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 2007-05-29

00:00:34Domonoky_somebody should buy preglow a gigabeat S :-)
00:01:03amiconneven more overkill hardware than the gigabeat f and x
00:01:16saratogawas just going to ask what you thought of this thing
00:01:37saratogastill, its pretty cheap
00:01:45 Join Pyromancer [0] (
00:01:54amiconnDoes it have a decent lcd? The f and x don't
00:02:34 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
00:02:35saratogathey're going around the 100-120 mark on ebay
00:02:56*amiconn rather pays more for a decent thing, like the h1x0.
00:03:07 Join sejanis [0] (
00:03:41 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
00:04:17saratoga"The screen is a 2.4” QVGA 320 x 240 pixel screen. I would compare the screen to something you would find on the latest Windows Mobile Pocket PC."
00:04:31preglowamiconn: sure, me too, but it's also fun with something new
00:04:51Joelybluebrother, you still there??
00:04:56saratogastill, i hope other players go this way in the future, instead of using proprietrary DSP cores
00:04:58Lloreansaratoga: As far as I'm concerned, most WIndows Mobile devices I've seen have worse screens than the daps I own
00:05:02bluebrotherJoely, yup
00:05:12amiconnsaratoga: I'm not interested in high resolution, or colour, or sth like that
00:05:30bluebrotherso why not port Rockbox to Windows Mobile and run that on the S? ;-)
00:05:30amiconnThe most important thing is a screen that is readable w/o backlight
00:05:42sejanisIriver H10 shows in windows my computer as an "audio device" with no drive letter to load the bootloader and firmware for rockbox, does anyone know a workaround?
00:05:44saratogaah, was going to ask what you looked for
00:05:45Joelybluebrother, i still have to edit that part of the wiki for a while...
00:05:57amiconnThe only rockbox targets which fulfill that are the non-colour ones, and as I recently learned, ipod video and nano
00:06:14Lloreansejanis: Don't use MTP mode, you have to be in UMS mode
00:06:30amiconnI even prefer the archos player screen over any of the other colour screens
00:06:31saratogai mostly just keep the player in my pocket anyway, so screens aren't a big deal to me
00:06:32bluebrotherJoely, looks like a nice start to read at least.
00:06:58bluebrotherthere are color screens around which you can read without backlight?
00:07:05saratogamy old gameboy had one
00:07:09preglowmy nano isn't half bad
00:07:14preglownot as good as the h120,but not bad either
00:07:57amiconnbluebrother: I already knew that there is this kind of colur lcds... and now know that the ipod video and nano fall into this category
00:08:09*amiconn has to check the video's display in direct sunlight
00:08:32saratogathat reminds me, I still think the Nintendo DS would be a fun target, since theres CF and SD card adapters for it, and the dev kit already has GPLed drivers
00:08:33 Quit My_Sic (Read error: 104 (Connection reset by peer))
00:08:37saratogapointless, but kind of fun
00:08:45preglowsaratoga: i've been thinking about it too...
00:08:48amiconn...a situation where all of H3x0, X5 and my mobile phone become totally unreadable, unlike all my archoses and the H1x0 and ipod mini
00:08:54 Quit sejanis ()
00:09:07saratogapreglow: you pick up a flash cart for one yet?
00:09:28preglowsaratoga: don't even have one, but it'd be one of the first things i did pick up
00:09:45saratogai jsut got a DS, lovely piece of hardware
00:09:52saratogacan't believe how much fun they are
00:10:02preglowi can, which is why i don't have one
00:10:14 Part maffe
00:10:23preglowgonna try one next weekend
00:10:46 Quit kokeradio ()
00:11:05saratogayeah there is a very real danger of your life being consumed by the DS
00:11:16saratogait really is that much fun, and you can bring it anyway and just play
00:11:20bluebrotherreally? Hasn't applied to me
00:11:29 Quit jhMikeS (Nick collision from services.)
00:11:34 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
00:11:40bluebrotherbut maybe that is also caused because I only have this brain jogging cartridge ;-)
00:12:01 Join My_Sic [0] (
00:12:20 Join maffe [0] (n=maffe@
00:12:35 Quit maffe (Remote closed the connection)
00:13:01 Join maffe [0] (n=maffe@
00:14:12 Quit jhulst (Read error: 54 (Connection reset by peer))
00:14:35preglowsaratoga: definitely looks like stock ffmpeg has a very decent chance of running on this thing, yes
00:15:20Domonoky_what about lame as an encoder :-)
00:15:56preglowit has an on-chip mpeg4 encoder
00:16:08preglowguess who has silicon to burn
00:17:09saratogai PMed toffe asking if hes had a look at the exploit
00:17:23saratogawhere there any other people with S models who were programmers?
00:18:18moosptw is the main guy
00:18:40moos"main"=the only one for now here :)
00:19:08moosnight all
00:19:12 Quit moos ("Glory to Rockbox")
00:19:30 Join Thundercloud [0] (n=thunderc@
00:22:29 Quit Daishi ("Client exiting...")
00:23:11 Quit saratoga ("CGI:IRC")
00:23:43chrisjs169what does 'PANIC Stkov sd' mean?
00:23:49preglowstack overflow
00:23:59chrisjs169so stack overflow on sd?
00:23:59preglowsd? don't know. sansa?
00:24:10preglowusing sd patch?
00:24:23chrisjs169yeah, but i don't have an sd card
00:24:37amiconnThat's odd then
00:24:46 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
00:25:09amiconnThe only situation where the sd thread needs signifiant amounts of stack should be a (re)mount of an sd card
00:25:28amiconnUnless the sd thread does something weird that the mmc thread on ondio doesn't
00:25:34Lloreanamiconn: The internal flash is "sd" too, I think
00:25:41LloreanAt least, I have an sd stack without the sd patch
00:25:46chrisjs169well, it didn't exactly happen to me (i don't have a Sansa with me because Sandisk is taking forever...) but it was a problem someone mentioned on ABi (they don't have a microsd card though)
00:25:49amiconnLlorean: That doesn't matter, since that can't be hotswapped
00:26:08chrisjs169( )
00:26:25Lloreanchrisjs169: The ABi people should learn to properly interact with the developers of the software, ie coming here and actually reporting bugs properly, I think.
00:26:50Lloreanamiconn: Yes, but could other activities mean a lot of use of the 'disk' stack?
00:26:54chrisjs169Llorean: yeah
00:27:13amiconnThe sd thread doesn't do much when no swapping is involved
00:27:27amiconnWithout hotswap, this thread doesn't even exist
00:27:40Lloreanamiconn: I don't have the sd patch installed, I have an sd thread on my sansa at 75%
00:27:51 Join Rob222241 [0] (
00:27:55 Quit lee-qid (Read error: 54 (Connection reset by peer))
00:27:59*amiconn wonders what it does, then
00:28:06 Join lee-qid_ [0] (
00:28:07 Nick lee-qid_ is now known as lee-qid (
00:28:11 Quit Pyromancer ("Leaving")
00:28:18chrisjs169if Llorean is right about the internal memory being sd based (I believe it is too) then is it possible that could be the cause?
00:28:48amiconnThe ondio internal flash is also mmc based, yet the thread is only necessary to handle hotswap
00:28:59LloreanIt could just be poor naming of a thread.
00:29:54*amiconn must be cautious with slapping forehead, in case he watches the channel ;)
00:30:00Lloreanchrisjs169: Though as far as I know, the stkov you're talking about doesn't happen to anyone using the official builds
00:30:22LloreanI know it's certainly never happened on my Sansa
00:30:23 Join pixelma_ [0] (i=pixelma@rockbox/staff/pixelma)
00:30:24amiconnThe thread also handles ata notification...
00:30:46 Quit pixelma (Nick collision from services.)
00:30:47 Nick pixelma_ is now known as pixelma (i=pixelma@rockbox/staff/pixelma)
00:31:20toer * Initializes the plugin. This is called from the main application
00:31:20toer * when the plugin is first loaded.
00:31:21chrisjs169Llorean: it's never happened on mine, but being an 'odd' error that someone mentioned, I figured I'd ask and see what it is (I use a custom build with the SD patch, but never have seen it)
00:31:29 Quit My_Sic (Read error: 104 (Connection reset by peer))
00:31:44pixelmapetur: on my Ondio's recording screen there is one empty line left now ("worst case line-in recording")
00:32:10Lloreanchrisjs169: Until it happens in SVN builds, it's very likely just a symptom of one of the patches.
00:32:24Lloreanchrisjs169: Meaning it's not a bug anyone here necessarily can fix.
00:32:29chrisjs169Llorean: ok
00:32:42LloreanEspecially if you have the SD patch installed.
00:33:03peturpixelma: well it's some time ago since I ran one in the sim, I thought I remembered the display being full
00:33:29 Part landan
00:33:51pixelmawell a bit of info was moved to the status bar some time ago
00:34:22pixelma(like quality and bitrate)
00:35:23 Quit Rob2222 (Read error: 60 (Operation timed out))
00:36:59 Quit austriancoder ("Kopete 0.12.4 :")
00:39:00chrisjs169Llorean: It seems to have happened to another user on ABi as well...
00:39:29Lloreanchrisjs169: Using an unmodified SVN build?
00:39:41chrisjs169I'm getting the impression, but they didn't say
00:40:08LloreanWell, feel free to ask them.
00:40:17chrisjs169I did
00:40:19LloreanBut I'm not sure why they're reporting bugs there if they're using official builds.
00:40:32*petur gets warned by his h300 that he removed an unenumerated usb device :)
00:41:02Lloreanpetur: So, progress?
00:41:04chrisjs169Yeah, but I guess they figure it'd be more active there then on Flyspray
00:41:31peturLlorean: yes, the usb stick now flashes twice :)
00:41:53Lloreanchrisjs169: That doesn't make any sense. How's a bug in the official version supposed to get fixed if people don't report it?
00:41:55peturbut I only see something happening when I remove it :/
00:42:11Lloreanpetur: Still, somewhat exciting news. :-P
00:42:40peturwell I found some time to work on it again :)
00:42:50chrisjs169Llorean: I know it doesn't really make sense, but maybe they figure it's common, even though Google shows nothing
00:43:31Lloreanchrisjs169: In the thread you're talking about though, he did say "it was working with sd earlier" (paraphrased) which suggests to me at least that he was using the patch.
00:43:55 Quit ender` (" To get as fewest unhappy people as possible, always bully the same ones.")
00:44:19 Quit Thundercloud_ (Connection timed out)
00:44:26chrisjs169Llorean: you're right, must have missed that part *now feels it may indeed be the SD patch*
00:45:44 Join BHSPitLappy [0] (
00:47:28 Quit z35 (Read error: 104 (Connection reset by peer))
00:48:24amiconnchrisjs169: A very quick look at the sd patch tells me why it crashes... and reveals other strangenesses which need to be fixed before it can go into svn
00:50:10 Part krush1704
00:50:19amiconn(quick == in the range of seconds)
00:50:25mpeccoriniamiconn: n1s asked me to use the list widget in a patch I made. Everything's done now, but I don't seem to understand the "actions" part. I tried "case ACTION_STD_OK" hoping for the SELECT_BUTTON to trigger that action, but it didn't. can you give me a clue?
00:50:30 Join krush1704 [0] (
00:51:03mpeccoriniamiconn: I ask you because apparently, you commited the action.c file :p
00:51:13bluebrothermpeccorini, the button is dependent on the context you are in
00:51:19amiconnJdGordon is Mr. action
00:51:53mpeccoriniamiconn: is he around?
00:52:06mpeccorinibluebrother: I don't even know what a context is in this case :(
00:52:11amiconnAnd btw, select button doesn't tell much w/o mentioning the target
00:52:24mpeccoriniLOL, you're right, it's an iPod nano
00:52:35 Join jac0b [0] (
00:52:43chrisjs169so if quick == a few seconds, very quick == ?? I've been away from rockbox for a while, so I can't see why it crashes...
00:53:01scorcheJoely: ping
00:53:11mpeccorinibluebrother: I don't know what a context is but I'm using CONTEXT_LIST ;)
00:53:26 Part jac0b
00:53:27bluebrothermpeccorini, that page might be helpful:
00:53:43bluebrothermight be outdated, but at least it has some text about it ;-)
00:53:45 Join jac0b [0] (
00:53:53amiconnchrisjs169: Stack is too small as the sd stack has to handle remounts with hotswap
00:54:10amiconnCompare ata-e200.c with ata-mmc.c ...
00:54:16chrisjs169amiconn: ok
00:54:17mpeccorinibluebrother: thanks
00:54:18 Join Thundercloud_ [0] (n=thunderc@
00:54:24amiconnThe sd patch doesn't change this
00:55:22amiconndisk_mount needs quite a bit of stack
00:55:55amiconnThe other strangeness is that the sd patch replicates the 'remove card...' behaviour of the Ondio
00:56:31amiconnThat's not desired behaviour, it's a technical shortcoming on the Ondio that doesn't apply to the Sansa afaik
00:56:32Joelyhey scorche
00:57:56scorcheJoely: see PM
00:58:15bluebrotherdoes anyone in here know this debugger?
00:58:26jac0bI am getting an error when I "make" my build with the hotswap patch
00:58:33jac0bmake[1]: *** No rule to make target `export/hotswap.h', needed by `/home/jacob/r
00:58:33jac0bockbox/build/firmware/common/disk.o'. Stop.
00:58:33jac0bmake: *** [build] Error 2
01:00:01otihBagder: ping
01:00:11jac0bcan someone help?
01:01:45mpeccorinijac0b, export/hotswap.h is probably included in another file and is not in the -I path in the makefile for that file
01:02:30jac0bso do I need to change the make file
01:02:30mpeccorinijac0b so make tries to compile it by itself and obviously doesn't find a rule to "create" a .h
01:02:55mpeccoriniyes, you need to add a new -I including the folder where "export" is
01:03:12mpeccorinior change the #include to point to the file
01:03:47*amiconn thinks that just a full rebuild might help, that regenerates the dep- files
01:03:53amiconn* dep-*
01:06:57 Part Domonoky_
01:09:01 Join Llorea1 [0] (
01:09:14 Quit Llorean (Nick collision from services.)
01:09:16 Nick Llorea1 is now known as Llorean (
01:11:53 Quit Thundercloud (Read error: 110 (Connection timed out))
01:12:35jac0bmpeccorini: I can't find a hotswap.h only a hotswap.o
01:13:36 Join major_works [0] (
01:13:52major_worksjac0b: I ran into the same problem earlier today.
01:14:09major_worksWhen the patch was reposted a couple of days ago, a bunch of stuff was omitted from it.
01:14:40major_worksYou need to compare the latest version to the version before and paste back in the stuff that's missing in the new version.
01:14:45jac0bmajor_works: the one posted from today
01:15:22jac0bokay so I need to merge the 2 files togther
01:15:30major_worksBasically, that's right.
01:15:46jac0bhow is that done?
01:15:54major_worksIt patched and compiled fine for me after I did so.
01:16:52major_worksOpen both versions of the patch in separate Wordpad windows and compare them section by section.
01:17:22major_worksUse the new one as your final version and paste in the stuff from the older one that's missing.
01:17:29major_worksThen patch with that file.
01:18:15*petur updated his usbotg page. No big breakthrough yet :(
01:18:29 Quit petur ("Zzzzz")
01:18:46major_worksjac0b: OK? Capische?
01:19:52 Join isaac_w [0] (
01:20:31 Join donutman25 [0] (n=chatzill@
01:20:48*Nico_P is hoping for a (small) green delta
01:21:40 Join My_Sic [0] (
01:21:52 Quit My_Sic (Client Quit)
01:22:49preglowNico_P: what, but that _was_ worth implementing? :P
01:23:25***Saving seen data "./dancer.seen"
01:23:29 Quit major_works ("major_works has left the building...")
01:23:32bluebrothermoving cuesheets into a different folder removes clutter?
01:23:49Nico_Pbluebrother: from the file browser, yes
01:24:04linuxstbWhy not just make cuesheets not a supported filetype?
01:24:09Nico_PI know I could just have display set to music instead of all
01:24:10bluebrotherI tend to disagree ...
01:24:22bluebrotherlinuxstb was faster −− just wanted to say the same
01:24:26Nico_Plinuxstb: because there are supported
01:24:42linuxstbI know there's a cuesheet viewers, but I can't see the point in that either...
01:25:18amiconnThere are other filetypes in rockbox which are supported in a way but not included in the 'supported' view
01:25:25amiconn...e.g. the .talk files
01:25:37LloreanThere's somewhat of a history of providing a folder for storing "supplemental" files in Rockbox already
01:25:43LloreanSpecifically the playlist folder for the catalog
01:25:49Nico_Pwell at least part of my commit is useful : the duplicate code removal
01:26:17Nico_Pif really everyone is against the other part I'll remove it
01:26:24Nico_Pbut I don't see how it hurts anyone
01:26:35preglowi can tell you of one cuesheet feature you can replace it with :>
01:27:18pixelmaNico_P: well first - some yellow
01:27:26Nico_Psilly me
01:27:27bluebrothercuesheets are overrated anyway :-p
01:27:31preglowi'd rather keep the cuesheet handling simple myself, though, if you ask me it isn't really logical to look for cuesheets any other place than in the dir where the audio track is
01:27:57amiconnIt's not logical to look for cuesheets at all :P
01:28:27*Llorean finds the FmPresets pages have become ridiculously complex.
01:28:43Nico_Phaha I see cuesheet support is probably rockbox's most mocked feature :)
01:29:02*amiconn rather uses −−nogap on hwcodec than using cuesheets
01:29:05bluebrotherI never understood what cuesheets are good for on a dap
01:29:17linuxstbNico_P: Yes, but I've also forum posts where people say how much they like cuesheet support...
01:29:23pixelmaLlorean: any idea to make them better?
01:29:34linuxstbbluebrother: Simply so people can play the same files everywhere.
01:29:35bluebrotherLlorean, maybe something similar to rockbox-themes for fm presets?
01:29:41Lloreanpixelma: Well, it's all the individual .zip files, mainly
01:30:01LloreanSomeone adding a .fmr file is required to update many files, rather than just adding his text update to the page, then updating the one "total" .zip.
01:30:10Nico_Pbluebrother: now I can listen to chaptered podcasts in a nice way
01:30:13LloreanIt really doesn't need all the divisions it's given, I think.
01:30:13bluebrotherand a script could generate the zip file on the fly
01:30:15amiconn(especially now that I have a nice automating transcoding script) :)
01:30:18preglowi think cuesheet support is great
01:30:21Lloreanbluebrother: A script would make it even easier, yes.
01:30:29preglowit even turned out better than i expected
01:30:38 Quit lee-qid (Read error: 110 (Connection timed out))
01:30:39pixelmaLlorean: s/he he can start by just uploading the .fmr
01:30:47bluebrotherI might even be interested in looking into it. When I manage to finish this uni stuff :o
01:30:48Nico_Ppreglow: how's that ?
01:31:26Lloreanpixelma: I just think the process should be simple: 1) Download the complete .zip. 2) Add your FMR to the appropriate place. 3) Go to the corresponding wiki page for the .fmr and upload it, then overwrite the complete .zip
01:31:41LloreanInstead we may often have a "USA" zip that contains presets the complete one doesn't, or vice versa.
01:31:45LloreanThere's even a Texas zip.
01:31:48jac0bmajor_works: sorry its my g/f's b-day had to cut the cake, but thanks for the help
01:31:56pixelmaLlorean: I just wanted them to state when it was uploaded so someone knows that this specific fmr might not be included in the zip
01:32:07bluebrotherLlorean, I would prefer to be able downloading individual files
01:32:09chrisjs169having nothing else to do, I was looking at the sd stack overflow problem, and noticed that there's also a mmc_stack in addition to sd_stack. what's the difference?
01:32:18Lloreanbluebrother: Downloading individual .fmr files should be fine.
01:32:22Lloreanbluebrother: But I don't like regional compilations.
01:32:32LloreanUnless they're automatically updated on .fmr upload
01:32:39bluebrotherso, I think it would be best to automate that.
01:32:51preglowNico_P: seeking performed better than i would have thought
01:33:09Nico_Ppreglow: in speed or precision ?
01:33:18Lloreanbluebrother: Automation would make me just as happy.
01:33:18bluebrothereither update compilations on file upload, or (that is a bit more what I'm thinking of) give the user a list with all files and let him tick those he wants.
01:33:24pixelmaLlorean: the main concern was just that the list(s) got very long
01:33:26bluebrotherand then generate a zip file with all those files on the fly
01:33:36Lloreanpixelma: I'm all for dividing the pages up so it's readable.
