| 00:02.36 | *** join/#kicad Wetmelon (~wetmelon@host-73-169.ncasgro.asheville.nc.us.clients.pavlovmedia.com) |
| 00:32.59 | *** join/#kicad cyborg_ar (~cyborg_ar@c-98-215-226-11.hsd1.il.comcast.net) |
| 00:40.21 | *** part/#kicad ridesa7___ (~quassel@pool-74-110-107-3.nrflva.fios.verizon.net) |
| 00:40.38 | decimad | I wonder if a class can put its members into the padding area of member objects ^^ |
| 00:41.59 | c4757p | o_O |
| 00:42.05 | c4757p | /why/ |
| 00:42.10 | c4757p | that's horrifying |
| 00:42.19 | mroszko_ | o.o |
| 00:42.24 | mroszko_ | i need more alcohol |
| 00:42.26 | decimad | consider class foo { bool a; bar b; } |
| 00:42.28 | mroszko_ | wheres the iv |
| 00:42.41 | decimad | chances are hight that bar has space left for a bool somewhere |
| 00:43.02 | c4757p | dumps a box full of 8GB RAM sticks on decimad's head |
| 00:43.09 | decimad | thank you! |
| 00:43.48 | decimad | memory is cheap, but exceeding cache lines unnecessarily is not ;) |
| 00:44.33 | c4757p | caches are getting larger too >_> |
| 00:44.57 | decimad | worksets get larger even more so >_> |
| 00:45.39 | decimad | I like your ascii arts... can you post the no-idea one? |
| 00:45.55 | c4757p | ¯\_(ã)_/¯ |
| 00:46.03 | decimad | :) |
| 00:48.22 | *** join/#kicad kow_ (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 00:52.56 | cirilo | that sounds Too Clever - or in human-speak: it will break with every version of every compiler imaginable |
| 00:53.11 | decimad | okay, there exist abis that at least reuse the tail padding of subobjects |
| 00:53.12 | cirilo | + break with every different optimization setting |
| 00:54.18 | cirilo | covering all the generic cases = hideously bloated and slow code |
| 00:54.55 | cirilo | it's difficult for me to imagine any benefits - it's a Classec Dickism |
| 00:55.01 | cirilo | Classic even |
| 00:55.33 | decimad | well, if an object gets smaller this way I don't find it hard to imagine benefits |
| 01:06.55 | decimad | cirilo: a classic dickism would be if you tried to to that manually ;) |
| 01:07.28 | cirilo | object = smaller, code bloat = far more than memory saved |
| 01:07.40 | decimad | code bloat? |
| 01:08.12 | cirilo | yes - the compiler has to take Other Crap into account |
| 01:08.58 | *** join/#kicad tom_fig (602556db@gateway/web/freenode/ip.96.37.86.219) |
| 01:09.17 | decimad | as in compiler code bloat? |
| 01:09.31 | decimad | I'd guess everything that changes would be the offset of a member? |
| 01:09.33 | cirilo | I'd imagine normal code would be bloated as well |
| 01:10.02 | cirilo | at least if the members are public or friend visible |
| 01:10.38 | decimad | still only some offset constant changes? |
| 01:12.04 | decimad | but yeah, I see a point when that way you spread members to different cachelines which are used together |
| 01:13.06 | decimad | but for boost::optional f.e. I could imagine it would be effective |
| 01:13.07 | cirilo | + possible data leak problems |
| 01:16.53 | decimad | but a profile guided optimization compiler could make use of this |
| 01:17.42 | decimad | anyways, this was a quick idea, not the attempt to file a standards proposal ;) |
| 01:19.37 | *** join/#kicad crazy_im1 (~mj@a89-182-6-27.net-htp.de) |
| 01:21.22 | xzcvczx | Yatekii: i linked the github repo |
| 01:22.39 | xzcvczx | wooo finally, bzr and svn stuff is GONE |
| 01:30.47 | cirilo | yay! now I can nuke bzr on my computer |
| 01:31.09 | cirilo | though I should have done that anyway as soon as kicad moved to git |
| 01:31.10 | xzcvczx | yeah i need to remember to nuke it |
| 01:31.38 | xzcvczx | checkcoding.py is still all bzr styles |
| 01:31.47 | xzcvczx | there are a few other references in comments to bzr |
| 01:35.10 | *** join/#kicad tom_fig (602556db@gateway/web/freenode/ip.96.37.86.219) |
| 01:38.09 | tom_fig | It seems Dick has heard our wailing and gnashing of teeth w.r.t. python scripting |
| 01:38.22 | xzcvczx | huh? in what way |
| 01:40.00 | tom_fig | https://code.launchpad.net/~dickelbeck/kicad/+git/swig-enhancements/+ref/master |
| 01:40.10 | mroszko_ | hmm |
| 01:40.18 | mroszko_ | we should integrate coreclr! |
| 01:40.29 | mroszko_ | *waits for xzcvczx to object* |
| 01:41.56 | xzcvczx | mroszko_: well based on the commits messages i am almost scared to look at his code, in which case coreclr might be better :S |
| 01:42.23 | tom_fig | I did try to make something the other day. I tried building a package for the Windows Store with KiCad; yes, a perversion of nature. |
| 01:42.50 | xzcvczx | well with the rate its going i see ms going walled garden at some point |
| 01:43.10 | mroszko_ | coreclr even has posix coding style conventions |
| 01:43.13 | mroszko_ | for the api |
| 01:43.16 | mroszko_ | due to community request |
| 01:43.18 | mroszko_ | lol |
| 01:44.05 | tom_fig | I don't know. I'm kind of hoping the packaging will be like Chocolatey for the store eventually. |
| 01:45.22 | mroszko_ | eh |
| 01:45.27 | mroszko_ | you just run the Centium |
| 01:45.33 | mroszko_ | you dont have to do the uwp packaging |
| 01:45.35 | mroszko_ | it does it for you |
| 01:45.47 | mroszko_ | we would probably create a centrally controlled kicad account though |
| 01:45.51 | tom_fig | Ah, wasn't aware of that |
| 01:45.57 | xzcvczx | how much does it cost to distribute free apps through the windows store? |
| 01:46.03 | mroszko_ | free |
| 01:46.08 | xzcvczx | really? |
| 01:46.10 | mroszko_ | yea |
| 01:46.17 | xzcvczx | no per year cost or one off cost at all :O |
| 01:46.20 | mroszko_ | nope |
| 01:46.38 | xzcvczx | hmmm i seriously think windows is gonna go walled garden at some point then |
| 01:46.43 | mroszko_ | o they renamed it |
| 01:46.43 | mroszko_ | https://msdn.microsoft.com/windows/uwp/porting/desktop-to-uwp-run-desktop-app-converter |
| 01:46.47 | mroszko_ | anyway |
| 01:46.50 | xzcvczx | and why the hell haven't we moved over to the windows store yet |
| 01:46.53 | mroszko_ | you literally tell it to run whjatever installer |
| 01:46.56 | xzcvczx | much better esspecially if its free |
| 01:46.59 | mroszko_ | nsis/inno/installshield/msi |
| 01:47.06 | mroszko_ | and it'll run it and create packages |
| 01:47.06 | tom_fig | Yeah, it did work. That's what I used the other day. |
| 01:47.23 | mroszko_ | xzcvczx, ceo of ms said their new model is open source it first |
| 01:47.30 | mroszko_ | make sales guys figure out how to make $$ later |
| 01:47.37 | xzcvczx | lol |
| 01:48.58 | mroszko_ | will it be true long term? |
| 01:49.00 | mroszko_ | who knows |
| 01:49.07 | mroszko_ | right now they are more open than google in some regards |
| 01:49.13 | mroszko_ | open sourcing all the things |
| 01:50.09 | xzcvczx | meh google hasn't oss'd anything really cool in a while now |
| 01:50.32 | xzcvczx | nor has apple really, but at least they oss'd clang which wasn't exactly a small project |
| 01:52.05 | mroszko_ | TFS even uses git internally now instead of whatever the fuck it had |
| 01:56.54 | *** join/#kicad Novus (~Novus@2605:e000:628b:1400:df2:8679:73d0:e704) |
| 01:57.06 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 01:58.14 | mroszko_ | welp |
| 01:58.19 | mroszko_ | time to rip on the microchip sales guy |
| 01:58.27 | mroszko_ | Micrel website used to be useful and give me the alt parts |
| 01:58.31 | mroszko_ | microshit instead gives me flashy animations |
| 01:58.35 | mroszko_ | because thats useful |
| 01:59.18 | tom_fig | There's probably a spy for them in here |
| 01:59.29 | mroszko_ | good |
| 01:59.31 | mroszko_ | they can go eat a dick |
| 01:59.36 | mroszko_ | have you ever seen the ICD3 schematic? |
| 01:59.44 | mroszko_ | i just got one more fucking dead one on my desk this week |
| 01:59.49 | tom_fig | No, you have it? |
| 01:59.58 | mroszko_ | i have like $5k in dead ICD3s on my desk |
| 02:00.00 | mroszko_ | yes |
| 02:00.01 | mroszko_ | want a copy? |
| 02:00.03 | mroszko_ | its a joke |
| 02:00.06 | mroszko_ | 6 layer pcb |
| 02:00.12 | tom_fig | Yes, yes I do |
| 02:00.37 | tom_fig | They have some questionable app notes; that I do know |
| 02:00.40 | mroszko_ | https://drive.google.com/open?id=0B9lz7NcXtZcOelZWZnF1Nkd0YWM |
| 02:00.42 | mroszko_ | there |
| 02:00.45 | mroszko_ | its the manufacturing docs |
| 02:00.51 | mroszko_ | scroll down like 40 pages and it has schematics |
| 02:01.21 | mroszko_ | they have a sepic converter implemented through a pwm chip, not even a regulator |
| 02:01.30 | mroszko_ | crazyass scheme to generate the target voltages |
| 02:01.39 | mroszko_ | the stupid thing is unstable too and will overshoot to 6V every time |
| 02:02.30 | mroszko_ | its almost going to be cost effective soon |
| 02:02.35 | mroszko_ | for me to design a replacement regulator to tack in |
| 02:02.44 | mroszko_ | and also solder on extra bulk caps |
| 02:02.52 | mroszko_ | because theres only 3x 1ufs for the FPGA+Micro |
| 02:03.05 | tom_fig | Bunch of cheapskates |
| 02:03.19 | mroszko_ | they basically took the ICD2 |
| 02:03.19 | mroszko_ | removed the serial port |
| 02:03.27 | mroszko_ | and had a idiot try and redesign the regulator |
| 02:03.34 | mroszko_ | the rest is the same |
| 02:05.02 | *** join/#kicad talsit (foobar@gromit.mixdown.ca) |
| 02:05.15 | mroszko_ | i wonder if they legally go after me |
| 02:05.22 | mroszko_ | selling "fixed" ICD3s with a $100 markup |
| 02:07.46 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 02:10.41 | cirilo | mroszko_, you mean they actually paid to make the ICD2 even *worse*? Do they have customers who demand crappier products or something? |
| 02:11.43 | c4757p | yes, they're called "Microchip customers" |
| 02:14.23 | mroszko_ | lol |
| 02:14.57 | mroszko_ | they spent alot of effort animating the search on their website |
| 02:15.05 | mroszko_ | none of it on readability of product pages |
| 02:15.18 | mroszko_ | the stuff you actually care about? docs, etc? tiny links at the bottom |
| 02:15.29 | mroszko_ | but dont worry all the marketing bullshit has 20 inches of whitespace |
| 02:15.32 | mroszko_ | and size 20 font |
| 02:18.07 | cirilo | mroszko_, they were inspired by Apple .. to think that once I thought governments had a monopoly of shit web pages with 99.9999999% crap |
| 02:19.03 | *** join/#kicad shaun413 (uid121475@gateway/web/irccloud.com/x-memosismwihrbdnz) |
| 02:19.06 | cirilo | I really don't get it how tech companies get dumbasses to make websites that don't present any of the information that an engineer would want to see |
| 02:20.33 | cirilo | then again, after the Apple2e I think Apple was always shit at presenting technical info - was always "but it has shiny doo-dahs!" |
| 02:24.43 | mroszko_ | ugh |
| 02:24.46 | mroszko_ | whats a good ethernet phy |
| 02:24.47 | mroszko_ | 100mbit |
| 02:24.50 | mroszko_ | mii |
| 02:24.59 | mroszko_ | i used to use micrel but fuck mchip's website |
| 02:25.13 | xzcvczx | cirilo: hahahahah @ aus govt thinking it can block overseas retailers who don't charge gst for them |
| 02:25.17 | cirilo | NFI - never had to put in my own PHY |
| 02:25.22 | xzcvczx | thats just gonna be funny seeing it completely fucking fail |
| 02:26.22 | cirilo | xzcvczx, WTF - where was that? Wouldn't surprise me though - bunch of ignorant dumbfucks in federal parliament. Even have a flat-earther and some git obsessed with muslims. |
| 02:26.45 | xzcvczx | https://www.choice.com.au/shopping/online-shopping/buying-online/articles/gst-overseas-website-block |
| 02:28.09 | cirilo | hah - epic WTF. The shits obviously pushed that Gerry Harvey law without mentioning it to the press |
| 02:28.56 | cirilo | this will be so funny - with Border Farce on strike demanding more money and goods coming in to port which they have to stop, check if overseas company is required to pay GST, etc ... oh, just too fucking funny |
| 02:31.03 | cirilo | what a lame excuse though - shit will still be more expensive in Australia because Gerry Harvey and his like will bilk people of every cent they can |
| 02:31.26 | cirilo | even with a 100% import tax, shit will still be more expensive in Oz |
| 02:32.32 | cirilo | oh jesus .. gets better - the dumbfucks imagine that we somehow have trade agreements that authorise taxation in foreign jurisdictions .. too funny. |
| 02:34.43 | xzcvczx | maybe its time to just attach bender to australia and let him take them off shit planet |
| 02:36.19 | cirilo | I say we send all Lib and Labor politicians to Nauru or better still Cambodia - should be a huge improvement here - national IQ might even go up a few points |
| 02:46.23 | decimad | If I were a member of this project I would be kind of upset if blobs of commits were just dropped in without further notice or any public plans or discussion |
| 02:48.47 | *** join/#kicad ayjay_t (~ajp_@24-151-28-56.dhcp.nwtn.ct.charter.com) |
| 02:50.10 | decimad | oh, it's just a branch |
| 02:53.07 | *** join/#kicad kyranf (~kyranf@24.70.125.46) |
| 02:54.28 | xzcvczx | lol |
| 02:54.50 | xzcvczx | decimad: but don't worry decimad shit does get dropped in without further notice or public plans or discussion |
| 02:54.56 | xzcvczx | esspecially from dick |
| 02:55.36 | xzcvczx | hence the clusterfuck that is openssl |
| 02:56.52 | decimad | forking got ever easier since git :P |
| 02:57.48 | xzcvczx | tbh some people think its more a project thats only worth is cherry picking the good shit out of it and starting over |
| 02:58.57 | *** join/#kicad C47 (~Xwx47@189.242.29.24) |
| 02:59.30 | decimad | Well, I do see a big risk failing with that |
| 02:59.46 | xzcvczx | and therein lies the problem |
| 02:59.59 | xzcvczx | the same thing can easily happen with a fork unforunately |
| 03:00.05 | decimad | Refactoring is manageable imho |
| 03:00.37 | xzcvczx | with refactoring your quickly are unable to merge new shit in kicad into your fork easily though |
| 03:00.53 | decimad | no I'm not talking about forking, that was kidding |
| 03:01.18 | c4757p | stabs kicad with a fork |
| 03:01.26 | cyborg_ar | lol |
| 03:01.34 | cyborg_ar | decimad, what blob of commits? |
| 03:02.02 | xzcvczx | cyborg_ar: dicks shit most likely |
| 03:02.05 | xzcvczx | re swig |
| 03:02.13 | xzcvczx | he was worried it was being direct commited |
| 03:02.28 | decimad | cyborg_ar: that swig stuff dick is onto... somehow I read master in that link and thought, what's up there! |
| 03:02.31 | cyborg_ar | was this a while ago? |
| 03:02.56 | xzcvczx | currently |
| 03:03.01 | cyborg_ar | last dick commit was he adding some fields |
| 03:03.08 | cyborg_ar | to the drc dialog |
| 03:03.22 | xzcvczx | cyborg_ar: yeah its a private branch |
| 03:03.42 | cyborg_ar | anyway swig needs to die |
| 03:03.46 | xzcvczx | was linked a bit ago in here, can't find it currently |
| 03:04.04 | cyborg_ar | i've never seen a project use it for python and have a decent api |
| 03:04.06 | *** join/#kicad mroszko__ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 03:04.10 | decimad | https://code.launchpad.net/~dickelbeck/kicad/+git/swig-enhancements/+ref/master |
| 03:04.22 | decimad | work on your grep-eyes xzcvczx |
| 03:04.33 | cyborg_ar | blender and the BGE used to use it before they came to their senses and implemented python support properly |
| 03:04.47 | decimad | what do they use? |
| 03:05.04 | xzcvczx | decimad: if my eyes had good grep i wouldn't need grep :) |
| 03:05.24 | cyborg_ar | the proper way |
| 03:05.27 | cyborg_ar | python.h |
| 03:05.39 | decimad | they're doing it manually? |
| 03:05.45 | cyborg_ar | yes, the proper way |
| 03:05.47 | xzcvczx | a lot of shit is |
| 03:05.55 | xzcvczx | even opencv does it properly |
| 03:05.57 | cyborg_ar | doing it any way else is half assed |
| 03:06.17 | xzcvczx | the api needs to be fixed first really though |
| 03:06.39 | decimad | is there an api? isn't it rather a blob of stuff here and there? :) |
| 03:06.54 | xzcvczx | kicad defines broke-ass api :P |
| 03:07.24 | decimad | First thing to add and provide long-time support is the KIWAY |
| 03:07.41 | cyborg_ar | some automated tools have improved a lot lately though |
| 03:07.46 | xzcvczx | i wish kiway would go |
| 03:07.47 | cyborg_ar | that pybind thing looks interesting |
| 03:07.53 | xzcvczx | i mean ffs just use damn .so's |
| 03:07.56 | cyborg_ar | would be a cleaner way |
| 03:08.14 | decimad | pybind is boost.python on c++11... there's such library for lua as well, etc. they exist since ages |
| 03:08.27 | cyborg_ar | xzcvczx, i think they did the kiface thing to sidestep the issue that cropped up with cirilo's plugins |
| 03:08.48 | xzcvczx | and which issue was that? |
| 03:08.52 | cyborg_ar | that being library paths, but it was figured out properly, i see no reason to keep doing it |
| 03:09.13 | cyborg_ar | *) Extend SWIG support deeper into the BOARD class. |
| 03:09.15 | xzcvczx | yeah breaking shit more ain't helping noone |
| 03:09.15 | cyborg_ar | lol |
| 03:09.41 | cyborg_ar | yes that's what kicad needs, deeper coupling with it's insides |
| 03:10.54 | cirilo | cyborg_ar, wut kiface thing? Was there some change I don't know about? |
| 03:11.25 | xzcvczx | cirilo: aka why the kifaces are in <prefix>/bin |
| 03:11.27 | cyborg_ar | cirilo, using .kiface and installing in /bin instead of using a .so installed in lib like you do |
| 03:11.42 | cyborg_ar | when you did it initially make install broke |
| 03:12.02 | xzcvczx | meh could use rpath if we really gave a damn |
| 03:12.12 | cyborg_ar | rpaths are not the answer though |
| 03:12.24 | xzcvczx | they are technically |
| 03:12.37 | xzcvczx | if you dont' want/can't add shit to /etc/ld.so.conf |
| 03:12.46 | xzcvczx | you shouldn't need to |
| 03:12.46 | cirilo | oh - that's got nothing to do with plugins |
| 03:12.57 | cirilo | sheer laziness |
| 03:13.13 | cirilo | but whenever I bring up the topic I gen nothing but excuses |
| 03:13.29 | xzcvczx | cirilo: he was theorising why the shit was in /bin though |
| 03:14.38 | cirilo | need to look in more detail of boost::context and how that code is actually loaded to be able to say |
| 03:15.51 | decimad | hrm? |
| 03:16.01 | xzcvczx | it looks in a logical place on osx |
| 03:16.09 | cirilo | it might have something to do with how that partucular code is loaded into process space |
| 03:16.30 | cirilo | whre is it on osx? |
| 03:16.42 | xzcvczx | <bundle>/Contents/Plug-Ins |
| 03:16.44 | xzcvczx | <PROTECTED> |
| 03:16.51 | decimad | boost::context just sets up a stack frame and runs a function ontop of it... how would they be related? |
| 03:16.58 | xzcvczx | decimad: are you a weendows user? |
| 03:17.04 | xzcvczx | (can never remember) |
| 03:17.12 | decimad | yes |
| 03:17.43 | xzcvczx | decimad: is the windows taskbar able to enumerate the kicad windows? |
| 03:18.39 | decimad | like showing all kicad windows screen minis when hoving over a single? |
| 03:18.44 | xzcvczx | yeah |
| 03:18.49 | xzcvczx | so you can easily swith between |
| 03:19.06 | decimad | works for me on master |
| 03:19.12 | xzcvczx | ok thanks |
| 03:19.24 | decimad | they're aligned anyways though |
| 03:19.59 | *** join/#kicad greyfiend (~greyfiend@203-57-212-112.dyn.iinet.net.au) |
