**** BEGIN LOGGING AT Sat Jul 30 02:59:56 2011 Jul 30 03:08:11 Ok, finally got it Jul 30 03:08:12 sheesh Jul 30 03:08:13 http://talk.maemo.org/showthread.php?p=1060924#post1060924 Jul 30 03:13:00 ok finally i flashed it Jul 30 06:00:41 i've updated my maemo to its latest version, here i see it has added sum new account types in its built-in conversation messenger. but seems yahoo doesn't work, how to fix it? Jul 30 07:30:26 hi how do i find out whats inside a mp4 container in commandline? Jul 30 07:38:53 mplayer with libcaca? Jul 30 07:39:12 but more seriously, _what_ do you want to find out? Jul 30 07:39:53 i got it ffmpeg -i Jul 30 08:28:12 what is the opposite of sudo gainroot Jul 30 08:28:53 i wanna put it but to normal mode not R&D mode Jul 30 08:44:56 how do i reverse sudo gainroot Jul 30 08:47:35 iluminator101: 'exit'? Jul 30 08:48:29 after did sudo gainroot my device is not charging...i plug it in nothing its not showing its charging Jul 30 08:50:37 opposite of gainroot != switching between normal mode and R&D　mode. To switch between normal mode and R&D mode you have to use the flasher utility Jul 30 08:51:27 and I personally don't think R&D　mode will prevent the device from charging. I have two N900, both of them can be charged even in R&D mode Jul 30 08:51:40 same here Jul 30 08:52:35 device is not charging nothing wrong with the plug Jul 30 08:52:58 I think there are at times the device does not accept charging. I　am not sure exactly why but sometimes unplugging and plugging the charger cable back in will get the device charging Jul 30 08:54:04 I've had to repeat plugging in/unplugging process at most 5 times. Though if after several times (beyond 5) the device will still not charge, I think it could be potential signs of dead USB port. Jul 30 08:54:46 I've found that BME is indeed flakey Jul 30 08:56:24 I thought there could be issues when one using pk, I'm using pk on both my N900 but mainly my older N900 is the one showing more prominent signs of requiring to plug USB cable in lots of times Jul 30 08:57:59 One thing you can do is, as root, "stop bme", wait 30 seconds, plug in the USB cable(LED should glow solid orange) "start bme Jul 30 08:58:14 If you did it right, the screen backlight will turn /off/, and it'll start charging Jul 30 08:59:48 talking about that robbiethe1st, how does one stop bme? killing it reboots the device Jul 30 09:00:00 Uh, "stop bme" as root Jul 30 09:00:33 * Sicelo is stupid :P Jul 30 09:00:52 thanks Jul 30 09:01:03 NP Jul 30 09:01:46 Also, depending, ShadowJK's charging script may also help Jul 30 09:02:29 yeah.. that's what i wanted it for, in order to do a proper learn cycle Jul 30 09:49:45 restarting bme with pk's battery kernel module loaded might fail or not work properly Jul 30 09:59:07 http://www.businesswire.com/news/home/20110728007223/en/Strategy-Analytics-Apple-Worlds-Number-Smartphone-Vendor go elop, go! :P Jul 30 10:00:01 from 38% to 15% on a growing market .. sounds like a plan Jul 30 10:02:05 Of course, Google's the real winner. Apple may have the highest per-vendor share, Google has the highest per-os share. Jul 30 10:02:41 and probably has a huge chunk of that 48+17% share Jul 30 10:03:12 what licensing does andoig cost? whats googles profit (we all know that M$ gets its share too) Jul 30 10:03:33 well i dont really care Jul 30 10:03:51 i only like to see some vendor grabbing the open platform model up Jul 30 10:04:12 apple wont, nokia denied ... Jul 30 10:04:21 leaves samsung, LG and few others Jul 30 10:10:07 Android costs nothing to license from Google. MS, however... Jul 30 10:10:18 $5/handset unless you've got patents to cross-license Jul 30 10:11:01 Personally, I want Kubuntu Mobile