Back
[00:04:13] -!- logger[mah] has quit [Remote host closed the connection]
[00:04:20] -!- logger[mah] [logger[mah]!~loggermah@ns1.mah.priv.at] has joined #linuxcnc-devel
[00:12:10] -!- ybon has quit [Ping timeout: 250 seconds]
[00:36:32] -!- Brandonian has quit [Quit: Brandonian]
[00:42:37] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[00:52:31] -!- rob_h has quit [Ping timeout: 255 seconds]
[00:57:01] -!- i_tarzan_ has quit [Ping timeout: 255 seconds]
[00:58:48] -!- alpha1125 has quit [Quit: Computer has gone to sleep.]
[01:00:02] -!- Nick001-Shop has quit [Remote host closed the connection]
[01:18:28] -!- ve7it has quit [Remote host closed the connection]
[01:19:16] -!- FinboySlick has quit [Quit: Leaving.]
[01:28:20] -!- asdfasd has quit [Ping timeout: 248 seconds]
[02:03:34] -!- Tom_L has quit []
[02:08:27] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[02:30:12] -!- asdfasd has quit [Ping timeout: 244 seconds]
[02:51:42] -!- i_tarzan has quit [Ping timeout: 264 seconds]
[02:52:30] -!- zzolo has quit [Quit: zzolo]
[03:37:47] zz_satyag is now known as satyag
[03:53:09] -!- mellon has quit [Ping timeout: 248 seconds]
[03:53:48] mellon_ is now known as mellon
[03:56:23] satyag is now known as zz_satyag
[03:56:27] -!- Keknom has quit [Quit: Leaving.]
[03:57:24] zz_satyag is now known as satyag
[03:59:31] -!- frysteev has quit [Ping timeout: 260 seconds]
[04:02:35] -!- alex_joni has quit [Read error: Operation timed out]
[04:03:24] -!- zzolo has quit [Quit: zzolo]
[04:08:36] -!- theos has quit [Disconnected by services]
[04:09:04] -!- i_tarzan has quit [Ping timeout: 260 seconds]
[04:10:02] -!- alex_joni [alex_joni!~alex_joni@emc/board-of-directors/alexjoni] has joined #linuxcnc-devel
[04:10:03] -!- mode/#linuxcnc-devel [+v alex_joni] by ChanServ
[04:13:32] -!- asdfasd has quit [Ping timeout: 244 seconds]
[04:15:24] satyag is now known as zz_satyag
[04:21:02] zz_satyag is now known as satyag
[04:21:24] -!- capricorn_1 has quit [Ping timeout: 276 seconds]
[04:21:26] -!- Loetmichel has quit [Ping timeout: 252 seconds]
[04:35:28] -!- phantoneD has quit [Ping timeout: 246 seconds]
[04:56:08] satyag is now known as zz_satyag
[04:56:49] <KGB-linuxcnc> 03seb 05v2.5_branch aa8618d 06emc2 10docs/man/man9/motion.9 * docs: list motion's tooloffset HAL pins in manpage
[05:11:30] -!- mellon has quit [Ping timeout: 264 seconds]
[05:11:34] -!- the_wench has quit [Ping timeout: 240 seconds]
[05:11:38] mellon_ is now known as mellon
[05:11:48] -!- archivist has quit [Ping timeout: 248 seconds]
[05:49:58] zz_satyag is now known as satyag
[05:50:58] -!- c60 has quit [Ping timeout: 255 seconds]
[05:51:36] -!- tayy has quit [Remote host closed the connection]
[05:58:12] <KGB-linuxcnc> 03TODO: deletor 052.5-package-replacement be1c433 06emc2 04. * branch deleted
[06:04:19] <KGB-linuxcnc> 03TODO: deletor 05hm2-watchdog 4c4b116 06emc2 04. * branch deleted
[06:08:04] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 2b64f2e 06emc2 10src/emc/rs274ngc/interp_find.cc * comment Interp::find_tool_pocket()
[06:08:05] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 fbaa6e3 06emc2 10src/emc/rs274ngc/interp_convert.cc * update comments for Interp::convert_tool_select (M6)
[06:08:07] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 d71aaee 06emc2 10src/emc/iotask/ioControl.cc * io: fix a misleading comment
[06:08:13] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 9de6d23 06emc2 10src/emc/nml_intf/canon.hh * rename the argument to SELECT_POCKET() to be more clear
[06:08:20] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 5dd0617 06emc2 10tests/ 10(15 files in 4 dirs) * add sim tests for the "toolno != pocket number" bug
[06:08:27] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 65fd665 06emc2 10src/emc/rs274ngc/interp_convert.cc * fix a bug in convert_setup_tool(), aka G10
[06:08:33] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 ad32937 06emc2 10src/emc/rs274ngc/interp_convert.cc * remove silly code in Interp::convert_setup_tool()
[06:08:40] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 46da1e7 06emc2 10src/emc/rs274ngc/interp_convert.cc * Interp::convert_setup_tool(): avoid a spurious tool copy
[06:08:47] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 8a4aff3 06emc2 10src/emc/task/emccanon.cc * fix a bug in emccanon's GET_EXTERNAL_TOOL_SLOT()
[06:08:54] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 9ecd9b3 06emc2 10src/emc/task/emccanon.cc * comment emccanon's GET_EXTERNAL_SELECTED_TOOL_SLOT()
[06:09:01] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 e9de409 06emc2 10src/emc/rs274ngc/interp_convert.cc * rename a variable in Interp::convert_tool_length_offset()
[06:09:07] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 c78ed43 06emc2 10src/emc/rs274ngc/interp_convert.cc * better comments in Interp::convert_tool_length_offset()
