<kristianpaul> hehe
<kristianpaul> ha !!
<kristianpaul> thansk to Clifford Kelley, i confirm
<kristianpaul> 1). I'm using same compresion format (two  4 bits in a byte)
<kristianpaul> 2). Decimation is not needed, or can even be considered
<kristianpaul> 3). Byte orientation is bytewise
<kristianpaul> Now i should take a look at the  correlation subroutine
<kristianpaul> fyi osgps is the only "soft" correlator that dint ask me for SIMD :-)
<rm> hello all
<kristianpaul> hi
<kristianpaul> roman?
<rm> yes :)
<rm> kristianpaul, do you also work on nanonote?)
<kristianpaul> yup :-)
<rm> I wanted to ask, does anyone know if there are Desktop PC (nettop?) available, which would use the XBurst SoC
<kristianpaul> i think there are
<kristianpaul> but not freed
<kristianpaul> I remenber tuxbrain have one?
<kristianpaul> rm: ie http://ur1.ca/3oj1r
<kristianpaul> but i dont remenber if is same as one from tuxbrain
<rm> I do not see such a thing at tuxbrain
<rm> kristianpaul, I mean not netbook
<kristianpaul> btw there are other people from russia here
<kristianpaul> rm: ah sorry, what you mean then?
<kristianpaul> ah nettop
<rm> but a box that would only connect to external monitor+keyboard+mouse
<kristianpaul> hmm
<kristianpaul> dunno :/
<kristianpaul> may be ask ingenic directly?
<rm> maybe
<rm> but that one is on a different CPU, not ingenic
<rm> and it uses a proprietary GPU
<rm> requiring either 'many megabytes' of proprietary code, or using just the fbdev
<kristianpaul> he, well ingenic is not very about his video aceleration in his soc either ;)
<rm> heh
<rm> and our yeeloong doesn't *have* any acceleration to begin with
<rm> so no problem here :D
<kristianpaul> ;-)
<rm> ok, I'll idle here for a bit, while I'll go get some sleep
<kristianpaul> sure
<kristianpaul> you are wellcome !
<rm> bye :)
<kristianpaul> bye
<kristianpaul> ha,!! first time osgps run and seems to analize something :D
<kristianpaul> rm: if you want a more close free cpu, you can buy a Milkymist One ;)
<kristianpaul> also it have vga out, mouse and keyboard if you need it :-)
<rjeffries> kristianpaul and others: how difficult to port Nanonote system to http://ur1.ca/3oj1r maybe ignoring wifi at first
<kristianpaul> dunno
<rjeffries> rm the Seagate Dockstar (no longer manufactured) use Marvel SOC (ARM) has good linux support and is cheap(!!)
<kristianpaul> 1 year more?..
<kristianpaul> same as nanote if you have similar people behind the effort
<kristianpaul> moment, a sample rate is defined by a TCXO..
<kristianpaul> ?¿
<kristianpaul> in Mhz and not in MSPS..
<kristianpaul> hmm sample clock
<kristianpaul> thats sound better
<kristianpaul> sample rate clock
<wolfspraul> DocScrutinizer: he :-) welcome back!
<DocScrutinizer> o/ :-D
<wpwrak> DocScrutinizer: heya, been a long time ! :)
<DocScrutinizer> yeah, adore my uptime. The channel is still on autojoin
<wpwrak> wow, then it's mightly long indeed :)
<DocScrutinizer> 07:34am  up 185 days 15:07
<DocScrutinizer> feels like I never moved my fat arse outa this seat since
<DocScrutinizer> s/fat/skinny/
<wpwrak> started skinny, now it must be too fat to move. do you have acquired moons already ? :)
<DocScrutinizer> the pizza service here is abysmal, so no chance to grow fat(ter) while watching this idiotic LCD
<DocScrutinizer> sighs when reading "gta04/diversity" in http://lists.openmoko.org/pipermail/openmoko-kernel/2008-June/002987.html
<wpwrak> DocScrutinizer: better read the tail-between-the-legs statement that "you can't 'make' hardware" and then see how far we've come already
<DocScrutinizer> who stated that?
