**** BEGIN LOGGING AT Mon Oct 19 02:59:59 2009 Oct 19 06:33:48 morning Oct 19 09:36:58 just wonder what exactly will the "debugger board" for the openmoko phone allow me to do or not to do ( yeah i know i most likely don't need it if i'm asking this but still ) Oct 19 09:39:28 the debug board page on the wiki should explain pretty well. does it miss something? Oct 19 09:40:09 i spotted it in the store let me check the wiki :) Oct 19 09:40:10 ty Oct 19 09:46:29 does this need updating, got it from wiki but doesnt work "mdbus -s org.freesmartphone.ogsmd /org/freesmartphone/GSM/Device org.freesmartphone.GSM.Device.GetInfo" Oct 19 09:46:37 supposed to show gsm firmware Oct 19 09:56:03 Psi: doesn't it work? Oct 19 09:56:56 im just rebooting the neo, i suspect something may have been screwed with when i was working on something else a few hours ago Oct 19 09:58:16 yep, its working now Oct 19 09:58:39 i ran that script to test for #1024 it must have left frameworkd in a weird state Oct 19 10:00:15 PaulFertser: any idea what sort of % current reduction deep sleep gives? Oct 19 10:00:21 Psi: you shouldn't run the script. If there's recamping, frameworkd will leave a trace in its log for sure. Oct 19 10:00:35 Psi: on gta01? Oct 19 10:00:37 ya Oct 19 10:01:02 Psi: well, i measured ~7mA in suspend without gsm iirc, does it correspond with your results? Oct 19 10:01:43 yeah, thats much less than what i got with deep sleep disabled Oct 19 10:01:57 just wondering will qemu run on the openmoko phone itself ( yes the goal is to run x86 stuff ) ? Oct 19 10:01:58 from memory it was around 30mA or something Oct 19 10:02:01 Psi: what about with gsm disabled? 7mA is with everything disabled. Oct 19 10:02:10 i get 3 mA if gsm is disable according to current_now Oct 19 10:02:17 lindi-: gta01? Oct 19 10:02:20 gta02 Oct 19 10:03:40 PaulFertser: i cant remember, it was a while ago, but i never got anywhere near 7mA Oct 19 10:03:41 Psi: i guess with deep sleep you'll have ~13mA with gsm on and ~21 without deep sleep. Oct 19 10:03:44 Superevildeath: its probably technically possible, but i wouldn't Oct 19 10:04:02 Psi: probably you didn't disable the amplifier which for some unknown reason wasn't turned off automaticalyl. Oct 19 10:04:14 yeah, i heard about that Oct 19 10:04:40 does that issue still exist or is it fixed in the unstable image now? Oct 19 10:05:07 Psi: not fixed, no reason to care about it until .31 is ready. Oct 19 10:07:19 fair enough Oct 19 10:08:36 Psi: is it hard to source a suitable cap for 1024 fix there? Oct 19 10:08:55 where? Oct 19 10:09:54 i dont think i have the issue, so haven't looking into soldering a cap in Oct 19 10:12:10 Psi: you'll never know if you have this 1024 issue or not. Esp not until you set calypso-deep-sleep=always, and let your Neo cool down to lower ambient temp Oct 19 10:12:32 ive set it to always Oct 19 10:12:37 thats why i was testing Oct 19 10:12:44 didnt know about the temp thing tho Oct 19 10:12:56 Psi: even then recamping might go unnoticed Oct 19 10:13:27 * Defiant- will never notice that bug Oct 19 10:14:02 the only way to detect is you're not reachable for inbound calls, your device is sending repeatedly frequently, and your bat drains really fast Oct 19 10:14:38 (if device in suspend) Oct 19 10:15:17 if device not in suspend, the "auto-heating" probably stops any 1024 Oct 19 10:15:24 Psi: you'll see recamping indication in the framework log Oct 19 10:15:32 i have the log open now Oct 19 10:15:50 does it use the word recamping? Oct 19 10:15:52 Psi: grep for -i recamping Oct 19 10:15:53 PaulFertser: no log during suspend Oct 19 10:16:04 DocScrutinizer-8: fo sure Oct 19 10:16:43 