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).

#rockbox log for 2002-09-10

00:00:24Zagoryup
00:01:36_Dino_The executable has all your names in it now, you vain people!
00:02:07Zagorhehe
00:02:33_Dino_I want them crunched! I want them in some kind of data overlay!
00:03:12_Dino_hmmm, code overlays can we do anything there me wonders...
00:03:15Zagoryeah. you'll save 200 bytes and add 400 bytes worth of code. good idea :-)
00:03:47_Dino_oh well :-)
00:04:08_Dino_I can get you a decruncher in < 400 bytes!
00:04:22_Dino_200 opcodes...
00:04:34Zagor200 opcodes is 400 bytes...
00:04:47_Dino_yeah, I know
00:05:18adi|homelol
00:05:28_Dino_just me thinking out loud... sorry
00:09:21 Quit _Dino_ ()
00:11:32 Join langhaarrocker [0] (~Philipp@Bdbe1.pppool.de)
00:15:05langhaarrockerelinenbe? awake?
00:17:28langhaarrockerHow do I debug things like I04:IllInstr at 0000200?
00:18:16Zagordebugging is hard without serial cable
00:20:49langhaarrockerFor that I have to connect the tx pin of the cpu somewhere, don't I?
00:22:06hardeepyou can look at the archos.map file to find out what function is at address 0x200
00:22:16hardeepand then put in debug messages
00:23:11hardeepactually, there are no functions at 0x200 are there...
00:23:46langhaarrockerThat's probably why it's illegal? ;)
00:24:15Zagorbingo :-)
00:25:53langhaarrockerThat archos.map file looks scary to someone who doesn't really want to learn something new ...
00:26:03Zagorhehe
00:28:38***Saving seen data "./dancer.seen"
00:28:39elinenbeI am here, but I am leaving in 5 min.
00:28:59elinenbelanghaarrocker: it crashes every few songs.
00:29:08langhaarrockerYes. I know.
00:30:16langhaarrockerOr do you mean the same as you wrote to the list? That it stops playing? _That_ bug is fixed. But the real crashes - I have no real clue.
00:30:58langhaarrockerProbably I'm writing somewhere next to the lcd buffer.
00:33:18EnnaN:) sounds zweet :)
00:33:49langhaarrockerThe lcd buffer sounds zweet?
00:33:59Zagorlanghaarrocker: did you ever post the cleaned-up wormlet patch? the one which works with gcc 3.0
00:34:09langhaarrockeryes
00:34:16langhaarrockeroh gcc?
00:34:33Zagorthe variable declaration thing
00:34:41langhaarrockeryes
00:35:22langhaarrockerhttp://rockbox.haxx.se/mail/archive/rockbox-archive-2002-09/0326.shtml
00:36:19Zagorbut there was no file attached to that mail...
00:37:01 Join gordo [0] (~gordo@adsl-34-191-86.bct.bellsouth.net)
00:37:26Zagorlanghaarrocker: you can send it to me directly if you want
00:37:31Zagordcc or mail
00:37:48*Zagor is going through his patch backlog
00:37:55langhaarrockerI prefer mail since I don' t even know what dcc is...
00:38:19Zagorhehe, ok. dcc is direct client connect, an irc feature
00:38:30Zagorsend to bjorn@haxx.se
00:41:30adi|homeZagor i would have thought your saying 'No, Im not accepting that patch'
00:41:34adi|homewould have been good enough
00:41:37adi|homebut apparently not
00:41:39*adi|home sighs
00:41:44Zagoradi|home: which patch?
00:41:59adi|homethe whole wrapping patch
00:42:20adi|home"I still dont see why peopel dont want the cursor to go to the top. So what will the cursor do? I must be missing something."
00:42:22Zagorah, right
00:42:54Zagorevidence of the confusion such behaviour would case
00:44:35hardeepThe archos firmware doesn't go back to the top when wrapping... that's why people may be requesting this
00:44:41Zagorah
00:44:54langhaarrockerZagor: after a short but violent discussion with my outlook I managed to force it send the wormlet diff
00:45:03Zagorhehe
00:45:35langhaarrockerHope it contains only worms and no bugs :)
00:47:10Zagorlanghaarrocker: the start screen doesn't work with fixed fonts
00:47:34Zagorsince that's the default build, it should
00:48:22langhaarrocker?!? It did on my Recorder
00:48:43Zagoraren't you using propfonts?
00:48:59langhaarrockerTried both.
00:49:18langhaarrockerWhen using propfonts the field of worms is a little bit bigger.
00:50:16Zagorthe start screen doesn't show the l/r up/dn and F1 texts completely with fixed fonts
00:50:40Zagoralso, it's *very* confusing with multiple worms. who am I controlling? maybe it should blink a few times first or something.
00:50:40langhaarrockerStrange. I'm investigating.
00:50:48gordoso
00:50:51 Nick gordo is now known as Jet8810 (~gordo@adsl-34-191-86.bct.bellsouth.net)
00:50:56Jet8810what is the status on fat write anyway?
00:51:06ZagorJet8810: in progress
00:51:22Jet8810cool
00:51:38Jet8810After that should be a cinch to sort by artist and albu based upon ID3 right?
00:51:46 Join Dexter [0] (~Dexter@Mix-Pointe-a-Pitre-105-4-17.abo.wanadoo.fr)
00:51:57langhaarrockerZagor: That reminds me that I must have forgotten to send you the manual for wormlet...
00:52:11ZagorJet8810: not really. it will still be very slow extracting id3 info
00:52:20Zagorlanghaarrocker: no, i got it. i just didn't read it :-)
00:52:22Jet8810hmm
00:52:23Jet8810cache?
00:52:29 Part Dexter
00:52:42Jet8810couldnt you keep a cache on the hard drive?
00:52:52ZagorJet8810: the plan is to have a pc program build a database of id3 info. no fat write needed for this.
00:53:02Jet8810hm
00:53:25langhaarrockerZagor: that's what manuals are for: not reading them.
00:53:25langhaarrockerThe problem with blinking worms is: which one should blink when in multiplayer mode?
00:53:26Jet8810after this pc program builds database, how does archos deal with it?
00:53:47Zagorlanghaarrocker: right
00:54:03ZagorJet8810: read, search, display
00:54:27Jet8810great
00:54:36Jet8810can archos handle it in a reasonable amount of time though?
00:54:39Jet8810(say 5-10 seconds?)
00:54:50Zagordepends on the database design. this is not done yet.
00:54:55Jet8810right
00:55:03Jet8810that would be the ULTIMATE feature for archos
00:55:19Jet8810I would probably have to go in to about 3 gigs of mp3s and manually put in id3 info though :(
00:55:48Jet8810dinner
00:55:52 Nick Jet8810 is now known as Jet8810_afk (~gordo@adsl-34-191-86.bct.bellsouth.net)
00:56:38langhaarrockerapropos ultimate feature and filewriting and so on:
00:56:38langhaarrockerHas anybody considered how to split files without having to rewrite mass data? Like copying a sector and relink?
00:57:21Zagorwe have considered, but then looked the other way :-)
00:57:41Zagorthe question is: do you want to risk the file while splitting it?
00:58:11langhaarrockeryes
00:58:23langhaarrocker;)
00:59:44langhaarrockerDo you fear inconsistent FAT because of drowned batteries, hard drive shock or something like that?
00:59:59Zagorexactly
01:00
01:00:21langhaarrockerBut where's the difference to normal writing - eg while recording?
01:01:53Zagorthe difference is you have intact data, then risk it. with recording, you don't have it intact
01:03:01Zagorbut we could try it
01:03:54langhaarrockerDoes that really matter when changing the fat? I mean isn't it possible you accidently corrupt data of old files while failing to write the fat for the newly recorded?
01:11:07langhaarrockerStrange: I recompiled wormlet with fix fonts and now the problems are as you reported.
01:11:41 Nick Jet8810_afk is now known as Jet8810 (~gordo@adsl-34-191-86.bct.bellsouth.net)
01:12:36langhaarrockerArgh! I see the problem: I used lcd_puts which is affected by the scroll bar of the menu!
01:12:58langhaarrockerWhen you turn off the scroll bar everything looks fine.