<DocScrutinizer> and who's "you"? ;-D
<wpwrak> DocScrutinizer: sean had a presentation on openmoko scheduled for i think it was just after the first bloodletting. at a conference in switzerland, if i remember right. and there he made a statement in that sense. i probably got the wording wrong, though.
<DocScrutinizer> aah, by "you" he meant OM mgmt ;-P
<DocScrutinizer> I.E. self()
<wpwrak> with the implied assertion of universality. it's all Kant in a way ;-)
<DocScrutinizer> it's never been the 'making' that's the problem. Always the 'marketing' though
<wpwrak> naw, marketing must have been easy, with all the hype surrounding openmoko. that was really incredible. wish we had half that for qi-hw.
<DocScrutinizer> qi-hw is missing the intriguing topic of "smartphone"
<wpwrak> oh, today it wouldn't get so much interest either
<DocScrutinizer> well, maybe qi is ahead of time, next buzzword been "tablet"
<DocScrutinizer> :-D
<wpwrak> yeah. let's make nanonote and milkymist the next big thing after the tablet
<DocScrutinizer> wearables - if I may dare a prophecy
<wpwrak> that may not be a bad guess. we're certainly heading this way.
<DocScrutinizer> qi should be at the exactly right location for this enterprise. Isn't all western clothing manufactured in far east? (yeah, or Bangladesh)
<wpwrak> wherever it is, the next qi-hw representative ain't far ;-)
<DocScrutinizer> lol
<DocScrutinizer> Qi-HW, Berlin London, New York, Taipei, Bangalore
<wpwrak> it gives me the chills though that i'd be the closest representative for antarctica. well, until rafa comes back
<wpwrak> barcelona, colombia, ...
<DocScrutinizer> seems even 'big players' only have EMEA, America, Asia
<wpwrak> we don't seem to have much coverage in africa and au/nz either, though
<DocScrutinizer> no coverage in Ozz? how's that?
<DocScrutinizer> quite unusual, I'm always amazed how many users are from there
<DocScrutinizer> curses missing regex grep on userlist of IRC client
<DocScrutinizer> o/ roh
<wpwrak> dunno. just haven't noticed anyone from down under
<wpwrak> DocScrutinizer: i think kyak is now the official IRC bot enhancer :)
<DocScrutinizer> aah, you think like
<DocScrutinizer> qi-bot: seen *au*
<wpwrak> has a nice ring to it, no ?
<DocScrutinizer> hihi cttv tests RL-twitter, with megaphone down in the streets. Strange, it doesn't work
<wpwrak> anyway, time for me to crash for a bit. had a busy day.
<DocScrutinizer> enjoy
<DocScrutinizer> time for 12648430  
<DocScrutinizer> here
<DocScrutinizer> (I'd love to design a few wearables together with qi-hw :-D )
<DocScrutinizer> meh, qi-bot is quite uncommunicative
<kyak> :)
<LunaVorax_mini> erp ?
<DocScrutinizer> kyak: nice :-D
<DocScrutinizer> kyak: the plan however was to find australian users in this chan. Seems it's not exactly simple to establish that using the !seen *xy* cmd
<DocScrutinizer> kyak: of course I tried "help" (plus a few others) in a /query qi-bot, which resulted in [2011-03-27 08:46:16] <DocScrutinizer> meh, qi-bot is quite uncommunicative
<DocScrutinizer> [So 27. März 2011] |[08:45:15] <DocScrutinizer>| help
<DocScrutinizer> [So 27. März 2011] |[08:45:22] <DocScrutinizer>| ?
