revident has quit [Quit: Combustible lemons? Bah, I bring you weaponized asparagus!]
<Mehhh>
Is there a debian wheezy image around that works with the mk802 (armhf) OOTB, or do i need to copy those 3 files on the nande partition and replace on the microsd from the mele or whatever image from doozm?
rsalveti has quit [Ping timeout: 245 seconds]
rsalveti has joined #arm-netbook
orly_owl has quit [Ping timeout: 264 seconds]
orly_owl has joined #arm-netbook
svp has joined #arm-netbook
rellla has joined #arm-netbook
Brandon15811 has quit [Excess Flood]
Brandon15811 has joined #arm-netbook
ZaEarl has quit [Ping timeout: 260 seconds]
rz2k has joined #arm-netbook
lerc has quit [Read error: Connection reset by peer]
lerc has joined #arm-netbook
svp has quit [Remote host closed the connection]
lerc has quit [Read error: Connection reset by peer]
<rz2k>
I believe I had pixelsequence set to 0 the whole time, I didnt manage to compile XBMC though
<rz2k>
but VLC works with it
<rellla>
r2zk: so i'll try to reset it to ps=0, sm=1 as it was at the beginning and discard gimli's advice. hoping to get rid of this (maybe) alpha problem with cedar.
<rz2k>
Wills VLC had no problems with ps=0
<rz2k>
might be that gimli/empat0 used configs from android and hardcoded pixel sequence conversion
<rellla>
didn't manage to compile vlc, yet.
<rellla>
might be. should ask empat...
hipboi has joined #arm-netbook
<oliv3r>
morning hipboi
<hipboi>
morning
<oliv3r>
hipboi: did you send out the mass mail yet? If you want, i can (temporarly) increase the maximum messages to 6000
ol1ver has joined #arm-netbook
<hipboi>
oliv3r: i am handling it now
<hipboi>
not that many mails this time
<hipboi>
i think if i send one by one, maybe the mail will not go to the spam folder
<oliv3r>
that won't matter, the client decides on that, and the client only sees one incoming message :)
<oliv3r>
anyway, the default max is still 1000, which is reasonable normally, if you need it higher, i will change it, just let me know ;)
<hipboi>
thanks
<RaYmAn>
oliv3r: not quite true for services like gmail though
<oliv3r>
also, I have received an e-mail from ali-express to pay my order, I guess that is an automated message, not initiated by you? I can now also pay via ali-express via CC if you prefer, just let me know and I will do it
<oliv3r>
RaYmAn: possibly, but if you relay 1000 message through an email server, with all addresses in the BCC field, the mail server will send out 1000 individual messages
<RaYmAn>
oliv3r: yes, but if 500 of those goes to gmail addresses (just an example) gmail's spam filtering will see 500
<oliv3r>
gmail MAY trigger on receiving 1000 identical messages, individually addressed those
<RaYmAn>
yes, that's my point :)
<oliv3r>
but that could happen on any spamfilter I guess, I dont' know if amavis does something that smart
<oliv3r>
which would make your comment untrue, as it would not be related to gmail specifically :D
<rz2k>
I believe for anything bigger than 300 messages or so, you need to use services from email sending guys (I recieve various newsletters from hw companies, they all use off the shelf services).
<RaYmAn>
oliv3r: 'like gmail' ;)
<RaYmAn>
rz2k: indeed. We use mailchimp's mandrill service here. They let you send 12k emails a month, for free
<rz2k>
because around a week ago I've got a mail from HardKernel saying that I've successfully paid the order in my gmail spam.
<rz2k>
(btw, anyone ordered something from their new restock of X2/U2 odroids? :p)
<oliv3r>
so for smalltime users, package mail in to 250messages per 6 ho urs or so :)
vgrade_ has joined #arm-netbook
<slapin>
hi, all!
<slapin>
do anybody have an idea of nice compile-time check for type size to be not less than sizeof(othertype)?
<RaYmAn>
you could define it =P so e.g. #if SIZELESSTHAN(sometype, othertype)?
<slapin>
awwwww
<slapin>
I thought there might be something like with negative array indexes... #if is strictly prohibited in that code by some SCM hook :(
* slapin
hates hooks for style
<oliv3r>
mnemoc: have you gave some thought to a template for the register tables?
