IRC log of #maemo for Saturday, 2013-02-16

cehteh	http://www.youtube.com/watch?v=5TLEVo1cEjQ .. crazy americans :P	00:00
*** shamus has quit IRC		00:00
*** shamus has joined #maemo		00:00
DocScrutinizer05	this one guy nuked away half his balcony with an exploding 18650	00:00
cehteh	well i decided that ammo boxes and charging on the balcony (at least when the weather becomes better) makse some sense	00:01
cehteh	heh	00:01
kerio	in other news, my laptop's battery is 50 cycles away from 2000	00:01
cehteh	well and this ammo boxes are very tight sealing .. i need to removed the seals or leave it open when charging at least	00:02
*** juozapas has joined #maemo		00:02
juozapas	hi	00:02
juozapas	where i can find voice files for omvoice server ?	00:03
jacekowski	i've had a fun with fire extinguisher i had one boxed in my car	00:03
jacekowski	and when it came to using it took me a good while to take it out of the box	00:03
*** ghostcube has quit IRC		00:03
cehteh	huh 18650 are small compared to the cells i have here .. 4S (14.4V) x 5000mAh ... over 500g each	00:04
DocScrutinizer05	cehteh: round or square?	00:05
cehteh	square	00:06
cehteh	lipo .. stack of 4 cells	00:06
DocScrutinizer05	usually only the cylindrical cels are really nasty grenades	00:06
cehteh	yes	00:06
cehteh	but the flat ones blow and make a lot mess and if it ignites it a big flame	00:06
kerio	Estel_: btw, how are the plans for the n900 with the two 18650s going?	00:06
DocScrutinizer05	square ones can expand and deform	00:06
DocScrutinizer05	and leak easily when doing so	00:07
cehteh	the round ones are liion in most cases	00:07
cehteh	yes	00:07
kerio	DocScrutinizer05: but as they expand, they become more cylindrical	00:07
kerio	and then they can explode!	00:07
cehteh	rc batteries are quite raw, no protection and stuff	00:07
kerio	cehteh: why? you'd think that considering the space, you could have protective circuits	00:08
DocScrutinizer05	yep, still they won't build up pressure like cyl ones	00:08
cehteh	kerio: price, weight, power	00:08
cehteh	you can draw serveral hundreds Amperes from those	00:08
DocScrutinizer05	rc batteries are built for crazy discharge rates	00:08
kerio	oh, so the circuit would have to be really big	00:09
*** LauRoman has joined #maemo		00:09
cehteh	in my case i use the low speced ones 25C only ... 5A*25 ... 125A already .. but there are cells with 65C available	00:09
cehteh	and 8000mAh	00:09
cehteh	usually you have a more complex charger and in the RC model you have some monitoring and do rather flat cycling	00:10
DocScrutinizer05	simple answer: you want only one mosfet between cell and motor, not two ore even more	00:11
cehteh	balancing charger .. this batteries have a 2nd connector with each single cell level coming out for monitoring and equalizing the charge levels	00:11
DocScrutinizer05	and that mosfet is a monster	00:11
cehteh	usually few of them in parallel	00:11
*** sixwheeledbeast^ has joined #maemo		00:12
DocScrutinizer05	:nod:	00:12
cehteh	havent counted but there are prolly a dozen fets on the speed controlers	00:12
DocScrutinizer05	nice of mosfet that you can parallel them thus simply	00:12
*** sixwheeledbeast has quit IRC		00:12
*** sixwheeledbeast^ is now known as sixwheeledbeast		00:12
DocScrutinizer05	the have pretty convenient characteristics for that	00:13
cehteh	http://0x.ca/sim/esc/Mystery_40A/img_4549.jpg	00:13
DocScrutinizer05	hehe	00:13
DocScrutinizer05	dang, that's a 3-phase controller	00:14
cehteh	sure, brushless motors, electronically commutated	00:15
*** raandoom has joined #maemo		00:15
DocScrutinizer05	those are cool beasts	00:16
DocScrutinizer05	go like hell	00:16
cehteh	standard meanwhile	00:16
cehteh	with hackable firmware :)	00:16
kerio	ooh, apparently my laptop uses a bq20z951	00:20
ShadowJK	I disassembled a dead laptop battery, and it had a bq-something :-)	00:20
kerio	that sounds familiar :D	00:21
eccerr0r	nots sure why my n900 takes so long to charge, can't imagine a dual 18650 2.8AH cell... or two in parallel...	00:30
ShadowJK	2 hours on fresh battery, 4 hours on worn battery	00:32
ShadowJK	probably 4 hours on fresh single 2.8Ah	00:32
*** loganbr` has joined #maemo		00:33
cehteh	my charging is just way off ... it takes hours over hours on 'almost full' .. and when i disconnect it when fully charged and plug it in few minutes later i'll charge for hours again	00:33
* cehteh just blames nokia and lets it be as it is		00:33
cehteh	battery still works fine	00:33
*** florian has quit IRC		00:34
*** shanttu has quit IRC		00:34
*** eichi has joined #maemo		00:34
*** loganbr has quit IRC		00:34
*** NeutrinoPower has joined #maemo		00:35
*** arcean_ has joined #maemo		00:37
*** topro_ has joined #maemo		00:42
topro_	is there a way to get qtmobility support in qt sdk for target maemo 5 (n900)?	00:43
*** muelli has quit IRC		00:43
eccerr0r	so it is true it takes longer to charge an old battery? thought a cc charger would hit finish voltage sooner on a worn cell?	00:44
DocScrutinizer05	cehteh: blame your screen on while charging	00:45
*** NIN101 has quit IRC		00:45
DocScrutinizer05	eccerr0r: we got CCCV charger though	00:46
DocScrutinizer05	eccerr0r: with a charge-end current threshold of like 100mA or sth	00:46
DocScrutinizer05	of which an enabled screen backlight already eats up to 200	00:47
eccerr0r	yes i've noticed that too, any foreground use slows down charge.... to the extent some computer usb ports won't charge/still discharge when the n900 is being used	00:47
DocScrutinizer05	so... never trips threshold for end-of-charge detection when screen is lit	00:47
DocScrutinizer05	yeah, that's because of battery is parallel to system power-in	00:49
DocScrutinizer05	or rather, this special effect is to limited power via USB, which will get split between system and charging, no matter what circuit config	00:49
DocScrutinizer05	is due to...	00:50
eccerr0r	what I don't get is why I can't seem to get it to draw the full 500mA on microusb	00:50
DocScrutinizer05	charge-end detection (or the lack of such) however is becuase charger suplies power for battery and system in parallel	00:51
DocScrutinizer05	hmm, it usually should do that	00:51
DocScrutinizer05	what makes you think it draws less than 500, and how much less?	00:51
eccerr0r	if I bump the voltage I can get it to draw more current... but a constant voltage smps should decrease current when voltage goes up	00:52
DocScrutinizer05	yes	00:52
eccerr0r	i've got it hooked up to a bench psu with voltage/current meters	00:53
DocScrutinizer05	I however wonder how you bump the voltage of a PC USB port	00:53
eccerr0r	behavior is NOT what I expect. ..	00:53
DocScrutinizer05	the bq24150 is not a virtual ideal component	00:54
eccerr0r	yeah for a pc port it'd be hard but not for a test jig :)	00:54
DocScrutinizer05	a test jig may fail on USB enum	00:54
DocScrutinizer05	which is mandatory to draw more than 100mA from USB without D+- short fastcharger signalling	00:55
*** Vlad_on_the_road has quit IRC		00:55
eccerr0r	(same jig that I killed the charging circuit on my 5230...)	00:55
eccerr0r	I shorted the d pins together.	00:55
DocScrutinizer05	then it should take way more than 500mA	00:56
eccerr0r	it's not...	00:56
DocScrutinizer05	if CCCV plus system usage allows for that	00:56
*** amospalla has quit IRC		00:56
*** eichi has quit IRC		00:56
DocScrutinizer05	you should use bq27200.sh or...	00:57
*** amospalla has joined #maemo		00:57
DocScrutinizer05	~bq27k-detail	00:57
infobot	rumour has it, bq27k-detail is http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27k-detail2	00:57
DocScrutinizer05	it gives an idea if cell is still in CC or already in CV segment	00:57
DocScrutinizer05	by merely looking at current-now and voltage-now	00:58
DocScrutinizer05	the transit from CC to CV is not ideal though, there's a zone of mixed characteristics where both apply	00:58
