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 2005-02-03

00:08:12 Quit Hadaka (Read error: 60 (Operation timed out))
00:10:20amiconnPlayer is now closed, still working. Phew! :)
00:10:25 Join Naked [0] (
00:10:29 Nick Naked is now known as Hadaka (
00:15:38 Nick jyp is now known as TheProphet (
00:16:22 Nick Strath is now known as Gmini (
00:16:27*HCl yawns and decides he's hungry
00:17:18***Saving seen data "./dancer.seen"
00:18:16 Nick Gmini is now known as Strath (
00:18:43 Nick TheProphet is now known as jyp (
00:21:08 Nick midk__ is now known as midk (
00:21:26 Nick ripnetUK is now known as ripnet (
00:31:02*amiconn is annoyed by usb 1.1 speed
00:31:07HCl :P
00:31:18HClpatience is a virtue
00:31:52Quelsarukamiconn: that's one of the reasons why i bought a usb2.0 V1 recorder when my old recorder6 began to die....
00:32:45amiconnI'm currently copying some music over to the player from my recorder 80 (usb 2.0). Player is always usb 1.1...
00:33:27amiconnI have to decide what to put on the player. I can't fit all tracks :(
00:34:03Quelsaruki had to cut half my music
00:34:32amiconnI want to use the player a bit, hoping to shed some light on the rld problem. The player has that Hitachi DK23DA disk
00:34:53Eodunis there some way to make it support 2.0? (stupid question, i imagine that is hard-coded into a rom chip...)
00:35:11Quelsarukchange the chip
00:35:15amiconnEodun: Nope. The usb bridge can only do 1.1
00:35:47amiconnQuelsaruk: I doubt that it's that simple. Probably different pinout etc...
00:36:09Quelsarukamiconn: i think it's easier to get a new recorder :)
00:36:29amiconnUnfortunately the disk is only a 10 GB, while my collection is around 15 GB...
00:37:06Quelsarukamiconn: stupid question/idea
00:37:18Quelsarukwhy don't you plug the drive into a external case
00:37:34Quelsarukand later on plug again into the player?
00:37:40amiconnErm, because I don't have one?
00:37:54Quelsarukgood point
00:38:44amiconnWell, I could copy the tracks over from the recorder to the laptop, then swap the other drive into the recorder, and copy back.
00:38:57Eodunhe, my collection is about 25Gb
00:38:59Quelsarukyes, you could
00:39:10*HCl 's collection is 13gb..
00:39:12Eodunbut it's been ages since I heard the first files...
00:39:35amiconnBut I don't want to mess around with the recorder too much. It's my main box.
00:39:40Quelsaruk136€ for a pair of SAPI5 spanish voices??
00:39:44Quelsarukthat's a lot!
00:40:06Eodunhave you seen AT&T offer?
00:40:29Eodunyou have to buy first the basic pack and then expand it with non-english voices
00:40:31Quelsarukbut AT&T are latin american
00:42:16amiconnQuelsaruk: I bought TextAloud; together with the AT&T base pack (enigne + 2 us english voices) this was 42,- €
00:42:35amiconnThen every additional voice cost 30,- €
00:42:59Quelsaruki understand that for commercial use, those prices are not too expensive
00:43:08Quelsarukbut just for rockbox....
00:43:13Eodunor for blind people
00:43:14HCli'm confused
00:43:16HClwhats sapi5?
00:43:24EodunTTS engine
00:43:29HClwhats tts XD
00:43:35Eodunmicrosoft's technology
00:43:40EodunText To Speech
00:43:47HClyou want that in rockbox?
00:43:52HCldoesn't that take too much space o.o;
00:43:55Quelsarukrockbox can speak
00:43:56Eodunwe HAVE that in rockbox
00:44:03HCl :P
00:44:05Eodunwell, not really lol
00:44:38QuelsarukHCl: we make a voice file in the computer, copy that to rockbox
00:44:56Quelsarukand rockbox can talk the menus
00:45:00Quelsarukthe folders, or the files
00:45:37amiconnQuelsaruk: Speaking about space - those at&t voices take around 600 MB per voice on the pc
00:46:04Quelsarukthat's why they are so nice
00:46:35amiconnWell, there are other TTS engines that sound rather nice too, and only take around 1/10 of that space per voice
00:47:01Eodunit hurts
00:47:04HCl600mb is a lot...
00:47:28amiconnI'd like to get hold of the ScanSoft RealSpeak engine & voices. Supports many languages and sounds rather nice...
00:48:19*amiconn is undecided whether he should update cvs. Those many warnings don't look good...
00:50:52jypupdate & fix :P
00:50:59Quelsarukthat company had nice spanish
00:51:08Quelsaruki tested spanish and japonese
00:51:13Quelsarukand both were great
00:51:30 Join bagawk [0] (~Lee@bagawk.user)
00:52:41amiconnjyp: I didn't cause that mess...
00:52:56jypI know, just kidding you ;)
00:53:32Eodunehm 1 thing: how do I put my photo in my profile?
00:53:40amiconnBut I always prefer a clean build table for when the daily builds are built (6 a.m.)
00:54:49 Quit bagawk ("Leaving")
00:55:15 Join bagawk_ [0] (~Lee@bagawk.user)
01:00:30HClits compiling :)
01:01:33Quelsarukamiconn: then, returning to that nasty issue
01:01:51Quelsarukwhat do you decide? Backward compatible, or clean fix?
01:03:04HCldoes anyone have a working copy of the bootloader?
01:03:08HCli'd like to compare some md5's
01:03:13amiconnI'd say a clean fix would be the way to go. Other languages may also have the same problem as spanish.
01:03:14 Quit jyp ("poof!")
01:05:01Quelsarukmy backward compatible idea is to keep LANG_PLAYINDICES_PLAYLIST, but just use it in the playindices() function. And create a new ID for LANG_PLAYLIST to use for voice and anywhere you require to use it. i.e. the other 3 current uses.
01:05:16Quelsarukoutdated languages won't have that id
01:05:21Quelsarukbut will use english one
01:05:26Quelsarukthat is, Playlist.
01:06:03Quelsarukthe Clean fix is merging both _LANG_PLAYINDICES_XXX IDs into only one
01:06:31Quelsarukbecause is nonsense to have 2 lines if the splash window can do that right now.
01:07:10amiconnI think we should just add LANG_PLAYLIST, and let both LANG_PLAYINDICES_PLAYLIST ids alone. This way, all non-updated languages would still display that splash correctly.
01:07:47Quelsarukand add a *WARNING* in the desc fields
01:07:52Quelsarukfor those 2 lines
01:08:04Quelsarukto say that is only for that particular function.
01:08:29amiconnImho a language cleanup (including bumping the language version, throwing out all deprecated stuff, combining everything like the playindices thing that is used by splash() now etc.) might be due
01:09:11Quelsarukmaybe for 2.5
01:09:13amiconnOf course, this would render all languages useless which don't get updated
01:09:35Quelsarukthat need to be done in an official release
01:09:47Quelsarukdoing a lot of work
01:09:48Quelsarukthat is
01:10:11Quelsarukediting those anguages to delete those id deprectated, and those merged...
01:10:39HClokay, who has an iriver and is feeling brave? :P
01:11:00*Quelsaruk points to HCl
01:11:06HClno :P i'm a wuss xD
01:11:24HClwell, not really, but, i don't plan on testing this just yet :P
01:11:26crash_HCl: i have on, you didnt tell the number ;)
01:11:28*HCl prods webmind with a stick
01:11:40HCl :P
01:11:48amiconnQuelsaruk: yup. I think such things should be done shortly before a release, to check whether everything still works as intended. Maybe this should be done for 3.0
01:12:10Quelsarukit's an option
01:13:14EodunIMHO, if no one updates them and no one complains, it's because nobody uses them
01:13:29Eodunso why not deleting outdated langs?
01:14:07Quelsaruk'cause we can't
01:14:19Quelsarukis you want for example to update polish
01:14:33Quelsarukit's easier if you have 40 id translated...
01:15:24amiconn14 out of our 24 languages weren't updated for more than a year :-/
01:15:59*Quelsaruk looks to his feet
01:16:09amiconnOnly 4 were updated after the 2.3 release...
01:16:10preglowanyone know where the __assert mentioned in assert.h is?
01:17:00Eodunenglish, spanish, french, german I assume
01:17:20amiconnAlmost. English, spanish, swedish, german
01:17:30Quelsarukdon't forget our swedish friends
01:18:07Quelsarukenglish is updated because is the official language
01:18:13DMJC-Lanyone flashed a 140?
01:18:16Quelsarukgerman.. because amiconn is here...
01:18:18preglowDMJC-L: yes, linus
01:18:21amiconnCzech, french, and wallisertitsch were last updated for 2.3
01:18:29Quelsarukspanish, because i updated
01:18:32amiconnEverything else is way outdated...
01:18:34DMJC-Lnot 120
01:18:35preglowDMJC-L: no, then
01:18:42preglowDMJC-L: but there shouldn't be a difference
01:18:42Quelsarukand swedish, because we have a lot of swedish guys here
01:18:53preglowDMJC-L: only difference between 120 and 140 are the physical dimensions and the hard drive
01:19:22HClDMJC-L: i got a flash image.. if interested..
01:19:34amiconnQuelsaruk: Swedish is maintained by matsl, as it looks
01:19:34DMJC-Ldcc it
01:19:56HClactually, let me change it back a little to it being 100% original first
01:19:58 Quit bagawk_ (Read error: 110 (Connection timed out))
01:20:03HCli added a not to the button check
01:20:07HClso it'd load the default firmware first
01:20:12HCli dunno
01:20:15HClwhat do you prefer? o.o;
01:20:25DMJC-Liriver default
01:20:31preglowi really don't think anyone should flash the bootloader until the swedes have tested it
01:20:34HCllet me just double check things.
01:21:33HClthis i don't like.
01:21:49HCltitania:/home/hcl/rockbox/rockbox-devel# tools/mkboot ihp_120.bin boot/bootloade
01:21:52HClr.bin new.bin
01:21:55HClWrote 0x1fa7cb bytes in new.bin
01:21:59HCltitania:/home/hcl/rockbox/rockbox-devel# tools/descramble -iriver new.hex new2.bin
01:22:01Quelsarukamiconn: have you tried loquendo tts voices?
01:22:02HClModel iHP-120/iHP-140
01:22:05HClFile decoded correctly and all checksums matched!
01:22:07HClOutput file contains all headers and checksums
01:22:10HCltitania:/home/hcl/rockbox/rockbox-devel# diff new.bin new2.bin
01:22:12HClBinary files new.bin and new2.bin differ
01:22:19amiconnQuelsaruk: Nope.
01:22:23HCli forgot the scramble bit
01:22:24HClnm xD
01:22:34Quelsarukthey sound nice
01:22:38HCloh wait, no i didn't..
01:22:42HClaren't the files supposed to match
01:22:46HClonce scrambled and descrambled?
01:22:50HCl*tries with original firmware*
01:23:17HClguess not.
01:24:52HCltitania:/home/hcl/rockbox/rockbox-devel# tools/scramble -iriver new.bin new.hex
01:24:52HClModel iHP-120/iHP-140
01:24:52HClFile encoded successfully and checksum table built!
01:24:52DBUGEnqueued KICK HCl
01:24:52HCltitania:/home/hcl/rockbox/rockbox-devel# tools/descramble -iriver new.hex new2.bin
01:24:55HClModel iHP-120/iHP-140
01:24:58HClFile decoded correctly and all checksums matched!
01:25:00HClOutput file contains all headers and checksums
01:25:03HCltitania:/home/hcl/rockbox/rockbox-devel# diff new.bin new2.bin Binary files new.bin and new2.bin differ
01:25:06HCli don't really like that...
01:26:02DMJC-Lnice upload
01:26:10DMJC-Lwhat's connection is THAT?
01:26:17HCl100mbit up/down
01:26:29DMJC-Laww :(
01:26:35HClfrom a quad p3 550mhz xeon
01:26:37HClwith 2gb ram
01:26:40HCland a raid5 set
01:26:40HCl :P
01:26:44DMJC-L1.5/256 here heh
01:26:51Quelsaruk512/128 here
01:27:01Quelsarukspain is different
01:27:03Eodundo you work at NASA or what?
