DocScrutinizer05 changed the topic of #neo900 to: http://neo900.org | conversations are logged to http://infobot.rikers.org/%23neo900/ and http://irclog.whitequark.org/neo900 | 2013-11-04 - the day our fundraiser reached its goal | 2014-05-01 360 devices 75k€| 0712 183 ~30k | 0810 300 ~49k | 0914 346 ~56k
<DocScrutinizer05> haha that's innovative use of a USB jack: http://wstaw.org/m/2014/11/19/plasma-desktopKM1987.png
<DocScrutinizer05> ((lots of tables)) anything worthwhile?
<DocScrutinizer05> *cough* intersting what weird certifications you seem to need: http://privatepaste.com/cb6e5e0342
<Oksana> ((anything worthwhile?)) I cannot make sense of which chips include what. Are some chip including only RAM, others only NAND, and some others both RAM and NAND? Like K4P8G304EB ... http://www.samsung.com/global/business/semiconductor/product/mobile-dram/detail?productId=7870&iaId=747
<DocScrutinizer05> quite decent power consumption: >> Power per fix: < 1.3 mAh, including cold start of GPS, GSM power-up and transmission via GPRS or SMS.<<
<DocScrutinizer05> Oksana: yes
<Oksana> K4P8G304EB is probably 1GB RAM without NAND ;-/ On page 21 (NAND + RAM) I can see only 162FBGA and 169FBGA, not 168FBGA?
<DocScrutinizer05> toldya
<DocScrutinizer05> why don't you believe me?
<DocScrutinizer05> [2014-11-19 Wed 00:26:21] <DocScrutinizer05> KCE00E00CA has a quote strange numbering that might indicate that it is a custom made product. also it seems Samsung doesn't build any FBGA168 MCP any more
<DocScrutinizer05> [2014-11-19 Wed 00:27:12] <DocScrutinizer05> OMAP3 (FBGA168) is obsolete
<Oksana> Strangeee... Why isn't 162FBGA obsolete, too? Does "NAND based MCP" sound right? Why does Micron (we do not have to use Samsung, right?) have 1GB NAND and 512MB DRAM, but not the other way round?
<DocScrutinizer05> well, I cannot answer any of your questions
<DocScrutinizer05> for "strange numbering" and CA part number ending, also see "KAT00F00DD-AE77"
<DocScrutinizer05> "-Axxx" indicates it also is a 168FBGA
<DocScrutinizer05> "DD" ending says nothing about speed I'd guess
<DocScrutinizer05> I have no faintest clue what SoC might use 162FBGA
<DocScrutinizer05> and I'm honestly not interested in finding out
<DocScrutinizer05> maybe even some of Samsung's own SoC, like the S3C64xx. You should be able to find out easily
kolp has quit [Read error: Connection reset by peer]
<Oksana> Same problem on Digikey: 8G NAND+4G RAM yes, 4G NAND+8G RAM no. Somebody is laughing...
<Oksana> All MCP products are customised... Toshiba: http://www.toshiba-components.com/memory/mcp.html
<Oksana> Hell... What's the difference between 168 and 169?
xes has quit [Quit: Going offline...]
<Oksana> JEDEC 14x14mm, 0.65mm, 152ball; OR; JEDEC 12x12mm, 0.5mm, 168ball; according to datasheet of Neo900's CPU?
che1 has quit [Ping timeout: 264 seconds]
askerix has joined #neo900
* askerix enter
askerix has quit [Client Quit]
<DocScrutinizer05> Oksana: seen all that before
<Oksana> So, waiting for GreenLightElectronics and hoping for a datasheet?
<Oksana> You contacted them, right?
<DocScrutinizer05> sure
<Oksana> :-)
<DocScrutinizer05> but there's still the problem that our funds vasically don't allow securing of "expensive" stuff like the PoP RAM
<DocScrutinizer05> s/vas/bas/
<Oksana> Another donation will be coming your way, soon :-)
<DocScrutinizer05> great! but we need to turn that into a campaign
<Oksana> Heh... :-)
<DocScrutinizer05> we have a ~380 preorders and we need to source/secure ~400..500 component sets already for them, each of which is in the range of 50 bucks at least
<DocScrutinizer05> the 100 bucks donation limit been a bit humble, not taking any such component securing into account
<DocScrutinizer05> actually they been somewhat on the low side even for mere R&D costs
<DocScrutinizer05> I probably should change that threshold to 200 right away, doesn't introduce any unfair changes to those who already donated
<bencoh> this might sound silly but ... do we have n950 schematics somewhere ?
<DocScrutinizer05> nope, not afaik
<DocScrutinizer05> and it doesn't sound silly
<DocScrutinizer05> :-)
<Oksana> Not yet. N900, yes, Neo900, yes. Not N950, not N9, not anything else.
<DocScrutinizer05> N9 is available
jonwil has quit [Quit: ChatZilla 0.9.91 [SeaMonkey 2.30/20141013232806]]
<DocScrutinizer05> just looked at it 20 min ago, studied audio
<bencoh> yes we do, I have it here as well
<Oksana> Ah... But not as interactive HTML, right?
<bencoh> pdf
<DocScrutinizer05> huh? we don't have any html schematics
<bencoh> hmm looks like they used rapuyama as well
<DocScrutinizer05> err yes, kinda
<DocScrutinizer05> BB5 been Nokia's universal WWAN chip, used in virtually everything they built
<DocScrutinizer05> basically just a customized OMAP chip aiui
<DocScrutinizer05> s/chip/SoC/
<DocScrutinizer05> ;-D
<bencoh> I wonder what it would take to clone a n9
<DocScrutinizer05> I googled a bit on that user and seem to recall n his CV he mentioned "Employer: TI, cooperation with Nokia to develop their new baseband chip based on OMAP" OWTTE
<bencoh> re-use case, screen, and .... mostly everything
<DocScrutinizer05> what for?
