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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2005-11-28

00:00:00Bgerit *could*
00:00:14ModernExecutivei understand
00:00:52ModernExecutivelostlogic: in case the dcc doesn't work, do you have a gmail?
00:01:01lostlogiccan email it to me
00:01:45Bgerheh, obviously u're not afraid of spam
00:02:25Wett...yeah this chan is logged
00:02:32lostlogicBger: bogofilter does a supurb job on my spam... 500 msgs a day, zero misspammed (outside of merchant mails I want) and only 1-2 per day that are mis-hammed.
00:02:48preglowBger: what overflow from 22db to 0?
00:03:12lostlogicreceived on server, awaiting local delivery
00:03:16Bgerpreglow: i mean when u're in bass settings and after u press "up" on 22, it goes to 0
00:03:19drumRBoy320whoah, i just got a virus detected
00:03:25drumRBoy320dang it
00:03:25Bgerand in my case there was scretch ...
00:03:38preglowBger: it does???
00:03:45Bgeron h300
00:03:59Bgeri can try again...
00:04:02preglowthat code should be the same on h1x0
00:04:11amiconnThe numeric settings wrap around since TiMiD's changes
00:04:11 Quit o0260o ()
00:04:13Bgeryes, just mentioning it
00:04:16 Quit cannard ("2.0 Build 3515")
00:04:25Bgerbye Jungti1234
00:04:28amiconn(not with repeat, but with single presses)
00:04:30 Quit Jungti1234 ("Bye Bye~")
00:04:33preglowamiconn: well, they most certainly shouldn't
00:04:39preglowand that needs to be fixed
00:04:40amiconnBass should go up to 24 dB though
00:04:44Bagderit sounds like an odd choice to me
00:04:52Bgeramiconn: i'm not sure that it was 22 or 24
00:04:58*amiconn notices Bagder around
00:05:01preglowBger: 24, i should imagine
00:05:12Bgeri can try again in a min or so
00:05:35amiconnWell, this wrappping is sometimes convenient, but sometimes unexpected
00:05:47BgerAR+RANLIB /home/guest/rockbox/build/libplugin.a
00:05:47Bgermake[2]: *** No rule to make target `<<<<<<<', needed by `/home/guest/rockbox/bu
00:05:47Bgerild/apps/plugins/dep-plugins'. Stop.
00:05:47DBUGEnqueued KICK Bger
00:05:47Bgermake[1]: *** [rocks] Error 2
00:05:47Bgermake: *** [all] Error 2
00:06:04amiconnYou applied a patch that created a conflict...
00:06:14Bgeri'll see
00:06:20preglowamiconn: i don't think wrapping is a good choice there
00:06:25Bger*how* to see
00:06:31preglowamiconn: if only because it can introduce drastic changes to the sound
00:06:36preglowamiconn: like wrapping from 0db to 24
00:06:50amiconnpreglow: Yes. Even worse volume wrapping from 0 to 100 - ouch!!
00:07:29Bgerhow to see the conflict ?
00:07:30amiconnIt creates one more special case. Binary and list-like settings should wrap, numeric settings should not
00:08:15amiconnBagder: I have a working self-extractor, but it has to be integrated into the build system...
00:08:32amiconn...and first of all, the proper sub-dir needs to be agreed on
00:08:50amiconn(I have it in firmware/decompressor for now)
00:09:16Bagdersounds good enough to me
00:09:28 Join ashridah [0] (
00:09:59Bagderamiconn: if you just write down the steps, I'll make sure to incorporate them into the build
00:10:00amiconnIt's just 3 files - decompressor.c, and a Make file
00:10:11lostlogicModernExecutive: check PM and email.
00:10:37amiconnThis Makefile needs some work - I don't quite get how all the directory variables are supposed to work
00:10:58amiconnFor now, it does everything within its own dir
00:11:01ModernExecutivejust realized I couldn't send msgs because i'm not registered
00:11:06ModernExecutive... i should do that i guess
00:11:21ModernExecutivei had said that the sum is different, but that I guess that's to be expected because it's a different fw
00:11:35lostlogicModernExecutive: nod.
00:11:43amiconnBagder: I should probably commit this stuff first
00:11:52Bagderamiconn: yes, sounds like a plan
00:11:59amiconn(and uclpack under tools/ )
00:12:06Bgercan someone explain me what does the following mean
00:12:06lostlogicas you can see, the steps followed were identical to those which I followed to make the firmware which I'm currently running, so if it's compatible, it'll work.
00:12:06Bger<<<<<<< SOURCES
00:12:06***Alert Mode level 1
00:12:06***Alert Mode level 2
00:12:06Bger>>>>>>> 1.65
00:12:50ModernExecutiveand i guess that's the big iff
00:12:51ModernExecutiveer if
00:12:58ModernExecutivethank you so much for your help
00:13:02preglowBger: collision
00:13:20Bgeryeah, but more preciselly?
00:13:21preglowBger: cvs detected you have done some changes in the same place as someone else, and you need to fix the collision manually
00:13:37preglowBger: the upper part is what your file says, the lower part what cvs says
00:13:47preglowBger: just edit it to say what you want
00:13:59drumRBoy320ok, i need some halp
00:14:09Bgerwhat do you mean with upper
00:14:16Bgerthe part above "====" ?
00:14:22drumRBoy320what is the easiest way to exit the navi menu without stopping the music
00:14:23 Join planetik [0] (
00:14:35lostlogicdrumRBoy320: just hit play
00:14:39BgerdrumRBoy320: stop ?
00:14:41lostlogictakes you back to the WPS screen if something is playing
00:14:41drumRBoy320i cant figure out how to do it without stopping the music...
00:14:54drumRBoy320play button does?
00:15:02drumRBoy320ok... i thought i tried that
00:15:08drumRBoy320ill try it again
00:15:11BagderBger: yes, the ==== separates the two parts
00:15:49drumRBoy320haha, it works, ty
00:15:55Bger10x :)
00:16:05 Nick CBM-away is now known as CheeseBurgerMan (n=BurgerBo@
00:17:53Bgerlostlogic: i'm compiling already, 10x anyway :)
00:18:46lostlogicahhhh, didn't figure otuwhat 10x meant
00:19:00Bger10x = thanks
00:19:01 Quit ModernExecutive ()
00:19:18lostlogicgot it now ;)
00:19:27Bgerheh, it's very popular
00:19:32lostlogichope the firmware I built for him works :-\
00:19:48lostlogicI really wish my cross-binutils would make the right executables.
00:19:50Bgeri hope too ...
00:20:12Bgerlostlogic: how did you compile your cross-dev
00:20:47lostlogicI used the gentoo linux crossdev package "crossdev -s1 -t m68k-elf" I -suppose- I could try doing it by hand instead.
00:21:09amiconnHmm :/
00:21:10Bagderso many people have tried the gentoo mess
00:21:12Bagderand failed hard
00:21:21Bgerlostlogic: forget about it
00:21:23lostlogicBagder: ah, has a history of failure
00:21:27amiconnAfaiu the ucl sources use automake
00:21:33Bgeri was among these people
00:21:36amiconn...and the whole package is ~2MB I have to find out how to cut this down :/
00:22:04Bgerlostlogic: the last time Bagder was affected by my crossdev
00:22:07***Alert Mode OFF
00:22:11Bgerbut mine problem was other ...
00:22:14Bgervery strange
00:22:30Bagderamiconn: oh yes it does... but I don't think that needs to be much of a problem to us
00:22:47 Quit planetik ("CGI:IRC")
00:22:47Bgeri started to write a plugin ... and with little code changes it run, without them - error
00:22:51amiconnWhat would you suggest?
00:23:05preglowamiconn: what's the status of −−nogap files in current rockbox decoder?
00:23:24amiconnDidn't test for quite a while
00:23:35preglowwell, if you do test, let me know
00:23:55amiconnDo you think you changed something for the worse?
00:23:58Bgerlostlogic: btw, if u do m68k-elf-gcc −−version, u'll see that this is not *plain* gcc
00:24:02amiconnLast time I tried it worked
00:24:03 Join linuxstb [0] (
00:24:07Bagderamiconn: I suggest we ignore the original makefile and write up our own
00:24:21preglowamiconn: yeah, i do think so
00:24:23lostlogicBger: what do you mean by not plain gcc?
00:24:34lostlogicBger: you mean because of all the gentoo patches?
00:24:37preglowamiconn: i now 500 samples or so from the start of every mp3
00:24:39amiconnArgh, another makefile mess :/
00:24:57Bagderit doesn't mean a mess
00:25:08amiconnWell, if you know how to do it
00:25:12Bgerlostlogic: yeah
00:25:17Bgeri suppose the problem is there
00:25:32preglowamiconn: insert 'cut' into that last sentence
00:25:47amiconnI think I understand makefiles well enough to do some changes, but writing one from scratch is a different matter...
00:26:35Bagderamiconn: if you just add and commit the source code, I'll write up the makefile stuff for it
00:26:44drumRBoy320what is the h300 screen res?
00:27:16Kyl3is there any way to, when you start rockbox, have it start on the WPS
00:27:38Kyl3rather then a file browser
00:27:40drumRBoy320tha'd be cool.
00:27:45BgerKyl3: it's on one key distance :)
00:27:59ashridahKyl3: you can have it automatically resume, that's about all i can think of. the WPS doesn't really activate unless something's playing.
00:28:15Kyl3ahh i see
00:28:19amiconnBagder: I can add my decompressor stuff right away,
00:28:43amiconnbut the ucl sources are several directories, most of which are probably not needed
00:29:30 Join skorpyon [0] (
00:29:31amiconnThe problem lies in this "probably"
00:29:54Bagderright src/ and include/ should be enough for libucl
00:30:12amiconnAt least one more: examples/uclpack.c
00:30:13Bagderand then the little uclpack example code
00:30:36amiconnThat's what we need (replaced by [IDC]Dragon's patched version)
00:30:48drumRBoy320will rockbox convert a color bitmap to b/w in the wps?
00:31:01drumRBoy320or should it be in b/w
00:31:16amiconnBagder: I'm not sure about the 'B' and 'asm' sub-dirs
00:31:18Bagderamiconn: I suggest a tools/ucl/ dir and then the src/ and include/ as subdirs within
00:31:31Bagderamiconn: ignore them
00:31:49amiconnShould I leave the src/ucl structure?
00:32:01amiconnWhat about the Makefile.* files, remove them?
00:32:13BgerdrumRBoy320: more probably it'll refuse to load it ... or load it not in the right way
00:32:17 Join jaydpb [0] (
00:32:26drumRBoy320ok, and is easywps compat w/ h300?
00:32:44Bagderamiconn: yes, I think we should keep the src/ and include/ as subdirs
00:32:54Bagderand you can remove the Makefile* s
00:33:03amiconnI mean the ucl dir within src/
00:33:22BagderI don't have any such
00:33:31amiconn1.01 ?
00:34:00amiconnBleh, I mean iclude/ucl
00:34:12Bagderah, yes keep that
00:34:26Bagderthe include files are used #include <ucl/[file]>
00:34:43 Join JdGordon [0] (
00:34:52amiconnOkay. uclpack.c goes directly into tools/ucl/ ?
00:35:30linuxstbI haven't tested it properly (it compiles), but I've stripped down ucl here:
00:35:45 Quit Kohlrabi ("Leaving")
00:35:47amiconnOkay. I'll check in the unmodified version first, then apply the patch
00:35:47JdGordonmorning all
00:35:53preglowmorning my ass
00:36:02Bagderamiconn: goodie
00:36:04JdGordon10.35am is still morning
00:36:12linuxstb(but I was probably too late....)
00:36:17preglow00.36 isn't :P
00:36:27Bagderlinuxstb: how stripped down?
00:37:23amiconnlinuxstb: This is 1.03 ...
00:37:28amiconnWe need 1.01
00:37:38amiconnNo difference in the compresion algo itself
00:37:46JdGordonhaha @ ur email linuxstb
00:38:47 Quit jaydpb ("CGI:IRC (EOF)")
00:39:54 Join el_taco [0] (
00:40:35lostlogic4 minutes till I get to try an unpatched manually compiled toolchain.
00:42:03JdGordondoes any1 know if the software doesnt check if buttons are pressed when hold is down? or if the hardware actually doesnt change the values?
00:42:29ashridahhey, which actual sector on disk does rockbox store its settings in?
00:43:04lostlogicashridah: I don't think it's a sector thing, it's just a file...
00:43:25ashridahlostlogic: except it is a sector thing, you can nerf .rockbox and the settings stick :)
00:43:46lostlogic::blink blink:: I see.
00:43:49ashridahthe only time they reset is if someone adds/removes a config setting and bumps the config version
00:44:02linuxstbashridah: It's normally sector 61
00:44:04JdGordonmaybe its saved to rom then?
00:44:21ashridahJdGordon: definently not
00:44:26ashridahlinuxstb: tar
00:44:38linuxstbThe first sector of your first partition is normally sector 63.
00:44:41amiconnBagder: UCL sources are there, complete with patch
00:44:52BagderI'm on the case
00:45:05 Quit skorpyon__ (Read error: 110 (Connection timed out))
00:45:45Bgerbtw, there is a bug with the "hdd activity" on the statusbar on h300 ... it's displayed only on the remote's lcd
00:46:14Bgerjust added a patch for h300's lcd button mappings
00:47:35Bgerbut i don't think it should go into cvs :)
00:48:59Bgeramiconn: what do you think about the remote "mistery" ?
00:49:51amiconnBger: It's only displayed on the remote because the H1x0 has a real LED on the main unit no symbol necessary
00:50:08 Join vik [0] (
00:50:17JdGordonis that was the circle thing in the status bar is? hdd activity?
00:50:20amiconn...and afaik the H3x0 also has a real LED (?)
00:50:32drumRBoy320no status LED in the h300
00:50:52lostlogicdid Linus mention why he lowered the H300 burst mode down to 90mhz from 140 or whatever it was?
00:50:57amiconnHmm. I remember Linus saying there is a LED??
00:51:06Bgeramiconn: no, there isn't
00:51:20JdGordonlostlogic: ye, so audio would work
00:51:21Bgerand linus said just that
00:51:38JdGordonmy h300 has no led
00:51:46Bgerbut he said also that only h100 has led ...
00:52:00amiconnHe said that only H100 has led *control*
00:52:09lostlogicJdGordon: ah... I was thinking it was because of heat or something, high-ish bitrate ogg/vorbis skips, was hoping cranking it back up would fix that, but I guess not.
00:52:26amiconn.. I read that as the H3x0 having a LED as well, just not controllable at all
00:52:27Bgerok, h300 doesn't have ANY led
00:52:44 Quit ender` (Read error: 104 (Connection reset by peer))
00:52:46Bgernot even one
00:52:52amiconnThe H1x0 LED power can be controlled, deactivating it completely
00:52:56JdGordonactually... i tihnk it has 6
00:53:04JdGordonbacklight leds :D
00:53:06amiconnSo for H3x0 LED_VIRTUAL should be defined
00:53:22amiconn#define CONFIG_LED LED_VIRTUAL
00:53:27JdGordon... speaking of which.. can the backlight be turned down?
00:53:55Bgerbrightness ?
00:54:00amiconnThe little circle should resemble a hard disk; if someone could come up with a better icon...
00:54:11amiconn(max size is 7x16 pixel)
00:55:04vikI'm trying to set up a toolchain for linux to play with the H300 version; do I actually need the source of gcc et al, or can I just install the binary (I'm running ubuntu)?
00:55:34JdGordonBger: no brightness settings
00:55:51amiconnBagder: decompressor committed as well.
00:55:54JdGordonand inverse mode lcd doesnt work
00:55:55lostlogicvik: you need the source unless ubuntu has a cross compiler for m68k-elf in binary
00:55:58BgerJdGordon: yep, i know, just asking whether you have this in mind
00:56:32Bgerthere should be such setting, we know it's controllable
00:57:12viklostlogic: I've got a deb gcc-3.3-m68k; will that work or do I have to use 3.4?
00:57:20lostlogicBger: well I'm goign to have to beat some ass over at the Gentoo camp, because manually built toolchain works great where crossdev built by Gentoo didn't.
00:57:42lostlogicvik: is that a cross compiler or just a straight m68k compiler?
00:58:01viklostlogic: crosscompiler
00:58:55ashridahvik: if it's the most recent 3.3, it should work. minimum was 3.3.something. can't hurt to try
00:58:59lostlogicshould probably work, need binutils cross for target as well
00:59:48amiconnm68k-elf-gcc 3.3.x ICEs somewhere iirc
01:00:06linuxstbvik: Most people who try to use cross-compiler binaries fail. I think you will save time in the long-run by just getting the source and compiling it yourself.
01:00:12ashridahamiconn: does it? thought we had it listed as working. or did we move to 3.4?
01:00:53lostlogicI must agree with linuxstb after wasting tons of time using the Gentoo packaging system's attempt to make a cross compiler, it took me < 20m to follow Linus' instructions and build a working toolchain from source.
01:01:21ashridahi'd have to agree tho. building it isn't too difficult, there's cookbook instructions on's wiki, and it won't contaiminate the rest of the system due to the way the tools get named when you're building crosscompilers
01:01:21Bgerlostlogic: i'm going to help with ass beating :)
01:01:36vikjust found some binaries for 3.4, but may as well do some compilin' - how long does it take to compile the compiler?
01:01:39 Quit ansivirus (Read error: 104 (Connection reset by peer))
01:02:01lostlogicBger: you a gentoo dev?
01:02:02Bgerdepends on your machine, but ...
01:02:02ashridahvik: not long, since you're restricting it to compiling a C compiler only (no c++/java/fortran/etc)
01:02:04Bger~ 20 min
01:02:33Bgerlostlogic: i'm gentoo user and i also want to beat someone's ass ... :P
01:02:41 Quit dpassen1 ()
01:02:41vikAnd on a Pentium 100?
01:02:43ashridahheh, being on dialup, it takes me longer to download the source than compile it :)
01:03:03linuxstbJdGordon: I've just committed your Sudoku patch for the h300.
01:03:11JdGordonkewl :)
01:03:12Bgeron AMD @ 1400 MHz ~ 10-15 min...
01:03:13ashridahthe internet works on a pentium 100? :)
01:04:04vikashridah: there's always lynx
01:04:07 Join ModernExecutive [0] (
01:04:09ModernExecutivehi guys
01:04:13lostlogicModernExecutive: any luck?
01:04:19ModernExecutivethe fw didn't take
01:04:24ModernExecutivei didn't brick my player
01:04:32Bagderamiconn: can you just try 'make' in the ucl dir now and see if it works fine for you?
01:04:38lostlogicModernExecutive: what do you mean didn't take?
01:04:39ashridahvik: bah. bloaty. telnet works well >:)
01:04:44ModernExecutiveI went through the upgrade screens
01:04:53ModernExecutivegot the "firmware upgrading, don't touch a thing" screen
01:04:54preglowamiconn: so sh has a vbr register as well?
01:05:07ModernExecutiveand then the player went to the shut down screen right after it finished reading the fw off the disk
01:05:22ModernExecutivebooted back up like nothing had changed
01:05:23amiconnpreglow: yes
01:05:32ModernExecutiveand nothing had changed
01:05:35***Saving seen data "./dancer.seen"
01:05:38amiconnWtf, now the decompressor hangs again :(
01:05:50lostlogicModernExecutive: DRM still works? You copied the new firmware file as H300.hex to the player?
01:05:59amiconnpreglow: SH1 even has a full VBR register, unlike coldfire
01:06:13ModernExecutiveas re: drm. I don't have any drm files, but the player isn't sending me the message that the drm key can't be found
01:06:16lostlogicvery weird
01:06:24ModernExecutiveso i presume it's still there
01:06:41Nixsosis there a wiki page with information about creating plugins for H3xx series?
01:06:58Nixsoscolor plugins that is...
01:07:16ashridahi was under the impression that the key would change, not that it'd be lost
01:07:17linuxstbNixsos: No, but the GraphicsAPI page is a good place to start.
01:07:35linuxstbAs well as looking at the source of the other plugins.
01:07:46amiconnBagder: 'make' in ucl/ works, but two strange things happen
01:08:04amiconn(1) uclpack.c and uclpack.o are copied one dir up (to tools/)
01:08:12amiconn(2) There is no uclpack.exe
01:08:45Bagdernow that is strange...
01:09:16amiconnCorrection, uclpack.exe is there
01:09:32amiconn...just ls -l u* confused me
01:09:49Nixsoslinuxstb: thanks... i'm totally new to this. i'll delve into it and maybe even write a tutorial for other developers starting out with writing plugins for RB
01:09:52ashridahModernExecutive: did you actually copy the firmware into the root of the player? :)
01:09:55amiconn..and there are no files one dir up
01:09:58ashridah(with the right filename)
01:10:09Bagderamiconn: I made it put uclpack in the tools/ dir just to keep all tools there
01:10:48amiconnGah, Why the *** does ls -l u* list both all files and dirs beginning with u _and_ the contents of these dirs??
01:11:03lostlogicamiconn: ls -ld
01:11:21ashridahamiconn: because the shell expands u* before ls -l sees iot
01:11:22Bagderbecause the shell does the wildcard expansion, and not ls itself
01:11:23ashridahit even
01:11:35linuxstbNixsos: One thing to remember is that Rockbox runs on lots of different types of players. Almost all the plugins run on all targets.
01:11:35ashridahso what ls runs as is ls -l ubla ubla2 udir1 udir2
01:11:44amiconnStill not very logical, imho
01:12:02preglowanother of my unix favourites
01:12:07preglowshell based wildcard expansion
01:12:16ModernExecutiveit must have been right
01:12:33Bagderok, I added uclpack to the main tools/Makefile now
01:12:41Bagderlet's hope it works fine
01:12:59ModernExecutivewhen there's no h300.hex file in the root of the player, trying to upgrade the firmware just returns you back to the general control screen
01:13:11Bgeramiconn: idea how to stop this silly autorun on plugging mass storage device on machine with winxp ?
01:13:17 Quit DocBrown ("CGI:IRC")
01:13:24ashridahModernExecutive: so the question becomes, did you copy the right file? :)
01:13:24ModernExecutivewhen i placed the patched us h300.hex in the root and tried to upgrade, it went to the firmware upgrade screen, telling me not to touch anything
01:13:24Nixsoslinuxstb: so you rather not have H300 only plugins?
01:13:39ModernExecutiveit was 4,094 kb.
01:13:40ashridahBger: install tweakuiXP
01:13:43amiconnYes, deactivate autorun for all drive letters which could be potential USB MSDs
01:13:53Bgerhm trying
01:14:10CassandraJust deactivate autorun full stop.
01:14:25Bgerok, 10x
01:14:49ashridahCassandra: that's too simple. can't have simple. just not done :)
01:15:11CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:15:11*amiconn now wonders why the decompressor works with cvs, but not with unicode-patched rockbox
01:15:35CassandraHeavens forbid you'd put a disk in your drive and it *not* do anything until you tell it to. I mean that'd be ... too confusing, or something.
01:15:47Bgerregarding unicode : will the default rb font include cyr & greek ?
01:15:58amiconnNot the built-in one
01:16:13amiconnThat's probably one of the next steps:
01:16:30amiconnHave a disk-based default font as full unicode
01:16:54CassandraDon't we actually need working unicode support first.
01:16:56*Bger didn't update the bulgarian because of missing cyr letters in many places
01:17:08ashridahCassandra: apparently that's getting close to merge
01:17:30ModernExecutiveother thoughts?
01:17:36markunI'll commit some unicode versions of some of the fonts that are included now after the unicode commit
01:17:37Bgerok, now i have virtual led on the main screen
01:17:59ashridahModernExecutive: compare the md5sum of the h300.hex from the disk against someone's working md5sum
01:18:08*Bagder hits bed
01:18:57markunthe latin1 fonts also work as unicode fonts, but the others will have to be replaced/converted
01:19:05linuxstbNixsos: I don't think anyone would tell you not to write a h300 specific plugin, just to think about making it portable when you do write it.
01:19:23markunGood night Bagder
01:19:30Bgernite, Bagder
01:19:33vikJust been reading the wiki - I can't quite get it clear - should I be using gcc 3.3 or 3.4?
01:19:45ModernExecutiveashridah: the md5sum would be different anyway. it was a us version of the firmware that was patched
01:19:51ashridahuse 3.4
01:19:57ashridahModernExecutive: oh.
01:20:26linuxstbvik: 3.3.x is for the Archos (SH1) targets, 3.4.x is for the iriver (Coldfire/m68k) targets.
01:20:46elinenbeamiconn: is this UCL commit preceding the unicode patch?
01:20:52JdGordon~var does bitwise inverse of var right?
01:21:04amiconnelinenbe: It's one step towards it
01:21:05ModernExecutivefwiw: the md5sum was 8747c0c420a977521e22c70d2b42baff
01:21:54markunamiconn: what's going wrong with the compressed unicode rockbox?
01:21:59amiconnI also committed the decompressor, but it seems to need more testing
01:22:08amiconnmarkun: It just hangs...
01:23:27amiconnGah, it also hangs on the player
01:23:53ashridahModernExecutive: how does that compare to the original 1.03 h300.hex?
01:24:30 Quit Moos ("Glory to Rockbox")
01:24:32JdGordony u using 1.03?
01:24:34ModernExecutivei've never done a md5sum on the 1.03 h300.hex
01:24:38ModernExecutivepersonal experiment
01:25:02ashridahModernExecutive: well, perhaps it might be an idea to check? :)
01:25:05ModernExecutivei want to see if it's possible to flash to a patched us version of the firmware and still use drm?