01:27:33HClsomething like that >.o
01:28:02DMJC-Lnorthern europe :p
01:28:26DMJC-Ldidn't they have protests demanding 100mbit?
01:28:36HClno, i haven't heard of those
01:28:44DMJC-Lsweden did I'm sure
01:28:45HClwe had protests against it
01:28:53DMJC-Lat least for 10 mbit
01:28:53HClbecause it'd make warez distribution too easy
01:28:56HClstupid people.
01:29:02HClbut we have an upload limit anyways
01:29:05HClof.. 15 gb.
01:29:07HCla week >.>
01:29:09amiconnI know what Bagder did to cause the warnings. His commit breaks all builds which don't use a mas 3587, i.e. player, Ondio SP, and all sims...
01:29:10DMJC-Lscrew warez
01:29:20DMJC-L100mbit is useful
01:29:28DMJC-Lstream movies from home
01:29:33DMJC-Lremote X
01:29:53HClnot to forget our filesharing network
01:29:56HClwe have running on 100mbit
01:30:08HCli can get pretty much any recent movie streaming to my xbox
01:30:11HCldirectly onto the tv
01:30:13HCloff our network
01:30:25HCldon't wet yourself xD
01:30:26Quelsarukamiconn: who does the change to fix the lang issue? are you going to do it or should i do it?
01:30:30DMJC-LI won't
01:30:31amiconnDMJC-L: Why would you need 100 MBit for remote gui? I do this often, and I only have 1024/128
01:30:35HClgood :P
01:30:49DMJC-Ltried fullscreen video on that remote gui?
01:30:56DMJC-Lor 3d stuff?
01:31:01HClhah o.o
01:31:08HClwhat person does fullscreen video on remote gui o.o;;;
01:31:10amiconnOf coz not, and I don't need that
01:31:52DMJC-Lheh you *think* you don't need it
01:32:00DMJC-Lwhen you actually do it it kicks ass
01:32:14HClno comment >.>
01:32:26HCl*is used to streaming divx off network into media player by wireless internet*
01:32:48DMJC-LI still need to build myself a tv box
01:32:55HCloh, did i mention we have a fully blown wireless network across a radius of a few square kilometres backed up by about 100 accesspoints? >.>
01:32:57DMJC-Lmy desktop does that very well already
01:33:07DMJC-LI need a digital tuner
01:33:08amiconnI don't play those 3d modern games, I don't even have a single one.
01:33:11HCli can drive to our local foodstore and have wireless internet >.>
01:33:15amiconnAnd I rarely watch video
01:33:35DMJC-Lwhereas I play doom3 and have anime on my machine..
01:33:38CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:33:38*HCl stops bragging about their network :X
01:33:42DMJC-Ldifferent uses
01:33:42 Quit mecraw (Connection reset by peer)
01:33:47HClwhat series do you got?
01:33:51EodunHCl, I'm thinking about moving to your country...
01:34:00DMJC-Lbleach, macross zero, elfin lied
01:34:04HClbleach is good
01:34:09HCli'm waiting for ep 17 :(
01:34:19HClelfin lied was a bit too sinister for my taste..
01:34:21HCli only saw ep 2
01:34:37DMJC-LI can't be bothered watching elfin lied
01:34:41DMJC-LI just have it heh
01:35:33HCloh, by the way, if you can't find a certain anime, its ok to ask me whether i have it :P as long as i'm not on my limit yet, and still have some to spare for my gf, i don't mind sending people stuff :p
01:36:01DMJC-Lmate of mine runs a 100mbit server in america
01:36:05HClokies :)
01:36:11DMJC-Lhe gets ANYTHING really fast heh
01:36:22HCl :p
01:36:37HCli'm still waiting for a good copy of daphne in the brilliant blue..
01:36:42DMJC-Lsince I have the fastest download connection out of his friends/himself, i get first priority
01:36:43HClif you happen to come across it, tell me :X
01:37:03HCli know how that goes, i'm usually reluctant to send stuff to people who're slower than 100k/s myself
01:37:42*HCl stares at new.hex and prods it gently.
01:37:50DMJC-L1.5mbit changes your net habits
01:38:04preglowi've got 100mbit
01:38:05HCltrust me, so does 100mbit xD
01:38:17HCli get annoyed when movies download with only 600k/s >.>
01:38:22preglow600k/s is nothing!
01:38:27preglowi demand 5 megs per second, minimum
01:38:34HCli pull 8mb/s from microsoft's site.. and our lan..
01:38:38preglowi want to be able to watch a movie at a minute's notice
01:38:47HClxboxes are great for that.
01:38:58DMJC-Lstream it off the server heh
01:38:58preglowlemme guess, you're on a uni line, yes? :P
01:39:07DMJC-Lxbox/mac mini are dumb ideas
01:39:08preglowso am i
01:39:14HClshared 1gbit at the moment..
01:39:15DMJC-Lthey need a digital tv tuner
01:39:30DMJC-LOR access to fast internet
01:39:37*HCl goes to play with his kitty cat
01:39:43HClcats are cute.
01:40:04*DMJC-L looks fir his seal point birman
01:40:22preglowcats rock
01:40:24preglowas opposed to dogs
01:40:49Quelsarukamiconn: did you read my last question to you?
01:41:20amiconnOoops, sorry. Was looking into Bagder's mess. Fix coming up...
01:41:33Quelsarukif you can't, i'll do
01:42:41amiconnI'd say fix it for english & spanish, I'll add german.
01:43:39EodunQuel, if you need help, just tell me
01:44:21amiconnDon't forget to remove the voice: from the old one - it's not needed then.
01:45:44QuelsarukLANG_PLAYLIST is a good ID for it, isn't it?
01:46:02Quelsarukbtw, hmmm
01:46:07Eodunoh, just 2 IDs
01:46:17Eodunyou won't need help for that :)
01:46:24Quelsaruknever messed with a ID with voice
01:46:56Quelsaruki suppose there's no problem with them
01:50:28HClDMJC-L: are you gonna try that file? (just out of sheer curiousity, so i know what to expect :P)
01:50:46 Quit hile (
01:51:02DMJC-Ldepends on what linus tells me
01:52:25HCli'm just, anxious for someone to try it, and i'm too much of a wuss to try it myself :P
01:52:31HClbut i guess it makes more sense to wait a little
01:52:36HClrockbox development seems to go fast
01:52:52NJoinhile [0] (
01:53:04HClshouldn't need to wait long before we have a loader declared somewhat stable
01:53:04 Quit hile (
01:53:09NJoinhile [0] (
01:58:24Quelsaruktesting changes
02:03:00*HCl yawns and sips his sake
02:04:48*HCl can't wait till rockbox runs well enough to start porting a gameboy emu
02:05:02HCldynarec is fun :x
02:05:32Quelsaruksubmiting changes
02:07:57amiconnOkay, my fix did fix all production targets, the simulator mess left as an exercise for Bagder ;-)
02:09:23Quelsarukyou don't touch tree.c or onplay.c, do you?
02:09:58amiconnNope, only mpeg.c
02:10:48preglowsake, noe bad
02:11:21preglowHCl: dynarec might not be essential for the gb emu
02:11:31preglowbut i don't know
02:11:39HClpreglow: i won't care xD
02:11:44HCli want to do dynarec darnit!
02:11:45HCl :
02:11:46HCl :P
02:11:48amiconnWhat cpu does the gameboy have
02:11:58amiconnWhat freq?
02:12:01HClno clue
02:12:08HCli've only done dos and n64 so far
02:12:18preglowor soemthing
02:12:27HCldecided there's not much point looking into gameboy while i can't compile anything
02:12:48amiconnThen it might be even possible to emulate a gameboy on the archos, only the display might be a problem
02:13:06preglowamiconn: how much ram does it have?
02:13:17amiconnThe archos? 2 MB
02:13:25HCl2 mb is *not* gonna be enough.
02:13:26preglowdamn, that's not much for an mp3 player
02:13:30HClat least
02:13:36HCli don't know how large gameboy roms are
02:13:44HClan interpreter *might* be possible
02:13:46preglownot large
02:14:04HClit might be possible to get an interpreter emu on archos
02:14:06amiconnHCl: Why? The Z80 itself can only address 64 KB, and a Z80 emu isn't that big
02:14:09HClbut you can forget about dynarec
02:14:12preglowbut fuck that, h120 is what's important :P
02:14:27HClamiconn: dynarec needs ram for its on the fly code generation
02:14:58amiconnWhy dynarec?
02:15:04preglowdynamic recompilation
02:15:36amiconnYeah, I know, I asked why? Z80 is sloo-oow, at least 4 cycles per instruction.
02:16:04preglowahh :-)
02:16:05DMJC-L32mb of ram on h120/140
02:16:13preglowbecause he wants to do dynarec
02:16:30 Quit edx (Read error: 60 (Operation timed out))
02:16:30amiconnThe archos has 11 MHz SH1. SH1 is risc, and does simple instructions in 1 clock cycle
02:17:00amiconnI know some Z80 asm, had a ZX Spectrum a looong time ago
02:17:18preglowit's not that far from a 6502, is it?
02:17:22***Saving seen data "./dancer.seen"
02:18:10amiconnI don't know much about 6502, only that it was used in the C64 and is 8 bit too...
02:18:21preglowc64 was 6510, but it's more or less the same
02:18:25HClbecause dynarec is cool when it works :P
02:18:32preglowtwo index registers, one accumulator, stack pointer, that'st hat
02:18:33*HCl knows from experience, n64 dynarec rocked :P
02:18:43HCltoo bad i never got to the compatibility of our interpreter core :/
02:19:25preglowwell, dynarec is a lot harder, so small wonder
02:19:51HCli dunno, i don't think it differs much...
02:19:59HClexcept the interrupts are much more tricky
02:20:14preglowwell, yes, and trapping memory writes?
02:20:22amiconnThere is a ZX Spectrum emulator for the Amiga (working interpretive, no dynarec). It runs about half realtime speed on a stock Amiga 500 (7 MHz 68000)
02:20:23*Eodun yawns
02:20:23preglowif that's needed much
02:20:25HClnot all games work well
02:20:31HClpreglow: was barely needed, in the n64, at least.
02:20:32DMJC-Lcould you cut off the ram chips and solder on more?
02:20:32DMJC-Lthat'd be kind of cool
02:20:42HClactually. i remember that we didn't have a dynarec implementation for that
02:20:49HClwe just wrote dynarec code that called the interpreter part
02:21:14preglowDMJC-L: yes, and risky
02:21:16*Eodun says good night everyone
02:21:38 Quit Eodun ("c ya!")
02:21:47HCli'll look into it
02:21:49HCland get it working :p
02:21:55HClinterpreter or dynarec
02:22:18preglowi'm struggling with mad, and it already seems it'll be a boring port, heh
02:22:25amiconnregisters missing and some instructions changes."
02:22:29amiconn"The CPU is a Z80 workalike running at 4.19 MHz. The CPU has several
02:22:37amiconnOops, swapped the lines
02:24:06HClwell, especially when you set it to black and white only
02:24:13HClso all color games will run in black and white mode
02:24:16HClit shouldn't take too much
02:25:43Quelsarukamiconn: commit: using deprecated info format strings. Convert your scripts to use the new argument format and remove '1's from your info file format strings <−−- is this something wrong? (i hope not)
02:26:07amiconnNo, everyone who commits something gets that
02:26:13amiconnIt's a server config issue
02:28:15DMJC-Lgameboy emulator..
02:28:32DMJC-Ltwo main codecs that are needed are mp3/ogg
02:28:40HClyea, thats what i'm gonna be trying to get to work :P
02:28:53DMJC-Lthen my existing media will work
02:28:56Quelsarukthen, i'm going to sleep
02:29:13Quelsarukgood night to all
02:30:00midknight Quelsaruk
02:30:12Quelsarukdidn't see you midk :)
02:31:03 Quit Quelsaruk ("I seek a life of honor and pride... that's why i use rockbox")
02:31:53 Quit preglow ("off")
02:38:41*HCl yawns.