01:33:41 Quit PaulJam (".")
01:34:04Lloreanpixelma: As I said, the only thing that makes me unhappy is that there are multiple .zip files that may or may not be in sync.
01:34:05 Quit Thundercloud_ (Read error: 54 (Connection reset by peer))
01:34:23*bluebrother puts a note to his todo list
01:34:27LloreanOne .zip file was already iffy enough, I thought, because an upload date just tells you when the last person added to it, not whether it's actually complete relative to the full list.
01:34:40*linuxstb wonders why people seem to be abbreviating place names to 8 characters in the FM presets
01:34:57bluebrotherman, will I ever get able cutting down that list before I'm really old?
01:34:59pixelmasurely I can understand that
01:34:59preglowNico_P: precision
01:35:25Nico_Pwell that's not my code anyway... it's all in the codecs
01:35:43linuxstbNico_P: I still get annoyed by the cue-sheet specific seeking behaviour though - i.e. not seeking back to the start of the current track when pressing PREV.
01:35:49Nico_Pbut I'm pleased you like it... funny how some people love it and other hate it
01:35:58preglowNico_P: how is it possible to hate it?
01:36:03preglowit's not even intrusive
01:36:08pixelmalinuxstb: guess most of it was my fault (still working from the assumption that they aren't allowed to be longer) :/
01:36:12Nico_Ppreglow: ask amiconn:)
01:36:26Nico_Plinuxstb: I am too but I really didn't find how to do it the right way
01:36:27preglowNico_P: he hates cuesheets, not the cuesheet support
01:36:45Lloreanpixelma: Perhaps the whole problem is more related to the fact that it's not manageable at all, anyway. =/
01:36:48preglowif you don't like cuesheets, i assume you keep the feature switched off
01:37:03Nico_Plinuxstb: the problem is mainly with VBR MP3 I think
01:37:12preglowi'd just like it to be always on and remove the setting
01:37:14Nico_Ppreglow: 100% agreed
01:37:42amiconn100% disagreed
01:37:47Lloreanlinuxstb: The wiki page actually says "(the name must be less than 8 characters long)"
01:38:01amiconnA waste of ram for no good... cuesheets are flawed by design
01:38:16pixelmaLlorean: that's outdated
01:38:17Nico_Pamiconn: I agreed on the "if you don't like cuesheets, i assume you keep the feature switched off"
01:38:18preglowstill doesn't prevent the fact that a lot of them exists
01:38:34preglowmp3 is also quite old and crummy, better codecs exist now
01:38:40preglowso we should remove support, right?
01:38:42LloreanOf course
01:38:43preglowwaste of good ram
01:38:58LloreanRemove support for "until the patent climate becomes less hostile to those using MP3 technology"
01:39:44amiconnpreglow: I don't say the feature should be removed... but the option to disable it has to stay
01:40:02preglowpft, what is an extra disk access these days anyway ;)
01:40:32amiconnStill, I haven't seen any cuesheets in the wild, only temporary ones (for ripping cds), and the one I created deliberately for testing the feature on archos
01:40:49preglowi've seen plenty, lots of people still don't know better
01:41:12idnarI've often seen CDs distributed as a single FLAC/wavpack/whatever file with a cuesheet
01:41:35 Quit bluebrother ("Leaving")
01:41:40idnarrather annoying, the first thing I usually do with it is to cut it up
01:41:43Nico_Pamiconn: I have now found a way to extract a cuesheet from a chaptered MP4 podcast... that's at least one major use I will have until real chapter support arrives (if it does one day)
01:41:48preglowamiconn: why are cuesheets flawed by design, btw?
01:42:00 Quit robin0800 (" HydraIRC -> <- Nine out of ten l33t h4x0rz prefer it")
01:42:01Nico_PI would never dream of cutting a podcast
01:42:05chrisjs169what's the difference between sd_stack and mcc_stack?
01:42:05Nico_Ptoo much of a hassle
01:42:19idnarah, yeah, something like a podcast is different, I guess
01:42:29linuxstbNico_P: Do you have any links to chaptered mp4 files I could download?
01:42:39preglowi don't want to cut anything, i want rockbox to be able to listen to my music as it is
01:42:49preglowif i have to pre-process anything, rockbox is wasting my time
01:43:02amiconnpreglow: Why shoudl I concatenate separate tracks and the have a list where the split points should be, rather than having the tracks separate as they should be
01:43:05Nico_Plinuxstb: sure.
01:43:18Nico_Pyou might not like the music though :)
01:43:18amiconnAlso, a bunch of smaller files is easier to handle than a single giant one
01:43:38preglowamiconn: that's a matter of preference, some people would say they prefer fewer files to many
01:43:49linuxstbNico_P: It's OK, I don't plan on listening to it... ;)
01:43:52Nico_Pamiconn: in the case of a podcast, the tracks are concatenated to begin with
01:43:53amiconnSomething like chapter marks _in_ the file make some sense... though not with music either
01:44:27Nico_Plinuxstb: today I wrote a program to extract the chapters as a cuesheet
01:45:01linuxstbWhy extract them, why not just read them in Rockbox?
01:45:03LloreanNico_P: Could that just be incorporated into an interaction between the metadata parser and cuesheet handling without an intermediary file?
01:45:19preglowamiconn: it depends, some might say one cd "belongs" in one file
01:45:40preglowamiconn: and stuff like dj sets also fit logically in one file
01:45:41Nico_PLlorean: I'd love that but my program actually uses an other one that outputs an SRT file and then parses the SRT to a CUE
01:45:50preglowwhile track info might still be nice
01:45:56amiconnpreglow: Then the metadata would belong there as well, and not in a separate file
01:46:01LloreanNico_P: I'm guessing the SRT one isn't GPL then.
01:46:21preglowamiconn: agreed, but i believe that qualifies more as "impractical" than "flawed by design"
01:46:24Nico_PLlorean: not sure... it's open source though : MP4Box from GPAC
01:46:45Nico_PI'll have a look at it
01:47:41LloreanNico_P: LGPL
01:47:56LloreanAt least, GPAC is described as that at their official site.
01:48:52Nico_PLlorean: then MP4Box is too
01:49:41 Join TrueJournals [0] (
01:53:16 Quit grndslm ("Leaving")
01:55:24 Join jhulst [0] (
01:57:13 Part isaac_w
01:59:54linuxstbNico_P: Does that example m4b file do chapters the Apple way or the Standard way?
02:00:06Nico_Plinuxstb: the Apple way
02:00:13 Quit amigan (Connection timed out)
02:00:34linuxstbNico_P: So you think mp4box can read Apple chapters, but not write them?
02:00:43Nico_Pthat's it
02:01:16Nico_Pbut I don't have an iPod or iTunes to check that
02:02:10Nico_Pand actually I think there's no real standard way of doing chapters in MP4
02:02:31LloreanIf we're going to add support for chaptered MP4, perhaps we should fix the long standing "MP4s over a certain size don't play" issue
02:02:48LloreanIf it's even fixable.
02:03:08*Llorean doesn't remember ever quite being told what caused it.
02:04:26 Part TrueJournals
02:07:16Nico_Pthat would be a good question for Lear
02:08:07LloreanRight now if I stumble across a long AAC format file, it's always spoken word anyway, so I convert it to speex first, and just be done with it.
02:08:16linuxstbNico_P: Are you sure they're chapters? They just look like subtitles - SRT is a subtitle format...
02:08:41Nico_PLlorean: mine are music
02:08:59Nico_Plinuxstb: you're right. they seem to be only subtitles
02:09:05Nico_Pbut there are also pictures
02:09:25Nico_Pmaybe adding an SRT track would be enough
02:10:00 Part pixelma
02:12:38linuxstbNico_P: Have you tried playing that file in Rockbox?
02:12:49Nico_Pit plays nicely
02:15:31 Quit BHSPitLappy ("Leaving")
02:15:56Nico_Pthe cuesheet I got from it seems to be screwed up though
02:16:05Nico_Pog I know why
02:16:21 Join Shak- [0] (
02:16:43Nico_Prockbox thinks it's half as long as it actually is, because of the frequency
02:17:05linuxstbI thought Lear fixed that recently?
02:17:33Nico_Phe fixed the playback speed but apparently there are still a few issues left
02:18:15Shak-how come theres no feature chart for the sansa e200 series on the rockbox site?
02:18:19Nico_Plinuxstb: if you want I can give you links to files that don't have that problem
02:18:27 Part krush1704
02:18:58linuxstbNico_P: No, that's OK.
02:19:29LloreanShak-: Because nobody's updated the page, and not all hardware features are implemented yet (no FM Radio, no SD slot, etc)
02:19:44LloreanShak-: Assuming you're talking about the feature comparison table, it's not updated often these days
02:20:40LloreanBasically, it existed because there was a good deal of difference between what the H100 and the Archos could do under Rockbox, I believe.
02:21:02LloreanNowadays, most software codec players are pretty much the same other than a few general rules like "If it doesn't have a radio chip, it can't do radio"
02:21:03amiconnIt compares rockbox vs. of
02:23:06amiconnNot many people can update that, as it requires knowledge about what the of can do
02:23:25*amiconn couldn't even update the comparison table for some of the targets he has
02:24:09 Join jchord [0] (
02:24:37linuxstbNico_P: Looking at that m4b file in hex editor, it seems full of all kinds of rubbish - xml files, pngs, jpegs, ... But it seems that the chapter information (at least the track titles) is multiplexed within the file, not neatly at the start. So reading it in Rockbox may not be very efficient.
02:24:41LloreanI think dropping the OF comparison aspect of it from the list entirely may be useful. Instead just a comparison of what various players can do under Rockbox
02:24:54Shak-is rockbox the only effort for new firmware on media players (besides ipodlinux for the ipods)
02:25:44amiconnLlorean: I don't agree. That would remove the main point of the chart, and would just convert it into a rockbox feature list
02:25:57Nico_Plinuxstb: yes, I read that somewhere about AtomicParsley: it would require a muxer/demuxer to be able to read/write this kind of chapter info
02:25:58linuxstbShak-: There's the archopen project for the newer Archos devices.
02:26:48linuxstbNico_P: If Rockbox can play the file, then it must contain demuxing code.
02:27:12Nico_Plinuxstb: yes, so maybe it would be possible to demux the SRT with not too much code
02:27:16linuxstbSo that would be a start - you would then need to change it to demux the subtitle stream instead of the audio stream.
02:27:38linuxstbAlthough the demux code will be in the codec, not in the core (get_metadata) where it's needed for chapter parsing.
02:28:43Shak-is the iriver clix on the to-do list at all for iriver devices?
02:29:37amiconnThere is no to-do list for targets
02:30:29 Join hanaurimasume [0] (
02:30:31Shak-do you have it in mind, rather
02:30:44Nico_PShak-: no rockbox developer owns it
02:31:02Nico_Pand no owner has manifested the desire to port rockbox on it
02:31:17Nico_Pnot that I know of at least
02:31:37Shak-do the rockbox developers get media player donations?
02:33:01linuxstbA lot of the money donated to Rockbox is used to buy hardware.
02:34:04linuxstbI think some people/companies have donated hardware - e.g. Sandisk donated a couple of Sansas.
02:35:00jac0bnow I am getting a different error on the sd-card patch
02:35:03jac0bpatch: **** malformed patch at line 111: @@ -1208,6 +1208,9 @@
02:35:21jac0bshould I change the 1208,9 to 1208,6
02:35:47Joelyjac0b which patch is it?
02:36:20jac0bthe sd-card patch
02:36:31 Quit jchord ()
02:36:31Nico_Plinuxstb: maybe a wiki page about MP4 chapters could be nice
02:36:50Joelyjac0b, url? i meant
02:36:51jac0bmajor_works said to compare the old and new diff files and paste the changes to the new one
02:37:10linuxstbNico_P: Yes, although I don't use mp4 for anything...
02:38:06Nico_Pmaybe I'll have a go at it tomorrow and upload my tools to it
02:38:16Nico_Pwith the information I found today
02:39:41 Join krush1704 [0] (
02:39:45 Part krush1704
02:39:47linuxstbNico_P: And some links to freely downloadable podcasts using such chapters could be useful as well.
02:40:02Nico_Pgood idea
02:40:13Nico_PI think atomicparsley has some example files
02:40:29linuxstbwho/what/where is atomicparsley?
02:40:52*linuxstb got an answer from google
02:41:09Nico_Pok :)
02:41:23Nico_Pit's dedicated to iTunes tags
02:42:10Nico_Pand it has example files :
02:42:36jac0bwhat does "(Stripping trailing CRs from patch.)" mean?
02:42:57Nico_Pjac0b: that means your patch had windows-style line endings
02:42:59 Quit kennethd (Remote closed the connection)
02:43:03Nico_Pand patch likes linux-style line endings
02:43:11jac0bits no problem right
02:43:36Nico_Papparently not, but it's better to have your patch files linux-style
02:43:50jac0boh okay
02:43:51Nico_Pactually it's "unix-style"
02:44:37jac0bjoely: did you check out the diff's
02:45:17Joelyyes....i still don't see what's the problem...
02:45:22Joelywhat are you trying to do?
02:45:24 Quit Nico_P ("bed time")
02:45:35 Join webguest43 [0] (i=18976e30@gateway/web/cgi-irc/
02:45:36jac0bpatch the current build with the sd-card patch
02:45:43jac0blook at line 103
02:46:00jac0bopps 111
02:46:08jac0bthat is where the problem is
02:46:19Joelyit went perfectly
02:46:36Joelysd? or hotswap?
02:46:56Joelyhotswap includes sd i think
02:47:12jac0byeah I think so
02:47:13 Join rushfan [0] (
02:47:20rushfanHow do I completely erase my database and rebuild it?
02:48:15Joelyjac0b are you on windows?
02:48:24 Quit webguest43 (Client Quit)
02:48:36Joelyusing cygwin or native utils to patch?
02:48:41rushfansupposing selecting initialize now should do that but it didnt work for me
02:49:09Lloreanrushfan: What do you mean by "it didn't work" exactly?
02:49:10jac0bthat is what I am using is cygwin
02:50:08Joelyuse wget to download then from cygwin
02:50:11rushfanLlorean: It added the new songs. But didnt erase the ones I deleted
02:50:32Lloreanrushfan: And what happens when you try to play one of the deleted songs, exactly?
02:50:39rushfanIt says No File!
02:51:11LloreanUsually when you "Initialize Now" if old songs show back up after the reboot, and the commit, that means they're in the recycle bin.
02:51:21rushfanrecycle bin?
02:51:25LloreanAfter "Initialize Now", did you shut down the device, turn it back on, and see "Committing"
02:51:34LloreanRecycle Bin, Trash Can, or some form of undelete hidden folder.
02:51:35rushfanshould I shut it down after?
02:51:56LloreanInitializing requires you restart to commit the first time, and re-initializing has the same requirements
02:52:07LloreanYou're probably looking at the old database, still loaded to RAM if you enabled that.
02:52:36rushfanthank you. Now its rebuilding the database :D
02:52:40rushfanIll remember that next time around
02:53:41 Quit jhMikeS (Nick collision from services.)
02:53:44 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:54:22 Part rushfan
02:56:20jac0bjoely: didn't work either
02:56:35Joelyhow are you patching?
02:56:51Joelypatch -p0 < hotswap.patch
02:57:16Joelyyeah, i just renamed it
02:57:27jac0bit fails on "make"
02:57:47Joelydid you _just_ sync with svn?
02:57:51Joelycheck your revision
02:58:01Joelyi'm at 13509
02:59:03Joelyhmm do you think it's a thing with cygwin?
03:01:49jac0bhow do I rename in linux/cygwin again
03:02:01Joelyyou could use mv
03:02:18Joelyor rename
03:02:59jac0bI am trying it with just cygwin
03:03:39jac0bb/c before I used wget but renamed the file with windows
03:03:44Joelyif you want i can build you a rockbox build...if you want to trust me. otherwise i guess i can just set up a native windows compiling env with patch and all the standard utils...
03:03:54jac0bnope still get the error
03:04:07jac0bmake[1]: *** No rule to make target `export/hotswap.h', needed by `/home/jacob/r
03:04:07jac0bockbox/build/firmware/common/disk.o'. Stop.
03:04:07jac0bmake: *** [build] Error 2
03:05:53jac0bthat is the same one I got before and major_works said I need to merge &
03:06:05jac0bto get it to work
03:06:24Joelywhy do you need both?
03:06:55 Join major_works [0] (
03:07:17major_worksjac0b: Are you sure you got all the missing bits from the older version into the newer one?
03:07:43major_worksOne of the missing bits, IIRC, was hotswap.h in total.
03:08:33major_worksMake sure there's a hunk in your combined hotswap.diff file labeled hotswap.h.
03:09:30 Join Daishi [0] (
03:10:30jac0b there is the merged hotswap.diff file
03:10:40jac0bI am pretty sure I got everything
03:11:25jac0bsorry but I have to go (get ready for dreded work)
03:12:09jac0bI will have to pick up on this tomorrow, but thanks everyone for the help
03:13:03major_worksPM me on ABi tomorrow and tell me where I can send you my version of the patch.
03:13:58jac0bthanks major_works
03:14:00 Join entheh [0] (
03:14:19major_worksOK, I'll send it now. Don't know if it'll help but you may as well try it.
03:14:27major_worksIt worked for me twice now.
03:15:12 Quit major_works ("major_works has left the building...")
03:19:22 Join aliask [0] (
03:23:29***Saving seen data "./dancer.seen"
03:31:33Joelyalso was hotswap-target.h and hotswap.c....
03:31:39Joelyoh wait, he left, d'oh
03:33:06 Quit jac0b ()
03:34:23XavierGrdoes the 8 character-long limitation exist on settings that require a path for files? (e.g fm presets, setting files etc.)
03:35:24XavierGrbecause the part on that says not longer than 8 characters name on the FmPresets wiki page was written by me. At the time all those files should be less than 8 characters in order for rockbox to remember them on the next boot
03:36:04XavierGrbut at that time the default settings file was written on a special disk sector instead of a simple file
03:36:59XavierGrso for sector capacity reasons the path to those files was not stored on te settings file (.cfg)
03:37:22XavierGrI should probably test it my self
03:37:51 Join z35 [0] (
03:38:39 Quit element_g (Read error: 110 (Connection timed out))
03:42:48XavierGrwell at least on radio level the limitation still applies, if you want the preset to be remembered after boot, it must be saved on the /.rockbox/fmpresets directory
03:43:01XavierGrI am not sure though if that limitation exists anymore on the saving of the configuration file
03:43:27 Quit jhMikeS (Nick collision from services.)
03:43:33 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:43:40 Join kubiix [0] (
03:48:47 Quit spiorf (Remote closed the connection)
03:52:12 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
03:53:28 Quit jhMikeS (Nick collision from services.)
03:53:34 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:53:57Joely <−− is that what major_works and jac0b were looking for?
03:54:15Joelythat patches fine for me
03:54:32Joelyhas all the needed files..
03:55:25 Join Soap2 [0] (i=41bd808d@gateway/web/cgi-irc/
03:55:50Soap2scorche - thanks for the added !s to the wiki
03:57:26Joelyah oops, i'm still working on what i added...sorry about that :-/
04:03:54 Quit Soap2 ("CGI:IRC (EOF)")
04:06:29 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
04:09:27 Quit gtkspert (Remote closed the connection)
04:10:26 Join gtkspert [0] (
04:19:39 Quit maffe (Remote closed the connection)
04:21:10 Join sando [0] (
04:25:01 Part Joely
04:28:04 Join lucaslucas [0] (n=lucasgeo@
04:28:54 Quit lucaslucas (Client Quit)
04:29:48 Join j1mc [0] (
04:39:05 Quit entheh ("^~")
04:44:00 Quit donutman25 ("ChatZilla [Firefox]")
04:52:37 Join Joely [0] (
04:56:57JdGordonany recording addits here?
04:57:22JdGordonand of course, by addict, I mean anyone who actually uses the recording stuff....
