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

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

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

#rockbox log for 2006-01-10

00:00:11 Quit t0mas (" good night")
00:01:24linuxstbmarkun: So you received the gigabeat source this morning?
00:01:34markunyes, I have
00:01:44markunbut not much of interest
00:02:03 Join webguest46 [0] (
00:02:23linuxstbIt sounds incomplete. Does it contain scripts to rebuild and install the gpl software that runs on your device?
00:02:33markunonly this file maybe:
00:02:44markunNo, it doesn't
00:03:39markunonly the tar balls of the open source software used in the gigabeat
00:03:56 Quit mikearthur (Read error: 104 (Connection reset by peer))
00:03:57linuxstbTo quote section 3 of the GPL: "The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable."
00:04:46linuxstbIn this case, "installation" would seem to have to mean installation on a real device.
00:05:26markunpreglow told me I should ask for more, but I don't really feel like bothering them all the time
00:05:49 Join ]RowaN[ [0] (
00:06:24]RowaN[guys where can i get one of those dual power+usb leads for h1x0 iriver series? there used to be some on ebay but i dont see them anymore
00:06:40linuxstbI understand how you feel, but they don't seem to be fulfilling their gpl obligations. It could be worth an email to the FSF to get their opinion about whether you are entitled to more or not.
00:06:40]RowaN[i managed to lose my big-ass wall charger over xmas some how =[
00:07:56linuxstbBut Toshiba could be in a tricky situation if installation of a modified kernel requires the user to encrypt the kernel using a secret key.
00:08:32webguest46hi, is it possible in rockbox to tell when the battey is full charged?
00:08:46webguest46on h3x0
00:09:37linuxstbI don't think that's been implemented yet. But others may know for sure.
00:10:11LinusNit isn't implemented
00:10:26webguest46ok, thank you
00:10:38 Quit petur ("here today, gone tomorrow")
00:10:54 Quit Strath ("Client closed")
00:11:50 Join Paul_The_Nerd [0] (
00:14:00 Join mikearthur [0] (
00:18:11lostlogicwebguest46: that's what I was looking at when I fried my player, will resume when Ig et a new one. It's pretty straightforward to read it from the I2C, but _when_ to read it is the harder question.
00:21:23 Join ashridah [0] (
00:22:20 Join Mongey [0] (
00:22:34Mongeyhi i have to ask a quick favour
00:22:55Mongeyanyone here?
00:23:01Bagderask away
00:23:02lokkiMmmm: great job
00:23:06*LinusN is waiting for the question
00:23:22lokkiMmmm: i'm extending it a bit more if it's ok
00:23:37Mongeyoh will someone compile Rb with a patch ill send them
00:23:42linuxstbBHSPitMonkey: Hello.
00:24:06MongeyPatch= working h300lcd remote
00:24:33LinusNMongey: you could try the beefed up h300 rockbox build at misticriver
00:24:48Mongeyi am using it
00:25:10BHSPitMonkeylinuxstb: noticed you made a remark earlier about a nano build... is the build working now, then?
00:25:11Mongeybut the h300 lcd remote doesnt work properly
00:25:29linuxstbBHSPitMonkey: Yes, it's been working for almost a week.
00:25:39Mongeyyes, dissapointing
00:25:51BHSPitMonkeythe last time i tried, the build was broken and i set the interest aside
00:26:03MongeyLinusN; will i dcc you the patch?
00:26:18BHSPitMonkeyi'm also holding out for iPL's loader to support rb
00:26:27 Quit mikearthur (Read error: 104 (Connection reset by peer))
00:26:28linuxstbBHSPitMonkey: You can see all the changes to Rockbox CVS on the home page at - that's the best way to keep up to date with the ipod port.
00:26:40LinusNMongey: shoot
00:26:43 Join mikearthur [0] (
00:27:01BHSPitMonkeyerr, what's the command to update a cvs again? cvs -Rp or something like that
00:27:09linuxstbcvs update -dP
00:27:19Mongeysending....waitinf for reply
00:27:35LinusNMongey: your firewall refuses connections
00:27:55 Quit DangerousDan (Read error: 110 (Connection timed out))
00:28:19linuxstbMongey: Where did you get the patch from?
00:28:26MongeyLinusN; i cant do pms
00:28:45LinusNMongey: see my /msg
00:28:49Mongeyok done
00:29:18Mongeyits the only one that has work w/t no flaws
00:30:01BHSPitMonkeylinuxstb: do you build from linux or windows?
00:30:02MongeyLinusN; if you could do it with the beffed up one, or is there no source?
00:30:04BHSPitMonkey(or mac)
00:30:13 Join akaidiot [0] (
00:30:58LinusNi can only build a standard rockbox
00:31:02MongeyLinusN; could you join #misticriver on zuh ?
00:31:11Mongeyso i can pm you
00:31:32 Part LinusN
00:31:53linuxstbBHSPitMonkey: Mainly Linux, but sometimes Mac OS X. You can also use Cygwin.
00:33:11BHSPitMonkeylinuxstb: just create a build directory, enter it, run the configure utility, then just build?
00:33:17 Join hd [0] (
00:33:36linuxstbYes - that's basically it.
00:34:16 Join Kyl3 [0] (
00:35:41 Join LinusN [0] (
00:35:45Mongeywb :P
00:36:14LinusNi thought xchat could connect to multiple servers
00:36:28 Join actionshrimp [0] (
00:37:02Mongeyhe i can see a pic of u LinusN
00:37:16LinusNwhich one?
00:37:50Mongeyand theres your number
00:37:54 Quit goa (Read error: 110 (Connection timed out))
00:37:55 Nick hd is now known as goa (
00:38:12Mongeyok gtg
00:38:15Mongeyty LinusN
00:38:20LinusNyou're welcome
00:38:30 Quit Mongey ()
00:42:35BHSPitMonkeyLinusN: Ctrl+T
00:43:07LinusNBHSPitMonkey: thx
00:44:31 Join Mongey [0] (
00:44:40 Quit Mongey (Client Quit)
00:46:48 Quit ender` (" Top reason why compilers are like women: Miss a period and they go crazy")
00:48:57BHSPitMonkeylinuxstb: happen to recognize this error?
00:49:23LinusNBHSPitMonkey: nope
00:50:53 Quit Thus0 ("Leaving")
00:51:50 Join hotwire [0] (
00:53:29 Quit mikearthur (Success)
00:53:52 Join BHSPitLappy2 [0] (
00:55:23 Join bagawk_ [0] (
00:57:47 Join Mongey [0] (
00:58:00linuxstbBHSPitMonkey: You're using an old version of gcc
00:58:45 Join mikearthur [0] (
00:59:00BHSPitMonkeythe cross-compiler?
00:59:17 Join BHSPitLappy [0] (
00:59:30linuxstbType "arm-elf-gcc -v"
00:59:40linuxstbYou should be using 3.4.x
01:00:45BHSPitMonkeyack, tis a PATH conflict
01:02:18 Quit bagawk (Nick collision from services.)
01:02:29 Nick bagawk_ is now known as bagawk (
01:02:32 Part lodesi ("Leaving")
01:03:04lokkii need help with peakmeter.c, anybody?
01:03:05BHSPitMonkeycool, building now
01:03:50BHSPitMonkeylinuxstb: "make[2]: *** No rule to make target `libTremor', needed by `all'. Stop.
01:03:58lokkii want to adjust the size of it...
01:03:58BHSPitMonkey"make[1]: *** [build-codecs] Error 2
01:04:56 Quit mikearthur (Remote closed the connection)
01:06:42linuxstbThat's a strange error. Did you re-run configure after fixing your PATH? You should probably delete your build directory and start again.
01:07:13BHSPitMonkeylinuxstb: yeah, i did, that's how i even got that much to happen.
01:08:21 Quit Mongey ()
01:08:24BHSPitMonkeyif it doesn't work this next time, i'm tempted to remove all vorbis/tremor content from the makefile...
01:08:27linuxstbWhere is your build directory? I think it needs to be in the same directory as "apps", "tools" etc
01:08:39BHSPitMonkeyit is
01:10:04 Join Strath [0] (
01:10:23BHSPitMonkeyoh, and when i try to build a simulator for nano it goes into the most horrible loop ever
01:11:18linuxstbYou could try downloading a fresh copy of rockbox-devel - it sounds as if something's wrong with your copy.
01:11:38linuxstbAre you doing all this under Linux?
01:11:56BHSPitMonkeycvs, updated not long ago
01:12:26linuxstbHave you tried building the bootloader?
01:12:48lokkiLinusN: i try to enable peakmeter for the remote (h-120) it's all fine, except for the size of it... do you know where i have to look in the code?
01:14:05BHSPitMonkeyshould I?
01:15:11linuxstbIt won't fix your problems building Rockbox, but you will have to install the bootloader if you want to try running it.
01:16:04 Quit Kohlrabi (Nick collision from services.)
01:16:05LinusNlokki: the width?
01:16:07BHSPitMonkeyalso why I really want ipodloader2 to support rockbox sometime :P
01:16:09 Join Kohlrabi [0] (
01:16:20lokkiLinusN: yes exactly
01:16:31linuxstbBHSPitLappy: Feel free to add it :)
01:17:05*BHSPitMonkey != genious
01:17:27lokkiLinusN: i see there is a variable width, i thought it depends on LCD_WIDTH, changing that to LCD_REMOTE_WIDTH doesn't help however
01:17:32LinusNlokki: gwps-common,c
01:17:38linuxstbYou just need to copy the function load_rockbox() from bootloader/ipod.c into the ipodlinux2 bootloader
01:17:42BHSPitMonkeyshould I try just cvs'sing the normal rockbox branch?
01:17:57lokkiLinusN: ok...thanks a lot i will look there
01:18:10BHSPitMonkeyand i will definitely look into that then...
01:18:10linuxstbBHSPitLappy: What do you mean? There is only one rockbox branch.
01:18:18LinusNlokki: replace LCD_WIDTH with display->width
01:18:36BHSPitMonkeylinuxstb: type? collection? whatever, in opposition to rockbox-devel
01:19:17lokkiLinusN: in peakmeter.c you mean?
01:19:38linuxstbBHSPitLappy: It's the same thing. But if I was you, I would try starting again with a clean checkout of rockbox-devel
01:19:40lokkiin gwps-common
01:19:52lostlogicheh, same problem happened when I made the A-B repeat patch for swcodec
01:20:05LinusNi guess peakmeter.c needs to adapt to the "display" concept
01:20:10lostlogicLinusN: a-b repeat patch could use merging for swcodec ;)
01:20:28LinusNlostlogic: :-)
01:20:59LinusNlostlogic: do you think we need to move some buttons around?
01:21:45 Quit Kyl3 ("( :: NoNameScript 3.81 :: )")
01:22:04BHSPitMonkeylinuxstb: also, i apologize for my limited familiarity with rockbox so far, but does rockbox build to a single file firmware? is there a filesystem?
01:22:09lostlogicLinusN: I believe at least one person who is using the patch moved them around. I dunno −− getting to settings would have to be inconsistent to deal with it currently.
01:22:22lokkithat was me
01:22:27lostlogicLinusN: either in rec or WPS there's going to be an inconsistency unless we use a long press for the settings.
01:23:15lokkii moved the menu button to rec, inconsistency in the rec screen
01:23:19 Quit BHSPitLappy2 (Connection timed out)
01:23:31lostlogicLinusN: do you think that there will be a record from wps function coming?
01:23:51LinusNrecord from wps? hardly
01:23:57linuxstbBHSPitLappy: There is a sort-of filesystem - typing "make zip" after "make" will create a file which contains rockbox.ipod (the main firmware binary), plus a .rockbox directory that contains fonts, plugins, codecs, themes etc
01:24:20lostlogicLinusN: then I'm inclined to leave it as the patch is and let people figure out that rec does A-B in WPS screen.
01:24:27linuxstbTo install, just then just type "unzip -d /mnt/ipod/"
01:24:46lostlogicbecause having a short press for the menu is a Good Thing
01:24:48 Quit Mark_ (Read error: 104 (Connection reset by peer))
01:25:03LinusNlostlogic: me too, but it sure feels awkward to not use the A/B button for the A/B function :-)
01:25:08lostlogicand consistency is so often lacking in other firmware products, but _not_ in rockbox
01:25:30linuxstbBHSPitLappy: But there is no partitioning of the type needed for ipodlinux - all files are stored on the FAT32 partition.
01:25:34LinusNrockbox is pretty inconsistent in many areas, but the menu button is the same everywhere
01:25:45BHSPitMonkeylinuxstb: i see
01:26:01lostlogichmm... I've only really noticed inconsistency in the plugins compared to the main OS
01:27:10webguest46lostlogic: thanks for your answer, i'm glad someone is working on it
01:28:58lostlogicwebguest46: It's not really a big difference, the functionality isn't likely to change, just the display.
01:29:50lostlogicLinusN: how much have you done with the PCF50606? Have you ever seen it get into a state other than the state it 'should be' left in by the iRiver firmware? I think I got it into such a state, which says to me we should set it's charge control registers on boot up rather than assuming they are set.
01:30:39LinusNlostlogic: we should set it the way we want to
01:31:09LinusNfor example, the exton wakeup can sometimes be disabled
01:31:57lostlogicLinusN: yeah, it actually has a pretty nice featureset that we are ignoring. I'll work through it more when I get my replacement H340.
01:32:28webguest46lostlogic: yes, i know the charging is done in hardware, but it would be good to know when the battery is full. many people say that it is bad to keep the charger plugged in when the battery is full.
01:33:57lostlogicwebguest46: As far as I can tell that's a falacy, the charger stops being in charge mode when it reaches full, I need to verify that it is not charging at all in this mode though.
01:34:36 Quit Kohlrabi ("Leaving")
01:34:41LinusNlostlogic: still, it will probably trickle charge
01:34:45lostlogicwebguest46: if necessary, I'll turn on the charge control feature of rockbox to have the software deactivate the charge completely when the battery is full.
01:34:59lostlogicLinusN: that would be really dumb on LiPoly, but you're probably right.
01:38:33LinusNseems the charger automatically switches off the charging after the fast-charge is done
01:38:51LinusNthen you have to select either trickle or switch mode afterwards
01:39:47LinusNhmmm, switch mode wasn't what i thought it was
01:39:56lostlogicthe MBCC registers during charge were 3F, 00 and 3A and the status was 5D or 1D depending on what part of the charge cycle it was in
01:40:45 Join BHSPitLappy2 [0] (
01:43:27 Join Jungti1234 [0] (n=jungti12@
01:43:42 Quit BHSPitLappy (Nick collision from services.)
01:44:22lokkiLinusN: your suggestion didn't work... but i guess it's because i used FOR_NB_SREENS and screens[] in peakmeter.c to replace lcd_* commands...
01:45:21lostlogicthat sounds like the right stuff, lokki
01:45:37 Join uwe_ [0] (
01:45:45lokkibut the peakmeter is still to big for the remote...
01:45:50lokkiim to dumb for this
01:46:34lostlogiclokki: you'll get it −− the code in peakmeter.c is probably called for each screen by the gwps-common.c, so you should only handle one screen in any given call to peakmeter.c
01:46:40lostlogicdoes that make sense for what you are working on?
01:47:03lostlogicyou probably need to add a 'struct screen * display' parameter to the peakmeter.c calls or some such
01:47:06 Join linuxstb_ [0] (
01:47:18 Quit linuxstb (Nick collision from services.)
01:47:22 Nick linuxstb_ is now known as linuxstb (
01:47:31 Quit akaidiot (Read error: 110 (Connection timed out))
01:48:39lokkiwell i don't quite understand, i'm editing peakmeter.c itself... so i have to add a struct there?
01:48:50lostlogiclokki: lemme take a glance.
