#emc | Logs for 2006-08-22

Back
[00:00:20] <jmkasunich> and I want to say "add 'foo bar blat' after line blah"
[00:00:20] <Jymmm> Talked to the RE Attorney this morning, It be tough to prove w/o having a copy of the will which isn't public document. Would require a lawsuit to be started and for us to subeanu the will.
[00:00:25] <SWPadnos> does halcmd spit out the offending line?
[00:00:52] <SWPadnos> Jymmm, look at the bright side - you didn't buy a Bridgeport :)
[00:00:57] <jmkasunich> you mean on an error?
[00:01:02] <jmkasunich> it only spits out the line number
[00:01:14] <SWPadnos> ok. maybe the offending line (or at least command) ould be a good thing
[00:01:15] <Jymmm> Our biggest problem is we have 21 birds, not easy to find a place for rental wise. We are looking at buying a MH.
[00:01:17] <jmkasunich> you were referring to echoing the line itself?
[00:01:21] <SWPadnos> yep
[00:01:22] <Jymmm> SWPadnos heh
[00:01:50] <jmkasunich> crap - I gotta get to the grocery store, was gonna do that an hour ago
[00:02:01] <SWPadnos> oh yeha - I have milk and donuts now
[00:02:10] <SWPadnos> time for a snack
[00:02:11] <jmkasunich> back in a bit
[00:02:17] <SWPadnos> see ya
[00:02:30] <SWPadnos> crap.
[00:02:39] <Jymmm> sour milk?
[00:02:51] <Jymmm> concrete donuts?
[00:02:57] <SWPadnos> as soon as I think I might be able to get my EMC development machine up and running again, my wife's computer dies
[00:03:08] <SWPadnos> no - haven't grabbed them yet
[00:03:16] <SWPadnos> argh
[00:03:28] <Jymmm> just ghost it
[00:05:16] <SWPadnos> the problem is that the easiest solution for fixing her computer is giving her the hardware in mine
[00:05:33] <Jymmm> video card?
[00:05:43] <SWPadnos> motherboard
[00:06:01] <Jymmm> ah
[00:06:08] <cradek> don't tell me you only have one computer's worth of parts around
[00:06:12] <Jymmm> are you in the boonies?
[00:06:13] <SWPadnos> the advantage is taht I'd be able to upgrade that machine ;)
[00:06:13] <cradek> good god man
[00:06:30] <Jymmm> are you in the technology boonies?
[00:06:35] <SWPadnos> I have only functional PC's around here (except hers)
[00:06:43] <SWPadnos> yes - Vermont
[00:06:52] <cradek> I've heard of that place
[00:07:00] <cradek> do you have stores there? or UPS?
[00:07:06] <SWPadnos> less than 1 mile from the factory that used to make 90% of the world's high density DRAM
[00:07:16] <SWPadnos> (wehen 64kbit was high density ;) )
[00:07:50] <Jymmm> SWPadnos Should I buy/ship you out some P3 600 machines for $15/ea ?
[00:08:04] <SWPadnos> I'll be placing an order to NewEgg as soon as my wife forgets that I spent $1400 on a ticket to Poland ;)
[00:08:25] <SWPadnos> it may be a while
[00:08:42] <cradek> eek
[00:08:50] <SWPadnos> that's what she sait
[00:08:52] <SWPadnos> said
[00:09:00] <Jymmm> The computers are ship, just think you'ld lose out in the shipping.
[00:09:06] <cradek> P3 600 machines are great for emc
[00:09:09] <SWPadnos> she was trying to say something longer, but her throat constricted
[00:09:10] <Jymmm> s/ship/cheap/
[00:09:24] <Jymmm> cradek and no onboard video/audio either
[00:09:29] <SWPadnos> www.retrobox.com
[00:09:53] <Jymmm> SWPadnos that's the Poland thing?
[00:10:10] <SWPadnos> no - that's cheap P3 computers
[00:10:17] <SWPadnos> Poland is for a wedding
[00:10:26] <cradek> thought you were already married
[00:10:36] <SWPadnos> in fact, I should call him - I have no hotel reservations yet, and I leave in two weeks
[00:10:47] <SWPadnos> sho'nuf
[00:11:14] <SWPadnos> a friend for 20+ years is getting married there, so what the hell
[00:11:21] <cradek> fun
[00:11:27] <SWPadnos> should be
[00:11:29] <cradek> visiting old friends is nice
[00:11:36] <SWPadnos> would be better if my wife were coming along, but oh, well
[00:11:37] <Jymmm> SWPadnos Just an excuse to justify seeing Poland, huh =)
[00:11:43] <SWPadnos> shhhh
[00:11:46] <Jymmm> =)
[00:11:51] <cradek> but I avoid weddings whenever possible...
[00:12:01] <SWPadnos> heh
[00:12:02] <Jymmm> cradek including your own?
[00:12:18] <Jymmm> or was there a shotgun involved?
[00:12:24] <cradek> Jymmm: um do you know something I don't?
[00:12:26] <SWPadnos> it's funny - the first movie Sharon and I saw together was "Four Weddings And A Funeral"
[00:12:32] <SWPadnos> we used to call it "Five Funerals"
[00:12:38] <cradek> ha
[00:12:53] <Jymmm> cradek oh, I thought you were married for some reason
[00:13:21] <cradek> nope, not married, not single either
[00:13:36] <SWPadnos> "bonded"
[00:13:49] <Jymmm> cradek: and jepler doesn't count
[00:14:09] <SWPadnos> no - he writes software to do that
[00:14:16] <cradek> jepler's not really my type - too brainy
[00:14:21] <Jymmm> lol
[00:14:51] <Jymmm> * Jymmm bets he could guess cradek's "type"
[00:15:12] <cradek> let's not play that game
[00:15:43] <Jymmm> cute, big tata's, can cook, keep a home. and overall fun spirited
[00:16:04] <Jymmm> how close did I get?
[00:16:13] <SWPadnos> I'm sure it's the same as everyone else: good looking (subjective), fun to be with (subjective), interesting (subjective) ...
[00:16:37] <cradek> fwiw, swp is definitely closer
[00:16:48] <Jymmm> lol
[00:17:15] <Jymmm> I never meant cover of a magazine.
