| 00:06.18 | GyrosGeier | hm |
| 00:06.31 | GyrosGeier | msys has massive overhead during builds, it seems |
| 00:08.02 | *** join/#kicad bugzc (~1@unaffiliated/bugzc) |
| 00:13.35 | sibiria | mingw cleaner environment |
| 00:14.11 | GyrosGeier | these are the mingw toolchains |
| 00:28.06 | DocScrutinizer05 | https://community.giffgaff.com/t5/General-Discussion/Anyone-heard-of-Copperhead-OS-Android/m-p/18387867#M2874952 |
| 00:29.33 | *** join/#kicad DL7AD (~sven@p4FD408EC.dip0.t-ipconnect.de) |
| 00:35.34 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 01:18.22 | *** join/#kicad alphaferret (~user@54034F08.catv.pool.telekom.hu) |
| 01:27.41 | *** join/#kicad VA3VNA (~mayday_ja@67.70.9.37) |
| 01:41.40 | *** join/#kicad crazy_imp (~mj@unaffiliated/crazy-imp/x-8704303) |
| 01:45.20 | *** join/#kicad eszett (~Eszett@dslb-084-059-074-203.084.059.pools.vodafone-ip.de) |
| 01:45.22 | eszett | hi |
| 01:45.58 | eszett | the word "footprint" doesnt appear in a 900 pages datasheet of an IC :-( |
| 01:46.27 | eszett | reality and datasheets are two different worlds, I suppose.. |
| 01:46.35 | c4757p | look for "package dimensions" |
| 01:46.39 | eszett | ok |
| 01:46.53 | *** join/#kicad kyranf (~kyranf@S0106bcd165673658.ok.shawcable.net) |
| 01:47.19 | c4757p | as for the actual /footprint/ (typically called a "land pattern"), that's usually left out of large datasheets and provided separately on the manuf website |
| 01:47.31 | c4757p | just google the name of the manuf and their silly custom name for the package |
| 01:47.35 | c4757p | it usually comes up :p |
| 01:47.37 | eszett | ye c47, i found the package dimensions, http://cache.freescale.com/files/shared/doc/package_info/98ASH00962A.pdf |
| 01:47.44 | GyrosGeier | hm |
| 01:47.47 | eszett | but it has no footprint dimensions :-( |
| 01:47.53 | eszett | for the land patterns.. |
| 01:48.11 | c4757p | (alternatively they may not provide it at all. that is to some extent considered the responsibility of the PCB designer.) |
| 01:48.16 | GyrosGeier | mechanical question: I have two plastic parts with holes, one of them is the crank of a servo, the other an extension rod |
| 01:48.24 | GyrosGeier | what is the best way to connect them? |
| 01:48.28 | GyrosGeier | wire? |
| 01:48.38 | eszett | c47: oh no, they dont provide it at all? |
| 01:49.26 | c4757p | they may or may not |
| 01:49.29 | c4757p | some do. |
| 01:49.39 | c4757p | > as for the actual /footprint/ (typically called a "land pattern"), that's usually left out of large datasheets and provided separately on the manuf website |
| 01:49.44 | c4757p | > just google the name of the manuf and their silly custom name for the package |
| 01:51.37 | *** part/#kicad hasuTMK (~hasu@58x13x224x139.ap58.ftth.ucom.ne.jp) |
| 01:52.58 | eszett | ye, got it :-) |
| 01:53.52 | eszett | c47: I guess i can somehow guess the land patterns by the details given for "base metal", here http://84.59.74.203/98ASH00962A.pdf, but can you tell me what's the difference between the yellow and the red values? |
| 01:54.29 | DocScrutinizer05 | http://cache.freescale.com/files/shared/doc/package_info/98ASH00962A.pdf is particularly nasty |
| 01:55.06 | eszett | nasty in which respect? hard to read? |
| 01:55.56 | c4757p | I'm not entirely sure. min/max at a guess? |
| 01:56.27 | DocScrutinizer05 | nasty in that it lacks any decent data for land pattern, not even for the solder area of the pins |
| 01:56.58 | eszett | yes, i've seen better made technical drawings.. |
| 01:57.19 | c4757p | so many are terrible though |
| 01:57.25 | c4757p | when I see a good one I get excited |
| 01:57.34 | eszett | c47: min/max, ye i guess that probably is it |
| 02:05.46 | DocScrutinizer05 | you possibly are better off searching for a standard footprint for that package, that seems to be what they expect you to do anyway |
| 02:06.25 | Plugh | eszett: Did you get the PCB for your keyboard finished? |
| 02:06.50 | DocScrutinizer05 | the footer prolly has best info of whole sheet |
| 02:08.02 | DocScrutinizer05 | LQFP 48 pin, JEDEC MS-026-BBC, SOT313-3. One of those might help |
| 02:08.58 | c4757p | all that and they couldn't be arsed to include a fucking IPC footprint name. I /really/ wish more people used those |
| 02:09.06 | c4757p | who the hell uses SOT* for QFPs?? |
| 02:09.45 | DocScrutinizer05 | http://www.nxp.com/packages/SOT313-2 Footprint for wave soldering http://cache.nxp.com/documents/wave_soldering/HTQFP-HLQFP-LQFP-MSQFP-WAVE.pdf |
| 02:10.01 | c4757p | at least IPC names are regular enough to figure out yourslef |
| 02:10.04 | c4757p | self |
| 02:10.17 | c4757p | QFP50P900X900X160-48 |
| 02:10.24 | c4757p | ^-^ |
| 02:16.06 | DocScrutinizer05 | http://www.infineon.com/dgdl/Processing_M-T-LQFP.pdf?folderId=db3a304412b91b910112b9a6854e000e&fileId=db3a304412b91b910112b9a6b54e004b |
| 02:16.31 | eszett | sorry i was it the basement looking for my cat |
| 02:17.12 | c4757p | did you find it? |
| 02:18.09 | DocScrutinizer05 | ^^^ is a good document. |
| 02:18.19 | DocScrutinizer05 | infineon FTW |
| 02:18.44 | eszett | c47: she made some noise probably was catching a mouse, dont know exactly |
| 02:19.00 | *** join/#kicad Eli2| (~Eli2@pdpc/supporter/professional/eli2) |
| 02:21.42 | eszett | docscrutinizer: browsing your given datasheets.. |
| 02:22.07 | eszett | well, i think i got the footprint ready now |
| 02:22.33 | c4757p | hm |
| 02:22.36 | c4757p | puts on his GyrosGeier hat |
| 02:22.42 | c4757p | I want a better way to do decoupling capacitors |
| 02:22.52 | eszett | my fabhouse said i can send the gerbers with 0.00 solder mask clearance to them. they adjust the value to their needs then. |
| 02:22.56 | GyrosGeier | lol |
| 02:23.04 | c4757p | I don't want to have to litter the schematic with a thousand individual capacitors |
| 02:23.20 | GyrosGeier | c4757p, I'm thinking of inventing a new schematic item |
| 02:23.23 | GyrosGeier | "table" |
| 02:23.29 | GyrosGeier | rendered as a table |
| 02:23.32 | c4757p | I want to just be able to say "C1000-C1200 are decoupling capacitors with these properties", and then plonk as many as I want onto the PCB |
| 02:23.35 | GyrosGeier | and creates netlist items |
| 02:23.38 | c4757p | GyrosGeier: ooh |
| 02:23.42 | c4757p | YES |
| 02:23.49 | c4757p | yesyesyesyesyesyes |
| 02:23.51 | GyrosGeier | that could also replace hidden pins |
| 02:23.58 | c4757p | yesyesyesyesyesyes |
| 02:26.05 | GyrosGeier | but I need to finish the other stuff first |
| 02:26.18 | cyborg_ar | lol |
| 02:27.15 | GyrosGeier | is building a GLaDOS lamp |
| 02:27.21 | GyrosGeier | http://ytec3d.com/glados-lamp/ |
| 02:27.37 | GyrosGeier | that's why I need the LT8500 |
| 02:27.53 | GyrosGeier | 9 RGBW LEDs and four servos |
| 02:28.42 | c4757p | that is quite the project |
| 02:29.07 | GyrosGeier | the 3d printing is done |
| 02:29.26 | GyrosGeier | now it's servo mechanics |
| 02:32.54 | GyrosGeier | is a tiny bit unhappy that the LT8500 needs an external clock |
| 02:33.25 | GyrosGeier | but I think I can derive that from the Arduino's core clock |
| 02:35.48 | GyrosGeier | hm |
| 02:36.21 | GyrosGeier | when I control the servos with the same PWM generator as the LEDs, I'm limited to 200 Hz for the LEDs |
| 02:38.08 | DocScrutinizer05 | c4757p: (("C1000-C1200 are decoupling capacitors with these properties")) isn't that what a component in lib does for you? |
| 02:38.18 | c4757p | what |
| 02:38.35 | c4757p | I mean a single thing that reserves that entire range |
| 02:38.37 | DocScrutinizer05 | component "decoupling capacitor" |
| 02:38.57 | c4757p | so I don't need to make a sea of 200 capacitors in the schematic to do that |
| 02:39.00 | DocScrutinizer05 | call it DC instead C? |
| 02:39.01 | c4757p | which serves no purpose |
| 02:39.04 | c4757p | ...what |
| 02:39.16 | c4757p | gives up |
| 02:39.16 | DocScrutinizer05 | aaah, you don't want to place them? |
| 02:39.21 | c4757p | not in the schematic, no |
| 02:39.36 | c4757p | I want to place /one object/ that reserves the /entire C1000-C1200 range/ |
| 02:39.44 | c4757p | and then just place parts out of that range dynamically onto the PCB |
| 02:39.56 | DocScrutinizer05 | hmmm, a tiny component maybe? with GND connection already integrated |
| 02:40.20 | wpwrak | that could be cute :) |
| 02:40.26 | c4757p | I don't care about the size, I'll make a dedicated sheet to hold them if I bloody have to |
| 02:40.32 | c4757p | I want two things |
| 02:40.36 | c4757p | 1) not to have to place them at /all/ |
| 02:40.52 | wpwrak | reminds me of the suggestion to manufacture diodes with built-in resistors, for the "maker" movement ;) |
| 02:40.54 | c4757p | 2) to reduce the number of silly things in my schematic. I'm /picky/ about them, I want them drawn a certain way |