01:15:38Zagorah. use lcd_setmargin()
01:15:47Zagori'll fix it
01:15:57langhaarrockerthx
01:19:51langhaarrockerIt's late. I must sleep. CU
01:19:59Zagorbye
01:20:12 Quit langhaarrocker ("Trillian (http://www.ceruleanstudios.com)")
01:22:00EnnaNlatorz /me to sleep
01:22:05 Quit EnnaN ("Leaving")
01:23:36PsycoXulanybody look into the problem i reported last night [or this morning for you guys i guess?] about playback stopping and WPS going blank?
01:23:47Zagornot me
01:24:05Zagordid we get a way to repeat it?
01:24:16PsycoXulit seems to happen at random, while playing in dirs .. i haven't tried with playlists
01:24:50Zagornever happened to me
01:25:08PsycoXulit's only happened in the last couple days
01:25:32Zagorodd
01:26:08pimlottcthe MIN macro is defined 4 times
01:26:29Zagorhehe
01:27:19pimlottcfirmware/system.h firmware/malloc/dmalloc.c firmware/test/i2c/main.c uisimulator/x11/uibasic.c
01:27:34pimlottcoddly MAX is only defined twice :)
01:27:38 Quit Jet8810 (Read error: 104 (Connection reset by peer))
01:31:30 Join Jet8810 [0] (~gordo@adsl-34-191-86.bct.bellsouth.net)
01:42:08 Quit Jet8810 (Read error: 104 (Connection reset by peer))
01:42:49 Join Jet8810 [0] (~gordo@adsl-34-191-86.bct.bellsouth.net)
01:52:31 Quit TotMach3r ()
02:00
02:15:14Zagorbed time. bye
02:15:15 Quit Zagor ("Client Exiting")
02:28:39***Saving seen data "./dancer.seen"
02:53:21 Quit Jet8810 (Read error: 54 (Connection reset by peer))
03:00
03:50:38 Quit hardeep ("Leaving")
03:56:58 Join hardeep [0] (hardeeps@sdf.lonestar.org)
04:00
04:20:10 Join mecraw [0] (~mecraw@67.41.113.92)
04:28:41***Saving seen data "./dancer.seen"
04:41:17 Join DJ4WWHR [0] (DJ4WWHR@ip009-56.pft.WKU.EDU)
04:41:42DJ4WWHRhey what is a popular channel where i would find a lot of people
04:43:30DJ4WWHRhello
04:43:30DJ4WWHRanyone there
04:43:30DJ4WWHRhola como estas
04:43:30DBUGEnqueued KICK DJ4WWHR
04:43:30DJ4WWHRyo yo
04:43:37DJ4WWHRsup dawgs
04:43:49DJ4WWHRwhat language do you all speak
04:43:53DJ4WWHRi know ur there
04:44:09 Part DJ4WWHR
04:44:55hardeeplaugh
04:45:09pimlottcthey always leave too soon
04:45:41mecrawif we're lucky, he might come back
05:00
05:05:08adi|homehehehe
05:13:11 Quit mecraw ("Trillian (http://www.ceruleanstudios.com)")
05:16:16 Join mecraw [0] (~mecraw@67.41.113.92)
05:18:28elinenbemecraw: are you working on repeat on/off?
05:18:42adi|homeme?
05:18:46adi|homeoh.. heheh
05:18:51adi|homemecraw sorry :)
05:19:15mecrawi just poked around with little success
05:19:42mecrawqualsaruk messaged me about it
05:20:28mecrawadi|home: you don't have to be sorry for wanting to be me :)
05:21:04mecrawelinenbe: have you looked at it?
05:21:09adi|homeaw shucks
05:21:17elinenbemecraw: nopes
05:23:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
05:23:10*mecraw is too busy rooting for the Pats to work on it
05:39:12 Nick _seb_ is now known as seb-sleep (user@bgp420584bgs.union01.nj.comcast.net)
05:57:06 Quit hardeep ("Leaving")
06:00
06:25:20 Quit edx|sleep ()
06:28:44***Saving seen data "./dancer.seen"
06:29:52 Join hardeep [0] (hardeeps@sdf.lonestar.org)
06:32:01 Join merwin [0] (~merwin@12.242.185.10)
06:32:06merwinyo
06:32:20pimlottcyo
06:32:22mecraw12yo
06:32:32*merwin just saw austin powers
06:32:33merwin3
06:32:34merwin:)
06:36:58pimlottcI think there's a bug in the linux usb driver
06:36:59pimlottc:/
06:38:03merwinthat sucks
06:38:10merwinI think PsycoXul would agree though
06:38:16mecraw12get windows then :D
06:38:30pimlottcbeh
06:38:50pimlottcI prefer to program under linux
06:39:21merwinI prefer cygwin
06:42:29PsycoXulyes there's something wrong with using the archos in linux.. it locks for no apparent reason sometimes and forces you to reboot to use it again
06:43:05pimlottcwell, it's not locked, but it won't mount anymore
06:43:09pimlottcsecond time this has happened
06:46:54 Quit merwin ()
06:47:53adi|homepimlottc: ive gottne that before too..
06:51:40pimlottcworked for dozens of times before that
06:51:51PsycoXulhmm
06:57:33 Join merwin [0] (~merwin@12.242.185.10)
06:57:44merwinwindows media player 9 beta is available from microsoft
06:57:47merwinneato
06:58:00merwinsupports mpeg4
06:58:27pimlottcwhat happened to 8
06:58:36merwin*shrug*
06:58:46pimlottc6 was good
06:59:06pimlottc7 they messed it up again with all the skins and silly things
06:59:26merwinyeah... I wanna see how 9 is
06:59:44pimlottcwhich was sad since it took so long to get to 6
07:00
07:00:39pimlottcremember the old media player... it was basically the 3.11 midi player with video and cd audio tacked on
07:01:13merwinI liked it
07:01:15merwinit was nice
07:01:21pimlottcit sucked for video though
07:02:17merwinvideo in general sucked on a 386
07:02:39pimlottcthe original win95 media player was basically the same thing
07:02:45merwinyeah
07:02:46merwinit was
07:03:25merwin150KB/s downloading office 2000 (own it but no cd, therefore kazaa :-)
07:09:10 Join LinusN [0] (~linus@labb.contactor.se)
07:09:19LinusNmorning
07:09:34mecraw12evening
07:09:51LinusN:-)
07:11:46mecraw12LinusN: has it been decided how to do the queueing of songs while already playing?
07:13:08LinusNwell, i don't know about decided, but i know how i would do it
07:13:10merwinmorning linus
07:13:24LinusNho merwin
07:13:34merwinho ender :P
07:13:41*merwin is an orson scott card fan
07:14:37mecraw12LinusN: does your way include appending to the current list?
07:14:46LinusNno
07:14:55mecraw12:(
07:14:59LinusNnot at the moment
07:15:22*merwin is staying out of this particular arguement
07:15:54LinusNmecraw12: do you mean "permanently" adding to the playlist?
07:16:01LinusNi mean for every repeat
07:16:02mecraw12i'm just thinking down the road when we can write playlists to disk, so we can create them on the fly
07:16:18LinusNof course we will be able to in the future
07:16:47mecraw12phew :)
07:17:28LinusNthis is just like the Archos queue function
07:17:39LinusNexcept it will work :-)
07:17:51mecraw12that would be a plus
07:22:41 Quit pimlottc ("reboot then zzzz")
07:32:37 Quit merwin ()
08:00
08:09:38LinusNhehe, i just realized that the queue function won't be perfect
08:09:58LinusNhere's why:
08:10:32LinusNthe mpeg thread preloads the next track before the current has ended, to avoid pauses between songs
08:11:01LinusNif a file is added to the queue after that, it won't be played until after the preloaded song
08:11:36LinusNor the user would have to expect a pause to reload the chosen song
08:13:30hardeepthe pause would only occur if you added to the queue in the last few seconds (dependant on bitrate) of the current playing song...
08:13:38LinusNyes
08:13:50 Join Bagder [0] (~daniel@as3-3-2.ras.s.bonet.se)
08:14:09LinusNthe problem is that the mpeg thread doesn't really have a "reload" function
08:14:18LinusNperhaps mpeg_play() will do it...
08:14:19hardeepyeah, it would have to be added
08:14:33hardeeplike a mpeg_flush_preloaded_tracks() or something
08:15:12LinusNwhat would the user dislike the most: the pause or having to wait for the queued track to play?
