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 2009-03-23

00:02:15 Join robin0800 [0] (
00:02:30pixelmaI wonder where diff takes the "\no newline at the end of the file" right before the last phrase...
00:06:48bluebrotherpixelma: that gets displayed if the file doesn't end with an empty line
00:07:09bluebrothergcc doesn't like files without an empty line at the end. I guess that's the reason for patch complaining about it
00:07:23pixelmabut it does now (it didn't before) and usually it is displayed at the end of the file but this time it's not
00:07:58bluebrotherurgh? Where did that happen?
00:08:11 Join StealthyXIIGer [0] (n=stealthy@
00:08:30pixelmasee the unidiff of the first of my lang commits today
00:09:00 Quit robin0800 (Remote closed the connection)
00:10:09rasherpixelma: I guess that's diff complaining that the *old* file didn't end with a newline then
00:10:27bluebrotherlooks to me that this was the old file
00:10:31*bluebrother too slow
00:10:43JdGordonnew version of 9795 is up and waiting for testers
00:11:27pixelmaok, I hope that's the only reason because I'll commit to the release branch too, just seems weird to me
00:12:03rasherpixelma: it's not a problem either way
00:13:16 Quit bluebrother ("tomorrow ...")
00:14:25 Join stripwax [0] (
00:15:59pixelmaunless you see it in a Rockbox menu somwhere... ;)
00:16:08pixelmasomewhere too
00:17:54rasherNo, it's really just diff output
00:22:43 Quit Chris_Black ("off")
00:23:06kugelany remote experts here?
00:27:12 Quit saratoga2 ("CGI:IRC (Ping timeout)")
00:32:56 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:32:56 Quit pixelma (Nick collision from services.)
00:33:00 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:33:01 Quit amiconn (Nick collision from services.)
00:33:09 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:33:18 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:36:22 Join ultrasonic [0] (n=manohar@
00:37:55 Quit flydutch ("/* empty */")
00:39:59 Quit tessarakt ("Client exiting")
00:41:05 Quit shotofadds ("Leaving")
00:41:27 Quit saratoga_chrome ("CGI:IRC (EOF)")
00:46:21 Quit ender` (" If the experiment works, you must be using the wrong equipment. -- Patrick's Theorem")
00:49:36 Quit n1s ("Lämnar")
00:49:40 Part CPrompt^
00:49:54JdGordonkugel: can 9817 be closed?
00:52:43kugelJdGordon: yea, looks like
00:55:15 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
00:55:40kugelamiconn: ping
00:57:03JdGordonhey Nico_P, see the patch update?
00:57:11Nico_Pjust got the e-mail
00:57:40JdGordonrighto, good thing you mentioned that comment... looks outdated so i removed it
00:58:16Nico_Pthere was another comment I wanted to make but right now I forget what it was
00:58:28Nico_Pmaybe it'll come back as I look at the new patch
01:02:30Nico_PJdGordon: oh I remember... it was about audio_next_track(). I don't recall how often it's called but the common case seems to include the bufread() call. what if the id3 data is already there?
01:04:11JdGordonif its already there it gets read into othertrack_id3 and that is returned
01:04:46JdGordonoh, misunderstood you
01:04:56JdGordonyou mean if its already read into othertack_id3?
01:05:35Nico_Pyes, maybe the memcpy that is done by bufread could be spared
01:06:05JdGordonwhat happens if you call bufread mulitple times? is that even safe?
01:06:15Nico_Pit should be
01:06:32 Part wincent ("Kopete 0.12.7 :")
01:06:51Nico_Pyes, it is
01:07:04Nico_Pbut I'm not sure how often audio_next_track is called, so maybe there's no waste here
01:07:39JdGordonaslong as bufread keeps the handle at the start it should be ok... the aim of this patch is to cut down how often audio_next_track() is called anyway
01:08:18JdGordoni dont think its called enough to worry about... checking now
01:08:59JdGordonyeah, hardly ever
01:09:07 Join PrettyColours [0] (
01:10:53 Quit MethoS- (Remote closed the connection)
01:11:14 Quit Thundercloud (Remote closed the connection)
01:13:11 Join MethoS- [0] (
01:18:05 Join QuickStart [0] (
01:18:17***Saving seen data "./dancer.seen"
01:18:59 Nick fxb is now known as fxb__ (
01:23:08 Join QuickStart- [0] (
01:26:21 Quit stripwax (Read error: 104 (Connection reset by peer))
01:27:32 Quit QuickStart- (Read error: 104 (Connection reset by peer))
01:29:19 Join stripwax [0] (
01:32:57Nico_PJdGordon: I'm putting a build with your patch on my beast for a train ride tomorrow, I should be able to give it a good test
01:33:56*kugel wonders when "if (!wps_state.id3 && !(audio_status() & AUDIO_STATUS_PLAY))" can ever get true
01:34:00JdGordonput a svn build on also just in case :)
01:34:16JdGordonkugel: it shouldnt be possible
01:34:26JdGordonwps_state.id3 should always be valid
01:34:35kugelyea, I thought so too
01:34:52JdGordonyeah, at least with my patch its impossible for it to be null
01:35:10JdGordonhwcodec it is possible though
01:35:58Nico_Pwps_state.id3 can be null when the wps first appears
01:36:10Nico_PJdGordon: I'll have other players with svn builds ;)
01:36:17kugelbut " !(audio_status() & AUDIO_STATUS_PLAY)" also means you have to be in the wps (else you couldn't pause or seek), and there's no wps without id3
01:36:23kugelor am I wrong here?
01:36:52JdGordonyeah, that sounds correct
01:37:09Nico_Pit does, yes
01:37:21kugelalso, I've never ever seen the splash which is within that if
01:37:49pixelmais a WPS with ID3 even if it doesn't display anything of it (thinking of test cases)?
01:37:56kugelI've removed that part in my little rework, but now I'm wondering how useless it really is
01:38:26JdGordonpixelma: even the track filename is in the id3 so no... everything fails if its NULL so it should never be possible
01:39:24pixelmaI already had WPS only show hold state (as I said for test cases only but it would be weird if it suddenly fails on those out of nowhere)
01:39:41Nico_PJdGordon: I've made the WPS to be out of sync with the playlist by skipping forward quickly
01:39:47kugelyea, every function that calls it has non-NULL id3 too
01:40:02JdGordonNico_P: svn or with my patch?
01:40:22Nico_Pcurrent and next track both showed next track info
01:40:32pixelmaseen that in SVN too
01:40:40 Join wincent [0] (n=wincent@
01:42:10JdGordonhow many do you need to skip ahad to see it?
01:42:21Nico_Psomething like 15
01:42:32JdGordonso not excatly a common case? :)
01:42:48Unhelpfulpixelma: should i deliberately mask the vertical scroll keys in the album list if they're not used for something else?
01:42:58JdGordonpixelma: you've seen what Nico_P saw? or your talking about the blank wps?
01:43:01Nico_Pbut I know the hard corner cases :)
01:44:26Nico_Phmm the ipod video seems to have gone back to being unresponsive after resuming on startup
01:45:04pixelmaJdGordon: what Nico_P saw, the confused next track info. What I see quite often is that something else is displayed after manipulating the already playing list (from WPS > file browser > insert some track), it usually fixes itself on next track change but not always - with SVN
01:45:34 Quit QuickStart (Read error: 110 (Connection timed out))
01:46:01*JdGordon needs to hook up a harddisk target
01:47:11Nico_Pspeaking of those, the WPS seems to lack updates when the disk is being spun up
01:47:33Nico_Phmm, maybe that's not new to the patch though
01:48:29pixelmaUnhelpful: I was confused a bit about it but also because I thought "up" at least would be used for something else, at least I think it should be masked if one of them is used. Can't think well atm and should *really* get some sleep...
01:48:49JdGordonI dont know why, but yeah I tihnk this patch screws up responsiveness a bit
01:48:55 Join timc`` [0] (n=aoeu@
01:48:57 Quit barrywardell ()
01:49:11Nico_PI'm sure it can be done without affecting responsiveness
01:50:00JdGordonhmm i seem to have lost next track info!
01:50:03Unhelpfulah, ok. i've pretty much decided to go with up as select for ondio in album list, i'll make sure to mask down there as well. i was going to use long menu for the app menu, and off for quit
01:50:32Unhelpfulhrm, maybe long off, in case somebody is used to the standard off->cancel mapping?
01:51:29pixelmawouldn't matter much too me (in theory) as a short "off" is used in some plugins too
01:51:31Nico_PJdGordon: the WPS should be updated on track skip
01:51:42Nico_Pin audio_skip()
01:52:20JdGordoninstead of?
01:52:35 Quit timc (Connection timed out)
01:52:36Nico_Pinstead of apparently not being updated
01:52:37 Quit miepchen^schlaf ()
01:53:33Nico_Pmost places where track_changed used to be set to true might need to be considered for WPS updates
01:55:39*kugel is curious about line 261+ in gwps-common.c
01:56:01 Join miepchen^schlaf [0] (
01:57:13JdGordonthe charcell check?
01:57:52kugeland the gui_wps_display() call
01:58:36JdGordonbah, crosstalk :p both your nicks are the same colour.. wondering wtf that line had to do with the playback patch :p
01:58:53kugelnothing, but still ;)
01:59:08kugelwhy would only charcell need a wps update?
01:59:21JdGordonscrew charcell :D
01:59:31 Join CaptainKewl [0] (
01:59:37kugelyea :D
02:00:22*kugel thinks that should be removed too
02:00:28JdGordonNico_P: depending on the call order... the wps should be updating as soon as the skip track is pressed
02:00:40JdGordonwhich might be useless if its too early or too late
02:04:05JdGordonremoving some wps updating.. checking if it makes it feel any slower/faster
02:05:46JdGordonwell... WPS updates fine on track change if the next track is buffered... otherwise it feels very slow
02:06:15Nico_Pyes, that's the case that matters IMHO
02:06:34JdGordonand worse... keep skipping and the track you stop at wont be the track that starts playing
02:08:02Nico_Pthere is a delay when the disk is spinning up to fetch the metadata for the unbuffered track the user skips to
02:08:43Nico_Pwhen dircache is on, it's possible to still show the filenames
02:09:33JdGordonis this a problem in svn also then?
02:10:19 Quit stripwax (Read error: 104 (Connection reset by peer))
02:11:22 Join bluefoxx_ [0] (
02:12:19 Quit moos ("Allez l'O.M !!!")
02:13:21kugelhm, 200bytes binsize save doesn't sound bad
02:13:42JdGordonarg, yes the patch is causing the wps to play wierd
02:13:49 Quit bluefoxx_ (Client Quit)
02:13:52JdGordonsvn works with and without dircache enabled
02:13:54 Join bluefoxx_ [0] (
02:14:52kugelJdGordon: I guess you don't have time to quickly test a patch? :)
02:16:24JdGordonput it up...
02:16:26Nico_PJdGordon: I need to go to bed, but I'll try to have a more thorough look at the patch tomorrow or in the week
02:16:35JdGordoncheers :)
02:17:06Nico_Pit might just be that a few more WPS updates are needed though
02:17:47JdGordonkugel: i got a clean svn... pass your patch....
02:17:55JdGordonquick before i repatch my patch :D
02:20:50JdGordonthat last part of the patch looks wrong
02:20:58JdGordonits going to refresh the first lcd's wps twice
02:21:16kugelwhich last part?
02:21:22 Join Neffi [0] (
02:21:44JdGordon@@ -702,8 +693,10 @@
02:21:49JdGordonthe last hunk of the pathc
02:22:03 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
02:22:10kugeluh, it shuld be gui_wps[i], iundeed
02:22:37NeffiI'm new to the whole project. I'm running an iPod 5.5 gen video, and I just installed RockBox on it. I'm curious, would FLAC playback be more battery-intensive than MP3 playback? \
02:22:41JdGordonafter this, whats the diff between update() and gui_wps_refresh() ?
02:22:46kugel&gui_wps[i], even
02:23:06JdGordonNeffi: the wiki has some battery benchmarks
02:23:13kugelJdGordon: update is gui_wps_refresh+track change handling
02:23:18JdGordonmp3 playback uses both cpu cores whereas flac doesnt
02:24:00kugelNeffi: flac is faster, but has bigger files, which means that the disk needs to spin more often
02:24:29NeffiSo does it about even out?
02:24:43 Quit Nico_P (Read error: 54 (Connection reset by peer))
02:26:43JdGordonkugel: is there any real reason to rush this?
02:26:59JdGordonit looks like its going to clash with my playback patch...
02:27:10NeffiAh, found the benchmarks
02:27:31NeffiThey aren't too thorough though. What would be the most efficient codec?
02:28:38kugelJdGordon: It seems your patch needs more work anyway, so why should I wait for it?
02:28:58kugeland I don't see why it should clash, other than a few failed hunks maybe
02:30:49JdGordonNeffi: there is no easy answer to that...
02:31:51 Join gartral [0] (
02:34:11JdGordonkugel: my patch completly changes those like to get it in as soon as its fixed up... i dont need extra work on it, unless there is actually a reason to do it
02:36:36kugelJdGordon: how does your patch handle cuesheets? It seems to me you could remove update() entirely, actually
02:37:05JdGordonthats pretty much whats being done
02:37:21JdGordoncuesheet is going to be removed from gpws once this is in
02:37:27kugelyea, that's the only part where our patches clash. and it's not more work anyway
02:43:08JdGordonyay for thread sync issues :/
02:48:47 Join bluefoxx__ [0] (
02:51:49 Join Strife89 [0] (n=nds@
02:53:31 Part wincent ("Kopete 0.12.7 :")
02:54:03 Quit Rob2223 ()
02:54:28gartralmy isp's dns is being retarded, whats the current rnumber?
02:56:10 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
02:59:59NeffiDoes anyone know of a good free FLAC->OGG converter?
03:01:50Unhelpfuloggenc :D
03:04:21 Quit bluefoxx_ (Read error: 110 (Connection timed out))
03:04:47 Join bluefoxx_ [0] (
03:06:05 Quit bluefoxx__ (Read error: 110 (Connection timed out))
03:07:49 Quit BHSPitLappy (Remote closed the connection)
03:10:54 Part Strife89
03:12:05 Join cspotcode [0] (
03:18:21***Saving seen data "./dancer.seen"
03:22:44 Join Rob2222 [0] (
03:23:24 Quit Neffi ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
03:24:42 Quit MethoS- (Remote closed the connection)
03:38:11 Part gartral
03:42:17JdGordonhow important is making sure the wps acts correctly when doing lots of quick track skips?