01:25:12ModernExecutivei'd be happy to
01:25:26ModernExecutivebut i don't know how. :(
01:25:52ashridahModernExecutive: eh? i meant the md5sum of the original .hex you used
01:26:11markunamiconn: do you decompress in place?
01:26:15ashridahyou just gave one for the modified one
01:26:21lostlogic380502405be24e3d58453c3b89d43872 H300-103US.hex
01:26:24amiconnmarkun: No, that's not possible
01:26:33lostlogic8747c0c420a977521e22c70d2b42baff H300-103US-rbv2.hex
01:26:34amiconnThere are 3 stages
01:26:44ashridahokay then
01:26:51*ashridah <−− out of ideas
01:27:05markunamiconn: So you move the compressed image to the back of the memory?
01:27:08CassandraAnd by a remarkable co-incidence, 3 stooges as well. Who'd have thought it?
01:27:14amiconn(1) Copy all code that's needed later into IRAM (otherwise it would overwrite itself)
01:27:16CassandraI should go to bed, I think,
01:27:21ashridahModernExecutive: my bet is that the jump to rockbox's firmware isn't ending up in the right place and thus, isn't booting. you're lucky it isn't bricking your player
01:27:33amiconn(2) Copy the compressed image to the upper end of DRAM
01:27:45amiconn(3) Decompress the image to the start of the DRAM
01:27:54amiconn..and run it
01:28:19 Join schnittn2005 [0] (
01:28:20markunMaybe you don't move it far enough and a part get's overwritten?
01:28:39markunHow much ram is there? 2MB?
01:28:44amiconnI move it to ram_end - compressed_image_size
01:28:49amiconnYes, 2MB
01:28:55markunok, should be no problem then.
01:29:51ModernExecutivei guess so
01:30:27ModernExecutivelucky me
01:30:34 Quit Wett (Read error: 110 (Connection timed out))
01:30:52Bgernite all
01:30:53markuncan't imagine why the unicode build fails to decompress, I would think only the size is different.
01:31:15 Quit Bger (Read error: 104 (Connection reset by peer))
01:31:49schnittn2005hi guys! can anyone tell me (query) a bit ybout this ucl?
01:31:58amiconnThere's something fishy, as the decompressor also hangs on the player
01:32:10amiconnI thought I solved it...
01:32:31markunamiconn: Does it hang during decompressing or after running rockbox?
01:32:57amiconnIt hangs somewhere within the decompressor. No screen output
01:33:05ModernExecutivethanks for your help, everyone
01:33:30markunand the ucl image decompresses fine with the uclpack tool?
01:33:43ModernExecutiveguess i'll wait to see if someone can crack that us firmware
01:34:01markunschnittn2005: what do you want to know about it? It's like zip
01:34:34markunso we can load a bigger firmware (there is a artificial size limit)
01:35:03 Quit mirak (Read error: 104 (Connection reset by peer))
01:35:49schnittn2005like zip, ah ok. thanks.
01:36:29schnittn2005is this artificial limit almost reached?
01:36:31 Quit ModernExecutive ()
01:37:14amiconnYes, on one of the archos targets
01:37:23amiconnThe recorder v1 to be precise
01:37:25ashridahschnittn2005: it's a barrier to merging the unicode patch because it makes it too big for some platforms.
01:37:56schnittn2005i see
01:40:11schnittn2005isnt there any crossfade tag yet? does anbody know?
01:40:21schnittn2005for the wps
01:40:36markunschnittn2005: what would it's use be?
01:42:57JdGordonok, how the hell do i go about debugging my plugin?
01:43:24ashridahJdGordon: get it running in the simulator? :)
01:43:26linuxstbJdGordon: Have you tried the simulator and DEBUGF?
01:43:46JdGordonit wont load int he sim :(
01:44:27linuxstbWhat does your plugin do?
01:44:36linuxstb(or not do...)
01:44:44*JdGordon is fiddling with the sudoku plugin again
01:45:01linuxstbSudoku works fine in the h300 sim for me.
01:45:24JdGordonit says not a valid win32 image when i try to load a ss int he win32 sim
01:45:30schnittn2005markun, its use would be: normally when i hear music for myself, i dont use it. when i play on a party, for example i wanna use it. just to see if its on or not. the same like shuffle or repeat.
01:46:00linuxstbJdGordon: Sounds like you are using a .rock compiled for the target.
01:46:13 Join Paul_The_Nerd [0] (
01:46:22linuxstbHave you run "make install" after running make to build the sim?
01:46:53*JdGordon slaps self
01:47:12elinenbeamiconn: good luck with everything...
01:47:26elinenbeamiconn: one question, would the font format change once unicode is supported?
01:47:43markunIt did not change much
01:48:06markunonly it now has support for very large fonts, but the old latin1 fonts should also work
01:48:34markuninternally there is a caching mechanism
01:49:23markunelinenbe: why?
01:49:32linuxstbDoes that go any way towards supporting multiple fonts in Rockbox?
01:50:39markunwell, the font space could be split for multiple fonts, but if it gets to small the glyphs will be read from the HD all the time..
01:51:11preglowmarkun: do you save the cache at shutdown?
01:51:27markunAnd when changing the font
01:52:27preglowhow's the efficiency? do you keep a glyph nr. -> cache position table?
01:52:45markunI'm not sure
01:53:25markunIt was stolen from the CJK patch and maintained by phaedrus :)
01:54:19markunpreglow: You can take a look if you want:
01:54:19 Quit unkle ("CGI:IRC (EOF)")
01:54:43preglownah, i'll pass
01:56:34markuncompiled on the ipod. I think it will also run because the code in lcd-16bit is also used by the H3x0
01:56:37JdGordonholy crap.. its working :O
01:58:33elinenbeJdGordon: what's working?
01:58:47JdGordonmy mods of the sudokuplugin
02:02:23JdGordon..almost working anyway
02:04:20JdGordonlinuxstb: is it possible to just compile a single file without having to compile everytihng?
02:04:40JdGordonby everything, i mean using make
02:05:22linuxstbJdGordon: typing "make" will just rebuild what has changed.
02:05:38JdGordonye, still takes a while to go through everything
02:05:43JdGordoneven if only the 1 file changed
02:06:08vikis it OK to build the cross-compiler using gcc4, or should I drop to 3.4?
02:06:48lostlogicvik: I check the CrossCompile wiki page.. I think it mentions that
02:08:01linuxstbJdGordon: I know - I don't think there's anything you can do to avoid that.
02:08:34viklostlogic: mentions it, but only related to building rockbox, not the cross-compiler
02:08:44amiconnvik: You can build m68k-elf-gcc 3.4.x using native gcc4
02:10:14vikand does the version of gdb matter? 6.1 or 6.3?
02:10:45ashridahvik: gdb isn't much use anyway, since we don't have a gdb serial connection to the iriver players.
02:10:53ashridahno-one implemented it
02:11:07viknot even for simulation mode?
02:11:26ashridahfor the simulator, whatever gdb goes with your system will work
02:12:02ashridahsince that compiles to native code
02:16:34vikashridah: I don't understand - code that runs on the simulator is compiled native to the machine that it runs on? i.e. it is not emulated?
02:17:01ashridahit's a simulator
02:17:02ashridahnot an emulator
02:17:18ashridahthere's no interpretation of targeted code
02:17:50ashridahthis means that plain-ole gdb works fine there
02:18:02JdGordonany1 wanna guess why this damn code works fine if curval is <7 then fucks up??
02:18:16JdGordonstate.possiblevals is an aray of ints
02:18:41JdGordonok, ignore that
02:19:33*JdGordon enjoys looking stupid.... incase you couldnt tell...
02:19:40JdGordonstupid fucking coding b4 breakfast :p
02:19:45vikis the coldfire emulator any use?
02:23:56JdGordonany1 here actually play sudoku?
02:24:11linuxstbJdGordon: I do occasionally.
02:24:22ashridahvik: for the cpu maybe, but it'd probably need work to add emulated hardware for the peripherals
02:24:23linuxstb(I wrote the plugin originally)
02:24:32JdGordonah ok
02:25:07JdGordonive almost finished my mods...
02:25:19linuxstbWhat are you changing?
02:25:40markunsudoku would be a lot nicer with a generator
02:25:47linuxstbI agree.
02:27:02JdGordonthe column on the left lets u mark which numbers could be valid in the cell....
02:27:32JdGordonu press rec to add a number there, so its still hard but makes it just as easy to play on rb than on paper
02:28:21JdGordonif there is enough room on the other players ill add it there also... but i doubt there is
02:28:57linuxstbI'm sure there is room - the grid is square, and all the LCDs are much wider than they are tall.
02:29:27JdGordonill have a look... should be prety simple to add to each.. just need to play with numbers
02:30:01vikJdGordon: u at Monash huh?
02:30:22JdGordonhow could u tell... :p
02:30:56Paul_The_NerdJdGordon Also, the scratch bar on the left doesn't have to be full size tiles, for smaller screens.
02:30:58vikJdGordon: I have uncanny powers of deduction
02:31:32linuxstbPaul_The_Nerd: That's not a problem - the scratch bar needs 9 cells, which is how high the main grid is.
02:32:03JdGordonwworst comes the worst, it could be done on a 2nd screen if the screen really s that small....
02:32:39 Quit el_taco ("Leaving")
02:32:45JdGordonk, ttyl ppl
02:32:47JdGordonbreakfast time
02:32:56linuxstbI'm sure there is always space to the side of the grid.
02:33:03Paul_The_Nerdlinuxstb: But there might not be enough width for a full cell on some screens. I don't know though, just saying if there's not, that's an option.
02:34:51linuxstbThe H1x0's LCD is 160x128, with a 122x122 grid. The Archos is 112x64 with a 82x64 grid. So plenty of room on both.
02:34:58Paul_The_NerdCool, cool
02:35:09Paul_The_NerdIt'd definitely be a useful addition
02:37:07linuxstbDo you think Sudoku on the H1x0's remote would be useful? Or would that be too small to be playable? (I don't have a remote)
02:37:50Paul_The_NerdIt'd be playable
02:37:54Paul_The_NerdBut navigation would be irritating
02:38:08Paul_The_NerdOne second
02:39:22Paul_The_Nerdlinuxstb: In the 5x8 font you get 8 readable lines + status bar, so theoretically you could cram 9 readable lines in, assuming you only separated by spaces.
02:39:30Paul_The_NerdIt would be barely readable, and hard to navigate, but doable. :-P
02:40:31linuxstbThe status bar would be disabled, giving us the same screen height as the Archos.
02:40:49Paul_The_NerdI imagined it would
02:41:11Paul_The_NerdYou'd basically not be able to draw horizontal lines, I think. You'd just have a grid of the numbers themselves, I imagine.
02:41:55linuxstbDo you know if the pixels are square on the remote? They are rectangular on the Archos LCDs, so the squares are in fact 8x6.
02:42:14linuxstbOn the Archos, I draw single and dotted lines for the grid.
02:42:33Benacoolthey looks square
02:42:35linuxstbI don't think it would be playable without gridlines.
02:42:36Paul_The_NerdI *think* they are square
02:42:42Paul_The_NerdIf the H120 main LCD is. They look almost exactly identical
02:44:46linuxstbOK, I'll assume they are square. I would like to play with that this week sometime.
02:48:22JdGordonlinuxstb: ok, if there is room it would be doable... but it sounds like the whole board would be needed to be moved to the right to fit it
02:50:45linuxstbPossibly, but that is just #define - XOFS
02:51:19linuxstbIf you position it on the far-left, then it may fit without moving the board.
02:52:13drumRBoy320are there any characters that the h300 would not display? i would like to make a word scroll in mt WPS, but i dont want to have see-able characters at the end of the spaces
02:52:15preglowanyone have any spare svn space i can borrow? ://
02:52:41*preglow misses his linux box
02:57:19 Join Soul-E [0] (
02:58:57Soul-Ecan i not join other chans with the webclient?
03:01:20vikgot everything compiled; does the simulator simulate anything useful for the h300?
03:02:05preglowdepends what you measn
03:02:12preglowit simulates menus, plugins, playback, etc
03:03:17vikHmmm... doesn't really do much so far - what are the keybindings for the h300?
03:04:26 Quit Soul-E ("CGI:IRC (Ping timeout)")
03:05:37***Saving seen data "./dancer.seen"
03:10:01*Paul_The_Nerd cannot use the sims effectively on his laptop.
03:10:29*CheeseBurgerMan suggests that Paul_The_Nerd watch the IRC channel instead.
03:10:54 Join Vlad0man [0] (
03:11:03Paul_The_NerdCheeseBurgerMan: "the sims" were referring to the H100/H300 simulators.
03:17:11Paul_The_NerdI've never actually played The Sims for more than about 5 minutes
03:20:09CheeseBurgerManI've seen it, I've never played it at all.
03:20:31Paul_The_NerdIt was... really boring.
03:20:45Paul_The_NerdI already live life, I don't need a second one that's very similar
03:20:55CheeseBurgerManYeah, that's what it looked like.
03:21:32CheeseBurgerManOnly vaguely interesting part was when the husband got burned by the oven, and then the wife couldn't eat because his remains were blocking the fridge door. :P
03:22:03CheeseBurgerManThat was interesting to see for the first few minutes.
03:22:59Paul_The_NerdI always want to torture them.
03:23:26CheeseBurgerManSorry, you can't get tortuing devices IIRC. ;)
03:23:41CheeseBurgerManAlthough I actually have no idea.
03:25:13 Quit JdGordon ("Miranda IM! Smaller, Faster, Easier.")
03:27:07 Quit _Vladoman (Read error: 110 (Connection timed out))
03:27:52Paul_The_NerdYes, but you can, say, surround someone with fridges pointing outward, and then watch them starve to death surrounded by food
03:28:05vikHmmm. the .rockbox folder wasn'
03:28:22viks/$/t created when I compiled the sim/
03:29:02vikwhere does the .rockbox folder live/come from?
03:29:12 Quit Benacool ()
03:29:28lostlogicvik: make install
03:30:07lostlogiclives in builddir/archos, iirc
03:30:48*vik slaps his forehead
03:30:57 Join JdGordon [0] (
03:31:24 Join iLyric [0] (
03:31:38 Join solexx_ [0] (
03:31:38 Part iLyric
03:32:03 Join iLyric [0] (
03:32:10iLyrichi guys
03:32:29iLyrickeep up the gud work
03:32:39iLyricbye bye
03:32:43 Quit iLyric (Client Quit)
03:33:33 Quit Kyl3 ()
03:33:55vikhow do you access the plugins in the sim?
03:35:57 Quit JdGordon ("Miranda IM! Smaller, Faster, Easier.")
03:38:15 Quit CheeseBurgerMan (Read error: 104 (Connection reset by peer))
03:40:00 Join CheeseBurgerMan [0] (n=BurgerBo@
03:40:44lostlogicinsert gets to the main menu IIRC
03:42:19vikcool - thanks
03:43:01 Quit solexx (Read error: 110 (Connection timed out))
03:46:30 Quit vik ("Leaving")
03:50:08 Join MohLeh [0] (
03:51:05 Join DreamTactix291 [0] (
03:51:09 Part MohLeh
03:52:29 Join Membrillo [0] (
03:55:20 Quit Membrillo (Client Quit)
03:55:37 Quit TCK ("I shall not cease!")
04:05:15 Join webguest15 [0] (
04:05:36 Part webguest15
04:17:45 Join iLyric [0] (
04:17:47 Join jaydpb [0] (
04:24:38 Quit Cassandra (Read error: 113 (No route to host))
04:26:15 Part iLyric
04:26:15 Join andrewmel [0] (n=andrewme@
04:26:32 Join dropandho [0] (
04:27:18dropandhohey all!
04:27:24dropandhoslasheri- you happen to be around?
04:28:39 Join elinenbe_ [0] (
04:28:39 Quit elinenbe (Read error: 104 (Connection reset by peer))
04:28:45 Nick elinenbe_ is now known as elinenbe (
04:33:32 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
04:37:07dropandhoany1 that worked on the remote tick bug in here?
04:52:29 Quit paugh ("Leaving")
05:01:03 Join Jungti1234 [0] (n=jungti12@
05:04:19 Join JdGordon [0] (
05:04:31 Quit Jungti1234 (Client Quit)
05:05:40***Saving seen data "./dancer.seen"
05:06:20 Part andrewmel
05:10:26 Join Kyl3 [0] (
05:10:34Kyl3theres screendump?
05:11:03Kyl3on h300?
05:11:16BBubdoesnt work yet afaik
05:11:28Kyl3 <-look there
05:11:41Kyl3last post
05:12:00BBubmaybe it works then :)
05:12:27Kyl3how do i use it? anyone with an H100 here?
05:12:57BBubjust enable it in the debug menu and when you connect the usb-port it will take a screendump
05:12:58Kyl3how do you use it on the H100?
05:13:06Kyl3o lol
05:13:12drumRBoy320haha... "take a screendump"
05:13:48BBubit dumps the frame-buffer into a file
05:13:52BBubis that better for you?
05:14:01drumRBoy320haha, ok ;)
05:15:52drumRBoy320i came up with a good idea... when i had stock h300 firmware, i mad emy front page have personal information... and i lost that when i upgraded, so i put my personal info in an unused space in my WPS
05:16:01drumRBoy320it scrolls... looks nice
05:17:09drumRBoy320says iriver h320 *rockbox* property of ""drumrboy"" Drumrboy@drumr.drumr or (555) 555-5555
05:17:14drumRBoy320on my last line
05:17:28BBubyou think anyone would return it if he finds it?
05:17:30BBubdream on ;)
05:17:30Kyl3No screendump
05:17:51drumRBoy320haha, well, they wouldnt appreciate rockbox, so they'd give it back
05:18:07Paul_The_NerdOr they'd say "Neat, I've always wanted one of these."
05:18:10Paul_The_NerdOr just selll it on ebay
05:18:22drumRBoy320seriously, who would look at rockbox who didnt know anything and say "wow what a nice text laden screen
05:18:30drumRBoy320haha, with my info on the WPS?
05:18:59Kyl3I'm too lazy ATM to do my own WPS
05:19:08drumRBoy320mine is nice... want the code?
05:19:08Paul_The_NerdWell, if they pawn shop it, they won't care about the WPS.
05:19:11BBubi like the boxes wps
05:19:15BBubiamp is also nice
05:19:27drumRBoy320%ac Now Playing:
05:19:27drumRBoy320%ac %it
05:19:27drumRBoy320%ac Codec:%fv %fc Bitrate: %fb
05:19:27DBUGEnqueued KICK drumRBoy320
05:19:27drumRBoy320%al %pc %ac %pt %ar %pr
05:19:28***Alert Mode level 1
05:19:30drumRBoy320%ac Next
05:19:32drumRBoy320%ac %It
05:19:34drumRBoy320%ac %Ia
05:19:36drumRBoy320%ac %Id
05:19:38drumRBoy320theres a line left for personal info
05:19:52drumRBoy320i use 'chicago' font
05:20:19Kyl3And you can just type your personal info?
05:20:40drumRBoy320what do you mean?
05:20:40Paul_The_NerdKyl3, they're just text files.
05:20:59drumRBoy320yea, so i put this... %s **Personal Info**
05:21:06drumRBoy320and it scrolls across the bottom
05:21:22drumRBoy320im actually gettin kinda comfortable with WPS coding
05:21:46drumRBoy320now i gotta start integrating with graphics
05:21:53drumRBoy320hmm, but for now, im off to bed
05:24:28 Quit DreamTactix291 (Read error: 110 (Connection timed out))
05:24:47 Quit dropandho ()
05:28:27 Quit Kyl3 ()
05:29:29***Alert Mode OFF
05:40:06 Join MohLeh [0] (
05:40:43MohLehI got a question...
05:41:00MohLehIf anyone is available to help me out...
05:43:06Paul_The_NerdWhat's the question?
05:43:52MohLehNever mind, i did another quick search and found out that the h300 doesn't yet support asian fonts... -_-
05:44:05Paul_The_NerdAh, yes.
05:44:34MohLehOr is there somethign?
05:44:35 Nick CheeseBurgerMan is now known as CBM-away (n=BurgerBo@
05:45:54Paul_The_NerdThere's a unicode patch that's being tested that *might* do it. I'm not really sure.
05:45:56ashridahMohLeh: unicode support is on the way
05:46:57MohLehAlright. Thanks for answering... can't wait for it to be availabe. (most of my songs are in korean)
05:47:37ashridahMohLeh: if you're feeling adventurous, you could patch and compile rockbox with it yourself. or just get someone else to do it for you :)
05:47:55MohLehi'd try it myself, but i dont know where to start.
05:50:49 Join o0260o [0] (
05:52:49MohLehhrmm. THANKS! for the link. I will definately try and do this. ^^
06:01:36 Quit MohLeh ()
06:04:20 Part JonSenior
06:09:52 Join Lost-ash [0] (
06:09:55 Quit ashridah (Nick collision from services.)
06:10:12 Nick Lost-ash is now known as ashridah (
06:25:48 Join BBub_ [0] (
06:37:28 Quit RotAtoR ()
06:37:42 Join Membrillo [0] (
06:38:57 Quit Membrillo (Client Quit)
06:42:45 Quit BBub (Read error: 110 (Connection timed out))
06:42:45 Nick BBub_ is now known as BBub (
06:46:25 Join skorpyon_ [0] (
06:46:59 Join webguest43 [0] (
06:47:15 Quit skorpyon (Read error: 110 (Connection timed out))
06:48:04webguest43phaedrus961: when do you plan to update the unicode patch?
06:48:25webguest43i can't generate the codepage files with newer patches done by others
06:49:12ashridahwebguest43: markun's been working on the unicode patch
06:49:39webguest43ashridah: i've got the patch from him few days ago but that has some bugs
06:50:18webguest43i'm missing codepage files... which is probably the reason why i see broken chars in wps
06:50:40ashridahthey're currently working on getting it mergable, one imagines he's got a newer copy someplace
06:50:48ashridahthey've just got to organise a chainloader to load a compressed firmware on some targets
06:51:09webguest43do you think i can just get the codepage files somewhere?
06:51:52ashridahi'm not sure, i'm not that familiar with the contents and function of the unicode patch, just that they're working on getting it merged
06:52:38webguest43do you have yours patched?
06:55:50ashridahI don't use unicode support.
06:56:08ashridahi just pay attention to this irc channel, and i've noticed them discussing it (them being markun and amiconn )
06:57:02webguest43ah isee
06:57:56 Quit webguest43 ("CGI:IRC")
07:00:56 Join Membrillo [0] (
07:01:51 Join DreamTactix291 [0] (
07:05:00 Quit perplexity ("*pop*")
07:05:13Membrillocan someone explain to me how to create a temporary playlist of my root -> music folder on my h3xx? I don't want to create an .m3u file but one thats create in the playlist_control folder
07:05:41***Saving seen data "./dancer.seen"
07:06:49ashridahif your music is all in one subdirectory, highlight it in the file viewer, hold navi, and then select 'playlist' and then 'insert'
07:07:11ashridahthat'll create a temporary one. i have no idea how to acheive the same with the root directory tho
07:08:05Membrilloyeah thats great thanks
07:08:06ashridahyou'll want to make sure it's set to recurse subdirectories in the playlist options menu tho
07:08:13Membrilloso highlight music and hold navi insert playlist
07:08:19Membrilloyep sure
07:09:08 Join XavierGr_ [0] (
07:09:10ashridahit takes a tiny bit of grinding for my 3929 songs, but not too long
07:09:29Membrilloummm i just found a bug, if you try to delete a file it says "play - yes any other - no"
07:09:36Membrilloyou actually have to press navi to delete it
07:09:44ashridahas i say, easiest if all you music is in a subdir of the root
07:10:07Membrilloyeah mine is in a subdirectory on root
07:10:08ashridahit's a quirk in the button assignments
07:10:30ashridahone that's been debated, but never really changed, not sure why
07:10:57Membrillofair enough
07:11:15Membrilloit probably should be play. Gives less chance of accidently deleting something
07:12:25ashridahit doesn't matter where you put the button, it'll be pressable in some situations.
07:12:25Membrilloso how do i access this playlist now ive inserted it
07:13:09ashridahMembrillo: hold 'navi' at the WPS and select 'playlist'
07:14:15Paul_The_NerdProbably because the strings aren't changed by build target, but by language file?
07:14:31 Join andrewmel [0] (
07:14:39ashridahPaul_The_Nerd: well, there's that and that the assignments were never really well thought out, and then kinda stuck
07:15:19Membrilloso every time i want to shuffle random songs on my whole iriver i have to create a temporary playlist?
07:15:55 Quit Paul_The_Nerd (Excess Flood)
07:17:37ashridahMembrillo: rockbox is playlist oriented
07:18:19 Quit schnittn2005 ()
07:19:00Membrilloyeah, i know that. It just takes about 20 seconds to create that playlist, and im sure it will get annoying having to wait everytime i turn on the player
07:19:55Membrilloseeing i generally use shuffle all the time
07:20:38andrewmelcan you make and save a playlist with all your songs in it and just play that?
07:20:53ashridahandrewmel: yes
07:21:39Membrillohmmm yeah
07:21:46ashridahgoign to get annoying if you add/remove music regularly tho
07:21:53Membrilloand just update it everytime i add songs
07:22:13Membrilloyeah. So there isnt a function like in the iRiver firmware that automatically shuffles all music on the player
07:22:22andrewmelis there a plugin that could be run that will grab all files and put them into a shuffled list?
07:22:45andrewmelyou would then only have to run it when you added music?
07:23:36Membrilloso is that what the iriver FW does when it boots up? searches out all the songs?
