Back
[00:23:02] -!-
Loetmichel has quit [Ping timeout: 260 seconds]
[00:31:23] -!-
rob_h has quit [Ping timeout: 256 seconds]
[00:41:24] -!-
phantoxeD has quit [Ping timeout: 245 seconds]
[00:58:36] -!-
n2diy has quit [Quit: Ex-Chat]
[01:24:47] -!-
iwoj has quit [Quit: Computer has gone to sleep.]
[01:38:11] -!-
iwoj has quit [Quit: Computer has gone to sleep.]
[01:40:03] -!-
dimas_ has quit [Ping timeout: 245 seconds]
[01:50:40] -!-
robin_sz has quit [Ping timeout: 276 seconds]
[01:58:44] -!-
skunkworks__ [skunkworks__!~chatzilla@str-bb-cable-south-3-102.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[02:01:55] -!-
tjb1 has quit [Quit: tjb1]
[02:06:25] -!-
alex_joni has quit [Ping timeout: 252 seconds]
[02:07:33] -!-
alex_joni [alex_joni!~alex_joni@emc/board-of-directors/alexjoni] has joined #linuxcnc-devel
[02:07:33] -!- mode/
#linuxcnc-devel [+v alex_joni] by ChanServ
[02:15:42] -!-
lerman has quit [Read error: Connection reset by peer]
[02:16:33] -!-
lerman [lerman!~chatzilla@24-151-1-146.static.nwtn.ct.charter.com] has joined #linuxcnc-devel
[02:48:52] -!-
demacus_ has quit [Ping timeout: 260 seconds]
[02:55:07] -!-
tjb1 has quit [Quit: tjb1]
[03:02:26] -!-
tjb1-linux has quit [Ping timeout: 272 seconds]
[03:24:02] -!-
skunkworks__ has quit [Ping timeout: 252 seconds]
[04:07:08] -!-
adb has quit [Ping timeout: 240 seconds]
[04:19:19] -!-
adb [adb!~adb@178-211-226-40.dhcp.voenergies.net] has joined #linuxcnc-devel
[04:25:28] -!-
FinboySlick has quit [Quit: Leaving.]
[04:32:42] -!-
vladimirek has quit [Remote host closed the connection]
[04:48:45] -!-
ve7it has quit [Remote host closed the connection]
[05:03:03] -!-
Fox_Muldr has quit [Ping timeout: 265 seconds]
[05:10:22] -!-
SWPadnos [SWPadnos!~Me@74-92-8-208-NewEngland.hfc.comcastbusiness.net] has joined #linuxcnc-devel
[05:16:28] -!-
alex_jon1 [alex_jon1!~alex_joni@81.196.65.201] has joined #linuxcnc-devel
[05:19:59] -!-
Ekken [Ekken!ekke@hilla.kapsi.fi] has joined #linuxcnc-devel
[05:21:25] -!-
capricorn_one has quit [*.net *.split]
[05:21:25] -!-
alex_joni has quit [*.net *.split]
[05:21:25] -!-
theorbtwo has quit [*.net *.split]
[05:21:25] -!-
djdelorie has quit [*.net *.split]
[05:21:25] -!-
i_tarzan_ has quit [*.net *.split]
[05:21:25] -!-
dunz0r has quit [*.net *.split]
[05:21:26] -!-
MattyMatt has quit [*.net *.split]
[05:21:26] -!-
pjm__ has quit [*.net *.split]
[05:21:26] -!-
Rogge has quit [*.net *.split]
[05:21:26] -!-
fatpandas has quit [*.net *.split]
[05:21:26] -!-
archivist has quit [*.net *.split]
[05:21:26] -!-
anonimasu has quit [*.net *.split]
[05:21:27] -!-
demacus has quit [*.net *.split]
[05:21:27] -!-
dimas_ has quit [*.net *.split]
[05:21:27] -!-
cylly2 has quit [*.net *.split]
[05:21:27] -!-
Valen has quit [*.net *.split]
[05:21:27] -!-
Nick001 has quit [*.net *.split]
[05:21:27] -!-
RagingComputer has quit [*.net *.split]
[05:21:27] -!-
mozmck has quit [*.net *.split]
[05:21:27] -!-
Ekkeri has quit [*.net *.split]
[05:21:27] -!-
SWPadnos_ has quit [*.net *.split]
[05:21:28] -!-
Connor has quit [*.net *.split]
[05:21:28] -!-
GeorgeH has quit [*.net *.split]
[05:21:28] -!-
cmorley1 has quit [*.net *.split]
[05:21:28] -!-
Tom_itx has quit [*.net *.split]
[05:21:28] -!-
X704 has quit [*.net *.split]
[05:21:28] -!-
Farthen has quit [*.net *.split]
[05:21:28] -!-
jdhNC has quit [*.net *.split]
[05:21:28] -!-
jackc has quit [*.net *.split]
[05:21:28] -!-
cjdavis has quit [*.net *.split]
[05:21:28] -!-
Jymmm has quit [*.net *.split]
[05:21:28] -!-
jepler has quit [*.net *.split]
[05:21:29] -!-
awallin has quit [*.net *.split]
[05:21:29] -!-
cradek has quit [*.net *.split]
[05:21:29] -!-
CIA-68 has quit [*.net *.split]
[05:21:29] -!-
WalterN has quit [*.net *.split]
[05:21:29] -!-
crib has quit [*.net *.split]