<DocScrutinizer> [2011-03-27 10:33:53] <qi-bot> ... All commands can sent to me in pm without "!" prefix.--
<DocScrutinizer> I.E. this isn't exactly true, 'help' doesn't work this way
<tuxbrain> DocScrutinizer:  (wearables) Yes!, also very in line with Arduino next move :), David Cuartielles one of his founders, has also near to publish a book about it :) , so some atmegas arround jacket and trousers and a brain/hmi based on qi-hw is not really a bat a idea at all
<DocScrutinizer> tuxbrain: :-D
<DocScrutinizer> I'm more interested in the 'embeeded' (enweaved) things though, not the hmi/mmi
<DocScrutinizer> ideally there's no such thing like a dedicated device for mmi. Just a 'service terminal' you hook up at home, to configure things
<tuxbrain> DocScrutinizer: and why not configure it directly on the ankward of you foreaarm with a wearable screen that make you able to do so on the go , and with an 6lowpan that make you able to even configure it with your desktop computer without need to plug anything? :)
<tuxbrain> or even better a cute light glasses to view and reserve the forearm for control/typing :)
<DocScrutinizer> tuxbrain: yeah, of course.
<DocScrutinizer> reminds me to file a few patents for my special glasses
<DocScrutinizer> damn, that's where hw-development starts to get annoyingly expensive and unfeasible for a mere mortal
<DocScrutinizer> special build of LCDs etc
<DocScrutinizer> possibly even non-standard LC itself (not that you'd be able to get such thing like "standard" LC in small quantities)
<DocScrutinizer> not to mention the glass electrodes coated with transparent conductive material
<DocScrutinizer> spacers, whatnot
<kyak> DocScrutinizer: --^ :)
<DocScrutinizer> nice
<DocScrutinizer> scribbles on a virtual postit note "install eggdrop" and sticks that note to some part of his neocortex
<tuxbrain> DocScrutinizer: please translate all above tech mambo jambo to an aprox amount.
<tuxbrain> and if you have a abstract of that idea please let me take an eye.
<DocScrutinizer> the simple bottom line: you actually can not build your own LCD
<DocScrutinizer> (as you'd need e.g. for a shutter glasses' glasses, or would you like to use goggles with two pcs 2.4" LCDs as glasses?)
<tuxbrain> Doc I'm a little bit lose on the vocabulary shutter? googgles? any let me search to see what you mean
<tuxbrain> googgles : mmm maybe something more integrate that this but with same spirit whould be no so bad idea http://f00.inventorspot.com/images/smartgogglesJSH_228x342.jpg
<DocScrutinizer> yeah, and the point is you can't built the needed shutter LCDs yourself. I'd have some rather smart (IMHO) ideas regarding combining shutter function, HMD (actual content mirrored in to the glasses via micro projector), and some ultra-geeky sunglasses function like selective dimming.
<DocScrutinizer> ~dict goggles
<DocScrutinizer> oops, no ibot here
<DocScrutinizer> !dict goggles
<DocScrutinizer> thought as much
<DocScrutinizer> tuxbrain: yeah, the last one is what my idea is based on
<DocScrutinizer> e.g. camera can detect direction of sun and selectively dim out that area in the goggle's shutter glasses
<DocScrutinizer> you can have true alpha channel in content mirrored into your view
<DocScrutinizer> get the picture?
<tuxbrain> I agree to , we can found a way that it can be addapted to existent glasses too? maybe without some ot the fancy fucntions like auto dark glass, just show me info device?
<tuxbrain> I have found the page of the guy of the googles :) http://www.isi.t.u-tokyo.ac.jp/~nakayama/index_en.htm
<tuxbrain> I'm thinking to aks him if there is more techinical information about the hardware he uses, DocScrutinizer you think it worth the meaning?
