**** BEGIN LOGGING AT Wed Dec 03 02:59:59 2014 Dec 03 07:24:19 haloooo Dec 03 07:26:52 av500: your fan is back! Dec 03 07:27:09 /o\ Dec 03 07:30:55 yep Dec 03 07:31:18 im here ready for new day with trying to kill bbb Dec 03 07:31:26 not to kill* Dec 03 07:34:02 try to kill it! Dec 03 07:34:17 the chances are higher that you will not succeede Dec 03 07:34:22 :P Dec 03 07:34:43 so em, some warmup question Dec 03 07:35:04 no Dec 03 07:35:11 come on :( Dec 03 07:35:12 the answer is just plain no Dec 03 07:35:20 its gonna be an easy one probably Dec 03 07:35:22 you havent paid you r debt Dec 03 07:35:32 i am trying to pay back the community Dec 03 07:35:40 neither did we receive chocolate, nor your imortal soul or your first born Dec 03 07:36:21 I told you, If i send yoo chocolate to zurich I would spend my entire month paycheck Dec 03 07:36:36 that's ok Dec 03 07:36:47 not for me i need to eat :O Dec 03 07:37:34 eating is overrated! Dec 03 07:37:43 +11 Dec 03 07:37:46 just look at av500. huge as a mountain, but doesnt eat anything! Dec 03 07:38:02 lol Dec 03 07:39:58 I am using debian on the BBB and I have a script that reads the time and date from my connected i2c RTC clock. For some reason the script doesnt work when set to be executed at startup (its soft link is placed in init.d directory, I then did rc update). Any ideas what could be the reason? If I execute the script by hand it works Dec 03 07:42:19 how did you "set it to be executed"? Dec 03 07:42:43 also, where is it located? Dec 03 07:42:45 path issue or sp Dec 03 07:42:46 so Dec 03 07:42:54 pastebin the script Dec 03 07:43:14 or maybe not in rc2.d Dec 03 07:43:21 or not correctly named or so Dec 03 07:43:23 http://pastie.org/9757628 Dec 03 07:43:35 no it is in my user folder and a soft link is placed in init.d Dec 03 07:43:47 well, init.d does not get executed Dec 03 07:43:59 init.d is the place where you should put this script Dec 03 07:44:00 dont you run systemd? Dec 03 07:44:08 oh.. right Dec 03 07:44:10 that thing Dec 03 07:44:19 I gues I do Dec 03 07:44:24 lol Dec 03 07:44:26 find out Dec 03 07:44:31 is that partialy why you hate it? Dec 03 07:44:36 No, no I do Dec 03 07:44:51 Is that a problem? Dec 03 07:45:05 as a certain plague put it recently: arm and x86 have one common enemy: systemd Dec 03 07:45:21 but then, he also like resisors Dec 03 07:45:35 get warm? Dec 03 07:45:47 bremenpl: so write a systemd unit Dec 03 07:45:52 1) it will work Dec 03 07:45:58 2) you learn how to do thatz Dec 03 07:46:02 3) .. Dec 03 07:46:05 4) profit Dec 03 07:46:20 you guys finish eachothers sentences, thats kinda sweet Dec 03 07:46:39 love is what keeps this channel alive Dec 03 07:46:47 :-* Dec 03 07:46:53 * KotH hugs a mountain Dec 03 07:46:55 ok, thanks Dec 03 07:46:56 :P Dec 03 07:46:57 and sarcasm Dec 03 07:47:06 I think the second thing mostly Dec 03 07:47:18 :P Dec 03 07:50:34 what do you think? http://pastie.org/9757637 Dec 03 07:51:01 * KotH has never learned how to systemd Dec 03 07:51:13 but isnt there a way to put everthing in there? Dec 03 07:51:21 ie. the echoe + hwclock calls? Dec 03 07:51:44 i dont understand Dec 03 07:53:33 put everything into that thingy, instead of calling a shell script Dec 03 07:53:48 systemd has ample documentation Dec 03 07:53:49 ohh, well I already had the scrript so didnt want to do more work Dec 03 07:53:58 I had to add Install label, now it works Dec 03 07:54:00 thanks Dec 03 07:54:03 even the man pages have man pages Dec 03 07:54:05 bad bremenpl! bad bremenpl! Dec 03 07:54:09 lol Dec 03 07:54:35 i still dont get the hatered thats put on systemd Dec 03 07:54:46 maybe beacause i am not aware of things, like kids Dec 03 07:54:53 it's mostly a philosophical issue Dec 03 07:54:57 and a personal issue Dec 03 07:55:10 there so many memes all over the internet about it Dec 03 07:55:11 bremenpl: its grumpy old men Dec 03 07:55:11 systemd is a big block of code at a very critical place Dec 03 07:55:22 KotH: like the kernel? Dec 03 07:55:27 I always hated that Dec 03 07:55:28 it's works differently then people are used to Dec 03 07:55:32 ah Dec 03 07:55:35 oh Dec 03 07:55:38 now we are getting closer :) Dec 03 07:55:47 and lennart (the main head behind it) has been known to disregard critisism Dec 03 07:56:00 and also, what av500 said Dec 03 07:56:06 and we are doomed Dec 03 07:56:10 wait, Apple is doomed Dec 03 07:56:11 go on Dec 03 07:56:20 we are doomed as well! Dec 03 07:56:30 some day the sun will swallow our lovely planet! Dec 03 07:56:42 pah, we will escape Dec 03 07:56:51 on fiery rockets Dec 03 07:57:35 u guys are potheads Dec 03 07:58:18 http://www.science.uva.nl/pub/solaris/solaris2.html#q3.24 Dec 03 07:58:47 :) Dec 03 07:59:42 any of you guys used to play or still does play paper rpg games like D&D? Dec 03 08:00:09 no, I'm not that much of a pothead Dec 03 08:00:19 well I was :D Dec 03 08:02:20 KotH: Do you have any expirience in building usb hubs? From IC's like USB2514B? Dec 03 08:04:02 bremenpl: we aren't pod heads. we work in embedded Dec 03 08:04:33 its hard to tell the difference :P Dec 03 08:05:05 there is no difference ;) Dec 03 08:08:52 bremenpl: talking about RPG: http://en.wikipedia.org/wiki/IBM_RPG Dec 03 08:09:46 guten morgen