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 2005-11-16

00:00:03davidc__the 0x00060000 all ahve the old lcd
00:00:08davidc__their hardware is very similar to the photo
00:00:10 Join webguest86 [0] (
00:00:23davidc__the 0x00060004's ahve a combo of old and new
00:00:30davidc__and detecting that is really strange
00:00:38davidc__preglow: we think there is some mailboxing code.
00:00:51linuxstbI saw leachbj make about 4 CVS commits before that worked.
00:01:05davidc__yeah.. its a bunch of bizzare gpio stuff
00:01:20linuxstbAnd there's no way for a user to know either, afaik.
00:02:00davidc__not that we know of
00:02:07linuxstbSo our current bootloader should work on your ipod.
00:02:20davidc__I'll check it out when I get home
00:02:51 Join davidd [0] (
00:03:09davidc__if we were going to adopt the one you guys have [that kinda bootloader is on our todo list, just hasn't been done yet] we would probably add hfs+ support
00:03:21davidc__how tight is the filesystem code in there?
00:03:25linuxstbThat would be very nice.
00:03:26davidc__is it easy to add a new filesystem?
00:03:34preglowmight be
00:03:44*preglow prods amiconn
00:03:51davidc__well, for the bootloader it would just need to be read only
00:04:02davidc__and if we were to be really lazy, just make it read the root dir only
00:04:31 Quit AliasCoffee ("Leaving")
00:04:35davidc__thats all you need to do to grab a single file
00:04:46 Quit matsl ("ERC Version 5.0 (CVS) $Revision: 1.776 $ (IRC client for Emacs)")
00:05:02linuxstbSomething we're missing is a GUI installer for Windows and Mac users - it seems you have that area covered already.
00:05:23 Quit webguest86 ("CGI:IRC")
00:05:24davidc__they're not really official projects of us
00:05:29davidc__others ahve written then
00:05:30 Join webguest15 [0] (
00:05:35davidc__and they horribly break nanos for your reference ;)
00:05:39ender`linuxstb: what kind of a GUI installer?
00:05:50webguest15NEED HELP: Does somebody uses IRC over Trillian???
00:06:04linuxstbender`: Something that performs raw writes to the hard disk in the ipod and repartitions the disk.
00:06:20webguest15i use now the web-irc from rockbox but i want to use it with trillian
00:06:27linuxstbIt will be needed for Windows users to install the ipod bootloader.
00:06:34ender`oh, for ipod only
00:06:51linuxstbYes, we're just talking ipod atm.
00:07:03webguest15do i need to register? i just added a random username and ... now i am connected but i cant open a chat window
00:07:23 Join speed123 [0] (
00:07:40speed123its working, ok
00:07:41linuxstbdavidc__: I think HFS would be very welcome in Rockbox. It would mean Mac users of any of the supported devices could switch to their native filesystem.
00:07:49 Quit webguest15 (Client Quit)
00:07:57amiconnpreglow: I think adding a completely new filesystem will be difficult. I'm not a filesystem expert, but the file and dir code is pretty much tailored for fat afaik
00:08:19 Part speed123
00:08:32 Join speed123 [0] (
00:08:36davidc__does rockbox rely on write support?
00:08:53davidc__if its gonna be really hard to add a new filesystem?
00:09:09CassandraAbsolutely. Lots of stuff gets written to disk.
00:09:19 Quit Mordov (Read error: 104 (Connection reset by peer))
00:09:40Bagdercan a music player do good without?
00:10:06Bagderresume, settings, etc
00:10:12amiconnrockbox doesn't rely heavily on write support, at least not for development
00:10:26amiconnThe current settings are stored outside the filesystem
00:11:00davidc__an easy way to do write support is to support not resizing files
00:11:00 Join Mordov [0] (
00:11:01amiconnRockbox did run happily on Ondio when there was no write support in the MMC driver
00:11:04davidc__thats almost as easy as pure read
00:11:06Bagderyes, it could boot fine without write
00:11:45amiconn...means it was able to play music etc. Of course no settings changes survived a reboot
00:11:55 Quit speed123 (Remote closed the connection)
00:12:01Bagderit was less featurepacked
00:12:16linuxstbdavidc__: Do the HFS ipods still have a large boot partition?
00:12:38davidc__on hfs+ ipods everything cept the kernel is on hfs+
00:12:46davidc__only the kernel goes into that firmware partition
00:13:01davidc__on fat ipods we need to create that ext2 stub because fat + /dev get bitchy
00:13:17linuxstbYes, I was talking about the firmware parition - i.e. the first partition.
00:13:30davidc__well, every ipod has that first partition
00:13:41davidc__and it has to be there because the ipods flash bootloader looks at that
00:13:48davidc__and we really would rather not tamper with flash
00:13:58davidc__you know how the ipod boot process works right?
00:14:16linuxstbSo ignoring Linux, the HFS and FAT32 ipods the same, apart from the filesystem on the second partition?
00:14:36davidc__well, there is one more variation
00:14:54davidc__mac ipods have an Apple Partition Table partition in the first partition
00:14:58davidc__just to make things more fun
00:15:20davidc__basically the way it ends up for us, the root partition is always partition 3
00:15:23davidc__which makes life a bit easier
00:17:24 Join speed123 [0] (
00:17:32 Part speed123
00:18:10 Join speed123 [0] (
00:19:52 Quit speed123 (Client Quit)
00:20:55 Join speed123 [0] (
00:21:02 Part speed123
00:21:10amiconnTeh 16 bit lcd driver needs loadsa work :/
00:22:36davidc__I gotta go back to work
00:22:40linuxstbamiconn: Yes.
00:22:40davidc__back in 45
00:22:52linuxstbdavidc__: See you later. Thanks for the info.
00:25:02linuxstbamiconn: What's your solution for the get_*ground functions?
00:25:55amiconnlcd_set_*ground() and lcd_get_*ground() need to be changed
00:26:16linuxstbThe green bitmask is wrong in those functions - it should be 0x07e0
00:26:19amiconnThey're supposed to take resp. return values in the lcd range, not in 0..255 range
00:26:57amiconn...but there's more, which I will work on as soon as I've committed 16 bit support for the win32 sim :)
00:27:18linuxstbAh OK.
00:27:51amiconn..which I am about to; just some final test builds to avoid RED
00:28:04linuxstbAre you building the ipod sims?
00:28:51amiconnNo, H300
00:29:51preglowdoes leach and the fellows document their findings in any way other than source code?
00:30:18amiconnFor building Ipod I would (1) need proper images and (2) a proper button assignment
00:30:39linuxstbOnly what's on the wiki pages I think. I think it's just in their heads.
00:31:38 Quit ender` (Read error: 104 (Connection reset by peer))
00:32:05amiconnIn fact adding 16 bit (and 24 bit) support to the windows sim is simple: Just replace the 8bit display bitmap with a 32bit one, supply a colourmap for when windows runs on a <=8 bit display, add the bit fiddling to lcd_update(_rect)() and you're done
00:32:44amiconnNow compare this to the x11 way of doing things ;)
00:37:44 Quit yngwi ("Chatzilla 0.9.68a [Firefox 1.0.7/20050919]")
00:43:10*Cassandra wonders how evil hardcoding a wps_reset when a request to load "/.rockbox/wps/rockbox_default.wps" is made.
00:43:52CassandraThat way we just dump a dummy rockbox_default.wps in the wps dir and it works for both platforms.
00:44:15CassandraOn the downside, user gets a surprise if they try to edit rockbox_default.wps, since it's never actually loaded.
00:44:29Cassandraall platforms, even.
00:45:33Paul_The_NerdYou might be able to remove the default file
00:45:54CassandraWell, you would, but then there'd be no way to select it for 'loading'.
00:45:58Paul_The_NerdAnd then, when a user chooses to browse WPS, add a "default" option that's there even though there's no file present.
00:46:16CassandraStill, if you muck about with files installed by rockbox, you're pretty much asking for trouble.
00:46:27Paul_The_NerdMuch like there's no *actual* .. folder, even though you often see a visual representation of it.
00:46:44CassandraAh, yes. That makes the code more complex though.
00:46:55CassandraI'll just go see how feasible that is.
00:47:05*Cassandra is the worlds laziest coder.
00:47:23Paul_The_NerdI wouldn't bet on that.
00:47:38Paul_The_NerdI wasn't going for the "easiest" solution... more a "What I'd say would be best." :)
00:51:46Paul_The_NerdOf course, if I really wanted it that badly, I'd go and do it. :-P
00:52:12CassandraYou know, I've just remembered why I hate the dirbrowse code.
00:54:18TiMiDwhy ?
00:54:33CassandraIt hurts my head.
00:54:34TiMiD(I also hate it ...)
00:54:43CassandraTo be fair a lot of Rockbox hurts my head.
00:54:46 Quit davidd ()
00:54:56TiMiDme too
00:54:58CassandraFull of special cases and other nastiness.
00:55:03TiMiDnot oo programming
00:55:43TiMiDlot of parts of the code could be made easier to read by just organizing the code better
00:56:15TiMiDtaht would allow programmers to use the other's code more easily
00:56:48CassandraRockbox evolved organically.
00:56:59CassandraIt really could do with a ground up rewrite.
00:57:00TiMiDorganically ?
00:57:07CassandraBut no-one really has the time.
00:57:13Paul_The_NerdAs in: Without a plan.
00:57:23TiMiDok :)
00:57:24Paul_The_NerdIt evolved based on need and want.
00:57:29TiMiDmessily :)
00:57:36CassandraNo-one was writing to a design spec, or if they were what it does now is vastly different from what it did then.
00:58:04TiMiDwhen I write something, I try to make the original code easier to read
00:58:18TiMiDbut it's difficult if you don't want to break everything
00:58:29TiMiDeverything is very tied
00:58:39Paul_The_NerdYeah, for me the process goes: Step 1: Make it work. Step 2: Make everything as simple to read as possible. Step 3: Try to get the hang of this whole 'comment' thing because i really suck at it.
00:58:45Cassandra*nods* Could do with a few more comments regarding what the code is actually for, really.
00:59:02TiMiDat least what each fn does
00:59:27TiMiDbut the pbl is that lot of function doesn' have a precise aim
00:59:28***Saving seen data "./dancer.seen"
00:59:34TiMiDif you see what I mean
01:00:07TiMiDI prefer when a fn does only one single tasj but does it well
01:00:12Paul_The_NerdHehehe... //This function does... err... something.
01:00:15CassandraRight - I can't be arsed to create a virtual file just now. I'll do it the easy way.
01:00:20CassandraSomeone may fix it later.
01:00:33TiMiDvirtual file oO
01:01:44TiMiDon what are you working ?
01:04:23Paul_The_NerdBasically, the idea was that when one went to "Browse .wps" there would be a choice there other than the files in the wps folder called simply "Default" (or something like that) that loaded the default WPS.
01:05:39ashridahi'd actually use the term 'Built in' over 'default', myself
01:06:49Paul_The_NerdMight be better, yeah
01:08:32ashridahjust not sure how well the distinction would translate to other languages
01:09:51Paul_The_NerdTrue. I still tend to think relative to one language only.
01:10:02amiconnHah, the inverse bar selector does look really distinct on a colour lcd :)
01:10:09Cassandrarockbox_default is consistant with the fonts.
01:11:00amiconnYes, inverse means complemented colours. The current colour choice is black text on "rockbox-blue" background
01:11:20amiconnThe inverse bar ends up with white text on red-brown background
01:11:44CassandraI wish the logo wasn't that disgusting shade of orange.
01:12:22Paul_The_NerdSo, actual inverse might not be the best idea?
01:12:31amiconnSomething is still wrong with mono bitmaps. bounce.rock looks way off
01:12:55amiconnShouldn't be hard to fix though
01:13:39CassandraDo I need to run a make install on the sim builds - I forget.
01:13:55amiconnYou should
01:14:13amiconnIt's basically a 'make zip' followed by auto-unzipping to the archos/ dir
01:14:27amiconnSo all your codecs, plugins etc get updated
01:17:28*Cassandra wonders who checked in default.wps.
01:17:46 Join Maxime`` [0] (
01:17:47 Quit Maxou` (Read error: 104 (Connection reset by peer))
01:18:58 Join tvelocity [0] (n=tony@
01:24:37Mordovhow do I get rid of the uppeermost line on iRiver when writng a .wps file,, I want to display the info elsewhere:)
01:24:56CassandraYou set "status bar" off.
01:25:24Mordov:) thanks
01:26:35ashridahthere really should be a way to flag that in the wps, since it seems to offset and cutoff images
01:30:03pinkutankdid anyone with a noninterrupting jpeg build come in while I was afk
01:30:37 Part tvelocity ("Leaving")
01:31:10amiconnlinuxstb: Hmm, the current lcd_mono_bitmap_part() implementation neglects a lot of things.
01:31:28amiconnIt only works right for <10% of all cases
01:31:32linuxstbI know.
01:31:34CassandraDammit! What are the keys for the iRiver sim?
01:31:48linuxstbI only just wrote the very minimum to get text display working.
01:32:20Cassandra(on X11)
01:32:24amiconnI am working on it
01:32:33amiconnIt's basically a reimplementation
01:32:39TiMiDPaul_The_Nerd: why not simply make the code load a "default.wps" file ?
01:32:51TiMiDand removing the built_in wps
01:33:00amiconnThat's bad
01:33:05TiMiDthat would be easier
01:33:13TiMiDwhy is it bad ?
01:33:20amiconnWe should never have to rely on specific files
01:33:25CassandraMeans rockbox falls over if you don't have access to the WPS file.
01:33:34amiconnThere must be builtin sane values
01:33:40CassandraRockbox doesn't require external files to boot.
01:33:58TiMiDwell if rockbox can't load a wps file it won't be able to play an audio file either ...
01:34:17amiconnIt would just use the default wps and go on...
01:34:25TiMiDwhen you have no .rockbox it doesn't work
01:34:43amiconnIt does, at least on hwcodec platforms
01:34:46TiMiDI assume that the user that removes system files want to go into troubles
01:35:01 Quit Kohlriba (Read error: 104 (Connection reset by peer))
01:35:01amiconnIt just complains several times that there is no playlist control file, but it works
01:35:22TiMiDbut what the interest for the user ?
01:35:30TiMiDto remove the .rockbox folder
01:35:45amiconnNo, but he might have deleted the file by accident
01:35:58TiMiDthen the user have to put it again
01:36:09amiconnOn a trip abroad?
01:36:36TiMiDthats a good point
01:37:01TiMiDbut if he deletes the firmware file
01:37:02 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
01:37:06TiMiDsame problem XD
01:37:10amiconnOf course rockbox cannot continue when an essential file is missing (like a .codec on swcodec pfms), but missing non-essential files shouldn't stop it
01:37:21Paul_The_NerdThat's the argument for hiding the firmware file in the .rockbox folder. there are fonts, .wps, languages...
01:37:33 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
01:38:10TiMiDwell what about the wps displays nothing when default wps is deleted ?