[06:09:14] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 2d4387c 06emc2 10src/emc/rs274ngc/interp_convert.cc * rename a variable in Interp::convert_cutter_compensation_on() for clarity
[06:09:21] <KGB-linuxcnc> 03seb 05g10-toolno-pocket-differ-2 b98f417 06emc2 10docs/src/code/Code_Notes.txt * docs: add Code Notes for tool changer
[06:09:41] <seb_kuzminsky> alright people
[06:10:24] <seb_kuzminsky> in that branch, check out commit 5dd0617 to see the test that triggers the bug
[06:10:31] <seb_kuzminsky> then check out the tip of the branch for the fix
[06:11:14] <seb_kuzminsky> some of the commits are fairly harmless cleanups/comments/documentation, the two actual bug fixes are 8a4aff3 and 65fd665
[06:11:55] <seb_kuzminsky> based on cradek's feedback on my last try i added a test with a random tc, and it worked, just like he said it would
[06:12:00] <seb_kuzminsky> this branch does not break it ;-)
[06:14:08] <seb_kuzminsky> if this looks good to everyone i'd like to push it to 2.5 and get some runtime in the field
[06:14:11] -!- johngag has quit [Client Quit]
[06:28:43] -!- vladimirek [vladimirek!~vladimire@95.105.250.72] has joined #linuxcnc-devel
[06:31:23] satyag is now known as zz_satyag
[06:45:00] -!- automata_ has quit [Ping timeout: 255 seconds]
[06:45:00] -!- automata has quit [Ping timeout: 255 seconds]
[06:51:48] <KGB-linuxcnc> 03chrisinnanaimo 05master 260c5b1 06emc2 10src/emc/usr_intf/gscreen/gscreen.py * gscreen -fix message dialogs
[06:51:49] <KGB-linuxcnc> 03chrisinnanaimo 05master 2a2d004 06emc2 10configs/sim/gscreen/gscreen.ini 10src/emc/usr_intf/gscreen/gscreen.py * gscreen -minor message dialog adjustments
[07:05:20] -!- tjb1 has quit [Quit: tjb1]
[07:05:32] -!- gag has quit [Quit: Page closed]
[07:18:27] -!- yuvipanda has quit [Quit: yuvipanda]
[07:42:25] -!- erictheise has quit [Quit: erictheise]
[07:47:06] zz_satyag is now known as satyag
[07:47:49] -!- kmiyashiro has quit [Quit: kmiyashiro]
[07:48:47] -!- morfic- has quit [Read error: Operation timed out]
[07:53:41] -!- yuvipanda has quit [Quit: yuvipanda]
[07:59:27] satyag is now known as zz_satyag
[08:04:12] -!- WalterN has quit [Ping timeout: 256 seconds]
[08:25:16] -!- neverho0d has quit [Ping timeout: 246 seconds]
[08:33:06] -!- neverho0d has quit [Ping timeout: 256 seconds]
[08:38:08] -!- psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[08:40:35] Tugge_ is now known as Tugge
[08:41:10] -!- neverho01 has quit [Ping timeout: 244 seconds]
[08:43:52] -!- maximilian_h [maximilian_h!~bonsai@84.157.231.39] has joined #linuxcnc-devel
[08:43:52] -!- maximilian_h has quit [Client Quit]
[08:45:07] -!- neverho0d has quit [Ping timeout: 260 seconds]
[08:49:59] -!- neverho01 has quit [Ping timeout: 256 seconds]
[08:58:04] -!- b_b has quit [Changing host]
[09:16:57] -!- racycle has quit [Quit: racycle]
[09:19:22] -!- capricorn_1 has quit [Read error: Connection reset by peer]
[09:19:31] -!- toner has quit [Read error: Connection reset by peer]
[09:24:24] -!- dhoovie has quit [Read error: Connection reset by peer]
[09:33:25] -!- rob_h [rob_h!~rob_h@027c009b.bb.sky.com] has joined #linuxcnc-devel
[09:49:01] -!- lufu has quit [Remote host closed the connection]
[09:49:55] -!- emel has quit [Excess Flood]
[10:00:07] -!- lufu has quit [Remote host closed the connection]
[10:17:38] -!- Cylly has quit []
[10:22:24] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[10:23:14] -!- lufu has quit [Remote host closed the connection]
[10:37:31] -!- H264 has quit [Read error: Connection reset by peer]
[12:03:35] <KGB-linuxcnc> 03jthornton 05v2.5_branch ef91afc 06emc2 10src/hal/components/time.comp * Docs: add info to description of component
[12:09:42] -!- skunkworks2 has quit [Read error: Connection reset by peer]
[12:41:27] -!- dhoovie has quit [Read error: Connection reset by peer]
[12:46:39] -!- skunkworks has quit [Remote host closed the connection]
[12:47:47] -!- fatpandas has quit [Ping timeout: 260 seconds]
[12:52:37] -!- adb [adb!~IonMoldom@178-211-235-11.dhcp.voenergies.net] has joined #linuxcnc-devel
[12:59:58] -!- mk0 has quit [Read error: Connection reset by peer]
[13:33:06] zz_satyag is now known as satyag
[13:37:07] -!- morfic- has quit [Ping timeout: 240 seconds]
[13:47:24] -!- mhaberler [mhaberler!~mhaberler@macbook.stiwoll.mah.priv.at] has joined #linuxcnc-devel
[13:47:36] -!- davec_ has quit [Ping timeout: 256 seconds]
[13:48:46] -!- mhaberler_ has quit [Ping timeout: 245 seconds]
[13:53:57] -!- gmouer has quit []
[13:59:31] satyag is now known as zz_satyag
[14:02:48] -!- psha[work] has quit [Quit: Lost terminal]
[14:06:05] -!- ybon has quit [Quit: WeeChat 0.3.8]
[14:10:02] -!- holst has quit [Quit: Leaving]
[14:18:24] -!- mozmck has quit [Quit: Leaving.]
