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

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

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

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

#rockbox log for 2008-04-04

00:00:07 Join daurnimator [0] (i=daurn@unaffiliated/daurnimator)
00:00:07mcuelenaere(it has to do with the ZVM port)
00:00:24mcuelenaerebecause Rockbox only understands FAT, not one of the Creative's proprietary file systems
00:01:08pixelmasomeone with an Ipod Video around who tested the WPS changes the last days (asking especially for the included UniCatcher and iCatcher). There is half of a report that it's "broken" though I'm not sure if he uses the included ones. I can't imagine that they are broken because it's been tested all in sims but who knows...
00:01:16 Quit thegeek (Read error: 110 (Connection timed out))
00:01:53 Join CyBergRind|w [0] (n=cbr@
00:02:10Nico_Ppixelma: UniCatcher works fine here
00:02:14 Join bughunter2 [0] (
00:02:54Nico_PiCatcher too
00:03:16pixelmathanks for confirming
00:03:40Nico_Pnicer icons would be good for themes other than cabbie...
00:04:32scorche|shbluebrother: what DerPapst said...hopefully i can do it this weekend =/
00:04:58bluebrotherscorche|sh: ok.
00:06:12pixelmaNico_P: who defines "nicer"? I also don't want to really change something in the looks now, those are not my creation...
00:07:01Nico_Ppixelma: I dare make the assumption that slightly bigger colour icons can be considered "nicer" :)
00:07:42Nico_PI was simply thinking of using the same ones as cabbie, although they probably would need adapting because of the antialiasing
00:07:52pixelmathat's a question of taste
00:08:37Nico_Pit is indeed
00:08:44pixelmaah... do you mean the icons in the menu?
00:09:16pixelmathose need thorough rework on greyscale though :\
00:09:28disorganizergood night to all
00:09:39 Quit disorganizer ("ChatZilla 0.9.81 [Firefox]")
00:10:42 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
00:11:29 Join stripwax [0] (
00:11:35 Quit Nevtus (Read error: 110 (Connection timed out))
00:18:30n17ikharen't there some better WPSs than the ones on rockbox-themes?
00:18:54n17ikhthe ones for the e2xx series are either too busy or just awful :/
00:19:00 Quit [CBR]Unspoken|w (Connection timed out)
00:19:48scorche|shn17ikh: check the wiki?
00:21:08n17ikhyeah, those are better
00:22:06n17ikhI don't really like them though... I would hate to have to learn the wps language to make my own
00:22:27bluebrotherseems you need to.
00:22:41gevaertsn17ikh: you could pick one where you like the basic design and change the background
00:22:49 Join hd [0] (i=jd@unaffiliated/helldragon)
00:22:49 Quit hd (Client Quit)
00:23:59*gevaerts has solved his wps problem that way
00:24:16n17ikhsome of these are... really ugly :/
00:24:46n17ikhcourse, from what I've seen the professional designers making the factory interface don't do any better
00:25:17krazykitn17ikh, if you want ugly, look at the OF on the gigabeat F :(
00:25:28gevaertskrazykit: that bad ?
00:25:49bluebrotherhmm. Seems domonoky has broken my local changes. Too bad ...
00:25:57krazykiti thought it was an eyesore for the few minutes i looked at it before putting on rockbox.
00:26:11*gevaerts has a gigabeat f, but installed rockbox before even looking at the of
00:26:15 Quit XavierGr ()
00:26:40ChristopherWn17ikh: check out the "2001" theme :)
00:26:43krazykitwell, the power was out when i got it ;)
00:28:57Bagderproof of concept rotated table =>
00:29:39bluebrotherBagder: kinda nice.
00:29:56Bagderstill requires scrolling, although vertical
00:29:59 Quit stripwax (Read error: 110 (Connection timed out))
00:30:11bluebrothernow make it colored and use alternating column backgrounds −− I think it would look quite good
00:30:14 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
00:30:26bluebrotherbtw, have you noticed my idea about the wiki stylesheet?