[05:21:29] -!-
ReadError has quit [*.net *.split]
[05:21:29] -!-
morfic has quit [*.net *.split]
[05:21:29] -!-
Patang has quit [*.net *.split]
[05:21:29] -!-
frysteev_ has quit [*.net *.split]
[05:22:03] -!-
cmorley1 [cmorley1!~chris@S010600c09fc019c2.no.shawcable.net] has joined #linuxcnc-devel
[05:22:03] -!-
mozmck [mozmck!~moses@client-74.117.92.175.dfwtx.partnershipbroadband.com] has joined #linuxcnc-devel
[05:22:03] -!-
Rogge [Rogge!~IceChat77@mail.tormach.com] has joined #linuxcnc-devel
[05:22:03] -!-
jepler [jepler!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc-devel
[05:22:03] -!-
awallin [awallin!awallin@lakka.kapsi.fi] has joined #linuxcnc-devel
[05:22:03] -!-
cradek [cradek!~chris@emc/board-of-directors/cradek] has joined #linuxcnc-devel
[05:22:03] -!-
CIA-68 [CIA-68!~CIA@cia.atheme.org] has joined #linuxcnc-devel
[05:22:03] -!- mode/
#linuxcnc-devel [+v cradek] by sendak.freenode.net
[05:22:12] -!-
RagingComputer has quit [Ping timeout: 244 seconds]
[05:36:02] -!-
GeorgeH has quit [Quit: Leaving]
[05:38:36] -!-
psha [psha!~psha@213.208.162.69] has joined #linuxcnc-devel
[06:10:05] -!-
micges [micges!~micges@adcv247.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[06:17:39] -!-
WalterN has quit [Ping timeout: 245 seconds]
[06:19:07] -!-
mk0 has quit [Client Quit]
[06:47:55] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has joined #linuxcnc-devel
[06:48:14] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has parted #linuxcnc-devel
[06:56:39] -!-
mk0 has quit [Quit: LinuxCNC 2.5.0!]
[06:57:01] -!-
mk0 has quit [Read error: Connection reset by peer]
[07:04:49] -!-
mk0 has quit [Quit: LinuxCNC 2.5.0!]
[07:07:57] -!-
maximilian_h [maximilian_h!~bonsai@anon-186-166.vpn.ipredator.se] has joined #linuxcnc-devel
[07:08:22] -!-
maximilian_h [maximilian_h!~bonsai@anon-186-166.vpn.ipredator.se] has parted #linuxcnc-devel
[07:20:45] -!-
mk0 has quit [Quit: Ex-Chat]
[07:22:29] -!-
mk0 has quit [Client Quit]
[07:25:43] -!-
capricorn_one has quit [Quit: Konversation terminated!]
[07:36:09] -!-
mk0 has quit [Quit: Ex-Chat]
[07:38:33] alex_jon1 is now known as
alex_joni
[07:38:44] -!-
alex_joni has quit [Changing host]
[07:38:44] -!-
alex_joni [alex_joni!~alex_joni@emc/board-of-directors/alexjoni] has joined #linuxcnc-devel
[07:38:44] -!- mode/
#linuxcnc-devel [+v alex_joni] by ChanServ
[07:43:09] -!-
iwoj has quit [Quit: Computer has gone to sleep.]
[07:51:27] -!-
micges has quit [Ping timeout: 252 seconds]
[07:57:08] -!-
toastydeath has quit [Ping timeout: 245 seconds]
[08:00:09] cylly2 is now known as
Loetmichel
[08:01:32] -!-
Loetmichel has quit [Disconnected by services]
[08:01:37] cylly2 is now known as
Loetmichel
[08:03:08] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[08:05:13] -!-
micges [micges!~micges@efm146.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[08:05:54] -!-
pingufan has quit [Read error: Connection reset by peer]
[08:21:15] -!-
mozmck has quit [Ping timeout: 244 seconds]
[08:22:32] -!-
maximilian_h [maximilian_h!~bonsai@anon-163-144.vpn.ipredator.se] has joined #linuxcnc-devel
[08:22:36] -!-
maximilian_h has quit [Client Quit]
[08:22:45] -!-
mozmck [mozmck!~moses@client-74.117.92.175.dfwtx.partnershipbroadband.com] has joined #linuxcnc-devel
[08:29:28] -!-
linuxcnc-build has quit [Ping timeout: 252 seconds]
[08:31:46] -!-
hm2-buildmaster has quit [Ping timeout: 272 seconds]
[08:43:00] -!-
linuxcnc-build [linuxcnc-build!~linuxcnc-@184-96-169-202.hlrn.qwest.net] has joined #linuxcnc-devel
[08:44:10] -!-
hm2-buildmaster [hm2-buildmaster!~hm2-build@184-96-169-202.hlrn.qwest.net] has joined #linuxcnc-devel
[08:48:42] -!-
WalterN has quit [Read error: Connection reset by peer]
[08:51:41] -!-
pingufan has quit [Quit: Konversation terminated!]
