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 2007-05-14

00:09:23markunpreglow: will you keep me informed on your progress? I'm really interested
00:11:06 Quit bluebrother ("Leaving")
00:14:47 Quit moos ("reboot")
00:18:14 Join moos [0] (
00:18:30 Quit moos (Client Quit)
00:18:58 Join moos [0] (
00:19:52preglowmarkun: will do
00:19:53preglowbut now, bed
00:20:04markungood night
00:20:54 Part krush1704
00:29:45 Join saratoga5 [0] (i=98039ac0@gateway/web/cgi-irc/
00:30:19 Join |cub| [0] (
00:36:11 Join scorche [0] (i=scorche@rockbox/administrator/scorche)
00:40:08 Quit idnar (Nick collision from services.)
00:40:10 Join idnar_ [0] (n=mithrand@unaffiliated/idnar)
00:42:39 Quit gromit` ("Coyote finally caught me")
00:43:05chrisjs169|brblinuxstb: I know JdGordon did it in the past, but since it wasn't commited (as far as I can tell) I'm trying to find what he used to do that
00:49:16 Nick Gursikh is now known as Khalsa (n=Gursikh@unaffiliated/gursikh)
00:49:26 Nick Khalsa is now known as Gursikh (n=Gursikh@unaffiliated/gursikh)
00:50:14***Saving seen data "./dancer.seen"
00:51:39 Quit Guile` ("KVIrc 3.2.0 'Realia'")
00:52:58 Nick Viljar is now known as Viljar`zZ (
00:53:33saratoga5anyone here have any ffmpeg experience?
00:53:45schulachrisjs169|brb: oh, are you makeing the rockbox releases on anythingbutipod?
00:53:47saratoga5i'm stuck on their API
00:54:09schula.oO(i just use vlc)
00:55:25 Quit Nico_P (Remote closed the connection)
00:56:27 Join TrueJournals [0] (
00:56:30 Part maffe
00:59:31 Join maffe [0] (n=maffe@
00:59:37 Quit Entasis ("Leaving")
01:00:35 Quit sergey (Read error: 110 (Connection timed out))
01:01:12 Join sergey [0] (
01:02:03 Quit |cub| (Read error: 54 (Connection reset by peer))
01:03:23 Part TrueJournals
01:04:10 Join |cub| [0] (
01:07:13 Quit robin0800 (" HydraIRC -> <- Wibbly Wobbly IRC")
01:08:03 Quit Siltaar ("Aurevoir -")
01:12:14 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:13:02chrisjs169|brbschula: uh....yeah?
01:13:11chrisjs169|brbJdGordon: I'm sure you can help me :D
01:13:31schulacool ;) the build is really nice :)
01:13:35chrisjs169|brbRemember way back when you figured out how to disable the database refresh on the OF from the Rockbox bootloader?
01:13:38chrisjs169|brbschula: thanks :)
01:14:31JdGordonchrisjs169|brb: actually it was barrywardell, but anyway, whats the problem?
01:15:44chrisjs169|brbIt may have been barrywardell...but what was the code used to do that?
01:16:30JdGordonbootloader/main-pp.c iirc
01:18:00chrisjs169|brbdo you remember the line of code that was used to write the '0' to the hidden firmware partition?
01:19:17 Quit moos ("Glory to Rockbox")
01:19:27 Quit |cub| (Read error: 54 (Connection reset by peer))
01:19:32JdGordonfrom line 463
01:21:47 Join |cub| [0] (
01:23:51chrisjs169|brbwow...didn't even know it was in there. what do i use to enable it?
01:23:55 Quit saratoga5 ("CGI:IRC (EOF)")
01:23:58*chrisjs169|brb is lost when it comes to c
01:26:56schulachrisjs169|brb: is it possible to get a version without the micro sd patch? gather runtime information does not work, and in the daily build it works, and the sd_patch disables tag cache..
01:28:48JdGordonchrisjs169|brb: turn on the sansa with usb connected
01:29:20JdGordonschula: TC works, but not if you try loading a file from he mSD
01:29:46 Quit HellDragon ("rebooting(58486456) - LY")
01:31:35schulahm strange, so i will build it on my own, and investigate the problem
01:31:42pixelmaJdGordon: really? Do I have to test if database still works on Ondio then?
01:32:02pixelma(with your patch/build)
01:33:21 Join zivan56 [0] (
01:33:57JdGordonpixelma: you can if you want, I was playing and noticed it worked o the e200, as long as the files were on the hard diska nd not the sd card
01:34:06JdGordonif they were it wold try pl;aying "no file"
01:34:23 Quit sergey (Read error: 110 (Connection timed out))
01:34:47 Join Thundercloud [0] (n=thunderc@
01:34:48JdGordonunfortunatly, i dropped my sd card last night and after 30min of searching i couldnt find it.. and it hasnt turned up this morning :'(
01:35:16chrisjs169|brbMicroSD card?
01:35:36chrisjs169|brbare they seriously that small?
01:35:48JdGordon1cm x 1.5cm x 1mm
01:36:16 Quit Thundercloud (Remote closed the connection)
01:38:03 Quit lee-qid ("aufwiederbyebientotsayonara")
01:38:07 Join Thundercloud [0] (n=thunderc@
01:38:45zivan56well the sansa radio somewhat works in rockbox, although it's reminiscent of those non-powered earphone radios :)
01:39:07chrisjs169|brbit somewhat works?
01:39:29 Quit MbS_ ("ChatZilla [Firefox]")
01:39:37zivan56yes, there is a faint signal when i force the audio chip to maximum gain
01:39:47 Quit Domonoky_ ("Trillian (")
01:39:50*chrisjs169|brb really wants to try :P
01:39:55zivan56you have to set the frequency in the OF though
01:40:27zivan56and the LCD makes noises, so it has to be off in order to hear anything :P
01:40:38pixelmaJdGordon: still seems to work (initialized again - and played a file that I knew is on the MMC)
01:40:53*chrisjs169|brb still wants to try :P
01:41:36zivan56well if someone managed to get it to tune and get the audio up, there will probably be a patch
01:42:10zivan56the datasheet is really lacking in any help except getting signal strength and chip id
01:42:41zivan56and there are some tedious calculations that need to be performed in order to get it to tune
01:43:24zivan56overall, a very bad chip design IMO, because you have to implement most of its functions in software
01:44:51schulaits cheaper
01:45:12zivan56definately, it probably costs them $3 a chip or something
01:46:41zivan56i dont mind the bad design, but the datasheet is not very helpful and very ambiguous
01:46:46jhMikeSzivan56: well, while you're at it I figure I might as well try to get it to record something. Which line in is the radio on, 1 or 2?
01:47:05pixelmaJdGordon: what I know from before is that updating an existing database (which contains files from the external memory) while the card isn't plugged in, confuses the database (and progress stalls) - in case someone experiences this
01:47:10zivan56jhMikeS, radio in on line1, mic is on mic1
01:47:24zivan56mic1 works perfectly and is very clear surprisingly
01:47:27jhMikeSdoes it actually have a line-in via the connector?
01:47:58 Join HellDragon [0] (n=JD@unaffiliated/helldragon)
01:48:38zivan56well the AS3514 chip has 2 line ins, one of which is connected to the radio
01:48:38zivan56there are 2 mics as well, one of them is the built in mic
01:48:47jhMikeSzivan56: this chip is an orgy of gain controls
01:49:07zivan56heh, i had to set gain to the very max in order to hear anything from the radio
01:49:14zivan56(for line in)
01:49:22jhMikeSI'm wondering if sansa sells something the has an actual line jack
01:49:33zivan56ah, not sure at all
01:49:40jhMikeSdid you use single or double ended input?
01:49:45JdGordonpixelma: ah ok
01:50:13zivan56not sure, not too familiar with the AS chip
01:50:24zivan56i uploaded the code and gave a link on rockbox-dev
01:50:33zivan56im off for a bit, i'll bbl
01:50:36jhMikeSbits 7,6 to 00?
01:51:35pixelmaJdGordon: just an idea/question - did you load database to ram or just tried without?
01:52:15 Quit HellDragon (Nick collision from services.)
01:52:41JdGordonyeah, the patch disabls ramcache
01:53:05JdGordonuni time, ttyl
01:53:15pixelmaok, just thought that there could be a difference since the Archos' can't store database in ram
01:53:50 Nick idnar_ is now known as idnar (n=mithrand@unaffiliated/idnar)
01:58:13 Join saratoga6 [0] (i=98039ac0@gateway/web/cgi-irc/
02:00:29 Join [1]zivan56 [0] (
02:00:29 Quit zivan56 (Read error: 104 (Connection reset by peer))
02:00:33 Nick [1]zivan56 is now known as zivan56 (
02:05:27 Join courtc_ [0] (n=court@unaffiliated/courtc)
02:06:23zivan56JdGordon, I set LINE_IN1_L[7,6] to both 00 and 10, but I got the same result...I think I am not writing to the FM chip properly
02:06:25 Quit courtc (Nick collision from services.)
02:06:29 Nick courtc_ is now known as courtc (n=court@unaffiliated/courtc)
02:07:27zivan56AUDIOSET1 was 0xF4 to enable line in 1 plus the regular stuff
02:07:54zivan56err sorry, meant jhMikeS
02:09:46jhMikeSsum is on...something that bothers me is it seems you can create a feedback loop
02:10:54jhMikeSwe need line out on? \
02:11:25zivan56i think that it is needed for headphone output
02:12:05jhMikeSno, only DAC is ever turned on and HP out
02:12:08zivan56or actually it may be used to bypass some headphone limit, I really didn't follow the development of the driver
02:12:49jhMikeSwould could comine the DAC gain and HP gain to make a big volume range
02:13:01jhMikeSs/would/we/ :P
02:14:10 Join [1]zivan56 [0] (
02:14:42[1]zivan56jhMikeS, perhaps, but is it needed? I guess for people with headphones that require more power it is
02:14:48 Quit zivan56 (Read error: 104 (Connection reset by peer))
02:14:48 Nick [1]zivan56 is now known as zivan56 (
02:15:21Soaplinuxstb: I haven't forgotten the test encodes, I've been busy the last 24 hours with getting stuff ready for my sister's wedding, should be done tomorrow.
02:19:21jhMikeSzivan56: you have the as3514 datasheet? if not, I don't feel I have the authority to decide who gets it. :\
02:20:22 Quit zivan56 (Read error: 104 (Connection reset by peer))
02:20:33 Join zivan56 [0] (
02:21:54zivan56jhMikeS, maybe ;)
02:22:19zivan56I dont think the problem lies with the AS3514 chip for my FM problem
02:22:26zivan56since the microphone works perfectly
02:23:11zivan56maybe i am not initializing the chip properly, as the only full datasheet I have is for the whole family of sanyo tuners
02:23:12jhMikeSthe chip should be fairly straightforward and there's no real difference in the inputs so, yeah, probably is the tuner
02:24:17zivan56it's hilarious that the DAC/PWM chip is much more simpler than the FM chip
02:24:32saratoga6wow Costco has the E280 for $139
02:24:39saratoga6looks like i'll be working on the sansa
02:24:47zivan56nice :)
02:24:52Soapdang, that is a fine price.
02:25:07saratoga6yeah i wonder why the baseline price on ebay is so high
02:25:14jhMikeSsaratoga6: where were you to tell me that when I got mine? :P
02:25:17Soap(that is the 8 gig model, isn't it?)
02:25:20saratoga6they've evidently been selling them nationwide at that price for months
02:25:25saratoga6yeah 8GB
02:25:55 Join Thundercloud_ [0] (n=thunderc@
02:26:04 Quit zivan56 (Read error: 104 (Connection reset by peer))
02:26:05jhMikeSindeed...that's really cheap...wonder if circuit city will let me return my e260
02:26:21SoapI wonder if the fact Costco has them at that price is a sign larger capacity ones are on the horizon. Sandisk, you would think, should be able to compete with apple on flash access...
02:26:28 Join zivan56 [0] (
02:27:22zivan56i bought my e250 for $220 when it first came out :/
02:27:39jhMikeSe260 for me: $119
02:27:45 Join TrueJournals [0] (
02:28:02jhMikeSthe 280 was $199.99 at the time
02:28:37zivan56still cheap compared to mine...well at least I have liquidmetal :)
02:28:46saratoga6i don't think theres a larger sansa coming soon
02:28:55saratoga6theres no more room on the PCM for memory
02:29:06saratoga6so they'll have to wait for larger memory modules
02:29:15Soapwhich don't exist?
02:29:22saratoga6i don't think so
02:29:32saratoga6i checked google a couple weeks ago to make sure
02:29:42saratoga6(didn't want to buy one if larger ones were coming out soon)
02:29:56Soapdo you know, is Apple in the same position? Or do they have room?
02:30:17saratoga6on the Nano 1G, they could actually stack chips, thus fitting 4 at once
02:30:22saratoga6i don't know about the 2G
02:30:43saratoga6i suppose sandisk could stack chips too, but they don't currently, so I doubt they will
02:31:04 Quit zivan56 (Client Quit)
02:31:28 Join zivan56 [0] (i=402e05dc@gateway/web/cgi-irc/
02:31:39saratoga6wow the ffmpeg IRC channel is useless
02:31:55saratoga6i've been asking questions there for days and so far have gotten only one reply
02:32:07saratoga6i don't understand how people use that project
02:32:15saratoga6theres no documentation, and no way to ask questions
02:32:26scorchesaratoga6: PM
02:34:10zivan56well, hopefully someone with a Sansa with FM will get this to work, it seems quite close arent identified...
02:34:48jhMikeSzivan56: where's the patch that does the basic FM enabling?
02:35:10saratoga6scorche: identified?
02:35:25zivan56 but its very hacky
02:35:33scorchesaratoga6: you must be registered with nickserv for me to see your PMs
02:35:47zivan56although i believe the read/write to be correct
02:35:52*jhMikeS wonders if someone's trying to replace the video decoder in mpegplayer ... and if so, why?
02:36:58saratoga6how do I do that?
02:37:21jhMikeSffmpeg _is_ another video decoder, right? :P
02:38:14saratoga6isn't mpegplayer already using ffmpeg?
02:38:15jhMikeSsaratoga6: rewrite a lot of code that uses the libmpeg2 right now
02:38:48 Join saratoga7 [0] (i=98039ac0@gateway/web/cgi-irc/
02:38:48 Quit saratoga6 ("CGI:IRC (EOF)")
02:38:54saratoga7sorry, i was refering to scorch
02:39:22scorchesaratoga7: well, first you must register with nickserv, and then you identify when you log in
02:40:12saratoga7any chance someone here has compiled the ffmpeg example code before?
02:40:40 Part pixelma
02:40:49jhMikeSzivan56: you're working based on a sanyo datasheet?
02:41:15zivan56err yes
02:41:25 Join Thundercloud__ [0] (n=thunderc@
02:42:45zivan56at least trying to, it doesn't explain key things such a tuning very clearly
02:44:17zivan56you kind of have to chase stuff though the document, and some things contract the model specific datasheet :)
02:44:20scorchesaratoga7: send me a PM
02:45:47zivan56yes, that one is model specific, AN2400S04.pdf tells you how to use it
02:46:33 Join Colby_Sollars [0] (
02:47:33jhMikeSzivan56: that's at sanyo's site?
02:48:21chrisjs169|brbtrying to make sansapatcher says `BLKSSZGET' undeclared (first use in this function) in sansaio-posix.c
02:48:33chrisjs169|brbon another computer, using the same exact files, it compiles fine
02:48:38zivan56it's on the rockbox site
02:48:48jhMikeSok, found it anyway right on rb
02:49:19zivan56get ready for a headache, that thing is horrible to read :)
02:50:16***Saving seen data "./dancer.seen"
02:51:18jhMikeS...and it says "Easy Radio IC" :P
02:51:49zivan56the only thing easy about is, is soldering it on and connecting it
02:52:57 Quit Thundercloud (Read error: 110 (Connection timed out))
02:53:27jhMikeSzivan56: you added one yourself?
02:54:07zivan56no, but it's probably easier than using :P
02:54:20zivan56you'll understand when you get to the tuning part
02:55:45 Quit jaebird ("Leaving")
02:55:48jhMikeSit has it's own volume control...but of course that's too obvious :)
02:56:33zivan56according to one datasheet, its "reserved" on the other, you can change it
02:56:47zivan56i have tried setting all volume to max
02:56:50 Quit Thundercloud_ (Read error: 110 (Connection timed out))
02:58:07 Join donutman25 [0] (n=chatzill@
03:01:58 Quit entheh ("^~")
03:03:34jhMikeSyeah, that little bugger has a lot of housekeeping to do
03:04:59zivan56yeah, its quite complicated to work with
03:09:35jhMikeShow is tuner presence determined?