01:38:25amiconnYou can boot rockbox with just the firmware file itself in place (even without it when booting from flash), and on hwcodec platforms it is even able to play music
01:38:36TiMiD(which shouldn't happen in normal use)
01:39:23amiconnOf course english only, default font, default wps, and complaining about playlist_control and .rockbox folder every now and then
01:40:20TiMiDat least when the default.wps is found it would use it
01:40:40TiMiDand when not just use built-in
01:41:01amiconnThat it could do, but then the packager needs to be able to package different default.wps per target
01:41:56TiMiDthat's just a problem at the makefile level ;)
01:42:11amiconnperl script in this case
01:42:27TiMiDwell perl script :p
01:42:43TiMiDnot a big deal at this level I think
01:43:03TiMiDjust rename the correct default file to default.wps
01:43:28TiMiDfor example a default-iriver.wps > default.wps
01:44:03 Join DJDD_ [0] (
01:45:43 Part Zoric
01:47:40 Join webguest47 [0] (
01:48:20webguest47was geht
01:48:55webguest47ist da jemand
01:50:20TiMiDsonna ji ni daremo ga itte iru :/
01:50:54 Join Zoric [0] (
01:50:55TiMiDboku dake XD
01:51:42 Part Zoric
01:52:44markunwebguest47: Englisch bitte :)
01:54:46webguest47whats uppppp????
01:55:08webguest47so what is the status of lcd driver for h3**?
02:00:03 Quit webguest47 ("CGI:IRC (EOF)")
02:00:31markunnot sure, did you see the picture?
02:03:05pinkutanklast update was the backlight i think
02:03:50pinkutankdid anyone with a noninterrupting jpeg build come in while I was afk
02:05:27pinkutanklol timid can overrule everyonje with his japanese
02:12:26TiMiDno I can't :(
02:13:19 Quit actionshrimp ("a bird in the bush is worth two in your house")
02:13:22TiMiDI was speaking with a japanese girl at this evening's party and I definitly can't speak japanese very well :(
02:13:56CassandraCan someone with a bleeding edge build check to see if they have a weird display artefact in the top left of the screen with the "boxes" WPS, please?
02:14:31pinkutankI tried to decipher what you just wrote, I can pcikup words but no go
02:14:51TiMiDuh ?
02:14:53pinkutankthings, everyone/noone , say , I, to
02:15:15pinkutank"sonna ji ni daremo ga itte iru , boku dake XD"
02:15:16TiMiDsonna > this kind of
02:15:23TiMiDji > hour
02:15:35TiMiDni > at
02:15:52pinkutankdaremo ga
02:15:54TiMiDdaremo > nobody
02:15:57pinkutankevberyone or noone
02:15:58pinkutankoh ok
02:16:02pinkutanknobody is
02:16:10pinkutanksaying something
02:16:22pinkutankwhat is boku dake
02:16:27TiMiDitte iru > be there
02:16:49TiMiD(more like "is used to be there", but my english is limited)
02:16:57TiMiDboku > I
02:17:01TiMiDdake > only
02:17:06pinkutankoh I udnerstand, carppy dictionary
02:17:18pinkutankI looked up itte iru , it said speak
02:17:22pinkutankoh well heh
02:17:29TiMiDyes dictionnaries are a little difficult to use in japanese
02:17:29pinkutankok I get it
02:17:39TiMiDsince a word can mean a lot of things
02:17:51pinkutankwell, I'll be better off when I start taking lessons after december
02:18:06pinkutankyay its 3:20 am again
02:18:19pinkutankgonna go to school in 3 hours
02:18:33TiMiD2:20 here
02:18:41pinkutankeveryday this shit happens, and I've done zen zen
02:18:51TiMiDI'm just finishing smth on the code I was working on
02:18:55TiMiDand sleeeep :)
02:19:37pinkutankI had to work for my english orals
02:19:39pinkutankoh well
02:19:44pinkutankher egoes anyhter day
02:19:45TiMiD(silly playlist code)
02:19:59pinkutankdo you use the jpeg viewer?
02:21:17pinkutankhoped youd have a compiled one patched with the patch applied
02:21:25TiMiDI use my iriver only in the street so remote only
02:21:29pinkutankI should learn compiling asap
02:21:44TiMiDyes it's not very difficult
02:21:51pinkutankdoes it take any time?
02:21:57pinkutankI already patched the jpeg.c
02:21:58TiMiDto lern
02:22:09TiMiDto compile
02:22:13pinkutankI dont know how to make that into the firmware
02:22:14TiMiDmaybe 1 minute
02:22:24TiMiDyou have a patched source ?
02:22:39TiMiDand a build environment ?
02:22:42pinkutankI have a patched source of the jpeg plugin of the daily build
02:22:49pinkutankbut I dont have anything else
02:22:59pinkutankand I have the programs installed of course
02:23:02TiMiDjust go on the wiki and download the build env
02:23:07TiMiD(it's an installer)
02:23:15pinkutankthe dev kit?
02:23:21pinkutankI have the dev kit
02:23:38pinkutankI have the tarball
02:23:49pinkutankI have a patched jpeg.c
02:23:51TiMiDthen move the sources tou the home/guest dir in the devkit
02:24:18pinkutankthe source is common right? I Will choose h120 sometime later?
02:24:52pinkutanki did move them along
02:25:03TiMiDthen lauch the devkit
02:25:14pinkutankIm looking for my jpeg c
02:27:50CassandraWaiting, waiting, for the build to make. / So many bloody platforms and so very quick to break.
02:28:00Cassandra</Les Mis>
02:28:51TiMiDLes Mis ?
02:28:53 Quit dpassen1 ()
02:28:53pinkutankwhat is tmpatname
02:29:09CassandraLes Miserables. A popular musical from the 80s.
02:29:11pinkutanki couldnt find my .c file, repatching, but the patch .exe is malfunctioning
02:29:12TiMiDpinkutank: I don't know XD
02:29:27TiMiDCassandra: T francais ?
02:29:29CassandraI may have taken some liberties with the words.
02:29:50CassandraIt's a French musical, yes. but no I'm not French.
02:29:51pinkutankhe asked if you speak french
02:30:07TiMiDI asked if he was french
02:30:07pinkutanki thought you were quoting cass
02:30:20TiMiDsince I'm french
02:30:22CassandraUn petit peut. Mais ma francais est tres mal.
02:30:32TiMiDnot a lor fo french people heure
02:30:42pinkutankI speak a little bit
02:30:47TiMiDI can understand at least :p
02:31:12TiMiD"Un peu, mais mon francais n'est pas bon"
02:31:27TiMiDor "Un peu mais mon francais est mauvais"
02:31:48CassandraIch kann auch ein bischen Deutsch.
02:32:18pinkutankun peu mais mon français are vraiment mal donc je ne vai(i forgot how to use voudrais) pas parles
02:32:25TiMiDnihongo wo hanasu koto ga sukoshi jozu ni dekimasu
02:32:27pinkutankdid i say are
02:32:29pinkutankoh lol
02:32:52pinkutanknihongo bansai
02:33:40TiMiD(I admint that to learn french must be a nightmare for a forigner)
02:34:12TiMiDbecase it's a nightmare for us to write it correctly :p
02:34:18pinkutankits not a nightmare
02:34:29pinkutankbut I only get like 40 mins per week of non serious ed.
02:34:51TiMiDcompared to korean or japanese, it's a nightmare :)
02:35:26CassandraWell English is no walk in the park either, I suppose.
02:35:35TiMiDit's easier
02:35:39CassandraBeing an English speaker makes one lazy.
02:35:54CassandraThe entire western world understands English, pretty much. :)
02:35:55TiMiDverbs are simple
02:36:13TiMiDit's because we have no choice :(
02:37:54CassandraWell you should have made a better job of taking over America then. You've only yourselves to blame, really. :)
02:38:47TiMiDI'm happy in france :)
02:38:47pinkutankso how do i compile timid
02:39:07CassandraI think if you compile timid he might get upset.
02:39:17pinkutankoh lol
02:40:07TiMiDwell just make a "build" folder inside your source tree
02:40:16TiMiDthen go t the tools dir
02:40:26TiMiDcd rockbox/tools
02:40:30TiMiDtype make
02:40:31pinkutankbuild folder? where
02:40:42TiMiDin the rockbox/ folder
02:41:21pinkutankam i operating commandline
02:41:28pinkutankor rockbox .bat
02:42:02pinkutankwhat do i wrtie
02:42:14amiconnlinuxstb: There. Shiny lcd-16bit.c :-)
02:42:23TiMiDtype cd rockbox/tools
02:42:33pinkutankthere inst a tools folder here
02:42:48TiMiDwhere rockbox is the source tree root folder
02:43:12pinkutankyes there isnt one, but there is one in the daily rarball
02:43:42TiMiDwell the name of the folder :)
02:44:08TiMiDnotice that when you are in command line you can press the TAB key and it completes the word you are typing
02:44:22TiMiDtype cd rockbox-daily-20051114/tools
02:44:31pinkutankand make
02:44:37pinkutankim already in the dir
02:44:39pinkutankthats all?
02:44:48TiMiDthat's the frst step
02:44:49 Quit tcmjr ("Tickle-Me Elmo uses The 7 Deadly Sins.. *giggle* *giggle* *giggle*   []")
02:44:56pinkutankit gave lots of errors
02:45:02pinkutanksaying this and that was udneclared
02:45:07CassandraAmazing. Nothing broke.
02:45:12CassandraRight, bed for me.
02:45:32pinkutankall was ipod errors tho
02:45:42TiMiDI get no errors
02:45:51TiMiDI will try to cvs update
02:46:07pinkutankletme try todays build
02:47:23pinkutanksorry fot taking your itme
02:47:31ashridahhm. regarding the latest cvs commit, wouldn't it be reasonable to provide for the situation where the user replaces the virtual file with a real 'default' wps?
02:47:50CassandraYes. Feel free to write a patch.
02:48:21CassandraThe user really shouldn't be replacing files in the default Rockbox build though. They'll get overwritten when they reinstall.
02:48:26pinkutankwhy do you need one, doesnt it automatically load the latest wps you loaded
02:48:42TiMiDpinkutank: I can't help you with the tools compile errors, it works here
02:49:06TiMiDCassandra: why not to make a real default file instead of a virtual one ?
02:49:06ashridahCassandra: but isn't the 'rockbox_default.wps' thing a virtual file?
02:49:07pinkutankim retrying
02:49:25ashridahTiMiD: then it won't work if there's no wps's at all, including the default file.
02:49:40Cassandraashridah: Because tree.c is bloody confusing and it seemed better to get a fix out there than to worry about niceties like that.
02:49:48ashridahCassandra: righto then
02:49:53TiMiDwhy would there be no files ?
02:50:05TiMiDin that case, it loads the default wps
02:50:47pinkutanknormally what should i be getting?
02:50:58TiMiDkevin@harkonnen ~/rockbox-devel/tools $ make
02:50:58TiMiDcc -O -ansi -g -c -o scramble.o scramble.c
02:50:58TiMiDcc -g scramble.o iriver.o -o scramble
02:50:58DBUGEnqueued KICK TiMiD
02:50:58TiMiDcc -DAPPLICATION_NAME=\"bmp2rb\" -g bmp2rb.c -o bmp2rb
02:50:58TiMiDcc -g convbdf.c -o convbdf
02:51:01TiMiDcc -g ipod_fw.c -o ipod_fw
02:51:03TiMiDtools done
02:51:19pinkutankhow can i copy my error window to show the results
02:51:38TiMiDright click->select text
02:51:43CassandraI admit it's a bit of a hack, but it has the advantage of working crossplatform and not confusing the end user unless they try to do something odd.
02:51:53TiMiDthen press enter to copy the selected text
02:52:02pinkutankI cant select text
02:52:04CassandraThere's no real need to change the default WPS since your last WPS selection is remembered anyway.
02:52:05pinkutankit wont letme
02:52:56TiMiDpinkutank: I can't help you a lot since I'm under linux
02:53:15TiMiDbut right click gives you a menu
02:53:31TiMiDand in this menu yoou can tell it to let you select
02:53:58pinkutanknot reaally in my window, but i think i found a way of copying it
02:54:55Midgey34TiMiD: I'm getting errors (well warnings) on linux when building the tools directory
02:54:59pinkutankgcc -g ipod_fw.c -o ipod_fw
02:54:59pinkutankipod_fw.c:28:19: errno.h: No such file or directory
02:54:59DBUGEnqueued KICK pinkutank
02:54:59pinkutankipod_fw.c:29:20: unistd.h: No such file or directory
02:54:59pinkutankipod_fw.c: In function `copysum':
02:55:00***Alert Mode level 1
02:55:00pinkutankipod_fw.c:140: error: `errno' undeclared (first use in this function)
02:55:02pinkutankipod_fw.c:140: error: (Each undeclared identifier is reported only once
02:55:04pinkutankipod_fw.c:140: error: for each function it appears in.)
02:55:06pinkutankipod_fw.c: In function `load_entry':
02:55:08pinkutankipod_fw.c:166: error: `errno' undeclared (first use in this function)
02:55:10pinkutankipod_fw.c: In function `write_entry':
02:55:12pinkutankipod_fw.c:185: error: `errno' undeclared (first use in this function)
02:55:14pinkutankipod_fw.c: In function `extract':
02:55:16pinkutankipod_fw.c:218: error: `errno' undeclared (first use in this function)
02:55:18pinkutankipod_fw.c: In function `lengthof':
02:55:20pinkutankipod_fw.c:240: error: `errno' undeclared (first use in this function)
02:55:22pinkutankipod_fw.c: In function `main':
02:55:24pinkutankipod_fw.c:289: error: `opterr' undeclared (first use in this function)
02:55:26pinkutankipod_fw.c:312: error: `optarg' undeclared (first use in this function)
02:55:28***Alert Mode level 2
02:55:28pinkutankipod_fw.c:372: error: `errno' undeclared (first use in this function)
02:55:30***Alert Mode level 3
02:55:30pinkutankipod_fw.c:422: error: `optopt' undeclared (first use in this function)
02:55:32***Alert Mode level 4
02:55:32pinkutankipod_fw.c:431: error: `optind' undeclared (first use in this function)
02:55:34pinkutankmake: *** [ipod_fw] Error 1
02:55:38pinkutankthe ones before ipod are all gcc
02:56:25amiconnYou're using the devkit I presume?
02:57:37TiMiDI thought this issue was fixed
02:58:07amiconnWell, bluechip fixed it internally, but a new devkit isn't out yet
02:58:31TiMiDwell since it works here I don't care :p
02:58:38amiconnAfaik someone promised him an arm-elf-gcc for inclusion
02:59:15TiMiDI have a question about Ipod
02:59:31***Saving seen data "./dancer.seen"
02:59:33amiconnpinkutank: You can safely ignore these errors in tools/ as long as you don't want to build the ipod targets (which won't work anyway atm because of the missing arm crosscompiler)
02:59:37TiMiDcould eventually ipod nano play ogg files ?
02:59:46pinkutanki dont care about ipod
02:59:51pinkutankbut what do i do know
03:00:00TiMiDI mean is it software codecs or hw codecs ?