02:39:44DMJC-Lso... as a guesstimate, how long till we see iriver's playing music?
02:40:59HCldon't look at me.
02:41:05HCli'm just waiting for a stable bootloader
02:41:13HCli'd almost say look at the checklist..
02:41:22HClaudio low level drivers haven't even been started on yet
02:42:18HClusb doesn't even work yet
02:42:41HClplus for the codecs the whole new api has to be done...
02:42:43HCli think
02:43:39HCltalking about music, i need some..
02:52:52*HCl wonders how many people are still awake
03:00:31 Join ashridah [0] (
03:43:09 Quit midk (Read error: 104 (Connection reset by peer))
03:43:37 Part amiconn
04:05:40 Join edx [0] (
04:07:05 Quit DMJC-L ("Leaving")
04:11:13 Join DMJC [0] (
04:16:42 Join shady`danky [0] (
04:17:08shady`dankyi have a question about a Archos Pm me to help
04:17:26***Saving seen data "./dancer.seen"
04:17:42 Quit elinenbe (Read error: 104 (Connection reset by peer))
04:19:50 Join elinenbe [0] (
04:30:45 Quit elinenbe (Read error: 54 (Connection reset by peer))
04:30:51 Join elinenbe [0] (
04:34:20 Quit shady`danky ()
04:37:14 Quit elinenbe (Read error: 104 (Connection reset by peer))
04:37:50 Join elinenbe [0] (
04:44:04 Join elinenbe_ [0] (
04:45:23 Quit elinenbe (Read error: 104 (Connection reset by peer))
04:45:24 Nick elinenbe_ is now known as elinenbe (
04:46:28 Join jorbond [0] (
04:56:31DMJCmmm.... oh man
04:56:38DMJCmy mouth is orgasming with taste
04:56:47DMJCbacon/eggs at 2:30pm
04:57:40 Quit elinenbe (Read error: 60 (Operation timed out))
05:03:24 Join webguest56 [0] (
05:03:42webguest56That is the sickest comment I've ever seen
05:03:45 Part webguest56
05:04:13 Join midk [0] (
05:27:54 Quit jorbond ("Leaving")
05:29:17 Join elinenbe [0] (
05:31:47ashridahi'm guessing he hasn't spent much time on irc. :)
05:48:07 Quit mbr (
05:48:52NJoinmbr [0] (
05:49:35 Quit elinenbe (Read error: 110 (Connection timed out))
05:55:27 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
06:07:08 Join elinenbe [0] (
06:17:30***Saving seen data "./dancer.seen"
07:13:57 Join elinenbe_ [0] (
07:14:12 Quit elinenbe (Read error: 104 (Connection reset by peer))
07:14:24 Nick elinenbe_ is now known as elinenbe (
07:27:58 Join LinusN [0] (
07:31:30 Quit ripnet (Read error: 110 (Connection timed out))
07:32:28 Quit MooMaunder (Read error: 110 (Connection timed out))
07:39:22 Join einhirn [0] (
07:46:37 Join elinenbe_ [0] (
07:47:05 Quit elinenbe (Read error: 104 (Connection reset by peer))
07:47:05 Nick elinenbe_ is now known as elinenbe (
08:00:32 Join elinenbe_ [0] (
08:00:45 Quit elinenbe (Read error: 60 (Operation timed out))
08:01:00 Nick elinenbe_ is now known as elinenbe (
08:08:23DEBUGEOF from server (Connection reset by peer) (snapshot: netstuff.c line 545)
08:08:23***Saving seen data "./dancer.seen"
08:08:24***Started Dancer V4.16
08:08:24***Connected to on port 6667
08:08:24***Logfile for #rockbox started
08:08:25***Server message 501: 'logbot :Unknown MODE flag'
08:08:25Mode"logbot :+i" by logbot
08:08:25 Join logbot [242] (
08:08:25 Join HCl [0] (
08:08:56Mode"#rockbox +t " by
08:08:56 Join mbr [0] (
08:08:56 Join edx [0] (
08:08:56 Join hile [0] (
08:08:56 Join Peter99 [0] (
08:08:56 Join QT [0] (as@area51.users.madwifi)
08:08:56 Join coob [0] (
08:08:56 Join courtc [0] (
08:08:56 Join Ka [0] (
08:08:56 Join izzy_ [0] (
08:08:56DEBUGParse error in OnTopic(from = "", line = "#rockbox :Rockbox - Coolest firmware around") (snapshot: server.c line 299)
08:08:59 Join Nibbler [0] (
08:08:59 Join Strath [0] (
08:09:11 Join DMJC [0] (
08:09:37 Join webmind [0] (
08:09:44 Join ze [0] (
08:09:53 Join elinenbe [0] (
08:09:53 Join Hadaka [0] (
08:09:53 Join crash_ [0] (
08:09:53 Join rasher_ [0] (
08:09:53 Join einhirn [0] (
08:09:53 Join pvh [0] (
08:09:53 Join Lynx_ [0] (HydraIRC@
08:09:53 Join gromit` [0] (~gromit`
08:10:05 Join Bagder [0] (
08:10:08 Join midk [0] (
08:10:43 Join desrt [0] (
08:10:55 Quit pvh ("Leaving")
08:12:17 Join dwihno [0] (~dw@
08:15:11 Quit einhirn (
08:15:11 Quit crash_ (
08:15:11 Quit Lynx_ (
08:15:11 Quit elinenbe (
08:15:11 Quit rasher_ (
08:15:11 Quit Hadaka (
08:15:11 Quit gromit` (
08:16:21NJoinelinenbe [0] (
08:16:21NJoinHadaka [0] (
08:16:21NJoincrash_ [0] (
08:16:21NJoinrasher_ [0] (
08:16:21NJoineinhirn [0] (
08:16:21NJoinLynx_ [0] (HydraIRC@
08:16:21NJoingromit` [0] (~gromit`
08:19:04 Quit elinenbe (Connection reset by peer)
08:19:14 Join LinusN [0] (
08:20:53 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
08:28:36 Join elinenbe [0] (
08:49:02DEBUGEOF from server (Connection timed out) (snapshot: netstuff.c line 545)
08:49:02***Saving seen data "./dancer.seen"
08:49:02***Started Dancer V4.16
08:49:02***Connected to on port 6667
08:49:02***Logfile for #rockbox started
08:49:04***Server message 501: 'logbot :Unknown MODE flag'
08:49:04Mode"logbot :+i" by logbot
08:49:04***Server message 477: 'logbot #rockbox :[freenode-info] please register your nickname...don't forget to auto-identify!'
08:49:04 Join logbot [242] (
08:49:04 Join Zagor [242] (
08:49:04 Join Nibbler [0] (
08:49:04 Join Bagder [0] (
08:49:04 Join dwihno [0] (~dw@
08:49:04 Join ze [0] (
08:49:04 Join desrt [0] (
08:49:04 Join HCl [0] (
08:49:04 Join webmind [0] (
08:49:04 Join Strath [0] (
08:49:04 Join LinusN [0] (
08:49:04 Join mbr [0] (
08:49:04 Join edx [0] (
08:49:04 Join hile [0] (
08:49:04 Join Peter99 [0] (
08:49:04 Join QT [0] (as@area51.users.madwifi)
08:49:04 Join coob [0] (
08:49:04 Join courtc [0] (
08:49:04 Join Ka [0] (
08:49:04 Join izzy_ [0] (
08:49:04 Join DMJC [0] (
08:49:05 Join Hadaka [0] (
08:49:05 Join crash_ [0] (
08:49:05 Join rasher_ [0] (
08:49:05 Join Lynx_ [0] (HydraIRC@
08:49:05 Join gromit` [0] (~gromit`
08:49:05 Join elinenbe [0] (
08:49:05 Join midk [0] (
08:54:03 Quit elinenbe (Connection timed out)
08:59:31 Join einhirn [0] (
09:00:28 Join elinenbe [0] (
09:14:06 Join ripnetUK [0] (
09:20:41BagderLinusN: tried my patch?
09:26:27 Join ashridah [0] (
09:26:37LinusNnot yet, had some sleep instead, for a change :-)
09:26:56Bagdergood choice
09:27:04Bagderbut not now while at work I hope! ;-)
09:27:08LinusNregarding the mp3end stuff, see how the stackend is defined in the link control file
09:27:36LinusNthe m68k linker doesn't want the . in front of the symbols
09:27:57LinusNjust define both .mp3end and mp3end
09:28:10LinusNlike with the stackend
09:28:51Bagderodd that they work so differently
09:29:09LinusNi guess it has to do with history and tradition
09:29:24Bagderprobably, yes
09:29:39LinusNgreat work, btw
09:29:57Bagderit really wasn't hard
09:30:07LinusNno, but it had to be done
09:33:49Bagderit was the _ that it doesn't like, but now it links fine
09:35:06LinusNah, silly me
09:35:20Bagder-rw-r−−r−− 1 daniel daniel 151608 Feb 3 09:35 rockbox.iriver
09:35:39Bagdercommit coming up
09:39:05BagderI wonder if this will start on the iriver now ;-)
09:39:17Bagderyou know if your main-init was done in the same order as the "real" one?
09:42:06 Quit elinenbe (Connection timed out)
09:43:18LinusNyes it was, but my working copy has a lot of extra port pin setup tricks that aren't yet committed to the proper files (power.c etc)
09:43:37LinusNso no, it won't start :-)
09:43:50Bagderbut perhaps your version will!
09:44:15LinusNafter resolving all the cvs conflicts... :-)
09:44:40Bagderyeah :-/
09:45:06LinusNi found the perfect connector for the iriver debug port btw
09:45:26Bagderthat's really good to hear
09:45:37Bagdercan't wait to try that on my unit ;-P
09:45:43LinusNso resurrecting a dead iriver will be pretty easy in the future
09:45:52QTLinusN: great work with iriver port! appreciated
09:45:54ripnetUKLinusN - cool. Does that mean that if I break my iriver, and solder a connector onto it, I can send it to you for bdm'ing???
09:46:30LinusNripnetUK: that's the plan, but i'll do the soldering if you can't find a connector
09:47:45LinusNQT: thanks
09:48:19QTwish i could help somehow but the only thing i can do is donate money $-)
09:48:43ripnetUKthats kind of you. Thanks :)
09:48:44LinusNQT: as soon as we have something running, you can help with testing
09:49:10QTyeah, if i do not fry my lovely player with all its 40GB content :-)
09:49:56QTi hang around here since a few days and do a /lastlog iriver from time to time. i saw that you committed the ATA part
09:50:46LinusNQT: you don't have a backup of the iriver content?
09:50:52QTripnetUK: i think that's least i can do if i wanna use your results soon
09:51:00QTLinusN: yeah, i do
09:51:04QTbut not with me
09:51:09*QT is travelling a lot
09:51:14ripnetUKqt - ?
09:51:26QTripnetUK: i mean donating....
09:51:40ripnetUKi see
09:52:12QTlike MisticRiver Jeff said it in "if you wanna see results and use enhanced FW, then donate"
09:56:28LinusNwe have had a great increase in donations since we announced the iriver port
10:01:05ripnetUKthe iRiver is a nice piece of kit... the Archos was just a /functional/ piece of kit :)
10:01:17dwihnoSo the rockbox '05 ibiza devcon is still on schedule? ;)
10:01:43dwihnoanyone of you got experience with usb pendrives?
10:09:54Lynx_dwihno: you plug them in, and they work ;)
10:12:21ashridahbooting from them is touch and go
10:12:27ashridahhighly motherboard dependent
10:12:46dwihnoashridah: touch and go?
10:12:57dwihnoashridah: I'm considering buying one and booting from it...
10:13:01QTripnetUK: absolutely! iriver H1x0 series is a great player!
