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 2006-04-27

00:00:06linuxstbjulius: f you look at the diff, it just enables for the 4g ipod what they used to do for the 3g ipods. Plus they've added some comments.
00:00:49juliuslinuxstb, so the code shoud be this one: ipod_i2c_send(0x8, 0x24, 0xf5);
00:00:52 Join Aditya|Nap [0] (
00:01:03amiconnThe if() in always evaluates to false even though sh.h 2593 and sh.h:2595 should set the flag
00:01:33 Part Aditya|Nap
00:02:54amiconnI have 2 choices without further hints what's going wrong here: (1) Drop the unsigned byte case and let gcc switch to word sized table for >127 or (2) Drop the signed case and always insert an extu.b for byte-sized table
00:02:55juliuslinuxstb, anyway, I will try to send this commands to my Ipod, lets see what happens, but before I can do that, where is the right place to send this commands using rockbox?
00:02:58preglowi wish i understood this gcc stuff
00:03:24preglowamiconn: i can't really see where the signed case would be useful
00:03:31preglowamiconn: ever seen a jump table with a negative jump?
00:03:32amiconnIt saves an instructin
00:04:26linuxstbjulius: There are three different writes - check the pcf50606 datasheet for the details of what they are doing. Basically, they enable the DCREG1, DCREG2 and DCREG3 to 3.0v, 3.3v and 3.0v respectively. So I'm guessing DCREG2 is the 3.3v power-out on the dock connector.
00:05:08amiconnBut yea, I've never seen a jump table with negative offset in gcc's output. But check SH1 memcpy, there I do use such a table...
00:05:27preglowi like the fact that the sh backend is actually bigger than the m68k one
00:05:31linuxstbjulius: Register 0x24 is DCREG1, 0x25 is DCREG2 and 0x26 is DCREG3
00:05:39preglowperhaps it's due to all the other processor revs
00:06:29 Quit PaulJam (".")
00:06:46 Quit midgey34 (Read error: 110 (Connection timed out))
00:06:47 Quit scott666|work (Read error: 110 (Connection timed out))
00:06:48amiconnpreglow: The sh family is quite large. SH5 is even 64bit iirc
00:06:57 Join goldschlager [0] (
00:08:07goldschlagerdoes anyone in here has x5? i just have a Q
00:08:22juliuslinuxstb, aaa ok, I see now what you mean, I will try to send all the tree commands, just to be sure :) anyway, where should I add this commands? (I did a cygwin devel platform on my win box, so I can recompile rocbox, I already did for some changes for my theme, but anyway, is there a .c file which handles this things in rockbox source?) (ps Thanks for your greate help, I really apreciate it!!!)
00:08:31linuxstbjulius: I'm not sure where this should be added into Rockbox. You could just add them to the end of i2c_init() in firmware/drivers/i2c-pp5020.c for testing.
00:08:32Bagdergoldschlager: ask and you'll find out
00:09:12goldschlagerok.. so after i turn off my unit my screen goes dark green..
00:09:15 Quit midkay (Read error: 104 (Connection reset by peer))
00:09:22goldschlagerso my q is wtf?
00:09:34 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:09:53Bagderand my a is:
00:09:57linuxstbjulius: No problem. Please let us know if it solves your problem.
00:10:04juliuslinuxstb, thanks a lot again!, I cross my fingers :), I'm a bit late because it's 1am here, but I will get back to you with the result as I know more
00:10:48goldschlagerno.. i asked on the forums someone told me its VGA controller chip..
00:11:07goldschlagerbut can it be firmware?
00:11:49BagderI don't even understand your first question
00:12:39goldschlagerhm.. my question was whats the problem?
00:12:53Bagderand what exactly is the problem?
00:13:43goldschlagerlike after i turn of my x5 instead of black my screen goes dark green
00:14:10goldschlagerand you can see vertical lines.. light green
00:14:12Bagderyou mean it actually did turn green forever
00:14:26goldschlagerno.. it works normally
00:14:41goldschlagerbut only when i turn it off
00:14:51goldschlagershis thing happens
00:15:59 Quit KlrSpz ()
00:16:29juliusThanks everyone for the greate support! Rockbox really rocks! :) so keep up the good work! I'm going to sleep now, by
00:16:43 Join midgey34 [0] (
00:17:47 Part julius ("Leaving")
00:18:49 Join myth1 [0] (
00:18:57myth1hi again
00:19:15myth1that mod bug bugged me so I couldn't get to bed ;)
00:19:27Mikachusorry :)
00:19:33myth1but i fixed it ;)
00:19:36dwihnoSleeping is just a waste of time anyhow :)
00:20:20myth1it is interesting that "char" seems to be unsigned with gcc
00:20:28Mikachui think it depends on target :)
00:20:29myth1with visual studio it is signed
00:20:31 Join ashridah [0] (
00:20:36Bagderdepends on target, yes
00:20:38Mikachuyou can also say -fsigned-char
00:20:49myth1well, that was the bug ;)
00:20:56BagderANSI C left it unspecified I believe
00:20:57 Quit elinenbe (Read error: 110 (Connection timed out))
00:21:45Mikachumyth1: which char?
00:22:17myth1Mikachu: row
00:22:28myth1I'm uploading a new diff
00:22:54Mikachuyou seem to have specified most of the other ones
00:23:48 Join klrspz [0] (n=klrSpz@
00:24:14klrspzdo you guys ever defrag your devices? I just defragged my ipod, wonder if it improves speed any?
00:24:15myth1also remove the #ifdef USE_EFFECT_E
00:24:26 Quit RedBreva_ ("using sirc version 2.211+KSIRC/1.3.12")
00:24:30Mikachui'll jsut wait for the diff
00:25:27 Quit klrspz (Client Quit)
00:27:40midkayany objections to resizing the EQ bar 'knobs'? a long time ago something was changed that made them much wider and i don't think it was intentional, or if it was, it's kind of ugly to see the bar slide past the end..
00:30:34myth1mmh, cvs diff takes ages
00:30:35 Join elinenbe [0] (
00:32:23 Join webguest75 [0] (
00:32:33myth1does cvs have some problems?
00:32:47 Quit muesli__ (Read error: 110 (Connection timed out))
00:33:01webguest75can someone help me?
00:34:01 Part goldschlager
00:34:15webguest75can someone help me?
00:34:16midkaywebguest75, without telling us your problem, no.
00:35:37webguest75wen i try to install rockbox on my video ipod i get to step f after i complete it i try to test my ipod and nothing happens
00:36:03 Part YouCeyE ("Leaving")
00:36:10midkaywebguest75, 'nothing' happens?
00:36:13webguest75rockboot.bin: No Such file or directory is what i recieve
00:36:24midkaythen you skipped a step or typed it incorrectly.
00:36:48webguest75wen i am supposed to see the Rockbox bootloader.
00:36:58webguest75i double checked i ddnt
00:37:31midkaywebguest75, there's no reason for it not to exist, if everything was typed correctly.
00:37:48midkaythe commands create the file, you don't have the file, so you typed the commands wrong somehow.
00:38:25webguest75what does You can now disconnect your ipod from your PC in the normal way. mean eject it from itunes?
00:38:32myth1the new mod.c is online
00:38:39 Join actionshrimp [0] (
00:38:52midkayhowever you normally disconnect it, yes.
00:40:45webguest75-rockbox.bin apple_sw_5g_rcsc.bin apple_os.bin bootpartition.bin
00:40:52webguest75those are the files it created
00:40:56webguest75what am i missing
00:41:34 Quit bleaked (Remote closed the connection)
00:42:03midkaywhat's "-rockbox.bin"?
00:42:13webguest75dont know it made it
00:42:33midkayi don't think that's any step in the instructions.
00:42:43webguest75k thx that was my mistake
00:43:13 Quit Moos (Read error: 110 (Connection timed out))
00:43:30 Join Nep [0] (
00:43:57 Quit Nepbaland (Read error: 104 (Connection reset by peer))
00:44:05 Nick thesoffork is now known as thesonork (
00:45:28thesonorkso, i finally managed installing rockbox on my ipod, BUT it crashes after 1 second or something,... the bootloader appears, then rockbox itself appears, and then it reboots, time and time again.... why?
00:45:50thesonorkno error-output or anything
00:46:17midkaythesonork, low battery possibly?
00:46:39thesonorksays full battery in disk mode
00:47:02thesonorkon the rockbox-screen a USB logo appears, thats all i see
00:47:18midkayyour build is recent?
00:47:47thesonorktook that 1
00:48:20midkaysure you have the right generation?
00:48:44thesonorkwell some1 told me, when the capacity is written on the back its 2nd gen
00:49:40thesonorkthink i shall try that 1st gen OS and bootloader?
00:49:42 Quit RotAtoR ()
00:49:50 Join RotAtoR [0] (n=e@rockbox/developer/rotator)
00:49:53midkayhm, i think the capacity on the back indicates 2G, yeah.
00:50:23midkayis the usb cable plugged in? you said you saw the USB logo..
00:50:35thesonorkthe usblogo appears even if it is not
00:50:39thesonorki jsut tried that
00:50:44thesonorkbut when its plugged
00:50:50thesonorkit automatically reboots to diskmode
00:51:03thesonorkdoes this mean "i have no songs" or something?
00:51:12thesonorkcuz my ipod is new-formatted
00:51:51midkayno.. did you install rockbox entirely? that whole zip needs to end up on the player, with full paths.. there should be a rockbox.ipod file and a .rockbox folder..
00:52:17thesonorki just copied everything into /dev/sda2
00:52:19 Quit myth1 ("leaving")
00:52:41*midkay scratches chin
00:52:59thesonorksame here
00:53:11thesonorksoem strange stuff isnt it
00:53:28midkaysounds like it should be working, yeah.
00:53:40thesonorkwhat about i try that daily build stuff?
00:54:51midkaywhat are you trying as it is? CVS build?
00:55:13midkayyou could give it a try, no reason the cvs one shouldn't work though.
00:55:52thesonorksame stuff
00:56:15linuxstbthesonork: one or two other people have reported similar problems with their 5g/video ipods - it seems that in a tiny number of cases (you are #3), Rockbox is mis-detecting USB, and always giving a false positive.
00:56:17midkaystill doesn't work, you mean?
00:56:51thesonorkya thats ewhat i mean
00:57:21thesonorkthats bad man
00:57:28thesonorkim just using a damned mini ipod ^^
00:57:42thesonorkany patch available or any suggestions to do?
00:58:06linuxstbI could easily make a patch for you (to disable USB detection) if you were able to compile Rockbox yourself.
00:58:19thesonorkusing make ?
00:58:25 Quit webguest75 ("CGI:IRC")
00:58:35linuxstbAre you running Linux?
00:58:51midkayi could compile it if you aren't set up to do it and if linuxstb doesn't want to.. just started VMware. :)
00:59:18thesonorkthat sounds great mates
00:59:22linuxstbThen it's almost as simple as just running make. You need to install a gcc cross-compiler.
00:59:33thesonorklinuxstb, if u could do that patch, would be really glad
00:59:55linuxstbmidkay: I don't have a clean source tree handy, so please go ahead and build it.
01:00:51midkaylinuxstb, sure.
01:01:00linuxstbJust edit the usb_detect() function in firmware/usb.c to always return false.
01:01:37thesonorkdoes that mean the usb-detection isnt working? so, how to apply songs and stuff?
01:02:10linuxstbIf you want to enter USB mode, you just need to manually reboot - either into the Apple firmware, or into the emergency disk mode. All Rockbox does on USB detect is to reboot into the emergency disk mode anyway.
01:02:32midkaylinuxstb, got it. by the way, not sure where it would go, but an ata_sleep before rebooting to disk mode for iPods would be nice. cutting power to the drive like that probably isn't very healthy. :)
01:02:33linuxstb(which is why you are getting the constant reboot cycles)
01:03:19linuxstbmidkay: We should do a proper shutdown - saving settings etc.
01:03:45midkaylinuxstb, even better.. should be very simple but i'm not sure where that's handled.
01:03:50 Quit herz42 (Read error: 110 (Connection timed out))
01:05:26linuxstbNo, I'm not sure either.
01:05:48Presenceaah, THATS the difference between "disk modes" from apple's vs rockbox's: Rockbox's is the "emergency" disc mode.
01:06:05linuxstbThat's what I call it anyway... But yes, there are two different disk modes.
01:06:20midkayPresence, yes, that's hilarious. :)
01:06:20linuxstbBoth of them are "Apple's" though.
01:06:35amiconnNow, let's see what happens...
01:07:04*amiconn changed the jumptable mechanism a bit
01:07:45midkaythesonork, k, i have a version for you to try. can you PM me your e-mail?
01:09:44thesonorktell me when sent
01:10:13midkaythesonork, did you try? i didn't get it. must be freenode blocking PMs from unidentified users...
01:10:14 Join DarkJesus [0] (
01:10:56thesonorktake it here
01:11:14midkayk :)
01:12:12 Quit DarkJesus ("Leaving")
01:12:21 Quit nudel ("At Argon, we're working to keep your money.")
01:12:51thesonorkk ill have it a try
01:12:59 Join DarkJesus [0] (
01:14:02DarkJesusDoes anyone feel like talking about an audioscrobbler plugin?
01:14:08DarkJesusIs it possible or not?
01:14:28DarkJesusI heard the time would have to be recorded, which is impossible on a H120
01:15:00thesonorkim thanking u badly guys
01:15:02 Join pihlopase [0] (
01:16:43 Join nudel [0] (
01:17:03amiconnmidkay: If you're registered, you can allow pms from unregistered users
01:17:28 Quit obo (Read error: 110 (Connection timed out))
01:17:48Inctime to re attempt rockbox
01:17:57Incdamn ipod fucked itself last time
01:19:07dwihnoI haven't used the apple fw for music.. Is it good? ;)
01:19:57midkayamiconn, hm.
01:20:22midkaythesonork, cool ;)
01:20:54thesonorkok :)
01:21:10thesonorkso now i can go to sleep and drive to work tmrw, with music ;)
01:21:44midkayyay :)
01:22:57DarkJesusDid anyone see my question?
01:23:39midkayDarkJesus, it's entirely possible, and two different attempts are being worked on, i believe.
01:26:45 Join matsl_ [0] (
01:27:18***Saving seen data "./dancer.seen"
01:28:57 Quit ashridah ("Leaving")
01:29:59 Quit Poka64 ("bbs")
01:33:38 Join TCK [0] (
01:34:58*preglow does another request for files that break the resampler
01:35:31preglowi actually believe lostlogic fixed what glitches there were
01:42:30DarkJesusThat's so cool
01:42:37DarkJesusI'd love that so much
01:42:43DarkJesusI'm gonna sleep now
01:42:56 Quit DarkJesus ("Leaving")
01:45:49preglowanyone opposed to me commiting toni's patches?
01:50:44 Quit matsl_ (Remote closed the connection)
01:51:33 Quit ender` (Read error: 113 (No route to host))
01:55:58 Join JdGordon [0] (
02:00:38 Quit adiamas (Read error: 110 (Connection timed out))
02:02:54preglowlostlogic: playback seems pretty functional right now
02:04:46 Quit Bg3r (Read error: 110 (Connection timed out))
02:07:45midkaypreglow, i'd be glad to see them in cvs :)
02:08:07*midkay yearns for the day at least EQ and hopefully crossfeed work well on iPods with MP3.
02:08:15 Join adiamas [0] (
02:08:53biffherook, am I an idiot? (don't answer that). I am using 'mv [a-z]*.mp3' , and it is moving the ones which start with [A-Z] too. ugh.
02:09:19Mikachucan you do [a-z] in a shell, i don't think so?
02:09:28zeyes, you can
02:09:33midkaybiffhero, there must be some switch for case-sensitive..
02:09:35Mikachuthere i go
02:09:47zemidkay: nope
02:09:57biffheromidkay I was looking, don't find it in 'man', nor in 'info'
02:09:58zethe only way i can see that'd work is if the filesystem itself is case-insensitive
02:10:24zelike a window filesystem
02:10:39biffherosigh. /dev/sda1 on /mnt/local/sdb1 type ext3 (rw)
02:10:56zeok well that makes no sense
02:11:15 Join Paul_The_Nerd [0] (
02:11:18zesure you didn't accidently put a space between [a-z] and *? :p
02:11:24biffheroso it seems like shell problems, eh? I bet it is something defined in /etc/bash_completion. That bugger is big.
02:11:30zewhat if you echo [a-z]*.mp3
02:12:01zebiffhero: thats for tab-completion.. if you're not hitting tab, it shouldn't be trigger
02:12:08Mikachu{2:11:53:/tmp/a}% echo [a-z]*
02:12:08Mikachua b c
02:12:08Mikachu{2:11:57:/tmp/a}% echo [A-Z]*
02:12:08MikachuA B C
02:12:09Mikachu{2:12:00:/tmp/a}% bash
02:12:09biffherogood call
02:12:11Mikachumikaelh@fartmonstret /tmp/a $ echo [a-z]*
02:12:13Mikachua A b B c C
02:12:18Mikachunot bash there is zsh
02:12:34biffheroboat_drinks.mp3 Boat Drinks.mp3
02:12:42Mikachuso try zsh ;)
02:12:47zeat least i think so, i dunno that much about bash's newfangled stuff
02:13:04zeMikachu: i've never had bash match case-insensitively
02:13:31Mikachuand that's with nocaseglob off
02:13:33luckzhow does rockbox decide whether to assume that a m3u file is in UTF or not?
02:13:46Mikachuluckz: it doesn't
02:14:28Mikachubut what?
02:14:38luckzdoes it always assume UTF?
02:14:47Mikachui'm not sure you understand your question
02:15:13luckzokay. I have a m3u file created with foobar pointing at file names using umlauts.
02:15:19luckzrockbox cant play those.
02:15:26luckzat least not using that playlist.
02:15:29luckz13:18:27 < preglow> luckz: 2) that's probably a codepage issue. rockbox might think your m3u files are utf-8 encoded
02:15:55zeer wtf
02:16:01zebiffhero: now my bash is doing like yours
02:16:04Mikachumaybe your player is mounted with another codepage in windows then
02:16:18Mikachuze: [] seems to be special
02:16:59biffheroze: there you go, I have cursed you.
02:17:17biffheroze: the bash manpage seems long. working my way through it now.
02:18:39Mikachubiffhero: try {a..z}*
02:18:46Mikachuit will complain about no such file, but it should work anyway
02:18:57zewell i found the shell option 'nocaseglob'
02:19:03zebut unsetting it doesn't seem to help
02:19:35zeactually it seems to be off by default
02:19:39zeif it were on, that'd explain it
02:19:45zebut its not on here and yet my bash still does that
02:20:09Mikachu$ ls [a]*
02:20:12Mikachu$ ls [a-a]*
02:20:14zeit doesn't seem to be respecting it
02:20:14Mikachu$ ls [a-b]*
02:20:14Mikachua A b
02:20:19Mikachui think i have realized the issue
02:20:36Mikachu$ ls [a-b]*
02:20:36Mikachua b
02:20:45Mikachuit simply uses the locale collation order
02:21:11Mikachuin C, a-z comes before A-Z, but in english, it's aAbBcC...
02:21:24Mikachubiffhero: ^
02:21:32zeah damn locale crap
02:21:49zeLC_COLLATE seems like the more appropriate one, but i dunno
02:21:52Mikachui don't feel matching should depend on locale
02:21:58MikachuLC_ALL overrides everything
02:22:02zeand/or LC_CTYPE
02:22:16zeyeah maybe LC_ALL is best
02:22:40Mikachuyou could also do [abcdefghijklmnopqrstuvwxyz]*
02:22:44Mikachuonly takes 3 seconds to write
02:22:58zeand yeah i didn't think matching was locale-specific
02:23:13zethough i gotta say in my experience, A-Z always came before a-z
02:23:16Mikachuit's just the expansion of a-z that is locale-specific, in bash
02:23:34Mikachuthe actual matching is not locale specific :)
02:23:53biffhero'ls {a..z}*' was interesting, it gave me errors for everyone I didn't have.
02:24:13zeah yes A-Z does come before a-z in LC_ALL=C
02:24:15biffheromy nocaseglob was the same as yours.
02:24:44zeheh what a bunch of crap
02:25:08Mikachubiffhero: {} is a generator, [] is a matcher
02:25:23biffheroMikachu: abcdefghijklmnopqrstuvwxyz.... 3 seconds.... how lazy do you think I am? I will spend 30 minutes figuring it out, instead of 30 seconds doing the other 5 times.
02:25:25zenocaseglob and locale-specific expansion/sorting both together seems screwy
02:25:41Mikachubiffhero: that was sort of a joke :)
02:25:45zebiffhero: indeed
02:26:01biffheroexport LC_COLLATE=C did it
02:26:08zei've many times spent an hour figuring out the 3 second way to do something that'd other wise take 10 seconds
02:26:24zejust on principle, its better to know those things
02:26:31zeespecially if you need to do it several times in a row
02:26:43zelearning is less tedious than the manual methods
02:26:46zeeven if it takes longer
02:26:48Presencemakes yah more useful on IRC, too.
02:27:17zeand yeah the bash man page is like entering some underground labyrinth
02:27:30biffheromost GNU docs is that way
02:27:39Mikachuman zshall is fun
02:27:56zeit always feels like some uncertain excursion fraught with peril
02:27:58 Join _Lucretia [0] (
02:28:00 Quit xmixahlx ("blah blah blah")
02:28:07 Quit _Lucretia_ (Read error: 110 (Connection timed out))
02:28:17zeand you always get the feeling you may get stuck down there for far longer than you intended
02:29:06zemaybe they oughtta do a nethack-like documentation system
02:29:08biffherolike when I play nethack
02:29:30zeyou might find the information you're actually looking for easier
02:29:54 Join eu4ic [0] (
02:31:00Mikachuscreen has options for nethack style messages
02:32:43 Quit eu4ic (Client Quit)
02:34:29 Quit pixelma (" HydraIRC -> <- :P")
02:36:11Hotfusionman I love the item duplication in oblivion
02:36:32HotfusionOT but I gots money
02:36:32Mikachuslightly offtopic?
