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 2007-08-23

00:00:13kyuubisealmy Sandisk is e250
00:00:21 Quit bdgraue (Remote closed the connection)
00:00:33linuxstb_Yes, it is. Assuming it's not an "R" model.
00:00:43kyuubisealit is sapported
00:00:55Bagderrasher: nice, I'll try it tomorrow
00:01:10kyuubisealwhat are the side effect's of rockbox?
00:01:28kyuubisealwhere do i go to download
00:01:32bluebrotherit works differently
00:01:38linuxstb_None - you can always start the original firmware (Rockbox can dual-boot on the Sansa)
00:01:48kyuubisealrealeases, currentbuild,extra's
00:01:51Chewie[]i just got an e250 on saturday and had rockbox on it within a few hours.. works great.
00:01:58 Quit k-man_ (No route to host)
00:02:03*bluebrother suggests FAQ and manual
00:02:15Chewie[]kyuubiseal: have you tried the default firmware yet?
00:02:35linuxstb_larsemil: You're saying that when playing the same file in Rockbox and the Apple firmware, Rockbox sounds worse?
00:02:42Chewie[]it feels flimsey compared to rockbox.. you don't really get a good volume or battery readout in the official firmware.
00:02:55larsemillinuxstb_: yes
00:02:58*amiconn wonders how 'blurry' sound is defined
00:03:07Chewie[]amiconn: sound is subjective anyhow,
00:03:25Chewie[]somebody look at flyspray and commit my lame patch. ;-P
00:03:35linuxstb_larsemil: Do you have any sound settings enabled in the Apple firmware? (and any in Rockbox)?
00:03:45kyuubisealanyone have a sandisk e250 2GB
00:03:55Chewie[]kyuubiseal: me?
00:04:01 Join x1jmp [0] (
00:04:03***Saving seen data "./dancer.seen"
00:04:06kyuubisealdid rockbox work
00:04:07larsemillinuxstb_: not that i know of in apple. and neither in rockbox. i am just new to rockbox
00:04:26Chewie[]kyuubiseal: using it right now.
00:04:33Chewie[]works wondefully.
00:04:44Chewie[]goes much louder and much softer than the default firmware too.
00:05:20rasherNote to pondlife: Check FS #7646 if you'd be so kind as to re-implement SAPI5 building with this new approach.
00:05:20Chewie[]where "softer" means "lower in volume but not muted".
00:05:22 Quit tvelocity ("Αποχώρησε")
00:05:48 Quit HellDragon (Client Quit)
00:06:08obopixelma: ping
00:06:10kyuubiseali am gonna download the firmware
00:06:12 Quit lee-qid ("aufwiederbyebientotsayonara")
00:06:25kyuubisealwhere do i download it
00:06:42Chewie[]kyuubiseal: look at - there's a whole manual for the Sansa.
00:06:43obocould you test some patches on an Archos please?
00:06:43parafinread manual at
00:06:45kyuubisealis it a open source jukebox
00:06:47 Quit miepchen^schlaf ("Verlassend")
00:06:57Chewie[]kyuubiseal: it ... absolutely is..
00:07:14 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
00:07:43pixelmaobo: sure, just have to try rebooting my comp, brb
00:07:48kyuubisealdo i install the bootloader
00:07:59kyuubisealor is it already installed
00:08:02Bagderkyuubiseal: the manual spells it out fine
00:08:44 Quit jgarvey ("Leaving")
00:09:00 Quit pixelma (" Reboot - who knows...")
00:10:16 Join ddalton [0] (
00:10:20 Quit qwm (Remote closed the connection)
00:10:22 Join qwm [0] (
00:10:42ddaltonamiconn: Should 9 in the morning be voiced as 9 o'clock?
00:10:47ddaltonin 24 hour format?
00:12:16*linuxstb_ has never heard "o'clock" used with the 24 hour clock
00:12:42Bjerrk13 o' clock.. works for me ;)
00:12:48amiconnSee this morning's log
00:13:14kyuubisealdo i download everything in the sansapatcher
00:13:18Bjerrkhm, does anybody know howto "fix" the blank screen bug in rockboy? The wiki says it's "doable", but the directions it gives are no longer valid..
00:13:35x1jmpSomeone here, who has applied FS #6697 (new dict) and gets a data aborts?
00:13:38kyuubisealdo i download all
00:13:44 Quit midkay (Read error: 110 (Connection timed out))
00:13:47parafinx1jmp, me
00:13:56kyuubisealdo i download all
00:14:13Bjerrkkyuubiseal: relax..
00:14:24Bagderkyuubiseal: that's not what the manual says, is it?
00:14:49kyuubisealnever mind
00:14:58kyuubisealdidn't read on
00:15:43bluebrother ...
00:15:58 Quit amiconn (" Fensteraufdatum...")
00:16:05*bluebrother wonders why we are writing a manual if people don't read it
00:16:39Bagderbecause it sounds cool! ;-)
00:17:10 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
00:18:15pixelmaobo: what do you want me to try?
00:19:00qwerux1mp: I do, I tested on Ipod nano and video as well as sansa - all give data aborts
00:19:11 Join kk [0] (n=kkit@
00:19:18 Join jhulst [0] (
00:20:10obopixelma: FS #6639 - there are 3 patches in there from "dreamlayers" - would it be possible to enabled logging and see if everything still behaves correctly?
00:20:18bluebrotherobo: just checked around in my Qt configuration ... in fact you don't need to manually add static to the config variable :)
00:20:43oboooo, nice :)
00:20:54pixelmaI'll take a look
00:21:31bluebrother(it's set in mkspecs/qconfig.prj)
00:21:39kyuubisealok i installed it and it worked but.......
00:21:48kyuubiseali can't listen to my music
00:22:30obothanks pixelma
00:22:45kyuubisealhelp please
00:23:00Chewie[]turn the volume up?
00:23:06bluebrotherdid you put any music on the player? How did you do?
00:23:18pixelmaobo: I'm a bit short of attention atm, do I need all 3?
00:23:19Chewie[]hmm, i presume too much.
00:23:40kyuubiseali put music on it when it was in the orginal software
00:23:43obopixelma: yup, all 3
00:23:46Chewie[]kyuubiseal: go into the official firmware (left button when you turn it on) to add music, then in rockbox, update the database.
00:24:05kyuubisealhow do i do that
00:24:18kyuubisealmy computer doesn't recognise it any more
00:24:28Chewie[]settings -> general -> database -> update now
00:24:37Chewie[]you need to use the official firmware to connect it to the computer.
00:24:53Chewie[]rockbox doesn't support much as far as the sansa and USB right now.
00:25:12 Join amiconn [0] (
00:25:14Chewie[]basically you use the official firmware to charge it and to transfer files.
00:25:45kyuubisealhow do i go back
00:25:47 Join My_Sic [0] (
00:25:52kyuubisealto offial filmware
00:26:01*petur curses H10 buttons
00:26:40advcomp2019they said the left arrow
00:26:51kyuubisealmy computer say's it does not recognise it
00:27:00*linuxstb_ wonders what bluebrother wondered 10 minutes ago..
00:27:03amiconnH10 buttons are not that bad, imho
00:27:08*jhMikeS curses H10 in general
00:27:28 Join Alonea [0] (
00:27:37kkthere's no way around the sansa making the "music" directory hidden?
00:27:50amiconnGetting used to them takes a while though
00:27:53Bagderkk: sure, just don't run the OF ;-)
00:27:57linuxstb_kk: Unhide it?
00:27:59peturamiconn: some mapping like in the recscreen is not nice, and the general layout on the device is just crap
00:28:00kyuubisealneed help
00:28:08advcomp2019while you turn it on, hold the left arrow
00:28:10kyuubisealuninstalling rockbox
00:28:29jhMikeSespecially when they stand in the way of so much improvement
00:28:33Aloneahmmm...ok, so GBSYSTEM is a hidden folder, but even with hidden folders turned on, I still can't see it. but if I type in the address to it, I can access it.
00:28:40linuxstb_kyuubiseal: That was a quick rejection...
00:28:41pixelmaobo: building (will take a while because of cygwin) - only thing I noticed is that the patch has the wrong lf
00:28:45Bagderkyuubiseal: if you don't read what we say and don't follow our advice, then what can we do for you?
00:28:46kklinuxstb_, it seems to rehide it
00:28:58kyuubisealtmy songs are still here
00:28:59linuxstb_kk: Then I refer you to Bagder's solution...
00:29:00amiconnpetur: Yeah, the layout is somewhat unusual. But the buttons provide better control than the H1x0 joystick
00:29:16kyuubisealk got it back
00:29:18Bagderkk: or really, don't put your music in the music directory
00:29:23amiconn...even with that odd scrollpad
00:29:30peturmapping menu on the power button troubles me a bit
00:29:34linuxstb_Alonea: The GBSYSTEM folder is both "hidden" and "system"
00:29:36kkyeah, that was my thought, really.
00:29:37obopixelma: yup, the whitespace is out in a few places as well - apart from that they look okay to me, but obviously I can't test them myself :(
00:29:45Aloneahave system turned on too...
00:29:49*jhMikeS won't even say what the scrollpad reminds him of ;)
00:29:55kyuubisealso the songs are in original filmware but how do i get to rockbox filmware
00:29:58AloneaI just turned it on...lemme check again.
00:30:06kyuubiseali just got the sandisk yesterday
00:30:07peturamiconn: is the scrollpad better with my change of yesterday?
00:30:08amiconnI can't comment on the recording screen. Never tried recording as I'm not much interested in it and I know it's broken on PP
00:30:19amiconnpetur: I can't feel any difference
00:30:31Aloneamain thing is, I cannot seem to start rockbox. first I was getting an error with some zeros and 98. now it seems stuck on the load screen
00:30:32amiconnMainly the pad just works as 2 buttons
00:30:42amiconnIt did before, and it does now
00:31:04peturyes, that wasn't changed, but the readings are much more stable
00:31:06kyuubisealhow do i use themes on rockbox and play games on rockbox
00:31:06Alonealike its in a loop cause I hear the hard drive spin and the side lighst blink
00:31:30amiconnpetur: Readings aren't stable now, and weren't before. But that has no influence on usability
00:31:31Bagderkyuubiseal: now stop asking questions here for 30 minutes and read all this in the manual
00:31:42amiconnIt worked before as two buttons, and it does now. Reliably.
00:33:03amiconnIf you put the fiinger in the centre, it toggles between up and down, exactly the same as before
00:33:06peturamiconn: not stable in the sense of keeping the same value, but staying in the same range. Before my change, the value could really swing a big range, even causing UP detected when touching down. At least on my unit
00:33:27amiconnNever had that on mine...
00:33:33linuxstb_Alonea: What is "in a loop" ? Is the Rockbox bootloader running?
00:33:50 Quit Bjerrk ("Dead Men Tell No Tales")
00:33:55Alonealinuxstb, remember that bootsplash.raw picture? Its stuck on that.
00:34:31linuxstb_I thought that was removed from the current GB bootloader?
00:34:38Aloneait flashes the yellow page with the version number at the bottom, then goes to that screen.
00:34:52Aloneai dunno. has it? I still have the file in there.
00:35:10linuxstb_Do you have the latest bootloader? That should be on
00:35:52Aloneayeah, I thought that was the problem. I got the new bootloader and thats when it stops at that screen. Before I got a 00000 (dont remember number of zeroes)98
00:36:00 Quit petur ("Zzzzzz")
00:36:20pixelmaobo: putting the new build on - all I have to do is enabling lastfm-log, reboot and, some music and watch the scrobbler.log? Anything particular I should keep an eye on?
00:37:08rasher interesting poll..
00:38:04Aloneayeah. its still doing it. Lemme try loading original firmware, then reputting rockbox on from the beginning.
00:38:42linuxstb_rasher: Where's the poll?
00:39:01advcomp2019on the side
00:39:05jhMikeSamiconn: which H10 model do you have?
00:39:28pixelmalinuxstb: I think he means the small box somewhere in the middle of the left side (took me a while to find it too)
00:39:37*linuxstb_ finds
00:40:03pixelmamuch better
00:40:10 Quit krazykit (Read error: 110 (Connection timed out))
00:40:11obopixelma: the .scrobbler.log may not be created straight away, it's cached, but should be flushed at shutdown or on USB connections. Could you check tracks appear in the log as expected (skipped tracks appear as 'S' etc)
00:40:16*jhMikeS just did something that massively speeds up video on e200
00:40:22Bagderhere's a nice quote from there => "i think rockbox is crap cuse you can't do what you want with it after you change a theme that's it you can't make anything else for it. it's stupid that the RB team has made it so you can't create anything for it."
00:40:32jhMikeSholy crap
00:41:00thegeekthis is torture jhMikeS
00:41:03BagderI get the feeling not everyone understands us ;-)
00:41:22linuxstb_Bagder: Reading those threads isn't recommended...
00:41:38Bagdermost people actually say rockbox is better/nicer/cooler than ipl
00:41:48linuxstb_" everythink you can do with rockbox you can do with "...
00:41:55rasher"rockbox is great. but cant figure out themes, or how to play music with it"
00:42:07linuxstb_"linux" was supposed to be at the end of my quote...
00:42:41pixelmaobo: might get interesting that I'm testing on an Ondio - because unlike the recorders it has no rtc
00:42:44FebsDamn, rasher, I was just about to post that!
00:43:47linuxstb_rasher: Well, if you're not aware that itunes is hiding your files in the iPod_Control folder, then Rockbox could be confusing...
00:43:51obopixelma: you should get a .scrobbler-timeless.log instead, and the last value per entry will always be 0
00:43:59rasher I'm not even going to quote that in here.
00:45:39pixelmaloading a new language gives me the spoken "new language" splash twice
00:45:45Aloneaoh yay! *huggles rockbox*
00:45:56AloneaI still wonder what caused it to go crazy though.
00:46:09amiconnlinuxstb: First thing that proves the statement wrong: Can you play mp3 music with ipl on G1..G3 without skipping? I'm sure you can't
00:46:17fxbthey always think "sleeping forever" is an error message...
00:47:14linuxstb_amiconn: You don't need to convince me... ;)
00:47:18bluebrotherwell, which human wants to sleep forever?
00:47:31bluebrotherbut humans are missing the reset button combination ;-)
00:47:34 Quit witz ()
00:48:01linuxstb_bluebrother: Isn't that the job of coffee?
00:48:13bluebrotherwell, to some degree ;-)
00:48:29bluebrotheror was that "resume from suspend"? ;-)
00:49:03rasheramiconn: If you don't go above 128kbit/s
00:49:28amiconnrasher: But is that music? ;)
00:49:44*amiconn can play 320kbps mp3 on his 2nd gen without skipping
00:50:12amiconn...even with sw tone controls enabled (it's struggling but it doesn't skip)
00:50:46Aloneaalright, back online, so whats special about the new bootloader?
00:51:33linuxstb_amiconn: Is that with or without a WPS with peakmeters?
00:51:48 Quit chrisjs169 (Connection timed out)
00:51:53 Quit jhulst (Remote closed the connection)
00:51:54amiconnBoth works (but with peakmeters buffering takes nearly forever)
00:52:09linuxstb_Yes, I've experienced never-ending buffering...
00:52:50*jhMikeS _really_ wants to remedy that ... blah
00:53:40 Quit kyuubiseal ("ChatZilla [Firefox]")
00:54:30*linuxstb_ senses more kernel complexity "discussions" coming up...
00:54:41 Join barrywardell [0] (
00:55:25 Join JdGordon [0] (
00:55:52jhMikeScomplexity beh. it's hardly that nor will be.
00:56:01 Quit ctaylorr (Read error: 110 (Connection timed out))
00:56:12amiconnjhMikeS: fyi: Recorder has 68 bytes of IRAM left in SVN
00:56:24jhMikeSthough my cache tester is running without incident now
00:57:15 Join Llorean [0] (
00:57:45amiconnLooking at the .map, I see 3 things which could be removed from iram. Probably tiny things though
00:57:53jhMikeSI do have an eye to straighten it out alot if the reason for the problems is fixed anyway.
00:58:19amiconn12 bytes...
00:58:41jhMikeShow important is IRAM to SH?
00:58:42pixelmaobo: I got no scrobbler-log (listened to an album: first track fully, then skipped a few, listened to the 8th track completely, stopped after beginning of the 9th track)
00:59:02Aloneaoh! would any lovely person point me in the direction of the file where I can remove the left right function that scrolls to the end/beginning of a text files in the textviewer?
00:59:41amiconnjhMikeS: It's used just for speed, so some often used low level stuff resides there
00:59:45 Quit midgey ()
00:59:53linuxstb_Alonea: I guess apps/plugins/viewer.c
00:59:56 Join ctaylorr [0] (
01:00:04 Quit ender` (" Some people have one of those days. I've had one of those lives.")
01:00:13amiconnThe ab repeat markers and charger_input_state surely don't belong there
01:00:16pixelmaobo: hmmm, wait a second, maybe I made a mistake
01:00:50amiconnthread.o is the biggest (and only) .ibss user
01:00:52jhMikeShmmm...if I invalidate the receiver of data first, then write the data, no cache problems happen
01:01:00 Join cujo8400 [0] (i=4826e575@gateway/web/cgi-irc/
01:01:06cujo8400hey guys
01:01:10Aloneathanks linuxstb, I will take a gander there
01:01:35 Join witz [0] (n=witz@unaffiliated/witz)
01:01:40cujo8400is there anybody free to help me out for a few minutes?
01:01:43 Join midgey [0] (
01:01:48 Quit midgey (Client Quit)
01:01:56witzWith what?
01:01:57rashercujo8400: don't ask to ask, just ask
01:01:57pixelmacujo8400: just ask
01:02:02AloneaI keep accidentally scrolling to the end.
01:02:24barrywardelllinuxstb: we'll need to release a new sansapatcher and bootloader since my sdhc commit. the old bootloader froze on ata_init() with my card inserted
01:02:30cujo8400i've tried a few times to initialize my database on my 4G colour
01:03:03amiconnbarrywardell: Does the sansa use sd or spi protocol?
01:03:05cujo8400it sits there for over 2 hours and the number gets over 40000000
01:03:16barrywardellamiconn: sd
01:03:22cujo8400the ipod gets really hot and then it shuts down and makes me start all over
01:03:35cujo8400any ideas?
01:03:35jhMikeSamiconn: it's got those thread entries all in iram but I don't really think they need to be except for dual core. it basically just holds pointers and they aren't references all that much.
01:04:00pixelmaobo: sorry, seems I tested with the wrong version, started again...
01:04:19amiconnWell, iram is for performance. We should probably need to find out what's accessed most
01:05:51obothanks pixelma - you can cheat by seeking to over half way through a track if you want ;)
01:06:00linuxstb_barrywardell: OK. That shouldn't be too much work, I can't recall much changing with sansapatcher.
01:06:11amiconnPutting code in iram makes quite a difference on SH, both because it's 32 bit (hence every other instruction fetch doesn't interfere with memory access), and because it enables efficient use of fast page mode (when running from dram, reading instructions and data alternatively cause a page miss everytime)
01:06:22jhMikeSthe running and sleeping lists by far but that implies the prev and next pointers too
01:06:25linuxstb_barrywardell: Although it could be the time to add 0xc to the supported partition types for the FAT partition....
01:06:53amiconnlinuxstb: ?