[00:17:41] <SWPadnos> covers of magazines aren't that fun to be with
[00:17:44] <Jymmm> what's that old saying... beauty is in the eye of the beholder
[00:17:58] <SWPadnos> "beauty is in the eye of the beer-holder"
[00:18:04] <Jymmm> after 2am
[00:18:12] <Jymmm> nobody's ugly
[00:18:24] <SWPadnos> "beer goggles"
[00:18:50] <SWPadnos> (it's funny that I remember those, because I really don't like beer
[00:18:53] <SWPadnos> )
[00:19:13] <cradek> beer is ok sometimes but coffee is better
[00:19:16] <Jymmm> you're not a drinker at all (for the most part) iirc
[00:19:25] <SWPadnos> Bailey's is about it
[00:19:32] <cradek> mmm
[00:19:39] <Jymmm> ever had a Bit O Honey?
[00:19:41] <cradek> coffee and bailey's is always better
[00:19:56] <SWPadnos> coffee and Bailey's is my preferred form
[00:20:02] <SWPadnos> milk is OK as well
[00:20:06] <SWPadnos> hmmm
[00:20:12] <Jymmm> Baileys and Butterscotch Schnapps. Very good even if you're not a butterscotch fan.
[00:20:15] <SWPadnos> too bad I just finished the donuts
[00:20:31] <cradek> what's your secret? seems you're always eating donuts or ice cream
[00:20:40] <Jymmm> Bit O Honey == Baileys and Butterscotch Schnapps.
[00:20:52] <SWPadnos> nice MB: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=260019736595
[00:21:24] <cradek> 4 pci express??
[00:21:43] <Jymmm> quad video?
[00:22:00] <cradek> no parallel?
[00:22:04] <SWPadnos> I don't think so - 2 x16, plus 1 x4 with an x8 slot
[00:22:41] <SWPadnos> it's basically the motherboard I have, but it has an Ht slot as well
[00:22:47] <SWPadnos> (HyperTransport)
[00:22:51] <SWPadnos> mine does have a parport
[00:22:56] <cradek> I've been a little tempted to get one of these and put two full consoles on it (running two X servers) for both of us to share
[00:23:07] <cradek> half the fan noise, HD noise, heat, etc
[00:23:28] <cradek> not necessarily this board, but one that handles two video cards
[00:23:32] <Jymmm> cradek two X servers?
[00:23:35] <SWPadnos> interestingly, there's a nice quiet power supply from PC Power & Cooling that's very quiet, and can run this board
[00:23:39] <cradek> sure why not
[00:23:41] <SWPadnos> $159
[00:24:07] <Jymmm> cradek you mean like dual consoles 2 kybd, 2 crts, 2 mice, ?
[00:24:11] <cradek> could use two dual-head cards even (I'm used to two screens now)
[00:24:14] <cradek> yes
[00:24:18] <SWPadnos> that board needs a larger than normal case though
[00:24:24] <SWPadnos> it's full EATX
[00:24:26] <cradek> it does look big
[00:24:41] <Jymmm> cradek sounds like you could use a vga terminal
[00:24:52] <Jymmm> or SVGA I suppose
[00:24:57] <cradek> ?
[00:25:21] <Jymmm> plugs into ethernet, and acts just like remote terminal.
[00:25:34] <cradek> oh X station? I have some of those, they're not fast.
[00:25:47] <cradek> they're from the old days
[00:25:57] <Jymmm> the newer ones allow you to upgrade the video cards
[00:26:04] <cradek> used to use one at work, no fan, no HD, totally quiet
[00:26:14] <Jymmm> yeah
[00:26:30] <cradek> the real killer is 8-bit color on mine
[00:26:44] <cradek> that used to be normal, now I doubt apps would even work right
[00:27:21] <Jymmm> http://www.imsltd.com/products/terminals.htm#book3
[00:27:33] <Jymmm> that was NOT the one I was looking for...
[00:28:05] <cradek> I use old laptops as X stations at home, bought a bunch the same, they boot from a floppy and load everything over the network
[00:29:01] <Jymmm> http://www.wyse.com/products/winterm/5150SE/index.asp
[00:29:06] <Jymmm> sorta like that one
[00:29:44] <cradek> interesting
[00:29:55] <cradek> probably no good as a primary machine, but might have its place
[00:30:26] <Jymmm> also found these (used wyse terminals "thin clients") http://www.interstateterminals.com/Terminals.html
[00:30:31] <cradek> 3270 emulation built in :-)
[00:31:24] <Jymmm> you know, if you can get pizza box cases, CF card to boot to the net wouldn't be a bad thing
[00:31:42] <cradek> or cdrom
[00:31:54] <cradek> you can do it with a floppy but it's very tight
[00:31:56] <Jymmm> CF I'd think be faster.
[00:32:07] <cradek> who cares how fast it is
[00:32:28] <Jymmm> instant on might be a nice touch
[00:32:57] <cradek> true but if the screens shut off, no harm in leaving them on
[00:33:29] <Jymmm> Yeah. I like the used laptop idea better. Seems suited for the task.
[00:34:06] <cradek> it's dirt cheap to buy laptops with no hard drive, no cdrom, no battery
[00:34:38] <Jymmm> Then you REALLY can toss in a CF card in place of the hdd and netboot
[00:34:50] <Jymmm> 8MB CF cards are cheap
[00:34:52] <cradek> sure
[00:34:57] <cradek> or a floppy :-)
[00:35:13] <cradek> yeah I bet the tiny cards are dirt cheap now
[00:35:33] <Jymmm> I bet ppl are giving them away if you ask.
[00:35:53] <Jymmm> I know I have an 8MB, 64MB and 110MB CF cards I don't use
[00:36:10] <SWPadnos> 110?
[00:36:11] <Jymmm> Hell, I have 30 pin simms for that matter
[00:36:13] <cradek> I have 10, 64, 128, I only use the 64
[00:36:25] <Jymmm> SWPadnos yeah, weird number, price was right though.
[00:36:40] <SWPadnos> I haven't seen any non-power-of-2 capacity before
[00:36:42] <Jymmm> I only use my 1GB cards
[00:37:04] <Jymmm> or my 512MB if It's tight.
[00:37:53] <Jymmm> maybe it's 128, dont know. All this "GET OUT" doens't have us thinking too straight the last couple of days.
[00:38:14] <SWPadnos> and to think that all I have is two 16M and one 512M card
[00:38:37] <SWPadnos> both xD though
[00:38:40] <Jymmm> They're for my Nikon D70
[00:38:41] <SWPadnos> XD
[00:38:43] <SWPadnos> heh
[00:38:52] <SWPadnos> I get a weird smiley for ex-dee cards ;)
[00:40:04] <Jymmm> heh
[00:40:38] <SWPadnos> damn: http://www.retrobox.com/rbwww/home/unit_view.asp?id=1735397&bin_id=world
[00:41:34] <SWPadnos> they have >200 computers that have PIII-600 or higher, 512M or more, and are $200 or less
[00:41:55] <SWPadnos> and 40-60 that are $100 or less
[00:42:25] <cradek> machines like that are everywhere
[00:42:41] <SWPadnos> I guess that'ws why they have them
[00:42:50] <cradek> well usually it seems like you have to add ram
[00:42:56] <cradek> but pc133 ram is everywhere too
[00:43:02] <SWPadnos> these have 512M, so that should be OK
[00:43:07] <cradek> yep
[00:43:16] <SWPadnos> and the CD-ROM is included
[00:43:26] <cradek> a disk?
