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

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

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

#rockbox log for 2007-02-24

00:04:24nlsBagder: thanks :-)
00:04:25pixelmanice green line in the binary size table, nls :)
00:04:54nlsgood night everyone! :-)
00:04:57 Part nls
00:06:14 Join winchester [0] (
00:06:37 Quit Mouser_X (Nick collision from services.)
00:06:46 Quit winchester_ (Read error: 104 (Connection reset by peer))
00:06:46 Join Mouser_X [0] (
00:11:54m0sis there GUI programs to make wps themes?
00:13:19winchesternotepad? =)
00:14:24winchester(there is some program which has most of the tags in it for easy reference but I don't think there's an actual gui designer window type thing)
00:14:56winchesterwell.... I don't know of any... wouldn't be hard to make so I'd expect there to be something
00:15:52 Quit Yoghurt (Remote closed the connection)
00:17:55 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
00:18:23 Join amigan_ [0] (
00:20:21 Join austriancoder [0] (n=christia@
00:23:10 Quit Mouser_X (Nick collision from services.)
00:23:19 Join Mouser_X [0] (
00:27:14m0soh ncie bubble bobble did come out for GBC
00:28:08gtkspertpreglow: You here?
00:28:39 Part redbreva
00:28:44gtksperti would fix it, but then i would get disconnected from the internet
00:28:53gtkspertoops, again...
00:30:51 Join pearldiver [0] (
00:31:19 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:31:23winchesterhow many touch positions are there on the ipod scrollwheel?
00:34:18linuxstb96 if your finger is small enough
00:35:21midkayBagder: around?
00:35:55winchesterIt is not realistic to divide it into 26 and turn it into a keyboard for renaming files, etc. is it?
00:36:04 Quit evilx__ (Read error: 104 (Connection reset by peer))
00:36:14*linuxstb is making a mess of the build table...
00:37:41 Join BobMorris [0] (i=446c7c29@gateway/web/cgi-irc/
00:38:33 Join Everybody| [0] (
00:38:49BobMorrisHello, I'm looking for Write Permissions for the Rockbox Twiki please... I'm registered and confirmed... just need permissions now : )
00:39:55midkayBobMorris: what's your wiki name? :)
00:40:19midkaylinuxstb beat me to it.. jerk :)
00:40:28BobMorrisfigured I'd keep it simple
00:40:34BobMorrishah thanks I appreciate it
00:40:43linuxstbmidkay: No need for that :)
00:41:00linuxstbBobMorris: You're added.
00:41:03BobMorrisand thanks for all your efforts on Rockbox... its a gem !
00:41:08midkaylinuxstb: just kidding! <3
00:41:16BobMorristhanks Linus
00:42:03 Quit BobMorris (Client Quit)
00:42:23winchesterooh..... being more realistic.... the scroll wheel could easily be like the mobile phone keypad.... much more convinient for text input and most people are allready used to it....
00:43:18winchesteror.... is everyone happy with the current text input?
00:43:52markunwinchester: just write a patch and convince people it's better then what we have now :)
00:44:21 Join Gigabeater [0] (i=464e0a82@gateway/web/cgi-irc/
00:44:27winchesteryeah I would, except I won't have time starting monday
00:44:40markunGigabeater: hi
00:44:50markunwinchester: be quick then! :)
00:45:33winchesteri am not a developer and barely know the rockbox api...... can't be quick at this stage.
00:46:11GigabeaterIs there a specific way I am supposed to install a new build.. like I have a build from earlier in the month on my Gigabeat. It has a couple of problems, so I want to put the latest one on. Can I do that without it erasing information?
00:46:23midkayGigabeater: absolutely, all you need to do is unzip the latest build to your player.
00:46:37Gigabeaterand I click Yes to ALL?
00:46:42midkayGigabeater: yes.
00:46:50midkayno problem.
00:47:48markunwinchester: firmware/target/arm/ipod/button-clickwheel.c
00:47:54GigabeaterJust unzip it to the .rockbox folder?
00:48:06markunGigabeater: no, just unzip it to the gigabeat
00:48:11Gigabeateroh okay
00:48:17linuxstbwinchester: And Rockboy uses the wheel-position APU - I think in rockboy/sys_rockboy.c
00:48:53GigabeaterDoes it erase my Doom saves?
00:49:14linuxstbGigabeater: If you're paranoid, just backup your .rockbox folder first.
00:49:19 Join Insectoid [0] (
00:49:21GigabeaterToo late :)
00:49:23GigabeaterOh well
00:49:27GigabeaterI was only at level 3
00:49:35linuxstbBut they should be fine.
00:49:37winchesterI will try (and fail)...
00:50:07winchesterand I actually was just skimming through firmware/target/arm/ipod/button-clickwheel.c
00:50:33InsectoidI've got a 30 gb Ipod 5g. I recently read a nice little article on booting xp from an external USB hdd. Anyone have any idea if this'll flub things up (I.E. overwriting the boot sector, etc?)
00:52:16 Join perldiver [0] (
00:52:40markunwinchester: you can also post a feature request, but writing some code yourself is a lot more fun of course
00:53:50 Quit Arathis|afk ("Bye, bye")
00:53:52 Quit Mouser_X (Nick collision from services.)
00:54:01 Join Mouser_X [0] (
00:54:03winchesterYeah... I kind of hesitate from feature requests..... since there are so many allready and I was asking because I was planning to do it myself... just to learn a bit.
00:54:31ShaidI like the 26 letters as scrollwheel thing
00:54:35Shaidsounds like fun
00:55:01winchesterI doubt you could press where you need properly since the area would be way too small
00:55:05GigabeaterIs there (and easy*) way to enlarge the font of the rockbox menu and possibly give it a shadow?
00:55:37 Quit austriancoder ()
00:55:43ShaidI was thinking more you scroll around to the spot. and then press the centre button
00:55:52Shaidbut that might work with mobile style 3 press keys
00:56:01Shaidbut then weíll ask you to do predictive text too :D
00:56:09 Quit Everybody|away (Read error: 110 (Connection timed out))
00:56:25 Join Everybody|away [0] (
00:56:31GigabeaterOr even just a shadow
00:56:40markunShaid: do you have a reboot scheduled some time soon? ;)
00:56:42GigabeaterIt's terribly hard to read with a custom backgroudn image
00:56:47winchesterit's not like you will be texting anyone..... doubt you will need predictive text
00:56:58midkayGigabeater: you can change fonts.. menu - general settings - display - browse fonts.
00:57:06midkaybut that will probably ruin whatever WPS you've selected.
00:57:09Shaidmarkun: probably.
00:57:27midkayGigabeater: maybe find something you like here:
00:57:43 Join BobMorris [0] (i=446c7c29@gateway/web/cgi-irc/
00:58:07GigabeaterI installed back theme
00:58:12Gigabeaterbut it doesnt change the font
00:58:15BobMorrisI'm back... looking for a little help on my uploaded WPS
00:58:41midkayGigabeater: what did you do after you installed it?
00:59:09GigabeaterUmm.. I can't recall exactly.. this was nearly a month ago.
00:59:21GigabeaterI think I installed it where I figured it should go
00:59:30GigabeaterThen booted the gigabeat to see if it worked
00:59:32Gigabeaterand it did
00:59:41 Quit matsl (Remote closed the connection)
00:59:41GigabeaterBut the actual font doesnt change menu text
01:00:04Gigabeateractual theme*
01:00:05BobMorrisMy uploaded picture is way larger than the others on the page... and I didnt understand how to leave my comments for the page
01:00:09midkayit should. (you need to select it from Browse Themes..) or else you maybe installed it wrong.
01:00:19BobMorrisI didnt know where to use this line = [[%ATTACHURL%/(Name of the][(Text to be displayed)]]
01:00:24 Quit bluebrother ("leaving")
01:00:36 Quit linuxstb ("CGI:IRC")
01:01:16midkayBobMorris: you have to follow the instructions, that means uploading the zip file then inserting the, well, name of the zip file where it belongs :)
01:01:27midkayas for the picture.. well. it looks like you enlarged it or something..
01:01:39BobMorrisyes I read that I dont know WHERE Im supposed to insert this
01:01:51midkayBobMorris: insert what?
01:01:51 Quit Gigabeater ("CGI:IRC (EOF)")
01:01:57BobMorrisI didnt know where to use this line = [[%ATTACHURL%/(Name of the][(Text to be displayed)]]
01:02:17midkaybelow the picture or something.. somewhere in your WPS section.
01:02:30midkayit'll be a link to the download.
01:03:07dan_aHmmm... I think I should be able to use the work linuxstb has been doing today to get the bootloader from my YH-820 :D
01:04:07midkaydan_a: cool!
01:04:24BobMorrisoh I bet the pic is so big cause I used the ZOON.bat to create the pic... I can fix that easy... just dunno about the link think... there is a link I click it and I get the download dialogue
01:04:33midkaydan_a: what's the res on that LCD? is it really 128x96?
01:04:45BobMorrisam I supposed to modify the HTML of the link?
01:05:07dan_amidkay: I believe so. It's definitely fairly low, but nice to look at
01:05:11 Join akaidiot [0] (
01:05:32midkaydan_a: :( darn samsung, skimping out on us like that..
01:06:04midkayBobMorris: paste that exact line below the image URL, and all you need to do is insert the name of the file where it says "name of the file here".
01:06:24midkayi.e. [[%ATTACHURL%/][Download PHK-nano here]]
01:06:36dan_aIf it had a model number on it, or worked the same way as one of the other LCDs it would be even better (and we'd have another PortalPlayer target without sound)
01:07:58midkaydan_a: indeed... *wishes companies would use more standard resolutions, etc*
01:09:21pixelmaI doubt it is important, but that's the same resolution as the X5's lcd-remote...
01:10:17midkaytoo bad it's monochrome though..
01:10:27 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
01:11:29pixelmathe Samsung one?
01:11:30 Join ioor [0] (
01:12:04midkaypixelma: X5's remote LCD.
01:12:09dan_aThe Samsung's colour
01:12:11 Quit pearldiver (Connection timed out)
01:12:11midkaysamsung's is color..
01:12:51pixelmamidkay: no the remote-lcd is frey... err.. greyscale ;)
01:13:14pixelmastill different though then
01:13:20midkaypixelma: oops, sorry :) i mean it's non-color :)
01:13:58 Join MonkeyTamer [0] (
01:14:09linuxstbdan_a: Seems I can reliably start both the "diskmode.ipod" and "diagmode.ipod" applications from within the bootloader (by copying them from the flash to DRAM_START, and then doing a return (void*)DRAM_START). So the problems are all caused by Rockbox/rolo....
01:15:11dan_aI had no luck whatsoever roloing the Apple OS
01:16:01*linuxstb wonders how the ipod Color bootloader went from 62040 bytes to 46488 bytes...
01:16:13BobMorrisk I think I got the idea now... and have modified the download ling... now I just gotta resize that picture : )
01:16:37dan_alinuxstb: When did it do that?
01:16:55linuxstbSince the last official release (a few weeks ago) and the bootloader I'm compiling today.
01:17:10BobMorrisback if I have probs re upping the new PNG
01:17:14 Quit BobMorris ("CGI:IRC")
01:18:27***Saving seen data "./dancer.seen"
01:18:46dan_aOh, by the way, what will stop dbg_save_roms from working on the Sansa?
01:19:22linuxstbI didn't know how the Sansa works - if there's a flash ROM there or not. I couldn't see one listed in the hardware components page.
01:19:47linuxstbYou could simply try it and see what happens. You'll need to add FLASH_SIZE to the config-e200.h
01:20:06linuxstbAll the ipods and the H10 have a 1MB flash afaics.
01:20:36dan_aI'm pretty sure there is - I'll push that on my stack of things to do... first I want to see if I can bring an unusable YH-820 back to life with the bootloader from a different one...
01:21:37 Join Gigabeater [0] (i=464e0a82@gateway/web/cgi-irc/
01:22:10MonkeyTamerI'm a bit interested in trying different formats, and I'm curious to know how efficient a flac music file compared to an mpc music file; I've tried flac myself, and with q8 or so, the cpu boosting is around 20-30%, but has anyone had any experience with mpc?
01:22:18 Quit akaidiota (Read error: 110 (Connection timed out))
01:22:45GigabeaterWould configuring the iCatcher theme to have a custom background image (As opposed to Background color: FFFFFF) be a difficult change?
01:22:58MonkeyTameralthough, I suppose it does depend on the hardware as well, but I'm looking for a relative comparison
01:23:02GigabeaterIs there a way for it to use /.rockbox/backdrops/back.bmp as it's background
01:23:36 Quit akaidiot (Read error: 110 (Connection timed out))
01:23:42linuxstbYes, just browse to back.bmp in the file browser, open the context menu (long-press on select on most players, not sure about the gigabeat), and select "set as backdrop".
01:23:43 Quit Vyrus001 ("Leaving")
01:24:23GigabeaterDoes that change the backdrop in the WPS?
01:24:32 Quit fasmaie (Remote closed the connection)
01:24:39linuxstbNo, so ignore me :)
01:24:50GigabeaterSorry, I should have specifyed
01:25:11linuxstbYou need the %X tag in the .wps file, and the backdrop should go in the same folder as all the iCatcher .bmps
01:25:35GigabeaterI don't know where they are
01:25:41GigabeaterOr how to do that
01:25:47linuxstb.rockbox/wps/iCatcher/ I think
01:25:47GigabeaterI'm pretty much compu-tarded
01:26:22 Part toffe
01:26:41GigabeaterBut the iCatcher theme doesnt use an image as it's backdrop
01:26:45Gigabeaterit uses a color
01:28:28winchesterhey..... what am I missing?
01:28:35winchestererror: ‚Äėstruct plugin_api‚Äô has no member named ‚Äėwheel_status‚Äô
01:28:45 Quit lee-qid_ (Connection timed out)
01:29:13 Quit acrid ("rcirc on GNU Emacs")
01:29:53linuxstbwinchester: Ah, you have a 1st gen mini?
01:30:12linuxstbWhich ipod then?
01:30:12winchesterI am pretty sure first gen doesn't have the touch thing
01:30:23GigabeaterIs there a way to change the background of the iCatcher theme's WPS from a color code (FFFFFF) to an actual image? If so, please explain.
01:30:27Gigabeatermini or nano
01:30:50linuxstbwinchester: Did you select the 2g mini when you were configuring Rockbox? The 1st gen doesn't have the wheel_status function, but the 2nd gen does.
01:31:10linuxstbGigabeater: I just did explain.
01:31:13winchesteryeah I did
01:32:04linuxstbGigabeater: The WPS itself doesn't specify a colour, it just uses the global colour set by the theme. If you add the %X tag (see the CustomWps wiki page for details), then you can use an image instead.
01:33:20linuxstbwinchester: Look in the Makefile in your build directory - does it say TARGET=-DIPOD_MINI or TARGET=-DIPOD_MINI2G ?
01:34:26 Quit Mouser_X (Read error: 110 (Connection timed out))
01:34:32linuxstbAh, are you compiling for the Simulator?
01:34:41winchesteryeah I am
01:34:44linuxstbThe wheel_status() function only works on a real ipod.
01:35:10winchesterwell.... that will make testing a bit annoying
01:35:23linuxstbThink more, test less...
01:35:54winchesterI tend to do things by trial and error
01:35:57winchesteror just error
01:36:06linuxstbThen yes, it will be a bit annoying...
01:37:03 Join Mouser_X [0] (
01:37:53dan_alinuxstb: What do the values that you OR the physical and logical addresses with for the MMAP regs mean? Or don't you know?
01:38:18linuxstbI don't know.
01:38:41Gigabeaterthank you linux
01:39:35 Quit Gigabeater ("CGI:IRC (EOF)")
01:42:59 Join JustJohnny68 [0] (
01:44:04 Join Gigabeater [0] (i=464e0a82@gateway/web/cgi-irc/
01:45:05GigabeaterLinux, sorry, but I still don't understand. I made an attempt though. I put the file back.bmp into my .rockbox/wps/icatcher folder, and in the icatcher cfg file i put %x|back.bmp|
01:45:11GigabeaterI'm being dumb somewherew
01:45:13Gigabeaterbut where?
01:45:23Gigabeaterthe "|"'s?
01:46:06pixelmayou have to add that line in the icatcher.wps file
01:46:24Gigabeaternot the cfg?
01:46:27Gigabeateror aswell?
01:46:37linuxstbJust the .wps
01:46:46linuxstbAnd it needs to be an upper-case X
01:46:56Gigabeaterit is
01:47:23 Join decayedcell [0] (
01:47:30linuxstbYou could also trying looking at your other .wps files, and see if they use %X
01:48:18Gigabeaterdo i have to put the root of back.bmp
01:48:21Gigabeateror just the file name
01:48:30linuxstbJust the filename
01:48:40Gigabeateri'll test
01:49:03linuxstbAnd the .bmp file must be exactly the same resolution as your LCD.
01:49:09Gigabeaterit is
01:49:21linuxstbThen it should work...
01:49:41Gigabeaterthere we go
01:49:52Gigabeaternow i just have to configure some of the other images in the theme
01:50:03Gigabeaterso they dont have white squares aroudn them
01:50:21winchesterwarning: passing argument 3 of ‚Äėrb->lcd_putsxy‚Äô makes pointer from integer without a cast
01:50:26winchesterwhat's that trying to tell me?
01:50:26Gigabeater.bmps can be translucest right?