hm.. it appears i do have the bug to some degree Oct 19 10:16:53 i have 1 recamp entry in the log Oct 19 10:17:18 a good idea is to get one of those "ring-extenders" that start yelling when detecting any GSM TX in vicinity Oct 19 10:17:46 yeah, i think i have one of those led antenna things somewhere Oct 19 10:18:03 might work if i put it up close to the ant Oct 19 10:18:20 or i could just put the phone on top of a amp :P Oct 19 10:18:35 Psi: if it starts going mad - like flashing every 30sec - you know the 1024 hit you Oct 19 10:20:02 will the tx power be the same as a phone call to the neo Oct 19 10:20:12 yep Oct 19 10:20:20 i put it on top of my amp and called it to get the volume right Oct 19 10:20:30 now i have it all set so i can hear if i have gsm act Oct 19 10:20:35 probably even higher, as registering works with max TX power iirc Oct 19 10:21:29 now you only need to set aircon to 8C ;-P Oct 19 10:21:42 nothing so far Oct 19 10:22:06 Psi: this will come at random Oct 19 10:22:07 phone is in suspend and amp is just normal hiss Oct 19 10:23:00 you need to observe for days or weeks. Told you you'll never know you are *not* affected by 1024 in the future Oct 19 10:23:06 sounds like it might be easier just to solder the cap in Oct 19 10:23:21 yep. you bet it is Oct 19 10:23:43 has the cap fix be proven to work? Oct 19 10:23:47 ive not really looked into it Oct 19 10:23:47 or stay with deepsleep=2 Oct 19 10:23:52 be=been Oct 19 10:24:05 yep, is certified to work Oct 19 10:24:14 kinda need the extra power saving on gta01 Oct 19 10:28:07 what size cap is that, (the existing one) 0603? Oct 19 10:28:43 hm. looks bigger Oct 19 10:29:40 Psi: you might want to check BOM in my people dir somewhere Oct 19 10:38:25 found it, 0805 Oct 19 10:38:49 nice and big, should be pretty easy to fix :) Oct 19 10:39:13 maybe i'll make a video of the fix and put it on youtube Oct 19 10:39:37 good Oct 19 10:41:12 though I'd suggest to add a 10uF instead of replacing the original C. Seems easier Oct 19 10:42:49 yeah Oct 19 10:46:56 min qty 25, heh Oct 19 10:47:12 not that i was expecting being able to buy 1 Oct 19 10:50:25 Psi: beware it was reported the cap is extremely hard to desolder on gta01 Oct 19 10:50:26 you might be able to find qty 1 at local electronic or repair shop Oct 19 10:50:57 PaulFertser: wont remove it, as DocScrutinizer-8 said, gunna be easier to add one Oct 19 10:51:21 Psi: you can place some insulator and add an additional cap soldering it to the left side of the original and to the can holder. Oct 19 10:51:23 DocScrutinizer-8: probably, but id rather just get 25 Oct 19 10:51:33 yeah, always listen to your Doc ;-) Oct 19 10:51:41 dont have any 10uf's in my collection of smt caps Oct 19 10:51:45 DocScrutinizer-8: on 2 gta02s i reworked i dismounted the original Oct 19 10:51:50 just got 1 and .1 Oct 19 10:52:45 this looks like what ill probably do http://wiki.openmoko.org/images/1/13/1024fix.jpg Oct 19 10:53:30 hard part is going to be getting some flux under the shield so the solder will stick Oct 19 10:54:51 yay, just found i do have some 10uf Oct 19 10:55:17 Psi: (flux) don't you have liquid flux? Oct 19 10:55:30 i have a pen i normally use when soldering to metal like that Oct 19 10:56:41 use a small ball of floss and soak it with flux. Then use a pair of tweezers to apply Oct 19 10:57:00 that sounds like a good idea Oct 19 10:58:18 i do have some liquid flux but its so damn strong i try not to use it as it tends to get all over me and ruin my cloths :P Oct 19 10:58:35 pulls the dyes out Oct 19 10:59:03 you can even cut and bend upper flat frame of can base, to form a convenient small contact to