[08:59:30] -!-
Aero-Tec has quit [Read error: Connection reset by peer]
[09:15:32] -!-
psha has quit [Quit: leaving]
[09:17:18] -!-
rob_h [rob_h!~rob_h@5ace7018.bb.sky.com] has joined #linuxcnc-devel
[09:20:49] -!-
mk0 has quit [Ping timeout: 250 seconds]
[09:32:00] -!-
Thetawaves has quit [Quit: This computer has gone to sleep]
[09:40:43] -!-
factor has quit [Read error: Connection reset by peer]
[09:44:55] -!-
A1Sheds has quit [Ping timeout: 260 seconds]
[10:08:29] -!-
Geissler has quit [Ping timeout: 245 seconds]
[10:13:38] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[10:49:38] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[11:34:38] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[12:09:30] -!-
dunz0r has quit [Remote host closed the connection]
[12:54:24] -!-
sumpfralle has quit [Quit: Leaving.]
[13:11:38] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[13:15:40] -!-
pingufan has quit [Quit: Konversation terminated!]
[13:30:12] -!-
mk0 has quit [Read error: Connection reset by peer]
[13:34:08] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[14:07:08] -!-
fliebel has quit [Remote host closed the connection]
[14:10:13] -!-
adb has quit [Ping timeout: 256 seconds]
[14:19:08] -!-
dunz0r has quit [Ping timeout: 240 seconds]
[14:31:10] <jepler> I'm looking at Sammel's latest grid patch
[14:31:29] <jepler> It looks like it places the grid with its origin at the g5x origin and always at 0 degrees
[14:31:48] <jepler> I wonder whether it should respect rotation or g92 to determine the origin
[14:32:40] <cradek> surely so
[14:33:13] <jepler> I also worry about showing it in perspective mode..
[14:33:20] <cradek> g5x, g92, and rotation together give you the coordinate system
[14:33:31] <cradek> why?
[14:33:42] <jepler> because you can't "measure" against it then
[14:34:23] <cradek> hmm
[14:34:38] <jepler> I assume that's the purpose of the grid
[14:39:12] <cradek> how do you select the grid scale?
[14:39:25] <jepler> in his patch there is a magic comment to set it
[14:39:51] <cradek> hmm, I don't like magic like that
[14:40:53] <cradek> we are close to needing a prefs screen. but so far they're all bools so the menu worked.
[14:41:31] <jepler> it seems entirely possible to me that the appropriate grid is related to the part program..
[14:42:02] <cradek> yes possibly
[14:48:58] <jepler> looking again at his patch, I think it's only shown in orthographic mode
[14:54:25] -!-
syyl has quit [Ping timeout: 256 seconds]
[15:11:50] <cradek> that solves the problem of needing an infinite number of lines to get a "full" grid
[15:12:06] <cradek> it also strikes me that the appropriate grid spacing might be related to the zoom level
[15:14:32] -!-
ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[15:21:32] -!-
joe9 [joe9!~joe9@c-24-99-80-97.hsd1.ga.comcast.net] has joined #linuxcnc-devel
[15:29:07] <jepler> the grid lines also stop at the machine limits
[15:29:14] <jepler> so you only ever need a finite number of them
[15:29:29] <cradek> ah, that's good
[15:48:06] -!-
Valen has quit [Quit: Leaving.]
[16:01:01] Connor1 is now known as
Connor
[16:07:13] -!-
sarariman_seb [sarariman_seb!~seb@67.51.249.130] has joined #linuxcnc-devel
[16:07:30] <sarariman_seb> good morning folks
[16:07:45] <cradek> hey
[16:08:23] <sarariman_seb> this is making me salivate with nerd lust:
http://fortcollins.craigslist.org/tls/2952112016.html
[16:08:35] <cradek> ooh
[16:08:49] <cradek> looks like a very nice size
[16:09:11] <sarariman_seb> yeah, pretty managable for a vmc
[16:09:20] <cradek> wonder if it has a high speed spindle
[16:09:28] <cradek> and what taper it is
[16:09:33] <cradek> and if it comes with tooling for that price
[16:09:41] <sarariman_seb> i mailed the guy asking, he seems to know next to nothing about it… :-(
[16:09:49] <sarariman_seb> i think a field trip is in my future :-)
[16:10:15] <sarariman_seb> he doesn't even know what model it is
[16:10:17] <cradek> yes obviously get in car, stop at bank, get cash, etc
[16:10:23] <sarariman_seb> haha
[16:10:51] <sarariman_seb> "help me lift it in the back of my truck"
[16:11:16] <sarariman_seb> i've never heard of leadwell before, do you know anything about them?