| 03:21.27 | decimad | Also it would be great if the build number would vanish so the filename fits on the taskbar button ;) |
| 03:21.49 | greyfiend | I'm trying to add the 'Connectors_JST.pretty' library from the official github repo via the wizard, but it gives me 'status invalid' any ideas? fresh 4.04 install |
| 03:23.23 | cyborg_ar | greyfiend, do the other libraries work? |
| 03:24.50 | greyfiend | @cyborg_ar apparently not status:invalid |
| 03:25.14 | xzcvczx | greyfiend: when you type in the path are you trying blah/kicad or blah/kicad/Connectors_JST.pretty? |
| 03:25.35 | tom_fig | Stupid msys |
| 03:26.28 | tom_fig | Great, just great. I can't build because of a cygheap base mismatch...it tells me I have an incompatible cygwin dll....except cygwin is not installed. So it's something else mucking it up |
| 03:26.29 | decimad | Maybe it would pose fewer problems using visual studio to create binaries than keeping msys running? :D |
| 03:26.38 | greyfiend | @cyborg_ar I'm selecting the library from the list in the wizard |
| 03:27.16 | cyborg_ar | something could have broken |
| 03:27.18 | tom_fig | Yeah, maybe. Does it work well? |
| 03:27.22 | cyborg_ar | can you access github just fine? |
| 03:27.30 | greyfiend | yep |
| 03:27.39 | decimad | you'd have me as a pseudo-regular dev that uses it for development, and could maybe fix stuff here and there |
| 03:29.21 | decimad | i think gyrosgeier only set up msvc builds for esoteric reasons, he doesn't use it, right? |
| 03:29.55 | tom_fig | Yeah, something like that |
| 03:30.45 | cyborg_ar | greyfiend, it works for me |
| 03:30.51 | cyborg_ar | sorry :( |
| 03:30.56 | tom_fig | He's got a kicad msi installer on his github, lol |
| 03:33.10 | greyfiend | @cyborg_ar added it manually and it worked, must be something whack with the wizard, cheers cyrborg_ar |
| 03:39.59 | cirilo | decimad for the moment mostly to get compiler warnings - gcc ignores a lot |
| 03:40.27 | decimad | yeah, I ran over these bombs as well ;) |
| 03:40.41 | decimad | plus the diagnostics that get triggered at runtime ;) |
| 03:40.52 | cirilo | but I think if the Python issues can be sorted we should have native MSVC builds and dump MSys - and if MS release free-to-use clang tools then we can even use MS-clang |
| 03:41.19 | *** join/#kicad cosmobird (~firebird@125.16.18.198) |
| 03:41.27 | cirilo | back soon - Bill wants me to reboot |
| 03:44.11 | xzcvczx | haha i dunno wayne seems pretty damn dedicated to msys |
| 03:48.41 | cyborg_ar | he might soften a bit, he got pretty annoyed the last few times it broke everything |
| 03:49.32 | *** join/#kicad mroszko_ (~mroszko@413335d3.cst.lightpath.net) |
| 03:50.45 | eszett | I have a question. When having a 0.5mm pitched IC, what with have the traces to be? It seems that DRC complains that 0.25mm wide traces are too thick (track too close to..) |
| 03:50.57 | eszett | with=width |
| 03:52.56 | decimad | Usually you set up the clearance rules, so it's not the DRC that enforces anything on you |
| 03:53.28 | eszett | ye, but im not sure whether to go with 0.25 wide traces or 0.2mm |
| 03:54.38 | decimad | the choice is easy if 0.25 violates your manufacturer's clearance rules |
| 03:56.34 | decimad | cirilo: bill was ten years ago :P |
| 03:56.46 | decimad | or 15 even? |
| 03:57.01 | eszett | decimad: ye |
| 03:57.52 | eszett | decimad: I cant find this in the specification, but i found that the min trace width is 0.089mm, ie. i go for 0.20mm wide traces, just to be sure |
| 03:58.27 | decimad | the manufacturer doesn't specify min clearance? |
| 03:59.17 | eszett | decimad: cant see it here: https://easyeda.com/Doc/Tutorial/PCBOrder.htm |
| 03:59.20 | eszett | do you see it? |
| 04:00.13 | eszett | maybe im dumb as a brick again.. but i cant see it listed in this capabilities list |
| 04:01.02 | decimad | they call it "space" or "spacing" |
| 04:01.52 | *** join/#kicad DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 04:01.58 | mroszko_ | nooooooooo |
| 04:01.59 | mroszko_ | fuck |
| 04:02.03 | mroszko_ | digikey has gone full retard |
| 04:02.07 | mroszko_ | their product list is awful now |
| 04:02.20 | mroszko_ | wait |
| 04:02.28 | eszett | decimad: ah, thanx! |
| 04:02.49 | cyborg_ar | they did do an improvement lately |
| 04:02.55 | cyborg_ar | now you can filter out obsolete parts |
| 04:03.02 | mroszko_ | http://i.imgur.com/MKqrFbm.png |
| 04:03.14 | mroszko_ | thats the new look |
| 04:03.17 | c4757p | wat? where do you see that? |
| 04:03.17 | cyborg_ar | ah |
| 04:03.31 | cyborg_ar | pictures for the attention span challenged |
| 04:03.32 | mroszko_ | it randomly gives it to you |
| 04:03.36 | mroszko_ | i.e. AB testing |
| 04:03.38 | c4757p | oh |
| 04:03.59 | c4757p | well, that's delightfully stupid |
| 04:05.15 | mroszko_ | yea... |
| 04:05.22 | cyborg_ar | they are starting to implement ranges for parameters though |
| 04:05.25 | cyborg_ar | that is nice |
| 04:06.12 | c4757p | yeah |
| 04:07.20 | mroszko_ | now i gotta figure out which category the actual subcategory is under :( |
| 04:08.17 | mroszko_ | they could at least have a power user tab/alt page |
| 04:08.21 | mroszko_ | rather than just removing it |
| 04:09.56 | cyborg_ar | another good thing would be to find a way to split off and normalize parameters for MCU peripherals |
| 04:10.21 | cyborg_ar | right now they are all lumped together so there's a lot of control-click to be done to narrow dow |
| 04:10.22 | cyborg_ar | n |
| 04:10.27 | mroszko_ | meh |
| 04:10.31 | mroszko_ | most have the same peripherals |
| 04:10.35 | mroszko_ | little point |
| 04:10.41 | mroszko_ | unless you are shopping 8051s |
| 04:10.46 | cyborg_ar | lol |
| 04:11.16 | c4757p | <mroszko_> unless you are shopping 8051s --- you do realize those are still used in decent volume...right? |
| 04:12.00 | mroszko_ | yes? |
| 04:12.08 | mroszko_ | but they are the most nonstandard-appspecific |
| 04:12.25 | mroszko_ | anything ARM will more or less have the same peripherals of some sort between all the manus |
| 04:12.41 | c4757p | that "more or less" can be important... |
| 04:13.40 | cyborg_ar | sometimes i would like to see which micros have USB |
| 04:13.51 | cyborg_ar | or an adc |
| 04:13.59 | cyborg_ar | or a specific adc |
| 04:14.14 | cyborg_ar | but there are many combinations of ADC, USB, and everything else |
| 04:14.24 | cyborg_ar | so i need to control-click all the ones that have what i need |
| 04:14.28 | cyborg_ar | to narrow down |
| 04:14.53 | cyborg_ar | perhaps a way to just grep and select the matching lines in the filter would be enough |
| 04:15.03 | cyborg_ar | no need to do much expensive data normalization |
| 04:15.32 | cyborg_ar | i'm wary of using text search because it tends to leave shit out |
| 04:24.02 | mroszko_ | honestly seems like a crappy way to design though |
| 04:24.10 | mroszko_ | its usually best tos tay with a manufacturer you always used |
| 04:24.16 | mroszko_ | to be able to reuse code for other perips :/ |
| 04:24.24 | c4757p | er, so? filter on manuf too |
| 04:24.25 | mroszko_ | not that important but eh |
| 04:24.27 | mroszko_ | lol |
| 04:24.51 | cyborg_ar | well if you're so marries to your manufacturer you can just use their parametric tools |
| 04:24.56 | mroszko_ | manuf websites usually have more detailed and up to date filtered |
| 04:24.57 | mroszko_ | yea |
| 04:25.00 | c4757p | cyborg_ar: if they're usable.. |
| 04:25.06 | mroszko_ | lol |
| 04:25.06 | cyborg_ar | it's usually better than using a distributor |
| 04:25.12 | mroszko_ | i think TI has made theirs worse |
| 04:25.17 | cyborg_ar | c4757p, in that case you can reevaluate your marriage |
| 04:25.20 | mroszko_ | they started creating sidebar filters |
| 04:25.26 | mroszko_ | in addition to table filters |
| 04:25.28 | c4757p | TI's always been awful |
| 04:25.32 | mroszko_ | with weird "key names" with no explanation |
| 04:26.32 | cyborg_ar | the problem with using manufacturers websites is that they are going to list every single part they could possibly manufacture |
| 04:26.47 | cyborg_ar | given enough MOQ and weeks of lead time |
| 04:27.04 | cyborg_ar | soyou need to cross check with distributors anyway |
| 04:27.35 | *** join/#kicad daey_ (~Flutterba@unaffiliated/day) |
| 04:27.46 | *** join/#kicad Sujith (~Sujith@117.209.200.28) |
| 04:29.49 | mroszko_ | hmm |
| 04:29.55 | mroszko_ | does type-c need a mux/switch IC |
| 04:30.00 | mroszko_ | for legacy usb over it |
| 04:30.13 | c4757p | no |
| 04:30.36 | c4757p | just short together the mirrored pairs close to the connector :P |
| 04:31.05 | mroszko_ | im too lazy to read the spec |
| 04:31.30 | c4757p | I spent rather a while reading it, though I don't know how much I remember ;0 |
| 04:31.33 | c4757p | ;) |
| 04:31.42 | c4757p | but no, no active hardware necessary to use it as a legacy port |
| 04:37.24 | mroszko_ | omfg |
| 04:37.33 | mroszko_ | so i ranted at the microchip sales guy to find me the latest phy |
| 04:37.36 | mroszko_ | and he CCed a FAE |
| 04:37.43 | mroszko_ | who now linked me a 5 year old chip |
| 04:37.47 | mroszko_ | instead of the 3 year old |
| 04:37.47 | mroszko_ | wow |
| 04:37.55 | mroszko_ | *drinks more alcohol* |
| 04:42.13 | cyborg_ar | hmm i really really dislike the `fixes` alias |
| 04:42.46 | cyborg_ar | since it uses ammend, if you use it in the wrong order it will make a mess |
| 04:43.31 | cyborg_ar | it should instead be a commit command, with the fix footer already pre-filled |
| 04:45.59 | *** join/#kicad Kevlan (~Kevlan@c-50-134-227-152.hsd1.co.comcast.net) |
| 05:09.19 | *** join/#kicad VA3VNA (~mayday_ja@67.70.9.37) |
| 05:23.52 | *** join/#kicad cirilo (~cirilo@CPE-120-144-53-157.lnse5.win.bigpond.net.au) |
| 05:24.34 | nickoe | mroszko_: What is a FAE? |
| 05:31.37 | *** join/#kicad ayjay_t (~ajp_@24-151-28-56.dhcp.nwtn.ct.charter.com) |
| 05:37.16 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 05:37.19 | mroszko_ | field apps engineer |
| 05:44.46 | *** join/#kicad C47 (~Xwx47@189.242.29.24) |
| 05:50.34 | Yatekii | good morning ladies and sirs |
| 05:50.58 | Yatekii | hmm is the footprint editor code at in the pcb new code? |
| 05:51.26 | nickoe | mroszko_: well, maybe he knows better :) |
| 05:51.37 | nickoe | Yatekii: what? |
| 05:52.47 | nickoe | mroszko_: there is a feedback button on that digikey page for you |
| 05:56.01 | nickoe | xzcvczx: there is a description of why it is in bin on the bug tracker |
| 05:56.39 | cirilo | ok, if I was looking for 'scopes .. what would people typically recommend? |
| 05:57.27 | cirilo | although I sometimes work with circuits > 400MHz the reality is I probably can't afford a 400 or 500MHz scope :( |
| 05:57.56 | cirilo | but 100 is uncomfortably close to frequencies I typically work with |
| 06:01.23 | *** join/#kicad rdmitry (~Dmitry@95-37-249-206.dynamic.mts-nn.ru) |
| 06:05.59 | archivist | cirilo, new or used |
| 06:07.26 | cirilo | archivist, new |
| 06:08.47 | archivist | I last bought a new scope....40 ish years ago |
| 06:11.41 | mroszko_ | just get a rigol |
| 06:11.46 | mroszko_ | and use the hax to get 100mhz |
| 06:13.26 | archivist | I picked up a used HP logic analyser stuffed with scope cards for not a lot |
| 06:14.00 | *** join/#kicad ETS_ (ad47c341@gateway/web/freenode/ip.173.71.195.65) |
| 06:14.13 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 06:16.40 | ETS_ | Need help with a grond plane zone. I don't understand what is necessary to do. I read that the "zone tool" can be clicked and then right-clicked and the fill of unfill zones choice comes up. Well, I can't get that consistantly. |
| 06:17.23 | ETS_ | Sometimes I get it, other times I can't, but I cannot pinpoint what the difference is between them. I get it that the fill will not work if there are drc errors. |
| 06:17.32 | cirilo | last time I got a used scope (in the bad old days) it was a lemon, so I don't trust them unless I can actually sit in front of them and check them out |
| 06:19.55 | mroszko_ | cirilo, get this |
| 06:19.56 | mroszko_ | https://www.rigolna.com/products/digital-oscilloscopes/ds1000Z/ds1054z/ |
| 06:20.00 | mroszko_ | then use the 100mhz hack |
| 06:20.03 | mroszko_ | \unlock code crack |
| 06:20.43 | archivist | if working at that freq then that is pushing it |
| 06:20.50 | mroszko_ | no |
| 06:20.54 | archivist | needs something better |
| 06:20.55 | mroszko_ | the 100mhz is a upgrade code |
| 06:21.11 | mroszko_ | but a bored person cracked the ec curves used to generate the codes |
| 06:21.37 | *** join/#kicad pointhi21 (~pointhi21@dyn166055.wlan.jku.at) |
| 06:21.47 | archivist | you forgot the "although I sometimes work with circuits > 400MHz" |
| 06:21.56 | mroszko_ | better get the bank account out :P |
| 06:22.09 | archivist | and but 100 is uncomfortably close to frequencies I typically work with |
| 06:22.20 | mroszko_ | cheapest will still be the rigols |
| 06:22.21 | mroszko_ | http://www.tequipment.net/rigol-4000-series-promotion/ |
| 06:22.27 | mroszko_ | best bang for buck too |
| 06:22.57 | *** join/#kicad decimad (~deci@ip5f5be505.dynamic.kabel-deutschland.de) |
| 06:23.14 | mroszko_ | but still run around 4-6k for the DSO |
| 06:23.22 | mroszko_ | cheaper than a techtronix for $20k |
| 06:23.32 | mroszko_ | + tectronix probe tax |
| 06:23.35 | mroszko_ | $500 or gtfo |
| 06:24.44 | cirilo | mroszko_, thanks - I'll have a look at Rigol again .. and if there's an Australia Tax I'll buy the fucking thing when I'm in England |
| 06:25.02 | trygvis | I wonder how many times I will forget the copper fill when producing PCBs |
| 06:25.18 | mroszko_ | tek fancy scopes which are like $20k |
| 06:25.21 | trygvis | is there a way through the python api to check if a zone is filled? |
| 06:25.22 | mroszko_ | also have "special" probes |
| 06:25.25 | mroszko_ | $500 each |
| 06:25.35 | mroszko_ | an adapter for normal probes is also $500 (just a box with a memory chip) |
| 06:25.36 | cirilo | Tek, like HP, always had "special" probes |
| 06:25.47 | mroszko_ | they also made the opening have a tighter fit |
| 06:25.51 | mroszko_ | so event hough it has a BNC |
| 06:25.56 | mroszko_ | not all normal probes fit |
| 06:26.03 | cirilo | though once upon a time many Tek models also took a non-special probe and it only meant you couldn't do certain things |
| 06:26.23 | trygvis | ETS_: I didn't understand your question .. or was it just frustration? :) |
| 06:26.58 | mroszko_ | i dont even hear agilents name anymore |
| 06:27.06 | mroszko_ | beyond reference to 20-30 year old equipment |
| 06:27.26 | mroszko_ | they priced themselves out to irrelevence |
| 06:27.37 | ETS_ | No, there is more, but wanted to see if others were watching before sending it. |
| 06:27.38 | cirilo | heh .. so for Rigol, 4Ch * 1GS/s = 4GS/s .. |
| 06:28.02 | *** join/#kicad erichvk (~esh@ppp121-45-127-106.lns20.adl6.internode.on.net) |
| 06:28.22 | cirilo | yeah - hopefully Agilent go bust. Weren't they bought by Keystone or something? |
| 06:28.29 | cirilo | the test equipment anyway |
| 06:28.51 | archivist | renamed keysight |
| 06:28.57 | cirilo | lots of companies that ripped people off in the 1980s with hellishly expensive equipment are now long gone |
| 06:29.07 | cirilo | Keysight - that's it |
| 06:29.59 | ETS_ | <PROTECTED> |
| 06:30.14 | cirilo | oh .. 2Ch 500MHz or 4Ch 500 MHz .. *sigh* |
| 06:30.24 | cirilo | though really I rarely ever need >2Ch |
| 06:30.56 | cirilo | though I've only ever had 4Ch DSOs before. :) |
| 06:31.01 | ETS_ | Then, when I do get a fill, I can't edit it or change it, nor can I make it unfill. Nothing makes sense, and I have not seen any instructional vids that I can actually follow to make it work. |
| 06:32.04 | ETS_ | My present problem is this: I have a ground plane that picked up all the boards grounds (GND), except for the grounds on one footprint. They were made the same way, and spelled the same. |
| 06:32.52 | ETS_ | The netlist has 4 devices but the "GND" lables only come across the netlist on the one footprint that will not pick up GND. Anyone seen this before? I don't even know how to troubleshoot this. |
| 06:33.51 | ETS_ | I have photos. Where should I post them? |
| 06:35.16 | trygvis | anywhere is probably fine |
| 06:35.27 | cirilo | ETS_, do you have fill zones inside fill zones? |
| 06:35.56 | *** join/#kicad kraiskil (~kraiskil@85-76-73-49-nat.elisa-mobile.fi) |
| 06:37.15 | ETS_ | I don't think so, it would only be by accident from trying to get one to flow. I do not have two zones on the same plane. |
| 06:38.41 | ETS_ | trygvis, I use one yesterday, but don't remember the site name. |
| 06:39.04 | trygvis | https://snag.gy/ seems useful |
| 06:42.30 | *** join/#kicad sandeepkr (~sandeepkr@111.235.65.5) |
| 06:42.31 | *** join/#kicad kuldeep (~kuldeep@unaffiliated/kuldeepdhaka) |
| 06:44.30 | ETS_ | <PROTECTED> |
| 06:46.15 | ETS_ | At the top you can see some thermos but the PCIe part has some GND symbols that aer not "picked-up" |
| 06:46.39 | *** join/#kicad xorly (~xorly@ip-89-176-10-118.net.upcbroadband.cz) |
| 06:47.00 | *** join/#kicad firewalker (~firewalke@46.246.152.93.dsl.dyn.forthnet.gr) |
| 06:47.08 | trygvis | fill the zone and they should be ok |
| 06:47.55 | *** join/#kicad cnx (b92ed638@gateway/web/freenode/ip.185.46.214.56) |
| 06:47.59 | cnx | Good morning |
| 06:48.27 | ETS_ | trygvis -- the first one WAS filled |
| 06:49.10 | cnx | In PCBNEW (v4.0.4) that after zooming there is an automatic slow vertical scroll which annoys me - somebody knows how to turn this off and scroll faster? |
| 06:49.15 | ETS_ | <PROTECTED> |
| 06:49.50 | trygvis | doesn't look filled to me |
| 06:50.03 | mroszko_ | hmm how would one drow a rotary encounter symbol |
| 06:50.17 | cnx | I've assumed that option "Center and warp cursor on zoom" has something to do with it but it doesn't. |
| 06:52.08 | ETS_ | Let me try this again. |
| 06:53.25 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 06:53.26 | trygvis | it should look like this when it is filled: https://snag.gy/vAlD6W.jpg |
| 06:53.56 | *** join/#kicad eduardas_m (~eduardas_@213.197.143.19) |
| 06:54.01 | ETS_ | https://snag.gy/A0GmnP.jpg |
| 06:54.14 | ETS_ | Any better? |
| 06:54.54 | trygvis | looks like a good fill on In2.Cu. is that net connected to GND? |
| 06:56.57 | ETS_ | follow the ratsnest lines and look at the pads. Are they all thermally relieved? |
| 06:57.43 | ETS_ | And yes GND was the selected net whened filled |
| 06:58.38 | *** join/#kicad mroszko__ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 06:59.18 | trygvis | hm, the drill doesn't look right for the lower two pin rows, slightly off form the pad's center |
| 07:02.15 | ETS_ | there is an underlying circle imported with the edge cut, and the footprint was made from the specification. Other pads are dead-nuts on. The Traces all are locked on to the grid. No drc errors. |
| 07:02.58 | ETS_ | Ant the lower rows are the problem |
| 07:03.06 | ETS_ | and |
| 07:03.35 | *** join/#kicad sandeepkr_ (~sandeepkr@103.49.155.114) |
| 07:04.39 | *** join/#kicad sandeepkr__ (~sandeepkr@111.235.65.5) |
