**** BEGIN LOGGING AT Mon Oct 20 02:59:59 2014 Oct 20 04:12:31 how do i get the my bbb using the ubuntu 12.04.4 image from http://www.armhf.com/boards/beaglebone-black/bbb-sd-install/ do networking via usb like the default angstrom image does? i tried changing /etc/network/interfaces to https://www.refheap.com/91983 and that isn't enough Oct 20 04:18:45 do i have to do some udev rules on my laptop? Oct 20 04:38:12 I'm having problems with the BBB audio cape: namely asking for PCM_S32_LE is acting like it works but the results are strongly biased around 1e8 instead of around 0. Any idea on the debian ALSA driver? Good not for audio cape rev b? Oct 20 07:48:50 I have a BB xM that doesn't boot properly with D6 flashing rapidly and D5 off. Oct 20 07:49:00 anyone know what those LEDs indicate? Oct 20 07:50:40 i honestly dont know Oct 20 07:54:26 ಠ_ಠ Oct 20 07:56:25 D5 indicates Power Oct 20 07:56:27 :| Oct 20 07:59:41 D6 is a user led. Oct 20 08:01:10 so something's broken. Oct 20 08:06:14 mrrh. Oct 20 08:06:23 apparently D5 is just shot and it's the sdcard. Oct 20 08:10:08 note to self: take multimeter next time I go debug a beagleboard. Oct 20 08:14:53 honestly: dont you know that an oscilloscope is a required debugging tool for software? Oct 20 08:15:44 >.< Oct 20 08:15:49 computers suck Oct 20 08:15:55 I'ma become a pumpkin farmer Oct 20 08:15:58 news at 11 Oct 20 08:16:17 hi koth Oct 20 08:16:28 hoi woglinde Oct 20 08:16:36 woglinde: where were you last week? Oct 20 08:16:46 I've spent 7 years at a university studying CS and doing computery things and it seems the only substantial thing I've learned is that everything sucks. Oct 20 08:17:58 I've also learnt a great deal of wonderful things about computing that would be great if computers didn't suck so terribly. Oct 20 08:19:18 koth at elce Oct 20 08:19:20 but I am shy Oct 20 08:19:50 there we go, it's up. Oct 20 08:20:01 I gotta petition to get class 10 sdcards for these things. Oct 20 08:20:49 honestly so give up and work in kindergarten Oct 20 08:20:52 woglinde: you didnt even follow the smell of The Chocolate Oct 20 08:21:03 I was in the UAV talk of koen Oct 20 08:22:05 woglinde: and you didnt say hi? Oct 20 08:22:09 woglinde: how rude! Oct 20 08:22:15 yes I know Oct 20 08:26:01 woglinde: no, the only thing that sucks more than computers is users Oct 20 08:26:08 pumpkin farming is where it's at Oct 20 10:40:52 Hi Oct 20 10:41:01 is there a sensor board (IMU) for beagle bone? Oct 20 10:41:08 that one can buy assembled? Oct 20 10:45:28 mkad: any 3.3V IMU board should do, like altimu-10 Oct 20 11:17:45 mkad: you can use anyone that interfaces with 3.3V logic. preferably using I2C or SPI Oct 20 11:27:27 Humpelstilzchen, thats not BB cape Oct 20 11:28:03 sorry I didnt formulate my question right Oct 20 11:28:17 Is there any sensor cape (IMU cape) for beagle bone that one can buy ? Oct 20 11:31:45 mkad: have you googled? Oct 20 11:31:56 https://www.google.ch/search?q=beagle+bone+cape+IMU Oct 20 11:31:59 first hit Oct 20 11:32:03 or second Oct 20 11:34:05 KotH, yes of course I googled, but that board is not for sale yet Oct 20 11:34:24 sorry cape Oct 20 11:34:42 then i dont know Oct 20 11:34:52 do you need it for work or hobby? Oct 20 11:35:16 actually it should not be in the capelist website, if it's not produced imo.. since it is confusing... Oct 20 11:35:24 KotH, hobby Oct 20 11:35:47 mkad: why do you need a cape? Oct 20 11:35:59 drones, for what else could it be? :) Oct 20 11:36:21 but why cape footprint Oct 20 11:36:25 I have my own imu cape based on max21100 and protoshield which has breakout board on the top of it, but it is chunky takes much space and it's only IMU, I was more interested in cape that has imu + baro Oct 20 11:36:38 also dedicated cape would be smaller format Oct 20 11:36:49 altimu10 is small ;) Oct 20 11:36:51 now I have like a tower of headers stack on another Oct 20 11:36:54 Humpelstilzchen: not everyone likes to do his own pcb Oct 20 11:37:08 also, soldering