| 02:41.01 | DocScrutinizer05 | somehow you have to define where they connect to, though |
| 02:41.05 | c4757p | sure |
| 02:41.12 | c4757p | "on this rail" is part of "these properties" |
| 02:41.22 | c4757p | obv the entire range is on the same rail |
| 02:41.26 | GyrosGeier | and ideally, the ratsnest lines would connect each power pin to its dedicated capacitor |
| 02:41.32 | GyrosGeier | so you don't lose track |
| 02:41.33 | c4757p | wpwrak: LEDs, you mean? |
| 02:41.51 | DocScrutinizer05 | aaah I see, for a power rail they are all basically same |
| 02:41.54 | c4757p | yes |
| 02:41.55 | wpwrak | c4757p: err, yes :) |
| 02:42.01 | c4757p | wpwrak: that's been done, a while ago |
| 02:42.22 | c4757p | I remember seeing those in ratshack |
| 02:42.25 | wpwrak | ;-) |
| 02:42.49 | c4757p | GyrosGeier: nah, I wouldn't want that |
| 02:42.59 | c4757p | too much maintenance making sure each power pin is associated with one |
| 02:43.11 | c4757p | decoupling is a layout thing anyway, I'll see there's a power pin lacking a cap at layout |
| 02:43.31 | c4757p | (and if two power pins are directly adjacent.../really/, there's no reason to place /two/ there, despite "one cap per pin" rules of thumb) |
| 02:44.24 | DocScrutinizer05 | so crreating a 200 identical capacitors on one sheet shouldn't be too hard, eh? |
| 02:44.38 | c4757p | not /too/ hard, I mean...I've obv done it |
| 02:44.43 | c4757p | it just seems silly to me |
| 02:45.18 | c4757p | if we made circuits with code instead of diagrams, nobody would copypaste a "decouple(power, gnd)" 200 times |
| 02:45.37 | c4757p | they'd just use whatever that language had for "for (i = 0; i < 200; ++i) decouple(power, gnd)" |
| 02:45.46 | DocScrutinizer05 | what do you eant? place *one* object is not exactly meaningful, unless that one object is a component considting of 100 capacitors |
| 02:45.53 | c4757p | it /is/ |
| 02:46.49 | *** join/#kicad Wetmelon (~wetmelon@host-73-169.ncasgro.asheville.nc.us.clients.pavlovmedia.com) |
| 02:46.56 | c4757p | basically, I want this "object" to be something akin to a text object that says "C1000..C1200: 100nF 0201" like used to be seen on well-drawn, /hand/-drawn schematics before CAD entered into this and ruined all the old readability tricks |
| 02:46.57 | DocScrutinizer05 | and yes, I instantly thought of a script asking you "how many? which capacity, voltage, footprint?" and then does the job for you |
| 02:48.11 | DocScrutinizer05 | alas... you know. eeschema and scripting :-/ |
| 02:48.15 | c4757p | I don't think it needs to be a script |
| 02:48.26 | c4757p | and we don't do scripting well, I /really/ don't want it to be one in /kicad/ |
| 02:48.48 | c4757p | all it /has/ to be is a component whose reference is a /range/ rather than a single point |
| 02:48.59 | wpwrak | c4757p: trance mode may be more efficient in the end :) the sort of mental state where you switch off all brain functions that get in in the way, and 'just do it' :) |
| 02:49.32 | c4757p | I don't have the bloody attention span for that |
| 02:49.43 | DocScrutinizer05 | 2^N gets you to a 200 capacitors pretty quickly |
| 02:49.46 | c4757p | if I have to manually place 200 capacitor symbols I'm going to start daydreaming of suicide methods by the hundredth |
| 02:49.51 | GyrosGeier | c4757p, I have an FPGA that needs 50 small caps |
| 02:49.56 | GyrosGeier | it's easy to miss one |
| 02:50.18 | c4757p | I have three of those :) |
| 02:50.36 | DocScrutinizer05 | I'll try how long it takes me... ;-P |
| 02:50.48 | c4757p | oh, sure, 2^N is efficient, and obviously I do that |
| 02:50.57 | c4757p | still, it's /silly/. it's redundant, and looks stupid on the schematic |
| 02:51.05 | c4757p | worst of all for me, it's /not how I envision my schematic/ |
| 02:51.20 | c4757p | IMO, if I can't make my schematic /exactly how I want it to look/, the tool is missing something |
| 02:51.21 | c4757p | it's a drawing |
| 02:51.27 | c4757p | I know better than the tool how to make those |
| 02:52.08 | wpwrak | c4757p: here's my secret: dinner consisting of crackers, tuna, some nice sauce, and a bottle of sweet white wine. when go to bed. when you wake up, you're clear in the head, relaxed, and excellently motivated. go easy on the caffeine, though. also a great condition for cracking some tough problems, since it boosts creativity (in a productive way). |
| 02:52.34 | wpwrak | 1st s/when/then/ |
| 02:58.11 | DocScrutinizer05 | as I understand it you could add components at will to pcbnew layout, no? - unless you get the error msg I just got while trying: http://wstaw.org/m/2016/08/28/plasma-desktopxJ2277.png |
| 02:58.33 | c4757p | you can, yes |
| 02:58.43 | c4757p | but managing parts on the PCB that don't match anything in the schematic is a PITA |
| 02:59.12 | c4757p | (and I don't want them /missing/ in the schematic, I just want to say for (i=0; i<200; ++i) decouple_shit() instead :) |
| 02:59.19 | DocScrutinizer05 | then that's the root problem, since you don't want those components to show up in schematics |
| 02:59.26 | c4757p | I /do/ want them to show up |
| 02:59.29 | c4757p | just not individually. |
| 02:59.30 | cyborg_ar | c4757p, :) |
| 02:59.58 | c4757p | I draw schematics for /humans/, not for computers. the computer reading it as a sort of graphical netlist is secondary to me |
| 03:00.03 | DocScrutinizer05 | you need a new class "joker component" |
| 03:00.09 | c4757p | you don't include piles of redundant info like that in diagrams for humans |
| 03:00.22 | cyborg_ar | c4757p, one thing i miss is being able to explode connectors in arbitrary pieces |
| 03:00.27 | c4757p | ^^^^^^^^^^^^ |
| 03:00.34 | c4757p | basically |
| 03:00.44 | c4757p | go look at a Tektronix service manual from the 1970s or 1980s |
| 03:00.51 | c4757p | anything in there that you can't do in eeschema |
| 03:00.57 | c4757p | I want to be able to do in eeschema |
| 03:01.01 | c4757p | :P |
| 03:01.36 | cyborg_ar | http://i.imgur.com/rcXVB8d.png this is from the embedded system i was talking about earlier today |
| 03:01.56 | c4757p | ooh! |
| 03:02.11 | c4757p | that's the first time I've actually seen that alternative Darlington symbol in application |
| 03:02.41 | c4757p | is easily distracted by schematic details |
| 03:02.50 | DocScrutinizer05 | oh wow, that's some awesome work |
| 03:03.19 | cyborg_ar | http://i.imgur.com/qCgfRjO.png this is the CPU board |
| 03:03.31 | cyborg_ar | it would be a lot more complicated with un-exploded connectors |
| 03:03.39 | c4757p | yeah, and it's already pretty busy |
| 03:03.50 | cyborg_ar | it should have been split in a couple pages lol |
| 03:03.54 | c4757p | I'd want more whitespace in there myself - but it's hand-drawn, so... |
| 03:04.04 | wpwrak | cyborg_ar: what would already be nice is to have a way to say that a component is to be guaranteed to be equivalent to another. i.e., that it would show up in ERC if they diverge. that way, you'd still have manual work, but at least you don't need to review N times. |
| 03:04.31 | cyborg_ar | it was originally A3 size |
| 03:04.56 | cyborg_ar | but yeah this could easily be 3 pages |
| 03:05.15 | c4757p | the exploded connectors are a readability tradeoff, though |
| 03:05.19 | cyborg_ar | but you can still follow it pretty well when debugging the board |
| 03:05.20 | DocScrutinizer05 | ((CPU board)) LOL |
| 03:05.23 | c4757p | it's much harder to work backwards and find out what pin 42 connects to |
| 03:05.32 | cyborg_ar | ah |
| 03:05.43 | cyborg_ar | but that was solved on the electrical plans of that same company |
| 03:05.50 | cyborg_ar | i dont have any on me unfortunately |
| 03:06.03 | cyborg_ar | they had dozens of connectors exploded all over the place |
| 03:06.20 | cyborg_ar | but at the end they had a table where they listed all the pages and coordinates where the ends were |
| 03:06.28 | c4757p | yeah, I was just going to say |
| 03:06.46 | c4757p | I've never seen this done for /connectors/, but that's an approach I like in general to resolve that sort of tradeoff |
| 03:06.47 | cyborg_ar | so if you wanted to know what was connected to J5 pin 22, you went to the tables at the end, find the page and coordinate, and then go to the right page |
| 03:07.02 | c4757p | optimize the schematic for reading in the direction it'll be most commonly read in, and then have a table for working it "backwards" |
| 03:07.30 | cyborg_ar | well, the most common way for us to read it was "backwards" so we always went to the tables first |
| 03:07.49 | cyborg_ar | they also had pictorial pages showing the physical pinout of the connector |
| 03:07.55 | c4757p | nods |
