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 2003-05-21

00:00:23Yeftwinamp is associated with it on my comp for some reason
00:00:46Stevie[FP]because '.mod' is the file extension of an old music file format, similar to MIDI (.mid)
00:00:57_aLF(from amiga)
00:01:09Yeftbut why wont my archos recognize it anyway
00:01:39Stevie[FP]did you make sure it's really named archos.mod ?
00:01:46Stevie[FP]what OS do you run? you said (xp)
00:02:07Stevie[FP]a 9x kernel (95/98/ME) or NT kernel (NT/2K/XP)?
00:02:16Stevie[FP]what drive letter is your archos?
00:02:22 Quit awy[baer] (Remote closed the connection)
00:02:26DBUGEnqueued KICK Yeft
00:02:26Yeft My Puter Stats OS-[Windows XP Home Edition (5.1 - 2600)]- Up Time-[day 10hrs 58mins 30secs]- CPU-[-Intel Pentium 4, 1593MHz, 256KB (9% Load)]- Memory-[Usage: 280/512MB (54.69%)]- Video Card-[Not Set]- Resolution-[ x 600]- Total Space Free-[.41Gb out of 71.67Gb]-
00:02:31Yeftif you really care
00:02:36Yeftf drive
00:02:38Stevie[FP]ok, that was really not necessary ;)
00:02:41Stevie[FP]ok, get a command prompt
00:02:43Stevie[FP]/run cmd
00:02:57Stevie[FP]chkdsk /f F:
00:03:14Stevie[FP]my brother had a problem with his Archos
00:03:17Stevie[FP]it wouldn't even boot
00:03:20Stevie[FP]i had to chkdsk it
00:04:48 Quit tracktheripper (Read error: 104 (Connection reset by peer))
00:05:27Stevie[FP]is the chkdsk almost done?
00:05:46Yeftwell, it took a while for me to figure out the usb was pulled out so its just starting...
00:05:46 Part LinusN
00:05:56Yeft52% complet
00:06:36Yeftit says "convert lost chains to files? (y/n)"
00:06:55Stevie[FP]and you need to start using unplug/eject
00:07:00 Quit [IDC]Dragon ("Leaving")
00:07:22Yeftyeh, it was probably once or twice, i usually do though (i was probably in a hurry as i am now lol)
00:07:27Yeftits done
00:07:39Stevie[FP]now do an unplug/eject
00:07:44Stevie[FP]and see if it works
00:08:20Yeftstill loads the old firmware
00:09:35Yeftit says the type of file is a winamp file which i am not happy about...but i have to go now...thanks (i may be back later)
00:10:46PsycoXularchos was dumb to name it .mod
00:10:57 Quit Yeft (" Info-[v9.4.2]- Released-[August 10, 2002]- Channel-[#Excursi")
00:11:11PsycoXulthats probably why they changed it for recorders
00:14:32Stevie[FP]psyco, how much exp. do you have reverse-engineering devices?
00:15:40CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:15:40*PsycoXul thinks...
00:15:55PsycoXulum, none.
00:15:56 Join LinusN [200] (
00:16:01PsycoXulpretty much
00:16:10Stevie[FP]linus, I have a hunch
00:16:35Stevie[FP]I have a sneaking suspicion that (due to the potential ROM-is-slow issue)
00:16:49Stevie[FP]the MAS normally copies its own firmware to RAM when it starts up
00:18:06Stevie[FP]there are instructions here
00:18:17Stevie[FP]on how to extract the contents of internal RAM
00:19:14LinusNStevie[FP]: reverse engineering a custom DSP...tricky stuff
00:19:58LinusNi doubt that it's custom though
00:20:00Stevie[FP]we're not doing anything permanent, right?
00:20:16LinusNthey probably licensed an existing core
00:20:25Stevie[FP]I'm betting it's a RISC core
00:20:37Stevie[FP]anybody wanna search for processors with a 20-bit instruction size?
00:26:26LinusNcare to send me the eagle files?
00:26:35Stevie[FP]eagle files?
00:26:44Stevie[FP]I just found that schematic on the web
00:26:59LinusNoh, i though you made it yourself
00:27:01Stevie[FP]and I regret to say that I dunno wtf an eagle file is
00:27:18LinusNEagle is an elcetronics cad program
00:28:30Stevie[FP]I wonder if we could use some sort of stockholder reports to find out where Micronas spends their money :D
00:28:40Stevie[FP]'Acme DSP Cores - $100,000'
00:28:56LinusNi searched the web a little
00:29:03LinusNseems like they do their own cores after all
00:29:30Stevie[FP]milestones: products:
00:29:33Stevie[FP]1992 - Open DSP architecture DIGIT3000
00:29:41Stevie[FP]open, eh?
00:30:34Stevie[FP]I wonder how hard-working these people are
00:30:48Stevie[FP]People tend to fall back on what they've built before
00:32:44 Join awy[baer] [0] (
00:33:38 Quit awy[baer] (Remote closed the connection)
00:37:49 Join awy[baer] [0] (
00:42:26 Quit Jet8810 ("Client exiting")
00:50:55 Join diddystar5 [0] (
00:51:51diddystar5sometimes when doing a cvs download i now get a error anyone have the same problem?
00:52:12LinusNsourceforge anonymous cvs access is choked
00:52:47diddystar5ok if i keep trying it eventually i get it
00:53:46LinusNour daily build script tries in a loop
00:55:31diddystar5thanks Linus bye!
00:55:44 Quit diddystar5 (Client Quit)
01:01:05Stevie[FP]do we have any idea about what clock the MAS runs at?
01:01:47LinusNaround 14MHz i think
01:04:09_aLFLinusN> I see somewhere that speed vary (25Mhz for decoding, 50Mhz for encoding)
01:04:22Stevie[FP]ah, the 3857 or whatever has a 18.432mhz clock
01:04:24Stevie[FP]rather high speed
01:04:37Stevie[FP]54ns per tick
01:05:25Stevie[FP]what do you think the odds are that they found a (relatively cheap) ROM chip that can deal with a 50ns access time?
01:06:41LinusNwell, if it's an ASIC, i think they wouldn't have a problem with that
01:12:08 Join tracktheripper [0] (
01:12:56tracktheripperhi LinusN
01:13:22tracktheripperLinusN when using timestretching to make a song faster it sounds like "bits" of the song are "missing"
01:13:26tracktheripperthis is using Soundforge
01:13:37tracktheripperlike bass thumps and the like
01:14:21tracktheripperwondered why that happens
01:15:23LinusNwell, i guess the algorithm isn't perfect
01:16:21tracktheripperLinus I feel every feasable feature request is in Rockbox now
01:16:29tracktheripperI can't imagine anything else new being added
01:16:41tracktheripperwhatever Rockbox cannot do I use the Archos firmware instead
01:16:47LinusNid3 editing?
01:16:55LinusNplaylist editin?
01:17:00LinusNrecord timer?
01:17:09 Join Yeft [0] (
01:17:12tracktheripperwhen are those requests u mentioned coming?
01:17:27Yefti'm back : )
01:17:35tracktheripperhi yeft
01:17:40tracktherippernice to meet you :-)
01:17:50LinusNtracktheripper: when? well, when it's ready
01:18:03LinusNwe have no time plan
01:18:35tracktheripperArchos don't seem to be updating their firmware anymore
01:18:55Yeftcause archos has such high quality standards : /
01:19:20tracktheripperLinusN will a "Move File" feature be intoduced one day?
01:19:21Yeftgreat machines....god awful firmware
01:19:32tracktherippermove files around different folders on the fly
01:19:37LinusNtracktheripper: sure
01:20:05Stevie[FP]maybe Archos does have high quality standards
01:20:17tracktheripperwhat about making the screen touchscreen like on a PDA for easier using of the built-in keyboard?
01:20:25Stevie[FP]and that's why they won't update their firmware −− they know they can't program, and they'll just fsck it up more
01:20:45*Stevie[FP] looks at tracktheripper
01:20:47Stevie[FP]wheres my LART?
01:20:55*Stevie[FP] digs around for it
01:20:59Yefti dont think you can make a non touchscreen touchscreen
01:21:02Stevie[FP]ah, here it is
01:21:05tracktheripperwhats a lart?
01:21:11*Stevie[FP] whaps tracktheripper repeatedly on the head with his LART
01:21:18tracktheripperouch :_)
01:22:03tracktherippermy head hurts
01:22:14Yeftok anyways...
01:23:01Yeftcan someone help me here?
01:23:09tracktheripperwhats up yeft
01:23:36Yeftmy archos thinks .mod is some winamp file : (
01:24:07LinusNYeft: does your Explorer show file extensions?
