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 2006-01-18

00:00:00Moosand better random thing could be visible for guys using it a lot, no?
00:00:45amiconnThe old talk is in
00:00:52Paul_The_NerdI think the question should be less "what tangible improvements does it offer" and more "are there any drawbacks?"
00:00:56Bagderamiconn: right, that's the Modified BSD as well
00:01:22BagderGPL compatible
00:01:30Moosgo go go :)
00:02:02amiconnI wonder what we should do about the clause about providing the license
00:02:35 Quit TCK- (Read error: 110 (Connection timed out))
00:02:43Bagderput it in docs/ and we make buildzip include it
00:03:49amiconnHmm, this may get ugly very quick if we decide to include more code with similar licenses
00:04:08 Join frederic_ [0] (
00:04:11Bagder we add them all in a "additional licenses" file
00:04:37 Join akaidiot [0] (
00:05:23 Quit frederic_ (Client Quit)
00:05:38amiconnHmm, sounds like a reasonable solution
00:05:58 Join markun [0] (
00:06:01lostlogicthat's what other projects I've worked on have done.
00:06:09MoosSalute markun
00:06:24lostlogicBagder: should I do the same with that license from the GCC file that I borrowed from for profiling?
00:06:27markunGood day!
00:06:42Bagderlostlogic: yes, I think so
00:06:46amiconnHehe, changing the algorithm will break 'in-flight' shuffled playlists during the transition
00:06:59*amiconn wonders how many people will notice & complain
00:07:15Bagderonly those who read this irc log
00:07:19Paul_The_Nerdin-flight shuffled? Huh?
00:07:27lostlogicamiconn: wait −− the in-flight shuffling is done by saving the random seed used to shuffle it??? that's brilliant!
00:07:37Bagderlostlogic: exactly
00:08:07Bagderso the playlist is only kept as a single index per entry in memory
00:08:11amiconnPaul_The_Nerd: I mean playlists that are unfinished when installing the first build with the new algorithm. Resuming them will re-shuffled the list
00:08:51BagderI reaaaly don't explain myself very good at times...
00:09:41amiconnlostlogic: Yes, exactly. The resume position is saved as playlist position and track time, and the random seed in case of shuffled lists
00:09:47 Join YouCeyE [0] (
00:10:13amiconnIn fact, it's a bit more complex when lists built up in multiple steps come into play
00:10:13Bagdernow who's good idea was that? ;-P
00:10:15lostlogicI was just wondering about that this morning.
00:10:30lostlogicrockbox playlist handling is pretty brilliant
00:10:47amiconn.playlist_control basically stores the instructions how to rebuild the list
00:12:08 Quit uski (Remote closed the connection)
00:12:16amiconnHmm, is it correct that rockbox spins up the disk after _every_ track as soon as there are queued tracks? I presume this is due to updating mentioned .playlist_control?
00:12:41mirakwhat is the interst of doing this if mpeg2dec_s is a struct : typedef struct mpeg2dec_s mpeg2dec_t;
00:13:07Bagdermirak: it makes a type out of a struct
00:13:20mirakyou can't instanciate a struct ?
00:13:27Bagderits like an alias
00:13:36Bagderyes you can
00:13:41Bagderbut people like to typedef them away
00:14:20BagderI noticed yet another license we should add to that new file of licenses
00:14:52Bagder(the X11 license for the tools/iriver.c file)
00:15:17amiconnHmm, is that distributed?
00:15:30Bagderwell, it is distributed as source
00:16:09Bagderand the source refers to COPYING actually
00:16:10amiconnImho it should be sufficient for build tools to have the license included in the source
00:16:31BagderI don't think we gain much by trying to separate them
00:16:47Bagderbut sure, it would be sufficient
00:17:18Bagderthe devkit guys can deal with their own license violations then
00:17:56Bagderor aren't they shipping binaries of these?
00:18:32MoosGood night at all
00:18:41 Quit Moos ("Glory to Rockbox")
00:19:22 Part cYmen
00:22:27 Quit YouCeyE ("Leaving")
00:23:47 Join YouCeyE [0] (
00:25:11 Quit YouCeyE (Client Quit)
00:26:12mirakis it possible to know the sizeof of a field of a struct ?
00:28:18 Quit Skydemon ("bye bye")
00:31:43 Quit ender` (" sex is like math: - the clothes, + the bed, / the legs, ...and pray to god you don't multiply.")
00:43:30 Join Skydemon [0] (
00:46:39 Quit akaidiot (Connection timed out)
00:54:40 Join Membrillo [0] (n=sam_kill@
00:57:28***Saving seen data "./dancer.seen"
01:01:22 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
01:11:06 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
01:11:10 Part Midgey34
01:14:22 Join ehntoo [0] (
01:17:42 Quit lImbus (Read error: 104 (Connection reset by peer))
01:23:45 Nick Rob2222 is now known as Rob2222_ (
01:23:50 Nick Rob2222_ is now known as Rob2222 (
01:29:08mirakI finally spotted what was wrong !!!
01:32:39 Quit Skydemon ("bye bye")
01:32:57 Join ^Guest37784 [0] (
01:33:36 Join alphakiller [0] (n=bohas@
01:34:16 Join Skydemon [0] (
01:35:27 Join blue_beastie [0] (
01:36:17 Quit blue_beastie (Client Quit)
01:40:43 Quit Skydemon ("Miranda IM! Smaller, Faster, Easier.")
01:46:30mirakWARNING, bad file name lacks slash
01:47:49 Join Skydemon [0] (
01:48:43markunmirak: what was the problem?
01:49:05mirakdon't know the simulkator bitch about that
01:49:29mirakthe problem was that there was a ALIGN stuf delcared in the struct
01:49:37mirakand it wasn't working on rockbox
01:49:52markundoes it work now?
01:50:43 Quit Paprica (Read error: 110 (Connection timed out))
01:52:00mirakexept for the saving to files
01:52:17mirakI am going to use the rgb conversion of xvid
01:52:30markunAnd then display the images?
01:52:48mirakwill try
01:52:59markunI wonder if there will be a big speed improvement over xvid
01:53:02mirakI would like to see if they are decoded fine
01:53:03 Join manhattan [0] (
01:53:07mirakI am sure there will
01:53:18mirakit decodes really faster on the pc
01:53:53manhattanis the Gmini based on SH7034 CPU?
01:54:14gromit`which one ?
01:54:29gromit`gmini 1xx and 2xx are samsung calmrisc
01:54:36manhattanokay then.
01:54:45manhattanso no rockbox for it?
01:54:54gromit`the port has been started
01:55:01gromit`by jyp
01:55:19gromit`but currently the project is suspended
01:55:24gromit`we need more people
01:55:35manhattanby more people oyu mean coders?
01:55:42 Quit Skydemon ("Miranda IM! Smaller, Faster, Easier.")
01:56:03mirakwhat should ve the maximum supported data alignement for coldfire ? markun
01:56:18markunI don't know, sorry
01:56:23gromit` for more information on the chips
01:56:31gromit`there is a link to the rockbox wiki specific page
01:56:35mirakwith as many device supported rockbox will become huge !
01:57:13 Join Skydemon [0] (
01:59:47manhattandamn getting a decent mp3 is a nightmare! I've been spending the last couple of weeks trying to get a decent mp3 player that will play FLAC, MP3, and M4U. But none exist, and the closest thing is to get something that you can put rockbox on. But all rockbox supported players are either extinct or a fortune! :'-(
02:00:21Membrillocheck eBay
02:00:33Membrilloeverything is on eBay and its always cheaper
02:00:51Membrillobuy a H3xx or a H1xx iRiver. You won
02:00:56Membrillowon't be let down*
02:01:20Membrillothat being h320 h340 h120 h140
02:01:24manhattanbut I can't afford those, they're so expensive. :'(
02:01:56Membrilloyou won't find a decent DAP cheaper than those
02:02:28MembrilloDAP being Digital Audio Player
02:03:13 Quit alphakiller ()
02:03:46manhattanthe h120 on ebay don't include the remote.
02:03:57 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
02:04:19Membrillowell with a remote as well, its going to be even more pricey
02:04:35 Join Paul_The_Nerd [0] (
02:04:55eEyewe need rockbox for the x5!
02:04:58manhattanbut don't you need it?
02:05:14manhattanthe remote I mean, isn't it like necessary?
02:05:23Xerionremote sux :p
02:05:34Xerioni don't use mine
02:05:39Paul_The_NerdIt's hardly necessary, but I'm quite fond of mine.
02:06:23eEyeremote for the x5?
02:07:45Paul_The_NerdI thought it was the h120 remote being mentioned by manhattan
02:08:43manhattanyeah I was asking about the h120
02:09:49Paul_The_NerdThe remote is by no means necessary, but it's certainly convenient.
02:09:56 Quit Skydemon ("Miranda IM! Smaller, Faster, Easier.")
02:11:11mirakwas vorbis broken ?
02:14:03XavierGrWhat? I use my remote alll-the time.
02:14:26XavierGrIn fact I tend to use the player more with the remote than with out
02:17:31Membrilloive never used my remote
02:17:54Membrilloto me its just more cords to tangle
02:17:55XavierGrI can't really imagine my player without a remote.
02:17:58Rob2222lostlogic: there?
02:18:45 Quit einhirn (Read error: 104 (Connection reset by peer))
02:18:50Rob2222anyone here with a H3x0 and the H300-optimized-rockbox on it?
02:18:52XavierGrMembrillo: It is the way that you use the player. I combine it with a banana type bag. The main unit is in the waist-bag(banana) and the remote is clipped on it.
02:19:32Membrillooh ok. Like one of those bags with the one strap that goes diagonally across you?
02:20:12MembrilloRob2222: yes i do
02:20:40Rob2222could you fire up the unit, and give me a register of the PCF debug screen pls?
02:21:00Membrillodo you want me to grab the latest optomized build?
02:21:09Rob2222no, not needed
02:21:09XavierGrMembrillo: Not diagonically across the waist. (though it can be worn diagonically)
02:21:38MembrilloRob2222: ok loading it up now
02:21:40Rob2222if you have VIEW pcf50606 in the debug menu its enough
02:22:01XavierGroptimized rockbox?
02:22:17Membrillotype out that screen?
02:22:48MembrilloMBCCX 0x3C003A
02:22:59MembrilloMBCS1: 0x1D
02:23:10Membrillo00CS: 0x59
02:23:41Rob2222thats enough
02:23:51MembrilloINTX: changing constantly between 0x000001 and 0x400001
02:24:08Rob2222its the seconds interrupt of the real time clock
02:24:14MembrilloBatt: 4.12
02:24:26Rob2222thank you very much
02:24:27Membrillothat what you wanted?
02:25:17Membrillocan i ask, why does the drive spin up when shutting down?
02:25:58Rob2222dont know, but i think the config i saved
02:26:03Rob2222to disk
02:26:08Rob2222but i dont know
02:26:28Rob2222im relative new here, too
02:26:43XavierGrHas the power consuption been solved? (in H300)
02:26:48Rob2222and yes, that was what i wanted :)
02:27:03 Join akaidiot [0] (
02:27:34XavierGrA problem where Rockbox in H300 devreases run time.
02:27:34Rob2222XavierGr: was/is it a problem (for you) ?
02:27:49Rob2222i didnt recognized.
02:27:58Rob2222but i am not a developer
02:28:02XavierGrRob2222: I don't have an H300, but I am thinking to buy one.
02:28:24Rob2222XavierGr: I have no problem with the power
02:28:30Rob2222Have a H300, too
02:28:47MembrilloXavierGr: sorry?
02:28:51Membrilloim not a dev
02:28:53Membrilloi have no idea
02:29:25Membrilloif your asking me about the power issue that is
02:29:33XavierGrI think that some said that Rockbox can give much less runtime than default firmware.
02:29:46XavierGrBut I am not sure.
02:29:52Rob2222but i think (and i think i have read devs writing about that) that rockbox is near at the iriver FW
02:29:59Membrilloi heard that too, i think its true, but not significantly
02:30:11Membrilloit wont make much difference
02:30:19Rob2222i would say that, too
02:30:23Rob2222not much difference
02:30:25Membrilloi dont use my player for extended periods of time so it doesnt affect me
02:30:40XavierGrAny major differences between the 2 ports (H100, vs H300)?
02:30:54Rob2222I installed RB on the ipod Video of my cohabitant. damn thats a nice display ...
02:31:09XavierGrI mean you can play all the supported formats as in H100 right?
02:31:24XavierGrany problems or serious bugs right now in H300?
02:31:32MembrilloXavierGr: h100 has been developped for a lot longer. quite a few features have been added in h100 that havent been in h300
02:31:40Membrillono, h300 is quite stable
02:31:54Rob2222yep, 100% usable
02:32:06Rob2222(for all i need)
02:32:15Membrilloyou can only expect crashes with some plugins which were made by external coders that arent rb devs
02:32:31XavierGrMembrillo: I know that, but H300 is nearlty similar to H100 and most of the features are the same. That is why I am asking if there is a major difference
02:32:45Rob2222no, not really.
02:33:01Membrillono, no real difference in that respect
02:33:03Rob2222i think you can buy a h300 without a risk and start using RB
02:33:32Membrilloa lot of features still need tuning and stability for the h300, but its a lot more useable than the iriver fw already
02:33:52XavierGrI already have my precious H100 and I am consider buying an H300. USB OTG is really sweet.
02:34:02Membrillowhere do you live?
02:34:12XavierGrme? Greece
02:34:14MembrilloUS h300s dont have USBOTG i believe
02:34:16Membrillooh ok
02:34:18Membrilloyour good then
02:34:35XavierGrYes if I buy one I will get it from
02:34:42Rob2222 <= read the DREAMS ... thats cool
02:34:54XavierGrThat's what I read and freaked out.!!!!
02:35:22Rob2222but the US version has DRM... ^^ ;)
02:35:44XavierGrDRM sucks big time!!!!!
02:35:47Membrilloholy shit
02:35:56Membrillothat would be crazy
02:36:01Rob2222hehe :)
02:36:03MembrilloTV on your h300... hmmmm
02:36:18XavierGrOr HID support?
02:36:21XavierGrmaybe joysticks?
02:36:28Membrilloi wouldnt use that as much
02:36:30XavierGrImagine playing Rockboy with a joystick
02:36:51XavierGryes I know but only the possibility is marvelous
02:37:03Membrilloyeah i agree
02:37:12Membrillohaha and add a camera\ video camera in there
02:37:16Rob2222think about that situation: sitting in a bus with a saitek cordless joystick and navigating trough your H300 thats hidden in your pocket. LOL
02:37:29XavierGrBut that that would require a major porting effort.
02:37:44Membrillolike something out of bond. sitting in class, twist the top of your pen and it changes song
02:37:47XavierGrEvery device would need special code. so...
02:38:05Rob2222... lets start coding ;)
02:38:18Membrilloi need to get familiar with C first :P
02:38:19XavierGrSecond that.
02:38:53Rob2222damn, near 3am again
02:38:59XavierGrI can code, and have some for rockbox, not low level though.
02:39:12XavierGrgrrr it's 4am here I must sleep.
02:39:17Rob2222i was tired at 1am but didnt wanted to go sleep. now i cant sleep
02:39:41Rob2222well ok, i try
02:39:50Rob2222good night all
02:40:01XavierGrHa, I must sleep too. I will have to send the money via Bank transfer so if I want an H300 I must get up to go to Bank!! ;)
02:40:27 Quit eEye ("fractal2 mirc script (ver 1.0betar2) ·")
02:48:55mirakwhy did they reduced the plugin ram size ???
02:50:20 Join DJDD__ [0] (
02:54:48 Quit birdfish (Read error: 104 (Connection reset by peer))
02:57:29***Saving seen data "./dancer.seen"
02:57:33Membrilloabsolutely no idea
02:57:38Membrilloit seemed a bit strange
02:57:58Membrillothere must be some incentive behind it though. they wouldnt just do it for sake of changing sometihgn
03:01:50mirakwhat means incompatible model error when loading a plugin ?
03:02:52Membrillothe plugin hasnt been adjusted for the new plugin loader
03:04:25mirakwhat am I missing ?
03:04:29mirakthat's a plugin I did
03:04:55Membrilloim not sure. i dont know about the new plugin loader. you will have to ask an experienced dev
03:12:16 Quit ashridah ("Leaving")
03:14:54 Join YouCeyE [0] (
03:17:51 Quit mikearthur ("Konversation terminated!")
03:21:53 Quit ehntoo ("Leaving")
03:22:54mirakmarkun: crap that's as slow
03:27:55 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
03:31:53Paul_The_Nerdmirak: <−−- This shows the change that needs to be made to plugins for the new plugin loader.
03:45:44 Quit akaidiot (Connection timed out)
03:45:58 Join eEye [0] (
04:38:21BHSPitMonkeymirak: heh heh
04:57:30***Saving seen data "./dancer.seen"
05:00:52 Join Jungti1234 [0] (n=jungti12@
05:01:32Jungti1234 <- iPod imitation product
05:01:41 Join Rob2222_ [0] (
05:02:31Membrillowhat are they?
05:02:42Jungti123442g, 512m, 1.5 65000 color, playtime 13h
05:02:54Jungti1234Made in China.
05:03:01Membrillothey arent ipods?
05:03:12Membrillowho makes them?
05:03:18Jungti1234I don't know
05:03:27Membrilloi think they are ipods
05:03:30Jungti1234record support
05:03:36Paul_The_NerdThey aren't
05:03:38Jungti1234FM radio support
05:03:44Paul_The_NerdThe middle button isn't labelled on ipods, for one thing.
05:03:49Jungti1234video support
05:03:54MembrilloPaul_The_Nerd: yeah i saw that
05:04:01Membrillobut it could be a photoshop job
05:04:02Paul_The_NerdThere's a lot of iPod knockoffs right now
05:04:13Paul_The_NerdEngadget covers 'em all the time
05:04:23Membrilloi mean the black background and the reflection at the bottom is typical ipod advertisement
05:04:25Paul_The_NerdUsually, more features, but less storage. 512mb-1gb on the flash ones.
05:06:16Membrillowhen you google Kmobile it comes up with the CIA lol
05:07:02Paul_The_Nerd <−−
05:08:01Jungti1234kmobile - CIA
05:09:38 Quit DreamTactix291 (Read error: 110 (Connection timed out))
05:10:25Membrillo“???? ??? ???”
05:10:38Membrilloi tried copying some korean
05:10:41Membrilloit didnt work
05:18:42 Quit Paul_The_Nerd ("Leaving.")
05:19:35 Quit Rob2222 (Read error: 110 (Connection timed out))
05:27:55 Join akaidiot [0] (
05:36:56 Join ashridah [0] (
05:38:42 Nick Rick is now known as Rick2 (i=rick@unaffiliated/Rick)
05:38:50 Nick Rick2 is now known as Rick (
05:44:27 Quit Membrillo ()
05:46:56 Quit RotAtoR ("zzzzzzz")
05:50:40 Join JoeBorn [0] (
06:24:39Jungti1234Is no one here?
06:36:39 Join Membrillo [0] (n=sam_kill@
06:36:44 Join Drumr|6006GAH [0] (
06:37:15Drumr|6006GAHi happened upon a 60 Gb apple HD (with an apple photo attached) for $15
06:37:25Drumr|6006GAHno internal headphones connector
06:37:29Drumr|6006GAHbusted battery
06:37:40Drumr|6006GAHtryin to get the HD to work in my h320 now.
06:38:25Membrilloyou wont get an apple harddrive working in your h320
06:38:34Membrilloif thats what your tryinhg
06:38:36Drumr|6006GAHwhy is that?
06:38:54Drumr|6006GAHi know the file system is different, but what else keeps it from working?
06:39:01Membrilloread this
06:39:07Membrilloits locked
06:39:51Membrilloread that thread though
06:42:51Jungti1234Doesn't fwpatcher work in windows 98?
06:43:01Jungti123498 -> 9x
06:43:38Drumr|6006GAHcan it be used somehow as external storage?
06:43:49Drumr|6006GAH...probably not
06:43:57Drumr|6006GAHwe need DVD jon nthis case
06:44:05Jungti1234You don't know?
06:44:11 Nick Drumr|6006GAH is now known as Drumr|Defeated (
06:46:21 Quit akaidiot (Read error: 110 (Connection timed out))
06:46:32Drumr|Defeatedhey, since RB is ipod compatable. can there be a patch written for ipod drives in the h300s?