and or debian with hildon desktop... but I doubt I'll be seeing that anytime soon Jul 30 10:12:08 maemo/meego would already be good enough, sans the nokia fuckup Jul 30 10:12:31 Yea Jul 30 10:14:44 there's cordia which may potentially bring hildon-desktop onto ubuntu environments Jul 30 10:20:08 Yea, and I think it's a good idea Jul 30 10:31:33 actualy powerkernel is broken regarding all i2c until most recently Jul 30 10:33:03 esp bme may refuse to work at random moments Jul 30 10:34:46 most recently would mean pk48 has addressed the issue? Jul 30 10:36:18 yes Jul 30 10:36:24 should Jul 30 10:38:57 hmm interesting, I　guess I may have to give pk48 a go. The last I read there was issues with bluetooth handsfree which directly affects me as I regularly use my BH-905 at work. Jul 30 10:41:19 v47 (back to v45 it seems) is actually *broken* in a very basic way, and it's almost impossible to tell which are the symptoms of this bug Jul 30 10:41:39 *could* even kill your CPU Jul 30 10:42:58 (smartreflex aka PMU VCC voltage, this "undervolt" thing for CPU power, also gets controlled via I2C) Jul 30 10:45:03 and on I2C there can happen random things on PK45..47 - though it's extremely unlikely something as nasty as ramp up CPU voltage to BRZZZZ level will happen Jul 30 10:49:55 That would explain random corruption, though. Some voltage not getting set quite right, one of those 300 times per second it ramps between 0 and 250mhz while in "sleep" mode Jul 30 10:51:55 yeah, matan introduced a "patch" (read botch) that basically implements an "always return ok" policy into I2C's "man 2 flock()" Jul 30 10:52:24 I.E. you CAN NOT lock I2C devices any more Jul 30 10:52:55 So, why is it not in extras-devel yet? Jul 30 10:53:11 Pali should be yelled at until he's uploaded it properly Jul 30 10:53:21 now add to that the fact that e.g PMU needs unlock sequence to allow writing to the VDD voltage settings etc Jul 30 10:53:56 pali whined to get help why it doesn't show up at all on repos Jul 30 10:54:05 nobody cared so far Jul 30 10:54:43 it IS on extras-devel AIUI Jul 30 10:54:53 huh Jul 30 10:54:55 it just doesn't show up correctly Jul 30 10:55:10 o.O Jul 30 10:58:03 s/VDD/Vcore, Vcc/ Jul 30 11:00:42 the nasty thing about the brocken locking in I2C API is: it's harder to tell what may happen from that than it is to *proove* "bug free" for the whole kernel Jul 30 11:01:17 I.E. impossible Jul 30 11:02:11 it's just a sure bet to claim: all I2C related stuff *may* break as bad as possible Jul 30 11:02:23 with this patch of matan Jul 30 11:02:37 outch Jul 30 11:02:44 What did it solve, though? Jul 30 11:02:59 there's got to be some reason -- if misguided -- for that patch Jul 30 11:03:01 bq27200 kernel drive bme breakage Jul 30 11:04:02 bme opens bq27200 chip over I2c, the kernel driver bq27200.ko opens it exclusively though. Result: bme breaks Jul 30 11:11:05 Well... don't load the module, then? Or unload it before charging? Jul 30 11:11:41 Pretty much Jul 30 11:12:17 robbiethe1st: just AIUI it's loaded forced on boot and seems hard to unload or blacklist Jul 30 11:12:23 The right solution (other than ditching BME) would be to put smarts into the driver, to treat an access to I2C port x as an access to the driver Jul 30 11:12:34 This is messy though. Jul 30 11:14:14 the right solution short term is binning the bq27200.ko and this insane patch. Long term the sane solution got suggested by shadowJK iirc and is a LD_PRELOAD overriding the ioctl access to those I2C APIs Jul 30 11:14:51 of bme Jul 30 11:15:37 I'm back to my universal fopen() interceptor.so Jul 30 11:17:31 somebody also suggested there are ways to open I2C in a way that's not completely blocking (iirc it was speedevil) Jul 