Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2006-11-19

00:00:42miepchen^schlafnano support would be nice
00:00:47Terinjokesand my 'Tast-o-meter' works now, reports 'your compatibility rating with Spug is "Medium"'
00:00:56Stroggthat's awesome
00:01:06StroggI'll have to check out your lastfm stuff. hehe
00:01:55TerinjokesSpug, sent you a friend request (grr..sounds like myspace.....)
00:02:36Terinjokesi'm user Terinjokes BTW, in case anyone want to look at what i listen to
00:03:05PaulPositionbwaaaha.. myspace. Was great for learning about events and shows, but now that lastfm does it, you think I'm ever going to load up a myspace page again? :p
00:03:26 Join hachi [0] (
00:03:31Terinjokesyep, mine, because mine is cool! :)
00:03:41Terinjokesoh, myspace, sucks
00:04:19hachican I get anyone to lay odds that rockbox will dual boot and work okay on the 5.5G 80G ipod in say... 4 months time?
00:04:49spugmy uneducated guess is that it'll work okay in 4 months
00:04:56spugbut the development is unpredictable really
00:05:04PaulPositionI'd say odds are good, the platform is interresting for sure so I'd say someone will get interrested.
00:05:06Terinjokesprogress has been slow...IDK really, its a whole new system, i'm very dobuous
00:05:54hachihow does the debugging of stuff get done, guess and check? or is there a way to attach a debugger of some kind?
00:05:55PaulPositionProgress was slow on the ARM too.. It's snowball effect, where some small things can lead to big things... then nothing new for a while until..
00:07:01hachiI love how the status page is like 'check the forums'... and the thread on the forums is 18 pages long
00:07:04hachithat's why I'm asking here
00:07:16PaulPositionguess and check, mostly, I think.
00:07:27spugyeah, well, as you can see on the forums, they're actively developing it
00:07:33spugthe thread is updated almost daily
00:07:41spugso i guess they're making progress steadily
00:07:49hachiI do want to smack people on the forums... every so often someone posts
00:07:54spugunlike what the nano 2G thread gives me the impression of
00:07:58hachi"I got rockbox working on my 5.5g"
00:08:24hachi*smack* what SIZE dimwit?
00:08:35PaulPositionfirst thing first.. Get a main() function to execute. Get it to lit up the lcd. Then hack away. (That's a few days work already me think)
00:09:10PaulPositionYeah, I guess that's why the devs don't use the forum all that much.
00:09:11hachiwell, then I'll go get one today... and hope that it gets working sometime soon :)
00:09:24PaulPositionWouldn't you rather wait.
00:09:37PaulPositionAnd buy it for a lower price anyway when the news breaks that it DOES work?
00:09:41Stroggbuy a used one on ebay and get something supported. hehe
00:09:46hachino, I'd rather try and force it to work... I'm an engineer, I like breaking things :)
00:10:08*Terinjokes wonders why recommended radio suddenlt turned french...
00:10:19 Join GreyFoux [0] (
00:10:55PaulPositionrecommended Radio ?
00:11:17hachihow do you reset ipods anyways? mash a few buttons?
00:12:05Terinjokesdoesn't work on the 3G, but IDK what does
00:12:21 Join Criamos [0] (
00:12:25rasher pretty thorough look at where the 5.5G 80gb ipod won't boot
00:12:52Terinjokesany 4.5 nano support?
00:13:46spugyou mean 2G?
00:14:04 Part nave7693
00:14:14spugnot yet
00:14:33Terinjokesi got a 1G, but really want a 2G
00:14:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
00:15:29JdGordonmorning all
00:15:45PaulPosition'night dude. ;)
00:16:19 Part Terinjokes
00:16:25 Join Terinjokes [0] (
00:16:41Terinjokesopps, *I* clicked the close button, LOL
00:17:41JdGordondoes anyone know how much over the rombox buids are?
00:18:37 Quit funky ("leaving")
00:18:38 Quit CriamosAndy (Read error: 145 (Connection timed out))
00:21:20 Quit merbanan (Remote closed the connection)
00:24:49 Join maquis [0] (n=erins@
00:24:56 Nick GreyFoux is now known as nGs|GreySleeps (
00:25:13maquisI have a question about setting up the rockboxui simulator on linux.
00:26:07rasherYou should ask it, then
00:26:18maquiswhenever i try to play an mp3 while in the simulator, it just displays "bad codec" on the simulator screen, and no sound comes out
00:26:24maquisrasher: sorry :)
00:26:41JdGordondid you do make install after the build?
00:27:02rasherAre you using x86_64?
00:27:05maquisthe errors say that it's trying to open mpa.codec, and that it's unable to open _temp_codec0.dll
00:27:19maquisrasher: i'm using x86
00:27:37StroggIs there a way to make rockbox read a conf file on bootup? or is there a default name that is always loaded?
00:27:43maquisJdGordon: for the UI? I thought so, i can try again, though.
00:27:57Stroggit's a nuisance to have to go to manage settings to load my config each time I turn it on
00:28:38maquisi was trying to do the simulator and the build at the same time...
00:28:42TerinjokesStrogg, are you properly shutting down the system?
00:28:43JdGordonStrogg: not yet, im working on this... and you dont have to.. you an just "run" the .cfg from the file browser
00:29:05StroggJdGordon: ooh.. I didn't know that.
00:29:24StroggTerinjokes: likely not, but still..
00:30:05Terinjokesdoes anybody know if rockbox and OF daul-boot correctly on the 1G nano? they both load on mine, but OF contains no files...
00:30:22Terinjokess /files/songs
00:30:22maquisJdGordon: looks like i must've messed that up... thanks
00:30:34rasherTerinjokes: have you put on any songs with iTunes?
00:30:39Terinjokesall of them
00:30:57rasherHm. Right, there was a problem like that once, I thought it was fixed
00:31:51Terinjokesthere was the option to downgrade the OF to a version that worked, and it did, but it would corrupt the bootloader every couple of boots
00:40:47StroggMan.. the battery life in the simulator is brutal. heheh
00:43:25TerinjokesStrogg, i still envy you
00:43:33*Terinjokes runs windows xp in the simulator
00:44:13Terinjokeshow do you double click with rockingwinbox?
00:45:38 Quit nGs|GreySleeps (Read error: 110 (Connection timed out))
00:46:46 Part Terinjokes
00:47:04 Quit lostnihilist (Read error: 110 (Connection timed out))
00:48:01 Join lostnihilist [0] (
00:48:59 Join spifwand [0] (
00:49:56obocould someone add "target/arm/iriver/*.[chS]" to /firmware/FILES - the h10 cannot be built from a source tarball - missing 2 .h files
00:50:43StroggHrmmm.. check the wiki.. I remember it saying something about not building the bootloader
00:51:46StroggAhh.. that's the other iriver's..
00:52:00rasherobo: done
00:52:10obothanks :)
00:52:49 Join bsk8er48 [0] (
00:56:11 Part bsk8er48
00:57:55 Quit getdow ("CGI:IRC")
00:58:05 Quit Criamos ("( :: NoNameScript 4.03 :: )")
00:59:30 Join Ribs [0] (n=freenode@
01:02:06***Saving seen data "./dancer.seen"
01:06:08 Quit Ribs2 (Read error: 60 (Operation timed out))
01:07:33spugthe iriver h300 has a Li-Pol battery?
01:09:03 Join Criamos [0] (
01:19:33SoapOSU FTW
01:23:29 Nick scorche` is now known as scorche (i=ScorchE@rockbox/administrator/scorche)
01:25:26blue_lizardspug: yes 1300ma
01:28:23spugokay, thanks
01:28:36 Quit Rondom ("Ex-Chat")
01:42:24 Quit lee-qid (Read error: 60 (Operation timed out))
01:43:20 Quit midkay_ ("*poff*")
01:51:32 Quit anathema (Read error: 145 (Connection timed out))
01:55:47 Quit Psiuyo (Read error: 104 (Connection reset by peer))
01:57:23 Quit EnterUserName (SendQ exceeded)
01:59:06 Nick Martian67_ is now known as Martian67 (
01:59:41 Quit JoeBorn (Read error: 145 (Connection timed out))
02:03:15 Quit ender` (" The problem with political jokes is they get elected. -- Henry Cate, VII")
02:08:38 Quit obo ("bye")
02:12:08 Join JoeBorn [0] (n=rootmeis@
02:14:26 Quit Arathis ("[rl_bot quit]")
02:20:04 Quit Strath (Read error: 104 (Connection reset by peer))
02:21:07 Quit Bjoern-Erik ("Lost terminal")
02:24:40 Quit dan_a (Read error: 110 (Connection timed out))
02:30:59spugonce i've flashed the firmware on the H300, i can safely delete the H300.hex file?
02:31:31spugok, thanks
02:36:32 Join JoeyBorn [0] (n=rootmeis@
02:39:07 Quit JoeBorn (Nick collision from services.)
02:39:16 Nick JoeyBorn is now known as JoeBorn (n=rootmeis@
02:54:25JdGordonjhMikeS: you around?
02:59:29 Join Id2ndR [0] (
03:00:41 Quit JoeBorn (Read error: 110 (Connection timed out))
03:02:09***Saving seen data "./dancer.seen"
03:06:20 Quit nudelyn2 ("At Argon, we're working to keep your money.")
03:08:13jhMikeSJdGordon: here and there
03:08:34JdGordondo the recording menus and stuff need fixxing after your big commit eariler?
03:09:28jhMikeSwhich of the big commits
03:09:43JdGordonthe recent recording overhauls
03:09:57JdGordonas in.. since sepetember
03:10:33jhMikeSthey shouldn't, why do you ask?
03:11:00JdGordonbecause im working on trying to get the settings and menu stuff commited this week and checking if that will need fixing
03:11:37jhMikeSoh you mean re: the settings changes. I think they should be updated to use real values at the least
03:11:43spugspeaking of recording settings:
03:12:52jhMikeSbasically means getting rid of global_to_encoder_config and encoder_to_global_config stuff
03:14:25jhMikeSI agree with that. Should be able to specify/browse to a directory and have it stick. I'd rather use the same dirs for both firmwares as well.
03:14:55 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)
03:15:14JdGordonyeah, i almost never record, but i thinks ts silly not being allowed to choose the folder
03:15:16Martian67do the general rockbox builds work on 5.5gen?
03:15:26Martian67as long as i have the bootloader properly installed?
03:15:36JdGordon5.5g is not ready yet
03:15:44rasherJdGordon: Well, get cracking on file-config then
03:15:52Martian67i had it working, i just wanted a newer rockbox build
03:16:04rasherIt's the only reason for the current method
03:16:16spugif i were better at programming, i'd make a patch :/
03:16:51rasherspug: It's quite involved. Unless you just want to add "RECORD"
03:16:55rasherAs a choice
03:17:09JdGordonrasher: well, it could be done now, just using the vkeyboard to type the name..
03:17:18rasherJdGordon: But how would you save the setting?
03:17:33spugrasher: that'd probably suffice
03:17:36Martian67JdGordon: if there is a working 5.5 patchset out there, why isnt it made offical?
03:17:46Martian67or at least avalible without hunting for 5 hours
03:17:48JdGordonrasher: same way as the wps/fonts...
03:18:15rasherSo there'd be an artificial limit on the path
03:18:16JdGordonMartian67: iirc its not working 100% yet so thee is no point
03:18:29Martian67JdGordon: whats not working about it, besides the apple firmware?
03:18:33rasher(and how much space is available in the config block anyway?)
03:18:43JdGordonnot much..
03:18:49JdGordonMartian67: i dont know
03:19:23JdGordonwould anyone like to try to settings patch now and help find missing settings and other fun bugs?
03:19:24Martian67JdGordon: because it works great for me :/
03:19:33JdGordon30gb model?
03:19:48JdGordonyeah, the 60gb model has bigger problems
03:19:53Martian6780 you mean?
03:20:38jhMikeSJdGordon: instead of converting the configs, they just need to be copied to the local instance and then the changed one copied back to global. The array in enc_config.c will need a size instead of a function pointer.
03:21:13Martian67JdGordon: what was that thing that supported the ipod style directorys by indexing the files?
03:21:23Martian67and do i need to rebuild to get it?
03:21:35JdGordontagcache and yes
03:22:41Martian67JdGordon: do you know if the 5.5 patchset works against those sources?
03:26:04Martian67JdGordon: do the daily builds have tagcatche or is it an external thing?
03:30:15scorchethey do
03:34:51 Quit |Rincewind| ("Cya")
03:35:57JdGordon:'( why is it soo damn humid here??