| 03:08.27 | cyborg_ar | so what we would do is first go the the pictorial page, check which wire is which pin number, the pictorial page also had coordinates and machine registers |
| 03:08.48 | cyborg_ar | so we could punch in the debug code in the console to test that IO pin |
| 03:16.32 | cyborg_ar | they spoiled me :( |
| 03:19.44 | *** join/#kicad Gud_ (~k@c-890ae555.021-17-73746f36.cust.bredbandsbolaget.se) |
| 03:20.29 | *** join/#kicad smoores (~smoores@c-67-188-210-139.hsd1.ca.comcast.net) |
| 03:20.57 | *** join/#kicad jhelwig_ (~jhelwig@50-39-183-7.bvtn.or.frontiernet.net) |
| 03:20.57 | *** join/#kicad jhelwig_ (~jhelwig@puppetlabs/development/jhelwig) |
| 03:21.32 | *** join/#kicad dashie (~someotter@octavia.celestia.licorn.eu) |
| 03:21.32 | *** join/#kicad dashie (~someotter@leloop/rhaamo) |
| 03:21.51 | *** join/#kicad esden_ (esden@repl.esden.net) |
| 03:22.16 | *** join/#kicad invultri_ (~leila@a82-95-232-65.adsl.xs4all.nl) |
| 03:23.06 | *** join/#kicad mkj_ (~johnsonm@people.danlj.org) |
| 03:23.24 | *** join/#kicad ants` (~nats@2001:4b98:dc0:41:216:3eff:fe8f:4e6f) |
| 03:23.25 | *** join/#kicad RaYmAn_ (rayman@rayman.dk) |
| 03:23.29 | *** join/#kicad flummer_ (~tf@web11.oemhost.net) |
| 03:23.33 | *** join/#kicad kakimir (kakimir@sienimetsa.wtf) |
| 03:24.18 | *** join/#kicad awallin____ (awallin@lakka.kapsi.fi) |
| 03:24.44 | *** join/#kicad kmehall_ (~quassel@droplet.kevinmehall.com) |
| 03:25.12 | *** join/#kicad nikomo_ (~quassel@nikomo.eu) |
| 03:27.42 | *** join/#kicad MrFahrenheit (~hamzadis@unaffiliated/rageofthou) |
| 03:27.44 | *** join/#kicad c4757p (c4757p@2604:a880:0:1010::631:8001) |
| 03:29.10 | *** join/#kicad Novus (~Novus@2605:e000:628b:1400:bd06:834a:37ed:fa6a) |
| 03:31.11 | DocScrutinizer05 | hmm, a lot of ping timeouts at 3:25 UTC |
| 03:31.54 | DocScrutinizer05 | c4757p: netsplit? |
| 03:32.11 | *** join/#kicad sftp (~sftp@unaffiliated/sftp) |
| 03:32.31 | c4757p | no, I dropped out completely. not sure what happened there |
| 03:32.48 | c4757p | figured it was on my end, didn't know it was a bunch |
| 03:32.51 | DocScrutinizer05 | (I only see parts of users who said something recently, so can't tell) |
| 03:33.27 | DocScrutinizer05 | [2016-08-28 Sun 05:25:59] <-- [Tux] (~H_Merijn@2001:981:9fe5:1:1826:a33b:8d85:a9a7) has left this server (Ping timeout: 258 seconds). (in #suse) |
| 03:33.52 | DocScrutinizer05 | [2016-08-28 Sun 05:25:59] <-- c4757p (c4757p@2604:a880:0:1010::631:8001) has left this server (Ping timeout: 258 seconds). |
| 03:34.28 | DocScrutinizer05 | common denominator: IPv6 ? |
| 03:34.35 | c4757p | ¯\_(ã)_/¯ |
| 03:34.50 | DocScrutinizer05 | nevermind |
| 03:35.07 | DocScrutinizer05 | is too curious |
| 03:35.35 | c4757p | I'm curious too, I just don't know what actually happened :p |
| 03:35.44 | c4757p | one of freenode's servers farted? |
| 03:36.26 | DocScrutinizer05 | did you ever hear of #fnalerts |
| 03:36.32 | c4757p | no |
| 03:37.45 | DocScrutinizer05 | pro tip: query the bot directly |
| 03:37.57 | DocScrutinizer05 | status |
| 03:38.45 | c4757p | unfortunately I'm not actually curious enough to do something to find out |
| 03:38.53 | c4757p | though admittedly that may be more a problem of laziness than curiosity |
| 03:39.05 | DocScrutinizer05 | it just tells you which servers are down, if any |
| 03:39.43 | DocScrutinizer05 | no downed servers, so no clue what happened at 3:25 |
| 03:40.58 | *** join/#kicad Shuggsy (~Shuggsy@c-73-7-71-211.hsd1.ga.comcast.net) |
| 03:41.49 | *** join/#kicad forrestv (forrestv@unaffiliated/forrestv) |
| 03:42.08 | DocScrutinizer05 | ((lazy)) I got a bookmark on #fnalerts ;-) |
| 03:42.49 | DocScrutinizer05 | only since bookmarks to queries don't work in my irc client. But just it occurs to me I could create a macro |
| 03:44.18 | DocScrutinizer05 | /msg fnalerts status |
| 03:44.23 | DocScrutinizer05 | :-) |
| 03:48.26 | DocScrutinizer05 | anyway funny, found another one: [2016-08-28 Sun 05:25:59] <-- azonenberg (~azonenber@chat.dmz.bainbridge.antikernel.net) has left this server (Ping timeout: 258 seconds). |
| 03:49.21 | DocScrutinizer05 | FN server fart sounds a reasonable explanation :-D |
| 03:58.09 | *** join/#kicad smoores (~smoores@c-67-188-210-139.hsd1.ca.comcast.net) |
| 04:01.42 | *** join/#kicad DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 04:06.40 | *** join/#kicad Novus (~Novus@2605:e000:628b:1400:bd06:834a:37ed:fa6a) |
| 04:06.56 | GyrosGeier | 20562 read(12, "digit\0__weisswurschtbrotzeit13\0A"..., 4096) = 4096 |
| 04:07.23 | GyrosGeier | debugging commercial code is fun |
| 04:07.53 | *** join/#kicad daey_ (~Flutterba@unaffiliated/day) |
| 04:08.56 | GyrosGeier | 20553 --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x2b59cffd} --- |
| 04:09.03 | GyrosGeier | -_- |
| 04:17.32 | *** join/#kicad jaggz (~jaggz@unaffiliated/jaggz) |
| 04:29.17 | DocScrutinizer05 | o.O |
| 04:34.29 | GyrosGeier | ModelSim fails |
| 04:34.40 | GyrosGeier | but only when invoked from Quartus |
| 04:43.30 | *** join/#kicad Tachyon` (tachyon@yuna.autie.net) |
| 04:44.34 | *** join/#kicad KarlZeilhofer (~Thunderbi@93-82-77-167.adsl.highway.telekom.at) |
| 05:06.44 | c4757p | woo |
| 05:06.50 | c4757p | fixed the corner artifacts issue in the raytracer |
| 05:11.21 | *** join/#kicad firebird_ (~firebird@125.16.18.198) |
| 05:18.10 | c4757p | lol, found the /actual/ problem |
| 05:18.28 | c4757p | fixed it by progressively untangling spaghetti code until it stopped being a problem, didn't know what was actually wrong >_> |
| 05:19.06 | c4757p | protip: std::vector<bool> is special-cased... >_> |
| 05:32.21 | GyrosGeier | learned something about servos |
| 05:32.32 | GyrosGeier | they are strong enough to destroy themselves |
| 05:33.04 | c4757p | oopsie |
| 05:35.37 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 05:55.37 | *** join/#kicad riq_ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 06:05.32 | *** join/#kicad mentar (~quassel@38.ip-51-254-125.eu) |
| 06:07.23 | *** join/#kicad mroszko_ (~mroszko@ool-4350d3a1.dyn.optonline.net) |
| 06:10.34 | *** join/#kicad ioch (~gyvunas@unaffiliated/ioch) |
| 06:16.46 | c4757p | ooh |
| 06:16.58 | c4757p | I like that the bugs are automatically marked as Fix Committed now |
| 06:17.09 | c4757p | thanks orsonmmz|away :) |
| 06:30.39 | *** join/#kicad Thor-Arne (~email@34.79-161-66.customer.lyse.net) |
| 06:48.48 | *** join/#kicad ioch (~gyvunas@unaffiliated/ioch) |
| 06:55.27 | *** join/#kicad rubdos (~rubdos@host-85-27-50-55.dynamic.voo.be) |
| 06:58.52 | *** join/#kicad tannewt (~textual@c-71-231-55-126.hsd1.wa.comcast.net) |
| 07:02.54 | *** join/#kicad berjan (~chatzilla@ip5455a1d7.adsl-surfen.hetnet.nl) |
| 07:27.54 | *** join/#kicad C47 (~Xwx47@189.242.13.67) |
| 07:36.26 | *** join/#kicad rohitksingh (~rohit@14.96.122.184) |
| 07:39.32 | cyborg_ar | GyrosGeier, my client learned that the other day, it will not only kill itself but take power supplies with it |
| 07:42.36 | GyrosGeier | well, it's a microservo |
| 07:42.58 | GyrosGeier | I'm a bit unhappy with the resolution though |
| 07:43.30 | archivist | change the encoder |
| 07:43.33 | GyrosGeier | I'd love to have smooth movements for the lamp |
| 07:43.46 | GyrosGeier | the encoder is an FPGA |
| 07:44.13 | GyrosGeier | I have 18 bit resolution in the encoder now |
| 07:44.19 | archivist | what is it reading for position |
| 07:44.44 | GyrosGeier | I just increment a counter by 25 each frame |
| 07:45.02 | GyrosGeier | frames are 125 Hz, roughly |
| 07:46.08 | *** join/#kicad stelarcf (~stelarcf@92.81.84.163) |
| 07:46.14 | GyrosGeier | but the movement is still jerky |
| 07:46.55 | cyborg_ar | did you grapth the error vs time? |
| 07:47.06 | GyrosGeier | not yet |
| 07:47.10 | cyborg_ar | also 125hz might be too slow for smooth motion |
| 07:47.21 | GyrosGeier | not sure I can go faster |
| 07:47.23 | cyborg_ar | my servo system used a 2000Hz torque loop |
| 07:47.30 | GyrosGeier | tries |
| 07:47.39 | cyborg_ar | watch the integral though |
| 07:48.05 | GyrosGeier | no, I'm trying linear motion |
| 07:48.11 | GyrosGeier | no PID yet |
| 07:48.17 | cyborg_ar | just P? |
| 07:48.21 | GyrosGeier | yep |
| 07:48.26 | cyborg_ar | you need to at least have PD |
| 07:49.01 | GyrosGeier | http://paste.debian.net/792861/ |
| 07:49.15 | cyborg_ar | give it enough D that it will feel like it's resisting when you try to move it, then rise the P until it gets there fast enough |
| 07:49.22 | cyborg_ar | shitty tunning but works well enough |
| 07:50.01 | GyrosGeier | it's one of the PWM controlled microservos |
| 07:50.23 | GyrosGeier | "move to position" is fast enough |
| 07:51.30 | cyborg_ar | so this is bang bang, will give you a 1 until it reaches position? |
| 07:51.50 | GyrosGeier | no, this is "1.5ms pulse encodes middle position" |