01:49:18 Join Jungti1234__ [0] (n=jungti12@
01:50:52lostlogiclokki: you'll want to modify the function signature of peak_meter_draw FROM int, int, int, int TO int, int, struct screen *, int and then pass in the variable display from gwps-common.c instead of LCD_WIDTH
01:51:10lostlogiclokki: then you'll call screen->function instead of lcd_function inside the peak_meter_draw function
01:51:31lokkithanks a lot i'll write that down and give it a try
01:51:32 Join Midgey34 [0] (
01:53:07 Join YouCeyE [0] (
01:53:18***Saving seen data "./dancer.seen"
01:55:03 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:55:55 Join Rob2222 [0] (
02:00:12Jungti1234who are you Jungti1234__?
02:00:40 Quit Jungti1234 ()
02:01:16 Join Jungti1234 [0] (n=jungti12@
02:01:26 Quit muesli__ (Read error: 110 (Connection timed out))
02:01:36Mode"#RockBox +o LinusN " by ChanServ (ChanServ@services.)
02:01:45LinusNJungti1234: want me to kick Jungti1234__ ?
02:01:55Kick(#rockbox Jungti1234__ :LinusN) by LinusN!
02:01:55 Join Jungti1234__ [0] (n=jungti12@
02:02:20lokkiLinusN: you are so mean :-)
02:02:54lokkiok i'll have to get some sleep...will look into this tomorrow...good night all
02:03:01LinusNnite lokki
02:03:01Jungti1234good night
02:03:05 Quit lokki ("CGI:IRC")
02:03:18LinusNthe iriver metronome patch is really nasty
02:04:38Jungti1234ah sorry
02:06:57 Quit Jungti1234__ ()
02:07:10 Join Janco [0] (
02:11:26 Quit Janco (Client Quit)
02:12:49 Quit Paul_The_Nerd ("Leaving.")
02:15:16 Join Janco [0] (
02:16:48JancoHello there, any one working on color bitmap support?
02:18:00Midgey34linusN: where s this patch? its not on the tracker
02:19:48LinusNMidgey34: in the rb forum
02:25:04 Quit Janco ("CGI:IRC (EOF)")
02:27:32 Quit ]RowaN[ (Read error: 110 (Connection timed out))
02:32:31 Quit ehntoo (Read error: 104 (Connection reset by peer))
02:34:38 Join ehntoo [0] (
02:37:17 Join webguest16 [0] (
02:41:33 Quit webguest16 (Client Quit)
02:41:40 Part Midgey34
02:41:47 Join Midgey34 [0] (
02:49:13 Join mikearthur [0] (
02:53:40 Quit webguest46 ("CGI:IRC (EOF)")
03:03:08 Join ataxy [0] (
03:04:23 Quit [San] (Read error: 110 (Connection timed out))
03:06:16 Quit ataxy (Client Quit)
03:30:10 Join akaidiot [0] (
03:45:32Midgey34well, I have the basics of a blackjack plugin complete. now to add some sort of GUI so you can see whats going on
03:45:45 Quit mikearthur (Read error: 110 (Connection timed out))
03:53:21***Saving seen data "./dancer.seen"
04:17:18 Quit akaidiot (Read error: 110 (Connection timed out))
04:42:51 Join San [0] (
04:43:22 Quit eli_sherer (Read error: 110 (Connection timed out))
04:49:33 Part LinusN
04:51:48 Join webguest56 [0] (
04:51:56webguest56What is Xobox?
04:52:32 Quit webguest56 (Client Quit)
04:52:53 Part Midgey34
04:53:27BHSPitLappy2linuxstb: ping?
04:53:52BHSPitLappy2linuxstb: will arm-elf-gcc 4.x be acceptable for the nano, or does it *have* to be 3.4.x?
04:54:24 Nick BHSPitLappy2 is now known as BHSPitLappy (
04:57:09ashridah4.x produces code that's a bit too large, so probably not
05:00:49 Join atudar [0] (
05:02:04atudarHey everyone..
05:02:51 Quit atudar (Client Quit)
05:06:23 Join sargos [0] (
05:06:42 Nick sargos is now known as Sargos (
05:07:19SargosHey there - I'm having bootloader troubles with my iriver h300
05:09:38ashridahwhat's the problem?
05:09:39 Quit Sargos (Client Quit)
05:09:42 Join atudar [0] (
05:11:33 Quit atudar (Client Quit)
05:15:14 Join atudar [0] (
05:15:28 Join Rob2222_ [0] (
05:19:09 Quit Rob2222 (Read error: 110 (Connection timed out))
05:21:39 Join mikearthur [0] (
05:22:26 Quit mikearthur (Remote closed the connection)
05:25:57 Quit RotAtoR ()
05:28:05 Quit ashridah ("Leaving")
05:29:20 Quit atudar ("CGI:IRC")
05:30:27 Quit BHSPitMonkey ("There's no place like")
05:44:45 Join BHSPitMonkey [0] (
05:47:01 Quit Jungti1234 ()
05:47:01 Quit Rob- (Connection timed out)
05:53:24***Saving seen data "./dancer.seen"
05:58:14 Join akaidiot [0] (
06:11:05 Quit actionshrimp (Read error: 110 (Connection timed out))
06:18:32 Quit hotwire (Remote closed the connection)
06:20:50 Quit akaidiot (Read error: 110 (Connection timed out))
06:22:26 Join eli_sherer [0] (
06:50:20 Quit ehntoo (Remote closed the connection)
06:53:07 Quit YouCeyE ("Leaving")
06:57:59 Join DJDD_ [0] (
07:05:29 Join ashridah [0] (
07:27:33Bgermorning :)
07:40:38 Join Paul_The_Nerd [0] (
07:50:14 Join ender` [0] (i=ychat@
07:53:25***Saving seen data "./dancer.seen"
08:00:28 Join B4gder [0] (
08:00:44 Join akaidiot [0] (
08:13:32BHSPitLappywoot, building from windows is actually working. too bad i can't figure out how to give the simulator a .rockbox dir...
08:14:07B4gderBHSPitLappy: 'archos' in the same dir you run it
08:14:21BHSPitLappybut it doesn't contain .rockbox
08:14:26B4gderrun 'make install'
08:14:30 Quit godzirra (
08:14:30 Quit Ismo_ (
08:14:40BHSPitLappybuilding for the device at the moment, but thanks!
08:16:46BHSPitLappymetronome caused a make error on the device build
08:17:27B4gderrecent enough code?
08:17:46B4gderwell, I can see a fix in CVS for that problem
08:18:07BHSPitLappyk, updated...
08:19:40 Join Ismo [0] (i=laitinei@
08:20:02 Join godzirra [0] (n=shawn@
08:20:03***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
08:21:57BHSPitLappyi can run the sim without errors now
08:22:09BHSPitLappybut it just shows the header stuff
08:22:16BHSPitLappybattery, volume, clock...
08:22:33B4gderyou have any files in your simulated root?
08:22:33BHSPitLappyoh...there is a message
08:22:46BHSPitLappyload_config_buffer: header & checksum test failed
08:23:01B4gderthat's quite expected
08:23:11BHSPitLappyyeah, i have codecs, codepages, docs, fonts......
08:24:41B4gderin 'archos' ?
08:24:51B4gderah, you mean in the .rockbox?
08:24:51BHSPitLappyin archos/.rockbox
08:25:02B4gderrockbox doesn't show the dot files by default
08:25:28B4gderyou need to change that setting if you want to browse it
08:25:49BHSPitLappywhat exactly am i supposed to see by defauly
08:25:59B4gderthe contents root directory
08:26:02B4gderof the
08:26:07BHSPitLappysince i'm new to trying this whole "rockbox" scene :P
08:26:18BHSPitLappyso should i drop a test file in archos then
08:26:33B4gderit'll give you more to look at ;-)
08:26:43B4gderbut you can try hitting the menu key
08:26:44BHSPitLappyso how do you run stuff? :P
08:26:49BHSPitLappyyeah, does nothing
08:26:58BHSPitLappynone of the buttons do
08:27:06BHSPitLappyi get random Backlight messages on the console
08:27:08BHSPitLappybut that's it
08:27:25B4gderin the iriver sim, INS pops up the menu
08:27:30B4gderx11 sim that is
08:27:49 Join Kohlrabi [0] (
08:28:01BHSPitLappy"nothing to resume"
08:29:19B4gderthat's the resume function, not the menu
08:29:56BHSPitLappydunno the controls, all i got is the numbers painted over the ipod skin :P
08:30:20B4gderwell, when I run the x11 sim it prints the keymap on start
08:30:38B4gderwhat keyboard keys that simulate what target buttons
08:30:39BHSPitLappyshows how much better win32 is then, eh? ;)
08:31:09BHSPitLappyi'm only on cygwin because my builds don't work on my linux box for some reason
08:31:17BHSPitLappywere you asking?
08:31:40B4gderI'm on linux 100% of the time I decide on my own
08:32:11BHSPitLappyhow is that a sentence? lol
08:33:25B4gdersilly people force to use windows at times
08:33:28B4gderlike at work
08:33:31B4gderforce me
08:34:01BHSPitLappyi'm on linux on my desktop more than this windows laptop
08:36:14BHSPitLappywow, i actually generated a firmware!
08:38:00BHSPitLappycopied to ipod!
08:38:22BHSPitLappynow the only thing left to do is get the ipodloader2 modified to load rockbox XD
08:39:11B4gderI'll put my money on the rockbox loader
08:40:23BHSPitLappydoes the rockbox loader dual-boot
08:41:32 Quit Paul_The_Nerd (Read error: 110 (Connection timed out))
08:42:07BHSPitLappymeh, i'll look into it objective is to work that function into ipodloader2
08:42:13BHSPitLappysleep time now
08:45:39 Join Paul_The_Nerd [0] (
08:45:48 Part Paul_The_Nerd
08:47:35B4gdergood morning!
08:47:37 Quit akaidiot (Connection timed out)
08:52:13 Join LinusN [0] (
08:53:54amiconnLinusN: I wonder about the double remote button defintions in the metronome plugin...
08:54:50amiconnThey are #defined at the top, yet there is a function that maps remote buttons to non-remote buttons in a hard-coded way
08:55:02 Quit B4gder ("time to say moo")
08:55:34 Join B4gder [0] (
08:55:47LinusNi didn't bother to loke at the remote code, now i see that i should have :-)
08:55:52 Join einhirn [0] (
08:58:04preglowwhat is the waveform in metronome?
08:58:21LinusNa pcm of the mp3 "tock" sound
08:59:01preglowand what is that? a proper metronome sample?
08:59:08LinusNthink so
09:00:00preglowjust making sure it can't be synthesized :>
09:04:15BgerLinusN morning
09:04:26Bgerdid you see the comment in my cvs commit ?
09:07:10Bgeriirc the problem is that lcd_init_device didn't set display_on to true ... so the update funcs didn't update the display
09:07:58 Join Janco [0] (
09:08:30Bgerbut i don't want to mess with the bootloader code
09:09:05LinusNi'll see if i can remove the lcd_enable() again
09:09:14 Join hd [0] (
09:15:05Bgerbtw, any plans for new bootloaders ? :)
09:18:11JancoI would like to contribute to rockbox development. I wrote code for color bitmap support, compatible with the current monochrome support. Unfortunately I have no previous experience with C, so there are some things I would like to ask. Can any one help me?
09:18:29B4gderask away
09:19:58JancoSome functions (in lcd-16bit.c)are defined twice, almost identically, but once followed by ICODE_ATTR. What does it mean?
09:20:20BgerJanco the first is declaration
09:20:35Bgerand the second is the actual definition/code
09:20:50Bgerthe first is a.k.a. prototype
09:21:13Bgerand ICODE_ATTR means to put the code of the function in the SRAM
09:21:30Bgersubstitute means with "tells gcc"
09:22:04Bgerthe SRAM is a fast ram situated in the CPU
09:22:59Bgerso we put the time-critical code there
09:23:13Bgerdata too
09:25:39 Quit hd (Read error: 131 (Connection reset by peer))
09:26:12 Join hd [0] (
09:28:10 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
09:28:16 Quit goa (Connection timed out)
09:28:16 Nick hd is now known as goa (
09:31:27JancoBger: thanks. Another question, I test on H340, because the simulator works, but not for the wps stuff. Any idea why?
09:31:51Bgerhuh ?
09:32:09B4gderit certainly _should_ work for WPS too
09:32:28Bgerbut the audio stops on your own after some secs, iirc
09:32:35Bgeron his own
09:32:37JancoI'll try again then.
09:32:41Bgeruf :)
09:33:38 Quit goa (Read error: 104 (Connection reset by peer))
09:34:12 Join goa [0] (
09:42:04 Join linuxstb_ [0] (
09:42:41linuxstb_Morning all. Do I just need to define some buttons to get Metronome working on the ipod?
09:42:56Bgerprobably ...
09:46:39LinusNif the pcm playback works, yes
09:47:35 Join fobbo [0] (
09:48:23preglowdoubt that
09:51:08linuxstb_I'm interpreting "working" loosely.... I just want it to compile.
09:53:27***Saving seen data "./dancer.seen"
10:02:03 Join muesli__ [0] (i=muesli_t@
10:02:21linuxstb_OK, Metronome "working" on the ipod now.
10:08:46 Quit linuxstb_ ("CGI:IRC")
10:09:18preglow*tick* *tick* *KWZZZZ* *tick* *tick* *tick* *KSZCHT*
10:09:39eli_shererLinusN: how did you get that sound() array (from a wave file?), and how?
10:10:00preglowit's from the mp3 snippet in the metronome plugin
10:10:04preglowjust decoded
10:10:52linuxstbpreglow: To fix the corrupt audio, just add "p=start" after the second call to get_more() in my patch (inside the if statement)
10:11:18 Join amiconn_ [0] (
10:12:02preglowstill skipping?
10:21:14 Quit linuxstb (Read error: 104 (Connection reset by peer))
10:22:23preglowi wonder why it's doing that, there should be enough cpu for decoding
10:25:14 Join Zagor [0] (
10:28:22 Quit amiconn (Read error: 110 (Connection timed out))
10:28:23 Nick amiconn_ is now known as amiconn (
10:30:09 Join akaidiot [0] (
10:34:11JancoBger: I got the simulator working now. Strangely enough the color bitmaps appear to be correct on the real target (H340) but have different colors in the simulator (B instead of R, G instead of B, etc.). Will have a look into that tonight, probably.
10:34:59LinusNJanco: you probably have an endian problem
10:35:57dwihnoHas the gmail invitation program ended?
10:37:14preglowi've got tons of invites
10:37:44dwihnoAre you certain?
10:37:57preglowi'm looking it at
10:38:01preglow100 of them
10:38:26dwihnoI don't want to be a begger, but can you share one? :)
10:38:37JancoLinusN: Probably, but since the colors are right on the real target, shouldn't this be corrected by the simulator? Or do I have to take this into account in my code?
10:38:39preglowsure, just gimme an addy
10:39:07LinusNJanco: the sim is little endian, the target is big, and your bmp reader probably gets it wrong
10:39:36B4gderJanco: if your code is correct (and the sim does right), the sim and target should of course look the same
10:43:39JancoAs I said, I have not a lot of experience with this kind of programming, so I suppose my code is not correct. I suppose I do it wrong twice (reading and displaying), which makes it correct for the target, but not for the simulator. Will have to study this problem.
10:44:34B4gderyou run the sim on a x86 cpu, right?
10:46:37JancoAlso, I used my own internal representation for color bitmaps (nothing fancy). This might not be what the simulator expects. Has a rockbox color bitmap format been defined (as was done for monchrome)?
10:47:11B4gderthe framebuffer use the target format
10:47:16JancoB4gder, that's correct (cygwin under XP).
10:47:29B4gderand we use bitmaps in the same format as the framebuffer wants them
10:49:32B4gderbut I'm not the expert on color bitmaps/framebuffers
10:50:10 Join gtkspert [0] (
10:50:12JancoI don't have time now, I will look into the matter this evening.