<oliv3r>
e.g. did you make one :D
<mnemoc>
oliv3r: no, totally been burned out at $work$
<rellla>
oliv3r: imho someone should start a10-integration in vdr. first time, that allwinner a10 is mentioned on vdr-mailing list ;-)
jquip has joined #arm-netbook
hipboi has joined #arm-netbook
<focus_well>
A question for HDMI electronics engineer aware techie - do you think it will be possible to get an A10 to read a HDMI signal and convert it to display on an LCD signal fast enough with an A10 and a bit of software? Preferably without adding glue logic. But its OK to some glue logic if it helps.
<focus_well>
A couple of SIPO registers for glue logic OK to use.
<focus_well>
The reason is then I can go build a cheap cheap cheap HDMI monitor module, but do some scaling on the fly to convert the HDMI to work on say 5" displays. Cost is everything in this solution.
<focus_well>
Lesser CPUs than A10 can work, but if I use A10, I can leave in extra functionality. e.g. $25 for electronics and $7 for 5" LCD, and still sell at profit for $50.
<slapin>
focus_well: do you want to build HDMI monitor?
<focus_well>
The small HDMI monitors are dammed expensive - about $100, when the VGA reversing monitors are $15 or less.
<focus_well>
slapin: It avoids having to build stuff with LCD interface. Almost all the SoCs out there support HDMI. So least amount of work is to attach HDMI monitor instead of LCD interface. So the logic goes :-)
<focus_well>
I assume because signals from reversing cameras are simple, the simple car reversing monitors are cheap. But I don't understand how that translates to $100 for small HDMI monitors
<rz2k>
I believe A10 will not handle 160MHz of HDMI at GPIOs.
<focus_well>
rz2k: what about if I sent it in through two 8 bit Serial In Parallel Out (SIPO) reigsters? the signals will then be 10MHz parallel bus of 16 bits. Is that slow engough for the GPIO's?
<focus_well>
Its possible there are ASICs out there that can do this very well, but I don't know of any :(
<rz2k>
use cheap FPGA
<rz2k>
or even CPLD
<rz2k>
you dont need processor here
<focus_well>
rz2k: I was thinking more generic modules that can handle 5" LCDs and the like - so a bit of fast video scaling required on the fly
<focus_well>
rz2k: hmmm... CPLD/FPGA - may be opencores.org can do this! :)
<rz2k>
buy any hdmi2vga box
<rz2k>
and see whats inside :p
<focus_well>
rz2k: google images - so many varieties - hard to choose
<rz2k>
just find one with hdmi/dvi compatible input
<focus_well>
rz2k: analog.com and its chips might be just it - thank you
jquip has quit [Read error: Connection reset by peer]
jquip has joined #arm-netbook
gsilvis has quit [Read error: Connection reset by peer]
<rz2k>
focus_well: check TI, national semiconductor and other major guys for these
gsilvis has joined #arm-netbook
tinti has joined #arm-netbook
ibrah has quit [Ping timeout: 264 seconds]
<rellla>
is categorytree-extension disabled in sunxi-wiki?
<mnemoc>
uhm.... seems to have died on an auto-update
ibrah has joined #arm-netbook
<focus_well>
rz2k: thank you - I will - may be onto something here - cheap HDMI monitor to eliminate the huge diversity of LCD monitors - at least when it comes to rapid prototyping of embedded systems using these new SoCs
<oliv3r>
rellla: i moved away from vdr for tvheadend :)
<rellla>
oliv3r: ;) did you already have regrets about that? did't use tvheadend, yet. i know that it is a backend. is xbmc the only frontend?
<oliv3r>
xbmc, 'vlc', showtime hts
<oliv3r>
well my VDR setup is ... horrid, so anything is an improvement
<oliv3r>
i have a DVB-T adapter, that ALWAYS fails to display an encrypted channel on boot. Also changing channels always causes sometimes channels to fail to tune. some simple GUI app works fine btw. FTA also works fine.
<oliv3r>
tvheadend works much much better so far (with oscam)
<oliv3r>
i can use vlc (on a10) mplayer on desktop and it works quite well (dvb-t reception still seems iffy at times) but atleast it always tunes
<oliv3r>
i find changing of channels to be much slower (in xbmc) compared to vdr
<rellla>
no problems here with dvb-s + dvb-s2 on my x86 machine.
<oliv3r>
also accelerated GUI is crap and the gui overal is ugly
<rellla>
changing/tuning speed is the reason, why a native output or something via libva/xine for vdr would be cool.