[14:19:12] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[14:20:14] -!- mozmck has quit [Client Quit]
[14:20:49] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[14:24:29] zz_satyag is now known as satyag
[14:31:45] -!- Nick001 has quit [Ping timeout: 276 seconds]
[14:34:13] -!- alani has quit [Ping timeout: 245 seconds]
[14:36:08] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[14:37:12] satyag is now known as zz_satyag
[14:53:19] <skunkworks> hey - how is it going?
[15:04:15] -!- bmwyss has quit [Quit: bmwyss]
[15:05:11] <seb_kuzminsky> hey skunkworks
[15:15:23] -!- syyl has quit [Ping timeout: 244 seconds]
[15:27:23] <pcw_home> seb_kuzminsky: If there's anything I can do to help update the hostmot2 firmware on the buildbot, let me know
[15:30:07] -!- yuvipanda has quit [Quit: yuvipanda]
[15:48:17] -!- maximilian_h [maximilian_h!~bonsai@stgt-4d03ba8f.pool.mediaWays.net] has joined #linuxcnc-devel
[15:56:55] -!- maximilian_h [maximilian_h!~bonsai@stgt-4d03ba8f.pool.mediaWays.net] has parted #linuxcnc-devel
[16:10:17] -!- Simooon has quit [Remote host closed the connection]
[16:27:10] -!- automata_ has quit [Ping timeout: 250 seconds]
[16:27:10] -!- automata has quit [Ping timeout: 250 seconds]
[16:28:16] zz_satyag is now known as satyag
[16:31:56] -!- JT-Shop-2 [JT-Shop-2!~John@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[16:31:56] -!- jthornton_ [jthornton_!~john@184-63-140-99.cust.wildblue.net] has joined #linuxcnc-devel
[16:31:57] -!- JT-Shop has quit [Read error: Connection reset by peer]
[16:31:57] -!- jthornton has quit [Read error: Connection reset by peer]
[16:36:10] JT-Shop-2 is now known as JT-Shop
[16:36:42] <skunkworks> logger[psha]_,
[16:36:45] satyag is now known as zz_satyag
[17:02:06] <skunkworks> cannot believe I still had that patch...
[17:07:00] <seb_kuzminsky> skunkworks: i don't know how rotary axes work, but would "wrapped rotary" be useful for that guy?
[17:07:16] <seb_kuzminsky> it would limit the amount of "silly travel" to 1 turn, right?
[17:07:35] <skunkworks> I don't know - But I still think it would move to whatever 0 degree's was
[17:07:38] -!- mhel has quit [Client Quit]
[17:07:48] <seb_kuzminsky> yeah... :-/
[17:07:48] <skunkworks> right
[17:08:36] <skunkworks> Hi seb
[17:08:43] <seb_kuzminsky> hi :-)
[17:08:55] <cradek> I thought if you just put three numbers in TOOL_CHANGE_POSITION it'd only move XYZ, but I didn't, uh, test it or look at the code
[17:09:12] <seb_kuzminsky> testing is for people who dont have confidence in their code
[17:09:20] <cradek> or memory
[17:09:40] <skunkworks> heh - I think that is why jepler gave me that patch for my situation
[17:10:23] <skunkworks> I think there is still an issue where the locked rotory will not work during a tool change. (doesn't unlock and ferrors) but it has been a while since I have looked at it.