07:23:58Membrilloandrewmel: something like that would be terrific although i wouldnt know how to code one
07:24:51andrewmelI could in Pascal......
07:25:14andrewmeland Java, but I dont get object oriented, I think in a data structured way.
07:26:20Membrilloi know a bit of VB. can that be used?
07:26:46andrewmeli am not sure. i think only C can be used
07:27:15Membrillomakes sense
07:27:20Membrillowhat format are the plugins in?
07:27:31andrewmelnot sure.
07:27:46andrewmelin fact i am more hear to learn than teach. i know little about rockbox
07:29:26Membrillothey seem to be in .rock format
07:29:32Membrillowhich is a variation of who knows what
07:29:45andrewmeli think it is their own
07:30:01Membrillobut it could be compiled from DLL or something
07:32:13Membrilloill probably just use that iRiver playlist editor for now
07:33:14 Quit XavierGr (Read error: 110 (Connection timed out))
07:34:16 Join Paul_The_Nerd [0] (
07:38:06Membrilloashridah: if your still here, another bug. When pause is pressed while playing, a pause image comes up, but when the song is resumed, the image doesnt disapear again
07:40:26Paul_The_NerdIn the WPS?
07:40:43ashridahMembrillo: don't look at me, i'm not a developer :)
07:41:18ashridahdoes it do that in all WPS's?
07:41:19Paul_The_NerdIf you have a Pause image in the WPS, but nothing when it's playing (IE: no play image) then it'll stay up, because that portion of the screen won't be changed until it's given something to change to, apparently
07:47:48Membrilloashridah: oops sorry:
07:47:59 Join B4gder [0] (
07:48:27MembrilloPaul_The_Nerd: well does it have to be replaced by an image? can't it just make that section not visible on play?
07:49:15andrewmelhmmm, i have a bit of a problem....
07:49:44andrewmelI dont know the names of the screens but will have a go at explaining
07:50:01andrewmelrunning v2 with todays latest build
07:50:33andrewmelwhen I boot all I get is the battery symbol, the volume and a square in the top line,
07:50:34andrewmelall on the left.
07:50:49Paul_The_NerdMembrillo: I think it's just the way the code works. If you don't tell it to change a pixel, it doesn't change that pixel. Very explicit.
07:50:59andrewmelif i press the a-b button i get the menu starting with recent bookmarks, sound settings ect
07:51:05Paul_The_NerdIf you want nothing to be there, create a white rectangle/square of the appropriate dimensions, and use that as the alternative.
07:51:09andrewmelpress it again and I am back to the start
07:51:28andrewmelthe only other button that works is the bottom left turn it off button.
07:51:32Membrilloandrewmel: hmmm it should show your root drive
07:51:32andrewmelany ideas?
07:51:44Paul_The_Nerdandrewmel: Do you have any other files on the player?
07:51:48andrewmelIt shoud\
07:51:57andrewmelyep when i boot to iriver firmware it looks fine
07:52:03andrewmeland works fine
07:52:04Paul_The_NerdHold A-B
07:52:15andrewmeli ge tthe three options,
07:52:16Paul_The_NerdWhen in Rockbox
07:52:21Paul_The_NerdWhat does it say at the bottom?
07:52:23andrewmelshuffle off
07:52:23andrewmelrepeat off
07:52:30andrewmelshow file id3 database
07:52:30Paul_The_NerdShow files?
07:52:34Paul_The_NerdChange that
07:52:36Paul_The_NerdTo supported
07:52:46Paul_The_NerdThen hit A-B again
07:52:50andrewmelhd humming.....
07:52:58andrewmelah. thanks!!!!
07:53:02Paul_The_NerdNo problem
07:53:09andrewmelwow many traps for those playing with the options :)
07:53:28Paul_The_NerdIt's a fairly common thing. The ID3 database has to be generated with an external tool, and isn't really all too 'finished' yet anyway.
07:53:32andrewmelyay, just got a loaded sudoko to work.
07:54:08andrewmelit has been a while since i asked questions about my lovely h320, i have been answering questions at misticriver for more than a year now :)
07:54:20andrewmelmuch better to not know, that means stuff to learn
07:58:28 Quit hardeep (" HydraIRC -> <- 100,000+ downloads can't be wrong")
07:58:48andrewmelI am guessing the developers know this one, but if i have the player turned off and plug the charger in it boots the iriver firmware
07:59:05 Join XavierGr [0] (
07:59:34ashridahlinus hasn't hooked that boot path yet
07:59:40 Quit XavierGr_ (Read error: 104 (Connection reset by peer))
08:00:06andrewmelmight be good to leave it?
08:16:16 Join Jungti1234 [0] (n=jungti12@
08:17:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:17:10*Paul_The_Nerd likes the new FLAC codec.
08:17:33DreamTactix291that's good
08:17:38Paul_The_NerdWith Crossfeed and Replaygain on, the only time it boosted while I was watching the audio thread was when reading from HD (of course) and during the backlight fade.
08:17:58DreamTactix291i still use wavpack :P
08:18:08Paul_The_NerdI batch processed over last night
08:18:45DreamTactix291i don't feel like doing that for a whole collection
08:18:59Paul_The_NerdMine was only about 450 songs
08:19:05DreamTactix291mine is ~5000
08:19:15DreamTactix291i'd lose 5gigs of space if i went from wavpack to flac
08:19:41DreamTactix291i've had some stuff drop almost 200kbps from flac to wavpack -hmx
08:19:43DreamTactix291but those are rare
08:19:46DreamTactix291average is 20-30
08:19:57Paul_The_NerdSee, mine were done "fast" so my wavpacks were actually bigger than the FLACs I made.
08:20:07DreamTactix291mine are mostly -hmx
08:20:07Jungti1234I have question.
08:20:43Jungti1234Why does not OGG Q10 become playing normal?
08:20:43Paul_The_NerdWhen I did lossless high they were skipping occasionally on rockbox.
08:21:04Paul_The_NerdJungti1234: Are you having it skip on an H320/340?
08:21:17DreamTactix291Paul_The_Nerd: i've never used much wavpack on my H140
08:21:28DreamTactix291it's mostly vorbis
08:21:34DreamTactix291like i said
08:21:38DreamTactix291i can't fit my lossless collection on it
08:21:45Paul_The_NerdIf I had more songs, I'd use Vorbis too
08:21:55DreamTactix291Lancer is great
08:21:56DreamTactix291so fast
08:22:00Paul_The_NerdBut right now, there's only about 1700 total, and 1300 the originals are MP3. =/
08:22:11B4gderJungti1234: because we enjoy reading people report the same problems over and over again
08:22:20DreamTactix291it's more like
08:22:23B4gdereven though we've explained why
08:22:30Paul_The_NerdJungti1234: The H320 is capped at 90mhz right now, for stability purposes. The Ogg Vorbis codec cannot quite reach realtime at Q10 right now.
08:22:37DreamTactix2914500 lossless and 1000 mp3 here on my PC
08:23:17Paul_The_NerdI'm just really slowly at finding music I like and acquring it.
08:23:27Paul_The_NerdSlow, rather than slowly.
08:24:04DreamTactix291so am I
08:24:10DreamTactix291i buy used CDs a lot
08:25:09Paul_The_NerdI do too, but I'm just not good at figuring out which to buy often.
08:25:10ashridahPaul_The_Nerd: i am too, i'll often find new stuff off friends.
08:25:20Paul_The_NerdThat's what I do too, ashridah.
08:25:34DreamTactix291i am the friend :|
08:25:40DreamTactix291so i don't have that luxury much
08:25:48DreamTactix291i'm a big prog rock fan
08:25:58DreamTactix291i look at who i like has worked with before
08:26:00DreamTactix291and try their stuff
08:26:03DreamTactix291and branch out like that
08:27:17Paul_The_NerdI wonder what determines whether something goes in the Sound Settings menu vs. the Playback menu.
08:27:36ashridahcoin flips :)
08:27:51Paul_The_NerdSeriously, "Beep Volume" seems in the wrong place to me. And *possibly* crossfade.
08:28:00Paul_The_NerdCoin flips would be a good way to do many things.
08:28:07ashridahcrossfade's a dsp effect
08:28:22Paul_The_NerdSo because of that it's in Playback?
08:29:18*Paul_The_Nerd shrugs.
08:29:27Paul_The_NerdI don't really go into "Sound Settings" ever anymore anyway.
08:30:44 Join Bger [0] (n=Bager@
08:30:46ashridahbut there has been debate about reorganisation of the menus
08:30:48Bgermorning :)
08:31:06B4gderthere's constant arguing about the menu layouts
08:31:17B4gderand my guess is that it will never stop
08:31:23ashridahB4gder: i didn't say it had a start or end. :)
08:31:30B4gderI just clarified
08:31:52Paul_The_NerdAh well. It matters not to me, really.
08:32:08Jungti1234Do you know 'Crucian carp bread'? :)
08:32:44 Join MrShlee [0] (
08:32:48Bgerhm, there was suggestion to make some kind of menu like "Play, Record, Radio, Settings, Browser"
08:33:02Bger(not in this particular order, of course)
08:34:08Bgerwhy ?
08:34:47Bgerin fact, i like this idea
08:35:21Bgerbecause i don't think that the browser is the *main* task of the player
08:35:51Paul_The_NerdThe music player is the main task.
08:35:56Paul_The_NerdBut to play music, you have to select it.
08:36:06Paul_The_NerdSo the browser is the only *necessary* prerequisite to playing music.
08:36:17Paul_The_NerdIf you've said to resume on boot, it boots straight into the music.
08:36:39andrewmeli love that aspect of it!!!!
08:36:43Bgerand if you want to go into radio (for example)
08:38:19Paul_The_NerdWhat about the radio?
08:38:29Bgeror recording screen
08:38:32Paul_The_NerdIf you want to do any other choice but play stored music, then you must go through options, yes.
08:38:51MrShleeAnyone had any problems with the H320 bootloader? is it safe?
08:38:51Paul_The_NerdIt's a music jukebox. Digital music is its primary purpose, no?
08:39:03Paul_The_NerdSafe is a very subjective word, MrShlee.
08:39:46MrShleehah :) I've got no-problems flashing it.. I mean is there a good chance of bricking :)
08:40:06Paul_The_NerdWell, I nobody's told *ME* that they've bricked theirs.
08:40:08MembrilloMrShlee: there have been no reports anyway of a bricked player
08:40:13BgerMrShlee: there *were* problems, but no bricks till now (at least i don't know for such)
08:40:22MrShleeCool.. I'll see how it goes.
08:41:05MembrilloMrShlee: i was worried about it, but i bit my tongue and went for it and its fine on my International H320 (1.28k)
08:41:07Bgerbtw, how did linus manage to almost brick his h320 ?
08:42:01Membrilloprobably fiddling witht the bootloader
08:42:09MrShleeive got an interat H320 1.28E. should be fine.
08:42:15Membrillohe can fix a bricked player though
08:42:25Membrillowith his tools :P
08:42:26Paul_The_NerdHow do you "almost" brick a player anyway?
08:42:28Bgerhm, he said his flash entered in tristate mode ...
08:43:04Bgerand was afraid that he would have to buy another player
08:43:51Membrillohe gets enough donations to buy a new one surely
08:44:13Bgeryes, he do, but ...
08:44:34Bgerwhy to waste money ...
08:45:23Membrilloits hardly wasting money. Without a player he cant develop
08:45:40MrShleerockbox.linux has to be on root right?
08:45:47MrShleeHere goes nothing.
08:45:51MrShleeor everything
08:45:54MrShleeatleast something.
08:46:08Membrilloeverything on root
08:46:15B4gderMrShlee: unpack a complete package in the drive's root
08:47:10MrShleeits booted fine. any idea how to change into FM mode?
08:47:25Membrillogood question, i was wondering that
08:47:31B4gderuse the menu
08:47:40Membrillowhich menu
08:48:08 Join skorpyon__ [0] (
08:48:31Membrillohow do you get to this menu
08:48:55MrShleeHolding down the main buttons doesn't do jack.. apart from create/rename and change the basic options.
08:49:04MrShleeeg navi or a-b :|
08:50:47MembrilloB4gder: ???
08:51:19MrShleeI want to play the plug-in games :*(
08:51:37Paul_The_NerdTapping A-B *should* bring up the menu, right?
08:51:54MrShleeit does.. with basic options
08:52:11MrShleelike shuffle, report and show files.
08:52:32MrShleeI need to figure out skin/fm and other fun stuff.
08:52:37MrShleebut so far. it looks very good :)
08:53:56MrShleeohh tap a-b
08:53:59MrShleenot press and hodll
08:55:44MrShleeThere should be a greater delay on the backlighting in an active plugin.. keeps going dark after 5 seconds..
08:56:04B4gderyes, the h3x0 needs much longer timeout
08:56:06BgerMrShlee: General Settings->Display->Backlight
08:56:07Paul_The_NerdYou can change the delay on the backlighting.
08:56:09B4gderbe default
08:56:21B4gdersilly colour LCD ;-)
08:56:26Paul_The_NerdI often set it to be "always on" while I'm playing a game, so I can stop and think without it fading.
08:56:56Bgerhm, Paul_The_Nerd, you have h1x0, haven't you ?
08:57:14Paul_The_NerdBut I often like a well lit screen.
08:57:16Bgerwith h3x0 it's worse ...
08:57:25Bgermorning, ami
08:57:43Paul_The_NerdI've seen how bad it is with an h3x0 Bger. That's why I prefer monochrome/grayscale screens. :)
08:57:47Paul_The_NerdMornin' amiconn
08:58:29B4gderimho, we should switch off the lcd completely instead of just switching off backlight on the h3x0
08:58:31*Bger uses a lcd remote and this lessens some of the badness ...
08:59:00B4gdersince the lcd is useless without it
08:59:37Paul_The_NerdCan you switch it to a monochrome mode?
08:59:46Paul_The_NerdSince WPS and all are currently monochrome anyway?
08:59:46B4gderI doubt that
08:59:51Bger? the display?
09:00:02B4gderand "currently" is the word here
09:00:09Bgereven if we do it, it again won't be visible
09:00:42 Quit Jungti1234 ("Bye~ -")
09:00:57*Paul_The_Nerd shrugs.
09:01:42Membrillohow do you get to Radio from the a-b menu
09:02:01BgerMembrillo: afaik it's not enabled yet on h3x0
09:02:27MrShleeI still can't access fm mode..
09:02:40Membrilloi figured thats what you meant by FM mode
09:03:24Membrilloin sudoku, when you press a-b an input thing comes up. whats that for
09:03:56MrShleeDoes the current build allow for battery % and time left?
09:04:18 Join Zagor [0] (
09:04:35MembrilloMrShlee: yep
09:04:49 Quit skorpyon_ (Read error: 110 (Connection timed out))
09:05:20Membrillocan anyone beat level 3 on star??? gar!!!!!!
09:05:45***Saving seen data "./dancer.seen"
09:06:26 Part Sando
09:06:53MrShleethis is very impressive so far :)
09:06:55MrShleeGood work team.
09:07:16 Join ender` [0] (i=ychat@
09:08:47Membrillosomeone see if they can do level 3 on star
09:08:50Membrilloits impossible
09:09:23 Join amiconn_ [0] (
09:09:50 Quit amiconn (Nick collision from services.)
09:09:50 Nick amiconn_ is now known as amiconn (
09:11:29JdGordonhey all
09:11:44JdGordonis there a way to convert the c array for a bmp back into the bmp?
09:12:03Membrillobejeweled is in colour
09:12:11Bgeryes, very nice :)
09:13:56MrShlee1 and 2 were easy.
09:13:59MrShleegive me a few minutes
09:14:33 Join DMJC [0] (
09:15:24MrShleethats impossible..
09:15:35Membrilloyeah thought so
09:15:39Membrillosomeone fix star lol
09:15:44Membrillo3rd level is impossible
09:16:40JdGordonwould any1 like to draw numbers for me :D
09:17:15ashridahfor what?
09:17:31JdGordonbloody sudoku :p its done buy t the numbers for the h300 are but ugly
09:17:47 Join Jungti1234 [0] (n=jungti12@
09:18:43Paul_The_NerdThe third level of star is not impossible
09:19:02Paul_The_NerdYou have to use the other block too. Move it so that it can stop you.
09:19:31MrShleemove it?
09:19:41 Quit skorpyon__ (Connection timed out)
09:19:50Paul_The_NerdThere is a button, I can't remember which, that switches your control from to the square
09:23:14Membrillocan you save in star?
09:23:30 Nick Lynx_awy is now known as Lynx_ (
09:23:39Paul_The_NerdI don't think so. :(
09:26:04MrShleebah.. 1 left
09:26:20MrShleebastard circle come-at-thee!
09:27:18 Join Bgr [0] (n=Bager@
09:28:15MembrilloDone it!!!!
09:28:43MrShleedone it!
09:28:44Membrillothis levels a bitch
09:29:45MrShleeBattery test is broken tho :(
09:30:04Membrilloummm on this next level ive got myself stuck
09:30:41Membrilloi really cant do a thing
09:31:06JdGordonany1 got the ondio?
09:32:50JdGordonhmm.. the ondio is seriously lacking buttons...
09:33:53 Join einhirn [0] (
09:38:44amiconnbejeweled in colour is fun :) (simulator)
09:38:53Bgryes :)
09:39:09amiconnJdGordon: I have two Ondios
09:39:09Bgrbtw, the color of the circles is a bit anoyng (IMHO)
09:39:20Paul_The_NerdWhat color are they?
09:39:21amiconnIt's a bit too bright
09:39:38Bgruf :(
09:40:34amiconnThere's also a small bug in the colourised bejeweled
09:41:06amiconnObviously the colour doest't get reset at the end (before the 'no more moves' splash)
09:41:06Bgrthe position of the marking square ? this splash appears in the colour of the last drawn jewel, and the main menu too
09:42:46 Join swunk [0] (
09:44:34 Quit Bger (Read error: 110 (Connection timed out))
09:46:27 Part swunk
09:48:34 Quit Membrillo ()
09:49:43MrShleeThere are a few delays with the gui. I'm guessing their arnt perfect threading in the irivers yet?
09:51:05B4gderthe threading is fine
09:51:23B4gderwhy do you think it is because of threading?
09:52:31Slasheriprobably he thinks the threading is pre-emptive but it's not..
09:52:42Slasheriso some other thread takes too much cpu
09:52:53MrShleebecause if I try and navigate folders the thing freezes for a matter of seconds. sometimes skips and audio and the charging image sticks until the gui is finally freshed.
09:52:57*Bgr suppose the LCD updates on h3x0 ...
09:53:07MrShleeooh k
09:53:16BgrMrShlee: this is the HDD
09:53:25Bgrjust enable the dircache
09:53:44Bgrhm, w8
09:53:55BgrMrShlee: does it happen when u scroll much ?
09:56:03MrShleedircache seems to fix it.
09:56:23 Join thegeek_ [0] (
09:57:40 Join Polo_o [0] (
09:58:59Bgrbtw, there is a lag when you scroll a lot ... even the audio stops sometimes (and the marker continues to go after u release the button)
10:00:08 Join LinusN [0] (
10:00:10linuxstbJdGordon: Why don't you use the text drawing functions of a bitmap editor to generate the numbers for Sudoku?
10:00:17MrShleeId love FM tho
10:00:54BgrMrShlee: btw, there are combinations to scroll one screen page at a time
10:01:09BgrLinusN: morning :)
10:01:33Bgrdid you see my last commnent on
10:01:37JdGordoncoz that would would be obvious :p
10:02:21JdGordonhowdy LinusN, huge thanx are in order for you :D i didnt touch the oriignal iriver firmeare all today :D
10:02:53MrShleeYeah, I've just flashed the H340.. Very impressive :)
10:02:58 Join skorpyon [0] (
10:03:13 Join DocBrown [0] (
10:03:26JdGordonwhats the lcd hieght of the archos recorder?
10:04:28linuxstbIt's 112x64
10:04:47JdGordon:'( i touched something and im getting unsupported lcd hieght on the archos rec
10:05:31*JdGordon slaps self again...
10:05:38LinusNBgr: saw your changes now
10:05:40JdGordonstupid #else went byby
10:06:06BgrLinusN: okay...any comments ? :)
10:07:32 Quit DMJC (Read error: 104 (Connection reset by peer))
10:07:40MrShleeLinusN : FM support is missing from the Iriver builds. any idea when they should be updated?
10:07:50LinusNfunny that the play/pause was moved
10:07:57 Join Sando [0] (
10:08:24JdGordonLinusN: when hold is on (on the h300's) does the hardware values of buttons not change? or does the software just ignore the changes?
10:08:31Bgryes, this is the main inconsistency between the main unit's and the remote's buttons
10:08:51Bgr(the fact that u power on it with navi on the remote, not with play/pause)
10:09:07JdGordoncoz i get really annoyed when i accidently stop my music when im playing and it would be awesome if the hold button could be used to just disallow any changes to the playing music.. but still let u play games and use menus
10:09:18Bgri also submitted a patch for the h300 remote in the patch tracker
10:10:39Bgrbtw, LinusN, what do you think about adding USB support asap for h300 ... switching to iriver fw is so annoying .. :)
10:10:47Rickanyone know of any online shops selling h3x0's?
10:10:56 Quit thegeek (Read error: 110 (Connection timed out)) ?
10:11:01*Rick looks
10:11:37Rickthanks Bgr, looks nice
10:11:57RickUK though D:
10:12:09MrShleelol LinusN has left the bulding. hes been flooded :P
10:12:15B4gderhey, buying an already supported device is cowardly, get an unsupported and port rockbox! ;-)
10:12:21JdGordonits tough being popular
10:12:24BgrB4gder hehehe
10:12:28RickB4gder: like what? :P
10:12:42B4gderiaudio, ipod, gigabeat, ...
10:12:53 Join [IDC]Dragon [0] (
10:12:59Rickdon't like any of those
10:13:17B4gderbut you like the h300?
10:13:18Bgrbtw, gigabeat looks very promising ...
10:13:26RickB4gder: sure (:
10:13:29Rickcolor LCD? :P
10:13:40B4gderRick: all the ones I mentioned have color LCDs
10:13:42Bgrand more powerful CPU
10:13:55*amiconn spots a rare guest :)
10:14:01Rickrare guest? where? D:
10:14:09[IDC]Dragonhi guys
10:14:22[IDC]Dragon1st day back to work
10:14:39[IDC]DragonI have no DSL at "home" any more
10:14:46[IDC]Dragonfor 4-6 weeks
10:14:53B4gderwhat a blow
10:15:03[IDC]Dragonlet alone time for rockboxing
10:15:16LinusNBgr: hmmm, my non-lcd remote doesn't show any signs of life except for the play/pause key
10:15:24[IDC]Dragonour new house is a mess atm
10:15:45Rickhas anyone figured out a way to distinguish remotes yet?
10:16:05LinusNMrShlee: i haven't started with the fm radio support yet
10:16:10[IDC]Dragonwhat's been happening during the last 2 weeks?
10:16:24LinusNJdGordon: the buttons are filtered in the button driver
10:16:29BgrLinusN: that's what i said too
10:16:43Nixsosi want to report that my intl H340 isn't working with Firmware V2. Disk spins up, but i get a black screen. is anyone working on this problem?
10:16:43LinusNusb support is my next task
10:16:49*[IDC]Dragon spotted something about iriver 3x0
10:16:59LinusNNixsos: i wish i could work on it
10:17:00BgrNixsos: try resetting it
10:17:21LinusNbut i can't make it happen on my h320
10:17:33Rickgigabeat looks nice
10:17:39Nixsosi already tryed everything. even the hold on, tap reset twice while pressing the on button trick
10:17:41BgrRick: no remote tho
10:17:43Nixsosdidnt work
10:17:44Rickoh damn
10:17:46Rickthat sucks
10:17:48*LinusN has ordered an iaudio x5
10:17:56BgrLinusN: go go go :)
10:18:06[IDC]Dragonis that the display-less one?
10:18:16NixsosBootloader V1 worked btw
10:18:17LinusNcolor lcd
10:18:19Paul_The_NerdNixsos: Does reset make *anything* happen?
10:18:30LinusNNixsos: v1 worked but not v2?
10:18:42NixsosLinusN: true
10:19:00Nixsosweird huh
10:19:12RickThe only trouble is that I've got no skills in writing bootloaders
10:19:20Rickso if I did get something I'd have to wait for someone else to start first :P
10:19:51NixsosPaul_The_Nerd: yeah, just shuts the device off nothing else... i'm able to boot original FW when the USB cable is plugged in
10:20:15RickLinusN: what's the iaudio x5 like?
10:20:15Bgrmarkun: how's gigabeat's battery time
10:20:25BgrRick: very like h100 and h300
10:20:29LinusNRick: haven't received it yet
10:20:30Paul_The_NerdNixsos: can you boot it by holding record AND play when you turn it on?
10:20:33B4gderBgr: it claims 16 hours
10:20:42Paul_The_NerdNixsos: I understand you have to hold them for several seconds.
10:20:48NixsosPaul_The_Nerd: nope, won't do a thing
10:20:51amiconn[IDC]Dragon: Btw, back when you wrote the flash bootloader, did you have problems with the ucl decompressor?
10:20:52*Rick looks
10:21:05RickHow much? (:
10:21:19amiconnThe thing is, I now made a selfectractor, but it doesn't work reliable so far
10:21:22LinusNNixsos: do you have a flashlight?
10:21:32amiconnSometimes it works, sometimes it just hangs
10:21:35Rick160x128 LCD :<
10:21:43NixsosLinusN: no.. would i be able to read the screen with that?
