**** BEGIN LOGGING AT Fri Dec 07 02:59:59 2012 Dec 07 03:10:51 Is it possible my n900 can't get 3G with any carrier in the bay area?! Dec 07 03:11:18 I switched to another that promises HSPA+, and now I'm assuming that all HSPA+ is on a different frequency than 1900mhz... Dec 07 03:11:21 Any ideas? Dec 07 05:39:28 internetishard: N900 supports several UMTS frequencies, but some US carriers abuse GSM/2G bands for UMTS and that's beyond what N900 BB5 can handle Dec 07 05:40:37 typically those carriers have no 2G service either, so they are completely incompatible to N900 Dec 07 05:55:45 N900 ONLY support 900 and 2100 in UMTS. Dec 07 05:55:59 that makes the N900 totally unusable in most countries except Europe Dec 07 05:57:12 works fine in the US Dec 07 05:57:19 with T-Mobile only Dec 07 05:57:25 (we're talking 3G by the way) Dec 07 05:57:40 in this side of the world 3G (UMTS) frequencies are either 850, 1800 or 1900 MHz Dec 07 05:57:46 and N900 3G is 900/2100. Dec 07 05:57:59 TMO is all that matters Dec 07 05:58:05 :) Dec 07 06:01:06 I will not do business with that orange abomination in the US Dec 07 06:02:19 bastards try to send me a bill like 6months after I canceled (within the first week) for usage 5 months later...wtf Dec 07 06:12:54 Sc0rpius: WCDMA Band I (2100) WCDMA Band IV (1700/2100) WCDMA Band VIII (900) Dec 07 06:13:05 ^^^N900 Dec 07 06:14:23 Sc0rpius: and "your side of the world" 850 and 1900 are originally defined as GSM frequencies Dec 07 06:15:25 carriers that use those bands for UMTS are basicaly violating international agreements Dec 07 06:23:30 http://www.developer.nokia.com/Devices/Device_specifications/N900/ and http://mitglied.multimania.de/bauersebi/BOS-Funk/Funkdienste/Bandplan2GHz.htm and http://en.wikipedia.org/wiki/GSM_frequency_bands http://en.wikipedia.org/wiki/GSM_frequency_bands#The_Americas Dec 07 06:24:45 http://en.wikipedia.org/wiki/Cellular_frequencies http://en.wikipedia.org/wiki/UMTS_frequency_bands Dec 07 06:26:47 don't blame Nokia for protectionist politics/policies of USA - they always prefer to run their own standards incompatible to the rest of the world, probably to protect their markets Dec 07 06:29:45 Sc0rpius: also it's not "rest of the world, except Europe" that disobeys those standards. Rather it's basically just "the americas" that are such a mess, while "rest of the world" has no such problems Dec 07 06:30:23 get it, USA is third world regarding mobile phones Dec 07 06:40:51 grbml, smartsearch using 100% cpu Dec 07 06:45:48 "smart" rulez Dec 07 07:07:43 continuing to idly mude about band allocation for mobile services, I notice that 900/1800 is a sane technically based decision (1. harmonic falls into upper band), while USA opted for a totally weird 850/1900 and "justified" that by "military use of 900" iirc. They never where short of lame excuses for their protectionist politics Dec 07 07:07:54 s/mide/muse/. Dec 07 07:11:57 >>GSM-850 and GSM-1900 are used in Argentina, Brazil, Canada, the United States and many other countries in the Americas.<< http://en.wikipedia.org/wiki/GSM_frequency_bands Dec 07 07:12:47 [while GSM900/1800 is used in the sane rest of the world, so this 'cinflict' is as old as GSM] Dec 07 07:14:30 the extremely funny part in all this: obviously UMTS on 850/1900 does NOT conflict with "military use" of same bands in USA Dec 07 07:15:09 * RST38h is sure it is a conspiracy Dec 07 07:15:22 it however conveniently conflicts with GSM use of same bands in rest of the world Dec 07 07:15:45 BTW, apparently they turned GSM off in the US Dec 07 07:15:46 RST38h: don't get me started Dec 07 07:16:03 At least, GSM phones no longer work Dec 07 07:19:25 the american mobile phone situation is the product of free market Dec 07 07:19:39 yep Dec 07 07:20:59 prototypically shows that free market tends to pervert, since it's driven solely by short term economic