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 2007-07-05

00:01:24 Quit Siltaar (Remote closed the connection)
00:01:58*amiconn wonders about bits 0..7 and 24..30 of 0x6007000 and 0x60007004
00:02:23gregjso, completly clean rockbox, all defult settings
00:02:27gregjlets see
00:02:34 Quit ender` (" To get as fewest unhappy people as possible, always bully the same ones.")
00:02:58gregjsame thing
00:03:19gregjjust right after the courtain goes up ( linuxstb_ will know when ) - sound is dead, and so are the keys
00:03:29*stripwax wonders if it could be a hardware revision thing - does that sort of info show up in debug menu for ipod photo?
00:03:38 Join LordPenguin [0] (i=4c675b5a@gateway/web/cgi-irc/
00:04:03gregjstripwax: it was one of the last , I bought it 2 months before video one was announced
00:04:15linuxstb_About the same time I bought mine...
00:04:24linuxstb_But your one is 20GB?
00:04:26gregjipod diagnostic
00:04:32gregjsrv mar 22 2005
00:04:40*gregj saw the 'other side of apple' :)
00:04:49 Quit ompaul (Client Quit)
00:05:04gregjhwVr 393220
00:05:13LordPenguinCould anyone help me with a problem that I'm sure you get all the time?
00:05:16gregjlinuxstb_: you know how to get to diagnostic mode?
00:05:28linuxstb_I surely do.
00:05:32gregjlinuxstb_: reboot, and keep middle+left until reversed apple doesn't show up
00:05:37gregjand go to syscfg setup
00:05:39gregjall info is there
00:05:43stripwaxLordPenguin - go for it
00:06:01gregjlinuxstb_: yes - it is 20GB one
00:06:10linuxstb_My model appears to be M9830. hwvr is the same as you.
00:06:21LordPenguinI have an iPod had .mp4 videos on it before I installed Rockbox. How do I view them?
00:06:32linuxstb_Using the Apple firmware
00:06:32 Quit jgarvey ("Leaving")
00:06:37TrueJournalsLordPenguin: Either convert them to mpg or use the apple firmware
00:06:40gregjthis hwvr is probably just 'g4-photo'
00:06:47LordPenguinOkay. Thank you.
00:07:11 Quit LordPenguin (Client Quit)
00:07:18gregjlinuxstb_: first number is a serial, but next two lines- god knows
00:07:57gregjlinuxstb_: if the number after M is hex, than a079 is obviously higher than 9830
00:07:57 Join jgarvey [0] (
00:08:01linuxstb_Was your ipod sold as a "Photo" or "Color" ?
00:08:13 Quit jgarvey (Read error: 104 (Connection reset by peer))
00:08:18stripwaxlinuxstb_ : - according to this, 9830 is 60GB ipod photo
00:08:23 Join jgarvey_ [0] (
00:08:24gregjlinuxstb_: I don't remember, to be honest
00:08:43gregjlinuxstb_: but I guess it was photo, I have the box somewhere - but about 2.5kKM away
00:09:08gregjon the back - it says: model no:a1099
00:09:15 Quit jgarvey_ ("Leaving")
00:09:31 Join jgarvey [0] (
00:09:57gregjlinuxstb_: I can show up in london over weekend, if you want - and have a look
00:10:09stripwaxSo that's the prev model, 9830 apparently replaced a1099(aka 9586)
00:10:11linuxstb_There is a small difference in the LCDs between the Photo and Color. (although I think some Color ipods have the older type of LCD). But I can't see how that could cause your problem.
00:10:23 Join thegeek [0] (
00:10:24stripwaxbut 60gb not 20gb according to that
00:10:32linuxstb_gregj: I won't be in London.
00:10:46linuxstb_You could join devconpub though next week.
00:11:11gregjnext week I am busy on sunday, but saturday - sure
00:11:16stripwaxdevconpub in london next week?
00:11:18linuxstb_As long as you stop complaining...
00:11:24*stripwax goes to check ..
00:11:40gregjmy spoken english isn't that good, so just for sake of that - I won't rant, promise :P
00:11:41preglowstripwax: yeap
00:11:45linuxstb_I think it's next wednesday/thursday.
00:11:49gregjwhat's devconpub :
00:11:53*gregj ducks
00:12:00preglowa ton of noobs having beers and talking rockbox
00:12:07stripwaxsounds like fun
00:12:53*linuxstb_ wonders how GodEater got on assessing venues this evening
00:12:53 Join fiftyfour123 [0] (
00:13:02gregjlinuxstb_: so what can I do, where can I try to tweak some code. I could start adding few sleeps for starters, to the other (non-video) thread
00:13:04 Join Xerion [0] (
00:13:08preglowwe'll probably find out in four hour's time
00:13:12preglowif we can decipher it
00:13:36gregjthe question about devconpub is not if I am going to rnt, but whether you guys are going to eat me alive or not
00:13:37 Quit pabs (Read error: 110 (Connection timed out))
00:13:52scorchegregj: buy beer for them and you should be fine
00:13:53preglowgregj: just show up two hours late or so, they'll be all mellowed out by then
00:14:16preglowprobably insisting on hugs before they'll even talk to you
00:14:32 Quit fiftyfour123 (Client Quit)
00:14:41gregjyou guys are going to do any hacking, or is it strictly getting pissed and hving fun kindof event
00:14:53preglowno hacking, i think
00:14:57gregjpreglow: tht sounds bit too queery, if you ask me
00:15:25gregj(the hugging thing)
00:15:32preglowhaha, not the hacking, you say?
00:16:14preglowbut no, a pub wouldn't be my venue of choice for hacking
00:16:16gregjwell, it would be good oportunity for me - to get to know the code better (and devels)
00:16:19preglowthe beer would be too disturbing
00:16:31gregjwe did hacking in gllug on beer events
00:16:33preglowit calling for me all the time
00:17:03gregjimagine whole downstairs level of pub just full of geeks, with tons of laptops
00:17:13gregjand beer just running in pipes downstairs
00:17:24preglowi could live with that
00:17:30gregjon one ocassion, we meet Z80 funclub
00:17:32gregjand what they do ?
00:17:50gregjthey show up in pub, with laptops, and (watch out!) Z80 - kind of laptops - hacking
00:17:56gregjwhat are the chances ?
00:18:04gregjbut it did happen
00:18:19stripwaxz80 or z88?
00:18:26gregjz80, spectrum
00:18:47preglowthose things that looked like oscilloscopes?
00:18:48gregjso they have spectrums, with HDs, and stuff
00:19:05gregjhere comes the youngster's question
00:19:15stripwaxZX Spectrum Orchestra is great. Playing a show in london on 16th Jul.
00:19:44*gregj had timex 2048
00:19:50*stripwax is an aged 48k hacker
00:19:55gregjspectrum comptivile
00:20:04stripwaxgregj - lol, that youtube advert is great
00:20:07preglowaged and matured
00:20:15gregjpreglow: too old to remember ? :D
00:20:33preglownever touched spectrums
00:20:37preglowc64 was my first encounter
00:20:39gregjand it was cheap, 180 quid
00:20:49gregjby today's standard obviously cheap
00:21:14gregjI wrote radio-telegraph for it, IM over CB :D
00:21:21 Join pixelma_ [0] (i=pixelma@rockbox/staff/pixelma)
00:21:51 Quit pixelma (Nick collision from services.)
00:21:51 Nick pixelma_ is now known as pixelma (i=pixelma@rockbox/staff/pixelma)
00:26:07gregjhmm, rockbox just crashed - not doing anything
00:26:13gregjon idle
00:27:19stripwaxgregj is your ipod stable on apple firmware?
00:27:26gregjrock solid
00:27:51 Join pabs [0] (
00:27:59linuxstb_Are you running a clean SVN build, or have you changed anything?
00:29:28 Quit thegeek ()
00:29:41 Quit PaulJam (".")
00:30:06 Part TrueJournals
00:31:07 Join krazykit_ [0] (
00:31:11gregjlinuxstb_: clean
00:32:06gregjwell, I was trying out some bits regarding serial port, but that code is not used at all - so I consider it clean.
00:32:08qwmuh oh what a beautiful day.
00:32:19gregjand certainly, I had problems before I started adding code anyway
00:32:56stripwaxgregj - want to try installing one of the official daily builds anyway?
00:33:19gregjgood idea
00:33:25amiconngregj: Just an idea, what arm-elf-gcc version are you using?
00:34:11amiconnThat might be the problem then
00:34:18gregjyou reckon
00:34:20linuxstb_4.0.3 is the recommended version.
00:34:29gregjokay, I'll try with 403
00:34:32stripwaxyou ever tried an official build or only built your own .. ??
00:34:50linuxstb_Yes, you could just download an official build and test that.
00:34:53gregjI started on official builds
00:35:07gregjbut than - trying to do anything, isn't easy :)
00:35:11amiconnStill, if rockbox behaves weird when built with 4.1.x, this should probably be investigated.
00:35:11linuxstb_But that was before amiconn's fixes?
00:35:23 Quit Rondom (Read error: 113 (No route to host))
00:37:15gregjlinuxstb_: yes
00:40:32gregjcan i be gcc 404 ?
00:41:10scorche<linuxstb_> 4.0.3 is the recommended version.
00:41:13gregjso far so good - much better
00:41:23 Join Soap_ [0] (
00:41:29*gregj kicks its own arse
00:42:24gregjok, I'll leave it on for a while - and if it was the compiler - than I am sorry, I owe linuxstb_ and few others quite few beer, and will get that even asap
00:42:44scorcheif you werent using the recommended, it liekly was
00:42:47 Join TrueJournals [0] (
00:44:00gregjwell, it advanced much further than it used to
00:44:03gregjso it must be the shit
00:44:05gregjlinuxstb_: sorry mate
00:44:17*gregj commits suicide
00:44:28*bluebrother calls 911
00:44:33*linuxstb_ calls 999
00:44:34 Quit krazykit (Read error: 110 (Connection timed out))
00:44:41*petur calls 100
00:44:42gregj911 doesn't do much in UK
00:44:46*gregj calls 112
00:44:49*bluebrother remembers it's not 911 here
00:45:40*gregj notices that audio thread is quite complicated - I am trying to figure it out :)
00:46:19JdGordongregj: I wouldnt bother, we are workling on a replacement atm
00:46:36gregjwhere is it ?
00:46:46bluebrotherJdGordon: the complete audio thread? I thought it was only MoB.
00:46:57JdGordonbluebrother: it will end up being both
00:46:58linuxstb_gregj: Basically it tries to keep a ringbuffer full of PCM samples, which is emptied via either DMA or a FIQ handler. (I'm assuming you're talking about the mpegplayer audio thread)
00:47:20bluebrotherhmm. So you basically also replace the playback engine?
00:47:21Nico_PJdGordon: not sure I'm up to rewriting the whole of playback.c :p
00:47:32JdGordonyou wait.. alot of it will be
00:47:48gregjlinuxstb_: of course mpegplayer
00:47:58JdGordongregj: oh, in that case ignore me
00:48:00Nico_PJdGordon: seen my latest changes ?
00:48:05*bluebrother remembers that playback unification thingy and curiously looks around
00:48:22JdGordonNico_P: just the log, i dont tihnk ill have time to play today though
00:48:36 Join SirFunk [0] (
00:48:54*gregj gets back to the idea of adding features to 'metronome' plugin
00:48:57*gregj ducks
00:49:12Nico_PI'm not really sure about what to do next... try to integrate the code to playback.c or experiment with metadata in the test app
00:50:17linuxstb_You could create a multi-threaded test plugin.
00:50:43gregjlinuxstb_: rock solid so far
00:50:50gregjlinuxstb_: test plugin ?
00:50:51Nico_Plinuxstb_: that sounds like a good idea
00:50:59linuxstb_gregj: I was talking to Nico_P
00:51:18JdGordonNico_P: or fix the test app to use mutpli threads...
00:51:38linuxstb_gregj: Glad that Rockbox is working for you though... Now you can start complaining about genuine issues ;)
00:51:49stripwaxlike how the SMP sucks ;-)
00:51:53Nico_PJdGordon: sounds like too much work compared to the plugin idea... I don't know how to manage threads in UNIX
00:52:15Nico_Pand the code would probably be totally different from what would need to be written in rockbox
00:52:25gregjI can surely rnt about gcc
00:53:00 Quit obo ("bye")
00:53:03 Quit bluebrother ("sleep")
00:53:23 Quit miepchen^schlaf ("Verlassend")
00:53:35*Nico_P starts looking at mpegplayer.c
00:55:12Nico_Plinuxstb_: thanks for the idea :)
00:56:17Nico_PI'm a bit too sleepy to start this now but I'll have a good go at it tomorrow
00:56:37gregjNico_P: what are you going to be doing there, just courius ?
00:56:49 Part TrueJournals
00:57:05Nico_PI'm working on the "metadata on buffer" Summer of Code project
00:57:45Nico_Pthe aim is basically to use the memory better in order to allow clean support for album art and other kinds of metadata
00:57:54gregjI see :)
00:58:01gregjhave fun :)
00:58:05Nico_Pthanks :)
00:59:10 Nick lnakioahd is now known as Anakin (i=0@
00:59:11Mouser_XI'd like to see memory handled better for smaller files, like SPCs, and NSFs, and similar.
00:59:33Nico_PMouser_X: it's one of the goals of MoB too :)
00:59:39gregjhow does the memory allocation work on rockbox ?
00:59:52gregjdo you guys have malloc/free and friends ?
00:59:54Nico_Pgregj: you mean malloc() ? :D
01:00:00Mouser_X(If I understand correctly, Rockbox only buffers 32 files in advance. 32 files (at 65 k each) isn't a whole lot, considering the amount of memory there is to work with.)
01:00:22Nico_PMouser_X: that is precisely what MoB will be changing
01:00:36gregjscorche: hmm
01:00:48gregjscorche: anything against mallocs() in particular ?
01:01:15peturwow.. that topic came up _late_ ;)
01:01:41Mouser_XGodEater was expecting to see it a few hours ago.
01:01:50Mouser_X(I think it was him that said that...)
01:02:03Nico_PI think so too
01:02:25petur16:36:24GodEaterI'm waiting for the, "and you should include malloc()" argument to start now
01:03:25gregjtalking about negative attitude
01:03:33scorchegregj: we get it a lot
01:03:44scorcheand it has fueled some entertaining debates
01:04:09gregjsomeone acused me just of that, so...
01:04:16petur... which we will not have now
01:04:28scorche"Rockbox has no support for such a function, and not only are there no plans to make one, there are very vocal people wielding pitchforks that hunt down and burn people that suggest we ought to include one."
01:04:46***Saving seen data "./dancer.seen"
01:04:59scorchepoint is, you wont win the debate =)
01:05:37gregjsee, I guess I haven't started it - because I understand that embeded stuff is bit different from desktop/server crap
01:05:56stripwaxa bit
01:05:58*gregj feels like GodEater got him wrong again
01:06:31*scorche wonders what is next on the list for hazing
01:07:09Nico_Pgregj: I wouldn't take it personnally... the malloc() question is frequent with newcomers. that's ven why there is a wiki page about it
01:07:10Mouser_XMaybe that should be added to the Wiki?
01:07:19*amiconn wonders whether he should just brute force check all non yet known gpio ports in order to find the disk poweroff :\
01:07:29Mouser_X(The list of hazing stuff, that is.)
01:07:52gregjNico_P: I guess I expect it - because for plugins, and gui things - idealy - you would imagine malloc would be usefull
01:07:53*stripwax sighs. ssx blur is hard
01:08:14gregjNico_P: lets say - plugin starts, and declares - reserve XMB of heap for me
01:08:20Nico_Pgregj: some plugins and codecs have malloc implementations
01:08:31gregjmpegplayer has
01:08:43gregjit is the same question I brought about DCT
01:08:54scorchemalloc wont be in the core though
01:09:03gregjhow many implemementation sof DCT are there exactly in RB
01:09:05Mouser_XSo, was WPS parsing tightened up? After removing the Album Art stuff, it loads. However, I know I've used this WPS on "current" (as of some months ago) builds, and it loaded just fine...
01:09:11gregjanwser - more and more :)
01:09:25stripwaxMouser_X : - yep
01:10:02gregjNico_P: so your work is aim towards brining album art stuff back, and working
01:10:13stripwax'back' isn't really it.
01:10:21scorchewell, it was never in there before, so one cant really bring it back
01:10:46Mouser_XAll the Album Art stuff I've run into is because I was using a custom build.
01:10:55Nico_Pgregj: AA will be made possible to implement thanks to MoB, but that's not the only benefit of it
01:10:55Mouser_X(Just FYI.)
01:11:06TTSbox_Hi... If we want certain file extensions to be opened with FLite
01:11:06TTSbox_like viewer
01:11:06TTSbox_where do we specify it
01:11:06DBUGEnqueued KICK TTSbox_
01:11:06TTSbox_I don't think it is in the plugin itself
01:11:23TTSbox_is there a config file for that?
01:11:31Nico_PTTSbox_: viewers.config
01:12:39 Quit entheh ("^~")
01:12:40Nico_Pgregj: about the DCT implementations, I have no clue but I think codecs just brought their own
01:12:55Mouser_XAfter a somewhat discouraging update, I think I've been able to rebuild Rockbox back to how it was before everything went downhill... Excluding the location in the playlist that I was at (which was somewhat important, being a 4,000+ long playlist and all...)
01:13:03Nico_PTTSbox_: are you the student working on the TextToSpeech SoC project ?
01:13:49CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:13:49*linuxstb_ notes that all four students have been around today - summer has arrived...
01:14:03gregjNico_P: other thing, they often use different floating point representation (fixed point)
01:14:13Nico_Plinuxstb_: good to see :)
01:14:25peturlinuxstb_: no, midterm evaluation is coming up ;)
01:14:35Nico_PI feel like I'm the latest on my work :s
01:14:35linuxstb_midterm already?
01:14:41Nico_Plinuxstb_: soon, yes
01:15:14 Quit borisyeltsin (Remote closed the connection)
01:15:17markunMouser_X: he is
01:15:24markuneh that was for Nico_P
01:15:31 Join borisyeltsin [0] (
01:16:01Nico_Pmarkun: how's the project going ?
01:17:21 Quit HellDragon (Connection timed out)
01:17:32 Quit Ryuu (Connection timed out)
01:18:57 Join HellDragon [0] (n=JD@unaffiliated/helldragon)
01:20:03preglowmidterm, yeah
01:20:04preglowthat reminds me
01:20:30preglowi wonder if that requires me to be around, i'm still a bit sketchy on if i'll have net access at all when i'm away
01:21:44peturbetter do it before then...
01:22:08preglowthat would pretty much require it to be done tomorrow
01:23:59preglowi can't see anything in the google interface that requires my action, though
01:24:23peturmaybe ask on the mentor ML
01:25:50Nico_PI think midterm evaluating starts on the 9th
01:25:53preglowthe timeline says midterm eval starts the nnth
01:26:11preglowtough luck, i go away the 7th
01:29:23 Quit petur ("Zzzzz")
01:29:25 Join Sega [0] (
01:30:00Nico_Ppreglow: you have till the 16th
01:30:11preglowi'm not back by then
01:31:31SegaCan I ask you guy's something?
01:32:10DerPapstSega: no questions are forbidden :-/
01:32:30Sega:0 Why?
01:33:22SegaSo can I or not?
01:33:36DerPapstsega: i just wonder everytime why people ask if they are allowed to ask ^^
01:33:36scorcheSega: read the guidelines linked in the topic
01:33:45Segai did
01:33:51scorcheyou did?