01:07:13amiconnRockbox ignores the partition type (well, it checks for != 0x05)
01:07:21barrywardelllinuxstb_: yes, should be just a matter of building the binaries
01:07:32rashercujo8400: Sounds strange, I don't really have any ideas, though I seem to recall other people having similar issues.
01:07:34barrywardellthe svn bootloader seems to be still working fine
01:07:53amiconnjhMikeS: A full memory access (opening a new page) costs 2 extra wait states compared to an access to an open page
01:08:15linuxstb_amiconn: sansapatcher checks the partition type.... The SVN version only checks for 0xb, but I think some people have reported having a FAT partition of type 0xc.
01:08:33Aloneai also would like to ask if there is any way to have multiple "bookmarks" for text files?
01:08:51barrywardelllinuxstb_: has anybody tested sansapatcher with a 0xc partition?
01:08:53cujo8400rasher: yeah, I haven't found much help for it around the internet ot in forums...I figured I should try asking around here because I really like rockbox
01:09:17linuxstb_barrywardell: No... But it was a change I made recently to ipodpatcher. Not sure it's in the released version of that though.
01:09:32linuxstb_It should be harmless.
01:10:13 Join Mikachu [0] (
01:10:32barrywardellin that case...go for it. can you build the binaries?
01:11:03 Join rotator [0] (n=e@rockbox/developer/rotator)
01:11:06barrywardellI don't have a windows or linux65 build environment
01:11:15barrywardelllinux amd64
01:11:35polygonalJdGordon: are you here? I have a question about the menu api
01:11:45jhMikeSjust serializing the scheduler with interrupts would simplify it. it wouldn't take dual core considerations out which are handled similarly but would likely cut fat on single core.
01:12:22Alonealike, each text file would have its own bookmark, or possibility to save to a bookmark?
01:12:32rashercujo8400: if you can manage to reach Slasheri, that'd be your best bet
01:12:33pixelmaobo: now... the log got created and looks correctly :)
01:12:41rashercujo8400: he's not around now though, I think
01:15:43 Nick kk is now known as krazykit (n=kkit@
01:16:52cujo8400rasher: thanks alot for your time though man
01:17:00linuxstb_barrywardell: Sure. But I'll wait until the morning - I've drunk a little too much tonight to be building releases...
01:17:15barrywardellno hurry on it. thanks
01:17:21linuxstb_(there are no Swedes around to put them on the download server anyway...)
01:17:57cujo8400I'll be back later
01:18:01 Quit cujo8400 ("CGI:IRC")
01:18:53Mikachulinuxstb_: do you know if the elephants dream isn't linked to my server anymore?
01:19:35linuxstb_Mikachu: Welcome back to Rockbox... :) I know we now have 2 fast download mirrors, so maybe it isn't.
01:19:55Mikachuthanks :)
01:19:58Mikachujust popping in
01:20:22linuxstb_I assume you're asking because your logs are showing no downloads?
01:20:34 Quit Soap (Read error: 110 (Connection timed out))
01:20:35Mikachunone linked from at any rate is now a round-robin DNS with 3 different hosts.
01:22:09 Quit tonyyarusso (Read error: 113 (No route to host))
01:22:29Mikachuthat does sound more reliable than my desktop computer :)
01:24:05 Part k-man__ ("Leaving")
01:25:14obopixelma: great - thank you again!
01:25:39 Join kyuubiseal [0] (
01:25:49kyuubiseallearned how to use
01:25:59kyuubisealand the manual was helpful
01:27:11pixelmaobo: you're welcome (gave me something to do instead of fighting with my box :/ )
01:27:12kyuubisealbut now i have a problem that it say's on my abort at 00005c00 and then it just froze and now it is frozen what should i do?
01:28:09kyuubisealcan anyone help me
01:28:35*scorche wonders what Llorean just deleted
01:28:37rasherkyuubiseal: Reset the player
01:28:56kyuubisealcan't player's frozen
01:29:08jhMikeSsomeone posted portal player docs? :P
01:29:23LloreanI deleted something?
01:29:35pixelmascorche: the Player still has rombox IIRC...
01:29:36*scorche shrugs
01:29:39rasherkyuubiseal: well there should be a key-combination that reboots it. Depends on the player. Consult the manual.
01:29:48jhMikeShis memory?
01:29:52 Quit HellDragon (Client Quit)
01:29:52scorchepixelma: did he say player?...bah
01:29:57rasherkyuubiseal: .. or state which player you have and wait for someone who knows.
01:30:01kyuubisealit just say's Data abort 000%C00
01:30:04 Join HellDragon [0] (
01:30:13pixelmascorche: "I am using my Archos Player" ;)
01:30:21Lloreanscorche: If you're referring to the forums, the last post I've deleted was 8:23 PM, US Central yesterday.
01:30:31kyuubiseali have sandisk sansa e200
01:30:43jhMikeShold power for 15sec or so
01:30:56amiconnYes, Player and Ondio SP still have rombox
01:31:14kyuubisealyep it worked
01:31:17kyuubisealnow what
01:31:29kyuubisealis it back to normal
01:32:01*scorche is getting good at glossing over the details
01:32:12kyuubisealwhat was the problem
01:32:54pixelmascorche: reading the forums too often? :P
01:33:08scorchejust a lot on my mind, i suppose..
01:33:19kyuubiseali do not specialize in media players or rockbox
01:33:28kyuubiseali do graphic desgning and so
01:33:42kyuubiseali am a graphic designer....
01:34:02 Quit obo ("bye")
01:34:38 Quit bluebrother ("leaving")
01:35:01 Quit My_Sic (Read error: 104 (Connection reset by peer))
01:35:05kyuubisealthe sansa's awesome
01:35:17 Join My_Sic [0] (
01:36:27 Quit linuxstb_ ("CGI:IRC")
01:45:29kyuubisealgood bye world
01:45:31 Part kyuubiseal
01:45:32 Quit x1jmp (Read error: 104 (Connection reset by peer))
01:48:09 Join H10_007quick [0] (
01:48:33H10_007quickbarrywardell: I have a question in regards to the buttons on the H10
01:48:55H10_007quickdoes the function for the scroll pad just restart every so many seconds?
01:49:12H10_007quickdoes it loop?
01:55:23 Join hannesd_ [0] (
01:56:02 Quit barrywardell ()
01:56:51 Quit GodEater (Read error: 110 (Connection timed out))
01:57:51 Join Soap [0] (n=Soap@rockbox/staff/soap)
01:58:17 Join chrisjs169 [0] (
01:58:22 Quit Alonea ("ChatZilla [Firefox]")
01:59:44jhMikeSamiconn: do you think trying starting the H10 with caching disabled and the patch applied would be definitive?
02:00:12 Join Merthsoft [0] (
02:00:13 Part KWhat
02:01:32MerthsoftI have a plugin question. I am doing key input using rb->button_status(), and then to move a simple object around the screen, I do: y += (button == BUTTON_SCROLL_DOWN) - (button == BUTTON_SCROLL_UP); x += (button == BUTTON_RIGHT) - (button == BUTTON_LEFT); THe thing is, this can only handle on at a time, and I want diagonal movement, how would I do that?
02:02:02Mikachuchange == to &
02:02:15Merthsoftso (button & BUTTON_LEFT)
02:02:27Merthsoftk, thanks
02:04:07***Saving seen data "./dancer.seen"
02:05:59 Join aliask_ [0] (i=82c20d6a@gateway/web/cgi-irc/
02:06:31MerthsoftMikachu: Any idea why that makes the x go the same speed it did before, but the makes the y go super fast?
02:06:51Mikachui added rockbox to cia some time ago (, they have an irc bot announce thinger too, what are your thoughts on that? (Bagder, linuxstb et al)
02:07:10Mikachu(i have a bot in another channel announce commits so i don't have to look at the web page :)
02:07:24MikachuMerthsoft: not spontaneously no
02:07:48Merthsoftwhen I just tap the button it goes way out of bounds
02:08:01Mikachuoh right
02:08:13Mikachuyou can add !! in front of every (
02:08:20Mikachuor write nicer code, but i'm lazy :)
02:08:35Mikachuthe == operator returns a 1 for true, the & returns the actual value
02:08:40Merthsoft!!... Is that just to slow it down?
02:08:41MerthsoftAh, ok
02:08:48Mikachu!! will make any value not zero into a 1
02:08:58MerthsoftThat makes sense
02:09:07 Join jhulst [0] (
02:09:20Merthsoftwhat would be a better way to do it?
02:09:36Mikachusome sort of switch or if construct i guess
02:09:59Merthsofti like this way
02:10:09Mikachuon second thought switch doesn't do & so it would have to be if
02:11:18MerthsoftNow, how do I make it go into a different folder when installed?
02:11:32Mikachuthat's not my department
02:11:43 Quit hannesd (Connection timed out)
02:11:43Merthsoftah, ok
02:11:44 Nick hannesd_ is now known as hannesd (
02:16:40 Quit chrisjs169 (Read error: 110 (Connection timed out))
02:24:54 Join barrywardell [0] (
02:28:34 Nick midkay_ is now known as midkay (
02:30:21 Quit polygonal ("CGI:IRC")
02:35:13 Quit barrywardell ()
02:41:51 Nick fxb is now known as fxb__ (
02:45:48 Join Bagder_ [0] (
02:51:01 Join scorche` [0] (n=scorche@rockbox/administrator/scorche)
02:55:49 Quit aliask_ ("CGI:IRC")
03:01:21 Quit scorche (Read error: 110 (Connection timed out))
03:01:21 Nick scorche` is now known as scorche (n=scorche@rockbox/administrator/scorche)
03:01:55 Quit Bagder (Read error: 110 (Connection timed out))
03:03:49 Quit jhulst (Remote closed the connection)
03:04:35 Quit H10_007quick ("ChatZilla [Firefox]")
03:06:53 Part Merthsoft ("Leaving")
03:13:45 Part pixelma
03:14:50 Join H10_007quick [0] (
03:16:55 Join datasleep [0] (
03:16:57H10_007quickcould someone check to make sure my code makes sense?
03:17:27 Quit datachild (Nick collision from services.)
03:17:51 Join saratoga [0] (i=9803c50e@gateway/web/cgi-irc/
03:19:54saratogalinuxstb: you still awake?
03:21:03H10_007quickSaratoga: could you check to make sure my code makes sense?
03:21:56LloreanH10_007quick: It's considered rather impolite to highlight people somewhat randomly
03:22:12H10_007quicki know that he knows C
03:22:20H10_007quickthats why I asked him
03:22:26H10_007quickWas not random
03:22:31LloreanLike, 90% of people who talk in this room know C
03:22:45LloreanIt's still impolite.
03:22:47H10_007quickbut none were looking at the screen at that moment
03:23:17 Quit HellDragon (Remote closed the connection)
03:23:40 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
03:23:40saratogaH10_007quick: sorry, i don't know anything about the button code
03:23:53saratogai really only know a bit about the audio codecs
03:24:17H10_007quickI just meant that the syntax was correct
03:24:26saratogaask gcc
03:25:04 Quit rotator ()
03:25:47saratogai mean compile it and see if gcc spits out an error
03:25:52saratogabest way to check syntax
03:26:27H10_007quickoh ok
03:26:31H10_007quickwill do
03:26:35H10_007quickthanks for the time
03:26:48saratogalinuxstb: i've noticed an interesting problem with seeking
03:26:57H10_007quickLlorean: was that poliet enough?
03:27:16saratogait seems that occasionally seekign will fail in such a way that the song gets corrupted and noisey music plays, but also such that it stays corrupted if i reload the track
03:27:30LloreanH10_007quick: Yes, but as a future notice, don't highlight people unless you know you should, such a someone you've been discussing something with before, or if you're asking about code they wrote.
03:27:43saratogai'm curious if you think the parser could be corrupting the data in the audio buffer?
03:28:04saratogai'll check the logs in the morning
03:28:16 Quit H10_007quick ("ChatZilla [Firefox]")
03:36:55 Quit gtkspert (Remote closed the connection)
03:37:08 Join gtkspert [0] (
03:48:54 Join Febs_ [0] (
03:52:16 Join H10_007quick [0] (
03:52:32 Quit H10_007quick (Client Quit)
03:56:23 Join barrywardell [0] (
03:56:38 Quit barrywardell (Client Quit)
04:00:20 Join Gursikh [0] (i=khalsa@tremulous/officialdevannoyer/khalsa)
04:00:55GursikhAnyone know the script the website uses to display the irc logs all fancy-like?
04:04:10***Saving seen data "./dancer.seen"
04:05:13LloreanI believe it's custom
04:06:26Gursikhi've been seeing it for ages
04:06:32GursikhIt's Nice ;-)
04:06:49 Quit Febs (Read error: 110 (Connection timed out))
04:06:51 Join Sambolina3 [0] (i=47fe6474@gateway/web/cgi-irc/
04:12:19 Quit Sambolina3 ("CGI:IRC (Ping timeout)")
04:27:39Chewie[]Gursikh: if you find out, tell me too, i was wondering the same thing a few hours ago.
04:27:48 Join BonBonTheJon [0] (n=jon@
04:29:03 Join BHSPitMonkey [0] (n=stephen@
04:34:17 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
04:39:49LloreanChewie[], Gursikh: Ask Zagor
04:44:44Chewie[]*makes mental note*
04:46:57 Quit Rick (Read error: 104 (Connection reset by peer))
04:47:46 Join Gibbed [0] (
04:49:08 Join jhulst [0] (
04:50:09 Join chrisjs169 [0] (
04:52:34 Quit BHSPitLappy (Connection timed out)
04:57:35chrisjs169somethings wrong with the maze plugin
04:58:50 Quit saratoga ("CGI:IRC (EOF)")
04:58:58krazykitchrisjs169, care to elaborate?
04:59:38chrisjs169krazykit: lol - in my build, it causes the _codec_ stack to overflow (??) and on the official build it causes a data abort
05:00:51krazykitwhat revision? you're playing music at the same time?
05:01:15chrisjs169or so i thought - i was selecting the official build from my build and it data aborted, but booting directly to the official build didn't cause any problems
05:01:17*chrisjs169 confused
05:02:25krazykitone question at a time then. what revision is your build? check system -> rockbox info
05:02:45chrisjs169when i originally tested it in my build, from a fresh boot (codec stack was 1%) it jumped to 100%
05:03:16 Join DC1 [0] (
05:03:44 Join midgey [0] (
05:03:45chrisjs169i need to answer my own questions first :P - in my build - I modified it to the revision number doesn't show (instead, it shows chrisjs169-[date]
05:04:26krazykityou could check in the source directory
05:04:47krazykitin .svn/entries, there'll be a number with the r#
05:05:10krazykitcuz i don't see such a problem with r14324
05:09:40*chrisjs169 looks at the time and decides it's probably just a bug that an update will fix
05:14:57 Join TerrorByte [0] (i=4a601198@gateway/web/cgi-irc/
05:15:13TerrorByteHi guys,.
05:16:34chrisjs169nope - my build still freezes on the maze plugin - oh well, i'll fix it tomorrow
05:16:40chrisjs169Hi TerrorByte
05:17:41TerrorByteJust one quick question...
05:17:56TerrorByteWhere can I find the bootloader for my MP3 Player?
05:18:25TerrorByteI remember getting the latest version for my iRiverH10 20 GB some time back and it said I needed a new bootloader or something to that effect.
05:18:48TerrorByteOh, I just found it in the Manual.
05:19:33*chrisjs169 's tired
05:19:40*chrisjs169 hears rain...?
05:20:05TerrorByteWait I just remembered something.
05:20:16TerrorByteI'm able to 'Dualboot' with the iRiver.
05:20:32TerrorByteIf I replace the bootloader, will I still be able to dualboot?
05:21:06chrisjs169it should still be able to - I don't have an iRiver, so I don't know enough about it
05:21:10 Quit ptw419 ()
05:22:10TerrorByteI see.
05:23:01TerrorByteOkay I think I screwed up this update completely.
05:27:29TerrorBytechris, are you still here?
05:33:15 Join rudman [0] (i=182fb504@gateway/web/cgi-irc/
05:33:25 Quit rudman (Client Quit)
05:33:54TerrorByteAh I got it!
05:34:21 Join rudman [0] (i=182fb504@gateway/web/cgi-irc/
05:36:42rudmanI've got a Sansa E200 with rockbox r14391-070818 and have seen some issues with themes that contain a backdrop
05:37:22rudmanafter you load a theme with a backdrop - and then load a theme without a backdrop - the backdrop stays
05:38:09rudmanit appears that all theme cfg files have to contain a backdrop: line even if it is empty
05:39:38 Quit TerrorByte ("CGI:IRC (EOF)")
05:40:30 Join TerrorByte [0] (i=4a601198@gateway/web/cgi-irc/
05:40:35TerrorByteYou still here?
05:40:43TerrorByteStill here?
05:41:05TerrorByteAbout that backdrop thing.
05:41:22TerrorByteYeah you're right that it doesn't change the backdrop. It doesn't on my iRiver H10.
05:41:30TerrorByteBut it's not that much of a concern.
05:41:34krazykitit doesn't on any targets
05:41:53krazykiti'd venture to say it's supposed to be like that. you can set a backdrop independant of a theme
05:42:02TerrorByteThat's what I do now.
05:42:17TerrorByteI converted all my wallpapers to my iRiver's resolution.
05:42:24krazykitwell, WPS, rather.
05:42:27TerrorByteAnd then I made config files for each of them.
05:42:29rudmanwhat happens is that you load a theme with a black backdrop and the next theme has a black font and you're fucked
05:42:37TerrorByteAnd I put them all in a folder.
05:42:46TerrorByteAnd now I just select the 'wallpaper' and the backdrop changes.
05:42:51TerrorByteI have over a hundred.
05:42:53 Quit parafin (Read error: 60 (Operation timed out))
05:43:11krazykitif you load a WPS without a backdrop, you don't want it to clobber the backdrop you've previously set
05:43:22rudmanthe default for all themes should have the null backdrop: line so if you change your theme - you can always then select your own backdrop
05:43:25TerrorByterudman: If you highlight an option it inverts color... If you have it set to that.. Then you're not fucked.
05:43:40TerrorByteI always select my own backdrop from the hundred or so that I have.
05:43:47TerrorByteJust make config files for them and place them in a folder.
05:43:50TerrorByteAnd select them from there.
05:43:57TerrorByteAnyways, I have to go.
05:43:58TerrorByteBye guys.
05:44:02 Quit TerrorByte (Client Quit)
05:44:25rudmanhmmm - inverse selection....let me find that
05:47:38rudmanI can only find bar selector vs pointer
05:48:58krazykitthe bar selector should invert it
05:51:55rudmanthe bar selector appears to only be in black
05:57:31 Quit Obnautikus (Remote closed the connection)
06:00:40Chewie[] - i'm getting that error, anyone have any ideas for fixing?
06:02:14 Quit midgey ()
06:04:12***Saving seen data "./dancer.seen"
06:08:21 Quit DC1 ("If Obi-wan ain't home then I don't know what the fsck we're gonna do. I ain't got no other connections on Tattooine.")
06:08:44 Join cendres [0] (n=ashes@2001:5c0:8fff:ffff:0:0:0:1)
06:09:09cendreshow is rockbox verse "wow", on the iriver?