03:50:08*kugel spots a cue sheet bug
03:50:14 Quit perrikwp (" ajax IRC Client")
03:50:20 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
03:53:18 Quit ultrasonic (Read error: 110 (Connection timed out))
03:53:21JdGordonwhat sort of bug?
03:57:38JdGordonhmm... thats not really surprising... but yeah a bug :)
03:58:49kugelJdGordon: I'm going to commit the thing, I hope it won't be too bad for you
03:59:05JdGordonyou'll find out if it is... :D
04:04:31 Join hd [0] (
04:04:33 Quit hd (Read error: 104 (Connection reset by peer))
04:11:29 Quit agaffney ("reboot")
04:12:46*kugel expects at least 1 red
04:18:43 Join KevBurnsJr [0] (
04:23:37kugelJdGordon: be careful with svn up now :)
04:24:01 Quit Llorean (Read error: 104 (Connection reset by peer))
04:24:19 Join Llorean [0] (n=DarkkOne@
04:29:47 Join SonGoKu [0] (n=SonGoKu@
04:29:52SonGoKuhello all
04:35:47 Join wangxiang [0] (n=wangxian@
04:37:08 Join tvelocity [0] (
04:37:54 Quit PrettyColours ("Leaving")
04:43:40 Join wangxiang_ [0] (n=wangxian@
04:47:10 Join hd [0] (
04:47:13 Quit hd (Read error: 104 (Connection reset by peer))
04:56:19 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner 1.9/2009020407]")
04:59:01 Quit HBK ()
05:00:30 Quit CaptainKewl ("( :: NoNameScript 4.02 :: )")
05:02:17 Join agaffney [0] (
05:02:40 Quit wangxiang (Read error: 110 (Connection timed out))
05:08:08 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
05:08:30 Quit cspotcode ("Leaving.")
05:12:05 Quit ze (Read error: 113 (No route to host))
05:13:38 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
05:18:24***Saving seen data "./dancer.seen"
05:19:56 Quit tarbo (Read error: 60 (Operation timed out))
05:20:57doomwakeis there a setting ? to have it boot 1st sould have to edit the code and recompile the bootloader to do that
05:21:46doomwakeo o
05:22:56 Join ze [0] (i=ze@
05:24:56 Join saratoga [0] (n=9803c264@gateway/web/cgi-irc/
05:25:31 Join webguest08 [0] (n=deffc8b1@
05:31:25 Part orsonj
05:32:06 Quit webguest08 ("CGI:IRC")
05:35:02 Join tarbo [0] (n=me@unaffiliated/tarbo)
05:40:10 Quit Horschti ("Verlassend")
05:53:45 Join webguest19 [0] (n=4a2f9b1b@gateway/web/cgi-irc/
05:54:10webguest19iwas wondering if someone could help me with a conversion issue im haveing
05:55:24Lloreanwebguest19: What kind?
05:56:28webguest19winff just stop converting all types of files
05:56:39webguest19it wont do avi mpeg or mov
05:56:57webguest19i was wondering if its a divex thing or what
05:57:33webguest19or if there was any alternatives to winff
05:57:42webguest19btw i have ipod nano 1st gen
06:03:38 Join davidfg4 [0] (n=david@
06:04:22webguest19im downloading media coder now to see if it will resolve me issue
06:06:06 Quit agaffney (Read error: 110 (Connection timed out))
06:07:26LloreanSupport for other peoples' software should be taken to them.
06:07:36LloreanWe don't write or maintain WinFF.
06:08:13webguest19i was just wondering if you guys have run into the same problem
06:08:45LloreanIt's not really a question that belongs in here. This channel is for questions actually about Rockbox itself or its development
06:09:07webguest19ok well where does it belong
06:09:34LloreanWell, there's an off-topic channel mentioned in the channel topic.
06:09:43LloreanBut for support with WinFF you really should just go to WinFF's support locations
06:21:58saratogawho should i talk to about file browser questions?
06:23:33 Quit fyrestorm (Read error: 104 (Connection reset by peer))
06:25:30 Quit webguest19 ("CGI:IRC")
06:43:16 Quit saratoga ("CGI:IRC (EOF)")
06:46:07 Quit SonGoKu (Read error: 110 (Connection timed out))
06:54:10 Quit shodanX (Read error: 110 (Connection timed out))
07:02:55 Join SonGoKu [0] (n=SonGoKu@
07:07:23 Nick fxb__ is now known as fxb (
07:18:27***Saving seen data "./dancer.seen"
07:22:21 Quit BUMBACL0T (Connection timed out)
07:24:07 Join n1s [0] (n=n1s@rockbox/developer/n1s)
07:25:51 Join BUMBACL0T [0] (n=ORF@unaffiliated/bumbacl0t)
07:28:45 Join shodanX [0] (
07:29:23 Join kadoban [0] (
07:31:11 Quit StealthyXIIGer (Read error: 145 (Connection timed out))
07:31:16 Quit SonGoKu ()
07:39:04 Nick fxb is now known as fxb__ (
07:55:34 Quit midgey ()
08:09:54 Join Mathiasdm2 [0] (
08:15:57 Quit wangxiang_ ("Leaving")
08:24:53 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
08:29:33 Join casainho [0] (n=chatzill@
08:29:46casainhohello :-)
08:30:00 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
08:30:15casainhogevaerts: here is the patch for Lyre: :: I am waiting for any suggestions ;-)
08:34:02 Join Rob2223 [0] (
08:34:27 Quit faemir ("Leaving")
08:35:26 Join bertrik [0] (
08:36:02 Quit homielowe ()
08:36:29 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:36:33 Join wincent [0] (
08:38:41linuxstbcasainho: Some cosmetic comments - "//" shouldn't be used (use /* .... */) and no TAB characters.
08:39:41linuxstbAlso, some files have extra ^M characters on some lines (but not all lines).
08:39:42casainhohmmm... ok, // are easy to find but I don't know how I will identify the TABs on every file....
08:40:21casainho^M ?? whats that?
08:40:21linuxstbPlus lines shouldn't be longer than 80 characters.
08:40:35linuxstbASCII 13
08:40:57 Join MethoS- [0] (
08:40:57linuxstbi.e. the files have a mixture of DOS and Unix line endings (in the same file)
08:41:19casainhooh, I don't know how to do that of 80 characteres, how to identify, I am suingEclipse, I must find a way for it show me the 80 char
08:42:57linuxstbAlso,if the patch is just for a bootloader, you don't need to include apps/keymaps/keymap-lyre.c
08:43:30casainhohow do I do that line endings? is with just an ENTER key?
08:44:18casainhocan you please write all this on comments on patch tracker? (I didn't know that were so many things to rectify....)
08:44:28linuxstbI don't use Eclipse. In emacs, I would just do a search and replace to replace a "^M" character with nothing.
08:45:02linuxstbThe docs/CONTRIBUTING document describes the coding guidelines.
08:45:47casainhookok. Thanks. I must go for now.
08:45:49 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.7/2009030422]")
08:46:42 Quit kachna|lappy (Read error: 113 (No route to host))
08:47:28 Quit bertrik ("Leaving")
08:47:46 Join ender` [0] (
08:51:10 Quit Mathiasdm2 ("Yuuw!")
08:52:30 Quit Rob2222 (Read error: 110 (Connection timed out))
08:53:31 Join petur [50] (n=petur@rockbox/developer/petur)
08:54:13 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:54:37 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
09:09:25 Quit bmbl ("Woah!")
09:12:23 Join Thundercloud [0] (
09:12:58 Quit Seed ("cu, Andre")
09:18:29***Saving seen data "./dancer.seen"
09:20:39B4gderRockboxPlayer => TargetLyre ?
09:20:46B4gderor PlayerLyre?
09:20:53B4gderor... ?
09:25:06linuxstbB4gder: ?
09:25:10B4gderthe wiki page
09:25:17B4gderit's still there using the old name
09:25:23linuxstbAh... Maybe LyreProject
09:25:44linuxstbThat's how casainho has titled his menu in configure
09:25:57B4gderah then I use that
09:29:30B4gder... hm, then there's a whole range of related pages using the same prefix
09:29:57B4gderI could just rename it similarly on all of them
09:30:03 Join kachna|lappy [0] (
09:30:11Zagoryeah, just s/RockboxPlayer/Lyre
09:32:05linuxstbYes, the main page could be LyreProject, and do what Zagor said for the rest.
09:34:06B4gderperhaps could just be removed?
09:34:27B4gderit seems to only contain a datasheet to be used in a player they're not working on
09:34:42scorcheZagor: could you allow the webclient to join #rockbox-gsoc ?
09:35:05B4gderoh well, I'll rename it first and then we'll see
09:35:28 Join BigBambi_ [0] (i=86ceaf40@rockbox/staff/BigBambi)
09:36:26 Join ultrasonic [0] (n=manohar@
09:36:48B4gderrename complete
09:40:40 Quit MethoS- (Remote closed the connection)
09:47:26Zagorscorche: sure
09:51:16 Quit Thundercloud (Remote closed the connection)
09:53:10 Join Seed [0] (
10:17:45 Quit nibbler_ (Read error: 113 (No route to host))
10:21:15 Quit BHSPitMonkey (Remote closed the connection)
10:21:20 Quit scorche (Nick collision from services.)
10:22:07 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
10:38:57 Join einhirn [0] (
10:43:15 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
10:45:25 Quit KevBurnsJr (" coming soon")
11:11:12 Quit wincent (Read error: 110 (Connection timed out))
11:11:12 Join AnonCoward [0] (
11:11:58 Quit AnonCoward ("leaving")
11:15:58 Join axios [0] (n=axios@
11:16:50axiosHi, I have tried to find a new device that runs with rockbox. A totally new lot of gigs device, anyone have some sugestions?
11:18:30***Saving seen data "./dancer.seen"
11:18:37axiosOh sorry, didnt read topic.
11:19:12 Quit Lss (Read error: 104 (Connection reset by peer))
11:20:03 Part axios
11:26:53 Join kushalone [0] (n=kushal@
11:27:42B4gderoh mp3hd...
11:28:12B4gdercountdown started until the question comes
11:28:48 Quit linuxstb (Read error: 113 (No route to host))
11:30:02B4gder"An mp3HD file is simply a ".mp3" file including a standard mp3 file and additional information that is stored in the id3 tag (where other info on the track like artist name, song name, etc.. is also stored),"
11:30:17B4gderbut but... a regular mp3 can't be 900kbps, right?
11:30:32 Join GodEater [0] (n=yeahrigh@rockbox/staff/GodEater)
11:33:46B4gder,2817,2343448,00.asp has that quote btw
11:42:52 Join Lss [0] (
11:44:42 Quit kushalone ("This computer has gone to sleep. Macs are like babies. They need a lot of sleep. :)")
11:45:26 Quit Joe__ (Read error: 110 (Connection timed out))
11:45:52 Join Joe__ [0] (n=joe@
11:47:35 Quit Nico_P (Remote closed the connection)
11:49:23 Join wincent [0] (
11:56:47 Quit kadoban (
11:56:47 Quit davidfg4 (
11:56:47 Quit jhMikeS (
11:56:47 Quit Kopfgeldjaeger (
11:56:47 Quit tmzt (
11:56:47 Quit Xerion (
11:56:47 Quit jfc (
11:56:47 Quit chrippa (
11:56:47 Quit webmind (
11:56:47 Quit suom1 (
11:56:47 Quit __lifeless (
11:56:47 Quit Bombe (
11:56:47 Quit lostlogic (
11:56:47 Quit Bagder (
11:56:47 Quit dionoea (
11:56:47 Quit crashd (
11:56:47 Quit preglow (
11:56:47 Quit fxb__ (
11:57:30NJoinkadoban [0] (
11:57:30NJoindavidfg4 [0] (n=david@
11:57:30NJoinjhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
11:57:30NJoin__lifeless [0] (n=lifeless@
11:57:30NJoinBombe [0] (n=droden@freenet/developer/Bombe)
11:57:30NJoinKopfgeldjaeger [0] (
11:57:30NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
11:57:30NJointmzt [0] (
11:57:30NJoinXerion [0] (
11:57:30NJoinjfc [0] (
11:57:30NJoinBagder [241] (n=daniel@rockbox/developer/bagder)
11:57:30NJoindionoea [0] (
11:57:30NJoincrashd [0] (
11:57:30NJoinchrippa [0] (
11:57:30NJoinsuom1 [0] (
11:57:30NJoinwebmind [0] (
11:57:30NJoinfxb__ [0] (
11:57:30NJoinpreglow [0] (i=thomj@rockbox/developer/preglow)
12:02:22 Quit bubsy ("Mrrrrreow!")
12:10:42 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:15:17 Quit mc2739 ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
12:15:19 Join robin0800 [0] (
12:18:51 Quit robin0800 (Client Quit)
12:19:09 Join robin0800 [0] (
12:31:31 Join dfkt [0] (n=dfkt@unaffiliated/dfkt)
12:40:42 Join Ubuntuxer [0] (
12:45:38 Quit robin0800 (Read error: 110 (Connection timed out))
12:54:03 Quit BUMBACL0T (Read error: 60 (Operation timed out))
13:01:48 Join BUMBACL0T [0] (n=ORF@unaffiliated/bumbacl0t)
13:08:06 Join AndyIL [0] (i=AndyI@
13:18:32***Saving seen data "./dancer.seen"
13:20:48 Quit Ubuntuxer ("Leaving.")
13:21:15 Quit AndyI (Read error: 110 (Connection timed out))
13:24:25 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
13:34:47 Quit ultrasonic ("Leaving")
13:37:24 Join archivator [0] (n=archivat@
13:38:17 Join LambdaCalculus37 [0] (n=44a04303@rockbox/staff/LambdaCalculus37)
13:40:56 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
13:41:43 Join pyro_maniac [0] (
13:43:57 Join Casainho [0] (
13:45:15Casainhohello :-) −− regarding the commentaries here: ; what should be the line ending on the code?
13:46:18B4gderunix style
13:46:20LinusNCasainho: LF only, i.e UNIX endings
13:47:11Casainhookok, thanks. I am using Eclipse and there is an option to convert it for Windows, Unix ow Macos style.
13:47:44 Part LinusN
13:49:31linuxstbCasainho: The problem with your files are that they are a mixture of the two - i.e. in the same file some have DOS (Windows) line endings, others have Unix. So you or Eclipse has done something wrong.
