Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2005-04-27

00:00:46 Quit silencer (Nick collision from services.)
00:00:48 Join silencer [0] (~silencer@zen.via.ecp.fr)
00:00:51 Quit silencer (Nick collision from services.)
00:01:46HClwe need some tofix list or so with glitches like that that need to be fixed..
00:01:59HClmaybe on our todo thing..
00:02:15Bagderyes
00:03:31Bagderperhaps I should have a go at it...
00:03:42HCli'm adding a glitch section to the thingstodo page
00:03:52HClso we have a place to document glitches like that so we won't forget about them
00:04:08amiconnhmpf. gcc strangeness
00:04:11t0masanybody good at regexprs here?
00:04:20Bagdert0mas: try me! ;-)
00:04:23t0mas:D
00:04:50HClhehe.
00:04:54t0masBagder: I have a string containing something like this: <tag1> info </tag1> bla bla bla <tag2> info </tag2> bla bla
00:05:05 Join Mirfle [0] (~chatzilla@ADSL222150.BRK.biu.ac.il)
00:05:11t0masI only want the info between the <tag1> and <tag2>
00:05:18t0masexcluding the tags themself...
00:05:49 Join silencer_ [0] (~silencer@zen.via.ecp.fr)
00:05:50t0masso endup with info info
00:06:05Bagderyou mean between the start tag and the end tag?
00:06:09t0masyes
00:06:51t0mas"<tag1> Some infor for Bagder here </tag1> bla bla bla <tag2> And something here too </tag2> bla bla" -> " Some infor for Bagder here And something here too "
00:10:10Bagder$r ~= s/<[^>*>(.?*)<\/[^>*)>[^<]*//;
00:10:12Bagderperhaps
00:10:19t0mas8)
00:10:23Bagder$r ~= s/<[^>]*>(.?*)<\/[^>*)>[^<]*//;
00:10:44Bagderbut that would need to be repeated and you'll get the middle text in $1
00:10:56Bagder$r ~= s/<[^>]*>(.?*)<\/[^>*]>[^<]*//;
00:11:04Bagdergee I type badly
00:11:17t0mashm... ok, and now I need a perl book
00:11:34t0mas(or can I use it in php's preg_match() ?)
00:11:39Bagder$r ~= s/<[^>]*>(.?*)<\/[^>]*>[^<]*//;
00:11:45HClthere, no more red :
00:11:46HCl :)
00:12:06Bagdert0mas: yes, if you can remove the matched part from the string and rerun the same regex on the string again
00:12:20t0masI can :)
00:15:12t0masWarning: Delimiter must not be alphanumeric or backslash in /home/tomas/dev/dict/regm.php on line 8
00:15:14t0mashmz...
00:17:16HClnight...
00:17:27Bagdernight!
00:18:22uskiday!
00:18:23uskioops
00:21:54BagderLinux host detected
00:21:54BagderSimulator environment deemed little endian
00:22:04amiconnHuh?
00:22:17amiconnWhy should Linux imply little endian
00:22:19amiconn?
00:22:26Bagderdeemed
00:22:30Bagderit tests for it
00:22:37amiconnah
00:23:32amiconnThough of PPC Linux...
00:23:38Bagderyes
00:23:40amiconnThought
00:23:47Bagderand mac os x is big endian too
00:23:50 Quit _ferenczy ()
00:24:20***Saving seen data "./dancer.seen"
00:24:20amiconnI once tried to get m68k Linux to run....
00:25:22 Join Sucka [0] (~NNSCRIPT@host81-157-72-184.range81-157.btcentralplus.com)
00:25:36Bagderis the sh1 little endian?
00:25:43amiconnnope
00:26:07Bagderyou know how the calmrisc is?
00:26:48Bagderseems to be big
00:29:26amiconnData sheet, p.25: "CalmRISC16 adopts a big endian memory format."
00:29:48Bagderthanks
00:29:53amiconnSo all our current targets are big endian
00:30:17 Quit Sucka (Client Quit)
00:30:43 Join lostlogic [0] (~lostlogic@node-4024215a.mdw.onnet.us.uu.net)
00:33:02preglowhmm
00:33:04preglowi want a 2001fp
00:33:11preglowBagder: satisfied with it?
00:33:24BagderI am
00:33:34Bagderbut I'm still using analog with it
00:33:43preglowwhy?
00:33:56preglowmy gfx card is dvi, so no worries there
00:33:58Bagdermy dvi out turned out too crappy, could only do 1280x1024
00:34:19Bagderit didn't occcur to me that it could be that bad until I tried ;-)
00:34:23amiconnOnly one tdms transceiver channel...
00:36:51preglowBagder: i can get 2001fp for 4900 NOK, so think i'll go for it
00:37:22Bagderpreglow: it is definately a very good screen for the money
00:38:16preglowwhat is dell's dead pixel policy?
00:38:26BagderI'm not sure
00:38:39BagderI've still to hear someone to get a 2001fp with a dead pixel
00:38:58preglowgood news
00:39:05preglowi don't think i'd take too kindly to dead pixels
00:39:39preglowthey seem to require seven dead pixels
00:39:47Bagderuj
00:39:54t0masEu laws...
00:40:24t0masa 1280z1024 screen with < 5 dead pixels isn't considered "dead"
00:40:28t0masz=x
00:41:08 Quit Aison ("( www.nnscript.de :: NoNameScript 3.72 :: www.XLhost.de )")
00:42:32Bagdermy endian take committed
00:42:42preglowhrmppohhhh
00:42:45Bagdernot used yet
00:42:55preglow5k is still a lot of money for me right now
00:43:45BagderI'll appreciate if a cygwin user can rerun a fresh configure
00:43:53Bagderto see if autoconf.h is made correctly
00:44:55amiconnI'll try.... target and sim?
00:45:04Bagderyes please
00:46:22amiconnconfigured for fmrecorder: #define ROCKBOX_BIG_ENDIAN 1
00:48:04amiconnfm-sim-win32: #define ROCKBOX_LITTLE_ENDIAN 1
00:48:10Bagderthanks
00:48:44 Quit t0mas ("goodnight")
00:49:58amiconnondiofm-sim-x11 works too
00:50:52amiconnHowever, when configuring the sims, I get an error message
00:50:54amiconnrm: cannot remove `/tmp/conftest-2100': No such file or directory
00:51:00amiconnThe number varies
00:51:09Bagderhm
00:51:36Bagderthe number is the process id, so that's expected to change
00:51:46amiconnAH, because on cygwin, this is called /tmp/conftest-2100.exe ...
00:52:01Bagderaaaah
00:52:51Bagderbut it says '1' when you run it, right?
00:53:18amiconnyes
00:53:39amiconnIt just doesn't get deleted
00:53:41BagderI'll fix
00:54:10amiconnYou can run programs under cygwin without the .exe extension, but not delete that way
00:55:21Bagderok, I could delete them with wildcard anyway
00:55:37Bagderlike "rm /tmp/conftest-$id*"
00:56:59Bagderfixed now
00:57:39 Quit cYmen ("zZz")
01:00
01:00:11preglowBagder: is this gfx card of yours old, btw?
01:00:20preglowi can't seem to find any max dvi resolution specs for mine
01:00:21Bagderits a matrox g450
01:00:48Bagderit is a bunch of years old
01:01:09preglowi find "2048 x 1536", but that might not be dvi related at all
01:01:26Bagderthey mostly often only mention the analog resolution
01:01:37Bagdermy card can do that res too
01:02:34amiconnAs I said, it depends on the number of tmds transmitters used for the dvi out. Dvi specs allow 1..4, but if it's only one, max. resolution is 1280x1024
01:02:47preglowhahaha, i LOVE people who think svhs == svideo
01:02:59preglowamiconn: how many does 1600x1200 need? 2?