03:10:52zivan56i guess you can check the tuner revision, it is 0x09
03:11:57zivan56tuner_write(BLK_SEL, BLK1); unsigned char version = tuner_read(CHIP_ID);
03:12:29jhMikeSmmmm...I was wondering if some GPIO bit was high or low depending or something
03:12:34zivan56the GPIO is not connect if it is not present, so the above should give 0?
03:13:11zivan56like another driver accicentally modifying it?
03:14:13jhMikeSno, just some overly simple indicator like that is just nice...was wondering
03:14:34 Part TrueJournals
03:18:20zivan56yep, that should work. Just send rockbox a message that device is not present, that way you dont need a seperate build
03:20:27jhMikeSradio_hardware_present is the current method of detection since the same build is used for x5 and x5v (no tuner)
03:21:50zivan56well at least we know how to do that so far :P
03:23:08jhMikeSit's clear I have to redo the way audio inputs are done in rb...probably by cap masks like samplerates are. there's no provision for no general line _and_ having FM radio
03:24:48zivan56is there any other target with a mic?
03:27:52jhMikeSanthing that records so far
03:28:36jhMikeSthey all had Line In as well, but this doesn't so I'll come up with a more flexible plan
03:29:12 Join digerati1338 [0] (
03:29:47 Quit Gursikh (Read error: 60 (Operation timed out))
03:29:59zivan56probably explains why I had to uncomment some "set source" calls for the radio option to work
03:36:51zivan56well this tuner is driving me crazy, so I guess I will leave it for another day
03:37:11zivan56maybe someone else can figure it out, it's probably something trivial
03:39:02jhMikeSa good break can help, fatigue doesn't :)
03:39:37zivan56yep, very true...
03:41:09zivan56someone should probably try to ask for a model specific datasheet from sanyo
03:43:01 Join TrueJournals [0] (
03:44:12chrisjs169|brbHow can I get a 'verbose' output of the bootloader? (sansa)
03:46:24 Quit TrueJournals (Read error: 104 (Connection reset by peer))
03:47:12chrisjs169|brbin the source it appears that bootloader messages should appear if any key is pressed, but it doesn't work
03:49:03 Join Thundercloud [0] (n=thunderc@
03:49:24 Join TrueJournals [0] (
03:54:24 Quit lids (Remote closed the connection)
03:55:17 Quit annulus_ (Read error: 110 (Connection timed out))
03:57:12 Quit Thundercloud__ (Read error: 60 (Operation timed out))
03:59:02 Join JdGor|uni [0] (i=82c20d68@gateway/web/cgi-irc/
04:01:19 Quit JdGor|uni (Client Quit)
04:06:17chrisjs169|brbnevermind...I fixed the problem
04:10:30 Join gtkspert [0] (
04:11:21 Join HellDragon [0] (n=JD@unaffiliated/helldragon)
04:11:50 Quit donutman25 ("ChatZilla [Firefox]")
04:15:13 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
04:15:59 Quit HellDragon (Client Quit)
04:19:59 Join HellDragon [0] (n=JD@unaffiliated/helldragon)
04:23:45 Quit amiconn (Read error: 60 (Operation timed out))
04:23:45 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
04:24:46 Join Llorean [0] (
04:25:17 Quit HellDragon ("reboot")
04:26:39CtcpVersion from freenode-connect!freenode@freenode/bot/connect
04:36:24 Join HellDragon_ [0] (
04:37:09 Nick HellDragon_ is now known as HellDragon (
04:43:13 Quit kkurbjun ("Leaving.")
04:44:10 Quit Thundercloud (Remote closed the connection)
04:47:32 Part TrueJournals
04:50:17***Saving seen data "./dancer.seen"
04:51:20digerati1338Does anyone know any good C resources on the web?
04:51:41digerati1338I know java very well but a reference for C would be helpful.
04:52:35 Join aliask|uni [0] (i=82c20d67@gateway/web/cgi-irc/
04:53:42 Join lids [0] (i=lds@gateway/tor/x-3072f2fb2c32873c)
05:00:22 Join gursikh [0] (n=khalsa@unaffiliated/gursikh)
05:00:23 Quit zivan56 ("CGI:IRC (EOF)")
05:04:43 Quit aliask|uni ("CGI:IRC")
05:16:17 Part digerati1338
05:23:07 Join dunder [0] (
05:23:51dunderi just bought a sansa e280, should i install rockbox on it?
05:24:02dunderi'm not too fond of the default interface
05:24:19Colby_Sollarsi just did it last week on an e280. worked like a charm
05:24:44dundersound quality is at least as good as stock sound?
05:24:48dunder..and battery life?
05:24:56dunderthose are 2 biggest worries
05:25:48Colby_Sollarsbeware tho: if the screeen does not say "writing" when you are moving stuff to and from it, it is not really doing anything, so make sure you get it to say "writing" before you unplug it
05:26:23Colby_Sollarssound quality is not quite as good, but the battery life seems to be about the same, maybe a hair better
05:26:25 Quit Llorean ("Leaving.")
05:26:55saratoga7I think sound quality is the same aside from the lower volume
05:27:13saratoga7at least Llorean's sansa had identical results in both firmwares
05:27:21dunderColby_Sollars: k, thanks for the do you feel the sound differs?
05:27:43dunderi've noticed the default sound is alright, but it's prolly a hardware limitation (DAC or something)
05:27:55dunder...could be a bit better
05:28:16dundersounds staticy sometimes...was hoping rockbox could solve that issue, but oh well...same sound is worth a shot
05:28:38dunderdefinitely if it's same battery life too
05:28:38Colby_Sollarsit is definitely quieter. And it could be a bit better
05:28:54dunderno eq...or other way to adjust volume?
05:29:09dunderstock sound is pretty damn quiet to me...i have to crank up the eq
05:29:15saratoga7theres a parametric EQ
05:29:25saratoga7EQ is not a good way to adjust volume though
05:29:29dunderdunno how e200r (rhapsody) players can live without the eq
05:29:52dunderit's better than barely being able to hear it
05:30:01saratoga7also battery life is generally poor on the PP based players, I expect the sansa to be the same
05:30:20 Join peanutb [0] (
05:30:40dunderi do have to tinkle, thanks for reminding me
05:30:47saratoga7portal player
05:31:58peanutbhow stable is the M5 port?
05:32:04dundersaratoga7: so, you're saying that battery life is going to drop from 20 to 15 hours or so??
05:33:23 Join kkurbjun [0] (
05:33:52dunderi assume that i'll lose all my music when i do this, right?
05:34:44Colby_Sollarsit keeps all your previously existing files perfectly intact
05:34:48saratoga7i have no idea what the change in battery life is
05:35:05 Join jhulst [0] (
05:35:06dunderhmm...Colby_Sollars, did you ever have a problem deleting songs from the player's default firmware??
05:35:21dunderi have this .Trash-dunder folder that i can't delete...keeps saying "read only"
05:35:30Colby_Sollarsare you on windows?
05:36:01dundernot now
05:36:08dunderi still can't delete only device
05:36:15dunderi've tried deleting it from windows also, tho
05:36:21Colby_Sollarswhat platform are you on?
05:36:24dundercan't figure out how the F to delete this trash folder
05:36:28dunderubuntu feisty right now
05:36:36dunderhave tried over xp also
05:36:56Colby_Sollarshave you tried 'sudo rm'?
05:37:16dunderpretty friggin sure...
05:37:46Colby_Sollarsas in "sudo rm 'Sansa e280/.Trash-dunder'" ?
05:38:37dunder[21525.642015] FAT: Filesystem panic (dev sdb1)
05:38:37dunder[21525.642036] invalid access to FAT (entry 0x00620061)
05:38:37dunder[21525.642042] File system has been set read-only
05:38:37DBUGEnqueued KICK dunder
05:38:37dunder[21525.642573] FAT: Filesystem panic (dev sdb1)
05:38:37dunder[21525.642580] invalid access to FAT (entry 0x00620061)
05:38:38***Alert Mode level 1
05:38:38dunder[21525.642937] FAT: Filesystem panic (dev sdb1)
05:38:40dunder[21525.642942] invalid access to FAT (entry 0x00620061)
05:38:50dundersorry for the paste, but that's dmesg output
05:39:35dunderhere's direct rm output −−> rm: cannot lstat `.Trash-dunder//VA-J.Armz-How_To_Be_An_Mc_Vol.40-(Bootleg)-2006-RAGEMP3/0': Input/output error
05:39:35Colby_Sollarshmmm... im running Dapper Drake, havent had any problems. What confuses me is that at some point your Feisty Fawn created the folder '.Trash-dunder' so at some point it must have been able to write
05:39:44dunderno kidding
05:39:59dunderi don't understand what the problem is...
05:40:30dunderColby_Sollars: so, if i wanted to keep the default firmware for right now and start over...i could reformat the device and the firmware would still be intact??
05:40:56saratoga7i don't know if you can format the sansa
05:41:07saratoga7however, there are uninstall directions in the manual
05:41:11saratoga7which I suggest you read
05:41:26dunderfor uninstalling what?
05:41:43dundersaratoga7 ^^
05:41:49saratoga7rockbox and it's bootloader
05:42:27Colby_Sollarsit seems like you're just having an issue with whater that how to be an Mc file
05:45:00dunderColby_Sollars: i typed sudo rm -r .Trash-dunder
05:45:16dunderit gives that same error for every file in every folder under .Trash...
05:45:25dunderrm: cannot remove `.Trash-grndslm/VA-Don_Cannon_&_Willie_The_Kid-Twenty_Eight_Grams_Pt._5-(Bootleg)-2006/00-va-don_cannon_and_willie_the_kid-twenty_eight_grams_part_5-bootleg-2006-ukp.sfv': Read-only file system
05:45:25dunderrm: cannot remove `.Trash-grndslm/VA-Don_Cannon_&_Willie_The_Kid-Twenty_Eight_Grams_Pt._5-(Bootleg)-2006/[VA-Don Cannon & Willie The Kid] Twenty Eight Grams Part 5.m3u': Read-only file system
05:45:25dunderrm: cannot lstat `.Trash-grndslm/VA-J.Armz-How_To_Be_An_Mc_Vol.40-(Bootleg)-2006-RAGEMP3/0': Input/output error
05:47:24Colby_Sollarsdunder: huh. You got me there. That might be a question for the Ubuntu forums. But until then, i dont think its really interfering with anything, is it? can you still add songs to your sansa?
05:48:39***Alert Mode OFF
05:48:42dunderColby_Sollars: i only have 6gigs wortha stuff on it (including the trash)...but i can see all the trash on a graphical disk map (gdmap)
05:50:27Colby_SollarsCan anybody tell me how to compile a single plugin and run it in my uiSimulator?
05:51:56peanutbbefore I install to my IAudio M5, should i charge it overnight?
05:52:03Colby_Sollarsdunder: I dont know what to tell you. You might be able to delete it once you get rockbox installed. It has all the functionality you need to delete, copy, or move files
05:52:51Colby_Sollarsdunder: in fact there have been a few times when, for some reason or another, i had to delete files that way
05:59:23tchanI would see how /dev/sdb1 is mounted by using the "mount" command. If that partition is mounted "ro" then that explains your problem.
05:59:58dunder/dev/sdb1 on /media/Sansa e280 type vfat (rw,nosuid,nodev,shortname=mixed,uid=1000,utf8,umask=077) [Sansa e280]
06:00:05dunderlooks like read-write to me
06:07:00 Join himitsu [0] (n=himitsu@
06:12:45 Join gromit` [0] (
06:37:35 Part toffe82
06:41:18 Quit saratoga7 ("CGI:IRC (EOF)")
06:50:18***Saving seen data "./dancer.seen"
07:11:22amiconnjhMikeS: Just fyi - radio_hardware_present() was not invented for the X5 / X5V, it was just a nice coincidence that we just had that method already
07:12:05amiconnX5 and X5V can be distinguished by the user, so we could have used different builds
07:12:24amiconnradio_hardware_present() exists because of the archos recorder V2
07:13:29amiconnThe first series were actually FM recorders, and that's something the user cannot see from the outside. On this first series of V2 recorders, the radio is usable in rockbox, but not in the OF :)
07:17:24jhMikeSamiconn: I see...and I guess one more needs it. BTW, I need to know what inputs available on each of those since I need more flexibility in deciding which input sources are available.
07:19:48amiconnMight be a bit difficult to come up with a scheme that works for all targets
07:20:15amiconnThe archoses which can record all have internal mic and external line in
07:20:18jhMikeSI've got one...same cap mask thing as is done for sample rates will make it easy
07:20:42 Join perplexity [0] (
07:20:43amiconn(i.e. archos recorder v1, v2 and fm, and archos Ondio fm)
07:20:57amiconnThe recorders also have s/pdif in
07:21:20jhMikeSand the fm is only for fm and ondio fm right?
07:21:27amiconnAnd the recorder fm and Ondio FM have a radio, which is simply hooked *in parallel* to the line input
07:22:05amiconnSO you have to set the recording circuit to line input if you want to record from the radio (and have no signal hooked up to the actual line in)
07:22:07jhMikeSoh, so you don't actually have a selection there in the Source list?
07:22:37amiconnYou can even use the line in as a line out and hear the radio in an external amplifier
07:22:45jhMikeSjust line in is used then and no special FM Radio?
07:23:19amiconnThe recording screen isn't used for radio recording on archos. You simply record from the radio screen - a whole lot easier for the user
07:23:26jhMikeSthe muxer could cover that by selecting line-in for playback but nothing imlements it that way
07:23:55amiconnSo there are 2 reasons why there is no separate fm source selection in the recording settings
07:24:17amiconnAnd the archos player is very special. Even though it can't record, it has a line input
07:24:26jhMikeSbut _only_ recorder v1, v2 have s/pdif in?
07:24:34amiconnNo, the fm as well
07:25:04amiconnrecorder fm/vm are practically identical hardware, just that the v2 (usually) doesn't have the radio mounted
07:25:44jhMikeSwhat's the line input used for? listening over headphones?
07:26:08amiconnThe only reason why we need 2 different builds is because of the firmware scrambling is different, so the archos loader on the fm wouldn't load a v2 binary and vice versa
07:27:20amiconnThe player line input is just mixed into the playback, so yes, it can be used for feeding some external signal you don't want to overhear while listening with headphones
07:27:37jhMikeSnot switchable?
07:27:44amiconnIt is switchable
07:28:21 Quit kkurbjun (Read error: 60 (Operation timed out))
07:28:24 Quit maffe (Remote closed the connection)
07:28:30amiconnRockbox has an option for this. The gain could also be changed iirc, but we don't do that
07:28:49jhMikeSbut if it's on, it's just mixed in so it's not really a separate source then
07:29:35 Join Flintheart|Glomg [0] (
07:30:39jhMikeSnothings really changing here but e200 has fm radio in but no line in so adding recording will not work with the current assumption that fmradio implies line in
07:30:45Flintheart|GlomgHey, I remember using an emulator for Rockbox earlier so that I could tinker with the WPS screens without syncing/rebooting my iPod over and over... but I can't seem to find it again. Does anyone know where to find it? :P
07:31:17amiconnAh no, gain can't be set.
07:31:42amiconnThe circuit is shown in the dac3550 datasheet, page 6
07:32:57 Join kaaloo [0] (
07:33:08 Part kaaloo
07:33:30 Nick davina__ is now known as davina (
07:36:48amiconnjhMikeS: Does the sansa have a mic?
07:37:42jhMikeSwouldn't be much problem there if it didn't anyway
07:40:58 Join maffe [0] (n=maffe@
07:41:26 Quit darkless (Read error: 104 (Connection reset by peer))
07:41:39 Join darkless [0] (
07:43:33jhMikeSI wonder if I should just #ifdef the recording screen to allow for no mic since it just assumes it now
07:44:43jhMikeSbah, forget it, it's an easy job if it comes up
07:45:30 Part maffe
07:49:58Mouser_XiPod 3G doesn't do video, does it?
07:50:17Mouser_XI'm asking, because someone said they had a 30 GB iPod video.
07:50:28Mouser_XAnd, Rockbox freezes on their player.
07:51:25Mouser_XFrom my recollection, Rockbox was reasonably stable on the iPod video.
07:52:02Mouser_XThey also said that SPCs freeze their player. I found that especially odd, since HCS made it a particular point to test SPC playback on his iPod.