10:13:15ashridahdwihno: as in 'may or may not work'
10:13:15dwihnoPerhaps linux, DOS, or whatever system you could fit on a 256/512/1024 meg stick
10:13:36dwihnoashridah: newer motherboards seem to work rather nicely
10:13:40ashridahbut generally windows, linux and macosX all deal with them fine, for relatively recent versions of all three
10:13:51ashridahdwihno: i still run into some that flake out, even when they claim to
10:16:01dwihnoashridah: as long as my new comp supports it... it will be for playing DOS games! :)
10:16:31dwihnoA new computer and pasta for a month ;D
10:16:32ashridahhope you don't have a usb keyboard then
10:16:46ashridahlast time i used a usb driver to access a usb drive under dos it disabled my usb keyboard the sodding thing
10:17:42QTLinusN: that is great i think that people are happy to donate in order to get enhanced FW
10:18:32ashridahit may or may not have been that particular driver, others may work better
10:20:46 Join elinenbe [0] (
10:26:48Bagderhm, 90 warnings ;-)
10:27:09 Join MooMaunder [0] (~me@
10:31:57 Join amiconn [0] (
10:37:33 Join [IDC]Dragon [0] (
10:37:52Lynx_Bagder: wasn't it less yesterday?
10:38:07Bagderbut I committed more iriver build stuff today
10:39:03[IDC]Dragonamiconn: r u there?
10:39:42Bagdernow rockbox builds "for real" on iriver
10:40:55amiconnBagder: Is there a reason why you didn't fix those warnings you caused yesterday? I think of ...
10:41:01DMJCis it safe to run yet?
10:41:20amiconn[IDC]Dragon: It seems so ;-)
10:41:34BagderI think the warnings in the iriver version is less critical
10:41:41*[IDC]Dragon is sick of that ravenheart song
10:41:54Bagderand I want to allow more people to help, and have my changes tested by Linus
10:42:02amiconnBagder: Yeah, I mean the warnings for the other "production" targets
10:42:23Bagderthose were not intended
10:42:30amiconnIf I wouldn't have fixed them, chances are the dailies for player & Ondio SP wouldn't have been functional
10:42:30Bagderjust sloppy tested
10:43:01[IDC]Dragona green build table looks sooo boring
10:43:03amiconnYeah, that's what I wanted to say with that link...
10:43:17Bagderwell, mistakes happen
10:43:22amiconn[IDC]Dragon: Still no error with that video?
10:43:48[IDC]Dragonexcept for the underrun, no
10:43:59[IDC]Dragona different question:
10:44:45amiconnReally strange
10:44:59[IDC]DragonI can't start with USB and MMC inserted without being prompted for re-plug. Did this come with my multivolume?
10:45:17amiconnErm, this was there all the time
10:45:31[IDC]Dragonaha. I didn't notice
10:46:00amiconnMaybe we can get rid of it now that we have multivolume
10:46:01[IDC]Dragonperhaps we should do a late mounting in case of USB while booting
10:46:22[IDC]Dragonwhy has it been there?
10:46:24amiconnWith multivolume it's no longer necessary to mount the external at boot
10:48:50 Quit elinenbe (Connection reset by peer)
10:49:05***Saving seen data "./dancer.seen"
10:49:39DMJCis it safe to install the bootloader yet?
10:49:45DMJCon IHP-140 devices?
10:50:11LinusNi don't know if it's safe
10:50:38amiconn[IDC]Dragon: This should in fact be simple. Skip the prompt when the external isn't initialized (and hence not switched to SPI mode)
10:50:52LinusNit works for me, but i have only tested on one device with one firmware version
10:51:20LinusNi can never guarantee that it is 100% safe
10:55:05amiconn[IDC]Dragon: The MMC driver does late init since the beginning, i.e. it checks on access whether the card is initialized. If it's not, the init is done "on the fly"
10:55:58DMJCdidn't you only test on a 120?
10:56:11LinusNyes i did
10:56:14DMJCtry it on your 140 then I might...
10:56:22LinusNthe 140 and the 120 are identical hardware wise
10:56:43DMJCwhich firmware version?
10:56:46DMJCI have 1.6 on mine
10:56:49LinusNbut i don't recommend anyone to install the boot loader yet
10:57:03LinusNi have 1.63
10:57:03DMJCstill waiting for the security features
10:57:14DMJCah good.. that's the same version I have
10:57:41DMJCsomeone built me a firmware with the bootloader set to boot the iriver firmware by default
10:57:46LinusN(i believe "e" stands for Europe)
10:58:06LinusNi will change that in the cvs too
10:58:37DMJCin time you'll set which one you want by default in rockbox?
10:58:39LinusNit's just more convenient for developers to start rockbox by default
10:58:42 Join elinenbe [0] (
10:59:10LinusNand for now, the boot loader is only for developers
10:59:15Bagderand there's no reason for non-devs to use for a while
11:01:00DMJCI could write hello world
11:01:13DMJCor a text based interface
11:01:14Bagderthat qualifies as a dev ;-)
11:01:30DMJCplay mp3 file y/n
11:01:38DMJCheh, i wrote a kernel installer that way
11:03:20DMJCshould port netbsd.. heh my girlfriend would love that
11:04:02 Quit Strath ("Client closed")
11:06:20 Quit ripnetUK ()
11:06:26 Join ripnetuk [0] (
11:14:11ripnetukI think I would prefer it if Rockbox started by default... that way, I NEVER have to re-flash once its working
11:15:48LinusNon the other hand, if the boot loader doesn't find rockbox.iriver, it would start the original anyway
11:15:49ripnetukLinus - which version of gcc are you using? some seem to be using 3.4.3 and some 3.3.4
11:16:09ripnetuki mean for m68k
11:16:11LinusNi *think* i use 3.4.x
11:16:31ripnetukok... i just want to make sure my enviro is as close to everyone else as possible
11:17:25LinusNand binutils from CVS 2004-09-22
11:17:53ripnetuksimilar to mine
11:18:03ripnetukbut not identical...
11:18:08BagderI could build fine with brand new cvs binutils
11:18:23LinusNyes, but could you *run* it? :-)
11:18:47ripnetuki built a toolchain yesterday, with the cvs yesterday, and it BUILDS perfectly (both rockbox and bootloader). I got the same md5 on the bootloader as others got the other day
11:18:49LinusNthe target code that is
11:18:54Bagderno, I need your outstanding commits
11:19:18LinusNso we're back again, where everything depends on me... :-(
11:19:21ripnetukbut when I built with 3.3.4 i got (unsuprisingly) a different md5, so the different versions definately produce doifferent object ocde.
11:19:37BagderLinusN: shouldn't be much more now though
11:19:47Bagdergotta go
11:20:22*LinusN fells the ulcer coming back
11:26:25LinusNthe safest way would be to offer binaries of the boot loader, along with instructions how to merge it with the original firmware
11:26:50LinusNand a list of proven firmware/bootloader combinations
11:28:07ripnetuki agree
11:28:20ripnetukand the md5s of the finished rom
11:28:41ripnetukor the ips patch to convert a rom, but we dont want to make it TOO easy :)
11:32:44ripnetuki get a different md5 on bootloader.bin now... aha, there have been overnight commits. What does the rockbox do now? i assume its more than a flashing backlight now - seemed to compile a lot more code after lat nights commit
11:33:16LinusNyes, Bagder made most of the stuff compile
11:33:28LinusNbut it doesn't run any of the code yet
11:34:56LinusNlunch time
11:35:05ripnetukso we still have the flashing backlight then
11:52:57Bagderactually no
11:53:11Bagderit doesn't flash anymore, it is replaced with the actual main
11:53:19Bagderwhich doesn't work quite yet ;-)
11:53:39Bagderso now it is less than flashing for a while
11:53:42*Bagder runs off again
11:56:15 Quit Nibbler ("blubber")
11:59:30 Quit elinenbe (Read error: 54 (Connection reset by peer))
12:04:43 Join elinenbe [0] (
12:05:09 Quit midk (Read error: 104 (Connection reset by peer))
12:15:49dwihnoBagder: you broke it! :)
12:15:57dwihnoyou broke it, you bought it? :)
12:25:45ripnetukbagder - how do you know it doesnt work? did Linus test it on his 120?
12:26:01Zagorripnetuk: it's missing some hardware setup code.
12:26:12ripnetuki see
12:30:02 Join DMJC-L [0] (
12:30:19ripnetukhi dm... did you manage to resist the temptation to flash :)
12:30:29DMJC-Lheh yeah
12:30:52DMJC-Lfigured i'd wait and see if someone gets mp3/ogg support soon
12:31:09DMJC-Lif I'm going to make it a brick it'll be in an attempt to play music
12:31:49DMJC-Lwill equalizer support be added?
12:31:57DMJC-Ljust your basic winamp style I mean
12:31:57ripnetuki would have htought so
12:32:03 Join preglow [0] (
12:32:08DMJC-LI don't use srs
12:32:23ripnetuki hate all euqalization... i always use natural
12:32:52preglowyo, would anyone happen to know where the __assert mentioned in assert.h is coming from?
12:32:57ripnetukim hoping for a 'feature manager' on Rockbox to turn OFF features i dont use
12:33:06ripnetukrather like the shuffle modes selector on iRiver
12:33:25DMJC-LI want otf playlist
12:33:33DMJC-Lotf, mp3/ogg
12:33:38DMJC-Lnothing else really mattersa
12:33:40ZagorDMJC-L: we have playlists already
12:33:47LinusNripnetuk: why turn off features?
12:34:00DMJC-Lon the fly? or just gay playlist files?
12:34:12ZagorDMJC-L: srs is garbage that will never come near rockbox
12:34:18DMJC-Lheh cool
12:34:20ripnetukfor example the shuffle modes on iRiver, there are about 6 different modes, and if I couldnt turn off the ones I never use, I have to press the shuffle select button like 6 times
12:34:26LinusNDMJC-L: yes, on-the-fly
12:34:29ripnetukto switch between no shuffle and shuffle all
12:34:35Zagorripnetuk: we don't fix iriver bugs, since we don't have them
12:34:57DMJC-Lwill navigation be similar to iriver's firmware?
12:35:01LinusNthe one-key shuffle toggle is an iriver firmware issue
12:35:06DMJC-LI found it to be pretty logical
12:35:12LinusNDMJC-L: you did???
12:35:25ripnetuki know, but its a great feature,
12:35:31ripnetuknot an issue :)
12:36:14LinusNi think the iriver UI is a nightmare
12:36:35DMJC-LI've never had an issue with it, only 2 things bug me
12:36:40ripnetuki think its OK apart from the lack of ability to go from deep in tree.c to wps.c
12:36:49DMJC-Lno OTF, and I want more format support heh
12:37:06ripnetukand no custom wps...
12:37:19*ripnetuk imagines 2 seperate custom wps's on remote and main
12:37:20LinusNwhile playing screen
12:37:23ripnetukwhile playing screen
12:37:28DMJC-Lah k
12:37:37ripnetukwith Rockbox, you can use format codes to choose EXACTLY what gets displayed
12:37:46DMJC-Lnot bad
12:38:40DMJC-LI don't mind as long as the buttons do what I'm used to for basic file selection/playback
12:39:09DMJC-Lcrap like radio and record, i don't care about
12:39:11LinusNone of the silliest things is the A/B key
12:39:37LinusNa dedicated key for just A/B playback
12:39:41DMJC-Lnot really.... I mean technically you could almost emulate a snes on the iriver..
12:39:51DMJC-Lit just lacks l/r buttons
12:39:52ripnetuki agree - a/b is a daft feature i have never used, and am forever accidently enabling
12:40:02ripnetukand a color screen
12:40:16DMJC-L300's have color screens
12:40:55ripnetukhow about a java virtual machine?
12:41:04DMJC-Lthe upcoming gameboy emulator almost makes me regret not getting a color..
