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

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

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

#rockbox log for 2009-02-28

00:01:33 Join robin0800 [0] (
00:02:19 Quit robin0800 (Client Quit)
00:02:41 Join gregzx [0] (
00:02:48 Join robin0800 [0] (
00:04:35midgeykadoban: did you try the patch above?
00:04:51kadobanmidgey: what revision is that against? can't get it to apply
00:07:49 Part gartral
00:08:18kadobanoh, git just didn't like it...trying it now
00:08:59 Quit bluebrother (Nick collision from services.)
00:09:04 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
00:09:46kadobanmidgey: yeah, that appears to fix it
00:12:00gotthardtmidgey: works on gigabeat f40
00:12:30midgeyyep, i already tested on my f40
00:12:55gotthardtthen - i concur!
00:13:20midgeyi have a slightly different version and if that works i'll commit it
00:13:20midgeyos x built fine, just testing on target
00:13:24 Quit bluebrother (Client Quit)
00:13:50 Join miepchen^schlaf [0] (
00:15:10 Quit robin0800 ("No Ping reply in 30 seconds.")
00:15:42 Join robin0800 [0] (
00:16:38 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:16:38 Quit pixelma (Nick collision from services.)
00:16:41 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:16:41 Quit amiconn (Nick collision from services.)
00:16:42midgeysorry for the breakage
00:16:50 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:16:59 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:19:02 Quit robin0800 (Client Quit)
00:19:39 Join robin0800 [0] (
00:19:46saratogakugel: I tested 4.3.2 on my Sansa, and both WMA and Ogg got slower
00:19:56saratogaMP3 and AAC crashed with undefined instructions
00:20:09saratogai'd still like to see Coldfire results though
00:26:37 Quit midgey ()
00:27:31 Join CaptainKewl [0] (
00:29:18 Part pondlife
00:29:40 Join rocko [0] (
00:29:55 Quit robin0800 ("No Ping reply in 30 seconds.")
00:33:09saratogaalso, do we have any theories why there are no logs for the second half of today?
00:33:44gevaertssaratoga: the raw logs are fine
00:35:25***Saving seen data "./dancer.seen"
00:36:39 Join avis [0] (n=ident@pdpc/supporter/student/avis)
00:42:57 Quit linuxstb (Remote closed the connection)
00:43:56 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
00:44:16 Quit homielowe (Client Quit)
00:51:07 Quit gotthardt ("ChatZilla 0.9.84 [Firefox 3.0.6/2009020911]")
00:52:49saratogagevaerts: how do i get to them?
00:53:15gevaertssaratoga: "View raw"
00:56:02 Join FlynDice [0] (
00:57:00pixelmaZagor had the theory that the reader chokes on parenthesis in URLs as in wikipedia links
00:57:11pixelmain some
01:01:08 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
01:01:44Unhelpfulgetting back into pictureflow... i think that after i have a "stable" cache manager to resolve the cache fighting, that would be a good place to commit before starting other stuff, like optional use of audio buffer, etc.
01:03:16Nico_Pis one of the swedes there?
01:04:23 Quit orsonj (Remote closed the connection)
01:06:40 Join orsonj [0] (
01:12:01 Quit toffe82 ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
01:18:04 Quit tyfoo (Read error: 104 (Connection reset by peer))
01:19:29 Quit FlynDice (Remote closed the connection)
01:19:54 Join FlynDice [0] (
01:22:53 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
01:32:32 Join iheartdogs [0] (n=424c70e3@gateway/web/cgi-irc/
01:34:16iheartdogshow do i do that trick where you charge the ipod thru usb without connecting?
01:34:38iheartdogsi know ya gotta hold menu but for how long? i have a 4th gen greyscale
01:36:11Unhelpfuliheartdogs: i believe only until you've actually plugged it in, or at least, that's how the "charge only" trigger works on my gigabeat
01:36:40iheartdogsso i let go as soon as it comes on or what?
01:37:46kadobanit's easier if you just start rockbox first, then hold <whatever> as you insert the plug (or at least it's like that on e200)
01:41:34iheartdogsi can't get the timing right, it always registers.. could that be because my ipod is almost dead?
01:42:28iheartdogsalso it doesnt seem to be charging regardless of whether i can do the trick, because it shuts down or gives me the low battery screen.
01:43:48iheartdogssrsly no matter how long i hold the button. won't do it. frustrating.
01:45:43 Quit iheartdogs ("CGI:IRC")
01:53:02 Join moos [0] (i=Mustapha@rockbox/staff/moos)
01:53:40 Quit faemir ("Leaving")
01:57:40 Quit avis (Remote closed the connection)
02:00:02 Quit Nico_P (Remote closed the connection)
02:06:29 Quit Rob2223 ()
02:08:56 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
02:10:04 Join webguestington [0] (n=424c70e3@gateway/web/cgi-irc/
02:10:33webguestingtonis there a known issue with charging problems for ipod 4g greyscale?
02:11:37 Quit saratoga ("CGI:IRC (EOF)")
02:14:15webguestingtonBecause I cannot get it to charge.
02:15:53webguestingtonI would try it in original firmware but i can't seem to load it. i know that was a problem at one point..
02:17:47webguestingtonWhat do you folks think about it
02:19:53 Quit webguestington ("CGI:IRC")
02:25:54LambdaCalculus37Wow, he couldn't even wait for an answer.
02:26:36 Join bs66_1 [0] (
02:35:28***Saving seen data "./dancer.seen"
02:43:41 Join Farglebargle [0] (
02:44:17 Join avis [0] (n=ident@pdpc/supporter/student/avis)
02:45:26FarglebargleGood Evening. Does a 700 MB+ .ipod bootloader file sound correct?
02:46:04Farglebargle753M 2009-02-27 20:39 bootloader-ipodvideo.ipod
02:51:11 Quit bs66_ (Read error: 110 (Connection timed out))
02:51:29 Join refuge [0] (
02:51:37FarglebargleYeah, that's sorta what I was thinking. we're looking for something closer to 738k, right?
02:52:36n1smore like ~50kB
02:53:29FarglebargleHmmm.... I'm pulling SVN, ../tools/configure, picking my 22 (ipod video ) default 32MB memory, and B for Bootloader.
02:54:11n1swhy are you building your own bootloader?
02:54:52FarglebargleThe refurb 5.5 I just got shipped to me has an 80 gig drive.
02:55:31FarglebargleGoogle points me at a posting about editing a header file to bring up the max sector sizes.
02:55:56refugeI have Rockbox 3.1 on my 4th Gen iPod Color/Photo and I love it but I'm having issues with charging. How do I get it to charge when plugged in to an A/C adapter?
02:56:00n1sdoes the released bootloader not work?
02:56:04FarglebargleI figure, what the heck, I have nothing better to do before Monday anyway....
02:56:51refugeI've looked at the applicable manual and I don't seem to have one of the battery settings it mentions. Weird...
02:57:01FarglebargleI get the dreaded : *PANIC* Unsupported physical sector size: 4096
02:57:14Farglebarglewith the stock bootloader.
02:57:40Farglebarglerefuge, if you put it into disk mode, does it charge ok?
02:58:11refugeRight! It'll only charge when plugged into my computer.
02:58:14 Quit Seed (Read error: 104 (Connection reset by peer))
02:58:50refugeIs this normal?
03:00:04FarglebargleI think I remember reading something about a charging circuit needing to have bits twiddled to get it to behave properly.
03:00:34refugeOh dear...
03:02:51refugeWell, I love Rockbox and I'll muscle through the minor bugs. Thanks!
03:04:08 Part refuge ("Hasta la vista, baby.")
03:11:26FarglebargleMUCH better: 48516 2009-02-27 21:10 bootloader-ipodvideo.ipod
03:12:04FarglebargleFWIW, *somehow* I was using the 4.3 toolchain NOT the 4.0.3 ( They're both installed... I guess I picked up the wrong profile...
03:24:59 Quit avis (Remote closed the connection)
03:25:28 Join refuge [0] (
03:25:45refugeI lied earlier. My old iPod _is_ charging... ;D
03:27:14 Part refuge ("Hasta la vista, baby.")
03:29:06 Quit Llorean (Read error: 54 (Connection reset by peer))
03:29:38 Join Llorean [0] (
03:36:06 Join dberg918 [0] (
03:36:41dberg918I have a quick question
03:37:07dberg918which player has more processing power...Gigabeat F40 or H10 20GB?
03:38:59dberg918I'm having skipping issues with my H10 playing certain FLAC files
03:39:24 Quit tvelocity ("Αποχώρησε")
03:39:54krazykitthe former, but the h10 should have plenty of power to decode FLAC (unless the disk is too slow?)
03:40:30dberg918it plays the majority of my FLAC files just fine, but there's a certain group that it can't handle right at the beginning
03:41:02dberg918it will play the first second, freeze for a moment, and then continue to play the rest of the file
03:41:26dberg918and it will be fine playing through tracks after that, unless I skip a track
03:45:39LambdaCalculus37How were those FLAC tracks encoded?