07:52:28Mouser_XHmmm. Seems that it's a little late...
07:52:35Mouser_X(Time wise.)
07:55:37jhMikeSMouser_X: I never heard him say anything about freezing
07:57:01Mouser_XNo, not HCS.
07:57:18Mouser_XI'm refering to someone on the OC ReMix forums.
07:57:51 Quit desowin (Connection reset by peer)
07:57:59 Join webguest08 [0] (i=434e70d0@gateway/web/cgi-irc/
07:59:00jhMikeSright, I'm just assuming he would have brought something up if his were freezing
07:59:47 Quit webguest08 (Client Quit)
07:59:55 Join calam [0] (i=434e70d0@gateway/web/cgi-irc/
08:01:06Mouser_XYes, of course. This is why I'm mentioning it, because that shouldn't be happening.
08:01:41calamhello :) brand new sansa e280 owner rockn rockbox since day i got it 4 days ago, chose player based partly(okay, alot) on rockbox commit to support :)
08:02:50 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:03:12jhMikeSwell, when it comes to PP targets, all sort of things shouldn't happen but do. given the fact there's so much of 0xC0EDBABE and 0xCACAD0D0 in there I have to question the mentality of the engineers. :)
08:05:00CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:05:00*LinusN puts on his straight jacket
08:05:03 Join Guile` [0] (n=Guile@
08:05:46Mouser_XI'm wondering if it might simply be their bootloader.
08:06:05Mouser_XI'm attempting to find info about that, but I'm not sure how current the online manual is.
08:06:53Mouser_XSpecifically, I'm looking for info on updating their bootloader.
08:07:31Mouser_X(Updatind Rockbox would probably be a good idea as well, depending on how old their version is.)
08:09:39calamI have a few small questions if I may... a: if I wish to try a daily build, will it also include ALL dailies up to that date, on top of latest stable build? B: to install it, may I simply paste over my current install of rockbox?
08:10:32calam(btw, first time in channel, and fairly noob to IRC in general, please go easy on me)
08:11:18Mouser_XAnswer to B: Download the most recent version, and unzip it to the Sansa's root directory, overwriting all.
08:11:23Mouser_XThat's how to update.
08:12:13calamgreat, thx you Mouser_X. do you know answer to A:?
08:12:35dunderwhat's this nonsense on the wiki about not having "ff/rw with sound"?
08:13:02Mouser_XAnswer to A: I'm not sure I understand the question. If you're asking what I think you're asking, then most likely yes. Rockbox's code source is constantly updated, and built upon.
08:13:35Mouser_XAs such, the "release" you're talking about, and any previous edits, fixes, etc. are also part of the recent build.
08:13:39calamokay, so daily build should include stable and all dailies up to that point then?
08:14:19calamokay, that is half of that part of the question then, good, thx.
08:14:37 Part Colby_Sollars
08:14:57dunderwhat's this nonsense on the wiki about not having "ff/rw with sound"? rockbox can ff & rw, right?
08:15:27Mouser_XYes, it can. However, I don't think it plays the music in a sped up form, which is what you make it sound like they want.
08:15:30 Join JurassicC [0] (
08:15:44calamoh wait, my bad, that answers ALL of A, so again, i thank you.
08:17:07dundersoo...i'm just gonna ask one more time before i start the rockbox install −−> *Do you think that the e200 sounds better with either stock firmware or rockbox?*
08:17:25dunderdisregard volume, i'm asking strictly of sound quality
08:17:59dunder...or are they too close to call, in your opinion(s)
08:18:15JurassicCdunder:close call
08:18:40Mouser_XUsually, if a player has the hardware necessary to sound better, then Rockbox would utilize it.
08:18:58Mouser_XIn other words, usually it will sound better with Rockbox, but this is not always the case.
08:19:10Mouser_XI don't know for the Sansa.
08:19:21dunderyes, i regret my purchase...i'm hoping rockbox will at least give a more acceptable interface without a temporary cut in sound quality
08:19:30calami think its better on mine with original FW, but i have not played much with the rockbox options yet for sound. im about to try mpeg play, as we speak as a matter of fact :)
08:19:31dunderSansa has a pretty crap DAC, i hear
08:19:47JurassicCdunder: it says here: that the Audio DAC driver does not yet sound as good as the original
08:20:25*jhMikeS must mention that after a reset, the AS3514 is ending up with some semi-random values
08:20:53calammy experience bears that out (but its only been 4 days), but there are sooo many new and gr8 options, i do not regret my purchase at all, and based it largely on rockbox support :)
08:21:10dundercalam: with e200?
08:21:24calamyed dunder, e280
08:21:32jhMikeSactually, when the tinniness is gone, the sound is superior to OF at least checking with MP3
08:21:44calami just cant find pacman roms i need ;)
08:22:57dundercalam: yes, i was pretty disappointed with stock FW, i got used to it after a while, but some good qulaity songs i have get pretty scratchy for me...and on 3 occasions, i've had audio cut out to one of my headphone speakers, but a restart fixed that (FW problem)
08:22:58 Quit calam ("CGI:IRC (EOF)")
08:23:01 Join calam [0] (i=434e70d0@gateway/web/cgi-irc/
08:23:43dunderi prolly woulda returned it to if i had the chance, but i waited past my 14 day return period
08:23:48JurassicCjhMikeS: what do you mean "when the tinniness is gone"? how do you get rid of the tinniness?
08:24:19Joelyohh that DAC on the sansa actually looks pretty least in specs (it says it can do 48 khz sampling!) so that means data transfers for the ir will be even faster. :)
08:24:21jhMikeSJurassicC: I just find it's there sometimes and not others
08:24:33dunderi'm hoping that rockbox makes everything all better...and nobody seems to be hatin' the rockboxed e200, here goes!
08:24:49calamDunder, i like it quite well, it has many options, and i will be learning more about them. in coming weeks. the two things i do not like on sansa are that i cannot seem to find a way to simply browse by folder trees to filenames only, like on cheap players, and video on orig FW sucks the big one.
08:25:22dunderyea, i don't plan to use this for video
08:25:26dunderaudio only
08:26:59 Join cynicalliberal [0] (
08:27:20jhMikeSH10 has mic and line?
08:27:45calami had very good results, and it stays in rockbox mode most of the time. rock boy crashes (freeze) alot, but i can get the gain up more, and FF/RW speeds increased (i like audio books alot) but i keep losing my settings somehow, and my gain goes back down :( bookmarks dont seem to work for me either, but i prolly just do not understand how to use it well yet
08:28:53JurassicCdunder: one more word of caution: i bricked my first sansa by thinking that it wasnt doing anything when i was installing the firmware. so i unplugged, and just as i did, the screen suddenly changed from "connected" to "writing". Apparently the sansapatcher didnt get a chance to do all of its thing, leaving me with only half an install and no usb support anymore, and i couldnt boot into the original firmware after that. so
08:29:09calami allmost got an iriver e10, but wanted a flashplayer, and roxkbox has answered most of my needs so far
08:30:00JoelyJurassicC : did you try out the manufacturing mode?
08:30:06Joelyto unbrick it
08:30:28dunderthanks for the tip JurassicC!
08:30:33calami cannot get connection in rockbox mode at all, i think i unplugged before "writing" message went away, but patcher had finished and dissapeared, so i thought i was okay, doh!
08:30:46JurassicCyeah. I tried everything. It wouldnt respond to anything. i every tip for unbricking memorized. 2 hours later i threw it out the window
08:31:07Joelyheh, i have a jtag could have sent it to me...
08:31:19calami do wish the patcher had a way to offer a choice of load first original FW or rockbox though, would be nice for some i bet
08:32:00JurassicC** i *had* every tip for unbricking memorized :)
08:32:26 Join Llorean [0] (
08:32:45JoelyJurassicC : you still have it? of it
08:33:49JurassicCi know. Im not the sharpest tool in the shed
08:41:14JurassicCcalam: I dont think the "writing" message ever goes away, even when it is finished writing, so you probably have a proper install. The Sansa e200 port of rockbox does not yet have complete USB functionality
08:45:21calamjurassic, ty, i figured that was right due to everything else did okay, however, i tried to install a version of rockbox i found on some forum, is was older version i found out when i rebooted :(, and wouldnt work, however i deleted rockbox folder and started over, and waited this time, not too long, but at least 5 minutes, and it stopped saying "writing", but still no usb. i alsoo do not think it will charge in r
08:45:34calami do not wish to spam
08:46:36Joelywait...usb functionality? has the usb stack successfully been imported or something?
08:47:28 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
08:48:07calamjoely, it does not seem to be able to connect for transfer or charging on my sansa, and i read somewhere that charging is not ready for prime time, at least on sansa
08:49:14Joelywell it connects to just will not transfer anything in rockbox, as there's no usb stack (you have to init the usb phy from software..)
08:49:19Joelyyou have to boot to OF
08:50:19***Saving seen data "./dancer.seen"
08:51:21 Quit XavierGr (Nick collision from services.)
08:51:24 Join XavierGr [0] (
08:51:33 Join Rob222241 [0] (n=Miranda@
08:53:31 Join lee-qid [0] (
08:55:39 Quit jhulst (Read error: 110 (Connection timed out))
08:56:52calamyes, i have to reboot to OF to charge. well, i just played the movie i encoded with winff, and though i cannot FF/RW, and there is some artifacting (black-levels maybe) it is MUCH smaller file, and according to the player, 23.x FPS. however if i quit play, and re-open file, i get hard stall with abort error. reproduced this x2 times, had to reboot to restore function. overall though, i am now okay with my sansa, a
08:58:55 Quit cynicalliberal (Remote closed the connection)
08:58:55 Quit juxtap (Read error: 104 (Connection reset by peer))
09:02:12calam=wonders if bluetooth sansas will become rockbox enabled for doom multiplayer !! :P \o/
09:03:38Joelyno, but ir will
09:04:05calamir? iriver, infrared...?
09:04:26Joelynot serial infrared
09:04:52calamwhat players support that, or whatever?
09:05:05Joelyit's going to support any player
09:05:15Joely_going_ is the key word though
09:05:21Joelyi'm still developing the prototype
09:05:37calamand where may i finds me a brewed IR xmitter :D
09:05:46calamyou?! sweet
09:05:54Joelyright after the dev con
09:06:00 Join pondlife [0] (
09:06:42Joelyhehe want to see my beginning CAD skills? i'm having to design a professional looking case for it, it's quite fun ;)
09:06:59 Quit gursikh (Read error: 104 (Connection reset by peer))
09:07:16 Quit perplexity (Remote closed the connection)
09:07:18Joely <−−that's what the ir adapter will plug into
09:08:04calamhmmm, plz send to me ANY hardware you come up with, my kids have 4 & 60 GB pods to play with here as well, we can..."test" it for you, lool
09:08:41 Quit Rob2222 (Read error: 110 (Connection timed out))
09:08:42calaminto sony jack?
09:08:56Joelyit goes into a 3.5 mm stereo jack
09:08:59calamHS!! thets crazy, :D
09:09:10calamwoots to you
09:09:12Joelyyup, that's what makes it compatible with any rockbox target
09:10:08Joelybut...i have to pay uber big amounts of money to get the `glue chips' (the actel fusions) which are $5 a piece...only one is needed per unit, but heh, i need samples!
09:10:37Joelythose will do post processing on the audio signal, to split it and to act as the led driver
09:10:39calamIR blaster is cheap and small and readily avail to connect to your adapter, or will i have to make my own?
09:11:00scorcheJoely: you mention devcon...are you going?
09:11:24calamwait, is that the device ON the 3.5 itself, or an adapter to a premade led?
09:11:25Joelyi know! i know! i was busy today working, i'll pm you
09:11:35 Join DarthShrine [0] (n=Angus@pdpc/supporter/student/DarthShrine)
09:11:37Joelyto scorche
09:11:50scorchethen edit the wiki
09:12:03 Part Llorean
09:12:08 Join norbusan [0] (
09:12:29Joelydon't you people remember what happened last time i edited the wiki!? :-P
09:12:53scorchei need to know who is coming and when....edit it ;)
09:13:05 Part norbusan
09:13:07 Quit XavierGr (Nick collision from services.)
09:13:11 Join XavierGr [0] (
09:13:12Joelyand back to calam, are you talking about a reciever?
09:13:42 Quit Flintheart|Glomg ("-=[ ReaperScript ]=-")
09:13:48Joelybecause it works for infrared remote controls, as well as irda...
09:13:54Joelythere's multiple leds
09:14:18Joelybut...the main purpose is for player to player communication
09:14:31Joelyanyways, time to edit wiki!
09:15:39calamyeah, lemme rephrase: is that pic the whole device itself, or an adapter ? and will it xmit and recieve from another one, i.e. allow doom multiplay for instance? and can i then bribe someone to add remote control functionality, so i can be pimpin like those who own iriver e10's, haha, only with a learning remote? heheh
09:16:35Joelythat pic is just a 3.5 mm jack... just showing off my cad skills, haha (i did that just today..)
09:16:41Joelyand second
09:17:02Joelyyou need two to receive and xmit
09:17:12calamokie, but is the functionality i mention possible?
09:17:20calamyes, understood
09:17:21Joelythrough the line in
09:17:39calamnif dude, truly
09:17:59calami wants me one STAT
09:18:31Joelyand then for bribes...i'm always open for people to generate a fund to get the fpgas...but heh. and yes, the original intention was to be a universal remote
09:18:39 Join norbusan [0] (
09:18:53Joelyone that you use the scrool wheel or other buttons in menus instean of thousands of buttons
09:19:01calamwait, my sansa has onboard mic, does the architecture supportr line in thrugh the phones as well as out then?
09:19:29Joelynope, for that...i'm going to have to make a revision 2.0 for usb
09:19:38Joely(which depends on the stack)
09:19:43Joelyhurry gsoc!
09:19:56Joelyso it would attach to the bottom of the sansa
09:20:02 Part norbusan
09:20:02Joelywhere you charge
09:20:18calam(head swims from acronyms i dont know, but it SOUNDS cool ;P )
09:20:26calamill take 3!
09:20:33Joelygoogle summer of code...
09:21:13calamy not just bluetoof an an adapter then to the port on bottom?
09:21:26 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
09:21:31Joelybecause....i'm going for _compatibility_
09:21:34calami can splice the wires
09:21:36Joelycross platform
09:22:26Joelyplus, bluetooth is actually slower surprisingly than this IR
09:22:46 Quit XavierGr (Nick collision from services.)
09:22:49 Join XavierGr [0] (
09:23:10calambut that would preclude making an adapter to my sansa, ifn you r gonna make it for all, unless its inline with usb maybe?
09:23:33calamthe meaning of all those acronyms added up perhaps?
09:24:13Joelycan you rephrase that...i didn't understand that
09:24:58pondlifeSlasheri: around?
09:25:07Slasheripondlife: hi
09:25:19JurassicCAPI question: i dont suppose somebody could tell me how to turn a file like "bitmap.bmp" int a fb_data * bm ?
09:25:28pondlifeQuick question - why does tagtree.c use the playback buffering/unbuffering events?
09:25:34JdGordonJurassicC: with bmp_read_file()
09:25:54Slasheripondlife: for gathering statistics (playcount etc.)
09:25:59calamk. does that mean then to get crossplatform, you will be making it an adapter that attaches to the usb of my adapter, or anyones, like ipod, whatever, the that to device, so i wont have to hack an attachment ?
09:26:03pondlifeSurely the stats are based on whether a track is played or skipped, not whether it's buffered or unbuffered?
09:26:20 Join aliask [0] (
09:26:34pondlifeI'm trying to work out how to make the buffering side of playback.c more independent.
09:26:35Slasheripondlife: yes, but disk is spinning when track is unbuffered
09:26:56calam&/or also thru the 3.5?
09:27:00Joelycalam: well, the `official' adapter is the 3.5 mm jack interface
09:27:22pondlifeOK, so it would be ok to use a "disk active" callback, along with "track played" and "track skipped"?
09:27:23Joelythen, for players that don't have the line in capability....they use usb otg
09:27:24Slasheripondlife: otherwise it would be necessary to cache those statistics to avoid spinning up disk everytime track changes
09:27:34pondlifeAh, I see.
09:27:43JurassicCjdGordon: Thank you. Can you tell me which file that's defined in?
09:27:46calamhow do i direct cmmts to you without spamming channel?
09:27:50JurassicCi can't seem to find it
09:28:04JdGordonJurassicC: apps/bmp.h I _think_
09:28:06Joely `/msg' me
09:28:18JdGordonapps/gui/bmp.h or firmware/export/bmp.h
09:28:43pondlifeI'd have expected the buffering callback to be an indicator of disk access, not unbuffering.