01:24:18LinusNthen make it
01:24:38_aLFgood night
01:24:43LinusN_aLF: nite
01:24:43 Quit _aLF ("bye")
01:24:51Yeftok, right on it
01:25:33Yeftand now it does
01:25:44 Quit tracktheripper ("Leaving")
01:26:06LinusNso is the file named ARCHOS.MOD?
01:26:19Yeft1 sec i mightve fixed the problem thanks to that
01:26:32Stevie[FP]i'm thinking acrhos.mod.mod
01:26:36Stevie[FP]or something like that
01:26:39Yeftand you are correct
01:26:50LinusNYeft: common mistake in the Windows world
01:26:51Yefti feel dumb
01:27:10Stevie[FP]i want to find the moron who came up with this 'hide file extensions' crap
01:27:26Yeftyeh me 2!
01:27:31Stevie[FP]and I want to cause him great pain
01:27:49Stevie[FP]I bet that half of the idiots who've run viruses named blah.txt.exe
01:27:57Stevie[FP]wouldn't have run them if they'd seen the .exe extension
01:29:05Stevie[FP]meh, stupid Micronas
01:29:35Yeftnow... took get the archos to boot up
01:30:06Yeftgood to go
01:30:33Yeftthats probably true (the .txt.exe)
01:34:06devZer0ooohhh - i feel so dumb, too :)
01:34:35devZer0hacking C - it`s not the world i`m living in :)
01:35:12*Stevie[FP] is away [gone] [KS-MsgLog Off]
01:35:13 Quit Yeft (Read error: 104 (Connection reset by peer))
01:35:28devZer0linus - i have found an issue with the rockbox simulator on win32
01:36:34LinusNdevZer0: yes?
01:36:54devZer0i use the sim on winxp with ntfs filesystem
01:36:56 Join tracktheripper [0] (
01:37:03tracktherippersorry got cut off
01:37:10 Join midknight2k3 [0] (
01:37:13tracktheripperLinus where can I read about the MAS chip in the Archos?
01:37:17tracktheripperhi mid
01:37:28midknight2k3It's undocumented isn't it?
01:37:40midknight2k3They have little info on it
01:37:44midknight2k3As I understand
01:37:51devZer0and the directory-icons don`t show up all
01:37:51midknight2k3I did see a link the other day though
01:38:16tracktheripperthanks linus
01:38:25devZer0the .attr seems to be different
01:38:46LinusNdevZer0: player or recorder?
01:39:06LinusNi didn't know that the icons were special
01:39:59devZer0how do i print a short in formatted way to debugf?
01:40:19devZer0then i can analyse this better
01:40:42LinusNdevZer0: %d should work
01:42:23devZer0the dir, where the icons are shown have a .attr with value 22
01:42:32 Join ferdy_ [0] (
01:42:38devZer0e.g. this is dir "Config.Msi"
01:42:58devZer0then i have dirs "admin" and "bb" - the have 8208
01:43:21tracktherippera lot to take in
01:43:26devZer0cygwin and cygwin-src have 10256 - all of them are without icons
01:43:38ferdy_hey, ermm, I'm just setting up the dev environment for win32 and have a few daft questions, anyone around be able to answer them?
01:44:10devZer0hey - depends WHICH win32 env. you are setting up ;)
01:44:37ferdy_easiest one possible, I'm just getting things going with cygwin
01:45:23 Quit tracktheripper ("Leaving")
01:45:49ferdy_I've got the source, done a configure, but I don't have gcc by default, should i have grabbed the gnush compiler?
01:46:06devZer0no - you can install cygwin gcc
01:46:24ferdy_OK, which i take it doesn't come with the default setup
01:46:39devZer0think so
01:46:41LinusNferdy_: of course not
01:46:53***Saving seen data "./dancer.seen"
01:47:00LinusNferdy_: sorry, i misunderstood
01:47:06LinusNyou need gcc
01:47:37LinusN_and_ the sh1 cross gcc
01:48:23midknight2k3I'll beback in a few minutes
01:48:25 Quit midknight2k3 ()
01:48:47ferdy_is that the gnush ?
01:49:15LinusNgnush is a ready-compiled cross gcc, yes
01:49:42 Join Yeft [0] (
01:50:13devZer0mhh - wasn`t sh1 cross-gcc "inside" gcc package?
01:50:25hardeepgnush doesn't work too well with cygwin though
01:50:31devZer0its some time ago that i installed that...
01:52:24Yefti could never be a computer helper guy, i am tooo impatient
01:52:42LinusNsame here
01:53:08ferdy_OK, from the web page it indicated that the compiler was picked up
01:53:28Yeftis there other archos firmware like rockbox?
01:53:35LinusNferdy_: that one is a special hack
01:53:51LinusNYeft: i never heard about anything else
01:54:00devZer0yes :) i made that some time ago. it is meant for people with small bandwidth
01:54:11devZer0and the impatient ones :)
01:54:26devZer0just follow the cooking-receipe - and you can compile rockbox
01:54:30ferdy_so is there a compiler in the package that I'm not finding one way or another?
01:55:06LinusNif you downloaded that package, yes
01:55:17ferdy_d'oh, where should it be?
01:55:23 Join tracktheripper [0] (
01:55:23devZer0that package is just a stripped-down cygwin installation
01:55:52tracktheripperLinusN i read about the MAS chip on the Micronas website. there is no way to make it read WMA
01:56:29devZer0does one need wma? ;)
01:56:43tracktheripperi have a few wma tracks
01:56:43Yeftbut maybe you could make it think it was reading mp3 but really reading an wma :)
01:56:57tracktheripperyea yeft rename the extension :P
01:57:47Yeftdamn extensions :
01:58:19Yeftmy archos needs some cooloff time, its been gettin a workout
01:58:22tracktheripperu would lose quality coding a 64kbs WMA to 128kbs MP3 anyway
01:59:54Yeftlinus, does logbot have a feature that can email logs (i've been looking for one that can)
02:00:08 Quit tracktheripper (Client Quit)
02:00:32DBUGEnqueued KICK Yeft
02:00:32Yeft Waaaaaas Uuuuuuuuuup #rockbox !!
02:02:43ferdy_devZer0, I have the sh-elf-ar.exe files, but none are aliases to gcc, should something have been set up like that ( this machine has had various cygwins on it at stages )
02:03:08 Join midknight2k3 [0] (
02:03:15midknight2k3I am back!
02:03:26LinusNYeft: no, logbot has no email feature as far as i know
02:03:32Yeftok ty
02:03:49LinusNbtw, logbot is a Dancer (
02:04:05LinusNgothenburg at the moment, but i live outside stockholm
02:04:47Yeftis it nice, never been of this continent lol
02:06:17 Quit Yeft (" Info-[v9.4.2]- Released-[August 10, 2002]- Channel-[#Excursi")
02:08:10devZer0ferdy. you should do a clean cygwin install. or download win32sdk and install that.
02:08:45devZer0perhaps try to reinstall gcc first
02:09:19 Join Yeft [0] (
02:10:48 Part Yeft
02:11:04 Join Yeft [0] (
02:11:57 Part Yeft
02:12:25 Join Yeft [0] (
02:12:31Yeftok i'm here for good now lol
02:13:28ferdy_devZer0: I did install the win32sdk, but like it says any other cygwin installation may create problems, thanks anyways
02:13:31DBUGEnqueued KICK Yeft
02:13:31YeftJukebox Tip of The Night do not stick the end of the charger in your mout
02:13:50midknight2k3uh oh
02:13:56midknight2k3what happened
02:14:02Yeftit hurt
02:14:27midknight2k3While it was plugged in?
02:14:40Yeftbut thank god it was just on the very corner........ yes
02:14:54midknight2k3you crazy or somethin?
02:15:40midknight2k3"Oh yes. How boring. Oooh! Archos charger! Looks absolutely delicious! Let me taste!! (ZAP) Ouch!"
02:16:34LinusNoh i feel bored today, hmmm...i'll just inflict some pain
02:16:46Yeftor "hmmm sucking on the wire.... sucking down the wire is that the end??? *zap*)
02:17:22midknight2k3Oh my GOD
02:17:39devZer0ferdy -just backout your current cygwin installation. move cygwin to a different dir and make shure, there is no other cygwin-dlls on your system.
02:17:41midknight2k3I wish the archos firmware could be modified to work on the rioriot
02:17:59devZer0after that, save your cygwin relevant registry settings
02:18:00midknight2k3Archos ROCKBOX firmware that is
02:19:46devZer0linus - you inflict som pain? how that?
