Back
[00:14:06] -!- logger[mah] has quit [Remote host closed the connection]
[00:14:13] -!- logger[mah] [logger[mah]!~loggermah@mail.mah.priv.at] has joined #linuxcnc-devel
[00:14:45] -!- pjm has quit [Read error: Connection reset by peer]
[00:16:20] <cradek> andypugh: yes exactly - the radiused top is a "circle" just like an end mill. it's just that the "origin" might not be in the center of it
[00:18:03] <andypugh> Is the "origin" set by tool orientation?
[00:18:35] <andypugh> So you could use the "wrong" orientation for subtle effects?
[00:18:47] -!- asdfasd has quit [Ping timeout: 246 seconds]
[00:20:56] <cradek> yes the origin is the tool orientation. it can be at a corner or midpoint of a side of the square that circumscribes the circle
[00:22:43] <cradek> radiused tIp of course
[00:23:21] <cradek> for subtle effects like your part's arc and taper features are totally wrong, yes :-)
[00:24:15] <andypugh> :-)
[00:26:20] -!- kwallace [kwallace!~kwallace@tmb-233.sonnet.com] has parted #linuxcnc-devel
[00:27:21] <andypugh> Woot! After a night of hacking I am exactly back where I started. (but now all my code is laid out completely differently)
[00:28:39] -!- kwallace [kwallace!~kwallace@63.83.74.233] has joined #linuxcnc-devel
[00:42:49] -!- kmiyashiro has quit [Quit: kmiyashiro]
[01:11:19] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[01:11:56] -!- ve7it has quit [Remote host closed the connection]
[01:18:48] -!- Tecan has quit [Changing host]
[01:18:48] -!- rob_h has quit [Ping timeout: 252 seconds]
[01:24:33] -!- ybon has quit [Quit: WeeChat 0.3.8]
[01:40:33] -!- andypugh has quit [Quit: andypugh]
[01:43:09] -!- skunkworks has quit [Read error: Connection reset by peer]
[01:47:16] -!- mhaberler has quit [Quit: mhaberler]
[01:49:59] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[01:51:45] -!- skunkworks_ has quit [Remote host closed the connection]
[02:17:05] -!- cmorley1 [cmorley1!~chris@S010600c09fc019c2.no.shawcable.net] has joined #linuxcnc-devel
[02:20:02] -!- cmorley has quit [Ping timeout: 245 seconds]
[02:20:19] -!- servos4ever has quit [Ping timeout: 265 seconds]
[02:22:54] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[02:38:33] -!- abetusk has quit [Quit: Leaving]
[02:45:42] -!- kwallace [kwallace!~kwallace@63.83.74.233] has parted #linuxcnc-devel
[02:45:56] -!- mhaberler has quit [Quit: mhaberler]
[02:48:44] <KGB-linuxcnc> 03seb 05v2.5_branch b60c201 06emc2 10docs/src/code/ 10(5 files) * docs: add some architecture diagrams
[03:03:34] -!- mevon has quit [Quit: Ex-Chat]
[03:04:39] -!- adb has quit [Ping timeout: 244 seconds]
[03:10:12] -!- cmorley [cmorley!~chris@S010600c09fc019c2.no.shawcable.net] has joined #linuxcnc-devel
[03:12:43] -!- cmorley1 has quit [Ping timeout: 246 seconds]
[03:17:37] -!- cmorley1 [cmorley1!~chris@S010600c09fc019c2.no.shawcable.net] has joined #linuxcnc-devel
[03:19:38] -!- cmorley has quit [Ping timeout: 244 seconds]
[03:34:26] -!- capricorn_1 has quit [Quit: Konversation terminated!]
[03:54:44] -!- jst has quit [Quit: No Ping reply in 180 seconds.]
[04:00:05] -!- Keknom has quit [Quit: Leaving.]
[04:20:17] -!- Loetmichel has quit [Ping timeout: 252 seconds]
[04:38:53] -!- tjb1 has quit [Ping timeout: 245 seconds]
[05:24:23] -!- FinboySlick has quit [Quit: Leaving.]