03:45:56krazykitare the flac files broken? flac -t will tell you (on a computer)
03:46:16dberg918yeah, I checked them with flac -t, they're all fine
03:46:22dberg918they play just fine on my Gigabeat
03:46:32dberg918and they play on my desktop too
03:46:52scorchejust to double check, are you running the lastest build?
03:47:01scorchelatest too
03:47:18dberg918yes, r20135
03:47:25 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:48:03n1sdberg918: that sounds like the decoder cohokes on something in the beginning of each file, any exotic metadata perhaps?
03:48:55 Quit z35 (Read error: 54 (Connection reset by peer))
03:48:57dberg918would it show up in EasyTAG?
03:49:06dberg918that's how I tagged them
03:49:35n1snot sure
03:49:40dberg918I'm not totally sure what you mean by "exotic"
03:49:53 Join z35 [0] (
03:51:14saratogalarge tags or binary data
03:51:15n1sdberg918: i mean anything that would be likely to cause problems basically, things like embedded album art, huge comment tags, lyrics, tags not belonging to that format at all etc
03:51:55dberg918nope, nothing like that
03:53:09saratogaif you can't think of anything, you could just to transcode to them to new flac files, as that will fix whatever is wrong
03:53:52dberg918will it? I was actually considering doing something like that, but I wasn't sure if would do anything
03:54:27krazykitit might. it can't hurt to try, anyway
03:54:52 Join Seed [0] (
03:54:54FarglebargleThat worked much better.
03:55:27dberg918I've never done this before, do I need to uncompress them, or can I just feed in the original .flac files?
03:56:32 Quit agaffney (
03:57:01 Join jeffronius [0] (n=jomegata@
03:58:00 Part jeffronius ("Time makes no sense")
03:58:11 Quit n1s ("Lmnar")
04:02:38 Part Farglebargle ("Leaving")
04:06:06NJoinagaffney [0] (n=agaffney@gentoo/developer/agaffney)
04:06:07 Join avis [0] (n=ident@pdpc/supporter/student/avis)
04:09:35 Quit nuonguy ("This computer has gone to sleep")
04:12:08 Join RaRe [0] (
04:13:12 Join blkhawk- [0] (
04:16:33 Join midgey [0] (
04:29:33 Quit blkhawk (Read error: 110 (Connection timed out))
04:35:05 Join SirFunk__ [0] (
04:35:14 Join midijunkie [0] (
04:35:33***Saving seen data "./dancer.seen"
04:38:47 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
04:39:01 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:40:14 Join domonoky1 [0] (
04:40:22 Part domonoky1
04:40:58 Join FunkyBob_ [0] (
04:41:52 Nick FunkyBob_ is now known as FunkyBob (
04:42:39FunkyBobmorning, all
04:43:20FunkyBobI was quite excited when I found RockBox (am and even more impressed now that I've installed it... very schmick!) ... as I've been toying with building my own music player, as a bit of an exercise
04:43:49 Join Barahir_ [0] (
04:44:19FunkyBobI was hoping to talk with some people who've dealt with the hardware on what they felt works best, gives good results, etc
04:44:48scorchewell, it all depends on what the purpose is
04:45:44scorchealso, this might be better in #rockbox-community as at this point, it isnt too topical to rockbox itself
04:46:01 Quit SirFunk_ (Read error: 110 (Connection timed out))
04:46:16FunkyBobmy original motivation was when I realised that through work I'd toyed with touch sensitive surfaces, small colour OLEDs, SD cards, and embedded ARM chips.... it all seemed so obvious :P
04:46:54linuxstbFunkyBob: You may be interested in this -
04:47:49FunkyBobfairy nuff... my thinking was that was more for social and this is more dev... and my questions are more dev
04:47:54FunkyBoblinuxstb: indeed I am... thanks :)
04:48:59 Quit avis (Read error: 104 (Connection reset by peer))
04:49:36 Join avis [0] (n=ident@pdpc/supporter/student/avis)
04:50:00dberg918okay, new development in the flac issue on my H10 20GB
04:50:03 Join hillshum [0] (n=hillshum@unaffiliated/hillshum)
04:51:23dberg918the player skips when using the cabbie v2 theme and my own personal theme "grunge," but not when using minimal themes like engineer2 or boxes
04:52:07hillshumWhat is needed for the e200v2 charging?
04:52:26linuxstbdberg918: Could these be high resolution files? e.g. 24-bit/96KHz?
04:52:48dberg918the images?
04:53:01linuxstbThe FLAC files.
04:54:47FunkyBoboh, my other question.... I have a little iPod<->Radio adapter (from Belkin)... was wondering what I could do to help to get support
04:54:52dberg918nope, I re-encoded them to make sure, they're 44100
04:55:37perrikwplinuxstb: were you the one making a Gigabeat S patcher?
04:55:45 Quit Barahir (Read error: 110 (Connection timed out))
04:55:47linuxstbperrikwp: Yes
04:56:01Unhelpfulare they 24-bit? i'm not sure we can handle 24-bit files from the newest flac reference encoder, yet
04:56:09perrikwplinuxstb: Did it work for windows?
04:56:23dberg918Unhelpful: how can I check?
04:56:54linuxstbYes - for one person. Although the problem others had was that Windows simply didn't recognise their S in MTP mode.
04:57:18perrikwplinuxstb: How can I help test it out?
04:57:23hillshumThe main problem with MTP
04:58:04Unhelpfuldberg918: i would expect any media player worth anything to be able to tell you if they're 24-bit? :/
04:58:34linuxstbperrikwp: You could try this -"> You also need this dll in the same directory -
04:58:47linuxstbperrikwp: Just attach your S to Windows (in MTP mode) and run beastpatcher.exe
04:59:02linuxstbIt will install a single-boot bootloader.
04:59:04perrikwplinuxstb: Ok thanks!
04:59:26*linuxstb leaves for a bit
05:02:09dberg918Unhelpful: I have no idea...what is "24-bit" referring to, a bit rate?
05:02:30FunkyBobthe sample size
05:02:51FunkyBobas opposed to 16bit (CD quality)
05:03:41 Quit LambdaCalculus37 ("Fwump")
05:04:04dberg918is 24-bit encoding a relatively new thing to FLAC?
05:05:53 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
05:07:31hillshumWhere is the e200(v2) charging handled?
05:08:17FunkyBobdberg918: yes, it does seem to be fairly recent
05:09:01dberg918then no
05:09:07dberg918they're 16bit
05:14:08 Quit avis (Read error: 60 (Operation timed out))
05:14:54hillshumshould I copy /firmware/target/arm/as3525/sansa-clip/powermgmt-clip.c to /firmware/target/arm/as3525/sansa-e200v2 ?
05:16:12 Quit midijunkie (Read error: 104 (Connection reset by peer))
05:17:20hillshumor /fimware/target/arm/sandisk/sansa-e200 ?
05:28:52 Part FunkyBob
05:37:44 Quit gevaerts (Nick collision from services.)
05:37:55 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
05:38:13dberg918maybe I'll file a bug report tomorrow
05:38:18dberg918I'm off to bed
05:38:41 Quit dberg918 ("Ex-Chat")
05:39:33 Quit linuxstb (Read error: 60 (Operation timed out))
05:48:22 Join avis [0] (n=ident@pdpc/supporter/student/avis)
05:50:09 Quit hillshum ("Leaving")
05:52:14 Quit Seed ("cu, Andre")
05:57:57 Join ashes [0] (
05:58:37ashesi just upgraded rockbox. the playing display is changed. i'd like it to display the current time of day.. like before
05:59:08midgeyyou'll have to modify the WPS
05:59:27ashescan you tell me where to find it in the menu?
05:59:59midgeywps is a file that describes the playing screen. it's how we have customizable now playing screens
06:00:23midgeyso your best off modifying it on a computer
06:00:33Unhelpful24-bit encoding in flac is not new, but recent releases (since a year ago or so, now, i think) encode these files differently, and i'm not entirely sure that we can handle this change yet.
06:01:23asheswhere is the WPS file located?
06:02:14midgeywell if you're using the default theme, it will be /.rockbox/wps/cabbiev2.wps
06:03:08midgeyhave at look at the wiki and it will show you what tags are valid
06:05:06ashes %cH:%cM:%cS
06:05:11ashesappears to be in the file
06:05:48ashesbut is not displayed during play
06:06:04midgeywhich wps are you using?
06:06:20ashesfrom last year's themes
06:08:38 Quit __lifeless (Remote closed the connection)
06:08:44midgeyis there anywhere I can find the theme?
06:08:54 Join __lifeless [0] (n=lifeless@
06:09:04 Quit Tristan (Remote closed the connection)
06:09:18ashesyou want me to upload it somewhere?
06:09:31midgeythat'd be helpful
06:11:57 Join Tristan [0] (
06:13:43ashesshould i try the copy from ?