06:10:25rudmanchewie: from the archives it looks like a user error
06:11:32 Join H10_007quick [0] (
06:11:48 Quit H10_007quick (Client Quit)
06:13:38Chewie[]rudman: well see, i don't know what i did wrong though :-P
06:15:59rudmanchewie: looks like your best bet would be to contact <>
06:17:05rudmanare you getting that exact error?
06:18:01LloreanIf it's "invalid option long calls" this almost always means you have the wrong version of the cross compiler
06:18:07LloreanWhat target are you compiling for?
06:19:19Chewie[]and yeah, i guess this GCC is *just a tad* out of date..
06:19:22LloreanIf you type "arm-elf-gcc −−version" what does it tell you?
06:19:30LloreanIt's not your normal gcc, it's your arm-elf-gcc that's important.
06:19:53LloreanSee, the development instructions say you must install 4.0.3
06:20:07LloreanSo there's your user error. :)
06:20:20Chewie[]now.. to find out which port freebsd uses for it.
06:22:30rudmanllorean: did you see my comments earlier about backdrop issues?
06:23:24Lloreanrudman: Simple solution: If your theme is not supposed to have a backdrop, put a backdrop: NULL in it. If you don't put a backdrop line in your theme, then it's assumed that the theme is not supposed to change the backdrop. Because NO .cfg changes any setting unless it includes a line for it.
06:24:46rudmanllorean: my thoughts exactly. I was suggesting that that be added to the guidelines
06:25:05LloreanWhat if the theme is supposed to be independent of the background?
06:25:06 Join ptw419 [0] (
06:25:15 Quit chrisjs169 (Read error: 110 (Connection timed out))
06:26:02LloreanSome themes don't set a backdrop, and intentionally don't care so that you can mix them with one of several backdrops
06:26:27rudmanno matter what, it will get tripped up by a background. light backdrop and light font....bad. same thing with dark
06:26:54LloreanNo matter what, it'll get tripped up if a user changes to a theme they don't know the font color of, yes.
06:27:03rudmansince you can always re-select your backdrop after you selected a theme, it would be better to null it out
06:27:04LloreanSo don't load a theme if it conflicts with your current backdrop, and it doesn't change one.
06:27:24LloreanYou can also always load the wrong backdrop, and have it conflict with the font color too.
06:27:30LloreanOr set a font color by accident that conflicts with the backdrop.
06:27:34 Join Rob2222 [0] (
06:27:39LloreanOr load a theme that doesn't set the font color, but does set the backdrop
06:27:44LloreanThere's a ridiculous number of possibilities.
06:27:47rudmanthis is the kind of thing that trips up noobs
06:28:15rudmanof course there are many ways to screw yourself up but you have to put yourself in the shoes of a noob.
06:28:31LloreanThere's two different ways to clear your settings if you screw them up.
06:28:42LloreanThere's no way to prevent a user from every possible, or even every easy, screwup.
06:29:15rudman"Hey I've got this new shiny mp3 player and I heard I can put this great software rockbox on it" "Wow! you can put different themes!" "Oh shit! After trying a couple of themes, my screen is black!" Panic! Delete Rockbox!
06:29:16LloreanWhy not let theme authors take advantage of the full flexibility of the .cfg system?
06:30:17 Join MonkeyTamer [0] (
06:31:03LloreanOut of curiosity, am I supposed to mind terribly if someone deletes Rockbox because they didn't pay attention to what theme they downloaded? The guidelines for the new theme page will require authors to include a backdrop: NULL statement if it's suppose to not have one, but there's no reason to do anything beyond asking the authors to do it.
06:32:05rudmando the guidelines already suggest that?
06:32:38LloreanI said for the new theme site.
06:32:43LloreanThere's no way to enforce it on the wiki.
06:32:53*scorche perks up
06:34:10rudmanAs long as there is a requirement for authors to do so that's fine. I didn't see anything in the existing pages and was simply suggesting that be added
06:34:30scorchewell, the existing pages wont be used...
06:35:05rudmanI'm new to rockbox and got burned by that on a number of the themes I've installed
06:35:29scorchei would hardly call it being burned...
06:35:44MonkeyTamerhow about "scorche"d?
06:36:52Lloreanrudman: Seriously, adding it to the guidelines without a way of enforcing it would do nothing
06:37:01LloreanI guarantee you the authors know that themes don't reset the backdrop unless told to.
06:37:15rudmanwell, burned in the sense of I had an "Oh fuck!" moment and then I had to mount it in order to delete the line in the cfg file
06:38:07scorcheyou can always reset the settings
06:38:10rudmanAdding it to the guidelines would let authors of themes what to do to play nice. If you were not aware of the consequences of not adding a blank backdrop: line in your cfg, you wouldn't do it
06:38:15Lloreanrudman: As it is, it's a wiki page, you're more than capable of updating any existing posted guidelines.
06:39:09rudmantrue. I just joined the site, now I have to find out who to contact to get write access
06:39:12LloreanI'd bet a full 2/3 of them know that backdrops don't get cleared unless you add a line to clear them, but it's just not very important to them.
06:39:23scorcheit says who to contact ont he page
06:39:46rudmanyes, yes, I'm sure it does, but like I said I just joined (like 20 minutes ago!)
06:40:19rudmanit should be important to them, it would make their themes more usable
06:41:06LloreanWell, tell them that, then.
06:41:18LloreanHalf the time they don't even change the font color, assuming it will be black.
06:41:40rudmanI did see that recommendation in the wiki....
06:41:54LloreanOf course, on that basis, many themes don't care about the font color or the background color because they intentionally don't care, being only a text-based WPS
06:42:42Lloreanrudman: Just because a recommendation is there doesn't mean *anyone* will ever see it.
06:43:28Lloreanrudman: As I said, guidelines mean nothing at all without an effective means to enforce them. You seem to think it's common that people get stuck with a black-on-black situation (since you've done it multiple times) yet at the same time assume theme authors wouldn't know that it's a bad idea.
06:44:29 Quit Rob222241 (Read error: 101 (Network is unreachable))
06:44:32rudmangeez, not multiple times, I just happened tonight!
06:45:20Llorean" got burned by that on a number of the themes"
06:45:25LloreanSounds kinda like more than once to me.
06:46:00LloreanAs I've said a few times, until there's a way to enforce it, it's a case of "most people know better, including the people using the themes, so life just goes on"
06:46:01rudmanwell, after it happened to me, I tested the other themes I had loaded and they all had the same problem
06:47:56 Quit MonkeyTamer ("ChatZilla [Firefox]")
06:48:57rudmanin writing my own theme, I scoured the sites to find out the right way to do it. hopefully other theme writers would do the same
06:50:51 Join webguest06 [0] (i=55a0d80d@gateway/web/cgi-irc/
07:02:35 Quit jhulst (Remote closed the connection)
07:04:15 Quit cendres (Read error: 54 (Connection reset by peer))
07:07:11webguest06es Prag
07:07:48webguest06krazykit haloo
07:08:23krazykitno, no, i'm just not sure what language prag is supposed to be
07:08:50 Quit webguest06 ("CGI:IRC (EOF)")
07:08:57krazykitguess it doesn't matter much now
07:09:00 Join webguest06 [0] (i=55a0d80d@gateway/web/cgi-irc/
07:09:52webguest06kr\yzikt your iv Prag?
07:10:07webguest06in Prague
07:10:22krazykitclosest i come is speaking crappy german
07:11:10webguest06krazykit czech nou? :-)))
07:11:35krazykitno, not at all. german and english is all i can do
07:11:51krazykitand there's no way i can support rockbox in german, sorry
07:12:52 Quit webguest06 (Client Quit)
07:17:17 Quit rudman ("CGI:IRC (EOF)")
07:33:41 Nick Gibbed is now known as Rick (i=rick@unaffiliated/rick)
07:38:53 Join parafin [0] (
07:46:45 Join perrikwp_ [0] (n=chatzill@
07:50:02 Quit Chewie[] ("z")
07:56:02 Join Thundercloud [0] (
08:04:16***Saving seen data "./dancer.seen"
08:05:11 Join MournBlade [0] (
08:06:44 Quit perrikwp (Read error: 110 (Connection timed out))
08:07:42 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:15:42 Join GodEater [0] (n=bryan@rockbox/staff/GodEater)
08:17:29ddaltonCan someone try p7641 and tell me if it compiles or not? I seem to get an error about a sleep timer. Anyway I will try it on the latest svn.
08:17:30 Quit Thundercloud (Read error: 104 (Connection reset by peer))
08:17:36ddaltondoes the code look ok?
08:32:11LinusNddalton: it was missing a brace, i am compiling it now
08:36:28ddaltonLinusN: Well is that all? it was giving me errors for the sleep timer.
08:36:36ddaltonanyway thanks.
08:36:47ddaltonDo you want to post an update or should I do that?
08:36:53LinusNi'll do it
08:39:14GodEaterhmm - had another instance of my ipod turning itself on in the middle of the night
08:39:31GodEaterwish I could figure out what might be causing it
08:40:53LloreanMaybe simply that iPods weren't designed to be turned off?
08:41:24 Join petur [0] (n=petur@rockbox/developer/petur)
08:41:28GodEaterthis may be true
08:41:41GodEaterin which case it's another nail in the coffin for battery life
08:41:59peturif austriancoder comes along, I'm at a customer the whole day - see ya tonight
08:41:59GodEaterunless you remember to always leave your ipod with the hold switch on to let the Apple firmware take care of sleeping it
08:42:03 Quit petur (Client Quit)
08:42:32LinusNGodEater: maybe the RTC has an alarm set?
08:42:57GodEaterI've checked that in Rockbox - I've made sure it's off
08:44:00LloreanHave you ever used the RTC?
08:44:11LloreanI seem to recall there was an initial problem with disabling it
08:44:13LloreanEr RTC alarm
08:44:27GodEaterno never have
08:44:33ddaltonanyone got p6044 to work?
08:45:08LinusNddalton: never tried
08:45:33ddaltonok just because it sounds like a good plugin but I haven't really got time to fix it.
08:45:41ddaltonthere were lots of errors.
08:45:44ddaltonwhen compiling
08:45:51GodEateralso, unless I'm missing it, there doesn't appear to be an alarm function in the OF
08:45:57GodEaterso it can't be that either
08:46:19ddaltonwell he says it works on the h300
08:46:39ddaltonbut I think the player has to be on the whole time.
08:46:43GodEaterhow can it work on the h300 if it won't even compile ?
08:46:49 Part larsemil
08:48:10 Join ender` [0] (
08:51:33 Join bdgraue [0] (
08:52:38GodEaterLlorean: did you get your hands on bioshock yet btw ?
08:53:01LloreanI was playing it yesterday which is why I brought it up in -community
08:53:30GodEateras good as you hoped ?
08:53:41 Join webguest91 [0] (i=c023110f@gateway/web/cgi-irc/
08:54:24GodEatercool - I might pick it up for the 360
08:54:37scorcheLlorean: ah, but you arent in -community though, are you? =P
08:55:15LinusNddalton: your time patch works fine after correcting the missing brace
08:55:23 Join davina [0] (
08:55:51*amiconn never had the effect that his ipod turned on by itself
08:55:56 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
08:56:14amiconnOnly the 1st/2nd gen weren't designed to be turned off - literally
08:56:18scorcheamiconn: you arent the only one
08:56:38GodEateramiconn: it's extremely hit and miss
08:56:48GodEatersince the last time it happened to me it's been about three weeks
08:57:02amiconnI have my mini G2 much longer
08:57:05GodEaterand before then I don't think I noticed it, because I wasn't leaving it with the hold switch on
08:57:21GodEaterso I could tell if it was powering itself on, and then off again when it noticed it was idle
08:58:18amiconnWhen I put it away, I usually do so with hold enabled
08:58:20LinusNGodEater: any other wakeup bits set in the PMU?
08:58:28LloreanI've turned on my Nano to find it already in the retail firmware, but I can never remember if I left it shutdown, or in AppleOS from my last transfer/charge/whatever.
08:58:34GodEaterLinusN: How would I tell ? :)
08:58:59LinusNyou would have to add some code to the debug menu i guess
08:59:00amiconnAnd since my last bootloader track-down, I have the OF reinstated
08:59:39*amiconn will probably remove it again after trying to track down pp502x battery issues
09:00:03amiconnThe increased boot time is kinda annoying
09:00:14GodEaterLlorean: that's a tougher one to track down I guess with the Nano since you have to use the OF for transfers
09:00:26GodEaterI always use disk mode since I find it's performance perfectly acceptable
09:00:36GodEaterso I KNOW I don't boot into Apple OS
09:00:49LloreanGodEater: Well, I also by habit charge in the Apple OS.
09:00:53amiconnI'd rather live with slow diskmode than slow booting (but mini G2 diskmode isn't slow)
09:01:11GodEaterLlorean: I only do that rarely
09:01:14amiconnI know how slow the nano/video diskmode is though
09:01:23*Llorean couldn't stand taking 3-7 minutes to transfer builds, back when they included fonts.
09:01:32*GodEater hasn't noticed the video's disk mode is particularly slow at all
09:01:37LloreanNow, minus the fonts, and in linux, it's a lot faster.
09:01:46LloreanGodEater: Oddly enough, it doesn't happen to everyone
09:01:50midkayvideo's disk mode is a bit slow.. nano's is unbelievably slow.
09:01:58midkayor so it's always seemed to me.
09:02:10GodEaterI've never witnessed it on a nano
09:02:24GodEaterand g/f has a 2g nano - so no point with Rockbox :)
09:02:47amiconnWell, video diskmode is still faster than Ondio
09:03:09amiconnUnzipping a to Ondio takes ~10 minutes on windows around 3 minutes
09:03:59 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:04:20GodEatergood lord that's slow
09:06:08amiconnOn linux it would probably be faster, but I rather live with the slowness than with forgetting to unmount and hence trashing my filesystem everytime
09:06:39amiconn(and I don't trust linux' usb capabilities anymore since it trashed flac files on transfer without any error message)
09:10:21*amiconn wonders why linux' file caching is only all-or-nothing
09:11:55GodEateramiconn: change it ;)
09:12:32 Join B4gder [0] (
09:19:55 Join PaulJam [0] (
09:23:50 Join desowin [0] (n=desowin@
09:25:45 Quit webguest91 ("CGI:IRC (Ping timeout)")
09:28:48 Join kubiixaka [0] (
09:31:55 Join aliask [0] (
09:36:53 Join Rob222241 [0] (
09:39:29 Join ie [0] (i=c27f0814@gateway/web/cgi-irc/
09:41:02ieHello. Do I understand it correctly that sansa's original FW only sees the music stored in the dir MUSIC (and its subdirs) but RB sees everything, i.e. also the files stored directly in the root folder
09:41:36aliaskRockbox views all files/folders on the disk
09:41:46aliaskNot sure about the OF
09:42:40iealiask: the manual says music should be stored somewhere under MUSIC
09:43:25aliaskie: That might be to facilitate using both firmwares
09:43:26Lloreanie: Not the Rockbox manual.
09:44:34ieLlorean: yes, the sansa manual. So if I want to see the same music with both FWs, I should store it in MUSIC, right?
09:45:04Lloreanie: I don't know about the official firmware. This isn't really the right place to ask for how to use the original firmware.
09:48:21ieLlorean: ok. Another question: I'd expect (under RB) that the context menu is called with the long press of the center button. Like in e.h. H120. The key mapping doesn't seem to be consistent across platforms at this point.
09:48:24PaulJamhi, does someone know where the fm radio stored its settings (like for example preset mode/scan mode) ?
09:48:57iePaulJam: nvram? (not sure)
09:49:19PaulJamwhy not the settings.cfg?
09:49:57Lloreanie: I've written a patch for this, but there's some contention as to whether it should be accepted.
09:49:59JdGordonshhuold be in config.cfg with the rest pf settings...
09:50:11GodEaterLlorean: why's that ?
09:50:35ieLlorean: could you tell the FS number?
09:50:52aliaskGodEater: Rockbox consistency vs what is written on the keys
09:51:06LloreanGodEater: Yeah, basically the "Power" button has the word Menu written just above it.
09:51:31GodEaterLlorean: but we ignore that on the ipod too
09:51:51LloreanGodEater: Personally, I think my patch makes it a little more usable. Actually, on the iPod, we coincide with all the labels, I think
09:52:05GodEaternot for context menu
09:52:06 Join miepchen^schlaf [0] (
09:52:20Lloreanie: I don't know the number any more
09:52:21GodEaterit's long push on centre button
09:52:30LloreanGodEater: Yes, but I had to move both context and main menu
09:52:46GodEateryou *had* to ?
09:52:47LloreanIf I moved the context to long-center, it leaves the "Playlist" button (it's a picture, could be interpreted as a menu, or context menu) functionless.
09:52:59LloreanSo I didn't *have* to, but I felt uncomfortable leaving a button entirely unused
09:53:09LloreanEspecially one that feels *so* much to me like it'd be perfect for the normal menu function.
09:53:17GodEaterI thought we had other targets with buttons unused too ?
09:53:18ieLlorean: But the other button has a picture of a menu on it!
09:53:24LloreanGodEater: Just "Record"
09:53:35Lloreanie: Yes, but it doesn't *say* menu.
09:53:43LloreanPlus, you're arguing my point, argue with the people on the other side
09:54:11GodEaterLlorean needs no help arguing ;)
09:54:39LloreanI *like* my patch.
09:54:45ieLlorean: I'm not argueing
09:54:48LloreanIt feels more comfortable to me. :)
09:54:48 Quit Rob2222 (Network is unreachable)
09:55:22 Join Rob2222 [0] (
09:55:47ieHe-eh. We could have a setting for keys layout. With values for OF, RB, emacs, vi, ...
09:56:19*GodEater gets the flame proof suit out to watch this....
09:56:30 Quit desowin ("use linux")
09:56:46 Join desowin [0] (n=desowin@
09:57:20ieGodEater: the two last words are dangerous indeed if they stand together
09:58:06 Quit Rob222241 (Read error: 110 (Connection timed out))
09:58:26 Join Nibbler [0] (
09:59:44 Join webguest91 [0] (i=c023110f@gateway/web/cgi-irc/
10:01:47ieBTW: is the long press of PLAY used on H120 somewhere? Besides FM radio?
10:02:45 Join ddalton [0] (
10:02:49PaulJamthe play button is used as a modifier in most of the other screens, so a function for long press would be propably bad
10:04:21***Saving seen data "./dancer.seen"
10:04:32ddaltonLinusN thanks for that update. I was rushing and couldn't fix it. Sorry.
10:04:42ddaltonis that how you want the time voiced?
10:05:10iePaulJam: huh? I didn't know that. What screens do you mean?
10:06:20LinusNddalton: sounds about right, i'm not sure if i want "hundred hours" or "o'clock" though, but i'm not a native english speaker
10:06:57ddaltonLinusN aren't you? You sound like you are. Anyway I don't know how you would say 8 a m.
10:07:19ddaltonI would think it would be 8 hundred hours.
10:07:29ddaltonbecause isn't o'clock 12 hour?
10:08:48 Join JdGordon_ [0] (
10:09:01PaulJamie: for example here
10:09:43LinusNddalton: from the wikipedia article: "Both "eighteen o'clock" and "eighteen hundred" are commonly encountered spoken English for 18:00."