03:00:18pinkutanki created a makefile
03:00:26pinkutankusing configure
03:00:42TiMiDok so it could play ogg
03:01:04pinkutankwhat do i do now?
03:01:05amiconnIt will take some time, rockbox is in its early stages on ipod
03:01:28TiMiDyes, but I know a friend that wanted a flash based player that plays ogg
03:01:38TiMiDand there is no choices
03:01:48TiMiDipod nano is the cheapest
03:02:11TiMiDand compared to the iriver T series, it's UMS
03:02:17pinkutanktimid how can i use my makefile
03:02:25TiMiDpinkutank: just type make
03:02:34pinkutankwhere should i store my makefile
03:02:39pinkutanki made it in another dir
03:02:39TiMiDI hope you created it in the build directory
03:02:46CassandraI suspect a video iPod will be my next Rockbox device.
03:02:46TiMiDgreat then
03:03:00TiMiDtype make in that dir and it woll be good :p
03:03:02CassandraBut to be honest, the H140 does everything I could want.
03:03:22TiMiDyes I love my h140 with rb XD
03:03:30TiMiDbut if you want a flash player
03:04:09CassandraiPod nano is probably your best bet. But probably not for 6 months or so.
03:04:22pinkutankit isnt working again :( crap
03:04:49TiMiDif I was kind I would compile it for you
03:04:50pinkutankhome guest firmware no such file ...
03:05:07TiMiDwhat patch is it ?
03:05:21pinkutankjpeg, bu tnevermin, I'm not gonna use it tomottow
03:05:26pinkutankI took enough of your time
03:05:29pinkutankgonna sleep now
03:05:33TiMiDme2 :)
03:05:34***Alert Mode OFF
03:05:39pinkutankand will learn by trial error tomorrow
03:05:56pinkutanki can udnerstand whats wrong
03:06:14pinkutankIn the config i put in the firmware directory wrong
03:06:22TiMiDamiconn: I'm going to commit the playlist-viewer patch, a big patch ...
03:06:27pinkutanki just need to edit the makefile a tad bit and ll will be good
03:06:43TiMiDpinkutank: you don't need to put it in the make file oO
03:06:54TiMiDthe configure script detects it for you
03:06:57pinkutanki know
03:07:01pinkutankbut it detected it as
03:07:21pinkutank"/firmware :D"
03:07:45TiMiDsomething is going wrong then :)
03:07:59TiMiDyou will maybe have other problems :(
03:08:45pinkutankwell I'll see tomorrow
03:08:50pinkutankif I dont die in school
03:08:56pinkutank2:30 till school
03:09:01pinkutankand I'm having ib orals
03:09:23pinkutankI'm really and asshole towards myself, complete idiot, baka baka baka
03:09:37TiMiDahou :)
03:09:44 Quit pinkutank (Read error: 104 (Connection reset by peer))
03:09:48 Quit Cassandra (" HydraIRC -> <- Go on, try it!")
03:09:59 Join Vlad0man [0] (
03:16:49 Join webguest89 [0] (
03:17:07 Part Midgey34
03:17:25 Join solexx_ [0] (
03:25:50 Quit Vladoman (Read error: 110 (Connection timed out))
03:26:50 Quit XShocK (Remote closed the connection)
03:30:15 Quit solexx (Read error: 110 (Connection timed out))
03:35:10 Quit webguest89 ("CGI:IRC")
04:06:23Mordovis there a display simulator for WSP fils? so I dont have to load it unto the iRiver al the time?
04:08:52lostlogic<3 H300 support is moving!
04:20:36Mordov# Each line has the format "setting: value". Values must be within the ranges specified in this manual for each setting.
04:20:43 Join Paul_The_Nerd [0] (
04:20:44Mordovcan someone tell me where to find this?
04:21:13Mordovis there a CFG index
04:23:27ashridahMordov: there's a simulator for rockbox. it should support displaying WPS's
04:23:35ashridahit's part of the same source, there might be daily builds of it
04:24:11ashridahhm. guess not
04:28:23Mordovdo you have a link?
04:32:43Mordovah don't bother.. I have to put this down a get some sleep :)
04:35:04ashridahthere's no daily builds of the simulators, it seems. but if you grab the devkit (for windows) or build a cross compiler (unixalikes) you could build your own, or ask someone to package theirs up and put it up for you to download.
04:39:31MordovI have put your last sentec in a text file on my desktop ready to understand tomorrow :)
04:40:10 Quit Seed (Nick collision from services.)
04:40:14 Join Seedy [0] (
04:50:18 Join Dragon_Fangs [0] (
04:59:13 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
04:59:32***Saving seen data "./dancer.seen"
05:13:04 Quit elinenbe (Read error: 110 (Connection timed out))
05:28:28 Quit Dragon_Fangs ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
05:32:25 Join frcsyk [0] (
05:56:47 Quit frcsyk ()
06:03:00 Join Lost-ash [0] (
06:08:56 Quit paugh ("bbiaf")
06:10:58 Join wacky [0] (
06:11:19wackyhey guys!
06:11:45 Quit ashridah (Nick collision from services.)
06:11:49 Nick Lost-ash is now known as ashridah (
06:13:20 Quit |joshn| ("KVIrc 3.2.0 'Realia'")
06:25:23 Quit wacky ("Bad BitchX, bad BitchX, whatcha gonna do? Whatcha gonna do when they come for you?")
06:26:35 Join Jungti1234 [0] (n=jungti12@
06:58:34 Quit ghode|afk (
06:58:34 Quit solexx_ (
06:58:34 Quit Vlad0man (
06:58:34 Quit Mordov (
06:58:34 Quit Zagor (
06:58:34 Quit linuxstb (
06:58:34 Quit BirdFish (
06:58:34 Quit ripnetuk (
06:58:34 Quit wubbla (
06:58:34 Quit crash_ (
06:58:34 Quit Igg-man (
06:58:34 Quit pill (
06:58:34 Quit thegeek_ (
06:58:34 Quit ]Maggie[ (
06:58:34 Quit crwl (
06:58:34 Quit pabs (
06:58:34 Quit Bagder (
06:58:34 Quit Seedy (
06:58:34 Quit Sandking (
06:58:34 Quit ansivirus__ (
06:58:34 Quit CoCoLUS (
06:58:34 Quit davidc__ (
06:58:34 Quit Strath (
06:58:34 Quit chiller (
06:58:34 Quit Nilisco (
06:58:34 Quit dwihno (
06:58:34 Quit bluebrother^ (
06:58:34 Quit ze (
06:59:33***Saving seen data "./dancer.seen"
06:59:50 Quit RotAtoR ()
07:00:52 Quit Jungti1234 (Killed by (Nick collision))
07:01:14 Join Jungti1234 [0] (n=jungti12@
07:01:14NJoinSeedy [0] (
07:01:14NJoinsolexx_ [0] (
07:01:14NJoinVlad0man [0] (
07:01:14NJoinMordov [0] (
07:01:14NJoinSandking [0] (
07:01:14NJoindavidc__ [0] (
07:01:14 Join Zagor [0] (n=bjst@pdpc/supporter/sustaining/Zagor)
07:01:14NJoinlinuxstb [0] (
07:01:14NJoinansivirus__ [0] (
07:01:14NJoinBirdFish [0] (n=bradbox8@
07:01:14NJoinCoCoLUS [0] (
07:01:14NJoinghode|afk [0] (
07:01:14NJoinStrath [0] (
07:01:14NJoinripnetuk [0] (
07:01:14NJoinwubbla [0] (
07:01:14NJoinchiller [0] (
07:01:14NJointhegeek_ [0] (
07:01:14NJoinIgg-man [0] (
07:01:14NJoinbluebrother^ [0] (
07:01:14NJoindwihno [0] (n=dw@
07:01:14NJoinpabs [0] (
07:01:14NJoinNilisco [0] (
07:01:14NJoinBagder [0] (
07:01:14NJoincrwl [0] (
07:01:14NJoin]Maggie[ [0] (
07:01:14NJoinze [0] (
07:01:14NJoincrash_ [0] (
07:01:14NJoinpill [0] (
07:01:16***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! ( )'
07:05:13CtcpIgnored 2 CTCP requests in 4 minutes and 1 second at the last flood
07:05:13CtcpVersion from freenode-connect!freenode@freenode/bot/connect
07:05:13***Server message 477: 'logbot #rockbox :[freenode-info] Private messages from unregistered users are currently blocked, except to network staff, to services (NickServ, etc.), and to registered users who have unblocked them. Please see the freenode website for more information:'
07:19:11 Part Jungti1234
07:20:04 Join Jungti1234 [0] (n=jungti12@
07:24:42 Join B4gder [0] (
07:37:32 Join speed123 [0] (
07:44:55 Join ender` [0] (i=ychat@
07:45:34 Part speed123
08:00:01 Join _FireFly_ [0] (
08:21:53 Quit _FireFly_ ("Leaving")
08:51:58 Join Gibbed [0] (
08:52:00 Quit Rick (Nick collision from services.)
08:52:10 Nick Gibbed is now known as Rick (
08:59:37***Saving seen data "./dancer.seen"
08:59:54 Join _FireFly_ [0] (
09:00:36 Join webguest61 [0] (
09:14:23 Quit chopped_pork (Remote closed the connection)
09:24:08 Quit Strath (Read error: 104 (Connection reset by peer))
09:46:29 Join chopped_pork [0] (
09:50:15Jungti1234At where can uisimulator download?
09:50:35B4gderthere are no binary downloads of the sim
09:50:41B4gderbuild your own
09:51:24B4gderthe sim is only useful for development
09:51:50Jungti1234What is this?
09:52:03markunB4gder: I found out that the CPU of the Gigabeat is SAMSUNG S3C2440AL-30. Found a PDF of the S3C2440A, but I don't know what the difference is.
09:52:24Jungti1234markun: hi
09:52:29markunJungti1234: morning
09:52:32B4gderARM9 I noticed
09:52:39*B4gder works with an ARM9 at work
09:53:05ashridahJungti1234: that's a view of the Attic for the win32 directory of the simulator. It's old files that've been removed from the current cvs tree.
09:53:13markunI was reading the datasheet and must say I like it a lot. Also a mac instruction like the coldfire.
09:54:00markunDid you even work with thumb instructions?
09:54:08B4gderI did
09:54:32B4gderthere's a gcc option for them
09:54:37markunI guess it's usefull to save space
09:54:53B4gderits especially useful for 16bit memories
09:55:24markunThe samsung has integrate RTC, flash memory and an LCD controller.
09:55:43B4gderno usb?
09:55:49markunYes, also
09:56:38markunIt can opperate at 400MHz, but the gigabeat runs at 300, should be more than enough
09:58:45B4gderhehe, indeed
09:59:55B4gder"Samsung's S3C2410X processor was used recently in Unication's Linux-powered Magpie PDA"
10:00:18B4gderthere might be code for some of its magic
10:01:09 Quit webguest61 ("CGI:IRC")
10:01:43markunWell, the Gigabeat is also Linux based, no free magic there I'm affraid :)
10:02:14B4gderpossibly their LCD/flash/USB stuff is similar to other ARM9-based microcontrollers
10:03:46markunIt's all in the datasheet (maybe it's not the datasheet, it says "user's manual")
10:04:19B4gderI was just thinking that existing code for it would be even better
10:04:56markunyes, maybe
10:07:03markunDo you know what kind of ARM the ipod uses?
10:07:32B4gderbut with dual cores
10:08:19markunI hope 2 75MHz ARM7 CPUs are slower than 1 300MHz ARM9
10:08:34B4gderno doubts there
10:11:58B4gderbtw, my ARM9 has 16K iram
10:14:46markunI only see 16K i-cache and 16K d-cache, but no mention of iram
10:15:27 Join linuxstb_ [0] (
10:15:46B4gdercalled isram here
10:17:37markunCould you take a look at the S3C2440A manual?
10:19:21B4gderyes, soon
10:19:47B4gderquite busy trying to get our arm9 to run ;-)
10:20:12B4gderwe're just in the transition of going from a dev board to our own board
10:20:21B4gderand things just don't work
10:20:35markunok, good luck then
10:21:13B4gdergcc -march=armv5 is your friend
10:21:49B4gderand objdump -d
10:23:06markunok :)
10:23:42B4gderlinuxstb_: you checked if your ipod port 16bit writes are done as strh or two strb ?
10:24:12B4gdergcc for arm defaults to two strbs
10:27:23linuxstb_B4gder: No. Why would gcc do such a thing? Alignment issues?
10:31:32 Quit linuxstb (Read error: 113 (No route to host))
10:33:46B4gderI think so
10:34:04B4gderor possibly some compatibility issue with old arm archs
10:34:34 Join LinusN [0] (
10:35:50LinusNwee, got the h300 ata driver working
10:36:14linuxstb_Were there many changes to the H1x0?
10:36:20LinusNexcept that as soon as i power the hd, the bdm connection is lost :-(
10:36:56LinusNvery minor changes needed
10:37:11linuxstb_B4gder: Which version of gcc for arm do you use?
10:37:22LinusNthey seem to have learned their lesson and now make use of the IORDY signal
10:37:47LinusNbut rockbox doesn't use it, and it works fine without it
10:38:04LinusNi actually got as far as displaying the dir browser
10:38:22LinusNbut i have problems with the button driver
10:38:59LinusNthe pcf50606 chip handles buttons, adc and rtc
10:39:08LinusNusing i2c
10:39:23LinusNwhich means that we have concurrency problems
10:40:12 Join Zoric [0] (
10:40:34LinusNthe rtc is polled in a thread, while the buttons and adc are read in interrupts
10:41:37B4gdermarkun: it is only usb 1.1 it seems
10:42:00B4gderthe microcontroller
10:44:52B4gder"Boot Internal SRAM (4KB)"
10:45:01LinusNthe h300 has 3 keys connected to GPIO pins: ON, Rec and A/B
10:45:14LinusN(and Hold of course)
10:46:13markunI'm pretty sure the the usb slave part is 2.0, but I can't find it now
10:46:27B4gderit might be an external chip
10:46:49LinusNB4gder: but it has both data and instruction cache
10:47:23B4gder"4-KB internal SRAM buffer can be used for another purpose after NAND flash booting."
10:47:47LinusNlcd dma
10:48:46markunnot a lot or iram..
10:49:04B4gderon this beast iram won't be that critical
10:49:08LinusNmarkun: no, but it has a data cache
10:49:56markunLinusN: what do you think about the gigabeat players as new rockbox platform?
10:50:41B4gderactually, it strikes me as pretty similar to what the Neuros N3 wants to become
10:50:58B4gderbut without the annoying TI part
10:51:00LinusNmarkun: looks promising
10:52:13markunyes, also :)
10:52:29B4gderits a super-cool platform
10:52:33B4gderand a way cool lcd
10:53:01linuxstb_markun: Does the original gigabeat firmware play video?
10:53:09markunlinuxstb_: no
10:53:09LinusN64-way set-associative cache!!!!!
10:53:34linuxstb_markun: That's good news - I'm guessing you read about the new ipod video and its video co-processor.