01:03:34amiconnI think so... it's basically a question of data transfer rate
01:06:05preglowi can't find this listed any place
01:06:12preglowdamn, that's a fine detailt not to include in your specs
01:06:34Bagderwhat card do you have?
01:06:59preglowasus v9999
01:07:17preglowi think that's all there is to it
01:07:29preglowadd a 'gt' as well, perhaps
01:07:56BagderHighest DVI output up to 1600x1200 resolution
01:08:07preglowwhere'd you find this? :V
01:08:09Bagdersince its a geforce 6800
01:08:12preglowahh
01:08:15preglowthen fine
01:08:18preglowi don't need more ;)
01:08:19preglowthanks
01:08:56Bagderthe 6800 GT is max 1600x1200 too
01:09:04preglowi think i'll order one of these 2001fps tomorrow
01:09:11preglowi desperately need a new display
01:10:57pregloware there any noticable differences between using vga and dvi?
01:11:14Bagderdvi gives a sharper image
01:11:49Bagderlinus did the switch a while ago and claims a noticable improvement
01:12:07Bagderwith his 2001fp
01:13:22preglowyeah, my flatmate uses his lcd in vga mode
01:13:26preglowand i think it looks smashing
01:13:52BagderI can't complain on my current setup either
01:14:07MoosCamarogood night all
01:14:17preglowMoosCamaro: nite
01:15:06MoosCamarogood choice for your future 2001fp
01:15:15MoosCamarobye
01:15:17preglowi think i'll agree
01:15:35 Part MoosCamaro
01:15:38preglowit shall be good to finally pitch this 17" old flickering piece of shit crt into the dumpster
01:19:10 Join Aison [0] (~hans@zux166-181.adsl.green.ch)
01:24:38HClurfl...
01:24:52HClflatmates ironing their laundry in the middle of the night.
01:24:57HClmakes a lot of noise >.<
01:29:19preglowironing makes noise?
01:29:25preglowthen he's doing it wrong :P
01:30:03HClits one of those steam things
01:30:08HClit makes a beepload of hissing noise.
01:30:25HClwhat kind of person goes ironing at 1:30 am >.<
01:30:34HClno way i can sleep like this :/
01:32:00*HCl puts some music up..
01:41:55 Quit Aison ("( www.nnscript.de :: NoNameScript 3.72 :: www.XLhost.de )")
01:57:22 Part Mirfle
01:58:44HClBagder: wouldn't your configure test program give 16777216 (0x01000000) on big endian environments? rather than 0
02:00
02:07:56 Join bagawk [0] (~Lee@bagawk.user)
02:10:32bagawkHello amiconn
02:10:44amiconnhi
02:15:15 Join MRJohnSmith [0] (~johnsmith@host-69-8-205-149.cybernetcom.com)
02:15:32MRJohnSmithHi
02:16:06MRJohnSmithAnyone have SourceAnywhere?
02:16:40 Join Strath [0] (~mike@dgvlwinas01pool0-a227.wi.tds.net)
02:17:58HClwhats sourceanywhere?
02:23:58MRJohnSmithI guess this is firmware. It is for soure control
02:24:05MRJohnSmithbut who knows
02:24:24***Saving seen data "./dancer.seen"
02:28:50*HCl goes to sleep.
02:30:06 Quit MRJohnSmith ()
02:38:10 Join asdsd_ [0] (asdsd@h-67-100-28-204.miatflad.dynamic.covad.net)
02:49:52 Quit gromit`` (Read error: 145 (Connection timed out))
02:56:14 Quit bagawk ("Leaving")
03:00
03:00:15 Quit Vicious_ ("be back in a tick")
03:06:22 Join TCK [0] (TCK@81-86-97-2.dsl.pipex.com)
03:08:36 Quit preglow ("e")
03:09:31amiconnhmpf
03:11:44 Part rasher
03:18:35 Join Vicious [0] (~irc@79190854f39f1117.node.tor)
03:22:40 Join silencer [0] (~silencer@81.56.71.56)
03:39:47 Join silencer1 [0] (~silencer@zen.via.ecp.fr)
03:39:55 Quit silencer_ (Read error: 110 (Connection timed out))
03:48:20 Quit silencer (Read error: 145 (Connection timed out))
03:50:13 Quit asdsd_ (Read error: 131 (Connection reset by peer))
04:00
04:04:02 Quit Vicious (Read error: 110 (Connection timed out))
04:06:13 Join QT_ [0] (as@area51.users.madwifi)
04:18:59 Quit QT (Read error: 110 (Connection timed out))
04:24:25***Saving seen data "./dancer.seen"
04:43:17 Join wes_ [0] (~WesleyBai@h153.45.55.139.ip.alltel.net)
04:43:29 Part wes_
06:00
06:09:39 Join amiconn_ [0] (~jens@pD95D1A33.dip.t-dialin.net)
06:11:27 Join _CoCoLUS [0] (~coco@h081217139221.dyn.cm.kabsi.at)
06:18:38 Quit CoCoLUS (Read error: 104 (Connection reset by peer))
06:18:41 Nick _CoCoLUS is now known as CoCoLUS (~coco@h081217139221.dyn.cm.kabsi.at)
06:24:28***Saving seen data "./dancer.seen"
06:28:10 Quit amiconn (Read error: 110 (Connection timed out))
06:28:10 Nick amiconn_ is now known as amiconn (~jens@pD95D1A33.dip.t-dialin.net)
06:50:21 Join Lynx_ [0] (HydraIRC@134.95.189.59)
06:53:33 Join matsl [0] (~matsl@1-1-4-2a.mal.sth.bostream.se)
06:53:58 Quit Lynx_awy (Read error: 145 (Connection timed out))
06:53:59 Nick Lynx_ is now known as Lynx_awy (HydraIRC@134.95.189.59)
07:00
07:37:58 Join ShyK [0] (Shy@80.74.127.61)
07:43:51 Quit einhirn ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
07:59:56 Join Harpy [0] (TyncDXZ4Dx@dsl-hkigw7wbb.dial.inet.fi)
08:00
08:00:57 Join webguest28 [0] (~c31ce021@labb.contactor.se)
08:15:09 Join gromit` [0] (~gromit`@ras75-5-82-234-244-69.fbx.proxad.net)
08:24:29***Saving seen data "./dancer.seen"
08:35:35 Quit edx (Read error: 60 (Operation timed out))
08:40:33BagderHCl: big ending should return > 1, that's all the test cares about
08:54:33 Join austriancoder [0] (~austrianc@m3.uclv.net)
08:54:39austriancodermorning
08:55:46Bagdergood morning
09:00
09:09:09 Quit matsl (Remote closed the connection)
09:18:28 Join edx [0] (edx@pD9EAB8B1.dip.t-dialin.net)
09:26:15 Join ashridah [0] (ashridah@220-253-122-7.VIC.netspace.net.au)
09:29:33 Join Aison [0] (~hans@zux166-181.adsl.green.ch)
09:45:33HClBagder: ah
09:53:03*webmind keeps reading badger :(
09:55:08*HCl is glad his alarmclock doesn't stop after 5 minutes
09:55:24HClit takes at least 10 minutes of constant loud beeping to wake me up a little :/
09:57:09HClwebmind: thats normal, there was this studies that showed that you can switch around any letters of a word, as long as you keep the first and the last letter at its correct position, and the word'll still be readable
09:58:24webmindouhuh
09:58:32webmind-o
09:58:38webmindbut his name is bagder
09:58:53Bagder:-)
09:59:17*Bagder has just been writing a driver for a thing called "badge"
09:59:31Bagderjust imagine my frequency of bad spelling of that
10:00
10:00:12 Nick Lynx_awy is now known as Lynx_ (HydraIRC@134.95.189.59)
10:02:01HClhmmm.