woglinde! Dec 03 08:10:13 god, you guys are not only potheads, but dinosaurs :P Dec 03 08:10:34 dinosaur? me? i'm not even as old as av500 and he is still a baby! Dec 03 08:10:58 how old are u guys? Dec 03 08:11:17 i'm 8 years younger than av500 Dec 03 08:12:50 that gives kind of an image Dec 03 08:17:08 if you do things right, in 10 years time you will be as nerdy and dinsaury as we are Dec 03 08:21:51 :D Dec 03 08:22:29 hi koth Dec 03 08:23:02 koth lol Dec 03 08:35:29 woglinde: hey! it's true! Dec 03 08:35:54 woglinde: 10 years are plenty of time to become old and grumpy Dec 03 08:36:14 woglinde: and learn everything about computers Dec 03 08:36:31 heck, i didnt own a computer 15y ago Dec 03 09:05:31 I owned one Dec 03 09:08:37 computers are overrated Dec 03 09:10:53 why did you bully one to do your math homework then? Dec 03 09:11:27 it only took me half of a computer science degree to start hating it Dec 03 09:13:09 to hate CS? i didnt have to study it for that Dec 03 09:13:41 you managed to hate it without learning anything at all about it? Dec 03 09:13:45 that's impressive Dec 03 09:14:04 i didnt say that Dec 03 09:41:19 electronic engineers also take a sip of computer science at university Dec 03 09:42:11 depends on the university Dec 03 09:42:32 sure Dec 03 09:42:34 at mine, we only took CS classes if we needed some cheap and easy to do classes Dec 03 09:43:15 :) Dec 03 09:43:18 well, we had programming as well as electrical courses Dec 03 09:43:29 yes, I learned pascal and C I think Dec 03 09:43:38 or maybe only C Dec 03 09:43:59 yeah we had c too Dec 03 09:44:11 * KotH did not learn c++ and java Dec 03 09:44:22 me too Dec 03 09:44:42 well, now im at electrical engineering master course and i only got matlab lol Dec 03 09:45:59 first programming language that we learn at school was foxpro Dec 03 09:46:26 no assembly? Dec 03 09:47:48 * woglinde learned the bases for software engeneering Dec 03 09:53:46 bremenpl: for an EE, matlab is the most important language Dec 03 09:53:54 bremenpl: everything else is stamp collecting Dec 03 09:54:06 yeah, I agree its easy to implement eaverything there (and fast) Dec 03 09:54:25 its just after iniversity its not used anymore Dec 03 09:55:37 not true Dec 03 09:55:53 maybe in big companys, that can afford matlab Dec 03 09:55:59 not true eitehr Dec 03 09:56:05 matlab is quite cheap Dec 03 09:56:08 well, thats how it is in poland Dec 03 09:56:28 not a single company uses it, because you can implement simulations in cheaper software Dec 03 09:56:48 that's true Dec 03 09:56:56 also depends what you do, for most things spice is ok. We got spice in altium Dec 03 09:57:16 scilab and octave are quite good replacements for the basic matlab functionality Dec 03 09:57:28 spice is an entirely different beast Dec 03 09:57:31 yeah scilab- thats the thing some use it instead of matlab Dec 03 09:57:41 also, spice was (and is) an OSS procejt, originally Dec 03 09:58:00 oss? Dec 03 09:58:21 open source software Dec 03 09:58:47 ah yea Dec 03 09:59:14 alghorithms are open Dec 03 09:59:43 matlab and spice do different things Dec 03 09:59:52 not sure you can do a kalman filter in spice Dec 03 10:00:16 you can Dec 03 10:00:25 it's damn awkward, but you can Dec 03 10:00:31 tunring complete is complete Dec 03 10:00:33 so yes Dec 03 10:00:38 turnip Dec 03 10:14:08 Guys, do you know any interrupts related tutorial more oriented to bbb? I have checked the ldd3 but its bit to general for my knowledge now and I dont hate to much time. Dec 03 10:14:27 what info are you looking for? Dec 03 10:14:36 its a linux system Dec 03 10:14:45 linux interrupt knowledge applies Dec 03 10:14:48 like example modules for hadling ext interrupts on bbb etc Dec 03 10:15:00 hmm? Dec 03 10:15:18 bremenpl: there is some quite concise irq docu in Documentation Dec 03 10:15:38 whoch documentation? Dec 03 10:15:52 bremenpl: read on request_irq() Dec 03 10:15:52 /usr/src/linux/Documentation/ Dec 03 10:16:05 ohh, ok. Thanks Dec 03 10:16:16 * KotH is probably the last one who still keeps a /usr/src/linux directory Dec 03 10:16:37 * KotH feels old and grumpy Dec 03 10:16:47 * KotH waves cane at fog Dec 03 10:16:48 bremenpl: linux has a generic irq layer. you need to read on linux irq stuff. not bbb Dec 03 10:17:52 KotH: /usr/src/linux is still very common. my distro even installs it for me Dec 03 10:18:17 derRichard: but it's unlikely to contain the full linux source Dec 03 10:18:22 derRichard: just the headers these days Dec 03 10:18:22 Yes but I need to setup the interrupt in hardware- thats not a problem to set the registers for the MCU, the thing is how to connect that to the userspace software Dec 03 10:18:51 derRichard: which has been superseeded by /lib/modules/**/headers or something Dec 03 10:18:57 KotH: rw@sandpuppy:~/linux (master $)> rpm -qf /usr/src/linux/Documentation/IRQ.txt Dec 03 10:19:00 kernel-source-3.12.0-34.1.ge8fa6b4.noarch Dec 03 10:19:08 derRichard: interesting! Dec 03 10:19:15 get a sane disro Dec 03 10:19:18 * derRichard hides ;) Dec 03 10:19:19 derRichard: lennart hasnt killed that yet? Dec 03 10:19:29 /care Dec 03 10:19:34 :) Dec 03 10:19:53 bremenpl: read the docu of the kernel Dec 03 10:20:01 bremenpl: it explains it all (ok, almost