[17:10:31] <cradek> no, my fevered imagination invented that feature
[17:11:02] <seb_kuzminsky> cradek: when/if you get a chance, i'd appreciate another code review of the new g10 branch i just pushed
[17:11:04] <skunkworks> but I should test and put it up on the bug tracker
[17:11:59] <seb_kuzminsky> it now features random-tc testing, like you suggested last time
[17:13:13] <cradek> seb_kuzminsky: can you help pcw get set up with git so we get firmware packages built? who cares if it's buildbot or manual invocation, I just think we need some peridiocially blessed packages in our apt repo so people get updates. the way people all have different unversioned firmwares is madness and will lead to suffering.
[17:13:48] <cradek> origin/g10-toolno-pocket-differ-2 ?
[17:13:58] <seb_kuzminsky> cradek: i'm talking to him in another window, we're making plans for that
[17:14:01] <seb_kuzminsky> yes that's the branch
[17:14:06] <cradek> yay!
[17:14:09] <seb_kuzminsky> and i agree, it's total madness
[17:14:18] <cradek> feel free to talk here if I can be any help
[17:14:28] <seb_kuzminsky> ok thanks
[17:14:45] <seb_kuzminsky> what i really want is "git clone seb" so i can send one of me to work and one of me can hack on linuxcnc ;-)
[17:15:13] <cradek> sudo make me a sandwich
[17:16:53] <skunkworks> ot (yes cradek - all babies do look the same.. but mine is cuter..) ;)
http://electronicsam.com/gallery/picture.php?/839/category/18
[17:17:20] <seb_kuzminsky> happy-looking kid :-)
[17:17:28] <skunkworks> a percentage of the time...
[17:17:32] <skunkworks> :)
[17:17:40] <cradek> haha you guys look alike
[17:18:06] <skunkworks> that is what everyone says.. poor girl..
[17:18:19] <cradek> you should send it to reddit so someone can do that thing where they switch the heads.
[17:18:29] <skunkworks> heh
[17:18:44] <cradek> or is it fark? one of those cesspools.
[17:19:33] <seb_kuzminsky> i spent a good evening on the 'machine porn' reddit, which is not what you think
[17:25:15] <cradek> seb_kuzminsky: I'm only on your first change so far :-/ the new comment isn't very precise or correct. I think when you say "return" you mean assigns through the int*pocket, but the last line is incorrect then, as in that case it actually errors/aborts
[17:26:59] <seb_kuzminsky> crap
[17:29:07] <cradek> and it doesn't clarify the special case !random && toolno==0
[17:29:42] <seb_kuzminsky> i don't understand that special case :-/
[17:30:14] <seb_kuzminsky> i *think* toolno==0 means "no tool" in this case, but returning OK with pocket==0 is just wrong
[17:30:17] <cradek> I guess I recommend just removing the comment, since it makes total comprehension worse
[17:30:34] <seb_kuzminsky> pocket 0 may well contain a tool
[17:30:38] <seb_kuzminsky> hm
[17:30:52] <seb_kuzminsky> ok i can drop that commit
[17:30:56] jthornton_ is now known as jthornton
[17:31:18] <cradek> in !random, I think tool 0 means no tool. T0 M6 should cause an empty spindle
[17:31:39] <seb_kuzminsky> oh
[17:31:46] <seb_kuzminsky> where would it put the tool that's in the spindle?
[17:32:00] <cradek> I think in random, T0 could be a tool, but is more likely to be the empty pocket (in order to be able to get an empty spindle, you need a pocket that's empty and remains tracked)
[17:32:01] <seb_kuzminsky> i haven't done any T0 testing at all :-(
[17:32:23] <cradek> so T0 in random is pretty much just like every other tool
[17:32:57] -!- mackerski has quit [Quit: mackerski]
[17:33:57] <cradek> I think for !random, where the tool goes on unload isn't particularly well defined
[17:34:47] <seb_kuzminsky> i think it goes to settings->current_pocket
[17:35:25] <cradek> brb
[17:35:34] <seb_kuzminsky> see you
[17:37:44] -!- psha [psha!~psha@213.208.162.92] has joined #linuxcnc-devel
[18:22:02] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[18:27:05] -!- syyl_ws has quit [Quit: Verlassend]
[18:43:51] -!- paideia has quit [Ping timeout: 256 seconds]
[18:46:25] -!- psha has quit [Quit: Lost terminal]
[18:55:36] <cradek> seb_kuzminsky: about changing GET_EXTERNAL_TOOL_SLOT's behavior: the commit message says it was wrong before - can you say how you knew that?
[19:01:55] <seb_kuzminsky> the only user fo GET_EXTERNAL_TOOL_SLOT() is rs274ngc_pre.cc:synch(), which assigns the return value to settings->current_pocket
[19:02:17] <seb_kuzminsky> current_pocket is consistently used as if thought i t has a pocket number, not a tool number
[19:02:24] <cradek> about code notes: IMO nonrandom is not as specific as you say. it may be like sam's machine where the tool is found in an implementation-specific way (barcodes in this case)
[19:02:57] <cradek> thanks
[19:03:19] <seb_kuzminsky> hm, can you expand on that? (here's the built docs btw:
http://buildbot.linuxcnc.org/doc/scratch/v2.5.1~g10-toolno-pocket-differ-2~b98f417/html/code/Code_Notes.html#_the_tool_table_and_tool_changer)
[19:03:49] <cradek> > nonrandom tool changers put the current tool back in the pocket it came from before...