01:50:50decayedcellGigabeater yeah
01:50:50GigabeaterOr will I have to change each one to a .png in both the filename and the .wps file?
01:50:53Gigabeateroh okay
01:51:21Gigabeaterthank you very much linux
01:51:40linuxstbFrom the CustomWPS wiki page - " * The color [R=255,G=0,B=255] is translated as transparent when used in BMP's, and can be used to create transparent images in the wps. "
01:52:03Gigabeaterthank you
01:53:38preglowgtkspert: am now
01:53:38 Quit Gigabeater ("CGI:IRC (EOF)")
01:54:52linuxstbHmm.... Rockbox's usb_init() function breaks the Apple disk mode code...
01:56:59gtkspertpreglow: I was having some issues with the gigabeat's eq
01:57:06 Quit inversions (Read error: 110 (Connection timed out))
01:57:12decayedcelllinuxstb that was discovered a while ago :D
01:57:21 Quit decayedcell (Remote closed the connection)
01:57:32 Join decayedcell [0] (
01:58:02decayedcellsometimes when you plugin the ipod in rockbox into a computer / AC charger, it will freeze when rebooting into Disk Mode correct?
01:58:22linuxstbThat's not what I'm doing.
01:58:22gtkspertpreglow: I enabled the eq, and changed the LS slider in the graphical eq, and i heard what sounded like a plane's engine starting...
01:58:31decayedcelllinuxstb okay what are you doing then =/
01:58:34linuxstbI'm trying to start Apple's disk mode code without rebooting.
01:59:02preglowgtkspert: can you make it do so again?
01:59:13preglowgtkspert: had the eq been switched on previously in that session?
01:59:19XavierGrCOMPUTER has been up for: -4457 day(s) <- Cool uptime.exe microsoft windows utility
01:59:20gtksperti did it with an mp3, and a flac file
01:59:27XavierGr-4457? LOL
02:00:05gtkspertpreglow: the eq was only turned on while the song was playing
02:00:57preglowbut the plane engine starting thing needs a bit of elaboration
02:01:07preglowlow frequency rumble? white noise?
02:01:32gtkspertok, well, it was like white nois, getting higher in pitch as i listened, and you can barely hear the song in the background...
02:01:45gtkspertscared me the first time i did it...
02:01:49preglowsounds like overflow
02:02:02preglowbut yeah, if you can make it do so again, that would be very helpful
02:02:04preglowhave you tried?
02:02:23gtksperti just tried then, and it is doing it
02:02:57preglowcan you make it do so consistently?
02:02:58gtkspertif i set the eq back to 0 it keeps playing normally
02:03:17gtkspertwhat do you mean by consistently?
02:03:18preglowsure, that switches it off
02:03:32preglowwell, if you can give me instructions on how to it myself, then maybe i can fix it
02:03:37preglowto do
02:03:38gtkspertok then
02:04:01preglowi've got an ipod nano which uses the same eq code as gigabeat
02:04:18gtkspertplay a song, enable the eq, change the LS slider in the graphical EQ
02:04:27 Nick Mouser_X is now known as Zombie_X (
02:04:54preglowwhat codec?
02:05:21gtkspertCurrently im in MP3, i think CBR
02:05:41preglowok, eq working fine for musepack and ogg
02:05:46preglowwill try mp3 now
02:06:45preglowthe ipod build feels so bloody sluggish these days :/
02:07:05gtkspert:/ heheh
02:07:06InsectoidEven with kernel on cop?
02:07:13preglowgtkspert: works fine here
02:07:19gtksperti got it happening with ogg
02:07:19preglowInsectoid: yeah, but not as bad
02:07:32preglowgtkspert: what build are you running?
02:07:38 Nick Zombie_X is now known as Mouser_X (
02:09:32gtkspertpreglow: I believe its not the newest build, but i've noticed it since audio was just working on the gb.
02:09:36gtkspertpreglow: r12159-070130
02:10:03dewdudewow, old build
02:10:28gtkspertHow old is that?
02:10:38preglowalmost a month
02:10:45gtkspert*does svn update
02:10:50preglowbut anyway, i've heard other people talk about eq problems with the gigabeat too
02:10:55Insectoidbounce me
02:10:59Insectoidbounce me
02:11:02gtksperti see
02:11:04dewdudethen again, mine's 2 weeks old, i need to do the same
02:11:06preglowbut i have no idea why it should affect only the gigabeat, the could should be the same
02:11:17*preglow bounces Insectoid s
02:11:35Insectoidwhat um
02:11:40InsectoidI thought I was in MUSHclient
02:11:52Insectoidtalk about strange strange
02:12:58preglowmarkun: yoes
02:13:13preglowmarkun: have you ever noticed any funniness with eq on gigabeat?
02:13:15Insectoid... silly ed
02:13:37gtksperttoffe has had some problems with the eq
02:13:50gtksperti was talking to markun earlier
02:13:54InsectoidAnybody ever play with mojopac? it lets you run xp off've a flashdrive (AKA Ipod) without buggering with the firmware. It is almost a VMware kinda thing. Works fine with rockbox.
02:14:36 Join Arathis [0] (
02:15:03preglowi can make the eq make fucked up overflow sounds, but i don't think that's what you're having
02:15:13preglowhow much gain did you have on the lowshelf filter?
02:15:49preglowthen it isn't that for sure
02:16:03preglowi have to use +24 before i ever get that, and it depends on the source material
02:16:36gtksperti see
02:17:07dewdudei don't think i've ever gotten noise from the EQ at any level
02:17:30dewdudei just get random codec failures and horrible static
02:18:15preglowon what target?
02:18:25preglownever gotten static
02:18:25dewdude5.5 30g
02:18:33dewdudewell, i'll get static on codec failures
02:18:45dewdudefrom i guess the player feeding the DAC the bitstream rather tahn PCM
02:18:48preglowi do get those without the static
02:18:58preglowdewdude: that's severely improbable
02:19:18preglowit's more likely the i2s interrupt isn't disabled somehow and it feeds garbage to the dac
02:19:39preglownot that it matters anyway
02:19:54 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)'s a minor inconvenice...but i've always said i'd rather use the buggy rockbox than the offical ipod
02:21:27 Quit Mouser_X (Nick collision from services.)
02:21:36 Join Mouser_X [0] (
02:23:23 Join fasmaie [0] (
02:23:43preglowi'd call it more than a minor inconvenience
02:23:47preglowit's simply not supposed to ever happen
02:24:06preglowthose are among the most severe bugs i can think of in a dap
02:24:20 Quit Nico_P (Remote closed the connection)
02:24:24preglowit should under no circumstance blast out white noise
02:25:12 Quit winchester ("Leaving")
02:25:39dewdudeheh, i almost wouldn't call it white noise
02:25:45dewdudeit has a hint of metallic to it
02:25:54 Join CriamosAndy [0] (
02:26:29dewdudewhich is why i said it COULD of been bitstream being fed to DAC..not sure how things work exactly, but it was diffently the wrong stream getting into the DAC.
02:26:57dewdudeit's an annoyance to most people, and you're right, it shouldn't happen, but it happens less often than not for me. i just reboot and get on with the tunes
02:27:03preglowthat would be the weirdest bug ever
02:27:17 Quit funky ("leaving") would be. and, you're right, it's probably not possible..but for some reason at random musepack won't decode and i'll get DAC noise
02:29:24dewdudehowever, i can't reproduce it at will or figure out what's different that makes it happen
02:30:33dewdudei'd still love to find out of this thing can pump out dvda quality stuff (96khz 24-bit)
02:31:43preglowsure it can, but it probably won't sound like it
02:32:14dewdudewell, the flac decoder can decode it...but from various sources in here rockbox resamples to 44/16
02:33:17preglowthat could be somewhat quickly changed, though
02:33:20preglowbut no one's done it yet
02:33:40dewdudeyeah, that's what i'm waiting for..becuase i ALMOST had it playing back the 96/24
02:33:54preglowi think we know how to do it, but so far our playback system relies on having just one output rate
02:34:20preglow24 bits you'll have to wait longer for, but that's the part that matters the least anyway
02:34:25preglowthe sample rate is more imptant
02:34:27preglowimportant too
02:34:35dewdudei disagree.
02:34:47dewdudebut i'm also an engineer and look at it differently
02:34:58preglowhave you read the specs on the ipod dac?
02:35:12preglow16 bit audio has a noise floor of about 96 dB
02:35:15dewdudei can't even find actual spec sheets on the ipod dac
02:35:18preglowthat's about exactly the same as the dac can handle
02:35:47preglowrockbox processes audio at a very high (bigger than 24 bits) resolution throughout, the conversion to 16 bits only happens at the last stage
02:35:55preglowso i don't think you'll hear much of an improvement
02:36:02preglowbut we'll probably find out for real some day
02:36:13dewdudein all honesty..most people can't really hear the difference between 44/16 and 96/24
02:36:14 Join Vague_Rant [0] (
02:36:25 Join x1jmp [0] (
02:36:41preglowindeed not
02:36:49preglowbut it all depends on how good the dac filtering is
02:36:49 Part Vague_Rant
02:37:02preglowyou might hear the highest frequencies better if the filter is poor
02:37:08preglowbut it's probably not that poor
02:37:17 Join austriancoder [0] (n=christia@
02:37:23dewdudewell, the noise floor isn't my issue for using 24
02:37:35preglowthen what is?
02:37:44dewdudethe higher resolution and more percise waveform recreation
02:38:06ShaidHeck, Creative sold soundcards for years claiming they did 24bit output, and no one noticed they were lying until someone did a waveform analysis on it...
02:38:28preglowdewdude: if the noise floor is the same, it doesn't matter
02:38:30markunpreglow: perhaps the EQ problems are just a side effect of some other problems on the Gigabeat
02:38:56markunI think we have some problem related to DMA/write buffer/caching
02:39:12dewdudepreglow, ehhhh....i don't want to get into it. i pretty much have a totally different opinion on the subject than everyone else. everyone starts talking about nyquist...i just keep yelling analog.
02:39:45preglowdewdude: we'll all find out the day we support 96/24 and we can do double blind tests anyway ;)
02:39:47markunfor example, when I leave a plugin the background code is reset correctly but the background pattern used for filling still has the colour of the plugin
02:39:56preglowmarkun: weird
02:39:59preglowmarkun: well, i hope so
02:40:04preglowit'll be a weird bug if that's not the cae
02:40:11markuns/background code/background colour/
02:40:19dewdudepreglow, indeed. however, the biggest part is going to be finding TRUE 96/24 sources..and not just upsampled 44/16
02:41:03 Join entheh [0] (
02:41:14 Join Nifl [0] (
02:41:30dewdudeand with the way the industry has been handling things...they don't need 24...they could easily get away with 8-bit and no one would notice
02:41:34 Quit Criamos (Read error: 110 (Connection timed out))
02:42:13preglowwith kickass noise shaping: sure
02:42:16decayedcellhmm system.c seems to have been updated with PLL changes?
02:42:19dewdudenot even that
02:42:25dewdudethere's no dynamics in modern music
02:42:30preglowgood point
02:42:30gtksperti need to head off, i'll talk to you guys later on...
02:42:37preglowgtkspert: one more thing
02:42:45gtkspertpreglow: yep
02:42:46preglowdoes it happen with other bands than ls
02:42:48 Join winchester [0] (
02:42:49preglowor just with ls
02:42:53NiflIf i install Rockbox on my iPod video 30gig, can i restore the original iPod firmware later ? like reseting the ipod?
02:43:02dewdudei mean, technically...the CD format has better sound than vinyl...16 bit has a lower noise's engineers that compress things.
02:43:04preglowNifl: yes
02:43:05gtksperti dont think so, hold on
02:43:20dewdudeNifl: the ipod firmware never goes anywhere. you can boot either it or rockbox.
02:44:03dewdudethe bootloader loads rockbox by default, but you can tell it to load the original ipod OS on don't have to reset the ipod to get that back.
02:44:11NiflIs it easy to remove rockbox, if i dont like it or something ? cuz i wanna try it
02:44:21dewdudesure. if you can follow directions.
02:44:35gtkspertpreglow: nope, it works fine on the other bands...
02:44:38dewdudeit's as easy as a single command at a command prompt.
02:44:44gtkspertanyway, talk to you guys later on...
02:45:44x1jmpCan someone explain me how to use playback_control with menu_run?
02:45:48preglowmarkun: that doesn't exactly point it out as one of those errors
02:45:55preglowthe ls band is the most problematic ones in terms of dsp
02:47:08dewdudeit also lacks the boom that putting band 2 at 60hz has.
02:47:41preglowthat's because putting a peaking filter at 60hz will give you the specified gain at 60
02:47:54preglowputting the ls filter at 60hz will give you the ls gain/2 at 60 hz
02:48:11NiflDoes rockbox only install stuff on the HD ?
02:48:19dewdudeyeah, i know. the only problem i've had with the eq is that LS band having no real oomph..but being quite literally a shelf filter
02:48:27preglowso pk filter at 12 db gain at 60 hz gives you 12 gain at 60 hz, ls filter at 60hz with 12 gain gives you 6 db gain at 60 hz
02:48:40preglowdewdude: which is what it's supposed to be? :-)
02:49:14dewdudepreglow, that's true..but it also means i have to use a band to add the oomph i like....i solved all my problems....i'm comboing the hardware and software EQ's
02:49:19dewdudesounds beter than i imagined
02:49:36preglowthere probably are better ways to implement the lowshelf filter, but i'm not skilled enough to know
02:49:39dewdudeNifl, yes, rockbox is installed to the HD..none of the code screened into the chips is affected
02:49:53preglowas it is, the rockbox eq filters should match pretty much all the other standard eqs in use out there
02:49:55dewdudedude, i'm only a "slide adjuster" according to my friend
02:49:57preglowdigital, that is
02:50:17dewdudeheh, if i had 12 or 13 bands, i could do magic with it.
02:50:23dewdudei mean, it's an excellent EQ.
02:50:25preglowif only because i use the basic filters they do, and indeed, probably even the same source formulas
02:50:43preglowthat rbj eq cookbook has gotten very popular...
02:50:55luckzShaid: around?
02:50:57preglowi might implement that many bands one day, but that will be as part of a dedicated plugin
02:51:11luckzhow did looking at the scrobbling stuff go?
02:51:15dewdudepreglow, excellent..keep me informed, i'd love to test it for ya.
02:51:16ShaidI had a look
02:51:17 Join SirFunk [0] (
02:51:26ShaidI need you to describe your issue in a little more detail for me :)
02:51:33preglowdewdude: robert bristow-johnson, the guy who published that particular doc
02:51:40Shaiddoes it happen with any track?
02:51:48preglowyou can google for rbj eq cookbook and see the source material
02:51:55dewdudeahhh....i don't code or even keep up with who's writing DSP code.
02:52:04dewdudei just...use the stuff
02:52:04luckzit seems to happen with every track. it just always declares a track skipped if you skip at any point, even if there's only a few seconds of silence/audience clapping left.
02:52:09preglownah, i just mentioned it because it's pretty well known
02:52:28luckzas far as I know specs say it should declare something played after 2m/50%
02:52:45 Quit m0s (Read error: 110 (Connection timed out))
02:52:54NiflDoes rockbox support the lyric displaying ? like the normal firmware does
02:52:59luckzother semi-issue is that it declares something played if I shut off the machine during playback, even in the first second of playback
02:53:03preglowNifl: no
02:53:04dewdudenot to my knowledge
02:53:32Shaidyeah, I think thereís a patch for that one on the tracker
02:53:38Shaidso Iíll patch to that, and then work from there.
02:53:40 Join saratoga [0] (i=98039a38@gateway/web/cgi-irc/
02:53:47luckzthat is 'fixed' by it not re-scrobbling when you resume playback
02:54:51 Part pixelma
02:55:08dewdudepreglow: but, back to the EQ stuff, aside from me personally not liking the LS...i do have to say it is a good software EQ, which is a switch from most players. of judgement is based on my Sansui analog EQ
02:55:13 Quit Nifl ()
02:55:39dewdudewow, i guess he wanted every every lame ipod feature or nothing.
02:56:20 Part JustJohnny68
02:56:25preglowonly filter i like is the peaking ones
02:56:32preglowbut that's because i ever only use the eq to play around habit from my younger days, i do like a tad more bass in playback than what's usually mastered in.
02:56:57preglownothing beats high q peaking filters in that regard
02:57:09luckzwell, it's a good argument for rockbox if it does all the OF does and more.
02:57:25dewdudebut my Shure's have a different sound than my sennheisers, so most of my EQ work is reflattening them out
02:57:29preglowluckz: sure, but who the hell wants to code lyrics support
02:57:37luckzin the case of the X5, it doesn't have the bass and whatever things. yes, I know why.
02:57:44preglowthe bass?
02:58:02luckz"moar bass", "less bass", "WOW bass (tm)"
02:58:07preglowi'm working on that, i'll be commiting bass/treble for x5 during the next couple of days, probably
02:58:10luckzand whatever else there was.
02:58:17preglownot wow bass, but you'll have treble/bass boost/cut support
02:58:21luckzdewdude: what shures now?
02:58:27dewdudeluckz, e3c
02:58:50dewdudeheh, thought they were something cheaper or something more expensive?
02:59:12luckzwell, I have the E4s and any tips on how to deal with them would be welcome. can never hurt to get some advice.
02:59:32 Quit austriancoder ()
02:59:35dewdudewell, it's mostly the mid range that's out of whack compared to my senny's
02:59:50scorchewhich senns?