02:37:01HotfusionI think when I get paid on my next pay day I'll get new head phones
02:37:04sharpeslightly, just slightly.
02:37:18Hotfusionthe default X5L ones clip at volume 21 default iaudio firm
02:37:34Presenceyikes, some of the more "active" WPS's really suck the battery down, huh?
02:37:35Mikachuyour mind is like a blender
02:38:11Hotfusionthough I do like the hanging phones from audio technica
02:38:11 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
02:39:14 Join tf4 [0] (
02:39:42tf4can someone assist me with a itunes alternative for my new rockbox ipod
02:39:57tf4i need something that can sync my music folder when i plug in my ipod
02:40:07tf4i dont want to have to drag and drop everything
02:40:16sharpegoogle for SyncToy. heheh.
02:40:21Mikachuwow, i never thought i'd see the day when someone was too lazy to drag and drop
02:40:22sharpetis by microsoft.
02:40:35tf4i want to have it on my puter first
02:40:41tf4then sync to pod when i plug it in
02:40:44tf4like itunes does
02:40:59tf4i sync podcasts all day long and sync before i go to work in the morning
02:41:12sharpeyeah, synctoy will do that i believe
02:41:12Mikachumaybe you could change your name to Mr Sync
02:41:37 Nick tf4 is now known as mrsync (
02:41:46Paul_The_NerdDoesn't that do syncing with UMS devices?
02:41:58mrsynccan it aggregate my podcasts too?
02:42:34Paul_The_NerdProbably. I couldn't say for sure though, I've never used it.
02:42:38Paul_The_NerdIt's supposed to be pretty powerful
02:43:05sharpei've never used any syncronization software...
02:43:07mrsyncdoes it have spyware? the websit looks a little iffy
02:43:26Paul_The_NerdIt's been recommended by a *lot* of people.
02:43:53mrsyncthat monkey looks like the 'punch the monkey' fucking banner so i was a little nervous
02:43:55Paul_The_NerdI just don't really like automation.
02:43:58Hotfusionfor TD3-Tagit has a better tagging interface than media monkey IMO
02:44:15TeaSeaI like amarok!
02:44:37mrsynci wanna use it for converting all my cds to flac then syncing to my pod
02:45:44mrsyncdoes amarok work with album art through rockbox?
02:46:06TeaSeamrsync: ?
02:46:32mrsynchmmm well i cannot get my rockbox firmware to show my album art
02:46:45mrsyncbut i see album art on amorok
02:46:46TeaSeaThere's a plugin for amarok which will generate cover.bmps for all the albums in your collection, but that's about it.
02:46:58TeaSeamrsync: Ah. Album art hasn't been implemented except via a patch.
02:47:06TeaSeaWait until rockbox 3.1, that'll have much better support.
02:47:08 Quit Nep ("—I-n-v-i-s-i-o-n— 2.0 Build 3515")
02:47:14mrsynci think the dudes who wrote rockbox should make a itunes replacent as well
02:47:20mrsyncthey did such a good job on the fw
02:47:32Paul_The_NerdI think they should keep making Rockbox better.
02:47:37TeaSeaI think so also!
02:47:46TeaSeaThere's far too many itunes replacements out there anyway.
02:47:48mrsyncim still getting used to it
02:47:56mrsynci was trying to shut it off today lol
02:47:57Paul_The_NerdEspecially since I don't think *any* of them like to use their devices in a non UMS way anyway
02:48:05mrsynci spend like 3 min trying to find out how to shut off my ipod
02:48:11Paul_The_NerdIt's in the FAQ
02:48:13Paul_The_NerdAnd the Manual.
02:48:19mrsynci dont RTFM
02:48:24Hotfusionyeah mp3 player manuals don't seem to be the best
02:48:38Paul_The_NerdHotfusion: Not the iPod manual. The Rockbox manual.
02:48:40mrsyncthings should be user intuitive enough to just press shit lol
02:48:57Hotfusionrockbox manual is much better than most manuals
02:48:58Paul_The_NerdYeah, well the iPod doesn't have a button labelled 'power'
02:49:00mrsynci figured it out... so it must be ok
02:49:18mrsynci like how you pause and the sound fades
02:49:21mrsyncthere some nice touches
02:49:30Mikachumrsync: it took you 3 minutes to find out it's the same button as in the apple firmware?
02:49:36Hotfusionyou'd think apple would print power on it. lol.
02:49:44mrsynci just am having a hard time when i dont see 'artists' 'albums' videos' etc
02:49:49MikachuHotfusion: thye should print Power on the play button?
02:50:00Mikachumrsync: maybe you should read the manual after all
02:50:02mrsyncmenu shuts off the ipod video
02:50:04mrsynciff you hold it
02:50:12mrsyncon the rockbox its the play
02:50:20Hotfusiondon't mind me mik, im just mouthin' off
02:50:23mrsyncthey changed that in the 5g's
02:51:11mrsynchow do you guys have your stuff arranged im curious? do you listen to podcasts a lot
02:51:15HotfusionI need to see if my player has a dual boot option for rockbox and iaudio x5 firm
02:51:21mrsyncdo you use a sep synching tool for podcasts
02:51:40Mikachui have a nano so i don't "sync"
02:54:22Paul_The_NerdHotfusion: I don't believe the iAudios dual boot yet.
02:54:31Paul_The_NerdI don't do podcasts.
02:54:52mrsynci love them. thats all i listen to i dont believe in radio anymore
02:55:07Paul_The_NerdMy DAP is full of music. If I want information of any sort, well, the internet is a big place.
02:55:21Hotfusionshoutcast radio for me
02:55:43mrsynci have a 1 hour drive in to work int he morning, watching a video podcast makes it go liek 10 min in my head ;)
02:56:55mrsyncdoes rockbox support video?
02:57:00mrsynci havent tried it yet
02:57:12 Join _Lucretia_ [0] (
02:57:27mrsyncwell thats geh
02:57:39sharpeyou drive for an hour, while watching your ipod?
02:57:53sharpethat's not safe.
02:57:53mrsyncwell the shit i watch is dudes in roundtable discussion
02:57:58mrsyncits not like i WATCH it
02:58:02biffheromrsync: where do you drive? :-)
02:58:03mrsynci glance at it once in a while
02:58:08mrsynccleveland oh
02:58:14biffherook, then I am safe
02:58:19 Quit _Lucretia (Read error: 110 (Connection timed out))
02:58:21mrsyncwhere were filming spiderman III baby!
02:58:22Hotfusionuntil you go off the road and go "oh shit" and carefully drive it onto the road again
02:58:35mrsynchaha i havent done that yet (knocks on wood)
02:58:35Paul_The_NerdRockbox is intended as an audio player anyway. While video support would probably be added if someone made it, it's not exactly a core concern.
02:59:13 Quit perldiver ("some games are better left unplayed")
03:00:13 Join InteliWasp [0] (
03:01:14biffherorockbox should get in with openfi, and subsume that project
03:02:07 Join v3l0ct [0] (
03:02:24TeaSeaHey guys
03:02:27 Quit tvelocity ("Ex-Chat")
03:02:38TeaSeaIs there any way to get rockbox to charge over USB?
03:03:22InteliWaspi keep getting a "Data abort at 00046FC0" on my ipod 4G, is there a workaround?
03:03:24biffheroMy 5G ipod does that.
03:03:33v3l0ctHi, i'm trying to install into a 1G 4gig mini but when I try to run ipod_fw it tells me "bash: ipod_fw: command not found"
03:03:40 Quit mrsync ()
03:03:47biffheroTeaSea: my 5G ipod charges over USB, I thought it was a function of the firmware.
03:03:56lostlogicpreglow: that's good, since I'll be completely computerless for the 4 days starting tomorrow at around midnight your time :)
03:04:19Paul_The_NerdInteliWasp: Under what circumstances does that happen?
03:04:30biffherolostlogic: say it ain't so, work will slow to a halt.
03:04:35Paul_The_Nerdv3l0ct: Well then you probably don't have the ipod_fw binary in the folder you're in.
03:04:45v3l0ctit is there
03:04:58InteliWaspPaul_The_Nerd: when i am playing music, either .mp3 or .flac codexes
03:05:02Paul_The_Nerdv3l0ct: You compiled ipod_fw.c from the tools folder?
03:05:12Paul_The_NerdInteliWasp: With what build of rockbox?
03:05:13lostlogicbiffhero: pfft, there are 30 other devs who could also pick up on the playback engine (I'd pray for them to do so if I was a religious man)
03:05:29biffherov3l0ct: are you running ./ipod_fw ?
03:05:50Paul_The_NerdTeaSea: The main problem isn't that it doesn't charge, but that it seems to draw more power than *some* USB controllers want to give it for some reason. For example, my Nano charges perfectly fine.
03:05:52v3l0ctPaul_The_Nerd, I created a folder named rockbox in my desktop (using linux) and copies ipod_fw.c and the two addition files
03:06:01v3l0ctbiffhero, no
03:06:02Paul_The_Nerdv3l0ct: And then you compiled them?
03:06:14TeaSeaPaul_The_Nerd: mm
03:06:20v3l0ctkaiwai, one sec
03:06:27biffherov3l0ct: it isn't a path problem?
03:06:40TeaSeaCheers anyway
03:06:42TeaSeaAnd now, BED! :)
03:06:53 Quit TeaSea ("Leaving")
03:07:11v3l0ctbiffhero, i forgot the ./ in front of ipod_fw. it ran now
03:07:31InteliWaspPaul_The_Nerd: i am using the 4g daily 20060421
03:07:34Paul_The_NerdInteliWasp: It would be appreciated if you'd confirm the problems exist with a clean install of the most recent build before reporting them. :)
03:07:59InteliWaspdo i just overwrite the old build?
03:09:07Paul_The_NerdThat's the normal upgrade process, but for the purpose of bug reporting, it'd be much appreciated if you installed it completely cleanly (delete the old rockbox.ipod and .rockbox folders, and reset settings while booting for the first time) to make sure that it's a clean test. You may want to back up your /themes and /wps folders, as well as any files that don't have the extension .rock in the /rocks folder
03:09:30 Join insulate [0] (
03:09:54biffheroPaul_The_Nerd: where are the settings set, so that you need to reset them? If deleting rockbox.ipod and ./.rockbox/ aren't enough?
03:10:05v3l0ctbiffhero, thanks :). bootloader installer
03:10:36Paul_The_Nerdbiffhero: The settings are stored in sector 63 on the disk. Settings are reset on iPod by flipping the hold switch to on immediately after powering it up (before the bootloader finishes)
03:11:05Paul_The_NerdI swear someone said 63 yesterday
03:11:15lostlogicmm, the lesser of 62 and start-of-first-partition - 2
03:11:22lostlogicfirst partition usually starts at 63
03:11:29Paul_The_NerdWell, 61 then
03:11:37lostlogicI might have that 'lesser' number wrong, it might be 63 or 61
03:11:37Paul_The_NerdEither way, you reset it like that on iPod.
03:11:51Paul_The_NerdOn the iRivers it's the Record button being held down
03:12:05InteliWaspah so that's why my settings sometimes don't hold :P
03:13:17Paul_The_NerdWell, *most* cases of settings not holding are people plugging in USB without shutting down properly first.
03:14:31 Join perldiver [0] (
03:14:49biffherosettings are only saved on a "shutdown"?? ugh.
03:15:12Mikachuthey are saved on disk spinup too
03:15:18biffherothat's cool
03:15:26Mikachubut that never happens on nano it seems
03:15:31biffherofunny that
03:16:25v3l0ctnew question, how do i mount the ipod once i'm ready to install rockbox
03:16:37biffheroplug it in, "mount /dev/sda2"
03:16:45biffherov3l0ct: on linux?
03:17:01v3l0ctbiffhero, yep. it asks for filesystem
03:17:12biffherov3l0ct: after you plug it in, wait 5 seconds for USB to settle down, run 'dmesg'
03:17:15v3l0ctmount: you must specify the filesystem type
03:17:21biffheroplugging mine in now.
03:18:23biffherorun this: dmesg | grep -i SCSI | tail -1
03:19:06v3l0ctsd 8:0:0:0: Attached scsi removable disk sdc
03:19:26biffherov3l0ct: run this: fdisk -l /dev/sdc | grep -i FAT32
03:19:54biffherov3l0ct: we do that to make sure that you have a FAT32 filesystem on that bad boy
03:20:26v3l0ct/dev/sdc2 * 6 497 3951990 b W95 FAT32
03:20:43biffherov3l0ct: mount /dev/sdc2
03:20:56v3l0ctmount: you must specify the filesystem type
03:21:05v3l0cti have auto in fstab
03:21:09biffherov3l0ct: mount -t vfat /dev/sdc2
03:22:37biffherov3l0ct: my fstab has /dev/sdb2 /media/ipod auto rw,users,noauto,sync 0 0
03:22:41 Quit NC-17 (Connection timed out)
03:24:08v3l0cti can't tell if it's mounting, i get a whole lot of text about usage of mount
03:24:26biffheromount -t vfat /dev/sdc2 gave you a bunch of messages?
03:24:31biffherowait one
03:26:43biffheroI get 'only root can do that' because I am specifying '-t vfat' but as root..... ok, bunch of errors. When you do that, you have to give it a mountpoint, it looks like mount doesn't look at the fstab file there. mount -t vfat /dev/sda2 /media/ipod/ worked for me as root. but I usually do it as a user... mount /dev/sda2 and the fstab entry from before takes care of the rest.
03:27:22***Saving seen data "./dancer.seen"
03:27:42 Join NC-17 [0] (i=NC-17@bb4win/nc-17)
03:27:51v3l0ctmount: wrong fs type, bad option, bad superblock on /dev/sdc2,
03:27:51v3l0ct missing codepage or other error
03:27:51v3l0ct In some cases useful info is found in syslog - try
03:27:51DBUGEnqueued KICK v3l0ct
03:27:51v3l0ct dmesg | tail or so
03:27:51 Part Paul_The_Nerd
03:28:21 Quit erich ("Client exiting")
03:28:21v3l0ctdid i hose the ipod?
03:28:25biffherov3l0ct: you used mount -t vfat /dev/sdc2 /mount/point ?
03:28:40IncYeah you hosed it
03:28:45v3l0ct mount -t vfat /dev/sdc2 /mnt/ipod
03:28:49Mikachuv3l0ct: grep vfat /proc/filesystems
03:29:04biffheroMikachu: good call
03:29:27Mikachuno luck
03:29:47biffherov3l0ct: dmesg | grep sdc | tail -5
03:30:30v3l0ctsdc: Mode Sense: 64 00 00 08
03:30:30v3l0ctsdc: assuming drive cache: write through
03:30:30v3l0ct sdc: sdc1 sdc2
03:30:30v3l0ctsd 8:0:0:0: Attached scsi removable disk sdc
03:30:31***Alert Mode level 1
03:30:31v3l0ctVFS: Can't find a valid FAT filesystem on dev sdc2.
03:30:44v3l0ctsorry for all the multiple line posting
03:31:15Mikachuv3l0ct: are you sure it's a windows ipod?
03:31:32v3l0ctIpod was used on a windows machine but i bought it as a refurbished unit from Apple
03:31:39biffherov3l0ct: you got me. your fdisk output shows /dev/sdc2 * 6 497 3951990 b W95 FAT32
03:31:42v3l0ctmaybe that's the issue
03:32:02v3l0ctbiffhero, i know
03:32:02Mikachuah, you already checked that, sorry
03:32:12Mikachumaybe they are crazy and have it as hfs+ anyway
03:32:26biffheroyeah, maybe that's the issue. maybe even though the partition table lists it as a vfat, the ipod software makes no difference, and mounts both ways. can he mount hfs+ under linux?
03:32:31v3l0cti did get some HFS+ when i did the original dmesg
03:33:03Mikachuif you boot apple firmware, does it say windows or mac?
03:33:23v3l0ctFrom dmesg: hfs: unable to find HFS+ superblock
03:33:36 Quit hardeep ("BitchX: cleaner, drier, protects even better")
03:34:06v3l0ctdoesn't say either, just ipod mini
03:34:21v3l0ctFAT: Filesystem panic (dev sdc2)
03:34:22v3l0ct fat_get_cluster: invalid cluster chain (i_pos 0)
03:34:22v3l0ct File system has been set read-only
03:34:22***Alert Mode level 2
03:34:22v3l0ctFAT: Filesystem panic (dev sdc2)
03:34:22***Alert Mode level 3
03:34:22v3l0ct fat_get_cluster: invalid cluster chain (i_pos 0)
03:34:28v3l0ctgot those from dmesg too
03:34:55Mikachuheh WTF
03:34:55biffheroI bet it tried to automount it on USB insertion, and it failed
03:34:58Mikachui booted apple firmware
03:35:02MikachuAbout >
03:35:04Mikachuthe title says
03:35:10Mikachuthat's pretty awesome
03:35:26Mikachuanyway, check Format at the bottom of that page
03:35:49v3l0ctversion is 1.3
03:36:00biffheroyep, someone else touched your ipod. even had ORGASM with it.
03:36:43v3l0ctwhen i plugged the usb cable it didn't mount, that's correct
03:37:24biffherov3l0ct: how much do you love your current data? got room to back it up?
03:37:34v3l0cton the ipod?
03:37:41v3l0ctall gone
03:37:49v3l0ctit's ok, i have it all on cd
03:37:53biffherohave a windows machine?
03:37:59v3l0cti dual boot
03:38:16v3l0ctwell my wife uses the windows partition
03:38:25biffherothe ipod folx have a 'convert to windows ipod' thing, iirc
03:40:13v3l0ctyeah they have something called 'ipod updater' i think
03:41:12biffheroupdate to windows ipod
03:41:50v3l0ctok. bbl. thanks for the help biffhero and Mikachu.
03:41:56 Quit v3l0ct ("Leaving")
03:44:23***Alert Mode OFF
03:51:51 Join _Lucretia [0] (
03:53:05 Quit _Lucretia_ (Connection timed out)
03:56:32 Join quobl [0] (n=quobl@tor/session/external/x-c7145a8466791083)
04:05:40 Join psi [0] (n=psi@
04:09:21 Quit Daishi ("Client exiting...")
04:10:03 Join Jungti1234 [0] (n=jungti12@
04:18:00 Quit new5guser ("CGI:IRC 0.5.7 (2005/06/19)")
04:21:47 Join _Lucretia_ [0] (
04:24:19 Quit _Lucretia (Read error: 110 (Connection timed out))
04:25:50JdGordonmidkay: u around?
04:31:43 Join Bg3r [0] (n=Bager@
04:32:28 Quit TCK (Read error: 104 (Connection reset by peer))
04:36:17 Join [TCK] [0] (
04:45:22 Join mc_365 [0] (
04:45:32 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
04:56:33 Quit [TCK] (Connection timed out)
04:59:58 Quit Jungti1234 (Read error: 104 (Connection reset by peer))
05:01:30 Quit insulate ("Leaving")
05:01:48midkayJdGordon, yeah?
05:01:54 Join Dyne [0] (
05:03:37 Join Rob2222_ [0] (
05:06:16 Join _Lucretia [0] (
05:09:30 Join scott666|work [0] (
05:09:38 Nick scott666|work is now known as scott666 (
05:09:55 Quit _Lucretia_ (Read error: 110 (Connection timed out))
05:17:05 Quit crwl (Read error: 110 (Connection timed out))
05:18:29 Quit Rob2222 (Read error: 113 (No route to host))
05:23:28 Join quobl_ [0] (n=quobl@tor/session/external/x-079e01206e6b8a89)
05:27:25***Saving seen data "./dancer.seen"
05:34:00 Quit _Lucretia (Connection timed out)
05:36:47 Quit quobl (Read error: 110 (Connection timed out))
05:39:46 Part midgey34
05:51:14 Join gursikh [0] (
05:55:39 Join Oetar [0] (
05:55:53Oetar... I might be an idiot, but I need to ask:
05:56:11OetarHow do I download plugins on the rockbox site?
05:56:29OetarI see fancy pages with plugin info, and they look so tasty, but I can't find a link. :(
05:56:50scott666care to cite an example?
05:57:04OetarPluginBobbles, at
05:57:29scott666have you looked in /.rockbox/rocks/?
05:57:44scott666or Browse Plugins on your player?
05:57:57OetarSo they all come pre-loaded? I feel really, really dumb now. :)
05:58:03OetarCool, thanks :)
05:58:19scott666i was playing bubbles today at wotk, its pretty cool
05:58:53Oetaryeah, it looks cool.
05:59:00scott666im sure ive played it somewhere else, but im not sure where
05:59:10scott666i think it was in an arcade like 10 years ago
05:59:17OetarHeh. I've certainly /never/ played it there :)
05:59:57CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
05:59:57*Oetar wanders off to try out his new rockbox install...
06:02:19 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
06:08:51 Join piroko [0] (
06:09:09pirokoipod... has been running... for an hour and 44 minutes...
06:09:29*piroko is... happy
06:09:33scott666i think thats a new world record
06:10:00pirokoscott666: the 4g ipod used to lock up after 71 minutes
06:10:21 Join TCK [0] (
06:11:04scott666really? that sucks
06:11:12pirokonot anymore! It's fixed!
06:14:21pirokowhere's lostlogic and linuxstb?
06:14:36pirokoand amiconn?
06:14:52 Quit gursikh (Read error: 104 (Connection reset by peer))
06:15:07 Join kehoe [0] (
06:15:21scott666whats up
06:15:37kehoei just recently installed rockbox and so far im loving it
06:15:53scott666what player?
06:15:59kehoeis there any more featrues about the playlist other than like insert and queue that r unique to it
06:17:08pirokothat's not a question for me, unfortunately. have you checked the manual?