05:17:31BHSPitMonkeyJdGordon, just ask your question... ;)
05:23:32***Saving seen data "./dancer.seen"
05:25:45 Quit amigan (Read error: 104 (Connection reset by peer))
05:26:29 Quit jurrie ("Leaving")
05:29:07 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
05:29:22 Quit Bawitdaba (Read error: 104 (Connection reset by peer))
05:29:31 Join miepchen^schlaf [0] (
05:31:04 Quit Daishi (Read error: 110 (Connection timed out))
05:37:05 Quit gtkspert (Remote closed the connection)
05:37:20 Join chainlynx [0] (
05:38:02 Join gtkspert [0] (
05:44:51 Quit relaxed (Remote closed the connection)
05:52:20 Quit sando ("Apply directly to the forehead")
05:59:41 Quit j1mc ("Leaving")
06:05:37 Quit alienbiker99 ("( :: NoNameScript 4.02 :: )")
06:27:13JdGordonBHSPitMonkey: :p
06:27:23JdGordonanyone wanna test a patch for the rec code?
06:28:57 Join alienbiker99 [0] (
06:30:11 Quit Shak- ()
06:33:47 Join corevette [0] (
06:34:51Joelyme me!!!
06:34:57Joelysansa right?
06:35:12JdGordonhehe no, not enablin recordin on the sansa
06:36:03Joelywhat's it for then?
06:36:12JdGordonsurprise :)
06:36:26JdGordoncommitintg it anyway seen as its so small and works :p
06:38:00 Quit jhulst (Read error: 60 (Operation timed out))
06:54:37 Join RMenes379 [0] (
06:55:18RMenes379Hello there!
06:56:38RMenes379Someone around?
06:57:48RMenes379I just registered on the Rockbox Wiki to do some editing and contribution.
06:58:06ZiderI've been planning to do that too
06:58:07Joelyhey RMenes379
06:58:16Ziderfor a couple of days at least ;)
06:59:05RMenes379Mine's longer term than that.
06:59:17RMenes379Mine's for as long as I'm using Rockbox.
06:59:24RMenes379Which will be for a good long time.
06:59:36ZiderI don't have a supported player, so.. :)
06:59:48RMenes379What are you using?
06:59:56Zideririver S10
07:00:28RMenes379Have you tried to look for any technical information about the player?
07:00:39RMenes379Perhaps you can help jump start a new port for it.
07:00:43ZiderI've found pictures of the interior
07:00:53Ziderthat's what I was planning on posting
07:02:02RMenes379Wait... the iriver S10... which one is that one?
07:02:07RMenes379I haven't seen that model.
07:02:09Zidertiny 2GB player
07:02:13Ziderquite new
07:02:27alienbiker99oh that one, thats really small
07:02:51RMenes379The only irivers that I've seen up close are the H120 and H320.
07:02:57RMenes379But that's what I want to post.
07:03:12alienbiker99i miss my H120
07:03:14RMenes379My girlfriend gave me her FM presets from her iriver H320 for the New York City region.
07:03:44RMenes379And I want to post it to the wiki, seeing that the only two regions posted for New York are both upstate.
07:04:28Ziderahh I found the pics again :)
07:04:43alienbiker99RMenes379 what stations?
07:04:59ZiderI dunno how to start anything at the wiki tho
07:05:11ZiderI suck at formulating stuff like that
07:05:13RMenes379All of the major stations from New York City and surrounding regions.
07:05:37alienbiker99i listen to some NYC stations, less now, i live in central jersey so i get a lot of stations
07:06:21RMenes379But isn't that a pain? I mean, whenever I travel across Jersey, right around halfway through the state I lose the NYC stations and start receiving the Philadelphia stations.
07:06:40RMenes379A pain, it is!
07:07:16alienbiker99it depends, it gets annoying but i dont travel to south jersey a lot but when i do usually just play stuff from my mp3 player
07:07:44 Join XbooX [0] (n=xboox04@
07:08:16alienbiker99i didnt even know there was a fm preset page on the wiki
07:08:38RMenes379You can reach it via the Extras menu.
07:08:50RMenes379At the bottom.
07:09:54alienbiker99on the wiki or in rockbox?
07:10:28RMenes379From the main Rockbox page, pick the "Extras" on the sidebar, then scroll to the bottom of the page, and you'll see the link to the FM Presets.
07:10:50alienbiker99oh ah isee
07:11:04alienbiker99i killed my H120 =( and my H320 battery is dead
07:11:14alienbiker99ballooned on me
07:12:08RMenes379A shame, that is. The irivers are some great little devices.
07:12:25RMenes379Did he blow it up like a beautiful balloon? :)
07:12:29Ziderthere's two things I don't like with this S10
07:12:34Zidersmall things tho
07:12:37alienbiker99hahah its like internally explodes sorta
07:13:01RMenes379What's bad about the S10?
07:13:13chainlynxhey all, I have a fourth-generation (grayscale) ipod and I'm installing rockbox for the first time... I've downloaded and installed the current build for my platform, but I don't (think) I have the fonts package installed yet... I've downloaded the fonts zip file but the docs don't say where to put it on the iPod... does anyone know where I'm supposed to stick the fonts files?
07:13:18Ziderthey use some weirdo playlist format, and not m3u/pls.. and they don't turn off completely, they just go down in "powesave mode"
07:13:18alienbiker99its a replacement battery and most of them balloon, but im in the process of tryingto find a working motherboard for the H120
07:13:30alienbiker99Zider pls or plp?
07:13:54alienbiker99ah thats what i meant, yeah its what they use on MTP devices
07:14:00RMenes379chainlynx: Open the fonts zip file, and just extract that folder to the root of your device.
07:14:10Ziderand it's a really dumb format too
07:14:19ZiderI hacked up a script to create playlists
07:14:38alienbiker99and it really only started going to sleep mode with the H10 ...trying to be ipod like
07:15:03Ziderit CAN be turned off completely.. but you need to access the reset button for that :P
07:15:11alienbiker99Zider, ive been looking for a script for my gigabeat, but trying to find one that doesnt copy the song a second time
07:15:15ZiderI'm nagging them for a setting for ut ;)
07:15:55RMenes379I never liked that sleep mode with the iPod.
07:15:59chainlynxRMenes379: ah, ok, as simple as that... thanks
07:16:09ZiderI never liked iPods ;)
07:16:21RMenes379But Rockbox makes them all the better.
07:16:34Ziderthey're still ugly
07:16:45Ziderand all Appley
07:16:53Ziderand HUGE ;)
07:17:02RMenes379Would you rather... a ZUNE?? *BUM BUM BUUUUUUM*
07:17:07Zidera what?
07:17:15RMenes379My thoughts exactly.
07:17:19chainlynxthey just passed a million sold, didn't they?
07:17:32Zidereww, a microsoft thing :P
07:17:34scorchethis is a bit offtopic..
07:17:39Ziderand even huger
07:17:55alienbiker99they really arent bad
07:18:01RMenes379At least Apple's less evil.
07:18:13*scorche coughs
07:18:23alienbiker99oh if somebody has a wiki account can they please change the note on the buyersguide about the 80gb not being compatible with rockbox
07:18:37Ziderscorche: we're not supposed to talk about mp3 players..?
07:18:43RMenes379I need permission to write to the wiki.
07:18:53RMenes379That's what I came for in the first place.
07:18:57scorcheZider: you are supposed to talk about rockbox
07:19:03scorchethis isnt a social channel
07:19:16Ziderscorche: sorry then :P
07:20:38scorchealienbiker99: done, but someone else will have to put it on the table
07:22:02scorcheRMenes379: what were you planning on editing?
07:22:18RMenes379My girlfriend gave me her FM presets from her iriver H320.
07:22:37scorcheand what is your wikiname?
07:22:44RMenes379She and I live in the New York City area, so the presets are for this region and surrounding areas.
07:22:53alienbiker99ok i would but no ipod for me =D
07:23:34scorcheRMenes379: done
07:23:35***Saving seen data "./dancer.seen"
07:24:06RMenes379Okay, so how to I upload the .fmr?
07:24:47RMenes379Wait... found it.
07:24:58RMenes379I'll be right back.
07:26:32RMenes379I attached the .fmr to the wiki, but I think something's off.
07:27:46RMenes379scorche: Can you see if it looks correct?
07:30:10RMenes379Actually, before I leave, I wanted to ask one question to any Rockboxers in New York City with FM radio targets.
07:31:16RMenes379Can any of you receive 100.3 Mhz (Z100) clearly on your players?
07:33:20 Quit corevette (Remote closed the connection)
07:36:43 Quit RMenes379 ()
07:37:30 Join SolidSnake [0] (
07:38:53 Join kaaloo [0] (
07:38:57 Part kaaloo
07:40:19 Quit SolidSnake (Client Quit)
07:47:48 Join My_Sic [0] (
07:54:01chainlynxabout rockbox's dual-booting: I can get the ipod to switch over to the original apple firmware from rockbox... how do I switch it back to rockbox? Powering down does no good... do I just have to wait until it goes into a deep sleep (wait a few hours w/o power to it)?
07:54:25LloreanHold Menu+Select to reboot
07:54:33LloreanThe Apple firmware doesn't power down naturally, it only 'sleeps'
07:55:51chainlynxLlorean: works beautifully, thanks
08:06:10 Join pearldiver [0] (
08:06:22chainlynxCan someone give me write access to the wiki, please?
08:17:14midkaychainlynx: sure.. your username?
08:18:24midkaychainlynx: done.. :)
08:21:08 Quit My_Sic (Read error: 104 (Connection reset by peer))
08:23:10chainlynxmidkay: thanks
08:26:17midkayno prob
08:26:36 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
08:28:42 Part toffe82
08:32:11 Join LinusN [0] (i=linus@gateway/web/cgi-irc/
08:34:11 Quit gtkspert (Remote closed the connection)
08:35:08 Join gtkspert [0] (
08:41:45JdGordonSlasheri: hey, do you tihnk it would take up reworking to add a tag in he tagnavi_custom.config file which would allow you to add a submenu into an exsisting menu?
08:43:42 Join Rob2222 [0] (
08:45:44 Join Clapa [0] (i=andrw@
08:53:09 Join ender` [0] (
08:53:17 Join B4gder [0] (
08:57:42 Join petur [0] (i=d4efd6a6@rockbox/developer/petur)
08:59:19SlasheriJdGordon: hmm, that should be possible to do
09:00:28JdGordonSlasheri: I tinhk that is better than just having the option of adding more menus, seen as the layout of the default menu is ussualy great, just needs some minor additions
09:00:40 Quit chainlynx ("Leaving.")
09:02:04 Quit Rob222241 (Read error: 110 (Connection timed out))
09:02:48SlasheriJdGordon: iirc sub menu entries are just pointer references so it should be able to allocate more of them afterwards
09:03:25JdGordoncoolies :) I'll have a play after I get some uni work done :p
09:03:28Slasheriand the pointer reference table is fixed in size, so menu entries could be even moved
09:03:34Slasherinice :)
09:05:33midkayquestion.. so tagnavi_custom.config is appended to tagnavi.config if it exists? or does it entirely replace it?
09:07:21 Quit aliask ("ChatZilla [Firefox]")
09:07:39JdGordonmy understanding is it replaces, unless you dont specify a new root, in which case it adds to it..
09:08:13*JdGordon thinks it shuold be read just before the end of tagnavi.config instead of at the begining so less menus need defining in the custom file
09:08:44midkayoh, i see..
09:09:55LloreanGodEater: I once had to have someone redirect the output of a directory listing on their player to a text file to finally get a straight answer as to what files were where.
09:10:23LloreanI though the custom one was just inserted, if you did it right.
09:10:39Llorean^−−- Speaking of Tagnavi
09:11:30GodEaterLlorean: I'll ask him later if he still has problems :)
09:12:50Slasherimidkay: it's appended
09:13:08Slasherimidkay: or included just there where the include clause is
09:13:09JdGordonit replaces if you specify a new rot though doesnt it?
09:13:21Slasheriyes, then it replaces the default menu structure
09:13:43Slasheribut everything defined in tagnavi.config (if any) is available also from tagnavi_custom.config
09:14:04 Join Thundercloud [0] (n=thunderc@
09:15:43JdGordonSlasheri: everything? or only the stuff that was efined before the %include line?
09:15:49midkaySlasheri: so there *is* a way to make it totally replace tagnavi.config if you so desire.
09:16:03*JdGordon vagualy remembers seeing a patch so even stuff after the %include could be used
09:16:26SlasheriJdGordon: only the stuff before %include
09:16:51Slasherimidkay: yes, if you mean replacing the default root menu
09:16:57Slasherijust use the %root_menu caluse
09:17:06midkaySlasheri: got it, was just curious. thanks.
09:17:14Slasherithen the default root menu will be hidden and inaccessible (but it's still there)
09:17:20JdGordonSlasheri: ok, then shohuoldnt the include be the last line of tagnavi.config? or that patch be commiteD?
09:17:58SlasheriJdGordon: it can't be the last line because the default root menu has the custom menu reference
09:18:07Slasheriso it needs to be included before that
09:19:30JdGordonSlasheri: then maybe 7003 could be commited? (goin just by the description though...)
09:20:33JdGordonhmm.. that actually sounds like it will do what I wanted t do and append to exsisting menus also :)
09:20:35 Quit GodEater ("CGI:IRC (Ping timeout)")
09:21:04SlasheriJdGordon: possible, there are few tagcache/browser patches on the tracker that are almost commitable
09:21:10 Join purplegreendave [0] (n=chatzill@
09:21:24 Join davina [0] (
09:21:33 Quit Clapa ("643")
09:21:37Slasherican't remember now which ones.. but the code of those needs little fixing, the use of buffers etc. isn't all correct
09:23:04purplegreendaveCan anyone help me please? I have rockbox on my iPod nano, and I can put an mpeg on and it plays fine, but it only displays a quarter of the video (the top left corner). I have tried using media coder to resize it but I cant seem to get it to work/
09:23:37***Saving seen data "./dancer.seen"
09:23:49SlasheriJdGordon: yes, i think #7003 is ok. but the code needs minor adjustments
09:24:02JdGordonwhats the use of havin a title and a in string i the struct root_menu? couldnt just the title be used for the id?
09:24:04 Quit miepchen^schlaf (Read error: 113 (No route to host))
09:25:54JdGordonactually... ignore that, I guess you could have multiple menus with the same name but unrelated
09:26:00*JdGordon bbs
09:26:23Lloreanpurplegreendave: You should probably ask somewhere that specialized in mediacoder for help with using it.
09:26:33Slasheriit could be used but might not be so clean way to implement it
09:26:56purplegreendaveok thanks & will do, but is there a way to zoom out on the iPod?
09:27:44LloreanThat would take more cpu power for scaling, and we're already tight in that area.
09:27:48LloreanYou should resize your videos in advance.
09:29:48purplegreendavemmkay thanks
09:29:52 Quit purplegreendave ("ChatZilla [Firefox]")
09:30:03 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
09:30:13 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
09:33:58 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
09:35:00GodEateris rbutil included in the source archive? Or only available through svn ?
09:37:49 Join maffe [0] (n=maffe@
09:38:02linuxstb_Just SVN I think.
09:38:35GodEaterwith my current lack of home internet I can't do my usual network hoop jumping to make svn work here :(
09:38:37linuxstb_I could host a copy if you wanted it.
09:38:58GodEatercool - I'm going to have another stab at making a statically linked version
09:40:22GodEaterthank you very much
09:41:28 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:44:24*linuxstb_ wonders why the distros are so slow to include wx-2.8
09:44:57B4gderdistros are generally slow at including new packages
09:45:22B4gder... and they often have valid reasons
09:45:52 Quit GodEater ("CGI:IRC (Ping timeout)")
09:47:26*linuxstb_ finds the valid reasons here -
09:48:16B4gder"Let me clarify: NEVER has
09:48:16B4gderthe first "stable" release of some new wx branch gone out the door
09:48:16B4gderwithout some critical flaw"
09:49:48linuxstb_I see the 2.8 branch is now up to 2.8.4...
09:51:54 Join crop [0] (i=c27f0812@gateway/web/cgi-irc/
09:51:55 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
09:53:17GodEatergentoo has available now
09:53:48cropHello. I find the new string "list acceleration speed" a little bit weird. IMHO just "list acceleration" or "list scroll acceleration" would be better. "acceleration speed" is a wrong term IMHO.
09:53:51linuxstb_Do they also have older versions available? I think that's what Debian is refusing to do.
09:54:18 Join Gibbed [0] (
09:54:22 Quit Rick (Nick collision from services.)
09:54:45cropamiconn: here?
09:54:53Lloreancrop: Wrong how?
09:54:54GodEateryes it does
09:55:21GodEater2.4.2, are available too
09:55:41crop"Acceleration speed" is double. "Acceleration" already means speed change per time unit.
09:56:18Lloreancrop: Yes, but acceleration has other attributes than just speed.
09:56:21crop"Acceleration start delay" is OK
09:56:21linuxstb_Acceleration speed means rate of change of acceleration.
09:56:31LloreanAs you mentioned, it has a time element and a speed element.
09:56:59LloreanHonestly there should be two settings, one that changes the time between increases, and one that changes the increase size (speed)
09:57:06 Quit linuxstb_ ("CGI:IRC")
09:57:14cropIf acceleration had speed it would change in value. But it doesn't change. So it has a value but not speed.
09:57:27LloreanWhat do you mean by "value"
09:58:02cropAcceleration has a constant value, e.g. 2x/3s. But its *speed* is zero since the acceleration doesn't change
09:58:22LloreanThe rate at which it accelerates changes.
09:58:31 Quit jhMikeS (Nick collision from services.)
09:58:32LloreanIn the dP/dT formulat, the ratio of dP to dT is changed.
09:58:37 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
10:00:05cropLlorean: well, we use 'logarithmic' acceleration but still it's constant (set by the option) What does change is the scrolling speed.
10:00:24 Quit Thundercloud (Read error: 104 (Connection reset by peer))
10:00:33LloreanWhat changes is "how fast it accelerates" in lay terms.
10:00:34linuxstbcrop: I think I agree. The seeking acceleration setting is just called "FF/RW Accel"
10:00:38 Join Thundercloud [0] (n=thunderc@
10:00:49Llorean"How fast" often translates to the word "speed"
10:00:55LloreanPerhaps "Acceleration Rate" might make more sense?
10:01:20cropLlorean: no, the acceleration remains const (if seen in the logarithmic way). The speed changes
10:01:43LloreanYes, you're picking that constant.
10:01:53LloreanYou're picking the rate of acceleration. You aren't picking the speed.
10:02:02JdGordonSlasheri: back to the id/title thing to save ram, would it be better to use a number instead of a string for the id? or not too worried about the ram usage?
10:02:22LloreanIf the acceleration changed it would be "Acceleration acceleration"
10:02:26crop"Acceleration rate" is better IMHO. It's correct and it does distinguish the option from the "start delay" option
10:03:10cropLlorean: yes. But since it doesn't change its speed is always zero, so the option doesn't make any sense
10:03:12JdGordonSlasheri: and lastly, could you put a comment on 7003 saying what would need to be fixed before it could be accepted please?
10:03:32Lloreancrop: That statement makes no sense.
10:03:40Lloreancrop: The acceleration changes when you pick it.
10:03:52LloreanThe acceleration has no acceleration, it's constant. You are picking that constant.
10:04:02LloreanThat constant is the "rate of increase of speed"
10:04:27cropLlorean: yes, but not "Acceleration speed".
10:04:29Lloreanrate means "a ratio between two different measurements", in this case, the change in amount scrolled per period of time, a rate of distance over time is also known as "speed" in the english language.
10:04:36croplinuxstb: how can I better explain it?
10:04:55linuxstbJdGordon: Looking at your patch to tools/configure - does that create the recpresets directory for targets without recording?
10:05:34cropThe best would be just "acceleration" or "scroll acceleration". But the former doesn't explicitly distinguish from the other setting
10:05:35*linuxstb goes to kitchen for coffee
10:06:49cropLlorean: so if I've chosen the option e.g. 3x/2s then the logarithmic accel is const and is 3/2 (in some appropriate unit). But it remains const, so its speed is zero.
10:07:21 Part odb|fidel_
10:07:27LloreanOut of curiosity, 3 units of measurement over 2 seconds = 1.5 units per second
10:07:30LloreanThis is a measurement of?
10:08:01 Quit ender` (Read error: 110 (Connection timed out))
10:08:13LloreanSince our units are distance, the right answer is "speed"
10:08:14JdGordonlinuxstb: yeah, I could change it to check, but is it realy a big deal if the exra folder is there?
10:08:52 Join barrywardell [0] (
10:09:07LloreanForget I said anything
10:09:13JdGordoncrop: the <num>x/2s is really only because it was the closest to the usage... without adding exra code tha was the closes to making sense.. it shhuoldnt be taken literally..
10:09:26JdGordonjust that the general idea is tha if that number goes up you go faster
10:09:27cropLlorean: with "distance" do you mean how far the cursor advances?