[05:28:11] <KGB-linuxcnc> 03seb 05v2.5_branch 707d337 06emc2 10docs/ 10src/gcode/gcode.txt 10src/gcode/gcode_de.txt 10src/gcode/gcode_es.txt 10src/gcode/gcode_pl.txt * docs: fix a wrong link in the G43.1 example
[05:36:18] zz_satyag is now known as satyag
[05:40:59] -!- crib has quit [Remote host closed the connection]
[05:41:21] -!- jpk has quit [Ping timeout: 252 seconds]
[05:44:58] -!- cmorley1 has quit [Quit: Leaving.]
[05:50:44] -!- zzolo has quit [Quit: zzolo]
[05:51:32] -!- mutilator has quit [Ping timeout: 265 seconds]
[05:54:42] -!- i_tarzan has quit [Read error: Operation timed out]
[06:03:24] -!- Fox_Muldr has quit [Ping timeout: 260 seconds]
[06:30:22] -!- Cylly has quit []
[06:33:06] -!- psha [psha!~psha@213.208.162.92] has joined #linuxcnc-devel
[06:39:01] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[06:42:34] satyag is now known as zz_satyag
[06:44:25] <KGB-linuxcnc> 03chrisinnanaimo 05master f8396da 06emc2 10src/emc/usr_intf/gscreen/gscreen.py * gscreen -fix dro units preference not being saved
[06:44:25] <KGB-linuxcnc> 03chrisinnanaimo 05master 6d14b99 06emc2 10configs/sim/gscreen_custom/gscreen.glade * gscreen config -add a vcp box, change the button signal
[06:44:28] <KGB-linuxcnc> 03chrisinnanaimo 05master 86e1091 06emc2 10configs/sim/gscreen_custom/gscreen_handler.py * gscreen config -change the metric select button behavior
[06:58:01] -!- tjb1 has quit [Quit: tjb1]
[07:03:11] zz_satyag is now known as satyag
[07:17:56] -!- r00t4rd3d has quit [Read error: Connection reset by peer]
[07:18:26] -!- r00t4rd3d has quit [Changing host]
[07:21:29] -!- cmorley [cmorley!~chris@S010600c09fc019c2.no.shawcable.net] has joined #linuxcnc-devel
[07:47:04] -!- dhoovie has quit [Client Quit]
[07:49:01] -!- toastydeath has quit [Read error: No route to host]
[07:56:41] satyag is now known as zz_satyag
[07:59:36] -!- herron has quit [Ping timeout: 276 seconds]
[08:16:08] zz_satyag is now known as satyag
[08:27:10] -!- beawesomeinstead has quit [Max SendQ exceeded]
[08:27:11] -!- phillipadsmith has quit [Read error: Operation timed out]
[08:33:01] -!- beawesomeinstead has quit [Changing host]
[08:33:01] -!- beawesomeinstead has quit [Changing host]
[08:33:25] satyag is now known as zz_satyag
[08:39:35] -!- theos has quit [Read error: Connection reset by peer]
[08:45:25] zz_satyag is now known as satyag
[08:50:36] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.21] has joined #linuxcnc-devel
[08:59:58] satyag is now known as zz_satyag
[09:04:06] -!- kmiyashiro has quit [Quit: kmiyashiro]
[09:06:36] -!- vladimirek has quit [Remote host closed the connection]
[09:10:06] -!- rob_h [rob_h!~rob_h@027c00e2.bb.sky.com] has joined #linuxcnc-devel
[09:14:09] theos is now known as Guest80686
[09:16:53] -!- Guest80686 has quit [Ping timeout: 252 seconds]
[09:18:48] -!- Adventsparky has quit [Client Quit]
[09:24:42] -!- pingufan has quit [Quit: Konversation terminated!]