13:50:39Casainhowell, I may used some code from other sources... −− my problem for now is to have a way for I see how are the lines ending
13:50:53archivatorEclipse also has an option to keep line endings as-is (are :)). If you select Unix though every file you edit and save will have the new style.
13:51:34Casainhoarchivator: do you know if there is a way for I see while editing the file, what are the ending character?
13:51:50Casainholinuxstb: and what you said: "Is any code in your patch (e.g. at91sam9260.h) taken from the Atmel SDK? If so, then you should include a header file from that project, including their copyright and license information." −−> you are correct ;-)
13:52:06archivatorHmm, not sure - try the paragraph symbol - the one that shows all whitespace.
13:54:14Casainhoarchivator: eheh, it works ;-)
13:54:24 Quit Joe__ (Read error: 145 (Connection timed out))
13:54:31 Join Joe__ [0] (n=joe@
13:54:47Casainhoarchivator: today I even learned how Eclipse show the line limiter for 80 characters ;-)
13:55:15archivatorThere is a Help menu. you know...
13:55:25B4gderand of course, all other line delimiters than "unix" are stupid :-)
13:55:52*gevaerts disagrees. All common line delimiters are stupid
13:56:48Casainhoarchivator: do you know if there is a way to visual identify the TABs on the line codes?
13:56:50archivatorgevaerts: not so. \n is explicitly a new line. \r is not. \n\r is reasonable but is outdated..
13:56:53gevaertsUsing printer control codes to indicate line endings is a bad idea!
13:57:09B4gderthat's only bad if you consider them printer control codes
13:57:15B4gdera single byte is sensible to use
13:57:28gevaertssure, but they should have chosen a different one
13:57:31B4gderand exactly which doesn't matter so the first one that was picked makes sense
13:57:42archivatorCasainho: That same mode should show tabs as an arrow and spaces as mid-line dots.
13:57:50 Join gartral [0] (
13:57:53 Part gartral
13:58:02 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
13:58:20gevaerts\036 is at least as reasonable as \012
13:58:28Casainhoarchivator: oh thanks!! Now I have all the tools needed for working on all questions about my patch ;-)
13:59:30CasainhoB4gder: about the LCD_DEPTH, does the Rockbox have an LCD driver for the 444?
13:59:53B4gderI spoke about the comment
14:00:04B4gderreading that code it sounds as if you have 64K colours
14:00:29B4gderI just think it perhaps could be wise to mention how things work in there in a comment
14:00:46 Join midgey [0] (n=tjross@
14:01:08B4gderCasainho: since you're the first using a 444 driver you'd be the first to write code for dealing with it internally, if you think that's a good idea
14:01:10CasainhoB4gder: ah, I understand now what you wanted to say ;-)
14:03:08CasainhoB4gder: I am fighting to get help for someone make a prototype for ourself developers, like 10 units for start, and we may change the LCD if possible.
14:05:21B4gderI think you'll get more people interested with a proper 16bit lcd that can show video if desired
14:11:48Casainhoyes, but no developer will build his own hardware
14:12:30CasainhoI think the best option for now is finding a partner for helping us, and selling to us, a few units/prototypes
14:13:10Casainholike this guy is selling his own boards on Ebay:
14:13:23Casainhoand his project page is this one:
14:15:10Casainhoand then we should try to use an better LCD, if possible −− if the partner have technology for soldering the LCD connector, etc
14:15:32Casainhodoes anyone have any idea of this? were we can get help?
14:16:00archivatorI asked this yesterday but I got conflicting answers - what's the best place to add a hook allowing plugins to read PCM data? Also, since the plugin callback would execute in a different thread context, is there a way to avoid mutexes? Further, is a callback the best way to pass data to the plugin (in way similar to how DSP functions work) or should the PCM side keep a separate plugin buffer and only send it when asked for it?
14:16:41LambdaCalculus37archivator: What are you planning on writing as a plugin, if you don't mind my asking?
14:16:49archivatorCasainho: who's doing your current PCBs ?
14:16:55archivatorLambdaCalculus37: a frequency spectrum analyzer
14:17:12LambdaCalculus37archivator: Oooh, nice! :)
14:17:51Casainhoarchivator: no one. We are using a development board + a custom board, which I have just made one unit for myself at my work, I can't do anymore :-(
14:20:16 Join pyro_maniac1 [0] (
14:21:12archivatorCasainho: technically, you *could* etch the boards yourself with a laser printer and some cheap chemicals but the board would be quite large - no way to do multiple layers either. :) You're in Portugal, right? And there isn't a single shop that does PCBs? That's kinda hard to imagine - here in Bulgaria I know of at least 6 places that do that. And that's just one city =/
14:21:43Casainhohey!! Olimex is from Bulgaria!! :-)
14:22:38Casainhowe need someone that design the board and have experience with some shop of PCBs
14:22:53 Quit pyro_maniac (Read error: 60 (Operation timed out))
14:23:05Casainhobut there is also the need for someone solder the components
14:23:27CasainhoI would prefer to pay for someone else do all that....
14:24:14archivatorOne of the places I've worked has automated component soldering - it's quite expensive on a small scale, though.
14:24:30archivator*worked with
14:25:48Casainhooh, I work on a company like that....
14:26:01Casainhobut that is no solution for our actual situation...
14:26:41CasainhoI think I will get more help from smaller companies / individuals...
14:28:11archivatorWell, there are these guys here, they do PCBs with a solder mark for 2 euro/dm2 - it's not impressive but does work for my side projects. I bet there is something like that where you live..
14:29:06webmindyay, another rockbox device \o/ (patched an ipod mini)
14:29:12webmindand good afternoon
14:30:19Casainhoarchivator: but they just do the PCB? or they also assembly?
14:31:06 Quit kachna|lappy (Read error: 110 (Connection timed out))
14:31:14webmindI know makes cheap pcb's
14:31:18webmindbut no assembly afaik
14:31:31webmindand it might take a while
14:31:58archivatorCasainho: PCBs only. No one would do SMD (or any other kind actually) assembly for fewer than 100 PCBs..
14:32:29Casainhowell, there is some guys that do oven reflow at home...
14:32:42webmindsmall oven works
14:33:26Casainhoif someone knows any potential shop/individual that would sell to us that service, please say ;-)
14:33:33webmindCasainho, you're trying to get a prototype for a media player build?
14:33:57archivatorCasainho: If I were you, I'd find an engineer who's done SMD assembly (has the tools and skills for the job) and hire him outside of his usual job.. Sounds reasonable to me..
14:34:03Casainhowebmind: yes, I am, for this one:
14:34:14 Quit jordoex (Read error: 60 (Operation timed out))
14:34:43webmindCasainho, nice
14:34:48Casainhowell, I have a guy here at work that do SMD soldering. but is not all the solution...
14:34:53webmindCasainho, have you tried asking arround hackerspaces?
14:35:07Casainhowebmind: no, I don't know what is even that.
14:35:49webmindCasainho, ah
14:35:51CasainhoI am a bit lost in what we should do, the best path...
14:36:07webmindCasainho, spaces where people do 'hacking' in the broad sense of the word
14:36:18webmindanything from knitting to soldering to coding
14:36:50webmindand automated fabrication seems to be all the hype
14:37:00webmindso maybe someone has made something for smd :)
14:38:14Casainhooh, I were now seeing the list here in Portugal −− I recognized 3 places of 4, I already meet some guys and I even did a presentation to Rockbox and Lyre in one of that places, recently
14:38:30webmindcool :)
14:38:43Casainhohere is more people arrount Linux, etc −− very few hardware people :-(
14:39:25CasainhoI found one guy/shop that works a lot with Arduino, but they do very simple hardware, much more simple that I can do... :-(
14:39:45 Join parafin [0] (
14:40:18webmindCasainho, you might want to poke
14:41:45webmindCasainho, you could also try the sparkfun forum
14:42:01CasainhoI already tried Sparkfun forum... no response :-(
14:42:13Casainhoand I am not getting response from Olimex either
14:42:39Casainhoso I think smaller ones will probably quicker help our project than that bigg ones
14:43:16CasainhoI didn't know aboyt liquidware...
14:43:55Casainhoarchivator: don't you want to help us?
14:44:19Casainhowebmind: do you want to talk with liquidware about this project?
14:44:43webmindI don't know them myself.
14:44:57webmindso that won't add anything, except maybe confusion :)
14:46:01archivatorarchivator: sorry, really don't have the time. Besides, I haven't done any EE work for the past 2 years, so I'm not that familiar with current market.. Might do something stupid and spoil the party for everyone =/
14:46:14Casainhooh, I am not having much free time and even I am bad at english language :-(
14:46:32Casainhoarchivator: okok. thanks :-)
14:46:48webmindCasainho, I think your english will be good enough :)
14:47:05webmindCasainho, I'll let you know if I find other options though
14:48:05Casainhowebmind: thnaks!! you can contact us on the forum:;topicseen#new
14:48:17Casainhowe really need help, more people joining us
14:48:33webmindCasainho, I'll keep it in mind
14:48:38Casainhothere is a lot of different work to do
14:48:42webmindbut you won't stay in the channel?
14:48:45CasainhoI must go now
14:48:52CasainhoI must go .- bye bye
14:48:54 Quit Casainho ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
15:00:16 Join robin0800 [0] (
15:06:22webmindbtw, I was wondering
15:06:40webmindis anyone looking in to porting rockbox to playing on cheap chinese 'mp4' players?
15:10:47 Join midijunkie [0] (
15:11:30markunwebmind: any brand in particular?
15:11:48markunthe ones we have been working on are not very cheap
15:12:16 Join |ahmina| [0] (
15:13:25webmindmarkun, these are 25-50 USD
15:13:36B4gder"cheap chinese mp4 players" are many different ones
15:13:45webmindB4gder, I know
15:13:52webmindthey're also rather brandless
15:14:05B4gderyes, or they're called "ipod" :-)
15:14:18webmindB4gder, yes :)
15:14:21markunB4gder: didn't you write a blog post about it?
15:14:25webmindcome to think about it
15:14:31webmindI could of bought 2 of those
15:14:42webmindand donate one :)
15:14:43B4gdermarkun: you mean the onda ones?
15:15:14markunB4gder: I thought it was about chinese brands in particular, but might be wrong
15:15:25B4gdernope, never done that
15:15:51webmindmarkun, what would you call cheap?
15:15:52markunwebmind: keep in mind that a port takes time and if that particular model is not very popular it's a bit of a waste
15:16:05markun< 50 euro
15:16:20B4gderand the unbrandedness is a problem too
15:16:29webmindmarkun, most of these are <50 dollar
15:16:31ZagorB4gder: I was just going to say that too
15:16:40webmindmarkun, true
15:16:55wincentHello all, I would like to participate in Summer of Code 2009. I have some trouble to choose the theme of the project. First one idea of mine was to port Rockbox to a Qt application, which would enable it for using both on Qtopia devices (OpenMoko, Sharp Zaurus et cetera) and on desktop devices (with Qt3 or Qt4). But as I read it from the IRC logs, another guy would like to do something very similar (the only difference is the he
15:16:55wincent to collide with other participants, I got another idea. Recently I ported a UI for a sound synthesis environment to Qtopia, which runs both on Sharp Zaurus (Linux PDA) and on x86 Linux. Now the idea is to port both the synthesis environment and the user interface to Rockbox to have something like this: . The question I would like to ask here is, would such a project be interesting for t
15:17:33gevaertswincent: please use shorter lines. They get cut off now
15:17:53linuxstbwincent: Do you have a link to that "sound synthesis environment" that isn't a video?
15:17:57webmindwincent, your sentace ends at 'nteresting for t'
15:18:27wincentOk. Again: would such a project be interesting for the developers and users of Rockbox? This is my question.
15:18:28B4gderthat video makes no sense to me - looking at it without sound...
15:18:33***Saving seen data "./dancer.seen"
15:18:42B4gderwincent: what exactly is the project?
15:18:43webmindwincent, is that pure data for an ipod?
15:18:57webmindfunky :)
15:19:05linuxstbwincent: Again: Do you have a link to that "sound synthesis environment" that isn't a video?
15:19:08 Quit |ahmina| (Connection reset by peer)
15:19:14Zagorit looks like a fun plugin, but it's not really something that furthers Rockbox
15:19:24 Quit lymeca (Read error: 104 (Connection reset by peer))
15:19:48wincentThis is the environment.
15:20:10 Join mcuelenae [0] (
15:20:10wincentMy UI is here: wincent/blog/">
15:20:22webmindZagor, well it could make a rockbox 'player' into a musical instruments if I understand correctly.
15:21:03webmindwincent, nice, does it run on angstrom aswel?
15:21:09wincentI thought about something like RjDj for iPhone.
15:21:42wincentwebmind: Probably. The sources are in SVN, if you get it to compile on angstrom, then...
15:21:54wincentBut we disgress here a little.
15:21:58webmindyes :)
15:22:32wincentSo, how much interest is to create musical instrument on Rockbox?
15:23:32wincentOr, maybe the better characterization, an environment where instruments created by users can be played.
15:23:47Zagorit's a little outside what I've had in mind, but that doesn't have to be a bad thing
15:24:11archivatorCan the instrument creation part be done on the device or does it have to be done on a PC?
15:24:32wincentIt has to be done on PC.
15:24:48wincentIt is comparable to RjDj −− both have PureData core.
15:28:19 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:29:03kugelrasher: haha, I never knew there's a memo service
15:29:08mcuelenaewebmind: that link on dealxtreme you posted earlier has a RockChip as chipset (says so in the description), which has almost no public documentation (and runs ARM+DSP, with DSP being some unknown core)
15:30:52kugelpetur: thanks for testing. The remote won't receive updates, unless the main unit doesn't have HAVE_LCD_ENABLE or HAVE_LCD_SLEEP
15:31:23*petur actually forgot to test and makes a note...