03:39:38 Quit spifwand (Read error: 104 (Connection reset by peer))
03:42:03 Quit Id2ndR ("Parti")
03:43:01 Join TeaSea [0] (n=thunderc@
03:43:23 Quit Criamos ("( :: NoNameScript 4.03 :: )")
03:45:16 Quit matsl ("Leaving")
03:51:14 Quit spiorf (Remote closed the connection)
04:00:53 Quit blue_lizard (Read error: 60 (Operation timed out))
04:00:57 Join blue_lizard [0] (
04:08:13Stroggjust sorta familiarizing myself with the plugin api.. wondering what's the standard way to output variables in text to the screen? It looks like you have to first sprintf() to a buffer and then lcd_putsxy(). Is that right or is there a better way?
04:13:16JdGordonStrogg: it would be better to use the screens[]. struct so it will work on the remote lcd screen... but yes,lcd_put* works
04:13:55StroggJdGordon: that was going to be one of my other questions.. What's a remote LCD?
04:14:21JdGordonthe x5, h100 and h300 models have remote controls which have an lcd screen on them
04:15:09JdGordonif your doing something purely text then you should have the lcd drawing function accomodate different lcd screens and fonts.. if yuor doing a game then dont worry about the rmeote lcd
04:15:19JdGordonunless you can be bothered of course
04:15:28 Join RoC_MM [0] (
04:16:06StroggJdGordon: I'm making something to view Go game files.. SGF files.. so.. it'll be game-like.
04:16:48StroggI'll try to keep things mostly to standard, but mostly first, I wanna get something working. Then can work on making it work everywhere. hehe
04:32:41 Join JoeBorn [0] (n=rootmeis@
04:42:08 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:42:19 Join miepchen^schlaf [0] (
04:42:20 Quit |AhIoRoS| ("Abandonando, see you")
04:58:27 Quit TeaSea (Remote closed the connection)
05:02:11***Saving seen data "./dancer.seen"
05:11:58 Join Rob2222 [0] (
05:14:48 Quit yipe (Remote closed the connection)
05:20:14 Quit Rob222241 (Read error: 145 (Connection timed out))
05:27:15 Join j1mc [0] (
05:28:17 Part maquis
05:28:23 Quit j1mc (Remote closed the connection)
05:28:26 Quit NickDe (Read error: 60 (Operation timed out))
05:39:07 Quit newbyx86 (Remote closed the connection)
05:39:47 Join newbyx86 [0] (
05:44:57 Join skipinschool [0] (
05:45:37*Strogg notices that his C is direly rusty.
05:45:48skipinschooli need an update with rockbox.... I had it like 3 months ago
05:45:58skipinschoolwhat all has improved since
05:46:12*Strogg knows nothing
05:47:03scorchebah...beat me to it
05:47:13*JdGordon rulezz :D
05:47:17skipinschooljust have like 2 questions
05:47:32skipinschoolalbum art fix?? can it play videos??
05:47:41JdGordonno and yes
05:47:57skipinschoolno to album art.... :-(
05:48:06StroggJdGordon: rockbox can do video now?
05:48:17*Strogg runs off to the feature list on the wiki
05:48:25JdGordonwell only a m2v files and only the video
05:48:31scorcheactually, run off to PluginMpegplayer
05:49:42JdGordonI have a ncie idea for the int settings screen, but scared the drawing code needed is going to be too big for it to work
05:49:52JdGordonI want to copy the speedo in cars..
05:53:06JdGordonanyone remember the formula to draw a circle?
05:53:12Strogghrmm poop..
05:53:13JdGordonactually a elipse
05:53:23skipinschoolits been awhile since ive used rockbox and wanna get it agian.... but i cant remember if i have to cange the song names
05:53:35StroggJdGordon: hrmm x^2 + y^2 = hrmm... something.
05:53:39skipinschoolfrom JKFJ.mp4 to name.mp4
05:53:44 Quit newbyx86 ("afk")
05:53:47StroggI wanna say r^2
05:54:21 Join newbyx86 [0] (
05:55:26*Strogg tries to figure out why his struct is full of garbage
05:57:27skipinschooldo i have to change my song names in my f folders on my ipod using godfather??
05:57:34Stroggthere are tools out there that will rename files based on the id3 tags
05:57:50scorcheand you can use tagcache
05:57:59skipinschoolso basically i just need to update my tags on iTunes
05:58:20skipinschooland then proceed with those other 2 steps you listed
05:58:26StroggWhy do you wanna keep the same ipod naming scheme?
05:58:46scorchewhy do you not have your tags correct?
05:58:53skipinschooli want it to display the name and title of the artist
05:58:57skipinschooland album art
05:59:19skipinschoolso... do not update my tages on iTunes?
06:00:48skipinschoolwhatever... ill figure it out
06:01:12Stroggcan't help yah on this one. I installed rockbox to get away from all the itunes stuff..
06:05:19skipinschoolwell, ive had rockbox... ive had senabs build, but i took it off because it kept freezing
06:05:23 Quit jaebird (Read error: 60 (Operation timed out))
06:05:35skipinschoolit wasnt made for dummies yet :-)
06:05:47skipinschooland i still think its not
06:06:24Stroggdepends what you wanna get out of it, I guess..
06:06:45 Join jaebird [0] (
06:08:40skipinschoolsee, theres too much access to critical stuff that could freeze it and stuff...
06:09:06scorcheoh wow...
06:09:09skipinschoolthats what i dont like.... it acts more like a harddrive rather than an Operating System
06:09:17*Strogg raises an eyebrow
06:09:48skipinschooldont get me wrong... i like the fact that you can do this to your ipod
06:10:06scorcheignoring your other comments, what ipod do you have?
06:10:07 Quit thegeek (Remote closed the connection)
06:10:08skipinschoolbut do you guys think it will ever develope into just a mere OS
06:10:10Stroggnah.. I'm just trying to figure out what you mean by harddrive-vs-OS
06:10:21skipinschooliPod video
06:10:34scorcheand by freeze you mean skip?
06:10:35Stroggthat's the same model I got. hehe
06:10:52skipinschoolwell, what i mean by hard drive...
06:11:04skipinschoolis you have access to stuff other than you need
06:11:19skipinschoolan OS is almost kinda like for dummies
06:11:25Stroggyou didn't you didn't have access to it before?
06:11:32skipinschoollike what iPod has right now ;-)
06:11:32Stroggerr.. *tries again*
06:11:37Stroggyou think you didn't have access to it before?
06:11:56skipinschooli know i had access to it before
06:12:01skipinschoollol.... now im confused
06:12:08skipinschoolwhere is this going??
06:12:08scorcheyou are the first person i have ever heard that would like less features
06:12:26skipinschoolno... features are GREAT
06:12:32Stroggskipinschool: are you talking about when it's plugged into USB, or when you're nav'ing around in ipod?
06:12:42skipinschoolbut access to critical stuff is like :-/
06:12:59skipinschoolwhen your just naving
06:13:12skipinschoolNOT pluged into usb
06:13:16Stroggso? Just go into your settings and turn that stuff off.
06:13:26skipinschoolno shit?
06:13:35*Strogg finds the option
06:13:37skipinschooli didnt know you can do that... lol
06:13:52scorchebe specific...i have no idea what you mean
06:14:11skipinschoolim a simple yet somewhat organized person
06:14:14StroggGeneral Settings -> File View -> Show Files -> Music
06:14:22scorcheoh...just that?
06:14:23StroggThen it shows only music.
06:14:33skipinschooli like the original iPod firmware, but its weak
06:14:45skipinschoolyou cant change the backround
06:15:13Strogghehe.. myself, I got nothing against the original ipod interface. It was seriously slick.
06:15:44skipinschoolyea... it is.... but you cant change the backround or anything like that
06:15:49skipinschoolnothing too confusing
06:16:06skipinschoolP.S. whats Senabs build website
06:16:14skipinschoolor is there a better suggested one
06:16:25skipinschoolfor the 5th gen 30 gig
06:16:56scorchesenab has stopped producing his buolds
06:17:13skipinschoolok... thats news...
06:17:26skipinschoolany alternatives?
06:17:53 Join EnterUserName [0] (n=dave@pdpc/supporter/student/GeekZoid)
06:25:44 Quit RoC_MM (Read error: 110 (Connection timed out))
06:39:49skipinschooluhm.... i think i fucked my iPod up, royally
06:40:05skipinschoolit will not load any firmware... not even the original
06:40:23scorcheand what does it do?
06:40:39skipinschoolrockbox boot loader
06:40:47skipinschoolthe all that other junk
06:40:56scorchethat is not junk
06:41:00skipinschoolloading error-
06:41:01scorchethat is what is telling you what is wrong
06:41:10skipinschoolloading original firmare
06:41:19skipinschooland then back to rockbox thingy
06:41:20scorcheyes...the whole line please
06:42:13scorcheas in there should be a number after error...what is it?
06:42:43scorchedo you have the rockbox.ipod and the .rockbox folder in the root of the device?
06:43:20skipinschooli didnt do that
06:43:28skipinschooli thought the firmware extracted that
06:43:42scorchethe instructions tell you to do that...
06:44:04skipinschoolyes, i relized that afterwords
06:45:47skipinschoolwell, shit.... my computer cannot even read it as a hard drive
06:46:03scorcheselect + play
06:46:14scorchealso in the instructions in case you screw up
06:46:21scorcheplease read =)
06:47:05skipinschoolselect and play didnt do anything... select and menu just restart it, but thats no help
06:47:19 Join t0dk0n [0] (
06:47:23scorchehold select and play...
06:47:41t0dk0nhmmm, there's no latest daily build on rockbox site
06:48:00t0dk0nhaha, I just take it that nothings been done for my player :P
06:48:13scorcheno...just a small glitch
06:48:19scorchenothing wrong with CVS builds
06:48:21t0dk0nooh, thought that as well
06:48:24skipinschooli held select and play.... and didnt do anything :-(
06:48:27t0dk0nYeah, I'll just update with that then
06:48:54t0dk0nI'm tired of that data abort error I kept getting from three days ago
06:49:39skipinschoolwell, shit.... i fucked up my ipods firmware... my mistake
06:50:02scorcheno you didnt (if you followed the directions)
06:50:08scorcheselect and play is hardware
06:50:18scorchereboot, then hold those 2
06:50:59*scorche sighs
06:51:02skipinschooli didnt reboot first
06:52:10skipinschooltank you sooo much!! i thought my iPod was done
06:52:31scorchethere hasnt been a single brick from rockbox...
06:52:51t0dk0nunless its a flaw in the user, or no?
06:53:03*JdGordon edited the ipod install page so extracting is before installing the bootloader
06:53:10scorcheipods are practically impossibly to brak permanently
06:53:23t0dk0nI meant in the booting part, and they hadn't backed up the original to start over in case they messed up
06:53:36scorcheJdGordon: it really doesnt matter...they are still plugged in
06:53:38t0dk0nor is that fixable as well?
06:53:44scorcheit is
06:53:55JdGordonscorche: no, the bootloader instructions tell you to disconnect and test!
06:54:01scorchethey do?
06:54:30*scorche shakes his fist at... <looks for a manual dev>
06:54:40t0dk0nhmm, on the cvs build part of the site, does green mean its been modified for that player?
06:54:51scorchegreen means no build errors
06:55:02t0dk0nthen what do the numbers mean?
06:55:09scorchenumber of errors
06:55:14t0dk0nah, alright
06:55:16t0dk0nthanks :)
06:55:33skipinschoolyeay!!! rockbox loaded
06:56:13t0dk0nhmm, when I update rockbox, unzipping it, does it overwrite my themes, backgrounds, and wps?
06:56:25JdGordononly if you want it to
06:56:44scorcheit will only add files and replace ones that have the same name
06:57:07t0dk0nhehe thats what I was thinking, but sometimes, it could just replace based on dir, rather then file
06:57:08scorcheif you made a theme and named it boxes.wps, then yes ;)
06:57:53JdGordonthe only files u need to update is rockbox.ipod, .rockbox/viewers/* and .rockbox/rocks/*
06:58:00JdGordonand .rockbox/codecs/*
06:58:35 Quit rotator ("zzzzzzzzzz")
07:00:59JdGordonok, who wants to play guenea pig?
07:01:21t0dk0nas in the Japanese snuff series? :P
07:01:29*JdGordon needs beta testers
07:01:37t0dk0nbeta testing for what?
07:01:40scorcheuh oh
07:01:43JdGordona patch for rockbox
07:01:44*scorche hides
07:01:52JdGordonyou know you ant to.. :D
07:01:55t0dk0nwhats this 'patch' do
07:02:01scorchewell, i usually do =P
07:02:14scorches/do/end up doing it anyway
07:02:17***Saving seen data "./dancer.seen"
07:02:31*Strogg commits his src tree before applying mysterious patchs. hehe
07:03:31StroggJdGordon: so what's this patch?