00:30:40 Quit lee-qid (Connection timed out)
00:30:44BagderI did yes, but that too is zagor's backyard more than mine
00:30:53bluebrotherah, ok :)
00:32:03 Nick fxb is now known as fxb__ (
00:32:26ChristopherWBagder: vertical scrolling is still better than horizontal scrolling
00:33:36 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
00:36:28*linuxstb has widescreen monitors, so prefers horizontal scrolling...
00:37:25*petur rather liked the idea of only showing targets with problems
00:37:43BagderI'll do some experients with that too
00:37:49linuxstbHow could that work?
00:38:02bluebrothersounds interesting, but it should be toggleable
00:38:09*bluebrother too slow :)
00:38:16 Join hd [0] (i=jd@unaffiliated/helldragon)
00:38:19Bagderwith a checkbox or something to display the zero-columns
00:38:58DerPapstthe table still fits om my screen :-P
00:39:20bluebrotherthe rotated one?
00:39:55*petur disables ata_lock (mutexes) in the ata driver and finds his bootloader booting again - still a crash from time to time :/
00:43:15peturgrrrr every time it crashes it is within threading code :(
00:43:53*roolku waits for the red...
00:45:21 Quit waldo ("Konversation terminated!")
00:46:22ChristopherWthreads are fun
00:46:41ChristopherWs/fun/really annoying to debug/
00:47:04pixelmaroolku: you did not fix the button map issues I pointed out
00:47:26roolkupixelma: sorry which ones?
00:48:12roolkuI overlooked it because you used an action
00:48:37roolkum5x5, right?
00:48:38 Quit JdGordon|zzz (Read error: 110 (Connection timed out))
00:49:21pixelmathe Ondio map has "menu" for select and several combos with menu without checking for pre conditions and releases and on the M5X5 pad only combos with power work (electrically) but it's not usable
00:49:48pixelma(because a "longer power" is also a hard power off)
00:49:53 Join Shaid [0] (
00:51:10pixelmaso basically you can't use button combos and have to try with long versus short button presses instead
00:54:06 Quit DerDome (Read error: 110 (Connection timed out))
00:55:34 Join Thundercloud [0] (
00:55:36 Quit linuxstb (Remote closed the connection)
00:55:53 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
00:56:31 Quit ColinT (Remote closed the connection)
00:56:44 Join ColinT [0] (
01:00:04 Quit mcuelenaere (Read error: 110 (Connection timed out))
01:00:39 Quit ender` (" How many Dragon Ball Z characters does it take to screw in a lightbulb? Only one, but it takes ten episodes.")
01:03:11kugellinuxstb: I think I have a version without reparsing at every redraw
01:04:51 Quit shotofadds (Read error: 104 (Connection reset by peer))
01:08:42 Quit n1s ()
01:11:18 Join Thundercloud_ [0] (
01:12:26 Quit robin0800 (" HydraIRC -> <- Like it? Visit #hydrairc on EFNet")
01:12:36***Saving seen data "./dancer.seen"
01:12:58kugellinuxstb: It's far from working well though
01:14:26linuxstbkugel: Did you also move the code that calculates the viewport sizes/positions?
01:14:49 Quit petur ("Zzzz")
01:18:06kugellinuxstb: Not sure what you mean
01:18:21kugellinuxstb: That's in viewport.c, since v8
01:19:02kugelin the geneic viewport structure parsing function
01:20:01 Join dandin1 [0] (
01:20:22kugellinuxstb: did you mean that?
