Back
[00:13:43] -!-
PCW has quit [Quit: ChatZilla 0.9.88.1 [Firefox 3.6.13/20101203075014]]
[00:18:55] -!-
rob_h has quit [Ping timeout: 260 seconds]
[00:21:17] -!-
Loetmichel has quit [Ping timeout: 245 seconds]
[00:39:37] -!-
Nick001-Shop has quit [Remote host closed the connection]
[00:42:31] -!-
skunkworks__ [skunkworks__!~chatzilla@184-158-63-58.dyn.centurytel.net] has joined #linuxcnc-devel
[00:42:58] <skunkworks__> logger[psha]:
[00:47:48] mybitcoin is now known as
ybit
[00:50:34] -!-
skunkworks__ has quit [Ping timeout: 245 seconds]
[01:20:09] -!-
stevegt_1 has quit [Ping timeout: 245 seconds]
[01:46:24] -!-
jdhNC has quit [Ping timeout: 245 seconds]
[02:12:45] -!-
andypugh has quit [Quit: andypugh]
[02:38:29] -!-
demacus has quit [Ping timeout: 245 seconds]
[02:46:10] -!-
mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc-devel
[02:46:41] -!-
GeorgeH has quit [Read error: Connection reset by peer]
[03:02:12] -!-
GeorgeH has quit [Read error: Connection reset by peer]
[03:02:15] -!-
stevegt_1 has quit [Ping timeout: 260 seconds]
[03:02:32] -!-
asdfasd has quit [Ping timeout: 245 seconds]
[03:21:52] -!-
mhaberler has quit [Quit: mhaberler]
[03:22:22] -!-
zlog has quit [Remote host closed the connection]
[03:22:25] -!-
Tom_itx has quit []
[04:55:24] -!-
psha[work] [psha[work]!~psha@195.135.238.205] has joined #linuxcnc-devel
[05:26:20] -!-
ThadiusB has quit [Ping timeout: 260 seconds]
[06:17:34] -!-
WalterN has quit [Read error: Connection reset by peer]
[06:57:58] e-ndy|afk is now known as
e-ndy
[07:25:03] -!-
Vq has quit [Ping timeout: 260 seconds]
[07:27:31] -!-
WalterN has quit [Read error: Connection reset by peer]
[08:27:34] -!-
Thetawaves has quit [Remote host closed the connection]
[08:28:14] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has joined #linuxcnc-devel
[08:28:22] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has parted #linuxcnc-devel
[08:41:12] -!-
mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc-devel
[09:03:29] -!-
mhaberler has quit [Quit: mhaberler]
[09:36:35] -!-
rob_h [rob_h!~rob_h@5ace70d6.bb.sky.com] has joined #linuxcnc-devel
[10:07:19] -!-
ysli [ysli!~ysli@125-227-124-249.dynamic.hinet.net] has joined #linuxcnc-devel
[10:07:46] <ysli> cradek: hi
[10:08:27] <ysli> I've created a feature request for merging s-curve to ja3 on the sourceforge tracker
[10:08:39] <ysli> https://6bc1d2f7-a-7f8637a5-s-sites.googlegroups.com/a/araisrobo.com/en/linuxcnc/g64-max-blending/2012-04-28_blending_vel.png?attachauth=ANoY7cpoxvj-CQy0MBjwSOMGAX5eSYVqjauimPjQsTFu4E-dh6oscJRiWIYPzy9meA30-EAFJD6G4qHIOzikPwbcA33M9Cpi0dLC6VZ5RXoCugu-Wjf1h8DrM64fQEocyfs2IZ9937sReoJG_FzNFj24vHeCG5b-zQMUsI40idYk7mO3SPtQOp19DxZ-BSDThAl6RlCBRcabir5xr0g3cHdq6Z3WJQCBU3mMi9myD1J7OJ9QjY8PxJnhj9JfDIJrT0tX-kgD34L7&attredirects=0
[10:09:11] <ysli> this is the screen shot for G64 velocity blending at max allowable velocity
[10:10:39] <ysli> the detail description is here:
https://sites.google.com/a/araisrobo.com/en/linuxcnc/g64-max-blending
[10:12:14] <ysli> the patch is here:
https://sourceforge.net/tracker/download.php?group_id=6744&atid=356744&file_id=442338&aid=3522196