06:15:29ashesits huge
06:18:39midgeyhmm that theme doesn't work at all on my iriver, i'd guess it needs some updating
06:19:03 Join sarixe [0] (
06:19:19ashesi used rbutilqt-v1.0.9 to upgrade
06:19:31 Quit mirak (Read error: 60 (Operation timed out))
06:19:41ashesthe gui says that themes are being reworked, and i should install manually
06:19:55midgeywhich build did you update to?
06:20:06ashesit upgraded rockbox itself though, to current
06:20:14ashestoday's svn/cvs version
06:20:16scorcheit says that the *theme site* is being reworked
06:20:34scorcheso you should install themes the way you always have done
06:20:41ashesit doesn't allow me to install themes
06:23:09*midgey kicks osx
06:23:23midgeyi can't build an h300 sim to test your wps, sorry
06:23:40midgeyi swear i will beat mac into shape
06:23:55ashesthe new themes rar file has a bunch of .bmp files
06:24:01ashesi don't know where to put them
06:24:35midgeycopy each folder into .rockbox on your h300
06:25:01ashesthere are no folders. the rar file extracts every file into the current directory
06:25:57midgeyhmm? when I extracted that rar I got backdrops, themes, and wps folders
06:26:28Unhelpfulashes: you need to see how to correctly extract folders with your rar program.
06:26:49scorcheif the archive was properly formed, it should be designed with a .rockbox folder in the root, and then the corresponding folders, etc....since that is from a 3rd-party site, it appears that they dont follow our rules to make it easier...
06:27:06 Quit rocko ("Leaving")
06:27:14scorchein that case, extract the wps, backdrops, etc folders into your .rockbox folder on your device
06:27:20ashesi used the "e" option with unrar, which extracts into the current directory
06:28:15ashesi should use "x"
06:32:40ashesno change
06:32:56ashesi unrar'd the new theme, shut down, and restarted
06:33:11ashesno clock in the playback screen
06:34:02ashes1 sec
06:35:11ashesmy .wps file has %cH:%cM:%cS
06:35:18ashesbut i don't see it
06:35:35***Saving seen data "./dancer.seen"
06:37:20Unhelpfuldid you add that to an existing wps file? and if so, where? added lines can very easily end up being displayed off-screen.
06:38:04ashesit's in the Arctic_Desert/wps/Arctic_Desert.wps file, it's part of the default theme
06:38:14ashesits default for the theme
06:38:42scorche[22:27:08] <scorche> in that case, extract the wps, backdrops, etc folders into your .rockbox folder on your device
06:38:58ashesi did that
06:39:02scorche.rockbox/wps.Arctic_Desert.wps not Arctic_Desert/wps/Arctic_Desert.wps
06:39:19ashes1 sec
06:41:26ashesi extracted the rar, then copied Arctic_Desert/* to ., in .rockbox
06:42:46ashesi do see the time of the audio file
06:42:57ashestime left, and time played
06:43:06asheswhat i want is the current time of day
06:43:55ashesa clock
06:45:50midgeydoes the time appear in the menu (in the upper right)?
06:46:24ashesit used to
06:46:44ashesthere is a stable release for iriver now?
06:46:56midgeygo to System > Time & Date
06:47:04midgeyand tell me what you see
06:47:16midgeyyes, 3.0 and 3.1 were both released for h300
06:47:27ashes00:47 feb 28, 2009
06:47:40asheswould i be better off with the stable release?
06:50:46scorcheno you wouldnt...there is nothing wrong with the latest build, but with the WPS you are using....
06:55:19 Join homielowe [0] (
07:11:58 Quit BHSPitLappy (Remote closed the connection)
07:17:18 Quit perrikwp (" ajax IRC Client")
07:21:23 Join webguestimo [0] (n=ce748651@gateway/web/cgi-irc/
07:21:40 Quit Zoxc ()
07:23:34 Quit webguestimo (Client Quit)
07:32:33 Join Seed [0] (
07:35:18 Quit Aurix_Lexico (Remote closed the connection)
07:37:55 Quit yosafbridge (Read error: 60 (Operation timed out))
07:45:35 Join yosafbridge [0] (
07:53:47 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
07:57:47 Quit avis (Read error: 113 (No route to host))
08:15:30 Quit Llorean (Read error: 104 (Connection reset by peer))
08:15:43 Join Llorean [0] (
08:24:25 Quit shodanX (Read error: 110 (Connection timed out))
08:28:56 Join rocko [0] (
08:35:36***Saving seen data "./dancer.seen"
08:38:55 Join shodanX [0] (
08:40:04 Quit shodanX (Remote closed the connection)
08:40:48midgeyand to bring a conversation over from -community, osx still has problems building certain sims
08:41:00midgeyh300 being one for sure
08:41:13amiconnmidgey: Rename the osx.dummy.bmp in remote_mono to e.g. osx.dummy2.bmp, and change SOURCES accordingly
08:41:15midgeythe produced error is:
08:41:17midgeyld: duplicate symbol _osx in /Users/tjross/code/rockbox-devel/builds/h300-sim/apps/bitmaps/remote_mono/osx.dummy.o and /Users/tjross/code/rockbox-devel/builds/h300-sim/apps/bitmaps/mono/osx.dummy.o
08:41:22midgeythat's the plan
08:42:02midgeythe dummy files were added for a quirk in 10.4, the quirk might be gone in 10.5
08:42:09amiconnThe osx toolchain is weird. Different section names than anything else, unable to deal with empty libs....
08:42:36midgeyif by weird you mean broken, then yes
08:43:14 Join nsfx [0] (
08:43:22midgeyunless things have changed recently, you'll get a ton of warnings build sims on 10.4
08:43:46midgeyi know i don't get them anymore on 10.5, but it might be a change with the build system
08:44:21midgeybasically it didn't like that we had functions that matched system library functions
08:46:32nsfxthe rockbox installer says to reboot the player into the original firmware. how do you do this with an iriver h1x0?
08:48:07midgeyhave you checked our lovely manual?
08:48:17midgeyspecifically section 3.1.3
08:48:57nsfxi was focusing on the "my orig firmware won't boot" section
08:49:03nsfxthanks midgey :]
08:49:09midgeyno problem
08:53:33*amiconn just found a little size optimisation for SH1
08:53:54amiconnIt's a compiler option - need to figure out where to put it
08:55:31amiconnHmm, looks like a job for configure
08:55:57amiconnStandard section alignment on SH is 16 bytes, but it can be changed to 4 bytes using -small
08:58:17 Quit Seed ("cu, Andre")
09:01:11nsfxof course now rockbox won't load at all
09:08:26 Nick Barahir_ is now known as Barahir (
09:11:00 Join shodanX [0] (
09:11:02 Quit __lifeless (Remote closed the connection)
09:11:12 Join _lifeless [0] (n=lifeless@
09:13:52amiconnHmm, -Os seems to already add -small to the assembler options
09:20:37amiconnSo no further savings...
09:28:12 Join bluefoxx_ [0] (
09:29:33 Quit bluefoxx (Read error: 60 (Operation timed out))
09:30:49 Join miepchen^schlaf [0] (
09:32:31 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:35:12 Quit rocko ("Leaving")
09:37:45 Join stoffel [0] (
09:42:36amiconnMaybe I should repeat my questions regarding plugin startup on the dev ml
09:45:53Unhelpfuli certainly missed seeing the questions entirely, others may have as well...
09:46:19 Quit midgey ()
09:48:51amiconnYesterday's log, 23:30 thru 23:48
09:52:26 Part nsfx
10:02:49Unhelpfullog seems to be truncated... but i think i found it. i'm not entirely sure i follow, other than apple being happy with __data rather than .data
10:04:26 Quit intrados_ (Read error: 104 (Connection reset by peer))
10:05:34 Join intrados_ [0] (
10:05:41amiconnUSe the raw log. The log viewer chokes on certain URLs and doesn't show the whole log in such a case
10:06:00amiconnBasically there is an inconsistency between plugin and codec startup.
10:06:37amiconnOn targets not using iram, codecs get their bss cleared by the loader, while codecs clear it themselves (in codec_crt0.c)
10:07:58amiconnOn targets using iram, codecs also init iram in codec_crt0.c, while plugins have to call plugin_iram_init() which is a core function
10:08:56amiconnIf a plugin actually uses iram, the plugin loader doesn't clear all .bss, but only the part that doesn't overlap the iramcopy. The latter part is cleared in plugin_iram_init() as well
10:09:41amiconnMy suggestion is to introduce a plugin_crt0.c that handles things similar to codec_crt0.c, and get rid of the clearing in the core
10:09:46 Quit SirFunk__ (Success)
10:10:20 Join SirFunk__ [0] (
10:10:54amiconnThe difference is that plugins which do use iram need to stop playback (because they use the same iram block as codecs). But this can be handled in crt0 by looking at the iram section size. If it is zero, no init and hence no playback stop is necessary
10:10:58Unhelpfuland this got exposed when the os x sim fix cleared the section attribute from ci and rb for builds without iram, right?