| 07:04.42 | trygvis | yeah, the other are connected properly. |
| 07:04.47 | ETS_ | It has me stumped. But then, Kicad has made a 100 hour project become a 2 month 20-hr/day oddesey |
| 07:05.20 | trygvis | I'm not sure what is up with that |
| 07:07.34 | ETS_ | if you will note above, the only footprint that has GND in the netlist is the one with the problem. and the netlist has ALL of the GND listed. |
| 07:08.16 | ETS_ | know of a better place to persue this? |
| 07:09.49 | trygvis | on the user's email list I guess, but people might come around here later on too |
| 07:10.11 | trygvis | http://tech.groups.yahoo.com/group/kicad-users/ |
| 07:11.12 | cirilo | I see wer're back to the Float Validator woes .. didn't anyone create a special float validator for KiCad which could also do magic like take units - or was that just all talk? :P |
| 07:13.14 | cirilo | ETS_, did you check (a) zone properties and the type of thermals etc? Then (b) PAD properties to see if someone set properties like "don't connect to zone" |
| 07:13.22 | xzcvczx | nickoe: i don't care how many descriptions there are... still doesn't make it right :) |
| 07:13.45 | cirilo | ah .. time to escape the computer .. I wish |
| 07:14.23 | xzcvczx | the matrix is inescapable |
| 07:16.07 | cirilo | definitely time to escape the computer - I made a design which *can* be assembled but it's impossible to perform the intended adjustment. :O |
| 07:16.28 | cirilo | when I start to make n00b mistakes it's time to get away |
| 07:16.49 | ETS_ | cirilo I'll give it a look. are these in the footprint file, or thr kicad-pcb file |
| 07:18.41 | *** join/#kicad mroszko_ (~mroszko@413335d3.cst.lightpath.net) |
| 07:19.56 | ETS_ | this leads me back to my first problem, how do I open the fill/unfill dialog window consistantly. I still just sort of stumble over it, but never when I WANT it. |
| 07:21.20 | cirilo | ETS_, in both, but the kicad-pcb file inherits pad properties from the footprint file |
| 07:21.51 | cirilo | but if you change the footprint you will have to reload all occurrences in the PCB design - but that's the fastest way if you have many instances of the problem component |
| 07:22.26 | cirilo | in the fill, your thermal reliefs also look very thin - so check to make sure they're a sensible width |
| 07:22.37 | *** join/#kicad mroszko__ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 07:22.48 | cirilo | I also see that some tracks laid to the pad have filled part of the thermal gap (not bad if you intend it to) |
| 07:25.39 | *** join/#kicad kraiskil (~kraiskil@145.247.15.238) |
| 07:31.57 | *** join/#kicad berjan (~chatzilla@ip54538619.speed.planet.nl) |
| 07:34.57 | ETS_ | the pad L10 has a wide trace to it that is supposed to connect one good pad to one bad pad to see if I was not getting a conversion from the netlist correctly. I have been f***ing with this for a week. |
| 07:35.24 | cirilo | ETS_, just 'b' and 'ctrl-B' I think - can't remember - but 'b' is definitely one of them |
| 07:36.01 | ETS_ | thank. that should be i the list of shortcuts? |
| 07:36.10 | ETS_ | i = in |
| 07:36.24 | cirilo | probly one of those problems that could be sorted quickly if someone else could sit down and show you .. normally takes ages on the internetz |
| 07:36.31 | cnx | In PCBNEW, is it possible to show the ratsnest only for the selected group of footprints? |
| 07:36.35 | cirilo | yup - should eb in shortcuts |
| 07:36.54 | ETS_ | I'll try it. |
| 07:38.19 | ETS_ | you were right b to add, CTRL=b to remove. |
| 07:41.31 | cirilo | cnx, no :( |
| 07:41.45 | cnx | cirilo, thanks! |
| 07:42.04 | cirilo | it would be nice to do taht though so we're not flooded with ratsnests we're not interested in |
| 07:42.18 | cnx | cirilo, exactly. |
| 07:42.44 | cnx | It's nice for what I call macro placement of larger groups of footprints |
| 07:45.28 | Yatekii | nickoe i dont know how to put that differently lol |
| 07:45.43 | Yatekii | its exactly what i asked :D |
| 07:49.06 | cyborg_ar | LOL |
| 07:49.18 | cyborg_ar | got a vendor step file and it's huge |
| 07:49.29 | cyborg_ar | found an easter egg under the epoxy http://imgur.com/a/PyVB9 |
| 07:54.52 | *** join/#kicad maiden (~maiden@c83-255-72-208.bredband.comhem.se) |
| 08:00.09 | cirilo | cyborg_ar, component or an electronic assembly? |
| 08:00.26 | cirilo | though even connector models can be >10MB :) |
| 08:00.42 | cyborg_ar | electronic assembly |
| 08:01.03 | cyborg_ar | including the construction detail under the epoxy is a bit too much |
| 08:01.14 | cyborg_ar | i'd have thought that was a trade secret |
| 08:01.54 | cyborg_ar | now i know the spectrometer is powered by joe, who eats worms |
| 08:02.34 | xzcvczx | cirilo: lol we all know your hate for the poor step format |
| 08:03.32 | cyborg_ar | cirilo, do you think the step plugin could be extended to accept gzipped step files? that should cut down a bit on the fat |
| 08:05.56 | *** join/#kicad riq_ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 08:07.56 | *** join/#kicad riq__ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 08:11.21 | *** join/#kicad ricardocrudo (~ricardocr@185.19.197.52) |
| 08:14.11 | ETS_ | cirilo-- Had to rebuild the footprint, and now it works. You gave me the idea when you questioned the grid. I guess some pins were off. A big Thanks to you. |
| 08:14.17 | *** join/#kicad Venemo (~venemo@fedora/Venemo) |
| 08:14.30 | *** join/#kicad crazy_imp (~mj@unaffiliated/crazy-imp/x-8704303) |
| 08:15.52 | cyborg_ar | hmm, xz compresses a step file to about 1/10 the size with default settings :D |
| 08:22.59 | xzcvczx | cyborg_ar: meh submit it to the standards commit to make stz zipped step :P |
| 08:23.32 | cyborg_ar | why |
| 08:23.40 | cyborg_ar | you just need to add .gz at the end |
| 08:23.46 | cyborg_ar | we've been over this |
| 08:23.47 | cyborg_ar | lol |
| 08:24.17 | cyborg_ar | hmm maybe i can use zlib overloading std::ifstream |
| 08:24.49 | inductiveload | cyborg_ar, a boost filtered stream can do deflate/inflate |
| 08:25.04 | inductiveload | it's the canonical example of a boost filtered stream |
| 08:25.25 | cyborg_ar | yeah, that would work |
| 08:30.41 | cyborg_ar | hmm |
| 08:30.44 | cyborg_ar | not so simple |
| 08:31.04 | cyborg_ar | the plugin passes around filenames up to oce instead of file-like objects |
| 08:31.05 | cyborg_ar | owell |
| 08:33.40 | inductiveload | patch oce ;-) |
| 08:35.30 | cyborg_ar | hell no |
| 08:35.38 | inductiveload | wuss |
| 08:35.43 | xzcvczx | meh just pipe it through unxz |
| 08:35.50 | cyborg_ar | yes |
| 08:35.59 | cyborg_ar | there is a cache dir already |
| 08:36.23 | inductiveload | what's the gain of xz over regular gzip? |
| 08:36.38 | xzcvczx | a shitload of space |
| 08:36.43 | xzcvczx | well ratio |
| 08:36.44 | cyborg_ar | there could be some logic at the highest level to uncompress any file with a .gz/bz2/xz extension before passing it on to the plugin |
| 08:37.00 | xzcvczx | cyborg_ar: there is already the option to archive projects |
| 08:37.05 | xzcvczx | it just doesn't do models |
| 08:37.07 | xzcvczx | stupid code |
| 08:37.07 | cyborg_ar | in a couple unscientific tests i ran it compresses step files about twice as much |
| 08:37.17 | inductiveload | not bad |
| 08:37.31 | cyborg_ar | xzcvczx, i would like to have the xzipped files in the library |
| 08:38.13 | cyborg_ar | but yeah the project archival thingy should copy over models too |
| 08:38.27 | cyborg_ar | and rewrite the paths to point to the new location |
| 08:45.36 | *** join/#kicad ziongate (~spritz@sysam.it) |
| 08:47.30 | *** join/#kicad berjan (~chatzilla@2001:67c:2564:a156:c03:f790:7285:8d82) |
| 08:48.44 | xzcvczx | meh effort |
| 08:48.51 | xzcvczx | the archiver doesn't even copy libs |
| 08:49.20 | cyborg_ar | well, it doesn't have to |
| 08:49.50 | cyborg_ar | perhaps another option would be to do something like the -cache.lib for 3d |
| 08:49.59 | cyborg_ar | a compressed file with all the meshes |
| 08:50.00 | xzcvczx | ewww no |
| 08:50.13 | xzcvczx | i would prefer it to just embed the step into the .kicad_mod file |
| 08:50.19 | xzcvczx | and archive that into the project folder |
| 08:50.25 | cyborg_ar | eww |
| 08:50.36 | cyborg_ar | that would make the kicad_mod absolutely massive |
| 08:50.40 | xzcvczx | so? |
| 08:50.56 | cyborg_ar | and you usally want to keep track of changes in those |
| 08:55.23 | cyborg_ar | i'm about to do something horrible |
| 08:55.39 | cyborg_ar | there's a python module for scripting GUI actions |
| 08:56.10 | cyborg_ar | i'm gonna script it to do plotting from the command line |
| 08:56.57 | xzcvczx | wasn't there someone in here earlier proclaiming their greatness for getting plotting done on command line? |
| 08:57.25 | cyborg_ar | on eeschema |
| 08:58.04 | xzcvczx | i didn't realise eeschema could plot |
| 09:00.33 | cyborg_ar | argh |
| 09:00.43 | cyborg_ar | eeschema doesn't say eeschema in the window title |
| 09:08.36 | cyborg_ar | oooh yay |
| 09:08.44 | cyborg_ar | this actually works surprisingly well |
| 09:08.54 | cyborg_ar | it is easier than coding against the pcbnew python api |
| 09:08.55 | cyborg_ar | lol |
| 09:09.06 | cyborg_ar | just tell it which buttons to click |
| 09:15.42 | cnx | Is it possible in PCBNEW to do "incremental selection" so that I can do multiple rectangular selects which then add up to one selected group? (e.g. when you have non-rectangular areas or overlaps with areas you do not want to select) |
| 09:15.53 | xzcvczx | no |
| 09:16.18 | cnx | Is there a feature wish-list somewhere? |
| 09:16.37 | xzcvczx | there is and i think thats on there somewhere |
| 09:16.43 | xzcvczx | but its not given much effort so far |
| 09:16.53 | cnx | Is there a voting button or something? ;-) |
| 09:16.55 | xzcvczx | the bug tracker on launchpad has wishlist items |
| 09:17.02 | xzcvczx | ha no one pays attention to that |
| 09:17.09 | cnx | aye |
| 09:17.49 | cnx | What is it they pay attention to? Bug-fixing? |
| 09:18.03 | xzcvczx | sometimes |
| 09:18.09 | xzcvczx | see bug ages to see |
| 09:18.09 | cnx | understandable. |
| 09:20.03 | xzcvczx | quite often bugs are upstream issues though |
| 09:20.22 | xzcvczx | and some of those upstream bugs also have patches provided upstream but they don't care about them |
| 09:20.31 | xzcvczx | so nothing happens and the bug sits there for ages going nowhere |
| 09:25.46 | *** join/#kicad tumnao (b91f64ed@gateway/web/freenode/ip.185.31.100.237) |
| 09:30.27 | *** join/#kicad xorly (~xorly@193.85.203.185) |
| 09:36.06 | *** join/#kicad NoGodDamnIdea (~King@188.241.239.53) |
| 09:36.06 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 09:36.14 | *** join/#kicad as_g5pw (~g5pw@mittelab/sudo/g5pw) |
| 09:40.06 | *** join/#kicad sandeepkr_ (~sandeepkr@103.49.155.114) |
| 09:53.07 | cnx | Is there a way to also display reference lables when moving footprints in PCBNEW? Its always annoying to move groups of footprints and then discover an overlap of the lables. |
| 09:57.41 | cirilo | cyborg_ar, other MCADs don't support zipped STEP files so I'm not so sure. + how do I know it's really a STEP file without partial decoding to check the format. |
| 09:58.42 | cirilo | cyborg_ar, on Linux you might be able to have a driver to provide a virtual filesystem where files are magically zipped/unzipped .. though this might be slow since STEP files are monstrous to begin with. |
| 09:59.55 | *** join/#kicad kuldeep (~kuldeep@unaffiliated/kuldeepdhaka) |
| 10:03.36 | xzcvczx | cirilo: kicad seems faster than freecad at least |
| 10:09.18 | cirilo | hehe - but for how long? :) |
| 10:09.30 | cirilo | 'k - time for the computers to go to sleep - see you guys tomorrow |
| 10:10.58 | cyborg_ar | yay |
| 10:11.07 | cyborg_ar | i finished my horrifying eeschema plotting script |
| 10:12.08 | cyborg_ar | xzcvczx, https://gist.github.com/iromero91/1f008c9c2b5571cfba2396baccaab9da |
| 10:12.33 | xzcvczx | lol |
| 10:12.49 | xzcvczx | i expect that sorta shit from c4..... i thought better of you |
| 10:13.09 | cyborg_ar | it works great though |
| 10:13.23 | cyborg_ar | gotta make another to generate the netlist for the bom tool |
| 10:13.44 | cyborg_ar | and then another for pcbnew and i'll have a fully automated fabrication folder generator |
| 10:13.47 | xzcvczx | so did big boy and little boy.... didn't mean they were a good idea |
| 10:15.00 | cyborg_ar | nah |
| 10:15.10 | cyborg_ar | this is more of a sarin kinda thing |
| 10:15.23 | xzcvczx | lol |
| 10:24.12 | *** join/#kicad sandeepkr__ (~sandeepkr@111.235.65.5) |
| 10:24.52 | *** join/#kicad sandeepkr_ (~sandeepkr@103.49.155.114) |
| 10:38.34 | *** join/#kicad cosmobird (~firebird@125.16.18.198) |
| 10:56.23 | *** join/#kicad stelarcf (~stelarcf@92.81.115.213) |
| 11:02.03 | xzcvczx | c4757p: so boards arrived yet? |
| 11:02.45 | *** join/#kicad SchrodingersGat (~Oliver@124-168-227-131.dyn.iinet.net.au) |
| 11:13.12 | *** join/#kicad orsonmmz (~orson@pcbe15262.cern.ch) |
| 11:14.22 | orsonmmz | nickoe: fyi, fp editor does not start centered since a long time |
| 11:14.58 | orsonmmz | c4757p: ping |
| 11:15.49 | *** join/#kicad mroszko (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 11:21.24 | xzcvczx | lol everyone wants c4 today |
| 11:21.58 | mroszko | ? |
| 11:22.01 | mroszko | what do people want to blow up? |
| 11:23.22 | xzcvczx | c4757's malfunctioning alarm clock maybe |
| 11:29.42 | *** join/#kicad rdmitry (~Dmitry@95-37-249-206.dynamic.mts-nn.ru) |
| 11:31.17 | *** join/#kicad sandeepkr__ (~sandeepkr@111.235.65.5) |
| 11:37.34 | *** join/#kicad ayjay_t (~ajp_@24-151-28-56.dhcp.nwtn.ct.charter.com) |
| 11:49.46 | *** join/#kicad Shuggsy_ (~Shuggsy@2600:1005:b000:fdf3:5340:b51b:27e8:16f) |
| 11:51.55 | *** join/#kicad berjan (~chatzilla@2001:67c:2564:a156:4532:bf97:af39:f1b5) |
| 12:18.14 | *** join/#kicad kuldeep (~kuldeep@unaffiliated/kuldeepdhaka) |
| 12:20.40 | *** join/#kicad Sujith (~Sujith@117.231.245.99) |
| 12:24.41 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 13:02.15 | *** join/#kicad decimad (~deci@ip5f5be505.dynamic.kabel-deutschland.de) |
| 13:04.52 | *** join/#kicad alphaferret (~user@54034F08.catv.pool.telekom.hu) |
| 13:25.11 | *** join/#kicad kraiskil (~kraiskil@85-76-73-49-nat.elisa-mobile.fi) |
| 13:25.51 | *** join/#kicad Kevlan (~Kevlan@c-50-134-227-152.hsd1.co.comcast.net) |
| 13:26.05 | *** join/#kicad N-i-c-o (~quassel@log77-7-88-185-146-198.fbx.proxad.net) |
| 13:37.18 | *** join/#kicad sandeepkr (~sandeepkr@111.235.65.5) |
| 13:51.29 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 13:51.52 | *** join/#kicad shaun413 (uid121475@gateway/web/irccloud.com/x-yywgyggmjccvzouj) |
| 13:52.33 | *** join/#kicad Kevlan (~Kevlan@2607:fb90:6c25:1f74:0:3f:91d3:b501) |
| 13:56.27 | *** join/#kicad Thor-Arne (~email@34.79-161-66.customer.lyse.net) |
| 13:58.46 | corecode | hi |
| 13:59.15 | corecode | i'm trying to import an svg image into a the silkscreen layer, but this seems to be close to impossible |
| 13:59.42 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 13:59.53 | *** join/#kicad twl (808d235d@gateway/web/freenode/ip.128.141.35.93) |
| 14:00.22 | c4757p | xzcvczx: my boards came! |
| 14:00.43 | *** join/#kicad cyborg_ar (~cyborg_ar@c-98-215-226-11.hsd1.il.comcast.net) |
| 14:00.45 | c4757p | corecode: it's a bit roundabout, but I'd render it to png first |
| 14:03.13 | alphaferret | buys c3p0 a beer |
| 14:03.36 | alphaferret | and a hooker |
| 14:03.37 | c4757p | yay! |
| 14:03.44 | c4757p | double yay! |
| 14:04.12 | alphaferret | congratulations man that's a big accomplishment |
| 14:05.05 | *** join/#kicad niki (~niki@0x3e2c87f8.mobile.telia.dk) |
| 14:05.14 | decimad | it seems as if intel hd completely ignores multisample requests in opengl |
| 14:05.40 | corecode | ah |
| 14:05.42 | corecode | svg2mod |
| 14:05.47 | corecode | with precision set to 0.1 |
| 14:05.54 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 14:06.21 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 14:07.38 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 14:08.21 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 14:09.18 | orsonmmz | c4757p: do you speak about your motherboard? |
| 14:10.06 | c4757p | yup |
| 14:10.58 | orsonmmz | congrats! |
| 14:11.26 | c4757p | https://misc.c4757p.com/mobo-pcb.jpg |
| 14:11.53 | orsonmmz | looks cool |
| 14:11.59 | orsonmmz | are you going to assemble it by hand |
| 14:12.00 | orsonmmz | ? |
| 14:12.25 | c4757p | yeah |
| 14:15.27 | alphaferret | how are you going to to those bga's |
| 14:16.13 | c4757p | stencil, paste, oven |
| 14:16.47 | c4757p | y'all should see the stencil. I accidentally ordered a full-size framed one like you'd put in a machine (for some reason it didn't cost much extra) |
| 14:16.49 | c4757p | shit's huge, yo |
| 14:16.52 | c4757p | https://misc.c4757p.com/mobo-stencil.jpg |
| 14:18.00 | *** join/#kicad stelarcf (~stelarcf@exenir.com) |
| 14:21.02 | twl | cool pcb ;-) |
| 14:21.25 | c4757p | :) |
| 14:23.12 | Sujith | c4757p: Nice pcb.. |
| 14:23.30 | Sujith | how long it took to develop in kicad? |
| 14:23.37 | c4757p | been working on it since June |
| 14:23.52 | c4757p | not non-stop of course, that's a hobby thing, not for work |
| 14:24.04 | c4757p | considering days I did nothing, maybe on average ~2h a day? |
| 14:24.07 | Sujith | great effort.. |
| 14:24.10 | c4757p | thanks :) |
| 14:24.20 | Sujith | omg...hobby itself like this? |
| 14:24.40 | Sujith | no words.. |
| 14:24.54 | Sujith | btw what is the application of it? |
| 14:25.05 | c4757p | if you want me to suggest some words, "haha asshole, now you have to /assemble/ it" would be good |
| 14:25.12 | c4757p | https://semianalog.com/c4_0/ |
| 14:25.36 | Sujith | ha ha...no i wont :-) |
| 14:25.53 | xl0 | So, say I've got 20 identical components and I want to change a bunch of fields. |
| 14:25.54 | *** join/#kicad jstein_ (~quassel@gentoo/developer/jstein) |
| 14:26.02 | c4757p | ugh |
| 14:26.07 | c4757p | vim, sed, or a quick Python script. |
| 14:26.08 | xl0 | On all of them, the same data. |
| 14:26.25 | c4757p | kicad has no batch edit :( |
| 14:26.33 | xl0 | https://www.youtube.com/watch?v=07So_lJQyqw |
| 14:29.05 | c4757p | yuuup |
| 14:29.21 | Sujith | so..its a pc using full of fpgas..? and what os it will run? |
| 14:29.30 | c4757p | leenucks |
| 14:31.01 | alphaferret | he copped out in not writing his own o/s, lazy b-stard |
| 14:31.20 | c4757p | yup, I am a lazy bastard |
| 14:31.51 | c4757p | I'd much rather have a working system with drivers and software :) |
| 14:33.15 | RaYmAn | so MicroBlaze or similar then? =P |
| 14:33.35 | c4757p | no |
| 14:33.51 | c4757p | the board isn't the project, it's the yak I had to shave. designing the CPU etc myself is the real project :) |
| 14:33.59 | RaYmAn | ah |
| 14:35.36 | Sujith | I found c4757p is the kicad dev Chris Pavlina... |
| 14:35.45 | c4757p | yes indeed |
| 14:35.49 | Sujith | there are lot of pcs from RPi to so many |
| 14:35.58 | Sujith | whats special in this actually? |
| 14:36.20 | nickoe | c4757p: ohh no, you chopse the wrong silk color compared to your 3D render. I am dissapointed! :D |