10:52:07JancoBut I must say, it looks nice with colors. But I expect stills lots of problems. Part of it is the wps format.
10:53:13JancoI think it could be revised to admit absolute position (in pixels) of strings in any font.
10:53:35B4gderyes indeed
10:53:56LinusNeither we extend it so it can handle both b/w and color bitmaps, or we store the b/w bitmaps in color format
10:54:08B4gderI vote the for latter
10:54:17LinusNme too
10:54:35LinusNespecially when we go for the one-bitmap skinning method
10:54:37B4gderI believe eli_sherer is also working on color images in the WPS
10:57:41 Quit gtkspert ()
10:57:52 Join gtkspert [0] (
10:58:01JancoI don't think either is necessary. My reader detects the bitmap depth and stores it with the image. The screen refresh functions (don't remember the name) then calls a new function with the colr_depth as a parameter.
10:58:34B4gderfair enough!
10:59:07JancoThis either calls the existing monochrome bitmap draw function or a new color bitmap draw function.
10:59:17 Join gtkspert_out [0] (
11:00:37 Quit gtkspert (Client Quit)
11:00:56 Join gtkspert [0] (
11:01:31JancoIt would be more memory efficient than storing monochrome as color, but is probably slower (bu I am unable to say what is more important).
11:02:01LinusNspeed is pretty important on the h300
11:03:12 Quit gtkspert (Client Quit)
11:03:25 Nick gtkspert_out is now known as gtkspert (
11:03:27JancoI guess, but what about memory?
11:03:32preglownot that important
11:03:42preglowit is important, but speed is more important on h3x0
11:03:51LinusN"32mb should be enough for everyone" ;-)
11:04:02JancoOk! Thanks.
11:04:22LinusNcolor bitmaps will be way faster on the h300
11:04:23preglowi had less ram than that on my pc for a very long while
11:04:25preglowof course it's enough!
11:05:40 Part gtkspert
11:07:26 Join Polo_o [0] (
11:09:00 Join gtkspert [0] (
11:09:55 Quit gtkspert ()
11:10:04 Join gtkspert [0] (
11:12:05Bgerhmm, good gui ftp client for linux, which supports moving of files on the server ?
11:12:13Bgerheh, can be console as well
11:12:27B4gder'ftp' ? ;-)
11:13:38B4gderthe plain bog normal ftp client tool
11:14:30Bgeryep, i undestood
11:14:37B4gderbut I figure all ftp clients can do renames
11:14:59Bgernot sure about windows' ftp
11:15:12B4gderI think even that
11:15:20B4gdersince it is a BSD port
11:16:13Bgerreally ?
11:16:39B4gderreally on what part?
11:17:09Bger"it is a BSD port"
11:17:28B4gderand yes, it has rename
11:17:45Bgeryes, i saw it
11:18:01B4gderat least older windows had bsd strings left in the binary
11:27:55 Join KN|stiff [0] (i=phhome@
11:38:47eli_shererLinusN: how do you plan to implement "one-bitmap-skinning" with the current state of WPS files?
11:42:19 Quit KN|stiff ("( :: NoNameScript 3.81 :: )")
11:47:25 Join aliask [0] (
11:47:36 Quit akaidiot (Read error: 110 (Connection timed out))
11:47:53aliaskeli_sherer: Did you ever get that patch I sent you at MisticRiver?
11:50:08eli_shereraliask: what patch?
11:50:24aliaskIt was for the bmpviewer plugin, hang on, I'll find it.">
11:51:14Jancoeli_sherer: "One-bitmap-skinning" as in "drawing one big bitmap comprising all static graphics and then writing all text and dynamic elements on top of it"?
11:51:18aliaskBasically just made it so it only exits on stop.
11:52:13eli_shererjanco: yes.
11:52:39 Join linuxstb [0] (
11:53:31***Saving seen data "./dancer.seen"
11:53:32eli_shereraliask: i don't see the point of it...
11:53:58aliaskThe backlight goes out after about 5 seconds (or whatever it's set at), there is no way to turn it back on without the plugin exiting.
11:54:18eli_shererand btw i intend of disabling the viewing feature of rockpaint because i want to make one program called viewbox to open all sorts of images
11:54:19 Quit Janco ("CGI:IRC")
11:54:33LinusNone-bitmap skinning is not about drawing a bit bitmap and fill in the text
11:54:33 Join Janco [0] (
11:54:43aliaskYes I saw that. I sent the patch before you started working on that :)
11:54:53eli_shererLinusN: the way winamp 2.xx is...?
11:55:01eli_shererLinusN: is it?
11:55:12LinusNit's about storing all the tiny bitmaps in one big bitmap for faster loading
11:55:31eli_shereri think that's waht he meant..
11:56:53eli_sherermy current project (and i'm revealing it for the first time) is...:
11:57:02eli_shererXONIX port for rockbox
11:57:31aliaskI wonder how many hits google just got for xonix.
11:57:42 Join KN|stiff [0] (i=phhome@
11:59:08eli_shererdon't search to much you can find the "original" windows game here:
11:59:10aliaskAh I know this game. I think Microsoft released it under a different name about 10 years ago.
11:59:44eli_shererthe original name is QIX and it was developed a long time ago it even has an ATARI port...
12:00:36eli_sherermine is called (not so originally and expected of me) XOBOX...
12:01:21JancoIt's also a kind of n mp3 watch. For a moment I thought you were going to port rockbox to a new platform.
12:02:26eli_shererthe problem with this game is that i need to fill spaces with recursive loops which i already did
12:02:48eli_sherermy problem now is to get the aliens inside the game to jump from hiting the walls
12:09:23 Quit San (Read error: 110 (Connection timed out))
12:20:08 Join gtkspert_ [0] (
12:24:01 Part gtkspert_
12:24:07 Join gtkspert_ [0] (
12:25:27 Quit uwe_ ("Leaving")
12:27:28 Quit gtkspert (Read error: 110 (Connection timed out))
12:35:03 Quit linuxstb ("CGI:IRC")
12:36:24 Quit eli_sherer (Read error: 110 (Connection timed out))
12:36:42 Join Rob- [0] (
12:48:34 Join DocterD [0] (
12:56:20 Join NicoFR [0] (
12:58:52 Quit muesli__ ("ich will Khe!!!")
13:00:31 Quit goa (Read error: 104 (Connection reset by peer))
13:00:55 Join goa [0] (
13:08:04 Join Lost-ash [0] (
13:11:54 Join mikearthur [0] (
13:14:56 Join petur [0] (
13:22:23 Quit mikearthur (Read error: 104 (Connection reset by peer))
13:22:42 Quit gtkspert_ (Read error: 110 (Connection timed out))
13:24:17 Quit ashridah (Read error: 110 (Connection timed out))
13:25:00 Join mikearthur [0] (
13:25:58peturI'm still not quite satisfied with my recording gain settings in the recording screen
13:26:19peturIt's not easy to combine analog and digital gain in a meaningfull way
13:26:30 Quit mikearthur (Read error: 104 (Connection reset by peer))
13:26:31peturI currently have it like this:
13:27:12peturGain Left: A: 0.0 dB D: 0.0 dB
13:28:30amiconnWhy not just have one large-range gain setting, which is split into analog and digital part internally?
13:28:32peturand clicking left/right first changes analog, then digital so the ranges are appended but you still see what the individual settings are
13:29:24peturbecause I (and others) care about knowing what analog/digital gain is used
13:29:36 Join mikearthur [0] (
13:30:04peturthat's why my current implementation acts as a big range but displays them separately
13:30:10 Join akaidiot [0] (
13:31:24amiconnWhy would the individual values be important, provided the chaining is done in a way that ensures maximum quality?
13:32:28peturbecause if you can control the source as well, you want to make sure you're not using digital gain if that can be avoided...
13:32:49 Quit mikearthur (Read error: 104 (Connection reset by peer))
13:33:36peturafaik, the rockbox mentatlity is to give the power to the user and not think in his place?
13:34:07 Join mikearthur [0] (
13:37:42peturamiconn: if you're using a digital camera, optical and digital zoom are concatenated, but you do get an indication which one you're using. Because you don't want digital zoom unless you have no other choice...
13:38:29 Join linuxstb [0] (
13:39:08Bgerhuh ... does someone have a good scan of h300 units ?
13:39:22peturso maybe append an (analog) or (digital) indication behind a single value? but that still doesn't tell you the analog range untill you switch range :(
13:39:32linuxstbIs applying digital gain whilst recording the equivalent of applying gain to the WAV file after the recording has finished?
13:39:38Bgerthe images of h300 in DeviceChart point to non-existing files
13:45:57peturyes, but I read somewhere that the decimator has a much higher quality because it is applied before conversion to pcm
13:47:51 Join San [0] (
13:48:53 Join DangerousDan [0] (
13:48:59 Join muesli__ [0] (i=muesli_t@
13:51:00 Join muesli- [0] (i=muesli_t@
13:51:00 Quit muesli__ (Read error: 104 (Connection reset by peer))
13:52:33linuxstbpetur: What (digital) format is the audio in before PCM? I always thought that the output of ADCs was PCM.
13:53:32***Saving seen data "./dancer.seen"
13:53:51amiconnMaybe the decimator works on >16 bit data
13:54:26petursorry, don't know where I read it and don't have time to read the datasheet atm
13:55:11preglowaren't you discussing gains here?
13:55:26preglowa gain isn't a decimation
13:55:55preglowand no, adcs do not generally have higher precision internally
13:56:03preglowquite the opposite, they're usually 1 bit internally
13:57:20linuxstbSo is applying digital gain in the ADC better or worse than post-processing the WAV file?
13:57:50preglowi'd expect digital gain in the adc to be about the same as postprocess
13:57:59preglowbut of course, in our case, we don't utilise the full width of the dac
13:58:00peturI understood it is better
13:58:00amiconnpreglow: Not only internally, but the UDA is able to deliver more bits (24) than we can use with DMA access (16)
13:58:10preglowso using the uda control would be better
13:58:27preglowamiconn: yes, exactly
13:58:47preglowpetur: how can it be better?
13:59:05preglowpetur: in our case it will be, since we don't use all the 24 bits, but if we did, the adc can't gain the signal any better than we can
13:59:43 Quit mikearthur (Connection timed out)
13:59:56amiconnThat means, digital gain down to -48 dB won't reduce the resolution of the final .wav (16 bit)
14:00:11 Join mikearthur [0] (
14:02:24preglowwhat's the range of the analogue gain?
14:02:57petur0 to 24 dB
14:03:07petur(for line in)
14:03:11preglowso it can't attenuate?
14:04:20 Quit akaidiot (Read error: 110 (Connection timed out))
14:04:59 Quit KN|stiff ("( :: NoNameScript 3.81 :: )")
14:05:30preglowi don't think two meters for digital and analogue is very elegant
14:06:23peturyou mean the tow combined or below each other?
14:06:36preglowone slider for each
14:06:43preglowi think that's cumbersome
14:07:03preglowi'd rather have one option for gain, and an indicator telling me if i'm gaining digitally
14:07:10preglowbecause one slider is better than two
14:07:19pregloweasier, more convenient
14:07:38 Join eli_sherer [0] (
14:07:56peturso just add an indicator to show if the current gain setting is analog or digital...
14:08:14markunAnd the digital gain will not be used as long as analog gain is still possible
14:08:33preglowi think this matter should be resolved in the manual anyway
14:09:13preglowpersonally i think there should be a limit where the analogue gain limit is
14:09:34markunpetur: maybe you can display the value after which the digital gain will kick in
14:09:39preglow24db should be enough for everyone!!
14:09:51preglowbut no
14:09:59preglowan indicator when you're doing digital gain would be nice
14:10:05preglowtwo meters is unnecessary
14:10:15preglowsliders, settings, whatever
14:10:21peturI'll think of something ;)
14:10:21preglowmy head isn't working today
14:11:39peturfyi, decimation filter is explained on page 14 of the UDA1380 datasheet...
14:13:29preglowyeah, so? that's not what you're talking about?
14:14:17peturyou're right
14:14:52preglowthat's the downsampling filter, which does indeed entail decimation
14:14:55 Join Zoric [0] (
14:15:52 Quit DangerousDan (Read error: 104 (Connection reset by peer))
14:15:59Zorichi all. Is there a program that can scan and add replagain info to mp3s on linux?
14:16:39B4gderif there was, I would consider playing it with sometime
14:18:00Zoricthat meaning you dont know?
14:18:12B4gderI've only seen people mention foobar
14:18:32 Join DangerousDan [0] (
14:18:39wubbla_there's foobar for linux?!
14:18:45Zoricyeah. Thats what I've been using on windows. Now that I've begun using linux I find myself without that...
14:19:12B4gderand... that's also why I've never played with replaygain
14:19:20crwli don't think there's a tool to add replaygain tags to mp3 files in linux
14:19:23markunyou can use foobar with wine if you really want to
14:19:34Xerionjust google for replaygain + linux :P
14:19:36B4gderI don't use wine
14:19:40crwl(mp3gain for linux doesn't do what foobar does)
14:19:51Xerionthere's at least a replaygain script for amaroK
14:20:11markunToo bad mp3gain still doesn
14:20:16markun't store the info in tags
14:20:21Zoricand that script can scan files and add info?
14:20:31markunIt now stores the undo info in apev2 tags
14:20:43Xeriondon't know, only googled it ;)
14:21:05crwlXerion, it's an ugly hack, and it only enables amarok to play back replaygained files correctly
14:21:06Zoricyeah read something about it on the wikipedia page i think
14:21:18Zoriccrwl: ok thanks
14:21:21crwllast time i checked, that script didn't add the replaygain info anywhere
14:21:55crwlit's amazing that the amarok people have been unable to add proper replaygain or gapless playback support to the player
14:22:05crwlthose must have been on the wishlist for more than year now
14:22:17Zoricthis is sad. Now im gonna get an album that has a lot more volume than all others...
14:22:17markunZoric: just switch to ogg vorbis :)
14:22:21crwlinstead they're fiddling with some half-useless database stuff etc
14:22:35Zoricmarkun: why would that solve it?
14:22:37crwlprobably they're just too busy fighting the horrible amount of crash/database corruption bugs the player has
14:22:38markunZoric: you can use mp3gain
14:22:51markunbecause the vorbisgain tool wors very well
14:22:58 Quit Bger (Remote closed the connection)
14:22:59markunworks even..
14:23:14 Quit DangerousDan (Read error: 104 (Connection reset by peer))
14:23:19Zoricoh ok. Good to know.
14:23:37 Join Bger [0] (n=Bager@
14:23:55ZoricIf I actually owned the cd I'd rip it into ogg, but now I dont... so its in mp3.
14:24:03crwlfor flac files you can use metaflac −−add-replay-gain
14:24:45 Part petur
14:24:46Zoricoh well. I'll just have to fix it next time I boot into windows.
14:25:02markunZoric: which linux player do you use?
14:25:13Xerionthere seems to be a linux replaygain scan binary on there...
14:25:39 Join DangerousDan [0] (
14:27:49markunI can't find if Rhythmbox supports replaygain for mp3..
14:28:37crwli haven't seen a single player that does...
14:28:56crwla player that supports replaygain info stored in id3 or ape tags, that is
14:28:58markunZoric: if you use mp3gain it will change the gain of the mp3 file, not store it as a tag. Is that what you want to do?
14:29:16 Quit DJDD_ ("Trillian (")
14:30:01preglowhahah, rude, me?
14:30:05Zoricno I want to store it in the tag
14:30:08preglowrecruiting people for rockbox in #ipodlinux
14:30:31Zoricand rhythmbox is supposed to be able to playback files with replaygain info
14:30:39 Join alphakiller [0] (n=bohas@
14:30:40markunI'm trying the same at, but they wont bite..