01:21:20linuxstbkugel: I mean the code in (e.g.) draw_title() that calculates the viewports (e.g. vp_icons and vp_text)
01:21:27linuxstb(plus similar in list_draw)
01:22:25kugelAh, that's what I meant with "not parsing at every redraw"
01:23:17 Quit ompaul (Client Quit)
01:24:46kugelThere's only "struct viewport *list_infos = list->list_info[display->screen_type];" now
01:25:15kugellinuxstb: My current approach is to calculate the viewport at list initilializing
01:25:39kugeland that gui_synclist contains that list_info vp
01:25:47linuxstbShouldn't that be ... = &list->list_info[display->screen_type];
01:26:08kugelI thought that too, but that gave warnings
01:26:33kugelincopatible ...
01:27:29 Quit Thundercloud (Read error: 110 (Connection timed out))
01:29:52 Quit bluebrother ("leaving")
01:30:54 Quit Jon-Kha (Remote closed the connection)
01:34:02linuxstbkugel: How is list_info declared?
01:34:11*preglow sees another *box plugin title :/
01:35:24 Join JdGordon|zzz [0] (
01:35:25pixelmapreglow: the topic came up before and gevaerts suggested rockpegs earlier ;)
01:35:55linuxstbNot helpful ;)
01:35:57*gevaerts runs away again
01:36:05*kugel thinks that pegbox and pacbox must lead to confusion
01:36:22*linuxstb apologies for the pacbox name
01:36:31*linuxstb apologises for the typo
01:36:53*pixelma thinks kugel didn't tested the latest patch (the one in the tracker) on his e200 but said "there were no bugs"...
01:37:25linuxstbpixelma: You're talking about pegbox?
01:38:12kugelpixelma: Which one?
01:38:29kugelpegbox? I said I never experienced an issue
01:38:35kugeland that's true
01:38:39pixelmathe one that was committed still had scroll up/down in it
01:38:54pixelmaprobably run your own build...
01:39:08scorche|shi thought we werent going to do pegbox...
01:39:12kugelI think it had it in once
01:39:52pixelmayes, and it was committed this way (probably because it was the last on in the tracker)
01:40:19roolkudon't panic it can still be fixed. :)
01:40:35linuxstbAnd renamed? ;)
01:40:41 Quit knittl (Read error: 110 (Connection timed out))
01:40:55*roolku couldn't care less about names
01:41:03kugelpixelma: I tested it heavily, and there was someone who tested it on ipod, and we didn't find an error
01:41:19kugelI just wonder why you come up with that old commit just today
01:41:40pixelmawhat are you talking about?
01:41:42linuxstb"old commit" ?
01:41:59pixelmaI'm talking about pegbox
01:42:11pixelmaI don't talk to your rename patch
01:43:21kugelWell, I don't quite get what you mean
01:43:23pixelmawhich was a good change, but the pegbox patch wasn't updated accordingly. That was my point
01:43:37linuxstbkugel: pegbox (as committed to SVN) doesn't compile...
01:43:46linuxstb(for the e200)
01:43:58kugelIt's not my patch dudes.
01:44:14kugelI'm not the maintainer, it's possible that the tracker version is out of sync
01:44:45pixelmabut you said there were no bugs earlier... that's all
01:44:58kugelIndeed, I have a private copy, which fixes that (but which is also more sync'd to my build)
01:45:00*DerPapst compiles to see what pegbox actually is
01:45:15gevaertsDerPapst: it's something like rockpegs :)
01:45:25 Join mud-rb [0] (
01:45:49kugelpixelma: Well, I said that there's no issue in general. That a patch can get out-of-sync is allways possible
01:46:05DerPapstgevaerts: ahhh now i know :-)
01:46:27 Join Thundercloud [0] (
01:47:12 Join Sader [0] (n=4a809077@gateway/web/cgi-irc/
01:47:37Saderhi :)
01:48:25SaderI've been using Rockbox on my ipod nano 1st gen, and i accedientally made it record .. i can't get it to stop. It won't let me listen to music and rebooting it doesn't help :S
01:48:33kugelAnyway, sorry if it was my fault, that one if you committed a not-sync'd version
01:49:56DerPapstsader: doesn't short pres leave the rec screen?
