer
[19:38:41] <les> leswatts@lmwatts.com
[19:38:50] <SWPadnos> hmmm - "it worked before" :)
[19:38:55] <les> heh
[19:39:31] <SWPadnos> on the way
[19:39:39] <Jymmm> blah.odf.jpg
[19:40:12] <les> ty
[19:40:12] <SWPadnos> mailto: ...
[19:40:33] <SWPadnos> the description language is very straightforward - it should be easy to do the rest
[19:40:57] <les> in the paper?
[19:41:06] <SWPadnos> now I've got to get to work on the stuff I couldn't do Thursday (because my computer died) or Friday (because I was sick) :)
[19:41:15] <les> ok
[19:41:17] <les> thanks
[19:41:36] <les> I guess I will buck up trees
[19:41:39] <Jymmm> SWPadnos your computer got a virus then gave it to you!
[19:41:52] <les> new husky chainsaw has a little low idle
[19:41:54] <SWPadnos> you can save as MathML, which works with browsers - it should be easy to import that way, or with screenshots (<ALT>-PrintScreen captures the current application to the clipboard)
[19:42:00] <SWPadnos> I guess so :)
[19:42:09] <SWPadnos> see ya
[19:42:11] <SWPadnos> SWPadnos is now known as SWP_Away
[19:42:17] <Jymmm> SHIFT+PrntScrn too
[19:42:24] <les> I guess it's not unusual to have to tweak idle on a new saw after breakin
[19:42:59] <Jymmm> les Just remember... where is says 'add oil here' to put in water
[19:43:10] <RevNealUmbly> bee odd if you didn't have to tweek a wearing tool
[19:43:10] <les> ha
[19:43:37] <Jymmm> les and where is says 'fuel', it wants diesel
[19:43:51] <les> the tractortoy does...
[19:44:42] <les> aw, it's starting to rain.
[19:45:49] <les> I wonder how long browsers have had mathml support
[19:46:09] <Jymmm> which, firfox?
[19:46:23] <Jymmm> firefox
[19:46:43] <les> IE...I have to write pages for that, because most people use it
[19:46:59] <Jymmm> it's probably a plugin
[19:47:10] <les> oh...not good
[19:47:25] <Jymmm> give me a url and I'll test it out
[19:47:44] <les> I don't have one
[19:47:57] <les> googleing mathml
[19:48:05] <Jymmm> going that now =)
[19:48:56] <Jymmm> les : http://www.mathmlcentral.com/Tools/FromMathML.jsp
[19:49:32] <les> looking
[19:49:40] <Jymmm> lets you generate a graphic
[19:50:31] <Jymmm> les be sure to look at the options on the left side of the screen too.
[19:50:40] <Jacky^> hi
[19:50:46] <Jacky^> time for dinner :P
[19:50:57] <Jacky^> Jacky^ is now known as Jacky^afk
[19:51:19] <les> neat
[19:51:39] <les> bookmarked that.
[19:52:02] <les> ok..lunch...lumberjacks get hungry
[19:52:12] <Jymmm> les you might have to enter as expression, then copy the results in MathMl generated to the text input
[19:52:19] <Jymmm> to generate a graphic
[19:52:39] <les> I like the gif..that's fine
[19:52:57] <cradek> wheee
[19:53:01] <cradek> I bought a lathe
[19:53:25] <les> got it?
[19:53:29] <cradek> yep
[19:53:35] <les> pictures pictures pictures
[19:53:53] <cradek> I have a picture of it here named P0007868.jpg
[19:54:08] <les> so tell me about the shape and what came with it
[19:54:16] <les> haha
[19:54:52] <cradek> it looks like it's in good shape. I had trouble measuring runout because my indicator wouldn't mount to the toolholder, and the inside of the spindle isn't entirely smooth
[19:55:32] <cradek> it came with two 3-jaw and one 4-jaw chuck
[19:55:36] <les> any burrs can be scraped out with a triangle scraper
[19:55:51] <cradek> a couple dead centers for the tailstock
[19:55:51] <les> that 3 jaw cost over $200 each!
[19:56:00] <cradek> no faceplate, no center for the headstock
[19:56:13] <cradek> two crappy jacobs chucks for the tailstock
[19:56:15] <les> easy to buy or make
[19:56:22] <cradek> but I have a new one
[19:56:32] <cradek> regular and cutoff tool holders
[19:56:52] <cradek> the motor on it is *huge*, I don't know if it's original
[19:57:10] <cradek> it's all bolted to a mish-mash metal table he welded up
[19:57:10] <Jymmm> if it says 'dayont' I doubt it =)
[19:57:13] <Jymmm> dayton
[19:57:23] <les> some times the handwheel bearing holes get wallowed out...np...just put in bronze inserts
[19:57:43] <les> I did that even on fairly new ones...just felt better
[19:58:12] <cradek> yeah the cross feed doesn't feel quite right
[19:58:21] <cradek> but the gibs are all misadjusted
[19:58:43] <les> even brand new they had a lot of clearance
[19:58:50] <les> yeah typical
[19:59:07] <cradek> all in all, I think it's pretty decent
[19:59:12] <les> easy to do though
[19:59:15] <cradek> it will be fine for what I want to do
[19:59:42] <les> oh I made a LOT of stuff with the two I had
[20:00:13] <cradek> I see how the teeth get knocked off the back gear!