06:17:40kehoei looked through it but it was massive, i can look again, thanks for the help though
06:17:55pirokoi'll look with you
06:18:13kehoethank u but u dont need to
06:18:20pirokoi know
06:18:27kehoei was just wondering if u knew off the top of ur head
06:18:33pirokoi have nothing better to do, really :P
06:18:38pirokoi'm curious myself
06:19:02 Join crwl [0] (
06:21:00 Quit amiconn (Read error: 110 (Connection timed out))
06:21:00 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
06:23:25 Join dj-fu [0] (
06:31:16pirokohi amiconn
06:31:35pirokoor maybe that was automatic...
06:31:55pirokoah well whoever fixed the 4g ipod THANKS A BUNCH!!! goodnight
06:32:10 Quit piroko ("telinit 0")
06:32:54DyneWhat was wrong with it?
06:33:57 Quit mikolas (Read error: 104 (Connection reset by peer))
06:34:03scott666apparentlly it crashed after 71 minutes
06:39:06 Quit JdGordon (Read error: 104 (Connection reset by peer))
06:40:43 Join Jungti1234 [0] (n=jungti12@
06:41:46dj-fumiiiikaaaa chuuuuuuuuuuuuuuuu
06:42:00Jungti1234what's meaning of 'Release i slutet av maj 2006'?
06:42:22 Join JdGordon [0] (
06:42:51 Join speacial_ed [0] (
06:42:52dj-fusomething about may 2006
06:42:55dj-fuI don't understand the language, though.
06:43:40dj-furelease is planned for may 2006, possibly?
06:44:12 Quit goffa (Read error: 110 (Connection timed out))
06:44:12 Quit kehoe ("CGI:IRC (EOF)")
06:45:05Jungti1234ok thanks dj-fu
06:45:29 Join goffa [0] (n=goffa@
06:47:09 Quit speacial_ed (Client Quit)
06:47:24 Join JBGood [0] (
06:54:34 Quit RotAtoR ("zzzzzzzzzz")
06:56:20ravonoooo, new mod patch. That will please that french guy.
06:57:45 Quit InteliWasp ("Leaving")
06:58:02 Quit Jungti1234 ()
07:04:36 Quit JBGood25 (Read error: 110 (Connection timed out))
07:05:26Presencelamer question: I have a dynamic playlist going, but I want it cleared and start fresh, skipping or dumping any tunes left in the previous list that hadn't played.
07:06:17Presenceright now, I manage playlist -> view playlist -> delete the song I'm listening to from the list and *somewhere magic has happened* and the playlist borks and is empty, and I rejoice and start adding new songs.
07:07:26 Quit darkless (Client Quit)
07:13:24OetarWow. Rocbox is so totally cool!
07:13:41scott666yes. yes it is.
07:14:20OetarIt would be even cooler, though, if it could get rid of my ant infestation.
07:14:20dj-fucertainly is
07:14:26OetarI have ants on my desk.
07:14:50ravonOetar: I can't promise anything, but I can try to write a plugin for it.
07:15:15scott666Oetar: theres actually a plugin that emits the frequency that causes ants to roll over dead
07:15:51dj-fuoh, I saw that one.
07:15:53dj-funeat stuff ;
07:16:11scott666I wouldn't use it with your dog in the room though..
07:18:02scott666I'm not sure why, but I really like stating misinformation as fact.
07:19:00scott666For example: Did you know that dogs can't look up?
07:19:13 Join Arrogant [0] (
07:19:24dj-fuAddicted to sarcasm, perhaps scott666 ?
07:19:36scott666think about it. have you EVER seen a dog look up?
07:19:37dj-fuIt is known that computer-types tend to become addicted to such things as sarcasm
07:19:40scott666I havnt
07:19:43 Join darkless [0] (n=darkless@
07:19:51dj-fuI certainly have. My fiance's dog looks up all the time
07:20:55scott666that's just hearsay
07:21:01scott666doesnt prove anything
07:21:27dj-fulol :D
07:21:37scorchemy dog looks up at me all the time
07:22:06scott666i take it no one present has seen Shawn of the Dead
07:22:16scorchei have :)
07:22:56scott666'yeah, well Big Eddy also says dogs can't look up' 'THEY CAN'T!'
07:25:12 Join [TCK] [0] (
07:27:28***Saving seen data "./dancer.seen"
07:30:49 Quit arf-arf ()
07:31:38 Quit dj-fu ("Leaving")
07:32:31 Join Rondom [0] (
07:32:38 Quit macdonalder ("bye bye ;D")
07:33:47 Join dj-fu [0] (
07:39:14 Quit TCK (Read error: 110 (Connection timed out))
07:40:46 Join shadou [0] (
07:43:14 Quit Fitzsimmons ()
07:44:03 Quit Oetar (Read error: 110 (Connection timed out))
07:44:03 Join gursikh [0] (
07:44:11 Quit dj-fu (Read error: 104 (Connection reset by peer))
07:44:13 Quit gursikh (Remote closed the connection)
07:44:22 Nick shadou is now known as dj-fu (
07:59:59 Quit dj-fu ("Leaving")
08:00:22 Quit perldiver (Read error: 110 (Connection timed out))
08:04:04 Join psii [0] (n=psi@
08:04:32 Join dj-fu [0] (
08:07:00 Quit Rondom (Read error: 113 (No route to host))
08:08:09 Quit adiamas ("Chatzilla 0.9.72 [Firefox]")
08:13:02 Quit dj-fu ("Leaving")
08:17:44 Join arf-arf [0] (
08:21:01midkayJdGordon. :(
08:21:13JdGordonwhy the sad face?
08:22:32midkayyou wanted me earlier?
08:23:12 Join RoC_MM [0] (
08:23:27JdGordondid u end up looking at the viewer patch?
08:27:00midkayJdGordon, yeah, i'm quite sure i'll have it in before 3.0, i've just been kind of staying away from it because i'm worried about having to deal with the saving/loading of settings :) (not specific to your patch)
08:27:13JdGordonah ok
08:33:11 Join PaulJam [0] (
08:33:29 Quit psi (Read error: 110 (Connection timed out))
08:33:29 Nick psii is now known as psi (n=psi@
08:35:28RoC_MMany Nano Rockbox users in?
08:36:10RoC_MMHow is it?
08:36:23RoC_MMAnd what is the most recent build that is completely stable for you?
08:37:41RoC_MM(going to install on friend's Nano, and need to have most stable rockbox build to provide)
08:38:48 Part YouCeyE ("Leaving")
08:38:57Galoishuh? the current cvs seems pretty decent.
08:39:10GaloisI haven't tried any fancy features like voice menus or fast forward yet :)
08:39:17 Quit Arrogant ("Leaving")
08:39:54RoC_MMfast forward is quite needed
08:40:55RoC_MMI've tried a couple April builds on my 4G, but usually these have bugs, or at least they did until the 15th when I stopped trying them...I'm currently running a March 30 you remember how stable the end of March builds were?
08:41:01Galoiswell I just fast forwarded/rewinded an ogg file without trouble
08:41:36Galoisvoice menu users swear by the march 23 build, iirc
08:41:43RoC_MMAny rough edges on the standard play/seek/bookmark/crossfade/browse sort of operations?
08:42:09Galoisug you want me to test bookmark/crossfade???
08:42:29RoC_MMBookmark not important
08:42:40RoC_MMIt would be helpful to know if crossfade works
08:42:44RoC_MMand what your build ID is
08:43:07Galoisthe bookmark looks broken
08:43:19RoC_MMSince Rockbox hasn't been released for the iPod target yet, I'm trying to find a "known good" build.
08:43:21Galoisif you select a bookmark it starts from the beginning of the file
08:44:50Galoishm I just crashed it by selecting a song with crossfade on
08:45:01Galoiswell, crashed the UI at least, the music is still playing
08:45:39Galoisthe last known good daily for me was april 6, I think
08:45:59RoC_MMHow certain?
08:46:11RoC_MMThanks for the help :-)
08:46:25GaloisI'm completely certain of the date, I'm not certain how good the build is :P
08:46:35Galoisbecause I normally don't use crossfade, or voice menus, or even seeking
08:47:50RoC_MMI suppose a end of March build might work if APril 6 does not... it was right before the freeze and code churn
08:48:11GaloisI thought a freeze was supposed to prevent code churn?
08:48:23RoC_MMDoesn't look like it from here.
08:48:35RoC_MMbut yes
08:49:19linuxstbLooks like Samsung have won the iPod CPU contract from Apple:
08:49:41linuxstbSo at least they are sticking with ARMs.
08:50:41 Join Acksaw [0] (
08:51:13Galoisyou're sure samsung only makes arms?
08:51:31linuxstbNo, but the news articles say that it will be ARM based.
08:52:15linuxstb"Samsung?s chip is based on 32-bit processor technology from ARM Holdings plc."
08:52:53Galoissounds good
08:56:56nudelweird thing is i think samsung also licence sigmatel chips
08:57:29nudelah the article says that :)
08:57:45Galoisrumours are just that.
09:03:50linuxstbInteresting side-effect of the distributed build system that someone mentioned on the forums - builds are now made in random time zones, so the file timestamps and the version info depends on the build server...
09:04:18linuxstbMaybe an "export TZ=Europe/Stockholm" in the build scripts could be useful.
09:04:22 Quit Hotfusion (Read error: 104 (Connection reset by peer))
09:06:05JdGordonmeh.. does it really matter?
09:06:31linuxstbIt's obviously not a huge problem, but if it can be fixed easily, then why not?
09:07:00JdGordonwould that change the timezone just for hte script?? coz if not it would be painful
09:07:04linuxstbAt least the date/time in the version info is useful when people report bugs.
09:07:16linuxstbJdGordon: Yes, it just sets the timezone for that script.
09:09:08 Join B4gder [0] (
09:10:30Presence the builds that are compiled in the future are SO MUCH BETTARS!
09:14:33 Join wehn [0] (
09:17:16 Quit Dyne (Read error: 104 (Connection reset by peer))
09:18:24 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
09:23:05 Quit RoC_MM ("Leaving")
09:25:15B4gderthe forums certainly are getting busy these days
09:26:11*amiconn is getting closer to what actually goes wrong in sh-elf-gcc jumptable generation...
09:26:23 Quit Acksaw (Read error: 104 (Connection reset by peer))
09:26:52 Join Acksaw [0] (
09:27:21B4gder188 posts/day in average during April
09:27:28B4gder110/day in March
09:27:30***Saving seen data "./dancer.seen"
09:27:41*amiconn stopped reading the forums regularly
09:27:46B4gder95/day in Feb
09:28:55amiconnB4gder: Did you see linuxstb's findings regarding distributed builds & timestamps?
09:29:10B4gderI replied to it just now in the forum
09:29:26B4gderthat timestamp is the one displayed within Rockbox, isn
09:29:30B4gder't it?
09:29:58B4gderslightly amusing I think ;-)
09:31:00amiconnIt is
09:31:14B4gderwe need to switch to UTC time
09:31:22amiconn(and also the timestamps of the files)
09:31:57amiconnEither UTC, or a suggested by linuxstb, Europe/Stockholm
09:32:16B4gderI think using Europe/Stockholm is a lot trickier
09:32:35B4gderand possibly more confusing to users
09:32:40linuxstbB4gder: Isn't it as simple as "export TZ=Europe/Stockholm" ?
09:32:41amiconnCheck the log starting 09:05
09:33:07B4gderright not trickier really, as you can just pass on the timezone in the time string
09:34:14linuxstbThe website seems to use GMT for timestamps, so the build system should probably do the same.
09:34:30B4gderwe try to use GMT all over
09:34:42B4gderin the CVS build table too etc
09:35:21 Quit windowsrefund (Read error: 104 (Connection reset by peer))
09:35:33 Join windowsrefund [0] (
09:39:13 Join JdGordon41 [0] (
09:39:32 Join Paul_The_Ner1 [0] (
09:39:45 Part Paul_The_Ner1
09:40:17 Join aegray_ [0] (
09:40:27 Quit Seed (Nick collision from services.)
09:41:29 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
09:42:27 Join BHSPitLappy2 [0] (
09:43:17 Quit kapheine (Remote closed the connection)
09:43:55 Quit HCl (
09:43:55 Quit mbr (
09:43:55 Quit BoD[away] (
09:43:55 Quit kackerlac (
09:43:55 Quit Galois (
09:43:55 Quit PiXEL8 (
09:43:55 Quit eaz (
09:43:55 Quit Ave (
09:43:55 Quit humulus (
09:43:55 Quit rob- (
09:43:55 Quit dark (
09:43:55 Quit steveb (
09:43:55 Quit markun (
09:43:55 Quit Ironhand (
09:43:55 Quit aegray_ (
09:43:55 Quit JdGordon41 (
09:43:55 Quit wehn (
09:43:55 Quit midkay (
09:43:55 Quit lostnihilist (
09:43:55 Quit linuxstb (
09:43:55 Quit solexx_ (
09:43:55 Quit Lynx_ (
09:43:55 Quit dpassen1 (
09:43:55 Quit Mikachu (
09:43:55 Quit webmind (
09:43:55 Quit Ribs (
09:43:55 Quit darkless (
09:43:55 Quit merbanan (
09:43:55 Quit Inc (
09:43:55 Quit mooch (
09:43:55 Quit pabs (
09:43:55 Quit pill (
09:43:55 Quit lalmeras (
09:43:55 Quit Hadaka (
09:43:55 Quit slarti (
09:43:55 Quit Slasheri (
09:43:55 Quit preglow (
09:44:08 Join jnc_ [0] (n=jnc@
09:44:11 Quit jnc (Connection reset by peer)
09:44:32 Quit FutureCrash ("changing servers")
09:45:01 Quit JdGordon (Nick collision)
09:45:06 Join scott666_ [0] (
09:45:06 Join Seed [0] (
09:45:06NJoinaegray_ [0] (
09:45:06 Join JdGordon [0] (
09:45:06NJoinwehn [0] (
09:45:06NJoindarkless [0] (n=darkless@
09:45:06NJoinmidkay [0] (n=midkay@rockbox/developer/midkay)
09:45:06NJoinlostnihilist [0] (
09:45:06NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:45:06NJoinRibs [0] (
09:45:06NJoinsolexx_ [0] (
09:45:06NJoinLynx_ [0] (
09:45:06NJoinInc [0] (n=inc@arcane/developer/inc)
09:45:06NJoinmbr [0] (
09:45:06NJoinrob- [0] (
09:45:06NJoinBoD[away] [0] (
09:45:06NJoinPiXEL8 [0] (
09:45:06NJoinkackerlac [0] (
09:45:06NJoindark [0] (i=deviled_@unaffiliated/darkx)
09:45:06NJoinsteveb [0] (n=steve@about/cooking/nakedchef/risotto/steveb)
09:45:06NJoinAve [0] (
09:45:06NJoinhumulus [0] (
09:45:06NJoineaz [0] (
09:45:06NJoinGalois [0] (
09:45:06NJoinHCl [0] (i=hcl@rockbox/developer/HCl)
09:45:06NJoinmerbanan [0] (
09:45:06NJoindpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
09:45:06NJoinmarkun [0] (
09:45:06NJoinMikachu [0] (
09:45:06NJoinIronhand [0] (
09:45:06NJoinmooch [0] (
09:45:06NJoinpabs [0] (
09:45:06NJoinpill [0] (
09:45:06NJoinwebmind [0] (
09:45:06NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
09:45:06NJoinHadaka [0] (
09:45:06NJoinlalmeras [0] (
09:45:06 Join preglow [0] (
09:45:06NJoinslarti [0] (n=tmartin@gentoo/developer/slarti)
09:45:08***Server message 505: 'logbot- :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
09:45:21JdGordoneveryone have a nice trip?
09:45:54 Join mikearthur_ [0] (
09:45:56scorchei didnt go anywhere
09:46:22wehnwhat was that?
09:46:25 Join TiMiD_ [0] (n=TiMiD[
09:46:47scorchewehn: ah....a new-type-person to irc i see
09:46:56scorchethat is something we like to call a "netsplit"
09:47:12 Join t0mas_ [0] (
09:47:15 Join crash___ [0] (
09:47:19wehni've seen it a few times, don't know what causes it. transer to new server?
09:47:25B4gderI call it banana split
09:47:25 Quit ze (Remote closed the connection)
09:47:37JdGordonsome1 over at the freenode server rooms tripped over a cable and nocked off a server or 3 :D
09:47:44B4gderwehn: no, disconnected servers
09:47:45 Join novimon_ [0] (
09:48:20wehnah. happens awefully regularly.
09:48:29 Join ze [0] (
09:48:29 Quit goffa (Read error: 104 (Connection reset by peer))
09:48:30scorchewehn: basically, a node is disconnected
09:48:35 Join gromit`` [0] (
09:48:54scorcheB4gder: node!...not servers i say!
09:49:08 Quit `3nergy ("changing servers")
09:49:23 Join `3nergy [0] (
09:49:38B4gderthe thing, different name
09:49:40B4gdersame thing
09:49:57 Join _slimeball [0] (
09:50:00scorchewell, i have to have something to argue about to keep me busy
09:50:24B4gderemacs is better than vi!
09:50:32B4gderLinux is cooler than windows
09:50:36*B4gder adds a few
09:50:43scorche<3 vim
09:50:52crashdnegative, vim > emacs
09:50:53scorcheand on the second, i agree
09:50:57crashdbut i'd rather use pico or nano anyday
09:50:59wehnnano! or zile!
09:51:04crashdvim is so overrated
09:51:07B4gderlook, it worked
09:51:07scorchevim wins!
09:51:19crashdB4gder: ;)
09:51:29scorchebut yes....i wouls rather use pico myself
09:51:32JdGordonvim is the shit!
09:51:38JdGordonlinux tho is just shit....
09:51:50 Join TeaSea [0] (
09:52:16scorchehavent actually tried nano myself
09:52:22 Quit Zagor (Read error: 110 (Connection timed out))
09:52:40 Quit gromit` (Connection reset by peer)
09:52:45 Quit _slimeball (Remote closed the connection)
09:52:50 Quit sharpe (Read error: 104 (Connection reset by peer))
09:53:06 Join kapheine [0] (
09:53:19 Quit crash__ (Connection reset by peer)
09:53:34 Join _slimeball [0] (
09:54:06 Quit novimon (Read error: 104 (Connection reset by peer))
09:54:27 Quit aegray (Connection timed out)
09:55:17 Quit mikearthur (Connection reset by peer)
09:55:18 Join myth1 [0] (
09:55:20midkayalright. alright. i have a _really_ good joke/riddle/WHATEVER. alright. listen up.
09:55:22myth1hi guys
09:55:30midkaywhat do you call coding for Rockbox?
09:55:41scorchei tried
09:55:50*scorche hugs midkay
09:55:53 Quit _slimeball (Remote closed the connection)
09:55:58midkayROCKet science!!
09:56:05 Join dwihno_ [0] (n=dw@
09:56:07crashdthat was awful.
09:56:15 Join FutureCrash [0] (
09:56:19midkayshut the .. i'm going back to bed. :(
09:56:23crashd: ) sorry dude
09:56:24crashdit was tho
09:56:34scorchecrashd: no it was "_really_ good"
09:56:36 Join _slimeball [0] (
09:56:44midkayfu all.
09:56:44crashdrockeally good!
09:56:47myth1i need more mem for my codecs, badly
09:56:50midkaybunch of queers don't appreciate a good joke.
09:56:53 Join qwx [0] (
09:56:54 Quit TiMiD (Read error: 104 (Connection reset by peer))
09:57:05 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
09:57:14B4gdermyth1: codecs normally have quite a lot of mem...
09:57:17 Quit scott666 (Connection timed out)
09:57:25 Quit dwihno (Read error: 110 (Connection timed out))
09:57:29*midkay is running back to bed. nite. :)
09:57:33crashdnn midkay
09:57:37myth1B4gder: I can only allocate about 512kb statically
09:58:06myth1B4gder: Do you know how much mem is available for rb->malloc()?
09:58:07B4gderyes, but why would a codec need more statically?
09:58:27myth1I've heard malloc is more or less a hack
09:58:32 Nick _slimeball is now known as jamezz (
09:58:38B4gderit is
09:58:38myth1So how much RAM can I get?
09:59:04 Nick qwx is now known as qwm (
09:59:53myth11 - 2 megs would be fine, is thi possible with malloc? It isn't with statical declaration
10:00:06 Quit BHSPitLappy (Connection timed out)
10:00:06 Join goffa [0] (n=goffa@
10:00:07 Quit amiconn (Success)
10:00:07 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
10:00:26 Join Spacksaw [0] (
10:00:38scorcheB4gder: I've heard rockbox is more or less a hack
10:01:10B4gderI heard it sounds like a bag of shit!
10:01:29 Quit t0mas (Connection timed out)
10:02:07 Quit slimeball (Nick collision from services.)
10:02:12 Nick jamezz is now known as slimeball (
10:02:20B4gderI noticed the guy who said that particular thing felt that he was flamed afterwards
10:02:49 Quit Acksaw (Connection timed out)
10:03:20scorchemeh....i like to create tension.....but i only do cause people dont realize i joke a lot to get reactions out of people
10:04:00crashdscorche: you should go to the doctors about that
10:04:25scorchecrashd: i did....they gave me some pills, but i never take them
10:04:34 Join lostlogi1 [0] (
10:05:02myth1hi lostlogic
10:05:22scorchemyth1: SHH!
10:05:24myth1lostlogic: Can you enlighten me now much ram I can allocate in codecsa?
10:05:30 Quit NC-17 (Read error: 104 (Connection reset by peer))
10:06:37 Quit [TCK] (Read error: 110 (Connection timed out))
10:07:54linuxstbmyth1: The codec itself can be up to 512KB (code and data), plus there is a 512KB malloc buffer, but we want to get rid of that malloc buffer.
10:08:08linuxstbIf you need more memory than that, you will somehow need to use the audio buffer.
10:08:24linuxstbBut your codec will be the first to do it.
10:09:21myth1how can I use the audio buffer for this?
10:09:46linuxstbSomeone will need to invent a way, and implement it.