03:00:02dewdudeslightly modified
03:00:04scorcheheh...than i cant comment
03:00:17dewdudeheh, why?
03:00:26 Part tyler_d ("Kopete 0.12.2 :")
03:00:30scorcheall i have experience with are the high end ones
03:00:45dewdudeoh, i've used the higher end ones
03:00:48ShaidIíve got a pair of those!
03:00:55dewdudei've used the orpheus system before
03:00:56Shaidblu-tack mod? :P
03:01:09 Join webguest75 [0] (i=46f570ef@gateway/web/cgi-irc/
03:01:11scorchewell, this is getting a bit off-topic anyway
03:01:18dewdudeyeah, sorry.
03:01:25*dewdude disappears
03:01:32webguest75hey, can someone help anwser some questions that i have about rockbox?
03:01:54luckzmy sennheisers are the HD580
03:02:18scorchenice....if they are amped
03:02:21markunwebguest75: sure, just ask
03:02:46webguest75i am thinking of installling rockbox on my ipod, but i am curious. will it whipe my music?, or more importantly, my files that are stored on it
03:02:55dewdudewebguest75, no.
03:03:08dewdudehowever, rockbox by default won't see your music that's been loaded with itunes
03:03:20dewdudeand..itunes can't see music loaded for rockbox
03:03:22Shaidwebguest75: which ipod?
03:03:33Shaidcause uh, if itís a 80g it wont work.
03:03:35webguest75well, the 5G 80gig
03:03:40dewdudewon't work.
03:03:44webguest75yea, i read that
03:03:57*Shaid is psychic.
03:03:58 Join pearldiver [0] (
03:03:58webguest75but wont it also somewhat work?
03:04:13Shaidif by somewhat you mean doesnít boot, yeah.
03:04:13dewdudei don't believe it works...period
03:04:21 Quit Mouser_X (Nick collision from services.)
03:04:22webguest75ahh, bummer
03:04:28dewdudeit's hard to load an os when you can't see the hard drive
03:04:29Shaiditís only a matter of time
03:04:30 Join Mouser_X [0] (
03:04:34luckz04:34:20 < BornToFrag> ok so i got my friend's ipodvideo to install rockbox 04:34:47 < BornToFrag> but then it got froze on the bootloader 04:35:48 < BornToFrag> TOSHIBA MK8010GAH
03:04:47luckz04:35:58 < Soap> 80 GB ipods are not supported. 04:36:04 < Soap> they don't work at this time. 04:36:18 < scorche> uninstall rockbox
03:04:51luckzokay? :p
03:05:14linuxstbdewdude: Rockbox can see itunes-loaded music fine in database mode.
03:05:18luckz(we had all that 22.5 hours ago)
03:05:36dewdudelinuxstb, yes, however, i said by default because someone installing rockbox should read it in the manual.
03:05:44webguest75well, guess ill just have to wait; thats kinda dissapointing though, since neither rockbox, nor iPL will work on the 80s
03:06:11dewdudeand if someone can't understand that the qualifier "by default" means that it still can, then, i'm sorry about thier stupidity
03:06:18linuxstbwebguest75: The reason's known, it's just a matter of time for Rockbox to be adjusted to it.
03:06:37webguest75linuxstb, and what would that reason be??
03:06:58 Join Plouj [0] (
03:07:11Ploujis what I need to download if I want to daily update my H320's firmware?
03:07:32linuxstbdewdude: "by default" to me means you need to do special things, and I don't consider the database special - it's one of the two standard ways Rockbox lets you browse your files.
03:08:11dewdudetrue...however it has to be initalized and enabled...but..since i don't want beef with you, next time i have to explain to someone, i'll make spencial mention of database mode
03:10:04linuxstbwebguest75: The disk in the 80GB ipod video can only be read two sectors at a time, compared to one sector for every other disk.
03:11:01 Quit perldiver (Read error: 110 (Connection timed out))
03:11:24webguest75well, i appericiate your guys help, thanks
03:11:54dewdudeno problem...keep an eye on the rockbox page though
03:12:10dewdudewhen it happens, then you can give it a try
03:12:18 Quit webguest75 ("CGI:IRC (EOF)")
03:14:40*preglow falls asleep
03:16:12markunwow, it's getting late..
03:18:30***Saving seen data "./dancer.seen"
03:19:26 Quit saratoga ("CGI:IRC")
03:20:48 Quit Insectoid (Read error: 104 (Connection reset by peer))
03:21:15 Join Thundercloud [0] (n=thunderc@
03:22:31 Part fasmaie
03:24:40luckzI need occupation of some kind.
03:25:01luckzI'm running a whole 3 apps. IE, so I don't have to restore my FF session. fb2k for music. putty for irc.
03:28:44 Join Thundercloud_ [0] (n=thunderc@
03:29:53Shaidputty does irc now?
03:30:00Shaidor youíve sshd somewhere for it
03:32:30luckzbut 90% of what I do with putty is irc
03:32:32 Quit entheh ("^~")
03:32:35 Quit decayedcell (Read error: 113 (No route to host))
03:37:04 Join decayedcell [0] (
03:42:31 Quit printfXh4 (Remote closed the connection)
03:43:25 Join printfXh4 [0] (
03:49:56 Quit Thundercloud (Connection timed out)
03:50:07 Quit Mouser_X (Nick collision from services.)
03:50:13 Join Mouser_X [0] (
03:53:42 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
03:54:36Ploujwhat is the rockbox.iriver file inside a rockbox build zip archive?
03:55:16luckzpossibly the actual unit of WhatMakesItWork(tm)
03:58:52 Quit decayedcell (Read error: 113 (No route to host))
03:59:05 Join toffe [0] (
04:08:20 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
04:09:13 Join aliask [0] (
04:18:46 Quit Mouser_X (Nick collision from services.)
04:18:55 Join Mouser_X [0] (
04:24:18 Quit Arathis ("Bye, bye")
04:29:50 Join x1jmp_ [0] (
04:33:30 Quit Mouser_X (Nick collision from services.)
04:33:39 Join Mouser_X [0] (
04:36:39 Join webguest74 [0] (i=46f570ef@gateway/web/cgi-irc/
04:37:56webguest74while i relize that rockbox does not work alone on the 5G, will it work if i install iPL first?
04:38:28 Quit XavierGr ("One firmware to rule them all!")
04:38:37 Join Owner [0] (
04:39:11 Nick Owner is now known as jordanahotgay (
04:39:53Soapwebguest74: I assume you mean the 5th generation iPod? In particular the 80GB model?
04:40:04jordanahotgayso wat is up everyone
04:40:09Shaidit. wont. work.
04:40:17Shaidany way you try it, it wont work.
04:40:21jordanahotgayis there some gays in here
04:40:27winchesterdo rockbox and the rb text editor use the same source file for text input?
04:40:40webguest74well, i think i will try it, just have to back up my 60+gigs of music/files first, lol
04:40:52Shaidwebguest74: donít bother. it wont boot.
04:40:59Shaidunless you really want to see the error messages.
04:41:19 Nick jordanahotgay is now known as jordanahotguy (
04:41:41Soapjordanahotguy: this is a discussion channel for the Rockbox audio player firmware.
04:42:12webguest74well, i think i found a unofficial way to install iPL on a 5G, and they also discussed rockbox, so i think ill give it a shot, sounds like it should work
04:42:21Soapit WON'T
04:42:28SoapRockbox is its own program.
04:42:30Shaidif youír referring to the bit on the ipl wiki, thatís for 30gs.
04:42:36SoapIt has nothing to do with ipod linux
04:42:40Soapthe two do not run at the same time,
04:42:53scorchewe are the ones who make the you not believe us for some odd reason
04:42:54 Join decayedcell [0] (i=3ba74f6a@gateway/web/cgi-irc/
04:43:05webguest74no, not the ipl wiki
04:43:20SoapIf or not IPL runs on the 80GB is completely ilrelevant to Rockbox's _known_ issues with the 80GB.
04:43:21winchesterhello? is there some main module for text input or do the plugins that need it code their own?
04:43:22 Join decayedcell_ [0] (i=3ba74f6a@gateway/web/cgi-irc/
04:43:36webguest74i kno, but since in order to get iPL to work i have to reformat,
04:43:48Soapforget IPL for a moment, webguest74.
04:43:50webguest74which could solve the issue with rockbox
04:43:55SoapIPL has NOTHING to do with Rockbox.
04:44:02scorchewinchester: main
04:44:03webguest74its forgotten
04:44:08Soapand that is not "the issue with rockbox"
04:44:11winchesterwhich file is it in?
04:44:58webguest74i was told like 1hour ago when i came in here that the reason that rockbox dosent work on the 80 has to do with rockbox not being able to see the 80gig HD properly
04:45:03Soapthe issue is known, it is being worked on, and if it was something as simple as a magic reformat it would have been solved, and we would not tell people the 80GB does not work with Rockbox.
04:45:23Soapwebguest74: and that is a matter of the HD's firmware, not of its format.
04:46:04webguest74ok, and i believe that you are probably right, but if nothing else i willl try to put on rockbox when i hopefullly successfuly install iPL
04:46:29webguest74you are right
04:46:36 Join XavierGr [0] (
04:46:42webguest74...but like i said, while im at it, i will try
04:46:51Shaidbut WHY?
04:46:54SoapI don't think ipl's kernel will even work yet - but that's your wasted time, not mine.
04:46:56scorcheand you will not get it to boot
04:47:10SoapTheir bootloader works IIRC, but the kernel hasn't been modified yet as I recall.
04:47:11 Quit x1jmp (Read error: 110 (Connection timed out))
04:47:16ShaidDataGhost went on holiday, the 80g kernel isnít done yet.
04:48:14Soapso it is all a waste of time webguest74. Accept it. Unless you can rewrite the ATA code for us before the people with actual experience at embedded developement do...
04:48:23 Quit decayedcell ("CGI:IRC (Ping timeout)")
04:49:01webguest74, im not saying that I CAN, im saying that i will try anyway, and i would not consider the time i spend on that wasted at all
04:49:33 Quit jordanahotguy ("Chatzilla 0.9.77 [Firefox]")
04:49:42scorchebut there is no purpose...other than if your goal is to see errors that we have all seen
04:49:45 Join Vyrus001 [0] (
04:49:53 Quit Vyrus001 (Connection reset by peer)
04:50:06 Join Vyrus001 [0] (
04:50:08Soapit is wasted because you don't even understand what the problem is.
04:50:10 Join Thundercloud [0] (n=thunderc@
04:50:23SoapYou ignore people who try and tell you that you are barking up the wrong tree.
04:50:42winchestershould I bother with making this input for the 2g mini?
04:51:09webguest74im not ingnoring you guys, plus, this will be a small side project
04:51:09 Quit bonbonthejon (Read error: 104 (Connection reset by peer))
04:51:25scorcheprojects typically have a goal
04:51:32aliaskwebguest74: have fun with it
04:51:34scorchethat goal is typically not to fail
04:51:37SoapUnless you can code, the only thing your project will be is trying to reinvent the wheel.
04:51:51SoapThe problem is know, the solution is known, it just isn't done yet.
04:52:18winchesterspeaking of reinventing the wheel >_<....... should I bother?
04:52:29webguest74in a way, arent you guys who work on rockbox trying to reinvent the wheel?
04:52:38aliaskwinchester: personally, i wouldn't
04:52:45markunwebguest74: and improve the wheel!
04:52:53aliaskIt'll be fiddly at best, and the 5 key isn't touch
04:53:00Soapwinchester: 13 degrees of a 1.5" wheel - not much room per letter.
04:53:17Soap14 degrees.
04:53:19winchesteractually I've tried it... it's quiet enough
04:53:46Soapreally? that thin of a pie slice per letter works, dang I never would have thought.
04:53:52aliaskbe prepared to do predictive text as well :P
04:54:09winchesternot per letter
04:54:12winchester45 degrees
04:54:26winchesterimitating the phone keypad
04:54:48Soapwhat does the "1" position do?
04:54:50 Quit Vyrus001 (Client Quit)
04:55:05winchesterand if you hold it... a menu for special characters perhaps
04:55:23winchestermissing a 0... but hey... who needs it
04:56:08SoapI totally grok your desire to maintain the phone paradigm, but I think the fact that "1" can't be pressed, and the fact "5" must be pressed makes "5" the better choice for space.
04:56:11scorchewiththat reasoning, whoneeds the 1 key?
04:57:45Soapsee, "5" has to be pressed, while 2, 4, 6, 8 _can't_ be pressed w/o invoking other actions - I have a feeling that will lead to inadventant presses.
04:58:20winchestermeh..... easier to use than the current method
04:58:27Soapand where is delete?
04:58:31winchester(dw I will probably drop this 20% of the way there)
04:58:47winchesterdelete? >.> that's just pushing it
04:59:06scorcheand special characters
04:59:36winchesterspecial characters would be 1.... and delete.... yeah I didn't think this through
05:01:07 Join fejfighter [0] (
05:01:16winchesterno way to press left without 4 being called right?
05:01:20SoapYou need also some way to press Menu, Play, <<, >> w/o registering as a letter - and whatever method you use to prevent letter registration on a push (as opposed to a touch) must not delay registering touches so much that you have a laggy interface.
05:02:27winchesterok..... since there will be delay to allow for the char to be entered.... if left or right is pressed within that time the char will be cancelled and backspace/delete will be called
05:02:33winchesterhow's that?
05:02:51SoapI don't know. ;)
05:03:03winchesteri'll give it a shot and see how it goes
05:03:27 Quit Thundercloud (Remote closed the connection)
05:03:39SoapAgain, if you were willing to break the phone SMS paradigm a bit, you could have short touch of 2=A, medium length touch = B and long touch = C, so you don't need to tap three times.
05:03:57Soapnah, that's not a good idea.
05:04:03winchesterwhy not?
05:04:20Soap'cause if it was a good idea some phone would alread do it.
05:04:31markunwinchester: will you ignore the loaded keyboard file?
05:06:27winchesteradd a setting?
05:09:21 Quit MonkeyTamer ("Chatzilla 0.9.77 [Firefox]")
05:10:41winchesterhmm..... does the current method allow the use of different languages>
05:11:35winchesteranyway.... suppose i will just make it for myself =/
05:11:48aliaskpost it to flyspray at least
05:11:56aliaskothers may like it
05:12:18winchesterthat's if I get it to anywhere near usable
05:13:11SoapI don't know if you are familiar with Milikeys - the alternative text input program for PDAs...
05:13:51SoapIt involves mapping regions of the touchscreen for user-customised keyboards...
05:14:34Soapyou _could_ map 30 (or 50, or whatever) regions to different letters, require a stylus to use.
05:14:49Soapand have people use a sticker overlay...
05:15:35Soap96 regions eh? you could treat the scroll wheel as a curved piano keyboard!
05:15:37winchesterdon't want to overcomplicate it for myself..... as I mentioned... I am not a programmer...this is just for the hell of it
05:17:18 Quit Thundercloud_ (Read error: 110 (Connection timed out))
05:17:18 Join Obsys [0] (
05:18:21 Quit Mouser_X (Nick collision from services.)
05:18:30 Join Mouser_X [0] (
05:18:34***Saving seen data "./dancer.seen"
05:19:47 Quit daurnimator ("Cyas later...")
05:20:09x1jmp_How can I delete all files in a folder that don't belong to the svn tree?
05:21:37winchesterhow would I count the 'ticks'?
05:22:53winchester(or how would I run a loop for a cirtain amount of time.... assume I would count ticks for that...)
05:23:27 Quit webguest74 ("CGI:IRC (Ping timeout)")
05:24:07 Quit fejfighter (Read error: 60 (Operation timed out))
05:24:38x1jmp_winchester: Did you have a look on that wiki article
05:25:00winchesterI cirtanly didn't
05:25:02winchesterthanks =)
05:26:46x1jmp_np, I never used these functions though, I didn't know about them when I would have needed them...
05:28:29 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
05:40:05 Quit linuxstb (Read error: 113 (No route to host))
05:41:28 Quit rotator ()
05:42:03 Quit Obsys (Remote closed the connection)
05:42:31 Quit decayedcell_ ("CGI:IRC (EOF)")
05:42:41 Join Obsys [0] (
05:47:35Mikachulinuxstb_: you already have a flash dump from nano?
05:51:49 Quit Obsys (Remote closed the connection)
05:56:48ShaidI think he got a couple last night
05:59:16 Join unwell_13 [0] (
05:59:27unwell_13easy installation of themes?
06:05:53 Quit Mouser_X (Nick collision from services.)
06:06:02 Join Mouser_X [0] (
06:07:30 Quit winchester ("Leaving")
06:08:12 Join bonbonthejon [0] (
06:11:24 Join Obsys [0] (
06:13:17 Join winchester [0] (
06:13:41winchesterhey..... what do I need to #include to use current_tick?
06:14:43winchesteror...... what am I missing that's not letting me use it?
06:14:56x1jmp_are writing a plugin?
06:15:01winchester"undefined reference to `current_tick'"
06:15:02 Nick x1jmp_ is now known as x1jmp (
06:15:15winchesterright now, yeah.
06:15:39Mikachu*rb->current_tick if your api pointer is rb
06:15:47winchestertouche >_<
06:17:26x1jmpmhhh... how do I update from rockbox svn, but commit to a local svn dir?