07:03:34JdGordon patch at the bottom..
07:03:42JdGordonreplaces all the settings and most menus
07:03:48JdGordonno more config block settings
07:03:56scorchedid you fix the boosting?
07:04:12JdGordonyeah :)
07:04:18scorchefigure it out, or just the recoding of the recoding did it?
07:04:24JdGordonhopefully gonna get it to a state where its ready for commit early this week
07:04:41JdGordonno, it was the dynamic menu stuff.. the sound menu wasnt plaing ncie
07:04:44JdGordonall fixed now tho
07:05:04scorcheoh...just the dynamic
07:05:13scorcheoh wait...i already heard you talking about that..
07:05:50scorchet0dk0n: stop laughing at me!
07:06:01t0dk0nsorry :'(
07:06:09t0dk0nI'm laughing with you :P
07:06:40scorcheof course...
07:07:28t0dk0neven if you're not laughing
07:08:52 Join RoC_MM [0] (
07:09:27t0dk0nhmm, if you made a playlist with files from the ipod, would it work with rockbox?
07:09:28 Quit RoC_MM (Client Quit)
07:09:44t0dk0nlike in a program like XMMS or exaile
07:09:51JdGordonyes.. if its a .m3u
07:09:53scorcheif it is m3u, yes
07:09:58t0dk0nno pls?
07:10:06scorchejust for that, i am not testing your patch!
07:10:14JdGordonim sorry
07:10:25 Join skipschoolandksk [0] (
07:10:34Strogghrmmmm my patch can't read your patch. hehe
07:10:39skipschoolandkskok... got my rockbox on :-)
07:10:54skipschoolandkskbut it aborts.... lol
07:11:14PaulPositionI wish I could help testing, Jonathan, but there's no recording *yet* on the portalplayer daps.. Wish you good luck with it, though.
07:11:21t0dk0n"data abort 'hex decimal numbers'"?
07:11:34JdGordonPaulPosition: no, this patch has nothing to do with recording
07:11:35scorchePaulPosition: recoding...not recording
07:11:44PaulPositionDoh. lol
07:11:58JdGordont0dk0n: what build and whats the hex number?
07:12:18skipschoolandkskjulis build
07:12:19t0dk0nooh, I was asking skipschoolandksk dude
07:12:27skipschoolandkskand its while im playing a song
07:12:36StroggJdGordon: How did you generate this patch?
07:12:42JdGordoncvs diff
07:12:42PaulPositionI thought it was about recording settings.. lol. −− Soon as I wake up tomorrow morning I download and build (and then read some, so I have at least an idea what to look for in terms of bugs and such..) :)
07:12:45t0dk0nI just put the newest cvs on mine, but it runs fine so far
07:12:45scorcheuhh...he wrote it? =P
07:12:59StroggJdGordon: should I be able to apply via patch -p2 < foo.patch ?
07:13:15scorchewell, depends where you are doing it
07:13:20skipschoolandkskhey... whatever... ill figure it out tommorrow
07:13:32PaulPositionskipshcool - hey
07:13:45JdGordonStrogg: crazy?
07:14:00scorcheskipschoolandksk: we request that you try with an official build before saying something is wrong
07:14:10StroggJdGordon: insane.. bizzarre..
07:14:23JdGordonwhat in particular is?
07:14:30PaulPositionWhen's the last Pi Julius build from?.. If it's between 14th and somewhere around 16th or 17th, the freeze might be caused by you using APE tagged files.
07:14:35Stroggit applied ok with -p0. hehe
07:14:47PaulPositionThere was some changes made and the latest builds resolve that.
07:14:52skipschoolandkskwell, for right now... where can i get the original ipod firmware so i dont have to restore
07:15:12t0dk0njust hold menu down after rebooting
07:15:16scorcheyou should have saved it
07:15:19t0dk0nand it goes into the origingal firmware
07:15:23t0dk0nor that
07:15:45skipschoolandkskuhm... i didnt save anything else besides the rockbox stuff
07:16:04scorchei *do* know that the instructions tell you to save that file
07:16:06t0dk0nDid you back up that bootpartition thing?
07:16:34skipschoolandkski dont know... i followed the directions
07:16:35scorcheyou dont need to back it is already just need to not delete it ;)
07:16:51skipschoolandkskthe only thing i forgot was the rockbox folder, but we resolved that
07:17:09skipschoolandkskok... well, i typed this... hold on
07:17:26scorcheYou should keep a safe backup of this bootpartition.bin file for use b if you ever wish to either upgrade the Rockbox boot loader or uninstall Rockbox
07:17:27scorchefrom your Ipod
07:17:46scorcheand you didnt follow that ;)
07:18:00skipschoolandkskipod_fw -o apple_os.bin -e 0 bootpartition.bin
07:18:08skipschoolandkskipod_fw -o apple_sw_5g_rcsc.bin -e 1 bootpartition.bin
07:18:17scorchethis is before that
07:18:24scorchewhen you first extract it
07:18:34skipschoolandkskipod_fw -g video -o rockboot.bin -i apple_os.bin bootloader-video.bin
07:18:49skipschoolandkskwell than no
07:18:55skipschoolandkski guess there merged or something
07:19:23scorchewe already know that...i was just pointing out that you did, in fact, not follow all the directions as you said ;)
07:19:32Strogghrmmm.. having a lot of issues compiling..
07:19:43scorcheStrogg: like?
07:19:46StroggJdGordon: I suspect that your build environment differs alot from mine..
07:19:58Stroggscorche: mostly path issues.. gimme a bit.. I'll get it working
07:20:03skipschoolandkskshit... im retarted....
07:20:12skipschoolandkski have 3 other iPods next to me
07:20:14JdGordonStrogg: ?
07:20:18scorcheStrogg: well, have you compiled rockbox before?
07:20:29skipschoolandkska long time ago
07:20:30Stroggscorche: yeah
07:20:50skipschoolandkskcan i pull the firmware from an 80gig??
07:20:58skipschoolandkskwould it work on the 30
07:21:18PaulPositionCould it be the tools/configure changes hitting you back in the face..? (see latest cvs..)
07:21:38skipschoolandkskwell, guys.... thanks for the help, im tired as hell, and ill deal with this tommorrow
07:21:46PaulPositionnite, then.
07:21:52skipschoolandksktake it easy...
07:21:57 Quit skipschoolandksk ()
07:22:15StroggThis is unrelated.. but.. should these be brackets instead of braces?
07:22:16Stroggexport DOCSDIR=$(ROOTDIR)/docs
07:22:16Stroggexport MANUALDIR=${ROOTDIR}/manual
07:22:44 Quit skipinschool (Connection timed out)
07:22:59*JdGordon no idea
07:23:06t0dk0ncrossfeed is sex for the ears
07:24:09 Quit PaulPosition ()
07:24:33 Join [sellout] [0] (
07:27:36StroggAnyone know offhand what error 127 out of make is?
07:33:41 Join webguest97 [0] (i=50e8759a@gateway/web/cgi-irc/
07:33:49 Quit webguest97 (Client Quit)
07:35:27JdGordonhmm... shutdown is happening too fast now :'( settings arnt getting saved
07:35:50t0dk0nhehe that needs to be fix0red
07:36:01JdGordonno, it needs to be fixed...
07:36:25t0dk0nhahaha, fine fine, 'fixed' it needs to be
07:36:39Stroggheh. There we go. I'm gonna have to add my all toolchain dirs to my PATH.. or make a script to do it.
07:36:50Stroggit was going nuts when it couldn't find arm-elf-ranlib
07:36:54t0dk0njdGordon: is that on your patch, or on the general cvs?
07:37:00JdGordonwith patch
07:37:04t0dk0nahh alright =D
07:37:13t0dk0nlol, good thing I didn't try it quite yet :P
07:38:41t0dk0nI don't wanna use my computers audio anymore, I'm addicted to the rockbox auido quality!
07:39:24scorcheyour computer should have better...
07:39:27t0dk0nit should
07:39:37scorchewhat audio card?
07:39:40t0dk0nbut I have great sennheiser headphones
07:39:57scorchejust plug the headphones into the computer ;)
07:40:01scorcheor get better speakers
07:40:03t0dk0nhaha yeah
07:40:13t0dk0nI have crappy generic harman/kardon speakers
07:41:24t0dk0nhmm, I'm testing to see how long I can just listen to audio before the battery runs out
07:43:02 Join sellout_ [0] (
07:50:17 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
07:50:37 Join miepchen^schlaf [0] (
07:50:54 Quit [sellout] (Read error: 113 (No route to host))
07:52:12 Quit sellout_ (Read error: 104 (Connection reset by peer))
08:02:25*Strogg gives up for tonight
08:05:07t0dk0nnight night
08:09:23t0dk0nhmm, the text is inside the image of this wps
08:09:55t0dk0nI wanna make it a little more aligned
08:10:11t0dk0nis that possible? or is it just straight up left, center, and right?
08:11:31t0dk0nthe images of the theme show it aligned correctly 0_o
08:13:35 Join midkay [0] (n=midkay@rockbox/developer/midkay)
08:14:22Stroggmaybe the positioning is font dependent?
08:15:03t0dk0nbut check it out
08:15:43t0dk0non my ipod, the text is crashing into the play part a little
08:17:38t0dk0nI was trying to look at the wps file, with the aid of the tag index in the manual
08:17:47t0dk0nbut couldn't figure out exactly what to do
08:47:53 Join Gnelik [0] (n=Miranda@
09:02:19***Saving seen data "./dancer.seen"
09:08:11GnelikGr. My sansa dont want to initilize lcd
09:18:50 Quit hannesd (Read error: 145 (Connection timed out))
09:20:39 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
09:21:21 Join miepchen^schlaf [0] (
09:23:21GnelikAny ideas why lcd could be initilezed in bootloader but not in normal firmware
09:25:10jhMikeSGnelik: Not totally sure but I think Rolo might need it reset at least. Other than that, no reason really.
09:26:07jhMikeSBut re: Sansa can't really say for sure. More familiar with Coldfire stuff atm.
09:26:11GnelikjhMikeS:(1) What is Rolo? (2) Screen becam yellow after init
09:26:50jhMikeSGnelik: Rolo restarts rockbox. You can invok it by selecting the firmware file in the browser and running it.
09:28:13jhMikeSGnelik: (2) Have no idea why. I don't have a Sansa. Don't know if anything I said was of any assitance :P
09:28:25 Join Blackfish [0] (
09:28:27scorchejhMikeS: not quite right...RoLo is a method of loading other firmwares
09:29:09GnelikI generate mi4 file and
09:29:13jhMikeSscorche: I've only seen it when updating the firmware or selecting another firmware (almost never). so yeah :)
09:29:22GnelikSandisk bootloader loads it
09:29:26scorchewell, what device?
09:29:46scorchethere is a bug on ipod where it will only load the same image that is currently running...
09:29:58BagderGnelik: well, Rockbox on sansa isn't even working so expect more flaws to remain
09:30:42GnelikBagder: But i cant understand why it dont wand to init lcd, we cant start debuging
09:30:59scorcheBagder: since when do you come in here this early on a sunday? =)
09:31:11Bagderscorche: ssshhh don't tell anyone ;-)
09:31:29Bagdermy youngest is sleeping and wife and daughter are out
09:32:01scorchei have to i an enabler?
09:32:28Bagderen enabler of what?
09:32:41jhMikeSenabler...haven't heard that term in a while
09:33:16scorcheone that enables another to achieve an end; especially : one who enables another to persist in self-destructive behavior (as substance abuse) by providing excuses or by helping that individual avoid the consequences of such behavior
09:34:10jhMikeSrockbox makes me forget to pay bills
09:35:33GnelikBagder: Can this be connected with memory remapings etc.?
09:35:58BagderI wouldn't guess so, but who knows
09:36:49 Join webguest00 [0] (i=5873b21c@gateway/web/cgi-irc/
09:38:46hachiwhy isn't there a link to the version controlled source on the website?
09:38:51hachior am I just missing it?
09:39:12Bagderlink to it how?
09:39:26Bagderhas two suitable links in the bottom
09:40:19hachiahh... I don't usually look in the instructions to find the urls
09:42:38 Join shnee [0] (
09:43:31 Quit ObsidianX (Read error: 104 (Connection reset by peer))
09:53:22 Join [sellout] [0] (
09:53:44 Quit shnee (Read error: 104 (Connection reset by peer))
09:57:47 Join shnee [0] (
10:03:59 Join shnee_ [0] (
10:06:44JdGordondoes ata_spin() force the disk to spin up if it is off?
10:07:40 Join My_Sic [0] (
10:07:58 Quit secleinteer (Remote closed the connection)
10:10:14JdGordon.. it doesnt :(
10:10:27JdGordonwhats the best way to force the disk to spin without reading.writing anything?