14:30:59 Quit linuxstb ("CGI:IRC (EOF)")
14:30:59preglowalphakiller: you ever built a cross compiler before?
14:31:14alphakillerfor tons of platforms
14:31:19 Join linuxstb [0] (
14:31:20preglowwhat you basically need is a gcc 3.4.5 for arm-elf
14:31:26preglowand newest binutils for the same platform
14:31:48preglowwe use only c and assembler
14:31:53preglowso no need to build c++ compiler, etc
14:31:57markunZoric: do you have wine?
14:32:08alphakillermaybe I might do something nasty on iPod :P
14:32:19alphakillerlike get java OS working on it
14:32:29*preglow crosses himself
14:32:46alphakillersome of my fun stuff
14:33:48 Nick Lost-ash is now known as ashridah (
14:34:12preglowwe've got quite a lot of stuff to working before we can do something interesting, though
14:34:30preglowcurrently button driver is working, threading, disk and some other stuff
14:34:32alphakilleryeah, I know
14:34:37preglowwe're having some trouble with the cpu cache
14:34:40alphakillerwhat's the funny part :D
14:34:48linuxstbalphakiller: If you want to help with Rockbox (or ipodlinux) on the ipod video, then the main skill at the moment will be reverse-engineering the Apple firmware. There is lots of new stuff in the video which isn't known yet - and obviously no public docs.
14:35:01preglowlinuxstb: still no info on the sound chip?
14:35:09linuxstbNot as far as I know.
14:35:21preglowwoops, bbl
14:35:41alphakillerlinuxstb: I'm just digging in coz I just wanna play with arm stuff around for a while ..
14:35:46linuxstbThe chip itself is a similar wolfson dac to the other ipods, but I don't think any of the IPL people have investigated it yet.
14:36:08 Quit fobbo (Connection timed out)
14:36:21linuxstbalphakiller: Rockbox is running on the video, so it's already in a usable state. The two main problems are audio (as I mentioned) and the LCD is currently operating quite slowly.
14:36:38alphakillermy next job is to develop an arm based platform, and develop a SOC, with a firewire port, USB and wifi
14:36:59alphakillerthat's why I am digging in on it :D
14:37:04linuxstbThe LCD is connected to a proprietory Broadcom chip - the instruction set isn't even known (but suspected to be MIPS-based). So that's a big hurdle.
14:37:05alphakillerso my boss pays me
14:37:25alphakillerto keep me up playing with Open Source Software
14:37:35alphakillerin order to develop new skills in order to work for them :D
14:37:43B4gdergo go go! ;-)
14:38:00alphakillerwell, I've played with mips (playstation 2 and playstation one)
14:38:28alphakillerjust dunno how the things are connected :P
14:38:42alphakillerI just grabbed an iPod video from japan ..
14:38:53alphakillermy brother didn't like my fathers gift heheheh .. he's a beast
14:39:02alphakillerso he just left it for me :P
14:39:27 Quit muesli- (Read error: 110 (Connection timed out))
14:39:39linuxstbI have a 60GB ipod color (the last one before the video was launched), plus occasional access to my girlfriend's 30GB video. I like the video a lot.
14:40:04alphakillermine is the 30gb video
14:40:17alphakillerI'm very interested on the task :D
14:40:32alphakillerthis is my opensource page
14:40:41alphakillerof things I've been doing so far
14:40:50alphakillermicrokernels, playstation ports, etc
14:41:33linuxstbIt would be nice to have a "full-time" video user helping - there's lots of stuff to do.
14:41:58alphakillerI've developed some strange things done too
14:42:12alphakillerdunno if my skills fullfil the profile of the requested developer
14:43:08alphakillerhmm ... I know asm for some architectures, C programmer and some base at eletric engineering. more on eletronic digital devices development (e.g. I developed my own cpu :P)
14:43:46B4gderthen you'll enjoy arm
14:43:52B4gderquite frienly asm
14:44:09alphakillerI've heard that :P
14:44:34B4gdersetup the cross-compiler, get the sources and enjoy! ;-)
14:44:38LinusNexo-kernel is pretty similar to the rockbox kernel
14:44:53alphakillerwhich exokernel ? mine?
14:44:57LinusNlongjmp task switch
14:45:28alphakilleryeah ... that was a little exokernel in order to avoid "knowledge" of specific processor target
14:45:47alphakillermy latest job is the Oxygen package
14:45:50alphakilleron the download area
14:46:17alphakillerI was working on something like a Media Object Oriented OS :P
14:46:22markunalphakiller: what kind of company do you work for?
14:46:52alphakillera small one in Brazil that develops all kind of embedded hardware/software for local banks here
14:46:57LinusNit doesn't mention what oxygen is
14:47:01alphakillerbanks and gouvernament
14:47:06 Join Maxime [0] (
14:47:09alphakilleroxygen is a microkernel
14:47:21alphakillerwith a OS-Layer sat on top of it
14:47:26alphakillerfor now, it's just an OS-Layer
14:47:29alphakillerand microkernel
14:47:54LinusNWarning: mysql_connect(): Can't connect to MySQL server on '' (111) in /home/groups/c/co/comos/htdocs/new/news.php on line 2
14:48:01 Quit Maxime` (Read error: 104 (Connection reset by peer))
14:48:18alphakillerthe sql db is broken :(
14:48:18alphakillerneed to fix it
14:49:00alphakillerwell, I've written some OS around there ...
14:49:16alphakillerthis Oxygen, my plan is to implement specific media objects in the kernel :P
14:49:32alphakillerto optimize media data manipulation
14:49:41alphakillerbut it would be very hardware specific
14:49:54linuxstbWhat kind of hardware do you have in mind?
14:50:05alphakilleriPod ?
14:50:39alphakillerI'm tending for hacking iPod :P
14:50:43linuxstb:) Looks like there will be a third OS for the ipod then...
14:50:54 Join elisherer [0] (
14:51:03alphakillerfor now, I would like to help you guys
14:51:08alphakillerI never worked with arm before
14:51:20alphakillerjust with mips, alpha, powerpc, and m68k
14:51:55 Quit B4gder ("time to say moo")
14:52:55alphakillerI guess I'll have to re-install my linux here
14:53:14alphakillerI was playing with windows to avoid work on my vacations
14:53:24alphakillerso I had the psychological excuse to don't code
14:54:21alphakillerwell, what I really need here ?
14:54:33 Quit Janco ("CGI:IRC")
14:54:46alphakillera unix box, nah ?
14:55:28alphakillerwell, first of all, If I were you, I would make a package called "toolchain" :P
14:55:36linuxstbYou need to install gcc (3.4.5) and binutils (2.16) cross-compilers (target=arm-elf) and the Rockbox source code from CVS.
14:55:38alphakillerwith buildscripts
14:55:51linuxstbYou can use either a Unix box or Cygwin.
14:55:59LinusNor mac osx
14:56:01LinusNor bsd
14:56:05linuxstb"Unix box"
14:56:11 Quit eli_sherer (Read error: 110 (Connection timed out))
14:56:22markunthe sdl simulator has some problems with bsd btw
14:56:30 Part Zoric
14:56:42alphakillerI don't have access to a mac ..
14:56:47LinusNmost *nix programs have some problem or another with bsd
14:56:53preglowgod, i hates it
14:56:57markunTrue :)
14:57:03alphakillerfor now, I just have a windowze box
14:57:09alphakillerI'll just return to work on march
14:57:21preglowi tried using a freebsd box for development, but after i started longing for death after using it half a day
14:57:25alphakillerI'll have from 21th january until march 3rd to hack
14:57:28linuxstbalphakiller: The only problem you'll have with Windows is installing the bootloader - you need raw access to the hard disk.
14:57:45preglowvim did not go very well with freebsd
14:58:01markunpreglow: what was the problem?
14:58:06preglowi have no idea
14:58:06linuxstbThis page contains the basic instructions for installing the Rockbox ipod bootloader:
14:58:09preglowcolours refused to work
14:58:19markunah, the TERM thing..
14:58:28alphakillerbut have you heard someone installing the bootloader on windows ?
14:58:29preglowand saving was tedious, since vim all the time reminded me it couldn't write a backup file, this despite me having turned off backup
14:59:03linuxstbalphakiller: No. preglow tried and failed. But I have an idea that may work - I just need to find the time to reboot into Windows and test it out.
14:59:05markunI can't remember having these problems.
14:59:13markunMaybe god hates you, not bsd..
14:59:53linuxstbThe problem is that the ipod's firmware is stored in a small (40MB-80MB) partition marked as "empty". Cygwin doesn't give you access to this device as a normal /dev/sd?? device
14:59:55preglowgod hates me, i hate bsd
15:00:12linuxstbs/to this device/to this partition/
15:00:17alphakilleryeah ..
15:00:20markunAs long as we all love rockbox..
15:00:22alphakillerI got the whole idea :P
15:00:41alphakillerI can't wait to hack my iPod :P
15:00:52 Quit DocterD (Remote closed the connection)
15:01:00alphakillerI just dunno what will I do :P
15:01:15preglowyou'll figure it out
15:01:22linuxstbI had the same problem on Mac OS X, but solved it by writing a small utility that accessed the "whole disk" device instead of the devices created for the individual partitions - /dev/disk1 on the Mac.
15:01:31aliaskonce you start using the firmware you will notice things that you might like fixed.
15:01:53markunalphakiller: port the flite text-to-speech program to rockbox
15:02:14linuxstbI haven't tested that program on cygwin, but I think it should work -
15:02:31markunalphakiller: or the speex codec
15:02:41linuxstbAlso, the ipod video needs slightly different instructions for installing the bootloader - I'll try and update the Wiki.
15:03:17linuxstb(a firmware image that runs on the Broadcom chip needs to be in the boot partition, along with the normal ipod firmware)
15:03:20alphakillerI'm not going to port a high level thing :P
15:03:32alphakillermy target around here, is to hack arm
15:03:44alphakilleran existing arm platform
15:04:25linuxstbThe ipod has two ARM cores - that could be an interesting aspect to explore. Currently, Rockbox puts the COP (coprocessor) to sleep and only uses the main CPU.
15:04:45alphakillerhmm ... like psp :P
15:04:54alphakillerpsp has two mips r4000 processors :D
15:04:59alphakillerone main cpu and another as cop
15:05:04linuxstbYou'll be right at home then...
15:05:22alphakillerhehehe ...
15:05:33alphakillerplaystation 2 has 5 or 6 cops
15:06:19markunany idea yet what the cop should do? It can only run at speed <= the main CPU, right?
15:06:20preglowyeah, but they're not cpus
15:06:29preglowthey're specialised units
15:06:41preglowyou can scale the cope somewhat
15:06:42alphakillermarkun: decoding ??
15:06:47preglowand i'm planning on using it for decoding and dsp
15:06:54preglowand fiq handling
15:06:59linuxstbI believe the cop can run at either 100% or 50% of the main CPU's speed.
15:07:12linuxstb(or 0% if it's sleeping)
15:07:32alphakilleryou just assign task for the cop and he will do markun
15:07:39alphakillerI just dunno how these tasks are assigned to the cop
15:07:39markunwhy not use the main CPU for decoding and the COP at 50% for some less intensive tasks?
15:07:53alphakillerin the mips based architecture you send like "packages" for the coprocessors
15:08:08alphakillerand you don't need to stop the current execution on the main processor
15:08:14alphakillerto send such data
15:08:19alphakilleryou just send the packages
15:08:23alphakillerand continue processing
15:08:39alphakillerno no ..
15:09:00alphakillerthe usage of cop, is minimize the cpu overhead
15:09:31alphakilleryou give it specific and hard tasks
15:09:40preglowyou can give it what you want
15:09:43preglowour cop is just another arm
15:09:52preglowspecced just like the main core
15:09:56alphakillerbut usually you give hard tasks ..
15:09:59preglowyou've got mailboxes and iram, that's the only means of communicating
15:10:06preglowsure, i want the cop to decode
15:10:38 Quit Vladoman (Read error: 110 (Connection timed out))
15:10:43markunpreglow: but then you also have the main CPU running at 100%, even if it's not needed..
15:11:17 Join Vladoman [0] (
15:11:44 Join grahamoz [0] (
15:11:55alphakillerI guess it won't affect your battery power consumption
15:11:57preglowmarkun: well, the option is doing it the other way around, having the cpu deal with decoding
15:12:04preglowalphakiller: it will
15:12:16markunpreglow: yes, that was what I was suggesting
15:12:17alphakillerdepends on how the things are connected...
15:12:41grahamozhmm rockbox ?
15:12:44grahamozfor ipd ?
15:12:54alphakillerI guess Apple's OS on iPod turns it up just for MPEG4
15:12:56markungrahamoz: what about it?
15:13:00alphakillerand bitmap processing
15:13:45alphakillermy best guess, is that mp3 is decoded by a FPGA
15:13:50preglowit's not
15:13:51preglowpure software
15:14:05preglowall codecs apart from video are decoded on the arm
15:14:13preglowvideo is decoded on the broadcom chip
15:14:27alphakillerthat you guess it's a mips based proc ?
15:14:37preglowwe actually have that covered, we can decode mp3/ogg/etc on ipod now, we just lack good sound output
15:14:38alphakillerI see
15:15:17alphakillerit's strange having a mips to do that :P
15:16:11alphakillerman, I pod is damn perfect :P
15:16:36alphakillerinterface, button
15:17:14markunIs it?
15:17:33*alphakiller just wants to hack it :D
15:17:44alphakillerplay on low-level stuff of it :D
15:18:13markunalphakiller: how far are you with compiling the cross compiler then? :)
15:18:53alphakillerI'm doing other tasks by now ..
15:19:07alphakillercollege admissions exams
15:19:09lostlogiconce we get rockbox on iPod, H3x0 and iAudio X5, someone needs to do a serious audio comparison between them.
15:19:47markunGave up the hope on the Gigabeat port? ;)
15:20:32lostlogicmarkun: gigabeat seems to be the least likely to get line-in recording, so I'm more inclined to ignore it than the others.
15:20:47 Join DocterD [0] (
15:21:17markunI thought you were talking about output quality..
15:21:17lostlogicmarkun: realistically though, the gigabeat port should be easier than the iPod port since it has just one higher clocked ARM rather than having the complexity of controlling two.
15:21:49markunAnd we have the documentation
15:22:02aliaskHows the gigabeat port going markun?
15:22:04alphakillerhehe, I still prefer two slower cpus than one :P
15:22:12markunnot working on it right now
15:22:34lostlogicmarkun: i was... but those three players have pretty comparable feature sets
15:22:44markunI was disassembling some of the programs that came with the firmware update yesterday
15:22:56markunlostlogic: I
15:23:05markunlostlogic: I'll mod mine to have line out then :)
15:23:08alphakillerhaving two cpus you have the advantage of having less OS code that does threading running .. that's where you lose performance
15:23:42preglowwe'll need more os code for threading
15:23:44preglowfor two cores
15:23:51lostlogicalphakiller: not with a round robin cooperative scheduler like rockbox
15:23:53preglowahh, right
15:23:54preglowlike that
15:23:55preglowforget me
15:24:06preglowlostlogic: he meant less time spent in context switch
15:24:38lostlogicah, right. yeah, having one core that can be dedicated to the single highest CPU task, on one thread is definitely beneficial.
15:25:08lostlogicpreglow: I would think that what I just said is the better way to deal with it than having the PCM output and codec on the COP...
15:25:10alphakillerheheh, Apple Engineers are the best around there :P
15:25:41alphakillerI have had lots of small devices around there
15:25:51alphakillerthe bests I had were from Apple
15:25:56LinusNi wouldn't want a transparent scheduling between the two cpu's
15:26:21preglowLinusN: you wont have it either
15:26:33preglowthe cop will run a fixed set of threads
15:26:36preglowperhaps only one
15:26:40alphakillerthat's what I said
15:26:45LinusNthen i'm with you
15:26:47 Nick elisherer is now known as eli_sherer (
15:26:55preglowif we didn't do this, we'd have cache issues
15:26:59pregloweach core has its own cache
15:27:04preglowand the bus isn't snooped
15:27:15alphakillerwell, that's what I would like to play with :D
15:27:21preglowthen hooray!