06:17:28 Quit Nibbier (Read error: 110 (Connection timed out))
06:17:56 Join webguest33 [0] (i=44455bef@gateway/web/cgi-irc/
06:18:12 Join Nibbier [0] (
06:18:24webguest33if i instal rockbox all i do is use the ipod disk to reinstal the ipod firmware right?
06:19:52webguest33sry for sayin hi so many times this is laggin or soemthing
06:21:38 Quit webguest33 (Client Quit)
06:32:11 Quit Kasperle (Read error: 104 (Connection reset by peer))
06:36:09 Join Kasperle [0] (
06:36:45 Join decayedcell [0] (i=3ba74f6a@gateway/web/cgi-irc/
06:40:25winchesterHey, is anyone on who knows what it is I was doing ( the scrollwheel keypad thing)?
06:40:32winchesterneed a bit of advice
06:41:22Mikachuwhat are you doing?
06:42:21winchesterwritting the function to proccess which character to return: char getchar (int key);
06:42:37winchesterkey being a number corresponding to teh 1-9 on the keypad
06:42:38Mikachuand less specifically?
06:43:07Mikachuyou mean like this?
06:43:07winchesterI think the way I am about to go about it is stupid.... wondering what the smart way is
06:43:17winchesterbecause what I would do is use a whole load of switches
06:44:04winchesterkind of
06:44:21winchesterthough that looks better than what I was thinking
06:44:23Mikachui've made a plugin that does 5165
06:44:31Mikachubut it's sort of crappy
06:44:40Mikachuyou can have a look if you want
06:44:52winchesterwhere can I find it?
06:45:37Mikachuyou'll need to comment out the set_trigger_delta calls since they don't exist
06:46:00winchesterok thanks
06:46:25Mikachuit really sucks with the delta at 10 though
06:47:45winchesterI don't really understand some of it
06:47:54winchesterthe alphabet is a matrix right?
06:48:30winchester(not so great with c.... just doing this for the sake of doing something)
06:48:56Mikachuif you look at the png in the tracker, where you put down your finger selects the honeycomb, and where you release selects the letter
06:51:38winchestergah..... don't understand the use of matrices for char mapping
06:52:03Mikachufirst one is the line, second is the column..
06:52:14winchesterwell yeah I get that much
06:52:36Mikachuit's easier to use a custom layout from a text file this way
06:52:55 Join menaztricks [0] (i=44e426c8@gateway/web/cgi-irc/
06:53:07Mikachualso, i know the drawline function is absolutely retarded :)
06:53:08winchesterI will just stare at the code until I get it
06:53:30 Quit Mouser_X (Nick collision from services.)
06:53:39 Join Mouser_X [0] (
06:54:02menaztrickshello there
06:56:22 Part toffe
06:56:31winchesterok I think thank helps and I can go on.... thanks Mikachu
06:58:33menaztrickscan rockbox add a search feature to my ipod photo?
07:04:45amiconnMikachu: What's wrong with drawline?
07:18:35***Saving seen data "./dancer.seen"
07:19:45 Quit menaztricks ("CGI:IRC (EOF)")
07:26:08Mikachuamiconn: it feels a little kludgy
07:27:56Mikachuamiconn: i'm referring to the one in my plugin
07:30:22winchestergod..... I am a horrible coder..... I get lost in my own code so easily..... and it's pretty short
07:30:33Mikachuwrite more comments
07:30:41winchesterthat's the plan
07:34:16amiconnMikachu: Why don't you just use the one in the core?
07:34:49Mikachuthe name is poorly chosen, it doesn't draw a line :)
07:35:08Mikachuin fact it draws 8 letters in a circle
07:35:53amiconnAha, hmm
07:36:04amiconnReally misleading name then
07:38:16 Quit Mouser_X (Nick collision from services.)
07:38:25 Join Mouser_X [0] (
07:38:55 Quit x1jmp (Remote closed the connection)
07:40:29winchesterwhich ipods don't support the touch scroll wheel thing?
07:40:44winchester1g mini and earlier?
07:41:04Mikachui think they all do?
07:41:20*amiconn doesn't fully understand what you mean
07:41:47winchesterthe scroll wheels which have the 96 touch spot things
07:41:52winchesterand nuh I know the 1g mini doesn't
07:41:57amiconnAll ipods have a wheel
07:42:17amiconn...and all of them are touch wheels, except the 1st gen ipod
07:42:37Mikachuwinchester: heh, i just realized my plugin won't work for you, BUTTON_TOUCH is not in svn :)
07:43:12amiconnBut I don't know whether all of them allow to read the touched position, if that's what you mean
07:43:16winchesterMikachu: nuh my code is completly different and for different targets.... I was just wondering how to get the alphabet right
07:43:29Mikachuright, just in case you wanted to try it
07:43:38winchesteryeah.... which ones don't let you read the touched positions?
07:43:53 Join combrains [0] (
07:44:05Mikachuwinchester: you just have to check #ifdef HAVE_WHEEL_POSITION
07:44:39Mikachuconfig-ipod4g.h:#define HAVE_WHEEL_POSITION config-ipodcolor.h:#define HAVE_WHEEL_POSITION config-ipodmini2g.h:#define HAVE_WHEEL_POSITION config-ipodnano.h:#define HAVE_WHEEL_POSITION config-ipodvideo.h:#define HAVE_WHEEL_POSITION
07:44:46amiconnI *think* 3rd gen and earlier, and the mini 1st gen
07:44:50winchesterah.... that would be the.... simpler way
07:45:47Mikachuyou currently have to poll the driver for the wheel position though
07:46:04winchesterwhat does that mean?
07:46:48amiconnMikachu: I think the touchwheel ipods could have an extra function in the button driver that switches between scroll events and position events
07:46:55winchesterI have got past the getting of wheel position.... it's in the api and all
07:47:31Mikachuamiconn: possibly, i just added a function for setting the delta value to 1, the scroll events already send the absolute position too
07:47:42Mikachuand also added sending events for touching / stopping touching
07:48:16amiconnMakes sense
07:48:19Mikachui called the other one BUTTON_UNTOUCH
07:48:35amiconnWhy not just BUTTON_TOUCH|BUTTON_REL
07:48:44Mikachui guess i didn't think that far
07:52:21winchestercan't find anything for BUTTON_UNTOUCH in button-clickwheel.c
07:52:37winchesteris it ok to just do while wheel_status != -1?
07:53:48winchesteroh wait nvrm..... that's not what I am doing.
07:54:11Mikachuthat's polling, but you'll want to sleep some time between each check
07:55:22winchesteryeah 'course
07:55:41winchesterah ok..... didn't know what polling was
07:55:47winchesternot up to scratch with the lingo
07:55:57Mikachuthe "opposite" is events
07:56:37winchesterI don't even know how events work in C >.>
07:56:46winchestereverything is just do while or do until for me
07:57:04Mikachuwell, when you call button_get you are waiting for an event
07:57:47winchesteroh...... so it doesn't go on to the next step if you have false as the argument?
07:58:11Mikachui don't remember if false or true makes it wait, but it doesn't return until the user presses a button in that case
07:59:19winchesterthat seems to be buggy
07:59:46Mikachuif you want to wait for the user to press a button it's pretty handy
08:00:37winchesterit ignores the first few keypresses sometimes though
08:00:53Mikachuah, you mean an actual bug, i thought you meant the concept
08:01:05Mikachuit sometimes doesn't work right if you just turned off hold
08:01:33winchesteryeah.... or just turned on the player
08:02:12winchesterand I think even if you just leave it for a while
08:06:00 Join Drumr [0] (
08:06:18winchesterdo you need to update the lcd after a splash?
08:06:55Drumranyone here on a 5g ipod? if so, what build do you recomend, if not the latest official build?
08:07:02 Join Entasis_ [0] (
08:08:33 Join puff [0] (i=puff@
08:08:41puffEvening. Anybody home?
08:08:52puffI have an 8gb ipod nano.
08:08:59Mikachuno chance, sorry
08:09:00decayedcellDrumr a build with the cop patch would be good
08:09:28puffThe first few months, I deliberately avoided rockbox because this is the first apple product I've owned since the Apple II+, and I wanted to get that "apple" feeling.
08:09:39puffOkay, so I've had it, so now, about rockbox... :-)
08:09:50Mikachupuff: it's a 2nd gen nano, isn't it?
08:09:54puffOh, yeah, come to think of it, IRRC rockbox didn't support mine.
08:10:08Mikachuit doesn't
08:10:17puffArgh. Okay, back to the ol' apple interface.
08:11:01puffI'm having some sort of werd problem, I can generally put music on the nano (though right now amarok and gtkpod are fighting) but nothing seems to manage to get a playlist onto the nano.
08:11:12puffThough maybe it's something obvious and I'm just "doing it wrong".
08:11:23Mikachuthat's not likely something anyone here will help you with
08:11:40 Join _Veseliq_ [0] (
08:11:45puffMother of... encrypted firmware?
08:11:49puffDamn, now that's annoying.
08:12:11decayedcelldon't worry it'll only be some time before someone figures out the encryption
08:12:22puffMikachu: Okay, just asking on the off chance. I'll be back by when rockbox supports my nano.
08:12:38puffdecayedcell: entropy, gotta love it.
08:12:43puffCiao folks.
08:12:46 Part puff ("ERC Version 5.0.2 $Revision: 1.726.2.11 $ (IRC client for Emacs)")
08:20:01winchesterkeymap [4][9]
08:20:12winchesteris that for 0 to 4 and 0 to 9?
08:20:20amiconnwinchester: splash updates automagically
08:20:21winchesteror 1 to 4 and 0 to 9?
08:20:29winchester*1 to 9
08:20:33amiconnNo it's 0 to 3 and 0 to 8
08:20:34winchesterand thanks amiconn
08:20:43winchesteroh =/
08:23:46winchesterwil I get kicked/muted for pasting 10 lines, should I use pastebin?
08:24:21scorchetheonly way you get kicked/muted is if an op does it
08:24:26scorchebut pastebin is polite
08:24:38winchesterbut... it's pretty quiet here right now...
08:24:47Mikachuuse it anyway
08:24:49scorche\but pastebin is polite
08:25:14Mikachuif you paste in channel, it will scroll off the screen, and then you have to scroll up and down to talk about it
08:25:25winchesterok thanks
08:25:55winchesternot much to talk about.... did something wrong with the matrix.... hang on
08:28:04winchesterhmm.... think I figured it out
08:30:39winchesterOk, I didn't.
08:30:54winchesterit's a pointer now.... but I am still getting a lot of errors around there
08:35:39 Join funky [0] (n=repulse@unaffiliated/funky)
08:38:19 Quit lostnihilist (Read error: 110 (Connection timed out))
08:39:21 Join menaztricks [0] (i=44e426c8@gateway/web/cgi-irc/
08:40:35winchester>.> I missed a comma..... sorry.
08:42:22menaztricksi just installed frostbox on my ipod, but the text isnt displaying right on the wps
08:42:39scorchehave you installed the font pack?
08:42:55menaztrickssure did
08:43:12scorchedoes it say that it needs certain patches to work?
08:43:18menaztricksit shows the text on the left side of the screen, on top of the album art, instead of the right need the scrolling margins patch
08:43:57menaztrickspoint me in that direction?
08:44:09scorcheeither build your own binary with the patch, or use an unofficial build with that patch in it
08:46:14menaztricksgonna have to find an unoffocial
08:46:28menaztricksi was lookin at the frost box authors page
08:46:42menaztricksbut he only has downloads for video ipods
08:46:55menaztricksneed one for the photo version
08:47:09midkayis it a photo-compatible theme?
08:47:17midkayi doubt it, if he only has 5G builds.
08:48:06menaztricksgimme a sec to make sure i didnt goof
08:48:06 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:48:55winchesterhmm.... debuggin my code.... am I the only one who makes so many careless, stupid mistakes?
08:49:05menaztricksyup, its right
08:49:09menaztricksgot it from there
08:49:56midkayseems like you'll need to patch your own build to use it.
08:50:04 Join Rob2222 [0] (
08:50:42menaztricksare there instructions on the rockbox site for that
08:51:11scorchesee the SimpleGuideToCompiling wiki page
08:51:14midkayyeah, but it's not that easy if you aren't at all familiar with compiling/patching code.
08:51:54menaztrickslast time i compiled something was about 7yrs ago
08:52:22midkayat least you know what it means to 'compile' - that puts you above most people who come in asking how to do it :)
08:53:32menaztrickshaha, its a start
08:54:15midkayyep, check out that page on the wiki, and there should be one for patching too.. WorkingWithPatches
08:54:21Mikachuhm preglow, can you send the diskmode.ipod from your updated flash?
08:55:24scorchemidkay: the SGTC page deals with that as well
08:55:47midkayscorche: good for the SGTC page. i imagine that if we have a page *dedicated* to patching it might cover it in a bit more detail.
08:55:53menaztricksthanks, im gonna get crackin on that
08:56:14scorchemidkay: oh, how i missed talking to you =P
08:56:23menaztricksits crazy how useful rockbox is
08:56:33midkayscorche: we must maptest sometime soon!
08:56:44scorcheyou are working on that again?
08:56:50menaztricksim finding myself saying "why didnt apple wanna let me do this?"
08:57:08Mikachumenaztricks: you bought it anyway, didn't you?
08:57:16midkaymenaztricks: well, rockbox is the result of hundreds of peoples' work over the course of several years.. i don't think Apple's team had so much time or manpower :)
08:58:33scorchemidkay: oooo..progress on the thread
08:58:45winchesterbut... they had all the technical hardware info.
08:59:05midkayscorche: haha. right. i've done a lot of work since then, but yeah :)
09:00:00scorchelooks so weird to see it again...been a long time
09:00:23midkayscorche: we haven't played for way too long... that's the problem!
09:00:28menaztricksyeah, bought it anyway cause it was cooooool
09:00:41menaztricksplus, i really didnt realize how limited it was until i had it
09:00:46scorchemidkay: only because you have spent way too long hating my guts!
09:01:12midkayscorche: i don't hate you, i hate your.. your......... something. but it isn't you!
09:01:19menaztricksfinding rockbox saved me though, cause i was about to get an iriver
09:01:48scorchemidkay: yeah isnt me...that is what you said last time we broke up =)
09:02:04midkayhaha. broke up. :(
09:02:17midkayalright, i'll get back together with you and give it a.. third? try.
09:02:39scorchethe second one didnt last long...
09:03:42midkayi can satisfy you this time.
09:04:03menaztricksi think you 2 need some privacy
09:04:18midkayhaha, oh god!
09:17:45 Quit decayedcell ("CGI:IRC (EOF)")
09:18:38***Saving seen data "./dancer.seen"
09:21:13 Join Audioengineer [0] (
09:24:34 Quit Audioengineer (Client Quit)
09:29:38 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
09:31:34 Join atsea-145193 [0] (i=ariel@gateway/tor/x-1ce21b587b8c88ce)
09:33:38menaztrickswhat does database is not ready mean?
09:34:19 Join Ribs [0] (n=ribs@
09:34:38midkayit hasn't fully updated/committed (applied) itself.. maybe you should re-initialize it.
09:34:43midkayand don't reboot until the disk stops.
09:35:37winchesterhey..... last question, I promis....
09:35:45winchester"(.text+0xa8): undefined reference to `memcpy'" what does that error mean?
09:36:00Mikachuit means you forgot to write rb->memcpy instead of memcpy
09:36:22winchesteri never used memcpy
09:36:24Mikachugcc decided to use memcpy for something
09:36:33winchestergcc decided to
09:36:36 Quit Mouser_X (Nick collision from services.)
09:36:45 Join Mouser_X [0] (
09:36:51Mikachusomeone had this problem recently, i don't remember how they resolved it though
09:37:43midkayi got that once recently.. it involved the usage of a char..
09:37:47Mikachutry something like this near the top of your file
09:38:40Mikachustatic inline void *memcpy(void *dest, const void *src, size_t n) { rb->memcpy(dest, src, n); }
09:39:07Mikachuer, that should be "return rb->..."
09:39:32Mikachuif that doesn't work, drop the inline
09:40:07Mikachuand if that doesn't work, wait until someone who knows what they're doing answers :)
09:40:15 Quit combrains (Read error: 110 (Connection timed out))
09:41:49winchesterstatic void *memcpy(void *dest, const void *src, size_t n) { return rb->memcpy(dest, src, n); }
09:41:52winchesterthat worked
09:41:56winchesterwell... it compiles....
09:44:35winchesterand runs :D
09:45:54 Quit atsea-145 (Read error: 104 (Connection reset by peer))
09:46:16winchesterand i'll shut up.... as promised =)
09:47:23midkaywe're here 4 u!
09:53:45menaztricksim so happy
09:53:50menaztricksmy ipod has a search :)
09:57:12 Quit Mouser_X (Nick collision from services.)
09:57:21 Join Mouser_X [0] (
09:58:12menaztrickslast dumb question of the night
09:58:24menaztricksi figured out how to go back to the list of songs
09:58:33menaztrickswhen on the wps
09:58:51menaztricksbut how do i go back to wps nonw
09:58:58menaztricksif i dont pick a new song?
09:59:29winchesterpress play while in the list of songs.... I think
09:59:45midkaythat's right.. you might want to check out the manual, menaztricks.
10:00:05menaztricksthats what i was doing
10:01:09menaztricksoh wait
10:01:26menaztricksim used to using the select button as play
10:01:59midkayselect = select :)
10:02:56menaztrickspast 1am
10:02:59menaztricksmy mind shuts down
10:03:47menaztricks30more mins and ill be using the hold button as play
10:04:34winchesterhey.... what's a way to divide the scroll wheel into 8 parts (mathematically) without using a switch statement?
