#raspberrypi IRC Log


IRC Log for 2012-02-07

Timestamps are in GMT/BST.

[0:01] <traeak> probably flight sim inputs would be my best bet i gather
[0:01] <traeak> hmm
[0:01] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) Quit (Quit: Leaving)
[0:07] * Martix (~martix@134.89.broadband12.iol.cz) has joined #raspberrypi
[0:07] * johnthebear (~Johnthebe@13-165-19.client.wireless.msu.edu) Quit (Quit: Computer has gone to sleep.)
[0:09] * DexterLB (~angel@ Quit (Ping timeout: 252 seconds)
[0:13] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) has joined #raspberrypi
[0:14] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) has joined #raspberrypi
[0:15] * TheShrew (~theshrew@87-194-161-58.bethere.co.uk) Quit (Quit: Nettalk6 - www.ntalk.de)
[0:16] * DexterLB (~angel@ has joined #raspberrypi
[0:17] * Martix (~martix@134.89.broadband12.iol.cz) Quit (Ping timeout: 265 seconds)
[0:19] * R` (~RHA@ip7.j-k.kund.riksnet.nu) Quit (Remote host closed the connection)
[0:21] * imnichol_ (~ian@ has joined #raspberrypi
[0:22] * koaschten (~koaschten@31-16-0-231-dynip.superkabel.de) Quit (Ping timeout: 255 seconds)
[0:22] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) Quit (Quit: lolwat)
[0:23] * imnichol_ (~ian@ Quit (Client Quit)
[0:24] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) has joined #raspberrypi
[0:35] <piofcube> tntexplosivesltd: Thought about using CG+? Don't know if it's what your wanting to do but... maybe ;-)
[0:35] * rafal_ (~rafal@ip-78-30-108-163.free.aero2.net.pl) has joined #raspberrypi
[0:35] * rafal__ (~rafal@ip-78-30-110-44.free.aero2.net.pl) Quit (Read error: Connection reset by peer)
[0:38] * victhor (~victhor@ Quit (Ping timeout: 276 seconds)
[0:39] * victhor (~victhor@ has joined #raspberrypi
[0:41] * nils_2 (~nils_2@unaffiliated/nils-2/x-2480262) Quit (Quit: WeeChat 0.3.7-rc1)
[0:44] * Laogeodritt (~Laogeodri@wikipedia/Laogeodritt) has joined #raspberrypi
[0:46] * victhor (~victhor@ Quit (Ping timeout: 252 seconds)
[0:47] * jmontleon (~jason@c-76-19-35-6.hsd1.ma.comcast.net) has joined #raspberrypi
[0:47] * victhor (~victhor@ has joined #raspberrypi
[0:47] * smw (~smw@ has joined #raspberrypi
[0:48] * smw (~smw@ Quit (Changing host)
[0:48] * smw (~smw@unaffiliated/smw) has joined #raspberrypi
[0:49] * MystX (~MystX@ec2-107-21-100-215.compute-1.amazonaws.com) has joined #raspberrypi
[0:50] <MystX> So, can we kick anyone that asks if it's out yet now that there's a date we know it wont be out by?
[0:50] <tntexplosivesltd> that sounds reasonable
[0:50] * jluisn (~quassel@ Quit (Read error: Connection reset by peer)
[0:51] <tntexplosivesltd> piofcube: the terrain gen is only a small part of it, and it all works in C++
[0:53] <Thorn_> is it out ye
[0:53] <Thorn_> t
[0:53] <Thorn_> ah ffs
[0:53] <mrdragons> MystX: Well, there's still no release, just a date as to when manufacturing will be done. :P
[0:54] <tntexplosivesltd> MystX: dairy at 2? Or are you "busy"
[0:54] <mrdragons> It is out yet?
[0:55] <traeak> mrdragons: you haven't got yours yet?
[0:55] <tntexplosivesltd> piofcube: but thanks for the suggestion, ;ppls interesting
[0:55] <tntexplosivesltd> * looks
[0:55] <MagicalTux> "[...] you should be able to buy them before the end of the month."
[0:55] <MystX> Yay flash sandboxing for FF
[0:55] <MystX> tntexplosivesltd: Ill see later
[0:55] <tntexplosivesltd> k
[0:55] <MagicalTux> so when people ask, just tell them "next month"
[0:56] * Soul_Est (~nolanhayn@CPE0017f2324fce-CM0012250096a8.cpe.net.cable.rogers.com) Quit (Remote host closed the connection)
[0:58] <mrdragons> lolol
[1:02] * esotera (~jamie@94-193-222-75.zone7.bethere.co.uk) Quit (Quit: Leaving)
[1:04] * feep (~feep@p5B2B51EA.dip.t-dialin.net) Quit (Ping timeout: 256 seconds)
[1:19] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) Quit (Remote host closed the connection)
[1:19] * prbz_ (~prebz@c83-254-51-59.bredband.comhem.se) Quit (Ping timeout: 240 seconds)
[1:20] * kynes- (~marty@adsl-75-0-8-178.dsl.renocs.sbcglobal.net) has joined #raspberrypi
[1:21] * prbz (~prebz@c83-254-51-59.bredband.comhem.se) has joined #raspberrypi
[1:22] * ShiftPlusOne (~Shift@124-168-104-90.dyn.iinet.net.au) Quit (Quit: Leaving)
[1:23] * rafal_ (~rafal@ip-78-30-108-163.free.aero2.net.pl) Quit (Quit: leaving)
[1:24] * izua (~izua@unaffiliated/izua) Quit (Quit: :tiuQ)
[1:31] * cerberos (~cerberos@ has joined #raspberrypi
[1:33] * iMatttt (~imatttt@cpc3-farn4-0-0-cust594.6-2.cable.virginmedia.com) has joined #raspberrypi
[1:34] * victhor (~victhor@ Quit (Ping timeout: 276 seconds)
[1:36] * Davespice (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) Quit (Ping timeout: 240 seconds)
[1:39] * cerberos (~cerberos@ Quit (Quit: cerberos)
[1:40] * kynes- (~marty@adsl-75-0-8-178.dsl.renocs.sbcglobal.net) Quit (Quit: leaving)
[1:41] <MystX> SO many people leaving D=
[1:41] * UnderSampled (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) has joined #raspberrypi
[1:41] <MystX> YAY
[1:42] <MystX> Also, i went on hacker news before: "RasberryPi release date February 20th"
[1:42] <MystX> NO
[1:42] * Hopsy (~Hopsy@ has joined #raspberrypi
[1:42] <Hopsy> yeeaass
[1:42] <Hopsy> finnaly
[1:42] <Hopsy> a release date thing
[1:42] <mrdragons> Vaporware.
[1:42] <mrdragons> I told you all.
[1:42] <Hopsy> is it out yet?
[1:42] * PiBot slaps Hopsy across the face with a cast iron pan.
[1:42] <piofcube> tntexplosivesltd: there's also the IML++ which is a C++ templated library which includes CG/CGS/BiCG/etc... http://math.nist.gov/iml++/ might be worth a quick browse ;-)
[1:43] * Hopsy (~Hopsy@ Quit (Client Quit)
[1:45] <piofcube> ah.. cool a broadcom pdf for the 2835 :-)
[1:46] <mrdragons> A godsend for those that want to try and roll a homebrew OS for it
[1:46] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[1:47] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[1:47] * ukscone (~Russell@cpe-68-173-234-162.nyc.res.rr.com) has left #raspberrypi
[1:47] * ukscone (~Russell@cpe-68-173-234-162.nyc.res.rr.com) has joined #raspberrypi
[1:48] <MystX> (ish)
[1:48] <mrdragons> A little bit of info is better than none
[1:49] <mrdragons> 205 isn't a pamphlet either
[1:49] * chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) Quit (Ping timeout: 252 seconds)
[1:50] * Davespice (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) has joined #raspberrypi
[1:50] <piofcube> a bit of bed-time reading :P
[1:52] <SpeedEvil> link?
[1:52] <DaQatz> Bot is now amended for Feb 20th
[1:52] <DaQatz> So after that day won't hit people with iron pans.
[1:53] <MystX> But it should do it until it's released?
[1:53] * SpeedEvil looks at the 3510 page document for his phone processor.
[1:53] <DaQatz> Just will start to say "Maybe" instea of pan bash.
[1:59] * victhor (~victhor@ has joined #raspberrypi
[1:59] * customtronics (~puppy@user-12l2tpe.cable.mindspring.com) Quit (Quit: Leaving)
[2:00] * iMatttt (~imatttt@cpc3-farn4-0-0-cust594.6-2.cable.virginmedia.com) Quit (Quit: TLS packet ect messages)
[2:01] <mrdragons> SpeedEvil: Okay, maybe it is a bit small considering what is inside the GPU...
[2:02] * Palbert (6f45c0fb@gateway/web/freenode/ip. has joined #raspberrypi
[2:02] <SpeedEvil> Where is the link?
[2:03] <SpeedEvil> ah
[2:04] * ali1234 (~ajbuxton@robotfuzz.co.uk) has joined #raspberrypi
[2:07] * PiBot (~Raspberry@c-24-147-35-37.hsd1.ma.comcast.net) Quit (Remote host closed the connection)
[2:07] * PiBot (~Raspberry@c-24-147-35-37.hsd1.ma.comcast.net) has joined #raspberrypi
[2:07] <DaQatz> Is it out yet?
[2:07] <PiBot> DaQatz: The boards will be completed on February 20th.
[2:08] <SpeedEvil> Oooh.
[2:08] <SpeedEvil> I wonder if the PWM pins are brought out.
[2:08] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[2:08] <SpeedEvil> Funky wierd DMA mode to write out a hundred megabits a second of arbitrary data.
[2:09] <SpeedEvil> SDR with just a filter.
[2:09] <SpeedEvil> (out)
[2:10] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[2:11] * gomiboy (~frodone@ppp-192-16.21-151.libero.it) Quit (Ping timeout: 260 seconds)
[2:12] <SpeedEvil> It would be realy, really awesome if it could do that inwards, but nope.
[2:13] * socialdefect_ (~socialdef@5418BB7F.cm-5-1c.dynamic.ziggo.nl) Quit (Quit: Konversation terminated!)
[2:17] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[2:21] * pizza-dude (~fake@dhcp-077-249-161-109.chello.nl) Quit (Ping timeout: 244 seconds)
[2:22] * wiiguy (~fake@unaffiliated/wiiguy) has joined #raspberrypi
[2:23] <victhor> BCM2835 has a 16 bit parallel memory controller...
[2:23] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[2:24] <victhor> too bad it's not like the OMAP's, where you can connect anything to there... I've seen CAN bus controllers, but mostly ethernet ones connected to that bus
[2:24] <victhor> maybe I'm wrong
[2:24] * LiENUS (~yes@2001:470:bbb3:12:cd37:f4d1:591b:2dcd) Quit (Quit: Leaving)
[2:26] * MinToN (ident@cpc1-oxfd7-0-0-cust545.4-3.cable.virginmedia.com) Quit (Quit: ???I-n-v-i-s-i-o-n??? 3.2 (July '10))
[2:28] * cerberos (~cerberos@ has joined #raspberrypi
[2:29] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[2:30] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[2:33] <victhor> this post is rather interesting... http://www.raspberrypi.org/forum/general-discussion/documentation-questions-i2c-and-second-uart/#p38852
[2:34] <victhor> according to the datasheet, pins 22-25 have suggestive alternate functions such as "SD1_CLK", "SD1_CMD", "SD1_DAT0", "SD1_DAT1"...
[2:34] <victhor> I hope that's what I think it is... a SD card interface
[2:38] <MystX> They're not connected to the existing sd card slot?
[2:39] * saua (~saua@193.81-167-3.customer.lyse.net) Quit (Ping timeout: 248 seconds)
[2:41] <victhor> doh. Maybe that's it...
[2:42] * wiiguy (~fake@unaffiliated/wiiguy) Quit ()
[2:42] <MystX> Lol
[2:50] * knotty (~void@loin.ailleurs.org) has joined #raspberrypi
[2:50] <DaQatz> victhor, With linux standard gpio can be used as SD.
[2:50] <victhor> if there's something I'd rather avoid doing is bit banging high speed interfaces. Besides it also supports SD cards through SPI, which is much better
[2:51] <SpeedEvil> DaQatz: Well - yes and no
[2:51] <SpeedEvil> DaQatz: and what victhor said
[2:58] <DaQatz> What'sthat beep?
[2:58] * roman3x_ (~roman3x@bband-dyn18.95-103-42.t-com.sk) Quit (Ping timeout: 245 seconds)
[3:00] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[3:01] * |uen| (~uen@ has joined #raspberrypi
[3:02] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[3:05] * uen| (~uen@p5DCB0FBB.dip.t-dialin.net) Quit (Ping timeout: 256 seconds)
[3:05] * jardiamj (~jardiamj@s64er2.p1.38.mtangel.net) has joined #raspberrypi
[3:07] * atts (~adam@ has joined #raspberrypi
[3:09] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[3:14] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[3:22] * victhor (~victhor@ Quit (Read error: Connection reset by peer)
[3:24] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[3:26] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[3:29] * Cemial (~hardcoreB@ip4da2366d.direct-adsl.nl) Quit (Ping timeout: 256 seconds)
[3:31] <ksx4system> will Raspberry Pi have RS232 serial port?
[3:31] <MystX> Nope
[3:31] <ksx4system> :(
[3:31] <ksx4system> so USB to RS232 interface is only way to do that, am I right?
[3:32] <Byan> wrong..
[3:32] <ksx4system> I'll only need four of nine pins (GND, DTR, TXD, RXD)
[3:32] <Byan> DTR?
[3:32] <ksx4system> Byan: http://www.grzesiek21.republika.pl/ds9097e.gif
[3:32] <MystX> Uhm. You could use SPI or stuff like that. But you would need to build the interface out of other interfaces, yes
[3:33] <Byan> ok, but what is DTR used for
[3:33] <Byan> it has UART ksx4system.
[3:33] <ksx4system> I have no idea, I don't fully understand this schematic... at least I can soldier it up
[3:33] <SpeedEvil> Data Transmit Ready
[3:33] <Byan> so that would be the control.
[3:33] <SpeedEvil> And you don't want to do that
[3:34] <SpeedEvil> there are more sensible ways to hook up 1-wire.
[3:34] <ksx4system> for example?
[3:34] <ksx4system> this one is fastest and cheapest I found
[3:34] <SpeedEvil> The easy way is USB->1wire bridge
[3:35] <SpeedEvil> The above will not work on a not 12V serial port
[3:35] <ksx4system> ...not quite easy 'cause I have all parts for schematic above and a bunch of USB<--->RS232 adapters
[3:35] <SpeedEvil> Well - then that works fine
[3:35] <ksx4system> SpeedEvil: wait, what? I can't hook it to ordinary PC's serial port?
[3:35] <ksx4system> like, linuxified thin client?
[3:36] <SpeedEvil> ksx4system: I thought you meant a serial port on a PI
[3:36] <Byan> well, I guess the question is.. since when is serial on a PC 12V? I always thought it was 5V.
[3:36] <SpeedEvil> It depends. Some modern serial ports may not output enough voltage for it to work well
[3:36] <SpeedEvil> Byan: Since forever
[3:36] <ksx4system> SpeedEvil: since you can't buy Pi *exactly now* and I'll soldier it all today/tomorow I'll use one of my current boxes to test it
[3:36] <SpeedEvil> except more recent ones.
[3:37] <MystX> What Byan its always been 'pretty much anything'
[3:37] <SpeedEvil> It won't work at all for the serial port on the pi.
[3:37] <ksx4system> SpeedEvil: even with external power?
[3:37] <SpeedEvil> As that is both very low current, and 3.3V
[3:37] <SpeedEvil> you can use an external USB->serial converter to run it
[3:38] <ksx4system> SpeedEvil: Bus 003 Device 002: ID 1a86:7523 QinHeng Electronics HL-340 USB-Serial adapter
[3:38] <ksx4system> works as good as Profilic PL2303 does
[3:38] <SpeedEvil> By the serial port on the pi, I mean teh serial port exposed on to the GPIO connectors.
[3:38] <Byan> honestly, ksx4system, you seem to be trying to "take the easy way out." Best to do more research and understand your schematic first..
[3:39] <ksx4system> Byan: well, you're right. I'm searching easiest possible way to make the hardware, software will not be a problem (owfs/digitemp uber alles!)
[3:40] <SpeedEvil> If you're using devices which can take external power - then do that
[3:41] <SpeedEvil> It simplifies the requirements on the interface lots
[3:44] <SpeedEvil> You can run 1-wire from a GPIO, at 3.3V
[3:44] <SpeedEvil> http://www.maxim-ic.com/app-notes/index.mvp/id/4206
[3:44] <SpeedEvil> Though I wouldn't recommend this dor other than short busses in an enclosure
[3:44] * lrncmdln (~puppy@125-239-187-192.jetstream.xtra.co.nz) Quit (Remote host closed the connection)
[3:44] * hamitron (~hamitron@just.screw-the.eu) Quit (Quit: Leaving)
[3:44] <ksx4system> SpeedEvil: I understand schematics and DS18B20 pinout well enough to modify it to use external power (hey, I can steal 5V from USB!)
[3:45] <SpeedEvil> Then you just hook up ground, +5V, and wire the data pin to a GPIO, and you're done
[3:45] <ksx4system> will it last for 2,5m four DS18B20 bus?
[3:46] <SpeedEvil> If you supply external power, yes
[3:46] <SpeedEvil> http://www.plugcomputer.org/plugforum/index.php?topic=163.0
[3:46] <SpeedEvil> a quick google revealed
[3:46] <ksx4system> external means stealing 5V from USB too or *really* external?
[3:46] * feep (~feep@p5B2B324D.dip.t-dialin.net) has joined #raspberrypi
[3:46] <SpeedEvil> any source of 5V at over 10mA or so
[3:47] <ksx4system> SpeedEvil, yay! thank you so much for this link
[3:47] <SpeedEvil> Also - read and understand some of the other maxim 1-wire bus documentation links - they're good
[3:47] <SpeedEvil> 'reliable 1-wire' is a good idea
[3:48] <ksx4system> only thing from them I have read was DS18B20 and DS18B20-PAR white papers
[3:48] <ksx4system> thanks again, I'll read more
[3:48] * SpeedEvil looks at his bag of 18b20s.
[3:49] <ksx4system> lol, these are around $3/one here...
[3:49] * ksx4system however got four of them for free
[3:53] * EastLight (t@5ac4afeb.bb.sky.com) Quit ()
[3:56] <SpeedEvil> I got some cheap from ebay
[3:56] <SpeedEvil> $.5 IIRC
[3:58] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[4:00] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[4:02] * jmontleon (~jason@c-76-19-35-6.hsd1.ma.comcast.net) Quit (Remote host closed the connection)
[4:06] * zgreg (greg@ Quit (Remote host closed the connection)
[4:08] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[4:09] * ksx4system (~ksx4syste@ancon.mac.info.pl) Quit (Quit: ZNC - http://znc.in)
[4:10] * ksx4system (~ksx4syste@ancon.mac.info.pl) has joined #raspberrypi
[4:11] * Palbert (6f45c0fb@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[4:11] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[4:14] * zgreg (greg@ has joined #raspberrypi
[4:16] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[4:19] * Skorpy (~sevanteri@ Quit (Ping timeout: 272 seconds)
[4:23] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[4:24] * customtronics (~puppy@user-12l2tpe.cable.mindspring.com) has joined #raspberrypi
[4:30] <MystX> tntexplosivesltd:
[4:31] * MystX (~MystX@ec2-107-21-100-215.compute-1.amazonaws.com) Quit (Quit: leaving)
[4:35] * inco (580ed4e9@gateway/web/freenode/ip. has joined #raspberrypi
[4:35] * atts (~adam@ has left #raspberrypi
[4:36] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[4:38] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[4:41] * piofcube (~piofcube@piofcube.1billionhex.com) Quit (Quit: Lost terminal)
[4:43] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[4:44] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[4:46] * rm (~rm@fsf/member/rm) Quit (Quit: ZNC - http://znc.sourceforge.net)
[4:47] * rm (~rm@fsf/member/rm) has joined #raspberrypi
[4:54] * Palbert (6f45c0fb@gateway/web/freenode/ip. has joined #raspberrypi
[5:01] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[5:01] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[5:03] * Hesmon (~fp@p4FE3A165.dip.t-dialin.net) has joined #raspberrypi
[5:05] * Hesmon_ (~fp@p4FDB77BF.dip.t-dialin.net) Quit (Read error: Operation timed out)
[5:08] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[5:09] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[5:11] * Transformer (~Transform@ool-4a59e397.dyn.optonline.net) has joined #raspberrypi
[5:12] * jardiamj (~jardiamj@s64er2.p1.38.mtangel.net) Quit (Remote host closed the connection)
[5:14] * Transformer (~Transform@ool-4a59e397.dyn.optonline.net) Quit (Excess Flood)
[5:21] * customtronics (~puppy@user-12l2tpe.cable.mindspring.com) Quit (Quit: Leaving)
[5:23] * Palbert (6f45c0fb@gateway/web/freenode/ip. Quit (Quit: Page closed)
[5:26] * ReggieUK (ReggieUK@5ac9ad24.bb.sky.com) Quit ()
[5:42] * MystX (~MystX@121-73-41-249.cable.telstraclear.net) has joined #raspberrypi
[5:54] * Skorpy (~sevanteri@ has joined #raspberrypi
[5:57] * DaMummy|PND (~veki@cpe-76-190-200-98.neo.res.rr.com) Quit (Ping timeout: 240 seconds)
[6:04] * Mowee (~Mowi@ Quit (Excess Flood)
[6:05] * Mowee (~Mowi@ has joined #raspberrypi
[6:13] * inco (580ed4e9@gateway/web/freenode/ip. Quit (Quit: Page closed)
[6:13] <SpeedEvil> http://www.dealextreme.com/p/td20-2d-movies-to-3d-converter-with-hdmi-1-4-output-black-115751 haha
[6:23] * relaxed (~relaxed@unaffiliated/relaxed) Quit (Quit: ??????)
[6:23] * mchou (~quassel@unaffiliated/mchou) Quit (Remote host closed the connection)
[6:31] * DaMummy|PND (~veki@cpe-76-190-200-98.neo.res.rr.com) has joined #raspberrypi
[6:32] * nemrod (~nemrod@unaffiliated/nemrod) Quit (Read error: Operation timed out)
[6:33] * nemrod (~nemrod@unaffiliated/nemrod) has joined #raspberrypi
[6:42] * brumako (~heraclitu@sa-175-68.saturn.infonet.ee) Quit (Quit: Leaving.)
[6:44] * relaxed (~relaxed@unaffiliated/relaxed) has joined #raspberrypi
[6:55] * mrdragons (~lucas@ Quit (Quit: nite)
[6:58] * imnichol (~ian@ Quit (Quit: Ex-Chat)
[7:18] * huene (~join@thedarkside.at) has joined #raspberrypi
[7:24] * CuriosLion (~stian@117.81-166-155.customer.lyse.net) has joined #raspberrypi
[7:25] * UnderSampled (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) Quit (Read error: Connection reset by peer)
[7:25] * CuriosTiger (~stian@117.81-166-155.customer.lyse.net) Quit (Ping timeout: 256 seconds)
[7:27] * fhmdgxs (~root@ has joined #raspberrypi
[7:46] * P4R4N01D (~johndoe@ has joined #raspberrypi
[7:48] * prbz_ (~prebz@c83-254-51-59.bredband.comhem.se) has joined #raspberrypi
[7:51] * prbz (~prebz@c83-254-51-59.bredband.comhem.se) Quit (Ping timeout: 248 seconds)
[7:59] * xmzgtx (~xmzgtx@ has joined #raspberrypi
[8:12] * R` (~RHA@ip7.j-k.kund.riksnet.nu) has joined #raspberrypi
[8:19] <Henchman21> SpeedEvil: http://www.dealextreme.com/p/compact-waterproof-usb-2-0-cmos-300kp-4-led-illuminated-snake-camera-endoscope-7m-115813
[8:19] <Henchman21> DIY prostate
[8:20] <MystX> ???_???
[8:21] <MystX> Do dicts always iterate in the same order in python?