[19:04:10] <seb_kuzminsky> did i say something else?
[19:04:10] <cradek> this in particular. nothing in the implementation requires or evern particularly expects this does it?
[19:04:15] <cradek> even
[19:05:18] <seb_kuzminsky> i think i see what you mean
[19:05:25] <cradek> I guess it depends. If you use the pocket hal output to find your tool, you better put them back where you found them. but if you use the toolno, you can search for it any old way your machine knows how
[19:05:55] <seb_kuzminsky> with nonrandom, the pocket of the tool doesnt change when you do Txx or M6
[19:06:16] <cradek> oh ok, I see
[19:06:28] <seb_kuzminsky> instead, a copy of the tool info is made in pocket 0 when you do m6, and the original tool is left behind in its old pocket
[19:06:58] <seb_kuzminsky> this is in contrast to random, where the tool info is *moved* to pocket 0 in the tool table, and the original pocket is over-written with "no tool"
[19:07:29] <cradek> I think my complaint was that you are free to ignore that pocket information in your implementation -- but I see now that's orthogonal to what the code notes say (document how the tool table gets changed or not)
[19:07:51] <seb_kuzminsky> i think you're saying that the tc logic in hal can ignore the pocket number, and still work
[19:08:08] <seb_kuzminsky> the pocket number then becomes an internal implementation detail, and the user doesnt need to care about it
[19:08:22] <seb_kuzminsky> and i think i agree with that
[19:08:28] <cradek> suggest something like "with nonrandom, the pocket number in the tool table is not changed as tools are loaded and unloaded"
[19:08:59] <cradek> then you can avoid talking about implementation of a particular tool changer
[19:08:59] <seb_kuzminsky> i see what you mean, that's a good suggestion, thanks
[19:09:31] <seb_kuzminsky> yeah my notes probably conflate the internal logic in linuxcnc with my (incomplete/incorrect) notion of tool-changing hardware
[19:09:50] <cradek> I agree and I think that summarizes my minor objection
[19:09:52] <seb_kuzminsky> i'll change it to focus on the internal logic, thanks
[19:10:02] <cradek> I will continue to look through it
[19:10:40] <seb_kuzminsky> thanks... appreciate the extra eyeballs
[19:10:50] <seb_kuzminsky> i really don't want to break 2.5 this late in the game
[19:11:03] <cradek> you aren't the only one who feels that way
[19:11:14] <seb_kuzminsky> or pollute the repo with incorrect comments & docs ;-)
[19:11:55] -!- zzolo has quit [Quit: zzolo]
[19:12:08] <cradek> I like how asciidoc source/diffs are utterly readable
[19:12:20] <seb_kuzminsky> yeah, <3 asciidoc for that
[19:14:45] H264 is now known as WalterN
[19:15:30] <cradek> more about code notes: is there really a limit of 99999 for tool number?
[19:17:12] <seb_kuzminsky> i'm not sure, but it's mentioned in a couple of places in the existing gcode docs
[19:18:33] <cradek> I don't see a relevant test in the code
[19:19:32] <cradek> nah, I can load tool 9999999 just fine
[19:19:45] <seb_kuzminsky> yeah i believe that
[19:19:54] <seb_kuzminsky> the T docs are even more wrong:
http://www.linuxcnc.org/docs/2.5/html/gcode/other-code.html#sec:T-Select-Tool
[19:20:34] -!- LeelooMinai has quit [Ping timeout: 240 seconds]
[19:20:36] <cradek> I agree
[19:21:41] <cradek> before random, we mixed tool/pocket willynilly and that probably WAS an error
[19:21:59] <cradek> so it's not wrong, it's just outdated :-)
[19:22:52] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 16.0.2/20121025205401]]
[19:24:18] <seb_kuzminsky> i've made a note to update the code docs & the t-code blurb
[19:26:30] <cradek> you might want to also specify units for TLO, as you did for diameter
[19:27:08] <cradek> in the very bad old days the units were interpreted according to the g20/g21 mode in effect when the offset was loaded ... gggngnnrnghhh
[19:27:10] <seb_kuzminsky> yep
[19:27:36] <cradek> coordinate systems were done that way too (!!!)
[19:27:42] <seb_kuzminsky> is that the sound of one tooth gnashing?
[19:27:46] <cradek> yes
[19:28:07] <cradek> just thinking about it makes me cringe, and it's probably 10 years removed
[19:28:31] <cradek> "oh you should really never use your non-native units" was the resulting advice
[19:28:39] <seb_kuzminsky> haha
[19:29:02] <seb_kuzminsky> about half my programs are in metric, and my machine is imperial, and i *like* it that way :-)
[19:29:21] <cradek> you can thank the 10 years for that :-)
[19:30:16] <seb_kuzminsky> i guess things do improve with age
[19:30:17] <cradek> I think my review is done, but mostly I looked for gotchas and I'm just trusting your tests
[19:30:51] <seb_kuzminsky> in the first branch you reviewed for me i use sai for testing, but it didnt have any random-tc test
[19:30:55] <cradek> I have a bit of a suspicious feeling revolving around T0
[19:31:16] <seb_kuzminsky> when i tried to add a random-tc sai test i ran in to a different bug in saicanon that confused me for a while
[19:31:18] <cradek> you already found some of it (H0, D0)
[19:31:39] <seb_kuzminsky> so i rewrote the test to use emccanon instead, and ran in to a bunch of problems with linuxcncrsh :-(
[19:31:48] <cradek> yurgh
[19:32:09] <seb_kuzminsky> i've got a pile of notes on all the bugs i tripped over on my way to fixing G10 L10...