<DocScrutinizer> sure
<DocScrutinizer> I'm *very* interested in HMDs
<DocScrutinizer> and shutter goggles, and related topics
<DocScrutinizer> wow, that's incredible: http://www.isi.t.u-tokyo.ac.jp/~nakayama/annotation_en.htm
<tuxbrain> the last one, scotland that is awesome (of course if is a scottish landscape :P)
<rejon> hi
<DocScrutinizer> hard to find good links, even harder to find vivid sites actually selling
<DocScrutinizer> >>wearable computing: Wearable computing hopes to shatter this myth of how a computer should be used. A person's computer should be worn, much as eyeglasses or clothing are worn, and interact with the user based on the context of the situation. With heads-up displays, unobtrusive input devices, personal wireless local area networks, and a host of other context sensing and communication tools, the wearable computer can act as an
<DocScrutinizer> intelligent assistant, whether it be through a Remembrance Agent, augmented reality, or intellectual collectives. (from the MIT Wearable Computing WWW page)<<  hardly anything to add - from http://www.hitl.washington.edu/scivw/visual-faq.html
<GNUtoo|bug20> does the arduino lilipad count in wearable computing?
<DocScrutinizer> http://www.vrealities.com/addvisor150.html not quite stylish, but meh. It's for the operation principles, for stylish see prev link
<DocScrutinizer> (the inition glasses)
<tuxbrain> GNUtoo|bug20: lilipad I think yes it is :)
<GNUtoo|bug20> bbs
<GNUtoo|bug20> bt kbd issues
<kristianpaul> hi
<rejon> nice post by xiangfu: http://www.openmobilefree.net/?p=875
<tuxbrain> What is the amount to create a free production Microoptical EG7 device?!!!
<rejon> tuxbrain ha
<kristianpaul> hmm, i need more data but i think i got the peak for two PRN codes..
<DocScrutinizer> tuxbrain: you need the display, the optics, the case. The rest is pfff
<kristianpaul> hmm, i need to fix the offset frequency for my particular clock.. so i need will need automate this calibrationa at some point..
<kristianpaul> How do usually is measured a clock  freq error?..
<kristianpaul> looks at wpwrak
<kristianpaul> I can put the scope there and see what freq i get, but dunno if this is the best procedure
<kristianpaul> hi DocScrutinizer :-)
<kristianpaul> gets an update almanac
<DocScrutinizer> hi kristianpaul
<DocScrutinizer> see man hwclock and man /etc/adjtime
<DocScrutinizer> you can't check clock freq base with a scope
<kristianpaul> what device i need?
<DocScrutinizer> the clock itself and a reference :-D
<DocScrutinizer> the way hwclock is doing that is the most accurate way to determine the actual exact freq of the clock osc
<DocScrutinizer> except if you're talking about sth like 20% off
<DocScrutinizer> which I guess is impossible you'll ever run into such a problem
<kristianpaul> the clock i need measure is a TCXO for a gps L1 rf front end
<DocScrutinizer> uhh, no idea
<DocScrutinizer> use a counter, connect it and probe for one hour
<kristianpaul> but according to the datasheet this is +- 1.5ppm
<kristianpaul> okay
<DocScrutinizer> *exactly* one hour. Then device the counts / 3600
<DocScrutinizer> devide*
<DocScrutinizer> as making sure the *exactly* one hour prerequisite might get tricky, you may want to probe for exactly 1 day, maybe 1 week ;-D
<DocScrutinizer> so an inaccuracy of 0.5s for the probing timespan has less impact
<steve|m> doesn't the TCXO of GPS-receivers automatically calibrate to the GPS clock, like GSM phones do as well?
<steve|m> some frequency counters have a GPS receiver as clock reference
<DocScrutinizer> yeah, afaik it's supposed to do
<kristianpaul> steve|m: yes, mine said +- 1.5ppm, but for the *front-end*, the receiver, well thats my job ;)
<steve|m> what's the frequency of the tcxo you're talking about? it's something odd, right?
<steve|m> 10.52?