06:46:43Drumr|Defeatedit cant be that far fetched
06:48:22ashridahwhy would they be incompatible?
06:49:09MembrilloDrumr|Defeated: no the actual harddrive has its own locking mechanism
06:49:22Membrilloapple are clever buggers
06:49:57ashridahah, sneaky
06:50:14Drumr|Defeatedyou cant "emulate" an ipod for the h300?
06:50:24 Join DreamTactix291 [0] (
06:50:31Drumr|Defeatedoh, and how does this little metal doohickey fit back in the iriver?
06:50:31Membrilloit would have to be a damn good emulator
06:50:48Membrilloyou would have to emulate the ipod firmware
06:50:55ashridahDrumr|Defeated: i imagine it requires support from the ATA controller in the unit
06:50:59BHSPitLappythere better never be a podzilla.rock...
06:51:01Membrilloleaving you with an ipod in fatter casing
06:51:21Membrilloa slower ipod, in fatter casing
06:51:31Drumr|Defeatedi meant a HD controller emulatoroh well
06:51:34Drumr|Defeatedguess not
06:51:54Membrillobut it could well be the firmware that locks\unlocks the harddrive
06:52:23Membrilloyou could risk unplugging the harddrive while the ipod is running and the harddrive is unlocked ;)
06:52:35Membrilloi wouldnt advise that though
06:52:47Drumr|Defeatedi dont care
06:52:48ashridahactually, it could well be using a standard ATA interface to lock/unlock it
06:52:53Drumr|Defeatedi dont have anything to lose really
06:53:02Drumr|Defeatedi need to get the pod working first though
06:53:09Membrilloa perfectly good harddrive
06:53:12Drumr|Defeatedbusted batt, wont lemme do anything else with it
06:53:27Drumr|Defeatedwould it be compatable with my new 60 gig ipod?
06:53:32Membrillowell a battery will cost 20 dollars. a new harddrive will cost hundreds
06:53:35Drumr|Defeatedif so ill keep it for a backup
06:53:37Drumr|Defeatedgod forbid
06:53:52MembrilloDrumr|Defeated: yes, apparently you can switch harddrives between ipods
06:53:59Membrilloas long as its a legit ipod harddrive
06:54:07Membrilloso i hear anyway
06:55:44Membrillothere are a tonne of ipods on ebay that you can pick up cheaply because they have dud harddrives
06:56:13Drumr|Defeatedah, i see
06:56:17Drumr|Defeatedwell, i have the ipod
06:56:25Membrilloyou could easily buy one of those and use the HD
06:56:37Drumr|Defeatedi just need one cable (found for ~30 somewhere) and a battery
06:56:50Membrilloi dont mind if you try the harddrive in your h320. it would actually be a good experiment
06:57:04Drumr|Defeatedand i have a (very scratched) ipod photo 60 gig
06:57:15Drumr|Defeatedthrow a little rockbox, or ipodlinux on top
06:57:19Drumr|Defeatedand its perfect
06:57:21Membrillobecause i was thinking of buying my friends old ipod and using that HD. until i started that post
06:57:34***Saving seen data "./dancer.seen"
06:57:40Drumr|Defeatedwell, now i have a backup
06:57:45Membrilloif you do try it. Please talk to me here or PM me on MR
06:57:56Membrillowith results
06:57:58Drumr|Defeatedyou dont think you can shove a 60 gig into a 30 case right?
06:58:22Membrilloim not sure
06:58:26DreamTactix291i doubt it
06:58:28Membrilloare they GAH or DAL
06:58:29DreamTactix291they're pretty tight as it is
06:58:43Membrillois theyre both gah then you could
06:58:47Drumr|Defeatedive ben trying for the last hour to get the ipod drive to work with the h320
06:58:51DreamTactix291if they're both GAH
06:58:51Membrilloalthough with the size of ipods, i doubt that
06:59:00Drumr|Defeatedi would think
06:59:44Membrilloah ok. Well do you mind if i report this on that post Drumr? to warn people wanting to try this?
06:59:54Drumr|Defeatedplease do
07:00:02Membrillowhats your MR username?
07:00:30Membrillo60gb ipod photo harddrive?>
07:00:35Membrillois that what your trying?
07:00:42Membrilloand it wouldnt read?
07:01:41 Join drumrboy [0] (
07:02:06DreamTactix291[00:00] <Membrillo> ok
07:02:07DreamTactix291[00:00] <Membrillo> 60gb ipod photo harddrive?>
07:02:07DreamTactix291[00:00] <Membrillo> is that what your trying?
07:02:07DBUGEnqueued KICK DreamTactix291
07:02:07DreamTactix291[00:00] <Membrillo> and it wouldnt read?
07:02:19drumrboywel also i should add
07:02:28drumrboyi dont know how much th eplayer was abused
07:02:36Membrillothe iPod?
07:02:49drumrboyand... it has a small amount of rust exterblly from touching a corroded battery
07:02:55drumrboythe ipod
07:03:15drumrboythe rust is all external, and not near any seams though also
07:03:17drumrboyso.. yea
07:04:10MembrilloDrumrboy320 has just tried to place a 60gb iPod harddrive in his H320 unsuccesfully. Although the iPod was in lower quality than an average iPod, it is still not a good sign at all if you want to try this. This is confirmation that you cannot take a harddrive out of an iPod or use any harddrive with an apple symbol on it, inside another DAP.
07:04:14drumrboyif i open up my ipod, would apple be able to figure that out?
07:04:30drumrboysounds good
07:04:38DreamTactix291i think there is a ribbon on the ipods
07:04:40Membrilloto open it up you have to bend the outer casing quite significantly
07:04:41DreamTactix291something to show opening
07:04:43drumrboyso i shouldnt, to try and make sure the drive works?
07:04:49DreamTactix291or that
07:04:55DreamTactix291i'm so glad i have an H140
07:05:15DreamTactix29110 screws
07:05:23Membrilloapple are a bunch of facists :P
07:05:28DreamTactix291that they are
07:05:34 Join matsl [0] (
07:05:39Membrillohaha my h320 doesnt even have that
07:05:49Membrillo5 screws and flimsy plastic clips :P
07:05:53DreamTactix291no clips
07:05:58DreamTactix291metal casing :P
07:07:27drumrboyi tried to pryit open
07:07:29drumrboydidnt work
07:07:32drumrboyoh well
07:07:42drumrboyi tryed by hand and it wouldnt budge
07:08:11DreamTactix291i never cared for the final H300 design much
07:08:18DreamTactix291the prototypes looked nice though
07:08:47Membrillowhere did you get pictures of the prototypes?
07:09:09DreamTactix291seen them many times at misticriver
07:09:33Membrillooh ok
07:09:52DreamTactix291i use your rockbox updating app btw :P
07:10:17Membrilloyou do? terrific :) its working fine?
07:10:32Membrillogood good
07:11:02Membrilloare you using save\load settings?
07:11:32DreamTactix291you released a new one again?
07:11:51Membrilloquite a lot in the latest version
07:11:55DreamTactix291i'll find the link
07:12:37DreamTactix291that works well too
07:14:27DreamTactix291nice updates
07:15:15Membrillojust making it more convinient for the rockbox community to be lazy :)
07:15:48DreamTactix291yeah back in the day i had to do it all manually :P
07:16:04Membrillohaha oh how stenuous!
07:16:23Membrillotried it out?
07:16:41DreamTactix291i had to add two more albums to my player anyway
07:16:46Membrillooh ok
07:17:14DreamTactix291i haven't filled my player up yet
07:17:18DreamTactix291but could if i chose too
07:17:20Membrillowhat ive done is made a shortcut on my desktop, so now i just plug the player in, open the shortcut and unplug and theres my new build
07:17:36DreamTactix291would work for me except
07:17:40DreamTactix291i have zero icons on my desktop
07:17:41Membrilloive filled my h320 unfortunately
07:17:49DreamTactix291i had an H120
07:17:52DreamTactix291then filled it
07:17:56Membrillowhy is that?
07:17:59DreamTactix291have them disabled
07:18:02DreamTactix291i don't like them :)
07:18:07Membrillofair enough
07:18:14Membrillowhat dont you like about them?
07:18:21DreamTactix291i just think they're cluttery
07:18:21Membrilloicons are useful
07:18:34DreamTactix291i'm more likely to access stuff by quicklaunch
07:18:36DreamTactix291or cmd
07:18:41 Quit Drumr|Defeated (Read error: 110 (Connection timed out))
07:19:09Membrillofair enough
07:19:15Membrillowell me and icons dont go well
07:19:38drumrboyi have anout 3 more open spaces for icons on my desktop
07:19:46Membrillothe res of my monitor is 1280x 1024 and ive filled the desktop
07:19:55drumrboymine is 1028 769
07:19:55DreamTactix291that's a lot of icons
07:19:59drumrboy3 left
07:20:00DreamTactix2911280x1024 here
07:20:05DreamTactix291wish it was higher
07:20:15drumrboyyou kiddin? look at me
07:20:49Membrillo1028x 768????
07:20:51Membrillothats strange
07:20:57drumrboy1024 i think
07:21:00Membrillowide screen?
07:21:01drumrboywait ill check
07:21:05drumrboyhold o
07:21:18Membrillo1028x768 is quite strange
07:21:35drumrboy1024 x 768
07:21:46drumrboysry, 4 off
07:21:56drumrboymines a square
07:22:00drumrboyor close
07:22:11Membrilloah ok
07:22:31Membrilloyeah 1024x768 is quite standard
07:22:37Membrilloi have a 20"
07:22:42DreamTactix291i have 19" LCD
07:22:57drumrboymy seccond (nonexistant) monitor will do... 2048 1536
07:22:59Membrillomines the same but 20"
07:23:06drumrboyi guess my maximum supported size
07:23:11Membrillothat isnt bad
07:23:12drumrboythats ridiculous though
07:23:16DreamTactix291i want 2048x1536
07:23:26BHSPitLappy1280x800 w00t
07:23:31Membrilloyour pc will struggle displaying that many pixels
07:23:31drumrboyif i had teh monitor to do it, it would be dont
07:23:45DreamTactix291Membrillo: that's why if you do that you have two graphics cards :P
07:24:09Membrillotwo graphics cards haha. your tower is getting pretty pricey now
07:24:15drumrboyi have one
07:24:20drumrboysupports dual monitor
07:24:29drumrboyATI all in wonder 9600 Pro xt
07:24:33Membrilloyeah i have dual monitor too
07:24:37BHSPitLappymine supports triple
07:24:48Membrilloso you can scroll from one monitor across to the other
07:24:55MembrilloBHSPitLappy: niiiice
07:25:05DreamTactix291i only have on graphics card
07:25:09drumrboyi used to... one 15 CRT one 15 LCD, then the crt dies
07:25:09DreamTactix291i don't game so i don't need a great one
07:25:12BHSPitLappyi have my tv in the s-video
07:25:32drumrboyi have RCA and S video out
07:25:37drumrboydual VGA
07:25:44drumrboyi could support three screens
07:25:53drumrboyone has to be a repeated screen though
07:25:58DreamTactix291i wouldn't need that many
07:25:59BHSPitLappyor i can use my desktop laptop as an additional monitor for my laptop
07:26:14BHSPitLappythe laptop SHOULD support triple out, but it won't do them simultaneously
07:26:28drumrboytoo bad
07:26:33BHSPitLappysince it has vga and s-video outs
07:26:34Membrilloim happy with one big fat monitor
07:26:39Membrillowell its LCD
07:26:43drumrboymy dads old inspiron has svideo and vga out
07:26:43Membrillobut its big
07:26:57drumrboyinspiron 8000
07:27:16 Join EvilDude [0] (i=EvilDude@
07:27:36DreamTactix291if i had a laptop i'd have to have a USB soundcard with it
07:27:59drumrboyyea, my dad has his laptop playin music in our studio for about 5 minutes
07:28:11drumrboylol, sound sucked.. buzzed like insane
07:28:13CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
07:28:13*BHSPitLappy = inspiron 6000
07:28:24BHSPitLappythese speakers sound great
07:28:41drumrboymy dads dell 9300 has a subwoofer built in
07:28:49drumrboysounds great also
07:29:15DreamTactix291i'm too used to bookshelf speakers to go to computer speakers of any kind :|
07:29:29Membrilloi play all my music on Xbox Media Center
07:29:48Membrillobetter than any PC media player
07:29:50drumrboyi play all my music on my... ipod.. or iriver
07:29:53drumrboyor radio
07:30:03BHSPitLappyhey EvilDude
07:30:09Bgermorning :)
07:30:11 Nick drumrboy is now known as DrumRBoy320|Away (
07:30:12EvilDude'lo :)
07:30:18Membrilloyou plug your iriver into your speakers?
07:30:41DrumRBoy320|Awaynote everyone
07:31:03DreamTactix291my solution is foobar2000 -> kernel streaming -> Chaintech AV-710 -> S/PDIF -> Onkyo receiver -> speakers
07:31:50MembrilloI have iRiver -> PC -> Network Router -> Xbox -> TV -
07:31:56Membrillo-> Speakers
07:32:00Membrillostupid keyboard
07:32:43DreamTactix291i don't have a TV
07:32:47DreamTactix291or an Xbox
07:32:58Membrillowell, i dont use my xbox for games
07:33:06Membrilloi used to
07:33:15Membrillobut i modded it and now just use it for video and music
07:33:25DreamTactix291i just use my PC for that
07:33:28DreamTactix291foobar2000 for music
07:33:31DreamTactix291MPC for video
07:33:56DreamTactix291and i don't game so
07:34:00DreamTactix291i don't use anything for that :)
07:34:02Membrilloi dont game either
07:34:07Membrillohavent in the last couple of years
07:34:12Membrillomoved on from that phase
07:34:46Membrillowhat speakers do you have? 2.1? 5.1? Stereo?
07:35:06DreamTactix2912 8" Pioneer bookshelfs
07:35:09BHSPitLappy5.1! pro logic! w00t!
07:35:12DreamTactix291kinda cheap but not that bad
07:35:26DreamTactix291i have a semi small subwoofer too
07:35:30Membrilloprologics arent bad
07:35:51DreamTactix291i'm not big on 5.1
07:36:06Membrilloyeah, ive never had 5.1
07:36:18DreamTactix291i have no room for 5.1 anyway
07:36:45Membrilloif i bought an amplifier i could use my 2.1 stereo and my sisters stereo which is the same but without the woofer and have 5.1 but im happy with my 2.1
07:37:09DreamTactix291i just have no room for 5.1
07:37:09Membrillo2x 250 watt speakers and a 350 watt woofer is good enough for me
07:37:24Membrillosmall apartment?
07:37:35DreamTactix291just small room
07:37:45Membrillooh ok
07:38:07DreamTactix291old house
07:38:23Membrillomy set up is in a family room
07:38:41DBUGEnqueued KICK Jungti1234
07:38:41Jungti1234[14:43:02] <Jungti1234> Doesn't fwpatcher work in windows 98?
07:38:48DreamTactix291my desk takes up the whole room here
07:38:51Membrillon oidea
07:39:02MembrilloJungti1234: no idea*
07:39:08DreamTactix291unicode one won't
07:39:13DreamTactix291non-unicode should
07:40:35Jungti1234Do you have 5.1?
07:40:41DreamTactix291i don't
07:40:53DreamTactix291same here
07:40:57Jungti1234I have 2
07:41:03DreamTactix2912 isn't bad
07:41:13Membrilloim happy with 2.1
07:41:16DreamTactix291what kinda music do you people have loaded up on your players?
07:41:23Membrilloi wouldnt want to fork out for 5.1
07:41:26DrumRBoy320|Awaydream theater
07:41:35DreamTactix291yeah i knew about your drumr :P
07:41:36Jungti1234I dont know 'kinda'
07:41:41DreamTactix291kind of
07:41:43DreamTactix291type of
07:41:46CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
07:41:46*DrumRBoy320|Away had to goto the bathroom, and is checking up again
07:42:00*Jungti1234 heh
07:42:19Membrillomostly jazz\alternative\funk\soul etc. Im not a fan of pop or heavy rock
07:42:36DreamTactix291mostly progressive rock/metal for me
07:42:46DreamTactix291listening to Yes right now
07:42:49Membrillonot my style there :P
07:42:52Membrilloyes are cool
07:42:53 Quit matsl (Remote closed the connection)
07:42:58Membrilloi dont mind yes
07:43:02DreamTactix291Yes is some old school prog rock
07:43:12Membrilloim listening to Ben Folds now
07:43:25DreamTactix291i was listening to Symphony X
07:43:40Membrillodont know symphony x
07:43:45DreamTactix291progressive metal
07:43:52Membrillooh ok
07:43:58DreamTactix291i love prog metal
07:44:49Membrillohow do you define progressive? i hear so many different definitions
07:45:15Jungti1234I like a delectable song.
07:45:22DreamTactix291i mostly define it as lots of time changes and general verbosiveness and length of the music
07:45:29DreamTactix291i'm not as snobbish about it as a lot of people
07:45:36DreamTactix291i like songs that are 10+ minutes a lot
07:45:42 Quit Rob- (Read error: 110 (Connection timed out))
07:45:43Membrilloyeah same
07:45:55Membrilloa lot of my jazz stuff is very length
07:45:56DreamTactix291listening to one that's 11 right now
07:46:04DreamTactix291Dream Theater :D
07:46:17Membrilloyeah, i have a song which goes for 32min lol
07:46:26DreamTactix291longest i have is 42 i think
07:46:42Membrillomost of mine are 5-10 though
07:47:50Membrillodo you know of James Brown?
07:48:25Membrilloim going to see him next week, and recording it with rockbox :)
07:48:45Membrilloshould be
07:48:54DreamTactix291i love bootlegs :D
07:49:07Membrillohaha where are you from on this globe?
07:49:19DreamTactix291Nashville, TN, US
07:49:38Membrillobootlegging is not as tollerated in australia as it is in US
07:49:49 Quit Jungti1234 ()
07:49:52DreamTactix291it's not really allowed here
07:49:55DreamTactix291people still do it
07:49:57Membrilloyou have to do it quite sneakily
07:50:19Membrillopeople do it here, but if you get caught they usually take your stuff or kick you out
07:50:21DreamTactix291i've heard some really good bootlegs
07:50:22BgerXavierGr ?
07:50:55DreamTactix291you guys have a really bad lack of fair use laws
07:51:09Membrilloi have tickets for seats at james brown, so i think ill tape my mic to the end of the arm rest
07:51:32Membrillowhat do you mean?
07:51:53DreamTactix291according to australian users it's technically illegal to rip a CD there
07:52:10EvilDudeit is
07:52:14EvilDudeAustralia has stupid laws
07:52:18Membrillopretty nasty fines as well if they catch you (although thats pretty unlikely)
07:52:32DreamTactix291here CDs are ok because they lack any sort of copy protection
07:52:34DreamTactix291DVDs though
07:52:50 Join B4gder [0] (
07:53:00Membrillowell, ours have the same protection as yours i think
07:53:01DreamTactix291the day i can't backup what i own though is the day i stop buying
07:53:19Membrilloits just we have harsher punishments
07:53:29DreamTactix291CDs lack copy protection
07:53:33DreamTactix291except for the corrupt ones
07:53:46DreamTactix291Sony really did it with those rootkit CDs :|
07:53:55DrumRBoy320|Away**xony** cough coigh
07:54:08DrumRBoy320|Awaywow i cant type
07:54:10Membrilloyeah, well the point they make is, they put flimsy protection on, and just bypassing that makes the action illegal
07:54:29Membrillowhat are the rootkit cds?
07:54:44DreamTactix291sony released audio CDs that autoinstalled a rootkit when you put them in
07:55:01DrumRBoy320|Away... thats the way to do it though... make it so easy to break the law... that you do it w/o thinking
07:55:20Membrillo"breaking the law" has lost all meaning
07:55:25BgerMembrillo there was a bug in the sony's drivers (that prevent cd ripping) which are installed with windows' autorun
07:55:32Bgeror something like this ...
07:55:36DreamTactix291i have a large legal lossless audio collection
07:55:41DrumRBoy320|Awaytry maliciousl software
07:55:45DreamTactix291for everything i have that's not legal i don't care
07:56:01DreamTactix291not much short of a virus really
07:56:09Bgeryes ?