<bencoh> DocScrutinizer05: wait ... I checked this page 2mins ago but the image didnt load :]
<bencoh> now it loads well ... wtf :)
<bencoh> DocScrutinizer05: just "comparing" to neo900
<bencoh> wondering whether it would be simpler or not
<DocScrutinizer05> nope
<DocScrutinizer05> au contraire
<bencoh> (housing/casing set apart)
<bencoh> why ?
<DocScrutinizer05> N900 is much better understood
<DocScrutinizer05> N9 we could only blindly clone according to schematics, and we run into insurmountable roadblock with... BB5
<DocScrutinizer05> Oksana: thanks for the tmo post! :-)
<bencoh> roadblock ?
<bencoh> hw or sw ?
<DocScrutinizer05> both
<DocScrutinizer05> you cannot source BB5, you have no datasheet for it, and you have no firmware for it
<Oksana> Is it also used in N950?
<bencoh> dont we have fw from n9/n950/n900 ?
<DocScrutinizer05> it's not even understoof exactly how it talks to APE
<DocScrutinizer05> bencoh: I'd guess N9 firmware won't fit at all on N900 BB5 modem, and vice versa
<bencoh> well, I guess there are some n9 out there ;)
<DocScrutinizer05> sure, but the firmware is crypto-signed
<DocScrutinizer05> and the BB5 carries the key
<bencoh> hmm
<DocScrutinizer05> and you cannot flash this key into a spare BB5 since you don't have it
<Oksana> Heh... Lily answered with 15 USD per piece, in stock and fast delivery; valid for next 5 days. Question is, when will she send a datasheet?
<DocScrutinizer05> question is, which quantity
<bencoh> not sure I get everything, but ... I'll think about that tomorrow morning ... 3am now :)
<DocScrutinizer05> she probably just founf the source we already know of
<bencoh> 'night :)
<DocScrutinizer05> night bencoh
<DocScrutinizer05> same time here ;-)
<DocScrutinizer05> local time, not _my_ time
<DocScrutinizer05> Oksana: and a further question is: will she be able to send a sample for us to evaluate?
<DocScrutinizer05> but yeah, it sounds not all that bad. Great job, thanks! please stay tuned and ask for a few days to decide
<DocScrutinizer05> would be actually quite interestng if she could offer real quantities as well, say 5k for example
<DocScrutinizer05> and I just have another weird idea, maybe some suppliers would offer ship sell in - say - 9 months from now, with a 5% paid right now and rest in shipping date. And they have opportunity to ask for earlier payment and shipping (e.g. when some other customer asks for their last 5k chips) but in that case we would get back the 5% down payment when we refuse to proceed with purchase. When however we don't want the chips after 9 months, then
<DocScrutinizer05> they may keep the 5% since it's us who break the deal
<DocScrutinizer05> Oksana: could you ask for MOQ and price on MOQ? For evaluation
<DocScrutinizer05> like 5 or 10 pcs
Kabouik has quit [Read error: Connection reset by peer]
louisdk has joined #neo900
nox- has quit [Quit: Leaving]
astr has quit [Ping timeout: 264 seconds]
astr has joined #neo900
jonwil has joined #neo900
<Oksana> Metaphysics of Quality? :-) The quote was for 1k, unless there was a fundamental misprint somewhere. No response about data sheet, yet. I will ask them about an evaluation sample, too.
<Oksana> quota*, even?
useretail has quit [Quit: Living in realtime. Thinking in binary. Talking in IP.]
useretail has joined #neo900
louisdk has quit [Remote host closed the connection]
<Oksana> Is Clarivue screen protector good for Nokia N900?
<DocScrutinizer05> MOQ == minimum order quantity
<DocScrutinizer05> no idea, got me some nice ones that were on a donation device
obsed has joined #neo900
<DocScrutinizer05> "Professional Screen Guard" "Japan PET"
Openbot has joined #neo900
<DocScrutinizer05> pretty much a noname product, "Materials from JAPAN, made in China"
<DocScrutinizer05> "the greatest guarder for lcd screen"
<DocScrutinizer05> it makes for a really nice anti-glare and no-fingerprints surface that's also pleasant to touch
<Oksana> Does not resist movement of stylus? Still, I cannot see it anywhere. Being sold for N900, I mean.
<Openbot> Docscrutinizer05: a silly question on how to find rx tx pads on any mobile board ? I mean these days strange markings come like D+ D- , X Y , even no labelings and worst intentional wrong markings? So how to reliably find such pads ?
<DocScrutinizer05> Openbot: you completely lost me
<Openbot> I am intrested in usb to serial
<DocScrutinizer05> Oksana: it somewhat has a nice "resistance" to sliding stylus, more like real writing with pencil
<DocScrutinizer05> Openbot: converter like that FTDI chip?
<Openbot> What ?
<Openbot> Wait let me rephrase
<DocScrutinizer05> Oksana: hmm, mine looks differently packaged, but for sume reasons I got this bookmark: http://www.ebay.es/itm/181091251333?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649#ht_3206wt_1132
<Openbot> oprekhp.blogspot.com.es/2011/05/virtual-usb-to-serial-com-port-for.html?m=1 go to that :) i wat to find the pinouts myself when they are not marked ar worst how to tell when they are wrongly marked ?
<DocScrutinizer05> Oksana: the (1) .. (5) sketches look exactly like the ones printed on backside of my package though
<Oksana> Darn! Lily ke has confirmed with the manager. This item is temporary out of stock.
<Openbot> Doom !
jonwil has quit [Quit: ChatZilla 0.9.91 [SeaMonkey 2.30/20141013232806]]
<Oksana> Apparently, at whatever pricelist Lily looked, it did not have up-to-date availability information. | So, waiting for GreenLightElectronics' reply.
<DocScrutinizer05> haha
<DocScrutinizer05> toldya
<DocScrutinizer05> :-(
<Oksana> So, these screen protectors do not randomly separate themselves from screen, get wrinkles or bubbles?
