#qi-hardware IRC log for Saturday, 2011-03-26

tuxbrainyep confirmed also happens with Nanomaps, if launched by ssh as command "ssh root@nanonote /usr/bin/Nanomaps" on exit it hangs00:04
kristianpaulxiangfu: are you there?00:04
kristianpaul(al que madruga dios lo ayuda)00:05
kristianpauli wonder how to said that in english00:05
Action: kristianpaul looks at tuxbrain 00:05
tuxbrainwho weakup early in the morning good help him00:05
tuxbraingood ->God00:06
kristianpaulah, ok00:06
tuxbrainliteral translation but surelly is a proper said on english00:06
kristianpauli was about to said that ;-)00:07
wpwraktuxbrain: the early bird catches the worm ? http://www.goenglish.com/TheEarlyBirdCatchesTheWorm.asp00:10
kristianpaulwpwrak: i like that :-)00:10
kristianpaulwpwrak: are you aware id gnuradio companion can do DSP by using blocks like in LabView?00:11
kristianpauls/id/if00:12
wpwrakkristianpaul: guis are for girls ;-)00:17
wpwrakkristianpaul: but yes, i've played a bit with it. it may be useful for documentation purposes. for "real work", i'd rather use C00:18
kristianpaul:-)00:19
kristianpaul(guis are for girls ) yes :-)00:19
kristianpaulokay so let me ask you, if have to make a oscilator in software.. that later will be mixed (multiplied) by the digital representation of the incoming signal00:21
kristianpaulwhat you take care about the oscilator, i dont get yet,or i'm bit lost on that part, mostly i think about how to "sync time" between both signals..00:23
kristianpaulalso i dunno if my question is clear :/00:23
kristianpauli think question will be how to implement a NCO in C00:26
wpwrakwhile (1) { y = amplitude*sin(t*omega); t += step; }   ?00:27
wpwrakor perhaps: while (1) { y = amplitude*sin(t*omega+phase); t += step; }   ?00:29
wpwraknow you can tweak "phase" to match the incoming signal00:29
kristianpaulhmm00:29
kristianpaullet me read slowly and carefully :-)00:30
wpwrak(that's for a real-valued signal, not I/Q)00:30
kristianpauli was to said that :)!!00:30
kristianpaulabout*00:31
Action: wpwrak grabs the DSP book00:31
kristianpauland tell me the name please ;-)00:32
Action: kristianpaul curious about wpwrak personal library 00:32
wpwrak"digital signal processing" by steven w. smith00:34
wpwrakactually ...00:34
wpwrakthis is what you want to read: http://www.dspguide.com/ch30.htm (and maybe the chapters following it)00:35
wpwrakch30 is just the basic math. doesn't have much meat to it.00:35
wpwrakalas, the book doesn't go into the more advanced math. it's more a hands-on book for lower-frequency problems.00:36
kristianpaulk00:38
tuxbrainwell the "grey screen" killed my pretension of making popup the app from the Qt-Creator with one click...., is not fair to make a video doing so if then when you exit the app NN stays in an unusable state..., so no video :(. but will start to write a tutorial on how to setup Qt creator....00:51
wpwraktuxbrain: just don't show application exit on the video ?01:05
wpwraks/on/in/01:06
tuxbrainwpwrak: that is what is not fair, this will give a false impresion that is working and someone that follow the instructions will end up with a blocked NN ... no thank you01:11
tuxbrainmmm I have an Idea for a workarround... what if there is a script in NN that launch the app?... I will try..01:13
kristianpaultuxbrain: did you looked at gmenu2x launch script for nanomap?01:13
kristianpauli remenber i had same quit problem time ago01:14
kristianpauls/i/it01:14
kristianpaulokay i need a opinion (again):01:15
tuxbrainkristianpaul: I will take a look01:15
tuxbrainkristianpaul: it launch Nanomap directly01:17
wpwraktuxbrain: there are lies, damn lies, and then there's marketing ;-)01:19
tuxbrainwpwrak: I was working serveral years in a mkt dpt.  I'm surelly will go to hell for that, but I think in qi we are trying making things a little bit different :)01:20
wpwraktuxbrain: :)01:24
tuxbraindammit.. it doesn't work01:28
tuxbrain:(01:28
tuxbraineven with ./nanokoder -qws >/dev/null 2>&1 &01:29
tuxbrainseems like if it doen't have a console to return the processes will not end...01:35