10:11:50webmindheh
10:19:22 Join t0mas [0] (~Tomas@ip503c08d1.speed.planet.nl)
10:19:32t0maslo
10:23:48tedboerhi.
10:24:00tedboeris this irc logged somewhere?
10:24:05Bagderyes
10:24:08tedboerok
10:24:13Bagderwww.rockbox.org/irc
10:24:18tedboerthanks
10:24:30***Saving seen data "./dancer.seen"
10:25:34 Join Patr3ck [0] (~patr3ck@p548CB4B9.dip.t-dialin.net)
10:40:40 Quit austriancoder ("using sirc version 2.211+KSIRC/1.3.12")
10:43:01tedboerhi t0mas. i see in the log you and preglow answered my question about full duplex operation.
10:43:15tedboeri got distracted right after i asked that question...
10:44:06t0maswell... I'm not sure
10:44:20t0masI'll get an email from Philips Nederland...
10:44:41tedboerok.
10:44:43t0masbut I don't think they work now... untill monday
10:44:46tedboerit would be nice.
10:44:57t0masyes
10:45:14t0maspreglow want's to implement a nice dsp system :)
10:45:28t0masbut I'm away now.. have to get some food :)
10:46:53tedboercu
10:48:38 Quit Aison ("( www.nnscript.de :: NoNameScript 3.72 :: www.XLhost.de )")
10:49:13 Quit ferenczy (Read error: 60 (Operation timed out))
10:49:16 Join ferenczy [0] (~ferenczy@fw.qcm.cz)
10:49:42 Join Aison [0] (~hans@zux166-181.adsl.green.ch)
10:50:05 Join bobTHC [0] (~foo@l04m-6-169.d1.club-internet.fr)
10:51:15bobTHCmornin' all !
11:00
11:14:19 Join tuco [0] (~martina@spurv.ii.uib.no)
11:15:02tucoHCl, if I want to try your new db tool, is that in the daily tarball?
11:16:49 Quit lostlogic (Read error: 110 (Connection timed out))
11:21:51tuconevermind, downloaded it from your link on the wiki and got it running just fine.
11:27:57 Part tuco ("Leaving")
11:31:33 Join lostlogic [0] (~lostlogic@node-4024215a.mdw.onnet.us.uu.net)
11:49:04*t0mas is back
12:00
12:08:11 Join MoosCamaro [0] (MoosCamaro@m214.net81-66-158.noos.fr)
12:08:22MoosCamarohi all
12:08:51 Join preglow [0] (thomj@s183a.studby.ntnu.no)
12:09:51preglowwhat, linus is still taking his car for a spin? :P
12:10:20MoosCamarohi preglow
12:10:31Bagderlinus is crossing the country today I believe
12:10:38Bagdergoing west
12:11:47preglownice weather for it
12:11:49preglowat least her ;)
12:11:51preglowhere <-
12:12:01Bagderit is great here too
12:12:55preglowarghhhhh
12:13:03preglowagain courier-imap decides to act up
12:24:33***Saving seen data "./dancer.seen"
12:28:15 Nick QT_ is now known as QT (as@area51.users.madwifi)
12:33:24 Join asdsd_ [0] (asdsd@h-67-100-25-193.miatflad.dynamic.covad.net)
12:34:11 Part asdsd_
12:40:18 Quit Strath ("Client closed")
12:43:41HClback
12:44:02HCllittle bit of good news, markuns gonna look at the grayscale patch :)
12:46:47HClwhere's rasher when you need him.
12:50:25 Quit uski ("Leaving")
12:59:43 Join cYmen [0] (~cymen@nat-ph3-wh.rz.uni-karlsruhe.de)
13:00
13:07:29 Join einhirn [0] (Miranda@bsod.rz.tu-clausthal.de)
13:19:47 Join T0mas_ [0] (~Tomas@ip503c08d1.speed.planet.nl)
13:20:54 Quit HCl ("Lost terminal")
13:21:54 Join hcl [0] (hcl@titania.student.utwente.nl)
13:22:32 Quit t0mas (brown.freenode.net irc.freenode.net)
13:22:32NSplitbrown.freenode.net irc.freenode.net
13:31:44 Quit ferenczy (Remote closed the connection)
13:33:53preglowhcl: wooot!
13:34:54hclmmm?
13:34:56 Nick hcl is now known as HCl (hcl@titania.student.utwente.nl)
13:34:56HClsup?
13:35:03preglowmarkun greyscale patch new
13:35:07HCloh, right.
13:35:11HClhe said he was gonna look at it
13:35:27preglowmake him join so we can badger him about it ;)
13:35:37HClhad a nice idea of using a lookup table to do 1bit -> 2bit bitmap conversion
13:35:57preglowdo you need a friggin lookup table to do that? :P
13:36:09HClif you want it fast, yes.
13:36:16HClyou need to convert an 11111111 byte
13:36:23HClto two bytes of 10101010 10101010
13:36:34preglowahh, yes, of course
13:36:35HClthats not easy to do with bit math
13:36:37preglowi was thinking single bits
13:36:37preglowheh
13:36:58preglowthat'll be a pretty sizeable table
13:36:59HClso he's gonna take the lower 4 bits and upper 4 bits
13:37:05HClin a 16 entry lookup table
13:37:08HClto produce the two bytes
13:37:13preglowahh
13:37:13preglowgr8
13:42:35 Join ripnetuk [0] (~george@82-70-100-230.dsl.in-addr.zen.co.uk)
13:42:40 Part ripnetuk
13:43:07 Join ripnet [0] (~mirc@82-70-100-230.dsl.in-addr.zen.co.uk)
13:51:59*HCl committed more untested functions to database.c that should work in theory :3
13:53:55NHeal(timeout) brown.freenode.net irc.freenode.net
13:58:29 Quit lostlogic ("Going to the moon")
14:00
14:01:16preglowwhat does a lcd panel need usb sockets for? :V
14:01:56Bagderit brings them closer to me
14:02:07preglowso it's a hub?
14:02:09preglowfriggin' leet
14:02:10Bagderyes
14:02:12preglowthat's exactly what i need
14:02:16preglowhaha
14:02:20preglowi need this display
14:06:37tedboerhi preglow
14:06:50tedboersorry i didn't get back to you yesterday
14:07:07preglowon what?
14:07:08preglowahh :P
14:07:09preglowremember
14:07:11tedboeri got called away right after i asked my question about fullduplex
14:07:36tedboeri read what you said in the log
14:07:58tedboerit would be really nice.
14:08:21preglowi agree
14:08:30preglowlet's hope it can do full duplex
14:08:32tedboeri think that definitely has all the numbers to be a killer app
14:08:54tedboerbut do you think that such an fx processor should run inside rockbox?
14:09:06preglowwhy not?
14:09:14preglowwe need drivers and a framework
14:12:44tedboerjust wondering.
14:13:40tedboerso you are thinking of a plugin?
14:13:50preglowi don't know how, yet
14:13:56preglowbut some kind of plugin, i guess
14:14:06preglowwe might end up with quite a number of plugins
14:14:12tedboer:-)
14:14:14HClheheh.