<kristianpaul> no
<kristianpaul> 16.384 Mhz
<kristianpaul> DSB321SDA_ZJ01092
<steve|m> hm, too bad.. I used a phone and osmocom-bb to calibrate my Racal GSM Test Set, by 'stealing' the clock of a commercial cell
<steve|m> had both clocks on my scope, triggered on one, and calibrated it until the second waveform wasn't moving anymore ;)
<kristianpaul> hehe
<DocScrutinizer> steve|m: cool
<kristianpaul> yup
<steve|m> do commercial gps receivers have the same frontend clock you could tap?
<kristianpaul> i have one with sirf chip i dunno
<DocScrutinizer> steve|m: even easier would've been to use one for X and the other for Y
<DocScrutinizer> then calibrate to get a lissajous that doesn't move
<steve|m> true
<steve|m> well, it kept moving a bit back and forth, was quite funny..
<steve|m> everytime when the AFC-DAC was updated
<DocScrutinizer> http://de.wikipedia.org/w/index.php?title=Datei:Lissajous.gif&filetimestamp=20070904162707
<DocScrutinizer> tuxbrain: http://www.vrealities.com/vrprost.html 10.500.- hehe
<wpwrak> DocScrutinizer: (lissajous) nice ! finally a good use for xy mode :)
<wpwrak> kristianpaul: if you can just count cycles, you can synchronize with ntp. ntp has very little drift but there's a possibly large random-ish offset when you take the sample. a few ppm for a measurement period of ~1h should be feasible, though.
<DocScrutinizer> wpwrak: yeah, but still you need a reference that's not too far off from what you try to calibrate
<wpwrak> DocScrutinizer: only if you're in a hurry :)
<DocScrutinizer> e.g. here in EU we got dcf77 which is a 77.5(?)kHz RF frequency and time normal
<wpwrak> yeah, dcf77 is nice
<DocScrutinizer> you can't get useful lissajous with that, for e.g 16Mhz
<wpwrak> ah, sure. i wasn't talking about lissajous. i was thinking of counting cycles over one ntp hour, then see how many you got, and tune accordingly. repeat until you either have it right or until you declare defeat in before overwhelming temperature drift :)
<DocScrutinizer> :-D
<rm> I wish nanonote would have built-in wifi :/
<wpwrak> wpan ! wpan ! wpan ! :)
<DocScrutinizer> gigs thru his boxes in attic to find the WPAN AP
<steve|m> who said jehova? ;)
<kristianpaul> rm, wpwrak already develop an copyleft alternative to clasic wifi (wpan)
<wpwrak> DocScrutinizer: you have to look in those labeled "future"
<DocScrutinizer> what's the name of those pills?
<rm> or bluetooth (but bluetooth is ewwww)
<rm> and also highly proprietary, no
<rm> point is, if it had wireless communication out of the box, it'd be an order of magnitude more attractive/useful
<DocScrutinizer> that's the point
<rm> (i am not a nanowar fan, just some I found)
<kristianpaul> also highly proprietary as wifi, same our lemote have ;)
<DocScrutinizer> suggests a few "embedded USB bays" for future devices of qi-hw
<rm> kristianpaul, afaik bluetooth requires licensing fee per device sold
<rm> does wifi?
<rm> I always thought that's why most cheap wireless mice/keyboards use their own thing instead of bluetooth
<kristianpaul> i dunno, ask to the wifi chip manufacturer
<kristianpaul> bu i guess it need at least to pay for the MAC addressing same as Ethernet
<wpwrak> kristianpaul: the MAC is cheap. and you need it for almost anything
<wpwrak> kristianpaul: well, you actually don't get a MAC but an OUI. then you manage your own MAC address space under that OUI
<wpwrak> kristianpaul: plus, they money has already been spent: http://en.qi-hardware.com/wiki/IEEE_OUI_assignments
<kristianpaul> btw this coming wpan linux support hopefully will allo to hookup a wpan to my linksys router isnt? same as with the RFM12B module
<steve|m> wpwrak: heh, it's already in the wireshark OUI-database as well :)
<kristianpaul> haha
<wpwrak> kristianpaul: does your linksys have usb host ?