10:53:45LinusN4-way, 16-word write buffer
10:53:52markunI did. But why is it good news?
10:53:55LinusNman, we don't need iram :-)
10:54:13linuxstb_Because it means it is unlikely to have a proprietory video DSP sitting between the CPU and the LCD.
10:55:22markunI posted another message on the gigabeat forum asking for more information of the components
10:59:39***Saving seen data "./dancer.seen"
11:01:20markunDoes the arm7 also have the MLAL instruction?
11:03:14Zagori'm not too fond of the lack of buttons though
11:03:52markunIt has 5 buttons on the right side and the scroll cross, also a hold switch on the top
11:07:13linuxstb_markun: I think so, and I certainly home so.
11:08:07 Quit goa (Read error: 110 (Connection timed out))
11:10:38Zagormarkun: yes but the touchy-feely navigation sliders scare me
11:11:24LinusNare they sliders?
11:11:57Zagortouch pads
11:12:59Zagorthe worst part of ipod becomes fashion...
11:13:07 Join amiconn_ [0] (
11:14:48 Join goa [0] (
11:16:28linuxstb_It should be possible to misuse the scroll-wheel on the ipod to allow the user to just touch the pad in different positions (i.e. not move their finger) to simulate normal buttons. So for each of the 4 button positions, we could have "touch", "long touch", "press", "long press".
11:17:15linuxstb_But maybe some people like the scrollwheel.
11:17:59Zagoryou're probably right. i still prefer clicky buttons though.
11:19:41_FireFly_TiMiD: the new playlist-viewer has some bugs.
11:19:56_FireFly_TiMiD: 1. no handling of remote-buttons
11:20:20_FireFly_TiMiD: 2. only one icon is shown
11:20:35Jungti1234hey markun
11:20:58wubblawow, even the ata driver is working now... congratulations to LinusN!!! :-D
11:21:44wubblawhat needs to be done before LinusN can start working on the bootloader?
11:24:38 Join Sandking_ [0] (
11:25:17B4gderhe works on it already
11:25:31B4gderit needs ata, buttons and lcd
11:26:51linuxstb_Will the patching of the iriver H3x0 firmware be done in the same way as the H1x0? Or will that be more work as well?
11:28:19ripnetukfirefly / timid - does that commit this morning include the firefly gui widget (remote wps) patch, or is that still not committed?
11:29:42 Quit amiconn (Read error: 110 (Connection timed out))
11:29:43 Nick amiconn_ is now known as amiconn (
11:30:11markunJungti1234 sent me this link:
11:30:23markunlooks like there are a lot of different LCD's to dupport :(
11:34:25linuxstb_markun: That doesn't look like fun.
11:34:26 Nick linuxstb_ is now known as linuxstb (
11:34:26 Quit Sandking (Read error: 110 (Connection timed out))
11:38:25markunsomeone is about to open his Gigabeat to take pictures of the internals
11:41:47markunlinuxstb: tapping the scroll pad also works on the H10. I think it's not so bad.
11:44:11linuxstbIt will be different to how the ipod currently behaves - but maybe that's not a bad thing. I would at least like to experiment with different ways of using the scrollwheel.
11:44:45linuxstbI don't like having to move my finger around the wheel when a simple "long touch" would do the same thing.
11:48:22amiconnRegarding the gigabeat - a 400 MHz CPU for a DAP sounds like a big waste to me
11:48:50B4gder400 is probably a maximum
11:48:53B4gderthat they don't use
11:51:31amiconnLinusN: What's the concurrency problem you mentioned? Afaiu we must not do i2c within an isr anyway?
11:52:21LinusNwell, that depends on if the pcf50606 clockstretches or not
11:52:37LinusNif it doesn't, we can do i2c as much as we want
11:53:13amiconni2c in isr means that the isr may break into a running i2c transfer from normal code
11:53:16LinusNas long as we don't interrupt an ongoing i2c operation
11:53:25amiconn...which will certainly mess up transfers
11:53:32LinusNthat's the concurrency problem
11:54:14LinusNwe need to synchronize the pcf50606 accesses
11:54:20amiconn...and iirc the pcf50606 i2c is shared with another component, so it seems i2c in isr is a no-no
11:54:38_FireFly_ripnetuk no it is only the playlist-viewer
11:55:00LinusNamiconn: it's shared with the eeprom, which we will not use
11:55:11_FireFly_it seams that currently no of the devs has the time to look at my wps-patch
11:55:48amiconnLinusN: So it could work okay from isr, provided that it's accessed from one isr exclusively
11:56:16amiconn...but in fact I would not like to block the ability to use the eeprom
11:56:35amiconnIt would be perfect for the binary part of the settings
11:56:46LinusN128 bytes
11:56:59 Join einhirn [0] (
11:57:02ashridahhas anyone tried using mingw to build rockbox?
11:57:02LinusNthe original firmware stores the settings in there
11:57:08 Quit einhirn (Client Quit)
11:57:09amiconnThe archos rtc has ~40 bytes, and we do use it
11:57:11 Join einhirn [0] (
11:57:38*ashridah imagines it's not that much different from cygwin, at least when building the firmware. simulators are probably a different matter)
11:57:57LinusNashridah: mingw uses the DOS shell
11:58:11ashridahnot if you couple it with msys
11:58:27LinusNtrue, but why mingw then?
11:59:16LinusNi haven't tried to build the cross compiler with mingw
11:59:24ashridahwell, obviously, for the firmware, it makes no difference, but it doesn't rely on a posix compatability library, you can build win32 applications with it. of course, that means it's not posix
11:59:37ashridahso i assume the simulators probably require the posix library cygwin provides.
11:59:45LinusNyes they do
12:00:50 Quit ashridah ("Leaving")
12:00:58 Join ashridah [0] (
12:00:59 Join Sando [0] (i=lolsteam@
12:01:18ashridahafter all that, and i still had the same ip address
12:01:20ashridahdidn't need to /reconnect :)
12:01:56 Part Zoric
12:09:51 Join elinenbe [0] (
12:10:10linuxstbashridah: There is also "msys" - which provides the normal build tools. Rockbox should compile with that.
12:12:02ashridahi know that msys exists, if you'll look carefully, i've already mentioned it once :)
12:12:25ashridahbut i was under the impression that's a toolchain, not a posix layer for windows :)
12:15:41linuxstbSorry, I didn't read the discussion properly. You're right of course.
12:16:37linuxstbIf it helps, I can cross-compile the windows sim under linux using i586-mingw32msvc-gcc
12:16:52linuxstbThis is probably the same as using mingw under msys
12:18:07ashridahlinuxstb: i don't really care that much, i was mostly curious
12:18:14*ashridah does all of his development on linux
12:22:45 Join Strath [0] (
12:24:26preglowmarkun: yes, the 64 bit math instructions is an arm5e feature
12:24:29preglowmarkun: so we've got it
12:25:11preglowi also need to use i2c in an interrupt, it seems
12:42:43amiconnB4gder: (1) Do you have an idea how to add colour support to the x11 sim? (2, related) How about adding a H300 win32 sim to the build table? Unlike the x11 sim, it won't go red, only yellow like the H300 target :-P
12:51:40preglowi'd advice you not to touch x11 if you're interested in retaining your sanity
12:55:25amiconnI'll probably work on removing the 1 yellow for H300 target and win32 sim tonight
12:55:33amiconnIt's caused by rockboy
12:58:39preglowwasn't someone planning to optimise rockboy?
12:59:43***Saving seen data "./dancer.seen"
12:59:48ashridahyeah, someone was working on dynamic recompilation.
12:59:53ashridahbut he gave up/got bored
13:00:52 Quit _FireFly_ (Read error: 110 (Connection timed out))
13:01:36 Join _FireFly_ [0] (
13:02:01 Quit _FireFly_ (Client Quit)
13:02:30preglowi don't think there's much to gain by that
13:02:33 Join _FireFly_ [0] (
13:02:44preglowi'd like to see an assembled optimised core and use of iram, though
13:03:35preglowoh, so there already is some use ofiram
13:04:06ashridahpreglow: the idea is to swap common bits of code with optimised versions
13:04:24ashridahso that they don't need to be interpreted
13:04:34preglowi know that
13:04:37preglowbut remember our code cache
13:04:41preglowit needs to be constantly flushed
13:04:53preglowwhen you write new code
13:05:08preglowthese cores can be hard to write
13:05:16preglowan assembler optimised interpretive core is far easier
13:05:25preglowand should be enough for at least h1x0
13:06:13ashridahwrite new code? it was mostly just a case of having it jump to prewritten routines iirc
13:07:19preglowplus, not even half of the iram is currently used
13:07:27preglowashridah: nah, that's not the case at all
13:07:42preglowashridah: you translate instruction for instruction to native code
13:08:11ashridahhe was working on replacing common routines, i don't know about the rest of it
13:08:42preglowcommon routines?
13:08:51preglowyou mean in the game code?
13:09:04preglowhow can you tell if there are any common routines there? each game might be wildly different
13:09:17linuxstbWhat's the status of sound in Rockboy? IIRC, it works but is currently disabled.
13:09:25ashridahapparently lots of them use a common library provded by nintendo
13:09:31preglowlinuxstb: yes, doesn't go too well with the playback code
13:09:46preglowashridah: but ok, that's indeed a more high level approach than dynamic recompilation alone
13:09:48B4gderamiconn: I have no idea how to fix the x11 sim, no, but then I don't even know how we do it on target
13:10:04B4gderand I'll add a sim build soon
13:10:21B4gderbusy now
13:10:50preglowashridah: still, i'm very certain an assembler core will do more than enough for us
13:11:07preglowand it'll work for all games, not just those using a library from nintendo
13:12:07Zagoryay. Name: Nick Smith Comment: Gigabeat owner - I work for Toshiba
13:12:51amiconnB4gder: The target code is in lcd-16bit.c, and the sim code needs to translate the 16bit packed rgb to whatever is a suitable truecolour format for the os.
13:13:26amiconnIt was rather simple to add support for this to the win32 sim since it uses a DIB for display, which is essentially a BMP in memory
13:14:02amiconnI have no idea how this could work with the delta-scheme and individual pixel drawing in the x11 sim...
13:14:58 Join NicoFR [0] (
13:16:36amiconnpreglow: I agree that an asm core for rockboy would be a good thing. Less complexity than dynarec.
13:16:50_FireFly_amiconn how is the 16bit packed rgb defined ??
13:17:20amiconnBut it isn't as straightforward as it seems - the gameboy cpu is a simplified Z80, but it does a lot of bankswitching
13:17:41amiconn_FireFly_: RGB565
13:17:42_FireFly_maybe i can help to add support for this for the x11-sim
13:18:10_FireFly_ah yes "normal" 16bit format of rgb
13:18:10preglowamiconn: nothing new there, all game systems of that era did a lot of bank switching
13:18:37amiconnYes, but did you already have a look at the gnuboy code?
13:18:45preglow'fraid not...
13:20:34_FireFly_amiconn i have a simple makro which converts 24bit rgb to 16bit rgb on x11 but currently not the other way
13:21:21amiconnThe conversion is trivial; the win32 sim does it too
13:22:07amiconnThe problem is that the x11 sim doesn't use a bitmap for display, and I didn't find out whether x11 can do that
13:22:35amiconnThe current solution uses two buffer, compares the changes, and draws each pixel individually if it changed
13:22:59amiconnProbably slow as hell for truecolour
13:23:21amiconnBesides, I have no idea how to make a window use truecolour
13:23:36amiconn...and then there is the remote lcd simulation, which stays palette based
13:23:46amiconnCurrently both displays share one window
13:23:58_FireFly_amiconn if i get the image data in 24bit rgb then i can generate a ximage or pixmap which can be easily and fast drawn to the window
13:24:25_FireFly_i have written a c++-class which handles this under linux
13:25:37LinusN_FireFly_: it would be great if you fixed the x11 sim
13:27:03 Quit Jungti1234 ("Bye Bye~")
13:27:11amiconnBut please w/o C++
13:27:51_FireFly_the class covert only the c functions to do this :)
13:28:06_FireFly_this class represents one loaded image
13:28:55_FireFly_which member function to draw it on a specified window or generate a 1bit shape mask for transparents
13:30:22_FireFly_amiconn under x11 a window or pixmap are boths drawables
13:30:30_FireFly_so you can direktly draw to the window
13:33:25amiconnThe x11 programming model looks quite cumbersome to me (not only the gfx part). Perhaps that's due to it's networking capabilites.
13:33:42_FireFly_the differenst between ximage and pixmap is the place where the data is stored because the x11-system is a server-client system.
13:33:44 Quit ender` (Read error: 110 (Connection timed out))
13:34:18_FireFly_but you can't compare directly the x11 api with the winapi
13:34:35 Join ender` [0] (i=ychat@
13:34:42_FireFly_because teh x11 api have no widgets as the winapi has
13:42:21 Quit chopped_pork ("leaving")
13:50:12 Quit ashridah ("sleep")
13:56:04 Join chopped_pork [0] (
13:57:34TiMiD_FireFly_: there is only one icon in playlist viewer
13:57:53TiMiDon the left side of the file currently playing
13:59:09TiMiDabout remote buttons, I just forgot them hehe (but browsing the playlist should work at least
14:04:40_FireFly_TiMiD browsing works but you can't left the viewer with the remote :)
14:05:27_FireFly_and i just found out that i can't get into the onplay menu with the remote in my wps-widget :)
14:05:29 Join Febs [0] (
14:06:11TiMiDthis is not a 'critical bug' :P
14:06:43_FireFly_yepp but it would be nice if it gets into cvs my widget :)
14:07:59TiMiDI didn't find a lot of time to read it completely
14:08:28TiMiDand the code is not very clear (since most of the patch is in fact old code :/)
14:09:25TiMiDbut basically, as far as I read, there is one thing that in my opinion could be better
14:09:57TiMiDmake the fn work for only one screen
14:10:07_FireFly_i have currently no idea how to improve the code
14:10:10TiMiDand then call them twice
14:11:08_FireFly_in the current version of update and format_display this isn't possible becuase their return value is checked iirc
14:11:33_FireFly_hmm maybe i find a way to do it
14:12:17TiMiDformat_display is void
14:12:29TiMiDand doesn't do any drawing
14:13:28_FireFly_hmm then i have the wrong function in mind but it's a function which is called in update afaik
14:13:43_FireFly_i have currently no access to my code
14:14:36 Quit Sandking_ ()
14:15:06TiMiDwell since I don't know the original wps code I'm maybe not the most competent here
14:15:19TiMiDjust another thing I just saw
14:15:21TiMiDdisplay->getstringsize(" ", &space_width, &string_height);
14:15:51TiMiDyou can get space_width with display->char_width
14:16:14_FireFly_this is only true when you have a fixed with font using
14:16:47TiMiDwhich font sn't fixed ?
14:16:52TiMiD(I never saw one)
14:18:10_FireFly_hmm thats true currently but with unicode support there could some fonts be variable-widthed e.g. asian fonts
14:18:26_FireFly_or i'm totaly wrong ??