[00:43:39] <SWPadnos> CD-ROM drive, and ~20G hard drive
[00:43:42] <cradek> wow
[00:43:53] <SWPadnos> did the link not work?
[00:44:02] <cradek> heh, I didn't try it
[00:44:04] <SWPadnos> here's the cheap one: http://www.retrobox.com/rbwww/home/unit_view.asp?id=1698578&bin_id=world
[00:45:01] <cradek> perfect
[00:45:21] <SWPadnos> yeah - a good terminal
[00:45:52] <Jymmm> Oh, I found a solution for my pedant problems... RF remote control
[00:45:58] <SWPadnos> they have a few with DVD-ROM and 512M for $110-ish
[00:46:18] <cradek> I've still not needed a dvd so far
[00:46:20] <SWPadnos> Jymmm, I'd be leery of that
[00:46:22] <jmkasunich> back
[00:46:25] <SWPadnos> hi jmk
[00:46:34] <jmkasunich> hi
[00:46:37] <SWPadnos> a DVD-ROM is good if you want to watch DVDs ;)
[00:46:55] <cradek> true I'm sure
[00:47:02] <Jymmm> SWPadnos, Yeah, I know. it's 300MHz Hopefully there won't be many garage doors within a MH park.
[00:47:06] <SWPadnos> there's now software that's distributed on DVD, but not that you'd run on that kind of machine
[00:47:46] <SWPadnos> not that - machine noise could cause the PC to lose the "key release", depending on how the remote and the PC side software are set up
[00:48:26] <Jymmm> SWPadnos: Oh, I'd only use it for jogging and zeroing out the axis'es
[00:48:38] <jmkasunich> key release on jog...
[00:48:40] <SWPadnos> jog + missed key release = travel without stopping
[00:48:43] <cradek> I agree, any wireless seems like a really bad idea
[00:48:52] <jmkasunich> push button, axis moves... let go, axis keeps moving
[00:49:18] <jmkasunich> of course, the most likely panic response would be to hit the key again, and you _might_ not miss that key release message
[00:49:18] <SWPadnos> I'm not sure that that kind of problem would happen, but I see it as a possibility
[00:49:24] <SWPadnos> true
[00:49:47] <jmkasunich> well, I should get boring
[00:50:02] <SWPadnos> man - they've got P4-2.8 GHz machines with 512M and CD-RW/DVD-ROM for $276
[00:50:06] <SWPadnos> you are boring ;)
[00:50:15] <cradek> heh their cheapest machine is 98c
[00:50:24] <SWPadnos> you achieved your goal rapidly :)
[00:50:35] <Jymmm> jmkasunich: See, I wasn't the smartass this time, SWPadnos was. (Though I was thinking about it =)
[00:50:40] <SWPadnos> heh
[00:51:00] <SWPadnos> .geez - that $0.98 machine could run a headless EMC
[00:51:07] <SWPadnos> just about
[00:51:10] <cradek> sure
[00:51:16] <jmkasunich> .98 + shipping
[00:51:19] <cradek> free for shipping
[00:51:19] <SWPadnos> I'm sure the $1.82 one could :)
[00:51:34] <cradek> but if you find them locally they're just free
[00:51:43] <Jymmm> craigslist.org
[00:52:26] <cradek> it was only a few years ago 300-400 MHz machines were our daily workers, they're still the same speed today
[00:52:31] <SWPadnos> it's sad that an equivalent to my panel PC is $25.38 or so
[00:52:59] <SWPadnos> software is slower though, so you've got to keep up
[00:53:03] <cradek> hell it was only a few years ago that 33-40 MHz machines were our daily workers
[00:53:07] <Jymmm> Computers are the only thing that are worht thousands today, and hundreds tomorrow.
[00:53:18] <jmkasunich> cradek: thats more than a few now
[00:53:19] <SWPadnos> same as cars, until they're old and unused
[00:53:30] <SWPadnos> I think it was 15 years ago or so
[00:53:32] <Jymmm> 67 Camero SS
[00:53:33] <cradek> SWPadnos: that software still works just as well too
[00:53:43] <cradek> well not all of it I guess
[00:53:49] <SWPadnos> hey jmkasunich, where's that photo of the truck that was upended by the pallet jack?
[00:53:54] <jmkasunich> when I started with Reliance Electric in 1991, I was assigned a 33MHz 386
[00:54:05] <jmkasunich> www.metalworking.com
[00:54:19] <jmkasunich> one of the retired files directories, maybe 2002? 2003?
[00:54:30] <jmkasunich> named truckfront.jpg or something like that
[00:54:37] <SWPadnos> ok. thanks
[00:54:50] <cradek> jmkasunich: you got your very own cpu? they must not have been using unix :-)
[00:55:00] <jmkasunich> no they werent
[00:55:03] <jmkasunich> messydos
[00:55:16] <SWPadnos> I remember one of our professors getting a 486-33 - man that thing was fast!
[00:55:24] <jmkasunich> there were a couple unix boxes, used only for Mentor PC board layout IIRC
[00:55:26] <cradek> when I started at this company about ten of us shared a 40 MHz 68040
[00:55:34] <jmkasunich> sparc 2's or something like that
[00:55:35] <SWPadnos> I think that's the one who has the second patent ever on a piece of software
[00:55:43] <cradek> it was old, but it still worked
[00:56:08] <Jymmm> http://www.interstateterminals.com/Terminals.html
[00:56:10] <cradek> before that I did cad on a sun 386i? i386?
[00:56:15] <Jymmm> http://www.metalworking.com/dropbox/_2003_retired_files/TruckBack.jpg
[00:56:24] <Jymmm> http://www.metalworking.com/dropbox/_2003_retired_files/TruckSide.jpg
[00:56:32] <jmkasunich> dat beez da one
[00:56:37] <SWPadnos> yep - found those
[00:56:51] <SWPadnos> though the back photo wouldn't load correctly from the file listing
[00:56:55] <SWPadnos> thanks
[00:57:03] <cradek> just recently I fired up that same 68040 to read some very old tapes!