15:31:47webmindmcuelenae, pity
15:31:53kugelI could play save, and only disable the updates for the main unit, whether the remote is readable or not
15:32:30kugelI didn't see sleep/enable for remotes at all, so I guess assuming all are readable is the safer assumption
15:32:32mcuelenaewebmind: there are some chinese players who run an Ingenic chipset, but most of them aren't as cheap as <50 eur though
15:32:55pixelmathe current remotes are readable without backlight because they are all monochrome or greyscale
15:33:12webmindmcuelenae, any idea on how hard a port would be if there where docs, and if this is feasable? (also considering usage)
15:33:30mcuelenaeI'm porting Rockbox to Ingenic based devices, and it takes quite some time
15:33:57mcuelenaeI think I started about a year ago
15:34:48mcuelenaehmm there are some Onda's ~ 60$ USD
15:34:55mcuelenaeyou get some for less on eBay I guess
15:34:59 Join Ubuntuxer [0] (
15:35:01mcuelenaes/get/could get/
15:35:26*webmind got himself an ipod mini recently
15:35:32webmindso I've got something again
15:35:45 Nick mcuelenae is now known as mcuelenaere (
15:35:47webmindjust thought it would be nice to have it on some hardware which is cheaply available
15:36:23mcuelenaerewebmind: the Onda is rather cheap
15:37:35webmindmcuelenaere, not 25 though
15:37:42mcuelenaerenot that cheap :)
15:37:55webmindand are these models still available?
15:38:02webmind(besides ebay :)
15:38:05kugelpixelma: so, all remotes are readable?
15:38:27pixelmacurrently supported lcd remotes are,yes
15:38:39mcuelenaerewebmind: the VX747 is getting EOF I think, but the VX777 isn't (yet) has some
15:39:50mcuelenaereyou could also look into other Ingenic Jz4740/Jz4732 players
15:40:02mcuelenaeremost of them shouldn't require much porting
15:41:40 Join _lifeless [0] (n=lifeless@
15:42:30kugelpixelma: thanks for the info, I'm going to only deactivate updates on the main unit then
15:43:01 Quit jhMikeS (Nick collision from services.)
15:43:07 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
15:44:33 Join toffe82 [0] (n=chatzill@
15:46:50archivatorNow that there are more people here, could you please look at today's logs - 13:15 GMT - I asked a lengthy question which I'd rather not repeat.
15:47:43archivatorBrief version - where do I place a hook that provides PCM data to plugins?
15:57:00 Join {phoenix} [0] (
15:57:56 Quit __lifeless (Read error: 110 (Connection timed out))
15:58:23webmindmcuelenaere, 60 usd, not a bad price indeed
15:58:25 Join kushal_12_27_200 [0] (n=kushal@
15:58:33mcuelenaerewebmind: which player?
15:58:44webmindvxr747 4gb
16:00:37 Nick fxb__ is now known as fxb (
16:00:46mcuelenaereit doesn't have audio fully working though atm
16:01:14 Quit kushal_12_27_200 (Client Quit)
16:03:31webmindah k
16:03:48webmindthen I won't feel fully ripped of on my ipod :)
16:04:26mcuelenaereother then that, the port is in quite a good state
16:07:01 Quit robin0800 (Remote closed the connection)
16:10:28 Join tyfoo [0] (
16:16:42 Join robin0800 [0] (
16:24:14 Join ultrasonic [0] (n=manohar@
16:35:30 Join jgarvey [0] (
16:41:56 Join dberg918 [0] (
16:42:56dberg918has anybody noticed that the H10 20GB build didn't update on the Current Builds page?
16:43:00 Join Conic [0] (
16:43:34dberg918everything else is at r20485, while the H20 is at r20483...
16:44:05dberg918sorry if this is old news...
16:44:37dberg918I take that back, the Ondio FM is also at 20483
16:44:52pixelmajhulst's server - if the zip is not there, it would also explain the delayed deltas...
16:45:30pixelmasame with the OndioFM
16:45:45B4gder"Warning: couldn't get zip from ... /build-ondiofm"
16:45:55dberg918I was able to download the zip for the H10, and it is version 20483
16:46:10B4gderyeah, that was from the previous build round
16:46:18B4gderthe 20485 failed to get transfered
16:46:25B4gderback to the build master
16:46:37B4gdereach build is done distributed by a remote build client
16:48:01B4gder-sh: zip: command not found
16:48:07B4gderon that host
16:48:51gevaertsYour fault :) You only ask for sdl and the compilers :)
16:49:13*B4gder takes full responsibility for every build server around the clock!
16:49:16 Join larsivi [0] (n=larsivi@
16:50:31larsivihai, do anyone know if work on 2nd or 3rd gen ipod nano has gotten further than what is described in the wiki?
16:50:48larsiviI currently have an unused 2nd gen and may be willing to experiment with it
16:51:37B4gderjhulst: you've got mail!
16:51:38dberg918check the forums
16:51:48krazykitany work done would be documented in the wiki or the relevant thread in the New Ports forum
16:55:00 Join cspotcode [0] (
16:57:03dberg918ok, I'm off to compile my own r20485 build
16:57:08dberg918happy Monday everybody
16:57:33 Quit dberg918 (Remote closed the connection)
16:59:44Unhelpfuldarn commie smurfs, ruining perfectly good wps themes :/
16:59:45 Quit toffe82 (Remote closed the connection)
17:01:21*Unhelpful tries to figure out how to resolve evilnick's complaint by adding a half a byte to his theme
17:06:46 Join saratoga [0] (n=9803c264@gateway/web/cgi-irc/
17:07:46 Quit Zagor ("Don't panic")
17:08:26 Join toffe82 [0] (n=chatzill@
17:14:58saratogawhere is the code for the file browser actually located in rockbox?
17:15:10JdGordonwhich part of it?
17:15:22JdGordontree.c and filetree.c and tagtree.c
17:16:57saratogaJdGordon: Llorean told me about an interesting feature in XBMC where you can merge different volumes together in the file browser
17:17:04saratogai wanted to see how hard that would be to do in rockbox
17:17:36Nico_Pisn't it already done?
17:17:36evilnicksaratoga: I've used that before in XBMC, it's pretty nifty
17:18:10saratogaNico_P: it recursively combines the file tree of different volumes
17:18:28saratogawhat we do now is basically just symlink other volumes as folders in the main volume
17:18:33Nico_Poh I see
17:18:35JdGordonsaratoga: you probably need to look into ft_load() for that i tihnk (that actually loads the directory into the buffer)
17:18:37***Saving seen data "./dancer.seen"
17:18:47JdGordoni dont know if it is doable though
17:18:53*JdGordon rejoins the convo in 20min
17:19:22 Join sanket [0] (i=cb6ef316@gateway/web/ajax/
17:19:47amiconnsaratoga: There is a reason why we "mount" and not merge. Merging potentially produces name collisions
17:20:30 Quit BigBambi ("No Ping reply in 30 seconds.")
17:20:58amiconnBesides, this is a filesystem feature, not a filebrowser thing
17:21:12 Part sanket
17:22:30saratogaamiconn: name collisions should only happen if you have the same song in both volumes, and can be delt with by simply ignoring the second copy
17:22:37saratogathis is what I believe XBMC did
17:22:41*rasher doesn't see *much* benefit in mimicking unionfs
17:23:02rasherUnless it can be done with little overhead/code
17:23:22mcuelenaeresaratoga: does this volume merging feature on XBMC have a specific name?
17:23:38saratogamcuelenaere: i don't think so
17:23:40rasherI'm willing to bet it's unionfs in action
17:24:46Unhelpfulcan't one simply use the DB browser?
17:24:51 Join BigBambi [0] (
17:24:52saratogai didn't even know it existed until Llorean saw my idea and said he'd been doing it for ages
17:25:41saratogawhat targets have multiple volumes? the sansas and one of the Archos?
17:25:55saratogaand i guess some upcoming iaudio players
17:25:57rasherSansas, Ondios
17:26:07gevaertsanything with a ramdisk :)
17:26:09rasherAnd any custom build with multivolume
17:26:36saratogaso its basically just flash players then?
17:26:38gevaertsah yes, so potentially all archoses (big disks need multivolume there)
17:28:01saratogathe archos with big disks have multiple partitions?
17:31:46gevaertsThe bridge chip doesn't handle more than 128GB, so if you have a bigger disk you partition it, only use the first partition over USB, and copy to the second partition within rockbox
17:31:59gevaertsThat trick also gets hard with merged volumes...
17:32:36saratogait might not be worth it on the Archos
17:32:42 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:40:25kugelNico_P: then all checks for id3 are useless. one of the first things in the wps is doing wps_state.id3 = audio_current_track();
17:40:43 Quit GodEater ("Leaving")
17:44:19Nico_Pkugel: IIRC it used to be possible for audio_current_track to be able to return NULL
17:44:43Nico_Palso I don't think the check costs much and it's good practice
17:44:57 Join JdGordon| [0] (i=836b0052@gateway/web/ajax/
17:49:49kugelNico_P: I find it more confusing than good practice, because half of the wps functions checks for id3, the other half not.
17:50:32 Join toffe82_ [0] (n=chatzill@
17:51:04 Part B4gder
17:51:45kugelI'm not saying checks are wasteful, but it's quite confusing to. I'd say a initial check (and all the subroutines assume valid id3 (which is safe, because id3 is valid once they're called)) should be sufficient
17:52:52 Join toffe82__ [0] (n=chatzill@
17:53:02kugelhm, mpeg.c can apparently return NULL in that function
17:54:56*JdGordon| said that yesterday
17:55:26*kugel doesn't remember
17:56:01kugelbut only if there's no tracks in the memory, which doesn't sound possible to me
17:57:49JdGordon|for that you need to talk to somone who actually knows how mpeg.c works...
17:58:00Nico_PI don't believe removing NULL pointer check to be a useful thing to do
17:58:01 Join kushal_12_27_200 [0] (n=kushal@
17:58:43kugelI think it's worse when it's not consistently being done.
17:59:30 Quit toffe82_ (Read error: 60 (Operation timed out))
17:59:54saratogai wonder if funman's unknown I2C audio device on the V2 clip might be an AS3517
18:00:03 Quit BigBambi_ (" ajax IRC Client")
18:00:15saratogathe registers he mentions are present on the 3517 and do things that ought to be useful
18:00:21Nico_Pkugel: then IMHO the missing checks should be added
18:00:30kugelsaratoga: older than what's in the V1 clip?
18:00:48saratogathe v1 has a 3514 built in
18:01:02saratogaand the v2 apparently does not, so the 3517 would be an obvious upgrade
18:01:54saratogai think funman isn't right about ARMv5 though, since I also so those ops in the V1 firmware, I think they may just be dead code in one of the codecs or some other binary component
18:01:54 Join flydutch [0] (
18:01:55kugelsaratoga: it has the 3514? How's the sound quality so much better compared to e200v1, then?
18:02:55 Join bertrik [0] (
18:03:15kugelI always thought we're just lucky that the as3525 is so compatible in the areas we re-use 3514 code
18:04:22saratogai think its actually the same device, but its possible they changed things on it
18:04:37 Quit petur ("work->home")
18:04:46saratogaeither that or integrating it onto the CPU fixes whatever interferance it had on the v1
18:06:44 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:06:56 Quit robin0800 (Read error: 110 (Connection timed out))
18:07:24kugelJdGordon|: so, was/is it hard to sync with my changes?
18:08:33 Quit toffe82 (Read error: 110 (Connection timed out))
18:10:29 Join faemir [0] (
18:10:44kugelNico_P: one could also do assert(id3), and define assert as nothing as long as we're sure there can't be NULL. if code is reused, only assert would be enabled.
18:11:25kugelbut yea, just removing all checks isn't the right thing to do, I agree
18:12:13 Join krazykit` [0] (
18:12:13 Quit krazykit (Read error: 131 (Connection reset by peer))
18:12:29archivatorAnyone got any ideas on how to expose PCM output data to plugins?
18:12:41Nico_Parchivator: did jhMikeS give (you or anyone else) pointers on how to tap into the PCM driver?
18:13:08Nico_Pyou want the data that's leaving the PCM buffer and being fed into the DAC
18:14:11Nico_Pkugel: I don't understand the point of defining assert as nothing just for the sake of one particular case. also, where would you put the assert(id3) call?
18:14:39archivatorWell, I could also use the data that's being fed to the PCM buffer though that wouldn't be realtime.
18:15:38mcuelenaeresaratoga: I don't think that post needed removal, but I'll PM him instead now :)
18:15:43 Nick krazykit` is now known as krazykit (
18:15:58mcuelenaereoh you moved it
18:16:19 Join cspotcode1 [0] (
18:17:02Nico_Parchivator: I guess as a first step, non realtime would be OK. at least that way you'll have PCM data to display
18:17:10kugelNico_P: in every function that wouldn't work properly with id3 == NULL
18:18:04kugelthe assert could also be used for the other structs which can basically never be null
18:19:01kugelanyway, I don't care much anymore, it's not so confusing for me anymore (but when I first looked at the code)
18:23:10saratogamaybe add a comment if its still ugly
18:23:29*mcuelenaere just did
18:23:48mcuelenaerenvm that last line
18:26:52 Join MethoS- [0] (
18:26:53 Quit killan (Read error: 104 (Connection reset by peer))
18:26:54kugelsaratoga: as3527 isn't our candidate either
18:27:07kugelthe datasheets for 353x need to be requested
18:28:13kugelthose would be armv5t too
18:29:10kugelsaratoga: also, 353x sounds too advanced for sandisk's needs to me
18:30:15 Quit perrikwp (" ajax IRC Client")
18:30:59Unhelpfulmaybe a core function could just set a callback to call right before (or maybe right after?) pushing a chunk of data to the actual DAC? and anybody implementing such a callback would do so with the understanding that it should return *quickly*?
18:32:42 Quit cspotcode (Read error: 110 (Connection timed out))
18:34:23 Join killan [0] (
18:39:21 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
18:39:32 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
18:39:53 Join bs66_ [0] (
18:41:21archivatorAm I right in assuming that the path from buffer to DAC is pcm_play_data -> pcm_play_data_start -> pcm_play_dma_start (driver side)? I think my hook should be in pcm_play_data* but I am still vague on the details.. Do I copy data to a buffer specified in the plugin or do I use callbacks of some sort?
18:43:01 Quit cspotcode1 ("Leaving.")
18:43:21JdGordon|kugel: havnt tried yet
18:48:40 Quit mcuelenaere ()
18:49:03 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:54:59saratogakugel: if you hex edit the v2 firmware, its got the strins AS3525 in it, so I think its probably not a 353x
18:55:33 Join Erant [0] (
18:56:29domonokyarchivator: both ways are possible, but maybe a callback would be enough. plugins can the still copy it in the callback into a buffer if needed.
18:57:01 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
18:57:08 Quit homielowe (Remote closed the connection)
18:59:13 Quit saratoga ("CGI:IRC (EOF)")
18:59:41archivatordomonoky: Right. I'll try it out in a few minutes.