[20:00:25] <les> but yours is ok?
[20:00:28] <cradek> yes, fine
[20:00:49] <les> you can do .0002" work on that lathe if it is set up tight
[20:00:51] <cradek> that's probably just for threading large diameters, right?
[20:00:52] <les> easily
[20:01:11] <les> yeah treading and drilling
[20:01:20] <cradek> oh yeah, big drills
[20:01:33] <cradek> can you tell I've always used miniature tools
[20:01:46] <cradek> I guess some would call this 6" miniature
[20:02:16] <les> Do the old taper test bar and shim the feet
[20:02:48] <les> then set up the tailstock offset
[20:02:56] <cradek> feet?
[20:03:08] <cradek> I understand how to do the tailstock
[20:03:10] <les> yeah touching the table
[20:03:33] <les> so the bed has no twist
[20:03:40] <cradek> ah
[20:03:51] <cradek> I wonder if this table is stout enough
[20:03:54] <les> normal setup...especially after a move
[20:04:48] <les> I use aboxford VSL in the shop now...bought 5 with someonne kept two, sold 3
[20:04:58] <les> oops
[20:05:03] <les> Boxford
[20:06:12] <les> well go forth and make round things
[20:06:33] <les> bigger round things!
[20:07:16] <cradek> I'll finally make that damn pulley for my pcb mill
[20:07:30] <cradek> remember you offered to make it for me months ago?
[20:07:32] <les> and of course take it all to pieces and clean it with toothbrushes ;)
[20:07:37] <les> yeah i remember
[20:07:57] <cradek> nah, I'll just use it. I don't wash my motorcycle either, I just ride it
[20:08:17] <les> ???? HUH? haha
[20:08:30] <RonB> * RonB likes cradek's attitude
[20:08:32] <les> I detail the tractor with a toothbrush...
[20:08:34] <les> heh
[20:08:38] <cradek> ha
[20:09:00] <cradek> a guy I work with (who rides a sport bike) totally freaks if it gets rained on
[20:09:03] <cradek> it's so funny
[20:09:04] <Jymmm> les So, how much chrome have you on the tractor now?
[20:09:08] <les> I know RonB I have a bit of a problem with that
[20:09:17] <les> heh
[20:09:37] <les> I got dirt on the front end loader bucket. Freaked me out.
[20:09:46] <cradek> haha you're a nut
[20:09:58] <RonB> I had about 1/2 million miles on my last truck ... got washed less than 10 times - that's waht rain is for
[20:10:21] <les> My tools are hHORRIBLY clean. antiseptic
[20:10:23] <les> heh
[20:10:29] <les> oops
[20:10:32] <cradek> RonB: with that many miles, eventually the rain will was the inside too
[20:10:35] <cradek> wash
[20:10:48] <RonB> I've seen picture of your shop... make that operating room
[20:10:57] <les> yeah heh
[20:11:13] <RonB> nah - cradek - live in a dry climate
[20:11:28] <cradek> RonB: oh, not me, we get serious rust in 10 years
[20:11:30] <RonB> BTW - I've always admired you work
[20:11:52] <RonB> we get serious DUST here - central TX
[20:12:23] <cradek> I think the killer is the salt/chemicals they use to melt the snow here
[20:12:39] <les> me RonB? tyvm
[20:12:58] <RonB> what is snow
[20:13:11] <Jymmm> white cold water
[20:13:13] <RonB> you are welcome Les -
[20:13:48] <RonB> I ran across your site when i started playing with CNC - about 97-98 I would think
[20:13:51] <les> of for some food...cradek go play with your lathe!
[20:13:57] <les> bye
[20:14:28] <les> oh...yeah. I have designed many tables over the years
[20:14:47] <les> later!
[20:15:11] <RonB> far more accurate than is needed for "never show on camera" woodwork
[20:47:59] <robin_sz> life isn't all beer and skittles
[20:49:04] <ValarQ> what?!?
[20:50:23] <robin_sz> "Life isn't all beer and skittles; but beer and skittles, or something better of the same sort, must form a good part of every man's education."
[20:50:35] <robin_sz> famous quote
[20:50:45] <ValarQ> oh, ok
[20:51:09] <robin_sz> it will confuse the americans though, who probably dont have skittles ;)
[20:51:53] <ValarQ> we don't have skittles either
[20:52:16] <robin_sz> its a common game in country pubs around here
[20:53:55] <robin_sz> http://www.radstockhotel.co.uk/images/radstockPicture7.jpg
[20:53:59] <robin_sz> some skittles ...
[20:54:19] <robin_sz> and ...