<Openbot> Docscrutinizer05?
<DocScrutinizer05> Openbot: seems http://oprekhp.blogspot.com.es/2011/05/virtual-usb-to-serial-com-port-for.html?m=1 has all the Q&A that possibly could get answered
<DocScrutinizer05> Oksana: I haven't seen this happening, but then I haven't seen or used many screen protectors in my life yet
<Oksana> Good night
<DocScrutinizer05> huh?
<Openbot> Docscrutinizer was just wondering if you could tell me more facts on finding such pads :)
brolin_empey has joined #neo900
<DocScrutinizer05> no, I had to do exactly what's written in Q&A as well
<Openbot> Ok bye i was just wondering on the board that i have on hand with x y marks lol :p
<Oksana> The battery charge indicator (the one in status bar) can easily go up from 1/8 to 3/4. :-) For no reason.
<DocScrutinizer05> honestly I fail to see the use of doing hacks on weird crappy flashertool boards
<Openbot> Oksana: be careful with chips i bought 10 spd 6820 (relatively new cpu)and you know none of em work !
<Openbot> Docscrutinizer05 lol you are high level you design em and i repair em ala low level rookie :p
<DocScrutinizer05> get a data cable, available for < 5 bucks
<Openbot> My knowhow = new chip available due to demand = hosed chips mostly
<DocScrutinizer05> it comes with decent low voltage UART <-> USB. Not some shady hack that operates with a 1n4007 and claims that silicon diode would introduce a voltage drop from 5 to 4.7V. I still don't get it what you do with those 4.7 err 4.3 actually volt anyway. low volatge UART is 1.8V or sth
<Openbot> Docscrutinizer05 you might jump out of your bed if i show you my inventry of cables :p
<Openbot> Bbl
<DocScrutinizer05> well, UART aka serial is used on e.g. Nokia F-BUS iirc
<DocScrutinizer05> and, surprise surprise, the UART3 on testpads is called F-Bus in several Nokia schematics
<Openbot> Yeah at times i flashed nokias with hand solderd cable
<Openbot> Just rip an fbs take a multimeter 10 min and ready !
<DocScrutinizer05> honestly, get an adapter called "data cable", attach a matching plug to it. You're better off than with frying your phone with some hacked Cruiser thingie
<Openbot> Hey no cruzier or insane stuff here just posted that link to rephrase my comment
<DocScrutinizer05> and about how to find RX/TX on a random phone PCB, I have no idea. Except "get the service manual!"
<Openbot> How neo900 gona flashed ?
<DocScrutinizer05> most easily by inserting a uSD with the image on it ;-)
<DocScrutinizer05> or by using DFU-util and connecting to uBoot
<DocScrutinizer05> via USB
<Openbot> Get the service manual ?huh once i mailed a vendor and the reply was like do yo do drugs ?
<DocScrutinizer05> or *maybe* even original NOLO istead of uBoot works and you can use flasher-3.5
<DocScrutinizer05> haha! google is your friend
<DocScrutinizer05> the vendor is definitely *not*
<DocScrutinizer05> seems even Nokia service manuals are only ever officially available to registered Nokia care centers and service points
<DocScrutinizer05> but all those manuals leak
<DocScrutinizer05> so search for then in google and it will take 30s to find them
<DocScrutinizer05> not that those service manuals always have useful info about UART testpads
<DocScrutinizer05> e.g N900 and N9 don't have any mentioning of UART3
<DocScrutinizer05> well, N9 SM sugguests that UART3 was inside TWL4030 on USB D+/-. I kinda doubt that's the _real_ CPU UART, what they call "UART3" there is well known serial mode of USB
<Openbot> K1 v2.0
<Openbot> 140510_n 02691w0 v12_0412_bj
<DocScrutinizer05> errwut??$v(856$ss?
<Openbot> ^ the marking on the slider mech i found a couple of days earlier
<DocScrutinizer05> fine... so?
<DocScrutinizer05> no idea what you're talking about
<Openbot> Of n900
<DocScrutinizer05> well, whatever that means
<Openbot> It was written on the slider mech of n900 available as replacememt slider
<DocScrutinizer05> ok. thnks for the info
<DocScrutinizer05> I don't know what to do with it though
<Openbot> And a185f written on the flex with rgb led
<DocScrutinizer05> there's a lot of weird numbers everywhere on every device
<DocScrutinizer05> nobody except the factory knows their meaning, and sometimes not even factory knows
<Openbot> I posted em as they are claimed to be oem by vendor
<Openbot> Also i got oem speakers as claimed i bought a pair :D
<Openbot> Things i could find were flex . Slider. Display. emmc. Casing without antena.
Openbot has quit [Quit: used jmIrc]
<DocScrutinizer05> sliders are rare
Openbot has joined #neo900
<Openbot> What ? He hot 2 !
<Openbot> *got
<Openbot> Maybe he sourced from nokia then
<Openbot> I mean here was a flood of symbians as cheap as $10 after the deal
<DocScrutinizer51> Two?! oh wow, heaps, mountains, abundance
<Openbot> Its a question of avaiblity as somewhere there may be larger stocks
<Openbot> Like your pop chips
<Openbot> 1110 type handsets are still avaible at $4
<DocScrutinizer51> whatever, we don't need 1110
<Openbot> But there were no n900s in the flood nither any of n9xx series
<Openbot> I mentioned that as availiblity of nokia parts became nonexistant during the time
Openbot has quit [Quit: used jmIrc]
Svetlana has quit [Quit: ZNC - http://znc.in]
sparetire_ has quit [Quit: sparetire_]
<freemangordon> DocScrutinizer05: something is fishy in xprot, if I don't tweak the stock parameters(that lengthy hex string), the variable called "T_coil_est" never changes from 0. no matter the ambient temp. seems like the algo for temp prediction is there, but not used, for some reason (too slow?).