01:50:07linuxstbSader: What happens when you reboot? (and how are you rebooting?)
01:50:16pixelmaI didn't say it was your fault, just a little reminder (with a wink) to be a bit accurate ;)
01:50:28pixelma^ kugel
01:51:22kugelI guess the one who'll correct that will be happy for the commit+1 anyway ;)
01:51:38 Join `Sader` [0] (
01:51:39*gevaerts gets to work ;)
01:52:00mud-rbby any chance is there a script somewhere that will build a .zip for each target?
01:52:26`Sader`Hey I've been using rockbox on my first gen nano, and its good, except i accedientally click record.. and i can't get it to stop when i reboot it doesn't do anything either.(Can't listen to music while its recording)
01:52:37`Sader`Any tips would be greatly appreciated.
01:52:58 Quit `Sader` (Client Quit)
01:53:19linuxstbSader: Still here?
01:53:45 Part toffe82
01:53:51 Join `Sader` [0] (
01:54:29`Sader` i accedientally click record.. and i can't get it to stop when i reboot it doesn't do anything either.(Can't listen to music while its recording)
01:54:52`Sader`(soz lag'd didn't mean to spam)
01:55:38Nico_Ppegbox is nice!
01:55:52linuxstb`Sader`: No, you can't listen to music whilst recording - that's expected.
01:56:26`Sader`I mean, i can't get it to stop recording
01:57:02linuxstbNot even by rebooting?
01:57:03mud-rb`Sader`: did you look in the manual for the correct way to exit the recording screen?
01:57:46`Sader`Yes - I looked in the Manual but don't anything about it..
01:57:47DerPapstiirc it's short menu for iPods
01:58:06DerPapstat least it's on mine
01:58:14pixelma`Sader`: what happens after a reboot?
01:58:30`Sader`It boots w/ apple screen, then turns on and its on the recording screen :S
01:58:47pixelmaah, maybe you set your start screen to recording?
01:59:08`Sader`If i did, it's by acciedent.(sp) anyway to un-do it?
01:59:27DerPapstclear your config or edit it on your PC
01:59:48pixelmayou could clear your setting by flicking the hold switch to on when the rockbox bootlogo appears (I think it was on Ipods) after a reboot or...
02:00:34pixelma...if edit/delete the config.cfg (in the .rockbox directory) from your computer
02:00:36kugellinuxstb: If you wanna take a look,
02:01:00kugelBut as I said, it's not working very well, and I'm still unsure the approach is good
02:01:18`Sader`Alright, I'm trying it now.
02:01:46mud-rbshouldn't it be possible to exit the recording screen even if it's set as the startup?
02:02:01 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
02:02:03pixelmait should
02:02:16mud-rbokay, just making sure
02:03:09pixelmaas already been said, it should be a "short press of menu" or IIRC "long play" works too
02:03:10`Sader`Pixelma, It worked :D
02:03:21`Sader`Thanks ALOT!
02:03:34pixelmayou're welcome :)
02:04:18`Sader`While i'm here, is there anywhere to get more games/demos like the ones that come defaultly?(I already have iBoy)
02:04:22pixelmamud-rb: but I _believe_ that there is a bug on Ipods that playback doesn't work after it was in recording mode or so
02:04:52linuxstbpixelma: Yes, not all ipods though (but the Nano is one of the ones affected)
02:05:04mud-rb`Sader`: all of the suppored ones are installed by default. you can look in flyspray for game plugins ('patches' on the left of the rockbox site)
02:05:17mud-rb`Sader`: but those will require you to build rockbox from source
02:05:34 Quit Thundercloud_ (Read error: 110 (Connection timed out))
02:05:43`Sader`Thanks for all of the help guys!
02:05:46 Quit `Sader` ()
02:06:10 Join corevette [0] (
02:06:41 Quit Sader ("CGI:IRC (EOF)")
02:07:39 Quit Mathiasdm ("Yuuw!")