some of the MID based devices can be a really pain Oct 20 11:37:24 KotH: you only need 4 wires Oct 20 11:37:34 KotH, I soldered this https://www.dropbox.com/s/r1n4ihivnodqp2e/Max21100.png?dl=0 Oct 20 11:37:44 KotH, 2x2mm Oct 20 11:38:02 mkad: 0.5mm pitch qfn? Oct 20 11:38:18 * KotH is impressed Oct 20 11:38:32 dont remember probably 0.5 Oct 20 11:38:38 even more so if you handsoldered it ;) Oct 20 11:38:47 yeah hand soldering Oct 20 11:38:54 first attempt and works Oct 20 11:38:59 woha! Oct 20 11:39:03 the problem is that now I have like a tower of headers stack one on to another Oct 20 11:39:14 and I dont like it Oct 20 11:39:57 Humpelstilzchen: i dont like the magnetometer being so cramped in that board Oct 20 11:40:17 been busy updating my Ublox gps cape this is what i got to now https://www.dropbox.com/s/gvdra0htj5xiy14/Screenshot%202014-10-20%2014.25.00.png?dl=0 Oct 20 11:40:37 KotH: still works, as long as you keep it 10cm away of everything ;) Oct 20 11:40:46 another gps cape? Oct 20 11:41:03 hewball: MAX? Oct 20 11:41:11 someone should really do imu cape :( Oct 20 11:41:21 with baro Oct 20 11:41:32 KotH: yeah max7 is what im using Oct 20 11:41:39 hewball: also useing an external power supply? Oct 20 11:41:46 using them for PPS gps Oct 20 11:42:01 KotH: nah thats got a dc to dc reg on there now too Oct 20 11:42:07 will take upto 30V Oct 20 11:42:07 KotH, what is the best way to read PWM singal from RC radio on BB ? Oct 20 11:42:08 uhmm.. there isn't a T variante of the MAX7, or is there? Oct 20 11:42:25 I guess ADC is not the best idea Oct 20 11:42:26 using the max7-q myself Oct 20 11:42:27 mkad: capture/compare units Oct 20 11:42:47 ok I will have a look if PRUs can do that Oct 20 11:42:49 thx Oct 20 11:42:55 hewball: ah.. how big is the jitter after sawtooth correction on those? Oct 20 11:42:58 hewball: 10ns? Oct 20 11:43:12 mkad: PRU is overkill Oct 20 11:43:18 mkad: use the capture compare units Oct 20 11:43:24 mkad: simpler to workwith too Oct 20 11:43:29 thanks Oct 20 11:44:16 yeah my PRU code does no work well, I can measure only output on one pin, I guess there is something with layout I might need to change Oct 20 11:44:40 KotH: yeah 10-20 i think is normal from what i see Oct 20 11:45:07 hewball: even after sawtooth? Oct 20 11:45:19 hewball: and how good is your antenna location? Oct 20 11:45:36 thats waht ntpq shows me anyway Oct 20 11:45:47 ah.. Oct 20 11:46:01 that's probably without sawtooth correction then Oct 20 11:46:18 * KotH has to build his GPSDO somewhen Oct 20 11:46:29 * KotH needs more time Oct 20 11:46:30 im reading pps with kernel and using that for ntp Oct 20 11:46:34 * KotH blames av500 Oct 20 11:47:00 hewball: are you on the time-nuts mailinglist? Oct 20 11:47:25 nope,this is just a hobby for working towards mlat with adsb Oct 20 11:47:40 current version im running http://hewball.com.au/hewballwordpress/wp-content/uploads/2014/08/2014-08-05-13.12.04.jpg Oct 20 11:47:45 hewball: mlat? Oct 20 11:48:04 multilateration Oct 20 11:49:10 hm? i dont see the connection of mlat with ads-b Oct 20 11:49:49 mlat allows positioning of say helicopters that dont have to broadcast position only height Oct 20 11:50:02 ah.. Oct 20 11:50:35 btw: i suggest you add more C to the power input of the MAX7 Oct 20 11:50:45 a 1n + 100n + 1u is the usuall approach Oct 20 11:50:54 give you more stable power -> better reception Oct 20 11:51:24 also, the SMA connector needs a standof, otherwise it shorts (RF) shorts the antenna signal Oct 20 11:52:10 also, i'm not sure the antenna track is actually 50R Oct 20 12:38:58 KotH, by the way, does it really matter how good is the latency for controlling ESC/motors via PRUs/external mcu or via PWM BB's ports ? Oct 20 12:39:50 mkad look at the ELCE talk about using the PRU Oct 20 12:39:59 KotH, or is it more about realtime, that OS might not provide that Oct 20 12:40:10 woglinde, thanks will google for it Oct 20 12:40:39 mkad or look at the RTmux talk too if you need more RT for motors