<freemangordon> jusa_: any idea? ^^^
<freemangordon> there is some variable in parameters, called t_mp (temp multiplier?) that is zero, but it needs to be not zero in order est temp to be calculated
<freemangordon> weird
<DocScrutinizer05> indeed weird
<freemangordon> (__sat_mul_add_16 is saturating 2*a*b)
<freemangordon> or somesuch
<DocScrutinizer05> are you sure t_mp isn't a decimal value?
<freemangordon> it is
<freemangordon> but it is 0
<freemangordon> int the parameters
<DocScrutinizer05> comma vs dot?
<DocScrutinizer05> 0,75 ?
<freemangordon> no matter, we multiply by zero :)
<DocScrutinizer05> huh?
<freemangordon> and stock functions give the same result
<DocScrutinizer05> yes, since they also see 0,75 as 0, 75
<DocScrutinizer05> try 0.75 instead (or vice versa), for t_mp
<jusa_> freemangordon: nope..
<freemangordon> what is the idea it is integer. and it comes from direct hexascii->binary conversion
ecloud is now known as ecloud_wfh
<DocScrutinizer05> I asked if it's a var with decimals
<jusa_> and also, I don't really know of the algorithm insides, just using them.. also I wasn't the guy sitting in the lab doing measurements
<freemangordon> sorry, maybe I didn't understand your question
<freemangordon> jusa_: any idea who can help? as it seems all we know what xprot is supposed to do is rumors :)
<freemangordon> I was really expecting that temp prediction to act somehow
<freemangordon> but it looks like disabled in params
<DocScrutinizer05> well, whatever it is that you read in as parameters there, obviously they are borked
<jusa_> freemangordon: I really don't know.. we just *got* the algorithm packages, and I don't know where the guys doing the tuning are nowadays..
<freemangordon> ok
<freemangordon> DocScrutinizer05: why borked, it was done on purpose. It is just my curiosity why :)
<DocScrutinizer05> maybe because you're running a profile where XPROT is not supposed to be active?
<DocScrutinizer05> I guess you're aware that xprot only makes sense for IHF
afics_ has joined #neo900
<freemangordon> yep
<freemangordon> and that's where I got the parameters from
<DocScrutinizer05> so, what profile are you running
<freemangordon> this is what I use
<DocScrutinizer05> and why do you thibnk that's the correct set of parameters?
<freemangordon> because this is the only place where xprot is enabled
<DocScrutinizer05> this doesn't look like a "place" to me
<freemangordon> and the same set of parameters are in meego and nemo
<freemangordon> anyway, believeme when I say those are the correct parameters
afics has quit [Ping timeout: 250 seconds]
<DocScrutinizer05> I asked which profile you're running. cmt_voice->IHF, or music->IHF, or systemtones->IHF, or music->headset, or...
<DocScrutinizer05> and evidently those are not correct
<freemangordon> that's irrelevant for xprot (what the input is), it uses only its own set of parameters
<DocScrutinizer05> since you say yourself, xprot is not working
* DocScrutinizer05 headdesks
<freemangordon> oh, it works, just not predicting the coil temp
<DocScrutinizer05> [2014-11-19 Wed 09:36:46] <DocScrutinizer05> I guess you're aware that xprot only makes sense for IHF
<freemangordon> look at the link ^^^
<freemangordon> "ihf.parameters"
<DocScrutinizer05> so obviously for other scenarios / profiles XPROT has to get disabled
<freemangordon> and it is
<freemangordon> xprot.displacement and xprot.temperature are 0 in all other profiles
<DocScrutinizer05> then what's the problem?
<freemangordon> toldya, my curiosity :)
<freemangordon> why the temp predictor is disabled
<freemangordon> however, gtg to work
<freemangordon> bye, bbl
<DocScrutinizer05> maybe because of[2014-11-18 Tue 22:13:12] <DocScrutinizer05> and I wonder where the heck the developers got the whole parameters from, since... who knows how much thermal capacity a Knowles TINY coil has, how hot it mey get, how's the thermal coupling factor to environment that cools coil down, etc pp
Pali has joined #neo900
<freemangordon> could be
<freemangordon> but then - why is that algo there at all? they didn;t manage to tune it?
<freemangordon> so it simply got disabled
<DocScrutinizer05> /* this is the ueber nifty coil temperature simulation and limiting function. (disabled since doesn't work as expected) */
<freemangordon> ok
* freemangordon is afk
* DocScrutinizer05 figures a Professor Dr Dr Dr doing a 2h talk explaining all the theoretically relevant parameters that determine when and how a speaker transducer emits magic blue smoke or throws confetti of the former membrane. Then 10 sw developers transform the formulas he wrote to the whiteboard into an algorithm
<DocScrutinizer05> then they notice that they lack data for half of the formulas
<DocScrutinizer05> I should have known already when I heard "maximum membrane excursion"
<DocScrutinizer05> who the heck cares in real life?
<kerio> hold on
<kerio> why does the audio system need to know the temperature
<DocScrutinizer05> good question
<DocScrutinizer05> there's 3 possible explanations: maximum allowable power output from amp (but we got a class-D amp that doesn't get hot from high volume), maximum allowable coil energy dissipation based on thermal balance of the transducer system, and finally point 3: speed of sound in air which depends on temeprature but also on altitude rsp barometric pressure
<DocScrutinizer05> and for thermal balance of transducer coil you simply don't have all the relevant parameters, and it probably doesn't make any sense at all in the end in real life to try and calculate temperature, when all you need to do is limit the average RMS power you feed to the speaker. And it's largely irrelevant if the ambient temperature is a 20 or 30 or 60°C
<DocScrutinizer05> when the coild has a max temp of maybe 250°C
<DocScrutinizer05> coil*
<DocScrutinizer05> the fun part is: the amp can do up to almost 2W under certain circumstances (high battery voltage and sufficient cooling), but the Knowles TINY can only cope with some few hundred milliwatt
<DocScrutinizer05> so XPROT rather should get info about battery voltage than about ambient temperature
<DocScrutinizer05> btw bme has no idea about the ambient temperature anywhere next to speakers. All that bme knows is chip die temp of the BQ24150 or 27200, and *maybe* also the battery temperature from NTC on BSI contact of battery
jonwil has joined #neo900
Oksana has quit [Read error: Connection reset by peer]
<kerio> i thought the middle contact was battery capacity
<DocScrutinizer05> nope
<DocScrutinizer05> the right one is
<kerio> there's three contacts
<DocScrutinizer05> middle is GND
<kerio> ok
<DocScrutinizer05> LOL!!! I made them envy? >>It’s with a great deal of excitement that we launch our Jolla Tablet crowdfunding campaign, the world's first crowdsourced tablet. We have a vision, we have a target, and now we’re looking to you, our incredible community, to help make this happen.<<
<DocScrutinizer05> >>As you know, we’re totally different to any other company you’ll support<< MHM!