10:16:54 Quit shnee (Read error: 110 (Connection timed out))
10:19:15 Join Siku [0] (
10:19:32 Part t0dk0n
10:29:46jhMikeSJdGordon: I already inquired about that. None.
10:30:11JdGordonI guess Ill have to explicitly call the save setting function on shutdown
10:33:33 Join matsl [0] (
10:34:41 Quit Gnelik (Read error: 110 (Connection timed out))
10:37:29 Join Gnelik [0] (n=Miranda@
10:39:05GnelikI hate my sansa :)
10:39:45 Join thegeek [0] (
10:42:18GnelikAnd yellow color too.
10:43:21JdGordonis rectrigger on all targets now?
10:44:59JdGordonjhMikeS: ?
10:52:29 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
10:52:43 Join miepchen^schlaf [0] (
10:59:09 Join merbanan [0] (
10:59:18 Quit Gnelik ("Miranda IM! Smaller, Faster, Easier.")
11:01:25 Join Gnelik [0] (n=Miranda@
11:02:04 Join GreyFoux [0] (
11:02:12 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:02:24***Saving seen data "./dancer.seen"
11:05:39GnelikAny ideas of sansa lcd_init bug
11:06:13 Join ender` [0] (n=ender@
11:06:55 Join Rondom [0] (
11:09:34jhMikeSJdGordon: sorry. yes?
11:09:45JdGordonis rec trigger on all tragets?
11:09:45 Quit Gnelik (Connection reset by peer)
11:09:59JdGordonand the recording settings are going to be a pita to move across :p
11:10:04jhMikeSoh...rectrigger. can't say.
11:10:36jhMikeSwhy are the recording settings going to be a pita?
11:11:04JdGordonjust because.. :p
11:11:14jhMikeSwere most of them already moved?
11:11:44JdGordonim not sure... I think only rectrigger (if thats swcodec) and frequency ned fixing
11:11:54 Join wehn [0] (
11:15:16jhMikeSshould properly #ifdef the global_settings structure while at it since a lot of stuff gets included that doesn't belong
11:18:26 Join Bjoern-Erik [0] (
11:18:51 Join dan_a [0] (n=dan-mirc@
11:19:11jhMikeSI don't see any problem with the frequencies atm on SWCODEC
11:21:25 Join spiorf [0] (
11:21:41jhMikeSIs there a problem putting those samplerate macros inside the others?
11:24:13 Join nudelyn [0] (
11:26:42 Quit Rondom ("Ex-Chat")
11:31:34 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
11:34:09 Quit amiconn (" HydraIRC -> <- Try something fresh")
11:38:38 Quit rconan ("Leaving")
11:38:42 Join rconan [0] (
11:44:25 Join n1s [0] (
11:47:29 Join lee-qid [0] (
11:55:31LearAnyone who knows how "aggressive" the dithering is on grayscale targets?
11:58:19jhMikeSIt should dither each 8 bit level to a unique dot pattern.
11:58:24 Join bernych [0] (i=50533e9c@gateway/web/cgi-irc/
11:58:47LearOn perfectly white (#FFFFFF) WPS backdrop, I get a light gray pixel every 16 pixels...
11:59:23jhMikeSSure the source image is perfect white?
11:59:33jhMikeSFEFEFE should produce that
12:00:37bernychHi guys, I have a X5 running rockbox, and I'm no happy with the radio reception quality, so I look into the code...
12:01:04jhMikeSbernych: It does seem a bit weak but I never looked myself. Find anything?
12:01:16LearQuite sure. It seems that the colors I picked (so that the top two pixels are right), aren't quite what the BMP loader expects, so it dithers.
12:01:25LearQuestion is, what does it expect... :)
12:02:34LearOuch, tried a slightly darker background, and that made things a lot worse...
12:02:48jhMikeSIt expects a bitmap. :P If it has to reduce color depth, it dithers levels that aren't representable by solid colors. White sould be solid though.
12:03:09bernychI don't know why , but amiconn added a 4kHz shift to the frequency selection and High Cut-Control is not active. Someone knows why?
12:03:30LearBut it isn't...
12:03:54 Join MadDog011 [0] (
12:04:01 Join TeaSea [0] (n=thunderc@
12:04:03jhMikeSbenych:hmmm...I have no clue why the shift is needed. I suspect the setup on iAudio and iRiver should be the same.
12:04:21 Quit MadDog011 (Remote closed the connection)
12:04:32jhMikeSLear: It should be. Ask amiconn. He wrote the new loader.
12:04:43LearI know, but he isn't around...
12:05:04 Quit thegeek (Read error: 104 (Connection reset by peer))
12:05:06 Join thegeek_ [0] (
12:05:31bernychFor me, without the frequency shift works better, and it is exactly what the philips datasheet says
12:06:32 Join bluey- [0] (
12:10:58jhMikeSLear: He said it's "Code Policed". I see the problem though
12:11:27bernychDoes someone has a decompiled listing of the original iaudio firmware, so that I can check what cowon did?
12:11:31LearYou do? Just started looking at it...
12:11:36jhMikeSbernych: could be. there might be a reason on iRiver and the reception is strong there. x5 should probably not use it.
12:12:17jhMikeSLear: need to do some minute
12:14:18 Join cobold [0] (
12:14:38jhMikeS(257 * (3 * brightness(*qp++) + delta)) >> 16. when brightness is 255 and delta is 0, the result is 2.999954224... which rounds to 2. That happens on 1 dot out of 256.
12:15:46LearInteresting... Not the only case with dithering (pure white), but perhaps the other cases have the same rounding problem.
12:15:51jhMikeSOther values in the matrix are > 0 so all the other pixels round to level 3
12:15:53 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
12:16:40 Quit lini ("lini has no reason")
12:16:41LearMaybe the fix is to replace 0 with 1 in the matrix?
12:16:47jhMikeS257*255 = 65535 so it runs short
12:16:59LearOr maybe that's the simple but wrong fix. :)
12:17:15jhMikeSLear: Definitely not. The matrix should have all levels 0-255
12:18:05LearI'll give it a quick test anyway. Would be interesting to see the effects...
12:18:21jhMikeSit will "fix" it. better would be to add 1 to the delta
12:19:01LearYou mean it should read "+ delta + 1"?
12:19:11 Quit cobold (Remote closed the connection)
12:19:15 Join Criamos [0] (
12:20:18 Join thegeek [0] (
12:20:32jhMikeSbut black won't be pure black any more then...there will be dark gray dots
12:20:44 Quit thegeek_ (Read error: 104 (Connection reset by peer))
12:21:45jhMikeSI never liked that 257*x / 65536 thing
12:23:44 Join lini [0] (i=pugsley@
12:23:59LearDo you have a better solution?
12:24:19jhMikeSthinking about it
12:25:09jhMikeSmore precision perhaps
12:25:53LearNope, "+ delta + 1" didn't fix the white. Changing the matrix worked though.
12:26:18 Join MarcoPolo [0] (
12:26:33jhMikeSthe result must be 255.003891051*257
12:27:22LearThe dots became darker, even. Black, rather than light gray.
12:27:41jhMikeSchanging 0 to 1 in the matrix would increase the delta by 1. If I calc it that way the result is 3.003875732 which should round to level 3
12:29:34jhMikeSThat's cause when delta is 255, adding 1 causes wrap around to black.
12:30:41n1sJdGordon: here?
12:30:43jhMikeSthe result is 4.003860474
12:40:04n1sJdGordon: anyway if you read the logs I found two bugs in your patch and reported them on the wiki page.
12:40:41 Join Strath [0] (
12:40:43 Quit robfitz (Read error: 110 (Connection timed out))
12:44:35 Join robfitz [0] (
12:47:12 Quit thegeek (Read error: 104 (Connection reset by peer))
12:47:16 Join thegeek_ [0] (
12:52:06 Join amiconn [0] (
12:52:28 Join Gnelik [0] (n=Miranda@
12:53:00LearAh, Amiconn. Backdrop dithering doesn't work quite right on grayscale.
12:53:15amiconnIt does work perfectly here
12:53:24amiconnH140, Mini 2G, and the sims
12:53:30Gnelikdan_a: hi Dan!
12:53:45dan_aGood morning, Gnelik
12:54:17Gnelikdan_a I tryed to init lcd befour all, but it became yellow
12:54:36LearI tried with a backdrop that was largely white. One out of 256 pixels wasn't white.
12:54:45GnelikAnd it don't display text
12:55:05dan_aGnelik: I saw that in the logs. I don't know why that would be. I don't think it is the memory remapping, because that gets done in the bootloader too.
12:55:19LearWhen delta is zero (as jhmikes noticed), the bright value is rounded down to two.
12:55:31*amiconn just checked the log
12:55:49amiconnYes, there is a very slight deviation
12:56:03Gnelikdan_a:Again we must serch for thing is'nt done in bootloader
12:56:07LearThe value to the left of the ">> 16" is exactly 3 too low for proper result (if delta is zero at least).
12:56:16amiconnIt results from approximating x / 255 by (x * 257) / 65536
12:56:55Gnelikdan_a: i try to look for 0x7000***
12:57:14Gnelikbut it didnt helped me
12:57:42dan_aGnelik: For flashing the LED?
12:57:43amiconnHowever, doing it this way (1) avoids a (costly) division and (2) also removes the need for explicit clipping
12:58:18amiconnWill think a bit about it
12:58:26 Join MadDog011 [0] (
12:58:38 Quit MadDog011 (Remote closed the connection)
12:58:47Gnelikdan_a: Nope. I thought thai may cause lcd driver to work bad, when doing something with such adreses.
12:59:58 Quit lini ("lini has no reason")
13:02:28JdGordonn1s: back, ok thanks
13:02:29***Saving seen data "./dancer.seen"
13:05:13Gnelikdan_a: Have you any ideas?
13:06:09dan_aNot at the moment
13:07:20JdGordonn1s: thanks, both are fixed, ill up a new version shortly
13:08:03n1sgoing afk, bot will test more later.
13:08:46 Quit rconan (Remote closed the connection)
13:11:10 Join hannesd [0] (
13:12:29GnelikAfk for a hour
13:12:52 Quit Gnelik (Read error: 104 (Connection reset by peer))
13:13:38jhMikeSit can be proven that the appoximation cannot work with any scaling and produce both solid black and solid white. you can have a matrix max of 255 with a brightness of 0 (should be black) and a matrix min of 0 and a brightness max of 255 (must be white).
13:16:13amiconnThe approximation can work when using the real division, but then we need clipping well
13:16:30amiconnBah, forget that
13:17:09amiconnThen it's no longer an approximation...
13:18:41 Join joe7d6 [0] (n=joe7d6@
13:19:23 Join lini [0] (i=pugsley@
13:21:59LearI tried adding 3 to the expression to the left of the ">>", and then all 4 colors in my test backdrop loaded correctly, without visible dither.
13:23:13amiconnTried a solid black area?
13:23:58 Join hannesd_ [0] (
13:24:08 Join anathema [0] (
13:25:47LearSorry, no, I was lucky. Tried some big solid blocks, and all but the white one had that 1 in 256 bad pixel.
13:26:26jhMikeSthe result of that with matrix 255 and brightness 0 would be 1.000030518.
13:27:20amiconnBut I think I know how to improve the approximation without introducing division and clipping
13:27:28jhMikeScourse you could always add the values 0-3
13:29:11 Join pixelma [0] (i=54bd5f96@gateway/web/cgi-irc/
13:30:23jhMikeS(257*brightness + delta + brightness/64) / 65536 ??
13:30:41jhMikeSoops forgot some parentheses there
13:36:46 Quit wehn ()
13:38:45jhMikeSamiconn: Instead of "Code Policed" you could have just said "jhMikeS don't you dare f**k with this" :D
13:39:11amiconnI referred to things like CamelCase and the like
13:40:36jhMikeS*joking* ... CamelCase?? Where's that in the commit message?
13:41:04 Quit hannesd (Read error: 110 (Connection timed out))
13:41:04 Nick hannesd_ is now known as hannesd (
13:41:38 Part joe7d6 ("Leaving")
13:41:46jhMikeSoh...good thing for wikipedia :P
13:45:11jhMikeSbool LeadCamelToWater() { return true; } ... bool MakeHimDrink() { return false; }
13:45:53 Join NickDe [0] (
13:46:03amiconn , section "Names"
13:46:13amiconn"Variables and function names should be all lower case."