<oliv3r>
xbmc opens a stream, so it tries to do some buffering etc, which causes things to be slow
<oliv3r>
on wired links, it shouldn't try to stream imo
<oliv3r>
or stream 3 channels in parallel channel 1 up channel 1 down from the current one
ibrah has quit [Ping timeout: 265 seconds]
<rellla>
i'm using vdr since 2003 now and do not appreciate a "nice" gui. so vdr is exactly what i was searching for. and it works perfectly most of the time (only on x86 :( )
<rellla>
imo it's time to get it run on smaller machines - without the need of xbmc. never tried streaming over vlc btw.
<rellla>
vdr development should go into another direction: embedded devices. there are smaller an cheaper things than ion and mini-itx. but software lacks.
<rellla>
sorry for vdr-spamming btw, should move to vdr mailing list...
ibrah has joined #arm-netbook
<oliv3r>
lol just some offtopic chatter :)
<oliv3r>
VDR's development model is way to old fashioned and I don't find it to be so stable. it does very ocassionally crash and has many issues io
<oliv3r>
also, I want to use a mele as frontend, which only works properly via xbmc. and xbmc is a pretty UI with a good WAF :)
<rellla>
WAF is only WAF, if there is no tearing, stuttering... yesterday i tested vdr-streaming to x86-client. stream stopped about 10 times, i had to ch+ and ch- everytime. believe, there wasn't any WAF at all.
<rellla>
today it's time to test WAF with a mele and xbmc as the client ;-)
<rellla>
i always have to explain, why we need those small china-things ;-) we have enough of these technical things...
<oliv3r>
LOL yeah
<oliv3r>
'what did you order again *grmbl grmbl*'
<oliv3r>
though my gf has learned to accept my buggy unstable 'because it has to be OSS" lifestyle
<rellla>
i should give mine kind of a crash course in ssh-killall-/etc/init.d/..., maybe reboot does :)
<rellla>
she calls you "my friends from the internet" - though neither i talked to anybody of them in reality nor i am connected via facebook ;-)
<oliv3r>
lol
<rellla>
funny
<rellla>
but should stop now. this grows kind of
<rellla>
VERY offtopic
nemik_ has joined #arm-netbook
<mnemoc>
rellla: categorytree should be working again
<rellla>
mnemoc: thanks, just noticed, that device-tree wasn't working.
nemik has quit [Ping timeout: 256 seconds]
hipboi has left #arm-netbook ["Leaving"]
popolon has quit [Ping timeout: 246 seconds]
The-Compiler has quit [Quit: WeeChat 0.4.0-dev]
The-Compiler has joined #arm-netbook
jquip has quit [Quit: jquip]
popolon has joined #arm-netbook
jquip has joined #arm-netbook
tinti has quit [Ping timeout: 264 seconds]
jquip has quit [Read error: Connection reset by peer]
ibrah has quit [Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121128204232]]
<oliv3r>
stupid template stuff :S
XenGi_ is now known as XenGi
XenGi is now known as XenGi_
XenGi_ is now known as XenGi
bsdfox has quit [Ping timeout: 248 seconds]
bsdfox has joined #arm-netbook
jelly has joined #arm-netbook
gzamboni has quit [Read error: Operation timed out]
XenGi is now known as XenGi_
jquip has joined #arm-netbook
gzamboni has joined #arm-netbook
vinifm has joined #arm-netbook
cat_x301 has quit [Ping timeout: 256 seconds]
Sv2 has joined #arm-netbook
sv has quit [Ping timeout: 276 seconds]
tinti has quit [Ping timeout: 245 seconds]
tinti has joined #arm-netbook
orly_owl_ has joined #arm-netbook
grevaill1t has joined #arm-netbook
Sv2 is now known as sv
RaYmAn has quit [*.net *.split]
orly_owl has quit [*.net *.split]
pwhalen has quit [*.net *.split]
grevaillot has quit [*.net *.split]
tinti has quit [Ping timeout: 264 seconds]
tinti has joined #arm-netbook
gzamboni has quit [Ping timeout: 264 seconds]
rellla2 has joined #arm-netbook
RaYmAn has joined #arm-netbook
Brandon15811 has quit [Excess Flood]
rellla has quit [Ping timeout: 250 seconds]
pwhalen has joined #arm-netbook
Brandon15811 has joined #arm-netbook
rellla2 has quit [Remote host closed the connection]
rellla has joined #arm-netbook
gzamboni has joined #arm-netbook
<rz2k>
Mali-400 awesomeness: es2tri dies with assertation errors, es1tri works.
tinti has quit [Ping timeout: 260 seconds]
<rz2k>
oh, es2tri cant compile first shader
<rz2k>
lol
<L84Supper>
is the cubieboard the only a10 or <$80 arm a8 (or better) board with gpio?