19:00:55 Part pyro_maniac1 ("Leaving.")
19:00:56 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
19:00:59 Quit kushal_12_27_200 (No route to host)
19:03:44jhMikeSarchivator: pcm_play_dma_get_peak_buffer gives the current playback position within the current buffer as accurately as possible. it's also a moving target afterall.
19:04:22archivatorjhMikeS: I'm not sure I follow.
19:05:27jhMikeSarchivator: the return value is the pointer of the next data to be fed to the DAC (laging by perhaps 4-8 samples), count is the amount remaining in the current chunk.
19:05:32 Quit parafin (Nick collision from services.)
19:05:40 Join parafin|away [0] (
19:05:58 Nick parafin|away is now known as parafin (
19:06:55archivatorjhMikeS: How do I know how long this next chunk is?
19:06:59jhMikeSthe pcmbuf_beep uses it in order to mix beeps with a few ms latency at most. see apps/pcmbuf.c
19:08:10 Join fml [0] (n=4fd3d5ee@gateway/web/cgi-irc/
19:08:20jhMikeSarchivator: 'count'−− what hasn't been played yet (in samples). mind that you could get 0 for count and NULL for the return value so that must be handled
19:09:01 Join kps [0] (n=40eb617d@gateway/web/cgi-irc/
19:09:19 Quit kps (Client Quit)
19:09:26fmlHello. Any comments about FS #9618 ? There was not a single reply on the mailing list. If it's not that important I'll commit it. Consistent WPS tag description is good IMO.
19:10:05Nico_PjhMikeS: what does "peak" mean in this context?
19:10:31archivatorjhMikeS: This wouldn't by any chance have a different behavior on a simulator? 'Cause that's what I'm currently using and the sim is getting new values every HZ/20 ticks..
19:10:58 Quit bzed (Read error: 101 (Network is unreachable))
19:11:09jhMikeSNico_P: It's rather atavistic now I think. It was put there so peak calculation could get the buffer for the current chunk but now it appears it's more useful than that.
19:11:55jhMikeSarchivator: It might due to audio card latency but it shouldn't be essentially different
19:11:55 Join kachna [0] (
19:12:30 Join webguest74 [0] (n=40eb617d@gateway/web/cgi-irc/
19:12:31bluebrotherdomonoky: any reasons to not release rbutil 1.1.1 tonight?
19:12:52 Quit webguest74 (Client Quit)
19:12:54domonokyno reason, but why not rbutil 1.2 ? :-)
19:13:15*rasher is compiling a static qt 4.5 on amd64
19:13:21bluebrotherthought about that too but it only seems like a minor update.
19:13:33rasherFor the nth time (ran out of space the other n-1 times).
19:13:40archivatordomonoky: did you see my latest pain-soaked attempt to integrate festival? :P
19:13:46bluebrotherany other opinions on this? I wouldn't mind making it 1.2
19:14:19 Quit linuxstb (Read error: 60 (Operation timed out))
19:14:29rasherIt's technically not much, but theme site integration (if the site goes live..) is rather major
19:14:49bluebrothertrue. That's what I was thinking about myself too ;-)
19:15:58domonokyarchivator: i only took a short look. didnt found much time to really try it out.
19:17:15archivatorRight, no hurry. Just wanted to let you know this is probably as good as it gets - I honestly don't see what else I could integrate short of duplicating the festival_client functionality..
19:17:56 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:18:41***Saving seen data "./dancer.seen"
19:19:45 Quit Ubuntuxer ("Leaving.")
19:19:50jhMikeSarchivator: the sim might be grainier than the targets though (how often the pointer is updated thus how much it advances each time and how much audio is sent per "request").
19:20:43domonokyarchivator: from the code, the changes look good. But i will need some time to really try it out :-)
19:20:54archivatorjhMikeS: Nah, I think it was a bug in my code, will see in a moment.
19:21:23archivatordomonoky: As I said, no worries. It's not exactly key functionality.
19:21:58 Join HBK [0] (
19:22:24domonokyby the way: i am think about changing how configuration of tts and encoders work.. ( i want to better seperate Ui and logic). Maybe some generic configuration dialog based on a Table :-)
19:23:18archivatorSounds good. I'll start work on a new TTS process some time next month.
19:23:37jhMikeSarchivator: beware though, some targets can hand back physical addresses (pp502x, imx31)
19:24:58archivatorI am using a PP - how do I read from those addresses from pluginspace?
19:25:19domonokyfor tts and encoder configuration, i am thinking about letting them present a data model, and use a generic table view (which shows dropdown boxes, spinboxes etc based on the models data.) But this will take a bit time to get to a good shape :-)
19:26:08 Quit lightbulbjim (Read error: 110 (Connection timed out))
19:27:46archivatorMVC would be of great help there - just make sure the new system is easier than the old one (or at least as easy) :) We don't want 10 LOC to get a single configuration value.
19:28:01jhMikeSarchivator: just read with the returned point. to translate any other pointers related to the buffer there is pcm_dma_addr. pcmbuf_beep apps/pcmbuf.c uses all that
19:28:34archivatorRight. Thanks, will try it out.
19:28:36*bluebrother is going to release rbutil 1.2 then shortly −− unless someone starts screaming *now*
19:29:59kugelFlynDice: nice work
19:32:22 Join Terinjokes [0] (n=a5a10310@wikinews/Terinjokes)
19:32:27Terinjokesheylo again
19:33:17 Quit Terinjokes (Client Quit)
19:35:35kugelFlynDice: I'm wondering how you tell it's running
19:39:01 Join Tamashii [0] (n=be9f50c6@gateway/web/cgi-irc/
19:39:35archivatorI was wondering, what's the official position on having a more general rbutil - let's call it rbtray for the purposes of discussion. I envision an app that detects when a rockbox device is connected, syncs it, does maintenance on its database, fetches albumart and that kind of stuff. No music management though, just automating the most repetitive of tasks. Kinda like a Rockbox on the device, rbtray on the PC bundle. Also, I don't think rbutil is a goo
19:39:35archivatord fit for such an application.
19:40:59TamashiiI'm wondering if somebody could help me.
19:40:59kugelrasher: the Portuguese update is flawed, it re-enables the usbstack strings
19:41:30mcuelenaereTamashii: not without a question
19:41:51bluebrotherarchivator: you might be interested on the last devcon writeup: (next to the bottom)
19:42:10BigBambiarchivator: I don't think some of those are really within the Rockbox remit - such as fetching album art. An argument could be made for some of the others, but I'd not personally use it
19:42:14bluebrotherTamashii: you need to click the other button, not that one. Then it will work.
19:42:30*LambdaCalculus37 is going to be receiving a donated Sansa c250 soon... anyone going to need one?
19:42:34TamashiiHehehe, of course: I've run thru a lot of IRC logs in the Rockbox page, and I've run thru some solutions well explained in that logs, however, I'm still unable to install Rockbox in my iPod nano 1st. gen.
19:42:51kugelLambdaCalculus37: v1?
19:43:01rasherkugel: oh, oops
19:43:10LambdaCalculus37kugel: I believe so. I'll find out for sure when it arrives.
19:43:36archivatorBigBambi: I realize that and that's why I'm not proposing extending rbutil in such a way. I was just thinking of Rockbox as an umbrella project and this new tool as a smaller project that fits but not quite.
19:43:43bluebrotherarchivator: my personal opinion: there are better programs for handling album art and stuff. I'd however like to see an option to change the rockbox configuration and update the database
19:44:06rasherbluebrother: What does this mean - RCC: Error in 'rbutilqt-lang.qrc': Cannot find file 'lang/rbutil_de.qm' ?
19:44:16bluebrotherTamashii: how exactly did you try installing it? And what's the error?
19:44:21TamashiiI already did the Restore configurations on iTunes, I already downloaded the 1.3.1 update package and installed it. No avail, the Rockbox Utility still is complaining about being unable to install the bootloader
19:44:31BigBambiarchivator: I don't think we really want anything that does anything outside direct Rockbox stuff (such as build database for instance) (just my opinion)
19:44:43BigBambiarchivator: Anything "official" that is
19:44:48domonokyrasher: ignore it, it warns because it cant find the compiled translations, but they will be built later.
19:44:51bluebrotherrasher: if building for a release this only warns because the language files haven't been lrelease-d yet. It will be done during the build process so this isn't a problem
19:44:55 Join kps [0] (n=40eb617d@gateway/web/cgi-irc/
19:44:59BigBambiarchivator: However, my view is minor :)
19:45:17bluebrotherTamashii: "not being able" is a wide field. Can you be more specific?
19:45:23Tamashii@bluebrother: I'm using Rockbox utility.
19:45:31bluebrotherTamashii: I already figured that ...
19:45:52kugelnot sure about the fetching, but I think at least converting album art would be a job for a rockbox-affiliated app, because rockbox needs special album art
19:46:10archivatorBigBambi: So such a project is better off as a project with rockbox integration than a rockbox project with side functions?
19:46:17bluebrotherTamashii: what OS are you using? Have you checked the output of About / System Info?
19:46:17kugeland if rbutil wants to be the housekeeping utility, I don't see why it shouldn't do it
19:46:32BigBambiarchivator: Depends what it does IMO
19:46:32TamashiiOk, I click the Complete Installation button... And then it downloads the bootloader
19:46:38JdGordon|I would have no problems with rbutil fetching album art... especially if AA ever gets linked to the db
19:46:52TamashiiAnd then: Failed to add bootloader.
19:46:58bluebrotherwhat's special on album art being in bmp format? There's nothing special at all, it's just one specific file format.
19:47:06BigBambiI just think there is a danger in RBUtil doing too much
19:47:10bluebrotherTamashii: what OS are you using? Have you checked the output of About / System Info?
19:47:20domonokyarchivator: my opinion about such a tool: if it doesnt stray too far from the rockbox needs, i am ok with integrating them into rbutil..
19:47:41 Join ibseco [0] (
19:47:44JdGordon|BigBambi: why cant rbutil be the swiss army knife for rockbox on the desktop?
19:47:46kugelbluebrother: only supporting bmp, and neither jpeg or embedded, *is* special
19:47:49bluebrotherI'd rather have rbutil use Qt plugins for such functionality.
19:47:56BigBambiIt gets more confusing to use, harder to maintain,etc - there is no need for it to handle the kitchen sink
19:48:10archivatorJdGordon|: where do you draw the line? AA is allowed but tag editing is not? Database update is allowed but music management is not? From a user perspective, doing one and not the other would create chaos IMO.
19:48:25domonokybluebrother: plugins is just a way to better structure the code in rbutil. Its still integrated into rbutil :-)
19:48:25*JdGordon| prefers not to draw lines
19:48:29rasherHm, seems you can manage to bring down rbutil source to about 17MB if you go on a deleting rampage
19:48:31Tamashii@ bluebrother: Yes, I've checked it. I'm using Windows XP SP3. In About/System Info: OS: Windows version 5.1, build 2600 (Service Pack 3)
19:48:38bluebrotherthe major problem with extending the functionality is simply that someone needs to maintain it. If we allow plugins to extend the functionality others can write plugins too
19:49:12JdGordon|oh yes, pluginise these stuff for sure
19:49:32Tamashiibluebrother: Username: Tamashii, Permissions: Admin, Attached USB Devices: VID: 046d PID: c049, USB Gaming Mouse, VID: 05ac PID: 120a, iPod [...]
19:49:41archivatorbluebrother: with the current codebase, creating a plugin platform is not worth it - it'd be easier to maintain a separate app
19:49:49domonokybluebrother: but who maintains those plugins ? i dont think there will be much devs outside building plugins, but not wanting to commit them to our svn.
19:50:16bluebrotherdomonoky: I'm not sure about this. Just remember this Rockbox Center ...
19:50:53BigBambiRockbox Center? What was that?
19:50:58domonokybut i am ok, with putting some functionality into plugins. there is even a prove of concept plugin (for python code) in the tracker... :-)
19:51:04bluebrotherBigBambi: can't remember :)
19:51:12BigBambihaha :)
19:51:38domonokyi built this python plugin for them, so they could extend rbutil with it, but they never used it...
19:52:20bluebrotherI guess it was too complicated to get the plugin running as it's not part of the official rbutil binary (yet)
19:53:08bluebrotherstill, a plugin concept might be good. We'll see how much it'll get used. And if it also helps having a better code structure the better
19:53:34 Quit ultrasonic (Connection timed out)
19:54:09domonokyno, i was in contact with them, and told them to tell me what they need... but it never happend..
19:54:16 Quit kps ("CGI:IRC (EOF)")
19:54:53domonokyso its not because it wasnt official, but because the lost interesset or similar :-)
19:56:37 Join taylor_ [0] (
19:58:07rasherbluebrother: we could change it in the theme site, but theme size is given in bytes...
19:58:59 Join Makuseru [0] (
20:00:40 Join Thundercloud [0] (i=thunderc@
20:02:48JdGordon|Nico_P: see my comment on 9795? what you rekon?
20:02:56 Join miepchen^schlaf [0] (
20:06:16 Join PaulJam [0] (
20:07:14 Join robin0800 [0] (
20:07:44 Quit Conic ()
20:08:06 Join petur [50] (n=petur@rockbox/developer/petur)
20:08:22 Quit Tamashii ("Thanks!")
20:08:42rasherbluebrother: modified rbutil.tpl to divide by 1000
20:09:31bluebrotherrasher: ok, though I'd prefer 1024. It's displaying as kiB ;-)
20:09:46rasherHurg. Well okay
20:10:58 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
20:11:01 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:11:12 Join Conic [0] (
20:12:44 Join Schnueff [0] (
20:13:06FlynDicekugel: re mmu I'm not sure it's running but there is a marked difference in the speed up to the point it fails so I _believe_ it is. I figured I'd just post the patch so maybe someone with a little more experience could tweak it into working...
20:16:41kugelFlynDice: your patch doesn't change the bootloader correct?
20:16:46 Quit Conic ()
20:17:17kugelI'm wondering where it fails, then. "no partition found" is a bootloader message, isn't it?
20:17:52kugelrasher: do you plan to fix portuguese ?
20:18:03gevaertskugel: no
20:18:08rasherkugel: yeah, not exactly an urgent matter
20:18:29gevaertskugel: "Insert USB cable and fix it" is the main firmware
20:18:44FlynDiceNo, it fails in main.c I believe where it tries to mount disks. I stayed away from the bootloader and just mimicked the OF in mmu-arm.s
20:18:51kugelI've never had this, so I don't know
20:18:53gevaertsIn this case of course it means that you should fix USB first ;P
20:20:40 Join casainho [0] (
20:21:41 Quit robin0800 (Remote closed the connection)
20:22:46kugelpetur: did you have a chance already to test the patch now?