10:10:03ddaltonok fine forget about the patch
10:10:08LinusNddalton: i'm from sweden btw
10:10:09ddaltonit is fine how it is now.
10:10:15LinusNi think so too
10:10:46LinusNbut i still want the "oh seven" feature
10:11:16ddaltonmaybe best in a separate patch.
10:11:21ddaltonwhat do you think?
10:11:29LinusNi think the patch should go in
10:11:40ddaltonwhat one?
10:12:38ddaltonok so are you happy to commit it how it is? or do I need to change it.
10:12:47ddaltonby the way is the indenting right?
10:13:39B4gderLinusN: tried the magic stick trick on the forums?
10:13:50 Quit JdGordon (Read error: 110 (Connection timed out))
10:13:56 Nick JdGordon_ is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
10:14:42 Nick perrikwp_ is now known as perrikwp (n=chatzill@
10:17:18 Part Llorean
10:18:13iePaulJam: ah, yes, there are combos PLAY + something. But that doesn't prevent us from using the long press of PLAY imho. If you want to press PLAY+UP, you wouldn't press and hold PLAY very long before you press UP (I think)
10:18:46 Quit webguest91 ("CGI:IRC (Ping timeout)")
10:19:34markunCan you rename the NonArchos wiki? Maybe something like PotentialTargets, or maybe there are better suggestions
10:20:40 Quit joshin_ (Read error: 110 (Connection timed out))
10:22:06 Join obo [0] (n=obo@rockbox/developer/obo)
10:26:31ddaltonLinusN are you there?
10:26:37LinusNyes i am
10:27:01ddaltonok so what were you saying about p7641? it should be committed?
10:27:59LinusNi think so
10:28:17LinusNi'll do it
10:28:27LinusNsoon :-)
10:28:27ddaltonis the coding ok?
10:28:33ddaltonok then
10:28:37ddaltonis the coding ok?
10:28:57LinusNi'll simplify it a little before committing, nothing fancy
10:29:30ddaltonok what can be simplified?
10:29:53LinusNbasically remove the call to talk_number(tm->tm_min, true) in both branches of the if(tm->tm_min < 10)
10:30:24ddaltonok I see what your saying.
10:35:50*linuxstb wouldn't say "9 o'clock am" - just "9 o'clock" or "9am"
10:36:25*LinusN smells a bikeshed discussion :-)
10:36:32*B4gder renamed NonArchos
10:37:14linuxstbLinusN: It's just that I've _never_ heard the time spoken that way...
10:37:49LinusNlinuxstb: i think i agree there
10:38:41LinusNbut 9 o'clock is rather ambiguous, isn't it?
10:39:04LinusN9am sounds better
10:39:10B4gderwell, spoken time frequently is ambiguous
10:39:15LinusNof course
10:39:20B4gderpeople rarely say am or pm
10:39:25LinusNbut here we need to be exact
10:39:27linuxstbI don't feel strongly about that - I'm generally aware enough to know if it's morning or afternoon...
10:39:29LinusNi think
10:40:16LinusNddalton: you have a talking watch?
10:43:16 Join Guile`` [0] (n=Guile@
10:44:53 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:45:37 Join webguest91 [0] (i=c023110f@gateway/web/cgi-irc/
10:46:49 Join bascule [0] (i=86920006@gateway/web/cgi-irc/
10:49:10B4gderI'm no longer on 10000 commits...
10:50:14 Join bluebrother [0] (i=2zXZzWCa@rockbox/staff/bluebrother)
10:50:22ddaltonLinusN yes I do. Why?
10:50:43LinusNjust wonder how it speaks the time
10:51:00LinusNdoes it say "9 o'clock am"?
10:51:02ddalton"7 47 pm"
10:51:21Mikachuwill this talking clock support 24-hour time too?
10:51:23ddalton"9 o'clock a m"
10:51:33LinusNMikachu: yes
10:51:49LinusNddalton: ok, just checking
10:52:41ddaltonI think the oh part is needed because that is how the watch says the time but it doesn't speak in 24 hour format.
10:52:50basculeAnyone aware that the forums are currently unavailable (HTTP 500 - internal server error)?
10:53:04basculefor me in the UK at least
10:53:05ddaltonpersonally I would say "4 hundred hours"
10:53:08LinusNi wonder if o'clock is acceptable in 24-hour mode instead of "hundred hours"
10:53:18ddaltonand maybe "9 hundred hours"
10:53:25ddaltonor 9 o'clock
10:53:29LinusNbascule: the server is b0rked
10:53:30ddaltonwhat is better?
10:53:33Mikachunobody says "hundred" in 24-hour mode countries, only the military in the us do that
10:53:42ddaltonor 09 hundred isn't it?
10:53:44LinusNMikachu: i believe so too
10:54:34linuxstbMikachu: How would you say 9am in the 24-hour clock?
10:54:41ddaltonok so your saying p7641 is not right? you just want me to fix the oh part and that is it?
10:54:45LinusNso would it be ok if i simplified it a little, and have it say "o'clock" in both 12-hour and 24-hour mode?
10:54:54MikachuLinusN: "nine" :)
10:54:58 Join scorche` [0] (n=scorche@rockbox/administrator/scorche)
10:55:06ddalton22 o'clock does that sound right?
10:55:15LinusNi dunno :-)
10:55:21Mikachui mean linuxstb
10:55:30linuxstbLinusN: I would vote for removing o'clock everywhere...
10:56:20pixelmaI don't know - if I hear "nine hundred hours", I always think of 900 as the number - but then I'm not a native speaker...
10:56:46ddaltonlook it is just meant to tell us the time. As long as we can tell what the time is it is fine.
10:56:50bluebrother900 hours is quite a long time ;-)
10:56:51basculethat's because it should be oh-nine hundred hours
10:56:57LinusNddalton: i agree with you
10:57:05amiconnddalton: Well, we had a talking clock (english only) which says o'clock in 24 hour mode. And according to wikipedia it is a correct way
10:57:06ddaltonnoone here says 22 o'clock. but it doesn't mean your wrong.
10:57:30basculepersonally, I would disagree with Wikipedia on that one
10:57:38linuxstbamiconn: And wikipedia is never wrong?
10:57:41bascule22 o'clock is never correct in britain
10:57:49LinusNwow, can you say "bikeshed" :-)
10:57:50linuxstbbascule: Agreed
10:57:59ddaltonAs I said before amiconn it is fine with me. what about 9 hours. and 22 hours. and for say 3 30 pm 15 30?
10:58:13 Quit BigMac (Read error: 110 (Connection timed out))
10:58:26ddaltonno and 22 o'clock doesn't sound right either.
10:58:41B4gderso the wikipedia article is wrong! ;-)
10:58:51LinusNas always :-)
10:59:06basculeif it's a 24-hr clock, it should always be oh-nine hundered, twenty-two hundres and
10:59:21ddaltonI think 22 hundred hours would be fine. But what do you guys think?
10:59:23basculeoh-nine-thirty, twenty-two-thirty
10:59:40linuxstbbascule: If you're saying to drop "hours", then I agree with you.
10:59:41Mikachuwhy not just say "twenty-two"?
10:59:42B4gderhundred sounds like the millitary way, not a way 24 hour users actually say
10:59:48LinusNhow about "22 sharp"?
10:59:54basculeyes, drop the 'hours'
11:00:07markunand add "sir!"
11:00:20pixelmabut hours are not hundred minutes ;)
11:00:22LinusNhow about "22 sharp"?
11:00:47basculeno, oh-nine hundred; oh-nine-oh-one
11:00:48GodEateris bascule missing his fix of requests from silly users ? :)
11:00:50ddaltonwell 22 sounds fine to me.
11:00:52linuxstbpixelma: I agree, it's not logical, but I can't think of any (sensible) alternative...
11:00:59basculeno, i'm bored
11:01:07ddaltonI would be using the 12 hour format anyway. that is why I implemented it.
11:01:31linuxstbLinusN: Just make a dictatorial decision and commit something ;)
11:01:37LinusNok, so we want to pronounce the leading zero for the hours as well?
11:01:38basculeif it's 12-hour format, it kindof has to include o'clock
11:02:01ddaltonLinusN: I know you won't agree with this. But why don't you committ the patch and see if anyone complains.
11:02:09 Quit scorche (Connection timed out)
11:02:15ddaltonDo you think most rockbox users actually care?
11:02:23ddaltonI don't know but would they?
11:02:52ddaltonwell if it sounds like an army format why does that matter?
11:03:02LinusNddalton: nah, i think most people would be happy with 7641
11:03:09ddalton22 hundred hours sounds better than 22 o'clock to me.
11:03:29LinusNby the way, doesn't it say "hour" in your patch?
11:03:35LinusNand not "hours"
11:03:45ddaltonyeah think so I can check hang on.
11:05:06Mikachuyou can have 24-hour, 12-hour and us military then
11:05:20LinusNit says "hour"
11:06:03morrijrI'd just use "Military" rather than specify the country
11:06:17ddaltonok just commit it then. If you want.
11:06:37LinusNwe use 24 hour format in sweden, and most of us are civilians
11:06:39 Quit Nibbler (Read error: 113 (No route to host))
11:07:03linuxstbSame in the UK - 24-hour time isn't considered a military thing.
11:07:19ddaltonhey just a quick question if I apply lots of patches and generate one big patch file how do I split up the patches into separate files?
11:07:24ddaltonso the big patch file
11:07:31preglowsame here
11:07:40LinusNddalton: you can't
11:07:50preglowhowever, pronouncing 13:00 as thirteenhundrer hours is :)
11:07:58ddaltonLinusN can you edit the patch file by hand?
11:08:14LinusNyes you can, but it is not recommended
11:08:15Mikachuthat's why you should be using git :P
11:08:19ddaltonsee 13 hundred hours sounds fine to me.
11:08:38LinusNis it worth it to add an "hours" voice entry?
11:08:42ddaltonhmmm I don't know what I should do then.
11:08:48Mikachudoes it round to nearest 5 minutes?
11:09:17markunddalton: I usually edit my patches by hand
11:09:22ddaltonok then
11:09:26Mikachuthen the chances of anyone hearing the o'clock is quite small, just round to :59 or :01 :P
11:09:26LinusNi'll commit 7641 as-is
11:09:28ddaltonI will give that a go.
11:09:37ddaltonok then.
11:09:50markunddalton: removing files and whole blocks of changes is no problem
11:09:56Mikachuddalton: as long as two patches aren't mixed in a hunk, it's usually easy
11:12:26ddaltonit looks like this one is going to be hard. But I have another idea.
11:13:18ddaltonyesterday everyone was telling me that isn't possible to voice plugins. Anyway my question is why can't the plugins use the talk functions from rockbox.
11:13:26ddaltonit isn't I mean
11:13:39ddaltoncan plugins use functions from rockbox?
11:13:42LinusNi just committed 7641
11:14:14Mikachuwould you want a bot announcing commits in here?
11:14:26ddaltonLinusN: I am updating p6138 if I post an update do you want to commit the patch?
11:14:37ddaltonI didn't write it
11:15:20 Quit My_Sic (Read error: 104 (Connection reset by peer))
11:16:16LinusNif we commit 7641, i think we would want another way to hear just the time
11:16:25LinusNsorry, 6138 i mean
11:16:40 Join tucoz [0] (n=chatzill@rockbox/staff/tucoz)
11:17:15ddaltonWell the time is going to be voiced straight after the battery level. then the disk free info.
11:17:21ddaltonWhat do you think.
11:18:15LinusNbut can it be interrupted?
11:18:30ddaltonAnd I have got a lot of patches from sdoyon here that he emailed to me. I am just updating them now.
11:19:02ddaltoncan what be interrupted??
11:19:22ddaltonAre you interest in those patches if I tell you what they are?
11:19:28LinusNi mean, to be able to leave the info screen when the time has been spoken
11:19:45ddaltonbecause I am using them now and being a blind user it gives me access to so much more.
11:19:59amiconnLinusN: Sure you can. Voice is asynchronous
11:20:02ddaltonum let me check
11:20:10ddaltonactually you just got an answer
11:20:20LinusNamiconn: but is the info screen interruptable? i think not
11:20:30ddaltonsorry about the '
11:20:38amiconnHuh? You can leave the info screen, can't you?
11:20:50LinusNah, yes, i see now
11:21:09amiconnBut the time you hear disk & time announcement, the info screen itself is long done and idling around
11:21:12ddaltonso any interest for it?
11:21:15LinusNamiconn: yes
11:21:26LinusNddalton: so they are not in the tracker?
11:21:45ddaltonLinusN the patches he sent me aren't yet.
11:21:55ddaltonactually they might be but over a year old.
11:21:56LinusNddalton: we want them in the tracker
11:22:20ddaltonok I will update them and put them there.
11:22:55ddaltonHe wrote them so I don't really want them to be under my name because really it is his work not mine. Would that be ok?
11:23:10ddaltonI don't want credit for something I didn't do.
11:23:22ddaltonwhen he deserves it.
11:23:43LinusNddalton: by the way, i find it amazing and uplifting to see a blind man contribute like you do
11:23:57*LinusN bows
11:24:30ddaltonLinusN: do You think I contribute much? I have only done 2 patches that got committed and I am not really a programmer yet.
11:26:01 Join joshin [0] (n=joshin@VDSL-130-13-8-105.PHNX.QWEST.NET)
11:26:08LinusNi'm just fascinated how you are even able to program without seeing
11:26:38markunddalton: btw, if you want to help out with porting espeak, I have some ideas on what needs to be done
11:27:23ddaltonactually I am very impressed with how sdoyon programs. He patches are almost perfect.
11:27:39ddaltonok what's that. I will certainly try.
11:28:32 Join Nibbler [0] (
11:28:55ddaltonLinusN: I am not sure how long these patches will take me. They all work on rev 14324 so most of them should work.
11:29:13ddaltonI think my patche with the 12 hour time messed a lot of things up.
11:29:16ddaltonsorry about that
11:30:29markunddalton: first thing would be to convert from c++ to c. The translator class should be replaced by a struct, that's about it
11:30:44ddaltonhmm ok where can I get the source from?
11:31:03ddaltonI haven't learnt c++ is that a problem?
11:31:10ddaltonI could if I need to.
11:31:59ddaltonso nothing has been done yet?
11:32:06markunnot on espeak
11:32:27markunsomeone else is working on porting flite, we'll have to see how that goes
11:32:46ddaltonso there all screen readers?
11:32:58Mikachuonly text-to-speech
11:33:36ddaltonwhich one will probably work on rockbox?
11:33:46markunboth could be made to work
11:34:04ddaltonso a blind user could have access to everything?
11:35:06ddaltonok I will give it a go. there might also be some blind users I know that might want to help me out.
11:36:31 Quit HeyImMatt (Read error: 110 (Connection timed out))
11:37:35markunddalton: we'll have to see how far we get. A plugin to read text files would be a good first goal, don't you think?
11:38:21ddaltonyes or even just to read hello world. Then we know it is working and then the text editor would be great.
11:38:44ddaltonI would really like to be able to read text files on my player
11:38:59ddaltonanyway this is a chalenge for me I think.
11:40:20 Nick fxb__ is now known as fxb (
11:41:01markunddalton: for me too, but if we invest enough time I'm sure we can do it
11:41:35ddaltonLinusN: some of these patches I don't even know what they do. So what should I write for them in the tracker. But he says for some to work you need others.
11:42:16ddaltonyes I will. So will people be able to help me out with code?
11:42:28ddaltonlike with the normal patches?
11:42:49LinusNddalton: just put them in the tracker and explain exactly that
11:43:27ddaltonok then
11:44:07ddaltonalso they are his patches so is it ok for me to upload them. Should I say they are by him?
11:44:58ddaltonwhat is the command to compare two dirs. and make a diff. So to go through every sub dir?
11:45:11ddaltonI know it is on the wiki but could someone tell me?
11:46:40ddaltonOk I will look it up.
11:47:50 Quit Jon-Kha (
11:47:50 Quit Slasheri (
11:48:15NJoinJon-Kha [0] (
11:48:15NJoinSlasheri [0] (
11:51:17GodEaterSlasheri: did you get a chance to investigate those dircache issues mentioned the other day ?
11:51:41ddaltonLinusN: Could you test the patch when I upload it. It is that I am just testing so many here that is all.
11:51:46ddaltoninfo screen one.
11:52:10LinusNno time right now, maybe later
11:55:41ddaltonLinusN: How do these patches sound? Say id3 info, voice more info under the info screen, voice fm radio more, playlist catolog voicing and that is all I will probably be able to do to night.
11:55:49ddaltono and the book marks screen.
11:56:29ddaltonhas anyone here got time to test a patch?
11:56:54GodEaterddalton: maybe, what's the FS number ?
11:57:22ddaltonit will be 6138 I just need to post my update.
11:57:44GodEaterok - just let me know when you're done
11:59:19GodEaterhmm - if I'm going to be testing voice, I'd best make sure my voice file is right
11:59:22ddaltonactually it will be a different one I told you the wrong patch sorry.
11:59:24GodEatercurrently it's all over the place!
11:59:35ddaltonwell it edits english.lang.
12:00:03GodEaterso I'll need to build my own probably
12:00:13GodEaterI just want to test how it works *now*
12:00:20GodEaterand then how it works after your patch
12:00:25GodEateror I'm not going to be able to tell you much
12:01:22GodEateris it my imagination or has configure changed now ?
12:01:31GodEaterno (V)oice option anymore ?
12:02:00 Quit aliask ("ChatZilla [Firefox]")
12:03:57ddaltonGodEater: p7647 I just need to know if it compiles fine.
12:04:09ddaltonbut to build a voice select advance then voice.
12:04:22***Saving seen data "./dancer.seen"
12:05:02GodEaterah ha - thanks
12:05:14ddaltonno problem can you test the patch?
12:05:52GodEaterI'll be about 30 minutes - work just arrived =(
12:06:04ddaltonshould time or disk free info be spoken first in the info screen?
12:06:11ddaltonok then
12:06:26 Quit ie ("CGI:IRC")
12:06:27LinusNi think time
12:06:59ddaltonyep well that's what I think to. I will just update this patch and submit it.
12:07:23 Join My_Sic [0] (
12:09:02 Join x1jmp [0] (
12:09:29 Join lee-qid [0] (
12:10:22GodEaterhmm - selecting "Voice" on my machine seems to hang...
12:10:26 Join kk [0] (n=kkit@
12:12:13ddaltonpress c then enter.
12:12:20ddaltonthat's what I found
12:12:52ddaltonactually I don't think p7647 modifies english.lang can't remember.
12:13:02ddaltondid that work?
12:13:51GodEaterpressing c then enter does indeed work
12:13:55GodEaterbut what on earth does it do ?
12:14:13B4gderLinusN: have you (tried to) contacted Jeff about the forum situation?
12:16:33 Join freqmod [0] (
12:17:53ddaltonwhat pressing c?
12:19:25 Quit My_Sic (Read error: 104 (Connection reset by peer))
12:20:14 Quit freqmod (Remote closed the connection)
12:20:32 Join Llorean [0] (
12:21:28 Quit krazykit (Read error: 110 (Connection timed out))
12:25:34 Nick scorche` is now known as scorche (n=scorche@rockbox/administrator/scorche)
12:34:18ddaltoncan someone please test p6138? because cygwin takes for ever here.
12:34:39ddaltonand it requires p7647 as well.