<rz2k>
L84Supper: beaglebone?
<L84Supper>
except for the pi
<jinzo>
offtopic: The whole OpenGL 3D programming is a bit... let's say lacking imo. From the tools to the documentation. I have background in developement and now I'm doing some preatty basic 3D stuff with OpenGL 3+ and really. What they call documentation and tooling I call kid's toys
<jinzo>
L84Supper, there's hackberry
<jinzo>
and gooseberry
<mnemoc>
the bone is $90+shipping
<jinzo>
A10 in them tho.
<mnemoc>
jinzo: he asked with GPIOs, pins
<L84Supper>
the bone is so old I forget what cpu it has
<rz2k>
mnemoc: really? I thought it was around 75 on farnell...
<jinzo>
mnemoc, I presumed they both support GPIOs - It's beyond me why a board like that wouldn't support them.
<L84Supper>
designers tend to skip laying out any GPIO or the PCIe on ARM boards
<L84Supper>
too busy? didn't think anyone would use them? I don't know
<jinzo>
but, but - then hackberry and gooseberry can only be used for what? media centers and alike?
<L84Supper>
Hackberry didn't include SATA?!
<mnemoc>
rz2k: 75E+16E to spain
<mnemoc>
rz2k: uh, those 16E are VAT, not shipping :|
<mnemoc>
the cubieboard is the only cortex-a8 with GPIOs for under $80
<L84Supper>
I don't get it, a designer does through the trouble of designing the board (not exactly compact either) and leaves out tons of IO? I can understand not stuffing unused parts, but not even leaving the pads for a general purpose board is just odd.
<bfree>
what about the olimex A13?
<mnemoc>
L84Supper: the gooseberry is the PCB of the lyf1 7" tablet
<mnemoc>
bfree: true, 55E+shipping
<mnemoc>
+var
<L84Supper>
mnemoc: so probably no GPIO then
<mnemoc>
vat
<mnemoc>
bfree: the A10 olinuxino will be hot
<bfree>
well 35+vat+shipping if the micro will do ... meets the A8 + gpio requirements but there may be more :-p
<mnemoc>
bfree: :)
<L84Supper>
the a13 is a crippled a10 , missing SATA and what else did they leave out?
<bfree>
mnemoc: yes, just noticed their A10 prototype last night ... if the 2GB "works" out (and price is remotely sane) it sure will be nice!
<L84Supper>
also missing ethernet
itamar_ has quit [Remote host closed the connection]
<mnemoc>
but I also hope they correct the mistake and sell the boards with the sata populated
<L84Supper>
what's actually shipping? I've been reading about some of these boards for months
<mnemoc>
nothing
<mnemoc>
fsl is still BSing with the availability of iMX6
<L84Supper>
what high volume product is shipping with it? any?
<L84Supper>
wondering if it's allocation or still broken silicon
<mnemoc>
hiapad claimed stock, but retracted some days after charbax's interview
<mnemoc>
same with the retina screen
<L84Supper>
will 2013 be the year of the useable ARM board? :)
<mnemoc>
?
<mnemoc>
there are many usable ARM boards
<mnemoc>
and if you want GPIOs and cortex-a8 you can buy a bone today
<mnemoc>
fully open source
<mnemoc>
and full specs
<L84Supper>
I should have just finished a different a10 design after dropping the EOMA stuff
<L84Supper>
or even fixing the bone design, the board layout is terrible
avernos has quit [Ping timeout: 246 seconds]
<L84Supper>
TI always makes their dev boards in a form factor that is terrible for integration
<L84Supper>
and/or make them too high priced or only in small quan like the Panda
avernos has joined #arm-netbook
avernos has quit [Changing host]
avernos has joined #arm-netbook
<mnemoc>
because they are only marketing tools
<L84Supper>
yes, so still no boards :)
<mnemoc>
tom is shipping soon-ishh :)
<L84Supper>
it would be nice
<L84Supper>
same story with all of them
<rellla>
rz2k: thanks, busy reworker ;-) would be of interest, if linux-sunxi/mali-libs -directories can be assigned...