[8:22] <rm> Waterproof <- I can't help myself but want to add "Manufacturer-rated" there :D
[8:22] * pygo (~pygo@fran.pygonia.com) has joined #raspberrypi
[8:22] <MystX> And a winky face
[8:22] <MystX> ;)
[8:24] <SpeedEvil> I sort-of-want one of those to answer some plumbing questions
[8:24] <MystX> Personal plumbing?
[8:24] <MystX> But really, i want one, but cant handle the ridicule
[8:26] * nemrod (~nemrod@unaffiliated/nemrod) Quit (Read error: Operation timed out)
[8:26] * nemrod (~nemrod@unaffiliated/nemrod) has joined #raspberrypi
[8:28] <SpeedEvil> For working out where my guttering goes
[8:29] <ahven> http://c-skills.blogspot.com/2012/02/lots-of-performance-webserver.html
[8:29] <ahven> this for a dedicated, yet simple, webserver? :)
[8:30] <Henchman21> in teh but hole
[8:33] * fhmdgxs (~root@ Quit (Ping timeout: 245 seconds)
[8:50] * Martix (~martix@134.89.broadband12.iol.cz) has joined #raspberrypi
[8:51] * Laogeodritt (~Laogeodri@wikipedia/Laogeodritt) Quit (Ping timeout: 244 seconds)
[9:04] * gabriel9 (~quassel@ has joined #raspberrypi
[9:11] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has joined #raspberrypi
[9:11] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has left #raspberrypi
[9:12] * diplo (~diplo@cpc1-trow1-0-0-cust13.aztw.cable.virginmedia.com) has joined #raspberrypi
[9:14] * MystX (~MystX@121-73-41-249.cable.telstraclear.net) Quit (Quit: Leaving)
[9:23] * Martix (~martix@134.89.broadband12.iol.cz) Quit (Ping timeout: 248 seconds)
[9:23] * zutesmog1 (~timh@CPE-124-179-5-14.lns3.ken.bigpond.net.au) has joined #raspberrypi
[9:24] * zutesmog (~timh@CPE-124-177-48-120.lns5.win.bigpond.net.au) Quit (Ping timeout: 252 seconds)
[9:24] <zutesmog1> MystX: iteration order of dicts should not be relied upon. (Current behaviour is an artifact of the implementation) If you want garunteed order use an OrderedDict.
[9:25] <zutesmog1> oops, he left.
[9:28] * paul- (u4804@gateway/web/irccloud.com/x-bcvafdruvzssbzql) Quit (Ping timeout: 240 seconds)
[9:30] * bolosaur (u5293@gateway/web/irccloud.com/x-ebucgamhnkzaefzx) Quit (Ping timeout: 276 seconds)
[9:33] <tntexplosivesltd> I'll tell him tomorrow
[9:35] * cosh (~cosh@gateway/tor-sasl/ryld) has joined #raspberrypi
[9:36] * nsc (~nsc@tardis.chantrell.net) has left #raspberrypi
[9:40] <RaTTuS|BIG> well we have a better date now...
[9:41] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) has joined #raspberrypi
[9:41] * yumaokao (~yumaokao@ Quit (Quit: leaving)
[9:43] * yumaokao (~yumaokao@ has joined #raspberrypi
[9:50] * kism3t (~kism3t@client-80-3-134-81.bsh-bng-013.adsl.virginmedia.net) has joined #raspberrypi
[9:50] * nsc (~nsc@tardis.chantrell.net) has joined #raspberrypi
[9:50] * nils_2 (~nils_2@unaffiliated/nils-2/x-2480262) has joined #raspberrypi
[9:52] * TSL (5b7ee6f9@gateway/web/freenode/ip. has joined #raspberrypi
[9:56] * paul- (u4804@gateway/web/irccloud.com/x-dukjziclpwcycdop) has joined #raspberrypi
[10:01] * koaschten (~koaschten@31-16-0-231-dynip.superkabel.de) has joined #raspberrypi
[10:04] * jzu__ is now known as jzu
[10:04] * xmzgtx| (~xmzgtx@ has joined #raspberrypi
[10:05] * xmzgtx (~xmzgtx@ Quit (Ping timeout: 247 seconds)
[10:07] <Ben64> how can i hook up a small lcd to the rpi, and which ones are supported
[10:09] <tntexplosivesltd> what do you mean LCD?
[10:09] <tntexplosivesltd> like a 16x2 green/black one
[10:09] * prbz_ is now known as prbz
[10:09] <RaTTuS|BIG> yes and none and YMMV
[10:09] <tntexplosivesltd> or a full-colour one
[10:09] <tntexplosivesltd> LCD is ambiguous these days
[10:09] <Ben64> full color
[10:10] <Ben64> theres not any definitive info around for it
[10:10] <Ben64> some people say the iphone lcd would work
[10:10] <RaTTuS|BIG> well all lcds could work - you just need to be able to drive them somehow
[10:10] <Ben64> theres apparently a connector on the rpi called "dsi" but that is very hard to search for thanks to nintendo
[10:11] <tntexplosivesltd> Ben64: google supports -
[10:11] <tntexplosivesltd> so -nintendo omits nintendo from the search
[10:11] <Ben64> yeah theres just no info
[10:11] <Ben64> tons of "maybe"
[10:11] <tntexplosivesltd> =(
[10:11] <rm> Ben64, try searching by its full name https://en.wikipedia.org/wiki/Display_Serial_Interface
[10:11] <Ben64> i don't want to go buy an lcd for no reason
[10:12] <RaTTuS|BIG> Ben64 wait until you have a RPi in your hands ;-p
[10:12] <Ben64> RaTTuS|BIG: but shipping times...
[10:12] <SpeedEvil> Ben64: On the GPIO without too many problems
[10:12] <SpeedEvil> Ben64: Ignore the above statements
[10:12] <RaTTuS|BIG> I dont think rpis have direct support from the off for any DSI
[10:13] <Ben64> from the off?
[10:13] <RaTTuS|BIG> but you will be able to do it
[10:13] <RaTTuS|BIG> at launch
[10:13] <tntexplosivesltd> SpeedEvil: using what interface?
[10:14] <SpeedEvil> tntexplosivesltd: On the GPIO interface
[10:14] <SpeedEvil> http://www.ebay.co.uk/itm/20x4-LCD-Module-White-Characters-Blue-Backlight-HD44780-/250987761700?pt=UK_BOI_Industrial_Automation_Control_ET&hash=item3a70095024
[10:14] <rm> what "on the GPIO"?
[10:14] <Ben64> http://dmkenr5gtnd8f.cloudfront.net/wp-content/uploads/2011/12/PC236526.jpg
[10:14] <SpeedEvil> This sort of device uses ~8 pins or so.
[10:14] <Ben64> so on the left side there, that is the "DSI"
[10:14] <rm> the discussion is about a full colour LCD
[10:14] <SpeedEvil> And is very slow, so can be interfaced easily on bit-banged.
[10:14] <Ben64> on here, http://www.dealextreme.com/p/genuine-apple-iphone-3gs-replacement-lcd-screen-module-42065
[10:14] <Ben64> that connector looks very different
[10:14] <SpeedEvil> Oops
[10:14] <SpeedEvil> nvm
[10:15] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[10:15] <SpeedEvil> Sorry - I misread tntexplosivesltd's comments as Ben64
[10:15] <tntexplosivesltd> SpeedEvil: lol
[10:15] <tntexplosivesltd> that kind's easy
[10:15] <tntexplosivesltd> some even use UART XD
[10:15] <SpeedEvil> There are however similar small mobile phone LCDs.
[10:15] <tntexplosivesltd> but full colour =/
[10:15] <SpeedEvil> Which do do color.
[10:15] <SpeedEvil> full color
[10:15] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[10:15] <SpeedEvil> on SPI
[10:15] <kism3t> [Ben64]: personally i am going to use an arduino with the raspi to run the LCD, easy :)
[10:15] <Ben64> kism3t: that seems unnecessarily complex
[10:16] <SpeedEvil> http://www.adafruit.com/products/358
[10:16] <SpeedEvil> SPI full colour 1.8" display
[10:16] <kism3t> very simple the arduino has support for LCD and you can talk to the LCD over serial
[10:16] <SpeedEvil> (there are vendors with much cheaper similar ones than that)
[10:17] <Ben64> kism3t: but that would cost many times the price of the pi
[10:17] <Ben64> i guess i'll just have to wait a few months and see what people confirm to work
[10:18] <kism3t> ?13 for a mini pro and nice and small
[10:18] <tntexplosivesltd> ???_???
[10:18] <tntexplosivesltd> why...?
[10:18] * yumaokao (~yumaokao@ has left #raspberrypi
[10:18] <SpeedEvil> The above sort of color LCD is available from about $5 as spares.
[10:18] <SpeedEvil> Full color LCD
[10:18] <SpeedEvil> SPI.
[10:18] <Ben64> ???_???
[10:18] <tntexplosivesltd> Ben64: heh
[10:18] <Ben64> i can unicode too :P
[10:19] <tntexplosivesltd> SPI is win
[10:19] <zutesmog1> http://www.seeedstudio.com/depot/320240-graphic-lcd-w-touch-screen-and-hw-accel-p-147.html?cPath=163
[10:19] <SpeedEvil> http://www.dealextreme.com/p/repair-part-replacement-lcd-screen-modules-for-nokia-n95-31022
[10:19] * roman3x_ (~roman3x@bband-dyn18.95-103-42.t-com.sk) has joined #raspberrypi
[10:20] <SpeedEvil> (again, this will not work out-of-the-box, it will require some wiring to the pi, and proper setup of a driver.
[10:21] <Ben64> i'd really like to use the iphone4 screen
[10:21] <Ben64> its cheap, plentiful, and high dpi
[10:22] * Davespice (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) Quit (Ping timeout: 276 seconds)
[10:22] <SpeedEvil> Well...
[10:22] <SpeedEvil> That sort of stuff is unfortunately hard.
[10:22] <tntexplosivesltd> probably some proprietary interface
[10:22] <Ben64> some people have said it would just work
[10:23] <Ben64> thats why i'm wanting definitive information
[10:23] <Henchman21> woohoo a release date
[10:23] <RaTTuS|BIG> until the RPis are out inthe wild nothing can be definate
[10:23] <SpeedEvil> Indeed.
[10:23] <Ben64> there are quite a few of them around
[10:23] <RaTTuS|BIG> not release - built date
[10:23] <SpeedEvil> And more importantly, a statement that DSI is or is not enabled in the GPU
[10:23] <zutesmog1> You might find an N900 replacement screen might be easier to work with. http://www.natisbad.org/N900/n900-commented-hardware-specs.html
[10:24] <Ben64> SpeedEvil: they wouldn't put the connector on there otherwise
[10:24] <Henchman21> http://images.cheezburger.com/completestore/2011/10/2/9eabe8a9-3c55-4052-acc9-fb6c702d27ec.jpg
[10:24] <SpeedEvil> Ben64: that's not always true
[10:24] <RaTTuS|BIG> which connector Ben64
[10:24] <SpeedEvil> The n900 screen is also DSI
[10:25] <Ben64> not for every device ever, but for the pi, yes
[10:25] <SpeedEvil> Which is identical.
[10:25] <SpeedEvil> err - similar to the iphone
[10:25] <RaTTuS|BIG> http://dmkenr5gtnd8f.cloudfront.net/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf
[10:25] <SpeedEvil> I wish we had a proper datasheet that went into all the details
[10:26] <Ben64> http://www.globaldirectparts.com/v/vspfiles/photos/NKIA3063750-4.jpg
[10:26] <Ben64> n900 lcd ^
[10:27] * SpeedEvil has one 4mm from him.
[10:27] <zutesmog1> me too ;-) Well not the replacement screen , an N900 ;-)
[10:27] <SpeedEvil> Most n900s contain screens.
[10:27] <zutesmog1> ;-)
[10:29] <SpeedEvil> I'm unsure of the 'n900 contains screen with frameb uffer'
[10:29] <SpeedEvil> I've never seen anything of that in looking at / reading/writing schematics.
[10:29] <SpeedEvil> Or source.
[10:29] <Stskeeps> n900 screen doesn't have it's own framebufer
[10:30] <Stskeeps> / lcd controller
[10:30] <SpeedEvil> I think it's a misunderstanding of the fact that tehre are data channels to the LCD
[10:30] <SpeedEvil> But in the same way that camera chips have nice slow-speed comms on them, these don't have the picture going through them
[10:32] * smw (~smw@unaffiliated/smw) Quit (Ping timeout: 252 seconds)
[10:35] * Davespice (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) has joined #raspberrypi
[10:39] <zutesmog1> I gather we need the specs for the DSI interface, and then see what DSI LCD's (like the N900 one) could be compatible, or what sort of conversion would be required.
[10:40] <SpeedEvil> Indeed.
[10:40] <SpeedEvil> And the confrmation that you can in fact turn the DSI on in some manner.
[10:42] <zutesmog1> And if you don't mind the size this is the cheapest I have seen so far for a HDMI display http://www.kogan.com.au/shop/16-led-tv-pvr-and-hd-tuner-kogan-elite-led16/
[10:42] <zutesmog1> In Aus ;-)
[10:45] <SpeedEvil> It's kinda silly that 7" HDMI screens are more expensive than 17"
[10:45] <SpeedEvil> Well - 24" even
[10:45] <zutesmog1> isn't it, I would imagine it's down to volume.
[10:45] <SpeedEvil> yeah
[10:47] * xmzgtx| (~xmzgtx@ Quit (Remote host closed the connection)
[10:51] * traeak (~bolsen@75-151-91-165-Colorado.hfc.comcastbusiness.net) Quit (Ping timeout: 240 seconds)
[10:51] <zutesmog1> This LCD subsystem is used on Beagleboards. http://www.4dsystems.com.au/prod.php?id=114 has it's own embedded controller. But it's more expensive than an HDMI TV.
[10:51] <tntexplosivesltd> o_o
[10:52] <rm> oh, I know I know
[10:53] <rm> sec..
[10:54] <rm> https://www.dealextreme.com/p/400a-4-0-tft-lcd-digital-monitor-for-vehicle-parking-reverse-camera-1440x272-12v-dc-57776
[10:54] <rm> this!!!
[10:54] <rm> it takes analogue video in
[10:55] <rm> just like the one that R Pi has
[10:56] <rm> another one https://www.dealextreme.com/p/ep-130np-professional-3-0-wide-screen-tft-lcd-with-built-in-speaker-and-rca-video-input-ntsc-pal-15001
[10:56] * diplo- (~diplo@cpc1-trow1-0-0-cust13.aztw.cable.virginmedia.com) has joined #raspberrypi
[10:56] <zutesmog1> buts its 1440 x 272, but takes in PAL or NTSC analogue. Which means any display would scaled strangely
[10:56] <rm> http://s.dealextreme.com/search/lcd+monitor.html?category=718
[10:56] <rm> and some more
[10:56] <rm> zutesmog1, the 1440 is most likely bollocks
[10:56] <rm> https://www.dealextreme.com/p/3-5-tft-lcd-monitor-visual-reversing-vehicles-reverse-camera-ntsc-pal-dc8-15v-49796
[10:57] <rm> here's 320x240
[10:57] <rm> and much cheaper
[10:58] * P4R4N01D (~johndoe@ Quit (Quit: Konversation terminated!)
[10:59] * diplo (~diplo@cpc1-trow1-0-0-cust13.aztw.cable.virginmedia.com) Quit (Ping timeout: 252 seconds)
[10:59] * diplo- is now known as diplo
[11:01] * chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) has joined #raspberrypi
[11:03] <zutesmog1> i predict many people will be unhappy with these small analogue displays unless they are just playing video. A UI output over analog at PAL/NTSC then scaled down to 320 x 240 will not be very crisp ;-)
[11:05] <SpeedEvil> It depends.
[11:05] <SpeedEvil> If you're viewing a 3" display at arms length, it's not an issue
[11:05] <rm> I believe you can run 320x240 in the first place
[11:06] <rm> and plan your UI accordingly
[11:06] <SpeedEvil> Indeed.
[11:07] <SpeedEvil> Sometimes two buttons and three lines of 12 chars is all you want on the display
[11:07] <tntexplosivesltd> or the current time
[11:07] <tntexplosivesltd> XD
[11:07] <tntexplosivesltd> making a clock that's never wrong
[11:08] <jzu> how much characters (LINES/COLUMNS) can you get on a 320x240 in console mode?
[11:09] <zutesmog1> can the analog ouput actually output 320 x 240 or is it a native PAL/NTSC analog output.
[11:09] <jzu> s/much/many/
[11:09] <tntexplosivesltd> jzu: depeds on text size XD
[11:10] <tntexplosivesltd> * depends
[11:10] <tntexplosivesltd> 2
[11:10] <tntexplosivesltd> 7
[11:10] <jzu> yeah, right :-)
[11:10] <tntexplosivesltd> who knows
[11:10] <tntexplosivesltd> XD
[11:10] <tntexplosivesltd> probably a standard somewhere
[11:10] <SpeedEvil> jzu: 9*13 s a reasonable font size.
[11:10] <SpeedEvil> And no, it's actually analog
[11:10] <SpeedEvil> But you can pretty much resolve 320 lines of black and white
[11:11] <R`> are there small screns with hdmi?
[11:11] <SpeedEvil> that is - 160 lines acrosss the picture
[11:11] <R`> like 7-10" ?
[11:11] <SpeedEvil> 7" - sure
[11:11] * Datum_Errata (Datum_Erra@ has joined #raspberrypi
[11:11] <R`> where can i find them?
[11:11] <SpeedEvil> Just ebay hdmi car
[11:11] <SpeedEvil> or hdmi 7"
[11:11] <RaTTuS|BIG> http://www.lilliputuk.com/monitors/hdmi/668gl/
[11:11] <RaTTuS|BIG> not cheep but ....
[11:12] <jzu> a 5x7 font could do on this display
[11:13] <jzu> just tried xterm -fs 5x7
[11:13] <jzu> not pretty, but readable
[11:13] <SpeedEvil> It depends
[11:14] <SpeedEvil> It's readable on a display with a decent transfer function.
[11:14] <rm> 80x25 is the standard console resolution
[11:14] <rm> I think that fits 320x240 very well
[11:14] <SpeedEvil> rm: Not on 320*240
[11:14] <SpeedEvil> At all
[11:14] <SpeedEvil> 40*25, yes
[11:14] <rm> ok
[11:14] <SpeedEvil> 80*25 doesn't look at all good over an analog display
[11:15] <jzu> ...bringing back memories of an Oric-1 on the TV set...
[11:15] <jzu> (yes, I'm that old :-)
[11:15] * SpeedEvil found his ZX81 the other day
[11:15] <drazyl> 64x24 in 256x192
[11:15] <drazyl> (spectrum)
[11:15] <SpeedEvil> umm
[11:16] <SpeedEvil> the spectrum was not 64 chars was it?
[11:16] <SpeedEvil> I thought it was 32
[11:16] <drazyl> not for it's console
[11:16] <drazyl> but a bitmapped display is a bitmapped display
[11:16] <SpeedEvil> Ah
[11:16] <SpeedEvil> yes. That was shit and eyestrainy
[11:17] <drazyl> true
[11:17] <SpeedEvil> I forget the name of the wordprocessor
[11:17] <drazyl> but it worked
[11:17] * traeak (~bolsen@75-151-91-165-Colorado.hfc.comcastbusiness.net) has joined #raspberrypi
[11:17] <SpeedEvil> For stuff output over analog ports, you don't generally want to go below 8*8
[11:17] * smw (~smw@unaffiliated/smw) has joined #raspberrypi
[11:17] <SpeedEvil> (where analog = displays that are bandiwdth constrained)
[11:17] <SpeedEvil> Not VGA
[11:19] <jzu> composite
[11:20] <SpeedEvil> yes
[11:20] <SpeedEvil> A good example.
[11:20] <tntexplosivesltd> hmmmm bed time night all
[11:20] <SpeedEvil> Night.
[11:23] * Draio_ITA (51ae1f5c@gateway/web/freenode/ip. has joined #raspberrypi
[11:23] <Draio_ITA> hi
[11:24] <Draio_ITA> isn't it out yet?
[11:24] <ahven> yes, everyone has 10 of them and the next batch is available in December :)
[11:24] <SpeedEvil> Not everyone has 10.
[11:24] <SpeedEvil> I got 18.
[11:25] <ahven> hoarder :P
[11:25] <Draio_ITA> i was joking :-)
[11:25] <ahven> we weren't ;)
[11:25] <Draio_ITA> i know ... i know
[11:26] <Draio_ITA> so ... who yesterday bet on a specific date?
[11:26] * smw (~smw@unaffiliated/smw) Quit (Ping timeout: 248 seconds)
[11:27] <SpeedEvil> I still think 14th of march
[11:32] * prbz (~prebz@c83-254-51-59.bredband.comhem.se) Quit (Read error: Connection reset by peer)
[11:34] <Henchman21> heh
[11:34] <Henchman21> negative nancy
[11:36] * saua (~saua@193.81-167-3.customer.lyse.net) has joined #raspberrypi
[11:36] * rodrigo_golive (quassel@nat/indt/x-ipfirbytmrfecjin) has joined #raspberrypi
[11:41] * M0GHY (~peterholl@ has joined #raspberrypi
[11:41] * victhor (~victhor@ has joined #raspberrypi
[11:45] <Davespice> I had put a bet on the 1st of December =)
[11:45] * Davespice shrugs
[11:45] <Davespice> 2011...
[11:47] <Ben64> Feb 06 2012 05:57:13 <ukscone> PiBot: i'm betting on a march 3rd release date at 2:15pm
[11:47] * MinToN (~ident@cpc1-oxfd7-0-0-cust545.4-3.cable.virginmedia.com) has joined #raspberrypi
[11:47] <R`> i bet on 2:13
[11:47] <R`> =)
[11:48] * victhor (~victhor@ Quit (Ping timeout: 260 seconds)
[11:48] <Ben64> Jan 30 2012 06:11:37 <R`> anyone else except me that plans to build a pi cluster?
[11:49] <Ben64> you're gonna have to wait a long time : /
[11:51] <R`> did you just search your logs for my nick ? :P
[11:51] <Ben64> yep
[11:51] <Ben64> what would you do with a cluster anyway
[11:52] <Kolin> epeen
[11:52] <R`> dunno, render adult meterial maybe? :D:D
[11:52] <R`> material*
[11:52] <R`> nah i dont know, just for fun? :)
[11:53] <Ben64> it seems like you'd need a couple hundred to match a normal desktop power
[11:53] <R`> :(
[11:54] <rm> clusters are fun when they are SSI
[11:54] <R`> ssi?
[11:54] <rm> I only tried that briefly back when Open
[11:54] <rm> I only tried that briefly back when OpenMosix was alive *
[11:54] <SpeedEvil> What might be interesting inprinciple is using it as a content addressible memory
[11:54] <rm> R`, Single-System Image, where you can use it as a normal computer
[11:55] <rm> and run normal apps
[11:55] <rm> and they migrate to execute on other nodes automatically
[11:55] <R`> cool
[11:55] <rm> as opposed to the 'usual' and boring clusters, which you have to program using cluster-specific tools like OpenMPI
[11:55] <rm> and only that one special app will be paralleled
[11:56] <zutesmog1> amoeba - or whatever came along later
[11:57] <zutesmog1> http://www.cs.vu.nl/pub/amoeba/amoeba.html
[11:58] <zutesmog1> an amoeba cluster based on Sparc servers (each one a is quite slow compared to the cpu in a raspberry pi)
[11:59] * cerberos (~cerberos@ Quit (Ping timeout: 252 seconds)
[11:59] <zutesmog1> Here is a comparison I did a little while back between an raspberrypi and a SPARCstation 2 - just for laughs https://docs.google.com/spreadsheet/ccc?key=0At1LTa6ONStgdE5RalZ5clJFa1Z2YkozU3VsMjVpcVE
[11:59] <rm> it asks me to log-in into Google
[12:00] <zutesmog1> haven't seen a SPECint/fp for a rapsberrypi to really see how it compares. --
[12:00] <zutesmog1> oh sorry
[12:01] <zutesmog1> its google docs spreadsheet - try this https://docs.google.com/spreadsheet/pub?key=0At1LTa6ONStgdE5RalZ5clJFa1Z2YkozU3VsMjVpcVE&output=html
[12:02] <rm> ok, works
[12:03] <zutesmog1> just when I joined Sun, we had benchmarked an SS2 (for my previous employer) and we could get between 50-80 concurrent active users running (about 200 logged in users) running Claims, GL, and Insurance broking apps.