[10:13:04] <ysli> it is tested for orthogonal vector blending only
[10:14:43] <ysli> and it requires the MAX vel, accel, jerk of AXIS_X and AXIS_Y to be identical
[10:27:31] -!-
ysli has quit [Quit: Leaving]
[11:26:31] -!-
elmo40 has quit [Read error: Connection reset by peer]
[11:30:51] -!-
asdfasd has quit [Ping timeout: 244 seconds]
[11:31:34] -!-
shdhdfghd has quit [Client Quit]
[11:34:34] -!-
mhaberler [mhaberler!~mhaberler@195.191.253.94] has joined #linuxcnc-devel
[12:06:23] -!-
elmo40 has quit [Read error: Connection reset by peer]
[12:44:24] -!-
skunkworks__ [skunkworks__!~chatzilla@str-bb-cable-south-3-102.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[12:45:46] -!-
i_tarzan_ has quit [Ping timeout: 276 seconds]
[12:48:40] -!-
andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust639.basl.cable.virginmedia.com] has joined #linuxcnc-devel
[12:51:27] -!-
psha[work] has quit [Quit: Lost terminal]
[13:12:35] -!-
the-jub has quit [Ping timeout: 246 seconds]
[13:19:37] -!-
syyl has quit [Ping timeout: 255 seconds]
[13:22:12] -!-
ysli [ysli!~ysli@125-227-124-249.dynamic.hinet.net] has joined #linuxcnc-devel
[13:38:04] -!-
demacus_ has quit [Ping timeout: 245 seconds]
[13:54:44] -!-
demacus_ has quit [Ping timeout: 245 seconds]
[14:01:20] -!-
psha [psha!~psha@188.65.232.26] has joined #linuxcnc-devel
[14:16:29] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has joined #linuxcnc-devel
[14:24:10] -!-
Nick001 has quit [Ping timeout: 260 seconds]
[14:52:18] -!-
Valen has quit [Quit: Leaving.]
[15:07:10] -!-
gambakufu has quit []
[15:19:19] cylly2 is now known as
Loetmichel
[15:20:58] -!-
maximilian_h [maximilian_h!~bonsai@130.255.104.157] has parted #linuxcnc-devel
[15:23:59] -!-
ysli has quit [Quit: Leaving]
[15:26:07] -!-
skunkworks__ has quit [Remote host closed the connection]
[15:38:30] -!-
vladimirek has quit [Remote host closed the connection]
[15:41:21] -!-
gambakufu has quit []
[15:59:29] -!-
mozmck has quit [Quit: Leaving.]
[15:59:46] -!-
skunkKandT [skunkKandT!~chatzilla@184-158-3-161.dyn.centurytel.net] has joined #linuxcnc-devel
[16:00:42] -!-
mozmck [mozmck!~moses@client-74.117.92.175.dfwtx.partnershipbroadband.com] has joined #linuxcnc-devel
[16:06:12] <skunkKandT> logger[psha]:
[16:07:48] -!-
skunkKandT has quit [Remote host closed the connection]
[16:11:10] -!-
skunkKandT [skunkKandT!~chatzilla@184-158-3-161.dyn.centurytel.net] has joined #linuxcnc-devel
[16:17:23] -!-
maximilian_h [maximilian_h!~bonsai@anon-180-79.vpn.ipredator.se] has joined #linuxcnc-devel
[16:35:52] -!-
maximilian_h [maximilian_h!~bonsai@anon-180-79.vpn.ipredator.se] has parted #linuxcnc-devel
[16:53:23] -!-
GeorgeH has quit [Client Quit]
[16:56:10] -!-
elmo40 has quit [Read error: Connection reset by peer]
[17:18:36] uwe__ is now known as
uwe_
[17:41:13] <skunkKandT> so - the s-curve acceleration patch only works in the x-y plane and the acc/vel for those axis needs to be the same? Also I kinda know why acc is 1/2 max in master (so you are sure to be able to stop at the end of the line segment) how is the s-curve patch getting around that? Is it actually calculating further ahead?