solder C to Oct 19 11:00:01 hm.. Oct 19 11:00:36 got some 10uf 16v x5r 1206 which should do the trick :) Oct 19 11:00:37 or siply solde a short wire on top of the upper flat area, and then push down so it's below plane of can lid Oct 19 11:01:13 yeah Oct 19 11:02:26 you should apply some (hot-)glue then to fixate the C Oct 19 11:02:57 got plenty of that Oct 19 11:06:49 Psi: I'd prefer a decent fotofootage over a video Oct 19 11:07:05 DocScrutinizer: heretic! Oct 19 11:07:13 Youtube will come and kill you. Oct 19 11:07:20 heh Oct 19 11:07:28 i can do both Oct 19 11:07:29 5fps 320*200 badly encoded noisy video is good enough for anyone? Oct 19 11:07:39 s/?/!/ Oct 19 11:07:39 SpeedEvil meant: 5fps 320*200 badly encoded noisy video is good enough for anyone! Oct 19 11:07:41 yeah, sure Oct 19 11:08:06 i can do nice clear photos with my EOS 40D Oct 19 11:08:09 esp when done without zoom and without macro Oct 19 11:08:27 Psi: that sounds way more useful Oct 19 11:08:39 easier to setup that video too Oct 19 11:08:43 that=than Oct 19 11:09:04 you can send to me in a PM and I'll upload to people.om, if you like Oct 19 11:09:10 cant do photos of my actually soldering it tho, as i dont have 4 arms Oct 19 11:09:16 *me Oct 19 11:09:43 video would be useful to people pondering how easy/hard it is to actually do the work Oct 19 11:10:10 * SpeedEvil kidnaps Psi, and implants an extra couple of arms. Oct 19 11:10:18 you think, by counting the swearwords during that video ;-P Oct 19 11:10:25 heh Oct 19 11:10:34 i dont suspect it will be that hard Oct 19 11:10:57 cant be harder than hand etching a board for 100TQFP and soldering it on Oct 19 11:11:54 hand-etching. Heh, you're the unknown one of X-MEN? Oct 19 11:12:07 I am lookin after another open source based phone instead of FR and have heard about the N900. Is that phone any good? Oct 19 11:12:22 beers: it is the last best hope for peace. Oct 19 11:12:30 N900 isn't open source Oct 19 11:12:32 beers: oh no that was the babylon stations. Oct 19 11:12:35 its not usable without prorpieatyr software Oct 19 11:12:38 n900 is largely open source. Oct 19 11:12:39 *proprietary Oct 19 11:12:48 There are currently large slices of the UI proprietory. Oct 19 11:12:57 yep Oct 19 11:12:58 however - from a users point of view - it's very opensource. Oct 19 11:13:00 the phone stack on the n900 is suppose to be fully open Oct 19 11:13:12 muhaha Oct 19 11:13:13 you can't chargea battery without proprietary software Oct 19 11:13:14 dns53, it's actually fully proprietary Oct 19 11:13:15 You can for example install gcc, and start developing apps on the phone. Oct 19 11:13:19 heh, well i mean home etching with the laser printer method and then hand etching in fecl3 Oct 19 11:13:24 dns53, though apparently they're working on a free replacement Oct 19 11:13:32 is there a phone with fewer problem than FR and more open source than N900? Oct 19 11:13:43 beers: yes Oct 19 11:13:47 mer does that work Oct 19 11:13:59 SpeedEvil: whitch one? Oct 19 11:14:00 not maemo/nokia though Oct 19 11:14:15 SpeedEvil: heh @ b5 comment Oct 19 11:14:18 mjr my mistake, i thought they released the full phone stack ages ago Oct 19 11:14:22 http://www.sparkfun.com/commerce/product_info.php?products_id=286 Oct 19 11:14:24 beers Oct 19 11:14:28 love how the later intro was "it failed" Oct 19 11:14:29 It is opensource. Oct 19 11:14:30 But... Oct 19 11:15:23 n900 is probably the least bad - by a long way - of 'current' phones. Oct 19 11:15:23 for now not even the api of phonestack is public Oct 19 11:16:22 most android handsets would be as open as n900s, if not more Oct 19 11:16:30 from what i've seen Oct 19 11:17:19 