10:05:12winchesternvrm, got it
10:10:30Shaideasy :P
10:10:57Shaiddonít use modulus
10:12:31winchesterhmm.... I'll make a horrible engineer =(
10:13:16 Quit Mouser_X (Nick collision from services.)
10:13:18Shaidmodulus would give you a rather, uh, unusable breakup of the wheel.
10:13:25 Join Mouser_X [0] (
10:46:02 Join darkless [0] (
10:51:03 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)
10:51:13daurnimator!seen jdgordon
10:51:30daurnimatorany? - linuxstb_?
10:51:36daurnimatorwhen was JD here last?
10:51:51 Quit menaztricks ("CGI:IRC (EOF)")
10:55:42 Join redbreva [0] (
10:56:00 Join bluebrother [0] (i=CEKAIAJc@rockbox/staff/bluebrother)
10:58:53 Join nls [0] (
11:00:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
11:01:13 Quit BHSPitLappy (Remote closed the connection)
11:12:39 Quit Mouser_X (Nick collision from services.)
11:12:48 Join Mouser_X [0] (
11:14:39 Join entheh [0] (
11:16:09 Quit aliask ("Chatzilla 0.9.77 [Firefox]")
11:18:47***Saving seen data "./dancer.seen"
11:22:25 Quit bk (Read error: 104 (Connection reset by peer))
11:28:19 Join fejfighter [0] (
11:29:28 Join decayedcell [0] (i=3ba74f6a@gateway/web/cgi-irc/
11:34:56 Quit decayedcell ("CGI:IRC (Ping timeout)")
11:40:24 Join decayedcell [0] (i=3ba74f6a@gateway/web/cgi-irc/
11:41:10 Quit decayedcell (Client Quit)
11:42:18linuxstb_dan_a: Are you around?
11:44:01winchesterhey, how do you round up a number in C?
11:45:51linuxstb_integer divisions will always truncate - are you sure you need to round up?
11:46:12linuxstb_You want to avoid floating point unless it's absolutely necessary.
11:46:29winchesterif I was to divide an integer by 12 and then round it up?
11:47:07Entasis_(int + 11) / 12?
11:47:07 Quit printfXh4 (Remote closed the connection)
11:47:33Shaidisnít that what mod is for?
11:47:37linuxstb_(int + 6)/12 may do what you want.
11:47:38Shaidor the counterpart.
11:47:41*Shaid shrugs
11:48:00winchesterand could I use ceil (x/12) or should I avoid that?
11:48:08winchesterand use (int + 11) / 12?
11:48:15linuxstb_Avoid it. Rockbox probably doesn't even have a ceil() function.
11:48:51linuxstb_Do you want 0-5 to give 0, 6-17 to give 1, 18-29 to give 2 etc etc ?
11:49:16linuxstb_Then (int + 6)/12 will do it.
11:50:32linuxstb_Not if you want the mapping you just said "yup" to.
11:50:48Entasis_+11 will round up, +6 will round to nearest
11:50:59winchesteroh nuh I need to round up
11:51:26winchesterI haven't slept in quite a while.... got carried away..... tend to get slow.
11:51:59linuxstb_But if you just do (int / 12), then 0-11 will give 0, 12-23 will give 1 etc etc.
11:52:30winchesternuh.... need to specifically round up
11:52:38linuxstb_Just tell me how the mapping should work.
11:53:28winchester13/12 =2
11:53:40linuxstb_12/12 = ?
11:53:55winchesternot important
11:55:18linuxstb_Then either (int + 11)/12 or (int + 12)/12 depending if you want 12/12 to be 1 or 2.
11:55:39 Join printfXh4 [0] (
11:57:30winchesterworks like a charm
11:59:24linuxstb_Mikachu: Thanks for the offer, but yes, I already have two flash dumps from Nanos.
11:59:58 Quit Xerion_ (Read error: 60 (Operation timed out))
12:00:16Shaidfound anything interesting in them, linuxstb_?
12:00:31ShaidSeem to be images stored in a rather raw format.
12:00:39Shaidand lots of free space.
12:01:03linuxstb_You mean images as in bitmaps?
12:03:46winchesteris "int[2]={-1,5};" a valid line or do I need to make a loop to fill the array?
12:03:57linuxstb_The most interesting thing is that the disk mode code seems to be a standalone application that we can potentially launch with Rolo. This means that implementing our own disk mode code is no longer a pre-requisite for flashing on the ipods.
12:04:20 Join decayedcell [0] (i=3ba74f6a@gateway/web/cgi-irc/
12:05:30 Join powr-toc [0] (
12:05:53linuxstb_winchester: Does that array change, or is it constant?
12:07:17linuxstb_Then you should define it using something like "static const int myarray[] = {-1, 5, 1, 4, 5 };" (with as many numbers as you want in the brackets).
12:07:39winchesterwhat does static do?
12:08:00Slasherilinuxstb_: that is great to hear. How big is that diskmode code image?
12:08:06winchesterAs in.... what's the diff between static and constant......?
12:09:10linuxstb_Slasheri: About 185KB.
12:09:30Slasherilinuxstb_: ouch, that's huge..
12:09:46 Join inversions [0] (
12:10:20linuxstb_That's on the 5g. For some reason, it's larger on my ipod Color (250KB) and the mini2g (228KB). The Nano is only 175KB.
12:10:40linuxstb_A fair proportion of that is probably the graphics.
12:10:54 Join lee-qid [0] (
12:11:10linuxstb_But yes, it is huge.
12:11:42Shaidimages an in bitmaps, yes.
12:12:02Slasherihowever, it could be still possible to find some interesting stuff from it such as the display initialization sequence (unless bootloader does that before)
12:12:02Shaidhow big is it on the 4g greyscale?
12:13:02linuxstb_Slasheri: There's also the "diagnostics mode" application - that's only about 95KB.
12:13:39linuxstb_You enter diagnostics mode by holding SELECT+LEFT when booting.
12:14:13Slasherilinuxstb_: oh, that would be much more interesting then.. i will try out that soon
12:14:38Slasheriand you could rolo that application also?
12:15:27linuxstb_Not reliably - rolo doesn't do anything with the COP currently - it only "rolos" the main CPU...
12:16:07linuxstb_But if you use my "flashsplit.c" file, that gives you "diskmode.ipod" and "diagmode.ipod" files. You can then write these to your ipod's firmware partition with "ipodpatcher -wf diagmode.ipod", and that will work.
12:16:13Slasheribut somehow it would be nice to modify the code and rolo the modified binary
12:16:27Slasherinice =)
12:16:54linuxstb_Not useful in practice, but it shows that rolo should in theory be able to launch them.
12:18:15linuxstb_And the ipods use a similar SST flash chip to the h1x0 - I was comparing your iriver_flash.c code with the datasheet for my ipod Color's flash, and they seem the same.
12:18:58linuxstb_But I shouldn't go there yet....
12:21:29bluebrotherwhat's your current goal? Extract the disk mode handling from the OF to analyze it?
12:21:44bluebrotheror is there already something working in RB itself?
12:26:08linuxstb_I don't think we need to analyse the OF disk mode code - I think MrH's discovery that the USB hardware in the portalplayer chips is the same as the i.MX31 which has a public datasheet is enough.
12:26:55linuxstb_But currently we reboot into Apple's disk mode code. If we get rolo working, we can cut out the reboot to enter disk mode.
12:27:13bluebrotherah. Nice idea
12:27:33linuxstb_As we've discovered that the disk mode code is in fact just a standalone firmware which is copied from flash to RAM and executed like any other ipod firmware.
12:28:26 Quit Mouser_X (Read error: 110 (Connection timed out))
12:33:08Slasherilinuxstb_: hmm, a public datasheet, where? i have found only the hardware datasheet about the electrical characters of that i.MX31 CPU
12:33:34Slasheribut not a programming manual
12:36:10Slasherihmm, now i found something more interesting..
12:36:57 Quit Ribs ("eh eh ehhhh!")
12:37:59Slasherilinuxstb_: ok, i found it. Last time i didn't see that on the freescale site
12:38:38 Join evilx [0] (
12:39:24evilxhow do i remove rockbxo from an ipod nano? do i just delete the bootloader? or do i have to rewrite the bootpartition? i know the old firmware seems to be on there, because rockbox like crashed and it went back to the old stock one
12:40:57linuxstb_Running "ipodpatcher -d" (with a recent version of ipodpatcher) will remove the bootloader.
12:41:57Shaidand then you just delete .rockbox folder and the rockbox.ipod file
12:42:56linuxstb_Slasheri: There's also a USB driver in the Linux kernel for the same chipset.
12:43:44evilxok that whati thought thanks
12:44:04dan_alinuxstb_: I'm here now
12:44:39linuxstb_dan_a: I want to help get rolo working, and was wondering what stage you're at - i.e. what you've unsuccessfully tried so far?
12:45:48linuxstb_Also, have you created "diskmode.ipod" and "diagmode.ipod" files for testing? Maybe they are less fussy than apple_os.ipod
12:47:10dan_alinuxstb_: I was testing with first roloing the running kernel, and in the tests where that worked then I'd try apple_os.ipod
12:47:39 Join anathema [0] (
12:47:39 Quit decayedcell ("CGI:IRC (EOF)")
12:50:48dan_aMy basic plan is that the CPU sends a message to the COP to reload, the COP invalidates then disables its cache, maybe turns off mmapping (but I think that mmap is common across both cores) and then jumps to either 0x0 or 0x10000000.
12:51:19linuxstb_When would it jump 0x0?
12:52:40linuxstb_Are you doing the rolo stuff in the cop_main() function?
12:52:41dan_aIf it is woken before the CPU clears the MMAP regs
12:52:55dan_aFor the COP, yes
12:53:15linuxstb_Have you moved it to IRAM?
12:54:11dan_aI just thought of that while I was explaining what I'd been trying
12:55:13linuxstb_So did I...
12:55:56dan_aThe basic idea is here:
12:56:43linuxstb_Is that using the mailboxes?
12:56:55linuxstb_Or just an iram variable?
12:57:28dan_aMailboxes - it means I don't have to explicitly wake the COP
12:57:44linuxstb_And the messages are getting through?
12:58:05 Quit redbreva ("Chatzilla 0.9.77 [Firefox]")
12:59:31dan_aWhen I messaged the COP before doing any cache/mmap stuff on the CPU, and then waited for a reply from the COP I would get the reply if I sent it immediately
12:59:52dan_a(urg - that's horrible English! Sorry)
13:01:04evilxthe english language is a horrible problem, so 2 wrongs do make a right
13:01:21Shaidpoint taken
13:02:26evilxi think regex makes more since then the english language.
13:05:08 Quit inversions (Read error: 110 (Connection timed out))
13:05:12evilxmy spelling sucks tonight
13:05:22evilxill just shut up now
13:07:02linuxstb_dan_a: Can you successfully rolo into a version of Rockbox including the kernel_on_cop patch?
13:07:43 Quit Rob2222 ()
13:08:02dan_aI don't get the main thread running on the COP (as you would expect)
13:08:49linuxstb_So the COP is still definitely the issue?
13:09:35dan_aIt's still definitely an issue
13:09:49linuxstb_That's what I meant.
13:12:46 Join Rob2222 [0] (
13:17:46 Join MobiLex [0] (
13:18:04 Quit powr-toc (Read error: 104 (Connection reset by peer))
13:18:48***Saving seen data "./dancer.seen"
13:22:03 Part evilx ("Leaving")
13:23:31dan_aIt looks like the mailboxes weren't working
13:23:58 Quit MobiLex (Remote closed the connection)
13:25:33dan_aThis will rolo Rockbox, but not the Apple OS:
13:25:55linuxstb_With the COP threads still working ?
13:26:13dan_aI'm just doing a build to check that
13:27:30linuxstb_Doesn't cop_main() itself need to be in IRAM? The main rolo() code will overwrite it IIUC.
13:29:01dan_aAh - that will explain why I'm hanging at "Executing" - the CPU is never getting its reply
13:29:35dan_aWhen I tried that, I got a compile error. I'll try waking the COP before loading the image
13:29:58linuxstb_You can't "bl cop_main" - because it's further than 32MB away.
13:30:20linuxstb_I think something like "mov r0, =cop_main ; mov pc,r0" is needed.
13:31:49 Join w1ll14m [0] (
13:32:26 Quit fejfighter ()
13:32:35dan_aIn the long run, I want the main cop thread to kill itsself though, so having a separate function is better. It also means that when we get mailbox interrupts we won't need to make big changes
13:32:45 Quit funky ("Changing server")
13:33:08 Join funky [0] (n=repulse@unaffiliated/funky)
13:35:10linuxstb_Also, is the COP's stack in IRAM? That could also cause problems.
13:36:18dan_aI think I would have put it in the same area as teh main stack. Why would that be a problem?
13:37:57linuxstb_If cop_main() is calling another function, that may involve writing some values on the stack. If the stack is in SDRAM, then you could overwrite the firmware you're roloing.
13:39:02dan_aIt's in IRAM
13:39:24linuxstb_So that's OK for this purpose, but is probably a waste of IRAM if the main thread isn't doing anything.
13:39:35linuxstb_^main cop thread
13:40:54 Join datachild` [0] (
13:43:36preglowhow much do we need before we've got romboxes again now?
13:46:21 Quit datachild (Connection reset by peer)
13:50:02dan_aI've just Rolo'd a dual core version of Rockbox...
13:50:31dan_a...but not the OF
13:50:59linuxstb_Do you have the diagmode and diskmode images?
13:51:02 Quit printfXh4 (Remote closed the connection)
13:51:22 Quit Everybody| ("( :: NoNameScript 4.02 :: )")
13:51:23dan_aNot yet
13:51:58linuxstb_You're testing on a 4g greyscale?
13:52:05winchesterhey how come I can't get (rb->button_get(false) == BUTTON_SELECT) to be true?
13:52:45 Quit anathema (Read error: 113 (No route to host))
13:56:13dan_aDiagmode hangs at Executing
13:56:21winchesternvrm me.... i'll try my best first.
13:57:03dan_aDisk mode hangs at Executing
14:00:37dan_aThe patch is here:
14:01:03dan_awinchester: I'd help, but I don't know enough about the button code
14:02:09winchesterdan_a: don't worry it's something in my code I am overlooking.... thought it was a wrong #define... but nope
14:02:43preglowwhat's the cop state at rolo image entry?
14:03:49dan_apreglow: You mean when we reboot?
14:04:55preglowjust wondering what the cop does when you rolo
14:05:20preglowthe diskmode and diag mode images assume both cpu and cop enter at 0 at about the same time
14:05:52preglowpreferable cop first, i think
14:07:21dan_aThey jump at pretty close to the same time... or they would if I wasn't being Mr Stupid today and forgetting to wake the COP
14:10:07preglowdidn't your parents ever tell you not to hang out with that guy? he's bad for you
14:11:23 Join Mouser_X [0] (
14:11:52 Quit XavierGr (Nick collision from services.)
14:11:55 Join XavierGr [0] (
14:14:30*dan_a pays a visit to wise old Mr. Coffee
14:14:40preglownow there's a healthy companion
14:15:23linuxstb_preglow: Did you say that diagmode and diskmode rolo reliably for you on your Nano?
14:20:02Mikachuthey do on mine, at least both started when i tried
14:20:47dan_aDiagmode: RoLo'd OK here...
14:20:55dan_aAll the tests pass
14:21:37Soaplinuxstb_: feel free to ignore this ignorant question, but outside of the cool factor (and it does seem cool) what pontential advantages can arise from your newfound ability to split out flash "modules" and rolo them? What potential legal issues?
14:21:50 Join inversions [0] (
14:22:23 Join anathema [0] (
14:23:01dan_adiskmode: RoLo'd, doesn't detect a connection
14:23:32Mikachui can add that i could mount and ls on my diskmode, i haven't tried transferring yet
14:23:53Mikachui first rolod, then plugged in
14:24:18Mikachu(and i'm not using the cop patch atm)
14:24:55dan_aApple OS: Stays on "executing" for a few seconds, then reboots
14:25:40linuxstb_Soap: The aim won't be to split them out to files, but to rolo them directly from the flash - i.e. copy from flash to RAM and execute.
14:26:15 Join Nico_P [0] (
14:26:32Soapahh, saving one reboot in the Rockbox->Diskmode->Rockbox cycle?
14:26:38Mikachui like running them from files, that means i can get the updated diskmode without having to actually flash
14:27:00linuxstb_dan_a: Yes, I think Rockbox's usb initialisation/detection breaks the diskmode code. You could test it by disabling Rockbox's usb code.
14:27:20linuxstb_Mikachu: You need to flash in order to extract them...
14:27:29Soapor could you binary patch so that Diskmode jumps to Rockbox code instead of rebooting?
14:27:37Mikachui am hoping i can get you or preglow to send it :)
14:28:21linuxstb_Soap: That's possible, but is a greyer legal area and the time would probably be better spent working on the native USB driver.
14:29:23SoapThank you linuxstb_, I was just curious about all this while reading the logs.
14:30:16Mikachuthe flash seems to be 1MB, would it be possible to fit a bootloader and rockbox in there?
14:30:53Mikachunot that it should do much difference on a nano i guess, no hd to spin up
14:31:03linuxstb_Yes, that would be the long-term aim.
14:31:25Shaidwouldnít that break OF compatibility?