[17:42:52] <skunkKandT> is it a good start?
[17:45:23] <andypugh> I don't think that running half max acc does make sense. As you tune for what the machine can handle, you are basically tuning half-acc to be as high as possible. What it really means is that the number is wrong, if I am interpreting correctly.
[17:45:31] -!-
micges [micges!~micges@djl170.neoplus.adsl.tpnet.pl] has joined #linuxcnc-devel
[17:46:10] <andypugh> And I doubt that S-curve only works in XY, it was written for mutli-joint robots.
[17:46:50] <skunkKandT> I guess I could post those questions on the list. ;)
[17:47:50] <skunkKandT> the .5acc is been in emc2 for a long time.
[17:48:08] <skunkKandT> for blending...
[17:48:10] <andypugh> Actually..
http://youtu.be/EwfhjVV7D7Q is an XY SCARA robot.
[17:48:20] <skunkKandT> exact stop uses full accelleration
[17:49:05] <skunkKandT> that is neat
[17:52:14] <skunkKandT> heh - won't find the spotting drill if it isn't in the tool chain...
[17:57:43] <cradek> I replied with some of my thoughts
[17:59:00] <cradek> I hope he continues to work on it.
[18:00:26] <cradek> I spent some time on it and was unable to get the single move case to work correctly
[18:00:53] <cradek> I can't tell if it's a fundamental problem or just a tweak needed.
[18:01:21] -!-
micges has quit [Quit: Leaving]
[18:04:36] -!-
iwoj has quit [Quit: Textual IRC Client:
http://www.textualapp.com/]
[18:13:21] <skunkKandT> is it strictly x-y atm?
[18:17:04] <cradek> I don't see any restriction like that
[18:17:30] -!-
iwoj has quit [Ping timeout: 260 seconds]
[18:18:40] <skunkKandT> oh - cool
[18:19:24] <skunkKandT> Why does he only say x vel/acc/jerk needs to be the same on x/y?
[18:19:42] <cradek> I think he means on all axes
[18:20:02] <cradek> I think he's made a special type of blending that works in very specific situations, and that's one requirement
[18:20:43] <skunkKandT> ah
[18:20:45] <cradek> I don't know how the general case works yet
[18:25:55] <cradek> I worked on eliminating the half-acc behavior recently but it didn't work right
[18:26:19] <cradek> in the current planner I mean
[18:26:48] <cradek> in fact it tested so badly I was worried
[18:27:02] <cradek> but when I took out my changes everything worked perfectly again
[18:27:21] <skunkKandT> heh
[18:27:52] <cradek> everyone hates our current planner, but damn if it doesn't work perfectly
[18:27:53] <skunkKandT> was half-acc an 'easy' compromise to make things work well?
[18:28:11] <cradek> yes because the accels add on reversal type moves
[18:28:30] <cradek> so if your path happens to give you perfect adding you don't exceed it
[18:29:18] <cradek> jepler tried to put in a check for this accel adding, using dotproduct<0, and while that seems obviously correct, it is an inadequate test
[18:29:22] <cradek> (we took it back out)
[18:31:20] <cradek> wow, that was in '06: b5e1ba20
[18:31:21] <skunkKandT> I agree that it works reliably.. (no issues here) the only real complaint seems to be HSM
[18:31:52] <cradek> yes maintaining feed is its weak point - and yishin's work does not affect that as far as I know
[18:32:20] -!-
motioncontrol has quit [Quit: Sto andando via]
[18:32:23] <cradek> I'm not yet convinced that jerk limiting is worth the complexity, but I'm happy to wait and evaluate it if he gets it fully working
[18:32:29] <skunkKandT> I think if you could test as you tried before - (if I understand it right) it would only use .5acc in situations that it needs to.. I bet most everyone else would be happy.