12:34:47ddaltonI said that in my comment
12:34:47LinusNB4gder: i sent him an email
12:35:10 Join My_Sic [0] (
12:45:33 Quit Llorean (Remote closed the connection)
12:46:00 Join Llorean [0] (
12:48:00ddaltondoes anyone know what a patch called "nailbutton" could be?
12:49:32 Quit tucoz ("ChatZilla [Firefox]")
12:49:43ddaltonLinusN: What do you think about the info screen being like the running time screen?
12:49:50GodEaterddalton: just trying yuor 7647 now
12:50:17ddaltoncan you try p6138 as well because p6138 uses p7647
12:51:02ddaltonand I don't think p7647 does much by itself.
12:52:11GodEaterwill do that as soon as I've done 7647 on it's own
12:52:30ddaltonbut for p6138 it needs p7647
12:53:04GodEaterno problem
12:53:48 Join advcomp2019- [0] (n=advcomp2@
12:53:57GodEateryou just want to know it compiles ok at the moment yes ?
12:54:26ddaltonyep pretty much I don't think it needs to be tested. I will probably do them all tomorrow
12:55:24ddaltondid it compile?
12:59:14 Quit My_Sic (Read error: 104 (Connection reset by peer))
13:00:08GodEater7647 comiles fine
13:00:16GodEater(sorry, just fighting with git at the moment)
13:01:21GodEater6138 fails to apply
13:02:34preglowforums dead?
13:02:43LinusNthe server is down
13:03:26markunwork is a lot more boring without the forums
13:03:54ddaltonGodEater: have you got the patch I submited about 10 minutes a go?
13:03:58LinusNthe forums are a problem for us, since they are hosted on a server we don't control
13:04:43 Join midgey [0] (
13:05:03GodEaterddalton: oops - my mistake - I'll try again ;)
13:05:40ddaltonok thanks
13:06:42ddaltonhow do I check what revision I have?
13:06:45ddaltonof svn
13:06:48ddaltonthe source
13:07:10GodEaterthat builds fine too
13:07:15GodEaternot sure
13:07:21GodEaterI don't use svn directly anymore
13:07:21obosvn info?
13:07:23Mikachusvn info|grep Revision
13:07:27ddaltonok and is that on latest svn?
13:07:40GodEaterddalton: yes, that was on the latest svn rev
13:08:42ddaltonso r14437? Because I thought the time patch that just got committed might mess it up.
13:08:53GodEater14437 correct
13:08:59GodEaterthe time patch seems to work with it fine
13:09:06LinusNddalton: is there a reason that stephane sent you the patches instead of updating them in the tracker?
13:09:25ddaltonok good. Do you have any more time I just have one more patch.
13:09:55ddaltonok if you can try p6239 that would be great. I think a few of these patches use it.
13:10:08GodEaterso you need me to apply the other two again first ?
13:10:11ddaltonIf it compiles then it is fine.
13:10:39GodEaterso apply it on it's own first ?
13:11:01ddaltonyep because I don't have any patches submited that use that one yet.
13:11:02 Quit advcomp2019 (Read error: 110 (Connection timed out))
13:11:11ddalton A lot of patches here to update.
13:12:20GodEaterthe talk.h part is slightly offset, but still applied ok
13:12:55preglowLinusN: why not host them locally?
13:13:45LinusNpreglow: that's an alternative, we just accepted the offer from jeff that he could host and administrate the forums for us
13:14:20preglowi guess it is nice not to have to bother about it...
13:14:28GodEaterexcept when they go wrong...
13:14:40GodEaterddalton: 6239 compiles ok too
13:15:00ddaltonso what needs to be fixed?
13:15:43GodEaterthe talk.h bit is offset by one line
13:15:49 Quit webguest91 ("CGI:IRC (Ping timeout)")
13:15:51GodEateryou'd need to look and see why - I can't really tell you
13:15:55 Quit JdGordon ("Konversation terminated!")
13:15:56ddaltonso is that a problem?
13:16:06GodEaterno - as I said, it still applied cleanly
13:16:31ddaltonso if someone wanted it in rockbox it would be excepted?
13:16:46GodEaternot my place to say
13:16:58ddaltonanyone know?
13:17:57 Part Mikachu
13:18:17 Quit kclaf (Read error: 110 (Connection timed out))
13:18:51LloreanDoesn't 6239 admit that it probably isn't a proper solution to the problem?
13:19:08ddaltonwell I can't even work out what it is for.
13:21:02LloreanIt looks like it's an attempt to fix a problem that doesn't happen yet, but might happen if other new code were added.
13:21:45ddaltonwell I think a couple of patches might use it. I am updating them all and then I will test what patches need other patches and then I will submit them.
13:21:55ddaltonBut the are really nice patches.
13:23:53GodEateranyone else seen this : ?
13:24:01ddaltonGodEater: does p6138 remove the new time voicing which was added? Because it isn't meant to.
13:24:27LinusNddalton: is there a reason that stephane sent you the patches instead of updating them in the tracker?
13:24:29GodEaterddalton: how would it do that, other than explicitly
13:25:37ddaltonLinusN: he sent them to me because I asked him for them. I think they needed updating and he hadn't got around to submitting.
13:25:45LloreanGodEater: I'm not sure why they'd do that, it seems like it'd introduce a lot of unnecessary overhead.
13:26:07GodEaterLlorean: one of the guys in the ipodlinux channel thinks they've already done it with the Nano 2G
13:26:22GodEaterthough he presented no evidence to this other than "nudge nudge, I know people, wink"
13:26:26LloreanI call shenanigans on that.
13:26:39ddaltonActually i don't think it has.
13:26:42GodEaterme too - but I didn't want to get into a pissing match over it
13:26:46LloreanYeah, no point.
13:27:02GodEaterbut if newer ipods are to mimic the iphone's UI then I guess it would make some sense
13:27:27ddaltonLinusN: is that ok? I thought I should update them.
13:27:46LinusNddalton: sure, i was just curious why stephane didn't do it himself
13:27:58LloreanYeah, but mimicing the iPhone UI seems a bit overkill, most of its UI relates to features the iPod won't have or use (unless they go for a touchscreen iPod, in which case it's officially a PDA and not a DAP)
13:28:30ddaltonLinusN: I have a playlist catalog patch and id3 viewer patch here. Are you interested in them.
13:28:30LloreanIf I were them, I'd keep them distinct products so that people buy both, instead of introducing too much overlap/similarity.
13:28:37scorcheone of the apple leaks site displayed the look of the new told them to take them down immediately, so it is likely that is a good chance...
13:28:41 Quit gunk ("Leaving")
13:28:46GodEaterLlorean: I remember seeing rumours of a wholey touch screen ipod a while back now
13:28:46ddaltonI haven't updated it yet but I will probably tomorrow.
13:29:11GodEaterlike there scorche ?
13:29:13LloreanGodEater: Yeah, but that was before the iPhone announcement, and it looked almost exactly like the iPhone.
13:29:59LloreanUgh, now *that* is ugly.
13:30:08GodEateryeah - can't say I'm a fan either
13:30:54GodEaterthe touch screen ipod mockup I saw did look a bit iPhone-ish, only without the ear / mic end pieces near the screen
13:31:07GodEaterof course, there's no telling how accurate it was
13:31:20LloreanWell, I guess we'll see what Apple announces soon enough
13:32:08GodEaterthe text in there says pretty much what we've just said
13:32:14GodEaterthat the new ipod will look like the iphone
13:32:21LloreanI'm actually kinda surprised at how quiet everyone's been about the question of Rockbox on iPhone
13:32:28GodEater4th paragraph to be accurate
13:32:50GodEaterLlorean: I can't imagine it's going to be any easier than getting RB on the 2G Nano though
13:33:07GodEaterunless we're just talking "Rockbox as an App" again
13:33:07LloreanActually, I imagine it'll be quite easy.
13:33:10GodEaterrather than as a firmware
13:33:21scorcheand considering there are already mockups (valid most likely) for the 3rd gen...
13:33:25LloreanI've been lead to believe that SDL runs on the iPhone, so just pick your sim of choice. Forget about the "as an app" bit. ;)
13:33:42ddaltonok I have updated about 6 of his patches and I will submit them tomorrow.
13:33:47ddaltonand update the rest.
13:33:49 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
13:34:22GodEaterLlorean: that's pretty neat then :)
13:34:25LloreanWe may need to prepare a "Please do not spam us about new iPod ports" sign, if there's new models coming up
13:34:39GodEaterI presume we'd still need to compile the SDL sim as an ARM arch though ?
13:34:51LloreanI imagine so.
13:35:02*GodEater wonders if anyone has ever tried that
13:35:09GodEateror how we'd test it
13:35:23*Llorean doesn't intend to ever own an iPhone.
13:35:28GodEaternor do I
13:35:34GodEatertoo big
13:35:39GodEaterI like my N73
13:35:50LloreanToo "lacking a tiny tiny keyboard, and replacing it with a screen covered in smudges"
13:36:47*scorche pats his treo
13:37:08preglowiphone is shit
13:37:33preglowif it wasn't apple making it, it would have been forgotten quickly
13:37:35scorchecouls i get you to take a more decisive side?
13:37:59LloreanI think the only decent thing about it is a full-featured web browser coming preinstalled.
13:38:22LloreanBut that's not worth the extra $200 over similar HTC phones with actual 3G radios, etc.
13:38:24scorcheindeed...its browser is one thing that has no equal in that realm
13:38:50GodEaterI wouldn't even consider a non-3g phone now
13:38:58GodEaterit'd be stepping back in time about 5 years
13:39:01preglowgym time
13:39:18GodEaterI've had it for the last three years and it's a massive improvement over the old GPRS system
13:39:24LloreanSo, I guess the new "feature" of the next iPod will be widescreen?
13:39:35GodEaterlooks that way
13:39:49*amiconn actually set one condition for his new phone to be non-3G
13:40:08GodEateramiconn: is that because of the weird charging in Germany amiconn ?
13:40:40GodEaterthen why ?
13:40:40amiconnIt's because a phone is a phone before all. It's not a camera nor a music player nor sth else
13:41:01scorchePDA functionality is damn handy
13:41:05GodEateramiconn: I like being able to get my email on my phone too, and also for web surfing
13:41:14amiconnIf I'd need a PDA, I'd buy a PDA
13:41:25amiconnBut I don't
13:41:25GodEaterone more thing to carry round....
13:41:33scorchei did that, however carrying 2 devides together is bulky
13:41:47scorcheand there is no loss in usage packing it into the phone
13:42:19GodEaterthe PDA I had before I had a 3G phone is now gathering dust
13:42:31GodEaterwhich is a shame, because it was a nice device
13:42:36GodEaterbut it's just not needed anymore
13:42:44scorcheperhaps we should be saying smartphone instead of 3g phone?
13:43:10amiconn3G is nice for data. But that's better covered with a 3G data card for the laptop
13:43:12scorcheindeed...i paid quite a bit for my palm T3, but having a treo is great
13:43:26 Quit midgey ()
13:43:27LloreanMy smartphone is GPRS though. Well "EDGE"
13:43:42scorcheyou dont have to have a data plan to see the immense benefits
13:43:59 Join Daemon [0] (
13:44:04LloreanIf I weren't using my phone for data, I'd have no interest in 3G
13:44:05scorcheand mine is CDMA (such a great technology if you ever get into it...)
13:44:30scorchei am referring to many 3rd party apps and PDA functionality
13:44:35 Join My_Sic [0] (
13:45:19amiconnI do use some of the most basic PDA functionality occasionally. But almost all current phones are able to do that, and it's nothing you need 3G for
13:45:29 Join The-Compiler [0] (
13:45:43amiconn(things like calendar or the occasional wap session)
13:45:58scorche[04:42:43] <scorche> perhaps we should be saying smartphone instead of 3g phone?
13:46:02amiconnIt's not essential functionality for me though, I could live without it...
13:46:15scorchewell, electricity isnt essential...
13:46:21amiconnIt is...
13:47:11Daemonim having some issues with playback on my iPod nano 1st gen, the playback is garbled and the songs continusly skip
13:48:17LinusNwoo, forums are awake
13:50:18LloreanDaemon: Many people are experiencing this problem right now. We haven't been able to identify what exactly is going wrong
13:51:39Daemonany idea when its gonna be fixed?
13:52:11bluebrotheryes: when the problem is identified and a proper way to fix it is known.
13:52:58 Quit bascule ("CGI:IRC 0.5.7 (2005/06/19)")
13:55:11markunB4gder: thanks for renaming the wiki page
13:55:40 Join webguest91 [0] (i=c023110f@gateway/web/cgi-irc/
13:59:06amiconnB4gder: Hmm, maybe ports which are actually in working condition should be removed from OtherTargets?
14:00:00amiconnWell, all ports in svn are "in progress" (except Elio and AV300...)
14:00:46 Quit sbeh (Read error: 110 (Connection timed out))
14:01:10bluebrotherhmm −− but that page doesn't list details about the targets
14:03:11 Part Llorean
14:04:27***Saving seen data "./dancer.seen"
14:06:27 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
14:07:39 Join chrisjs169 [0] (
14:08:21 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:09:23 Quit JdGordon (Client Quit)
14:10:05 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
14:13:16B4gderamiconn: I think working ports should be removed from OtherTargets, yes
14:13:28B4gderI just happened to get an error when I edited it...
14:13:35B4gder... and haven't solved that yet
14:14:13amiconnWell, otherwise some ports would even be missing, e.g. iAudio M5
14:15:38B4gderright, but then I think we should make a wiki page for that, or add info about it to an existing
14:16:19amiconnIt could be added to IaudioPort
14:16:34B4gdersounds like a good place yes
14:16:56*amiconn should also add the M5 pcb photos
14:21:14 Quit x1jmp (Remote closed the connection)
14:22:40 Join x1jmp [0] (
14:23:17 Join JdGordon [0] (
14:23:40 Join Nibbl [0] (
14:23:59 Quit Febs_ (Read error: 110 (Connection timed out))
14:23:59 Join jon_ [0] (n=jonathan@
14:30:24 Quit jon_ ()
14:33:55 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:34:26 Join JdGordon [0] (
14:36:23 Quit linuxstb_ ("Client Exiting")
14:39:07 Quit Guile`` ("KVIrc 3.2.0 'Realia'")
14:40:52 Quit Nibbler (Read error: 110 (Connection timed out))
14:46:12 Join sbeh [0] (
14:55:25B4gder"it seems that intel removed all pxaxxx documentation from their web-site." ...
14:55:39B4gder"And marvell does not seem to publish any documentation to non registered users that did not signed an NDA."
14:55:57B4gderanother one gone secret
14:56:26B4gderpxa == xscale
14:56:41rasherWait, they had documentation on their website, and are now trying to make it secret, retroactively? I don't see that working very well..
14:57:05B4gderhehe, I guess not. I saw this cry for help on the arm-kernel list
14:57:27B4gderintel sold xscale a while ago so it makes sense that they remove the docs from their site
14:57:45 Join seablue [0] (
14:58:13markunsamsung removed the docs for the S3C2440 (Gigabeat CPU) a few weeks after I downloaded it
14:58:23markunmaybe even a few days
14:58:31B4gderthe world is against us!
14:59:54markunthe chinese company who made the gigabeat hardware for toshiba left their intranet open to the outside world long enough so google could cache the spreadsheets with all the part numbers :)
15:01:17markunbut I see it got removed from our wiki (I should still have it somewhere)
15:01:51rasherB4gder: had any time to look at
15:02:08B4gdernot yet, no
15:05:50 Quit hcs (Read error: 110 (Connection timed out))
15:09:42 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
15:10:01 Part Daemon
15:10:59rasherI wonder if the MAINTAINERS file should not contain an email address
15:11:04rasherOf each maintainer, that is
15:11:27rasherObfuscated as much as possible without looking awful, I guess
15:15:42*rasher stares blankly at VOICE_OH
15:15:54rasherHow on earth am I supposed to translate this
15:16:28B4gderprobably "zero" or similar on most languages
15:16:45*amiconn repeats his request for nice ideas on language dependent voicing of certain things (like numbers)
15:16:50rasherThe desc: field really is under-used
15:17:25rasherdesc: spoken only, for wall clock announce .. That should be desc: spoken only, for wall-clock announce. Voiced between hours and minutes if minutes < 10
15:17:29rasheror something like that
15:19:52 Quit Soap ()
15:21:55 Quit B4gder ("It is time to say MOOO")
15:21:56LinusNrasher: well, it isn't restricted to wall clock announcing
15:22:21LinusNbut anywhere you want to pronounce zero as "oh"
15:22:33LinusNlike a telephone number for instance
15:23:30 Join samsggz [0] (
15:23:34 Nick samsggz is now known as spunkei (
15:23:38spunkeiHi guys
15:23:44markunhi spunkei
15:23:53spunkeii got a problem with my iaudio x5
15:24:10spunkeiI just flashed your rockbox firmware on it, everything works, but I have no sound..
15:24:53spunkeiwhat have I done wrong?!
15:25:04markunI have no clue
15:25:13markunmaybe someone with an x5 can answer
15:26:03spunkeiok, i'll wait
15:26:24 Part Gursikh
15:26:48markunspunkei: you could search the forums while you are waiting:
15:27:38markunand maybe post your question there as well
15:28:17 Part LinusN
15:28:47spunkeioh I found a topic!
15:29:17markunwith a fix?
15:29:24spunkeinope =(
15:29:37spunkeionly two responses =(
15:31:26 Quit tedr0ck (Read error: 110 (Connection timed out))
15:35:45pixelmado you have an X5 or an X5V (without radio)?
15:37:42 Join enyc_ [0] (
15:37:42 Join perrikwp_ [0] (n=chatzill@
15:40:00spunkeian X5L
15:40:21spunkeiand I got no line-out or headphone-jack signal..
15:42:03pixelmaspunkei: so you installed the x5_fw.bin and also unzip the X5 build completely to the root of your player?
15:42:19spunkeiyes I did
15:43:49 Quit enyc (Success)
15:44:36 Join sarixe [0] (
15:47:01 Quit sarixe (Client Quit)
15:47:15pixelmaif you play a file - what happens, what do you see on the screen?
15:53:45 Join barrywardell [0] (
15:54:58 Quit kubiixaka (Read error: 110 (Connection timed out))
15:56:39pixelmaspunkei: please give as much details as possible, it's very hard to remotely help (Just to make sure: you installed an official build from the site?)
15:57:08 Quit perrikwp (Read error: 110 (Connection timed out))
15:57:36 Quit JdGordon ("Konversation terminated!")
16:01:58 Quit Zagor ("Client exiting")
16:04:30***Saving seen data "./dancer.seen"
16:06:18rasherbluebrother, Domonoky: I got a response from my post on the tango-artists list. Someone interested in doing tango icons for our supported devices
16:06:57 Quit hcs ("Leaving.")
16:07:41 Quit pill (Nick collision from services.)
16:07:56 Join pill [0] (
16:08:22morrijrnice error from their website: Can't contact the database server: Access denied for user 'tangomw'@'localhost' (using password: YES) (
16:08:37 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
16:09:36rasherYeah, I think they're having some hosting trouble at the moment
16:10:26 Quit BonBonTheJon (Remote closed the connection)
16:13:42 Quit pill (Nick collision from services.)