[19:32:38] <seb_kuzminsky> i've got a queasy feeling about T0 too, because it's used so inconsistently in the code and in the docs
[19:33:10] <seb_kuzminsky> do you think on random, T0 is just another valid tool (by convention used to track the empty pocket)
[19:33:14] <seb_kuzminsky> and on nonrandom it's invalid?
[19:33:20] <cradek> am I correct that 2.5 currently works ok for random AND for nonrandom where T=P for each tool in the tool table?
[19:33:44] <cradek> yes that's exactly my expectation for T0 on random
[19:33:57] <cradek> which possibly makes H0, D0 smell weird
[19:34:42] <seb_kuzminsky> i think that statement ("if T==P, random & nonrandom work in 2.5") is correct, but i dont have a test that shows it
[19:35:17] <cradek> if so, it strikes me that one thing we might choose to do in 2.5 is nothing
[19:36:04] <cradek> although I'm currently leaning toward fixing it in 2.5
[19:38:21] <seb_kuzminsky> i'd rather fix it, if we can convince ourselves taht we're not breaking something else
[19:41:05] <seb_kuzminsky> i dislike it that T0 is sometimes valid and sometimes invalid, depending on ini file settings
[19:41:25] <seb_kuzminsky> i dont think we should change this is 2.5 thought, probably
[19:43:22] <seb_kuzminsky> ugh, that also means that there's no way to ask for G43 H0 and actually get tool 0, you get whatever tool you currently have loaded instead
[19:43:49] <seb_kuzminsky> i dont know why you *would* ask for G43 for a tool you dont have loaded... but the current behavior seems very confusing
[19:44:00] -!- mozmck has quit [Quit: Leaving.]
[19:45:26] <seb_kuzminsky> so depending on the state of the machine and the ini files, T0 can be invalid, or mean "current tool", or "the tool whose number is 0"
[19:45:27] -!- micges has quit [Quit: Wychodzi]
[19:47:05] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc-devel
[19:49:15] <cradek> we could add an error if tool 0 has nonzero offsets
[19:49:46] <cradek> I bet the feature of asking for a different tool's entry for G41/42/43 is widely used
[19:50:45] <cradek> many tools have more than one useful cutting surface you might want to measure to
[19:50:57] <cradek> and tweaking diameter is an easy way to get roughing/finishing
[19:52:32] <seb_kuzminsky> but wouldn't you do that with g41.1/g42.1/g43.1 instead of changing the tool table? i guess sometimes its simpler to run the program twice while monkeying with the tool table rather than program in explicit passes with explicit TLOs
[19:52:52] <cradek> you could use those, yes
[19:53:52] <cradek> but imagine you've got a slitting saw mounted. you might want a tlo for the top and a tlo for the bottom, and you'd pick depending on which half of your cut you were hoping to keep
[19:54:05] <seb_kuzminsky> g41.1 x[#5401 + 0.020] ; roughing pass
[19:54:15] <cradek> a ball end mill might have a useful offset at the tip and another at the center of the ball
[19:54:17] <seb_kuzminsky> right, makes sense, "kerf compensation
[19:55:33] <cradek> a more traditional user will find it more natural to use two tool table entries than to use g41.1; I understand that's the way people have done it forever
[19:56:08] <andypugh> How would folk feel about a magic number for "don't tool-home this axis"?
[19:56:14] <seb_kuzminsky> ok, so i've got three items to fix in the code notes (description of nonrandom behavior and generally unconfuse code vs tc hardware, remove bogus tool number limit, tlo units for tlo), and add some t0 testing
[19:56:39] <cradek> andypugh: what is tool-home?
[19:56:41] <andypugh> (alternatively, initool should populate a variable with a bitmask of axes to be homed...)
[19:57:15] <cradek> oh you mean TOOL_CHANGE_LOCATION
[19:57:17] <seb_kuzminsky> i think andy means "dont move this axis to the tool-change position during a tool change"
[19:57:21] <andypugh> TOOL_CHANGE_POSITIO
[19:57:41] <andypugh> Yes, that's the ones.
[19:57:44] <cradek> I suggest using remap instead of TOOL_CHANGE_POSITION if you need anything nontrivial. solved problem (in master).
[19:58:28] <andypugh> That's probably a solution for the guy who asked.
[19:58:37] <cradek> in 2.5 some people may be able to get by with TOOL_CHANGE_QUILL_UP (sp) instead of TOOL_CHANGE_POSITION
[19:58:48] <andypugh> But it still seems wrong to zero all angular axes if not told to do so.