02:21:24midknight2k3<Yeft> Jukebox Tip of The Night do not stick the end of the charger in your mout
02:21:41midknight2k3<Yeft> bored
02:21:41midknight2k3<Yeft> confused
02:21:41midknight2k3<midknight2k3> "O
02:22:01midknight2k3<LinusN> oh i feel bored today, hmmm...i'll just inflict some pain
02:22:15Yefthmmm so i guess the jukebox6000 can't play videos huh
02:22:24devZer0what is the problem with sticking the charger in your mouth? no danger at all... :D
02:23:16midknight2k3While it's PLUGGED in
02:23:44devZer0yes - where is the problem?
02:23:54LinusNanyone here got the "rec fls: -114" error?
02:24:02midknight2k3I think I used to
02:24:07midknight2k3after that defrag
02:24:17devZer0your tongoue will prickle a little bit
02:24:27midknight2k3a little bit yes
02:24:48devZer0linus: no - me not.
02:26:53midknight2k3Linus, do you think it's possible to make custom firmware for the rioriot which uses an EXE file to install the upgrade on the player?
02:27:07midknight2k3That seems pretty impossible, to make up the firwmare and an EXE to install it
02:27:14midknight2k3you can't even see the files on the drive
02:28:10devZer0maybe the exe patches the firmware to flash-memory?
02:28:49midknight2k3that's what it does
02:29:17midknight2k3but it seems hard, knowing you can't see the drive (totally hidden) and you cant take stuff off of it
02:30:08LinusNit is certainly possible
02:30:16LinusNit just takes some work
02:30:27midknight2k3I don't know how you'd look at it though
02:30:38midknight2k3The code is basically built into the installation exe
02:30:39LinusNi'd open it
02:30:49midknight2k3open... the exe file?
02:30:49LinusNand examine the flash contents
02:31:00LinusNthe rioriot
02:31:12midknight2k3how can you see the contents of a chip?
02:31:40LinusNby connecting a logic analyser, for example
02:32:01LinusNi guess examining the EXE file would be easier though
02:32:08midknight2k3I have no equipment of the sort
02:32:16devZer0if it is an ordinary flash-rom, you could use a hardware, to read the "contents"
02:32:18LinusNme neither, but i can borrow
02:32:53midknight2k3I know there's people out there that can do this... the rioriot is the cheapest 20GB mp3 player
02:33:00midknight2k3I don't get why they don't
02:33:36devZer0is it ? i thought archos is
02:33:52midknight2k3perhaps bye a fine hair
02:34:07midknight2k3the rioriot has a nice 2.25" screen (i think thats the size)
02:34:17midknight2k3And a very good navigation system
02:34:37midknight2k3As opposed to.... 24 character cells that can't be modified more than 4 at a time
02:35:13Yeftany archos wont boot!
02:35:31midknight2k3I don't want to sound dumb CALLING them and asking about when the new firmwareis gonna be out
02:35:34midknight2k3uh oh
02:35:40midknight2k3HD ERROR?
02:37:20devZer0sorry guys - i`m very tired. gonna sleep now. good night!
02:37:55midknight2k3see you
02:37:58midknight2k3good night
02:38:02midknight2k3all that...
02:38:09 Quit devZer0 ("Leaving")
02:38:28Yeftmeh hd error
02:38:35midknight2k3Third one this month for me.
02:38:44Yeftmonth? third one this day
02:38:56midknight2k3third person to tell me about it
02:39:24midknight2k3try to: hook up to usb, delete rockbox, put on archos latest firmware
02:39:27midknight2k3should go away
02:39:32midknight2k3thenput on rockbox again
02:39:49Yeftlol ok
02:39:51LinusNmidknight2k3: did you reformat to get rid of the -114 error?
02:40:26 Quit ferdy_ ("ChatZilla 0.8.11 [Mozilla rv:1.2/20021126]")
02:40:33midknight2k3no way!
02:40:47midknight2k3I just one day put on the new rockbox and kazam it was gone
02:41:10LinusNthat was an interesting clue
02:41:32LinusNwas you jukebox close to full?
02:41:49midknight2k36.2GB left or so
02:45:09midknight2k3Me love me archos
02:45:13midknight2k3Me love me archos
02:45:38midknight2k3Me love me rioriot but me rioriot has no firmware upgrade cause of sonicstinkingblue
02:46:20 Quit Yeft (Read error: 104 (Connection reset by peer))
02:47:40 Join Yeft [0] (
02:48:19midknight2k3not working?
02:57:04LinusNmidknight2k3: was it a defrag that triggered the -114 bug?
02:57:40DBUGEnqueued KICK Yeft
02:57:40Yeftit makes noise like it wants to load, and it holds at the "Jukebox 6GB Ver : 5.02" screen
02:58:02LinusNit's funny, i defragged my fm several times to trigger the bug, to no avail
02:58:10midknight2k3are you working out that bug?
02:58:19midknight2k3Well it was like two months ago
02:58:20LinusNmidknight2k3: trying to figure out what it could be
02:58:25midknight2k3it was an older version
02:58:33midknight2k3before 2.0
02:58:33midknight2k3way before
02:58:42LinusNpractically nothing has changed since then
02:58:54midknight2k3welp I haven't an idea then
02:59:02midknight2k3i only got about 12% through it though
02:59:09midknight2k3perhaps that's it
02:59:11midknight2k3I stopped it
02:59:13LinusNthrough the defrag?
02:59:41midknight2k3it moved stuff but left some stuff in odd places?
03:00:02LinusNi bet the fsinfo structure was corrupt...
03:00:35LinusNfsinfo is an info block that tells how much space is left etc
03:01:00LinusNwindows doesn't always update it correctly
03:01:20midknight2k3there you go then that's gotta be it
03:05:54Yeftok format i come
03:06:58midknight2k3give it a go
03:22:47 Join Yeft|Away [0] (
03:22:47 Quit Yeft (Read error: 104 (Connection reset by peer))
03:22:54 Nick Yeft|Away is now known as Yeft (
03:23:17Yeftthe formatting screwed up, and now i no longer have rights to the F drive : (
03:24:47Yeftthis stinks
03:27:57LinusNscrewed up?
03:28:25Yefti got it to work...
03:28:42Yeftnow i have to load everything back on (wishes usb was as fast as firewire)
03:28:57LinusNUSB2.0 is faster
03:29:15Yefti dont have 2.0 : (
03:29:17LinusNbut the player is usb1.1 :-(
03:30:10Yeftcrap lol i loaded it wrong
03:32:28Yeftits gonna be like an hour : (((((((((
03:37:27adi|homelinus you still around?>
03:37:46adi|homeyou got time/want to talk about sokoban?
03:38:00adi|homek.. whats up?
03:38:17adi|homei realized the offset part wasn't working...
03:38:26LinusNi looked at your code for the loadable levels and i feel it's a bit complicated
03:38:34adi|homei was more worried aobut the segfault bit...
03:39:39LinusNfirst of all, is it a problem to have a numlevels=xxx first in the file?
03:40:05LinusNor is the levels.txt unmodified from the original?
03:40:05adi|homeit is if we want/expect the user to add/edit levels
03:40:16adi|homeits rather modified
03:40:28adi|homebut not drastically
03:40:44LinusNis it hard to use them unmodified?
03:40:51adi|homethe impression i took from zagor was that the # of levels shouldn't be fixed.
03:41:04adi|homewell... unmodified they are in the .h file
03:41:22LinusNhe meant fixed, as in hard-coded
03:41:52adi|homebut how do i know when i've reached the last level unless i count them?
03:42:07adi|homejust make the assumption that the user with never edit the levels?
03:43:39 Join Stevie-O [0] (
03:43:48Stevie-Ophjear! muhaha
03:45:11Stevie-Ois anyone even here?
03:45:27adi|homeno.. he left about an hour ago
03:45:32midknight2k3oh really lol
03:45:37adi|homebut if you give it some time.. everybody will be back
03:45:44adi|homeand he was going to bring his friend noone
03:46:00midknight2k3I see 20 people in here
03:46:07midknight2k3i've talked to about 3 of them so far
03:46:11midknight2k3and it's been awhile
03:46:21Stevie-OI haven't checked the very latest builds
03:46:24Stevie-Obut in 5/02
03:46:31Stevie-Owhen you stop recording
03:46:36Stevie-Owe generate an xing header
03:46:56***Saving seen data "./dancer.seen"
03:47:31midknight2k3like 'zing'?
03:47:35midknight2k3or 'ex-ing;
03:48:07CtcpIgnored 3 channel CTCP requests in 57 minutes and 8 seconds at the last flood
03:48:07*Stevie-O rolls his eyes
03:48:29midknight2k3that's it right?
03:48:44*Stevie-O LARTs midknight2k3
03:49:30midknight2k3is that an abbreviation?
03:49:37midknight2k3Or like 'farts on'?