[20:54:33] <robin_sz> http://www.thefourhorseshoes.co.uk/images/skalleysm.jpg
[20:54:38] <robin_sz> a skittle alley
[20:55:08] <robin_sz> think of it as english ten-pin bowling
[20:55:23] <Jymmm> We have skittles, but it's a candy =)
[20:55:39] <robin_sz> ah yes, we have that too
[20:56:47] <Jymmm> only 11gb left, damn! Hope it's enough room
[20:56:56] <robin_sz> the key to skittles is that 1) its manual (taking turns to put up the pins is part of the evening)
[20:57:13] <robin_sz> 2) its played in teams
[20:57:40] <robin_sz> 3) the ball is much smaller than a bowling ball and can be thrown very fast ...
[20:57:49] <anonimasu> hi
[20:58:15] <Jymmm> hey anonimasu! robin_sz was just talking about you. Were your ears ringing?
[20:58:26] <robin_sz> nah htat was ages ago
[20:58:28] <anonimasu> Jymmm: no I bought a coke at the kiosk..
[20:58:28] <anonimasu> ;)
[20:58:38] <Jymmm> ah, o
[20:58:39] <Jymmm> k
[20:58:43] <anonimasu> and I thought I was going to pop in and see what improvements to the tp you've made :D
[20:59:02] <Jymmm> anonimasu it's gonna be scrapped
[20:59:07] <anonimasu> *grins*
[20:59:20] <robin_sz> taken out the back and shot ... its the only humane thing to do
[20:59:22] <Jymmm> dont believe me? ask robin_sz
[20:59:41] <anonimasu> no seriously, got any conclusions at all?
[20:59:51] <robin_sz> no, been asleep :)
[21:00:01] <anonimasu> ok, I am going to finish my logging stuff for tc.c
[21:00:01] <anonimasu> ;)
[21:00:12] <Jymmm> anonimasu les worked up some formulas and SWP_Away was/is playing with them
[21:00:20] <anonimasu> Jymmm: the ones in the paper?
[21:01:00] <Jymmm> anonimasu couldn't tell you, I didn't keep much track during the conversation.
[21:01:13] <Jymmm> anonimasu it's like a boing 737
[21:01:33] <Jymmm> boeing
[21:02:00] <anonimasu> Jymmm: are you sure you are not talking about the paper
[21:02:14] <Jymmm> anonimasu 30k feet and climbing
[21:02:20] <anonimasu> gah.
[21:02:41] <SWP_Away> (waaaaaay over his head :) )
[21:02:42] <anonimasu> I wonder where I get a book about trajectory planning
[21:02:49] <anonimasu> SWP_Away: wb
[21:02:57] <Jymmm> SWP_Away bingo =)
[21:03:03] <SWP_Away> I'm not here - this isn't the SWP you're looking for
[21:03:07] <SWP_Away> move along
[21:03:19] <anonimasu> heh
[21:03:30] <SWP_Away> I guess my jedi IRC tricks aren't working
[21:03:48] <Jymmm> Star Wars is so, so, 80's
[21:03:51] <anonimasu> SWP_Away: they are but my force is stronger
[21:03:57] <SWP_Away> '70s, actually
[21:04:05] <Jymmm> nah, really?
[21:04:40] <Jymmm> I could swore 80's
[21:04:46] <SWP_Away> yep - around the same time as "Logan's Run"
[21:04:59] <Jymmm> * Jymmm shutters *
[21:05:00] <SWP_Away> 1977
[21:05:18] <SWP_Away> Logan's Run was 1976
[21:05:38] <SWP_Away> "Close Encounters of the Third Kind" was also 1977
[21:05:38] <anonimasu> *goes away*
[21:05:48] <Jymmm> CE was good
[21:05:54] <SWP_Away> yeah
[21:06:15] <SWP_Away> we got the enhanced DVD - brought back memories
[21:06:23] <Jymmm> heh
[21:06:34] <SWP_Away> they have the Budweiser clydesdales on TV in the background
[21:06:49] <Jymmm> Well, I like fucking around in areas I'm not really suppose to be in, so....
[21:07:01] <SWP_Away> gonna try with the TP?
[21:07:14] <robin_sz> if only there was more info on tp in general
[21:07:25] <SWP_Away> we just bought 36 rolls
[21:07:29] <Jymmm> what about the ORIGINAL code?
[21:07:40] <robin_sz> nah
[21:08:13] <SWP_Away> it's a really trivial problem in theory - like Y=mX + b
[21:08:14] <Jymmm> SWP_Away : I've only been escorted out of an area by M16 once.
[21:08:38] <SWP_Away> heh - I've been excorted with an M16 many times (we all had them :) )
[21:08:42] <SWP_Away> escorted
[21:09:08] <Jymmm> SWP_Away : Well, this wasn't a military base (per se)
[21:09:18] <SWP_Away> ah - that is different
[21:09:55] <Jymmm> SWP_Away : This is a 'in the middle of nowhere on a single way in JEEP ONLY trail' sorta area
[21:09:57] <robin_sz> SWP_Away: theres OODLES of robaotics papers on the web that cover tp in detail
[21:10:01] <anonimasu> SWP_Away: it's true
[21:10:12] <SWP_Away> SWP_Away, is getting a lot of messages :)
[21:10:34] <anonimasu> yeah, feel like moving the tp discussion somwhere else to keep it easier to track?