15:27:28preglowi certainly wont stop you
15:27:30alphakillerdevelop the "driver" to send stuff to the cop
15:27:32amiconnImho the scheduler for the COP should be easy. What's more complicated is the cache coherency, as we surely need to pass data between both cores
15:27:42preglowamiconn: not much data, i'd think
15:27:44LinusNand the thread synchronization
15:27:47preglowamiconn: and for what little there is, iram is good
15:28:00alphakillermy idea is not having a scheduler on the cop
15:28:14alphakillerthe cop just will run a thread
15:28:20alphakillerdetermined and packed by the cpu
15:29:11markunToo bad the speeds can't be canged independantly..
15:29:12alphakillerso a main task sends a thread to the cop
15:29:12alphakillerthat's the idea
15:29:12alphakillerthat's where you reduce the overhead of context swithc
15:29:12DBUGEnqueued KICK alphakiller
15:29:36alphakillermy idea is this:
15:30:45alphakillerMain Task −−> ask for cop −−> turns it on what speed it would likes to −−> package thread −−> send packet to cop −−> do another thread from the task in parallel on the cpu, while cop works on another thread
15:30:57markunHow does such a mailbox work btw? You can write something to it and the other CPU gets an interrupt?
15:31:25alphakillerI've never worked with mips before
15:31:38markunwith arm.. right?
15:31:41alphakillerI dunno even the global memory is shared between the two processors
15:32:33preglowmarkun: don't really know, you write data somewhere, then the cop gets an interrupt
15:32:34alphakillerif we have the global memory shared between them, we could stablish a global pointer where cop tasks will write to
15:32:42preglowalphakiller: cache issues
15:32:54preglowalphakiller: we'd have to flush the cache before we did that
15:33:01preglowalphakiller: it's better to use the iram, since that isn't cached
15:33:13alphakillerhmm ..
15:33:38alphakilleryeah, but my idea is to have an unified global point down somewhere
15:33:41alphakillerirqs are pretty slow :P
15:33:48preglowwe don't need to use irq
15:34:30markunit's faster to check the memory location from time to time?
15:34:32preglowthe cop can be stuck in a loop, reading a variable from iram to see if it should wake, and then the cpu can send it to sleep, set the variable whenever it needs to, then wake the cop again, so it can work
15:34:47preglowsee if it should work, that should be
15:34:59preglowthe cpu can freeze the cop whenever it wants to, we should use that
15:35:07eli_shererprobably gonna cause the use of "wait" and "signal" if there is a mutual resources...
15:35:19alphakillerhmm .. and rather easy to do that :D
15:35:56markunalphakiller: well, what are you waiting for then, get to work :)
15:35:59alphakillerbut it should have a data flux from the cop to the cpu and vice-versa
15:36:47alphakillerand something to input routines ...
15:37:01alphakillermy idea is to do something like:
15:37:12alphakillerhook_CopCode(pointer function)
15:37:25alphakillerunhook_CopCode(pointer function)
15:37:54preglowjust keep in mind that the cpu and cop can't share data via regular memory
15:38:11preglowthat will require us to flush the cache all the time, which isn't very wise
15:38:34alphakillerso we need a way to copy thread code somehow
15:38:45LinusNalphakiller: unhook?
15:38:52alphakillerand store it in a fixed address
15:39:02alphakillerLinusN: link and unlink code
15:39:09preglowyou don't need to copy any thread code
15:39:14LinusNyes, but why unlink?
15:39:15preglowjust load it, have the cop flush its cache, then execute
15:39:52LinusNand what is the pointer in the unlink function?
15:40:14alphakilleroh .. forget heheh
15:40:18alphakillerthe pointer()
15:40:22alphakillerit's void :D
15:40:29alphakillerunhook(void) :P
15:40:34LinusNok, then i'm with you
15:40:46alphakillerand the pointer it self is the flag
15:40:53alphakillerunhook sets pointer to null
15:41:08alphakillerand hook setspointer to something
15:41:09LinusNok, so we could just have one function: link(ptr)
15:41:09alphakillerwe just have a loop
15:41:21LinusNand link(NULL) to unlink
15:41:22 Quit ashridah ("Leaving")
15:41:36alphakillerand then loop forever
15:41:45alphakillerif pointer null, do nothing
15:41:50alphakillerelse, call function :D
15:42:16LinusNperhaps cop_run(ptr);
15:42:22alphakilleryeah :P
15:42:23alphakillermore pretty
15:42:39alphakillerpreglow ?
15:42:44LinusNand a few others, like cop_put_msg(ptr) etc
15:42:59alphakillerwe are discussing cop issues :P
15:43:09preglowsure, i follow you
15:43:14alphakillerI was waiting for your point :D
15:43:22LinusNcop_run(NULL) <−− cop killer ;-)
15:43:31alphakilleryeah :D
15:43:32preglowLinusN: no, it'll just reset :)
15:43:52preglowthat's where the reset vector is, heh
15:43:57LinusN- there she is, set for "stun"
15:44:13alphakillerso, it's rather easy having the cop in a loop
15:44:15preglowi wonder what the cpu will do when the cop goes in for reset...
15:44:30alphakillernothing :P
15:44:32preglowbut yeah, just a loop waiting for a flag on where to go
15:44:43preglowand when the cpu knows the cop is in this loop, it also tells it to sleep
15:44:47preglowso we don't waste too much battery
15:44:51alphakillerI prefer the loop idea :D
15:44:54alphakillerand sleep
15:45:03alphakillerso we would have 2 procedures:
15:45:04alphakillerwake up
15:45:07LinusNloop: sleep;jmp loop;
15:45:08alphakillerand assign task
15:45:13 Join akaidiot [0] (
15:45:22alphakillerif assign task == NULL
15:45:31alphakillerthe cop put it self to sleep
15:45:34LinusNpretty much like the rockbox kernel then
15:45:56alphakiller(I think that's possible, putting it self to sleep)
15:46:11preglowyes, it can
15:46:24 Quit eli_sherer (Read error: 104 (Connection reset by peer))
15:46:32alphakillerhehehe, so we are agreed to write that ?
15:46:55alphakillerI've done something like that on my Trap hooker on x86/powerpc
15:51:35 Quit DocterD (Remote closed the connection)
15:53:35***Saving seen data "./dancer.seen"
15:53:48linuxstbalphakiller: When do you plan to start work on your ipod? I need to update the installation instructions to cover the video.
15:54:07alphakillerI plan on january 21th
15:54:12alphakillerthat's when my exams are over
15:54:50linuxstbOK. I should hopefully have installation from Windows/Cygwin sorted out by then.
15:56:07preglowtrap hooker?
15:56:20preglowisn't trap a software int on powerpc?
15:56:59alphakillerall the vectors are shared
15:57:21alphakillerso I need a way to call of them, and anyone handle the vector as he wishes
15:57:45alphakillerthat's my idea :P
15:59:04alphakilleranyway lunch time
16:00:14aliaskHrm, out of interest, what are people's upload speeds? I'm stuck here downloading the sim at 9kb/s because my friend's linux box is on a 128kbit upload.
16:07:11preglowupload speed to where?
16:07:19preglowcurrently, it's 100mbit/s
16:07:25preglowif you just mean in the general case
16:08:02markunfor me too
16:09:49aliaskTo the general internet?
16:10:02alphakillerpreglow: saw the sample code?
16:10:08alphakillerlinusN ?
16:10:46alphakillerso we just have a permanent "nanokernel" running on the cop
16:10:51alphakillerthat calls the thread
16:10:56alphakillerotherwise, sleep
16:11:27alphakillerthen, the cpu awakes the COP before assigning a new task
16:12:33alphakillerthis is the simple part I guess ... the complex part it would be develop an API to share data
16:12:53preglowi'm a bit busy, but it looks good
16:13:11LinusNalphakiller: looks pretty much like i imagined it
16:13:38alphakillerat least seems I have something to contribute :P
16:17:32alphakillercan't wait for start to play with :P
16:20:28alphakillerI just dunno if I can work properly with my iPod Video
16:23:43alphakillerhas anyone wrote something here for iPod video ?
16:23:58preglowrockbox works ok enough
16:24:06preglowso you've got something to build on
16:24:25alphakillerhave you wrote a hello world for it ?
16:25:38alphakiller(at least)
16:26:38alphakilleroh, sorry for my terrible english :P
16:26:42alphakillerI'm not a native speaker
16:27:31LinusNneither are we :-)
16:30:41 Quit Seed (Read error: 110 (Connection timed out))
16:33:07 Join saa[b_r]ider [0] (n=saab_rid@
16:33:42preglowwe've got hello world, bejeweled, mp3 decoders, etc
16:34:07preglowif it's assembler you're interested in, there's not much of that in rockbox for arm yet, though
16:34:16preglowbut rockbox does work for ipod video
16:34:21preglowit just doesn't do everything it should yet
16:34:23preglowlike not crash often
16:35:01alphakillerI guess we are going to play with asm on arm :P
16:35:14saa[b_r]iderdoes audio already work on the ipod video?
16:35:18alphakillerin order to awake and sleep
16:36:50 Join DrMoos [0] (
16:39:37preglowi have already played with asm on arm
16:39:46preglowcrt0.S and firmware/threads.c has some asm i wrote
16:40:21saa[b_r]iderpreglow: does audio already work on the 5G iPod?
16:42:41alphakillerI'm downloading cygwin
16:43:44preglowsaa[b_r]ider: no
16:43:54preglowwe have no idea have audio even works in 5g ipods
16:44:15alphakillerI have no idea if the code I imagine will work
16:44:35saa[b_r]iderwho owns a 5G ipod from the developers or testers?
16:44:44preglowno one, ipodlinux has access to one, i believe
16:44:48preglowbut i don't think anyone else has got one
16:45:06linuxstbI have access to one.
16:45:06saa[b_r]iderI was only aware of your nano, and linux's 4G.... ah
16:45:28linuxstbIt's my girlfriend's, but as we live together, I have 24-hour access if I want.
16:46:11saa[b_r]idershe allowed you to fiddle with it?
16:47:27preglowlets hope so
16:47:30 Quit akaidiot (Read error: 110 (Connection timed out))
16:47:33saa[b_r]iderI'll be getting one for the mrs., but I think it will be a while before I'd be allowed to put rockbox on it ;)
16:47:55linuxstbAs long as I restore it to the original state, she doesn't mind.
16:50:03linuxstbAnd to answer the original question, the existing ipodlinux audio code doesn't work on the 5G.
16:53:31saa[b_r]iderI've never owned an iPod, but I've heard of iPodlinux.. what's the connection between RB and ipodlinx when it comes to iPods? is it needed to load RB?
16:55:31linuxstbThey are completely independent.
16:55:46linuxstb(apart from the fact that a lot of ipodlinux code is now in Rockbox).
16:55:47 Join Kohlrabi [0] (
16:56:11alphakillerlinuxstb: I'm downloading cygwin
16:56:25linuxstbThe Rockbox bootloader on the ipod can triple-boot between the original Apple firmware, Rockbox and a linux kernel.
16:57:50alphakillerlinuxstb: is cygwin fine for dev ?
16:58:05linuxstbYep - I think preglow spends most of his ipod time in Cygwin.
16:58:17linuxstbBut it's slow compared to Linux.
16:58:51linuxstbBut you can use whichever is more convenient.
16:59:09alphakillerdid you see our multicore idea ?
16:59:21alphakiller( I guess you didn't read all stuff ) :P
16:59:44linuxstbYes, I read that. I'm trying not to think about too many things though...
17:00:03alphakillerhmm :D
17:00:13alphakillerwhat people here are active devs ?
17:00:32alphakillermost of ppl are just enthusiasts and just expect things happen around here?
17:00:46aliaskOr not so active devs.
17:00:53amiconnlinuxstb: Any plans to remove the ugly in() and out() macros?
17:01:11alphakillerwho founded the project?
17:01:18linuxstbCheck the CVS logs for who is active:
17:02:16linuxstbBagder, LinusN and Zagor founded the project. preglow and me are active on the ipod. amiconn is very active generally, but won't touch an ipod :)
17:02:40alphakillerhmm :P
17:02:48alphakillerI'm just interested on the iPod :D
17:02:55linuxstbamiconn: I would love to remove them, but it relies on giving names to addresses - and in most cases I have no idea what the addresses should be called.
17:03:15alphakillerwell ... I'll be back later
17:03:24alphakillerneed to do some personal stuff here ;D
17:03:31linuxstbUnless we just give names like MEM_0x80001002
17:08:03saa[b_r]idermarkun is working on the Gigabeat
17:08:58 Join Mmmm [0] (
17:11:03 Join ST [0] (
17:12:37 Join DocterD [0] (
17:13:04 Quit aliask ("G'night everyone")
17:13:28 Quit ST (Client Quit)
17:20:18 Quit Maxime ()
17:21:32alphakillerbut why support tons of players ?
17:21:52alphakillersince near 80% of ppl has iPods :D
17:22:39DrMoosmaybe because iPods aren't the best devices ;-)
17:22:54Bgeram i right that pressing a button quickly generates 2 events : BUTTON_NAME and BUTTON_NAME | BUTTON_REL ?
17:23:49lostlogicBger: yes
17:24:08lostlogicBger: see the gwps-common.c code for examples where you need to handle both separately (seek vs. skip)
17:24:33Bgeri have an idea from tree.c
17:24:48Bger10x :)
17:25:48 Join Maxime [0] (
17:26:09 Join actionshrimp [0] (
17:28:45alphakillerDrMoos: in my opinion they are :D
17:29:16 Quit San (Read error: 110 (Connection timed out))
17:29:57markunalphakiller: yes, we can remove the other players from rockbox when the ipod port is finished!
17:30:42CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:30:42*amiconn slaps markun
17:30:56markun*auch* :)
17:31:41MmmmAgghhhh Linus... what happened to the lovely 'tock' sound? it has turned into a not quite so lovely 'tick' sound! :)
17:32:02DrMoosmarkun: I slap you too :)
17:32:17markunI slap myself :)
17:33:52markunI hope the average user doesn't like rockbox
17:34:14saa[b_r]iderwhy markun?
17:34:31markunthe average ipod user
17:35:03saa[b_r]iderhow would it be bad if they did like rockbox?
17:35:09markunBecause it would get too busy in here with all the people complaining that we don't support itunes
17:35:12lostlogiccan you imagine the irritation?
17:35:25DrMoosmarkun: completly agreed here :)
17:35:33DocterDhow l33t you are :)
17:35:39markunAint I :)
17:35:50*saa[b_r]ider despises itunes
17:35:56linuxstbIgnoring itunes is a definite feature, not a bug.
17:36:43markunDocterD: Did I offend you with my stupid elitist talk?
17:36:48saa[b_r]iderit took a million hours to add music to my itunes library
17:36:54DocterDno but i dont like
17:37:05DocterDwhen people think they are better than other
17:37:10DocterDonly because a other software
17:37:18DrMoosit didn't said this
17:37:24linuxstbIf you _only_ use itunes to manage your music, and nothing else, then it works fine. Just don't try and think differently to Apple.
17:38:01markunDocterD: yes, it would be very bad if everything I said was true..
17:38:07DrMoosDoctorD: if you can, stay here and ask for all potentialy iPods question user ;)
17:38:24saa[b_r]iderlinuxstb: what do you mean "don't try and think differently to apple"?
17:38:31DocterDeh i dont care about ipod etc and what people want to buy
17:39:47DocterDbut it's its annoying those flames everywhere (kde vs gnome, windows vs linux etc)
17:39:50markunDocterD: which device do you have btw?