08:15:25LinusNi guess the uncertainty is worst
08:15:27hardeephaving to wait... :) at least if the user is me =)
08:16:04LinusNwe can't expect the user to understand why the queued track doesn't play when he thought it would do
08:16:45hardeepyeah... this is similar to the open bug when you switch shuffle modes
08:16:59LinusNexactly
08:17:13LinusNthe flush() function would probably solve that too
08:17:22hardeepyep
08:17:35LinusNhardeep: care to implement the flush()?
08:17:55hardeepsure, it shouldn't be that hard
08:18:24LinusNprobably accompanied by a mpeg_num_preloaded_tracks() function?
08:18:43LinusNmaybe not
08:19:13hardeepwe already have that...num_tracks_in_memory()-1
08:19:18LinusNok
08:20:38LinusNwhat would we like to happen when we press Prev?
08:20:53LinusNi suggest that the queue is to be cleared
08:21:46mecraw12it seems to me it should behave like it was actually inserted into the playlist
08:22:30BagderRED
08:22:36LinusNi see it as a function that lets you play tracks "outside" the playlist
08:22:54LinusNBagder: wow
08:23:05Bagdermucho
08:23:10 Join merwin [0] (~merwin@12.242.185.10)
08:23:40hardeephey, rec worked... who cares about everything else... =)
08:24:32mecraw12LinusN: you're only affecting how it will play this time through
08:25:19LinusNmecraw12: show me the code! :-)
08:25:37LinusNseriously, that is way harder
08:25:53mecraw12i know... i'm just expressing my wishes :)
08:26:27mecraw12besides, i don't think you'd want to see my code
08:26:47LinusNi think users would like to be able to do both
08:27:19LinusNadd en extra track to be played once, or add it to the playlist
08:27:37Bagderand what if the playlist is shuffled? ;-]
08:27:51LinusNbut if the user wants to add it to the playlist file too, then they'll have to show me the code
08:27:57mecraw12what about switching the array over to a linked list?
08:28:14Bagderwould take heaps of more memory
08:28:15LinusNmecraw12: sure, but the list size will double
08:28:35LinusNand the scrambling will take longer
08:28:41LinusNnot to mention the sotring
08:28:44LinusNsorting
08:28:45***Saving seen data "./dancer.seen"
08:29:04LinusNit would need to be dolble-linked
08:29:11LinusNyuck!
08:29:55mecraw12how long would it take to push every entry down a sport?
08:30:03mecraw12s/sport/spot
08:30:15LinusNnot that long
08:30:58LinusNit might be feasible
08:31:38LinusNbut would the user really want the Queue function to insert the song into the playlist?
08:31:47LinusNi'm not usre
08:31:49LinusNsure
08:32:01mecraw12like you said, they should be able to do either
08:32:07mecraw12so... show me the code!
08:32:33LinusNi think insert-into-playlist is a very different thing than the queue function
08:32:47LinusNand i'm working on the queue now
08:35:56hardeephmmmm, playlist handling needs to be changed slighly in mpeg thread... playlist_next() should only be called when the song starts playing else the playlist index will be wrong after flush
08:36:29LinusNhow does it work now?
08:36:40adi|homewhich makes more sense
08:36:41hardeepit calls it as soon as new song is loaded
08:36:53adi|homeerasing an entire line of text and redrawing it to delete one char
08:36:58adi|homeor just deleting the one char?
08:36:59LinusNhardeep: ouch!
08:37:19LinusNadi|home: i suggest the simpler code
08:37:30adi|homeokay.. cool.. line deletion it is ;)
08:40:19LinusNblah, the queue function is quite tricky to implement...
08:41:25rwoodI got dem ol' code police got me blues - i thought the speed improvement was fairly significent
08:41:39LinusNrwood: i agree
08:41:52rwoodLinusN: it will be some time next week or the week after before i get the i2c stuff for the player
08:42:10LinusNrwood: dont worry, i am in no hurry
08:42:32rwoodLinusN: we are moving to a new building this weekend and production is completely packed up
08:42:48merwinrwood: heh... in the speed of things, an extra 5ms doesn't really make too much of a difference in loading rolo
08:42:48LinusNbtw, i got the recorder logs
08:43:10rwooddid the software work ok under NT?
08:43:15LinusNyup
08:44:17merwinLinusN: make a mental note not to EVER use any NTFS -> FAT32 conversion software :-) I lost all of my LFN's doing it.
08:44:36merwinI'm just finishing up reinstalling everything
08:44:37merwinyuck
08:44:44rwoodmerwin: 5ms? - on mine it looked more like a second or two
08:44:54adi|homehow expensive is calling lcd_update?
08:45:01merwinrwood: really? it only takes about 1 second in the first place for me
08:45:05LinusNadi|home: takes about 25ms
08:45:20adi|homeso avoid extra calls to it if possible?
08:45:25merwinadi|home: yup
08:45:25LinusNi would
08:45:27adi|homek
08:45:44Bagderadi|home: also, try using lcd_update_rect() if possible
08:45:54adi|homeahh okay..
08:46:17merwinWindows Media Player 9 is a letdown :(
08:46:22merwinit's the same old crap
08:46:28LinusNmerwin: what did you expect?
08:46:34merwinnow, Winamp3 on the other hand... that's a piece of work
08:46:54merwinLinusN: I was expecting a little more than that with all the hype :-)
08:47:05LinusNi haven't seen any hype :-)
08:47:24merwinmicrosoft hype...
08:47:27merwinarticles, etc
08:48:04LinusNoh, that stuff
08:48:07LinusNi don't read that
08:48:11merwinnothing huge :)
08:48:34rwoodeven less mp3 support?
08:49:07merwinI don't think they could get much worse
08:50:47 Quit Synthe ("Connection Lost - Excess Blood")
08:52:38 Join Synthe [0] (Synthe@galt.synthe.net)
08:52:46adi|homeodd.. im getting undefined ref to lcd_update_rect
08:53:46adi|homeahhh.. its not defined for the sim...
08:53:48adi|homegrrrrr...
08:53:52Bagderoh
08:54:00*adi|home slaps bagder
08:54:11adi|home:)
08:54:16*Bagder ducks
08:54:52adi|homeand the onlything using it is status.c
08:55:00Bagderyes
08:55:00adi|homeand the keyboard if i have it in there...
08:55:08Bagderbut it is relatively new
08:55:13adi|homenods
08:55:18Bagderit should be used much more
08:55:26adi|homeany chance you could hack a copy into the lcd-x11.c?
08:55:33Bagderyes, me fix
08:55:37adi|homety :)
09:00
09:03:25adi|home> [RockBox - the God of all things small, with blue bumpers]
09:03:28adi|homei love that sig
09:03:52merwinhahaha
09:04:29*merwin has sh1-cygwin up and running again
09:04:30merwinwoo
09:07:19Bagderadi|home: I'm adding it now without testing ;-)
09:07:31Bagderwhat can possibly go wrong? B-P
09:07:35adi|homethats fine.. ill test ;)
09:07:56adi|homelet me know when its in
09:08:19Bagderjust committed
09:08:30 Join langhaarrocker [0] (~Philipp@dsl-2129324-055.ip.versatel.de)
09:09:45BagderI'll fake something up for the win32 sim too
09:09:56adi|homecompileing right now
09:10:01merwinBagder: you'd better :-)
09:16:51adi|homeseems to work fine bagder.. but im going to double check
09:17:00Bagdergoodie
09:17:06 Join bobTHC [0] (~bobTHC@AMarseille-206-2-1-9.abo.wanadoo.fr)
09:17:14bobTHChi all!
09:20:25LinusNhi
09:22:10bobTHCu know how many vote for the tshirt contest
09:22:11bobTHC?
09:22:18Bagder55 people
09:22:57 Quit merwin ()
09:23:07bobTHCcool... and it's always the classical 6 who win ?
09:23:35Bagderwell, 24 and 23 are closing in
09:23:54LinusNone of the voters didn't want the Archos name on the shirt
09:24:02Bagderwhich makes me hesitate, as they are both very similar
09:24:10LinusNhe called it "unnecessary free advertising"
09:24:18bobTHC;)
09:24:35LinusNbut if we only wrote "Jukebox", which one would we mean?
09:24:52BagderAr**os ? ;-)
09:24:57bobTHCbut Rockbox IS archos advertising !
09:25:23bobTHCAR??OS
09:30:31bobTHCwho try the ozilo patch?