[16:11:37] <micges> sarariman_seb: hi
[16:11:42] <cradek> nothing
[16:11:47] <sarariman_seb> hi micges :-)
[16:12:12] <micges> sarariman_seb: I'll be retrofiting simmilar size machine in 4 weeks :)
[16:12:24] <sarariman_seb> cool!
[16:12:28] <cradek> do you know if it's linear rail or dovetail?
[16:12:47] <sarariman_seb> nope
[16:13:07] <sarariman_seb> looks like there are a couple of tool holders lying on the table in the first picture
[16:13:42] <cradek> ah you're right
[16:13:45] <cradek> BT30 maybe
[16:14:15] <sarariman_seb> is that the same as cat30?
[16:14:28] -!-
tiago has quit [Remote host closed the connection]
[16:14:28] <cradek> same taper, different flange
[16:14:56] <cradek> I really can't tell for sure from the picture
[16:15:03] -!-
Farthen has quit [Ping timeout: 250 seconds]
[16:15:39] <sarariman_seb> just what i need, another milling machine
[16:15:54] <sarariman_seb> then i just need another garage to put it in
[16:19:35] <sarariman_seb> aw shucks, i'll just call the dude and talk to him, that can't possibly hurt
[16:20:53] <cradek> the model is "westlake"
[16:22:29] <sarariman_seb> oh! i saw it on there but didnt realize it was the model name
[16:22:40] <sarariman_seb> bt-40 then probably
[16:23:09] <cradek> possibly says "westlake series II"? travel 22x16, "Meldas" control?
[16:23:28] <sarariman_seb> that sounds right
[16:23:51] <cradek> http://www.astratool.com/machines/ph4577.htm
[16:23:53] <sarariman_seb> the seller said the old dead control was a "mitsubishi meldas MO"
[16:24:13] <cradek> definitely BT, possibly 40
[16:24:30] -!-
robin_sz has quit [Ping timeout: 260 seconds]
[16:24:39] <cradek> possibly 8000 rpm (wheeee)
[16:24:44] <sarariman_seb> wooo!
[16:24:59] <sarariman_seb> hih, the one on astratool went for $3250
[16:25:04] <sarariman_seb> *huh
[16:25:10] <cradek> well it might have worked
[16:25:22] <sarariman_seb> … which would make it better than this one!
[16:25:31] <cradek> you're very unusual in not caring whether it works :-)
[16:25:47] <sarariman_seb> http://www.linuxcnc.org/index.php/english/component/kunena/?func=view&id=559&catid=30#559
[16:26:05] <sarariman_seb> i hope it doesnt work, then i won't feel bad retrofitting it ;-)
[16:26:16] <cradek> this one also came with a nascar bumper sticker
[16:26:26] <sarariman_seb> i wonder if i can sell the bp, if i buy the leadwell
[16:27:00] <sarariman_seb> there are a couple of bridgeports for sale in my area currently, so it might be hard… otoh, they're all manual
[16:27:30] <cradek> trying not to think about how it would fit in my basement...
[16:27:38] <sarariman_seb> hehe
[16:28:58] <cradek> even though this is a tiny vmc you might be shocked to see how big it is :-)
[16:29:52] <sarariman_seb> i'd guess about 6.5 feet tall, 4 feet deep, 8 feet wide or so? judging from the cart and gas bottle next to it
[16:30:50] <cradek> I bet it's 8x8x8'
[16:31:27] <cradek> although so is a bridgeport - it's just that the cube isn't quite so full.
[16:32:11] <sarariman_seb> i bet he doesn't have the nice wiring diagrams like you did
[16:32:30] <cradek> you usually have to trade dollars for those
[16:35:39] <sarariman_seb> if you can even get them… leadwell seems to have discontinued the westlake model
[16:37:32] <cradek> oh, just saw the linuxcnc.org thread
[16:37:49] <cradek> sounds like resolvers but very fast (10m/min rapids)
[16:38:30] <cradek> unfortunately not linear rail, since he mentions turcite
[16:38:46] <sarariman_seb> not sure if the one robh converted was a westlake, but it seems like a similar vintage at least
[16:39:21] <cradek> yeah, smells like the same machine, but not sure.
[16:39:39] <cradek> same control, same umbrella tool changer
[16:39:54] <cradek> same? X travel
[16:40:40] <sarariman_seb> the control panel looks the same too
[16:40:51] <sarariman_seb> same model of the controller possibly
[16:41:24] <cradek> I really bet your candidate machine will be resolvers too then
[16:41:32] -!-
sumpfralle has quit [Client Quit]
[16:41:34] <cradek> not that there's anything wrong with that
[16:41:44] * sarariman_seb google resolver
[16:42:17] <sarariman_seb> it's like an encoder, but analog?