14:14:15Bagderplugins are goodness
14:14:16preglowplugin types
14:14:23preglowlike codec plugins, dsp plugins
14:14:30*HCl made the initial plans for his build-a-playlist-from-a-database-search plugin
14:14:41preglowi guess this could be a dsp plugin capable of taking realtime input
14:14:42HClits gonna be interpreter based :)
14:14:57tedboerchainable...
14:14:59preglowHCl: haha, how?
14:15:02HCland allow stuff like
14:15:08HCl"all songs where :"
14:15:16preglowtedboer: sure, but don't expect too much from this thing, the hardware is quite limited in ways
14:15:27HCl"year >= 1980 and (title contains albumname)"
14:15:28HCl :P
14:15:41HClit'll allow almost anything, and it should be fairly fast too o.o
14:16:13HClmight require too much disk activity on archos, on iriver, i'd have to load big parts of the database into ram entirely, to prevent disk spinnin
14:16:16HClg
14:17:18Bagderyou'll only get that amount of ram by using the audio buffer
14:17:20HClpretty much, its gonna be an interpreter that accepts a tiny boolean expression language and applies them to all the songs in the database
14:17:25Bagderand that you can do on the archos too
14:17:33HClBagder: i know, it wouldn't be able to run while playing mp3
14:17:39tedboerpreglow: sure, but i'd say that thie coldfire is more than powerful enough to do some not to complicated time-domain dsp. some guitar fx are not to heavy on the cpu.
14:17:41Bagderyeps
14:17:43HClthere are simpler search engines possible as well
14:17:51HClbut for now, i'm going for possibility, not speed
14:17:57HClwe can always build a more limited, faster one later
14:18:01Bagderindeed
14:18:04preglowtedboer: well, yes, we'll be pretty tied to time domain unless someones codes a _really_ fast fft
14:18:17preglowlots of fun to be done in the time domain anyways
14:18:18tedboerfftw...
14:18:25preglowhaha
14:18:34preglowi don't think fftw is very optimized for the coldfire
14:19:02tedboerbut a 68000 fftw should work just fine, right?
14:19:12preglowi have no idea
14:19:16preglowi've just used fftw for x86
14:19:50tedboerhmm. a friend of mine coded an fft in assembly on the mac II years ago.. i'll ask him.
14:20:02preglowit'll have to be recoded
14:20:13preglowcoldfire doesn't have all the instructions the 68k has, nor all the addressing modes
14:20:23tedboernot sure if that was for the 68020 or for an audiomedia card (56000)
14:20:27preglowand our coldfire has some multiply-accumulate instructions that are pretty good
14:20:51tedboerok... i really should have a look at the coldfile docs.
14:21:24 Join Patr3ck_ [0] (~patr3ck@p548CB52E.dip.t-dialin.net)
14:21:25tedboerfire
14:22:26preglowCFPRM.pdf is pretty nice
14:22:37preglowhas an instruction set reference, at least
14:22:39preglowso you see what's there
14:24:35***Saving seen data "./dancer.seen"
14:26:50 Quit Patr3ck (Read error: 60 (Operation timed out))
14:32:30 Join lolo-laptop [0] (~lostlogic@68.251.84.226)
14:35:37 Part ripnet
15:00
15:27:16 Quit webguest28 ("CGI:IRC")
15:29:44 Quit mdeboer ("Leaving")
15:29:58tedboerthat instruction set looks very familiar. nice.
15:30:05amiconnHCl: Concerning your 'obvious optimizations', database.c: lines 201, 228: Rockbox provides a nice macro, which does esentially the same as your code, but is more readable (imho)
15:30:39amiconnE.g. line 228 would become: amount=MIN(1024, bytes);
15:31:16amiconnOf course there is also a MAX() macro
15:32:39preglowtedboer: if you're looking to do dsp coding, have a look at the emac instructions as well, it pays to be familiar with what they can do
15:33:10HClamiconn: mm, i figured something like that, feel free to update it
15:33:14*HCl has to go do stuff
15:33:15HClbbl
15:33:25*HCl actually already started on the searchengine :X
15:33:28HClbbl.
15:33:31*preglow shrieks
15:33:42preglowcracks have started forming in my vi reference mug!
15:37:47tedboerpreglow: yes, the emac instructions look very familiar to the M56K
15:38:35tedboeryou'll need a vim mug. that's the improved version
15:39:11preglowif this cracks, i'll get it
15:39:14preglowbut yeah
15:39:23tedboer:-)
15:39:26preglowmac.l and msac.l in fractional mode are more or less all you need for lots of things
15:39:55preglowi'm using them with lots of speed improvement in optimizing codecs
15:40:04preglowbut i have to go
15:40:05preglowlater
15:40:09*preglow disappears
15:40:58tedboerok
15:46:00 Quit ashridah (Read error: 113 (No route to host))
15:46:53 Join ashridah [0] (ashridah@220-253-121-86.VIC.netspace.net.au)
16:00
16:05:56 Quit ashridah ("Leaving")
16:18:45 Join ismi [0] (~50db05a8@labb.contactor.se)
16:18:51ismihi
16:20:06bobTHChi
16:20:29ismispeak someone german?
16:21:13bobTHCnot me but some other ...
16:21:38ismiaha
16:22:15ismii live in swizerland. and i have a question about the Iriver H120
16:23:08ismii think, that i will buy it, but i dont know about the Firmware
16:23:20bobTHCit's a good choice
16:23:31ismiis rockbox running god?
16:23:40bobTHCbut u seems to speak english well
16:23:48bobTHC;)
16:23:50ismithanks
16:24:06ismiit must be work!
16:24:38***Saving seen data "./dancer.seen"
16:24:39ismii had 3 years English lessens yet.
16:24:44ismi;-)
16:24:57bobTHCso if i were u i'll buy the H140, because it's support standart 1,8 inch HDs
16:25:25ismiand what supports the H120?
16:26:02bobTHCnon standard hd , so u cannot upgrade it "easily" in the futur
16:27:54bobTHCin all case rockbox will work alike on the 2 H1xx
16:29:33ismiyou know, i`m a student, and i haven`t so much money. but i must look... perhaps it will be enough...
16:29:42ismiand whats about the H10
16:31:18bobTHCi understand, so the H120 is a good choice if you want to use rockbox quickly, because the H10 is not supported by rockbox for the moment
16:33:38ismihm... but the H10 is newer than the H120. and it have a colour display. but it have yust 5GB...
16:34:23bobTHCthe H1XX series rockbox port is pending but for other iriver... i haven't much info
16:34:47ismido you knew how the feeling is when you have it in your hands? does it feel robust, etc...
16:35:17ismiok
16:35:43bobTHCi think the H3XX serie will be the next port but after....
16:36:30ismiok. thanks. now i must go.
16:36:34ismisee you later
16:36:37bobTHCcu
16:36:37ismiby
16:36:55 Quit ismi ("CGI:IRC (EOF)")
17:00
17:25:56 Quit bobTHC ("Smoke Weed Every Day !")
17:33:47 Join MeTaXo [0] (~3ea7dd0b@labb.contactor.se)
17:34:42 Quit MeTaXo (Client Quit)
17:34:51 Join MeTaXo [0] (~3ea7dd0b@labb.contactor.se)
17:34:51 Quit MeTaXo (Client Quit)
17:34:58 Join MeTaXo [0] (~3ea7dd0b@labb.contactor.se)
17:35:35 Quit Nibbler ("blubber")
17:36:42 Quit MeTaXo (Client Quit)
17:40:30 Join Sucka [0] (~NNSCRIPT@host81-156-157-52.range81-156.btcentralplus.com)
18:00
18:04:05 Quit Patr3ck_ ()
18:24:39***Saving seen data "./dancer.seen"
18:38:18 Join TCK- [0] (TCK@81-86-99-255.dsl.pipex.com)
18:56:45 Quit TCK (Read error: 110 (Connection timed out))
19:00
19:11:30HClhello
19:12:08preglowa h10 port is looooong away
19:12:32HClheh.