all) Dec 03 10:22:49 KotH: Ok, ill try this Dec 03 10:23:20 KotH: Its just- it makes me wonder I havent found anyone doing this on bbb. Dunno if its so hard, not worth it or I dont know Dec 03 10:23:22 bremenpl: you dont connect interrupt to user space Dec 03 10:23:31 and don't even think if messing with /dev/mem as some "howtos" suggest Dec 03 10:23:36 I know its a file operation thing Dec 03 10:23:37 IRQs stay in the kernel Dec 03 10:23:46 and your driver handles them Dec 03 10:23:53 the driver might have an API to user space Dec 03 10:24:04 well yeah some ppl say dev/mem is faster. I use a library for gpio that uses it Dec 03 10:24:25 of course its faster Dec 03 10:24:28 you bypass the kernel Dec 03 10:24:33 and access the raw hardware Dec 03 10:24:37 yeah, the interrupts are ment to be fast too Dec 03 10:24:44 have fun messing with interrupts this way Dec 03 10:24:49 they are fast Dec 03 10:24:52 inside the kernel Dec 03 10:24:57 not in user space Dec 03 10:25:03 its not a microcontroller Dec 03 10:25:11 i know Dec 03 10:25:17 av500: well, it is.. a big uC but it is :) Dec 03 10:25:24 but lets consider this example Dec 03 10:25:30 bremenpl: see UIO Dec 03 10:25:45 On an external interrupt I need to stop a motor, if the interrupt is slow a motor wont stop in right place Dec 03 10:25:46 using uio you can create a sane kernel stub to have a driver in userspace Dec 03 10:26:12 for this the userspace gpio stuff should be fast enough Dec 03 10:26:13 bremenpl: write a small kernel driver that handlse the hw and time critical stuff, and do everything else in userspace Dec 03 10:26:14 bremenpl: then you need to handle that in the kernel Dec 03 10:26:17 not in user space Dec 03 10:26:22 user space can block for any time Dec 03 10:26:31 or handle it in the PRU on the BBB Dec 03 10:26:35 that is BBB specific Dec 03 10:26:38 and raw hardware Dec 03 10:26:55 maybe we ask him first for his timing constraints? Dec 03 10:27:10 so the texas cpu that bbb uses can have a hardware set to do certain things with hardware if some events occur? Dec 03 10:27:23 time is not too important Dec 03 10:27:35 but important enough :P Dec 03 10:27:47 theres a table driven by a stepper motor Dec 03 10:27:53 use the PRU, it's made for this Dec 03 10:29:31 bremenpl: what are you doing today? Dec 03 10:29:42 bremenpl: yesterday it was video, today stepper motors... Dec 03 10:29:54 tomorrow he learns french Dec 03 10:30:03 very likely Dec 03 10:30:17 well Dec 03 10:30:24 i have a really, really big project Dec 03 10:30:35 the things I need to do are in a high amount Dec 03 10:30:44 http://www.metrolyrics.com/machine-lyrics-violent-femmes.html Dec 03 10:30:52 bremenpl: university? Dec 03 10:30:52 i had to create this cape boar from strach and now i need to program it Dec 03 10:31:03 yes its my masters work Dec 03 10:31:07 * av500 likes boars Dec 03 10:31:13 makes for a nice roast Dec 03 10:31:19 av500: +100 Dec 03 10:31:24 =p Dec 03 10:31:37 I cant handle the video for now, so I am trying to get other things done Dec 03 10:31:49 what is the scope of it all? Dec 03 10:31:52 video and interrupts are main obstacles for me Dec 03 10:32:00 well its a driving system Dec 03 10:32:08 that drives motors Dec 03 10:32:11 bremenpl: you created a board from starch? Dec 03 10:32:23 meant to be used in some university machine its a part of it Dec 03 10:32:24 * KotH is seriously impressed Dec 03 10:32:41 ok, not such stratch :P Dec 03 10:33:23 most important factor is to drive a stepper motor and read the encoder signals using eQEP Dec 03 10:33:44 the table that moves has their maximum and minimum position, that is indicated by a sensor Dec 03 10:34:02 if the table covers the sensor its in minimum position and an interrupt should be handled Dec 03 10:34:08 ok Dec 03 10:34:09 an external interrupt Dec 03 10:34:13 so do that on the PRU Dec 03 10:34:16 not with linux Dec 03 10:34:19 thats why its a bit time critical Dec 03 10:34:20 thats the wrong approach Dec 03 10:34:26 I see Dec 03 10:34:46 but I should know in userspace that the event occured, even if the motor was stopped in hardware Dec 03 10:34:57 sure Dec 03 10:34:58 so i can handle the state machine etc Dec 03 10:35:03 user space can talk to PRU Dec 03 10:35:10 Thats right Dec 03 10:35:18 but I would do the whole control loop on the PRU Dec 03 10:35:24 and you control it high level Dec 03 10:35:30 like drive to 1024 at speed 16 Dec 03 10:35:41 sound ok Dec 03 10:35:55 also make that part standalone and open source Dec 03 10:36:01 but can I set PRU do it such specific tasks? Dec 03 10:36:01 useful for others Dec 03 10:36:13 pru is a 200mhz general purpose CPU Dec 03 10:36:17 yeah if that work out i would post it on hackaday Dec 03 10:36:18 2 cores in fact Dec 03 10:37:27 http://www.element14.com/community/community/designcenter/single-board-computers/next-gen_beaglebone/blog/2013/09/14/bbb--super-accurate-small-motor-control-with-a-beaglebone-black Dec 03 10:37:58 thank you Dec 03 10:38:00 has all you need to get started on PRU Dec 03 10:38:05 thank sergey and larry Dec 03 10:38:14 they surf the web all day and write it down Dec 03 10:38:30 who are they Dec 03 10:38:43 :) Dec 03 10:38:47 google knows Dec 03 10:38:58 the code has no