01:33:57DerPapstsure you are allowed to ask.
01:34:09DerPapstif not you would already have broken a rule
01:34:16DerPapstand then being kicked :P
01:34:40SegaOK Does rockbox have tv out?
01:35:15DerPapstmaybe one should add "Don't ask to ask" to the topic ^^
01:35:17scorcheSega: if you read the guidelines before, did you purposefully choose to ignore them then?
01:35:21Segaand sorry if i broke a rule!:-(
01:35:31scorcheSega: but you just told me that you read tehm
01:35:32DerPapstyou didn't.
01:35:45scorcheDerPapst: technically, that is in the guidelines
01:35:50SegaI'm reading them now....
01:35:59Segafor the second time...
01:36:36DerPapsti have read them once.. but completely forgotten their content :P
01:38:38SegaOk i'm done!
01:39:46SegaOne more thing to ask... Is there anymore games besides the ones all ready in rockbox?
01:40:09scorchethere are a few unfinished ones in the tracker, but no
01:40:25XavierGryou might find some more on various forums and such but you will have to search around more
01:40:31DerPapstunless you get lots of different roms or wads for rockbox and rockdoom
01:40:37preglowoh well, i'll try to catch saratoga tomorrow and see
01:41:18TTSbox_Nico_P: yes I am the one
01:41:26TTSbox_sorry saw your question now
01:41:35Nico_Pno problem ;)
01:41:36SegaOk DerPapst where can i more wads for rockdoom?
01:41:42Nico_Phow's the project going ?
01:42:13DerPapstfor wads google "freewads"
01:42:31DerPapstand afaik therea re some pages that have free roms too.
01:43:00Segathanks Derpast see you guys!
01:43:27 Quit Sega ("Quitting!")
01:43:29TTSbox_Thanks... I still don't have something working
01:43:46TTSbox_BUt I guess when I have something working, I will be almost done
01:44:23TTSbox_though I don't know How much different are the requirements for porting as a plugin
01:44:28TTSbox_vs to the core
01:44:45Nico_Pwhat kind of requirement ?
01:45:48TTSbox_for example I have limitations like using the api interface for all functions
01:45:58 Join Segadude [0] (
01:46:04TTSbox_I don't know yet anything about putting things to the core
01:46:14preglowit's simpler than doing it in a plugin
01:46:27preglowwell, i think so
01:46:32Nico_PI agree
01:46:34preglowno memory requirements, all the api is there
01:46:36TTSbox_so why are we porting it as a plugin?
01:46:50preglowdon't ask me :PP
01:46:57linuxstb_I wouldn't agree - porting to the core means you have far more memory and binary size constraints.
01:47:00Nico_PTTSbox_: probably because it is something not everybody wants in the core
01:47:01TTSbox_My mentor Markun suggested that
01:47:16linuxstb_In a plugin you can relax...
01:47:22Nico_P... or what linuxstb_ said
01:47:26preglowwell, also true...
01:47:39 Quit Segadude (Client Quit)
01:48:00 Join webguest62 [0] (i=186e1ca3@gateway/web/cgi-irc/
01:48:18TTSbox_we could still put it as an option to compile...
01:48:19Nico_PI agree a plugin feels like a better testing ground for some things
01:48:26TTSbox_to the makefile
01:48:32TTSbox_make -withTTS
01:48:52Nico_PTTSbox_: true. that's managed with #defines usually
01:48:56XavierGrfor the time beeing yes but not when finished
01:48:59TTSbox_something like this
01:49:01linuxstb_I think the preference is to avoid compile-time options - it just gives more and more combination of builds to test.
01:49:09webguest62Hello people. Is it a good idea to download the lastest build or wait for more cleanup of the WMA codec?. gigabeat user here. thanks
01:49:22Nico_Pwebguest62: both ?
01:49:25linuxstb_Why wait?
01:49:35XavierGrwma works quite well on gigabeat, though no seeking
01:49:44linuxstb_If you're ever in doubt, just backup your .rockbox folder first, and you can always restore it.
01:50:08 Quit seeee (Read error: 110 (Connection timed out))
01:50:21Mouser_X(Wish I had thought of that...)
01:50:23TTSbox_hmm, so the difference with core and plugin for enduser is that plugins is not loaded to memory at startup, is that it?
01:51:08linuxstb_Yes, there's a portion of memory (512KB on most targets, 32KB on the Archos targets) reserved for plugins. One plugin at a time can be loaded and run from that buffer.
01:51:09Nico_PTTSbox_: well actually the user has to run the plugin
01:51:31webguest62What does 400% realtime mean anyhow? I saw that listed in one of the builds concerning the WMA codec
01:51:44linuxstb_If a plugin needs more than that amount of memory it needs to stop playback and claim the entire main memory.
01:51:52Mouser_Xwebguest62: It means it's running 4 times faster than necessary
01:51:53DerPapstcam 2 plugins run at the same time?
01:52:05Nico_Pwebguest62: it means the codec decodes in 1/4 of the length of the track
01:52:15linuxstb_webguest62: It means it works...
01:52:32Mouser_XSo, it doesn't mean it's decoding 4 times faster than necessary?
01:52:48Mouser_XI mean sure, when actually listening to it, it'll run full speed.
01:52:48XavierGrMouser_X: the more % the better
01:53:01Mouser_XOf course. It means you have more headroom for stuff.
01:53:03linuxstb_It means that a 4 minute track can be decoded in 1 minute - so the CPU is only working at 25% capacity.
01:53:05Nico_PMouser_X: when you say that you make it sound like it's not a good thing, or at least useless
01:53:13DerPapstlinuxstb_: then i wonder why it would be better to put TTS into a plug in. lots of plungin have a menu that wouldn't get read if the TTS plugin can't run at the same time...
01:53:27DerPapstor have i missed something improtant? :S
01:53:47Mouser_XNico_P: Ah. You're right, but I was going to clarify. I should have worded it different in the first place.
01:53:58webguest62so there is the codec is not limiting in regards to general playback? only the absence of seeking
01:54:02XavierGrDerPapst: some think that developing in a plugin is easier than on core
01:54:09Nico_PMouser_X: you have the idea right though
01:54:13linuxstb_I don't think that's been decided yet. IMO, we would probably want a dedicated "TTS plugin buffer" which could be used by TTS if enabled, or reclaimed as part of the audio buffer if not.
01:54:21XavierGrDerPapst: maybe it is personal preference too
01:54:22 Quit hcs ("Leaving.")
01:54:50Nico_Pwebguest62: if you don't use WMA, the WMA codec will change nothing to you
01:54:57Nico_Pwhatever its state
01:55:16TTSbox_I think having it a compile time option is good. People who don't want it probably wouldn't want to bloat their firmware size too
01:55:43linuxstb_You can make it optional without needing to make it a compile-time option.
01:55:47DerPapstlinuxstb_: ah ok. :) i'm not sure i want TTS in the core also.
01:56:12preglownot having it in the core would be a huge waste, if you ask me
01:56:16linuxstb_The other problem with compile-time options of course is that you can't disable them at run-time...
01:56:18preglowthink of the possibilities for blind people
01:56:54TTSbox_linuxstb_:how You can make it optional without...
01:56:55webguest62I have wma files already on my player. It just says no file. So I am gladly going to download the latest build
01:57:09linuxstb_TTSbox_: What I said 2 minutes ago
01:57:30TTSbox_can you clarify that
01:58:32DerPapstpreglow: i aware of it.. that's why i asked if 2 plugins (TTS and something else) can run at the same time. So that menus of plugins still get read.
01:58:37TTSbox_the TTS code's binaries will be in firmware,
01:59:18TTSbox_if it is not a compile time option
01:59:30TTSbox_put it is like 2,5 MBs
01:59:39DerPapstpreglow: also i hardly can imagine how TTS is supposed to work if it's not in the core.
01:59:51TTSbox_that is why I thought compile time would be better
01:59:55preglowDerPapst: well, you point it to a text file or something
02:00:32DerPapstpreglow: i thought TTS should also read menus and such
02:00:54TTSbox_yeah that is also planned...
02:01:03preglowthen i can't see how it's supposed to be a plugin
02:01:10TTSbox_Firt plugin, then to the core
02:01:16preglowthen good
02:01:18TTSbox_and reading manus filenames
02:01:20preglownothing wrong with that
02:01:30 Join stripwax_ [0] (
02:01:44DerPapstso if it is a plugin how should it read the menus then? running in the background?
02:01:49TTSbox_but I was thinking plugin programming was easier, since it was adviced to me
02:02:25TTSbox_putting it to the core has only no malloc limitation?
02:02:35TTSbox_is that the only one?
02:02:55webguest62Thanks for answering my questions people and thanks for the WMA codec. I prefer Mp3 VBR but some of my library is WMA and I don't want to go through the conversion again so this is Major for me.
02:03:08XavierGrDerPapst: as soon as he manages to make it working as a plugin then I don't think it would be too difficult to convert it for the core
02:03:49linuxstb_I'm not talking of a plugin in the current sense, but a new type of plugin, with its own buffer area. If the user has enabled TTS, the TTS plugin is loaded to that buffer, otherwise that space is taken by the core for the audio buffer.
02:03:51scorcheTTSbox_: core has no malloc at all
02:03:55XavierGrwell webguest62 wma was one of the final unexplored strongholds of rockbox so we are all happy with it :P
02:04:30preglowwebguest62: playback isn't exactly smooth yet
02:04:49linuxstb_It's pretty good on Gigabeat...
02:04:56preglowlinuxstb_: oh, right...
02:05:19*linuxstb_ is happy preglow hasn't got one of those yet...
02:06:17 Quit Nico_P (Remote closed the connection)
02:06:23linuxstb_55 GBP was a cheap price to pay to not have to worry about optimising...
02:06:25preglowi'm just happy i never ended up with anything archos :>
02:06:27*Mouser_X is looking forward to a S port. 533 mhz CPU? Where's my Snes9x port?
02:06:28 Quit bdgraue_ (Remote closed the connection)
02:06:37stripwax_55 for a gigabeat? not bad
02:07:06linuxstb_An F20 from ebay. It cost me more than double that for an 80GB disk for it...
02:07:16TTSbox_I just bought a F20 for $76 from Ebay
02:07:25TTSbox_plus 12 shipping
02:07:27preglowwhat the hell
02:07:36preglowyou don't want to mention these prices to me
02:07:48scorchecheap cheap cheap
02:08:18Mouser_XI paid $175 USD for my F40, but apparently that was overpriced.
02:08:34Mouser_X(It did include shipping though.)
02:08:35webguest62wonder why they made the gigabeat with such a powerful CPU 300mhz and the newest ipods dont even have that power?
02:08:39TTSbox_still going for that range
02:08:53preglowwell, the nano2g has a pretty kickass cpu too
02:09:00scorchewebguest62: the new nano is quite...bah...ybeat me to it
02:09:31linuxstb_And the ipod video has two 80MHz ARM cores, plus a 150MHz (I think) [insert speculation here] core with DSP
02:09:37preglowseveral hundred mhz arm, its own dsp core, tons of sram
02:09:54scorchethey went a bit overboard =P{
02:09:54DerPapstXavierGr:the problem i have is that once TTS is fully running that it has to go into the core atm since there is no way to have a plugin running in the backgrund yet. so the binarysize of the core would be increase a lot. and i also dont think everyone whats to have TTS as a feature. So a plugin would be the best solution. and according to linuxstb it is possible to get TTS running in the backgrund as a plugin.
02:09:57preglowa 150mhz sparc, you say??
02:10:24toffe82webguest62: they did it just for rockbox people ;)
02:10:31preglowDerPapst: there is a way to have a plugin running in the background, there is
02:10:36preglowtsr plugs
02:10:43preglowlike the battery bench thing
02:10:51DerPapstah.. didn't know that.
02:10:56preglowbut now you do!
02:10:58webguest62I paid 99 for my F40
02:11:09TTSbox_but when you run another plugin it is terminated
02:11:19 Quit jgarvey ("Leaving")
02:11:22DerPapstso it is possible to run 2 plugins at a time.. one in the backgrund and the other not :P
02:11:28preglownope, just the one
02:11:38TTSbox_but probably we can change that for TTS
02:11:56DerPapst^ that is what i think would be the best.
02:11:57TTSbox_as I think linuxstb suggested
02:12:03linuxstb_How much of that 2.5MB is code, and how much is data?
02:12:18TTSbox_all incorporated into code
02:12:26 Join Ryuu [0] (i=JD@unaffiliated/helldragon)
02:12:28TTSbox_data is also on the code
02:12:42*linuxstb_ doesn't understand
02:12:48webguest62Where does the F40 with the newest of Ipods in regards to processing power?
02:12:58TTSbox_but it is like 2 MB
02:13:19TTSbox_less than .5 MB is code
02:13:46linuxstb_webguest62: I would say the current ipods are more powerful, but Rockbox can't take advantage of that power. All the gigabeat's power is in a single ARM CPU, which makes things very easy.
02:14:47TTSbox_for example on my linux setup flite executable is 3.22 MBs
02:15:17TTSbox_I have downloaded a port for Windows mobile which is 2.5 MBs
02:15:29 Quit stripwax_ ("Miranda IM! Smaller, Faster, Easier.")
02:15:33TTSbox_The linux one has two voices I guess
02:15:41linuxstb_I'm wondering how big just the code would be if you loaded the data from an external file at runtime.
02:15:46TTSbox_but all is in the executable
02:16:02TTSbox_it would be smaller
02:16:07TTSbox_very much smaller
02:16:24TTSbox_you are thinking it so that the firmware will be smaller
02:16:37linuxstb_Yes, we wouldn't want to put the data in the core.
02:16:48preglowoh no siree, we wouldn't
02:16:50TTSbox_most of it's size is voices and lexicon
02:17:19DerPapsthow many languages does "it" speak?
02:17:35TTSbox_Than I will try to learn how we can achieve that after plugin succeeds
02:17:41TTSbox_only English yet
02:17:49DerPapstok :)
02:17:49 Quit webguest62 ("CGI:IRC (Ping timeout)")
02:17:58TTSbox_but any of FESTVOX voices can be converted
02:18:34TTSbox_there we have spanish
02:18:50TTSbox_I remember seeing German too
02:19:23 Quit stripwax (Read error: 110 (Connection timed out))
02:19:59*preglow bed
02:21:03DerPapstwow... they sound scary :D
02:21:27TTSbox_I have just seen that there are works for Indian languages, titles "Building Hindi and Telugu Voices using Festvox"
02:21:47TTSbox_they are intelligeble at least
02:22:03TTSbox_what do you think?
02:22:45TTSbox_for me that comes first
02:22:59DerPapsthaving problems with that page...
02:23:02TTSbox_naturalness is also important
02:23:16DerPapstmy sytesized wave files are under one 1 second...
02:23:22 Quit Thundercloud (Remote closed the connection)
02:23:46DerPapstthe only one that worked was half of the indian one :D
02:23:54DerPapstand that sounded scary ^^
02:24:03linuxstb_That page crashed firefox for me...
02:25:04DerPapstor they don't work at all -.-
02:25:08DerPapstempty files
02:25:31TTSbox_same happened to me now
02:25:44TTSbox_I guess, it can not handle 4 people :)
02:25:55 Join Daishi [0] (
02:25:56TTSbox_it should
02:26:55*Mouser_X prepares to crash Firefox.
02:27:01Mouser_XI'm going to check out the page...
02:29:29Mouser_XWorked for me. I tried "Hello World." I understood the Hello part, but World sounded like whirl.
02:29:32DerPapsthave used the premade ones
02:30:16DerPapstsome i understood good.
02:30:37DerPapstbut for some i had to read the text to undersant what it said
02:31:22*amiconn starts to understand PP somewhat more
02:31:45DerPapstand the german examples are most worst ever heared.
02:31:54DerPapstespechially the male one :D
02:32:06DerPapstbut they are still in development.
02:32:18DerPapstso i'm not complaining
02:33:28DerPapstamiconn: still reading lots of arm asm?
02:33:53amiconnlinuxstb_: Could you perform a test on your photo?
02:34:18 Join scorche` [0] (n=scorche@rockbox/administrator/scorche)
02:34:33amiconnI would like to know what happens if you set bits 30 and 31 of register 0x600060a0 to zero
02:34:38DerPapstthe only asm i ever read and undersand was some pretty trivial asm for an am<sup>0</sup> machine.
02:35:38linuxstb_amiconn: Sure.
02:36:00DerPapstin fact i have "written" a C0 compiler in haskell that made am0 code ^^
02:36:02 Quit scorche (Nick collision from services.)
02:36:07 Nick scorche` is now known as scorche (n=scorche@rockbox/administrator/scorche)
02:37:12DerPapsthaskell is a pretty strange programming language/concept.
02:37:21amiconnAccording to the color/photo is the only target where these 2 bits get set - and hence it's the only target that needs them reset temporarily during clock change in order to fix the freezing
02:37:23DerPapsti've forgotten most of it already.
02:39:01linuxstb_amiconn: Do you have a patch you want me to test? I'm busy with some real-work at the moment... (and I'll be away for four days starting tomorrow).
02:39:34amiconnNo patch... just a quick test. But it's not especially important
02:44:45 Join Thundercloud [0] (
02:52:58 Part pixelma
02:53:02 Quit DerPapst (Nick collision from services.)
02:53:05 Join DerPapst [0] (
03:04:48***Saving seen data "./dancer.seen"
03:11:20 Quit krazykit_ (Read error: 104 (Connection reset by peer))
03:12:26FebsHmmm. Rockbox on my iPod Video just crashed for about the 5th time in the past day. Undefined instruction at EDBAE044.
03:12:57DerPapstmine did that too yesterday 0037e5 though
03:13:32DerPapstadn i have another funny issue...
03:15:51DerPapstif browse to e.g. /iPod_Contor/Music/F01/BLAH.mp3 i get a splash saysing that my installation is uncomplete and that .rockbox is missing. tehn it jumps to the wps and plays the song as there has never been an error. also i can still use all codecs and plugins afterwards.
03:16:35DerPapstalso backdrop and wps are still the ones i've selected.
03:16:47 Join krazykit [0] (
03:17:03DerPapstFebs: do you have a similar issue as well?
03:17:24FebsI have not noticed anything like that. Does it matter where you browse to?
03:18:06DerPapsti only hve music in iPod_Control... but i can test it though
03:20:28DerPapstbah.. rockbox froze...
03:20:33 Quit Shaid (Read error: 110 (Connection timed out))
03:21:58DerPapstheh odd...
03:22:06DerPapstdoesn't happen anymore
03:22:37DerPapstmight have to do something with the filename checking code and Lear's commit
03:22:52DerPapstfor fixing a bug.
03:23:37DerPapstmeh rockbox crashed again -.-
03:24:23DerPapstalways uf i enter the filebroser after i quit it because getting confused with the contols.
03:25:07 Quit borisyeltsin (Remote closed the connection)
03:25:26 Join borisyeltsin [0] (
03:28:12XavierGrDerPapst: well I wouldn't wan't to see TTS as a tsr plugin, if voice on rockbox is on core, I think that TTS should be on core too
03:28:31FebsNow this is really strange. I had Rockbox running but wasn't looking at it. Next time I looked, it was frozen on the boot splash screen.
03:31:03 Quit Thundercloud (Remote closed the connection)
03:32:14DerPapstXavierGr: i was concerned about adding 2.5MB to the core... but linuxstb suggested to spit the code from the data and add only the code to the core.. which isn't as worse as 2.5MB ;)
03:32:43DerPapsthow the hell do i exit this text viewer...