[12:03] <zutesmog1> I wonder if we could do that with a Pi
[12:03] <zutesmog1> ;-)
[12:04] * victhor (~victhor@ has joined #raspberrypi
[12:04] <rm> maybe not on an SD card
[12:04] <zutesmog1> yeah ;-)
[12:04] <rm> but an USB HDD could work
[12:04] <rm> most likely even an USB interfaced modern disk will be a lot faster than those old disks in that machine
[12:05] <rm> SD cards might have horrible erase block sizes to support concurrent random write effectively (or maybe not)
[12:06] <zutesmog1> Definatley. I haven't seen any specifics on I/O bandwidth, or hardware contexts etc... for a Pi. The SS2 had Sync SCSI2 interfaces, but realistically you could only get 1-3MB/sec tops with random I/O. A bunch of disks hanging off USB2 interface would be way faster.
[12:06] <rm> regarding RAM, it became clear today that the minimum GPU RAM portion is 32 MB
[12:07] <zutesmog1> I am pretty sure the old 104MB disks where <4000rpm
[12:07] <rm> and video won't be accelerated with that size
[12:07] <rm> so kinda unfortunate
[12:07] <rm> while 128 MB is already in the red zone, 96 MB is even more sharply into there
[12:07] <zutesmog1> Can you drop RAM from the GPU if your not using it ?
[12:08] <rm> 32 MB is the absolute minimum static allocation, from what I see in the datasheet
[12:08] <rm> it is decided at boot time
[12:08] <SpeedEvil> rm: you've got the unredacted datasheet?
[12:08] <rm> it's in the published one
[12:08] <SpeedEvil> Ah
[12:08] <zutesmog1> bummer, thought that might be the case. (My plans don't include using the GPU)
[12:09] <rm> The split between ARM and GPU memory is selected by installing one of the supplied
[12:09] <rm> start*.elf files as start.elf in the FAT32 boot partition of the SD card. The minimum amount
[12:09] <rm> of memory which can be given to the GPU is 32MB, but that will restrict the multimedia
[12:09] <rm> performance; for example, 32MB does not provide enough buffering for the GPU to do
[12:09] <rm> 1080p30 video decoding.
[12:09] <SpeedEvil> It'd be lovely to be able to tun the GPU off
[12:09] <SpeedEvil> A number of apps I have don't require it
[12:09] <rm> oh, btw
[12:10] <rm> there's a way to use video RAM as swap
[12:10] <mjr> haha
[12:10] <SpeedEvil> ahhahahahahahahhahha
[12:10] <zutesmog1> but does that mean 32MB is the minimum for a framebuffer that works, as opposed to just hardwiring a minimum of 32MB ?
[12:10] <rm> not sure if that will fly on the R Pi specifically
[12:10] * SpeedEvil remembers swapping on video RAM back in the day.
[12:10] <rm> https://wiki.archlinux.org/index.php/Swap_on_video_ram
[12:10] <mjr> apropos I also recall some patches to make swapping to Gravis Ultrasound memory possible
[12:10] <SpeedEvil> Back when 256M was important
[12:11] <SpeedEvil> 256K
[12:11] * kcj (~casey@unaffiliated/kcj1993) Quit (Quit: Procrast...)
[12:14] * jluisn (~quassel@ has joined #raspberrypi
[12:17] * smw (~smw@unaffiliated/smw) has joined #raspberrypi
[12:18] <zutesmog1> Does anyone know if all of the GPIO pins will be mapped to /sys/class/gpio ?
[12:19] <SpeedEvil> yes
[12:19] <SpeedEvil> The exposed ones
[12:19] <zutesmog1> and i2c ?
[12:19] <zutesmog1> etc...
[12:19] <zutesmog1> I did do a scan around for docs, but came up short
[12:19] <rm> don't try to access the GPIO pins, Neo; you just have to understand - there are no pins
[12:20] <zutesmog1> ;-) I can wield a soldering iron.
[12:21] <SpeedEvil> There are actual pins
[12:21] <zutesmog1> I am interested in sticking an RFM12B direct, rather than using an Arduino as a gateway to talk to all the other JeeNodes I have.
[12:21] <SpeedEvil> That was changed
[12:21] <zutesmog1> which is addressed over SPI
[12:22] <SpeedEvil> Tehre is SPI brogught out onto the GPIO connector. I'm unsure how it's used.
[12:23] <zutesmog1> Yeah saw that, can't wait to see more docs on this area - or getting hold of the hardware ;-)
[12:24] <SpeedEvil> There is a datasheet of sorts linked from the r-pi pages
[12:25] <zutesmog1> Yeah saw that, - more interested in api's The one example I have seen on the wiki is accessing gpio via memory mapped io rather than via dev filesystem.
[12:26] <zutesmog1> http://elinux.org/Rpi_Low-level_peripherals
[12:29] * smw (~smw@unaffiliated/smw) Quit (Ping timeout: 252 seconds)
[12:34] * oenime (~oenime@ has joined #raspberrypi
[12:34] <oenime> hello.
[12:35] <oenime> will the schematics for the raspberry pi pe public? (/are public)
[12:36] <rm> maybe
[12:37] * zutesmog (~timh@CPE-124-179-5-14.lns3.ken.bigpond.net.au) has joined #raspberrypi
[12:37] <Ben64> i believe they already are
[12:37] * zutesmog1 (~timh@CPE-124-179-5-14.lns3.ken.bigpond.net.au) Quit (Ping timeout: 244 seconds)
[12:38] <rm> but that's of debatable usefulness to most people, BCM2835 is not available in small quantities, and soldering is next-to-impossible
[12:39] <oenime> Ben64, got a link?
[12:39] <oenime> i'm not designing in small quantities, and i have access to pnp machiens and reflow owens.
[12:40] <ahven> afaik big quantities = starting 100k-1m pcs
[12:41] <Ben64> could just pay rpi the $25
[12:41] <oenime> can i buy 100k rpis?
[12:41] <ahven> eventually that could be possible, in theory :)
[12:42] <oenime> i need a home-designed/hacked sbc solution. raspberry looks nice, but we don't want to buy the boards from someone else when we can manufacture them
[12:42] <Ben64> i think they said a few months to get big numbers
[12:42] <RaTTuS|BIG> http://dmkenr5gtnd8f.cloudfront.net/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf
[12:42] <Ben64> you won't be able to do it any cheaper
[12:42] <SpeedEvil> Ben64: yes you will
[12:43] <oenime> i need a home-designed solution, or something evolving from an existing product.
[12:43] <SpeedEvil> Ben64: If you are sourcing in large volumes, and can leave off unrequired parts.
[12:43] <Ben64> manufacturing isn't cheap
[12:43] <SpeedEvil> Ben64: yes it is
[12:43] <SpeedEvil> Ben64: If you're making 50K.
[12:43] <SpeedEvil> The cost comes _way_ down
[12:43] <oenime> manufacturing isn't cheap.\
[12:43] <SpeedEvil> And no, it's not cheap
[12:44] <oenime> however, it's not on hobby/personal money.
[12:44] <SpeedEvil> But it's nowhere near as nasty as making one is
[12:44] <oenime> so - where can i find the schematics for the rpi?
[12:44] <Ben64> "we don't want to buy the boards from someone else when we can manufacture them"
[12:44] <SpeedEvil> I don't think they've been published.
[12:44] <oenime> hm.
[12:44] <Ben64> unless you outsource to china, it will be more expensive
[12:44] <RaTTuS|BIG> scamitics not yet available
[12:44] <ahven> nope
[12:44] <SpeedEvil> Ben64: well - obviously.
[12:44] <Ben64> which is what rpi does
[12:44] <Ben64> so whats the point of bypassing them
[12:45] <oenime> i'm not handling expenses, i'm just looking for a nice design.
[12:45] <SpeedEvil> Buying your own design of board based on the rpi, to your spec, assembled in china, is not the same thing.
[12:45] <SpeedEvil> For example - if you want to connect a specific display to the rpi, you may beed a $30 converter. Or you can plop a $3 chip on the same board, and the specific connector.
[12:45] <oenime> it could be cheaper - in theory - but that's not really what i mind.
[12:46] <SpeedEvil> And get it more reliable too
[12:46] <oenime> i just don't want to redisign it from scratch and mess around with kernel hacking for a few months.
[12:47] <RaTTuS|BIG> what are you wanting to do and how many are yoou wanting to make
[12:47] <SpeedEvil> oenime: I'm in a similar position, though I want to make ~10
[12:47] <SpeedEvil> oenime: More as a learning exercise.
[12:47] <oenime> we need a device like the rpi, small sbc, ethernet + sdcard + arm, but it should be home built. not my choice, really, company policy.
[12:48] <oenime> to the final design we'll need to add some other hardware, and we don't want to stack boards, but have everything on the same pcb.
[12:48] <zutesmog> do you actually need a gpu ?
[12:48] <SpeedEvil> Indeed - for some things, it's not sane to do otherwise.
[12:48] <SpeedEvil> Looked at the beaglebone?
[12:49] <zutesmog> if no gpu required thats what I was going to suggest. Schematics are published
[12:49] <oenime> nope, not really, just a webserver
[12:49] <oenime> beaglebone looks way overkill imo, studied it earlier.
[12:49] <SpeedEvil> The ethernet on the r-pi is on USB
[12:49] <SpeedEvil> that is probably a bad plan
[12:49] <oenime> mhm.
[12:50] * rafal_ (~rafal@ip-78-30-110-199.free.aero2.net.pl) has joined #raspberrypi
[12:50] <SpeedEvil> If you're designing from scratch, you'd likely want to find a SoC with an onboard phy.
[12:50] <oenime> are there any other sbc boards out there?
[12:50] <oenime> i know of the gumstix, but that's rather esoteric
[12:50] <SpeedEvil> Depressingly many.
[12:50] <zutesmog> Surely gutting beaglebone would be the easiest way. No GPU to worry about.
[12:50] <SpeedEvil> Almost every 32 bit microprocessor manufacturer has a board which can run linux.
[12:51] <SpeedEvil> It's usually $500 or so.
[12:51] <SpeedEvil> And sometime will come with a GPL kernel under NDA
[12:51] <oenime> yeah, that's why i liked rpi
[12:51] <SpeedEvil> The _nice_ part about r-pi is that you can base your design on an existing community, and you know if it's not booting, it's the circuit, not the software.
[12:52] * ksaua (~saua@193.81-167-3.customer.lyse.net) has joined #raspberrypi
[12:53] <zutesmog> beaglebon has on chip Phy, and it also has a community, (that doesn't mean you shouldn't look at r-pi - but I think the beaglebone doesn't have any close blobs - can anyone confirm this statement ;-)
[12:54] <oenime> no it doesn't, it has a 8710A phy chip
[12:54] * victhor (~victhor@ Quit (Read error: Connection reset by peer)
[12:55] * saua (~saua@193.81-167-3.customer.lyse.net) Quit (Ping timeout: 248 seconds)
[12:55] <zutesmog> oops sorry
[12:57] * victhor (~victhor@ has joined #raspberrypi
[13:00] <chris_99> whats Phy mean?
[13:00] <SpeedEvil> err - MAC, I mean, not PHY
[13:00] <SpeedEvil> PHY = physical layer of ethernet - it connects to the ethernet cable.
[13:00] <RaTTuS|BIG> what is that 486 embedded thing ...
[13:00] <SpeedEvil> MAC = logical side of internet
[13:00] <SpeedEvil> RaTTuS|BIG: bifferos
[13:00] <RaTTuS|BIG> yarg
[13:00] <chris_99> ah cheers
[13:01] <RaTTuS|BIG> http://bifferos.co.uk/
[13:03] <SpeedEvil> logical side of ethernet
[13:03] <SpeedEvil> I'm not awake
[13:03] <SpeedEvil> Packet framing and timing and stuff
[13:03] * Martix (~martix@134.89.broadband12.iol.cz) has joined #raspberrypi
[13:04] <oenime> hmm.
[13:04] <oenime> bifferboard seems pretty closed source
[13:07] <chris_99> that seems kind of cool, what processor is that, if its x86
[13:07] <RaTTuS|BIG> 486sx
[13:07] * koaschten (~koaschten@31-16-0-231-dynip.superkabel.de) Quit (Ping timeout: 244 seconds)
[13:07] <RaTTuS|BIG> ie no FPU
[13:08] <RaTTuS|BIG> I've never used one but 've just been asked here about something that that may be veryuseful for
[13:08] <chris_99> its not an intel one
[13:08] <chris_99> though
[13:08] <RaTTuS|BIG> no - but cloe enough
[13:09] <chris_99> i'm curious what it actually is
[13:09] <SpeedEvil> It's a part intended for stuff like thin clients
[13:09] <SpeedEvil> there is a website, from teh chipmaker somewhere linked
[13:09] <SpeedEvil> it's a small one in china
[13:10] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[13:11] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[13:12] * koaschten (~koaschten@ has joined #raspberrypi
[13:16] * smw (~smw@unaffiliated/smw) has joined #raspberrypi
[13:18] <chris_99> 'S3282 BGA CPU @ 150MHz (R861x/R321x equivalent). 486SX instruction set (no maths co-processor).'
[13:18] <chris_99> i can't find any info really on that chip
[13:26] * rafal__ (~rafal@ip-78-30-102-48.free.aero2.net.pl) has joined #raspberrypi
[13:27] * smw (~smw@unaffiliated/smw) Quit (Ping timeout: 245 seconds)
[13:27] * rafal_ (~rafal@ip-78-30-110-199.free.aero2.net.pl) Quit (Read error: Connection reset by peer)
[13:28] <koaschten> is it out yet?
[13:28] <PiBot> koaschten: The boards will be completed on February 20th.
[13:28] <koaschten> i mean, good morning :D
[13:28] <koaschten> ooh wait, where is my pan, i need to make some eggs :(
[13:28] * Caver (~Caver@nat1.nipltd.com) has joined #raspberrypi
[13:29] <RaTTuS|BIG> channel owner should update the topic to say Feb20
[13:29] <Caver> true
[13:30] <Caver> though thats the next estimate date I think
[13:30] <RaTTuS|BIG> yeah - keeps a running total as it where
[13:32] * WASDx (~AINOR@81-229-79-162-no64.business.telia.com) Quit (Ping timeout: 252 seconds)
[13:33] * WASDx (~AINOR@81-229-79-162-no64.business.telia.com) has joined #raspberrypi
[13:33] * Cemial (~hardcoreB@ip4da2366d.direct-adsl.nl) has joined #raspberrypi
[13:37] * drazyl (~drazyl@60-142.dsl.data.net.uk) Quit (Remote host closed the connection)
[13:38] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) Quit (Ping timeout: 260 seconds)
[13:39] * _spt_ (~postmaste@unaffiliated/-spt-/x-5624824) has joined #raspberrypi
[13:40] <_spt_> hello folks
[13:40] <_spt_> I've not been in here for a while.. whats new?
[13:40] * Datum_Errata (Datum_Erra@ Quit (Ping timeout: 252 seconds)
[13:41] <koaschten> just asj if it's out yet?
[13:42] * IT_Sean (~cdi-1fors@applefritter/IRCStaff/UltimateMacUser1) has joined #raspberrypi
[13:43] <Caver> _spt_, have a look at the website, there has been some recent news on things
[13:43] * Tachyon` (~tachyon@cpc1-york2-0-0-cust489.7-1.cable.virginmedia.com) Quit (Quit: Rebooting/Upgrading)
[13:44] * IT_Sean runs to the website
[13:46] <ukscone> morning all
[13:47] <IT_Sean> 'ello ukscone
[13:47] <koaschten> u no have twitter? ;)
[13:47] <IT_Sean> me?
[13:47] <IT_Sean> no... i do not twit.
[13:48] <ukscone> yeah IT_Seanisn't a twat a twit he's just american
[13:48] <IT_Sean>
[13:48] <ukscone> anyone else read the data sheet yet?
[13:49] <ukscone> i was looking for the framebuffer info but so far haven't spotted it just wondered if i was blind or it's not in there
[13:50] * Tachyon (~tachyon@cpc1-york2-0-0-cust489.7-1.cable.virginmedia.com) has joined #raspberrypi
[13:50] <RaTTuS|BIG> http://dmkenr5gtnd8f.cloudfront.net/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf
[13:50] <mjr> I doubt there's anything about graphics there, that's the secret crown jewels
[13:50] <RaTTuS|BIG> na nothing about FB other than 32MB is the smallerst it can go
[13:50] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) has joined #raspberrypi
[13:51] <koaschten> holy shit, that looks quite similar to the trafo blowup in december at that footballstadium... just a lot bigger.. http://www.youtube.com/watch?v=2C-xbcAJ9mE
[13:53] <huene> IT_Sean: you don't have to have an account to read the tweets - i also don't have one, you can read them on the twitter page
[13:53] <ukscone> mjr: there shoudl be something as we have already been told you "poke" a few bytws into a location and that sets up thee fb
[13:54] <Caver> yeah I couldn't see anything in it either
[13:54] <Caver> I guess there will be more info to come on the subject
[13:55] <Caver> a lack of kernel messages while it's booting, will be very annoying
[13:56] <Caver> lots of interesting stuff about GPIO though
[13:56] * drazyl (~drazyl@60-142.dsl.data.net.uk) has joined #raspberrypi
[13:59] <ukscone> who using linux at the moment wants to check that these commands do what he "claims" thy do? :) https://plus.google.com/111560558537332305125/posts/d64bp6rkZMC
[13:59] <friggle> framebuffer should be easy to work out from the raspi framebuffer driver in the sources on github
[13:59] <IT_Sean> sorry... running winblows
[14:00] <IT_Sean> :/
[14:01] <drazyl> ukscone - looks reasonable
[14:01] <ukscone> friggle: yup should be all in there i'm just too lazy to look and i like reading data sheets, design manuals and programming guides but kernel source just puts me to sleep
[14:03] <Caver> IT_Sean, oh thats easy then - cmd /c format c:
[14:05] <slaeshjag> ukscone: 7. doesn't work
[14:05] <Caver> luckly for most of those you need to be root, or sudo to do too much damage
[14:07] * Martix (~martix@134.89.broadband12.iol.cz) Quit (Ping timeout: 260 seconds)
[14:08] * octeris (~Adium@13-99-245.client.wireless.msu.edu) has joined #raspberrypi
[14:08] <jzu> only one of them works (or has worked in the past - when I tried it) when run as std user: the fork bomb
[14:08] <octeris> Is it normal behavior that I wasn't emailed by the mailing list about the recent blog post on rpi.org?
[14:08] <_spt_> thanks,
[14:09] <IT_Sean> Caver: i'm on my work laptop
[14:09] <IT_Sean> so... no reformatting C
[14:09] <Ben64> ukscone: you want people to actually try them? lols
[14:09] <Caver> octeris, yes I think the mailing list is much more for "it's released" ... type stuff, for more local type stuff, twitter feed is useful
[14:09] <Caver> IT_Sean, thats no fun
[14:10] <octeris> Caver: Okay, thanks.
[14:10] <ukscone> Ben64: it'll seperate the wheat from the chaff when looking for guinea pigs for other reasons :)
[14:10] <Caver> sniggers
[14:10] <IT_Sean> Caver: It also boots linux
[14:10] <IT_Sean> But, i needed to log into some windows only soffware this morning
[14:11] <Caver> ick
[14:11] <Caver> thats what vmware/virtualbox is for ...
[14:11] <Caver> or wine :P
[14:11] <IT_Sean> Unfortunatly, my boss is staunchly anti-linux
[14:11] <octeris> Caver: But what about video games? :P
[14:12] <haltdef> here be fanboys
[14:12] <Caver> what about them?
[14:12] <Ben64> ukscone: numbers 1-3,5-12 won't work right without sudo
[14:12] <octeris> Caver: Not all video games run under wine.
[14:13] <koaschten> ukscone wtf ... "yes > /dev/sda"
[14:13] * wiiguy (~fake@unaffiliated/wiiguy) has joined #raspberrypi
[14:13] <Ben64> koaschten: its a really stupid slow way to erase a drive
[14:13] <koaschten> This command fill your hard disk with the character 'y'
[14:13] <Ben64> almost as bad as /dev/urandom
[14:13] <koaschten> so whats the "yes" command doing?
[14:13] <Ben64> prints infinite "y"
[14:14] <koaschten> ...
[14:14] <Ben64> /dev/urandom speed = 47166464 bytes (47 MB) copied, 5.51668 s, 8.5 MB/s
[14:14] <jzu> a few seconds is enough to bork your partition table
[14:14] <Ben64> /dev/zero speed = 10885173248 bytes (11 GB) copied, 6.88593 s, 1.6 GB/s
[14:14] <Ben64> winner = /dev/zero
[14:14] <koaschten> dev/zero is not cpu bound Ben64
[14:15] <ukscone> number 4 was/is actually quite useful
[14:15] * Da|Mummy (~Veki@cpe-76-190-200-98.neo.res.rr.com) Quit (Ping timeout: 245 seconds)
[14:15] <Ben64> number 4 still works on many systems, at least temporarily
[14:15] <ukscone> used to use a variation of it which the mknod hack to priv escalate in the olden days
[14:15] <jzu> and /dev/urandom is actually useful to get rid of discs with potentially sensitive data on them
[14:15] <Ben64> mine for example will stop it after a minute or so
[14:16] <Ben64> jzu: not very useful, too slow to be effective
[14:16] <octeris> jzu: Probably better and faster tools for that than piping /dev/urandom to the hard drive.
[14:16] <jzu> I have done exactly that a few times
[14:16] <koaschten> jzu well data recovery actually is pretty overrated considering the amount over "over writes" to make data unrecoverable
[14:16] <octeris> jzu: Not saying you can't do it..
[14:16] <Ben64> not saying it won't work; saying its slow and lame
[14:16] <koaschten> just google "the great zero challenge"
[14:16] <ukscone> if you don't have a copy of dban around using urandom can do in a pinch
[14:16] <jzu> where I'm working, only the paranoid survive :-)
[14:17] <Ben64> take hard drive out of computer, place discs in microwave for 10 seconds
[14:17] <Ben64> problem solved
[14:17] <Ben64> or take a reciprocating saw through it
[14:17] <Ben64> or even a few holes with a drill
[14:18] <ukscone> Ben64: that's no fun -- take hd out and spend a few minutes introducing it to a sledgehammer and chisel are more fun
[14:18] <koaschten> http://hostjury.com/blog/view/195/the-great-zero-challenge-remains-unaccepted
[14:18] <jzu> unfortunately, I dont have htese tools at hand
[14:18] <drazyl> open up drive, remove platters, apply hammer
[14:18] <jzu> otoh I have a Linux computer
[14:18] <octeris> koaschten: A counterpoint? http://www.tablix.org/~avian/blog/archives/2008/10/the_not_so_great_zero_challenge/
[14:18] * octeris (~Adium@13-99-245.client.wireless.msu.edu) has left #raspberrypi
[14:19] * octeris (~Adium@13-99-245.client.wireless.msu.edu) has joined #raspberrypi
[14:19] <octeris> whoops, closed the wrong active window
[14:20] <Ben64> i always just tell people to zero a drive out
[14:20] <koaschten> octeris ...
[14:20] <Ben64> if someone wants your data that bad, then you're screwed anyway
[14:21] <koaschten> octeris what do you do if you don't have a screenshot?
[14:21] <IT_Sean> Ben64: not if you put the drive thru a woodchipper :p
[14:22] <Ben64> they could get to you other ways
[14:22] <koaschten> apply thermite, light it, enjoy show
[14:22] <Ben64> if you piss off the right people
[14:22] <octeris> koaschten: Ignore that shitty visual detective stuff. I was mostly interested in the idea that an data recovery company probably would have to spend more money to recover the drive than the prize money.
[14:22] <koaschten> http://www.youtube.com/watch?v=mgKk6vOVZAA
[14:23] <Ben64> i've always wanted to make thermite
[14:23] <Ben64> :O
[14:23] * ReggieUK (ReggieUK@5ac9ad24.bb.sky.com) has joined #raspberrypi
[14:23] <Ben64> i could make an iron case for the rpi!