kyaktuxbrain: it's a known bug of qt apps with uClibc 0.9.30.106:38
tuxbrainkyak:(qt bug) any hope to be solved soon?09:43
kyaktuxbrain: i doubt it will be fixed in that uClibc version. The trunk is using uClibc 0.9.32 now, and qt should work there10:11
rohhmmm.10:36
rohplayed around (mechanically) a bit with atben10:36
rohcould imagine some rubber block with a slot in it.. but that all doesnt give any mechanical support10:36
rohi am tempted to just smear something like sugru on it10:49
rohtuxbrain: what do you think about that? sell the atben with a pack of sugru, color of choice10:49
kyaklars_: do you have an idea why scrollback is not working in 2.6.37 framebuffer?12:23
wpwrakroh: is sugru removable ?13:22
wpwrakroh: the idea of selling it as something a little diy-ish has a nice appeal to it13:24
rohi dunno.. depends13:27
rohi think its adhesive somehow there is a list what it doesnt stick on13:28
wpwrakjust reading about its properties. hmm, sounds at it it wouldn't come off without a fight.13:29
wpwrakroh: have you worked with it yet ?13:30
rohi have seen it. not tried myself13:30
wpwrak"if swallowed, seek medical advice and show this pack" oh yes, i can already imagine the faces of the doctors ;-)13:34
rohwell.. standard text for something chemical.. check your washing powder13:34
wpwrakthere's one difference: they've heard of washing powder before. for sugru, they'll have to refer you to the ufology department :)13:35
rohhttp://blog.makezine.com/archive/2010/11/reader-research-diy-casting-silicon.html13:37
rohthere is also the diy way... if i can ever find out wtf these units are (teaspoons?!) .. ounces? (which ones?)13:37
rohand what they exactly mean with chaulk (different types, right one to be 'sniffed out')13:38
wpwrakthe exact formula would depend on your paint thinner anyway. "paint thinner" is about as specific as "dinner"13:41
wpwrakand yes, the same applies to caulk13:42
wpwrakbut i'm happy that someone else tried the paint thinner + off-the-shelf sealant approach. i experimented a little with it (for dipping), and the results were fairly nice13:43
wpwraki'm not so sure about the starch, though. might attract fungi and such13:44
rohone could always paint it with something water repellant.. acryllic paint?13:45
rohon the other side.. i wouldnt want to sell self mixed chemical stuff. in the end i would have no clue how stable it stays13:48
wpwrakif you can paint it. at least regular caulk is not paintable.13:48
wpwrakwell, silicone caulk. if you get polyurethane, a lot of things change. alas, also the price.13:49
rohwell.. you can mix it in. and you can grind it, so painting should be possible13:49
wpwraka problem with self-mixing may also be permitted ingredients. at least some of the paint thinners they sell around here contain quite nasty stuff13:53
rohhehe.. isnt it just acetone or so? means something which qualifies as 'loesungsmittel13:54
rohsomething that evaporates out13:55
wpwrakthey're not very specific. here's a relatively weak one: "mezcla de hidrocarburos alifaticos (principalmente parafinicos y oeofinicos) y aromaticoscon contenido menor de oxigenados"13:57
wpwraka more aggressive one just says "mezcla de hidrocarburator aromaticos, alifaticos, y oxygenados" and they mention that there's no benzene or chlorated (?) solvents. at least something ;-)14:01
rohhm. hydrocarbons. well... duh.14:04
wpwrakthere's also a more evil brew, but i don't have that one around at the moment. that one also has MEK.14:05
rohi think acetone would be the right one14:06
wpwraki've used acetone. i think it didn't produce as smooth a solution as one of my paint thinners, but the result looked quite okay.14:07
wpwrakone problem with acetone is that it's used for production of illegal drugs, so it's not trivial to obtain in quantity14:08
wpwrakdifficulty in argentina ranges from having to show an ID document when buying a liter bottle (not sure if there's a limit for how much they'd sell you) to the shop requiring a written declaration on the intended use14:09
rohheh14:09