| 14:36.21 | c4757p | that I get to design it from scratch. I /want/ to ;) |
| 14:36.50 | corecode | c4757p: oh nice |
| 14:37.11 | c4757p | besides, I get enough ARM during the day at work >_> |
| 14:37.31 | corecode | c4757p: how many layers? |
| 14:37.34 | c4757p | 6 |
| 14:37.42 | corecode | where is RAM? |
| 14:38.04 | c4757p | it goes in the thing that looks suspiciously like a SODIMM slot ;) |
| 14:38.12 | corecode | ah, sodimm |
| 14:38.18 | nickoe | orsonmmz: hmm, yeah about the fp editor, but why? |
| 14:38.40 | corecode | c4757p: does it work? |
| 14:38.59 | c4757p | work? I just got the unpopulated PCBs half an hour ago |
| 14:39.27 | Sujith | which supplier? |
| 14:39.40 | corecode | ah |
| 14:39.41 | corecode | heh |
| 14:39.48 | c4757p | Sujith: *shudder* pcbway |
| 14:40.00 | nickoe | Hmm, xzcvczx what descriptions are we talking about? |
| 14:40.27 | *** join/#kicad kraiskil (~kraiskil@85-76-73-49-nat.elisa-mobile.fi) |
| 14:40.43 | orsonmmz | nickoe: you mean why it is not centered? |
| 14:40.49 | corecode | c4757p: you don't need series resistors for the sodimm? |
| 14:41.03 | nickoe | orsonmmz: yes |
| 14:41.09 | orsonmmz | nickoe: sincerely, no idea :P |
| 14:41.18 | c4757p | shouldn't. I'm copying a known working design, and the original just used the FPGA internal ones |
| 14:41.23 | orsonmmz | nickoe: but I think I have fixed the qa problem |
| 14:41.41 | corecode | c4757p: ah! |
| 14:41.44 | c4757p | and DDR3 has the terminations at the other end on-board |
| 14:41.50 | nickoe | orsonmmz: ahh, great :) |
| 14:41.54 | c4757p | or, on-chip |
| 14:42.08 | corecode | c4757p: that's where i got stuck when i wanted to design a high speed circuit |
| 14:42.20 | c4757p | but yeah - I'm not adventurous enough to design my own DDR3 SODIMM interface from scratch |
| 14:42.23 | corecode | c4757p: went down the IBIS simulation rathole |
| 14:44.05 | nickoe | orsonmmz: we will se in a moment |
| 14:45.38 | orsonmmz | stares nervously at jenkins |
| 14:46.31 | c4757p | wise. don't get too comfortable with Jenkins, that's how it'll get you |
| 14:46.59 | decimad | That jenkins guy is a friend of Chuck Norris |
| 14:47.01 | GyrosGeier | guys |
| 14:47.03 | GyrosGeier | http://ci.kicad-pcb.org/job/windows-kicad-msvc-head/1216/warnings39Result/new/ |
| 14:47.04 | nickoe | orsonmmz: yay, it is fixed. See the kicad-qa build |
| 14:47.15 | GyrosGeier | http://ci.kicad-pcb.org/job/windows-kicad-msvc-head/1216/changes |
| 14:47.27 | GyrosGeier | there are some objects that are now linked twice |
| 14:47.47 | nickoe | GyrosGeier: maybe poke orsonmmz |
| 14:47.57 | c4757p | I'm glad you said that. none of what you just linked means anything to me :P |
| 14:48.12 | GyrosGeier | c4757p, ~200 new warnings |
| 14:48.24 | nickoe | 188 |
| 14:48.27 | GyrosGeier | all of them of the form "*@3$&@(!8@%$&#*&@(% already defined" |
| 14:48.36 | *** join/#kicad firebird (~firebird@125.16.18.198) |
| 14:48.37 | c4757p | nickoe: how's it coming getting builds going on my Mini? |
| 14:48.53 | orsonmmz | GyrosGeier: emm, is LNK4006 the whole error message msvc output? |
| 14:49.14 | orsonmmz | ah, details! |
| 14:49.16 | orsonmmz | there you go.. |
| 14:49.31 | orsonmmz | this is madness.. |
| 14:50.07 | GyrosGeier | it's a linker error |
| 14:50.07 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 14:50.12 | decimad | how does that happen really? is the same cpp source file part of multiple targets? |
| 14:50.16 | GyrosGeier | yes |
| 14:50.44 | c4757p | bleh, pcbway messed with both my mask and paste layers |
| 14:50.47 | GyrosGeier | gcc will not notice if there are static libraries involved |
| 14:51.08 | decimad | well then that should be easy to fix, right?^^ |
| 14:51.18 | nickoe | c4757p: :O in what way? |
| 14:51.21 | GyrosGeier | needs a good manufacturing service that does plugged vias |
| 14:51.30 | c4757p | mask is okay, I guess - looks like this six-layer board was done on a higher-end process than what they quote specs for, so they tightened up the mask expansions a bit |
| 14:51.36 | GyrosGeier | I have a 5x5 BGA with 0,4mm raster |
| 14:51.43 | GyrosGeier | and two layers |
| 14:51.56 | c4757p | they're better that way, but they should have just bloody been honest about their capabilities in the first place and I'd have sent it to them like that |
| 14:52.09 | corecode | GyrosGeier: is that cheaper than using 4 layers? |
| 14:52.12 | c4757p | they widened the paste layer, I'm not too pleased with that. I wanted less paste on some of those pads |
| 14:52.25 | GyrosGeier | corecode, I'd still need plugged vias then |
| 14:52.39 | GyrosGeier | or very small microvias |
| 14:52.44 | corecode | GyrosGeier: hm yes |
| 14:52.52 | corecode | c4757p: it seems that paste layers are never used for stencils |
| 14:52.57 | decimad | I wonder if in a hundred years we'll have gerber files that take gigabytes to circumvent the formats huge limitations |
| 14:52.59 | GyrosGeier | I'm not sure I should trust dirtypcbs to get it right |
| 14:53.01 | c4757p | corecode: oh? |
| 14:53.04 | corecode | c4757p: they just use the copper pads and resize them |
| 14:53.11 | corecode | c4757p: in my limited experience |
| 14:53.12 | c4757p | ooh, it does look like my Mini is doing things |
| 14:53.22 | c4757p | the build log didn't load at first :) |
| 14:53.40 | c4757p | corecode: huh. does the Gerber format actually say what is and isn't a pad? I thought it'd be rather hard to identify just from the artwork |
| 14:53.54 | corecode | because i've got stencils back where "pads" i didn't put on the paste layer were present |
| 14:53.57 | c4757p | corecode: also, I've done funny paste patterns before and the stencils came out right |
| 14:54.00 | c4757p | so... |
| 14:54.19 | corecode | c4757p: might depend on the fab tho |
| 14:54.24 | c4757p | actually, yes |
| 14:54.34 | c4757p | the windowpaned paste areas are done correctly |
| 14:54.37 | c4757p | so they did use the paste layer |
| 14:54.59 | c4757p | they just applied an expansion to it, god knows why |
| 14:55.50 | corecode | what design did you base this off? |
| 14:56.33 | c4757p | well, "this" = just the DDR3 controller, that's all I 'borrowed' |
| 14:56.35 | c4757p | lemme find it |
| 14:57.43 | c4757p | hm, their website's screwed up now |
| 14:57.49 | c4757p | here's the opencores page http://opencores.org/project,ddr3_sdram |
| 14:58.07 | c4757p | they sell the hardware, but their website's just farting at me now |
| 14:58.33 | c4757p | there we go http://www.sioi.com.au/shop/product_info.php/products_id/57 |
| 15:00.21 | c4757p | ahh, I see, link's broken because they upgraded the FPGA on the board and gave it a new part number |
| 15:01.12 | c4757p | shame they never ported it to 7-series, I wanted to use those :( |
| 15:02.01 | decimad | https://twitter.com/whyaltiumsucks |
| 15:02.17 | c4757p | lol |
| 15:02.34 | decimad | i searched for "Gerber format sucks" :) |
| 15:02.42 | decimad | I like such search queries a lot |
| 15:02.54 | corecode | so the length matching feature works well now? |
| 15:03.02 | c4757p | mostly |
| 15:03.06 | c4757p | it still doesn't account for vias :( |
| 15:03.07 | *** join/#kicad riq_ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 15:03.33 | corecode | c4757p: what about bus topologies? |
| 15:03.38 | c4757p | (though that's a can of worms...I just made sure matched groups had a matched number of vias, which is really the best I could do) |
| 15:03.59 | c4757p | (which really comes naturally as they're on the same layer too :P ) |
| 15:04.31 | c4757p | it doesn't understand bus topologies in any meaningful way, it just goes point to point |
| 15:04.48 | nickoe | Also, we need a proper stackup tp handæe length matching wit vias properly |
| 15:05.09 | c4757p | we need a lot more than that ;) |
| 15:05.12 | c4757p | like I said...can of worms |
| 15:05.16 | nickoe | Yahh! |
| 15:06.11 | c4757p | the impedance going through the via may not be the same as for the trace, so it doesn't necessarily count with the same ratio of length to delay |
| 15:06.36 | *** join/#kicad nitrousnrg (~quassel@190-103-222-12.cepanet.com.ar) |
| 15:06.44 | *** join/#kicad pointhi (~pointhi@heim-032-115.raab-heim.uni-linz.ac.at) |
| 15:06.45 | c4757p | just adding it to the trace length would be a better approximation than making it /zero/, though |
| 15:06.48 | c4757p | so. |
| 15:07.11 | nickoe | Apropos, I wonder what the status is about azonenbergs skew coloring |
| 15:07.18 | c4757p | ooh, I forgot about that |
| 15:08.28 | corecode | well, if you length match per layer, you're good |
| 15:08.35 | corecode | which you should do anyways |
| 15:08.37 | c4757p | yeah |
| 15:08.59 | c4757p | wonder how easy it would be to make the length matcher stop at a via, so you can just do the segments between them |
| 15:09.30 | corecode | probably not too hard? |
| 15:09.53 | c4757p | yeah, but I'm always surprised on this project by the sort of "should be easy" things that end up being quite difficult ;) |
| 15:10.48 | decimad | twl : I noticed cursor hoppen when in router mode, grid snapping is active and you're close to a segment. Is that behaviour intentional to hit the grid line that is closest to the cursor? |
| 15:11.18 | decimad | hoppen -> hopping |
| 15:11.56 | twl | decimad: yes |
| 15:12.27 | twl | it should align to grid, but if not possible to align exactly to a 'round' grid point it picks the nearest aligned axis |
| 15:12.34 | twl | someone asked for it a long time ago, so I did it |
| 15:13.10 | decimad | twl: Okay, I prepared a patch that would align to the closest grid line of the nearest point on the segment, that will probably be bad then |
| 15:13.36 | twl | yeah, that's how it should work |
| 15:13.48 | twl | it it broken now? |
| 15:14.15 | decimad | It takes the grid line that is closest to the cursor in my master |
| 15:15.16 | decimad | So when you move around the cursor it the cross will jump around wildly if you are not very close to the segment center line |
| 15:19.11 | *** join/#kicad tom_fig (42bce4eb@gateway/web/freenode/ip.66.188.228.235) |
| 15:22.14 | corecode | china has some IC marketplace website, but i keep forgetting their domain. some number, i think |
| 15:22.51 | *** join/#kicad rdmitry (~Dmitry@95-37-249-206.dynamic.mts-nn.ru) |
| 15:23.59 | twl | which commit? |
| 15:24.04 | *** join/#kicad birukun (~b@cpe-76-176-69-151.san.res.rr.com) |
| 15:24.31 | decimad | twl: hrm? my master head? |
| 15:25.07 | decimad | twl: it's 104a066268f4210e2cde6464d19348bb73f2807b |
| 15:25.55 | corecode | 1688.com! |
| 15:26.01 | *** join/#kicad dumle29 (~quassel@64.137.168.54) |
| 15:31.06 | *** join/#kicad k\o\w (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 15:32.44 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 15:36.22 | xl0 | corecode: www.szlcsc.com - closest you get to digikey. Which is sad. Or hqew.com for checking stock. |
| 15:36.35 | xl0 | 1688 is alibaba for inside China. |
| 15:37.04 | c4757p | mh |
| 15:37.07 | c4757p | hm |
| 15:37.16 | c4757p | as a programmer it just feels /weird/ for a name to start with a digit |
| 15:38.33 | decimad | yeah, they should be trailing and ideally always be 24 ^^ |
| 15:39.09 | xl0 | c4757p: Some people are not comfortable dealing with latin letters. |
| 15:39.45 | c4757p | I wasn't being serious :P |
| 15:43.00 | *** join/#kicad berjan (~chatzilla@ip54538619.speed.planet.nl) |
| 15:46.26 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 15:47.13 | *** join/#kicad alphaferret (~user@54034F08.catv.pool.telekom.hu) |
| 15:48.03 | GyrosGeier | if(${name}_LIBRARIES) |
| 15:48.14 | GyrosGeier | can we please set cmake on fire and start from scratch? |
| 15:48.34 | c4757p | I don't really mind cmake |
| 15:48.37 | c4757p | it's got its problems |
| 15:48.56 | c4757p | I ditched it for my EC firmware, too much of a pain to set up with a mixed-language environment... |
| 15:49.09 | c4757p | it does a fine job with big PC applications though :) |
| 15:52.24 | GyrosGeier | stat("/usr/arm-linux-gnueabihf/usr/local/lib/arm-linux-gnueabihf", 0x7fffaf5b89f0) = -1 ENOENT (No such file or directory) |
| 15:52.53 | GyrosGeier | I don't really want to understand why it does this |
| 15:52.59 | GyrosGeier | but my customer does |
| 16:03.40 | *** join/#kicad Eli2_ (~Eli2@pdpc/supporter/professional/eli2) |
| 16:04.47 | *** join/#kicad phinxy (~tehhhd@unaffiliated/phinxy) |
| 16:09.36 | GyrosGeier | hmmk |
| 16:09.38 | GyrosGeier | now it works |
| 16:09.57 | GyrosGeier | http://images-cdn.9gag.com/photo/anYZ9Eo_700b.jpg |
| 16:10.22 | c4757p | :) |
| 16:14.10 | dumle29 | just got myself an MX master |
| 16:14.32 | dumle29 | I've noticed that in kicad, with the thumbscroll wheel, the scrool bar at the bottom moves, but the window ignores it |
| 16:14.49 | tom_fig | That's a long-standing bug |
| 16:15.12 | dumle29 | ah, ok. I'd imagine it isn't high priority ;) |
| 16:15.37 | dumle29 | Ok that's extra weird. Now it just works like the normal whell |
| 16:15.53 | dumle29 | That has to be on my mouse though |
| 16:16.07 | dumle29 | wat, now it scrolls sideways again |
| 16:17.00 | c4757p | https://kicad.c4757p.com |
| 16:17.06 | dumle29 | xD |
| 16:17.14 | dumle29 | I love that you have an image for it xD |
| 16:17.38 | dumle29 | oh well. Small issue. I usually navigate with the main wheel anyways |
| 16:17.39 | c4757p | my digital logic prof had one for Xilinx :D |
| 16:17.48 | dumle29 | "Zoom out, zoom in" |
| 16:17.52 | dumle29 | :P |
| 16:18.06 | decimad | I don't consider my code working when in the second state... ;) |
| 16:18.48 | dumle29 | that said. The MX master is awesome :3 |
| 16:19.29 | c4757p | it looks silly |
| 16:19.34 | c4757p | says the guy who uses a trackball |
| 16:19.40 | dumle29 | xD |
| 16:19.51 | dumle29 | that thumb rest is super comfortable |
| 16:20.04 | dumle29 | and theres a button under the thumb, that'll open the exploded window view in windows |
| 16:20.11 | dumle29 | basically the default menu in gnome 3 |
| 16:20.51 | decimad | there are logitechs with thumbrest, that button and without trackball ;) |
| 16:22.15 | decimad | I like the performance mx quite a lot |
| 16:24.19 | *** join/#kicad drod (~ldm@ip-95-221-235-169.bb.netbynet.ru) |
| 16:26.19 | *** join/#kicad thinkl33t (~thinkl33t@thinkl33t.co.uk) |
| 16:27.00 | thinkl33t | OK, so i'm pretty sure I must be being stupid here, but how do you create a new component part library? |
| 16:27.31 | c4757p | create the component, then use the "save in new library" button |
| 16:27.34 | thinkl33t | I can add components to existing libraries fine, but having real trouble actually creating a new empty one |
| 16:27.54 | c4757p | I don't think there is a way to directly createe a new empty library |
| 16:27.57 | c4757p | -e |
| 16:28.12 | thinkl33t | Weird. OK. thanks :) |
| 16:28.15 | nickoe | c4757p: did you do something to your mini just now? |
| 16:28.35 | nickoe | ssh seems very slow |
| 16:29.08 | thinkl33t | Will i get my head bitten off if i put it in a feature request? :) |
| 16:29.30 | nickoe | hmm, c4757p seems like it was solely because it was building with adams scripts... |
| 16:29.39 | c4757p | <PROTECTED> |
| 16:29.57 | nickoe | hmm, yeah, I just cancled the build and it is still a bit .. |
| 16:30.00 | c4757p | are you transferring a large file to/from the mini? |
| 16:30.12 | nickoe | no |
| 16:30.19 | nickoe | just a ssh sesstion with tmux in it |
| 16:30.21 | c4757p | huh |
| 16:30.39 | c4757p | nothing's using heavy bandwidth that I know of. wonder what's going on |
| 16:30.40 | nickoe | sometimes it is responsive |
| 16:30.58 | c4757p | it's rather irritating, this is messing up my IRC too :P |
| 16:31.01 | nickoe | some dwarfs is throwing away packets maybe? |
| 16:31.08 | c4757p | lol |
| 16:32.13 | c4757p | if it's still acting up when I get home later, I'll have a look at it |
| 16:33.32 | nickoe | ok |
| 16:34.51 | c4757p | huh, avg bandwidth out eth0 is like 10kB/s |
| 16:35.09 | c4757p | throughput* |
| 16:35.11 | c4757p | fuck knows what's going on |
| 16:35.29 | c4757p | I've got neighbors moving in this week, maybe the ISP is over fucking around with the cables trying to get them connected |
| 16:37.23 | c4757p | interesting https://misc.c4757p.com/throughput-holes.png |
| 16:40.30 | nickoe | Does this look more right? https://github.com/KiCad/kicad-website/pull/173/commits/e0164c6cd9e30687223bdfd7978dcbba0f81068f |
| 16:41.44 | c4757p | yes. merge it |
| 16:44.12 | nickoe | ok |
| 16:47.50 | *** join/#kicad awolf3 (4925e60a@gateway/web/freenode/ip.73.37.230.10) |
| 16:47.55 | awolf3 | nickoe: are you around? |
| 16:48.19 | c4757p | points at nickoe, hiding behind the sofa |
| 16:48.48 | awolf3 | ha |
| 16:49.11 | awolf3 | literally 1 hour after I start rewriting the osx scripts, nick sends me an email saying "I'm getting this set up on our official jenkins" |
| 16:50.01 | c4757p | I assume he means jenkins on the Mini I "donated" |
| 16:51.06 | awolf3 | so |
| 16:51.08 | awolf3 | there are 2 pieces of work here |
| 16:51.11 | awolf3 | 1) rewriting the scripts |
| 16:51.23 | awolf3 | 2) rebuilding the deps because of the 10.7/10.9 change |
| 16:52.38 | nickoe | awolf3: hi |
| 16:52.42 | awolf3 | I had a big long email i was wokring on, when I decided to just jump in here |
| 16:52.56 | awolf3 | I have a meeting in 8 minutes |
| 16:53.01 | nickoe | awolf3: no, I mean I am playing with it :) |
| 16:53.04 | awolf3 | ok |
| 16:53.10 | awolf3 | it would be awesome to have another person using the official builder |
| 16:53.16 | *** join/#kicad Laksen (~Laksen@unaffiliated/laksen) |
| 16:53.17 | awolf3 | then we can actually make changes on the builder without breaking everyones stuff |
| 16:53.28 | nickoe | sure, we can do that on c4757p's host |
| 16:53.32 | awolf3 | a+++++++ |
| 16:53.39 | nickoe | awolf3: but do we have a 10.9 to test on? |
| 16:53.44 | awolf3 | I have a meeting, then I'll send this email to you with details |
| 16:53.46 | awolf3 | What do you mean? |
| 16:53.53 | nickoe | if that is our minver target |
| 16:53.54 | weatherhead | hi all, there are a few things I was thinking of working on in eeschema, but it seems it's undergoing major re-factoring churn at the moment |
| 16:54.02 | awolf3 | I could make one |
| 16:54.05 | awolf3 | bbl, sorry :( |
| 16:54.13 | *** join/#kicad skoocda (~skoocda@76.74.139.126) |
| 16:54.14 | nickoe | awolf3: xzcvczx says that we can build to it, but it might not work :O |
| 16:54.31 | tom_fig | I feel the same, weatherhead |
| 16:54.45 | nickoe | awolf3: I am not sure I understand the exact issue witht he 10.7/10.9 deps. And which deps are affected. |
| 16:54.46 | c4757p | weatherhead: some of it is more a seemingly permanent state of "I'm about to refactor this, seriously, I am"! |
| 16:54.53 | c4757p | do you plan to touch file IO? |
| 16:54.55 | awolf3 | nickoe: I will handle that |
| 16:55.14 | awolf3 | I will keep doing what I was doing, with new excitement because there's another builder we can install it on :) |