13:46:32 Join obo [0] (
13:46:44jhMikeSsomething I readilty observed by example
13:48:15 Join My_Sic [0] (
13:55:09 Quit Blackfish (Read error: 104 (Connection reset by peer))
13:55:58 Join Blackfish [0] (
13:58:59 Join funky [0] (
14:04:04 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
14:06:27 Quit bluey- ("Leaving")
14:08:30n1sJdGordon: I've got a "Not Done Yet!" in the Recording settings, the stuff I reported earlier works fine now with the new patch :-)
14:09:09JdGordonfor the reocrding settings.. im hopeing I can get someone to fix them all for me :D
14:10:35amiconnjhMikeS: I think I understand what's going on. Looks like your suggested fix is fairly good
14:10:57jhMikeSJdGordon: someone?
14:11:26 Quit matsl (Read error: 110 (Connection timed out))
14:11:40jhMikeSNever saw "Ferris Beuhler's Day Off"?
14:12:08jhMikeSoh...never mind then :p
14:12:55amiconnFerris Bueller's Day Off
14:13:01 Join rconan [0] (
14:13:02jhMikeSJdGordon: Which patch should I use?
14:13:13JdGordonthe latest on the wiki
14:13:16 Quit GreyFoux ("Le vrai danger, c'est quand les hommes penseront comme les ordinateurs")
14:13:45jhMikeSthink I need a separate checkout to do this...
14:14:23jhMikeSyou said trigger and sample rates right?
14:14:37JdGordoni tinhk so, im not really sure whats missing
14:17:26jhMikeSthere's esperanto as a
14:17:42JdGordonyeah, but no bork bork!
14:17:48jhMikeSlet's add Klingon
14:17:55JdGordonnerd! :D
14:19:26*amiconn is slightly confoozled :/
14:19:50amiconnAdding the correction doesn't influence code size on h1x0
14:20:30JdGordonthats good isnt it?
14:20:46amiconnIt's strange...
14:21:17jhMikeSIs the code size byte aligned or rounded up?
14:21:59JdGordonspeaking of bin size.. all the global static const vars go staright into the bin and not ram right?
14:22:04amiconnNot sure.. maybe the linker aligns sections to longwords or even lines
14:23:09JdGordonoh woops... its not consted.. bugger
14:23:17*JdGordon talking to self again... :p
14:24:04jhMikeSJdGordon: What's the p level on it?
14:24:56jhMikeSnever mind. was in the wrong dir :p
14:25:06JdGordonoh.. p0
14:25:27JdGordonamiconn: there is no semi-circle drawing routines in rb are there?
14:25:51amiconnWhy would you need that?
14:26:06JdGordonI had a nice idea for int settings, but adding the needed code would make it slow and bloated
14:26:15JdGordondraw it up like a speedo in the car
14:26:55amiconnThe int settings are fine as a list. Why would you want to give up simplicity and precision?
14:27:08JdGordoncoz its ugly and boring...
14:27:50 Join Rondom [0] (
14:27:56amiconnIt's not ugly at all imho
14:28:04jhMikeSAnalogue guages for settings?
14:28:49amiconnIf you could have a look at archos recorder retailos "int" settings (that _does_ use sliders) you would run away immediately
14:29:00JdGordonit would look cool, but it would be slow and bloated so I havnt even bothered coding it to test
14:29:53jhMikeSEvery setting should be from 0% to 100% so as not to confuse people ;)
14:31:25JdGordonspeaking of code bloat.. why is checkbox() in the code? its only used by plugins
14:31:53JdGordonactually, i dont tihnk its used ever
14:31:53amiconncheckbox was once introduced as an idea for boolean settings
14:33:15amiconn...but since _only_ boolean settings would look okay when put directly into the settings list, this idea was never completed
14:33:38JdGordonis it worth keeping the code?
14:33:45amiconnI think checkbox() should be removed
14:34:33jhMikeSNot sure why only boolean setting would look ok that way.
14:35:13 Quit Ribs ("eh eh ehhhh!")
14:35:15amiconnAll other settings use text of varying width
14:35:34amiconnThat looks like crap on smaller LCDs and/or with larger fonts
14:35:56JdGordonrecorder/wigets.c can be removed totally then
14:35:59jhMikeSI can think of a way to do it for anything at all
14:36:05amiconnThe recording trigger screen does that, and it's ugly as hell
14:36:11JdGordononce anything calling scrollbar() calls gui_scrollbar_draw()
14:40:37 Quit Strath ("User pushed the X - because it's Xtra, baby")
14:42:04 Quit pixelma ("CGI:IRC (Ping timeout)")
14:42:10 Join pixelma [0] (i=54bd5f96@gateway/web/cgi-irc/
14:42:14jhMikeSone advantage of sliders with a number is it gives you a sense of the range of a setting at one glance
14:42:37 Join Gnelik [0] (n=Miranda@
14:43:09GnelikI am back again
14:45:53Gnelikdan_a: are you doing something with sansa, if not i'l don't truble you?
14:47:48dan_aGnelik: At the moment I'm doing other work. I'll let you know when I look at the Sansa, and if I find anything
14:48:00dan_aFeel free to ask questions, though.
14:49:03markunjhMikeS: what, you don't speak esperanto?? ;)
14:51:27Gnelikdan_a: Where must i seach a bug?
14:52:52GnelikOr i can't find it at all?
14:56:49dan_aGnelik: I don't know at the moment. I don't think it will be in crt0-pp.S, because that is the same for all the PortalPlayer targets (apart fromt the problem I found yesterday)
14:57:12dan_aI'd guess that something is going wrong in lcd_device_init, but I don't know what.
14:59:08 Quit bernych ("CGI:IRC (Ping timeout)")
15:02:32***Saving seen data "./dancer.seen"
15:02:32JdGordonthe comment at the end of the plugin struct /* Keep these at the bottom till fully proven */, is that still wnated at the bottom?
15:02:40JdGordonall the recording stuff
15:03:08jhMikeSJdGordon: yes ... 'till preglow does a running DSP plugin
15:03:19JdGordonokey dokey
15:04:11jhMikeSmarkun: not a word. maybe could pick a few words out if esperanto has any Spanishiness
15:05:11jhMikeSLet's add Elvish too while we're at it
15:05:38JdGordonnext time im bored im adding "drunked yobbo aussie"
15:05:47JdGordoncomeplte with full voice pack
15:06:18jhMikeSEbonics too
15:06:27 Join Strath [0] (
15:08:19 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
15:10:29 Quit spiorf ("Read error: 104 (Connection reset by peer)")
15:10:35 Join spiorf [0] (
15:15:05Gnelikdan_a:One another question: when i turn off the LED, in init() method it don't evem makes screen whie
15:15:17 Part psiborg
15:15:22Gnelikwhie = white
15:16:10dan_aGnelik: The LED (on the scroll wheel) shouldn't affect the screen at all.
15:16:49GnelikBut it does
15:17:30GnelikIt only turns LED off and not clear the screen how it did without led controling.
15:18:28dan_aHow are you turning the LED off? Can you post the code on
15:20:49Gnelik volatile unsigned int* ptr;
15:20:49Gnelikptr = (volatile unsigned int*)0x70000020;
15:20:49Gnelik*ptr |= (1 << 13);
15:20:57Gnelik*ptr |= (1 << 13);
15:21:12Gnelik*ptr &= ~(1 << 13);
15:21:24Gnelikin diferent places
15:23:30dan_aThat has never done anything to the LCD screen on my Sansa
15:24:53GnelikI doesn't do anything
15:27:15GnelikIt doesn't clear lcd
15:28:21GnelikI am stupid
15:28:47Gnelik> /*lcd_init();*/ /*Turn it off for debug*/
15:30:54 Quit blue_lizard (Read error: 110 (Connection timed out))
15:31:22 Join blue_lizard [0] (
15:32:01JdGordonoh wtf?? i fixed these bloody warnings locally :'(
15:34:07amiconnjhMikeS: I finally found the correct dithering formula :D
15:34:37amiconnFor 4-grey it is: level = (3 * bright + (bright >> 6) + delta) >> 8
15:36:35 Join bernych [0] (i=50533e9c@gateway/web/cgi-irc/
15:39:10amiconnJdGordon: RED!
15:39:21JdGordonmr impatient :p
15:39:24JdGordonjust commited fix
15:39:43*JdGordon blames the find in files feature of kate for that
15:39:56JdGordonbloody typical that the buidls i tested worked :p
15:40:45oboJdGordon: kate has mangled a name in video.c
15:41:36Gnelikdan_a: it's to hard for me :(
15:41:49preglowyou guys need to stop using kate
15:41:55preglowit's obviously completely unsuitable as an editor
15:42:10JdGordonapart from tiny quirks its fine
15:42:24preglowtiny quirks like changing stuff you don't touch yourself
15:42:30obokscope does the same thing :(
15:42:44JdGordonyeah, apart from that
15:43:01JdGordonI thought i was doing so well to check for that before commiting tonight :'(
15:43:07bernychamiconn, I'm trying to fix the radio poor reception in the X5, can you please have a look to
15:43:35JdGordoncan someone fix that (c) message for me please? once its commited i cant..
15:44:16amiconnbernych: I wonder what shift by 4kHz you're talking about
15:45:17amiconnBesides, the X5 doesn't use the philips tuner chip
15:45:29amiconnIt uses DA-202
15:45:34bernychlooking at the CVS is you that added +16384 in the line that calculate the PLL counter
15:45:47amiconnYes, for philips
15:46:00bernychYes, but the registers are the same
15:46:01amiconnHowever, that's just for rounding
15:46:34amiconnThe probklem with the 32768Hz xtal is that you cannot tune exactly to a frequency
15:46:37 Quit JdGordon ("Konversation terminated!")
15:46:42bernychFor you this allowed a better reception?
15:47:34amiconnInteger arithmetics just round down, adding 16384 makes it round to the nearest frequency instead
15:48:36bernychand for the High cut controller, there is a reason why is not enabled?
15:49:19amiconnI don't know
15:50:26bernychI would like to check what the original firmware does, but I don't know how to disassemble it....
15:57:21 Quit pixelma ("CGI:IRC (Ping timeout)")
16:13:55Gnelikdan_a: on what are you working now if it isn't a secret
16:15:10dan_aGnelik: Not Rockbox - work I get paid for, but don't want to be doing on a Sunday!
16:15:32 Quit hannesd ("Client suicide")
16:16:18jhMikeSamiconn: that's certainly cleaner
16:19:12 Quit merbanan ("Leaving")
16:19:42dan_aGnelik: When I get some free time I'll have a look at the LCD thing, and let you know when I've got that working.
16:27:57jhMikeSJdGordon: I can't see how the Encoder Settings is coming up with a Not Done yet!
16:32:00 Join actionshrimp [0] (
16:35:41jhMikeSI see it now
16:38:10 Join Id2ndR [0] (
16:41:59 Quit Gnelik (Read error: 110 (Connection timed out))
16:53:04 Join Gnelik [0] (n=Miranda@
16:57:02 Join My_Sic [0] (
17:02:09 Join Blackfish` [0] (
17:02:36***Saving seen data "./dancer.seen"
17:04:19 Join Aldoliel [0] (
17:12:40 Join pixelma [0] (i=54bd5f96@gateway/web/cgi-irc/
17:26:10 Quit Blackfish` ("( :: NoNameScript 4.03 :: )")
17:26:16 Quit bernych ("CGI:IRC (EOF)")
17:39:11 Quit pixelma ("CGI:IRC 0.5.7 (2005/06/19)")
17:40:46Gnelikdan_a: what turn's led off *ptr &= ~(1 << 13); or *ptr |= (1 << 13);
17:44:01 Join NoRemorse [0] (
17:46:55Strogghrmm somewhere, something is calling as instead of arm-elf-as and it's botching my build..
17:50:41 Join MarcoPolo [0] (
17:51:46dan_aGnelik: I think that when bit 13 is on, the LED is off - so *ptr |= (1 << 13);
17:51:52 Quit jhMikeS (Nick collision from services.)
17:51:58 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
17:52:03GnelikI have
17:52:23GnelikSome result it executes to set_irq_level(0);
17:52:35 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
17:54:18dan_aExcellent news!
17:54:30Gnelik5 blinks
17:55:39GnelikComment it and you'l see yellow scren
17:55:55GnelikIt comes to logo
17:56:28Stroggman.. Makefile syntax is so crazy
17:57:47StroggCan someone explain what this make line does?
17:57:52Strogg $(call PRINTS,AR+RANLIB $(@F))$(AR) ruv $@ $(OBJS) >/dev/null 2>&1
17:58:48 Join merbanan [0] (
17:59:42Gnelikdan_a:Greate news
18:00:06dan_aGnelik: Do you mean you actually see the Rockbox logo?
18:00:33GnelikBecause of lcd driver
18:00:59GnelikBut it goes very far
18:01:46GnelikI didnt make blinks after that
18:02:15Learstrogg: Calls ar to build an archive?
18:02:47LearForget the first $() part, it just echoes the command.