<DocScrutinizer05> sigh
<jonwil> Just found this interesting video on how advanced SMT circuit boards get made https://www.youtube.com/watch?v=bTij7Juj5qE This should show people why components need to be ordered for mass production in big lots (i.e. whole reels/trays/etc of components)
<DocScrutinizer05> >>We want you to tell us how you want your Jolla Tablet to be. It doesn’t matter who you are, you’ll get your voice heard in our community forum together.jolla.com.<< MEH!
<DocScrutinizer05> jonwil: no doubt it's relatively easy to explain to people why there's MOQ etc. But who will explain to them that sourcing can take months, or even half a year or longer? And that components may get available today out of thin air, after 12 months of unobtainium state, and in 7 days they vanish just ike they appeared
<DocScrutinizer05> jonwil: 55minutes? :-o
<jonwil> it IS a complete tour of a manufacturing plant that makes SMT PCBs and stuff
<DocScrutinizer05> hmm, I know all that stuff, so I wonder if there's any particular candy in it for me
<jonwil> its fascinating to see how a modern factory (in a country where labor is expensive and capital is cheap)
<jonwil> I bet in China some of those machines dont exist because its cheaper to use people
<DocScrutinizer05> yup
<DocScrutinizer05> hmm, no vapor phase soldering
<DocScrutinizer05> oh sheeeet >>EEVblog #678 - What is a PCB Spark Gap?<< who's using spark gaps nowadays? they are so 1950
drathir87 is now known as drathir
<jonwil> people still use them including the mob shown in that factory video
<jonwil> btw the guy that does those videos is by far the best electronics geek guy I have ever seen, really knows his stuff. He even got not only a tour of a decommissioned analog TV transmitter but he got to take parts of the old transmitter and tear them apart.
nicksydney has joined #neo900
afics_ has quit [Ping timeout: 250 seconds]
afics has joined #neo900
che1 has joined #neo900
kolp has joined #neo900
wazrus has joined #neo900
arcean has joined #neo900
kolp has quit [Remote host closed the connection]
wazrus has quit [Quit: Lost terminal]
che1 has quit [Ping timeout: 245 seconds]
Kabouik has joined #neo900
Pali has quit [Ping timeout: 265 seconds]
Pali has joined #neo900
infobot has quit [Read error: Connection reset by peer]
infobot has joined #neo900
che1 has joined #neo900
che1 has quit [Ping timeout: 272 seconds]
Pali has quit [Read error: Connection reset by peer]
che1 has joined #neo900
<freemangordon> jonwil: hi. I am finished with libxprot, any idea what else xprot related needs to be finished?
<jonwil> not sure off the top of my head :)
<freemangordon> is PA part ready? or it should be in -voice?
<jonwil> I think some of the PA stuff is ready
<jonwil> but not the temperature stuff
jonwil has quit [Quit: ChatZilla 0.9.91 [SeaMonkey 2.30/20141013232806]]
<DocScrutinizer05> intel SoC, me already starts smelling Aegis_2.0
<bencoh> do I read 603 k$ ?
<DocScrutinizer05> dunno, I closed the window without refreshing, yesterday they been at... I forgot. Some 100+ devices/pledges
che1 has quit [Ping timeout: 250 seconds]
<bencoh> $603,255USD
<bencoh> raised of $380,000 goal
<DocScrutinizer05> congrats
<DocScrutinizer05> and that with a 4700mAh battery
<DocScrutinizer05> so they actually already got ~2000 preorders? sounds like great success for their project
<bencoh> a bit depressive too, but yeah
<bencoh> depressing*
<DocScrutinizer05> depressing? why?
<bencoh> comparing that to neo900 "preorders"
<DocScrutinizer05> because donors again fall for the "we're different, we're oh so open and free" bait?