[14:23] <Ben64> thermite + mold
[14:23] <koaschten> oh wait, myth busters http://www.youtube.com/watch?v=hw1xhRt9GvU&feature=related
[14:24] * datagutt (~datagutt@140.80-202-130.nextgentel.com) has joined #raspberrypi
[14:24] * datagutt is now known as Guest15120
[14:25] * Guest15120 (~datagutt@140.80-202-130.nextgentel.com) Quit (Client Quit)
[14:26] * datagutt_ (~datagutt@140.80-202-130.nextgentel.com) has joined #raspberrypi
[14:26] * datagutt_ (~datagutt@140.80-202-130.nextgentel.com) Quit (Changing host)
[14:26] * datagutt_ (~datagutt@unaffiliated/datagutt) has joined #raspberrypi
[14:26] * rafal_ (~rafal@ip-78-30-120-89.free.aero2.net.pl) has joined #raspberrypi
[14:26] <mjr> hmm, thermite case, in case your facilities are compromised
[14:26] * webuser2453 (57fca21c@gateway/web/freenode/ip. has joined #raspberrypi
[14:27] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) Quit (Quit: Leaving)
[14:27] <IT_Sean> I ould think that having a thermite case would likely compromise your facilities, at some point :p
[14:28] <IT_Sean> OOPS! I pressed the "blow shit up" button by accident!!!
[14:28] <octeris> IT_Sean: Doesn't that button involve a high amount of temperature and flames?
[14:28] * tsdedst (~tsdedst@a95-92-116-133.cpe.netcabo.pt) has joined #raspberrypi
[14:29] <Ben64> could have it controlled by the rpi
[14:30] * rafal__ (~rafal@ip-78-30-102-48.free.aero2.net.pl) Quit (Ping timeout: 272 seconds)
[14:32] * octeris (~Adium@13-99-245.client.wireless.msu.edu) has left #raspberrypi
[14:36] * _spt_ (~postmaste@unaffiliated/-spt-/x-5624824) has left #raspberrypi
[14:37] * Mazon (~Mazon@ has joined #raspberrypi
[14:45] * Tachyon is now known as Tachyon`
[14:48] * uriah (~uriah@unaffiliated/uriahheep) has joined #raspberrypi
[14:48] <uriah> so.... feb 20
[14:49] <Caver> hopefully :)
[14:49] <Kolin> yeah, they wont ship on the 20th
[14:49] <uriah> :P
[14:49] <uriah> indeed
[14:50] <Caver> IT_Sean or ukscone is it worth updating the channel title, with the joyous news?
[14:50] <uriah> but who knows, maybe they'll finish manufacturing before then
[14:50] <uriah> like... a few days beforehand
[14:50] <IT_Sean> topic is full.
[14:50] <IT_Sean> ... i believe.
[14:50] <ukscone> Caver: dunno know personally i'd prefer to change it to "bog off you annoying little toe rags" but i don't think it'll fit in the space we have left in the topic
[14:51] <IT_Sean> hhahahaha
[14:51] <Caver> a man who's tired of irc is tired of life ...
[14:51] <uriah> agreed
[14:51] <Caver> meh you can remove the may be kids here
[14:52] <Caver> the taster than chocolate dipped bacon, has to stay of course
[14:52] <ukscone> Caver: not tired of irc hate it. it all went downhill once the yanks got hold of the source code. it was a much nicer place when it was just australians and europeans on it and it was still called relay
[14:52] <IT_Sean> we still couldn't fit the 'bog off you...' bit
[14:52] <Caver> you'd have to explain what bog off means for starters :P
[14:53] <ukscone> IT_Sean: i'm thinking we shoudl password the channel. so you, shiftplusone and I need to have a committee meeting about it
[14:53] <uriah> wait, irc isn't called internet relay chat anymore?
[14:53] <uriah> since when?
[14:53] <uriah> <-- canada
[14:53] <ukscone> uriah: when it first appeared it was just called RELAY
[14:53] <Kolin> its called mirc duh
[14:53] <Caver> I think ukscone is showing his age :P
[14:53] <RaTTuS|BIG> it all went downhill from chat actully
[14:53] <uriah> :P
[14:54] <ukscone> and it was mostly accessible via JANET, KOMETH, BITNET and IPSS
[14:54] <Caver> why is something getting worse "downhill" surely it ought to be uphill ...
[14:54] <uriah> RaTTuS|BIG: pulling my leg, are ya?
[14:54] * webuser2453 (57fca21c@gateway/web/freenode/ip. has left #raspberrypi
[14:54] <IT_Sean> ukscone: i'm off to a meeting... Perhaps this weekend? or an evening ?
[14:54] <ukscone> well KOMETH wasn't really a network, more of a gateway but it got you places
[14:54] * Caver shudders at the distant memories of TCP/IPX and PAD's out to JANET etc
[14:54] <RaTTuS|BIG> ;-p yeah
[14:54] <uriah> :P
[14:55] <ukscone> Caver: remember the tymnet outdials?
[14:55] * IT_Sean (~cdi-1fors@applefritter/IRCStaff/UltimateMacUser1) has left #raspberrypi
[14:55] <uriah> ukscone: oic. didn't know that
[14:56] * jmontleon (~jmontleon@c-76-19-35-6.hsd1.ma.comcast.net) has joined #raspberrypi
[14:56] <Caver> ukscone, only if you force me to at gun point
[14:56] * Caver defers to ukscone's longer unix beard
[14:56] * cerberos (~cerberos@ has joined #raspberrypi
[14:57] <Caver> outdials - as in being able to telnet over to a pad where you could call a local area number with a modem?
[14:57] <ukscone> Caver: i had it cut yesterday. people start calling the police on me for being a terrorist if i let it grow too long
[14:58] <Caver> sacrilege ...
[14:58] <ukscone> i have it shaved every two years whether it needs it or not
[14:58] <ukscone> Caver: once i go from being a charlie manson lookalike through the saddam lookalike stage and start to get to the gandalf lookalike stage
[14:59] <Caver> *grins* ... it's charactor
[14:59] * vipkilla (~chatzilla@unaffiliated/t-dot-zilla/x-2830497) has joined #raspberrypi
[14:59] <uriah> ukscone: reminds me of napoleon :P
[14:59] <ukscone> or people start going "ruh roh shaggy" behind my back when i am walking in the street
[14:59] <uriah> "whether it needs it or not"
[14:59] <Caver> personally I always liked the one about Sun Microsystems banning random drug testing, as it was removing way too many of their best programmers
[14:59] <ukscone> the 2 years whether i need it or not works for socks and underwear too
[15:00] * Caver has gone for the traditional geek's goaty style of facial fur
[15:00] <Caver> of course
[15:00] <uriah> bbl
[15:00] * uriah (~uriah@unaffiliated/uriahheep) Quit (Quit: leaving)
[15:00] <Caver> complete with jesus creeper sandles?
[15:01] <ukscone> Caver: i once cause a drug problem at a governement lab in the states. sent a friend some bags of hops from the uk to the states at his work address. the drug sniffing dogs thought it was pot and my friend and 3 others got called in to the security heads office and asked why they were ordering pot from the uk
[15:02] <Caver> what did you reply?
[15:02] * whatisnot_ (ada79aac@gateway/web/freenode/ip. has joined #raspberrypi
[15:03] <whatisnot_> quarts crystal manufacturing delay?
[15:03] <Thorn_> terrible excuse
[15:03] <Thorn_> rpi is vapourware
[15:03] <Thorn_> you might as well all unsubscribe from the mailing list right now...
[15:04] <Caver> lol
[15:04] <Caver> fine by me!
[15:05] <Caver> I have to admit, it's quite a star trek sounding excuse
[15:05] * vipkilla (~chatzilla@unaffiliated/t-dot-zilla/x-2830497) has left #raspberrypi
[15:05] <whatisnot_> is the soft shipment date still around feb 20?
[15:05] <Thorn_> no
[15:05] <Thorn_> june
[15:05] <rm> which year
[15:05] * rzr (~rzr@rzr.ww7.be) has joined #raspberrypi
[15:05] <ukscone> Thorn_: it's all your fault.
[15:06] <Thorn_> is it out yet
[15:06] <PiBot> Thorn_: The boards will be completed on February 20th.
[15:06] <Thorn_> what
[15:06] <rzr> hi, is the fosdem presentation online ? i'd like to share it w/ pple
[15:06] <Thorn_> u'd like to share it w/ apple?
[15:06] <rzr> :)
[15:07] <Caver> hmmm apple pi has a certain ring about it
[15:07] <rzr> stop w/ fruits
[15:08] <Thorn_> never liked fruitshoot, odd taste
[15:08] <whatisnot_> too american sounding, they need something more like where the board is being made, maybe "every other thing in the world" pi
[15:08] <Caver> how about with cream on top?
[15:08] <Caver> which is all fine until you remember the pi is in fact python
[15:08] <rzr> would qtmediahub support harmattan ?
[15:08] <ukscone> really poor spelling skills in the foundation
[15:09] <Thorn_> raspberry piss
[15:09] <Thorn_> mmm
[15:09] <drazyl> loganberry pi
[15:10] <whatisnot_> I wonder if they're ever going to come out with the usb stick version the origional was
[15:10] <ukscone> w00t! raspi story on slashdot and i have mod points
[15:10] <Caver> unlikely
[15:11] <Caver> oh now ... cue web server over loading in 10 .. 9 .. 8 .. *sniffs* ... 6 ... - can anyone smell burnt fruit?
[15:12] <Thorn_> you do realise someone's going to do the two electrodes thing and run a raspberry pi off of raspberries now...
[15:12] <Thorn_> it's the potato web server all over again ;\
[15:13] <Caver> oooh ... I think your onto something there
[15:13] <Caver> :)
[15:14] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[15:15] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[15:15] * atts (~asheehan@static-99-136-171-68.axsne.net) has joined #raspberrypi
[15:16] <whatisnot_> Thorn: do you mean the mesh potato: http://villagetelco.org/mesh-potato/
[15:16] <whatisnot_> I think RP would be great for that.
[15:21] <Caver> 1W ... might take quite a few raspberries
[15:21] * cerberos (~cerberos@ Quit (Quit: cerberos)
[15:22] <Caver> according to the lemon battery on wiki, we'd need the equiv of 5000 lemons to sustain 1W of power ...
[15:22] <Caver> not sure the lemon to raspberry conversion factor though
[15:22] * ksaua_ (~saua@193.81-167-3.customer.lyse.net) has joined #raspberrypi
[15:24] <Thorn_> you can't mix fruit like that
[15:24] <Thorn_> that's inzest
[15:24] <Thorn_> sorry :(
[15:26] <jzu> must be related to the fruit's pH
[15:26] * ksaua (~saua@193.81-167-3.customer.lyse.net) Quit (Ping timeout: 260 seconds)
[15:27] * rafal__ (~rafal@ip-78-30-111-164.free.aero2.net.pl) has joined #raspberrypi
[15:28] <Caver> *groans*
[15:29] <jzu> lemon juice has a pH of 2 to 2.6, rasperries are between 3.2 and 4
[15:30] * rafal_ (~rafal@ip-78-30-120-89.free.aero2.net.pl) Quit (Ping timeout: 252 seconds)
[15:30] * jzu would do anything rather than work today
[15:33] * R` (~RHA@ip7.j-k.kund.riksnet.nu) Quit (Ping timeout: 276 seconds)
[15:34] <Caver> lol ... clearly
[15:35] * roman3x_ (~roman3x@bband-dyn18.95-103-42.t-com.sk) Quit (Quit: Leaving)
[15:40] * DexterLB (~angel@ Quit (Read error: Connection reset by peer)
[15:45] * DexterLB (~angel@ has joined #raspberrypi
[15:47] * troth (~troth@nat/hp/x-pitruzulluptuahv) has joined #raspberrypi
[15:50] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[15:50] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[15:58] * jzu__ (~jzu@ has joined #raspberrypi
[15:59] * ShiftPlusOne (~Shift@124-168-104-90.dyn.iinet.net.au) has joined #raspberrypi
[16:01] * jzu (~jzu@ Quit (Ping timeout: 272 seconds)
[16:02] <OneFix_Work> So, it will be the 20th before the first run is finished...
[16:04] <atts> supposedly
[16:04] <atts> and on sale by the end of the month
[16:05] <OneFix_Work> Hopefully this is the last setback...not upset just anxious to get my hands on a board :)
[16:05] <ShiftPlusOne> setback, what setback?
[16:05] <atts> what are you planning to do with your rpi?
[16:05] * imnichol (~ian@somsen-129-32.winona.edu) has joined #raspberrypi
[16:05] <SpeedEvil> The first run being finished does not inherently imply that is the first point at which they can go on sale.
[16:05] <Caver> I want to make a automated garrage door opening controller
[16:05] <whatisnot_> shiftplusone: quarts crystal manufacturing delay
[16:06] <OneFix_Work> ShiftPlusOne: The fabricator had problems sourcing a part in China
[16:06] <ShiftPlusOne> ah yeah, that
[16:06] <OneFix_Work> I'm thinking about using my first one as a ZoneMinder server
[16:06] <atts> they mention it being on sale by the end of the month in the article, but still tentative i guess
[16:07] <ShiftPlusOne> Not really a setback... something which should've been accounted for by Hofstadter's law.
[16:07] <Caver> so 20th = finished in china, plus air fraiting time to I presume the UK, then final dispatching
[16:07] <whatisnot_> sounds about right
[16:08] <whatisnot_> are they being shipped directly from china though?
[16:08] <ShiftPlusOne> whatisnot_, nope
[16:08] <Caver> how do you mean?
[16:08] <ShiftPlusOne> to UK, then distributed from there
[16:08] <whatisnot_> ahh
[16:08] <whatisnot_> thanks
[16:09] <ShiftPlusOne> (correct me if I am wrong)
[16:09] <OneFix_Work> ShiftPlusOne: Well, this issue seemed to be related to the fact that China seems to be ahead of the curve in getting the newest, cheapest parts
[16:09] <Caver> so end of month for UK I think is reasonible - perhaps a couple days longer for USA and further afield
[16:10] <whatisnot_> I'm guessing they'll take all the orders then start shipping when supply runs dry, not ship per order. But who knows.
[16:10] <ShiftPlusOne> nuh, that would be a shipping nightmare... much easier to send them out a few times a week.
[16:11] <OneFix_Work> whatisnot_: It also depends on if they plan on shipping from China initially, since they will be there, it seems almost a waste to ship from China -> UK -> Somewhere else
[16:11] * Caver raises hand to help packing 'n' dispatching
[16:11] * UnderSampled (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) has joined #raspberrypi
[16:11] <Caver> this is of course assuming they don't get mugged from the airport
[16:11] <whatisnot_> OneFix_Work: that's what I was thinking
[16:12] <Caver> OneFix_Work, I'm 99% sure they're comming to the UK 1st
[16:12] <whatisnot_> but that cuts out a lot of QAQC hands on time
[16:13] <ukscone> hi UnderSampled -- how goes the zipit hacking?
[16:14] <OneFix_Work> whatisnot_: That's what they said in the post, but sell a few hundred as "Ship from China" at a markup to cover travel expenses
[16:14] <ukscone> http://www.raspberrypi.org/forum/general-discussion/new-post-on-the-front-page-a-date-thatll-interest-you-and-a-datasheet-that-will-too/page-7#p39061 hehehehehhehehehe
[16:14] <Caver> lol
[16:15] <whatisnot_> interesting, so "ship from china" with a markup and less QAQC... sounds.... bad
[16:15] <Caver> OneFix_Work, is this some cunning plan to get yours before we get them in the uk?
[16:15] * taaz (~dlehn@pool-96-240-177-123.ronkva.east.verizon.net) Quit (Ping timeout: 240 seconds)
[16:16] <SpeedEvil> whatisnot_: QA is a per-board process.
[16:16] <SpeedEvil> whatisnot_: they are not going to hook them up for a couple of weeks running memtest.
[16:17] <whatisnot_> ohh i doubt that, I'm guessing 10% on the chinese side, random 10% on the UK side, it's just not cost effective to test every single part of 10k boards
[16:17] <OneFix_Work> Caver: Well, my point is that we know some people were willing to pay like thousands of dollars for reference boards, so I just figured some people would be willing to pay ~$75/$100 to get them shipped direct from China...I actually probably wouldn't pay extra to get mine first, but I'm sure enough people would...
[16:17] <Caver> mem test on 256Mb ... lol
[16:18] <OneFix_Work> SpeedEvil: It makes more sense to pull random samples...like 20 or 30 boards, and then do more testing if you come up with any problems in the random testing
[16:19] <SpeedEvil> OneFix_Work: I'd actually be doing every board.
[16:19] <whatisnot_> OneFix_Work: exactly, that's how almost all inspection QAQC works.
[16:19] <SpeedEvil> OneFix_Work: If you do it right, it's ~10s/board.
[16:20] <Caver> well I guess with a Pi it's fairly easy to make up a SD card, plug in both power and ethernet, and get it to self run tests, and put the result to a centeral server db, with the mac address as the key
[16:20] <SpeedEvil> And the fixture for the pi wouldn't be very complex.
[16:20] <Caver> do it as they get marked up for sending out
[16:20] * UnderSampled (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) Quit (Read error: Connection reset by peer)
[16:21] <SpeedEvil> Caver: for example - plug in SD, press lever on jig which inserts all cables.
[16:21] <SpeedEvil> Caver: device boots, uses USB webcam to look at HDMI screen, sees it's good, goes beep through network device, unplug, ship.
[16:21] <Caver> yup
[16:22] <Caver> I've done it manually with floppy disks (oh boy was that dull!)
[16:22] <whatisnot_> and all of this is why I'm more than likely going to wait for RPv2 or even 3 before I buy over 10 of them, 1 at v1
[16:22] <Caver> I used to use all the PC's in the office to do the occasional disk duplication run
[16:22] <SpeedEvil> Caver: BTDT
[16:23] * taaz (~dlehn@pool-96-240-177-106.ronkva.east.verizon.net) has joined #raspberrypi
[16:23] <Caver> :) you poor bastard
[16:23] * taaz (~dlehn@pool-96-240-177-106.ronkva.east.verizon.net) Quit (Client Quit)
[16:23] <SpeedEvil> Though in my case, it was floppies downloading slackware disks.
[16:23] * mchou (~quassel@unaffiliated/mchou) has joined #raspberrypi
[16:24] <SpeedEvil> Point each PC at a different file
[16:24] <Caver> nods :) done
[16:24] <mchou> so the RP has no standoff holes on the board?
[16:24] <Caver> mchou, none ... I guess it's hot glue, or friction mounts
[16:25] <mchou> bah!
[16:25] <whatisnot_> didn't even think of that
[16:25] <OneFix_Work> Caver: BTW, it would be easier to make up a USB flash drive to boot for QA testing
[16:25] <OneFix_Work> BTW, any ideas when the GertBoard will be available for purchase?
[16:26] <Caver> OneFix_Work, you'd still need a SD card .. remember the GPU boots 1st looks at the SD card and runs what ever is on there as a boot loader for the CPU
[16:26] <Caver> just putting a usb drive it, and no SD card, ... nothing will happen
[16:26] <mjr> did I mention by the way that that's a little perverse? ;)
[16:27] * taaz (~dlehn@pool-96-240-177-106.ronkva.east.verizon.net) has joined #raspberrypi
[16:27] <Caver> heheh ... it is, but there you go
[16:27] * rafal__ (~rafal@ip-78-30-111-164.free.aero2.net.pl) Quit (Read error: Connection reset by peer)
[16:27] <Caver> OneFix_Work, no I don't know anything re gertboard yet
[16:28] <mchou> ok, a lil question abou how the RP boots...Is it possible to netboot via PXE?
[16:28] <Caver> http://elinux.org/BCM2835_datasheet_errata that was quick
[16:28] <mjr> I suppose it fits with the GPU being the main CPU and the ARM core being just ancillary ;]
[16:28] * rafal_ (~rafal@ip-78-30-122-13.free.aero2.net.pl) has joined #raspberrypi
[16:28] <mchou> who needs SD when you have NAS!
[16:28] <Caver> mchou, not directly, but nothing stoppnig you making a SD card that boots up, then sucks a PXE image down and reboots into that
[16:29] <Caver> like I say the booting is very primative - it's literally on, read SD card, run that ... no options, no user bios, no choices
[16:29] <Caver> (the added advantage, that it makes it harder for kids to mess up!)
[16:29] <Caver> http://elinux.org/BCM2835_datasheet_errata <-- ok that was a) nerdy b) fast
[16:30] <OneFix_Work> Caver: Guess you're right...I suspect the first run will be the only run that will need testing in China, but later runs will just get tested in the UK, and hopefully eventually, they can be made outside of China
[16:31] <Caver> for the PXE idea you wouldn't need a very large or fast SD card, so it's not a major problem!
[16:31] <Caver> plus when it doesn't work ... you just pull out the SD card and put in a bigger one with full OS on it, and ... you can see whats up
[16:31] <mchou> Caver: you can tell boot loader to boot off usb instead of SD though, right?
[16:31] <OneFix_Work> Caver: Yea, PXE is probably the way to go...you could do it with a small laptop as the PXE server...
[16:32] <Caver> mchou, no only off SD
[16:32] <OneFix_Work> mchou: You would still have to boot from the SD and then you could reboot to the USB
[16:32] <Caver> like I say the boot process is fixed in the GPU's rom
[16:32] <mchou> oh, didn't realize that
[16:33] <haltdef> you should be able to have a tiny sd with uboot on it or something that loads an OS on a usb drive
[16:33] <Caver> thats ok
[16:33] <ShiftPlusOne> pxe? what are you talking about? There's no PXE for ARM. O_o
[16:33] <OneFix_Work> Remember, this chip was originally meant to work only as an embedded chip
[16:33] <whatisnot_> there's no way to reflash the gpu?
[16:33] <Caver> nope
[16:33] <Caver> its a rom
[16:34] <ShiftPlusOne> just specify the rootfs as the USB stick and you're done... no need for uboot or anything fancy
[16:34] <Caver> ShiftPlusOne, well not literally PXE ... I just meant as a general idea
[16:34] <ShiftPlusOne> or an NFS share if you want to boot off a network
[16:34] <Caver> tftp :)
[16:34] <haltdef> that's true, kernel will easily fit on a 128MB SD won't it
[16:34] <haltdef> silly haltdef
[16:35] <ShiftPlusOne> whatisnot_, nope, and if there was, it would have to be signed.... so there would still be a fair bit of hacking to do.
[16:36] <whatisnot_> so it's not impossible, just improbable
[16:36] <Caver> signed?
[16:36] <Caver> hmm ... as far as I know it's a ROM, not flash ...
[16:37] <ShiftPlusOne> actually yeah, the files which the GPU uses are checked to make sure they're proper
[16:37] * tom_say (~pain@cpe-68-203-248-184.stx.res.rr.com) has joined #raspberrypi
[16:37] <ShiftPlusOne> so if you want to modify them, it probably won't work
[16:37] <whatisnot_> what is the function? Does it just initiate hardware then handoff to the OS on the SD card?
[16:37] <ShiftPlusOne> i forget what that security method is called... first something or other....chain... I don't know.
[16:38] * customtronics (~puppy@user-12l2tpe.cable.mindspring.com) has joined #raspberrypi
[16:38] * oenime (~oenime@ Quit (Ping timeout: 244 seconds)
[16:38] <OneFix_Work> whatisnot_: My guess is it's reading the microcode for the chip off of the SD Card...which could also contain licenses for CODECs and what not
[16:39] <ShiftPlusOne> whatisnot_, that's exactly right.... it loads up the firmware for the GPU and loads the kernel. Someone has looked at some calls found within the files, and there is some OpenMAX, FAT stuff and so on.
[16:39] <whatisnot_> so even the kernel won't be replaceable/reflashable?
[16:39] <whatisnot_> this is sounding less and less open.
[16:40] <ShiftPlusOne> "Chain of trust" that's what I was looking for. http://en.wikipedia.org/wiki/Chain_of_trust so yeah, modified code won't work.