30 11:18:41 so bq27200 *could* do the "right thing" and lock the I2C resource only for milliseconds when a read to the bq27200 sysnodes of this driver actually requires reading a value from the chip Jul 30 11:18:52 That too, yes Jul 30 11:19:26 A thingy that could do arbitrary ioctl remaps would be _so_ cool. Jul 30 11:19:31 As well as file Jul 30 11:19:59 bme seems "smart" enough to survive occasional -EAGAIN or -EBUSY on read to the I2C Jul 30 11:20:30 and if it's not it simply segfaults and gets restarted - not a big thing either Jul 30 11:20:44 unless it does that all the time Jul 30 11:22:38 the bq27200.ko however allocates the I2C resource exclusively and permanently on module init Jul 30 11:23:33 aaaah wait - and bme also keeps a handle/control open but not an exclusive one, so it used to work as long as bme gets started before bq27200.ko module Jul 30 11:25:32 stop bme; start bme -> doom guaranteed, for pre-matan-I"C-breaking-patch Jul 30 11:26:05 now doom is permanently pending though harder to trigger in a reliable way Jul 30 11:26:45 also doom is pending in *all* I2C access now, until recent V48 PK Jul 30 11:28:19 V48 restores the pre-matan.patch behaviour and also should remove bq27200.ko from the list of modules that get autoloaded on boot Jul 30 11:28:54 I.E modprobe bq27200; stop bme; start bme; -->doom guaranteed Jul 30 11:31:21 no big thing as for now nobody should rely on bq27200 module anyway Jul 30 11:31:36 s/rely/depend/ Jul 30 11:31:36 DocScrutinizer meant: no big thing as for now nobody should depend on bq27200 module anyway Jul 30 11:33:09 a useless module anyway, as iirc it not even provides full access to the chip's wealth of data Jul 30 11:34:12 hmm Jul 30 11:34:31 i have a device with Processor : Feroceon 88FR131 rev 1 (v5l) Jul 30 11:34:39 and it's like 6x faster than n900 cpu Jul 30 11:34:57 by which metrics? Jul 30 11:35:19 just calculation Jul 30 11:35:23 running pi Jul 30 11:35:34 also Watts/drystone ? Jul 30 11:35:51 time that it takes to calculate pi to few spaces after . Jul 30 11:35:56 and: can it suspend? Jul 30 11:36:01 no Jul 30 11:36:10 and it's NAS not a phone Jul 30 11:36:17 tablet* Jul 30 11:37:20 hmm, I have access to a device that is ~300 times faster than N900 cpu on such tasks Jul 30 11:38:26 jacekowski: how about RE bme a bit, and patch in more sane voltage threshold? Jul 30 11:39:46 Indeed, I'd kill for a BME with a battery threshold of 3200mV Jul 30 11:39:50 (kill BME) Jul 30 11:40:06 3100 even Jul 30 11:41:34 but honestly, result of such effort should be a "tool" patchbmevoltagethreshold Jul 30 11:42:20 quite possibly a sed script tuned to patch binaries ;-) Jul 30 11:42:53 (the approach I failed on epically, when it was about patching the smslib Jul 30 11:42:55 ) Jul 30 11:44:01 it's daunting ther seems to be not ONE unix tool / set of tools capable of simple hexedit-alike patching in a scripted environment Jul 30 11:45:17 messybox doesn't exactly help on the issue either Jul 30 11:46:21 messybox is quite in line with harmattan aegis in this regard: allows only what some maintained figured the users may need Jul 30 11:46:25 bspatch is not what you're looking for? Jul 30 11:46:30 maintainer* Jul 30 11:46:46 Atarii: sounds good, never heard of it Jul 30 11:46:51 http://www.daemonology.net/bsdiff/ Jul 30 11:47:43 first random factoid: not installed on my pretty comprehensive desktop x86 linux developer system Jul 30 11:48:13 vim on a binary file Jul 30 11:48:25 Though that's not quite what you're looking for Jul 30 11:48:30 err, I bet that will work :-D Jul 30 11:48:39 It pops up a hex screen Jul 30 11:49:02 well, I bet vi *could* eat scripts as well Jul 30 11:49:19 MEMMOVE=0x$(objdump -S -j .plt $INPUT | grep memmove |awk '{ print $1 }') Jul 30 11:49:20 [ $MEMMOVE = "0x" ] && echo "Can't find memmove call in $INPUT PLT" Jul 30 11:49:22 something like that? Jul 30 11:49:34 I have to admit I managed since ~20 yeras to not touch vi(m) Jul 30 11:49:37 there is also xdelta for binary patching Jul 30 11:50:28 ruskie: yeah, just check messybox awk, to feel with my frustration Jul 30 11:50:29 and then using printf on stuff Jul 30 11:54:57 I had a solution along od and sed, but it also had "issues" - one of them the nasty habit of od to spit out a trailing line with addr but no value Jul 30 11:57:24 ruskie: s/grep memmove|awk/awk '/memmove/ Jul 30 12:48:09 can i make the packaging call more verbose? im using the QtSDK to deploy via mad, but i get: mad dpkg-buildpackage -nc -uc -us' failed.Exit code: 2 Jul 30 12:51:06 DocScrutinizer: o you happen to recall teh samplerate of the voltmeter on the bq27200 Jul 30 12:51:15 ah - nvm - I should look itup Jul 30 12:51:47 ill try to run mad from the console, not within qtsdk Jul 30 12:56:38 oh - 2.5s Jul 30 12:59:17 SpeedEvil: some ticks faster than the 5.2s Jul 30 12:59:27 yeah. Jul 30 12:59:41 I'm wondering about a hacked up battery analyser to properly spit out impedence Jul 30 12:59:54 hmm Jul 30 13:00:05 impedence spectroscopy Jul 30 13:00:06 *should* work Jul 30 13:00:17 I'll knock out a proto tonight. Jul 30 13:00:22 cool Jul 30 13:00:36 ping me if you need sparring, review, help, whatever Jul 30 13:01:55 SpeedEvil: OT but nice: http://r0ket.badge.events.ccc.de/ Jul 30 13:02:25 looking forward to wear one of those in 10 days Jul 30 14:37:23 hello all Jul 30 14:37:41 How do I edit the .mo files in /usr/share/locale/? Jul 30 14:40:35 you can't Jul 30 14:40:46 hatake_kakashi: surely there's some way to manipulate it? Jul 30 14:53:11 seems man 1 localedef is providing some info. Anyway you need the input files I'd think Jul 30 14:54:54 I was about to reply but he left. .mo files are precompiled translation files. He'll need to obtain the .po files which allows editing with some translation software Jul 30 14:56:25 also note /usr/lib/locale/locale-archive Jul 30 14:56:40 a PITA Jul 30 15:03:37 23MB of cruft, on rootfs where space is precious Jul 30 15:04:57 and no way t get rid of (sorry) abominations like zh_CN Jul 30 15:05:28 I suppose that could be optified (and maybe later on use binding mount). Much like /var/lib/dpkg where if one has extras-devel enabled or whatever its causing it, it chews up more rootfs space Jul 30 15:07:28 well, looking at the FQN starting with /usr/, and at Jul 30 15:07:32 ~optification Jul 30 15:07:33 from memory, optification is a inventive duct tape workaround to reclaim space in fs root, done due to the fact the systeminit *and* partitioning is FUBAR, http://wiki.maemo.org/Documentation/Maemo_5_Developer_Guide/Packaging,_Deploying_and_Distributing/Installing_under_opt_and_MyDocs, or ""OMG - I wish somebody had looked into FHS and moved /usr to eMMC"", http://www.pathname.com/fhs/pub/fhs-2.3.html#PURPOSE2 bullet1,2 and fhs-2.3.html#PURPOSE16 dot3 Jul 30 15:07:51 I'd guess yes it *should* be optifyable Jul 30 15:08:37 as I don't give a dang about early boot having access to a full set of translations of every possible string to every supported locale Jul 30 15:08:59 it maybe tricky though :) I guess if its not required around boot time it may not cause bootloops and may also not affect with the various UI Jul 30 15:09:09 but, you know, that's maemo and Nokia Jul 30 15:09:12 hmm or maybe the rest could be optified Jul 30 15:12:37 I bet there's only PIN and lockcode entering dialog, and firstboot wizard for setting date/time/kbd/location that's affected by missing /usr/lib/locale/locale-archive Jul 30 15:14:54 hopefully the worst case scenario with optifying it would be that one gets stab_me_*, push_this_*, etc lookalike UI　:) Jul 30 15:15:07 s/the/in the/ Jul 30 15:15:08 hatake_kakashi meant: hopefully in the worst case scenario with optifying it would be that one gets stab_me_*, push_this_*, etc lookalike UI　:) Jul 30 15:18:41 wazd, what's the licence of http://zhil.in/terminal.svg? Jul 30 15:47:43 hi all Jul 30 16:03:18 hahaha, I created /etc/rc.local, within it I　added mount /opt/var/lib/dpkg /var/lib/dpkg -o bind. Upon reboot I get the fancy looking dialog as I mentioned before :D Jul 30 16:26:30 damn.. lol.. kept updating the date in the changelog but forgot the day of week.. apparently the autobuilder then simply ignores the date and takes the date of the next upcoming, matching day of week xD Jul 30 16:58:49 who provided the original DONT_PANIC bootvideo? I'd like to give proper credit on http://maemo.cloud-7.de/maemo5/bootvideo-DONTPANIC/ Jul 30 16:59:19 is there a way to just disable mr hands? Jul 30 16:59:51 or is a 1-frame video the best way to do that? Jul 30 17:02:00 kerio: see note in http://wiki.maemo.org/User:Joerg_rw/tools Jul 30 17:02:38 just one line :-) :: /etc/X11/Xsession.d/10hildon_welcome = shaking-hands-video Jul 30 17:02:50 how to mount root partition of n900 on my pc? Jul 30 17:03:48 TimmyT, use sshfs Jul 30 17:08:08 mgedmin: through wifi? Jul 30 17:08:16 sure, why not? Jul 30 17:08:32 (you could also use it over usbnet, if you want greater speed) Jul 30 17:08:56 mgedmin: that's why i need to do this through usb Jul 30 17:09:10 mgedmin: it is a little bit slow Jul 30 17:15:48 is there any music player with graphic equalizer? Jul 30 17:17:18 jacekowski, what MHz does your Feroceon run at? Jul 30 17:17:46 * ShadowJK has a 1200MHz sheevaplug with "Feroceon 88FR131 rev 1 (v5l)" Jul 30 17:37:24 TimmyT: use compression with sshfs, it will be quicker for most files in / Jul 30 17:39:18 lol, I managed to make N900 play BOTH the DONT_PANIC and the Hands video X-P Jul 30 17:42:02 ShadowJK: 1200 Jul 30 17:43:08 kerio: refresh my tools page Jul 30 17:50:16 TimmyT, I think rockbox has graphic equaliser Jul 30 18:09:59 btw, encoding a ~60s wav to mp3 with lame: 1200MHz Feroceon armv5: 2m13s, N900 omap3 @ 600MHz: 1m28s Jul 30 18:10:02 :-) Jul 30 18:11:27 ShadowJK: how come? Jul 30 18:11:34 how come what Jul 30 18:11:45 ShadowJK: why is the omap3 so much faster? Jul 30 18:12:25 Neon, and does more work per clock overall Jul 30 18:16:51 how get kernel of the maemo ? Jul 30 18:17:08 compiled or not? Jul 30 18:17:50 jacekowski: not compiled Jul 30 18:18:59 apt-get source linux-kernel Jul 30 18:19:03 or something like that Jul 30 18:19:13 i don't remember package name Jul 30 18:19:35 im not use n900. Jul 30 18:19:45 in scratchbox Jul 30 18:20:32 http://repository.maemo.org/pool/fremantle/free/k/kernel/ Jul 30 18:20:54 jacekowski: tanx Jul 30 18:23:08 fvahid, looking for anything in particular? Jul 30 18:23:45 ShadowJK: yes Jul 30 18:27:15 fvahid: ok im back Jul 30 18:28:10 fvahid: i think u are better to buy a n900, rather than finding a girlfriend, kuz whoever she be, u can't install irssi on her Jul 30 18:28:13 :p Jul 30 18:30:34 TimmyT: http://maemo.org/downloads/product/PC/vpc-sdk/ Jul 30 18:31:17 fvahid: maskhare :D. akhe morche chie ke kale pachash chi bash Jul 30 18:31:18 e Jul 30 18:34:25 when i try to install sum apps like smplayer or extra decoders or nano-opt , it says that sum dependencies are missed and are not available in the repository? what is this? :D how to fix it? is there any hidden key to install apps from maemo's repository? Jul 30 18:35:49 there are like 3 repositories on maemo.org Jul 