19:12:34HClwhy?
19:13:04 Quit HCl ("Lost terminal")
19:13:07preglowi think getting specs will be harder this time around
19:13:10preglow...
19:13:25 Join hcl [0] (hcl@titania.student.utwente.nl)
19:13:30 Nick hcl is now known as HCl (hcl@titania.student.utwente.nl)
19:13:33preglow19:13 < preglow> i think getting specs will be harder this time around
19:13:38HClokay
19:13:42preglowand it's arm
19:13:42HClsorry, screen malfunctioned
19:14:21preglowwhich is a wholly new cpu arch for rockbox
19:15:32preglowbut i'm out again, anime time
19:15:33 Quit HCl (Client Quit)
19:16:05 Join hcl [0] (hcl@titania.student.utwente.nl)
19:16:10hcli knew that was coming.
19:16:12hclscreen crashed.
19:16:15 Nick hcl is now known as HCl (hcl@titania.student.utwente.nl)
19:19:08 Join Lynx0 [0] (HydraIRC@134.95.189.59)
19:19:34 Quit Lynx_ (Read error: 145 (Connection timed out))
19:19:35 Nick Lynx0 is now known as Lynx_ (HydraIRC@134.95.189.59)
19:33:45*HCl is confusing himself :/
19:44:10HClthere, much better ^^
19:44:13 Join webguest75 [0] (~d86bc3d0@labb.contactor.se)
19:44:59webguest75hi all
19:45:49webguest75I'm hoping for a bit of help with what I thought'd be an easy operation: mounting an archos under linux.
19:46:17webguest75cfdisk sees it as /dev/hdb, but mount won't seem to accept it
19:46:29webguest75oops, /dev/sdb
19:47:27webguest75mount -t vfat /dev/sdb /mnt/archos was the command I tried.
19:47:39 Join rasher_ [0] (rasher@zork.zork.net)
19:47:55webguest75hi
19:48:06rasher_hclmerf
19:48:19rasher_HCl: you called?
19:48:42HCloh
19:48:49HClyea, i need hash stuff..
19:48:59HClwebguest75: try /dev/sdb1
19:49:24HClrasher_: are there any tests yet on how unique hashes are?
19:49:37HClspeed isn't important anymore, according to basher we'll have to search for filename either way
19:50:04rasher_well if you use 1024 bytes, they should be unique enough
19:50:22rasher_512 bytes produced 2 duplicates on my collection
19:50:24HClwell, we should probably pick a bit more..
19:50:41HClanyways, whats the status of the code for that?
19:51:10rasher_(in those cases I used half from the beginning, half from the end)
19:51:26rasher_I don't have anymore than what you've already seen
19:51:28HClBagder said we couldn't use half of the end
19:51:32HClwhat was the url again?
19:51:33webguest75HCl: Thanks, that worked. :)
19:51:41HClnp :)
19:51:43rasher_rasher.dyndns.org/~rasher/hash.c">http://rasher.dyndns.org/~rasher/hash.c
19:51:46HClk
19:51:51rasher_19:51 <HCl> Bagder said we couldn't use half of the end
19:52:00rasher_how so?
19:52:10HCli dunno, something with the ends not being read in or so
19:52:25HClfor now i'm having too much fun coding my search interpreter
19:52:26HClxD
19:52:39rasher_oh, you mean when loading the file?
19:52:56HClyea
19:52:59HCli think
19:52:59rasher_I still think we should hash music data
19:54:45 Part webguest75
19:55:52rasher_I could add that
19:56:23rasher_(seeking to first mpeg frame)
19:57:19HClyes, please
19:57:32*HCl (thinks) he's making good progress on his parser
19:57:37HCli love compilers / interpreters
19:57:42rasher_guess I could also move the crc table to iram
19:57:45rasher_how's that done?
19:59:51rasher_finding an mpeg frame is just seeking to first 0xff , isn't it?
20:00
20:00:43HClno idea :x
20:02:29 Join ismi [0] (~50db05a8@labb.contactor.se)
20:02:37ismihi
20:02:59rasher_ismi: iriver 120 accepts 1,8" harddrives as well
20:03:16rasher_just thinner ones than h140, so they can't be as large
20:07:32 Join MeTaXo [0] (~dgydg@62.167.221.11)
20:09:06rasher_(single-platter)
20:09:11 Nick rasher_ is now known as rasher (rasher@zork.zork.net)
20:09:12ismithanks rasher
20:09:27ismi;-)
20:10:11ismican i test the rockbox firmware on my pc???
20:10:53ismiwith an emulator or somthing like???
20:11:15rasherThere is an emulator, yes
20:11:44rasherWindows?
20:12:04ismiyes, windows XP Professional
20:14:42ismiwher i can download this Emulator?
20:15:37MeTaXohi all
20:16:29MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140
20:16:29MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140
20:17:09HClits not for end users yet.
20:17:19MeTaXo.....
20:17:29MeTaXobut u can use it ???
20:17:38HClit doesn't play any music yet.
20:17:43MeTaXooch
20:17:56MeTaXoso it still dosent
20:18:05HClno.
20:18:09HClit doesn't.
20:18:23MeTaXohow come thats the most important
20:18:29MeTaXo!!!
20:18:31MeTaXohehe lol
20:18:52rasherismi: I don't know where the simulator hides - don't use windows.. sorry
20:19:38ismiand whats about Linux?
20:20:25rasherthere's a simpler simulator, but you'll have to build it yourself
20:20:38MeTaXoah linux dose anyone know were 2 download a boot up manager
20:21:20MeTaXo2 start linux i reinstalled win xp
20:23:01MeTaXohello ?????????????????????????????????????????????????????????????????????????????????????????????????????
20:23:24rasherTake it easy
20:23:35rasherwe don't always answer within 109 seconds
20:23:40rasherbesides, this is hardly a linux helpdesk
20:23:43MeTaXokk
20:23:45MeTaXosyr
20:23:51rasher10 seconds, that is.
20:24:12rasherBoot your distros install cd and choose to install boot loader
20:24:13MeTaXo i thoult this was a char
20:24:27MeTaXoi dont have a cd any more
20:24:30MeTaXohehe
20:24:38MeTaXo:)
20:24:41***Saving seen data "./dancer.seen"
20:24:47 Quit MeTaXo ()
20:25:59 Join MeTaXo [0] (~dgydg@62.167.221.11)
20:26:42MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140
20:26:44MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140
20:27:08HCldon't repeat yourself.
20:27:19MeTaXowell no one ansers
20:27:27HClrepeating isn't gonna help.
20:27:35MeTaXoand dont care
20:27:35rasherasking again won't help you
20:27:42MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140$
20:27:47MeTaXocan some one help me i dont know how 2 get rockbox 2 work on my i river ihp 140$
20:27:48*HCl twitches.