| 16:55.17 | c4757p | that's the only part currently undergoing serious refactoring. |
| 16:55.49 | nickoe | isn't it possible to have multiple xcode SDK's coexisting? |
| 16:56.06 | nickoe | weatherhead: what things do you want to do? |
| 16:56.23 | weatherhead | c4757p, I have a big wishlist which I'm willing to devote some time to trying to implement |
| 16:56.36 | tom_fig | Python scripting in eeschema? |
| 16:56.37 | nickoe | weatherhead: What are the top two items on that list? |
| 16:56.43 | weatherhead | first, I'd like some kind of bus-splicing, which allows to splice busses with different numbers together |
| 16:56.43 | c4757p | I'd love to see some improvements come to eeschema. I was working on that and now I have no more time. |
| 16:56.56 | nickoe | tom_fig: someone linked a repo where dick made a few commits to enable scripting in eeschema. |
| 16:56.59 | c4757p | weatherhead: yessss |
| 16:57.05 | tom_fig | Yeah, I linked it lol |
| 16:57.11 | weatherhead | e.g. I have a bus in a hierarchical sheet labelled A[0..3], and would like to splice into a buss on the top sheet labelled B[4..7] |
| 16:57.20 | weatherhead | without doing another sheet specifically to splice the buss |
| 16:57.25 | nickoe | tom_fig: :) |
| 16:57.26 | tom_fig | So, sticking with SWIG? |
| 16:57.30 | c4757p | weatherhead: yes yes yes |
| 16:57.31 | weatherhead | second, I'd like support for altium-like "harnesses" |
| 16:57.36 | c4757p | yessssssssssssssssss harnesses |
| 16:57.39 | c4757p | hugs weatherhead |
| 16:57.53 | weatherhead | I don't think either of these things are particularly difficult but I don't know the code well enough to know whether it's worth waiting before trying to start on this |
| 16:58.01 | c4757p | if you seriously want to do the work, make a detailed proposal on the mailing list |
| 16:58.14 | c4757p | this is going to touch the new file format work, so get it in quickly |
| 16:58.23 | weatherhead | lack of both of these have hit me several times in last few projects |
| 16:58.33 | weatherhead | thirdly, I'd like to be able to annotate net types in schematic |
| 16:58.36 | c4757p | I'll support you 100% :) |
| 16:58.38 | weatherhead | e.g. annotate nets as a diff pair |
| 16:58.47 | c4757p | that's already planned |
| 16:58.48 | weatherhead | ok |
| 16:58.58 | c4757p | /fuck/ I want harnesses |
| 16:59.01 | nickoe | There was some discussion about that on the list stated by Tomasz I think |
| 16:59.14 | nickoe | you should probably read that. |
| 16:59.16 | weatherhead | fourthly I'd like proper back annotation, i.e. annotate by board position. But I think that's beyond my level of commitment and knowledge of the code base atm |
| 16:59.36 | c4757p | weatherhead: I'm somewhat familiar with that code, though it's been a while |
| 16:59.44 | c4757p | it won't be /trivial/, but if you're a decent programmer it shouldn't be too hard |
| 17:00.09 | weatherhead | c4757p, I have ~10 years of c++ experience, but in physics rather than as a jobbing programmer, so I'm still pretty bad ;-) |
| 17:00.11 | c4757p | look at eeschema/netlist.cpp, I think most of your work (except UI) will be in there |
| 17:00.53 | weatherhead | c4757p, the problem with e.g. harnesses is that it can be done in quite a few mutually exclusive ways that some people might argue about |
| 17:01.00 | c4757p | you'll have to create a new object type to represent the harness, but that's mostly just graphical - then netlist.cpp will have to track them down and figure out what connects to what |
| 17:01.07 | c4757p | yes. I want a /red/ bikeshed! |
| 17:01.11 | weatherhead | e.g. do you need defined harness "types" |
| 17:01.24 | weatherhead | or just have a harness object that auto connects sub-wires based on name |
| 17:01.53 | weatherhead | I've seen good arguments for both, to the point that in an ideal world I'd have two separate features |
| 17:01.57 | c4757p | I'm not convinced those are mutually exclusive |
| 17:02.02 | weatherhead | but almost nobody would end up knowing the difference |
| 17:02.19 | weatherhead | kicad is bad enough at this stuff already, what with the current "switch canvas types to do things" |
| 17:02.30 | *** join/#kicad c4758p (d1d9d0f5@gateway/web/freenode/ip.209.217.208.245) |
| 17:02.44 | c4758p | fuck, my dodgy connection is driving me insane |
| 17:02.54 | weatherhead | well, should it be a DRC error to connect a harness called "I2C" to one called "SMBUS" |
| 17:03.03 | weatherhead | which contain identical wire names |
| 17:03.07 | c4758p | that's easy |
| 17:03.18 | c4758p | give them both a name and a type, and allow type to be blank |
| 17:03.32 | c4758p | if type is defined, then enforce connecting like with like |
| 17:03.35 | c4758p | otherwise, don't |
| 17:03.42 | weatherhead | OTOH if you implement it as types, then all connection lookups can be just by number rather than name |
| 17:03.42 | Yatekii | guys is the code for the footprint editor the same as for the pcb editor? |
| 17:03.50 | c4758p | Yatekii: it shares a lot, but not all |
| 17:04.03 | c4758p | the OpenGL one shares much more than legacy |
| 17:04.16 | c4758p | weatherhead: well, just omit that optimization then |
| 17:04.23 | c4758p | sure, you can do it by number, but you don't have to :P |
| 17:04.31 | Yatekii | c4757p: hmm well becaus i can only find the dir for the pcb one |
| 17:04.45 | Yatekii | but i want to define shortcuts for the footprint editor |
| 17:04.48 | c4758p | Yatekii: in same directory, because kicad is well organized like that |
| 17:04.52 | c4758p | >_> |
| 17:05.07 | Yatekii | ok i see |
| 17:05.11 | weatherhead | c4758p, what are the implications of the "new file format" |
| 17:05.15 | c4758p | look for things containing "mod" / "module" |
| 17:05.25 | c4758p | weatherhead: well, we're drafting a new sch/schlib file format |
| 17:05.27 | Yatekii | an people are like: why change the ghetto if the ghetto works atm ... |
| 17:05.33 | c4758p | it'll have to contain all these things |
| 17:05.50 | Yatekii | ok |
| 17:05.58 | Yatekii | why new file format? |
| 17:06.18 | c4758p | because the old one is poorly designed and limiting |
| 17:06.31 | *** join/#kicad rohitksingh (~rohit@119.82.107.167) |
| 17:06.56 | weatherhead | c4758p, but surely the most important thing in a new "long term" file format is to have capability of doing later versioning |
| 17:06.57 | Yatekii | hmm i heard some guys like being limitted in here ... |
| 17:07.18 | weatherhead | anyway, ok, I will try and write a proposal over the next few days |
| 17:07.36 | Yatekii | also whats really confusing: in the place menu all shortcuts are written with shift as modifier |
| 17:07.41 | c4758p | weatherhead: yes |
| 17:07.47 | Yatekii | but in reality you dont need shift |
| 17:07.55 | c4758p | but among some of the devs, there's still a lot of inertia wrt adding things |
| 17:08.01 | c4758p | it'll be way easier for you to get it in now |
| 17:08.14 | weatherhead | I mean a simple harness proposal could be as simple as allowing busses with names for members instead of numbers |
| 17:08.34 | weatherhead | but I guess we also want different pin types for each one |
| 17:10.41 | c4758p | weatherhead: permitting names is somewhat nontrivial, as buses right now basically only look at numbers |
| 17:10.51 | c4758p | ie A0 on one bus can connect to B0 on another if you join them |
| 17:11.01 | c4758p | so you'll either have to break that, or only allow either all-name or all-number |
| 17:11.08 | c4758p | I'd be more in favor of a totally new object type |
| 17:11.18 | weatherhead | c4758p, I think that is a better idea yes |
| 17:11.54 | weatherhead | c4758p, so you think it's sensible to go ahead and write proposal/start work? |
| 17:12.13 | weatherhead | Because the impression I get from some mailing list posts is that you basically shouldn't touch anything in eeschema until it's "cleaned up" |
| 17:12.18 | c4758p | yeah, if you've got the time to implement it, go ahead |
| 17:12.20 | c4758p | meh |
| 17:12.31 | c4758p | they keep saying that. they've been saying that for over a year |
| 17:12.32 | weatherhead | which as a veteran of some large projects I know will be literally never, but there are degrees of cleanliness |
| 17:13.15 | c4758p | the work is just not getting done anywhere near as quickly as it should, it shouldn't totally stall development like that. and they have been flexible, if you have a reasonable proposal and are willing to cooperate |
| 17:13.44 | weatherhead | c4758p, ok I'll start looking in detail at netlist.cpp and see if I can come up with a sensible way to do it |
| 17:14.00 | c4758p | thumbs-up |
| 17:14.21 | c4758p | only one request |
| 17:14.30 | c4758p | can you go back in time and implement it /before/ I start work on my last board? |
| 17:14.33 | c4758p | could have really used it. |
| 17:14.37 | weatherhead | haha me too |
| 17:14.59 | weatherhead | almost cracked and went Altium for the last one, but booting up windows was too much irritation in the end |
| 17:15.38 | c4758p | lol |
| 17:16.39 | weatherhead | c4758p, is c++11 usage allowed in the code yet? |
| 17:16.44 | nickoe | not to forget how long time it takes to "boot" altium |
| 17:16.52 | weatherhead | nickoe, ;-) |
| 17:16.58 | nickoe | weatherhead: yes, for stuff that is not meant to be backported to 4.0 |
| 17:17.35 | weatherhead | nickoe, also are patches acceptable for e.g. changing french variable names to english? |
| 17:17.43 | nickoe | I think so |
| 17:17.53 | c4758p | if you stay out of things being actively refactored, yeah, I'd accept them |
| 17:17.57 | weatherhead | ok, still a lot of "Objet" around in this file :-) |
| 17:18.00 | c4758p | hehe |
| 17:18.30 | c4758p | <-- has commit privileges, by the way, so I /can/ accept them ;) |
| 17:18.34 | weatherhead | c4758p, and how do I check what's being worked on, just mailing list? |
| 17:19.01 | nickoe | yeah, or the roadmap to some degree. |
| 17:19.07 | xl0 | Is there any work being done on a BOM manager? |
| 17:19.12 | c4758p | mailing list, or ask in here, or even just check the git log to see how recently it's been poked at |
| 17:19.24 | nickoe | I think Wayne has his own branch he pushes to and sometimes pushes some stuff to master |
| 17:19.36 | c4758p | I'd ignore the roadmap. lots of things it's a bit optimistic about that nobody's actually touching at the moment. |
| 17:20.04 | nickoe | true, but it is there, there is not reason it can not be kept up to date |
| 17:20.07 | xl0 | Riht now BOM maintenance is the most annoying and time wasting thing for me. |
| 17:20.15 | c4758p | nickoe: well, keep it up to date then :> |
| 17:20.28 | c4758p | xl0: not that I'm aware of. |
| 17:20.38 | nickoe | it is sort of up to date |
| 17:21.02 | nickoe | c4757p: What do you think about https://github.com/nickoe/kicad-source-mirror/commit/22fa2e4d7d38b3aa4a366ee9553dac9982c112e3 ? |
| 17:21.13 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 17:21.19 | xl0 | Does a simple table with components / fields sound like a solution? |
| 17:21.45 | nickoe | xl0: How do you maintain the BOM? Have you tried KiCost or similar? |
| 17:22.05 | c4758p | nickoe: looks like a pretty routine update, not sure what sort of opinion I'm supposed to have :P |
| 17:22.20 | xl0 | nickoe: I export it once and just use the spreadsheet. |
| 17:22.23 | nickoe | c4757p: would you merge it? |
| 17:22.34 | nickoe | and is the description about the new folder accurate enough? |
| 17:22.35 | c4758p | ofc |
| 17:22.48 | c4758p | oh, heh "Old crap" |
| 17:22.53 | c4758p | I'd...rephrase that |
| 17:22.55 | c4758p | missed that |
| 17:22.59 | nickoe | :P |
| 17:23.00 | weatherhead | c4758p, is there a place to look for "logic" of how class implementations are arranged into files |
| 17:23.10 | c4758p | weatherhead: yes |
| 17:23.18 | weatherhead | e.g. netlist.cpp has definitions for methods of about 5 different classes, all defined in different headers |
| 17:23.20 | c4758p | get really drunk, then get high on something dodgy, then try to imagine how you'd do it |
| 17:23.36 | c4758p | that should be as much of an authority on that as anyone |
| 17:23.43 | *** join/#kicad xorly (~xorly@ip-89-176-10-118.net.upcbroadband.cz) |
| 17:24.04 | nickoe | c4757p: but it has essentially not been updated in four years (except for a search and replace ifdef for wx 3.0) |
| 17:24.12 | c4758p | nods |
| 17:24.30 | c4758p | nickoe: personally I'd prod the list and see if anyone minds just deleting the 'new' folder first |
| 17:24.44 | c4758p | rather than updating the readme to describe how it's "Old crap" ;) |
| 17:25.28 | nickoe | it is very old |
| 17:25.44 | nickoe | Don't befooled by the naming of this mysterious folder, it is old! |
| 17:25.51 | c4758p | yes |
| 17:25.59 | c4758p | it's "new" in the sense of "pcbnew" |
| 17:26.04 | c4758p | i.e. not at all :) |
| 17:26.14 | weatherhead | c4758p, ok I'll get tanked up later and see what happens |
| 17:26.25 | c4758p | hehe |
| 17:29.31 | *** join/#kicad kraiskil (~kraiskil@a91-152-241-34.elisa-laajakaista.fi) |
| 17:29.57 | *** join/#kicad pointhi (~pointhi@212.33.36.212) |
| 17:30.32 | weatherhead | c4758p, so I can roughly see how I would go about this. Couple more pointers before I start hacking and try and come up with something rough |
| 17:30.45 | weatherhead | 1) re: c++11, is it ok to e.g. replace enums with scoped enums? |
| 17:31.14 | weatherhead | or are those kind of changes not allowed for compat reasons |
| 17:32.09 | c4758p | they're allowed. we've been slowly migrating things |
| 17:32.21 | *** join/#kicad rohitksingh (~rohit@119.82.107.167) |
| 17:32.32 | weatherhead | ok, I'll write a brief proposal and throw it on the list |
| 17:33.02 | c4758p | devel branch doesn't need to maintain compatibility with stable branch. the corollary is that bugfixes should be done first against stable, and ported forward |
| 17:34.26 | weatherhead | c4758p, currently my source is pulled from github.com/kicad-source-mirror, does "devel" correspond to master branch there |
| 17:35.19 | c4758p | yep |
| 17:36.05 | c4758p | no need to use the mirror anymore though, we've fully migrated to git - if you want to go straight to the source (heh) and avoid the mirror lag you can use git+ssh://git.launchpad.net/kicad |
| 17:36.20 | c4758p | (I imagine you can use https:// if you only want fetch, not sure) |
| 17:37.28 | weatherhead | c4758p, ok https works (I don't have launchpad account, and don't really need one unless it's absolutely necessary)( |
| 17:37.41 | c4758p | :) |
| 17:40.40 | nickoe | c4757p: there is only about two seconds lag between the mirror and lp |
| 17:40.47 | c4758p | yes, on a good day |
| 17:40.48 | c4758p | :) |
| 17:41.34 | nickoe | ok, ok, no more than five seconds then, http://ci.kicad-pcb.org/job/any-kicad-mirror-head/buildTimeTrend |
| 17:41.59 | c4758p | fair enough :P |
| 17:42.08 | c4758p | it still seems rather unnecessary to me |
| 17:42.19 | c4758p | nickoe: you're not using the Mini right now, I assume, with the connection this dodgy? |
| 17:42.34 | nickoe | c4757p: I am building stuff on it, but I can try to connect again |
| 17:42.38 | c4758p | please don't |
| 17:42.43 | c4758p | :P |
| 17:42.46 | nickoe | was already connected. I can disconnect then. |
| 17:42.54 | nickoe | bit it is at 100% !!! |
| 17:43.01 | c4758p | will that abort the build? |
| 17:43.05 | nickoe | no |
| 17:43.08 | nickoe | I am in a tmux |
| 17:43.13 | c4758p | okay. I want to sacrifice a chicken and reboot the router |
| 17:43.16 | c4758p | see if that improves matters |
| 17:43.19 | c4758p | okay to do that now? |
| 17:43.31 | nickoe | woops, I git ctrl+c... haha |
| 17:43.48 | nickoe | I am disconnected now |
| 17:43.56 | c4758p | yes, you are >:) |
| 17:43.59 | c4758p | reboots router |
| 17:44.13 | nickoe | c4757p we will miss you! |
| 17:44.26 | c4758p | I highly doubt the problem is actually the router, but it's pissing me off and that's the best I can do from 20 miles away |
| 17:44.41 | nickoe | What router do you have? |
| 17:44.50 | c4758p | custom router in a Debian box |
| 17:45.30 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 17:47.01 | c4758p | mm, it's taking its time coming back up |
| 17:47.44 | nickoe | You break it, you buy it! |
| 17:48.00 | c4758p | I wonder whether that's "ISP's shat itself" or just "been a long time, fsck forced" |
| 17:48.01 | decimad | what does if( future ) mean in cmake= |
| 17:48.23 | nickoe | decimad: is that how it is written literally? |
| 17:48.29 | decimad | yes |
| 17:48.32 | nickoe | decimad: Where exactly do you see that? |
| 17:48.45 | decimad | cmakelists.txt in common |
| 17:48.55 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 17:49.28 | nickoe | decimad: look at 2c67c3ff |
| 17:49.55 | c4758p | there we go, it's 1) back up, and 2) still dodgy |
| 17:49.59 | c4758p | grumble. |
| 17:50.43 | nickoe | c4757p: dammit, you stressed me with that disconnect order |
| 17:50.52 | c4757p | o_O |
| 17:51.03 | nickoe | not it is rebuilding from scratch, because I guess awolf3's script builds a clean build every time. |
| 17:51.06 | nickoe | *now |
| 17:51.08 | nickoe | :)I |
| 17:51.47 | c4757p | mm, it's /less/ dodgy though |
| 17:51.50 | c4757p | ...oh wait, no it isn't |
| 17:52.00 | nickoe | decimad: I guess it is essently just for "development" |
| 17:52.43 | nickoe | decimad: like there is supposed to be a lib_kicad in the future, but it can't work right now. |
| 17:53.56 | nickoe | it is a bit confusing https://github.com/blairbonnett-mirrors/kicad/blob/master/new/drawing.png |
| 17:54.11 | nickoe | ... not sure why I linked that from bliarbonnet... but meh |
| 17:54.39 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 17:54.39 | c4758p | ... |
| 17:54.45 | c4758p | squints at that diagram, tilts head |
| 17:55.20 | nickoe | c4757p: I think it is describing "SWEET" which is pretties for eeschema. No idea why that needs to be two seperate concepts. |
| 17:55.38 | c4758p | because kicad |
| 17:55.51 | nickoe | I think mostly because Dick right there. |
| 17:55.53 | c4758p | see also: why they're called "pretty" and "sweet" |
| 17:55.58 | c4758p | well, yeah |
| 17:56.15 | c4758p | but most things that are "because kicad" are really either "because Dick" or "because JP" |
| 17:56.16 | nickoe | where is that? I think I saw something like that before. |
| 17:56.16 | c4758p | >_> |
| 17:57.35 | c4758p | where is what? |
| 17:59.00 | nickoe | something like https://github.com/KiCad/kicad-source-mirror/wiki/Modular-KiCad-with-Alternate-Top-Level-Launchers just about that sweet stuff |
| 17:59.15 | *** join/#kicad firebird (~firebird@125.16.18.198) |
| 18:01.55 | nickoe | the road map also mentions this mysterious sweet thing http://ci.kicad-pcb.org/job/kicad-doxygen/ws/Documentation/doxygen/html/v5_road_map.html#v5_sch_sweet |
| 18:02.14 | c4758p | it's mysterious? |
| 18:02.22 | c4758p | sweet is to eeschema as pretty is to pcbnew |
| 18:03.04 | nickoe | yeah |
| 18:03.58 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 18:07.33 | decimad | Oh great, you can organize sub-projects into folders with cmake, sweet! |