30 18:35:54 + loads of private repositiories Jul 30 18:38:07 jacekowski: i only have extra fremantle-1.3, wat are other things? Jul 30 18:38:47 extras testing Jul 30 18:38:49 and devel Jul 30 18:39:58 jacekowski: so are u saying those dependencies are available in testing and devel repos? Jul 30 18:43:41 jacekowski: i promise it is my latest silly question im asking u Jul 30 18:46:49 some are Jul 30 19:06:15 DONT_PANIC! :-D http://talk.maemo.org/showthread.php?p=1061232 Jul 30 19:08:43 I thought there was some app in the repos for managing bootmenus Jul 30 19:10:25 anyone here knows anything about the offline caching patch for modest in CSSU Jul 30 19:10:26 ? Jul 30 19:10:51 ruskie: err, managing bootmenus? Jul 30 19:12:09 erm bootvids even Jul 30 19:12:30 im asking cos i just saw this http://forums.internettablettalk.com/showthread.php?s=35a18607e31620b9cfc1477f3001b6bb&t=55771&page=4 Jul 30 19:12:53 DocScrutinizer, btw wrt to that cron job... you could just have fcron on the n900 and run the job that way? Jul 30 19:15:30 which cron job? Jul 30 19:17:16 the backup-cronjob that runs on my PC? no, I can't, as that would mean N900 knows how to log in to my home IT, without me entering the password. And that's sth I definitely never will do Jul 30 19:17:49 ahh Jul 30 19:19:15 so... cssu has apparently had a modest patch added recently. but my apt says my modest is the latest version. what do i do to get this patched version? Jul 30 19:19:33 install CSSU? Jul 30 19:19:58 i have it already...? Jul 30 19:20:00 or simply get and dpkg -i the .deb Jul 30 19:21:12 yeah but the .deb by the developer is not the same as the .deb for cssu, but i cant find anything about modest at the cssu website... Jul 30 19:21:15 HAM will automatically pull updates from CSSU repo, so check if there are updates available in HAM Jul 30 19:21:22 ahhh Jul 30 19:21:25 thanks! Jul 30 19:21:52 DocScrutinizer: why the don't panic video? is it some sort of meme? Jul 30 19:23:12 DocScrutinizer51: sorry, ive already checked HAM for updates Jul 30 19:23:38 ~42 Jul 30 19:23:39 extra, extra, read all about it, 42 is the answer to life the universe and everything, see also http://en.wikipedia.org/wiki/the_answer_to_life,_the_universe,_and_everything Jul 30 19:24:58 http://en.wikipedia.org/wiki/The_answer_to_life,_the_universe,_and_everything#Don.27t_Panic Jul 30 19:26:25 >>DON'T PANIC (always upper-case) is a phrase written on the cover of The Hitchhiker's Guide to the Galaxy.[31] The novel explains that this was partly because the device "looked insanely complicated" to operate, and partly to keep intergalactic travelers from panicking.[32] It is said that despite its many glaring (and occasionally fatal) inaccuracies, the Hitchhiker's Guide to the Galaxy itself has outsold the Encyclopedia Galactica Jul 30 19:26:27 because it is slightly cheaper, and because it has the words "DON'T PANIC" in large, friendly letters on the cover<< Jul 30 19:26:33 yeah I read that Jul 30 19:26:39 should i do some kind of new install of cssu? Jul 30 19:28:37 is there actually a way to rotate the screen to unlock the phone (lock code screen) Jul 30 19:28:41 ask MohammadAG Jul 30 19:29:06 trying and typing that one 1 handed is difficult when holding it upright :/ Jul 30 19:29:47 hm reading the cssu changelog the patch doesnt seem to have been implemented at all. so problem solved Jul 30 19:36:10 hello Jul 30 19:36:26 what is best game for n900? Jul 30 19:38:30 the one you're playing the most Jul 30 19:41:01 the one? Jul 30 19:43:48 ? Jul 30 19:48:18 i ask for best game for n900 Jul 30 19:49:52 it is unethack Jul 30 19:50:02 do i need to uninstall CSSU before i can switch to the testing version? Jul 30 19:50:19 