03:50:14Yeftstevie-o = stevie[FP]?
03:50:29Stevie-Owhat gave it away?
03:50:37Yeftummm the LART actually
03:51:00midknight2k3what is lart?
03:51:14midknight2k3Loving Aour Rockbox Team?
03:51:39midknight2k3but then you can't LART someone..
03:52:29midknight2k3that's mean
03:52:43Yefthalf hour and my files are loaded
03:53:04*midknight2k3 larts mr. stevie-oeh
03:54:43LinusNadi|home: you currently search the file for levels when you start the game, and then for every time you load a level
03:55:21midknight2k3you're modifying sokoban?
03:55:49midknight2k3loadable levels?
03:55:55midknight2k3sorry yeft you can't play ;)
03:56:11 Nick Yeft is now known as Iogbot (
03:56:19Iogbotthats funny
03:56:25LinusNStevie-O: what about the xing header?
03:56:52Stevie-Owell it seems that we don't generate a TOC
03:56:52Iogbotwhat is a xing header?
03:57:07midknight2k3dont interrupt
03:57:07Stevie-Obut, I'm trying to find where the vbr info is calculated
03:57:11midknight2k3they're doing important work
03:57:16LinusNIogbot: an information block in the mp3 file to aid playback of VBR files
03:57:34adi|homeLinusN: well.. thats not the way its _supposed_ to work :)
03:57:38LinusNStevie-O: create_xing_header()
03:57:40adi|homewe do a level count at start
03:57:50adi|homethen find our level (and store its offset)
03:58:03adi|homeif we increase levels, we start the load at the last offset
03:58:17 Nick Iogbot is now known as Yeft (
03:58:29LinusNadi|home: but why do we need to know the number of levels from the start?
03:58:35adi|homeif we decrease we start from the start (or i could start at the offset - X)
03:58:43Stevie-Onumber of levels?
03:58:46*Stevie-O rubs his head
03:58:48adi|homeso that we know when you have 'won' the game
03:58:52LinusNStevie-O: sokoban
03:59:01*midknight2k3 days, "Stevie do it again"
03:59:13adi|homeeither that... or i could just keep reading till we run out of levels
03:59:17*Yeft punches midknight
03:59:19LinusNadi|home: if you get EOF when trying to load the next level...
03:59:19midknight2k3I liked the old way where it said YOU WON and the screen flashed
03:59:26midknight2k3now it just... leaves the game
03:59:48adi|homethe only caveat with that is that if we are loading the first level, and we get EOF we have a prob (and i tell the user and exit ) :)
04:00:04adi|homewould that make the complexity less?
04:00:15Stevie-Oif we load the first level and we get EOF
04:00:19Stevie-Othe user has some issues
04:00:26LinusNbtw, all you do is concatenate the rows, when loading the level?
04:00:38adi|homebasically, yup
04:01:21LinusNadi|home: this offset caching, how much do you gain?
04:01:38adi|homewell.. you don't have to reparse the entire file everytime
04:01:49LinusNbut how long does that take?
04:01:54adi|homeits the size of an int versus reparsing the entire file everytime
04:02:20LinusNi don't care about the int, i care about the code
04:02:31adi|homeshould't be much.. i just figured it was more convenient...
04:02:48midknight2k3any more games in the works, ppl?
04:03:03LinusNmidknight2k3: we have a few
04:03:10adi|homemidknight2k3: we need sokoban level loading working first ;)
04:03:25midknight2k3i hope this comes out
04:03:29midknight2k3i love the current ones
04:03:42midknight2k3Sokoban is my least favorite but it's still okay
04:04:59Stevie-Omeh! I lost the xing url
04:05:06Stevie-Ostupid scrollbacks
04:05:13LinusNcheck the log
04:05:55LinusNStevie-O: what about it, anyway?
04:06:02midknight2k3make an easter egg in rockbox it would be cool
04:06:25LinusNopen source easter eggs aren't that funny imho
04:06:52midknight2k3not funny, hidden
04:07:18midknight2k3I have an idea, make rockbox closed-source
04:07:19LinusNhiding an easter egg in the open source is hard
04:07:44midknight2k3i won't look at the source! lol
04:08:05midknight2k3how about a hidden rockbox concept logo?
04:08:48midknight2k3me miss easter egg
04:08:53Stevie-O(and *then* tell me that casting to/from char* is abuse :D)
04:09:21LinusNStevie-O: tried programming ini Shakespeare?
04:09:45midknight2k3I think he knows what you mean
04:10:36Stevie-O#define SHAKESPEARE (bb||!bb)
04:11:28Stevie-Oso wait, does this xing header actually store the average bitrate, or what?
04:11:31Yeftgmmg Hk˧? gm tH |m
04:11:54Stevie-Oyeft, don't make me hurt you
04:12:00LinusNStevie-O: it stores the stream size and the frame count
04:12:28Stevie-Oassuming that it doesn't switch between layers I/II/III
04:12:47Stevie-Oand samplerates (the latter of which is a pretty good indicator)
04:12:53LinusNnow why would it?
04:13:03Stevie-Obecause it's closed-source :P
04:13:05*midknight2k3 says "Hurt me! gm tH |m"
04:13:47LinusNStevie-O: LAME is not closed source
04:13:49Stevie-O1 1 2 3 5
04:13:58Stevie-Oi thought the encoder was the MAS chip
04:14:18LinusNin the recording case, yes
04:14:39Stevie-Owell that's what I was referring to
04:14:45Stevie-Oour generation of the vbr header
04:14:54Stevie-Owhen recording is finished
04:14:55LinusNi'm pretty sure that no encoder on earth switches sampling rate or layer
04:15:15Stevie-Oour counter appears to stop at 1048575 frames?
04:15:20Yefti dont get that
04:15:24Yefti dont get much though
04:15:30Yeft(wooooooosh) over my head
04:15:53LinusNStevie-O: yes
04:15:53Stevie-O1048575 frames * 1152 samples per frame == 1207958400
04:16:04Yeftall i know, is that wasnt in romeo and juliet, and in romeo and juliet they meet and BOTH die...sooooooo its not real
04:16:29Stevie-O1207958400 samples / 44100 samples/s = 2739.134694 seconds
04:16:38Stevie-O45mins 39secs
04:17:06Stevie-Oso we can record for about 45 minutes before we no longer generate a particularly effective vbr header (since num_frames is 0)
04:17:25LinusNabout 26ms per frame
04:17:52Stevie-Ohow often do we have to pull data from the encoder?
04:18:08midknight2k3My guess is 1 second
04:20:08midknight2k3well it's worthless
04:20:12Stevie-Oi know
04:20:20midknight2k3i better stop blurting in like a clumsy janitor
04:20:22Stevie-Owhich is why I don't understand you offering it
04:20:24midknight2k3continue and sorry
04:21:04 Quit kassoolet (Remote closed the connection)
04:21:12midknight2k3uh oh
04:22:21Stevie-Odorks at micronas
04:22:29Stevie-Oytf doesn't the counter wrap
04:23:47LinusNStevie-O: if the counter did wrap, we wouldn't have a problem
04:23:51Yeftdoes rockbox play .wav?
04:23:57LinusNYeft: no
04:24:19Stevie-OLinusN: we wouldn't have one if the stupid thing could be explicitly reset either
04:25:00LinusNguess so
04:25:35Stevie-OI just can't imagine that there's no way to do that
04:25:56Stevie-Oit says something about audio data time-outs
04:26:13LinusNStevie-O: the counter stops at 7ffff, btw
04:26:16Stevie-Othat's 510, yeft
04:26:33Yefto thanks, i always get that part messed up
04:26:39Stevie-Owait, it stops at 524287?
04:27:23Yeftlol i think the billionth digit is 0
04:27:43Yeftbillionth or trillionth or something
04:27:50Stevie-Ogood, cuz if we put a 1 there
04:28:02Stevie-Owe'd be off by 10^-(10^6)
04:28:11Stevie-Owhich would be perilous for our circle calculations
04:28:47Yeftbut you know what estimation i hat
04:29:26LinusNStevie-O: that is about 3 hours and 47 minutes
04:29:47Stevie-Owhat, 524287 frames?
04:29:55LinusN26ms per frame
04:30:37Stevie-O3hrs 47mins 11secs
04:30:46Stevie-Oi wonder how i got the other number
04:30:51LinusNme too
04:31:12LinusNanyway, we use the frame counter if it is less that 7ffff
04:31:32LinusNotherwise we don't include the frame count in the xing header when we generate it
04:32:02Yeftlol i cant stop my archos!
04:33:07LinusNi gotta go to sleep now
04:33:22Stevie-Onite linus
04:33:26LinusNStevie-O: any more questions about xing header generation?