18:03:26StroggLear: hrmm ok, thanks
18:05:34Stroggok.. here's my problem.. I know what it is, but I can't figure out how to solve it..
18:06:14Stroggwhen you first build from a clean source, the first thing it does is build the tools. For this, it needs a gcc and as for the host system.
18:06:27StroggI've got gcc and as in my path, so all is good.
18:07:03Stroggbut later when it's building the firmware, it's trying to use as instead of arm-elf-as, and it's giving me an error.
18:07:32LearYou mean, despite using arm-elf-gcc?
18:08:33Stroggis there a way to make 'make' more verbose? I only see CC ata_idle_notify.c
18:08:41Stroggas: unrecognized option `-mcpu=arm7tdmi'
18:09:48LearUse 'make V=1'
18:10:15StroggHrmm. yeah, despite using arm-elf-gcc
18:10:31LearCould be that you have wrong binutils. (Similar problems happen on m68k with a too new binutils.)
18:10:31StroggCan I paste the command and error?
18:10:32 Join Blackfish` [0] (
18:10:47Strogg(it's a big gcc line. hehe)
18:11:11Stroggactually.. I better not.. it's pretty huge.
18:12:10 Quit Blackfish` (Client Quit)
18:12:14Gnelikdan_a: it seems ata_init hangs up
18:12:45LearYep, I do suspect a "bad" version of binutils. What does "arm-elf-as -v" say?
18:13:08StroggGNU assembler version 2.15 (arm-elf) using BFD version 2.15
18:13:08dan_aGnelik: Try using the version of ata-e200.c that is in CVS
18:14:29*Strogg wonders why no one else gets these errors when they build
18:14:45LearStrogg: Old version of binutils, it seems. 2.16.1 is known to work.
18:15:37Strogghrmm ok.. I'll try and make a new toolchain and try again
18:17:37 Join Ribs [0] (n=freenode@
18:18:26 Join sando [0] (i=lolsteam@
18:21:51 Quit Blackfish (Read error: 110 (Connection timed out))
18:22:01Gnelikdan_a: ata_init() == 0 - no errors?
18:22:36 Quit NoRemorse ("( :: NoNameScript 3.81 :: )")
18:23:16dan_aGnelik: I think so - but I can't remember for certain. I've not put error handling in the NAND driver yet
18:24:25Gnelikdan_a:We come to Partition detection
18:25:22 Join mirak [0] (
18:25:45dan_aWith the driver in CVS (which does nothing,) partition detection will fail.
18:26:02GnelikI know
18:26:02 Quit midkay (Read error: 110 (Connection timed out))
18:26:07GnelikIt writes somethin
18:26:34GnelikBut only it can write is partition error
18:27:40GnelikWe need NAND read to go further
18:27:45GnelikAnd lcd driver too
18:29:41dan_aGnelik: OK. Which version of the Sansa do you have?
18:31:58dan_aI've been doing some more work on the patch, to clear up some warnings and to stop it including wmcodec.c - I'll post up a patch soon and make sure that the ATA driver is correct.
18:32:18dan_aIt's possible that whatever is stopping the LCD driver working is stopping the ATA driver too.
18:34:00GnelikHaw can i update with next patch
18:34:07Gnelikwith curent?
18:34:52 Quit sandoaway (Read error: 110 (Connection timed out))
18:35:42dan_aI don't think there is an easy way to do it. Have you made many changes?
18:36:11Gnelikoly in main.c
18:39:07dan_aThe patch doesn't touch main.c, so you can roll back the old patch (patch -R -p1 < e200_3.patch) and then apply the new one.
18:40:13Gnelikdan_a: Farest point for now dircashe init
18:41:40dan_aAgain, that will rely on a working disk driver
18:46:40 Join Arathis [0] (
18:47:35Gnelikdan_a:Gona go now.Bye.
18:47:51dan_aGnelik: Bye. And thanks for your work on this.
18:48:11 Join senab [0] (
18:48:15 Join _FireFly_ [0] (
18:48:34Stroggman.. I've gotten all sorts of images to build, but none of them run on my ipod..
18:48:43 Quit senab (Client Quit)
18:48:53 Join senab [0] (
18:49:33senabafter installing the cross compiler, is it safe to delete build-rbdev?
18:49:46senabcus it's taking up 400mb
18:49:53 Quit Gnelik (Read error: 104 (Connection reset by peer))
18:50:38StroggOk.. so, I've built an image from CVS, and I get an error on bootup.. "Prefetch abort at DEADBEEE".. I searched the forum, and there's a lot of erefences to a bug that's been fixed..
18:50:47Strogganyone know the info on this one?
18:51:01senabfor what build?
18:51:07Stroggipod video
18:51:31senabany patches?
18:51:54Stroggno patchs.. straight from CVS, with a toolchain that's built from the script in the tools directory
18:52:21dan_aThat sounds like some kind of stack error
18:52:56Strogglemme try again, just to make sure it's not some crazy screwup on my end..
18:57:32 Quit senab (Remote closed the connection)
19:02:08 Quit mirak (Remote closed the connection)
19:02:39***Saving seen data "./dancer.seen"
19:03:46 Join My_Sic [0] (
19:03:47obojhMikeS:"> :D
19:04:18[g2]what command-line linux tool do you guys recommend for manipulating mp3 tags ?
19:15:03 Quit [sellout] (Read error: 104 (Connection reset by peer))
19:17:12 Join Gnelik [0] (n=Miranda@
19:18:24GnelikI am back again :)
19:19:14*Strogg rolls to disbelieve
19:19:43 Quit ompaul (Read error: 104 (Connection reset by peer))
19:22:13 Quit actionshrimp (Read error: 104 (Connection reset by peer))
19:24:05 Join actionshrimp [0] (
19:29:30 Join bluebrother [0] (i=o5Wo4CNx@rockbox/staff/bluebrother)
19:29:41atrusmostly ot: i've got an iaudio x5 (rockbox working nicely thank you :) ). somewhere in the back of my brain something's telling me i once heard something about this being compatible with some ipod accessories. anybody know more than i do in this regard?
19:31:12Soapwell, if the accessory uses the dock connector...
19:32:05Gnelikdan_a: new patch is fully without Nand driver?
19:32:21atrusthe dock looks vaguely similar (at a distance :) )
19:33:33dan_aGnelik: It is. The NAND driver is not ready to be committed yet, but I think that this patch can be, soon
19:34:23Gnelikdan_a: What in driver isnt working yet??
19:36:18dan_aI want to get reading above 4Gb working properly before it's committed. Also writing doesn't work and there is no error checking (which we will need to get fixed quickly) and the microSD slot isn't supported (which can wait a while)
19:36:39 Quit Ribs (Read error: 113 (No route to host))
19:37:27 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:38:57Gnelikdan_a: can it read and find partitions?
19:40:32GnelikDo you have a time to work on sansa now?
19:40:35 Join Deef [0] (
19:40:46GnelikWe need to invistigate lcd bug
19:41:09dan_aI do
19:42:27 Quit sando (Read error: 60 (Operation timed out))
19:46:32 Quit TeaSea (Read error: 110 (Connection timed out))
19:49:22Gnelikdan_a: I can't see any reason why it isn't working
19:50:55dan_aGnelik: I've got one idea (I don't know if it's a good idea or not, though) - it might be that only certain areas of memory will work as the framebuffer.
19:51:21dan_aI think that the areas used should be in the .map files
19:54:25amiconnThe sansa lcd hookup is serial??
19:54:32 Join lost|X40 [0] (
19:54:49ompaulis old the current "stable release" or is it something else?
19:55:33dan_aThere's a serial command connection (bit banged GPIO), but it reads the framebuffer itsself.
19:55:44 Quit Gnelik (Read error: 104 (Connection reset by peer))
19:56:19 Join Gnelik [0] (n=Miranda@
19:57:33dan_aThere's a big difference in the addresses for the framebuffer in the bootloader and normal builds I have here... I wonder if that is it
19:57:55amiconnThat's bad...
19:58:25amiconnMany parts of rockbox rely on framebuffer updates not showing up on th elcd immediately
19:58:56amiconnGuess the Sansa needs 2 framebuffers then, the rockbox framebuffer and the hardware framebzffer
19:59:20dan_aamiconn: I don't know - we tell the hardware when to read the FB
19:59:21amiconnAt least updates will be simple, just one (or a bunch of) memcpy's
19:59:45 Quit XavierGr ()
19:59:45amiconndan_a: Can you tell it to only read a rectangular area?
19:59:58dan_aNot that we know of
20:00:30amiconnHmm, perhaps that's not necessary, depending on how fast the hardware reads
20:01:14dan_aIs lcd_update_rect just used for performance, or could there be things outside that rectangle being updated?
20:02:10amiconnIt's primarily used for performance, dunno whether there are parts of code which rely on outside areas not being transferred
20:02:26amiconnThe update recatngle isn't pixel exact on all targets anyway
20:03:24dan_aI guess we'll wait and see if there are strange artifacts when we've got RB booting properly, and if neccessary introduce the double buffer.
20:03:46amiconnOn mono & greyscale targets lcd_update_rect() may update more than the specified area
20:03:58amiconnIt rounds up to full bytes
20:04:23 Quit lostnihilist (Read error: 104 (Connection reset by peer))
20:04:55amiconnThe monochrome format consists of 1x8 pixel blocks, greyscale can be 1x4 pixel blocks (h1x0), 4x1 pixel blocks (greyscale ipods), or 1x8 pixel blocks (x5 remote)
20:06:37Gnelikdan_a: nothing strange in .map files
20:07:37dan_aGnelik: Look at the addresses for lcd_framebuffer in and - on mine they are a long way apart.
20:09:33dan_aI'm going to see what happens if I put the framebuffer in IRAM, just as an experiment
20:09:52 Join midkay [0] (n=midkay@rockbox/developer/midkay)
20:10:26amiconndan_a: There might be another problem, provided that you remap the ram the same way as on the other pp targets
20:10:49amiconnThe lcd controller might not follow the remapping, or have its own remap registers
20:11:49dan_aamiconn: We do remap in the same way. That would explain it not working. But isn't there some remapping happens in the bootloader?
20:12:51 Join midkay_ [0] (
20:14:24[g2]dan_a any info on the JTAG BSDL for the sansa ?
20:15:02dan_a[g2]: Sorry, I don't know anything about JTAG
20:15:44[g2]dan_a it sounds like you are doing some sansa development
20:16:08Gnelikdan_a: look at
20:16:25Gnelik[g2]: We do
20:17:40[g2]Gnelik dan_a last time I e-mailed Daniel S. a couple weeks ago there wasn't any information about the BDSL file for using the JTAG on the Sansa e series
20:18:07 Quit midkay (Read error: 145 (Connection timed out))
20:18:32*[g2] would like to be able to hook up JTAG then be able to do HW debugging and have complete control over unit
20:19:05[g2]it'll be interesting what NV does with the PP line
20:19:16Gnelik dan_a: Ha
20:19:42GnelikTry this: copy #elif CONFIG_CPU == PP5024
20:19:42Gnelik #define DRAMORIG 0x10000000
20:19:42Gnelik #define IRAMORIG 0x40000000
20:19:42DBUGEnqueued KICK Gnelik
20:19:42Gnelik #define IRAMSIZE 0x18000
20:19:42Gnelik #define FLASHORIG 0x001f0000
20:19:43***Alert Mode level 1
20:19:43Gnelik #define FLASHSIZE 2M
20:20:26dan_aGnelik: does that work?
20:20:46GnelikI didn't try
20:20:51GnelikBut it can
20:22:42dan_aGnelik: I'm just on the phone at the moment - I'll try when I'm off the phone. I suspect amiconn might be right though.
20:22:47 Quit Deef ()
20:25:34Gnelikdan_a: I am mega programer
20:25:45GnelikI see logo!!!!!!!!!!!!!!!!!
20:28:24 Quit lee-qid ("Trillian (")
20:29:44***Alert Mode OFF
20:37:08[g2]Gnelik did you just get the sanse LCD working ?
20:37:30GnelikOn normal buils
20:37:39[g2]~praise Gnelik dan_a
20:37:57[g2]I guess there's not bot in here
20:38:04[g2]that does that
20:38:21[g2]Gnelik dan_a _congrats_ !
20:38:51 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
20:38:53[g2]Gnelik is that an svn build ?
20:39:02 Quit Aldoliel (Read error: 104 (Connection reset by peer))
20:39:47dan_aGnelik: Why does that work?
20:40:05Gnelikdan_a: I did that i tell you
20:40:47dan_aI know. The other PP targets don't need to do that - why does Sansa?
20:41:13GnelikI don't know
20:41:39dan_aI think it proves amiconn right.