10:09:32B4gderJdGordon: yes, people get annoyed by empty dirs in release packages
10:09:35Lloreancrop: I said forget I said anything.
10:09:45JdGordonB4gder: ok, ill fix that now then
10:10:47cropJdGordon: the way it works is OK, I only complain about how it's names
10:11:05JdGordonyeah, I didnt like the name either, but meh, close enough
10:12:13*Llorean thinks the problem is just a case of the physics definition of speed, which isn't appropriate here, vs the colloquial definition of speed, which is acceptable
10:14:46SlasheriJdGordon: if you find some alternate way for the id, just please do it :) however, numbers in configuration file instead of strings could be confusing
10:15:29JdGordon:) if we arnt worried about the mem usage then im happy if you're happy :)
10:15:34SlasheriJdGordon: the fix should be fairly simple and i have intended of doing it while committing it.. but it just has been delayed, i will try to do it soon
10:17:03cropLlorean: maybe. But since I've dealt a bit with the terms the difference is very clear to me :-)
10:17:53Lloreancrop: In my defense, I confused what the 2x referred to in the ratio, you were right that the unit isn't a measure of speed.
10:18:23LloreanBut I still don't think the phrase is inappropriate for how it's used considering it'll make sense to users more this way, I think. Though I think "rate" would be better.
10:19:54cropLlorean: "rate" in the sense of "value". And the accel unit is logarithmic :-)
10:21:02Llorean"Rate" in the sense of "how often the speed is changed"
10:21:36LloreanIn the end, it's apparently "double the speed ever X seconds", so it's just a frequency.
10:21:57LloreanAt least, according to that menu
10:22:23 Quit Gibbed (Read error: 60 (Operation timed out))
10:22:38LloreanI'm not explaining clearly now
10:22:41LloreanIt is much too late.
10:22:48cropLlorean: yes. So the linear acceleration isn't const but the logarithmic one is
10:23:37cropBut if the accleration (the log one) is const then its speed is zero
10:24:11LloreanSee, I see "60 cycles per second" as a speed. Or "1 trigger every two seconds". It's how fast an even occurs.
10:24:35cropThat's why I think just "acceleration" would be the best, without specifying that it's really a logarithmic one.
10:24:38LloreanSince you can't change both the 2x and the / X seconds aspects, you can only change the speed of events, rather than the whole acceleration
10:25:04amiconnJdGordon: Did you spot my remarks regarding the sd patch in the log?
10:25:22cropLlorean: ah, then you mean acceleration frequency
10:25:27JdGordoni must have mised them, i did read the log though :p
10:25:36JdGordonamiconn: roughly what time?
10:25:37cropI meant acceleration as applied to the cursor movement
10:25:40LloreanBut frequency is velocity within an oscillatory path anyway.
10:26:03LloreanIt's still distance over time, it's just a case of distance to reach the zero point again.
10:26:10LloreanWell... sorta
10:26:23*JdGordon thinks he will just change the setting to a regular int to shut everyone up
10:26:26 Quit barrywardell (Remote closed the connection)
10:26:26cropIt's frequency, i.e. how many times an event occurs within a time unit
10:26:34JdGordon"scan accell speed" 1...10
10:26:49amiconnJdGordon: 00:48..00:56 cest today
10:26:55*linuxstb wishes English had two different words for scroll
10:27:03Lloreancrop: Yeah, but in computers it's a speed: How fast you go through a specific series of instructions
10:27:08LloreanOr repeat a specific series.
10:27:13JdGordonmmm... chocolate scroll :p
10:27:22Lloreanlinuxstb: English could use separate words for a _lot_ of things
10:27:25LloreanLike "free"
10:27:34*Llorean doesn't need to hear that speech a million more times.
10:28:00Lloreancrop: Anyway, I think "Acceleration" is valid, just a bit less clear for the users.
10:28:03B4gderyes, english needs many more words for beer! ;*P
10:28:31GodEaterenglish *has* lots of words for beer :)
10:28:35Lloreanscorche: Actually, at this thing I was at, the CEO of Red Hat said "Proprietary software companies give away of lot of demos, the 'free beer' approach as you might say. Open Source tends to take a more sober approach."
10:28:41LloreanI swear, I was the only person at the even who laughed.
10:28:45JdGordonamiconn: the sd thread without hotswap is the ata thread... so it does the idle notify code
10:29:27JdGordon... /me shold have continued reading...
10:29:46 Join barrywardell [0] (
10:30:19Lloreancrop: Anyway, sorry for all the arguing
10:30:29linuxstbHow does "scroll" translate into other languages? Can you use different words for scrolling up/down lists and the horizontal scrolling of text?
10:30:38cropLlorean: you don't have to be sorry, it was a pleasure!
10:30:56linuxstbcrop: You just came here for an argument? ;)
10:30:58croplinuxstb: cursor movement?
10:31:04GodEaterwe could say "panning" for horizontal movement
10:31:09GodEaterI think that's valid
10:31:12*petur gets a late trigger and blames work for it
10:31:18LloreanPanning might work.
10:31:22B4gderlinuxstb: in Swedish we prety much use the english word "scroll" in daily speak actually
10:31:22 Join spiorf [0] (
10:31:37JdGordonpanning is also 1 less byte!
10:31:53cropamiconn: if "input source" is the "most important" setting why not put it to the top?
10:32:00linuxstbpanning also means to heavily criticise..
10:32:15linuxstbBut I guess people won't confuse the meaning :)
10:32:25LloreanThat's a rather uncommon use of the word over hear at least.
10:32:33GodEaternot here it isn't
10:32:44GodEaterbut I don't see how you could possibly confuse it as a setting :)
10:32:47linuxstbRockbox gave that text a good panning...
10:33:08GodEaterthat's a bit of a stretch
10:33:41LloreanOr as Crop suggested we could just call the list-version the "cursor acceleration" or whatnot, rather than scrolling.
10:33:59GodEaterthat's far too many bytes :)
10:34:01LloreanSince there's a lot of code and discussion referring to "text scrolling" as "scrolling" anyway.
10:36:10 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:38:29 Part Llorean
10:40:46linuxstbGodEater: Regarding your last forum post - does party mode make the lock icon appear?
10:41:05scorcheugh @
10:41:25scorchelinuxstb: aye....i didnt think it did either
10:42:59GodEaterlinuxstb: I don't know actually - I've never used it
10:43:09GodEaterlet me check...
10:43:38scorcheit sounded like his switch was actually broken, or Rockbox froze, if he was moving it and it didnt do anything
10:43:38linuxstbI think I remember someone else mentioning that problem on a video ipod (lock icon appearing when the hold switch wasn't on)
10:44:20GodEaterNo it doesn't
10:44:25GodEatermy bad
10:44:34GodEaterit's not a bad idea though :)
10:44:44linuxstbWe need a party-hat icon...
10:45:01linuxstbOr a beer icon...
10:45:07GodEaterhahah - yeah
10:45:22*JdGordon likes the idea of the lock icon showing if in party mode
10:45:43scorchei dont think petur/preglow would be able to have Rockbox run without party mode then...
10:45:59 Quit atsea- (Remote closed the connection)
10:46:26GodEaterthey don't have some sort of beer based theme already ?
10:47:20GodEaterwhat happened to the patch to make rockbox read themes straight from a tarball ?
10:47:53linuxstbNico_P updated it recently, but it didn't seem to give any noticable speedup. It's in the patch tracker, and could do with more testing.
10:48:19*amiconn wonders what crop is talking about
10:48:20linuxstbI'm not sure if it's worth committing if it doesn't speed up theme loading.
10:48:39 Join atsea- [0] (i=ariel@gateway/tor/x-281cecb780ac8bc3)
10:48:42GodEaterIt might not speed it up
10:48:52GodEaterbut it might make *some* of our users lives easier
10:49:07linuxstbIt wasn't the whole theme in a tar, just the WPS bitmaps.
10:49:56markunI wonder if Rockbox will be mentioned in this thread:
10:49:56 Quit Maz (Read error: 104 (Connection reset by peer))
10:50:01GodEaterI was thinking it might be nice if the theme could be loaded all from one file - it might encourage more theme authors to package them up nicely
10:50:13cropamiconn: hey! What are you wondering about? I meant the recent change in the wavrec menu
10:50:34 Join The-Compiler [0] (
10:51:49 Join bluebrother [0] (i=aJD0HkfO@rockbox/staff/bluebrother)
10:52:42B4gdermarkun: I doubt it
10:54:51markunSlasheri: why can't the tagnavi
10:54:59 Join thegeek [0] (
10:55:22markunSlasheri: why can't the tagnavi_custom.config not use UTF-8 but uses ANSI instead?
10:55:41markun(at least according to
11:04:06amiconncrop: The most important item is 'start recording', but the items are sorted in logical order. First you set parameters, then you start recording
11:09:15 Join Entasis [0] (
11:22:18 Join Nico_P [0] (
11:23:40***Saving seen data "./dancer.seen"
11:29:10 Quit linuxstb ("Leaving")
11:33:51cropamiconn: ah, right, the item number is zero based! My fault. This will probably change once we have a dedicated record button, right?
11:36:33Slasherimarkun: hmm, that needs to be fixed
11:37:04 Quit Nibbier (Read error: 110 (Connection timed out))
11:39:02markunSlasheri: I think so too, so people can have a korean tagnavi file for example
11:39:30 Join Nibbier [0] (
11:39:41 Quit petur ("worrrk stkov")
11:40:06 Join lee-qid [0] (
11:40:10 Quit lee-qid (Read error: 104 (Connection reset by peer))
11:40:25 Join lee-qid [0] (
11:44:33 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
11:48:58LloreanIn regards to a discussion while I was out: even if part of it was in jest, I do think some sort of visual display that party mode is active would be helpful.
11:51:00JdGordonSlasheri: (soz keep bugging you), just tried out 7003 and it works great. my custom config is tiny compared to before now :)
11:55:13cropamiconn: why is the entry to start recording in the sttings at all? Shouldn't it be "factored out", i.e. be not in every recording settings?
11:56:36amiconnwavrecord only has one menu
11:56:47amiconnThis is tech preview stuff anyway
11:56:52 Join DataGhost [0] (
11:56:54LloreanDoes anyone have any special love for the current Sansa keymap?
12:00:41 Part maffe
12:01:00 Join maffe [0] (n=maffe@
12:03:45LloreanQuestion about the button things, what's the difference between a button | BUTTON_REPEAT with a prereq of the same button vs one without a prereq of the same button?
12:07:11markunSlasheri: it looks like the tagnavi file works fine with UTF-8
12:09:52 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
12:12:41 Join kubiixaka [0] (
12:15:50 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:16:55LloreanAnyone with an e200 want to let me know what they think of this? (and if they think I did something wrong with setting the keymappings, I've not tinkered with those before)
12:19:44JdGordonLlorean: there is a sublte difference which i cant remember right now :p
12:20:27LloreanJdGordon: I couldn't determine the difference just from reading the buttons and thinking "how do I uses these buttons"
12:20:46LloreanSo I pretty much went with "if there's a |BUTTON_REPEAT or |BUTTON_REL, use a prerequisite of the button being pressed"
12:21:04LloreanAnyway, I want to remap the keys so that the buttons make a wee bit more sense.
12:21:11JdGordonrepeat isnt always needed, I cant really remember why though
12:21:25 Quit kubiix (Read error: 110 (Connection timed out))
12:21:32JdGordonI disagreee with moving stop to power and menu to the down button
12:21:46*JdGordon finds menu much more uefull tha a stop button
12:22:15LloreanMy goal is to bring it more into line with our other targets.
12:22:18JdGordonalso, I like the fact that I dont have t hold the buton to get the contxt menus
12:22:32 Quit Soap ()
12:23:03LloreanTo me, holding down the select button makes a kind of sense anyway, it's almost like right clicking.
12:23:30LloreanMeanwhile, on many of our targets the power and stop buttons are the same button.
12:23:46LloreanThe only target I know of that uses a long press of Play/Pause is the iPod because there isn't a button we can use on its own.
12:24:23JdGordonI really dont see the need for a stop button... pause is good enough, and if you want t clear the playlist, then a long pause is just as simple
12:25:05LloreanLong pause doesn't make much sense though, people frequently enough ask how to stop playback on iPods already
12:25:23LloreanAlso, why is "I like it better" a good reason to make it more confusing for people moving from other Rockbox targets?
12:25:36JdGordonlooking at the keymap... the |button_rel is only needed if its likly that the button is used in another context but it is still needed in this one.. e.g the power button in context_standard
12:25:46linuxstbI'm just looking at a picture of an e200 - what's the icon on the "down" button meant to be?
12:25:54JdGordona drop menu
12:26:03JdGordonits used for the context menus in the OF
12:26:05amiconnLlorean: The difference of a button|BUTTON_REPEAT with and without prereq is very simple:
12:26:18amiconnWith prereq, it triggers only once, i.e. action on long press
12:26:35Lloreanamiconn: Ah, forces only one trigger?
12:27:14Lloreanlinuxstb: My patch basically makes that the "Menu" button as it is on most of the other players.
12:27:19amiconnWithout prereq, it triggers continuously times until the button is released. This is used for actions that should repeat (like increasing/decreasing volume, seeking in wps etc
12:27:36Lloreanamiconn: Then for my changes, adding the prereq was appropriate it seems. Thank you.
12:27:36*JdGordon thought there was a geater argument for using the buttons as they are labbeled, than using them to get consistancy between targets
12:27:57LloreanJdGordon: The only label I contradict is the small word "Menu" above the power button.
12:28:09LloreanErrr, below
12:28:23JdGordonI know :) I really think hat should be used as it is no
12:28:24linuxstbI think there's an advantage to keeping the "power" button as menu - it means that in plugins where you need four directions, you still have a button free for the menu (or exit).
12:28:38Lloreanlinuxstb: My patch doesn't change the plugins at all
12:28:47linuxstb(that's a pain on the ipods)
12:28:54LloreanIt just aligns the keymappings in the WPS and tree/menus
12:29:10LloreanSo that long-menu gets you the quick menu, long-select gets you the context menu, and the power button serves as stop
12:29:44*JdGordon back in an hourish
12:30:10linuxstbLlorean: Consistency between core and plugins is a nice thing to have...
12:31:07Lloreanlinuxstb: I agree, but I think having the quick menu be an different button from the main menu, and the context menu be a different button from select causes a lot of confusion for those who come from other targets.
12:31:52LloreanTo me, these are part of the fundamental controls of Rockbox, and they're mostly the same on other players.
12:31:55linuxstbIs there a button on the side (top-left) of the sansa?
12:34:15LloreanIt just seems to me there should attempt to be some fundamental consistencies about how Rockbox works on the players. A sort of input paradigm, where possible, rather than somewhat drastically different controls per target.
12:36:38LinusNusing the power button for menu makes it somewhat risky, since you could hold it for too long to get to the context menu, and turn it off instead
12:36:52LloreanLinusN: Context menu was a short press of a different button
12:36:57LloreanIt was not mapped to the same button as the main menu
12:37:12LloreanErr, quick menu, rather
12:37:13linuxstbWhat's record used for?
12:37:18LloreanRecord isn't used at the moment
12:37:29LloreanIt's sorta in the same state as the H1xx
12:38:35 Quit Rob2222 ()
12:39:16LloreanTo me, if you're going to present users with the same screens, there's a value in having the same button relations.
12:39:43LloreanWhere the number of buttons permits
12:40:27LloreanThe "Menu / Quick menu" pairing and the "Select / Context menu" pairing are, to me, at the core of this.
12:40:39LloreanWhich is kinda a point I argued regarding the Gigabeat too
12:45:46 Join PaulJam [0] (
12:46:38 Join jurrie [0] (
12:47:26 Join Rob2222 [0] (
12:47:32 Quit HEx (Read error: 110 (Connection timed out))
12:47:35PaulJamhi, i noticed, that now when paged scrolling is enabled, at the end of a list, the last item now isn't always aligned with the bottom of the screen. is this intended?
12:50:12PaulJamand when you then mark the second to last idem and hold DOWN, then the list jumps down so that the last item is at the bottom of the screen.
12:54:01 Join david11 [0] (n=david52@
12:55:51*david11 greets everyone
12:57:43 Quit z35 (Read error: 60 (Operation timed out))
13:06:06linuxstbPaulJam: Is that something that changed with the recent list acceleration commit?
13:08:15 Join HEx [0] (
13:20:44 Join Domonoky [0] (
13:22:44PaulJamlinuxstb: i haven't noticed it before, should i try it with an older build before the list acceleration?
13:23:43***Saving seen data "./dancer.seen"
13:28:39linuxstbPaulJam: It would be nice to know if it's an old or new problem.
13:29:23PaulJamlinuxstb: with the 2007-05-27 dayly build it does not happen.
13:29:52*amiconn thinks that the list acceleration is the culprit
13:29:59amiconnIt has more problems than just that
13:30:28linuxstbI've only used it briefly (in a sim), and got some strange jumping between items.
13:34:41*JdGordon back
13:35:05JdGordonthe strange jumping isnt selecting the wrong item... just moving the wrong one to the top.. needs some fixing
13:35:42linuxstbJdGordon: Can you explain briefly how the acceleration works?
13:36:04linuxstb(neither your flyspray comments or commit message described the details)
13:36:59 Join Jon-Kha [0] (
13:37:26JdGordonafter a wait time (where the direction is held down) the number of items skips each buton press increases
13:39:26 Join petur [0] (i=d4efd6a6@rockbox/developer/petur)
13:40:00linuxstbJdGordon: So the number of items skipped doubles every X seconds (specified in the acceleration setting) ?
13:41:16JdGordonnot really... I jus used that formatting string in the setting list because its close enough and saves writing extra code...
13:42:01linuxstbSo what does happen, and what does the setting mean?
13:42:24JdGordonevery <that val>*HZ/2 the skip value is incremented
13:42:36 Quit elinenbe (" Like's GUI? Then try HydraIRC -> <-")
13:42:45JdGordonwait a sec.. that doesnt sound right
13:42:50 Quit idnar (Nick collision from services.)
13:42:53 Join Rob222241 [0] (
13:42:53 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
13:43:12JdGordonno, yeah, thats right
13:43:24 Part Llorean
13:44:28linuxstbOK, so the "2x/2s" setting means that the items_to_skip value is incremented once a second?
13:44:49bluebrothercould this also cause FS #7230?
13:44:59PaulJamthere is another fummy behaviour: if paged scrollung is disabled and you scroll down item by item near the end of a list, the view jumps too early to the start of the list (with the cursor still being at the end of the list).
13:45:30JdGordonbluebrother: yeah, saw that in the thread... it does cause it...
13:45:46JdGordonthat shold be a simpe fix.. just have to find 5 min to d it
13:46:39linuxstbJdGordon: Did you try any other algorithms? e.g. doubling the skip items at each interval?
13:47:01JdGordoneach which interval? the start delay interval?
13:47:29linuxstbThe interval at which you change the skip value.
13:48:01JdGordonthe code to decide how much we skip is gui/list.c:814-835...
13:48:16Nico_Plinuxstb: that would be enourmous acceleration, wouldn't it ?
13:48:43linuxstbNico_P: It depends on the value of the skip interval - it would obviously need to be longer than the current method.
13:48:49JdGordonlike I said, I cheated with the 2nd setting formatter to save bin space... the actual string the uer sees is mostly irrelvant, I could have used numbers instead...
13:49:03JdGordonit only increments.. never doubles...
13:49:18JdGordonif it did double it would be way to fast, as Nico_P said
13:49:18Nico_Plinuxstb: I think it would feel jumpy at one point, but probably still worht testing
13:49:30linuxstbJdGordon: That's what you've just told me - I'm asking if you've experimented with other algorithms, or just went with the first.
13:49:47JdGordoncant really remember, probably just went with the first
13:50:17JdGordonThe cause of the bugs is in the function which replaced the old select_previous and select_next functions
13:51:56*Nico_P is off ot class (and going to get $500 from google too :) )
13:52:04JdGordonlucky bugger...
13:52:08JdGordonnot to the first part though ;p
13:54:25JdGordonlinuxstb: do you ask bcause the acceleration isnt smooth enough? or not fast enough? or?
13:55:56GodEaterjust to add my 2 cents - it doesn't appear to kick in in an untuitive way either
13:56:01linuxstbJdGordon: My original question was just so that I understood the algorithm you've chosen to use - there are lots of possible ways to implement this acceleration.