| 18:08.16 | decimad | I think I can make kicad usable with msvc++^^ |
| 18:09.52 | c4758p | stealthily replaces decimad's msvc++ with a potato, wonders how long it'll take him to notice |
| 18:09.55 | *** join/#kicad pointhi (~pointhi@212.33.36.212) |
| 18:10.42 | nickoe | decimad: What do you mean? |
| 18:11.28 | *** join/#kicad erichvk_ (~esh@ppp118-210-78-135.lns20.adl2.internode.on.net) |
| 18:11.31 | decimad | currently the generated msvc project has a flat list of subprojects ... pcbnew ... pcbnew_kiface ... something ... something_lexer |
| 18:12.22 | decimad | I can organize those into separate folders so I don't have to look at the whole list all the time |
| 18:12.40 | nickoe | oh |
| 18:12.53 | decimad | like, resources, lexer-files, plugins, ... |
| 18:13.36 | *** join/#kicad sftp (~sftp@unaffiliated/sftp) |
| 18:15.38 | weatherhead | nobody likes cmake, but it's hard to find an alternative that's acceptable |
| 18:16.12 | weatherhead | anybody who doesn't learn the lesson and implements some superbly beautiful and elegant custom build system is doomed to not be used by anyone |
| 18:16.51 | c4758p | there's one positive thing you can always say about cmake |
| 18:16.56 | weatherhead | better than autotools? |
| 18:17.00 | c4758p | in fact, I think this should be its slogan |
| 18:17.05 | c4758p | At Least It's not Autotools⢠|
| 18:17.12 | weatherhead | or, "it's not autotools" |
| 18:17.13 | weatherhead | lol |
| 18:17.16 | c4758p | grins |
| 18:17.24 | weatherhead | or "at least it's not autotools on cygwin" |
| 18:18.43 | weatherhead | I feel like any buildsystem that tries to hard to have a nice language / DSL behind it will always have haters who hate that particular philosophy |
| 18:18.54 | weatherhead | but since everyone agrees cmake is complete garbage, at least people are kind of willing to accept it |
| 18:19.19 | c4758p | ...oh, crapola |
| 18:19.33 | c4758p | just spent rather a long time building a linux kernel, to discover he built the wrong version |
| 18:20.47 | decimad | i'd like if the projects would be eeschema_standalone and eeschame vs. eeschema nad eeschema_kiface ... |
| 18:22.26 | decimad | actually it would already be enough if the word kiface weren't there ;) |
| 18:26.49 | c4758p | oh, hm, I feel even stupider now |
| 18:27.07 | c4758p | I have no idea why I built that kernel on the box I'm going to run it on, when I could have built it on my much more powerful workstation |
| 18:27.58 | tom_fig | Tell me if you get kicad working with msvc++ Decimad, I'd really like to use that as well |
| 18:28.37 | decimad | tom_fig: it "works" already, I'm just trying to setup the cmake files so actually developing with it is practical |
| 18:28.53 | decimad | all I did so far was with msvc |
| 18:33.24 | nickoe | I like cmake |
| 18:33.56 | c4758p | yeah, but you're a pervert |
| 18:36.01 | tom_fig | Some dude tried to get Eclipse setup for KiCad |
| 18:36.08 | tom_fig | I never heard from him again |
| 18:36.58 | c4758p | hah |
| 18:37.00 | nickoe | haha, tom_fig, inductiveload had good luck with, it also worked for me. But I don't really like eclipse. I mean, it is not magic. |
| 18:37.21 | inductiveload | worked fine for me :-) |
| 18:37.23 | nickoe | https://gist.github.com/johnbeard/895bad60e4716f7f9c77 |
| 18:37.55 | nickoe | I tried to use vscode, but I can't figure out how to make the symbol lookup work |
| 18:38.32 | inductiveload | you can also just use the eclipse generator with cmake, that works well too (I think better than it used to) |
| 18:41.27 | *** join/#kicad phinxy (~tehhhd@unaffiliated/phinxy) |
| 18:45.13 | decimad | unless your try to compile that swig crap there is really not much to do to get kicad compatible with msvc, and much of that is also an error on the gcc side |
| 18:47.31 | *** join/#kicad kow_ (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 18:48.32 | decimad | pretty and sweet huh? |
| 18:50.12 | decimad | I mean seriously... both of the names give no hint of what they're about |
| 18:54.09 | nickoe | it is just names |
| 18:55.04 | c4758p | but giving hints about purpose is the point of names |
| 18:55.10 | c4758p | names are /important/ in software design |
| 18:55.16 | nickoe | What is ths purpose of c4757p? |
| 18:55.23 | c4758p | "in software design" :P |
| 18:55.43 | decimad | when did you change the 4 to an 8 btw? |
| 18:56.37 | c4758p | it was a 7 |
| 18:56.43 | c4758p | I increment it if I have multiple connections ;) |
| 18:57.35 | *** join/#kicad jjrh (~jjrh@216.106.104.252) |
| 18:57.47 | decimad | nickoe: Pretty georgeous huge flawless excellent? |
| 18:58.01 | decimad | See how that sentence makes no sense, even though I only changed words? |
| 18:58.48 | decimad | Naming is _very_ important |
| 18:59.07 | decimad | Much more than doxygen comments |
| 18:59.55 | *** join/#kicad k\o\w (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 19:00.07 | *** join/#kicad danw_ (~danw@213.218.195.7) |
| 19:01.23 | awolf3 | this is so so awesome |
| 19:01.30 | nickoe | awolf3: Whats up? |
| 19:01.41 | awolf3 | just working on the new script for building |
| 19:01.44 | nickoe | So which scripts are you currently using for the nightlies? |
| 19:01.53 | *** join/#kicad weatherhead (~danw@213.218.195.7) |
| 19:02.00 | awolf3 | basically what you had |
| 19:02.05 | nickoe | I just built it, but it seeems that it requires a "manual" step for the docs |
| 19:02.06 | awolf3 | but there are dependency issues |
| 19:02.12 | awolf3 | one mo, yes |
| 19:02.18 | awolf3 | it tells you, right? |
| 19:02.21 | awolf3 | it says to put the docs in a directory |
| 19:02.23 | nickoe | I am still confused about what folder it refers to, becsaue I see build-docs under wx folders! |
| 19:02.33 | awolf3 | that is weird |
| 19:02.34 | ETS_ | I have a question about power planes. When thermals are attached to a pin, but I still have a ratsnest line, is it connected or not? |
| 19:02.50 | c4758p | ETS_: uh, yesno |
| 19:02.50 | awolf3 | nickoe: the new scripts are coming along very nicely |
| 19:02.55 | c4758p | the copper is connected, but kicad doesn't think it is |
| 19:03.02 | awolf3 | they have 2 years of improvements baked in :) |
| 19:03.03 | c4758p | because connections are determined by connection to center of pad |
| 19:03.12 | c4758p | the zone fill must not have reached the cetner |
| 19:03.13 | c4758p | center |
| 19:03.13 | nickoe | awolf3: http://dpaste.com/10X93JP |
| 19:03.22 | nickoe | I so those two build-docs folders, are they the ones? |
| 19:03.31 | awolf3 | it sould be the root one |
| 19:03.32 | c4758p | you'll want to either adjust the zone parameters, or just draw a little trace from the center of the pad to where the zone actually did reach >_< |
| 19:03.44 | awolf3 | neither of those, nickoe |
| 19:04.07 | nickoe | awolf3: but as you see there are not build-docs in the cwd |
| 19:04.15 | ETS_ | no, there is a gap from the x in the center to the rest of the thermel |
| 19:04.15 | awolf3 | yeah, make it |
| 19:04.20 | awolf3 | mkdir build-docs |
| 19:04.36 | nickoe | awolf3: are you going to modify that to pull it from http://docs.kicad-pcb.org/master/kicad-doc-HEAD.tar.gz instead? |
| 19:04.44 | decimad | do you know which cmakelists generates the lexer targets? |
| 19:04.46 | awolf3 | That's what my jenkins job does, but it doesn't do it in the script |
| 19:04.51 | nickoe | ok |
| 19:04.54 | awolf3 | i.e. I have chain of jenkins jobs |
| 19:04.54 | c4758p | ETS_: oh, interesting |
| 19:04.54 | nickoe | I see. |
| 19:04.57 | c4758p | not sure what's going on then |
| 19:05.05 | ETS_ | which zone parameter adjusts that. I can't seem to make it change |
| 19:05.08 | awolf3 | $ find build-docs build-docs build-docs/help build-docs/help/en build-docs/help/en/cvpcb.pdf |
| 19:05.11 | awolf3 | that idnd't paste right |
| 19:05.15 | awolf3 | inside of build-docs should be help |
| 19:05.19 | nickoe | ok |
| 19:05.21 | awolf3 | inside of help should be the languages |
| 19:05.58 | *** join/#kicad kow_ (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 19:06.10 | nickoe | are you going to switch to git your for scripts? |
| 19:10.06 | *** join/#kicad eszett (~Eszett@dslb-084-059-074-203.084.059.pools.vodafone-ip.de) |
| 19:11.45 | ETS_ | c4758p=======+> https://snag.gy/hvsuoR.jpg |
| 19:13.07 | nickoe | that looks strange |
| 19:13.17 | ETS_ | https://snag.gy/14Wea3.jpg |
| 19:13.27 | ETS_ | settings added |
| 19:13.46 | nickoe | can you share your pcb or just with that footprint on it? |
| 19:14.19 | ETS_ | i think so, what d0 I need to do? |
| 19:15.00 | ETS_ | all of my feetprints <:7) do it |
| 19:15.11 | *** join/#kicad mgottschlag (~quassel@HSI-KBW-134-3-46-128.hsi14.kabel-badenwuerttemberg.de) |
| 19:15.21 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 19:15.45 | c4758p | ETS_: that might just be a rendering artifact in "outline" mode |
| 19:15.56 | c4758p | I can't test right now because my kicad is screwed up |
| 19:16.21 | c4758p | needs to rebuild |
| 19:16.30 | Thor-Arne | hm... what's the gray circle? |
| 19:17.04 | c4758p | oh, yes, how strange |
| 19:17.06 | c4758p | what is that? |
| 19:17.15 | Thor-Arne | from the legend on the right it looks like edge cuts |
| 19:17.19 | c4758p | yep |
| 19:17.33 | c4758p | ETS_: you didn't draw on Edge.Cuts on those pads, did you...? |
| 19:17.45 | ETS_ | no, I turned off pad filled mode ant that is there. Also, I have moved the drawing around on the screen, so,,, |
| 19:17.58 | c4758p | any chance you could export one of those footprints and send it to me? |
| 19:18.08 | c4758p | (could just send the whole PCB, if it's not sensitive) |
| 19:18.14 | weatherhead | is there a way to add arbitrary "drawing" layers in a PCB? |
| 19:18.15 | c4758p | (ideally that) |
| 19:18.17 | c4758p | weatherhead: no |
| 19:18.27 | weatherhead | I have some DXF files from mechanical guys I want to look at whilst not spoiling my own dwgs |
| 19:18.32 | c4758p | sadly the mechanical layers are fixed |
| 19:18.38 | weatherhead | ok. Well that's way down my own implementation list, but would still be nice ;-) |
| 19:18.39 | nickoe | weatherhead: but there are multiple |
| 19:18.47 | c4758p | I just pick one of the mech layers nobody uses, like Eco1/Eco2 |
| 19:18.49 | nickoe | weatherhead: you can use the eco layers for example |
| 19:18.53 | weatherhead | o |
| 19:18.55 | weatherhead | *ok |
| 19:19.03 | ETS_ | Edgecuts has a circle. It was used to align the footprint to the board. Placement was strictly controlled |
| 19:19.09 | c4758p | remove the circle |
| 19:19.13 | c4758p | do not draw on edgecuts there |
| 19:19.39 | c4758p | that signals that you want the PCB to be routed out in that spot. the zone fill is pulling back away from it so the router bit doesn't fuck up the copper :P |
| 19:21.07 | weatherhead | tbf that's what I used to do in altium in these situations, |
| 19:21.20 | ETS_ | 'k |
| 19:22.17 | ETS_ | That did it! Genius... sheer genius. what else are those circles going to interrupt |
| 19:22.43 | ETS_ | ? |
| 19:22.53 | weatherhead | ETS_: edgecuts should signal where the fibreglass is |
| 19:24.18 | ETS_ | yes, but I didn't think the circle would interrupt the program. I guess I need to delete them, eh? |
| 19:25.48 | ETS_ | FYI, they are two half circles, with a -180 degree angle, why minus I don't have a clue |
| 19:26.13 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 19:26.35 | ETS_ | Anyway, Thanks so very much. |
| 19:27.59 | weatherhead | just trying to build latest master from source, I get the following errors: "m_caller declared as void" |
| 19:28.05 | weatherhead | in coroutine.h |
| 19:28.24 | weatherhead | is my boost version too old? I have 1.54.0 |
| 19:28.38 | nickoe | yeah! upgrade |
| 19:28.44 | weatherhead | god damnit |
| 19:28.47 | weatherhead | ;-) |
| 19:28.50 | weatherhead | bleeding edge |
| 19:28.58 | nickoe | What system are you on? Debian old stable? |
| 19:29.05 | weatherhead | opensuse 42.1 |
| 19:29.13 | weatherhead | normally it's pretty up to date with things |
| 19:29.33 | nickoe | I advise to use at least boost 1.56 IIRC |
| 19:30.27 | nickoe | c4757p: am I correct, or am I confused with the numbering? |
| 19:30.49 | weatherhead | nickoe: installing newer version .At work I'm on ubuntu14.04 still, even worse ;-( |
| 19:30.57 | nickoe | :O |
| 19:31.12 | Thor-Arne | ETS_; if you need some drawings to align holes to use the Dwgs.User layer, the Edge.Cuts is meant to be cuted with the router |
| 19:31.52 | nickoe | you can't build kicad master on 14.04 IIRC, becasuse the compiler does not support some of the new c++11 features we are using. |
| 19:33.16 | ETS_ | Thanks for the tip. I had imported th board from a solidworks drawing I was given, |
| 19:34.16 | Thor-Arne | ah, then you need to make the pads that size and then delete the edge.cuts |
| 19:34.39 | weatherhead | nickoe: yes it's a weird install, we have latest icc installed so it will likely support everything, but libraries are ancient |
| 19:34.44 | *** join/#kicad drawkula (~username@p5B0C446E.dip0.t-ipconnect.de) |
| 19:35.03 | ETS_ | I am doing it now. Again, Tanks to all |
| 19:35.15 | ETS_ | er, Thanks |
| 19:35.33 | Thor-Arne | np |
| 19:36.14 | ETS_ | Just incase you were thinking tankards for all. Not a bad Idea, but No. |
| 19:36.59 | kakimir | why does kicad load wrong version of component |
| 19:37.06 | kakimir | preview is ok |
| 19:37.31 | kakimir | and when it placed it's the unmodified version of the component in another library |
| 19:38.18 | c4758p | it uses the first component it finds by that name, traversing the library list in order. reorder your libs in the library preferences |
| 19:38.31 | c4758p | (yes, we know it's broken. yes, it's going to be fixed. ;) |
| 19:41.41 | *** join/#kicad leizhou (~kudlit@2.89.234.167) |
| 19:41.52 | c4758p | lol wayne |
| 19:43.12 | kakimir | *sighs* |
| 19:43.22 | nickoe | c4757p: when are you going home? |
| 19:44.02 | c4758p | nickoe: getting tired of my connection? :) |
| 19:44.09 | nickoe | yea! |
| 19:44.12 | c4758p | lol |
| 19:44.30 | c4758p | fairly soon. I've basically been shaving yaks all day, I can do that from home |
| 19:46.19 | decimad | haha |
| 19:46.25 | decimad | what's wayne's irc name? |
| 19:46.52 | nickoe | ok, packaigng the last dmg now |
| 19:47.09 | nickoe | decimad: :P I guess Bruce |
| 19:47.49 | c4758p | I'd guess he doesn't have one - back when we were arguing about the digikey thing apparently someone sent him some excerpts and complained about them, I suspect I was in them ;) |
| 19:48.09 | nickoe | ohh, no, failure. |
| 19:48.30 | c4758p | skims through the nicklist for anything waynelike anyway out of curiosity~ |
| 19:49.00 | decimad | 3d-viewer is rather a plugin than a lib, right? |
| 19:49.15 | c4758p | sees nothing, but grins and waves just in case he's being watched |
| 19:49.15 | c4758p | ~ |
| 19:49.23 | c4758p | hm, define plugin :P |
| 19:49.31 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 19:49.52 | decimad | well, it handles its own life as an integral part of something else |
| 19:50.22 | decimad | I'll add it to the plugins folder |
| 19:50.25 | decimad | :P |
| 19:50.37 | *** join/#kicad bvernoux (~Ben@chl26-1-88-183-104-56.fbx.proxad.net) |
| 19:51.15 | nickoe | hmm, cp: mnt/kicad/./packageblablablablablaPT-2.54mm_2pol.wings: No space left on device |
| 19:52.24 | c4758p | mnt/kicad? o_O |
| 19:53.17 | nickoe | c4758p: something in here makes that a 1GiB mount point http://bazaar.launchpad.net/~adamwolf/+junk/kicad-mac-packaging/view/head:/package_kicad.sh |
| 19:53.25 | nickoe | /dev/disk2s1 1.0Gi 1.0Gi 0Bi 100% 267225 0 100% /Users/kicad/test/kicad-mac-packaging/packaging/mnt |
| 19:53.45 | nickoe | Header is ofc, Filesystem Size Used Avail Capacity iused ifree %iused Mounted on |
| 19:53.47 | c4758p | hah |
| 19:53.50 | c4758p | waves at ridesa7__ |
| 19:54.01 | c4758p | nickoe: ...mnt? |
| 19:54.09 | nickoe | :P |
| 19:54.10 | c4758p | is that on my box? |
| 19:54.13 | ridesa7__ | i'm waving back |
| 19:54.15 | nickoe | yeas |
| 19:54.28 | nickoe | woo |
| 19:54.28 | c4758p | wtf |
| 19:54.49 | c4758p | ah, I see |
| 19:54.57 | c4758p | forgot osx packages are actually disk images |
| 19:55.05 | nickoe | yeah |
| 19:55.16 | nickoe | but I wonder why it is magically limited to 1GiB there |
| 19:55.24 | c4758p | er |
| 19:55.28 | c4758p | because a disk image has a size :P |
| 19:55.43 | c4758p | it's not setting one anywhere that I can see, so that's probably the default |
| 19:55.57 | c4758p | 'man hdiutil' and see if you can find the option to increase it |
| 19:56.25 | nickoe | c4758p: if just your connection was not borked it was easy |
| 19:56.34 | c4758p | anyway, I'm heading home from work, will be home in ~20min to fix the connection. I'll have to boot you off to do that, so wrap up whatever you're doing in that time |
| 19:56.45 | nickoe | seems like there is a -size option |
| 19:56.52 | c4758p | I won't reboot the mini or anything silly like that, but I will be unplugging connections and swearing like a sailor ;) |
| 19:56.55 | nickoe | c4758p: sure |
| 19:57.10 | nickoe | then there are absolutely no issues |
| 19:57.14 | c4758p | :) |
| 19:57.14 | nickoe | working in tmux anyways |
| 19:57.42 | decimad | is there some tool to work with cmake trees? |
| 19:57.54 | c4758p | a hammer |
| 19:57.56 | decimad | browsing around the filesystem searching stuff sucks |
| 19:57.57 | nickoe | trees? |
| 19:58.09 | c4758p | or, given they're trees, perhaps a saw would be more appropriate |
| 19:58.14 | nickoe | you mean all the carious cmakelists? |
| 19:58.30 | decimad | where you open the top cmakelists.txt and you can browse the subdir tree of what it includes |
| 19:59.10 | nickoe | decimad: can't visual studio do that? |
| 19:59.22 | nickoe | I even think source insight would do that. |
| 20:00.15 | nickoe | awolf3: you there? Any idea of why that image fails now, other thatn we got more 3d models possibly? |
| 20:00.26 | decimad | I'll try the cmake tools for visual studio, however it seems only to do highlighting |
| 20:02.01 | *** join/#kicad berjan_ (~chatzilla@vpn096105.vpn.utwente.nl) |
| 20:03.03 | *** join/#kicad Kevlan (~Kevlan@android-efc9c8459d6a725c.mines.edu) |
| 20:04.20 | kakimir | c4757p: that is a major bug I say |
| 20:04.35 | kakimir | you may accidentally use wrong component and not even know it |
| 20:04.44 | c4757p | ...yes |
| 20:04.54 | c4757p | that's why I said it's known, and going to be fixed |
| 20:05.40 | c4757p | it is a major bug that is surprisingly difficult to fix. the fix is planned, and involves a full reimplementation of the library system to mirror the one in pcbnew |
| 20:05.43 | *** join/#kicad fellbuendel (~furry@unaffiliated/fellbuendel) |
| 20:06.05 | *** join/#kicad C47 (~d3bug@189.242.29.24) |
| 20:08.45 | ridesa7__ | i'm working on that. hopefully it will be fixed soon |
| 20:10.47 | *** join/#kicad opamp_rider (b2fad688@gateway/web/freenode/ip.178.250.214.136) |
| 20:12.26 | kakimir | c4757p: I bet it's hard |
| 20:15.03 | mozmck | ridesa7__: hi! |
| 20:15.50 | ridesa7__ | mozmck: what's up? |