16:13:57 Join _pill [0] (
16:13:58markunanother rockbox review:
16:18:06desowin"iPods can't play OGGs."
16:19:09 Nick enyc_ is now known as enyc (
16:19:20 Nick _pill is now known as pill (
16:19:23rasherDoes Database use voice in any way?
16:20:09 Quit pill (Nick collision from services.)
16:20:37 Join pill [0] (
16:20:37 Join advcomp2019- [0] (n=advcomp2@
16:21:17pixelmayes, it definitely spells, don't know what else
16:22:06hcs"ad finitum"?
16:22:32hcswhile more technically accurate, probably not what they meant
16:23:00 Join courtc_ [0] (
16:24:26 Quit pill (Nick collision from services.)
16:24:37 Join pill [0] (
16:25:29 Quit pill (Nick collision from services.)
16:25:38 Join pill [0] (
16:27:46 Quit pill (Nick collision from services.)
16:28:08 Join _pill [0] (
16:29:44 Join saratoga [0] (i=98039a9c@gateway/web/cgi-irc/
16:30:01 Quit courtc (Read error: 111 (Connection refused))
16:30:21 Join einhirn [0] (
16:32:44 Quit webguest91 ("CGI:IRC (EOF)")
16:37:24 Quit PaulJam (".")
16:37:33 Quit advcomp2019 (Read error: 110 (Connection timed out))
16:39:26 Join samstown [0] (
16:42:49 Quit lostnihilist (Read error: 104 (Connection reset by peer))
16:45:30 Join lostnihilist [0] (
16:46:41 Quit samstown ()
16:49:43 Quit My_Sic (Read error: 104 (Connection reset by peer))
16:50:02 Join Domonoky [0] (
16:50:13 Join mbr [0] (
16:54:03 Nick courtc_ is now known as courtc (
16:58:21 Join n1s [0] (
17:00:18 Join linuxstb_ [0] (i=d556da1b@rockbox/developer/linuxstb)
17:01:44 Quit spunkei (Read error: 110 (Connection timed out))
17:04:50 Quit n17ikh|Lappy (Connection timed out)
17:07:09 Quit barrywardell ("Konversation terminated!")
17:07:46 Join kubiixaka [0] (
17:14:43 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
17:16:40 Join linuxstb_ [0] (i=d556da1b@rockbox/developer/linuxstb)
17:22:50 Join H10_007quick [0] (
17:23:07 Quit H10_007quick (Client Quit)
17:40:14GodEater <−− linuxstb, seen this ?
17:46:11GodEaterbluebrother: how come you don't have a developer badge yet ? :)
17:46:45bluebrotherGodEater: maybe I haven't earned it yet? ;-)
17:47:00*GodEater thought committer == developer badge
17:47:34bluebrotherwell, when I got commit access I was working on the manual almost exclusively
17:47:39GodEatermaybe Llorean has got lost with who has what badge again
17:47:52GodEateryeah, but you've been adding proper code for a while now :)
17:48:00bluebrotherpossibly. My cloak also doesn't say "developer" (but "staff" instead)
17:48:19GodEaterdid you ask for that to be changed last time scorche was doing them ?
17:48:24bluebrotherand it didn't bother me enough to ask about it ;-)
17:48:29GodEaterthat'll be a no then
17:48:53*GodEater is actually just plotting to reduce the number of people who are a pretty green colour in the forums
17:49:41rasherbluebrother: Are we still interested in tango icons of devices? I got a mail from someone who'd be willing to make them
17:50:14bluebrotherIMO it would be a nice addon.
17:50:28GodEateryeah that would be very nice
17:50:34 Join lee-qid_ [0] (
17:50:41bluebrotherSo if someone is willing to draw some we could integrate it −− shouldn't be to hard
17:51:34GodEaterwho did the existing pictures ?
17:51:52rasherexcept the Sansa E200
17:51:52GodEatertalented girl :)
17:51:59 Quit desowin (Read error: 110 (Connection timed out))
17:52:02 Join Thundercloud [0] (
17:52:08bluebrotherindeed. The images are really nice.
17:52:13rasherSomeone recently posted an E200 to the tango-artists lists actually
17:52:15*pixelma bows :)
17:54:05GodEaterwhat were they done in pixelma ? Photoshop?
17:54:17 Quit Anakin (Read error: 110 (Connection timed out))
17:54:31pixelmano they are vector drawings -> CorelDraw and Inkscape
17:54:49GodEaterah - no wonder they look so smooth
17:55:04GodEaterI ought to try getting the hang of Inkscape at some point
17:55:05rasher −− rename to tar.bz2. They're not GPL though, but ss-by-sa afaik
17:55:12rashereh, cc-by-sa
17:56:10rasherIt's drawn to be an illustration, rather than completely realistic, so it doesn't look as good as "our" svg
17:56:17rasherBut for an icon, it's probably better
17:56:38bluebrotherrasher: file tells me it's gzip, not bzip2
17:56:47rasherAh, you're right
17:56:55rasherOr rather, file is
17:57:04bluebrotherhmm, looks nice
17:58:02bluebrotherit only lacks the Rockbox logo on the display *g*
18:00:14 Join BigMac [0] (
18:03:20 Quit hcs ("Leaving.")
18:04:34***Saving seen data "./dancer.seen"
18:04:57 Join freqmod [0] (
18:08:08 Quit lee-qid (Read error: 110 (Connection timed out))
18:08:31rasherbluebrother: That e200 was done by the guy who contacted me, in fact
18:08:45rasherHadn't noticed that until now
18:14:17 Nick _pill is now known as pill (
18:15:08 Quit linuxstb_ ("CGI:IRC (EOF)")
18:16:48 Join Lear [0] (i=chatzill@rockbox/developer/lear)
18:17:09 Join desowin [0] (n=desowin@
18:22:19 Quit thegeek ()
18:26:07 Join spiorf [0] (
18:27:25 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
18:33:51 Quit obo ("bye")
18:36:22 Join ObeLisK_12183 [0] (
18:38:11 Quit ObeLisK_12183 (K-lined)
18:43:26 Join thegeek [0] (
18:43:36 Quit amiconn (Read error: 110 (Connection timed out))
18:43:37 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
18:45:41einhirnDoes the Archos Recorder support 64Bit Integers? I have a feeling it doesnt.
18:46:12einhirnSomehow my replaygain code does things it shouldnt ;)
18:46:40LearPretty sure it does. Just needs the proper help from the compiler..
18:46:54LearWhat's the problem then?
18:47:32einhirnlike playing back a piece with gain +17db very very quiet while the same piece set to -15db Gain is played rather loud...
18:48:01einhirnOne moment - I've got to recompile ;)
18:48:09LearSounds like a sign problem. Or possibly overflow.
18:50:29einhirnyes something like that.
18:50:48 Nick perrikwp_ is now known as perrikwp (n=chatzill@
18:50:52einhirnI still seem not to get to what I want:
18:51:39einhirnThe gain value is between -48 and +17dB, or max. 7.079...
18:51:52einhirnas a factor
18:53:10 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
18:55:19einhirnOk I am sure I don't make any sense ;)
18:56:50LearNot much I can help you with, given the small amount of information...
18:57:01einhirnOk, lets try a different approach:
18:59:09einhirnIf I turn the Preamp higher than, say, +9db, the overall volume goes down again
18:59:19einhirnSo that would be an overflow, right?
18:59:30 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
18:59:44 Nick datasleep is now known as datachild (
18:59:46 Quit amiconn (Nick collision from services.)
18:59:46 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
18:59:49amiconneinhirn: Do you already have an idea how to fix the latency problem?
19:00:25einhirnWell, only a "dirty one" - how much latency has the mas when rebuffering?
19:00:45amiconnIt's not constant, it depends on *current* bitrate
19:00:59amiconn...because the buffer has a fixed size
19:01:29amiconnI can only imagine a crude approximation, using the average bitrate
19:01:52amiconnEverything else would require additional infrastructure we do not have in place atm
19:02:19einhirnOk, here is what I thought could be done: Just stop the data stream to the mas after one mp3, wait until the mas signals "empty buffer", set replaygain, stream next piece to mas.
19:02:40amiconnThen you will get gaps in playback, which we do not want
19:04:29einhirnOk, so the MAS won't start playback "immediately".
19:04:36amiconnIt will
19:04:58amiconnBut it has to resync to the stream, which will need a few frames
19:05:12einhirnOh, there's the Gap, I see.
19:05:34amiconnThe mas signals "I need more data" as soon as there is enough room in the buffer for another frame
19:06:03amiconnThen there is enough time for the CPU to react, while the mas continues to play from its buffer
19:06:33amiconnThis time varies with bitrate, for 8kbps mp3 (okay, not very common) it is >1 second
19:06:38 Quit lee-qid_ ("aufwiederbyebientotsayonara")
19:06:53 Join lee-qid [0] (
19:07:05einhirnOk, that leaves us with the rough estimate...
19:07:06amiconnIf it hits the end of the buffer and the cpu still didn't feed new data, it gets out of sync
19:07:29amiconnThat would be detectable, but then it's already too late to avoid a gap...
19:07:58einhirnsince the MAS would need to resync. I got that.
19:08:26einhirnWell, maybe someone sees an obvious problem in here:
19:08:27amiconnYes, and the signalisation wold be via i2c polling, which also needs its time
19:10:19einhirnSo, the whole "Time elapsed/Duration" thing displayed in the WPS is guesswork?
19:10:55einhirnI think I saw something like a "duration" field somewhere in ID3 or so
19:11:26einhirnand since the mpeg thread keeps track of the "track time" anyway, maybe we could use that?
19:11:56amiconnIt's not guesswork, it's simply based on what we're currently transferring to the mas
19:13:03amiconnFor typical music, we are (roughly) 50 ms ahead - you can't see such a tiny difference on the lcd
19:13:30amiconn...especially since the lcd adds some lag, compensating the fact that we're slightly ahead
19:13:57amiconnBut switching gain 50 ms too early might be audible, if the gain difference is large enough
19:13:58 Join My_Sic [0] (
19:14:21 Quit My_Sic (Read error: 104 (Connection reset by peer))
19:14:23amiconn...and the track end isn't silence
19:14:48einhirnOf course.
19:15:14amiconnI can't spot a mistake in the code, but I also can't really follow all this back & forth shifting
19:15:31 Join robin0800 [0] (
19:15:39einhirnheh ;)
19:16:13LearThere's a get_replaygain_mode that could be handy here...
19:16:21amiconnI usually work out such formulas on paper, reducing it as far as possible
19:16:35amiconnI would also avoid 64 bit division by all means
19:16:58LearAnd the snippet doesn't say from where all the different values come from (replaygain_preamp etc.).
19:17:03amiconnI'd rather use a multiplication so that the division is a power of 2, and then shift
19:17:37amiconnIt's not so much about execution times (this division is just executed once per track), but about binsize
19:17:56amiconnIirc, the 64 bit division from the gcc lib is a >3KB blob for SH1
19:18:37einhirnyes I can of course use >>4 to divide by 8. That would equal about 18.15 dB as an upper bound.
19:18:59amiconnThat's not what I mean
19:19:12amiconn(and btw, >>4 would be /16)
19:19:41LearComparing against DEFAULT_GAIN after scaling to 0..1 is a bad idea.
19:19:48amiconnI would also think that it should be possible to calculate replaygain with using 32 bit variables only
19:20:08amiconnThe range isn't that large
19:21:03LearCould be a bit tricky to do, without loosing precision, I guess.
19:21:15amiconnI don't think so
19:22:20einhirnLear: I do that because the Scaling to 0..1 is based on the assumption "get_replaygain_int" returns max. 7.1, but there are bigger values possible because of preamp
19:22:46amiconnThe whole replaygain range just equals 11 bits, so using 32 bit vars leaves 21 bits of precision
19:23:49einhirnsince I can only attenuate with the MAS I have to define an attenuation that is used for "1x Gain". In my case that would be -17dB
19:25:04einhirnAmiconn: but when you multiply two 32Bit vars, wouldn't an intermediate value be 64bit at most?
19:25:41amiconnOnly if you need the whole range of both vars, which I doubt
19:26:42amiconnI didn't go through the whole calculation, so I can't be sure whether it's possible. But I would think it is
19:28:54einhirnLear: To answer your questions from about 19:16 - I don't use "get_replaygain_mode" since I can't call from Firmware to App. So I just copied the code and found out that there were many redundant checks.
19:28:59amiconnOn one hand I think it's a good idea to implement replaygain on archos because it's possible, and leaving out a possible feature from a target isn't nice. Otoh I'll probably never use it. I have no replaygained tracks at all...
19:29:58n1sI use it all the time on my h300 it's very nice
19:30:14*bluebrother doesn't want to miss ReplayGain
19:30:42LearAh, you're in firmware, didn't think about that... Could be moved to replaygain.c though.
19:30:56*n1s remembers mix MDs he created several years ago that were a pain when certain tracks came as they nearly blew his eardrums...
19:31:16amiconnLear: The hwcodec playback is in firmware/, so it cannot call code in apps/
19:31:34einhirnLear: The different Values are just copies of the global_settings and of the Replaygain values from the track. Again something with being in firmware.
19:32:13amiconnI want to keep the level differences between tracks, even if it requires to adjust volume a bit between albums
19:32:36 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
19:32:46rasherIsn't that what album mode replaygain is for?
19:33:03amiconnYes, but I also want to keep the differences between albums
19:33:12 Join fnakicoel [0] (i=0@
19:33:23rasheramiconn: then you most be weird
19:33:30amiconnI rarely adjust volume while listening (unless the environmental noise changes)
19:33:32Nico_Pbluebrother: I might have spotted a bug in the rbutilqt theme preview resizing
19:33:45Leareinhirn: Can't say I follow the last steps in your code, but I don't know what values mas wants...
19:33:54Nico_Pbluebrother: the fullsize preview isn't the right size after the dialog has been resized
19:34:02amiconnLear: s0.19 fixed point
19:34:33bluebrotherNico_P: you're right
19:35:09einhirnLear: And from what I understand only values from -1.0 to 0.0 should be used to avoid clipping
19:35:10amiconnAnd it's common to use negative values for straight gain, although not required
19:35:22amiconneinhirn: That's not correct
19:35:52amiconnThe *total* gain per channel should not exceed +/- 1.0
19:35:56einhirnThen I misunderstood the MAS Datasheet.
19:36:25amiconnThat's why stereo width >100% reduces straight gain
19:37:25amiconnBut replaygain doesn't need to care about that, as it would be "multiplied into" the existing values
19:38:02amiconnIt just needs to produce a value between 0.0 and 1.0 in some fixed point representation
19:38:08 Join H10_007quick [0] (
19:38:13 Quit H10_007quick (Client Quit)
19:39:13amiconnrasher: Why?
19:39:27bluebrotherNico_P: should be fixed now.
19:39:39Nico_Pbluebrother: cool :)
19:39:39 Quit HellDragon (Client Quit)
19:39:48amiconnBesides, gain differences between most albums are minor. Just some very old stuff is a little quiet
19:40:31Nico_Pbluebrother: seems fixed here
19:40:32 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
19:40:50linuxstbDoes anyone else only see part of the page when viewing ? i.e. the footer at the bottom isn't there. Attempting to download the page with wget only gets about 80% of the page before it freezes...
19:40:55bluebrotherthat was a quite easy one. But I'm still not sure if I like the way the fullscreen preview works anyway ;-)
19:41:04rasheramiconn: just kidding, but that's the "problem" album gain is supposed to fix, and you're telling you'd rather not have it fixed.
19:41:05chrisjs169if I wanted to make a diff of all modifications made to the local copy of the source, how would I do that?
19:41:40amiconnrasher: The thing is that it imho it isn't a problem, it's intended...
19:41:56linuxstbchrisjs169: If you're using svn, then you do "svn diff > mypatch.diff"
19:42:00pixelmalinuxstb: yeps, me too (even in e different browser)
19:42:01einhirnAmiconn: the multiplying is the part I can't wrap my head around: Maybe I just don't think in 2's complement enough, but I don't get "Larger Values" for the MAS-Register when Multiplying with Gain Values <1...
19:42:05bluebrotherlinuxstb: same here
19:42:27 Quit HellDragon (Client Quit)
19:42:58einhirnIf the MAS can also work with [0 .. +1.0] then the Multiplication would be simple for me ;)
19:43:20linuxstbpixelma, bluebrother: Thanks for checking. I guess we need Bagder/LinusN/Zagor to investigate...
19:43:24amiconnIt doesn't matter whether it's +1 or -1
19:43:27chrisjs169linuxstb: that's what I originally did, but if I added something like the quick/custom menu patch, files from it aren't added - any way to get it to include them?
19:43:32*Lear suspects amiconn doesn't own any "hot" albums.
19:43:41linuxstbchrisjs169: Do "svn add mynewfile.c" for all the new files.
19:43:55 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
19:44:01einhirnHmm. Ok, then I really didn't get 2's complement ;)
19:44:05chrisjs169linuxstb: ok, thanks
19:44:16pixelmalinuxstb: I remember that I'd seen it once before, then the next day it was ok again
19:44:18amiconnLear: Well, digital level can't exceed what's possible on CD, and pratically all albums go near that or even reach it
19:44:40linuxstbpixelma: I wonder if it's a caching issue...
19:44:43amiconnDynamics might make the volume appear a bit different, but it's really not much from my experience
19:44:56amiconnThat includes albums from 1979 through 2007
19:45:42linuxstbpixelma: Yes, it was. I refreshed the cache for that page, and now it seems fine...
19:45:43amiconnlinuxstb: Does the partial page have a "special" filesize, like e.g. exactly 1MB?
19:45:44bluebrotherlinuxstb: looks like it was a caching issue −− I called the twiki cache update and it's fine again (at least for me)
19:46:17amiconnah ok
19:46:20linuxstbamiconn: No, about 160KB...
19:46:38 Quit Nico_P (Remote closed the connection)
19:46:39Learamiconn: A "hot" album is where the volume is raised a lot during mastering, but then limited to avoid clipping. The perceived volume is then much higher.
19:46:49bluebrotherlinuxstb: beat me :)
19:47:03pixelmalinuxstb: ah yes, helped here too
19:47:17amiconnLear: I've never seen/heard/had such an album it seems
19:48:12amiconnI don't consider differences of about 3dB being much
19:48:35amiconn(that's my usual adjustment for some old Midnight Oil albums)
19:50:45LearOn some albums I have, the replaygain can be about zero, or even above, on others, -10.
19:51:08amiconnvery odd...
19:52:08amiconnMaybe the actual difference between my albums is higher, but that's one more reason for me *not* to level them out
19:53:20XavierGrI dislike the fact that in all Rockbox reviews that I read, it is stressed that in order for Rockbox to look good you have to do your research quite a bit. That's why I think that it is time to select default fonts/wps/themes for all targets.
19:54:27linuxstbXavierGr: The release of rbutil (and the upcoming relaunch of should also help a lot with that.
19:55:00XavierGramen to that.
19:55:21XavierGrit is just that a thing so trivial is stressed out very much in the end
19:55:33linuxstbXavierGr: One problem with shipping a default theme that doesn't use the system font is that we'll need to always include that font in the "light" Otherwise a large number of users will boot into Rockbox before installing that font, breaking the theme...