10:21:58NixsosLinusN: so it could be a problem with the backlight..
10:22:02BgrRick: just like the h100's resolution
10:22:10amiconnIt's quite hard to debug such low-level stuff, with no LCD output...
10:22:11Rickyes I know
10:22:15RickI want something bigger D:
10:22:20NixsosLinusN: anyone else had a simmilar problem?
10:22:35JdGordonLinusN: JdGordon: the buttons are filtered in the button driver <- does that it would be possible to use the hold button to only allow menus and games?
10:22:44Nixsosofew, so i'm not on my own then :d
10:23:13LinusNJdGordon: yes
10:23:37JdGordon.. could it be done? or too much work to worry about it?
10:23:56BgrJdGordon: firmware/drivers/button.c
10:24:27JdGordonright... thats what im playing with next
10:24:59LinusNJdGordon: i think it's too much work
10:25:11NixsosI'm sad... i want to write a plugin. it would be great to be able to test it on my player. but i guess i'll have to use the simulater until a new bootloader is out :(
10:25:17Rickx5 looks nice but the LCD is not very large :(
10:26:09JdGordonlinuxstb: the scratchpad thing is done on all players cept the ondio coz no free button
10:26:30*JdGordon is gone
10:26:35 Nick JdGordon is now known as Jd|away (
10:29:07RickLinusN: which x5 did you get?
10:29:09[IDC]Dragonamiconn: the decompressor worked flawless out of the box
10:29:19LinusNRick: haven't received it yet
10:29:28LinusNit's the 20gb version
10:29:37[IDC]Dragonwell, I substantially thinned it out
10:29:43Ricklooking at the 60gb one
10:29:56[IDC]Dragonafk, meeting
10:30:41linuxstbJd|away: Maybe amiconn can suggest a free button combo if you explain how it works on the other targets.
10:32:29amiconn[IDC]Dragon: I copied the decompressor from flash/bootloader/bootloader.c almost verbatim
10:32:49 Quit DocBrown ("CGI:IRC (EOF)")
10:32:58amiconnOnly thing is that I used standard datatypes
10:33:52LinusNfeels unnecessary to build uclpack for the iriver builds
10:34:35B4gderbut I didn't see any quick way to avoid that
10:34:47ashridahwould it improve boot time?
10:34:50*ashridah assumes no
10:35:01amiconnThe strange thing is that it works with my (almost-)cvs recorder v1 build, but when I try unicode-patched recorder or cvs player it hangs
10:35:39amiconnIt seems some more LED-blinking debugging is due...
10:35:42*B4gder has 40+ files changed for gcc4 warning removals
10:37:12amiconnB4gder: Speaking about gcc4 warnings - I wondered about some casts you introduced
10:37:38B4gderremove the -f option and try out ;-)
10:38:29B4gder-W even
10:38:33amiconnI mean e.g. this one:
10:39:00amiconnIt now takes a char* but returns an unsigned char* ???
10:39:57B4gderI personally am against the very frequent use of unsigned char * that we do, very often without a useful purpose
10:40:25amiconnIt seems we need to decide on our global char format
10:40:38B4gderchar pointer
10:41:03amiconnEhm, the two are related, aren't they? ;)
10:41:11B4gderthey are, sure
10:41:25linuxstbThe string.h functions use plain "char". But does unicode (utf8) require unsigned char for strings?
10:42:21amiconnImho using unsigned char would be better, if for portability
10:42:36amiconnCould also be explicit signed char of course
10:42:57linuxstbIn which case, we should change string.h to use unsigned char - which isn't POSIX.
10:43:04BgrLinusN: btw yesterday i've got very strange behaveour from mkboot under cygwin (DevKit from bluechip) ... i didn't recieve the correct md5sum from the patched hex ...
10:43:24Bgryes :(
10:43:26Bgrsee logs ...
10:43:37Bgror i'm really dumb (maybe this is the reason:))
10:43:47LinusNmost likely :-)
10:44:10*amiconn wonders why the C language designers decided to introduce this ambiguity
10:44:54amiconnAll integer types are signed by default, only char is ambiguous
10:44:57BgrLinusN: see yesterday log 23.51.17
10:45:07LinusNi see it
10:45:28amiconnlinuxstb: Hmm. Then we have 3 options
10:45:30linuxstbWe could use the int8_t and uint8_t types if we specifically need a numeric 8-bit value, leaving char for strings.
10:45:41amiconn(1) Deviate from posix and go for unsigned char
10:46:03 Join KN|stiff [0] (i=phhome@
10:46:08amiconn(2) Make plain char the default, and be aware of possile catches
10:46:15amiconn(3) Live in casting hell
10:46:25B4gderhow is 2 and 3 different?
10:46:47B4gderyou mean (2) is ignore the warnigns?
10:47:02BgrLinusN: can u give me the md5sum of bootloader-h300.bin
10:47:13amiconn(2) is to convert as much code as possible to use plain char instead of unsigned char
10:47:38Bgri'll check it tonight (it's on other machine)
10:47:54Bgr10x a lot
10:48:21amiconnB4gder: Like in the bidi.c example: bidi_l2v() should then return a char *
10:48:42B4gderI'm in favour of approach (2)
10:49:09amiconnBtw, I'm still not convinced that gcc4 is worth the hassle, but that's a different matter
10:49:24B4gderimho, its not a question about gcc4 or not
10:49:34LinusNthe downside of (2) is of course that the code isn't as portable as (3)
10:49:45B4gderLinusN: why?
10:49:52B4gderwe still need lots of casts in (2)
10:50:13LinusNi assume (2) requires a gcc switch for the default signedness of char, right?
10:50:18linuxstbI thought the idea was that we use plain char when we know either signed or unsigned will work.
10:50:23amiconnImho (2) doesn't make sense when too many casts survive
10:50:46B4gderwell, the diff between 2 and 3 is only that 2 is an effort to clear up the mess while 3 would be to just add casts
10:50:46amiconn(3) is plain ugly
10:50:55LinusNor did i misunderstand?
10:51:21B4gderI think of (2) as linuxstb said
10:51:48LinusNok, so (2) is to not assume signedness at all
10:52:11B4gderyes, and to use plain char * whenever possible
10:52:14LinusNthen i vote for (2) as well
10:52:24*LinusN makes coffee
10:52:34preglowgood idea
10:54:08linuxstbFor numeric values, I would prefer to explicitly state the signedness (so people are clear that "char" is not always signed). For strings, just use char.
10:54:35B4gderbut in general, numericals should not use chars in the first place ;-)
10:54:58amiconnImho they should if it is to save space
10:55:00linuxstbI agree - it's just used in arrays to save space.
10:55:13preglowlinuxstb: if we are to do that, i'd rather we just used int8_t and uint8_t
10:55:16preglowunsigned char is long :>
10:55:37amiconnDo you care about this?
10:56:04amiconnstatic const unsigned char * const is long...
10:56:59preglowi've always wanted to use the inttypes.h types, at least it forces you to consider how many bits you truly need
10:57:13preglowand is portable
10:57:45linuxstbI agree - but only if you need a specific sized variable. Most of the time a plain int is best.
10:57:49B4gdersure, but it also fool people to use 8-bits instead of ints
10:58:09B4gderit happens in all projects
10:58:45amiconnYes, often the native int size is best
10:59:01preglowB4gder: how can you be fooled into doing that?
10:59:20amiconnThat's how my latest optimisation for the grayscale libb worked - unsigned int instead of unsigned char saves 3 instructions in the loop
10:59:41preglowint is almost always best, but you do often need strict control of size
10:59:47amiconnLinusN: Btw, did you try my lcd drawing optimisations?
11:00:02LinusNhaven't had time, sorry
11:00:26amiconnIt surely works, and it surely is faster - the only question is how much faster?
11:00:37B4gderpreglow: just people thinking "oh my cool counter goes from 0 to 53, that fits in 8 bits so I use an 8 bit variable type"
11:00:50LinusNamiconn: does it matter how much faster it is?
11:01:00amiconnHmm, not really...
11:01:11amiconnYOu think I should just commit it?
11:01:15LinusNdo that
11:01:19Bgramiconn: commit, commit :)
11:02:12LinusNin fact, being forced to choose between signed or unsigned is often not a good thing
11:02:21amiconnLinusN: Hmm, still no installer builds? Didn't you get my message concerning the test?
11:02:34 Quit ashridah ("Leaving")
11:02:47LinusNin most projects i've worked in, it has made the code larger and slower
11:03:05LinusNi got your message, and i thought the builds worked
11:03:28LinusNdamn, what's wrong?
11:03:54SlasheriBagder: yes, that is a common thing with devices with very limited iram.. Just for example some avr microcontrollers which have only 0.5 - 4 KiB on internal ram
11:04:24B4gderits common on all devices
11:04:30B4gderI see it everywhere
11:05:03Slasheritrue.. i also use always char on avr when possible but not so often on devices with more ram
11:05:27B4gderSlasheri: but often using ints (on stack) makes smaller code than using chars
11:05:40Slasherithat's interesting
11:05:49***Saving seen data "./dancer.seen"
11:06:30amiconnRegarding signedness, the SH1 is quite special
11:06:57 Part Polo_o
11:07:06amiconnUsually signed makes smaller code, because the SH1 sign-extends by default, but for variables that are shifted a lot it's better to use unsigned
11:07:29amiconnShifted by fixed amounts that is
11:08:48markunJungti1234: morning
11:12:47andrewmelbye all, tomorrow :)
11:12:50 Quit andrewmel ()
11:13:46MrShleeWeird.. my battery is fulled charged on original firmware but 88% on the rockbox.. how does it calculate batt life?
11:15:07LinusNit calculates it from the battery capacity and the assumed power consumption
11:15:51MrShleeok I'll let the thing keep charging in Rockbox. See if it hits a peak
11:16:04LinusNbtw, did you notice that the "Battery level" column on IriverPort is empty?
11:16:19LinusNi have never tried charging in rockbox, i have no idea if it works or not
11:16:54MrShleeoohh bonus. I'll see if its ok.
11:19:21 Join Bger [0] (n=Bager@
11:22:02MrShleewell if detects that its 98% filled with 8 hours life..
11:22:08MrShleeDoesn't seem to charge tho
11:22:32MrShlee*9 hours
11:23:21preglowunit needs to be switched on for charging to work on h3x0?
11:23:24amiconnLinusN: Maybe I did misunderstand you last week. The H3x0 has no ATA LED at all?
11:24:19LinusNno leds whatsoever
11:24:31amiconnHow does it indicate charging?
11:24:43 Join Polo_o [0] (
11:24:44LinusNlike the archos recorder
11:24:48B4gderwow, an installer!
11:25:06LinusNB4gder: it doesn't work when run from cron :-(
11:25:17B4gderLinusN: probably a permission thing
11:25:22 Join mirak [0] (
11:25:38B4gderLinusN: I'll check tomorrow's logs
11:25:53LinusNcheck todays logs
11:25:58LinusNor yesterdays
11:26:01B4gdercan't, already gone
11:28:12amiconnLinusN: How do you mean like the archos recorder? v2/fm?
11:28:29amiconnI presume charging is handled in hardware for safety reasons...
11:28:54LinusNthe player turns on and displays a charging animation, like the recorder
11:29:07LinusNthat's how it indicates charging
11:29:50LinusN"the player" == h300
11:29:54amiconnI think there must be a method to read the charging state from the charging controller
11:30:06LinusNyes there is
11:30:17LinusNwe can control and monitor lots of things
11:30:25amiconnunlike the H1x0 :/
11:31:21amiconnIs charging monitoring already implemented?
11:31:28 Quit CoCoLUS (Read error: 104 (Connection reset by peer))
11:31:40markunamiconn: no luck with ucl?
11:31:52amiconnNot really...
11:31:53LinusNi haven't worked anything on the charging
11:32:04 Join CoCoLUS [0] (
11:32:09Bgerso does it charge or not under rockbox ?
11:32:16LinusNnot, it seems
11:32:38Bgerbtw, the "charging" indicator (the fulling of the battery) works
11:33:35mirak47 ?
11:33:40BgerLinusN btw did you tryed other freq between 90MHz & 124MHz ?
11:33:42MrShleeyeah.. it detects the voltage but doesn't seemt to indeciate if it actually chargers or not.
11:33:53LinusNBger: yes
11:33:57Jungti1234LINUX install stopped.
11:34:24BgerJungti1234 what linux
11:34:36Jungti1234redhat 9.0
11:34:43mirakdo you think it should be possible with rockbox on H300 to have USB charging while connected as UMS on a PC ?
11:34:52mirakactually you can't have both
11:35:03Bgerat least with iriver fw
11:35:09LinusNMrShlee: can't have both?
11:35:20LinusNi don't see why
11:35:23 Quit Bgr (Read error: 110 (Connection timed out))
11:35:36Bgerit doesn't show that it's charging, at leas
11:35:50 Join linuxstb_ [0] (
11:36:10 Quit linuxstb (Nick collision from services.)
11:36:12mirakBger: I have seen a moded cable that can do it. There is a cable that plug into the alimentation hole. this little plug just extract of the usb cable
11:36:13 Nick linuxstb_ is now known as linuxstb (
11:36:38mirakBger: and it doesn't charge, the battery cgoes down very fast when doing transfers
11:37:13Bgermirak: i think the problem is that the hdd uses much current (especially when reading/writing)
11:37:13mirakthat's why my guess is that it doesn't
11:37:26LinusNif it isn't charging in usb mode, i believe iriver tries to be on the safe side, since charging and powering the hard drive at the same time would exceed the USB current limit
11:37:34 Join ashridah [0] (
11:37:45LinusNjust a guess
11:37:55miraklike I said I have seen on an online store, a usb cable that do it
11:38:01mirakI wonder if anyone tried that
11:38:12mirakI search it
11:38:17LinusNit sure looks like the h300 is charging
11:38:25LinusN(in rockbox)
11:38:26BgerLinusN: back to my favourite remote topic: do you have any ideas about non-lcd remote and distinguishing remote types ?
11:38:35LinusNnot yet
11:38:57LinusNrockbox thinks my non-lcd remote has Hold active
11:39:01Bgeris there any free pin on the h100 lcd remote connector
11:40:22amiconnLinusN: It seems we really need to detect the remote type somehow
11:40:29mirakmmm the H300 are discontinued ?
11:40:35LinusNeither that, or add a setting
11:40:36mirakby iriver ?
11:40:37Bgermirak: seems so
11:40:43amiconnThe only remote with an 'active' hold switch is the H1x0 remote
11:41:11amiconnI would prefer detection. Iriver does it so it must be possible
11:41:19LinusNamiconn: i prefer that too
11:41:39LinusNbut until we find out how, i suggest we add a setting
11:42:03Bgeri can do this tonight ...
11:42:10Bger(at least i can try to do it)
11:42:13LinusNfind out how?
11:42:18Bgernoo, add a setting
11:42:36amiconnBah, yet another setting :/
11:42:37Bgeri whish i could find out ...
11:43:25Bgerhm, we should use different icons for remote and main unit holds
11:43:45LinusNamiconn: ok, let's not add another setting
11:44:04LinusNlet's wait with the h300 remote support
11:44:18mirakI got it −−- >
11:44:29*Bger is using the remote all the time :(
11:44:31mirakit's retractable and supposed to powersupply at the same time
11:44:50mirak15$ ahah
11:45:08mirakminiSyncâ„¢ is the single cable that will simultaneously charge and synchronize your handheld, all in an ultra-portable form factor!
11:45:30mirakthat's interesting but a bit expensive. I like the size factor. My cable is way to long
11:45:54LinusNyup, the h300 charging works
11:46:01 Join muesli- [0] (n=muesli_t@
11:46:19LinusNit won't stop the animation when it is finished, though
11:46:33mirakLinusN: is the bootloader working for european H300 ?
11:47:01mirakis it to early to not risk a H300 bricking ?
11:47:04LinusNi have heard reports of people running it on their eu model
11:47:15LinusNthere will always be a risk
11:47:30 Quit KN|stiff ("( :: NoNameScript 3.81 :: )")
11:47:34*amiconn ponders to visit Saturn and buy a H3x0 right away...
11:47:41mirakas long as its the same than on H120 it's ok ^^
11:47:46preglowamiconn: ipod!
11:47:46Bgeramiconn: just do it ;)
11:48:00LinusNas i see it, the risk is only the flashing itself
11:48:05amiconnNo iPod for sure
11:48:25Bgeramiconn: why? it's interesting target ...
11:48:43amiconnNot really
11:48:57amiconnFirst I don't really like the design
11:49:03Bgerwhy? arm, dual core ... :)
11:49:11preglowbut you like the archos design? :-)
11:49:20Bgerit's massive :)
11:49:28amiconnYes, but it looks technical
11:49:43preglowi think that's negative, heh
11:49:55amiconnThen, portalplayer isn't documented
11:49:57preglowanywho, i don't really care I can't check datasheets and correct other people's mistakes ;)
11:50:17preglowh3x0 would probably be a better choice
11:50:39preglowlinus being the only other dev i know about that has one
11:50:43amiconnFinally, it's more fun to show off what limited hardware can do
11:51:39amiconnCoding for powerful hardware can lead to laziness, and is often just plain boring
11:51:45crwlimho H300 is a huge brick... buy an iAudio X5 :)
11:52:46preglowi like the hardware just powerful enough
11:52:59preglownano seems to fit me well there, too bad it doesn't have a line in plug :/
11:53:29mirakanyone have a flexiskin from Boxwave ?
11:53:50Paul_The_NerdWhat's the process for putting rockbox on the Nano. (I don't have one, I'm just curious how you install outside firmware.)
11:53:57mirakI have seen on amzon they sell a bundle with the retractable cable plus a flexiskin, that's interesting
11:54:05amiconnThe nano _would_ be interesting (in black) - if it wasn't as susceptible to scratches...
11:54:12preglowPaul_The_Nerd: you pretty much need to run linux
11:54:27preglowamiconn: if that is an issue, then stay far, far away from all ipod
11:54:41preglowi treat my daps very gently, and even then, the nano has become scratched
11:54:58amiconnUnlike your H1x0 I presume?
11:55:00Bgerhm, is there any reason why my dircache sometimes is running, and sometimes isn't
11:55:09preglowmy h1x0 is pretty much in perfect condition
11:55:44preglowBger: do you sometimes upload huge amounts of files?
11:55:50Bgerno :(
11:55:51LinusNBger: no, it should always work
11:55:54preglowthen i have no idea
11:56:02Bgeronly updating rockbox
11:56:02preglowtalk to slasheri
11:56:05Paul_The_Nerdpreglow: You mean linux as a PC OS, or iPodLinux?
11:56:10preglowPaul_The_Nerd: pc os
11:56:19Paul_The_NerdAh. Why's that?
11:56:22preglowPaul_The_Nerd: you need to manipulate a partition directly
11:56:30preglowPaul_The_Nerd: you CAN do it from windows as well, i just wasn't able to
11:56:39Paul_The_NerdI dual boot.
11:56:50Jungti1234iAudio A2 is good.
11:56:52preglowPaul_The_Nerd: that's just for the bootloader part, after you've got that in, you can just whatever os you like, you update rockbox just like on the other units
11:56:58amiconnMy archos recorder is now more than 3 years old, and the only sign of wear is that some of the silver colour peeled off the buttons, and around the USB socket
11:56:59BgerLinusN: it was saying : Cache initialized: No; Scanning Took: 0 sec
11:57:04Jungti1234It's PMP.
11:57:07Paul_The_NerdWell, my dekstop does. My laptop seems to have issues relating to the wireless card under linux, still. :(
11:57:23Bgerthe unit powered off, i powered it on again and now it's ok
11:57:23preglowBger: i'd say go poke slasheri a bit
11:57:41Jungti1234My friend bought it.
11:57:59Jungti1234Very nice. :)
11:58:02preglowamiconn: and to make matters worse, the only protective case i know of that's available for the nano (the official apple ones), don't have a screen protector
11:58:15preglowamiconn: and to make matters worse again, you need to buy these cases in 5 packs!!!
11:58:26 Join KN|stiff [0] (i=phhome@
11:58:31amiconnwow indeed
11:58:49*Bger can't comment this
11:58:49*amiconn just reinstated blink_byte()
11:59:18amiconnHopefully this helps to track down the decompressor problems...
11:59:20Paul_The_Nerdpreglow: There's another one available. It's designed for a keychain, and it's very solid metal. But still no screen protector.
11:59:31preglowPaul_The_Nerd: that's so incredibly braindead
11:59:35preglowthe only thing i care about is the scree
11:59:45preglowthe other parts can be as scratched as they like, i never look at it anyway
11:59:53Paul_The_NerdI agree completely.
12:00:25 Quit muesli- ("ich will Kühe!!!")
12:00:46mirakso anyone have a flexiskin for is H300 ?
12:01:50Paul_The_NerdI'm thinking about getting this for my H120:
12:01:58markunpreglow: zcover has silicon covers with screen protectors:
12:01:58Paul_The_NerdIt looks pretty nice
12:02:34Paul_The_NerdThey've got iSkins with screen protectors for the Nano as well, now that I look at it
12:03:02Paul_The_NerdEven have port covers to keep out dust.
12:03:11mirakPaul_The_Nerd: seems there is a screen protection
12:03:23markunzcover was a manufacturer for iskin, now they make their own covers which look exactly the same. The was a law suit, but iskin lost it.
12:04:11Paul_The_Nerdmirak: The thing without screen protection is a keychain case for the iPod Nano, if I could remember where I lost the link. Heh.
12:05:21mirakis there in rockbox something that allow mp3 sorting by name, but with skipping of "The" in the begin of names
12:06:06mirakok that's stupid
12:06:41 Quit phaedrus961 (Read error: 110 (Connection timed out))
12:07:33preglowmarkun: hmm, thanks
12:08:45ashridahmirak: not that i know of. one could probably make a viewer plugin for m3u files that did that tho.
12:09:18ashridahi suppose it could be shoehorned into the id3 database, if anyone gets around to fixing it (assuming it's actually broken)
12:09:47Jungti1234I am going to send email to you.
12:10:17amiconnashridah: This would be quite language specific...
12:10:17mirakI was wondering if there was a way to have iripdb run automatically when unmounting the H300 on linux. I did a little unmount script, but I must click on it. I wonder if anyone ever though about that
12:10:40mirakI also wonder if anyone though of adpating gtkpod to iriver H300 or other dap
12:10:58mirakif anyone heard of such project ...
12:11:29LinusNmirak: why would we do that?
12:11:44ashridahmirak: why would you want gtkpod? rsync works just fine ;)
12:11:47mirakLinusN: wich one ? mount or gtkpod ?
12:11:56mirakashridah: I like graphical stuffs
12:12:00SlasheriBger: How much you have files? Try remove something temporarily to get the cache working and then check the cache size
12:12:14mirakLinusN: well not necessarily people on this channel ;)
12:12:21BgerSlasheri: it sometimes works, sometimes doesnt ...
12:12:37Bgerthe only files i have changed recently is the .rockbox itself
12:12:45SlasheriBger: what do you do to make it not to work? USB connection and you change some files?
12:12:47MrShleeSF has a few projects for IRIVER firmware.
12:12:49MrShleemight be worth a look.
12:12:52mirakLinusN: well I know gtkpod was first made because ipod is not UMS for is music, but well synchronisation features are ncie I think
12:12:54SlasheriHmm, weird
12:13:08mirakLinusN: nice
12:13:15BgerSlasheri: if i find how to reproduce it, i'll tell u
12:13:26Bgerdon't think about it for now
12:13:36LinusNmirak: i see
12:13:44SlasheriBger: If you can, try compile rockbox with logf enabled and check from the log what went wrong
12:14:06Bgeri can, but not now
12:14:10Slasheriok, good :)
12:14:11mirakLinusN: I shuld try to do that myself. gtkpod uses some external libs, I will see how modular it is
12:18:49Bgerwhere should "dump rom contents" go ? :)
12:19:29amiconnIt does what it says
12:19:41SlasheriBger: /internal_rom...
12:19:56BgerSlasheri: that's what i asked, 10x ;)
12:20:38SlasheriBger: you could have checked that easily from debug_menu.c too.. :)
12:20:54Bgerheh, u're right
12:21:07*amiconn is confused
12:21:35amiconnThe ucl decompressor seems to read the compressed data okay. It returns the correct uncompressed length
12:21:55amiconn...but the uncompressed data written to the destination is all wrong
12:22:09LinusN*all* wrong?
12:22:19LinusNoverlapping issue?
12:22:23amiconnWell, I only checked the first 4 bytes
12:22:24ashridahthat's the technical term? :)
12:22:29amiconnSurely not
12:22:45amiconnThe compressed data is copied to ram_end-image_size
12:22:57amiconn...then uncompressed to ram_start
12:22:58LinusNthe technical term is "wacky" :-)
12:23:13amiconnThe first 4 bytes should be 0x09000200
12:23:22ashridahthought that was spelled whacky.
12:23:26amiconn...but with the unicode .ucl I get 0x000000f1
12:23:31LinusNashridah: not the technical term :-)
12:23:32*ashridah checks his notes
12:23:48preglowwhacky? i've never seen an h there
12:23:51ashridahah. a colour/color thing?
12:24:13linuxstbWhack (with a h) means to hit.
12:24:19amiconnEven more strange, if I flash that very .ucl, the flash bootloader boots it correctly
12:24:27 Join Jamjam [0] (
12:24:28Slasheriamiconn: Hmm, why not just make a bootloader instead of using ucl compression? That wouldn't have a size limit at all
12:24:30amiconnIt's the very same decompressor...