interest Dec 07 07:22:40 so why use 900/1800 and allow cheap chinese mobile phones, when you can opt for 850/1900 and sell your own special proprietary phones for several times more money Dec 07 07:24:10 forget about global roaming, 99% of US citizens not even know there's land behind the huge waters starting at the US coasts Dec 07 07:25:29 DocScrutinizer05: comeon, that's not true, they are very well aware where they came from :P Dec 07 07:26:01 afaik you can't even legally use your own phone on US mobile networks Dec 07 07:26:04 or something Dec 07 07:26:12 I bet they think Russia is North of Canada, and Europe and Asia are part of that, and Africa an island south-east of europe russian province Dec 07 07:28:47 sorry, got my sarcasm day it seems. Take it easy, my US resident friends ;-) Dec 07 07:32:12 kerio: afaik that's only true for Japan and Korea Dec 07 07:33:23 there might be other countries with similar policies (only registered approved IMEI allowed to log in), but USA definitely isn't one of them Dec 07 07:34:40 hm, i definetely remember something like that for the US, though Dec 07 07:34:45 perhaps it was landlines Dec 07 07:34:47 such policy isn't feasible in countries with 2000 small carrier supply areas like in USA Dec 07 07:44:06 http://en.wikipedia.org/wiki/GSM_frequency_bands#Multi-band_and_multi-mode_phones is quite enlightening Dec 07 07:49:30 probably reding a lot of boring reports about ITU and 3GPP meetings, and see how many US-based carries DGAF, might also shed some light on the issue Dec 07 07:52:51 reeding* Dec 07 07:55:39 http://en.wikipedia.org/wiki/Alliance_for_Telecommunications_Industry_Solutions (US member of ITU and 3GPP) >>In the United States, ATIS competes with a number of other network equipment standards developers, including IEEE and the Telecommunications Industry Association.<< Dec 07 07:57:07 that's free market for you Dec 07 08:09:17 kerio: (landlines) for sure not, since historically all the "free" landline terminal equipment like phones, TAM, FAX, modem, all came or were targeted at US market mainly Dec 07 08:09:39 came from* Dec 07 08:10:02 RJ11 is an american invention I'd think Dec 07 08:14:44 kerio: http://en.wikipedia.org/wiki/RJ11#History_and_authority Dec 07 08:15:09 oic Dec 07 08:16:25 >>Bell System owned the phones and did not allow interconnection of separate phones or other terminal equipment<< >>This began to change with the case Hush-A-Phone v. United States [1956] and the FCC's Carterfone [1968] decision...<< Dec 07 09:54:50 hey , need deb of skype for n900 Dec 07 10:16:20 I'm not sure it's available as .deb Dec 07 10:16:37 ~pkg Dec 07 10:16:38 pkg is, like, http://maemo.org/packages/ Dec 07 10:19:03 maybe http://maemo.org/packages/view/skyhost-bin/ and http://maemo.org/packages/view/skyhost-vengine/, in addition to all those pkgs showing up under searchterm "skype" in pkg interface Dec 07 10:24:56 Devilinside: ^^^ Dec 07 10:45:03 ovi repo isn working here. yeah, i found those two packages. Dec 07 10:45:49 skype is avaolable on ovi, but it usually doesnt work. Dec 07 10:49:02 those two packages are already the nwest version. Dec 07 10:51:46 hey, is there deb of netcat? i was looking in extras,testing and devel but no luck Dec 07 10:53:43 get nmap, there's also ncat inside Dec 07 10:58:34 thanks tadzik, i have it already installed, just figured sec that is ncat Dec 07 10:59:26 netcat vs ncat driving me nuts every once in a while Dec 07 10:59:49 I usually mv ncat netcat Dec 07 11:00:04 nc another one Dec 07 11:00:47 prefer nc, less to type :) Dec 07 11:01:57 alas iirc that's ambiguous Dec 07 11:02:49 nc, mc, dc, bc... Dec 07 11:02:59 wc Dec 07 11:13:02 thnx its done :) Dec 07 11:36:27 Devilinside: it worked? Dec 07 11:43:18 yup Dec 07 11:46:07 yup Dec 07 11:46:45 thnx.. :) Dec 07 11:48:17 Bah, i need an OTG cable for the Galaxy S3 :/ Dec 07 