17:39:59DocterDiaudio m3 :S
17:39:59markun..vs FreeBSD..
17:40:31markunIs the hardware of the m3 similar to the X5?
17:40:42DocterDa bit but it has no display
17:40:43DocterDonly a remote
17:41:28DocterDbut it sucks because the remote has only 3 month warrenty and a new one costs 50 lol
17:42:10 Quit DrMoos ("CGI:IRC (EOF)")
17:45:50markunStrange to only have the LCD in the remote. They changed it for the M5.
17:47:48DocterDyeah i guess they wanted to have it not very thick
17:47:52 Join webguest58 [0] (
17:48:56webguest58Not that strange if you use it like I use my h120 - it's in my coat pocket 99% of the time. The main LCD is just eating power
17:49:47markunWe could turn off the screen after a while
17:50:31 Join akaidiot [0] (
17:50:32DocterDyeah but the player depense on the remote and that isn't really great
17:50:59amiconnThe power consumption of a monochrome LCD (w/o active backlight) is neglectible
17:51:01mikearthuris it just me
17:51:20mikearthuror does Rockbox on the H140 produce a loud "pop" when turning on, when plugged into the remote, but not in line out?
17:52:27linuxstbsaa[b_r]ider: I just mean that Apple's software (itunes and the ipod firmware) isn't flexible. It has a small feature-set - which is both its strength and weakness, depending on your point of view.
17:53:39***Saving seen data "./dancer.seen"
17:53:41saa[b_r]idervery true..
17:53:53saa[b_r]iderwhat's good for someone may not be good for someone else
17:54:15 Join PaulJ [0] (
17:54:23 Quit webguest58 ("CGI:IRC (Ping timeout)")
18:00:27markunI added CP1256 (arabic) support. According to Saab all arabic id3 tags use this (and not iso-8859-6)
18:00:45markunShall I replace the arabic codepage from the settings with CP1256?
18:01:06 Quit DocterD ("Leaving")
18:01:59saa[b_r]iderI'm almost sure that over 99% of arabs use windows mainly
18:02:19saa[b_r]iderand under windows, CP1256 is the used codepage
18:02:32saa[b_r]iderI couldn't say for sure about apple
18:02:49saa[b_r]ideralthough macs aren't that popular any more
18:03:04saa[b_r]ider(in the arab world)
18:03:30markunI'll just commit it Saab, as you are the only rockboxer with arabic mp3 that I know of :)
18:03:48saa[b_r]iderhehehe :)
18:03:57saa[b_r]iderI feel special :)
18:05:20 Part Polo_o
18:05:31amiconnIs Win-1256 just extended iso8859-6, or are some encodings different (like with cyrillic Win-1251)?
18:08:31markunquite different from iso8859-6 I think
18:09:20saa[b_r]iderwith iso8859-6 I wasn't able to view my ID3s properly
18:09:50saa[b_r]iderI'd get arabic letters, but they were random, and had nothing to do with what was supposed to be within the tags
18:11:03 Quit PaulJ ("CGI:IRC (EOF)")
18:11:20 Join PaulJ__ [0] (
18:12:20 Join Seed [0] (
18:13:47markunsaa[b_r]ider: it's in cvs now
18:13:49 Nick PaulJ__ is now known as PaulJ (
18:15:30 Quit grahamoz (" X Info~[v10B.3.1]~ Released~[November 24 2005]~ Channel~[#Exc")
18:16:05saa[b_r]iderit would be interesting if someone came and complained about the change from iso-8859-6 to cp1256 :D
18:16:39saa[b_r]iderthank you for making this a special "Eid" markun :)
18:17:16markunno problem, and yes, that would be very interesting :)
18:18:25markunMy ears are getting worse I think. There is a new firmware for the gigabeat and everyone is raving about the improved sound quality. I can't hear a difference.
18:19:06 Quit edx__ (Read error: 110 (Connection timed out))
18:19:32lostlogicmarkun: are they the same people who rave about SRS or other audio quality reductions?
18:20:28linuxstbmarkun: Any idea if the gigabeat supports MP2 files? It seems that no DAPs support that any more.
18:21:31markunIt doesn't say so in the specs, but I could try it to make sure.
18:21:49markunOnly I don't have a windows computer here to upload music to it right now..
18:21:59thegeeksrs sucks
18:22:07linuxstbI guess the reason is that software MP3 decoders generally don't include layer-2 support - unlike the hardware decoders.
18:22:30markunIsn't SRS something like crossfeed?
18:24:13saa[b_r]iderarabs looking for Unicode support on iPod to display arabic:
18:24:21preglowit's really weird
18:24:30preglowadding layer 2 support to a layer 3 decoder is not very hard at all
18:24:36 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
18:24:50markunsrs wow sounded terrible with my oggs on my H120, but maybe that got fixed in the newer firmwares.
18:25:49preglowsrs tends to amplify encoding artifacts for me
18:25:53saa[b_r]idermp2 as in layer II?
18:27:21markunsaa[b_r]ider: yes
18:30:52saa[b_r]iderlinuxstb: why are you interested in MP2 files?
18:31:18preglowbecause there is tons of mp2 around?
18:31:26linuxstbBecause I've been archiving digital radio broadcasts in their original MP2 format for the past 3-4 years.
18:31:29preglowdab radio, for example, is mp2
18:31:48linuxstbAnd I need some way to listen to them...
18:32:56linuxstbI would also like to be able to recommend DAPs to others that can natively play MP2 files. Rockbox devices are the only ones I'm aware of. IIRC, the Archos firmware would only play them if you renamed them to .mp3
18:33:14lostlogiclinuxstb: what, and you don't want to transcode them:):):)
18:33:17saa[b_r]ideraha, wasn't aware that it was the common format for radio broadcasts
18:33:47linuxstblostlogic: What do you think? :)
18:33:54saa[b_r]iderthe iRiver *might* play them...
18:34:09linuxstbAt least not the h1x0
18:36:00linuxstbsaa[b_r]ider: In Europe, almost all digital broadcasts (both TV and Radio) use MP2 audio. In the USA, the broadcast standard is AC3. I don't know about the rest of the world.
18:36:04saa[b_r]iderok, I found my MP2s, I;ll try
18:36:12preglowbroadcast ac3?
18:36:16preglowwhat bitrate?
18:36:22linuxstbVaries from 128kbps upwards.
18:36:35preglowi imagine it doesn't sound very good
18:36:41preglowac3 isn't a very complicated codec
18:37:16linuxstbIt's not too bad. No worse than 128kbps MP2.
18:37:44linuxstbBut generally, I think it's 384kbps.
18:37:54linuxstb(for HDTV broadcasts)
18:38:08preglowahh, t
18:38:17linuxstbYep, TV.
18:38:19preglow128 kbps mp2 is something different agin
18:38:29preglowit isn't a transform codec, so doesn't sound so bad when mangled
18:39:20saa[b_r]iderok, H300 doesn't display MP2 files, I'll try to rename it to MP3 and see if it plays
18:40:20preglowit doesn't play them
18:40:22preglowdon't worry
18:40:37preglowi bet they're using the motorola codec, and that only supports mp3
18:40:56saa[b_r]ideryou're right... just wanted to reconfirm :)
18:42:55saa[b_r]iderany idea if iPodlinux supports unicode?
18:43:09linuxstbNo idea.
18:43:28 Quit ender` (Read error: 110 (Connection timed out))
18:44:27preglowman, gtk1 is so incredibly ugly
18:44:41mirak_hello !!!
18:45:40markunmirak_: any mpeg2 news?
18:46:00mirak_still trying to know what is going bad
18:46:13mirak_a working simulator could help
18:46:15linuxstbpreglow: Do you think ac3 would be the best choice for a Rockbox video player?
18:46:21mirak_maybe I could build a sim for H100
18:46:33mirak_since I display nothing it should not matter
18:46:36preglowlinuxstb: well, it certainly is fast
18:46:51preglowand can probably be made faster
18:46:57markunmpeg2+ac3.. sounds like we're making a micro dvd player :)
18:47:05preglowconsidering the small amount of work i did to make it as fast as it is today
18:47:14linuxstbAnd I _think_ it has very small RAM requirements as well.
18:47:34linuxstbMaking co-existence with a video codec easier.
18:48:20*alphakiller is building cross compiler
18:48:22mirak_mpeg2 plug is ported. I don't know why the decoding goes bad. I tried it on my mac with linux so it's bigendian.
18:48:33mirak_and it works
18:48:37 Quit Rob2222_ (Read error: 104 (Connection reset by peer))
18:48:42mirak_I disabled altivec and other cpu optimisations
18:49:04mirak_I will compare the code, maybe I removed a line somewhere without caring
18:49:15 Join Rob2222 [0] (
18:49:36preglowit's big endian?
18:49:46preglowyou sure it's not using any endian flags that rockbox don't use?
18:49:51 Join eli_sherer [0] (
18:50:17mirak_preglow: well I checked that but maybe I missed something
18:50:32 Join ender` [0] (i=ychat@
18:50:56mirak_ /* Define to 1 if your processor stores words with the most significant byte
18:50:56mirak_ first (like Motorola and SPARC, unlike Intel and VAX). */
18:50:56mirak_#define WORDS_BIGENDIAN 1
18:51:22mirak_that's in the config file of the mpeg2
18:52:00linuxstbHave you compared the speed of the mpeg2 and xvid decoders on your PC?
18:52:03preglowand now for the high point of the day
18:52:08mirak_it's used only in a yuv conversion method
18:52:11*preglow goes to the pub
18:52:33mirak_linuxstb: that's faster
18:52:54mirak_linuxstb: and mpeg2 code looks less complex. It fit's also in 100ko
18:53:24linuxstbSo libmpeg2 is both faster and smaller?
18:53:31mirak_seems so
18:53:44mirak_the idct is the same
18:53:50mirak_on both codecs
18:53:51linuxstbAnd is there much difference in quality for the same bitrate?
18:54:19mirak_linuxstb: I encoded a video from a divx in 220*176 at 300kb and it looks fine on pc
18:54:24 Quit DangerousDan (Read error: 110 (Connection timed out))
18:54:45mirak_I think that the divx bitrate used on the H300 are way to high
18:54:51mirak_I mean in xvid
18:55:01linuxstbIt all depends on the quality of the encoder I think.
18:55:06mirak_with 300kbit for xvid image is perfect, there is no artifacts
18:55:23linuxstbBut it sounds promising.
18:56:10mirak_if anyone is interested in finding why mpeg2 fails on H100 or H300 it would be nice
18:56:24linuxstbDo you have a patch somewhere?
18:56:29mirak_it's not easy to trace from the code at wich moment it breaks
18:56:40mirak_linuxstb: not yet, since it doesn't work )
18:56:44 Quit darkless (Client Quit)
18:56:54mirak_but I can put it on the tracker
18:57:10mirak_I the code is not clean I have put some string output and sleeps
18:57:19 Join iSTHM [0] (
18:57:31mirak_I will do a kompare first to see If I have not done a dumb move
18:57:35linuxstbIf you're still struggling in a couple of days, let me know.
18:58:00mirak_it's been 4 days already :-/
18:58:36iSTHMCan anyone help? The firmware upgrade process failed because of low battery. I can't turn on my player now. What should i do? I have iriver t30 mtp
18:59:24saa[b_r]iderT30? this doesn't have anything to do with RB does it
18:59:27linuxstbiSTHM: Cry?
18:59:50mirak_In file included from /usr/include/sys/stat.h:105,
18:59:50mirak_ from lcd-x11.c:26:
18:59:50mirak_/usr/include/bits/stat.h:70: error: field `st_atim' has incomplete type
19:00:01mirak_hem still broken for X11
19:00:26linuxstbYou could try the new SDL uisimulator - it's in CVS now.
19:00:26iSTHMsaa[b_r]ider: its iRiver player
19:00:41mirak_LinusN: ok, what's the module name ?
19:01:17saa[b_r]ideriSTHM.. I honestly don't know what to say. have you tried pluging your AC adapter?
19:01:40saa[b_r]iderand I know it's an iRiver, only your problem is not RB related
19:01:42linuxstbmirak_: It's just an extra directory in the uisimulator module. You'll need to update tools/ and uisimulator/
19:02:04linuxstbThen just run configure and select SDL under simulator types.
19:03:10iSTHMmaybe you know a forum or a irc channel where could get help?
19:03:30linuxstbYou could try or #misticriver
19:04:46iSTHMok, thanks
19:04:51 Quit iSTHM ("Ever stop to think, and forget to start again?")
19:05:11saa[b_r]iderwas I rude?
19:06:43linuxstbTime to go. Back later.
19:06:47 Quit linuxstb ("Client Exiting")
19:06:54mirak_linuxstb: I still have the same error on lcd-x11.c with SDL also: I still have the same error on lcd-x11.c with SDL also
19:07:13mirak_mmm don't know why it repreated
19:09:08 Join einhirn [0] (
19:16:31 Join PaulJ_ [0] (
19:19:47saa[b_r]idereli_sherer: the suspense is killing me :)
19:21:28 Join ender1 [0] (i=ychat@
19:21:52 Join San [0] (
19:21:59 Join [San] [0] (
19:22:03 Quit [San] (Remote closed the connection)
19:23:19eli_shereri'm finishing it up... :)
19:23:47eli_shereri have a last moment bug! :(
19:27:42alphakillerwhat's is this thing mirak_?
19:28:40eli_shereri think i found it...
19:28:46 Join einhirn_ [0] ( but it's part of the problem
19:33:02 Quit ender` (Success)
19:33:21mirak_alphakiller: ?
19:33:43 Join EDG4 [0] (
19:33:43 Quit PaulJ (Read error: 110 (Connection timed out))
19:35:56 Quit einhirn (Read error: 113 (No route to host))
19:39:08 Quit akaidiot (Read error: 110 (Connection timed out))
19:41:31alphakillerarm-elf binutils built
19:41:35alphakillerjust missing gcc :P
19:42:12 Quit einhirn_ ("Miranda IM! Smaller, Faster, Easier.")
19:43:40 Join darkless [0] (
19:46:02 Join mymomthelush [0] (
19:47:47BgerBagder ? some people are complaining about missing things like "scramble" in the daily builds' source
19:49:50 Join ste__ [0] (
19:51:08 Join akaidiot [0] (
19:53:40***Saving seen data "./dancer.seen"
19:57:58 Quit mymomthelush ("Trillian (")
19:58:30 Quit San (Read error: 110 (Connection timed out))
20:00:59NicoFRhas anyone noticed snkae2 crashing at startup lately ?
20:01:07NicoFRit does on the h300 bleeding edge build
20:02:30saa[b_r]iderdoes it? you got today's bleeding edge?
20:02:52NicoFRwell it was before the last commit
20:03:44NicoFRi first noticed it on my custom build so I tryed the bleeding edge and it did the same
20:03:55 Quit Mmmm ()
20:04:02NicoFRsomeone on misticriver has also mentioned the problem
20:04:35alphakillerRockbox is only intended for iPod and Archor ?
20:05:01NicoFRno it also works on iriver iHp-1xx and H3xx
20:05:14saa[b_r]iderI barely play snake2 on my H300, since it's not optimized for its screen
20:05:22NicoFRand is being worked on for other platforms
20:06:08saa[b_r]iderrockbox can be ported to any platform, as long as someone is willing to port it
20:06:18alphakillerhmm ..
20:06:24alphakillerthe current "main target" is iPod ?
20:06:36alphakillerwhen I was here hehehe, everybody was only talking about iPod
20:07:03saa[b_r]ideractually, only two guys are working on the iPod....
20:07:08saa[b_r]ider3 max
20:07:20NicoFRsaa[b_r]ider: i don't play it either, i removed it from my build for the time being
20:07:29NicoFRjust wanted to point out the probelm
20:07:53saa[b_r]iderhow long can you play it before it crashes?