[09:26:40] -!- djdelorie has quit [Ping timeout: 246 seconds]
[09:27:48] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[09:45:46] -!- b_b has quit [Changing host]
[09:54:01] -!- djheinzNO has quit [Read error: Connection reset by peer]
[09:56:29] -!- asdfasd has quit [Ping timeout: 252 seconds]
[10:01:05] -!- theorbtwo has quit [Ping timeout: 255 seconds]
[10:40:42] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[10:56:29] -!- mackerski has quit [Quit: mackerski]
[11:00:06] -!- holst has quit [Quit: Leaving]
[11:08:14] -!- maximilian_h [maximilian_h!~bonsai@130.255.104.21] has parted #linuxcnc-devel
[11:12:42] -!- Flipp has quit [Ping timeout: 264 seconds]
[11:59:03] -!- paideia has quit [Read error: Connection reset by peer]
[12:25:14] -!- dhoovie has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
[12:29:54] -!- herron has quit [Ping timeout: 240 seconds]
[12:59:54] -!- b_b has quit [Changing host]
[13:11:32] -!- gambakufu has quit [Ping timeout: 252 seconds]
[13:12:59] -!- mozmck has quit [Quit: Leaving.]
[13:14:18] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[13:35:40] -!- b_b has quit [Changing host]
[13:49:10] <mhaberler> awallin: Gladevcp+rtapi+HAL on raspberry: with remote X display - works just fine
[13:49:22] <mhaberler> with local console: still researching
[13:58:11] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[14:31:13] <mhaberler> Gladevcp standalone apps on raspberry: work with local console, too ;)
[14:33:34] -!- emel has quit [Read error: Connection reset by peer]
[14:35:20] -!- paideia has quit [Quit: Leaving]
[14:46:15] -!- alpha1125 has quit [Quit: Computer has gone to sleep.]
[15:01:58] -!- the_wench has quit [Ping timeout: 246 seconds]
[15:02:14] -!- archivist has quit [Ping timeout: 240 seconds]
[15:09:37] zz_satyag is now known as satyag
[15:11:50] -!- syyl has quit [Ping timeout: 268 seconds]
[15:15:54] -!- automata has quit [Ping timeout: 240 seconds]
[15:16:12] theorb_ is now known as theorbtwo
[16:22:30] <awallin> mhaberler: good news!
[16:22:56] <awallin> by local console, do you mean a HDMI-display connected to the pi, with usb-keyboard/mouse ?
[16:23:03] -!- adb [adb!~IonMoldom@178-211-235-11.dhcp.voenergies.net] has joined #linuxcnc-devel
[16:26:00] -!- skunkworks [skunkworks!~chatzilla@str-broadband-ccmts-ws-26.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[16:26:20] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[16:26:53] -!- psha has quit [Quit: Lost terminal]
[16:27:39] satyag is now known as zz_satyag
[16:32:35] -!- Simooon has quit [Ping timeout: 255 seconds]
[16:32:47] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 16.0.2/20121025205401]]
[16:33:31] -!- markvandenborre has quit [Read error: Operation timed out]
[16:36:06] -!- bmwyss has quit [Quit: bmwyss]
[16:38:25] -!- micges [micges!~micges@adcs119.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[16:40:25] -!- tjb1 has quit [Quit: tjb1]
[16:49:29] -!- kwallace [kwallace!~kwallace@smb-55.sonnet.com] has joined #linuxcnc-devel
[16:54:24] <kwallace> cradek: This is what I tried and works with any fuss:
http://wallacecompany.com/tui/Screenshot-2.png
[16:54:45] <kwallace> One pass with, and one pass without compensation.
[16:55:52] <cradek> cool
[16:58:40] <cradek> it's kind of funny to see what happens to arcs - it doesn't always look intuitive
[17:03:43] -!- vladimirek has quit [Ping timeout: 244 seconds]
[17:06:03] -!- ve7it has quit [Remote host closed the connection]
[17:17:10] -!- tayy has quit [Remote host closed the connection]
[17:18:11] <kwallace> My application tries to make lathe roughing passes into a fillet and to allow for all or most reasonable parameters.
[17:22:11] <awallin> what compensation is that? looks like compensation only in the x-direction, not z?
[17:25:25] zz_satyag is now known as satyag
[17:26:28] <cradek> awallin: the compensation is for the tool radius and origin
[17:30:01] -!- mackerski has quit [Quit: mackerski]
[17:30:56] <kwallace> Just in case, I believe the compensation is more pronounced when the tool contact point is away from the normal South and West edge of the tool radius, which defines the tool's control point or touch off points. In my example, the contact point is South-West and has the most compensation, then moves West where there is no comp.