03:33:12*DerPapst looks that the source now.
03:35:12FebsAnother crash. Undefined instruction at 081D2DA4.
03:35:18 Quit spiorf (Remote closed the connection)
03:35:41linuxstb_XavierGr: The core is currently around 300KB-500KB on swcodec (depending on LCD size and CPU). It sounds like TTS will be at least 500KB of code...
03:36:44linuxstb_But I think it's too early to decide how to implement it - we need to see how it performs, and what the resource requirements are first.
03:36:50*DerPapst fails at quitting the text viewer
03:38:43 Quit Xerion (" ")
03:39:10XavierGrDerPapst: the plugin text_viewer?
03:40:35FebsAnother crash.
03:40:51DerPapstdoh.. found it now
03:41:12*DerPapst starts hating the text_viewer plugin.
03:41:42XavierGrfor me a simple stop press quit the plugin
03:42:34DerPapstnot on the iPod :P
03:43:34FebsPressing Menu should bring you to the menu, from which you can select "quit."
03:44:22DerPapstmenu brings me to some sort of editing menu
03:44:39DerPapst<< brings me to a meu where i can quit the prugin.
03:45:17*DerPapst thinks about changing that though
03:45:24DerPapstbut not now.
03:45:29DerPapst3:50am ;)
03:45:41DerPapstgood night everyone :)
03:45:48 Quit DerPapst ("So Long And Thanks For All The Fish!")
04:18:28 Join jhulst [0] (
04:27:15 Join webguest44 [0] (i=186e1ca3@gateway/web/cgi-irc/
04:28:35webguest44july 4th build for gigabeat gives undefined instruction error and freezes the unit for some of my WMA coded files. They are all at least 64k/bit
04:29:47 Quit TTSbox_ (Read error: 110 (Connection timed out))
04:30:26krazykitwebguest44, wma support is very very young.
04:31:37krazykitdo you know what version wma these are?
04:33:17 Join Lynx [0] (
04:33:44 Quit webguest44 ("CGI:IRC (Ping timeout)")
04:35:02 Join webguest44 [0] (i=186e1ca3@gateway/web/cgi-irc/
04:35:39webguest44I don't know the version of the WMA codec it is.
04:36:34krazykitwell, wma playback is really new, so don't be surprised that it isn't working well. it should be much better by the end of summer
04:39:26 Quit webguest44 (Client Quit)
04:40:07 Join hcs [0] (n=agashlin@
04:46:54 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
04:48:36 Quit Lynx_ (Read error: 110 (Connection timed out))
04:48:36 Nick Lynx is now known as Lynx_ (
04:49:58 Quit Soap ()
04:50:48toffe82I create a playlist of all my songs (2800) but the playlist only contains 1920 songs, is there a limit ?
05:00:13 Quit jhulst (Remote closed the connection)
05:04:50***Saving seen data "./dancer.seen"
05:11:36Mouser_Xtoffe82: My limit is set at 10,000.
05:11:59Mouser_XIf I had all the SPCs on my Gigabeat, 20,000 would still be about 100 or 200 short...
05:15:12 Quit tedrock (Client Quit)
05:18:46toffe82Mouser_X: the limit is at 10000 and I have only 2800 songs
05:19:11Mouser_XI figured. I have no idea.
05:19:18toffe82the browse limit is at 1000
05:19:37 Join tedrock [0] (
05:19:41Mouser_XI have mine at 3,000 (my OCR directory is big).
05:20:11 Join iPodPhoto [0] (n=46311c65@
05:20:14toffe82if you creat the playlist, you have all your songs ?
05:20:59Mouser_XIn my OCR directory? As far as I've noticed, yes.
05:21:11iPodPhotohello. is there anyone here who has twiki access
05:22:47toffe82Mouser_X: I put 3500 in browse directory , and it give me a lot of error and the playlist is now 10000 songs ???
05:23:39iPodPhotoI will take that as a no. Thank you kindly. I sha'll check another time
05:23:45 Part iPodPhoto
05:25:02toffe82Mouser_X: I change it to 2000 and it saya me no rockbox directory, installtion incomplete and start the wps screen with 10000 songs but nothing to play ?
05:27:30 Join piroko [0] (
05:32:52toffe82I think I do with this now, but I have to foind why he doesn't create a playlist with all the songs
05:36:46Mouser_XOn the Gigabeat?
05:36:52*Mouser_X is worried now...
05:36:56Mouser_XThat's the build I'm using.
05:37:02Mouser_XWell, it's from today anyway.
05:39:07 Quit piroko ("leaving")
05:39:16 Join aliask [0] (
05:40:56 Join jhulst [0] (
05:44:31 Join piroko [0] (
05:55:49 Nick Soap_ is now known as Soap (n=Soap@rockbox/staff/soap)
05:57:02 Quit piroko ("leaving")
06:04:45 Quit lazka (Remote closed the connection)
06:09:22 Quit atsea- (Remote closed the connection)
06:11:39toffe82Mouser_X: on the gigabeat X, it is with the buid from today but I have it like this for 2 weeks more or less
06:17:40 Quit Nibbier (Read error: 110 (Connection timed out))
06:18:17 Join Nibbier [0] (
06:20:23 Join atsea- [0] (i=ariel@gateway/tor/x-911b0fcdc7971417)
06:26:49 Join linuxstb__ [0] (
06:35:36 Join piroko [0] (
06:37:53 Quit Febs (Read error: 110 (Connection timed out))
06:40:03 Join BK-- [0] (
06:41:01 Quit linuxstb_ (Read error: 110 (Connection timed out))
06:49:48 Join MrKeuner [0] (n=kudo@unaffiliated/mrkeuner)
06:50:45MrKeunerhi, is there a whatsnew file that I can find all of the changes to releases?
06:50:59scorcheMrKeuner: the commit log on the front page
06:51:11scorcheand we dont put out releases, so...
06:52:00MrKeunerthat's good enough thanks
06:53:08 Quit MrKeuner (Client Quit)
06:53:54 Quit amigan (
06:53:54 Quit XavierGr (
06:53:54 Quit bospaadje (
06:53:54 Quit maffe (
06:53:54 Quit crashd_ (
06:53:54 Quit Weiss (
06:53:54 Quit gtkspert (
06:53:54 Quit Seed (
06:53:54 Quit FOAD (
06:56:25toffe82scorche: do you have an idea on my problem of playlist ?
06:56:42scorchei havent been here
06:56:47*scorche scrolls up
06:57:08 Quit piroko ("brb")
06:57:39toffe82scorche: to resume, I have 2800 songs and creating the playlist give me only 1920
06:57:47 Join piroko [0] (
06:58:02scorcheare all 2800 valid?
06:58:23toffe82yes I can listen to them if I browse
06:58:44NJoinSeed [0] (
06:58:50scorchecan you pick out ones that arent in the database, but are in the browser?
06:58:57NJoincrashd_ [0] (
06:59:18NJoinWeiss [0] (
06:59:21toffe82hold on I have to check wich one is not in the file
07:03:48 Quit hcs (Read error: 110 (Connection timed out))
07:04:38toffe82I can play what I want if I browse
07:04:54***Saving seen data "./dancer.seen"
07:05:12BHSPitLappythe DAP on today has a lot of smart features
07:05:45scorchetoffe82: were there some that you could see that werent in the database, but are in the browser though?
07:06:03toffe82I don't use the database
07:06:20toffe82I just creat the playlist of all the songs
07:06:40toffe82yes I play some that are not in the play list
07:07:13 Join atsea-49 [0] (i=ariel@gateway/tor/x-d2673f1c0994371c)
07:08:12NJoinamigan [0] (i=dcp1990@unaffiliated/amigan)
07:08:12NJoinXavierGr [0] (
07:08:12NJoinbospaadje [0] (
07:08:12NJoinmaffe [0] (
07:08:12NJoingtkspert [0] (
07:08:12NJoinFOAD [0] (
07:08:23scorchetoffe82: have recursive dir turned on as well?...i forgot what the limit was for number of folders deep..
07:11:28toffe82I just have a structure like this : music\artist\album\song
07:12:12toffe82I change the browse limit and the max playlist and I have some strange behavior
07:13:25 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
07:14:28 Part atsea-49
07:14:53 Quit piroko ("Time to dream of electric sheep")
07:15:41 Quit goffa (Remote closed the connection)
07:16:53 Quit atsea- (Read error: 104 (Connection reset by peer))
07:21:02 Part toffe82
07:39:57 Join printfXh4 [0] (
07:45:49 Join erchino [0] (n=erchino@
07:46:16erchinoim trying to install rock box on my 30gb video
07:46:20erchinoand it doesnt work
07:46:39erchinowhen I unzipped... in the ipod root doesnt appear .rockbox
07:46:43erchinoas it says in the manual
07:46:47erchinowhat can I do?
07:47:15 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
07:48:18erchinoim using mac osx 10.4
07:49:01erchinoanyone? can help...
07:49:22 Quit XavierGr (Nick collision from services.)
07:49:25 Join XavierGr [0] (
07:50:49erchinohelp... :(
07:52:37scorcheholy spam...
07:53:11scorcheplease read the guidelines linked in the topic
07:53:54erchinoI have readed it
07:54:01erchinoand it doesnt work ... really
07:54:10scorchewhat doesnt?
07:54:16erchinothe installing...
07:54:40erchinoinstalling rock box on my 30 gb video
07:54:42scorcheif you are on a mac, you wont see the .rockbox folder because it is hidden
07:55:02erchinothe root folder doesnt unzipped anything named .rockbox
07:55:22erchinoI did exactly everything it says
07:55:36erchinoafter the unzipped I download the patcher
07:55:43erchinoand it patched
07:55:52erchinoand even than it doesnt work
07:56:13erchinothan I downloaded mbr-video30gb.bin
07:56:14scorchewhat do you mean by "doesnt work"
07:56:29erchinowhen I reset the ipod it looks just the same
07:56:52erchinoEnter i to install the Rockbox bootloader, u to uninstall
07:56:52erchino or c to cancel and do nothing (i/u/c) :i
07:56:52 Join entheh [0] (
07:56:53DBUGEnqueued KICK erchino
07:56:53erchino[Process completed]
07:57:20erchinodamn... i know what is going on
07:57:27scorchewhat does the screen look like?
07:57:28erchinoit says I have to convert the ipod to 32fat
07:57:45erchinobut if I convert to it does it going to work on my mac?
07:58:04scorcheyes it will
07:59:03 Quit Daishi (Remote closed the connection)
07:59:21erchinodo i have to erase everything???
07:59:26erchinonow that hurts...
08:03:22erchinohow do i formaat my ipod on mac?
08:04:49 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
08:05:10 Quit joshin (Read error: 110 (Connection timed out))
08:05:14erchinoi saw that to
08:05:31linuxstbThen what step are you stuck on?
08:05:38erchinobut how do I know if mine is 512 bytes or 2048 bytes?
08:05:57linuxstbipodpatcher displays that information when you run it.
08:06:10erchinolet me see
08:06:52[Ray]"NOTE: If you have the 30GB video ipod, you need to determine your sector size. This can be done with the ipodpatcher command - after identifying your ipod, type "ipodpatcher /dev/diskN" (replacing diskN with the device name assigned to your ipod). This should then display the sector size.
08:07:07linuxstbBut if you're not comfortable in the terminal, I would recommend trying the "easy way" - i.e. finding a Windows PC with itunes installed and using that.
08:07:53 Join pondlife [0] (
08:09:02erchinocan I do this on disk utility?
08:09:28linuxstbSome steps yes, but not all.
08:10:00erchinogod damn it...
08:10:04 Join huzz [0] (
08:10:15 Quit hcs ("Leaving.")
08:10:39erchinodd if=mbr-video30gb.bin of=/dev/diskN
08:10:39erchinodd: mbr-video30gb.bin: No such file or directory
08:10:47erchinoim new on terminal
08:11:30linuxstbWhere did you save the mbr-video30gb.bin file to?
08:11:48erchinoon the desktop
08:12:05erchinoand is dev/disk2
08:12:13linuxstbThen type "cd Desktop" in the terminal
08:13:19erchino cd desktop
08:13:19erchino-bash: cd: desktop: No such file or directory
08:13:55erchinowhat do i have to type on the terminal?
08:14:00erchinocd desktop
08:14:04LinusNerchino: type "ls"
08:14:15LinusNwithout the quotes
08:14:20erchinoaprobacion.rtf mbr-video30gb.bin telefono Tonala.rtf
08:14:20erchinoconsul de noemi.rtf rockbox-ipodvideo.pdf
08:14:37 Join Rob222241 [0] (
08:15:00erchinoit says dd unknow operation...
08:15:52linuxstbDid you type the command correctly?
08:16:16linuxstbCan you copy and paste what you typed into the terminal to here?
08:16:33erchinodd if=mbr-video30gb.bin of=/dev/disk2
08:16:37erchinoI typed this...
08:17:06LinusNand the error message was?
08:17:07linuxstbAnd what is the exact error message?
08:17:24erchinodd: unknown operand cd
08:17:49LinusNthen you can
08:18:05LinusN't have typed exactly what you said you did
08:18:12 Join bdgraue [0] (
08:18:45LinusNjust try it again
08:18:56linuxstbOn an empty line.
08:20:02 Quit linuxstb ("CGI:IRC")
08:20:09 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
08:20:56erchinoit says resource busy now
08:21:12linuxstbThen you need to unmount your ipod - you can do that in disk utility.
08:22:25 Quit borisyeltsin (Remote closed the connection)
08:22:33linuxstbBe sure to click "unmount", not "eject".
08:22:46 Join borisyeltsin [0] (
08:23:35huzzrockbox rox :/
08:24:42markunhuzz: indeed :) which player do you have?
08:24:54erchino1+0 records in
08:24:54erchino1+0 records out
08:24:54erchino512 bytes transferred in 0.000108 secs (4740582 bytes/sec)
08:25:10erchinothis is what it said...
08:25:18erchinois it alright?
08:25:28erchinoor did i did something wrong??
08:25:43markunerchino: I think it's fine
08:26:10erchinoit says creating images
08:26:20erchinohey is it worth it all this
08:26:22 Join wmaaddingmusic [0] (i=474872e4@gateway/web/cgi-irc/
08:26:58LinusNerchino: no, it's not worth it - rockbox sucks :-)
08:27:07BK--markun, i have it on a sansa e270, and it really does rock
08:27:10huzzmarkun, iPod
08:27:31wmaaddingmusiccan someone help me figure out why adding WPS makes sound files top playing?
08:27:35huzzi've been an iPod user for less then 48 hours
08:27:37wmaaddingmusic*stop playing
08:27:40huzzi thought i almost bricked mine :/
08:27:53huzzjust had to press the back cover and wala *magic* :p
08:28:08LinusNwmaaddingmusic: they stop alltogether?
08:28:22erchinoit wont boot...
08:28:29wmaaddingmusicyes they will not play sound at all and no time movement or anything
08:28:30erchinoI think I broke my pod...
08:28:48LinusNerchino: now i believe you have to perform a restore operation with itunes
08:28:58huzzerchino, just press the front panel from all sides, give it a little thumb, then press the reset keys will work like magic :)
08:29:10huzzif it won't boot you can't restore via itunes
08:29:12wmaaddingmusici am using a 30gb video ipod with senabs build of wps and yesterdays version of rockbox that added wma....
08:29:24linuxstberchino: Did you format the new FAT32 partition?
08:29:57wmaaddingmusici really dont want to try a fresh install becasue it takes a long time for the music to transfer over again....
08:29:58linuxstberchino: You need to force your ipod into disk mode - hold MENU+SELECT for about 5 seconds to reboot it, then immediately press and hold SELECT+PLAY
08:30:11CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:30:11*linuxstb runs - back on Sunday evening.
08:30:14huzzthis just happened to me like 5 minutes ago :/
08:30:17 Quit linuxstb ("CGI:IRC")
08:30:30LinusNwmaaddingmusic: i don't see why you would have to transfer the music again just because you reinstall rockbox
08:30:59 Quit Rob2222 (Read error: 110 (Connection timed out))
08:31:10pondlifeRegarding TTS - I wonder if the TTS plugin could use the voice buffer? I don't suppose you'd want to use both TTS and lang.voice..? Might be a squeeze though.
08:31:32wmaaddingmusicoh well you are right lol i wasnt thinking right.. so use ipodpatcher to uninstall it then delete .rockbox?
08:31:36LinusNpondlife: i think it should use the voice buffer
08:31:51pondlifeIt makes a lot of sense.
08:31:55LinusNwmaaddingmusic: ipodpatcher is only for the bootloader - no need to use that
08:32:27LinusNwmaaddingmusic: just unpack the entire file to the player
08:32:33pondlifeHopefully TTSBox will get to know this at some point.
08:32:47linuxstb__How big is the voice buffer?
08:32:53pondlifeAs big as you want.
08:32:58LinusNpondlife: i think it's a waste of time to work with it as a plugin
08:33:00pondlifeAs big as the voice file (plus a bit)
08:33:22wmaaddingmusic.... well what should i do then? just re-extract everything? I have tried when i do it to the rockbox it will play musci but then the wps doesnt work, if i do it to teh wps the sound doesnt work
08:33:25 Join erchino_ [0] (n=erchino@
08:33:31pondlifeLinusN: True, I meant more that it could be a run-time option (needing a reboot), not a compile-time option.
08:33:41pondlifeJust like loading a voice file.
08:33:43linuxstb__But is the voice buffer at a fixed location?
08:33:54LinusNwmaaddingmusic: doesn't senab's wps require a special senab build as well?
08:34:08linuxstb__So how can it be used to load code?,,,,
08:34:22wmaaddingmusicyes i downloaded it and each time i unpack it the sound quits working
08:34:49LinusNwmaaddingmusic: but does senab
08:34:53erchino_it says file not found...
08:34:58LinusN's build contain the wma codec???
08:35:36LinusNwmaaddingmusic: senab has to update his build with the new wma codec, else it won't work
08:35:37linuxstb__erchino_: That's good - you now need to extract the file.
08:35:42*linuxstb__ really runs
08:35:45 Quit linuxstb__ ("Leaving")
08:35:50erchino_i did...
08:36:07erchino_on disk mode? right???
08:36:07wmaaddingmusici downloaded his senab build called, 5g_32_070430 and that is what i unpack, the WPS i use is called phktape i dont know if it is a senab WPS
08:36:10LinusNerchino: how did you go about when unzipping the file?
08:36:29wmaaddingmusicextract all to root of player
08:36:38wmaaddingmusicwhoops nevermind
08:36:49erchino_I dont have that option on mac...
08:36:51LinusNwmaaddingmusic: as you can see, that build was made on april 30, long before the wma codec
08:36:54erchino_extract all
08:37:08LinusNerchino: that's the mac curse for you
08:37:23wmaaddingmusicyes but i was just using it yesterday, and then i added more themes and it stopped
08:37:25erchino_but Now im stuck
08:37:27LinusNyou can do it in the terminal
08:37:35erchino_cna I do that?
08:37:41LinusNtype "ls"
08:37:52LinusNis there?
08:38:20LinusNtype "unzip"
08:38:29LinusNwhat did it say?
08:38:42erchino_it says whole bunch of options
08:38:54LinusNis your ipod attached and unmounted?