[21:10:42] <SWP_Away> smoothing should be easy - you have a starting velocity, and an ending velocity
[21:10:47] <SWP_Away> SWP_Away is now known as SWPadnos
[21:10:51] <SWPadnos> damn - I'm back
[21:11:02] <SWPadnos> the difference is delta V
[21:11:14] <anonimasu> yeah cpos-newpos/time
[21:11:29] <SWPadnos> you want to get through the delta V as quickly as possible - always use max accel to do it
[21:11:41] <SWPadnos> (that keeps corners as sharp as possible)
[21:12:11] <SWPadnos> it takes N periods to get from V1 to V2, where N is deltaV / maxAccel
[21:12:11] <anonimasu> yeah
[21:12:19] <robin_sz> here ...
[21:12:22] <robin_sz> read this ;)
[21:12:24] <robin_sz> batman.mech.ubc.ca/~ial/publication/theses/sonja.pdf
[21:12:29] <SWPadnos> been there
[21:12:35] <SWPadnos> done some of that :)
[21:13:04] <robin_sz> chapter 4 seems fun
[21:13:17] <anonimasu> *sigh*
[21:13:25] <SWPadnos> I think I stopped reading somewhere in chapter 1 ;)
[21:13:26] <anonimasu> why didnt I have that paper before.
[21:14:03] <robin_sz> its the best one I found on the subject during this afternoons brief search of the web
[21:14:07] <robin_sz> then I fell asleep
[21:14:09] <anonimasu> * anonimasu likes it
[21:14:52] <anonimasu> this is nice stuff
[21:16:26] <anonimasu> wb
[21:19:11] <SWPadnos> OK - so basically I just mentioned equation 4.1
[21:20:47] <anonimasu> :)
[21:23:27] <robin_sz> the bibliography is awesome
[21:23:40] <robin_sz> that sort of research alone would take you months
[21:24:24] <SWPadnos> yeah - 3 1/2 pages of that stuff - damn
[21:27:05] <anonimasu> I am going to print that out..
[21:27:20] <anonimasu> I'd love to have a copy of it as a book..
[21:27:42] <robin_sz> any more tasks for me today?
[21:28:00] <SWPadnos> yeah - fis the TP ;)
[21:28:02] <SWPadnos> fix
[21:28:11] <anonimasu> * anonimasu is having trouble starting to do anything today
[21:28:15] <anonimasu> bloody saturday.
[21:28:16] <robin_sz> no chance ...
[21:28:27] <robin_sz> first I would have to put the code into order
[21:28:30] <SWPadnos> heh - well then, you're fired
[21:28:37] <anonimasu> laters robin_sz
[21:28:38] <robin_sz> excellent
[21:28:45] <SWPadnos> more free time now
[21:28:50] <anonimasu> if you cant do that simple stuff, just go and laser something ;)
[21:28:55] <robin_sz> can I have my back pay please?
[21:29:02] <anonimasu> robin_sz: no way!
[21:29:12] <SWPadnos> what he said
[21:29:18] <robin_sz> dang
[21:29:21] <anonimasu> robin_sz: what are we other going to buy drinks from ;)
[21:29:25] <anonimasu> s/drinks/booze
[21:30:07] <anonimasu> SWPadnos: however looking at 4.1
[21:30:21] <anonimasu> err at the paper..
[21:30:36] <SWPadnos> yeah
[21:30:38] <anonimasu> the tp isnt the broken thing I think, it's the velocity calculation..
[21:30:44] <anonimasu> the planning works..
[21:31:04] <anonimasu> just not the blending
[21:31:11] <SWPadnos> I'm sure there's a bug in the implementation, not necessarily in the idea
[21:31:34] <anonimasu> yeah but if you look at the plots the commanded position is where it should be if I remember it right...
[21:31:39] <anonimasu> but the velocity is way off..
[21:32:06] <robin_sz> I think we need to find lloyd and haywards paper ..
[21:32:23] <anonimasu> yeah
[21:32:36] <anonimasu> I wonder if I can get this stuff as a book..
[21:33:03] <robin_sz> Lloyd, J., Hayward, V., "Trajectory Generation for Sensor-Driven and Time-Varying Tasks",
[21:33:03] <robin_sz> International Journal of Robotics Research, Vol. 12, No. 4, (1993): 380-393.
[21:33:28] <anonimasu> I wonder if I could use a copy of it at work ;)
[21:33:58] <anonimasu> robin_sz: I meant the sonja paper ;)
[21:34:01] <robin_sz> http://www.cs.ubc.ca/spider/lloyd/papers.html
[21:34:06] <anonimasu> the velocity stuff rocks..
[21:34:07] <robin_sz> heh .. goo d links page
[21:34:48] <anonimasu> robin_sz: I am just curious, would it be possible to do the tp calcs on a FPGA?
[21:35:06] <robin_sz> probably ... but .. err, why?
[21:35:32] <anonimasu> just curious..
[21:36:44] <SWPadnos> with sufficient precision, it becomes a single add per order of the formula (ie, third order = 3 additions)
[21:36:48] <robin_sz> ooh
[21:36:58] <robin_sz> the Lloyd of that paper ..