07:56:12MembrilloBger: ah... stupid sony
07:56:21Membrillosorry, accidently quoted your name haha
07:56:37Bgerhaha never mind
08:12:25 Join ender` [0] (i=ychat@
08:20:12 Quit `3nergy (Connection reset by peer)
08:25:54 Join LinusN [0] (
08:29:05 Join stamppot1975 [0] (
08:35:28 Quit perplexity (Read error: 110 (Connection timed out))
08:50:04 Join perplexity [0] (n=joust@
08:51:29 Join jborn_ [0] (
08:51:47 Join EvilDude9 [0] (i=EvilDude@
08:52:34*B4gder does the rare booting-with-jffs2-rootfs-on-his-cfi-flash dance
08:53:53dwihnoB4gder: flash what who the toot boot?
08:53:56 Quit EvilDude9 (Client Quit)
08:54:28B4gderbooting with roofs on the flash of my customer's board
08:55:02dwihnoWhat kind of flash?
08:55:03B4gdersome parts of linux is just so without docs
08:55:14B4gdera 16MB CFI intel flash
08:55:21dwihnoI got some flash toys yesterday
08:55:32 Join jborn__ [0] (
08:55:58dwihnoSounds like a lot of fun to play around with! :)
08:56:09B4gderit is
08:56:31B4gderlinux 2.6.15 on this arm9. 32MB ram
08:57:36***Saving seen data "./dancer.seen"
08:57:52 Join Zagor [0] (
08:58:12B4gdernow if I could only _write_ to the fs too...
08:59:29dwihnoSmall steps, grasshopper :)
08:59:50B4gderveeeery small
09:00:08 Join einhirn [0] (
09:08:07 Quit JoeBorn (Read error: 110 (Connection timed out))
09:08:34 Join JoeBorn [0] (
09:10:08Zagorhi joe
09:12:25 Quit EvilDude (Read error: 113 (No route to host))
09:13:00 Quit jborn_ (Read error: 110 (Connection timed out))
09:13:28Bgerstupid q, but ... what's the "empty" string for s/something/"empty"/g ?
09:14:01B4gder"s/something//g" ?
09:14:17B4gderthat is the replacement
09:14:33 Quit jborn__ (Read error: 110 (Connection timed out))
09:14:38B4gders/replace this/with that/g
09:14:44Bgeryep, but ...
09:15:09Bgerokay, i want to remove the +-es of the diff (entire new file)
09:15:17Bgerin vim
09:15:28B4gder\+ probably
09:15:34B4gder+ is regex operator
09:15:49Bgerin this case - no
09:16:22B4gderand you do want to remove "+ " if I'm not mistaking
09:18:01Bgerit seems it's not extended regexp
09:18:15dwihnovim's regex is fishy
09:18:21Bgerah, i'm dumb (as usual)
09:18:31Bgerforgot the % for all lines
09:18:48Bger10x anyway
09:18:56dwihnoPerl compatible regex in vim would be swell :)
09:19:10B4gdersed is your friend! ;-)
09:19:27BgerB4gder why in sed when i can do it in vim :P
09:19:40B4gderhey, you couldn't ;-)
09:19:52*B4gder ducks
09:20:21 Nick paugh is now known as AliasCoffee (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
09:27:06Bgeryep ? ;)
09:27:59 Join bshift- [0] (
09:27:59 Quit eEye (Read error: 104 (Connection reset by peer))
09:29:21preglowdo a quick plug/deplug/plug with usb, and rockbox will hang
09:31:24AliasCoffeepreglow: tried with iriver f/w? maybe the hardware doesn't like it.
09:31:32Bgerthis is a feature :D
09:31:41BgerAliasCoffee i don't think it's a hardware problem
09:32:20 Join Febs [0] (
09:36:53B4gderI talked to Linus about the SWAP_WORDS in ata.c yday
09:37:11B4gderI think we should rename it to ATA_LITTLE_ENDIAN or similar
09:37:56B4gderit'll make the code more obvious
09:39:23 Join eEye [0] (
09:39:24 Quit bshift- (Read error: 104 (Connection reset by peer))
09:41:15 Quit uwe_ (Read error: 104 (Connection reset by peer))
09:49:44 Quit DreamTactix291 (Read error: 110 (Connection timed out))
09:53:49 Quit AliasCoffee (Remote closed the connection)
10:02:47 Join BHSPitLappy2 [0] (
10:02:55 Quit BHSPitLappy (Nick collision from services.)
10:03:03 Nick BHSPitLappy2 is now known as BHSPitLappy (
10:11:14 Join Kohlrabi [0] (
10:11:20LinusNamiconn: u there?
10:15:10 Join amiconn_ [0] (
10:29:58mirakso I managed to make mpeg2 decompress
10:30:03mirakbut that's as slow as xvid
10:30:48markunmirak: with some things in IRAM?
10:31:49mirakthe ram seems to be a boottleneck for sure
10:31:52Bgeras slow as xvid without IRAM or as slow as xvid with IRAM
10:32:07mirakas slow as xvid without iram
10:32:24mirakwell maybe 3 fps when xvid is 2
10:32:43 Quit amiconn (Read error: 110 (Connection timed out))
10:32:43 Nick amiconn_ is now known as amiconn (
10:32:43mirakI didn't do a precise benchmark
10:32:55mirakthat's pointless anyway
10:33:15Bgerbtw what xvid files did you test ?
10:33:30Bgerfor example the iriver's decoder doesn't support some kind of frames ...
10:34:08mirakBger: I tried a xvid stream I used for the iriver firmware
10:34:29mirakand they have sound in plus of that ...
10:34:59mirakI imagine them working as nuts 24/7 to have that think optimised
10:35:01Bgerbut i guess the sound takes only 20MHz or so of the CPU
10:35:09XavierGr3 fps wow that is slow?
10:35:26XavierGrany known problems?
10:35:33mirakwell I think that's more 2 anyway ...
10:36:00Bgerbut they use the motorola's decoder ...
10:36:10mirakalso it displays nothing, I just display the frame number
10:36:14 Quit JoeBorn (Read error: 110 (Connection timed out))
10:36:17XavierGrwell you have to optimize it by at least 8-10 fps. no?
10:36:19Bgerand our mad wants ~ 45MHz for 128kbps mp3
10:36:21mirakBger: what motorola ?
10:36:45mirakBger: you name is like Bigger ?
10:36:50Bgermirak there is a rumour that iriver h1x0 use the motorola's mp3 decoder
10:37:02Bgermirak that's not my name :)
10:37:10mirakBger: ? there is a hardware decoder ?
10:37:11Bgerspell it as u like
10:37:14Bgermirak nooo
10:37:24B4gderit is quite likely they just bought a codec, yes
10:37:25Bgersoftware one, but not license-compatible with rockbox
10:38:18preglowi think iriver has free access to motorolas decoder
10:38:19Bgerall i wanted to say is that we will need much more work to achieve 10fps .....
10:38:26preglowwell no shit
10:38:42Bgerand i don't know whether this is possible with mad for sound
10:38:43mirakok that's just 2 fps, I lied
10:39:24 Join linuxstb [0] (
10:39:39mirakalso the plugin ram was shrinked to 512kB
10:39:47mirakxvid doesn't fit in that
10:40:01Bgermirak that's the least problem now ... :)
10:40:03mirakat least not as it is
10:42:14mirakthe problem I had with the code was that there was a ATTR_ALIGN(64) in a structure
10:42:20 Quit DJDD__ ("Trillian (")
10:42:25mirakit wasn't working
10:43:04linuxstbB4gder: About the SWAP_WORDS, I agree it's a confusing define. I'ld be happier with defining the endianess of the ATA hardware as well.
10:45:24 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
10:45:53linuxstbWe can just use that, along with the letohe16() and betohe16() macros instead of swap16(). I can look at this this evening if no-one else wants to do it.
10:46:23 Join JdGordon [0] (
10:51:37Zagorntohs() is the standard macro/function for that
10:51:59B4gderfor network order, yes
10:52:42linuxstbZagor: We've got a set of byte-swap macros in system.h - they were added a few months ago
10:52:49Zagorwell do we different macros for different purpose?
10:53:11B4gderin this case I think the macros fulfill a purpose
10:53:57linuxstbThey are generally used when dealing with binary file formats - e.g. the codecs and bitmap saver.
10:54:30Zagorfine with me
10:57:35 Quit linuxstb ("CGI:IRC")
10:57:39***Saving seen data "./dancer.seen"
11:10:00 Join needleboy [0] (
11:10:15needleboygood morning
11:10:30needleboyanyone patient enough to help an idiot?
11:10:46needleboyi need some help with creating .patch files
11:14:09ashridahi can try
11:14:38 Quit Xerion (Read error: 110 (Connection timed out))
11:16:59needleboynevermind, seems i've handled it for now
11:17:01needleboythanks anyways!
11:17:04 Join Xerion [0] (
11:20:21amiconnZagor: ntohs() & co are only for converting between network (i.e. big endian) and host order. The have no complement for little endian.
11:20:36amiconnletoh16() & friends have that, and they seem to be somewhat standard
11:21:56JdGordonhey all, finally back at home from holidays
11:22:04 Join DreamTactix291 [0] (
11:22:22JdGordoni managed to update rb a few times and 1 thing stumped me... why is the volume -30% instaed of 70%?
11:22:45Zagorit's not % anymore. it's dB.
11:32:20 Join uwe_ [0] (
11:36:28mirakis there a variable that can tell the stack size ?
11:37:47preglowbut you can check rockbox sources and see what you see
11:37:50 Quit Febs (Read error: 113 (No route to host))
11:38:36preglowcodec stack size, for example, is DEFAULT_STACK_SIZE + 0x2000
11:39:04preglowand DEFAULT_STACK_SIZE is 0x400
11:39:12preglowwhich is what most parts of rockbox use for their stacks
11:39:30preglowhowever, plugins use the main stack, i think
11:39:57mirakI already asked something like that, but when a a struct is declared and the size of that struct is around 16kB is it put on the IRAM stack ?
11:40:19preglowit is put on the stack if you put it here
11:40:25preglowthat is, if you declare it a function scope
11:40:34preglow16kb sounds too much
11:40:46mirakthere is an array in it
11:40:53pregloweven codec stack would be smashes by that
11:41:03mirakso how is it possible ?
11:41:09preglowit isn't, you can't use the stack for it
11:41:19mirakbut it works !
11:41:30preglowthen it's not on the stack
11:42:01preglowdepends on how big the main stack is
11:42:10preglowi kind of assumed it wasn't bigger than the codec stack
11:42:16mirakthe stack growth in IRAM ?
11:42:32 Join `3nergy [0] (
11:43:09mirakwhere is the stack ?
11:43:50mirakI think I will let that optimisation thing to interested people
11:43:51 Quit needleboy ("CGI:IRC (EOF)")
11:44:26preglowmain stack is 0x2000 big
11:44:43preglow8192 bytes
11:44:44preglowso no
11:44:49preglowthat struct is not on the stack
11:44:56 Join Paul_The_Nerd [0] (
11:45:07preglowif you define a variable in a function, then it's on the stack, unless you declare it static
11:45:08 Join steveb [0] (n=dave@ircguide/director/about/fish/about/uk/about/seo/steveb)
11:45:15amiconnmirak: The main stack size is 0x2000 (8KB)
11:45:32amiconnThis is an embedded system; don't put large structs on the stack
11:45:42preglowyou can in stuff like decoders
11:45:45preglowbut be careful
11:45:59stevebHi, i just patched my firmware as per the instructions. it said it patched it fine. i installed all the rockbox files in to the root of my iRiver H340 but it only loads the iriver fw...
11:46:03stevebany ideas?
11:46:55mirakamiconn: so why isn't it just crashing ?
11:47:05preglowlet me see the code
11:47:20mirakok I put it on the tracker since it works
11:47:32ashridahsteveb: did you actually flash the firmware using the iriver's firmware flash option?
11:47:50stevebashridah: yep. it shut itself down and everything
11:48:13stevebodd huh?
11:48:32ashridahi'd try patching it again, i guess
11:48:39preglowbrb, lunch
11:49:01ashridah(patching the .hex again then flashing, i mean)
11:49:14ashridahmaking sure you copy it to the unit and overwrite the one that's there
11:49:27stevebi patched it in situe
11:50:43Paul_The_NerdDoes it boot straight into iRiver, or does it very briefly show something else first?
11:50:55stevebstraight in to iRiver
11:52:59JdGordonwhat bootloader is the h300 upto?
11:53:13Paul_The_NerdThen it is very likely that somehow either the firmware file wasn't patched, or the file you copied over and flashed wasn't the right one.
11:53:36JdGordonanything majoy change since 3?
11:55:05ashridahbootloader usb mode. although people have reported the occasional bug with v5 (nothing like v4 tho) so user discretion is advised
11:56:59Membrilloanyone here a MR mod?
11:57:25LinusNashridah: "user discretion"?
12:00:20stevebok it isnt loading up the patched fw for some reason...
12:00:36ashridahLinusN: between v3/v5 i guess
12:01:06LinusNashridah: why not use v5?
12:01:19ashridahLinusN: didn't some people report some minor bugs with it?
12:01:39LinusNyes, sometimes the bootloader usb mode doesn't work
12:01:56LinusNbut then you don't use the bootloader usb mode
12:02:10stevebmeh... just tried patching again and the damn thing just stays on the old one... i wonder whats up with it
12:02:26mirakok I uploaded a mpeg2 plugin on the tracker. If someone is interested to optimise that ...
12:02:32stevebit should be called H300.hex right?
12:02:33LinusNafaik, there are no bugs that should make you go back to v3
12:02:50ashridahLinusN: aah, righto
12:03:22ashridahdoesn't fwpatcher.exe create a copy of the file and patch that?
12:03:26*ashridah can't remember)
12:03:39stevebthe filesize changed
12:03:43steveband modified date
12:04:41preglowit creates a copy and patches that, yes, but then it overwrites the original one
12:08:20stevebwell for some reason my firmware isnt being patched at all because when it starts up it says 1.29E when i patched a 1.28E fw
12:10:58preglowPaul_The_Nerd: any negative experiences with using gcc4 on arm?
12:11:57Paul_The_Nerdpreglow: So far, not a one.
12:12:22preglowi'm considering asking one of the big boys to put gcc4 on the server
12:12:38Paul_The_NerdBoth the bootloader and the normal build no problem, and everything seems to work at least as well as before.
12:12:43preglowi don't know if it's even worth testing it in and out, gcc3 makes wrong code, gcc4 does not
12:13:30stevebgcc3 is evil.
12:13:32JdGordongaaaahhhh... the h300 optimized build on mr has a horrible colour..
12:13:40preglowgcc3 is nice, but it has bugs on arm, apparently
12:13:50stevebi am not a fan.
12:13:55Paul_The_Nerdpreglow: It may affect the IFP port though, isn't that Arm as well?
12:14:00B4gdersteveb: for coldfire, gcc3 makes better code
12:14:06preglowPaul_The_Nerd: indeed it is
12:14:07B4gderfan or not
12:14:12preglowfor coldfire gcc3 is recommended, yes
12:14:23stevebmeh. dont really know anything about that ;) i have had many problems with gcc3 though in the past
12:14:33B4gderand not with gcc4? ;-)
12:14:41stevebnot so many surprizingly
12:14:46B4gder4.0 was horrible
12:15:15mirakanyone have an ipod here ?
12:15:21stevebtrue. i never used that one
12:15:32Paul_The_Nerdmirak: Nano here.'
12:15:54mirakit would be intersting to have a test of mpeg2 with the ARM idct.c assembly optimised of ffmpeg
12:16:13mirakto see if this give a significant performance boost
12:16:31mirakbefore trying anything for coldfire
12:16:33preglowgcc4 wont reach its full potential until at least 4.2
12:18:07JdGordongrr... the reset button is too damn small
12:18:15JdGordoni cant find a pin
12:18:27Paul_The_Nerdpreglow: ifp port compiles without complaint, but it's not like that says too terribly much.
12:18:35Paul_The_NerdI don't exactly have one to test 'er on.
12:19:24JdGordonoh goody... it seems to crash loading the wps...
12:21:38preglowdoesn't matter that much, the server doesn't create dailies for ifp anyway
12:21:50preglowand i seriously doubt gcc4 will start making wrong code for ifp
12:21:53 Join steveb___ [0] (
12:22:05 Part steveb___ ("oops")
12:22:41B4gderI'll change the h300 sim build to use SDL in the build table
12:22:55B4gderand the other color target sims too
12:25:08Paul_The_Nerdpreglow: Well, GCC 4.0.2 gets my vote.
12:25:30preglowanything major left to do for the sdl sim?
12:25:45B4gderpreglow: verifying the cygwin part
12:25:56Bgerpreglow are you sure it's gcc's problem ?
12:26:16preglowBger: pretty much
12:26:36 Join Jungti1234 [0] (n=jungti12@
12:26:36preglowif not, the problem we have is even weirder
12:26:56Bgerand what about the case of (for example) some kind of buffer overflow (lol), and just the 4.0.2's code doesn't exploit it
12:27:43Bgersorry if saying/talking bullsh.ts, i don't know the situation...
12:28:02 Join akaidiot [0] (
12:28:20 Join webguest73 [0] (
12:28:22preglowit's far more low-level than that
12:28:33 Quit webguest73 (Client Quit)
12:28:39preglowPaul_The_Nerd did a pretty in-depth test, and it all boils down to conditionals
12:28:42preglowno buffers
12:28:58Bgerthat's why i said "sorry":)
12:29:09preglowdidn't see that
12:31:11Membrilloyay!!! ive managed to clean 850mb of stuff off my h320! that leaves..... 830mb free
12:31:20Membrillo880 sorry
12:31:28Membrillohaha that was embarresingly stupid
12:36:09Membrillowhat does windows media player have against my iriver.... whenever i right click on a file on my iriver it pauses the song im listening to.
12:36:22Membrillobloody bill gates
12:36:31Membrillofacist dictator he is
12:36:39Membrillomakes me pause my songs
12:36:45B4gderyet you use his products
12:36:51Membrilloill go and pause his song if you know what i mean
12:37:16ashridahwouldn't bother bitching too much. usb stacks tend to be fairly heavy on the processor
12:37:21ashridah(for usb storage anyway)
12:37:36Membrillooh well
12:37:39ashridahdoesn't help that you can't really do simple dma transfers
12:38:03ashridahget a better music player that caches more of the file while decoding
12:38:31B4gderthere are others? ;-P
12:38:44 Join ST [0] (
12:38:59ashridahi've always stuck to quintessential when using windows (doesn't happen too often)
12:39:01B4gderthe latest rockboy patch doesn't build on h1x0
12:39:04Membrillolol im too lazy. i dont listen to music on my PC very often
12:39:24B4gderregion PLUGIN_IRAM is full
12:39:48Membrillothats probably because ram was dropped for pluggins earlier. read that somewhere
12:40:24B4gdernot at all
12:40:40B4gderIRAM is not plugin ram
12:40:55B4gderand it builds fine for h300
12:41:23Membrillowell i had no idea what i was talking about. Thought i could try and make myself useful though
12:41:31Membrillo :)
12:42:37MembrilloSuper Mario Bros 2- 6 Golden Coins runs really well on H3xx rockboy now
12:43:20 Quit perplexity (Read error: 110 (Connection timed out))
12:43:27preglowfull for h1x0 but not h3x0?
12:43:37B4gder 0x100180ec iend = .
12:43:44B4gdercompared to
12:43:46B4gder 0x10017e98 iend = .
12:43:55B4gder0xec bytes too many
12:44:18preglowwell, so it's got some CONFIG_CPU based iram decisions, then?
12:44:27B4gdermust have
12:44:27 Join Rob- [0] (
12:47:36B4gderthe lcd code for example
12:47:47 Join petur [0] (
12:49:49 Join linuxstb [0] (
12:50:03preglowlinuxstb: what do you say we put gcc4 for arm on the server?
12:51:43B4gderok, the problem is that the 'scan' struct uses a larger buffer for the 2 bit lcds
12:52:18linuxstbpreglow: I haven't had chance to test it.