aftewr what google did to some boy who was makde a image of adnorid I would not suport them Oct 19 11:17:22 Kamping_Kaiser: Only the 'dev mode' handsets Oct 19 11:17:42 Kamping_Kaiser: and by open, I mean the default openness when you buy it. Oct 19 11:18:06 http://en.wikipedia.org/wiki/Android_%28operating_system%29#Community-based_firmware Oct 19 11:18:31 read about TI going after some people who made a new OS for the TI calculators week ago or so Oct 19 11:19:21 SpeedEvil: that i can't comment on Oct 19 11:19:25 im guessing they were scared of their calculators being banned from school exams Oct 19 11:19:46 new OS for TI clculators? Damn, that's really needed Oct 19 11:20:45 Psi: lol Oct 19 11:21:11 yeah Oct 19 11:21:34 some of their calcs have an exam mode that locks out some features Oct 19 11:21:39 The emitters of the oscillator transistors are connected to ground, the collectors to 100R resistors to +, the bases to 10K resistors to +, and the capacitors are connected from the emitter of one transistor to the base of the other. Oct 19 11:21:39 oops Oct 19 11:22:00 oh, didn't know about that Oct 19 11:22:44 SpeedEvil: he astable multivibrator that's called here Oct 19 11:23:22 hm.. getting late here, better sleep. Well.. BSG then sleep :) Oct 19 11:23:24 and the C are from collector of one to base of other, not emitter Oct 19 11:23:25 nite Oct 19 11:23:53 Psi:.au? Oct 19 11:23:58 nz Oct 19 11:24:01 err - yes Oct 19 11:24:06 Psi: later mate Oct 19 11:24:06 Psi: anyway, nite Oct 19 11:24:19 see ya Oct 19 11:24:32 night Oct 19 11:27:07 SpeedEvil: you knew you can sync astab MV with a much higher external signal? Like 1Hz synced to the freq of mains power Oct 19 11:27:35 does not look like the n900 is avalebly in denmakr yet Oct 19 11:28:30 nor in Germany Oct 19 11:29:03 10-30 is current scheduled date of shipping Oct 19 11:29:42 DocScrutinizer: yes Oct 19 11:31:31 * DocScrutinizer-8 off for daily SOP to rescan nokia shop for another delay announced :-S Oct 19 11:31:49 heh Oct 19 12:30:22 openmoko: 03werner * r5695 10/trunk/gta02-core/docs/ecn/ (STATUS ecn0010.txt): Oct 19 12:30:22 openmoko: Various ECN0010 updates: Oct 19 12:30:22 openmoko: - use IO_3V3 to feed LDO34IN (CODEC_3V3 and IO_2V8) Oct 19 12:30:22 openmoko: - rename CODEC_3V3, i.e., to CODEC_VDD Oct 19 12:30:22 openmoko: - add 0R to LDO34IN, so that we can switch to Vsys if 3 V are not enough Oct 19 12:30:24 openmoko: - moved to "Execute" Oct 19 13:18:27 beers: the guy was obviously violating google Oct 19 13:18:34 beers: google Oct 19 13:18:43 beers: google's copyright Oct 19 13:19:48 beers: how would you like if someone took let's say SHR and make it closed source? That would be also copyright violation Oct 19 13:25:06 radekp: one thing is if you violate copyright of evil guys and other thing is if you violate copyright of good guys. Oct 19 13:26:12 anyone found a DECT system with user-modifiable software? Oct 19 13:26:18 radekp: laws exists not to follow them blindly, rather to try to bring a bit more fairness and sanity to our lives. If the law is wrong i can't see how it can be seen immoral to violate it. Oct 19 13:28:00 * amarsh04 could use a DECT handset that didn't have brain-dead timezone settings Oct 19 13:28:30 PaulFertser: i agree, but if google is going to loose money because of it (at least they say it) then to me it's perfect ok what they did Oct 19 13:30:04 radekp: first of all, they didn't prove they migth lose money. Oct 19 13:30:48 radekp: second, they still have their shit-store to earn enough. Even on non-"googleexperience" devices. Oct 19 13:31:54 radekp: third, restricting noncommercial apps distribution (afai understand