12:24:51LinusNamiconn: sounds like a really trivial bug
12:25:01amiconnSlasheri: We have to keep down the size anyway
12:25:02LinusNamiconn: no gdb?
12:25:04linuxstbSlasheri: Compression is far neater IMO.
12:25:05MrShleeI wonder if a linux based firmware would be worth it :|
12:25:15MrShleeno real point tho.
12:25:17amiconn...and 2-stage boot would introduce more possible points of failure
12:25:17Slasheriamiconn: yes, true..
12:25:17LinusNMrShlee: not at all
12:25:23B4gderMrShlee: would good what that bring?
12:25:32B4gderwhat good would
12:25:37MrShleeclusters :)
12:25:37*B4gder types crappy
12:25:56Slasheribut some day we would probably reach the ucl limit too
12:26:04dwihnoAnyone uses gcc4 for production stuff?
12:26:11JamjamHey, im just stopping by to fall to my knees and cry thanks to you all for finally bringing gapless playback to the H3XX
12:26:16Jamjamyou are all gods among men
12:26:17B4gderdwihno: yes sure
12:26:17Jamjamhave some pie
12:26:23linuxstbRe Linus - it was bring us a large number of USB drivers (for H300) and support for all known filesystems. But that's about it. I'm not advocating it.
12:26:26LinusNJamjam: you're welcome
12:26:31amiconnSlasheri: The ucl limit will be around 350KB uncompressed, and we better stay away from that
12:26:59dwihnoB4gder: Major version numbers always scare me :)
12:27:07dwihnoB4gder: At least the first 2-3 revisions
12:27:13SlasheriHmm, and now rockbox is around 250 KiB? =) That's not so far away..
12:27:25amiconnSlasheri: It's just under 200KB on archos
12:27:26dwihnoRockbox 2.0 - Now THAT was scary!
12:27:31Slasheriah, ok
12:27:40amiconn(recorder v1 that is)
12:27:49amiconnfm recorder and v2 are slightly above
12:28:01mirakhum there is sound on H100 ?
12:28:08amiconnondio fm is ~191KB, ondio sp ~170KB, player ~148KB
12:28:16LinusNmirak: yes
12:28:23dwihnoJust considering cell phones using the WinCE sw... Windows Mobile 2003 SE is really nice, and I've read a lot about the 5.0 being evil stuff! :O
12:28:29B4gderh100 has had sound for many months
12:29:06 Quit Jamjam (Client Quit)
12:29:51Paul_The_Nerdmirak: H100 has sound, radio, sound recording, and even limited lossless wavpack encoding.
12:30:37 Join Jamjam [0] (
12:30:39 Join Wett [0] (
12:32:51 Join Jungti1234__ [0] (n=jungti12@
12:33:27 Quit Jungti1234 ("Bye~ -")
12:33:43 Nick Jungti1234__ is now known as Jungti1234 (n=jungti12@
12:33:58Paul_The_NerdI think almost all excuses to switch to the iRiver FW are gone.
12:34:26B4gderI haven't used the original in months
12:34:38Paul_The_NerdI still use it for recording in MP3 format.
12:34:39LinusNme neither
12:35:18Paul_The_NerdAnd the couple times I end up there after finding a new way to lock it up again.
12:37:10mirakWett: hi
12:37:22Jamjamhmmm...theres a guy on misticriver asking if it would be possible to port scummVM to rb.....unlikely?
12:37:35mirakit's prettier than rockbox
12:37:37mirakok I am out −−-> [] ^^
12:38:16Bgerhm, opensource :)
12:38:24preglowSlasheri: afaik, not all archos units can be flashed
12:38:31Paul_The_NerdY'know, ScummVM is something I'd never imagined playing on an MP3 player
12:38:42MrShleethats the only real benifit of porting to nix.. Emuation and greater software support over the long term.
12:38:46Paul_The_NerdThe screens *are* generally speaking too low of a resolution.
12:38:52mirakhem I downloaded gcc debian sources, but there is a gcc archive plus a gpc archive. What's gpc ?
12:39:01linuxstbJamjam: It's written in C++ which could be a problem. But it seems to have been ported to mobile phones and PDAs.
12:39:09Jamjamoh? cool
12:39:20Paul_The_NerdI used to use ScummVM on my ancient HP Jornada 565
12:39:28*Jamjam is blissfully unaware of the limitations of programming languages
12:39:50Jamjammobile phones?!
12:39:59Jamjamsurely the screens are far too small...
12:40:36Slasheripreglow: yes, but still the bootloader can be loaded from disk and that can load rockbox itself.. It doesn't need to be flashed :)
12:40:37linuxstbJamjam: C++ has larger overheads than C - Rockbox is written in C. A port of scummvm would either need to make C++ plugins work in Rockbox or convert scummvm to C.
12:40:45preglowSlasheri: true enough
12:40:49preglowi say go for the compressed way
12:40:52Paul_The_NerdJamjam: I imagine someone clever could set it up to display only the areas immediately surrounding the character. It wouldn't be stricly accurate any more, but that's usually enough for those game.
12:41:01preglowi can't imagine what we'd have to add to make the archoses reach 350kb
12:41:11Slasheriyep, that sounds better atleast at the moment because the image size is still so small
12:41:39preglowbut sure, it'll probably happen
12:42:03amiconnI would really prefer to not get anywhere near 350KB
12:42:12preglowyou're not alone in that
12:42:35JamjamPaul_The_Nerd: Aha, an ingenius scheme
12:42:47amiconnEvery byte we consume for the firmware isn't available for buffering music
12:42:53Slasherihehe, possibly yes.. as "640 KiB of ram should be more than enough anybody is ever going to need on a pc" ;)
12:43:00amiconn...and I would prefer to stay below 228KB
12:43:09preglowwhy 228?
12:43:11LinusNJamjam: how do you play the game? don't you have to enter text? order not to give up ROM execution
12:43:25Paul_The_NerdLinusN: All SCUMM games are played purely with a mouse.
12:43:31Paul_The_NerdOr should be able to be.
12:43:37JamjamLinusN: Yeah, they're mainly point and click
12:43:39amiconn(for those lucky ones with a flashable box)
12:43:51LinusNok, so the joystick would be enough
12:43:53JamjamLinusN: there is some typing, but only for saving games
12:45:06LinusNlooks tricky to squeeze into a 220x176 screen :-(
12:45:21Paul_The_NerdYeah, the games are 320*200
12:45:23preglowday of the tentacle on h120, that would be something
12:45:29Paul_The_NerdIf I recall
12:45:33preglowthey were
12:46:00Jamjamon their site, it seems to be in windows, fedora, debian, slackware, mandriva, mac, os/2, psp, smybian, amiga, morph, dreamcast and beOS format
12:46:08Paul_The_NerdYou could implement so basic edge scrolling I suppose. (or perhaps pressing play scrolls the screen 100 pixels left/right, and rec scrolls it 16 up/down)
12:46:10Jamjamso it seems to be quite flexible
12:46:33Paul_The_NerdJamjam: There's an official PS2 port as well.
12:46:45JamjamPaul_The_Nerd: Is there? excellent
12:47:04Paul_The_NerdI wonder if maybe the idea of porting it to rockbox might be better proposed to them?
12:47:20Jamjamill have a look in thier forums...
12:48:05Jamjamwell, they seem to have some variety of "other ports" sub forum
12:48:11linuxstbI think we would need to use C++ to compile it though - it seems very object-oriented.
12:48:44Jamjamgoddamn it! i was following the conversation up until you said that
12:50:29JamjamOut of interest, how do you stop people abusing the open sourceness of rb? i mean, if people can just submit code, then surely they could cause serious problems with peoples players, if they were so inclined
12:50:48 Join tvelocity [0] (n=tony@
12:50:58B4gderJamjam: we don't just let any code in
12:51:04LinusNJamjam: only a select group of people can change rockbox
12:51:14Paul_The_NerdNow *there's* a common Open Source misconception.
12:51:37B4gderJamjam: and then we review each other's changes
12:51:46Jamjamwell, the word "open" tend to give a hint of openness
12:52:00B4gderwell, open doesn't mean everyone-can-do-everything
12:52:04LinusNopen, as in "public"
12:52:24MrShleedepends.. you got anything against a fork? not that it would happen :|
12:52:37MrShleeunless the dev team break up SCO style.
12:52:39LinusNit has happened
12:52:49B4gderMrShlee: a fork splits the dev team, that's a downside
12:52:59Paul_The_NerdOpen means the public can download the source code. Then they can make changes for their personal use. If they think those changes are ones everyone would like, they can submit them to the patch tracker, where they may be reviewed, and if functionally sound and useful and wanted, a core developer may then integrate them into the project.
12:53:30Paul_The_NerdWell, generally speaking
12:53:42B4gderMrShlee: but we have no legal means to stop a fork and wouldn't want to, so if someone wants to they can run off and make their own Rockbox branch
12:53:57B4gderlike openneo
12:54:12Jamjamwell that clears that up then
12:54:13B4gderbut then also risk that they, like openneo, lags behind in no time
12:54:31Paul_The_NerdWhy did open neo fork anyway?
12:54:44B4gderPaul_The_Nerd: because they wanted their own race
12:54:48LinusNthey have a hardware that is very different
12:55:00Jamjamcan someone breifly tell me what openneo is?
12:55:00B4gderthey actually never even tried to play ball with us
12:55:19LinusNJamjam: an in-car mp3 player
12:55:21B4gderJamjam: a rockbox fork, a firmware for Neo
12:55:30preglowwasn't it a car audio player?
12:55:31Jamjama splinter organisation
12:55:40B4gdermainly in-car
12:55:46B4gderbut they did have a portable too
12:55:53B4gderI'm not sure if openneo works on that
12:56:10B4gderthey have a 4 line charcell lcd
12:56:18LinusNopenneo is much like an in-car OSCAR player
12:56:24B4gderand 256K ram iirc
12:56:38B4gder... and they added malloc() ;-)
12:56:54Jamjamwhich page should i quote to the scummVM guys about asking for a port, to give them an idea of how rb works?
12:57:25LinusNB4gder: and increases the pool size every now and then
12:57:33B4gder64K+ and counting
12:58:17Paul_The_NerdJamjam: You may just want to point them to the main page, and mention that Rockbox is slowly spreading to more MP3 players, and a port to it is more a port to a category of devices rather than any specific one.
12:58:42JamjamPaul_The_Nerd: Ok, good idea
12:58:43 Join webguest81 [0] (
13:00:11Paul_The_NerdOkay, it's bad enough that sites aren't fully functional in Firefox, but it's iritating when they just refuse to let you in unless you're running IE 5.5+
13:00:39B4gderthere's that nice user-agent selector plugin ;-)
13:01:17ashridahdoesn't always help if they use horrible javascriptery to do it tho
13:01:36B4gderpersonally I seldom visit such sites
13:01:38Paul_The_NerdI don't want to do that though. It's not an important site.
13:01:49linuxstbPaul_The_Nerd: Don't use the site and complain to the company concerned. If you don't complain, the people in charge won't realise there is a problem.
13:02:01Paul_The_NerdI used to just mildly dislike IE. Then I learned CSS to make a page for a friend. Then I saw what that page looked like in IE. Then, I cried.
13:02:07 Quit webguest81 (Client Quit)
13:02:20Jamjamif you're running the beta of firefox, i think you can report non working pages
13:02:32mirakwhen I install binutils and gcc, it says that this file is conflicting /usr/local/lib/libiberty.a I use checkinstall to have the possiblity to uninstall, but it conflicts. I wonder if I can overwrite it
13:02:47Paul_The_NerdJamjam: It's not non-working
13:02:49mirakif it's a comon file or not
13:02:51Paul_The_NerdThey don't want non-IE users.
13:02:56Paul_The_Nerd Very happy about it too. Terrifying
13:02:57JamjamPaul_The_Nerd: Ah
13:03:01linuxstbmirak: Are you installing a cross-compiler?
13:03:04LinusNmirak: which gcc are you installing?
13:03:11miraklinuxstb: I try to install gcc coldfire
13:03:19mirakLinusN: gcc 3.4.4
13:03:24linuxstbThen you should specify a −−prefix
13:03:24mirakwith binutils 2.16
13:03:46linuxstbe.g. pass −−prefix=/usr/local/coldfire to the configure script.
13:04:04mirakwhat is the make uninstall command ?
13:04:12LinusNthere is none
13:04:30linuxstbmirak: rm -fr /usr/local/coldfire (if you had used a prefix....)
13:04:41miraklinuxstb: I didn't ...
13:04:54B4gderyou should follow the instructions closer
13:05:04mirakLinusN: I used this page, but I though the prefix was for conveniant usage
13:05:06B4gderit saves you from grief
13:05:15linuxstbmrak: Have you run "make install" yet?
13:05:18mirakit doesn't say why you should use it
13:05:28B4gderno, but it says you should use it
13:05:34miraklinuxstb: yes, but I used checkinstall, so I can just remove the .deb I think
13:05:50***Saving seen data "./dancer.seen"
13:06:28mirakB4gder: Feel free to use any prefix it says, so I felt really free ;)
13:06:47B4gderwell, you *are*
13:06:53LinusNfreedom has a cost :-)
13:07:00B4gderbut then you had already installed something in that same path
13:07:09B4gderit isn't really the fault of the instruction
13:07:38mirakdon't know, I wasn't expeting binutils to have some comon files with gcc.
13:07:51mirakB4gder: no it's not, but someone did a mistake ^^
13:07:52LinusNif you deviate from the instruction, you need to know what you are doing
13:08:17mirakLinusN: sorry I don't have a /home/linus/ folder ^^
13:08:32B4gderthen make one darnit! ;-P
13:08:45linuxstbHave you checked? You don't know what Linus has been up to...
13:09:08LinusNyou are 0wn3d!
13:09:57mirakwhat is the interest of building in a different folder than the source ?
13:10:11LinusNmakes it easier to reconfigure
13:10:18linuxstbmirak: I don't know what checkinstall does, but I'm guessing it hasn't installed any files yet. So just delete your build directory and start again.
13:15:48 Quit [IDC]Dragon ("CGI:IRC")
13:17:21mirakLinusN: ok but if I use the same folder for gcc and binutils, I will have the same conflict isn't it ?
13:17:51linuxstbNo. They should both be installed in the same place.
13:19:08 Join webguest00 [0] (
13:19:11LinusNthey should have the same prefix
13:19:28webguest00tnx for the good job linus !
13:19:38LinusNyou're welcome
13:19:42webguest00but why is the music so low in the h3xx ?
13:19:51LinusNdecrease the bass/treble
13:19:56webguest00is it because of the 90 frequenz?
13:20:02LinusNdecrease the bass/treble
13:20:03ashridahmirak: as long as they're both on the path, they should work, but convienience is just putting them both in the same prefix.
13:20:32ashridahwebguest00: crossfeed will lower the volume a bit because of the way it's implemented if you've got that on
13:20:39webguest00yeah because if i play in 100 % in rockbox and play like 50 % in iriver the iriver is higher:D
13:20:47Jungti1234damn linux
13:21:22webguest00but linus qill you inplate colour later on?
13:21:35mirakthe volume in the iriver firmware uses the max output possible, or is there some power left ?
13:21:48Jungti1234My regrettable 24 hours
13:21:51mirakit's pretty weak on the H300
13:21:52 Join Zak1392 [0] (
13:22:08LinusNwebguest00: yes, there will be more colors eventually
13:22:15Zak1392WAZZUP GUYS
13:22:19ashridahmirak: the iriver firmware will turn off bass/treble as the main volume increases, rockbox won't, it'll clamp the volume instead
13:22:23Zak1392sorry for caps
13:22:59webguest00LIKE IN THE PLAYBACK MODE?
13:23:05webguest00sorry for caps:S
13:23:13webguest00and in rockboy?:D
13:23:24Zak1392hardy har-har
13:23:28linuxstbmirak: A fair comparison will be to turn off all sound processing options in both iriver and rockbox firmware, and then turn the volume up to the maximum.
13:24:28 Join DMJC [0] (
13:24:54webguest00but its still not even close to the original volume in iriver:(
13:25:07linuxstbHave you done that comparison then?
13:25:35webguest00got it now
13:25:39webguest00sorry :d
13:26:58webguest00i just saw that the 16bit LCD driver: faster drawing routines update
13:27:04webguest00what does that mean?
13:27:21B4gderwhat do you think?
13:27:32B4gdercan it be said any clearer?
13:27:42MrShleeit means if can push 16bits of data at the LCD per cicle..
13:27:51MrShleemore data = less time to transfer
13:28:11webguest00sorry but im a n00b:D
13:28:23amiconnThe *drawing* speed is increased, not the transfer to the lcd
13:28:30B4gderwebguest00: then you probably should ignore the CVS commit messages all together
13:28:52B4gderthey are source code changes
13:29:03amiconnThe latter requires an actual unit to tune it, but imho there's quite some room
13:29:10MrShleemeh.. I didn't read the cvs. educated guess.
13:29:15webguest00so now the rockboy is faster?
13:29:27Jungti1234Do not have person who convert bdf to fnt to me?
13:29:30amiconnrockboy does its own drawing
13:30:05B4gderJungti1234: tools/convbdf
13:30:06webguest00so if we want faster rockboy somebody most work with the rockboy code?
13:30:30Jungti1234B4gder: I know but I use Windows.
13:30:32linuxstbamiconn: Is rockboy on the h300 the full height? (I'm assuming you removed the line-skip code)
13:30:36B4gderJungti1234: so?
13:30:43amiconnlinuxstb: yes
13:30:46B4gderJungti1234: lots of people develop on windows
13:30:53Jungti1234Don't know whether use it how.
13:31:04amiconnlinuxstb: It's centered within the display
13:31:21B4gderJungti1234: so you rather bother others than learning yourself?
13:31:52Jungti1234Is trying.
13:32:10Jungti1234Is so many as don't know however.
13:32:54 Quit skorpyon (Read error: 110 (Connection timed out))
13:33:09linuxstbamiconn: Any idea if adding color support to rockboy would slow it down significantly?
13:33:09 Join skorpyon [0] (
13:33:56amiconnI have no idea
13:34:14amiconnI wasn't the one who disabled palette support, HCl did that
13:34:36amiconnThe gnuboy code isn't really transparent...
13:36:15linuxstbCan you remember the resolution of the gameboy?
13:36:46amiconnErm, 160x144
13:37:10linuxstbI prefer your first answer - the Nano's screen is 176x132
13:38:26preglowplaying games on the nano screen quickly becomes tiring anyway
13:39:41Paul_The_NerdWhy's that?
13:39:51preglowwell, it's small
13:40:08preglowactually, it's also got a lot to do with the unit itself
13:40:18preglowit's too small for comfortable controls in the long run
13:41:22 Quit Zak1392 ()
13:41:58preglowi prefer a larger unit when i want to use it for games
13:42:16amiconnpreglow: It's not smaller than the Ondio, except the thickness
13:42:24 Quit webguest00 ("CGI:IRC (EOF)")
13:42:30preglowamiconn: then i probably wouldn't like to play games on that either
13:42:35 Quit goa (Read error: 110 (Connection timed out))
13:43:13amiconnThe drawback with games on the Ondio is the lack of backlight, the controls aren't a problem
13:43:30amiconn(except when a game needs many of them)
13:44:12preglowondio has a different shape than nano, yes+
13:44:14preglow? <-
13:44:33Jamjamwould it be possible to create some sort of music sequencer, with samples? like Ejay?
13:44:40preglowJamjam: it's already been done
13:44:47Jamjamhas it?
13:44:57mirakin theory in wich file should I add the PATH variable values ?
13:44:57mirakon linux
13:44:57mirak.bash_profile or .bashrc ?
13:44:58Jamjami thought i went through all the plugins
13:44:59preglowsomeone made a step sequencer
13:45:47Jamjampreglow: Thanks for a build that's already got it
13:47:13preglowit's kind of old, though
13:47:29 Quit Jungti1234 ("bye")
13:47:38Paul_The_Nerd"Kind of"?
13:47:46Jamjamso if i ran that, i would lose other new features?
13:48:00Paul_The_NerdIt's about two months.
13:48:01preglowrasher seems to have vanishes
13:48:04Paul_The_NerdQuite a bit has changed...
13:49:07 Join phaedrus961 [0] (
13:49:24Jamjampity so many people vanish
13:49:58preglowthey usually come back
13:50:59mirak/usr/local/coldfire/lib/libiberty.a conflicts between binutils and gcc. I will just overwrite it since a make install would also overwrite
13:51:24LinusNmirak: checkinstall?
13:51:58mirakLinusN: it's a script that do a debian package from install script. so it's easier to remove stuffs
13:52:31mirakLinusN: you see what I mean ?
13:53:21LinusNnot really
13:54:08LinusNi think it's safe to overwrite though
13:54:11mirakwhen you do a make install in a source folder, it put the file where you said you wanted to put them. So checkinstall installs the files and do a package with the installed file.
13:54:12Slashericheckinstall is quite useful, it automatically creates and installs the debian package, and it does same as make install (at least it should do)
13:54:37mirakLinusN: then you can remove the things you isntalled by using dpkg -r
13:54:38linuxstbmirak: It seems to be a GNU thing - libiberty is shared by various packages, so each package includes a copy.
13:55:14linuxstbOverwriting the binutils version with the gcc version should be fine - everyone else must do that.
13:55:22mirakSlasheri: yes. However I wonder how it works, because it seems it installs the files before creating the package. It's weird
13:55:33linuxstbIt probably uses fakeroot.
13:55:48Slasherimirak: yes.. probably it looks what files was installed by the script
13:55:50miraklinuxstb: it fails if I am not root
13:55:55Slasherii am not quite sure how it works
13:56:08mirakbut it works anyway ^^
13:59:09mirakhem where should I put the path variable to it's in the PATH always, and even for root ?
14:05:03linuxstbYou don't need it to be in the PATH for root - only when you do the make install for gcc.
14:06:15 Quit KN|stiff (Read error: 110 (Connection timed out))
14:13:46 Join Jungti1234 [0] (n=jungti12@
14:14:45Jungti1234markun, are you here?
14:14:45dwihnoIs it possible to check what is wrong with a mp3? I just checked a file which has several broken frames... :/
14:14:45Jamjamwould it be possible to quickly knock together a rb equivilent to this:
14:14:45dwihnoI know which frames are borken
14:14:45dwihnoIs there any way to check the rest of the MPEG data being intact?
14:15:56linuxstbJamjam: There is a runtime database in Rockbox. So you would need a PC-based application to query the runtime database. That's all I know about it though.
14:16:52Jamjamis that the "gather runtime data (experimental)" option?
14:17:03linuxstbI expect so.
14:17:15Jamjami see
14:17:20linuxstbNo-one is actively working on it at the moment as far as I know.
14:17:45Paul_The_NerdIt's useless unless you've got the database set up. It's only functional for OGG and MP3 files though, to my understanding.
14:17:53*Paul_The_Nerd may be wrong
14:18:16Paul_The_NerdIt also hates me, very very much, because *somehow* my MP3s got APE tagged, and I'm not quite sure how.
14:19:27*Jamjam looks blankly at "APE", hoping someone will enlighten him
14:19:51Paul_The_NerdAPEv2 specifically, I believe. Like ID3, really
14:19:51MrShleeAPE = lossless codec
14:19:52MrShleeor something
14:20:30Jamjamare lossless codecs the ones with only like a 1:2 compression ratio?
14:20:33*ender` 'd like to see APEv2 tag support in rockbox...
14:20:34Paul_The_NerdIt's another form of music file tagging. They're native for Monkey's Audio lossless .ape files.
14:21:02dwihnoI really DO hate every new tag in mp3 files.
14:21:16dwihnoWasn't id3v2 supposed to be sufficient?
14:21:28ender`id3v2 is overbloated piece of crap
14:21:41Paul_The_NerdWhy do you say that?
14:21:42ender`plus 90% of the taggers get it wrong
14:21:57 Quit Jungti1234 ("bye")
14:22:06ender`ever tried writing an id3v2 parser?
14:22:49Paul_The_NerdBut that doesn't mean you can't give me the "layman's terms" version of what you mean by "overbloated" and especially "piece of crap" since those are rather subjective terms.
14:22:49ender`try it once
14:23:28LinusN90% of the taggers don't do ape tags, iirc
14:24:09LinusNsounds to me it's much like the vhs/betamax issue
14:24:19LinusNid3v2 is vhs
14:24:23ender`90% of MP3s i get with id3v2 tags are unreadable because they use whatever windows ansi encodin the user had instead of using iso-8859-1 or unicode (which are the only two officially supported)
14:24:46preglowonly thing i don't like about id3v2 is that it's put at the start of the file
14:24:48preglowwhich is rather braindead
14:24:57Paul_The_NerdSo, your problem with ID3v2 doesn't relate to the format, but rather the implementations being wrong?
14:25:01LinusNender`: is that id3v2's fault?
14:25:04B4gderpreglow: its good for rockbox
14:25:11 Join muesli- [0] (n=muesli_t@
14:25:21LinusNpreglow: why is that braindead?
14:25:36preglowbecause it requires every bloodu codec to check for mistagged id3v2
14:25:52ender`LinusN: not really, but it requires me to rewrite the complete file to fix that, because the tag is at the start of file without any padding
14:26:09LinusNand is that id3v2's fault?
14:26:20amiconnpreglow: It doesn't matter whether wrong tags are located at the start or end
14:26:27ender`because nobody sane should put tags at the beginning of the file
14:26:29LinusNender`: that it isn't padded?