*** xes has quit IRC		00:59
DocScrutinizer05	again, since bq24150 is a cheap real silicon component, not something ideal virtual	00:59
DocScrutinizer05	it's just "good enough2 for charging LiIon	01:00
DocScrutinizer05	s/2/"/	01:00
infobot	DocScrutinizer05 meant: it's just "good enough" for charging LiIon	01:00
ds3	DocSCrutinizer05: what do you consider wrong with that chip?	01:00
*** amospalla has quit IRC		01:01
*** amospalla has joined #maemo		01:01
DocScrutinizer05	nothing wrong with it, it's just not something mathematical	01:02
DocScrutinizer05	but real hardware	01:02
ds3	mathematical?	01:02
DocScrutinizer05	ideal	01:02
ds3	hmmm ok... still not sure in what aspects	01:03
DocScrutinizer05	like a diode with 0.0000V Vforward, infinite Vreverse	01:03
ds3	oh	01:03
ds3	got a specific example on that chip?	01:03
DocScrutinizer05	it has no sharp switching from CC to CV, both blend into each other	01:03
ds3	ah... okay. that i understand	01:04
DocScrutinizer05	on CV it doesn't have 0 Ohms source impedance	01:04
DocScrutinizer05	on CC it has dependencies on cell voltage as well	01:04
DocScrutinizer05	so neither voltage (in CV) nor current (in CC) are really constant	01:05
DocScrutinizer05	and particularly not in the zone between CC and CV	01:05
DocScrutinizer05	the watchdog timer specified as 32s might actually be as short as 15s	01:07
DocScrutinizer05	iirc	01:07
DocScrutinizer05	it's all just good enough to do the job it's built for	01:07
DocScrutinizer05	and that job means "5V USB"	01:08
DocScrutinizer05	not "3-9V USB"	01:08
DocScrutinizer05	so no surprise eccerr0r found it drawing more current when he increases Vusb	01:09
DocScrutinizer05	chip isn't built to handle that	01:09
eccerr0r	i just wish it could draw the full 500mA @ 5V, maybe it'd do 500mA at 9V	01:09
DocScrutinizer05	it can draw way more than 500mA at 5V	01:10
DocScrutinizer05	evidently	01:10
eccerr0r	i've never seen it do so	01:10
DocScrutinizer05	I did	01:11
eccerr0r	what did you use to measure/what source?	01:11
DocScrutinizer05	simple math and bq27200.sh to read out charge current of battery	01:12
DocScrutinizer05	plus some knowledge of system power consumption, also derived from previous bq27200.sh runs	01:14
eccerr0r	need orthogonal measurement... I cant trust bq24150.	01:14
DocScrutinizer05	when I see battery getting charged at 800mA (3.3V) and I know screen backlight uses e.g. 200mA at 4.1V, then the match is simple to calculate how much the USB is currently delivering minimum (even on a neglected converter efficiency that's not 1 either)	01:16
*** topro_ has quit IRC		01:17
* kerio has seen 900mA, measured by bq27k		01:18
eccerr0r	also something weird:I don't see a current swing when locking the device... which corresponds to why having the screen on slows charging...	01:18
kerio	fastcharger, battery almost empty	01:18
DocScrutinizer05	but charging current for battery is limited to 850mA iirc	01:18
kerio	DocScrutinizer05: not 1A?	01:19
DocScrutinizer05	bme sets it to 850 iirc	01:20
eccerr0r	all of this experimentation is just to build a fast charger... the d+- short wasn't enough	01:20
*** raandoom has quit IRC		01:20
DocScrutinizer05	well, see my last 2 lines	01:20
DocScrutinizer05	usually you can't charge batter with more than 0.7C	01:22
kerio	what does the nokia wallcharger do? constant 5V?	01:22
eccerr0r	I'd be happy to see 850mA	01:22
DocScrutinizer05	and that's not even taking into account the system power which gets subtracted from that	01:22
eccerr0r	It's nowhere close...	01:22
DocScrutinizer05	kerio: sure	01:23
kerio	the battery charges at around 500 from around the midpoint onwards	01:23
kerio	so deal with it	01:23
kerio	you only get the truly fast charge at the beginning, when it's almost empty	01:23
DocScrutinizer05	it will charge with 800 or more only while <3.6V	01:23
DocScrutinizer05	yes, exactly	01:24
eccerr0r	ok... that may be the fact needed.	01:24
DocScrutinizer05	and no matter what you do, LiIon takes its time to charge to 100%	01:24
eccerr0r	i'll discharge a battery down to 3.4 and see if it behaves different...	01:25
DocScrutinizer05	the faster (higher current) you charge it initially, the longer it will take for the CV segment	01:25
kerio	also, don't push more than 5V through the usb	01:25
kerio	it's not really meant to do that :s	01:25
DocScrutinizer05	which largely kills any benefit of "fast charging"	01:25
*** Darkchaos has quit IRC		01:25
DocScrutinizer05	use nokia fastcharger and switch off all powerhogs in system, to get fastest possible 'normal' charging which will need ~3h to from empty to battery almost full. Use ShadowJK's charge21.sh (with edited CC parameter) to charge battery with 1C instead 0.7C, though that's not recommended since it wears out battery quite fast	01:29
DocScrutinizer05	nothing you could do to USB-charger will change that	01:30
eccerr0r	hmm... honestly you guys are confusing me more, there are definitely some facts that mesh with what I know but others seem to clash... then again I'm coming from a completely-hardware aspect...	01:30
cehteh	DocScrutinizer05: when idle, no screen on, LED blinking	01:31
DocScrutinizer05	ideal charger delivers constant 5.25V @1.5+A and has D+- short	01:31
eccerr0r	but that is also true, fast charging wears batteries faster.	01:31
*** waelsy123 has quit IRC		01:32
*** LauRoman has quit IRC		01:32
eccerr0r	the BL5J is 4.9WH. At 5V1A it should take about 1 hour at constant power to charge, if it were possible.	01:32
kerio	you'd wreck it	01:33
*** NIN101 has joined #maemo		01:34
eccerr0r	it is technically "possible" of course but that is true, you cant charge the last bit at constant power...	01:34
eccerr0r	I guess I need to see that the thing will eat 800mA while charging a flat battery, if this is true, then this will all make sense.	01:36
eccerr0r	and wonderful, t-mobile disappeared, not getting GSM coverage right now...	01:38
*** NIN101 has quit IRC		01:40
*** Ethernin has joined #maemo		01:45
*** eichi has joined #maemo		01:53
*** sardini has quit IRC		01:53
MohammadAG	meh, $25 for publishing to Google Play	01:54
merlin1991	of what?	01:55
MohammadAG	apps :P	01:55
DocScrutinizer05	meh	01:57
*** disco_stu has quit IRC		02:00
*** OkropNick has quit IRC		02:10
*** shamus has quit IRC		02:17
*** shamus has joined #maemo		02:17
deepy	kerio: I think I need to sell my N900	02:23
deepy	I just won another phone	02:23
*** rcg has quit IRC		02:26
*** hardaker has quit IRC		02:31
SpeedEvil	:-\	02:35
*** goldkatze has quit IRC		02:35
SpeedEvil	what?	02:35
*** eijk has quit IRC		02:41
*** LaoLang_cool has joined #maemo		02:44
*** Hurrian has joined #maemo		02:49
deepy	SpeedEvil: I got a N900 to replace my Nexus One	02:52
deepy	and then I won a phone	02:52
deepy	and now I've won another one	02:52
*** LaoLang_cool has quit IRC		02:58
*** otypoks_ has quit IRC		02:58
*** dafox has quit IRC		03:03
*** valerius has quit IRC		03:15
*** robbiethe1st has joined #maemo		03:18
*** dos1 has quit IRC		03:22
*** dos11 has joined #maemo		03:22
*** nslu2-log has quit IRC		03:22
*** Aoyagi has left #maemo		03:28
*** NeutrinoPower has quit IRC		03:31
*** Martix has quit IRC		03:37
*** bef0rd has joined #maemo		03:39
*** M4rtinK has quit IRC		03:50
merlin1991	deepy: haxx0r	03:51
*** kolp has quit IRC		04:00
Estel_	ShadowJK, thanks a lot	04:09
Estel_	and gregoa for anotherr lightweight httpd	04:10
Estel_	btw, using wireshark, I've also noticed a simple way to start this certain game "rebuilt 2" outside embedded sites, without need for fake domain	04:10
Estel_	after some tweaks, it works great offline using N900 browser with cursor mode (100% fluent), so wee just got another complex game (rebuilt 2 is post-apocalyptic sim, quite complicated one, despite being, technically, flash game)	04:12
Estel_	now I just need to ask developer's permission, and I can package it with tweaks included, to non-free - if repos ever start working again without hash mismatch, that is	04:12