14:31:45Mikachuno, but it would get rid of the apple logo :)
14:31:47linuxstb_Maybe, maybe not. But it's not a concern.
14:31:59desowincurrently all codecs use the same id3 code ?
14:32:06Soapit would be funny to have to launch OF from disk while Rockbox is in flash.
14:32:07jhMikeSpreglow: I don't get any boost any more turning on all the fancy dsp stages...what'd I do wrong? ;)
14:32:44linuxstb_Flashing would always be an optional way to install Rockbox, so if users wanted to keep OF compatibility, they simply wouldn't flash (if indeed flashing broke the OF).
14:33:03Mikachuit would help on nano since the apple flash always loads the whole apple os from the firmware partition...
14:33:17linuxstb_You can easily stop it doing that.
14:33:35SoapMikachu: someone could overwrite the internal apple logo now if they wanted to, could they not? Or does the flash check its own checksum/hash?
14:34:01linuxstb_Mikachu: ipodpatcher -rf apple_os.bin ; cp apple_os.bin /mnt/ipod/ ; ipodpatcher -wf bootloader-ipodnano.ipod
14:34:27linuxstb_Soap: I've found one checksum related to the logo, but I don't know if there are more.
14:35:11linuxstb_I'm not sure I would want to risk it, but replacing the Apple logo looks relatively easy.
14:35:27Mikachumaybe it has a checksum for the whole flash image too
14:35:45Mikachuthe question is what it would do when the checksum fails?
14:35:50SoapNo kidding. (on them checksumming a bitmap) - I'll stop pulling you into tangents. Thanks again for the info.
14:36:11 Join fasmaie [0] (
14:36:16 Quit datachild` ("the dismal monotony of dissapointments")
14:36:45linuxstb_Mikachu: That's my thought as well. The only reason I can think of for a checksum on the whole flash would be to prevent tempering. If it's corrupt, then there's nothing the bootloader can do....
14:37:19Mikachufunny, roloing the diskmode.ipod didn't let me mount now
14:37:20linuxstb_Although in diagnostics mode, there's a "flash checksum" option which spits out a 16-bit checksum.
14:37:23Shaidhow are you planning to reflash the flash?
14:37:41linuxstb_The flash is a standard component with datasheets. It's easy to flash.
14:37:50linuxstb_But obviously very risky...
14:38:07Mikachuit doesn't let you flash by just writing to the address you read it from, right?
14:38:17linuxstb_You just write a magic sequence of values to magic addresses in the flash, and that allows you to erase/write.
14:38:17Shaidhypothetically: if you modify it, and it breaks your ipod and stops it booting, will it be easily fixable?
14:38:58linuxstb_The ipod's flash is very similar to the flash in the h1x0 - look at the apps/plugins/iriver_flash.c plugin for how it works.
14:39:22linuxstb_ - the writes to addresses 0x5555 and 0x2AAA
14:40:08linuxstb_Shaid: No, it will be bricked.
14:40:23Shaidthatís what I thought./
14:40:30Shaidgoing to be a fun thing to test.
14:40:33preglowlinuxstb_: realibly indeed
14:40:35preglowjhMikeS: dAMN
14:40:43preglowjhMikeS: what codec? including eq?
14:41:25 Join datachild [0] (
14:41:51Mikachusomeone broke compiling the ipod bootloader with multivolume support
14:41:59Mikachu(possibly me :)
14:42:25jhMikeSpreglow: I said everything (expect for dither though since it's now it's own output)
14:42:38 Join lostnihilist [0] (
14:42:41jhMikeSmp3 and flac
14:43:17jhMikeSvorbis boosts of course but only an additional 3-5% with everything on
14:43:40preglowthat's pretty tight
14:44:07jhMikeSdon't laught at all the emac always proved to have alternative positive uses
14:44:46jhMikeSbut pregain was still the c stuff
14:45:13JdGordonwas the broken gcc 4.1.1 server fixed?
14:45:17dan_aMikachu: Why would people comile it with multi-volume?
14:45:18preglowemac abuse as in i'm going to have hell porting the code to arm?
14:45:20preglowor good emac abuse? :)
14:45:37 Join printfXh4 [0] (
14:46:02Mikachudan_a: i would because i shrinked the firmware partition, but i don't want to risk overwriting the start of my main partition if the apple firmware decides to hibernate
14:46:09 Quit inversions (Read error: 113 (No route to host))
14:46:41Mikachuputting IF_MV(0) in disk_init() seems to fix it
14:49:16jhMikeSpreglow: not really...good abuse. It rules for clamping and scaling in the final output over anything else. And help do line bursts to the pcm buffer.
14:49:53Mikachulinuxstb_: you wrote cp apple_os.bin /mnt/ipod, i don't have to do anything to make it .ipod first?
14:50:33linuxstb_Mikachu: Sorry, I mistyped - just rename it. The -rf command will write it in .ipod format.
14:50:58jhMikeSI switch it to integer mode for that and multiply everything by 1 << (16 - scale) and then take the upper 16 bits
14:51:13Mikachuwow yeah, that was faster
14:52:24preglowjhMikeS: ahaha, so that's asm too
14:52:37Mikachuhrm, how long should it take to rolo the OF?
14:52:49preglowjhMikeS: when're you going to commit?
14:53:04linuxstb_Rolo'ing it won't work, but the bootloader can successfully load it - the bootloader looks for apple_os.ipod before looking for an "internal" copy.
14:53:11Mikachuah right
14:54:18jhMikeSpreglow: got a big lump now...and will continue some more after that. I added a setup to make sending dsp params to the asm routines easier as well.
14:54:25jhMikeSso soon
14:54:45preglowall the dsp plus a codec with no boost is quite insane, though
14:55:05Mikachufunky, i get the No RetailOS detected message, but with a light blue background
14:55:28linuxstb_So it's not working?
14:55:33Mikachunot as such, no
14:55:42Mikachubut it's probably my fault somehow
14:55:47*jhMikeS _is_ insane though
14:56:26linuxstb_Mikachu: You'll need a pretty recent version of the bootloader.
14:57:07preglowi don't think we'll get such a boost on arm, though, but we'll see
14:57:14Mikachui compiled it just now
14:57:26 Quit Mouser_X (Nick collision from services.)
14:57:35 Join Mouser_X [0] (
14:57:35jhMikeStrying several variants of asm rountines and testing with and without takes some time to get through :)
14:58:37 Quit fasmaie ()
14:58:51jhMikeSpreglow: Given all that, can you guestimate or do I have to wait for results?
14:58:57 Join Hub441 [0] (
14:59:19Hub441does rockbox play ogg on a g4 greyscale ipod?
14:59:34linuxstb_Ogg vorbis, yes.
14:59:52preglowjhMikeS: i've given up estimating for portalplayer chips, they always surprise me in some way
15:01:08Hub441linuxstb_: i havent seen rockbox live so far, but is it usable as the original firmware? i gave ipodlinux a try but it's just unusable (menue not available while playing a song, long silence between two songs..)
15:01:10jhMikeSpreglow: I should say the down side is those output routines are bit on the big side. :\
15:01:33preglowwell, i'd rather have them anyway
15:01:44preglowdsp asm is good
15:02:13pregloware you gonna commit it today, though?
15:02:19jhMikeSThey be even better if the source and dest are line aligned but not with the current pcm buffer
15:02:25preglowi'll try to commit the tone controls stuff today, i hope
15:02:28preglowwe'll see if i have time
15:02:43jhMikeSI probably can, all the changes seem to be working well
15:02:47preglowthe alarm stuff still needs more work, i found out
15:02:51zylcheThe watch is in the tree.
15:03:03preglowzylche: roger that, the goose is aloft
15:03:10jhMikeSpreglow: so you slept late ;)
15:03:18preglowjhMikeS: excessively
15:03:23preglowsaturday today :)
15:03:33MikachuHub441: rockbox is a lot better than ipl at actually playing music
15:03:34preglowbut no, the problem was that it turned on despite me not wanting to
15:03:47preglowat the same time i set for friday morning
15:05:01Hub441Mikachu: and that's exactly what i want to do with my ipod! ;)
15:05:10jhMikeSso it failed to deactivate even though turned off in the ui?
15:05:12preglowi think zylche is confering with some fellow spies who're in tere as well
15:05:34zylcheThe crow is off, The crow is off!
15:05:36preglowjhMikeS: when the alarm switches on once, it's supposed to deactivate itself
15:05:37 Part zylche ("-")
15:05:52preglowin five minutes, we'll see a nuclear explosion somewhere
15:05:53 Join zylche [0] (
15:06:03dan_a\msg BigBoss Preglow let slip that there are spies in Tere
15:06:23*jhMikeS knows there's always spies
15:06:29*zylche starts on 1984...
15:06:42jhMikeSbut that's what the meds are meant to help ;)
15:07:21ShaidShalamaneser is real!
15:07:33preglowanother cup of coffee before i'll start cleaning this place
15:08:13*jhMikeS has to clean badly...too many chip doc printouts all over the floor
15:09:24 Join ender` [0] (
15:11:38*dan_a tries ROLO-ing the Sansa OF. It didn't work, but in a much prettier way than the iPods
15:13:02 Quit Mouser_X (Nick collision from services.)
15:13:13 Join Mouser_X [0] (
15:18:49***Saving seen data "./dancer.seen"
15:21:33preglowmight not have time to commit today after all
15:21:34preglowoh well
15:25:00 Nick printfXh4 is now known as setrlimitXh4 (
15:26:50jhMikeSpreglow: what filter cutoffs did you settle on? The hw ones are 250/300Hz and 1.5/3.0kHz.
15:27:02preglowwhy the slashes?
15:27:20preglowwhere'd you see that?
15:27:25jhMikeSin the datasheet
15:27:27 Nick setrlimitXh4 is now known as fwopenXh4 (
15:27:31preglowreally now
15:27:35preglowi didn't see that
15:27:37jhMikeSp. 15
15:27:45preglowwhich do we use?
15:27:57jhMikeSIn one of the printouts on the floor
15:27:58preglowi'll try out 200 and 4000
15:28:05preglowand see how those sound
15:28:09jhMikeSI set like 120 and 8000 hrmph
15:28:18preglowtoo extreme for my tastes
15:28:23preglowbut like i said, i'll check it out
15:28:36preglowbut looks like i'll have company later on, so probably won't be until tomorrow
15:29:28*jhMikeS imagines what kind of company would keep someone from rb...and realizes what would keep _him_ tied up
15:31:38Mikachui'd say... any company? :)
15:31:48*jhMikeS hasnt' been tied up in weeks :(
15:32:08preglowi seldom tell company to fuck off because i wanna code rockbox, no
15:32:16preglowin this case it's just a couple of mates, i'm afraid
15:32:29preglownothing too exciting
15:32:36Mikachu"can't talk, haxxoring"
15:32:39jhMikeSahh...don't come to my house when I'm trying to solve a problem...unless you've got tits
15:32:58Shaidwords to live by
15:32:59preglowi'm easy to distract
15:33:07preglowbut tits do it extra well, true
15:33:49preglowtoo well
15:33:53preglowbut enough about that!
15:34:23preglowit's hard enough to not loose my concentration anyway, and right now i need to concentrate on these bloody floors
15:36:14Shaidhave fin
15:36:40 Quit Hub441 (Connection timed out)
15:40:41JdGordoncan you put a #if inside a #define?
15:41:36 Part Plouj
15:42:46jhMikeSsure: #def#ifine
15:43:03*JdGordon is too tired for sarcasm :p
15:43:18JdGordonthe player doesnt have the button bar does it?
15:43:26jhMikeSno, I don't think you can...but then there's always just try it out
15:43:37Soapso what is this about no boosting with DSPing?
15:44:54jhMikeSOn coldfire I've got it tuned up enough that all the effect add little or no boost depending on the codec
15:46:14Soapahh, coldfire. :(
15:46:22Soapstill impressive.
15:46:37Soapjust making me jelous.
15:46:40jhMikeSpreglow is gonna take on the ARM after I do the revamp
15:47:57Soapis there much hope for similarly impressive results with ARM?
15:47:58 Quit zylche (Read error: 110 (Connection timed out))
15:50:33JdGordonamiconn: ping?
15:50:51jhMikeSSoap: he thinks it wont be as drastic as coldfire
15:51:21jhMikeSbut also says you can't coun't on that maybe
15:51:27Soapeven a 20% speedup in EQ would give you another band for "fee", though.
15:51:39Soap"free" even
15:52:22jhMikeSthat's already taken care of...other stages were kind of slowish I found.
16:02:18elinenbehas there been any interest from the google summer of code?
16:04:05 Quit Mouser_X (Read error: 110 (Connection timed out))
16:04:27 Join Thundercloud [0] (n=thunderc@
16:05:50 Quit JdGordon ("Leaving.")
16:06:45 Join Domonoky [0] (
16:11:50 Join redbreva [0] (
16:13:20winchesterdoes break leave a function?
16:13:37winchesterwait nvrm
16:13:55 Quit Lynx_ (" Try HydraIRC -> <-")
16:23:19 Join zylche [0] (
16:25:07 Quit entheh ("^~")
16:35:31 Join Hub441 [0] (
16:36:19Hub441just installed rockbox on my ipod g4. really nice so far. the last thing i'm missin is support vor various artists/compilations
16:37:19Hub441but ogg support really rocks ;)
16:38:15 Quit Nico_P (Remote closed the connection)
16:39:03 Join Nico_P [0] (
16:39:35 Quit ender` (Read error: 104 (Connection reset by peer))
16:42:41 Join ender` [0] (
16:46:20 Join pilot000 [0] (i=55023c17@gateway/web/cgi-irc/
16:52:39 Quit pilot000 ("CGI:IRC (EOF)")
16:59:02bonbonthejonHub441: what do you mean, missing support for various artists?
17:03:15bluebrotherI guess he means grouping albums with various artists in the database view
17:03:32 Quit linuxstb_ (Read error: 110 (Connection timed out))
17:03:59bonbonthejonbluebrother: oh, i dont use database view, so I wouldnt be abel to help
17:04:03 Join Vyrus001 [0] (
17:06:11 Join Arathis [0] (
17:06:30 Quit Hub441 (Read error: 110 (Connection timed out))
17:07:36 Join Ribs [0] (n=ribs@
17:07:36 Join toffe [0] (n=chatzill@
17:07:43 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
17:07:44winchesterhey.... in c..... if I want to make a structure with a boolean and int components.... how would I do that, and how would I define them?
17:08:01winchesterhey.... in c..... if I want to make a structure with a boolean and int components.... how would I do that, and how would I define them?
17:08:09winchestergah, sorry. lag
17:08:10 Nick Arathis is now known as Arathis|afk (
17:11:36bluebrotherwinchester, where's the problem? struct foo { bool a; int b; int c; };
17:11:53bluebrotheror did I get your question wrong?
17:12:25winchester(problem being I don't know C), and how would I give a a value and retreive it later?
17:12:45winchestersince I am sure you can't just foo.a=.... in C
17:13:06bluebrotheryou need to define a variable of that structure.
17:13:11bluebrotherstruct foo name;
17:13:27bluebrotherand then name.a is the bool
17:13:41bluebrotherif name is a pointer to the structure you need to write name->a
17:13:55 Quit Domonoky (Read error: 104 (Connection reset by peer))
17:14:02winchesterah... I get so confused with pointers
17:14:19*bluebrother loves pointers
17:14:27winchesterbut thanks bluebrother, I get the idea.
17:14:38winchesterO_o what's there to love about them?
17:14:58markundon't know, but I love them too :)
17:15:00bluebrotheryou can also omit the structure name if you declare a variable immediately, i.e. struct { elementsofstruct} variablename;
17:15:21bluebrotheryou can do great things which would get otherwise much harder
17:15:28bluebrotheronce you understood how they work ;-)
17:16:51linuxstb_winchester: The easiest thing is just to search other plugins for the word "struct" and see examples of them being used.
17:17:07winchesterI'll get back to my patch when I finish the C book I am reading then.
17:18:26winchesteror I'll just get back to it when I have spare time..... does anyone remember when I started?
17:18:51***Saving seen data "./dancer.seen"
17:18:55winchesterthink i've just spent almost 24 hours coding programming O_O
17:22:39 Join the_miker [0] (
17:22:50winchestermy girlfriend goes on camp for 3 days and look what happens
17:23:24winchesterbut anyway.... 'night
17:23:25 Quit winchester ("Leaving")
17:28:15 Join Criamos [0] (
17:30:49 Quit dan_a (Remote closed the connection)
17:30:59 Join Vyrus [0] (
17:31:02 Quit Vyrus (Read error: 104 (Connection reset by peer))
17:32:13 Join dan_a [0] (n=dan_a@
17:32:39 Quit Vyrus001 (Read error: 104 (Connection reset by peer))
17:39:10 Quit Arathis|afk (Remote closed the connection)
17:39:33markunThe Ogg Vorbis guys are thinking about using this surround sound technique, quite interesting:
17:39:58 Join Mouser_X [0] (
17:47:14 Join Arathis [0] (
17:49:59 Quit unwell_13 ()
17:50:59preglowand thusly all was sparkling
17:51:25preglowmarkun: i happen to be working with that technology
17:52:07 Join linuxstb__ [0] (
17:52:10preglowat least from time to time
17:52:30markunpreglow: great, there is a lot of talk about it on the vorbis-dev mailinglist
17:52:37preglowmarkun: as in currently?