[16:40] <ShiftPlusOne> whatisnot_, kernel is fully replacable
[16:40] <ShiftPlusOne> whatisnot_, doesn't even need to be a linux kernel
[16:40] <OneFix_Work> whatisnot_: No, the point is that everything will be replaceable...down to the microcode
[16:41] <whatisnot_> "everything will be replaceable" except the things that are not?
[16:41] * nemrod (~nemrod@unaffiliated/nemrod) Quit (Read error: Operation timed out)
[16:41] <OneFix_Work> whatisnot_: Most people are used to having intel chips with a BIOS that already contains a base microcode for the processor...this simply loads everything off of the SD card first
[16:41] <ShiftPlusOne> whatisnot_, everything except for the GPU firmware (for now). That's how it has always been with ALL hardware, so it's nothing new.
[16:41] <mjr> OneFix_Work seems a bit confused. The GPU stuff is proprietary as hell, even on the Linux side. And the GPU controls boot. But you can make it start any kernel you want afterwards.
[16:42] * nemrod (~nemrod@unaffiliated/nemrod) has joined #raspberrypi
[16:42] <whatisnot_> I realize all of that, someone does need to play angel's advocate once in a while
[16:43] <mjr> (personally I'd feel better if just the Linux side drivers and libraries were free even if there were proprietary firmware still, but meh)
[16:43] <OneFix_Work> mjr: Not exactly....the GPU is proprietary, probably because of the licensable CODECs that is contains...
[16:43] <ShiftPlusOne> yeah, the userspace blobs... I am not a fan of, but they don't really get in the way.
[16:44] <ukscone> ctyler: ping
[16:44] <ShiftPlusOne> OneFix_Work, no, the GPU is proprietary because Broadcom doesn't want competitors to steal their ideas... so keeping it as closed as possible minmizes that risk.
[16:44] <ukscone> noone other than ctyler in here has an alpha or betaboard do they?
[16:44] <OneFix_Work> mjr: The SD card really contains the "brains" of the processor...because presumably, you could turn off certain features of the chip to save on power if you really wanted to...
[16:45] * oenime (~oenime@ has joined #raspberrypi
[16:46] <ShiftPlusOne> ukscone, try vgrade and that guy who bought the ebay one... forget the name...
[16:46] <whatisnot_> What are some of the planned projects all of you are thinking about?
[16:47] <ukscone> ShiftPlusOne: tah -- forgot about vgrade and really need someone with the "standard" distro too but if i build static it should be ok
[16:48] <ShiftPlusOne> MagicalTux, ping?
[16:48] <ShiftPlusOne> ukscone, yeah, MagicalTux was the other one.
[16:49] * victhor (~victhor@ Quit (Ping timeout: 276 seconds)
[16:49] <ukscone> oh great. nothing serious just want them to check devmem2 before i start work on porting my pxa2xx register/gpio bit twiddiler to the bcm2835
[16:51] * whatisnot_ (ada79aac@gateway/web/freenode/ip. Quit (Quit: Page closed)
[16:52] <mchou> sigh
[16:52] <mchou> so close and yet so far
[16:57] * jzu__ is now known as jzu
[16:58] * jzu is now known as Guest23400
[17:02] <ukscone> i know it's the sun newspaper but ROTFLMFAO!!!!!! http://www.thesun.co.uk/sol/homepage/news/3544485/Thieves-targeting-vans-on-council-estate-are-confronted-by-four-SAS-men-on-a-stakeout.html
[17:04] * Faperdaper (~Faperdape@dhcp-077-250-027-109.chello.nl) has joined #raspberrypi
[17:05] * cuesilence (ada79aac@gateway/web/freenode/ip. has joined #raspberrypi
[17:05] * gabriel9 (~quassel@ Quit (Read error: Connection reset by peer)
[17:07] <Aquilus_> Haha
[17:14] * djazz (~djazz@78-72-49-85-no186.tbcn.telia.com) has joined #raspberrypi
[17:18] <Caver> ukscone, that was april last year!
[17:19] <ukscone> Caver: yeah but it takes time for news to reach the colonies
[17:20] <ukscone> and also shows how often i read the sun :D as i have better taste
[17:20] <Caver> was funny though .... "a bit of a slap" I wonder if they'll ever walk again
[17:20] <Caver> at least one a year, so don't get too high 'n' mighty :P
[17:20] <ukscone> it's like a thing that happened in NYC last year
[17:20] <ukscone> a guy threatened 4 people
[17:20] <Caver> oh?
[17:21] <ukscone> who jusy happened to be marines
[17:21] <Caver> *sniggers* poetic justice then
[17:21] <ukscone> the police blotter in the newspaper said when police arrived they took him to hospital for his injuries
[17:22] <ukscone> broken leg, 5 missing teeth, fractured ribs, broken nose, and other injuries
[17:22] <Caver> "he tripped on the cell stair case" "you honor"
[17:22] <ukscone> the police report responded with "it is believed the guy had fallen off the curb and caused those injurires himself"
[17:22] <Caver> "repeatedly"
[17:23] <Caver> :) close then
[17:23] <ukscone> yup
[17:23] <ukscone> one marine had a small knife cut on his hand
[17:23] <koaschten> UK got some serious curbstones o.O
[17:23] <koaschten> ;)
[17:23] <ukscone> and was taken to hospital for a bandaid
[17:23] <SpeedEvil> 'After he was released, he was later re-arrested, and charged with damaging police truncheons.
[17:24] <Caver> sucks teeth ... not something to be taken lightly
[17:25] <cuesilence> RP article on ostatic: http://ostatic.com/blog/low-cost-raspberry-pi-linux-devices-to-arrive-this-month
[17:27] <Caver> I think "supprisingly powerful" @ 700MHz might be overstating things a bit, but oh well
[17:28] <SpeedEvil> It's fastish for a 700MHz device in some cases
[17:28] <cuesilence> "In other words, this ultra low-cost device isn't as underpowered as you might think. In fact, it could possibly meet the computing needs of millions of people who could never afford computers. " agreed.
[17:29] * M0GHY (~peterholl@ Quit (Quit: Leaving.)
[17:29] * rafal__ (~rafal@ip-78-30-102-52.free.aero2.net.pl) has joined #raspberrypi
[17:29] <Caver> well it's defo fast in comparison to your typical electronics dev board
[17:29] <slaeshjag> The CPU is kinda crappy (it's a ARM11..), but the GPU makes up for it
[17:29] <SpeedEvil> Bullshit.
[17:29] <SpeedEvil> If you can't afford a computer, you can't afford a HDMI monitor
[17:29] <Caver> LOL
[17:30] <Caver> I just checked .... I've only got one hdmi screen in the house
[17:30] <Caver> ah well
[17:30] <cuesilence> Then that's the next step, a 7 to 10 inch monitor with hdmi in for as cheap as possible
[17:30] <SpeedEvil> cuesilence: Indeed.
[17:31] <SpeedEvil> CuriosLion: Ipad2 interface board.
[17:31] <Caver> heheh 1080p @ 10" would be true retina stuff
[17:31] * Edek (~Edek@5357E55B.cm-6-8d.dynamic.ziggo.nl) has joined #raspberrypi
[17:31] <SpeedEvil> ^display/touchscreen
[17:31] <cuesilence> naa, i'm guessing most people would be fine with 800x480 @ 10", I would be.
[17:32] <SpeedEvil> The ipad display may actually be cheaper
[17:32] * rafal_ (~rafal@ip-78-30-122-13.free.aero2.net.pl) Quit (Ping timeout: 272 seconds)
[17:32] <cuesilence> How much is a replacement ipad display? is it compatible?
[17:32] <SpeedEvil> http://www.ebay.co.uk/itm/Original-LCD-Display-Screen-Replacement-iPad-3G-Wifi-16-32-64-GB-/330636656686?pt=UK_Replacement_Parts_Tools&hash=item4cfb7b002e
[17:33] <SpeedEvil> And no, it will need hackery
[17:33] <cuesilence> no controller board etc
[17:33] * M0GHY (~peterholl@ has joined #raspberrypi
[17:33] <SpeedEvil> There is no controller board for modern LCDs
[17:33] <SpeedEvil> they don't worlk like that
[17:35] <cuesilence> I was thinking more like some of the 7" screens from this guy: http://www.ebay.co.uk/sch/njytouch/m.html?_nkw=&_armrs=1&_from=&_ipg=&_trksid=p3686
[17:36] * esotera (~jamie@94-193-222-75.zone7.bethere.co.uk) has joined #raspberrypi
[17:36] <Caver> yeah if we could find something that could be hooked to the DSI connector that would be great
[17:36] <SpeedEvil> If you're designing something, you want something available in volume, with security of supply
[17:36] <SpeedEvil> the ipad dipsly is it
[17:37] <SpeedEvil> I guess that's likely a RGB display
[17:38] <SpeedEvil> it will need a HDMI ->RGB adaptor
[17:38] <Caver> but but ... it's the wrong flavoured fruit!
[17:38] <cuesilence> So is there any project currently working on a HDMI in --> ipad panel?
[17:40] <Caver> http://store.apple.com/us/product/TY600VC/A
[17:42] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) Quit (Ping timeout: 276 seconds)
[17:45] <mchou> seriously, only if RP had mpeg2 license, I wouldn't really care how slow the CPU was
[17:48] <Caver> true
[17:48] <RaTTuS|BIG> $2.5 for each device IIRC
[17:49] <haltdef> ouch
[17:49] <Caver> well I gather you can still code up a program that can use the GPU to help with the decode operations, just can't throw a complete file at it, and say display that
[17:49] <mchou> no mpeg2 means pretty much means RP won't have much of a market in North America
[17:49] <Caver> yay
[17:49] <Caver> more for us
[17:49] <Caver> it's not designed as a player for that
[17:49] <haltdef> not for raw recordings, I can't imagine many people using an htpc for those though
[17:49] <Caver> it's not designed as a media player
[17:50] <haltdef> separate DVR
[17:50] <mchou> yeah well
[17:50] <haltdef> only SD is mpeg2 in sane countries though :P
[17:50] <haltdef> not quite sure how well the ARM would deal with that
[17:50] <Caver> is HD tv mpeg2 in america?
[17:50] <mchou> haltdef: no mpeg2 frontend
[17:50] <haltdef> yeah
[17:51] <mchou> Caver: yes, all of north America
[17:51] * imnichol (~ian@somsen-129-32.winona.edu) Quit (Ping timeout: 252 seconds)
[17:51] <haltdef> they've had HD for so long, I don't think h.264 was around at the time
[17:51] <haltdef> that or decoding hardware was too expensive
[17:51] <mchou> haltdef: it was around, but it was a bit computationall expensive back then
[17:52] <mchou> haltdef: had to get heavy iron for that back in the day
[17:53] <Caver> nods ... well you never know .. someone might find some cunning hack or leak something from broadcom later on - who knows
[17:53] <mchou> pfft
[17:53] <traeak> steal android codec
[17:53] <mchou> not holding my brath
[17:53] <Caver> they do say that necessity is the mother of invention
[17:53] <mchou> breath*
[17:53] <traeak> cell phones have this chip
[17:53] <traeak> so use their codec
[17:53] <mchou> good luck hacking the rom
[17:54] <traeak> i don't care :-p
[17:54] <Caver> I gather the Roxo 2 which uses the same chip doesn't have MPEG2 either
[17:54] <mchou> Caver: yup, true
[17:54] <Caver> but seems to sell
[17:54] * imnichol (~ian@ has joined #raspberrypi
[17:54] * imnichol (~ian@ Quit (Remote host closed the connection)
[17:54] <traeak> what's price on roxio2 ?
[17:55] * imnichol (~ian@ has joined #raspberrypi
[17:55] <mchou> Caver: which iswhy it makes more sense to buy a blu-ray player than roxo
[17:55] <mchou> Caver: that's because pople don't know better :)
[17:55] <traeak> boo-ray
[17:55] <traeak> heh
[17:55] <Caver> $100 I think
[17:56] <traeak> too mcuh of course
[17:56] <mchou> it's less than that
[17:56] <mchou> bluray players now are around $100
[17:56] <Caver> though it does come with a remote control etc
[17:57] <Caver> wifi too
[17:59] <Caver> well I'm hoping that it will manage SD mpeg2 stuff with out too much trouble
[17:59] * cosh (~cosh@gateway/tor-sasl/ryld) Quit (Quit: WeeChat 0.3.6)
[17:59] * prebz (~prebz@c83-254-51-59.bredband.comhem.se) has joined #raspberrypi
[18:00] <mchou> Caver: usb remote controls can be had here for $6
[18:01] <mchou> reasonably nice ones too
[18:02] <Caver> well for what I'm doing I think a old remote and use the HDMI CEC functionality will work fine
[18:02] <mchou> actually just need the usb dongle because any learning remote control has the same IR protocol
[18:02] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) has joined #raspberrypi
[18:02] <mkopack> Hey gang
[18:02] <Caver> (can get the TV to pass though all IR signals back down the HDMI cable to the pi)
[18:02] <mkopack> So, looks like the site got slash dotted again...
[18:03] <Caver> surprise surprise!
[18:03] <Caver> yup
[18:04] <mkopack> lol.. Let's just hope when they FINALLY go on sale, nobody posts to slashdot so the shop doesn't get slash dotted too!
[18:05] <mchou> there will be DOS when RP goes on sale
[18:05] <mchou> no doubt about that :)
[18:06] <mchou> they should just hand out lottery #'s beforehand
[18:06] <mchou> prevent any rampage
[18:07] <cuesilence> I like that quite a bit, but more like random email with a purchase link
[18:07] * diplo (~diplo@cpc1-trow1-0-0-cust13.aztw.cable.virginmedia.com) Quit (Quit: Leaving)
[18:07] <mkopack> Yeah, but then you know somebody will bitch that they got left out, or that the way the "random" emails was sent out wasn't fair, etc.
[18:08] <mkopack> Eh, honestly, I can wait for the 2nd or 3rd prod run...
[18:08] <mkopack> so it won't be the end of the world to me if I don't get in on the first batch
[18:08] <cuesilence> no matter what someone will be left out and will bitch, that's just how it goes
[18:08] * CuriosLion (~stian@117.81-166-155.customer.lyse.net) Quit ()
[18:08] <mchou> yeah, wait for others t o work out the bugz :)
[18:08] <mkopack> Thankfully, even at $35, if they do find issues it's not like I wasted tons of $$$
[18:09] <Caver> yup!
[18:09] <mchou> it aint the money. it's the time :)
[18:09] <mchou> time spent on debugging is painful
[18:10] <cuesilence> isn't that kind of the point of the board, to explore, debug, program, etc
[18:10] * kism3t (~kism3t@client-80-3-134-81.bsh-bng-013.adsl.virginmedia.net) Quit (Quit: Leaving)
[18:10] <mchou> cuesilence: maybe for you :)
[18:10] <mchou> I already have sheevaplug :)
[18:10] <cuesilence> i meant more for the classroom
[18:10] <Caver> mchou, what are you wanting to do with it?
[18:10] <mkopack> different strokes for different folks - I want to use it for robotics work???
[18:10] <mchou> Caver: not much really
[18:10] <mkopack> So I don't want to spend my time debugging the Pi, but rather debugging my robotics control code running on it
[18:10] * smjms (~janne@dyn2-212-50-133-158.psoas.suomi.net) has joined #raspberrypi
[18:11] <Caver> I can see it being very nice for robotics work
[18:11] <mkopack> mchou: Yeah, I have one of those. What a waste of $99 that was for me
[18:11] <mchou> Caver: just would like something with more peripherals than sheevaplug
[18:11] <Caver> especially if we can get a nice touch screen driver working
[18:11] <mkopack> Just not enough support for non-linux gurus
[18:11] <mchou> mkopack: I actually have the dockstar (on sale for $20)
[18:11] <mchou> mkopack: dockstar is way better than sheevaplug
[18:12] <Caver> nods ... well python ought to be reasonably well developed out the box for the GPIO controls
[18:12] <ali1234> no i2c driver yet though
[18:12] <mkopack> I FINALLY figured out how to get a newer version of linux on it (I have the original sheevaplug dev kit model) and then I made the mistake of doing an apt-get update and it filled the damn onboard flash and crashed. At that point I gave up on it
[18:12] <ali1234> kind of important if you want sensors
[18:12] <cuesilence> dockstar $100 US on amazon
[18:12] <feep> DOS, Denial of Sale
[18:12] <mchou> mkopack: dockstar at least doesn't burn up and keeps on running
[18:12] <ali1234> is anyone working on that?
[18:12] * M0GHY (~peterholl@ Quit (Quit: Leaving.)
[18:12] <Caver> nope but it's in the datasheet now, so no reason it can't be done
[18:13] <ali1234> i know
[18:13] * Yf (5c0596fb@gateway/web/freenode/ip. has joined #raspberrypi
[18:13] <mchou> cuesilence: you missed it last year when seagate was clearing them all out
[18:13] <ali1234> i don't want to waste my time doing it if someone else is working on it though
[18:13] <mkopack> I want to get ROS running on the PI
[18:13] <mchou> cuesilence: fire sale
[18:13] <cuesilence> apparently!
[18:13] <zer0her0> i'd second a nice touch screen driver
[18:13] <Caver> yup it would be nice
[18:13] <mkopack> No reason the base ROS code won't work??? but things like point clouds, planning, and such probably won't be doable given the limited RAM
[18:13] <Caver> expecially if I could make it run my RepRap
[18:14] <Caver> there are ROS linux isn't there?
[18:14] <Caver> or good as
[18:14] <mkopack> Yes , but ROS typically is built to run on Ubuntu and x86...
[18:14] <mkopack> So it'll be a recompile from source
[18:15] <Caver> mkopack, you think the ram will be a problem then?
[18:15] <Caver> thats a pity
[18:15] <mkopack> Caver - BTW, since you mentioned wanting to control a RepRap - one of the guys in my robotics club had a board from STMicro that had like a 400Mhz ARM processor on it, about the size of an Arduino Mega, and it was like $17 from digikey...
[18:16] <Caver> yeah I'm not sure it's going to be possible as I think the reprap needs more than the Pi has in gpio lines
[18:16] <mkopack> Caver: Well, like I said, the base messaging infrastructure that ROS uses for passing data between Nodes should be doable no problem. But things like SLAM, point cloud processing, path planning, etc. Those are require a lot of ram and CPU, and we have neither...
[18:16] <cuesilence> could you just emulate x86, I've run windows xp on my android phone that way with qemu
[18:16] <Caver> unless I get clever with daisy chaining stuff
[18:16] <Caver> ah I was meaning real time os's not robotic os
[18:16] <mkopack> So I think for basic interaction with sensors and interfacing to other machines, the Pi running ROS will work.
[18:17] <Caver> yup - sounds like you need something thats fundermentally going to cost more!
[18:17] <mkopack> I'm also going to look into using multiple Pi's together to see if I can do some of the more processor/memory intensive ROS stuff as a cluster
[18:17] <Caver> hmm!
[18:18] <Caver> I can't see that being at all easy
[18:18] * jardiamj (~chatzilla@221-228-212-206-wiban.onlinemac.com) has joined #raspberrypi
[18:18] <Caver> especially as the network latency isn't going to be fast
[18:18] * NIN101 (~NIN@ has joined #raspberrypi
[18:18] <mkopack> Given what I'm currently using for my telepresence robot (an old core duo laptop)'s power budget, I think I could use a cluster of like 10 Pi's no problem and use the same amount of power.. and be way more parallel in operation, and with no moving hard drive to die with all the vibrations of driving over carpet
[18:18] <Caver> in all honesty why not use a basic netbook instead?
[18:18] <Caver> loads more power, still small
[18:19] <Caver> like a EEE901 or something
[18:19] <Caver> SSD ..
[18:19] <ali1234> because of the terrible battery life and weight
[18:19] <mkopack> $$$
[18:19] <mkopack> We'll see???
[18:19] <DaQatz> Connect all the pi's in the bramble via their GPIO's.
[18:19] <Caver> hehe cheap, small, powerful - choose two
[18:19] <Caver> ick ... GPIO's can't be made quick!
[18:20] <ali1234> at least use i2c
[18:20] <DaQatz> Yeah but then you have no external networking, and can use just A's
[18:20] <Caver> i2c is ok for small amounts of data
[18:21] <ali1234> i2c is good for up to 400kbit
[18:21] <ali1234> per second
[18:21] <ali1234> less protocol overhead
[18:21] <Caver> for short distances remember
[18:21] <ali1234> between to pis it should be much faster
[18:21] <DaQatz> Umm think the gpio on pi is actually a bit faster then that.
[18:21] <ShiftPlusOne> model a has ethernet, doesn't it?
[18:21] <DaQatz> Oh wait
[18:21] <DaQatz> Nvm
[18:21] <Caver> ShiftPlusOne, no model A - no ethernet
[18:22] <ali1234> how many instructions does it take to toggle a gpio? how many cycles?
[18:22] * johnthebear (~Johnthebe@13-165-19.client.wireless.msu.edu) has joined #raspberrypi
[18:22] <ali1234> that's the limiting factor
[18:22] <ShiftPlusOne> Caver, hmm.... was that always the case?
[18:22] <Caver> yes
[18:22] <ali1234> you'd just end up bit banging something extremely similar to i2c anyway
[18:22] <DaQatz> ShiftPlusOne, yes
[18:22] <mkopack> Caver: http://search.digikey.com/scripts/DkSearch/dksus.dll?KeywordSearch&FV=fff40028,fff8014d,fffc01f1&k=STM32
[18:22] <ShiftPlusOne> hmm, strange. thanks
[18:22] <mkopack> I believe the board I was talking about is one of the first ones in the list there
[18:22] <ali1234> i don't think the broadcom supports multimaster i2c anyway so you might as well just use the serial uart
[18:22] <ali1234> which will go up to 6mbit iirc from the datasheet
[18:22] <Caver> Model A = 128Mb, no Ethernet - Model B = 256Mb, ethernet built in via the internal usb bus
[18:23] <mkopack> I think it's the second one
[18:23] <Caver> nods
[18:23] <DaQatz> 6mbit is pretty good for that.
[18:23] <Caver> well I'm defo no expert in robotic controls
[18:24] <Caver> though I do fancy trying a linux CNC setup sometime
[18:24] <mjr> presumably in B the ethernet/hub chip is put on the soc's usb controller whereas on A the soc usb is just passed through to the usb jack?
[18:25] <mkopack> Caver: Same here. Buddy of mine was interested in doing CNC and so he got me looking and some of these $100 setups look interesting. Would be nice to be able to make nicely etched circuit boards with one.
[18:25] <DaQatz> Pi should handle CNC operations fine/
[18:25] <mkopack> mjr: I believe that's the case. The Ethernet driver chip provides the 2nd USB port
[18:25] <Caver> mjr, yes ...
[18:26] <Caver> thats it .. I believe the chip has a internal usb hub, which it then attaches a usb<>ethernet device and exports the another 2 usb lines, which are connected to the edge of the board :)
[18:26] <mjr> rather, it provides both of the ports in B :]
[18:26] <traeak> heh
[18:26] <mjr> Caver, yeah, that's what I figured
[18:26] <Caver> http://www.smsc.com/media/Downloads_Public/Data_Briefs/9512db.pdf
[18:27] * Space_Man (~space_man@ has joined #raspberrypi
[18:27] * Draio_ITA (51ae1f5c@gateway/web/freenode/ip. Quit (Quit: Page closed)
[18:27] <mjr> (pretty nifty chip)
[18:27] <Caver> page 3 has useful block diagram
[18:28] <Caver> yeah I think they chose very well
[18:28] <mkopack> Yes and no???
[18:28] <mkopack> But for the price range, yeah, I think they did
[18:28] <mkopack> I think may of us would have preferred to a stronger ARM processor with more RAM??? but you're not going to get that in the price range they targetted.
[18:28] <mkopack> I think may of us would have preferred to a stronger ARM processor with more RAM??? but you're not going to get that in the price range they targeted.
[18:29] <Caver> hehe oh the amount of "but I wanted something with 1Gb/different arm cpu/something".... all of which is well thats nice, but it woudl have bumped the price up
[18:29] <Caver> there's a reason all the other boards cost a lot more
[18:29] <mkopack> Exactly...
[18:29] <mkopack> So like I said, for the target price range, I think they chose well
[18:29] <Caver> and if you want that .... go buy one!