12:41:05ripnetukor a .net compact framework :)
12:41:06dwihnoa/b key is silly
12:41:35DMJC-LI'm sure you'll find a use for it
12:42:03LinusNoh yes, we need all the keys we can get
12:42:08ripnetukthere are a hell of a lot of inputs on the remote... i make it 11+hold
12:42:15LinusNit's just that the button labling is silly
12:42:20DMJC-Lah heh
12:42:36ripnetukthe archos had what? 10? including the hardware off button
12:42:47LinusNf1-f3 is nice, like on the archos recorder
12:43:05LinusN10, yes
12:43:26LinusNthe off button is used as a "normal" button as well
12:43:40 Quit Peter99 ("CGI:IRC (EOF)")
12:43:46ripnetukprovided it doesnt need to be held
12:44:27ripnetuki think the off (stop) on iRiver is software driven... i managed to crash it (only) once, but giving it a bad mp3, and i had to use the reset button
12:46:38ripnetuki was out on my bike at the time - had to find a tree with thorns to pree the reset :)
12:46:48 Join Patr3ck [0] (
12:47:02LinusNit is s/w only, yes
12:47:24DMJC-Lmy h-140 is the single coolest thing I own
12:47:58ripnetukdmjc - mine comes close to the top of my list
12:48:11DMJC-Lrockbox is going to make these things crap all over ipod
12:48:38ripnetukrockbox converted the distinctly average archos to be a great mp3 player
12:48:51ripnetukso, it will convert the great iRiver into a superb mp3 player
12:49:06***Saving seen data "./dancer.seen"
12:49:26ripnetuki hope we get a c64 sid player
12:49:35LinusNcoll indeed
12:49:45ripnetukMAgical Sound Shower
12:49:59DMJC-LI want .mod support..
12:50:02DMJC-Lthat'd be l33t
12:50:11DMJC-Lactually .mid would be more l33t
12:50:18LinusN.mod is not impossible at all
12:50:33ripnetuk.mod just involves being able to mix 4 samples does it not?
12:50:43LinusNstandard mod, yes
12:50:58DMJC-L.mid is possible in theory afaik
12:51:03DMJC-Lif you use soundfonts
12:51:03LinusNthere are 8-track variants too, iirc
12:51:13ripnetukfor amiga i remember
12:51:15LinusNripnetuk: exactly
12:51:23ripnetuki think it software mixed 4 pairs
12:51:41DMJC-Lamiga was a 7 or 14 mhz machine?
12:51:51DMJC-Lit was pretty slow afaik
12:52:08ripnetuki loaded winuae on my laptop the other day to watch "jesus on e's" demo. They have implemented the floppy disk drive noises in the emulator ;)
12:52:10LinusN.mid is harder, because it requires polyphonic voices
12:52:35DMJC-Lpolyphonic voices?
12:52:42LinusNyou know, chords
12:53:06LinusNi think mid requires at least 16 channels
12:53:39LinusNand each channel can have polyphonic voices
12:53:53ripnetuksurely if the 7mhz amiga could mix 8 channels into 4, our coldfire should be able to merge 16 into 1?
12:54:28LinusNbut in the midi case, it's more like 32 or 40
12:54:33LinusNin theory
12:54:44ripnetukand there IS a repository of game music on the web in mid format, which is cool
12:54:59DMJC-Lheck i have a bunch of it here
12:56:30 Join Nibbler [0] (
12:57:17DMJC-Lmainly wing commander stuff
12:59:09ripnetuki always like the ninteno music on mario games
13:03:42dwihnothe iriver software bugs on some mp3s
13:03:53ripnetukit doesnt like mp2s
13:04:05LinusNincluding "Kebab" :-(
13:04:21ripnetukif it hangs on some mp3s, maybe there is potential for a buffer overrun attach
13:04:59DMJC-Lvirus infected iriver now?
13:06:36LinusNripnetuk: probably
13:06:48LinusNi haven't had the time or energy to investigate though
13:07:33dwihnoLinusN: It doesn't like kebab?! We have a major software issue! :(
13:07:47ashridahmmmm. lamb sandwich
13:07:47ripnetukgive it some beer. then it will LOVE kebab
13:07:49dwihnoLinusN: I got a regular mp3 file which is "ok", but the software just plays a couple of secs and then skips
13:10:56LinusNdwihno: it hates Starka Såsen
13:11:24dwihnoiriver needs rockbox - "because it's starka såsen compatible(tm)"
13:11:41LinusNa strong sales pitch indeed
13:12:24 Join [1]Patr3ck [0] (
13:13:52dwihnoLinusN: have you done any work on some kind of pcm driver?
13:14:17LinusNhaven't had time
13:14:43LinusNbut that is the next thing to do when the boot loader is finished
13:15:45dwihnodo you think it will be hard?
13:15:50Lynx_what's a pcm driver? :)
13:15:50ripnetukhows the minimon coming on then?
13:15:59amiconnLinusN: "starka såsen"?
13:16:08ripnetukpulse coded modulation - uncompressed (almost uncompresed) audio
13:16:10LinusNripnetuk: i think i'll ditch the minimon
13:16:22dwihnokebab is a good song for testing mpeg playback
13:16:22Lynx_ripnetuk: a, thx
13:16:32DMJC-Lditch the minimon?
13:16:34ripnetukdid you say you were pointing the exception vector to the original firmware instead?
13:16:45LinusNas you said, there are few occasions where it will be useful
13:17:10ripnetuki agree really, its only really useful if rockbox fails between minimon coming up, and the check for the record button
13:17:12LinusNinstead, i'll start the original firmware
13:18:01ripnetuki dont mean when will it be ready, i mean when will you start the original fw
13:18:07LinusNin the exception handlers
13:18:21LinusNand in panicf()
13:18:40LinusNbasically, in every situation where rockbox crashed
13:19:16ripnetuki see... so the offical advice is still to wait until thats done before flashing then? and wait for Zagor and Badger to flash as well?
13:19:56LinusNanother safety net would be to write a cookie to the internal ram early in the startup, and clearing it when reaching the rec-key loop
13:20:27LinusNso if the cookie is there when it reboots, something must have gone wrong and reset ws pushed by the user
13:20:29ripnetukand have another thread checking the cookie?
13:20:38ripnetuki see :) smart
13:20:48LinusNso if the loader hangs, you can reset and it will start the original
13:21:04ripnetukthat sounds a lot safer...
13:21:08LinusNbetter play it safe
13:23:44ripnetukdo we have any spare hardware interupts? could we point one of them at the original, so we can connect +ve to a pin and have it jump there?
13:25:05LinusNinteresting idea
13:25:29ripnetukkind of like one of those heart machines... SHOCK... alive
13:28:19 Quit Patr3ck (Read error: 110 (Connection timed out))
13:28:20 Nick [1]Patr3ck is now known as Patr3ck (
13:29:06LinusNno general irq:s are routed to a nice external port or key
13:29:20LinusNhowever, a watchdog is of course a good idea
13:33:29ripnetukoh well
13:33:58ripnetuki guess the cookie thing should cover it... all other suggestions rely on being able to successfully jump to the original code, and on our code running at all
13:34:18ripnetukwhich as far as I can see are the only failure possibilities for the cookie thing anyway
13:34:37ashridahdoesn't mean you can't glue a timer to a software interrupt very early in the bootup
13:34:47LinusNyes you could
13:35:08ashridahhave the vector point to a bit of code that kills the timer, then jumps to the normal firmware
13:35:28ashridahneed to remember to kill it in rockbox tho ;)
13:35:33ripnetukbut does that give us anything the cookie doesnt?
13:35:43ashridahripnetuk: it gives you a poor mans watchdog
13:36:01ashridahthe cookie is failure detection, not endless loop detection
13:36:20LinusNthere is a watchdog timer in the coldfire
13:37:01LinusNit can reset the cpu
13:37:18LinusNso we can extend the cookie functionality to include the entire boot loader
13:37:31ripnetuki dont follow
13:37:41LinusNi.e we reset the cookie right before leaving the boot loader
13:38:12LinusNthe watchdog resets the cpu and the cookie is not reset, so the original is immediately started
13:38:50ripnetukso it just saves having the manually press reset?
13:39:03 Join jyp [0] (
13:40:01LinusNhmmm, you're right, we only need the cookie protection up to the point where the rec key sensing works
13:40:42LinusNand the cookie protection is somewhat simpler than the timer solution
13:41:27ripnetukso long as the reset doesnt clear / invalidate the internal ram
13:41:47LinusNi doubt it very much
13:41:49ripnetukbut you can easily test that with your bdm
13:42:44ripnetukjust out of interest, what does the bdm consist of? is it a custom chip?
13:42:48ashridahripnetuk: easy way to find out. stop the iriver half-way through a song. then hit the reset button on the bottom, and see if it forgets where it was
13:43:04LinusNripnetuk: it's a quite simple design
13:43:09ashridah(assuming the track position is written there)
13:43:20ripnetukash - i dont know if it stores that on disk or ram
13:43:30LinusNmainly only a buffer for the parallel port, with some extra logic for the clocks
13:44:04ripnetuki see
13:44:33LinusNthere are schematics and PAL code out there, if you want to build your own
13:44:35ashridahripnetuk: i don't think there's really any place that it can safely store that kind of thing in a fat32 partition.
13:44:49LinusNashridah: how about a file? :-)
13:45:04ashridahLinusN: there doesn't seem to be one, unless it deletes it before allowing a pc to connect
13:45:19ashridahof course, it forgets when you do so
13:45:26LinusNi think they do it in one of two ways:
13:45:44LinusN1) The Rockbox way, i.e in an unused sector on disk
13:45:52LinusN2) In the I2C EEPROM
13:46:00ashridahinternal ram seems sanest, since then you can use it for things like FM and not have the disk on at all
13:46:15ashridahnvram rather
13:47:08LinusNthere is no nvram, afaik
13:47:11ripnetuki dont this we can tell for sure without implementing the cookie, and deliberately making it hang after writing the cookie, then reseting and seeing if it works
13:47:15ashridahmy first iriver H140 hada defect where it'd refuse to read the drive's contents if it was warm when it got turned on (either warm from use or charging)
13:47:38ashridahbut it seemed to still be able to remember its settings
13:48:00ashridahripnetuk: took me a few days to pin down how to reproduce it. got it replaced quickly enough once i did tho
13:48:25ashridahwas annoying tho, i had to go home without files on it twice in a row and minimal battery life for listening to the radio :)
13:48:26ripnetukyeah... failed hdd's can be nasty... i think (although still not proved) that a failed hdd was responsible for my recent spate of Linux hangs
13:48:34LinusNi also think the DRM key on the 300 is stored in the i2c eeprom
13:48:39ashridahripnetuk: oh, it wasn't a failed drive
13:48:54ripnetukfailed controller?
13:48:56ashridahthe drive worked fine no matter what when you plugged the unit into usb
13:49:15ripnetukmight have been a loose joint and thermal expansion
13:49:15ashridahit just refused to read the disk properly itself when you turned it on or gave it control
13:49:26ashridahit'd wig out when it got to the directory tree, yet still turned off with the stop button
13:49:40ashridahripnetuk: well, yes, since it only happened when warm, i assumed it was thermal expansion
13:49:46ashridahthe replacement's worked fine for almost a year now
13:49:49ashridah(10 months)
13:49:57ripnetukmine is about that age
13:50:12ripnetukmore like 8 months actually
13:50:49*ashridah triggered a wave of iriver purchases at his uni when he got his >:)
13:50:50 Join Sporkboy [0] (
13:50:56 Quit Sporkboy (Client Quit)
13:50:58ashridahLinusN: did they add DRM support to the 3xx?
13:51:18ashridahi haven't been following firmware updates since they changed the site to flash-based hell
13:52:20 Quit Nibbler (Read error: 113 (No route to host))
13:55:03LinusNthe us version of the 300 has drm support
13:55:21LinusNbut the drm key is lost if you flash non-us firmware
13:56:37LinusNand it doesn't help to reflash the us firmware
13:57:21ripnetukintereswting... by i think th ebest way to handle drm is to buy the music and then download a non-restricted version
13:57:45LinusNthat's probably not the spirit of the drm scheme :-)
13:57:53ashridahhm. the 3xx series doesn't have the optical out anymore does it?
13:58:16ashridahpity, since you'd be able to make basically identical copies using optical-in on something else
13:58:21LinusNwhich makes it suck enormously
13:59:28LinusNi don't use the s/pdif to copy, i use it to connect my player to my hifi
13:59:38ripnetukme too, via minidisc
13:59:50ripnetuksounds so much better than the internal dac
13:59:51ashridahmy digital satellite receiver has an optical/analog output.