<bencoh> yea
<bencoh> exactly
<DocScrutinizer05> see my rant above
<DocScrutinizer05> I'm quite sure community again has not a single vowel and for sure no word of saying in decision how the HARDWARE will look like, and which parts of bootloader are open
<DocScrutinizer05> and we seen on HARM how you provide an "open" linux that nevertheless cannot get altered or simply just messed around a tiny bit with e.g. config and initscripts
<DocScrutinizer05> btw the above photo has one interesting detail: a bright pinkish LED that no human eye can see, it's the proximity detector IR LED
<bencoh> :)
<DocScrutinizer05> I'd probably even buy one of those jollblets IFF they would provide sufficient info about the mere hw so I could install my own OS on it
<DocScrutinizer05> no computer hardware in my house where I cannot install a dual boot system at least
<DocScrutinizer05> sailfish may be nice and all, but the last thing I want is to feel locked in on that one OS
<DocScrutinizer05> so either enough docs so I can install a dual boot with true linux as my main and primary OS, with full access to the complete user data of the sailfish OS, or they can keep their device
<DocScrutinizer05> and please don't tell me stuff like "you can use linux, but then it won't charge battery, and you cannot revert to sailfish clean without reflashing"
<DocScrutinizer05> and for sure I'm not even sending in my hardware for reflash even when hell is freezing
<DocScrutinizer05> a device I cannot bootstrap-install from "brick" to "100% operable" is broken by design, and so far I can't see any statement regarding those details of the jollblet
<DocScrutinizer05> so they won't see a pledge from me. Did once, got hurt once
<DocScrutinizer05> I just hope they learned their lesson, the FOSS world would need it
* DocScrutinizer05 glares at jolla phone collecting dust
* DocScrutinizer05 notices N9 and N950 right next to jolla phone collecting almost just as much dust
* DocScrutinizer05 wonders when was last time the PC had a 24/7 opened ssh over WLAN opened to N9(50), while T900 next the the other 3 devices, while collecting 3 times as much dust as all three of the others, is actually "used" every single day, since several years now
<DocScrutinizer05> t900:~# uptime
<DocScrutinizer05> 17:22:28 up 174 days, 6:28, load average: 0.00, 0.00, 0.00
<DocScrutinizer05> t900:~# uname -a
<DocScrutinizer05> Linux t900 2.6.28.10power46 #1 PREEMPT Sun Dec 12 03:11:24 EET 2010 armv7l unknown
<DocScrutinizer05> ;-D
che1 has joined #neo900
RzR is now known as rZr
* DocScrutinizer05 idly wonders if we could _contract_ Jolla to do things for us. Like building new case incl display and kbd, etc
<DocScrutinizer05> they must have brought quite a nomber of nice industry contacts with them from Nokia
<DocScrutinizer05> number*
louisdk has joined #neo900
che1 has quit [Ping timeout: 258 seconds]
<DocScrutinizer05> ok, I would almost like to say "I knew it". Trying to disassemble my N9, it turned out that the internal TX-4 screws are so massively tightened that a alost 10EUR worth TX-6 high quality screwdriver was only able to damage 2 of them so now they sit in there and I have to come up with some way to get them out
<DocScrutinizer05> :-((
<DocScrutinizer05> also the steel shield that is fixed by all those scres has several tiny very weird looking welding marks all around parts of the screw holes :-o
che1 has joined #neo900
<ds2> prehaps they are loctite'd?
<DocScrutinizer05> dunno, several that opened were very tight as well
<DocScrutinizer05> what really niffs
<DocScrutinizer05> err
<DocScrutinizer05> what really pisses me off is that I damaged my 45EUR kipex cutter pliers
<DocScrutinizer05> when trying to somehow deal with those screw critters
<ds2> could be loctite
<DocScrutinizer05> knipex actually
<ds2> prehaps try heating it briefly with a soldering iron?
<DocScrutinizer05> yeah, thought about that
<DocScrutinizer05> >>[2014-11-19 Wed 17:05:07] <DocScrutinizer05> 17:22:28 up 174 day<< ouch! the sysclock is some 20 minutes off :-S
<DocScrutinizer05> or wait, is that the sub-day fraction of uptime?
<DocScrutinizer05> nope
<DocScrutinizer05> seems I should run openntpd on that one
<DocScrutinizer05> or rather a "cronjob" doing netdate once a week
* DocScrutinizer05 heats up the evil soldering station. 400°C
Pali has joined #neo900
<DocScrutinizer05> yeah, red locktite or whatever
<DocScrutinizer05> but heating doesn't help much, since you cannot try to drive out the screw and same time heat it up
<DocScrutinizer05> I guess I have to drill those suckers out. Or cut the steel sheet
<DocScrutinizer05> thank you whoever had that brilliant idea of lock-gluing the screws
<DocScrutinizer05> btw new N9 still go for >>400EUR
<bencoh> wtf
<DocScrutinizer05> well, seems madness isn't selective on where it shows up: http://www.ebay.com/itm/Nokia-N-Series-N900-32GB-Black-Unlocked-Smartphone-/310733109557
<bencoh> huhu well ...
<bencoh> and that's a HK one
<Humpelstilzchen> nice finally an unlocked n900
<Humpelstilzchen> uhm what ever you are searching for..
Kabouik has quit [Ping timeout: 258 seconds]
<DocScrutinizer05> is that an ebay promo?
sparetire_ has joined #neo900
Kabouik has joined #neo900
<DocScrutinizer05> bencoh: ((depressing etc)) Lumia has 120 employees now according to their press release, and yet they outsource hw R&D to some hw-house in china afaik. We are basically ~3, and we do all hw R&D inhouse
<DocScrutinizer05> s/lumia/jolla/ ;-P Freudian
<useretail> in full hd!
<bencoh> DocScrutinizer05: hw r&d or hw sourcing ?
<DocScrutinizer05> while dragonbox and Neo900 go direction WWAN/phone, jolla boggles and does a jollbletwithout any cellular, after starting with a supposedly LTE phone, which at least to me is unclear if it meanwhile can do LTE in all the regions it got sold to
<DocScrutinizer05> they get complete jolla phone hw by outsourcing R&D and manufacturing to whomever
<DocScrutinizer05> afaik
<bencoh> bleh
<DocScrutinizer05> maybe jolla specified the outer look of case, and the display site and resolution
<DocScrutinizer05> size*
<bencoh> they chose the soc as well
<DocScrutinizer05> I doubt that
<bencoh> and a few other things
<DocScrutinizer05> I guess they took what got offered to them
<bencoh> thats what stskeeps said iirc
<DocScrutinizer05> prolly Foxcon offered "build phone based on our standard design 3C, with modifications as specified by user, listed below in points 3 to 12"
<DocScrutinizer05> s/user/customer/
che1 has quit [Ping timeout: 244 seconds]
<DocScrutinizer05> I talked with several of the Jolla guys about working for them, the final info been "there's not even a hardware R&D department in our company, so no position for an EE existing inside jolla whatsoever"
<DocScrutinizer05> doesn't really matter if jolla aske d 20 chinese hw-houses "what can you offer for LTE?" or if they asked "can you offer LTE snapdragon?"