09:29:13pondlifeunbuffering just indicates that an buffered entry is about to be lost/overwritten.
09:29:17JdGordonpondlife: thats whay the playback code needs redoing!
09:29:34pondlifeJdGordon: I'm working on an API
09:29:44JdGordoncool :)
09:30:09pondlifeNo idea if it will be of any use though!!
09:30:14calamdid that go thru joely?
09:32:29 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:34:36 Join webguest87 [0] (i=92972988@gateway/web/cgi-irc/
09:36:13petur .
09:36:21 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:36:54JurassicCjdGordon: TY (it was actually read_bmp_file() just in case anybody was reading this and wondering the same thing)
09:37:04JdGordonah ok :)
09:43:22 Part JurassicC
09:43:50 Quit GodEater ("CGI:IRC (Ping timeout)")
09:46:08 Quit XavierGr (Nick collision from services.)
09:46:11 Join XavierGr [0] (
09:54:09 Join My_Sic [0] (
09:54:28 Quit lee-qid (Read error: 110 (Connection timed out))
09:58:07 Quit webguest87 ("CGI:IRC (EOF)")
09:58:09Joelyscorche, wiki updated!
09:58:20dunderwhoopsy daisy, no fm support on sansa...can't rockbox now, damn
09:58:44JdGordonnot yet... but looks like its close
09:59:04Joelyscorche, please make sure my commits to the wiki aren't tragic...
09:59:12Joelyi don't trust myself with wikis
09:59:28dunderhmm...JdGordon, do you know how close?
09:59:56JdGordonsomeone is working on it atm...
10:00:08JdGordonso could be hours, or days.. or he could fail :p
10:00:38 Join webguest99 [0] (i=3e58b3c0@gateway/web/cgi-irc/
10:00:46dunderhe could suck balls
10:00:58dunderwho knows...i hope he doesn't, tho
10:01:09dundercuz then i'd have to start gay bashing
10:01:42 Join combrains [0] (
10:02:16linuxstbFrom today's logs: 01.38.45 # <zivan56> well the sansa radio somewhat works in rockbox, although it's reminiscent of those non-powered earphone radios :)
10:02:55dunderthank God, I don't have to bash him
10:03:31 Quit webguest99 (Client Quit)
10:04:35 Join Llorean [0] (
10:04:35calamreally?! is that (sansa FM) in todays build?
10:04:46 Join miepchen^schlaf [0] (
10:04:49 Quit My_Sic (Read error: 104 (Connection reset by peer))
10:05:54 Join w1ll14m [0] (
10:10:06 Quit combrains ("Rockbox Rocks :)")
10:11:11scorcheJoely: PM
10:15:06 Join obo [0] (
10:16:59 Join My_Sic [0] (
10:20:18 Join sergey [0] (
10:22:24 Quit w1ll14m (Remote closed the connection)
10:24:40 Join Nico_P [0] (
10:28:36XavierGrhmm just got a strange response from my Ondio FM. As it was playing it stopped and said to remove the MMC, then it went into USB mode (of course without the usb calbe on). I powered off and then on, only to find out that once again entered USB mode, after a while it got fixed by itself... (it was running from normal rechargable AAAs)
10:28:57XavierGramiconn: do you think that it might have to do with the heat on my car?
10:29:17JdGordonNico_P: hey, why not commit the wps volume patch you put on the tracker?
10:29:40Nico_PJdGordon: because I wasn't sure everyone would agree on it
10:29:55Nico_Phas it been discussed ?
10:29:55JdGordongood enough reason :p
10:30:22JdGordonI havnt seen any talk... but its unlikely to get much discussio unless it breaks something
10:31:48Nico_PI'm especially interested in Llorean's opinion about i because he was the one who asked for the feature
10:32:05linuxstbNico_P: Do you know which targets don't go above 0db?
10:32:27Nico_Plinuxstb: at least the irivers H100 and H300
10:32:40Nico_Pall the details can be found in sound.c
10:32:53JdGordonthe only thing that worries me is where the same screen size could be used for targets where one can go over and one cant, so you couldnt use the same wps...
10:33:06linuxstbSo that means that the WPSs are no longer portable between irivers and ipods with the same LCD characteristics?
10:33:26linuxstb(what JdGordon just said....)
10:33:32Nico_Pah right. so maybe I shouldn't make a distinction between players
10:33:37JdGordonthinking about it, you shuold always have 0db being the second last one, regardless of the target
10:33:45Nico_Pprobably yes
10:34:19Nico_Pthat's what I had at first but the problem is it looks a bit weird on existing themes
10:34:58Nico_PI tested on iCatcher and increasing the volume from -1 to 0 dB makes the little circle display two more dots at once
10:35:49 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
10:36:01 Join miepchen^schlaf [0] (
10:40:34pondlifeJdGordon: If you're interested, my sooper-simple proposal is at Probably over-simplified at the moment :(
10:41:22JdGordonanything is more iteresting than working on assignemtnes...
10:41:27*JdGordon has a look-see
10:41:30pondlifelostlogic: If you're around, could you look at it too?
10:41:50calamnite all, very cool indeed, keep up the good work :)
10:42:02 Quit calam ("CGI:IRC (EOF)")
10:42:12pondlifeDon't be concerne with the implementation too much, as long as we have a suitable interface.
10:42:17Nico_Ppondlife: does this include MoB considerations ?
10:42:35pondlifeAlthough feel free to point out what is missing!
10:43:03Nico_Pif you plan on implementing this soon we'll have to work in close collaboration :)
10:43:07pondlifeI don't want to do anything much other than simplify playback.c
10:43:08Nico_PI'll have a look
10:43:22JdGordonimmediate thoughts, returning -1 for all errors is giong to be bad...
10:43:28pondlifeI don't think I'll have time to actually implement this, but I'm very happy to help
10:43:39pondlifeJdGordon: Why? Do we need error codes?
10:44:11pondlifeCurrently pretty much any error just gives "Codec failure" !
10:44:30pondlifeBut we have the entire negative range to play with
10:45:13pondlifeUpdated to say < 0 rather than -1
10:45:17JdGordonpondlife: buffer full shhuold give the same error as an invalid filename?
10:45:39Nico_PLlorean ?
10:45:46pondlifeHmm, bufopen would give an error on invalid filename, bufread on buffer full...
10:46:01oboCould anyone look at FS #5495 and FS #6213 for me please?
10:46:04 Join w1ll14m [0] (
10:47:55JdGordonpondlife: ok, also errors like cant open any more files, umm, I dunno
10:48:12pondlifeYep, no problem.
10:48:22pondlifeI'm keeping it quite vague I know!
10:48:48pondlifeJust wanted to fan some playback flames and see if was useful.
10:49:18JdGordonwell, it looks like its about as simple as it could be made which is good :)
10:49:44JdGordon"Access to the global variables audiobufend and audiobuf should become read-only via this API" <- no access at all...
10:50:05peturobo: I've been looking at it but ran completely out of time, will try to have another look
10:50:18pondlifeWell read-only at least (i.e. functions to bufgetstartptr() or something)
10:50:21***Saving seen data "./dancer.seen"
10:50:41obothanks petur
10:50:43 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
10:50:45LloreanNico_P: Present
10:50:48pondlifeI'd like them to be totally hidden, but plugins probably need them.
10:50:56Nico_PLlorean: have you seen my volume patch ?
10:51:06LloreanThe one for the WPS?
10:51:46JdGordonpondlife: any idea if the implementation could stop rebuffering the same track/codec more than once?
10:51:57pondlifeIt certainly could.
10:52:18pondlifeJdGordon: But like I said, initially we should maintain the current algorithm.
10:52:51pondlifeOtherwise it'll probably never get off the ground :/
10:52:56JdGordonimo the only thing to keep the same is the codec api... scrap the rest so we dont bring bugs across
10:53:00JdGordonstart from a clean slate
10:53:12JdGordonin a seperate svn branch...
10:53:15LloreanNico_P: Yes, I like it quite a lot, but then, I'm the one who made the feature request so I may be biased. ;)
10:53:27pondlifeJdGordon: We can migrate to this API
10:53:34LloreanI haven't tested it yet, though.
10:53:54*Llorean hasn't updated in over a week. =/
10:53:59JdGordonpondlife: oh, you want t d what I did with the menus? have both working together and then wrap and migrate?
10:54:04jhMikeSpondlife: mpegplayer will already implement this in some way but doing multiple streams is no treat and the amount of data requires sparse buffer since there no way it should have to say, buffer from the beginning to a seek point
10:54:22*JdGordon gone
10:54:25Nico_PLlorean: You're the most suited person to tell me if I did it right :) But it turns out having two different syntaxes is a bad idea
10:54:48LloreanI was going to suggest that it might be a bad idea.
10:55:02pondlifejhMikeS: We can have sparse buffers. Can you maybe add any required API? Or would multiple calls to bufseek be enough?
10:55:29LloreanWhy not just leave the >0 case in for all WPSes, and on the targets where you can't get greater than 0, the author can leave it blank (or include a spare image, like the stop images so many people used to include)?
10:55:46LloreanThat makes WPSes more cross-target anyway.
10:55:49pondlifeCertainly nothing there should preclude an implementation which actually just pre-buffersma the most suitable parts of files...
10:55:56jhMikeSwe need a seek pos and the amount of lookahead needed since it will have to scan for start codes at first
10:56:32Nico_Pobo: you gave the same link twice
10:56:44pondlifeI did consider having both size and minsize parameters to bufread()... would that help?
10:57:05pondlifei.e. amount of data, ideal and minumum...
10:57:15jhMikeSprobably something like that
10:57:47pondlifeThen I thought it was maybe an unneeded complication!
10:59:08 Join Entasis [0] (
10:59:10jhMikeSI'm also doing it with no wrapping indexes, only the pointer wraps. Otherwise it's file window that streams cannot exceed.
10:59:30oboNico_P: huh? Where abouts?
10:59:51Nico_P[10:46] <obo> Could anyone look at and for me please?
11:00:07pondlifejhMikeS: I wasn't going to wrap indexes either
11:00:22pondlifeBut at pointer wrap there would be a short bufread
11:01:35jhMikeSit seems any buffer capable of handling an mpeg video needs some stream awareness, I can't figure a way out of that
11:02:15pondlifeProbably, but you could implement multiple streams with multiple bufopen()s?
11:02:16 Join spiorf [0] (
11:02:58Nico_Pobo: sorry it's my client's fault
11:03:03oboNico_P: that's not what the log says :)
11:04:05Nico_PLlorean: but I don't really see how the WPS creators could handle it nicely
11:04:36Nico_Ptake iCatcher for example, we'd have a problem with the H300 and iPod color
11:04:37jhMikeSpondlife: but we don't want the file opened twice
11:04:49pondlifeWe don't have to
11:05:10pondlifeThe buffering implementation could handle that
11:05:26pondlifeI'm just concerned that the API is suitable at the moment.
11:06:46 Quit bospaadje ("great minds run in great circles")
11:06:52LloreanNico_P: If the WPS just had <mute|..|..|0dB|max>, then on the iPod Color you'd see the max image, on the H300, it'd be there but you'd never see it, and the WPS would work cross platform, right?
11:06:53jhMikeSwill it have it's own thread?
11:07:03pondlifeSo more work for you there ;)
11:07:31pondlifeThe idea is just to give it a list of work to do and hope that it will have all the data ready in time,
11:07:37pondlifeso as not to block too much!
11:08:13pondlifeIn reality, the buffering thread will be reading the disk whilst the calling thread will block in bufread().
11:08:15Nico_PLlorean: that means the circle in iCatcher would never be full for the H300
11:08:58Nico_Pof course that's not a real issue but it means some WPSs will need adapting
11:08:59jhMikeSpondlife: sort of like now? :)
11:09:06pondlifeBut it should be easier to contain the buffer access than the current setup, which has all of my "favourite" bugs
11:09:14LloreanNico_P: WPSes would need adapting to add in the Max image anyway.
11:09:25jhMikeStwo streams can fight too
11:09:59pondlifeTrue, but the implementation should be able to cope.
11:10:21pondlifeI'm not too concerned with video initially, as long as the API is ok
11:10:26LloreanBut I think the special case for the 0dB maximum targets can be removed, WPS authors can handle having a useless case in their conditional I believe and it keeps things more flexible I think.
11:10:37pondlifeIt's the main playback which is buggiest.
11:11:03Nico_PLlorean: i agree
11:11:09jhMikeSFrankly I'd like this COP stuff resolved to allow both cores direct access to file APIs as well
11:12:27jhMikeSAnd if all that can be done, only need for a buffering thread should be background buffering...but mostly my head is video oriented atm
11:13:46pondlifeNo problem, I don't see much changing in the near future. Like I said, I don't have much spare time myself, but wanted to dump my ideas onto the wiki.
11:14:12pondlifeIt's all about getting playback.c to a place where it can be worked on without breaking some feature.
11:14:56Nico_Ppondlife: (about the "audio buffer" definition) maybe the audio buffer could simply be renamed to "main buffer" ?
11:15:27jhMikeSget rid of higher level code from it like playlists, directories, etc.
11:15:39pondlifeNico_P: Yes
11:16:09pondlifejhMikeS: That should be done too.
11:16:14Nico_PI know it's completely unimportant but it just felt much more appropriate
11:16:17*Nico_P reads on
11:16:56jhMikeSI was thinking that it should be given an abstract interface to call for directions on what to play. it cares not what it points to.
11:18:36pondlifeYes, a prearranged playlist?
11:18:39jhMikeSthere's always that push vs. pull decision. I suppose filenames could be pushed to it but that would be sort of its own private playlist
11:18:54pondlifeCertainly shouldn't care about directories
11:20:07jhMikeSright now it's hard-coded to calling playlist_* functions which bugs me a bit
11:20:36pondlifeI don't mind that so much.
11:22:51JdGordonpondlife: perhaps add a size and offset param to bufopen() to tell it how much and from where to start buffering?
11:23:03jhMikeScould always have an audio_play_file which queues it
11:23:22pondlifeJdGordon: bufopen(), the bufseek() to set offset...
11:23:28Nico_Ppondlife: I'm not sure why this has nothing to do with MoB ?
11:23:29 Join br3nda|mobile [0] (
11:24:00pondlifeNico_P: It should make MoB easier, but doesn't actually implement any of it.
11:24:02jhMikeSI'd like to see playlists run the playback and never the other way
11:24:10JdGordonpondlife: so bufopen() only opens the fd and doesnt read anything from the file?
11:24:16Zagorwe just got porn spam in the tracker
11:24:36jhMikeSbout time :P
11:24:43Zagorlooks like we can't trust unverified users to attach files :-(
11:24:46*petur hurries over ;)
11:24:50pondlifeJdGordon: Ah, I see what you mean... it would be silly for it to start buffering from the start each time.... I'll add a start offset
11:24:50Nico_Ppondlife: true but it seems closely related to me. If there can be all sorts of files in the "main buffer" with handles and so on, MoB will be quite straightforward
11:25:00 Nick br3nda|mobile is now known as Br3nda (
11:25:31Zagorif anyone has a better idea than just "disable file attachments", I'm all ears
11:25:56JdGordoncan you limit file types?
11:26:27JdGordonpondlife: also a minimum size to buffer...
11:26:33ZagorJdGordon: this was a html file. should we limit those?
11:26:46pondlifeJdGordon: A minimum?
11:26:53pondlifeDo you not mean a maximum?
11:26:54JdGordonI cant think why a html would should be uploaded onto a task
11:26:59LinusNi don't see a need to allow html files
11:27:38JdGordonpondlife: yeah, minimum... if the buffer doesnt have enough room for the header (for example) then it shuold fail...
11:27:38LinusNin fact, the only allowed types should be .diff, .patch and .zip imho
11:27:50JdGordonand .png .jpg ./bmp
11:27:54ZagorLinusN: filenames are easy to fake
11:28:04Zagorwe must use 'file' on them
11:28:16pondlifeJdGordon: bufopen wouldn't read the file, just queue it for reading.
11:28:43JdGordonoh, so when would it start reading?
11:28:49jhMikeSI've posted other stuff in the forums by renaming the extension
11:29:00JdGordonon next spinup? or at the next bufseek()?
11:29:04Zagora very simple way is to make all attachments content-type application/binary. then you must save it and open it manually to view it.
11:29:12Zagora bit of a bother for patches though
11:29:33aliaskYes, that'd be quite irritating...