[17:32:13] <cradek> yes
[17:33:14] <cradek> the compensated arc has all different radius, center, and endpoints
[17:47:03] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[17:48:02] -!- b_b has quit [Changing host]
[18:01:41] satyag is now known as zz_satyag
[18:10:47] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[18:17:29] -!- hewball has quit [Ping timeout: 246 seconds]
[18:20:05] -!- psha [psha!~psha@213.208.162.92] has joined #linuxcnc-devel
[18:32:58] -!- alpha1125 has quit [Quit: Computer has gone to sleep.]
[18:36:36] -!- mazafaka has quit [Ping timeout: 252 seconds]
[18:36:59] -!- plushy has quit [Quit: Leaving.]
[18:48:47] -!- kwallace has quit [Read error: Operation timed out]
[18:49:39] -!- kwallace [kwallace!~kwallace@smb-168.sonnet.com] has joined #linuxcnc-devel
[18:52:52] -!- ve7it has quit [Remote host closed the connection]
[18:55:19] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[19:13:00] -!- grummund has quit [Ping timeout: 276 seconds]
[19:17:09] -!- yuvipanda has quit [Read error: Connection reset by peer]
[19:20:41] <mhaberler> awallin: around?
[19:21:50] <mhaberler> yes, HDMI screen, mouse, usb - or 'export DISPLAY:<ipaddr>:0' and run remote
[19:22:04] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc-devel
[19:24:32] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 16.0.2/20121025205401]]
[19:43:52] -!- syyl_ws has quit [Remote host closed the connection]
[19:50:44] -!- kwallace1 [kwallace1!~kwallace@smb-236.sonnet.com] has joined #linuxcnc-devel
[19:51:27] <awallin> mhaberler: ok thanks. now I'm just left wondering if the delivery will be 1 week or 1 month or longer...
[19:51:46] <mhaberler> slow.. this aint a cnc platform - this is HAL only
[19:51:55] <mhaberler> no motion, no gui, no interp
[19:52:16] <mhaberler> that is pushing the envelope a bit, esp soft stepping
[19:52:27] -!- kwallace has quit [Ping timeout: 260 seconds]
[19:53:08] <awallin> well HAL and some working SPI would be a good start
[19:53:33] -!- horatio_cromwell has quit [Quit: Leaving.]
[19:55:07] <mhaberler> that shouldnt be far off; you might gain some inspiration from ftp://ftp.berlios.de/pub/natld/rpi_cnc.tgz
[19:56:02] <mhaberler> it has a HAL i2c driver for the pi (I'd say 'in principle', otherwise I would habe integrated it and I have no i2c peripherals)
[19:57:03] <awallin> ok, I'll these things when I get the pi..
[19:58:31] <andypugh> How much work is motion?
[19:58:41] <andypugh> (I mean CPU work)
[20:00:11] <awallin> atom 330 runs linuxcnc with ease... but I guess the pi is slower
[20:01:54] <pcw_home> probably about 1/4 the speed
[20:02:18] <micges> idle motion loop is about 4-10k cpu instructions
[20:02:46] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[20:08:27] <mhaberler> xenomai shows cpu usage per RT thread; with SW stepgen on an atom, base-thread is around 20% and servo-thread around 4-5% CPU
[20:09:29] <mhaberler> the pi processor is about 1/6th of an atom core in term of 'umph' .. this isnt scientifc measurement though
[20:10:12] <awallin> so 1ms servo-thread with a lightweight GUI would be doable?
[20:10:18] <mhaberler> motion is ok I think but you run out of steam with a base thread
[20:10:21] <mhaberler> could be
[20:12:13] <mhaberler> UI's are work - Python, OpenGL, tkinter/tcl/tk versioning..
[20:15:05] <mhaberler> thanks to our non-existing networked interface API between HAL and motion, and splitting between task and motion is currently not an option and progress display too
[20:15:09] <andypugh> keystick ?
[20:16:13] -!- yoyoek1 has quit [Ping timeout: 252 seconds]
[20:17:11] <awallin> is keystick similar to touchy?