08:39:04LinusNthen do it
08:39:24LinusNeh, not unmounted, sorry
08:39:42erchino_I attached again
08:39:56LinusNnow to find out the path to the ipod
08:41:04 Join Soap_ [0] (n=Soap@rockbox/staff/soap)
08:41:56LinusNdo you know what the ipod is called on the desktop?
08:42:02erchino_Sung Bae
08:42:27LinusNit's a long time since i used a mac...
08:43:26LinusNhang on
08:43:47GodEaterearning your stripes today LinusN ? :)
08:43:52 Quit Rob222241 (Read error: 104 (Connection reset by peer))
08:43:56LinusNGodEater: :-)
08:44:01 Join Rob2222 [0] (
08:44:32 Quit [Sur`DataGhost] (Read error: 104 (Connection reset by peer))
08:44:34GodEaterLinusN: ipodpatcher −−scan ?
08:45:04LinusNGodEater: does it tell the path to the ipod filesystem?
08:45:14GodEaterno - but it would give you the /dev/disk
08:45:16 Join DataGhost [0] (n=dataghos@
08:45:18GodEaterand then you could run "mount"
08:45:24GodEaterto find out where in the filesystem it's gone
08:45:26GodEaterI suppose
08:45:27LinusNit's already mounted
08:45:38 Quit erchino (Read error: 110 (Connection timed out))
08:45:40LinusNerchino_: does the disktool tell you the path to the ipod?
08:45:48GodEateryeah, running mount with no arguments should tell you where it's mounted
08:45:50erchino_ not really
08:45:58LinusNGodEater: good idea
08:46:04LinusNerchino_: type "mount"
08:46:15*GodEater is also macless though is and is just guessing
08:46:46*GodEater hopes it works the same as other POSIX systems
08:47:04erchino_it says
08:47:43*GodEater likes the dramatic pause
08:47:58LinusNoh, the tension....
08:48:29erchino_it says dev/disk2s2
08:48:39erchino_ on /volume/sung bare
08:48:50LinusNhere's what you do:
08:49:48GodEatermore dramatic pauses...
08:49:49LinusNtype "unzip -d "/volume/sung bae""
08:50:10scorcheyou guys haing fun? =P
08:50:18LinusNscorche: :-)
08:50:50erchino_it says
08:50:56markunerchino_: you're almost there
08:51:04erchino_cannot find or open bae,
08:51:16GodEatererchino_: did you put the "" in ?
08:51:17markunerchino_: try "unzip -d "/volume/sung bare""
08:51:19markunwith the R
08:51:31LinusNyou were only supposed to remove the outer quotes
08:51:45markunLinusN: and you made a typo
08:51:59LinusNthe quotes around "/volume/sung bae" were supposed to be there
08:52:06erchino_cannot create extraction directory: /volume/sung bae
08:52:08markunLinusN: sung baRe
08:52:15LinusNmarkun: aaaaah :-)
08:52:18erchino_is SUNG BAE
08:52:30markun08:48 < erchino_> on /volume/sung bare
08:52:40erchino_I did type that
08:52:50erchino_is sung bae
08:52:50LinusNmarkun: see the line after that
08:52:54erchino_SUNG BAE
08:52:55 Join atsea- [0] (i=ariel@gateway/tor/x-07a9c2f8e51aa18a)
08:53:01markunoops ;)
08:53:14erchino_and it says cannot create extracion directory
08:53:31markunmaybe you don't have write permission then
08:53:49LinusNerchino_: ls -l "/volume/sung bae"
08:53:54erchino_I can read and write
08:54:42 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
08:54:42erchino_is says ls [-abcf...] file
08:54:52linuxstbIt's /Volumes/...
08:54:58 Quit linuxstb (Client Quit)
08:55:22erchino_its volumes
08:55:31erchino_let see if it works
08:55:39erchino_cause its unzipping
08:55:54erchino_what do I do now?
08:56:01erchino_do I eject and try it out?
08:56:02GodEaterhahahaha - love linuxstb's "Hit and Run" support
08:56:08LinusNerchino_: yes
08:57:17LinusNerchino_: a word of advice for the future: when we ask what a command says, you must COPY and PASTE, not retype it
08:58:28LinusNlet's hope it works now
08:58:37 Join pseudo_ [0] (
08:58:44erchino_its on
08:58:47erchino_now what??
08:58:52erchino_put music???
08:58:55erchino_and thats it?
08:58:56LinusNnow enjoy rockbox :-)
08:59:28wmaaddingmusicno dont enjoy rockbox... love rockbox
08:59:29 Quit Mouser_X (Read error: 104 (Connection reset by peer))
08:59:55erchino_the menu is that small???
09:00:10scorchechange the font
09:00:14LinusNerchino_: maybe you could have used the Rockbox Utility:
09:00:15GodEatererchino_: it's now time to read the manual....
09:00:17erchino_i attached the ipod and doesnt appear... on the desktop as usual
09:00:42erchino_like what utility?
09:01:05 Join ender` [0] (
09:01:13LinusNit's a utility that helps you install rockbox, and themes
09:01:44erchino_so I dont use itune anymore to put music on?
09:02:17LinusNyou can, if you want though
09:02:34 Quit entheh ("^~")
09:03:17 Quit Soap (Read error: 110 (Connection timed out))
09:03:38wmaaddingmusiclike i said love rockbox
09:03:52LinusNwmaaddingmusic: good to hear :-)
09:04:08erchino_so when I opened the rbutil and click on themes
09:04:17erchino_it downloads automatically?
09:04:35erchino_when it says fetching?
09:04:41LinusNi believe so
09:04:55***Saving seen data "./dancer.seen"
09:05:02erchino_let me see...
09:06:08 Quit printfXh4 (Connection timed out)
09:07:39erchino_ok let me copy some music to see how it works...
09:07:43 Join Mouser_X [0] (
09:07:52erchino_thanks a lot linusn
09:07:56erchino_god bless
09:07:57LinusNyou're welcome
09:08:27LinusNsometimes i regret selling my mac
09:08:36wmaaddingmusicLinusN: I know i just really love rockbox, i know the wma and WPS thing is wonky cause its new but i love drag and drop and i love the functionality, these people are amazing for writing such a great piece of software
09:08:42erchino_why is that?
09:08:43 Quit wmaaddingmusic ("CGI:IRC")
09:08:46GodEaterLinusN: and then you wake up :)
09:09:22LinusNerchino_: because i lack the knowledge to help guys like you, for instance
09:09:34LinusNand i kind of like os x
09:10:05LinusNwhat i *don't* miss is os 9
09:10:22erchino_this is my first mac
09:10:23 Join webguest92 [0] (i=c023110f@gateway/web/cgi-irc/
09:10:30erchino_so I dont really know about oc9
09:10:31 Quit blithe (Remote closed the connection)
09:10:48LinusNi once had to write an Illustrator plugin that had to work on OS9, OSX and Win32
09:11:02LinusNwith networking support
09:11:09LinusNthat sucked
09:11:42LinusNit involved porting curl to os9
09:11:58LinusNusing a posix portability layer
09:12:19erchino_is it stable the rockbox?
09:12:24erchino_does crash?
09:12:25LinusNi bought a mac for that sole purpose
09:13:08LinusNerchino_: it's fairly stable on most targets, but unfortunately ipod is one of the most unstable targets...
09:13:22LinusNit should work fine though
09:13:54LinusNipod is a hard target to support for several reasons
09:14:15erchino_too much apple??? i guess??
09:14:18GodEatererchino_: I had a crash on my iPod yesterday - but that's the first one since I first installed at the beginning of June
09:14:20LinusNthe lack of hardware documentation
09:14:37GodEatererchino_: and they're not hard to recover from - simply reset and continue
09:14:42erchino_when it crash you do the same as original?
09:14:45erchino_menu and select?
09:14:58erchino_ok thanks
09:15:26erchino_about the sound setting?
09:15:32erchino_and all thos things
09:15:38erchino_any good advices??
09:15:44GodEateryes: read the manual :)
09:15:45LinusNerchino_: i suggest you spend some time with the manual
09:16:01erchino_hehe I guess I have to...
09:16:14 Join kaaloo [0] (
09:16:36 Join blithe [0] (
09:17:08GodEatererchino_: if you have questions *after* you've read the manual - then please come back. We like questions from people who demonstrate they've done their homework :)
09:17:18 Part kaaloo
09:17:43 Join davina [0] (
09:19:41 Join pseudo__ [0] (
09:20:08erchino_how about video and pics
09:20:16erchino_doesnt say anything about that...
09:20:25scorchesure it does
09:23:06 Join petur [0] (n=petur@rockbox/developer/petur)
09:23:53*petur spots 3 oops commits on todays frontpage :)
09:24:09peturactually 4...
09:24:15LinusNwe're getting sloppy in vacation mode
09:25:08peturLinusN: is it in FAT spec that the . and .. entries are always first and second? Or shall I read the spec tonight ;)
09:25:10*jhMikeS doesn't need vacation to be sloopy :P
09:25:35LinusNthe fat spec says they have to be in entry 1 and 2 when you create a directory
09:25:46peturah ok.... thanks
09:26:06 Join miepchen^schlaf [0] (n=hihi@
09:26:13*petur should read the spec *before* trying to fix things
09:27:26*jhMikeS wonders if the build is getting sticky again
09:27:35 Join stripwax [0] (
09:28:22LinusNsomething is wrong with the builds - it repeatedly send the 8MB recorder builds to the servers, but nobody builds it
09:28:39 Quit pseudo_ (No route to host)
09:28:45peturhaha nobody wants it anymore?
09:31:06jhMikeSnicely kicked now?
09:31:23 Quit stripwax (Client Quit)
09:31:23LinusNi killed it softly :-)
09:31:28*petur wonders why Febs kept a hijacking post in one thread and locked the same Q that was posted in a separate one
09:31:34jhMikeSkill it harder
09:31:45LinusNpetur: me too - he must be in vacation mode :-)
09:31:54 Quit aliask ("ChatZilla [Firefox]")
09:32:23jhMikeSvacation = anagram for vacant io
09:33:27LinusNjhMikeS: :-)
09:34:20peturthings are silent today, no gregl ?
09:34:41LinusNi think he has many good points
09:34:58petursure, but I don't think we need a revolution
09:35:15LinusNwe have been quite sloppy for some time
09:35:45LinusNbut we don't have the same overall goals as he might think
09:37:39GodEaterworld domination being one of them
09:37:52LinusNregarding stability and having a solid code base, i think we have been too eager to port to new targets instead of making the swcodec code more solid
09:38:20LinusNbut then again, porting is a lot more fun than bugfixing a complex playback engine
09:39:03GodEaterI don't see how you can force people to work on something that they don't want to do
09:39:10LinusNme neither
09:39:11jhMikeSisn't JdGordon separating the buffer code? that should make it much easier to go through playback. it's full of race conditions and needs mutexing between codec/audio
09:39:16GodEaterbugfixing of the swcodec code will happen IF someone wants to do it
09:39:30LinusNjhMikeS: yes, that work is important
09:39:42erchino_thanks guys again
09:39:44erchino_see you later
09:39:46 Part erchino_
09:40:23jhMikeSif any new yields appear somewhere or aren't present in a build, the timing falls apart between those two threads
09:40:47LinusNjhMikeS: it's horrible as it is now
09:40:53pondlifeThe way that most playback bugs don't appear on the sim is a bit of a giveaway..
09:41:49LinusNhowever, the audio playback in rockbox *is* complicated
09:42:03pondlifeHopefully a careful rewrite (as Nico_P and JdGordon are doing, with the rest of us keeping an eye on their code) will simplify it..
09:42:06jhMikeSI think audio playback is complicated no matter what
09:42:15LinusNsince we need two simultaneous codecs
09:42:19GodEateris this because it was a "feel our way" implementation that's never been tidied up, or there's simply no way to make it any less complex ?
09:42:51LinusNwe have a lot of requirements that make it complex:
09:43:02pondlifeMore the former, I think. It's evolved...
09:43:11jhMikeSwhere's the two simultaneous codecs things for?
09:43:14LinusN1) voice playback mixed with audio
09:43:14*GodEater sees differing opinions here :)
09:43:43jhMikeShow about three? voice/audio1/audio2 (for low memory crossfade)?
09:43:53LinusN2) needs to buffer as much as possible to reduce disk usage
09:44:09pondlife2 ?
09:44:17LinusN3) needs fast response to user input (skipping, winding etc)
09:44:36LinusN(2) and (3) are somewhat in conflict
09:44:37jhMikeSI'm doing some PCM stuff to get it to work on arbirary cores and I'll be comfortable mixing up cores between audio/codec
09:44:42pondlifeLinusN: for (2), why is a codec involved in buffering?
09:44:54pondlifeMetadata parsing, yes, but no need to run the codec...
09:44:55LinusNdid i say that?
09:45:16pondlifeNo, I thought you were giving reasons for simultaneous codecs.
09:45:31pondlifeBut now I realise you mean complexities
09:45:32GodEaterno - they were reasons why audio playback is complex :)
09:45:34jhMikeShaving low crossfade mem requirements would be awesome for low latency output implementation otherwise it's really demanding on ram
09:45:36 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:45:49LinusN4) crossfade :-)
09:46:04GodEaterI was going to ask why 2) is anything to do with audio playback at all - but I see it is because of 3) now
09:46:10pondlifeHopefully jhMikeS's ideas can result in voice being moved well out of the way...
09:46:20LinusNthe codec swapping is a nasty complexity villain
09:46:40 Join spiorf [0] (
09:46:56jhMikeSbut can we have a minimal voice codec implementation with just enough to decode voice files?
09:47:00GodEaterLinusN: are these points wikified anywhere ?
09:47:02LinusNi would like the playback to be (re)built from the ground up
09:47:08pondlifeWhat's stopping us having a special voice codec linked to a different address?
09:47:21LinusNso, for example, pcm playback wouldn't depend on the playback states
09:47:36LinusNpondlife: IRAM usage
09:47:49LinusNthe codec swapping is about IRAM usage
09:47:51pondlifeDid anyone attempt voice playback without IRAM?
09:47:59jhMikeSpcm playback depends on playback states?
09:48:16LinusNjhMikeS: example: no voice playback when paused
09:48:17pondlifeThere was a theory that mono, low-bitrate stuff could do without IRAM.
09:48:23jhMikeSah, yes
09:48:46LinusNjhMikeS: and the recent bug with the headphone unplug
09:48:47jhMikeS11kHz to boot
09:49:13LinusNi'd like a separate voice codec built in the core
09:49:19jhMikeSodd that headphone unplug is taken as a button event :\
09:49:27GodEateranother dataabort. That's two in as many days. And here I was saying the ipod version wasn't that bad.
09:49:41LinusNbut then the binary-size fetishists will kill me
09:49:55GodEaterLinusN: all one of him ? :)
09:50:03LinusNjhMikeS: not button event, just event
09:50:06jhMikeSit should be an interrupt...I hate all this polling stuff when IRQs are more efficient
09:50:47LinusNjhMikeS: it still needs to be sent as an event to the main thread, irq or not
09:50:49jhMikeSbuild needs a big kick in the pants
09:51:02 Quit midkay ("Leaving")
09:51:15 Join midkay [0] (n=midkay@rockbox/developer/midkay)
09:51:28LinusNjhMikeS: one of our servers is misbehaving...
09:51:29jhMikeStrue but it only takes CPU time when it happens then
09:51:43 Join Wiwie [0] (
09:51:52*GodEater hopes it's not his server
09:52:39 Quit perplexity (Read error: 113 (No route to host))
09:55:20*jhMikeS jumps angrily on the ground hoping the shockwaves reach the other side and restart the builds
09:55:52LinusNi suspect
09:56:06jhMikeSmust be with Ms. Piggy
09:56:28pondlifeNow I need early beer.... :/
09:56:39pondlifeTo wipe it all away
09:56:44LinusNit never finishes the ifp7xx sim build
09:57:16pondlifeerror: build config too obscure
09:57:48pondlifebad joke, don't worry
09:57:54*pondlife goes to work
10:00:11 Quit XavierGr (Nick collision from services.)
10:00:11 Join XavierGr_ [0] (
10:03:43 Join kretender [0] (
10:03:54huzzwhat formats can be played for video via rock box
10:04:05huzzi know mp4 doesn't work :/
10:05:03dionoeacheck the MpegPlayer wiki page
10:05:34dionoea+hi everyone
10:06:15LinusNjhMikeS: yellow build for gigabeat
10:06:27 Join desowin [0] (
10:07:01pondlifeInteresting colouring in the build log...
10:07:25jhMikeSthat's an odd one given I hadn't touched anything re: LCD_OFFSET :\
10:07:56jhMikeSwoops, guess I did :P
10:08:30 Join XavierGr [0] (
10:21:22 Quit bdgraue (Remote closed the connection)
10:22:52 Quit XavierGr_ (Read error: 110 (Connection timed out))
10:23:10 Quit XavierGr (Nick collision from services.)
10:23:13 Join XavierGr [0] (
10:24:26 Join bluebrother [0] (i=hZte9kWn@rockbox/staff/bluebrother)
10:24:41 Join bdgraue [0] (
10:36:46 Nick pseudo__ is now known as printfXh4 (
10:42:34 Join obo [0] (
10:44:46 Quit bdgraue (Read error: 104 (Connection reset by peer))
10:45:07 Join bdgraue [0] (
10:51:33 Quit Bspec (Remote closed the connection)
10:52:27pondlifeIs there a wiki page (or somewhere) that shows (current) codec performance? A table of percentages with targets across the top and codecs/bitrates down the side...
10:53:42amiconnpondlife: For good results we need that standard test track...
10:54:22pondlifeThere must be some freely licensed 3 minute song we could use....
10:55:12pondlifeAha, missed that one.
10:55:40pondlifeSearching for "codec performance" (without quotes) didn't find it.
10:58:34*jhMikeS could donate one of his MIDIs in PCM format but fears laughter :p
11:00:15amiconn3 minutes would be too much
11:00:39 Quit printfXh4 (Read error: 113 (No route to host))
11:00:50amiconnThe track should be <= 30MB (better <= 14MB) uncompressed
11:01:48 Join Entasis [0] (
11:03:00*jhMikeS has one that's 1:02 but it's too weird
11:04:32 Quit Nibbier (Read error: 110 (Connection timed out))
11:04:58***Saving seen data "./dancer.seen"
11:06:23markunI found some creative commons tracks which I think could be used, but have the bookmark at home
11:07:11markunah, found it:
11:07:19amiconnLinusN, jhMikeS: Regarding the voice codec, there is another potential conflict I am not sure what to do about
11:08:02amiconnNow that we have a speex codec, it could make sense to switch to speex for voice files
11:08:17amiconnBut then the .talk clips would also need to be in speex format...
11:08:30 Join Arathis [0] (
11:08:38 Join haemmy [0] (n=stefan@
11:09:25jhMikeSwe could use both with a little finessing but not at once
11:09:48LinusNi don't get it, what is the problem with .talk clips in speex format?
11:10:29jhMikeSvoice in mp3 and talk in speex would make a lot of nasty decoder switching
11:10:30amiconnspeex is a specialised voice codec
11:10:45amiconnAnd the .talk clips can be anything
11:10:45 Join Beta-guy [0] (
11:11:07LinusNwho uses .talk clips for non-voice audio?
11:11:08amiconnPlus it may be confusing for the user that the .talk clips don't work
11:11:23LinusNwhen would they not work?
11:11:23amiconn...which could happen if they selected the wrong .talk format in voicebox
11:11:50Beta-guyI've got a question about rockbox, I have an Ipod 5.5G if I run into a problem installing rockbox or want to go back to the apple firmware can I just update it with itunes?