13:56:02 Quit crop ("CGI:IRC")
13:56:37GodEaterlooking at the menu options only, it appears we start accelerating after a given amount of time yes ?
13:56:48linuxstbJdGordon: Does it work on the Sansa's wheel ?
13:56:54JdGordonyeah, works great
13:57:01JdGordon_if_ you can keep your thumb going :p
13:57:03GodEaterI'd find it more intuitive if it were kicked in when you really go for it with the wheel
13:57:22GodEateri.e. a great deal of movement in a short space of time
13:57:36GodEaterrather than just, has the wheel been moving for >500ms
13:57:46JdGordonthe button driver speeds up the rate it sends repeat events anyway
13:59:40*GodEater wonders if that's the best approach
14:00:07linuxstbJdGordon: I think the setting needs to change to more accurately describe what it does. e.g. give a time in 500ms increments.
14:00:33 Quit Rob2222 (Read error: 110 (Connection timed out))
14:01:18JdGordonlinuxstb: ok, but then what shuold the setting be called?
14:01:49linuxstbMaybe "list acceleration interval" ?
14:02:20*linuxstb waits for crop to complain
14:04:32linuxstbHow do the repeat events work for the Sansa's wheel? IIUC, it's a mechanical wheel that you turn, so you can't detect a user touching it?
14:04:48JdGordonjhMikeS got repeat events working
14:05:10JdGordoni assume goign by the time difference on the interupts
14:05:14 Quit GodEater ("CGI:IRC")
14:05:27 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/
14:05:50JdGordonI tihnk I fixed 7230, but no multiline lists long enough, still other wierd bugs though
14:06:20linuxstbYes, looks that way. Something similar should probably be done in the ipod wheel driver, so a user momentarily untouching the wheel doesn't stop repeat events being sent.
14:07:32PaulJamJdGordon: tried the id3 info screen with a large font? it uses double lines.
14:08:18JdGordonI forced the tree view to use 2 lines, and now always have the first line selected first, so seems t work
14:08:54JdGordonpage scrolling is funny :)
14:09:16 Join sando [0] (
14:17:00 Join Llorean [0] (
14:17:13LloreanJdGordon: Y'know, I'm still having problems using Icons reliably.
14:17:32LloreanIs it still "not done" or is it expected to be working at this point?
14:17:50JdGordonwhats wrong wth them?
14:17:56*JdGordon thought all bugs were fixed
14:18:02LloreanWithout a .icons file, they don't quite work right
14:18:08LloreanI'm using the white "default" icons
14:18:20LloreanAnd the viewers ones for .txt files and .mpeg files are showing up as question marks
14:18:48LloreanIt's a duplicate of the normal icons, only white, and they're in the same order.
14:19:08JdGordonok, 1 thing at a time... :)
14:19:13LloreanIt is one thing
14:19:24JdGordonlists first :)
14:19:49LloreanJust wanted to know if it was "known" or if I should remember to bug you about it later. :-P
14:20:00LloreanI was testing out this "Ultima" theme, Denny's default entry
14:20:14 Quit slarti ("leaving")
14:20:25 Join slarti [0] (i=tom@gentoo/developer/slarti)
14:21:33linuxstbI assume I'm not the only person getting "Internal Server Error" when trying to use viewvc? e.g.
14:21:56petursame here
14:22:05JdGordonI dont remember anyone else reporting icon bugs... Llorean if you want t put it on FS and assign to me that shuold be enough to remind me ;p
14:22:15JdGordon500 error here also
14:22:21*JdGordon 's commit broke svn :D
14:22:41*linuxstb pings LinusN, B4gder and Zagor
14:22:42LloreanJdGordon: This is with an absolutely current build, so it's possible they've resurfaced recently, or it's possible nobody noticed since I think this is the only iconset without a .icons file
14:22:46*Zagor is looking at it
14:22:52LloreanJdGordon: I'll post a flyspray entry
14:23:28JdGordonZagor: hey, are emails in FS viewable to all users or just the developers group?
14:23:57Zagorviewable? they are all sent out to the mailing list, so anyone who subscribes can see them
14:24:27JdGordonwere they always?
14:24:47linuxstbJdGordon: Are you talking about emails or email addresses?
14:24:47JdGordonhmm, ok then
14:25:08LloreanJdGordon: Okay, task 7233
14:25:13Zagoroh, they shouldn't be visible. are they?
14:25:37JdGordonI dont know...
14:25:57JdGordonI was wondering why I got an email from someone who must have got it from FS
14:26:42linuxstbI know the email addresses used to be visible in flyspray emails, but I think that's been fixed at some point.
14:27:30ZagorI think the new version fixed that
14:27:51PaulJamif you are logged in and click on a name in flyspray the emails are visible
14:27:52linuxstbYes, checking my archives, the 19 May 2007 was the last email with the addresses listed.
14:28:13linuxstbOne I received on 20 May was just addressed to me.
14:28:18 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
14:29:01linuxstbPaulJam: I think that's new though...
14:29:23PaulJamis there any way to hide the email adress from others?
14:30:15 Quit pearldiver ("some games are better left unplayed")
14:31:26linuxstbI'm not sure if we would want to - I don't mind my address being there, and think it's useful to be able to directly contact people who have opened tasks or commented on flyspray.
14:31:48linuxstb(it's no different to using one of the mailing lists)
14:32:10david11where is the difference between single platter and dual platter, exept of the size and weight?
14:32:16ZagorPaulJam: I'll fix that
14:32:35Lloreandavid11: Um, other than the fact that there's two platters, size, and weight, what are you looking for?
14:32:59david11maybe it needs more energy
14:33:16david11or is faster at reading
14:33:42LloreanEnergy use and read speed varies from drive to drive even among those with the same number of platters.
14:33:45otihB4gder: ping
14:33:59*B4gder schpongs
14:34:02GodEaterZagor, how much work would adding a cookie the the be so it can remember things like autoscroll, and show/hide joins ?
14:34:15david11how about the price
14:34:32otihB4gder: linuxstb has checked my server yesterday, and everythink worked!
14:34:36LloreanDrives with more capacity always cost more.
14:34:56LloreanSingle platter drives with the same capacity as dual platter drives tend to cost more because they tend to use more advanced technology.
14:35:00B4gderotih: yes, but how does that help us?
14:35:01ZagorGodEater: not much. I figured you might want to explicitly enable autoscroll, though, since it can be somewhat annoying to turn off again...
14:35:09david11ok, thanks#
14:36:00GodEaterZagor, true - but when I hit refresh currently to renew my timed-out connection, the autoscroll box remains ticked, but doesn't actually do anything until I click it again. At which point autoscrolling is now working, but the setting is indicating it isn't.
14:36:21otihno idea, but failure is not on my servers side ... i´m now reading the scripts, perhaps i could find something ... would you check my server too? or test it again in the build process?
14:36:23GodEaterZagor: you could always get the text being drawn into a frame and keep the settings at the top ?
14:36:53B4gderotih: we have like 20 servers and your server showed this error, the others don't
14:36:56Zagoryeah, but I don't like frames
14:37:03B4gderand no I have not tracked anything since we talked yday
14:37:05GodEaterZagor: fair enough
14:37:19otihB4gder: but why?
14:37:24B4gderwhy what?
14:37:33*Zagor curses python for changing everything every release
14:37:51otihwhy there is a failure? linuxstb didn´t find something
14:37:57B4gder"and no I have not tracked anything since we talked yday"
14:38:04PaulJambtw, would it be possible in the to always open a new window/tab for links?
14:38:16*preglow curses python
14:38:47*GodEater thinks perhaps we should open a feature request for in flyspray ;)
14:40:07 Part maffe
14:40:20otihB4gder: perhaps there is a failure in my sshd config, could you parse me yours? perhaps i find something?!
14:42:47 Part david11
14:45:28ZagorPaulJam: flyspray email address is now hidden for non-admins
14:45:57PaulJamZagor: ok the i can change it back, thank you.
14:46:17 Join david11 [0] (n=david52@
14:46:48Zagorhmm, would it be ok if developers can see all email addresses? sometimes you want to mail someone questions.
14:47:31LloreanI think it's somewhat important that at least developers can get in direct contact with patch posters.
14:47:33linuxstbI can't see a problem with always showing email addresses - treat flyspray the same as the mailing lists.
14:47:36LloreanAnd/or bug posters.
14:47:57LloreanI don't see a problem with having normal users able to see them either, honestly, as linuxstb says
14:48:10PaulJamZagor: now i have a problem. the email entry on the "edit my details"-Page is now removed too. so i can't enter my email adress now.
14:48:16LloreanPeople who try to fix bugs aren't always in the Developer group
14:48:24ZagorPaulJam: haha oops :)
14:49:43Zagorlinuxstb: yeah but flyspray doesn't even try to obfuscate the address. personally I don't care, but I know we're going to get complaints from people.
14:49:44PaulJamLlorean: i don't like it if everyone can see my emailadress.
14:50:08B4gdertrying to hide email addresses is futile and will fail no matter what we do
14:50:23 Join aliask [0] (
14:50:24linuxstbZagor: But the difference is that you need to be registered and logged into flyspray to see the addresses.
14:50:36GodEaterPaulJam: Have you ever investigate TMDA ?
14:50:44LloreanPaulJam: And I don't like it when someone reports a bug, then never checks their flyspray task, and doesn't have it set to email them when someone posts a comment on it.
14:50:49ZagorGodEater: don't recommend tmda please. it's broken by design.
14:51:04LloreanPaulJam: You still have to be registered for Flyspray to see the addresses.
14:51:10GodEaterit wasn't a recommendation, just a suggestion. I don't use it any more myself.
14:51:25GodEaterthere are people more paranoid than me out there
14:51:41*Zagor goes back to bashing on python
14:52:08linuxstbPaulJam: So you've never used the Rockbox mailing lists for that reason?
14:52:48*preglow joins in on the python bashing
14:52:56*B4gder lines up
14:53:37PaulJamLlorean: but everyone can register at flyspray. and the emil notification on comments is enabled by default. if i knew that my email is visible for everyone i wouldn't have made a flyspray account.
14:53:41Zagorif anyone knows a svn browser that doesn't use python, please speak up now
14:53:59B4gderI believe there are
14:54:14 Quit aliask (Client Quit)
14:54:23 Quit kubiixaka (Read error: 104 (Connection reset by peer))
14:54:43LloreanPaulJam: If your email has been harvested once, ever, your spam amount will continue to increase with time, ad infinitum anyway.
14:55:15B4gderand even if not, any friend who get virus infected will spread it
14:55:21B4gderetc etc
14:55:39 Join z35 [0] (
14:55:42linuxstbWas there a deliberate decision to make flyspray excempt from the real-name policy?
14:56:00LloreanPaulJam: Give me good reasons it shouldn't be available, other than you don't want yours available. I'm not sure keeping your account there is a good reason for us to make it hard for those who need more details to contact those who have the details.
14:56:09B4gderlinuxstb: not really, we just never bothered there
14:56:17 Join jac0b [0] (
14:56:21peturlinuxstb: it's for users too, not contributers only
14:56:32peturjust like the forum
14:56:40JdGordonand any friend who is nice enough to add you to their myspace/freindster/etc list...
14:56:41linuxstbpetur: But users turn into contributors...
14:56:55peturand then we ask their real name
14:57:03B4gderthat we do
14:57:19linuxstbBut then we get awkward cases of patches posted anonymously...
14:57:40peturmaybe we could encourage them to use real names if they plan to contribute
14:58:00Zagorlinuxstb: yes we do, but we ask for real name before accepting them
14:59:02LloreanThe problem lies in the fact that people post a patch, not realizing we'll ask for a real name.
14:59:22LinusNi don't see it as a big problem
14:59:29Zagorotoh I certainly have nothing agains adding a little text about our real-name policy on the sign-up page
14:59:39LloreanZagor: That's about what I feel should happen.
15:00:24XbooXanyone know if Rockbox has better battery life on a 5.5G 80GB iPod?
15:00:26*linuxstb doesn't feel strongly one way or the other
15:00:30LloreanXbooX: No, it doesn't.
15:00:43 Quit jac0b (Client Quit)
15:00:46LloreanXbooX: We're quite clear on the fact that on iPods Rockbox has worse battery life than the original firmware
15:01:35XbooXLlorean: do you have a ratio?
15:01:48LloreanXbooX: See the IpodRuntime wiki page.
15:03:21XbooXthanks Llorean, now i know what to use if i cant charge for a while
15:05:52 Join webguest07 [0] (i=d061b707@gateway/web/cgi-irc/
15:06:48 Join ducbian [0] (n=ducbian@
15:07:47 Join pondlife [0] (
15:07:47 Quit webguest07 (Client Quit)
15:07:50 Quit The-Compiler (Read error: 60 (Operation timed out))
15:07:57JdGordonok, now the selection bugs have been fixed.. whats the final decision on the acceleratio settings language?
15:08:04 Join The-Compiler [0] (
15:08:47pondlifeHmm, did anyone note that the SVN diffs aren't working?
15:09:03linuxstbYou mean viewvc ?
15:09:17JdGordonLlorean: ah, your icons problem is by design... without the .icons file there is no way to know that the icons are actually the corect height, and order
15:09:27pondlifelinuxstb: "SubversionException: ("Can't open file '/sites/svn/rockbox/format': Permission denied", 13"
15:09:37LloreanJdGordon: I thought it assumed they were the correct order, and the same height as the main icons?
15:10:06linuxstbpondlife: Zagor said he was investigating.
15:10:11JdGordonill double check, but I dont think so
15:10:39pondlifeNo problem, just thought I'd pop in and mention it as I couldn't see the error in the IRC logs. Bye for now.
15:10:46 Quit pondlife (Client Quit)
15:12:21 Join webguest06 [0] (i=d0438e04@gateway/web/cgi-irc/
15:13:29LloreanJdGordon: I thought that the viewer icons were supposed to work _if_ they were in the same order, and the right size. That's at least what you said originally
15:14:03JdGordonI said that? ok, ill modify the code..
15:14:07ducbianIs there a reason why it's a waste of my time to attempt to get a build running on my g2 ipod or is it simply because no current developer has hardware to develop for?
15:14:08 Quit webguest06 (Client Quit)
15:14:34linuxstbducbian: You mean the 2nd gen Nano?
15:14:59ducbianNo, I mean a 2002 gen2 touchwheel, kind of old but apparently has the same processor as the g3
15:15:10linuxstbThen no, it's not a waste of time at all.
15:15:11LloreanJdGordon: I _thought_ you did. I could be wrong.
15:15:23JdGordonLlorean: no, it doesnt make sense..
15:15:42JdGordonumm... hmm
15:16:05LloreanWell then, everything's working fine
15:16:19 Join BlueChip [0] (
15:16:19linuxstbducbian: What you will need to do is to look at the source code to ipodlinux (mainly the kernel, but the original podzilla source code could also help), see where different codepaths/functions are used on the g2 compared to g3, and port those changes to Rockbox.
15:16:27GodEaterlinuxstb: will ipodpatcher work on a 2nd gen ?
15:16:41JdGordonLlorean: yeah, it wont play nice if we allow it without the .icons file
15:17:12JdGordonI tihk, if you create the empty .icons file it will work as expected though
15:17:15linuxstbGodEater: Not sure. If it does, it will install the 3rd Gen bootloader, which may or may not work on the 2nd gen...
15:17:35GodEaterthat'll be interesting =/
15:17:41BlueChipHi. A very quick visit. I threw a guy on the m-list a link for SVN seems svn is dead - or maybe just viewvc? Thought it would be helpful to mention it here
15:17:48GodEateris there a hardware based disk mode in the early ipods ?
15:18:04GodEaterBlueChip: we're aware of the issue ;)
15:18:11BlueChipOK dude ;)
15:18:25linuxstbThe firewire chip is documented (I think), so I think a firewire driver could potentially be written.
15:18:28 Part BlueChip
15:18:35linuxstbducbian: Did the 2nd gen have USB, or just firewire?
15:18:41ducbianJust firewire
15:18:47GodEaterdocumentation, on an iPod component? Whatever next!
15:18:49JdGordonLlorean: definatly, if you create the .icons file with the only entry bei he highest value in the viewers.config file it will work
15:19:05linuxstbGodEater: It's external to the portalplayer chip IIRC.
15:19:25LloreanJdGordon: Highest value?
15:19:38LloreanJdGordon: As in, the largest number so it knows how many icons there are?
15:19:45linuxstbducbian: Have you tried installing the 3rd gen version of Rockbox on your ipod?
15:20:09JdGordonLlorean: yes
15:20:14JdGordonLlorean: your a lefty?
15:20:26LloreanActually, ambidextrous, but with a preference to my left.
15:20:40JdGordonok, startin to make thing clearer now...
15:20:42LloreanI'll catch a ball with my right if it's coming from that way, and that's a common test.
15:21:05JdGordonI ca see how tring to get thepower button with the left hand is a pita
15:21:18ducbianNot yet, I'm currently at work and our linux guy was bullying me about my moaning rather than using some of our resources to attempt to get rockbox working for my hardware.
15:22:16linuxstbducbian: That's probably the first thing to try, and take it from there... The pp5002 ipods are in need of love and attention - no-one is maintaining/developing them currently.
15:23:44***Saving seen data "./dancer.seen"
15:24:58JdGordonLlorean: at the expense of a few bytes, I could add an optoin so if there is a line in the .icons file to the effect of "use viewers.config" it would stop reading and assume its all correct, wold tha be ieal?
15:26:03LloreanJdGordon: What if the .icons file existed but was empty?
15:26:09LloreanInstead of looking for a line
15:26:39 Join Faemir [0] (
15:27:13JdGordonthat could work also
15:27:26LloreanI'm not sure though
15:27:35LloreanI mean, it's fairly easy for authors to just create the file. Heh.
15:27:48 Join pearldiver [0] (
15:28:02LloreanI don't think we're going to be getting _too_ many more iconsets anyway, interest in it seemed to be somewhat shortlived
15:28:18 Part david11
15:28:23ducbianOk thanks for the information linuxstb, I'll be back outside of business hours as I've probably used up my lunch break poking around the docs and chatting here. My other question was if there is a copy of the docs/contributing.html page because the linked one seems to be awol.
15:29:17 Join Arathis [0] (
15:30:18 Part Domonoky
15:30:45 Quit Thundercloud (Read error: 60 (Operation timed out))
15:30:49linuxstbducbian: That's just a temporay problem with the SVN viewvc script. Apart from coding guidelines (tabs, line lengths etc), the main contributing rules are: 1) Write a patch and submit it to the patch tracker; 2) No anonymous contributions are allowed, so you need to use your real name when contributing patches.
15:30:56*JdGordon cant remember how the .icons file worked :p
15:35:27ducbianFair enough.
15:39:31 Part LinusN
15:40:08 Part Llorean
15:40:14amiconnlinuxstb: Calling emergency diskmode when detecting firewire should be straightforward.
15:40:23*amiconn still wants to do that for the mini
15:40:41 Join kubiix [0] (
15:46:34 Quit pixelma (Read error: 110 (Connection timed out))
15:52:43 Join Psilonaut [0] (
15:54:34B4gderbtw, I've been offered a sansa e200 R model
15:54:42 Join EspeonEfi [0] (
15:55:06B4gderand if I get it I'll offer it to anyone who feels like poking it
15:55:47 Quit EspeonEefi (Read error: 110 (Connection timed out))
15:59:03linuxstbDid MrH have some success with it?
15:59:18B4gderhe doesn't have one
15:59:34B4gderwe have not been able to put our own bootloader on it
16:01:32Zagorviewvc running aain
16:01:57 Quit mpeccorini ("User pushed the X - because it's Xtra, baby")
16:02:44*linuxstb is drowning under new devices to work on, so should pass
16:03:04Zagorgotta go
16:03:06 Quit Zagor ("Client exiting")
16:03:15preglowi'm almost tempted
16:03:22preglowbut can't really guarantee poking time, hrm
16:03:57preglowwhat's the status on the thing anyway? didn't someone manage to flash it with ordinary e200 firmware?
16:04:20JdGordonI thought dan_a managed to put a regular firmware on it?
16:04:31B4gderwe can make a vanilla e200 into an R and back
16:04:42B4gderbut we can't modify an R at all
16:04:55linuxstbDo the DRM features work on a vanilla e200 turned into an R?
16:05:07B4gderI don't think anyone tried
16:05:37B4gderthe R models can't upgrade the bootloader from recovery mode
16:05:47B4gderand they don't have it in a second partition
16:06:19B4gderand they fixed the dummy signature flaw
16:06:23preglowwhere does that leave us? :-)
16:06:35linuxstbDo we have an unencrypted R bootloader?