[20:17:33] <mhaberler> it's more similar to vi, I believe
[20:18:42] <awallin> ok :)
[20:19:44] <andypugh> Just change to Keystick in your INI and look, it's pretty minimalist.
[20:21:30] <awallin> nothing but sim/axis seems to work for me now..
[20:23:26] <andypugh> Keystick is running on sim/axis-mm for me.
[20:23:56] -!- rob_h has quit [Quit: Leaving]
[20:24:19] <awallin> I tried master on quantal so I guess it's not very stable... 2.5 might be better
[20:26:37] <andypugh> This is Master on Lucid
[20:27:27] <andypugh> I can't do much with it, no Home key on my keyboard, but it uses mainly the same keyboard shortcuts as Axis. And that is all it uses.
[20:27:57] <awallin> well I guess touchy is better and doable on pi
[20:28:18] -!- mhaberler has quit [Quit: mhaberler]
[20:28:38] <andypugh> Hmm, not sure, it uses Graphics.
[20:29:39] <andypugh> Screenshot:
http://imagebin.org/236630
[20:29:59] <andypugh> It's actually quite a nice bit of work.
[20:35:09] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[20:43:19] -!- mevon has quit [Read error: Connection reset by peer]
[20:51:54] <mhaberler> pcw_home: q on test rig - I'd try to get away without real hw for now - would't be integ.9 (integrator) a substitute for stepgen in velocity mode in your suggested setup? (background: mesanet drivers are RTAI only currently, and RTAI is rough to inject servo thread noise for me since its kernel threads; much easier with xenomai-user in a de-facto sim setup)
[20:52:06] <alex_joni> lcnc on this one next:
http://www.theregister.co.uk/2012/11/20/oldest_digital_computer_deboots/
[20:52:14] -!- vladimirek has quit [Remote host closed the connection]
[20:52:47] <mhaberler> that would be a problem per se: _remaining_ locked in to it would
[20:52:49] -!- psha has quit [Quit: Lost terminal]
[21:05:32] <alex_joni> whee, been a while since I saw people actually using tkemc/tklcnc
[21:09:18] -!- mevon has quit [Ping timeout: 245 seconds]
[21:13:02] -!- zzolo has quit [Quit: zzolo]
[21:15:06] -!- PCW [PCW!~chatzilla@99.88.10.65] has joined #linuxcnc-devel
[21:17:40] <cradek> a dekatron is not a valve, and they do not flash.
[21:18:17] <cradek> but that's wonderful that the original designer met the restorers
[21:19:52] <cradek> I hope they still had the tubes with it because I doubt there are 828 spare dekatrons existing in the world
[21:20:20] <PCW> mhaberler: I dont think integ will work since its running on the same clock, but you should be able to inject jitter with a jitter comp addf-ed before anything else
[21:21:24] <mhaberler> I understand the 'same clockcycle' thing
[21:21:42] <mhaberler> my idea was to inject jitter into thread_task from a noise source
[21:22:22] <PCW> I have one (dekatron) and I briefly had a calculator here that used dekatrons
[21:22:56] <mhaberler> that without jittercomp should show up in ferror noise, and it should improve if jittercomp is turned on (so the theory)
[21:23:37] <cradek> PCW: I built a clock from them (there are /10 and /12 versions) but never got the (triode) interstage amplifiers to work reliably. it would only keep time when you were watching it.
[21:23:52] <PCW> Yes (or PID error since IMHO ferror has a bug)
[21:25:09] <PCW> Never seen a /12 I think most of the ones Ive seen were from radiation counters (scalers)
[21:25:57] <cradek> PCW: they had two carry outs opposite each other, so you could also do /6 with them. They were probably mostly made for counting british money?
[21:26:17] <mhaberler> breaking german enigma code, I think
[21:26:18] <PCW> Wow
[21:26:37] <cradek> mhaberler: I don't think they're that old
[21:27:08] -!- motioncontrol has quit [Quit: Sto andando via]
[21:27:10] <mhaberler> ah right, that is the sucessor
[21:27:12] <PCW> l 50s, early 60s maybe
[21:27:37] <mhaberler> I visited Bletchley park a few years ago, geek heaven
[21:27:49] <cradek> mhaberler: yeah, I'd love to see it someday
[21:28:13] <cradek> I should try to fix up my clock one of these days. I doubt anyone else has actually made one.