10:11:40 Join daurnimator [0] (
10:11:41 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
10:13:04amiconnMy suggestion would make things easier for plugin devs, at the price of losing the ability to delay iram init, e.g. for asking the user first
10:13:50amiconn*but* this special trick is potentially dangerous with the current section layout anyway, and making bss and iram not overlap increases ram usage of plugins (if they use iram)
10:14:14Unhelpfulhrm, yes, "i have to stop playback, is that ok" would be nice, but not if it would be unstable
10:14:15amiconnMy suggestion would also reduce core binsize a bit, at the cost of increasing plugin binsize a bit
10:15:27amiconnUnhelpful: It is unstable if something (e.g. this request) starts using bss before iram is inited, and the part of bss it uses actually falls into the iram copy
10:16:43amiconnThe request could be added to my plugin_crt0 idea as well, at the cost of more binsize increase, even for plugins not using iram at all
10:17:43Unhelpfulhrm, i don't like *that* solution
10:20:49Unhelpfulam i right in thinking that making a request as the first action in the plugin, and immediately doing iram init if permission is granted, or exiting if not, should be safe?
10:25:15Unhelpfulalso, how many plugins both 1) use iram and 2) do *not* obviously require stopping audio playback anyway (ie, because they play videos, or are games with sound)
10:31:14 Join nuonguy [0] (
10:35:38***Saving seen data "./dancer.seen"
10:47:49 Join insanepotato [0] (
10:56:10 Join awake_ [0] (
10:56:59*amiconn found why the log viewer barfs on URLs with parentheses in them
11:00:15 Join Darksair [0] (n=user@
11:05:13amiconnHmm, or maybe not
11:09:53Unhelpfulamiconn: i think i've found why PF is horrid on my e200... it's not the new slide cache, it's apparently the last change i put up on FS, to fetch the current title on demand.
11:13:56 Quit BigBambi (Read error: 113 (No route to host))
11:14:34 Join BigBambi [0] (
11:14:55insanepotatoi think my lcd is broken. im getting streaks of orange across it, even when in OF
11:15:17 Quit markun (Read error: 104 (Connection reset by peer))
11:15:31 Join HackyKid [0] (n=HackyKid@
11:15:51 Join nibbler_ [0] (
11:17:17 Quit nibbler (Read error: 60 (Operation timed out))
11:18:31Unhelpfulhrm, fixing the incorrect cod that was intended to guard against fetching the same cover repeatedly makes it work faster, but there's a visible stall each time the album name changes. there was no problem on beast, even with that guard broken, but that might be down to the higher clock speed
11:18:42 Quit z35 ("Leaving")
11:19:53 Join markun [50] (n=markun@rockbox/developer/markun)
11:22:23 Join zimba42 [0] (
11:27:23 Quit zimba42 ("Leaving.")
11:29:06 Join stripwax [0] (
11:29:18 Part insanepotato
11:38:46 Join faemir [0] (
11:40:05 Quit stoffel (Read error: 113 (No route to host))
11:45:58 Quit nuonguy ("This computer has gone to sleep")
11:53:53 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
11:57:13 Join jaykay [0] (
11:57:36 Join ender` [0] (
11:58:18jaykaypixelma: were there still problems with
12:00:45 Quit stripwax (Read error: 104 (Connection reset by peer))
12:04:15 Join stripwax [0] (
12:14:16 Join gregzx [0] (
12:14:43 Quit nibbler_ (Read error: 101 (Network is unreachable))
12:15:05 Nick fxb__ is now known as fxb (
12:34:07 Quit BigBambi (Read error: 54 (Connection reset by peer))
12:35:43***Saving seen data "./dancer.seen"
12:37:51 Join BigBambi [0] (
12:42:40 Join Rob2222 [0] (
12:44:33 Quit BigBambi (Read error: 54 (Connection reset by peer))
12:47:45 Join BigBambi [0] (
12:53:58 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
12:55:51*B4gder welcomes committer #70
13:08:26 Quit miepchen^schlaf ()
13:15:43 Join n1s [0] (n=n1s@rockbox/developer/n1s)
13:28:33 Join miepchen^schlaf [0] (
13:44:01pixelmajaykay: well, me having internet connection problems and then forgetting about it. I'm still not sure what to do with the "mp3 frames" and "Festplatte/Datenträger" - it is possible to be different to English there and I'd prefer it
13:44:58pixelmaif I don't forget, I will commit the non-controversial part tomorrow
13:45:15pixelma(don't have much time noe)
13:45:15jaykaypixelma: i also dont know about the mp3-frames, but datenträger/festplatte is only syncing to the english lang
13:45:32pixelmayes, but it's not necessary
13:46:33*BigBambi welcomes fml
13:48:58pixelmajaykay: you can diverge from the master and to me it makes sense in German (IIRC it was put there on purpose)
13:48:58jaykaypixelma: but its redundant and could be removed... anyway, im also happy if the rest gets committed :)
13:50:28pixelmaI don't think it's redundant, just more precise (maybe more personal to the reader with an HD based target)
13:50:35 Quit Barahir (Read error: 110 (Connection timed out))
13:50:45 Quit BigBambi (Read error: 60 (Operation timed out))
13:51:13pixelmaI don't know what happens on CF modded devices though... I'll think about it :)
13:55:46 Join fdinel [0] (
13:58:21 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
13:59:28gevaertsIs there any progress on the 1st gen ipod scrollwheel issue? Is anyone actually still thinking about it?
14:00:26B4gderI can't recall seeing anyone
14:02:54gevaertsThe patch in FS #8778 looks pretty simple. Are there risks of it suddenly not identifying G2s properly any more?
14:03:22 Join stoffel [0] (
14:03:36B4gderafaik, amiconn is the only one who ever use one of those targets
14:04:58n1sgevaerts: i think that's it, yes
14:05:15gevaertsEvery now and then an actual first gen user turns up. It would be a pity to drop support for only this reason
14:05:36 Join BigBambi [0] (
14:06:33gevaertsAny reason not to make two builds then, one normal and one "force 1st gen"?
14:06:38pixelmaIIRC there was a problem with this
14:07:08pixelmacan't remember the details, amiconn knows more
14:07:13 Join tyfoo [0] (
14:07:49 Join stripwax5443 [0] (
14:11:57 Quit stripwax5443 (Client Quit)
14:12:50 Quit B4gder ("It is time to say moo")
14:14:16 Join stripwax5443 [0] (
14:20:51 Quit miepchen^schlaf ()
14:28:14 Quit stripwax5443 ("")
14:32:08 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
14:32:16 Quit domonoky (Read error: 104 (Connection reset by peer))
14:35:44***Saving seen data "./dancer.seen"
14:36:39 Quit ender` (" Sex is dirty - if you do it right. -- Isaac Asimov")
14:38:36gevaertsDoes screens[x].transparent_bitmap() just draw a bitmap on top of what's already there at the position specified?
14:39:36 Quit markun (Read error: 104 (Connection reset by peer))
14:43:59 Quit FlynDice (Remote closed the connection)
14:45:25 Part RaRe
14:46:02 Quit Nico_P (Remote closed the connection)
14:47:14 Join mcuelenaere [0] (
14:47:38 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
14:53:16awake_does rockbox support embedded cuesheets in flac files?
14:57:50BigBambier, no
14:57:52BigBambinot embedded
14:58:34BigBambiIf they are seperate they should be fine (certainly for MP3, I assume for FLAC)
15:00:18awake_ah okay, thanks
15:06:46 Join {phoenix} [0] (
15:12:57 Join avis [0] (n=ident@pdpc/supporter/student/avis)
15:13:22 Join Aurix_Lexico [0] (
15:15:13 Join midijunkie [0] (
15:16:07 Quit gevaerts (Nick collision from services.)
15:16:16 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
15:16:43 Join miepchen^schlaf [0] (
15:20:41*rasher wonders what the build system is up to
15:20:47 Quit J-23 (Excess Flood)
15:20:58 Join J-23 [0] (
15:25:38 Quit midijunkie ("?(???~~)?")
15:25:48rasherBagder: ?
15:25:55 Join midijunkie [0] (
15:29:31 Quit stripwax (Read error: 104 (Connection reset by peer))
15:32:11 Join robin0800 [0] (
15:32:54 Quit stoffel (Read error: 113 (No route to host))
15:33:13 Quit robin0800 (Remote closed the connection)
15:34:30 Join stripwax [0] (
15:39:49CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
15:42:52CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
15:45:04 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
15:45:47 Quit stripwax (Read error: 104 (Connection reset by peer))
16:03:27 Quit Horscht (Read error: 110 (Connection timed out))
16:11:11 Quit mcuelenaere (Read error: 54 (Connection reset by peer))
16:11:18 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
16:12:10 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
16:13:33 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:15:09 Join ender` [0] (
16:17:50 Join miepchen^schla [0] (
16:18:20 Join mcuelenaere_ [0] (n=mcuelena@rockbox/developer/mcuelenaere)
16:18:27 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
16:23:15 Join Seed [0] (
16:26:09 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
16:28:34soap13:33 < kugel> Our benches are meaningless anyway, since they aren't scientific who cares!