10:10:01myth1Currently I have the mod codec with 512kbyte and this not very much. I'm thinking about doing XM, but 512kb with xm would be far too less
10:10:13linuxstbWhat do you need the memory for? Is it to hold the mod file, or for working memory?
10:10:32myth1ok, then this is my official request: Someone please invent a way for more memory and I will do a working xm player
10:10:37myth1I need it for the samples
10:10:44myth1and for the pattern data of course
10:10:51linuxstbWhere do these come from?
10:11:00myth1the samples come from the file
10:11:06crwlwouldn't it be easier to get DUMB working than to write your own XM player (and later perhaps IT, which is even more difficult...) :)
10:11:10linuxstbIn which case, they are already in RAM in the audio buffer.
10:11:25linuxstb(assuming the whole file is buffered)
10:11:37myth1mmh, this is an interesting idea
10:11:54myth1doing an own xm player is no big deal, it works on my intel machine, I just have to port it
10:12:07myth1same as for the mod player which now works quite nice
10:12:26linuxstbYou should try and avoid copying the contents of the audio buffer to your own buffer if you can.
10:13:14myth1how can I directly access the audio buffer?
10:13:49myth1lets say I want a pointer to the complete 1MB file, how do I get it?
10:14:50linuxstbYou use the ci->request_buffer() function to get a pointer to your file. But you may have to call this twice if the file is at the wraparound point in the audio buffer (it's a ring buffer).
10:15:18linuxstbBut calling it twice won't work....
10:16:23myth1so what will work?
10:16:29linuxstbSo I think we either need to make sure that .mod files are always stored in a contiguous block of RAM in the audio buffer, or add a new API function which will potentially return two pointers.
10:17:01 Quit lostlogic (Read error: 110 (Connection timed out))
10:17:22myth1two pointers would slow down the mixing-routine, i would not recomend this. a flat buffer is highly recommended
10:18:12myth1if you could make sure .mod files are always stored in a contiguous block, i can continue working on the tracker routines
10:19:56linuxstbI'm not sure if storing some files as a contiguous block is desirable behaviour though. I'm curious to know what others think.
10:20:07 Join ashridah [0] (
10:20:28B4gderI figure there will be some codecs that would prefer that
10:21:12B4gderbut I'm not the man to judge the impact
10:21:39myth1if this is only for some files the case impact shouldn't be too large
10:21:58 Join RedBreva [0] (
10:22:04B4gderwell, it will require some new intelligence in the music data loader system
10:22:07linuxstbThe main impact IMO is how much this would complicate the buffering code. But I don't think the RAM impact will be that great.
10:22:33B4gder"oh this type of music needs to be loaded into one chunk so we need to take these actions..."
10:23:11B4gderfun like when you want to load N megabytes into a single chunk and there's only N-1 available
10:23:27B4gderand it is at the end of ram
10:23:44linuxstbmyth1: If I was you, I would continue developing the codec, and assume that your mod file is in a contiguous block. So you can just return an error if ci->request_buffer() doesn't give you the entire file.
10:24:50linuxstbBut lostlogic or Slasheri are probably the people to judge the impact of this change.
10:27:09myth1if I implement the codec with ci->request_buffer() and assume i get the whole mod file I guess some modules will randomly not play, is this correct?
10:27:24myth1at least with the current behaviour when the ringbuffer is at its end
10:28:28linuxstbYes - that's the problem.
10:28:56myth1is there a chance to reload the file so it won't use the "ringpart" of the buffer?
10:29:12linuxstbI don't know if lostlogic is still around to comment - he said he can't do any more Rockbox coding until next Tuesday.
10:29:35linuxstbNo, I don't think so.
10:30:14myth1honestly, I do't want to publish a codec which randomly skips playing files ;)
10:31:11myth1so I guess for now the 512kb have to suffice until we have a solution
10:31:14linuxstbWell, it's easily documented. The only other solution which would work today would be to change it to be a plugin. You can then claim the entire audio buffer.
10:32:07JdGordondoes any1 know how to setup samba so my windows boxes can connect without any passwords?
10:32:48myth1if we can not find a solution for the ringbuffer I will think about coding a plugin, however direct playback is more what I would prefer
10:33:09B4gderJdGordon: you can set passwd level required somewhere, I don't remember exactly but I did it years ago
10:33:22linuxstbI'm sure we can find a solution - it's just that the people that know about that part of Rockbox aren't around at the moment.
10:33:44myth1should be open a task in the tracker for this?
10:34:04myth1so it doesn't get lost?
10:35:41linuxstbYou could try posting your problem to the rockbox-dev mailing list - lostlogic and Slasheri will hopefully see it there if they miss this conversation in the logs.
10:36:16myth1how can I join this mailing list?
10:36:17linuxstbIMO, it's more of an issue for discussion, than a bug/feature request.
10:37:20myth1ah, its mentioned at the rockbox page
10:38:14 Join phaedrus961 [0] (
10:38:26myth1if you have time it would be nice if you could post it as you have more internal understanding of the ring buffer
10:38:35 Join Paul_The_Nerd [0] (
10:39:52linuxstbI'm about to leave home now, but if I don't see a post from you, I'll try and remember to do it during the day today.
10:40:15myth1ok, thanks
10:40:20myth1btw: Where are you from?
10:46:35*B4gder recalls linuxstb being a Londoner
10:47:09crashdpoor bastard ;)
10:47:34crashdill be joining the fray in a few month tho, so who am i to talk ;\
10:48:26myth1ok buys, be seeing each other
10:48:28 Quit myth1 ("leaving")
10:49:34 Join ender` [0] (i=ychat@
10:51:21Paul_The_Nerdamiconn: You around?
10:53:39 Join muesli__ [0] (n=muesli_t@
10:55:58 Part Paul_The_Nerd
10:58:54 Join Paul_The_Ner1 [0] (
10:59:14markunIs the backlight fading we do logarithmic?
11:01:07SlasheriHmm, that kind of contiguous ram loading should be possible, but i am not sure how easy it's to implement regarding the latest changes to the engine. But it shouldn't be hard
11:10:52wehnYay! the H300 backlight flash was fixed, thanks Petur!
11:13:46 Join webguest50 [0] (
11:13:52webguest50hey people
11:13:57webguest50i need help please
11:14:16Paul_The_Ner1webguest50: Where are you experiencing problems?
11:14:38webguest50dealing with rockbox for my ipod photo
11:14:53Paul_The_Ner1I meant specifically, what problem have you encountered?
11:16:17webguest50i get a menu containing calenders, contacts, notes, photoes, and rockerbox, but no music, and when i hover over rockerbox then click the next button it flashes the word loading"and does nothing
11:16:56Paul_The_Ner1Okay, you're not seeing a menu
11:16:59Paul_The_Ner1You're seeing the filetree.
11:17:14Paul_The_Ner1That file is the "rockbox.ipod" you installed, but it's already running at this point which is why clicking on it does nothing.
11:17:22wehnOr maybe he's still in AppleOS?
11:17:27Paul_The_Ner1wehn: No, he's definitely not
11:17:54wehnsorry, not familiar with whats on the iPod HD.
11:17:55webguest50ok, how can i get to rockbox then?
11:18:03Paul_The_Ner1webguest50: I just told you, it's already running.
11:18:12Paul_The_Ner1webguest50: I'm guessing you mean "How do I play my music?"
11:18:18webguest50then how can i access my music ?
11:18:50webguest50are you there man?
11:18:53Paul_The_Ner1webguest50: There are two solutions here. One is that you can use our database, TagCache, to index the music that iTunes has hidden on your iPod. The other is to clear the iTunes music off, copy over your files into a reasonable directory structure, and browse using the filetree.
11:19:11Paul_The_Ner1Which would you prefer?
11:19:45Paul_The_Ner1The Database method will give you "Artist, Album, Genre, All Songs" like you're familiar with.
11:19:58Paul_The_Ner1The filetree lets you organize your music however you want, just by where you put it on the disk.
11:20:15webguest50cool then the second, man, thanks for being so helpful
11:20:17 Nick Paul_The_Ner1 is now known as Paul_The_Nerd (
11:20:48Paul_The_NerdWell in that case you simply want to use iTunes to remove the music from your iPod, then just copy the music over in disk mode wherever you want to place it. Just drag and drop MP3 files or folders to it.
11:21:08Paul_The_NerdWhen you're in Rockbox, you're actually seeing the layout of your disk, so you'll be able to browse into those folders, and load whole folders or individual songs to play.
11:21:09webguest50thanks man
11:21:34Paul_The_NerdIf you want to try the database, instructions are here: it can be used alongside filetree mode as well.
11:26:13webguest50that's really good
11:26:25webguest50how to switch between rockbox and apple firmware?
11:27:33***Saving seen data "./dancer.seen"
11:29:14Paul_The_NerdTurn off Rockbox by holding play/pause.
11:29:29Paul_The_NerdWhen you press Menu to turn on your iPod, just hold it down instead of just tapping it.
11:29:50Paul_The_NerdOnce it says "Loading original firmware" you can let go of menu.
11:29:58 Quit darkless (Client Quit)
11:31:06Paul_The_Nerdlinuxstb: "freezing issue previously attributed to cpu scaling has gone." <−−- Any idea at all what change recently might have magically done this?
11:31:21dwihno_How is the rockbox charging compared to the ipod stock fw?
11:31:45Paul_The_Nerddwihno_: I think the actual charging is hardware controlled. (I may be mistaken.)
11:33:18wehnPlayback in the current daily works superbly everyone, i'm impressed.
11:33:35wehnSmooth seeking and quicker track changes. :)
11:33:58muesli__just seen a software that let u convert protected apple songs into non-protected
11:33:59dwihno_Paul_The_Nerd: hm, strange...
11:34:09muesli__for anyone who didnt know before..
11:34:20dwihno_Paul_The_Nerd: Might be related to rockbox playing until the battery is completely drained then.
11:34:24crashdhmm, i just got some really weird behaviour with on-the-fly playlists
11:34:41Paul_The_Nerddwihno_: What might be related to that?
11:35:00Paul_The_Nerdcrashd: What sort of behaviour?
11:35:12crashdwell, i was making a playlist using the insert>into playlist
11:35:27 Quit webguest50 ("CGI:IRC")
11:35:35crashdwhich ive done loads before, only this time, when it got to about 20 seconds into the second track, it skipped about 6 1/2 tracks and startedc playing the middle of another track
11:35:38crashdvery strange
11:35:40crashda quick tap of << fixed it
11:35:49dwihno_Paul_The_Nerd: Well, rockbox will die due to low battery... Need to load apple fw to charge for a while
11:35:58Paul_The_Nerdcrashd: How new/old is your build?
11:36:12crashdPaul_The_Nerd: it's a cvs build from yesterday with SID and the libmad patches
11:36:32crashdtbh, it might well have been fixed by one of brandon's commits from yesterday
11:36:38crashdonly, he's not here to hassle about it ;)
11:36:43Paul_The_NerdThere were a lot of commits yesterday yes.
11:36:49Paul_The_NerdAnd a bug like that has been reported before.
11:37:13Paul_The_NerdBut I don't think I've heard anyone mention it again in the last 24 hours, so I don't know if it's been fixed, or if they just haven't stumbled across it again yet.
11:37:37crashdwell, as i say, this is a cvs build+patches from yesterday
11:37:45crashdso it could be that the stuff that got commited fixed it
11:37:54Paul_The_Nerddwihno_: Oh, so you're just saying that Rockbox doesn't really work with *extremely* low battery, and you need to charge in Retail a bit first? That doesn't really surprise me, I think.
11:38:09dwihno_Paul_The_Nerd: Doesn't surprise me either...
11:38:29dwihno_Paul_The_Nerd: On the archos, rockbox won't play if the battery is beneath some treshold
11:39:05Paul_The_NerdThere's quite a bit of battery-related tuning Rockbox on Ipod needs.
11:39:54wehnI think rockbox on the H300 discharges deeper than the iriver software does.
11:40:25wehnIt'll boot into rockbox while iriver firmware refuses to boot.
11:42:19dwihno_Paul_The_Nerd: true, true
11:49:25PaulJamwehn: according to my observations the iriver firmware shuts down the device at ~3.65V , at this point rockbox shows 10-18% battery left
11:50:50 Join aziiz [0] (
11:52:20wehnPaulJam: Didn't realise it was such a big difference. I wonder if rockbox goes below recomendend discharge depth?
11:54:46PaulJami could imagine that the original firmware shuts down earlier, because below 3.65V the display flickers at disk activity
11:56:15 Quit wehn (Remote closed the connection)
11:57:16 Join wehn [0] (
12:01:22 Join bluebrother [0] (
12:01:44wehnI used to notice the display flickering in iriver firmware but didn't associate it with battery discharge
12:03:22wehnlithium ion battery life seems to be linked to discharge depth (along with other things):
12:03:51wehnBut I didn't notice much difference to runtime after 4 consecutime 100% till shutoff cycles.
12:04:56wehnmaybe 4 minutes less out of nearly 9 hours. exactly same WPS, songs on repeat etc.
12:09:02 Part Paul_The_Nerd
12:12:18 Quit bluebrother ("leaving")
12:14:11PaulJamif i understand the article correctly, the battery would not work at all annymore if rockbox would discharge it below the recommended level
12:14:30 Join ark_112 [0] (
12:14:36ark_112hey all
12:14:53ark_112anyone know anything about development on the ipod 3 gen port?
12:15:05B4gderark_112: yes, it has stalled
12:15:12ark_112o ok :(
12:15:15ark_112do u know y?
12:15:21ark_112major problems?
12:15:25B4gderpretty simply: no one works on it
12:15:42lostlogi1PaulJam: there's an emergency disable circuit in most LiION batteries, but that's not till 2.7 volts I believe. All DAPs I'm aware of have hardware cutoffs in the power circuitry at 2.9v so they don't get down there.
12:15:46ark_112bumma, wat6 langauge is it written in?
12:15:52B4gderark_112: C
12:16:13 Join Paul_The_Nerd [0] (
12:16:19B4gderark_112: the single guy that has done the 3 gen work so far has been busy doing other stuff and no one else has stepped up to do it
12:16:23ark_112does anyonw hav the source for the last release?
12:16:32B4gderark_112: ask cvs
12:17:12B4gderthat's how you get the latest sources
12:18:45lostlogi1wehn: it's true that Lithium based batteries do not like to hold at full or empty for extended periods, deep discharge and full charge do not generally do Bad Things (TM) to the battery though unless the battery is held at that charge level afterwards (or worse allowed to self discharge lower).
12:19:04 Join ProgramZeta [0] (
12:19:06 Join _FireFly_ [0] (
12:21:40lostlogi1B4gder: any idea how we might get UTC timestamps from all build servers? ntp might be a reasonable requirement, although the exact parameters to require are hard to determine (My server's clock is currently within 2ms of 4 stratum 2 servers, but I may be more obsessive than most)
12:22:00 Join Poka64 [0] (
12:22:22B4gderwell, firstly we could make the build just use UTC times for the time stamp
12:22:29B4gderinstead of local time as it does now
12:22:59B4gderif we wanted the times to be _exact_ the same on all builds we could use a time stamp passed on from the build server
12:23:09 Quit ark_112 ("CGI:IRC (Ping timeout)")
12:23:11B4gderI mean from the master
12:23:24B4gderit already kind of passes on the timestamp
12:23:31B4gderas it is used for the cvs update command
12:23:43B4gderto make sure all servers update to the exact same state
12:23:48wehnlostlogi1: I think what the article (and others I read) said was battery lifespan depended on number and depth of discharges
12:24:26wehnlostlogi1: ie x full discharges or 2x half discharges = battery lifespan
12:24:44wehnso more frequent less deep charges are better
12:24:55lostlogi1I think one of two things makes sense: if we're not going to care about precision, then we should just use the CVS timestamp as the build's stamp, or if we want to precisely know when the actual build happened then we should require NTP and use UTC timestamps from the servers.
12:24:58wehnBut that said, I'm happy with rockboxes current discharge cutoff
12:25:12Paul_The_NerdIt's theoretically possible that two *different* builds could have the same time stamp with just UTC if a slow server's a few minutes off in one way, a fast one in the other, and they're two builds in a row. That's probably a bad thing.
12:26:03wehnPaul_The_Nerd: perhaps NTP would help there.
12:26:07Paul_The_NerdIt would.
12:26:09lostlogi1wehn: If I understand correctly, a lithium ion battery constantly cycled in the 40-80% charge range will last push a lot more useable current out it's leads than one constantly cycled 0-100%
12:26:29 Join actionshrimp [0] (
12:26:45lostlogi1the chemistry doesn't like the extremes of charge
12:26:56Paul_The_NerdI vote for the timestamp though, since you've already got it around and since it's then also reflects exactly which CVS co a user could use to get the same sources as the build, right?
12:26:56B4gdersince the builds are started by the master, it really can't have the same time stamp on two builds
12:27:00 Nick lostlogi1 is now known as lostlogic (
12:27:25JdGordoncant u just change the filestamps when the zips come back to the main server?
12:27:38Paul_The_NerdJdGordon: The one that shows in the credits plugin.
12:27:52wehnlostlogi1: yep, i agree. (though not sure on the 80% bit.)
12:28:06B4gderPaul_The_Nerd: yes the match with the cvs version is a very compelling reason
12:28:34B4gderI'll bring out my stick and poke it tonight
12:29:08wehnlostlogic: maybe in poor chargers that force into overcharge 80% would be better. anyway.
12:29:56 Join tvelocity [0] (
12:32:34PaulJamwehn: i think the charging is controlled by an internal chip and not by the external charger. And i think the user shoud be able decide how he wants to treat the battery.
12:34:26PaulJamwhat i would like to see in rockbox is a beep every few minutes if the voltage falls below a certain (user configurable) level
12:35:45 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
12:51:53wehnPaulJam: yep, of course the power adaptor only supplies power. And I agree the user controls charging
12:52:40wehnPaulJam: I like the current cut off level since I haven't seen the capacity loss I've come accross in other products from an uncontrolled deep discharge
12:55:00wehnnot sure about battery warnings, if your running that low it's probably because you can't get to a charger. :)
12:57:10PaulJamwehn: of course, the warning should be optional, but i often forget to connect the charger at home. so i would prefer a warning because i dont want to cischarge the battery too low
13:03:47 Join Jungti1234 [0] (n=jungti12@
13:06:41 Quit Ribs ("Leaving")
13:16:51XavierGrmuesli__: are you here?
13:18:50 Join Ribs [0] (
13:22:00 Quit RedBreva ("CGI:IRC")
13:23:15 Join webguest65 [0] (
13:23:54webguest65is there any way that I can display the battery voltage in root, without going to wps ?
13:24:31JdGordonturn on the status bar
13:25:10webguest65hmm, batt voltage in status bar....
13:26:03_FireFly_afaik there is no voltage display in the statusbar only a % value
13:26:09_FireFly_or graphical
13:26:17Paul_The_NerdI didn't even know there was an actual voltage display in WPS. Is there?
13:26:31webguest65yes was looking for voltage display specifically, more meaningful
13:27:06 Join perldiver [0] (
13:27:24Paul_The_NerdBut you said "without going to WPS" which somewhat implied you meant something that was currently available in WPS.
13:27:31Paul_The_NerdYou can check the battery status in the debug menu though
13:27:37***Saving seen data "./dancer.seen"
13:27:43JdGordonu can put voltage on the wps..
13:28:30webguest65hmm, if I'm doing file work and its connected to the pc, it would be nice to have voltage displayed in the usb logo, or something like that
13:28:32Paul_The_NerdOh, %bv, gotcha
13:28:45amiconnwehn: It's true that LiIon does like full charge less than a medium charge state
13:29:06amiconnI.e. the extremes are bad, as already mentioned
13:29:24Paul_The_Nerdwebguest65: Why not just make sure it's charging while plugged in, since you're bound by a cable anyway?
13:29:34 Part tucoz ("Leaving")
13:29:50webguest65H140 cant charge via usb
13:29:51amiconnPaul_The_Nerd: what did you want to ask me?
13:30:23Paul_The_Nerdwebguest65: I didn't say you should charge by USB. I said that you might as well make sure it's charging.
13:31:00Paul_The_Nerdamiconn: I was curious if the replacements of the #pragma stuff might have other effects than simply removing some bugs with -Os
13:31:17webguest65ok, would it be possible to display battery voltage in the usb logo whilst connected to a pc ?
13:31:41Paul_The_Nerdwebguest65: It's open source, so yeah you could make that change.
13:32:05 Join Lynx [0] (
13:32:06webguest65vesibility is good, now to find a coder :)
13:32:31webguest65feasibility even
13:32:36Paul_The_NerdWhy do you need actual voltage though?
13:32:52amiconnPaul_The_Nerd: It shouldn't. In fact, the replacement shouldn't have any effect if there were no gcc bug
13:33:19webguest65well, I do alot of file work without the charger, so thought it would be nice to know the status of the battery whilst connected to a pc
13:33:22wehnamicon: ok. hadn't heard that before. Not really a practical charge pattern though.
13:33:28Paul_The_Nerdamiconn: Okay. It's just that people are reporting that the iPod 4G freezes vanished, and it's either that or playback engine changes.
13:33:42amiconnPaul_The_Nerd: These changes were all archos only
13:33:45Paul_The_Nerdwebguest65: "Battery status" is different than voltage though. A % full should be valid for status.
13:34:03webguest65mea culpa, meant battery voltage
13:34:22webguest65battery voltage is way more meaningfull than %
13:34:35amiconnThe bug is that a #pragma interrupt turns _all_ subsequent functions into interrupt handlers instead of just the next one with sh-elf-gcc 3.4.x and any optimisation level > 1
13:34:42Paul_The_Nerdwebguest65: How so if it's just while you're using the unit as an external HD?
13:34:57_FireFly_webguest65: not really if you don*t know the with which voltage the battery is empty
13:35:07amiconnpreglow: around?
13:35:08Paul_The_Nerdamiconn: Aaah, ick. Well, I guess the playback engine overhaul is what fixed the cpu scaling bugs, at least supposedly.