11:29:41pondlifeJdGordon: The buffering thread will likely drive the spinups
11:30:18JdGordonok, but I tihnk I would expect that fairly shortly after bufopen() my file would be ready to use..
11:30:21jhMikeSwell, I don't see how text files can irritate too much. do spammers really want to post text?
11:30:48ZagorjhMikeS: html is text
11:30:49jhMikeScan html links be made to pop up as source?
11:30:52pondlifeJGordon: Yes, so would I. Don't want to block in bufread() at all if possible.
11:31:21Zagorflyspray doesn't have any content sensitifity at the moment, so I'd have to patch that it
11:31:31JdGordonI wouldnt tihnk it would, bufread() would be the blocking task
11:31:57Nico_PZagor: speaking of flyspray, are you planning on an upgrade to the new version ?
11:32:00jhMikeSI know that...I'm thinking so spammer content can't be readily viewed as the spammer intended
11:32:34pondlifeJdGordon: You'd bufopen() first, and then bufread() would block until the data was ready.
11:32:44ZagorjhMikeS: I can't say I'm planning it, but I'll do it eventually.
11:33:06JdGordonyeah, umm.. hmm
11:33:48ZagorjhMikeS: that could be solved by forcing attachments to be type text/plain, but that breaks images.
11:34:13 Join juxtap [0] (
11:34:27LloreanWhy not just limit it to extensions.
11:34:42LloreanAnyone with the appropriate privileges should be able to delete files that are inappropriate but renamed.
11:35:07LloreanIt's not perfect, but if the extensions are limited it does say "These are our expectations of you" and the majority at least should follow them.
11:35:41Zagorthe majority already follows it. the problem here is spammers who actively seek to circumvent the rules.
11:35:56Zagoralso we didn't find these links ourselves. I got an external report about it.
11:36:17Zagorso I'm not sure we can rely on finding and fixing uploaded spam
11:37:06*jhMikeS would like to be able to delte attachments on his own wiki page himself for starter :\
11:37:31Zagoryou can't do that? an access denied
11:38:13jhMikeSI can just hide them from view
11:38:25LloreanI think that as long as we leave it possible for new people to contribute, spam will be possible, won't it?
11:39:36LinusNwe have effectively stopped spam in the wiki with the registration process, but we don't want to make it that hard to report bugs
11:39:49LloreanWell, you don't need attachments to report bugs
11:39:59LinusNno, but patches
11:40:25LloreanWould it be hard to put into place a step like used for the wiki for posting patches?
11:41:14Zagorin flyspray only admin can change a users access rights
11:41:19*JdGordon gone again
11:43:36Nico_Panyone against me committing ?
11:44:45LinusNNico_P: looks ok to me. but why can't it go further above 0dB?
11:45:46Nico_PLinusN: you mean why can't we have several bitmaps for volumes above 0 dB ?
11:46:24scorcheLinusN: i forgot...when does bagder get back again?
11:46:25 Join n1s [0] (
11:46:41LinusNscorche: thursday i think, why?
11:46:41jhMikeSspeaking of that, why not use a two digit hex value instead of A-Z and a-z for ids?
11:47:13LloreanjhMikeS: Why does it even need to be hex?
11:47:16 Join ender [0] (
11:47:17LinusNscorche: anything i can help with?
11:47:40*ender curses the national telecom
11:47:42Nico_PLinusN: well obviously I need to know the position of the 0 dB case, so I decided that was the best way to keep at least some compatibility with current WPSs. Otherwise I could decide the 0 dB case is the last one but that would require adapting all the WPSs... unless I'm missing something ?
11:47:53jhMikeSscorche: cause hex looks cooler and we can have 256 elements, not 100 or 52
11:48:15enderthey screwed up something on wensday, and i've been without internet ever since
11:48:19jhMikeSthe dependency on codec order in the core should be gone too
11:48:30Nico_PjhMikeS: 100 is already better than 52
11:48:31LloreanjhMikeS: That was me who asked. But people often run out of memory before they reach 52 images anyway, right?
11:49:04LinusNNico_P: ah, so it's about calculating the enum range?
11:49:16jhMikeSLlorean: as little as I've done with that, some WPSs came close enough to the limit that it got awkward
11:49:21Nico_PLinusN: yes
11:49:27LloreanjhMikeS: I think giving them 100 gives them more than enough rope with which to hang themselves. ;)
11:50:02dunderwhat's the difference between Sansa e200 - Boot, Sansa e200 - Normal, & Sansa e200 - Sim ?
11:50:04LinusNNico_P: tricky one
11:50:15LloreanLinusN: When I made the feature request, in my mind the value over 0dB was more of a "Warning: Clipping" image than anything else.
11:50:34jhMikeSThere should be labelled enums that use the IDs defined in the codec array instead of the current type being index dependent
11:50:44Lloreandunder: Bootloader, Rockbox itself, and the UI Simulator for various testing purposes.
11:50:59dunderso, i just download the first two
11:51:03dunderhow can i get svn?
11:51:17*jhMikeS noticed this stuff when having to fix a bunch of WPSs after changing the codec order :\
11:51:17dunderi tried clicking on the 0s, but i get a buncha crap...
11:51:39Lloreandunder: Read the text at the top of the page.
11:51:49Nico_PjhMikeS: the codec order shouldn't change
11:51:52Lloreandunder: And the manual tells you what you need to download.
11:52:13jhMikeSI think it should be irrelevant really and if it weren't for the WPSs it would be
11:52:39n1sLinusN: has FS #6934 just been forgotten or is it not needed anymore?
11:53:34LinusNi guess it's been forgotten
11:55:31Nico_PjhMikeS: labelled conditionals wouldn't be very easy to add to the WPS syntax... but I agree it's a bit clumsy
11:55:40 Join bospaadje [0] (
11:56:58 Quit w1ll14m (Remote closed the connection)
11:57:43n1sLinusN: ok, I can confirm that the bug is still there in svn, should I commit?
11:57:53jhMikeSNico_P: what's the biggest hurdle?
11:58:17pondlifen1s: I'd like that committed, FWIW.
11:58:34n1spondlife: ok, I'll go ahead then :-)
11:58:54pondlifeHah, you can blame me! :)
11:58:55LinusNn1s: shoot, and don't forget docs/CREDITS
11:59:24*pondlife heard a rumour that Bagder wouldn't be back until the build table was completely green
12:00:21pondlifeWe may have that in only 12 perfect commits!
12:00:39LloreanJust be eeeextra careful
12:01:03pondlifeAs always... ;)
12:01:11Nico_PjhMikeS: enums were designed to be numerical. adding the possibility of labelling might be possible but I think it would be a little complicated
12:02:36Nico_Plike storing a table of the different labels or having to reread the whole conditional for labels each time it's evaluated
12:02:55XavierGrLinus: btw why the new bootloader isn't released yet?
12:03:04pondlifeSlasheri: I'm still having dircache blocking lock up my database initialisation on the sim :/
12:03:28pondlifeSurely dircache_is_initializing() should be "return dircache_initializing && thread_enabled"
12:03:40 Part LinusN
12:04:06pondlifeThat certainly fixes it for me.
12:05:13 Quit sergey (Read error: 110 (Connection timed out))
12:05:25jhMikeSI was thinking like %?fc<%xdr:AIFF|%xds:WAV|%xdt:Ogg....> or something. Anything not labelled gets its numerical order for some backward compat.
12:05:45 Join sergey [0] (
12:26:26 Quit spiorf (Remote closed the connection)
12:37:39 Join maffe [0] (n=maffe@
12:41:35 Join Psilonaut [0] (
12:50:22***Saving seen data "./dancer.seen"
12:50:58 Quit sergey (Read error: 110 (Connection timed out))
12:53:30 Quit Llorean ("Leaving.")
12:55:23 Quit DarthShrine ("Leaving")
12:55:58 Join norbusan [0] (
12:56:15 Part norbusan
12:57:17 Join Arathis [0] (
12:58:54 Part maffe
12:59:13 Join RockingD [0] (
13:00:09 Quit Seed (Nick collision from services.)
13:00:14 Join Seedy [0] (
13:06:44 Quit Br3nda (Remote closed the connection)
13:13:04 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
13:15:42 Join hostf4cekilla [0] (
13:21:16 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
13:25:54 Quit hostf4cekilla ("ChatZilla [Firefox]")
13:30:40 Join Compaq [0] (n=chatzill@
13:30:50 Nick Compaq is now known as jSwetz (n=chatzill@
13:35:27 Quit jSwetz (Client Quit)
13:36:45 Quit pondlife ("disconnected has pondlife")
13:38:23 Join pondlife [0] (
13:38:33 Quit pondlife (Client Quit)
13:41:58 Join pondlife [0] (
13:48:07 Join Llorean [0] (
13:48:50 Join Thundercloud [0] (n=thunderc@
13:49:29 Quit pondlife ("disconnected has pondlife")
13:49:47 Join bluebrother [0] (i=PzRnUgbX@rockbox/staff/bluebrother)
13:54:24 Join riddlebox [0] (
13:54:46riddleboxhello, is the rio sport supported by rockbox?
13:55:24LloreanThe only players supported are those listed on the front page.
13:55:35riddleboxdang, I really wanted to play ogg on it
13:58:12 Quit Rob222241 ()
14:08:19*GodEater remembers listening to ogg on his Rio Karma - on the stock firmware no less.
14:10:42GodEatershame the hard drive sucked so much
14:22:53 Join GEWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
14:23:05 Quit GodEater (Nick collision from services.)
14:23:13 Nick GEWeb is now known as GodEater (i=c2cbc95c@gateway/web/cgi-irc/
14:28:36 Quit riddlebox ("Leaving")
14:41:23 Quit livingtm (Remote closed the connection)
14:42:21 Join webguest90 [0] (i=c023110a@gateway/web/cgi-irc/
14:42:42 Join lee-qid [0] (
14:44:17aliaskIs there an option in Flyspray for users to edit their own tasks?
14:46:38 Quit GodEater ("CGI:IRC (Ping timeout)")
14:48:01 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
14:49:21LinusNonly for some
14:50:06aliaskWhat does that mean? Regular users can't, but admins can? :P
14:50:25***Saving seen data "./dancer.seen"
14:50:58LinusNdevelopers can
14:51:54LinusNaliask: you want those powers too?
14:52:05aliaskI'd love them.
14:52:12 Quit Arathis ("Bye, bye")
14:52:24LinusNaliask: *bing*
14:52:26aliaskWould I be able to close tasks, or is that reserved for admins?
14:53:04LinusNyou can close tasks too
14:53:35aliaskCool, I keep seeing ones I think should be closed every once in a while...
14:53:44aliaskThanks for that LinusN :)
14:53:46LinusNdon't abuse your powers though
14:54:02aliaskOf course not.
14:54:04preglownot unless it's for fun!
14:54:20LinusN"with great power comes great responsibility" :-)
14:55:02aliaskWoops, looks like default severity is critical...
14:55:27LinusNyeah, why else would you report a bug? :-P
14:55:35aliaskIt was a patch!
14:55:38 Quit pill (Nick collision from services.)
14:55:58aliaskOh well. At least I can fix that now...
14:56:06 Join _pill [0] (
14:56:32 Join Domonoky [0] (
15:04:16 Join Arathis [0] (
15:04:29 Quit RockingD (Read error: 110 (Connection timed out))
15:16:09 Join spiorf [0] (
15:18:13 Join Rob2222 [0] (
15:26:07 Join kaaloo [0] (
15:31:28 Part kaaloo
15:31:34 Quit aliask ("ChatZilla [Firefox]")
15:32:00 Join desowin [0] (
15:36:28JdGordonanyone around?
15:36:51JdGordonLinusN: amiconn?
15:37:24 Quit Psilonaut (Read error: 60 (Operation timed out))
15:37:55JdGordonI have a really crazy idea for the audio buffer and want to know how stupid it is... is our FAT drivr fast enough to be used so we "format" the audio buffer as a fat disk and use the fat read/write functions to buffer data?
15:38:34bluebrothersounds like much overhead to me
15:38:38*JdGordon shuold have prempted that by saying im half-asleep and half-drunk so probably very stupid idea
15:39:32JdGordonwell, Im tryin to think of a way to use the buffer in a non-ring buffer fasion, and it sort of clicked
15:39:46 Join kkurbjun [0] (
15:39:51JdGordonthe code is there already, so its just a matter of weather its fast enouigh to be feasible
15:40:07petursounds like bad weather to me
15:40:23preglowi don't see the point...
15:40:38bluebrotherfrom what I did with FAT I think the overhead is quite bad, and it's also kinda complex
15:40:54peturJdGordon: memory fragmentation
15:40:55bluebrotherif you want to use it as buffer
15:41:53JdGordonpetur: yeah, I figured the fat driver needs to worry about that anyway so the code for that is already there.... if the chunk sizes are big enough then fragmentatio isnt such a big deal
15:42:34bluebrotherhow do you plan to deal with files that are bigger than the buffer? You'd need to implement some sort of partial files for that ...
15:42:50 Quit webguest90 ("CGI:IRC (EOF)")
15:43:26LinusNand what about the simulator?
15:44:09JdGordonhmm... ok
15:44:21LinusNactually, i still don't see the point
15:45:10 Join webguest90 [0] (i=c023110a@gateway/web/cgi-irc/
15:45:48 Quit webguest90 (Client Quit)
15:46:05 Join maffe [0] (n=maffe@
15:49:12 Quit bawb2 ("seacrest out!")
15:49:46 Part maffe
15:50:12 Join maffe [0] (n=maffe@
15:50:53 Quit linuxstb (Remote closed the connection)
15:51:00 Join Thundercloud_ [0] (n=thunderc@
15:52:11 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
15:55:27 Quit spiorf (Remote closed the connection)
16:00:54 Join ctaf [0] (
16:01:06 Part ctaf
16:06:00 Quit Thundercloud (Read error: 110 (Connection timed out))
16:12:38 Quit GodEater ("CGI:IRC")
16:12:39 Join gursikh [0] (n=khalsa@unaffiliated/gursikh)
16:13:02 Quit Entasis (Read error: 104 (Connection reset by peer))
16:14:59 Quit |cub| (Remote closed the connection)
16:17:40*Nico_P is about to commit the volume enum patch from
16:19:58XavierGrjust great, I spent about 50 euros on car chargers and usb cables (etc) to mod my ondio for in-car use without batteries and now when the audio is connected to the sound system and the car lighet it is full of static :(
16:20:16XavierGr^car lighter
16:20:42peturground problem?
16:21:05XavierGrI am not so sure, I am not an expert in these stuff
16:21:32XavierGrthe moment the (-) is connected to the ondio (with the line out on the car stereo) I can hear hissing sounds
16:21:57XavierGreven if the ondio is off
16:22:15XavierGrof course when I turn on my ondio these hiss sounds go rampant
16:23:17JdGordonNico_P: is the big problem with album art loadin the AA file? or having an extra buffer to store the bmp in?
16:23:32Nico_PJdGordon: the buffer space
16:24:02Nico_Pbecause we want to store the bitmaps for all the buffered songs, otherwise we cause additional disk spinups
16:25:16JdGordonok, so pondlife's api propasal could work for that, just load the bmp before the audio file...
16:25:44JdGordonnot sure how it would work for metadata which doesnt come from a file easily though
16:25:57 Join |cub| [0] (
16:27:16Nico_Pyes. I'll focus on getting external files at first and then maybe try to add support for embedded art
16:27:48 Quit Zagor ("Client exiting")
16:27:48Nico_Pwhat I was thinking is that maybe it could be possible to use a variation of the jpeg viewer plugin to decode jpeg pics into a bmp buffer
16:28:33amiconnXavierGr: A ground loop for sure
16:28:39amiconnThis doesn't work well in the car...
16:31:32 Join Rincewind [0] (
16:35:20 Join toffe82 [0] (
16:39:58 Part LinusN
16:40:49 Join Rincewind_ [0] (
16:48:50 Join syof [0] (i=420046b6@gateway/web/cgi-irc/
16:50:07 Join pixelma [0] (i=d5494c9a@rockbox/staff/pixelma)
16:50:25 Quit JdGordon ("Konversation terminated!")
16:50:29***Saving seen data "./dancer.seen"
16:55:00 Quit Thundercloud_ (Read error: 54 (Connection reset by peer))
16:55:05 Join Thundercloud [0] (n=thunderc@
16:56:05 Join webguest49 [0] (i=3fad1312@gateway/web/cgi-irc/
16:56:42 Quit syof ("CGI:IRC")
16:57:06 Join jgarvey [0] (
16:58:11 Quit Thundercloud (Remote closed the connection)
16:58:23 Join Thundercloud [0] (n=thunderc@
17:04:48 Quit Rincewind (Read error: 110 (Connection timed out))
17:08:18 Join The-Compiler [0] (
17:11:07 Join Guest88856 [0] (
17:11:14 Join bluey- [0] (
17:11:16Guest88856THIS IS BLASPHEMY!