HAM says 'CSSU is already installed' Jul 30 19:50:26 ludens: there is only the tsting version Jul 30 19:50:40 orly Jul 30 19:50:54 :) Jul 30 19:51:13 rly Jul 30 19:52:20 well then i am very congfused Jul 30 19:53:40 ill ask in #maemo-ssu Jul 30 20:00:13 ruskie: I actually don't care about any app to manage bootvideos, it's not like I change them 5 times a day. I'm not even booting 5 times a month. And for sure I wouldn't want to explain how to use that app to get "my" DONT_PANIC movie installed and running Jul 30 20:03:14 don't boot videos just increase boot time? Jul 30 20:03:41 Not by the length of the video Jul 30 20:03:53 Simply as other stuff happens in the background. Jul 30 20:04:13 It is additional CPU though Jul 30 20:04:25 a bit Jul 30 20:04:41 as that's mostly DSP/GFX Jul 30 20:04:47 Very little in the case if the DSP is used. Jul 30 20:36:23 what the tmo mods should do is some parenting for all the chaotic users. every app should be in the bug tracker for example. then, if someone 'complains about something not working' the post should be deleted and a note dropped that this belongs on the bug tracker. feature requests, too. Then, everytime someone writes a guide and posts it on the forum, it should be moved to a new wiki page and deleted. Jul 30 20:40:17 Actually most of them are too incoherent for bugtrackers Jul 30 21:08:02 SpeedEvil, how goes your impedance/ir meter? Jul 30 21:08:50 ShadowJK: I ended up working on flooring in the spare room. Just got a sketch done. Hopefully tomorrow. Jul 30 21:09:39 a sketch? Jul 30 21:10:26 Outline of code, on a bit of A4 Jul 30 21:16:08 DocScrutinizer, MohammadAG is still having problems opening sites Jul 30 21:16:19 and my buffer goes back to ask MohammadAG Jul 30 21:21:27 MohammadAG: solved, was sth about a patch to modest in cssu that turned out to be never applied Jul 30 21:21:45 so the whole story was a non-issue Jul 30 21:22:13 MohammadAG: so let me open DHL for you once more :-D Jul 30 21:22:56 12 Clearance delay TEL AVIV - ISRAEL 21:46 Jul 30 21:22:59 :-S Jul 30 21:25:00 DocScrutinizer, TLV, delayed there Jul 30 21:25:36 probably cause it's a saturday Jul 30 21:36:53 MohammadAG:Hi! Did you notice the bug in phone-control? Jul 30 21:38:58 I was in TLV today Jul 30 21:40:42 DocScrutinizer: do you have a fitting sound for the don't panic vid? :D Jul 30 21:40:46 mase76, "the" bug? Jul 30 21:40:46 what bug Jul 30 21:41:28 I posted last week. The ifconfig part is wrong. And unloading and loading the wifi driver causes a mac change. Jul 30 21:41:57 merlin1991: nope Jul 30 21:43:43 MohammadAG: read #maemo-ssu Jul 30 21:47:47 merlin1991, DocScrutinizer, MohammadAG is still having problems opening sites Jul 30 21:58:13 wtf? Jul 30 21:58:18 opening sites Jul 30 21:58:25 what does that have todo with channels? Jul 30 22:04:04 merlin1991: I gather he meant he's not able to open chanlog Jul 30 22:05:36 merlin1991: MohammadAG uses a bouncer, and afaik a history buffer that's defeating the purpose of keeping backscroll in that bouncer and get it via playback Jul 30 22:06:14 that's so short so it's defeating... Jul 30 22:36:28 MohammadAG: wtf is wrong with your internet? Jul 30 22:43:14 Argh. Jul 30 22:43:24 * SpeedEvil manages to confuse himself majorly. Jul 30 22:43:35 o.O Jul 30 22:43:45 Getting accurate measurements of current consumption is hard when you've got the USB lead plugged in. Jul 30 22:44:00 yeah Jul 30 22:49:28 Hi, there, I found this game http://paxbritannica.henk.ca/ was written purely in Lua. But my limited knowledge cannot allow me to run it on N900. Anybody can test it out? Jul 30 23:46:18 DocScrutinizer ShadowJK - 0th cut - http://www.mauve.plus.com/t/ohm.sh Jul 30 23:46:52 I get 