16:06:46JdGordonand just to be annoying, I think the new OF for the e200 doesnt use a byte for the usb rebuild...
16:06:58B4gderusing e200tool we could rip the bootloader from memory
16:07:29B4gderand we even have a binary-patched version with supposedly removed signature checks
16:07:30linuxstbe200tool allows us to upload and run code on the device, commicating via USB?
16:07:45 Join mpeccorini [0] (
16:08:02linuxstbSo the problem is installing that bootloader?
16:08:56linuxstbHas anyone compared them physically?
16:09:13B4gderyou mean the hardwares? no I don't think so
16:09:21preglowwhat size are the e200 rs?
16:09:32 Join jgarvey [0] (
16:09:37preglowas in flash capacity
16:09:45B4gderpreglow: they look identical on the outside from the photos I've seen and they have the same flash sizes
16:10:15B4gderand they have radio
16:10:57preglowbut no, i can't really promise time for hacking
16:11:07preglowso i'll skip this one
16:12:03 Join JETC- [0] (
16:14:08 Join VPN-User [0] (n=Miranda@
16:14:18VPN-UserHello there
16:15:17VPN-UserI still experience charging problems with rockbox an my recorder v1 :(
16:15:39VPN-UserIt only charges for about half an our from 5% to 100%
16:16:03*preglow pats hardware controlled charging
16:16:21B4gderVPN-User: batteries in mint condition?
16:16:53 Join PaulJam_ [0] (
16:17:03VPN-UserThere are relatively new
16:17:18VPN-UserWhen I charge them in a dedicated charger, I get 15h playtime
16:17:32VPN-UserBut when charged with rockbox, only about one hour :(
16:17:48linuxstbVPN-User: Which version of Rockbox are you using?
16:18:01VPN-Userits a recent daily build
16:18:06VPN-User3 days old
16:18:20*B4gder haven't used his recorder in ages
16:18:25VPN-Useri also flashed it
16:18:32JdGordonanyone with a sansa and not the latest OF and 3min spare time? can you connect the sansa to the comp but not mount it, then reboot back into the of? does it rebuild the db?
16:19:03B4gderno sansa here
16:19:34linuxstbB4gder: Have you given them all away, or do you just not have one with you?
16:19:45B4gderI still have one at home
16:20:52 Quit mpeccorini (" HydraIRC -> <- Go on, try it!")
16:21:15VPN-UserI really love my v1 :)
16:21:21 Join mpeccorini [0] (
16:21:27VPN-UserBut with rockbox - currently unusable
16:21:43VPN-UserHas anyone a flashable stock firmware for me?
16:21:46linuxstbWhat if you charge with the original firmware?
16:21:55amiconnWhat rockbox version are you using?
16:22:01VPN-UserBecause when I ROLO into stock firmware, charging works fine
16:22:05amiconnIf it's 2.5, consider upgrading
16:22:14VPN-Userits a 3 days old nightly
16:22:22VPN-Usertried several
16:22:47amiconnWhat batteries are you using? Very high capacity ones?
16:22:53VPN-Userfirst i thought i have the solder-problem
16:23:23VPN-Userbut then it also wouldnt charge with stock firmware, nor?
16:23:33 Join Febs [0] (n=chatzill@
16:23:55amiconnNot necessarily. The stock firmware brute-force charges afaik least for a certain amount of time
16:24:56amiconnIt looks like rockbox thinks the batteries are full when they actually aren't, or even stopping completely because of some odd behaaviour
16:25:38amiconnIf it would just detect full, it should at least peform the 2 hour top-off charging...
16:28:08Faemirdoes anyone know what battery mAh a ipod video 5.5g is? or how to find out?
16:33:22 Quit PaulJam (Read error: 110 (Connection timed out))
16:34:09linuxstbFaemir: Have you read the section in the manual about that setting?
16:36:20 Join haemmy [0] (i=stefan@
16:38:27VPN-UserThe voltage increases very fast with these batteries while charging
16:38:32JdGordonhmm... viewcvs is still fubar
16:38:32VPN-Usermaybe thats the problem?
16:38:51 Join printfXh4 [0] (
16:39:46VPN-UserThese are batteries specially for MP3 players
16:40:02VPN-Useror electronic devices that need a constant high voltage
16:40:08B4gderJdGordon: where how?
16:40:50JdGordonviewing a diff
16:41:20B4gderPermission denied: '/dev/null'
16:41:49 Quit printfXh4 (Remote closed the connection)
16:42:07Faemirlinuxstb, i see.
16:44:53 Quit VPN-User ("Miranda IM! Smaller, Faster, Easier.")
16:45:54 Join printfXh4 [0] (
16:54:50 Join My_Sic [0] (
16:56:49 Join toffe82 [0] (
16:57:01 Quit printfXh4 (Read error: 113 (No route to host))
17:02:16 Join ender` [0] (
17:03:38 Quit B4gder ("It is time to say MOOO")
17:05:18 Join pixelma [0] (i=pixelma@
17:06:29 Join desowin [0] (
17:11:25 Join dimmok [0] (n=dimmok@
17:15:24 Quit Entasis (Read error: 104 (Connection reset by peer))
17:16:03 Quit JdGordon (Remote closed the connection)
17:16:25 Join Llorea1 [0] (
17:16:28 Join crop [0] (i=c27f0812@gateway/web/cgi-irc/
17:16:34croplinuxstb: ping
17:19:15 Join rift [0] (
17:20:46linuxstbcrop: hi
17:23:46***Saving seen data "./dancer.seen"
17:26:11 Quit lee-qid (Read error: 110 (Connection timed out))
17:27:02 Join david11 [0] (n=david52@
17:28:16croplinuxstb: I wanna complain :-) About "interval."
17:29:28linuxstbDo you have a better name, given jdgordon's description of how it works?
17:32:31 Quit dimmok ()
17:32:33 Quit Seed (Nick collision from services.)
17:32:40 Join Seed [0] (
17:32:41 Join Rick [0] (
17:32:51haemmyi have a question concerning album art: which are the requirements to see album art (i know that i need an build with albumart patch(i got senabs one), and a theme supporting it(black glass aa)
17:33:12linuxstbhaemmy: Have you seen this page?
17:33:18haemmyi have tagged mp3s
17:33:39haemmyi'll have a look at it ;)
17:34:49mpeccorinilinuxstb: is there a way to list all the documents in the wiki?
17:35:15mpeccorinilinuxstb: I'm planning to write a comprehensive guide to use the list widget, but I wonder if one already exists :p
17:35:58linuxstbI doubt there is such a document. You can search the wiki, or view the very long index (see that links at the top of every wiki page)
17:36:19 Quit shnee (Connection timed out)
17:36:45mpeccoriniAhhh, that's exactly what I was looking for, thanks
17:39:25mpeccoriniOk, there's no guide about how to use the list widget. It's confirmed. Do you guys think it's worth writting it?
17:40:01mpeccoriniI would have used one yesterday, it would have taken me 1 hour to do what I did in 3 LOL
17:40:33haemmy@ linuxstb: the link helped me thx
17:40:56 Join miepchen^schlaf [0] (
17:41:03 Join lee-qid [0] (
17:42:06 Quit Febs ("ChatZilla [Firefox]")
17:42:45 Quit amigan (Read error: 110 (Connection timed out))
17:44:13 Quit petur ("later")
17:45:12 Join shnee [0] (
17:46:13haemmyanother nooby question on ipod rockbox: is there a plan to use artwork providet by the itunes artwork db?
17:46:35preglowalbum art is going to be supported soon
17:46:53preglowrockbox is not going to care how you get that artwork
17:47:10GodEaterit won't be loaded from the itunesdb directly however
17:47:14preglownot all formats will be supported, though
17:47:20 Join jaczehack [0] (i=d572f7b7@gateway/web/cgi-irc/
17:47:36haemmyi thougt that to...
17:47:37linuxstbpreglow: IIUC itunes creates a database on the ipod which stores (as uncompressed bitmaps I think) the album art used by the Apple firmware.
17:47:39croplinuxstb: ehmm... sorry, was a bit off. Where is the description of how it works?
17:47:55preglowlinuxstb: not the worst arrangement ever
17:47:59linuxstbcrop: The IRC logs just before I invited you to complain :)
17:48:08preglowlinuxstb: it's probably going to be keyed by something we can't use, though
17:48:52linuxstbpreglow: Possibly. But the principle is very close to Rockbox's proposed album art support - just that Apple use a database instead of the filesystem.
17:50:32linuxstbhaemmy: I think it's unlikely that Rockbox will read album art directly from the itunes db - but it would be nice if someone wrote a program (or maybe even a plugin) to extract it to files Rockbox could use.
17:50:59haemmywould be very nice *g*
17:51:45croplinuxstb: just read the explanation. It doesn't tell much since it's not specified _how_ the items_to_skip is incremented.
17:52:43 Join nls [0] (
17:52:44preglowlinuxstb: since retailos uses itunesdb, i kind of doubt how much use we can find in the artwork db. we'll be unable to figure out what bitmap goes with what files anyway
17:52:46cropSorry, I must leave. Will read the log. And hope to continue the talk tomorrow :-) Bye
17:52:49 Quit crop ("CGI:IRC 0.5.7 (2005/06/19)")
17:53:12preglowlinuxstb: that is, of course, unless people just use the database
17:53:18preglowi never do that, so kind of forgot about it, heh
17:53:27haemmyif rockbox could use the itunes album art db, then it would not be far away to completely integrate itunes
17:53:51haemmythe itunes data on ipod
17:53:57GodEaterI'm not sure we'd be too keen to do that - not a lot of point
17:54:10GodEaterit's only one of our targets that uses iTunes
17:54:19preglowwell, if the integration is by the way of using the database and external tools: sure
17:55:07 Join Rondom [0] (
17:56:16haemmyso i'll transfer my songs + artwork to iPod with Anapod and then listen to them with Rockbox +g+
17:56:24 Join helixxo [0] (
17:56:47 Join Thundercloud [0] (n=thunderc@
17:56:52haemmyit's not a pity for me dont having album art everywhere
17:57:17haemmyif i want album art, i can start with original firmware +g+
17:57:45haemmyanother question: what about the battery?
17:57:51nlshmm is it known that the diff links on th efrontpage don't work?
17:58:23linuxstbnls: Yes. At least, B4gder knows.
17:58:40haemmyhow long can i listen non stop with an ipod video 30gb? would be interesting
17:59:11nlslinuxstb: ah, good, I guess he'll fix it when he has time to do so
17:59:31haemmyis there a significant loss of playing time?
18:00:10linuxstbhaemmy: Yes, Rockbox has bad power consumption on ipods. Around 7 hours I think is usual for the 30GB video according to this page -
18:01:42haemmyah.. turing down the display light will help a bit i think?
18:01:45helixxojust loaded rockbox onto my gigabeat, when i browse the files my directories are still there, but when i open them i just get a blank screen, no songs to select. is there a troubleshooting page that could help me here?
18:02:30 Join PaulJam__ [0] (
18:02:33helixxooh i see rockbox cant read sat
18:02:36 Nick PaulJam__ is now known as PaulJam (
18:05:59preglowhaemmy: possibly, but the biggest cause of power drain is unknown as of yet
18:06:33 Part david11
18:07:51haemmyunknown? -.- sounds mysterious
18:09:23 Quit shnee (Connection timed out)
18:09:27intgrHas Apple contributed any code recently? :)
18:11:15haemmyintgr : is apple supporting rockbox project?
18:11:32intgrOf course not, I was joking.
18:16:32desowinAustraiMicrosystems is only one supporter (from player makers)
18:20:37linuxstbI wouldn't really call them a supporter - companies that freely publish datasheets for their chips have contributed a lot more to Rockbox.
18:21:12 Quit PaulJam_ (Read error: 110 (Connection timed out))
18:21:14preglowwell, they are supporters in the sense that we've had an active dialogue with them
18:21:25pregloweven though we'd prefer it if the datasheets were freely available or not
18:23:04haemmyaustria :)
18:23:14linuxstbI agree, it's nice they made us send people to go and collect datasheets from their offices... ;)
18:23:33ducbianWhy do they keep them a secret, are they scared of Chinese knockoffs that are protocol / pin compatible or is it just reflex?
18:23:34haemmyi'm from austria too ;)
18:23:43nlsdesowin: SanDisk gave the project 2 or 3 sansas actually, we didn't get any docs on them though...
18:24:16preglowducbian: i think they're afraid users would fry their players writing illegal combinations of values to their chips
18:24:18desowinyou got developer boards didn't you ?
18:24:25preglowducbian: but that's rather silly, since that risk is even greater when you have no info
18:24:36linuxstbducbian: The cynic in me thinks that companies with strict NDAs regarding their documentation have something to hide.
18:25:01 Join shnee [0] (
18:25:05preglowi just think they have retarded management
18:25:14nlsdesowin: I think bagder got one with a jtag connector soldered on
18:25:34ducbianlinuxstb, I suspect it's actually more than many of these companies adopt a default position similar to everyone elses because it is easier than exploring the ramifications of making such documents publicly available.
18:27:19 Quit ducbian ("boom")
18:27:23 Join jac0b [0] (
18:28:48*amiconn tends to agree with linuxstb (cynic part)
18:28:50 Join JerryLange [0] (
18:29:17 Quit jac0b (Client Quit)
18:29:23JerryLangehello. is there anybody in here? i have a very important problem
18:29:51JerryLangemy ipod wont turn on at all
18:30:28markunis the battery charged?
18:30:52JerryLangeit was last night when i turned it off.
18:30:57PaulJamhave you already tried to reset it?
18:31:26JerryLangewith menu+select?
18:32:13JerryLangeok that worked.
18:32:28 Quit lee-qid (Read error: 110 (Connection timed out))
18:32:42AtGsomebody know if Duke Nukem 3D will work on Rockbox - H320 ?
18:32:57JerryLangewhy did it work like that?
18:33:03PaulJamif someone ports it, them maybe
18:34:13PaulJamJerryLange: i could imagine that the wasn't rally off, but that it hung with backlight being off.
18:34:14JerryLangemay i ask my my ipod would only turn on that way the first time today?
18:34:56preglowthe safest and cheapest way is to require ndas, so i understand them well
18:36:52helixxosweet i got it working :D
18:37:03helixxothanks for the neast program, guys
18:38:08markunAtG: if it already works?
18:38:12 Quit shnee (Read error: 60 (Operation timed out))
18:38:14markunor if it's possible to port it?
18:40:29 Quit JerryLange ("Leaving")
18:40:42 Join shnee [0] (
18:45:33 Join dan_a [0] (n=dan_a@
18:45:35 Quit dan_a (Remote closed the connection)
18:46:38 Quit jhMikeS (Nick collision from services.)
18:46:44 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
18:46:46 Join barrywardell_ [0] (
18:48:54*nls wonders if he made the top 10 highscore?
18:51:44mpeccorinin1s: hi buddy, thanks a lot for suggesting the List Widget. I did look a lot better. Did you test it?
18:52:02nlsmpeccorini: yep, looked good so I comitted it :-)
18:52:46mpeccorinin1s: gr8, I've got plenty other features to add ;-)
18:53:13nlsmore chessfeatures?
18:54:58 Quit GodEater ("CGI:IRC (Ping timeout)")
18:56:38mpeccorinin1s: yes, storing games you play against GnuChess, switching between viewer and game modes, a configuration screen with several options like autosave ...
18:57:35 Join pepie34 [0] (
18:57:38nlsmpeccorini: ah ok, sounds nice I guess you play a lot of chess on the go :-)
18:58:16mpeccorinin1s: yep, I'm a lousy player, I hope my iPod can help me LOL
18:58:45mpeccorinithese are not my ideas though, they're documented here
19:00:04nlsYeah I remember the guy that made the initial port of chessbox, he had a lot of ideas, but he vanished
19:00:35 Part shnee ("Konversation terminated!")
19:01:57 Quit Arathis ("Bye, bye")
19:02:13 Quit helixxo ()
19:02:34 Join n17ikh|Lappy [0] (
19:02:52mpeccorinin1s: with some time, they'll get done by someone
19:03:29mpeccorinin1s: he does have some very bold ideas though
19:05:17nlsyeah, people who are not regulars tend to do a couple of drive-by patches for the plugins so maybe one day they'll all get implemented :-)
19:08:21 Quit pepie34 ("Ex-Chat")
19:10:18n17ikh|LappyI have a question for people with the iriver H10 20gb. I have problems playing wavpack and high-bitrate (320kbit) mp3s
19:10:33n17ikh|Lappyseems like every once in a while the thing just gets overwhelmed and pauses for a few seconds and resumes
19:11:00n17ikh|Lappyonly happens in the WPS regularly, not in the file browser, although it does happen in the file browser occasionally
19:11:08n17ikh|Lappynot that I have a very complicated WPS
19:11:19nlsn17ikh|Lappy: are you using eq, crossfeed or dithering?
19:11:47n17ikh|Lappynot sure about EQ, but no crossfeed or dithering
19:11:59n17ikh|Lappylet me check
19:13:00n17ikh|Lappynot using the EQ, but I am using the coarse bass/treble adjustment
19:13:45n17ikh|Lappybut then again, mine isn't the very latest build, it's a week or two old
19:14:08nlsn17ikh|Lappy: I don't think the bass/treble controls should affect it, the problem is that eq and the other dsp options use a fair bit of cpu power
19:14:20n17ikh|Lappyyeah, I realize that
19:14:24n17ikh|Lappybut I'm not using any :/
19:14:44nlsand as the codec also wants a fair bit of cpu power, especially with higher bitrates there is not enough
19:14:48amiconnDoes the H10 have hardware tone control?
19:14:50n17ikh|LappyI thought the portalplayer CPUs were more powerful than that though
19:14:57n17ikh|Lappyno clue, amiconn
19:16:02nlshmm, seems they are software tone controls
19:16:16n17ikh|Lappyso turning those to zero would probably help
19:16:19nlstry torning bass/treble off (set them to 0)
19:16:26n17ikh|Lappyyeah, I'll give it a shot
19:16:32amiconnn17ikh|Lappy: Do you have peakmeters on your wps?
19:16:39n17ikh|Lappyit's the uh
19:16:39nlsn17ikh|Lappy: yeah it should help a little bit at least
19:16:40n17ikh|Lappyone sec
19:18:14n17ikh|Lappy"boxes" WPS that comes with rockbox
19:19:10n17ikh|Lappyno peak meters, it does have some graphical stuff that I could get rid of if I wasn't too lazy to look for/write a good wps
19:19:39nlsn17ikh|Lappy: the graphics shouldn't make a big difference
19:20:12n17ikh|Lappyhow much of the screen does it redraw every refresh?
19:20:22n17ikh|Lappy(I don't know how rockbox does screen updates)
19:20:27amiconnIt's dynamic
19:20:37n17ikh|Lappyso just what's changed then
19:20:43amiconnIt only refreshes what's changes, approx. twice per second
19:20:54amiconnOnly peakmeters are drawn more often
19:21:48n17ikh|Lappywell, I guess it's just a limitation of the portalplayer CPU :/
19:22:04n17ikh|Lappywonder if it's possible to get it to clock up when plugged in
19:22:20 Join a1titude [0] (
19:22:21n17ikh|Lappyperversely, I don't have the problem when on battery because I have the screen set up to time out
19:22:28nlsn17ikh|Lappy: also rockbox does not make full use of the second core in the processor yet
19:22:39n17ikh|Lappyso, eventually it might do better
19:22:41amiconnThe problem is that we can't use the pp like we'd want to atm
19:23:20n17ikh|LappyI see.
19:23:50***Saving seen data "./dancer.seen"
19:24:23n17ikh|Lappythat's part of the summer of code project, right?
19:25:48a1titudeI was actually going to ask a question about battery life, I was wondering how long do you immagin that it will take for you (the rockbox developers) to find the drainage problem and solve the short batterylife for the portalplayer chip based targets? - I'm reluctent to use rockbox for audio playback until it gives bettter battery life. :(
19:26:23nlsa1titude: we don't make ETAs
19:26:46nlsbecause we don't know
19:27:15a1titudeahh, well thanks
19:28:40nlsa1titude: although one problem that is known is that nobody knows how to power down the harddrives in the ipods (does not affect h10 where that works or flash based players)
19:29:50a1titudeapple should make it easyer for rockbox developers when making the hardware! haha jk lol
19:33:28 Quit toer (Remote closed the connection)
19:35:05 Part a1titude
19:45:45 Join f00f [0] (
19:45:56 Nick f00f is now known as zobo (
19:47:57mpeccoriniguys, is Kevin Ferrare still contributing to Rockbox?
