**** BEGIN LOGGING AT Wed Jan 04 03:00:02 2017 Jan 04 03:32:01 ericxdu: e = force-enable Jan 04 03:32:06 ericxdu: https://www.kernel.org/doc/Documentation/fb/modedb.txt Jan 04 04:17:25 Thx Jan 04 04:19:58 they can really make crazy shit nowadays Jan 04 04:20:33 apparently the pressure sensor on the BBE uses "a mono-silicon membrane above an air cavity with controlled gap and defined pressure" Jan 04 04:21:32 Yeeep crazy Jan 04 04:21:35 Very sci-fi Jan 04 04:28:26 zmatt: I'm trying "D" nstead of "e" not sure if it'll make a difference Jan 04 04:28:36 Kinda thought the screen looked fuzzy Jan 04 04:28:38 Before Jan 04 04:29:36 it won't make a difference since there's no analog output anyway. you don't happen to know the native resolution of the monitor? Jan 04 04:30:20 getting 1920x1080 to work can be tricky, sometimes you get very suboptimal choices by default Jan 04 04:31:18 e.g. a problem I had with some monitor was that it would use something x 1024, and instead of putting small black borders at the top/bottom of the screen the monitor decided it would be a good idea to rescale 1024 lines to 1080 lines Jan 04 04:31:19 zmatt: documentation I found said it's 720p and can do 1080p by scaling, so it sounds like its 720p native Jan 04 04:31:30 1024x768 and 1440x900 also work Jan 04 04:32:01 native resolution is easy to determine by using some evil test pattern with vertical and horizontal lines spaced 1 pixel apart Jan 04 04:32:35 :-o Jan 04 04:32:41 Sounds cool! Jan 04 04:33:43 for native 1080p monitors a useful resolution can be 1440x1080, which the bbb can sustain more easily at higher refresh rates than actual 1080p Jan 04 04:35:01 iirc 1920x1080R@50 (note the R) is also within valid range, but not comfortably Jan 04 04:35:24 but if the monitor is native 720p then it obviously *should* be sharpest if provided with 720p Jan 04 04:37:38 I'm not sure how to tell if it's "not sharp" or if my eyes just don't like it Jan 04 04:37:50 Could just be lack of antialiasing in some areas Jan 04 04:37:52 “Native 720p” monitors can sometimes instead be 1360 or 1366 x 768. Jan 04 04:37:57 So be careful with that. Jan 04 04:38:32 ericxdu: grab a magnifying glass to be able to see individual pixels clearly? Jan 04 04:38:33 Tenacious-Techhu: There is some confusing documentation that mentoins 1366x768 for this, but it didn't work when i entered it in Jan 04 04:38:45 You really need to find out what the panel itself actually is, regardless of what the monitor itself likes as input. Jan 04 04:38:45 zmatt: I shall try Jan 04 04:39:23 Is there anything in the Beaglebone documentation about ignoring edids? Jan 04 04:40:04 Hmm. I'll have to borrow a magnifier I don't recall if I have one :-P Jan 04 04:40:12 the kernel lets you provide an override-edid if necessary Jan 04 04:40:29 it also has a few built in for common resolutions Jan 04 04:40:45 Ignoring the reported edid is often a necessary first step in getting a Raspberry Pi to drive a VGA monitor at a native resolution. Jan 04 04:40:51 This sounds like a similar problem. Jan 04 04:41:03 dunno, normally just providing an override resolution suffices Jan 04 04:41:28 It sounds like you need to supply an override edid for this monitor so it will be driven at the 1366x768 resolution. Jan 04 04:42:34 Macs can have a similar problem when driving obscure displays; there’s a whole Hackintosh thread about it. Jan 04 04:42:37 I thought I was overriding EDID already by setting in uEnv.txt Jan 04 04:43:30 ericxdu: you are, but you're not *providing* an edid Jan 04 04:43:40 I see Jan 04 04:43:40 you're just telling it to disregard the edid Jan 04 04:43:53 Telling it what resolution to use won’t do any good if the monitor’s EDID has a hissy-fit and says you should use the resolution it wants you to. Jan 04 04:45:00 So xrandr will not actually *know* what resolution is actually displaying? Jan 04 04:45:29 of course it knows Jan 04 04:45:45 or rather, it knows which resolution is being sent to the monitor Jan 04 04:45:53 It’s sort of