11:11:52LinusNyou would not select the format in voicebox
11:12:01LinusNyou would select the target
11:12:03Beta-guyalso is there any change I'll brick the ipod?
11:12:11amiconnHow would the .talk clips work on archos then?
11:12:28LinusNbecause they would be in mp3 when you select archos as the target
11:12:52LinusNand it would be included in rbutil
11:14:07*amiconn doesn't like all-in-one tools
11:14:24LinusNok, so keep voicebox separate then if you wish
11:14:34LinusNit would still work the same
11:14:57LinusNi.e you select the target platform and it uses the appropriate format for the .talk clips
11:16:30bluebrotherBeta-guy: you'll always be able restoring using itunes
11:16:33LinusNit would be perfect to have the speex codec builtin, without iram to avoid swapping
11:16:39LinusNif possible
11:16:49amiconnVoicebox (or whatever) could tyr to autodetect the target (if it's run directly on the target drive)
11:17:05LinusNas it does today
11:17:29LinusNrbutil autodetects the target
11:17:54amiconnOnly for ipods afaik
11:17:59 Quit borisyeltsin (Remote closed the connection)
11:18:23 Join borisyeltsin [0] (
11:18:37amiconnAnd it needs administrative privileges to do so
11:19:00peturnope, it detects my h380 fine
11:19:13LinusNautodetection.cpp uses the rockbox-info.txt file
11:19:29LinusNand it works for all targets
11:20:01LinusNamiconn: but you are correct when it comes to the autodetection for the bootloader installation
11:20:15LinusNhowever, when rockbox is installed, it can detect all targets
11:20:55LinusNso it would work perfectly for .talk clip creation
11:21:22amiconnI made some suggestions how the archos targets can be autodetected for installation (in most cases)
11:21:48LinusNsearching for ajbrec.ajz/archos.mod?
11:21:51 Join carlitoco [0] (
11:21:54LinusNand comparing sizes?
11:22:55amiconnchecking for archos.mod / ajbrec.ajz to detect an archos. Checking the ajbrec.ajz header for the type
11:23:31amiconnIf no ajbrec.ajz is present, Ondios can still be autodetected in most cases, by checking for its settings file
11:23:35LinusNperhaps it would suffice to check the size of ajbrec.ajz to find out if it's recv1 or fm/v2?
11:23:47amiconnfm/v2 must still be told apart
11:23:54LinusNah, yes...
11:24:03amiconnAnd checking the header would work even if rockbox is already installed
11:24:31amiconnThe header check is quite simple, we know the values
11:25:56LinusNhow is the speex performance nowadays?
11:25:57pondlifeGoing back a bit, what's the great advantage of speex over mp3 - less CPU intensive? Less RAM use?
11:26:33LinusNbetter compression iirc
11:26:34pondlifeComparing mp3 without IRAM vs Speex without IRAM.
11:28:14preglowbetter compression for speech, potentially faster given optimisation
11:28:39preglowless ram too, i think, the entire narrowband decoder should fit in iram these days
11:28:55preglowLinusN: performance isn't too great right now, i think
11:29:29 Quit Beta-guy ("Leaving")
11:30:28pondlifeCould not we build (via #ifdef hell?) an mp3 decoder for voice (mono, no IRAM), resulting in a smaller codec.
11:31:50LinusN"realtime on gigabeat" - surprise
11:32:01pondlifeThe main advantage of speex is the patent issue, IMHO. Given that the Archos targets that would benefit from better compression can't use it. files will use one cluster anyway.
11:32:29 Join tedr0ck [0] (
11:32:41LinusNwell, i see a huge benefit of having the voice codec builtin with no iram
11:33:02LinusNeven if llorean hits me in the head
11:33:13pondlifeGet rid of that stupid swapping.
11:33:26pondlifellorean? amiconn, surely...
11:33:29 Quit haemmy ()
11:33:50LinusNllorean is the big binarysize defender in the forums :-)
11:34:02pondlifeAha, amiconn's attack dog :)
11:34:14carlitocoSorry did you know wich Filesystem the Ipod mini 4 gen. got?
11:34:48LinusNcarlitoco: it depends
11:35:20LinusNcarlitoco: it is either fat32 or that mac filesystem whatever it is called
11:35:51amiconnIpod mini 4 gen??
11:35:53LinusNand rockbox only works with fat32
11:35:59pondlifeI'd think it not much work to build a second copy of mpa.codec with the IRAM disabled and associate it directly with .talk files. Then see how that works on various targets...
11:36:27pondlifeIf performance is ok, link it to a different address and ....
11:36:28LinusNpondlife: interesting idea
11:36:42pondlifeIf no, get optimising (mono, limited bitrate)...
11:37:09pondlifeJust some more #ifdef in the existing codec source.
11:37:20LinusNi don't think you can get much speed optimization by stripping the codec
11:37:34LinusNbyt you can cut some space
11:37:55pondlifeYes, then we'll be able to plan a sensible memory layout.
11:38:16 Quit Wiwie (Read error: 113 (No route to host))
11:38:31pondlifeJdGordon, Nico_P: The buffering code looks good, works fine here. Nice that bufread() is no longer needed too.
11:39:41 Join webguest94 [0] (i=589b0063@gateway/web/cgi-irc/
11:40:10pondlifeLinusN: What was the devcon decision regarding the position of codecs in memory? Any changes planned?
11:40:25pondlifeI saw the diagrams, but not the outcome.
11:40:25amiconnI am not sure anymore whether having the voice codec built in would be a good idea
11:40:41amiconnThe added binary size will hit the iFP
11:40:45 Join printfXh4 [0] (
11:41:05webguest94Hi, there is a simple way to edit the language files (.lng)? I just want to change something there
11:41:26scorchewebguest92: any text editor will do
11:42:13pondlifeamiconn: Will we not be able to free up the space used for swapping at the moment? I'd hope that it wouldn't increase binary too much.. Although the iFP is going to be on the sharp end, indeed.
11:43:37pondlifeHopefully it could also be planned to only eat into the buffer if voice is going to be used.
11:43:50 Quit webguest94 (Client Quit)
11:44:00amiconnHmm true, the swap buffer can be removed...
11:44:50pondlifeAnd all the must be using a little bit of battery up...
11:45:06pondlifeswapping got removed from that sentence.
11:49:39pondlifeReal basic question - does IBSS_ATTR indicate IRAM usage?
11:50:52JdGordonpondlife: bufread will be used also, its just nice to have 2 sepearte ways to get data
11:51:06 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
11:51:17pondlifeI was hoping that would be needed for normal playback - no copying required...
11:51:22amiconnpondlife: Yes, if enabled by the #define
11:51:23pondlifenot be needed I mean
11:51:40JdGordonnot for playback, but it might be useful, for metadata stuff
11:51:41*JdGordon gones
11:54:12safetydanThere's some some obvious optimisation points in speex with the fixed point functions.
11:54:24safetydanthere's already some arm opts, just lacking coldfire
11:55:36 Join webguest13 [0] (i=c1bcd4f0@gateway/web/cgi-irc/
11:56:04 Join Thundercloud [0] (
11:56:31LinusNhonestly, i couldn't care less about the ifp
11:56:46 Quit webguest13 (Client Quit)
11:57:06LinusNbut that's me
11:57:26LinusNthe ifp is such an oddball compared to the other targets
11:57:48LinusNand the ifp constraints shouldn't affect the other 20 targets
11:58:11LinusNthe same goes for the archos player imho
12:00:40 Part maffe
12:00:44 Join maffe [0] (
12:03:51 Join pseudo_ [0] (
12:04:02 Quit tedrock (Connection timed out)
12:04:25 Join Wiwie [0] (
12:11:35peturwhat, no fork/freeze archos discussion now? :p
12:12:03 Quit printfXh4 (Read error: 110 (Connection timed out))
12:15:49safetydanOn arm, what's the difference between add and adc?
12:20:22dionoeathe last letter
12:20:27*dionoea hides
12:20:38safetydandionoea, well yes... :P
12:21:34safetydanah, add with carry. Now I wonder why the speex fixed point functions use adc, and the libwma ones just use add?
12:27:51 Join perplexity [0] (
12:31:38oboCould someone close a couple of flyspray entries for me please?
12:32:26peturobo: no, it's time you get the needed rights there too ;)
12:32:56*petur summons Zagor or LinusN
12:45:29 Join Febs [0] (n=chatzill@rockbox/administrator/Febs)
12:47:50 Join joshin [0] (n=joshin@VDSL-130-13-252-16.PHNX.QWEST.NET)
12:51:11 Join crop [0] (i=c27f0812@gateway/web/cgi-irc/
12:52:45cropDoes RB still strive to support iFP? I thought the port is dead since RB doesn't really work on that DAP and there is no progress. And if some decisions are influence by the fact that RB "supports" iFP it could be abandoned IMHO
12:53:45amiconnThe iFP port is a one-man thing
12:53:56amiconnAnd tomal isn't around here often
12:53:58 Join Nico_P [0] (
12:54:05pondlifeHi Nico_P
12:54:46pondlifeAs I said a while ago: "The buffering code looks good, works fine here. Nice that bufread() is no longer needed too."
12:55:16Nico_Pcool :)
12:55:35 Part maffe
12:55:40pondlifeAny progress with multi-thread testing/incorporating?
12:55:44Nico_Ptoday I think I'll try tunring it into a plugin
12:55:54Nico_P... no
12:56:00cropamiconn: but isn't that DAP so special and different from the other supported platforms that it's nearly pointless to try to have full sized RB on it? It also has quite different constraints (mem etc.)
12:56:38pondlifeYes, a plugin would be a good next step. To do the same file spooling. I assume?
12:57:00Nico_Pprobably, but with better simulation of playback vs buffering
12:57:13Nico_Pand maybe add a bit of metadata
12:59:29pondlifeMaybe just start a straight port (with input filenames hard-coded, or read from a .txt file)... then put the output code on a separate thread
12:59:53pondlifeIt'll be easier to spot corruptions while you're just dealing with file i/o.
13:00:48pondlifeAny problems will likely appear when the buffering code has to wait for some free buffer
13:01:12pondlifeOr the output code is blocked waiting for the buffering code.
13:01:28Nico_Psomething I haven't been able to test i keeping one handle on the buffer whilst dealing with others
13:01:34 Join B0FF [0] (
13:01:51Nico_Pbut with my latest changes I expect it to at least almost work
13:04:54B0FFhi, i need some help for evilg fusion, best a german, cause my english isn't that good, thx
13:05:02***Saving seen data "./dancer.seen"
13:05:17*Nico_P takes off for lunch
13:09:49amiconncrop: I don't think it's that special. It just has less resources (RAM and CPU power) than the other swcodec targets
13:10:08amiconnMaking rockbox run on the neo would be harder...
13:10:57LinusNstill, there hasn't been much progress on the ifp lately, has there?
13:11:26LinusNi think the ifp has to adapt to rockbox and not the other way around
13:11:38amiconnBut unlike the gmini, rockbox plays music on the ifp afaik
13:12:35LinusNand in this case, i think we could just disable voice playback on the ifp
13:13:11LinusNyou really can't have it all on such small targets
13:16:10LinusNpetur: i'm here now
13:19:09 Quit Wiwie (Read error: 113 (No route to host))
13:21:50pondlifeLinusN: peter was hoping you could extend obo's Flyspray rights...
13:22:56cropamiconn: yes, and that makes the full RB hardly possible on iFP
13:25:59cropLinusN: he, how can ifp adapt to RB?
13:26:36cropBy committing a suicide?
13:26:42LinusNby stripping features etc
13:27:41cropBy that would still mean RB adapting to ifp. Then ifp will have "The pure rockbox" (tm) :-)
13:28:18LinusNpretty much, yes
13:29:01cropMaybe we should do this for all targets. Yesterday you mentioned that RB becomes bloarware ;-)
13:29:03LinusNbut we shouldn't let the ifp constraints be a burden until ifp is a first-class rockbox citizen
13:29:33crop...which will never happen I think
13:29:46LinusNi don't think so either
13:30:29*petur executes RFL instruction (return from lunch)
13:31:19croppetur: ... and is now probably executing RTL instructions (read the log)
13:31:50peturyup - but I didn't miss much... gregl still not here? :p
13:32:43peturand pondlife brought over my message to LinusN so all is well :)
13:33:07croppetur: wasn't it gregj (l -> j)?
13:33:26*Nico_P executes RFL too
13:33:54peturah yes, and he's here
13:34:47safetydanwoo, a simple copy and paste from Tremor to speex and coldfire goes from 200% realtime to 220%
13:35:06Nico_Psafetydan: what did you copy and paste ?
13:35:27cropNico_P: performance statistics from gigabeat? :-)
13:36:10safetydanNico_P, one of the multiply functions
13:36:53safetydanIt's probably not correct, as I'm clueless when it comes to asm, but it works and doesn't seem to break playback.
13:37:39 Join spiorf_ [0] (
13:38:10 Quit miepchen^schlaf (Read error: 113 (No route to host))
13:38:55 Quit spiorf_ (Client Quit)
13:39:12 Join spiorf_ [0] (
13:39:49 Quit B0FF ()
13:40:39peturBOFF, not BOFH?
13:41:09Nico_Pwhat's BOFH ?
13:41:24peturwhat???? you don't know BOFH?
13:42:26*petur fears we won't see Nico_P for the rest of the day....
13:42:33 Quit safetydan ("Ex-Chat")
13:42:43peturgo read
13:43:31Nico_Ppetur: nover heard of it :p
13:45:42 Quit spiorf (Read error: 110 (Connection timed out))
13:58:36 Join Lear [0] (i=chatzill@rockbox/developer/lear)
14:04:18 Quit bdgraue (Remote closed the connection)
14:04:54 Join seeee [0] (
14:06:43Nico_Pmy plugin compiles but doesn't do anything yet :p
14:07:36GodEaterNico_P: what's it going to do ? :)
14:08:53Nico_PGodEater: it will be a multithreaded test of the buffering API
14:09:01Nico_Pnot very interesting :)
14:11:11 Quit carlitoco (Read error: 104 (Connection reset by peer))
14:12:17 Quit desowin (Read error: 113 (No route to host))
14:12:53 Join desowin [0] (
14:13:11cropNico_P: why does it have to be tested as a plugin? Why not on PC?
14:13:28Nico_Pcrop: it was tested on PC up until now
14:14:02Nico_Pyou can see the code in svn://
14:14:19cropBut then there's no need for tests as plugin. The code is not platform or env specific in any way I suppose?
14:14:56LinusNcrop: no, but the threading in the target is rather specific
14:14:56Nico_PI want to start introducing multithreading
14:16:01cropAnd isn't any buffer inherently not thread safe? I.e. waht needs to be tested are the "users" of the buffer, i.e. they must synchronize their access
14:17:31Nico_Pthat's exactly why I want to start dealing with the multithreading
14:17:42Nico_Pwhich is why I need to write a plugin
14:17:57Nico_Pthe porting was quite straightforward really
14:18:08Nico_PI needed to add a bunch of "rb->"
14:19:01 Join Wiwie [0] (
14:19:43cropSo you actually want to test the callers? And not the buffer itself? I'm just curious.
14:20:49 Quit Wiwie (Client Quit)
14:20:50Nico_Pwhat do you mean ?
14:20:53cropOr will there be a buffer wrapped by a sync layer? So that the callers don't have to worry about ot?
14:21:03cropOr will there be a buffer wrapped by a sync layer? So that the callers don't have to worry about it?
14:21:21Nico_Pthe buffer should be transparent to the callers
14:21:58cropI mean, to test the caller syncing, you don't need a buffer, just an int would do.
14:23:28Nico_PI want to test more than that
14:23:59cropNico_P: e.g. what happens if two callers simultaneously call bufgetdata()? If they are not synced the data will probably get corrupted. So the callers must sync
14:24:59cropSo I don't understand why do you need multiple threads to test the *buffer*
14:26:17Nico_PI'm testing the buffering API more than the buffer
14:26:40cropThe buffering API itself can (and IMHO should) be tested single threaded since syncing is not the task of that API
14:27:00cropIIUC of course
14:28:08Nico_Pthe buffering API should end up being the means of interacting with a buffering thread
14:28:52Nico_Pthe buffering thread will manage the buffer level and fill it when needed
14:29:28cropNico_P: aha, so other threads will send messages to the buffering thread? And those messages will get processed sequentially?
14:29:40Nico_Pat least that's how I see it
14:29:42ender`can somebody suggest me a good visual diff tool?
14:29:50Nico_Pender`: kompare ?
14:30:08cropThen this buffering thread (or queueing in RB) is the sync layer I meant above
14:30:24Nico_Pcrop: yes
14:30:26ender`preferrably something that runs (or compiles) on windows :)
14:30:34Nico_PI'm not really familiar with the vocabulary :)
14:30:44Nico_Pender`: oh sorry... then I don't know
14:31:11peturender`: WinMerge
14:31:41ender`thanks, i knew there has to be a free tool for this :)
14:31:48peturand somebody once suggested KDiff3
14:32:38petur...but I like WinMerge more
14:32:52 Join ep0ch [0] (i=57c2fa65@gateway/web/cgi-irc/
14:33:17cropBut then again: to test the buffering API, you don't need multiple threads. Single thread would do. Since queueing (=syncing) can be taken as tested.
14:33:56cropYou'd only have to ensure that the buffering API is directly accessed only by that thread
14:34:18Nico_Pcrop: you mean the buffer ?
14:34:25ep0chender: for file comparison on windows Beyond Compare is v.good
14:35:00cropNico_P: I mean the funcs from the buffering API
14:35:18Nico_Pcrop: well these are meant to be called from other threads
14:35:29Nico_P... I think
14:35:47 Quit SirFunk (Remote closed the connection)
14:36:05Nico_Pnot in their current state of course
14:36:15cropNico_P: so these funcs will send messages to the buffering thread? Err... I thought the funcs are here to access the "raw data"
14:36:32cropAnd syncing must be done in another layer / by other means
14:37:04cropE.g. by accessing the funcs only via sending messages and only allowing the buffering thread to directly call the funcs
14:37:12 Quit perplexity (Read error: 113 (No route to host))
14:37:13 Quit alberink_ (Read error: 110 (Connection timed out))
14:37:36cropBut I'm not an expert in the API you've designed :-) You should of course know better
14:38:17cropIt should be only clearly stated what are the responsibilities of the API and what are the caller expected to do
14:38:47Nico_Pcrop: I didn't design it, but the way I see it, the functions that are in the wiki page are meant to be used by external threads. They queue messages for the buffering thread, which will then write the data in the buffer
14:39:28Nico_Pso the code that's currently in the functions would have to be moved to functions that aren't accessible externally
14:39:42Nico_Pand replaced by code that sends messages to the buffering thread
14:40:52Nico_Pso the buffering thread would be the syncing layer, wouldn't it ?
14:40:52cropNico_P: I couldn't find any infos about who'd be responsible for syncing on that wiki page. As it stands there I'd rather assume that the callers are.
14:41:34Nico_Pwhat exactly do you mean by "syncing" ?
14:41:55Nico_Pensuring the buffer isn't written by two different callers at the same time ?
14:41:56cropcaller synchronization. A long word, hard to type :-)
14:42:52Nico_Pcan you ellaborate ?
14:43:48cropNico_P: if two callers call a function (or two different funcs from that API) at the same time what is supposed to be the result?