04:33:34Stevie-Onot atm, thanx
04:33:39LinusNnite then
04:33:42 Part LinusN
04:35:32Yeftsee ya!
04:37:01 Quit Yeft ("i'd rather be rich than stupid")
04:40:40 Join Yeft [0] (
04:40:51Yefthow do i change the file limit?
04:41:01Yeftlol quick! i have to go........
04:41:33midknight2k3file limit?
04:41:39midknight2k3max files in directory?
04:41:42Yeftlol i dont get how to change the 400 file limit
04:41:50midknight2k3what version you using?
04:41:52Yefto well, g2g ill find out tomorrow
04:41:56midknight2k3no wait
04:41:56Yeft(newest build)
04:42:01midknight2k3heres how
04:42:09Yefttype spedily
04:42:14midknight2k3MENU>General Settings>system (i think)
04:42:20midknight2k3down 3 or for
04:42:27Yefti got thanks!!!!!!!!
04:42:38 Quit Yeft (Client Quit)
04:55:49 Quit hardeep ("[BX] Just do it like Nike... BEEATCH!")
05:10:02midknight2k3yeah WOOOOH\
05:10:05 Quit midknight2k3 ()
05:46:59***Saving seen data "./dancer.seen"
07:04:53 Join hardeep [0] (1098@
07:17:15 Join [keno] [0] (
07:23:32 Quit Stevie-O (Read error: 110 (Connection timed out))
07:47:03***Saving seen data "./dancer.seen"
07:58:27 Join matsl [0] (
08:32:54 Join Bagder [241] (
08:34:18 Join tracktheripper [0] (
08:35:48Bagdera fine morning it is
08:36:01tracktheripperits nice and sunny here in in England
08:36:13Bagderhere too
08:36:48tracktheripperany latest news?
08:37:07Bagdernot really, no
08:37:31tracktheripperi said to linus yesterday i think every feasible request has been implemented
08:37:37tracktherippercan't think what else can be added
08:38:04Bagderonce we get the plugin system there, we can add lots of more stuff
08:38:22tracktheripperwhat do you have in mind?
08:38:28Bagderpda-stuff for example
08:38:28tracktheripperas in extra features?
08:38:36Bagdercalender etc
08:38:37tracktherippercool :-)
08:38:51tracktheripperbut i feel there is only so much u can do with a 1bit mono LCD
08:39:19tracktheripperwill the Player have a plugin thingy as well?
08:39:30tracktheripper((feels player owners are being left out in the cold))
08:39:35Bagderit'll probably work the same on all models
08:40:10tracktheripperwhat the hell Archos were thinking with 22 character cells?!
08:40:34Bagderwell, the player is a cheap box
08:41:03tracktheripperare character-cell LCDs really a lot cheaper to make than bitmap LCDs?!
08:41:44tracktheripperwell Archos has discontinued the player anyway now
08:43:15tracktheripperi almost got a Player myself before plumping for the Recorder
08:54:43 Quit tracktheripper ("Leaving")
09:37:03 Quit hardeep ("[BX] Does your mommy know you use BitchX?")
09:37:47 Join mecraw [0] (
09:39:16webminddamn.. that 40gb hd is quiet... and stable
09:39:41BagderI should get one too
09:40:11webmindit's sweet :)
09:47:04***Saving seen data "./dancer.seen"
09:47:47 Join awy[baer]_ [0] (
09:56:05webmindhmm. any sugestsion on how to redisgn archos zo that it would be easy to change batts and hd's? :)
09:56:52Bagderwell, walkmans, remote controls etc usually have the batteries easy swapped
10:01:39 Nick dw|gone is now known as dwihno (
10:02:13webmindit's less easy on the archos
10:02:33webmindi could make it a 3 part device.
10:02:42webmindbatt, controll and hd pack
10:03:44 Join nicks [0] (~sten@
10:04:15 Part nicks
10:04:40 Quit awy[baer] (Read error: 110 (Connection timed out))
10:04:42 Quit dwihno ("Lost terminal")
10:05:35 Join dwihno [0] (dwihno@
10:06:24 Join ken0 [0] (marklar2@
10:06:41 Quit [keno] (Read error: 104 (Connection reset by peer))
10:28:27 Join Guest [0] (
10:28:53 Nick Guest is now known as erahurts (
10:29:11erahurtsAnybody here?
10:30:08erahurtsThanks for all your work omn Rockbox. Definitely makes the archos worthwhile.
10:30:34*Bagder bows
10:30:40Bagderwe're a large crowd of hackers now
10:31:00erahurtsSo I noticed in the credits.
10:31:40dwihnoBagder: 1337 h4x0rs? :) rs ;-)
10:32:05Bagdertry to pronounce that! ;-)
10:32:38erahurtsWorks in English. Not Swedish?
10:32:55dwihnoBagder: hah, you're so wasted :)
10:33:44erahurtsSo, if I changed the hdd in my fm recorder, no soldering is required?
10:33:58Bagderbut you void the warranty
10:34:56webmindchanging hd's works fine :))
10:34:59erahurtsColleague of mine just got an Ipod. Very nice interface. I'd get one, if I could hack it.
10:35:18webminderahurts, has a harder to get hd though.. and hell of job to open it
10:35:51erahurtswebmind: hard to get one?
10:35:57Bagderand there are only 30GB ones afaik
10:36:14webminderahurts, ipod uses pcmcia disks.. not 2.5" laptop disks
10:36:31Bagder1.8" disks aren't they?
10:36:49erahurtsOh, Ipod. Don't care. What about the archos. I can get a 2.5 60GB, right?
10:36:56Bagdereven 80
10:37:05webmindBagder, dunno.. but the ipod itself is smaller then a 2.5" disk
10:37:07 Nick erahurts is now known as earHurts (
10:37:24earHurtsYeah, I noticed the Ipod is small. And light.
10:37:25webmindBagder, do they 80'ers now? cool :)
10:37:36Bagderyeps, but still very expensive
10:37:40webmindearHurts, and has pretty nice interface.. but not quite hackable
10:37:49webmindBagder, awell.. my 40gb is fine :)
10:37:53earHurtsDoes the Fujitsu with the quiet option come in 80GB?
10:39:33earHurtsI like being able to add my own patches to Rockbox. Need to finish a few, and I think Badger wanted me to resubmit one
10:40:29BagderI did? ;-)
10:40:56earHurtsYeah, the id3 patch. (submitted by TPDiffenbach)
10:41:32earHurtsI was surprised when the patch was the entire file. ;)
10:41:44Bagderyeah, not very friendly
10:42:14earHurtsI just followed the patch tracker instructions. ;)
10:42:37earHurtsFigured there had to be a reason; forgot about my uing DOS line endings.
10:43:22Bagderdiff -b
10:43:58webmindbtw.. the prints doesnt seem to like the lack of casing..
10:44:29Bagderthe prints?
10:44:54webmindthe part where the batteries are connected
10:45:02earHurtsYeah, lemme try to remeber what files that involved. I've got about tree other files in transition −− unmerged.
10:45:05webmindif u have no casing they tend to bent quite a lot
10:46:56earHurtsWhat's the best way to manage mutiple changes? I'd like to fetch files from sourceforge's CVS, but be able to save tre files to my own cvs server? any easy way to do this?
10:47:22BagderearHurts: your changes will be merged with the ones you get from cvs
10:47:36Bagderjust 'cvs update' and it'll be made automaticly
10:47:46Bagderin case of conflicts, you'll be told
10:48:02earHurtsBut I can't update to sourceforge.
10:48:27Bagderyou mean you want to store your changes on a different cvs server?
10:48:40earHurtsThen I could version my changes.
10:48:53Bagderok, I see
10:49:11BagderI can't think of any easy way of doing that comfortably
10:50:03earHurtsYeah. Could be useful for open source type projects in general.
10:50:46earHurtsAnybody using MP3BookHelper?
10:51:49earHurtsVery nice Id3 editor. And the uthor is reat about adding features.
10:52:14webmindah k
10:52:50webmindah ok :)
10:53:03*webmind thinks of making a perl script for his id3 tags
10:53:26*Bagder did ;-)
10:53:37earHurtsthe ~200KB limit: is that because of a limit in the onboard memory space for code.
10:53:59earHurtsMp3bh does regexps.
10:53:59BagderearHurts: well, its hard to tell, but the loader refuses to load larger images
10:54:19Bagderwe can rolo larger ones
10:54:40earHurtsAccording to the docs for the chip, iirc, at least some versions have considerably more onbaord memory?
10:54:52Schnueffif one would have unscrambled firmware, would they load significantly faster?
10:54:58earHurtsHow large an image can you rolo??