20:09:17 Join edx [0] (
20:09:24 Quit ste__ (Read error: 110 (Connection timed out))
20:09:45alphakillertoolchain built :P
20:10:29 Join San [0] (n=test@
20:14:11NicoFRit crashes at startup
20:14:18NicoFRI don't get to play
20:14:23NicoFRI have to reset the player
20:16:58 Join ender` [0] (i=ychat@
20:18:21eli_shererI found that pasky bug...apperantly i changed something and i forgot
20:18:39eli_shererconcerning my project sorry for interfering
20:22:37 Join _FireFly_ [0] (
20:25:23Saneli, when are going to have it released?
20:25:27Sanlooks good
20:29:36saa[b_r]idereli, san, has been posted??
20:30:27eli_shereron the site...just wait a few moments and it's there (i'll announce it don't worry)
20:30:39Santhank you.
20:30:48 Join linuxstb [0] (
20:30:54Sanyou are good at this. Did you learn it in school or in classes?
20:31:04Sanyes Saab
20:31:11saa[b_r]iderawesome :)
20:31:14 Join kkurbjun [0] (
20:33:03eli_shererit's out...goto:
20:33:13linuxstbalphakiller: There is no "main" target for Rockbox. The original targets were the Archos Jukebox range - with a hardware MP3 decoder and Hitachi SH1 cpu. About 12 months ago, work started on a port to the iriver H1x0 series - which does all audio decoding in software using its Coldfire CPU. This port was recently extended to support the iriver H3x0 devices (very similar hardware to the H1x0, but a colour LCD). About 3 months ago, I started work o
20:34:07linuxstbI think the build system currently gives you a choice of 15 targets...
20:34:19Sangot it
20:34:30alphakillerlinuxstb: I built the cross compiler
20:34:35alphakillerunder cygwin
20:34:36Sanwhat games does it play?
20:34:38alphakillerpretty easy
20:34:43eli_shereri'm now building an h1xx sim to check it out on the h1xx
20:34:49linuxstbYep. The bootloader is next then...
20:35:10Saneli, bug
20:35:14Sanbig bug
20:35:25Sanwhen you hit into a wall
20:35:30Sanit shuts the player down
20:35:48eli_shererwhat player do you use?
20:35:51 Quit linuxstb ("CGI:IRC")
20:35:59eli_shererthat's odd..
20:36:17eli_shererit can be cause by loss of memory but i calculated everything to work
20:36:47Sanyep, happend again
20:36:50 Join linuxstb [0] (
20:38:02alphakillerlinuxstb: what's the next step ?
20:38:45mirak_linuxstb: anyidea why lcd-X11.c fails for the uisimulator ?
20:39:15linuxstbalphakiller: Follow the instructions at - but they are slightly different for the Video....
20:39:19EDG4Do you know #misticriver server?
20:39:20EDG4[9:39:14 PM] hp says: ir wsio
20:39:20eli_sherersan: have you changed anything concerning the plugin buffer allocation?
20:40:57eli_shererit happened to me now...i can see that...
20:42:25linuxstbmirak_: Try removing #include <sys/stat.h> from uisimulator/x11/lcd-x11.c - that seems to cause your error, and I don't know why it's there.
20:42:58Saneli, I didn't change anything
20:43:17linuxstbalphakiller: Those ipodboot instructions won't work on cygwin because cygwin doesn't give you access to the ipod's boot partition (because it is marked as type "0" - i.e. empty).
20:43:49linuxstbI've written a utility which may work, but it hasn't been tested on cygwin yet:
20:43:59alphakillerI downloaded that
20:44:08linuxstbHave you compiled it?
20:44:12alphakillernot yet
20:44:22alphakillerchecking out cvs
20:46:07linuxstbYou can test diskdump by typing "diskdump /dev/sda" - that should show you the partition map for that disk. You can try different devices (/dev/sdb, /dev/sdc etc) until you find your ipod.
20:46:39alphakillerhmm I see
20:46:42linuxstbIt should be safe because in that mode it only opens the device read-only.
20:47:05eli_shererSan: youre right it happend to me too...i'm working on an optiomization function right now...
20:47:19linuxstbYou obviously need your ipod plugged into your PC, and "disk mode" activated. You may also need to close itunes if you have it installed.
20:48:34 Quit ender1 (Connection timed out)
20:48:40alphakillerhmm :)
20:49:16 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
20:49:17linuxstbThis part of the process is much easier under Linux.
20:49:43alphakillerok, I have disk mode activated too :D
20:50:22eli_shererSan: i need to have some sort of handle the flood fill
20:50:43Xerioneli_sherer, on my H140 it seems to work fine :)
20:50:52eli_sherernow i'm using recursive loop
20:52:48linuxstbalphakiller: Any luck with diskdump?
20:53:08eli_shererXerion: that's because you have a smaller screen and the flood fill don't need much memory
20:53:48alphakillerI'll compile it now
20:55:23alphakillerlinuxstb: worked :D
20:55:24kkurbjunamiconn: I saw your name in the credits for rockboy: I've created a patch that enables color on color targets, speeds it up, and fixes the sound. If you get a chance could you look at it and tell me if the patch is ok or if there are changes that need to be made?
20:55:33alphakillergot it on /dev/sdb
20:55:40linuxstbCool :)
20:55:51alphakillertype 0x0 ?
20:55:54alphakillerfirst partition
20:55:54kkurbjunor if anyone else can look at it that would be fine
20:56:03alphakiller78mb long
20:56:10alphakillerand another partition with 28gb
20:56:14alphakillerthat should be my ipod
20:56:17alphakillerI have only one hd here
20:56:20alphakillerno other disk
20:56:24linuxstbYes - sounds perfect.
20:56:43alphakillerso I'll send to /dev/sda?
20:56:51alphakillerthe bootloader
20:57:00linuxstbNow you want to backup the original contents of the boot partition to a file - "diskdump -r /dev/sdb bootpartition.bin"
20:57:26linuxstbThis isn't just a backup - you'll need to extract some data from that file.
20:57:41alphakillerhmm I see
20:58:42 Join Moos [0] (
20:59:34alphakillerok done :D
20:59:55linuxstbI'm happy that worked. Did diskdump display any warnings about being unable to seek?
21:00:09alphakillerno ... didn't saw the error
21:00:32 Quit EDG4 (Read error: 104 (Connection reset by peer))
21:00:33alphakillerSeek error: reading 63 sectors instead
21:00:40linuxstbThat's fine - it's just a warning.
21:00:56linuxstbIs bootpartition.bin 78mb long?
21:01:53alphakillerjust finished downloading cvs
21:01:57alphakillernow build toolchain ?
21:02:03linuxstbOK, now you need to extract two files from bootpartition.bin. In the Rockbox cvs, "cd" to the "tools" directory, and type "make ipod_fw"
21:02:37alphakillernext step?
21:02:50linuxstbThen go back to the directory containing bootpartion.bin and type:
21:03:25linuxstb...../tools/ipod_fw -o apple_os.bin -e 0 bootpartition.bin
21:03:38linuxstb...../tools/ipod_fw -o apple_sw_5g_rcsc.bin -e 1 bootpartition.bin
21:04:00linuxstbNow you can build the Rockbox bootloader...
21:04:45alphakillerdid that .. how to build the bootloader ?
21:05:17linuxstbCreate a build directory (e.g. "build-boot") in the same directory that contains "tools", "apps", "firmware" etc in the Rockbox source.
21:05:31linuxstbThen cd into that directory and type ../tools/configure
21:05:33 Join EDG4 [0] (
21:06:06linuxstbEnter "15" for ipod video, then "B" for bootloader, then type "make"
21:06:30alphakillerhehe nice :P
21:06:50darklesslinuxstb: you should throw all that into a wiki page for future requests :)
21:07:19linuxstbI'll just point people to the IRC logs :) But yes, I need to update the IpodBoot page for the video.
21:07:25alphakillerstrange ..
21:07:44alphakillerNo such file or directory ../bootloader
21:08:07alphakillerjust after:
21:08:09linuxstbDid you checkout "rockbox" or "rockbox-devel" from cvs?
21:08:38linuxstbYou are missing the bootloader directory then. You should just be able to type "cvs co bootloader" from inside the rockbox directory.
21:08:41alphakillerdid I picked up the wrong one ?
21:08:45linuxstbYep :)
21:09:11alphakilleroh ok
21:09:16alphakillerwith cvs co bootloader
21:09:35alphakillerI have a bootloader.bin
21:09:38alphakilleron /bootloader
21:10:24linuxstbAlmost there then. Now "cd" back to the directory containing "bootpartition.bin" and the other files.
21:10:31 Join IRCSteve [0] (
21:11:30linuxstbAnd type "...../tools/ipod_fw -g video -o rockboot.bin -i apple_os.bin ...../build-boot/bootloader/bootloader.bin"
21:11:56*EDG4 firmware ugrade process failed because of low battery, i can't turn on my player (iriver t30).Can somebody help me please?
21:12:05linuxstbThis will create a rockboot.bin which you can then write to your ipod using "diskdump -w /dev/sdb rockboot.bin"
21:12:11eli_sherernow i really can't think of ay better algorithm for flood fill that will fit the h300 memory abilities
21:13:02alphakillerhmmm that same Seek Error
21:13:09alphakillerwill this blow up my iPod ?
21:13:16*linuxstb hides...
21:13:38linuxstbNo developer (either Rockbox or ipodlinux) has destroyed an ipod yet.
21:13:38alphakillerWritting input file to device
21:14:07alphakilleris that hard to code my own bootloader ?
21:14:11alphakillerlike a hello world bootloader ?
21:14:35eli_shererCan someone who knows a little computer science help me with this one...?
21:14:42linuxstbNo. The place to look is bootloader/ipod.c - this uses code (lcd driver, ata driver etc) from the firmware/ directory.
21:15:07alphakilleryeah, but I guess there's a ready framebuffer, or no ?
21:15:11 Nick Paprica[sleep] is now known as Paprica (
21:15:14alphakillerwrote ..
21:15:22alphakillerunplug ?
21:15:24alphakillerand test ?
21:15:25linuxstbSort of. Look in firmware/drivers/lcd-ipodvideo.c
21:15:36 Quit San (Read error: 110 (Connection timed out))
21:15:37linuxstbplus the code in lcd-16bit.c
21:15:39alphakillerhmm .. that's what I was interested :D
21:15:51alphakillerI'll play with that later
21:16:02alphakillerthe important thing now is how I can make both cores work on your kernel
21:16:20alphakillerI might use that "nanokernel" idea, to put it on the other core, and let it in piece :D
21:16:34alphakillerlinuxstb: can i deattach ipod?
21:16:35alphakillerand test ?
21:16:39alphakillerit's written
21:16:47linuxstbYou should also get a copy of all the ipodlinux source you can find - that will help with the low-level stuff.
21:17:28linuxstbalphakiller: Yes - it should display some logos (but they are corrupt - an easy bug that I haven't fixed yet) and then load the Apple firmware.
21:17:54alphakillerhmm ... you killed my iPod :@
21:18:02alphakillerit turns on, then off
21:18:58linuxstbOK... Try plugging it back into your PC and seeing what "diskdump /dev/sdb" tells you.
21:19:38alphakillerit can read my ipod
21:19:43linuxstbAlso, can you still see the normal FAT32 partition when you plug it in?
21:20:39alphakillerit locked up my explorer
21:20:43alphakilleron windows ..
21:20:50linuxstbAh windows...
21:21:03alphakillerbut the partition table isn't changed
21:21:16linuxstbI've heard people talk about "ipodservice.exe", and having to restart it sometimes.
21:21:39linuxstbYou could try restoring "bootpartition.bin" back
21:21:47alphakillerthat's what I'm trying to do
21:22:29 Join Lear [0] (
21:22:40alphakillerseems it didn't read perfectly
21:22:46alphakillerand it might messed up stuff
21:23:38alphakillernow unlocked
21:23:43alphakillerI can see the FAT 32
21:23:55alphakillerI probably know the cause ..
21:24:02linuxstbYou're the first person to try this on Windows....
21:24:20alphakillerwell, seems it was locking my windows, coz I just restored the FAT32 FS
21:24:29alphakillersomething overwrote that on our image
21:24:40linuxstbIf you do end up with an unusable ipod, you can download the "ipod restorer" from the Apple website, and restore your ipod using that.
21:24:41alphakillermy guess
21:25:34alphakillerI wish I could write something on my ipod :(
21:25:38alphakillera demo ..
21:25:41 Nick IRCSteve is now known as SteveL_ (
21:25:57linuxstbGetting the bootloader running is the hardest part - you can then code away...
21:26:09 Nick SteveL_ is now known as SteveL__ (
21:26:25linuxstbSo are you able to restore bootpartition.bin ?
21:26:26alphakillerthe shitty part is that my usb is damn slow
21:26:31alphakillerI'm trying
21:26:34alphakillermy usb is slow ..
21:26:41alphakillerusing the usb on the front of my pc
21:26:55alphakillerif these ones are slower than the backwards ones
21:26:57alphakillerI think no
21:27:33alphakillerI think it's a shared USB controller
21:27:35alphakillerand it's 1.1
21:28:53alphakillermy computer isn't that old
21:29:02alphakillerit's a dell dated in 2003
21:29:24alphakillerstrange a usb 1.1 port
21:30:10linuxstbI have a similar Dell at work
21:30:21linuxstbMust have been the last models with usb 1.1
21:30:41ender`in 2003 1.1 was still common - and you usually got 2 2.0 ports at the back, plus 1-2 1.1 ports in front, and sometimes additional 2 1.1 ports at the back
21:32:20 Join jlo [0] (
21:32:31jlohi all
21:35:12 Quit edx (Read error: 110 (Connection timed out))
21:37:30jlohi preglow : pls check crossfeed schematics on, let me know if it's clear
21:39:40alphakillercan't find apple restore
21:42:11alphakillerjeez ..
21:42:12alphakillerI'm bored :(
21:42:28linuxstbDid restoring bootpartition.bin fail?
21:42:54alphakillerI can't access my ipod
21:42:55alphakilleras devie
21:44:10alphakillerjeez it can't detect my ipod
21:44:12alphakillerI'm scared
21:44:38 Quit EDG4 ()
21:44:40 Join EDG4 [0] (
21:44:42 Quit EDG4 (Remote closed the connection)
21:45:14 Join linuxstb_ [0] (
21:45:30linuxstb_Don't worry - it's always retrievable.
21:45:53linuxstb_You probably need to force the ipod into disk mode - see the comments at the end of the IpodBoot wiki page.
21:46:16 Quit Rob2222 (Read error: 104 (Connection reset by peer))
21:46:57 Join Rob2222 [0] (
21:49:19alphakillerno effect :(
21:49:40 Quit linuxstb_ (Client Quit)
21:49:56alphakillerit detects when I plug in
21:50:01alphakillerbut it isn't going to disk mode
21:50:53 Join linuxstb_ [0] (
21:51:02 Quit Nibbler (Success)
21:51:28 Join Nibbler [0] (
21:51:33linuxstb_So if you enter disk mode, then plug your ipod into your PC, WIndows doesn't recognise it?
21:52:04linuxstb_You mean the ipod isn't entering disk mode?
21:53:41***Saving seen data "./dancer.seen"
21:54:22alphakillerseems that I'm on a serious trouble :'(
21:54:35Moosjlo: died link
21:54:55mirak_is there a way to make gcc tell all the functions who are called ?
21:54:59mirak_to have a trace
21:55:19alphakillerenabling debug option and using gdb
21:55:28linuxstb_Sorry, I'm not understanding - if the ipod isn't entering disk mode, then you are not pressing the key combinations correctly. It never fails...
21:56:55linuxstb_You need to be very fast to press SELECT+PLAY after the ipod reboots.
21:57:00mirak_alphakiller: don't know how to use gdb ...