17:11:19Guest88856THIS IS MADNESS!
17:11:52 Quit Guest88856 ()
17:12:43 Quit webguest49 ("CGI:IRC (EOF)")
17:13:53 Join yossa [0] (
17:15:43 Quit pixelma ("CGI:IRC 0.5.7 (2005/06/19)")
17:17:52 Part maffe
17:18:08 Quit Rincewind_ ("Cya")
17:18:10 Join maffe [0] (n=maffe@
17:18:26 Part maffe
17:25:27 Quit lids (Remote closed the connection)
17:25:28 Join pondlife [0] (
17:25:30 Join saratoga3 [0] (i=98039921@gateway/web/cgi-irc/
17:32:46 Part |cub|
17:33:35linuxstbpondlife: I've just read your buffering proposal, and am confused. Are the API functions you list intended to be used by the playback engine (to start files buffering) or by the codecs (to read data from the buffer) or both? There is already a simple API to access the audio buffer present in the codec API.
17:34:14 Quit eck (Connection timed out)
17:35:27 Join joa [0] (
17:37:38pondlifelinuxstb: For the playback engine to use.
17:37:52pondlifei.e. To handle all disk -> buffer transfers
17:39:45 Join lids [0] (i=lds@gateway/tor/x-248ac76469130f8b)
17:40:01 Quit bluey- ("Leaving")
17:40:40linuxstbOK, so what's bufread() for?
17:41:18pondlifeTo obtain a pointer to the next bit of data
17:41:29pondlifeAnd to ensure it's actually in memory
17:41:40linuxstbWhy does the playback engine need that?
17:42:05pondlifeDoesn't it need to feed a block of data to the codec?
17:42:21linuxstbNo, the codecs read from the buffer.
17:42:22pondlifeThis is SWCODEC only
17:43:29pondlifeHmm, so the playback engine is responsible for setting up the buffer ready for the codecs... including switching codecs as required and also removing metadata?
17:43:43pondlifeMaybe this is more MoB related than I thought.
17:44:14 Quit joa (Read error: 60 (Operation timed out))
17:44:25 Join joa [0] (
17:44:41amiconnpondlife: Please keep hwcodec in mind (not the current engine, but the possibility to use one engine for both)
17:44:48pondlifeOf course.
17:45:22pondlifeI'm hoping this will simplify playback.c almost out of existence.... which should help unification.
17:45:25amiconnHwcodec *should* be simpler in many cases (e.g. there can't be crossfeed as there is only one codec
17:46:15amiconnAnd there are no loadable codes, at least not in the same manner as on swcodec
17:46:37amiconnPCM will be handled differently, in a MAS state machine
17:46:50pondlifeHopefully we can see MASCODEC as a special MP3 codec, ultimately?
17:47:21pondlifeOr rather have an MP3 (and a WAV) codec for MAS.
17:47:46saratoga3are there still devices with the mascodec being made?
17:48:16pondlifesaratoga3: If not, there are still devices with the mascodec in everyday use.. ;)
17:48:52amiconnNot as a loadable codec
17:49:11amiconnThe MAS decodes & plays at once, you can't separate the 2 layers
17:49:43amiconnSo on hwcodec, there needs to be a MAS playback buffer system (similar to swcodec's pcm_playback)
17:50:05 Join maffe [0] (n=maffe@
17:50:21pondlifeBut HWCODEC needs encoded data, right? Not PCM...
17:50:45amiconnThe only extra work on the playback engine level is that the engine does not only need to tell the sample rate & mono/stereo to the underlying system, but also the format (MPEG audio or PCM audio)
17:50:47pondlifeI see the PCM part as being the bit that's removed.
17:51:12amiconnHow do you think PCM output should be handled on swcodec then?
17:51:27pondlifeI meant on HWCODEC.
17:51:33amiconnI think the PCM layer is necessary, as it will do the mixing for voice, crossfeed
17:51:34yossalinuxstb: I saw you mention the CEC hi x8 player in one of the logs - are you or were you trying to develop rockbox for this player?
17:51:46pondlifeI see HWCODEC as being the same as SWCODEC, but without PCM buf.
17:51:51amiconnpondlife: On hwcodec there will be a similar layer
17:52:01linuxstbyossa: I was curious about it, but never did any work on it, and don't plan to.
17:52:09amiconnIt is needed as the data from the playback buffer can't be fed directly to the MAS
17:52:25amiconn..because of the necessary bitswap
17:52:34 Join Lear [0] (i=chatzill@rockbox/developer/lear)
17:52:51pondlifeSo for HWCODEC the "codec" and "PCM" must be the same code, right?
17:52:53amiconn...unless we can add the bitswap directly to the buffering system
17:53:13yossaah, do you have any idea if Raenye or LinusN are still on it?
17:53:21amiconnNo, the 'codec' layer is what will not be present on hwcodec
17:53:48pondlifeWhat about switching between MP3 and WAV? Isn't that potentially possible by programming the MAS?
17:53:53amiconnOr rather, the 'codec' will be the bitswap
17:54:02pondlifeYep, that's what I meant.
17:54:04amiconnYes, but the transition isn't gapless
17:54:17pondlifeIn either case the buffering can be the same code though.
17:54:29amiconnSwitching MP3->WAV will take 1..2 seconds (10Kword download to the MAS via i2c)
17:54:50amiconnSwitching back will be faster (a couple of ticks)
17:55:08pondlifeShould still be possible, and analogous to loading a codec in SWCODEC.
17:55:15linuxstbyossa: No, I don't think anyone is.
17:55:20amiconnNot quite.
17:56:00amiconnLoading a codec in swcodec is asynchronous. As soon as codec #1 finishes decoding (several seconds before the end of the track), codec #2 already starts decoding
17:56:05yossalinuxstb: OK, thank you
17:56:10amiconnThis is necessary for crossfeed
17:56:23amiconnOn hwcodec, however, the switching must happen synchronous
17:57:00linuxstbpondlife: In my mind, the main difference is that on swcodec, codecs give data back to the core for further (DSP) processing and passing to the DAC. On hwcodec, the codec is linked directly to the DAC - i.e. you give the MAS some MPEG audio frames and you never see the audio again.
17:57:14amiconnIt would be best if we wouldn't need an extra buffer for the MAS engine, but that would mean to do the bitswap in-place, which might complicate things for the buffering engine
17:58:09amiconnlinuxstb: Also not quite correct. When switching, you need to make sure the mas actually played all data you passed it *before* switching to the other format.
17:58:24pondlifeIt would be easy for the buffering engine to deal with, surely... is it a byte-reversal or a multi-byte swap?
17:58:34amiconnOtherwise you would cut the end of a song as there is a buffer in the mas
17:58:53pondlifeamiconn: That's just an extra wait loop #ifdef MASCODEC somewhere.
17:59:12pondlifeOK, buffering could do that
17:59:36pondlifeJust an flag parameter for bufopen() ;)
17:59:37amiconnThe buffer can be longer than a second for low-bitrate mp3 stuff (e.g. audio books in 16kHz mono)
17:59:40 Join lowlight [0] (i=c730190a@gateway/web/cgi-irc/
18:00:16amiconnpondlife: All audio data (mpeg audio or raw pcm) needs to be bitswapped, metadata of course not
18:00:23 Quit yossa ("ChatZilla [Firefox]")
18:00:47amiconnByte swapping (for supporting wav vs. aiff) can be handled by the mas, but you need to tell it what it has to play
18:00:56pondlifeHmm, another reason to deal with this as part of MoB.
18:03:33 Quit Nibbier (Read error: 104 (Connection reset by peer))
18:03:39Nico_Ppondlife: looks like I'm going to need to be writing buffering.[ch] and get some help for the rest
18:04:30Nico_P'cause rewriting the whole playback engine sounds like a biiig task for me
18:04:45amiconnBut afaik all the stream-related metadata is already stpored anyway (I mean things like sample rate, mono/stereo etc)
18:04:47 Join Nibbier [0] (
18:05:42 Quit lowlight ("CGI:IRC (EOF)")
18:06:50 Quit petur ("home")
18:08:51pondlifeNico_P: Ignore what I just mentioned about bitswapping, I think that will need to be done using a seperate buffer on the way to the MAS.... the buffering code shouldn't care about such things.
18:09:16Nico_Pphew :)
18:10:52pondlifeI didn't realise that the codecs took data direct from the buffer though, I thought that they did a callback to the playback code whenever they needed more data.
18:11:01pondlifeSo some basic assumptions are wrong :(
18:11:28pondlifeBeen too long with my head out of the code.
18:12:09linuxstbWell, it's both - a callback to the playback code to get data from the buffer...
18:12:33linuxstbSee the functions defined in codecs.h
18:12:34pondlifeOK, so why can't that callback be routed through as a bufread() call?
18:12:50linuxstbIt could, but there is more to the api than just that function.
18:13:06pondlifeIndeed, but the playback code will still be there to handle that...
18:13:19pondlifeI'm trying to break the monolith up a bit.
18:13:36 Join syof [0] (i=420046b6@gateway/web/cgi-irc/
18:13:37amiconnpondlife: I agree that a separate buffer for the bitswapped data would be better, but theres a reason why I would like to see it happen in the main buffer
18:13:52linuxstbpondlife: I can understand that, but what else does the playback engine do apart from buffer files and give data to codecs?
18:14:21pondlifelinuxstb: Handle metadata, crossfeed, voice....
18:14:31pondlifeThese are things it probably shouldn't do of course!
18:14:42amiconnWith a separate buffer, we'd need a separate thread that does the copy-and-swap operation, plus the amount of extra buffer ram will probably become a problem
18:14:55pondlifeIf we can section things off, the problem code will be easier to resolve tidily... I hope
18:15:00syofWhen doing the initial database build on a 5gen ipod i get a "prefetch abort at 83C383C2" message and rockbox freezes. I have no copy-protected music or videos and no mp3s with album art. What am I missing?
18:15:18amiconnDepending on the latency we can guarantee, the buffer might be too much for the little archos ram
18:15:28pondlifelinuxstb: I'm hoping this won't result in lots of new code required, just a sensible partitioning of the existing code.
18:15:49pondlifeamiconn: How low a latency (samples) can we get away with?
18:16:01linuxstbpondlife: I'm all for a more modular playback engine...
18:16:17amiconnDepends on how fast the thread is rescheduled...
18:16:36pondlifePlus, if the API is suitable, this might allow for better buffering strategies and buffering files for video and perhaps other purposes.
18:17:06pondlifeI'm trying to think of it as more than just a playback split, although that will be the initial intention.
18:17:27amiconnBut 0.1 second of 48kHz 16 bit audio is already 19KB, and we need 2 buffers at least as we need to bitswap before handing the buffer to the mas
18:17:37pondlife2 buffers?
18:17:53pondlifeWhy not just copy and swap from the audio buffer to a MAS buffer?
18:17:58 Join lowlight [0] (i=c730180b@gateway/web/cgi-irc/
18:18:15pondlifeOne buffer playing, while another fills... I assume
18:18:16amiconnThe mas will need to be fed continuously
18:18:40pondlifeNot a single ring buffer? Timing would be troublesome I suppose.
18:18:51amiconnThe mas is fed via dma...
18:19:01amiconn(controlled by 2 ISRs)
18:19:12 Part maffe
18:19:30 Join maffe [0] (n=maffe@
18:19:37lowlightwould things be easier/simplier if the codecs were "pushed" data to decode (instead of the codec "pulling" data)?
18:19:40amiconnSure, one could try to swap-behind. But this sounds nasty...
18:20:00pondlifeHmm, sounds a little similar to the plan jhMikeS was making for mixing on SWCODEC...
18:20:15linuxstbamiconn: How is the bitswapping done now - at the same time as reading from disk? i.e. read a chunk, bitswap it, read next chunk, ....
18:20:38pondlifeMust be done post-metadata parsing.
18:20:46amiconnlinuxstb: No, it's read big chunks until buffer is filled, then swap
18:21:03lowlighthave the codec register a decode_frame callback which playback feeds data to
18:21:06amiconnThis saves battery
18:21:19linuxstblowlight: I've always thought that would be a more logical way to do it.
18:21:49amiconnlowlight: I can't see how that changes buffering requirements...
18:21:59lowlightit seems for logical for video playback too
18:23:02amiconnlinuxstb: The only situation where it swaps directly after reading a small chunk is when the buffer runs really low
18:23:18linuxstbamiconn: We're back to the old idea of codec-specific loading functions. For the MAS, this could handle the bitswap after it's finished reading the data.
18:24:18pondlifeIt sounds like the bitswap should be done by the code that parses metadata - codec-dependent of courstr.
18:25:13linuxstbThe metadata code currently only parses the metadata - nothing more.
18:25:50pondlifeYes, but we'd want all codec-dependent code in one place so a single decode-and-prepare pass would be possible.
18:27:43pondlifeForget that too...
18:27:47pondlifeI wasn't thinking!
18:29:07lowlightamiconn: I was thinking more about simplifying playback unification by switching to a playback model that feeds the codec/mas
18:29:22 Join Elim [0] (
18:29:46pondlifelowlight: That's kind of how I assumed SWCODEC was working already :/
18:30:02linuxstblowlight: What do you think about separating the container-parsing (and seeking) code from the codecs themselves?
18:30:05 Join bawb2 [0] (
18:32:43lowlightcan't say I've thought everything through :)
18:33:26 Quit Elim ("Leaving")
18:35:20 Join Elim [0] (
18:35:55 Join Thundercloud_ [0] (n=thunderc@
18:37:29 Join raphi [0] (
18:43:12 Quit Thundercloud (Read error: 60 (Operation timed out))
18:45:17 Join kaaloo [0] (
18:45:52 Part kaaloo
18:47:09 Part maffe
18:47:18 Join maffe [0] (n=maffe@
18:50:21 Quit raphi (Read error: 60 (Operation timed out))
18:50:31***Saving seen data "./dancer.seen"
18:52:48 Quit obo ("KVIrc 3.2.6 Anomalies")
18:58:25 Part linuxstb ("Leaving")
19:04:45 Quit syof ("CGI:IRC")
19:04:48 Join pixelma [0] (i=d5494c0d@rockbox/staff/pixelma)
19:08:23 Join petur [0] (n=petur@rockbox/developer/petur)
19:09:47 Join spiorf [0] (
19:10:47Nico_Pdoes someone know if it's possible to have an optional item in a menu without having it have a dynamic text ?
19:11:16Nico_Pby optional I mean that can dynamically be shown or hidden
19:11:55 Part maffe
19:11:58 Join maffe [0] (n=maffe@
19:12:46 Join Psilonaut [0] (
19:14:38Nico_Papparently it is :)
19:15:18amiconnlowlight: How is the initiator of data transfer (codec vs. playback engine) going to make buffering more or less complex?
19:17:56 Part Llorean
19:21:33XavierGramiconn: any more info on this "ground loop" and how I may solve it?
19:22:05XavierGrI am googling it atm
19:22:13 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
19:22:21amiconnYou need to break the loop somehow. As you radio obviously connects audio ground to true ground, you have 2 options
19:22:39amiconn(1) Don't power the Ondio with the car charger
19:23:58XavierGryeah (1) doesn't produce noise, but that's the whole point, to be able to avoid batteries
19:24:06amiconn(2) Put a decoupling transformer for NF between the Ondio audio out and the radio line in
19:24:23 Join mykl [0] (i=c74048fc@gateway/web/cgi-irc/
19:24:30myklHi guys
19:24:39myklany1 can help plz ?
19:24:39XavierGrhmm let me google that too haven't heard of it
19:25:07myklI've installed latest version of rockbox-ipodvideo to my 30GB ipod
19:25:09amiconnHmm, NF isn't correct english
19:25:24XavierGramiconn: isn't there something I can do with the radio wiring? maybe I could provide audio ground to the radio?
19:25:26myklbut the light doesn't switch on after booting :(
19:26:06 Join obo [0] (
19:26:12amiconnXavierGr: The radio does have audio ground. The problem in an electrically noisy environment like a car is that ground isn't always ground
19:26:16myklany idea about it ? thx!
19:27:16XavierGramiconn: so I guess my options are limited to (2) then?