09:32:14Hes24 and 23 are really cool IMO, they're very similar so probably the votes got distributed between them... if only one of them had been there it would have had lots more votes probably.
09:33:05BagderHes: yes, that's kind of my thinking too
09:33:44bobTHCno back design ?
09:34:04 Join quelsaruk [0] (~zaknafein@faerun.ugr.es)
09:34:09quelsarukhi
09:34:19LinusNyo
09:34:27Bagderthe contest rules did specificly mention single-side design
09:46:22 Nick quelsaruk is now known as quel|out (~zaknafein@faerun.ugr.es)
10:00
10:00:54 Quit rwood ()
10:01:42langhaarrockerWith some designs I see the problem that they cover too much of the t-shirt. When wearing them half of the design will be hidden in the trousers
10:04:52LinusNwe will probably shrink some of them to fit both the shirt and our wallets :-)
10:06:17bobTHCfor the 5 for exemple it's possible to reduce the design size easyly..
10:06:52Bagderwell, we'll sort out a winner first, then we'll see what kind of "tweaks" we'll need to do
10:07:04langhaarrockernew design?
10:07:04hardeepanyone have 3-4 mp3's where all fit in memory at same time... i'm too lazy to generate some
10:07:43langhaarrockerNo. Sorry our songs tend to be 4-5 minutes...
10:08:53BagderI think Linus has some farbror frej ones :-)
10:09:09bobTHCi can't listen a 247Mb 192kbps 44Hz CBR file... with Rockobox 1.3....
10:09:27BagderbobTHC: what happens?
10:10:13bobTHCthe hd begin loading and freeze on the loading window with HD led on .....
10:10:48adi|homesoooo
10:11:12langhaarrockeryou don't accidently have the osci demo turned on?
10:11:15langhaarrocker;)
10:11:39bobTHCimpossible to browse, to listen... the only way to change track is to stop the device.
10:12:05adi|homeokay.. so who wants to test the new keyboard app ;)
10:12:09bobTHCand tto restart it without resume
10:12:10*adi|home pokes Bagder
10:12:12adi|homeand linus
10:13:08bobTHCon the 1.3 version witout any patch
10:13:42adi|homebagder so who are the top 3 tshirts?
10:14:01Bagder6, 24 and 23
10:14:06adi|homenice
10:14:14adi|homenext question
10:14:20adi|homewho wants to test the keyboard?
10:14:20langhaarrockeradj|home: I prefere topless t-shirts..
10:16:31LinusNhardeep: i have some
10:16:56hardeepLinusN: thanks, I found some :)
10:17:42LinusNbobTHC: try the latest build
10:18:20langhaarrockerDo you think it would be a good idea to enhance the lcd api with optimized functions to horizontal / vertical lines? The standard line algorithm is a bit overkill for that.
10:18:39bobTHClinus : i'll test it this night...
10:19:02LinusNlanghaarrocker: do it
10:19:12langhaarrockerok, tonight.
10:19:29bobTHCASM is requiered?
10:19:46LinusNno
10:20:01hardeepLinusN: okay, i've got flush and reload code that appears to be working... tested with playlist shuffle and everything's working
10:20:23LinusNhardeep: cool! did you have to change much?
10:21:19hardeepLinusN: just the playlist handling in mpeg thread so that playlist_next is called when track changes... everything else was pretty simple. I'll send you a diff so you can look over it
10:24:44adi|homewell.. fine.. the hell with all of you...
10:24:46adi|homeim going to bed ;)
10:24:48*adi|home smirks
10:24:51 Join Zagor [0] (bjst@as9-5-6.k.s.bonet.se)
10:24:56adi|homeZAG
10:24:58adi|homeperfect timing
10:24:59Zagoryo
10:25:04adi|homekeyboard ver1 is done
10:25:08adi|homeready to be beta'd
10:25:11Zagornice
10:25:12hardeepsalut Zagor!
10:25:41adi|homeyou want me to add it to apps/recorder and you can play with it at will?
10:26:37 Join TotMacher [0] (tot@p508DA36C.dip.t-dialin.net)
10:27:13Zagoradi|home: wait a bit. we'll need it for both player and recorder. i'm not sure we should use one or two files.
10:28:00adi|homenods
10:28:07adi|homewell right now it is coded only for player
10:28:10adi|homeas proof of concept
10:28:23adi|homerecorder rather
10:28:25adi|homesorry.. tired..
10:28:47***Saving seen data "./dancer.seen"
10:29:35Zagoreverything in apps/ is linked into the daily builds, so it's not a good place for test code
10:29:42adi|homeokay...
10:29:52Bagderunless #ifdef'ed of course
10:30:02Zagorpost it on the list, then we'll have it handy when we need it later.
10:30:07adi|homek...
10:32:01hardeepLinusN: okay, how should I send you diff?
10:33:07adi|homesent Zagor...
10:33:09adi|homebedtime
10:33:14Bagdernight adi
10:33:28langhaarrocker don't let the bugs bite...
10:38:32hardeepLinusN: it's at http://hardeeps.freeshell.org/flush_reload.patch
10:39:36Zagorhardeep: what is the purpose of this?
10:40:28hardeepflush preloaded tracks if playlist changes... or new songs are queued
10:41:56Zagorah, right
10:42:57LinusNlooks good
10:43:53LinusNZagor: any objections?
10:44:00Zagornope. just curious.
10:44:17LinusNhardeep: go commit!
10:44:56hardeepsir yes sir! =)
10:45:06bobTHCqueue function work ?
10:45:30LinusNbobTHC: no, not yet
10:46:09LinusNi discoovered that it was quite tricky to do
10:46:30bobTHCok it's just optimization for playlist (and in the futur for queue...)
10:46:50bobTHCnear futur for linus ;)
10:54:58 Join Hawk [0] (~bmorley3@washdc3-ar3-4-35-205-062.washdc3.elnk.dsl.genuity.net)
10:55:04LinusNrolo is soooooo handy when using gdb
10:56:43 Quit Hawk (Client Quit)
11:00
11:00:04langhaarrockerTo use gdb you have to modify the JB, don't you?
11:00:22LinusNyes
11:03:15langhaarrockerfor the recorder you probably have to sacrifice the spdif jack?
11:03:37Hesor the line in
11:03:52Heswhichever you prefer
11:03:53langhaarrockerHas anybody done a switchable modification?
11:04:36LinusNi only know of only two people that have done this mod, I and Zagor
11:04:46LinusNand we haven't made it switchable
11:05:05hardeepoh hey, that reminds me... i tried digital out and it worked fine
11:05:19hardeepi saw a post on one of the message boards saying it didn't work with rockbox
11:06:10Hesit works for me
11:06:41langhaarrockerIs there a spare output port of the cpu? If so I think it should be possible to make a very small pcb board with a 4066 on it to switch between spdif / serial.
11:06:43LinusNyeah, several people have reported success
11:06:47Hesi use at home, playing through the quality DAC on my Edirol UA-5 A/D/A-box
11:07:39LinusNlanghaarrocker: there may be a spare port, but i dont know at the moment
11:08:02LinusNlanghaarrocker: i have searched for the SYNC signal, but it doesn't seem to be connected... :-(
11:09:09langhaarrockerBoohooohooo! No SYNC!
11:09:09langhaarrockerConcerning digital in out: Or do you mean that you can connect spdif / serial simultaniously and they don't interfere to much when only one is operating?
11:12:42LinusNthey will not work simultaneously
11:13:22langhaarrockerLinusN: any more bad news?
11:13:35LinusNi have lots of them :-)
11:13:49langhaarrockerOh. That's bad news.
11:13:54LinusN:-)
11:14:15LinusNi found an interesting "feature" in the MAS3587F
11:14:26LinusNyou know the files without VBR header?
11:14:43langhaarrockerYes, I produced some.
11:14:59LinusNif they are encoded with the MAS3587F, there is timing information embedded in the ancillary data
11:15:06ZagorHes: have you had time to test skipping?