like trying to tell a dragon to roll over without bitch-slapping it with it’s true name first. Jan 04 04:45:54 Yeah Jan 04 04:46:02 what it will do with that stream of data is up to the monitor of course Jan 04 04:46:15 Tenacious-Techhu implies the monitor may ignore that? Jan 04 04:46:25 That's how I understood it Jan 04 04:46:45 it can do anything it wants with the data ultimately Jan 04 04:47:32 If you just try to tell it a resolution, it may ignore you; you have to associate that resolution with the EDID of the display somehow, or instead use a different EDID with that resolution that is compatible with your monitor. Jan 04 04:47:34 but providing an override-edid isn't particularly useful here, it just provides different info to the kernel Jan 04 04:47:50 Tenacious-Techhu: ehh Jan 04 04:47:58 Tenacious-Techhu: what you just said is completely incoherent Jan 04 04:48:08 No, it isn’t. Jan 04 04:48:20 edid is just a data structure the kernel reads from the monitor to know what resolutions it claims to support Jan 04 04:48:56 Hmm http://www.imatest.com/docs/screenpatterns/ Jan 04 04:49:58 zmatt, yes, it is, but it is also related to the resolutions that the monitor will accept. Jan 04 04:50:27 Monitors may support resolutions not reported in the EDID for whatever reason. Jan 04 04:50:29 in an ideal world it would truthfully describe the abilities of the monitor Jan 04 04:50:33 yes I know Jan 04 04:50:43 Sometimes, those resolutions are the actual native resolutions a person may want to use. Jan 04 04:50:55 http://www.lagom.nl/lcd-test/display_settings.php Jan 04 04:50:59 yes I know, been there done that Jan 04 04:51:42 Getting a native resolution to work in spite of the EDID can sometimes be more complicated than just trying to drive at that very resolution. Jan 04 04:52:04 I'm fairly certain that tilcdc ignores the edid entirely if you provide an override resolution Jan 04 04:53:09 in fact it will happily configure and enable the output even if no monitor has been connected yet Jan 04 04:53:15 (hence it cannot possibly have read the edid) Jan 04 04:55:26 Yes, but just because the Beaglebone may choose to ignore the EDID, that doesn’t mean the monitor will allow it if you don’t ask it in the right way. Jan 04 04:55:38 I have not tried 1360x768 yet, I wonder if I should Jan 04 04:55:45 You should. Jan 04 04:56:10 hi Jan 04 04:56:17 ericxdu: the test images for sharpness on that website are a nice test (vertical and horizontal lines) Jan 04 04:56:45 You might also care to try doing it with a Raspberry Pi, if only to see whether the monitor will accept it or not. Jan 04 04:57:02 i am trying to connect to the beaglebone black wireless using wifi Jan 04 04:57:14 Tenacious-Techhu: I've tried 1366 though haha Jan 04 04:57:21 I’m more confident in the Raspberry Pi’s ability to handle this sort of problem correctly than the Beaglebone Black’s. Jan 04 04:57:31 Tenacious-Techhu: my Pi is a lot better at working with this monitor. I should definitely check Jan 04 04:57:35 ericxdu, it won’t hurt to try, so you should do it anyway. Jan 04 04:57:53 Yeah, try the Pi at the native resolution, and see if it takes it. Jan 04 04:58:01 hello i cannot connect to the beaglebone wifi it is asking for a password Jan 04 04:58:03 zmatt: I'll check the sharpness image a few times Jan 04 04:58:05 If it does, then you know the problem is with the Beaglebone video driver somewhere. Jan 04 04:58:36 olivia, sorry, we hear you, but I don’t know anything about that; how about you, zmatt? Jan 04 04:58:41 to be honest it's a bit tiring if monitor manufacturers can't be bothered to fill in the edid structure correctly... how much fucking effort is it.. -.