16:28:57soap - Huh? I'm not the only one who has taken great care to test only one variable at a time.
16:29:04 Join SndBstrd [0] (
16:29:08 Quit SndBstrd (Read error: 104 (Connection reset by peer))
16:29:13rashersoap: It was a quote
16:29:17rasherOf some troll
16:29:25gevaertssoap: that's a reference to a troll who claimed that at some point
16:29:31 Join sarixe` [0] (
16:29:37 Quit Seed (Nick collision from services.)
16:29:55 Join Seedy [0] (
16:30:18 Join Seed [0] (
16:30:25 Quit Seed (Remote closed the connection)
16:30:31 Quit Seedy (Remote closed the connection)
16:30:41 Join Seed [0] (
16:32:26Nico_Pgevaerts: what's a LUN?
16:32:33 Join robin0800 [0] (
16:33:02rasherSome SCSI communication point I believe
16:33:29gevaertsYes. It stands for Logical Unit Number. Maybe I should have called it "drive" in the commit message
16:35:45***Saving seen data "./dancer.seen"
16:35:50 Nick mcuelenaere_ is now known as mcuelenaere (n=mcuelena@rockbox/developer/mcuelenaere)
16:36:08gevaertsIf someone who's reasonably familiar with rockbox graphics code feels bored, have a look at FS #9969
16:36:50 Join markun [50] (n=markun@rockbox/developer/markun)
16:45:56 Quit sarixe (Read error: 110 (Connection timed out))
16:46:38 Quit robin0800 (Remote closed the connection)
16:49:02 Join robin0800 [0] (
16:49:31 Quit robin0800 (Remote closed the connection)
16:49:51 Join robin0800 [0] (
16:50:22 Quit robin0800 (Remote closed the connection)
16:51:16*gevaerts thinks he knows a way to do activity indicators as well
16:52:12rasherAnd on-the-fly transfer speeds?
16:53:14gevaertsmore or less, yes
16:53:44*Llorean wonders how many bug reports our transfer speed not matching what the host says will generate.
16:54:11*gevaerts would never show actual numbers
16:54:20 Quit avis ("maybe when we drown, the fish will be our friends")
16:54:35gevaertsOK, *you* can have numbers :)
16:54:38 Join avis [0] (n=ident@pdpc/supporter/student/avis)
16:57:22soapis anyone aware of an "offline" editor for Twiki tables? My google-fu has turned up nothing, and tables can quickly become a pain to edit.
16:57:55 Quit Aurix_Lexico ("Leaving.")
16:59:45 Join robin0800 [0] (
17:02:31 Join nibbler_ [0] (
17:02:47 Join Aurix_Lexico [0] (
17:08:16 Quit robin0800 (Remote closed the connection)
17:11:36 Join stoffel [0] (
17:13:44rasherMy e250 apparently fails to write the firmware image I copy over in recovery image, giving the error "main_image write failed". Does anyone have any ideas what might be wrong? (also posted here:
17:15:33LloreanHave you tried also copying the paired .rom image over at the same time?
17:16:36rasherI believe so. Not in this round of attempts though. Let me retry
17:21:39rasherWriting the bootloader fails as well
17:23:47soapLlorean: any thoughs on my proposed changes to the IpodAccessories wiki page as outlined in my discussion with Dave Woyciesjes in the User Mailing List?
17:25:00soapon thought 3A - perhaps "Audio In" could also be tracked under "Other" as only the "Recording" class of devices have this - and IMHO the table is getting overly huge as is.
17:25:31LloreanAudio Out always ought to be "yes" right?
17:25:55LloreanI mean mostly we're dealing with accessory protocol issues. Controls, etc?
17:26:02*Llorean hasn't looked at the page in ages.
17:26:07 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
17:26:24rasherWhat on earth has happened on that page. Why are there old tests in the new table?
17:26:30 Quit nibbler_ (Read error: 110 (Connection timed out))
17:26:46*gevaerts points rasher to FS #9969
17:27:35gevaertsI'd post a screenshot, but I have no idea how to make screenshots in rockbox while USB is connected
17:27:55rasherI expect it's not possible
17:28:10*gevaerts finds a bug
17:28:54mcuelenaereis it normal for the power thread to have a high stack usage? (like 94%)
17:32:29Lloreankugel: If I saw an option called "Recognize numbers while sorting" and I chose "false" I would expect it to then ignore numbers entirely.
17:33:05rasherIs the Sansa E200 recovery mode part of the bootloader? Or does it reside somewhere else? It seems odd that manufacturer mode is able to write the bootloader, but recovery mode isn't.
17:33:06Nico_Pgevaerts: isn't the USB supposed to disappear after the device is unmounted?
17:33:20Nico_Ps/USB/USB screen/
17:33:27gevaertsNico_P: no. Only after eject
17:34:01Nico_Pit still doesn't
17:34:31gevaertswhat OS and what target?
17:34:37Nico_Poh I need to use sudo for eject to actually work, sorry
17:35:46soapLlorean: Audio Out is not always a yes. Some accessories _appear_ to await two way communication before working at all.
17:36:13soaprasher: there are old tests in the new table becaue those items work and don't need reproven.
17:36:38rasherOh ah
17:37:25soapOld positives should equal current positives. Whereas old negatives might be current positives.
17:37:49rasherHadn't looked close enough to catch that. Makes sense
17:38:29 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
17:39:00 Quit kugel (Nick collision from services.)
17:39:02 Quit bmbl ("Woah!")
17:39:03Lloreangevaerts: Is the USB screen expected to go away with ejection on current builds? Or with something special enabled?
17:39:04 Join kugel [0] (n=kugel@rockbox/developer/kugel)
17:39:06 Quit ender` (" Cluster bombing from B-52s is very, very accurate -- the bombs always hit the ground.")
17:39:19gevaertsLlorean: current builds
17:39:37LloreanOkay, doesn't work for me in Windows at least.
17:39:56 Quit jaykay (Read error: 110 (Connection timed out))
17:40:06 Join toffe82 [0] (
17:43:12gevaertsThe fancy USB screen causes bus resets :(
17:43:27 Quit kadoban (Remote closed the connection)
17:43:36rasherAww. Ejecting works here (Linux)
17:43:56 Join kadoban [0] (
17:44:09LloreanDoes Windows handle ejecting differently? I'm sure I've seen devices become usable after a windows "eject".
17:44:13 Join Thundercloud [0] (n=thunderc@
17:44:19rasherNo idea, really.
17:44:40LloreanOh, I was testing with a Gigabeat S. Just thought I should ask, is this device-specific at all for the moment (besides requiring swusb)?
17:44:47 Quit Thundercloud (Remote closed the connection)
17:44:50gevaertsit shouldn't be
17:45:04 Join ender` [0] (
17:45:08 Join Thundercloud [0] (n=thunderc@
17:45:43*Llorean tests on a Nano
17:46:11 Quit Thundercloud (Remote closed the connection)
17:46:37 Join Thundercloud [0] (n=thunderc@
17:47:32 Quit Thundercloud (Remote closed the connection)
17:47:32LloreanHas anyone looked at the iPod charging patch?
17:47:52LambdaCalculus37Llorean: Which one is it again?
17:47:52gevaertsLlorean: how do you eject? "Safe remove" or "Eject"?
17:47:54 Join Thundercloud [0] (n=thunderc@
17:48:45Lloreangevaerts: Safely remove.
17:48:52 Quit Thundercloud (Remote closed the connection)
17:49:10gevaertsLlorean: can you see if "eject" (from the drive context menu) works?
17:49:16LloreanLambdaCalculus37: I can't find it. Maybe I'm crazy
17:49:18 Join Thundercloud [0] (n=thunderc@
17:49:34LambdaCalculus37Llorean: I'll go search.
17:49:42*gevaerts has to look what windoes does exactly on safely remove
17:49:46LloreanLambdaCalculus37: I just looked through all the still open tasks by him
17:50:21 Quit Thundercloud (Remote closed the connection)
17:50:32Lloreangevaerts: There's not an "Eject" option for HD targets.
17:50:38mcuelenaereanyone familiar with settings.c/h around?
17:50:45LloreanThey don't show up as "Removable Disk" so they only get "Safely Remove"
17:50:49 Join Thundercloud [0] (n=thunderc@
17:50:57gevaertsah yes. The beast is explicitely non-removable...
17:51:10LambdaCalculus37Llorean: I see FS #6891 (nano charger support), but that's it right now.
17:51:22Lloreangevaerts: Both "Eject" AND "Safely Remove" work for my Nano, same SVN revision
17:51:34LambdaCalculus37Ahh, FS #9295 also looks interesting.
17:51:48LambdaCalculus37"Charger detection for 2G iPod (and 1G?)"
17:52:10gevaertsLlorean: that actually makes some sense. Of course windows won't send an eject command to a non-removable drive...
17:52:19LloreanIt's a patch without its own task
17:52:35Lloreangevaerts: Is there some way you can tell it's been safely removed but not ejected?