11:15:13langhaarrockerI've read that, too
11:15:15HesNope
11:15:22Hesi've been at home, caught flu in the weekend
11:15:22hardeepyeah, i noticed that... was trying to figure out some way to make use of that info
11:15:26Hesso no driving around 8-(
11:15:27hardeepre: LinusN
11:15:37ZagorHes: ok
11:15:42HesI did install a new build yesterday though
11:16:06hardeepAlso, they say it's a configurable option... I wonder if the archos fw enables it
11:16:47LinusNhardeep: i don't know
11:17:06LinusNbut since they don't produce a TOC, i think they activate it
11:17:23LinusNi'd rather have a TOC
11:17:36hardeepheh, yeah
11:18:19hardeepactually, yeah, they probably enable it... I think ffw/rew works correctly with those files on the archos fw
11:19:20LinusNbut if you edit the files, the timecode will be wrong
11:20:48LinusNi wonder how it is implemented
11:21:07LinusNi mean, you aren't guaranteed to have room for ancillary data in each frame
11:21:52LinusNso the encoder is probably wasting space to guarantee at least 20 bits of ancillary data
11:21:54langhaarrockerI've split archos generated mp3s with mpDirectCut and the time info became weired.
11:22:57langhaarrockerLike telling the file was much longer and it was fun to see the seconds pass on the JB in double time.
11:23:05LinusNlanghaarrocker: can mp3directcut regenerate Xing headers?
11:23:14langhaarrockerprobably not.
11:23:28LinusNnot many programs can, for what i can see
11:24:02langhaarrockerDo you know some?
11:24:18LinusNi think having to read the ancillary data all the time when playing is a waste of time and resources
11:25:02LinusNlanghaarrocker: i have heard of a program called "vbrfix"
11:26:02hardeephmmm, why would they need to read it all the time? isn't the timecode only needed when you're seeking?
11:26:20langhaarrockerIll have a search.
11:26:21langhaarrockerIf it's a complete time stamp I think it's a useful feature for documenting purposes like "During the last practice we had this idea around 19:00. What was it?"
11:27:02LinusNit is relative from the start of the recording
11:27:37LinusNlanghaarrocker: it is a great idea!
11:27:55langhaarrockerWell, we could save some rtc info in the mp3 header.
11:28:13LinusNi think there is a ID3V2 tag that holds the absolute time
11:28:29LinusNwe should definitely create one when recording
11:29:20langhaarrockerCool: a 'goto' function with absolute time!
11:29:31LinusNlunchtime
11:29:37 Nick LinusN is now known as LinusN|lunch (~linus@labb.contactor.se)
11:37:41 Quit TotMacher ()
11:38:09 Quit hardeep ("Leaving")
11:52:53 Part elinenbe
11:53:01 Join elinenbe [0] (trilluser@user-0cces0l.cable.mindspring.com)
11:53:13 Nick elinenbe is now known as elinenbe|sleep (trilluser@user-0cces0l.cable.mindspring.com)
12:00
12:28:51***Saving seen data "./dancer.seen"
12:38:34 Join Blaster_Master [0] (~noone@firewall.scribona.com)
12:41:13 Join mecraw_ [0] (~mecraw@67.41.113.92)
12:47:47 Quit adi|home (Read error: 110 (Connection timed out))
12:49:44 Quit mecraw12 (Read error: 60 (Operation timed out))
12:52:41 Quit Blaster_Master ()
12:55:02 Nick LinusN|lunch is now known as LinusN (~linus@labb.contactor.se)
12:58:08quel|outlinus you have lunch when i have breakfast :)
12:58:41quel|outreally different ways of life
12:59:15 Quit mecraw (Read error: 110 (Connection timed out))
13:00
13:07:54 Join Phil [0] (~Philipp@dsl-2129324-055.ip.versatel.de)
13:08:19PhilAnd I just had some chicken at lunchtime which acutally was my breakfast.
13:08:22 Join b0bTHC [0] (~bobTHC@AMarseille-206-2-1-9.abo.wanadoo.fr)
13:08:37 Quit quel|out (brunner.openprojects.net irc.openprojects.net)
13:08:37NSplitbrunner.openprojects.net irc.openprojects.net
13:08:37 Quit bobTHC (brunner.openprojects.net irc.openprojects.net)
13:08:37 Quit langhaarrocker (brunner.openprojects.net irc.openprojects.net)
13:08:37 Quit elinenbe|sleep (brunner.openprojects.net irc.openprojects.net)
13:08:37 Quit dwihno|gone (brunner.openprojects.net irc.openprojects.net)
13:08:37 Quit datazone (brunner.openprojects.net irc.openprojects.net)
13:08:37 Quit laotan (brunner.openprojects.net irc.openprojects.net)
13:08:54 Join alkorr [0] (alkorr@ASte-Genev-Bois-110-1-2-27.abo.wanadoo.fr)
13:09:10alkorrhi
13:09:28alkorrsomeone tries to investigate with MMJB firmware ?
13:09:40Zagoralkorr: no
13:09:52PhilHi. Who am I. Am I really Phil? I used to be langhaarrocker a few moments ago.
13:10:05Zagori looked at it, but only so much I saw they use another scrambler
13:10:21 Quit Phil (Client Quit)
13:10:23alkorrthe size of the file is set to offset 8 and in little endian format like PC
13:10:38alkorrso it is sure it isn't the same scrambler
13:11:07NHealbrunner.openprojects.net irc.openprojects.net
13:11:07NJoindwihno|gone [0] (dwihno@Bald067.Baldakinen.Umea.SE)
13:11:42alkorrthe header is 16 bytes according to is real size and the stored size
13:11:49alkorrits
13:12:54 Join Phil [0] (~Philipp@dsl-2129324-055.ip.versatel.de)
13:12:56alkorrthere is no visible text so there's chance this file is crypted...
13:13:05NJoinelinenbe|sleep [0] (trilluser@user-0cces0l.cable.mindspring.com)
13:13:05NJoinquel|out [0] (~zaknafein@faerun.ugr.es)
13:13:05NJoinbobTHC [0] (~bobTHC@AMarseille-206-2-1-9.abo.wanadoo.fr)
13:13:05NJoinlanghaarrocker [0] (~Philipp@dsl-2129324-055.ip.versatel.de)
13:13:05NJoindatazone [0] ([NyzPmBte4@207.136.36.203)
13:13:05NJoinlaotan [0] (~jesse@markham.openflows.org)
13:13:07 Quit langhaarrocker (Read error: 104 (Connection reset by peer))
13:13:08Zagoryup
13:14:07 Quit bobTHC (Read error: 104 (Connection reset by peer))
13:14:44 Nick Phil is now known as langhaarrocker (~Philipp@dsl-2129324-055.ip.versatel.de)
13:20:15 Join Schnueff [0] (~mah@wjpserver.cs.uni-sb.de)
13:20:31Schnueffmoin
13:22:07LinusNyo
13:23:33 Nick seb-sleep is now known as _seb_ (user@bgp420584bgs.union01.nj.comcast.net)
13:27:36alkorri'm wondering if they use a compression
13:29:24alkorrmaybe something like struct hdr { u32 signature; u32 usize; u32 csize; u32 crc };
13:30:18alkorrcsize is file size (281555) minus 16 (header size).
13:30:34Zagorlooks reasonable
13:32:14langhaarrockerHave we ever gotten any feedback from Archos wether they like their Jukeboxes being hacked?
13:32:29Zagorlanghaarrocker: nope. not a word.
13:32:53langhaarrockerHas anybody tried to provoke some kind of comment?
13:33:01LinusNwhy?
13:33:13langhaarrockerJust for fun.
13:34:28langhaarrockerMaybe they like the idea and would even provide us with info about the MMJB scrambling.
13:34:29LinusNi'm happy as long as we don't get any heat from Archos
13:36:38langhaarrockerUse the right power adapter ;)
13:38:21LinusN:-)
13:41:00 Join gizz [0] (gizz@lyon-1-a7-62-147-19-215.dial.proxad.net)
13:41:10gizzhi all
13:41:22LinusNyo
13:46:42 Quit Hes (brunner.openprojects.net irc.openprojects.net)
13:46:42NSplitbrunner.openprojects.net irc.openprojects.net
13:46:57 Quit Bagder (brunner.openprojects.net irc.openprojects.net)
13:46:57 Quit gizz (brunner.openprojects.net irc.openprojects.net)
13:46:57 Quit Zagor (brunner.openprojects.net irc.openprojects.net)
13:46:57 Quit fraggle (brunner.openprojects.net irc.openprojects.net)
13:47:03NHealbrunner.openprojects.net irc.openprojects.net
13:47:03NJoingizz [0] (gizz@lyon-1-a7-62-147-19-215.dial.proxad.net)
13:47:03NJoinZagor [0] (bjst@as9-5-6.k.s.bonet.se)
13:47:03NJoinfraggle [0] (~fraggle@pc2-glfd1-4-cust151.gfd.cable.ntl.com)
13:47:05NJoinHes [0] (~hessu@hessu.zedi.sonera.fi)
13:47:16NJoinBagder [0] (~daniel@as3-3-2.ras.s.bonet.se)
13:48:03gizzwhat happened ? massive disc ?