[20:00:47] <cradek> I don't want a new feature like that in 2.5, and I mostly think it's silly to do in master
[20:01:13] <cradek> your simple remedy is to just not use TOOL_CHANGE_POSITION
[20:01:44] <cradek> (or use sam's patch!)
[20:01:52] <seb_kuzminsky> cradek: you mean implement the part of TOOL_CHANGE_POSITION that you want in a remapped M6?
[20:02:18] <andypugh> remapping is not something I would describe as "trivial"
[20:02:19] <cradek> yes, do whatever series of moves you need, probably using g53
[20:02:48] <cradek> andypugh: actually remapping just M6 is not too bad
[20:03:09] <andypugh> So, remove TOOL_CHANGE_POSITION from Master? It duplicates remappable behaviour, and doesn't do what it says.
[20:03:18] <seb_kuzminsky> i've run into problems with g53 not doing what i want, when a previous program left the machine in metric units and my G53 G0 was giving instructions in imperial
[20:03:28] <cradek> you can call the old M6 inside it. so your remap sub would be g0 g53 .....; m6; ...
[20:03:56] <andypugh> seb_kuzminsky: All the more wierd as internal units appear to be metric.
[20:04:15] <seb_kuzminsky> my machine's user units are imperial
[20:04:29] <cradek> seb_kuzminsky: need more details...
[20:04:33] <cradek> andypugh: not sure if serious
[20:04:35] <andypugh> Ah, but it was G53-ing in metric?
[20:05:00] <andypugh> cradek: Which part?
[20:05:08] -!- mozmck [mozmck!~moses@client-204.235.45.161.wcfltx.partnershipbroadband.com] has joined #linuxcnc-devel
[20:05:31] <cradek> andypugh: I don't see a reason to remove it -- and I don't see why you think it doesn't do what it says
[20:05:39] <seb_kuzminsky> i have a sub that does "G53 G0 X9 Y12", to center the table and move it out towards the user (x9 is the center of x travel, y12 is the max y)
[20:05:47] <andypugh> I am semi-serious about TOOL_CHANGE_POSITION not acting as expected, as it moves axes it isn't told to move.
[20:06:20] <seb_kuzminsky> when i run that sub after switching interp to metric, it goes 9 mm & 12 mm instead of 9 & 12 inches
[20:06:34] <seb_kuzminsky> i realize it's my own fool fault
[20:06:42] <cradek> seb_kuzminsky: is that different from what you expect? ... oh
[20:06:55] <seb_kuzminsky> andypugh: you're right, the behavior you're describing is at odds with the docs
[20:06:55] <cradek> seb_kuzminsky: so put a g20 on that line
[20:06:58] <seb_kuzminsky> http://www.linuxcnc.org/docs/devel/html/config/ini_config.html#_emcio_section_a_id_sub_emcio_section_a
[20:07:37] <cradek> seb_kuzminsky: hmm, that documented behavior is the behavior I remembered
[20:07:54] <seb_kuzminsky> it's not perfectly clear, but it suggests (to me) that if you supply 3 numbers it will only move x y and z
[20:07:55] <cradek> (although it's kind of silly)
[20:08:01] <cradek> yes I agree that's what it says
[20:08:09] <cradek> I don't know if we used to do that
[20:08:38] <seb_kuzminsky> i dont know what the current behavior is, and i bet we don't have a test for it ;-)
[20:10:40] <andypugh> The sub should be able to store the imperial/metric mode and reinstate it at the end? #<_metric> exists in Master.
[20:12:23] <andypugh> As far as I can see readToolChange counts arguments and carefully sets the non-included ones to zero.
[20:12:24] <cradek> yes, the remap sub can save and restore all the modal stuff
[20:12:42] <seb_kuzminsky> i should totally do that
[20:12:57] <seb_kuzminsky> oh, i'm still on 2.5 in the garage
[20:12:59] <cradek> no you should run 2.5 and use it every day
[20:13:15] <andypugh> As far as I can see the only data passed back out of that routine is the offset, not how many axes were involved in the move,
[20:14:24] <andypugh> (And it errors if there are not 3, 6 or 9 arguments.)
[20:15:19] <cradek> http://www.linuxcnc.org/docs/devel/html/remap/structure.html#_making_minimal_changes_to_the_builtin_codes_including_tt_m6_tt
[20:15:47] <seb_kuzminsky> andypugh: my machine has a knee, which i've considered motorizing as the W axis, so i'd want to supply X Y Z W as a tool change position, i think
[20:18:11] <cradek> I can't find the mode save/restore in the docs, although I know it's there
[20:18:13] <andypugh> Can someone more versed in C++ have a look at initool.cc / readToolChange and tell me what happens to the values put into have_tool_change_position and have_tool_holder_clear ?
[20:18:19] <andypugh> Ah, found them.
[20:18:43] <andypugh> emcglbl.h defines system globals? Userland-only I assume?