[21:37:08] <robin_sz> he has a library of motion control routines :)
[21:37:21] <robin_sz> http://www.cs.ubc.ca/spider/lloyd/rccl.html
[21:38:39] <anonimasu> neat
[21:38:47] <anonimasu> I wonder how much data I can stuff into memory
[21:38:56] <SWPadnos> depends on your beain
[21:38:58] <SWPadnos> brain
[21:39:06] <anonimasu> ah I was thinking about newVel
[21:39:09] <SWPadnos> (apparently, not much into mine :) )
[21:39:23] <anonimasu> cant do anything during the tcruncycle..
[21:39:39] <robin_sz> hmmm ... might need to email him to get it though
[21:39:43] <les> ah tp talk
[21:39:55] <robin_sz> hey les ..
[21:40:09] <anonimasu> just tmp[i++]=newVel;
[21:40:14] <SWPadnos> those are .tar.gx links on that page
[21:40:18] <SWPadnos> gz, that is
[21:40:19] <les> Paul spent ages here pouring over the Sonja thesis
[21:40:28] <les> and many others
[21:40:29] <robin_sz> ahh right
[21:40:33] <robin_sz> its a good paper
[21:40:37] <anonimasu> les: you've got it in print?
[21:41:13] <les> had one but Paul took it with him...a little light reading for the trip back
[21:41:19] <anonimasu> *jealous*
[21:41:20] <les> it's a pdf
[21:41:24] <anonimasu> yeah
[21:41:25] <robin_sz> les interestingly .. the Lloyd of the "Lloyd-Hayward blending algoritm" has a motion control sw library ;)
[21:41:33] <anonimasu> * anonimasu would love to have a hardcopy as in book..
[21:41:54] <les> I have seen some sw library stuff but not free
[21:42:09] <les> just print the pdf anders
[21:42:21] <anonimasu> les: yeah, but it's still not a book, with a real paperback :)
[21:42:33] <les> oh...yeah
[21:42:51] <anonimasu> I'll go to work and print it tomorrow as I dont have a fast printer at home
[21:43:10] <les> Cubic and quintic planners are not that bad mathwise either
[21:43:37] <anonimasu> I didnt look at the 3d motion planning stuff
[21:43:43] <anonimasu> just at the blending algorithms..
[21:43:47] <anonimasu> but they seem very very simple in theory
[21:43:56] <Jacky^afk> Jacky^afk is now known as Jacky^
[21:44:01] <les> think is no one can figure out how to plug new code into emc
[21:44:04] <Jacky^> evening
[21:44:10] <les> hi jacky
[21:44:37] <les> I figure at this point only hope for emc is to fix what's there
[21:45:01] <anonimasu> les: that's not true.. I think we've gotten pretty far with fixing the tp..
[21:45:03] <SWPadnos> man - hows this for a paper title: "Desingularization of Non-Redundant Serial Manipulator Trajectories Using Puiseux Series"
[21:45:21] <anonimasu> les: just that people are looking togther is a huge improvement
[21:45:33] <les> that's what I said...fix the tp...
[21:46:11] <les> I would like a quintic...but fixing the existing one would go a long way
[21:46:32] <les> at leasst it is cbic sub interpolated
[21:46:38] <les> cubic
[21:46:39] <anonimasu> if somone had understanding of the current tp and how to add a new one we could build a quintic one eventually..
[21:47:02] <anonimasu> paul's probably the only one right now..
[21:47:33] <les> yes
[21:53:11] <Jymmm> anonimasu what book?
[21:53:17] <anonimasu> Jymmm: the sonja paper
[21:54:31] <Jymmm> anonimasu : I mean book, like ISBN # ?
[21:54:52] <les> it is a mechanical engineering master's thesis
[21:55:03] <les> fairly easy to read
[21:55:15] <Jymmm> les if you're sadistic =)
[21:56:35] <Jymmm> anonimasu this one? http://www.mech.ubc.ca/~sonja/Macfarlane_ICRA.pdf
[22:01:25] <robin_sz> if we had a neat little block ..
[22:01:32] <robin_sz> where points went in
[22:01:39] <anonimasu> hm, the other one..
[22:01:44] <robin_sz> and commanded velocities came out
[22:01:58] <anonimasu> robin_sz: you cant call stuff from the tcruncycle.. ;)
[22:02:05] <anonimasu> not much atleast...
[22:02:08] <robin_sz> and it was neat and discrete ... writing an new tp and playing with it would be simple(er)
[22:02:36] <Jymmm> http://cadcam.eng.sunysb.edu/~ge/publications/Low-harmonic_rational_Bezier.pdf
[22:02:52] <robin_sz> then we could even experiment with pluggin in differetn tp's
[22:03:09] <anonimasu> robin_sz: the tp isnt that hard, it's algorithmizing the calculations..
[22:03:14] <robin_sz> as it stands ... its hard to see where it begins and ends, let alon how it works
[22:03:34] <anonimasu> well tpRunCycle is a good starting point..
[22:03:47] <anonimasu> ;)
[22:04:03] <les> the bug is prob in tpruncycle()
[22:04:18] <Jymmm> This looks semi interesting (fwiw) http://cadcam.eng.sunysb.edu/
[22:04:18] <anonimasu> les: wouldnt it be tcruncycle?