14:00:07ashridahi've recorded some long sessions off the radio channels they distribute
14:00:12ashridahgives me some variety
14:00:27ashridahonly via analog atm tho, need to get an optical cable
14:01:21ripnetukin the UK, freeview and a dvb card gives me a very good way to record radio. Gives me a unrestricted mp2 stream :)
14:04:49ashridahhm. midnight, gotta crawl out of bed at 8am. night
14:04:50 Quit ashridah ("sleep")
14:08:20HCli didn't know those things about 3xx
14:09:45*HCl reads backlog
14:12:59LinusNi learned it from the misticriver forum
14:16:53HClok, done :P
14:19:55HClLinusN: can you tell me something about how lcd accesses work? does it have a framebuffer? or?
14:20:19HClits somewhat important for the gameboy emu whether you can write to it directly or not..
14:20:24LinusNthe lcd has its own frame buffer
14:20:30HClgood :)
14:20:48ripnetukare we going to have a gameboy emu before mp3 playback :)
14:20:52HCl :P
14:21:07LinusNyou set the buffer pointer and then you write the data, autoincrementing the pounter
14:21:11HClnah, wouldn't say that, just wanting to collect some info
14:21:14HClLinusN: *nods*
14:21:30LinusNthe rockbox lcd driver has its own frame buffer
14:21:35HClits ashame it misses a few pixels then..
14:21:43HClcompared to the gameboy
14:21:58LinusNin which we can render fonts and graphics, then we transfer the whole buffer to the lcd
14:23:12ripnetukhow do you book out from cvs as it was on date x aat time y?
14:23:14HClwell, i'm somehow gonna have to intercept and remap writes to certain lines.. skipping 16 pixels in the beginnning/end shouldn't be hard, but. skipping 16 evenly distributed lines throughout the screen will be harder..
14:23:31LinusNripnetuk: cvs co -d <date>
14:23:36ripnetukthanks :)
14:23:41HClah well.
14:23:42LinusNor cvs up -d <date>
14:23:42 Join bobTHC [0] (
14:23:51bobTHChi mates!
14:23:54*HCl waits till the bootloader is declared stable and stuff
14:23:56LinusNhi ho
14:24:03LinusNlong time no see
14:25:13bobTHCyep, very busy this days....
14:25:48bobTHCand u seems to be busy with the iriver too ;)
14:35:05Lynx_[IDC]Dragon: should your RockVideo tool be able to convert xvid avi's?
14:38:24HCldoes any archos / iriver even have the cpu to do xbvid..?
14:38:51LinusNyes, at 128x64 pixels
14:39:48HClwell.. personally i think that might only prove useful on a h3x0...
14:40:06HClor cartoons, or something
14:42:45HClcan you disable the whole drm stuff on a h3x0?
14:43:19LinusNi dunno
14:48:47 Quit ripnetuk (Read error: 60 (Operation timed out))
14:48:52 Join ripnetuk [0] (
14:49:10***Saving seen data "./dancer.seen"
14:52:21 Join R3nTiL [0] (~zorroz@
14:59:05ripnetukRe booking out from CVS for a given date, I just cant get the syntax correct... the line I use to book out already had a -d in it. I use
14:59:07ripnetukcvs -z3 co rockbox-all
14:59:13ripnetukfor a normal bookout...
14:59:47LinusNthe -d must be after the "co"
15:00:00ripnetuki see. thanks :) i will document that on my wiki of tech notes
15:00:39ripnetukthat worked - thanks :)
15:04:41HClthe faq is a bit out of date... with the upcoming iriver port..
15:04:54HClcause it says stuff can't be done cause the archos doesn't support it.. but like, the iriver will.
15:05:41 Join El_Barto2 [0] (
15:06:12El_Barto2anyone interested in an archos recorder 20?
15:06:26HClLinusN: q19, for instance
15:06:26LinusNyou're selling?
15:06:48El_Barto2i sell it or will it change with a gmini 120
15:06:58LinusNHCl: ah, yes
15:07:12LinusNHCl: feel free to update, it's a wiki
15:07:29HCli dunno, i'm not sure whether i trust my documentation writing abilities
15:08:11HCllmao @ q27 xD
15:09:48 Join NibbIer [0] (
15:15:40amiconn[IDC]Dragon: r u there?
15:19:18 Join midk [0] (
15:32:57 Quit R3nTiL ()
15:34:10ripnetukcvs question - when you book out for a given date, is that the START of the day of the end?
15:34:15ripnetukor the end
15:34:28LinusNthe start, methinks
15:34:40LinusNah, no, the end
15:47:50 Quit jyp ("poof!")
16:02:05 Part LinusN
16:13:46Zagorsounds like gmail is out of beta. everyone got loads of invites.
16:19:34Lynx_why do they keep this funny invite scheme?
16:19:58Zagorgood question
16:21:16Lynx_hmm, would you pay 5 euro more for 8mb harddisk cache instead of 2 mb ?
16:21:37Zagorfor a computer, yes. for an mp3 player, no.
16:21:40bobTHC2.5' ?
16:21:58Lynx_it's more for backup purposes, so i guess no...
16:22:39ripnetuki would as its only 5 euros and you might want to use it for something else in future
16:23:26Lynx_ripnetuk: that's what i just thought. on the other hand, in the future it may not even fit a longhorn 2 installation ;)
16:27:21 Join methangas [0] (
16:30:31Lynx_i ordered the 8mb one :)
16:32:39El_Barto2it would be better to order the gmini 120 ;)
16:32:47El_Barto2it has standart 8mb cache
16:33:17El_Barto2and you dont have to open the player ;)
16:34:29Bagderbut you don't get much storage
16:35:02 Quit Ka (Read error: 60 (Operation timed out))
16:35:09Lynx_El_Barto2: i was referring to a 200 gb standard hard disk...
16:35:10Bagdera 120 is only 20GB, isn't it?
16:35:33El_Barto2and what should have 200?
16:35:39Bagdera hard drive
16:35:57El_Barto2i thought you were talking about an archos player ;)
16:42:38 Join mecraw [0] (~mecraw@
16:49:12***Saving seen data "./dancer.seen"
16:55:54[IDC]DragonLynx_, amiconn: now I'm here
16:56:09 Quit El_Barto2 ()
16:56:53 Join funkymonkey [0] (
16:56:53 Quit methangas (Read error: 104 (Connection reset by peer))
16:56:57 Nick funkymonkey is now known as methangas (
16:57:07Lynx_[IDC]Dragon: howdy :) Have you ever used RockVideo with an xvid file?
16:57:12[IDC]DragonLynx_: the video converter should be able to convert anything you can play in the Media Player
16:58:22Lynx_[IDC]Dragon: i investigated further with that graphedit thing. It does not allow me to connect the xvid decoder to your filter, although the out of xvid and in of your filter list the same 'format', YUV2 or somethign like that.
16:58:48Lynx_[IDC]Dragon: if i just use the gui it crashes, and sound goes to the standard direct sound device.
16:59:26Lynx_however: i wanted to try to use the divix decoder instead, and by accident the graphedit connected the xivid decoder to the divix, and then to your filter. and that works...
17:00:09ripnetukwierd stuff happening... I think there is something wrong with the iriver scrambling code - i scrambled a .bin to a .hex, and then descrambled the .hex to .bin.check (to check) and the files differ
17:00:11Lynx_Does that make sense? My vocabulary is probably off, I never do anything like that
17:00:52 Join Digital007 [0] (
17:01:46[IDC]DragonLynx_: I very vaguely remember something similar
17:02:01[IDC]Dragonand fixing something
17:02:11ripnetukthey differ LOADS - is there some random factor in the scrambling?
17:02:40Lynx_[IDC]Dragon: well, the message is i guess that in the version i downloaded today it does not work, and the connections that graphedit makes by default are wrong (no video, audio to std device)
17:02:41[IDC]Dragondo you have a short xvid which doesn't convert but play?
17:03:56Lynx_[IDC]Dragon: 183 mb...
17:04:46 Quit Digital007 (Client Quit)
17:05:05 Part Zagor
17:05:52amiconn[IDC]Dragon: Any news concerning the mas wav codec?
17:06:13[IDC]DragonI nagged again just a few days ago
17:06:26[IDC]Dragonit's "in progress"
17:06:42amiconnHmm. :-/
17:08:01Lynx_[IDC]Dragon: ok, the divix->xvid->rvf version creates a rvf file, but that displays as a black screen on the box, sound is ok
17:08:05 Join Stryke` [0] (~Chairman8@resnet-241-86.resnet.UMBC.EDU)
17:08:27[IDC]DragonI don't understand divix->xvid
17:08:45[IDC]Dragonthe first decodes, and the second?
17:09:00amiconnLynx_: I also had problems when testing the DirectShow filter (usually I do use the command line tool), also with a DivX video
17:09:11amiconnI have the 3ivx codec installed.
17:09:18HClripnetuk: mmm, i remember that happening yesterday when i did a check for that before i sent the .hex to DMJC-L
17:09:47amiconnThe problem was that there is a setting to "force" the codec to a hardware overlay... which obviously fails when it should render to a file...
17:10:09[IDC]Dragonah, I remember
17:10:20Lynx_[IDC]Dragon: i have no idea what i am doing, really. i wanted to use the Divx decoder instead of the xvid one in graphedit. when i connected that, i still had the xivid one open in the program, and it connected it after the divx one. so i thought i would just try that, and it didn't give ma an error and created the file.
17:12:47ripnetukhcl - its wierd, I can decode and re-encode the original fw and get the same file. If i encode new.bin then decode it, its VERY different
17:13:06HClripnetuk: it says something about building checksum tables the first time
17:13:14HCli'm figuring new.bin doesn't have those the first time?
17:13:24HClhave you tried reencoding the decoded new.hex?
17:13:33ripnetukno :) did you?
17:13:48HClshould be tested, really
17:13:53ripnetuki guess so
17:13:58Lynx_[IDC]Dragon: Am i correct that the 'XForm Out' pin of the xvid decoder should be connected to the YUV Video pin on the rfv renderer?
17:14:29[IDC]Dragonis there any other?
17:15:25Lynx_[IDC]Dragon: no, but that is the connection graphedit does not allow, "the codecs could not agree on a connection..."
17:15:58[IDC]Dragonthat's the interesting point
17:16:12[IDC]Dragonare there any properties at the xvid?
17:17:04Lynx_[IDC]Dragon: yes, as i said both list YUY2 and YV12, but the xivid has more details, sec
17:17:17Lynx_Major Type: Video - Sub Type: YUY2 - Format: YUY2 512x384, 16 bits,
17:17:18Lynx_Aspect Ratio: 512x384,
17:17:18Lynx_Interlace format: Frames
17:17:18DBUGEnqueued KICK Lynx_
17:17:32Lynx_sorry, that was copied as on one line!
17:20:55HClripnetuk: they match the second time
17:21:03HClso its the building of the checksum stuff
17:24:17ripnetukthat sounds better :)
17:24:34[IDC]DragonLynx_: no, I mean a property page where you can configure stuff
17:24:55[IDC]Dragonproperty of the filter, not the output pin
17:26:19ripnetuki want to know how on earth people managed to work out the scrabling alogrithm for the iRiver
17:26:52ripnetukmaybe the descrambler is unencrypted in the rom?
17:28:19HClwith the bdm thingy you can dump the iriver flash rom
17:28:30HCland after that you can reverse engineer the flashing process
17:28:31ripnetukbut it was cracked WELL before the bdm came into being
17:28:33HClwhich decodes it
17:28:41HCli guess
17:28:53ripnetuksome people were tampering with the bitmap images ages ago
17:29:24ripnetukim guessing the rom starts with the decryption routine in plaintext - otherwise the iRiver would require 2 seperate firmware roms, one to bootstrap the other
17:31:56HClwouldn't it decrypt it while flashing?
17:32:59ripnetukeven so, someone managed to work out the alogrithm without a bdm
17:33:22HClinsider? who knows.