- Jan 04 04:58:53 Inorite? Jan 04 04:59:16 olivia: I don't have experience with the bbbw but I'd assume the default is documented somewhere Jan 04 05:00:13 at least on my BBE it was something fairly trivial, but I'm afraid I don't remember what and no longer have those config files Jan 04 05:00:33 its asking for the 8 digit pin on the router label Jan 04 05:00:53 uhh Jan 04 05:01:06 but that pin is not working Jan 04 05:01:29 oh you're trying to connect it to a wireless network? Jan 04 05:01:40 yes Jan 04 05:01:42 that's not the same as connecting to the beaglebone's wifi Jan 04 05:01:54 (it is in access point mode by default) Jan 04 05:02:17 i am trying to connect to the beaglebone AP Jan 04 05:02:26 from my laptop Jan 04 05:03:02 then what "router label" are you referring to? Jan 04 05:05:46 ok there's really an appalling lack of documentation afaict :/ Jan 04 05:06:32 jkridner: where's any sort of documentation on how the wireless stuff is supposed to work on bbbw (or iot images in general) Jan 04 05:10:21 ok bye olivia :P Jan 04 05:11:34 and fuck IIO too, I'm really getting sick of big hairy balls of complication where any info you find is either useles or no longer correct Jan 04 05:12:06 I'm done for tonight, afk Jan 04 05:14:38 IIO? Jan 04 05:14:45 Hehe Jan 04 05:14:58 Documentation is kinda sad Jan 04 05:15:06 For the Beagle line Jan 04 05:15:16 Anyway 1360 works! Jan 04 05:15:51 I'm gonna save this uEnv.txt somewhere safe... I have all the resolutoins that worked written and commented out Jan 04 05:16:33 Interesting 1360x768 looks all-around sharper than 720p... Jan 04 05:16:45 but still not quite sure. Some letters look bolder than others Jan 04 05:16:58 my laptop is 1366x768 Jan 04 05:17:05 zmatt: mine too Jan 04 05:17:11 I like that res Jan 04 05:18:06 yes, 2 * a prime number, lovely Jan 04 05:19:15 what the heck is a BBE ? Jan 04 05:19:18 guaranteed to be insufficiently aligned for most purposes, which also means the line stride for framebuffers on my laptop is not width * (bytes/pixel) but larger Jan 04 05:19:24 bealgebone extended or something ? Jan 04 05:19:27 enhanced Jan 04 05:19:38 never heard of it until now Jan 04 05:19:58 ericxdu, 1360 works but not 1366? Jan 04 05:20:10 sadly their indiegogo wasn't hugely successful, but they delivered as promised Jan 04 05:20:24 Tenacious-Techhu: yes Jan 04 05:20:36 Interesting. Jan 04 05:20:38 zmatt: Oh this is some sort of kickstarter ? Jan 04 05:21:00 https://www.indiegogo.com/projects/sancloud-beaglebone-enhanced Jan 04 05:21:11 Tenacious-Techhu: Is that strange? I figured there was just some disagreement between manufacturers Jan 04 05:21:18 By a few piixels Jan 04 05:21:41 About which was offically 16:9 Jan 04 05:21:55 ericxdu: some people just love a 683:384 aspect ratio Jan 04 05:22:07 The answer is clearly 1366 Jan 04 05:22:30 zmatt: :-o Jan 04 05:23:34 Honestly? Not sure. Could just be SNAFU... Jan 04 05:23:53 actually 1366 is just 768*16/9 rounded to the nearest even integer (since an odd number of pixels was probably too horrid to seriously consider, and/or a hw technical problem) Jan 04 05:24:54 I think it has more to do with rounding to the nearest RAM chip; do the math. Jan 04 05:25:28 1366 isn't rounded to anything nice whatsoever, as I mentioned above it's 2 * a prime number Jan 04 05:25:46 1360 will have arisen from a need for better alignment Jan 04 05:26:04 since it's a multiple of 16 pixels Jan 04 05:27:00 Oh? Jan 04 05:27:12 1360x768/1024=1020. Jan 04 05:27:30 Ugh. I think it's still wrong Jan 04 05:27:40 Tenacious-Techhu: uhh... and? Jan 04 05:27:42 And I was hoping I'd finally settled on 720p Jan 04 05:27:45 -_- Jan 04 05:27:46 1366x768/1024=1024.5. Jan 04 05:27:57 I'll get this figured out. Maybe it's just 1024x768 Jan 04 05:28:35 Tenacious-Techhu: congratulations, you know how to multiply and divide! (or at least enter some random numbers in a calculator and copy/paste the answer) Jan 04 05:30:07 I suspect it’s actually 1365, but the marketing on 1366 is better. Jan 04 05:30:33 As I said, it’s rounded to the nearest RAM chip. Jan 04 05:30:57 All displays at all resolutions are like that. Jan 04 05:32:36 Oh right! I was going to see what my Raspberry Pi had to say about all this Jan 04 05:33:13 except neither 1366 nor 1360 is. 1366 exceeds it (as you just showed) and 1360 is lower than necessary. it