19:49:18 Join maffe [0] (n=maffe@
19:49:44desowinmpeccorini: why not ask him directly ?
19:49:58*amiconn hasn't seen him for ages
19:50:26nlshis last commit was on Jul 2nd 2006
19:50:42 Join PaulJam_ [0] (
19:50:49haemmyhm is there a build, where ipod could shut down the display during playing audio until a button is pressed?
19:51:07mpeccoriniI just wrote and I was hoping he could take a loot at it
19:51:25mpeccorinibut I should probably just go get a life :p
19:51:54desowinhaemmy: atleast on mini it didn't gave me any change, if you have grayscale iPod I could check if my old patch still applies
19:52:10desowin(change in battery life)
19:52:35amiconnSwitching off the lcd doesn't make sense wrt battery life
19:52:47amiconnWhat sucks power is the backlight, not the lcd itself
19:53:22amiconnAnd if the lcd stays readable w/o backlight, it doesn't make sense to switch it off. That applies to all ipods except the color/photo
19:53:25haemmydesowin : no i got a ipod video 30gb
19:54:29desowinhaemmy: sorry, I can't help then
19:56:10*desowin was one who thought he know it better, but it just paid me many hours on trying to make patch and then notice that there's no difference, but it's good, I no longer argue with you guys
19:57:21haemmysry, didn't know that
19:58:44 Quit EspeonEfi ("Leaving")
20:00:30 Quit Nibbier (Read error: 110 (Connection timed out))
20:01:19 Join ducbian [0] (
20:02:58 Join Nibbier [0] (
20:03:46 Quit n17ikh|Lappy ()
20:04:12zobodoes anyone know if rockbox implements some sort of buzzer?
20:05:02scorchethere is a ptch on the tracker for the piezo, if that is what you mean
20:05:03 Quit haemmy ()
20:08:09 Join midkay_ [0] (
20:09:56zoboscorche, the piezo patches deal with button presses, correct?
20:10:48desowinthose are old from what I know, probablt won't apply to svn now, someone would have to sync them
20:10:59 Quit PaulJam (Read error: 110 (Connection timed out))
20:11:14desowin(and correct)
20:12:35riftfor doom, i put the .wad on .rockbox/games/doom or in .rockbox/doom ?
20:13:00linuxstbWhatever the manual says...
20:13:48zobobut say I want to generate beeps at specific time intervals as opposed to buttons being pressed.
20:14:31linuxstbAre you talking about beeps in your headphones, or the ipod's piezo?
20:14:53 Quit barrywardell_ (Read error: 110 (Connection timed out))
20:15:24zobolinuxstb, I just need an audiable beep to be created at certain time intervals.
20:15:57 Join david11 [0] (n=david52@
20:17:12zoboFor example, I often do High-intensity Cardio w/ my Sansa. I run for a minute, then walk for a minute, then run for a minute again... etc.
20:17:27linuxstbSo you want it whilst listening to music?
20:17:48zoboIt would be much more prctical to hear when the interval has elapsed than having to glance at my stopwatch every couple of seconds
20:18:00zobolinuxstb, ideally, yes.
20:18:22zobofade music, hear beep.
20:18:36zobois there a plugin for something like this already?
20:19:03 Quit sando ("Apply directly to the forehead")
20:19:26linuxstbI don't think such a plugin exists already, but there's the pcmbuf_beep() function available to plugins, so a plugin could be written to use that. (although I'm not sure how well that function works)
20:26:03 Quit midkay (Read error: 110 (Connection timed out))
20:26:14 Join toer [0] (
20:27:46Nico_Plooks like I' not getting any Flysspray notifications anymore...
20:28:42 Quit atsea- (Read error: 104 (Connection reset by peer))
20:29:51linuxstbNico_P: You mean personal emails or the mailing list?
20:30:01Nico_Poh actually it's ok I got one today :)
20:30:02 Join Lear [0] (i=chatzill@rockbox/developer/lear)
20:30:20Nico_Pit seems the notifications were disabled with the new version of FS
20:30:37Nico_Plinuxstb: I was talking about personal emails for watched tasks
20:31:22Nico_Plinuxstb: btw, what do you want to do with your checkwps tool ?
20:32:04linuxstbI'm happy to commit it, but want someone (you?) who knows the WPS code to help improve it - it's just a proof-of-concept really at the moment, rather than a useful tool.
20:32:53Nico_Plinuxstb: what kinf of improvements do you have in mind ?
20:32:53 Quit ducbian ("Leaving")
20:32:53linuxstbMainly two things: 1) Check that the changes to core rockbox code are OK, and that they can't be done any less intrusively; 2) Change the application itself to provide more useful output.
20:33:18Nico_Pwhat sort of output ?
20:33:46linuxstbYou think it's fine as it is? I don't write WPSs, so don't know what WPS authors will want to be told.
20:34:38*linuxstb can't remember exactly what the output looks like, apart from the fact that it seemed very verbose
20:35:01Nico_Pit is very verbose... maybe that can be changed with command line arguments
20:35:32Nico_PWPS creators seemed to be quite happy with the output
20:36:03linuxstbFair enough then.
20:36:27Nico_PI'll check the core parts
20:37:07 Join webguest01 [0] (i=43ad8a99@gateway/web/cgi-irc/
20:37:38webguest01hi can anybody give me a link to the current build for my ipod video plz?
20:38:33 Join aa234238 [0] (
20:39:13aa234238Good evening all, I was hoping someone could advise me - in its current state, is Rockbox on a 5G video iPod capable of decoding and mixing two streams of audio for >30 secs?
20:39:13 Quit zobo (Read error: 54 (Connection reset by peer))
20:39:27 Join zobo [0] (
20:40:08linuxstbwebguest01: Click on "Current Build" in the menu on the left of the Rockbox website, and scroll down until you see a photo of your iPod. Then click on it.
20:40:57webguest01thank u
20:41:33 Quit webguest01 (Client Quit)
20:41:41linuxstbaa234238: There's no feature to do that currently. Are you asking if the ipod hardware would be up to it if it was implemented?
20:41:43 Join webguest75 [0] (i=43ad8a99@gateway/web/cgi-irc/
20:42:58aa234238yeah basically.. well my idea (/dream, it would take a long time for me to get anywhere near implementing but i love a good challenge..) is to have precalculated info on bpm/beat timings of songs and for the ipod to resample the next song to match the tempo and x-fade the two
20:43:00aa234238dj style
20:43:04aa234238just for fun really ;)
20:43:47aa234238see, x-fade obviously exists, but i'm unsure if it actually decodes two streams at once or whether it uses the last few seconds when the other track has been decoded to start decoding the next one
20:43:58aa234238i did have a look at the sources a while ago but it was a bit confusing, thought i'd see if anyone knew the answer
20:44:02 Quit Rob222241 (Read error: 110 (Connection timed out))
20:44:20aa234238my guess is that mp3 decoding would have to be happening at well over 200% realtime for this to even be feasible
20:44:25linuxstbYes, crossfade starts to decode the second track after it's finished decoding the first. Only one codec is running at a time.
20:44:35 Join lee-qid [0] (
20:44:49linuxstbaa234238: This page shows some decoding speeds (on the iriver H10, but it has the same CPU as the ipod video) -
20:45:25aa234238ah excellent, thanks for that
20:46:19linuxstbNote that the MP3 figures are for a version of the decoder with some as yet uncommitted optimisations. Current SVN is slower.
20:46:20 Quit zobo (Read error: 54 (Connection reset by peer))
20:46:40*linuxstb just notices the page has two sets of MP3 figures....
20:46:46 Quit Faemir (Read error: 104 (Connection reset by peer))
20:46:55aa234238180%.. hmm, that would allow ~3 mins to decode the next song on a six min song... but then there's the memory overheads to consider
20:46:55 Join zobo [0] (
20:46:55 Join Faemir [0] (
20:47:21aa234238and woudl i be right in thinking that resampling works by changing something in the hardware, i.e. its a global thing rather than a "per-stream" thing?
20:47:41linuxstbNo, the DAC is fixed at 44.1KHz - all resampling is in software.
20:48:02aa234238ahh, interesting
20:48:21aa234238so what i'm suggesting might not be impossible, as long as memory isnt the limiting factor
20:48:30aa234238i might look in to it a bit more
20:48:39aa234238as i say, it'll be a fun challenge if nothing else ;)
20:48:41 Quit pearldiver (Read error: 60 (Operation timed out))
20:49:03aa234238- is the overhead of resampling in s/w high do you know btw?
20:49:14linuxstbI'm not sure why you need to decode the whole song in advance though - you can resample the second song as it's playing.
20:49:44 Join pearldiver [0] (
20:49:49Nico_Plinuxstb: I don't see a problem with the core parts of the patch... But I get weird warnings about printf when I build the tool
20:49:49aa234238yea, you're absolutely right. i was more thinking in terms of i'd like there to be a ~45sec period of time where both songs are playing
20:49:54aa234238which looks like it might not be impossible
20:50:19linuxstbYou could just increase the current crossfade buffer.
20:50:41linuxstbNico_P: Yes, I get those too.
20:50:59aa234238indeed, then write software on the pc side to analyse the tracks, and routines on the ipod/rockbox side to use this info to synchronise the tracks (assuming they are perfectly in sync)
20:51:01 Join atsea- [0] (i=ariel@gateway/tor/x-3a1e6f8a6f60ee95)
20:51:01 Quit webguest75 ("CGI:IRC (EOF)")
20:51:08aa234238i think it'd be a quite cool feature, i'll look in to it more
20:51:10aa234238thanks for you advice mate
20:51:16Nico_Plinuxstb: no idea how to fix them, or even where they come from ?
20:51:18linuxstbNico_P: I think it's a problem with the system printf getting in the way of the Rockbox printf.
20:51:18aa234238if i carry on with it i guess you'll see more of me here ;)
20:51:31Nico_Pshouldn't we be using the system printf ?
20:52:02linuxstbNico_P: Yes, we should - and I think we are... I probably meant to say the Rockbox printf is getting in the way of the system one.
20:52:11Nico_Pok :)
20:52:20 Quit zobo (Read error: 104 (Connection reset by peer))
20:52:27aa234238linuxstb: by the way, ignoring obvious preferences like UI etc, will i have a harder time getting a build environment going under linux or under windows?
20:52:38 Join zobo [0] (
20:52:54Nico_Paa234238: the build env is meant for linux
20:53:00 Quit zobo (Read error: 54 (Connection reset by peer))
20:53:27 Join zobo [0] (
20:53:35linuxstbaa234238: Linux is easier (and compiles faster). On Windows, your options are cygwin or Linux running in vmware.
20:54:13aa234238yeah, i was thinking i could use colinux under windows (not checked if that project is still alive tho..) - however been looking for an excuse to use XGL more so linux it is :)
20:54:24linuxstbAh yes, colinux works as well.
20:54:51linuxstbI think you'll find info on using colinux for compiling Rockbox on the Rockbox website.
20:55:07aa234238cool, i'll have a good read of the wiki and look through the sources
20:55:09aa234238thanks for the help
20:55:31linuxstbyou're welcome.
20:55:54linuxstbNico_P: I'll commit checkwps-v3.diff then?
20:56:10 Quit zobo (Client Quit)
20:56:43Nico_Plinuxstb: I'm fine with it. maybe someone will come up with a solution for the warnings
20:56:53 Quit aa234238 ()
20:57:02 Quit secleinteer (Read error: 110 (Connection timed out))
20:57:14Nico_Pand the functions you moved feel more at home in wps_parser.c
20:59:06 Join chainlynx [0] (
20:59:41 Quit XbooX (Read error: 110 (Connection timed out))
21:00:53linuxstbNico_P: OK, committed...
21:01:01Nico_Pcool :)
21:01:49preglowis gigbeat s still sold?
21:02:38david11oh sorry
21:03:00 Join krush1704 [0] (
21:04:51linuxstbpreglow: google seems to be telling me no.
21:05:00preglowdamn, shelflife is getting short
21:05:08Nico_Ppreglow: there are some on ebay
21:05:19Nico_PI think the S is still sold in some places
21:05:30preglowhmm, anyone know if the gigabeat s code injection tool works with zune? :/
21:05:49Nico_Ppreglow: it doesn't
21:05:51linuxstbIt doesn't. I think the exploit on the S doesn't exist on the Zune.
21:05:59Nico_PI was going to add that :)
21:06:09preglowwould be seriously cool with an fpu target for rockbox...
21:06:09nlsis someone tempted by the 500MHz arm + vector fpu? ;-)
21:06:29Nico_Ppreglow: get startedon the port then :)
21:06:37LearThat's what the Zune has?
21:06:41preglowLear: aye
21:06:52preglowNico_P: i'd need a unit first
21:06:53preglowalso time
21:07:04linuxstbLear: And the Gigabeat S.
21:07:23 Part The-Compiler ("Connection reseted by god")
21:07:36*linuxstb is starting to see his Gigabeat F as slow...
21:07:40Nico_Ppreglow: toffe82 might be able to get you one
21:09:13 Join Tachikoma [0] (
21:09:28toffe82why a zune ? the only difference with the S is the wifi if I remember
21:09:45linuxstbIt's a currently available device?
21:09:46Tachikomadoes the 4th gen ipod have a colour screen?
21:10:03nlslinuxstb: not on this side of the pond
21:10:23nlsTachikoma: some do, they are called ipod photo/color
21:10:26linuxstbTachikoma: The one with the colour screen does... There are two 4th gen - one is an greyscale LCD (normally just called "4th gen") and one with a colour LCD - called the Photo or Color.
21:10:28LearHe, found a page claiming that the Broadcom chip is a 300 MHz ARM11 and a 150 MHz "video core"...
21:10:43preglowi'm pretty sure it's arc based
21:10:45Tachikomai ordered one off ebay
21:10:59Tachikomaand can tell if its the grayscale or the photo
21:11:02preglowbut that might be bs
21:11:13linuxstbTachikoma: If it's just advertised as "4th gen" then it will most likely be the greyscale version.
21:11:52Tachikomawell i was looking on youtube and this guy had a video of a 4th gen with colour
21:13:07linuxstbThe 4th gen with colour is more often called "ipod Color" or "ipod Photo".
21:13:34Tachikomai know about the ipod photo
21:14:21linuxstbIf the ebay advert didn't advertise colour, then I would be very surprised if it was.
21:14:49Tachikomaalot of them don't
21:14:58Tachikomait looks like it has colour
21:15:00 Join otih_ [0] (
21:15:34Tachikomaonly reason i am getting it it to install linux on it
21:16:09linuxstbYou mean Rockbox? ;)
21:16:10Tachikomai am a huge linux nut
21:16:20Tachikomapodzilla or rockbox
21:16:52linuxstbpodzilla is an application that runs on the ipodlinux kernel. Rockbox is a complete standalone firmware replacement - nothing to do with Linux.
21:17:14Tachikomai want
21:17:53 Quit Jon-Kha (Read error: 145 (Connection timed out))
21:18:11Tachikomai just download the installer from the rockbox site?
21:18:57 Join robin0800 [0] (
21:19:38nlswow, the gigabeat S is almost 9mm thinner than my h320 and slightly smaller in the other dimensions too /me wants one
21:19:50 Join Buschel [0] (
21:21:58*slarti likes his iaudio x5l... the battery life is incredible
21:23:35 Join shiftplusone [0] (
21:23:51***Saving seen data "./dancer.seen"
21:25:11 Quit thegeek (Read error: 110 (Connection timed out))
21:25:34 Quit Tachikoma ("KVIrc 3.2.0 'Realia'")
21:25:55 Join perldiver [0] (
21:26:29 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
21:26:55 Quit linuxstb (Nick collision from services.)
21:26:58 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
21:28:55 Quit otih (Read error: 110 (Connection timed out))
21:30:20linuxstbmpeccorini: Have you thought about updating the manual for chessbox?
21:34:18 Part david11
21:36:10 Join netmasta10bt [0] (
21:37:45netmasta10btis FM working on sansa?
21:38:21nlsnetmasta10bt: not in the official builds, there's a patch for it though
21:38:33netmasta10btnls: great thx much
21:39:41 Join secleinteer [0] (
21:39:49 Join thegeek [0] (
21:41:33 Quit thegeek (Read error: 104 (Connection reset by peer))
21:42:22 Quit pearldiver (Read error: 110 (Connection timed out))
21:42:55mpeccorinilinuxstb: no I had not thought about that, but now that you mention it, I definitely will
21:43:20 Join thegeek [0] (
21:43:31Buschelprelow: you are there?
21:44:47mpeccorinilinuxstb: however, I think I'd rather spend some more time adding the features listed in the wiki and take care of the manual when they're done
21:44:59mpeccorinilinuxstb: what do you think?
21:46:01linuxstbmpeccorini: It's up to you what you want to work on - I just thought I would mention it as I see you had a desire to document things....
21:48:20 Quit jaczehack ("CGI:IRC (Ping timeout)")
21:48:25mpeccorinilinuxstb: hehe, if you mention that because of the ListWidget in the wiki, that was because I would have been VERY greatful if that had already existed ;-)
21:48:58mpeccorinibut yes, I will update the manual, there's no point in adding features if no one can use them
21:49:10linuxstbOr if no-one knows about them...
21:50:03linuxstbIf you want to become an A-list patch submitter, you could even include manual patches with your code patches...
21:50:44mpeccorinisounds good, I didn't know the manual was in the SVN repository ...
21:50:57mpeccoriniwhat format is it written in?
21:51:01mpeccoriniplain text?
21:51:04linuxstbIt's LaTeX.
21:51:45nlsmpeccorini: it's quie easy to add stuff to it, just look at what's already there
21:51:59mpeccoriniI will
21:52:25mpeccoriniand if adding them to the patches is better, I'll do it that way
21:52:39mpeccorinithem = the changes to the manual :p
21:53:23nlsmpeccorini: yeah sure, although I would like to have the manual changes in a seperate patch which of course can be posted in the same task
21:53:43nlsbut either way is absolutley fine :-)
21:54:16mpeccoriniif it works better for you, I'll have it in a separate patch
21:54:43mpeccorinibut in that case, I'll have a patch just for the manual so it's updated with the features already added
21:55:44*mpeccorini thinks LaTeX is a little cryptic, even more than the C language
21:56:26*linuxstb suspects mpeccorini hasn't done much Perl
21:56:45*mpeccorini hasn't done any Perl at all
21:56:53*preglow doesn't get why people think perl looks cryptic
21:57:02 Join Jon-Kha [0] (
21:57:14preglowBuschel: i'm here now, but i've got to go pretty soon and haven't got time for too much
21:57:24*mpeccorini thinks it's probably because they don't know it
21:57:36preglowalso, it uses $ and @ characters
21:57:54Buschelpreglow: that's ok. you've made up your mind reagrding the decoder optimizations?
21:57:56preglowbut they really do clarify instead of obfuscate
21:58:28preglowBuschel: haven't had time for rockbox stuff at all, sorry
21:58:39linuxstbpreglow: I don't think it looks cryptic, I think it is cryptic - i.e. I have no idea what's going on.... But that's obviously because I don't know it well enough.
21:59:02preglowBuschel: probably won't have time until after the weekend either
21:59:04Buschelpreglow: :o) same to me, our customer is pushing a lot
22:00:19Buschelpreglow: just to let you know -> even if still using full precision (all 32bit-defines unused) the performance is +10% on ipod and about +3-4% on coldfire
22:01:19mpeccorinilinuxstb: how can I preview the changes to Tex files, can you recommend a good/light/free/easy-to-use viewer?
22:01:40mpeccorini... for windows
22:03:54preglowBuschel: yeah, aware of it
22:07:54amiconnThis is musepack, correct?
22:08:14linuxstbmpeccorini: You "compile" it to either PDF or html.
22:08:41 Quit rift ("Lost terminal")
22:09:09linuxstbmpeccorini: Create a build directory (e.g. build-manual), run ../tools/configure, and select "M" for manual. Then type "make" (for PDF) or "make html".
22:09:32amiconnI mean, the various mp3 decoders don't produce bit identical output, there are those various levels of compliance
22:10:00mpeccorinilinuxstb: ok, thanks
22:10:28 Part maffe
22:10:45 Join maffe [0] (n=maffe@
22:11:27Buschelamiconn: yes, i was talking of musepack
22:11:58linuxstbBuschel: (you may already have done this) but regardless of what gets enabled when your changes are committed, it would be nice to have a #define to control the accuracy, so both options are still available in the code.