the mod is noncommercial) is plain stupid because it can't be practically done. Oct 19 13:33:13 radekp: fourth, even if they're loosing money that way, it's good because they would deserve that by treating their users bad (i.e. providing uncool non-hackable firmware). Oct 19 13:33:51 radekp: not to mention goodle is evil and immoral anyway, i think the deal with chinese goverment is enough to prove this point. Oct 19 13:34:12 PaulFertser: they are probably loosing money because they denied the apps distribution - but explain that to sales manages :) Oct 19 13:34:40 radekp: that's not my point Oct 19 13:35:08 radekp: my point is that google is evil and hence you must blame it, not defense ;) Oct 19 13:35:53 defend Oct 19 13:36:01 PaulFertser: oki :) Oct 19 13:37:23 * mwester mutters something about situational ethics Oct 19 13:41:51 mwester: hey :) Oct 19 13:42:07 mwester: managed to break on through to the other side (of the firewall)? Oct 19 13:42:37 mwester: (in fact i can hardly imagine a corporate so draconian that it doesn't allow https and if it does, you can get ssh through it) Oct 19 13:43:14 mwester: (ethics) do you have any special education in this field? Oct 19 13:58:22 no special training required to understand the basics of ethics, and why they are not ethics if you can flex them to rationalize any actions. Oct 19 13:58:53 And the corporate firewall terminates https connections that are "long-lived". :( So no reliable ssh there. Oct 19 13:59:03 (I'm working from the home office this week, though) Oct 19 14:04:03 Hi! Is there somebody who could provide me with the rootfs and kernel for "Qt Extended 4.4.3"? The link on the website is down. Oct 19 14:35:14 PaulFertser: violating copyright is illegal, and immoral. that rule isnt bendable as long as you ever want to be taken seriously Oct 19 14:35:35 asfor googles "with google apps" yes - they have the right to go after those distributing them without a license. Oct 19 14:36:19 there is no way to make it acceptable to do that as long as you want people to obey your licenses - like gpl etc. Oct 19 14:37:13 of course google has pulled a bit of a swift one by going around all over the place that andorid is an "open platrform'and yet closing off parts of it that have been distributed as standard so far in real situations Oct 19 14:37:39 but they are allowed to enforce their rights Oct 19 14:38:18 just like you can enforce your right over your front yard if a bunch of people are camped out in it - you can kick them out. it's your yard. you dont have to prove they are causing harm or costing you money. Oct 19 14:38:45 you have a right and you may exercise it. its your front yard. same with their "with google" apps. Oct 19 15:00:33 openmoko: 03werner * r5696 10/trunk/eda/fped/ (Makefile gui_inst.c gui_style.c inst.h): Oct 19 15:00:33 openmoko: Drawing all the pad layers as filled areas can cause confusion. Now we draw Oct 19 15:00:33 openmoko: the solder mask with lines. Oct 19 15:00:33 openmoko: - gui_inst.c: draw solder mask with lines, not with filled rectangles and arcs Oct 19 15:00:33 openmoko: - gui_style.c: increase line width for solder mask from 1 to 2 pixels Oct 19 15:00:35 openmoko: - Makefile: added manual/concept-inst.png (forgot to commit this before) Oct 19 15:04:29 hey. is there any way to send files to other phone (not OM) with bluetooth? Oct 19 15:04:31 (violating copyright immoral) well, i do not want to be taken seriously then. Neither RMS obviously since he said many times that it's better to violate copyright than to disappoint a friend. Oct 19 15:04:56 Ddorda: yes, obex works as on any other GNU/Linux system. Oct 19 15:08:56 