13:48:25LinusNyeah, probably a hub that went down at freenode
13:48:28 Join Snorlax [0] (lamamba_@h135n1fls34o883.telia.com)
13:48:36LinusNhappens all the time
13:49:11gizzahh
13:49:16gizzOK.
13:50:57gizzBye Linus (have to hangup :)
13:51:04LinusNbye
13:51:20 Quit gizz ()
14:00
14:09:00 Nick _seb_ is now known as seb-school (user@bgp420584bgs.union01.nj.comcast.net)
14:13:21quel|outis there any possibility to make AJR have usb 2.0?
14:13:41langhaarrockerno, that would need another usb chip
14:13:59quel|outok
14:14:18quel|outand none has made suck a mod yet, ..
14:14:20 Join pimlottc [0] (~pimlottc@a1-1b012.neo.rr.com)
14:15:22Zagornot realisticlly
14:16:32quel|outthat doesn't look quite optimist
14:16:49 Quit b0bTHC ("Trillian (http://www.ceruleanstudios.com)")
14:17:03alkorris there someone who has two different version of MMJB firmware ?
14:17:15alkorri need to compare them
14:18:13Zagorhttp://groups.yahoo.com/group/archosmultimedia/files/
14:18:17langhaarrockerCan't we just use rolo to load the mmjb firmware and have movies on the jb? ;)
14:18:26Zagorhaha
14:20:19langhaarrockerZagor: Seems you didn't find much to niggle about when code-policing wormlet?
14:21:09Zagornothing worth changing
14:21:18langhaarrocker:)
14:22:14Zagoryou should keep the lines below 80 columns
14:27:59alkorrwell they don't seem to use a partition scrambling
14:28:54***Saving seen data "./dancer.seen"
14:30:22LinusNBagder: can you perform a battery test with ATA_SLEEP vs ATA_STANDBY?
14:31:10 Join edx|sleep [0] (~edx@pD9EAB531.dip.t-dialin.net)
14:31:20 Nick edx|sleep is now known as edx (~edx@pD9EAB531.dip.t-dialin.net)
14:35:45alkorrdifferences starts at offset 64, 68, then from 6D to the end of file. Maybe they used a better encryption
14:38:10 Quit langhaarrocker (Read error: 54 (Connection reset by peer))
14:39:23LinusNi wonder how Archos Ondio looks inside
14:39:32ZagorOndio?
14:39:38LinusNhttp://www.archos.com/uk/archos_news.html
14:39:51LinusNand FM Recorder
14:40:14pimlottcnice 'backplate'
14:40:18LinusNJust when you thought Jukebox technology could not go any further, Archos get seriously retro with The Jukebox FM Recorder 20 and the Ondio
14:40:22LinusN:-)
14:40:27 Join langhaarrocker [0] (~Philipp@dsl-2129324-055.ip.versatel.de)
14:41:27Zagorugh, can you imagine the clean-up necessary to get reasonably good fm reception in this thing?
14:42:06Zagornice to see them do new stuff, though. looks like they plan to stay in business a while.
14:42:36LinusNi wonder how much different the FM recorder is
14:42:53LinusNi guess we would have to support that one... :-(
14:43:50Zagorwell, we don't have to do anything. it it's simple, we'll do it. otherwise, not.
14:46:18langhaarrockerWith the only difference that the person who does it gets NO pudding: He won't have the time to eat any.
14:47:31alkorrfinally they plan to add FM feature
14:48:37LinusNi'm surprised thay haven't done it earlier
14:48:50alkorrme too
14:49:03LinusNit's such an obvious feature (to me)
14:49:06alkorri'm sure they would be more attractive
14:51:11Zagornow we only await them publishing full specifications :-)
14:51:44LinusNi guess a familiar place will freeze over before that happens
14:52:28langhaarrockerWhat leads to the question: Is there a heaven for hackers?
14:52:48LinusNyeah, with C64's all over the place
14:53:15langhaarrockerHey, I've got 1 1/2 of them in my cupboard.
14:53:24alkorrwell just imagine they use a new encryption :(
14:54:05langhaarrockerHas Rockbox started before the MMJB?
14:54:29LinusNlanghaarrocker: we started before we knew about the MMJB
14:54:33alkorryes
14:54:42alkorrno i'm speaking about FM recorder
14:55:10langhaarrockerI just feared we might have risen the ARCHOS ENCRYPTION MONSTER.
14:55:16LinusNpossibly
14:55:26Zagorlanghaarrocker: that's very possible
14:55:38LinusNbut it isn't that much of a problem
14:55:38alkorrthey could decide to have a different firmware as recorder and player have different firmwares
14:58:20alkorrmaybe not. I don't think they would do it because the internal rom (not encrypted) need to unscramble the firmware in the external ROM. It would need to change all their SH1 because of the rom content
14:58:41LinusNexactly
14:59:08LinusNand even if they changed it, we could just monitor the memory accesses to see how the decryption is done
14:59:17Zagori think the FM recorder uses the same as the current recorder. but the multimedia may have harded protection.
15:00
15:01:29 Join xam [0] (~xam@tqpc.ucc.ie)
15:04:39alkorri'm really wondering if they are using a compression
15:04:55alkorrseveral reason to think so :
15:05:25alkorrwhen zipping their firmware, compression is unefficient
15:05:51alkorreach firmware seems to have both sizes
15:06:22alkorrthe size of the file and a bigger size (nearly twice) in their header
15:07:10edxare you talking about the archos multimedia player?
15:07:17LinusNyes
15:07:22alkorrthe difference between two firmware big sizes is nearly twice as big as the filesize too
15:07:30edxanyone of you got one?
15:07:42LinusNnot me
15:08:07langhaarrockerI don't need it either.
15:08:22LinusNhow can one not need a geeky toy?????
15:08:44alkorrgeeky ? what does that mean ?
15:09:00edxhmmm.. what do they cost?
15:09:17alkorrtoo expensive for the moment I suppose
15:09:32alkorrfor what they can bring to us
15:09:47edxwell... :)
15:10:22edx400 Euros... only??!! wow.. i had exspected much more actually...
15:10:23edxcool :D
15:10:51alkorryes but i would buy it if i cannor reverse-engine it :)
15:10:56alkorrNOT
15:11:01alkorrwouldn't
15:11:09edxyup
15:11:09edxhehe :)
15:11:23LinusNalkorr: a catch 22 indeed
15:11:24edxwell those toys have two fun-sides - using and disassembling
15:11:41alkorrthe fact that it uses a ARM7 and a color screen is something i like
15:11:54langhaarrockerDo you all belong to those people who disassemble a toy _before_ trying it?
15:11:57Zagoryes. and what seems like a programmable dsp
15:12:14langhaarrockerCan we do ogg on it?
15:12:18alkorrand a TI DSP too we can program
15:12:21alkorrsurely
15:12:45langhaarrockerOk: I _need_ a MMJB!
15:12:47alkorrso it is why i'm trying to see if we can hack it
15:14:21edxhmm where do you have the firmware from? i cant find it on the archos site
15:14:56alkorrthe picture i found at http://groups.yahoo.com/group/archosmultimedia/files/107-0763_IMG.JPG shows us a TMS320
15:15:19alkorrfrom http://groups.yahoo.com/group/archosmultimedia/files/
15:15:26LinusNcool
15:15:38alkorrdriver downloadibg section in ww.archos.com
15:15:40alkorrtoo
15:16:36edxah ok
15:16:44edxhmm .. was on the wrong way.. :)
15:18:29alkorrhey they use a CPLD
15:18:39alkorri'm wondering for what...
15:21:05langhaarrockerHave you seen the scar on the thumb of http://groups.yahoo.com/group/archosmultimedia/files/107-0764_IMG.JPG
15:21:05langhaarrockerThat reminds me of my thumb half a year after I poked a hot soldering iron into it...
15:21:18edxhmm alkorr, you have been talking about compression of the firmware...