20:23:06kugelI changed to not disable updates on remotes (for now, at least), which should be safe
20:23:26casainhohello :-) −− for a bootloader first patch of Lyre, do I need to commit also the ?
20:24:08domonokyFlynDice: i probably fails because sd-access uses dma, which might have problems with caching...
20:25:31 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
20:26:35 Quit Makuseru (Read error: 104 (Connection reset by peer))
20:26:39domonokygsoc student applications are now open, i wonder how will be the first ? :-)
20:26:42amiconncasainho: RGB444 isn't difficult, as that would still need a 16 bit framebuffer. I.e. simply use lcd-16bit.c for all the drawing stuff, and have your target specific driver take care of some details
20:27:03*amiconn once considered going RGB444 for c200
20:28:19amiconnThe display isn't that great anyway, and it would allow to speed up YUV blitting by ~30% (that's a low-level driver thing). But then it was possible to speed up the low level driver enough to keep RGB565 and still have mpegplayer run smoothly
20:29:00casainhoamiconn: I am really busy with others things now, like a first patch and put audio working. I don't really know what can be better, since this is my first time working with LCD. I also think that later I could try to use DMA for sending the data for the LCD.
20:30:25casainhoamiconn: I am looking for alternatives, for some individual/small company to assembly and produce first 10 units prototypes and on that time, we will change the LCD if we can (we desire it)
20:31:39rasherWow, the rbutil sources compress down to 600KB
20:32:22casainhoamiconn: do you want to work on the LCD code? ;-) :-)
20:32:41rasherIf you strip everything but the needed files from apps/codecs/libspeex and tools/, and exclude .svn dirs
20:32:50casainhoamiconn: and maybe find one for future hardware?
20:33:55 Join Conic [0] (
20:35:08domonokybluebrother: rbutil-1.2 branch ready for building ?
20:35:16 Join tessarakt [0] (
20:35:28 Quit jhMikeS (Nick collision from services.)
20:35:28 Quit tessarakt (Read error: 104 (Connection reset by peer))
20:35:34 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
20:35:53bluebrotherdomonoky: yes
20:37:16rasherNo changes needed in release versions?
20:37:45 Join Lear [0] (i=chatzill@rockbox/developer/lear)
20:38:03domonokyno, rbutil now gets the rockbox release numbers from the server.
20:39:27 Quit kugel (Nick collision from services.)
20:39:32 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:40:15peturkugel: will test very soon...
20:40:16bluebrotherthat change was the reason for the first minor release in about two years :)
20:43:39domonokyBagder: rbutil 1.2 mac binary"> please move to the download server
20:44:25kugelpetur: nice, thanks
20:46:17bluebrotherBagder: non-mac binaries for rbutil 1.2 here:
20:46:56 Quit Horscht ("Verlassend")
20:46:56rasherbluebrother: you're forgetting about 64bit!
20:49:40bluebrotherrasher: bah, who needs 64bit? :P
20:51:23scorche|shpeople with >=4GB of RAM?
20:51:34 Quit {phoenix} (Remote closed the connection)
20:51:43 Join jaykay [0] (
20:51:44taylor_64 bit is awesome, nevermind it cracks encryption faster ;)
20:52:09*scorche|sh shoos taylor_ away into a different channel
20:52:34 Join MT [0] (n=MTee@
20:53:11taylor_scorche|sh: what is your problem with me here?
20:53:23*scorche|sh points to the topic
20:53:33MTsaratoga : ping
20:54:48Lloreantaylor_: This is #Rockbox. Random off-topic comments don't belong here.
20:54:56kugeljaykay: r20355 *is* backported
20:55:51*taylor_ reminds everyone scorche|sh how they were talking about 64-bit processors
20:56:28scorche|shtaylor_: and what does 64-bit processors breaking encryption have to do with rockbox?
20:56:39Lloreantaylor_: No, they were talking about 64-bit support for RBUtil.
20:56:59JdGordon|... to break encyrption
20:57:39jaykaykugel: why are 20256 and the ~40commits after r20357 not shown int the "normal" commit lists?
20:57:45MTlinuxstb : ping
20:58:07rasherjaykay: Because they didn't happen in any of the places that show up there.
20:58:13taylor_To break enryption for devices? I don't know. The point its, rasher asked why someone might need a 64-bit processor and I gave one reason. Now move on.
20:58:52kugeljaykay: is r20356?
20:58:53jaykayits quite confusing when 40 commits are missing...
20:59:08Lloreanjaykay: Then follow the commits by hand
20:59:13LloreanThe frontpage only tracks commits to the "trunk"
20:59:23 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
20:59:24jaykaykugel: i meant that, sorry... and its not shown...
20:59:27LloreanIf you want to watch other branches, you have to manage it yourself.
20:59:48jaykayLlorean: that doesnt change the fact that missing commits are confusing
20:59:49kugeljaykay: then lookup what it did, you'll notice that it exists, but not every r is shown on the front page
21:00:08Lloreanjaykay: Duplicate commits would be just as confusing to people *less* technically knowledgeable.
21:00:19Learjaykay: You can subscribe to commit mails, where you see all commits.
21:00:22kugelyou'll get used to it, and alternative svn log tools/techniques
21:00:25bluebrotherthat ~40 commits were the theme site stuff
21:00:36jaykayLlorean: the commits for the theme site are not duplicates
21:00:41Erantscorche: So I was interested in possibly doing a project for rockbox, it being one of the precious few embedded organisations in GSoC. How much of the hardware on the players is currently supported?
21:00:44Lloreanjaykay: The backported one would be.
21:00:48LloreanNon-theme, I mean
21:00:52rasherjaykay: they also don't belong on the frontpage.
21:01:07rasherThe frontpage lists commits to rockbox trunk. Nothing confusing about that.
21:01:12Lloreanjaykay: Seriously. The frontpage serves a specific purpose, it's *not* an "all commits" list.
21:01:37scorche|shErant: it depends on the device, but are you thinking about perhaps supporting currently unused hardware on a device?
21:02:22Erantscorche|sh: That being one of my interests. Software/hardware boundary.
21:02:49Erantscorche|sh: Ofcourse, porting SCUMM is awesome too ;)
21:02:50domonokyErant: on "supported" targets nearly all hardware works. or else they wont be supported. But there are many upcoming ports which could need more driver work :-)
21:03:03jaykayof course its your decision what is shown and what not, but i have the opinion that "commits" shouldn't show up as commit anywhere... no need to hide them then
21:03:05peturkugel: what FS was that again?
21:03:24rasherjaykay: They're not hidden.
21:03:26jaykayand backporting a fix is worth mentioning in the commit logs
21:03:27bluebrothercommits shouldn't show up as commits?
21:03:27scorche|shErant: you could look at, say, getting USBOTG support working, or some other USB bits
21:03:31Lloreandomonoky, Erant: Though sometimes hardware is supported only to the degree necessary to get our features working. It could be possible to expand use of it (USB Host type stuff, perhaps better charging options)
21:03:37bluebrotheras what should they show up instead?
21:03:38jaykayrasher: on the main page, yes
21:03:40*bluebrother is confused
21:03:42rasherjaykay: Backporting a fix *is* mentioned in the commit logs.
21:03:43Erantdomonoky: Got any interesting ports? Ones that possibly need some reverse engineering?
21:03:44 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
21:03:46jaykaybluebrother: i think they shouldnt be commits
21:03:54*Erant slaps Horscht
21:04:00rasherjaykay: the front page is not the main source of commit messages.
21:04:03Lloreanjaykay: "Not showing" is not the same as "Hidden". There are multiple parts of SVN, we show commits to the "Main Rockbox" part of SVN on the frontpage.
21:04:06kugeljaykay: the font page commit list is to track actual developement, backporting a commit shown to another branch is not actual developement
21:04:10HorschtErant, you keep stalking me!
21:04:10bluebrothercommits shouldn't be commits? :o
21:04:14ErantHorscht: Long time no see ;)
21:04:23LloreanHorscht: This is off-topic. Please take random social chatter elsewhere.
21:04:25jaykaybluebrother: exactly :)
21:04:37scorche|shErant: the broadcom chip on the 5th gen ipods is currently unused, but i dont think that it would be a good project...if i were you, definitely look at our USB ideas on our ideas page though =)
21:04:53bluebrotherso a commit now becomes ... nothing?
21:04:54*domonoky thinks working on new targets might be a difficult gsoc project. you never know when you stumble about some "unsolveable" problem. (like chips without datasheet)
21:04:55Erantscorche: Sounds interesting. Not done much with USB, would be a good place to learn.
21:05:10gevaertsErant: I wouldn't recommend things involving too much reverse engineering for GSoC. The risk of failure is *very* high on those
21:05:14jaykaybluebrother: yes
21:05:20LloreanErant: Right now the AMS Sansas are the closest to complete of the interesting ports. I don't know how much more RE is needed. There's also the potential to RE the broadcom chip in the iPod 5G/5.5G. That's very much "supported player with not supported bit"
21:05:27scorche|shdomonoky: indeed...though there are some targets (such as archopen devices, etc) where we know it would be possible, so..
21:05:36*bluebrother doesn't understand what jaykay is intending to accomplish
21:05:39Erantgevaerts: It is hit and miss, but I have done reverse engineering.
21:05:46bluebrotherexcept confusing people in this channel
21:06:10kugeljaykay: use if you're curious about not mentioned commits
21:06:36rasherjaykay: or subscribe to the -cvs mailing list.
21:06:48gevaertsErant: I'm not trying to imply you wouldn't be able to do it :). Getting stuck for months is not unheard of though, and in GSoC terms that means failing
21:07:02Erantgevaerts: Unfortunatly, stuff like RE isn't taught in university, so you gotta pick up stuff where you can :/
21:07:08scorche|shgevaerts: nice..
21:07:21Erantgeaverts: Very true.
21:07:40domonokyErant: do you have an rockbox able mp3player ?
21:07:49LloreanErant: For SoC projects, we're most likely to choose in favour of ones where a timeline can be realistically approximated.
21:07:51 Join casainho_ [0] (n=chatzill@
21:08:09linuxstbMT: Pong
21:08:26amiconnThe broadcom chip isn't unused... otherwise rockbox wouldn't be able to display anything</nitpick>
21:08:39ErantI have an older iAudio I can get my hands on, yeah. Ofcourse, hardware would depend more on the type of project.
21:09:15kugelWhen is release time?
21:09:31kugelonly some 3-4 ours left, I guess?
21:09:42ErantStuff like USB host would require a specific unit, obv.
21:10:06gevaertsyes. Not *very* specific, but still not just anything
21:10:15Lloreankugel: Whenever Bagder's around to do it, mainly.
21:10:42*kugel practices the release-dance once more
21:11:13MTlinuxstb : a fourier-related question :) ; I guess I will not have to deal with the math in the initial phase of the project, not until I start converting the decoder to fixed point, correct ?
21:14:48MTlinuxstb : good, because my studies up till now were all about Fourier series and transforms for time continuous signals, no DFT, FFT ... I got myself a good book though so I was just checking to see that I'll have time to be good at them by the time I need them.
21:15:15 Join GodEater [0] (n=yeahrigh@rockbox/staff/GodEater)
21:17:57 Quit midgey ()
21:18:45***Saving seen data "./dancer.seen"
21:19:42 Quit CaptainKwel (" ajax IRC Client")
21:22:07MTlinuxstb : could you please check my application one last time to tell me if I'm good to go ? ( )
21:22:44MTand if you have any comments, on the 'reporting' section specifically ?
21:23:49peturkugel: my remote seems to behave as it should
21:24:05 Quit casainho (Read error: 110 (Connection timed out))
21:24:57kugelpetur: cool. You can verify that it's working for the main unit by commenting out the queue_post in the hook function. Then, the update after backlight on is slightly delayed
21:29:23linuxstbMT: I would swap steps 2 and 3 - implement the standalone rm-to-wav first (using floating point), then convert to fixed point. (and then port to Rockbox).
21:29:48rasherWas natsort ever disabled in the release branch?
21:30:20linuxstbMT: (I am looking at 2.2.1)
21:30:23MTlinuxstb : yes, done.
21:32:36 Join taylor___ [0] (
21:33:07taylor___*lost power for two seconds :)
21:33:19 Join raphi [0] (
21:33:41domonokyMT: as a hint: you can still change you application after you submitted it to google :-)
21:33:50linuxstbMT: I have no other comments - it looks good to me.
21:33:57BigBambitaylor___: For the last time, please stop the random not related comments in here
21:34:03doomwakewhat picure format does rockbox support?
21:34:11kugelpetur: what irivier do you have?
21:34:29Unhelpfuldoomwake: bitmap.
21:34:32Lloreandoomwake: jpeg, non-progressive, for displaying. Bitmap for using in WPSes
21:34:33bluebrotherdomonoky: for album art? bmp.
21:34:46MTdomonoky : Thanks for the hint :)
21:34:55MTlinuxstb : thanks :)
21:35:11kugelpetur: good then. Did you verify it is working ok for the main unit?
21:35:21taylor___BigBambi: Real sorry about that, I actually did mean to post that on another channel with '/me'.
21:35:37taylor___wont happen again :|
21:39:28peturkugel: actually, it doesn't...
21:39:56taylor___Just to let everyone know, I tried running the 'notes' file under debug mode, but it crashes debug mode, so pretty useless for now :P
21:40:12peturif I initiate playback via the remote, and then wake up the main unit screen, the backdrop isn't loaded
21:41:13peturif I skip to the next/prev song it loads
21:41:35kugelweird. setting the backdrop doesn't happen in the function that I made returning earlier if the lcd is inactive
21:42:05kugelis that with commenting the line out? or with the plain patch too?
21:42:14peturplain patch
21:42:24peturI also saw it with the modified one
21:42:40kugelit's working with svn, I assume?
21:42:51peturI thought is was because of the commented-out line but on reverting it was still there
21:42:52linuxstbtaylor___: What's debug mode?
21:43:01peturwill try now with svn
21:44:10taylor___debug mode, its pretty easy to find?
21:44:38taylor___like I said, its useless howver
21:44:46Lloreantaylor___: What player are you referring to. Of the known modes for various players, I can think of DFU, Manufacturing, Diagnostic, and Disk.