13:35:16webguest653.3 vollts is disc cutoff no?
13:35:38webguest65that means more to me than one bar
13:35:44Paul_The_NerdOne bar?
13:35:48Paul_The_NerdIt's shown as 0-100%
13:35:57_FireFly_Paul_The_Nerd: or graphical
13:36:03_FireFly_depending on the setting
13:36:07Paul_The_NerdWell, yeah, but it can be shown as a percent
13:36:07webguest65I prefer voltage :)
13:36:16_FireFly_webguest65: then code it ;)
13:36:25webguest65but ofcourse
13:36:32Paul_The_NerdYeah, there's not much call for that being added into the core, since it's very definitely "fluff"
13:36:52webguest65tharts your opinion
13:37:40webguest65thanks for the info, everything is possible, as long as you do it yourself :)
13:37:43 Part webguest65
13:38:01B4gderwebguest65: yes, YOU wrote rockbox didn't you?
13:38:25*B4gder mutters to the whiners
13:38:35Paul_The_NerdI'm still not quite sure how an actual numeric voltage while USB is plugged in is that useful. I mean, what is he going to actually *do* with the number?
13:39:04B4gderhe'll have his own translation table on the table next to his keyboard ;-)
13:39:12wehnruntime or % might be vaguely more useful
13:39:46 Join romanos_melodos [0] (
13:40:38Paul_The_NerdAh well
13:40:41romanos_melodosrockbox-ipodnano and shure e2c = killer combination
13:41:15Paul_The_Nerdrockbox'ed *anything* with headphones you like == killer combination. ;-) At least any target so far.
13:41:18romanos_melodosanyone know how can i create playlist using rockbox?
13:41:29B4gderthe manual knows!
13:41:41Paul_The_NerdThey're basic M3U playlists if you want to make them on your computer
13:41:49*wehn promises he's truely grateful B4gder, <3 060417 rockbox daily
13:41:52Paul_The_NerdIf you want to do it in Rockbox, the manual covers it a lot better than we could here.
13:42:05Paul_The_Nerdwehn: Why a 10 day old daily?
13:42:05romanos_melodosoh! so can i just use winamp playlists?
13:42:13B4gderromanos_melodos: yes
13:42:14Paul_The_NerdIf'n yah want to.
13:42:16wehn27 27!!
13:42:29Paul_The_NerdI was just curious if some bug that was bothering you hadn't been resolved yet
13:42:40romanos_melodosjust another small question. Why rockbox doesn't support ape?
13:42:44wehnoohh playback seeking, backlight flash gone, it's all so nice now.
13:43:04JdGordonstill souns like a bag of shit tho
13:43:08Paul_The_Nerdromanos_melodos: It's not suitable for portable devices. The format itself wouldn't be able to run full speed.
13:43:33 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
13:43:37Paul_The_NerdThough doesn't one of our future potential targets have like a 400mhz core?
13:43:44 Nick slimeball is now known as jamezz (
13:43:57wehnJdGordon: lol, where'd that quote come from?!
13:44:12JdGordonthe mailing list somewhere.. its a good one :D
13:44:33B4gderPaul_The_Nerd: yes gigabeat
13:44:55 Quit Lynx_ (Connection timed out)
13:45:08linuxstb_Paul_The_Nerd: I never believed the CPU scaling was the cause of the reported crashes. Maybe it was something like a race condition in the playback code which only happened when the CPU was running at 30MHz, so disabling CPU scaling hid that problem.
13:45:14romanos_melodosjust a final one... if i'm using both ipl and rockbox, does that affects rockbox perfomance?
13:47:11Paul_The_Nerdlinuxstb_: Well, I just called it the "cpu scaling problem" since that's what made it apparent. :)
13:47:26Paul_The_Nerdromanos_melodos: It shouldn't, unless there's something wrong.
13:49:07wehnB4gder: Makes me laugh everytime I think of it. I like the 5th one on that page too.
13:49:40B4gderyeps, but it seems he might be proven wrong soon
13:49:49 Quit quobl_ (Remote closed the connection)
13:49:56romanos_melodoswell basically, i haven't installed ipl yet but it's on my future plans. The main reason is the video playback feature.
13:50:20romanos_melodosany chance for including the ipl video playback in the next rockbox release?
13:50:27B4gderis uncompressed video really that cool?
13:50:54B4gderI mean, it would probably be rather easy to support that in Rockbox too
13:50:57romanos_melodosi just wanna test it..
13:51:14Paul_The_Nerdromanos_melodos: The video feature they use is actually pretty miserable if you put it in context. If Rockbox were to do video in the future, my hope is that we'd do it quite a bit more impressively. :) But that's not something to expect for a *long* time if ever.
13:51:43romanos_melodosis it that difficult paul_the_nerd?
13:51:58linuxstb_Or we just wait for IPL to release their next video player (I know a few people are working on things) and steal that.
13:52:04romanos_melodosi mean , they made it for uncompressed..
13:52:22B4gderlinuxstb_: a good plan! ;-)
13:52:26romanos_melodosthey just have to put the damn codecs in it
13:52:37Paul_The_Nerdromanos_melodos: Never ask if it's really that hard. If *you* can't do it, by definition it's "that hard" at least. :)
13:53:14Paul_The_NerdI've found that it's basically impossible to judge how hard something is once it's at the point where it's "too hard for me to do"
13:53:21romanos_melodosyeah.. i'm sure they've already done much..
13:53:32romanos_melodosi mean , i can play doom on my nano..
13:53:47linuxstb_romanos_melodos: The ipods currently struggle to decode some audio files in realtime, so imagine what would happen if you tried to decode audio and video at the same time.
13:53:52romanos_melodosso that made me a little bit greedy
13:55:08romanos_melodosis that clearly a matter of ipod's hardware?
13:55:25romanos_melodosi mean cpu, ram etc...
13:56:25linuxstb_Yes, and how efficient the codecs are, and how well they've been optimised.
13:57:21linuxstb_Also, for the 4G Photo/Color and 5G ipods, the LCD updates are very slow.
13:57:56romanos_melodosis nano a 4G?
13:58:41romanos_melodoswhat is it then?
13:58:45Paul_The_NerdIt's the Nano.
13:59:02romanos_melodosi was just about to say that
13:59:17Paul_The_NerdThere's 1G - 4G, then 4G Color and Photo (which are the same thing), Mini 1G and 2G, Nano, and Video.
13:59:21romanos_melodosbut i wasn't sure if nano is a generation itself
13:59:52 Quit ProgramZeta (Read error: 110 (Connection timed out))
13:59:54 Quit Lynx (Read error: 110 (Connection timed out))
14:00:05romanos_melodosso if i want to compare a nano with another genaration, what generation would that be?
14:00:14linuxstb_The Nano shares some hardware with the 4G and some hardware with the 5g.
14:00:58 Join darkless [0] (
14:01:00linuxstb_e.g. it uses the same CPU and backlight as the 5g, but the same LCD controller as the 4G Color.
14:01:12linuxstb_But it doesn't have the Broadcom video DSP that the 5g has.
14:01:30romanos_melodosthanx for the info
14:01:55 Join Lynx_ [0] (
14:02:35romanos_melodosi'm sure people back in apple are working on it.. i mean the video dsp thingo
14:02:54romanos_melodosfor getting that on the next nano release..
14:03:30 Join RedBreva [0] (
14:04:06linuxstb_I hope not. We don't need more undocumented chips... A faster main CPU would be more interesting, and potentially more useful for Rockbox.
14:05:11romanos_melodosoh ok then :P
14:05:25 Quit kaiwai (Remote closed the connection)
14:06:23Jungti1234bye all
14:06:33 Quit Jungti1234 ()
14:07:07 Join ProgramZeta [0] (
14:07:36 Quit romanos_melodos ()
14:11:25*B4gder sent off the mail
14:11:39B4gderwith the P-word
14:13:09Paul_The_NerdOh, got the mail just now
14:13:46Paul_The_NerdI agree by the way
14:14:40B4gder1st of May feels very close
14:14:50 Join Rondom [0] (n=Rondom@
14:14:53preglowlinuxstb_: btw, you mentioned something about a data abort with the new libmad patches at on time, yes?
14:15:38*amiconn notices proeglow around
14:16:07Mode"#RockBox +o B4gder " by ChanServ (ChanServ@services.)
14:16:08linuxstb_No - I just asked if you had tested them on ARM. I thought there might be an issue because the bit-reading functions had been changed to use 32-bit reads.
14:17:01amiconnpreglow: More news regarding the jumptable issue. The offset_unsigned flag itself isn't the problem, it's evaluated correctly
14:17:06Topic"Latest news: extended freezing period suggested" by B4gder (n=daniel@rockbox/developer/bagder)
14:17:35 Nick t0mas_ is now known as t0mas (
14:17:39amiconnWe just hit a corner case where the offset is increased by 1 _after_ deciding about the signedness
14:17:41Mode"#RockBox +o t0mas " by ChanServ (ChanServ@services.)
14:17:47t0mascome on..
14:17:53amiconnThe real bug is in the jumptable size calculation
14:18:26amiconnI'm 99% sure, only have to find the place where the jumptable size is calculated
14:18:52Topic"The Rockbox Channel - Latest news: extended freezing period suggested - Have a server? -->" by t0mas (n=tomas@rockbox/developer/t0mas)
14:19:14B4gdert0mas: I find the extra info redundant
14:19:21t0maswe still need some :)
14:19:28B4gderthis being #rockbox should be evident
14:19:35 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
14:19:38B4gderand the build server info hardly gets us more servers anymore
14:19:38t0masah, you mean that part
14:20:05B4gderbut sure, it can still be mentioned
14:20:10amiconnt0mas: I'm not sure whether additional build servers are that useful.
14:20:25B4gdera larger amount of new servers will make a difference
14:20:28t0masme neither, but I think we're getting short in bandwidth soon
14:20:32B4gderonly a few will hardly be noticable
14:20:33amiconnIt seems there's a saturation effect in the build times, due to ccache and the handout of buiöds
14:21:00B4gderI don't think we'll have a bandwidth problem
14:21:09B4gderthe problem we have is the slowest servers
14:21:10amiconnThe more build servers, the less likely it is for a specific build server to get the same build twice in a row
14:21:27amiconn...and that makes ccache less efficient
14:21:35B4gderand the slowest servers that get a second build very late in the game and that build takes a long time
14:21:52B4gderamiconn: ccache by default fits all builds in the cache
14:22:14tucozI see that a file in the manual is marked as DOS in emacs. How do I convert this (from within emacs) to unix?
14:22:26amiconnB4gder: yeah, but ccache doesn't help if a source file needs to be recompiled due to changes
14:22:48B4gderamiconn: right, but then how is it a ccache issue?
14:23:02amiconn...and more build servers extend the period for each build server in which this recompilation will hurt the build time
14:23:34amiconnB4gder: I didn't say it's a ccache _issue_
14:23:50B4gderyou said you could make it less efficient somehow
14:23:53B4gderand I don't see how
14:23:57amiconnI said the build distribution among more build servers makes ccache less efficient
14:24:09B4gderonly if the cache isn't big enough, right?
14:24:46B4gderoh, you mean subsequent builds would be faster if done on the same server as the last lap
14:25:02linuxstb_Yes, and with more servers, that is less likely to happen.
14:25:14amiconnIf one server does all the builds, a big change (e.g. in a .h file) makes it refill ccache completely, and the next round will be fast again for all builds
14:25:14B4gderI think that's a very minor penalty
14:25:24wehntucoz: don't know about emacs, but maybe: sed 's/.$//' filename > converted
14:25:28B4gderoverall I mean
14:25:29amiconnIt's clearly visible in the build stats
14:25:42tucozwehn, thanks. I ran dos2unix on the file :)
14:26:22B4gderyes its visible, but it isn't very frequent
14:26:35B4gderis it?
14:26:46amiconnMy impression is that it is quite frequent
14:27:39amiconnWould it be possible to define some kind of automatic build preference?
14:28:13amiconnI.e., make it more likely that a server gets a certain build in the next round if it did this build in the current round
14:29:02B4gderit could be a bit hairy to get it right
14:30:40B4gderI think we could also benefit from handing out the slowest builds first
14:30:58B4gderto top up with smaller and smaller
14:31:27amiconnYeah, swcodec target & sim first, then hwcodec, bootloaders last
14:36:24ravonWhere can I read up on char -> ushort sound conversions?
14:40:49linuxstb_You mean "signed char" to "unsigned short" ?
14:41:00linuxstb_Or unsigned char...
14:41:10ravonSigned char
14:41:21ravonI have my own way, but the sound gets all fuzzy
14:48:01preglowamiconn: no clues on what the real problem is, then?
14:48:31amiconnpreglow: [14:17:41] <amiconn> I'm 99% sure, only have to find the place where the jumptable size is calculated
14:49:05amiconnAll SH1 insns are 16bit, so jumptable must have even size
14:49:09 Join Siku [0] (
14:49:28ZagorB4gder: a simple server speed toplist could be done (with statistics) and start with both the big jobs and the big servers
14:49:33amiconnByte sized jumptables need a padding byte if the number of cases is uneven
14:49:34 Quit wehn (Read error: 113 (No route to host))
14:50:26amiconnIn the case where the bug hit us, there were 5 cases, and the offset to the last one was 127, since the padding isn't considered in the table size decision
14:50:40B4gderZagor: yes but basing on collected data like that can be a bit cumbersome to work with
14:50:50amiconnThe padding in the insn output stage then bumped the offset to 128 - boom
14:50:55B4gderZagor: just sorting the builds in size-order from what we know is a lot easier
14:51:56ZagorB4gder: yes, but simply storing the build times in a file shouldn't be very hard.
14:52:00amiconnThe easy fix is reducing the maximum allowed distance for byte-sized table by 1. I tested it, it works
14:52:13amiconnBut the proper fix would be to fix the table size calculation...
14:52:33B4gderZagor: no, just that I don't think we'll gain a lot
14:53:17Zagorperhaps not
14:53:58B4gderI mean, the order of the builds would be rather static given enough time
14:55:35 Join quobl [0] (n=quobl@tor/session/external/x-04a2e00b75605d05)
14:56:43 Join Bjoern-Erik [0] (
15:01:34 Join gtkspert [0] (
15:01:55 Quit perldiver (Read error: 104 (Connection reset by peer))
15:02:58 Join DpK|Cronic [0] (
15:03:16DpK|Croniccan anyone give me a good theme?
15:03:27DpK|Cronicwith a good code?
15:03:47Paul_The_NerdThe themes are all to be found in the WPS Gallery in the wiki. :)
15:04:46DpK|Cronicy i see it but thats not sooo good ^^ and the iPod Green5g are not functionaliti
15:04:57DpK|Cronicthe text are not on the box
15:05:04DpK|Cronicthere is on the left site..
15:07:04 Quit ProgramZeta (Read error: 110 (Connection timed out))
15:07:57DpK|Cronici would an better menu :( are you can help me Paul_The_Nerd ?
15:08:34Paul_The_NerdDpK|Cronic: I'm not understanding what you're asking. Some themes require patches for text alignment. For the menu you can only set backdrop and font
15:09:22 Part kapheine
15:09:25DpK|Cronicahh the patches...where must they in?
15:09:26ravonOh speaking of that. Will those patches be a part of the main tree after 3.0?
15:10:42Paul_The_NerdDpK|Cronic: They must be applied to the source code, and then rockbox must be compiled. It's described in the wiki.
15:11:20Paul_The_Nerdravon: The album art patch isn't being included because it doesn't load album art into memory with songs the way it's been decided would be optimal.
15:12:19pregloware anyone looking at what wpses we should bundle?
15:13:07Paul_The_NerdHas anyone actually submitted any WPSes through the proper process recently?
15:13:28B4gderI think we've hesitated to accept any
15:13:54B4gderI'd rather have us arrange them in a new way
15:14:10B4gderwith ~3 WPSes for each target LCD
15:14:50B4gderI wish someone would bother to setup a dedicated site and service for Rockbox themes
15:15:47DpK|Cronicthr pszvhrd in G:\.rockbox
15:16:08DpK|Cronicthe patches in g:\.rockbox ?
15:16:43JdGordonso... um... when is rockbox.blaa moving into .rockbox??
15:16:55ravonDpK|Cronic: You need to patch the sources before you build them.
15:16:57Paul_The_NerdDpK|Cronic: No, I told you they have to be applied to the source code, and compiled. See and
15:17:25linuxstb_Paul_The_Nerd: The album art patch seems much better now - it doesn't waste RAM, but it causes an extra disk spinup just before each track change, and the album art is loaded into the WPS image buffer.
15:17:48linuxstb_But this also only happens if the image is different to the previous one.
15:17:48Paul_The_NerdB4gder: A non-wiki site, where people submit them, the maintainer reviews the WPS to make sure it handles things properly (nothing extra in the .cfg, it explicitly sets font and background colors if that is required for the theme to work well, etc)?
15:18:13B4gderwith previews, sizes, zips to download, all targets
15:18:24B4gderand user rating
15:18:29Paul_The_Nerdlinuxstb_: Well, I understand that lostlogic wants to make metadata on buffer work properly, which would probably then be ideal for integrating the album art into, right?
15:18:30JdGordonPaul_The_Nerd: that could be done with a script...
15:18:39B4gderlike those winamp/xmms/etc skin sites
15:19:01linuxstb_Paul_The_Nerd: Yes, I think that's everyone's preferred approach.
15:19:47preglowB4gder: how would that site differ from what we have now?
15:19:57 Quit Spacksaw (Read error: 104 (Connection reset by peer))
15:20:10B4gderwell, we don't have such a service now
15:20:32 Join Acksaw [0] (
15:20:33B4gderthat would be the difference ;-)
15:20:47linuxstb_I would be surprised to see a volunteer for that job though.
15:21:13B4gderI pushed for it a bit at MR
15:21:15preglowi could have done so, but i haven't got the webspace
15:21:20preglownor the time :>
15:21:27Paul_The_Nerdpreglow: There would be a human filter (or a few) to help make sure WPS submissions actually *work.* A lot of the ones on the Wiki have had errors (the .cfg points to a non-existing .wps, or assumes the text is the default color, etc)
15:21:28preglownot the interest in wpses :>>>
15:21:47B4gderI have the space and skill, but I don't feel like taking on yet another task
15:22:06B4gderwould be neater to get a new fresh web guy to do it
15:22:15preglowthe only webspace i've got available only have mod_php and _MOD_LISP_
15:22:20preglowand i require mod_perl for proper operationm
15:22:34B4gderwell, we have many gigabytes to offer... :-)
15:23:50 Join Spacksaw [0] (
15:24:01Paul_The_NerdI'll gladly help with checking over submissions for guideline adherence if a web design guy does come along. I'm learning PHP, but I'm still very limited to "The aspects of it necessary to help with my ex-girlfriend's Geophysics project"
15:24:12 Quit Acksaw (Read error: 104 (Connection reset by peer))
15:24:40preglowwe need to finally decide on the wps patches issue as well
15:25:02preglowpeople will need the time to actually use them before release
15:27:14Paul_The_Nerdlinuxstb_: It doesn't show a checkmark in the Rockbox column by Mini 2G, probably simply because Rockbox wasn't on Mini yet when the page was made or something
15:27:19 Quit DpK|Cronic ("get satisfied! • :: ««« (Gamers.IRC) »»» ::")
15:27:41***Saving seen data "./dancer.seen"
15:28:39Paul_The_Nerdpreglow: Are you talking about the WPS-specific patches that are out there? There's only like two left right (at least in wide circulation), the scroll margins and album art? Or are there others.
15:30:50_FireFly_Paul_The_Nerd: also my combined-bitmap-support ;)
15:31:06_FireFly_but currently no updated patches arround
15:31:07tucozIf one makes changes to a binary file (image) and want to commit the changes, is there something special I need to do to commit that?
15:31:17 Join Kohlrabi [0] (
15:31:44tucozor will the current cvs version simply be overwritten?
15:32:10tucoz...if I cvs commit?
15:32:42 Join DpK|Cronic [0] (
15:32:46preglowPaul_The_Nerd: scroll margins would be the more important one, if you ask me
15:32:49DpK|CronicPaul_The_Nerd you are german or?
15:32:57amiconnNothing special, just commit like normal
15:33:04tucozamiconn, thanks :)
15:33:15preglowtucoz: hey, congrats on the little one
15:33:24tucozpreglow, thanks :D
15:33:37Paul_The_Nerdpreglow: Honestly, a lot of people seem to use it, and I'm somewhat in favour of it being added as a stopgap until proper viewports as long as it can be integrated in a way that allows it to be backed out cleanly when the time comes.
15:34:04preglowPaul_The_Nerd: i think wps is going to see a major overhaul, so i don't really think it needs to be backed out as such
15:34:05Paul_The_NerdDpK|Cronic: American.
15:34:14DpK|Croniccan anyone install meThe Rockbox Source Code in the remote desktop pls?
15:34:16DpK|Cronicah ok
15:34:35DpK|Cronicpls can anyone install me The Rockbox Source Code in REMOTE DESKTOP from windows?
15:34:41Paul_The_Nerdpreglow: Well, in that case, I'm definitely in favour of it. It solves what I see as the single most common *universal* complain about WPSes (as I see Album Art as probably a smaller but more vocal group. I'm not sure though)
15:35:18preglowi don't care too much about it
15:35:27preglowbut scroll margins i do care somewhat about
15:35:36preglowespecially since my current wps needs it badly :P
15:35:50Paul_The_NerdYeah, Scroll Margins would actually be a WPS feature I'd use.
15:36:14DpK|CronicPaul_The_Nerd can you install me The Rockbox Source Code ? ^^
15:36:24lostlogicDpK|Cronic: read wiki, do it yourself.
15:36:48DpK|Cronici read it ! but i dont understand the cygwin O.o
15:37:08Paul_The_NerdDpK|Cronic: Have you installed Cygwin yet?
15:37:16DpK|Cronicyes on C:
15:37:41Paul_The_NerdAnd what exactly are you having problems with?