20:27:55MeTaXohehe lol
20:27:58pabsMeTaXo: there's a link on the rockbox iriver porting page
20:27:59*HCl forces a smile and goes back to work
20:28:09rasherMeTaXo: http://www.rockbox.org/twiki/bin/view/Main/IriverBoot
20:28:10MeTaXoi dont get that
20:28:11 Nick lolo-laptop is now known as lostlogic (~lostlogic@68.251.84.226)
20:28:11pabsMeTaXo: but it's a little bit of work to get things going
20:28:19rasherread that
20:28:20pabsactually, a fair bit of work
20:28:22MeTaXoah ill have ao look tnx
20:28:29rasherif you still can't figure out how to do it, don't even try
20:28:41MeTaXohehe ok
20:29:03MeTaXojep i dont get that stuff
20:29:25MeTaXoextraxt to so and so........... ??????'''
20:29:26rasherseriously, if you're not very technically inclined, there's not much fun in using rockbox right now anyway
20:29:27 Join asdsd_ [0] (asdsd@h-67-100-30-2.miatflad.dynamic.covad.net)
20:29:38amiconnHCl: Hashing file content relative to the end is a bad idea imho, for at least 2 reasons
20:29:40pabsMeTaXo: it'll be better in a couple of months probably
20:30:10MeTaXowaht u mean technically inclined ?????
20:30:26pabsMeTaXo: like, know how to build a cross-compiler, patch your firmware, that sort of thing
20:30:28MeTaXoi dont realy want 2 wait that long
20:30:29 Quit Sucka ("a bird in the bush is worth two in your house")
20:30:45MeTaXoprogramm ????
20:31:01pabsMeTaXo: there's a lot of wrk that needs to be done to rockbox before it'll be ready for non-programmer types to use it on the irivers
20:31:17MeTaXok
20:31:21pabssorry!
20:31:37MeTaXoand is there a driver 4 i river a better one pls
20:31:48pabsnot that i know of
20:31:53HCl"driver" ?
20:31:55MeTaXodamn
20:32:10MeTaXofirmware !!
20:32:11pabsMeTaXo: just give it a few more months
20:32:14MeTaXohehe
20:32:19MeTaXokk
20:32:22MeTaXoill see
20:32:26amiconnHCl: (1) The end of the file might not be in ram, either a really big file (think .wav or .flac), or simply the beginning of a file
20:32:38pabsMeTaXo: alrighty
20:32:45HClamiconn: yea, i don't really mind, i was just wondering.
20:32:56 Quit Zagor (Remote closed the connection)
20:32:59MeTaXooch sad was exsited
20:33:04amiconnHCl: (2) If you calculate position relatively to the end, this end might change if you trim away some gibberish
20:33:33ismidoes anybody have the new player from Olympus?
20:33:43MeTaXonope
20:33:45amiconnHashing from the beginning, after id3v2 tags if any, should be fine
20:33:47MeTaXobye all
20:33:58 Quit MeTaXo ()
20:34:20rasheramiconn: sounds like the Right Way
20:35:17HClrasher: think you could thange that...?
20:35:27*HCl thinks he finished the base of his search engine o.o.
20:35:47HClit just needs to be linked to the database, and it needs something to feed it a query in the right token format
20:36:10pabsi was thinking about porting sqlite to rockbox
20:36:34HClheh.
20:36:38HCli doubt thats fast enough :P
20:36:44rasheramiconn: can you confirm that mpeg frames (and nothing else) starts with 0xff?
20:36:48HCloh.
20:36:49HClhmm.
20:36:55*HCl forgot one string operator
20:37:00HCl"equals"
20:37:28 Join webguest29 [0] (~d86bc3d0@labb.contactor.se)
20:37:37rasherheh
20:37:39pabsHCl: i'm assuming you haven't actually uesd sqlit
20:37:43pabse
20:38:16rasheryou assume correctly
20:38:55webguest29HCl: You helped me earlier mounting the archos. I can see it, but I can't write to it.
20:39:07pabsanyway, looks lie the memory allocation tsuff would need to be fixed, and a couple other things
20:39:18pabsotherwise it should be a fairly straightforward port
20:39:27pabsthe question is whether or ot it'd be useful to anyone
20:39:40pabscertainly for building a database of music it'd be useful
20:41:09rasherWe already have that :)
20:41:45webguest29HCl: root@knoppix-hp1:/mnt/archos# touch testfile touch: cannot touch `testfile': Read-only file system
20:43:22webguest29but mount says: type vfat (rw)
20:50:30HClawwww. my kitty brought a kitty friend home xD
20:50:42HClwebguest29: dunno, mount -o remount,rw ?
20:51:01webguest29Oh no. dmesg says: Directory 4676: bad FAT Filesystem panic (dev 08:11). FAT error File system has been set read-only
20:51:12rasherHCl: soon, they'll be having all the sex
20:51:22HClthey're both male >.>;
20:51:29rasherand you'll end up with a gazillion kitties
20:51:33rasherph33r
20:51:37rasheroh.
20:51:45webguest29I wonder why mount doesn't report it as read-only
20:51:46rasherThat sortof ruins my horror scenarios
20:51:50HCl :P
20:52:36rasherwebguest29: guess it's time to fsck.vfat :)
20:54:08webguest29I didn't know fsck worked on non-linux filesystems.
20:54:33rasherthere's a vfat version
20:54:42rasherit may not be installed by default though
20:54:44amiconnrasher: I'm not sure whether you can rely on that. Rockbox already contains an id3 tag parser which can handle id3v2 (how else would rockbox be able to display id3v2 tag info)
20:54:54amiconnfirmware/id3.c
20:55:26rasherafaik id3v2 doesn't allow 0xff to occur anywhere?
20:55:52amiconn[20:54:45] <amiconn> rasher: I'm not sure
20:56:03rasherwho'd want a ÿ anyway? ;)
20:56:15rasherFair enough.
20:56:17amiconnI didn't deal with id3 parsing so far
20:56:38rasheryou should be grateful
20:56:50pabsid3v2 has a length associated with fields
20:57:15rasherI think it defines a total length as well
20:57:20pabsyeah it does
20:57:20rasherin the header
20:57:20rasherif memory servers
20:57:28pabsi guarantee 0xff can appear in id3v2 headers
20:57:31rasherserves
20:57:57rasherDammit
20:58:16rasherthat makes things so much more complicated
20:58:55amiconnJust reuse the existing parser to skip the tags
20:59:30amiconnOr, still better: hook into the buffer refill routine to extract the start of actual music data
20:59:57HClpabs: no :P
21:00
21:00:24pabsHCl: go ahead and embed an image in your id3v2 headerthen and see how far that gets you
21:00:38pabsHCl: i'll be sitting here watching with a smirk on my face
21:02:25amiconnrasher: Btw, what is an "A with tilde - reversed question mark"?
21:03:14rasherThat's a gem
21:03:20rasherIn the font I submitted?
21:03:35amiconnNo
21:03:41amiconn[20:56:04] <rasher> who'd want a ÿ anyway? ;)
21:04:03webguest29HCl: looks like I'm gonna need to spend some time w/archos on AC(mains) power to back it up.
21:04:15rasheroh, that's an y with an umlaut thingy
21:04:35rasherin n any decent charset :)
21:04:42webguest29I might just take this a a reason to upgrade the HD. Thanks for the help.
21:05:27rasherah - it's "0xff 0x00"
21:05:42rashereh.. no
21:05:47rasherI'm an idiot
21:05:49amiconnrasher: Argh, utf-8
21:05:52rasheranyway, this is probably not done
21:05:56rasherwso yes
21:06:06 Join stevenm [0] (~steve@176-46.mam.umd.edu)
21:06:29rasherOh, sorry :)
21:06:44HClpabs: ?