[21:28:47] <mhaberler> they had the 'Tunny' - a tube based computer running off paper tape program; it looked a bit like the the one on the picture and that was used to break the Lorenz encoded typewriter
[21:28:49] <cradek> (lots of folks use them as a readout for a uC or in other equally silly ways, after the nixie clock fads)
[21:28:59] <mhaberler> rebuilt and almost running at the time
[21:29:06] -!- wboykinm has quit [Remote host closed the connection]
[21:29:14] <andypugh> Bletchley is very interesting. Enigma was actually broken with purely mechanical machines (bombes) designed by the Poles. Colossus was for breaking Lorenz.
[21:30:45] <andypugh> Arguably Collossus was the first programmable computer, but as they were broken up and hidden after the war the US may be forgiven for thinking that they invented the computer.
[21:31:01] <cradek> http://www.tube-tester.com/sites/nixie/datdekat/GC12-4B_ETL-baird/gc12-4b-etl-ba.htm
[21:31:41] <cradek> andypugh: it's silly to worry about who invented what type of computing machine first, but people are so very concerned about that
[21:32:04] <andypugh> The Germans are in with a shout too, with the Zuse machines.
[21:32:58] <mhaberler> still, given some 10.000 people overall worked in the BP decrypt pipeline I consider it an exceptional cultural achievement that _none_ of these talked about it until it was declassified in the seventies or so
[21:33:12] <mhaberler> nowadays you'll probably have it on cnn in a week
[21:33:57] <andypugh> Yes, that part is astonishing, and denied Tommy Flowers the recognition he deserved. (Not just for having the idea, but for building it despite being told not to)
[21:35:30] <mhaberler> I think I read just about everything about allied scientifc intelligence in ww2 what was ever printed; my father was sorta on the other side if you know what I mean
[21:35:53] <mhaberler> building jammers for american airborne radar
[21:36:04] <PCW> speaking of dekatrons
[21:36:05] <PCW> http://www.extremetech.com/extreme/140953-worlds-oldest-original-digital-computer-is-turned-back-on-after-61-years
[21:38:43] <cradek> a dekatron based machine is SO slow but generates little heat and uses little power - it must've been a very attractive technology.
[21:39:39] <cradek> do we know what they used for interstage amplification? I suppose it must've still have normal valves for that. but instead of two triodes per bit you had one (or two?) per decimal digit.
[21:40:30] <PCW> the Anita calculator used a couple of them
[21:41:09] <cradek> PCW: isn't that the one that only scanned the keyboard with the dekatron?
[21:41:49] <cradek> (maybe they used trigger tubes for interstage, but it couldn't have been very reliable if so...?)
[21:42:43] <andypugh> cradek: It might say here:
http://www.tnmoc.org/special-projects/harwell-dekatron-witch/restoration
[21:42:46] <PCW> I had one here briefly And just noticed the dekatron (i think it used little cold cathode thryratrons as well so maybe it dit only have one dekatron)
[21:43:36] <PCW> the one here was flaky...
[21:44:18] <PCW> But was just in its way to a museum probably never to be powered again
[21:44:37] <cradek> > Keeping the machine running long-term will eventually require a modification of these circuits to use a more reliable type of trigger tube.
[21:46:06] <PCW> well I dont _think_ anyone is making new dekatrons
[21:46:25] <cradek> the dekatrons didn't fail, the trigger tubes did
[21:46:47] <andypugh> There is that French chap who makes tubes by hand, perhaps he can help?
[21:47:03] <andypugh> (Have you seen the video?)
[21:47:26] <seb_kuzminsky> andypugh: yeah that dude is nuts
[21:47:31] <PCW> I have, its very nice
[21:47:33] <cradek> yes
[21:47:46] -!- micges has quit [Quit: Leaving]
[21:47:47] <andypugh> Mixed opinions here, I see. :-)
[21:47:54] <seb_kuzminsky> andypugh: hey you asked about the two copies of bitfile the other day
[21:48:01] <andypugh> Yeah.