[18:30] * rafal_ (~rafal@ip-78-30-102-40.free.aero2.net.pl) has joined #raspberrypi
[18:30] <drazyl> it's not intended to be all things to all men, but if it fits your need as well as the foundations...
[18:30] <Caver> heh sorry .. get bored with the amount of "I want X, it *must* have Y or I'll sulk" stuff
[18:31] <mkopack> Yeah agreed??? If it's not the right tool, fine, move on??? they're making what THEY want to make for THEIR reasons.
[18:31] <Caver> it doesn't have MPEG2 ... well yes .. but *why* does a educational board need MPEG2? ... mpeg4 is fine and probably more reliant anyway
[18:31] <ali1234> actually it does have mpeg2
[18:31] * datagutt_ is now known as datagutt
[18:31] <Caver> are you sure?
[18:32] * rafal__ (~rafal@ip-78-30-102-52.free.aero2.net.pl) Quit (Ping timeout: 245 seconds)
[18:32] * oenime (~oenime@ Quit (Ping timeout: 252 seconds)
[18:32] <ali1234> the gpu firmware contains references to a list of 4ccs
[18:32] <drazyl> and personally I can't wait
[18:32] <ali1234> including things like 3gp, webm etc
[18:32] <ali1234> whether those will be available on the pi is another question
[18:32] <Caver> ali1234, I thought they'd not been unlocked, as the foundation can't afford the license for them
[18:32] * oenime_ (~oenime@ has joined #raspberrypi
[18:32] <mkopack> I don't believe they've paid for the licensing to support MP2 in the OS though
[18:32] <Caver> correct
[18:33] <ali1234> yes, that's quite possible
[18:33] * pistacik (~pista@ip-213-220-206-109.net.upcbroadband.cz) has joined #raspberrypi
[18:33] <Caver> right .. so for the purposes of this channel it doesn't
[18:33] <cuesilence> what do you guys think: http://www.alibaba.com/product-gs/462382030/10_inch_hdmi_monitor.html
[18:33] <ali1234> where's the channel for bcm2835 reverse engineering?
[18:33] <ShiftPlusOne> Caver, I think many people will be interested in hacking it.
[18:34] <Caver> dev?
[18:34] <Caver> ShiftPlusOne, oh defo
[18:34] <ShiftPlusOne> ali1234, this channel is fine.
[18:34] <Caver> very welcome to talk about hacking it on here
[18:35] <Caver> I just didn't want people getting the wrong idea
[18:35] <mkopack> What sort of price on it cue?
[18:35] <ali1234> are any firmwares officially pulbicly available yet?
[18:35] <ShiftPlusOne> ali1234, yes and no... (just no)
[18:35] * zutesmog1 (~timh@CPE-124-176-144-101.lns5.lon.bigpond.net.au) has joined #raspberrypi
[18:36] <ShiftPlusOne> but there are some rootfs archives floating around
[18:36] <ali1234> i mean start.elf
[18:36] <Caver> are there :)
[18:36] <ShiftPlusOne> none 'official' though
[18:36] * zutesmog (~timh@CPE-124-179-5-14.lns3.ken.bigpond.net.au) Quit (Ping timeout: 245 seconds)
[18:36] <ShiftPlusOne> ah right... that firmware
[18:36] <ali1234> has anyone seen that "VC CPU BUS" diagram on the datasheet
[18:36] <ali1234> memory mapped at 0x00000000, 0x40000000, 0x80000000, 0xc0000000
[18:36] <Caver> which page?
[18:36] <ShiftPlusOne> just what people got preloaded on SD cards I think... haven't seen anything online yet though
[18:37] <ali1234> smells like MIPS
[18:37] <cuesilence> they don't say but I'm guessing based upon others: $20 - $80 by volume with a minimum of 1000, I'm sure at $80x1k, probably $20x100k, something like that
[18:37] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has joined #raspberrypi
[18:37] <ali1234> Caver: page 5 iirc
[18:37] <ShiftPlusOne> cuesilence, it says 100 minimum
[18:38] <ShiftPlusOne> but normally you have to ask to get a price sheet, obviously
[18:38] <cuesilence> Same monitor $100 a unit / min order 10 units: http://www.alibaba.com/product-gs/526123177/10_inch_HDMI_monitor_with_High.html
[18:38] <cuesilence> So the origional one might be $80 with 100 min
[18:38] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has left #raspberrypi
[18:38] <ShiftPlusOne> cuesilence, be careful with alibaba... lots of scams
[18:39] <cuesilence> That's the shame of it
[18:39] <zer0her0> w/ a name like that don't you sort of expect it would have scams? ;)
[18:41] <cuesilence> alibaba use to be quite reputable, it was only a handful of incidents that shined a bad light, they just couldn't be as big or popular or successful as they are if they were bad.
[18:41] * smjms (~janne@dyn2-212-50-133-158.psoas.suomi.net) Quit (Quit: Konversation terminated!)
[18:43] * nils_2 (~nils_2@unaffiliated/nils-2/x-2480262) Quit (Read error: Operation timed out)
[18:43] <ShiftPlusOne> cuesilence, maybe, but I know from first hand experience, so I think it would have to be fairly common.
[18:43] * nils_2 (~nils_2@unaffiliated/nils-2/x-2480262) has joined #raspberrypi
[18:44] <cuesilence> I think the problem is there is no way to verify the individual seller on there, and I doubt alibaba.com itself does
[18:44] <cuesilence> so it's a gamble
[18:44] <ShiftPlusOne> yup
[18:45] * octeris (~Adium@ has joined #raspberrypi
[18:47] * Datum_Errata (Datum_Erra@ has joined #raspberrypi
[18:50] * MuNk (~MuNk@2001:470:1f09:1ec1:210:75ff:fe1a:bdb0) has joined #raspberrypi
[18:51] * djazz (~djazz@78-72-49-85-no186.tbcn.telia.com) has left #raspberrypi
[18:51] <DaQatz> >.>
[18:51] * Hopsy (~kvirc@ has joined #raspberrypi
[18:51] <DaQatz> <.<
[18:52] <DaQatz> Umm spam.
[18:52] <DaQatz>
[18:52] <DaQatz> <Wulfe> ????????????????????????????????????????????????????????????????????????
[18:52] <DaQatz> <Wulfe> ?????????????????????????????????????????????????????????????????????
[18:52] <DaQatz> <Wulfe> ???????????????????????????????????????????????????????????????
[18:52] <DaQatz> <Wulfe> ????????????????????????????????????????????????????????????
[18:52] <DaQatz> >.>
[18:52] <DaQatz> <.<
[18:52] <Caver> *quints*
[18:53] <rm> http://www.aliexpress.com/wholesale/wholesale-hdmi-monitor.html?SortType=total_tranpro3_desc&SortType=y
[18:53] <rm> use aliexpress
[18:53] <rm> no MOQ
[18:53] <rm> and you can see who already ordered and their feedback
[18:53] <cuesilence> ohh no kidding, any kind of protection or ratings would be great
[18:54] <cuesilence> rm: thanks for that
[18:54] <octeris> is it out
[18:54] <octeris> :P
[18:54] <octeris> slap me bot!
[18:55] * Caver slaps you with a raspberry jelly flavoured frying pan
[18:55] <octeris> You know it's actually an especially dumb question now that a release window has been announced.
[18:55] <octeris> Sort of, anyway.
[18:56] <Caver> assuming no more hickups
[18:56] * thrawed (5ec57f89@gateway/web/freenode/ip. has joined #raspberrypi
[18:58] * MuNk (~MuNk@2001:470:1f09:1ec1:210:75ff:fe1a:bdb0) Quit (Ping timeout: 256 seconds)
[18:58] * oenime_ (~oenime@ Quit (Quit: Leaving)
[18:59] <DaQatz> is it out yet?
[18:59] <PiBot> DaQatz: The boards will be completed on February 20th.
[18:59] <DaQatz> Or is the raspberrypi out?
[18:59] <DaQatz> Or is the raspberrypi out yet?
[18:59] <PiBot> DaQatz: The boards will be completed on February 20th.
[18:59] <DaQatz> see the "yet"
[19:00] <Caver> completed in *china*
[19:00] <Caver> don't put the idea into people's heads that their ready to be send out then
[19:00] * cuesilence (ada79aac@gateway/web/freenode/ip. Quit (Quit: Page closed)
[19:01] * MuNk (~MuNk@2001:470:1f09:1ec1:210:75ff:fe1a:bdb0) has joined #raspberrypi
[19:01] <DaQatz> That's why it does not say for sale.
[19:02] <Caver> yeah but doesn't say half way around world
[19:03] <DaQatz> Being that it's not saying that it's for sale that part is not relevant.
[19:05] <ShiftPlusOne> to be fair the only use for that bot feature is Thorn.... so it really doesn't matter what it says
[19:06] <Caver> just trying to make the darn thing vaguely useful *sulks*
[19:07] * imnichol (~ian@ Quit (Ping timeout: 252 seconds)
[19:08] <DaQatz> It is vaguely useful. It's just not obnoxiously verbose with unrequested info.
[19:10] <nrltd> Cant wait to get my hands on raspberry pi ;)
[19:11] <merlin1991> hmmm datasheets
[19:11] <Caver> yup :)
[19:12] * iMatttt (~imatttt@cpc3-farn4-0-0-cust594.6-2.cable.virginmedia.com) has joined #raspberrypi
[19:12] <merlin1991> though one on the gpu would have been far more interesting
[19:13] <merlin1991> since there is the part we need to replace fsckd openmax binary blobs
[19:13] <Caver> yeah or even how to write to the framebuffer
[19:14] * Tachyon` (~tachyon@cpc1-york2-0-0-cust489.7-1.cable.virginmedia.com) Quit (Quit: Rebooting/Upgrading)
[19:15] * merlin1991 especially would like to stream textures into gpu memory from the cpu side without going through nasty apis
[19:15] <pygo> ^
[19:16] <merlin1991> pygo: do you mean you too or what? :D
[19:16] <mkopack> BBL...
[19:16] * DaQatz la la la's along.
[19:16] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) Quit (Quit: mkopack)
[19:16] <DaQatz> nah no one will ever reverse the blob.
[19:17] * Hopsy (~kvirc@ Quit (Read error: Connection reset by peer)
[19:17] <DaQatz> >.>
[19:19] * Hesmon (~fp@p4FE3A165.dip.t-dialin.net) has left #raspberrypi
[19:20] * Tachyon (~tachyon@cpc1-york2-0-0-cust489.7-1.cable.virginmedia.com) has joined #raspberrypi
[19:20] * ShiftPlusOne (~Shift@124-168-104-90.dyn.iinet.net.au) Quit (Ping timeout: 244 seconds)
[19:24] * thrawed (5ec57f89@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[19:30] <Henchman21> its futile
[19:30] <huene> resistance? *g*
[19:30] * rafal__ (~rafal@ip-78-30-108-231.free.aero2.net.pl) has joined #raspberrypi
[19:30] <Henchman21> no resistance is your only hope for future blobs
[19:30] * imnichol (~ian@somsen-129-37.winona.edu) has joined #raspberrypi
[19:31] * rafal_ (~rafal@ip-78-30-102-40.free.aero2.net.pl) Quit (Read error: Connection reset by peer)
[19:32] <Henchman21> probably wouldnt have gnu/linux if stallman didnt write gnu free software
[19:32] <Henchman21> and linus releasing his homebrew kernel
[19:32] * esotera (~jamie@94-193-222-75.zone7.bethere.co.uk) Quit (Remote host closed the connection)
[19:33] <Caver> no but we might have gnu/hurd
[19:34] <Caver> and bsd
[19:34] <relaxed> more like BSD
[19:34] * Davespice_ (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) has joined #raspberrypi
[19:36] * Davespice (~quassel@cpc13-haye17-2-0-cust146.haye.cable.virginmedia.com) Quit (Ping timeout: 276 seconds)
[19:36] <Caver> btw if you loathe it *that* much your very welcome to find a more open board
[19:37] <rm> there is not a more open board
[19:37] <rm> not at this price point
[19:37] <Caver> thats kinda my ponit
[19:37] <Caver> *point
[19:38] <rm> doesn't mean we can't or shouldn't voice our opinions :p
[19:38] * mjr hopes that the Lima driver project for Mali will take sufficiently off for future Mali-containing ARM boards, but they're just it, possible future
[19:39] <mjr> well, techincally Pi's also "possible future", but fairly likely to be not very far off at this point ;)
[19:41] * octeris1 (~Adium@ has joined #raspberrypi
[19:41] * octeris (~Adium@ Quit (Read error: Connection reset by peer)
[19:41] * octeris1 (~Adium@ Quit (Client Quit)
[19:42] * octeris (~Adium@ has joined #raspberrypi
[19:43] <mchou> mjr: look, I know everyone hopes for the best but forward looking statements can be construed as vaporware
[19:43] <Caver> http://en.wikipedia.org/wiki/RTAI interesting - supports Arm7
[19:44] <Caver> they *are* vaporware, but we're allowed to be hopeful
[19:44] <mjr> that's pretty much what I said, no?
[19:44] <Caver> I wouldn't be at all supprised if we get another hickup
[19:45] <Caver> but I think they will come out
[19:46] * octeris (~Adium@ Quit (Client Quit)
[19:46] <rm> I wouldn't be too surprised if the first batch will have some h/w issue
[19:46] <rm> like e.g. lock-up under load
[19:46] <rm> or overheating
[19:47] <rm> probably I'm an 'optimist'
[19:47] <Caver> lol
[19:47] <Caver> or realist ...
[19:47] <mjr> as an OpenMoko early adopter, it wouldn't be that surprising, but I'm willing to throw $35 away
[19:47] * NIN102 (~NIN@ has joined #raspberrypi
[19:47] <ewan> ah, the openmoko
[19:48] <ewan> I recall trying to buy one, but my order got lost in the crush
[19:48] <ewan> dodged a bullet in the end
[19:49] * outsidein (ada79aac@gateway/web/freenode/ip. has joined #raspberrypi
[19:49] * chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) Quit (Read error: Connection reset by peer)
[19:50] * mjr dodged the early Touchbooks which had USB power issues at least
[19:50] * NIN101 (~NIN@ Quit (Ping timeout: 248 seconds)
[19:50] * chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) has joined #raspberrypi
[19:50] <mjr> here the USB output power was pretty directly taken from the input, was it?
[19:51] <mjr> (well, one wouldn't want to power much stuff through the Pi anyway)
[19:51] <tntexplosivesltd> exactly
[19:51] <tntexplosivesltd> one is advised to use a powered USB hub
[19:53] <outsidein> So is there a power supply that can both power the RP and a usb hub?
[19:53] <Tobias|> Undoubtedly, outsidein
[19:53] <ali1234> http://www.maplin.co.uk/bench-power-supply-with-lcd-screen-219129
[19:54] <outsidein> nice, less cords is always better
[19:56] <outsidein> http://gizmodo.com/5637181/lenovos-new-laptop-charger-has-four-usb-ports-built-in
[19:56] <outsidein> something like that
[19:57] <ali1234> what toolchains are people using for kernel builds?
[19:57] <Caver> lol @ $75
[19:58] <Caver> I'm using a $20 sata hard disk thing, thats got a powered hub built in
[19:58] <outsidein> do you have a link?
[19:59] <mjr> outsidein, you might be able to find a USB hub that'll power the Pi while also serving as the hub. But be wary, it would have to provide sufficient power without negotiation. Probably the case for many because it's easy to do the dumb way, though.
[19:59] * tero (~0@aether.info) has joined #raspberrypi
[19:59] <mjr> and of course, it'll have to provide more than the USB spec requires for model B, unless you go for a USB 3 hub.
[20:00] <outsidein> mjr, if you, or anyone, finds a reliable one, I'm guessing most people ordering the RP would want one.
[20:00] <tero> http://www.raspberrypi.org/ is down?
[20:00] <Tobias|> up for me
[20:00] <outsidein> http://www.raspberrypi.org/archives/619
[20:00] <outsidein> new article on it, thats why
[20:01] <Caver> sadly it's not made any more ... but similar to this http://www.win-star.com/eshop/goods.php?id=69
[20:02] <mjr> outsidein, that'll probably have to wait until there's hardware around to experiment with ;] I'll probably try it with my hub...
[20:02] * wirehead (~a@ has joined #raspberrypi
[20:03] <outsidein> I appreciate it and understand where the teacher is coming from, if you only want the RP for teaching a given programming language, do you need to know all the hardware, software, etc?
[20:04] <tero> hm this might be a very stupid question, but if you a learning for example python, why would you want to do that on pi?
[20:04] <tero> and not on a "simple" pc?
[20:04] <tntexplosivesltd> no idea
[20:04] <tntexplosivesltd> cost?
[20:04] <ali1234> this is a very good question
[20:05] <outsidein> yep, cost
[20:05] <ctyler> ukscone: pong
[20:05] <ctyler> 'sup?
[20:05] <Caver> cheap enough you can give to a kid and say take this home and practice
[20:05] <outsidein> exactly
[20:05] <ali1234> if you try to sell this thing to schools and they turn around and say "we already have windows PC, why do we want this?" well how do you answer that?
[20:05] <Caver> also the GPIO ports give you the ability to wire up stuff and do real press a switch and make the program do something type stuff too
[20:06] <Caver> GPIO on a PC is a faff, and if you over volt something you can write the entire PC off, not a $35 board
[20:06] <rm> some load of drivel
[20:07] <rm> with a picture of a cow
[20:07] <ali1234> GPIO is a faff in any full blown operating system including linux
[20:07] <Caver> ?
[20:07] <rm> on the blog
[20:07] <rm> :)
[20:07] <tntexplosivesltd> rm: it's actually prettu well-written
[20:07] <tntexplosivesltd> * pretty
[20:07] * xlq (~apropos@89-168-179-153.dynamic.dsl.as9105.com) has joined #raspberrypi
[20:07] <tntexplosivesltd> ali1234: he wasn't talking about the OS
[20:08] <tntexplosivesltd> he was talking about damaging the hardware
[20:08] <ali1234> what was he talking about then?
[20:08] <Caver> GPIO + python seems not that hard to me
[20:08] <tntexplosivesltd> 08:05 < Caver> GPIO on a PC is a faff, and if you over volt something you can write the entire PC off, not a $35 board
[20:08] <ali1234> why is it a faff then?
[20:08] <tntexplosivesltd> ruining a computer vs an r-pi
[20:08] <ukscone> ctyler: sorry i was napping
[20:09] <ali1234> answer: it's a faff because you need special drivers to access the parallel port from windows to use it as GPIO
[20:09] <Caver> got to plug something into the PC, have the right drivers, and bits
[20:09] <tntexplosivesltd> ali1234: who even uses windows for that?
[20:09] <Caver> erk well if you try the parallel port, you make it very easy for the little darlings to blow up
[20:09] <ukscone> ctyler: will you have physical access tot he alpha board in the next 24 hours or so?
[20:09] <ali1234> it's no different under linux
[20:09] <tntexplosivesltd> ...
[20:09] <tntexplosivesltd> it's much much easier
[20:10] <outsidein> heh
[20:10] <Caver> I'd say 90% of schools are going to be running windows setups on their PC's
[20:10] <tntexplosivesltd> just ask ukscone
[20:10] <tntexplosivesltd> >.>
[20:10] <ctyler> ukscone: yes
[20:10] <ukscone> ctyler: if you do can you give these to a quick spin please http://russelldavis.org/RaspberryPi/devmem2 http://russelldavis.org/RaspberryPi/devmem2.static -- just want to know if they run and if they look ok on the register area of the board
[20:10] <tntexplosivesltd> ukscone: was it you saying about the GPIO fiddler yesterday?
[20:11] <ukscone> tntexplosivesltd: yes
[20:11] <tntexplosivesltd> and using the GPIO under linux is pretty easy?
[20:11] <ukscone> if devmem2 works ok i'll add the raspi's gpio registers and stuff to my bit twiddiler
[20:11] <ctyler> ukscone: that are ... what? Linux binaries?
[20:11] * Caver (~Caver@nat1.nipltd.com) Quit (Quit: home time!)
[20:11] <ukscone> ctyler: yes built on the f14raspi alpha rootfs
[20:11] <ali1234> ukscone: how many lines of code is that?
[20:11] <ctyler> ah. Need any special permissions?
[20:11] <ukscone> just did the static in case that wasn't the ver you had any more
[20:12] <ctyler> k
[20:12] * HansH (5f2468c4@gateway/web/freenode/ip. has joined #raspberrypi
[20:12] <ukscone> ctyler: might (will need root)
[20:12] * rafal_ (~rafal@ip-78-30-109-252.free.aero2.net.pl) has joined #raspberrypi
[20:12] <ukscone> but maybe not if /dev/mem is world writable
[20:12] <ukscone> readable...
[20:12] <ctyler> ukscone: so... you want me to run unknown binaries as root, eh? :-D
[20:13] <ukscone> ali1234: that's about 50 lines but most are blank lines
[20:13] <ukscone> ctyler: well you could build it from source ya self :)
[20:13] <ukscone> just saving you the bother :)
[20:14] <ukscone> ctyler: it's standard devmem2.c -- no mods yet
[20:14] <ukscone> want the source instead?
[20:14] * octeris (~Adium@ has joined #raspberrypi
[20:15] <ukscone> the ver i have that has all the registers and bit masks labels is about 16000 lines for the pxa270 15550 of them area text
[20:15] * octeris (~Adium@ has left #raspberrypi
[20:15] * rafal__ (~rafal@ip-78-30-108-231.free.aero2.net.pl) Quit (Ping timeout: 248 seconds)
[20:17] <xlq> So how does the BCM2835 now compare to the OMAPs in terms of openness?
[20:17] <ali1234> still not as open, but not bad for a first try
[20:18] <ali1234> (the trm for opam3430 is over 3000 pages)
[20:18] <ali1234> *omap
[20:18] <ukscone> ctyler: if you'd prefer the source here is it http://russelldavis.org/RaspberryPi/devmem2.c
[20:19] <ali1234> the omap chips don't have nearly as much "secret" hardware in them either
[20:19] <ctyler> ukscone: np, I was just pulling your leg. Reimaging an SD is pretty trivial, and that's the worst case :-)
[20:19] <ukscone> ctyler: and it's not as if you use it for your online banking anyway :)
[20:21] <ctyler> (well, you could!)
[20:22] <xlq> minimal system = more trustworthy = more secure
[20:22] <xlq> raspberry pi = minimal system
[20:22] <ukscone> ctyler: you could but for that to happen you'd need to have a bank account
[20:22] <rm> except 18 megabytes of proprietary code in RAM at all times, having absolute level access to everything else on the machine
[20:23] <ctyler> xlq: yeah, "This is my banking SD card" :-)
[20:23] <ukscone> i keep my filthy lucre under my matress
[20:23] <ali1234> xlq: well except for the whole other operating system running on the secret second cpu core that has full access to the entire memory, yes, it is quite trustworthy
[20:23] <ctyler> as long as you don't keep your lucre under a filthy matress, you're doing ok
[20:23] <xlq> rm: What, on the GPU?
[20:23] <xlq> Does that have access to everything? I suppose it does, with shared RAM and all.
[20:23] <rm> I think so too
[20:23] <ali1234> xlq: that "GPU" is actually a full CPU + GPU
[20:24] <xlq> Not that you don't get the same problem with modern machines.
[20:24] <ctyler> ali1234: heck, as long as the other code doesn't have control of the NIC it's all good
[20:24] <rm> from the sound of it, ARM core is a guest in that small world
[20:24] <ali1234> pretty much
[20:24] <xlq> You can compromise a NIC and run malware on it. Goodness knows what you could do to a southbridge.
[20:25] <rm> there are IOMMUs and stuff
[20:25] <rm> on a regular computer add-on cards don't DMA from/to where they aren't allowed to
[20:25] <xlq> Not on many computers, and on most computers with an IOMMU, it's disabled.