license file, so you might ask the guy Dec 03 10:39:04 or only use it for inspiration Dec 03 10:39:08 ok Dec 03 10:44:03 wait a minute Dec 03 10:44:15 PRU can be programmed like a regular mcu? Dec 03 10:44:57 ohh prucode is in assembly, this is getting more fun every hour :O Dec 03 10:47:34 yes Dec 03 10:47:39 but you are not the first Dec 03 10:47:46 to write motor control Dec 03 10:47:49 or to the PRU Dec 03 10:48:03 just ask sergey and larry Dec 03 10:49:31 u mean google :P? Dec 03 10:49:34 also: http://processors.wiki.ti.com/index.php/PRU_Projects Dec 03 10:49:39 never heard of google Dec 03 10:49:42 I should google that Dec 03 10:51:33 thanks for the link Dec 03 10:52:09 https://www.google.com/search?q=PRU+motor+control Dec 03 11:14:26 so like to verify if I get this straight- The PRU is a programable module that I can communicate with from userspace and I can compile regular c code to a .bin file to program it also from userspace? Dec 03 11:15:39 the PRU is a CPU Dec 03 11:15:43 programmed in assembly Dec 03 11:15:49 linux can talk to it Dec 03 11:15:55 via shared memory Dec 03 11:16:05 and thus in the end user space can also talk to it Dec 03 11:16:17 there is also a c compiler Dec 03 11:16:20 so I can only use assembly langue, cant compile c code? Dec 03 11:16:23 ahh Dec 03 11:16:23 I dont know how useful it is Dec 03 11:16:40 but you have all the above listed PRU projects to study Dec 03 11:16:44 but is the base principal to have the PRU code dead simple and fast? Dec 03 11:16:49 sure Dec 03 11:17:00 you can do what you want on the PRU :) Dec 03 11:17:08 I was wondering either it is possible to for example debug the assembly code inside pru Dec 03 11:17:57 if a user space application can be reached externally via tcp for debbuging, can PRU be as well? Dec 03 11:18:47 https://github.com/mmcdan/visualpru Dec 03 11:18:48 maybe Dec 03 11:18:51 no idea Dec 03 11:19:37 heh this one looks cool Dec 03 11:19:55 looks are not all Dec 03 11:42:03 this pru seems to be the answer to all my problems, i need to make it wooork :< Dec 03 11:44:11 bremenpl: yes, some work is expected from you Dec 03 11:54:17 av500: "some assembly required" Dec 03 11:54:28 yup Dec 03 11:54:58 bremenpl: pru is not a solution for everything :) Dec 03 11:55:10 getting it right is rather hard Dec 03 11:55:31 implementing device drivers in pru is not fun and communicating with kernel drivers is also not trivial Dec 03 11:57:49 derRichard: getting things done is hard, news at 11 Dec 03 11:58:41 KotH: fair point :) Dec 03 12:10:13 i gues if i wanted easy i would write in c# for a living Dec 03 12:11:06 node.js Dec 03 12:15:43 cloud9 Dec 03 12:16:48 wolke7 Dec 03 12:25:06 KotH: you're german? Dec 03 12:26:08 wötsch me beleidige? Dec 03 12:26:13 mi* Dec 03 12:26:44 so you must be austrian :D Dec 03 12:26:49 alpenkäskopp Dec 03 12:26:51 LOL Dec 03 12:26:56 even worse insult Dec 03 12:27:17 derRichard: be carefull or i get av500 to kick you! Dec 03 12:27:22 nah Dec 03 12:27:29 I like his way of thinking Dec 03 12:27:30 av500: you'll get a bar of chocolate Dec 03 12:27:38 oh Dec 03 12:27:40 ok Dec 03 12:27:41 schwitzerdeutch Dec 03 12:27:43 3 Dec 03 12:27:46 2 Dec 03 12:27:50 1 Dec 03 12:27:58 KotH: where is it? Dec 03 12:28:04 isn't av500 bavarian? Dec 03 12:28:10 the amazone drone got lost Dec 03 12:28:20 so many insults today Dec 03 12:28:34 * derRichard is from tyrol, so you can call me schluchtenscheißer :P Dec 03 12:29:04 secretly, we already do Dec 03 12:29:26 :) Dec 03 12:36:56 Hi is anyone there Dec 03 12:37:01 no Dec 03 12:37:05 all gone for lunch Dec 03 12:37:08 Ha Dec 03 12:37:39 all gone for sun Dec 03 12:37:40 I need help cloning a git repository to my beagle bone black from my macbook Dec 03 12:37:50 haeh? Dec 03 12:38:00 I'm getting a http error, and can't even ping google Dec 03 12:38:06 can anyone help Dec 03 12:38:19 setup network correctly Dec 03 12:38:35 I thought I had Dec 03 12:38:44 when you cannot ping Dec 03 12:38:46 you havent Dec 03 12:38:47 "cant even ping google" Dec 03 12:38:58 maybe google is at lunch? Dec 03 12:39:03 I know that now, but can't see what I have done wrong Dec 03 12:39:07 can you ping 8.8.8.8? Dec 03 12:39:26 no Dec 03 12:39:32 says "network is unreachable" Dec 03 12:39:44 This is the first time I have used this and just need help figuring it out Dec 03 12:40:54 ? Dec 03 12:41:09 hint: actually ask a question. Dec 03 12:41:21 Sonny_: stupid question: did you plug it in? Dec 03 12:41:33 Sonny_: and if yes, what does ifconfig say? Dec 03 12:42:13 and it is no bbb question at all Dec 03 12:42:40 also, what is plugged into what? Dec 03 12:43:31 sie müssen nur den nippel durch die lasche ziehn.....