19:55:57XavierGrbut a necessity nonetheless
19:56:44linuxstbI'm not 100% convinced it is. If rbutil becomes the standard install method, it could take care of it, and keep the fonts out of
19:57:41XavierGrif rbutil is the only official way to install rockbox sure
19:57:49Domonokyit should be easy to make a function in rbutil to generate a good config..
19:58:13einhirnAmiconn: If I mulitply 0x80000 (-1.0) by say 0.5, I get 0x40000 because the compiler doesn't know that the 20th bit is the sign. But I would want 0xC0000... So I'd need the original value to be 0xFFF80000 in a signed variable, right?
19:58:13XavierGrbut still I haven't used rbutil lately how is the user prompted about the theme?
19:58:20rasherXavierGr: it doens't have to be the only official way. The manual way could just put more emphasis on installing the fonts
19:58:25bluebrotherwe should add such a functionality anyway
19:58:25*amiconn would rather like to keep the ability to install without special utilities
19:58:42XavierGrwhat if the user just don't want to install a theme thinking that it comes with a default
19:58:56bluebrotherit does :)
19:58:56XavierGrI know I am becoming finicky but the default theme idea sounds more kiss to me
19:59:19XavierGrbluebrother: yeah that default is a fine one on iPod Video :P
19:59:28Domonokythe (not ready) complete installation button, should give you all.. so fonts, a theme you choose, and of course rockbox..
19:59:42bluebrotherrbutil should allow the user to select some sensible configuration
20:00:05bluebrotherand users who don't want to always use a gui tool (like me) can still continue to use the current method
20:00:06amiconneinhirn: Yes, sure you need to care about sign extension
20:00:28preglowthe default font is kinda small for tons of targets, anyone even considering to work on solving that?
20:00:36bluebrotherXavierGr: if you want to try it, I put a binary of the current state online:
20:00:47bluebrothergotta go now. cu later.
20:00:50amiconnBut don't limit yourself by starting at s0.19. It's sufficient if the final values is s0.19
20:01:01XavierGrbluebrother: I have tried it in the past, but that was a long long time ago
20:01:14amiconnpreglow: ALl user screens need to be converted to using the UI font
20:01:18bluebrothera lot has changed :)
20:01:30*bluebrother out now
20:01:50amiconnThat's a necessity, not so much because the builtin font is small, but because it doesn (and shouldn't) have glyphs for all languages
20:01:59XavierGrpreglow: I don't think it is so difficult to just make the include a font that should NOT replace the normal tiny one
20:02:12XavierGrthe UI font is needed in the end for many screens
20:02:23amiconnFor all except debug
20:02:24preglowbtw, i thought icatcher was the default wps now. isn't it?
20:02:35preglowhmm, i wonder where i got that from, then
20:02:39amiconnNo it's not, and I wouldn't like it to be
20:02:40LearUgh, the floppy doesn't sound nice when I click browse (to select my player). :)
20:02:45preglowwell, we should make a default wps, that's for sure
20:02:59preglowall users think the current one sucks rod
20:03:08preglowand yeah, i'm not even hesitant when saying "all users" here
20:03:10XavierGrit is for sure
20:03:22preglowi've gotten used to it, but it sucks badly still
20:03:31*amiconn actually uses the default wps on most of his targets, because he's lazy, and the default does what he wants
20:03:42XavierGrthe defualt is cruel
20:03:43preglowamiconn: yeah, me too, but it's stopping people from wanting to use rockbox
20:03:52amiconnOnly on archos I have my own
20:04:02preglowthat is, my h120 actually uses icatcher, now that i think about it
20:04:07preglowbut i so seldom look at wps :>
20:04:08amiconn(even 2 different ones on bitmapped archoses)
20:04:18XavierGra very bad first impression on those that nonbelievers :P
20:04:38***Saving seen data "./dancer.seen"
20:04:48amiconnI don't like icatcher for a number of reaosns
20:05:05 Join My_Sic [0] (
20:05:12XavierGramiconn: I am sure that all users don't like the default for more reasons that your. But that's not the point
20:05:25XavierGrThe point is to make a default, icatcher or not
20:05:36 Join petur [0] (n=petur@rockbox/developer/petur)
20:06:28XavierGr(a new default that is)
20:06:40amiconn(1) It has no peakmeters. I like peakmeters, except on targets where it slows down things too much atm. (2) It has no disk activity display. (3) The battery and volume displays are weird. (4) depending on disk layout, it needs considerable extra load time
20:08:30XavierGrmost of these are subjective but I agree on (4). Though there was some patches to solve this problem (not only icatcher but all wps's that use many bmps_
20:08:34 Join obo [0] (n=obo@rockbox/developer/obo)
20:10:45 Join H10_007quick [0] (
20:10:51einhirnamiconn: I definitely got an overflow problem somewhere...
20:10:58XavierGrand the sad truth is that if we want a good first impression the new default wps should have a lot of colour and bmps. (for colour targets of course)
20:11:08rasheramiconn: I don't think a default wps should include peakmeters, to be honest.
20:11:22preglowthe peak meter i'd rather be without
20:11:27preglowi never sit just looking at the wps
20:11:33linuxstbWe can always keep the existing default wps for Archos targets for example, where it's a reasonable default.
20:11:35H10_007quicksame here
20:11:38preglowi'd have to be seriously bored to even consider it
20:12:06 Quit H10_007quick (Client Quit)
20:12:22preglowand we want skinnable peak meters anyway :P
20:12:29linuxstbBut do we want to introduce a large UI font into
20:12:33rasherThe important bit is to keep an eye on the difference between "What I'd like" and "what's a reasonable default"
20:12:43XavierGrlinuxstb: I don't think it is possible right now
20:12:47preglowlinuxstb: well, if we can't have another default wps without doing so, then hell yes
20:12:51preglowwhy does it even matter?
20:12:51XavierGrsome screens are hardcoded for that font
20:13:22XavierGrpreglow: to have a new default wps we don't need to replace the UI font
20:13:30preglowi know
20:14:04amiconnXavierGr: huh?
20:14:23XavierGreh sorry I meant replacing the current default font with a larger
20:14:25*linuxstb notes the nimbus fonts are about 30KB, so a negligible increase in the size of
20:14:39XavierGrthat is not needed and would be quite hard to do so.
20:14:49preglowlinuxstb: i'd expect most of them are of negligible size
20:15:06rasherlinuxstb: they also support cyrillic and hebrew, which is not unimportant
20:15:07preglowso i don't see a problem here
20:15:13amiconnXavierGr: The default (builtin) font, yes
20:16:08preglowanyway, are there any nice small fonts we can use as builtin on targets with bigger screens than archoses?
20:16:14amiconnImo it would be more important to convert all user screens to use the UI font before even thinking about packaging a different default theme
20:16:17preglowcp850 8x16, for example :>
20:16:24XavierGrFONT_UI is the tiny one (builtin) and FONT_USER the is the selected right?
20:16:41amiconnFONT_UI is the the user selected one
20:16:57XavierGrand the default? (builtin)
20:17:03preglowsaratoga: yoes, you decided whether to commit seeking or no?
20:17:05 Join SirFunk [0] (
20:17:14amiconnThe builtin one is FONT_SYSFIXED
20:17:23XavierGrah okay forgot the names
20:17:47XavierGrwell I meant that we shouldn't touch FONT_UI
20:17:56linuxstbsaratoga: To answer your question about the audio buffer, if the file contains one audio payload per packet (which all my test files seem to, apart from that broken v1 file), then no, nothing is modified in the audio buffer.
20:17:59saratogapreglow: i want to at least get it seeking in reverse
20:18:00amiconnAs soon as FONT_SYSFIXED is only used for debug stuff, I would even cut it down to pure ascii
20:18:15XavierGrdamn I misused the tags again
20:18:36saratogalinuxstb: 1 payload/packet is normal right?
20:18:37preglowis it used by a lot of user screens?
20:18:40preglowseems really nasty to me
20:18:44XavierGramiconn: if I remember it is used by many screens
20:18:52linuxstbsaratoga: Yes.
20:19:04XavierGrso changing FONT_SYSFIXED should be quite difficult
20:19:08saratogathen i'll implement reverse seeking for now
20:19:12amiconnXavierGr: look back ~3 minutes
20:19:19saratogathat error i was getting is really strange though
20:19:24amiconnI do not want to change it
20:19:43amiconnIts use should be limited to debug screens and severe error messages
20:19:44saratogaeven weirder, it crashes the uisim, but on the target, it just messes up the track position graphic
20:19:44linuxstbsaratoga: You can test by searching for the occurance of "memmove" in wma.c, and add a DEBUGF to tell you when that memmove is being called. In my test files, it never was.
20:19:51preglowXavierGr: any examples? can't say i've noticed it
20:19:53rasheramiconn: changing it would be nice for screens where there's plenty of room though
20:20:11XavierGrpreglow: EQ?
20:20:15linuxstbsaratoga: Is your patch still frame-walking the file?
20:20:20amiconnrasher: All user screens should use the UI font for consistency and language support
20:20:36*amiconn gets the feeling he's talking chinese
20:20:37XavierGrpreglow: rec screen, quickscreen
20:20:52amiconnThe quickscreen uses the UI font
20:21:06rasheramiconn: yes, but still. The builtin font being tiny is silly. Even if it's (eventually) only used in the debug menu
20:21:23linuxstbIt's also in the status bar...
20:21:23XavierGrideally FONT_SYSFIXED should only be used for debug and system errors as amiconn said, but that's not the case right now
20:21:28saratogalinuxstb: yes it is
20:21:32XavierGrand I can't imagine how it is possible to change that
20:21:34preglowrec screen will be a hairy one to fix
20:21:40amiconnXavierGr: So that's the important part to work on...........
20:21:49saratogai don't see any reason to change that, its quick enough even on PP
20:22:04linuxstbsaratoga: Even with large (e.g. 100MB) files?
20:22:05amiconnIt sure is possible, see wps, browser, bookmark selection, settings, ...
20:22:21 Quit ptw419 ()
20:22:25XavierGryeah all is possible but it will be quite a difficult task :\
20:22:25saratogalinuxstb: I didn't try those, but i think in that case, the disk spini up time will dominate
20:22:30amiconnlinuxstb: Yes, and button bar, but all that is fixable
20:22:44saratogai only read a few bytes from each packet, so the cpu time should be tiny
20:22:49linuxstbsaratoga: IMO frame-walking is very bad. We don't do it for MP3 for example.
20:23:03saratogawhats the disadvantage?
20:23:13 Quit perrikwp ("ChatZilla [Firefox]")
20:23:21linuxstbSpeed. We don't want to read 100MB of data from the disk to seek to that point in a file.
20:23:29linuxstb(when an lseek() will do)
20:23:29saratogaoh i see what you mean
20:23:40amiconnIt's the same (in)famous thing all the time: people are thinking about high levels, even if the basics aren't straight and solid
20:23:57preglowlinuxstb: are wma frames timestamped?
20:24:02linuxstbpreglow: Yes.
20:24:02*amiconn is annoyed
20:24:03saratogayes they are
20:24:17saratogathat reminds me
20:24:28saratogait seems the present parser doesn't deal with the last packet properly?
20:25:04preglowseek and resync should do nicely then
20:25:55amiconnXavierGr: It being a difficult task is no excuse
20:26:00saratogaso basically, skip ahead a certain number of packets, and check the time step to see how close you are?
20:26:18preglowwell, if i understood linuxstb correctly, packets are fixed size
20:26:28preglowso you can basically pretty much seek straight to where you're going
20:26:31saratogawhich is a problem if vbr is used?
20:26:34linuxstbsaratoga: Based on the average bitrate, seek to approximately where you think the correct packet will be, then adjust.
20:26:44saratogaok thats what i assumed
20:26:45preglowbinary search if you have to
20:27:02linuxstbpreglow: Yes, or that...
20:27:05saratogaVBR in WMA is probably quite constrained anyway
20:27:09pixelmapreglow: the current default wps not liked by a number of users (those who care only about the looks) could also be an argument for keeping the current one ;D
20:27:25saratogaanyway, i'm somewhat concerned about the last packet problem
20:27:40saratogais it just me or is the parser unable to decode the last packet of every stream?
20:27:45preglowpixelma: i doubt there are many people who care _only_ about the looks
20:28:07XavierGramiconn: wouldn't be a simpler solution (for the new default theme) to just change FONT_UI to the one that will be shipped with
20:29:03XavierGrexcept if you are reffering to the fact that FONT_SYSFIXED has to be changed to be cleaner.
20:29:05amiconnXavierGr: The UI font should be selected per target and set as default to the one shipped, but that's the last step imo
20:29:07pixelmapreglow: but if the wps is the reason to scare people away... (I'm not entirely serious though)
20:29:18amiconnBefore doing that, we need to get our screens straight
20:29:47pixelmaXavierGr: there is another thing about iCatcher, for my M5 it still uses a too small font...
20:29:48preglowpixelma: if you're not entirely serious i won't bother arguing with you :D
20:29:50amiconnThat will also fix the nasty LANG_SYSFONT_* entries
20:30:06preglowbut anyway, good looks never hurt
20:30:11rasherYeah, those aren't nice
20:30:11amiconnImo that's a more serious lack than the default wps. Really and honestly
20:30:13preglowand in our case, the wps looks plain crappy
20:30:20XavierGramiconn: yeah but I don't find those too things so much related
20:30:27XavierGrtoo = two
20:30:37amiconnRight now you can't use russian, or greek, or arabic on the recording screen, in the button bar etc
20:30:57 Join Phalangees [0] (
20:31:07linuxstbamiconn: I agree, but changing the default wps is a 2-minute job... Little work for a big gain in the eyes of new uesrs IMO, so why not do it?
20:31:10preglowbut how to set up the recording screen to use arbitrarily sized fonts?
20:31:17preglowhear, hear
20:31:38*XavierGr says wrs :P
20:31:57PhalangeesI've got a question. When setting the foreground color in a plugin, what exactly do you pass?? rb->lcd_set_foreground(????)
20:31:58amiconnlinuxstb: Because it's the second step before the first
20:32:09preglowamiconn: why does the order even matter?
20:32:11rasheramiconn: the first step being?
20:32:23XavierGrMy opinion is that yes IDEALLY we should change FONT_SYSFIXED too but that is not an obstacle for a new default theme on targets
20:33:00XavierGrrasher: changing FONT_SYSFIXED and making all screens work okay with the different font
20:33:09preglownew wps first or ui font everywhere, it doesn't matter which comes first
20:33:19amiconnAgain, touching FONT_SYSFIXED would be a really really bad idea imo
20:33:32XavierGrnow I got confused...
20:33:43XavierGrah I think I got it
20:33:58rasheramiconn: What needs to be done before changing the default WPS, and why?
20:34:08amiconnIt should stay the same size, and fixed width, and the same width and height on all targets
20:34:21amiconnrasher: The talk was about default theme, not default wps
20:34:23preglowamiconn: even though it's unreadable on ipod video?
20:34:39amiconnpreglow: It's perfectly readable on ipod video...
20:34:43rasheramiconn: okay, default theme then, I dont see how that changes my question?
20:34:51XavierGrpreglow: probably amiconn means that FONT_SYSFIXED should be used only on debug an errors
20:34:56amiconnIt does, because a theme includes a font
20:35:31amiconnA theme cannot look good if it uses different fonts
20:35:33XavierGrbut again I don't see why fixing those screens comes first
20:35:35preglowthis is a consistency problem, users have dealt with inconsistencies before, it's nothing new
20:35:36rasherI don't understand why we can't set a user font and a theme as default. Right now
20:36:04amiconnIt makes the UI inconsistent
20:36:05preglowa new theme just makes the problem more visible, perhaps that'd even make us fix it faster :>
20:36:14rasheramiconn: It will be no different from what users are seeing when they change the theme manually
20:36:28XavierGramiconn: the UI is inconsistent even if you use FONT_SYSFIXED
20:36:30amiconnYes, but tnow they need to mess it up manually
20:36:41XavierGrit is unreadable on larger targets there is no point in that
20:36:48rasherI really think you're making an issue out of something that isn't one
20:37:06amiconnMy opinion is: do it properly or let it altogether
20:38:04rasheramiconn: There's no reason to hold up changing the default theme. It'll be better, right now. Not perfect, but still better. Switching all other screens to use FONT_UI can be done independently, and will gradually make things even better.
20:38:05preglowand now: anyone agree on what the default theme should be? :PP
20:38:42linuxstbpreglow: Of course not...
20:38:45XavierGrit could be different for every target but oh please we must get on that process (of picking them) at some point
20:39:30preglowyes, and how are we going to do that?
20:39:37XavierGrthe wps could be different. The themes should be different obviously
20:39:37preglowit can't be done by arguing about it, that's for sure
20:39:51XavierGrmake the big three decide :P
20:39:55pregloweither someone big make a decision and do it, or we need to vote in some way
20:40:12*amiconn shouldn't have dived into that discussion
20:40:23amiconnNow I'm too annoyed to get back to rockbox coding :((
20:40:41XavierGrwow you are actually annoyed by it? relax...
20:41:11amiconnYes I am. Severely.
20:41:38*einhirn thinks that gcc for sh somehow doesnt cope well with 64bit-Integers...
20:41:48pregloweinhirn: it should
20:41:48rasherI just think we're seeing a case of "perfect is the enemy of the good", with nothing real gained by waiting for the perfect.
20:41:56preglowagain, agreed
20:42:05pixelmamaking a wps that looks good and keeps at least the same feel on all targets is very hard... and it should at least have that, so that someone who knows it from one target will recognise it on the other :\
20:42:19XavierGrIMHO someone shouldn't take those things so seriously. we are just saying our opinions, nothing to be annoyed about.
20:42:24preglowpixelma: i think it's so hard we shouldn't necessarily go for that approach
20:42:33preglowpixelma: icatcher does do a decent job of it, though
20:42:35preglowthe consistency
20:42:50pixelmayes, iCatcher comes close
20:43:19pixelmastill nedore-9 is not very readable on my 160x128x2 display
20:43:21einhirnpreglow: Yes, but when I use "int64_t" and set a Value like 0xFFFF FFFF FF80 0000 the compiler complains about integer constant is too large for 'long' type
20:43:43amiconnYou need to write 0xffffffffff8000ll
20:43:46pregloweinhirn: ll
20:43:49preglowyeah, what amiconn said
20:43:58amiconn(or ull for unsigned)
20:43:59einhirnoh. Ok.
20:44:04preglowc compilers assume all constants are long type
20:44:15preglowamiconn: is LL and ll the same?
20:44:21amiconnI think they assume int type?
20:44:25amiconnpreglow: yes
20:44:27*einhirn always keeps learning
20:44:29preglowamiconn: yeah, they assume int, sorry
20:44:42amiconnOn 32 bit you're lucky that int == long
20:44:51preglowindeed... plenty of people are lucky that's the case...