12:52:52amiconnLinusN: Speaking about bootloader USB mode - the H300 bootloader seems to cut hd power for a short time before entering USB mode which leads to a nasty click
12:52:59amiconn..aka HD emergency park
12:53:11linuxstbBut yes, I think we should put gcc4 on the server - but we can wait a couple of days, to give us all a chance to use it and make sure there are no problems.
12:53:19LinusNamiconn: see "known issues" on IriverBoot
12:53:46linuxstbAnd also, the latest Rockboy patch doesn't fit in the ipod's IRAM either.
12:54:13preglowno surprises there
12:54:26linuxstbNope. There is quite a bit of ICODE in Rockboy now I think.
12:54:26XavierGrBagder: seems like a nice, is it realtime now?
12:55:15MembrilloWhat kbps on OGG should I aim for that is around the equivalent file size of 160kbps mp3?
12:55:30mirakis there a patch size limit on the tracker ?
12:55:35linuxstbMembrillo: Er, 160kbps?
12:55:44mirakMembrillo: q 5
12:56:09ender`wouldn't q4 be closer?
12:56:16Membrillohmmm. if i go 160kpbs the filesize is a lot smaller than the mp3
12:56:25mirakender`: probably yes
12:56:28mirak4 or 5
12:56:29Membrilloits strange, because i didnt think it should be so
12:56:42ender`4,5 :)
12:56:42Membrillousing dBpowerAMP
12:56:54preglowMembrillo: so what? quality is better
12:56:58preglowdbpoweramp ://
12:57:05ender`let's go fractional :)
12:57:20mirakpreglow: is there a size limit on the tracker ?
12:57:26mirakof the patches
12:57:28Membrilloill go 192 or 224
12:57:32preglowmirak: no idea
12:57:41***Saving seen data "./dancer.seen"
12:57:43preglowMembrillo: why do you care so much about sie?
12:57:58peturmirak: zip it?
12:58:03Membrillobecause i only have 800mb left after deleting a whole lot of stuff
12:58:12Membrillospace is something im short on
12:58:18preglowthen you shouldn't try to use more of it
12:58:21preglowlike you're doing right now
12:58:34preglowif the files end up smaller than 160kbps mp3s, then hooray!
12:58:40mirakpetur: it's zipped but I included a mpeg2 stream test file
12:58:44Membrillohaha fair enough
12:58:45mirakof 2megs
12:58:52preglowq5 will sometimes make smaller files than 160kbps, sometimes larger
12:59:11mirakI think you should use q6
12:59:12preglowlike the parameter hints at, it's a quality thing
12:59:18Membrilloim new to ogg and im not sure how it equivalates
12:59:21preglowand i think he should try it out himself
12:59:26peturmirak: why not put a link to that test file in the comment and put the file somewhere on the net
12:59:32preglowi use q5 and it's just dandy most of the time
12:59:43Membrillogood good
12:59:46Membrilloill have a play around
12:59:47mirakpetur: well that somewhere on the net is the problem, I have no webspace
13:00:04Membrilloi just got shure e2cs and i wont to be able to use them properly
13:00:17Membrillobut ive also got to balance with space
13:00:20Membrilloits a nasty issue
13:01:05Membrilloare Q4 and Q5 etc only for VBR?
13:01:43crwlthey're all VBR modes
13:01:44Membrillofor CBR it seems to be measured in kbps but when i changed to VBR its measured in Qx
13:02:39crwlall the VBR modes have a "nominal bitrate" (for example q5 = 160 kbps, q4 = 128 kbps) but they don't necessarily mean much
13:02:55Membrilloah ok
13:02:57crwlit's just a bitrate they'll quite often end up around
13:03:08crwlan average bitrate, that is
13:03:27Membrilloim encoding to Q5 now. ill tell you the ratio of lossless to Q5
13:03:29preglowMembrillo: there is no cbr for vorbis, so just forget it
13:03:31Membrilloin this case
13:03:46Membrilloreally? dbpower says i can have CBR or VBR
13:03:52preglowdbpoweramp is crap
13:03:58Membrillobad dbpoweramp
13:04:09Membrillowhat should i use to convert between formats then?
13:04:12preglowyou can have pseude-cbr
13:04:17preglowbut you don't want that
13:04:21mirak patch is here, test file is here
13:04:25preglowit's still vbr, just worse quality, and tries to be cbr
13:04:34Membrillooh ok
13:04:35Paul_The_NerdWhy's this dbpoweramp crap?
13:04:45Membrilloso what freeware do you suggest for encoding?
13:05:03MembrilloPaul_The_Nerd: i dunno, it seems to be pretty good for me
13:05:06mirakMembrillo: q6 is cd quality I think
13:05:11Membrillobut apparently its crap
13:05:22mirakat q7 I can't do a difference between a wav and ogg, and the size maybe 10% bigger
13:05:33mirakthan q6
13:05:44preglowand that's you, there's no Grand Right answer with encoders, you need to test yourself
13:05:58Membrillofrom lossless flac to Q5 ogg VBR the filesize has gone from 17.711mb to 2.706mb
13:05:58preglowunless you want to waste space, in which case you can just listen to anyone who goes the highest
13:06:00crwlif q6 is CD quality, then q7 must be better than CD, eh? :)
13:06:08mirakI am against bitrates below 192kbits :D
13:06:20crwllossless is CD quality
13:06:25preglowi'm against people who think in terms of bitrate
13:06:28preglowbut that's just me
13:06:38Paul_The_Nerdmirak: The mpeg2 patch, which targets should it be aimed at?
13:06:54mirakPaul_The_Nerd: I think any should work
13:06:55Membrilloif i had a chunky harddrive i would stick at 192 or higher
13:07:14crwlfor my Iriver I encode to ~128 kbps Vorbis (q4) with the newest AoTUV whatever encoder (because I hear it's better and somewhat faster) and I think it's just fine for portable use
13:07:19Membrillobut now im at mostly 128kbps mp3s and some oggs. yuck
13:07:26crwlfor home use, I use FLAC
13:07:40mirakPaul_The_Nerd: I have used IRAM IDATA_ATTR I probably shoudln't have used them since I am not sure if it will work with other devices
13:07:42crwlthis is a good combination, because it's so easy to encode from flac to vorbis
13:08:02mirakMembrillo: what's your device ?
13:08:08Membrillothe size of this album has gone from 293mb to 45.3
13:08:29mirakMembrillo: 20G is a lot of space man !
13:08:33Jungti1234Can't rockbox go to folder of a song that returned to life on last if stop a song?
13:08:36mirakI have one too
13:08:46Membrilloive filled it unfortunately
13:08:51Membrillowith 128kbps mp3s
13:10:27Paul_The_Nerdmirak: I'm not entirely certain what to do with the files in your patch.
13:10:35*ST np: Rage Against The Machine - Killing In The Name Of [05:14m/128Kbps/44KHz]
13:10:57JdGordon.. fuck you i wont do what you tell me.. fuck you i wont do what you tell me.. fuck you i wont do what you tell me
13:11:00B4gderST: please shut off that
13:11:20*Membrillo likes that song
13:11:33*Membrillo doesnt like much RATM but Membrillo likes that song
13:11:42*petur joins in the singing :D
13:11:59mirakPaul_The_Nerd: hem you never dev ?
13:12:09Membrillowe can ban and kick people
13:12:31Membrilloalthough im not mean
13:12:44Paul_The_Nerdmirak: It's more, which do I need to add to sources?
13:12:53mirakPaul_The_Nerd: you must put the folder into apps/plugins/
13:13:12mirakadd a viewever intry and change the make file of the plugins
13:13:30Paul_The_NerdAh, leave it in its own folder.
13:13:33Paul_The_NerdOkay, I think I got it.
13:13:49 Join Pi___ [0] (
13:14:29Membrillodamnit back down to 87mb left
13:14:34mirakyou must tell to plugin make file to add the folder mpeg2
13:14:54Membrillodoes anyone have a spare H340 back plate i can buy off them?
13:15:00mirakMembrillo: like rockboy folder is added
13:15:13mirakMembrillo: but the plugin displays nothing ;)
13:15:17*ST np: The Alchemist - 1st Infantry - 19 - Different Worlds [04:10m/192Kbps/44KHz]
13:15:43Paul_The_Nerdmirak: I got it.
13:15:52Mode"#rockbox +o B4gder " by ChanServ (ChanServ@services.)
13:16:02markunHere we go :)
13:16:04Paul_The_NerdAttempting to build it now
13:16:14B4gderST: shut off the script or I'll do it for you
13:16:29BgerB4gder binlang has a bug :)
13:16:29Membrillohe might not be watching the channel
13:17:11preglowjust kickban him, he's likely not watching
13:17:18*ST np: Cuban Link - Flowers For The Dead (Tribute To Big Pun) [04:42m/192Kbps/44KHz]
13:17:23preglowyes, there you go
13:17:26preglowST: and out you go
13:17:28Bgerif there is a space at the end of the line like "LANG_BLABLA " it says
13:17:30BgerFound no LANG_BRIGHTNESS in english file!
13:17:31BgerMissing ID in ../apps/lang/korean.lang: LANG_BRIGHTNESS
13:17:36Mode"#rockbox +b *!*i=STx@* " by B4gder (
13:17:36Kick(#Rockbox ST :B4gder) by B4gder!
13:17:43Membrillobwa ha ha ha haaaaa
13:17:46Bgerbad B4gder
13:17:52preglowgood badger
13:17:56Membrilloalmighty powerful B4gder... we worship you
13:18:10B4gderkneel when you say that!
13:18:14markunLogbot is the one to worship here
13:18:16Paul_The_Nerdmirak: Gosh, it sure has a lot of warnings. Is that normal?
13:18:35mirakPaul_The_Nerd: yes
13:18:37BgerB4gder did you see the problem ?
13:18:41stamppot1975almighty powerful B4gder, we sacrifice virgin untouched sliderules in thy name...
13:18:48mirakPaul_The_Nerd: I didn't tried to fix them
13:18:59Paul_The_NerdI'll copy it over and try your test stream in a few minutes then
13:19:22B4gderBger: I don't know what problem you're referring to
13:19:33mirakPaul_The_Nerd: this comes from the compile options probably because the same code doesn't complain with a basic make on x86
13:19:40mirakor the compiler
13:20:17BgerB4gder if in a translation file there is a space at the end of the "id: LANG_ID" line (like "id: LANG_ID ") the binlang complains
13:20:30B4gderfeel free to fix
13:20:37Bgerperl ...
13:21:12Bgerprobably the best i can do is to make it behave worse :D
13:24:10B4gdertry this:
13:24:14B4gder- if($_ =~ /^ *([a-z]+): *(.*)/) {
13:24:14B4gder+ if($_ =~ /^ *([a-z]+): *(.*)\s*\z/) {
13:24:23 Join gtkspert [0] (
13:24:31B4gderperhaps not
13:24:38Paul_The_Nerdmirak: Don't think it works.
13:24:50*B4gder forgot that .* is "greedy"
13:25:26*B4gder goes to fetch a cup of that black stuff instead
13:27:41mirakPaul_The_Nerd: what's not working ?
13:28:00Paul_The_NerdOne second, I think I mistyped my change to viewers.config.
13:29:29Bgeri really like this kind of commits ....
13:30:26Paul_The_NerdFirst, mpeg2_dec.rock is in the rocks folder, is that okay?
13:30:26Paul_The_NerdSecond: The new part of viewers.config should just be this, right? mpeg2,rocks/mpeg2_dec, 00 00 00 00 00 00
13:31:24stamppot1975As an aside: I've found that the RTC is not exactly "Realtime"... As in: I can set it all I want, but it doesn't keep track. Shutdown the player and start it up the next day −−> time is waaaay off.
13:31:31stamppot1975this is on a H320 by the way.
13:31:51LinusNstamppot1975: how far off?
13:32:17stamppot1975Well, my local time is 13:32, and the clock is at 22:17...
13:32:44peturI've noticed it too but haven't bother to investigate
13:32:56preglowipod rtc is just fine :>
13:32:56stamppot1975Oh, and my date seems to have gone back to Jan. 07 2040...
13:32:59Bgerstamppot1975 heh, it changes realtime :P
13:33:08preglowstamppot1975: you've booted iriver firmware...
13:33:13Jungti1234hey stamppot1975
13:33:22stamppot1975Now that you say so...
13:33:22Jungti1234Rockbox likes 2040.
13:33:23Mode"#rockbox -b *!*i=STx@* " by B4gder (
13:33:27Mode"#rockbox -b *!*n=hotwire@* " by B4gder (
13:33:34preglowBagder: awww :/
13:33:36stamppot1975Indeed I have...
13:33:36Mode"#rockbox -b *!*@81.222.176.* " by B4gder (
13:33:54preglowstamppot1975: iriver firmware has buggy date handling currently, perhaps that messes up the time as well
13:34:03Bgerwow B4gder almighty
13:34:13stamppot1975Can be... time to flash bootloader v5 then...
13:34:22preglowb4gder: btw, who are opers in here? you, linus and zagor?
13:34:28stamppot1975No need to end up in iRiver anymore then...
13:34:30LinusNit is also quite likely that the h300 rtc drifts
13:34:52LinusNsince it uses the internal free-running oscillator instead of an accurate external oscillator
13:35:00stamppot1975I am correct in assuming that bootloader V5 will boot into Rockbox when returning from USB mode?
13:35:07preglowinternals oscs can be pretty accurate these days
13:35:10LinusNstamppot1975: yes
13:35:17Bgermine is nearly accurate
13:35:28stamppot1975LinusN: agreed, but drifting 34 years into the future in just 2 days?
13:35:44LinusNstamppot1975: ;-)
13:36:14stamppot1975Would make a nice time-travel device... Another unexpected benefit of flashing to Rockbox... :)
13:36:17Bgeru've probably traveled with vel near to the speed of light :P
13:36:23preglowthey tried tuning it to 32.768khz, but ended up on 10mhz
13:36:29preglowthey figured no one would notice
13:36:59 Join ST [0] (
13:37:20preglowST: so, have we turned off our "omgz lol see what i'm playing" script feature now?
13:38:20Paul_The_Nerdmirak: The viewer doesn't seem to load the file, and if I long-click on the file to choose a viewer manually it's not in the list.
13:38:40SThehe sorry about that, i use it for msn, and didnt realise it was displaying here as well
13:39:25preglowturns out it was
13:39:36 Join mikearthur [0] (
13:39:41peturyuck... msn...
13:40:01mirakPaul_The_Nerd: you added an entry to the viewer file ?
13:40:13LinusNST: just using msn should get you banned ;-)
13:41:10mirakPaul_The_Nerd ok I didn't saw your previous answer
13:41:14*petur hands over his copy of psi and whispers 'jabber'
13:41:19Paul_The_Nerdmirak: Yes. mpeg2,rocks/mpeg2_dec, 00 00 00 00 00 00
13:41:31Paul_The_NerdI also tried moving it to the viewers folder.
13:41:36Paul_The_NerdAnd making the appropriate change.
13:41:44mirakyou use the sim ?
13:41:51Paul_The_NerdNo, why?
13:41:55 Quit darkless (Read error: 104 (Connection reset by peer))
13:42:12 Join darkless [0] (
13:42:14mirakPaul_The_Nerd: you do make && make zip and then unzip it as usual
13:42:21mirakdon't know what's wrong
13:42:42Paul_The_NerdWhat should be different?
13:43:02mirakdon't know
13:43:09mirakif you got the rock file it should be ok
13:43:27mirakyou have an H320 right ?
13:43:33mirakdid you updated your cvs ?
13:43:47linuxstbPaul_The_Nerd: You could try a make clean. I had a similar problem a few days ago trying to add a new viewer, but I forget how I fixed it...
13:43:54Paul_The_Nerdmirak: ipod
13:44:04Paul_The_NerdI did get the rock file.
13:44:20Paul_The_NerdAnd linuxstb: Any time I try something interesting, I make clean first. ;-)
13:44:34Bgergood strategy
13:45:35 Quit akaidiot (Connection timed out)
13:45:51linuxstbPaul_The_Nerd: Nice work finding the Nano bootloader problem by the way.
13:46:18Paul_The_Nerdlinuxstb: All I really did was help narrow down what it's not to "anything in the actual code"
13:46:49linuxstbBut it's another "show-stopper" bug fixed for the ipod port.
13:46:52Paul_The_NerdShould my "open with" include two instances of "mp3_encoder"?
13:47:51STi dont understand what this patch does:
13:48:41Paul_The_NerdST: If you click the thumbstick (navi) button on songs, it inserts them into the current playlist, rather than creating a new one with them as the first song, I believe
13:49:55STok nice
13:50:11preglowPaul_The_Nerd: no
13:50:22linuxstbIf I understand correctly, the current behaviour is for both NAVI and RIGHT to do the same thing - i.e. create a new playlist and start playing that track. That patch makes NAVI insert the track into the currently playing playlist. Which seems sensible to me.
13:50:23preglowbut then again, mp3_encoder is weird
13:51:19Paul_The_Nerdmirak: Okay, I got it to show up in the viewers this time
13:51:32Paul_The_NerdWhat do you need to know?
13:52:38 Quit ST ()
13:54:29ashridahpreglow/linuxstb: what's the vague status of the ipod port atm? kinda updating the TODO list
13:55:28Bgershould anything stop me initializing a plugin's global var with its declaration ?
13:56:08Bgeri mean struct { blabla,... } ala[] = { 0, 5, 42.. };
13:56:51Bgergcc doesn't call memset for this, yep ?
13:57:15 Join mirak_ [0] (
13:57:29Paul_The_Nerdpreglow: Actually, several of the viewers show up more than once in the Open With list.
14:00:37mirak_Paul_The_Nerd: probably you can try first the H300 simulator target
14:00:49mirak_to check if that work
14:00:56Paul_The_Nerdmirak: I said it seems to be working.
14:01:05Paul_The_NerdAssuming I'm supposed to get several lines of text.
14:01:07 Join DJDD__ [0] (
14:01:24mirak_on ipod this work ? or the sim ?
14:01:28mirak_I got disconnected
14:01:39Paul_The_NerdI get about a screen full of text.
14:01:44Paul_The_Nerdipod nano specifically.
14:02:46Paul_The_Nerdstarting with: 0, 2: 4194304, a B7B04
14:02:46Paul_The_Nerdending with: 5922593, s: 1859, a 65DA25
14:03:00mirak_ok that's the fake mallocs in action
14:03:17mirak_you might need to change the line that can be ouput to
14:03:43Paul_The_NerdWhere is that located?
14:04:03mirak_in mpeg2_rb_api.c
14:04:20mirak_in function write_text
14:04:25mirak_if(line==22) line=0;
14:04:33mirak_the nano screen is tiniy
14:05:05Paul_The_NerdShouldn't I just take the write_texts out of the malloc?
14:05:06Jungti1234Korea and UAE play soccer today.
14:05:18linuxstbashridah: We basically just need an audio driver for it to be usable. This is relatively easy for the Nano and Color (ipodlinux has audio for those devices), but not easy for the video (no code in ipl to steal).
14:05:19 Quit Jungti1234 ()
14:05:21Paul_The_NerdOr do you want those?
14:05:28mirak_Paul_The_Nerd: if they annoy you you can remove them
14:05:38mirak_but the text display is circular
14:05:43mirak_there is no scrolling
14:06:27Paul_The_NerdShould it do anything other than displaying the text?
14:06:52mirak_you can save to files
14:06:56 Quit mirak (Read error: 110 (Connection timed out))
14:07:05mirak_the screen should show the frame number beeing decoded
14:07:57Paul_The_NerdLet me compile it without the outputs from within the fake malloc then
14:09:53linuxstbPaul_The_Nerd: There is a bug in the ipod ATA driver which may affect the mpeg decoder. A quick fix is to change line 383 of firmware/drivers/ata.c to "#if defined (SWAP_WORDS) || defined(ROCKBOX_LITTLE_ENDIAN)".
14:10:11linuxstbI'll commit a better fix tonight.
14:10:54preglowashridah: vague status is very unfinished
14:11:30Paul_The_NerdThanks linuxstb
14:12:11mirak_how much ram have the ipod ?
14:12:28linuxstb96kb for the Color/Photo and 128kb for the Nano and Video
14:12:39mirak_ram ???
14:12:44mirak_not iram I mean
14:12:54linuxstbSorry, 32MB, 64MB for the 60GB Video.