[21:48:10] <seb_kuzminsky> one is the old one used by the old userspace tools that jmk wrote
[21:48:20] <andypugh> (And the double-include, too)
[21:48:22] <seb_kuzminsky> the other is the newer one i wrote for the hm2 in-kernel firmware loader
[21:48:47] <seb_kuzminsky> i think i'll remove the userspace stuff in the 2.6 timeframe, it's just clutter now
[21:48:53] <andypugh> My main question was whether having too files with the same name was a problem.
[21:49:15] <andypugh> (two)
[21:49:37] <seb_kuzminsky> hm, it could be, but i think in this case it's not because they
[21:49:55] <seb_kuzminsky> 're in different include directories and we manage that stuff with command-line arguments to cpp
[21:50:33] <seb_kuzminsky> i think the headers live in the same dir as the c files that use them, and the c files include with "" instead of <>, so they find "their" header file instead of any other
[21:50:49] <andypugh> I guess the include, then include with an include of the same file is also not a problem, as things work.
[21:51:06] <PCW> which reminds me, should mesaflash be included in the firmware distribution?
[21:51:23] <andypugh> I think it should.
[21:51:51] <andypugh> Though only if the firmware distro is going to include 5i25 / 6i25 bitfiles.
[21:52:26] <seb_kuzminsky> both of the bitfile.h files are protected from multiple inclusions by the "#ifndef MUMBLE" at the top
[21:52:28] -!- Yarl has quit [Remote host closed the connection]
[21:52:35] <PCW> (well and any newer cards like 7I80)
[21:52:49] <PCW> 7I76E
[21:54:00] <andypugh> seb_kuzminsky: I probably ought to start working out how do that sort of stuff.
[21:54:05] -!- micges [micges!~toudi@adcs119.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[21:54:47] <seb_kuzminsky> andypugh: look at the three lines that mention __BITFILE_H in the hm2 version of bitfile.h
[21:55:02] <seb_kuzminsky> they say: "ignore this file if you've seen it before"
[21:55:38] <andypugh> Yes, I see where the endif is now.
[21:55:57] <seb_kuzminsky> it's a very common idiom
[21:57:03] <andypugh> But... bitfile.c includes firmware.h before bitfile.h has chance to turn itself off.
[21:57:36] <andypugh> (But then, that also has the same stuff at the top)
[21:58:49] <andypugh> So, in general, you can include anything you think you might need, and not be too concerned that another header also tries to include it?
[21:59:20] <seb_kuzminsky> yes, in general, headers are written with that three-line idiom to make it safe to include them multiple times
[21:59:58] <seb_kuzminsky> it doesnt matter if firmware.h includes bitfile.h first, or if the main C file does it, in either case one will be first and will define __BITFILE_H, and after than any other include of bitfile.h will be a no-op
[22:05:13] -!- DJ9DJ has quit [Quit: bye]
[22:05:36] -!- FinboySlick has quit [Quit: Leaving.]
[22:12:56] <mhaberler> hm: folks dream up all sorts of things:
http://code.google.com/p/rt-8p8c/
[22:14:16] -!- Nick001-Shop has quit [*.net *.split]
[22:14:16] -!- tayy has quit [*.net *.split]
[22:14:16] -!- PCW has quit [*.net *.split]
[22:14:17] -!- Tom_itx has quit [*.net *.split]
[22:14:17] -!- morfic- has quit [*.net *.split]
[22:14:43] -!- PCW [PCW!~chatzilla@99.88.10.65] has joined #linuxcnc-devel
[22:24:33] -!- kwallace1 has quit [*.net *.split]
[22:24:34] -!- servos4ever has quit [*.net *.split]
[22:24:34] -!- mozmck has quit [*.net *.split]
[22:24:34] -!- beawesomeinstead has quit [*.net *.split]
[22:31:06] -!- horatio_cromwell has quit [Quit: Leaving.]