15:37:45Paul_The_NerdWhich step confuses you?
15:38:21_FireFly_wasn't there a rockboxdev package with an modified cygwin installation
15:38:40lostlogic_FireFly_: was dropped in favor of the vmware image I believe
15:38:43Paul_The_Nerd_FireFly_: There's a set of packages available from Rockbox's servers that include all the crosscompilers, etc.
15:38:47DpK|Cronicthe step in The Rockbox Source Code From CVS
15:38:47DpK|CronicCheck out source -
15:38:58Paul_The_NerdDpK|Cronic: Those are commands to type inside of cygwin.
15:39:21DpK|Cronichmm.. but when i open cygwin they closed automaticle
15:39:48preglowPaul_The_Nerd: you know where the scroll margins patch is?
15:39:57Paul_The_NerdDpK|Cronic: Then you have not installed cygwin properly
15:40:17preglowit's the paprica one?
15:40:23preglowyup, found
15:40:32DpK|Cronicok install the 2nd...
15:40:56preglowugh, it really does touch a lot of stuff
15:41:08DpK|Cronicinstall with DOX/TEXT ?
15:41:30DpK|Cronicor unix?
15:41:44Paul_The_NerdThe default, which I believe is Unix.
15:41:55amiconnHmm, who's currently in chargethe vmware image? of
15:42:06 Quit Spacksaw (Read error: 104 (Connection reset by peer))
15:42:07 Join Acksaw [0] (
15:42:17amiconnIt will need updating with the patched sh-elf-gcc soon...
15:44:00lostlogicamiconn: so will all the build servers
15:44:03lostlogicand won't that be an adventure
15:45:08amiconnPerhaps it will
15:45:25amiconnThe good thing about gcc 3.4.x for sh is that we don't need to build with newloib
15:45:35lostlogicPaul_The_Nerd: this dr.dns dude must have the most convoluted brain on earth.
15:45:43lostlogicamiconn: but we do need to patch it, yesno?
15:45:43DpK|Cronici install it again but its not gone <.<
15:45:50JdGordonis the specs of the build servers on the web anywhere?
15:46:08lostlogicPaul_The_Nerd: afaics his problem is that auto change directory fucks up sometimes.
15:46:25preglowdr. dns?
15:46:26amiconnlostlogic: Yes, to fix a bug that is present in all sh-elf-gccs and hits at optimisation levels >1
15:46:42lostlogicpreglow: forum poster.
15:46:47lostlogicamiconn: aye
15:47:17DpK|Cronicthey arent a .bash_profile ? how i can load it ? o.o
15:48:01amiconnPerhaps I'll post a patch to the gcc bugtracker, but (1) I don't really expect them to fix their trunk and (2) they surely won't fix the 3.4.x branch since it's officially closed
15:48:16amiconn...and we cannot use 4.x :-(
15:48:52ravonWho's dr.dns?
15:49:03ravonI keep hearing about him.
15:49:18lostlogicravon: forum poster... with spaghetti in his head.
15:49:26lostlogicamiconn: did they drop sh in 4.x?
15:49:28ravonlostlogic: lol. On the rockbox forums?
15:49:34Paul_The_Nerdlostlogic: Some of these bugs are getting more uncommon. "Auto resume with autobookmark only fails when the unit idle powers off."
15:50:01lostlogicPaul_The_Nerd: yeah, I saw that... and I liked your next test request.
15:50:42Paul_The_Nerdlostlogic: I never realized how boring I was. I actually turn my player on and off by hand, and listen to music in either a sequential or shuffled playlist.
15:51:10lostlogicPaul_The_Nerd: I know... me too... which both work fine
15:51:40preglowlostlogic: would reworking the way rockbox works to be compatible with gcc 4 be a great bother?
15:51:44lostlogicPaul_The_Nerd: the only thing I actually do that sometimes breaks it is skipping back and forth in certain patterns related to buffer levels and fill times and stuff.
15:52:24 Join ProgramZeta [0] (
15:52:28DpK|CronicPaul_The_Nerd can you qry me pls?
15:53:29 Quit linuxstb (Remote closed the connection)
15:54:02 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
15:54:38ravonFigured out 8->16 got all fuzzy, I think. Someone slap me when you have time.
15:54:46JdGordonok, is there anything painfully simple but boring that needs to be done for mayday?
15:54:49ravonAnd my sentences make no sense either.
15:55:26ravonnvm, just got called "trash!" for it
15:56:17*JdGordon abliges and slaps ravon
15:56:42Paul_The_NerdJdGordon: Bugfixes, bugfixes, bugfixes.
15:57:05lostlogicJdGordon: <−− the bugs listed there... although I can't call them painfully boring :-D
15:57:11Paul_The_NerdJdGordon: Oh, the "View ID3 information" option in the Context Menu should show "Composer" along with everything else, as composer is a supported tag in WPS and TagCache. I think that qualifies as "Simple" maybe?
15:59:06JdGordonPaul_The_Nerd: any idea what file thats in?
15:59:23amiconnlostlogic: No, but they took a silly design decision that blocks us from using 4.x for sh rockbox.
15:59:31Paul_The_NerdJdGordon: Not a bloody clue. I just remember someone was quite upset about it's lack. :)
15:59:38JdGordonah ok
16:00:39amiconnpreglow: I have absolutely no idea how to work around that silly gcc 4.x thing
16:01:11ravonweee, it worked.
16:04:51preglowamiconn: perhaps we should patch our own gcc4 as well
16:05:12preglowamiconn: 'cause there's pretty much no way whatsoever that they're going to revert that
16:06:26 Part Paul_The_Nerd
16:09:08preglowi like the way in which pinski just suddenly stops responding to the bug
16:13:59lostlogicbagh, gcc people.
16:16:51godzirraHrm... I cant delete a folder off my h120. Anyone have any suggestions? it tells me the folder/disk can't be read.
16:17:02godzirraI did a scandisk on it and it seems to show up fine.
16:17:02 Part tucoz ("Leaving")
16:18:54 Quit darkless (Client Quit)
16:20:01 Join Farpenoodle [0] (
16:21:21JdGordonthe langv2 changes are in place now right? i.e it shuold say NAVI = yes on the h300?
16:25:50JdGordonoops.. can some1 change to a patch.. pressed ok to quickly
16:26:30amiconnpreglow: Actually I talked to him y'day. He says that in his opinion this is 90% not a bug, and we're doing something in our code which isn't considered legal code
16:26:59amiconn..but he didn't tell me what isn't legal about it
16:27:46amiconnI don't ask to remove this feature, since it probably makes sense in most cases, but they could have added an option to disable it...
16:28:50preglowwell, they too don't like having seldom-used options in their code
16:28:58ashridahamiconn: wait, you dont' have the specs he's working from memorized? i thought everyone did.
16:29:00preglowbut i don't think this option in particular makes much of a difference
16:29:40godzirraAnyone have any ideas?
16:30:20ashridahgodzirra: inside rockbox, or when connected to the pc?
16:30:40godzirraashridah: connected to the pc.
16:31:18ashridahweird. did your scandisk include a surface scan?
16:31:32godzirraI'm doing a scan for bad sectors now.
16:31:37godzirraits still running.
16:38:10preglowi found another way to detect incoming sample rate for spdif
16:40:16godzirraI still cant delete the file :(
16:40:20godzirraI dunno what the hell is up.
16:42:01mikearthur_is anyone here familiar with Wolfson Microelectronics?
16:42:41 Join quobl_ [0] (n=quobl@tor/session/external/x-87ab636a50becea1)
16:42:49preglowmy pc's spdif is 48005 hz \o/
16:42:53lostlogiconly in that they make the codecs in the ipods and don't publish specs for the specific ones in the ipod.
16:42:57 Quit JdGordon ("Miranda IM! Smaller, Faster, Easier.")
16:43:39preglowmy cd player spdif varies between 44097 and 44099 :/
16:44:11 Quit _FireFly_ ("Leaving")
16:44:12linuxstbmikearthur_: What's your question?
16:46:06mikearthur_linuxstb: just curious if anyone is familiar with them
16:46:11mikearthur_I may be working there in the summer
16:46:23preglowhave them publish more specs
16:46:41mikearthur_I could mention it, doubt it would happen
16:46:55preglowwe're fine anyway
16:46:58preglowwe've figured out most
16:47:02linuxstbI'm not that worried about the specs - most of their codecs seem to be register-compatible with each other, and we've found datasheets that seem to match all the ipod codecs.
16:47:19preglowcan't you get work at portalplayer instead? :P
16:47:19 Nick mikearthur_ is now known as mikekarthur (
16:47:29linuxstbOr Samsung now...
16:47:31mikekarthurbecause portalplayer haven't given me an interview :p
16:47:38preglowlinuxstb: is it official?
16:47:45linuxstbI don't think portalplayer will be doing much hiring in the near future...
16:48:11linuxstbpreglow: Yes - Samsung have announced it. Search google news for samsung and ipod.
16:48:27mikekarthurbut anyway, other than anything else, hopefully my C skills will be boosted greatly, and I could help with rockbox :D
16:48:35linuxstbI think all they've said is that it will be ARM-based.
16:48:42preglowno surprise
16:48:49preglowapple will wont to modify as little code as possible
16:49:19preglowSamsung’s chip is based on 32-bit processor technology from ARM, which is a bit sexier than the PortalPlayer chip.
16:49:21 Join klrspz [0] (n=klrSpz@
16:49:25preglow"that is so"
16:50:05linuxstbBut I would be surprised if Apple's current firmware didn't use large portions of the PP firmware development kit.
16:50:44preglowit does
16:50:53preglowthe one they use is old as hell
16:50:56preglowso i doubt they care much
16:51:23preglowunless they're the "don't fix it if it isn't broken" type of people, of course...
16:52:41preglowi want their mp3 decoder...
16:53:34preglowSlasheri: here?
16:53:58Slasheripreglow: yes :)
16:54:35 Quit ProgramZeta ("Trillian (")
16:55:07 Quit actionshrimp (Read error: 110 (Connection timed out))
16:55:27preglowSlasheri: could you do some 96khz testing for me afterwards?
16:55:40preglowi found a new method to measure the sample rate directly, and want to see if i've got it right
16:55:57 Join PaulJam_ [0] (
16:56:14 Quit elinenbe (Read error: 104 (Connection reset by peer))
16:57:25Slasheripreglow: sure
16:57:54Slasheriif i just manage to get the cables work :)
16:58:01preglowjust finishing the code now
16:58:21Slasherigood, just tell me when it's ready and i will try out :)
16:58:33 Join elinenbe [0] (
16:58:58 Quit quobl (Remote closed the connection)
16:59:21 Quit ashridah ("Leaving")
17:01:05 Quit Hadaka ("upgrade upgrade")
17:03:13 Join Naked [0] (
17:03:33 Quit RedBreva ("CGI:IRC")
17:04:20 Join RedBreva [0] (
17:11:26 Quit PaulJam (Read error: 110 (Connection timed out))
17:12:25 Quit DpK|Cronic ("get satisfied! • :: ««« (Gamers.IRC) »»» ::")
17:12:39 Join Genre9mp3 [0] (n=yngwiejo@
17:13:11 Quit muesli__ (Read error: 110 (Connection timed out))
17:14:57 Nick Naked is now known as Hadaka (
17:18:04RedBrevaJust reading the logs, anyone have any objection to me having a go at a mini themes site (as a proof of concept)
17:18:18B4gderon the contrary!
17:18:54B4gderI'll even offer you a name for your server in the domain! ;-)
17:21:18RedBrevaheheh, I'll knock something up in my own domain initially and see if it looks OK first ;-)
17:21:42B4gderoh sure, just showing my enthusiasm! ;-)
17:22:17preglowSlasheri: will break settings
17:22:46preglowSlasheri: this time, the sample rate displayed on the recording screen should update itself after what it the spdif input frequency is measured to be
17:22:53preglowplease test it with as many as you can :)
17:23:14Slasheripreglow: hehe, i will try :)
17:25:09 Join PaulJam__ [0] (
17:25:32 Nick PaulJam__ is now known as PaulJam (
17:27:45***Saving seen data "./dancer.seen"
17:28:38Slasheripreglow: btw, there is still the bug that recording will crash when trying to stop unless signal is found
17:29:07preglowi know
17:29:10preglowi haven't fixed that
17:29:16thesonorki installed rockbox yesterday, and i mentioned that it does not Auto-Off the display so it sucks my battery down, what to do?
17:29:23preglownothing has changed except the freq measurement
17:31:07 Quit klrspz ()
17:31:27thesonorkany ideas?
17:31:39preglowwhat player?
17:31:49Slasheripreglow: works well! frequency even updates correctly when changing the freq on the fly
17:31:53thesonorkipod mini 2g
17:31:58Slasherii tested with 44.1 48 and 96 kHz
17:32:00linuxstbthesnork: What is your "backlight timeout" set to?
17:32:02 Quit Zagor ("Client exiting")
17:32:10preglowSlasheri: cool, thanks for testing
17:32:11Slasheriand yes, optical out has to be enabled for optical in to work
17:32:14Slasherinp :)
17:32:15linuxstbAnd the backlight doesn't go off?
17:32:22preglowhe said display
17:32:27thesonorknono, not the backlight, the display....
17:32:36preglowdoes the apple firmware switch off the display?
17:32:41 Nick jnc_ is now known as jnc (n=jnc@
17:32:50preglowno, it sleeps the entire unit, not just the display
17:32:58preglowanyway, the display doesn't really suck much power
17:33:07thesonorkthen what sucks my power down :P
17:33:08preglowwe've got far worse problems than that
17:33:10 Join solexx [0] (
17:33:16linuxstbThe Apple firmware turns off the display when music is playing?
17:33:17preglowsomething else
17:33:18preglowwe don't know
17:33:31thesonorklinuxstb, no, on idle
17:33:38thesonorkor on "hold"
17:35:35thesonorkit was just like i kept my ipod in the car, (all the time loaded while driving so 100%) after work i got in my car... only 67% left, so this means that i cannot use zthe ipod for 2 days without loading it
17:35:49 Quit vca ("leaving")
17:37:48preglowit does shut itself off after a given amount of time
17:37:54preglowbut it doesn't sleep itself like apple does
17:37:56preglowwe might fix it in the future
17:38:08preglowbut my ipod boots so fast anyway, i don't see the point in wasting power in sleeping it
17:38:39thesonorkthis sleep feature works
17:38:48thesonorkbut my ipod INSTANTLY reboots after sleeping
17:39:04thesonorkso no effect and another waste of power
17:39:40preglowhm? your ipod instantly reboots when it turns itself off? in rockbox?
17:40:08thesonorki dont know
17:40:11preglowmine sure as hell stays off when it turns itself off
17:40:17preglowyou got anything connected to the expansion port?
17:40:29thesonorkno nothing, i just tried it in the kitchen
17:40:52thesonorkgot that sleep timer to 1min, and waited, it turned off and booted again
17:41:09thesonorkso i turned that sleeptimer off
17:42:02 Quit PaulJam_ (Read error: 110 (Connection timed out))
17:42:48preglowSlasheri: if i disconnect the spdif cable enough times, suddenly the peak meter (and hence also actual recording) dies. i wonder why that happens...
17:42:52linuxstbWhat happens if you shutdown Rockbox manually - i.e. press and hold the play/pause button for a few seconds.
17:43:42 Quit B4gder ("time to say moo")
17:43:46thesonorklinuxstb, instant reboot
17:44:15preglowamiconn: you recognize this? :>
17:44:28linuxstbthesonork: You have a very confused ipod...
17:44:33 Join dpro [0] (
17:44:59thesonorkwasn_'t u the one patching my rockbox last night? caused to that usb stuff?
17:45:07thesonork@ linuxstb
17:45:16thesonorksee ^^
17:45:22linuxstbmidkay actually did the patch for you though.
17:45:52linuxstbYou seem to have the only ipod mini in the world where the Rockbox USB detection gives a false positive...
17:45:57thesonorkknow that, just wanted to recover the visions in my head
17:46:42dprocan anyone enlighten me again on how to use dd instead of ipod_fw ... my x-dev environment now moved to my freeBSD server and my ipod is connected to my powerbook with only a linux live cd due to a hd failure ...
17:46:44thesonorkI am taking this as a complement
17:47:29dprodamn ... I can't even crosscompile pcfdisk .. maybe parted will do the job ...
17:47:38linuxstbdpro: Yes. Just use "dd if=/dev/sda1 of=bootpartition.bin" to read the firmware partition, and "dd if=rockboot.bin of=/dev/sda1" to write the modified version with the bootloader at the end.
17:47:50linuxstb(replacing /dev/sda with the device assigned to your ipod)
17:48:21linuxstbYou did say you were running Linux?
17:49:00dprolinuxstb: yup
17:49:02thesonorka linux live cd
17:49:18dprowell ... linux on the powerbook and freebsd on the box to compile rockbox ...
17:49:20 Quit PaulJam (Read error: 110 (Connection timed out))
17:49:57linuxstbIt's just that Linux is the only OS that will give you direct access to the first partition. Cygwin and Mac OS X will hide it from you.
17:50:07linuxstb(because it's marked as type 0 - empty).
17:50:33dproah ok, no that shouldn't be a problem then
17:50:36linuxstbThat's why diskdump and ipodpatcher are needed - they access the partition via the whole-disk device.
17:50:56linuxstbThere is an IpodInstallationFromLinux wiki page...
17:51:39 Quit solexx_ (Read error: 110 (Connection timed out))
17:52:43dprobut that also uses ipod_fw, I only need it to merge bootloader and original firmware to make my shiny new video triple boot, but I can do that on the server and then dd it from the laptop (not exactly practical /me knows)
17:53:28*dpro wanders if the live cd supports fuse via ssh that would make all this a little more convenient
17:54:18dprohopefully a friend will give me an unused 2.5" hd tonight so I can start kind of working again tomorrow or so
17:55:05Slasheripreglow: for me, it's enough just to disconnect the cable once and peakmeter dies. However, frequency still updates correctly even after this
17:55:22dprobut if it's just for the partitioning requiring pcfdisk I might be able to find another linux box for those 5 minutes
17:55:34linuxstbSo you're installing IPL as well?
17:55:48*dpro isn't sure if parted likes an empty partition type ...
17:55:52dprolinuxstb: yes
17:56:04linuxstbNo, I think I've read that parted chokes on the empty partition.
17:56:07Mikachulinuxstb: what if you change it to non-empty while doing stuff to it?
17:56:23linuxstbI've never tried, but I can't see why that wouldn't work.
17:56:32 Join freqmod [0] (
17:56:37linuxstbI always use Linux and fdisk, and that's happy to do anything.
17:57:02 Quit quobl_ (Remote closed the connection)
17:57:30dproparted is crap ... it only tells you where /about/ partitions start and end ... very helpfull when trying to recover from a crash ...
17:59:20Mikachui remember many moons ago when i used windows, there was a program called ranish' partition manager, it was pretty good
17:59:39Mikachuit may or may not have been dos-only though
17:59:56thesonorkused partition magic on windows, long long ago ;)
18:00:01 Quit tvelocity ("Ex-Chat")
18:00:14dproah nice ... so I just copy bootpart.bin over to the server an then use ipod_fw to extract all the other stuff needed to merge into a new bootpartition right ?
18:01:04preglowSlasheri: yeah, recording and freq estimation are two completely different things
18:01:09linuxstbYep. And then copy rockboot.bin back at the end of it.
18:01:18 Quit SereR0KR ("XChat Aqua")
18:01:35dprolinuxstb: excellent !
18:01:46 Quit HCl (Remote closed the connection)
18:04:36 Quit RedBreva ("CGI:IRC (EOF)")
18:05:37 Join bluebrother [0] (
18:05:44 Join RotAtoR [0] (n=e@rockbox/developer/rotator)
18:06:41 Join SereR0KR [0] (
18:09:57 Join HCl [0] (
18:21:18 Quit linuxstb ("Client Exiting")
18:26:26 Join PaulJam [0] (
18:27:16 Quit Acksaw ("I'm off, see ya later!")
18:28:08 Join Acksaw [0] (
18:30:40 Join tempi [0] (n=tempi@unaffiliated/tempi)
18:31:40 Join romanos_melodos [0] (
18:32:33romanos_melodoshello! Does anyone know how can i have greek language support on rockbox?
18:33:47 Join tab [0] (
18:34:41Mikachuchange to a font with greek characters
18:34:53Mikachutry unifont
18:35:13Mikachuif you're on linux you need to mount the fat partition with -o utf8
18:35:46romanos_melodosoh.. thanx for that
18:35:49 Quit tab (Client Quit)
18:36:59 Join Farpnut [0] (
18:37:06thesonorkGuys, I sorted my songs using Folders like: /Band/Album/songs.mp3, but in folder view rockbox does not shuffle through the folders, and when i create a playlist, rockbox does not sort UIsing the folders i made... what to do? ^^
18:37:23 Quit Genre9mp3 ()
18:37:30thesonorkUIsing = using
18:39:39goffathesonork: you mean like when you select a song?
18:39:41 Join plop [0] (n=charle@
18:39:48goffait just plays songs from that folder?
18:39:54thesonorkwhen i browse through my songs
18:40:08thesonorkthats exactly my problem
18:40:11 Part tempi ("Leaving")
18:40:19goffayeah... that's default behavior currently :(
18:40:26goffabut... you can load your root playlist
18:40:37goffathen in file browser hold down your navi button
18:40:38thesonorksee, i wasted hours in making my music archiev really cool,... like /band/year-Albumname/Tracknumber-trackname
18:40:48goffathen playlist
18:40:50goffathen insert
18:40:51thesonorkso i have lots and lots of songs starting with tracknumbers
18:41:15plopi smell
18:41:30goffaand then it will go random through all dirs
18:41:37thesonorkgotta do that for every folder?