14:13:40mirak_why would the nano need that much ram if it have a flashram ?
14:14:19mirak_it doesn't need that much buffer
14:14:26linuxstbFor a start, it means apple can run exactly the same firmware.
14:14:43stamppot1975Slasheri: did you get any further with your tagcache proposal? I seem to recall that you wanted to work on it during the weekend...
14:15:01linuxstbmiurak_: The flashram is accessed in the same way as a hard drive via an ATA interface.
14:15:10 Join perplexity [0] (i=heh1807@
14:15:11Paul_The_NerdYup, the Nano even has the same bugs in its firmware as the Minis, apparently.
14:15:20 Quit Pi___ ("bye")
14:15:23Slasheristamppot1975: sorry but i had almost no time at the weekend and very busy still.. maybe in next week
14:15:48stamppot1975Slasheri: Don't sweat it, just curious...
14:17:23*linuxstb spots the identical ipod bug in the ATA write code as well as the read.
14:17:34Paul_The_Nerdmirak: Other than the write_text call in static_malloc, it gives no output at all.
14:17:48Paul_The_Nerdmirak_ even...
14:18:51Paul_The_Nerdlinuxstb: Is the Nano supposed to still do endless rebooting when USB is plugged in, by the way? (until I hit Menu at the right time and go to retail?)
14:19:05preglowSlasheri: any idea why dircache hangs ipods?
14:19:24Slasheripreglow: Hmm, can you have the logf output?
14:19:44B4gderpreglow: the unaligned ata problem?
14:19:58preglowB4gder: ahh...
14:20:14mirak_linuxstb: does it draws a lot of power ?
14:20:32mirak_Paul_The_Nerd: probably it's failing before
14:20:56dwihnoIf you have a cable with two atx power connectors, does it matter which end you connect to the mobo or psu?
14:21:06mirak_mirak_: if you check the function main function in mpeg2dec.c there is the main loop
14:23:54Paul_The_Nerdmirak_: I reenabled some of the write_text calls in there, I'll see if I can tell you how far it gets.
14:25:22 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:25:59linuxstbpreglow, B4gder: Sadly, fixing the ata bug didn't fix dircache for me. But I've just noticed that copy_write_sectors also has that bug - I only tested it with the bug in copy_read_sectors fixed.
14:26:40linuxstbPaul_The_Nerd: Yes, the USB "reboot-to-diskmode" feature doesn't work on the Nano or the Video. We need to think of a solution to that.
14:27:03linuxstbOn my Color ipod, it reboots and enters the "black and white diskmode".
14:27:10preglowlinuxstb: how do you reboot do disk mode?
14:27:20preglowjump to an address?
14:28:11linuxstbNo. The code writes some strings to a certain address and then reboots (something like DEV_RS|=4).
14:28:57Paul_The_Nerdlinuxstb: Would it be possible to just tell the bootloader to boot into RetailOS?
14:29:00preglowman, that usb thing is going to take work
14:29:26Paul_The_NerdSince "Disk Mode" is significantly slower on the Nano anyway, so wouldn't be ideal even for a short-term fix.
14:29:29preglowat least its a good way to learn ida pro
14:30:42mirak_Paul_The_Nerd: you tryed the sim for nano ?
14:31:07Paul_The_Nerdmirak_: No, I have not tried the sim yet.
14:31:09linuxstbPaul_The_Nerd: That's a good point. So you would prefer to go to retailos?
14:31:23Paul_The_NerdVery definitely.
14:32:34linuxstbI suppose we could add usb detection to the bootloader, and if USB is detected, then always go to retailos (for all targets).
14:32:43B4gderthe sims for color targets are now SDL in the build table builds
14:33:26B4gderon next commit we'll see how that works!
14:33:54linuxstbpreglow: ipodloader2 has a "diskmode" option. I'm pretty sure it uses the same code I've used in Rockbox, but it could be worth a quick check, and a chat with slowcoder.
14:34:28linuxstbBut we've just decided we don't care about diskmode on the nano... Ignore me.
14:35:38Paul_The_Nerdmirak_: In sim, it seems to STATE_INVALID on every frame
14:35:57Paul_The_NerdOn the iPod it does STATE_BUFFER and STATE_SEQUENCE and then stops there before ever displaying a frame number.
14:36:08 Quit ashridah ("sleep")
14:37:07mirak_Paul_The_Nerd: check the variable size
14:37:24mirak_when it reads the file to the buffer
14:37:44mirak_the sim doesn't seem to work at all for ipod
14:38:58mirak_Paul_The_Nerd: you mean it crashes ?
14:40:23Paul_The_NerdNo, it doesn't crash.
14:40:38Paul_The_NerdI'm not sure whether it's frozen, or just not progressing.
14:42:09 Quit kernelsensei (Read error: 104 (Connection reset by peer))
14:42:11 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
14:42:38 Join Kohlrabi [0] (
14:45:13Paul_The_NerdYou wanted the value of "size"?
14:45:24 Nick Rob2222_ is now known as Rob2222 (
14:46:24Paul_The_NerdMirak_: size==2146304
14:47:51mirak_Paul_The_Nerd: if your buffer is bigger than the file size you should have the file size
14:47:55mirak_seems it's correct number
14:49:30mirak_Paul_The_Nerd: maybe you want to remove all the IRAM stuff I used ICONST_ATTR ICODE_ATTR IDATA_ATTR
14:50:02mirak_this provide just a little boost anyway, almost unoticeable
14:50:51 Part petur
14:52:57 Join Moos [0] (
14:57:43***Saving seen data "./dancer.seen"
14:57:49Paul_The_Nerdmirak_: What should the state be on the third iteration of the do...while in sample4?
14:58:02LinusNi don't really like that playlist.c contains gui code
14:58:32B4gderI agree
14:58:37mirak_Paul_The_Nerd: STATE_INVALID_END probably
14:58:51Paul_The_NerdOkay, it's not doing that.
14:59:52Paul_The_NerdIt does STATE_BUFFER, then STATE_SEQUENCE, then three loops where it doesn't go into any of the named states.
14:59:54mirak_at least it should enter in this case
15:00:18mirak_becase STATE_SLICE goes to STATE_INVALID_END also
15:00:33Paul_The_NerdYes, it never goes to STATE_INVALID_END
15:01:08 Quit stamppot1975 ("CGI:IRC")
15:02:59Paul_The_NerdIt does Buffer, Sequence, Default, Default, then on the 5th iteration it starts the do loop, but does not reach the switch.
15:06:08preglowPaul_The_Nerd: did you need to do anything clever to make gcc4 compile, or just use ordinary gcc3 procedure?
15:06:25Paul_The_Nerdpreglow: Nothing special at all.
15:06:29linuxstbI've just committed the fix to ata.c for the ipod. I've just done a simple fix, but may look at changing the SWAP_WORDS define sometime.
15:07:20B4gderyes, SWAP_WORDS is not very a descriptive name
15:08:52 Quit Membrillo ()
15:08:52mirak_Paul_The_Nerd: I have a seg fault in the sim now, I don't even know why
15:09:55Paul_The_Nerdmirak_: Strange, it seems to work in my sim.
15:15:11amiconnB4gder: Blame LinusN ;)
15:15:36*B4gder blames LinusN
15:15:58*B4gder slaps the "blamed" sticker on LinusN's forehead
15:15:59LinusNso kill me ;-)
15:19:50mirak_Paul_The_Nerd: works again from I used a bad macro
15:22:33Paul_The_NerdWhat do I need to use this bootloader to load iPodlinux... my curiosity is getting to me. Or is that just there for planned expansion?
15:24:19XavierGrCan disk_spindown set to 0?
15:24:47XavierGrbecause the apps/setting.h says so but settings won't let me do it.
15:24:51markunPaul_The_Nerd: Are the filesizes bigger for gcc4 compared to gcc3 for ARM?
15:25:50 Join NicoFR [0] (
15:27:43Paul_The_Nerdmarkun: The bootloader was slightly smaller, and I *think* so was the firmware itself, but I'm not 100% certain on the latter.
15:27:47 Join akaidiot [0] (
15:30:35preglowi'm going to investigate a bit myself
15:33:21lostlogici mean ... good morning.
15:33:39XavierGrgood morning
15:33:50 Join Febs [0] (
15:34:22XavierGrlostlogic I think you have hallucinations (or dreams) with coding languages. What a way to start your day?
15:34:50Paul_The_Nerdpreglow: can the bootloader be made to load ipodlinux?
15:35:00preglowPaul_The_Nerd: ask linuxstb
15:35:49Paul_The_Nerdlinuxstb: What would I need to try out ipodlinux with this bootloader?
15:36:16lostlogicXavierGr: It's a mutt finger tick −− I login to my computer, move the mouse to my screen terminal, and hit c<ENTER> to change to my inbox... except when I left screen here instead of mutt.
15:36:54mirak_what the guy of ipodlinux are thinking about rockbox prt to ipod ?
15:36:55 Quit uwe_ (Read error: 110 (Connection timed out))
15:37:21preglownot much
15:37:24XavierGrlostlogic: Now you are obsessed with programming! ;p hehe
15:37:26preglowthey're nice guys
15:38:04linuxstbPaul_The_Nerd: You need to repartiton your Nano's drive to create an ext3 partition for linux to use, format it, copy the userland files there, copy the kernel image to the root of the FAT32 partition (call it linux.bin) and then hold PLAY whilst booting.
15:38:23linuxstbpreglow: Did you read in #ipodlinux about the Nano's "suspend to disk" mode?
15:38:34preglowlinuxstb: no, i mostly just follow this channel while working
15:38:49preglowlinuxstb: btw, didn't you shrink your bootpart?
15:38:59mirak_can the tagdb be created from rockbox or it's not possible ?
15:39:13linuxstbThe deep sleep in the Nano and Video is actually a suspend-to-disk mode, where the RAM image is written to the firmware partition.
15:39:36linuxstbWhich is why the bootloader isn't run when the Nano resumes from sleep.
15:39:37BgerLinusN is there any reason for having lines 394 and 400 in firmware/usb.c for iriver h300 ?
15:39:42preglowlinuxstb: yes, i kind of figured that
15:39:58Paul_The_NerdOn a related note, know a tool for win32 that'd be good for such a repartitioning by random chance?
15:40:04preglowlinuxstb: so firmware checks for a file on disk and loads from that if it can?
15:40:20linuxstbSeems that way.
15:40:32linuxstbSo on a Nano (and video), you shouldn't change the size of the boot partition.
15:40:40preglowyes, 'cause either it definitely runs, or they just paste the apple logo onto the screen to make it look like it
15:40:44linuxstbGNU parted could probably be used to shrink the FAT32 partition.
15:40:46LinusNBger: no, the second one is redundant
15:41:05preglowso how does deep sleep work on your 4g?
15:41:07Paul_The_Nerdlinuxstb: It's not too important that it's nondestructive. The ipod's empty.
15:41:18BgerLinusN want me to remove it ?
15:41:26mirak_does the sim work for ipod nano ?
15:41:35linuxstbPaul_The_Nerd: Then just backup the files and repartition using fdisk
15:41:42LinusNBger: go ahead
15:42:06linuxstbPaul_The_Nerd: You should only need to give about 32MB or so to the Linux partition I think.
15:42:18mirak_Paul_The_Nerd: probably you should download a livecd of linux and do the partitioning from there
15:42:52linuxstbpreglow: The retailos deep sleep is the same as the power-off code in Rockbox.
15:42:58linuxstb(on my ipod)
15:43:17Paul_The_Nerdmirak_: Okay. I've got a knoppix disc around.
15:43:20linuxstbi.e. you press MENU or SELECT, and it runs the bootloader
15:44:18preglowwell ok
15:44:24preglowi can confirm that bootloader works with gcc4
15:44:33mirak_I have buld the sim for a nano, but rockbox load then shows nothing in the root drive
15:45:02linuxstbYou probably need to select "show all files"
15:45:09preglowlinuxstb: and yeah, btw, perhaps we want to disable the lcd before we poweroff in rockbox? the lcd image stays for a good and dissipates over time now
15:45:23mirak_linuxstb: I can't acces the menu. Wha key is it ?
15:45:29mirak_I have tried all the keys :)
15:45:32linuxstbI think it's INSERT.
15:45:38linuxstb(a long press)
15:46:15mirak_ok thanks !
15:46:42linuxstbpreglow: I'm not sure we know how to disable the lcd - maybe just fill it with a black background.
15:48:24 Join tvelocity [0] (
15:49:53 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
15:50:04preglowjust clearing it might do
15:52:50 Part LinusN
15:54:17 Join actionshrimp [0] (
15:56:22 Quit thegeek (Read error: 104 (Connection reset by peer))
15:57:58 Join thegeek [0] (
15:58:45 Quit Bger ("BitchX: the ONLY bug-free client")
16:03:34 Quit Febs (Read error: 104 (Connection reset by peer))
16:05:28BHSPitLappyhey linuxstb, hey preglow
16:05:43BHSPitLappymornin' to you both
16:07:12linuxstbGood afternoon.
16:08:56 Join Bger [0] (n=Bager@
16:10:58 Quit Bger (Client Quit)
16:18:51 Join akaidiota [0] (
16:20:59 Join uwe_ [0] (
16:23:09 Join Matze [0] (
16:30:42 Join webguest44 [0] (
16:31:25 Join curio [0] (
16:34:32 Quit kernelsensei (Remote closed the connection)
16:34:34 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
16:34:34 Quit akaidiot (Read error: 110 (Connection timed out))
16:34:55 Quit webguest44 (Client Quit)
16:35:59 Quit B4gder ("time to say moo")
16:40:23 Join akaidiot [0] (
16:42:30 Join Paul_The_Nerd [0] (
16:43:02Paul_The_NerdWell, in the end, that didn't work so well. Heh.
16:44:53linuxstbWhat's that? ipodlinux?
16:45:30Paul_The_NerdNah, I just screwed up and accidentally deleted the boot partition.
16:45:44Paul_The_NerdSmall typo of doom. Heh.
16:45:55Paul_The_NerdBut the restore utility was being... irritating.
16:46:12Paul_The_NerdGot everything back to "normal" now. I suppose I'll try the ipodlinux thing some other day.
16:46:23 Quit akaidiota (Read error: 110 (Connection timed out))
16:49:55linuxstbPaul_The_Nerd: Does everything seem fine with ipodpatcher.exe now? Do you ever have problems accessing your ipod?
16:51:43Paul_The_Nerdlinuxstb: Ipodpatcher has been great. No problems at all.
16:51:47 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
16:52:48linuxstbPaul_The_Nerd: Cool. I'm guessing you tested it a lot yesterday trying to get the bootloader working.
16:53:06Paul_The_NerdOh yeah
16:53:30darklessfeature request, wondering how much it would take to implement
16:53:33darklessadd some sort of album/track blacklist to the rockbox firmware, so it will skip certain tracks I deem bad while allowing me to keep the full albums on my iriver
16:54:32linuxstbTwo alternatives that would work now 1) Rename those files to (e.g.) .mp3.skip 2) Create a playlist for that album that excludes the bad tracks.
16:54:35 Join BHSPitLappy2 [0] (
16:54:43Paul_The_NerdI was going to suggest the playlist idea.
16:54:52darklessit's a pain to run a blacklister over a dynamically generated playlist
16:55:13darklessand renaming would also be a pain since people I might share the music with would need to rename it again
16:55:20Paul_The_NerdIs there any functionality for adding a playlist to a playlist (in such a way that it scans the playlist, reads the tracks, and then only adds entries for those tracks. Nothing dynamic, so you can't create a loop)
16:56:39darklessPaul_The_Nerd: I've been wondering about that too
16:57:17linuxstbdarkless: I have a feeling that Rockbox will ignore files that have the "hidden" or "system" attribute set on them.
16:57:44 Quit XavierGr ("Trillian (")
16:57:47***Saving seen data "./dancer.seen"
16:58:22linuxstbI don't think the idea of a blacklist would fit well into Rockbox's file browser based interface But it could be a feature that's implemented in the tag database part of Rockbox.
16:58:46darklessthe problem is that I wouldn't want to compromise my music files by fiddling with filesystem attributes and/or renaming
16:58:59amiconnlinuxstb, darkless: Rockbox ignores hidden files and unix-hidden files (.dotfiles) *unless* 'Show files' is set to 'All'
16:59:41darklessI wouldn't mind trying to implement it myself, I just need to know what would likely need to be modified (since my knowledge of the inner workings of rockbox is not good)
17:00:03Paul_The_NerdYeah, it'd almost have to be part of the tag database.
17:00:31Paul_The_NerdYou could implement a rating system, then have a threshold option "No songs under x of 10"
17:00:46linuxstbFor the file browser to ignore the files, it would need to keep all the filenames in memory - so we would have to set aside a fixed amount of memory for this purpose, which would be wasted for most users. It would also slow down a lot of the features.
17:01:03darklessthe ratings system would work for me
17:01:06linuxstbPaul_The_Nerd: Yes, that's the kind of thing I was thinking of. People have often requested a ratings system.
17:01:43darklessthat way I'd just rate stuff I dislike "zero"
17:01:57linuxstbBut the tag database is currently morphing into tagcache (we hope...) So I don't think it's worthwhile adding ratings to the current tag database.
17:02:16Paul_The_NerdWell, the current tag database and I don't get along anyway.
17:02:33darklesshow's the progress on tagcache?
17:02:45linuxstbNo idea. It's just one person's pet project.
17:03:42Slasheridarkless: hi, i haven't had time for it in the past two weeks. Maybe next week will be less busy
17:04:14linuxstbHave you thought about ratings? I guess that would have to be part of the runtime database.
17:05:09Slasherilinuxstb: yes, i have had some thoughts about that in mind. For example, rating the song based how long it was listened can be done (something similar amarok already does)
17:05:31Slasheriand of course saving the play counts in the db
17:07:26 Join webguest89 [0] (
17:08:24 Quit uwe_ (Read error: 110 (Connection timed out))
17:08:44webguest89Slasheri: Do you think it will be possible to read existing ratings from tags when building the tagcache?
17:09:38linuxstbI'm sure it's possible, but is it desirable? If someone gave you an mp3 file, would you want their rating?
17:10:26mirak_Paul_The_Nerd: any succes ?
17:10:43Slasheriwebguest89: i think it would be more practical to allow syncing the ratings with other pc software (amarok for example)
17:10:44webguest89Well I guess in my case, I've rated my songs outside of Rockbox and am looking for a way to get that rating in the WPS
17:11:03 Join uwe_ [0] (
17:11:04webguest89Slasheri: that would work too
17:11:11 Quit BHSPitLappy (Read error: 110 (Connection timed out))
17:11:20Paul_The_Nerdmirak_: I have not yet found where it's stopping. I'll have to stop now, but I'll look at it again this evening and see if I can figure out what it's doing.
17:11:51 Join ehntoo [0] (
17:13:56 Quit Paul_The_Nerd ("Leaving.")
17:18:30webguest89Does amarok right rating to tag or is it stored just within in the program?
17:19:37crwlwithin the program, i think
17:20:25webguest89I guess the trick too would be to support a program or two that us poor windows users have access to ;)
17:22:02webguest89Based on my very limited Rockbox programming knowledge, I assumed that using the tag would the easiest way
17:22:59webguest89Or at least the most universal way
17:23:10 Quit akaidiot (Read error: 110 (Connection timed out))
17:23:24mirak_you can install linux
17:23:34mirak_just buy another hard drive :)
17:23:44mirak_for 50$
17:24:03linuxstbwebguest89: Any idea if the "rating" tag in your files is standard, or if all PC software does its own thing?
17:24:21lostlogicamiconn: are you adding that licenses dir to docs and zip build?
17:27:06webguest89I believe it's standard, but only a few PC programs actually use it. Many seem to build their own non-tag based internal database of ratings.
17:28:56 Nick Lynx_awy is now known as Lynx_ (
17:32:47 Quit Zagor ("Client exiting")
17:33:49webguest89For vorbis comments the tag it assigns is 'RATING', in multiples of 10 from 0 to 100
17:34:57markunI've seen a rating or track_rating tag being used in foobar2000
17:36:40webguest89I tagged them with MediaMonkey, but foobar also recognizes them
17:37:49DreamTactix291that's because foobar shows all tags
17:38:07DreamTactix291it probably says RATING=n correct?