17:35:39 Join LinusN [0] (
17:36:24LinusNripnetuk: the descrambled original firmware still contains all the checksums from the scrambling
17:36:49LinusNhowever, new.bin doesn't have those before scrambling
17:36:50 Quit bobTHC ("( :: NoNameScript 3.81 :: )")
17:37:07LinusNso if the differences are in the end of the file, there is nothing to worry about
17:38:14ripnetukthanks Linus - one less thing to worry about :)
17:38:59LinusNif you scramble the newly descrambled file again, and then descramble it, the result is the same
17:39:41LinusNgotta go
17:39:43 Part LinusN
17:42:24Lynx_[IDC]Dragon: yes, there is a property page, where i can set the colourspace and some other things
17:44:08Lynx_[IDC]Dragon: weird, if i select "render pin" on the file pin or the xvid out pin, it connects it to the rvf MPEG audio pin...
17:45:45[IDC]Dragonthe audio input should be a bit mory picky
17:46:13[IDC]Dragonare you using a not too old version?
17:46:23[IDC]Dragonfrommy webspace
17:46:37[IDC]Dragonnot that I updated it recently
17:47:04Lynx_[IDC]Dragon: i got it from your t-online page, the link that is in the wiki
17:47:07Lynx_[IDC]Dragon: today
17:48:06Lynx_[IDC]Dragon: rockvideo is 1.0.1
17:49:18 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:50:44Lynx_[IDC]Dragon: when the xvid out pin is not connected to anything, the properties say YUY2 (what I set it to), and when it gets connected the rvf audio pin, the props on the xvid out say audio, mpeg1Payload, WaveformatEx
17:55:51 Quit ripnetuk (Remote closed the connection)
17:56:27[IDC]Dragonso xvid is too flexible
17:56:39[IDC]Dragonconnecting to an audio input even
17:57:39[IDC]DragonI could do a change to the filter to accept RGB video as well, but that's less efficient
17:58:37Lynx_[IDC]Dragon: well, but why does connecting xvid by hand not work with YUY2?
17:59:15[IDC]Dragonthere is no config option at xvid?
18:01:12Lynx_[IDC]Dragon: there is, that's where i set it to YUY2. It also offers YV12 and two RGB.
18:03:53Lynx_[IDC]Dragon: AHA! The xvid properties have an 'compability renderer' checkbox, if i check that i can connect it. Seems to work, i'll try the result in a minute...
18:09:00Lynx_[IDC]Dragon: ok, that was it, now the rvf file plays fine. well, not too great, but ok.
18:10:37Lynx_[IDC]Dragon: however, the RockVideo app still crashes with the file...
18:12:44[IDC]Dragonso, you can do the proper conversion with graphedit?
18:13:15Lynx_[IDC]Dragon: Yes, by activating "Compatibility renderer" in the xvid options
18:13:31[IDC]Dragonand it generated a nice file?
18:14:13Lynx_[IDC]Dragon: I don't have much to compare it to, the contrast could be higher, and every 2 seconds a snowy bar goes from the bottom of the screen to the top
18:14:24Lynx_but it is watchable
18:14:53[IDC]Dragonthe snowy bar is normal
18:15:12[IDC]Dragonyou can only make it slower and smaller
18:15:12Lynx_ok. it's actually more like 2 times a second
18:15:24[IDC]Dragonthat's slow already
18:15:46Lynx_ok :). how far can the general flickering be reduced?
18:16:13[IDC]Dragonit can't, this is no video display
18:16:40 Quit Patr3ck (" HydraIRC -> <- Chicks dig it")
18:17:01Lynx_i meant is there still the option to set 'some value' to make it better? i remember you asking in the mailing list to collect the individual values for the players...
18:17:23Lynx_the value that can be set in you rvf renderer
18:17:50[IDC]Dragonyou can try to match your fps better
18:18:05[IDC]Dragonand you can adjust the gamma
18:18:06Lynx_[IDC]Dragon: is there a way to find out, or is it trial and error?
18:18:26[IDC]DragonI once made a plugin for this, uncommitted
18:19:02[IDC]Dragonit'd manual, but helps to "tune in"
18:19:16Lynx_is it still available online?
18:19:38Lynx_an how would i go about changing the gamma?
18:19:42[IDC]Dragonbut the fps of the display yaries with temperature
18:19:58[IDC]Dragonisn't it in the property page of the filter?
18:20:06Lynx_[IDC]Dragon: ah, right, that was the thing about the temperature...
18:20:41Lynx_no, there's only archos playback fps there
18:22:56[IDC]Dragonthen it's waiting to be implemented...
18:23:34Lynx_hehe. The possibility to watch simpsons on the box is great already :)
18:25:22Lynx_gotta go. thanks for the help (and the plugin)
18:25:59 Nick Lynx_ is now known as Lynx_awy (HydraIRC@
18:25:59DBUGEnqueued KICK Lynx_awy
18:30:00 Join einhirn [0] (
18:40:25 Quit preglow ("leaving")
18:47:43amiconnHmm, now that is strange: I just discovered that the archos remote stops working after rockbox is playing for a while (on my Studio 10). The only way to make it work again is to reboot rockbox :-/ ??
18:49:02 Join jyp [0] (
18:49:15***Saving seen data "./dancer.seen"
19:15:08 Join ripnetuk [0] (
19:29:00 Join preglow [0] (
19:44:01CtcpIgnored 2 channel CTCP requests in 1 hour and 8 minutes at the last flood
19:44:01*jyp shakes his head
19:45:29 Quit [IDC]Dragon ("CGI:IRC")
19:50:36HClgod how i hate some people.
19:50:46HClthere's like, this poster of a cat thats been missing for weeks
19:50:53HCland some people took a red marker and drew blood on the picture.
19:55:48ripnetukmost people dont realise how cool cats ar
20:07:04 Quit ripnetuk (Remote closed the connection)
20:11:45 Join shano [0] (
20:12:16shanoDoes anyone know where I can get information on rockbox causing the battery to fry in archos jukebox 2.0?
20:13:01HCli'd say, check the wiki?
20:13:59shanoWhere/s that?
20:14:34shanofound it. Thanks. Didn't look hard enough before.
20:14:47HClout of sheer curiousity
20:14:49HClwhats the url?
20:14:57HCli'm curious to hear about archos frying
20:16:07Bagderrockbox does not fry the battery
20:16:52HCli figured as much
20:17:45shanothe link is on the left hand side.
20:18:57shanoi'm looking at the forums right now. my jukebox fried last year. i took out the battery and found one of the terminals completely fried. i just sent it in to get fixed a couple of weeks ago. i charged it last week with the regular firmware and it was fine. i loaded rockbox a few days ago, now when i plug in the charger, the unit is EXTREMELY hot. i'm going to do some more testing.
20:19:07shanoit's supposed to be 9V, center positive polarity, correct?
20:19:46 Join _aLF [0] (
20:21:05HCli'm not sure, i'm more into iriver
20:22:15HClthe jukebox falls under the players, right?
20:22:47HClDoes Rockbox charge the batteries more/less/worse/better than the Archos firmware?
20:22:50HClPlayer: Rockbox doesn't affect the charging on Players, it is all done by hardware logic out of software control.
20:36:56 Join Quelsaruk [0] (~kvirc@
20:37:59HClhey quelsa
20:38:17HClhas anyone looked at writing a gdb stub for iriver?
20:40:59shanoi don't think that's right. if you look at the rockbox manual, it specifically lists deep discharge option which doesn't charge until the battery is at 10%. Or the trickle charge option. So rockbox must affect charging somehow.
20:41:26Quelsarukcharging the recorder is made by firmware
20:41:32Quelsarukplayer is made by hardware
20:41:46Quelsaruk(archos models)
20:45:17HClshano: only for the recorders
20:45:31HClit clearly says that in the battery faq
20:45:55amiconnTo be precise, for the v1 recorder only.
20:46:10amiconnv2 and FM have hardware charging too
20:46:16HCl :)
20:46:35shanoMine is a recorder.
20:47:05shanoamiconn, V2 is supposed to be okay?
20:47:26HClthey're all supposed to be okay
20:47:33HClits just that only with v1 does rockbox control charging
20:49:17***Saving seen data "./dancer.seen"
20:49:54shanoCan you show me where it says that?
20:50:00shanoTell me, I mean.
20:53:55amiconnshano: First, see , Q1 and Q3
20:54:15amiconnThen you need to know that this is for the player and recorder v1 *only*
20:54:17 Join davFr [0] (robind@
20:54:38amiconnWhen this faq was put together, rockbox did only support these 2 models
20:55:01amiconnThen see , Q7
20:56:05davFris there some R.E. work going on around the rio SP-250?
20:57:08Quelsaruki don't think so
20:59:02davFrhi Quelsaruk, i am trying to get the original SP-250 firmware disassembled, so i am trying to figure out if it is scrambled, and how. Maybe there some stuff similar to firmware encryption of HD iriver products?
21:00:25Quelsaruki know you can decramble iriver firmware
21:01:36davFrdo you know links to useful ressources, please?
21:01:51Quelsarukbut i don't follow the rockbox portability to iriver... so i don't know really nothing interesting about that
21:02:06Quelsarukmaybe amiconn can help there
21:02:32amiconnSorry, I can't
21:03:37davFrok :o)
21:03:46amiconnI guess the scrambling will be different tho.
21:04:48amiconnEach manufacturer in fact wants his own, proprietary algorithm. That's what scrambling the firmware is about - make it harder to reverse engineer
21:05:56HClwhy would the iriver scrambling be anything like the rio scrambling....?
21:06:03davFri thought it would have been decrypted since sp250 was released
21:06:46davFrrio was just putting "riovolt" label on iriver products, that's my guess
21:07:21HClwell. you can always go at it with a disassembler
21:07:23davFri don't feel rio people used to know much about sp250 internals
21:07:28HClsee if it makes any sense
21:07:36HClofcourse you'd need to know its architecture for that
21:08:25davFrsp250 is based on a Cirrus Logic chip. which is build on an ARM720T (evolution of the ARM7TDMI core)
21:08:52HClmy laptop battery's almost empty, brb
21:09:05davFrSo it should be disassembled by a ARM gcc tool
21:09:42 Quit shano ("Leaving")
21:10:07HClwell, objdump will disassemble, which is part of binutils..
21:10:24davFrnow i got the s/n : it's a cirrus logic EP7312
21:10:46HCli used to have an arm toolchain, but i don't have it installed at the moment
21:11:34davFra guy on figured out some stuff about the firmware image, like checksums, but did not get how to disassemble the code.
21:11:40rasher_What's the latest news on the iRiver? I see a bunch of things checked into cvs
21:12:04davFrcan't say, i don't have a HD-based iRiver
21:17:50Quelsaruki think it can boot
21:18:25QuelsarukHCl: did you have a iriver jukebox?
21:21:00Quelsarukrasher_: AFAIK,
21:21:03Quelsarukthat's the status
21:21:07Quelsarukbut i'm not sure
21:24:26 Join Eodun [0] (
21:25:31*Eodun waves his hand
21:27:51Quelsarukdinner time
21:28:15 Nick Quelsaruk is now known as quel|dinner (~kvirc@
21:29:09 Quit Eodun (Client Quit)
21:29:16 Join markun [0] (
21:29:53markunI get this error when I try to compile the bootloader:
21:29:56davFrsame chinese takeway time!
21:30:02markun/usr/home/karl/rockbox/bin/../lib/gcc/m68k-elf/3.4.2/../../../../m68k-elf/bin/ld: region IRAM is full (/usr/home/karl/tmp/src/rockbox/builds/boot/bootloader.elf section .bss)
21:32:04markunHCl: what version of binutils did you use to compile gcc and the bootloader?
21:35:41HClare you gonna try it? *curious*
21:37:41markunNo, but I would like to have a working compiler ready.
21:38:56 Join mecraw_ [0] (~mecraw@
21:39:23HClwell, i just followed the instructions, worked fine
21:40:32markunFirst compiled CVS binutils and then gcc-3.4.2?
21:41:12markungcc didn't want to compile, only with binutils-2.15
21:42:29HClyea, cvs binutils..
21:42:31HClnot 3.4.2
21:42:39HCl3.3.4 is recommended
21:42:44HClcause 3.4 generates larger code
21:43:18HClyou need the binutils update because it contains the specs of the iriver cpu
21:44:03HClotherwise you'll just get "unrecognised machine blahblah" when you try to compile with your new gcc
21:46:26HClhmm... *wonders whether he can upgrade his h140 to a 60gb hdd*
21:49:22davFr(back with chinese food)
21:50:56rasher_share and enjoy
21:55:30 Quit mecraw (Connection timed out)
22:00:18 Join LinusN [0] (
22:02:41markunLinusN: did you use gcc 3.4.2 for the coldfire compiler?