*** MetalGearSolid has joined #maemo		04:15
*** arcean_ has quit IRC		04:22
*** loganbr`` has joined #maemo		04:24
*** eichi has quit IRC		04:25
*** dos11 has quit IRC		04:27
*** loganbr` has quit IRC		04:28
*** nox- has quit IRC		04:31
*** FlameReaper has joined #maemo		04:37
*** jrocha has quit IRC		04:41
*** Pali has quit IRC		04:47
*** Hurrian has quit IRC		04:56
*** maybeWTF has joined #maemo		05:00
*** nslu2-log has joined #maemo		05:02
*** maybeHere has quit IRC		05:03
*** hardaker has joined #maemo		05:18
*** shamus has quit IRC		05:24
*** shamus has joined #maemo		05:24
*** guampa has quit IRC		05:25
*** MentalistTraceur has left #maemo		05:32
*** guampa has joined #maemo		05:38
*** disco_stu has joined #maemo		05:48
*** shamus has quit IRC		05:54
*** shamus has joined #maemo		05:54
*** FlameReaper has quit IRC		05:58
*** dockane_ has joined #maemo		06:00
*** Hurrian has joined #maemo		06:01
*** lxp1 has joined #maemo		06:01
*** DocScrutinizer05 has quit IRC		06:02
*** DocScrutinizer05 has joined #maemo		06:02
*** dockane has quit IRC		06:03
*** lxp has quit IRC		06:04
*** teotwaki_ has quit IRC		06:05
*** Ex-Opesa has quit IRC		06:10
*** FlameReaper has joined #maemo		06:23
*** uen\| has joined #maemo		06:34
*** uen has quit IRC		06:38
*** uen\| is now known as uen		06:38
*** waelsy123 has joined #maemo		06:40
*** radic has joined #maemo		06:46
*** radic_ has quit IRC		06:50
*** shamus has quit IRC		07:15
*** waelsy1231 has joined #maemo		07:15
*** shamus has joined #maemo		07:16
*** waelsy123 has quit IRC		07:16
*** waelsy1231 has quit IRC		07:22
*** waelsy123 has joined #maemo		07:22
*** robbiethe1st has quit IRC		07:24
*** hardaker has quit IRC		07:28
*** iluminator105 has joined #maemo		07:30
iluminator105	is vlc in the repos	07:30
*** Ex-Opesa has joined #maemo		07:30
*** hardaker has joined #maemo		07:32
*** MetalGearSolid has quit IRC		07:33
*** hardaker has quit IRC		07:38
*** MetalGearSolid has joined #maemo		07:43
*** waelsy123 has quit IRC		07:48
*** shamus has quit IRC		08:07
*** shamus has joined #maemo		08:07
*** ron0062000 has joined #maemo		08:11
*** bef0rd has quit IRC		08:22
*** ron0062000 has quit IRC		08:36
*** MetalGearSolid has quit IRC		08:50
*** emma has joined #maemo		08:52
*** sLumPia has joined #maemo		08:53
*** MetalGearSolid has joined #maemo		08:56
*** bef0rd has joined #maemo		09:04
*** bef0rd has quit IRC		09:14
*** leochencipher has joined #maemo		09:15
*** leochencipher has quit IRC		09:20
*** MetalGearSolid has quit IRC		09:23
*** iluminator105 has quit IRC		09:31
*** Hurrian has quit IRC		09:47
*** LauRoman has joined #maemo		09:49
*** rcg has joined #maemo		09:55
*** SmilyOrg has joined #maemo		09:55
*** Smily has quit IRC		09:59
*** Hurrian has joined #maemo		10:02
*** eichi has joined #maemo		10:20
*** croppa is now known as croppa_		10:22
*** croppa_ is now known as croppa		10:22
*** leochencipher has joined #maemo		10:25
*** sixwheeledbeast is now known as sixwheeledbeast_		10:25
*** sixwheeledbeast1 has joined #maemo		10:26
*** sixwheeledbeast_ has quit IRC		10:26
*** Darkchaos has joined #maemo		10:28
*** sasquatch has quit IRC		10:29
*** sasquatch has joined #maemo		10:29
*** Vlad_on_the_road has joined #maemo		10:29
kerio	i want steam on fremantle!	10:34
*** NIN101 has joined #maemo		10:53
*** eichi has quit IRC		10:55
*** eichi has joined #maemo		11:04
*** githogori has quit IRC		11:04
*** githogori has joined #maemo		11:07
*** Pali has joined #maemo		11:08
*** hubutm20 has joined #maemo		11:13
*** FlameReaper has quit IRC		11:16
*** fuz_ has quit IRC		11:23
*** guampa has quit IRC		11:25
*** OkropNick has joined #maemo		11:28
*** fuz_ has joined #maemo		11:35
*** guampa has joined #maemo		11:38
*** leochencipher has quit IRC		11:43
*** muelli has joined #maemo		11:44
*** M4rtinK has joined #maemo		11:45
*** dvoid_ has quit IRC		11:48
*** dvoid has joined #maemo		11:48
*** sLumPia has quit IRC		12:18
*** kolp has joined #maemo		12:18
*** fortytwo has joined #maemo		12:23
*** xes has joined #maemo		12:25
*** xes has joined #maemo		12:25
*** RzR has quit IRC		12:26
*** RzR has joined #maemo		12:28
*** xes has quit IRC		12:31
Raimu	You know your culture from your trash	12:35
Raimu	You know your plastic from your cash	12:35
*** xes has joined #maemo		12:35
*** xes has quit IRC		12:35
*** xes has joined #maemo		12:35
*** githogori has quit IRC		12:38
*** githogori has joined #maemo		12:52
*** Vrm1 has joined #maemo		13:11
*** mrglinux has joined #maemo		13:21
*** mrglinux has left #maemo		13:21
*** LauRoman has quit IRC		13:23
*** ghostcube has joined #maemo		13:28
*** xes has quit IRC		13:32
*** Dynamit has quit IRC		13:36
*** rcg has quit IRC		13:36
*** muelli has quit IRC		13:39
*** Dynamit has joined #maemo		13:40
*** fuz_ has quit IRC		13:41
*** till- has quit IRC		13:43
*** MetalGear has joined #maemo		13:52
*** florian has joined #maemo		14:00
*** Vrm1 has quit IRC		14:00
*** fuz_ has joined #maemo		14:05
*** brzys has quit IRC		14:06
*** brzys has joined #maemo		14:09
*** arcean has joined #maemo		14:09
*** till- has joined #maemo		14:13
MohammadAG	wow, my router's firmware is hardcoded to only accept Israeli upgrades	14:16
MohammadAG	and the original firmware is nowhere to be found	14:16
*** croppa has quit IRC		14:17
kerio	MohammadAG: not even through tftp?	14:18
juozapas	where i can find voice files for omvoiceserver?	14:18
MohammadAG	kerio, not sure actually	14:18
*** techlife has quit IRC		14:18
MohammadAG	kerio, I found some forum thread with someone having the same problem	14:18
MohammadAG	they changed the header so the web UI accepted it	14:19
*** till- has quit IRC		14:19
MohammadAG	and the router died :)	14:19
MohammadAG	oh nice, busybox on the router doesn't have grep	14:19
*** till- has joined #maemo		14:20
*** techlife has joined #maemo		14:25
*** FlameReaper has joined #maemo		14:26
*** MohammadAG has quit IRC		14:29
*** Aoyagi has joined #maemo		14:30
*** MetalGearSolid has joined #maemo		14:34
*** FReaper has joined #maemo		14:41
*** FlameReaper has quit IRC		14:43
*** zl2tod has quit IRC		14:44
*** FReaper is now known as FlameReaper		14:44
*** eMHa__ has quit IRC		14:44
*** eMHa__ has joined #maemo		14:44
*** xes has joined #maemo		14:51
*** xes has quit IRC		14:51
*** xes has joined #maemo		14:51
*** eijk has joined #maemo		14:54
kerio	i'm really stumped :(	15:07
kerio	should i just reflash?	15:07
kerio	i have so many customizations :(	15:07
*** jrocha has joined #maemo		15:15
*** dos11 has joined #maemo		15:17
*** yana has joined #maemo		15:21
yana	hello! i just installed webos on my n900 but i cant understand how to download ipk files, from where and how to put the on my phone.... any suggestions? thanks in advance	15:22
yana	them*	15:22
Aoyagi	I uh... don't think WebOS question is very appropriate in #maemo. But maybe someone will know...	15:23
yana	i hope so ...	15:25
*** konelix has joined #maemo		15:27
*** konelix has quit IRC		15:33
*** NeutrinoPower has joined #maemo		15:34
*** yana has quit IRC		15:37
*** otypoks has joined #maemo		15:46
*** MetalGearSolid has quit IRC		15:47
*** dos11 has quit IRC		15:48
*** konelix has joined #maemo		15:51
*** dos1 has joined #maemo		15:53
*** Hurrian has quit IRC		15:54
*** valerius has joined #maemo		15:55
*** topro_ has joined #maemo		15:55
*** dos1 has quit IRC		15:56
*** dos1 has joined #maemo		15:58
*** xes has left #maemo		16:00
*** xes has joined #maemo		16:05
*** gonzo__ has joined #maemo		16:06
*** gonzo__ has left #maemo		16:07
*** l3jk has joined #maemo		16:08
*** xes has quit IRC		16:09
*** jrocha has quit IRC		16:13
*** dos1 has quit IRC		16:15
qwazix	WebOS on the N900?	16:15
Aoyagi	He/she is gone.	16:16
kerio	deepy: why bother selling it	16:16