14:18:32B4gderuwe even
14:18:44_FireFly_B4gder ??
14:19:01TiMiDchinese and japanese fonts are fixed on my system
14:19:47TiMiDI don't know about korean or arabic ones
14:20:13LinusNwe have plenty of proportional fonts
14:22:12TiMiDat least it's handled in font.c
14:22:23B4gderuwe_prop for example
14:22:54preglowi wonder if we should try doing some anti-aliased fonts for colour screens
14:22:56LinusNdisplay->char_width is the maximum char width
14:23:02preglowthey look pretty nice
14:23:26LinusNand i doubt that getstringsize(" " ...) wants that
14:24:10TiMiDdepends on what you want to do
14:24:51LinusNin this case the coder wanted the width of the space character, no?
14:28:28TiMiDI can't say, it's not my code ^^
14:28:49TiMiDwell if it works like this it's ok
14:28:51amiconnTiMiD: Some remarks regarding the playlist viewer from my side: (1) The size gain is less than you promised ;-| (2) I see that you changed the 'queued' and 'move' icons into real icons, so they don't need to resemble letters anymore. My idea for 'move' would be a double-arrow (up-down)
14:29:33TiMiDabout the icons, I just choosed quickly since I'm not very good when it comes to design
14:29:52TiMiDabout the code size did you compared the two versions ?
14:29:59TiMiDI recompile it just to see
14:30:28amiconnI did a full rebuild from before your commits, noted the sizes, cvsupped, then rebuilt again
14:30:36TiMiDthe most critical is the Recorder isn't it ?
14:30:49amiconnThe gain on recorder is 1644 bytes
14:31:12TiMiDwell it orth it ;)
14:31:20amiconnyes of course
14:31:30TiMiDit depends of the target
14:31:34amiconnI'm looking forward to further gains :)
14:31:40TiMiDit's "roughtly 2kb"
14:32:35TiMiDnext thing I will do : remove splash calls since they are exacltky the same as gui_syncsplash regarding the parameters
14:33:43TiMiDnext thing I will od : eat first ;)
14:34:57linuxstbTiMiD: Have you thought about the plugins? What happens now when you run a plugin?
14:37:25amiconnThe plugins should handle multi-display themselves, as they see fit
14:37:58amiconnThe only thing needed in the core is clearing both lcds before and after running a plugin, instead of just main
14:38:30LinusNit would be *so* cool to be able to control the music via the remote when running a plugin
14:38:53LinusNand should be pretty simple with the default_handler()
14:38:53chopped_pork(exactly what i was thinking)
14:38:59_FireFly_LinusN then we need separate gui threads
14:39:13LinusNno, i don't mean the wps
14:39:19LinusNjust the buttons
14:39:22amiconnLinusN: Urgh, that would be awkward to implement
14:39:26linuxstbI still want to find the time to look at the plugin-menu system I proposed a while back.
14:40:03_FireFly_LinusN ups missunderstanding :)
14:52:10 Quit B4gder ("time to say moo")
14:53:36elinenbepreglow: anit aliased fonts sound GREAT! :)
14:54:27elinenbeI have a question for anyone who can answer... how are the versions of of rockbox showing the asian fonts? Are they using the unicode patch?
14:54:57markunelinenbe: The chinese patch doesn't apply cleanly right now, the unicode does.
14:55:07markunYou could use either of them
14:58:30markunSome pictures of a naked Gigabeat X30:
14:59:42preglowdamn, 400mhz arm
14:59:45***Saving seen data "./dancer.seen"
14:59:47preglowyou can do some REALLY funky stuff with that
14:59:49 Quit ender` (Read error: 104 (Connection reset by peer))
15:00:20 Join ender` [0] (i=ychat@
15:00:22preglowdoes it have line in?
15:00:34preglowif so, i think i've found the perfect portable effects unit hardware
15:00:35markunpreglow: There are also a 533MHz and a 300MHz version, maybe this is the 300MHz one..
15:01:08markunNo, but you can connect a line-in to the dock connector.
15:02:22markunDo you recognize any of the chips?
15:03:26preglowthere's an usb bridge there, seems
15:04:25linuxstbYes, the Cypress cy7c68300b
15:14:34 Join mashalla [0] (
15:18:06markunThe WM875X.S is a Wolfson DAC
15:20:46linuxstbYes, but is that the correct part number? The ipod has Wolfson dacs with numbers like wm8721, wm8971 and wm8975. The latest ipod uses the wm8758
15:21:30markunI couldn't find this exact partnumber on the wolfson site
15:22:01TiMiDoh about playlist viewer, I have a small question
15:22:19linuxstbIt's probably very similar to the wm8758.
15:22:27TiMiDmy last changes made it use the global setting show_icons
15:22:52TiMiDbut before, there were a separate setting for playlist
15:23:05TiMiDmust I handle that or should ijust follow the global setting ?
15:23:05linuxstbmarkun: Have you decided to buy one, or are you still thinking about it?
15:23:15markunStill thinking
15:23:50markunWell, the X30 is not for sale here, so that gives me some time to think :)
15:25:01linuxstbDoes it have a standard USB connector, or is that via the dock connector as well?
15:25:25markunThere is also a connector on the unit
15:26:18 Join justsomeperson [0] (
15:27:00 Join ep0ch| [0] (n=ep0ch|
15:28:16markunK4S56163PF is 32MB SDRAM
15:29:07amiconnTiMiD: I think this was because the icons in the playlist viewer are somewhat more important than elsewhere, but I have no strong opinion on that. I always had icons enabled.
15:32:42TiMiDI ask this because the gui_list relies partially on the setting show_icons to decide wether it should draw icons or not
15:33:21TiMiDso making it use the other setting would add some small code (well not a lot)
15:35:25_FireFly_maybe you could use a parameter on gui_init to determine if icons should shown or not
15:35:53 Join novimon [0] (
15:36:02novimonhow to enable the crossfeed?
15:36:14_FireFly_it is in playback-settings
15:37:04TiMiDwell I could also just not set the icon callbach in that case
15:39:06TiMiDwell about your patch, I will look more deeply this evening
15:39:16TiMiDand maybe commit it if other devs see no objections :)
15:40:46_FireFly_TiMiD k
15:41:56 Join XavierGr [0] (
15:46:45 Join muesli_- [0] (
15:49:21*LinusN is browsing files on his h300 :-)
15:50:03ep0ch|dare you to play some tunes :p
15:51:37TiMiDLinusN: Congratulations !
15:51:40NicoFRlinus your pgrogress is really impressive :D
15:51:59XavierGrIndeed Gongratulations from me too.
15:52:07NicoFRmaybe a pic ?
15:52:13 Join webguest98 [0] (
15:52:14XavierGrYou did that only in 2 days?
15:52:29muesli_-LinusN 4 president :D
15:52:58_FireFly_TiMiD: on the tracker isn't currently the latest version of the widget
15:53:14muesli_-btw there's no h300 ;)
15:53:16linuxstbI wonder if that news is on misticriver yet...
15:53:26ep0ch|meh where is sourceforge :(
15:53:38TiMiD_FireFly_: upload it for this evening :)
15:54:02 Quit Sando (Read error: 104 (Connection reset by peer))
15:54:07muesli_-linuxstb just done ;9
15:54:26linuxstbmuesli_-: You're getting slower.
15:54:28amiconnLinusN: I'll commit some lcd fixes tonight, e.g. the pb that the default background is rockbox blue, but becomes white after running a plugin
15:54:30_FireFly_TiMiD k
15:54:31TiMiD(well maybe this evening I will be in a pub drinking some beers, so if that the case, I won't look at your patch :D)
15:54:46TiMiDbut if I'm home then ok :)
15:54:49amiconnIt involves 2 new #defines, LCD_DEFAULT_FG and LCD_DEFAULT_BG
15:55:11preglowthat took, what, a couple of minutes?
15:55:32linuxstbamiconn: You will need to be careful because RGB_PACK can't be evaluated at compile time on the ipod. Maybe we should try and fix that.
15:55:54 Join Moos [0] (
15:56:06MoosHi all!
15:56:20amiconnlinuxstb: I know. I'm thinking about using the standard lcd_set_*ground() fns instead of RGB_PACK()
15:56:47amiconnJust slightly more code
15:56:49linuxstbOK, I'm sure you'll find a nice solution. Just wanted to make sure you knew about the issue.
15:57:34amiconnlcd_set_drawinfo(DRMODE_SOLID, LCD_DEFAULT_FG, LCD_DEFAULT_BG); should in fact do the trick
15:58:58linuxstbHow about defining RGB_PACK as a macro for compile-time use, and "rgb_pack" as a function for run-time use.
15:59:06 Quit Strath (Read error: 104 (Connection reset by peer))
16:01:10 Join muesli- [0] (
16:01:35preglowLinusN: how far is the bootloader from being ready for release, then?
16:01:46LinusNnot very far
16:03:12 Join Strath [0] (
16:03:14amiconnlinuxstb: the lcd_set_*ground() fns do exactly what RGB_PACK() does at runtime
16:03:16preglowgood news, good news
16:03:53SlasheriHmm, debug (logf) build fails:
16:03:55Slasheridsp.c:456: error: can't find a register in class `ADDR_REGS' while reloading `asm'
16:03:55linuxstbamiconn: So should we rewrite RGB_PACK so it can always be evaluated at compile-time?
16:04:09amiconnImho we don't need it
16:04:37preglowSlasheri: that's right, i use the frame pointer
16:04:53Slasheriah, ok :)
16:04:54linuxstbamiconn: So colours will always be stored as a "struct rgb" in the source and converted to the 16-bit packed format at runtime?
16:05:36preglowSlasheri: if there is such a thing as a debug define, you could add that to the ifdef line
16:05:49Slasheripreglow: i will now check the one ogg file if it still crashes and send it to you
16:06:27ep0ch|preglow: did you find out why mpc hung?
16:06:34preglowep0ch|: no
16:06:37amiconnlinuxstb: Hmm, when I think of it... Sounds like a waste.
16:06:50amiconnI'll think about it, perhaps there's a better solution
16:07:44linuxstbI was thinking that all colours should (wherever possible) be stored as packed rgb565, instead of struct rgb.
16:08:00amiconnpreglow, Slasheri: I wonder why logf builds don't specify -fomit-frame-pointer
16:08:32Slasheriamiconn: i wonder that as well
16:08:52Slasherii think that option should be added because there is no gdb stub available currently
16:09:02amiconnlinuxstb: Yes, however, that'll change the parameter format for some functions. I'll think about the implications
16:09:31amiconnSlasheri: Yeah, and logf() isn't the option for when you're going to use gdb
16:10:46amiconnlinuxstb: The packed rgb565 should always be big endian, yes?
16:11:39Slasheripreglow: do you want the file now?
16:11:43Slasherii will dcc :)
16:11:45preglowSlasheri: might as well
16:12:11Slasheriit crashes immediately player, cuts hdd power and prints screen full of lines
16:12:23preglownice crash
16:12:29 Join Mongey| [0] (
16:12:34Slasheriah, i have to register yet again to dcc :)
16:12:35preglowyou tried it in the sim?
16:12:38Slasherinot yet
16:12:56Slasherii have some modifications that sim doesn't compile currently.. but i will try soon
16:13:00linuxstbamiconn: Yes. Unless we change the lcd_update() for the ipod to byte-swap.
16:13:20amiconnOkay. I don't see why this shouldn't be possible at build time
16:13:21Mongey|LinusN you rule, nice work on the h340
16:13:38XavierGrHmm interesting I am doing a battery benchmark.
16:14:01linuxstbI don't like using htobe16() for it though - I think we should make the code clearly by having #if CONFIG_LCD==LCD_IPOD_4G || CONFIG_LCD==LCD_IPOD_NANO and using swap16
16:14:27XavierGrI started 5.30 yesterday it is 5.15 and the player says 3h 19m left - 14 % level
16:14:43XavierGr128kbps mp3 50 volume
16:14:45linuxstbAnd we would replace the swap16 function by a macro.
16:14:53amiconnI think about using a similar mechanism as for the bmp header for screendump()
16:15:02amiconnThis is always little endian
16:15:19linuxstbI've discovered that we could write rgb565 bmp files - at the moment I (poorly) convert it to 24-bit.
16:15:39XavierGrit exided 23h 45m that the first estimated time was.
16:15:40preglowXavierGr: well, that's not bad, is it?
16:15:51preglowXavierGr: the estimated time is pure guesswork
16:15:57XavierGr1900mAh battery, no misunderstandings
16:16:06XavierGryes it seems very good.
16:16:06preglowXavierGr: what capacity is the default one?
16:16:18preglowwell, sounds decent to me
16:16:20amiconnlinuxstb: Yes we can, and if you look at the bmp2rb source, you'll see that I know about this format
16:16:41XavierGrit has some juice left and I want to see how far it can get.
16:17:07XavierGrthough I think that I set the volume too low
16:17:15XavierGris 50 compared to 20 of iriver fw?
16:17:17 Quit _FireFly_ (Read error: 110 (Connection timed out))
16:17:37SlasheriXavierGr: no, the scale is non-linear
16:17:47XavierGrahh then bummer
16:17:58XavierGrthis is then below standarts
16:18:27 Part Mongey|
16:18:27XavierGrI usually set the volume more than 70-75 to listen to music
16:18:34 Quit muesli_- (Read error: 110 (Connection timed out))
16:18:37preglowi honestly don't think the volume will affect things much
16:18:40XavierGrand more than 85 if it is crowded
16:18:45XavierGryou do?
16:18:55XavierGrbecause some say it does though I am not sure.
16:18:55 Join ehntoo [0] (
16:18:57preglownot much, no
16:19:39XavierGralso the album is 54 MB long is it right for a benchmark?
16:19:46Slasherii think bass might be more important considering power consumptions
16:19:57XavierGrno flat EQ, always flat
16:20:25Mordovhey, I*m looking for a iHp-120 simulator so I can easyer make WPS files.. is there any?
16:20:48linuxstbMordov: Yes, but you will need to compile it yourself.
16:20:58XavierGralso I really want to see the benchmark battery data from my plugin, which is running simultanesly
16:21:11XavierGrI want to see the graphs
16:21:28Mordovhmmmm, linuxstb can't I copie it from someone?
16:21:34 Join kaouete [0] (
16:21:45MordovI'm no programer... wps are difficult for me ;)
16:22:43linuxstbMordov: Yes, if you can find someone to give you a copy. But there are no official versions distributed.
16:23:15Mordovah, ok, so I'll just write a post in the forum and hope :)
16:23:33Mordovgot tired of mounting the player :)
16:23:39kaouetei have a strange thing with my archos recorder with rockbox 2.5 : sometimes, (when the battery charge is under 15% sometimes, but this morning it was 30% !) the songs playing just stop, and even when no songs are played i get the message "press stop to shutdown" : it looks like someone press the stop button in fact, but i'm not :]
16:23:47 Join _FireFly_ [0] (
16:23:52 Join jelle-k [0] (
16:24:01kaoueteis it a know bug ? is it even a bug ?