12:07:42 http://thedailywtf.com/Articles/(Situation)-(location)-has-been-cleared.aspx#Pic-3 Dec 07 12:08:02 There's just so much win in that picture! Dec 07 12:08:19 "My Beethoven playlist on Pandora picked a song from Braveheart, on a Star Trek album, with Titanic album art. Where is this CD and how do I get it?" Dec 07 12:37:46 a song from braveheart in star trek? Dec 07 12:37:55 must be from Insurrection Dec 07 12:38:42 oh, khan Dec 07 12:53:13 dammit, why can't the skype plugin be removed easily? Dec 07 12:55:23 kerio, delete file list, keep deb, dpkg happy Dec 07 12:55:31 MohammadAG: no u Dec 07 12:55:34 dpkg is the law Dec 07 12:55:43 hm, apparently the problem is skype-ui-l10n-* Dec 07 12:56:04 wtf, why does rtcom-accounts-ui depend on that? Dec 07 12:56:26 cause maemo deps are fucked up? Dec 07 12:56:37 or they packed a library in the language files Dec 07 12:56:37 usually dependencies like that are actually legit Dec 07 12:56:42 whichever seems funnier Dec 07 12:56:55 i guess it's something stupid like using a localized string in skype-ui-l10n-mr0 Dec 07 12:56:56 hmm Dec 07 12:57:06 well, most probably Dec 07 12:57:52 oh well, i just need to leave the localization there i guess Dec 07 12:58:01 everything else gets removed fine Dec 07 13:10:21 DocScrutinizer05, hi, did you tried to backup nolo, kernel & rootfs to one fiasco file via 0xFFFF? Dec 07 13:11:09 nope Dec 07 13:12:27 Pali: that kind of backup would be useless Dec 07 13:12:33 you need the exact optfs too Dec 07 13:13:27 kerio, ok, so tar -czf opt.tag.gz /opt/ do that job? Dec 07 13:13:42 yes, but the problem is in restoring that :) Dec 07 13:14:21 kerio, maybe creating some kernel image with initfs image which will export mmc via usb... Dec 07 13:14:23 We need something like Android's recovery Dec 07 13:14:33 Pali: rescueOS :) Dec 07 13:14:42 or at least, something built upon rescueOS Dec 07 13:14:42 kerio, one option Dec 07 13:14:57 didn't we have a hidden partition at some poitn? Dec 07 13:14:59 but at that point you don't really need to make fiasco images Dec 07 13:15:04 or am I mixing up devices Dec 07 13:15:04 MohammadAG: we do have initfs, actually Dec 07 13:15:10 it's kinda small afaik Dec 07 13:15:16 how small? Dec 07 13:15:22 hold on, checking Dec 07 13:15:29 we can send initfs image via flasher or 0xFFFF Dec 07 13:15:40 2 or 4 MB Dec 07 13:15:40 MohammadAG: same as kernel Dec 07 13:15:44 is initfs partition Dec 07 13:15:49 if same as kernel then 2MB Dec 07 13:15:50 2mb Dec 07 13:15:58 hm, it's not a lot Dec 07 13:16:04 actually, it kind of is Dec 07 13:16:17 it'd hold the bare minimum Dec 07 13:16:24 the rescueOS initrd is 12mb Dec 07 13:16:30 ubooted Dec 07 13:16:31 better to use flasher/0xFFFF to send own kernel with own initfs without flashing Dec 07 13:16:51 rescueOS isn't bare minimum then :P Dec 07 13:17:05 MohammadAG: it's enough to get a fbcon shell Dec 07 13:17:11 usb mass storage, usb networking Dec 07 13:17:18 wifi (if you load the firmware from somewhere) Dec 07 13:17:35 MohammadAG: the current on-device restore still requires a working *something* Dec 07 13:18:03 uboot + sd is enough to load rescueOS Dec 07 13:18:09 which is enough for everything Dec 07 13:18:20 what you'd need is working microSD, which is provided by kernel, and tar Dec 07 13:18:26 plus a scripted UI Dec 07 13:18:29 so you'd need xloader+nolo+uboot Dec 07 13:18:45 MohammadAG: aka a "standalone" version of backupmenu Dec 07 13:19:05 MohammadAG: you also need a way to boot it, though Dec 07 13:20:17 kerio, https://www.youtube.com/watch?feature=player_detailpage&v=7-HqCbTanQM#t=59s Dec 07 13:20:45 MohammadAG: how much can be wrecked and still get that to function? Dec 07 13:21:00 well, you need a working bootloader Dec 07 13:21:05 and a kernel Dec 07 13:21:07 but that's about it Dec 07 13:21:11 MohammadAG: because backupmenu doesn't require a lot Dec 07 