10:55:10BagderSchnueff: no, the unscrambling is very fast
10:57:42BagderearHurts: I don't know, but basicly there's no limit except the 2mb ram
10:58:25Bagder"basicly" because of course the code that copies the data needs to fit too etc
10:58:50earHurtsBut those 2 megs is all the ram there is? There's none hidden on the chip?
10:59:11Bagderthere's 4K internal ram
10:59:37Bagderbut that's all, yes
11:01:22earHurtsHmm. 5000 * 128 = 10000 * 64 = 640000 = .7 mb
11:02:04earHurtsHow easily can the mpeg memory buffer be dynamically resized?
11:02:22Bagdernot easy at all
11:02:38Bagderit really needs to be one huge chunk
11:02:48Bagderor we'll get things very complicated
11:03:16earHurtsThere's not a pointer to the start and a pointer to the end, like a C streambuf?
11:03:34Bagderthere is
11:03:53Bagderbut the area is contiguos, from the start all the way to the end
11:04:10Bagderwhich makes it hard to allocate dynamicly
11:04:36earHurtsYeah, but could one bump the pointer to start by x bytes, to (temporarirly) shrink the buffer, and grab some ,e,ory?
11:05:02Bagderif you stop the music and flush the area, then yes
11:05:07webmindwow.. max files in dir has become an option? cool :)
11:05:39earHurtsWhat if I, uh, wait umntil the music being read is past the point I want to bump to?
11:06:17webmindproblem.. i got write failed when i want to write my config ?
11:06:44BagderearHurts: the buffer is used in a circular fashion
11:07:16webmindanyone knowhow i can fix this ?
11:07:22earHurtsSo i aassumed. If I bump the start pointer, won't the new start be used when it wraps? I know this will mean more frequent disk reads.
11:07:39earHurtswebmind: you got space on the drive?
11:07:49webmindearHurts, 36gb of them
11:08:03BagderearHurts: the point is that it might already have "wrapped"
11:08:33Bagderimaging playing at position Y, and we have data filled to position X: |−−−−−−X−−−−−−-Y−−−−−−-|
11:08:55earHurtsYeah, I can't advance the start pointer until I know the memory has been read into the decoder. sure.
11:09:21Bagderbut of course, given some careful considerations it can be done
11:09:26earHurtsAnd I need to tell the disk to re-read.
11:09:33earHurtsOr do a memcpy.
11:09:41Bagderand we have a function for re-reading
11:10:25earHurtsAnd it uses the start and end pointers, so it'll do the right thing?
11:10:56Bagderthe start and end pointers are set "dynamicly" at startup so everything must use them
11:11:37Bagderthat's how the 400-limit no longer is fixed
11:12:23earHurtsWhat I want is a enough memory to do what the Ipod does: allow songs to be found by Artist/album/Genre/Composer/Title
11:13:03Bagder"to be found" you mean by a search or when browsing?
11:13:30earHurtsI meana simple database. A bunch of red-black trees.
11:14:38Bagderseveral people have requested and worked on a feature like that
11:14:54earHurtsWhat's been the general direction?
11:14:56Bagderso it would be appreciated, I think
11:15:13Bagderthey've mainly worked on getting a search feature
11:15:19Bagderto allow searching for a given song
11:15:50earHurtsI think a display of all song titles alphabetical is preferrable to a search per se.
11:16:08BagderI think I agree
11:16:33earHurtsAgain, if we have 5000 songs, and a title size of say 128, that's at least .7 mb
11:17:06Bagderso you should optimize on as little as possible needed in memory
11:17:09earHurtsAssume the average title is only 50 chars, it's stilll .3 mb
11:17:13Bagderyou can't design it to have it all in mem
11:17:23Bagderthere are 80GB disks
11:17:31Bagderit easily makes 14000 songs
11:17:42earHurtsYeah. But disk access is slow, and coslty in battery.
11:18:43Bagderwell, in that case I'd recommend you shut off the music completely and use the full mp3 buffer for your purpose
11:19:19earHurtsYeah, but that's annoying isn't it?
11:19:35Bagderprobably, if you wanna use it to browse around for files to listen to...
11:21:27earHurtsoh for satan, my ear hurts.
11:21:42earHurtsDo they say "oh for satan" in sweden?
11:22:12Bagderyes! but with "fr"
11:22:25earHurtsNo kidding?
11:22:47Bagderits the truth
11:23:04earHurtsI had this Danish girlfriend who said it all the time.
11:23:20Bagderdanish and swedish are similar
11:23:23earHurtsI found it very, uh, odd.
11:23:39 Join quelsaruk [20] (
11:24:44Bagderquelsaruk: its not morning anymore, its lunch
11:24:57Bagderand we are in the same time zone! ;-)
11:25:03quelsarukit's coffee time
11:25:12*Bagder eats an apple
11:25:29quelsarukhow can you have lunch at 11:30?
11:25:38webmindhmm. if i want to write a config file i need a .rockbox dir i guess right ?
11:25:41Bagderwell, almost lunch then
11:25:46Bagderwebmind: yes
11:25:51webmindBagder, heh :)
11:26:40quelsarukBagder: you'd better not come here to live... we have our meal at 14:30 or even later
11:27:02webmindhmm. rockbox manual not available in html ?
11:27:12Bagderwebmind: nope
11:27:30earHurtsMAke one, webmind
11:27:46Bagderthe fact that we at least HAVE one is good enough for me ;-)
11:28:08quelsarukBagder: what about that website that converted openoffice to html?
11:28:09*Bagder slaps the "manual hero" sticker on quelsaruk's forhead
11:28:28Bagderyeah, that's left to examin close
11:28:40quelsarukwith care Bagder :P
11:32:55quelsarukBagder: 2.1 is still far in time, no?
11:33:24Bagderwell, we haven't even talked about it for real yet
11:34:24quelsarukbecause i have to update the manual, and i will be away for a month.. more or less :)
11:34:56quelsarukexams and so on
11:36:09 Nick quelsaruk is now known as quel|study (
11:36:14quel|studycu l8r
11:38:39 Quit TotMacher (Read error: 54 (Connection reset by peer))
11:47:05***Saving seen data "./dancer.seen"
11:55:30webmindearHurts, uhm.. pdf2html conversion not that great :)
12:12:48 Quit Bagder ("")
12:17:44 Quit earHurts ("Leaving")
12:32:01 Join LinusN [200] (
12:32:02 Quit ken0 (Read error: 54 (Connection reset by peer))
12:32:17quel|studymorning LinusN :)
12:32:23quel|studythx for your help yesterday
12:32:30LinusNyou're welcome
12:33:10quel|studybtw, what happened with that 'cold hearted' version?
12:33:43LinusNforgot about it :-)
12:37:25LinusNso many projects, so little time :-)
12:44:50 Join ken0 [0] (
13:18:09LinusNquel|study: regarding the manual, i have updated it a lot myself, mainly template stuff
13:18:25LinusNthe template was lost when converting from Word to Openwriter
13:47:08***Saving seen data "./dancer.seen"
13:59:48quel|studythx LinusN :)
14:00:32quel|studyi'd like to have a look at it, if possible :)
14:05:38dwihnoLinusN: How hard do you think it is to flash rockbox?
14:07:01 Nick quel|study is now known as quel|lunch (
14:07:10quel|lunchcu later guys
14:14:50 Join TotMacher [0] (
14:17:16 Join ken0__ [0] (
14:17:16 Quit ken0 (Read error: 54 (Connection reset by peer))
14:24:13 Join kassoulet [0] (
14:25:54LinusNdwihno: not very hard
14:26:33dwihnoLinusN: And the risk involved?
14:27:00LinusNwell, you need to make sure that you can restore the original contents of course
14:27:30dwihnoSo how would you commence the flashing?
14:27:39dwihnoVia the serial mod, or just a special ajz?
14:34:14LinusNwe are examining the boot ROM (inside the CPU) to see if it has a flashing utility
14:34:38LinusNor if it at least has a code download feature
14:34:45dwihnoNeato. I'm keeping my fingers crossed. Would be excellent to run RomBox :)
14:34:51LinusNso we can download a flash programmer
14:44:07 Join nelliep [0] (
14:44:53nelliepLinusN: Could you quickly answer a question for me?
14:47:00nelliepor anyone else come to that. It's about the sleep timer
14:52:41dwihnoshoot's just I was wondering why the sleep timer just freezes the song and not switch off the unit
14:56:24nelliepOK...look I've really got to go so if you answer then I'll look it up in the IRC Log
14:56:42nelliepthanks...great job by the way!