[16:42:22] <cradek> mesa has a resolver version of the 6-encoder 6-dac card now
[16:42:32] <cradek> yes it's analog and absolute positioning
[16:42:58] <sarariman_seb> absolute within one revolution, i guess?
[16:43:06] <cradek> yes or half a rev
[16:43:11] <sarariman_seb> ok
[16:43:56] <cradek> which makes them quite immune to transient noise
[16:44:06] <cradek> they are a good idea
[16:45:16] <cradek> aha, it's spelled 7i49
[16:46:04] <sarariman_seb> didn't andy pugh add support for that recently? i've only been half paying attention
[16:46:20] <cradek> yes I think it either works fine or almost works fine
[16:47:29] <sarariman_seb> cool
[16:48:59] -!-
syyl_ws has quit [Quit: Verlassend]
[16:51:38] -!-
sumpfralle has quit [Client Quit]
[17:02:11] -!-
sumpfralle has quit [Quit: Leaving.]
[17:03:17] -!-
sumpfralle has quit [Client Quit]
[17:07:52] -!-
sumpfralle has quit [Client Quit]
[17:14:50] -!-
bedah has quit [Quit: bye]
[17:33:43] -!-
adb [adb!~adb@178-211-226-40.dhcp.voenergies.net] has joined #linuxcnc-devel
[17:34:41] -!-
phantoxe has quit [Remote host closed the connection]
[17:46:27] -!-
andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc-devel
[17:59:17] -!-
iwoj has quit [Quit: Computer has gone to sleep.]
[18:10:01] -!-
phantoxe has quit []
[18:13:41] -!-
motioncontrol has quit [Ping timeout: 248 seconds]
[18:20:56] -!-
erasmo [erasmo!~erasmo@62.244.154.19] has joined #linuxcnc-devel
[18:37:28] -!-
IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 11.0/20120310193829]]
[18:45:42] -!-
A1Sheds has quit [Changing host]
[18:58:06] <skunkworks> andypugh: didn't you create or setup a config using a fake index?
[18:58:19] <skunkworks> (software generated)
[19:00:10] -!-
iwoj_ has quit [Client Quit]
[19:02:35] -!-
iwoj has quit [Ping timeout: 255 seconds]
[19:10:34] -!-
jthornton has quit [Read error: Connection reset by peer]
[19:10:34] -!-
JT-Shop has quit [Read error: Connection reset by peer]
[19:11:03] -!-
JT-Shop [JT-Shop!~John@184.20.140.167] has joined #linuxcnc-devel
[19:11:06] -!-
jthornton [jthornton!~john@184.20.140.167] has joined #linuxcnc-devel
[19:13:05] -!-
motioncontrol has quit [Ping timeout: 252 seconds]
[19:13:39] <andypugh> skunkworks: I vaguely recall something along those lines
[19:13:57] <andypugh> But isn't it built-in to sim-encoder/
[19:14:00] <andypugh> ?
[19:17:59] -!-
jv4779 [jv4779!~jv4779@c-98-201-6-55.hsd1.tx.comcast.net] has joined #linuxcnc-devel
[19:18:29] -!-
vladimirek has quit [Ping timeout: 245 seconds]
[19:19:43] <jv4779> I am hoping someone can help me out with a custom realtime component and getting access to the current [TRAG]LINEAR_UNITS. I am dealing with an input of dots per inch and have the axis scale, but I need to know what units the axis scale is in.
[19:20:16] <cradek> how do you have the axis scale?
[19:20:29] <jv4779> passed in via a paramter
[19:20:51] <cradek> then do the same?
[19:21:03] <cradek> but beware it can be a number or a string
[19:21:19] <jv4779> I dont know where to get it from and the setp of [TRAG]LINEAR_UNITS is a string.
[19:21:46] <jv4779> I am not aware of any way to get a string into a component via parameters or pins
[19:22:07] <cradek> right, you'd have to do it at module load time, not via pins or params
[19:22:19] <cradek> I thought that was what you meant by "passed in"
[19:22:40] <jv4779> hmm, ok i can explore that. I don't have any module load options right now
[19:22:55] <cradek> pretty sure that is the only way to pass a string.
[19:23:21] <jv4779> i was hoping there was some major shared memory in the component that gave system parameters
[19:23:30] <jv4779> major=magic
[19:23:41] <cradek> no, they are totally different layers
[19:25:09] <jv4779> can you give any insight into why stepgen.rawcounts is a parameter instead of a pin? I can get the counts at base-thread if I run capture-position in base thread in effect giving me rawcounts as a pin, but that is just major overhead of something that is already there, just not a pin
[19:25:48] <jv4779> or a way to get a parameter from one component to another
[19:26:22] <cradek> no
[19:26:32] <jepler> I think basically rawcounts was envisioned as a debug aid only
[19:27:25] <CIA-68> 03jepler 07master * r074467ddda5d 10/configs/sim/axis/axis_foam.ini: foam: use appropriate file in place of splash screen
[19:27:26] <cradek> I don't know what you're doing, but I wonder if you could write your component in terms of position, not counts.