| 07:52.45 | GyrosGeier | 0.5 to 2.5 seem to be the value range for this one |
| 07:52.52 | cyborg_ar | sorry, i dont really understand HDLs very well yet |
| 07:52.57 | archivist | RC servo? |
| 07:53.00 | GyrosGeier | yes |
| 07:53.04 | archivist | toy |
| 07:53.10 | GyrosGeier | sure |
| 07:53.17 | cyborg_ar | you can't tune those very well |
| 07:53.24 | archivist | servo to me is a real thing on a cnc :) |
| 07:53.25 | GyrosGeier | but then again I need to move 16g of PLA and a LED |
| 07:53.30 | cyborg_ar | i thought you had motors and encoders |
| 07:53.46 | decimad | any reason not to use a process? |
| 07:53.57 | archivist | get a real one, can be simple |
| 07:54.04 | GyrosGeier | decimad, it's throwaway code |
| 07:54.15 | decimad | okay |
| 07:54.18 | decimad | just wondered |
| 07:54.19 | cyborg_ar | yeah you might get a lot better performance if you replace the controller board in the servo |
| 07:54.30 | GyrosGeier | normally I wouldn't divide the clock by hand if I could avoid it |
| 07:54.33 | cyborg_ar | those RC servos are usually simple PD thingies |
| 07:54.47 | archivist | with a pot for feedback |
| 07:54.52 | cyborg_ar | yeah |
| 07:54.55 | cyborg_ar | and deadband |
| 07:55.00 | cyborg_ar | to prevent oscillation |
| 07:55.10 | GyrosGeier | ah |
| 07:55.14 | cyborg_ar | very shitty tunning and positional accuracy |
| 07:55.18 | GyrosGeier | the deadband is the problem then |
| 07:55.30 | cyborg_ar | but good enough for someone who's controlling ailerons with an RC stick |
| 07:55.46 | archivist | not for real work though |
| 07:55.52 | GyrosGeier | hm |
| 07:56.00 | GyrosGeier | I can still change the design for the controller |
| 07:56.11 | GyrosGeier | so I can still change servo type |
| 07:56.35 | cyborg_ar | GyrosGeier, do you have your own encoders, or just rely on the pot? |
| 07:56.41 | archivist | I did have to make replacement gears for some, a certain motorcycle company was using them in a racing bike on the throttle |
| 07:56.45 | GyrosGeier | cyborg_ar, just the pot |
| 07:56.52 | cyborg_ar | ah |
| 07:57.06 | decimad | does gcc fill up deleted memory with a pattern in debug builds? |
| 07:57.10 | GyrosGeier | decimad, no |
| 07:57.29 | GyrosGeier | decimad, use valgrind if you want to track memory errors |
| 07:57.34 | cyborg_ar | GyrosGeier, how big are the servos? |
| 07:57.36 | archivist | dc motor, h bridge driver, optical feedback to counters in the fpga |
| 07:57.52 | archivist | pwm driving the h bridge |
| 07:58.22 | GyrosGeier | cyborg_ar, http://www.hobbyking.com/hobbyking/store/__662__HXT900_Micro_Servo_1_6kg_0_12sec_9g.html is the bigger one |
| 07:58.44 | GyrosGeier | the FPGA is just for testing |
| 07:58.56 | GyrosGeier | ideally I'd connect it to a free PWM channel from the LEDs |
| 07:59.09 | cyborg_ar | hmm |
| 07:59.29 | cyborg_ar | i wonder if you could stick one of those tiny teeny magnetic encoders AMS makes |
| 07:59.40 | cyborg_ar | they're magic |
| 07:59.51 | GyrosGeier | the problem seems to be the deadband |
| 08:00.16 | cyborg_ar | GyrosGeier, the problem is the shitty servo loop, a quick and easy solution is to get a good digital servo |
| 08:00.59 | cyborg_ar | they put the deadband there because the pot has so much hysteresis that it would grind it's gears clean if left to oscillate |
| 08:01.15 | GyrosGeier | mmh |
| 08:02.06 | GyrosGeier | no idea how to find one that is better though |
| 08:02.18 | GyrosGeier | building my own is going to delay the project even further |
| 08:02.45 | *** join/#kicad firewalker (~firewalke@athedsl-05196.home.otenet.gr) |
| 08:02.50 | cyborg_ar | is looking on hobbyking |
| 08:04.15 | GyrosGeier | do they have a parametric search beyond "size"? |
| 08:04.25 | cyborg_ar | nope |
| 08:05.13 | archivist | there is so much backlash in the gear chain as well |
| 08:05.37 | GyrosGeier | that should be fixable with metal gears, probably |
| 08:05.45 | archivist | not really |
| 08:06.21 | cyborg_ar | metal gears will make it better |
| 08:06.32 | cyborg_ar | and with a known load it can probably be mostly tuned away |
| 08:06.46 | GyrosGeier | the load is minimal |
| 08:06.57 | archivist | clearance in gear design is pretty constant |
| 08:07.03 | archivist | http://www.collection.archivist.info/searchv13.php?searchstr=cnc+racing |
| 08:09.31 | cyborg_ar | hmmm |
| 08:11.11 | cyborg_ar | i'm watching a video on servos, analog vs digital, maybe just getting a digital will improve things a lot https://youtu.be/FeCmUwTdfH8?t=16m14s |
| 08:11.32 | cyborg_ar | didnt know the analog ones were so slow |
| 08:14.59 | archivist | methinks factual errors in that vid |
| 08:15.43 | archivist | pwm rate is separate form the feedback and internal response |
| 08:16.02 | cyborg_ar | yeah they may be conflating things |
| 08:16.17 | cyborg_ar | but it is real that the cheaper servo has a suprisingly slow loop |
| 08:16.36 | cyborg_ar | 50 Hz is slooooww |
| 08:17.16 | cyborg_ar | and i believe it is 50 because they made them hum and the slow one made a very familiar 50hz hum |
| 08:22.13 | cyborg_ar | http://www.hobbyking.com/hobbyking/store/__8303__Hextronik_MG_14_14g_2_6kg_0_11sec_Digital_Aircraft_Servo.html that could be a good step up |
| 08:23.27 | GyrosGeier | mmh |
| 08:23.30 | archivist | cheap for what is in there |
| 08:23.37 | GyrosGeier | adds some to the shopping list |
| 08:25.02 | *** join/#kicad alphaferret (~user@84.3.79.8) |
| 08:26.43 | alphaferret | hmm you got me paranoid now about not being able to fit the cards in easily, i might change the input jacks to a smaller one that sits flush or slightly recessed fromc cutout. the only problem is those don't make contact with the chassie so... |
| 08:26.52 | GyrosGeier | hm |
| 08:26.56 | alphaferret | maybe a pogo right there would help |
| 08:26.58 | GyrosGeier | that one is larger than the 900 |
| 08:27.30 | cyborg_ar | yeah it has more depth |
| 08:27.34 | cyborg_ar | and slightly more width |
| 08:27.49 | GyrosGeier | mmh |
| 08:27.55 | GyrosGeier | I think it should still fit |
| 08:30.04 | alphaferret | http://uploadpie.com/NDCjN |
| 08:30.40 | alphaferret | trashed the switchers over input jacks, right side is dc module, jack, ldo for rails, left side boost and ldo |
| 08:31.54 | alphaferret | phantom boost switcher on left sits next to either output jack or near to summing amp, i supposed that's better than charge pump right on top |
| 08:36.55 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 08:45.49 | *** join/#kicad Wetmelon (~wetmelon@host-73-169.ncasgro.asheville.nc.us.clients.pavlovmedia.com) |
| 08:54.44 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 09:06.42 | *** join/#kicad leila (~leila@a82-95-232-65.adsl.xs4all.nl) |
| 09:13.41 | invultri_ | do references start from 0 or 1 ? |
| 09:21.03 | trygvis | usually 1 |
| 09:21.04 | cyborg_ar | normally 1 |
| 09:21.26 | trygvis | I've nevery seen 0, but some start with 1000 and increment 1000 per sheet |
| 09:23.26 | invultri_ | then I did it correctly :) |
| 09:25.14 | invultri_ | and is it allowed to give them some extra identification other than a single number ? |
| 09:26.27 | alphaferret | i think it has to end in a number |
| 09:26.34 | alphaferret | anything can go before it though |
| 09:26.45 | mgottschlag | it is rather unusual though |
| 09:27.36 | invultri_ | its just for some specific things. I have X-PINK, X-RING, X-MDDL and X-INDX for instance |
| 09:34.37 | *** join/#kicad antipsychiatry (50f68aac@gateway/web/cgi-irc/kiwiirc.com/ip.80.246.138.172) |
| 09:35.47 | invultri_ | should be something like XFU<number><groupingletter> FU being finger unit, number to identify it and a grouping letter to indicate that things belong to eachother |
| 09:36.10 | invultri_ | or maybe A is better since that is an assembly |
| 09:37.45 | invultri_ | but the grouping letter seems to annoy the checker |
| 09:38.14 | *** join/#kicad eldarkg (~eldar@94.180.177.80) |
| 09:39.26 | *** join/#kicad oofus (~chris@oofus1-adsl.demon.co.uk) |
| 09:43.49 | cyborg_ar | what about swapping them? |
| 09:43.56 | cyborg_ar | put letter then number |
| 09:51.14 | invultri_ | that would have worked. I just made the unofficial agreement that odd number is top and odd number+1 is bottom |
| 09:51.36 | invultri_ | apparantly output connected to output is a bad thing |
| 09:52.51 | antipsychiatry | What u do for the secret service not using MIND READING against us ?????????????????????????????? How can we protect against their mind reading technology?????? Read: stopeg.com |
| 09:55.18 | invultri_ | wut? |
| 09:56.21 | mgottschlag | hm, related: Who in this channel is an op? :D |
| 09:57.29 | invultri_ | look for the @ signs |
| 09:57.43 | invultri_ | which is only chanserv at the moment :) |