14:56:56 Quit nelliep ("IceChat IRC Client - Download at")
14:57:17 Join Guest [0] (
14:59:36 Quit Guest (Client Quit)
15:08:21*Stevie[FP] is back from [gone] [gone 13hrs 33mins 8secs] [KS]
15:08:25*Stevie[FP] yawns
15:11:31Stevie[FP]I bet nobody's here
15:12:11Stevie[FP]would ata_spinup_time be the amount of time it took to spin up the drive last time?
15:16:59 Quit TotMacher ()
15:17:05 Join TotMacher [0] (
15:23:49LinusNStevie[FP]: yes
15:24:07Stevie[FP]I wonder if we shouldn't do some sort of running average
15:24:26Stevie[FP]to be slightly more accurate...
15:24:46LinusNnelliep: it stops the song if the device is connected to a charger, because then we can't turn it off. it's a hardware limitation
15:25:08LinusNStevie[FP]: des it differ much for you?
15:25:30Stevie[FP]but I wonder
15:25:43Stevie[FP]has anyone checked the spinup time after the end of a 'red led dead' scenario?
15:28:43Stevie[FP]hmm, we don't have anything akin to a WPS for the recording screen, do we?
15:31:01LinusNStevie[FP]: no WRS, no
15:39:45Stevie[FP]any guesses as to the average amount of time it takes to read/write MAS memory?
15:40:11LinusNlong time since i worked on that
15:40:19LinusNi'd guess 1-2ms
15:40:25Stevie[FP]that's not too shabby
15:40:45*Stevie[FP] tries something
15:40:58dwihno\o/ Stevie \o/
15:41:03dwihnoStevie Wonder? :)
15:42:11Stevie[FP]... no.
15:43:59Stevie[FP]hm, we don't seem to enter USB mode at the Recording screen
15:44:29Stevie[FP]thank god for UnxUtils
15:47:09***Saving seen data "./dancer.seen"
15:55:12Stevie[FP]hm, I wonder what these unused commands do in the MAS
15:55:50LinusNStevie[FP]: no, we don't enter USB mode in the rec screen, partly for safety reasons
15:56:48LinusNStevie[FP]: "unused commands"?
16:08:20Stevie[FP]i'm still having trouble understanding how this device is accessed (the datasheet is not very clear)
16:08:32Stevie[FP]but even if I ignore the fact that we only use one nybble for the command
16:09:01Stevie[FP]the 'command code'
16:09:10Stevie[FP]the 3785f supposedly has 4 run commands (0-3)
16:09:23Stevie[FP]and it skips commands 4, 8, and 9
16:11:17 Quit TotMacher ()
16:11:25 Join TotMacher [0] (
16:14:01Stevie[FP]ok so the MAS is at i2c address 0x3C
16:14:11Stevie[FP]we read from 0x3C and write to 0x3D
16:16:09Stevie[FP]this is way more complex than the EEPROM on our boards
16:20:30Stevie[FP]err, swap read&write there
16:25:26LinusNStevie[FP]: it is device address 0x3c. the LSB is the R/W bit
16:25:36 Nick quel|lunch is now known as quelsaruk (
16:28:26LinusNthere are four 'run' commands because the highest 2 address bits are in the command nibble
16:32:01Stevie[FP]man, I hate how these guys decided to describe this stuff
16:34:23Stevie[FP]ok, address >> 8
16:34:50Stevie[FP]so that gives us 0x0000 thru 0x3FFF
16:35:38Stevie[FP]16 kilowords, hmm
16:36:08 Join nelliep [0] (
16:36:13Stevie[FP]now that's odd
16:37:48 Nick dwihno is now known as dw|gone (dwihno@
16:38:55nelliepLinusN: Thanks for the reply. Yes it works just fine without the charger.
16:48:03Stevie[FP]is there any major reason why the recording screen uses FONT_SYSFIXED?
16:54:09quelsarukummm... because LinusN wanted that everyone could see all the info
16:54:28quelsarukif you have a 10x10 font.. you'd miss some info :)
17:00:42 Join hardeep [0] (1098@
17:10:26Stevie[FP]but I have a 5x7 font
17:10:44Stevie[FP]and frankly, a 10x10 font is nuts on that tiny little screen
17:11:00*Stevie[FP] makes a note to add a patch for that
17:12:32 Quit awy[baer]_ (Remote closed the connection)
17:16:20 Join nelliep_ [0] (
17:21:45 Join kassoolet [0] (
17:23:12 Quit nelliep_ ("IceChat IRC Client - Download at")
17:30:28 Quit nelliep (Read error: 113 (No route to host))
17:39:49 Quit kassoulet (Connection timed out)
17:46:38 Join elinenbe [0] (
17:47:10***Saving seen data "./dancer.seen"
17:48:38 Join awy[baer] [0] (
17:49:34Stevie[FP]hey, there's a bug here!
17:52:02Stevie[FP]it calls mas_devread()
17:52:36 Quit ken0__ (Read error: 104 (Connection reset by peer))
17:52:52Stevie[FP]mas_devread() truncates the value to 20 bits, since that's what it ought to do for most requests (registers and memory)
17:53:03Stevie[FP]but the version command returns a full 32-bit value
17:53:06 Join ken0__ [0] (
18:01:26Stevie[FP]anybody have any guesses about these unnamed commands?
18:09:22*Stevie[FP] drops a pin
18:10:32hardeepshhhh, we're trying to sleep in here
18:11:04*Stevie[FP] ponders
18:26:49 Quit matsl ("Client Exiting")
18:27:01 Join Guest [0] (
18:27:03 Quit Guest (Client Quit)
18:27:11 Join tracktheripper [0] (
18:27:39*Stevie[FP] is away [food n stuff] [KS-MsgLog Off]
18:27:41tracktheripperhi Hardeep
18:39:30hardeepanyone up for alpha-testing a new feature?
18:42:29tracktheripperYES me!!!!
18:42:34tracktheripperill test it :-)
18:42:54hardeepdon't even care what it is? =)
18:43:00hardeepwhat device do you have?
18:43:09tracktherippera Recorder 10
18:44:07elinenbehardeep: what is the new feature?
18:44:35hardeepit's the dynamic playlist stuff
18:44:46hardeepinsert/queue/append tracks
18:44:54hardeepalso, save dynamic playlists
18:45:01elinenbehardeep: you rock!
18:45:20elinenbeso now you can queue up a song to play next or at the end of the queue?
18:45:39hardeepat end of playlist
18:46:00tracktheripperhang onn
18:46:02hardeepalthough adding code to put it at end of queue is also pretty easy... i'll put that on my todo
18:46:12hardeepbtw, diff is at:
18:46:40elinenbehardeep: I will go over it with a fine toothed comb and tell you what is missing :)
18:47:22hardeepthe big thing missing right now is inserting playlists
18:48:33 Join _aLF [0] (
18:50:06tracktheripperhardeep what about queing songs BEFORE playback?
18:50:37hardeeptracktheripper: do you mean before we begin playing a playlist?
18:51:06tracktheripperi mean just queue up a set of songs BEFORE playback rather than DURING playback
18:51:36elinenbehardeep: it does not work on directories −− I am unable to queue up a directory
18:52:07hardeepelinenbe: yeah, directories/playlists are not in yet... that's the next thing i'm going to work
18:52:30hardeeptracktheripper: that would also be possible, it's not in there yet though
18:52:44tracktheripperThe Creative Labs Nomad enables you to do that
18:54:03_aLFhardeep> you made a patch about playlist ?
18:56:00hardeep_aLF: yeah, dynamic playlists...
18:56:11elinenbewhat are the differences between queue/queue append/insert/append ?
18:56:12hardeep(still a work in progress
18:56:55_aLFhardeep> is there a ajz for FM recorder ?
18:57:03hardeepelinenbe: queued songs are deleted after they are played and they aren't included in the playlist when you request a save
18:57:22hardeepqueue/insert add the track after the current playing song
18:57:30hardeepqueue append/append add it at end of playlist
18:57:48elinenbeah −− now it makes more sense.
18:58:15hardeepbtw, the save dynamic playlist option is on the main menu Playlist options->save dynamic playlist
18:58:30hardeepyou need to specify a valid absolute path right now
18:58:47tracktheripperhardeep can u save playlist in directory of ur choice?
18:59:53hardeeptracktheripper: so long as you specify the correct full path
19:00:02elinenbeall I think is missing now is a playlist editor (for doing static playlists, and deleting songs) and the ability to queu up directories or playlists.
19:01:19hardeepelinenbe: agreed, those are the two features I want too =)
19:01:41tracktheripperwhat about an option to reshuffle the playlist when it repeats in repeat mode?
19:02:04hardeeptracktheripper: that's unrelated to this and there's already an open feature request for that
19:02:16tracktheripperbut its still a playlist option
19:07:45elinenbehardeep: I like the progress.