Oct 20 12:41:35 mkad: your latency in the AM335x will be less than your mechanical system can react Oct 20 12:41:58 mkad: ie you can runn shorter loop times than your system can handle Oct 20 12:42:20 mkad: also, if you are going that way: pretty please have a look at some control theory textbooks Oct 20 12:42:59 KotH, you mean AM335 is too fast for mechanical system, or too slow... Oct 20 12:43:09 mkad: neither Oct 20 12:43:41 mkad: have a look at koen kooi's talk at this years ELCE Oct 20 12:43:52 ok Oct 20 12:43:54 thanks Oct 20 13:20:11 mkad, you could look at the way the ardupilot guys are doing it on the beaglebone black. Oct 20 13:59:31 eh Oct 20 13:59:55 mistawright: mkad : the last time i took a look at ardupilots bbb code its just been a mess Oct 20 14:00:08 i strongly recommend not to copy anything from there Oct 20 14:00:35 generally: bbs pwm pins are totally fine for controlling escs Oct 20 14:00:37 did this a lot Oct 20 14:10:47 anyone have experience with audio cape rev b? Oct 20 14:25:38 gurki, yup that's why I wanted to write my own PRUs code Oct 20 14:26:20 gurki, but I am busy at work, and it's a hobby project, so I need to research a bit more this topic if this is really needed... Oct 20 14:26:53 gurki, which means watching those talks mentioned by KotH :) Oct 20 14:27:19 this is not going to happen till weekend I guess;) Oct 20 14:29:33 pfff... Oct 20 14:29:35 screw work Oct 20 14:29:39 watch talks! Oct 20 14:31:57 lol Oct 20 14:32:31 skip work do hobby instead Oct 20 14:33:45 If Only. Oct 20 14:34:13 I kinda lucked out with this current contract. They are paying me to do BBB related work. So I am effectively getting paid to work on a hobby. Oct 20 14:34:55 * MK_FG waits for someone to suggest pumpkin farming instead Oct 20 14:35:45 mine does crossover alot which is nice Oct 20 14:35:51 There's some land for sale or lease in central california just waiting for pumpkins! Oct 20 14:36:12 VirG, Are you building an army of drones to subjugate humanity using BBBs? Oct 20 14:36:35 ssshhhhh Oct 20 14:36:45 no talkee secret stuff. Oct 20 14:37:04 BBB enabled pumpkins! Oct 20 14:37:16 Oh wow Oct 20 14:37:19 but my interest in gps is totally not work related Oct 20 14:39:30 hewball: hint: do not go down the path i've gone Oct 20 14:39:40 hewball: you will not return the same as you were Oct 20 14:40:19 supporting users has to be worse Oct 20 14:40:21 which reminds me, i need to buy more gps antennas Oct 20 14:40:47 can never have enough Oct 20 14:40:58 yeah.. especially L1/L2 antennas Oct 20 14:41:58 i just need to find a different bank account to support my hobby Oct 20 14:42:10 your boss'? Oct 20 14:42:24 i'm pretty sure he has too much money Oct 20 14:42:34 couldnt get away with that sadly Oct 20 14:44:25 i do like the newer max series getting qzss , need to look into that more Oct 20 14:44:41 i doubt you can get qzss where you are Oct 20 14:45:02 it flies over where i am Oct 20 14:45:08 huh? Oct 20 14:45:16 dont they turn around slightly below the equator? Oct 20 14:45:26 http://upload.wikimedia.org/wikipedia/commons/c/c2/Qzss-45-0.09.jpg Oct 20 14:45:40 oh.. Oct 20 14:45:42 cool! Oct 20 14:46:05 im to the left a little Oct 20 14:46:38 "the left"? east coast? Oct 20 14:46:45 west coast Oct 20 14:46:52 er.. yes Oct 20 14:47:02 i guess that's the sign that i should go home Oct 20 14:47:09 haha :D Oct 20 15:52:41 KotH, I have multiple hobbies, where first is job related and third of them is electronics including doing something with BBB ;) Oct 20 17:37:33 mkad : eh ... using the bbbs own pwm ports for controlling escs is just fine Oct 20 17:37:55 ive just stated that the ardupilot bbb port is a mess ;) Oct 20 17:38:04 not that bbbs own pwm d be unusable or sth Oct 20 17:41:44 (if anybody wants to prove me wrong here ... feel free. i stopped digging into the code when i found out that it wont work for me as they actually try to address avr registers on a bbb...) Oct 20 17:51:56 gurki, how so? I know they have gotten