15:22:44alkorrTMS320DSC21, there are several PDF about it : ARM7+TI DSP -> http://www.ti.com/sc/docs/news/2000/00104.htm
15:23:35alkorrso there is no reason we cannot have a OGG on it too
15:23:47alkorredx, yes
15:23:50edxwhat have you found about the compression?
15:23:50alkorri'm not sure
15:23:56edxyou were talking about two sizes?
15:24:08alkorryes they have a header of 16 bytes
15:24:16alkorrthe first word is a signature ZAZA
15:24:24edxyes that was about what i was going to throw into the conversation :)
15:24:38edxand after that a little less than the actual file size.. or something, right?
15:24:54alkorrthe second word looks a size around 520 Kbytes
15:25:15alkorrthe third is exactly the file size minus 16 bytes comming from the header
15:25:22edxyup
15:25:27alkorrand the fourth probably a CRC
15:25:57edxaha
15:26:11alkorrfile sizes are approximatively 280 KBytes
15:26:16edxso then the second word might be the unpacked filesize?
15:26:38edxdo you have any other firmware version?
15:26:48alkorrperhaps because it is always twice as large as the filesize
15:26:55alkorrthree versions
15:27:03edxoh cool
15:27:13edxcan you still d/l older versions on the site?
15:27:27alkorrhttp://groups.yahoo.com/group/archosmultimedia/files/107-0764_IMG.JPG
15:27:34alkorrsorry bad link
15:27:48alkorrhttp://groups.yahoo.com/group/archosmultimedia/files/
15:27:57edxhmm i dont have a yahoo id *shame*
15:28:01edxwell im gonna get one
15:28:07alkorrdcc capability ?
15:28:34edxhmm im behind a server
15:28:40edxgenerelly yes
15:29:13alkorrindeed four version but one is the factory installed version (0.66)
15:29:29alkorri'm making an archive of them
15:29:43Hadakahm
15:29:48edxok - can you send it to me then?
15:35:29alkorras you can see compression is not efficient
15:35:50alkorrcannot get it ?
15:36:48alkorredx: wake up !
15:37:17edxoh sorry
15:37:25edxi was pretty busy signing up at yahoo
15:37:32edxthanks a lot :)
15:43:37 Join RipnetUK [0] (~george@ripnet.fsnet.co.uk)
15:44:01alkorrmaybe they used gzlib, it would be so nice :)
15:45:01Zagoralkorr: I doubt it. that would put them in a difficult legal position.
15:45:24LinusNZagor: really?
15:46:02alkorri'm not sure
15:46:13alkorrmaybe a LZSS :)
15:46:28Zagorit depends which lib you are talking about
15:47:44Zagorthis lib would be ok: http://www.gzip.org/zlib/zlib_license.html
15:48:01alkorrzlib yes
15:48:19edxhmm .. are there any command sequences typical for the processor of the archos multimedia?
15:48:35Zagoredx: look up ARM7. it's a very common cpu core
15:48:35LinusNlook for the vector table
15:48:53LinusN(here we go again...)
15:48:57edx:)
15:49:38LinusNeven if cryptanalysis is fun, i'd rather work on rockbox a little bit more...
15:50:05edxtrue.
15:52:07alkorredx: the fact we cannot see any text indicates us that they used a compression and/or encryption schema.
15:53:24alkorri prey that they really use libz to compress their software because it would make easier our task
15:57:57LinusNgotta go now! cu!
15:58:35 Part LinusN
16:00
16:11:48 Join mecraw12 [0] (~lmarlow@67.41.113.92)
16:12:52Bagderif they used compression it should be easy to figure that out
16:15:22 Join gizz [0] (gizz@lyon-2-a7-62-147-20-77.dial.proxad.net)
16:16:57gizzhi
16:17:06gizzis diddystar here ?
16:22:04 Quit xam (Read error: 113 (No route to host))
16:28:56 Nick Zagor is now known as Zagor|away (bjst@as9-5-6.k.s.bonet.se)
16:28:59***Saving seen data "./dancer.seen"
16:39:46Snorlaxy0z0rZ?!?
16:39:59alkorrgrumpf ?
16:40:21langhaarrockerwas that the MMJB decryption?
16:40:35Snorlaxwell have ya mixed withthe sound on the playerZ latly?!
16:41:07 Quit mecraw_ (Read error: 104 (Connection reset by peer))
16:43:40alkorrokay the fourth word is CRC for the file minus header
16:44:55alkorrwhen i compute the sum of bytes in the file except the first 16 bytes, i always get a sum equals to the fourth word
16:45:43Bagder:-)
16:47:09alkorrso we know about 3 of the four word in header. I'm still unsure about the purpose of the second word except that is always nearly twice as large as the filesize. total size of the firmware in RAM (i.e, with BSS section included) ? i don't see why this piece of information is necessary
16:48:25alkorrwhat i fear the most is there is an decryption to do before a possible decompression
16:49:38 Join Phil [0] (~Philipp@dsl-2129326-034.ip.versatel.de)
16:54:14 Quit langhaarrocker (Read error: 60 (Operation timed out))
16:54:27 Part Phil
17:00
17:00:08pimlottcalkorr the hakorr
17:11:04gizzbye all !!
17:11:18 Part gizz
17:15:51 Quit edx (Read error: 110 (Connection timed out))
17:18:22Snorlaxgotta go to IKEA!
17:18:25SnorlaxYAY SWEDEN!
17:18:31Snorlaxbye!
17:18:35Bagderbye Snorlax
17:18:35 Quit Snorlax ("gittar ny!")
17:22:33 Join hardeep [0] (hardeeps@sdf.lonestar.org)
17:22:53hardeepmorning all
17:23:08alkorrBagder: i want to link zlib.a
17:23:37Bagderyou mean to compress the image?
17:23:45alkorrto uncompress
17:23:57Bagderuse bzip2 instead
17:24:02Bagderits a better compress
17:24:14alkorrno i'm trying to uncompress the firmware
17:24:36alkorri must compile a modified descramble.c
17:25:05Bagderand?
17:25:24alkorrzlib.h is okay, but i must specifiate a library to link but i cannot do with a -l since zlib is not standard naming for library
17:25:43Bagderright, its normally called libz.a
17:26:02alkorrokay i try it
17:26:51alkorrmaybe 'uncompress' is not in zlib.h... should not be so
17:27:40alkorrit is part
17:28:15Bagderremove the header and run a linux 'file' on it
17:28:29Bagderthen you might see what kind of compression it is
17:29:03alkorrarf !
17:29:48alkorrwell try but there is no reason there are signature after teh header to allow 'file' to recknolize
17:30:29Bagderif they used one of the formats zlib recognizes, I believe it could be possible
17:30:36Bagderbut I don't know
17:31:27alkorrdo you know how i can do with 'dd' to strip off the header ?
17:31:33alkorri'm lazy tpo program :)
17:31:41Bagderhehe, lemme see
17:32:20Bagderskip=BLOCKS
17:33:21alkorrhow to set how many bytes for BLOCKS ?
17:34:04Bagderyou need to set blocksize to 1byte or something, like in "bs=1 skip=16" I think
17:34:10alkorrok
17:35:34alkorrgood
17:36:01alkorrit says me 'data' nothing much
17:36:13Bagderok
17:39:12 Join langhaarrocker [0] (~Phil@dsl-2129326-199.ip.versatel.de)
17:39:28alkorrwell it seems uncompress is defined in zlib.h but not in libz.a :/
17:45:50Schnueffcan't u just pipe it into gzip for a test?
17:46:12alkorrno i just wanted the compressed data to uncompress
17:46:27alkorrwith gzip you need a header
17:46:51Schnueffah yes
17:48:34alkorrf*cking gcc, you need to put -lz elsewhere in the command line !
17:49:15alkorrthe time for the truth...
17:49:28Schnueff...and?
17:50:21alkorreeeck seg fault :/
17:50:25fraggleto use zlib?
17:51:36alkorrmust be my program... maybe a null pointer. it's so common in C
17:52:30alkorroh yeah piece of code i forget to remove
17:55:41alkorrfailure
17:56:14hardeepheh, two snake games now
17:56:20hardeepwe can't get enough snakes!!
17:57:54langhaarrockerno, its three
17:58:45hardeepwhat's the third?
18:00
18:00:17langhaarrockersorry, I mixed it up: I thought you were counting worms instead of games.