13:21:32 but those devices have recovery partitions larger than 2MB Dec 07 13:21:39 lemme check how much the S3 has Dec 07 13:21:47 MohammadAG: the n900 has up to 64gb of storage Dec 07 13:21:59 mtd blocks, not so much Dec 07 13:22:06 the emmc is always there Dec 07 13:22:09 I'm talking about something that could run without any external storage Dec 07 13:22:18 ok then, 32gb Dec 07 13:22:24 that's external :p Dec 07 13:22:34 go back to the OneNAND Dec 07 13:22:36 it's an EMBEDDED memory card Dec 07 13:22:47 still, it's user data Dec 07 13:22:53 you could break something when partitioning it Dec 07 13:22:53 not /home :) Dec 07 13:23:26 5MBs Dec 07 13:23:29 MohammadAG: i bet that you can't fit a pseudographical restore system in 2mb Dec 07 13:23:33 for an advanced touch recovery Dec 07 13:23:43 also, where are you going to backup/restore? Dec 07 13:23:53 if it's a sd, then just require something to be put there, ffs Dec 07 13:23:55 that's where you could mount the 32GB Dec 07 13:24:08 like MyDocs Dec 07 13:24:21 MohammadAG: if mydocs is hosed, you want to repartition it, reformat it, and restore from usd Dec 07 13:24:30 and now you also need fdisk, mkfs.vfat and mkfs.ext3 Dec 07 13:25:01 hmm, didn't think of that Dec 07 13:25:11 CWM recovery just formats, then unpacks an .img Dec 07 13:26:43 that's what backupmenu does for /home Dec 07 13:27:10 it does a rm -rf and an unpack for rootfs, because ubifs is tough as fuck Dec 07 13:27:17 and because nobody understands how to fix it Dec 07 13:27:54 actually no, it's because to load backupmenu you need rootfs to be at least partially working, and that includes not being fucked up Dec 07 13:29:07 DocScrutinizer05; hey have you seen some people "fixing" cellmo issue by adding spring to push down cellmo chips? Dec 07 13:29:17 DocScrutinizer05; http://www.jabawok.net/?p=14 Dec 07 13:29:49 ShadowJK: nope, that's news to me Dec 07 13:36:11 wow, awesome! Dec 07 13:36:18 bookmarked Dec 07 13:36:31 DocScrutinizer05: so the issue is really a bad soldering Dec 07 13:36:45 it's not that some batches have defective chips Dec 07 13:37:17 well, that's already expected Dec 07 13:37:53 though it *could* as well be poor bonding combined with poor sealing of the BGA balls Dec 07 13:38:23 btw, won't pushing on the chip eventually cause it to fail? Dec 07 13:38:35 but incomplete soldering under BGA chips is far form being unheard Dec 07 13:38:45 kerio: shouldn't Dec 07 13:38:54 especially if, for instance, you drop your n900 on the ground Dec 07 13:39:07 nvidia had some issues with some of their chips Dec 07 13:39:28 the chip's sealing shouldn't be plastically deformable Dec 07 13:40:11 it's supposed to be ceramic material that doesn't permanently deform Dec 07 13:42:00 btw this extremely hard case material of the chips creates a lot of the problems seen during reflow and when dropping the final product thus bending the PCB Dec 07 13:42:45 i see Dec 07 13:42:56 can't we make everything out of that material? :3 Dec 07 13:43:17 hmm, I think there are PCB made of ceramic Dec 07 13:43:43 but probably that would just crack into pieces when you drop the device Dec 07 13:44:08 hm Dec 07 13:44:22 do you want your device to be flexible then? Dec 07 13:44:34 it could deform if you make it too flexible Dec 07 13:45:24 I actually just a minute ago started thinking about chips with spring loaded BGA soldering balls Dec 07 13:46:28 which could mitigate a lot of those problems Dec 07 13:47:57 0.1mm bearing play up/down and sideways was already sufficient I guess Dec 07 13:49:40 of course building those microsprings into the PCB pads and not soldering the chip at all but simply pressing it on the pads could also work quite nicely Dec 07 13:50:23 probably you'd want gold plated BGA balls on the chip for that Dec 07 13:50:46 * kerio is happy of studying math instead of EE Dec 07 13:50:56 hehehe Dec 07 