20:45:05preglowit's gotten better, though, i had tons and tons of those error in pretty much all the code i tried some years ago
20:45:15preglowwhen first going amd64
20:46:49 Quit spiorf (Connection timed out)
20:46:57pixelmabtw. I believe (though the creator should say) that he chose nedore-9 for those displays (especially H1x0 and X5) to have enough lines on the rwps - would be nice if one could chose a different font size for main/remote
20:47:26pixelmawhich is another problem, I'm aware
20:47:34preglowhmm, i wasn't
20:47:37 Join spiorf [0] (
20:47:39XavierGryeah that would be good too
20:53:03 Join spiorf_ [0] (
20:53:49pixelmaor at least I could imagine that it was one of his reasons - at least I find it weird to either chose the same font for the H300/H100 wps so that they can use the same rwps or have a different rwps for H300 with less info (or more sublines/whatever to show the same info)
21:00:13preglowi pretty much forgot remotes existed until you mentioned it :>
21:00:21preglowi don't even know where mine is, heh
21:00:32XavierGrwell I can't imagine the word of rockbox without a remote
21:00:44XavierGrthis is why I still use my H300 over my gigabeat
21:00:57XavierGrthe lcd-remote makes a major difference
21:02:41einhirnWhere would I apply the Settings from "Playback Menu" on Boot?
21:04:06einhirnI always seem to have to change my "enable replaygain" setting to yes again to really enable it, even though it says that it's enabled...
21:04:44einhirn(I have to copy the "Replaygain Enabled" value from global_settings, so where is global_settings initialized?
21:05:01*einhirn greps around in code
21:05:29amiconnglobal_settings is initialized at boot, you don't need to care about that
21:05:50amiconnBut since your code resides in firmware/, you need to add some calls to settings_apply()
21:06:18 Quit spiorf (Read error: 110 (Connection timed out))
21:09:20einhirnHmm - ok, I already added stuff to settings_apply, so maybe it already works and I didn't find out yet ;)
21:20:00 Join andy^rb [0] (
21:21:56 Join merbanan [0] (n=banan@
21:22:01 Join tedrock [0] (
21:23:00jhMikeSamiconn: why the would a function end up in IRAM that isn't declared ICODE_ATTR?
21:23:43Leareinhirn: search for dsp_set_replaygain
21:24:03amiconnIt wouldn't be dsp_* for hwcodec
21:24:18amiconnI think it should go into sound_settings_apply() there
21:24:31einhirnyes - it is, but it somehow doesnt work
21:24:32LearYes, but the call to the hwcodec code should be done at the same place.
21:25:08einhirnAnyway im tired of looking at the code because I can't figure out my problem.
21:25:16LearThen it will always be called when needed. (And that was settings_apply...)
21:25:42einhirnLear: "settings_apply" contains "sound_settings_apply"
21:26:09linuxstbjhMikeS: Is that something in SVN, or something you're working on?
21:26:42jhMikeSqueue_wait shows up in iram but it is not declared that (fm recorder build)
21:26:44Leareinhirn: To debug, you could make your calculations in the sim and use a debugger (or even DEBUGF) to inspect your values, to make sure they're what you expect.
21:26:47 Join tayk [0] (
21:27:00einhirnk, will try sometime else
21:27:04einhirnNow I gtg.
21:27:10einhirnThanks anyway.
21:27:16 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
21:28:12linuxstbjhMikeS: Yes it is - see the top of kernel.c...
21:29:14jhMikeScrap, I was looking in the sim kernel. :P hehe
21:29:50linuxstbsvn blame points to me for adding that line...
21:29:54jhMikeSthey look the same now in what I am working on
21:30:16jhMikeStoo rare a call to worry about really
21:31:35 Join jgarvey [0] (
21:31:55jhMikeSthat gets 88 bytes back
21:32:06jhMikeSeven compiles now
21:32:45amiconnOn SH1 it's usually better to put code in iram than data, unless the data is accessed *really* often
21:32:54amiconn(like the swaptable in bitswap() )
21:33:40taykwhat you type to talk to the robot for a user + Pass
21:35:15 Quit BigMac (Remote closed the connection)
21:35:18jhMikeSI rearranged the thread tables by member size as well to make sure they're packed and aligned and that saved a good bit too
21:36:14jhMikeSIf either queue function is more iram worthy it should be queue_wait_w_tmo anyway
21:38:50pixelmatayk: I don't understand completely.. do you search for a specific user and want to know how to ask logbot, or you want to register/identify your nick or something else?
21:38:58amiconnIn fact I think the whole scheduler isn't work using iram
21:39:07amiconn(on SH1 at least)
21:40:12taykis't there a place to ftp to to find the rockbox files?
21:40:43amiconnIt's only executed on thread switch, and the code that does it doesn't contain extensive loops or such
21:41:02jhMikeSperhaps switch_thread or check_sleepers and of course on pp it must.
21:41:30amiconnDualcore is another thing of course (extra headache)
21:42:17amiconnWay back in archos only times I did a test how many yield() round trips are typically executed
21:42:18jhMikeSI simplified things a good bit anyway
21:42:45amiconnI got ~600 round trips per second, i.e. roughly 4800 yields
21:43:08amiconnThat was almost without load, with load it gets lower
21:43:24amiconnAnd of course it was with the old, simple round-robin scheduler
21:43:28 Quit Galois (Remote closed the connection)
21:43:41jhMikeSof course most threads spends their time not even being looked at now
21:43:45 Join Galois [0] (
21:44:54amiconnI doubt that this actually speeds up things
21:45:36amiconnSome logic still has to check whether to wake up a thread, and that logic became more complex
21:45:59jhMikeSit probably keeps the cores asleep longer though
21:46:21jhMikeSthat is a very rare event
21:47:30 Quit Lear ("ChatZilla [Firefox]")
21:47:30jhMikeSI think it's less logic being executed by far than running every thread until one drops out
21:51:34jhMikeSa yield just does a thead->next and mostly there's only one thread running if playing. otherwise none unless viewing interacting.
21:52:19 Join tvelocity [0] (
21:55:04 Quit The-Compiler (Remote closed the connection)
21:57:05 Join BigMac [0] (
21:58:03 Join sarixe [0] (
21:59:40 Join oxygen77 [0] (
22:00:56preglowthe scheduler is still round-robin, no?
22:01:02preglowjust prioritised
22:01:07 Quit kubiixaka (Read error: 110 (Connection timed out))
22:01:57jhMikeSif that's enabled
22:03:13 Join przemhb [0] (i=530d0d73@gateway/web/cgi-irc/
22:04:41***Saving seen data "./dancer.seen"
22:04:59jhMikeSI'm gonna put a counter in switch_thread. Just curiously.
22:05:15 Quit HellDragon (Client Quit)
22:07:47 Join PrzemyslawHolubo [0] (i=530d0d73@gateway/web/cgi-irc/
22:08:39 Quit przemhb ("CGI:IRC (Ping timeout)")
22:13:44 Quit tayk ()
22:14:05 Quit PrzemyslawHolubo ("CGI:IRC (EOF)")
22:14:28 Join przemhb [0] (i=530d0d73@gateway/web/cgi-irc/
22:15:51 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
22:18:54 Join Merthsoft [0] (
22:19:13MerthsoftHow do I control what folder a plugin gets put into during make ?
22:20:20 Quit bdgraue (Remote closed the connection)
22:21:34 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
22:23:03przemhbMy name is Przemyslaw Holubowski; I am professional electronic engineer from Poland. I would like to contribute to CodecsPerformaceTest Wiki page. Could someone be so kind and grant me access to Wiki?
22:23:41 Quit merbanan ("Leaving")
22:24:04 Join agm3nt [0] (
22:24:36Bagder_sure, tell us your wiki name
22:25:42pixelmaprzemhb: done
22:26:04pixelmaBagder_: he did...
22:26:33*Bagder_ is old and slow ;-)
22:26:38 Nick Bagder_ is now known as Bagder (
22:27:14pixelmaif you say so ;)
22:28:39 Quit HellDragon ("brb")
22:29:03 Quit BobShiel1 (Read error: 104 (Connection reset by peer))
22:31:31 Join H10_007quick [0] (
22:33:05 Part oxygen77
22:33:29oboBagder: has the master build server IP changed in the last day?
22:34:01Bagder91.191.140.131 it is
22:34:07 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
22:35:32 Join ie [0] (i=d9b96768@gateway/web/cgi-irc/
22:35:56obohmm, wonder why the master build server can't connect to mine
22:35:58 Quit przemhb ("CGI:IRC (Ping timeout)")
22:36:08BagderI'll check...
22:36:43Bagdersvn up
22:36:44Bagdersvn: Can't open file '.svn/lock': Read-only file system
22:37:17 Join Febs [0] (n=chatzill@
22:37:51Merthsoftworked fine for me, I hink
22:37:54Bagder(that's on your host)
22:37:57oboeek, dmesg doesn't look good
22:38:04 Quit sarixe ("Peace")
22:39:39 Quit H10_007quick ("ChatZilla [Firefox]")
22:40:23 Join Domonoky_ [0] (
22:41:23 Join sarixe [0] (
22:41:48 Join przemhb [0] (i=530d0d73@gateway/web/cgi-irc/
22:44:56 Quit andy^rb ()
22:46:26ieHello. Has it been noticed that sansa lacks a key for toggling mode in the pitch screen (small steps vs. semitones)
22:47:16 Quit przemhb ("CGI:IRC (Ping timeout)")
22:48:12linuxstb_ie: It has now...
22:48:32 Quit tchan (Read error: 110 (Connection timed out))
22:48:40ielinuxstb: err.... have you just committed a change?
22:49:10 Join przemhb [0] (i=530d0d73@gateway/web/cgi-irc/
22:49:17linuxstb_No. I just meant that everyone has now noticed it - because you mentioned it.
22:49:30linuxstb_What button would you suggest?
22:49:51ieLong SELECT or REC?
22:49:55pixelmaobo: fun thing that fixing a feature (really enabling it) decreases bin size a bit :)
22:50:04ieREC might be better
22:50:22linuxstb_I noticed that REC was free, but I've never used a Sansa.
22:50:42 Quit pill (Nick collision from services.)
22:50:54 Quit freqmod (Remote closed the connection)
22:51:04 Join _pill [0] (
22:52:39przemhbpixelma: Thank you for allowing me to edit Wiki.
22:54:15ieBut keys on sansa seem to be a passionately discussed topic. See :-)
22:56:42 Part ie ("Off to play with my new sansa :-P")
22:57:36obopixelma: it was a bit of a bonus :)
22:57:43 Quit przemhb ("CGI:IRC (Ping timeout)")
22:58:40 Quit Domonoky (Read error: 113 (No route to host))
23:01:12 Quit sarixe ("Peace")
23:02:18 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
23:02:24 Join midgey [0] (
23:04:44 Join bdgraue [0] (
23:05:07 Join jac0b [0] (
23:05:51jac0bI am having a issue trying to compile my gigabeat build
23:06:39jac0bI get this
23:06:40jac0b../tools/configure: line 1490: arm-elf-gcc: command not found
23:06:40jac0bWARNING: The compiler you must use (arm-elf-gcc) is not in your path!
23:06:40DBUGEnqueued KICK jac0b
23:06:40jac0bWARNING: this may cause your build to fail since we cannot do the
23:06:40jac0bWARNING: checks we want now.
23:06:57jac0bI checked my profile and I have it just like the wiki
23:07:08Domonoky_jac0b: you have to add the compiler to your path..
23:07:09 Join ddalton [0] (
23:08:27jac0bhere is my path
23:08:28jac0b../tools/configure: line 1490: arm-elf-gcc: command not found
23:08:28jac0bWARNING: The compiler you must use (arm-elf-gcc) is not in your path!
23:08:28***Alert Mode level 1
23:08:28jac0bWARNING: this may cause your build to fail since we cannot do the
23:08:28***Alert Mode level 2
23:08:28jac0bWARNING: checks we want now.
23:09:15Bagderwas that the output of "echo $PATH" ?
23:10:19jac0bopps I think I need to restart my cygwin
23:10:32jac0bthe paths aren't in there
23:11:56 Quit jac0b ()
23:13:26Phalangeeshow do you do MOD in C?? like 7 MOD 2 = 1
23:13:34Phalangeesmodulus division
23:13:43Bagdera = b % c;
23:13:55saratogais there a dynamic range compression option in rockbox?
23:13:57Phalangeesthanks badger. I didn't know for sure if it was that
23:18:05 Join kyuubiseal [0] (
23:18:26kyuubisealdo guys now any websites where u can get more rockbox games
23:18:29***Alert Mode OFF
23:18:36Febssaratoga: ask Andhyka. ;)
23:20:10linuxstb_Febs: ?
23:20:54linuxstb_kyuubiseal: No
23:21:11kyuubisealanyone else
23:22:39 Quit kyuubiseal (Client Quit)
23:22:58Phalangeesis there an easy way to round in C. I'm getting errors by doing modulus division
23:23:37jhMikeShrm...almost every context switch results in running the same thread
23:24:31jhMikeSout of 540 or so per second at idle, only about 10-12 pick a new thread
23:25:36peturjhMikeS: that most be the priorisation....
23:25:39linuxstb_Phalangees: C integer division will always round down. So for example if you're dividing by 10, you can round to nearest by adding 5 - i.e. (x+5)/10
23:25:58bluebrotherPhalangees: just do it the manual way ... add .5 and cut the decimal digits
23:26:15PhalangeesJeez I feel dumb. I completely forgot about that method.
23:26:38 Quit Galois (Remote closed the connection)
23:26:38 Quit agm3nt (Read error: 104 (Connection reset by peer))
23:26:49 Join Galois [0] (
23:27:06peturlinuxstb: in my experience failing to do that is a very common bug...
23:27:08jhMikeSpetur: I think you're probably right. It's not correct behavior though.
23:28:51 Quit petur ("Zzzzzz")
23:29:24saratogaFebs: yeah i posted in that thread
23:29:34saratogai just didn't remember if there was actually a compression option or not
23:32:01FebsNot that I'm aware of.
23:33:35 Join SoapSud [0] (
23:38:14ddaltoncan anyone connect to using svn up?
23:38:25ddaltonits ok it worked now
23:38:49 Join Bjerrk [0] (
23:39:27BjerrkHm, which device would you say that rockbox works the best on? Performance and battery wise, for example..
23:40:31Bagderbattery wise, the M5L...
23:40:48linuxstb_performance - gigabeat F
23:40:57linuxstb_(as in the target with fastest CPU)
23:41:21BjerrkAh, thanks. How pricey is the M5L nowadays?
23:41:36PhalangeesSorry for interrupting again but is there away to test if a button is being held down rather then just pushed?
23:42:58pixelmasee BuyersGuide wiki page, it's a bit hard to find one (found only one ebay shop which sells a bunch of them every now and then; mp3-direct it's called - an uk store)
23:44:30linuxstb_Phalangees: You can check the current status of the buttons - button_get_status() (or something like that...)
23:44:36BjerrkDamn.. That's a cheap one. But the shipping is $20 :P
23:44:46PhalangeesOk thanks.
23:45:02pixelmaBjerrk: that one - it only has the disadvantge that the charger has UK plugs...
23:45:05linuxstb_Phalangees: Or you can check for "repeat" events
23:45:12SoapSudHello, will there ver be support for the chine replica ipods, as I have searched the internet for firmware upgrades, but cannot find any. i think a change to linux (ie rockbox) is the only choice?
23:45:41Bjerrkpixelma: Hm, crud.. Well, i think i've got an adapter for those, actually.
23:45:46BagderSoapSud: 1) rockbox is not linux 2) ports don't happen by accident
23:45:49linuxstb_SoapSud: If a user (or users) of such devices port Rockbox, then yes. Otherwise it's extremely unlikely.
23:46:14SoapSudah ok, can i go about porting it myself?
23:46:23 Join charge|Roflcpt3r [0] (
23:46:35charge|Roflcpt3rHi, quick question for ya
23:46:42BjerrkBut, feature wise, how well does the M5L work with rockbox? No broken features or grave bugs?
23:46:47SoapSudright, thanks
23:46:49charge|Roflcpt3rIs rockbox compatitlbe with the gmini 402?
23:46:57Bagdercharge|Roflcpt3r: no
23:47:10charge|Roflcpt3ri see
23:47:12charge|Roflcpt3rthank you
23:47:27linuxstb_Bjerrk: The only limitation I can think of is that there is no dual-boot - you need to uninstall Rockbox if you ever want to use the original firmware.
23:47:40pixelmaBjerrk: works very well, just not something for you when you need - colour screen or radio, as I said the BuyersGuide in the wiki gives an overview
23:47:41BjerrkThat's okay by me.
23:47:49linuxstb_Or maybe USBOTG? (does the M5L have that?)
23:47:51 Quit charge|Roflcpt3r (Client Quit)
23:47:53BjerrkBuyersGuide.. aha :p
23:48:01pixelmalinuxstb: you don't really need dual boot on M5...
23:48:59pixelmait doesn't have USBotg but instead of that you have a "normal" usb connection without sub-pack (all in the BuyersGuide... ;) )
23:49:48Bjerrklinuxstb_ and pixelma: The reason i'm asking is that i really like using rockbox on my ipod video, but the battery life is just really, really bad :-(
23:50:25dionoeathat's going to get better :)
23:50:30CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:50:30*dionoea crosses his fingers
23:50:31 Join JdGordon [0] (
23:50:40pixelmayou can use the sub-pack if you have one (for line-in for example) - it's not recommended according to the cowon to establish both USB connections at the same time ;)
23:50:53linuxstb_Bjerkk: How long do you get? People seem to be reporting up to about 10 hours on the ipod video.
23:50:56pixelmas/cowon/cowon manual
23:51:26Bjerrklinuxstb_: 5 hours, i think. If i listen to musik all the time, and fiddle around with the menus once in a while
23:52:01linuxstb_What format music?
23:52:11BjerrkMostly mp3, some ogg.
23:52:19BjerrkNo aac, m4a or anything like that.
23:52:59linuxstb_30GB model?
23:53:45linuxstb_OK... According to this page, about 7 hours seems normal -
23:54:04 Join kapoot [0] (
23:54:24 Part kapoot
23:54:38Bjerrklinuxstb_: Yeah, and if i use the menus once in a while i'm down to maybe 4 and a half hour. :-(
23:54:45 Join kapoot [0] (
23:54:48*dionoea would vote closer to 5hours than 7hours of runtime (I'd have to run a real battery bench to confirm)
23:54:55 Quit Phalangees ("Bye")
23:55:01pixelmaBjerrk: the M5L gives really great battery life ~45hours normal use (I can't remember when I charged last time :D )
23:55:05 Part kapoot
23:55:28 Join H10_007quick [0] (
23:55:32 Join kapoot [0] (
23:55:43Bjerrkpixelma: I want :P
23:55:47kapootsorry, script is getting angry ... anyways i have a question
23:55:57kapootmy nano is a toaster
23:56:20 Join midkay_ [0] (
23:56:25kapootactually it just doesn't play music
23:56:30H10_007quickas all ipods should be
23:56:30kapootany thoughts?
23:56:48 Quit H10_007quick (Client Quit)
23:56:56kapootits the rockbox firmware
23:57:10Bjerrkpixelma: Battery life seems to be one of the most important things to me.. But i can't find the M5L on the iaudio site. I see the M5 there, tough.
23:57:37linuxstb_kapoot: Current Rockbox seems to be unhappy on a minority of Nanos. I think the only solution is to revert to an older build - see the "archived builds" page.
23:57:57kapootlinuxstb −−- i've tried going back to like 14000 and its still fubar
23:58:00pixelmawhere? Basically the M5L is the M5 with a Large battery so to speak
23:58:08kapooti don't think the archived builds go back that far
23:58:35 Join H10_007quick [0] (

Previous day | Next day