18:00:32hardeepoh, heh
18:00:38alkorrstrangely enough, the difference always starts at offset 64 for each firmware file. So between offset 16 and 64 we have the same bytes : what does that mean ?
18:02:49 Join Dexter [0] (~Dexter@Mix-Pointe-a-Pitre-102-3-134.abo.wanadoo.fr)
18:03:53 Part Dexter
18:04:18alkorroh sorry 64 is in hexadecimal !
18:04:19 Part Bagder
18:07:37Schnueffso its 100
18:08:15Schnueffhm
18:09:46alkorr96 bytes untouched...
18:10:30 Join EnnaN [0] (bla@cc53164-b.groni1.gr.nl.home.com)
18:10:46EnnaNhi
18:11:15alkorrARM code always have conditional flag on their opcode. Most time they would be unconditionnal so an opcode like Exxxxxxx (opcode are 32-bit in ARM)
18:11:40alkorrbut i can frankly find such occurence in their file...
18:11:43alkorrcannot
18:11:51EnnaNah, people in here...mind if i ask a question? (another 1 than this :)
18:12:10alkorrtry it
18:13:24EnnaNi'm (ofcourse :) using the rockbox (release 1.3) but was wondering if any of the daily builds is stable and better...i know thse builds are xperimental, but still...
18:13:55alkorrwell the people who can answer you are gone out, sorry :(
18:14:19EnnaNtoo bad..
18:14:31EnnaNyou not using anything?
18:15:45langhaarrockerI'm always using the newst build plus my own stuff.
18:16:10langhaarrockerAnd it's always my own stuff that's not stable... ;)
18:16:33EnnaNany special features on newer daily builds that's really worth "the trouble", or "trying" ?
18:16:54langhaarrockerYes: two snake games.
18:17:50EnnaNprob not on player?
18:18:05hardeeplaf... a big feature in the daily builds not in 1.3 is ROLO which lets you load the archos fw from rockbox
18:18:19hardeepalthough, it's probably not as useful if you have a player
18:18:50EnnaNWould be usefull on recorder because? ... (rockbox doesnt record, right?)
18:19:11hardeepright... so load up archos fw to record then go back to rockbox to play
18:20:01langhaarrocker:)
18:20:19hardeepoh, other big feature in the daily is customizable wps
18:20:28hardeepif you don't like the default
18:21:08 Join xam_ [0] (~xam@tqpc.ucc.ie)
18:21:35EnnaNcustomizable in what way?
18:21:51hardeepin every way. :) you can specify what gets displayed where...
18:22:16 Nick xam_ is now known as xam (~xam@tqpc.ucc.ie)
18:22:23hardeepsee: http://rockbox.haxx.se/docs/custom_wps_format.html
18:22:24EnnaNah, right..
18:23:19EnnaNyeah..is there a listing of those docs? cause if i go to rockbox.haxx.se/docs, there is no listing of these fine docs...
18:23:37EnnaNthats why /me no found custom_wps_fromat.html..
18:24:02hardeepthe faq has a link to it
18:24:49EnnaNcheck **shame**
18:25:06hardeepelinenbe: snake game broke rec sim build
18:29:02***Saving seen data "./dancer.seen"
18:31:55langhaarrockerI think its the makefile ../x11/Makefile
18:32:43langhaarrockerIn line 93 there probably is a snake.c missing
18:35:03 Join edx [0] (~edx@pD9EAAB59.dip.t-dialin.net)
18:42:43 Quit EnnaN ("Leaving")
18:51:24 Nick edx is now known as edx|away (~edx@pD9EAAB59.dip.t-dialin.net)
19:00
19:03:57 Part langhaarrocker
19:12:25 Nick edx|away is now known as edx (~edx@pD9EAAB59.dip.t-dialin.net)
19:18:28 Quit hardeep ("Leaving")
19:21:47 Quit alkorr (Read error: 110 (Connection timed out))
19:27:49 Quit RipnetUK ()
19:31:54 Join merwin [0] (merwin@bespin.org)
19:31:57merwinanyone around?
19:37:03quel|outme
19:37:04quel|outbut must go
19:37:08quel|out:(
19:50:10 Join hardeep [0] (hardeeps@sdf.lonestar.org)
19:56:43 Quit merwin ("BitchX-1.0c19 -- just do it.")
20:00
20:16:03 Join EnnaN [0] (bla@cc53164-b.groni1.gr.nl.home.com)
20:17:31dwihno|gonequel thalas
20:25:18 Join merwin [0] (~merwin@12.242.185.10)
20:29:06***Saving seen data "./dancer.seen"
20:44:08 Quit hardeep ("Leaving")
20:59:21 Join adiamas [0] (~adiamas@as5300-9.216-194-23-184.nyc.ny.metconnect.net)
20:59:49 Nick adiamas is now known as adi|home (~adiamas@as5300-9.216-194-23-184.nyc.ny.metconnect.net)
21:00
21:01:07 Quit merwin (Read error: 110 (Connection timed out))
21:04:32 Nick seb-school is now known as _seb_ (user@bgp420584bgs.union01.nj.comcast.net)
21:07:22 Quit quel|out (Read error: 104 (Connection reset by peer))
21:08:57 Join quelsaruk [0] (~zaknafein@faerun.ugr.es)
21:11:29adi|homeis John Wood here?
21:15:00 Nick edx is now known as edx|test (~edx@pD9EAAB59.dip.t-dialin.net)
21:15:52 Nick edx|test is now known as edx (~edx@pD9EAAB59.dip.t-dialin.net)
21:29:07dwihno|goneedx: Are the gnush makefiles ready for the new game?
21:29:57quelsaruknew game?
21:30:20adi|homeyup
21:30:31quelsarukwhich one?
21:30:49adi|homesnakes
21:31:07quelsarukit's similar to wormlet?
21:31:32dwihno|gonesnake0rs
21:31:39adi|homesimilar yeh..
21:31:43adi|homebut this is more buggy...
21:31:48adi|homeleast it was when i just played with it
21:31:54quelsarukjeje
21:32:03quelsarukwell, i must go, but i'll try it tomorrow
21:32:04quelsaruk:)
21:32:12quelsarukcu
21:33:22 Quit quelsaruk ("KVIrc 3.0.0-beta1 "Eve's Avatar"")
21:44:37edxdwihno: what new game?
21:44:40adi|homeelinenbe|sleep: you around yet?
21:44:41adi|homesnake
21:44:47adi|homeeric commited it
21:44:49adi|homebut its buggy
21:44:51dwihno|goneya, snake0r
21:46:50 Quit xam ("ChatZilla 0.8.7 [Mozilla rv:1.0.0/1]")
21:47:07edxhmm snake
21:47:09edxill try.. :)
21:48:34edxok - the gnush makefiles are snake compatible
21:53:16*edx wonders why he does not get commit information mails any more
21:53:24edxdo you still get those mails?
22:00
22:16:43 Quit EnnaN ("Leaving")
22:20:44 Join hardeep [0] (hardeeps@sdf.lonestar.org)
22:29:10***Saving seen data "./dancer.seen"
22:29:25 Nick Zagor|away is now known as Zagor (bjst@as9-5-6.k.s.bonet.se)
22:31:10edxzagor, i dont get cvs commit information mails regularly - is there something wrong?
22:32:46Zagori get them, so it seems the list works
22:34:04edxhmm would you check whether my e-mail address is still on the list?
22:35:04Zagoryup
22:35:10edxhmm strange
22:35:15edxvery strange..
22:35:47Zagorwhat's your address?
22:48:33 Join edx|notebook [0] (~edx@pD9EAAB59.dip.t-dialin.net)
22:49:04 Part edx|notebook
23:00
23:01:21adi|homeshowering.. be back
23:03:59 Nick edx is now known as edx|sleep (~edx@pD9EAAB59.dip.t-dialin.net)
23:10:45 Nick _seb_ is now known as seb-away (user@bgp420584bgs.union01.nj.comcast.net)
23:21:22adi|homewe really need to consolidate the make files for x11 and win in the sims
23:34:40 Quit Zagor ("Client Exiting")
23:37:29 Join Lowfiler [0] (H@pD95423E6.dip.t-dialin.net)
23:37:31Lowfilerheya
23:38:32 Quit edx|sleep ()
23:40:03Lowfilersome1 here

Previous day | Next day