[00:57:08] <cradek> still works fine of course
[00:58:01] <SWPadnos> aha - they have a capital D in dropbox on the directory link
[00:58:22] <jmkasunich> gotta yell at steve stallings about that
[00:58:36] <SWPadnos> hmmm - nope. I got a bad version of the directory listing
[00:58:40] <SWPadnos> now it works fine
[00:59:10] <SWPadnos> that's so weird - TCP is supposed to be a reliable transport. I wonder how I get erroneous web pages
[01:01:02] <jmkasunich> pages plural? it happens a lot?
[01:02:01] <SWPadnos> I see it daily, I think
[01:02:15] <SWPadnos> sometimes it's on generated pages, like eBay search results
[01:02:22] <Jymmm> SWPadnos: TCP is, but you have to remember about dropped packets.
[01:02:44] <SWPadnos> you're supposed to get an error if a packet is dropped and no replacement arrives
[01:03:03] <Jymmm> it might take 15 dropped/resent packets to give the final results
[01:03:05] <SWPadnos> it's weird that in this case, I had one bad bit - a 'd' changed to a 'D'
[01:03:41] <SWPadnos> sure, so I'll have to wait longer, but if the browser is finished loading the page, then I should have all the information, and it should be correct, as far as a CRC32 can tell
[01:04:00] <Jymmm> SWPadnos assuming the http server isn't fubaring
[01:04:28] <SWPadnos> sure - I'm not sure I've seen it on static pages, so it could be any number of problems at the server enc
[01:04:30] <SWPadnos> end
[01:12:18] <cradek> I don't recall ever seeing that
[01:12:27] <cradek> maybe your machine is somehow bogus?
[01:12:47] <cradek> oh
[13:05:46] <steves_logging> steves_logging is now known as steve_stallings
[14:05:13] <NickServ> This nickname is owned by someone else
[14:05:13] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[15:20:17] <steve_stallings> steve_stallings is now known as steves_logging
[15:29:54] <mecheng> Hello!
[15:30:37] <mecheng> I stopped in because I have a hang-up I hope someone may be able to help with.
[15:31:27] <mecheng> The company I work for is switching to linux for workstations and I will get a sweet new computer with Pro NC.
[15:31:28] <alex_joni> mecheng: describe the problem
[15:31:57] <mecheng> However I can't seem to find any good posts or gcode editors or anything for linux.
[15:32:06] <mecheng> post processors I mean
[15:32:40] <mecheng> pro-e uses Gpost which doesn't work under linux
[15:33:26] <mecheng> any info on where to start or look for info would help.
[15:36:44] <cradek> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?Cam
[16:15:54] <mecheng> I will read through it today thank you very much.
[16:25:02] <cradek> I hope it helps (I don't know much about the stuff there)
[16:25:21] <cradek> this channel is about emc which is just a machine controller, most of us don't use cam
[16:25:41] <SWPadnos> not that we wouldn't like to ...
[16:26:02] <SWPadnos> I think you'll find that there aren't a lot of inexpensive options for CAM on Linux
[16:26:39] <SWPadnos> well, I should rephrase that - there are free options for CAM on Linux, but they probably don't interface to your CAD software, and the user interface won't be what you'd like
[16:27:08] <SWPadnos> for Linux CAM that is easy to use and interfaces to CAD, I think there are only expensive options
[16:27:19] <SWPadnos> (but I'd love to be wrong about that)
[16:29:22] <mecheng> yes that is what I have found as well.
[16:30:36] <mecheng> We already have good CAM software for linux (Pro-Engineer) just need to convert APT-CL files to G-Code.
[16:31:14] <mecheng> and maybe have a nice editor to read the gcode after it is converted.
[16:32:03] <SWPadnos> there was some work done on a G-code syntax hilighter for Kate (the KDE editor)
[16:32:22] <mecheng> yeah something like that.
[16:32:24] <SWPadnos> it's straight text, but at least it would show different kinds of commands in differnt colors
[16:32:36] <mecheng> That is exactly what I need
[16:32:51] <mecheng> I can have IT Dept customize it as well
[16:33:04] <SWPadnos> it didn't work well for me, but I may not have had the final version
[16:33:43] <mecheng> any way to get a hold of it?
[16:33:47] <SWPadnos> I'd love to see something that could collapse groups of similar codes - like a bunch of G1 moves (with a little +/- icon next to the group
[16:34:02] <SWPadnos> hmmm - I'm not sure where it is, except on some BDI discs
[16:34:29] <mecheng> no problem I will probably make something anyway
[16:35:05] <mecheng> haha that would be very cool (the +/- thing)
[16:35:18] <SWPadnos> the hilighting file format for Kate (and probably GEdit) is a relatively easy to understand XML file, with regular expressions for determining the tokens to hilight
[16:35:44] <mecheng> bbiab gotta go work for a sec.
[16:35:49] <SWPadnos> heh
[16:38:32] <SWPadnos> wow - this one looks great: http://www.cnc-syntax-editor.com/
[16:42:08] <mecheng> yeah no,linux support though. There are plenty of good free ones for windows.
[16:42:24] <SWPadnos> I like the "mirror" dialog box
[17:35:21] <mecheng> yes mirror seems nice but I never seem to use it.
[17:35:53] <mecheng> well time to go flip my part and hit cycle start again :)
[17:36:01] <SWPadnos> more useful for people who write code by hand, I imagine
[17:38:07] <Lerneaen_Hydra> 'lo
[17:39:04] <SWPadnos> hiya
[17:41:45] <mecheng> yeah hand writing code is a pain. Especially on a 5 axis machine :)
[17:42:02] <SWPadnos> hard to get the math right all the time ;)
[17:42:43] <mecheng> yeah and switch on +/- and KABOOOOOOOOM!!!
[17:43:39] <mecheng> APTOS seems interesting but old.
[17:45:19] <davidf> hi...
[17:45:32] <davidf> SWPadnos, Thanks for the help yesterday.
[17:45:43] <SWPadnos> there sure - did you get it going?
[17:45:47] <SWPadnos> -there
[17:46:23] <davidf> I went back to the transcript and digested it a bit. Now it make sense. Not that bad once you see a few simple relationships.
[17:46:33] <SWPadnos> yep
[17:46:52] <SWPadnos> you can probably decrease the BASE_PERIOD by quite a bit as well
[17:46:53] <davidf> I haven't started making changes yet.
[17:46:58] <SWPadnos> what's the CPU speed again?
[17:47:00] <davidf> But I wanted to ask...
[17:47:41] <davidf> The BASE_PERIOD I left at the default I think- 50 uS (50000 nan0Sec)
[17:48:33] <SWPadnos> right - but you can get faster (or smoother) operation if you decrease BASE_PERIOD
[17:48:37] <davidf> Q is: is there a rule of thumb or formula for determining the theoretical min. BASE_PERIOD based on uP speed or sdomething?