raster: so i can't agree with your point Oct 19 15:09:22 Good not to take RMS seriously, he's "over the rainbow", to use a figure of speech. Oct 19 15:11:39 mwester: i'm not sure copyright law is inherently ethical it's just that it was the most suitable to protect freedom given current legislation. Oct 19 15:16:02 PaulFertser: * Cannot find package obex. Oct 19 15:16:27 Ddorda: come on, read the wiki Oct 19 15:16:35 my bad Oct 19 15:18:28 PaulFertser_: is there any GUI tool? Oct 19 15:18:47 Ddorda: i think SHR has some but i'm not sure. Oct 19 15:21:20 PaulFertser: i will ask there Oct 19 15:41:44 Could someone please supply me with the rootfs of "Qtopia Extended 4.4.3" or "Qtopia Extended Improved"? I need them Oct 19 16:27:15 opimd-messages doesn't seem to show the newest message but all the others... can I do something about that? Oct 19 16:32:43 maybye You runned put of free space for messages? Oct 19 16:36:25 openmoko: 03werner * r5697 10/developers/werner/bl-pwr/ (. bl led.py): Backlight power efficiency measurements. Oct 19 16:37:28 nope, df shows loads of free space. and messages *do* appear, just the first one's always missing. looks more like a wrong sql query... Oct 19 16:57:42 * DocScrutinizer-8 wishes opimd would use sql-queries Oct 19 17:00:28 actually opimd uses sort of generic cache datastructure internally. dos1 is hardly convincable a "cache" based on sqlite would greatly simplify the whole thing Oct 19 17:01:51 and what about the sqlite backend settings in frameworkd.conf? Oct 19 17:01:54 DocScrutinizer-8: i'm already convinced. i just don't have time, motivation and skills to do that Oct 19 17:03:57 Tanuva: that's exactly that - a backend. Probably read out once on opimd start with "SELECT * FROM *" Oct 19 17:18:43 ah.. :/ Oct 19 17:19:04 freesmartphone.org: 03mickey 07cornucopia * r42369873c8db 10/fsogsmd/src/ (6 files in 2 dirs): fsogsmd: define Constants.CALL_INDEX_MAX. fsogsmd will have support for up to 7 concurrent calls. Oct 19 18:40:15 freesmartphone.org: 03mickey 07cornucopia * r4e516bfddf2a 10/fsogsmd/src/ (5 files in 2 dirs): fsogsmd: more work on call handling Oct 19 18:44:16 fsogsmd will have support for up to 7 concurrent calls. Oct 19 18:44:41 that will become expensive. Oct 19 18:58:43 hehe Oct 19 20:56:48 http://dev.haiku-os.org/changeset/33664/haiku/trunk/build :) Oct 19 20:57:50 wow, thx Oct 19 21:06:48 android is looking good Oct 19 21:08:47 hello Oct 19 21:10:04 slaxxer, what about smelling?) Oct 19 21:23:46 freesmartphone.org: 03mickey 07cornucopia * r3e9f338e9db2 10/fsogsmd/src/ (4 files in 2 dirs): fsogsmd: org.freesmartphone.GSM.Call.SendDtmf() done Oct 19 21:53:50 freesmartphone.org: 03mickey 07cornucopia * r65f60278725e 10/fsogsmd/src/lib/ (atcall.vala atcommands.vala): fsogsmd: more work on call handling, pending conference, transfer, and join Oct 19 22:07:15 freesmartphone.org: 03mickey 07cornucopia * r5c7d2aa7013d 10/fsogsmd/src/lib/ (atunsolicited.vala modem.vala): fsogsmd: call: enable call waiting URC and retrigger call syncing upon detection Oct 19 22:08:45 max_posedon: android stinks Oct 19 23:49:26 freesmartphone.org: 03mickey 07cornucopia * r376c357db750 10/fsogsmd/src/ (7 files in 2 dirs): Oct 19 23:49:26 freesmartphone.org: fsogsmd: org.freesmartphone.GSM.Network.GetStatus() done. Oct 19 23:49:26 freesmartphone.org: NOTE: Look at atmediators.vala:625-680 if you want to see a real beauty! Oct 19 23:49:26 freesmartphone.org: It's the monster of a mediator, but still pretty readable thanks Oct 19 23:49:26 freesmartphone.org: to our coroutines (and the stellar fsogsmd architecture, of course ;) **** ENDING LOGGING AT Tue Oct 20 02:59:57 2009