rcinput working through the pru as well as pwm etc through pru. Oct 20 17:52:41 I am pretty much waiting for a stable version of the pixhawk firecape to be available so I can get the board fabbed and try flying my beaglebone black with a logitech c920 attached Oct 20 17:54:17 when did they? Oct 20 17:54:29 the last time i took a look at it is about half a year ago Oct 20 17:54:44 sounds like i should ve another look Oct 20 17:58:18 they have even gotten to the point of actually showing videos of testing Oct 20 17:58:43 https://www.youtube.com/watch?v=YDFamxgsb88 Oct 20 17:58:51 erle copter based on the beaglebone black Oct 20 17:59:22 https://www.youtube.com/watch?v=HD1kXlNbInw Oct 20 17:59:29 good stuff on the way Oct 20 17:59:49 yea ... looks like they put quite a bot of work into this Oct 20 18:00:15 no product = no work Oct 20 18:00:17 ^^ Oct 20 18:00:20 lol Oct 20 18:00:22 https://www.youtube.com/watch?v=RIFAZd42r9k Oct 20 18:00:24 huh? Oct 20 18:00:39 well. concerning the fact that its just been a mess the last time i looked at it Oct 20 18:00:49 and that this looks kinda good Oct 20 18:00:51 ... ;) Oct 20 18:01:04 actually you can do whatever you want, until you dont have results which is working product it's like you did nothing Oct 20 18:01:06 you theoretically should be able to get sensors etc and start using it yourself minus the cape. I have though about grabbing mpu-6050 and maybe the adafruit ultimate gps breakout Oct 20 18:01:39 did u find an eagle layout or sth? Oct 20 18:01:40 Hey guys, I was wondering if someone with more experience could help me figure out if the beagleboard-xm is the right tool for me. I'm quite inexperienced and haven't gotten it yet, but that's because I don't know if it can do what I want it to for the project I have. I know the beagleboard-xm can run at 1GHz, but I can't tell how fast it will be able to grab data from the GPIO pins. Essentially, my situation is the following: Oct 20 18:01:56 I have some custom hardware that sends 6 bits of data over GPIO at a frequency of 50MHz. It sends one packet at a time, with a pakcet beign 144 bits. So that's 24 rounds over GPIO. Would the beagleboard-xm be able to accomodate these sorts of speeds? This information then needs to be passed along to an arduino. Like I said, I'm a bit of a beginner and though I've looked through search engines and forum posts, I can't quite decipher s Oct 20 18:02:20 what the fuck are you trying to do? Oct 20 18:02:22 (sry for language) Oct 20 18:02:24 https://github.com/diydrones/PXF Oct 20 18:03:06 me? There's a touch panel that's connected to custom hardware, which sends the data over at a rate of 50MHz. I need some sort of hardware to be able to catch data at that speed. Oct 20 18:03:16 not sure if the latest bb-xm images run at the rated 1GHz anymore :/ Oct 20 18:03:29 mistawright, why on this movie he is flying back and forth.. I get headage when I watch this, my AVR based copter can fly better ^^ Oct 20 18:03:56 mkad, testing altitude hold Oct 20 18:04:09 Oh, ok Thurgood. Is there a way I can confirm that? Also, is there a way to figure out how fast I'd be able to sample the GPIO pins? Oct 20 18:04:40 mkad, but can your avr copter use opencv or a logitech c920, which handles all the encoding 1080p flight recorder Oct 20 18:05:20 mistawright, of course not, but I can attach gopro to it Oct 20 18:05:47 or some fpv set Oct 20 18:06:57 derpkit: you can force it into 1GHz mode with some uEnv.txt coding to write some values to the dm3730 if it's you really need that extra bit of performance Oct 20 18:07:56 derpkit, I've never tested gpio read speeds personally... so dunno about that Oct 20 18:09:13 the upside of the xM is the extra exposed header pins, the downside is the price of the board Oct 20 18:10:23 I see. Thanks for the information thurgood. Appreciate it! :) Oct 20 18:12:49 50mHz data for atouchpad then spit into a arduino seems like poor planning. Oct 20 18:14:31 I opted for the arduino due since it has HID stuff built into it and I'm kind of a beginner. The problem with the Arduino due was that it wasn't able to handle 50MHz Oct 20 18:14:48 so now I'm trying to figure out how to make sure I don't miss any of the data if it's coming in at 50Mhz Oct 20 18:15:28 can you slow down the deice? 