17:52:56gevaertsnot sure. I need to test that
17:53:34LambdaCalculus37Our hero dreamlayers to the rescue once again! :)
17:54:41LloreanIt looks like it may need an option similar to the H300.
17:54:53LloreanOr has one from the description
17:55:50gevaertsrasher: manufacturer mode doesn't write the bootloader
17:56:01gevaertsAlso, have you tried sansa.fmt?
17:56:27rashergevaerts: Oh darn. So I guess the memory is broken?
17:56:32rasherI have. :\
17:59:14gevaertsrasher: maybe the partition table is just bad in a way that sansa.fmt can't handle. That's what happens on c200
17:59:21*gevaerts speculates wildly
17:59:30rasherI'm open to any suggestions
18:00:34gevaertsIf that's the case, an e250 port of the c250 eraser may help
18:00:56LambdaCalculus37gevaerts: I say go for it.
18:03:22rashergevaerts: I'd love that... I doubt it's a task for me, though
18:03:43 Join nibbler [0] (
18:03:44 Quit Thundercloud (Remote closed the connection)
18:10:33 Join fml [0] (n=4fd3c4ef@gateway/web/cgi-irc/
18:10:45fmlAny flyspray admins here?
18:11:23BigBambifml: To be given rights you want Bagder
18:11:32BigBambifml: congrats btw
18:12:36fmlBigBambi: thanks. This was frankly speaking a surprise
18:12:49fmlBagder: I want you!
18:13:22BigBambifml: We clearly *someone* thought it deserved :P
18:13:25Lloreangevaerts: So "Safely Remove" can fail to 'eject' an e200 as well, it seems.
18:13:28fmlDoes anybody know how the font is chosen for the default theme?
18:13:45fmlIt's too small IMHO for sansa e200
18:13:45gevaertsrasher: I *think* that if you change tools/configure line 1624 to say -DSANSA_E200 (instead of C) and then build 52\ne\n5\n it should work
18:13:57 Join rocko [0] (
18:14:02rashergevaerts: I'll try
18:14:21gevaertsrasher: you need the .bin file with e200tool, not the .mi4
18:14:37rashergevaerts: I trust it this is not really likely to screw things *more* up?
18:14:51soapfml, as I recall A font was chosen for the default theme, and then the size was determined by attemping to maintain the "look" of the WPS across targets as best possible.
18:14:55rasherconsidering things aren't working as it is
18:16:37gevaertsrasher: hm. e250 seems to be smaller than c250, so that partition table won't work...
18:18:54 Nick fxb is now known as fxb__ (
18:19:14gevaertsrasher: comment out all storage_write_sectors() calls in bootloader/main-c250wipe.c, and then try. It won't do anything at all, but you'll know if the approach works. Then, run a c240erase build (after the same configure changes of course), and try that. This will give you an e240 of course, but once you have that rockbox should work and you can use rockbox usb (or even the OF) for further work
18:19:46rashergevaerts: Sounds good.
18:20:08fmlsoap: was it discussed for each target individually or is there some rules (LCD W+H -> Font size)?
18:21:10soaphaha - I do not believe there was any science behind the choice. Let me see if the discussion is still in the WPS forum, or if it got "archived".
18:21:40 Quit kugel (Nick collision from services.)
18:21:45 Join kugel [0] (n=kugel@rockbox/developer/kugel)
18:21:58Lloreansoap: Shouldn't be in the archives. That's strictly limited to pre-move threads, I believe.
18:22:13fmlsoap: since a 12-font (the glyphs are even smaller) is hardly readable (for me) I use 19 on sansa e200
18:22:38LloreanFor what it's worth, I think the font size is about right. Maybe *slightly* larger, but 19 would be huge.
18:22:40soapso every bit of the (non-IRC) discussion should be in the (massive) "ONGOING DEFAULT THEME CONTEST!!!" thread?
18:22:48Lloreansoap: Yeah
18:23:17 Join saratogahome [0] (n=41becb3b@gateway/web/cgi-irc/
18:24:02rasherfml: Might I interest you in my theme (Wide Cabbie)?
18:24:15Lloreanfml: Don't forget that the default theme uses album art, and so the font size needs to be "big enough to read, but small enough to fit certain information even while AA is displayed"
18:24:44soapok, seaching simply for "font" and only on that thread takes you down to 7 pages of threads. That removes 90% of the entries.
18:24:48fmlrasher: what am I supposed to do? I can look at it, but I myself use a very simple, text only WPS
18:24:54 Join ender [0] (
18:25:26rasherfml: Ah, I thought you were looking for a "pretty much cabbiev2, but larger font!" theme
18:25:27fml Llorean: that's a valid point, yes. I never use AA.
18:25:33 Join jaykay [0] (
18:25:44*rasher thinks the font in cabbiv2 is tiny
18:25:59*kugel welcomes fml :)
18:26:02*BigBambi thinks it is too (especially in gigabeat f/s)
18:26:18BigBambiIn fact, it was bigger in gigabeat F/S, then Nico_P changed it
18:27:12kugelsaratogahome: Doesn't surprise me that 4.3.2 is slower too
18:27:17*Llorean thinks that in both those cases, they're at what he would consider "the smallest size I don't yet consider too small"
18:27:26fmlrasher: I just saw the font and tried to switch the languages. It lacks many areas of the unicode table. So a new user might be "catched up". And then I also thought it was too small.
18:27:35saratogahomekugel: yeah
18:27:36fmlkugel: thanks
18:27:59saratogahomei think a lot of versions of cabbie with different font sizes were uploaded to the wiki, so you could probably dig out who proposed what sizes
18:28:03BigBambiLlorean: I use my own version of cabbie v2 anyway with a bigger font, as I do consider them too small :)
18:28:27saratogahomebut basically everyone just sort of argued about it and eventually came to agree, and then i took those fonts and hard coded them
18:28:34kugelfor me, the cabbiev2 font is perfect
18:28:46BigBambikugel: On which player?
18:29:19fmlIt depends on the screen size. And on the eyes. And on the taste of course. I couldn't live with it.
18:30:05rasherI was having a go at adding a "LANG_CHARACTERS" string to the language files, that the font/language loader could use to check if the font and languages match
18:30:47soapIMHO the problem is worse on those portrait screens. Wrong orientation for displaying info, IMHO. Large fonts force scrolling too easily.
18:31:02kugelBigBambi: e200, fuze (same font)
18:31:05fmlCould we agree on a measure? I.e. x height should be no less that ... mm (real millimeters) on target? I heard that in printed docs, font size under 8pt is considered too small. How big is it in mm?
18:31:15BigBambikugel: I still think it is slightly small on that, but not as bad as on the gigabeat F/S
18:31:23kugeleven on the clip it's perfect, as it matches well with the dual-color screen
18:31:25 Quit HackyKid (Read error: 110 (Connection timed out))
18:31:31 Join HackyKid [0] (n=HackyKid@
18:32:01BigBambisoap: yes, portrait is worse. Personally I prefer scrolling over too (IMO) small fonts, but as with so many things it is a matter of taste
18:32:16Lloreanfml: That might be unrealistic. On targets like the Clip with *very* tiny screens, a smaller font size is much more acceptable, even expected.
18:32:23LloreanIt needs to be chosen on a target-by-target basis.
18:32:58soapfml: If all targets had the same pixel density perhaps a hard rule could be made, but forcing low density and low resolution screens to be considered with the same "rule" as high-density high-resolution screens (not to mention orientation) can't work.
18:33:08kugelI prefer more items on the screens over non-scrolling (i.e. <3 portrait for lists)
18:33:10saratogahomethe optimal size for print doesn't really apply here since smaller screens are held closer to the face, and thus the characters subtend a larger angle to the viewer
18:33:19fmlLlorean: yes. But just as an objective meause. Just for "orientation"
18:34:30soapThe design goal of the default theme would need to be changed. The current default theme was designed to appear as consistent across targets as possible.
18:35:21soapPerhaps a new (default?) theme could be worked out favors maximizing the capabilities of each particular screen over attempts to maintain a "universal" feel.
18:35:35soap*which favors
18:35:47***Saving seen data "./dancer.seen"
18:36:00rashergevaerts: e200tool recover bootloader.bin? (with a e250erase build, commented out all storage_write_sectors in bootloader/main-c250wipe.c?
18:36:31gevaertsrasher: yes. Maybe with e200tool init first. That's not needed on c200, but I don't know about e200
18:36:37Lloreanfml: As I said, objective measure doesn't mean anything.
18:36:50rashergevaerts: I don't think it's needed on e200 either
18:37:04kugelbut it can't hurt. I had more success with it
18:37:18 Quit ender (" In politics, absurdity is not a handicap. -- Napoleon Bonaparte")
18:37:26Lloreansoap: A fair idea.
18:37:43rashergevaerts: It's counting down .. seems to have stopped at "256 kB left"?
18:38:47rashergevaerts: It also skipped past "Press any key to shutdown", then appeared to write some more. Without me having to press anything?
18:39:32gevaertsrasher: it shouldn't do that, Are you sure it isn't due to the not very advanced screen layout I used?