02:08:36 Quit Nico_P (Remote closed the connection)
02:09:40 Quit gevaerts ("sleeping time")
02:15:17 Join thegeek [0] (
02:20:01 Join Hilton [0] (
02:20:31 Quit kugel ("Leaving.")
02:34:41 Join Thundercloud_ [0] (
02:34:49 Quit Hilton ("ChatZilla 0.9.81 [Firefox]")
02:35:58 Join toffe82 [0] (n=chatzill@
02:36:13 Join ryran [0] (
02:38:45 Quit amiconn (Nick collision from services.)
02:38:51 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
02:43:31 Quit jhMikeS (Nick collision from services.)
02:43:37 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
02:44:00pixelmahmm... not very patient
02:45:20DerPapstheh.. i guess sie just wanted to know if the channel exists
02:51:21 Quit Thundercloud (Read error: 113 (No route to host))
02:51:30roolkupixelma: I'll look at the button combinations tomorrow - now sleeptime
02:52:41pixelmanice, thanks :)
02:53:41 Quit ol_schoola ()
02:54:25 Quit ali_as ("zzzzzzzzzzzzzzzzzzzzzzzzzZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZz")
02:54:39 Join webguest74 [0] (n=411adf9d@gateway/web/cgi-irc/
02:54:41 Quit roolku ()
02:55:02 Quit webguest74 (Client Quit)
02:55:09 Join ol_schoola [0] (
02:56:39 Quit ol_schoola (Client Quit)
02:58:14 Quit toffe82 (Remote closed the connection)
03:04:30 Join toffe82 [0] (n=chatzill@
03:04:48 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
03:06:14 Join Thundercloud [0] (
03:06:45 Join cool_walking_ [0] (
03:12:38***Saving seen data "./dancer.seen"
03:12:42 Join mossimo2k3 [0] (n=18eda03b@gateway/web/cgi-irc/
03:13:14mossimo2k3how do i load patches?
03:19:01 Join Tordre [0] (
03:21:01 Quit Tordre (Client Quit)
03:22:27 Nick qwm is now known as qm (
03:22:29 Join DavidSG [0] (
03:22:30mud-rbmossimo2k3: you need to check out the rockbox source from subversion, apply the patches, and make a build from source
03:22:38mud-rbmossimo2k3: look in the wiki, there's a howto
03:23:16 Quit Thundercloud_ (Read error: 110 (Connection timed out))
03:24:05DavidSGHello I am David Gilbert I applied for the Summer of Code project to create a WPS and theme Editor.
03:24:06ChristopherWlol, there's two people so far that say "hello" and leave :)
03:24:42DavidSGsorry that first one was me, i thought it would be better to use my real name.
03:24:46mud-rbChristopherW: yeah, it seems pretty common...apparently people can't look to the right and realize that there's people here or something
03:25:23ChristopherWI guess if they really need help they'd say something more than just "hello"
03:28:13DavidSGHas anyone here read my proposal, any feedback would be appreciated.
03:28:49 Join linuxstb_ [0] (
03:32:07 Join crzyboyster [0] (n=6210f06b@gateway/web/cgi-irc/
03:32:49crzyboysterDo viewport conditionals work right now? I thought I read something in the forums about it, but I may have misunderstood...
03:33:00linuxstb_crzyboyster: No.
03:33:34 Quit jhMikeS (Nick collision from services.)
03:33:38crzyboysterCan someone explain what this means >
03:33:40 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
03:33:44*pixelma missed the part about "conditionals" and decides that this indicates the need for sleep
03:33:50linuxstb_DavidSG: You've come at the wrong time of day (night) - most of the possible Rockbox mentors are European, and it's the middle of the night here (yes, I should be sleeping...)
03:34:25pixelmalinuxstb: wanted to say something along these lines too, but then I saw you around...