16:24:01jelle-khi all
16:25:23XavierGralso did anyone tried or checked my battery benchmark plugin? A benchmark on 128kbps mp3 at 50 volume would be good to be done (with or without the plugin) to check if the results match. Also I have to do now the same test on the iriver fw.
16:28:14LinusNwee, the remote lcd driver just worked without modification on the h300
16:28:17 Quit muesli- (Read error: 110 (Connection timed out))
16:28:56LinusNkaouete: did it do this with 2.4 as well?
16:28:58preglowmarkun: i can speed up crossfeed a bit more by using interleaved delay buffer data
16:29:07chopped_porkXavierGr: i could do some tests as i already did in the past... i'd do flac and vorbis though
16:29:14chopped_porkXavierGr: how do i get/use your plugin?
16:29:23kaoueteLinusN: i dont think but i'm not sure : all that i am sure is that before (maybe 2.3) it was not doing this
16:29:32preglowchopped_pork: flac has already been done
16:29:37XavierGrchopped_pork: do you know how to compile?
16:29:49markunpreglow: yes that would speed it up. I had only thought of interleaved input, but that is not an option.
16:29:51chopped_porkXavierGr: i will be setting up the crosscompiler today
16:29:59chopped_porkXavierGr: i do know how to compile :)
16:30:04XavierGrelse I could you a daily biuld with it
16:30:26chopped_porki really really want to see unicode in action :)
16:30:40linuxstbXavierGr: Do you want to update this page with info about your plugin:
16:30:42preglowmarkun: i can speed up the fetch/store and free a register like that
16:30:45markunchopped_pork: we all do :)
16:31:18XavierGrchopped_pork: please tell me if you want to use the patch or I could sent you the fw file with the plugin
16:31:43XavierGrlinuxstb: I will go and update it, as soon as I see the battery log, just to be sure that all is fine with the plugin
16:31:57chopped_porkhm i could try musepack too, since it's slow it'll drain the battery fast anyway :)
16:32:00linuxstbI've just updated the IriverRuntime to add that I had the volume at 80 when doing my tests.
16:33:08XavierGrBy the way how the battery ETA is calculated?
16:34:07chopped_porkoh boy i remember back in the day, it was some week or two after there actually was flac playback in rockbox i did a test and the runtime was some 8 hours afair
16:34:43chopped_porkwhich was already way cool
16:34:50linuxstbchopped_pork: That's interesting. I would have expected the FLAC codec in Rockbox to have progressed more than that.
16:35:34linuxstbBut I guess the main battery drain is the disk, and disk usage hasn't changed much since the first playback code.
16:36:00chopped_porklinuxstb: i would have never expected more, because of disk usage
16:36:08chopped_porkyou just pwn
16:38:53SlasheriHmm, the disk eats battery quite much. It takes ~500-600 mA at buffering, so after 5-6h of flac playback 600 mAh of power would have been consumed at least
16:39:02Slasheriplus the cpu current
16:42:55linuxstbI've still got a couple more tests I want to do with FLAC - a test with the normal CPU frequency set to 24MHz, and a test with Rockbox sitting in the file browser during playback, instead of having a WPS (with peakmeter) displayed.
16:43:26 Join tucoz [0] (
16:44:06linuxstbWavpack will be an interesting test - it uses slightly more CPU than FLAC, but the files are slightly smaller.
16:44:39XavierGrlinuxstb: are interested in using my plugin?
16:44:47tucozIn the cross compiler howto it is written that Rockbox does not currently work with GCC 4.0.x. Is this still true?
16:44:54tucozThat is in the wiki
16:45:03chopped_porkshould i use binutils 2.16 or 2.16.1?
16:45:09Slasheritucoz: i would not recommend gcc 4 yet
16:45:39tucozSlasheri, ok. Then it is best to keep that there.
16:45:44Slasheriit should work, but might have some problems
16:47:44linuxstbXavierGr: After you tell me that it doesn't decrease performance :)
16:50:15tucozchopped_pork, I think 2.16 is good enough
16:50:28XavierGrok I will try to make the same test without the plugin
16:50:35chopped_porktucoz: k thanks
16:52:04tucozI cannot find a changelog that states what has been changed in 2.16.1
16:52:13tucozBut, 2.16 works fine
16:52:34 Quit webguest98 ("CGI:IRC (EOF)")
16:54:07 Join lamed [0] (
16:55:02 Quit justsomeperson ("CGI:IRC")
16:56:25 Join webguest47 [0] (
16:56:33preglowtucoz: gcc4 seems to works just fine
16:56:50preglowtucoz: i ran a gcc4 compiled build for days
16:58:10lamedanyone knows what's a "C" means at the start of a line from "cvs update -dP"
16:58:13lamedi'm getting:
16:58:16lamedC apps/gui/list.c
16:58:20linuxstbC is bad :)
16:58:31preglowlinuxstb: conflict :/
16:58:32tucozpreglow, I see. Well, I can change that statement to "seems to work just fine" instead of "does not currently work" ;) Or what do you say?
16:58:34lamedu mean it didn't updated the file..
16:58:35preglowlamed: conflict :/
16:58:37lamedright conflict.
16:58:52preglowtucoz: seems to work just fine, yes
16:59:15linuxstbYes, it updated the file, but there will be parts of it marked with "<<<<<<<<<<" - you need to manually edit and resolve the conflicts.
16:59:37linuxstbOr if you want to discard your local changes, just delete the file and do update again.
16:59:46***Saving seen data "./dancer.seen"
17:00:36lamedyes, here it is...
17:02:42lamedwoop, yeah, i've got totally to different files there :/ but good thing that the gui widget is getting worked on. have a gday!
17:03:03_FireFly_lamed: which gui widget ??
17:03:11tucozpreglow, this is only for coldfire, right?
17:03:26lamedthe gui_list
17:03:30preglowtucoz: yes
17:03:35preglowtucoz: wont work at all for sh
17:04:48 Quit lamed ("CGI:IRC")
17:06:23_FireFly_TiMiD: i'm going to update my widget on tracker
17:07:27_FireFly_argh sf is slow
17:08:49_FireFly_ok updated
17:11:47 Join hardeep [0] (
17:21:00 Join Sando [0] (n=lolsteam@
17:22:13 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
17:23:29XavierGrend of benchmark: Total runtime 24h 52 minutes (wtf lets say 25h) 1900mAh battery mp3 128kbps EQ flat, 50 Volume
17:24:05ep0ch|was that buffering from disk too?
17:24:43XavierGrhmm just checked the log, last successful right was at 24h 22minutes
17:26:24 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:26:53_FireFly_TiMiD: currently here ??
17:27:04XavierGrthat was the result from the builtin timer in rockbox 24h 22m
17:27:48XavierGrhere is the log
17:27:54XavierGrbattery log
17:29:17XavierGrhmm 51 succesfull right on the disk.
17:30:31XavierGrOne write every 28,6 minutes
17:30:49XavierGramazing more or less like the buffer size...
17:31:35ep0ch|yeah i think 1mb of 128kbps is about a minute
17:31:46XavierGryes exactly
17:31:56 Quit webguest47 ("CGI:IRC (EOF)")
17:33:17XavierGralso when the battery reached 0% or ETA 0 minutes there was no successfull HD write.
17:34:06XavierGrSo the error on the benchmark was exactly as it should. 30 minutes!!!
17:34:14XavierGrfor mp3 128kbps
17:34:55XavierGrvery accurate results :D
17:35:58XavierGrthe disk managed to make 51st write. That was logged. Then the player continued for another ~30 minutes and it stutdonw on the 52 HD access which failed. (and the log couldn't be updated)
17:35:59 Join Kohlrabi [0] (
17:36:42XavierGrSo the error of the mesurement comes from the bitrate of the files played.
17:37:17XavierGrAs bitrate gets bigger the error is smaller.
17:38:09XavierGrthen the error will be at about 1hour incase 64kbps mp3 are tested
17:39:59XavierGralso the voltage dropped till 3.46. Strange last time I pushed the player to 2.73. But then it shut in front of my eyes. So I think that 3.46 was the voltage 30 minutes before shutdown.
17:40:20XavierGrnow let's make some graphs
17:44:03tucozXavierGr, you should probably change the decimal separator to from , to .
17:44:07tucozin the voltage
17:44:18XavierGryes, sorry localization issue
17:44:50tucozIt is the same here
17:45:49tucozNorway at the moment. It is the same in Sweden (where I am from), as well.
17:46:10XavierGrI will search in which countries it is a '.'
17:46:20XavierGrI think UK and US for sure
17:46:47SlasheriHmm, probably it's almost alwhere in europe , (at least here in finland). But i use . always instead, because for example programming languages don't like ,
17:47:42chopped_porkit's , in poland too
17:47:59tucozSlasheri, and it isn't very good to use in maths as well. For instance to describe a set of floats (3,14, 2,618, 1,14)
17:48:06chopped_porki like dots better anyway
17:48:16Slasheritucoz: yes, true
17:48:31XavierGrwell I use dots to describe thousands
17:48:31tucozwonder why they settled for the ,
17:48:33Slasherior one could use 1,23 ; 1,50 but that is confusing too
17:48:48_FireFly_XavierGr: you should use a seperator which will be correctly understand in a spreadsheet app
17:48:55linuxstbOr change the units - 1230mV
17:49:02_FireFly_or that
17:49:16XavierGryes probably mV is better.
17:49:24linuxstbWhy solve a problem when you can avoid it?
17:49:27amiconn_FireFly_: Yes, and e.g. OpenOffice wants , when set to german, but . when set to english
17:49:39XavierGrmy spreadsheet wants ','
17:49:40tucozlinuxstb, KISS
17:50:02_FireFly_amiconn: also when you import a csv file ??
17:50:18amiconnyes, afaik
17:50:38_FireFly_if so then avoid a seperator :)
17:51:08tucozSeems to drop quite quickly when reaching 3.7 volts
17:52:12_FireFly_yepp it is always when going to "empty"
17:53:04_FireFly_a good battery should hold the voltage as long as it could and then rapitly drop to "empty-voltage"
17:53:16tucozI see
17:53:43tucozsee you
17:53:46 Part tucoz ("Leaving")
17:55:08 Join tucoz [0] (
17:56:14tucozI just remembered. Do any of you know why the iriver is silent at 96Mhz (radio). A user in the forums is saying that he listens to a station at that frequency on other devices.
17:56:27wubblais it possible to compile rockbox using the debian sarge package: prc-tools-m68k
17:56:51_FireFly_amiconn: you are right i have just tested it
17:57:59XavierGrHmm the ETA calculation I think can be better (and the % level). If the battery drops so quickly from 3,46.
17:58:39XavierGrBecause while the ETA measurement at start is very accurate then, at about the end of the battery it is completely wrong
17:58:56Mordovwhere do I report bugs?
17:59:03XavierGrBut I don't know how this caclulation is done, so if anyone has info...
18:00:26tucozMordov, You could just mention your bugs here, and someone will pick it up
18:00:46 Part tucoz ("Leaving")
18:02:54 Quit Febs ("CGI:IRC (EOF)")
18:03:20Mordovbug report: I deleted some files in teh Tag mode yesterday, tehy didn't get delete but just alterd. this caused the player to lock up when browsing and I could not uppdate my DB file bacause when getting to the songs that was alted I got a java out of heap memmory bug. I deleted the files manualy using windows exploer and all was fine after that. ONe schoud not be alowed to delete files in Tag mode if it doen's work ;)
18:03:53HClwhat do you mean delete files in tag mode?
18:05:20MordovI can delte files... I press teh joystick and get a delte option
18:05:39MordovI can delete something... oviusly not the file :)
18:05:54HClas far as i know that should just delete the file
18:05:59HClshouldn't affect the database
18:06:11Mordovyapp, but it doen't
18:06:16Mordovthe file is still there
18:06:23MordovI havn't tested that today
18:06:28MordovI'll cherck noe
18:06:33Mordov-e +w
18:06:57HCli never really liked the way rockbox browses the database, or the way the database uses raw offsets, in the first place
18:07:03HClits extremely unclean and unmanagable
18:07:15_FireFly_HCL write a new one ;)
18:07:27_FireFly_if you can
18:07:30HCli *wrote* the current design
18:07:39HCli wanted to make it more abstract
18:07:45HClbut there was heavy resistance from the other developers
18:07:49HClso i had to compromise
18:08:24Mordovtoday I can't delete in tag mod :) only playlist and open with visible :)
18:08:26HClthe offsets aren't that much of a disaster, aside from changing small bits of the database
18:08:49HClthe gorey code is in the database browsing, which i didn't touch much aside from updating it to work with the new format
18:09:06Mordovmy biggest irritation is that the Tag browsing changing is a bit slow
18:09:15XavierGroverall I am very glad with the benchmark results. This evening I will make the same test without the plugin, to see if it matters. Then I will update the wiki.
18:09:15HClyea, thats on a todo list somewhere
18:09:28HClits mostly cause of the gorey design of the database browsing
18:09:32HClthat its hard to get the way we want
18:09:42HCl(extra folder like thing in the root directory that allows you to browse the database)
18:10:01Mordov:D think it would make the whole firmware more stady,,, I have had some hangs during browsing
18:10:21HClthe problem of that mostly lies with the gorey design of the database.
18:10:40HClits a bit of a trade off between memory+cleanness and performance.
18:10:45HClmost developers prefer performance
18:11:09HCloh wait, its cleanness and performance+memory
18:11:18HClmost developers prefer perfomance/memory
18:11:33Mordovwe al do :)
18:11:36HClwhere i'd rather have it work clean and right,more slowly, than quickly and dirty :)
18:11:41MordovI know where the reset butin is ;)
18:12:23MordovIf I coud chose between faster changing betwene tag's in differnt foldewrs and more stady system I would go for the first
18:12:42HClyea, thats what was decided on
18:12:42Mordovbut not in any extreme ways...
18:12:55HClbut an effect of it is that the code is unmanagable, and glitchy
18:13:17Mordovhehe, taht is not good :)
18:13:29Mordovunmanagable, and glitchy sounded frustrating for you :)
18:13:41HClyea, one of the reasons why i practically refused to touch that part of the database
18:13:47HCli only updated it to work with the new format.
18:16:56HClmy cat is so cute
18:18:23 Join cora61 [0] (
18:18:27XavierGrmeaow! I love cats.
18:19:29HCl :)
18:20:20 Part cora61
18:27:44 Quit mashalla (Read error: 110 (Connection timed out))
18:44:22 Quit NicoFR ()
18:45:19 Join ehntoo_ [0] (
18:50:32 Nick Seedy is now known as Seed (
18:56:41 Quit t0mas (Read error: 113 (No route to host))
18:59:13 Quit ep0ch| ("gone")
18:59:50***Saving seen data "./dancer.seen"
19:00:39 Quit ehntoo (Read error: 110 (Connection timed out))
19:00:44 Join justsomeperson [0] (
19:01:15 Quit justsomeperson (Client Quit)
19:06:42 Quit hardeep (" HydraIRC -> <- IRC with a difference")
19:16:42 Join gromit` [0] (
19:22:35 Join pinkutank [0] (n=ddd@
19:22:53 Join einhirn [0] (
19:23:13 Quit einhirn (Client Quit)
19:24:59pinkutankhey ho guys
19:25:40pinkutankI've compiled a jpeg viewer that allows both scrolling and skipping to a enxt/prev file and allows you to listen to music while doing so
19:30:02wubblais there a difference between m68k-linux-gcc and m68-elf-gcc?