18:39:49rasherAh right, probably
18:39:57rasherOh yes, I see it now.
18:40:13rasherThat appear to have "worked"
18:41:32rasherSo now I do the same with 240?
18:41:51 Quit ender` (Read error: 110 (Connection timed out))
18:42:52gevaertsI'd try that yes
18:44:17rasherHere we go
18:46:33 Join flydutch [0] (
18:47:46 Join Zoxc [0] (
18:52:50fmlHow are the fonts built? I.e. when (in the make process)? And how are the errors/warnings processed?
18:56:02fmlRe FS #9931: am I right that the font size in RB is the number of pixels between two lines (between their base lines)? And that the size doesn't necesserily tells anything about the height of the glyphs (e.g. x height)?
18:57:41 Quit Darksair ("Use the Force, Luke!")
18:59:44 Quit markun (Remote closed the connection)
19:03:46 Join ender` [0] (
19:08:13 Quit fml ("CGI:IRC 0.5.9 (2006/06/06)")
19:10:24 Quit BlakeJohnson86 ("Leaving.")
19:10:48 Join BlakeJohnson86 [0] (
19:16:07 Join SndBstrd [0] (
19:19:16 Quit jaykay (Read error: 110 (Connection timed out))
19:21:17 Quit saratogahome ("CGI:IRC (EOF)")
19:22:28 Quit ender` (Read error: 104 (Connection reset by peer))
19:22:37 Join ender` [0] (
19:24:49 Join webguest83 [0] (n=d8e8955c@gateway/web/cgi-irc/
19:25:49 Quit webguest83 (Client Quit)
19:36:00 Join nuonguy [0] (
19:37:09 Part SndBstrd
19:44:19 Part toffe82
19:47:48 Join stripwax [0] (
19:48:07 Join z35 [0] (
19:59:43 Quit midijunkie (Read error: 104 (Connection reset by peer))
20:03:33 Quit nibbler (Read error: 60 (Operation timed out))
20:05:22 Quit stripwax ("")
20:05:58 Join __lifeless [0] (n=lifeless@
20:09:12 Quit SirFunk__ (Read error: 60 (Operation timed out))
20:13:05 Join SirFunk__ [0] (
20:14:49 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:20:24 Quit stoffel (Read error: 113 (No route to host))
20:20:39 Quit _lifeless (Read error: 110 (Connection timed out))
20:25:54 Quit kugel (Read error: 104 (Connection reset by peer))
20:26:00 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:28:37 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
20:30:50Lloreangevaerts: Did you have any thought about the e200 note regaining control when "Safely Removed" only "Ejected" (while Nanos regain control on both)?
20:31:42gevaertsLlorean: not really. Was this with a microsd card inserted?
20:33:10LloreanIf I eject the internal memory drive, I'm back in the Rockbox menu (even though the "removable drive" still shows in My Computer just without a 'disk' in it). If I safely remove the device while leaving the cable connected, it stays in USB mode, even though both drives are gone from My Computer
20:35:50***Saving seen data "./dancer.seen"
20:45:05Nico_Pgevaerts: RoLo failed to load a firmware file I had just transferred (without the status indicator patch), is this suspicious?
20:46:44gevaertsNico_P: it's certainly not good. Does it fail again after rebooting?
20:47:06Nico_Pgevaerts: I didn't try and it works OK now :(
20:47:19Nico_PI mean I sent another firmware file
20:47:49kugelhow are the odds for 5 title to fit exactly into the audio buffer? i.e. data_rem: 0
20:48:07kugelthis stupid boost bug appears again.
20:48:29gevaertsNico_P: in that case I don't really have ideas...
20:49:50gevaertsLlorean: I can reproduce this (Windows safely remove not leaving rockbox USB mode). I need to trace this, but unfortunately the tracer software runs on windows, using the same PC to trace and be traced doesn't work well, and I don't have a second windows machine available right now...
20:50:00gevaertsI'll look at it as soon as I can
20:54:58 Quit sarixe` ("Leaving")
20:55:16kugelNico_P: would it be a good idea to always set filling to something != STATE_FILLING (so that boosting is canceled), no matter of the actual error code?
20:55:54 Join sarixe [0] (
20:55:58Nico_Pkugel: if the error causes the buffer to stop filling, then yes
20:56:16kugelit's weird. the boosting goes away again if I do it this way at metadata buffering
20:56:38kugeleven though bufopen for metadata can only return ERR_BUFFER_FULL, or a valid id
20:57:38kugeland I can't use logf, since that makes the boosting go away too (less audio buffer)
20:58:08 Quit avis (Remote closed the connection)
20:58:15kugelwell, I don't know if a failed bufopen should cause buffering to stop in any event
21:01:14kugelso, it actually shouldn't matter if I do == ERR[..], or < 0, for metadata buffering; but it apparently does (/me doesn't quite understand
21:01:50kugelbut I'm going to commit that anyway, this was before me there anyway, and I changed it to ==
21:02:55LambdaCalculus37Does anyone have any objections to FS #8778 being committed? According to tests done, it seems to fix the scroll wheel problem on 1G iPods, and it doesn't break anything on 2G iPods or unaffected 1G iPods.
21:03:42gevaertsI'm in favour
21:04:50kugelLambdaCalculus37: I think amiconn had objections
21:05:15LambdaCalculus37kugel: Do you remember what he had said about the patch?
21:06:09LambdaCalculus37I've tried it on my 2G iPod last night with r20137 and have seen no problems with it at all.
21:06:11kugelNico_P: seems the boosting goes away solely by slight changes of the binsize
21:06:27Nico_Puh, that's not very nice :(
21:06:59kugelseems there's a very tiny window where add_handle doesn't work as expected
21:07:15kugelprobably only if data_rem == 0 too
21:07:38Unhelpfulok, here's how FS #9919 is fetching album titles... am i doing this, is there a faster way, or do i need to go back to caching the title when i build the album index in the first place?
21:07:41 Join avis [0] (n=ident@pdpc/supporter/student/avis)
21:07:44LambdaCalculus37amiconn: Ping
21:08:30 Quit moos ("Rockbox rules the DAP world")
21:08:31 Join Keripo [0] (
21:09:06KeripoHello, I was wondering if Rockbox will be participating this year in Google Summer of Code 2009
21:09:20Nico_PKeripo: we'll be applying
21:09:31kugelLambdaCalculus37: No, would be easy to find if he would bother to raise his opinion on the tracker too. But it should be somewhere in the logs
21:09:42LambdaCalculus37Keripo: Watch the mailing list and the front page for the news.
21:09:47Unhelpfuldoing this *wrong*, that is :/
21:09:53LambdaCalculus37kugel: I'll check the logs out.
21:10:22KeripoNico_P: Same/similar projects as last years? I'm interested in porting Rockbox to the PSP platform or possibly just as a generic, portable app
21:10:57BagderKeripo: the projects are always up to the students to suggest
21:11:10Bagderwe as projects only provide ideas
21:11:57Bagderbut of course, the best is if done in communication student <=> project
21:12:00*Bagder runs off
21:12:05KeripoAhh, that's good. Always thought it was project-decided. Looking forward to it then ^_^
21:12:25gevaertsIt is, in a way. The project decides which proposals get accepted
21:12:41kugelah, since USB, developing is so much better
21:12:59gevaertskugel: for you, maybe ;)
21:13:00kugelno need to boot into the OF endlessly. just drop the binary and rolo it
21:13:09kugelI love it
21:13:26bluebrotherthe binary?
21:15:02Horschtisn't rebooting from OF to rockbox the more tedious reboot?
21:15:55UnhelpfulHorscht: much more so, on my e200... also, RB seems to get storage available much more quickly than OF
21:16:55kugelnot to forget that stupid database refresh
21:17:42*kugel does the USB dance once more _o| \o/ |o_
21:21:00LambdaCalculus37kugel: I see what amiconn said, but I want to speak to him about it directly as well.
21:21:24LambdaCalculus37I think the patch is very useful, and will finally allow people to use their 1G iPods without worry.
21:21:53kugelthough, I think yapper has some good point
21:22:11 Part Keripo
21:23:04LambdaCalculus37I do as well. However, we also don't have very many developers or staff that own a 1G iPod, so I don't know if anyone's tested this patch on the 1G iPod itself.
21:24:22 Part Llorean
21:29:36 Join Llorean [0] (
21:40:50 Join kubiix [0] (n=511bc91c@gateway/web/cgi-irc/
21:41:33 Quit ender` (" To err is human; to really foul things up takes a computer.")
21:44:30 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
21:48:08 Join stoffel [0] (
21:53:43 Nick miepchen^schla is now known as miepchen^schlaf (
22:00:19 Join FlynDice [0] (
22:01:08 Quit miepchen^schlaf ()
22:01:23 Join bandan [0] (
22:02:43 Part bandan
22:03:18 Quit kugel (Remote closed the connection)
22:04:59 Join miepchen^schlaf [0] (
22:06:10kubiixanyone can help me understad working with gpio states ?