18:41:48goffaoh.. to queue every folder
18:42:10goffago into playlist options -> create playlist
18:42:17goffathen load root.m3u
18:42:30goffathere's also an option in there for recursively insert directories
18:42:31 Join BOBqwerty [0] (n=bob@
18:43:24goffaso if you have 5 albums by an artist and want to listen to that artist you can insert that dir
18:43:24goffaand it will insert the rest recursively
18:43:25BOBqwertyhi plop
18:43:29ploprock my boy
18:43:45thesonorkhm i think that just worked
18:43:46goffathere has been some discussion on this thesonork
18:43:56romanos_melodosunifont is working great for reading the greek characters only in folders. What about id tags?
18:44:03goffathere seems to be some interest in implementing my feature
18:44:29goffabecause to me its frustrating when you pick a song out of file browser and your playlist goes Poof
18:44:56thesonorkwhat the holy hell ?!
18:45:01lostlogicthat was amusing
18:45:09goffanot really
18:45:11BOBqwertyi no
18:45:13BOBqwertythnkyou all
18:45:15BOBqwertyi bow
18:45:25 Join julius [0] (n=julius@
18:47:09 Quit Farpenoodle (Read error: 110 (Connection timed out))
18:47:13 Quit romanos_melodos ()
18:47:48juliusBOBqwerty, plop are you bored? can I help you somehow? :)
18:47:58BOBqwertyyes we are
18:48:06plopim bored butnope unless u got a huge amount of porn u cud share
18:48:27 Quit Ribs (Read error: 104 (Connection reset by peer))
18:48:55juliusthe problem is that I don't have a high speed internet connection, so you would have to wait for hours for just one movie :)
18:48:55BOBqwertyim afraid i do not have any on me i could share
18:49:17juliusso hi everyone
18:49:39 Join Ribs [0] (
18:50:00BOBqwertywhat were you tlking about b4??
18:50:09 Part plop
18:50:10BOBqwertyanything interesting??
18:50:50 Part BOBqwerty
18:51:17 Quit JBGood (Read error: 110 (Connection timed out))
18:51:32goffathat crap gets annoying
18:51:35juliusi have a litle problem with rockbox, yesterday I had greate support from you guys, and I found out how to turn the power on on the ipod 5g dock connector, but I still have one question
18:51:43juliuscould someone help me with this please?
18:52:03goffa<- knows nothing about ipods
18:52:13goffabut you can ask your question
18:52:22juliusok, so here is the situation:
18:52:29dwihno_asking the question, always a good idea is!
18:53:00thesonorksubject-verb-object dwihno_
18:53:15juliusi found out (with the help of linuxstb) that this code:
18:53:17julius ipod_i2c_send(0x8, 0x24, 0xf5);
18:53:17julius ipod_i2c_send(0x8, 0x25, 0xf8);
18:53:17julius ipod_i2c_send(0x8, 0x26, 0xf5);
18:53:51juliusturns on the power on the ipod 5g connector, so I have power on pin 17 and some other above 17
18:54:11juliusso this is greate, but I tested my itrip with it and it's still not working
18:54:22juliusthan I made some test and found out
18:54:35juliusthat there are two pins which are required by the itrip
18:54:39juliuspin 11 and 17
18:54:49dwihno_hardware stuff!
18:54:50*dwihno_ hides
18:55:19juliusthe only problem is that on the pin 11 I don't have any power if I use rockbox, but with apple's firmware I have 2.9v or something like that on the pin 11
18:56:27juliusso do someone knos how to turn on the power on 11 pin too?
18:56:29lostlogicwhat is pin 11 called in the pinout?
18:57:12juliusit's Serial GND
18:57:29juliusyes, and another thing
18:57:36juliusI have 3.3 on the pin 17 and not on 18
18:57:48 Join JBGood [0] (i=Johnq@
18:58:25juliusor am I counting them wrong?
18:58:32 Join klrspz [0] (n=klrSpz@
18:59:38juliuslostlogic, if I try to increment the 0x26 in the code ipod_i2c_send(0x8, 0x26, 0xf5); could I hurt my ipod?
18:59:58juliusyou know, just to make some tests, maybe one of them would turn the 11 pin on
19:00:16lostlogicjulius: it's not likely that you'd damage it, IMO, but it is possible.
19:01:10juliusI don't really know any other way to found out which one is for pin 11, do you?
19:01:25lostlogicnope, since we don't have datasheets ;)
19:01:40amiconnDisasm the original firmware and start hunting...
19:02:55juliusyep, I could do that, but they said that would be a big jurney :) and I really don't want to go abroad :)
19:03:28klrspzwhat does the the 11 pin do?
19:04:34 Join petur [0] (n=petur@rockbox/developer/petur)
19:05:02juliusklrspz, I'm not sure, I just noticed that I have a power out on it with 3v and it is used by the itrip fm transmiter. And i found out the both pin 18 and 11 are required for the itrip to work (or it seems like that)
19:13:33 Quit Xerion (" ")
19:23:18 Join JBGood25 [0] (i=Johnq@
19:25:34scott666_Is the only thing causing playback stutter on a 5G ipod while listening to music and running a plugin unoptimized code?
19:27:49***Saving seen data "./dancer.seen"
19:28:38 Quit Farpnut (Read error: 110 (Connection timed out))
19:28:46 Join Farpenoodle [0] (
19:29:26 Join ep0ch_ [0] (n=ep0ch@
19:30:35 Join muesli__ [0] (n=muesli_t@
19:33:23 Quit JBGood (Read error: 110 (Connection timed out))
19:33:25 Join kimberly [0] (
19:38:06 Join DarkJesus [0] (
19:46:14 Join macdonalder [0] (
19:49:37preglowscott666_: pretty much
19:50:20scott666_but that's being worked on, right?
19:51:20scott666_if you can't play the games while listening to music theres not a lot of point in having games
19:51:34 Quit JBGood25 (Read error: 110 (Connection timed out))
19:51:50DarkJesusWhat about an ipod style music quiz?
19:52:02preglowyou can play them without listening to music
19:52:24preglowluckily, on our other targets, it's not that much of a problem
19:53:07 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:53:26scott666_never was on my archos
19:53:45linuxstbThe archos doesn't use the CPU for audio decoding.
19:54:13 Quit Acksaw (Read error: 104 (Connection reset by peer))
19:54:23 Quit thegeek ("( :: NoNameScript 4.02 :: )")
19:54:27linuxstbBut hopefully once we start using the ipod's second CPU, it shouldn't be an issue on the ipods either.
19:55:09linuxstbjulius: Glad to hear you've at least got pin 17 working. Have you found out which of the three commands enables it?
19:56:46preglowamiconn: didn't you have some ideas on how we should go about moving stuff to the other core, btw? :>
19:57:50linuxstbjulius: In the IPL code I pointed you to yesterday, there is some code to init ser0 and ser1 on the 4g. You should try that. It's the last modification on this page:
19:58:36 Join Acksaw [0] (
19:58:40 Join romanos_melodos [0] (
19:58:42juliushi linuxstb I'm glad to see you!
20:00:13juliuslinuxstb, ok, I will try that one too, btw: I didn't test the tree command to se which one enables which pin, but I will do so after I got pin 11 to work
20:00:33romanos_melodosr there any other games i can play in rockbox except from the ones already provided?
20:03:19Presenceromanos, I play "spin the bottle" at parties with the ipod, but thats not really dependant upon rockbox.
20:04:04linuxstbromanos_melodos: Not really. Almost every plugin that gets written for Rockbox gets added into CVS.
20:04:47romanos_melodosare there gonna be any new features in the new release?
20:04:58romanos_melodosi'm talkin' about verion 3
20:05:05 Quit Farpenoodle (Read error: 110 (Connection timed out))
20:05:06 Join Xerion [0] (
20:05:09linuxstbThere's a feature freeze...
20:05:25linuxstbThe idea is that no new features are added, but bugs are fixed in all the existing features.
20:05:37linuxstbWe then release and start again...
20:05:42amiconnpreglow: I think the idea was to run just a single thread on the second core. If we want more than one, we could run a second scheduler on the second core, and let create_thread() have an argument specifying the core.
20:06:04romanos_melodosyou're right linux
20:06:41amiconnBut I think the biggest problem will be the data exchange. Iiuc this has to be done through iram, unless we find a way to invalidate the data cache
20:06:47romanos_melodosi'm still having some problems during playback and got an annoying sound each time i'm skipping to another dir to hear a song
20:06:50preglowamiconn: perhaps running only the codec thread on the second core does make sense
20:06:57preglowamiconn: playback can stay on the first core
20:07:17 Join Lear [0] (
20:07:18preglowthen i _think_ the only transfer that needs to take place is the decoded data
20:07:23preglowand that is always in iram now anyway
20:07:25linuxstbWe also have the issue of CPU boosting - the two CPUs share the same clock, but I think the COP can be run at half the speed of the main CPU.
20:07:44romanos_melodosare there 2 cpu's on ipod?
20:07:46amiconnCould we actually swap cores then?
20:08:05linuxstbYou mean run the UI on the COP?
20:08:18linuxstbI don't see why not.
20:08:19 Join DpK|Cronic [0] (
20:08:29linuxstbThey both have equal access to the peripherals.
20:08:41amiconnRun everything on the cop except the codec thread. Then we could throttle the ui to 1/2 whenever codec boosts
20:08:42DpK|Cronichi anyone here how can me install the source for rock box over WINDOWS DESKTOP REMOTe?
20:09:12linuxstbThat's assuming we can figure out how to run the COP at half-speed. I think I just read this in the PP product briefs.
20:09:23ep0ch_dsp should also be on the fast cpu?
20:09:34romanos_melodosare there 2 cpu's on ipod?
20:09:40linuxstbromanos_melodos: Yes.
20:09:45amiconnep0ch_: dsp isn't on a separate thread
20:09:47scott666_how does the apple firmware delegate it?
20:10:00amiconnpreglow: What thread is handling dsp?
20:10:04linuxstbscott666_: I'm not sure anyone knows for sure.
20:10:19scott666_ah, i see
20:10:19preglowamiconn: codec thread
20:10:34DpK|Cronichi anyone here how can me install the source for rock box over WINDOWS DESKTOP REMOTE? PLEASE :(
20:10:45preglowamiconn: exactly
20:10:50*scott666_ has no idea how much of the apple firmware has been reverse engineered
20:10:57scott666_DpK|Cronic: no
20:11:07romanos_melodoshow the heck did u guys managed to chroot inside ipod and other mp3 players anyway.. :P
20:11:21linuxstbscott666_: I think IPL mainly work with the bootloader and diagnostics mode code. That's a lot more managable than retailos.
20:11:23DpK|Cronicwuah scott666_ :( you are bad ^^ my cys does not start !
20:11:48ep0ch_did anyone test boost ratio for .mod on ipod?
20:12:26amiconnpreglow: Uhm, does dsp also handle voice mixing? That might cause difficulties...
20:12:28DpK|Cronicscott666_ or can you give me a step by step documention from yoU? ^^ also in qry..
20:12:38preglowamiconn: no, playback engine does that, i'm afraid
20:12:42ep0ch_i see that .mod playback has some icode in it so i may do another test with and without linuxstbs no icode patch
20:13:23amiconnpreglow: Why afraid?
20:13:36romanos_melodosi wanna lay back and smoke pot
20:13:45klrspzmeeeeeeeeeee too
20:13:53preglowamiconn: because that means data sharing
20:13:57ep0ch_preglow: going to commit the libmad opts sometime?
20:14:02preglowep0ch_: yeah, today
20:14:23amiconnpreglow: Didn't you say output buffers are in iram?
20:15:23 Join xmixahlx [0] (n=xmixahlx@
20:16:25 Quit romanos_melodos ()
20:16:29preglowamiconn: not after dsp
20:16:43preglowamiconn: then they're stuffed in the pcmbuf
20:16:52ep0ch_hmm if core 2 speed is half of core 1, then core 2 would be sh!t slow when unboosted
20:16:53amiconnAnyway, there's another problem with voice & dual core operation:
20:16:59 Join thegeek [0] (
20:17:48amiconnToday, codec & voice are interlocked and swapped in & out as necessary. This is no longer possible (or at least very undesirable) with dual core operation,
20:18:40amiconnep0ch_: The cores are equally fast, but the second core can be throttled
20:18:51 Join darkless [0] (n=darkless@
20:19:59DpK|Croniccan anyone help me with the installation of cygwin?
20:20:45scott666_Mr. can
20:22:12DpK|Cronicwhich paket i needed?
20:22:14juliuslinuxstb, I inserted the whole block of code from here: see Line 362, without the if statement, but I still don't have any power on pin 11. do you have any more ideas?
20:23:04DpK|Cronicare this the paket?
20:23:05 Join tabh120 [0] (
20:23:11preglowwe'll need to experiment a great deal with the cache to find out how it works
20:23:18preglowif it's write-through or what
20:24:13DpK|Cronicscott666_ where is the paket how i need to install ?
20:25:08scott666_what the hell is a paket?
20:30:12klrspzpacket as in set of instructions
20:30:14klrspzi'm sure
20:30:43scott666_in that context though?
20:31:21klrspzfuck if i know
20:32:20 Quit tabh120 ("Chatzilla 0.9.73 [Firefox]")
20:33:19*goffa thinks he should get nap-time at work after lunch
20:33:30goffakinda like kindergarden
20:35:23goffai should be awake after riding my bike uphill against the wind
20:37:56DpK|Cronic" Set local Packet Directory" scott666_ that was i mean
20:40:48 Join Nepo [0] (
20:40:56Nepohill ^^
20:41:01Nepohi@all ^^
20:41:38 Join mirak [0] (
20:41:51DpK|Cronicwhere is the PACKET how i need to install cygwin? :(
20:42:30 Join charkins [0] (n=casey@pdpc/supporter/active/charkins)
20:42:36preglowthis is weird
20:42:42lostlogicgoffa: I always tell them we shoul dhav enap time after work
20:42:46lostlogicwe all enter food coma any way
20:42:49PaulJamDpK|Cronic: in about 10 minutes i can try to help you
20:43:02preglowsomeone skilled in coldfire recording, help from, request
20:43:02 Quit Nepo (Client Quit)
20:43:11DpK|Cronickk :) PaulJam can you give me german supp on qry? or dont speak german?
20:43:33goffayeah... we can't get anything done anyway :)
20:44:19*goffa thinks its the cold turkey (tryptophan)
20:44:19PaulJamDpK|Cronic: yes, i'm from germany
20:44:32DpK|Cronicnice one kannst mir qry support geben?
20:45:01 Join Fitzsimmons [0] (
20:45:41 Quit Fitzsimmons (Client Quit)
20:49:06 Join Fitzsimmons [0] (
20:50:37 Nick Acksaw is now known as Acksaw|Coursewor (
20:54:40 Quit kimberly ()
20:55:43preglowbluebrother: ok, so we should use \setting instead of emph or ``'' now?
20:55:58preglowor is that just for describing menu entries?
20:56:05bluebrotheryes, thats the intention.
20:56:11preglowformer or latter? :>
20:56:16bluebrotherthe first ;-)
20:56:25preglowi'll start changing some stuff in what i've written
20:56:38bluebrotherso we have all menu entries the same style −− atm it's kind of mixed.
20:57:05bluebrotherI was just about adding this to the LatexGuidelines wiki page ;-)
20:59:48*goffa wants some of what the people at nintendo are smoking.. must be good stuff
21:01:05 Join myth1 [0] (
21:01:35DarkJesusYou lot don't like the name?
21:01:40DarkJesusI think it's fine
21:01:50ep0ch_wii fine?
21:02:22ep0ch_its stupid and they're having a laugh
21:02:45goffaits just silly
21:03:54preglowi rather think they think they're being really clever now
21:04:05preglowcan't say i like it
21:04:17DarkJesusIt's gonna be a brilliant console anyway, so who cares?
21:04:33goffa*puff* Further, Nintendo said, the written name is meant to visually conjure up the image of two people standing together. *exhale*
21:04:49preglowit fails in that
21:04:53 Join qwx [0] (
21:05:10preglowunless their shiny logo is going to typeset the is as people standing together
21:05:46DarkJesusWho cares?
21:06:30 Quit DarkJesus ("Leaving")
21:07:22goffai just don't think that the name is going to sell the box
21:07:29goffai think revolution was a better name
21:07:38preglowbluebrother: perhaps we should small case the string before feeding it to \textsc
21:07:52preglowi don't like the capped first letter
21:07:57goffai think they are going to sell a lot of them though... and so the new buzz word will be wii
21:08:23preglowwell, it sure looks like it'll be fun
21:08:49preglowand if they stick through with making a snes emu on it, then hell, i just might be one for kicks
21:08:53preglowsnes > *
21:09:01goffa"whee" would have been better but i guess that doesn't put a mental picture of people standing in line to get one :)
21:09:12bluebrotherpreglow: we can do latex this for us.
21:09:26bluebrother*let do
21:09:37preglowbluebrother: yeah, that's what i meant, i just don't know how
21:10:01bluebrotherso if we decide to switch to something else (maybe italics or something else) the spelling should be like in the device itself.
21:10:19bluebrotherafaics this is every first letter of menu entries is capitalized.
21:10:56bluebrotherI need to look how we can lowercase everything, but I think we should wait some days and wait for the reactions to the new style ;-)
21:10:59*petur wonders why there's an ipod bugreport in the 3.0 buglist
21:13:16preglowbluebrother: i like it, i just think the first letters should be small caps too
21:14:20preglowwhich means we need to lowercase them before sending them to \textsc, but yeah, no hurry
21:16:43 Quit qwm (Read error: 110 (Connection timed out))
21:17:01 Join jbauman [0] (
21:17:34 Join herz42 [0] (
21:21:43 Quit ep0ch_ (Read error: 110 (Connection timed out))
21:22:16peturhmmm I just installed a voice file and now it hangs at boot - AAARGH
21:24:39 Quit myth1 (Remote closed the connection)
21:26:49 Join qwm [0] (
21:27:10 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
21:27:53***Saving seen data "./dancer.seen"
21:28:13tucozbluebrother, your new commit looks good
21:28:46bluebrotherI decided to commit it so others could have a look on it ;-)
21:29:05tucozI would be surprised if people do not like it
21:29:40bluebrotherhehe ...
21:30:14bluebrotherI'd vote also for removing paragraph indents and instead using a greater parskip. But this seems to be somewhat controversical.
21:30:33preglowi like parindent
21:30:36klrspzbig update?
21:30:50bluebrotherbtw, I wasn't aware of the capital "H" modifier for floats. Really nice.
21:31:06bluebrotheras I said, its somewhat controversical ;-)
21:31:25tucozYes, it seems to work better than the old way. I am looking at the nonfloat package at the moment
21:33:03tucozwhat does parskip do? more white space between the paragraphs?
21:33:42preglowit just uses parskip instead of parindent to mark paragraphs
21:33:50tucozI see. I have no strong opinion in that camp. What looks good works for me :)
21:34:01preglowthe usual way is for parskip=0 and a parindent of around one em
21:34:15preglowbut i guess i can forgive parskip paragraphs in a manual
21:34:29preglowin other literature i'd shoot however used parskip :)
21:34:43bluebrotherI have this in my local build (to get a vision of it and me likes this better for the manual):
21:34:53tucozI'll try it
21:35:05 Quit charkins ("Leaving")
21:35:55bluebrotherpreglow: I agree with you on other literature. But for the manual I find the parindents somewhat strange.
21:36:06preglowwouldn't you use one entire baselineskip for the parskip, btw?
21:36:17 Join charkins [0] (n=casey@pdpc/supporter/active/charkins)
21:36:31preglowone should always skip a whole line at a time
21:36:42bluebrotherI played with it a bit and liked this spacing most as it's otherwise really much space.
21:37:06bluebrotherwe're not on typewriters anymore, so no need to hold on lines ;-)
21:37:09 Quit SereR0KR (Read error: 104 (Connection reset by peer))
21:37:24tucozLooks good
21:37:26 Join SereR0KR [0] (
21:37:29preglowbluebrother: ah, no, it's just overall a general typographic rule
21:37:37 Quit qwx (Read error: 110 (Connection timed out))
21:37:48scott666_DpK|Cronic: heres the file you were asking for yesterday:
21:37:49preglowtry to keep material baseline aligned so it doesn't look weird when you're viewing several pages at once
21:38:04preglowwhere the hell did that 'material' come frmo
21:38:28preglowbut anywho, it's just nitpicking
21:38:33tucozbeen drinking again, eh ;)
21:38:51preglowoh, i wish
21:38:56 Join aaronslife [0] (
21:38:57preglowi almost need to to handle this dreary weather
21:39:20tucozFor a change, it's pretty nice here atm
21:39:35 Quit hotwire (Success)
21:39:57aaronslifehello, does anyone know when the iPod Video will have better charging capabilities with rockbox?
21:40:19 Join hotwire_ [0] (
21:40:21 Nick hotwire_ is now known as hotwire (
21:40:27klrspzhow often should i update my binary?
21:40:39tucozbluebrother, what do you think of some guidelines when referring to a key-table. Sometimes it is Keys: followed by the table
21:40:56tucozand sometimes we simply show the table
21:41:05bluebrothersorry, need to go for some time. See you later.
21:41:12tucozSee you
21:41:39aaronslifeklrspz I'd say every month or week. depending on if your really into getting new features.
21:41:40tucozklrspz, I would recommend: As often as you like
21:41:58tucozI update almost daily :)
21:42:09klrspzyeah iwas thinkin of doin that
21:42:46 Join pixelma [0] (n=M_@
21:42:51aaronslifequick question... whats rockbox programmed in?
21:43:05preglowanyone know what the idle timeout triggers on?
21:43:09tucozBut if you are happy with the current state, then just keep an eye on the front page for fixes to bugs
21:43:15aaronslifethx klrspz
21:43:26 Join qwx [0] (
21:43:33tucozor target specific changes
21:43:34aaronslifesee you guys later!
21:43:38 Quit aaronslife (Client Quit)
21:43:57 Join sharpe [0] (
21:44:13klrspz060426-2123 is the latest bleeding build?
21:44:59 Quit psi (Nick collision from services.)