14:44:29Nico_Pcrop: two messages get sent to the buffering thread, which will treat them sequentially
14:44:30cropIf the API does syncing the first caller gets through and the second will wait nd then also get through. But the callers don't have to sync explicitly
14:45:10cropIf the API doesn't sync, both get through but the data will be in an inconsistent state, so the result is unpredictable
14:46:06cropNico_P: from your answer I conclude that, from the callers' perspective, we have the situation A, i.e. they don't have to sync
14:46:13Nico_Pso the second should be blocked until the API has finished with the first one ?
14:47:10cropBut if the syncing is done via queues (which are asynchronous API iiuc) how can you return values?
14:47:16Nico_Pcrop: some calls would have to be blocking
14:47:38cropIs anyone else following the discussion? Does it still make sense?
14:48:21Nico_Pfor example bufopen() could return a handle id immediately, but bufread or bufgetdata can't return a value at any time
14:48:27cropNico_P: but if the calls are done via messages how can return values be returned?
14:49:18Nico_PI was thinking of a queue of handles, which get added to the buffer when there is free space
14:49:39Nico_Pbufopen() would add a handle in the queue and return its ID
14:50:03 Join Febs_ [0] (n=chatzill@
14:50:23Nico_Por it would add a request for file A to be buffered
14:50:26cropNico_P: IMHO the semantics of the API has to be clarified. Then it will be easier to develop and to test
14:51:25 Quit Febs ("ChatZilla [Firefox]")
14:51:30 Nick Febs_ is now known as Febs (n=chatzill@
14:51:44Nico_Pprobably... thingh is I don't have any experience in threaded programming, so I don't really know what should or shouldn't be done
14:52:16cropNico_P: what happens if two threads call bufread() with the same handle? Who wins?
14:53:00Nico_Pthe first to make the call ?
14:53:01 Quit desowin (Read error: 113 (No route to host))
14:53:12cropNico_P: I'd suggest to design the API as NOT thread safe. And then add a syncing layer if necessary.
14:54:01cropNico_P: ah, I forget that we don't have preemtive theading in RB...
14:54:07Nico_Pcurrently it's not thread safe
14:54:09Nico_Pno we don't
14:55:38cropSo we don't need any snying at all if only we guarantee that the buffer data is only accessed via the API and that no funcs from the API call yield()
14:55:57crop*..any syncing
14:56:47Nico_Pthat's basically how I was trying to build the API
14:57:00Nico_P... as the only way to access the buffer
14:57:35 Quit XavierGr (Nick collision from services.)
14:57:38 Join XavierGr [0] (
14:57:45Nico_Pwell by access do you mean read, write, or both ?
14:59:32cropNico_P: both. I mean, you'll have a byte array somewhere. That array may only be accessed directly (e.g. data[i]) from within the API funcs.
14:59:58 Quit atsea- (Read error: 104 (Connection reset by peer))
15:01:15Nico_Pthe codec thread will need to read the buffer directly
15:01:29pondlifeThe thread safety problem comes from the use of bufgetdata() as it's currently specced. - "Note that there is no guarantee how long that the data will remain valid." is the killer phrase.
15:02:06cropNico_P: hm... wait... but there is bufgetdata() which returns a pointer directly to the data. Who will block the buffer so that the next bufopen (e.g.) won't overwrite the data? I.e. there should be something like lock/release.
15:02:07pondlifeThere's no way to lock part of the buffer while it's accessed by a codec.
15:02:09Nico_Ppondlife: is there a way to solve this problem ?
15:02:30croppondlife: he, you read my mind :-)
15:02:35Nico_Ppondlife: we could add a lock mecanism to a handle
15:02:51pondlifeYes. you need bufgetdata to return some kind of handle to the block , and a bufblockunlock() call to free it
15:03:11Nico_PI'm not sure this is even needed
15:03:46pondlifeIt might not be. We don't have this at the moment, and it's not a major cause of problems AFAICT
15:03:51Nico_Pthe way I wrote things, data can only be overwritten once it has been explicitly declared old by bufseeking
15:04:01crop but then it's possible that a block gets locked forever (if something goes wrong) So we need another thread to sweep over... Oh no =:-O
15:04:28pondlifeSo data is locked by bufgetdata() and unlocked by bufseek()....
15:04:41 Part BK--
15:04:46Nico_Pthat's about it, or bufclose
15:04:56pondlifeThat would work for streaming, but maybe not for non-streaming (which I don't think we use yet).
15:05:07***Saving seen data "./dancer.seen"
15:05:19FebsHas anyone else noticed frequent crashes with recent builds on the iPod Video 60GB?
15:05:19Nico_Ppondlife: non straming is when you need the whole file in one block ?
15:06:07pondlifeNot necessarily, but total random access, with potentially large locked blocks.
15:06:21 Join matsl [0] (
15:06:22Nico_Ppondlife: and the data isn't actually locked by bufgetdata... it's locked when it's buffered, until it's released by bufseek or bufclose
15:06:27 Join atsea- [0] (i=ariel@gateway/tor/x-b004dedb9b0dfc0b)
15:06:40pondlifeWith our memory we could never assume an entire file is available.
15:07:19Nico_Pwhat does that imply ?
15:07:38pondlifeNot a lot, just that a file may be > 32MB..
15:08:36 Join Xerion [0] (
15:08:42pondlifeSo " the whole file in one block" is impossible.
15:09:47 Join Soap [0] (n=Soap@rockbox/staff/soap)
15:10:08Nico_Pbut I don't see that as a problem
15:10:08pondlifeSo... don't worry about that case!
15:10:19pondlifeIt's not, that's what I'm saying.
15:10:30Nico_Pok :)
15:13:00 Join The-Compiler [0] (
15:14:10oboFebs: yes, every few hours for me
15:15:05Nico_Ppondlife: have you looked at the code ?
15:15:19pondlifeA little
15:15:23 Quit tedr0ck (Read error: 110 (Connection timed out))
15:15:28pondlifeAnything in particular I should look at?
15:15:33FebsI've had 3 crashes in the past half hour with r. 13795. I cannot replicate these crashes using the archived daily build from July 1. I haven't had a chance to check further to see if I can narrow down when the problem was introduced.
15:15:49Nico_Ppondlife: err not really
15:16:07Nico_PI'm porting the app as a plugin right now
15:16:13FebsThe most recent crash was a data abort at 00030784, but there have been other errors as well.
15:16:28pondlifeIt might be good to put your plugin onto Flyspray or the wiki for a bigger audience to eyeball at some point.
15:16:45oboone was in trigger_cpu_boost for me
15:16:52Nico_Ppondlife: I agree
15:17:56pondlifeAs the playback rewrite occurs, the new code should be well-reviewed. Hopefully lostlogic and jhMikeS will be able to look over it for you too.
15:21:12 Nick Arathis is now known as Arathis|afk (
15:21:41Nico_Ppondlife: I hope you're not expecting me to rewrite the whole platback code alone :)
15:21:50pondlifeHaha, well....
15:22:12pondlifeJdGordon will help I expect. I doubt I'll have much time to write stuff sadly.
15:22:18 Quit seeee ()
15:22:24LinusNmy ipod video crashes all the time with the latest build
15:22:40Nico_Panything to do with amiconn's changes ?
15:22:59Nico_Ppondlife: looks like I'm going to have lots of fun :p
15:23:05Nico_Pthe plugin works now
15:23:13pondlifeNico_P: DON'T PANIC
15:23:15Nico_Pexactly the same as the standalone app
15:23:29pondlifeGreat, is it going into SVN?
15:23:33*Nico_P runs away screaming
15:23:47pondlifei.e. JdGordon's SVN
15:24:06pondlifeLinusN: If it's crashing regularly, should be possible to work backwards...
15:24:22LinusNit either freezes or crashes
15:24:28pondlifeBetter than intermittent stuff either way
15:24:30Nico_Ppondlife: the only problem is the list of files to read is hardcoded
15:24:42oboMine isn't freezing, just data aborts, or undefined instructions
15:24:44Nico_Pmaybe I should add support for a playlist
15:24:44LinusNpondlife: i haven't yet found a pattern
15:25:12oboI don't have my map file to lookup the latest one
15:25:17pondlifeNico_P: Can you do wildcards (IN*.mp3 or something).
15:25:26LinusNi should build my own and keem the .map file
15:25:44Nico_Ppondlife: not sure... how would I do that in C ?
15:26:10 Join SirFunk [0] (
15:26:24 Quit Soap_ (Read error: 110 (Connection timed out))
15:26:27pondlifefindfirst()/findnext(), but I don't know if Rockbox has them.
15:26:59pondlifeLook in the file browser...
15:27:29Nico_Pyou mean the source code of it ?
15:27:38pondlifeYes, I'm just looking myself..
15:28:00peturor the properties plugin...
15:28:15croppetur: he-he
15:29:29*petur was being serious: the code is small and simple - faster to find what you need ;)
15:29:44pondlifeCertainly smaller than tree.c...
15:29:59 Part LinusN
15:30:14pondlifepetur: Do you know if there's any directory enumeration functions available?
15:30:34 Nick Arathis|afk is now known as Arathis (
15:30:51croppetur: I just thought that everybody speaks about what he knows :-)
15:31:09peturthat too... *sigh*
15:31:53peturpondlife: afaik, properties.c does it the easiest way possible
15:32:08pondlifeYep, just found it.
15:32:33Nico_Poh great, the linked list test fails on target
15:32:48pondlifeThat's what tests are for :)
15:34:27Nico_Pit probably has somthing to do with the values I hard-coded for the handle moving
15:34:41Nico_Pafter commenting this part of the test out, the plugin runs fine
15:36:00Nico_P...and only some md5sums are correct
15:36:21dionoeawhich is better than none :)
15:36:23 Quit RaRe (Read error: 110 (Connection timed out))
15:37:04Nico_Pindeed :)
15:37:06 Join thegeek [0] (
15:40:02 Join perplexity [0] (
15:44:56 Join Arathis_ [0] (
15:46:00 Quit Arathis (Nick collision from services.)
15:46:06 Nick Arathis_ is now known as Arathis (
15:48:37 Quit ep0ch ("CGI:IRC (EOF)")
15:49:04amiconnpondlife: Many non-streaming formats require the whole file in one contiguous block
15:49:24amiconnIf the file is then larger than the available buffer, it simply cannot be played
15:51:32pondlifeYes, that was the point I was failing to make.
15:52:51amiconnHowever, the buffering system needs to handle files larger than the buffer for streaming formats
15:56:07 Join RaRe [0] (
15:56:18 Join bdgraue [0] (
16:00:36Nico_Pamiconn: it does
16:01:05*Nico_P is building a logf build to investigate the cause of the problems on target
16:03:56 Join TrueJournals [0] (
16:04:23 Quit webguest92 ("CGI:IRC")
16:04:57 Quit crop ("CGI:IRC")
16:05:26 Quit Zagor ("Client exiting")
16:08:00pondlifeNico_P: One thing to bear in mind is that file i/o calls may yield().
16:08:05 Quit Lear ("ChatZilla [Firefox]")
16:10:18 Part TrueJournals
16:13:40 Join desowin [0] (
16:15:40Nico_Phmm logf isn't very helpful
16:16:26 Part austriancoder ("Kopete 0.12.4 :")
16:18:43 Join jack|Argon [0] (
16:19:40jack|ArgonDoes anyone know where I might be able to contact the writer of ipodpatcher?
16:21:15Nico_Pjack|Argon: here, he's linuxstb
16:21:37jack|Argonok, thanks
16:22:21Nico_Pyou can try mail too
16:22:31jack|Argonexcept he's not on right now. I guess I'll wait (and post something in the forums while I'm doing it). Whats his email?
16:22:47 Quit matsl (Remote closed the connection)
16:26:04Nico_Ppondlife: I have a static value in a function that seems to be corrupted... sounds plausible ?
16:26:37pondlifeWhat precedes it?
16:30:04Nico_Pwhat precedes it where
16:30:09 Quit jack|Argon ("Ex-Chat")
16:30:29*JdGordon back
16:30:31JdGordonwhat i miss?
16:30:45Nico_Pwhat happens is that suddenly the handles ids get crazy
16:31:06Nico_PJdGordon: I have started porting the app to a rockbox plugin
16:31:30Nico_Pon sim it works fine but on my gigabeat some files don't come out right
16:31:53JdGordonjust dumping files? or actually getting data into the codecs?
16:32:00pondlifeNico_P: Put a big, unused static byte array in front of the corrupted variable - does that help ;)
16:33:03Nico_PJdGordon: just dumping files
16:33:26Nico_Ppondlife: I'll try
16:33:52pondlifeIf not, you'll need to use logf to watch the value I guess.
16:34:00Nico_PI am
16:34:26pondlifeAh, so you should be able to find in which routine it changes?
16:34:44pondlifeOops, static function var... sorry
16:34:55pondlifeTemporarily make it a global static?
16:35:09JdGordonNico_P: can you put the code in svn quickly so i can have a look before bed?
16:35:22Nico_PJdGordon: ok
16:35:27Nico_Pseparate file
16:35:33pondlifeI'll have a look too
16:36:23*JdGordon looks forward to monday when his interstate friends go back home and i can go back to being a hermit :p
16:37:49Nico_Pthere is a bufopen() call that returns the id 184549376
16:38:00Nico_Pright after the id 10 :p
16:38:06 Join kaaloo [0] (
16:38:12 Part kaaloo
16:38:37JdGordonNico_P: ah, you just did a striaght copy across... not tryied spliting into threads yet
16:38:40 Join lazka [0] (
16:41:01pondlifeadd_handle() allocates the id, right?
16:41:20pondlifeSo it's cur_handle_id that's going wacky...?
16:41:27Nico_PJdGordon: yes
16:41:31Nico_Ppondlife: that's right
16:41:51 Join ivan` [0] (i=ivan@unaffiliated/ivan/x-000001)
16:42:17ivan`hi. can rockbox navigate by folders, but then show NN Title instead of song filenames?
16:42:27Nico_Por maybe not, because this handle id doesn't seem to be recognised
16:42:37Nico_Pivan`: no
16:42:42dionoeaivan`: no. You can only do that in database mode.
16:42:51Nico_Pivan`: there is a patch for it in the tracker though IIRC
16:43:19ivan`i'll just write a script to rename for now
16:44:11amiconnNico_P: 184549376 is 0x0b000000. Looks quite special...
16:45:12amiconn0x0b would be 11...
16:45:43amiconnI don't know the code, but this looks like it's endian swapped...
16:47:04JdGordonthe handle count is a unsigned int isnt it?
16:47:29JdGordonwhy would it use the worng endianess? and would 10 be wrong also if thatw as the case?
16:47:52JdGordoni.e it would give 0x0a00000000 instead ?
16:48:05 Join AnarkiNet [0] (n=AnarkiNe@about/csharp/regular/AnarkiNet)
16:48:17Nico_PJdGordon: it's signed
16:48:25Nico_Pbecause we might want to use -1
16:48:41JdGordonok, but that shouldnt change endianess...
16:48:42amiconnWrong access somewhere that only happens sometimes (e.g. on wrap), i.e. accessing a char as int or vice versa?
16:49:18JdGordonstatic int cur_handle_id = 0; is the trouble maker?
16:49:28Nico_PJdGordon: maybe
16:49:37JdGordonmaybe pull it out of the function?
16:49:44Nico_PI'll try
16:50:34JdGordonadd a logf() after cur_handle->id = cur_handle_id++; to see whats happening
16:53:55 Join nl1 [0] (
16:54:14 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
16:55:16Nico_Pstill the same problem with the variable global instead of just static in the function
16:55:37Nico_Pbut the logf right after cur_handle_id is changed shows it keeps the right value
16:56:21Nico_Pcur_handle_id: 12, immediately followed by added handle : 184549376 and then buffer_handle(184549376)
16:56:58Nico_Pwhich means 11 became 184549376 somehow in bufopen
16:58:25JdGordonbetween the add_handle() call and return h->id?
16:59:02Nico_Punless the value is actually right but it is displayed wrong by the logf call
16:59:33Nico_Pwhich seems impossible seeing as it's displayed right at some other place
17:00:21*JdGordon too tired to be helpful
17:00:33JdGordongnight, ill be back sunday arvo sometime
17:00:41 Quit JdGordon ("Konversation terminated!")
17:02:14 Quit kretender (Read error: 104 (Connection reset by peer))
17:02:38 Join MrKeuner [0] (n=kudo@unaffiliated/mrkeuner)
17:03:19MrKeunerhi, I get constants reboots in ipod nano 1st gen, until(I guess) the battery is recharged to a certain level. Is that a bug?
17:03:34 Part The-Compiler ("Connection reseted by god")
17:03:54peturyou're not the first one to mention this
17:05:08***Saving seen data "./dancer.seen"
17:06:29dionoeaMrKeuner: it's because the iPod doesn't have enough power to spin the hard drive
17:06:46 Quit desowin (Read error: 110 (Connection timed out))
17:06:51MrKeunerdionoea: mine is a flash based I think
17:07:25 Join jgarvey [0] (
17:07:35 Join My_Sic [0] (
17:08:06peturwhat's "ATA error: -1" again? RB not found or HDD connection failure?
17:09:09dionoeaMrKeuner: arf ... well it must be something similar to not being able to spin the hard drive :)
17:09:44pondlifeNico_P: Mystified here... does it work better if you make cur_handle_id a global?
17:09:51Nico_Ppondlife: no
17:10:06Nico_Plooks like h->id somehow gets corrupted in bufopen
17:10:23pondlifeAh, so cur_handle_id itself is fine?
17:10:32Nico_Plooks like it
17:10:36pondlifeWhat happens to h.
17:11:12Nico_PI commented out the strncpy and I'm testing that now
17:12:01Nico_Phmm, still get wierd ids :(
17:12:17 Quit amiconn (Read error: 110 (Connection timed out))
17:12:18 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
17:12:21 Join BK [0] (
17:12:29pondlifeWhat happens if you rearrange the struct memory_handle to put id later?
17:12:47BKhey, any reason why my .kbd file won't load on boot? it's in the .rockbox folder
17:13:04BKi can load it manually and it works fine
17:13:05Nico_Ppondlife: trying now
17:14:24pondlifeOr what if you logf the values of both h and h->id both inside add_handle and on return?
17:15:53Nico_Ppondlife: looks like after changing the position of id in the struct I get the same thing... I'll try logging h and h->id
17:17:06peturhahaha seems the MR wiki got spammed:
17:17:19Nico_Pwhat's the format for a pointer ?
17:17:59pondlifeI don't know if we have one... Maybe %lX and cast it to an unsigned long??
17:18:07dionoea%p ?
17:18:22amiconnRockbox printf doesn't support a special format for pointers
17:18:29dionoeaah ...
17:18:30pondlife%p wasn't implemented when I last looked
17:18:30amiconnJust use %lx
17:18:55amiconn...and if the warning annoys you, cast explicitly
17:18:56 Join desowin [0] (
17:19:01pondlifeYou only need to see if the value changes anyway.
17:19:22*amiconn would in fact use %08lx
17:20:14peturamiconn: do you know of a guide to repair the broken soldering of the daughterboard (h1x0)?
17:22:26 Quit ivan` ("Leaving")
17:22:27pondlifeNico_P: If things are changing "by magic" you may want to ensure you don't have database or dircache enabled and try again...
17:22:29Nico_Pthat's strange... the values don't change
17:22:50pondlifeOK, so add more logfs... and see when they do.
17:23:45 Join The-Compiler [0] (
17:25:08 Quit lazka (Remote closed the connection)
17:26:31Nico_PI don't get it
17:26:36The-CompilerWhat does "Fix red" or "Get rid of yellow" (in the SVN-changes) mean?