may very well have been a factor, but clearly not a decisive one since there's plenty of 1366x768 out there Jan 04 05:34:43 anyway, I'm off Jan 04 05:35:09 Yes, but do you really think they’re putting another RAM chip in there for that extra column of pixels? If you do, you’re kidding yourself. Jan 04 05:36:41 Take the aspect ratio, and apply that to a given convenient RAM size, and you get a common display resolution; it’s that simple. Jan 04 05:37:24 I don't think my laptop even has dedicated VRAM ? if it does it definitely won't be exactly the size of one framebuffer Jan 04 05:38:05 argh, why do I get dragged into this idiotic discussion again Jan 04 05:38:18 I just explained that your theory does not match the facts Jan 04 05:38:28 why do you continue sprouting nonsense? Jan 04 05:38:34 It’s not going to be exact; it’s going to be rounded to the nearest chip. Jan 04 05:38:57 Why do you continue to live in denial? Jan 04 05:39:15 except there's nowhere in a laptop (which is where this resolution typically occurs) any ram that's sized a framebfufer anyhow Jan 04 05:39:42 some monitors *might* buffer a whole frame but even that seems doubtful Jan 04 05:40:15 normally you buffer a line and then send that to the column drivers Jan 04 05:40:25 a vertical scaler only needs to buffer a few lines Jan 04 05:40:50 Maybe your laptop uses system RAM? Jan 04 05:40:50 the graphics card otoh typically has waaaay more ram than one framebuffer, and needs to support multiple resolutions anyhow Jan 04 05:41:44 If it’s supporting higher resolutions than the display, work from the max res. Jan 04 05:41:56 * zmatt sighs Jan 04 05:44:32 It’s all about reducing the parts cost to the minimum necessary required. Jan 04 05:45:16 zmatt: ah right I take that back. I have heard of that board . . . dont think I'd have in interrest in that one Jan 04 05:47:25 Don't get me wrong I thnk the BBB, or BBG for that matter are both great boards. But when we start to talk about more RAM, and GbE . . . theres other SBCs that make better sense IMHO Jan 04 05:51:10 I agree, really. I’d rather see more cores on the CPU, more inputs, better community and library support, and maybe simpler PRU-type subunits. Jan 04 06:31:13 The PRU's though, by deiniftion are not meant to be "simple" at least not in use. Jan 04 06:31:22 I'm ok with the community Jan 04 06:35:09 PRUs are trivial to use Jan 04 06:35:18 the SoC is the complex part Jan 04 07:01:05 ds2 I tend to agree. But an Cortex M3 is easy to use by it's self too. When not tethered to an application based processor through an interconnect ;) Jan 04 07:01:41 actually, I've not used a PRU yet, but have read somewhat extensively about it Jan 04 07:02:15 well thats mostly correct. I've setup, and have used the sample apps from Jason github repo Jan 04 07:02:24 Jason's* Jan 04 07:25:03 The Pi likes 720p Jan 04 07:25:58 cortex-m3's are more complex Jan 04 07:26:30 i.e. a cortex-m3 doesn't have IO abilities w/o the associated SoC Jan 04 07:26:37 a PRU does. Jan 04 15:17:19 Hey guys Jan 04 15:18:10 I'm playing with the idea to make a system that runs qt gui and at the same time has sensors interface Jan 04 15:18:33 Is the beaglebone up to that task? Jan 04 15:18:39 Or will i be straining it? Jan 04 19:50:03 JP___: that's a fairly common setup Jan 04 20:19:35 hy folks anyone here good with beagle bone black i seem to be missing my cape manager folder in sys/devices/ Jan 04 20:20:41 for information am quite junior and following through derek molloys book for experience Jan 04 20:21:02 post your kernel log 'dmesg | pastebinit' Jan 04 20:21:43 not sure how to use that? Jan 04 20:22:03 is your beaglebone on the Internet? Jan 04 20:22:05 recoginze the name from the text lol. in awe Jan 04 20:22:07 yes Jan 04 20:22:07 can you log in the shell? Jan 04 20:22:14 yes i ssh into it Jan 04 20:22:40 running debian in oracle vm Jan 04 20:22:41 * jkridner feels unworthy of awe. Jan 04 20:22:58 you r more than worthy in presence of greatness Jan 04 20:25:26 ivorm: looks like it moved to /sys/devices/platform when capemgr got merged into the mainline kernel Jan 04 20:25:41 but the docs all still talk about the old location Jan 04 20:27:33 im trying to copy and paste message with no luck Jan 04 20:29:48 ive tried editing my .profile as on pg 223 of text then try to cat $SLOTS nothing Jan 04 20:31:56 ok ive cd' ed into /devices/platform and found capemgr. is this where it should be or should i mv it? Jan 04 20:32:23 try stuff out, young padewan Jan 04 20:34:02 bk to the drawing board all this to light and LED and press a button.. ive been doing this for a week Jan 04 20:34:52 I've been doing this shit for 15 years Jan 04 20:34:58 and it still doesn't work Jan 04 20:35:02 it never works :-/ Jan 04 20:35:06 haha Jan 04 20:35:08 except when it works, and then it's great :-) Jan 04 20:35:13 ^^ bitter old padewan Jan 04 20:35:28 should i do export SLOTS=/sys/devices/platform/bone_capemgr.9/slots?? Jan 04 20:35:33 biot: careful who you call padewan Jan 04 20:36:06 i dont even know what padewan means t be honest no offence taken im learning.. Jan 04 20:36:14 jkridner: this cape-universaln that's getting loaded by default in the newer debian images is real fancy schmancy Jan 04 20:36:19 ivorm: it's a Star Wars reference Jan 04 20:36:37 but it's telling me I got /dev/spidev* when their pins are in fact in gpio mode Jan 04 20:36:46 treachery! Jan 04 20:36:53 biot: you can use config-pin script to set them. Jan 04 20:36:59 yeah, I did, works fine now Jan 04 20:37:07 :-) Jan 04 20:37:44 ivorm: yes, it moved in newer kernels. Jan 04 20:37:55 ivorm: wish there was a way to update the book. :-) Jan 04 20:39:01 i know and the problem with internet hv to be carefull what blogs i read. it took me a month to configure bb on ethernet Jan 04 20:39:34 everyone has there own way of doing things but not everyone has written a book. Jan 04 20:40:11 im using your image on my bb Jkrinder had to update Jan 04 20:46:33 another question on the same not should i edit my .profile file with the environment variables or the uEnv.txt file? Jan 04 20:50:56 tell you what ill try it alternately doing both. Whats the worst that can happen.. Jan 04 21:37:52 trying to load DTO and getting error message /sys/kernel/debug/pinctrl/44e10800.pinmux/pins: not a valid identifier Jan 04 21:38:09 any ideas? Jan 04 22:13:20 ello Jan 04 22:18:10 jkridner, do you have an ETA on the Beaglebone Blue? Jan 04 22:27:14 jkridner: is there any docs online on how the bbb-wireless is supposed to work out of the box? I've noticed quite a few people with wireless trouble, but haven't found any remotely useful info Jan 04 22:39:46 bbl Jan 04 22:44:39 jkridner: btw: the " this IRC survival guide" link on https://beagleboard.org/chat is 404 Jan 04 22:44:51 jkridner: or rather, dead server Jan 04 23:46:44 jkridner, since I feel like your my inside source. when are expecting that beaglebone blue robotics kit. You need a beta tester? Jan 04 23:47:02 when are you Jan 05 00:29:29 KotH: I found the new location a while ago... there were good tips in that guide, but sometimes presented a bit silly and the thing is inevitably "tl;dr" (like the smart questions faq) Jan 05 00:29:45 I'd rather see a few key points embedded directly on https://beagleboard.org/chat Jan 05 00:32:47 i.e. just ask your question in a clear and self-contained way, and BE PATIENT since people are on all kinds of different schedules and may only peek in now and then Jan 05 00:33:10 what i pretty much do now Jan 05 00:33:14 don't ask if anyone is here, don't ask "can anyone help me?" or similar Jan 05 00:34:25 way too often I see people just saying hello and then leave again, or asking a question but disappearing before it can be answered Jan 05 00:34:49 Because chats are supposed to be chats; not forum posts. Jan 05 01:49:39 interesting idea there Jan 05 02:06:28 Oh yeah, real interesting, things living up to what they are supposed to be. Far too rare these days. **** ENDING LOGGING AT Thu Jan 05 03:00:00 2017