21:44:51LloreanAnd Recovery
21:45:02taylor___there is a small 'debug' mode..
21:45:08gevaertsLlorean: and bricked :)
21:45:17taylor___that can be activaded in the notes area
21:45:26 Join merbanan [0] (
21:45:32linuxstbtaylor___: Remember this channel covers about 30 different DAPs, so some context to your comments is helpful...
21:45:46*GodEater has no idea what taylor_ is talking about
21:45:50LloreanEspecially if you're talking about an OF that most of us never touch
21:46:06BigBambiGodEater: Nor does anyone else
21:46:11taylor___gavearts: because of the notes area it would be extremely hard for the average user to break an ipod unless you open it up and smash with a hammer :)
21:46:15GodEaterah, I'm in good company then
21:46:22 Quit taylor_ (Read error: 110 (Connection timed out))
21:46:23taylor___Because of the DFU i meant to say
21:46:40GodEaterso you're talking about some sort of ipod then ?
21:46:43BigBambitaylor___: For god's sake, what player?
21:47:06taylor___hold on a getting the key combo right now, and don't swear please
21:47:18*GodEater wonders who swore
21:47:28evilnicktaylor___: The generation would be helpful. And saying god is not swearing.
21:47:30BigBambia) I didn't swear, and b) for god's sake, what player?
21:47:55archivatorRegarding my plugin work, good news first - I was using the right method to access PCM data all along, had a bug that messed up all the data. Bad news - something is inherently wrong with my FFT implementation - the graphs I'm getting look weird - peaks at 1/4, 1/2, and 3/4 - will have to find another implementation. A (simple) frequency spectrum analyzer is coming soon.. :)
21:48:12taylor___Please give me a sec to open the documentation
21:48:17taylor___its on most ipods btw
21:48:32BigBambiah, so you are talking about ipods
21:48:39*taylor___ thinks everyone is a little impatient ;)
21:48:41GodEateryes, but he won't say which one
21:48:48GodEatermaybe it's a secret
21:49:00taylor___How can I say which one when its 'ALL'?
21:49:18GodEaterwe know of 5 generations which don't have one
21:49:19 Quit fml ("CGI:IRC (EOF)")
21:49:25evilnicktaylor___: List the ones that it does affect?
21:49:26BigBambitaylor___: All players ever?
21:49:28GodEaterso it's hardly ALL
21:49:41taylor___Ok... let me tell you how to do this....
21:49:58taylor___its not a very indepth 'debug' mode. Apple just calls it this
21:50:09taylor___say you have a note file or a picture you want to check for errors..
21:50:10peturkugel: SVN is ok
21:50:22taylor___*and its in the notes folder*
21:51:03MTlinuxstb : I started extracting cook decoder from ffmpeg yesterday, looks like it will need a lot of files with it. I will probably need to write a makefile, right ?
21:51:22evilnicktaylor___: Please please please please please can you let us all out of our misery and say which generation you are referring to?
21:51:47*taylor___ already said ALL
21:51:58GodEaterand we already told you it DOESN'T apply to all
21:52:01kugelpetur: thanks. I don't quite know where the bug is
21:52:12BigBambitaylor___: ALL WHAT!
21:52:16kugelsetting the backdrops is a different function, not affected by the patch
21:52:22BigBambitaylor___: certainly not all players, nor all ipods
21:52:51taylor___GodEater: How can you know when you don't know yet what I am talking about? And if everyone could stop yelling, I could explain this MUCH faster. Thanks
21:53:08GodEaternot all ipods even support notes
21:53:23*BigBambi bangs his head on the desk in disbelief
21:53:24domonokyMT: yes a little makefile might be needed, but that isnt too hard :-)
21:53:58 Join webguest31 [0] (n=568032ad@gateway/web/cgi-irc/
21:53:59taylor___every ipod that can view notes has this, got it?
21:54:02taylor___can I go on?
21:54:14gevaertstaylor___: *nobody* has been stopping you
21:54:21taylor___which is roughly 90% of ipods
21:54:49rashertaylor___: Please don't use the enter key as punctuation.
21:54:55BigBambiI wish you would stop
21:54:57linuxstbMT: You might want to look at the apps/codecs/libffmpegFLAC directory - that is the FLAC decoder that I extracted from ffmpeg a few years ago, with a Makefile (Makefile.test) used to build it as a standalone test program. main.c is my standalone test proram.
21:54:59MTdomonoky : thanks :)
21:55:24taylor___Ok, first you create a notes doc
21:55:28MTlinuxstb : ok, will look there. Thanks.
21:56:00linuxstbMT: And yes, you might (at first) need to copy a lot of files from ffmpeg, but you should try and clean them (i.e. delete any unused code from them).
21:56:12peturkugel: I'll have a look
21:56:18 Quit casainho_ (Remote closed the connection)
21:56:56webguest31I think the 3.2 draft release notes have a contradiction: it says at the top that USB has been taken out for the portalplayer devices as it isn't stable enough. It then explains that USB has been enabled on most targets since 3.1
21:57:17rasherwebguest31: clearly I forgot to remove some bits
21:57:19linuxstbwebguest31: We're just teasing the users...
21:57:39gevaertswebguest31: thanks for spotting this :)
21:58:13MTlinuxstb : I was thinking that later I could rewrite the decoder to decrease the number of files a bit.
21:59:03 Quit jaykay ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
21:59:47linuxstbMT: Yes. I wouldn't think too much about that yet though - just get something working. Although it would be a nice "bonus" if you could offer your conversion to fixed-point back to ffmpeg. Also, check the ffmpeg-devel mailing list archives - I think I remember someone posting some (unfinished) patches to convert cook to fixed point.
22:00:34 Quit _Auron_ (Read error: 104 (Connection reset by peer))
22:01:59 Quit webguest31 ("CGI:IRC (Ping timeout)")
22:02:26archivatorI guess I might as well ask here - can anyone give me a plausible reason why I'm getting spikes in the FFT of the actual sound output -
22:02:43 Join _Auron_ [0] (
22:03:02rasherarchivator: And you're playing regular music?
22:03:12archivatorProblem is, I don't have the needed skillz to decipher the implementation. Would be helpful if I knew what algo they're using.
22:03:27archivatorrasher: tried several songs - similar effects.
22:03:41 Quit Llorean (Read error: 104 (Connection reset by peer))
22:03:47rasherarchivator: Just checking that you weren't playing some sort of generated signal
22:03:55kugelpetur: that would be nice
22:04:02 Join Llorean [0] (n=DarkkOne@
22:05:12 Quit raphi ("leaving...")
22:06:00MTlinuxstb : by the way, could you please give me an idea about how packets are further divided to sub-packets ? just some pointer would be fine.
22:06:30 Quit bs66_ (Read error: 110 (Connection timed out))
22:07:01 Quit XavierGr (Nick collision from services.)
22:07:12 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
22:07:40linuxstbMT: Sorry, I've no idea. I never got that far with my work...
22:11:14kugelpetur: can you try adding a display->clear_display before the return line in gui_wps_display?
22:11:15 Join BigBilly [0] (i=185b52cd@gateway/web/ajax/
22:11:20 Quit GodEater (Remote closed the connection)
22:11:35 Part BigBilly
22:12:18MTlinuxstb : never mind. When testing the decoder what I'll need to do, generally is : call its init() function, go through a loop which sends audio frames to the decode_frame() function, till the end of the audio data, and then call close() to clean up, right ?
22:12:20kugelpetur: or, maybe even better, move the one that's already in that function down
22:12:42 Quit taylor___ ("Leaving")
22:12:45 Part archivator
22:13:12Bagderanyone has anything in particular pending before release?
22:13:50rasherBagder: I was hoping to be done with a 64bit version of rbutil, but it looks like the compile will take ages.
22:13:55 Join bs66_ [0] (n=sysuser@
22:14:13Bagderthe 3.2 build will take >30 mins anyway
22:14:38rasherIt's been going on for hours. Could be done any minute of course, but I don't think so
22:14:43LloreanJust to verify - USB is disabled and the sorting option has been removed?
22:14:50LloreanWith the default back to ASCII for this release?
22:14:56gevaertsUSB is disabled. I don't know about sorting
22:15:10rasherkugel: Has natsort been disabled?
22:15:18kugelnot by me
22:15:29rasherSo I would guess it's still there.
22:15:55kugelwell, it wasn't my idea, and I'm not entirely convinced of the idea
22:16:10kugelso I didn't/don't feel responsible for deactivating it
22:17:03 Quit merbanan ("Leaving")
22:17:05*rasher wasn't involved in the discussion but was under the impression that it had been agreed to disable it for the release
22:17:10bluebrotherwell, the feature isn't stable in terms of how it's supposed to work
22:17:25*BigBambi also thought it was agreed to disable
22:17:36kugelwell, then let's do it
22:18:14rasherBagder: Do we want the date and revision in the version?
22:18:38BagderI don't have a strong opinion either way
22:18:45LloreanI'm voting against it.
22:18:54LloreanThe revision number is next-to-useless anyway since it's a branch.
22:19:04rasherI think it looks nicer without, and it's not like it's terribly important since it's ... yes, a branch
22:19:06kugelis it enough to deactivate the setting in the settings menu? (it could prossibly be activated then using config.cfg), or revert the initial commit entirely?
22:19:18Lloreankugel: Deactivate the setting, and revert the default to ASCII
22:19:20BigBambiIt should just be v3.2 IMO
22:19:21Bagderok, "3.2" it is
22:19:35rasherkugel: It's both enough and better (less change of breaking stuff and people who have already enabled it won't notice)
22:19:53Lloreanrasher: Well, they will. Since the default was "numeric" most people won't have it visible in their config.
22:19:55 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
22:20:07rasherLlorean: Good point. Oh well.
22:20:28rasherBut I don't think the code should be removed, just for fear of breaking stuff
22:20:29LloreanStill, removing the setting and fixing the default is probably the simplest way
22:20:30bluebrotherthat's even better, so no accidential turning on by an svn config file
22:20:32 Join Munkie [0] (n=29b13251@gateway/web/cgi-irc/
22:20:36 Join taylor_ [0] (
22:20:54Lloreanbluebrother: Well it could've been nice to have people currently using it not have it vanish with an upgrade to 3.2
22:21:12MunkieWhen is the movie chip on the ipod video going to be supported?
22:21:35taylor_BTW - I'm not sure if cmwslw already told you, but the annotations are finished for the HW specs
22:21:43rasherMunkie: When someone writes the code for it.
22:21:51bluebrotherLlorean: well, if the release notes tell the feature has been disabled I'd expect it to be disabled when using the release, regardless of the old setting
22:22:07LambdaCalculus37So are we ready to release yet? :)
22:22:14Lloreanbluebrother: True. I just don't expect most people upgrading from SVN to read them (of course many of those will skip the release version anyway)
22:22:15rasherbluebrother: They don't mention the feature at all (currently)
22:22:17 Quit XavierGr ()
22:23:39kugelhm, seems like a nice way to not upset people liking the sort anyway too much, too
22:24:11kugeljust in case someone comes here and complains
22:24:28LloreanIf they complain they can just use SVN builds.
22:24:28gevaertsthey will complain about usb first, don't worry
22:24:35 Quit Munkie (Client Quit)
22:25:10kugelLlorean: well, if they want a release because of the stable-ness, getting a current build is probably not an option for them
22:25:12 Quit Horscht ("Verlassend")
22:25:14kugelwho knows
22:25:21rashergevaerts: Feel free to put a public apology in the release notes
22:27:25 Quit taylor_ ("Leaving")
22:28:00kugelso, should it be in the config file options part of the manual then?
22:28:12gevaertsprobably, yes
22:28:22rasherThat's the truth, so it should
22:28:54kugelbut I still remove the normal manual description of it
22:29:31 Join Ubuntuxer [0] (
22:29:57 Quit midijunkie ("?(???~•~)?")
22:31:31 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
22:34:04rasherBagder: building?
22:34:30Bagdernot yet
22:35:14gevaertswe still need the sort commit I guess
22:35:41rasherAh right
22:35:58 Join taylor_ [0] (
22:36:06Bagderok, I'll get to work
22:36:37JdGordon|so sort and usb were both disabled for 3.2?
22:36:41BagderAt revision 20501
22:36:45taylor_n1s: sorry I didn't answer your question. As I said before, I keep cutting out because of the power :(
22:36:57kugelBagder: damn, 20500 looked nicer :(
22:36:57 Quit taylor_ (Client Quit)
22:37:34 Join taylor_ [0] (
22:37:40kugel20502 would look good too, some one has a commit?
22:38:36taylor_Ok, so say you have a notes file with this in it: <a href="kklklk/ >. Thats obviously an error in your file, so all the 'debug' mode will do will pop up a blank screen that says something like: Bad link error(5)
22:38:53 Quit Thundercloud (Remote closed the connection)
22:39:02 Join akur [0] (
22:39:08 Part akur
22:39:11taylor_I thought when Apple calls it 'debug' mode, it would actually be helpful :P
22:39:37n1sso it is some kind of debugging for the notes viewer only?
22:40:00 Join akur [0] (
22:40:04 Part akur
22:40:06taylor_yeah basically, but I didn't know that when I first checked out Apple's docs
22:40:26taylor_I actually had hope for a few seconds
22:40:37n1sit is documented?
22:41:30taylor_place <ERRORS> at the beginning of a notes doc, thats it
22:41:40taylor_thats all you have to do
22:42:15*amiconn points Bagder to ReleaseChecklist
22:42:20taylor_of course, you will need to purposely place an error (like bad link) in a notes file for it to notice
22:42:45Bagderamiconn: why?
22:43:01amiconnWe want new flash packages...
22:43:22UbuntuxerHi, I have written a patch for the games brickmania, rockblox and snake. The patch uniforms the menu and add some new features. e.g I have added in brickmania and rockblox a highscore list. Please could someone from the developers look at the code and tell me and help me to improve the code. Thank You. I have problems to upload a comment with the code in the tracker, so I have uploaded it here.
22:43:47Bagderamiconn: I'm not the suitable person to fix them
22:43:57LambdaCalculus37Ubuntuxer: Can you make a patch and post it to Flyspray, please? And give us the ticket number when you do.
22:44:05taylor_Ubuntuxer: cool!
22:44:22UbuntuxerI can't upload something, I will try it again
22:44:28BigBambiUbuntuxer: What problems with the tracker? It will get lost of not on there
22:44:39amiconnThe necessary steps are mentioned, and the order of steps is not arbitrary...