<DocScrutinizer05> the latter just a tad worse still than the former
<DocScrutinizer05> since jolla, consisting of mainly sw engineers, should have known about the issues with baseband firmware part of the user storage area and thus the whole device gets cumbersome to flash by users
<DocScrutinizer05> with snapdragon
<useretail> o_O didn't knew that jolla doesn't have hw dept
<DocScrutinizer05> well, that's what guys told me. I can only quote that
<freemangordon> 500$ for n900?!?
<DocScrutinizer05> actually I forgat it myself when a few hours back I pondered if Jolla could do tasks for Neo900 UG
<DocScrutinizer05> freemangordon: well, that seems pretty much the original unreduced price for original unopened package
<freemangordon> yep, looks like... but still, it is 2014
<DocScrutinizer05> or wait, isn't it already 30% discounted?
<useretail> freemangordon: yeah, don't tell me you haven't saw that iphone 2g for 6500$
<freemangordon> I haven't seen it
<useretail> sometimes stuff on ebay is sooo overpriced
<freemangordon> ~seen jonwil
<infobot> jonwil <~jonwil@27-33-80-219.tpgi.com.au> was last seen on IRC in channel #neo900, 5h 22m 20s ago, saying: 'but not the temperature stuff'.
<freemangordon> damn time zone difference :(
<freemangordon> DocScrutinizer05: I don;t have tools for visualizing audio data, would you do it for me if I put raw files on my web server?
<DocScrutinizer05> well, all i have is audicity
<DocScrutinizer05> but yeah, maybe I find a timeslot to do
<DocScrutinizer05> dang audicity/PA pretty much raped me yesterday, when I had kmixer open when deleting a rule in audacity, which froze complete KDE window handling
<DocScrutinizer05> since mixer been covered by audacity and audacity didn't take focus to close
<DocScrutinizer05> so very funny deadlock I only been able to solve with a killall audacity on console
<useretail> that's why i prefer gnome
<freemangordon> not that it doesn't have bugs
<useretail> agree
<DocScrutinizer05> well, sorry. I did a vow that I don't use any software suite starting with "g" ;-)
<DocScrutinizer05> gnome, gconf, gvfs, g* - all makes my toenails roll up
<freemangordon> DocScrutinizer05: not bad, eh? http://pastebin.com/nULHizBY
<DocScrutinizer05> SuSE once switched from KDE iirc 2, to gnome as default desktop, which been when I tried that stuff for a 2 or 3 days. I been faster than SuSE to switch back to KDE back when ;-)
<DocScrutinizer05> freemangordon: nice
<DocScrutinizer05> (faster than SuSE) but only a few months ;-)
<DocScrutinizer05> I heard they now adopted gnome again, for their SLE* stuff
<DocScrutinizer05> which might actually make some sense, given the unbearable monotone decay of reliability and quality in KDE
<DocScrutinizer05> all the more funny that SuSE SLE* ships standard browsers: FireFox and Konqueror ;-P
<useretail> never tried suse - i've been using debian all the time
<DocScrutinizer05> I boggled form migrating to SLED when I read it also already uses systemd and doesn't support KDE read: Kontact, anymore
<freemangordon> DocScrutinizer05: http://46.249.74.23/xprot/
<freemangordon> test.raw is 32bit, testout is 16
<freemangordon> oh, wait, my bad
<freemangordon> wait a minute
<DocScrutinizer05> hmm
<DocScrutinizer05> just wanted to ask
<DocScrutinizer05> not that it makes much difference, but actually sample rate and format would help to
<DocScrutinizer05> I can already tell you that this test.raw is not suited to test xprot
<freemangordon> now out file is the correct one
norly has joined #neo900
<DocScrutinizer05> -30dB, how#s that relevant for xprot?
<freemangordon> which file is that?
<freemangordon> input or output
<DocScrutinizer05> play -n -c2 synth 0:0:30 pinknoise synth trapezium amod 2 0 0 40 100
<freemangordon> how to redirect the output to a file?
<DocScrutinizer05> use sox instead play, append "outfile" parameter
<freemangordon> with the same parameters?
<DocScrutinizer05> yes
<DocScrutinizer05> sox [global-options] [format-options] infile1
<DocScrutinizer05> [[format-options] infile2] ... [format-options] outfile
<freemangordon> ok
<DocScrutinizer05> play [global-options] [format-options] infile1
<DocScrutinizer05> [[format-options] infile2] ... [format-options]
<DocScrutinizer05> mn sox
<DocScrutinizer05> man*
<DocScrutinizer05> you'll probably want to specify outfile format
<DocScrutinizer05> >> sox recital.au -b 16 recital.wav channels 1 rate 16k fade 3 norm
<DocScrutinizer05> performs the same format translation, but also applies four effects (down-mix to one channel, sample rate change, fade-in, nomalize), and stores the result at a bit-depth of 16.
<DocScrutinizer05> >> sox -r 16k -e signed -b 8 -c 1 voice-memo.raw voice-memo.wav
<DocScrutinizer05> converts `raw' (a.k.a. `headerless') audio to a self-describing file format
<freemangordon> I guess rec is what is needed
<DocScrutinizer05> rec records from a audio source, like default microphone
<DocScrutinizer05> but yeah, to record output of XPROT you may use | rec foobar.wav
<freemangordon> ok, I think I did it
<DocScrutinizer05> sorry, afk, need dinner
<freemangordon> sox -n -r 48000 -b32 -c2 xprot.raw synth 0:0:30 pinknoise synth trapezium amod 2 0 0 40 100
<freemangordon> DocScrutinizer05: ready, on the same url
<freemangordon> reuploaded the out file, only half of the samples were processed in the first one, now it is ok
che1 has joined #neo900
Openbot has joined #neo900
Openbot has quit [Ping timeout: 264 seconds]
Oksana has joined #neo900
Oxyd76 has quit [Ping timeout: 250 seconds]
Oxyd76 has joined #neo900
<Oksana> ((don't use any software suite starting with "g")) What about gconftool-2 on Fremantle? | ((TX-4 screws damaged by TX-6 high quality screwdriver)) No TX-4 screwdriver close by? | ((ambient temperature is a 20 or 30 or 60°C)) Well, somebody insane could hold the device over a boiling teapot (close to 100°C), but it would be crazy considering that many components do not tolerate temperatures...