[19:27:54] <sarariman_seb> for a while there was an effort to convert hal parameters to hal pins, not sure what the status of that is now
[19:28:06] <jv4779> I set out to reengineer laser raster engraving under linuxcnc via pure gcode instead of using halstreamer. the more direct way was to build a bit mask of which steps are on and off.
[19:28:38] <jv4779> I can build the mask and as the axis sweeps I need to track the rawcounts in base-tread to toggle laser on and off
[19:28:40] <cradek> sarariman_seb: (we stopped because we always seemed to fuck it up)
[19:29:19] <jv4779> the original system used stepgen.position-fb but that required fp and capture-position in base-thread
[19:29:32] <cradek> jv4779: what will you do when a servo machine wants to run a laser? It seems like this really ought to be in terms of position.
[19:29:43] <cradek> or even just a machine with hardware stepgen
[19:29:46] <andypugh> jv4779: I am playing around with a way to not use G-code at all for laser rastering. But I have got a bit bogged-down with a jerk-free motion planner.
[19:30:14] <jv4779> if you have hardware stepgen or servo you are on your own majoing a component taht works :)
[19:30:33] <sarariman_seb> andypugh: do you mean a finite jerk planner? ;-)
[19:30:43] <jv4779> as far as I know you can't get position-fb at base-thread with most harware solutions anyways
[19:31:21] <cradek> that is probably correct
[19:31:26] <andypugh> The idea is to have a HAL component that outputs positions to a stepgen, synchronised with an image file loaded into shared-memory, which generates constat-speed rasters internally.
[19:31:34] <andypugh> sarariman_seb: Err, yes.
[19:32:43] <jv4779> I am solving the planner issue by sending bitmap data to the component (wrappered in a float) using M68/67. since I can send 53bits at a time the G1 segments are pretty long
[19:33:12] <jv4779> and the bitmap data is at img dpi and then scaled to steps internally, each bit is even longer
[19:33:20] <cradek> funky!
[19:33:45] <jepler> XY grid lines (for master):
http://pastebin.com/EQgwEp6q http://oi48.tinypic.com/245forq.jpg
[19:34:15] <cradek> jepler: I see they are inch squares. what do metric configs get?
[19:34:24] <jepler> It's still based on magic-comment
[19:34:32] <jepler> that's G21 ... (GRID,25.4)
[19:34:43] -!-
factor has quit [Ping timeout: 252 seconds]
[19:35:01] <cradek> I like the result but very much dislike the magic comment scheme
[19:35:18] <cradek> it's the opposite of discoverable
[19:35:38] <andypugh> jv4779: hmm, I think I might have a neater way. My realtim component has direct access to the whole image file.
[19:36:11] <sarariman_seb> it seems like the grid would be most useful if it was dependent on the ui view, not the program
[19:36:27] <jepler> sarariman_seb: I think cradek agrees with you
[19:37:10] <sarariman_seb> some program (heekscad maybe) had finer grid lines if you zoomed in between the regular grid lines, which was pretty handy
[19:37:24] <jepler> qcad has that as well
[19:37:28] <sarariman_seb> but i'm not doing the work, so i'll shut up now
[19:37:40] <jepler> this code doesn't consider what the current view extents are, so it can't do that
[19:37:49] <jv4779> andypugh: my goals were to allow mixed raster and vector in the same file, which could be generated externally from postscript or via a .py script loaded vi axis. pure gcode with M67/M68 was the only thing I could find to do it
[19:37:51] <sarariman_seb> yeah
[19:38:01] <cradek> someone else I know has misused magic comments smellily:
http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commitdiff;h=0fc41d50c1bce2bdecc8a90e6e5057af5153d66f
[19:38:43] <andypugh> jv4779: Do you want to look at my config? You would need to compile the realtime component, but then it is just a case of halrun -I -u laser.hal
[19:38:52] <sarariman_seb> at least you got to make a function called ARBARC
[19:39:07] <sarariman_seb> ugh, back to work for me
[19:39:10] <sarariman_seb> see you guys
[19:39:14] -!-
sarariman_seb has quit [Quit: Leaving]
[19:39:16] <jepler> see you
[19:39:19] <jepler> too slow
[19:39:26] <andypugh> It's very much a dedicated raster thing though, as it doesn't use any of EMC2 except HAL.