17:52:59markunjan and feb
17:53:08preglowhmmm, perhaps i could get the uni guys to pay me further to research how it works with vorbis...
17:53:16markunlook for "5.1 surround channel coupling"
17:53:18preglowambisonics is quite dependent on relative phase between signals
17:54:03preglowi'm on and off writing on an open source linux encoder/decoder for it, actually
17:55:31markunpreglow: do you think we could do some nice things with it in rockbox? like using a hrtf to make it sound 3d with headphones and turning in the sound field?
17:56:05preglowmarkun: it's something i'm planning to research, actually, so we'll see
17:56:15preglowit's pretty computationally expensive
17:56:28*jhMikeS hopes he doesn't get spanked for the extra 120-200 bytes coming up
17:56:40preglowfirst you need ambisonics encoding, then decoding, then hrtf filtering for a good number of channels
17:58:07markunpreglow: I hope either the Gigabeat will be powerful enough to do some nice things or else maybe a future target with a dsp?
17:58:15preglowalso add in the fact that you need a quite high order for it to sound good, but i don't know, really
17:58:18preglownever tried
17:58:26preglowthe gigabeat can do it for sure
17:58:54 Quit the_miker ()
17:59:12jhMikeSis this an insane thing for me code try doing in assembly? if so, I'm up :)
17:59:49preglowit would have to be asm
17:59:56preglowambisonics really just is a lot of gaining
18:00:02preglowwith optional iir/fir filtering here and there
18:00:19preglowthe hrtf part will be tricky to get right, though, being an individual factor
18:00:23jhMikeSsounds like the stuff...should be more room for it anyway
18:00:24preglowwhat sounds good for some will sound bad for other
18:01:00markunpreglow: we can decode the B format to stereo easily I think and then just listen to it with or without crossfeed.
18:01:10jhMikeSthis adds spacial queues to stereo headphones, right?
18:01:42jhMikeSby "ambisonics" it certainly conveys that....will have a look in a moment
18:01:46preglowmarkun: decoding b format to stereo, how? via hrtfs?
18:02:11preglowi assume so, since it's really the only good way
18:02:53markunpreglow: I was thinking of this (but really don't know much about it yet)
18:04:25 Quit linuxstb_ (Read error: 113 (No route to host))
18:04:44preglowmarkun: hmm, that looks fun
18:05:04preglowbut that'll only work for first order ambisonics, it seems
18:05:13preglowthe main area of resarch now is higher order ambisonics
18:05:28preglowbut anywho, i've gotta go, feel free to drop some comments, i'll read later
18:06:13markunok, perhaps I find some interesting things
18:08:09jhMikeSpreglow: can the eq bands be applied in parrallel?
18:21:14 Quit Obsys ("Chatzilla 0.9.77 [Firefox]")
18:22:10 Quit bezlemin (Read error: 110 (Connection timed out))
18:24:27 Quit Arathis ("Bye, bye")
18:25:30 Join Arathis [0] (
18:27:40 Quit lee-qid ("aufwiederbyebientotsayonara")
18:45:42 Quit Mouser_X (Read error: 110 (Connection timed out))
18:46:05 Join bezlemin [0] (
18:49:27 Join MobiLex [0] (
18:54:24preglowjhMikeS: yeah, but then the filters need to be different
18:59:00 Join Hub441 [0] (
19:01:16 Join daurn [0] (n=fake@unaffiliated/daurnimator)
19:01:59 Quit daurnimator (Read error: 60 (Operation timed out))
19:02:13jhMikeSpreglow: least it keeps later bands from compounding side effects of earlier ones.
19:02:43 Join Anon0052 [0] (
19:03:34pregloweach filter affects phase which will yield phase effects between filters
19:04:06Anon0052Is the the dual processor patch in the source yet(Kocp)?
19:04:13 Nick Anon0052 is now known as BigMac (
19:04:38jhMikeSso in other words in this case you're avoiding combing
19:05:11BigMacfor the5g*
19:05:28 Nick linuxstb__ is now known as linuxstb (
19:05:32BigMacAh sweet
19:05:42linuxstbBigMac: No.
19:06:03MobiLexNot then
19:06:04BigMacAh ok, then who is the author of the patch?
19:06:12linuxstbMainly dan_a
19:06:23BigMacOk thank ya
19:07:01BigMacIs he Daniel Ankers?
19:08:30BigMacHmm is his patch in the tracker? I just put him in the author field and searched and came back with no results
19:09:37jhMikeSpreglow: but of course they do, but I've seen it done with "shelf" and "peak" filters. wouldn't yet know much about how to get it right or much about the different kinds.
19:09:39BigMacOk thank you
19:10:21Hub441is a systemfreeze quite normal on a g4? already had two ;)
19:10:54preglowjhMikeS: dinner, will read later
19:11:22jhMikeSyumyum...I'm hungry and can't find anything good :(
19:12:20BigMacjhMikeS:Ramen is what I've got :)
19:12:53jhMikeSI'm out ... :\
19:13:31jhMikeSpreglow: at least watch the pcm buffer stay pegged at the top with almost all the dsp stuff running on your H120 while you eat :)
19:13:40 Quit bonbonthejon (Read error: 113 (No route to host))
19:14:06 Quit SirFunk (Read error: 104 (Connection reset by peer))
19:17:11 Join CriamosAndy [0] (
19:17:11 Quit Criamos (Read error: 104 (Connection reset by peer))
19:17:34linuxstbHub441: Yes, it's a known issue on the 4g, but avoidable if you use this build (or compile your own) -
19:18:53***Saving seen data "./dancer.seen"
19:21:39 Join miepchen^schlaf [0] (
19:23:41*jhMikeS found chicken pot pies in the fridge downstairs...mmm :D
19:24:44MobiLexDoes 4g affect photo too
19:26:36linuxstbThe Photo has the same problems as the 4g.
19:27:24MobiLexMy friend complains about that
19:29:49 Join tick [0] (i=95e13cfc@gateway/web/cgi-irc/
19:30:09tickFS admins here?
19:30:19Mikachulinuxstb: i eventually had to remove the checks and just return DRAM_START to get the bootloader to start, it might still be my fault though :)
19:30:28Mikachulinuxstb: to start the OF, that is
19:30:31Mikachutick: yes
19:31:06tickCould you please add a patch category in FS for "just code improvements", i.e. not a feature addition but just code cleanup etc.
19:31:14Mikachunot that much of an admin
19:31:45tickOr where should such patches go?
19:32:18tickE.g. fixing a typo in a comment.
19:33:00 Join Thundercloud_ [0] (n=thunderc@
19:36:47tickMikachu: so you are not an FS admin?
19:37:00Mikachui can close bugs, and that sort of stuff
19:38:21 Quit Thundercloud (Read error: 60 (Operation timed out))
19:38:39BigMacMikachu: Is your nick pronounced MEEKACHU or MIKACHU
19:38:40tickAha, ok. I hope Bagder & Co. will read this.
19:38:54 Join SirFunk [0] (
19:39:06MikachuBigMac: the second alternative is sort of ambiguous :)
19:39:46 Quit Hub441 (Read error: 113 (No route to host))
19:41:36 Quit desowin ("use linux")
19:44:29tickMikachu: do you know who is/are that sort of FS admins?
19:44:40 Quit MobiLex (Remote closed the connection)
19:44:51Mikachumaybe zagor or bagder
19:46:39tickOk, then... waiting for Godot ^H^H^H^H^H Bagder :-)
19:46:42 Join web-taz [0] (
19:47:39 Join x1jmp [0] (
19:50:41 Join DerPapst [0] (
19:50:42x1jmpWhat's the way to create a patch for an already patched rockbox version with svn?
19:51:11 Join DataGhost [0] (
19:51:32 Join Mouser_X [0] (
19:51:53 Quit tick ("CGI:IRC (Ping timeout)")
19:52:08*DerPapst wonders what happend to the AustriaMicrosystems talk...
19:58:28 Quit Mouser_X (Nick collision from services.)
19:58:35DerPapstwas the meeting already?
19:58:37 Join Mouser_X [0] (
20:02:07Kasperle says "The visit will be on 9th of March and RenePeinthor and I (ChristianGmeiner) will went to them... stay tuned"
20:03:03 Quit FOAD_ ("I'll be back")
20:03:57 Quit BigMac ("MadPenguin forever!")
20:07:10 Join mightybrick [0] (i=44bb32e5@gateway/web/cgi-irc/
20:14:59Nico_Pcould we please have a link to on the tracker ? i just closed a request for rockbox on the zen vision m...
20:17:06 Quit toffe (Remote closed the connection)
20:23:11 Join toffe [0] (n=chatzill@
20:25:05 Quit _Veseliq_ (Remote closed the connection)
20:27:50 Quit Mouser_X (Nick collision from services.)
20:28:01 Join Mouser_X [0] (
20:30:17 Join Thundercloud [0] (n=thunderc@
20:30:43 Join desowin [0] (
20:33:32 Quit Thundercloud (Read error: 104 (Connection reset by peer))
20:33:45 Join Thundercloud [0] (n=thunderc@
20:37:57Presencefixed.cfg is the greatest thing ever for rockbox.
20:38:24Presencelate at night I turn the brightness down to like uber low. in the morning, its right back up to "blind the fuck outta myself".
20:40:53 Join FOAD [0] (
20:41:01 Join busa_blade [0] (
20:52:28 Join bluey- [0] (
20:53:35 Join blippe [0] (
20:54:18blippei got some questions before installing rockbox on an ipod: first, does it matter if i upgrade the ordinary ipod sow
20:55:21nlsblippe: if you do the update after you install rockbox, you will need to reinstall the rockbox bootloader
20:55:37blippeok. nice
20:55:39 Quit Thundercloud_ (Read error: 110 (Connection timed out))
20:56:45 Quit Mouser_X (Nick collision from services.)
20:56:56 Join Mouser_X [0] (
20:57:08 Quit mightybrick ("CGI:IRC")
21:02:47 Join mightybrick [0] (i=44bb32e5@gateway/web/cgi-irc/
21:04:53 Join petur [0] (n=petur@rockbox/developer/petur)
21:08:08 Quit scorche (Remote closed the connection)
21:13:09 Join Thundercloud_ [0] (n=thunderc@
21:14:16 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
21:18:55***Saving seen data "./dancer.seen"
21:24:00 Quit bluey- ("Leaving")
21:28:59 Join Compaq_Administr [0] (
21:29:00 Quit Mouser_X (Nick collision from services.)
21:29:05 Quit FOAD ("I'll be back")
21:29:08 Join Mouser_X [0] (
21:29:09 Nick Compaq_Administr is now known as SkB (
21:29:34SkBI have a question :P
21:29:44SkBDoes RockBox require use of the hold switch?
21:29:50SkBon 5g ipods
21:32:56 Join FOAD [0] (
21:33:14nlsSkB: you use the hold switch for 3 different things, 1) button hold (surprise) 2)To boot apple firmware when starting 3) to clear settings when rockbox starts
21:33:26nlsnone of which are vital exactly
21:33:52SkBCan you get into apple firmware without hold switch?
21:33:55SkBMine is broken
21:34:14nlsnot with the rockbox bootloader...
21:34:25nlsit is possible with ipod linux bootloader however
21:34:34SkBCan you use that bootloader without ipl?
21:34:46nlsyes, i believe so...
21:35:38 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
21:35:57 Quit midkay (Read error: 110 (Connection timed out))
21:36:10nlsSkB: or you can wait until someone figures out how to "RoLo" apple firmware...
21:36:24linuxstb_SkB: Yes, there are two ways to start the Apple firmware - holding MENU as you boot, or turning on the hold switch.
21:36:29funkyjust pressing menu button while rebooting, you go into apple firm
21:36:35funkyhold button is not needed at all
21:36:44SkBreally now...
21:36:46nlsoh, didn't know the MENU button was used still
21:37:10SkBso I DON'T need loader2
21:37:42funkythats it
21:38:03SkBI'm using the loader X because I can't DL from the rockbox website...
21:39:22linuxstb_What's wrong with the website? doesn't work for me
21:40:08SkBand I can't download the manuals
21:40:36 Quit Thundercloud (Read error: 110 (Connection timed out))
21:41:08SkBhere I go
21:41:16linuxstb_Odd, everything's working fine for me.
21:41:23 Quit redbreva (Remote closed the connection)
21:42:00SkBDo I have to boot my ipod into disk mode to install?
21:42:36 Quit Drumr ()
21:43:03SkBHow; I forgot
21:43:12SkBMan I'm such a n00b
21:45:27SkBgot it.
21:46:11linuxstb_This link doesn't work for you ?
21:47:08SkBFirefox said the connection timed out
21:49:10 Join johnd0e [0] (i=42beea7b@gateway/web/cgi-irc/
21:49:32johnd0eI have a quick question
21:50:15peturI see
21:50:40johnd0eI have a 5g 30gb ipod and I just bought a iluv battery pack and a wall charger, and when I plug them in, rockbox persistently boots into disk mode, is there any way to turn this off?
21:50:57johnd0eIs it hard coded or in the config?
21:51:11linuxstb_Are you running a recent version of Rockbox?
21:51:36 Quit Ribs ("eh eh ehhhh!")
21:52:01johnd0ethe build date was Jan 16
21:52:17linuxstb_Try installing a more recent version - that was a bug which should be fixed now.
21:53:01johnd0eSweet. I thank you linuxstvb_ . If I have any problems, I'll check back
21:53:09SkBhow do I install this damn thing
21:53:27SkBjust boot into disk mode?
21:53:50linuxstb_If you can't access the websites, then you won't be able to...
21:53:51johnd0eboot into disk mode and copy files over, next step depends on what mp3 player you have
21:54:10SkBI'm using a custom installer
21:54:19johnd0ewhat custom installer?
21:54:31johnd0eI wasn't aware there was a custom installer out...
21:54:39scorchean unofficial one
21:54:57SkBRockBox Installer X
21:55:48SkBit's only for ipods
21:55:57SkBand I think it comes with a build
21:56:04SkBbecause I can't DL one
21:56:04preglowjhMikeS: so what you say is that if we ran the different filters in parallel, the freqency response consisting of all five filters would somehow look different than it does now?
21:56:15scorcheand is unsupported by us...if you have issues with it, dont come to us
21:56:30SkBNo worries :)
21:56:33linuxstb_SkB: Can you access ?
21:56:41 Quit linuxstb_ ("CGI:IRC")
21:56:47SkByes I can
21:57:05 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
21:57:17peturany fat32 specialist around?
21:57:40linuxstb_SkB: Unzip it, and run ipodpatcher.exe
21:58:27SkBand then I have rockbox?
21:58:37linuxstb_Then you have the bootloader.
21:58:55SkBthat doesnt do anything If I don't have rockbox :-/
21:59:21 Join midkay [0] (n=midkay@rockbox/developer/midkay)
21:59:31SkBSo, what do I do next
21:59:51scorchelinuxstb_: might be easier hosting a link to the manual as well =/
22:00:10 Quit Mouser_X (Nick collision from services.)
22:00:16 Join SkB_ [0] (
22:00:17linuxstb_scorche: I was about to...
22:00:21 Join Mouser_X [0] (
22:00:31linuxstb_SkB: I'll download everything you need and zip them up for you...
22:00:44SkBSo, just install the bootloader now/
22:00:52 Quit SkB_ (Client Quit)
22:00:58SkBstill here.
22:02:06 Quit Kitt0s (Read error: 104 (Connection reset by peer))
22:02:06 Quit FOAD ("Lost terminal")
22:02:19linuxstb_SkB: Wait, and read the manual I'll give you.
22:02:22 Join Kitt0s [0] (
22:02:56linuxstb_Sorry, yes you can install the bootloader - just double-click on ipodpatcher.exe and say yes.
22:04:24SkBbootloader installed succesfully
22:04:56linuxstb_ - read the pdf but you can skip the bootloader part of the install.
22:05:42jhMikeSpreglow: huh? the phase response would. Why would the frequency response by different if the parellel filters each have the same frequency domain characteristics? There is the phase I guess I go back to square one and say "yes" because of combing.
22:05:52SkBtimed out
22:06:10 Join FOAD [0] (
22:06:19SkBWhat's my problem?
22:06:22SkBI timed out
22:06:25preglowjhMikeS: they'd be massively different if the filters were they same
22:06:39preglowjhMikeS: you can't make the two resulting spectrums equal, but not with the same filters
22:07:03johnd0elinuxstb_: Just wanted to say it works beautifully in the new version, thanks again, best wishes
22:07:15jhMikeSpreglow: your typing seems slurred :\
22:07:39 Quit busa_blade (Remote closed the connection)
22:07:59linuxstb_johnd0e: That's good to hear.
22:08:09SkBlinuxstb_: I can't download it; I timed out again.
22:08:09preglowapart from "can't" that should be "can", it was all correct
22:08:10linuxstb_SkB: I can't fix your internet connection...
22:08:27 Quit johnd0e ("CGI:IRC (EOF)")
22:08:27preglowjhMikeS: i probably misunderstood something you said a long way up
22:08:31Soapturn off the bittorrent client!
22:08:44SkBCan you host it somewhere else?
22:09:33jhMikeSNot sure why some are preferring to do parellel if reinforcement and cancellation are a problem with summing the varying phases at output.
22:09:44linuxstb_If you can't download from two independent websites, I tend to blame your connection...
22:10:09preglowjhMikeS: it's not really a problem, it's pretty unavoidable if you use iir filters
22:10:16SkBI blame gremlins.
22:10:35SkBI never have problems like these.