19:30:37Maxime``probably, as they haven't the same name :s
19:31:34wubblaand what is the difference? :D
19:32:10Maxime``dunno ^^ sry :x
19:35:47wubblak ;-)
19:37:38pinkutankwhere is the spalsh image in the source to edit?
19:38:09 Join Sandoaway [0] (n=lolsteam@
19:45:38 Join Jungti1234 [0] (n=jungti12@
19:46:08Jungti1234markun: are you here?
19:50:25amiconnlinuxstb: In fact your idea to always use packed rgb colour values solves a fair number of problems, e.g. with the multiscreen code
19:51:03amiconnThis way the data format for a pixel value is identical for both colour and greyscale lcds
19:51:19amiconn(unsigned int)
19:51:54 Join muesli_- [0] (
19:52:38 Join |beowulf| [0] (
19:56:43 Quit Sando (Read error: 110 (Connection timed out))
20:01:09 Quit muesli_- (Read error: 104 (Connection reset by peer))
20:03:00 Join Sandking [0] (
20:07:38pinkutankwhich file do i need to chanfe to change the spalsh image?
20:07:57pinkutankwhile compiling
20:09:20 Quit Jungti1234 ("Bye Bye~")
20:09:39 Join shx [0] (
20:13:03 Quit gromit` ("Leaving")
20:13:16TiMiDI've a question
20:13:44TiMiDhow do you make scanf read strings with a white space as one single string instead of stopping the reading
20:14:08 Quit shx (Client Quit)
20:15:00TiMiDfor example I have a string "foo1 foo2"
20:15:17TiMiDand scanf first reads "foo1" then "foo2"
20:17:06 Quit Moos ("Glory to Rockbox")
20:19:23Mordovcan I do anything to make the ramote work better on my iHp-120?
20:20:20 Join DangerousDan [0] (
20:21:19 Join muesli_- [0] (
20:22:04_FireFly_TiMiD: updated version of my widget is on tracker (if you have some free time :) )
20:31:46_FireFly_TiMiD: it seams it isn't possible (scanf) look at man scanf
20:32:32_FireFly_because scanf returns on each white space it finds in the input-stream (stdin)
20:38:11pinkutankhey timid
20:38:33pinkutankive succesfully compiled a build
20:39:07pinkutankand not stopping there, I combined to plugins that otherwise interfered with each other, and released a combo patch to the website :)
20:41:37TiMiD_FireFly_: no party this evening, but I must help my sister with her C programming project so maybe later :)
20:41:55_FireFly_k, ;)
20:42:05TiMiDpinkutank: so you have the plugin patched as you wanted ?
20:42:09TiMiDgreat :)
20:46:16pinkutankyes, I even have it up on the forum if anyone wants it
20:46:37pinkutankwhich file do i need to edit so I can change the splash logo?
20:49:13pinkutankapps/gui/icon.c ?
20:51:45pinkutankdo i extract them via a decompiler?
20:51:50TiMiDLinusN: I think the dir recorder / player are quite confusing
20:51:57LinusNwhat is it that you want to do?
20:52:07LinusNTiMiD: of course it is
20:52:12pinkutankchange the graphics that the palyer uses, currently aimin at tthe rockbox logo
20:52:13TiMiDit could be renamed to bitmap/charcell ?
20:52:25amiconnLinusN: What do you think is better: RGB_PACK() or LCD_RGBPACK() ?
20:52:26TiMiD(I know it's for historical reasons)
20:52:27LinusNTiMiD: not without causing huge problems
20:52:44LinusNamiconn: it doesn't matter to me
20:52:55LinusNTiMiD: cvs doesn't have a rename function
20:53:09pinkutankwhat kind of unpacker should i use
20:53:19TiMiDand a delete ?
20:53:29TiMiD(I assume delete is logged too)
20:53:43LinusNTiMiD: the files will lose all cvs history
20:54:32 Join RotAtoR [0] (
20:54:48 Join Mordov_ [0] (
20:55:02Mordov_is rockbox for iPod nano out?
20:55:12_FireFly_with svn it is possible to rename (move) a file or directory ;)
20:55:42 Join linuxstb_ [0] (
20:56:01pinkutankhow can I create bitmaps from the resoruce files
20:56:34linuxstb_Mordov_: No. There is a working bootloader, but Rockbox itself isn't working yet.
20:56:49LinusNpinkutank: create a monochrome bmp file and convert it using tools/bmp2rb
20:57:07LinusN_FireFly_: yes
20:57:23amiconn't need to be monochrome
20:57:31 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
20:57:36LinusNtrue, but then you don't know how it will look
20:57:50Mordov_linuxstb_: ok, my frined just bought nano and have been bitching abaout the firmware and soware for an houer :)
20:57:57amiconnLinusN: It will map the brightness levels to the closest possible ones
20:58:02Mordov_one haw to have iphoto to make photo work
20:58:03pinkutankis there a way to dump the ones it has in it so i can see them
20:58:21amiconnIn fact the .bmp for h1x0 should have more than 2 colours
20:58:29LinusNamiconn: true
20:59:00amiconnIt's no more the old bmp2rb where you never knew how it comes out ;)
20:59:36amiconnpinkutank: Just do a screendump (possible on target and simulator)
20:59:51***Saving seen data "./dancer.seen"
21:00:16 Quit Mordov (Read error: 110 (Connection timed out))
21:00:25pinkutankdoeswsnt screedump dump the hole screen
21:00:35pinkutanki think the rockbox site had a wiki page
21:00:38SlasheriHehe, i got almost zero-latency wps working
21:00:59SlasheriNow it will display track title (parsed from filename) immediately without delays
21:01:07Slasheriwhen skipping tracks fast
21:01:30Slasherii will commit later after i have checked for a few possible problems
21:01:35_FireFly_how ??
21:01:53t0masfrom filename
21:01:58linuxstb_Slasheri: Do your changes help with the ALAC/AAC .m4a problem?
21:01:59t0masso not the id3-tag I assume?
21:02:00Slasheri_FireFly_: just updating playlist position immediately and letting audio catch later
21:02:11Slasherilinuxstb_: unfortunately no :/
21:02:17Slasherit0mas: yep, no
21:02:32Slasheriwe generate a temporary id3-entry from fly until real entry is generated
21:02:46linuxstb_Also, do you check the return value of get_metadata() ? If it returns false, the file is unplayable.
21:02:46t0maswe can also add that to the dircache?
21:03:05t0masit would make the cache bigger...
21:03:13Slasherimaybe, but that will be more complex
21:03:17t0masso it should be an option... but I think some people might like it
21:03:19Slasheriyes, much bigger
21:03:46Slasherii keep that in mind too :)
21:05:09pinkutankwhich file has the splash logo itself
21:05:30 Quit chopped_pork ("leaving")
21:06:58pinkutankconst unsigned char rockbox112x37[]={
21:07:01pinkutankthis ?
21:07:23LinusNthat's the archos version
21:07:59pinkutankconst unsigned char rockbox160x53x2[]
21:08:26LinusNyes, that's the iriver h120 version
21:08:35_FireFly_for the main display
21:15:26 Join MrStaticVoid [0] (
21:18:04 Quit ender` (Read error: 104 (Connection reset by peer))
21:22:30SlasheriHmm, the playlist needs some modification too to support dircache. Then it shouldn't need to save the current playlist on disk
21:23:47wubblasomething is weird
21:24:00wubblai cannot compile anything in tools/*
21:24:06wubblausing the rockbox devkit on win32
21:24:13Slasheriwhat is the problem?
21:24:46wubblai'm getting an 16bit-MS-DOS-error
21:25:15wubblasome ugly pop-up appears...
21:25:30Slasheriah, that's windows :/
21:26:19Mordov_could I do something to make my ramote better?
21:26:20Slasheritry to find out which of the programs crashes.. make, m68k-elf-gcc etc
21:26:44wubblasomething about the NTVDM-CPU detecting a bad command (i've no idea what it's called in english, as my system's language is set to german)
21:26:48wubblagcc crashes
21:27:06wubblai manually tried to "gcc -o iriver.o iriver.c iriver.h"
21:27:08SlasheriMordov_: if you want a wps on the remote, _FireFly_ might have a patch for you
21:27:12wubblaand that crashed...
21:27:20Slasherioh, something is really wrong then
21:27:21Mordov_Slasheri :) thanks
21:27:34Slasheriwubbla: hmm, you should not give header files for gcc
21:27:51 Quit |beowulf| ("KVIrc 3.2.0 'Realia'")
21:27:53Slasherijust try "make"
21:28:18wubblajust a second
21:28:34wubbla/cygdrive/c/Rockbox/rockbox/tools>gcc -o iriver.o iriver.c
21:28:34wubbla/usr/lib/gcc-lib/i686-pc-cygwin/3.3.1/../../../../i686-pc-cygwin/bin/ld: install
21:28:37wubblaation problem, cannot exec `/usr/lib/gcc-lib/i686-pc-cygwin/3.3.1/../../../../i6
21:28:40wubbla86-pc-cygwin/bin/ld': Invalid argument
21:28:42wubblacollect2: ld returned 255 exit status
21:32:44 Quit Sandking ()
21:33:21 Join ender` [0] (i=ychat@
21:34:05pinkutanki used bmp2rb
21:34:10pinkutankand converted a monochrome bmp
21:34:17pinkutankto 4 level h1x0
21:34:28pinkutankbut it appears scrambled in the player
21:34:57pinkutankI copied the text from a cmnd window and pasted it in the icons file by a hex editor, anything wrong there?
21:35:26_FireFly_hex-editor ??
21:35:41pinkutanki use a hex editor instead of notepad, easier to keep track of line #
21:40:23Slasherieh.. you should use a real programming editor, that undestands the language you are working with :) (sorry, but i can't recommend anything for windows users)
21:40:53*amiconn recommends ConTEXT
21:41:00_FireFly_jedit notepad+ ConText
21:41:22pinkutanki use ultraedit -32
21:41:35dpassen1i also use ultraedit
21:41:51linuxstb_pinkutank: Is your replacement bitmap the same size as the original?
21:42:23pinkutankshould be
21:42:25pinkutanklets see
21:42:45pinkutank160x 53
21:42:48 Join petur [0] (
21:45:24pinkutank 0x00, 0x00, 0x00, 0x30, 0x30, 0xff, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30,
21:45:24pinkutank 0x30, 0x30, 0x30, 0x30, 0x30, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
21:45:25pinkutank 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0xcc,
21:45:25DBUGEnqueued KICK pinkutank
21:45:25pinkutank 0xcc, 0x0c, 0x30, 0x30, 0xc0, 0xc0, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
21:45:25pinkutank 0x00, 0x00, 0x00, 0x00, 0x00, 0xc0, 0x30, 0x30, 0x30, 0x0c, 0xcc, 0xcc, 0xcc,
21:45:25***Alert Mode level 1
21:45:25pinkutank 0x30, 0x30, 0x30, 0xc0, 0xc0, 0xc0, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30,
21:45:27pinkutank 0x30, 0xff, 0x30, 0x00, 0x00, 0x00, 0x00, 0x00, 0x30, 0x30, 0x30, 0x30, 0x30,
21:45:29pinkutank 0x30, 0x30, 0xf0, 0x3c, 0x33, 0x30, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
21:45:31pinkutank 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
21:45:32 Quit pinkutank (Excess Flood)
21:45:44 Join xmixahlx [0] (n=xmixahlx@
21:46:52 Join pinkutank [0] (n=ddd@
21:46:58pinkutankI got the boot
21:47:51 Join NicoFR [0] (
21:48:56pinkutankis there a way for bmp2rb
21:49:05pinkutankto save instead of output text
21:50:27linuxstb_Just redirect the output. i.e. type "bmp2rb-f 2 bitmap.bmp > outputfile.c"
21:50:28LinusNredirect the output
21:55:26***Alert Mode OFF
21:57:03 Quit MrStaticVoid (Read error: 110 (Connection timed out))
21:58:05 Join MrStaticVoid [0] (
22:01:03 Quit jelle-k (Read error: 104 (Connection reset by peer))
22:01:05 Quit Kohlrabi (Nick collision from services.)
22:01:11 Join Kohlriba [0] (
22:02:52 Quit muesli_- ("ich will Kühe!!!")
22:11:04 Join hardeep [0] (i=hardeeps@norge.freeshell.ORG)
22:15:00 Join mashalla [0] (
22:15:42 Join Cassandra [0] (
22:16:53pinkutankit worked that way
22:17:08pinkutanknow a monkey greets me
22:17:24pinkutankis there a way to create a bigger splash image?
22:18:00_FireFly_for what ??
22:18:11 Quit Cassandra (Client Quit)
22:18:18 Join Cassandra [0] (
22:21:47pinkutankh120, rockbox
22:22:07pinkutankor do i have to use the default size, there is a lto of space on the screen
22:22:37 Join Wonderboy [0] (n=Test@
22:22:43 Nick Wonderboy is now known as San (n=Test@
22:24:03 Join yngwi [0] (
22:28:33Sanhow is the H300 port coming along
22:28:56Cassandra_FireFly_, you've got some C++ style comments in your WPS widget.
22:29:15_FireFly_could be
22:30:22CassandraPlease have a read of that and amend your code as appropriate. Thanks.
22:32:25CassandraApart from that, your patch scares me. It is big and complex. :)
22:34:06_FireFly_currently i have only made a multi-screen interface to the old wps-code and needed modification
22:34:23_FireFly_in the code so that it works with the interface
22:36:10 Quit MrStaticVoid ("leaving")
22:36:15Cassandra*nods* I'm not sure I feel competant to review it. :(
22:37:29CassandraI may adopt the "suck it and see approach" - compile for Archos and see if anything breaks.
22:38:07CassandraCan't see anything obviously wrong with your coding style. Lots of lovely comments. (We like comments. Rockbox needs more comments.0
22:38:26_FireFly_it shouldn't break anything on other targets but who knows we are only humans :)
22:38:27 Join ^BeN^ [0] (
22:38:33Sanhas LinusN got the browser on the H300 working?
22:39:04CassandraDid he? News to me.
22:39:16Sangood work man
22:39:26NicoFRthis is real good :D
22:39:30Sanand the remote needs no porting I heard?
22:39:38NicoFRdo you happen to have pics ?
22:39:52_FireFly_Cassandra: updated
22:40:09Sanmade any more progress linusN?
22:40:14Sanor working on anything else?
22:40:16CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:40:16*amiconn hopes his LCD changes do work on the colour targets
22:40:22amiconnThey do work on the sim
22:40:23CassandraLinusN, any opinions on the wps-widget patch. I'm tempted to apply it and see if it works for Archos, and commit if it does.