19:27:24amiconnYou have a loop: radio audio ground −−(audio cable)−− Ondio −−(power cable "minus")−− car charger −−(car) −−- radio audio ground
19:27:43amiconnYou can break this loop (for dc) at two points
19:28:09amiconnEither in the audio cable, between radio and Ondio, this would be (2)
19:28:27amiconnOr in the power cable, by simply not having it. This would be (1)
19:28:40XavierGryes indeed
19:29:01*amiconn simply runs his rockbox of choice from batteries in the car
19:29:09XavierGrah so you had the same problem?
19:29:19amiconnOr rather, will do so again after my trusty aux-enabled radio is mounted again
19:29:30amiconnNo, I simply didn't try it
19:29:37amiconnToo much of a hassle
19:30:06amiconnThis won't only happen with the Ondio, but with any target
19:30:07myklyes, it seems like ground loop to me too...
19:30:08XavierGrwell the way I had imagined it, it wasn't a hassle at all
19:30:33XavierGramiconn: yeah definitely, I read some reports with the same problem on other devices too
19:30:45toffe82I will say it is better to let the ground on the power cable and cut it on the audio cable
19:30:45amiconnWell, there would be another option: a car charger that is dc decoupled. But I don't know whether such beasts exist
19:30:58myklany user of this file here, please ?
19:31:20XavierGrmy plan was to leave my ondio on the car all the time, fire up the engine push the on button and play
19:31:21amiconnTechnically they're not very difficult to implement nowadays
19:31:33XavierGrbut with batteries I will have to take recharge batteries
19:32:03XavierGrand given ondio's low battery skills this is dissapointing
19:32:59amiconnThe Ondio battery performance isn't bad... (unless you have the battery-sucking problem)
19:33:16XavierGrfor me 10 hours is mediocre to bad
19:33:25XavierGrI am used to my H340 with wooping 30 hours
19:33:39amiconn30 hours?
19:33:49amiconnThe H300 is more like 20...
19:33:57XavierGr1900 mah battery
19:34:07amiconnIf you want really long runtime, try to get an M5L
19:34:53XavierGryeah it is pretty awesome in that aspect
19:34:59*amiconn really wonders whether dc decoupled car chargers exist
19:35:17XavierGrtoffe: so you suggest to make an audio cable without audio ground and couple the ondio with the radio with it?
19:35:52toffe82yes or try a 1:1 transformer
19:36:02 Join kaaloo [0] (
19:36:10XavierGrhmm interesting
19:36:10toffe82FGA-40 Ground Isolator is 1:1 audio isolation transformer with 10 kohm impedance from Monacor (Best.-Nr. 06.4370)
19:36:22 Part kaaloo
19:36:24toffe82check this site :
19:36:28myklyup, mouser isolator will help too
19:36:44XavierGrthanks guys I will see what I can do.
19:37:13XavierGrsad part is that I wasted 50 euros on car chargers cables and USB staff for it
19:37:30myklplease what can I do to get my iPOD display light on when booting rockbox?
19:38:35myklI do not use bootloader from apple and rockbox firmware does not switch light on :(
19:38:55 Quit gursikh ("Leaving")
19:39:16toffe82XavierGr: don't worry ,on mine, I go to the car audio through a cassette and I hear some electronic noise coming from the electronic of the car (each time I speed up or down, I heard a noise)
19:39:17myklI have to wait untill time set for background lightning and then all works fine
19:40:08XavierGrtoffe: sounds like the alternator to me (from googling my problem I read what you are describing)
19:40:13amiconntoffe82: Ugh, cassette adapter usually sound rather bad. I had that once, and never want to go back once I switched to using AUX in
19:40:36amiconnEven without induced noise that is
19:41:09toffe82I have an old audio system and it is a company car so I can't modify but it is way better than fm transmition
19:41:53toffe82I don't have any problem in the car of my wife with the cassette, it works really good
19:41:53XavierGryeah, wasted my money on an fm transmitter once
19:42:29XavierGrit couldn't tune to the specified frequency even next to the antenna, horrible
19:42:31toffe82I can't use a fm transmitter because there is no free frequency where I live :)
19:42:50toffe82it is all use between 88 and 107 mhz
19:43:05 Join webguest35 [0] (i=3fad1312@gateway/web/cgi-irc/
19:43:47webguest35austriancoder: Is there any chance that the enabling of line1 in the as3514 driver broke your audio-API patch?
19:44:14webguest35Since you committed that change, the patch causes compile errors for me.
19:45:29 Quit Psilonaut (Read error: 110 (Connection timed out))
19:45:57 Part mykl
19:47:02saratoga3does anyone here have any experience working with ffmpeg's codecs?
19:47:06saratoga3i'm somewhat stuck at the moment
19:47:38 Quit Viljar`zZ (Read error: 110 (Connection timed out))
19:49:17XavierGrnice find, now I will have to see if I can get it on my local store
19:49:46XavierGrah amiconn: btw, I found an EL foil on my local store, but... it was huge and needed 12 volts to operate :(
19:49:53 Quit pixelma ("CGI:IRC 0.5.7 (2005/06/19)")
19:50:18amiconnEL needs ~100 volts AC ~400Hz to operate
19:50:28amiconnThat's why you need an inverter circuit
19:50:50amiconnEither onboard or separate
19:50:53 Quit darkless (Read error: 110 (Connection timed out))
19:50:56 Quit webguest35 ("CGI:IRC")
19:51:00amiconn(onboard the Ondio I mean)
19:51:02 Join webguest35 [0] (i=3fad1312@gateway/web/cgi-irc/
19:51:07XavierGrah so even if I could get the original EL foil then I couldn't be able to mount it without the inverter?
19:54:10 Quit webguest35 (Client Quit)
19:54:21 Join Psilonaut [0] (
19:56:38austriancoder webguest35; jep... it could broke it. But I will come with an improved version of the API-patch today or tomorrow..
19:57:26 Join Rondom [0] (
19:57:32 Join ctaf [0] (
19:57:44 Part ctaf
20:00:14XavierGrthe lower end of the audio jack (not the tip) is the ground right?
20:00:41Nico_Plooks like we have a stallen build
20:01:04 Join webguest86 [0] (i=3fad1312@gateway/web/cgi-irc/
20:01:36webguest86austriancoder: (I asked the question earlier...) Thanks... no sweat. Just wanted to be sure it wasn't me. :-)
20:03:03 Join chooowner [0] (
20:03:03 Quit webguest86 (Client Quit)
20:04:06 Quit bluebrother ("Leaving")
20:04:54 Part chooowner
20:06:44toffe82XavierGr: yes, the contact closest to the cable
20:06:53 Quit Guile` ("KVIrc 3.2.0 'Realia'")
20:07:33 Join ptcemetary [0] (
20:12:35toffe82amiconn: you have a ground loop take your hands from the keyboard ;)
20:13:18amiconnNo, Grip didn't like me and didn't tell me I really cannot use " in a filename
20:13:37amiconnSo I have to double-check all my flac rips and rerip missing tracks
20:14:00*amiconn will switch back to windows for ripping
20:15:32 Join entheh [0] (
20:16:09 Join Domonoky_ [0] (
20:23:56 Join major_works [0] (n=a08e9562@
20:26:18 Join moos [0] (
20:27:24 Quit major_works ("leafChat IRC client:")
20:27:37lexhmm hey, if i put a new firmware into my ipod with ipodpatcher, does the bootloader still remain there?
20:27:49lexor do i need to take a backup of the bootloader and flash it again then? :o
20:28:19 Join major_works [0] (n=a08e9562@
20:29:39 Quit Spec (Nick collision from services.)
20:29:42 Join Spec_ [0] (n=spec@
20:32:44 Join rigel [0] (
20:33:33rigelyall are aware that the h10 20 gb version is broken as of last night, yes?
20:33:36rigel"codec failed"
20:33:44 Quit Domonoky (Read error: 110 (Connection timed out))
20:36:22LearAnd you made a proper install?
20:36:35 Quit atsea- (Connection reset by peer)
20:39:44rigeli unzip it straight to the .rockbox directory on my h10
20:42:25LearAnd the bootloader is recent enough? (Don't have an H10 myself, so I can only give general help...)
20:43:39 Join midkay_ [0] (
20:50:06 Part major_works
20:50:24 Join major_works [0] (n=a08e9562@
20:50:34***Saving seen data "./dancer.seen"
20:51:28 Part major_works
20:54:10dunderwhy are their no plugins listed for e200??
20:54:13 Join major_works [0] (n=a08e9562@
20:55:00rigelyeah, the bootloader is part of it, iirc. i have it set up for dual boot
20:55:10rigelive never had a problem with the codecs before
20:57:59LearCodec problem is often a version mismatch between the core (rockbox.<something>) and the codecs.
20:58:41 Quit major_works ("major_works has left the building...")
21:00:02 Quit midkay (Read error: 110 (Connection timed out))
21:02:00 Part maffe
21:02:39 Join maffe [0] (n=maffe@
21:02:59 Quit maffe (Remote closed the connection)
21:03:29 Join maffe [0] (n=maffe@
21:03:46 Part maffe
21:03:46dundercan someone please elaborate on the svn page...
21:04:07dunderi can't download svn 100% of the's like a 5min window a day i can download the current svn??
21:04:13 Join maffe [0] (n=maffe@
21:04:19 Join atsea- [0] (i=ariel@gateway/tor/x-d828d6fa7f65ab4e)
21:04:21 Quit maffe (Remote closed the connection)
21:04:22badsheepyits always worked for me fine
21:04:30badsheepyusing the instructions from the svn page
21:04:47 Join maffe [0] (n=maffe@
21:05:20badsheepyjust svn updated and it worked fine too
21:05:48dunderbadsheepy: ok, i don't understand the instructions other than "or download an older build from the Archived Daily Builds page", which i don't wanna do
21:06:04dunderall i see is a lotta zeros...
21:06:04badsheepytheres only like 3 lines, where have you got to
21:06:57dunderok...i get it now
21:07:09badsheepylol ok, and i just loaded the page too
21:07:54 Join bluebrother [0] (i=roNPwhiE@rockbox/staff/bluebrother)
21:09:23 Part maffe
21:10:00 Join maffe [0] (n=maffe@
21:14:31 Join syof [0] (n=Jeff@
21:14:43syofWhen doing the initial database build on a 5gen ipod i get a "prefetch abort at 83C383C2" message and rockbox freezes. I have no copy-protected music or videos and no mp3s with album art. What am I missing?
21:17:44 Nick chrisjs169|brb is now known as chrisjs169 (n=jack@unaffiliated/chrisjs169)
21:20:32 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
21:22:55preglowmarkun: you haven't read about a good way to interpolate hrtfs, have you?
21:25:20 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
21:25:40 Quit amiconn (Nick collision from services.)
21:25:40 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
21:26:03 Join darkless [0] (
21:26:59feisarjust updated to latest build on sansa and I can't map the scrollwheel in rockboy anymore :(
21:27:14feisarit doesn't work on default mapping
21:27:57feisarused to work ok
21:29:55 Join vader [0] (
21:30:03 Quit dunder ("Leaving")
21:30:50 Quit Psilonaut (Read error: 104 (Connection reset by peer))
21:32:32 Part vader
21:33:13 Join Psilonaut [0] (
21:34:27 Join bluey- [0] (
21:37:41 Join dunder [0] (
21:37:56dunderi don't get it...i did the sansapatcher and restarted, shouldn't rockbox load now??
21:38:34 Quit Siku ()
21:38:37feisaryep installed 29.4 version and scrollwheel works ok
21:38:44 Nick midkay_ is now known as midkay (
21:40:46 Join MIG [0] (, i got it installed...
21:45:53dunderbuilt the database
21:45:57dunder..but i can't find the songs
21:46:54dunderAND I got a "Data abort at 00040F58"
21:49:22feisaryeah I noticed those data abort crashes also earlier build didn't crash even once like that
21:49:37badsheepyyou are having all the problems today arentcha dunder :o
21:51:19dunderwhere is the music supposed to be?
21:52:24badsheepyunder database for the database
21:52:41badsheepyif its not obvious id say somethings gone wrong, but im not qualified to diagnose what :)
21:53:59syofyeah, i have the same database problem. googling suggests the problem is copy-protected files OR album art choking the db build. i removed my copy-protected music and stripped my album art. still no joy
21:54:24 Quit Lear ("ChatZilla [Firefox]")
21:55:03badsheepydid you try an older build?
21:56:15syofno, which build do you recommend i try? i'm on 5gen(v1) ipod
21:56:25bluebrothersyof, did you remove the database files before retrying?
21:56:52syofbluebrother: no, i did not. where are those files?
21:58:07badsheepyin /.rockbox/ i think
21:58:41syof(that makes sense... I decided to ask the obvious question whilst plugging the ipod in)
21:58:56 Quit dunder (Read error: 104 (Connection reset by peer))
21:58:57bluebrother /.rockbox/*.tcd
21:59:25syofok, i have no .tcd files in the main .rockbox dir
21:59:38syof(or any subdir)
22:00:30syoffor what it is worth, it chugs along and finds like 700 files before croaking with the "prefetch abort" message
22:04:39syofwhere can i download previous builds?
22:06:38Slasherisyof: that most likely means you have a broken file and have found a bug in metadata parser or a disk problem
22:07:22syofSlasheri: can i get into some sort of debug mode to determine which file it is (so i can remove it and/or send it to someone for further diagnosis)?
22:07:57Slasheriunfortunately that is not possible with ipods yet (but would be very useful indeed)
22:08:23Slasherithe best you could is to find the broken file manually (removing half of files and so on)
22:08:36 Join Guile` [0] (n=Guile@
22:08:42syofbummer. can you direct me to a place to dl older versions of rockbox? the navigation on the site isn't helping me
22:08:57Slasheriand then send the file to me or somebody to fix the issue
22:09:08 Join BigMac [0] (
22:09:16Slasherisyof: i don't think an older version is going to help
22:09:39syofI'll go about it the way you suggested, then
22:09:54BigMacHey on my gigabeat f40, with the current svn and album art applied, I can't access the files directory
22:10:05BigMacI am going to try the clean svn right now
22:10:15Slasheriand when you find the file, please send it to me (dcc works or at email)
22:10:19 Join gursikh [0] (n=khalsa@unaffiliated/gursikh)
22:10:19Slasherineed to go now, night
22:10:28BigMacbut will report back if it doesn't work because this is pretty disturbing
22:11:21 Join hostf4cekilla [0] (
22:13:02bluebrotherBigMac, checked the file types setting already?
22:13:04 Quit lowlight ("CGI:IRC (EOF)")
22:13:45 Quit hostf4cekilla (Client Quit)
22:13:56BigMacyah and it works fine with clean svn (and apparently leds were fixed in recent change)
22:14:04 Join hostf4cekilla [0] (
22:14:06BigMacso I think I am going to try extracting again
22:15:19badsheepyfor tagcache to find a file in ram by filename does dircache have to be on?
22:15:51badsheepyi was looking at find_entry_ram and it appears to need it
22:15:59badsheepyis there something better to use?
22:16:19 Join webguest00 [0] (i=4a475751@gateway/web/cgi-irc/
22:16:49webguest00hey has there been any development for fastforward/seek in mpegplayer
22:17:09 Quit The-Compiler (Read error: 110 (Connection timed out))
22:21:46bluebrotherwebguest00, not yet. But it's planned.
22:21:51BigMacalright seems to work with current svn
22:22:59 Join robin0800 [0] (
22:24:22 Quit FOAD ("I'll be back")
22:25:15BigMacDoes anyone know if album art and bmnp resize work in equillibrium with each other ?
22:25:52 Quit bluey- ("Leaving")
22:28:47 Join FOAD [0] (
22:29:05 Quit petur ("Zzzzz")
22:29:53webguest00so i imagine there would be no way to estimate when there would be seek/ fastforward support
22:30:51bluebrotherno. Estimating timeframes usually doesn't work ;-)
22:31:17webguest00sigh, ok well thanks anyway
22:31:32webguest00can i make a suggestion for it though
22:31:33schulabtw. is it just me, or does the sound under the original firmware sounds better then on rockbox (sansa)
22:31:46webguest00make fastfoward more similar to that of an ipod rather than that of my sansa
22:32:04webguest00it fastforwards by like 10 secoonds and is not helpful really
22:33:03webguest00to send someone a message you do /msg right?