[19:40:05] -!-
ve7it has quit [Remote host closed the connection]
[19:40:11] <cradek> andypugh: seems like you could use limit3 as-is, and perhaps write limit4 later
[19:40:17] <jv4779> I think I have solved all the techinal problems for my solution, the last was getting linear_units, whcih I can do via a module load parameter
[19:40:27] -!-
DJ9DJ has quit [Read error: Operation timed out]
[19:41:13] <andypugh> jv4779: And the HAL file can read that in from the INI, in the usual way.
[19:41:23] <jv4779> I would to have a better way to get large amounts of data from gcode into a component, maybe a pin with array type or ability to do more than one float per M68
[19:41:24] <cradek> I wonder if hm2_<BoardType>.<BoardNum>.read can be done without fp
[19:41:59] <cradek> (the hostmot2 manpage does not say which funcs need fp)
[19:42:04] <andypugh> cradek: It was largely an excercise in GladeVCP/python/HAL. I don't even have a laser.
[19:42:06] <jv4779> getting float and extra processing out of base-thread was a goal too, but that isnt' possible with stepgen.counts the way it is
[19:43:14] <andypugh> cradek: Almost certainly not.
[19:43:15] <cradek> jv4779: well you could always change stepgen
[19:43:55] <cradek> andypugh: just wondered if one could get sub-servo-cycle fb position on a servo system
[19:44:06] <andypugh> cradek: Some hm2…read functions will do fp operations.
[19:44:12] <cradek> ah ok
[19:44:14] <cradek> then no
[19:44:30] <andypugh> Might be interesting to see which ones, though.
[19:44:51] <jv4779> I understand that my solition, and probably any software linuxcnc one, will only work with steppers
[19:46:00] <andypugh> jv4779: I think mine will work with servos. But how rather depends on how I handle >servo-thread pulse-rates to the laser (currently I just don't)
[19:46:44] <jv4779> you could probably extrapolate where the beam is between servo-thread cycles
[19:47:08] <cradek> or simplify your "servo" thread enough that it can be run fast enough
[19:47:16] <cradek> seems like andypugh's scheme is more along these lines
[19:47:38] <jv4779> I alrady have to crank up my servo thread for 1080in/min
[19:48:10] <jv4779> which isn't even fast by laser raster engrave standards, most are >150in/sec
[19:48:39] <cradek> if I was serious about this I'd work on a hardware-assisted scheme
[19:49:07] <cradek> you know, pass a servo cycle's worth of image data to the fpga which clocks it out according to the encoder input
[19:49:58] <cradek> .001 resolution at 150in/sec is only 150 samples per cycle
[19:50:28] <cradek> and transferring 150 bits to the fpga seems like nothing
[19:50:40] <jv4779> lots of options when you start doing the sub servo-cycle work in a fpga
[19:51:04] <jv4779> sending bitmap data isn't somethign that HAL excels at
[19:51:20] <cradek> that is very true
[19:51:28] <jv4779> cramming 53bits into a float is the best I could do
[19:51:44] <jv4779> hal float, ie a double
[19:52:15] <andypugh> jv4779: I loaded it onto a separate area of shared memory within the python GUI. Then the HAL component only needs to look for the shared-memory key to find the data.
[19:53:26] <jv4779> I am trying to stay with pure gcode to simplify the users process
[19:53:56] <andypugh> I am trying to eliminate G-code from the process altogether, as writing a raster in G-code is perverse.
[19:53:56] <jv4779> using a laser with linuxcnc is already so much harder than a commercial product with a print driver.
[19:54:25] <jv4779> how to handle touch-off and positioning if it is outside of axis ?
[19:54:53] <jv4779> I agree that it is perverse
[19:57:19] <andypugh> jv4779: That (and scaling) would be handled by the GladeVCP/Python in the GUI. Basically jog to bottom left, press a GUI button, jog to top-right, press a GUI button, fill that rectangle with the loaded image. I am not sure you need to home.
[19:57:25] <jv4779> on a completly different topic... most dsp laser cutters have an option to do an non-cuting outline of the current jobs extents, ie run box. I can see that axis shows the limits of the current gcode, but is there a way to get that in a O,M,or component to do the box automaticly ?
[20:00:12] <jv4779> my current idea is go write a sub O<box> into each gcode file and just run that from a gui button
[20:08:43] <jv4779> cradek>: do you know of an exmaple component in linuxcnc that gets a string via module init ?
[20:09:58] <andypugh> jv4779: look at bldc.comp
[20:09:58] -!-
fliebel has quit [Remote host closed the connection]
[20:10:41] <andypugh> cradek: DO you know what is allowed in the base thread? ie, is comparison of string values allowed?
[20:11:03] <cradek> sure it would be
[20:11:44] <andypugh> Actually, so far, I reckon pwmgen is base-thread safe
[20:12:52] <andypugh> Basically every hostmot2 sub-driver has a …read() function. If those are float-free for all the modules you have loaded, then I think you can call the Hostmot2_read in the base thread.