17:26:37Nico_Ppastebin coming
17:26:51*pondlife waits
17:27:02dionoeaThe-Compiler: fixing compile errors and warnings on
17:27:13nl1The-Compiler: red == build errors, yellow == warnings
17:27:24The-Compilerah thanks
17:27:34pondlifeWe prefer green
17:28:50Nico_Ppondlife: the h in the logf output was outputted by bufopen
17:29:44pondlifeOK..."cur_handle->id = cur_handle_id++;" is the suspect, right?
17:30:09Nico_Pdon't see anything else
17:30:19pondlifeWhich order is the output in, it's confusing
17:30:31Nico_Pthe output reads from bottom to top
17:31:05pondlifeThe logf on line 4 gives: cur_handle_id=11
17:31:54Nico_PI'll try moving the incrementation out
17:31:58pondlifeThe logf on line 6 gives: cur_handle->id= 184549376.
17:33:28pondlifeHmm, can't see why the post-increment would cause problems... any gcc gurus around?
17:33:48Nico_Pcan't see either, but can't see any other possibility
17:33:53 Join tedrock [0] (
17:34:04Nico_PI'm testing with the incrementation moved
17:34:57amiconnIs the whole thing available somewhere, including variable declaration and the struct?
17:35:23pondlifeYes, it's in an SVN repository on JdGordon's PC...
17:35:25Nico_Pstill the same
17:35:28*pondlife hunts the url
17:35:41Nico_Pamiconn: I can sends you my version with all the logf calls
17:35:51 Join Wiwie [0] (
17:35:59pondlifeMight as well just pastebin the whole thing.
17:37:18Nico_Phmm, should've formatted it as C :
17:39:03 Join piroko [0] (
17:39:13amiconnIs it guaranteed that (buffer + buf_widx) is always 32 bit aligned?
17:39:43amiconnBecause, if it's not, you fell victim of the arm rotate-on-unaligned-store/load feature
17:40:04pondlifeThat's the problem. Nice feature.
17:40:17Nico_Pamiconn: I doubt it's aligned because I didn't do anything for it to be
17:40:31amiconnOn ipod, where the loader enables unaligned exception, your code would just crash with a data abort
17:40:56amiconnYou must make sure the struct is 32 bit aligned
17:41:01pondlifeNeed to use (buffer + buf_widx + 3) & 4 ?
17:41:20amiconn& ~4
17:41:22pondlifeOR keep buf_widx as a multiple of 4
17:41:32pondlifeamiconn: Oops, silly me
17:41:42amiconnHehe, I made that mistake once
17:42:00amiconnOtherwise I would have had the G5.5/80 running at devcon already
17:42:17Nico_Pwhat does " & ~4" mean ?
17:42:30 Join Segadude [0] (
17:42:33Nico_P~ is bitwise NOT ?
17:42:48SegadudeHey guys!
17:42:51amiconnIt saves you from writing & 0xfffffffc
17:43:03amiconn(and the latter would be incorrect on 64 bit archs)
17:43:09pondlifeShouldn't it be ~3 ? Just to clear the bottom 2 bits?
17:43:16amiconnoh, of course
17:43:20SegadudeDoes rockdoom and rockboy have sound?
17:43:47Nico_Pamiconn, pondlife: so (buffer + buf_widx + 3) & ~3 ?
17:43:51nl1Segadude: yes, no music in doom tho
17:44:11pondlifeYes, but be careful how you increase buf_widx
17:44:29Segadudei see so how do i turn on the sound effects?
17:44:57Nico_Ppondlife: is it enough if I just make sure the alignement is right when I write the structs in memory ?
17:45:19pondlifeI think so, yes.
17:45:28pondlifeAs long as blocks won't overlap!
17:45:49 Join MySic [0] (
17:46:12Nico_Pah yes, potential problem
17:46:31Nico_PI'd better keep buf_widx a multiple of 4
17:47:16pondlifeamiconn: Are there any targets that require other alignments (e.g. 64-bits)?
17:47:18Segadudebye guys!
17:47:27 Quit Segadude ("Quitting!")
17:48:30pondlifeOr.. are there any macros to help with alignment?
17:49:15Nico_Pwhat's the elegant way to keep a variable a multiple of 4 ?
17:49:24pondlifeSame equation
17:49:53Nico_Pbuf_widx = (buf_widx + 3) & ~4 ?
17:50:14Nico_Por even buf_widx = (buf_widx + 3) & ~3 ?
17:50:51dionoeaanything &~3 is a multiple of 4
17:51:36dionoeathe +3 is to make it round to the neirest multiple of 4 >= to the value
17:51:58pondlifeWe should have a macro to do this.
17:52:05Nico_Pthanks :)
17:52:33*Nico_P feels stupid for forgetting about alignment issues
17:52:39Nico_Pthey were mentioned earlier
17:52:41 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
17:52:58pondlifeIt's easy done when you're used to desktop systems.
17:56:16pondlifeWhere would an ALIGNPTR(x) macro live? I assume it should be possible to vary the definition by target.
17:56:34Nico_Pwoohoo !! now all the files come out right
17:56:46 Quit My_Sic (Read error: 113 (No route to host))
17:56:53pondlifeGreat - now for some threading fun!
17:56:55Nico_Pthanks amiconn and pondlife
17:57:03pondlifeHuh, I did nothing
17:57:07Nico_PI don't think I'll have time for that, sadly
17:57:19Nico_Ppondlife: I wasted your time, that's enough to thank you :)
17:57:19 Quit thepenguin (Remote closed the connection)
17:57:30pondlifeWhenever. Update SVN with the alignment..
17:58:04Nico_Pabout that, is it possible to use a symlink to the file in apps/plugins/ ?
17:58:15Nico_Pit would avoid me the copying
17:59:04pondlifeWhy not just move it into the root - split into testplugin.c and buffering.c
17:59:44Nico_Pit's to compile rockbox with the plugin that I'd need a link
18:00:16Nico_Pseems to work
18:01:25 Quit MrKeuner ("Ex-Chat")
18:02:03*nl1 loves the ~900 lines long kbd_input function...
18:04:20XavierGrpetur: thank you for mentioning BOFH, these stories are great and hilarious! :D
18:05:14peturcheck for a new one (almost) each friday...
18:06:07pirokoHmm... who should I ask about a problem regarding the nsf plugin?
18:07:52 Join Beta-guy [0] (
18:08:20Nico_Ppondlife: before updating the SVN I want to find out why the LL test fails
18:08:51pondlifeI thought you had a full good run-through.
18:08:55Nico_Pit fails around the move_handle tests
18:09:04Nico_PI needed to comment these tests out
18:09:25Beta-guyI really want to install rockbox on my Ipod I'm just afrid of bringing it also I have the Ipod 30 GB 5.5 gen would the rockbox 30 gb video be the right update to download?
18:10:51markunBeta-guy: it's always possible to unbrick, and yes, it's the right update
18:11:47pondlifeAh, brick, not bring..?
18:12:03 Join Quequero [0] (
18:12:06 Join sujdik [0] (n=Reprogra@
18:12:10peturyes, bring it to us :)
18:12:12 Join Alonea [0] (
18:12:20markunI just assumed it was a typo
18:12:40markunpondlife: or did you really think he was afrid?
18:12:45Beta-guyyeah it was a typo
18:13:21 Part sujdik
18:13:24Beta-guygrrr I shouldn't type when I'm tired :P
18:13:41pondlifeAnd I shouldn't read when I'm tired :/
18:14:20Beta-guyok, can rockbox play the MP4 video's that's on the Ipod ATM, and last question b4 I install it will I lost the files currently on on the ipod?
18:14:57 Join TTSbox_ [0] (
18:14:57nl1no, you can't play those videos on rockbox and no you wont lose your files
18:15:13 Quit The-Compiler (Read error: 104 (Connection reset by peer))
18:15:23pirokoBeta-guy: But it's always a good idea to backup your stuff anyway ;)
18:15:36Beta-guytrue enough :D
18:15:41peturBeta-guy: you can (dual-)boot into apple OS to watch your movies
18:15:53 Join Siltaar [0] (
18:16:12Beta-guyI saw mention of that in your FAQ I think
18:17:07markunBeta-guy: and maybe you shoulddo it when you are not tired anymore
18:17:43Beta-guyI just need caffine :P
18:18:21pirokoBeta-guy: Maybe you should just get more sleep ;)
18:18:59peturmarkun: tired too? :p
18:19:51markunpetur: a bit :)
18:22:05Beta-guylooks pretty straight forward, do I need to do anything special to setup Dual boot rockbox/apple?
18:22:07 Join TrueJournals [0] (
18:26:26nl1would anyone be against moving all the config*.h files from firmware/export to a new dir for example firmware/config ?
18:26:35peturyou could use RockboxUtility to install everything
18:26:59 Join bobbyguzik [0] (
18:27:02 Quit bobbyguzik (Read error: 104 (Connection reset by peer))
18:27:22 Join Domonoky_ [0] (
18:27:54 Join bobbyguzik [0] (
18:27:56 Quit bobbyguzik (Read error: 104 (Connection reset by peer))
18:28:38 Quit Nico_P (Remote closed the connection)
18:28:40 Quit pondlife ("disconnected has pondlife")
18:31:52 Quit Xerion (" ")
18:33:52 Quit obo ("KVIrc 3.2.6 Anomalies")
18:38:11 Part TrueJournals
18:38:11 Quit Alonea ("ChatZilla [Firefox]")
18:49:00 Join pixelma [0] (i=d5494cce@rockbox/staff/pixelma)
18:51:32 Quit petur ("work->home")
18:51:32 Quit Domonoky_ (Read error: 104 (Connection reset by peer))
18:53:13TTSbox_in viewers.config some files are associated to many programs
18:53:21TTSbox_what does it mean?
18:53:40TTSbox_uses the first existing program in the list
18:56:23Beta-guyis the bootloader how the Ipod dualboots?
18:56:39pirokoBeta-guy: Yep :)
18:56:44Beta-guygood :D
18:56:58 Join laaw [0] (n=laaw@
18:57:01Beta-guyI'm using rockboxutility BTW
19:01:13 Join pondlife [0] (
19:01:35laawHei, got a little problem with my Sansa e270 and rockbox. The case is that every time I run the original firmware it hides the Music folder, so that rockbox cant find it. It aint the bigest problem but kinda irritating to un-hide the folder evry time i connect the device to my computer. Any wise heads out there that can help me?
19:02:45 Quit laaw (Remote closed the connection)
19:03:15 Join laaw [0] (n=laaw@
19:03:59pirokolaaw: Maybe stick the stuff you want rockbox to see into a different folder?
19:04:52 Join PaulJam [0] (
19:05:07laawyou know, i cant see why I didnt think about that. Ofcourse rockbox dont need the music files to be in the music folder :D thank you
19:05:10***Saving seen data "./dancer.seen"
19:05:29alienbiker99has anybody checked out the isignia 4gb player?
19:05:46PaulJamlaaw: you could also set the fileview to show all files. afaik this shows hidden folders too.
19:06:39 Join kretende2 [0] (
19:06:56pirokolaaw: Yep. Both would work
19:09:05laawthank you both.
19:09:17pirokolaaw: You're welcome :)
19:10:37 Quit laaw ("Bersirc 2.2, for external use only. [ - Open Source IRC ]")
19:12:41 Join maffe [0] (
19:13:58 Join Lear [0] (i=chatzill@rockbox/developer/lear)
19:14:34 Join My_Sic [0] (
19:14:35 Quit My_Sic (Client Quit)
19:17:37pirokoIt's so much fun checking out an entire svn tree on dialup :-D
19:18:16Beta-guyI'm now more awake :) what should I do if I do brick it (to unbrick it)
19:19:15pirokoReboot the ipod into diskmode by holding menu and select until the ipod reboots, and then hold select and play to force diskmode. Then hook it up to your computer and run "restore" from itunes
19:19:47pirokoOnce the bootloader is installed though, it's quite difficult to brick your ipod
19:19:50Beta-guygotcha thanks
19:20:18Beta-guyI've always thought it's better saf then sorry :P
19:20:29pirokoYep :)
19:20:58 Join obo [0] (
19:21:24 Join laaw [0] (n=laaw@
19:21:28 Part BK
19:21:49Nimdaeit appears kermit fell out of the build server pool :S
19:22:23Beta-guyRockbox says I have Apple Ipd 5th gen video, it's true I have 5.5 but I was told I needed to download the 30 GB video firmware will rockbox know what it needs to download?
19:23:06Nimdaethe 30gb firmware works on the 5g and 5.5g
19:23:14Nimdaewell 5.5g 30gb
19:23:41 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
19:23:43Nimdaeactually, the different firmwares between 5g and 5.5g is the 32mb mem vs 64mb mem
19:24:15LinusNNimdae: yes, i removed kermit, because the build got stuck for some reason
19:24:41Nimdaelemme look at the server
19:25:10LinusNamiconn: my ipod video is quite unstable
19:25:31Nimdaei don't see any hung up processes
19:25:45Nimdaeand i did a manual build on it earlier
19:26:01LinusNi can't have it play more than a few minutes before it either freezes or crashes withl illinstr or data abort
19:26:12LinusNNimdae: let's try to add it agan then
19:26:18 Join petur [0] (n=petur@rockbox/developer/petur)
19:26:20alienbiker995.5 has 64mb while the 5 only has 32?
19:26:23 Join merbanan [0] (n=banan@
19:26:32Nimdae5.5 30gb has 32, 5.5 80gb has 64
19:26:44alienbiker99oh ok thats what i though
19:26:51LinusNNimdae: done
19:27:10NimdaeLinusN: ok, if it happens again, i'll schedule a reboot on it since it probably needs it anyway
19:27:19LinusNoki doki
19:27:58LinusNgtg, cu later
19:28:00 Part LinusN
19:28:08 Quit ptw419 ()
19:28:45Nimdaepetur: remember my mahna mahna picture?
19:29:08pondlifedoo do du do doo
19:29:22pondlifeNooo, twice in one day :p
19:29:42laawGot a question that might be a bit out of the rockbox topic, but anyways. I got some problems with safely removal of my sansa e270 in winVista, so everytime i disconnect the device it has to rebuild the files, witch thakes about two-three minutes. Anyone know a solution?
19:29:42Nimdaei shaved my face off
19:30:11*petur tries (not) to remember
19:30:38 Join Domonoky_ [0] (
19:30:52 Join saratoga [0] (i=980398fe@gateway/web/cgi-irc/
19:38:21TTSbox_the code I am porting has malloc and free
19:38:38TTSbox_they had been wrapped around other functions
19:40:16TTSbox_now I see them. What is better? to implement malloc functions myself? I don't think trying to eliminate them is a good idea.
19:40:48Domonoky_TTSbox_: if it is possible to eliminate them, its better to remove them..
19:42:10Domonoky_as a plugin you could make your own malloc or i think there could be one in the pluginlib... in the system malloc is forbidden :-)
19:43:59 Part pixelma
19:45:59 Join webguest40 [0] (i=93151003@gateway/web/cgi-irc/
19:46:12 Quit webguest40 (Client Quit)
19:46:25 Join fasteddie [0] (i=93151003@gateway/web/cgi-irc/
19:47:09 Quit fasteddie (Client Quit)
19:47:56pirokosaratoga: Do you know what was wrong with that wma I linked to you?
19:49:16 Quit laaw ("Think your current client is sexy? Check out Bersirc 2.2! [ - Open Source IRC ]")
19:51:52saratogapiroko: linux said it caused problems with his parser, also it was a v1 file and the v1 decoder doesn't work right
19:52:08 Nick pseudo_ is now known as printfXh4 (
19:52:11pirokoSo I should try v2?
19:52:33saratogaI think basically all wmas are v2
19:52:43saratogaso we never tested the v1 part
19:52:56saratogaits almost exactly the same, but theres some minor changes I need to make to support it
20:00:36 Quit TTSbox_ ("Leaving")
20:21:13 Part kretende2
20:23:52 Quit pondlife ("disconnected has pondlife")
20:25:54Quequerohi guys
20:27:52Quequerodo you know who's working at the ipod nano 2G encryption and how I can get in touch with him/her?
20:28:19 Quit piroko ("Stuff to do that involves the telephone... :P")
20:32:12 Quit thegeek ()
20:34:40saratogaQuequero: I don't believe anyone is working on that
20:34:52 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
20:34:56Quequerohi saratoga, what a bad news
20:35:03Quequeroand do you know why?
20:35:17peturQuequero: unless you have some usefull info to share? ;)
20:35:18 Join nerd_boy [0] (
20:35:41Quequeroi did an extensive cryptanalysis of rc4, for several months
20:35:56Quequeroi know the algorithm and its weakness very well
20:36:00saratogaQuequero: I think we just never found anyone who was interested in it
20:36:06Quequeroi tried to contact blackblox
20:36:21nerd_boySorry to play the part of the n00b, but Google search results turned this channel up when I did a search on "CHIP8 IRC". I don't suppose anyone on here *does* know CHIP8... :x
20:36:22Quequerobut he didnt answer (yet) so i'm looking for someone else ;p
20:36:47peturQuequero: we're certainly interested...
20:36:56Domonoky_nerd_boy: there is an Chip8 emulator in rockbox :-)
20:37:03 Part tzafrir_home ("Leaving")
20:37:27nerd_boyI mean more like someone who knows the ins and outs of the language. The internets is giving me conflicting descriptions for the CXKK instructions. :x
20:37:32Quequerowell i'd like to contact someone who's accustomed with ipod internals, i have some questions, and i need some starting-hints to start the analysis by myself
20:38:02saratogaQuequero: well I think what we know about the nano2g is in the thread
20:38:17Quequerothis one you mean: ?
20:38:27nerd_boyAnd upon actually looking at your site, I see you guys are more about jukeboxes... :x
20:38:28peturQuequero: the problem is that the internals og the 2nd gen are quite different from previous ipods
20:38:34nerd_boySorry to have taken up your time. :)
20:38:54peturQuequero: I also think some people at IPL have been looking at it
20:38:58 Part nerd_boy
20:39:40Quequeroyep i saw there's an arm9
20:39:51Quequerobut it's ok for me, i work with arm cpus ;p
20:40:12Quequeroanyway i'm just interested in studying and decrypting the firmware for now
20:40:31peturQuequero: we can always use people with a clue, welcome!
20:40:42Quequerohey petur can i query you for a moment?
20:40:56saratogai think at this point very little investigation of the nano has been done, so you'll probably be figuring out a lot of it's hardware yourself
20:41:10Quequerothanx petur and saratoga
20:42:06*Domonoky_ detects his brother had written the chip8 emulator.. but he is not in the Credits.. :-/
20:45:31 Part AnarkiNet
20:55:34peturDomonoky_: is there a name to be added to CREDITS?
20:56:09Domonoky_petur: yes, he is my brother :-) firstname: alex
20:57:49XavierGrDomonoky_: I think petur will need the full name :P
20:58:04 Join Wiwie [0] (
20:58:04peturAlex Wenger ?
20:58:23Domonoky_XavierGr: see petur isnt dumb :-)
20:58:47*petur wonders if the original author shouldn't be in there too
20:58:47 Quit Mouser_X (Read error: 104 (Connection reset by peer))
20:59:20 Join Mouser_X [0] (
20:59:50 Quit printfXh4 (Read error: 113 (No route to host))
20:59:51Domonoky_petur: i think normally we dont put the name of original authors (of game port) to the credits file..