[22:04:24] <robin_sz> you know what will happen?
[22:04:29] <robin_sz> it will get fixed
[22:04:41] <robin_sz> and then no one will ever dare touch it ever again, lest they break it
[22:04:54] <les> tc yeah
[22:05:04] <les> robin: probably
[22:05:17] <Jymmm> robin_sz L Heh, I bet you Fred said that when he write it initially =)
[22:05:29] <anonimasu> he's comments surely helps :D
[22:06:14] <les> Fred says that tp/tc is very bad and he wished they could have the funding to do it right
[22:06:39] <les> it was intended as a first cut to just get some iron moving
[22:06:43] <les> nothing more
[22:06:47] <robin_sz> ive got 10 quid ...
[22:06:52] <robin_sz> will that help?
[22:06:54] <les> haha
[22:07:03] <anonimasu> I could contribute a couple of $100 ;)
[22:07:10] <anonimasu> would that get the tp fixed?
[22:07:11] <anonimasu> :D
[22:07:26] <anonimasu> how many minutes/secinds/nsecs can that pay fred ;)
[22:07:27] <RonB> might help in private industry ..
[22:07:29] <les> let's see...mmm...a man year to get it right
[22:07:38] <les> yeah that seems fine robin
[22:07:54] <anonimasu> bah, year ;D
[22:07:56] <Jymmm> "Five-axis Tool Path Generation for Sculptured Surface Machining using Rational B�zier Motions"
[22:08:07] <RonB> but with a government program you start off with a pile of zeros and put numbers in front
[22:08:10] <anonimasu> Jymmm: explain it to me please
[22:08:13] <les> RonB we have explored some private funding
[22:08:27] <les> but...
[22:08:34] <anonimasu> not enough?
[22:08:36] <RonB> have to be private amnagement too
[22:08:40] <les> do that and you will have a boss
[22:08:52] <Jymmm> anonimasu It's on the link I gave under publications
[22:08:56] <les> not that that is automatically bad
[22:09:11] <anonimasu> Jymmm: no time to look right now ;)
[22:09:12] <anonimasu> coding
[22:09:16] <RonB> well - everybody's got to eat
[22:09:23] <Jymmm> anonimasu no, fuckign around on irc =)
[22:09:39] <Phydbleep> les: Yeah, It's nice to have a scapegoat. :)
[22:09:41] <les> I wanted to use emc in manufacturing machinery in a very large corporation
[22:09:59] <les> but won't push the idea right now
[22:10:06] <les> for obvious reasons
[22:10:13] <anonimasu> Jymmm: seriously I am coding
[22:10:19] <anonimasu> * anonimasu fwaps jumm with a manual
[22:10:26] <Jymmm> anonimasu : seriously, I'm pulling your chain
[22:10:28] <anonimasu> fork();fork();fork();fork();fork();
[22:10:36] <Jymmm> * Jymmm wants to fork on the tbale!
[22:10:47] <RonB> real EMC question -
[22:11:06] <Jymmm> I haven't looked at the code, but is any of it extensively commented?
[22:11:12] <anonimasu> Jymmm: no
[22:11:16] <Phydbleep> anonimasu: Will you please quit forking around and get back to work?
[22:11:16] <Jymmm> less the TC/Tp
[22:11:19] <RonB> on steppers - is it usual parctice to have isolation on printer port?
[22:11:20] <Jymmm> anonimasu bummer.
[22:11:23] <anonimasu> /* if tc->preMax was calculated correctly this check is
[22:11:23] <anonimasu> redundant.
[22:11:23] <anonimasu> (Just because I'm paranoid doesn't mean they are not out to get me!) */
[22:11:30] <anonimasu> Jymmm: helpful isnt it?
[22:11:35] <Jymmm> anonimasu I remember that =)
[22:12:00] <Jymmm> RonB it's a 'good idea'
[22:12:05] <les> Isolation is ok but not required
[22:12:09] <Jymmm> RonB just like fuses
[22:12:31] <RonB> Well - for testing I may not use it
[22:13:06] <RonB> but - just thinking witha ll the electricity running about it may be a good idea
[22:13:27] <les> won't hurt
[22:13:43] <Jymmm> Especially if your testing initially building, isolation is best. saves you from your own mistakes =)
[22:13:57] <les> I don't use steppers for machine tools...but I use them for other things.
[22:14:15] <anonimasu> well, test time
[22:14:36] <les> The initial product test stations for the encoder product we have done use steppers
[22:14:54] <les> plan was to go to emc...
[22:14:58] <les> but...
[22:15:19] <Jymmm> les alternative?
[22:15:52] <les> A commercial control...this is 100,000 units per yyear
[22:15:59] <Jymmm> ah
[22:16:07] <RonB> I'm finding a lot in Linux is not ready for pubic use ... or maybe I should say windoze users use
[22:16:27] <anonimasu> well general use..