rohhttp://www.pollin.de/shop/dt/ODY3OTg0OTk-/Werkstatt/Werkstattbedarf/Spray_Schmierstoffe_Reiniger/Aceton.html14:10
roh4E/l .. mail order14:10
wpwrakthe latter is a place where you get ~30% peroxide without trouble. mix that with HCl and etch your PCBs in mere seconds :)14:10
rohbut one woudlnt use such bad stuff for drugs. thats out of style14:11
wpwrak(unfortunately, with very uneven performance. also, the reaction is a bit violent)14:11
jlamotheI've been thinking about doing that myself.  I'm told that ferric-chloride works well.14:11
jlamothe...but I don't know much on the subject.14:11
jlamothe...and I'm not sure what to do with it once it's spent.14:12
wpwrakjlamothe: one you've tried HCl+H2O2, you'll never go back to ferric chloride ;-)14:12
rohbtw http://video.google.com/videoplay?docid=8639555925486210852#14:12
wpwrakjlamothe: just use the "weak" peroxide they sell at pharmacies. not the industry-grade stuff.14:12
rohvery interresting series from bbc (6 parts)14:12
rohseemingly applies to hw and sw development as well.14:12
jlamotheYou mix it with the HCl?14:13
wpwrakwell, unless having some supercharged acid boil in your kitchen is your idea of frun ;-)14:13
jlamotheThat's what the lab is for.  :P14:13
jlamotheAh, it's 1 part HCl to 2 parts H2O2 (based on what Google tells me).14:16
wpwrakjlamothe: about one part HCl (~30%), two parts peroxide (~5-10% i think). it's a clear solution that stays clear. it loses its potency after a few hours, though. but you can restore it by adding more peroxide.14:16
wpwrakjlamothe: ah, you found it ;-)14:17
jlamotheAlso it seems that I should add the acid to the peroxide, not the other way around.14:18
jlamotheNot entirely sure why, but they're very insistent on this point.14:18
kristianpaulmorning :-)14:19
jlamotheMorning.14:19
jlamotheCan I still use a toner transfer method with this mixture?14:20
wpwrakjlamothe: (toner transfer) yes14:21
jlamothewpwrak: Great.  Thanks.  :)14:22
jlamotheI generally use a ground plane, to minimize the amount of copper I need to etch, so it should last me for a while.14:23
wpwrakjlamothe: (acid into peroxide) the idea is that the concentrated acid gets diluted by pouring it into the diluted peroxide. also, if you have spills, they will be mainly peroxide, not acid. the peroxide is harmless, the acid isn't.14:23
jlamotheMakes sense.14:23
wpwrakjlamothe: i just use very small quantities of acid. afeter use, i let the water evaporate and dispose of it as solid garbage. the crystals are easy to remove.14:24
jlamotheThe first boards I did, I just had a third party make them for me.  Too many tiny traces and vias.  Not exactly a good starter project.14:25
jlamotheI've got another one that's much simpler though.14:25
wpwrakfor me, toner transfer works well for 8 mil traces/clearance. the main issue is how the toner's shape changes during transfer. if you have too much heat, the substrate may expand and produce a larger than desired toner footprint.14:34
wpwrakyou can fix this by manually removing the excess, though. if you can pick larger traces/clearance, even better.14:35
jlamotheI'v actually got at least 10mil traces and clearance everywhere.14:36
jlamotheIt's more an issue of drilling out the many, many 20mil dia vias.14:37
wpwrak20 mil holes or rings ?14:37
jlamotheholes14:37
wpwrakoh, 20 mil holes are easy :)14:37
jlamotheFor a first attempt?14:38
wpwrakmy holes are nominally 10 mil, the rings 30 mil, the drill i use 13.5 mil14:38
wpwrakyeah, 20 mil is still dremel-friendly :)14:38
jlamotheI've actually got a drill press at my disposal... although, the bits tend to migrate. :/14:39
wpwraki've switched to using my CNC mill for drilling, though. more nicer to watch the machine do it ;-)14:39
wpwraks/more/much/14:40
wpwraklock the door :)14:40
wpwraklars_: can you please give me commit access to the qi-kernel project ? also, Richard Sharpe would need commit access. i don't think he has an account on projects yet, though15:22
wpwraklars_: well, or i think i can also help myself to the necessary access, with the superpowers invested in me, if you don't mind15:26