03:34:40 Quit crzyboyster (Client Quit)
03:35:52 Part pixelma
03:42:55DavidSGlinuxstb_: thank you will try back tomorrow
03:43:47 Join mud-rb_ [0] (
03:45:49 Quit mossimo2k3 ("CGI:IRC (Ping timeout)")
03:46:41 Quit argumentD (Read error: 110 (Connection timed out))
03:47:45 Quit Thundercloud (Remote closed the connection)
03:48:50 Join Thundercloud [0] (
03:57:50 Part ryran ("Leaving")
04:00:01 Quit DerPapst ("So Long And Thanks For All The Fish!")
04:00:51 Join MobileShaid [0] (
04:03:48 Part linuxstb_ ("Leaving")
04:04:10 Quit mud-rb (Connection timed out)
04:05:25 Join FOAD_ [0] (
04:18:37 Quit Shaid (Read error: 110 (Connection timed out))
04:18:38 Nick MobileShaid is now known as Shaid (
04:21:28 Join miepchen^schlaf_ [0] (
04:22:07 Quit FOAD (Read error: 110 (Connection timed out))
04:22:07 Nick FOAD_ is now known as FOAD (
04:23:23JdGordonhas someone left part of rockbox usb in the svn builds?
04:23:44JdGordonpluggin my sansa in gmesg says a 2gb usb storage device was connected!
04:24:57JdGordonactually... maybe not..
04:25:02JdGordonstupid lack of timestamps
04:31:54 Quit corevette (Remote closed the connection)
04:36:40 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:37:40 Join wyoak [0] (n=wyoak@
04:38:44wyoaki'm out of ideas...on both of my sansa players (c250 and e250), i can't get the DB to initialize, even on a blank player
04:39:21linuxstbI'm not sure if the database works with a blank player... (although that bug may have been fixed)
04:39:32wyoaki'm trying the e250, and it counts up all the files on the player (around 1700 in total, with themes and stuff), and never moves. the .rockbox directory has the database.ignore on it
04:39:47wyoaklet me put a few songs on there.....
04:41:03n17ikhhow does rockbox calculate the battery life?
04:41:18n17ikhI just got done charging the battery in the OF on my e260 and it says 60% in RB
04:43:07wyoakyeah, even with songs on there it still seems like it's trying to add the .rockbox directory to the database
04:43:23wyoakstuck on 377 found (didn't add themes this time)
04:43:44 Join corevette [0] (
04:44:50n17ikhaccording to a very old mailing list post it's based on a moving average and is incorrect immediately after boot... is that still correct?
04:47:59A-4n17ikh: battery life is based on the voltage the battery currently has. 4.5 V = 100%; 22.5 volt = 0%
04:48:34wyoakthe stock firmware builds its database fine
04:48:47wyoaki wonder if my computer is doing something wierd, since it happens on both players
04:52:05A-4n17ikh: the battery % just after boot is supposed to be a bit low because booting consumes quite a bit of enery which causes the voltage to drop temporarily, i belive right after boot 5% is added to the battery life to make up for this voltage drop
04:52:14n17ikhbleh... that's a pretty terrible way to calculate power remaining in a li-ion
04:52:52n17ikhseems silly to me to build hardware using modern batteries that doesn't keep track of mWh
04:56:20A-4wyoak: you could put a few files in some subdirs of .rockbox and check if they are also appear in the database
04:56:47wyoakthe database never finishes initializing
04:57:00wyoakit just counts all the files on the player and stops there
04:57:27linuxstbwyoak: Do you restart your player when that happens?
04:57:43wyoaki've left it on for a while before, tried restarting it too
04:57:47wyoaknever can get past that
04:57:54wyoakthe files view works fine though
04:59:06 Join ol_schoola [0] (
05:02:03 Quit joris (Remote closed the connection)
05:08:27 Quit Horscht (Read error: 104 (Connection reset by peer))
05:12:26 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
05:12:43***Saving seen data "./dancer.seen"
05:13:43 Quit Thundercloud (Remote closed the connection)
05:17:50 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
05:18:40 Join piga [0] (
05:26:20 Join freelance [0] (n=fr33l4nc@
05:26:34 Quit dandin1 ()
05:28:05 Quit hd (Client Quit)
05:28:34 Join midgey [0] (
05:30:06*midgey notices his pegbox was committed...
