#linuxcnc-devel Logs

Oct 18 2018

07:21 AM jepler: If network during installation is *NOT* a requirement of the canonical debian live images then there's some chance it's a "simple" problem amenable to resolution.
07:23 AM jepler: but unfortunately the bug is upstream https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=866629
08:55 AM JT-Shop: andypugh: Loetmichel ran into a problem with his spindle and master this morning or day for him and you. his config is http://www.cyrom.org/palbum/main.php?g2_view=core.DownloadItem&g2_itemId=17217
09:02 AM JT-Shop: you can read about the problem he had on #linuxcnc
12:12 PM seb_kuzminsky: rene_dev_: does the bug447-fix-blend-tangent-fallback branch get rid of the problem described in #447?
12:16 PM rene_dev_: seb_kuzminsky reading the issue again, Im not sure. you want to try?
12:16 PM rene_dev_: now trying to make travis happy about the pr
12:19 PM andypugh: Am I right that the Buildbot hasn’t made anything since Sunday?
12:23 PM rene_dev_: seb_kuzminsky I wont merge it until Im certain that it will fix the problem. but I wanted to create a pr first, to make things easier...
12:25 PM rene_dev_: If it isnt fixed I will ask him, I will see rob next week...
12:25 PM hazzy-m: it seems the docs aren't being updated, is the buildbot down?
12:43 PM andypugh: Multispindle supports 8 spindles. I don’t know why I chose that number. Can anyone think of a more defensible number?
12:45 PM rmu: 10^500
02:54 PM Roguish_ is now known as Roguish
02:57 PM seb_kuzminsky: rene_dev_: i won't have time to re-test it any time soon, sorry
04:01 PM sync: andypugh: ULONG_MAX
04:59 PM andypugh: Interesting that the commit message automatically closed the issue
05:02 PM sync: github is trying to outsmart you
05:08 PM pcw_mesa: Thanks for fixing that Andy!
05:14 PM andypugh: pcw_mesa: It would be useful to check the fix, it seemed too easy :-)
05:15 PM andypugh: pcw_mesa: Did you see the 8C20 question on the forum? Does it actually exist?
05:17 PM pcw_mesa: At one time but we dropped it
05:18 PM pcw_mesa: I'll test the popup error now
05:18 PM andypugh: I have the impression that the 8i20 isn’t a hige seller, and the 8c20 was a small fraction of that?
05:24 PM pcw_mesa: I think maybe 200 or so 8I20s so not that many
05:24 PM andypugh: I have a significant proprtion then (5)
05:25 PM pcw_mesa: WD bite popup works (would have saved a long debugging thread on the forum)
05:25 PM andypugh: Indeed.
05:25 PM andypugh: I recognised the impetus for the bug report
05:25 PM pcw_mesa: NASA is using some now for a building printer
05:25 PM andypugh: Nice
05:26 PM andypugh: I have to admit that on any new conversion I would probably use STMBL.
05:26 PM pcw_mesa: Yeah
05:26 PM andypugh: Though there is no reason that Mesa couldn’t make and sell STMBL.
05:26 PM andypugh: (it even works with smart-serial)
05:28 PM pcw_mesa: I really need to extend sserial to 8 registers instead of 4 (and add separate read and write packets)
05:30 PM andypugh: I think that the driver is ready for that.
05:31 PM pcw_mesa: Busy with other stuff though (RPI /master has been up about 3 months now... seems quite reliable if you dont do huge graphic stuff and run out of memory)
05:32 PM pcw_mesa: The guy that did the SPI interface(Bruno?) did a good job
05:33 PM pcw_mesa: too bad the SPI driver stuff has to be redone for almost every host
05:35 PM andypugh: Yes, that does seem a shame.
05:48 PM Jin^eLD is now known as Jin|away
05:51 PM andypugh: How would one update dgarr/external_offsets to include the latest fixes from master?
05:51 PM andypugh: (and then push to the repository)
05:56 PM andypugh: I guess that first we need the commits in to 2.7, then 2.8
05:57 PM seb_kuzminsky: andypugh: the best way would be to convince dgarr to rebase his external_offsets on top of origin/master
05:57 PM andypugh: Or to do that for him?
05:57 PM andypugh: Actually, better still would be to merge EO into master..
05:57 PM seb_kuzminsky: well, you could fetch his repo, rebase his branch, push the result to your repo, and then convince him to drop his branch in favor of yours
06:04 PM seb_kuzminsky: or if you're planning to merge it, just merge it
06:10 PM andypugh: I already broke master enough with Multispindle :-)
06:11 PM seb_kuzminsky: but you're dilligently fixing the bugs so it's all good
06:13 PM pcw_mesa: when you have multiple spindles what determines which spindle is used for spindle synchronized motion?
06:20 PM andypugh: The $ command
06:20 PM andypugh: So G76 K…… $3
06:21 PM andypugh: (Surprsingly important for a counterspindle lathe, to be able to thread both ends of the work)
06:24 PM andypugh: G96 S100 $0 / G96 S200 $1 / M3 $0 / M4 $1 has two spindles turning opposite directions at different speeds, but both varying according to X position. Probably not very useful
06:26 PM andypugh: Hmm, external-offsets does apply cleanly and compile without errors on current master.
06:37 PM rene_dev_: https://imgur.com/a/YkiSmdD
06:49 PM andypugh: And runtests all pass with external-offsets on top of master.
06:49 PM andypugh: I think we should merge it
06:50 PM andypugh: rene_dev_: How’s the latency?
06:59 PM rene_dev_: andypugh https://imgur.com/a/1LN9OeS
07:22 PM andypugh: I’ve seen worse
07:25 PM andypugh: Actually, I haven’t. I mis-counted decimal places. 17mS latency?
07:52 PM hazzy-m: rene_dev_ (IRC): Whoa! That's a neat trick! Just for fun or what is the motivation?
07:54 PM rene_dev_: just for fun, my other laptop broke
07:54 PM rene_dev_: its really easy. just install wsl ubuntu, and compile. im using the cygwin x server
07:57 PM rene_dev_: so next time pepole complain that linux is too complicated, we can tell them that they can just run linuxcnc on windows
07:58 PM rene_dev_: it died, because I spilled the leftover cider from emf on the keyboard :D andypugh
07:59 PM andypugh: Is it super-dead, or in recovery ?
08:02 PM andypugh: Time to sleep