13:51:11 this real world things seem really tough! Dec 07 13:51:38 by the way Dec 07 13:51:48 what's the usb-connector fix? Dec 07 13:51:48 indeed that's exactly the notion I detected on a lot of programmers coming from the math side as well Dec 07 13:51:52 Is it fixable? Dec 07 13:52:05 teotwaki: "kinda" Dec 07 13:52:07 teotwaki: hmm? Dec 07 13:52:13 ~usbfix Dec 07 13:52:14 from memory, usbfix is http://talk.maemo.org/showthread.php?t=75920 - and **NEVER** use epoxy (unless you want to seal your device for underwater) Dec 07 13:52:16 the better you are at soldering, the greater the damage you can fix Dec 07 13:52:17 or what? Dec 07 13:52:34 I haven't had a single usb port fail, it might be due to how I unplug Dec 07 13:52:36 My N900, currently in use by my girlfriend, has been showing "charging" a lot, sometimes. Dec 07 13:52:40 DocScrutinizer05: the math side has different problems :3 Dec 07 13:52:47 teotwaki: huh? Dec 07 13:52:52 oh, it thinks it's charging?= Dec 07 13:52:58 not really Dec 07 13:52:59 does the physical connector work? Dec 07 13:53:01 when it's plugged in Dec 07 13:53:05 oh :( Dec 07 13:53:15 there seems to be a bad connection Dec 07 13:53:49 either plug or receptacle or soldering of the latter Dec 07 13:54:06 or your PSU is defect ;-D Dec 07 13:54:15 or wire, my first car charger's wire got bad Dec 07 13:54:18 or cable broken Dec 07 13:54:32 * DocScrutinizer05 highfives ShadowJK Dec 07 13:59:27 hmm Dec 07 13:59:29 will check it out Dec 07 13:59:31 teotwaki: first test a new/different charger, or simply PC Dec 07 13:59:58 yeah, will use the heavy duty Palm charger. Dec 07 14:00:01 if it still fails in same way, see Dec 07 14:00:04 ~usbfix Dec 07 14:00:05 somebody said usbfix was http://talk.maemo.org/showthread.php?t=75920 - and **NEVER** use epoxy (unless you want to seal your device for underwater) Dec 07 14:00:50 but also check the USB receptacle's contacts, sometimes they get bent or corroded or simply dirty Dec 07 14:01:57 it's not unusual that a defect USB plug destroys the receptacle's contacts Dec 07 14:02:30 or some debris got into the receptacle Dec 07 14:07:10 I almost did that the other day Dec 07 14:07:24 but spotted the crapp in n900 usb port in last second Dec 07 14:15:23 * DocScrutinizer05 wonders to which temperature you'd need to pre-heat the complete N900 PCB so you could successfully reflow the BB5 chips with hot air Dec 07 14:15:45 heh Dec 07 14:17:22 probably finding out about the reflow temperature profile used in N900 PCB assembly would be rather useful for that Dec 07 14:18:12 friggin reflow is kinda voodoo magic Dec 07 14:19:05 evidence: Nokia failed on several BB5 chips when doing reflow Dec 07 14:19:17 are the boards moved through ovens with different temperaures? Dec 07 14:19:28 yep, usually Dec 07 14:19:51 do these have airflow too? Dec 07 14:19:55 or still air Dec 07 14:20:26 several minutes pre-heat, then ramp up to max during 30s, keep it at max for 10s, ramp down to chillout temp and keep it there for another minute, then slowly cool down Dec 07 14:20:39 usually IR and still air Dec 07 14:20:50 hm Dec 07 14:21:04 or very smooth airflow Dec 07 14:21:10 on top of IR Dec 07 14:21:24 honestly I don't know much about rwflow Dec 07 14:21:36 it's an own art Dec 07 14:21:55 the concept of temperature in presence of IR seems daunting to me Dec 07 14:22:47 well, in every environment you got air- and radiation-temperature Dec 07 14:23:08 the trick with reflow is to balance the two Dec 07 14:23:32 and to have a proper profile Dec 07 14:24:07 is the temperature profile supposed to represent air temperature or actual temperature of some board component.. Dec 07 14:24:17 vapour phase reflow has also been used, and has some advantages Dec 07 14:24:24 bit is uncommon Dec 07 14:24:25 http://www.nxp.com/wcm_documents/about/corporate-social-responsibility/environment/lead-free-halogen-free/images/diagram.gif