*** xes has joined #maemo		16:16
*** SmilyOrg is now known as Smily		16:17
qwazix	Aoyagi I just saw that nobody was surprised and I thought wtf? is that so normal?	16:18
Aoyagi	I was.	16:18
Aoyagi	But I try to not voice my opinions about things I know very little about.	16:19
*** merlin1991 has quit IRC		16:19
qwazix	I just wanted to know more about that. Technically it's possible to install webos on the N900	16:19
*** merlin1991 has joined #maemo		16:20
qwazix	In fact the Pre had pretty similar hardware.	16:20
*** loganbr`` has quit IRC		16:20
kerio	merlin1991: µ	16:20
merlin1991	hm?	16:20
kerio	~moo	16:21
* infobot mooooooooo! I am cow, hear me moo, I weigh twice as much as you. I am cow, eating grass, methane gas comes out my ass		16:21
*** sq-one has joined #maemo		16:22
*** pvanhoof has joined #maemo		16:23
*** loganbr`` has joined #maemo		16:24
kerio	~µ	16:28
kerio	~literal µ	16:28
kerio	infobot: you suck	16:28
infobot	and very well I might add	16:28
*** konelix has quit IRC		16:30
*** konelix_ has joined #maemo		16:30
*** MetalGearSolid has joined #maemo		16:30
*** konelix_ is now known as konelix		16:35
*** FlameReaper has quit IRC		16:35
*** l3jk has left #maemo		16:41
*** mvp_ has joined #maemo		16:45
*** topro__ has joined #maemo		16:47
*** topro_ has quit IRC		16:50
*** topro__ is now known as topro_		16:50
*** arcean_ has joined #maemo		16:59
*** arcean has quit IRC		17:02
*** waelsy123 has joined #maemo		17:07
*** loganbr`` has quit IRC		17:07
*** loganbr`` has joined #maemo		17:09
*** rcg has joined #maemo		17:10
*** jrocha has joined #maemo		17:12
*** Milhouse has joined #maemo		17:14
*** hardaker has joined #maemo		17:15
*** WizardNumberNext has joined #maemo		17:19
*** FlameReaper has joined #maemo		17:20
WizardNumberNext	hey guys I need execute something every set number of seconds, including fractions (like 2.56s). any way to do so without sleep command? I need quite accurate timming, preferably up to 1ms, if possible	17:22
*** Martix has joined #maemo		17:22
kerio	"execute" how?	17:24
kerio	also, afaik topro investigated this, and ended up figuring out that the kernel timers have a resolution of 50ms	17:25
kerio	you'll have to use a soundcard timer	17:25
topro_	really interesting, we just had that topic ;)	17:30
topro_	WizardNumberNext: i couldn't get kernel based timers to run at higher res then 50ms	17:30
topro_	but my task was to detect peaks in an audio stream recoreded from mic. so I went with counting samples	17:31
WizardNumberNext	even 50ms would do, 1s resolution of sleep is unacceptable	17:31
WizardNumberNext	I would execute it in script	17:31
topro_	ever heard of usleep() ?	17:31
WizardNumberNext	do not really want to program	17:32
kerio	most likely, loading the binary and launching it will take more than 1ms	17:32
topro_	then I'm off, don't even know if there is a way to access timers in script. you would need a binary that does the job. is there something in gnu? don't now	17:33
WizardNumberNext	kerio, I expect everything takes time, so loading and executing (especially loading)	17:33
WizardNumberNext	then only way would be to put work into kernel, as I want to access i2c	17:34
WizardNumberNext	bq27x00	17:34
WizardNumberNext	to be more precise - I want nice battery logger	17:34
kerio	...WHAT	17:35
WizardNumberNext	like I had on N95 - Nokia energy profiler - something not exactly similar	17:35
xes	you mean a battery burner..	17:35
kerio	i don't even know if bq27k updates once per second	17:35
kerio	let alone once every millisecond	17:35
WizardNumberNext	Nokia Energy Profiler wasn't really battery burner - yes it was stilling some energy, but it wasn't bad	17:35
WizardNumberNext	kerio, no, not every ms - every 2.56s	17:36
WizardNumberNext	more often is completely pointless	17:36
*** dhbiker has joined #maemo		17:36
topro_	I'd like to understand that 2.56s constraint	17:37
WizardNumberNext	I want to log every sample of average current, which happens to be updated every 2.56s	17:37
*** qwazix is now known as qwazix_900		17:37
topro_	come on, why not sample every 2 or every 3 seconds. the error will level out over time, don't you think so?	17:37
WizardNumberNext	I have been reading datasheet some time ago and decided that this would best approach to get some real usage	17:37
*** qwazix_900 is now known as qwazix		17:37
kerio	do 5 seconds	17:37
kerio	so you'll have 120ms to do your processing :)	17:38
WizardNumberNext	then, I am loosing every second sample - useless	17:38
WizardNumberNext	all processing needed is to save it to file (might be even in ram/tmpFS	17:39
*** loganbr`` has quit IRC		17:40
*** dos1 has joined #maemo		17:42
*** loganbr`` has joined #maemo		17:42
*** jrocha has quit IRC		17:45
ShadowJK	WizardNumberNext; it's updated every 5.12s, approximately	17:47
WizardNumberNext	there's some value considering current updated every 2.56s	17:47
ShadowJK	I would expect bq's oscillator isn't terribly accurate anyway	17:48
ShadowJK	running bq27200.sh in a 5s loop does redcue idle battery lifr by a factor of 5 or so :)	17:49
xes	WizardNumberNext: first of all try a "time yourscript.sh"...	17:50
*** dhbiker has quit IRC		17:50
*** dhbiker has joined #maemo		17:51
ShadowJK	ah yes, -2.2% ... 1.5% stated accuracy.. so that ends up as 5.01 - 5.20s update interval, if accessing the chip doesn't postpone it updating :)	17:52
*** NIN101 has quit IRC		17:53
*** bef0rd has joined #maemo		17:56
kerio	ShadowJK: well, it could be your script's fault :P	17:58
ShadowJK	I suspect a "while true; do sleep 5; done" has similar impact on battery life :)	17:59
kerio	well, bq27200.sh does a lot of calculations in shellscript	18:00
kerio	and a lot of requests with i2cget	18:00
ShadowJK	ya	18:01
ShadowJK	but wakeups do more for power use really	18:01
WizardNumberNext	rigth - I have seen all of it. I have my own script which is reading sysfs file and it have quite a latency	18:01
*** NIN101 has joined #maemo		18:02
*** loganbr`` has quit IRC		18:04
*** loganbr`` has joined #maemo		18:05
DocScrutinizer05	kerio: use bq27k-detail, it only has one i2cdump command instead of lots of i2cget	18:11
DocScrutinizer05	thus it should be considerably more lightweight regarding that	18:11
kerio	DocScrutinizer05: hm, i wonder if it could be made to use bq27x00_battery	18:12
DocScrutinizer05	it should almost work OOTB, since that's basically what it does on Openmoko	18:13
*** loganbr`` has quit IRC		18:13
kerio	DocScrutinizer05: hm, what's the path to that script, in your site?	18:13
DocScrutinizer05	so if the developer of bq27x00_battery followed my advice, there shall be a similar sysnode to dump raw bq27200 register content	18:14
kerio	oh, maemo5/usr/local/sbin	18:14
DocScrutinizer05	~bq27k-detail	18:14
kerio	DocScrutinizer05: there is!	18:14
infobot	bq27k-detail is probably http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27k-detail2	18:14
*** loganbr`` has joined #maemo		18:15
kerio	DocScrutinizer05: http://0paste.com/264 this is the content of the registers file	18:17
DocScrutinizer05	AYAYAYAYAYAY	18:18
DocScrutinizer05	waaaay too much "wisdom" in that one	18:18
DocScrutinizer05	why are some 16bit values while others are 8bit?	18:19
kerio	ah balls, rmmodding bq27k caused hald-addon-bme to shut down my phone	18:20
DocScrutinizer05	that's not raw, that's obfuscated interpreted semantic of the chip	18:20
*** Neutron19 has quit IRC		18:20
kerio	Pali: why are some values 16bit and other values 8bit?	18:20
*** loganbr`` has quit IRC		18:20
Pali	where?	18:20
kerio	bq27200-0/registers	18:21
Pali	because some are 16bit and some are 8	18:21
Pali	see bq datasheet	18:21
*** loganbr`` has joined #maemo		18:21
Pali	kerio, hald-addon-bme shut down device?	18:22
DocScrutinizer05	no, I2C has no 16bit registers	18:22
kerio	Pali: i assume so	18:22
DocScrutinizer05	at least not on this chip	18:22
DocScrutinizer05	raw dump is not meant to interprete semantics of chip	18:22
Pali	kerio, kernel oops or dsme reboot?	18:22