17:38:19 Join XavierGr [0] (
17:38:20webguest89I think a good rating system in Rockbox could address darkless' desire to skip lower quality tracks
17:38:53lostlogicHmm... is there a reason to only offer battery status updates to the UI every 60s even though they are read every 1/2s?
17:38:55webguest89DreamTactix29: I think so, but I don't have access to my regular compuer right now to check
17:39:13 Quit XavierGr (Read error: 104 (Connection reset by peer))
17:39:14DreamTactix291well that's my guess
17:39:19DreamTactix291because tags are simply text
17:39:36DreamTactix291if you look at a Vorbis file'
17:39:43DreamTactix291you'll see the tags like this
17:39:58DreamTactix291ARTIST=.... ALBUM=.... etc
17:39:58 Join XavierGr [0] (
17:40:29webguest89The question is then, do id3 tags have a standardized rating tag?
17:42:57DreamTactix291no idea
17:43:15DreamTactix291ID3 isn't a very well thought out system but i think there is a rating tag
17:49:20mirak_why isn't the rating done according to songs you play the most ?
17:49:45webguest89I think that's more along the lines of playcount, similar to the runtime db in rb
17:49:49linuxstbmirak_: That's not the same thing (IMO).
17:50:15mirak_why ?
17:50:50mirak_well anyway I would not use that, I prefer discover new songs instead of sticking to the same
17:51:00linuxstbHow ofter you have played the track on your DAP is not the same as how much you like the track. An obvious example is a brand new track you have just copied on.
17:51:41mirak_linuxstb: then you can weigth by the age of the file on your dap
17:52:02mirak_number of time you play divided by the time you have it on the drive
17:52:06linuxstbOr, if you normally play whole albums (like I do), the tracks will have the same playcount, but I will obviously have favourites.
17:52:25webguest89There would be value to being able to create a playlist with say only ratings 80-100 etc.
17:52:39mirak_linuxstb: yes if you skeep them before they are finished also
17:52:48linuxstbmirak_: How do you know how long a track has been on the player?
17:53:00mirak_linuxstb: hehe :)
17:53:07mirak_linuxstb: the date of creation on the fat
17:53:10mirak_for exemple
17:53:27mirak_of course if you remove it and put it back
17:53:38mirak_that will be a weird number
17:54:32mirak_personnally the rating would need to be wired to my brain to be in contact with my mood
17:54:56mirak_a song that I really like at one moment I will probably not want to hear it at another moment
17:55:08mirak_well I am sure of that in fact :D
17:55:16webguest89It is done then, we need an ESP rating system! ;)
17:55:26mirak_esp ??
17:55:39webguest89extra sensory perception
17:56:32webguest89It knows what you think and will adjust Rockbox accordingly :D
17:58:57mirak_I am bit disapointed about mpeg2 :/
17:59:26mirak_respect to iriver devs
17:59:42 Join needleboy [0] (
17:59:48needleboyhi everyone
17:59:56 Nick DrumRBoy320|Away is now known as DrumRBoy320 (
17:59:59needleboycan anyone help me with a tiny bit of code?
18:00:26mirak_anyway that it's xvid or mpeg2 the bootleneck is the bidimentional idct. That's to much
18:00:33markunmirak_: well, time to continue with xvid then :)
18:00:42needleboy+#define LCD_DEFAULT_FG LCD_WHITE
18:00:46needleboyi need to know how i change it so this:
18:00:49mirak_markun: mpeg2 looks way easier to optimise
18:01:11mirak_mirak_: the code is simpler
18:01:14needleboywill read the color from global_settings.fg instead of defining it
18:01:27markuntalking to yourself? :)
18:01:37mirak_markun: I really don't think we can do better than iriver though iaudio managed to reach 15fps
18:01:42mirak_markun: :)
18:02:59markunIf you get the the player in a working state (with output to the screen) I bet some people will optimize it (preglow, linuxstb :)
18:03:06mirak_markun: you where talking about the gigabeats often. In fact they are not distributed in France, but I knew what they where.
18:03:16mirak_I don't know why we can't find them here
18:03:31markunThey are only sold in japan, us and uk
18:03:51needleboyno one here to help me?
18:03:57mirak_markun: wonder why
18:04:21markunMaybe they don't want to setup a support line for other languages
18:04:52markunOr maybe Toshiba is not sure if they want to stay in the DAP bussines
18:04:55linuxstbneedleboy: That isn't a trivial change. LCD_DEFAULT_FG is not a variable, so you can't read the value from the settings file on the player.
18:05:15mirak_here some shops are lobying against ipods and apple
18:05:40 Join JoeBorn [0] (
18:05:50mirak_they have a conflict interst with there music store and apple store. You can't use their tunes on ipods directly
18:06:09markunI think the ipods are nice devices, but not for me (no itunes, only Ogg Vorbis)
18:07:37mirak_markun: they say that gigabeats menus are in french german spanish also
18:08:12linuxstbmirak_: I always copy files with the "-p" option - to preserver the creation date. Also, not all Rockbox players have an RTC.
18:08:15mirak_I use itune on an old mac
18:08:16 Join jborn_ [0] (
18:08:57markunmirak_: ah yes, you are right about the menu. Also Spanish.
18:09:20 Quit ehntoo ("Leaving")
18:09:36mirak_markun: I think we will not see ogg players before sometime now
18:10:34mirak_markun: hem you have an H300 or an ipod ?
18:10:45mirak_H100 maybe ?
18:11:00markunMaybe korean companies have ogg support in their players, but I think they are all very unknown.
18:11:52webguest89The lack of ogg support is really a shame...makes the need for Rockbox all the greater
18:12:18needleboylinuxstb: then how does the Color Changer patch read and writes it?
18:12:48linuxstbLook at the patch and find out :)
18:12:55needleboynothing i can see
18:12:57markunSome of the players:
18:13:23needleboyproblem is, i have no clue in C... or any other programming language for that matter :)
18:13:51needleboyi've been reading this code the whole day and still not close to a solution....
18:15:08linuxstbneedleboy: Why don't you just apply the existing patch?
18:16:19 Join DT291 [0] (
18:16:20 Quit DreamTactix291 (Read error: 104 (Connection reset by peer))
18:17:46needleboyi'll explain the problem, it's a bit long. do you use the H300 Optimized build?
18:17:56needleboyanyway, we have a Color Changer plugin
18:17:58 Quit YouCeyE ("Leaving")
18:18:00mirak_needleboy: what's optimised ?
18:18:09needleboyit changes the FG and BG color
18:18:29mirak_ok :-|
18:18:34linuxstbIMO, optimised is a bit of a misleading name. "Experimental" is probably more accurate - it includes lots of patches not committed to CVS.
18:18:41needleboymirak_: it's a version i compile with loads of patches and plugins made for the H300
18:18:58mirak_needleboy: ok I will test that probably
18:19:32needleboyanyway, you set the BG color, say, to blue, and the FG color to white
18:19:50needleboynow everything including the WPS displayed blue and white
18:19:56needleboythere's another patch
18:20:10needleboythat's meant to change the color of the 1bit BMPs in the WPS
18:20:25 Join webguest83 [0] (
18:20:39needleboyonly when using it, every time you get to the WPS, it changes the BG to blue and FG to white
18:20:41Moosmaybe now that lostlogic have CVS access he will commit some of those patchs
18:20:48needleboysorry, FG to BLACK
18:21:10needleboythat's what these two lines in the patch do
18:21:19needleboy+#define LCD_DEFAULT_FG LCD_WHITE
18:21:20needleboy+#define LCD_DEFAULT_BG LCD_BLUE
18:21:32needleboyin the LCD.H file
18:22:38needleboyi need to change the patch so that it doesn't change the color to what's defined in the code, but instead, reads it from the and global_settings.fg files made by the Color Changer\
18:23:16linuxstbThose two defines are the _default_ colours. What I think you want to do is this patch: which introduces the concept of "current colours".which are stored in the settings, and initialised to the default colours.
18:23:20needleboyif you want to take a look at the patches:"> ,
18:23:43needleboyinuxstb: checking it out, thanks
18:23:59needleboyso there's no way to do it from the patch itself?
18:24:28needleboylinuxstb: err, that's the Color Changer patch i use...
18:24:36linuxstbThen what's the problem?
18:24:37needleboythat's what i meant...
18:24:44needleboythe problem is in the other patch
18:24:54needleboythe one meant to change colors of BMPs only in the WPS
18:25:24 Quit JoeBorn (Read error: 110 (Connection timed out))
18:25:37needleboyit's right at the end, in the LCD.H part
18:26:13needleboy+#define LCD_DEFAULT_FG LCD_WHITE
18:26:14needleboywhat do i change this line to?
18:26:27needleboyso it reads from the ready made setting files?
18:28:05 Quit webguest89 ("CGI:IRC (EOF)")
18:29:04needleboytough, eh?
18:29:28needleboynah... only for idiot me :<
18:29:56linuxstbI don't understand why you need to change LCD_DEFAULT_FG and LCD_DEFAULT_BG. The first patch changes Rockbox so that it doesn't use those values, apart from setting the initial defaults when you run Rockbox for the first time.
18:30:40needleboyi tried removing these lines altogether...
18:30:49linuxstbOr at least, it should...
18:30:51needleboydidn't change anything, the colors were still changed
18:32:32 Nick DT291 is now known as DreamTactix291 (
18:32:48needleboyi removed the whole LCD.H section
18:33:25needleboyrecompiled, and changed the colors to blue and white. went to the WPS, they changed back to blue and black
18:34:23needleboyyou have to be right... it's illogical...
18:34:28needleboybut that's what happens
18:34:32DrumRBoy320in order to just test it out, could i wire say four AA batterys in succession and connect it in place of the battery just to see if an ipod works
18:34:57needleboyit works on an iriver...
18:35:09needleboybut the battery pack we get might contain some circuit...
18:35:16needleboyso i don't know
18:35:20DrumRBoy320well, idk, ill try it
18:35:34DrumRBoy320tell you'all how it goes
18:35:57needleboylinuxstb: any clue? sorry for bothering...
18:40:31linuxstbIt's probably because the WPS code isn't using settings.fg and settings bg - it's still using LCD_DEFAULT_FG and LCD_DEFAULT_BG. i.e. the first patch is incomplete.
18:41:41linuxstbSo the fix is to finish implementing what the first patch started to do - replacing all occurrances of LCD_DEFAULT_FG and LCD_DEFAULT_BG in Rockbox with global_settings.fg and
18:41:53DrumRBoy320i wish i had a multimiter
18:42:23DrumRBoy320is there any way to test steady 6 volts?
18:42:34needleboyjust replace LCD_DEFAULT_FG with global_settings.fg
18:42:51needleboyas is, no " around or whatever?
18:43:01needleboy:) sorry, idiot, as mentioned...
18:51:15needleboyok, i got something going i think. compiling.
18:51:20preglowlostlogic: welcome to the club, then
18:51:35lostlogicpreglow: thanks!:)
18:52:35 Join bobwise [0] (
18:52:43 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
18:52:43DrumRBoy320just took a bit ogf the covering off of the battery, and there is a circuit
18:53:02needleboyby the way, why have 3rd party plugins started returning "incorect model" again today?
18:53:08needleboyplugin loader changed again?
18:54:21needleboyDrumRBoy320: yeah, i figured there would be one...
18:54:41DrumRBoy320im going to disconnect the tabs
18:54:45 Quit jborn_ ("Leaving")
18:54:53DrumRBoy320and connect my battery to where the tabs were
18:55:03DrumRBoy320... i dont know wher epositive negative is though
18:55:06needleboyit's an iPod, so trash the hell out of it ;)
18:55:08DrumRBoy320it might be marjked
18:55:19DrumRBoy32060 gig photo 4th gen
18:55:25DrumRBoy320got it for 15 dollars
18:55:35DrumRBoy320missing some parts though
18:55:38needleboyeven more of a reason to trash it
18:55:40DrumRBoy320like, one internal connector
18:56:03lostlogicDo we need an audio event listeners queue for swcodec machines like hwcodec has? (Specifically would be helpful for doing AB-repeat better)
18:56:06DrumRBoy320its the connector the goes to the headphones / hold switch
18:56:50preglowlostlogic: i don't know what that entails
18:57:14lostlogicpreglow: basically similar to the tick event queue would be my expectation
18:57:49***Saving seen data "./dancer.seen"
18:57:49lostlogicpreglow: ab-repeat can be done just by adding lines to the elapsed callback and the endoftrack callback but that's not really elegant (but is fast)
18:58:11needleboyit's working! thanks linuxstb!!!
18:58:23linuxstbI've been thinking for a while that I would prefer the codec api to be more "push" oriented. i.e. moving the main decoding loop out of all the codecs, and using an api with a main function like "decode_frame()"
18:58:45preglowthen what about stuff like seek table loading?
18:59:11linuxstbThat will be part of an init_track() function.
18:59:24preglowok, so you more or less propose we extend the plugin header functionality for codecs?
18:59:58lostlogicyeah, the decoding loop duplicated across the 12309821 [codec].c files is probably less than great.
19:00:14linuxstbpreglow: I'm not sure what you mean.
19:00:53preglowlinuxstb: we'd need a function pointer table to have several callable functions in one codec plugin
19:01:24preglowright now we have one, which i've also got to admit i don't know how we get the address of
19:01:34linuxstbOr could plugin_start() return that table?
19:01:47 Nick Lynx_ is now known as Lynx_awy (
19:02:00preglowit could, but i don't see the point, might as well keep it in the plugin hedaer
19:02:13lostlogicthen we have 2 apis involved for codecs, one going each direction :-\
19:02:13preglowor do we currently use the same loader for plugins and codec plugins?
19:03:21linuxstblostlogic: If possible (it may not be), I would like the codec api to be in one direction only - i.e. you pass the codec a buffer with compressed data, and it returns some PCM data.
19:03:39linuxstbThis would mean separating the container format parsers from the codecs.
19:03:43 Quit bobwise ("CGI:IRC")
19:04:11linuxstbBut maybe we're getting off-topic from your original question...
19:04:17lostlogicahh, so change the direction completely over, does that gain us anything tangible though?
19:04:38lostlogichaha, yes −− which way it goes doesn't change the possible need for an event listener queue
19:04:40linuxstbIt means, for example, that plugins could play compressed audio.
19:04:48preglowlinuxstb: for those codecs when you can do that, libmusepack is completely dependent on fetching its own data
19:05:38linuxstbWhy is that? Are you saying it's not possible to just pass it a compressed frame of audio?
19:05:48preglowpossible yes, requires rewriting libmusepack, yes
19:07:05 Join Lear [0] (
19:07:30linuxstbSo how does libmusepack get data at the moment? Does it use the codec api internally to get data from the audio buffer?
19:07:46linuxstbi.e. using the read() equivalent.
19:08:33preglowit uses callbacks for all io operations, and i just use read_filebuf
19:08:40preglowfrom within the read
19:09:22linuxstbThat's fine then. decode_frame() would pass the codec a compressed frame, and you would have an internal read_filebuf that gives libmusepack data.
19:10:40 Join Paul_The_Nerd [0] (
19:12:05linuxstbMy idea to implement this would be to slowly change all the codecs, so that the main codec.c file is identical for every codec. Once that is done, we would be in a position to move the main decoding loop into playback.c. But it's obviously a lot of work.
19:13:02 Join SteL [0] (
19:13:10linuxstbI would imagine that this is also closer to how playback works with the MAS.
19:13:44preglowwell, this might work for all codecs, just be careful not to generalise too much so we make everything but the most common cases impossible
19:15:18preglowi don't see why playback would have to hand the codec a frame
19:15:28preglowwhy can't it just hand it the buffer and let it do whatever it wants?
19:15:36preglowit can keep track of what is done with it anyway
19:15:37 Join ehntoo [0] (
19:16:33Paul_The_NerdYeah, remember you also have codecs for things like midi/mod/etc as potentials as well. I imagine those might not fit into the same kind of generalization.
19:16:47preglowand stuff like hybrid wavpack
19:17:03linuxstbBecause I think that makes the codecs more complicated. I would prefer to keep them simple.
19:17:45preglowbut ok, exactly what do you say playback should provide the codec with?
19:18:32linuxstbA pointer to the next X (fixed size - e.g. 32kb, the same as the current GUARD_BUFFER_SIZE) bytes of the compressed data.
19:19:03linuxstbThe codec will then return the uncompressed PCM samples, and say how many bytes of the input data it has consumed.
19:19:10preglowwhat does the codec do if it can't sync?
19:19:14preglowbitstream errors, etc
19:19:20preglowahh, well
19:19:35preglowi just don't see the need to constrain the input buffer to any sise
19:19:39linuxstbJust return a "can not sync" error, along with how much data has been consumed looking for the start of a frame.
19:19:51linuxstbpreglow: We do - the audio buffer is a ringbuffer.
19:20:57linuxstbThis is similar to the ci->request_buffer() and ci->advance_buffer() system.
19:21:40linuxstbSo the first change would be to make all codecs use ci->request_buffer() instead of ci->read_filebuf().
19:21:47 Nick BHSPitLappy2 is now known as BHSPitLappy (
19:21:50preglowstill don't know how musepack would sit with this without a rewrite
19:21:56linuxstbIf that's possible, then I think my suggestion would be possible.
19:22:00preglowbut ok
19:22:13preglowany redesign/rethink of the codec api is going to include non-streaming codecs
19:22:31linuxstbI wouldn't be against having two apis.
19:22:58linuxstbi.e. streaming codecs along the lines I suggest, and non-streaming where we just allow the codec to use the memory themselves as they see fit.
19:23:15preglowonly tricky thing with non-streaming codecs is the loader thing
19:23:32preglowrest is just modifications to the buffering system
19:23:58 Join Matze41 [0] (
19:28:05linuxstblostlogic: Going back to your question, what is the audio_event_queue used for on the Archos devices?
19:31:22 Quit linuxstb ("CGI:IRC")
19:32:26preglowloading problem with non-streaming codecs are that: 1. codecs themselves need to parse their data and save it in the codec buffer in ready-to-use form, instead of just loading the file to the buffer as-is, since the files are usually not usuable in this form and 2. that this loading needs to be done while other codecs are playing :>
19:32:29 Quit needleboy ("CGI:IRC (EOF)")
19:33:12Matze41preglow: why does this loading need to be done when other codecs are playing?
19:33:39Matze41parsing a non-streaming file is mostly done in under one second.. so it can be done AFTER the last file of another filetype
19:33:53preglowMatze41: well, what about parsing loads of them?
19:34:05preglowMatze41: i don't want to parse just one, i want to parse as many as i can fit in the buffer
19:34:08preglowMatze41: which can be _a lot_
19:34:20Matze41hmm i dont understand why?
19:34:21preglowwe'll have to limit that due to limited number of metedata structs, of course, but still
19:34:25Paul_The_NerdMan, can you imagine buffering 25+mb of midis?
19:34:46Paul_The_NerdOr sids.
19:34:49preglowsamples will take quite some space
19:34:51preglowsids will be worse
19:34:52preglowbut still
19:34:54Paul_The_NerdAh, that's true.
19:34:56preglowwe'll only get ten, i think
19:35:06preglowwhich is the amount of metedata structs we have at the moment
19:35:16Matze41well they can just be buffered as before... and then when it should be played the non-streaming codec tells system "i need 500kbyte of space now for parsing" and rest will be just filebuffer as now...
19:35:18preglowthat can be changed in a jiffy, but still
19:35:29amiconnpreglow: We have 30 on swcodec
19:35:38preglowamiconn: cool
19:35:46Matze41but ONLY the file actually played needs to be parsed
19:35:51preglowMatze41: so ok, when i load my 18 megabyte xm, what should the loader do?
19:36:03Matze41ah you mean ONE file with 18meg?
19:36:03 Quit gtkspert ()
19:36:10preglowthat's a possible scenario
19:36:13Matze41well that is...
19:36:17preglowthen there's not enough memory and it has to give up
19:36:24preglowit's inefficient and i don't like it
19:36:24Matze41i dont have an xm file over 500kbyte
19:36:30preglowi have tons that are large
19:36:35preglowcheck out mellow-ds stuff
19:37:11amiconnMatze41: Some parsing steps must be done at load time, and load time != playback start time
19:37:41amiconnThinking of the various "Tracker-Packer" formats, like ProRunner etc pp
19:37:52Matze41amiconn: i dont know xm... but i think, every parsing that needs to be done to start playing (step from file-load to play) could be done in nearly no time
19:37:53Paul_The_NerdMatze41: mellow-d's, possessive, I think.