19:08:10elinenbehardeep: when you save a playlist, it will only save the songs that have been inserted/appended?
19:08:33hardeepelinenbe: no, all the songs in the current dynamic playlist
19:09:22hardeepplease let me know if you find any bugs, i went a little overboard on the playlist changes and there are bound to be issues I missed
19:11:35 Quit tracktheripper ("Leaving")
19:20:09 Nick quelsaruk is now known as quel|out (
19:24:56quel|outbtw, is it possible to 'continue on target directory after queue'?
19:26:01quel|outi mean, you queue a file, and when you finish playing your current file, it starts that file (like normal queue), BUT, when that song ends, rockbox continues with next song on *that* directory.
19:26:25quel|outmore or less like archos queue :)
19:26:52elinenbehardeep: so, if you queue up a few songs and then instert a few songs, and then let a few play through and then save it, what happens?
19:27:11elinenbealso, what happens with shuffle and repeat?
19:31:18 Quit kassoolet (Remote closed the connection)
19:32:02hardeepelinenbe: save will save everything except the queued songs... no matter what's played
19:32:40hardeepif you unshuffle, all queued/inserted songs will be put at end of list
19:32:53hardeepa re-shuffle will shuffle everything randomly
19:33:39hardeepactually, now that I think about it, queued songs should probably always be the next song to play, irregardless of shuffle
19:33:52hardeepwhich is how they work on the existing implementation
19:35:29hardeepquel|out: there's an open feature request for that... personally, i wouldn't enable that feature. :)
19:36:40hardeepquel|out: actually, would it be sufficient if you just queue the entire directory? that's something i'm going to be adding soon
19:42:31 Quit hardeep ("[BX] Reserve your copy of BitchX-1.0c19 for Windows CE today!")
19:47:13***Saving seen data "./dancer.seen"
19:51:18 Join groovingandi [0] (
19:52:20 Join josh966 [0] (
20:03:08 Quit ken0__ (Read error: 110 (Connection timed out))
20:22:33*Stevie[FP] is back from [food n stuff] [gone 1hr 54mins 54secs] [KS]
20:29:03Stevie[FP]anybody even here?
20:31:56Stevie[FP]haven't seen you before
20:32:17groovingandii haven't been here too often yet
20:32:33Stevie[FP]well I'm Steve
20:32:47 Join [keno] [0] (
20:32:56Stevie[FP]do you do any development?
20:33:11quel|outgoing home
20:33:15quel|outcu tomorrow
20:33:19 Part quel|out (""Eve's Avatar - Avatar's Eve"")
20:33:40groovingandihi steve, i'm andi, i don't code yet but plan to
20:38:57groovingandiI'm sorry, i have to leave now, bye!
20:39:06 Part groovingandi
20:40:36*Stevie[FP] ponders
20:42:14*mecraw bites
20:42:19mecrawPondering what?
20:44:32Stevie[FP]if you examine the list of command codes for the MAS3785F (or 3875 or 3857 or whatever tf it is)
20:44:41Stevie[FP]the command code is one nybble
20:45:02Stevie[FP]and three values are missing from the table
20:45:06Stevie[FP]4, 8, and 9
20:45:15Stevie[FP]so I'm trying to find out wtf they do
20:47:47mecrawwhat's a nibble? (i'm only an application developer)
20:47:47 Quit [keno] (Read error: 54 (Connection reset by peer))
20:48:18Stevie[FP]a nybble is four bits (i.e. half a byte)
20:49:45Stevie[FP]hm, interesting
20:51:50 Join hardeep [0] (1098@
20:52:03Stevie[FP]wb hardeep
20:52:51Stevie[FP]how goes it?
20:53:27hardeepit goes
21:04:54 Quit josh966 ("ChatZilla 0.8.26 [Mozilla rv:1.4b/20030514]")
21:47:14***Saving seen data "./dancer.seen"
22:05:04Stevie[FP]who wrote the code that uses the MAS?
22:05:15Stevie[FP]who wrote the code that uses the MAS 3507?
22:06:20mecrawmost likely Linus
22:06:27Stevie[FP]this mas_run stuff
22:06:30*Stevie[FP] rubs his head
22:13:37 Join kassoulet [0] (
22:19:12 Join tracktheripper [0] (
22:29:41tracktheripperhello steve
22:29:51tracktheripperjust came back from the boozer
22:32:59LinusNStevie[FP]: i wrote pretty much all MAS code
22:34:15tracktheripperabout time linus wrote some mas code :-)
22:35:07Stevie[FP]LinusN: do you know anything about this mas_run() stuff?
22:35:38LinusNwhat do you want to know?
22:35:42Stevie[FP]blah, this page is in swedish =/ I think it is, at least
22:35:50Stevie[FP]just about everything? :P
22:35:56Stevie[FP]there's three calls to mas_run()
22:36:18Stevie[FP]two are hardcoded to 0x475 which the spec sheet says is something about PLL control
22:36:25Stevie[FP]the other is 0xfcb or something
22:36:55LinusNit is a call to DSP code that changes the PLL settings
22:37:40LinusNthere are different versions of the mas, and the DSP code that does the job is at different addresses in the different versions
22:38:08LinusNi don't know anything about it, i just do what the data sheets tell me
22:38:28tracktherippertry paying Micronas for the data sheets
22:38:43Stevie[FP]are there any DSPs where we don't need an NDA to find, say, the opcodes supported?
22:39:27LinusNStevie[FP]: DSP's are often pretty closed
22:39:37Stevie[FP]that doesn't surprise me
22:39:50Stevie[FP]but all we need to do is find *one* sufficiently open DSP decoder
22:39:54Stevie[FP]but all we need to do is find *one* sufficiently open MP3 decoder
22:39:58LinusNthe PLL settings took quite some detective work to find out
22:40:34Stevie[FP]well, let's say we were designing a player
22:40:50tracktheripperas if micronas will believe that :-)
22:40:51Stevie[FP]We know that the opcodes of product XYZ can provide the support we need
22:41:16tracktheripperhi linus
22:41:32Stevie[FP]How do we even KNOW if the 3587's instruction set would support what we need?
22:42:01Stevie[FP]fact is −− the only information I have to go on indicates that it Ogg has a serious chance at not working
22:42:41LinusNStevie[FP]: fact is, the MAS is an MP3 codec, it just happens to be implemented with a DSP
22:42:56Stevie[FP]that masc 3500 guy mentioned a difficulty implementing lookup tables, which I believe Ogg uses a lot of?
22:43:25LinusNwe KNOW that the MAS can do MP3, that's what the data sheets tell us
22:43:29tracktheripperit cant play wma
22:43:31tracktheripperor ogg
22:43:32LinusNTremor uses >100kbytes of tables
22:43:32tracktheripperjust mp3
22:43:48LinusNthe integer ogg implementation
22:43:56Stevie[FP]that's a lot of tables :-o
22:44:42Stevie[FP]okay, there was that other thing
22:44:58Stevie[FP]that mp3 player
22:45:53LinusNi don't follow you
22:46:35Stevie[FP]there was an mp3 player you mentioned with similar hardware
22:46:46LinusNyes, the m3po
22:46:50Stevie[FP]i have a few .bin files from a firmware update
22:46:55tracktherippera terretack device
22:46:58Stevie[FP]does it use the same cpu?
22:47:13Stevie[FP]ok, sh-1
22:47:17LinusNSH1 + MAS3507
22:47:19Stevie[FP]any schematics?
22:47:43Stevie[FP]how much do the damn things cost?
22:47:55LinusNi don't know
22:48:00tracktheripperhi linusN
22:48:13Stevie[FP]wait... how do we know it uses similar hardware?
22:48:18tracktheripperwhy im i being ignored?
22:51:27 Join [IDC]Dragon [0] (
22:51:41 Nick LinusN is now known as LinusN|food (
22:51:46 Quit [IDC]Dragon (Client Quit)
22:59:34tracktheripper"the player may display album covers if you squeeze an entire cover into one of its character cells. If u like u could display multiple covers side by side a.k.a. MusicMatch Library"
23:11:20 Join gnc [0] (crash@
23:13:04gncanyone alive here at the moment?
23:19:33 Quit gnc ("karsts plombiira piiraadzinju burbonaa meerceetu zhenjshenja saknju siirups")
23:19:52Stevie[FP]i can't find any pricing information on this 'm3po'
23:27:25*Stevie[FP] is away [gone] [KS-MsgLog Off]
23:30:27 Quit tracktheripper ("Leaving")
23:47:09 Join DJBaz [0] (
23:47:15***Saving seen data "./dancer.seen"
23:56:43 Join Jet8810 [0] (

Previous day | Next day