| 20:16.07 | *** join/#kicad spirelli (5ed80b6d@gateway/web/freenode/ip.94.216.11.109) |
| 20:16.19 | mozmck | Just saw your post on the mailing list and then saw you here. |
| 20:16.53 | ridesa7__ | i've been lurking for a while, thought join in the fun! |
| 20:16.54 | mozmck | I've been mostly lurking on this channel since it had about 2 people in it |
| 20:17.47 | mozmck | IRC can be quite helpful for development collaboration if you can ignore the noise. There is a kicad-dev channel that almost never gets used. |
| 20:17.56 | ridesa7__ | my time is limited so i've avoided yet another distraction until now, i'll see how much of time sink it becomes. |
| 20:18.24 | mozmck | yeah, I'm the same way. I lurk but don't look at it much. |
| 20:19.15 | mozmck | the #linuxcnc-devel seems to keep a pretty high signal-to-noise ratio |
| 20:19.31 | mozmck | I'm Moses McKnight in real life btw |
| 20:19.35 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 20:20.05 | *** join/#kicad pointhi (~pointhi@heim-032-115.raab-heim.uni-linz.ac.at) |
| 20:20.11 | ridesa7__ | i was wonder who belonged to the nickname |
| 20:23.36 | *** join/#kicad m4v (~znc@ubuntu/member/m4v) |
| 20:24.03 | decimad | <3 https://s10.postimg.io/5py3kmal5/vstree.png |
| 20:24.21 | decimad | now the tedious part will be adding the headers to the sources |
| 20:25.02 | c4757p | for what it's worth, this channel is pretty okay at switching back on topic when the topic comes up. we just have a lot of offtopic side convos when it's otherwise quiet ;) |
| 20:25.12 | c4757p | but that's most irc channels I've seen :p |
| 20:26.25 | decimad | there's kicad-devel? |
| 20:27.17 | nickoe | kicad dev seems to multiplex the other discussions in here without problems |
| 20:27.35 | c4757p | decimad: yeah, but nobody has bothered switching over |
| 20:27.45 | ridesa7__ | c4757p: from what i've seen, this channel doesn't seem too noisy. it's kind of neat seeing what folks are working on at work and at play. |
| 20:27.57 | c4757p | I think a lot of people would move over entirely and leave this one dead |
| 20:27.57 | nickoe | mozmck: But you know the other place is a secret :P |
| 20:28.09 | decimad | especially the retarded discussions are worth it |
| 20:28.29 | nickoe | Yeah, lets have more discussions about retarded things |
| 20:28.38 | c4757p | personally I'm not really interested in user discussions very much, I'd leave if devel talk moved :p |
| 20:29.18 | eszett | :-(( I try to download Kicad Schematic symbols from the official kicad source, but I get a "not found" for http://library.oshec.org/compressed/all.tar.gz |
| 20:29.27 | eszett | Is the server down? |
| 20:29.42 | nickoe | oshec is not official |
| 20:29.45 | nickoe | as such |
| 20:29.50 | eszett | hm, what a pity |
| 20:29.50 | ridesa7__ | let me know if the dev talk is going to move to kicad-dev, i'll head over there |
| 20:30.18 | nickoe | ohh, oshec, isn't that the eagle stuff? That is in fact totally unsupported. |
| 20:30.56 | eszett | nickoe: I just need a simply Zener 323 symbol |
| 20:30.58 | nickoe | eszett: if it is the eagle lib thing, then there are strange things in there that makes kicad barf |
| 20:31.39 | eszett | nickoe: I guess its fast when i draw it by hand, than continueing the search in that jungle of 404 and "website not found" |
| 20:31.51 | eszett | faster |
| 20:31.53 | nickoe | eszett: so add a ZENER or ZENERsmall and put the proper part number on it |
| 20:32.01 | eszett | ye ok |
| 20:32.09 | awolf3 | Hi Nick! |
| 20:32.15 | nickoe | eszett: why don't you just use the one that should be preinstalled? |
| 20:32.18 | awolf3 | sounds like you ahve some questions about the mac packager |
| 20:32.27 | nickoe | awolf3: I am having issues with package_kicad.sh |
| 20:32.32 | awolf3 | ok |
| 20:32.40 | *** join/#kicad drod (~ldm@ip-95-221-235-169.bb.netbynet.ru) |
| 20:32.44 | awolf3 | what are the issues? |
| 20:32.48 | nickoe | it seems to barf when the tmp mountpoint gets to 1GiB |
| 20:32.50 | awolf3 | it is actually really, really complicated. |
| 20:32.59 | nickoe | awolf3: :O |
| 20:33.03 | awolf3 | like |
| 20:33.04 | nickoe | KISS FFS |
| 20:33.10 | awolf3 | that's apple's fault |
| 20:33.15 | c4757p | if you two want to kiss, get a room |
| 20:33.23 | nickoe | looking at http://bazaar.launchpad.net/~adamwolf/+junk/kicad-mac-packaging/view/head:/package_kicad.sh |
| 20:33.38 | awolf3 | Do you see how it uses a template DMG? |
| 20:33.51 | nickoe | mmm not really. |
| 20:33.53 | awolf3 | ok |
| 20:33.56 | awolf3 | that's part of the problem :) |
| 20:33.59 | c4757p | nickoe: lol, the connection is much more solid connecting /out/ |
| 20:34.09 | c4757p | I was home from work yesterday, I wonder how long this went on without me noticing |
| 20:34.22 | awolf3 | so there's a template dmg that we start from |
| 20:34.22 | nickoe | awolf3: whats inside it? Can't it just be created everytime? |
| 20:34.24 | Yatekii | ok guys, today is the day i might implement a hotkey yay! |
| 20:34.25 | awolf3 | no. |
| 20:34.36 | Yatekii | code finally compiles and i got timw |
| 20:34.56 | nickoe | Yatekii: hotkey for what? |
| 20:35.01 | awolf3 | When users open the DMG, it opens the window to a certain size, it has a custom background, it changes the icon size |
| 20:35.09 | awolf3 | it has the drag arrow |
| 20:35.29 | Yatekii | nickoe: setting the grid in the footprint editor |
| 20:35.35 | Yatekii | and the pcb editor |
| 20:35.37 | nickoe | awolf3: And can't that just be copied into a new image (dmg) every time? |
| 20:35.44 | awolf3 | I mean, if you want to write it |
| 20:35.46 | Yatekii | its one of the features i use the most |
| 20:36.00 | awolf3 | All the people I talked to who do mac install dmgs use a template. |
| 20:36.08 | nickoe | awolf3: The problem being that it seems I can't package it as is, because the image is too small |
| 20:36.08 | awolf3 | There are a few unmaintained template dmg creators |
| 20:36.16 | awolf3 | first off, it works for me :) |
| 20:36.22 | awolf3 | second, you can enlarge the template dmg. |
| 20:36.23 | nickoe | I don't believe you |
| 20:36.38 | nickoe | I will try to get a proper log. |
| 20:36.46 | nickoe | awolf3: did it work for you just now? |
| 20:36.54 | awolf3 | ? |
| 20:36.57 | awolf3 | I'm not building right now |
| 20:37.04 | awolf3 | you mean like last night? |
| 20:37.10 | Yatekii | also what i would fancy is the ability to press tab when placing parts or traces or pads so you can edit the component you will place ibstead of placing it and then manually editing them all :) |
| 20:37.25 | Yatekii | its a fearure i liked in altium |
| 20:37.59 | awolf3 | It worked for me a week ago, for sure. |
| 20:38.15 | nickoe | awolf3: I mean today |
| 20:38.32 | nickoe | I get not space left errors |
| 20:38.40 | nickoe | *no space |
| 20:38.56 | reportingsjr | hahaha, Wayne's comment about c4757p raking him over the coals on here :D |
| 20:39.04 | c4757p | looks innocent |
| 20:39.09 | c4757p | whistles |
| 20:39.33 | awolf3 | so I recently updated the extras dmg to be 1.5G |
| 20:39.43 | awolf3 | I will check to see if it worked today |
| 20:39.47 | awolf3 | if not, I'll expand teh DMG again |
| 20:39.54 | awolf3 | its one line on the CLI.... |
| 20:40.10 | nickoe | awolf3: http://dpaste.com/0JQN083 |
| 20:40.13 | awolf3 | don't forget though, OS X treats dmgs over 1 G different than ones under 1G! |
| 20:40.27 | awolf3 | over 1G, OS X will start indexing it, so you can't eject it the same way |
| 20:40.37 | awolf3 | you'll have to add metadata_never_index and fseventsd/no_log |
| 20:41.10 | awolf3 | (Does this seem crazy to anyone else?) |
| 20:41.11 | nickoe | awolf3: I am on revno 94 of your scripts |
| 20:41.38 | nickoe | awolf3: is there no option to have it auto resize? |
| 20:41.41 | awolf3 | no |
| 20:41.47 | awolf3 | that is not how OS X works, nickoe :) |
| 20:41.51 | nickoe | awolf3: how do I resize it? |
| 20:42.07 | awolf3 | hdiutil |
| 20:42.12 | awolf3 | take the template dmg bz2 |
| 20:42.13 | awolf3 | expand it |
| 20:42.19 | awolf3 | then use hdiutil to expand it |
| 20:42.20 | c4757p | that's how disk images work |
| 20:42.22 | awolf3 | then bz2 it |
| 20:42.29 | *** join/#kicad kraiskil (~kraiskil@a91-152-241-34.elisa-laajakaista.fi) |
| 20:42.30 | c4757p | try working inside a disk image on windows or linux and see it it autoresizes /there/ ;) |
| 20:42.48 | awolf3 | if it gives you errors, thats just more of how OS X works, and I'll show you how to recreate the template DMG in Finder |
| 20:43.03 | nickoe | awolf3: huh, you said it was one line before.. |
| 20:43.29 | awolf3 | hdiutil resize -size 25G /PATH/TO/DISK/IMAGE.dmg |
| 20:43.31 | nickoe | awolf3: I am not going to use the grapical interface. The VNC connection is awfully slow here. |
| 20:43.35 | nickoe | wee |
| 20:43.44 | awolf3 | I stored all that information on google.com |
| 20:43.48 | ridesa7__ | reportingsjr: c4557p wasn't the only one joining in on the Wayne bashing |
| 20:43.49 | awolf3 | :p |
| 20:44.28 | awolf3 | nickoe: OK. If that command doesn't work for you, and it is not obvious why, let me know and I will recreate the template dmg to be larger |
| 20:44.41 | awolf3 | I will also check to see if maybe somethign else is wrong in your setup, if it is working in my setup |
| 20:45.03 | awolf3 | Due to the 10.7/10.9 dependencies issue on my build box, I haven't been watching to see if the builds work well |
| 20:45.18 | awolf3 | Are you planning on building for 10.11/10.12? |
| 20:45.22 | awolf3 | or for 10.9+? |
| 20:45.23 | c4757p | ridesa7__: perhaps, but if you characterized me as a bit more of a grouch than most people I'd hardly protest ;) |
| 20:45.32 | nickoe | awolf3: yeah, I am testing it now, it seems to take some time to resize. |
| 20:45.32 | c4757p | nickoe: bleh. I have no idea what's wrong with the connection |
| 20:45.36 | c4757p | might have to get the ISP involved :\ |
| 20:45.45 | nickoe | c4757p: wireshark? |
| 20:45.48 | awolf3 | nickoe: I will try to sit on this channel for hte next few weeks |
| 20:45.53 | c4757p | dude wireshark isn't going to help |
| 20:45.56 | nickoe | c4757p: or tcpdump |
| 20:46.03 | awolf3 | I am hoping to have the new scripts out very soon, like... monday? |
| 20:46.05 | c4757p | packets just suddenly stop going through |
| 20:46.05 | nickoe | awolf3: :) |
| 20:46.17 | nickoe | awolf3: when are you most likely to be responsive? |
| 20:46.27 | awolf3 | 7AM CST to now |
| 20:46.29 | c4757p | I'm guessing it's an issue on the RF side. either the ISP, or the roofing company that was over the other day, "interacted with" the cables |
| 20:46.31 | nickoe | c4757p: yeah, maybe you should kick that midget out |
| 20:46.41 | nickoe | ok |
| 20:46.58 | corecode | did the file format on master change to be incompatible with 4.0.2? |
| 20:47.00 | c4757p | nickoe: in the meantime, want me to forward a port for mosh? |
| 20:47.07 | c4757p | it's much better than ssh on an unreliable connection :P |
| 20:47.32 | ridesa7__ | c4757p: i never thought of you as being grouchy just mildly opionated :) |
| 20:47.32 | nickoe | I don't think mosh needs another port |
| 20:47.40 | c4757p | yes, mosh needs another port |
| 20:47.44 | c4757p | it uses a UDP port in the high range |
| 20:47.51 | c4757p | in addition to ssh-based auth |
| 20:48.23 | c4757p | the only port you have access to is tcp:22, so even if it could use that port you wouldn't get its strengths |
| 20:48.26 | c4757p | which come from it using udp :P |
| 20:48.50 | nickoe | are you sure, I think I have used mosh with only the ssh port |
| 20:48.53 | nickoe | but I won't say no then |
| 20:49.05 | c4757p | well, you are welcome to give it a try >:) |
| 20:49.16 | awolf3 | I ahve another question |
| 20:49.17 | c4757p | but let me know when it doesn't work and you want that port forwarded ;) |
| 20:49.20 | awolf3 | where is the Jenkins stuff hosted? |
| 20:49.27 | awolf3 | My company has a spare mac mini |
| 20:49.33 | awolf3 | We can ship it somewhere if that helps |
| 20:49.41 | c4757p | awolf3: oh, nice |
| 20:49.43 | awolf3 | and we can make this all official and tied into thee real jenkins |
| 20:49.46 | c4757p | I just donated my Mini |
| 20:49.53 | c4757p | and I mean that literally, it's kicad's now |
| 20:49.59 | c4757p | so y'all can do whatever you want with it |
| 20:50.05 | awolf3 | that works too |
| 20:50.13 | c4757p | not sure if we'd have a use for two |
| 20:50.17 | awolf3 | no problem |
| 20:50.21 | nickoe | awolf3: c4757p's box it already connected to kicad's jenkins |
| 20:50.22 | awolf3 | it'd be nice to have one on 10.9 |
| 20:50.26 | awolf3 | but if we had one on 10.9, it'd get owned fast |
| 20:50.29 | awolf3 | ok |
| 20:50.30 | eszett | I have a question ... In Eeschema library editor I can set the grid size in MILs, how can change it to Millimeters? Not possible? |
| 20:50.38 | awolf3 | are we trying to get official packages built out of it? |
| 20:50.48 | c4757p | eszett: not possible |
| 20:50.51 | nickoe | we can build more! |
| 20:50.52 | awolf3 | if so, we have to be super super particular about every thing that gets installed |
| 20:50.58 | eszett | c47: ok then i will live with that |
| 20:50.59 | awolf3 | i.e. brew and stuff like that have to be hand-edited |
| 20:51.22 | c4757p | awolf3: if you want me to do a fresh OSX install on it at some point so you can be very careful about the exact setup, let me know |
| 20:51.29 | awolf3 | how about this |
| 20:51.39 | nickoe | awolf3: not sure whey I get those errors at the end http://dpaste.com/3RFRMFJ |
| 20:51.39 | awolf3 | how about right now, we set htat box up to do builds, targeting the current osx version |
| 20:51.43 | nickoe | awolf3: How about signing, what is needed for that given we get the nightlies running properly agian? |
| 20:51.44 | c4757p | I strongly recommend using proper management software to control it, if you can, when you want it to be official |
| 20:51.57 | awolf3 | I have ansible controlling mine, c4757p |
| 20:52.00 | c4757p | is familiar with saltstack, I think puppet and chef also support osx |
| 20:52.02 | c4757p | ah, good |
| 20:52.23 | awolf3 | nickoe: I followed the official rules for signing apps |
| 20:52.29 | awolf3 | I paid apple the $100 or whatever |
| 20:52.31 | awolf3 | and it says |
| 20:52.34 | awolf3 | "Unknown error" |
| 20:52.39 | awolf3 | it's due to how we include python |
| 20:52.47 | awolf3 | Someone was working on that. |
| 20:52.51 | awolf3 | ok |
| 20:53.03 | nickoe | awolf3: well, really I am just trying to understand what your issues are for now and playing around with it. I expected you to fix it :P |
| 20:53.03 | awolf3 | it's clear to me, the first priority is to get the next version of these scripts out |
| 20:53.21 | awolf3 | once the next version of the script is out, we can get it working on mulitple setups |
| 20:53.26 | awolf3 | i.e. mine, and the official jenkins |
| 20:53.42 | awolf3 | I really like the idea of having the official jenkins one produce builds for the current osx version |
| 20:53.50 | awolf3 | by doign that, you can just do regular sofwtware updates |
| 20:53.52 | awolf3 | and it wont get owned |
| 20:54.05 | awolf3 | mine is on a special network, firewalled to hell because I never update anything on it |
| 20:54.22 | c4757p | mine's firewalled to hell too, even though it /is/ updated |
| 20:54.29 | c4757p | because I'm letting random kicad people access it ;) |
| 20:54.38 | awolf3 | ha, yeah, thats another thing |
| 20:54.46 | awolf3 | the jenkins user can do a lot of stuff on that box |
| 20:54.49 | awolf3 | anyway |
| 20:54.50 | c4757p | actually I could tighten the firewall a bit more |
| 20:55.00 | nickoe | meh, still errors |
| 20:55.20 | nickoe | awolf3: why is this still 1G? /dev/disk2s1 1.0Gi 1.0Gi 0Bi 100% 267225 0 100% /Users/kicad/test/kicad-mac-packaging/packaging/mnt |
| 20:55.21 | awolf3 | nickoe: |
| 20:55.25 | awolf3 | i am not sure |
| 20:55.28 | awolf3 | but those errors with permission denied |
| 20:55.35 | awolf3 | are because osx treats 1G+ dmgs differently |
| 20:55.52 | awolf3 | I'll pause working on my scripts to get this working for you, but I gotta head home |
| 20:56.14 | awolf3 | how big did you expand it to? |
| 20:56.22 | nickoe | awolf3: 25G |
| 20:56.22 | awolf3 | i.e. how big is the directory you're trying to fit into the dmg? |
| 20:56.24 | awolf3 | haha |
| 20:56.39 | awolf3 | so OS X thinks thats like an external HD |
| 20:56.43 | nickoe | but when I du -hs packaging/kicadtemplate.dmg it is still 1.1G |
| 20:56.59 | awolf3 | so it's a sparse image, or should be |
| 20:57.06 | awolf3 | it was one, at least :) |
| 20:57.14 | awolf3 | so the disk space used on disk is not the same as the capacity of the image |
| 20:57.25 | awolf3 | if you have a sparse image with no contents, it will be like 30k |
| 20:57.32 | *** join/#kicad darkstar-20011 (~matt@217.155.229.6) |
| 20:57.55 | awolf3 | i love how this channel is getting a deep dive into the complexities of like 1/10th of what goes into packaging KiCad for OS X |
| 20:58.30 | nickoe | yeah, good. But why does df -h still say it is max 1G then? I would expect that to report the resized size and not the on disk one. Maybe we resized the wrong partiion on it, maybe? |
| 20:58.36 | awolf3 | I am not sure. |
| 20:58.45 | awolf3 | I resize the images like every 6 months |
| 20:58.54 | awolf3 | I will take a look and try to figure out what's going on, tomorrow morning |
| 20:59.09 | nickoe | I simply did hdiutil resize -size 25G packaging/kicadtemplate.dmg and then rerun the script |
| 20:59.12 | awolf3 | also, any way I can get onto this machine too? |
| 20:59.17 | awolf3 | oh |
| 20:59.34 | awolf3 | I suspect the script re-extracted the bz2 right over your new template |
| 20:59.40 | reportingsjr | ridesa7__: :) |
| 21:00.47 | ridesa7__ | reportingsjr: hi! |
| 21:00.55 | awolf3 | anyway, check that nickoe |
| 21:01.01 | *** join/#kicad tumnao (b91f6476@gateway/web/freenode/ip.185.31.100.118) |
| 21:01.02 | awolf3 | I gotta head out, but I'll be back tomorrow! |
| 21:01.05 | nickoe | http://dpaste.com/1FRREQV |
| 21:01.13 | nickoe | awolf3: ohh |
| 21:01.24 | ridesa7__ | gotta run folks, catch up with you tomorrow |
| 21:01.27 | *** part/#kicad ridesa7__ (~quassel@rrcs-98-101-140-98.midsouth.biz.rr.com) |
| 21:01.43 | nickoe | c4757p: your connection seems to work a bit better for the time being |
| 21:01.49 | c4757p | nickoe: huh, my activity plot looks a bit nicer now. is my connection treating you any bet------ah |
| 21:01.51 | c4757p | good |
| 21:01.53 | c4757p | :) |
| 21:01.55 | c4757p | thumbs up |
| 21:02.31 | awolf3 | if you are still having problems with this tomorrow morning, lmk if I can get in on this machine too :P |
| 21:02.41 | c4757p | ...damn, I come home and spontaneously got a headache |
| 21:02.44 | c4757p | wtf |
| 21:02.46 | c4757p | rubs head |
| 21:02.48 | awolf3 | if hdiutil starts acting really wonky on you, you'll need to reboot |
| 21:03.39 | nickoe | awolf3: is it the uncompressed.dmg that I should resize or the template one? |
| 21:04.01 | awolf3 | hmm |