[17:48:47] <SWPadnos> yes, but it doesn't work ;)
[17:48:52] <davidf> ha.
[17:49:11] <davidf> Sorry I was typing while you asked-
[17:49:14] <SWPadnos> I think the limit is in the 6-10 microsecond range (for most computers)
[17:49:21] <davidf> uP speed = 733 MHz
[17:49:51] <SWPadnos> and it gets wirse as you move to slower computers (I think Alex had it down to 6 uS on an Athlon XP 1800 or so)
[17:49:53] <SWPadnos> worse
[17:50:04] <SWPadnos> it really depends on the chipset as well
[17:50:07] <davidf> So I could go 5 - 8 times faster?
[17:50:20] <SWPadnos> I have a Celeron 500 that can reliably do 20, and I'm pretty sure I've had it down to 16 or so
[17:50:34] <SWPadnos> you can probaably get to 20 easily enough
[17:50:51] <davidf> this is a P3 @ 733 (I think thats right)
[17:50:59] <SWPadnos> you can tell that you've gone too far if the machine locks up when you start EMC ;)
[17:51:13] <davidf> Just got it used awhile ago. Have to double check that.
[17:51:40] <davidf> OK, just like WQindows then. ;)
[17:51:47] <davidf> Windows.
[17:51:54] <SWPadnos> sort of
[17:52:10] <SWPadnos> you'll notice a slowdown in X when you're getting close to the limit
[17:52:44] <SWPadnos> I'd try 20, and con't bother going further if that works
[17:52:51] <SWPadnos> s/con't/don't/
[17:52:57] <davidf> why x only? First to process or what? Just curious.
[17:53:21] <SWPadnos> the problems occur when the execution of the base thread takes up too much CPU time
[17:53:51] <SWPadnos> if the thread period is shorter than the execution time, the machine locks up (it never finishes that thread, since it restarts before it's done)
[17:54:18] <SWPadnos> if the thread takes up a lot of PU time, it just makes applications slow - it's basically loading the CPU a lot, in the kernel
[17:54:23] <SWPadnos> s/PU/CPU/
[17:54:37] <davidf> I see.
[17:55:30] <SWPadnos> so if you go down to 25 uS, and X is still reasonably snappy, you can probably go further
[17:56:32] <davidf> OK I'll try that. Might as well start with an optimal system before I do all the calcs for MAX_VEL etc.
[17:56:35] <SWPadnos> I think the default used to be 20 or 25 uS, but it was changed to 50 when there was a bug in RTAPI that would take several microseconds per thread (for time measuring, ironically)
[17:56:58] <davidf> heh
[17:57:36] <SWPadnos> you should figure out what step rate you want, and choose the base period for that - no sense going much further than necessary, since it will impact the performance of other parts of emc (such as the UI)
[17:58:43] <davidf> OK. By the way, the "resonance I thought I had seems to be actually due to software kinematics settings. When I approach moving faster than the numbers allow, that's when I start seeing roughness.
[17:59:03] <davidf> OK, that makes good sense.
[18:00:35] <davidf> Main thing is getting the BASE_PERIOD up fairly near to the optimum (with your last note taken into consideration) so I can increase the uStepping rate to a good level.
[18:01:04] <davidf> That really helps a lot for quiet operation. BIG difference.
[18:01:24] <davidf> Even at 2K compared to 400.
[18:02:26] <davidf> Thanks again - I'll give it a try. I think I understand enough to get it right now. I really apprecite that. Immensely.
[18:02:59] <davidf> Later..
[18:07:43] <SWPadnos> no problem. good luck
[18:16:18] <davidf> SWPadnos, or anyone - one more question please?
[18:16:35] <SWPadnos> sure
[18:17:05] <davidf> What determines the velocity sent as the starting velocity on an axis during an accelerated move?
[18:17:24] <davidf> Is it the DEFAULT_VELOCITY in the TRAJ section or what?
[18:17:51] <davidf> Or just same always?
[18:18:02] <cradek> davidf: I don't understand the question
[18:18:13] <SWPadnos> err -that was what I was going to say :)
[18:18:17] <davidf> (& accel at MAX_ACCEL?)
[18:18:31] <cradek> hierr, hium, hi guys
[18:18:37] <cradek> sorry to jump in, I only read the last line or two
[18:18:42] <davidf> well, I guess the starting vel. is 0 of course.
[18:19:07] <davidf> But is the vel simply dependent on the acceleration?
[18:19:32] <SWPadnos> is your question "what is the default acceleration used on a move, and how does it change?"
[18:19:48] <davidf> hi er hello err howyadoin cradek
[18:20:15] <davidf> Well I guess I didn't think about the Q enough first.
[18:20:15] <cradek> wonder how I typed control-Us in there
[18:20:42] <SWPadnos> heh - I was wondering what the [\0x15] was for (I assumed it was a color thing - that whould be bright white)
[18:20:56] <cradek> acceleration is limited (for all moves) by the axis acceleration (from AXIS section) and the tooltip acceleration (from TRAJ section)
[18:20:58] <jepler> ctrl-U is "clear the line" in irssi
[18:21:01] <jepler> and many other terminal programs
[18:21:09] <cradek> not right then it wasn't
[18:21:13] <SWPadnos> ah
[18:21:13] <cradek> although it's working again
[18:22:04] <davidf> ok, so say you set the acceleration to 0. Then you cant make a move at any velocity, right?
[18:22:15] <SWPadnos> right
[18:22:16] <cradek> right
[18:23:12] <davidf> My question is really - Is there an independent velocity figure for the axis or is the vel at any given moment, including the start, entirely dictated by the time and Accel?
[18:23:38] <SWPadnos> I thikn the answer may be that the overall accel for the move will be governed in the same way as the velocity, except that there's no limit from the G-code
[18:24:13] <davidf> OK.
[18:24:42] <SWPadnos> for a single-axis move, the TP will accelerate at the maximum of the following, with the limitation that it won't try to accelerate at a rate that would overshoot hte endpoint:
[18:25:00] <SWPadnos> the following being the axis MAX_ACCEL and the TRAJ MAX_ACCEL
[18:25:11] <davidf> k
[18:25:29] <SWPadnos> cradek can correct me if I'm way off - he wrote it ;)
[18:25:33] <jepler> davidf: each servo cycle, emc commands the machine to a new position. The new position is such that you can move from the old position to it while respecting the velocity and acceleration limits.