22:12:29amiconnHow would the amount of deviation caused by the additional optimisations relate to what the mp3 compliance levels allow?
22:12:34Buschellinuxstb: that's exactly the way the patch is done
22:12:44linuxstbBuschel: I'll stop stating the obvious then ;)
22:12:47amiconnDo all mpc decoders up to now produce bit identical output?
22:13:40 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
22:13:52Buschelamiconn: no, normally the difference is +/-1 over 16bit range
22:14:22saratogahow many MPC decoders are there? The official one and peters right?
22:14:44 Join TrueJournals [0] (
22:15:19saratogapreglow: I saw a used gigabeat S go on ebay for < 100 shipped the other day
22:15:26 Join PaulJam__ [0] (
22:15:29linuxstbsaratoga: Do you have a stripped-down wma2wav based on the current ffmpeg?
22:15:35Buschelthere is a long history of official ones −− there was lots of optimizing in the past (especiall around 2000-2002)
22:15:45 Join pepie34 [0] (
22:15:52saratogaif you're interested in the port, I'd be willing to kick in 30 USD for one, maybe others would
22:16:15saratogalinuxstb: yes I do, as well as an updated patch from the old code
22:16:20saratogado you want either?
22:16:35linuxstbI would be interested in the version based on current ffmpeg.
22:16:38amiconnBuschel: I just want to get an idea what enabling these additional optimisations means regarding sound quality
22:16:45saratoga(i'm afraid the rockbox decoder still has issues with the huffman coding though)
22:17:01saratogaok, i'll upload it
22:17:02amiconnIt might not be bad; lossy audio compression is always about making compromises
22:17:16toffe82saratoga: it was a working S ?
22:17:25linuxstbsaratoga: Is it just a program that you link with ffmpeg, or is it using the ffmpeg .c files directly, with your own makefile.
22:17:28amiconnIf one doesn't want to make compromises, he'd go lossless
22:17:30 Quit saratoga (Client Quit)
22:17:35 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
22:17:41 Quit Psilonaut (Read error: 110 (Connection timed out))
22:17:59saratogamy own makefile
22:18:07saratogai do not build ffmpeg
22:18:32Buschelamicon: if you take a look at the images i submitted in the patch-description, you will see the amount of additional noise which is committed by the 32bit-optimization
22:18:34preglowamiconn: the differences seem to be unnoticable
22:18:36 Join Alldogship [0] (
22:18:51Buschelamicon: noise is below masking...
22:19:34linuxstbsaratoga: Have you spent much effort in stripping it down, or is it just at "first-attempt" stage?
22:19:45*amiconn didn't use mpc so far
22:19:45Buschelamicon: or better -> additional noise is below noise which is added through encoding
22:19:56saratogaa fair amount of effort, but it still needs a lot more
22:20:18saratogatheres a lot of crap left in the various structs, and some of the ffmpeg abstraction stuff is still there for the decoder initialization
22:20:40saratogabut its more stripped down then marsdaddy's
22:21:11 Quit PaulJam__ (".")
22:21:31saratogai'm actually very impressed with John Paul's decoder
22:21:51saratogai think it should be very fast if he'd bothered to do the fixed point stuff in asm and to use iram
22:22:17saratogagiven the issues he had with c fixed point and running his look ups in DRAM, i'm not surprised it was slow
22:23:04saratogabut even in 32 bit precision, he has no overflows at all in my test files, and only occasional problems with some as yet unidentified bug in the huffman coding I think
22:24:35saratogahere you go
22:25:21preglowyeah, he mentioned one bug, i remember
22:25:27preglowbut why would huffman decoding bug out?
22:25:40saratogano idea
22:25:44saratogabut it appears to
22:25:54saratogararely, it'll return garbage for a couple hundred frames
22:26:05saratogaat least i think its huffman, or unpacking the exponents or something
22:26:13saratogaits not in the MDCT or windowing at least
22:26:28saratogai spent all night going through those and they appear to work correctly
22:26:34linuxstbAre you able to create a very small WMA file that exhibits the problem?
22:26:39saratogai think he just added a bug somewhere
22:27:08saratogaupdated the wmaplay zip
22:27:16saratogaso redownload it
22:27:26saratogathe one i put up was seeked to a problem spot in a test sample
22:27:43saratogatest sample:
22:27:58linuxstbsaratoga: That looks a managable number of files...
22:28:33saratogaa lot of them are half #if 0 and not much else too!
22:28:36 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:28:46 Quit amiconn (Nick collision from services.)
22:28:47 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:29:00 Quit schula (Read error: 104 (Connection reset by peer))
22:29:11saratogawmatest2.c is the loader, its pretty simple once the ffmpeg stuff is out of the way
22:30:09saratogabasically, theres an open on the input file and codec which call the init functions in the asf and wma c files respectively
22:30:11linuxstbHmm, I get a segmentation fault running "./wmaplay g_128k.wma"
22:30:27saratoga./wmaplay g_128k.wma out.wav
22:30:32linuxstbAh, OK
22:31:08linuxstbStill a segmentation fault...
22:32:19saratogaoh me too
22:32:22saratogalet me see
22:32:46saratoga ./wmaplay g_128k.wma out.wav b
22:32:53saratogasorry, dumb debug code i was just trying
22:33:10saratogai made it look for a third input that triggers different debug screens
22:33:13saratogai should comment that out
22:33:26linuxstbNot to worry.
22:34:33saratogaas you can see i was just trouble shooting the MDCT and window length code in that build, so you get a lot of crap
22:35:00saratogabeen dumping the coefficinets in both versions to see whats wrong with the integer one
22:35:31saratogawhat were you looking for anyway?
22:35:35preglowsaratoga: you checked out if the bugging happens both on target and computer?
22:35:49saratogai can't build for the target yet
22:35:52saratogafor lack of a target
22:35:57 Quit PaulJam_ (Read error: 110 (Connection timed out))
22:36:08saratogai ordered a sansa and a gigabeat, but the holidays in the US mean the mail is very slow
22:37:13preglowyou have no rockbox targets? :
22:37:22toffe82saratoga: what gigabeat ? S30 or S60
22:38:03 Join jac0b [0] (
22:38:44 Quit jac0b (Client Quit)
22:39:15 Quit shiftplusone ("Leaving")
22:39:24linuxstbsaratoga: The reason I asked was that I'm thinking (but not really that seriously) about porting the realaudio Cook decoder. So thought your stripped down ffmpeg could help...
22:39:48saratogatoffe82: sorry, just an F20
22:40:28saratogalinuxstb: sure, could help you too
22:40:30toffe82I thought you were interested in a S ;)
22:40:34 Quit Rondom (No route to host)
22:40:40saratogaI am, but no time to port that now
22:40:47saratogagot to get this WMA thing working
22:40:56saratogaone project at a time
22:41:00*bluebrother returns and notices that some LaTeX talk has happened ...
22:41:17 Quit Alldogship ("Leaving")
22:41:37 Join Rondom [0] (
22:41:37saratogabasically, my stripped down version would work great, just got to change it to register the RA demux and decoder instead of ASF or WMA
22:42:16linuxstbsaratoga: That's what I thought...
22:42:20 Quit desowin ("use linux")
22:42:31saratogawhat is cook anyway?
22:42:35saratogai thought real used AAC now
22:42:43preglowcook is an older format, afaik
22:42:55saratogaits already fixed point?
22:43:24linuxstbI thought it was, but just checking the code now, it doesn't seem to be... Maybe the patch wasn't committed.
22:43:35preglowby the same dude that said he'd tackle wma
22:43:38preglowlinuxstb: oh?
22:45:02saratogathe GSOC projects have now started right? has austriancoder said anything about his USB project?
22:45:52 Join webguest51 [0] (i=18976e30@gateway/web/cgi-irc/
22:46:16preglowsaratoga: they started as of yesterday, afaik
22:46:46preglownico_p is around, so i assume he's started his. the speec synth guy was busy until june 10th, and ac i haven't heard from
22:47:43 Part TrueJournals
22:47:49saratogai created a wikipage for my project in order to save people digging through IRC
22:48:04saratogaif anyone sees AC, bug him to do the same, so i can stop asking about it
22:48:44linuxstbpreglow: Unless I'm missing something, it seems the patch was never committed in the end...
22:49:03saratogaironically, id rather his project work then mine
22:49:32saratogaoutside of some sansa users, who cares about wma
22:49:36preglowi'm really quite excited about all four projects
22:49:49saratogaits just mp3 with no subbands and variable MDCT window sizes
22:49:56preglowsaratoga: hell, i do, i have a wma file i'd like to be able to play :)
22:50:01linuxstbsaratoga: wma is popular amongst iriver users as well...
22:50:17bluebrotherwho needs users anyway? ;-)
22:50:19*linuxstb hopes he can still find his wma file
22:50:33Nico_Ppreglow: I haven't realy started mine
22:50:35preglowsaratoga: that pretty much sums up aac too...
22:50:35saratogai linked my one WMA file if anyone needs it
22:50:54linuxstbSo the three people showing interest in WMA have one file each? ;)
22:51:08saratogaits amusing how no one has really come up with a way to really beat MP3
22:51:12preglowNico_P: get going!"¤!"¤!"
22:51:17saratogajust fix the stupid bugs in the format
22:51:27saratogabut still leave the idea intact
22:51:41preglowsaratoga: turns out mp3 was pretty flexible after all, yeah
22:51:52Nico_Ppreglow: :) I need to figure some things out
22:52:12Nico_Pif you know the playback engine I'll appreciate your help
22:52:15saratogathe one cool thing about WMA is the many block sizes you can use, I haven't heard of that before
22:52:22preglowNico_P: i happen to know absolutely nothing about it
22:52:32preglowNico_P: try getting hold of jhMikeS, he knows a hing or two
22:52:34preglowthing, yes
22:52:34 Quit webguest51 ("CGI:IRC (EOF)")
22:52:48 Join jimh_jimh [0] (i=18976e30@gateway/web/cgi-irc/
22:52:48Nico_Pwill do
22:53:04preglowsaratoga: as in it uses more than two in one file?
22:53:29saratogathey're variable in powers of two from 128 to 4096 IIRC
22:53:38preglowvorbis also has tons of window sizes, but only two can be used in one file
22:54:14 Join XbooX [0] (n=xboox04@
22:54:14 Quit XbooX (Remote closed the connection)
22:54:19nlsNico_P: lostlogic also knows a fair bit about the playback engine, although he's not around much nowadays
22:54:33saratogashould give wma very good transient handling I guess, but i have to wonder how they deal with the MDCT reconstructon, since they'd need a ton of different windows
22:54:40Nico_Pyeah, I'll try to ask him for advice if he shows up
22:54:57 Join stripwax [0] (
22:55:04Nico_Pwhat I'd like to do is set up a branch for my work so that people who know the code can review it and maybe help out a bit
22:55:33preglowsaratoga: well, from what i hear, wma has nasty pre-echo tendencies which point to anything but a good transient handling
22:55:44 Join XbooX [0] (n=xboox04@
22:55:53preglowsaratoga: different window sizes is no problem for mdct as long as you overlap them nicely
22:56:15saratogayeah, but you'll need a window for each possible window transistion right?
22:56:31saratogai guess they have some formula to convert between them
22:56:35 Join hannesd [0] (
22:56:47preglowwell, kind of, but the new windows can still be constructed from existing ones
22:56:53preglowso it's not like you need to generate new coefs or anything
22:56:53saratogathe code says something about resampling them, but i don't know how that wokrs
22:57:20saratogamaybe it is just as easy as zero padding/stuff to step between sizes
22:57:22preglowi don't really know, all my playing about with mdct has been done with fixed-size blocks
22:57:49saratogabeen meaning to ask: is there some way to get gcc to spit the output from the preprocessor out?
22:57:56saratogaas in preprocessed .c files
23:00:18 Quit Lear ("ChatZilla [Firefox]")
23:00:33linuxstbsaratoga: Is some of that zip file you sent me from an old ffmpeg?
23:01:41 Join TrueJournals [0] (
23:02:05jimh_jimhguys - I've been working on enhancing the mpegplayer plugin - is anyone here familiar with this code?
23:02:35saratogalinuxstb: yes, the asf stuff is parly from it
23:03:00saratogasince the newer stuff mainly added more abstractions I didn't want or need
23:03:06linuxstbsaratoga: OK, that explains why the cook decoder doesn't work...
23:03:24saratogathe codec interfaces haven't changed though
23:03:37saratogathey still only expose themselves through a function pointer table
23:03:48 Join p3tur [0] (n=petur@rockbox/developer/petur)
23:03:49linuxstbI'm sure it's fixable. Just wanted to make sure I was understanding what I was looking at.
23:04:03saratogathough the function names in futils and utils.c will be different, moved around and changed some
23:04:17linuxstbYes, that's what I was hoping you had reconciled...
23:04:24saratogaactually, i never got the current SVN ffmpeg to work
23:04:42saratogaso i took the WMA code from it and went with that
23:05:08saratogathey changed something in utils.c that broke all the example code i could find, eventually i gave up in fustration
23:05:55linuxstbI can understand that...
23:06:03linuxstbjimh_jimh: What's your question?
23:06:23saratogaprobably the code you have will have a mountain of crap in utils and cook.c related to "probing" which is how they figure out what a container actually contains
23:06:43saratogaif you strip that out and just hard code Cook to .RA, then most of the changes since then don't really matter
23:07:01saratogathe codec and container init stuff is relateively simple and not changed very much
23:07:48jimh_jimhlinuxstb: I'm trying to get ff/rew to work - I've managed to get the sound thread to jump around to the new playing point, but the video won't jump ahead/back in the buffer - any suggestions?
23:07:56linuxstbsaratoga: OK, but I think I need to leave it now.
23:08:15 Part Llorea1
23:08:41linuxstbjimh_jimh: I know jhmikes has done a lot of work on the buffering code in mpegplayer (and was also planning to implement seeking at some point). He's probably the best person to talk to.
23:09:51 Quit davina (Remote closed the connection)
23:10:00jimh_jimhlinuxstb: thanks - I've also written some code that allows for resuming - I'll try to get in touch with him in case he wants to incorporate it with his work
23:12:41 Quit pepie34 ("Ex-Chat")
23:12:51TrueJournalsjimh_jimh: You should post your code on flyspray... maybe some people could help there
23:13:33 Join Llorean [0] (
23:13:52 Nick p3tur is now known as petur (n=petur@rockbox/developer/petur)
23:16:41saratogaugh my Sansa is in a building less then 55 minutes from my door, yet the post office is predicting another week to get it to me
23:16:49saratogai don't understand how this is possible
23:18:40 Join ducbian [0] (
23:19:25 Join ivan` [0] (i=ivan@unaffiliated/ivan/x-000001)
23:19:49ivan`hi - is there any easy way to take something like and create an 'flatter' equalization for the DT990 on rockbox?
23:19:56ivan`without manually playing with Q values and listening all day
23:22:17 Join Wiwie [0] (
23:22:18 Quit Wiwie (Read error: 54 (Connection reset by peer))
23:22:30 Join Wiwie [0] (
23:23:03 Part TrueJournals
23:23:55***Saving seen data "./dancer.seen"
23:24:37 Quit Wiwie (Client Quit)
23:26:01 Join ShedAdatiCheapSl [0] (
23:26:47 Nick ShedAdatiCheapSl is now known as use1ess (
23:26:55 Quit Thundercloud (Read error: 54 (Connection reset by peer))
23:27:55 Join Thundercloud [0] (n=thunderc@
23:29:40 Quit jimh_jimh ("CGI:IRC (EOF)")
23:32:34linuxstbLlorean: Around?
23:36:44Lloreanlinuxstb: Yes
23:37:00 Join jhulst [0] (
23:37:15 Join alexsaway [0] (
23:37:39alexsawayanyone here?
23:39:08linuxstbLlorean: I was trying to modify a forum post containing a DOS directory listing so it was readable, but got some odd behaviour I thought you might understand. The <DIR> part of the message was being interpreted as markup - but only in the real post, not the preview. i.e. the preview was different to the final post...
23:39:11stripwaxivan` - I can't think of any easy way..
23:39:27ivan`that's ok, i mostly got it with the peaks
23:39:42 Quit thegeek (Read error: 113 (No route to host))
23:39:51ivan`my ears are way too sensitive to highs
23:39:57Lloreanlinuxstb: Yeah, Anything wrapped in a <> is considered markup, but for some reason not in the preview, and I haven't really gotten to look into it.
23:40:08stripwaxI sortof feel there should be a way to plot the frequency response of the sw filter though .. preglow?
23:40:10Lloreanlinuxstb: The easiest solution, as a workaround, is to add spaces < DIR >
23:40:24linuxstbLlorean: That's what I did in the end.
23:40:51alexsawayuhm i think im abit screwed.. i cant boot in to normal boot , i cant get it connected to pc becuse rockbox doesnt support that yet on sansa
23:40:53alexsawayany ideas?
23:41:12Lloreanlinuxstb: TBH, I'm kinda hoping that it'll just be 'fixed' with an update.
23:41:41Lloreanalexsaway: Why can't you boot into the official firmware? What does it say when you try?
23:41:46 Quit atsea- (Remote closed the connection)
23:42:02alexsawaywell , i installed that thing that auto boots in to orginal
23:42:31alexsawaynow suddenly it just boots into rockbox
23:42:36alexsawaycant press < button either
23:42:52linuxstbLlorean: I think the problem is that the forum software isn't quoting <> - so "<DIR>" is being sent to the browser as-is, and is processed as a tag.
23:42:56Bagderalexsaway: you can use recovery mode to restore the original
23:43:24alexsawaythanks <33
23:43:45Lloreanlinuxstb: It just seems odd to me that the preview mode handles it differently than the actual post mode.
23:45:14linuxstbLlorean: So people can use HTML in messages?
23:46:22 Join scorche` [0] (i=scorche@rockbox/administrator/scorche)
23:46:39 Join Thundercloud_ [0] (n=thunderc@
23:47:42 Join donutman25 [0] (n=chatzill@
23:47:42Lloreanlinuxstb: I mean, though, the whole point of preview is to show you what the message will look like when posted, right? So any time the preview and message don't look the same, it's a bug?
23:48:39linuxstbLlorean: Sorry, my question wasn't an answer to your message... I agree preview should do what the name says.
23:49:15linuxstbBut my question was whether people can use html in messages - I at least seem to be able to, without the [html] markup.
23:49:54Buschelstripwax: you there?
23:50:14Lloreanlinuxstb: People aren't supposed to be able to, the option is unchecked.
23:50:20stripwaxlinuxstb - what is the status of the MPA patch in ? Other than the profiling, (and the potential small diffs due to the new dct), is it committable?
23:50:21stripwaxBuschel - ay
23:51:14linuxstbLlorean: Am I (I'm not sure what group I'm in) supposed to be able to?
23:51:20Buschelstripwax: just saw your post regarding the battery type patch. is there any way to differ 60/80GB iPODs from 30GB ones?
23:51:27Lloreanlinuxstb: As far as I know, it's not group specific
23:51:37Buschelstripwax: at compiletime?
23:51:39LloreanIt's supposed to be disabled across the board.
23:51:42*Llorean checks his dummy account
23:52:02stripwaxBuschel - when you configure, you get the option of specifying memory size. 60GB and 80GB have 64MB. Could use that?
23:52:15stripwax(i.e. ../tools/configure asks for ipod video memory size)
23:52:42amiconnSomeone should really check how much RAM the G5.5 30GB has
23:52:49stripwaxwas just about to say same
23:52:56stripwaxbattery too.
23:53:17Lloreanlinuxstb: It seems any account can do it.
23:53:19amiconnShould be simple - just install a 64MB build. If it runs, it has 64MB, if it crashes, it has only 32
23:53:40 Join atsea- [0] (i=ariel@gateway/tor/x-f928e5e1df681f0c)
23:53:43Buschelbatteries are 400mAh for 30GB and 600mAh for 60/80GB (Li-Ion)
23:53:54stripwaxBuschel - 5.5G too?
23:54:11Buschelyep, that's what apple sites sais
23:54:31 Quit Thundercloud (Read error: 60 (Operation timed out))
23:55:28 Quit alexsaway ("( :: NoNameScript 4.02 :: )")
23:56:10saratogastripwax: been wondering about that patch too
23:56:16saratogaseems like a good thing to have
23:58:25*amiconn would like to see mpegplayer support sample rates != 44100 Hz

Previous day | Next day