20:42:55Gneliki need nand driver
20:43:06Gnelikcan you give me what you have
20:43:14preglowyou use that in the lds file and still remap memory?
20:46:40dan_aGnelik: I've just msg'd you a place you can download them from
20:47:11dan_aI don't think that putting that in the lds file is right, even if it works.
20:48:41GnelikI don't think so to
20:49:12dan_aIt looks like we need to give the LCD driver the unremapped address of the lcd_framebuffer. Is there a clean way to find that?
20:49:16preglowis the pp5024 lcd controller external?
20:49:28dan_aNo, it's built in
20:49:33preglowthen weird
20:49:50preglowbut no, no clean way i can think of apart from just applying the hard coded offset
20:50:10preglowwhich is 0x10000000
20:50:29preglowtry putting the framebuffer in iram
20:51:12 Join readonmylips [0] (i=58a10615@gateway/web/cgi-irc/
20:51:42dan_aI thought I had in a test I did - but I've just checked the .map file and it is still in DRAM... oops!
20:55:24dan_aOdd. I've got "#define IRAM_LCDFRAMEBUFFER IDATA_ATTR" in my config-e200.h, but the framebuffer is going into DRAM.
20:56:24dan_aAh, found it!
20:57:15 Quit newbyx86 (Remote closed the connection)
20:57:44Gnelikdan_a: What
20:58:03GnelikI cant write to screen but Show colo can
20:58:26amiconndan_a: There is a reason why we need to remap for mainline: the arm interrupt vector table can only reside at address zero
20:59:05dan_aGnelik: edit firmware/export/config.h and add PP5024 to the list of defines for IRAM, at the bottom of the file
20:59:28 Quit amiconn (" Want to be different? HydraIRC -> <-")
20:59:39GnelikAnd kill my defenations?
21:00:56dan_aGnelik: Yes
21:02:05Nimdaeboot with hold to clear settings seriously needs to be reconsidered D:
21:02:12dan_aAh. There's a problem with putting the framebuffer in IRAM.
21:02:42***Saving seen data "./dancer.seen"
21:06:19Gnelik dan_a: it don't working
21:06:46dan_aGnelik: No. The framebuffer is too big
21:11:08GnelikCan we do something wit it
21:11:16Gnelikwit = with
21:12:33dan_aRight now I'm testing it by putting the framebuffer in IRAM, and setting IRAMSIZE to 0x20000 - which doesn't leave any IRAM for the plugins.
21:13:26dan_aWhat I'm going to do is test applying an offset to the framebuffer address that we give to the LCD driver, and having the framebuffer in DRAM
21:13:34Gnelikat 0x18000 it works normaly
21:14:27dan_aThat leaves 0x2000 bytes for plugins.
21:14:29readonmylipssansa port dev −−> YOU ROCK !
21:14:39dan_aThank you!
21:14:46 Quit Siku (Read error: 145 (Connection timed out))
21:14:53GnelikWe are %)
21:15:24Nimdaei think they smell funny
21:15:27Gnelikwhy framebuffer is so big
21:15:32readonmylipsjust finished the daily log and its far better than any others lectures (readings?)
21:15:42Gnelikisn't it bigger in iP Video?
21:16:05 Quit rconan (Remote closed the connection)
21:16:48GnelikScreen i biger there?
21:17:03Gneliki = is
21:17:05dan_aIt is, but it doesn't put it in IRAM
21:17:29dan_aWhat did you need to take out of init() to get show_logo to work?
21:17:33GnelikLets put it there
21:17:38 Join lee-qid [0] (
21:18:41 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
21:18:47 Quit linuxstb (Read error: 110 (Connection timed out))
21:18:48dan_ais that all?
21:18:49readonmylips test on real hardware will block me to read with "normal" firmware and actual mp3 library ?
21:19:14Gnelikdan_a: yep
21:19:37dan_areadonmylips: At the moment, yes - but when you put the original firmware back on, you can read your MP3 library again
21:20:46 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
21:20:55 Quit Id2ndR ("Parti")
21:21:00dan_aHahahaha! "No partition found. Insert USB cable and fix it" I think that means I'm writing to the screen :D
21:21:46readonmylipsok. and what do have to do to test your firmware ? my cvs dir -with rockbox-dev is done and updated
21:21:53GnelikWhat did you do
21:22:20 Join tomal [0] (
21:22:55dan_aGnelik: I put the LCD framebuffer in IRAM. Next to move it back to DRAM
21:23:45dan_areadonmylips: You need to patch with e200_4.patch from
21:24:07Gnelikit didnt write me about partitions
21:24:42dan_aI'm using the dummy ATA driver, not the one which reads the disk.
21:26:05Gnelikdan_a: i made buffer lager but it don't shows a logo
21:26:49dan_aGnelik: Have you edited firmware/export/config-e200.h to put the framebuffer in IRAM?
21:27:12dan_aTry that
21:28:43 Join rconan [0] (
21:28:43GnelikI can't see the place
21:29:32dan_a#define IRAM_LCDFRAMEBUFFER IDATA_ATTR /* put the lcd frame buffer in IRAM */
21:31:08GnelikHow do we put buffer to DRAM?
21:32:05dan_aComment that line out, like it is in CVS
21:33:11GnelikGrrr. It dont swo log
21:33:24GnelikIt don't shows logo
21:33:54 Quit jhMikeS (Read error: 104 (Connection reset by peer))
21:37:20 Quit BHSPitMonkey (Remote closed the connection)
21:37:39 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
21:38:26 Quit linuxstb (Read error: 110 (Connection timed out))
21:40:34Gnelikdan_a: When i changed
21:40:43Gnelikit was in DRAM i think
21:40:53dan_aIt was.
21:41:32GnelikDRAMORIG 0x10000000 was changed
21:41:42Gnelikmaybe this effects?
21:43:19dan_aIt does. In a normal build, DRAM is remapped from 0x10000000 to 0x00000000. But the LCD controller doesn't know about this.
21:43:44 Quit actionshrimp (Read error: 54 (Connection reset by peer))
21:44:01dan_aIf you tell the linker that DRAM is at 0x10000000, then it will put the framebuffer where the LCD controller can see it, but you will break all sorts of other things
21:44:23CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:44:23*Bagder just handed out some more sansa research
21:44:30 Join actionshrimp [0] (
21:46:00GnelikGona go now
21:46:03dan_aSee you later
21:47:04 Quit Gnelik (Read error: 104 (Connection reset by peer))
21:47:38 Quit merbanan (Remote closed the connection)
21:48:14 Part n1s
21:48:56dan_aWoo! Manufacturing mode!
21:49:40readonmylipswhats that ?
21:50:34dan_aIt could potentially help us unbrick a Sansa
21:53:29 Join GreyFoux [0] (
21:54:25tomalHey, I've just put a set of patches to FS (#6357). They change codecs memory layout and save some memory.
21:55:01tomalCould someone test and comment them? Rockbox on iFP doesn't yet work well enough.
21:58:49preglowtomal: the last one looks interesting
21:59:09LearYes, but even considering temporary mallocs, Tremor can be quite memory hungry...
21:59:17preglowwell, that's how it was designed
22:01:50 Quit SUSaiyan (Remote closed the connection)
22:01:57 Quit Febs ("Chatzilla 0.9.76 [Firefox 2.0/2006101023]")
22:02:32LearApparantly that was based on a guess on how embedded platforms would evolve...
22:03:23preglowyeah, i read that somewhere
22:03:32preglowproblem is, he wasn't really wrong
22:03:34preglowjust half wrong :>
22:03:47LearAnother point, AAC is actually quite malloc hungry too, for large files. The current way the seek tables are created uses lots of memory.
22:03:52 Join SUSaiyan [0] (
22:03:56preglowugh, aac
22:04:01preglowi really want to be rid of faad
22:04:41 Quit JoeBorn ("Leaving")
22:05:10LearWe could ditch SBR and PS. Neither the iPod nor the Zune supports that anyway...
22:05:25 Nick spug is now known as slappfisk (
22:05:50preglowwe probably should, yes
22:06:08tomalmaybe then it would fit in iFP memory
22:06:09preglowat least for now
22:06:18 Join rotator [0] (n=e@rockbox/developer/rotator)
22:06:30preglowthe only format apart from aac-lc that is used, is aac+sbr+ps
22:06:35LearBtw, are SBR files realtime on iPod as it is now?
22:06:35 Nick slappfisk is now known as spug (
22:06:38preglowand i don't think we can make that work realtime
22:06:47preglowLear: i sincerely doubt it, even lc files are barely realtime
22:06:53preglowsbr adds considerable complexity
22:07:11 Join roolku [0] (
22:07:15preglowbesides, i have no sbr files with which to test
22:07:33LearYou can create them with the Nero encoder.
22:07:49preglowis there a linux binary around
22:08:21LearThere's been talk about it at least, but I don't know the status.
22:08:27preglowtried sbr on iriver?
22:08:57preglowi wonder what the hell's up with aac in ffmpeg
22:09:08LearYep. Not realtime. Several tables not in IRAM is one reason.
22:09:16preglowhow far from realtime?
22:09:30preglowtables? does it need lots of them? sounds weird
22:10:14LearPretty bad, but I haven't any measurements. I'd guess less than half realtime.
22:10:25preglowdoesn't sound unreasonable by far
22:10:46preglowi really wouldn't use sbr for anything but streaming anyway
22:10:56Martian67do plugins for rockbox have to be specificly designed for rockbox
22:11:03LearI don't remember exactly, but table constants, like sine or cosine tables.
22:11:04Martian67or is there a decent way to compile generic code?
22:11:09preglowand as far as i know, there's no point in using aac at so low bitrates. vorbis both sounds better and is easier to decompress than aac+sbr
22:11:25preglowMartian67: well, the api interface is rockbox specific, sure
22:11:45preglowif you mean an easy way to just port some c program, then no
22:11:47Martian67was thinking how cool a scummvm port would be for higher end players
22:12:02preglowi think someone's actually had a look at scummvm
22:12:02Martian67but it would require alot of work
22:12:04preglowbut no, not easy
22:12:19Martian67it should be possible
22:12:22Martian67its runs of the DXZ
22:12:41preglowscummvm is c++
22:12:48preglowthen it sure as hell won't be easy
22:12:58Martian67nope :(
22:13:02Bagderor we'll get a busload of extra linkage
22:13:20roolkuhi, is anybody here who has experience with the h120 rs-232 connection and/or the commit by LinusN to enable logf output through it?
22:13:38Martian67would have to be a reimplentation
22:13:40Martian67aka alot of work
22:13:44roolkuI have built an adapter and it works with the OF as described at but not with rockbox :(
22:13:44Bagderroolku: afaik, only LinusN has used that
22:14:13Martian67just imagine how cool maniac mansion would be :)
22:14:20roolkubadger: okay, I'll see if I can catch him
22:15:05preglowbut c++ in rockbox shouldn't be exactly impossible, at least for plugins. you'd need to not use iostream, also we'd need to code a malloc and use it to wrap operator new
22:15:18preglowstuff like exception handling we'd want to disable, i gues
22:15:25Nimdaei think i discovered something concerning usb...not all usb ports are created equal...i was getting a net gain on battery charge on a computer at work on my ipod
22:15:32Nimdaeit charged to 100% while playing
22:16:01 Join newbyx86 [0] (
22:16:05 Join [sellout] [0] (
22:16:14Nimdaedoesn't do that on my computers at home
22:16:24Martian67or imagine a GNASH player for higher end models :)
22:16:33Martian67that could be pretty cool
22:17:24preglowyes, flash, that's what i need on my dap...
22:17:46 Quit _FireFly_ ("Leaving")
22:17:50Martian67preglow: who cares, its all pretty gimmiky anyway :)
22:18:01Martian67who need sudoku or chess :P
22:18:29Martian67can hardly wait for a full 5.5 port
22:18:50Martian67although no comlaints on the current, ipod firmware sucks anyway
22:20:51dan_aCan anyone tell me how the bitmap build system works? The Rockbox logo seems to be being made as an RGB565SWAPPED image, which doesn't look good on an RGB565 screen
22:21:38Bagderdan_a: it should be controlled by the configure file
22:21:46Bagderbmp2rb_native="$rootdir/tools/bmp2rb -f 5"
22:22:02dan_aThank you
22:22:39dan_aI've got the cache flush written, and we're starting to get somewhere
22:24:46 Quit spiorf (Remote closed the connection)
22:25:05 Join secleinteer [0] (
22:27:02 Quit MarcoPolo (Remote closed the connection)
22:36:12 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
22:40:42 Quit Rondom ("Ex-Chat")
22:41:30 Join niskel [0] (
22:41:57niskelI am having my trouble booting my iPod in Rockbox or Original firmware!