[18:26:37] <jepler> davidf: however, "stepgen" issues all the pulses during a particular servo cycle at the same rate (+- 1 base period), not an accelerating or decelerating rate during a single servo cycle
[18:27:37] <cradek> at the start of a long move (when coming up to cruise velocity) you are restricted by just acceleration limits
[18:27:57] <davidf> OK so in a G01 say, the move starts at v=0 and simply increases velocity over time according to max_accel respecting the vel limits.
[18:28:02] <cradek> I think that might have been the question, still not sure
[18:28:13] <cradek> right
[18:28:21] <davidf> yes that was the Q.
[18:28:35] <cradek> and when you get near the end, you have to start decelerating at the same max_accel in order to stop in the right place.
[18:28:52] <davidf> 2B || !2B =Q. :)
[18:31:26] <pier_casa> hi everyone
[18:31:42] <SWPadnos> hello
[18:32:01] <pier_casa> could someone give me a hand with make emc2 pls?
[18:32:18] <SWPadnos> yep - ask away
[18:32:24] <pier_casa> thanks
[18:32:50] <pier_casa> I get an error "Depending rtapi/rtai_ulapi.c"
[18:32:58] <pier_casa> in make process
[18:33:14] <pier_casa> don't know how to handle it
[18:33:16] <SWPadnos> did you apt-get build-dep emc2 emc2-axis ?
[18:33:33] <SWPadnos> (I'm assuming this is an ubuntu installation)
[18:33:49] <pier_casa> as a matter of fact no... I am still haing a go
[18:34:11] <pier_casa> with installing emc2 on my slack distro
[18:34:18] <SWPadnos> ah
[18:34:35] <jepler> please put the full error on http://pastebin.ca/ -- the message "Depending rtapi/rtai_ulapi.c" is not an error message, it is a normal step of compiling emc2.
[18:34:45] <pier_casa> ok
[18:34:52] <SWPadnos> well - I'm not familiar enough with the build process to help you there - jepler is the man
[18:35:13] <pier_casa> thanks all the same
[18:39:14] <pier_casa> jepler: http://pastebin.ca/144867
[18:43:10] <jepler> on my system line 299 of rtai_config.h reads
[18:43:10] <jepler> on my system line 299 of rtai_config.h r 299#include "/usr/src/linux-headers-2.6.12-magma/include/linux/autoconf.h"
[18:43:14] <jepler> oops
[18:43:20] <jepler> #include "/usr/src/linux-headers-2.6.12-magma/include/linux/autoconf.h"
[18:44:05] <jepler> this may indicate that there is a problem with your rtai installation, or that it requires an extra -I argument that we are not using
[18:44:43] <jepler> that #include on line 299 must find the file that is associated with the correct kernel, and the one in /usr/include/linux/config.h is generic, as described at http://ep09.pld-linux.org/~mmazur/linux-libc-headers/doc/FAQ
[18:48:49] <pier_casa> mine on line 299 reads: #include <linux/config.h>
[18:50:04] <pier_casa> perhaps the search path is not correct
[18:56:42] <jepler> it is true that we don't use the flags printed by '--lxrt-cflags
[18:56:50] <jepler> er, 'rtai-config --lxrt-cflags'
[18:59:35] <pier_casa> root@darkstar:/home/pier# find /usr/src/RT_linux-2.6.15.1/ -name config.h
[18:59:42] <pier_casa> returns
[18:59:50] <pier_casa> /usr/src/RT_linux-2.6.15.1/include/linux/config.h
[19:00:13] <pier_casa> and /usr/src/RT_linux-2.6.15.1/include/config/i2o/config.h
[19:00:26] <jepler> what does 'rtai-config --lxrt-cflags' print?
[19:01:25] <pier_casa> weird
[19:01:32] <pier_casa> permission denied
[19:01:44] <pier_casa> run as root
[19:02:09] <jepler> $ /usr/realtime-2.6.12-magma/bin/rtai-config --lxrt-cflags
[19:02:09] <jepler> -I. -I/usr/realtime-2.6.12-magma/include -O2 -I/usr/src/linux-headers-2.6.12-magma/include -Wall -Wstrict-prototypes -pipe
[19:02:14] <jepler> here's what it prints on my ubuntu system
[19:04:07] <jepler> well, you may wish to manually add -I/usr/src/RT_linux-2.6.15.1/include in Makefile's CFLAGS
[19:04:32] <pier_casa> ok
[19:07:46] <pier_casa> in the INCLUDE variable?
[19:08:14] <jepler> in CFLAGS
[19:08:15] <Lerneaen_Hydra> hi jepler
[19:08:29] <jepler> hi Lerneaen_Hydra
[19:10:08] <pier_casa> make went further now!
[19:11:46] <alex_joni> hi all
[19:11:48] <pier_casa> http://pastebin.ca/144938
[19:12:20] <alex_joni> sounds like the wrong include
[19:13:04] <alex_joni> seems like a missing -DULAPI for halui ?
[19:13:20] <Guest69086> Guest69086 is now known as DaeMoN
[19:13:34] <DaeMoN> DaeMoN is now known as ElBit0r
[19:13:38] <ElBit0r> ciao
[19:13:50] <alex_joni> hello
[19:13:58] <alex_joni> pier_casa: can you pastebin your Makefile.inc ?
[19:14:22] <pier_casa> ok right away
[19:16:36] <pier_casa> http://pastebin.ca/144946
[19:17:29] <jepler> I dunno -- my bitops.h doesn't seem to include <linux/config.h>
[19:17:30] <alex_joni> pier_casa: reading it now, just as a small matter, it might be wiser to try to run it in place first
[19:17:51] <jepler> maybe, but it won't fix the compile errors
[19:17:51] <alex_joni> then afterwards try to install it to the proper location .. I suspect it will save you some problems first
[19:18:06] <alex_joni> no, of course not.. but it helps dividing possible issues on the way
[19:18:41] <alex_joni> jepler: do you need to include /usr/realtime/include for CFLAGS?
[19:18:44] <alex_joni> seems odd
[19:19:43] <jepler> alex_joni: there are probably lots of things there that are superstition...
[19:19:56] <alex_joni> jepler: I hear ya ;)
[19:24:32] <pier_casa> perhaps it is not correct... but wouldn't a ln -s to /usr/src/RT_linux-2.6.15.1/include/linux/config.h /usr/include/linux/config.h do?
[19:25:22] <alex_joni> pier_casa: when doing make prepare_all, that should be done automatically I think (or at least something similar involving symlinks..)
[19:27:00] <alex_joni> pier_casa: I'm talking about the kernel make..