| 09:58.04 | invultri_ | and ... good thing that error popped up. I was putting an output signal unbuffered on the bus |
| 09:58.07 | invultri_ | that is bad |
| 09:59.20 | mgottschlag | invultri_: yeah, but in many channels, there are people who have access to +o but don't use it all the time |
| 09:59.51 | archivist | freenode recommends not keeping ones hat on |
| 10:00.33 | *** join/#kicad KimK (~Kim__@2600:8803:7a85:6d00:adb5:f24a:73fe:51f) |
| 10:01.10 | *** mode/#kicad [+o archivist] by ChanServ |
| 10:01.24 | *** mode/#kicad [+b *!*50f68aac@*.com/ip.80.246.138.172] by archivist |
| 10:01.24 | *** kick/#kicad [antipsychiatry!~archivist@host81-149-189-98.in-addr.btopenworld.com] by archivist (antipsychiatry) |
| 10:01.46 | *** mode/#kicad [-o archivist] by ChanServ |
| 10:04.02 | *** join/#kicad phinxy (~tehhhd@unaffiliated/phinxy) |
| 10:07.21 | invultri_ | I can agree with that |
| 10:16.24 | *** join/#kicad hasuTMK (~hasu@58x13x224x139.ap58.ftth.ucom.ne.jp) |
| 10:20.55 | *** join/#kicad oninoshiko (~ahettinge@199.103.3.236) |
| 10:21.10 | invultri_ | so yesterday I was in a discussion that all cmos inputs should have a pullup or pulldown. |
| 10:22.51 | invultri_ | is that also true for input - output interchip lines ? |
| 10:23.48 | mgottschlag | I've only heard of that as a method to reduce the power consumption of floating wires and the following CMOS stages |
| 10:23.48 | *** join/#kicad stelarcf (~stelarcf@92.81.84.163) |
| 10:24.09 | mgottschlag | so I don't think it's necessary when the wire is always pulled to a defined level by the output of the other chip |
| 10:24.20 | mgottschlag | (but I missed that discussion) |
| 10:26.15 | invultri_ | but I suppose it would not hurt to have them either |
| 10:26.41 | invultri_ | just 1M or so to define the level |
| 10:29.58 | *** join/#kicad rubdos (~rubdos@host-85-27-50-55.dynamic.voo.be) |
| 10:41.00 | *** join/#kicad ricardocrudo (~ricardocr@2a02:8109:8c80:7bfc::2) |
| 10:50.50 | cyborg_ar | lol |
| 10:51.07 | cyborg_ar | there's been a few nutjos straying into here for some reason |
| 10:51.17 | cyborg_ar | and i'm not talking about xzcvczx |
| 10:51.44 | archivist | kicks cyborg_ar |
| 10:51.57 | cyborg_ar | D: |
| 10:52.29 | cyborg_ar | cmos inputs benefit from being pulled to a defined level |
| 10:52.56 | cyborg_ar | they have such high impedance that any stray fields can make them flap around and waste power or worse |
| 10:54.21 | cyborg_ar | for microcontrollers the best for open pins is to either make them analog imputs or digital outputs |
| 11:05.39 | alphaferret | can someone point me to a right angle low profile board to chassis spring contact? |
| 11:10.58 | *** join/#kicad erichvk_ (~esh@ppp118-210-141-198.lns20.adl6.internode.on.net) |
| 11:14.10 | invultri_ | cyborg_ar: I just made sure to add pull downs on all the lines. |
| 11:14.43 | cyborg_ar | alphaferret, for what, grounding? |
| 11:15.14 | cyborg_ar | the tradition is to use pull ups, but pull downs work too |
| 11:15.22 | alphaferret | yea i was thinking changing the 3.5mm mini connector to a bushingless one, but the problem then is no chassis contact |
| 11:15.56 | alphaferret | so looking for an elegant board to chassis contact for the balanced cable shield pin |
| 11:16.01 | invultri_ | I can always change the 3gnd to 3.3v |
| 11:16.17 | invultri_ | is there a screw terminal in the default libs |
| 11:16.18 | invultri_ | ? |
| 11:16.44 | cyborg_ar | alphaferret, http://www.digikey.com/product-search/en/connectors-interconnects/contacts-spring-loaded-and-pressure/1442508?FV=fff40016%2Cfff802cc&mnonly=0&newproducts=0&ColumnSort=1000011&page=1&stock=1&pbfree=0&rohs=0&quantity=10&ptm=0&fid=0&pageSize=25 |
| 11:16.49 | cyborg_ar | one of those? |
| 11:17.16 | alphaferret | how do those mount to pcb? |
| 11:17.31 | alphaferret | i need right angle i think |
| 11:17.35 | cyborg_ar | solder |
| 11:17.40 | cyborg_ar | read the datasheet |
| 11:17.45 | cyborg_ar | it's a drawing |
| 11:17.50 | *** join/#kicad jstein_ (~quassel@gentoo/developer/jstein) |
| 11:18.07 | alphaferret | no that won't work |
| 11:21.14 | cyborg_ar | http://www.digikey.com/product-detail/en/avx-corporation/009155002541006/478-5450-1-ND/2001011 |
| 11:21.19 | cyborg_ar | what about something like that |
| 11:21.55 | alphaferret | boom, a little big but sort of perfect idea |
| 11:22.35 | alphaferret | i'm looking at these pistons because they are cool as shit but right angled are pretty expensive |
| 11:24.14 | cyborg_ar | http://media.digikey.com/pdf/Data%20Sheets/ITT%20PDFs/Universal_Contact%202.5mm.pdf you could probably get away with soldering one of these on it's side |
| 11:26.02 | alphaferret | spring travel needs to be right angled to mounting plane |
| 11:26.27 | cyborg_ar | hence why i suggest mounting it on its side |
| 11:27.08 | alphaferret | ok i see, but it's sort of an important contact don't want to take chances |
| 11:27.47 | alphaferret | battery one will work, i'll find rom |
| 11:27.50 | alphaferret | room |
| 11:33.22 | alphaferret | cyborg_ar, i took your advice and reworked whole p/s layout |
| 11:36.42 | *** join/#kicad rohitksingh (~rohit@14.96.122.184) |
| 11:45.03 | *** join/#kicad fellbuendel (~furry@unaffiliated/fellbuendel) |
| 11:45.46 | alphaferret | ooooh clever solution: use the same bushing connector but let it sit flush against chassis hole, but instead of bushing on outside, bushing is on inside of chassis and unscrew it to make chassis contact |
| 11:55.51 | *** join/#kicad firewalker (~firewalke@athedsl-05196.home.otenet.gr) |
| 12:12.37 | *** join/#kicad jstein_ (~quassel@gentoo/developer/jstein) |
| 12:14.35 | *** join/#kicad rohitksingh (~rohit@223.227.167.134) |
| 12:40.24 | *** join/#kicad cyborg_ar (~cyborg_ar@c-98-215-226-11.hsd1.il.comcast.net) |
| 13:16.02 | *** join/#kicad rohitksingh (~rohit@223.227.248.6) |
| 13:37.29 | *** join/#kicad decimad (~deci@ip5f5be545.dynamic.kabel-deutschland.de) |
| 13:38.08 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 13:47.51 | GyrosGeier | hm |
| 13:48.04 | GyrosGeier | I have 96 connections on a small PCB |
| 13:48.48 | GyrosGeier | option 1) pin header forest |
| 13:48.54 | GyrosGeier | option 2: solder pads |
| 13:49.10 | GyrosGeier | option III, something inconsistent |
| 14:08.18 | *** join/#kicad mgottschlag (~quassel@reactos/tester/phoenix64) |
| 14:11.44 | *** join/#kicad cyborg_ar (~cyborg_ar@c-98-215-226-11.hsd1.il.comcast.net) |
| 14:13.43 | *** join/#kicad riq_ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 14:22.01 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 14:23.05 | *** join/#kicad blaze (~blaze@unaffiliated/blaze) |
| 14:36.15 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 14:53.40 | *** join/#kicad smoores (~smoores@c-67-188-210-139.hsd1.ca.comcast.net) |
| 15:03.20 | *** join/#kicad shaun413 (uid121475@gateway/web/irccloud.com/x-ocdairtojgrdyzqe) |
| 15:04.04 | *** join/#kicad superlou (~superlou@ool-457c146e.dyn.optonline.net) |
| 15:07.50 | *** join/#kicad Novus (~Novus@2605:e000:628b:1400:bd06:834a:37ed:fa6a) |
| 15:26.20 | *** join/#kicad alphaferret (~user@54034F08.catv.pool.telekom.hu) |
| 15:26.58 | alphaferret | how do you make a footprint for a part that has pads internally connected without routing wires on the board |
| 15:37.55 | *** join/#kicad ioch (~gyvunas@unaffiliated/ioch) |
| 15:38.04 | *** join/#kicad rohitksingh (~rohit@180.151.119.206) |
| 15:50.15 | GyrosGeier | alphaferret, usually you are supposed to connect all the pads in parallel |
| 15:56.21 | *** join/#kicad decimad (~deci@ip5f5be545.dynamic.kabel-deutschland.de) |
| 15:56.36 | c4757p | ^ |
| 15:56.54 | c4757p | I don't know why people insist on using internal connections as...connections |
| 15:56.59 | c4757p | the part has multiple pins on the same net for a /reason/ |
| 15:57.40 | archivist | sometimes to use as a jumper! |
| 15:57.49 | c4757p | thanks |
| 15:58.01 | GyrosGeier | the only place where it makes sense is using the shield of an RJ45 connector as a jumper on a single-sided PCB |
| 15:58.52 | c4757p | o_O |
| 15:59.00 | c4757p | an RJ45 on a single-sided PCB? |
| 15:59.17 | archivist | 0ohms on single sided boards, lots of heatsink/ high current pins etc |
| 15:59.37 | *** join/#kicad Laksen (~Laksen@3e6b798f.rev.stofanet.dk) |
| 15:59.37 | *** join/#kicad Laksen (~Laksen@unaffiliated/laksen) |
| 16:04.55 | alphaferret | sorry i'm being dyslexic on this one, so do you just give the pins or pads the same net name? I shouldn't have to route extra traces |
| 16:13.05 | *** part/#kicad talsit (foobar@gromit.mixdown.ca) |
| 16:14.46 | *** join/#kicad scubasonar (~smuxi@c-24-5-230-176.hsd1.ca.comcast.net) |