*sing* Dec 03 12:43:59 eth0 Link encap:Ethernet HWaddr d0:39:72:4e:b0:20 inet6 addr: fe80::d239:72ff:fe4e:b020/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:3 errors:0 dropped:0 overruns:0 frame:0 TX packets:69 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:282 (282.0 B) TX bytes:14289 (13.9 KiB) Interrupt:40 Dec 03 12:44:13 Thank you @av500 I appreciate your help on this Dec 03 12:44:25 Sonny_: what does this tell you? Dec 03 12:44:46 I have a mini usb to usb going from BBB to Mac, and an ethernet to ethernet Dec 03 12:45:35 ah... Dec 03 12:45:40 Sonny_: look at networking 101 Dec 03 12:45:43 KotH Im not sure this is my first time using such things. there is no loopback? Dec 03 12:45:56 you dont want loopback, you want proper routing Dec 03 12:45:58 Is that a site KotH? Dec 03 12:46:08 Sonny_: ethernet does not magically work if you plug a cable into two devices Dec 03 12:46:21 101 = first year class on a certain topic Dec 03 12:46:49 Ok thank you Dec 03 12:46:53 is it clear to you whats wrong? Dec 03 12:46:57 yes Dec 03 12:47:06 what LetoThe2nd said Dec 03 12:47:09 Sonny_: if you don'T want to really learn how it works, plug both devices into a network that a) offer proper routing b) offers dhcp Dec 03 12:47:22 e.g. grab a cheapo router. Dec 03 12:47:29 again what LetoThe2nd said Dec 03 12:47:52 derRichard: btw: LetoThe2nd ist unser quoten bayer Dec 03 12:48:05 KotH: quotensandworm, please Dec 03 12:48:15 LetoThe2nd it first must works on layer1 Dec 03 12:48:27 hey ich bin auch ein quotenbayer. naja. frange. :D Dec 03 12:48:41 to much german sentences today Dec 03 12:48:43 KotH and LetoThe2nd: Thank you for your pointers I will look in to it now Dec 03 12:48:51 LetoThe2nd: sorry, my fault Dec 03 12:49:01 Sorry I come across as such an amateur, but I do so because I am Dec 03 12:49:04 derRichard: btw: LetoThe2nd ist unser quotensandwurm Dec 03 12:49:36 Sonny_: nobody blames you for that Dec 03 12:50:05 Sonny_: but you have a problem at a very basic level, that is exected knowledge when you deal with linux or networked devices in general Dec 03 12:50:18 KotH: mkay :D Dec 03 12:52:20 so we need #beagle_südlich_vom_main Dec 03 12:52:32 so we need #beagle_suedlich_vom_main Dec 03 12:52:54 and #beagle_chocolate_lovers Dec 03 12:53:41 though I'm a Hibbdebächer Dec 03 12:54:11 btw: remeber my questions on the dht11/22 sensor? with my patches applied it works fine on the bbb Dec 03 12:54:14 https://lkml.org/lkml/2014/12/2/868 Dec 03 12:54:54 derRichard: oh, you got that running? nice Dec 03 12:55:11 sure Dec 03 12:55:50 * tbr should play with his somewhen Dec 03 12:56:17 tbr: using humidty as air quality measure? :) Dec 03 12:56:31 KotH: on the toilet ;) Dec 03 12:57:23 tbr: well, it'll measure 100%. always. you know? ;) Dec 03 12:58:20 well. measuring 100% air humidity is easy Dec 03 12:58:28 bbb is broken: 100% Dec 03 12:58:32 bbb is not broken: not 100% Dec 03 12:58:32 :P Dec 03 13:02:25 hey! IoT is a serious application! Dec 03 13:06:01 i ve no idea what the iot fuzz is all about -_- Dec 03 13:06:17 like my fridge would need a lan connection. Dec 03 13:06:29 iot is cool Dec 03 13:06:40 havent you subscribed to twitter yet? Dec 03 13:06:57 https://twitter.com/iotoilets Dec 03 13:07:01 :D Dec 03 13:07:38 what the ... ^^ Dec 03 13:10:34 welcome to the brave new world of IoT Dec 03 13:14:22 no one has connected that to public toilets at airport? Dec 03 13:15:13 hitlin37: we had the idea to do a massive deployment during ELCE in duesseldorf Dec 03 13:16:08 well, fosdem is approaching. Dec 03 13:16:12 but currently it's just connected to the big white telephone down the hallway, in relation to me sitting on my sofa Dec 03 13:26:04 tbr: big white bakelite telephone? with rotary dial? Dec 03 13:48:31 btw guys, are you here everyday? Dec 03 13:50:50 not on Dec 24th Dec 03 13:51:08 av500: which year? Dec 03 13:51:39 nice Dec 03 13:52:03 tehn i gonna be here everyday too, and we are going to be like famili, and after a while i am also going to make systemd jokes Dec 03 13:52:16 but we love systemd Dec 03 13:52:34 you dont only joke from what you hate Dec 03 13:53:13 not sure we want you here all the time :) Dec 03 13:53:20 how about tuesdays and fridays? Dec 03 13:54:09 i feel like parents devorce al over again Dec 03 13:57:33 bremenpl haha your hate comes from missunderstandings and FUD Dec 03 13:57:43 systemd service very well Dec 03 13:57:50 woglinde: nobody hates here Dec 03 13:57:56 I will not allow it! Dec 03 13:58:19 i dont hate anyone/ anything :P Dec 03 13:58:27 than all is fine Dec 03 14:00:51 hi all, i have a Q about serialport in node.js and need a bit of guidance. i'm reading the serail port and but need a way to read at my own request not on an event.. Dec 03 14:01:38 node.js is evnet driven, no? Dec 03 14:02:59 i g2g guys. thanks for help and se you tommorow :> Dec 03 14:04:01 thanks for replying, i understand node.js is event driven i would just like to make a request then have the read to happen in a call back Dec 03 14:05:20 I have no idea about node.js Dec 03 14:06:07 at the moment. the event with my data is not called ever time there is new data read to read. Dec 03 14:24:32 sdavey: you could probably set up a a polling routine with setTimeout if that's what you're trying to do Dec 03 14:24:55 probably not the best way to handle things in node though Dec 03 14:38:12 hi Dec 03 14:38:21 i am new to beaglebone Dec 03 14:39:05 welcome to the club Dec 03 14:39:19 may i introduce you to av500, he is our local bear Dec 03 14:39:35 overthere sits LetoThe2nd, he is our local sandworm Dec 03 14:40:14 and hiding behind him is mdp, he is ...well.. Dec 03 14:41:53 well, i didnt get what you said. But still Dec 03 14:42:20 i dont know how to store analog outputs in the