[20:19:07] <seb_kuzminsky> i gotta pay attention to real life/work, talk to you guys later
[20:19:22] <seb_kuzminsky> thanks for the code review chris
[20:19:33] <cradek> welcome
[20:20:33] <andypugh> Hmm, we _could_ make have_tool_change_position be a bitmask of axes to move...
[20:21:22] <andypugh> But I was always told that global variables were evil :-)
[20:22:27] <andypugh> (and while we are there, TOOL_HOLDER_CLEAR is undocumented?)
[20:22:47] <cradek> that's historical and no longer used
[20:23:41] <andypugh> Seems to relate to the Rack toolchanger?
[20:24:29] -!- JT-Shop has quit [Read error: Connection reset by peer]
[20:24:36] <cradek> I think it was for some particular machine - it was before my time
[20:25:04] <cradek> pretty sure you'll find those values unused even if they are read
[20:25:19] <andypugh> Is remapping in 2.5.1?
[20:25:26] <cradek> no
[20:25:35] <seb_kuzminsky> i really like remap for things like that, because it moves stuff that's particular to a specific machine out of the core of linuxcnc and out to the config for that machine
[20:25:49] <cradek> yes
[20:25:59] <seb_kuzminsky> if it turns out to be generally useful we can consider moving some variety of it into the core
[20:26:05] <cradek> seb_kuzminsky: GET TO WORK SLACKER
[20:26:05] <seb_kuzminsky> oh wait, i'm supposed to be working
[20:26:07] <seb_kuzminsky> by
[20:26:09] <seb_kuzminsky> bye
[20:26:49] -!- JT-Shop [JT-Shop!~John@184.63.140.99] has joined #linuxcnc-devel
[20:27:31] <andypugh> The current readToolChange code seems simply strange to me.
[20:34:26] -!- erasmo [erasmo!~erasmo@77-254-232-156.adsl.inetia.pl] has joined #linuxcnc-devel
[20:41:32] -!- erasmo has quit [Remote host closed the connection]
[20:52:01] -!- vladimirek has quit [Remote host closed the connection]
[20:54:05] -!- Loetmichel has quit []
[21:02:17] -!- theorbtwo has quit [Remote host closed the connection]
[21:07:18] -!- automata has quit [Ping timeout: 264 seconds]
[21:07:18] -!- automata_ has quit [Ping timeout: 264 seconds]
[21:11:17] -!- rasher has quit [Changing host]
[21:26:38] -!- theos has quit [Disconnected by services]
[21:43:05] -!- jpk has quit [Ping timeout: 255 seconds]
[21:43:23] -!- JT-Shop-2 [JT-Shop-2!~John@184.63.140.99] has joined #linuxcnc-devel
[21:43:24] -!- jthornton has quit [Read error: Connection reset by peer]
[21:43:24] -!- JT-Shop has quit [Read error: Connection reset by peer]
[21:43:28] -!- jthornton_ [jthornton_!~john@184.63.140.99] has joined #linuxcnc-devel
[21:56:19] -!- maximilian_h [maximilian_h!~bonsai@f052123232.adsl.alicedsl.de] has joined #linuxcnc-devel
[21:56:31] -!- maximilian_h has quit [Client Quit]
[22:00:19] <andypugh> ROFL, so Our "Programmer" with the need for 50 parameters changed the code, but forgot to recompile?
[22:08:56] -!- FinboySlick has quit [Quit: Leaving.]
[22:10:02] -!- chillly has quit [Quit: Leaving]
[22:11:32] <cradek> ey, don't make me say "who hasn't done that?"!
[22:11:38] -!- DJ9DJ has quit [Quit: bye]
[22:14:13] -!- syyl_ has quit [Quit: Leaving]
[22:19:01] -!- wboykinm has quit [Ping timeout: 248 seconds]
[22:21:42] -!- pjm has quit [Read error: Connection reset by peer]
[22:30:20] JT-Shop-2 is now known as JT-Shop
[22:31:30] -!- skunkworks has quit [Read error: Connection reset by peer]
[22:38:11] -!- ch4o [ch4o!~chatzilla@c-68-52-244-25.hsd1.tn.comcast.net] has joined #linuxcnc-devel
[22:38:44] -!- WalterN has quit [Ping timeout: 248 seconds]
[23:09:34] -!- zzolo has quit [Quit: zzolo]
[23:12:00] <andypugh> I have, many times. But not after getting on my high horse about being a programmer. And generally in the middle of dozens of compiles.
[23:17:12] -!- alpha1125 has quit [Quit: Computer has gone to sleep.]
[23:23:35] -!- sumpfralle1 has quit [Read error: No route to host]
[23:25:14] -!- sumpfralle has quit [Ping timeout: 255 seconds]
[23:29:52] -!- sumpfralle2 has quit [Read error: Connection reset by peer]
[23:30:03] -!- sumpfralle has quit [Read error: No route to host]
[23:36:06] -!- sumpfralle1 has quit [Ping timeout: 240 seconds]
[23:41:34] -!- racycle has quit [Quit: racycle]
[23:54:36] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[23:58:27] -!- sumpfralle1 has quit [Read error: No route to host]