| 16:16.22 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 16:20.17 | *** join/#kicad Thaolia (~thaolia@80.90.49.230) |
| 16:29.31 | *** join/#kicad blaze (~blaze@unaffiliated/blaze) |
| 16:35.33 | *** join/#kicad stefan_slight (~Thunderbi@89.15.237.44) |
| 16:56.17 | *** join/#kicad NoGodDamnIdea (~King@unaffiliated/nogoddamnidea) |
| 16:59.07 | Laksen | What version of wxWidgets does the current Kicad master require to build? |
| 17:01.18 | DocScrutinizer05 | invultri_: pulldown R only on (possibly) open pins |
| 17:04.22 | c4757p | Laksen: 3.0.2, allegedly |
| 17:05.01 | DocScrutinizer05 | ((<alphaferret> ooooh clever solution: use the same bushing connector but let it sit flush against chassis hole, but instead of bushing on outside, bushing is on inside of chassis and unscrew it to make chassis contact)) will you use special custom made plugs to fir in there? |
| 17:05.42 | alphaferret | uhh, not settled on that idea, but no idea for custom plugs |
| 17:05.54 | alphaferret | no need* |
| 17:06.19 | alphaferret | you mean socket or plug? |
| 17:06.44 | alphaferret | it's just a standard 3.5mm stereo jack connecter |
| 17:06.46 | DocScrutinizer05 | err, sure. Usually the plug needs to go in all the way until it sits flush on what you call bushing |
| 17:07.51 | alphaferret | the bushing in this case is the threaded part of the jack that would have protruded from the case |
| 17:08.04 | DocScrutinizer05 | you can't add 'distance' there by case wall thickness and/or screws you turn outward to make some contact |
| 17:08.26 | Laksen | Ah okay. Latest release. Sounds reasonable.. :) |
| 17:08.26 | alphaferret | screws would be a good idea but this is a flimsy mint tin |
| 17:09.14 | cyborg_ar | lol using internally connected pins as jumpers |
| 17:09.23 | DocScrutinizer05 | OUCH! |
| 17:09.40 | DocScrutinizer05 | absolute NOGO according to all datasheets |
| 17:10.00 | cyborg_ar | yeah they are rarely rated for such use |
| 17:10.12 | DocScrutinizer05 | except for clearly defined feedthrough like e.g. in the ESD prot part invultri_ found |
| 17:10.32 | cyborg_ar | and you dont want to be the guy debugging bizzarre board behavior because of a cold joint in a shield of an unrelated connector |
| 17:11.09 | cyborg_ar | or pushbuttons |
| 17:11.29 | cyborg_ar | i've been tempted to use pushbuttons as jumpers before but always managed to come to my senses |
| 17:11.43 | DocScrutinizer05 | connecting GND zones via chip GND pins is a rather classical notorious error that causes lots of bizarre problems |
| 17:12.44 | DocScrutinizer05 | at least TI (iirc) is very explicit on that |
| 17:15.00 | DocScrutinizer05 | alphaferret: for audio you probably want jacks with plastic bushing and internal GND contact for the plug, to avoid poor GND connections to case, or - worse - GND loops |
| 17:15.08 | cyborg_ar | i would like a thing for SMD connectors that have mechanical pads so i can have them connected to GND (or the dominant pour) without having to put an extra pin in the symbol, or mandate that i absolutely must connect it |
| 17:15.30 | *** join/#kicad ayjay (~ajp_@24-151-28-56.dhcp.nwtn.ct.charter.com) |
| 17:16.10 | cyborg_ar | having them connected to the pour would improve adhesion and avoid breaking it unnecessarily |
| 17:16.22 | DocScrutinizer05 | :nod: |
| 17:16.52 | alphaferret | i think mill-max makes things like that |
| 17:17.44 | DocScrutinizer05 | alphaferret: generally try to strictly stick to a star topology for your GND, with star "hub" next to most sensitive input of your device |
| 17:18.16 | alphaferret | i'm using a plane |
| 17:18.34 | DocScrutinizer05 | well that's a pretty fat star but still... :-) |
| 17:18.51 | DocScrutinizer05 | avoid GND loops by all means |
| 17:19.20 | DocScrutinizer05 | e.g. via multiple jacks mounted to case steel with their bushing |
| 17:19.52 | alphaferret | the input jack is signal shield, not 0v |
| 17:20.09 | alphaferret | phantom reference goes to chassis |
| 17:20.24 | DocScrutinizer05 | :nod: so even more important not to make bushing contact the case |
| 17:20.26 | alphaferret | 0v plane and they will connect in one spot |
| 17:20.42 | alphaferret | no, i wanted it to contact |
| 17:21.29 | DocScrutinizer05 | just telling you age old audio design gossip |
| 17:21.48 | alphaferret | balanced signal shield to chassis |
| 17:22.00 | alphaferret | 0v ground plane |
| 17:22.11 | DocScrutinizer05 | might work |
| 17:22.38 | DocScrutinizer05 | but then you need to make damn sure there's no other possibly obscure contact of chassis to 0V |
| 17:22.50 | alphaferret | yes |
| 17:23.28 | alphaferret | except for the single junction |
| 17:23.35 | DocScrutinizer05 | I hope you have no switches, potentiometers etc that need fixing to case |
| 17:23.49 | alphaferret | pots are plastic |
| 17:23.53 | DocScrutinizer05 | good |
| 17:24.12 | alphaferret | dc power jack i need to be careful |
| 17:24.19 | DocScrutinizer05 | what about output? |
| 17:24.25 | alphaferret | it's plastic |
| 17:24.31 | DocScrutinizer05 | good |
| 17:24.36 | alphaferret | output is unbalanced stereo mini |
| 17:24.41 | alphaferret | i can use the same plastic bushing |
| 17:25.05 | alphaferret | hmm actually not, good catch |
| 17:25.09 | DocScrutinizer05 | make sure the plug can't make contact to case |
| 17:25.13 | alphaferret | i need secure bushing on the out |
| 17:25.19 | *** join/#kicad berndj (~berndj@197.242.93.84) |
| 17:25.54 | alphaferret | ok got to rethink that one |
| 17:26.02 | DocScrutinizer05 | use plastic screws or isolation washers |
| 17:26.17 | alphaferret | no i need a better solution |
| 17:26.29 | DocScrutinizer05 | make sure the plug bottom plane has a distance of at least 1mm to case |
| 17:26.38 | alphaferret | i haven't really gotten to the final ground scheme thought process yet |
| 17:26.56 | alphaferret | yes bottom plane will be taped or coated |
| 17:27.35 | *** join/#kicad berndj (~berndj@197.242.93.84) |
| 17:29.11 | DocScrutinizer05 | there are those https://ixquick-proxy.com/do/show_picture.pl?l=english&rais=1&oiu=http%3A%2F%2Fimages.monoprice.com%2Fproductlargeimages%2F6672.jpg&sp=72aaa02b99885ca1bcf86b28ea4bab26 but also those https://ixquick-proxy.com/do/show_picture.pl?l=english&rais=1&oiu=http%3A%2F%2Fwww.parts-express.com%2FData%2FDefault%2FImages%2FCatalog%2FOriginal%2F090-292_HR_0.jpg&sp=0f5cd650ba7021ba05baa61345a5c356 plugs |
| 17:29.14 | alphaferret | last thing i read was that chassis acts as extension of balanced cable shield |
| 17:30.09 | alphaferret | right i need to be careful of metal plugs |
| 17:30.11 | DocScrutinizer05 | even those: https://ixquick-proxy.com/do/show_picture.pl?l=english&rais=1&oiu=http%3A%2F%2Fwww.parts-express.com%2FData%2FDefault%2FImages%2FCatalog%2FOriginal%2F092-0150_HR_0.jpg&sp=fd3d009f81a9d1abf1d027b4fef34d09 |
| 17:30.16 | alphaferret | will cogitate on that |
| 17:31.16 | DocScrutinizer05 | GND design is usually surprisingly non-trivial |
| 17:31.32 | alphaferret | i know |
| 17:31.39 | alphaferret | that's why i'm putting it off |
| 17:31.51 | alphaferret | have other fish to fry first |
| 17:31.59 | alphaferret | power and ground have taken the most time so far |
| 17:34.47 | DocScrutinizer05 | don't know, from my guts feeling I'd say it's a good thing to have power and GND sorted before you even start frying other fish :-) YMMV |
| 17:36.03 | DocScrutinizer05 | changing either power or GND concept later on usually means to re-evaluate a lot of things |
| 17:36.37 | DocScrutinizer05 | particularly when either one is non-trivial |
| 17:37.03 | DocScrutinizer05 | as frequently with audio |
| 17:37.32 | *** join/#kicad berndj (~berndj@197.242.93.84) |
| 17:38.15 | DocScrutinizer05 | the standard approach is: star topology for GND, hub next to input, connect case to hub there via a dedicated contact. Use isolated jacks everywhere |
| 17:38.18 | alphaferret | i'm just working on the phantom supply layout, rail power is dc module (isolated) |
| 17:38.46 | alphaferret | and what is standard for phantom reference vs 0v |
| 17:39.14 | DocScrutinizer05 | keep them strctly separate |
| 17:39.36 | alphaferret | so i could just make another bus trace instead of using chassis |
| 17:39.40 | alphaferret | no big deal |
| 17:39.59 | *** join/#kicad antipsychiatry (50f68b3e@gateway/web/cgi-irc/kiwiirc.com/ip.80.246.139.62) |
| 17:41.35 | DocScrutinizer05 | you often see PE or cable shields etc connected to GND via 1M R || small capacitor for RF |
| 18:11.06 | cyborg_ar | lol frying |
| 18:11.12 | cyborg_ar | i see what you did there |
| 18:13.46 | reportingsjr | mmm, rust is C++ the way it should have been done |
| 18:14.23 | cyborg_ar | is it even done yet though? |
| 18:15.56 | DocScrutinizer05 | my old server is named rust :-D |