<L84Supper>
Olimex doesn't have any pics of the PCB layers, he also seemed baffled by the a10 early on, so y confidence in that board isn't exactly too high
<rz2k>
I've assigned couple of libs
<L84Supper>
they also went with 4 layers vs 6 while using KiCAD
avernos has quit [Ping timeout: 246 seconds]
<L84Supper>
they can always try to run it slower to get it stable
avernos has joined #arm-netbook
<rellla>
rz2k: i meant, todo would be: find out, if/which libs in the "official" summary in linux-sunxi-repo correspond to the ones listed (tom.linux../dl.linux-sunxi.org)
<rz2k>
I will get there some day
<rellla>
all in all listing is more clear imho. +1
<libv>
"make VERSION=r3p0 ABI=armhf x11" is completely pointless if you are going to do it manually again later on
<libv>
also. abi and versioning should happen automagically
<libv>
so make that "make x11"
<libv>
if this did not work for either of you, _say_ so
<rellla>
libv: i haven't tried or took a look into reorganized mali-repo, yet. can i tell make where to install libs? and does this work for a cross-compile environment, too?
<rz2k>
libv: fixed
<libv>
rellla: why would you want to do this in a cross-compile?
<libv>
rellla: the versioning program talks to /dev/mali
<libv>
rellla: these makefiles are all about installation, not compilation
<libv>
rellla: gnu standard prefix and libdir env variables are being used to force installation to some places
<rellla>
libv: thats, what i wanted to know.
<rellla>
i'm too slow in writing ;-)
<rellla>
compiling was the wrong word, of course
<libv>
rz2k: ok, so this was not some issue you ran into when trying this yourself?
<rz2k>
I believe additional cp -r was from old instructions, I've added yours above old ones
<rz2k>
just didnt test, usual stuff...
<libv>
oh, ok :)
<libv>
i do not think many people have tested this automated installation properly, as i was expecting a few complaints and i so far have received none
<rz2k>
could you please add your quake3 results with some info to Benchmarks page?
<libv>
i will get back to playing with q3a in the next few days
<libv>
and will do so then
<rz2k>
cool, thanks
<rz2k>
you can also restructurize benchmarks page if you want, I've done it pretty raw, might missed something or added useless info.
<libv>
right now i am trying to speed up existing lima demos, a dirty hack tells me that i can get things to run 4x as fast when doing ""pageflipping"" to fb
<rellla>
libv: it's faster to do a cp on another directory then adapt a newly structured system.
<libv>
rellla: yes, reading is hard
<libv>
rellla: luckily irc does not involve reading
<libv>
oh wait.
<rz2k>
lol
<rellla>
empat0's xbmc is doing cp, still. should also change.
<libv>
that readme really is not information overload
<rellla>
but must done :-)
<rellla>
be done
<rz2k>
libv: if you will get somewhere around pixel sequence/conversions/scaler mode - document it somewhere, please.
<rz2k>
could be useful
<libv>
rz2k: that does not parse in respect to the mali stuff
<libv>
rz2k: that sounds like the broken /dev/disp device
<libv>
rz2k: the pseudo-page flipping is purely /dev/fb
<rz2k>
oh, I thought you are trying to implement lima stuff on A10 using default crap scaler/fliping stuff
<libv>
that would be too chip specific
<libv>
and... i do not want to depend on something that should vanish in the mid-term
<rz2k>
ok :)
Turl has quit [Excess Flood]
Turl has joined #arm-netbook
rz2k has quit []
newbie has joined #arm-netbook
newbie is now known as Guest95387
tinti has joined #arm-netbook
gzamboni has quit [Ping timeout: 260 seconds]
hp__ has quit [Ping timeout: 276 seconds]
Sv2 has joined #arm-netbook
tinti has quit [Quit: Leaving]
sv has quit [Ping timeout: 276 seconds]
Mehhh has quit [Ping timeout: 245 seconds]
vinifm has quit [Quit: Saindo]
revident has quit [Quit: MHmmm... Raspberry Pi]
Sv2 is now known as sv
rellla has quit [Quit: rellla]
ZaEarl has joined #arm-netbook
<penguin42>
lkcl: I think you need to add a year to the shedule for your chip
<penguin42>
lkcl: From completing a chip and having your samples working to mass prod is probably a few months - so forget about mid-'13
ZaEarl has quit [Ping timeout: 260 seconds]
cat_x301 has quit [Ping timeout: 245 seconds]
rastko has joined #arm-netbook
Brandon15811__ has quit [Quit: Leaving]
<xxiao>
ULP vs EOMA, anyone?
rastko has quit [Remote host closed the connection]
j1nx_ has quit [Quit: Hard work pays off in the future, laziness pays off now]
<hno>
xxiao, similar but quite different aims.
<hno>
xxiao, EOMA aims at end-user replaceable modules, while ULP aims at OEM designs. Both have merit, but ULP do not address any of what EOMA tries to address. EOME do however indirectly address part of what ULP aims at.