Dec 07 14:26:01 http://www.nxp.com/about/corporate-social-responsibility/environment/lead-free-halogen-free/understanding.html Dec 07 14:27:37 no mention of tin whisker Dec 07 14:28:06 seen: http://nepp.nasa.gov/whisker/anecdote/af114-transistor/index.html ? Dec 07 14:29:03 ShadowJK: >>Note: All temperatures refer to topside of the package, measured on the package body surface<< Dec 07 14:32:41 SpeedEvil: dafaq! tin whiskers :-o Dec 07 14:33:36 http://nepp.nasa.gov/whisker/anecdote/2009busbar/index.html Dec 07 14:33:37 this friggin germanium transistor looks strangely familiar to me, even inside ;-D Dec 07 14:33:41 is fun too Dec 07 14:33:55 it's not just a small signal problem Dec 07 14:34:34 DocScrutinizer05; hm, so when usin IR, the board temp would be different depending on the board mask color, even if top of package temp was same? :-) Dec 07 14:34:49 ShadowJK: well, yes and no. Dec 07 14:35:01 ShadowJK: colour is far IR colour Dec 07 14:35:03 ShadowJK: that's the magic Dec 07 14:35:18 right Dec 07 14:35:19 ShadowJK: most organic materials have emissivities close to 1 Dec 07 14:35:30 in the far IR Dec 07 14:35:50 to the point that making low emissivity paints is really hard Dec 07 14:36:44 in one factory I visited a reflow street, and they had those test objects there, with sophisticated temperature sensors of all kinds on the board Dec 07 14:37:48 really experienced guys tune the oven for half a day, so results obtained with such test objects were inside the defined profile Dec 07 14:39:02 and the factors that influende such tuning are even such things like "shadowing" by shielding blades on the board etc Dec 07 14:40:03 of course board thickness and other board properties (copper ground planes etc), chips' size, whatnot else Dec 07 14:41:32 and when you've tuned the oven with those test objects, you run a number of sample boards, then check the soldering with xray etc, then do some finetuning of the profile Dec 07 14:42:05 run another number of samples... rinse, restart Dec 07 14:42:06 The supplier of dies at $work switched to a surface finish with low emissivity and high reflectivity, had been totally opposite before. We usually use IR sensors to check whether thermocouples are still giving reasonably OK readings or not. It was kinda confusing at first before we realized what was going on. Dec 07 14:42:45 hehe Dec 07 14:43:34 IR is amusing Dec 07 14:43:55 for the accuracy *inside* the die it's for sure better when they have reflective surface Dec 07 14:44:02 take a kettle, place a sheet of aluminium foil over it, and boil it. Dec 07 14:44:13 Switched/switching to pt100, nicer stability over time, and failure modes are nicer Dec 07 14:44:31 point ir at sheet, measure temp of ceiling Dec 07 14:44:42 :nod: Dec 07 14:44:52 spray with a tiny coat of oil - 100c Dec 07 14:45:01 :nod: Dec 07 14:45:02 :-) Dec 07 14:45:49 I want to make a IR thermometer good for measuring the temp of single BGAs balls. Dec 07 14:45:54 this is hard. Dec 07 14:46:03 ouch Dec 07 14:46:15 seems terribly hard Dec 07 14:46:15 as part of a longer term project to do laser dead bug soldering Dec 07 14:46:32 o.O Dec 07 14:47:08 muktilayer PCBs are expensive Dec 07 14:47:17 how would you even *look* at a BGA ball that's supposed to be/get soldered to a pad? Dec 07 14:47:20 I'm wondering on alternatives Dec 07 14:47:27 dead bug Dec 07 14:47:54 nfc Dec 07 14:48:30 apply tinned 0.1mm wire to surface of ball, straight down. hit with laser to melt the ball, and feed wire in. Dec 07 14:49:01 hmm Dec 07 14:49:10 ideally it ends up with a nice solder joint, with the enamel remaining almost until it goes into the ball Dec 07 14:49:32 yup Dec 07 14:50:02 but what you gonna do with a chip with wires all over it? Dec 07 14:50:30 automatic point to point wiring Dec 07 14:51:13 YAY Dec 07 14:51:29 it's 1940 again :-) Dec 07 14:51:41 PCB making roof-down :-) Dec 07 14:52:24 sounds like a nice alternative concept Dec 07 14:52:47 if it works, then 20 wires per minute is fairly plausible. Dec 07 14:53:00 glue all the components to a board upside down, then wire the while mess Dec 07 14:53:14 and though that's not a really great production technique... Dec 07 14:53:17 whole* Dec 07 14:53:55 meh, it sounds great regarding some aspects Dec 07 14:54:27 there are major issues Dec 07 14:54:34 but it's missing one extremely important detail: shielding/heatsinks Dec 07 14:54:40 impedence control is 'interesting' Dec 07 14:54:49 and that. Dec 07 14:55:46 potting in thermal compound against a plate goes some way. Dec 07 14:55:59 but has its own issues Dec 07 14:56:40 well, on BGA the balls are also thermal vias Dec 07 14:57:27 I mean potting the wiring side Dec 07 14:57:30 while thermal resistance to top of case is undefined and usually poor Dec 07 14:57:36 aaah Dec 07 14:58:14 it's never going to be a good replacement generally. Dec 07 14:58:24 could it be useful - perhaps Dec 07 14:59:07 well, for certain usecases it is for sure a good rapid design alternative Dec 07 14:59:34 signal separation though is a nightmare Dec 07 14:59:47 crosstalk rather Dec 07 14:59:58 very much Dec 07 15:00:19 one very silly idea I had. Dec 07 15:00:28 conformal coat it. Dec 07 15:00:39 fill with low melt solder Dec 07 15:00:58 in practice, likely not to work due to thermal expansion Dec 07 15:03:01 * DocScrutinizer05 waves Dec 07 15:03:33 Hi Dec 07 15:03:46 180min til meeting, need to get some shopping etc done Dec 07 15:04:22 I have some problems with /usr/sbin/browserd - it's eating my cpu -,- Dec 07 15:04:33 hah Dec 07 15:04:40 killall browserd Dec 07 15:04:51 Doesn't work Dec 07 15:04:56 have fun Dec 07 15:04:57 o.O Dec 07 15:05:29 anyway, bbl Dec 07 15:05:39 :-) Dec 07 15:05:45 DocScrutinizer05: the process is starting anew itself Dec 07 15:05:56 sure Dec 07 15:07:14 It renders browser stuff, and also Conversations Dec 07 15:07:32 Is any way to kill it? Dec 07 15:07:44 permanently* Dec 07 15:08:38 you realize that it would make Conversations stop working? Dec 07 15:11:31 okay, I wouldn't want to stop Conversations Dec 07 15:12:19 a) find out which of the several browserd processes eats CPU, b) find out what it's doing, c) fix the source of the problem Dec 07 15:12:42 PID TTY STAT TIME COMMAND Dec 07 15:12:43 1697 ? Ssl 0:00 /usr/sbin/browserd -d Dec 07 15:12:44 2177 ? Sl 0:03 \_ /usr/sbin/browserd -s 2177 -n browserui Dec 07 15:12:46 9487 ? Sl 0:04 \_ /usr/sbin/browserd -s 9487 -n RTComMessagingServer Dec 07 15:14:11 strace is your friend Dec 07 16:25:18 guys, what do you use to block ads? Dec 07 18:39:55 iDont: 'lo Dec 07 18:40:18 kerio, hi! Dec 07 18:40:32 it's Friday :) Dec 07 18:59:52 4 minutes to saturday here Dec 07 19:00:26 sux! got 4 hours! Dec 07 19:57:58 http://arstechnica.com/information-technology/2012/12/richard-stallman-calls-ubuntu-spyware-because-it-tracks-searches/ Dec 07 20:09:34 Aoyagi: clearly, that means that you must buy a n900 really quickly Dec 07 20:12:03 kerio: I'm working on fixing mistake of not having one yet as fast as I can. Dec 07 20:12:55 i wonder if i should install the actual flashblock and drop this silly css adblocker for microb Dec 07 20:13:09 and maybe add some adblocking to the hosts file Dec 07 20:13:41 Which reminds me Dec 07 20:14:03 Any idea what version of Opera run on N900? Dec 07 20:14:11 opera mobile 12 Dec 07 20:14:26 which is a surprisingly competent mobile browser Dec 07 20:14:46 Very. Dec 07 20:14:55 And great. Dec 07 20:15:10 about as capable as opera mini on a s60 phone, but also very, very light Dec 07 20:15:58 Well, Mini is lesser version. Dec 07 20:16:17 Java or something. Dec 07 20:16:17 opera mobile only adds local rendering, pretty much Dec 07 20:16:27 And some options. Dec 07 20:16:27