22:33:16webguest00it keeps going to a new window with just htat person
22:33:32bluebrotherbut you need to be registered to freenode if you want to send pm's
22:33:39 Part gursikh ("Leaving")
22:34:29webguest00k i am going to go
22:34:41webguest00jus twanted to say thanks for all the work put into rockbox
22:34:51webguest00i really appreciate the time nad effort
22:35:02 Quit webguest00 ("CGI:IRC")
22:36:09 Join __IRCMonkey__ [0] (
22:37:54 Quit amigan (Read error: 104 (Connection reset by peer))
22:38:00 Join amigan_ [0] (
22:38:02 Quit amigan_ (Client Quit)
22:45:55 Part rigel
22:48:27__IRCMonkey__I have a 5G iPod Video and have a few questions about the MPEG Player Plugin. I've read the page and the relavant FAQ pages.
22:48:33 Part BigMac
22:50:30__IRCMonkey__AFAIK, there is only support for MPEG-1/2; Will there eventually be support for MPEG-4 (part 2, 10)?
22:50:38***Saving seen data "./dancer.seen"
22:52:35Nico_P__IRCMonkey__: if someone decides to work on it
22:54:39__IRCMonkey__Nico_P: so currently no one is working on it OR have there been some experimental patches already in svn/cvs?
22:55:01robin0800Soap, this is not correct if you rember about battery matters on ipods
22:55:14Nico_Pno, there is nothing for MPEG4 ATM
22:56:13 Quit Elim ("Leaving")
22:56:19 Quit pondlife ("disconnected has pondlife")
22:56:58robin0800Soap, i think battery capacity is not working properly
22:57:46chrisjs169also, why's the build hanging?
22:58:48XavierGrah crap! grounding loop didn't get fixed, with either an audio wire without ground, or car charger with out ground going to my ondio :(
22:59:38robin0800Soap,if it is changed it simply alters the time which would only be correct if Current is taken into account
23:00:27__IRCMonkey__I'm a very novice developer. SO i'm asking: 1) How hard would be to port and/or VLC/Mplayer into Rockbox? 2) If this could be ported, would the developers allow this?
23:01:54robin0800Soap,running time should be capacity divided by average current?
23:03:44robin0800Soap, I think average current is hard coded and not different for differrent models
23:03:54 Quit joa (Read error: 104 (Connection reset by peer))
23:04:50toffe82XavierGr: without the power connected , you have the noise also ?
23:05:05bluebrother__IRCMonkey__, hard.
23:05:07robin0800Soap,this is why the Ipod has the right time But the wrong Caoacity 1300 not 600
23:05:22bluebrotherthe main problem with pc software is that it uses floating point operations
23:05:59bluebrotherwhich would be dog slow on the Ipod / Irivers
23:06:15bluebrotherplus, your ressources are much more limited.
23:07:55__IRCMonkey__bluebrother: right, but how about "The Core Pocket Media Player" (
23:08:49bluebrotherfrom the wikipedia overview it doesn't look like it would be license compatible
23:09:14bluebrotherRockbox is GPL, thus all code that gets integrated needs to be GPL compatible at least
23:10:04__IRCMonkey__right, I think TCPMP is GPL compatible, lemme check....
23:10:26chrisjs169Anyone know why the builds are hanging?
23:11:12bluebrothertheir website seems to be down, at least I can't access it atm
23:11:27preglow__IRCMonkey__: rockbox _is_ a media player, asking "can we port this media player to rockbox" doesn't make sense
23:11:48__IRCMonkey__yeah, it is down. But The last time I checked, there is both a GPL version and a licensable OEM version (dual-license like Qt)
23:12:04bluebrotheranyway, as their targets are various versions of windows I bet it uses float operations.
23:12:45 Join joa [0] (
23:13:53 Part syof
23:14:37__IRCMonkey__hmmm... I would have thought iPod 5G would have a CPU with good enough floating point unit(s)
23:14:45preglowthey don't have floating point units
23:14:53preglownone of our current targets have
23:15:00preglowall our codecs use integer maths
23:15:17bluebrotherwhat about the gigabeat?
23:15:21*chrisjs169 thinks he found out why the builds aren't done....
23:15:28preglowgigabeat s does, i believe
23:15:33preglowbut none of the ones we currently support
23:18:03__IRCMonkey__hmmm.. Am I understanding this correctly: No one can (easily) port any of the existing GPL media players because they rely on floating point units. Instead it would be easier to implement MPEG-4 etc. from scratch?
23:18:47saratoga3theres probably interger mpeg4 decoders out there somewhere
23:19:22bluebrotherif there is no fixed point implementation someone needs to either convert an existing one to fixed point or start a new one
23:19:42Nico_Pchrisjs169: why's that ?
23:19:43bluebrotherdepending on what's around it depends what's the faster / easier solution
23:20:37chrisjs169Nico_P: I was going through each of the builds that were up, and found that the iPod 4G grayscale wasn't done, which was one of the three my computer was supposed to build
23:20:42pregloware we talking mpeg4 video?
23:20:55 Join ac_2 [0] (
23:21:08 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
23:21:10ac_2MIG: hi
23:21:23preglowi think video codecs in general are easier to port to fixed point than audio codecs, and we have lots of them. but then again, video codecs weren't usually used on devices with no fpu
23:21:25__IRCMonkey__mpeg4 vide
23:21:34ac_2MIG: its me.. austriancoder
23:21:38__IRCMonkey__From a hardware engineering standpoint, is there a reason why companies like Apple left out floating point units - reduce power consumption?
23:21:45preglow__IRCMonkey__: and cost
23:22:08saratoga3h.264 decoders are integer based anyway, so porting one of those should be fairly trivial
23:22:20saratoga3assuming you had a fast enoughcpu
23:22:24preglowh264 is wildly complex, thugh
23:22:40saratoga3i think simple profile isn't much slower then mpeg2
23:22:44preglowmight no be
23:22:45saratoga3its supposed to be scalable
23:22:52preglowwhat profile is used the most?
23:23:19saratoga3apple uses the simple profile, I think bluay uses one of the more advanced ones
23:23:22saratoga3i think
23:23:34saratoga3i got about halfway through the h.264 spec before i got bored
23:23:47preglowi did a bunch of stuff related to h264 my final year at uni
23:23:50preglowbut forgot most of it by now
23:23:56preglowvideo codecs never were my major interest
23:24:01saratoga3same here
23:24:15saratoga3well i never did anything with them, but i share your disinterest
23:24:31 Join __IRCMonkey___ [0] (
23:25:12preglowplus, the lecturer was so occupied talking about how superior their design proposals always were, so i soon got tired of the entire course ;)
23:25:15 Join zivan56 [0] (
23:25:16__IRCMonkey___another question: will the MPEG player plugin eventually become builtin to Rockbox (i.e. not a plugin anymore)?
23:25:26preglow__IRCMonkey__: i can't see why not
23:25:33amiconnMost probably not
23:25:39preglow__IRCMonkey__: main reason it's aplugin now is because it's experimental
23:25:41amiconn(unless on a future PMP target)
23:25:54__IRCMonkey___right, but when it becomes stable enough... ?
23:26:08preglowamiconn: the argument that current targets weren't MEANT to play videos is really thin, if you ask me
23:26:11amiconnIt's also a plugin because video isn't considered a core feature on a DAP
23:26:22saratoga3seems like someday rockbox will gain some PMP abilities, though it might be a long ways to go until CPUs are fast enough and people are interested
23:26:49bluebrotherRockbox is (at least currently) about playing music. If it should have video in the core some things need to change too
23:27:02XavierGrtoffe82: my problem is that: whenever the [audio of radio and ondio are coupled] and [car auxiliary power is coupled with my ondio] I get very loud hissing sounds. (even if the ondio is turned off)
23:27:23XavierGrtoffe: I tried what you suggested though without results, I mean:
23:27:59XavierGrtoffe: from the car auxiliary power (+) to ondio (-) nowhere and audio as usual
23:28:24saratoga3XavierGr: presumably this is because your player shorts the signal and power ground which your radio does not like?
23:28:38__IRCMonkey___really?!? only about music... I think that most people would be hesitant to remove Apple firmware and put Rockbox because Rockbox cannot do everything Apple can do (H.264 video) (but obviously it can do much more - FLAC, OGG etc.) - If this problem were overcome, I think more people would switch to Rockbox
23:28:42XavierGrtoffe: and car auxilliary as normal and audio cable without ground
23:29:14saratoga3i don't think Ipod Video users are really a concern to development, beyond getting rockbox working on the video
23:29:23bluebrotherat least I don't see a need to convert Ipod users to use Rockbox.
23:29:36 Quit Rondom ("Ex-Chat")
23:29:45XavierGrI guess I will have to contact the man that installed my radio for more info
23:29:58bluebrotherit's done by volunteers because of the fun of doing it, not because of converting Ipod users ...
23:30:03XavierGror buy (or make) an isolator
23:30:13preglow__IRCMonkey__: it's not really a problem, if you ask me, it's just a question about maturity. the video player plugin is not currently fully developed
23:30:27preglowi can see no reason not to put it in the core when it one day is stable
23:30:58jhMikeSif we did pick a "ready made" solution, it's my guess we'd have to redo a good portion of it anyway to make it meet our needs
23:31:12preglowjhMikeS: a very solid portion at that
23:31:19jhMikeSbeyond the floating point issue
23:31:43preglowoh yes, not even counting that
23:32:05 Quit zivan56 (" HydraIRC -> <- Chicks dig it")
23:32:05saratoga3see: wma codec
23:32:20jhMikeSJust to start with, most of these implementation assume a disk the never spins down and right there you've got problems in buffering
23:32:43preglowsaratoga3: heard anything from the ffmpeg wma fixed point guy?
23:32:50n1sis there a significant advantage in size or quality of simple profile mpeg4 over mpeg2 on the kinds of low resolutions we have?
23:32:57preglowi am so hoping that'll be done for us
23:32:57saratoga3no, in fact i haven't seen anything at all about him
23:33:02XavierGrthe whole day today, was quite depressing too, nothing came out. Even the velcro mod I did on my ondio (to hold it on the console) failed completely. what a waste...
23:33:06preglowmerbanan: nothing new, i take it?
23:33:38saratoga3there seems to be no way to find things out about ffmpeg
23:33:53preglowwell, there's an irc channel and a mailing list
23:33:55jhMikeSvelcro mod? :\
23:34:02saratoga3no documentation, mailing list is almost impossible to search, and no one uses the IRC channel
23:34:21__IRCMonkey___right, but I think that Rockbox has potential to be as widely used as VLC/Mplayer/MPC (since there are so many PMP owners). maybe improving video functionality, might help it reach this potential.
23:34:31jhMikeSwith these open source libs, no documentation seems the rule
23:34:52saratoga3so far, I'm pretty sure even their "apiexample" program is broken
23:34:55preglow__IRCMonkey__: and people are indeed on it, so hooray
23:34:57*bluebrother doesn't see this potential
23:35:03saratoga3since it encodes a file, and then errors out trying to decode it
23:35:31toffe82XavierGr: I was saying just connect the ondio without power from the car so it will run on its battery and with the standard audio cable , you still have the noise ?
23:36:00preglowsaratoga3: tried linuxstbs wmadec?
23:36:09saratoga3yeah i used that initially
23:36:13saratoga3its the only thing i got working
23:36:20saratoga3however, its completely obscolete
23:36:20XavierGrjhMikeS: yeah I tried to stick velcro on the console (with 2 sided sticker paper) and then do the same to the back of my ondio. That way you can attach the device on upright position (like magnets on refrigerators)
23:36:51XavierGrtoffe82: no of course not! If the device is working from an external power supply sound is perfect :)
23:37:05jhMikeSXavierGr: and what, no sticky?
23:37:18saratoga3and attempting to use it with the current SVN code causes the asf parser to fail to detect that file as WMA, presumably because something doesn't get initialized properly
23:37:33XavierGrwell the heat of the sun melted the 2 sided sticker paper and the velcro fell off
23:37:33saratoga3and theres no example code at all for the file parsers
23:37:48XavierGrnext time I will try to use super glue
23:37:52preglowsaratoga3: sounds like it's approaching time to badger them on a mailing list, then
23:37:57XavierGrfor 2 hours it was quite cool
23:38:00saratoga3yeah i'm doing that
23:38:03saratoga3will send more emails
23:38:12saratoga3maybe one of them will get an answer
23:38:20jhMikeSnot suprising...dont' know if you have access to an automotive body shop supply. you can get hi bond foam tape that'll mount anything for good
23:38:44XavierGrhmm let me see what is this...
23:39:15jhMikeSthen you could mount a cradle or something
23:39:21schulahm, when i try to build rockbox from scratch i get an linker error: /usr/lib/gcc/arm-linux-gnu/3.4.6/../../../../arm-linux-gnu/bin/ld: error: no memory region specified for loadable section `.got' (binutils 2.17)
23:39:29 Join __IRCMonkey____ [0] (
23:39:51bluebrotherschula, you need to use gcc 4.0.3 for arm targets
23:39:51XavierGrit is always nice to find new items for various jobs, but it is always hard to discover the name in Greek in order to search for it here :P
23:39:58schulaah ok
23:41:17bluebrothercheck the wiki for the required toolchain version.
23:41:32saratoga3did Toni happen to mention if his SD patch for the sansa works with SDHC or whatever that standard is called
23:41:53 Quit __IRCMonkey__ (Read error: 110 (Connection timed out))
23:42:05saratoga3i'm loving the idea of getting an 8GB flash card for my sansa in a year or two
23:42:09 Nick __IRCMonkey____ is now known as __IRCMonkey__ (
23:42:40bluebrother SDHC isn't required to support SPI mode any more afaik
23:42:45jhMikeSit's for mouning body moldings/emblems etc.
23:43:01saratoga3without it though you're limited to 2GB though IIRC
23:43:16saratoga3which is pretty useless on an 8GB player
23:43:38schulabluebrother: oh ok, thanks :)
23:44:33saratoga3mostly wondering if I'll be able to upgrade the sansa to 16 or 24GB in a couple years
23:45:48*jhMikeS doesn't want the build server stuck for the next commit which could possible make alot of red :\
23:46:05chrisjs169jhMikeS: it's been stuck for an hour...
23:46:39*jhMikeS wants to give it swift kick
23:46:56chrisjs169For some reason the build master isn't getting the ipod 4g grayscale build from my server, even though it's done
23:47:36preglowchrisjs169: that might be cool info to have for debugging
23:48:01preglowbut i doubt it'll be fixed, only linus, bagder and zagor have access to the build server, and bagder is on vacation
23:48:34preglowjhMikeS: what's coming up?
23:49:30jhMikeSpreglow: just a redo to make recordable input selection more flexible using the cap masks like sample rates do now, but it touches every port so I want to be able to fix any problems
23:49:48chrisjs169preglow: apparently the 4g grayscale only starting building not more than half an hour ago, so I dunno what it was doing for the 30 mins before that (apart from two other builds compiling)
23:49:51jhMikeSe200 is the first port with FM Radio in but no Line In and this must be accomodated
23:50:32chrisjs169is there any way to test the FM radio up to this point?
23:50:45saratoga3i was thinking of adding a GSOC progress page in order to encourage people to keep everyone updated on teh progress of each project
23:50:49saratoga3does this seem like a good idea
23:51:08 Quit ompaul ("Today is only half as nice as tomorrow")
23:51:56jhMikeSusing the bitmasks makes the config file more compact and another header in the config.h expands it all into HAVE_xxx_IN for use in the code
23:53:10preglowsaratoga3: it does seem like a good idea, yes
23:54:20 Join dunder [0] (
23:55:16*jhMikeS thinks linus, bagder and zagor need a paging system to wake them up in an emergency :)
23:55:44 Quit __IRCMonkey___ (Read error: 110 (Connection timed out))
23:55:58preglowi think more people should have access to the build system
23:56:08chrisjs169i don't
23:56:29chrisjs169the more people that have access to the private ssh key, the more unsecure it could become
23:56:39jhMikeSyes, a limited account for developers to restart it if nescessary
23:56:46preglowwell, yeah, but since having a paging system is completely unrealistic...
23:57:03jhMikeSsomeone already mentioned a solution for this but I don't recall the name of it :\
23:57:04chrisjs169well, a limited account would be fine, i thought you meant full access to everything
23:57:05preglowi'm not saying people should have full access
23:57:12preglowbut there should be a system for resetting the whole thing
23:57:50jhMikeSthere's something the gives certain admin priveledges to certain users but is secure
23:58:25 Quit jgarvey ("Leaving")
23:58:57chrisjs169supposedly, the build master is supposed to issue a build to another server if it fails on the one it was assigned to, but it seems it isn't doing that
23:59:00preglowwe don't really even need that

Previous day | Next day