21:45:07 Join psii [0] (n=psi@
21:45:09klrspzyeah that's what i got
21:45:25tucozthose are the latest bleeding edge builds
21:45:40klrspzk, guess i'm up to date then
21:45:59klrspzif i can finish installing this mod on this guy's ps2 tonight, i'm going to build me some wps'
21:46:45tucozAnd especially now, there have been lots of fixes to the playback-system. So, a bleeding edge could potentially be worse than your previous build
21:48:31scott666_i think rockbox could use micro-releases
21:48:42scott666_just daily build that are 'pretty good'
21:48:59scott666_so you can get one of those instead of the bleed edge with loots of new bugs added
21:49:22scott666_but maybe thatd just be more trouble than its worth
21:49:39klrspztucoz, i saved my previous binary just in case :)
21:50:12tucozklrspz, I think the last 30 or so builds are available on the server
21:50:49klrspzi was seeing if i could rename the files on the fly then reboot and use the binary
21:51:54klrspzand i can
21:52:32klrspzso while i'm stuck in traffic after work (a 20mi drive takes me 1.25 hours in this shitty area) i can switch binaries if it bails out on me
21:53:14preglowlibmad patches commited
21:53:21preglowplease try and give feedback
21:53:28klrspzyay for smoothie day
21:53:32 Nick klrspz is now known as KlrSpz (n=klrSpz@
21:54:06 Join webguest84 [0] (
21:54:15 Join qwx_ [0] (
21:54:21*petur swears at the voice stuff
21:54:53 Quit qwm (Read error: 110 (Connection timed out))
21:55:29KlrSpzyeah the voice thing just kept saying "x" to me
21:55:31KlrSpzover and over
21:55:34KlrSpzit was kinda creepy
21:55:41 Quit lostnihilist (Read error: 113 (No route to host))
21:55:45webguest84my browser shows the latest bleeding edge build to be 26 Apr 20:22 is that the latest, someone mentioned 21:23
21:55:49 Join hardeep [0] (i=hardeeps@otaku.freeshell.ORG)
21:56:48KlrSpzi dl'd 060426-2123 from the link
21:56:52KlrSpzbut it was yesterday's date
21:57:12peturmy bleeding edge is just a minute ago :P
21:57:16KlrSpz2006-04-27 19:53 seems to be the latest timestamp on my browser
21:57:17webguest84ah ok, thanks
21:57:26KlrSpzerr wait, that's not built yet
21:57:37webguest84I see there's one in the pipeline as we speak
21:57:43KlrSpzBuild expected to complete around 19:59:45 (in 6mins 35secs)
21:57:54KlrSpzyeah, i'm gunna grab that one in a bit
21:58:16KlrSpzi was looking at the diff, not sure if i can be much help now... looks like a bunch of magic numbers
21:58:16 Quit PaulJam (".")
21:58:52KlrSpznot THE diff, but A diff, sorry
21:59:05 Part webguest84
21:59:15 Nick jamezz is now known as slimeball (
21:59:57amiconnpreglow: I've finally found the spot of the jumptable size bug in the gcc sources. Now I only need a less hackish way of fixing it...
22:00:18KlrSpzone thing i noticed, is that when i change tracks, that BEEP sound only is audible like 1 out of 7-10 times i click the button...
22:00:46amiconnpreglow: Do you know by chance which macro tells me the minimum required alignment in the .text section?
22:00:58 Join JBGood [0] (
22:01:40 Join damaki [0] (
22:05:17peturamiconn: had a weird freeze at boot after I copied a voice file, but am unable to reproduce :( know anything about this?
22:05:36 Quit qwx (Read error: 110 (Connection timed out))
22:06:15amiconnHmm. Do you have resume on startup enabled, or manually tried to resume before rockbox was fully booted?
22:07:10peturused bootloader usb to copy the file over, then let it boot but got an empty screen and hang
22:08:16 Join PaulJam [0] (
22:08:25peturwanted to try the bugreport about recording and voice and ended up experiencing this
22:08:28amiconnThen I don't know. However, I do know that there are still quirks like that in playback and voice
22:09:51lostlogicpetur: there are still a few people having problems with bootup with a voice file present, but most people don't... I hate bugs like that.
22:10:40*lostlogic hopes the bugs in voice and playback resolve themselves while he's gone for 4 days starting now
22:10:49peturmust be voice init because it froze without me having anything voice related enabled
22:11:13 Quit jbauman (Read error: 110 (Connection timed out))
22:11:14*petur wishes lostlogic a nice time in Paris
22:11:51peturlostlogic: beware, sometimes it's cheaper to order wine than water there!
22:12:02 Join obo [0] (
22:12:09peturand don't expect much of the food
22:12:10Mikachulostlogic: have fun
22:14:01lostlogicpetur: yeah, I've heard about the lack of free water in restaurants... so I'm bringing my 3 litre hydration pack.
22:14:04lostlogicMikachu: thanks :)
22:14:16lostlogiclater rockbox :)
22:15:01peturlostlogic: restaurants normally give tap water free - not always
22:15:29peturoh, and drive three hours extra and visit me :)
22:17:54 Quit SereR0KR ("XChat Aqua")
22:18:04preglowlostlogic: wha, you're goin away?
22:18:35lostlogicyes, as soon as I finish shutting down and packing up around my workspace, I'm going to Paris, with no computers for the weekend.
22:18:44preglowparis, of all places
22:18:52preglowi thought you had seen europe now :)
22:21:13*petur pours himself a cognac
22:21:30 Join saratoga [0] (
22:21:43saratogahay, i just noticed the new libmad changes
22:21:53saratogait looks like an amazing amount of work was done on it
22:22:14saratogaare any benchmarks available for arm or coldfire for the new changes?
22:22:22preglow19% gain for coldfire
22:22:27preglow10% less boost at 320kbps for arm
22:22:39saratogahow much boost is needed?
22:22:45preglowdepends on the file
22:22:50preglow320kbps on arm = 62% boost
22:22:53preglowway too much still
22:22:58preglowbut it used to be higher
22:23:01saratogayeah thats a lot
22:23:15preglowon coldfire it's very fast now
22:23:49saratogawhat makes coldfire so much faster?
22:23:54preglowgod knowsn
22:23:58preglowa more sane arch
22:24:15saratogai was looking into how loads are handled on the arm7
22:24:17saratogaits retarded
22:24:21saratogajust to save a few transistors
22:24:26saratogabut still
22:24:31preglowi'm not too fond of load-store
22:24:38saratogais coldfire all that much better ?
22:24:44preglowwell, i think it's slightly better
22:24:50pregloweven though you do end up with more code most of the time
22:25:04saratogathats why i assumed, but it makes such a huge difference running the same c code?
22:25:12saratogais there a lot of 68k assembly or something
22:25:16preglowwell, apparently
22:25:30preglowsometimes the same code is faster on arm, sometimes not
22:25:32preglowhard to see a pattern
22:25:42preglowbut dsp seems to be faster on coldfire altogether, at least
22:26:00saratogawell arm can't pipeline loads or stores
22:26:11saratogaso its going to lose anytime you have to work on a large data set
22:26:23saratogasince you basically stall on every load and store
22:26:30*amiconn realises that we only use -O for the core on all archs
22:26:41 Join adiamas [0] (n=adiamas@
22:27:07amiconnMaybe we should use -O2 instead (except archos of course, where we'll most likely switch to -Os)?
22:27:30 Nick thesonork is now known as thesoffork (
22:27:46 Quit xmixahlx ("blah blah blah")
22:28:48 Quit DpK|Cronic ("get satisfied! • :: ««« (Gamers.IRC) »»» ::")
22:28:58saratogadoes O2 make a big difference?
22:29:10adiamasi have the cvs pull from 4/11 and my volume seems weird. on the archos rec 20 for the vol to be resonable I have to bring the setting down to -14. Anyone having this?
22:29:35 Nick thesoffork is now known as thesonork (
22:29:43amiconnadiamas: Seems you didn't update for a loooooooooong time?
22:29:56adiamasto say the least :)
22:29:57amiconnVolume uses dB for quite some time now
22:30:09preglowamiconn: worth a shot
22:30:12preglowamiconn: i thought we got errors
22:30:21preglowsaratoga: yeah, the loads seem to kill a lot of the advantage
22:30:24adiamashehe that would make more sense then..
22:30:28*amiconn is building with -O2 for H300 right now
22:30:29preglowalso, the 64 bit multiplier seems to be slow
22:30:31preglowhooray for that
22:30:33adiamasstill confusing for me since I don't know dB lol
22:30:56 Join webguest02 [0] (
22:31:10saratogaIt looks like apple is ditching portal player
22:31:18saratogaso maybe we'll get arm9 on future ipods
22:31:21*goffa guesses new patch will improve battery life?
22:31:28saratogathat has sane pipeling
22:32:05webguest02hello anybody
22:32:07saratogaalso, what 64 multiplier ? ARM7 doesn't have one :)
22:32:13tucozhello webguest02
22:32:20 Quit webguest02 (Client Quit)
22:32:32 Join webguest02 [0] (
22:32:55 Part PaulJam
22:34:21preglowsaratoga: it does?
22:34:28preglowsaratoga: sure arm7 does
22:34:32preglowsaratoga: look up smull and smlal
22:35:38Mikachurandom filenames are showing up in apple's about screen
22:35:42saratogai don't remember that, have to look it up
22:35:51Mikachuwhat is the title supposed to be in that screen?
22:35:57 Quit webguest02 (Client Quit)
22:36:10preglowsaratoga: i use them all the time, so at least i know they exist :)
22:36:41saratogaare you doing a lot of assembly code for the ipod?
22:37:10amiconnpreglow: warnings with -O2 ...
22:37:16amiconnplayback.c:694: warning: dereferencing type-punned pointer will break strict-aliasing rules
22:37:29amiconnThat's in fact the only one, twice for the same line
22:38:05amiconnrockbox.iriver size increases by ~5%
22:39:00 Join lee-qid [0] (
22:39:09 Quit Ribs ("Leaving")
22:39:20 Join PaulJam [0] (
22:40:09preglowamiconn: i get a zillion of those for doom with 4.1
22:40:25preglowamiconn: can you think of any way to measure performance increases?
22:40:53amiconnNot really.
22:41:37preglowperhaps the lcd benchmark plugin?
22:41:40amiconnBoost ratio is a rather coarse measure
22:41:41preglowor is that all asm, these days
22:41:56amiconnThe data transfer is all asm on H300
22:42:02preglowlostlogic: does the null output patch still work? :>
22:42:21amiconnThe upper layer is C, but I doubt it has much influence
22:42:38amiconnThe drawing routines might profit from -O2
22:42:51preglowbut yeah, the codecs already are -O3
22:42:52preglow-O2 <-
22:43:04amiconnYes, as well as plugins
22:43:13 Join Purgatory [0] (
22:43:22amiconnThat's why I know that the sh-elf-gcc bug also hits with -O2
22:43:23preglowi can't think of any testable scenarios where the core has much to say
22:44:14amiconn-O2 build crashes within a second
22:44:26amiconnIBC: at 32F001C2
22:45:20Purgatoryi've got a problem with installing on my nano
22:45:46amiconnThis is codec ram...
22:45:54Purgatorywhen i type: ipodpatcher 0 in cmd
22:46:11Purgatoryit says: Drive is not an ipod, aborting
22:47:10 Join ProgramZeta [0] (
22:47:31Purgatorygot it :)
22:47:35Purgatorymy bad
22:47:57 Quit charkins (Remote closed the connection)
22:50:54 Quit freqmod (Remote closed the connection)
22:52:44preglowamiconn: weird
22:52:56amiconnpreglow: gcc...
22:54:21 Quit Lear ("Chatzilla 0.9.72 [Firefox]")
22:57:26bluebrothertucoz: I'm back for some time ...
22:57:29scott666_hooray, i just complied rockbox
22:58:56bluebrothertucoz: I was thinking we should generally not write "the following table" but refer to it with either its caption numbering or let the user find the table itself.
22:59:17bluebrotherwhich would imply giving all tables reasonable captions.
22:59:48 Join lostnihilist [0] (
23:04:02 Nick scott666_ is now known as scott666|work (
23:09:00 Join RotAtoR_ [0] (
23:09:18amiconnpreglow: The optimised libmad reduces boosting of 320kbps mp3 files to 4% on H300. With Hold enabled, it's 0%
23:09:42preglowwhy would hold affect it?
23:09:43Purgatoryrockbox rocks!
23:09:59amiconnpreglow: Hold disabled the ADC read from the PCF
23:10:12preglowwhat the hell? that requires so much?
23:11:12 Quit RotAtoR (Nick collision from services.)
23:11:16 Nick RotAtoR_ is now known as rotator (
23:11:17amiconnWell, it's a bit-banging i2c driver, called HZ times per second
23:11:40preglowstill sounds like a lot to me
23:11:42amiconnThat's _after_ my optimisation work, imagine what the load was before...
23:12:02goffai don't know who gave me the tip to get out of wps and go into a menu to save battery, but thanks
23:12:08preglowand why bit-banging? weren't these morons clever enough to use a pin that's got a hardware interface?
23:12:11 Join momo3 [0] (
23:12:20goffaplayer has been playing for an hour (after it shut off)
23:12:27amiconnpreglow: Nope. i2c in irivers is quite strange
23:12:50 Quit momo3 (Remote closed the connection)
23:13:03preglowBagder: these hex address writes of yours makes us look like ipl!
23:13:05amiconnIn fact I was wrong. The difference in boosting is caused by the backlight being on vs. off, i.e. lcd updates
23:13:09BagderI know
23:13:25BagderI'll search/replace them to something nicish soon
23:13:28 Join TCK [0] (
23:13:40preglowamiconn: makes a ton more sense
23:13:50*goffa wonders what the least cpu intensive wps is
23:13:57goffai thought engineer2 would be pretty low
23:14:00preglowgoffa: an empty one
23:14:15amiconnStill, the pcf i2c draws considerable power. About 3..4% at normal & high frequency
23:14:27goffadoes next track eat up that much cpu?
23:14:37goffaor playing next or whatever?
23:14:43Bagdergoffa: make a patch that stops updating everything when the LCD/backlight is off to compare
23:14:55 Part Purgatory
23:15:15goffaBagder: i'll do that right after i learn c :)
23:15:28Bagderah, that'll be tomorrow morning then? ;-P
23:15:47amiconnBagder: What do you think about the switch path to gcc 3.4.x and -Os for archos?
23:15:53amiconnImho we have 2 options
23:16:49goffaBagder: uh... doubt it
23:17:28goffai think i may have sams learn c in 24 hours at home too
23:17:42goffamaybe i'll skim that and learn it in 4... :P
23:17:50amiconn(1) The solid & slow way. First, require a patched gcc 3.3.x. When all build servers are upgraded, introduce -Os, and check for patched gcc in configure. Finally, require a patched gcc 3.4.x
23:18:05 Quit phaedrus961 ("Leaving")
23:18:41goffathat is a good idea though... not update when lcd is off
23:18:50Bagdergoffa: yes it is
23:18:58Bagderbut somewhat tricky
23:18:59goffawould save a TON of battery
23:19:33amiconn(2) The fast way. Start with few build servers: Install patched gcc 3.4.x, update configure with the check, and add -Os. Disable sh builds for all servers which are not yet upgraded
23:19:47preglowi say (2)
23:19:50preglowless time wasted
23:19:51BagderI say (2)
23:19:54 Join ashridah [0] (
23:20:05amiconn(1) has the advantage of a smooth transition, but requires an additional step of patching & installing gcc 3.3.x
23:20:27amiconn(2) is faster, but requires upgrading gcc in a short timeframe
23:20:29Bagderso why is that smooth?
23:20:52Bagderyou still need to patch and rebuild gcc
23:21:00 Join ep0ch_ [0] (
23:21:10amiconn(1) is smooth because all build servers can continue to build for archos, upgrading one by one
23:21:25goffaBagder: is your lcd/wps idea in the feature request yet? (probably is but figured i'd ask)
23:21:39amiconnThe problem is that we can't switch to -Os without the patch and that we can't swotch to gcc 3.4.x without -Os
23:21:46Bagderamiconn: I thought you said (1) would "require a patched gcc 3.3.x". how can any build server continue then?
23:22:10ep0ch_is everyone elses ipod hard disk dog slow?
23:22:14Bagdergoffa: I don't think it is, but we've discussed it a lot so it is "common knowledge" among the devs
23:22:37amiconnBagder: (1) is done in steps. The first step would be to install a patched gcc 3.3.x. At this point it's not yet required because we don't use -Os yet
23:22:48Bagdernow I get it
23:22:52amiconnOnly after all build servers are updated, we would introduce -Os
23:23:06amiconnThen we would see the first benefit
23:23:10goffathere was that guy in the forum looking for something to do... maybe i should suggest it to him...
23:23:19amiconn(recorder v1 binary size reduced by ~8KB)
23:23:34muesli__damn it..last experimental build crashes after every song
23:23:42BagderI still say (2)
23:23:45preglowand that's why it's experimental!
23:23:48amiconnAfter that, the build servers would be upgraded to a patched gcc 3.4.x, which gives another ~4KB
23:24:06crashdpreglow: has those libmad patches been submitted yet?
23:24:19preglowcrashd: yeah
23:24:21Mikachuamiconn: how much did my inline stuff give?
23:24:24muesli__preglow i wouldnt use it..but its the only way to get my h3ll lcd working properly :o
23:24:43amiconnMikachu: Ahem, don't remember. Does it still apply?
23:24:48Mikachudon't know
23:24:53goffai threw the idea at him.. we'll see what happens (if he still checks the forum)
23:25:02 Join phaedrus961 [0] (
23:25:27Mikachuamiconn: probably not the hunks in playback.c :)
23:25:52 Quit macdonalder (Read error: 104 (Connection reset by peer))
23:25:58amiconnBut these aren't relevant for archos (yet)
23:26:28Mikachu14:05 <amiconn> your patch saves 496 bytes on archos recorder
23:26:56amiconnHmm, not bad as another extra gain...
23:26:57Bagdernow there's a man in control of his logs! ;-)
23:27:12*amiconn should fetch the latest logs
23:27:20amiconnWay more convenient for searching
23:27:31preglowMikachu: waht pathc?
23:27:35Bagdermaybe I should setup the logs for rsyncing
23:27:47Mikachupreglow: (outdated)
23:27:55amiconnBagder: Not necessary, Firefox + DownThemAll is sufficient
23:27:56***Saving seen data "./dancer.seen"
23:27:57Mikachubut you get the general idea
23:28:16Bagderamiconn: well, rsyncing is easier for automatic jobs
23:28:26preglowjust a bunch of inlines
23:28:31Mikachupreglow: yes
23:28:37preglowwhy does that make the binary smaller? :>
23:28:44Mikachupreglow: those functions are only called from one place
23:29:46amiconnMikachu: Perhaps the gain is much less when using -Os, since gcc might figure out these things itself
23:30:33Mikachui don't know, there is a -finline-functions, but it's only enabled at -O3
23:31:17amiconnHmm, worth a test
23:31:53amiconnThe only thing holding me back from providing the gcc patch is that I want to do a proper fix. :/
23:32:16amiconnI know where the fix has to go, just not which macro (or function) to use...
23:33:18amiconnMikachu: The tagcache part probably won't apply as well
23:34:21preglowamiconn: tried asking the gcc people?
23:34:29amiconnhmpf :/
23:35:22*amiconn much prefers the helpful people at #rockbox over the gcc people
23:35:28preglowyes, indeed
23:35:42preglowbut we don't know much about gcc :P
23:36:22amiconn\: fpmh
23:37:30preglowi'd truly love to try to understand it one day, but as of now, no
23:37:43sharpewhat about making the, rbcc? rockbox c compiler? ahah... i'm so funny... :(
23:38:01preglowwe might end up with that
23:38:12preglowi know of a couple of coldfire based patches we might want to include as well
23:39:30amiconnHah, the bug is still present in 4.10, and it's in a generic source file
23:40:00amiconn(although in a part that is #ifdef CASE_VECTOR_SHORTEN_MODE, which in turn is only defined for SH)
23:40:03 Join DarkJesus [0] (
23:43:23ep0ch_Slasheri: are there any notes on dircache? its not working for me anymore and i was wondering what the limits of it are...
23:43:27 Quit KlrSpz ()
23:44:11preglowlimits should be ample
23:44:13preglowcheck the debug screenm
23:44:45 Quit saratoga ("CGI:IRC (EOF)")
23:45:26ep0ch_aint working for me, says cache isn't initialized, last size=458132 B, scanning took 0s, entry count = 8152
23:46:11dwihno_dir cache + on boot resume = lots of buffer underruns on ipod
23:46:17preglowthat's quite a number of files
23:46:21 Quit julius ("Leaving")
23:46:29ep0ch_yeah sids and mods init :)
23:46:42 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
23:46:46preglowthose will _kill_ dircache, you know
23:47:04preglowrape it right up the ass
23:47:14ep0ch_i propose rockbox has a non dircache directory
23:47:19preglowperhaps some hybrid mode should be devised :P
23:47:29preglowthat's hardly KISS, though
23:47:37preglowso it really vote against it
23:47:49preglowunless slasheri tells me it's very simple to do
23:48:13 Quit ProgramZeta (Read error: 110 (Connection timed out))
23:49:37 Quit petur ("'night all")
23:50:11ep0ch_4 channel mods boost by 21% on ipod
23:50:37ep0ch_i though arm chips were super effecient :s
23:50:47preglowthat doesn't bode well at all
23:51:09ep0ch_there's something not right going on well at all with codec performance
23:51:19ep0ch_yeah that made sense
23:52:25ep0ch_there's a massive overhead with mp3 irrelavant of the bitrate
23:53:31preglowi know
23:53:39preglowi'm going to find out when i write the profile parser
23:56:51 Join KlrSpz [0] (n=klrSpz@
23:58:33preglowgod, i'm starting to hate arm
23:58:36*preglow misses coldfire

Previous day | Next day