21:06:57HClwe don't search for images o.o.;
21:08:00pabsall i'm saying is that if you want to extract content from id3 headers you'll need to actually parse them
21:09:49rasherWe're not even talking about that
21:09:53rasherwe just need to skip the header
21:10:17rasherbut amiconn's suggestion sounds like more fun
21:10:25pabshten read the first 10 bytes, get the length, and skip it
21:10:49rasherread from the audio buffer
21:10:54pabsi think there can be several id3v2 headers in a file though
21:10:56pabsah yeah, even better
21:11:24amiconnrasher: Supporting multiple formats with different tagging systems will be fun...
21:11:50rasherbut that will be handled elsewhere though?
21:11:55rasherif we
21:11:55rasher're reading from the audio buffer
21:12:21rasherBut yes, that'll be great fun
21:12:42rasherat least I think id3 is the worst format :)
21:12:46rasherso the worst part is done
21:12:56rasherworst *parser* part
21:15:27pabsid3v1 is awful
21:15:49pabsfollowed immediately by id3v11, which is possibly even worse
21:15:56pabsthen id3v22
21:15:57rasherNot if you're a parser
21:16:08pabsid3v23 and id3v24 are actually okay, in my opinion
21:16:46rasherid3v1(.1) is like 3-4 lines of code :)
21:17:00rasherv2 is a horrible horrible mess
21:17:01amiconnid3v1/v1.1 is plain & simple
21:17:22pabsyes, after you seek to the end of the file
21:17:34rasherat first glance, it's not too bad, but then arises all the special-cases
21:17:37pabsand golly i hope your music fits nicely into the tiny list of genres
21:17:49rasherthen yo0u kill yourself slowly with a spork
21:17:52pabsand boy i sure hope your artist name isn't too long and album name
21:18:22rasherNoone said id3v1 is a pleasure to use
21:18:33rasherbut it sure is easy to parse
21:18:36pabsmy two major gripes with id3v2 is the change from 3 => 4 character codes from v22 to v23,v24, and the character encoding stuff kind f annoys me
21:18:57ismii have a question about the Iriver H120
21:18:59rasherand id3v2 otoh is a right nightmare to parse
21:19:13rasherask ahead
21:19:13 Quit webguest29 ("CGI:IRC (EOF)")
21:19:27ismidoes .m3u lists or other funktions?
21:20:04pabsrasher: i agree that it takes more work to parse, but i think it's a fundamentally less broken tagging format
21:20:05rasherSortof
21:20:11rasherit's a bit of a kludge though
21:20:11 Nick Lynx_ is now known as Lynx_awy (HydraIRC@134.95.189.59)
21:20:42rasherismi: but yes, you can play playlists
21:21:19rasherIn my opinion id3v1 and v2 are both broken, but on each end of the horror scale
21:21:29ismioh. cool
21:21:29rasherwith "sane" being the mid-value
21:22:00rasherismi: of course, when rockbox is ready, there'll be nicer playlist support
21:22:04rasher*much* nicer
21:23:36ismiand what`s about the equalizer?
21:24:13ismiin the Firmware from Iriver where just two funktions. Bass Boost and Bass trebler
21:24:20rasherHaven't really used it.. there are some presets, and I think maybe (?) a user-defined
21:24:36rasherNo
21:24:41rasherThat's not right
21:25:35ismiwhat isn`t right?
21:29:35 Quit ismi ("CGI:IRC (EOF)")
21:30:19 Join ismi [0] (~50db05a8@labb.contactor.se)
21:30:35ismiups, opera was crashd
21:33:43 Quit TCK- (Read error: 110 (Connection timed out))
21:40:16HClhrm..
21:44:06ismiwhat?...
21:46:13preglowwuuut
21:51:49HClwell, with a bit of (bad) luck, i'll have to implement my interpreter search engine as a module sooner than i had planned
21:53:07HClis there a proper multiple-files-plugin structure thing yet?
21:53:10HClamiconn: ?
21:55:56 Quit stevenm ("Leaving")
22:00
22:00:08 Join Strath [0] (~mike@dgvlwinas01pool0-a239.wi.tds.net)
22:03:03HClpabs: id3 tags *are* getting parsed. we have a tagdb generator for it.
22:03:07ismii wish you all a good night
22:03:11ismibye
22:03:16HClcya
22:03:50 Quit ismi ("CGI:IRC (EOF)")
22:04:59HClmeh.
22:05:00HCli need a name.
22:06:03preglowyou've forgotten your old one? :V
22:07:47HClno
22:07:49HCli mean
22:07:52HClfor my searchengineplugin
22:08:08 Join Tang_ [0] (~chatzilla@ARennes-252-1-48-213.w83-195.abo.wanadoo.fr)
22:08:20HClit seems i'm forced to actually implement it on the target sooner than i had planned
22:09:12rasherdbsearch?
22:09:27Tang_Hi
22:09:33Tang_:)
22:09:43HClmeh.
22:09:47HCltoo plain :/
22:09:50*HCl wanted something humerous
22:09:51Tang_the DB seems to be quite advanced indeed
22:10:01Tang_nice
22:10:02HClyou ain't seen nothing yet :)
22:10:21Tang_eh eh
22:10:29Tang_still motived HCl?
22:10:33Tang_;)
22:10:57HClyupyup, writing an interpreter based boolean expression search engine was fun
22:11:06HClunfortunately, i think i've almost finished it and need to start testing o.o
22:11:12Tang_Great
22:11:19HClbut i need a name.
22:11:22HCldbsearch is too plain.
22:11:36HClbut i guess the mystery of canned beef in japanese is too long and people will object to that too.
22:11:39Tang_i guess rioDJ will be "game over"
22:12:03HClwe'll even be able to do "find me all songs in which the title contains the album name
22:12:21HCl :p
22:13:09HClthat reminds me, i forgot to add the equals string operator
22:13:13*HCl goes to implement
22:14:11rasherI think plugin names should be descriptive.. since you have nothing else to go by
22:16:39HCltrue.
22:16:52HCli need to find something funny with search :/
22:19:02 Join Sucka [0] (~NNSCRIPT@host81-156-157-52.range81-156.btcentralplus.com)
22:21:02HCli'll just call it searchengine
22:21:03HCli guess.
22:24:42***Saving seen data "./dancer.seen"
22:25:43Tang_you need a name for what?
22:25:57Tang_or the plug?
22:26:56Tang_why not DBoogle
22:26:58Tang_lol
22:27:02Tang_(i'm out)
22:27:17HCl :P
22:27:19HClhmmm.
22:27:20HCl :P
22:28:47Tang_with search
22:28:56Tang_hard to find pun
22:29:18Tang_specialy when we don't speak english very well
22:29:20Tang_:D
22:38:57HCli'll just keep it searchengine
22:43:01amiconnrockbase ;-)
22:46:43preglowdatabox!
22:47:57Tang_datbox is nice to me
22:48:02Tang_databox sorry
22:48:04rasherdackbose
22:48:13Tang_dackbose?
22:48:43rasheralternating 2 letters from each word :)
22:50:24preglowkabaoaxe!
22:50:26Tang_ah lol
22:50:30Tang_too subtle for me
22:51:06rasherrotabax
22:51:50rasher g = f.toNFA().toNFAM().reverse().determinize().toNFA().toNFAM().reverse().determinize();
22:51:56rasherJava scares me
22:52:08preglowjava :///
22:52:44rasherUni is completely obsessed with Java
22:53:52HCljava is nice
22:54:05HClbut thats abuse of it o.o.