[19:27:09] <pier_casa> ok
[19:27:23] <alex_joni> but it's been quite a while, can't really recall
[19:28:05] <pier_casa> I did that long ago so I can't recall either
[19:28:44] <alex_joni> pier_casa: sorry, can't say I know what to tell you..
[19:29:02] <pier_casa> nevermind :)
[19:29:06] <pier_casa> thanks
[19:29:07] <alex_joni> but it seems odd that you get that message, it sounds like it thinks you're compiling a kernel module
[19:29:32] <alex_joni> which you aren't (halui is only user-space : defined by -DULAPI)
[19:30:15] <alex_joni> maybe jepler knows how to make it more verbose
[19:30:34] <alex_joni> maybe there is a -DMODULE in the cmd line
[19:30:46] <pier_casa> spooky thing is that I managed to compile once
[19:31:06] <pier_casa> but giving the ./script/emc command
[19:31:16] <pier_casa> on the old program returns
[19:31:30] <pier_casa> pier@darkstar:~/emc2$ ./scripts/emc
[19:31:30] <pier_casa> EMC2 - 2.0.1
[19:31:30] <pier_casa> ./scripts/emc: line 129: /usr/local/share/emc/tcl/bin/pickconfig.tcl: No such file or directory
[19:31:48] <alex_joni> oh, sounds like you didn't do 'make install'
[19:32:01] <pier_casa> so I decided (perhaps stupidly) to compile again
[19:32:01] <alex_joni> emc is compiled for installed, and you are trying to run it in place
[19:32:05] <alex_joni> try 'emc'
[19:32:12] <alex_joni> without the scripts/ before it
[19:32:25] <pier_casa> co
[19:32:29] <pier_casa> sorry
[19:32:29] <alex_joni> should be in /usr/local/bin/emc (if you did 'make install' )
[19:32:35] <pier_casa> ok
[19:33:04] <pier_casa> no.. there isn't such file
[19:33:08] <jepler> this will print the compiler commandline used: objects/hal/utils/halcmd.o
[19:33:11] <jepler> argh
[19:33:15] <jepler> I just can't paste right today
[19:33:27] <alex_joni> jepler: we can understand, you've been away lots :D
[19:33:28] <jepler> make -n objects/hal/utils/halcmd.o
[19:33:35] <alex_joni> pier_casa: try 'make install'
[19:33:43] <pier_casa> ok
[19:33:44] <jepler> on my machine it prints (will probably be cut off):
[19:33:45] <jepler> gcc-3.4 -c -I. -Ilibnml/linklist -Ilibnml/cms -Ilibnml/rcs -Ilibnml/inifile -Ilibnml/os_intf -Ilibnml/nml -Ilibnml/buffer -Ilibnml/posemath -Irtapi -Ihal -Iemc/nml_intf -Iemc/kinematics -Iemc/motion -Iemc/ini -Iemc/rs274ngc -I/usr/realtime-2.6.12-magma/include -I/usr/include/python2.4 -O -g -Wall -DULAPI -Wdeclaration-after-statement -DEMC2_BIN_DIR="\"/home/jepler/emc2/bin\"" -DHAL_RTMOD_DIR="\"/home/jepler/emc2/rtlib\"" -DMODULE_EXT="\".ko\"" h
[19:33:55] <alex_joni> halui.o
[19:34:07] <alex_joni> make -n objects/emc/usr_intf/halui.o ?
[19:34:58] <pier_casa> make install returns always root@darkstar:/home/pier/emc2/src# make install
[19:34:58] <pier_casa> Depending emc/usr_intf/halui.cc
[19:34:58] <pier_casa> In file included from /usr/include/asm/bitops.h:8
[19:35:03] <pier_casa> erc etc
[19:36:17] <jepler> did you alter your rtai installation since you first compiled emc2?
[19:36:55] <pier_casa> no... I compiled a new kernel 2.6.17.1 but now I am running the rtai patched one
[19:37:16] <alex_joni> pier_casa: did you re-run ./configure ?
[19:37:23] <pier_casa> yes
[19:37:38] <pier_casa> configure gives an encouraging message :)
[19:37:42] <alex_joni> that's the reason why it wants to compile again
[19:38:03] <alex_joni> what's in /usr/src/linux ?
[19:38:15] <alex_joni> the old (patched) or the new 2.6.17.1 kernel ?
[19:38:48] <alex_joni> I see KERNELDIR = /usr/src/linux in your Makefile.inc
[19:39:20] <pier_casa> /usr/src/linux points to
[19:39:39] <pier_casa> /usr/src/RT_linux-2.6.15.1/
[19:40:07] <alex_joni> ok, that sounds legit
[19:41:08] <alex_joni> can you try what jepler suggested?
[19:41:18] <alex_joni> make -n objects/emc/usr_intf/halui.o
[19:41:35] <pier_casa> already done
[19:41:41] <alex_joni> what does it print?
[19:41:41] <pier_casa> I'll try again
[19:42:02] <pier_casa> from src dir
[19:42:05] <pier_casa> returns
[19:42:06] <alex_joni> right
[19:42:21] <pier_casa> root@darkstar:/home/pier/emc2/src# make -n objects/emc/usr_intf/halui.o
[19:42:21] <pier_casa> Depending emc/usr_intf/halui.cc
[19:42:21] <pier_casa> In file included from /usr/include/asm/bitops.h:8,
[19:42:21] <pier_casa> from rtapi/rtapi.h:188,
[19:42:21] <pier_casa> from emc/usr_intf/halui.cc:29:
[19:42:23] <pier_casa> /usr/include/linux/config.h:1:2: #error "Compilation aborted. Please read the FAQ for linux-libc-headers package."
[19:42:26] <pier_casa> /usr/include/linux/config.h:2:2: #error "(can be found at http://ep09.pld-linux.org/~mmazur/linux-libc-headers/doc/)"
[19:42:29] <pier_casa> make: *** [depends/emc/usr_intf/halui.d] Error 1
[19:42:31] <pier_casa> sorry for flooding
[19:42:38] <alex_joni> pier_casa: no issue
[19:43:11] <alex_joni> what make version are you using ?
[19:43:35] <pier_casa> GNU Make 3.80
[19:43:37] <jepler> hm, yeah, that won't work if it doesn't get past 'Depending'...
[19:44:25] <jepler> well, remove the "@" from the lines under "Rules to make .d (dependency)" files in Makefile
[19:44:36] <jepler> the leading @ on each of 4 lines
[19:44:47] <pier_casa> ok
[19:44:51] <jepler> then it will print the commands
[19:46:53] <pier_casa> three lines...
[19:48:33] <alex_joni> pier_casa: care to share what it says now?
