#emc-devel | Logs for 2010-07-22

[02:18:22] <ries_> ries_ is now known as ries
[02:33:27] <KimK> cradek: In the tool table, the (integer) tool numbers and pocket numbers (notwithstanding the current EMC2 limit of 55 or 56 or whatever), do they really go to 99,999 or do they stop at 65,535 ? (Q for the manual again, just making sure it's right.)
[02:33:48] <cradek> that depends on whether the tool changer is random
[02:34:14] <cradek> tool numbers do not have an upper limit (well maybe it's 4 million)
[02:34:21] <cradek> pocket numbers on NONrandom do not have a limit
[02:34:33] <cradek> pocket numbers on random have a limit of 55 or whatever it is
[02:36:40] <cradek> in both cases the total number of entries in the table is limited to that 55 or whatever it is
[02:36:47] <cradek> is that clear as mud?
[02:52:00] <KimK> No, very clear, and very helpful, thanks! I was off looking for a link I thought might amuse you. I was going to kid you and say, "Too bad we have a limit of 55 tools, John's brother has one of these across the street" (HMC with 70-tool carousel. No, he's not thinking of retrofitting it.) http://www.haascnc.com/details.asp?ID=EC-400PP&webid=HMC_EC&plain=1
[03:21:04] <cradek> neat. I bet stuart's enshu (sp?) also has >55 tools. simple matter of recompiling (I think)
[03:31:58] <KimK> OK, wait a minute, I thought I understood, but maybe not. In a randomizing TC, what changes in the tool table as tools exchange? Does (for example) T1 stay at the head of the list, but its following pocket number changes? Or do the pockets stay in order but everything else moves around? Or something else entirely?
[03:33:14] <cradek> in a random tc, as soon as the tool change is complete, emc rewrites the tool table
[03:33:38] <cradek> it is always sorted by pocket
[03:33:42] <cradek> pocket 0 is the spindle
[03:38:21] <KimK> OK, always in pocket order, thanks, that helps. And must the (carousel/chain/whatever) pockets start with 1? I mean, certainly I would, as a general rule, but I haven't seen every machine out there, and I'm trying to think of a case where you would want (for some historical reason?), say, 16 pockets running from 100 to 115? (No, I can't recall seeing a machine like that, but who knows?)
[03:39:21] <cradek> I think they have to be sequential
[03:40:07] <cradek> if you want to play with it (and you should if you're writing docs!) you can run the sim/random_tc config
[03:40:19] <cradek> looks like it uses the tool table simpockets.tbl
[03:41:03] <cradek> also - thanks for writing docs - that's so heroic
[03:42:27] <KimK> Sequential pocket numbers are probably OK. So, as a practical matter, the internal pocket limit of 99999 will be a great plenty it seems. Yes, maybe I will try fiddling with the random TC simulator.
[03:42:54] <cradek> but there is nothing special about 99999
[03:43:10] <cradek> on random, pockets go from 0 to 55 or whatever it is
[03:43:44] <cradek> on nonrandom, pocket numbers are whatever you want for your own bookkeeping - emc doesn't use them for anything, but it does remember them for you
[03:45:29] <KimK> Ha, I hope I don't have to write very much, lol! I hope to mostly edit, we'll see. But clearly I need to write a little more on the TC business, especially the random TC. I just want to clarify the unclear. And fix those typos.
[03:46:07] <cradek> excellent
[03:49:52] <KimK> I read the developers coding guide about committing one change at a time (in code), but I presume there would be no objection to (in docs) one big chunk of typo fixes? (I saved my abandoned patch list from before to find them again.) After this time there shouldn't be this many at once.
[03:51:04] <cradek> yeah I think that's fine - "Fixed all the typos" is one conceptual change
[03:52:11] <KimK> Ha, I wish I "fixed all the typos", lol. I'm pretty sure there's a bunch more waiting for me.
[03:53:22] <KimK> I should put "fixed 1.25 liters of typos". That'll confuse 'em.
[03:59:31] <cradek> goodnight
[04:01:24] <KimK> goodnight
[05:17:20] <ries> ries is now known as ries_away
[13:29:34] <ries_away> ries_away is now known as ries
[17:30:04] <skunkworks> SWPadnos: your email to the list is cryptic...
[17:30:13] <SWPadnos> and accidental
[17:30:21] <skunkworks> heh
[17:30:39] <SWPadnos> my phone did it
[21:42:38] <andypugh> Does anyone recall the cause of "the remote end hung up unexpectedly" when trying a git clone?
[21:42:54] <SWPadnos> no ssh keys
[21:43:33] <andypugh> OK.
[21:46:48] <jepler> you probably should be using a clone URL that starts with git://
[21:48:06] <andypugh> I am using git clone git://git.linuxcnc.org/git/emc2-git emc2.5-dev
[21:48:58] <andypugh> I am sure that git works on the other machine, and I don't recall setting up SSH keys on that
[21:49:27] <SWPadnos> I know I have had that error when I had forgotten to copy over my "standard" .ssh/* files
[21:49:32] <jepler> emc2.git, not emc2-git
[21:49:44] <SWPadnos> click. buzzzzzzzzzzzzzzz
[21:49:57] <andypugh> Aha!
[21:50:01] <andypugh> That works.
[21:50:35] <andypugh> Mark that down as User Error
[21:50:51] <andypugh> (Not the clearest of error messages, though)
[21:51:22] <jepler> indeed not