kerio	Pali: shutdown	18:23
kerio	i turned it back on with the power key	18:23
Pali	shutdown? not reboot?	18:23
topro_	is there a way to get qtmobility support for qt sdk maemo toolchain?	18:23
*** dos1 has quit IRC		18:23
kerio	and no oopsies	18:23
*** dos1 has joined #maemo		18:23
kerio	Pali: yeah	18:23
Pali	seems like "battery too low" and EDV1 flag	18:24
Pali	and if not there is bug...	18:24
Pali	or temperature?	18:24
kerio	Pali: well, hal returns 0 if there's no matching device right?	18:24
kerio	Pali: idk, i always rmmod rx51_battery	18:24
Pali	for which "returns 0"?	18:25
kerio	the function to get a value from hal	18:26
Pali	yes	18:26
kerio	DocScrutinizer05: fwiw it's not supposed to be a raw i2c dump	18:26
kerio	it's supposed to be the file containing the registers of the chip	18:26
DocScrutinizer05	I honestly suggest a format similar to result of `i2cdump -y 2 0x55 b\|tail -n +2\|cut -d ' ' -f 2-17`	18:27
DocScrutinizer05	kerio: the registers of that chip are all 8bit	18:27
Pali	but some values are 2bytes	18:28
DocScrutinizer05	as you can tell from address on left side	18:28
DocScrutinizer05	Pali: that is irrelevant for a dump	18:28
DocScrutinizer05	it's wisdom about chip semantics that is misplaced in a dump	18:29
Pali	that sysfs file in kernel driver was for shell scripting instead calling i2cget	18:29
kerio	Pali: fwiw, if you're merging 2-byte registers, then you should also split bitfields	18:29
DocScrutinizer05	it's the software using the dump content that's supposed to make assumptions about such properties as pairing of some of the registers	18:30
kerio	DocScrutinizer05: not if you're converting something that used i2cget to get a 2-bytes value	18:30
Pali	btw, proper kernel interface for that is regmap	18:30
DocScrutinizer05	kerio: exactly	18:30
Pali	and I'm not going to change it in bq27x00 battery driver	18:30
DocScrutinizer05	(for bitfields)	18:30
kerio	Pali: what's that? like a conventional way to present registers?	18:31
DocScrutinizer05	kerio: "eh??" for the remark about i2cget	18:31
Pali	something like that	18:31
kerio	DocScrutinizer05: i2cget has a way to get a word instead of a byte	18:31
Pali	interface for accessing some kind of registers	18:31
DocScrutinizer05	kerio: so what?	18:32
DocScrutinizer05	kerio: only if YOU tell it so	18:32
DocScrutinizer05	so it's YOU who has the knowledge, NOT i2cget or even the chip itself	18:32
Pali	registers file in bq27x00_battery kernel driver is for bq27x00 chip	18:32
kerio	i think the point was that you'd "always" use that to get the appropriate value	18:32
Pali	so output from that file is correct	18:33
DocScrutinizer05	no, it's garbled	18:33
*** valerius has quit IRC		18:33
DocScrutinizer05	it introduces sematnics where there shouldn't be any	18:33
DocScrutinizer05	a dump is a dump is a dump	18:34
kerio	it's not called "dump", it's called "registers"	18:34
DocScrutinizer05	a dump has no idea about register pairing	18:34
*** _rd has joined #maemo		18:35
DocScrutinizer05	a dump sysnode is supposed to have a uniform format that allows access to registers WITHOUT knowing about their properties	18:35
DocScrutinizer05	IOW it should be identical for all chips, no matter what register pairing or other properties those registers have	18:36
DocScrutinizer05	If I were interested in highly sophisticated interpretation of the registers' meanings, I'd not use the dump sysnode but the named sysnodes which are supposed to do all interpreting for me	18:38
DocScrutinizer05	dump is meant to do no interpreting for me	18:39
DocScrutinizer05	see kerio's remark about bitfields	18:39
DocScrutinizer05	just adding a random subset of semantic interpretation to dump sysnode renders it garbled and basically useless	18:40
DocScrutinizer05	how the fsck am I supposed to know quite basic things like e.g. if the paired register is big-endian or little-endian and if that garbled dump already swapped registers for me or not?	18:41
*** bef0rd has quit IRC		18:42
DocScrutinizer05	please fix that, it's a bug	18:42
DocScrutinizer05	kerio: it doesn't matter if it's called dump or registers. Registers also are all 8bit in this chip. Actually I2C has no means to access 16bit registers, they are handled as a pair of 8bit registers and it needs prior knowledge to tell those apart form plain 8bit registers. I2C or chip doesn't provide any flag signalling the paired property of some of the registers, you need datasheet to learn about it	18:47
kerio	DocScrutinizer05: otoh, bq27x00_battery does indeed "have" the datasheet	18:47
*** loganbr`` has quit IRC		18:47
DocScrutinizer05	and info from datasheet has no room in either dump or register sysnodes	18:47
*** valerius has joined #maemo		18:48
DocScrutinizer05	it's like if `wc -c` would count UTF16 as one byte	18:49
kerio	you mean wc -m	18:50
kerio	:)	18:50
DocScrutinizer05	no, I mean -c	18:50
kerio	wc -m does indeed count the characters, according to your locale	18:50
DocScrutinizer05	-m is what you get on sysnodes like capacity and current and temperature	18:51
DocScrutinizer05	let me put it as simple as possible: that sysnode is called "registers", that chip has no 16bit registers	18:52
DocScrutinizer05	ETX	18:52
*** bef0rd has joined #maemo		18:52
*** dimir has quit IRC		18:55
*** topro_ has quit IRC		18:55
DocScrutinizer05	you're aware that "registers" sysnode is meant to be r/w? you're also aware that there might be (and actually are) situations where you don't want to write 16bit to a paired set of registers in unknown sequence and timing?	18:56
*** topro_ has joined #maemo		18:56
kerio	DocScrutinizer05: RST	18:56
kerio	you can't TX after ETX dude	18:56
*** _rd has quit IRC		18:57
*** _rd has joined #maemo		18:58
DocScrutinizer05	shall I SYNC and retransmit?	18:59
*** hardaker has quit IRC		19:01
eccerr0r	hmm... is the n900 port of pidgin source archived on maemo.org, or is it integrated into the main distribution?	19:03
eccerr0r	need to see if I can get it to stop pinging the servers every 30 seconds...that will kill the battery :P	19:03
DocScrutinizer05	~pkgs	19:04
DocScrutinizer05	~pkg	19:04
infobot	methinks pkg is http://maemo.org/packages/	19:04
DocScrutinizer05	~#maemo pkgs is <reply>see #maemo pkg	19:04
infobot	okay, DocScrutinizer05	19:04
DocScrutinizer05	~pkgs	19:04
infobot	i heard #maemo pkg is http://maemo.org/packages/	19:04
eccerr0r	not sure how debian works, is the code actually rebuilt automatically once in a while?	19:05
eccerr0r	on the repository that is (if it's OSS)?	19:06
DocScrutinizer05	errr, it is when you upload to drop aka autobuilder	19:06
DocScrutinizer05	I don't think it's usually ever rebuilt later on without any trigger	19:06
*** dimir has joined #maemo		19:07
DocScrutinizer05	since, what would change?	19:07
eccerr0r	Ok... wonderring if it's easier to get the maintainer to stop it from pinging, I see it dropping an ssl packet every 30 seconds	19:07
eccerr0r	that seems excessive...	19:07
kerio	eccerr0r: make your own package	19:07
DocScrutinizer05	locally	19:08
eccerr0r	hence finding the code :D	19:08
kerio	or try to look for a way to increase the keepalive frequency	19:08
DocScrutinizer05	period	19:08
DocScrutinizer05	lower frequency	19:08
kerio	er, yes	19:08
kerio	i'm sorry, mr engineer :(	19:08
eccerr0r	I just don't know if the maemo patches made it into the main distribution	19:08
* DocScrutinizer05 suggests gconftool and hexedit		19:08
kerio	eccerr0r: just use the same source that's in maemo extras	19:09
DocScrutinizer05	eccerr0r: most probably they didn't	19:09
kerio	like, apt-source pidgin or whatever	19:09
eccerr0r	Ok, yeah that's what I was hoping...	19:09
kerio	no, apt-get source	19:09
kerio	after you've enabled a source repo	19:09
eccerr0r	again this is still very new to me... in Gentoo I get the source only so it's easy to find the code :)	19:09
kerio	or just download tarball, dsc and patches	19:09