450mR on a old cell, and 220 on one I got on ebay a couple of days ago. Both at 3.7V Jul 30 23:46:58 Charge state is probably important Jul 30 23:47:44 Synchronizing? Jul 30 23:48:00 v1.1 uploaded - minor bug. Jul 30 23:48:03 Synchronising? Jul 30 23:48:35 Yeah elaborate a bit on that Jul 30 23:49:05 The current measurement updates every 5.12s. You can't change that. Jul 30 23:49:38 So polling more often than 2* that risks non-isolated measurements Jul 30 23:50:17 One of the biggest annoyances I encountered, was that voltage is sampled by bq every second (?), while current is sampled every 5s, and often times I'd see high current use for a second or two, during that 5s period, so unless I oversample voltage to not miss anything, it becomes a bit unreliable to load variations Jul 30 23:50:27 voltage every 2.5s Jul 30 23:50:30 I think Jul 30 23:50:40 And yes Jul 30 23:50:50 This assumes idle, with screen off for best use Jul 30 23:50:52 (overpoll and discard non-stable readings) Jul 30 23:51:06 * SpeedEvil checks if it actually matters Jul 30 23:51:31 I only tried to do this on paper, and then I did it with a .8A drain and compared to .8A charge :) Jul 30 23:54:49 when I did it on "paper" it was more of analyzing large amounts of logs :/ Jul 30 23:54:54 but this is cool Jul 30 23:55:03 Seemingly 10% out of the idle figure when on and doing stuff Jul 30 23:55:10 for the average Jul 30 23:55:28 Hmm - my newest battery looks like 220mR Jul 30 23:55:33 why am I getting negative mR Jul 30 23:55:44 It's meant to do that! Jul 30 23:55:50 It means it's discharging Jul 30 23:56:10 (or I can't be bothered to read the sign bit) Jul 30 23:57:12 3652mV starting, 271, 259, 299, 385, 209 301, 233, 234, 359, 306: 296 Jul 30 23:57:24 (10% SOC) Jul 30 23:57:32 Fairly old? Jul 30 23:57:48 155 cycles Jul 30 23:57:55 1234mAh Jul 30 23:58:24 let me find my "fairly old" battery.. Jul 31 00:01:23 Interesting - it gives fairly consistent results when plugged in Jul 31 00:01:37 (not if charged of course) Jul 31 00:02:22 plugging it in prevents cpu from sleep Jul 31 00:02:52 But it can still measure delta-I Jul 31 00:04:33 it's a bit more complicated, because there's the max current limit from powersupply, and the phantom resistance between bq24150 and rest of system Jul 31 00:04:54 I know Jul 31 00:05:06 It's in the 'sort of works' category Jul 31 00:05:17 * ShadowJK can get somewhat accurate estimates of system load while charging by the voltage drop across that phantom resistance Jul 31 00:05:18 As long as it's far from v_taper Jul 31 00:11:15 751 for the fairly worn :) Jul 31 00:11:32 Wow. Jul 31 00:15:10 (I calculated 1000 for that one with 900mA charge/discharge difference Jul 31 00:15:24 +900 / -900, so 1.8A delta Jul 31 00:15:29 approx Jul 31 00:15:59 ionic diffusion, and all that voodoo, I guess. Jul 31 00:20:52 Mugen 2.4Ah, among the first produced, ~100 cycles before I stopped using it: 259 Jul 31 00:23:46 Seems sane Jul 31 00:24:21 The newest one I have is from ebay - but I suspect it's a knock-off. 250 Jul 31 00:30:56 Eastmaze, ~70 cycles: 220 Jul 31 00:31:50 now if bme was sane, it probably should adjust shutdown threshold by a function(AI) to account for the voltage drop in battery and parasitic R of circuit Jul 31 00:32:10 or adjust battery meter Jul 31 00:32:38 Indeed. Jul 31 00:33:19 The one I just got came flat. Jul 31 00:33:31 It may also havebeen in storage ages I guess. Jul 31 00:34:18 how flat? Jul 31 00:34:21 0v Jul 31 00:34:29 send back Jul 31 00:34:30 too flat Jul 31 00:34:32 yeah Jul 31 00:34:38 Not actually 0v Jul 31 00:34:47 actually - no - I think I measured it Jul 31 00:34:49 250 seems good though Jul 31 00:34:52 it was at 2.8 or so Jul 31 00:34:58