22:42:46Nimdaedescribe "trouble"
22:42:59niskelRockbox error: -1
22:43:09niskelLoading original firmware...
22:43:12niskelthat is it
22:43:20niskeleven if I hold down the menu button
22:43:34Nimdaeholding the menu button will boot the original firmware
22:43:42Nimdaedo you have the .rockbox folder on your ipod?
22:43:55Nimdaeand the rockbox.ipod file?
22:44:01niskelI should
22:44:12niskelI cant get into the disc to do anything about it
22:44:23Stroggyou can't plug it into USB?
22:44:29Nimdaeyou should be able to
22:44:38scorchereboot and hols select and play
22:44:51Stroggyeah.. the ipod has a rescue disk mode..
22:45:10niskelscorche: thanks
22:45:11Stroggwell.. -the +my
22:45:13niskelthat let me in
22:45:19Nimdaewell, even still, you can get to it from the apple firmware too
22:45:39niskelthat was the problem... it wouldnt even boot apple firmware
22:45:51Nimdaedid you give it some time?
22:46:00niskel+1 minute...
22:46:28tomalOk, I have to go, if anyone takes a look at my patches, please comment them in FS.
22:46:28scorchedont worry about it for now...copy the .ipod file and the .rockbox folder for now, and then you can test it later ;)
22:46:35 Part tomal
22:46:53 Join Terinjokes [0] (
22:47:11Terinjokesanyone know where itunes 7 keeps the firmware files?
22:47:45niskelThere, back in. The zip file didnt unzip properly :/
22:47:57niskelor the iPod didnt unmount properly
22:48:20niskelall is good now. Thanks!
22:48:24 Part Terinjokes
22:48:32Nimdaeyeah, be sure to "eject" it before disconnecting ;)
22:48:39spugevery time i plug out the USB cable, my H300 says "scanning directories..." or something, even though i always unmount it first
22:49:15spughow come?
22:49:27 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
22:50:27Nimdaeare you "ejecting" or "unmounting"? these are different things, at least in linux
22:51:18niskelI am using linux
22:51:27niskelI usually just use KDE to unmount/eject
22:51:35niskelthey call it "Safely remove"
22:51:45Nimdaeyeah, ubuntu says eject
22:51:51Nimdaebut they basically evoke the eject command
22:52:01Nimdaewhich "safely" unmounts it
22:52:33niskelstrange thing is though, is that it never shows the checkmark on the ipod
22:52:39 Join Terinjokes [0] (
22:52:41Nimdaei accidentally discovered that using unmount was bad, heh
22:52:47niskelit always still shows the do not eject symbol
22:52:55Terinjokesanyone? iPod Updater no longer exists
22:52:57niskelwhat is wrong with unmount?
22:53:10Nimdaewhen i use eject in ubuntu and rhel4 it gives me the checkmark
22:53:20Nimdaeniskel: it doesn't flush the buffers to disk
22:53:29Nimdaei ended up with file corruption
22:54:04Nimdaewell, at least for removable devices
22:54:12niskelso to properly eject, would just doing an 'eject /dev/sdg' work?
22:54:17niskelwhere /dev/sdg is the ipod
22:54:22amiconnHrmm, tomal never stays for long :/
22:56:58Nimdaeniskel: also, with my new 4gb usb drive, i noticed that even though the os says it's done copying, the drive itself indicated it was still writing to the drive...~2gb kept showing the indication for about 2 minutes after os said it was done...unmount would kill that immediately, whereas eject would allow that to finish
22:57:03Nimdaeat least in my experience
22:57:06Terinjokesany ideas?
22:57:22Nimdaeon what?
22:57:43Terinjokeswhere i can find the ipod firmware, now that ipod updater is gone
22:59:07niskelNimdae: I can only eject if I am root :/
22:59:45NimdaeTerinjokes: this?
23:01:11amiconnLear: Tried your colour stripes with the updated dithering yet?
23:01:30spugso in linux, is eject and umount the same or not?
23:01:49LearNo, but I can do a quick test...
23:01:49spugNimdae: you said both that it was and that it wasn't?
23:02:20blue_lizardit is not
23:02:31amiconnIt should work. I tried with (85,85,85), (170,170,170) and (255,255,255)
23:02:32blue_lizardeject is the command to pen the cd drive
23:02:45blue_lizardor an device that can be opened
23:02:47***Saving seen data "./dancer.seen"
23:03:02amiconnAll these give solid greylevels as they should
23:03:07blue_lizardunmount is the command to detach an filesystem from the system
23:03:16Terinjokesgreat! thanks, now only if i had windows
23:03:43amiconnBest thing is that the new formula even saves a few insns on coldfire
23:03:48blue_lizardsome systems have a newer command that allows an user to mount unmount
23:03:55spugbut if i first eject the device, i can't unmount it afterwards?
23:03:56blue_lizardpmount and pumount
23:04:36blue_lizarddepends on the device
23:04:40 Join nick89 [0] (
23:04:45blue_lizardwhich device had you ejected?
23:04:54spugiriver H300
23:05:05blue_lizardand irive cannot be ejected
23:05:09amiconnjhMikeS: I guess you use white foreground colour. That makes the selection frame in solitaire vanish completely...
23:05:10Learamiconn: yes, those were the colors I found to be dither free too.
23:05:14blue_lizardit only can be unmounted
23:05:24spugi can successfully run the eject command on it though
23:05:51blue_lizardyou can argument as long as you want
23:05:51amiconnkkurbjun removed too much...
23:05:56 Join Siku [0] (
23:05:56amiconndionoea: around?
23:05:56blue_lizardyou have problems not me
23:06:00Learamiconn: looked fine.
23:06:41spugblue_lizard: i'm not saying that eject works fine, because i get the scanning directories message, just saying that the command works with no error messages and that's why i thought it worked okay
23:06:52spugbut pumount is perfect, thanks a lot
23:07:39 Quit GreyFoux (Read error: 60 (Operation timed out))
23:08:38Nimdaespug: eject will unmount the device, but it will flush to disk first, allowing safe unmount, so no, they aren't the same
23:09:27StroggNimdae: umount doesn't flush first?
23:09:46blue_lizardof course it does
23:09:46Nimdaenot in my experience
23:10:07Nimdaei would have to dump my usb drive to disk and play with it to see for sure
23:10:09blue_lizardumount and pumount remain blocked til the device is flushed
23:10:20niskel... eject is the only way I can get my iPod show a check mark indicating safe removal if that means anything
23:10:37 Join mirak [0] (
23:10:45Nimdaethe only time i've had umount be blocked is when it was in use
23:10:48 Part Terinjokes
23:11:09Nimdaewhen a device was in use i mean
23:11:09Stroggeject likely sends some kind of usb-eject command to the device.
23:11:12spugi couldn't umount my iriver because it wasn't in fstab, so i always used eject
23:11:24spugi didn't know about pumount of course
23:11:37Nimdaewell, i've had file corruption using umount ;)
23:11:48Nimdaemaybe it was just a coincidence though
23:11:48StroggI didn't think umount cares about fstab.. I thought it only looks at mtab
23:11:53blue_lizardpmount and pumount is the reason the device is not needed in fstab anymore
23:11:59Stroggor maybe even /proc
23:12:11Nimdaeafaik if an entry is not in mtab, it checks proc, then fstab
23:12:14*Strogg will have to check out pmount sometime
23:12:22StroggNimdae: that sounds sane
23:12:33blue_lizardgnome-volume-manager and afaik even kde uses pmount
23:12:47Nimdaei'll have to look at pmount as well
23:12:56niskelkde does use pmount
23:13:02Stroggooh. nice. debian comes pre-setup with it
23:13:19Nimdaeah, pmount looks to just be a non-superuser wrapper for hotpluggable devices
23:14:12Nimdaeyou specify the device, it grabs the label if a label wasn't specified, and mounts it under /media
23:14:26Nimdaethat's good to know
23:17:04 Join ndansmith [0] (
23:17:32blue_lizarda further nifty thing are custom rules for usb devices
23:18:00blue_lizardi as an example have my iriver allways mounted as USB-HD-iRive
23:18:24blue_lizardand the other known devices hav their own names too
23:19:13Nimdaehave there been advances in reducing battery load on the ipod?
23:19:23Nimdaei seem to be getting a net gain on battery charge on my desktop now
23:19:28Nimdaevery small
23:19:50Nimdaebut after quite some time of playing (including some skipping, playlist changing, etc), i have a 6% increase
23:20:13 Quit [sellout] (Read error: 110 (Connection timed out))
23:20:58Nimdaemaybe my ipod is special
23:23:22 Quit niskel (Remote closed the connection)
23:24:10 Quit webguest00 ("CGI:IRC")
23:24:13 Quit Lear ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
23:28:05mirakrockbox recent cvs are crashing on wav playing on iriver H300
23:28:41 Quit pagefault ("Leaving")
23:32:43 Join Terinjokes [0] (
23:33:07Terinjokesany ideas on how to extract the firmware from iPod_14.1.3.ipsw?
23:33:41Terinjokesthat's the latest firmware for the 1G iPod Nano
23:35:01Bagdernoooo, Rockbox is more recent ;-)
23:35:40TerinjokesBagder, true, except i would like 14.1.3 OF and rockbox ;-)
23:36:00 Quit actionshrimp (Read error: 54 (Connection reset by peer))
23:36:02BagderI figured there was a catch somewhere... :-)
23:36:28Terinjokeshopefully apple restored playback in OF when rockbox is also on the ipod :-)
23:37:26 Join scorche` [0] (n=ScorchE@rockbox/administrator/scorche)
23:37:26 Quit lee-qid ("Trillian (")
23:37:28 Join MarcoPolo [0] (
23:37:31Terinjokesany ideas on extracting the firmware?
23:37:32rasherWhat the..
23:37:46rasherRockbox is playing a song with a path of length > MAX_PATH
23:37:49rasherHow does that work
23:38:24rasherUnless my counting's off.
23:41:02Terinjokeshrm, maybe...
23:43:37rasherIs rockbox strlen utf8 aware, or does a utf8-aware strlen exist?
23:43:55 Part Terinjokes
23:44:16*ompaul congradulates the devs of this rather nice piece of software
23:44:23ndansmithI am completely stumped as to how to use the virtual keyboard to save a .cf on my iPod Mini 2g
23:44:39 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
23:44:40markunrasher: utf8length()
23:44:58 Quit scorche (Read error: 145 (Connection timed out))
23:45:18rashermarkun: Ah, excellent. Do filenames entered with the vkeyboard get input as utf8?
23:46:07markunyes, I think so
23:46:57rasherSo if ( utf8length(newname) + 1 > MAX_PATH ) {
23:47:58rasherwould be a sensible check before creating a new fat entry?
23:48:18rasher(I'll have to check the filename has not been converted to .. uh.. whatsit.. UCS.. something.. yet
23:48:38amiconnfat entries are limited to 255 ucs-2 chars
23:49:05amiconnThat limit is from fat specifications
23:49:12rasherOh right, the name there is the individual entry, not the full path
23:49:15amiconnvfat I mean
23:49:30rasherIs it checked that the full path is not > MAX_PATH as well?
23:49:56amiconnPath length is not part of the vfat specs
23:50:07rasherIt is according to
23:50:11amiconnI don't know whether rockbox checks this, but I guess it does
23:52:13amiconnIndeed it is
23:52:27rasherI still think someone with a clue should extend fat_checkname to check for illegal characters and such
23:52:38rasherRight now it just checks that the name doesn't end with "."
23:52:48amiconnHowver, the specs are unclear about what characters it refers to
23:53:32amiconnI _guess_ it refers to ucs-2 as well
23:53:37rasherI would assume that as well
23:53:44rasherAs in, the characters that the user sees
23:53:59amiconnA 260-char ucs-2 path can well exceed 260 bytes when expressed as utf-8
23:54:38rasherWorst case, it could be 4*259+1, right?
23:55:07amiconnucs-2 can only represent the unicode bmp
23:55:17rasherAh okay
23:55:26amiconn..and the 260 chars include the trailing \0
23:55:39rasherso 3*259+1 th elast char is always a single byte
23:56:03rasherThat's... rather a lot more than 260 bytes
23:56:18amiconnHmm, in fact a few bytes less
23:56:46rasherOh yeah, there's a leading slash?
23:56:50amiconnA filename can't be longer than 255 chars, so a path longer than 255 chars must contain at least one path separator
23:57:12rasherso 3*257+3
23:57:13amiconn...and rockbox paths are always absolute, so the first is also a slash
23:57:29rasherStill a hefty increase
23:57:47rasherI say we go back to pure ASCII for everything.
23:58:16rasherWell, is Rockbox doing anything smart here, or is it simply assuming that everything goes well?
23:58:27amiconnI dunno

Previous day | Next day