22:44:53*rasher wonders why amiconn hasn't done this
22:45:20UbuntuxerTracker 9953
22:45:40amiconnBecause I never know when the release actually happens, and it shouldn't be done too much in advance
22:45:54rasherIt was announced on the -dev list..
22:46:16rasherAnd I don't see why it'd be bad to do it a bit in advance in the 3.2 branch
22:46:29UbuntuxerMy patch is based on all_changes.patch from mud, which find also in this tracker
22:46:34kugelamiconn: so you wait until Bagder already has his finger on the big red button?
22:47:32Bagderamiconn: let me also remind you that it says "if" we want new flash packages, and I haven't seen anyone saying we want them
22:47:43Bagderand yes I asked
22:47:51rasherThere have been 2 commits in the past 24 hours, both in apps/lang/ ...
22:48:10amiconnWe want them since the last release already. I clearly mentioned that, several times
22:48:26BagderI haven't taken notice of that
22:48:32Bagderhence I asked
22:48:47Bagderon the -dev list
22:49:00Bagderabout what's left before relase
22:49:24*kugel doesn't see the question
22:49:52 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.5b4pre/20090321055010]")
22:50:17amiconnThat was about 11 hours ago. Who checks mail that ofetn?
22:50:29JdGordon|you mean you dont check every 3 min?!
22:50:31amiconnBut I've been around in IRC
22:50:50Bagderwell, you didn't reply to any of the mails about the release
22:50:51rasherHe also asked 30 minutes ago
22:50:57kugel11h is *often*?
22:51:22*amiconn usually checks mail about once per day, sometimes less
22:51:36kugelisn't checking mails a matter of running thunderbird in the background the whole day anyway?
22:51:39JdGordon|your doing it wrong! :D
22:52:04kugelJdGordon|: I agree
22:52:08amiconnBut anyway, what to do now?
22:52:47amiconnThe bumping of LATEST_BOOTLOADER_VERSION must be done before the main builds are made.
22:53:05BagderI can restart the builds
22:53:51JdGordon|on that topic... we should do a simliar thing for other targets to make sure old bootloaders are killed off
22:54:19rasherJdGordon|: You mean refusing to boot with an old bootloader? Doesn't sound wise...
22:54:33rasherwhen there's no technical reason it wouldn't work, that is
22:54:43 Quit Ubuntuxer ("Leaving.")
22:54:46Lloreanrasher: Maybe BOOTLOADER_MINIMUM_VERSION?
22:54:47rasherDoing it just to force users to upgrade is stupid.
22:55:14JdGordon|there has been cases where the old bootloader caused problems
22:55:22Nico_PJdGordon|: about your comment on FS #9795: I agree the process of skipping is convoluted, and I've been wanting to simplfy it for some time
22:55:22JdGordon|or just a waring if there is a newer version or something
22:55:28amiconnBagder: Only the archos buuilds will need restarting
22:55:36taylor_Has anyone gotten a sansa to fully work with itunes yet? I'm just curious on the progress of that.
22:55:37amiconnI'm preparing the necessary commit
22:55:47LambdaCalculus37Bagder: Is 3.2 going to display "Version 3.2" as its version number?
22:56:20JdGordon|Nico_P: I started a email last night but didnt send it because im not sure I'm the best person to get discussion going.... bassically we need to talk about how we want swcodec to evolve and get cleaned up
22:56:24Nico_PJdGordon|: I had made a graph of how it works:
22:56:32mcuelenaeretaylor_: AFAIK that should apply to all Rockbox SW-USB capable devices, not only the Sansa's
22:56:44LambdaCalculus37Bagder: Okay, because I wondered why 3.1 didn't display "version 3.1" anywhere.
22:56:49taylor_Very nice.
22:56:56Nico_PJdGordon|: you're perfectly welcome to make suggestions
22:56:59BagderLambdaCalculus37: that was just me not doing it right
22:56:59taylor_I'm just asking because I might get a Sansa
22:57:17pixelmaBagder: and have a small enough voice file at least for the releaso on the FM/V2-Recorders
22:57:27pixelmarelease too
22:57:35rashertaylor_: "Fully work" is pushing it. It did mostly work though.
22:57:48taylor_And if I do get a sansa I am hoping on porting rockbox to it. Or at least helping, if there is help needed?
22:57:50Bagderpixelma: well, that'd require that someone would work on making it happen...
22:57:54gevaertstaylor_: for the record, I'm still against doing this in regulare builds
22:58:03Lloreantaylor_: These experiments are being done on a well supported sansa.
22:58:11Lloreanand they apply to all software USB targets, not just Sansas.
22:58:33taylor_good to know
22:58:40pixelmaBagder: well, what do you mean? Speed and quality settings can be changed in the build environment
22:58:49JdGordon|Nico_P: is where i stopped... i'm not sure how that sounds seen as for sure people know more about how it shuold and does work than me
22:58:57Bagderpixelma: yes, but changed to what?
22:58:59*JdGordon| will check out that pdf tonight
22:59:33Nico_PJdGordon|: it's very short and rather simple
22:59:48Nico_Pseems you forgot to mention the buffering thread btw :)
22:59:54rasherpixelma: it's *not* easy with festival
23:00:08rasherUnless you resort to compressing more
23:01:15pixelmaI don't know festival but if that's the only thing that can be done currently, I'd think it's better than providing a non-working voice file
23:01:26rasherI agree.
23:01:31LloreanMe too
23:01:35pixelmaunless it's not understandable at all
23:01:46rasherpixelma: you basically have to reach into the bowels of festival and write Lisp scripts to configure it
23:01:53rasher(not kidding)
23:02:43JdGordon|Nico_P: right.. where does buffering really fit in? is there any reason it shouldnt be merged with the playback thread once the playlist part of that thread is moved to where it shuold be?
23:02:57amiconnBagder: Committed. When the archos builds including this change are done, I'll need them for the final version of the packages
23:03:32amiconnMeanwhile I'll build the bootloaders and check the rombox addresses
23:03:42Bagderbuilds restarted
23:03:43 Quit ibseco (Read error: 60 (Operation timed out))
23:04:07Bagderthe second time flies so much faster
23:04:49 Quit LambdaCalculus37 ("Food time is now")
23:04:57pixelmaor - don't provide a voice file - or maybe it already helps lowering that one setting that needs quite some RAM, "max files in directory" I believe it was
23:05:26Nico_PJdGordon|: buffering could probably be merged back into the same thread as playback, but I'm not sure how useful that would be
23:06:45JdGordon|1 less thread to keep in sync
23:07:18JdGordon|especially if we can move to useing the ata idle callback to do rebuffers instead of watermarks (unless thats back in already?)
23:07:43Nico_Pthere are watermarks IIRC
23:08:26Nico_PI think the main issues are between playback and playlist though
23:09:11JdGordon|and ui->playback more generally
23:09:41Nico_PIMHO audio_check_new_track() is the place where it's worst because it has to handle a lot of different cases
23:10:11 Quit evilnick (" ajax IRC Client")
23:10:21 Quit Schnueff ("leaving")
23:11:47Nico_Prefactoring all that is going to be a HUGE task. we better have a solid design before we start actually doing it
23:11:48JdGordon|can you put up thoughts on the ml so we can get discussion and hopefully fixes happening?
23:11:56amiconnBootbox for v2 does not build....
23:12:03amiconn(in the release branch)
23:12:03*n1s wonders if Bj?\195?\182rn will fix the unicode stuff on the frontpage eventually ;P
23:12:23JdGordon|Nico_P: yeah, for sure :)
23:13:03Nico_Pn1s: hahaha. I think that's a problem in the way svn handles the emails I think
23:13:16kugelNico_P: that's not JdGordon's business normally though
23:13:21kugelhaving a solid design ;)
23:13:54Nico_Pn1s: it's fine in
23:14:10*JdGordon| punches kugel in the face
23:14:21Nico_PJdGordon|: I will try to write some thoughts down, but not today. I have an RC5 implementation I need to finish...
23:14:26JdGordon|anything serious of course has to be designed properly
23:14:28n1sNico_P: yeah, i've noticed :)
23:15:49 Join moos [0] (i=Mustapha@rockbox/staff/moos)
23:16:32kugelscorche|sh: banning ain't nice :(
23:16:36Nico_PBagder: the version string at boot below the logo will be "3.2-rXXXX", right?
23:17:06Bagder3.2 only
23:17:20Nico_Pah yes, I saw the second commit too late :)
23:17:26 Quit bertrik ("Leaving")
23:17:39Bagder28 builds remade now, soon they're all done
23:17:57Bagderall bin zips at least
23:18:32amiconnBagder: Lools like it would be a good idea to have all bootloaders built in the build table...
23:18:47Nico_Pso we're keeping USB and natsort for the next release it seems?
23:18:49***Saving seen data "./dancer.seen"
23:19:00amiconnfmrecorder and recorderv2 *should* be identical apart from the scrambling, but one builds, the other doesn't :(
23:20:04 Join robin0800 [0] (
23:20:16amiconnLast time I made flash pacakges both of them built
23:20:35 Join casainho [0] (n=chatzill@
23:21:33domonokymabe we should make sure the voicefiles work for archos recoder (are small enough) this time :-)
23:21:51casainhohello again :-)
23:22:26casainhoI just commited the Lyre patch again. I would like to know if the license on the "at91sam9260.h" file is ok for Rockbox....
23:22:47pixelmadomonoky: you didn't read logs ;)
23:23:35Bagdercasainho: the license seems BSD'ish and that's fine
23:23:47Bagderpacked with // comments though...
23:24:26casainhoBagder: okok, thanks. I think that file is the only that is with wrong format style... and is big... :-(
23:24:27 Join Thundercloud [0] (
23:24:52casainhobut I will start working on it :-)
23:24:57Bagdercasainho: a decent editor has a macro function that fixes it easilt
23:25:21Bagdersearch for //, replace with /*, go to end of line add */, next
23:25:23JdGordon|vim not being one of them :D
23:25:40Bagderis that really an editor? B)
23:25:49Nico_Pvim can do that perfectly fine
23:26:08domonokyor just a editor which supports search&replace with regexp :-)
23:26:23*amiconn blames jhMikeS
23:27:14bluebrothervim :)
23:27:23domonokya true hacker would ofcourse write a rockbox plugin todo this task :-)
23:28:04casainhoBagder: thanks, I already that that search and replace :-)
23:28:57linuxstbcasainho: Still a few tabs in your patch - and crt0.S
23:28:57*Bagder starts the manual builds
23:29:50*gevaerts reminds Bagder of the need to hold the archos manuals (or rebuild them later)
23:30:10Bagderit'll be easier to rebuild them later
23:30:16 Quit petur ("Zzzzz")
23:30:32n1seeh, should we perhaps mention in the release notes that users will have to rebuild their databases(?|, they will no?)
23:30:39amiconngevaerts: why?
23:30:52rashern1s: if so, then yes that's probably a good idea
23:31:05LloreanWhy do they need to?
23:31:16kugelso, we release 3.2 on 23rd of march, in Hawaiian time? ;)
23:31:24gevaertsamiconn: uhm, because I didn't check...
23:31:26*Bagder starts the voice builds too
23:31:46rasherIt's still the 23rd for 90 minutes
23:31:54 Join webguest39 [0] (n=4f445323@gateway/web/cgi-irc/
23:31:57n1sLlorean: i think the FS #9977 patch caused that, kugel should know though
23:32:01amiconnhumm-hommm :\
23:32:14Lloreann1s: Ah, right
23:32:30*Llorean has many hours of 23rd left.
23:32:58kugeln1s, Llorean: yes, there's a quite increase in ram usage due to this
23:33:15kugelrasher: in gmt, yea
23:33:36rasherWhich is what counts!
23:33:52 Join lymeca [0] (
23:34:02n1skugel: from the fs task the increase didn't seem very large but it's correct that databases must be rebuilt, right?
23:34:14kugelthat too, yes
23:34:56n1sso where to put that in the release notes, rasher any suggestions?
23:35:42linuxstbn1s: In the "How to upgrade" section?
23:35:44rashern1s: Good question, either What's new or How to upgrade I'd say
23:36:07casainholinuxstb: I couldn't find any TAB on crt0.s −− but I found one ;-)
23:36:33linuxstbcasainho: There is one...
23:36:33 Quit JdGordon| (" ajax IRC Client")
23:36:34n1sguess I'm tired, didn't even see the "how to upgrade" section :/
23:38:24casainholinuxstb: I couldn't found it again. Please say what is the line
23:39:00linuxstbcasainho: mov sp, r1
23:39:06linuxstb(after the r1)
23:40:37 Quit webguest39 ("CGI:IRC")
23:42:44 Quit taylor_ ("Leaving")
23:43:17casainholinuxstb: thanks. I got it ;-)
23:46:39 Quit Llorean ("Leaving.")
23:47:52 Join Llorean [0] (
23:50:21 Join tessarakt [0] (
23:52:31 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.7/2009030422]")
23:54:09rasherBagder: did you move rbutil to the download servers?
23:54:10 Quit tessarakt (Client Quit)
23:54:17 Join tessarakt [0] (
23:54:19Bagderno, where are they?
23:54:50Bagderah found them
23:54:56rasherBagder: and
23:55:48*rasher wonders why bluebrother tars a single file
23:55:49 Join taylor_ [0] (
23:56:30taylor_Does anyone know if any of the ipod's cpus are related to the Sansas? (I know they are both ARM)
23:56:44rashertaylor_: Check the wiki.
23:57:18rasherBagder: Do we want a tarball as well? If so,">
23:57:38taylor_ok thanks, Basically Im wondering if it would be anywhere close to easy turning the Sansa emu you guys have into an ipod one?
23:57:38Bagderyou mean with the code?
23:57:57BagderI guess it can't hurt
23:58:19linuxstbWhat's needed? rbutil/ apps/ and tools/ ?
23:58:24bluebrotherrasher: does it include the parts of rockbox sources that are needed?
23:58:24Bagderrasher: the url doesn't work
23:58:25taylor_Are those responses directed to me?
23:58:29linuxstb(or rather, parts of apps and tools) ?
23:58:36rasherbluebrother: yes
23:58:47Bagdertaylor_: it depends a little on what sansa you speak of
23:58:51rasherlinuxstb: apps/codecs/libspeex and tools/rbspeex and a few other files in tools
23:58:57rasherBagder: hm hang on

Previous day | Next day