21:57:33alphakillerit's on disk node
21:57:36 Quit HCl ("Lost terminal")
21:57:42alphakillerbut windows doesn't recognizes the disk
21:58:15linuxstb_Do does diskdump display anything?
21:59:08linuxstb_Also, do you have itunes installed on this PC?
21:59:25linuxstb_Have you tried restarting ipodservice.exe ?
21:59:48 Quit linuxstb (Read error: 110 (Connection timed out))
22:00:13 Nick linuxstb_ is now known as linuxstb (
22:00:16 Join solexx [0] (
22:04:35alphakillerit simply can't detect my ipod
22:04:49 Quit alphakiller (Read error: 104 (Connection reset by peer))
22:06:14mirak_damn I was wasting time putting printf in the code while there is gdb ...
22:06:20mirak_I though is was complicated
22:12:20 Quit solexx_ (Read error: 110 (Connection timed out))
22:17:13 Join hshah [0] (
22:18:24*eli_sherer presents H300 OPTIMIZED ROCKBOY (COLOR,REMAPPED KEYPAD, FRAMESKIP) copyrighted to an anonymous user who was kind enough to put some effort into it!
22:18:49 Quit linuxstb ("CGI:IRC (Ping timeout)")
22:19:07*eli_sherer forgot SOUND SUPPORT
22:19:20Xerionkkurbjun was the one who made it ;)
22:21:22kkurbjunI have a thread on mistic river with the compiled version
22:21:53lostlogicoptimized sounds like the wrong word for it ... I mean ... it's not faster... it just has some extra features...
22:21:53 Join drumrboy [0] (
22:22:08kkurbjunI did speed up the lcd updates
22:22:13kkurbjunwithin rockboy
22:22:21kkurbjunbefore it was updating a line at a time
22:22:29*eli_sherer bows to kkurbjun
22:22:34kkurbjunnow it does the whole screen all at once
22:22:36 Quit _FireFly_ ("Leaving")
22:23:16eli_shereri'll upload your version because mind seems to be 300kb...
22:23:16Papricakkurbjun, no chance to set it for all the screen?
22:23:18kkurbjunand when the sound is diabled it is alot faster then before.. it skips alot more code now
22:23:36kkurbjunahh scale it?
22:23:42kkurbjunsounds good
22:23:56 Join _FireFly_ [0] (
22:24:04 Join muesli__ [0] (i=muesli_t@
22:24:15Papricacan you do it?
22:24:23 Join drumrboy320| [0] (
22:24:32kkurbjunpaprica: it's possible, gnuboy has native support for it.. maybe at somepoint, but it's not really a priority for me right now
22:24:53kkurbjunit would slow it down also
22:25:47*eli_sherer changed the link to
22:26:03 Part hshah ("Leaving")
22:27:12 Quit Nibbler ("can't stop the signal!")
22:27:18 Join Nibbler [0] (
22:28:03Learlostlogic: 6i is about the same in bass as the E2c. :)
22:28:04 Join linuxstb [0] (
22:31:52eli_shereri think if the sound or the color isn't enough a reason to add it to the cvs the frame skipping is!
22:32:09linuxstbkkurbjun: I haven't looked at your patch, but have you made more use of IRAM?
22:32:46linuxstbI don't think rockboy currently uses very much IRAM - big speed improvements could possibly be gained in that direction.
22:33:28kkurbjunlinuxstb: I don't know how to use IRAM .. its that some code cache? and do you have an example in rockbox that you can point to that uses it?
22:34:16Learkkurbjun: see codecs
22:34:32kkurbjunok, I'll look at that
22:34:37Learassuming it works for plugins, that is... :)
22:35:50linuxstbYes, it works the same in codecs and plugins
22:36:18linuxstbIRAM is the fast 96KB of internal memory on the coldfire. Putting data in iram means it can be accessed a lot faster than from the slow SDRAM.
22:36:28Learbut what if you try to run a plugin that needs iram when playing music?
22:36:38linuxstb48KB of IRAM is used by core Rockbox, 48KB can be used by a codec or plugin.
22:36:39kkurbjunahh, so data cache
22:36:53Bagdernot cache, just fast ram
22:37:01linuxstbrockboy stops playback anyway. I'm not sure how the details work, but I'm sure they do.
22:37:01Bagderor rather less slow ;-)
22:37:25Bagderrockbox stops playback because it nicks the buffer
22:37:33Bagderrockboy even
22:37:54Learit ought to nick the buffer before trying to use iram then...
22:38:23BagderI believe it does
22:38:49 Join alphakiller [0] (n=bohas@
22:38:59linuxstbkkurbjun: Search for ICONST_ATTR (for constant data) and IBSS_ATTR (for variables) and ICODE_ATTR (for code) in the source.
22:39:04alphakillerlinuxstb: just restarting the machine
22:39:07 Quit akaidiot (Connection timed out)
22:39:12alphakillerit entered on the stuff again
22:39:15BagderLear: because it needs the buffer to load the ROMs in
22:39:17alphakillerI'm trying to upload rockbin again
22:39:17linuxstbalphakiller: I guessed that :)
22:39:44linuxstbLook at your ipod's display - can you see a spinning disk symbol in the top-left corner?
22:40:07linuxstbMake sure that has disappeared before you unplug your ipod.
22:40:18alphakillerok :)
22:40:27 Quit drumrboy (Read error: 110 (Connection timed out))
22:40:49alphakillerit takes a while this shit :P
22:41:11kkurbjunlinuxstb: thanks I'll look into that
22:41:21alphakiller(trying to rewrite rockboot.bin)
22:42:30alphakillerjeez ..
22:42:36alphakillerit takes half a century :P
22:43:50alphakilleroh .. your diskdump is working
22:44:07alphakillerwhen I put send it ... it starts to spin symbol on ipod
22:44:51 Join Mark_ [0] (
22:45:54 Quit darkless (Client Quit)
22:49:25markunlinuxstb: what's the difference between mov pc, 100 and b 100 ?
22:49:38markunb 100 is relative?
22:49:54Bagderb sets LR too, doesn't it?
22:50:46linuxstbI don't know, I haven't done any ARM assembler yet...
22:54:17alphakillermarkun: my guess is that both are absolute
22:54:37alphakillercan you do add pc, 100 ?
22:54:41alphakillerthat will be relative
22:55:00markunyes, sure
22:55:07 Part jlo
22:55:30alphakilleroh ... you're not interested on doing a relative ... you just wanna know the operation, nah ?
22:56:32markunI'm looking at some code and want to know why sometimes b is used and sometimes a value is writen to PC
22:56:41 Join ehntoo [0] (
22:57:03LinusNrockboy on h300 with sound ROCKS!!!
22:57:09linuxstbIs it a constant value being written to the PC, or the contents of a register?
22:57:25BagderLinusN: yay!
22:59:10markunlinuxstb: both I think
22:59:46linuxstbI think Bagder sounds right - b sets LR
23:01:05 Join lodesi [0] (
23:02:47 Join Mmmm [0] (
23:03:53MmmmLinus: Have you noticed that the metronome tock sound isn't a meaty as it was?
23:04:26MmmmSounds like maybe the whole array isnt being played ??
23:05:07 Join Kohlrabi [0] (
23:06:10alphakillerlinuxstb: no idea of what can I do :(
23:06:15alphakillerto put it on ...
23:06:57linuxstbCan you write the original bootpartition.bin back? Just to confirm (or not) that diskdump is doing its job correctly?
23:07:32linuxstbIf that works, then I can either try your rockboot.bin on my ipod, or give you a copy of my working rockboot.bin
23:08:08linuxstbI'm assuming your second attempt failed as well?
23:08:16 Quit mirak_ (Connection timed out)
23:08:18LinusNMmmm: didn't think of that
23:08:56MmmmYou noticed that then?
23:09:43drumrboy320|im fearing the fact that this will prob sound stupid... is there any 5g ipod support?
23:09:53alphakillerI'm restoring with apple restorer ..
23:09:59linuxstbalphakiller: Do you have access to a linux machine?
23:10:08alphakillernot here ..
23:10:16alphakillernot a physical access
23:10:17alphakillerI mean
23:10:23alphakillerto attach my iPod
23:10:30alphakillerI can boot a live cd :P
23:10:35drumrboy320|theres always a live distro, iff needed
23:10:41drumrboy320|lol, nvm
23:10:45linuxstbdrumrboy320l: Yes, the ipod port supports the 5g.
23:10:57linuxstbBut the ipod port isn't very usable yet.
23:11:27drumrboy320|ahh, i see, well then, i will stay very posted for new developments, and i have an ipod now
23:12:33alphakillerat least I can restore my iPod ..
23:12:45MmmmLinus: is it something to do with the values passed to timer_register? I don't really understand timer_register :(
23:12:59alphakillerman I have no idea what I could do
23:13:01LinusNMmmm: i think i know what it is, hang on
23:13:05alphakillerI deleted my rocketboot.bin
23:13:17alphakillerI think I might have done something wrong with that ..
23:13:29linuxstbThis channel is logged -
23:14:01 Quit eli_sherer ()
23:14:30linuxstbIf you try this again under Linux, you can use dd if=/dev/sdb1 of=bootpartition.bin and dd if=rockboot.bin of=/dev/sdb1 to read/write the firmware partition.
23:15:07 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
23:15:39 Quit Lear ("Chatzilla 0.9.69 [Firefox 1.5/0000000000]")
23:16:54 Join edx [0] (
23:18:32alphakillergonna try just to see if it works
23:19:19BHSPitMonkeylinuxstb: what's the rockbox bootloader like
23:19:33BHSPitMonkeyi just might use it instead of ipl's for now since it triple-boots
23:20:07 Quit t0mas (" brb.. switching accesspoints")
23:20:10linuxstbThere is no menu like ipodloader2 - the default is to boot Rockbox, hold MENU to boot the Apple firmware, or PLAY to boot Linux.
23:20:34linuxstbJust try it - if you don't like it, or it doesn't work for you, you can easily restore the ipodlinux loader.
23:21:04BHSPitMonkeycan you change the default?
23:21:19BHSPitMonkeyi guess i'll look at the source instead of asking like a dweeb :P
23:21:42linuxstbInstructions are here:
23:21:53linuxstbAnd no, you can't change the default unless you edit the source.
23:22:23linuxstbSorry, I meant here:
23:22:28LinusNMmmm: fixed
23:22:38LinusNMmmm: i forgot that the tock sample was mono
23:23:51MmmmLinus:...ooh great...Looking forward to seeing what fixed it :D
23:24:47alphakillerlinuxstb: to do thouse operations you told me, it should be on the diskmode, nah ?
23:25:26linuxstbThe diskdump commands need to be. Nothing else does.
23:25:38alphakillerI think that might caused the failure
23:26:13alphakillerhow do I read the apple firmware ?
23:26:16linuxstbBut if the ipod wasn't in diskmode, I would have expected diskdump to have failed.
23:26:19alphakillerI can't remember
23:26:41linuxstbCheck the IRC logs here: (times are GMT - it's currently 22:26 GMT)
23:26:45alphakillerlet's try again :P
23:27:05 Join darkless [0] (
23:27:32 Join RotAtoR [0] (
23:28:25LinusNMmmm: the timer_register() call allocates a high-resolution timer interrupt
23:29:12LinusNwhich calls timer_callback() 1024 times per second
23:29:59alphakillerlinuxstb: I did something wrong
23:30:13alphakillerI'll try again :p
23:30:34linuxstbI hope so :)
23:32:36 Part SteveL__
23:32:38MmmmLinus: Ahh, right... I'll think about that one :) Looked at your fix, I wouldnt have though of that one :D good job i asked! :D Ta linus, Goodnight...
23:33:18alphakillerlinuxstb: wasn't I supposed to do something with apple_sw_5g_rcsc.bin ?
23:33:45linuxstbNo - ipod_fw loads that file (from the current directory) when you create rockboot.bin
23:34:06 Quit Mmmm ("CGI:IRC")
23:34:09BHSPitMonkeylinuxstb: I'd just use my regular make_fw in leu of downloading ipod_fw, correct?
23:34:17linuxstbipod_fw is a utility from the ipodlinux people.
23:34:39linuxstbBHSPitMonkey: Yes - ipod_fw and make_fw are identical. We've just put a copy in the Rockbox CVS for convenience.
23:34:47BHSPitMonkeymake_fw is what I know it by... ok cool
23:35:11linuxstbAnd obviously "make_fw" is not a good name for use in Rockbox.
23:35:33linuxstb(considering all the different targets)
23:36:39alphakillerhmm strange
23:36:57alphakillerit said that it was generating firmware for iPod photo
23:37:06alphakillerand iPod mini
23:37:11alphakillernot for iPod video
23:37:25 Quit _FireFly_ ("Leaving")
23:37:28alphakillerand 4g
23:37:32linuxstbThat's fine - it's a bug in ipod_fw
23:37:43alphakillerlet's try to update :P
23:38:09linuxstbHow big is rockboot.bin - it should be approximately the combined size of the two apple files and bootloader.bin
23:39:32alphakillerit's only 10 mb
23:39:37alphakiller10 mb and half
23:39:48alphakillertwo apple files and bootloader.bin are 15 mb
23:39:54BHSPitMonkeylinuxstb: where does the bootloader look for a linux kernel?
23:40:14linuxstbBHSPitMonkey: A file called linux.bin in the root of your FAT32 partition
23:40:32alphakillerlinuxstb: failed :(
23:40:37alphakillerI think
23:40:42alphakillerthe firmware generation
23:40:49alphakilleras I said it's not 15 mb
23:40:52alphakillerit's 10 mb
23:40:55alphakillerand each apple file is 5mb here
23:40:59alphakillerplus 5mb from bootloader
23:41:11linuxstbalphakiller: My rockboot.bin is 11204608 bytes
23:41:39linuxstbThe bootloader is tiny - about 66KB
23:41:55alphakilleryours has the same size as mine
23:42:09alphakillerops yeah I told you bullshit
23:42:19alphakillerwell, mine is exactly the samesize as yours now
23:42:49linuxstbTry writing it...
23:42:58alphakilleryeah, I'm already writting it
23:43:00BHSPitMonkeyalphakiller: really, don't kill the IRC buffers so much
23:45:11alphakillerlinuxstb: failed :(
23:46:08linuxstbMmm. Seems that diskdump doesn't work then.
23:46:19linuxstbTime for that Linux Live CD
23:46:23 Join Janco [0] (
23:50:23 Quit Bagder ("Off to search for that connect-resetting peer guy!")
23:50:27 Join Bagder [0] (
23:50:31alphakillerI think I'll give up :(
23:51:16 Quit BHSPitMonkey (Remote closed the connection)
23:51:19linuxstbAll I can say is that it works for me under Linux and Mac OS X - no-one has succeeded under Windows yet.
23:52:09alphakillerpartcopy :P
23:52:11linuxstbAll you need to do is to restore your ipod, then boot linux, then use "dd" to write your rockboot.bin to /dev/sdb1 (or whichever device it appears under Linux)
23:53:20linuxstbIf partcopy works, then let us know.
23:53:42linuxstbBut the problem is the ipod's boot partition is marked as "empty" - so lots of utilities ignore it.
23:53:45***Saving seen data "./dancer.seen"
23:53:52alphakilleror dd under cygwin
23:53:59alphakilleranyone has tried ?
23:54:14linuxstbYes - cygwin doesn't give you a device for empty partitions.
23:54:33linuxstbSo /dev/sda will only have one partition visible to cygwin - the fat32 partition.
23:55:03*Xerion dances to the megaman sound from rockboy ;)
23:55:16linuxstbYou could try changing the type of the partition from 0 to something else, then use dd, and then change it back to 0.
23:55:54alphakilleryeah :P
23:56:00linuxstbWe'll be developing a GUI installer to do this under Windows - but no work has been done on it yet.
23:57:56alphakilleror rawrite :D
23:58:22 Join akaidiot [0] (

Previous day | Next day