05:30:35 Join hd [0] (i=jd@unaffiliated/helldragon)
05:31:59midgeyi had no idea anyone was interested in that, i would have updated it or at least cleaned it up
05:32:07midgeyand renamed it to pegs
05:41:53 Quit Horscht ("IRC is just multiplayer notepad")
05:43:58 Join ctaylorr [0] (
05:44:48mud-rb_midgey: haha, congrats on getting it submitted anyway :) looks pretty cool
05:56:54 Quit wisp_sh (Read error: 110 (Connection timed out))
05:58:47midgeymud-rb_: most of my code seems to be committed by someone else :P
06:04:50midgeyif anyone has anything to say about the code leave it in the logs and i'll get to it when i have a chance
06:09:30*midgey sleeps
06:09:32 Quit midgey ()
06:11:12 Quit piga ("Leaving")
06:11:58 Quit ctaylorr (Read error: 110 (Connection timed out))
06:26:04 Quit freelance ("Leaving")
06:30:33 Quit ol_schoola ()
06:32:02 Join ol_schoola [0] (
06:33:10 Quit ol_schoola (Client Quit)
06:38:23 Quit wyoak ("quit")
06:46:16 Join atm [0] (n=atmb4u@
06:48:19 Join courtc_ [0] (
06:48:24 Part atm
06:48:58 Quit courtc (Read error: 113 (No route to host))
06:55:43 Quit mud-rb_ (Remote closed the connection)
07:03:00 Join lee-qid [0] (
07:07:04 Join wisp_sh [0] (
07:12:45***Saving seen data "./dancer.seen"
07:13:06 Quit Loplin (Read error: 104 (Connection reset by peer))
07:23:11 Quit bughunter2 (Read error: 104 (Connection reset by peer))
07:23:57 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:27:01 Quit Zarggg ()
07:37:13 Join argumentD [0] (
07:37:48 Join at1 [0] (n=atmb4u@
07:38:30 Quit lee-qid (Read error: 110 (Connection timed out))
07:40:37 Part at1
07:42:25 Part toffe82
07:46:38 Quit TMM (Remote closed the connection)
07:53:28 Join knittl [0] (n=knittl@
08:09:28 Join kugel [0] (n=54bd72d6@gateway/web/cgi-irc/
08:10:58 Quit BigBambi (Remote closed the connection)
08:10:58 Quit miepchen^schlaf_ ()
08:11:33 Quit kugel (Client Quit)
08:12:04 Join ol_schoola [0] (
08:19:38 Join Rob2223 [0] (
08:20:59 Join lenny [0] (n=4cd1df26@gateway/web/cgi-irc/
08:32:29 Join spiorf [0] (
08:37:25 Join atm [0] (n=atmb4u@
08:37:52 Quit Rob2222 (Read error: 110 (Connection timed out))
08:38:46 Quit Chipsaru (Read error: 110 (Connection timed out))
08:48:38 Quit XavierGr (Nick collision from services.)