[22:32:40] -!- kwallace [kwallace!~kwallace@smb-236.sonnet.com] has joined #linuxcnc-devel
[22:33:41] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[23:04:01] -!- DearLeader has quit [Ping timeout: 252 seconds]
[23:07:16] -!- syyl_ has quit [Quit: Leaving]
[23:23:26] -!- Simooon has quit [Quit: Leaving]
[23:25:21] -!- Brian__ has quit [Quit: Leaving]
[23:30:43] -!- The_Ball has quit [Ping timeout: 268 seconds]
[23:37:36] -!- mhaberler has quit [Quit: mhaberler]
[23:41:12] <andypugh> Why don't they do these things _inside_ the project?
[23:42:49] <andypugh> Is anyone looking at 7i80 support?
[23:42:59] <micges> yes
[23:43:24] <andypugh> I thought you were, but didn't want to assume my memory worked :-)
[23:44:03] <micges> quite hard
[23:46:18] -!- adb has quit [Ping timeout: 264 seconds]
[23:47:19] <andypugh> Working hard, or hard to do? (I can believe both)
[23:48:39] <micges> hard to do, design of driver + rtnet + linuxcnc is incompatible (as my state of knowlege)
[23:49:17] <andypugh> Fitting it into the Hostmot2 abstraction I can see being tricksy
[23:49:41] <micges> yes
[23:50:02] <seb_kuzminsky> does the 7i80 have a register file that you read by explicit read commands from the pc? or does it have some other communication interface?
[23:50:03] <micges> I've tried to change lcnc to fit driver + rtnet and it failed
[23:50:19] <micges> now I'm changing hm2 to fit rtnet + lcnc
[23:50:37] <andypugh> That project linked earlier was using rtnet, but it isn't clear if it actually works.
[23:50:50] <micges> it works
[23:51:42] <andypugh> I was meaning the 8p8c project, rather than rtnet
[23:51:45] <micges> seb_kuzminsky: it has lbp16 inside udp
[23:52:08] <seb_kuzminsky> what's lbp16?
[23:52:11] <micges> andypugh: yes 8p8c
[23:52:22] <micges> seb_kuzminsky: simmilar to sserial
[23:52:29] <seb_kuzminsky> ah
[23:52:29] <andypugh> lbp is PCWs private command format
[23:52:44] <andypugh> I think LBP _is_ SSerial
[23:52:59] <micges> probably yes :)
[23:53:07] <seb_kuzminsky> sounds neat!
[23:53:15] <andypugh> Well, SSLBP is.
[23:53:45] <micges> real problem in first design was:
[23:53:46] -!- ve7it has quit [Remote host closed the connection]
[23:53:57] -!- plushy has quit [Quit: Leaving.]
[23:54:11] <andypugh> So, Seb, did you code-in provision for RTnet and SSLBP when you did hm2?
[23:54:22] <seb_kuzminsky> heh
[23:54:36] <andypugh> (I am smirking here, that wasn't a serious question)
[23:54:37] <seb_kuzminsky> yeah i got an email from you in 2012 with some suggestions for it, so i put it in ;-)
[23:55:04] <micges> 7i80 hm2 driver loads simmilar like 8p8c but rtnet communication MUST be done in rt task, but hm2 driver starts comm at load time ( to discover hw)
[23:55:37] <andypugh> PCW seems to be able to bring out hardware quicker than we can support it.
[23:55:58] <andypugh> Ah, that's my fault.
[23:56:37] <andypugh> The smart-serial probong stuff?
[23:56:52] <micges> andypugh: you're fault is that there is few thousands reads at start
[23:57:17] <micges> but even I disable sserial, there is quite few to read idrom and etc
[23:57:26] <andypugh> Thousands? If there are, then I must have messed up.
[23:57:58] <seb_kuzminsky> a few hundred i can believe easily, for the idrom and lots of pins and modules
[23:58:27] <micges> at first time there was 1500000 after 20 seconds :D
[23:58:43] <andypugh> I feel guilty that micges is wallowing in the morass of me learning C. (I think I might be 20% competent now)
[23:58:44] <micges> (with sserial)
[23:58:44] <seb_kuzminsky> well, 20 seconds is a long time
[23:59:06] <seb_kuzminsky> 20,000 invocations of hm2_read()