sd card of my BBB Dec 03 14:42:28 can you help me with it Dec 03 14:44:24 what is an "analog output"? Dec 03 14:44:55 a sensor output such as a temperature sensor Dec 03 14:45:41 i want to measure the temperature of my room and store that value in sd card as a text file using my GPIO Dec 03 14:45:45 ah.. Dec 03 14:45:47 Jags: LM75? Dec 03 14:45:54 ya LM35 Dec 03 14:46:06 look up the following therm "ADC" Dec 03 14:46:21 you are not the first one who does that Dec 03 14:46:37 isn't LM35 5V? Dec 03 14:46:58 ya i know that..i have been surfing the net for long to try and figure out a way to store data in SD card Dec 03 14:46:59 Humpelstilzchen: depends on the temperature ;) Dec 03 14:47:13 right Dec 03 14:47:28 but all search engines give results of booting the BBB with an OS in the SD card Dec 03 14:48:08 Jags: uh.. Dec 03 14:48:15 yes LM35 is a 5V sensor Dec 03 14:48:22 Jags: do you have basic linux knowledge? Dec 03 14:48:32 Jags: and basic electronics knowledge? Dec 03 14:48:37 yes, a lil bit Dec 03 14:49:12 then what part of writing data to an SD card is the problem? Dec 03 14:50:02 well, i am able to create a text file in the SD card Dec 03 14:50:07 yes Dec 03 14:50:10 good Dec 03 14:50:15 but no data is being stored Dec 03 14:50:27 oh..kay... Dec 03 14:50:32 let me guess Dec 03 14:50:33 and what if i add a BMP085 to it Dec 03 14:50:37 you just pull the sd card out? Dec 03 14:50:45 without unmounting it first Dec 03 14:50:46 it is a barometric pressure sensor Dec 03 14:50:58 and uses I2C Dec 03 14:51:03 ok Dec 03 14:51:06 you are asking very basic electronics questions Dec 03 14:51:07 pulled out Dec 03 14:51:14 and i guess your linux knowledge is about the same level Dec 03 14:51:21 yes i said i am new to Beagle bone Dec 03 14:51:22 yes, it definitly is Dec 03 14:51:27 no, you are new to computers Dec 03 14:51:29 and electronics Dec 03 14:51:37 please learn linux and electronics first Dec 03 14:51:41 thank you Dec 03 14:51:48 Jags: ever tried to write a file under linux on your desktop pc? Dec 03 14:51:59 Humpelstilzchen: he wont see the problem there Dec 03 14:52:16 ok where can i get to learn in depth about linux. Dec 03 14:52:18 Humpelstilzchen: he doenst pull out the harddisk without shuting down the pc first Dec 03 14:52:20 it's been a long time since I've seen anybody consider temperature measurement as an analog task. considering that our building blocks have long hidden that detail from us. Dec 03 14:52:27 "detail" Dec 03 14:52:37 mdp: really? Dec 03 14:52:46 mdp: you should talk to time-nuts more often Dec 03 14:53:07 mdp: you cannot do temp control down to 0.001§C with digital control :) Dec 03 14:53:18 0.001°C Dec 03 14:53:29 KotH: I know of them Dec 03 14:53:45 mdp: i think i know one of those you know ;) Dec 03 14:53:46 KotH: they are the 0.00000001%ers, if you will Dec 03 14:53:50 hehe Dec 03 14:53:52 yes Dec 03 14:53:54 KotH: freaks :P Dec 03 14:54:09 please help me. I am interested to learn more Dec 03 14:54:17 mdp: i recently got asked about how to control the temp of a room to better than 1°C Dec 03 14:54:33 mdp: apparently, that's hard enough that noone offers it commercially Dec 03 14:54:41 Jags: get a good linux book Dec 03 14:54:43 Jags: there are many Dec 03 14:54:54 Jags: there are also many books on how to learn electronics Dec 03 14:55:07 Jags: i usually recommend the book by tietze & schenk Dec 03 14:55:08 ok. Can you name a few. Dec 03 14:55:10 KotH: imagine the cooling/heating panels you need :o Dec 03 14:55:16 I will get that right away. Dec 03 14:55:26 day: so? Dec 03 14:55:34 you cant just blow air into the room sub the target temp Dec 03 14:55:40 day: we are talking about equipment that costs millions in parts Dec 03 14:55:46 ah i see Dec 03 14:55:57 day: and yes you can Dec 03 14:56:02 then its obviously not that difficult :P Dec 03 14:56:04 day: everything is running at steady state Dec 03 14:56:14 day: the problem starts when people enter Dec 03 14:56:28 day: that shifts the power flowing into the room by 100W Dec 03 14:56:40 KotH: hm if you blow air in sub target temp (>1°C difference) you break the rules Dec 03 14:56:55 day: in a room that usually goes by in the 10kW range Dec 03 14:57:27 day: you dont talk about blowing air in at a specific temp Dec 03 14:57:34 any good book to start with linux? Dec 03 14:57:43 day: you talk about the amount of heat energy you take out Dec 03 14:57:50 Jags: do you know how to google? Dec 03 14:58:34 KotH: true, but the fastest way to do that is pumping air into the room with lower temperature Dec 03 14:58:48 'fastest feasable way' Dec 03 14:58:48 Jags: sorry if i sound unfriendly, but i have a short temper with indans... and you just confirm my prejudies Dec 03 14:58:55 ya but i wouldnt know if the books that are shown by google are the ones i need to read.. Dec 03 14:59:09 the problems are the airbubbles with different tempertures Dec 03 14:59:12 i would rather be wasting my time if i read the unnecessary books Dec 03 14:59:15 day: yes. that's the only way you can take out heat energy :) Dec 03 14:59:25 Jags: read reviews Dec 03 14:59:26 KotH: not really Dec 03 14:59:42 day: you cannot take heat out by blowing in hot air ;) Dec 03 14:59:46 you could plaster the