50mHz... prety fast for moving a finger around a pad. Oct 20 18:15:39 s deice/device Oct 20 18:16:12 I'm not in charge of that component unfortunately :(. All I was told was that when the touch event was registered, the data for it would be sent over at 50 MHz so I'm trying to find something that can handle that Oct 20 18:16:24 Unfortunately, I'm struggling a bit given my lack of experience Oct 20 18:16:54 i dunno but fpga comes to mind. Oct 20 18:17:26 Yeah, it seems that way from the research I'm been doing Oct 20 18:17:31 that an fpgs is the best bet Oct 20 18:17:36 *fpga Oct 20 18:18:27 one issue with BB's is that the os is not real time and there might be data misses of it can go a 50mHz Oct 20 18:18:51 data misses if it can go 50mHz Oct 20 18:19:13 Right, that makes sense Oct 20 18:19:46 my fingers are dyslexic today Oct 20 18:20:11 heh, happens to us all Oct 20 18:50:59 data misses at 50mHz? Oct 20 18:51:21 what kind of crazy numbercrunching do you do that even 50mHz is too much? Oct 20 18:52:38 KotH: then you might help him out? Oct 20 18:53:39 djlewis: let me first verify you really mean 50mHz as in fifty mili hertz Oct 20 18:55:35 scroll further up, it's Mega Oct 20 18:55:38 50 milliHertz? srsly? Oct 20 18:55:45 KotH: not me dude. Oct 20 18:56:19 mashiara: ok.. that makes more sense Oct 20 18:57:30 Sending data generated by human interaction at 50MHz does not, but OP cannot change that (for whatever reason) Oct 20 18:58:37 o_0 Oct 20 19:03:01 derpkit: is the protocol clocked separately like I2C/SPI, or is the clock recovered from the data like in UART ? Oct 20 19:03:21 Sorry, I'm back. Unfortunately the spec I was given was 50 MHz as in 50 megahertz Oct 20 19:03:34 data sent over serially Oct 20 19:03:39 uhm Oct 20 19:04:05 Sorry for not being super informed. I'm not quite sure Oct 20 19:04:09 Here's what I know though Oct 20 19:04:29 8 pins are used: 6 for data, 1 for clock and 1 for sync Oct 20 19:04:40 when the sync is asserted, that's when I need to recognise that valid data is coming in Oct 20 19:05:16 then I need to grab data from the pins ever clock cycle thereafter. data available on the falling edge Oct 20 19:05:47 Admittedly I'm not very informed, so I understand it can be frustrating for you trying to help me. I appreciate the patience Oct 20 19:06:02 If there are any questions you think I need to ask my partner to clarify, please let me know :) Oct 20 19:06:43 so 6-bit parallel interface with 50MHz clock (and "Chip-Select" or "Slave-Select" as it's called in SPI) Oct 20 19:07:52 I don't recall seeing any chip select, unless that's the term you're using for sync Oct 20 19:07:56 could GPMC help here? Oct 20 19:07:58 yes Oct 20 19:08:08 that was for derpkit Oct 20 19:08:20 "sync" is "chip-select" Oct 20 19:08:31 if you think in SPI-terms Oct 20 19:09:06 anyway 300Mbit/s for touchpad data ? why ?? Oct 20 19:09:48 To be frank, I'm just an intern student. I can't really give more information on the specifics cause I don't know them :( Oct 20 19:10:13 what are you supposed to do with this data ? Oct 20 19:10:43 All I know is that, the sync is asserted to show that valid data is coming through. Then the data is transmitted at 50 MHz. This data is then used to configure an HID device (like using the arduino due to emulate the mouse) to have mouse events on screen Oct 20 19:11:56 anyone recall the max bandwidth for USB HID devices ? I'm pretty sure it's nowhere near 300Mbit/s Oct 20 19:12:58 Oh, of course. I don't expect mouse events to happen that fast Oct 20 19:13:01 They won't Oct 20 19:13:14 It's just that the data coming off the custom hardware occurs at 50MHz and I can't miss any of it Oct 20 19:13:37 I just need to be able to collect