| 21:04.03 | nickoe | c4757p: Don't die! |
| 21:04.09 | c4757p | I'm trying not to |
| 21:04.25 | c4757p | I think I'll have more leftover Indian food. that makes everything better |
| 21:04.26 | awolf3 | what I have always done is take the template.dmg.bz2, extract it to just a dmg, resize it, and then re bz2ed it |
| 21:04.35 | nickoe | it is your c4poooter your are supposed to use the hammer on, not your head. |
| 21:04.36 | awolf3 | and then re-ran the script, and everything works a++++ at that point |
| 21:04.39 | c4757p | nickoe: lol |
| 21:04.53 | awolf3 | anyway |
| 21:04.57 | awolf3 | ttyl |
| 21:05.14 | nickoe | ok, awolf3 cyl |
| 21:07.08 | *** join/#kicad nitrousnrg (~quassel@190-103-222-12.cepanet.com.ar) |
| 21:10.18 | decimad | Would it perhaps make sense to let the cmake scripts be c programs, considering someone setting up a c project probably can program c but not cmake? :) |
| 21:15.09 | corecode | oh my cmake |
| 21:15.12 | corecode | i had so much hope |
| 21:15.20 | corecode | but you disappointed me |
| 21:19.29 | *** join/#kicad Kevlan (~Kevlan@android-efc9c8459d6a725c.mines.edu) |
| 21:20.43 | nickoe | ok, resizing that dmg worked! |
| 21:42.36 | corecode | asking again: did the pcb file format change between master and 4.0? |
| 21:42.53 | corecode | i can't open master saved files in 4.0 |
| 21:43.01 | corecode | which is very inconvenient |
| 21:45.38 | decimad | Why do you hop forth and back? |
| 21:47.36 | *** join/#kicad kmehall (~quassel@droplet.kevinmehall.com) |
| 21:48.10 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 21:48.14 | decimad | <3 https://s10.postimg.io/xfpvscpyh/vsheaders.png |
| 21:48.27 | *** join/#kicad Laogeodritt (laogeodrit@wikipedia/Laogeodritt) |
| 21:48.34 | decimad | the standard cmake files get nowhere near close to that |
| 21:49.04 | *** join/#kicad trygvis (trygvis@knot.trygvis.io) |
| 21:49.49 | corecode | decimad: working with other people |
| 21:50.06 | *** join/#kicad Kevlan (~Kevlan@2607:fb90:6c3d:4623:0:4f:69d1:c701) |
| 21:50.13 | *** join/#kicad m4v (~znc@ubuntu/member/m4v) |
| 21:50.34 | *** join/#kicad Jacob______ (sid27928@gateway/web/irccloud.com/x-pvdmziibpglfmros) |
| 21:50.55 | *** join/#kicad dingbat (uid70835@gateway/web/irccloud.com/x-luhmkilguxdqgvlm) |
| 21:51.00 | *** join/#kicad leizhou (~kudlit@2.89.234.167) |
| 21:51.06 | *** join/#kicad mrnuke (~mrnuke@2601:2c1:8580:c9:da50:e6ff:fe00:4a96) |
| 21:51.46 | *** join/#kicad DemoFreak (~hannes@pdpc/supporter/active/demofreak) |
| 21:51.48 | *** join/#kicad filt3r (~filter@2a03:b0c0:2:d0::7d:7001) |
| 21:51.48 | *** join/#kicad mosinut (mosinut@unaffiliated/mosinut) |
| 21:51.53 | *** join/#kicad rje (~rje@emanuele.us) |
| 21:52.03 | *** join/#kicad Nutter (~hehe@2605:6400:2:fed5:22:f62c:b508:6f00) |
| 21:52.05 | *** join/#kicad TD-Linux (~Thomas@about/essy/indecisive/TD-Linux) |
| 21:52.13 | *** join/#kicad seve (~seve@2604:180:2:1c3::27bd) |
| 21:52.25 | *** join/#kicad abcrawf (~abcrawf@apollo.acrawford.com) |
| 21:53.13 | corecode | it's the diff_pair setting -_- |
| 21:53.52 | nickoe | corecode: also be aware of round rect pads |
| 21:54.27 | nickoe | gnight |
| 21:55.25 | corecode | well this is poop |
| 21:55.38 | corecode | it's because of this brainless parser |
| 21:55.48 | corecode | which flips out if there is a token it doesn't know |
| 21:56.02 | corecode | this is not elegant |
| 21:56.32 | corecode | as long as new features aren't used, this shouldn't prevent older versions to open the newer file |
| 21:59.24 | *** join/#kicad reportingsjr (~reporting@2604:a880:800:10::11e:d001) |
| 22:00.25 | *** join/#kicad helaughsatfear (~this@2602:306:ccba:8e0:203:25ff:fe22:67d7) |
| 22:00.36 | *** join/#kicad helaughsatfear (~this@unaffiliated/helaughsatfear) |
| 22:01.00 | xzcvczx | no i missed adam |
| 22:01.08 | *** join/#kicad intart (~intart@148.103.127.48) |
| 22:02.51 | xzcvczx | you wanna know whats funny.... c4's boards are actually only 2-layer they just figured out that c4 was dumb enough to accept his boards just didn't work so they just left out the other 4 layers to save money |
| 22:09.43 | decimad | xzcvczx: okay, so you wanna say they only produced 2 layers because the board wouldn't work anyways? |
| 22:09.52 | xzcvczx | yup |
| 22:10.01 | decimad | good, I needed to ask ;) |
| 22:11.21 | x29a | where are you guys having your (small amounts of) pcbs manufactured? |
| 22:11.31 | xzcvczx | pcbshopper.com |
| 22:12.04 | c4757p | xzcvczx: I actually double-checked my layer legend to make sure all six were visible :D |
| 22:12.12 | c4757p | mostly out of excitement for having my first six-layer back :D |
| 22:12.55 | xzcvczx | haha |
| 22:18.22 | reportingsjr | x29a: oshpark for most stuff |
| 22:18.42 | c4757p | oh, right, someone asked a question |
| 22:18.43 | reportingsjr | c4757p: damn, you went to six layers? How much did that run ya? |
| 22:18.55 | c4757p | x29a: oshpark for quality or tiny boards, elecrow for cheap |
| 22:19.02 | c4757p | reportingsjr: bit over $400 |
| 22:19.04 | x29a | r |
| 22:19.07 | reportingsjr | ouch |
| 22:19.36 | c4757p | shrugs |
| 22:19.47 | decimad | show-off! :P |
| 22:20.05 | c4757p | it'll probably be the only substantial hardware project I get to work on for quite a while |
| 22:20.29 | reportingsjr | fair enough |
| 22:20.41 | reportingsjr | The most expensive boards I have ordered for myself were $80 |
| 22:20.46 | reportingsjr | but I got 10 of them :P |
| 22:20.50 | c4757p | heh |
| 22:20.51 | c4757p | I got 5 |
| 22:21.27 | c4757p | the quality is pretty good on these. mask has a few nicks, and silk is a bit misaligned |
| 22:21.29 | xzcvczx | how much would you have payed for 1 or 2? |
| 22:21.34 | c4757p | 5 min |
| 22:21.38 | c4757p | so...the same |
| 22:21.51 | decimad | I will absolutely send this diff as a patch to the list :P |
| 22:21.55 | xzcvczx | ah lol |
| 22:22.11 | Bird|otherbox | c4757p: where in the world did you get them from? |
| 22:22.12 | decimad | I wonder if MSVC can be endoded hexadecimal in cmake or something |
| 22:22.14 | c4757p | I think I'll end up using three - |
| 22:22.18 | xzcvczx | c4757p: why did you bother with smd jtag ports? |
| 22:22.23 | Bird|otherbox | also, mind linking the elecrow folks? |
| 22:22.29 | c4757p | I may fasten two of them together and use them as a rigid mounting plate to go under the main one |
| 22:22.48 | c4757p | I was going to use aluminum for that, but these are surprisingly stiff, I think two of them together will do and will save me cutting and drilling the plate |
| 22:22.55 | c4757p | Bird|otherbox: *shudder* pcbway |
| 22:23.22 | c4757p | Bird|otherbox: http://www.elecrow.com/services/pcb-prototyping.html |
| 22:25.22 | Bird|otherbox | c4757p: the elecrow stuff looks interesting for a cheap solution. |
| 22:25.30 | c4757p | quality's pretty okay |
| 22:29.53 | decimad | which language is "librairi" ? |
| 22:30.31 | c4757p | none that I'm aware of |
| 22:30.33 | c4757p | :P |
| 22:30.47 | c4757p | "librairie" is French. perhaps poorly spelled French :) |
| 22:31.30 | decimad | I thought so to, but the online translators only knew bibliothèque |
| 22:32.31 | decimad | oh, the other way around the find it, yeah |
| 22:32.52 | c4757p | they don't translate it as "library" :) |
| 22:33.29 | c4757p | interesting, though. it doesn't seem to be used for "library" in a computing context /either/ https://fr.wikipedia.org/wiki/Biblioth%C3%A8que_logicielle |
| 22:33.34 | c4757p | so... /quirky/ misspelled French? |
| 22:33.37 | c4757p | ¯\_(ã)_/¯ |
| 22:34.47 | decimad | you could hear me speak french here ;) |
| 22:34.54 | decimad | and laughing ;) |
| 22:35.43 | decimad | okay, I'm super happy with my msvc setup now, celebrate \o\ /o/ |
| 22:36.23 | c4757p | hah |
| 22:36.44 | reportingsjr | yay! congrats :) |
| 22:37.54 | c4757p | hm |
| 22:38.05 | c4757p | I've never done an i2c /slave/ on xmega before. wish me luck >_> |
| 22:38.09 | c4757p | grabs the manual and vim |
| 22:38.19 | reportingsjr | can't be that bad! |
| 22:38.23 | c4757p | probably not |
| 22:38.30 | c4757p | I <3 the xmega peripherals, they're all so nice to work with |
| 22:38.43 | reportingsjr | Can't be as bad as working with STM32s |
| 22:39.03 | c4757p | I've heard terrible things about I2C on STM32 :) |
| 22:39.21 | reportingsjr | s/I2C/everything/ |
| 22:39.23 | decimad | isn't it usually the standard lib there that is kind of strange? |
| 22:39.31 | c4757p | heh, STM32's pretty popular though |
| 22:39.35 | reportingsjr | They're so nice and cheap, but man the documentation is awful awful awful |
| 22:39.51 | c4757p | ah, yes |
| 22:39.55 | c4757p | ST docs are bad |
| 22:40.05 | decimad | well, the ptp clock functionality wasn't documented well... the other stuff was usable |
| 22:40.08 | c4757p | I don't really use STM32, but from the other ST parts I've used... |
| 22:40.19 | *** join/#kicad Shuggsy (~Shuggsy@c-73-7-71-211.hsd1.ga.comcast.net) |
| 22:40.36 | decimad | still my head exploded because of all the cross-reading between multiple documents |
| 22:40.45 | reportingsjr | You basically have to have a good working knowledge of 700+ pages of documentation to use a specific STM32 chip and not constantly trip over stuff |
| 22:40.51 | reportingsjr | decimad: ugh |
| 22:41.17 | reportingsjr | especially when one document covers most of topic X, except for this one little important sentence in a completely separate document. |
| 22:41.37 | decimad | hehe, yeah |
| 22:50.41 | xzcvczx | reportingsjr: you aer starting to sound like c4 |
| 22:52.15 | xzcvczx | cirilo: DOWN WITH VRML!!! |
| 22:57.35 | reportingsjr | aer? Stop trying to act scottish |
| 22:57.48 | reportingsjr | first the sheep fucking, no you're starting to spell like them, sheesh! |
| 23:01.36 | Yatekii | lol i don't get a single words of the comments on how to add a new shortcut :P |
| 23:03.29 | *** join/#kicad bugzc (~1@unaffiliated/bugzc) |
| 23:04.20 | c4757p | raises eyebrow at reportingsjr |
| 23:04.23 | Yatekii | I'll need some sleep and work tomorrow I guess :D |
| 23:04.51 | reportingsjr | c4757p: I gotta throw shade back at xzcvczx |
| 23:05.15 | c4757p | lol |
| 23:10.24 | xzcvczx | lol |
| 23:11.25 | *** join/#kicad k\o\w (~fffffff@CPE3085a93a4319-CM00fc8db094f0.cpe.net.cable.rogers.com) |
| 23:24.13 | *** join/#kicad cirilo (~cirilo@120.144.53.157) |
| 23:27.06 | GyrosGeier | ow |
| 23:27.23 | GyrosGeier | the people at the hackerspace want me to finish the controller for the CNC mill |
| 23:27.29 | cyborg_ar | stm32 micros are nice, but you keep stepping on garden rakes with them |
| 23:27.33 | GyrosGeier | so many projects, so little time |
| 23:27.37 | cyborg_ar | then finish it |
| 23:27.43 | cyborg_ar | stop adding safeties |
| 23:27.50 | cyborg_ar | let nature take it's course |
| 23:28.03 | GyrosGeier | nature is a 6NM motor in this case |
| 23:28.23 | cyborg_ar | yup exactly |
| 23:31.00 | archivist | use linuxcnc, all the safety needed is built in |
| 23:31.05 | GyrosGeier | lol |
| 23:31.10 | decimad | what kind of motor? |
| 23:31.11 | *** join/#kicad mroszko_ (~mroszko@413335d3.cst.lightpath.net) |
| 23:31.14 | cyborg_ar | lol yeah |
| 23:31.23 | cyborg_ar | let's put the safety in the software |
| 23:31.33 | archivist | it is actually running some very big iron |
| 23:31.51 | GyrosGeier | two Berger-Lahr VRDM3913 50LNBOO |
| 23:32.03 | GyrosGeier | and a Berger-Lahr VRDM368 50LNBOB |
| 23:32.11 | cyborg_ar | archivist, i dont doubt that, but i also doubt _all_ the safety is in linuxcnc |
| 23:32.27 | *** join/#kicad metacollin (~anonymous@c-174-56-101-29.hsd1.nm.comcast.net) |
| 23:32.36 | archivist | well has estop inputs any external stuff direct |
| 23:32.46 | cyborg_ar | a minimal hardware safety is to put the machine in a safe state if the charge pump signal is lost |
| 23:32.59 | cyborg_ar | ie, linuxcnc loses control |
| 23:33.01 | archivist | has that too |
| 23:33.03 | decimad | thanks |
| 23:33.27 | *** join/#kicad mroszko__ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 23:33.38 | cyborg_ar | then a second safety is encoder feedback on all axis, plus servo control limits, if a motor stops responding, bam |
| 23:33.47 | decimad | can I somehow make git give me an alert when a new header file is added? |
| 23:33.57 | cyborg_ar | decimad, make a hool |
| 23:34.00 | cyborg_ar | hook* |
| 23:34.14 | decimad | The question included "can I" :D |
| 23:34.18 | cyborg_ar | grep for .h(pp)$ |
| 23:34.34 | cyborg_ar | you definitely can |
| 23:34.37 | decimad | Got no grep ;) |
| 23:34.40 | cyborg_ar | it's just a shell script |
| 23:34.42 | cyborg_ar | install it |
| 23:34.45 | decimad | nooo |
| 23:34.51 | cyborg_ar | also if you have git |
| 23:34.53 | decimad | I'm windooozeee! |
| 23:34.54 | cyborg_ar | you have grep |
| 23:35.00 | archivist | cyborg_ar, I do believe this is all servo https://www.youtube.com/watch?v=mxxdq6y8z8M |
| 23:35.04 | cyborg_ar | how are you running git with no bash or grep |
| 23:35.15 | decimad | right, I've got grep |
| 23:35.27 | decimad | I never asked for grep! |
| 23:35.31 | cyborg_ar | then stop whining and write your shell script |
| 23:35.42 | decimad | Actually I'm laughing ;) |
| 23:37.59 | cyborg_ar | one of GyrosGeier's worries was of power issues causing uncommanded (non) motions, adding closed loop feedback to the motors would allow you to detect that immediately, regardless of cause, if the difference between commanded position and actual possition differs for over a couple steps, estop |
| 23:38.29 | cyborg_ar | that would catch all problems from power supply issues, to controller failures and wiring problems |
| 23:38.57 | decimad | that's why I'd never let a blind stepper on my table ;) |
| 23:39.08 | cyborg_ar | since this is just a safety feature you don't even need to use super high resolution encoders |
| 23:39.14 | cyborg_ar | just keep the following error in check |
| 23:39.54 | archivist | linuxcnc catches that with servo control (ferror limit) |
| 23:40.03 | decimad | that would actually be open loop feedback though? |
| 23:40.03 | cyborg_ar | yeah |
| 23:40.13 | cyborg_ar | decimad, sorta |
| 23:40.25 | *** join/#kicad mroszko__ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 23:40.28 | cyborg_ar | the stepper motor closes the "loop" |
| 23:40.49 | cyborg_ar | but true, you dont act on the feedback except for shutdown |
| 23:41.11 | decimad | I have a few AMS magnetic encoders here, they work well, but only the new ones support 16bit transfers, and stm only supports 8/16bit -.- |
| 23:41.12 | archivist | steppers are open loop! |
| 23:41.28 | cyborg_ar | yes they are |
| 23:41.29 | decimad | archivist you can surely do closed loop stepping |
| 23:41.38 | cyborg_ar | and yes you can |
| 23:41.46 | cyborg_ar | but the motors are open loop of course |
| 23:41.51 | archivist | encoders close the loop, and yes people have used encoder feedback on steppers |
| 23:41.54 | cyborg_ar | you need a sensing element to close the loop |
| 23:42.11 | archivist | steppers cog though |
| 23:42.12 | cyborg_ar | in this case i propose an encoder as just a safety device |
| 23:42.29 | cyborg_ar | to evaluate the ferror |
| 23:42.55 | archivist | some use dual feedback encoder and linear scale |
| 23:43.46 | cyborg_ar | it should protect you from all unconmmanded motions, as long as you make the estop state truly fail safe |
| 23:47.40 | cyborg_ar | then on the charge pump side, you need a supervisory circuit, a dumb tone detector triggers estop if the controller hardware stops emiting a tone of the right frequency. on the controller side, if you're not using linuxcnc you need to toggle a pin every servo loop cycle after doing sanity checks, put all your software effort in making sure that the cpu cannot get stuck in a state where it will toggle the pin at exactly the right |
| 23:47.40 | cyborg_ar | frequency for the tone detector to keep plowing |
| 23:49.46 | archivist | that is only the safety stuff, the other part is also the trajectory planning, coordinated movements, and kinematics |
| 23:50.07 | *** join/#kicad bronson (~bronson@50-1-51-14.dsl.dynamic.fusionbroadband.com) |
| 23:51.04 | decimad | hrmmm, actually I don't see a hook that would match the events I need... I would need a post-fetch or something |
| 23:51.44 | cyborg_ar | decimad, merge hook? |
| 23:51.56 | cyborg_ar | when you pull you merge |
| 23:52.32 | decimad | I thought I wouldn't... Don't I rather fetch and rebase? |
| 23:52.49 | cyborg_ar | then you need to run the hook on the rebase |
| 23:53.00 | cyborg_ar | because that's when your tree changes |
| 23:53.50 | GyrosGeier | cyborg_ar, I think watching the motors' power draw and checking that the optocouplers work should be mostly sufficient |
| 23:54.36 | GyrosGeier | but an optical encoder would indeed be a nice addon |
| 23:54.37 | archivist | GyrosGeier, that misses a steppers most common problem |
| 23:54.59 | GyrosGeier | skipping |
| 23:55.01 | GyrosGeier | yep |
| 23:55.04 | archivist | yes |
| 23:55.18 | GyrosGeier | that shouldn't happen though |
| 23:55.19 | cyborg_ar | and the encoder is a great catch all |
| 23:55.24 | GyrosGeier | hmm |
| 23:55.29 | archivist | it will happen :) |
| 23:55.35 | GyrosGeier | well |
| 23:55.36 | cyborg_ar | would allow you to catch all the issues you try to catch at the sources |
| 23:55.49 | GyrosGeier | not all of them |
| 23:56.00 | archivist | trust me it does happen |
| 23:56.16 | GyrosGeier | the most important is to kill power to the brake if the Z axis power supply fails |
| 23:56.18 | *** join/#kicad bombledmonk (~bombledmo@thief-pool3-123-31.mncable.net) |
| 23:56.25 | GyrosGeier | because the Z axis falls down then |
| 23:56.29 | archivist | just needs a stiff axis to cause a stall |
| 23:56.34 | cyborg_ar | GyrosGeier, that's what the encoder catches |
| 23:56.42 | GyrosGeier | cyborg_ar, not quickly enough |
| 23:56.49 | cyborg_ar | GyrosGeier, plenty quick |
| 23:57.03 | cyborg_ar | it will catch it as soon as it does 1 whole step of uncommanded motion |
| 23:57.05 | archivist | it will be ms to catch |
| 23:57.12 | cyborg_ar | boom |
| 23:57.15 | GyrosGeier | I still have 200kg of inertia when the power fails |
| 23:57.16 | cyborg_ar | lockout |
| 23:57.21 | GyrosGeier | when I see that moving... |
| 23:57.32 | archivist | that needs a brake |
| 23:57.37 | cyborg_ar | GyrosGeier, slam on the brakes is the best you can do anyway |
| 23:57.44 | GyrosGeier | archivist, it has a brake |
| 23:57.50 | cyborg_ar | GyrosGeier, ^ |
| 23:57.59 | GyrosGeier | I just need to disengage it when moving the Z axis |
| 23:58.13 | GyrosGeier | which is also the most likely point for the PSU to fail |
| 23:58.15 | cyborg_ar | it will slam on the brakes and cut power to everything as soon as the z axis has an uncommanded motion |
| 23:58.31 | archivist | pwer fail will drop in the brake any way as it should be powered to open |
| 23:58.51 | GyrosGeier | the Z axis has no accessible part though |
| 23:59.07 | GyrosGeier | and it's 100 steps/mm |
| 23:59.29 | GyrosGeier | well |
| 23:59.34 | archivist | steps per mm not relevant to safety |