[20:17:42] <andypugh> pwmgen is safe as it doesn't have a read() function.
[20:20:18] <jv4779> to pass a float into a component at module load time it would have to be a string that I then convert to a float ?
[20:22:47] <andypugh> No, you can pass a float in directly.
[20:23:22] <andypugh> (or, rather, I think that what happens is that the string from the INI/HAL is converted to a float.
[20:24:11] <jv4779> I need to use RTAPI_MP_* defines to get parmas into the module ? I see INT, LONG, and STRING
[20:24:13] <andypugh> No, wait, I am wrong.
[20:24:23] <andypugh> Yes, there is no float.
[20:24:49] <jv4779> haha, I always seem to be going down the path less traveled
[20:25:07] <andypugh> So, yes, load it in as a string then convert.
[20:26:42] <jepler> I don't think there's strtod or atof in kernel space, either
[20:27:00] <jv4779> ugh
[20:27:52] <andypugh> jepler: Does the init code run in kernel space?
[20:27:58] <jepler> yes
[20:29:51] <andypugh> So, an iterated search through the string multiplying by 10, then dividing by 10^i if a decimal point is found?
[20:30:06] <jepler> oh, a proper strtod is much much worse than that
[20:30:12] <andypugh> Sounds like fun.
[20:31:02] <jv4779> I am just going to leave the floats are params, only need to get a string in which has to be done at load time
[20:36:19] -!-
erasmo has quit [Remote host closed the connection]
[20:37:03] <cradek> jv4779: units are mm or inch. why not just do the easy thing and make the user say metric=yes or metric=no.
[20:38:31] <jepler> How to read floating point numbers accurately
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.45.4152&rep=rep1&type=pdf
[20:40:04] <jv4779> i am just following the convention that if something is in the .ini, then use that in the .hal
[20:40:40] <andypugh> jepler: I can't help feeling that approach would be overkill in this context.
[20:41:19] <jepler> it depends how much you care whether "1.448997445238699" is correctly converted to double, or whether it is incorrectly converted to 1.4489974452386991.
[20:41:57] <jv4779> I care so I am going to let setp handle it
[20:42:23] <jepler> but mostly I mention it because I geek out about the deep level of understanding needed to do even the simplest-sounding things
[20:46:45] -!-
motioncontrol has quit [Quit: Sto andando via]
[20:49:48] <andypugh> I would tend towards the opinion that if I am writing my own HAL module to parse my own INI file, and I use a exponent notation in the INI and fail to handle it my component, then I deserve it to fail.
[20:50:26] -!-
Thetawaves has quit [Quit: This computer has gone to sleep]
[20:53:05] -!-
iwoj has quit [Quit: Computer has gone to sleep.]
[20:54:03] -!-
DJ9DJ_ has quit [Quit: bye]
[20:56:50] -!-
motioncontrol has quit [Remote host closed the connection]
[21:05:23] Mono is now known as
Guest91334
[21:28:29] -!-
theorbtwo has quit [Ping timeout: 245 seconds]
[21:28:39] theorb is now known as
theorbtwo
[21:35:02] -!-
sumpfralle has quit [Read error: Connection reset by peer]
[21:46:12] -!-
motioncontrol has quit [Quit: Sto andando via]
[21:50:04] -!-
Guest91334 has quit [Quit: Page closed]
[21:50:07] -!-
factor has quit [Read error: Connection reset by peer]
[21:50:23] <jv4779> can I allocate memory in a component ?
[21:50:44] <jv4779> realtime component I should say
[21:51:43] -!-
servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[21:52:28] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has joined #linuxcnc-devel
[21:52:30] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has parted #linuxcnc-devel
[21:57:06] -!-
syyl_ has quit [Read error: Connection reset by peer]
[21:57:08] <alex_joni> with hal_malloc you should be able to
[21:57:49] <jv4779> is it bad form to allocate stuff out of there that isn't going to be shared ?
[22:03:48] <andypugh> jv4779: No, you will generally need to hal_malloc in any component which isn't an obligate singleton.
[22:04:54] -!-
jepler has quit [Read error: Operation timed out]
[22:18:25] -!-
Farthen has quit [Client Quit]
[22:23:17] -!-
The_Ball has quit [Ping timeout: 248 seconds]
[22:24:18] -!-
jepler [jepler!~jepler@emc/developer/pdpc.professional.jepler] has joined #linuxcnc-devel
[22:41:14] -!-
i_tarzan has quit [Ping timeout: 252 seconds]
[22:49:41] -!-
FinboySlick has quit [Quit: Leaving.]
[23:37:46] -!-
micges has quit [Quit: Leaving]
[23:57:52] -!-
WillenCMD [WillenCMD!~cnc@174-125-9-87.dyn.centurytel.net] has joined #linuxcnc-devel
[23:58:38] <WillenCMD> Im going to sound like an idiot, but where does one find the AVR include files?