kerio	eccerr0r: http://repository.maemo.org/extras-devel/pool/fremantle/free/source/p/pidgin/pidgin_2.10.6+nix-1.tar.gz	19:10
*** dhbiker has quit IRC		19:10
DocScrutinizer05	+nix???	19:10
DocScrutinizer05	wtf is that?	19:11
eccerr0r	ah cool...	19:11
DocScrutinizer05	or simply follow the links on ~pkgs	19:11
eccerr0r	need to get vbox set up next.	19:11
kerio	unpack that tarball in your scratchbox, do your modifications, dch -i and increase the version by adding a +eccerr0r1 suffix, dpkg-buildpackage -rfakeroot	19:11
kerio	and it'll give you a tarball	19:11
kerio	er	19:11
kerio	a .deb	19:11
kerio	DocScrutinizer05: nix is apparently the maintainer	19:12
DocScrutinizer05	duh	19:12
*** dafox has joined #maemo		19:12
kerio	it's still wtf	19:12
topro_	sorry for asking again, but I wonder why no one reacted to this as i think that would bot only be interesting to me: is there a way to get qtmobility support for qt sdk maemo toolchain?	19:12
kerio	topro_: idk, i have qtmobility installed i think	19:12
* DocScrutinizer05 looks at (duck and cover!) http://maemo.org/packages/search/?org_maemo_packages_search%5B1%5D%5Bproperty%5D=name&org_maemo_packages_search%5B1%5D%5Bconstraint%5D=LIKE&org_maemo_packages_search%5B1%5D%5Bvalue%5D=pidgin&org_maemo_packages_search%5B2%5D%5Bproperty%5D=title&org_maemo_packages_search%5B2%5D%5Bconstraint%5D=LIKE&org_maemo_packages_search%5B2%5D%5Bvalue%5D=same and BURPs		19:13
kerio	?	19:13
topro_	kerio: is there a howto, ressource, doc?	19:13
kerio	topro_: hm, they should be in the sdk already	19:14
kerio	cssu ships some bugfixes i think	19:14
* kerio laughs at version 1:1.0.2-maemo4+0m5+0cssu1+thumb2		19:14
kerio	what the fucking fuck, really	19:14
topro_	kerio: my maemo on n900 has it, but its missing from qt sdk maemo target toolchain	19:15
kerio	topro_: i can't grasp how scratchbox works, sorry	19:15
kerio	:s	19:15
*** guampa has quit IRC		19:15
kerio	ask freemangordon maybe	19:15
DocScrutinizer05	what's "qt sdk maemo target toolchain"?	19:15
DocScrutinizer05	we usually have scratchbox (and some say madde too)	19:16
topro_	DocScrutinizer05: using qt sdk updater I installed maemo target.	19:16
DocScrutinizer05	not supposed to work afaik	19:16
topro_	its missing qtmobility, obviously. so i installed madde as stated somewher on tmo, which has no qtmobility either	19:17
DocScrutinizer05	madde isn't really well supported	19:17
DocScrutinizer05	it should be possible to install qtm in it though, I think	19:18
DocScrutinizer05	not that I really had a clue about madde	19:18
topro_	i found some references stating that the only way to build qtmobility applications for maemo atm is to use scratchbox, which doesn'T integrate into qt-creator	19:18
DocScrutinizer05	sounds about right	19:18
*** topro_ has quit IRC		19:19
*** tzafrir_n900 has joined #maemo		19:21
*** topro_ has joined #maemo		19:22
topro_	got disconnected	19:22
ccxCZ	uh, I installed xev, but I don't get any keyboard events at all	19:22
*** MetalGear has quit IRC		19:23
topro_	[18:18:25] <topro_> i found some references stating that the only way to build qtmobility applications for maemo atm is to use scratchbox, which doesn'T integrate into qt-creator	19:23
*** xes has quit IRC		19:23
topro_	so I'm looking for a way to bild qtmobility-aware apps using qt-sdk	19:24
topro_	s/bild/build/	19:24
infobot	topro_ meant: so I'm looking for a way to build qtmobility-aware apps using qt-sdk	19:24
topro_	it works when compiling for targets simulator and harmattan, but not for maemo	19:26
topro_	as maemo toolchain is missing qtmobility support	19:26
*** loganbr has joined #maemo		19:28
*** waelsy123 has quit IRC		19:36
*** pvanhoof has quit IRC		19:40
*** sardini has joined #maemo		19:42
*** emma has quit IRC		19:48
*** tzafrir_n900 has quit IRC		19:51
*** tzafrir_n900 has joined #maemo		19:54
*** povbot has joined #maemo		19:56
*** dos1 has quit IRC		19:59
*** dos11 has joined #maemo		19:59
*** valerius has quit IRC		20:02
*** konelix has quit IRC		20:10
tzafrir_n900	apt-get update fails due to expired keys. where can I find up-to-date ones	20:16
*** _rd has quit IRC		20:16
*** topro_ has quit IRC		20:17
*** FlameReaper has quit IRC		20:20
*** jrocha has joined #maemo		20:20
*** valerius has joined #maemo		20:26
sixwheeledbeast1	tzafrir_n900: expired keys should be a warning.	20:29
*** valerius has quit IRC		20:37
*** markinfo has joined #maemo		20:37
*** _rd has joined #maemo		20:41
*** emma has joined #maemo		20:41
*** tzafrir_n900 has quit IRC		20:42
*** valerius has joined #maemo		20:47
*** tzafrir_n900 has joined #maemo		20:49
*** jrocha has quit IRC		20:49
DocScrutinizer05	topro: sounds about right	20:53
DocScrutinizer05	topro: [2013-02-16 18:18:44] <DocScrutinizer05> sounds about right	20:53
DocScrutinizer05	~listkeys expired	20:55
infobot	Factoid search of 'expired' by key returned no results.	20:55
*** tzafrir_n900 has quit IRC		20:55
DocScrutinizer05	~#maemo expired-key is <reply>apt-get update soits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	20:56
infobot	DocScrutinizer05: okay	20:56
DocScrutinizer05	~#maemo ek is <reply>see expired-key	20:56
infobot	DocScrutinizer05: okay	20:56
DocScrutinizer05	~ek	20:57
infobot	apt-get update soits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	20:57
DocScrutinizer05	infobot: no, #maemo expired-key is <reply>apt-get update spits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	20:57
infobot	DocScrutinizer05: okay	20:57
DocScrutinizer05	~ek	20:57
infobot	apt-get update spits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	20:57
DocScrutinizer05	~listkeys gpg	20:58
infobot	Factoid search of 'gpg' by key (6): gpg sign multiple files ;; .gnugpg/options ;; gpg ;; agpgart ;; gpg upload ;; pgpgpg.	20:58
DocScrutinizer05	~gpg	20:58
infobot	rumour has it, gpg is the GNU Privacy Guard, a free public-key encryption program available from nonUS.debian.org . or http://www.gnupg.org, but apt-get install gnupg	20:58
DocScrutinizer05	[notice] lists.maemo.org rebooted	20:59
DocScrutinizer05	something seriously fishy with all that stuff on our xen-grid	21:00
Pali	DocScrutinizer05, now https://lists.maemo.org/ not working	21:12
Pali	Not Found The requested URL / was not found on this server.	21:12
DocScrutinizer05	Pali: sorry, CND. http://lists.maemo.org/pipermail/maemo-developers/2009-September/020830.html WFM	21:13
Pali	hm, that working	21:14
Pali	but this not, try $ wget lists.maemo.org	21:14
DocScrutinizer05	https://lists.maemo.org/pipermail/maemo-developers/2009-September/020830.html works too	21:14
DocScrutinizer05	after nodding off the borked-cert warning	21:14
DocScrutinizer05	sorry, I'm not interested in wget of this page	21:16
*** sobukus has joined #maemo		21:16
DocScrutinizer05	if you found a bug, please analyze and report to bmo	21:16
sobukus	Hi, folks. Dies someone have a hint on how to circumvent https://bugs.maemo.org/show_bug.cgi?id=11384 ?	21:17
povbot	Bug 11384: Diablo extras & extras devel badly signed with Fremantle repository key	21:17
sobukus	can I tell apt-get to just ignore the signature?	21:17
DocScrutinizer05	~ek	21:17
infobot	apt-get update spits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	21:17
sobukus	This is not it. It's about a bad signature on the Maemo Extras repo.	21:18
sobukus	The bug report describes my experience: Apt refuses to work with the extras repo because of bad sig.	21:19
DocScrutinizer05	please file a ticket	21:19
* sobukus notes that this is an N800 with diablo		21:19
DocScrutinizer05	ooh you did	21:19
DocScrutinizer05	ping merlin1991	21:19
sobukus	DocScrutinizer05: no, the ticked is already there since years:-/	21:19
DocScrutinizer05	o.O	21:19
merlin1991	pong DocScrutinizer05	21:20
DocScrutinizer05	^^^	21:20
merlin1991	yeah that bug is OOOLD	21:20
sobukus	I guess this issue should be easy to fix ... just change some script to create the correct sig?	21:20