| 18:16.23 | cyborg_ar | i had an old server made out of an old desktop computer |
| 18:16.29 | cyborg_ar | i called it jalopy |
| 18:18.01 | *** join/#kicad tannewt (~textual@c-71-231-55-126.hsd1.wa.comcast.net) |
| 18:39.27 | *** join/#kicad sandeepkr (~sandeepkr@103.49.155.114) |
| 18:40.58 | *** join/#kicad bugzc (~1@unaffiliated/bugzc) |
| 18:43.26 | *** join/#kicad Wetmelon (~wetmelon@host-73-169.ncasgro.asheville.nc.us.clients.pavlovmedia.com) |
| 18:55.27 | *** join/#kicad maicod (~mailadd@d55059.upc-d.chello.nl) |
| 18:55.37 | *** part/#kicad maicod (~mailadd@d55059.upc-d.chello.nl) |
| 19:00.33 | *** join/#kicad eeriegeek (~eeriegeek@73.31.123.0) |
| 19:00.55 | reportingsjr | cyborg_ar: define done |
| 19:01.08 | reportingsjr | The stable version is stable |
| 19:01.22 | reportingsjr | but like basically every language currently used it is still being updated |
| 19:05.58 | DocScrutinizer05 | idly wonders when all possibly imaginable programming languages finally have been invented and developers agree on which half dozen of them are best and get used |
| 19:07.04 | DocScrutinizer05 | https://xkcd.com/927/ |
| 19:09.27 | trygvis | are there any tools to mass edit component attributes? like setting a custom field (e.g. tollerance) on a bunch of resistors |
| 19:11.36 | decimad | text editor |
| 19:12.40 | reportingsjr | ^^ |
| 19:12.50 | *** join/#kicad pointhi_ (~pointhi@88.116.72.14) |
| 19:13.39 | trygvis | yeah.. from the .sch file, what are the different fields here: F 5 "123" H 1600 3250 60 0001 C CNN "wat" |
| 19:13.50 | trygvis | it is a custom field I just added from the gui |
| 19:16.33 | DocScrutinizer05 | text editor since the generic attribute inspector requester isn't implemented yet :-) |
| 19:22.41 | *** join/#kicad KarlZeilhofer (~Thunderbi@188-22-39-246.adsl.highway.telekom.at) |
| 19:32.00 | c4757p | https://i.imgur.com/dP4dcLn.jpg |
| 19:33.30 | trygvis | what is that? |
| 19:33.36 | c4757p | /art/ |
| 19:33.39 | trygvis | hehe, yeah |
| 19:34.04 | c4757p | there are /so/ many things wrong with that. like dozens. but I don't /care/, I just think it's /adorable/ |
| 19:34.22 | c4757p | that is the best bad board I've seen in a while |
| 19:35.07 | *** join/#kicad IanTLopp (~IanTLopp@mc24636d0.tmodns.net) |
| 19:37.51 | trygvis | what does it do? |
| 19:38.12 | c4757p | apparently it's a datalogger of some sort |
| 19:38.19 | c4757p | dunno, not my board, it was shared in #otherchannel |
| 19:40.32 | *** join/#kicad AndyBotwin (~Gustavo@unaffiliated/andybotwin) |
| 19:41.07 | reportingsjr | that looks cool as hell |
| 19:41.27 | reportingsjr | ever heard of.. topor? |
| 19:41.32 | reportingsjr | I think that is what it is called |
| 19:41.47 | reportingsjr | yeah, that's it: http://eda.eremex.com/products/topor/ |
| 19:41.59 | c4757p | yeah |
| 19:42.23 | c4757p | I like topoR, though most people seem to think I'm a horrible person for it :D |
| 19:43.44 | reportingsjr | I kinda like it too |
| 19:44.51 | reportingsjr | wonders when EDA will start having net parameters as part of footprints |
| 19:44.57 | reportingsjr | (commonly) |
| 19:45.26 | reportingsjr | stuff like "the output pin of this device switches high currents and will cause traces near it to be noisy" |
| 19:45.36 | reportingsjr | and "this pin needs a low noise path" |
| 19:45.50 | Sync | topor is pretty cool |
| 19:51.37 | GyrosGeier | c4757p, oh noes, 82ed0fde32d5e85b238e821af1e542574d8e9fe7 breaks MSVC builds |
| 19:52.22 | GyrosGeier | (OpenMP loop variables need to be signed according to the OpenMP spec, but gcc has a nonstandard extension there) |
| 19:52.37 | c4757p | I tested that in every compiler we support |
| 19:52.58 | GyrosGeier | which are gcc and gcc |
| 19:53.17 | GyrosGeier | adds it to the MSVC patch stack |
| 19:53.36 | c4757p | no, clang too |
| 19:53.40 | c4757p | as that's used to build on osx |
| 19:53.44 | GyrosGeier | ah |
| 19:53.47 | GyrosGeier | they follow gcc |
| 19:54.21 | c4757p | * GyrosGeier adds it to the MSVC patch stack --- if it broke due to incompatibility with an actual standard, feel free to give me a patch and I'll merge it, not going to pull "we don't support msvc" there |
| 19:54.42 | c4757p | gcc and clang should have emitted a warning for that if it's something they want to allow that's against the standard. shame on them. |
| 19:55.20 | GyrosGeier | we already have one of these signed/unsigned in ratsnest_data.cpp |
| 19:55.48 | GyrosGeier | I was just hoping to eventually get rid of differences |
| 19:56.08 | GyrosGeier | problem is if you make these variables signed, you get additional warnings from other code |
| 19:57.43 | *** join/#kicad cyborg_ar (~cyborg_ar@c-98-215-226-11.hsd1.il.comcast.net) |
| 19:58.12 | *** join/#kicad nitrousnrg (~quassel@190-103-222-12.cepanet.com.ar) |
| 19:58.17 | *** join/#kicad nitrousnrg_ (~quassel@190-103-222-12.cepanet.com.ar) |
| 20:01.33 | DocScrutinizer05 | KOTKEYS_H hotkey_id_commnand ID_COMPONENT_BUTT m_Curseur honestly... :-S |
| 20:04.11 | DocScrutinizer05 | the first one maybe only looks strange for Germans |
| 20:04.31 | DocScrutinizer05 | last one maybe doesn't look strange to French |
| 20:06.41 | DocScrutinizer05 | hotkeys.* |
| 20:10.26 | DocScrutinizer05 | too much french in sourcecode for me |
| 20:10.50 | c4757p | that's how you identify the age of the code |
| 20:10.53 | c4757p | percent of French |
| 20:13.57 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 20:15.30 | DocScrutinizer05 | onleftclick must be ancient |
| 20:16.29 | DocScrutinizer05 | /* Traite les commandes declench�e par le bouton gauche de la souris, |
| 20:16.30 | DocScrutinizer05 | <PROTECTED> |
| 20:16.31 | DocScrutinizer05 | <PROTECTED> |
| 20:16.45 | *** join/#kicad riq_ (~riq_@c-24-6-115-254.hsd1.ca.comcast.net) |
| 20:18.40 | mrnuke | anyone aware of a library for VQFN chips? |
| 20:21.15 | mrnuke | example: http://www.nxp.com/documents/outline_drawing/SOT762-1.pdf |
| 20:25.59 | DocScrutinizer05 | wow, NXP datasheet with proper footprint info |
| 20:38.17 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 20:45.52 | *** join/#kicad ricardocrudo (~ricardocr@2a02:8109:8c80:7bfc::2) |
| 21:03.54 | mrnuke | hmm, anyone knows what scripts are used to generate Housings_DFN_QFN.pretty? |
| 21:14.34 | GyrosGeier | hm |
| 21:15.00 | GyrosGeier | all these driver ICs only go to 3.8V for '1' if Vcc is +5V |
| 21:25.27 | *** join/#kicad saxicola_afk (~saxicola@192.150.115.87.dyn.plus.net) |
| 21:27.08 | *** join/#kicad mozzarella (~sam@unaffiliated/sam113101) |
| 21:49.02 | *** join/#kicad cirilo (~cirilo@CPE-120-144-53-157.lnse5.win.bigpond.net.au) |
| 21:50.07 | cirilo | c4757p, you're absolutely right from a technical viewpoint - however, beware the long arm of the Dick. People have had their commit access revoked for asking milder questions. :O |
| 21:50.19 | c4757p | that's okay |
| 21:50.35 | c4757p | I have a very simple rule for contributing to community projects |
| 21:50.58 | c4757p | you either want me to contribute, and accept that my opinions are valid and worth hearing, or you can feel free to show me the door and we'll both have a better time apart |
| 21:51.19 | c4757p | Dick reaching his long arm in and revoking commit access would simply make me not want it anyway |
| 21:51.21 | c4757p | ^-^ |
| 21:51.40 | cirilo | Lorenzo did just that .. but then he already had a few years' experience maintaining his own fork of KiCad with features. :) |
| 21:52.24 | cirilo | but Wayne's excuse for generated files in the source tree is just lame dumbfuckery |
| 21:52.35 | c4757p | I'm unusually grumpy this weekend for some reason, but I'm starting to get pissed at this project. |
| 21:53.00 | cirilo | oh, kicad is always frustrating. :) |
| 21:53.05 | c4757p | I'm fond at the description "bikeshedding by people who want to paint the shed orange with green polka dots to make it easier to find" |
| 21:53.34 | c4757p | yes, but I don't fucking have /time/ to wrangle a frustrating project anymore |
| 21:53.52 | c4757p | so I'm at the "deal with it /my/ way and they can either fix their shit or send me packing" stage. |
| 21:54.15 | c4757p | I'm going to continue voicing my frustration with all the moronic shit they do until they either stop doing it or demand /I/ stop, at which point I'll be done, and happier. |
| 21:55.04 | cirilo | yeah; all the nit picking on introducing the 3D plugin system has put me off investing in the PCB API |
| 21:55.18 | c4757p | it really is bikeshedding, though |
| 21:55.23 | c4757p | I don't have much trouble introducing largish changes |
| 21:55.34 | c4757p | it's "hey, let's fix the version strings" that gets picked to death by morons |
| 21:55.37 | cirilo | all that thought goes into proper design and you get told "we want to do things in a way even a retard wouldn't do" |