[20:26] <rm> and curiously, none other than Broadcom has some very very smart NICs, that contain a whole full-featured MIPS core on them :)
[20:26] <xlq> *panic* :P
[20:26] <rm> afaik even Intel ones don't
[20:26] <xlq> http://en.wikipedia.org/wiki/Intel_AMT :(
[20:26] <xlq> "is a deliberate BIOS rootkit for remotely managing and securing PCs out-of-band"
[20:30] * HansH (5f2468c4@gateway/web/freenode/ip. Quit (Quit: Page closed)
[20:30] <ali1234> xlq: in simple terms the bcm2835 architecture is similar to a dual core windows machine running linux inside vmware with a dedicated CPU core
[20:30] <ali1234> except instead of x86 the two cores have different instruction sets, one of which is completely unknown
[20:30] <ali1234> and istead of windows, it's some RTOS you;ve never heard of
[20:32] * imnichol (~ian@somsen-129-37.winona.edu) Quit (Quit: Ex-Chat)
[20:32] <xlq> I still think all the stuff a modern southbridge does - and a graphics card - is equivalent to all the GPU-ey stuff.
[20:32] <ali1234> well you're wrong
[20:32] <xlq> Why?
[20:32] <ali1234> ever seen a GPU that compiles shader code *on the gpu*?
[20:33] <xlq> No, I'm talking about in terms of what can be compromised.
[20:33] <ali1234> the "GPU" can access anything the arm can access and more
[20:34] <xlq> Same with a GPU in a PC.
[20:34] <ali1234> the "GPU" in fact controls what the arm can and cannot access
[20:34] <ali1234> no
[20:35] <xlq> What do you mean, no? With bus mastering (or whatever the PCIe equivalent is called) has full access to main memory.
[20:35] <ali1234> ok, that's true
[20:36] <ali1234> but if nvidia wanted to exploit your computer they would have to put that capability in the silicon
[20:37] <xlq> "In the silicon" = in flash memory.
[20:37] <xlq> Intel AMT shows just how not-hard-wired a modern southbridge is.
[20:37] <ali1234> yes
[20:37] <ctyler> ukscone: well, at first blush it seems to be doing something along the lines of what I'd expect it to be doing
[20:37] <ali1234> intel AMT is specifically designed to do that
[20:38] <ali1234> nvidia cards are not
[20:38] <ali1234> the bcm2835 is
[20:38] <ali1234> do you see what i'm saying?
[20:38] <xlq> Perhaps.
[20:38] <ali1234> if i even had an intel motherboard you can be sure i would disable AMT right away
[20:38] <xlq> Can one not upload firmware to an nVidia or ATI GPU?
[20:38] <ukscone> ctyler: ok great. means mmap() isn't going to be a problem --didn't expect it to be but it was a pain on one arm device i had
[20:38] <xlq> I suspect their proprietary driver might, but who knows :-\
[20:38] <ali1234> you can't disable this stuff in the bcm chip because then the chip doesn't work at all
[20:39] <ukscone> so i might as well start typing stings then
[20:39] <ali1234> xlq: nvidia cards can be reflashed but there is a limit in what the silicon can do still
[20:39] <xlq> ali1234: True. Although if you disable AMT there's probably still going to be a lot of stuff going on on the same chip.
[20:39] * barr5790 (~alan@cpc1-belf7-0-0-cust981.2-1.cable.virginmedia.com) has joined #raspberrypi
[20:39] <ukscone> so i can twiddle bits
[20:39] <ali1234> xlq: for example you can't reflash a nvidia card to run x86 code
[20:40] <xlq> ..
[20:40] <xlq> O_o
[20:40] <xlq> The BCM2835's ARM core is just software?
[20:40] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has joined #raspberrypi
[20:41] <ctyler> xlq: no
[20:41] * zma (~zmac@c2s31-2-83-152-88-41.fbx.proxad.net) has left #raspberrypi
[20:41] <xlq> Oh..jumping ahead there :P
[20:42] <ali1234> xlq: real GPUs use significantly different architecture to CPUs
[20:42] <xlq> I know.
[20:43] <ali1234> for example. a PC video has access to all of memory
[20:43] <ali1234> ok. how are you going to exploit that?
[20:44] <xlq> Anyway, there's something to be said for being able to trust software running on a 3510-transistor CPU.
[20:44] <ali1234> let's say you want to snoop on ethernet card traffic
[20:45] <tntexplosivesltd> o.O
[20:45] <xlq> ali1234: Modify kernel in main memory; insert trojan?
[20:46] <xlq> Hmm, I see how it could be easier to do that with the BCM.
[20:47] <ali1234> that would require you to implement a x86 disassembler + patcher in opencl... good luck with that :)
[20:48] <ali1234> you could perhaps patch the interrupt tables
[20:48] <ali1234> i bet those are fairly consistant across various operating systems
[20:49] <ali1234> you still need to be able to figure out what interrupts are for which hardware though
[20:49] <xlq> Probably, but you've got to know where to point them, and that won't be consistent.
[20:49] <ali1234> you end up needing to run a whole OS on the GPU if you want to do anything really nefarious
[20:49] <ali1234> well guess what, this GPU already runs a whole OS :)
[20:49] * joink (joink@toppe.net) has joined #raspberrypi
[20:49] <xlq> Hmm.
[20:51] <tntexplosivesltd> ali1234: what OS is that?
[20:51] <ali1234> also, i bet that intel thing just patches the bios to create a fake keyboard and mouse
[20:51] <xlq> It's a seekrit!
[20:51] <ali1234> tntexplosivesltd: it runs express logic threadx
[20:51] <ali1234> wwww.rtos.org
[20:52] <ali1234> www.rtos.com sorry
[20:52] <xlq> ali1234: From what I can gather, the southbridge *is* the (emulated PS/2) keyboard and mouse.
[20:52] <tntexplosivesltd> I see
[20:53] <xlq> So the CPU sees a PS/2 keyboard/mouse, which is emulated by the southbridge either using a USB keyboard/mouse or through AMT.
[20:53] <ali1234> xlq: yeah, sounds about right
[20:54] <ali1234> which means a remote user only has the same access as a local user
[20:55] <xlq> Assuming all is well, mwahaha :P
[20:57] * ushioet (~ushioet@ppp-105-168.32-151.iol.it) has joined #raspberrypi
[20:58] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) has joined #raspberrypi
[21:02] * |uen| is now known as uen
[21:04] * Hopsy (~Hopsy@ has joined #raspberrypi
[21:07] * ushioet (~ushioet@ppp-105-168.32-151.iol.it) has left #raspberrypi
[21:10] * kcj (~casey@unaffiliated/kcj1993) has joined #raspberrypi
[21:13] <zgreg> does intel's AMT use an embedded cpu of some sort?
[21:13] * rafal__ (~rafal@ip-78-30-120-44.free.aero2.net.pl) has joined #raspberrypi
[21:13] <ali1234> zgreg: i doubt it
[21:14] <ali1234> perhaps some very very simple one
[21:14] <zgreg> >Intel Active Management Technology (AMT) is a deliberate BIOS rootkit
[21:14] <zgreg> BIOS rootkit? it works on the hardware level!
[21:14] <ali1234> er... no
[21:14] <ali1234> bios is software
[21:14] * datagutt (~datagutt@unaffiliated/datagutt) Quit (Quit: kthxbai)
[21:14] <ali1234> the bios chip is just a rom
[21:14] <zgreg> yes, but AMT is not implemented in software
[21:15] <ali1234> with code in it... that runs on the main cpu
[21:15] * rafal_ (~rafal@ip-78-30-109-252.free.aero2.net.pl) Quit (Ping timeout: 245 seconds)
[21:15] <zgreg> from what I know, AMT is implemented with specialized hardware in the chipset
[21:15] <xlq> [20:12] <zgreg> does intel's AMT use an embedded cpu of some sort?
[21:16] <xlq> AMT is implemented on a CPU in the southbridge.
[21:16] <zgreg> that's why it also works when the PC is switched off
[21:16] <xlq> [20:14] <ali1234> with code in it... that runs on the main cpu
[21:16] <xlq> Not true.
[21:16] <ali1234> what
[21:16] <xlq> As zgreg says, AMT can work with the main CPU off, or infinitely looping, etc.
[21:16] * wirehead (~a@ Quit (Ping timeout: 256 seconds)
[21:16] <zgreg> on the main cpu, it would be pretty useless, too
[21:16] <ali1234> i'm not talking about AMT, i;m talking about the BIOS
[21:16] <xlq> I don't think it even uses SMM.
[21:17] <xlq> Oh.
[21:17] * Datum_Errata (Datum_Erra@ Quit (Remote host closed the connection)
[21:17] <xlq> Ah, misread.
[21:17] <zgreg> since it's quite easily possible to circumvent/disable it
[21:17] <zgreg> ali1234: yes, but apparently you misunderstood
[21:18] <zgreg> I was criticizing that wikipedia calls AMT a "deliberate BIOS rootkit"
[21:18] <ali1234> why?
[21:18] * Datum_Errata (Datum_Erra@ has joined #raspberrypi
[21:19] <zgreg> because it's separate from the BIOS
[21:19] <zgreg> it's a hardware feature
[21:20] <ali1234> it can't do anything without hooking the bios/acpi subsystems
[21:20] <ali1234> if you want to remotely power off a machine
[21:20] <ali1234> if it was hardware only you could just cut off the power
[21:20] <ali1234> if you want to do a graceful shutdown, you need to hook acpi
[21:20] <ali1234> acpi is handled by the bios
[21:20] <zgreg> well, it certainly can interact with ACPI, but that does not change the fact
[21:20] * mrdragons (~lucas@ has joined #raspberrypi
[21:21] <zgreg> not quite sure, but it's probably also possible to simply cut power
[21:21] <ali1234> well you could, but then it would only work with the right drivers installed, which is kind of pointless
[21:21] <zgreg> that's interesting in case the OS locks up
[21:21] <ali1234> the bios can do that too
[21:21] <zgreg> drivers? it does not run on the x86 cpu
[21:21] <ali1234> exactly
[21:21] <ali1234> my
[21:21] <ali1234> point
[21:22] <traeak> !w
[21:22] <PiBot> traeak: in Parker, CO on Tue Feb 7 13:53:00 2012. Temp -6??C. Condition: Light snow, Humidity: 78%, Later -3??C - -13??C. Condition: Mostly Sunny.
[21:22] <ali1234> if it doesn't run on the main CPU then the only way it can interact with high level OS is through a driver or the bios
[21:22] <traeak> snow is done
[21:22] <ali1234> and if it used a driver it would suck, there's already a million ways to do that
[21:22] <ali1234> therefore it must use the bios
[21:23] <traeak> epic win is for rasppi competitor based on an armv7 soc with a mali gpu
[21:23] <outsidein> can't you just run sudo shutdown -h ?
[21:23] <ali1234> well quite
[21:23] <xlq> [20:22] <ali1234> if it doesn't run on the main CPU then the only way it can interact with high level OS is through a driver or the bios
[21:23] <xlq> Though a driver, yes, but not an AMT-specific driver.
[21:23] <zgreg> I still not understand your argument that seems to be that AMT is a BIOS feature, since that's not the point of it
[21:23] <traeak> if the price isn't super high that is
[21:23] <ali1234> xlq: yes, an ACPI driver :)
[21:23] <xlq> PS/2 driver, even Intel graphics driver.
[21:23] <ali1234> zgreg: is a keyboard a bios feature?
[21:23] <zgreg> the point of AMT is that it provides way more control than a software solution
[21:24] <xlq> I think I agree with zgreg, "BIOS rootkit" is misleading.
[21:24] * rafal__ (~rafal@ip-78-30-120-44.free.aero2.net.pl) Quit (Quit: leaving)
[21:25] * KrimZon_2 (~krimzon2@super.duper.reetleet.org) Quit (Ping timeout: 248 seconds)
[21:25] <ali1234> and i still doubt it has anything resembling a CPU
[21:26] * Hopsy (~Hopsy@ Quit (Read error: Connection reset by peer)
[21:27] * KrimZon_2 (~krimzon2@super.duper.reetleet.org) has joined #raspberrypi
[21:27] <outsidein> anyone with a link to the armv7 board with the mali gpu?
[21:27] <traeak> ugh
[21:28] <zgreg> ali1234: why not? it's the easy, provides much flexibility, and it's cheap
[21:28] <zgreg> s/the//
[21:28] <traeak> AMLogic 8726-M
[21:29] <traeak> that's one of the SOCs
[21:29] <outsidein> ok, looks like the cotton candy board has it: http://en.wikipedia.org/wiki/Cotton_Candy_%28single-board_computer%29
[21:29] <ali1234> it also adds complexity, points of failure, and vulnerabilities
[21:29] <ali1234> and uses more power
[21:29] <zgreg> as if a 100mW CPU core matters in a desktop chipset
[21:29] <outsidein> but that's $200
[21:29] <traeak> allwinner a10 also has a mali400
[21:29] <ali1234> 100mW is LOADS man
[21:29] <zgreg> but right, this can also be risky
[21:30] * Faperdaper (~Faperdape@dhcp-077-250-027-109.chello.nl) Quit (Quit: Quit)
[21:30] <zgreg> ali1234: on a 10W part? please...
[21:30] <ali1234> they could use a microcontroller and it would use like 10mW
[21:30] <ali1234> but even that is more than this function needs
[21:30] <mkopack> Anyone try doing one of the Emulated setups to emulate the Arm processor and run Linux on it?
[21:30] * wirehead (~a@d154-5-144-145.bchsia.telus.net) has joined #raspberrypi
[21:30] <ali1234> mkopack: i've done it before
[21:30] <zgreg> sure, 100mW is just a guess, could definitely be a lot less.
[21:30] <traeak> youch 199usd for cotton candy
[21:30] <mkopack> I'm reading this thread http://www.raspberrypi.org/forum/general-discussion/emulator-vs-the-real-thing
[21:30] <ali1234> mkopack: qemu?
[21:31] <zgreg> outsidein: there is none, if you mean the "board" from rhombus tech
[21:31] <mkopack> And was wondering if that's something I can do on my Mac...
[21:31] <zgreg> outsidein: it's vaporware
[21:31] <ali1234> mkopack: yes, just rebuild qemu
[21:31] <traeak> http://rhombus-tech.net/allwinner_a10/
[21:31] <traeak> yes it's from rhombust
[21:31] <traeak> but it's a good sheet on that soc
[21:32] <outsidein> zgreg: been reading about that for a while, it does seem unlikely it will see the light of day
[21:32] <ali1234> mkopack: http://www.kberg.ch/qemu/
[21:32] <zgreg> the rhombus tech thingie is made by some freetard nerd with no clue/experience/money and extremely unlikely to ever see the light of the day
[21:33] <mkopack> Eh, ok, looks like i's more hassle than it's worth. I'll just wait for actual hardware
[21:33] <zgreg> and the 15$ number is pure hyperbole
[21:33] <zgreg> that's the price of the BOM for mass production (>= 100k)
[21:33] <outsidein> along the lines of "winning the lotery"
[21:33] <zgreg> this is not and end-user price AT ALL
[21:33] <zgreg> and 100k units per run is unlikely to say the least
[21:34] <mjr> Yeah the rhombus tech device could be quite nice if it exists at some point.
[21:34] <traeak> or someone uses that SOC for a stick computer
[21:34] <outsidein> if $15 were the price, I'd buy $75 worth
[21:34] <traeak> i don't care about rhobust
[21:34] <traeak> rhombust i mean
[21:35] * mrdragons (~lucas@ Quit (Ping timeout: 252 seconds)
[21:35] <zgreg> yeah, but the rhombus tech guy still has the audacity to compare his non-existing vaporware to raspberry pi and concludes that it's cheaper
[21:35] <mjr> vapor is pretty cheap
[21:35] <traeak> i need to go look on the forums again for why the rpi folks didn't release the gumstick version
[21:35] <outsidein> based on what the overo?
[21:36] <traeak> rhombust are stupid, but both the amlogic and the allwinners are there today just waiting to be stuck on a pcb
[21:36] <ukscone> traeak: cost and room for connectors like gpio
[21:36] <traeak> ukscone: a bunch of the rpi users don't care about gpio
[21:36] <traeak> ukscone: and as i mentioned yesterday if the original were released anyways there owuld already be a software ecosystem for the rpi
[21:36] <ali1234> there already is
[21:37] <mjr> Allwinner does seem like a winner. Presumably it will end up in some decently priced board _at some point_. Until then...
[21:37] <zgreg> traeak: I'm wary of these chinese chip makers
[21:37] <ukscone> traeak: cost as well. smaller is so much more expensive
[21:37] <traeak> ali1234: if the rpi were released in gumstick form 6 months ago there'd be TONS more software today
[21:38] <traeak> ukscone: could be true, yeah, but that gumstick had way fewer connectors, jsut the hdmi, usb, sd and power
[21:38] <ali1234> traeak: no there wouldn't. all free software that anyone cares about was ported to arm years ago
[21:38] <ali1234> that's a hell of a lot of software btw
[21:38] <traeak> xbmc ?
[21:38] <ali1234> yes xmbc has already been ported
[21:39] <ukscone> traeak: and the gunstick costs how much?
[21:39] <ukscone> gumstick even
[21:39] <traeak> ukscone: i wish i knew, the rpi guys killt it (where's the post for why they did that?)
[21:39] <ali1234> Qt and Gtk both already run fine on arm, so that's basically everything
[21:40] * mrdragons (~lucas@ has joined #raspberrypi
[21:40] <traeak> heh, they need to move the comments away from the faq page, it kills load time
[21:41] * Edek (~Edek@5357E55B.cm-6-8d.dynamic.ziggo.nl) Quit (Read error: Connection reset by peer)
[21:42] <zgreg> rpi doesn't support opengl, though, so there's quite a bit of software that might benefit from a port to support opengl es
[21:42] <ali1234> Qt supports ES
[21:42] <ali1234> i suspect cairo does too by now
[21:42] <ali1234> which means gtk does
[21:42] <zgreg> yes, but a lot of software uses opengl directly
[21:42] <ali1234> actually very little software does that
[21:42] <zgreg> e.g. games
[21:42] <ali1234> games for linux lolololol
[21:43] <zgreg> but there's quite a bit more (vice comes to my mind)
[21:43] <mkopack> Ist' that an oxymoron?
[21:43] <ali1234> vice the c64 emulator?
[21:43] <ali1234> i guess it could have an opengl output plugin
[21:43] * mrdragon1 (~lucas@ has joined #raspberrypi
[21:43] <ali1234> but seriously... it was already ported to n900 which uses es
[21:44] <ali1234> as were all those other emulators
[21:44] <zear> vice has a software sdl renderer, i guess n900 uses that
[21:44] <ali1234> maybe. is that a problem?
[21:44] <zear> not at all
[21:44] <zear> sorry for jumping into your discussion guys ;)
[21:45] <zgreg> vice can use opengl for scaling, etc.
[21:45] <mjr> I suppose it could benefit from OpenGL scaling
[21:45] <mjr> echo...
[21:45] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) has left #raspberrypi
[21:45] * jardiamj (~chatzilla@221-228-212-206-wiban.onlinemac.com) Quit (Quit: ChatZilla 0.9.88 [Firefox 10.0/20120129021758])
[21:45] <mrdragons> We have sdl on this right?
[21:45] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) has joined #raspberrypi
[21:45] <zear> zgreg, yes, but it doesn't essencially mean that vice uses it
[21:45] <zear> on n900
[21:45] <mjr> mrdragons, I don't know but I'm going to say "yes" anyway
[21:45] <ali1234> sdl can work on anything
[21:45] <mjr> there's really no way we wouldn't ;P
[21:45] <zgreg> well, at the moment a much more serious problem than opengl (es) support is X support
[21:46] <traeak> mrdragon yes sdl, but unaccelerated
[21:46] <mrdragons> I'm just making sure. :P
[21:46] <ali1234> zgreg: X support? it has X support
[21:46] <zgreg> and it's bad -- onldy fbdev, so it's super slow. have fun watching the screen redraw!
[21:46] <traeak> don't run anything fancy like opaque move/resize on the rpi and it shoudl be okay as frame buffer
[21:46] <zear> there's not many games that require X
[21:46] <ali1234> there are a million youtube videos of people running open office on it
[21:46] * Yf (5c0596fb@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[21:46] <traeak> wireframe move and resize helps a lot
[21:46] <zgreg> ali1234: yes, but performance is extremely bad
[21:47] <traeak> but scrolling is ugly i'm sure
[21:47] <ali1234> zgreg: and that will not be fixed unless broadcom releases gpu documentation, which will never happen
[21:47] <traeak> mali400!!
[21:47] <zgreg> it's completely unaccelerated, everything is done completely in software
[21:47] <traeak> unless the mali400 driver eternally sucks
[21:47] <mrdragons> Oh...
[21:47] <zgreg> and there's no egl/opengl es integration, so you can't run opengl es applications in a window
[21:47] <zgreg> (not cleanly at least)
[21:47] <mrdragons> Everything done in software on this isn't good
[21:47] <ali1234> zgreg: again, won't be fixed without documentation
[21:48] <traeak> youch, no ability to do gles contexts ?
[21:48] <zgreg> ali1234: it's possible to use opengl es for X acceleration
[21:48] <traeak> multiple contexts i mean
[21:48] <ali1234> traeak: i don't think that's really true
[21:48] <zgreg> traeak: that should be possible I assume, but it'll simply draw over your X screen
[21:48] <traeak> ahh
[21:48] <ali1234> i think it actually draws into the framebuffer
[21:49] <zgreg> ali1234: there's a project by intel just for this, but it's not mature at all and does not work on the pi (yet?)
[21:49] <ali1234> (nor anything else lol)
[21:49] * R` (~RHA@ip7.j-k.kund.riksnet.nu) has joined #raspberrypi
[21:49] <traeak> hmm...back to the "if we'd already had this 6 months ago even without all the gpio pins...
[21:49] <mjr> Well, that is actually an interesting tidbit about the X situation. Not killer, though. We all knew already the GPU is restricted. :(
[21:49] <zgreg> well, I jist think X accel is VERY important but there's no progress on this front at all
[21:49] <zgreg> it's a bit sad
[21:50] <traeak> past is past
[21:50] <traeak> zgreg: when the rpi is no longer vapor i'm sure people will step up and provide x with opengles backend
[21:50] <zgreg> hopefully
[21:50] <ali1234> there is absolutely no reason why people need hardware to do that
[21:51] <zgreg> I just wonder what the fuck the people with beta boards are doing
[21:51] <mrdragons> Yeah, it is a bit early to speculate where the project will end up
[21:51] <mkopack> yeah, exacty. It's hard to really make much progress on anything beyond the basics without having hardware to try things on
[21:51] * TSL (5b7ee6f9@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[21:51] <ali1234> gl|es is a standard
[21:51] <ali1234> there are loads of boards with it
[21:51] <mkopack> You KNOW there's going to be somebody who will reverse engineer stuff for the GPU - decompiling the blob, etc.
[21:51] <ali1234> anyone who actually cared about making an X backend would have done it by now
[21:51] <traeak> hopefully they'll get something useful out of the decompilation
[21:51] <mrdragons> zgreg: the user that goes by MagicalTux has one, and he says they've been working with devs to work out a couple kinks, and there's that one guy on reddit
[21:52] <mkopack> right now we're in a lot of limbo waiting
[21:52] <mrdragons> there's 3 boards accounted for
[21:52] <traeak> perhaps those working on the rpi already decided that it's not worth using it for a dumb terminal
[21:52] <ali1234> magicaltux has a raspberry pi? the guy who runs mtgox?
[21:52] <mrdragons> That I know of
[21:52] <mjr> ali1234, presumably the more devices out there the more motivation around, so rpi might well make a difference. Or not, we'll see.
[21:52] <zgreg> ali1234: that's my point -- there are a couple of beta boards in circulation with developers, and X accel is very important. so there really SHOULD be someone working on this, but apparently that's not the case.
[21:52] * outsidein (ada79aac@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[21:52] <ali1234> zgreg: ad there's also thousands of beagle boards and n900 and etc etc etc
[21:52] <mkopack> And think of it this way - with these production delays, it's given the guys with the beta boards more time to get support software built and tested and working.
[21:53] <mrdragons> ali1234: I have no idea, possibly. You'll have to ask him next time he stops by.
[21:53] <ali1234> mrdragons: on irc?
[21:53] <mrdragons> Yep
[21:53] <ali1234> well assuming you spelled the nick right then it's him lol
[21:53] <ali1234> and assuming i remember the correct spelling
[21:53] <traeak> kernel patches
[21:54] <traeak> media player work
[21:54] <ali1234> i wonder if he's mining bitcoins on it :)
[21:54] <mjr> the closed gpu is broadcom's ploy to mine bitcoins. There's a hidden client there that crunches away with spare cycles!