DocScrutinizer05	you know anything about that?	21:20
DocScrutinizer05	merlin1991: impact, severity, fix?	21:21
DocScrutinizer05	merlin1991: workaround?	21:21
merlin1991	impact apt goes foobar	21:21
Pali	that bug was funny :D:D	21:21
merlin1991	workaround tell apt to ignore keys	21:22
merlin1991	fix fix the friggin repo script	21:22
Pali	fremantle cssu repo was signed by diablo key	21:22
sobukus	?	21:22
Pali	and diablo repo is signed with fremantle key :D	21:22
sobukus	I guess the real problem is not the signing key but having the wrong signature file.	21:22
sobukus	merlin1991: I tried apt-get --allow-unauthenticated update ... to no avail	21:23
sobukus	is there another way?	21:23
*** Ethernin has joined #maemo		21:27
*** dos11 has quit IRC		21:28
*** Ex-Opesa has quit IRC		21:32
*** Ex-Opesa has joined #maemo		21:33
kerio	sobukus: APT::Get::AllowUnauthenticated "yes";	21:33
kerio	in /etc/apt/apt.conf	21:33
kerio	merlin1991: just delete Release.gpg?	21:34
sobukus	kerio: that AllowUnauthenticated will only work after Release.gpg is deleted, right? because, as long as there is a sig, it is not unauthenticated and apt still complains (it looks like that to me)	21:38
DocScrutinizer05	sounds correct. I suggest renaming instead of deleting though	21:41
kerio	yes, yes	21:41
kerio	Release.gpg.bak	21:41
*** bef0rd has quit IRC		21:44
sardini	hello trusted.gpg ou trustdb.gpg	21:44
sardini	in /etc/apt/ ?	21:44
*** GuySoft has quit IRC		21:46
kerio	sobukus: if the bug comments are correct, you don't even need to explicitly ignore the authentication	21:47
kerio	because diablo repos aren't supposed to be signed anyway	21:47
sobukus	Well, this one is; but I admit I am confused. apt-get update is not very clear on the success of the operation	21:48
*** mvp_ has quit IRC		21:48
sobukus	but since I was able to install usbcontrol, I presume I have the extras repo working?	21:49
sobukus	What I really was after was either diffutils or an updated busybox; is that available in the maemo repos for diablo?	21:49
*** GuySoft has joined #maemo		21:50
*** MetalGearSolid_ has joined #maemo		21:53
*** MetalGearSolid_ has quit IRC		21:55
*** MetalGearSolid has quit IRC		21:55
* sobukus notes he didn't have extras-devel, only devel		21:57
sobukus	now I see busybox-power!	21:57
*** bef0rd has joined #maemo		21:57
*** nox- has joined #maemo		22:02
*** dos11 has joined #maemo		22:15
*** sq-one has quit IRC		22:16
*** sq-one has joined #maemo		22:16
*** sq-one has joined #maemo		22:21
*** valerius has quit IRC		22:23
DocScrutinizer05	~pkg	22:28
infobot	hmm... pkg is http://maemo.org/packages/	22:28
DocScrutinizer05	sardini: sorry?	22:29
sardini	I have same issue with maemo. on apt-ge update with skeiron.org repository gpg error and bzip2 error code 1	22:31
*** sq-one has quit IRC		22:34
*** sq-one has joined #maemo		22:34
*** Martix has quit IRC		22:37
DocScrutinizer05	~ek	22:37
infobot	apt-get update spits out Warnings about Nokia GPG key expired, it's not an Error that should break anything. Ignore it	22:37
sardini	yes I see you repeat 4x this, :) sorry	22:37
DocScrutinizer05	bzip2 error iirc is related to busybox-power, do you have this installed?	22:38
*** Martix has joined #maemo		22:38
sardini	how to know if busybox is installed?	22:38
DocScrutinizer05	if you installed busybox-power, you done it on purpose	22:38
sardini	busybox-power is already the newest version	22:39
DocScrutinizer05	dpkg -s busybox-power	22:42
DocScrutinizer05	if that shows you installed busybox-power, then I suppose your bzip2 problem is caused by that	22:42
kerio	sardini: which skeiron repository?	22:43
sardini	I must remove busybox ?	22:43
*** valerius has joined #maemo		22:43
DocScrutinizer05	no, it's about busybox-power, not about busybox	22:44
DocScrutinizer05	and I doubt removing busybox-power will help	22:44
kerio	busybox-power uninstalls very cleanly	22:44
DocScrutinizer05	not my call, I won't touch it ever	22:44
sardini	kerio, from first post here http://talk.maemo.org/showthread.php?s=dc0666b43c4824cb010f17e370f12d13&t=88707	22:45
kerio	sardini: yeah, but which one? extras? extras-devel?	22:45
sardini	kerio, all	22:46
DocScrutinizer05	sardini: you also installed CSSU?	22:46
sardini	docScrutinizer yes a long time ago	22:46
sardini	DocScrutinizer05, yes a long time ago	22:46
kerio	DocScrutinizer05: decompressiong Packages.bz2 worksforme	22:46
kerio	*decompressing	22:46
DocScrutinizer05	I think CSSU 'conflicts' with busybox-power	22:47
kerio	ah yes	22:47
kerio	sardini: open a terminal, apt-get install --reinstall busybox-power	22:47
kerio	if it was an old version of bb-p	22:47
sardini	how can I install package the not works ?	22:47
DocScrutinizer05	so maybe reinstalling busybox-power can fix stuff for you. But again, I'm not the best expert you can find, since I never touched bb-p	22:47
kerio	sardini: hmm	22:47
DocScrutinizer05	hah, kerio been faster than me	22:48
kerio	sardini: open a terminal	22:48
kerio	it's probably better if you do it on your computer, via ssh	22:48
kerio	do you have root access?	22:49
DocScrutinizer05	when he got bb-p? I'd hope so	22:49
*** Martix has quit IRC		22:49
sardini	yes sudo gainroot	22:49
DocScrutinizer05	use "root"	22:49
kerio	ok, as root:	22:49
DocScrutinizer05	not "sudo gainroot"	22:50
kerio	wget http://repository.maemo.org/extras-devel/pool/fremantle/free/b/busybox-power/busybox-power_1.20.2power5_armel.deb && dpkg -i busybox-power_1.20.2power5_armel.deb	22:50
*** Martix has joined #maemo		22:50
kerio	no, actually	22:51
*** Martix has quit IRC		22:51
DocScrutinizer05	hmm, well then you should do /cd /home/user/MyDocs before	22:51
kerio	much much easier	22:51
kerio	rm /bin/bzip2	22:51
kerio	and i mean much easier	22:51
kerio	:D	22:51
DocScrutinizer05	aha	22:51
DocScrutinizer05	o.O	22:51
DocScrutinizer05	hmm	22:51
kerio	sardini: rm /bin/bzip2, then apt-get update as root, then apt-get install --reinstall busybox-power	22:51
DocScrutinizer05	might work	22:51
sardini	ok I will tell you	22:52
kerio	DocScrutinizer05: my apt prefers gzip, it's lighter and not much bigger than bzip2	22:52
DocScrutinizer05	kerio: NEVER suggest to rm anything. EVER	22:52
*** Martix has joined #maemo		22:52
kerio	:(	22:52
DocScrutinizer05	use mv!	22:52
kerio	oh please	22:52
kerio	"ln -s /bin/busybox /bin/bzip2" vs "mv /bin/bzip2.bak /bin/bzip2"	22:53
kerio	it's just one character longer	22:53
sardini	all system use bzip2	22:54
kerio	sardini: and yet the stock fremantle works without it just fine	22:55
*** eichi has quit IRC		22:58
sardini	I have moved bzip2, apt-get update, sub-process gzip returned error code 1	22:59
DocScrutinizer05	kerio: and when his bin/bzip is not a symlink?	22:59
kerio	sardini: wtf	22:59
kerio	it's either skeiron that's busted, or...	22:59
DocScrutinizer05	no	22:59
kerio	DocScrutinizer05: gzip is from the stock busybox	23:00
kerio	DocScrutinizer05: if it wasn't a symlink to a busybox without bzip2, it would've worked	23:00
DocScrutinizer05	and probably as messed up by cssu install as bzip	23:00
kerio	DocScrutinizer05: nope	23:01
kerio	unless the cssu busybox doesn't ship gzip	23:01
DocScrutinizer05	honestly, don't invent new leete stuff on helping users	23:01
DocScrutinizer05	stay on tested path	23:02
kerio	sardini: what's the output of "busybox bzip2"?	23:02
sardini	I wait N900 is becom very slow	23:02
sardini	error with apt-get in a term, but the packagemanager say me now updates	23:03
sardini	oh even operamobile update :)	23:03
sardini	ok in fact I have just 'remove' bzip2	23:04
kerio	sardini: what does "busybox bzip2" say?	23:05
kerio	even as the normal user	23:05
sardini	community SSU PAckage 21.2011.38-1Tmaemo7.2 is the last	23:05
*** _rd has quit IRC		23:06
*** Martix has quit IRC		23:06
kerio	sardini: is it?	23:06
sardini	kerio sorry n900 update cssu I can't type anymore	23:07
kerio	i don't really get what's happening	23:07