22:54:27rasherit's so verbose :-\
22:54:29*HCl might rename it to databox one day
22:54:54HClits odd when you start coding in pseudocode
22:54:58HClthen slowly convert it into actual code
22:55:00HCl :|
22:56:02Tang_bye
22:56:05Tang_i've to go
22:56:07 Quit Tang_ ("Chatzilla 0.9.66 [Mozilla rv:1.7.5/20041108]")
22:56:29preglowi've never used pseudocode :/
22:57:44rasherThere's a lack of good compilers 
22:57:47HClits easy, its just a description of what that code at that point is supposed to do.
22:57:52HClrasher: write a new one ! :D
22:58:05HClthe searchengine is based on compiler theory :)
22:58:13HClimplemented as an interpreter
22:58:15preglowi tend to keep everything in my head until i simply start writing code
22:58:18preglowwhich is not always optimal, heh
22:58:28HCli have that, but i often lose stuff and need to correct other stuff.
22:59:05rasher<my-programming-language-prof> WUse SML!
22:59:29rashers/W//
22:59:56 Quit einhirn ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
23:00
23:01:45*preglow orders a lcd panel ://
23:05:38HClugh.
23:05:50HCli hate how there's no way to kill a plugin or quickly exit.
23:06:00HClis there any way to export an goto location? *forces a smile*
23:06:25HClugh.
23:13:25HCltitania:/home/hcl/rockbox/rockbox-devel/iriver# ls -al searchengine.rock
23:13:25HCl-rwxr-xr-x 1 root root 70372 Apr 27 23:10 searchengine.rock
23:13:28HClpheer.
23:14:14*preglow ph33rs
23:16:24rasherjesus h. christ
23:17:00HClmm?
23:17:08HClyes, my son? :P
23:17:53rasherthat's a largeish rock
23:18:01HCl :x
23:18:04HClmerf.
23:18:08amiconnIt's too large :(
23:18:08HClit'll still load on iriver.
23:18:21HCli'll look at size later..
23:18:42preglowdaym
23:18:43HCllike i said earlier, i'm gonna focus on possibility first, we can always build a limited searchengine later
23:18:44preglowwhat's in it?
23:18:54preglowthat's a lot of code
23:19:07HClum, a parser. oh. and an array of 256 of entries that i could move to mp3buffer easily.
23:19:10amiconnHCl: Go ahead and implement setjmp/longjmp if you want a quick way out of plugins (and other routines)
23:19:10HCllet me delete that one :3
23:23:06HCl-rwxr-xr-x 1 root root 2808 Apr 27 23:20 searchengine.rock
23:23:08HClbetter? XD
23:23:22HCli don't know how much it stripped though o.o;
23:23:38preglowyou stripped a rock??
23:23:38HCli don't understand how it could cram all that in only 2808 bytes o.o;
23:23:41HClno
23:23:42HClo.o
23:23:47HCli just changed my static array to a pointer
23:23:52preglowahh
23:25:17HClhow can i pluck a rock apart again?
23:25:19HCljust objdump?
23:25:27preglowhmm
23:25:31preglowa rock is a flat binary image
23:25:47preglowit's better to objdump the .o file
23:25:58HClbut i know the .o's are okay..
23:25:59HCli think.
23:26:22HCl-rwxr-xr-x 1 root root 39234 Apr 27 23:20 searchengine.elf
23:26:28HClhow can it go from that to only 2808 ?
23:26:41preglowthere's a lot of cruft
23:26:45HClmrf...
23:26:53HClwell, i guess i'll go test it
23:26:56preglowand bss might be removed
23:27:10HCl(all it does is say "hi! i'm searchengine v0.1 o.o" "bye."
23:27:13HCl:P
23:27:23preglowbut what's the humongous table for?
23:27:34HClmeh, it was my tokenstream
23:27:42HClits better non static anyways
23:28:10HCli don't have a program to create a good tokenstream yet
23:28:17HClthe first version will probably have a search query hardcoded
23:28:45HClbut yea
23:28:59HCli had 256*264 bytes in that array
23:29:03HClso i guess thats about right :P
23:29:26HClin fact
23:29:29HClits exactly right :P
23:29:29HCl23:26 <+cu2q> 256*264 = 67584
23:29:29HCl23:26 < HCl> !calc ans+2808
23:29:29DBUGEnqueued KICK HCl
23:29:29HCl23:26 <+cu2q> answer +2808 = 70392
23:30:06HClwell, almost exactly o.o.
23:30:07CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:30:07*preglow prefers his ti89 to irc bots
23:30:58 Join amiconn_ [0] (~jens@pD9E7FF7E.dip.t-dialin.net)
23:31:14*HCl goes to fetch his iriver
23:31:16 Quit amiconn (Nick collision from services.)
23:31:16 Nick amiconn_ is now known as amiconn (~jens@pD9E7FF7E.dip.t-dialin.net)
23:33:38 Quit lostlogic ("Client exiting")
23:39:21 Part T0mas_
23:39:41 Join ashridah [0] (ashridah@220-253-121-225.VIC.netspace.net.au)
23:40:57 Quit Harpy (Read error: 60 (Operation timed out))
23:41:35 Quit gromit` (Read error: 104 (Connection reset by peer))
23:41:52HClhow much should i keep account of limited ram on archos?
23:42:12HCli'm tempted to dedicate quite some memory to lower disk accesses and speed up searches
23:42:35HCli guess 8mb archos recorders should be okay, but i don't know about other archos players.
23:42:57*HCl prods amiconn
23:43:54preglowthey have at least 2 megs of mp3 buffer, i think
23:43:59HClyea...
23:44:05amiconnAll SH1 based archoses have 2 MB of RAM; 8 MB is a (quite involved) hardware mod
23:44:08preglowbut hell, will you need more than that?
23:44:10HCli dunno about other people, but my tagdatabase is 5.8mb...
23:44:12Bagder1.7MB anyway
23:44:17preglowyikes
23:44:20HClmostly due to having an 715 songs album
23:44:26HClthat causes all other albums to be that size
23:44:32HClbut ofcourse, i don't care about the songs an album has
23:44:34preglowelegant
23:44:42HClso i guess it might be much less.
23:44:55HClshall i just dedicate ram to it
23:44:58HCland we'll see how it goes.
23:45:06amiconnThe "MB are total ram, so the mp3 buffer is around 1.7 MB depending on the other settings and whether running from RAM or flash
23:45:08HCli wrote it flexible enough to load it from disk instead
23:45:12HClamiconn: yea.
23:45:49preglowBagder: you wouldn't happen to know what 'dual tone' means, do you? i'm talking about dell lcds here
23:45:59preglowBagder: is that their gray black colour scheme?
23:46:45Bagderpreglow: I'm not sure, but that's the one with the greyish plastics I think
23:47:09BagderI'll check
23:47:19preglowwell, the other one is called midnight grey...
23:47:49Bagderthe midnight is blackish
23:47:59Bagderand tco99 while the light grey is tco03
23:48:17Bagderwhatever that means in real life;-)
23:48:40preglowhaha
23:48:44preglowtheir site isn't very informative
23:49:46BagderI have the "dual tone"
23:50:31preglowyeah, i'll go with that as well
23:52:42amiconnhmpf
23:52:51 Join gromit` [0] (~gromit`@ras75-5-82-234-244-69.fbx.proxad.net)
23:53:22amiconnI hate it when 'standards compliant' products aren't really compliant :(
23:53:55HCllag x.x
23:54:40amiconnI had a nasty bug with my mmc driver bugging on write. Turns out the card needs longer timeouts that it tells in their csd register...

Previous day | Next day