[21:55] <mjr> with your electricity!
[21:55] <mrdragons> :o
[21:55] <mrdragons> I plan to use one to mine bitcoins actually.
[21:56] <mjr> are you deranged? just asking.
[21:56] <ali1234> there's no opencl support is there?
[21:56] <zgreg> how very useful, it's likey 1/100 of the speed of a substandard PC
[21:56] <zgreg> ali1234: no
[21:56] <ali1234> pity
[21:57] <mrdragons> mjr: Quite possibly. :P
[21:57] <ali1234> the GPU would probably be quite power efficient
[21:57] <zgreg> also, let me tell you that opengl es is nearly useless for gpgpu applications
[21:57] <mjr> write shader code for it! *does not know if possible*
[21:57] <mkopack> ali: No, I've asked ,but they say that's not going to happen??? they MIGHT try to expose some sort of GP-GPU api of some sort, but it's going to be something proprietary, and not high on their To-Do list.
[21:57] <zgreg> for the few people that think about this
[21:57] <ali1234> zgreg: of course it is, it's totally different :/
[21:57] <mkopack> (although I feel that's a HUGE mistake given the power of the GPU and lack of it in the CPU, unleashing that GPU power would help a LOT.)
[21:58] <zgreg> right, but later opengl versions can be abused for some gpgpu applications pretty nicely
[21:58] <zgreg> opengl es, no
[21:58] <mjr> yeah it's really very much like the CPU is just a helper processor there...
[21:58] <DaMummy|PND> is it out yet
[21:58] <PiBot> DaMummy|PND: The boards will be completed on February 20th.
[21:58] <ali1234> later opengl versions are almost identical to ES
[21:58] <DaMummy|PND> no wai!
[21:58] <ali1234> they are trying to converge
[21:58] <DaMummy|PND> thx PiBot
[21:58] <zgreg> ali1234: compatible, but not identical at all
[21:59] <zgreg> opengl es of course skips a whole lot of opengl features
[21:59] <ali1234> most of the fixed function pipeline, what ES lacks, was done on CPU anyway
[21:59] <zgreg> yeah, but GLSL is also stripped down a whole lot
[21:59] * mrdragon1 (~lucas@ has left #raspberrypi
[22:00] <zgreg> and in opengl es 2.0, it's based on the ancient GLSL 1.20
[22:00] <traeak> other than for utilizing the gpu for image processing for robotics, why would the embedded folks using the GPIO pins give one rip about the GPU on the rpi ?
[22:00] <zgreg> regarding bitcoin mining, for example, it is completely useless (no integer support)
[22:01] <traeak> and because of the binary blob you can't do image processing on the gpu anyways
[22:01] * atts (~asheehan@static-99-136-171-68.axsne.net) has left #raspberrypi
[22:02] <traeak> so for the guys interested inthe gpio pins the fast gpu is pretty useless? please tell me how i might be wrong here
[22:02] <ali1234> traeak: they wouldn't. and nobody with any sense will be using those pins as GPIO, they'll be using i2c and spi
[22:02] * rodrigo_golive (quassel@nat/indt/x-ipfirbytmrfecjin) Quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
[22:02] <traeak> okay, i2c and spi then, so why would they care about the rpi having really fast opengles ?
[22:02] <ali1234> they wouldn't
[22:02] <ali1234> other than computer vision as you said
[22:02] <traeak> which they can't because the GPU is too closed to do computer vision processing
[22:03] <mrdragons> Yay for closed source
[22:03] <ali1234> well not quite
[22:03] <traeak> anyways, just a small criticism of the rpi
[22:03] <ali1234> i bet you could still do some damage in just the framebuffer gl|es context
[22:03] <ali1234> i hear it writes into the shared framebffer so you can read back the result
[22:03] <mrdragons> I don't think many people would be able to understand the internals of that GPU though, it's brutally complicated from what I've heard
[22:04] <ali1234> not 100% sure on that
[22:04] <mkopack> traek: You actually don't do much CV on the GPU.
[22:04] <zgreg> ali1234: of course you can do that, (you can also simply use a pbuffer and framebuffer objects), but it's still quite limited
[22:04] <ali1234> sure you do
[22:04] <mkopack> I'm sure you CAN do some of the algorithms using a GPU normally, but most of them are done on CPUs
[22:05] <zgreg> GLSL ES simply is not very useful for most signal processing applications
[22:05] <ali1234> GPUs were made for running convolution kernels
[22:05] <traeak> opencl perhaps
[22:05] <zgreg> ali1234: no
[22:05] <mkopack> Sure but that doesn't mean you HAVE to do them on GPU
[22:05] <zgreg> in fact, convolution is quite inefficient with GLSL
[22:05] <mkopack> Most classes teach you how to do them on CPU
[22:05] <ali1234> sure. it's just a hell of a lot faster
[22:05] <zgreg> it can be made much, much faster with a proper gpgpu environment
[22:06] <traeak> gpus have been extended to be able to handle some convolution kernels hehe
[22:06] <ali1234> convolution is ideal for SIMD architectures
[22:06] <traeak> i've seen to many examples of peopel trying to do geometry on gpus and epically faililng
[22:06] <ali1234> which is what GPUs are
[22:07] <zgreg> ali1234: yes, but it's a specially optimized and restricted flavor of SIMD
[22:07] <ali1234> it is in fixed function pipeline gpus
[22:07] <zgreg> so, no, this generalizazion does not work, sorry
[22:07] <ali1234> but nobody makes fixed function gpus any more
[22:07] <zgreg> no, I am talking about the programmable pipeline (which still includes many fixed parts, FYI)
[22:08] * Datum_Errata (Datum_Erra@ Quit (Ping timeout: 255 seconds)
[22:08] <mkopack> The biggest issues with using GPUs for processing is their memory - usually with a lot of the AMD Stream cores or the NVidia cores, they each have VERY little RAM per core, and access out to the main board memory is slow and restricted. So you have to be able to break things up into really small subsets of data to allow each to process
[22:09] <zgreg> some newer (desktop) GPUs provide more flexibility, but that is not exposed by opengl/glsl
[22:09] <zgreg> for example, there's no shared memory support in glsl
[22:09] * Tachyon is now known as Tachyon`
[22:09] <ali1234> this is all fine
[22:09] <zgreg> mkopack: yes, and now imagine that you cannot access the local memory at all :D
[22:09] <zgreg> mkopack: which is the case in GLSL
[22:09] <mkopack> yeah. exactly...
[22:10] <ali1234> so no texture support then?
[22:10] <ali1234> seems unlikely
[22:10] <zgreg> ali1234: I seriously doubt you've done any GPGPU programming
[22:10] <mkopack> Plus to REALLY make effective use out of them ,you need highly parallelizeable data and algorithms, which many things do NOT lend themselves to
[22:10] <zgreg> texture loads are supported, but they are not a panacea
[22:10] <ali1234> zgreg: and i seriously doubt you've done any CV work
[22:11] <ali1234> if you think running an edge detect filter on an arm cpu will be faster than doing it in a shader
[22:11] <traeak> the fun part here is that sucking more gpu like stuff into the general cpu helps solve lots of these problems
[22:12] <traeak> AVX, etc
[22:12] <mkopack> ali: I don't think anyone is saying it's going to be faster on an ARM CPU than in a shader - we're just questioning whether we're going to have the API capability to do it on the GPU at all
[22:12] <traeak> mkopack: don't get your hopes up
[22:12] <mkopack> YEah,. I'm not
[22:12] <mrdragons> Lol at new blog post
[22:12] <zgreg> ali1234: doing it in a shader + uploading data to the gpu + shader launch overhead + downloading the data
[22:13] <traeak> mkopack: why i think rpi could get hurt badly if someone releases a tiny computer based on the allwinner a10 or the amlogic soc's
[22:13] <ali1234> zgreg: shaders are only compiled once and the data ends up in the shared memory framebuffer
[22:13] <traeak> considering those socs are probably only 1-3USD more per unit *IF* you can get them in low volume (that's the critical part)
[22:13] * Da|Mummy (~Veki@cpe-76-190-200-98.neo.res.rr.com) has joined #raspberrypi
[22:14] <zgreg> ali1234: there's still an overhead inherent to executing the shader. I know it's not compiled over and over again.
[22:14] <zgreg> ali1234: and framebuffer access might be slow (we don't know yet)
[22:15] <traeak> also with gpgpu programming there's the in herent problem of being harder to maintain, requiring more tools to copmile shaders, etc
[22:15] <traeak> and there's the problem of gpus having unpredictable performance from generation to generation, and unpredictable drivers and stability....all that stuff has been flushed out ages ago in CPUs
[22:15] <ali1234> the driver includes the compiler
[22:15] <zgreg> anyway, gpugpu on the pi with the currently available tools is basically useless, except for very simple things (but you can do these on the CPU as well)
[22:15] <mrdragons> "statistically, every PowerPoint I???ve have ever seen looks like it was made by a five month-old chimp by flinging its own faeces at the monitor."
[22:15] <ali1234> it's part of the es spec
[22:16] * ShiftPlusOne (~Shift@124-170-21-63.dyn.iinet.net.au) has joined #raspberrypi
[22:16] * Davespice_ is now known as Davespice
[22:17] <ShiftPlusOne> hello
[22:17] <traeak> ali1234: just my typical arguemnt against gpgpu programming in general...i wouldn't do it unless there's lots of money behind it and it would target one specific architecture for a specific customer
[22:17] <mkopack> ali1234: Yeah, don't remind me. I couldn't wait for OpenCL to come out with OSx Snow Leopard for my 2009 Mac Pro. Then when I tried a couple of the example OpenCL programs, I came to realize that my 4 core (8thread) Nehalem CPU could execute the code multiple times faster than the Radeon 4870 I got with the machine!
[22:18] <ShiftPlusOne> is it just me or is the last raspberrypi.org post a little misguided?
[22:18] <mkopack> Shift: How so?
[22:18] <traeak> mkopack: exactly! when i look at benchmarks generated online they tend to be guys running gt560's ruinning then with core2 duos and bragging about how their 560 trashes their cpu
[22:18] <ali1234> mkopack: now compare power usage of that CPU and GPU :)
[22:19] <traeak> gt560 ~ 400W
[22:19] <traeak> core i7 max 135W
[22:19] <mkopack> ali: Yeah, the CPU uses a LOT less!
[22:19] <zgreg> mkopack, traeak - for some things gpus are awesome, though. but they're no magical bullet at all
[22:19] <traeak> that's pretty easy
[22:19] <zgreg> and hard to develop for
[22:19] <traeak> zgreg: they are a bigger PITA than people let on is all
[22:19] <zgreg> opencl might make code portable in the sense that it runs everywhere
[22:19] <mkopack> Of course, that was also when OpenCL first came out, and the NVidia cards were WAY better at it than the Radeons...
[22:19] <zgreg> but it won't perform well everywhere, that still requires very specific optimizations
[22:20] <mkopack> I have a newer 6870 in my Mac Pro now, but it's a hacked one, so I don't think I can do OpenCL on it
[22:20] <traeak> and running a radeon 4xxx compared with a 5xxx series the performance varies wildly
[22:20] <ShiftPlusOne> mkopack, I don't know... just made me cringe. I don't want to be negative about, was just wondering if anyone else found it a bit odd.
[22:20] <zgreg> e.g. look at the global memory coalescing requirements of amd vs nvidia
[22:20] <traeak> if you are doing it for high performance
[22:20] <mkopack> Yeah, you have to design your code based on the properties of the GPU in question to real get the most out of it
[22:20] <ali1234> since when is a radeon 4870 a gt 560?
[22:20] <traeak> same thing gf4xx vs gf5xx
[22:20] <mkopack> Shift: what are you talking about? The post by the teacher? Or the "two things" post?
[22:20] <ali1234> radeon 4870 uses nowhere near 400W
[22:21] <mkopack> ali :It's at least 200??? Required 2 aux power jacks
[22:21] <ShiftPlusOne> mkopack, the 'Clive' one.
[22:21] <zgreg> well, radeon 4xxx is no good for gpgpu
[22:21] <zgreg> it's not really support by amd, because the hardware cannot properly support opencl
[22:21] <zgreg> *supported
[22:21] <mkopack> Shift: Ok, what's wrong with it? It's just something written by a teacher??? Except I cringe when he says he teaches computing yet seems THAT clueless about computing
[22:22] <mkopack> zgreg: true???
[22:22] <mkopack> It was like the very first gen that could
[22:22] <mkopack> and only barely
[22:22] <zgreg> basically radeon 4xxx does not support the __local memory model of opencl
[22:22] <zgreg> and it uses __global memory as fallback
[22:22] <zgreg> which is SLOW
[22:22] <mkopack> I should give the example code a try now on my 6870 to see if it'll run...
[22:22] <traeak> and the radeon 6xxx will make the 5xxxx look like a POS as well
[22:22] <traeak> :-p
[22:22] <traeak> etc etc
[22:22] <ShiftPlusOne> mkopack, that's exactly it.
[22:22] <mkopack> Not sure it will since it's a hacked PC card, not an actual Mac EFI card
[22:22] <traeak> no biggie
[22:23] <traeak> !w
[22:23] <mkopack> Shift: But see, that's the thing??? In england (and many parts of the US) they've taught ICT (basically, how to use MS Office and Windows basics) as "Computing" for the last 20 years.
[22:23] <mkopack> So they were able to get away with "teachers" like that...
[22:24] * PiBot (~Raspberry@c-24-147-35-37.hsd1.ma.comcast.net) Quit (Ping timeout: 244 seconds)
[22:24] <mkopack> I tell ya, I <3 my Mac Pro, but god I HATE how it's nearly impossible to get a video card upgrade, and what little is out there for it costs $500 and is already outdated.
[22:26] * Hopsy (~Hopsy@ has joined #raspberrypi
[22:27] <mkopack> Anyhow, in the schools I attended in NY State, we had both a "Business" area, where you could focus in that and do the MS Office type stuff, and take accounting and such. Or there was the Technology track, where you did woodworking + drafting + CAD, and small engines, and there was the Science track where you could take BASIC and Pascal and computer technology/electronics
[22:27] <mkopack> It sounds like the UK is starting to realize they need to do something similar
[22:28] <ali1234> yeah
[22:28] <ali1234> but the problem is there are literally no teachers who have a clue at that level
[22:28] <Hopsy> is it out yet?
[22:29] <Tachyon`> heh
[22:29] <Tachyon`> if you think they had a clue in the late 80s...
[22:29] <ali1234> if they do they are most likely teaching a "real" subject like science or maths
[22:29] <Tachyon`> I had a computing teacher in 1987 or so who taught programming but couldn't write a line of code, he just copied things out of books and they worked or they didn't
[22:30] <mrdragons> wtf
[22:30] <ShiftPlusOne> jesus
[22:30] <ali1234> pretty much all my science teachers knew more about computers than the one IT teacher at my school
[22:30] <Tachyon`> yeah, I'd like ask him why something wasn't working
[22:30] <Tachyon`> and he'd tell me to type it in again from scratch
[22:30] <ali1234> granted that was 20 years ago
[22:30] <ShiftPlusOne> this is depressing. I am going to go sleep.
[22:30] <ShiftPlusOne> 'night
[22:30] * UnderSampled|tab (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) has joined #raspberrypi
[22:30] <mrdragons> The one competant IT teacher at my school just left. Knew how to code C and everything. :\
[22:31] <ali1234> actually one of my science teachers quit to become a software developer :)
[22:31] <mrdragons> Understood network security, knew linux, rootkits, etc.
[22:31] <Tachyon`> we had a good teacher at my school, taught maths, english, programmign and electronics
[22:31] <Tachyon`> but he was made redundant
[22:31] <mrdragons> ali: For reals? That's pretty cool, who's he develop for?
[22:31] <Tachyon`> because he didn't get on with the new head
[22:31] * ShiftPlusOne (~Shift@124-170-21-63.dyn.iinet.net.au) Quit (Quit: Leaving)
[22:31] * Tachyon` sighs
[22:31] <ali1234> mrdragons: no idea
[22:32] <ali1234> it was 20 years ago
[22:32] <mrdragons> Oh yeah, lol
[22:33] <ali1234> i think he founded his own software company actually
[22:33] * pistacik (~pista@ip-213-220-206-109.net.upcbroadband.cz) Quit (Ping timeout: 245 seconds)
[22:33] <ali1234> so he was probably writing accounts software you've never heard of or something like that
[22:33] <mrdragons> no shame in that
[22:33] <ali1234> not at all
[22:33] <ali1234> and i bet it paid better than teaching
[22:38] * Jaseman (5f93f112@gateway/web/freenode/ip. has joined #raspberrypi
[22:40] <Da|Mummy> is there something like this available, but both ends being hdmi male? http://images2.monoprice.com/productmediumimages/37331.jpg
[22:40] <Da|Mummy> would work nice for rpi
[22:40] * kism3t (~kism3t@client-80-3-134-81.bsh-bng-013.adsl.virginmedia.net) has joined #raspberrypi
[22:42] * kism3t (~kism3t@client-80-3-134-81.bsh-bng-013.adsl.virginmedia.net) Quit (Client Quit)
[22:42] <mjr> meaning to let the pi board hang from the monitor's hdmi connector through that?
[22:42] * Jaseman (5f93f112@gateway/web/freenode/ip. Quit (Ping timeout: 245 seconds)
[22:43] <Da|Mummy> right
[22:43] <mjr> doesn't sound like the most stable arrangement, considering the other cables
[22:43] * LiENUS (~yes@2001:470:bbb3:12:88de:e2e9:5292:b3dc) has joined #raspberrypi
[22:45] * Hopsy (~Hopsy@ Quit (Read error: Connection reset by peer)
[22:45] * Hopsy (~Hopsy@ has joined #raspberrypi
[22:48] <mkopack> Yeah, to think that my 9th grade math teacher gave me a D on the stupid programming projects she gave us (using what at the time was already a 10 yr old C= Pet computer!) and here I am as a Sr. Software Engineer at a major robotics research lab??? Would LOVe to rub that in her face...
[22:49] * outsidein (ada79aac@gateway/web/freenode/ip. has joined #raspberrypi
[22:52] * MystX (~MystX@ec2-107-21-100-215.compute-1.amazonaws.com) has joined #raspberrypi
[22:52] * outsidein (ada79aac@gateway/web/freenode/ip. Quit (Client Quit)
[22:53] * mjr (mjr@nblzone-241-33.nblnetworks.fi) Quit (Ping timeout: 245 seconds)
[22:56] <mrdragons> http://www.opensource.apple.com/source/Libc/Libc-262/stdio/putchar.c
[22:57] <mrdragons> Lol, 50 lines of license for 5 lines of code
[22:58] * mjr (mjr@nblzone-241-33.nblnetworks.fi) has joined #raspberrypi
[23:00] <ReggieUK> a license for putting a char on a device, outstanding how damn trivial companies will be over IP
[23:01] * NIN102 (~NIN@ Quit (Quit: NIN102)
[23:02] <mjr> ReggieUK, 'scuse me, what was that about?
[23:02] <mjr> (and yes, yes they will)
[23:02] <mrdragons> http://www.opensource.apple.com/source/Libc/Libc-262/stdio/putchar.c
[23:02] <mrdragons> I just found it funny that there's a massive license for something so small
[23:03] <mjr> oh yeah, one of those
[23:03] <mjr> mind boggles why they have that FILE *so = stdout
[23:04] <mjr> but at least it's register!
[23:05] * multipoint (ada79aac@gateway/web/freenode/ip. has joined #raspberrypi
[23:07] <traeak> it could jsutbe a canned copyright message
[23:07] <traeak> that's what it is...canned copyright message
[23:09] * multipoint (ada79aac@gateway/web/freenode/ip. Quit (Client Quit)
[23:12] <ctyler> #include cannedcopyright.
[23:14] * |e_e| (~|e_e|@ppp-105-168.32-151.iol.it) has joined #raspberrypi
[23:16] * kismet (~kismet@client-80-3-133-201.bsh-bng-013.adsl.virginmedia.net) has joined #raspberrypi
[23:16] <traeak> heh
[23:17] * UnderSampled|tab (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) Quit (Quit: Leaving.)
[23:17] <traeak> btw looks like a 4870 is like 166W under load....lots worse than a core i7
[23:18] <mkopack> traeak: LOL, like a pit bull with a bone, you had to go track that down huh? LOL
[23:18] <Da|Mummy> is it out yet
[23:18] <mkopack> NO
[23:18] <Da|Mummy> fuckin pibot!
[23:19] <mkopack> sometime after the 20th...
[23:20] * genbattle (~quassel@203-114-137-9.wir.sta.inspire.net.nz) has joined #raspberrypi
[23:20] <traeak> hehe, no pissed off at trying to figure out this matlab code example, i did something i might have some success at
[23:22] * UnderSampled|tab (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) has joined #raspberrypi
[23:25] * kismet (~kismet@client-80-3-133-201.bsh-bng-013.adsl.virginmedia.net) Quit (Quit: Leaving)
[23:28] * Hopsy (~Hopsy@ Quit (Read error: Connection reset by peer)
[23:28] <acfrazier> he good news is that this finally means we have a date for the first batch: the boards will be finished on February 20.
[23:28] <acfrazier> YES
[23:29] * Hopsy (~Hopsy@ has joined #raspberrypi
[23:29] <mkopack> acfrazier: ASSUMING there are no problems with them!
[23:29] <acfrazier> don't jinx it
[23:29] * wiiguy (~fake@unaffiliated/wiiguy) Quit ()
[23:29] * acfrazier sets my calendar
[23:29] <acfrazier> heh third person fail
[23:29] * acfrazier sets his calendar
[23:31] <mkopack> Yeah, I know, but at this point, I wouldn't be surprised if the chinese screw it up
[23:31] <mkopack> That's why I'm hoping Eben + Liz go over to China THIS week, rather than waiting until the end of the production run
[23:32] <ReggieUK> the bit that I found funny was they actually felt the need to copyright putting a char on a screen, like it's an original idea
[23:34] <xlq> You can't "copyright putting a char on a screen".
[23:34] <xlq> "Putting a char on a screen" isn't a creative work. It's not even a work.
[23:38] <MystX> Well, you SAY that
[23:38] <MystX> But much less creative things have been copyrighted/patented
[23:38] * jluisn (~quassel@ Quit (Read error: Connection reset by peer)
[23:39] <xlq> Patented, maybe. Ideas aren't a matter of copyright.
[23:40] <xlq> http://www.gnu.org/philosophy/not-ipr.html
[23:40] <xlq> (because he rants better than I ever could :P)
[23:41] * |e_e| (~|e_e|@ppp-105-168.32-151.iol.it) has left #raspberrypi
[23:41] * |e_e| (~|e_e|@ppp-105-168.32-151.iol.it) has joined #raspberrypi
[23:43] * aerius (~aerius@host137-2.natpool.mwn.de) has joined #raspberrypi
[23:46] * aerius (~aerius@host137-2.natpool.mwn.de) Quit (Client Quit)
[23:46] * aerius (~aerius@host137-2.natpool.mwn.de) has joined #raspberrypi
[23:51] * |e_e| (~|e_e|@ppp-105-168.32-151.iol.it) Quit (Quit: |e_e|)
[23:51] * UnderSampled|tab (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) Quit (Ping timeout: 252 seconds)
[23:51] * mkopack (~mkopack@173-133-152-214.pools.spcsdns.net) Quit (Quit: mkopack)
[23:53] * Hopsy (~Hopsy@ Quit (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/)
[23:54] * johnthebear (~Johnthebe@13-165-19.client.wireless.msu.edu) Quit (Quit: Computer has gone to sleep.)
[23:54] * UnderSampled|tab (~UnderSamp@cpe-174-097-224-028.nc.res.rr.com) has joined #raspberrypi
[23:56] <traeak> i'd be curious how long a run of 10k takes
[23:57] <traeak> production line setup
[23:57] <traeak> and jamming them through
[23:58] <genbattle> also depends on the size of the setup
[23:59] <genbattle> i think the facility rpi are using is relatively small, at least in terms of production facilities
[23:59] <genbattle> sounds like total production time is about 3-4 weeks

These logs were automatically created by RaspberryPiBot on irc.freenode.net using the Java IRC LogBot.