that data at 50Mhz, the processing for mouse events will happen more slowly Oct 20 19:15:58 will you need to emulate USB-mouse ? or just move a cursor on the local display of the BeagleBoard ? Oct 20 19:16:09 I'm hoping to emulate USB-mouse Oct 20 19:16:12 using an arduino Oct 20 19:16:18 (Arduino Due) Oct 20 19:16:27 Or, if there's another option I don't mind using that either Oct 20 19:17:35 If you need to actually emulate the mouse, you will probably have to flash the firmware on the arduino to get it to correctly show up as a mouse. Oct 20 19:17:52 afurbee that's least of derpkits problems ;) Oct 20 19:17:58 The arduino Due has that ability built in. I was playing around with it Oct 20 19:18:09 That's why I'm opting for it :P Oct 20 19:18:40 ah okay, im not that experience with them :) Just offering a tip (apparently useless) Oct 20 19:18:49 the problem is I don't think you'll be able to sample that 50MHz signal correctly on a Due which has 84MHz clock Oct 20 19:19:02 No, not at all! I'm a complete beginner. I wouldn't have thought of something like that if I hadn't known. I appreciate it Oct 20 19:19:23 Yeah. The due can't handle the speed, which is why I need to look for something that does, and then link that to the due Oct 20 19:21:10 a faster MCU will also be able to emulate mouse if it has USB Oct 20 19:21:27 I haven't yet tested how fast the STM32 Nucleo F4 boards can run Oct 20 19:21:33 both those are cheap Oct 20 19:21:42 s/both/but/ Oct 20 19:22:07 of course it's not "Arduino easy" Oct 20 19:22:23 Yeah, I realise. but I think it might be easier for me than fpga haha Oct 20 19:22:45 I think where my difficulty is is that, I see a clock/proc speed, but I can't understand how that would translate to reading the GPIO pins Oct 20 19:23:27 well you set an interrupt to the signal clock input pin going low whenever sync is asserted (and remove that interrupt when sync is released) Oct 20 19:23:37 then in the interrupt-handler you sample the data pins Oct 20 19:24:50 So, I guess, as an example. I thought the Due at 84 MHz would be able to possibly sample the GPIO pins at 50 MHz. However, I learnt that wasn't possible. When I see some fast speed that seems like it will work, is trying the only way to find out if it'll be able to manage sampling at 50MHz? Oct 20 19:25:00 Sorry about the noob question haha Oct 20 19:25:02 but getting to the interrupt-handler will take many clock cycles so you will need to run at much higher speed than 100Mhz Oct 20 19:25:10 I see Oct 20 19:25:45 so the beagleboard-xm, or even raspberry pi have high speeds. But they aren't good choices because they aren't real time systems? Oct 20 19:25:52 yes Oct 20 19:25:59 Thank you :) Oct 20 19:26:05 I guess, I need to do a lot of research Oct 20 19:26:22 It'd be nice to have a really fast microcontroller to work with. I'll admit, I'm afraid to work with fpgas haha Oct 20 19:26:51 STM32F405 at least will run at 148MHz Oct 20 19:27:02 I'm sure there are plenty of options Oct 20 19:28:03 though the GPIO clocks will run at different speed from the core, so the core speed is not the whole truth Oct 20 19:28:37 Yeah, I just discovered that recently haha Oct 20 19:28:52 I'll look through the datasheets and can't find anything about GPIO speeds, so then I feel discouraged Oct 20 19:29:03 I'm not sure how to gauge that aspect if it's not listed anywhere Oct 20 19:29:12 Other than bother people like this :P Oct 20 19:29:26 some people have run benchmarks, google will help Oct 20 19:30:25 Yeah. I suppose I just need to persevere more Oct 20 19:31:26 at 50MHz you'll need to think about the wiring for the signals pretty thoroughly as well Oct 20 19:32:17 parasitic inductance and capacitance will cause issues, don't even think about using a solderless breadboard Oct 20 19:33:46 Thanks for the tips! None of this would have crossed my radar Oct 20 19:34:17 how many bits is a packet ? Oct 20 19:34:53 the packet overall is going to be 144 bits Oct 20 19:35:23 how many packets