14:26:32amiconnWe still have to skip them properly
14:26:38preglowamiconn: well, yeah, you don't need to care if it's at the end
14:26:45preglowno you don't
14:26:48ender`(besides, padding = waste of space)
14:26:54linuxstbamiconn: Not really. Most codecs will stop decoding at the end of the stream (they store the length in the header).
14:26:55 Join Mirfle [0] (
14:27:16amiconnThat's... strange
14:27:32amiconnI thought a stream is a stream because it doesn't have a global header
14:27:44linuxstbamiconn: That's what containers do.
14:27:50pregloweven mp3 has a header, mostly
14:27:54preglowcomplete with a frame count
14:28:02 Join Kohlrabi [0] (
14:28:20preglowonly mp3 is stored containerless
14:28:24preglowwell, and flac
14:28:43linuxstbEven FLAC has a header.
14:28:44amiconnpreglow: You mean the xing header? That's an afterthought, and not necessary for playback
14:28:50preglowamiconn: know
14:28:55preglowagain, this is an mp3 problem
14:28:59 Quit jaydpb ("CGI:IRC (EOF)")
14:29:02preglowalmost all other formats have containers
14:29:11 Join Benacool [0] (
14:29:11preglowand even for mp3 you don't have to skip it
14:29:16preglowid3v1, that is
14:29:24preglowit'll resync and continue playing
14:29:31ender`anyway, is there something wrong with the apev2 support patch in the forums?
14:29:33amiconnYes, and then you have a gap
14:29:57linuxstbWhy would that cause a gap? Or are you talking about the MAS?
14:30:12preglowno reason for there to be a gap if the resync is successful
14:30:33preglowstill, the point is clear by now
14:30:38preglowan id3v2 tag is intrusive
14:31:51amiconnlinuxstb: Then tell me why mp3 'nogap' gapless didn't properly work on iriver until LinusN added id3v1 skipping?
14:32:19 Quit Vlad0man (Read error: 110 (Connection timed out))
14:32:20linuxstbI've no idea. Presumably the resync didn't work correctly.
14:32:55LinusNiirc, pretty much all mp3 decoders output silence when they fail to resync
14:33:05linuxstbOr maybe a resync causes libmad to reset itself, losing the bit reservoir.
14:33:14 Join Vlad0man [0] (
14:34:00LinusNlinuxstb: sounds likely
14:35:41Benacoola little question... when you're in the browser, you can hold press de A-B button to see the repeat/shuffle menu... but why I can't do the same thnig when I'm in the WPS screen ?
14:35:51Benacoolthe* not de
14:36:14ashridahBenacool: try holding navi?
14:36:19Paul_The_NerdHrm... H3x0 series?
14:36:47Benacoolnot the same menu when i hold NAVI
14:37:04 Quit Febs ("CGI:IRC")
14:38:25Benacooli talk of the menu with Shuffle Mode :, Repeat Mode:,
14:38:33Benacool and Show Files:
14:38:50Paul_The_NerdIt's odd that it doesn't work in wps...
14:39:14Paul_The_NerdBy odd I mean "I have an H120 and it works there" really... Heh.
14:39:46mirakwhere exactly are the build info ?
14:39:56mirakI have the environnement for build and CVS
14:41:22Benacoolwhere we can post this kind of bug ?
14:42:13 Quit MrShlee ("Changed by popular Request - "I am the heart of your darkness, you stinking oompa-loompa morons!"")
14:45:34ashridahBenacool: posting bugs is on the pointless side, key assignments is on the TODO list like a lot of things
14:46:23ashridah(unless, of course, you post a patch with the bug, those are welcome :) )
14:48:30mirakwhere is rbconf ?
14:48:35mirakI can't find that file
14:49:01amiconnashridah: A patch _with_ the bug? ;)
14:49:14ashridahamiconn: bug report, rather.
14:49:34ashridahmirak: it's not part of rockbox's source and toolkit thati can see, nor the crosscompiler
14:49:38ashridahwhat're you running that's giving out the error?
14:50:01mirakashridah: I did a configure in tools, from a folder
14:50:11mirakI choosed simulator as target
14:50:21mirakI commited simultaor from cvs
14:50:32mirakmake[1]: /home/karim/Prog/src/rockbox/rockbox/tools/convbdf: Command not found
14:50:38mirakthat's after trying to do a make
14:50:48LinusNmirak: cd tools;make
14:50:56mirakLinusN: I have done that already
14:50:59ashridahwhat he said
14:51:10 Join perplexity [0] (
14:51:10ashridahwhere does rbconf kick in?
14:51:27mirakok now I have it
14:51:43mirakashridah: I just read the building how to
14:51:46LinusNiirc, rbconf is a silly script in bluechips cygwin devkit
14:51:50linuxstbmirak: AFAIK, rbconf is part of the devkit and is just a wrapper around the configure script.
14:52:11mirakI try to follow that but it seems not much in sync
14:53:04linuxstbIgnore that page :) The instructions are specific to the devkit (a stripped down version of cygwin).
14:53:07mirakhem what exactly the simulator consists in ? Does it simulates the architecture ?
14:53:37LinusNno, it just runs rockbox with a simulated lcd and buttons
14:53:57LinusNusing windows/x11 stubs
14:53:57mirakok but it's nt specific to the device ?
14:54:24chopped_porkwoops sorry
14:54:30LinusNeach device has its own lcd type and set of buttons
14:54:43LinusNso the simulator has to know which device you are simulating
14:54:44chopped_porkis [ 1267994 ] Unicode patch the patch that is going to get implemented?
14:54:54amiconnYes, and the x11 sim still doesn't work for H3x0
14:54:59chopped_porkor rather commited to cvs
14:55:10amiconnIt runs, but you don't see anything on the main lcd
14:55:23mirakI am running it
14:56:08LinusNchopped_pork: eventually, yes, i think so
14:56:19Paul_The_NerdQuestion: What does "No wps module present, can't do the WPS magic!" mean, and how do I make it go away?
14:56:29LinusNPaul_The_Nerd: cvs co wps
14:56:31linuxstbcvs co wps
14:56:56mirakhe is not happy
14:57:08LinusNthe wps is now included in the "rockbox" module, but only when you check it out fresh
14:59:09mirakrahh that's beautiful I got sound !
14:59:22mirakanyone use that as a player on linux ?
14:59:38mirakah it crashed when chaging song
15:00:57mirakhum but what about assembly parts in the code ? how can they be simulated ?
15:02:02linuxstbThere are always C alternatives.
15:02:39 Nick Jd|away is now known as JdGordon (
15:05:53***Saving seen data "./dancer.seen"
15:08:33JdGordonwoohooo im in the credits :D
15:14:18JdGordonany1 with an ondio...
15:14:49markunchopped_pork: I didn't update the patch in the tracker yet. Unicode will be committed soon, so just wait a bit
15:15:40amiconnJdGordon: The image link yields a 403...
15:16:07JdGordonmy stupid webhost..
15:16:19JdGordonu got an idea for a button combo?
15:16:46 Quit ashridah ("sleep")
15:17:04amiconnI first wanted to see what that feature looks like. I can't really imagine just from the description
15:18:10linuxstbamiconn: If you copy and paste the url of that image into your browser it will work.
15:18:37JdGordonye, my web ost settings is screwed.. dunno whats wrong
15:19:25amiconnHmm, so how is that scratchpad supposed to be used?
15:19:55JdGordonu press a button when u have a number u tihnk it might be and its added to the pad
15:20:26amiconnWhere do you press the button
15:20:34linuxstbSo on the irivers, you press record, and that adds the number in the current cell to the scratchpad for that cell?
15:20:47amiconnI mean, how does the scratchpad relate to the main grid?
15:21:04linuxstbSo how do you remove numbers?
15:21:14JdGordonby pressing the button on the number again
15:21:40linuxstbI'm guessing there are 81 scratchpads, one per cell.
15:22:10JdGordon81*shorts.. uses a bit of mem for it :p but tis worth it
15:22:17amiconnAh, so the 1 column shown is a view to one of 9 columns (?)
15:22:47JdGordonno the colum on the left shows you the numbers u tihnk might be possible in the cel that the cursor is over
15:23:05amiconnAh, now I got it
15:23:43amiconnWell, since the scratchpad is shown at the left side, I'd suggest (BUTTON_MENU | BUTTON_LEFT) on Ondio
15:24:05amiconnA bit tricky to get, especially since (BUTTON_MENU | BUTTON_REPEAT) is also used, but it should work
15:25:06mirakhem song skip with right arrow crash rockbox simultaor
15:25:34JdGordonu want to try it out?
15:25:40B4gdermirak: fix it and post a patch!
15:27:17Mirflewhen I try to compile rockbox for the h120, I get a compile error in the tools/ipod_fw.c
15:27:30 Quit DreamTactix291 (Read error: 110 (Connection timed out))
15:27:34linuxstbMirfle: Either ignore it or upgrade to the latest version of the devkit.
15:28:05 Quit San (Read error: 104 (Connection reset by peer))
15:28:17LinusNit's hard to ignore nowadays when tools/ is built automatically
15:28:24B4gdermake -k
15:28:33 Join skorpyon_ [0] (
15:28:59mirakB4gder: ok, but I need to eat
15:29:02 Join BBub_ [0] (
15:29:21B4gdernonono, fix first *then* eat! B-]
15:29:28Mirflethanks' I didn't realize there's a new version...
15:29:38JdGordonis there a more automatic way of using the bmp2rb file with more than 1 bmp?
15:30:08B4gderJdGordon: only if you write up a script for it
15:30:10linuxstbfor a in *.bmp ; do bmp2rb $a > $a.c ; done
15:30:15B4gderlike that
15:30:28*linuxstb can't live with the bash for loop.
15:30:31 Quit skorpyon (Read error: 110 (Connection timed out))
15:30:43LinusNlinuxstb: "with"?
15:31:04*linuxstb realises he should learn more bash commands.
15:31:06*B4gder hands linuxstb an "out"
15:31:10linuxstbWhat's "with"?
15:31:32linuxstbI'm confused, but have now caught up...
15:31:40*linuxstb can't live without the bash for loop.
15:31:49LinusNthat's better
15:32:41JdGordon~/tools>for a in rb\ sudoku/*.bmp ; do bmp2rb $a > $a.c ; done
15:32:41JdGordonBASH: $a.c: ambiguous redirect
15:32:41JdGordonBASH: $a.c: ambiguous redirect
15:32:47 Quit muesli- ("ich will Kühe!!!")
15:32:59B4gderha, spaces
15:33:20linuxstbYep. Use "$a" and "$a.c"
15:34:01linuxstbThat's why it's best to avoid spaces in file/dir names.
15:34:05B4gderoh yes
15:34:14B4gderspaces in file names make life harder
15:35:07*amiconn is annoyed by the selfextractor
15:39:03 Quit DMJC (Read error: 110 (Connection timed out))
15:39:52 Join DMJC [0] (
15:41:52chopped_porkhm where can i get unicode fonts for rockbox? (i'm testing a patched build from the forum, but it's lacking unicode fonts)
15:42:20chopped_porkgoogle isn't very helpful :/
15:43:25markunchopped_pork: I have some for you. Is there a specific language you need it for?
15:43:45JdGordonDONE... finished.. caput
15:44:50JdGordonlinuxstb: can i email u the changes again?
15:44:56chopped_porkmarkun: well i'm most interested in baltic (polish to be precise) stuff, as in 8859-2, and also japanese, besides that it's mostly eastern europe (as in 8859-1)
15:45:02linuxstbJdGordon: Sure.
15:45:03JdGordonnew neater numbers and the scratch thing is finished
15:45:13JdGordoni think thats enough for that
15:45:33linuxstbI forgot to ask you one question - why did you make the cells 16x16 instead of the maximum possible size (18x18) ?
15:45:34markunchopped_pork: do you want the BDF fonts or converted to .fnt?
15:45:55chopped_porkmarkun: .fnt would be cool if you could :)
15:45:55JdGordondunno :p
15:46:18 Quit BBub (Read error: 110 (Connection timed out))
15:46:18 Nick BBub_ is now known as BBub (
15:46:56chopped_porkbtw- will utf-8 ever be an option in the default encodings menu?
15:47:47chopped_porkfor all the unicode freaks who have everything in unicode (filenames and all tags;)
15:48:07MirfleI downloaded devkit v3.14, and with a fresh cvs I now get this:
15:48:09Mirflegcc -O2 -Iinclude -c -o uclpack.o uclpack.c
15:48:10MirfleIn file included from /lib/gcc-lib/i686-pc-cygwin/3.3.1/include/syslimits.h:7,
15:48:11Mirfle from /lib/gcc-lib/i686-pc-cygwin/3.3.1/include/limits.h:11,
15:48:13Mirfle from include/ucl/uclconf.h:46,
15:48:14Mirfle from include/ucl/ucl.h:39,
15:48:16Mirfle from uclpack.c:36:
15:48:18Mirfle/lib/gcc-lib/i686-pc-cygwin/3.3.1/include/limits.h:122:75: limits.h: No such file or directory
15:48:19MirfleIn file included from uclpack.c:37:
15:48:21Mirflelutil.h:34:19: ctype.h: No such file or directory
15:48:22Mirflemake[2]: *** [uclpack.o] Error 1
15:48:24Mirflemake[1]: *** [uclpack] Error 2
15:48:26Mirflemake: *** [tools] Error 2
15:48:27Mirfleany ideas?
15:48:28markunchopped_pork: try and franken.fnt
15:48:36JdGordonanything interestgin in the newwest builds?
15:48:51chopped_porkmarkun: awesome, thanks
15:48:53markunchopped_pork: yes, will be an option
15:49:01markunchopped_pork: let me know if it works
15:51:19markunJdGordon: you can see all the changes are on the
15:51:43JdGordonye.. but that takes effort :p
15:55:01 Join ^BeN^ [0] (
15:55:12linuxstbMirfle: The problem is that the devkit ( which is "unofficial" ) is a stripped-down version of cygwin, and it's obviously been stripped down too much.
15:55:25 Join JdGordon41 [0] (
15:55:25 Join chiller_ [0] (
15:55:54 Quit chiller (Read error: 104 (Connection reset by peer))
15:55:59 Nick chiller_ is now known as chiller (
15:56:14chopped_porkmarkun: yes, fantastic:D can't wait for a official commit and utf8 in the filebrowser :)
15:56:20chopped_porkmarkun: cheers ;D
15:56:31Mirfleso what can I do (I don't need that ucl stuff anyway, 'cause I'm compiling for h100)
15:56:36linuxstbMirfle: The easiest solution for you will be to edit the file tools/Makefile and remove the word "uclpack" from the line near the top of that file starting "TARGET :="
15:56:53linuxstbI mean TARGETS :=
15:56:53markunchopped_pork: utf8 in the filebrowser? fat32 stores everything in UCS-2 (UTF-16)
15:56:59MirfleI'll try that
15:57:44chopped_porkmarkun: hum... i've been mounting my iriver with utf8 codepage under linux (i set it to default to utf8 in kernel)
15:57:49Benacoolis it possible that weird caracters in a id3 tag can prevent the tag to be recognized in rockbox ?
15:58:13linuxstbJdGordon: I see you've moved the bitmaps into .h files again...
15:58:15chopped_porkthat's a bad thing to do i presume?
15:58:23markunchopped_pork: I do that too. It converts all the utf8 filenames to utf16.
15:58:38markunYou can't see polish filenames?
15:58:43^BeN^there is some program that i can preview the wps before i put it in the player (h300)?
15:58:44JdGordon41ye sorry.. no u moved em back out...
15:59:22BenacoolAlegría ... the í
15:59:33Benacoolall the tag of this album don't load
15:59:49chopped_porkmarkun: not only polish, all special characters show as two weird chars, as if it was trying to read utf as ascii
16:00:14markunthat's really strange..
16:00:35markunI have greep and russian filenames that show up fine.
16:00:41markungreek :)
16:00:51chopped_porki guess that's some error on my side
16:00:57chopped_porki'll check
16:00:58 Join DreamTactix291 [0] (
16:03:11miahw00 i must say 2.5 has certainly fixed the "red light of death" issue for me. with 2.4, i'd have that happen so much that using my mp3 player was annoying, and I replaced the hdd thinking the drive had issues..
16:03:45linuxstbJdGordon: You've also added ^M and TAB characters... I removed them from the first version I checked into CVS.
16:04:02linuxstbAlso, have you thought about how to add the state of the scratchpad to the saved games?
16:04:09JdGordon41sorry.. blame vis studio and cygwin
16:04:23JdGordon41bah, no, too hard
16:04:31preglowyou can set vs to use spaces and not tabs
16:05:10linuxstbIt seems that vs adds ^M characters to the lines it changes, but not to every line in the file...
16:05:30linuxstbSo it ends up with a file that is neither Unix or DOS.
16:06:00^BeN^there is some program that i can preview the wps before i put it in the player (h300)?
16:07:14JdGordon41the simulator?
16:08:01chopped_porkmarkun: ah it was of course my fault:) mount -outf8 does the trick
16:08:22markunok :)
16:11:22 Quit Jamjam ("Chatzilla [Firefox 1.5/undefined]")
16:13:02 Quit ^Guest37784 (Read error: 110 (Connection timed out))
16:14:14JdGordon41gnite all
16:15:09 Quit JdGordon41 (Read error: 104 (Connection reset by peer))
16:17:00Benacoolgnite... it's 10:17 am here ;-)
16:17:14linuxstbpreglow: How would you implement playspeed control? Does that imply that if a user wanted x5 playback speed (which the iriver firmware can do IIUC), Rockbox would need to decode the file at more than x5 realtime?
16:17:19 Quit JdGordon (Read error: 110 (Connection timed out))
16:17:37linuxstbOr are tricks possible?
16:18:03Benacoolthe x5 speen in orignal firmware is not really 5x realtime ...
16:18:25linuxstbWhat is it then?
16:18:59Benacooldon't know but it's certainly not real 5x speed
16:19:08preglowlinuxstb: indeed
16:19:24preglowlinuxstb: currently, i would just implement it using the resampler
16:19:40preglowlinuxstb: set a resampling rate based on the speedup/slowdown that's specified
16:21:51 Quit CBM-away (Read error: 104 (Connection reset by peer))
16:21:52Benacool(in the original firmware the voice are a little bit faster and with a higher pitch? at 5x)
16:23:22linuxstbSo we would struggle to get much more than 2x realtime on the iriver, and Ogg probably couldn't even achieve that?
16:23:56preglowlinuxstb: who the hell wants 2x speedup?
16:24:01preglowit'll be pretty hard to catch what's happening
16:24:26Benacoolthe fun is just to have the voice with higher or less pitch ...
16:24:36preglowlinuxstb: there's no way around the fact thay you need to decode faster
16:24:43linuxstbThe person on the forum talked about wanting greater than x5 playback.
16:24:57preglowthen he's obviously drunk and/or on speed
16:25:08Benacool... fastforward is the same thing lol
16:25:10linuxstbAh, it's "+5", not "x5".
16:25:24preglow+5 what?
16:25:30Benacoolgood question ...
16:25:31preglow5% ?
16:25:31linuxstbGood question...
16:25:39preglowif so: easy
16:25:45amiconnOn archos recorder we have 50..200%
16:25:55preglowamiconn: yeah, but you overclock the mas
16:25:59Benacoollet's go for that then lol
16:26:06preglowwe can do 200%, probably
16:26:08preglowbut not much more
16:26:34preglowbut yeah
16:26:49preglowthe only thing keeping me from implementing it is the firmware/app thing
16:26:55linuxstb200% would be too much for some codecs/bitrates though.
16:27:12preglowthough i don't see why i shouldn't be able to hack in a quick ifdef in iriver settings
16:27:13Benacooljust by curiosity, is the todo list for the h300 is somewhere on ?
16:27:18preglowlong time since i looked at this
16:27:29preglowlinuxstb: it would also be an effect used primarily for comic relief
16:27:48preglowor by the most ardent speedcore djs
16:27:55B4gderBenacool: I don't think so
16:28:07B4gderthe todo list is pretty much "get it working"
16:28:09B4gderso far
16:28:12Benacoolok ;-)
16:28:19Paul_The_NerdYeah, but if you know in advance you're going to want something sped up, you could always encode it at double speed, 'eh?
16:28:22linuxstbBenacool, there is a sort-of to-do list for the h1x0 - the IriverStatus page.
16:28:23B4gderand for all else: add what you feel like
16:28:51preglowlinuxstb: i think we can cap ours at 150%
16:28:54Benacoolyea this one i know but it was for the bugfix or thing like that
16:29:11preglowlower limit: whatever you want, but we'll need a better resampler
16:29:15Bgerpreglow: seems impossible for vorbis -q10 ...
16:29:23preglowBger: indeed
16:29:44linuxstbMaybe we should call 150% a random number like iriver - e.g. +8
16:29:55Paul_The_NerdForget numbers.
16:30:01preglowdunno how we should handle that, perhaps just cap it dynamically based on if dropouts are occuring
16:30:02Paul_The_NerdHave "Normal" "Fast" "Faster" and "Fastest"
16:30:27Paul_The_NerdFastest can be "as fast as the codec will allow" and the other two can be arbitrary points between that and normal. :-P
16:30:53Bgergotta go, bye for now
16:30:54Benacooland if we want to slow down playback ... lol
16:30:55 Quit Bger (""In the other world, in paradise, the beauty of women surpassed even the beauty of Bulgarian women" Adaloloddin Mohammed Balh)
16:31:19preglowno, we need accurate limits
16:31:20Paul_The_Nerd"Slow" "Slower" "My five yeard old nephew playing it on a xylophone"
16:31:20BenacoolSlow Slower Slowes
16:31:30preglowi'd like it one could use rockbox for djing
16:32:01Benacool"My five yeard old nephew playing it on a xylophone" XD
16:32:21 Join Febs [0] (
16:34:00MirfleI finally managed to build. I added the unicode patch, and I see it made a "codepages" directory. what's that for?
16:34:22markunMirfle: For very large codepage convertors like chinese and japanese
16:34:49Mirflehebrew wouldn't need. right?
16:35:48perplexityJust another datapoint for the archives.. H3x0 can't play vorbis encoded at -b 256 and access the disk at the same time without skipping.. testing continues
16:36:43markunno, non of the iso-8859 fonts
16:36:53markuneh, codepages
16:37:43markunMirfle: does hebrew also have a mirrored question mark?
16:37:52markun(like arabic)
16:38:57Mirfleno. it's just the same as in english
16:39:50markunHm, that sucks a bit. then the bidi code should know if it's dealing with hebrew or arabic
16:40:25wubblaLinusN: ?
16:41:27preglowlinuxstb: i'll do a playback speed test now
16:41:42linuxstbUnscientific test - +5 playback speed using the iriver firmware takes 40 seconds for a 45 second track.
16:41:51linuxstbSo it's about a 10% speedup.
16:42:44Mirflecan I automatically create a .rockbox folder with all the stuff inside? (fonts, wps, themes, etc.)
16:42:57B4gderMirfle: make zip
16:43:19linuxstb...and then unzip that zip file.
16:43:57markunAfter selecting a theme, selecting a different wps and font and rebooting I get the original theme back..
16:43:59Mirflei did, but all the folders are empty
16:44:37B4gderthen you need to tell us more about your setup
16:45:02linuxstbMirfle: Do you have the "wps" directory (at the same level as apps, tools, firmware etc) ?
16:46:07Mirflesame with fonts
16:46:25B4gderand you can build rockbox fine?
16:46:38B4gderand you run 'make zip' in a build dir?
16:46:48Mirfleyes and yes
16:47:07B4gderthen start figuring out the details, cause I can't ;-)
16:47:22 Join webguest99 [0] (
16:47:27B4gderthat's how the daily/bleeding builds are made
16:47:31webguest99Hello everybody
16:47:34mirakI am reading some docs, but how can I know if it's currently how it works ?
16:47:43mirakabout multitheading for exemple
16:47:51webguest99who is Mikka?
16:47:59mirakhow much in this page is up to date ?
16:48:13ghode|afkwebguest99: Slasheri
16:48:24webguest99Mikka: thanks for your modification of yesterday, now the boot is faster
16:48:40webguest99a ok, then Thanks Slasheri :)
16:49:37webguest99is the same guy made playback for irivers and crossfade...
16:49:50webguest99Slasheri: we like you
16:50:05webguest99and like all others dev too ;)
16:50:44ghode|afkwebguest99: yes
16:51:17webguest99I love so much Rockbox
16:51:38webguest99I will make donation ;)
16:52:05webguest99cheers all
16:52:08Mirflewell, it compiled the plugins, but the zip thing didn't put them in rocks dir, and didn't even create a viewers dir
16:52:08 Quit webguest99 (Client Quit)
16:52:29Mirflefonts were'nt compiled at all
16:52:35mirakis rockbox multhreaded ?
16:52:36B4gderMirfle: something is weird in your setup
16:52:40B4gdermirak: yes
16:52:50mirakB4gder: so how much in the docs is up to date ?
16:52:57mirakwhat is the best ressource ?
16:53:05B4gderthe source is the best resource
16:53:09B4gderit is up-to-date! ;-)
16:53:36MirfleI didn't change anything unusual
16:53:43mirakB4gder: what is accurate in that stuff ?
16:53:51B4gdermost of it I'd guess
16:53:57B4gderthe threading has been the same for many years
16:54:04markunMirfle: except adding the unicode patch, right?
16:54:12mirakso why is he discussing about hwo it should be ?
16:54:21linuxstbMirfle: What is contained in the zip file when you do "make zip" ?