<wpwrak> kristianpaul: or at least spi / gpios for bit-banging ?
<kristianpaul> i WILL *not* use USB :-D
<kristianpaul> wpwrak: yes spi :-)
<wpwrak> who "*not*" usb ?
<wpwrak> s/who/why/
<kristianpaul> I consider un necesary
<kristianpaul> for the troughput you may get
<wpwrak> (spi) then you'll be able to hack an atben to do your bidding :) well, you'll also need a few more gpios for control signals. reset (semi-optional), standby (optional), interrupt
<kristianpaul> hmm
<wpwrak> the advantage of usb would be that you could just plug in an atusb :)
<kristianpaul> thast true
<wpwrak> (and increase tuxbrain sales in the process :)
<kristianpaul> oh sure :-)
<rm> what games would you suggest
<rm> to try on "simple" devices like nanonote (and for me right now on yeeloong)
<kristianpaul> jlime muffinman ships nice games
<kristianpaul> there is also a #jlime channel in wich you can get better recomendations about games ;-)
<kristianpaul> dingux also use same Xbusrt SoC as nanonote and is a game console oriented device
<kristianpaul> wpwrak: http://home.earthlink.net/~cwkelley/OSGPS_chapter_3.pdf (check page 11, nice graph about that old signmag discussion)
<tuxbrain> DocScrutinizer: wow that was very very pricy!, I'm afraid the how much the beautiful Microoptical EG7 can cost
<DocScrutinizer> yeah
<tuxbrain> wpwrak: atusb on wifi AP, yummi yummi
<tuxbrain> wpwrak kristianpaul, that means instant inet access on NN :)
<DocScrutinizer> at those hefty price tags I actually start to rethink my position about "you can't build your own LCD"
<DocScrutinizer> hell, for 10k$ I can get one handmade to my specs
<DocScrutinizer> I guess selling two of those per week actually is a working business model
<DocScrutinizer> also I'd like to investigate about usability of DLPs for this purpose
<kristianpaul> tuxbrain: exactly :-)
<DocScrutinizer> fritzbox
<wpwrak> kristianpaul: nice description ! too bad it gets a little sketchy right after page 11
<DocScrutinizer> wpwrak: this should fill in the gap :-) http://www.colorado.edu/geography/gcraft/notes/gps/gps_f.html
<kristianpaul> DocScrutinizer: great link!
<DocScrutinizer> grrr
<DocScrutinizer> I should recommend this one http://www.colorado.edu/geography/gcraft/notes/gps/gps.html rather than the *_f.html framed one
<wpwrak> DocScrutinizer: one question is a bit more basic: what do the Q/I components really signify ?
<wpwrak> DocScrutinizer: it's more clear in the FFT. but in the time domain, what does the imaginary part really mean ?
<DocScrutinizer> wpwrak: where?
<DocScrutinizer> usually iirc the I component is the 90° shifted part so basically the phase
<wpwrak> DocScrutinizer: (where) for example in the chip kristianpaul uses. there, he gets Q and I
<wpwrak> DocScrutinizer: and i have the same in the USRP. so i'm kinda curious what this really means, too :)
<DocScrutinizer> sorry, I can't follow. is that a signal, or what?
<wpwrak> DocScrutinizer: (phase) yes, i understand that for the frequency domain. but what does it mean in the time domain ? how does even the concept of a phase enter in individual samples ?
<DocScrutinizer> NFC
<wpwrak> DocScrutinizer: my DSP book explains the complex time domain also more or less in terms of the frequency domain, so this doesn't really help
<DocScrutinizer> sorry, I never seen any I and/or Q on e.g. an AD converter
<kristianpaul> I/Q is part of a downconvertion
<DocScrutinizer> so I can't comment on whatever you're asking me
<kristianpaul> s/downconvertion/serialization
<kristianpaul> but i'm not sure of this at all
<kristianpaul> ADC gives two bit sign/mag data
<kristianpaul> Downcoveter / Serializer  provides I/Q