22:07:38 Part Llorean
22:11:39 Join Llorean [0] (
22:15:35 Join miepchen^schla [0] (
22:21:22saratogakubiix: whats your question?
22:25:31kubiixsaratoga: i know that on ipod photo connecting the charger causes GPIOD change from 50 to 58, how can I handle this in charging_state() in ipod-power.c
22:26:12 Quit rocko (Remote closed the connection)
22:26:45kubiixthere is used only GPIOB, but it looks it is not used for it on iPod Color/Photo
22:32:32 Quit miepchen^schlaf (Read error: 113 (No route to host))
22:33:43 Join ender` [0] (
22:35:52***Saving seen data "./dancer.seen"
22:38:52saratogakubiix: does it work if you check GPIOD?
22:44:31kubiixsaratoga: thinking about it, I am not sure now that the problem is in charging_state(), because the all the powermanagement in last builds is weird, i can charge iPod only from OF, the battery indicator reports always empty battery and the estimated runtime is arount 180 mins
22:46:06saratogakubiix: is the problem that it doesn't charge or that you don't get accurate battery voltage readings?
22:47:25 Join rocko [0] (
22:48:08kubiixsaratoga: i am sure it is not charging
22:50:03 Quit HackyKid (Read error: 104 (Connection reset by peer))
22:50:07 Join HackyKid [0] (n=HackyKid@
22:59:42 Quit LambdaCalculus37 ("Fwump")
22:59:48amiconngevaerts: The problem is that the current method stays on the safe side by checking for a 1st gen, and only enabling the wheel power if it finds one
23:00:04 Join fml [0] (n=4fd3c4ef@gateway/web/cgi-irc/
23:00:19fmlWhere is the file with manual credits?
23:00:24amiconnIf we change it the way the patch does, and a 2nd Gen will be misidentified, I don't know what happens when power is applied to that pin
23:00:44bluebrotherfml: docs/CREDITS-MANUAL
23:01:03amiconnAnd the reason why there aren't two builds but just one is that the signatures are identical, so that ipodpatcher/rbutil cannot distinguish them at all
23:01:16amiconnApple also uses a single firmware for both
23:01:24bluebrotheror are you referring to the credits section? That gets generated from CREDITS
23:02:24 Quit rocko ("Leaving")
23:02:28rasherCREDITS-MANUAL isn't actually used for anything, is it?
23:02:39amiconnIt would be possible to check the apple SysInfo structure in RAM at boot, but I'd rather not rely on apple magic here
23:04:24amiconnImo the proper method would be to analyze what checks the apple ROM is doing in order to build the SysInfo struct. That requires some RE, preferably using a ROM dump from such a misidentifed 1st Geb
23:04:55amiconnI guess that updating the OF on a problematic 1st Gen would fix the problem as well, but I don't think we should require that
23:05:16saratogathat actually sounds like a pretty reasonable thing to ask people to try
23:05:22 Quit stoffel ("leaving")
23:06:29fmlbluebrother: thanks
23:06:59amiconnsaratoga: It is certainly worth a test, but imo it shouldn't be the final solution
23:07:25amiconnAnd once you upgraded, you cannot downgrade again to test other solutions...
23:07:40Lloreanamiconn: Under what conditions could a 2nd gen be misidentified?
23:07:57fmlNow that more people are here I'd like to repeat my question: How are the fonts built? I.e. when (in the make process)? And how are the errors/warnings processed?
23:08:20 Quit FlynDice ("Gotta go pay the bills!")
23:08:23amiconnPerhaps there are oldish ROMs for 2nd Gens as well, which also do not have the SysI info in ROM
23:08:28 Join FlynDice [0] (
23:08:41Lloreanfml: The best way to ask questions that only one or two people may know a good answer for is to do it on the -dev mailing list. It also helps if you explain why you're asking, so people know exactly what parts you're most interested in.
23:09:16Lloreanamiconn: Does Apple depend on the SysI info, or do they do some sort of hardware detection we don't know about?
23:09:17*amiconn would perhaps be able to answer, but doesn't understand fml's question
23:10:01amiconn[23:04:23] <amiconn> Imo the proper method would be to analyze what checks the apple ROM is doing in order to build the SysInfo struct. That requires some RE...
23:10:11amiconnIow, I don't know yet
23:10:21LloreanHm, I even read through the backlog and somehow I missed that line.
23:10:32*Llorean needs to check his eyes it seems.
23:10:44fmlamiconn: it's about FS #9931. I think convbdf should print errors (or at least warnings) if a font is "bad" (whatever that is). Will it break the build process?
23:11:06Lloreanamiconn: Could you hazard a guess as to what the worst negative effect of misidentifying a 2G would be?
23:11:40amiconnIirc, the 1st Gen wheel power pin is the same as the 2nd Gen headphone detection
23:12:08fmlAnd what is better: produce a font with some glyphs displaed incorrectly or to refuse to process the font?
23:12:08amiconnDepending on how headphone detection works, switching that GPIO pin to output and high could mean short-circuiting that pin
23:19:26*amiconn would simply close #9931
23:19:41amiconnIt's not really our bug - it's a bug in the imported fonts
23:20:08amiconnWe could drop them, if we deem the remaining fonts enough
23:21:41 Quit cbecht (Remote closed the connection)
23:22:29 Quit kubiix ("CGI:IRC")
23:22:38 Join kubiix [0] (n=511bc91c@gateway/web/cgi-irc/
23:22:44LloreanWell, it's not like we haven't modified fonts before (adding characters and such).
23:22:54amiconnThe more interesting question is why X has them this way...
23:23:09LloreanMaybe x handles that better than we do?
23:23:36LloreanDo they have the same sort of graphical glitches when used elsewhere, or is it something to do with how we render them?
23:23:49amiconn'better' in what way? These are bitmap fonts, and some glyphs don't fit the bounding box
23:25:15LloreanBut from the screenshots it looks like we don't simply clip them at the bounding box edge.
23:27:51amiconnThere might be a bug in convbdf then.
23:28:31*amiconn doesn't fully understand those screenshots
23:30:01 Join podliy [0] (i=podliy@
23:31:28amiconnImo convbdf should simply clip if a character exceeds the bounding box
23:32:13amiconnChanging the actual font heights might be problematic if themes use them, and if we adjust for individual characters, adding characters may change the font height
23:32:58amiconnIt may print a warning, but it certainly shouldn't stop
23:34:47amiconnFor some fonts we could change the height if the clipping is too heavy, but only if those fonts aren't used in themes yet.
23:35:24LloreanOkay, I'm confused. What are the font ascent and descent exactly? Is that something of ours?
23:36:08amiconnNo, those values come from the BDF. Ascent is the size above the font's base line, descent is the size below it
23:36:33amiconnThe font height is the sum of both, and rockbox doesn't care about the baseline, only about the total height
23:36:49LloreanHm, I need to find something else to read then, the doc I was reading about BDF didn't mention it.
23:37:36amiconnActually those measured are used in other font formats as well
23:37:57LloreanThat's what I'd assumed they were, I was just looking for a specific description of them in regard to BDF.
23:38:16 Quit Horscht ("Verlassend")
23:38:46LloreanWhat I'm reading now says "ascent" and "descent" are used for spacing, not bounding.
23:38:50amiconnI think the proper way of handling those things in a more sophisticated font rendering engine would be to make such glyphs actually overlap the previous/next line, but rockbox has no provision for that
23:39:04Lloreanit sounds like a character can extend outside of the ascent/descent.
23:39:19LloreanDoesn't BDF also define a bounding box?
23:39:47 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
23:40:18Lloreanamiconn: So we should clip those glyphs now (fixing whatever bug causes them to wrap or whatever it is that's shown in the _1 images) and suggest that the proper fix is allowing the glyphs to overlap?
23:41:54amiconnAllowing the glyphs to overlap would introduce a ton of problems wrt viewports, scrolling text etc. What height should a viewport for a single line of text have?
23:42:23LloreanViewports can be less height than a line of text, currently.
23:42:29LloreanI don't see a problem with it, they handle their own clipping.
23:42:44amiconnHow does text scroll when it overlaps other lines?
23:42:44fmlamiconn: yes, clipping might be the best we can do. Each glyph is a mini viewport!
23:43:29 Nick blkhawk- is now known as blkhawk (
23:43:40fmlamiconn: I suppose a remnant would stay on the previous line
23:44:10fmlIIUC, now the whole rendering in RB assumes that lines do not overlap
23:44:14LloreanSo maybe it's just not worth allowing overlap.
23:44:36LloreanSince even if we did, it'd basically be restricted to nonscrolling lines.
23:44:50amiconnfml: Yes it does, and it also assumes that glyphs don't overlap
23:45:03amiconnThis is why combining marks don't work
23:45:43 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
23:46:38fmlamiconn: ok. So would you agree with clipping and printing a warning? Would they be catched by the build script?
23:50:49 Join fyrestorm [0] (
23:53:28 Quit kubiix ("CGI:IRC")

Previous day | Next day