walls with thermoelectric elements :P Dec 03 15:00:08 oh, absolutely no problem KotH..i dont mind Dec 03 15:00:18 day: if that wouldnt be so expensive, people would do it :) Dec 03 15:00:24 day: actually, people do that Dec 03 15:00:31 well you said multimillions :P Dec 03 15:00:54 day: hydrogen masers and other stuff that is sensitive to temp variations are in boxes that have termoelectric coolers for temp stabilization Dec 03 15:01:23 to get the maser outside stabilized to <0.1°C Dec 03 15:01:45 thanks a lot guys..i will be back Dec 03 15:17:11 Hi, does anybody know: Dec 03 15:18:31 KotH: btw. how big is this room we are talking about? Dec 03 15:19:37 day: depends Dec 03 15:20:11 day: the swedish just converted an old bunker into a lab, which is around 10m^2 and iirc 3m high Dec 03 15:20:29 day: the upper end is probbably 1000m^2 and several m high Dec 03 15:20:39 KotH: why would they even want to stabilize the temperature so precisely in such a large room? Dec 03 15:20:54 because it makes life easier :) Dec 03 15:21:05 I understand environmental control in testchambers? But the whole lab? :D Dec 03 15:21:11 well Dec 03 15:21:39 if your environment is more stable, then your "test chamber" is also more stable, because it sees less variation at its outside Dec 03 15:22:02 (also "test chamber" isnt quite what these are. these are not tests, but continously running equipment) Dec 03 15:23:53 im amazed that these machines dont stabilize the temp where its needed themselves Dec 03 15:24:57 then again..a completely temperature stabilized room via thermoelements....nerdgasm :D Dec 03 15:28:46 day: they do Dec 03 15:28:48 they do Dec 03 15:29:02 the interior of them is stabilized to sub 1mK Dec 03 15:29:29 but you have a amplification factor from isolation shell to isolation shell Dec 03 15:30:20 if you can take the temp control of the room from +/-5C to +/-1C you stabilize the temp inside the device as well by a factor 5 Dec 03 15:30:54 ie it goes from +/-1mK to +/-200uK Dec 03 15:31:55 KotH: i see. Dec 03 15:34:17 day: think about having a whole hall full of damn expensive equipment Dec 03 15:35:04 day: if you can make the hall better stabilized for a couple 100k, then that's well worth it as another isolation shell per device costs easily several 10k to a few 100k Dec 03 15:35:47 and each shell means the guy doing maintenance has another shell to deal with and open to get to the main equipment Dec 03 15:36:04 just put sweden inside an isolatin shell Dec 03 15:36:06 isolation Dec 03 15:36:15 it already is Dec 03 15:36:21 it's called "tundra" :) Dec 03 15:36:49 by that standard NK should be a research leader :P Dec 03 15:40:00 in that field? Dec 03 15:40:30 no, that's the germany (PTB), france (SYRTE), switzerland (METAS) and USA (NIST) Dec 03 15:40:47 oh.. and italy (INRIM) Dec 03 15:45:20 also USA (USNO) Dec 03 15:46:26 not anymore Dec 03 15:46:36 USNO uses mainly a bunch of 5071s Dec 03 15:46:47 that's good enough for them Dec 03 15:47:10 there are plans to upgrade to a couple of H-masers though Dec 03 15:47:37 as they will be measuring their paper clock against the galileo reference clock at INRIM Dec 03 15:47:42 ah. they're a source of time but not a research leader then Dec 03 15:47:52 yes Dec 03 15:48:36 USNO is mostly concerned with navigation Dec 03 15:49:02 KotH: PtB/Germany is good at something? :o Dec 03 15:49:06 im...amazed Dec 03 15:49:16 and the instabilities in GPS are mainly atmospheric disturbance related than stability of clocks Dec 03 15:49:32 day: they are one of the world leaders in time/frequency metrology Dec 03 15:50:10 day: as you might know, the also run the DCF77 transmitter which most of europe uses Dec 03 15:51:27 well dcf77 is kinda...meh compared to the gps time signal :X Dec 03 15:51:41 the dcf77 amplitude signal is complete trash Dec 03 15:51:59 lol Dec 03 15:52:10 the dcf77 signal is actually quite good Dec 03 15:52:25 well the error is quite high Dec 03 15:52:36 the original AM modulation is simple enough that you can decode it with very low complexity Dec 03 15:52:58 its good for consumers yes. Dec 03 15:53:04 the PSK signal is good if you are in a low reception zone or need better uncertainty Dec 03 15:53:18 But you dont need to be a leading research organisation to host a AM time signal Dec 03 15:53:40 yes. its better. But iirc. its still much worse than the GPS signal Dec 03 15:53:49 and given that WWV completly messed up their PSK signal when they introduced it 2-3 years ago, i'd say DCF77 with their 20 year unchanged signal is pretty good Dec 03 15:54:09 yes, DCF77 cannot compensate for atmospheric conditions Dec 03 15:54:21 GPS can (and does) to some extend Dec 03 15:54:46 so in essence..the us military has flying labs that work better than a ground based research station :X Dec 03 15:55:05 not really Dec 03 15:55:13 it's the transfer mechanism that sucks with DCF77 Dec 03 15:55:36 no one forced them to use it :) Dec 03 15:55:45 actually, there was Dec 03 15:56:00 no other system could get the reach with the limited budget they had Dec 03 15:56:18 GPS is a couple of magnitudes more expensive than DCF77 Dec 03 15:56:46 KotH: AM yes. PSK? The decoder look quite expensive Dec 03 15:56:54 i'm not sure