19:38:10amiconnMatze41: It's not about time, it's about memory
19:38:32Matze41ah i found mellow-d..
19:38:37amiconnIf another track is already loaded behind the current one, the current track can't be expanded
19:38:38*Paul_The_Nerd isn't sure.
19:38:44Matze41ah well
19:38:50Matze41now i understand
19:38:53preglowalso a good point
19:39:17amiconnWe always want to buffer as many tracks as possible. We don't want to waste battery power
19:39:27Matze41yep thats right
19:39:54Matze41so maybe a kind of "precodec" or even the codec itself must be able to tell the buffering system how much space is needed for playing a file...
19:40:06Matze41and that space has to be kept free
19:40:25Matze41but that doesnt solve the "18-meg xm file" problem
19:40:32Matze41because playing this would be nearly impossible
19:40:43amiconnpreglow: Btw, with my new loader system, it should be quite simple to split the codecs into loader and decoder w/o actually splitting the file
19:40:44Matze41with just 30megs of codec buffer
19:41:05amiconnWe could just extend the header to include 2 (or more) entry points
19:41:16preglowamiconn: yeah, but how to load tracks while other tracks are playing?
19:41:30 Join webguest28 [0] (
19:41:30preglowamiconn: and also, you'd need to load an entire codec just to load a file
19:41:38amiconnHmm, valid point..
19:41:40preglowif there's a bunch of different formats, you need to load several codecs
19:41:47preglownow, if plugins were relocatable...
19:41:52amiconn...and yes, some parsers would be common between codecs
19:42:09*manhattan is debating between the iAudio x5 30GB for $268+$5 s&h OR the 1GB iAudio U3 for $170+$5 s&h. He ponders.oO(The plus of U3 are the 100 bucks cheaper price tag, and tiny player the minus are the no LFAC support and small 1gig. The plus for the x5 30GB! & FLAC. The minus 100 dollars more! it's a bigger piece and it uses a dock!!. Oh decisions decisions!) :'(
19:42:19amiconnSo it seems if we want to go for modularity, we will have both loaders and codecs
19:42:53amiconn(e.g. ogg -> vorbis/flac, m4a -> alac/aac (iirc))
19:43:02preglowproblem is we probably want these loaders to stay in the core all the time
19:43:05amiconnogg -> vorbis/flac/speex ;)
19:43:14preglowfor which we yet again need a relocatable plugin format
19:43:38amiconnWe could try position independent code
19:43:47preglowshould work well
19:43:51preglowespecially for arm
19:43:57preglowwhich is more or less position independent as it is
19:44:34amiconnI want to try pic anyway, as I would like the number-voicing algorithm to be localised, which would mean to embed it in the voice file
19:44:56preglowyeah, good idea
19:44:59amiconnOf course that will make voice files platform dependent, but we'll get there anyway
19:45:05amiconn...with l10n v2
19:45:10preglowgerman counting is backwards, that's right
19:45:30amiconnNot always. 10s and 1s are swapped
19:45:55amiconne.g. for 21 (twenty-one) the german equivalent is one-and-twenty
19:46:03preglowyou can do that in norwegian as well
19:46:03 Join merbanan [0] (
19:46:17preglowprobably because of german influences, heh
19:46:24LearAnd then there's danish and french... :)
19:46:36amiconnFrench is weird regarding numbers
19:46:53amiconnI don't speak french, but I once read about french numbers a bit
19:47:04LearYep, but danish is even worse, afaik...
19:47:15preglowlets not think about that
19:50:16Moosamiconn: hehe twenty-and-one for your exemple in french :)
19:51:32 Join BHSPitLappy2 [0] (
19:51:38preglowand something incomprehensible using a completely different number base in danish
19:52:11mirak_Moos: yes
19:52:28MoosSalut mirak!
19:52:41 Join ender [0] (i=ychat@
19:53:12mirak_in french 90 is spellled eigty seventeen
19:53:37MoosI love our language ;)
19:53:47mirak_there is not one word to say 90
19:53:52mirak_in french 97 is spellled eigty seventeen
19:53:57mirak_not 90
19:54:22Learand eighty is four-twenty...
19:54:37mirak_Lear: yes, how could I miss that
19:54:43mirak_so used
19:54:50Moosyes indeed
19:55:14mirak_also in switerland they say nonante for 90
19:55:21mirak_or nine, I don't remember
19:55:33mirak_though they speak french but they cound differently
19:55:59mirak_in fact even 70 is 6*10
19:56:06mirak_sixty ten
19:56:17mirak_well n
19:56:21preglowcrazy people
19:56:23Moosnot just french spoken in Switzerland, +german+italian... maybe they mixed
19:56:30mirak_I was never good in mental calcul
19:56:49mirak_70 is 60+10
19:56:57mirak_sixty ten
19:57:06mirak_sixty + ten
19:57:21 Quit ender` (Nick collision from services.)
19:57:25 Nick ender is now known as ender` (i=ychat@
19:57:42Moosmirak: but I assume, that like when you was child you didn't learn the numbers that way
19:58:09mirak_when you think to it it's weird, but I don't think anybody always think 80 as quatre vingts (four twenty)
19:58:16DrumRBoy320... intesesting if we sould ever have GBA support
19:58:34mirak_Moos: I think yes we did
19:58:51Mooswe learned arbitrary
19:58:56mirak_Moos: that's why we are retarded
19:59:01Mooshehe :)
19:59:41mirak_I remember learning with cubes that we were stuffing together
20:00:51*Moos is remenbering his childhood language learning :)
20:01:17 Quit webguest83 ("CGI:IRC")
20:02:20 Join TCK [0] (
20:04:42 Quit tvelocity (Connection timed out)
20:07:44 Quit BHSPitMonkey (Read error: 110 (Connection timed out))
20:07:49 Join BHSPitMonkey_ [0] (
20:09:00 Quit BHSPitLappy (Read error: 110 (Connection timed out))
20:11:08 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
20:13:37 Join Paul_The_Nerd [0] (
20:21:23Rob2222DrumRBoy320: Have you killed you ipod, yet?
20:21:41 Part Paul_The_Nerd
20:23:03Rob2222If you wanna connect AA batterys to the lithium jack you should take 2 AA batterys, ane 1 AA NIMH/NICD rechargable battery
20:23:16Rob2222that should give you 4,2V
20:27:24DrumRBoy320didnt work
20:27:51DrumRBoy320i used 4 AAA's and it didnt fry anything
20:29:36Rob2222how do you know, that you didnt fry it?
20:30:02Rob22226V instead of 4,2V ... hmm
20:30:59Rob2222DrumRBoy320: it could be, that it didnt work cause the voltage was to high
20:32:31Rob22221 AA/AAA alkaline battery (1,5V) and 2 AA/AAA NIMH/NICD batterys (2x1,2V) = 3,9V should be definitely in the voltage range that must work
20:33:53 Join linuxstb [0] (
20:34:23amiconnI'd rather use 3 AA NiMH cells to replace one LiIon cell
20:34:42 Join Paul_The_Nerd [0] (
20:34:47Rob2222ok, should be fine, too
20:34:49amiconn4.2V is the *maximum* voltage of a LiIon cell, the nominal voltage is 3.6V
20:34:59Rob2222whats the shutoff value at LI-ION ?
20:34:59preglowwhy is the sim dir called archos? :PP
20:35:00amiconnSame with NiMH...
20:35:16Paul_The_NerdHrm... is the bootloader supposed to hang if the disk is an unrecognized file system, or should it boot into RetailOS?
20:35:43preglowno idea
20:35:59 Nick Rob2222 is now known as Rob2222_ (
20:36:00pregloware you sure the first partition is set up correctly?
20:36:03 Nick Rob2222_ is now known as Rob2222 (
20:36:05amiconnLiIon shutoff is somewhere below 3.0V, depending on the cell type, but never below 2.5V
20:36:07preglowthe boot part, that is
20:36:08Paul_The_NerdThe first partition was ext2 at the time
20:36:14Paul_The_NerdOh, the REALLY first one was fine
20:36:20Paul_The_NerdI could force it into retail
20:36:22Paul_The_NerdAnd that worked
20:36:28Rob2222so the iriver will go down to 3,0v? really?
20:36:43amiconnIt even goes below
20:36:49amiconn(with rockbox)
20:36:57Paul_The_NerdBut the bootloader wouldn't auto-retail, it just hung on "loading rockbox" even though the partition was... ext2 I think.
20:37:20curioPaul_The_Nerd hello
20:37:30curioPaul_The_Nerd i made the same mistake some days ago
20:37:35amiconnRockbox (and the bootloader) don't care about the partition layout, as long as it finds a *primary* FAT32 partition
20:37:44linuxstbpreglow: I'm about to remove the bootloader logos (both of them) and rename the bootloader to just say "Rockbox bootloader". Any objections?
20:38:00Paul_The_NerdYeah, I'm just saying, if the primary isn't fat32, the current action is to just sit there and continue trying to load rockbox.
20:38:18preglowsdl sim playback works fine here
20:38:19lostlogicAny problems with me creating a docs/licenses dir and putting the license for the profiling code in there?
20:38:24preglowlinuxstb: none
20:38:35curioPaul_The_Nerd, and if you try to boot the iriver firmware ?
20:38:48amiconnlostlogic: Hmm, I would prefer one file with multiple sections
20:39:01Paul_The_Nerdcurio: I'm playing with an iPod. It'll go into the iPod's OS just fine.
20:39:19Rob2222my iPOD building process (cygwin/gcc3.4.5) iss hanging at OBJCOPY sokoban.rock?! What does that should say?
20:39:20lostlogiclinuxstb: the audio_event_listeners on hardware codecs are the set of functions that are run when the hardware reports an audio event
20:39:21curioPaul_The_Nerd, oops i thought it was with H300 ;)
20:39:34lostlogicamiconn: LICENSES then?
20:40:09BagderI'm in favour of a single file too
20:40:19BagderLICENSES seems fine
20:40:22Paul_The_NerdOn an unrelated note, apparently all your settings are saved on the fat32 partition, rather than in the special partition, oddly enough.
20:40:36amiconnToday we have COPYING, which is the gpl
20:40:49amiconnI'm not sure whether LICENSES is clear enough
20:41:11lostlogicpreglow: want me to take off the profiling start and stop calls from the codecs so that the individual developer can just turn it on for the what he wants to profile and commit the stuff?
20:41:12Bagderyou mean we should add them to COPYING ?
20:41:56amiconnEither that, or put the gpl in LICENSES too, and remove COPYING
20:41:56Bagderlostlogic: can't you have them present or not with some ifdef/macro magic?
20:42:06lostlogicBagder: that's the current structure.
20:42:25lostlogicBagder: but it's kinda ugly to have it in all the codecs when few people will ever use it
20:42:29preglowlostlogic: i think i'd prefer that
20:42:34preglowLear, linuxstb: what do you say?
20:42:36 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
20:42:57preglowi don't want to clutter the codecs with code only a select few are going to need
20:43:13linuxstbI agree - as long as it's documented somewhere, I would prefer to keep the code clean.
20:43:29amiconnpreglow: Depends on how much clutter it actually is. Today we already do so with DEBUGF() and LOGF()
20:43:44Bagder... and how hard it is to add again
20:43:49lostlogicI'll add some of the content of my varios postings o n the patchtracker to a docs file on how to use the profiling regardless.
20:43:52lostlogicBagder: very easy
20:44:00lostlogicthey are just 'start' and 'stop' calls (two lines)
20:44:01preglowit's just a few lines, no changes needed
20:44:10lostlogicand they can really be added anywher eyou want to put them to decide what exactly to profile
20:44:38Rob2222could somebody give me a hint when a building nis freezing at OBJCOPY sokoban.rock?
20:45:06BagderRob2222: build with 'make V=1' to see the command lines
20:45:24Rob2222ah ok. thx, will try that
20:45:54lostlogicamiconn: the logf and debugf calls are helpful in documenting code aside from their usually inactive functionality.
20:49:05linuxstblostlogic: Did you get ice problems when trying to profile libfaad?
20:49:16lostlogiclinuxstb: yes, and alac iirc
20:50:24 Join LinusN [0] (
20:51:18LinusNlostlogic: are you certain that avgbat won't overflow with the new battery precision patch?
20:52:07amiconnLinusN: I had the same thought. A 32 bit int won't overflow
20:52:15LinusNok, good
20:52:17lostlogicdo we have any smaller than 32 bits?
20:52:23lostlogicyeah, it's only 1.6m max
20:52:34amiconnlostlogic: It's more
20:52:48linuxstbpreglow: Have you tried enabling the dircache since my ata fix?
20:52:54Bagderlostlogic: we did cleanup some of the int/long use in the past since the Gmini uses 16bit ints
20:53:06Bagderbut the gmini port faded away...
20:53:16lostlogicok, good to know.
20:53:19amiconnarchos recorder: ~6500 (scalefactor) * 1023 (adc_max) * 32 (number of averaged values)
20:53:22linuxstbAre gminis still sold by Archos?
20:53:35amiconniriver h300: ~2600 * 255 * 32
20:53:40amiconniriver h300: ~26000 * 255 * 32
20:53:55preglowlinuxstb: will now
20:54:59 Quit linuxstb ("CGI:IRC")
20:55:03 Join linuxstb [0] (
20:55:15lostlogichmm, yes, I was off by 2 powers of ten in tha tquick calculation I did now, because I forgot battery voltage is in centivolts, not volts.
20:55:19preglowwe might want to just disable dircache for nano, heh
20:55:37linuxstbpreglow: Good point.
20:56:21preglowhang, seems
20:56:43preglowbonafide hang, yes
20:56:46linuxstbMmm. I assume you get the same problem loading a theme?
20:57:00preglowwhat's in common?
20:57:07preglowhmm, how did i fix this again?
20:57:09preglowright, bump
20:57:18linuxstbMaybe lots of small disk accesses.
20:57:45linuxstbWe should probably implement a keypress on startup of Rockbox that resets the config block for ipod.
20:57:52***Saving seen data "./dancer.seen"
20:57:57preglowperhaps i should provide slasheri with a logf dump
20:58:01preglowif it even gets saved
20:59:05linuxstbWe could try adding some lcd_puts() commands to see whereabouts it hangs. I've never looked at the dircache code though.
20:59:15 Part curio ("Leaving")
21:00:27linuxstbSeems like we just need to #define SETTINGS_RESET in main.c to assign a key to the "reset settings" functions.
21:01:02linuxstbWhat should we use? The centre button?
21:01:14preglowi'd like a combo
21:01:37preglowright and center sounds good
21:01:40preglownot used for anything else
21:03:07 Join petur [0] (
21:03:40linuxstbOK. Shall I do it?
21:03:41 Quit webguest28 ("CGI:IRC (EOF)")
21:05:05 Join Kaggen [0] (
21:05:33lostlogicpreglow: ok for me to commit the profiling stuff with the changes we've discussed (1 commit for the return path changes in codecs and 1 for the profiling code)?
21:05:46linuxstbpreglow: What does the arm do if there is an unaligned memory access?
21:11:25preglowtoss an exception
21:11:35preglowi suppose i could add better support for displaying exceptions
21:11:53preglowlostlogic: sounds good
21:12:18 Join linuxstb_ [0] (
21:12:41linuxstb_preglow: Could that be the dircache/theme bug then? Any idea what happens on other targets?
21:12:51preglowcoldfire slows down, archos crashes
21:13:07linuxstb_Are there any themes for the Archos devices yet?
21:13:17*linuxstb_ checks
21:14:08preglowi just can't get ipod_fw to work now
21:14:09preglowfread failed: Success
21:15:12 Quit linuxstb ("CGI:IRC (Ping timeout)")
21:16:21 Join linuxstb__ [0] (
21:16:37amiconnpreglow: No UIE() for arm yet?
21:17:20linuxstb__amiconn: Did you ever try dircache on an Archos device?
21:17:28amiconnHmm, no
21:17:39linuxstb__So it's not been tested for unaligned memory accesses?
21:17:53amiconnWhat does it need? Just a #define in config-*.h ?
21:18:19preglowamiconn: no
21:18:36preglowamiconn: i think i'm going to try do an UIE handler now
21:18:48amiconnNm, found it
21:18:57amiconnIt's in config.h, based on the RAM size
21:20:12amiconnBuilding dircache-enabled recorder v1 target...
21:20:17linuxstb__It seems like there are no themes with bitmaps available for the archos devices. Maybe the bmp loader uses unaligned memory accesses as well.
21:20:55preglownow why the hell would ipod_fw/make_fw start acting up like this
21:21:42preglowwhat the hell
21:21:47preglowmy apple_os.bin is suddenly 0 bytes
21:22:34amiconnlinuxstb_: I doubt that. There is a bitmapped wps for archos - boxes
21:23:52linuxstb__amiconn: OK. It's not an endian problem - I've just loaded a theme in the ipod sim.
21:25:50 Join Paul_The_Nerd [0] (
21:26:26 Join akaidiot [0] (
21:30:25preglowboot sequence is nice and fast for nano
21:30:31preglownot much use for the splash screen anymore
21:30:55linuxstb__Are you running gcc4 now
21:31:18 Quit linuxstb_ (Read error: 110 (Connection timed out))
21:31:52*linuxstb__ visits
21:32:41Paul_The_NerdI find if I want to boot RetailOS, I have to hit Menu before the apple logo even displays
21:32:50preglowthat's what you need to do
21:32:54preglowno way in hell it works any other way
21:33:17Paul_The_NerdI would've expected it, I dunno... at some point when you're aware the unit's powered. ;-)
21:33:33linuxstb__At least it works now...
21:34:20amiconndircache adds ~5KB to the binary
21:35:29preglownow how to do the uie on ar
21:35:52amiconnEnabled cache, rebooted -> I09 CPUAdrErr
21:36:06preglowthen we know what's wrong
21:36:15preglowghah, i forget uie isn't supposed to return
21:36:17*amiconn slaps Slasheri
21:36:21 Quit akaidiot (Read error: 104 (Connection reset by peer))
21:36:42linuxstb__amiconn: boxes.wps needs 160x128 and boxes.rwps need 128x64. So I don't think it works on the Archos.
21:36:43*amiconn goes hunting for the exact spot in the .map
21:36:59 Join akaidiot [0] (
21:37:12SlasheriHmm.. that's interesting. I think it could be align problem then.. because the cache structure is not aligned
21:38:50 Quit akaidiot (Read error: 104 (Connection reset by peer))
21:39:09amiconnGah, how can that be?!? :confused:
21:39:25 Join akaidiot [0] (
21:39:43amiconnThe address where the CPUAdrErr was caught is - the CPUAdrErr exception routine (!!!)
21:39:55Slasherihehe :D
21:40:15 Quit akaidiot (Read error: 104 (Connection reset by peer))
21:40:20amiconnHow can that be...
21:40:21 Join akaidiot [0] (
21:42:03 Quit linuxstb__ ("Leaving")
21:42:10 Join akaidiota [0] (
21:43:32eEyehey now
21:45:35LinusNwe have a rec v1 charging patch in the tracker, and i see two options:
21:45:47LinusN1) commit it
21:46:03LinusN2) build a separate test build for users to try
21:46:13LinusNwhich ine do you guys prefer?
21:49:06 Join YouCeyE [0] (
21:49:36 Join linuxstb [0] (
21:49:43lostlogicwish there was a build like h300-optimized for that target.
21:52:16preglowLinusN: comes to stuff like battery, i think i'd go for 2, but then again, i don't have an archos
21:52:31preglowand send off a mailing list post with an url
21:52:57LinusNpreglow: i agree
21:56:20 Quit Matze41 ("Miranda IM! Smaller, Faster, Easier.")
21:57:04preglowarm doesn't have too many fatal exceptions
21:59:00amiconnOn the other targets, an UIE is always considered fatal
21:59:01preglowthe docs luckily doesn't say very much on what triggers an exception...
21:59:07preglowprefetch abort, for example
21:59:23preglowis that a memory error?