[22:16:30] <Jymmm> linux is awesome as a server
[22:16:40] <Jymmm> nix
[22:16:42] <RonB> damn nice desktop
[22:16:44] <Jacky^> ghghghg
[22:16:51] <Phydbleep> RonB: A lot of linux is not for the faint of heart or the mentally challenged. :)
[22:16:51] <les> I can call it I guess...but I would be commiting suicide to try to use emc in a highvolume production environment
[22:16:52] <Jacky^> yeah !
[22:16:53] <Jymmm> eh, it's ok as a desktop
[22:16:57] <Jacky^> as wk
[22:17:01] <Jacky^> :)
[22:17:07] <les> I would like to...I wish I could.
[22:17:37] <RonB> * RonB mentally challenged could Phydbleep be referring to me?
[22:19:05] <Phydbleep> RonB: No, more like my boss... Give him an NT/ME/2K box and he's OK.. Set him at a console and you get a blank look. :)
[22:20:21] <RonB> there are a lot of winbloze "expurts" out there than know nothing more than virus installation programs
[22:21:07] <Jacky^> lol
[22:21:12] <RonB> had one come buy the other day and was amazed I ran no anti-virus on my windoze machine
[22:21:34] <RonB> was furthure astounded I ahd no porblems
[22:21:45] <Jacky^> I'm using an antivirus on linux ..
[22:21:46] <RonB> "I don't go there"
[22:21:54] <Jacky^> to sban windows machine on lan :P
[22:21:59] <Jacky^> scan*
[22:22:10] <Jacky^> very useful, clmav
[22:26:13] <RonB> well - out for a bit
[22:54:55] <robin_sz> I have a much simpler method ... I have linux boxes for things that matter
[22:55:05] <robin_sz> and windows boxes for things that don't
[22:55:29] <robin_sz> by not connecting the 'doze machines to the net, they seem to be safe
[23:11:47] <Jacky^> :-)
[23:13:01] <anonimasu> *yawns*
[23:14:14] <Jacky^> anooon :)
[23:15:11] <anonimasu> *hits the tp with a hammer*
[23:15:21] <Jacky^> lol
[23:19:11] <dpy> hi guys
[23:19:18] <dpy> is it possible to control a 6 wire bipolar stepper motor as if it were a 6 wire unipolar motor ?
[23:22:00] <cradek> nope
[23:22:16] <Jymmm> if you like popcorn
[23:22:30] <Jymmm> =)
[23:22:34] <cradek> bipolar motors don't have a center tap on the windings
[23:26:06] <anonimasu> hm, logging from tc.c is tricky..
[23:26:20] <robin_sz> here ,
[23:26:24] <robin_sz> have a bigger hammer
[23:26:27] <anonimasu> yeah
[23:26:32] <anonimasu> * anonimasu gives up for tinight
[23:26:34] <anonimasu> tonight
[23:27:00] <anonimasu> I need to call the routine from somwhere else to write the stuff out
[23:27:10] <anonimasu> any delay makes it too slow :)
[23:28:23] <anonimasu> well, no big deal..
[23:28:24] <anonimasu> tomorrow
[23:28:41] <LawrenceG> open file an ramdisk?... makes for quick writes, or is it printf lib call that is too slow?
[23:29:02] <anonimasu> yep
[23:29:12] <anonimasu> it's the sprintf..
[23:29:14] <anonimasu> that's too slow..
[23:29:34] <anonimasu> anything other then data[x]=newVel; is too slow..
[23:29:57] <anonimasu> must put the data in some global variable and read it out from somewhere else
[23:31:02] <anonimasu> LawrenceG: well, I am getting tired of reading tc/tp code for today
[23:31:09] <LawrenceG> nasty.... maybe dump array at some condition (program end?) ( zero velocity ?)
[23:31:19] <anonimasu> not possible
[23:31:27] <anonimasu> not possible to call from the tp
[23:31:28] <anonimasu> :)
[23:31:40] <anonimasu> the tp has a deadline for eveything it does..
[23:31:50] <LawrenceG> I know... I have been reading as well... working through the sonja report now...
[23:32:08] <LawrenceG> going blind... :}
[23:32:08] <anonimasu> I'll just dump it to memory somewhere.. and grab it :)
[23:32:42] <LawrenceG> hey custom mcodes!... start logging and end logging?
[23:32:47] <anonimasu> hehe
[23:32:58] <anonimasu> that wouldnt be bad
[23:33:16] <anonimasu> but way too much work :)
[23:33:33] <LawrenceG> would make it easy to find the dump code...
[23:33:45] <anonimasu> you mean where it breaks?
[23:34:10] <anonimasu> hm, the tp should log all moves..
[23:34:25] <anonimasu> err all steps and put them into a file..
[23:34:29] <LawrenceG> emc2 has user m codes... dont know how hard it would be in emc1
[23:34:35] <anonimasu> 1 : 0.1
[23:34:38] <anonimasu> like that
[23:35:06] <anonimasu> or just velocitites
[23:35:14] <LawrenceG> tp runs every ms, correct?
[23:35:17] <anonimasu> yeah
[23:35:33] <LawrenceG> or is that just servo loop?
[23:35:42] <anonimasu> the tp runs every servo cycle..