whether DCF77's cost already reached that of one GPS satellite Dec 03 15:57:07 day: why does it look expensive? Dec 03 15:57:11 thats not fair :o Dec 03 15:57:23 KotH: the PSK decoder ive seen look quite complex Dec 03 15:57:33 what complex? Dec 03 15:57:40 electronics? decoding process? Dec 03 15:57:53 dunno google for one. Its not a simple circuit Dec 03 15:58:27 GPS receivers are build on mass. They ve become very cheap Dec 03 15:58:35 en mass* Dec 03 15:58:51 uhmm... Dec 03 15:58:57 i doubt the volume for DCF77 psk receivers is equally big Dec 03 15:58:58 i have a problem guys Dec 03 15:58:59 could be there are no single chip solutions Dec 03 15:59:15 but you can build a DCF77 PSK decoder using and ADC and an uC Dec 03 15:59:19 and that has been done too Dec 03 15:59:32 i see. didnt know that Dec 03 15:59:36 AhmedHelmy: state your problem and send 100g of good swiss chocolate this way Dec 03 15:59:56 i only saw the ridiculously lare decoders, and noped out :D Dec 03 16:00:00 large* Dec 03 16:00:01 heh Dec 03 16:00:42 As i'm trying to connect my beaglebone black with EasyN ip camera in a network using a router to connect beagleboard with the IP camera , but i couldn't do that at all , i want to connect them with each other to use the EasyN ip camera in face detection program using python , how can i do this project ?? thanks in advance. Dec 03 16:01:40 AhmedHelmy: step #1: check whether the network of both camera and BBB is correctly configured Dec 03 16:01:54 AhmedHelmy: setp #2: try to narrow down where your problem comes from Dec 03 16:02:08 AhmedHelmy: note #1: "does not work" is not an error description Dec 03 16:02:27 day: i've been reading on that kind of stuff for a long time Dec 03 16:02:55 day: so i've seen a few designs that look good and easy Dec 03 16:03:30 day: and i also know that most people dealing with DCF77 barely know what they are doing. it's so simple that you can do it by trial and error Dec 03 16:03:46 day: that doesnt work with gps, that's why the stuff there is designed way better Dec 03 16:04:00 KotH: is the frequency that slow that you can sample it sufficiently with a µC adc? :O Dec 03 16:04:21 Koth , #1 they are correctly configured Dec 03 16:04:58 AhmedHelmy: what does not work? Dec 03 16:05:07 you have to be specific Dec 03 16:05:12 day: 77.5kHz Dec 03 16:05:18 Koth i want to have a stream video from this camera to my beaglebone Dec 03 16:05:26 day: oversample it by a factor of 5 and you are still good with a uC Dec 03 16:05:35 in germany, schoolkids sample DCF77 by hand Dec 03 16:05:40 lol Dec 03 16:05:42 in 4th grade Dec 03 16:06:00 av500: that explains a lot about your mental health Dec 03 16:06:11 @av500 # i can't open this cam from my beaglebone Dec 03 16:06:18 define "open" Dec 03 16:07:37 "clicking on button labled 'open'" Dec 03 16:07:55 when I read open, I think screwdriver :) Dec 03 16:08:22 what should i write in command line of beaglebone , to see live stream from this cam ? Dec 03 16:09:08 ah Dec 03 16:09:46 1. Why everybody uses address 0x80001000 (base: 0x80000000 offset: 0x1000) for transfering data from PRU --> DDR? Dec 03 16:11:46 https://www.google.com/search?q=pru+0x80001000 Dec 03 16:12:07 2. Is there a recent & complete article which explains why we can not transfer more than > ~24KB from PRU DRAM --> DDR @0x80001000 ? Dec 03 16:12:57 Cengiz: maybe the address map demands that Dec 03 16:13:07 I am googling since monday... Dec 03 16:13:32 mdp: werent you looking for some distraction? Cengiz needs your help Dec 03 16:13:46 mdp: and he is a friendly turkish guy who can make you a nice Döner as well! Dec 03 16:14:07 http://e2e.ti.com/support/arm/sitara_arm/f/791/p/233891/820919.aspx Dec 03 16:14:11 does this one help? Dec 03 16:14:22 My OS is BeagleBoard.org BeagleBone Debian Image 2014-05-14 Dec 03 16:19:45 KotH: pass! Dec 03 16:21:48 mdp: no döner for you? Dec 03 16:40:18 KotH: you might find the NTP histogram local clock offset graphs interesting - http://blog.dan.drown.org/beaglebone-black-timer-capture-driver/ Dec 03 16:51:42 ddrown: please write MHz instead of mhz Dec 03 16:51:58 sure Dec 03 16:52:07 ddrown: mhz can be confused with mHz which is something completely different Dec 03 16:52:13 right Dec 03 16:54:18 ddrown: nice writeup! Dec 03 16:54:25 thanks Dec 03 16:54:26 * KotH approves Dec 03 16:55:28 ddrown: btw: anything running with 70MHz? Dec 03 16:55:47 ddrown: HZ timer or soemthing? Dec 03 16:55:58 ddrown: that would explain the n*14ns spikes Dec 03 16:56:11 I'm wondering if the 14ns/29ns spikes are from a side effect of the 41.6ns quantization (3-11 points used in linear regression) Dec 03 16:56:32 hmm.. Dec 03 16:56:44 I don't think anything is running at 70MHz, but it is on a desk with other stuff Dec 03 16:57:06 only if there is some intermodulation between the 41.6ns quantization clock and some other clock with a difference of about 70MHz Dec 03 16:57:34 anyways..gtg Dec 03 16:57:37 bbl Dec 03 16:57:39 the gps module (MTK MT3339) claims "10ns jitter" Dec 03 16:59:05 hm, I should try moving it to the other side of the house Dec 03 19:23:49 is it possible to use the hardware RTS and CTS signals for uarts? Dec 03 19:24:08 Yes. **** ENDING LOGGING AT Thu Dec 04 02:59:59 2014