<Oksana> ...higher than 85°C.
<Oksana> Good morning
<Oksana> In MCP, I understand RAM, but I do not understand NAND. Read-only memory for images or music? What for?
Oxyd76 has quit [Remote host closed the connection]
Oxyd76 has joined #neo900
Oxyd76 has quit [Remote host closed the connection]
Oxyd76 has joined #neo900
<DocScrutinizer05> huh?
<DocScrutinizer05> s/TX-6/TX-4/
<DocScrutinizer05> I'm not THAT silly ;-)
<freemangordon> DocScrutinizer05: did you see my last post ^^^ ?
<DocScrutinizer05> I even doubt you could damage TX4 screws with a TX-6 driver
<DocScrutinizer05> yes
<freemangordon> ok
<DocScrutinizer05> alas in a hurry
<freemangordon> definitely xprot does something, sounds like high-pass filter
<freemangordon> ok
wazrus has joined #neo900
<DocScrutinizer05> HPF makes a damn lot of sense, been what we already expected to see
<DocScrutinizer05> given we have a class-D amp with f0 ~ 10Hz or lower
<freemangordon> yep
<freemangordon> but I *guess* that band depends on the volume
<freemangordon> maybe I should make a pack of output files with different volume setttings. someday :)
<DocScrutinizer05> that's why the pink noise is ramped up in that test file
<DocScrutinizer05> you know the uniform "pink" input spectrum and the volume, and you can analyze output with FFT
<freemangordon> naah, I'm too old for this shit :P (FFT)
<freemangordon> I have forgotten almost everything
<DocScrutinizer05> audacity does it for you, among others
<freemangordon> hmm, maybe I should install it
<DocScrutinizer05> alas it froze a second time already, when I tried to analyze the initial (wrong?) test.raw files
<freemangordon> hmm, it should be xprot.raw
<DocScrutinizer05> maybe I should update it
<DocScrutinizer05> afk, bbl
<freemangordon> and xprotout.raw
<DocScrutinizer05> [2014-11-19 Wed 20:58:39] <freemangordon> DocScrutinizer05: http://46.249.74.23/xprot/
<DocScrutinizer05> [2014-11-19 Wed 20:59:03] <freemangordon> test.raw is 32bit, testout is 16
<freemangordon> yep, but in the meanwhile I changed the files
<DocScrutinizer05> I know
<freemangordon> ok
<DocScrutinizer05> alas it froze a second time already, when I tried to analyze the ***initial (wrong?)*** test.raw files
Oxyd76 has quit [Remote host closed the connection]
<freemangordon> no
xes has joined #neo900
<freemangordon> there are xprot.raw and xprotout.raw
<freemangordon> something's wrogn with your browser
<freemangordon> try to refresh :)
Oxyd76 has joined #neo900
Oxyd76 has quit [Ping timeout: 250 seconds]
<DocScrutinizer05> I know
<DocScrutinizer05> just showing you that you're too slow to fix errors before I go active. Sometimes
<freemangordon> sorry about that :)
<DocScrutinizer05> your >>[2014-11-19 Wed 20:59:43] <freemangordon> oh, wait, my bad<< came too late and without highlight, so I didn't see it since I already downloaded the stuff and started analyzing it
<freemangordon> anyway, fft looks weird
<freemangordon> it has a peak at about 2000 hz
<DocScrutinizer05> that's the FFT of your weird music
<freemangordon> no, the fft of xprotout.raw. I installed audacity
<DocScrutinizer05> aaah
<DocScrutinizer05> great :-)
<DocScrutinizer05> o/ bbl
<freemangordon> but I still would like you to look at it as well
<DocScrutinizer05> 2kHz .. 20kHz they are identical, below the HPF kicks in. Looks good to me
jonwil has joined #neo900
<freemangordon> :nod:
<freemangordon> jonwil: hi!
<jonwil> hi
<freemangordon> xprot algo seems like ready :)
<jonwil> neat
<jonwil> whats next
<jonwil> ?
<freemangordon> jonwil: you tell me :)
<Oksana> :-)
<freemangordon> you played with PA parts
<freemangordon> maybe finish PA part of xprot?
<jonwil> I think the temperature stuff in the PA bits is not done yet
<jonwil> but the other part of xprot should be
<freemangordon> bme stuff?
louisdk has quit [Remote host closed the connection]
<freemangordon> jonwil: BTW, in case you're curious what exactly xprot does: http://46.249.74.23/xprot/
nicksydney has quit [Remote host closed the connection]
<jonwil> yes the temperature stuff that talks to bme
<freemangordon> ok, will look at it. do you have any progress with the big struct?
<jonwil> nope
<jonwil> been busy
<freemangordon> ah, ok
<freemangordon> however, /me needs sleep
<jonwil> after the xprot stuff, the eq stuff makes sense to do next
<freemangordon> jonwil: yep
<freemangordon> will you have time to look at the PA bits soon?
arcean has quit [Quit: Konversation terminated!]
<jonwil> dont know
<jonwil> probably not before Wednesday next week at the least
<jonwil> I have other important non-code stuff to do in the next few days
<freemangordon> sure, no hurry
Oxyd76 has joined #neo900
<Oksana> Just donated. Sent email, too.
wazrus has quit [Quit: Lost terminal]
wicket64 has joined #neo900
nox- has joined #neo900
xes_ has joined #neo900
xes has quit [Ping timeout: 250 seconds]
xes_ has quit [Client Quit]