22:40:29^BeN^LOL=\, no anyone from rockbox crew here?
22:40:37LinusNno general opinions, no
22:40:41CassandraSame goes for you amiconn.
22:41:10LinusN^BeN^: what a funny question
22:41:10CassandraSo this is not a bad plan then?
22:41:22LinusNi don't think it is
22:41:29CassandraNo, none of us have anything to do with Rockbox.
22:41:47^BeN^mmm sorry but i dont see any op
22:41:57CassandraWe just hang out on #rockbox because, erm, ......
22:42:02Mode"#rockbox +o LinusN " by ChanServ (ChanServ@services.)
22:42:03CassandraWe're French.
22:42:09^BeN^ok dude
22:42:09CassandraYes, that's the reason.
22:42:21_FireFly_Cassandra: the only thing which could break a build on other targets could my fix for the not get cleared images on wps
22:42:25NicoFRi'm french... is that a problem ? ^^
22:42:34t0masit is
22:42:40t0masif you try to speak that language...
22:42:43t0masI'll freak out
22:42:45CassandraOnly if you think it is.
22:42:59NicoFRI can do more than try
22:43:06SanBTW, that image of the H300 with the rockbox logo, was that real? How did you get it up? I notice a ribbon wire out the side. Is that for testing?
22:43:12t0masI had a 7,9 out ot 10 on my last test... but still I don't understand a thing in French :)
22:43:44NicoFRit can't be that bad
22:43:51pinkutankdid preglow spurt the eq yet?
22:45:02LinusNSan: that's the BDM connection
22:45:13LinusNand yes, that was real
22:45:28Sannice :D
22:45:35Sanso, port coming soon?
22:45:53Cassandra_FireFly_, patches cleanly, anyway
22:46:04NicoFRI was wondering... once the bootloader is done, how much more work will be required to have rockbox functionnal on the h300 ?
22:46:15NicoFRI mean same level as h100
22:46:44_FireFly_Cassandra: the build on non lcd-bitmap target will fail
22:46:45t0masnot to much
22:46:47 Quit Kohlriba (Read error: 104 (Connection reset by peer))
22:46:54t0masNicoFR: a lot of hardware is the same as the H1xx
22:47:14t0masso if the LCD is working... and the bootloader is working, we won't have to wait long for audio playback
22:47:16_FireFly_because i have forgotten to ifdefout my fix for non lcd-bitmap-targets
22:47:20XavierGrok second battery test started at 11:45 p.m local time
22:47:30Cassandra_Firefly_, just spotted that, yes.
22:47:35_FireFly_i will update the patch on the tracker
22:47:39XavierGr(without my plugin lets see if it shows something different)
22:47:45CassandraOK. I'll revert in the mean time
22:48:22pinkutankXavierGr, I combined your jpeg patch with someone elses no we all can have non interrupting file scrolling jpeg viewers, I posted it on the site, giving credit to you, I hope you dont mind
22:48:26CassandraYou might want to check for a rogue comment in dbtree.c before line 167 as well.
22:48:58XavierGrI hope you gave credit ot the other guy too.
22:49:14pinkutankof course i did
22:49:35XavierGrand no I don't mind. I am happy to see a little of my work posted for others. Thanks
22:49:54 Join Kohlrabi [0] (
22:50:08XavierGrSometime I will look at it, maybe I will try and make it even better. My patch has space for optimizations.
22:51:20_FireFly_hmm my search for c++-style comment found many files
22:52:07Bagdersloppy reviews let them through
22:52:12pinkutanknormally, the two patches overlapped, giving errors, so I merged them, referring to 4 different jpeg.cs to check, because I dont know c++
22:52:29pinkutankanyway, thanks a lot
22:52:50_FireFly_Bagder: these comments are mostly in codecs-sources
22:52:54Cassandra_FireFly_, and the culprit of every single one of them should be roundly spanked for not reading the coding style. :)
22:52:55Bagderah, right
22:53:04Bagdercodecs of course don't follow our guidelines
22:53:36CassandraThere's actually a fair few in apps. Very naughty.
22:53:49 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
22:55:26NicoFRthx for the answers
22:55:40NicoFRand keep on the good work ;)
22:55:45 Quit NicoFR ()
22:56:17BagderI've added a h300 sim win32 build now
22:56:37amiconnNice :)
22:57:27Bagderwe just need a commit now to see it appear
22:58:09LinusNi just booted rockbox on my h300!!!!
22:58:13*Bagder buckles up
22:58:19CassandraSomeone detach Linus from the mains.
22:58:37*Bagder writes down 22:58
22:59:04 Join Kohlrabi [0] (
22:59:05Bagderyou just rock LinusN
22:59:40ender`LinusN: congratulations!
22:59:46LinusNdamn, that felt goooooood
22:59:53***Saving seen data "./dancer.seen"
23:00:11CassandraBah. Now I want an H3xx so I can admire the coolness.
23:00:16ender`Bagder: it was 23:00:10 according to my (NTP-synchronized) clock :)
23:00:57BagderI think your ntp-synced clock is wrong :-)
23:01:01CassandraI'll just commit a WPS or two while I'm waiting for a revised patch from _firefly_, I think.
23:01:05CassandraI agree with Badger.
23:01:13CassandraWindows time sync bloweth the goats.
23:01:22 Join Creatine [0] (
23:02:23ender`my linux box is syncing with 2 stratum 2 servers, and my windows box is synced from the linux box with net time
23:02:45BagderI checked several hosts, they all agree with my time
23:02:49Bagderand I just resynced
23:03:16Bagderso, I guess we're on a 2 minute time zone diff ;-)
23:03:32ender`i stopped ntpd and ran ntpdate just to be sure on mnie
23:03:34amiconnCongrats Linus! :-)
23:03:56 Join webguest68 [0] (
23:04:19webguest68psst. the log: 22.58.01 # <LinusN> YIIIIIIEEEEHAAAAAAA!!!!!!!!!
23:04:32Bagdermark my words ;-)
23:04:50 Part webguest68
23:05:47ender`argh, you're right, gentoo init scripts overwrote my ntp.conf and put my isp's servers in ... which are 129,24 seconds off
23:08:57_FireFly_ender`: i don't think so :) maybe you have overwritten yourself the config with etc-update ;)
23:09:51ender`i'm pretty sure i haven't - i hacked my etc-update to color all files that aren't known to me as safe for updating in red
23:10:33_FireFly_ah good that i just done a build test :) i have found some typos of my last changes
23:12:58linuxstb_The H300 news was on the mistic river thread 5 minutes before Linus announced it...
23:13:18linuxstb_Either they've got ntp problems or Linus needs to close his windoes.
23:13:22LinusNwe have a leak! :-)
23:13:29Bagdera mole!
23:13:39LinusNor a psychic!
23:14:01Sanhehe, who posted it in the MR thread?
23:14:15Bagderactually, it was posted 4 minutes after LinusN
23:14:56LinusN11:02 PM
23:15:11 Join aliask [0] (
23:15:16peturam setting things up to help out on the H300 (oh no not another windows user)
23:15:34peturwhen building the crosscompiler, what target does one specify?
23:15:43XavierGrLinusN: Through BDM ir bootloader?
23:15:51LinusNXavierGr: bootloader
23:15:54*Bagder blushes
23:16:14XavierGrNow I am jealous of H300 users :(
23:16:48aliaskLinusN: Congrats! Now, we expect it in CVS so I can compile it and show it off at school... in say... 10 mins :)
23:17:17XavierGrand more jealous with you, that you own all targets and you made the port for Hxxx iriver series...
23:17:18LinusNdream on :-)
23:17:51aliaskHahaha. I've been following the port for a while now (before it was on the H100s) and this is really big news. Thanks so much for all your work.
23:18:01peturany cygwin user know what to do with cc1.exe failing with 'The procedure entry point _impure_ptr could not be located in ...'?
23:18:46Bagderah, a commit
23:19:03BagderLinusN: I guess a h300 boot build makes sense too now?
23:19:24*linuxstb_ can't read the time...
23:19:43_FireFly_argh is damn slow
23:19:45 Join ^Guest37784 [0] (
23:20:20elinenbeLinusN: safe for testing?
23:20:50Bagderthe build table is becoming wiiiiide
23:20:55_FireFly_Cassandra: the patch should be updated on tracker
23:21:04_FireFly_Cassandra: ok is updated
23:21:14CassandraJust had to correct my clock from being half an hour out.
23:21:15CassandraGet a better ISP.
23:21:15Cassandra*blink* Erm, OK.
23:21:16DBUGEnqueued KICK Cassandra
23:21:16CassandraA leak with a time machine!
23:21:16CassandraThe Amazing Adventures of TIME MOLE in the 21st Century!
23:21:17***Alert Mode level 1
23:21:17CassandraDon't mind me, I'm in a silly mood tonight.
23:22:13 Quit pinkutank (Read error: 104 (Connection reset by peer))
23:22:32LinusNelinenbe: nah, the usb mode doesn't work
23:22:42LinusNBagder: yes
23:22:53BagderI've added it just now, just after the commit
23:23:48XavierGrLinusN: Just curious, and maybe silly, do you have playback? I guess no... it would be too good news to be true.
23:24:02elinenbeLinusN: does it have the iriver fallback like the H100 does?
23:24:56 Join frederic [0] (
23:25:10peturnevermind... crosscompiler build started :)
23:25:18LinusNelinenbe: yes
23:25:58CassandraHow big are the user definable characters on the player?
23:26:42Creatinepetur: check your PM.
23:26:51Bagderaren't they 8 x 5? or was it 8x6?
23:26:59amiconn7 x 5
23:28:30elinenbeLinusN: So, then there is no harm in loading it up?
23:30:14LinusNelinenbe: well, it's feels brick-safe if that's what you mean
23:30:31LinusNXavierGr: no playback
23:31:02peturCreatine: nothing received (excuse me, this is all quite new to me)
23:31:16 Join uski [0] (n=uski@
23:31:18***Alert Mode OFF
23:32:04 Quit frederic ("Chatzilla [Firefox 1.5/undefined]")
23:32:36 Join PaulJ [0] (
23:32:51_FireFly_Creatine petur did you register yourself by freenode and identified when connecting to the server???
23:33:10Creatinepetur: I'm talking about IRC :P, there should be a private message for you from me
23:33:43CreatineI just connected, changed my nick because it was in use, and came here.
23:33:47CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:33:47*Creatine np: with - effects [00:51m/1032Kbps/44KHz]
23:34:26Creatineis there a problem?
23:34:46_FireFly_priv messages are block for non registered users
23:34:51_FireFly_to avoid spam
23:34:59_FireFly_on the freenode irc servers
23:35:17peturCreatine: sorry busy, back later - will look at it
23:35:20Creatinewell, thank you for informing me
23:37:31 Join gromit` [0] (
23:38:31 Quit ^BeN^ (Read error: 110 (Connection timed out))
23:38:41Bagderimage too small
23:38:58*Bagder points to the build table
23:39:17BagderI guess I'll cut off "iriver" from them
23:39:23Bagderanyone objects?
23:39:45_FireFly_from me no
23:39:48XavierGrfrom where?
23:39:50LinusNdo that
23:40:24_FireFly_XavierGr: to cut of the irver string of the build table
23:40:36XavierGryes just saw
23:40:43XavierGririver will be mad hehe...
23:40:50XavierGrbut who cares.
23:41:01Bagderthe other names are all just model names
23:41:55XavierGrwow the H300 sim is working!
23:42:05amiconnThere's a space in front - intentionally?
23:42:13Bagderno, I just fixed it
23:43:19amiconnH300 and H300 - Sim - WIn will soon turn green :)
23:43:49Bagdergreen we like
23:44:14XavierGrwhat stupid design is this? Who told them to write "multi-codec jukebox" on the side of the screen?
23:44:29XavierGrit looks odd...
23:44:32BagderI did
23:44:46XavierGrno I mean on the sim
23:44:53XavierGron the unit it self
23:44:54LinusNi *knew* it was a silly person
23:45:05Bagderthe sim uses a pic of the actual unit, afaik
23:45:06XavierGririver designed this no?
23:45:14XavierGryes that's what I am saying
23:45:39XavierGririver designed it. The company is to blame for that side screen string
23:45:52CassandraRight, I've plonked out some suggested guidelines for submitting WPSen to the patch tracker on the dev list. Comments welcome.
23:45:55Bagderok, I admit I lied it wasn't me ;-)
23:46:06CassandraNow I'll have another go at firefly's patch.
23:46:23*t0mas goes to bed :)
23:46:31 Quit t0mas (" good night!")
23:46:32XavierGrnight t0mas.
23:46:40Cassandra(I'm so lazy, I want the patch submitters to do all the work for me.)
23:46:52_FireFly_Cassandra: it should work or i'm very blind for my mistakes;)
23:47:06CassandraIf it works, I will commit it.
23:47:40CassandraAnd then EVERYTHING will break!
23:47:57_FireFly_Cassandra: give me cvs write access i will do it after i setup a cross-compileer for the other targets and tested at least if it will build on all targets
23:48:17XavierGrpreciousss write access...!
23:49:39_FireFly_i know but how should i do it as cassandra said without write access ?? ;)
23:51:22 Join Simtech [0] (
23:51:39 Part Simtech
23:52:40XavierGrwhat was that all about?
23:52:48aliaskChecking for bots I think.
23:53:03aliask!list is a command used by a lot of file serving bots
23:53:36Cassandrafirefly: You should be able to set up a cross compiler without write access to CVS.
23:54:05amiconnCassandra: You mixed up character width and height for player. It should be 14x55, not 10x77
23:54:24Mordov_patch-tracker in the wps-widget could some one direct me to this ?:)
23:55:07_FireFly_Cassandra: hmm then i has missunderstood what do you meant with your sentence "(I'm so lazy, I want the patch submitters to do all the work for me.)"
23:55:10*Bagder narrowed the title images of the build table
23:55:14Cassandrafirefly: Syntax error at line 258 of gwps.h - something odd about the struct I can't spot.
23:55:21Cassandrastructs make my head hurt.
23:55:35 Join ashridah [0] (
23:55:51Mordov_think I found it.. was looking for 10 min postest sentence and found it.....
23:55:57Cassandra_FireFly_, I was talking about people submitting WPS files.
23:56:10 Quit Kohlrabi ("Leaving")
23:56:36BagderCassandra: I think your guidelines for WPS submissions sound great
23:56:50_FireFly_Cassandra: unsigned char wps_progress_pat[8];
23:57:09CassandraGet rid of the initialisation then?
23:57:28_FireFly_had forgotten to delete it
23:57:49LinusNBagder: not to draw a nice picture in the build table by introducing compile errors :-)
23:58:06_FireFly_i think i should setup additional cross-compilers to get rid of these type of failures
23:58:26Sanwould it be hard to add a password system to the iriver ports? Like when the iriver boots, and if you get the P/W wrong 3 times, it turns of
23:58:47CassandraI think so too. I've hit another one in gwps.c
23:58:56_FireFly_line ??
23:59:00LinusNSan: a nice way to brick a player

Previous day | Next day