16:54:24mirakif it's already like that :D
16:54:29MirfleI tried also w/o and it's the same thing
16:54:31B4gdermirak: people are free to edit or discuss whatever they want
16:54:38B4gderI didn't read that page
16:54:58B4gderthere's not much to say nor know about the threading
16:55:03B4gderits is cooperative multi-tasking
16:55:15B4gderall tasks must yield() or sleep()
16:56:25 Join muesli- [0] (n=muesli_t@
16:56:27linuxstbmirak: If you go into the Debug menu in Rockbox, and then select the "View OS stacks" option, it will list all the threads along with their stack usage.
16:56:31B4gderMirfle: tools/ is what you need to debug
16:56:35Mirfleit didn't create a zip file, only a .rockbox dir with empty subdirs, and a file called "perl.exe.stackdump" which seems to be empty to
16:56:45B4gdersounds like a perl crash
16:57:19miraklinuxstb: I am running the simulator, I don't know how to go to the general menu
16:57:34linuxstbPress INSERT (I think).
16:57:43mirakthat's it
16:57:43B4gderthe threads are not identically treated in the simulator
16:57:47mirakand undocumented ^^
16:57:50B4gderonly simulated
16:58:44linuxstbAh yes, and there is no Debug menu in the sim.
16:58:51B4gderhehe, right
16:59:08linuxstbSo that wasn't a useful suggestion....
17:00:45 Join Cassandra [0] (
17:02:16markunDidn't get my Gigabeat today :(
17:03:00 Quit Zagor ("Client exiting")
17:03:04linuxstbAnyone know any shops selling iriver h1x0 remotes? Or is ebay the only option?
17:04:21markunlinuxstb: are you going to connect it to your ipod? :)
17:05:07mirakwhen the iriver goes on sleep mode, who is controling it ? The iriver firmware ?
17:05:22 Join lamed [0] (
17:05:22XavierGrwhat sleep mode?
17:05:23mirakIt shouldn't be possible to do an alarm clock isn't it ?
17:05:26mirakXavierGr: off
17:05:30mirakoff mode
17:05:43XavierGrwhen it is off it's off, hwat irver fw you talk about?
17:05:45mirakit's still running since there is a clock
17:05:54mirakXavierGr: it's never off
17:05:57***Saving seen data "./dancer.seen"
17:06:00XavierGrif it has an rtc
17:06:17XavierGroh and you are talking about h300 rght
17:06:18amiconnThe h300 has an rtc, the h1x0 doesn't
17:06:27mirakif it's off and on hold if you push a key it says it's olded
17:06:33XavierGrbut the h1x0 doesn't have a sleep mode...
17:06:36mirakso does it means it not really of ?
17:06:50lamedxaviergr: do you have the h120 battery specifications? I'd like to order one of those from a local battery store, so I need the model name, not just a site that sells them.
17:06:56amiconnThe CPU is completely off on both when they are off
17:06:59XavierGrand hasn't got implemented an alarm,clock function so no need to sleep mode
17:07:28lamedahhh nm gtg.
17:07:31 Quit lamed (Client Quit)
17:07:46mirakamiconn: how can it display that image ? with the locker ?
17:08:07amiconnWhat image?
17:08:25XavierGrmirak: when it is on hold the bootloader starts and sees the hold button, then the text is displayed
17:08:32 Quit Cassandra (" HydraIRC -> <- Go on, try it!")
17:08:35preglowthere is no BUTTON_PLAY for h1x0?
17:08:36 Join Cassandra [0] (
17:08:52XavierGrthere is BUTTON_ON
17:09:11preglowwell, how do i cope with code that uses BUTTON_PLAY?
17:09:15preglowaliasing button defs is bad, they say
17:09:26XavierGr#ifdefs :)
17:09:41XavierGrwell that is done in all the plugins iirc
17:09:57amiconnpreglow: Most app code doesn't use BUTTON_* directly
17:10:12XavierGrbut I've seen that in main code too.
17:10:24B4gdermost code should use rather use button aliases
17:10:25amiconnInstead, there are symbolic #defines
17:10:26preglowamiconn: screens.c does
17:10:38XavierGryou can always make a big define in the start of the code and use those instead
17:12:31amiconnpreglow: pitch_screen() does this because it was recorder-only so far
17:12:57 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
17:13:00preglowyeah, so i see
17:13:13amiconnIf it gets extended, it will need to have symbolic #defines as other app code
17:13:53amiconnThe player doesn't support pitching (well, technically it could support pitching by a few percent),
17:14:12amiconnand the Ondio doesn't have enough buttons to call the pitch screen from the wps
17:14:34amiconnHmm, it could be put in the context menu...
17:17:49 Quit skorpyon_ (Read error: 110 (Connection timed out))
17:18:00 Join skorpyon [0] (
17:19:33 Join bobwise [0] (
17:19:38preglowbut argh, i just wanted this to be a quick hack :/
17:20:52preglowsome commenting out of everything that mentions BUTTON_ON, and we're alright
17:20:54preglowno, we're not
17:21:13amiconnThe pitch screen is quite special
17:21:36B4gderah, so the devkit can't build uclpack
17:21:45B4gderhow fun
17:22:02amiconnOn archos, it's displayed as long as you hold ON, allowing to pitch up/down by 0.1% and temporarily pitch by +/-2 %
17:22:12t0maswish Bluechip luck ;)
17:22:25amiconnIf you let go the ON button, you're back in wps
17:22:47preglowon iriver i basically need to use the PLAY button for this
17:23:06amiconnIf we want it to work the same way, that is
17:23:14preglowwell, we do, i assume
17:23:23preglowbut, what to do with BUTTON_PLAY usage, then
17:23:24amiconnWell, on Ondio we can't
17:23:27preglowit's used all over the place
17:23:32mirakare the codec totally rewrote or mos of the original code can be used ?
17:23:38preglowmirak: most
17:23:44B4gdert0mas: I'm more worried about the flood of people that'll come nagging about this ;-)
17:23:46preglowmirak: with a tiny bit of new code
17:23:53Benacoolamiconn: why you don'y put it in same menu that where is the repeat, shuffle ans show files ?
17:23:54amiconnThe pitch screen is not yet converted to the gui api...
17:24:01mirakpreglow: they seem huge in size
17:24:02B4gderI better do some damage control
17:24:06preglowmirak: codecs often are
17:24:09preglowmirak: but not all are huge
17:24:21preglowamiconn: which means i can't use it? or?
17:24:39amiconnYou can... but it should be converted
17:24:45mirakpreglow: my ultimare goal is to do a video codec but I am far from that lol
17:24:52preglowmirak: yes, i can imagine
17:25:11mirakpreglow: Wett is working on that I have seen a sample video of his work
17:25:19amiconnB4gder: Do we support the devkit?
17:25:33amiconnIt surely is the devkit that needs to be fixed
17:25:38Benacoolmirak: xvid, divx or soemthing else ??
17:25:47B4gderyes, but the uclpack doesn't need to build for iriver
17:25:49 Join San||Away [0] (
17:26:04B4gderso I can at least limit the impact
17:27:05mirakBenacool: I think it's just uncompressed video
17:27:21mirakBenacool: lcd refreshrate seem to be a problem
17:27:31mirakI can't find the link to the video
17:27:34linuxstbB4gder: How do you know which target(s) the user wants to build for when they type "make" in tools?
17:27:45B4gderhehe, I don't
17:27:45linuxstbOr are you planning to only compile uclpack when it is needed?
17:27:46Benacoolmirak: it's a beginning ^^
17:27:58B4gderlinuxstb: but most people will build tools from a build dir
17:28:05B4gderand then I know
17:28:11linuxstbWhat do the devkit instructions say?
17:28:26B4gderyou're right, it'll be troublesome
17:28:32Benacoolmirak: ok
17:30:04Benacoolmirak: so we'll wait a while before we can see something like the original firmware can do ?
17:30:43mirakBenacool: certainly, and I don't even know if I can succeed
17:30:58mirakBenacool: but Wett did that on the H100 with greyxcale video
17:31:14mirakBenacool: it looks nice, I need to refind that video
17:31:31linuxstbWhat format video can the iriver h300 firmware play?
17:31:37markunyou could try to get ffmpeg's mpeg-4 codec to compile
17:31:38mirakonly xvid
17:32:02markunIt has ARM asm optimisations
17:32:03mirakxvid 10 fps 128kbit sound and the resolution is the same than the screen
17:32:22mirakbitrate max is 500 kbit for video
17:32:39mirakbut even at 150 it's still nice
17:33:09mirakI don't know if the problem is LCD refresh problems or cpu power to uncompress xvid
17:33:10linuxstbmarkun: You'll be OK on your 400MHz arm then :)
17:33:34markunIt's clocked at ~290 MHz I think
17:33:47mirakmarkun: what's your device ?
17:34:05markunThe screen is 2x as big as the H300's, so maybe 2x as slow?
17:34:15markunmirak: Gigabeat F40
17:34:29B4gderthat's 18bit too isn't it?
17:34:36 Join Philip [0] (
17:34:40markunIt should have arrived today..
17:34:57preglowamiconn: on iriver there is the problem that we don't have any more buttons that can be pressed simultaneously
17:35:06preglowamiconn: like for the BUTTON_PLAY | BUTTON_ON case
17:35:08miraklinuxstb: on the H300 the only problem with video is the low framerate. It's watchable but more frames would be perfect and we can't change the size of the screen so ...
17:35:13preglowno, i'm wrong
17:35:20preglowplay CAN be used for this
17:35:29amiconnpreglow: The pitch screen should be simpe on iriver
17:35:39amiconnBUTTON_ON stays the same
17:35:59amiconnThe directions stay the same as well
17:36:17miraklinuxstb: they could have give it to H100, it would have been grayscale :)
17:36:25PhilipI have noticed when deleting files it says press play to delete when you have to press navi...
17:36:46markunPhilip: known problem
17:37:00*amiconn want's to run around screaming
17:37:12preglowamiconn: nothing's stopping you
17:37:15markunamiconn: you fixed it?
17:37:25markunor just getting crazy?
17:37:44muesli-when i danger or ib doubt run in circles and shout :D
17:37:51amiconnI have no idea what's going wrong
17:38:57amiconnThis ucl code is strange. Sometimes it works correctly, sometimes it returns the correct length but the decompressed data is wrong, sometimes even the beginning of the decompressed data is correct
17:39:11muesli-mirak kewl
17:39:11preglowamiconn: at random?
17:39:11amiconnSometimes the decompressed length is _almost_ correct
17:39:39amiconnNot completely random, for the very same build it stays the same
17:39:52mirakmuesli-: that's Wett work
17:40:03Benacoolmirak: thanks ^^
17:40:14muesli-thats awesome..8fps
17:40:20muesli-and it looks nice though
17:40:37amiconn...but whenever I change something, it's impossible to predict how it will behave afterwards
17:41:33amiconnIt must be some really stupid mistake, but it seems that I am more stupid...
17:41:55amiconnIt's despairing
17:42:05Benacoolin color on the h2x0 it will be cool lol
17:42:54muesli-mirak are there further movies?
17:43:11*preglow changes pitch
17:43:15mirakmuesli-: ask Wett
17:43:24mirakmuesli-: I am trying to contact him since yesterday
17:43:27muesli-wett are there further movies?
17:43:42 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
17:44:10linuxstbWett's video player plays Rockbox's own ".rbv" format videos. These are uncompressed video files with mp3 audio (I think).
17:44:22XavierGrpreglow I had done a test on the iriver fw about the pitch changes
17:44:44preglowXavierGr: what is there to test?
17:44:50muesli-linuxstb will that player be implemented soon?
17:45:08linuxstbI've no idea - ask Wett...
17:45:19XavierGrI had recorded a 440hz pulse and then with a tuner I wrote down how frequency changes with the playback speed, though I dont know if this can tell us anything on speeds .e.t.c
17:45:24preglowso, anyone wanna test pitch changing?
17:45:26linuxstbmuesli-: AFAIK, he's the only person working on it.
17:45:54miraklinuxstb: I forgot to say that on iriver firmware video lenght is contrained to 1 hour. You need to split them.
17:46:14mirak1 our of video is around 80 mega byte
17:46:28mirakuncompressed video would be bigger than a divx !
17:46:45Mirfleabout my perl problem: when I run from the build dir i get this:
17:46:46markunmirak: yes, so get ffmpeg working :)
17:46:47Mirfle 5 [main] perl 2920 fixup_mmaps_after_fork: WARNING: VirtualProtectEx to return to previous state in parent failed for MAP_PRIVATE address 0x480000, Win32 error 87
17:46:48Mirfle 401 [main] perl 2920 fixup_mmaps_after_fork: WARNING: VirtualProtect to copy protection to child failed forMAP_PRIVATE address 0x480000, Win32 error 87
17:46:50DBUGEnqueued KICK Mirfle
17:46:50Mirfle 560 [main] perl 2920 fixup_mmaps_after_fork: ReadProcessMemory (2nd try) failed for MAP_PRIVATE address 0x480000, Win32 error 87
17:46:51MirfleC:\Naftali\Iriver\ROCKBO~3\bin\perl.exe (2920): *** recreate_mmaps_after_fork_failed
17:46:53Mirfle 12 [main] perl 3160 fork_parent: child 2920 died waiting for dll loading
17:47:03linuxstbMirfle: You have windows/cygwin/perl problems....
17:47:21muesli-80mb for 1h..great deal
17:47:28preglowi can do 150% pitch change at vorbis q5
17:47:40mirakmarkun: my main motivation joinging rockbox is to be able to play over 10fps videos ^^
17:47:47Mirfleany ideas what I do about it?
17:47:47Benacool<mirak>: one hour is more 200 mb
17:48:02preglowmirak: then congrats, you've got quite a bit of work ahead of you
17:48:03mirakBenacool: uncompressed certainly
17:48:14muesli-doesnt matter..there's enough space on 60gb drive :D
17:48:15linuxstbIt will be very hard work getting compressed video playing at even 10fps. Let alone more.
17:48:16Benacool<mirak>: in xvid
17:48:22mirakBenacool: I am talking about the xvid video I used
17:48:33Mirflegotta go. bye
17:48:36 Part Mirfle
17:49:03mirakBenacool: ho sure in big resolutions, but for H300 it's something like 180*130
17:49:10Benacooli have a video oh 1h15 of 275 mb
17:49:15preglowvorbis never boosts at 50% speed :PPP
17:49:26Benacoolin xvid
17:49:31linuxstbpreglow: Have you solved the AAC problem then? :)
17:50:39Benacoolno the h300 is 220x176 ... 10 fps ... this video is in 400 kbps
17:50:43amiconnpreglow: ogg -q10 @200% ?
17:50:44linuxstbIf you can't make a codec realtime, then slow down time...
17:50:44mirakwhat EID do you use ?
17:50:58preglowamiconn: MAN, it takes long to change speed to 200%
17:51:13preglowa good minute
17:51:13BenacoolVideo: XVID 220x176 10.00fps [Stream 00]
17:51:13BenacoolAudio: MPEG Audio Layer 3 44100Hz stereo 128Kbps [Stream 01]
17:51:22Benacool220 x 176
17:51:27Benacool= 275 mb
17:52:03amiconnGah, who broke the pitch screen on recorder?
17:52:04Benacooloh damn
17:52:08Benacoolmy bad
17:52:18Benacoolit's the file not cut i two xD
17:52:22 Quit bobwise ("CGI:IRC (EOF)")
17:52:27preglowamiconn: broken how? works on h120
17:52:36amiconnI can't enter it
17:52:38Benacool36 min = 133 mb
17:53:41amiconnme silly
17:53:44amiconnIt does work
17:54:44mirakBenacool: I have one of 85mb for 130 kbits/s I think
17:55:02mirak46 minutes
17:55:02Benacooloh me its more 400 kbps
17:55:04linuxstbpreglow: Does the resampler handle 96KHz audio yet?
17:55:20mirakBenacool: yep I did test and realised high bitrate doesn't help much
17:55:31mirakBenacool: you have rockbox on your H300 ?
17:55:42mirakBenacool: it's working ?
17:55:56mirakBenacool: what's your firmware ?
17:55:59Benacoolabout music you talk ?
17:56:04Benacool1.28 k
17:56:14mirakBenacool: you have a european or US ?
17:56:40mirakmusic works ?
17:56:53 Quit tvelocity ("Leaving")
17:57:02mirakwhere you afraid to brick it ?
17:57:05Benacoolwith booltloader v2 and the lastest builds
17:57:05preglowlinuxstb: no, but it will very soon
17:57:08mirakdid you built it
17:57:18Benacoolnot really
17:57:22preglowlinuxstb: just trying to figure out how to deal with it, i'm thinking of introducing some 64 bit maths to make it precise enough
17:57:29mirakBenacool: ok I will try that
17:58:29mirakBenacool: what is not working ?
17:58:38Benacoola rare bug can happen (rockbox firmware is unable to boot ... black screen) but you can normally boot on the original firmware by plugging the player with the adapter or the usb plug
17:58:58mirakmmm I don't understand
17:59:10Benacoolbut you do this at your own risks !
17:59:25mirakwhat do you mean by pluggin on usb ?
17:59:56Benacoolrokbox doesnt support usb for now
18:00:11Benacoolso when you plug it when the player is off the origina lfirmware load
18:00:31Benacoolbut sorry i have to go
18:00:46Benacoolall the instructions are on the rockbox website anyway ...
18:05:56muesli-*PANIC* mount: 0 means?
18:07:29preglowthere, 200% pitch
18:07:31preglowsounds like a dream
18:07:45linuxstbmuesli-: What gives you that error message? The bootloader?
18:08:19preglowMAN, vorbis struggles at 200%
18:08:44 Part Polo_o
18:08:51muesli-i've just plugged my 140 in...rockbox was empty..recopied the bleeding edge, unplugged and got his msg
18:08:59linuxstbpreglow: Have you tried many codecs? Does everything apart from vorbis work OK at 200%?
18:09:00preglowa dropout or two at q5 and 200% pitch
18:09:07preglowlinuxstb: yeah
18:10:02 Join Polo_o [0] (
18:10:26preglowi need to enter pitch screen again for the pitch change to take effect at track change
18:10:31linuxstbmuesli-: It's an error caused by the usb code not being able to access the hard drive. Is your battery low?
18:10:42 Part Polo_o
18:11:00muesli-no its not..but the hdd may not be connected properly
18:11:19muesli-so its possible that there was no proper connection
18:11:22preglowlinuxstb: 200% is clearly a fringe case
18:11:39preglowlinuxstb: ogg has occasional dropouts, lossy wavpack _just_ manages to survive
18:12:14preglowflac now has a small boost ratio ;)
18:12:32preglowstill below 10%
18:12:43linuxstbI love that decoder.
18:12:46preglowme too
18:12:50preglowit's insanely efficient by our standards
18:13:14preglowflac could take 300% easily
18:13:31preglowwhat to do?
18:13:44preglowjust let skipping occur, or go for 150% for now?
18:13:52linuxstbI think you should let skipping occur.
18:13:58preglowthe code isn't exactly elegant as it is either
18:14:03preglowso i don't know if i should commit it
18:14:47amiconnPerhaps make it work across track changes before? Or better after commit?
18:15:14preglowmake it work across track changes first
18:15:15amiconnHow about mp3?
18:15:20preglowmp3 works fine
18:15:47preglow223kbps at 200% works just fine
18:16:00preglowit's straining, but that is to be expected
18:16:13amiconnBtw, getting to 200% takes almost a minute on archos too
18:16:16preglowalso: currently it overrides ordinary resampling, i need to fix that
18:16:52preglowbut yeah, i've basically just implemented sound_set_pitch and sound_get_pitch in dsp.c
18:17:02preglowthey should never be seen by a HWCODEC platform anyway
18:17:20preglowignore the fact that that define doesn't exist, please, i'll look stupid
18:18:01preglowpitch change works across track changes on archos, yes?
18:19:04preglowand also, will anyone kill me for using defined(IRIVER_H100_SERIES) instead of CONFIG_KEYPAD == H100_KEYPAD?
18:19:09preglowi think the first is cleaner
18:19:49amiconnyes it does work across track changes
18:20:04preglowamiconn: of course, it's because of SET_SAMPLE_FREQUENCY
18:20:51amiconnWhere do you want to used that ifdef?
18:21:11amiconnFor the keymapping, CONFIG_KEYPAD should be used
18:21:34amiconn...and btw, I don't see a reason why it shouldn't work on H300
18:21:41preglowamiconn: with the same keys?
18:21:59preglowamiconn: and yeah, currently i just did the button remapping with one #define in the start of the routine...
18:22:27amiconnYes, with the same keys
18:22:47preglowthought i'd leave that for someone with a unit, but if you say the keys will work, then i guess we're ok
18:22:53amiconnH300 can potentially use more buttons at the same time
18:23:32mirakis there reports of bricked H300 with the new bootloader ?
18:24:42amiconnH1x0 has Play and Stop on port pins, H300 has Play, Rec and Mode on port pins
18:25:07linuxstbpreglow: There are a lot of ((CONFIG_KEYPAD==H100_KEYPAD) || (CONFIG_KEYPAD==H300_KEYPAD)) conditions already in Rockbox.
18:25:25amiconnYes, because the two are almost identical
18:26:20linuxstbpreglow: What key mapping do you use?
18:27:58preglowlinuxstb: just #define BUTTON_PLAY BUTTON_SELECT
18:28:13amiconnYou shouldn't do that
18:28:21amiconnAliasing is BAD
18:29:09amiconnJust introcude a symbolic #define PITCH_PAUSE
18:29:31amiconn...and set it to BUTTON_PLAY for archos recorders, and to BUTTON_SELECT for iriver
18:29:45linuxstbSo how do you change the pitch?
18:30:03amiconnlinuxstb: Press & hold Play, the use the joystick
18:33:26 Join Polo_o [0] (
18:35:53preglowamiconn: the recoder gaps a bit when changing playback speed, yes?
18:36:14amiconnThat's unavoidable
18:36:34 Join goa [0] (
18:38:42 Quit B4gder ("time to say moo")
18:40:49preglowok, i need to start comparing the delta to 1 instead of comparing the sample rate to NATIVE_FREQUENCY
18:41:02preglowfor this to work
18:41:06preglowas usual, more work than expected
18:41:08preglowthink i'll go make dinner
18:48:18 Quit Philip ("bye everybody")
18:49:00 Join skorpyon_ [0] (
18:50:36 Quit muesli- ("ich will Kühe!!!")
18:51:21 Join _FireFly_ [0] (
18:51:25 Part Polo_o
19:00:26 Nick San||Away is now known as San (
19:00:37SanAny changes in the latest builds?
19:03:44 Join Bger [0] (n=Bager@
19:03:56Sanhey Badger
19:05:30 Quit skorpyon (Read error: 110 (Connection timed out))
19:06:01***Saving seen data "./dancer.seen"
19:07:21 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:08:43 Join jaydpb [0] (
19:10:42 Join webguest66 [0] (
19:10:46Sanis there a screendump on the H300 yet?
19:12:02webguest66San, if you want to know what the changes are, click on the appropiate build, there is always a brief descroption of the change
19:12:57^BeN^the last change was in 28 Nov 10:05?
19:13:00webguest66or click on the daily buld wherein all the builds with decsriptions are listed
19:14:18webguest66go to the daily builds of rockbox's site look, all the info is there, just a click away
19:14:42Sanyou meen there?^^
19:14:59^BeN^i know but yesterday the page isnt update for me
19:15:07webguest66there ^^^
19:15:09^BeN^you can check for a moment?
19:15:26 Join KN|stiff [0] (
19:17:06webguest66click on the day date stamp, it'll list who and what has been done
19:20:08 Join DangerousDan [0] (
19:21:23 Part webguest66
19:23:19CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
19:23:19*LinusN has an iaudio x5
19:23:46linuxstbHow is it?
19:23:49LinusNi must say i like the form factor
19:24:05LinusNthe joystick feels better than the h100
19:25:33linuxstbHooked up the bdm yet?
19:26:04linuxstbDoes it have an lcd remote?
19:27:06ghode|afkLinusN: that isn't the austriancoder was working on is it?
19:27:27LinusNit has an lcd remote, but it is optional :-(
19:27:58LinusNghode|afk: yes, he was working on iaudio x5 iirc (or was it m3?)
19:28:26ghode|afkit was the x5
19:34:13 Join _DangerousDan [0] (
19:35:46linuxstbAnyone know what the record key is mapped to in the h300 win32 sim?
19:36:49LinusNi assume it's the same as the h100 sim
19:37:49linuxstbI found it - it's the backslash on the numeric keypad.
19:37:51_FireFly_f1 or divide id it's the same as the h100
19:39:37_FireFly_linuxstb: or f1 :)
19:40:30linuxstbYes, or F1 :). I'm learning...
19:40:36 Quit Cassandra (" HydraIRC -> <- IRC with a difference")
19:40:48_FireFly_button.c in win32-sim
19:44:49 Join Guest229461 [0] (
19:44:49 Quit KN|stiff (Read error: 104 (Connection reset by peer))
19:44:55 Quit Guest229461 (Remote closed the connection)
19:49:13 Quit Febs ("CGI:IRC (EOF)")
19:50:28 Quit DangerousDan (Read error: 110 (Connection timed out))
19:54:32 Quit Bger (Read error: 110 (Connection timed out))
19:58:10 Join Bger [0] (n=Bager@
20:00:15 Join mrah [0] (
20:00:20SanCan I make a suggestion?
20:00:54SanYou Know the crossfade, I think you should make it instad of seconds, in miliseconds
20:01:29XavierGrI guess users can never be happy....
20:02:48XavierGrI can accept half-second steps but milliseconds is too much IMHO!