[19:49:25] <pier_casa> I am afraid I made a mess
[19:49:28] <pier_casa> root@darkstar:/home/pier/emc2/src# make
[19:49:29] <pier_casa> Makefile:109: *** commands commence before first target. Stop.
[19:49:58] <alex_joni> can you pastebin the whole Makefile ?
[19:50:04] <pier_casa> ok
[19:50:09] <pier_casa> sorry for that
[19:50:57] <jepler> pastebin may not make the error visible, because in Makefiles a tab is different from 8 spaces...
[19:51:14] <alex_joni> oh, right.. what editor did you use?
[19:51:24] <alex_joni> might be one of those that converts tabs to spaces
[19:51:36] <pier_casa> http://pastebin.ca/145011
[19:52:08] <alex_joni> pier_casa: that's still Makefile.inc, not Makefile
[19:52:18] <pier_casa> god... sorry
[19:52:23] <alex_joni> pier_casa: :)
[19:53:20] <alex_joni> gotta run for half an hour.. jepler knows this better than me anyways .. bbl
[19:53:45] <jepler> well maybe but my real work is also crying for my attention
[19:54:00] <pier_casa> thaks alex_joni
[19:54:21] <pier_casa> ok ok thanks
[19:54:42] <pier_casa> perhaps I might run in here again to get thi sorted out
[20:21:23] <alex_joni> pier_casa: back now
[20:46:15] <danex> hello all
[20:46:20] <alex_joni> hi danex
[20:47:10] <danex> alex_joni, I have not been on when your were for some time, must be the time difference
[20:47:31] <danex> How are your projects?
[20:47:45] <alex_joni> my projects?
[20:47:51] <alex_joni> not sure what you mean ;)
[20:48:03] <alex_joni> but most are well ..
[20:48:19] <danex> I think you were working on halui the last time
[20:48:32] <alex_joni> oh, yeah.. I like to think that's finished now
[20:48:38] <alex_joni> till someone proves me wrong ;)
[20:48:41] <danex> great
[20:49:15] <danex> I was told today that the machine I am working on now is going to Mexico
[20:50:06] <alex_joni> oh .. nice
[20:50:34] <danex> Nice for me, A free vacation to Mexico, to install and train
[20:50:59] <alex_joni> sure sounds nice ;)
[20:51:20] <danex> I have never been to Monteray
[20:52:06] <danex> Two weeks will be nice, and the company pays the bill for both my wife and I
[20:52:10] <alex_joni> danex: same here, if that counts for anything :)
[20:53:49] <danex> What is your local time now, I am in eastern US zone
[20:55:40] <alex_joni> Day changed to 23 Aug 2006
[20:55:42] <alex_joni> just now ;)
[20:55:51] <alex_joni> GMT+3 right now
[20:55:57] <alex_joni> 00:01 < alex_joni> GMT+3 right now
[20:56:03] <cradek> your clock's fast
[20:56:12] <alex_joni> cradek: ntpdate might be sleeping
[20:56:49] <alex_joni> ok
[20:57:00] <alex_joni> 2 more minutes till tomorrow :)
[20:58:03] <danex> cradek, did you see the note I left yesterday?
[20:58:18] <cradek> umm
[20:58:57] <danex> Had some storms yesterday, so the internet went up and down a lot
[20:59:23] <danex> anyway I got the brake signal working for each axis
[20:59:35] <cradek> great
[20:59:52] <alex_joni> danex: working as it should?
[21:00:16] <danex> It,s Wednesday for alex_joni now
[21:00:24] <alex_joni> danex: sure is
[21:00:42] <danex> Yes, but needs to be tweeked
[21:02:08] <danex> The question I have is about the min and max for wcomp. What are the values in
[21:03:10] <danex> I.E. pulses, volts, etc.
[21:03:15] <alex_joni> depends on what you want to compare
[21:03:22] <alex_joni> it's just a window comparator
[21:03:23] <cradek> if they're the pos ddt, they're in units/sec
[21:03:54] <danex> O.K.
[21:03:58] <alex_joni> you feed in a signal, then you have a min and max (if the signal is position, then unit is mm or inches, if it's speed then it's units/sec, etc)
[21:04:54] <danex> so based on what you intend to compare
[21:05:51] <alex_joni> right
[21:08:42] <danex> The other problem I have is that the axis overshoots by 10 micron +/- and then hunts till it is exact
[21:09:22] <cradek> servos?
[21:09:33] <danex> Yes
[21:09:35] <alex_joni> sounds like PID tuning not OK
[21:09:36] <cradek> that's just a tuning problem then
[21:09:56] <alex_joni> danex: first hunch is that P is too low, but I haven't done this that often
[21:10:45] <alex_joni> maybe a bit of FF1 can help too
[21:10:45] <danex> Some CNC i have worked on in the past have a parameter for inposition ?
[21:10:47] <cradek> might play with FF1 too (it's very sensitive)
[21:11:15] <alex_joni> danex: emc2 has too (HAL pin motion.in-position iirc)
[21:12:26] <danex> What effect doe FF1 have in relation to FF0?
[21:12:55] <alex_joni> FF0 is position feedforward, and FF1 is speed feedforward?
[21:13:12] <alex_joni> danex: I only read about it a few times.. so don't bet any on my word :D
[21:14:35] <danex> HAL pin motion.in-position iirc? any documentation on this one?
[21:14:57] <alex_joni> danex: try "halcmd show pin motion"
[21:15:12] <alex_joni> it will tell you the HAL pins starting with "motion.."
[21:15:22] <alex_joni> iirc = if I remember correctly
[21:16:24] <danex> I will have to look at that one in the morning, after the Fire Department lets us back in the building
[21:17:35] <danex> A forktruck driver hit a Gas tank, now it has to be inspected before we are allowed back in
[21:18:16] <danex> Might be 10:00 before I can get to the machine
[21:22:27] <danex> What computers do you suggest for emc?
[21:22:56] <alex_joni> danex: 7-800 MHzish
[21:23:02] <alex_joni> maybe P3 or better
[21:23:17] <alex_joni> very modern processors aren't that good
[21:23:37] <alex_joni> I am running one I really like (Athlon XP 1400+, a couple years old)
[21:24:20] <danex> In the past I have used AMD only
[21:24:45] <danex> They seem to work good
[21:41:08] <danex> alex_joni, sorry about the disconnect, had a power failure
[21:42:24] <danex> I am going to shut down now
[21:42:47] <danex> Thanks alex-joni and cradek for the help
[21:42:58] <alex_joni> bye
[21:43:03] <alex_joni> * alex_joni heads to sleep too
[22:51:37] <SWPadnos> testing notification for SWPadnos_