sardini	but I see in the update Enhanced BusyBox shell	23:07
sardini	question is why now is ok in the PackageManager and no in the apt-get	23:08
*** Martix has joined #maemo		23:09
sardini	term reboot? I pray	23:09
sardini	*?	23:09
*** croppa has joined #maemo		23:10
*** waelsy123 has joined #maemo		23:11
*** Martix_ has joined #maemo		23:12
*** Martix has quit IRC		23:15
sardini	kerio, busybox byp2	23:21
sardini	applet not found	23:21
kerio	alright	23:21
kerio	you said that there's an update for Enhanced Busybox?	23:22
kerio	do that update too	23:22
sardini	vers 1.20.2power5	23:23
sardini	65Mo updates...	23:23
*** waelsy123 has quit IRC		23:23
sardini	kerio, DocScrutinizer05 thanks a lot :)	23:24
*** valerius has quit IRC		23:25
*** messerting has joined #maemo		23:25
*** _rd has joined #maemo		23:27
sardini	ah busybox-power update say to me that: /bin/busybox has been modified since installing poxer (invalid md5 checksum). Can be result busybox update from cssu	23:27
sardini	I say ok, yes do it my wonderfull n900	23:27
*** jrocha has joined #maemo		23:41
Estel_	WizardNumberNext, cheesus, 50ms interval for gathering battery data? for fucks sake!	23:45
Estel_	wanna have 10x less battery life?	23:45
SpeedEvil	haha	23:45
SpeedEvil	50ms may be interesting to do impedence spectrometry.	23:46
SpeedEvil	but basically insane	23:46
Estel_	WizardNumberNext, the funniest thing is that you don't need battery profiler in N900	23:46
Estel_	SpeedEvil, yea	23:46
Estel_	well, he wanted 1 ms before	23:46
Estel_	until got info about kernel timers being 50ms accurate	23:46
WizardNumberNext	Estel_ did you read all what I wrote? I need anything below 2.56s, but be as accurate as possible, as I do not want to miss any sample	23:46
Estel_	hehe, of course I've seen it	23:47
Estel_	now the funniest part:	23:47
SpeedEvil	WizardNumberNext: the BQ* chip samples current every 5 seconds	23:47
Estel_	why do you want to mimic what hardware bq27x00 is already doing?	23:47
WizardNumberNext	Estel_ and I knew about kernel timers as soon as it was written here - IU tend to read whole backlog	23:47
SpeedEvil	sampling faster will not result in more readings	23:47
Estel_	profiler from n95 is dumb-phone equivalent (less useful) of what we have in hardware	23:47
Estel_	+ what speedevil said	23:48
SpeedEvil	and the chip has a hardware integrator	23:48
Estel_	it's 5.12 second interval updated, 10% accurate	23:48
*** arcean__ has joined #maemo		23:48
Estel_	now, as ShadowJK said, polling it's data even every 5 seconds is decreasing idle time 5x	23:48
WizardNumberNext	Estel_ there is no damn way to mimic what bq27x00 is already doing without another bq27x00 alike device. All I want is to store those samples, but not missing single one and not storing doubles	23:48
Estel_	4mA idle usage is becoming 20+mA	23:49
Estel_	WizardNumberNext, hat do you want to achieve, really	23:49
kerio	sample every 5 seconds and skip duplicates?	23:49
kerio	idk	23:49
Estel_	bq27x00 stores mAh used like intelligent LiIon discharger, using variable current	23:49
Estel_	i just don't get what you want to achieve	23:50
*** arcean_ has quit IRC		23:50
Estel_	by some script storing samples. For what reason?	23:50
WizardNumberNext	Estel_: simple answer: LOG	23:50
*** LauRoman has joined #maemo		23:50
Estel_	for?	23:50
WizardNumberNext	battery log	23:50
Estel_	and what to do with it?	23:50
WizardNumberNext	read it	23:50
Estel_	log containing current stored every 5 second?	23:50
Estel_	for what fckn sake?	23:51
WizardNumberNext	yes	23:51
SpeedEvil	WizardNumberNext: there is no battery log.	23:51
SpeedEvil	you need to read the chip every 5 seconds	23:51
WizardNumberNext	I want to see what would be affecting and how my current drain	23:51
Estel_	use powertop	23:51
Estel_	;)	23:51
SpeedEvil	http://wiki.maemo.org/N900_Hardware_Charge_Meter	23:52
Estel_	you will get detailed log what is causing what and how often	23:52
SpeedEvil	http://wiki.maemo.org/N900_Hardware_Power_Consumption	23:52
Estel_	~powertop	23:52
infobot	i guess powertop is available from http://talk.maemo.org/showthread.php?p=491333, or http://www.2shared.com/file/11135428/bbe19274/powertop_111-1_armel.html	23:52
WizardNumberNext	SpeedEvil: that is what I want to do - read bq27x00 every 2.56, as there is some current value, which is updated every 2.56s	23:52
Estel_	infobot, no, powertop is available from repos	23:52
infobot	okay, Estel_	23:52
SpeedEvil	http://wiki.maemo.org/N900_software_power_management	23:52
SpeedEvil	WizardNumberNext: it's every 5.12s	23:52
*** Hurrian has joined #maemo		23:52
WizardNumberNext	I simply want full log	23:53
WizardNumberNext	which is average only then	23:53
WizardNumberNext	I want "instant" as well	23:53
Estel_	such log is utterly useless for anypurpose	23:53
SpeedEvil	read the above pages	23:53
Estel_	no instant in this device, period	23:53
Estel_	+ what SpeedEvil said	23:53
WizardNumberNext	Estel_ wrong it is usefull	23:53
SpeedEvil	(of which I'm the main author)	23:53
Estel_	WizardNumberNext, again, for your purpose, use powertop	23:54
WizardNumberNext	I want to know exact capacity of battery - you cannot know excact capacity of battery, if you measure capacity of battery in mAh	23:54
SpeedEvil	it's only useful if you correlate it with running processes, and track all enabled peripherals.	23:54
Estel_	there is no such thing as exact capacity of battery	23:54
Estel_	capacity is always relative	23:54
SpeedEvil	WizardNumberNext: the charge meter is designed for that	23:54
Estel_	capacity is relative to temperature, discharge current, etc.	23:55
*** dos11 has quit IRC		23:55
WizardNumberNext	I agree it would be relative capacity, as exact would be full-empty and empty Li-ion is dead Li-ion	23:55
Estel_	nope	23:55
WizardNumberNext	but I would get knowledge of mWh instead of useless mAh	23:55
SpeedEvil	WizardNumberNext: the charge meter measures and integrates accurately the entire consumption of the device over the whole charge cycle	23:56
Estel_	even from hyphotetical full-empty, it's still relative	23:56
WizardNumberNext	Estel_: then do it: discharge Li-ion below 2.7V and charge it back - get fire extiguisher first	23:56
SpeedEvil	to do accurate milliwatt-hour measurements, you need about ten voltage measurements over the cycle	23:56
SpeedEvil	that is all.	23:56
SpeedEvil	and even then it's wrong	23:56
SpeedEvil	as the battery can provide fewer miliwatts hours if used to provide a higher power	23:57
SpeedEvil	read above pages	23:57
Estel_	yea, in practice, what SpeedEvil said is what you need and what I tried to tell you from beginning - bq27x00 does just that, measure capacity via variable current logging, it's as accurate aas you will get	23:57
Estel_	and if we're talking about battery's theory, thefre is no "exact" capacity ever, even theoretical	23:57
Estel_	it's always relative	23:58
WizardNumberNext	SpeedEvil: I am aware of internal resistance ad all other parametres, which you even have no idea have some impact on battery as well	23:58
Estel_	to temperature, to discharge, to pressure, etc	23:58
SpeedEvil	to do better, you need to do impedence spectroscopy. and have a accurate battery model. I started on this, but got bored.	23:58
SpeedEvil	WizardNumberNext: I suspect I know way, way more about lithium ion charge cycles than you.	23:59
WizardNumberNext	battery is like every other piece of wire - have all possible electric parametres including resistance, impendance, being coil and so and on and on ....	23:59
Estel_	WizardNumberNext, what you want to achieve is some hoax about accurate battery measuring, using wong assumptions and wrong methods 0_o	23:59
kerio	>battery is like every other piece of wire	23:59
Estel_	battery is not even remotely related to wire, btw	23:59
kerio	it's a bunch of fucking chemicals	23:59
* Estel_ nods		23:59

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!