20:03:52Sanyou see, i listen to dt, and all their songs link into one another, and there is no time for the millisecond but I got it crossfading into it and it is alright for now
20:04:02SanI will just re-rip them
20:04:24 Join iobound [0] (
20:04:40ioboundhi, just wanted to report a successful rbx install on a int H340
20:05:01XavierGrSan: are you talking about gapless?
20:05:53XavierGrwell then you don't need crossfade
20:06:11Sanbut, the cross fading is nice aswell, when skipping from say.,... acdc to DT
20:06:18XavierGryou need to rip those songs with a stable lame encoder.
20:06:20^BeN^what is gapless?
20:06:28XavierGryes that's nice too.
20:06:34ioboundis radio supported on h300 yet? how do i start it?
20:06:37Sanwhere there is no gaps in between the songs
20:06:43Santhere is no radio yet
20:06:46XavierGrThough crossfade will ruin the mood in 2 gapless songs
20:06:51ioboundoh ok, the wiki said 90% done
20:07:02DBUGEnqueued KICK San
20:07:02*San 14Skinny Dips in a sea of 07Dream Theater - Live Scenes From New York Disc - Through Her Eyes1415
20:07:24XavierGrso i prod slasheri to add another crossfade option which will enable crossfade only on change of tracks.
20:07:40Sanin milliseconds
20:07:55Sancause 1 is too much and 0 is too little
20:08:08_FireFly_crossfade does only work on track change
20:08:15_FireFly_why then an extra option ??
20:08:25Sanyeah, but from one band/directory to another different one
20:08:40Sanif it is the same band or directory, leave it gapless
20:08:53Sandoes vorbs have gapless on RB yet?
20:09:09XavierGrno it doesn't work only on track change
20:09:25Sandoes AAC?
20:09:27XavierGrI meant on manuall track change
20:09:33linuxstbSan: No.
20:09:36XavierGrwhen you ffw or rw
20:09:56XavierGrbecause if there are 2 songs with no gaps the whole mood is spoiled.
20:09:57linuxstbSan: That's a limitation of AAC files - silence is added at the start and end of every track.
20:10:07XavierGr(and crossfade set to on_
20:10:16ioboundit'd be nice if the wiki mentioned that flashing the firmware of an h300 resets the iriver firmware's settings
20:10:23linuxstbSan: We can probably do something about it, but the priority is to optimise the decoder first.
20:10:50linuxstbiobound: It's a wiki....
20:11:04linuxstbYou can make that change.
20:11:09ioboundoh but it's such a hassle to set up an account and figure out the wiki dialect etc
20:11:18Sanok, thanks linuxstb
20:11:41ioboundif you think it's a useful addition, i'll do it
20:11:43ghode|afkyeah it's also such a hassle to make a firmware from scratch, why bother :P
20:12:24Saniobound, would you be able to speed up rockboy?
20:12:34Sanlike, get it at normal speed?
20:12:38ioboundi don't know? :)
20:12:48CtcpIgnored 2 channel CTCP requests in 0 seconds at the last flood
20:12:48*XavierGr reboots
20:12:49ioboundi haven't read the source
20:12:57ioboundit would be cool to add (back?) color support too
20:13:07 Part XavierGr
20:13:17Santhat rould rock15
20:13:26Sanwould it be hard
20:13:34Sando you think you could do it?
20:14:13ioboundthat's hard to say when i haven't read the source. i guess we'd have to try to figure out where the bottlenecks are, experiment with gcc's various -O flags, and perhaps maybe try to hand-write inner loops in 68k asm?
20:14:34Sandoes anyone know whats wrong with it? Cause a guy my dad works with codes programs in C for the company. He could do it probally.
20:15:00ioboundmaybe, maybe not
20:15:28ioboundsan, i don't think there's anything "wrong" with it? maybe it just is that slow, maybe there are clever ways to do more work with less effort to make it run faster, who knows..
20:17:25 Join JoeBorn [0] (
20:18:20ioboundyeah i've got scummvm running on my 6600 phone
20:18:38Sani tought you said H300 and i was like, WTF?!
20:18:45 Join XavierGr [0] (
20:19:24ioboundi guess it would probably be easier if rockbox shipped with a regular C library :)
20:19:29 Join Amar [0] (
20:19:29Sanwould it be hard to port?
20:20:00_FireFly_iobound: why to blowed the firmware with a big c-library ??
20:20:39ioboundwhy bloat up the firmware with a big c library?
20:20:59_FireFly_that was the word : bloat
20:21:05ioboundi just assumed that scummvm probably relies on a lot of standard C library routines
20:21:43Santhey said the H300 has the right specs to run it and it can be ported
20:22:14ioboundi'm sure it can, after all, the original monkey island ran on a regular amiga 500....
20:23:53ioboundbut seeing as scummvm is written for a regular computer with a full blown C library environment, you'd probably have to write a lot of boring plumbing (c library stuff)
20:25:22Sango for it
20:25:36ioboundi'd rather try to get libsidplay or libdumb working
20:25:50ioboundscummvm on a mobile unit is fun for five minutes, then you get tired of the tiny screen and lack of mouse
20:25:56linuxstbiobound: scummvm is worse than that - it's C++
20:26:15 Quit Maxime` (Read error: 104 (Connection reset by peer))
20:26:29ioboundok, then it's even more of an effort :)
20:26:45linuxstbMost of the standard C library is already in Rockbox.
20:26:46 Join solexx [0] (
20:26:57 Join Maxime [0] (
20:27:00linuxstbAt least, everything that's relevant to an embedded device.
20:27:36ioboundbut there's just a "dummy" malloc with a no-op free(), isn't it? i'd imagine things like scummvm would do allocs and reallocs all the time and you'd probably run out of memory fairly fast
20:28:19_FireFly_malloc is a bad thing for embedded devices with small amount of ram afaik
20:28:21linuxstbIf scummvm needed malloc, then it could have it's own implementation. We don't want it in core rockbox.
20:28:36ioboundof course not, i'm just saying it would have to be implemented somewhere
20:28:49ioboundmaybe dietlibc or whatever could be used
20:29:52linuxstbEven dietlibc would probably be overkill for Rockbox - most of the functionality of a C library is already present.
20:30:51ioboundyeah. but we wouldn't need to re-invent a full malloc() algorithm :)
20:31:03_FireFly_i have a small function set for reading files from a zip-file with working crc :)
20:31:23linuxstbNo - we take a simple malloc implementation from somewhere. I believe Bagder has stated in the past he has one waiting if Rockbox ever needed it.
20:31:31o0260on't need to re-invent a full malloc() algorithm :)
20:31:31o0260o<_FireFly_> i have a small function set for reading files from a zip-file with working crc :)
20:31:31o0260o<linuxstb> No - we take a simple malloc implementation from somewhere. I believe Bagder has stated in the past he has one waiting if Rockbox ever
20:31:39 Part o0260o
20:31:48chopped_porkthat was cool
20:31:55linuxstbIs there an echo in here?
20:32:13 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
20:35:29ioboundwhen i started charging, the familiar iriver firmware with its "charging.." screen popped up. did the rbx bootloader detect charging and switch to that, or how does that work?
20:35:29 Quit Maxime (Read error: 104 (Connection reset by peer))
20:36:21iobounddoes the win32 sum support audio?
20:36:34_FireFly_but only uncompressed zip-files because i don't think that a uncompressen algo would be efficient
20:36:46_FireFly_on the current rockbox targets
20:36:52ioboundfirefly, well, for MS-DOS is only 2700 bytes or so
20:37:32linuxstbThat's a lot of bytes for Rockbox.
20:37:43_FireFly_iobound: but you need afaik malloc for an uncompressen algo
20:37:54 Join Maxime [0] (
20:38:11ioboundthere's no "malloc()" provided to an MS-DOS .com file..
20:38:40 Join Lear [0] (
20:39:06 Quit solexx_ (Read error: 110 (Connection timed out))
20:39:11iobounddoes the WPS logos flicker on your H300s too?
20:39:22_FireFly_i don't have an h3xx
20:42:25linuxstbiobound: Yes, but I think you have to enable sim sound in the autoconf.h file that the configure script generates in your build directory.
20:42:37ioboundi just tried editing autoconf.h and the compile completed at least
20:43:05Sanany chance of getting color wps's? (I actually prefare the black and white ones though, just wondering)
20:43:58linuxstbOf course. It's just that very few developers have colour devices at the moment.
20:44:42mirakwhat is the problem with the floating point unit ?
20:44:49mirakor the lack of it
20:45:16_FireFly_does read_bmp_file already support color-bitmaps ??
20:46:05linuxstbmirak: That's the problem - the lack of an FPU. It means that floating point calculations are done very slowly in software.
20:46:06ioboundwhat is the proper way to populate the archos/.rockbox directory when using the simulator? just manually copy apps/codecs/*.codec etc?
20:46:15linuxstbiobound: "make install"
20:46:16_FireFly_make install
20:46:20ioboundah, thanks
20:46:57 Join Cassandra [0] (
20:46:57ioboundugh, i don't have cygwin unzip... gotta re-run setup.exe
20:49:06ioboundcool, sound in win32 simulator works
20:49:14linuxstb_FireFly_: No, read_bmp_file() only supports mono bitmaps.
20:49:58_FireFly_then this needs updated first to have color-images in wps workable ;)
20:50:22linuxstbThe bmp2rb.c program will probably be a big help.
20:50:52amiconnAdding support for all possible bmp formats should be straightforward
20:51:07_FireFly_also for RLE-encoded bmps ??
20:51:11ioboundwhy does some plugins appear twice in the "Open with.."?
20:51:14amiconn..but that's the least problem to solve for colour support in wps
20:51:37amiconn_FireFly_: No, but all uncompressed variants
20:51:44_FireFly_ok :)
20:52:00amiconnI would leave out some space wasters though, like 32 bit
20:52:19linuxstbAre tools that specific in letting you choose the bmp format?
20:52:56_FireFly_yepp max support color-depth would be the max supported color depth of the color-lcd-screen
20:53:08 Quit dpassen1 (Read error: 110 (Connection timed out))
20:53:30_FireFly_afaik is without compressen the only difference in bitmaps the color-depth
20:53:49mirakwho won the 2002 world cup ?
20:54:13amiconnI want to support 1..24 bit bmp on all devices
20:54:18amiconnNo major problem
20:54:41amiconnIt's basically the same as bmp2rb, just without the generous intermediate buffers
20:55:24_FireFly_yeah and the bitmap-buffer is then fbdata instead of char
20:55:25amiconn...but I don't bother as long as there are other major problems which prevent colour (or even greyscale) support in wps
20:57:07amiconnlinuxstb: For truecolour images, Photoshop Elements (was included with my laptop) offers 16bit RGB555, RGB565, RGB444, 24 bit RGB888 and 32 bit RGB888
20:57:47_FireFly_a first step to reduce the used ram-space of the wps-format buffer would be to commit the new wps-file-loader which is on tracker
20:58:20linuxstbI just tried the gimp, and it just gave me "Windows BMP image". No options at all.
21:03:35Sandoes shorten work on the H300?
21:04:10 Join iobound_ [0] (
21:04:35mirakLinusN: what kind of floating points calculs needs to be done ?
21:05:23mirakLinusN: sorry wrong nick
21:05:38miraklinuxstb: what kind of floating points calculs needs to be done ?
21:06:03***Saving seen data "./dancer.seen"
21:06:57 Join solexx___ [0] (
21:08:37Wettmirak: you wanted to ask me smthg ?
21:08:37 Join iobound__ [0] (
21:10:00preglowlinuxstb: well, just have a look at an mpeg decoder and you'll see
21:10:34preglowmirak: you don't have to do the calculations in floating point, but i think that's how most of the source code you'll find does it
21:14:04 Join JonSenior [0] (
21:16:55 Quit solexx (Read error: 110 (Connection timed out))
21:19:10amiconnNow I made a test plugin for the ucl decompressor, and it works on the very first try!
21:19:27 Join jochen [0] (
21:19:35jochengood evening
21:19:50_FireFly_amiconn: congratulation
21:19:58jochenis anyone here investigating the remote ticking?
21:19:59amiconnCopy & paste, some file operations around that, thrown into my player's viewer dir
21:20:38amiconnBut why the %&$§ doesn't it work in the decompressing loader then?
21:21:09 Quit iobound (Read error: 110 (Connection timed out))
21:21:53^BeN^where i need to do this:
21:22:42jochenwell, is there any chance that the remote ticking issue is related to the compiler used?
21:22:58 Join muesli_- [0] (
21:23:38 Join petur [0] (
21:23:52 Quit iobound_ (Read error: 110 (Connection timed out))
21:24:28 Join iobound_ [0] (
21:25:08iobound_I'm trying to write a plugin that links with libdumb, but I'm getting "Undefined reference to _debugf" when it tries to "DLL dumb.rock". any hints?
21:25:57jocheni've "installed" the latest daily build on my player this morning and afterwards i had that remote ticking. then i've built rockbox from cvs and it was gone. so maybe it has something to to with me using a different compiler.
21:27:47iobound_nevermind, i got it working by commenting out the DEBUGF stuff in that one dumb file
21:27:58preglowwill anyone shoot me if i combine the separate channel resampler passes into one pass which does both?
21:28:10_FireFly_me not
21:28:18*LinusN loads his gun
21:28:33LinusNpreglow: why would we shoot you?
21:28:50jochenis that a trick question? *lol
21:28:59iobound_how do I "register" my plugin so it appears in "Open with.." ?
21:29:17preglowiobound_: edit viewers.list
21:29:21preglowin .rockbox
21:29:26Sanwhat plugin did you make?
21:29:36iobound_viewers.config ?
21:29:40preglowyeah, whatever
21:29:47iobound_what do those numbers mean
21:29:52preglowit's an icon
21:30:33jochenwell then, just for the logs, i used gcc-3.4.1
21:30:34*San decides to neaten up and rearragne all the files on his PC
21:30:38preglowwe'll probably end up with slightly slower mono resampling and slightly faster stereo resampling
21:31:08LinusNpreglow: a fair tradeoff imho
21:31:22amiconnI tested with a number of .ucl files now. They were all uncompressed correctly.
21:31:34LinusNwhat was the bug?
21:31:42amiconnNo, I mean in my test plugin
21:31:43iobound_yey my plugin crashed and burned rockboxui.exe :(
21:31:51amiconnIt still doesn't work in my loader
21:32:23amiconnI made a simple test plugin (as a viewer)
21:32:32LinusNwhat does your loader do?
21:32:41amiconnIt contains literally the same uncompressing code as my loader
21:32:42LinusNhow is it linked?
21:32:48amiconnIt's in cvs
21:33:05amiconnIt's not yet integrated into the build system
21:33:11^BeN^linusn: where i need to write this:
21:33:33 Join mashalla [0] (
21:33:49linuxstb^BeN^: As that page says, in the .bash_profile
21:34:02linuxstbYou can find that in your cygwin home directory.
21:34:13^BeN^thereis no a file in this name
21:34:19jochen^BeN^: ~/.bash_profile
21:34:25linuxstbThen just create one.
21:34:38jochenit's a hidden file.
21:35:03^BeN^ahh ok
21:35:20 Quit iobound__ (Read error: 110 (Connection timed out))
21:35:24^BeN^no file
21:35:38LinusNamiconn: you put the image in .data?
21:35:38preglowLinusN: though i really should make two separate routines, where one does mono, and one does stereo, but that'll mean some copied and pasted code
21:35:48 Quit jaydpb ("CGI:IRC (EOF)")
21:35:49amiconnLinusN: Yes
21:35:57*petur is happy he installed tortoisecvs on his windows box - much better
21:36:00preglowi don't know what your position is on that kind of ugliness
21:36:01LinusNhow does it end up in the ajz file?
21:36:30linuxstbpreglow: Can you use macros or inline functions to reduce the duplication?
21:36:30amiconnWell, the image is converted into an .o by objcopy magic
21:36:35LinusNobjcopy doesn't include the .data segment when converting to binary
21:36:37amiconn(in the Makefile)
21:36:47LinusNonly the data copy
21:36:57LinusNput the image in rodara instead
21:37:17*amiconn wonders how all other targets do work then
21:37:23LinusNthey have a data copy is included
21:37:30amiconn.bss is not
21:37:45amiconnThe datacopy routine is only for rom-based execution
21:38:07preglowlinuxstb: well, i guess so
21:38:07 Quit Amar ("CGI:IRC (EOF)")
21:38:23preglowbut it wont be pretty...
21:38:49LinusNamiconn: have you double checked the ajz image with a hex editor?
21:39:06LinusNamiconn: how can i try it?
21:39:09amiconnThe .ajz is scrambled; of course I checked the .bin
21:39:48amiconnWith current cvs, put a v1 rockbox.ucl (not rombox.ucl) into firmware/decompressor
21:39:54amiconnThen make
21:40:05amiconnIt should produce a compressed.bin
21:40:16amiconn...which then needs to be scrambled into .ajz
21:40:32amiconn*sometimes* it works...
21:41:47^BeN^[22:33:15] <^BeN^> LinusN: where i need to write this:
21:42:28LinusN^BeN^: jochen answered your question
21:42:41^BeN^i see but i dont see the file
21:42:45^BeN^in the folder
21:42:47LinusNls -a
21:42:47linuxstbCreate one.
21:42:50jochen^BeN^: just type "echo PATH=${PATH}:/opt/sh1/bin >> ~/.bash_profile"
21:43:10^BeN^ok now its clear
21:44:40preglowlinuxstb: the macro would pretty much have to take some code as an argument...
21:45:00linuxstbThat doesn't sound nice.
21:45:04preglowindeed not
21:45:17preglowa macro with another macro as argument = w00t
21:45:27amiconnYou could use similar trickery as the ucl sources
21:45:38amiconnThis uses recursive #include ...
21:46:14preglowi'd rather just bloody copy and paste each time
21:46:25amiconnIndeed, some files use recursive #include
21:46:47amiconnIn th first pass they define some macros, and even the header of a function
21:47:04amiconn...which is then filled in in the second pass
21:47:30t0massounds like a slow-as-hell-but-it-works Microsoft approach?
21:47:47jochenmicroWHAT? *fg*
21:47:52iobound_well my newbie attempt at making a dumb plugin crashes, when I run rockboxui.exe in gdb, bt shows a crash in _libwinmm_a_iname() (during loading of the .mod). any hints what this means?
21:47:54amiconnIt works, and is easier to understand than one might think
21:47:59linuxstbpreglow: My vote goes with copy and paste. You (or someone else) can always simplify it in the future.
21:48:19*t0mas has been reading msdn docs the whole day...
21:48:41t0masyou just start to believe anything about MS...
21:48:54t0masanybody heard of the NT-2-bit-patch?
21:49:06JonSeniorLinusN: When we spoke the other day, you suggested USB as an area to work on. I've had a look at the Wiki and can't seem to find any differences between the basic USB (ie. Non-USBTG) setup on either. I'm sure I remember seeing something somewhere though, so if I've missed something, please point me at it.
21:49:21jochendon't do that, i'm not sure whether this will affect your programming skills in a positive way t0mas *g*
21:49:22t0maschanging a few bytes in the NT kernel can disable the whole security system...
21:49:47t0masjochen: well... it was about hacking the NT kernel... writing drivers... in light of the sony rootkit
21:50:27t0maswhat sony did was a simple version... it can be done with a much broader effect... not just hiding files...
21:50:36jochenand you've found information on kernel hacking in the msdn?
21:50:42iobound_well what do you expect, when you modify the kernel, you modify the kernel! :)
21:50:47t0masssst ;)
21:50:50LinusNJonSenior: i discovered that the usb mode needs two power sources
21:50:54t0masI just read that for function calls
21:51:04jocheni didn't say a thing.
21:51:04iobound_i'm sure you could do the same for linux with modules or just by patching /proc/kmem
21:51:18LinusNand we might do it wrong in the h100 as well
21:51:27t0masiobound_: as root yes
21:51:34amiconnHmm, two power sources?
21:51:35JonSeniorLinusN: Ahh
21:51:37t0masbut do you know a lot of people who work as root the whole day?
21:51:47JonSeniorTime to start reading data sheets then!
21:51:55linuxstbt0mas: Sadly, yes.
21:51:56amiconnLinusN: Maybe that's the reason for the USB hang some users experienced? (H1x0)
21:51:57iobound_don't you need Administrator credentials to write to the NT kernel?
21:52:03jocheni'm afraid i can imagine that t0mas...
21:52:10t0masiobound_: jup, but everybody is admin in NT...
21:52:16t0mastake a look at win XP home...
21:52:21amiconnt0mas: huh?
21:52:25iobound_sure, but that's a HOME OS
21:52:35t0masI dare to day that at least 80% of the dumb users is running XP Home
21:52:53amiconnXP Home also knows the difference
21:52:56iobound_so if everybody ran linux, they would never need to run as root?
21:53:03jochenthat's why i like diskless clients, mounting the root fs ro from a nfs server. ;)
21:53:14t0mastry to start X as root in some "first use" distro
21:53:17t0masit will warn you...
21:53:19LinusNamiconn: yes
21:53:29iobound_i'm sorry, i fail to see any argument here :)
21:53:53t0maswell... you have to agree that almost all win XP users are using it as an Admin user?
21:54:04amiconnThe point is, you can run windows as admin all the time, and you can run linux as root all the time
21:54:10amiconnBoth is wrong
21:54:14t0masIn linux this is far less common... some first-timer distro's even make it impossible
21:54:33amiconnThat's an entirely different story
21:54:35jochenwell, not impossible, but harder ;)
21:54:38iobound_but all that is just gui wrapping on top, no need to criticize the NT kernel :)
21:54:41t0masand I have to admit... I'm using my windows workstations as an admin...
21:54:45amiconnIt's not the system that's a fault here
21:55:02preglowlovely latency
21:55:13t0masit is... you can't normally work in XP w/o admin rights... checkout XP home... it's very limited or admin... nothing in between
21:55:13preglowtakes a couple of secs before pitch change settings take effect
21:55:16amiconnIt's user habits (and factory defaults - these are debatable)
21:55:19iobound_anyway, does anyone know what might be the cause for a segfault in "_libwinmm_a_iname()" in my plugin?
21:55:29amiconnt0mas: YOu can.
21:55:41amiconnGuess what 'Run as..' is for
21:55:47t0maswell.... I did that in W2k
21:56:04amiconnW2K is *very* different from XP in that respect
21:56:17t0masjup, it's kind of workable
21:56:45amiconnYou can even create links to an app with 'run as' settings
21:57:12jochenhowever, i use win for playing games only.
21:57:17amiconnThat's practically the same as e.g. Gnome asking you for the root password when you try to start e.g. synaptics as user
21:58:19amiconnThe only problem I see is that the first user is admin by default.
21:58:28Bgeramiconn: this "run as" showed up in win2k for the firs time ..
21:58:49amiconn...and software companies relying on that and fail to write applications in a way that they run correctly with user rights
21:59:18amiconnBger: Yes, but that was 6 years ago...
21:59:18jochenwell, just imagine where applications save their settings.
21:59:33Bgerbtw, if you're in domain, u can't authenticate to other pc as different user than the one u're logged in ... (iirc)
21:59:44amiconnjochen: If they are written correctly, they save either in the user registry or user profile
21:59:56t0masyes, mmc.exe with a runas option is nice
21:59:56t0masI use em both... mixed
21:59:57t0masamiconn: but in windows you have admin rights by default...
21:59:57DBUGEnqueued KICK t0mas
21:59:57t0masand a lot of things just won't work as a normal user
21:59:58jochenyeah, IF. ;)
21:59:58t0masthat's a designflauw in the beginning...
21:59:58***Alert Mode level 1
21:59:59amiconnBger: YOu can
21:59:59***Alert Mode level 2
21:59:59t0maswell... a really lot of software is just written to work with admin priv. because everybody is used to having them... A nasty leftover from Win9x
22:00:04t0masLinux (and Mac OS afaik?) started in the right direction... so they don't have to solve this problem...
22:00:06t0masBger: hardly
22:00:14t0masyou have to disconnect fully from that pc
22:00:22linuxstbamiconn: Making the first user an administrator sounds like the main problem to me.
22:00:37amiconnlinuxstb: [21:58:17] <amiconn> The only problem I see is that the first user is admin by default.
22:00:47t0masoh wait... Bger
22:00:47t0masyou're right
22:00:47t0masin a domain you can't do it...
22:00:47***Alert Mode level 3
22:00:47t0masonly if you're denied access...
22:01:06t0masbut that's just a problem in the SMB protocol...
22:01:14amiconnYou can't connect to the same PC with more than one account
22:01:17t0masor at least in the way MS uses it
22:01:19peturmain problem is most software installs for the current user only, sometimes that needs to be admin, so you end up being admin
22:01:21Bgeryes, but it is
22:01:27iobound_are you sure you can't use "NET USE \\server\whatever /USER:someusername" ?
22:01:31amiconn...but you can connect to several PC with a different account each
22:01:38t0masiobound_: you can't.... 1 user at a time
22:01:40peturand you can't copy settings from one user to another
22:01:42Bgeramiconn: that's right
22:01:59iobound_oh, you mean if you've already connected to the same server as someone else..
22:02:01t0maspetur: there you have another miss...
22:02:06t0masthe registry
22:02:19t0masiobound_: I assume that's what he ment...
22:02:30peturthe _ugly_ you mean
22:02:32iobound_any tips on how I can do debug logging from deep within apps/codecs/dumb/src ?
22:02:41amiconnpetur: You can copy settings from one user to another
22:02:45amiconnIt's rather easy
22:02:50iobound_DEBUGF apparently doesn't work from there...