21:59:32preglow"undefined instruction" i get
21:59:58amiconne.g. if we get a timer interrupt, but no module implements an isr for that, it's considered fatal as well
22:00:12preglowthe rest are "prefetch abort" and "data abort"
22:00:18preglowthat's that
22:00:24 Quit akaidiota (Read error: 110 (Connection timed out))
22:00:29 Quit akaidiot (Read error: 110 (Connection timed out))
22:00:38preglowamiconn: yeah, sure, but i'm just handling the internal arm exceptions for now
22:00:53peturCVS problem for windows fanboys: you can't have a file and a directory share the same name. The issue is docs/licenses
22:01:11amiconnpreglow: Do they need special handling?
22:01:30lostlogicpetur: don't you have cvs up -dP it should be OK
22:01:42 Join solexx_ [0] (
22:01:45lostlogic(this is lostlogic screws things up his first day situation)
22:01:50amiconnOn both coldfire and SH1, exceptions and interrupts are handled the same
22:02:01preglowamiconn: there are seven exception types on arm, one of the is interrupt
22:02:01 Join webguest70 [0] (
22:02:26preglowi guess we can use the same handler for all of them, but we need some wrapper code arokund
22:02:32 Join gtkspert [0] (
22:02:44 Quit gtkspert (Client Quit)
22:02:49webguest70LinusN: I tried to disable the HDD LED on my H-120 the other day, but couldn't get it to work.. are you sure that GPIO 9 is the one?
22:02:52peturlostlogic: nope
22:03:06preglowmainly to calculate the exception address, arm requires different return code based on which exception type you've got
22:03:10 Join San [0] (n=test@
22:03:17LinusNwebguest70: pretty much, but i'll double check one of those days
22:03:34Sanhey linusn
22:03:36preglowlike irq return is subs pc, lr, #4, data abort needs subs pc, lr, #8
22:03:49lostlogicBagder: you run the CVS server? Can you physically delete that dir I added wrongly that is giving petur a problem? (docs/licenses)
22:03:52webguest70LinusN: thanks :)
22:04:11Sani have a question, for RockPod (hehe) could rockbox support drag and drop? Or does it have hardware limitations?
22:04:33preglowdo you mean the ipod port?
22:04:37*petur advises aginst dropping a DAP ;)
22:04:50Sanyeah preglow
22:04:51preglowwhat does rockbox have to do with drag and drop?
22:05:19Sanwell, i tought you could get rid of the MTP
22:05:41linuxstbThere is no MTP on the ipod.
22:05:59linuxstbitunes just copies the files normally to the fat32 filesystem.
22:06:29 Join webguest31 [0] (
22:06:33Sanbut, could it be UMS?
22:06:39linuxstbYes, the ipod is UMS.
22:06:54Sanlike, external disk?
22:06:56webguest70LinusN: It should be possible to use the LED freely (on/off at will) if you to keep the HDD spinning and then disable/enable the led?
22:07:11webguest31LinusN: is the shebbs playlist patch going to be incorporated ?
22:07:12preglowlinuxstb: you got some reset settings code coming?
22:07:14LinusNwebguest70: not sure
22:07:15Paul_The_NerdSan: If you go into itunes, right click your ipod, choose ipod options, you can enable disk mode
22:07:22LinusNwebguest31: i don't know yet
22:07:28linuxstbpreglow: No - it seems your button driver doesn't detect the button press
22:07:35Sani dont have one, I am asking for a friend
22:07:38Paul_The_NerdYou can't just drag music onto it, though
22:07:55linuxstbBecause you need to press it before Rockbox starts.
22:07:58Paul_The_NerdYou will be able to with Rockbox, but the retail firmware won't see files without them being in the database.
22:08:00preglowsmall wonder then
22:08:08preglowwe need a polling driver for that
22:08:28preglowi wonder if the interrupt handler can be called on its own
22:08:51Rob2222are you talking about the ipod BL?
22:09:22lostlogicLinusN: is it just me or are we setting IOREGC to 2.9V when the comment says 3.3V (in pcf50606.c)?
22:09:34Santhanks paul
22:09:37 Join DangerousDan [0] (
22:12:21 Part webguest31
22:13:45 Quit solexx (Read error: 110 (Connection timed out))
22:15:02preglowdoesn't arm pass first four parameters as regs?
22:15:21 Quit YouCeyE ("Leaving")
22:19:55LinusNlostlogic: yes, the comment is wrong
22:20:06lostlogicLinusN: better the comment wrong than the code wrong :)
22:21:17lostlogicLinusN: btw, the PCF50606 is giving me horrible nightmares with not really reporting it's true status on the MBCC registers (ie being charging when MBCC1 bit 0 is 0), so I'm not sure I can finish my proper charge detection and setting stuff until I can listen for interrupts from it.
22:22:48peturLinusN: did you put the checking of H3xx ISP1362 CS/RD/WR connections on your busy schedule somewhere?
22:23:07LinusNpetur: yes, at the very bottom ;-)
22:23:44lostlogicpetur: what's ISP1362?
22:24:14lostlogicahhh :)
22:25:03preglowok, let's see
22:25:22preglowno data abort exception :/
22:28:12 Quit Lear ("Chatzilla [Firefox 1.5/2005111116]")
22:29:17preglowdata abort exception!
22:29:38*petur hates exceptions
22:30:45preglowaccording to the map file, there's nothing there...
22:32:08lostlogicpreglow: map file only shows non-statics? might need to combine map with object like I did for profiling?
22:34:08preglowdircache_build ain't static
22:35:28preglowand anyway
22:35:41preglowmap file says audiobuf starts at 0x0009asomething
22:35:58preglowmy crash point is at 0x0009b364
22:36:54preglowgod, i'm a retard
22:37:11 Join matsl [0] (
22:38:22 Nick BHSPitMonkey_ is now known as BHSPitMonkey (
22:38:34 Nick BHSPitMonkey is now known as BHSPitMonkey_ (
22:38:58preglowthat's where it's on
22:39:58 Quit webguest70 ("CGI:IRC (Ping timeout)")
22:40:14 Quit NicoFR ()
22:42:24LinusNlostlogic: the pcf50606 irq is on gpio6 (it was already in the wiki)
22:42:36 Quit linuxstb ("CGI:IRC (Ping timeout)")
22:44:47preglowi think my UIE will have to take exception/interrupt number as a parameter
22:45:12DrumRBoy320i tried to hook up that lil battery pack i made, now i smell burning
22:45:43Rob2222did you used 3 batteries?
22:45:45DrumRBoy320yay, still turns on when you plug it in
22:45:58Rob2222why you do that
22:46:02Rob2222its to much
22:46:13DrumRBoy320most battery packs use 4 AAA
22:46:16Rob2222more then only me said that
22:46:38Rob2222well, get a screw between the 4th connectors
22:46:42Rob2222or a wire
22:46:48Rob2222or something
22:47:05lostlogicLinusN: woah, what wiki page :-\ cool.
22:47:07DrumRBoy320i made the thing myself
22:47:15DrumRBoy320with AAAs and foil and tape
22:47:30preglowthis really isn't pretty, long calls are used for bloody everything
22:47:37 Nick BHSPitLappy2 is now known as BHSPitLappy (
22:50:53amiconnpreglow: dirchache.c, line 100 (in dircache_gen_next):
22:51:02amiconnnext_entry = allocate_entry();
22:51:19preglowamiconn: based on your own findings?
22:51:33amiconn..and allocate_entry casts a (char *) to (struct dircache entry *)
22:51:37amiconn=> boom
22:51:49amiconnNo, based on your section name
22:52:06preglowat least my data abort handler works, then
22:52:10amiconnSomehow the SH1 CPUAdrErr doesn't behave as it should
22:52:26LinusNpetur: CS3
22:52:57Sandoes H300 support #include <cstdlib> ?
22:53:04LinusNRD/RW are connected as usual
22:53:07preglowbut we've really got to fix this long call stuff
22:53:18lostlogicgah, the PCF50606 might not even be in charge of charging the battery.
22:53:20preglowthree instructions are used for every call
22:53:23preglowas opposed to one
22:53:27preglowin short call mode
22:53:38Rob2222LinusN: Have you looked at the charge system of the H300 closer? It lookes like the LX2201 chip charges the unit, not the PCF50606. Can it be?
22:53:42LinusNSan: rockbox is C, not C++
22:53:58peturLinusN: thanks. Either your schedule was pretty empty or you've got strange priorities :D
22:54:24LinusNi just felt sorry for you :-)
22:55:04peturdon't. before you know it I'll start nagging about my patches again ;)
22:55:38 Join akaidiot [0] (
22:56:01Rob2222can the current of a multimeter kill a chip when doing measurements on a PCB?
22:57:14Rob2222while tracing some pins
22:57:15 Join Skydemon [0] (
22:57:28 Join Midgey34 [0] (
22:57:47LinusNRob2222: i doubt it
22:57:53***Saving seen data "./dancer.seen"
22:58:24amiconnBah, can't cvs up anymore
22:58:43lostlogicamiconn: :( my fault, LICENSES licenses
22:58:48 Join einhirn [0] (
22:59:02lostlogicamiconn: Bagder can remove the dir from the server, or someone else with shell access, didn't know CVS windows was so stupid.
22:59:04 Join the_winch [0] (
22:59:09LinusNRob2222: haha, that may very well be true :-)
22:59:10amiconnSomeone has to remove this, either directly on the server or from a linux box
22:59:26LinusN(the lx2201 being the charger)
22:59:35Rob2222yeah, i think so, too
22:59:49amiconnlostlogic: This has nothing to do with cygwin cvs, but with the filesystem
22:59:51Rob2222after the PCF experiments from lostlogic and me ..
22:59:57BHSPitLappypreglow: sup, anything cool going on?
23:00:06amiconnBoth fat and ntfs preserve case, but don't distinguish it
23:00:11preglowBHSPitLappy: adding illegal exception handler
23:00:23*Bagder removed the licenses dir
23:00:28lostlogicBagder: thanks!
23:00:28Rob2222if the lx2201 is the charger, that is the cause of all the strange PCF behaviour.
23:00:33BHSPitLappypreglow: kick-ass!
23:00:38amiconnSame would happen on linux if you'd put a cvs working dir on a fat partition
23:00:43Rob2222But the LX2201 isnt programmable :(
23:00:44preglowBHSPitLappy: why? it just tells you when something went bad, heh
23:00:54lostlogicRob2222: that's fine, at least we know who to ask what is going on now instead of the PCF
23:01:04LinusNhow ironic
23:01:26*amiconn fires up vmware
23:01:28lostlogicamiconn: hmm, trunuf. sorry about that.
23:01:46Rob2222yep. :( But i wished to had a chance to program the charge mode for myself
23:02:23lostlogicRob2222: me too.
23:03:23Rob2222we only need to connect the battery to the pcf. lol
23:03:49lostlogicRob2222: so, how do we ask the lx chip about the battery status?
23:03:58Rob2222so H300 Charging can be closed... ^^
23:04:18Rob2222hmm, i think the battery + is connected to the ADC of the PCF, too.
23:04:25Rob2222but i am not sure
23:04:39lostlogicvoltage doesn't tell us full battery.
23:04:40LinusNok, RCP is 100k, giving Ibat(max)=0.65
23:04:45preglowcompiling rockbox with anything but -O on arm makes it crash
23:04:50preglowon a reloc error
23:05:47amiconnMore than -O doesn't work for SH1 either
23:05:57preglowin thread.c
23:06:05preglowi've never liked depending on that inline magic
23:06:54Rob2222lostlogic: voltage is enough information, if you didnt need to control the charger, or?
23:07:09Rob2222when discharging you only had the voltage, too.
23:07:19LinusNcharge level=4.1V
23:08:04lostlogicLinusN: are you measuring the pins of the lx2201 currently?
23:09:14lostlogicnice −− that seems to indicate that our battery scale factor and battery voltage-to-percents are somewhat off if it's only charging to 4.1V but we often see 4.17V?
23:10:14LinusNgpio23 = charging status (1 = charging)
23:12:17Rob2222charging statis? has the LX2201 a status output pin? *still looking for that*
23:12:32amiconnLinusN / Bagder: thx
23:12:45Rob2222ah, yes, indeed. stat
23:13:05lostlogicwella t least we can listen to the charge status...
23:13:33Rob2222LinusN: Do you put that infos in the wiki?
23:13:40lostlogicI wonder what the MBC pins of the PCF50606 are connected to then :-\ that they feed current to when it's saying it's charging (and lying)
23:13:43Rob2222Or should/may I?
23:13:59 Join RotAtoR [0] (
23:14:49LinusNit's there
23:15:02Rob2222hehe ok
23:15:56Rob2222maybe should be updated, too. that the UNKNOWN CHIP _is_ the charging chip.
23:16:51LinusNme == silly, the charge level is 4.2V
23:17:38amiconnCan the lx2201 be switched on & off
23:18:01*amiconn wonders how the iriver firmware switches charging
23:18:59lostlogicyes it can
23:19:08LinusNyes it can, but it seems VID0 is tied to VCC
23:19:10lostlogicif the VID pins are controllable
23:19:13lostlogicso no it can't
23:19:44lostlogicthe chip does the 'right thing' and lets the battery level fall while the device is plugged in for long periods of time before topping off
23:19:59lostlogicso I dont' see a reason to control it (as that's what I would have done on the PCF if it was the charger)
23:20:53amiconnSo what's the deal with the iriver firmware's 'usb charging' mode?
23:21:03Rob2222thats VID1
23:21:48amiconnAfaiu, the lx2201 will always charge when an external power source is connected
23:21:57amiconnOr did I miss sth?
23:22:31Rob2222hmm, i was to fast with VID1, maybe...
23:22:40lostlogicamiconn: it looks that way, probably is still charging when USB is connected even if it's in disk mode... but reduces to low level perhaps?
23:22:48lostlogic(only draw 100mA from USB)
23:23:20 Quit darkless (Read error: 110 (Connection timed out))
23:23:57amiconnI think USB only feeds the charger and the unit runs from battery
23:24:01 Join darkless [0] (
23:24:23Rob2222would meet the specs. USB low curreent limit 93mA
23:25:45preglowlooks like i'm going to have to pass the exception address around as a parameter
23:26:14preglowno stack frame to look into to see where we came from
23:26:29lostlogicamiconn: from the block diagram, the VDD can be supplied from either VIN or VBAT
23:26:44preglowoh, this is how archos does it anyway
23:27:22Rob2222"VID1 logic input selects USB charge currents."
23:28:28Rob2222LinusN: have you checked if VID1 is wired, too?
23:30:13Rob2222It looks like the LX2201 always charges when connected to USB. But you can select the charge current with VID1.
23:30:42Rob2222Typical 90 / 460 mA.
23:31:07 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
23:32:24preglowwe'll be running the arm in supervisor mode, mostly, which makes it highly unrecommended to use software interrupts
23:32:39preglowshould i make them call UIE?
23:33:01 Join amiconn_ [0] (
23:33:48 Join linuxstb [0] (
23:34:16 Quit amiconn_ (Client Quit)
23:36:52 Part SteL
23:38:00 Join nico__ [0] (
23:39:28 Part the_winch ("Leaving")
23:39:31nico__hello !
23:40:01nico__i have some problems with ogg-vorbis file
23:40:40nico__when i try to read it, the player stop immediatly and say then len is 0 seconds
23:40:44 Join gtkspert [0] (
23:40:57LinusNi'm trying to find out where vid1 is connected, but so far no luck
23:41:32nico__my player is an iRiver H210 with the daily buil of the 17 janury of 2006
23:41:45preglownico__: you'll have to give us access to that file somehow
23:42:52Rob2222LinusN: have you checked the GPIOs?
23:43:08LinusNof course i have
23:43:11lostlogicLinusN: the iRiver firmware has that HUB vs. Desktop mode selector int he menu so it is connected somehow.
23:43:11nico__there is known bug about the vorbi codec ? its a music i have encoded myself
23:43:43lostlogicnico__: not a known bug
23:44:31amiconnlostlogic: Do you have an idea what that setting actually does? I don't...
23:44:37preglowif it was a known bug, i wouldn't have asked to see the file
23:44:41 Part Paul_The_Nerd
23:45:10lostlogicamiconn: most likely changse between HUB (100mA) and Desktop (500mA) USB charge modes...
23:45:15nico__yes ... i search how to give you this music (maybe i can restart my ftp server but i don't do it for few month ...)!
23:45:15amiconnIt's absolutely ambiguous. Both a desktop and a hub should be able to provide 500 mA, unless it's a bus-powered hub
23:45:22lostlogicamiconn: right
23:45:53Rob2222hub selected max 100mA from USB
23:45:53 Join NightCat [0] (
23:46:01NightCatHi All!
23:46:11Rob2222for that they used VID1 of the LX2201
23:46:50Rob2222LinusN: How do you know if you have a hit? 1 ohms? But well, maybe is my only H300 not the unit i should start tracing lines.
23:46:51preglowhah, prefetch abort works as well!
23:47:20Rob2222with 1 ohms i meaned 0 ohms
23:48:30NightCatCan Anybody explain what is a new profilling (profiles) plugin and other?
23:48:37preglowlinuxstb: Checksum in bootloader doesn't fit on nano screen
23:48:58lostlogicNightCat: profiling isn't a plugin. it's fro performance analysis.
23:49:05XavierGrOk here is the deal (I need help). I have a thread going on (on a plugin) what if I want to end the thread if the user connects the player with the USB cable? I see from the examples that I can use button_get and, but this clogs up the thread and makes the player sluggish. Am I doing something wrong or there is another path for it?
23:49:20linuxstbpreglow: How many chars are missing?
23:49:43lostlogicXavierGr: just don't check the button that often, use the non-blocknig button_get
23:49:47lostlogic(block = false)
23:49:50preglowa bit hard to say, it boots so fast, perhaps one? two?
23:50:25BHSPitLappydoes BHSPitLappy need to take a picture of his? *grumble*
23:50:28linuxstbI could just replace "Checksum:" with "Sum:"
23:50:41preglowor put it on a separate line, we've got tons of them nw
23:50:50stevebok for some reason i just cannot get my h340 to load a patched firmware. it does the firmware upgrade, shuts down, but just loads up in the origonal one i had on before
23:51:00LinusNRob2222: i have a beeping multimeter
23:51:02linuxstbpreglow: Do you want to do it?
23:51:09preglowsure, can do
23:51:16lostlogicRob2222: most multimeters have a 'continuity' beep
23:51:34Rob2222Yeah, ok, I have a beep, too.
23:51:47Rob2222does it beep at 0 ohm? never checked
23:51:51LinusNfound a pullup
23:52:02Rob2222LinusN is the master
23:52:03XavierGrwhat button_get_w_tmo does?
23:52:12lostlogicXavierGr: button get with timeout
23:52:27lostlogicLinusN: pullup -> always high, can't configure USB charge current?
23:52:57NightCatXavierGr: What are you working at now? (Just my interest :) )
23:53:02XavierGrlostlogic: So it will sniff a button once every x seconds?
23:53:26 Part Kaggen
23:53:50Rob2222lostlogic: no, a pullup means only, that a chip can pull that line to GND
23:53:51XavierGrNightCat: I am just making some modifications on my battery_bench plugin. I want to make it more secure. Then maybe Bger will commit it if noone else has objections.
23:54:18Rob2222And it means, that Linus is on the right way.
23:54:31NightCatXavierGr: Thx for answer :)
23:55:44LinusNlostlogic: nope, i saw that it was low when i started the h300, so it is definitely controlled
23:55:53lostlogicXavierGr: except it will block on that call for the timeout unless a button is pressed
23:57:13stevebok. i just tested it. my H340 doesnt want to load a patched firmware.
23:57:16XavierGrlostlogic: Sorry but what do you mean exactly with "block"?
23:57:23stevebany one have any ideas about why its not likeing it?
23:57:43stevebit doesnt give me any errors and the patcher said it did it sucessfully...
23:58:07lostlogicXavierGr: means that your program will stop at that button_get_w_tmo function call until either it gets a button, or the timeout expires
23:58:15XavierGrsteveb: It will ignore the file or it will not boot?
23:58:23stevebXavierGr: it boots
23:58:30stevebit just doesnt upgrade the firmware
23:58:54stevebit doesnt tell me it hasnt upgraded though - it turns itself off and everything
23:58:55XavierGrdid you tried to upgrade an unpatched file. (without rockbox to see if you can upgrade)

Previous day | Next day