21:00:11peturwell we put the 'team' name at the end
21:00:17XavierGrDomonoky_: I didn't know that petur knew your last name :P
21:00:28Domonoky_ah.. the teams.. thats new :-)
21:00:37*petur read chip8.c
21:01:06peturbah... gotta run, back in 30
21:03:24 Quit Quequero (Read error: 104 (Connection reset by peer))
21:05:11***Saving seen data "./dancer.seen"
21:09:39 Quit MySic ("Miranda IM! Smaller, Faster, Easier.")
21:13:23 Quit ompaul ("all in faviour")
21:21:15 Join My_Sic [0] (
21:21:44 Join foleyjd [0] (
21:24:00 Quit My_Sic (Client Quit)
21:25:28 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
21:26:03 Join alberink [0] (
21:27:07amiconnNimdae: The 5G 60GB also has 64MB
21:27:15 Join My_Sic [0] (
21:27:25 Quit My_Sic (Read error: 104 (Connection reset by peer))
21:27:33Febsamiconn: I've tested a number of builds on my iPod Video, and the instability appears to have started between r13761 and r13769.
21:28:24FebsThis probably isn't a surprise, but I figured I'd report what I'd found.
21:28:53amiconnIt is a surprise
21:29:51amiconnWhy the flaming hell do these clock setup changes make all the PP5020/PP5022 targets more stable, and only destabilise the 5G (NOT the 5.5G!)
21:30:19 Join My_Sic [0] (
21:31:52 Join DerPapst [0] (
21:32:11DerPapsti got a dataabort on my 5.5G too last night.
21:33:03amiconnDoing what?
21:33:14amiconnIf you used test_codec, that doesn't count
21:33:25DerPapst0x00037e50 nothing.. it didn't do anything.
21:33:44amiconnFebs: Could you tell me the CLOCK_0x2C and CLOCK_0xA0 values from debug->view i/o ports?
21:34:01DerPapsthaven't used it in that moment. and it didn't play anything. but it was in the wps
21:34:22amiconnAnd you didn't use test_codec before in this session?
21:34:29DerPapsti'm still trying to figure out how to use the .map file ^^
21:34:52oboamiconn: both those clocks are 0x0 for me
21:35:01amiconnDerPapst: Do you have a map of the build you're running?
21:35:20DerPapstwanna have it?
21:35:39amiconnyes please
21:36:00DerPapstok.. one sec
21:36:14amiconnobo: 0 is ok
21:36:43oboamiconn: I think from my map that the data aborts I got today were in pcmbuf_request_buffer, but had one yesterday in trigger_cpu_boost
21:36:57amiconnThey're 0 on everything else than ipod color/photo (CLOCK_0xA0 == 0xC0000000) and ipod mini G2 (CLOCK_0x2C = 1)
21:37:03Febsamiconn: likewise, they're both 0 for me.
21:37:29 Join Benedikt [0] (
21:37:38Benedikthi somebody there?
21:37:45Benedikti have a problem or an question ^^
21:37:55pixelmanope ;)
21:38:07pixelmaand just ask
21:38:09DerPapstwe're all dead
21:38:23amiconn0x37e50 is in dsp_input_count(). That doesn't make sense....
21:38:49amiconnI wonder which part of my changes could trigger things like this...
21:39:30amiconnMaybe it's the 78MHz (??!??)
21:40:01Benedikti have rockbox on my 5th gen ipod and i want to use it in my car, now i want to buy me a Load cable for my ipod, to get power from my car batterie..
21:40:22Benediktbut my ipod doesn't play any music, it just say, that it's loading..
21:40:22DerPapstdon't know.. i mean the PP5022 should be able to run 80MHz with ease
21:40:34pixelmaFebs, DerPapst, obo: is that always when you are in the wps?
21:40:54pixelmaor playing music
21:40:58oboBenedikt: let your ipod boot, then whilst holding down the menu button plug in the cable
21:41:07obokeep holding menu for 10+ seconds after you plug it in
21:41:22DerPapstamiconn: i remeber one guy over at iPL recompiled a kernel so that the iPod run with nearly 100MHz and it run stable about an hour or so untill it get too hot.
21:41:52amiconnYes, and the PP5022 is specced 100MHz
21:41:55DerPapsti had it only once... in the wps while the somg was paused.
21:42:36pixelmaand what revisions are all of you using (trying to narrow it down a bit)?
21:42:51Febspixelma: it's definitely been while playig music. But I haven't used Rockbox for anything other than music recently.
21:42:59obopixelma: has been so far, but I tend to leave it on the wps
21:43:16amiconnFebs: Could you perhaps try to find the exact rev where this uncouth behaviour starts?
21:44:05DerPapstpixelma: i think r13791 including gbs patch
21:44:07Beta-guyI installed the bootliader it says it's been installed on my device (ipod) then the next line says unsupported bootloader instal method
21:44:16amiconnobo: With a pure power cable, it's not necessary to hold MENU while plugging
21:44:16Beta-guycan you give me more info?
21:44:17Febsamiconn: I'll try, but I'm going to be away from my computer for much of the rest of the afternoon, so I won't have time to do much today.
21:44:55Benediktobo and the ipod will be still loading up the batterie or?
21:45:06oboamiconn: I thought a pure power cable wouldn't cause those that trouble at boot?
21:45:22FebsI confirmed that the problem exists with the 7/3 daily build, which is r13767. I was unable to reproduce with r13761.
21:45:24Beta-guyI need help is my Ipod screwed?
21:45:32Benediktobo, ah oke thx it works ^^
21:45:37DerPapstpixelma: it's r13771
21:45:45Benediktbye and thx for the help!!
21:45:49 Quit Benedikt ("get satisfied! :: (Gamers.IRC) ::")
21:46:00Domonoky_Beta-guy: i think you used rbutil ?
21:46:33Domonoky_try the normal ipodpatcher installation..
21:47:09amiconnSome more questions: 30GB or 60/80GB models? If the latter, 32MB or 64MB build? What format(s) were playing, and were there any special things in the wps (peakmeters, ...)?
21:47:24Beta-guywhere can I get ipodpatcher?
21:47:33DerPapstamiconn: 5.5G 80GB
21:47:46Domonoky_Beta-guy: ready the install instructions in the manual..
21:47:47obo60Gb, 64Mb (self) build - no peakmeaters, mixture of oggs and mp3s
21:48:11Domonoky_Beta-guy: which ipod do you have ?
21:48:13Febsamiconn: 64 MB build, 60 GB model. Playing MP3s. No peakmeters. Crashes happened with at least three different WPSes.
21:48:34Beta-guy30 GB 5.5 G
21:49:42amiconnFebs: I'm especially interested in whether 13764/65/66 has that problem
21:49:59amiconn(that was after the first bunch of clock setup changes)
21:50:47Domonoky_Beta-guy: have you tried you rockbox install ?
21:51:07Domonoky_and it doenst work ?
21:51:23LearHm, still get "Filetype array full". Anyone else with test_codec compiled that gets it?
21:51:29Febsamiconn: OK. We don't archive each of the current builds, do we? So I'll have to revert SVN to that revision and then compile to test each of those revisions?
21:51:41Beta-guyI haven't turned off the ipod yet because if the bootload was incorrectly installaded I may not get it to boot
21:51:47amiconnFebs: yes
21:51:58pixelmaLear: I think XavierGr reported that too
21:52:07DerPapstBeta-guy: you can always reset it
21:52:09Domonoky_Beta-guy: the error message is wrong.. i just checked the rbutil code..
21:52:10amiconnUnfortunately no daily build was made between the first and second part of changes
21:52:18Domonoky_Beta-guy: it should be fine..
21:52:31Beta-guyok... I'm rebooting the ipod now...
21:52:31Domonoky_Beta-guy: install also rockbox, bevor rebooting..
21:52:45Febsamiconn: not a problem. I should be able to test it. It just may not be until tomorrow morning.
21:52:52Beta-guyI did even after the error message
21:53:04Domonoky_Beta-guy: then it should work..
21:53:22Nimdaelooks like kermit is working again :)
21:53:48Nimdaeamiconn: <delay response> yeah, i realized that after i thought about it
21:53:52 Join piroko [0] (
21:54:28pirokoIs there an easy way to find out the irc name of one of the developers?
21:54:34Beta-guylooks like it works!
21:54:35amiconnFebs: The strange thing is that I tested my changes on 5.5G/80 with 64MB build (r13772). Ran a full runtime test. Rock solid operation for 12h 15 min
21:54:50Domonoky_piroko: there is an IRC nicknames page
21:55:01amiconn(lame vbr mp3, ~195kbps, wps without peakmeters)
21:55:16pirokoGot it :)
21:55:17pirokoThank you
21:55:31pirokoCrud. hcs isn't here :(
21:56:23Domonoky_piroko: he is there..
21:56:41pirokoI mean in the chat room. :-D Sorry
21:56:42Febsamiconn: that is strange. I just had another data abort (E800007A) with r13767. I'll step backwards from that revision as soon as I can and report back to you.
21:56:50 Quit maffe (Remote closed the connection)
21:57:38Beta-guythanks for your help!
21:58:30Domonoky_Beta-guy: no problem. feedback is always nice :-) ( i just fixed the false error message) :-)
21:58:56pirokoSo Adam Gashlin doesn't get on here much?
22:01:50 Quit Wiwie (Read error: 104 (Connection reset by peer))
22:03:03Domonoky_piroko: i think there was a command to ask logbot about when someone was last seen here in the chat.. but i dont remember the command..
22:03:22amiconn/msg logbot_ seen <nickname>
22:03:30***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! ( )'
22:03:47amiconnYOu have to allow messages from unregistered user in order for this to work
22:04:00amiconnlogbot_ isn't registered
22:04:03pirokoApparently I'm not doing that because I didn't get a reply
22:04:15 Quit Mouser_X (Read error: 113 (No route to host))
22:04:33***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! ( )'
22:04:44pirokoIs that a per-client setting or a server setting?
22:04:57amiconnIt's a per-nick setting freenode
22:05:21amiconnYou need to be registered and identified in order to set it
22:05:23pirokoAh. So how do I let logbot_ give me a message?
22:05:24pirokoI am
22:05:56amiconn/ns set unfiltered on
22:06:06amiconnthen repeat the /msg logbot_ thing
22:06:28amiconnYOu only need to change this setting once. Freenode remembers it for you
22:06:35pirokons is an unknown command for me :(
22:06:47 Quit SirFunk (Remote closed the connection)
22:06:51pirokoI got it
22:07:18pirokoOh wow. He was just on 13 hours ago ^_^
22:07:21pirokoThank you amiconn
22:07:55amiconn/ns is a shortcut for /msg nickserv
22:08:07pirokoGotcha. I'm gonna alias it right now
22:08:10DerPapstnice thing.. didn't know that :D
22:08:19amiconnHydra has that by default
22:08:34amiconnLikewise /cs for /msg chanserv
22:09:05 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
22:09:08pirokoI learn something new everyday
22:10:06pirokoI decided that it's faster to just svn update on dialup rather than download a new build everytime a change is made
22:13:09 Quit piroko ("Gotta run!")
22:15:28*amiconn wonders whether something is different between his own builds and teh official ones
22:18:16pixelmajust wondering (don't think it's important) but there is someone in the forums who claims that he got freezes with playing flacs but his mp3s would work... could there be any differences?
22:19:04pixelmaon a 5.5g Ipod 30GB - "recent build" (wasn't that exact)
22:19:32 Join maffe [0] (
22:20:15 Join pondlife [0] (
22:20:35DerPapstamiconn: maybe give Febs a build you've made?
22:21:56 Join boten_anna [0] (i=54bd60bc@gateway/web/cgi-irc/
22:23:17 Join SirFunk [0] (
22:25:09DerPapstwow.. fading out a wma file until playback stops (pausing) takes about 6 seconds o.O
22:25:49 Join RoC_MasterMind [0] (
22:26:09amiconnDerPapst: Not very surprising...
22:28:51 Quit crashmatrix (Read error: 110 (Connection timed out))
22:29:24 Quit SirFunk (Read error: 113 (No route to host))
22:29:52DerPapststill irritating if you think playback stops emmediately.
22:32:42amiconnMy own build of latest svn differs from the official one by 64bytes
22:33:19amiconnThis is with no core changes for ipod, and the recommended arm-elf-gcc and binutils
22:33:47amiconnInstalling it....
22:38:14 Join My_Sic [0] (
22:41:19 Quit RoC_MasterMind ("Leaving")
22:42:30 Join miepchen^schlaf [0] (
22:44:12 Join SirFunk [0] (
22:46:14*petur wonders why kkurbjun didn't commit his rockboy sound patch
22:46:38amiconnI got some kind of crash....
22:47:10amiconnI played jewels with music playing in the background
22:47:24amiconnNow the lcd froze, but music is still playing....
22:47:35amiconnBacklight reacts to keypresses
22:47:59 Join stripwax [0] (
22:49:40ender`amiconn: i've seen a crash like that a few times when skipping/inserting new song into playlist (i've got H120)
22:54:07DerPapstamiconn: mhh.. so it's a problem of the compiler on one of the build servers?
22:54:55 Quit merbanan ("Leaving")
23:00:03 Part nl1
23:05:13***Saving seen data "./dancer.seen"
23:07:34markunsome GSoC announcements:
23:10:28 Join chrisjs169_ [0] (
23:10:52 Join phpwner [0] (
23:10:56phpwnerhow2add songs
23:11:09amiconnTrying yet another register setup thing now...
23:11:34 Quit jhulst (Remote closed the connection)
23:12:35Beta-guywhile paying around with rockbox I tried to resume playback and it game me an error message
23:12:52Beta-guyData Abort at 400013EC
23:12:57Beta-guyhope that helps
23:13:13bluebrotherBeta-guy: do you have the lastest bootloader installed?
23:13:31Beta-guyyes I installed it an our ago
23:14:13DerPapstbluebrother: afaik these errors have nothing to do with the bootloader
23:14:31 Quit SirFunk (No route to host)
23:14:33Beta-guyhow do I restart the ipod after I get that error
23:14:33bluebrotherhmm. Recent problems I guess?
23:14:42Beta-guyit does seem to want to reboot lol
23:14:48DerPapstbluebrother: check the logs from 3 hours ago to now
23:15:17Beta-guygot it :)
23:15:45Beta-guyI'll be a beta-tester :D
23:15:45DerPapstbluebrother: 21:27:33 to be precise ;)
23:16:13 Quit Domonoky_ (Read error: 110 (Connection timed out))
23:16:27 Quit desowin ("use linux")
23:16:35Beta-guyhow often do you guys update the firmware?
23:17:18bluebrothervaries. Sometimes daily, sometimes weekly
23:17:23 Join Wiwie [0] (
23:17:25pondlifeSeveral times a day, I'd say
23:17:45bluebrotherless than hourly when I'm working on something ;-)
23:17:49pondlifeDo you mean bootloader?
23:17:56bluebrotherbut due to the lack of time ...
23:18:14Beta-guyone thing I've been puzzled on how do I exit games? menu and back don't seem to work
23:18:28*petur keeps the usb cable attached to the player to avoid wearing out the connector
23:18:30bluebrothersome games exit on Menu + Select
23:18:48Beta-guyah thanks
23:19:00*bluebrother needs his player when not at home, especially at uni
23:19:41peturwell I use it too every day when biking to work and at work via wireless headphone
23:20:44amiconnI don't think these usb connectors will wear out too fast
23:21:13peturstill... can't be too carefull ;)
23:21:25amiconnFor some tests I plugged/unplugged my archos recorder ~100 times per day, and the usb socket still works fine
23:22:01amiconnAnd plugging/unplugging at the device side is more convenient most of the time
23:22:07 Quit phpwner ()
23:23:02*Beta-guy listens to buddy holly on his Ipod... :P
23:23:16stripwaxBeta-guy : - rockbox is under "continual" development. changes take place at all hours of the day, and each time a change is made the build on the website is updated automatically.
23:23:42stripwaxBeta-guy : - but there hasn't been an "official stable release" for quite some time. we still do new in-development builds every day though
23:24:06Beta-guyI think I'll just do a daily update :P
23:24:07 Quit chrisjs169 (Read error: 110 (Connection timed out))
23:24:10*stripwax wonders if phpwner got an answer
23:24:39stripwaxBeta-guy : - yeah, probably only worth doing if something important changed. I usually update my ipod every few weeks, or less
23:26:07Beta-guyis there a matrix thing for rockbox?
23:26:16*amiconn updates rather irregularly
23:26:19stripwaxwhat's a "matrix thing"?
23:26:30bluebrothera matrix like screensaver?
23:26:51Beta-guylol nuts! :P
23:26:51bluebrotherno. But rasher has some plugin at his website.
23:27:00bluebrotherno idea if it still applies though.
23:27:30stripwaxBeta-guy : - the starfield plugin is neat, and pulses in time to the music, if that's your bag
23:27:45Beta-guyI like plasma :)
23:28:11Beta-guylooks like a lava lamp :P
23:29:04*stripwax wonders if rockbox needs a viz plugin
23:29:46 Join entheh [0] (
23:30:40 Join chrisjs169|afk [0] (n=jack@unaffiliated/chrisjs169)
23:30:41 Join TrueJournals [0] (
23:31:08*amiconn wonders what that should be
23:34:09Beta-guyany way to make Sunbird compatible with the calender feature in Ipod?
23:37:29dionoeais that rockbox related ?
23:37:53*scorche votes no
23:38:02dionoea(few people here use the original firmware so you're unlikely to get an answer)
23:38:05Beta-guynot really, just curious lol
23:38:24scorcheBeta-guy: read the guidelines linked in the topic
23:38:28dionoeaI'm sure that someone has a module/plugin for sunbird to enable iPod compatibility
23:38:32dionoeaon google
23:39:44 Join Entasis_ [0] (
23:40:32 Join SirFunk [0] (
23:40:48Beta-guy"Please refrain from using odd formatting such as special characters when they are not needed. This makes the conversation harder to follow by many people." OMG I just thought of Mschat when I saw this lol
23:43:00*scorche sighs
23:43:20amiconn@ those who experience instability on G5/G5.5: anyone fancy testing a patch? (very tiny one, yet another register set up the way the OF does)
23:43:27 Quit midkay ("Leaving")
23:45:13oboamiconn: sure
23:46:18 Quit saratoga ("CGI:IRC (Ping timeout)")">
23:46:29Beta-guywhat kind of instability?
23:47:23amiconnI can't make the G5.5/80 crash here
23:49:23*dionoea never had any crash on his G5/30... only some keyboard lockups sometimes.
23:50:38 Join Mouser_X [0] (
23:53:29 Quit boten_anna ("CGI:IRC")
23:55:28stripwaxMy G5/60 rarely crashes and certainly not recently.
23:56:00stripwaxOh. except, immediately after I turn the hold switch off, the play/pause key doesn't work unless I scroll the scrollwheel first
23:56:19amiconnDo you use a build from June 3 or newer?
23:56:44 Quit Entasis (Read error: 110 (Connection timed out))
23:56:52stripwaxbut I'm pretty sure it's done that since a while back
23:57:57amiconnThere are some quirks regarding hold and buttons on ipod
23:58:02amiconnNothing new
23:58:21stripwaxok cool.
23:58:26 Join chuklz [0] (n=chuklz@
23:58:29 Quit Febs ("ChatZilla [Firefox]")
23:58:45stripwaxso is it worth checking your change to see if things are just as stable? :)

Previous day | Next day