[23:35:50] <anonimasu> as far as I remember
[23:36:16] <les> it can
[23:36:38] <les> usual set ups run it 5-10 times slower than servo rate
[23:36:44] <LawrenceG> ok... how about writing to a system fifo.... not sure what the overhead is but it should be quicker than file i/o and you can have a slower process dumping the fifo in chunks to disk
[23:36:55] <dpy> cradek: a 6 wire motor is always unipolar ?
[23:37:08] <anonimasu> LawrenceG: I thought about that but I am still unsure if it's fast enough
[23:37:21] <anonimasu> it's the sprintf that takes time..
[23:37:24] <anonimasu> most of it..
[23:37:38] <LawrenceG> a binary dump would avoid that...
[23:37:54] <anonimasu> yeah..
[23:38:39] <anonimasu> I'll try it with a fifo..
[23:38:51] <anonimasu> LawrenceG: reading with a text editor/gplot directly is easy..
[23:38:58] <anonimasu> I'd rather not dump it as binary..
[23:39:40] <LawrenceG> push into fifo in binary, but have the puller convert to text gnuplot format
[23:39:57] <LawrenceG> and dump to disk
[23:40:10] <anonimasu> I'll see how much faster just having a fifo is..
[23:40:54] <LawrenceG> good luck... I'm trying to wrap my brain around this paper... I'm starting to twitch with all this talk of jerk!
[23:41:14] <les> heh
[23:41:27] <anonimasu> LawrenceG: when you've read through it and I've read through it we should have a talk :)
[23:42:30] <LawrenceG> hi les... the paper seems well written... so far, it has been a good intro 2 several motion strategies I had not really thought much about
[23:43:00] <les> sometimes you need jerk in a cnc path...just not every segment!
[23:44:14] <les> You must mean the sonja thesis...mine is just a simplified explanation of wht emc does now (or should do)
[23:44:24] <LawrenceG> my old driving instructor taught that we should try and drive a standard as if we had a hot cup of cooffee in our lap.... learning to double clutch really brought that home!
[23:45:07] <les> I can do a roll in an aircraft and not spill my coke.
[23:45:11] <les> sometimes
[23:45:13] <les> haha
[23:45:59] <les> We do zero g parabolas then the coke drifts around in globules.
[23:46:24] <LawrenceG> cool... if I was a passenger, I wonder what barf looks like floating around the cabin?
[23:46:44] <les> I know.
[23:46:50] <les> It is not good.
[23:47:04] <LawrenceG> hard not to get any on you :}
[23:47:07] <les> I'm here to tell you
[23:47:56] <les> one time I opened the door in flight. it was making me sick.
[23:48:05] <les> serious breeze
[23:48:22] <les> oh well
[23:48:33] <RonB> especially if it is a single engine and the power is on
[23:49:03] <LawrenceG> yep... I tried sky diving once.... kind of blacked out after letting go of the plane...
[23:49:15] <les> yeah. better than that smell when a kid passenger barfs though.
[23:49:38] <Jymmm> ewwwwwwwwww
[23:50:00] <les> skydiving...whoa...I am to chicken for that.
[23:50:05] <les> too
[23:50:14] <Jymmm> LawrenceG blacking out AFTER you let go of the plane is NOT a good thing
[23:50:29] <les> static line or tandem I hope
[23:50:46] <RonB> you can get one of those chutes that open on impact
[23:50:52] <les> hahaha
[23:50:58] <LawrenceG> it was great fun..... but it took about 10 seconds for me to get reoriented... thank goodness for the static line
[23:51:23] <les> I like zero g...funny feeling. We can do it for 15-30 seconds
[23:51:52] <les> your sinuses suddenly stop up.
[23:52:10] <LawrenceG> I figured nasa should have hired me as a barf dummy... I'd love to spend a week in space
[23:52:41] <les> I can take it for 15 seconds...100 days...I don't know
[23:52:46] <anonimasu> hehe
[23:53:23] <les> I don't get the full impact because I am tightly strapped in
[23:53:37] <les> I have done the m&m thing
[23:53:43] <les> floating around
[23:53:58] <les> also lots of dirt from the floor carpet
[23:55:10] <les> people's cheeks kinda puff up
[23:55:55] <les> in little planes we have to pay for it though.
[23:56:04] <les> 3 g pullout
[23:56:04] <anonimasu> ?
[23:56:09] <anonimasu> ouch
[23:56:23] <les> the dirt and m&ms drop down REAL fast
[23:56:42] <Jymmm> attack M&M's !
[23:56:46] <LawrenceG> especially if you are tto low
[23:56:49] <les> 3g is not bad
[23:56:56] <les> I have done 6
[23:57:03] <les> it hurts.
[23:57:22] <LawrenceG> when the wings fold at 4.5g?
[23:57:59] <les> The usual planes I fly are rates for +4.4 -1.5
[23:58:26] <les> I had some acro lessons though in a +6 -3 though.
[23:58:45] <les> Not with pax
[23:59:08] <les> more than about 1.5 and it's a barf fest.
[23:59:49] <les> I have never understood that
[23:59:50] <LawrenceG> makes one kind of nervous in wind shear and thunderstorms.... I hate watching the wings flap on an airliner (but its a good thing they do)