22:04:43 Nick quel|dinner is now known as Quelsaruk (~kvirc@
22:06:00LinusNusing it now
22:07:00markunDid you see the error I got? Maybe because I had to build gcc using binutils-2.15 and the bootloader with binutils from cvs..
22:07:31LinusNiram is full?
22:07:56markunyes, that one.
22:09:01LinusNi have no idea what that could be
22:09:12LinusNcan you send me
22:09:37LinusNemail it to linus at
22:10:00markunI can also send it through IRC, but mail is ok too.
22:10:20LinusNi have problems with dcc
22:10:28LinusNbut try it
22:10:53 Quit preglow (Read error: 104 (Connection reset by peer))
22:11:49markunI could also be because I made some changes to the lcd_puts code..
22:13:00LinusNdcc worked .-)
22:13:16markunYes, and very fast.
22:13:19LinusNyou seem to have a huge font
22:13:42markunIt's quite big.. so that's the problem..
22:14:05amiconnWth does the font do in iram ???
22:14:15LinusNbeats me
22:14:32LinusNit's in bss, probably for the loadable font
22:14:32 Join preglow [0] (
22:14:36 Nick NibbIer is now known as Nibbler (
22:14:39preglowthe tale of the crashing server
22:15:21LinusNthe boot loader uses the firmware code for most things
22:15:39LinusNso if the font buffer is large, it won't fit in ram
22:15:48markunI had to change the #define MAX_FONT_SIZE for now to support my unicode fonts
22:15:50LinusNand the iram is all the bootloader has
22:16:19LinusNyou should build the bootloader from a clean cvs checkout
22:16:20preglowshouldn't the iram be more than enough?
22:17:37LinusNit should, but not with a huge font buffer
22:17:48preglowwhy do we need a font buffer in a boot loader?
22:18:02LinusNi explained it
22:18:06LinusNthe boot loader uses the firmware code for most things
22:18:08preglowait, i'll check it out
22:18:23LinusNand the firmware code has a font buffer
22:18:56preglowi read the irc log, got it
22:19:47amiconnWhy does the bootloader need a font at all? Doesn't it only check buttons and acts accordingly? I thought the bootloader doesn't do lcd at all...
22:20:52LinusNit displays error messages if something goes wrong, for example
22:21:41amiconnmethinks about the bootloader for archos rockbox in flash. This one doesn't do such fancy things...
22:21:45 Join Patr3ck [0] (
22:22:03LinusNyeah, this one is a lot more complicated
22:22:12amiconnBut then the archos bootloader doesn't include disk access etc...
22:22:25amiconn..because it doesn't need to.
22:22:57LinusNthis one == the iriver bootloader
22:23:31 Join webguest32 [0] (
22:23:42amiconnIt wouldn't need to for iriver rockbox as well if rockbox would be always flashed, coexisiting with iriver firmware in rom.
22:23:57DMJC-Lis the bootloader safe?
22:24:11LinusNamiconn: what makes you think that we can do that in this early stage?
22:24:29LinusNDMJC: soon
22:25:39amiconnLinusN: Why not? Imho starting from flash is even easier than loading from disk.
22:25:43preglowi'd rather have the bootloader for now, eys
22:25:52preglowit would be hell to develop if you had to flash it
22:26:13preglowi wonder how many flash cycles to chip in the h120 can take
22:26:19LinusNamiconn: because the we would have to reflash every time we want a new rockbox
22:26:33LinusNor i will keep on doing this all by myself
22:26:36LinusNwith the bdm
22:26:50markunI now compiled the bootloader from a clean cvs checkout, thanks. I got: MD5 (bootloader.bin) = 7002d30a58bc5f7ed28864caf8b3263a
22:26:56amiconnYes of course. However, I don't consider this a problem, if the flash chip is specced like the one in the archos
22:27:08 Nick Quelsaruk is now known as Quel|wc3 (~kvirc@
22:27:16amiconnThe archos flash chip is specced for 100 000 cycles...
22:27:19LinusNamiconn: there is no way of resurrecting the iriver if the flashing goes wrong
22:27:21preglowi think it can take quite a lot, yes
22:27:25preglowthey're usually under-specced as well
22:27:45amiconnLinusN: I don't think of flashing the *whole* rom, only the part where rockbox resides.
22:27:54preglowyou probably wouldn't have room for both
22:28:00LinusNand if rockbox crashes?
22:28:17DMJC-Lhow about making the damn thing work before trying to tweak it?
22:28:23LinusNDMJC: agreed
22:28:40DMJC-Lno end user cares what happens if it can't play their files
22:29:07amiconnLinusN: Hmm, good point. While you would still be able to run the box with the iriver fw, you might not be able to flash a new rockbox...
22:29:16webguest32amiconn: What are advantages of falshing it to the rom in the dev stage?
22:30:22 Join Ka [0] (
22:31:07amiconnI think it's easier (fewer points of failure to start) to start from rom than loading from disk
22:32:15LinusNyes, but since rockbox is very shaky in the dev stage, it is very likely to not work
22:32:45LinusNit will crash, and you'll have to reflash the entire rom with the flashing procedure in the original firmware
22:32:59LinusNrisking your iriver every time
22:33:06amiconnYes, understood
22:34:31amiconnOT: I'm currently investigating my strange "mmc dirs vanish if file is read past a certain point" issue. Strangely enough, while this is perfectly reproducable with 2 Ondios (mine and my sister's), Jörg didn't manage to reproduce it.
22:34:49LinusNgotta go
22:34:52 Part LinusN
22:34:56amiconnNow I have a test plugin, just reading the file to a certain point
22:40:41 Part markun ("Leaving")
22:48:25Quel|wc3amiconn: i think i can sell you some goats
22:48:45Quel|wc3if you find a bug no-one can reproduce.... you need a sacrifice ;)
22:49:20***Saving seen data "./dancer.seen"
22:55:04amiconnI was absolutely right about the position where this happens - it happens when reading the 11857920th byte, therefore completing the 2895th cluster
22:55:35amiconnAt this point there is a discontinuity - a new cluster chain starts at a different position on the card.
22:56:42amiconnI'd need a fat.c guru...
22:57:48Quel|wc3then don't look at me :)
22:58:02 Join ripnetuk [0] (
23:04:44*Bagder added quick-links at the top of each build-log
23:04:57Bagdergot tired of scrolling
23:05:53amiconnNice one :)
23:07:08*Bagder bows
23:07:24Quel|wc3quick link to the warnings and errors, isn't it?
23:08:51amiconnBagder: I rethought my fix for the failing player/ Ondio SP builds. I think the presence of these recording related functions should mainly depend on HAVE_RECORDING. Within that #ifdef, we can differentiate further (MAS3587 / MASNONE).
23:09:14amiconnIt wouldn't make a differece for now, but it would be the clean way, imho.
23:09:20BagderI agree
23:10:12amiconnMaybe it's not really worth to do this now, I presume that not much of the current mpeg.c will remain once the iriver port does it's impact...
23:10:16 Quit davFr ("ChatZilla 0.9.61 [Mozilla rv:1.7.5/20050105]")
23:10:34Bagdertrue, we should consider these changes temporary to get things to build
23:11:26amiconnOn the other hand, many "temporary" fixes may also lead to a mess. Temporary solutions tend to stay much longer than intended...
23:11:36Bagderthat is also true
23:12:05Bagderbut in the mpeg.c case, I don't doubt it'll change
23:12:40preglowany of you guys know where the __assert extern in assert.h comes from?
23:12:59Bagderprobably a gcc internal thing
23:13:06preglowwhy should assert be internal?
23:13:25preglowand no, gcc vomited on it, so it probably has to be in a lib somewhere
23:14:30preglowand if it was internal, why is it declared with EXFUN ?
23:14:41Bagderwell, I use assert() all the time and I never use any particular lib to get it to work
23:15:24preglowno, nor should you have to, i always thought assert() was pure macro magic
23:15:39Bagderit isn't in gcc
23:15:47Bagderit has more knowledge
23:16:42preglowi tried to compile mad with m68k-gcc yesterday, and it kept complaining about that
23:17:27Bagderread the comment in assert.h just above the extern?
23:17:30Bagder"The following is not at all used here but needed for standard
23:17:31Bagder compliance."
23:17:55preglowthat comment is not here :V
23:17:57Bagdernot that it helps
23:18:47preglowbut yes, i did a -nostdinc compile, and pointed to the rockbox include dir
23:18:58preglowas malloc is more or less the only extern mad needs, i thought that should work
23:19:52Bagderin a program I build here that uses assert...
23:20:08Bagder__assert_fail is the only __assert symbol
23:20:20Bagderwhen using nm on it
23:21:09*amiconn spots a strange formula in fat.c ...
23:21:16Bagdernm /usr/lib/*.a | grep __assert | grep -v fail
23:21:46Bagderthere's one
23:21:49amiconnBagder: Could you please double-check whether I'm not seeing ghosts - this is in for almost 3 years now.
23:22:26amiconnSee for a comparison of 2 very old fat.c versions
23:22:39*Bagder checks
23:22:48amiconnLook into fat_sec2cluster(), calculation of max_cluster
23:22:57preglowi'll just wait until the rockbox core is more operational for the 68k before i try again
23:23:12amiconnIsn't there a pair of parentheses missing in the newer version?
23:23:19Bagderpreglow: shouldn't take too long now
23:24:14amiconnI.e. int max_cluster = (bpb->totalsectors - bpb->firstdatasector) / bpb->bpb_secperclus + 1;
23:25:11BagderI agree it looks like that
23:25:31Bagderotherwise it makes no sense
23:25:45amiconnThat shouldn't cause the strange effect I get though, it's just a safety measure.
23:26:29 Quit methangas (" HydraIRC -> <- Try something fresh")
23:29:25 Join XShocK [0] (
23:29:53 Part webguest32
23:31:48 Quit ripnetuk ()
23:40:11 Join ripnetuk [0] (
23:42:51 Join [IDC]Dragon [0] (
23:43:15amiconnhi Jörg
23:43:26[IDC]Dragonhi Jens :-)
23:43:31amiconnBagder: Now it doesn't work correctly ...
23:46:58[IDC]Dragonmy Ondio backlight behaves well now, 23 mA :-)
23:47:18amiconnNice :)
23:47:32amiconnDid you read about my experiments with that strange file?
23:47:49[IDC]Dragona bit, catching up...
23:48:47 Join ashridah [0] (
23:50:30amiconnHmm. Now the file behaves strange - it stops a few seconds after the point where the directories vanish
23:51:04[IDC]Dragonamiconn: stops video, or your test plugin?
23:51:05amiconnMaybe it did this for a while now - didn't always listen & watch
23:51:11amiconnSTops video.
23:51:23[IDC]Dragonan underrun?
23:53:06 Join amiconn_ [0] (
23:53:46 Quit amiconn (Nick collision from services.)
23:53:47 Nick amiconn_ is now known as amiconn (
23:54:21amiconn[23:51:42] <amiconn> No, it simply stops, and falls back to the browser
23:54:49[IDC]Dragonthat is very strange
23:55:01[IDC]Dragonand the browser is OK?
23:55:18[IDC]Dragonmaybe a read error is treated like the file end
23:55:34amiconnIt shows the file. When I go up one level, the dir contents is gone.
23:55:53[IDC]Dragoncontent is that 1 file, right?
23:56:30amiconnNo, one level up it's the card's root, which contains 3 dirs
23:56:53amiconn...which are shown before video playback, and gone afterwards.
23:57:00[IDC]Dragonno, I mean the "It shows the file" part
23:57:12amiconnYes, the one file.
23:57:34amiconnThis is probably still in the browser list cache
23:58:09 Quit ripnetuk (Remote closed the connection)
23:58:16 Join ripnetUK [0] (
23:58:33ripnetUKyay the iriver sim builds again

Previous day | Next day