| 21:56.38 | cirilo | yeah - reading through those I thought "wtf - can't they see it's easy enough to find the commit in git and that the bzr revno is worse than useless?" |
| 21:57.24 | c4757p | (anyway, re: long arm of Dick - I've been /way/ nastier about Dick before and he didn't do anything.) |
| 21:57.29 | c4757p | /directly/ about Dick. |
| 21:57.32 | cirilo | in a way it was reminiscent of Linus' early git emails trying to get people to understand that the cvs/svn/bzr revisions really weren't a Good Idea |
| 21:57.56 | c4757p | Linus was in the enviable position of being able to say "fuck you, we're doing it this way" ;) |
| 21:58.30 | cirilo | I think he put an email out to that effect eventually |
| 21:58.46 | c4757p | it's Linus, he probably said that verbatim :D |
| 21:58.58 | cirilo | along with the "if you really want to continue using bitlocker that's *your* problem" |
| 21:59.50 | cirilo | but man, was he sore with Tridge for years after that |
| 21:59.55 | c4757p | https://misc.c4757p.com/fucking-launchpad.png |
| 21:59.59 | c4757p | roduct-git indeed |
| 22:00.30 | cirilo | hahaha; well, Launchpad is "special" |
| 22:08.00 | decimad | yay, the single trace router kind of works again! |
| 22:08.20 | cirilo | I wonder what it would take to make kicad stop printing inner layer pads when those pads have no connection anyway |
| 22:08.43 | cirilo | with proper DRC support so DRC doesn't prevent people from routing tracks where those pads would be |
| 22:09.37 | decimad | inner layer pads? |
| 22:10.09 | *** join/#kicad CarlPoirier (~CarlPoiri@modemcable019.249-59-74.mc.videotron.ca) |
| 22:10.13 | c4757p | the DRC would be a bit awful |
| 22:10.21 | cirilo | yeah, kicad's idea of a thru-hole pad or via is to put large annular rings on all inner layers |
| 22:11.35 | decimad | and kicad should not prevent routing inner tracks through the holes? |
| 22:11.57 | cirilo | copper pours make the rules a bit trickier as well |
| 22:12.49 | cirilo | decimad, no - when you put vias or thru-pads under a component to break out leads, you don't want annular rings with no electrical connection on that layer to interfere with the routing |
| 22:13.37 | cirilo | so effective size of an annular ring with no connection should be drill size + 2*plated thickness |
| 22:19.22 | decimad | yeah, and then the DRC would only use the keepaway rules and everything is good, I guess |
| 22:19.29 | *** join/#kicad ferdna (~ferdna@cpe-69-23-129-73.elp.res.rr.com) |
| 22:19.50 | decimad | But drills may easily be off by a few mu |
| 22:22.45 | cirilo | that's why design rules take fabrication reproduction errors into account |
| 22:23.40 | c4757p | gah |
| 22:23.45 | c4757p | tries /not/ to reply to this one |
| 22:24.30 | cirilo | is there another one? :) |
| 22:24.33 | *** join/#kicad talsit (foobar@gromit.mixdown.ca) |
| 22:24.36 | c4757p | he replied to me |
| 22:24.55 | cirilo | heh .. "I am too right! |
| 22:25.01 | cirilo | yeah - better ignore it |
| 22:25.04 | c4757p | I can't think of anything to say back that isn't snippy and biting, so I think I'll hold my tongue. |
| 22:25.12 | *** join/#kicad BillSussman (~Gustavo@177.204.22.106.dynamic.adsl.gvt.net.br) |
| 22:25.13 | c4757p | until the grumpiness eats at me to the point where I can't anymore, that is. |
| 22:25.24 | cirilo | let him keep that bit of dumbfuckery |
| 22:25.37 | c4757p | I mean, I suppose I may have won this one |
| 22:25.41 | decimad | c4757p: Hrmmm, so I reworked the revision handling of the router. But I'm not really happy with the naming. The change set of one revision to another I called "DiffState", and the spatial revision (ie. where your're colliding stuff in etc. during pns) is called RevisionNode, it always has the state up to and including the diffstate of its revision. |
| 22:25.42 | c4757p | it doesn't look like he's reverting it. |
| 22:25.44 | cirilo | there are more intersting thing to do/discuss |
| 22:25.47 | *** join/#kicad mozzarel1 (~sam@unaffiliated/sam113101) |
| 22:26.19 | cirilo | be quiet and he might forget to revert it :) |
| 22:26.22 | c4757p | the stupid decisions just wear me down |
| 22:26.26 | c4757p | this project has so many |
| 22:26.38 | c4757p | and honestly, it really turns people off. the public perception of kicad is that it is /very/ unpolished |
| 22:26.51 | cirilo | yeah .. but still - what are its free competitors like? |
| 22:26.53 | c4757p | there is a /reason/ for that. |
| 22:26.57 | decimad | So now the spatial state walks the diff tree, no more multiple and interlinked spatial states (ie. the one you have is always consistent) |
| 22:27.03 | c4757p | <cirilo> yeah .. but still - what are its free competitors like? --- so? |
| 22:27.12 | c4757p | we might be less bad than the alternative |
| 22:27.15 | cirilo | best of a bad bunch |
| 22:27.17 | c4757p | but I want to be /good/, not less bad |
| 22:27.49 | cirilo | I know .. and it's a long road full of idiotic decisions made by people holding the reins |
| 22:28.30 | *** join/#kicad antonioclj (~Thunderbi@189.41.66.166) |
| 22:29.04 | *** join/#kicad kakimir (kakimir@sienimetsa.wtf) |
| 22:29.11 | cirilo | current python scripting needs to die and be replaced by one based on a sane C API |
| 22:29.59 | *** join/#kicad dingbat (uid70835@gateway/web/irccloud.com/x-eukrryennkkchbbq) |
| 22:30.13 | cirilo | need to get MCAD export in there - will be shoehorned as an "intermediate" solution but ideally should be a proper plugin (which may never happen) |
| 22:30.43 | cirilo | the lack of true arcs representing track segments drives me batty ... |
| 22:30.52 | *** join/#kicad Jacob______ (sid27928@gateway/web/irccloud.com/x-ugavjdazbetiunfk) |
| 22:32.07 | *** join/#kicad dsal (sid13060@gateway/web/irccloud.com/x-ilyhjknydecqqujo) |
| 22:32.17 | cirilo | but I remember discussions years ago when I'd say "we need to do X to get kicad in shape for future stuff which everyone, including hobbyists. will need" |
| 22:32.37 | cirilo | and 90% of the responses are excuses for doing things in a thoroughly idiotic way |
| 22:33.32 | *** join/#kicad antonioclj (~Thunderbi@189.41.66.166) |
| 22:33.37 | cirilo | just look at the mess we have of outlines. Rather than cutouts + outlines, which would make it trivial to support multiple bodies and flex-rigid - we have this 1970s era single outline + cutouts on a single layer |
| 22:34.12 | cirilo | well, admittedly I can't imagine hobbyists using flex-rigid at this point in time :) |
| 22:37.14 | *** join/#kicad DemoFreak (~hannes@idefix.v6.chdstudt.de) |
| 22:37.14 | *** join/#kicad DemoFreak (~hannes@pdpc/supporter/active/demofreak) |
| 22:38.51 | *** join/#kicad x29a (~x29a@unaffiliated/x29a) |
| 22:39.24 | Sync | cirilo: well, flex rigid has got suprisingly cheap |
| 22:39.25 | *** join/#kicad wpwrak (~werner@181.171.122.65) |
| 22:51.13 | cirilo | Sync, yeah, but unless the hobbyist happens to be an EE, what are the odds they'd use flex-rigid? |
| 22:53.05 | c4757p | I've seen non-EE but quite serious hobbyists do some heavy stuff :) |
| 22:56.46 | cirilo | yeah, but they're pretty rare :) |
| 22:57.22 | cirilo | anyway .. wonder if I should kick up a stink on the dev list about steps towards flex-rigid support .. *again* |
| 22:58.42 | c4757p | I'm all for kicking up stinks on the list |
| 23:00.13 | cirilo | I can already imagine the responses to "let's ditch the current idiotic edge definitions" |
| 23:01.06 | Bird|otherbox | cirilo: sounds like something the zoidberg would do :p |
| 23:01.19 | cirilo | heh |
| 23:01.54 | cirilo | 'k - gotta run |
| 23:06.10 | *** join/#kicad DemoFreak (~hannes@idefix.chdstudt.de) |
| 23:06.10 | *** join/#kicad DemoFreak (~hannes@pdpc/supporter/active/demofreak) |
| 23:06.44 | decimad | c4757p, was there something I didn't catch? my brain was somewhere else |
| 23:07.02 | c4757p | so was mine |
| 23:07.03 | c4757p | ¯\_(ã)_/¯ |
| 23:07.38 | decimad | I reread and there were a few "he"s I could potentially attribute to me |
| 23:08.30 | decimad | oh, that was regarding wayne's complaint, right? |
| 23:13.04 | *** join/#kicad dingbat (uid70835@gateway/web/irccloud.com/x-ycovieoiirmwuqkz) |
| 23:15.41 | c4757p | yes, I was just bitching and moaning. |
| 23:21.00 | *** join/#kicad ne0phyte (sid18336@gateway/web/irccloud.com/x-cqbemxpkuavdjqfg) |
| 23:26.03 | decimad | https://en.wikipedia.org/wiki/File:Olympus_Stylus.jpg hahaha |
| 23:32.27 | *** join/#kicad Jacob______ (sid27928@gateway/web/irccloud.com/x-awgtilytantzkmfr) |
| 23:40.17 | *** join/#kicad Plugh (~kcozens@CPE68b6fc4e50b3-CM68b6fc4e50b0.cpe.net.cable.rogers.com) |
| 23:54.09 | *** join/#kicad smoores (~smoores@c-67-188-210-139.hsd1.ca.comcast.net) |