22:10:37preglowjhMikeS: people used to like using parallel filters in the days of analogue electronics so that the complete filter response wouldn't be too affected by component tolerances
22:10:48preglowjhMikeS: in dsp, series is what's mostly used
22:11:33preglowthe result would be exactly the same if we used parallel filters, afaik, but we'd have to use bandpass, lowpass and highpass filters instead of the ones we use nmow
22:11:38jhMikeSI see...not that caps such wouldn't suffer the same. Mine idea was more from an implementation pov anyway.
22:12:17SkBLinuxstb_: Can you post it on
22:12:19 Join digriz [0] (
22:12:33preglowbut anyway, why would you want to do them in parallel?
22:12:46preglowsounds like it'd result in more memory pressure to me
22:13:36jhMikeSwondering if I can be more insane...and if it might be better at least for emac...but series but be pipline compatible anyhow :\ gotta think.
22:13:43SkBLinuxstb_: I got it!
22:14:23SkBDon't know why I couldn't get it the first time
22:14:35SkBI *think* it was trying to open it as a webpage.
22:14:39jhMikeSWell anything IRAM is scarcely memory pressure. Doesn't seem too intense.
22:14:55 Join JPMaximilian [0] (
22:15:07preglowjhMikeS: true enough, but i still can't see much of a point
22:15:14preglowyou still need to do the same filtering
22:15:29jhMikeSI'm sure I'll make one at some point...:D
22:15:31preglowand anyway, all our targets don't have emac :/
22:15:47jhMikeStrue...harse reality for them :)
22:15:52jhMikeSharsh even
22:16:18preglowyou'll need an extra buffer too
22:16:21linuxstb_Some just have a 300MHz ARM...
22:16:22digrizquick question how long will a battery last generally on a 5g ipod using rockbox to just play mp3's
22:16:28digrizthe faq is vague
22:16:31 Join JPMaximilian_ [0] (
22:16:34preglowsince all filters need the same input
22:16:42jhMikeSlinuxstb_: yes...but those don't count ;)
22:16:52preglowyou either need to run all filters in parallel, one sample at a time, at would give more memory pressure
22:16:58preglowor you need an extra buffer for input samples
22:17:12linuxstb_digriz: Try the IpodRuntime wiki page for some test results.
22:17:35jhMikeSIf I get insane enough and get a chance to do a good study...perhaps something will dawn on me. I dunno.
22:18:11*jhMikeS wants time warpage
22:18:30*petur joins jhMikeS
22:18:33digrizwow it cuts battery time in half
22:19:35preglowjhMikeS: i'd love to be enlightened anyway, so do tell if you discover something :)
22:19:37jhMikeSpetur: what year? dial it in on the eq
22:19:51preglowanywho, brb
22:19:53preglowor more like bbl
22:19:54jhMikeSpreglow: I need much enlightenment myself
22:20:05 Join amiconn_ [0] (
22:21:34SkBdo you extract the fonts package to the root directory of the player?
22:22:09*petur excuses himself for adding some bytes to the binary
22:22:19jhMikeSI guess I have to work out the equations and reduce them in series
22:22:21 Quit JPMaximilian_ ("I Shall Return")
22:22:26 Quit JPMaximilian ("I Shall Return")
22:22:32 Join JPMaximilian [0] (
22:22:39*jhMikeS added bytes today too
22:22:51SkBAlso: does the menu method of booting into the apple firmware still work with version 2?
22:22:52 Quit FOAD ("Lost terminal")
22:23:24 Quit DerPapst (Read error: 110 (Connection timed out))
22:23:55linuxstb_version 2 of what?
22:23:56 Join FOAD [0] (
22:24:28linuxstb_But yes, it should work with anything.
22:25:52SkBversion 2 of the bootloader
22:28:25Bagderpetur: isn't "snprintf(num, 9" a bit too strict
22:28:39SkBthe font package goes in the root directory of the ipod or the rockbox directory?
22:28:47Bagderafair, that'll limit it to use only 9 bytes, ie 8 + zero
22:29:20Bagderwhich doesn't fit with the line above "cnt %= 9999999" which uses 9 digits
22:29:33MikachuSkB: yes and no, the font files go in /.rockbox/fonts, but you extract it to / because the zip has the directory structure
22:29:38*jhMikeS can't believe what an utterly superior sounding playback device is to the H120 and yet an utterly inferior recording device overall at the same time.
22:29:45jhMikeSthe x5 that is
22:29:54MikachuBagder: wouldn't you rather use %= 1000000 ?
22:30:16peturBagder: it is 8.3 so we want ~and max 7 digits, no?
22:30:22Mikachui mean petur:
22:30:41Mikachu%= 999 means from 0 to 998
22:30:55 Join Thundercloud [0] (n=thunderc@
22:31:04 Quit FOAD ("Lost terminal")
22:31:11peturah well..
22:31:57peturfeel free to create 9999999 files ;)
22:32:14Bagderpetur: sorry, I counted the digits on the line before snprintf() wrong
22:32:19Mikachuit's not a bug, but it's suboptimal :)
22:32:36SkBHell yeah!
22:32:46SkBLinuxstb_: Much thanks
22:33:02peturMikachu: should I change it? ;)
22:33:03jhMikeSIt should error out if it can't find a unique name perhaps?
22:33:30 Join FOAD [0] (
22:33:37peturjhMikeS: yes, but there's more stuff in there that isn't 100% ok
22:33:38Mikachupetur: i don't think it matters much, but maybe it confuses someone looking at the code
22:36:30 Nick Everybody|away is now known as Everybody (
22:36:46jhMikeSrrr...the x5 battery scale is all wrong for a non L model again. :\ How to solve that one?
22:36:53 Quit amiconn (Read error: 110 (Connection timed out))
22:36:53 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:37:16MikachuEverybody: didn't you forget something?
22:37:32jhMikeS...unless my battery is really dyning that fast now
22:37:59jhMikeSor "dyning" less is it may be
22:37:59bluebrotherEverybody is quite quiet ... except for name changes :)
22:38:09peturjhMikeS: keeping track of the fact that we tried all possible shortnames will cost some bytes, and I fear the code will never get executed ( as you get 10000000 possible names)
22:38:16*jhMikeS is saying all kinds of crap noones listening to :)
22:38:28 Quit FOAD ("Lost terminal")
22:38:49bluebrotherhe is alive!
22:39:02SkBHow do you get songs that are already on your ipod from itunes into rockbox
22:39:03Everybodyyeah I need to get around to putting this server into my mIRC server list so I can change away status settings for it
22:39:05Everybodyjust for you
22:39:21 Quit midkay ("Leaving")
22:40:13SkBdatabase -> initialize now?
22:40:46*petur hopes Mikachu will be satisfied
22:41:05Mikachui hope so too
22:42:00jhMikeSpetur: it's meant to be called in a search loop by something that actually reads the disk?
22:43:35peturbut you could start at a high number and wrap around, so you need to keep track of it. The calling code does goto restart:, maybe a counter can be used there.
22:43:38jhMikeShow can a short filename have a name "greater" than x~999999.ext ?
22:44:05 Join FOAD [0] (
22:44:09peturgood question ;)
22:44:23peturby using it as long filename
22:44:52jhMikeSummm...then why worry about a limit?
22:45:13jhMikeSbut then it's not a "dos" name anymore ;)
22:45:22peturif you create x~999999.ext, and _then_ create x~999999.ext2 you'll get a wrap
22:45:46peturbecause x~999999.ext was used as shortname because it was valid
22:46:48 Quit Thundercloud_ (Read error: 110 (Connection timed out))
22:46:50jhMikeSyes, as it should. dos filenames strip spaces and lfn only character too
22:47:11 Quit Mouser_X (Nick collision from services.)
22:47:20 Join Mouser_X [0] (
22:47:44Mikachucan't you just have "invalid" characters in the generated short filenames? or would that break windows?
22:48:05peturit's against the spec
22:48:31jhMikeSinvalid ones would yes
22:48:53peturbut it depends on the version of windows you use, and wether you try in on command line or in the browser (really!)
22:49:22Mikachugood thing they implement the filesystem spec in userspace
22:49:28jhMikeSof course any short filename generator replaces the path components as well
22:49:58peturI've seen it convert weird symbols into a strange series of numbers - must still check if that was the char code
22:50:26 Join webtaz [0] (
22:50:37peturhehe, using the correct constant saves some bytes :)
22:50:42jhMikeSanother thing with long filenames is if the bit after the last "." contans spaces, it's not an extension, nor is it if it's a directory
22:51:36*petur preps for an early night to bed
22:51:50jhMikeSthose are my observations anyway...
22:52:42jhMikeSalmost 11pm there?
22:53:06peturhave been doing hard work outside all day - my muscles hurt
22:53:14jhMikeSpretty early I'd say...:)
22:53:34peturyes, for me that is early but the body says it is time
22:53:48jhMikeSthen you'll sleep for sure
22:54:18 Quit petur ("Zzzzzz")
22:56:24 Quit FOAD ("Lost terminal")
22:57:18 Join FOAD [0] (
22:57:52 Join entheh [0] (
23:06:22 Quit bluebrother ("Verlassend")
23:08:26 Quit web-taz (Read error: 110 (Connection timed out))
23:15:06Everybodycome to think of it
23:15:09 Quit SkB ("Chatzilla 0.9.77 [Firefox]")
23:15:20Everybodywould you guys prefer it if I just didn't idle in here
23:15:25Everybodyit's not like you need idlers
23:15:32Everybodyand I don't really need help any more
23:17:08 Quit Mouser_X (Nick collision from services.)
23:17:17 Join Mouser_X [0] (
23:18:51preglownow, let's try to update
23:18:54Soapwhy not idle in here?
23:18:58***Saving seen data "./dancer.seen"
23:19:06preglowjhMikeS: funnily enough, conflict
23:19:39 Join x1jmp_ [0] (
23:19:44preglowjhMikeS: what's up with all the empty lines at ends of structs?
23:19:47 Quit x1jmp_ (Client Quit)
23:19:51EverybodySoap because everyone hates my nick changes
23:19:53preglowjhMikeS: forget that...
23:20:16SoapEverybody: I don't see how one excludes the other.
23:20:23Everybodymy nick changes happen much less frequently than joins and quits
23:20:26Everybodywhat do you mean
23:21:03jhMikeSpreglow: ?? I forgot it but what did I forget not?
23:22:50preglowjhMikeS: what i asked for above that, "empty" last lines of structs
23:22:54preglowthey're not empty, they have comments
23:23:17preglowunless you were being ironic, in which case you need forget i'm saying this too
23:23:52jhMikeSperhaps both...and neither :)
23:24:34jhMikeSyou mean the stuff with the offsets listed by things used by asm?
23:25:06preglowthings have definitely happened in dsp.c, yes
23:25:16jhMikeShelp me keep track of things...maybe all the comment are so I won't forget ;)
23:27:13preglowso now i can count on the dsp pointer always being right?
23:27:28jhMikeSyes...I believe so.
23:28:00jhMikeSI listed dependencies in the comments
23:28:20jhMikeSAnd we do want voice going throught the eq right? I left it that way on purpose
23:29:46preglowreally? i can see only one eq state
23:30:06jhMikeSVoice and audio share it
23:30:11preglowwell, that's not really right
23:30:12 Join redbreva [0] (
23:30:27preglowfirst, the history will contain audio samples at first when you do voice samples
23:30:33preglowand vice versa
23:30:48preglowalso, since voice almost always has a different sample rate, the filters won't be right at all
23:30:50jhMikeShmmm...that is true
23:30:57 Join sounds [0] (i=549969c8@gateway/web/cgi-irc/
23:31:03jhMikeSit's already resampled by then though
23:31:14preglowok, then the first issue is all that matters
23:31:20preglowsounds like a waste of cpu, though
23:31:26preglowmost processing should be done before resampling
23:31:33 Join Thundercloud_ [0] (n=thunderc@
23:31:39preglowbut then we'd need more code to support multiple sample rates, i guess
23:31:40jhMikeSWell...if tone controls go through both, then hey
23:31:57preglowmy sw tone controls don't, at least not right now
23:32:02jhMikeSI don't wanna do that at 88 or 96 khz for god sakes :)
23:32:05soundsjhMikeS: so you reckon that the X5 has better sound than the H1xx series, in what way ?
23:32:48jhMikeSsounds: It's very much the point sounds in the heaphones can sound joke
23:33:07soundsI have h140 only, just asking as I might get a X5, if its that better
23:33:09preglowjhMikeS: i guess it's right with regards to resampling, btw, if you're using a lower sample rate, you should have cycles to burn on the eq anyway
23:33:20preglowjhMikeS: and if you're using a higher sample rate, you probably want to apply it post resampling anywa
23:33:34soundsbut if tis a SW decoder how can there be a difference
23:33:41*jhMikeS really doesn't want the nightmare of actual switching
23:33:57preglowonly probably, though, with the current resampler, you really want to apply it before resampling so you won't be amplifying aliasing frequencies
23:34:04jhMikeSsounds: different codec chip....more basic and that's probably part of the reason
23:34:20 Join x1jmp_ [0] (
23:34:29 Quit x1jmp (Read error: 110 (Connection timed out))
23:34:37 Nick x1jmp_ is now known as x1jmp (
23:34:44jhMikeSThe sound from a H120 sounds smeary in comparison
23:34:47soundshmm, what earphones do you use, and what is there impedence ?, please,
23:35:30jhMikeSpx100 - not expensive and I didn't want isolation if I'm walking around so they're really quite perfect
23:35:42*sounds looks on ebay for an X5
23:35:56soundsrockbox works well with it ?
23:35:59jhMikeSI recommend it if playback is the most important thing
23:35:59 Quit Thundercloud (Read error: 60 (Operation timed out))
23:36:04jhMikeSsounds: oh yes
23:36:48jhMikeSstill needs some things worked out...which I plan on doing soon but nothing too serious
23:37:21soundsok, playback is the most important for me, replay gain, gapless, and eq
23:37:46jhMikeSit will handle that just fine especially after the commit today which sped things up
23:38:08soundssounds good ;)
23:38:23jhMikeSpreglow: You know, I never really heard anything bad from sending the voice through the eq that really didn't seem to matter in reality
23:40:44preglowyou're probably unlikely to hear in such a context, no
23:40:50preglowhaving the audio interrupted by voice anyway
23:40:52 Quit sounds ("CGI:IRC (EOF)")
23:41:02preglowin not a too nice manner with the amplitude being suddenly cut anyway
23:41:12preglowbut if we start doing fading, we'll probably hear it
23:41:16preglowespecially with high q eq filters
23:41:52 Quit Shaid (Nick collision from services.)
23:41:53jhMikeSwell I plan at least soft muting on things
23:42:00 Join Shaid [0] (
23:42:05jhMikeSno hard stops or start on a channel
23:42:05Shaidguess what.
23:42:08ShaidI rebooted.
23:42:14ShaidForced to, actually
23:42:32ShaidI forgot to turn my cpu fan back up after putting it in quiet mode and then went and ripped some FLACs
23:42:39Shaidand it shutdown due to overheating. woo!
23:42:48Shaidbut more importantly!
23:42:59ShaidThese's work now: ' ' ' " " "
23:43:10SoapWe really should remove the part about Release version 2.5 from section 2.2.2 of the ipod manual.
23:44:14 Join ALTiDEL [0] (
23:44:19jhMikeSman...I'm wondering if Cowon killed my battery or if new limits were put in cause it's shutting off on me way too soon
23:44:31nlsSoap: I agree, and am on it :-)
23:45:20 Quit miepchen^schlaf (Read error: 113 (No route to host))
23:45:28SoapjhMikeS: isn't one of the hallmarks of a dying Li-ion battery that it "falls off a cliff" voltage wise? That you lose the bottom 20-30% of capacity?
23:45:33 Join miepchen^schlaf [0] (
23:45:43jhMikeSholy crap...they destroyed it...
23:45:45Soap(and also the top 20-30%)
23:46:08jhMikeSSoap: I really don't know but it was fine when it got sent out and the curves in powermgmt.c haven't changed
23:47:36SoapI hope they take responsibility.
23:47:55jhMikeSthere's something wrong with the filtered voltage...I see it going way up all of a sudden...
23:48:05linuxstb_Shaid: Welcome to ASCII-world.
23:48:29jhMikeSSoap: They did the first time...wonder how many times they will againif it turns out to be that
23:48:42ShaidSilly Mac keyboards and their odd keymappings.
23:48:51Shaidlinuxstb: thanks. ;)
23:48:54Soapstill, what a pain.
23:49:40jhMikeSNow I show 3.70V which is the drop right after a charge to 4.20 which is correct
23:51:44Soapdropping half a volt quickly = failing battery. Then the curve should flatten out close to new, then plummet again at the end.
23:52:30jhMikeShmmm...I'll have to watch it.
23:53:33 Join Domonoky [0] (
23:54:12markunjhMikeS: I think the Gigabeat also sounds pretty good :)
23:54:13jhMikeSsticking at 3.70 with backlight on and it's been on awhile with the backlight
23:54:45markunbut I can't compare it to a X5 (because I don't have it)
23:54:46jhMikeSmarkun: really? and 300Mhz to boot? Wish I could listen to one.
23:57:40markunfunny, in #vorbis there is a bot which tells about the latest svn commits
23:57:52Mikachumany channels have those
23:57:58Mikachuis it called CIA-xx?
23:58:08markunwould it be nice to have it here too?

Previous day | Next day