a second ? Oct 20 19:36:19 I'm not sure. It's bursty data. It depends on when someone touches the board. One data packet of 144 should contain all the information, from what I gather Oct 20 19:37:55 I don't think it will be 2M packets per second that the data rate would suggest... Oct 20 19:38:57 I'll be having a talk with the others I gain a bit more understanding about the whole thing. Cause as of now, I'm not quite sure how everything works myself ¯\_(ツ)_/¯ Oct 20 19:39:09 And apparently my googling sucks as well Oct 20 19:40:31 the clock speed is just insane, even at 1kHz update rate from the touchpad (which ought to be way,way faster than touchpad needs to update) the clock would need to be 24kHz Oct 20 19:40:48 which is easily in the realm of just about any microcontroller Oct 20 19:41:09 and definitely easy for anything with HID capable USB Oct 20 19:42:51 so either there is some huge misunderstanding or the interface is meant to force you to solve this in FPGA... Oct 20 19:43:13 It might. I think they're open to that, but suggested I look at microcontrollers first Oct 20 19:43:25 There might be a misunderstanding though! Oct 20 19:43:32 So I will definitely be touching base with them at some point Oct 20 19:44:04 Anyway, I just wanted to thank you for your assistance. Thanks for being so patient and explaining to this ultra noob Oct 20 20:15:19 Anyone have input for best linux distro to run for chrome? Oct 20 20:26:36 pretty subjective question Oct 20 22:10:39 can anyone give me a little help with building angstrom from setup-scripts? Oct 20 22:11:31 i dont know which files, in which order, to copy to mmc Oct 20 22:12:53 I forgot my login username and password to my beagleboard-xm, is it possible to reset it without need of rebuilding the kernel? Oct 20 22:13:00 so i have a quesiton Oct 20 22:13:18 I downloaded the latest ubuntu 14.04 image for a bbb and I don't have a /sys/devices/*cape* Oct 20 22:13:25 I don't seem to have any cape manager AT ALL Oct 20 22:13:36 Should I? Oct 20 22:18:00 CanyonMan: i dont know anymore specifics, but have alook at https://www.mail-archive.com/beagleboard@googlegroups.com/msg16681.html Oct 20 22:18:16 ok thanks Oct 20 22:18:20 note that it was for 12.04 Oct 20 22:18:32 i sorta wonder if maye I've got the wrong kernel. It's kernel 3.14.something so much later than 3.8.something Oct 20 22:19:03 I didn't see any kernel .config options to turn on/off the cape manager Oct 20 22:20:44 huh so cape manager isn't in the kernel proper it's added by somebody? Oct 20 22:21:14 right, seems to be so Oct 20 22:21:51 i think i'm too lazy to build a kernel and I will find one to steal Oct 20 22:22:31 oh perhaps there's a 3.15 version Oct 20 22:23:13 i guess 3.15 is dev-only, 3.16 is suposed to be the next stable release Oct 20 22:24:46 not completly sure, though Oct 20 22:26:15 i'm so lazy in fact that i didn't set up a cross compiler Oct 20 22:26:19 i'm compiling the kernel on my bbb Oct 20 22:26:20 haha Oct 20 22:26:26 pretty bad eh Oct 20 22:27:14 can anyone point me the difference between u-boot.img and u-boot.bin? I am getting .img file from yocto build Oct 20 22:27:42 hahahaha I am about to do that, this build thingy is so anoying! Oct 20 22:28:03 it's comically slow compiling a kernel on the bone Oct 20 22:28:06 but it seems somehow proper Oct 20 22:28:08 circle of life etc Oct 20 22:29:20 way better then going to sleep, wake up hoping for a build a minor errror showing up on your screen Oct 20 22:30:03 ok here we go Oct 20 23:23:26 i'm having trouble with a beaglebone black running debian and a USB device. It's a USB RFID reader - the BBB doesn't do a good job of powering it. If I leave it plugged in when I power up the BBB, it beeps and is receiving power, but shortly after it loses power Oct 20 23:23:42 connecting it doesn't make anything show up in dmesg - i'm having trouble finding useful resources Oct 20 23:24:16 the reader works fine on my laptop **** ENDING LOGGING AT Tue Oct 21 02:59:59 2014