lars_wpwrak: go ahead15:31
wpwraklars_: on http://www.kernel.org/pub/software/scm/git/docs/user-manual.html#maintaining-topic-branches they recommend adding a "mytree" entry in .git/config. this isn't really necessary as long as i don't have branches with commits i don't want to push "upstream", is it ?15:46
wpwraklars_: and which branch should i track ? jz-2.6.38 ?15:50
kristianpaulhum http://blog.garbe.us/2008/02/08/01_Static_linking/15:50
lars_wpwrak: yes15:56
lars_kristianpaul: thats partially nonsense i would say16:01
wpwraki concur :)16:02
kyaklars_: do you have an idea why scrollback is not working in 2.6.37 framebuffer?16:02
wpwrakat least many of the statements are way too general16:02
lars_kyak: nope, sorry.16:03
kristianpaullars_: why? i dont even understan well the difference between static and dynmic linking..16:03
xiangfukyak: you mean in command terminal like /dev/tty1 ?16:04
kyakxiangfu: yeah16:04
kyakit works in 2.6.3216:05
lars_kristianpaul: with static linking you but all the code needed by the executable in that one file16:05
kristianpaulyes16:05
kristianpauli used that some times i remeber16:05
lars_kristianpaul: with static objects library code is in shared objects and gets loaded at runtime16:05
kristianpaulso the binary is more portable isnt?16:06
lars_well, it doesn't depend on the shared objects, yes16:06
kristianpaulpersonally i like the part "they will run on the same architecture even in 10 years time."16:06
kristianpaulcan you tell me more what is the nonsense part for you?16:07
xiangfukyak: I remember we should enable something in busybox.16:07
kristianpaulstatic linking saved my ass at work, when i have to install same software agent in different OS, i liked that16:08
xiangfukyak: needs check in detail.16:08
lars_"Statically linked executables use less disk space" - they'll use more disk space, because all of the code required to run the binary is inside the binary16:08
kyakxiangfu: hm... are you sure busybox has something to do with that?16:08
lars_"Statically linked executables consume less memory because their binary size is smaller." - also wrong for the same reasons16:08
kristianpaullars_: oh yes thats true, i have my doubts about that part indeed16:08
kyakxiangfu: i actually took the config.full_system, so all "our" busybox options should be there anyway16:09
xiangfukyak: hmm... not sure. at first we don't the scroll back in terminal. after we enable something. it's work in 2.6.3216:09
kristianpaulin any case i found this sta.li, looks interesting for emdebbed i think16:09
lars_kristianpaul: i have a static busybox install on my laptop as a rescue system16:12
kristianpaulcool :-)16:13
lars_saved me once, when i deleted /lib64 by accident16:15
kristianpaulif understand well you have busybox asociated with your bootloader and is in the same disk or ?16:15
lars_no, it's on the same disk as the rest of the system. i just point the kernel cmdline 'init' parameter to it, when i want to use it16:18
wpwrakyay, branch creation worked ! ;-)16:19
kristianpaul:-)16:19
kristianpaulwpwrak: why you we're afraid dont?16:19
wpwrakkristianpaul: there's always ample potential for mishap when trying something new ;-)16:26
LunaVoraxHi everyone18:14
qi-bot[commit] kyak: w3m: update to 0.5.3, rework patches http://qi-hw.com/p/openwrt-packages/a34928022:40
kristianpaullol digikey.com.co22:51
kristianpaulpatetic.. -_-22:51
tuxbrainkristianpaul: what happens with digikey.com.co?23:29
tuxbrainroh:the sugru think seem very inline with diy hardware stuff :), and the idea of select your color is also not bad , then it will also be on the costurmer choise to hide or not the electronics.... I will take a futher look.23:49
kristianpaultuxbrain: i tought the have stock in colombia, but is just s poor frontend for a minimun 100usd shipping from the US..23:51
wpwrakkristianpaul: when you wrote "pathic", i thought is was outright phishing or such :)23:52
--- Sun Mar 27 201100:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!