[18:33:11] <cradek> yes I'd like to do that - but as I said, it didn't work out as easily as I thought it might
[18:33:26] <cradek> 'accel seems low in auto mode' is a common complaint
[18:33:32] <skunkKandT> it never does... ;)
[18:34:51] <cradek> jeff's change had a fundamental mistaken assumption (not acute is safe to add) and an implementation problem (blended poorly when adjacent moves had different accels)
[18:35:17] <cradek> both should be fixable
[18:35:23] <cradek> but 6 years later...
[18:36:32] <andypugh> I am messing about with a standalone HAL raster generator. I _was_ going to make that jerk-limited just for fun.
[18:37:32] <cradek> would that be running joints at certain fixed speeds? that'd be pretty easy I think
[18:39:00] <andypugh> Yes, that was my thinking. It seems totally pointless to generate G-code to perform a rectangular raster.
[18:39:11] <cradek> yep
[18:39:38] <andypugh> And the wiki laser rastering page describes such a complex kludge...
[18:39:47] <cradek> and if it doesn't have to be an exact size, even easier
[18:40:17] <cradek> you could hardcode it (jerk up a certain number of steps, then accel a certain number, then jerk down, then cruise ...)
[18:42:11] -!-
kent_ has quit [Ping timeout: 272 seconds]
[18:43:51] <skunkKandT> damn - probing is nice
[18:44:20] <andypugh> You want to be careful where you say that
[18:45:09] <skunkKandT> heh
[18:46:01] <andypugh> jepler: is option userinit in comp effective for realtime components? It seems to only be considered if option userspace is set.
[18:51:11] -!-
elmo40 has quit [Read error: Connection reset by peer]
[19:22:06] -!-
theorbtwo has quit [Remote host closed the connection]
[19:22:13] theorb is now known as
theorbtwo
[20:02:19] -!-
skunkKandT has quit [Remote host closed the connection]
[20:17:38] -!-
vladimirek has quit [Remote host closed the connection]
[20:31:54] <andypugh> I just realised, my Youtube channel has 55 subscribers. That's a little scary.
[20:32:59] <andypugh> Ah, slightly wrong tab wrong tab.
[20:40:47] MrZulu is now known as
Tecan
[20:42:54] -!-
Thetawaves has quit [Quit: This computer has gone to sleep]
[20:43:47] -!-
syyl_ws has quit [Remote host closed the connection]
[20:46:52] -!-
DJ9DJ has quit [Quit: bye]
[20:48:50] -!-
Tecan has quit [Quit: live long and phosphor]
[21:19:56] -!-
the-jub has quit [Quit: ill be back soon! :-/ ....maybe..]
[21:31:27] -!-
psha has quit [Quit: Lost terminal]
[21:53:31] -!-
skunkKandT [skunkKandT!~chatzilla@184-158-3-161.dyn.centurytel.net] has joined #linuxcnc-devel
[22:07:33] -!-
mhaberler has quit [Quit: mhaberler]
[22:14:14] e-ndy is now known as
e-ndy|afk
[22:15:14] <skunkKandT> logger[psha]:
[22:30:16] Tecan is now known as
RUBBERDUCK
[22:30:44] RUBBERDUCK is now known as
RUBBERN
[22:35:28] -!-
fliebel has quit [Remote host closed the connection]
[22:51:34] -!-
KimK has quit [Read error: Connection reset by peer]
[22:54:48] -!-
KimK [KimK!~Kim__@209.248.147.2.nw.nuvox.net] has joined #linuxcnc-devel
[22:58:52] RUBBERN is now known as
Nikola
[22:59:39] Nikola is now known as
Tecan
[23:00:44] -!-
bedah has quit [Quit: bye]
[23:03:50] the-jub is now known as
the-useless
[23:06:49] -!-
crib has quit [Ping timeout: 245 seconds]
[23:10:06] -!-
archivist has quit [Ping timeout: 272 seconds]
[23:18:52] -!-
archivist has quit [Ping timeout: 276 seconds]
[23:45:50] -!-
sumpfralle has quit [Quit: Leaving.]
[23:50:36] -!-
skunkKandT has quit [Remote host closed the connection]