08:48:49 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
08:50:51 Quit Shaid (Read error: 110 (Connection timed out))
08:52:01 Quit lenny ("CGI:IRC (Ping timeout)")
08:54:35 Join ender` [0] (
08:56:47 Join at1 [0] (n=atmb4u@
08:56:54 Part at1
09:05:15 Quit atm (Read error: 110 (Connection timed out))
09:11:53 Nick JdGordon is now known as JdGordon|afk (n=jonno@rockbox/developer/JdGordon)
09:12:47***Saving seen data "./dancer.seen"
09:16:51 Join Jon-Kha [0] (
09:18:28 Quit csc` ("If you can't do anything smart, do somethin right ~Serenity")
09:23:05 Join petur [50] (n=petur@rockbox/developer/petur)
09:25:02 Join davina [0] (
09:30:48 Quit argumentD (Read error: 113 (No route to host))
09:31:19 Quit wisp_sh (Read error: 110 (Connection timed out))
09:33:32 Join atm [0] (n=atmb4u@
09:36:02 Join wisp_sh [0] (
09:46:30 Join ender [0] (
09:49:17ChristopherWI'm basically looking for feedback on my patch:
09:50:52ChristopherWactually it's more of a feature request with a working proof-of-concept patch
09:51:17 Join ender1 [0] (
09:52:06 Join qwedsa_ [0] (
09:52:45 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
09:55:12 Quit atm (Read error: 110 (Connection timed out))
09:58:10peturChristopherW: not that I iknow the GUI code or wps parser, but is that "We have consumed 1 byte" return still correct?
09:58:52ChristopherWI believe it is, because the return value indicates how many bytes are read from the wps file, and that doesn't change
09:59:04peturah, right
10:00:10 Quit Nico_P (Remote closed the connection)
10:04:40 Quit ender` (Read error: 110 (Connection timed out))
10:04:58 Quit ender (Read error: 110 (Connection timed out))
10:07:52 Join kritter` [0] (
10:08:43 Quit qwedsa (Read error: 110 (Connection timed out))
10:16:42 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
10:23:29 Join pixelma_ [50] (i=pixelma@rockbox/staff/pixelma)
10:24:40 Join gevaerts [0] (
10:26:02 Quit pixelma (Nick collision from services.)
10:26:02 Nick pixelma_ is now known as pixelma (i=pixelma@rockbox/staff/pixelma)
10:38:33 Join midkay_ [0] (
10:38:37 Quit midkay (Nick collision from services.)
10:38:39 Nick midkay_ is now known as midkay (
10:45:40Mouser_XCool, an update for FS #8647
10:45:59*Mouser_X would like to see this one commited sometime "soonish."
10:48:46*gevaerts would think that this instruction page should be finished first
10:49:38Mouser_XVery true. I did say "soonish" for a reason.
10:49:54gevaertsDoes it have all keymaps ;)
10:50:16Mouser_XNo idea, but it works rather well on the Gigabeat.
10:50:51Mouser_X(Though, the button to access the map needs to be changed. The version I have has it on the LONG down.)
10:51:31Mouser_X(I don't know what the new version has it at. I'll have to find someone to build it for me. I have no build environment right now..)
10:51:43*Mouser_X hopes to remedy this "soonish."
10:52:00Mouser_X(As in, within the next 3 months.)
10:52:12Mouser_X(Perhpaps less.)
10:54:11 Join FOAD_ [0] (
10:58:19amiconnThe amaze patch has some unrelated changes. Also, the drawing code can be optimised quite a bit
10:58:59amiconn(e.g. using lcd_vline() resp. lcd_hline() where appropriate instead of lcd_drawline())
10:59:33amiconnOn mono and greyscale targets, it could use the greylib
11:05:08 Quit cool_walking_ (Remote closed the connection)
11:06:10 Quit FOAD (Read error: 110 (Connection timed out))
11:06:10 Nick FOAD_ is now known as FOAD (
11:12:49***Saving seen data "./dancer.seen"
11:18:12linuxstbChristopherW: Regarding 8844 - it looks good to me (it was something I originally wanted to implement, but wasn't sure how to do it). You should try and point Nico_P to the patch when he's next around, as he knows the wps parser best.
11:18:15*gevaerts just talked to the real USB guy here, and has some new insights on the USB problem
11:19:38pixelmacool, maybe you should try to get him interested in Rockbox... :)
11:19:46gevaertsWhat _probably_ happens is that the controller can't access RAM fast enough, which causes it to send out the entire packet a bit too late
11:20:12gevaertsSo the bit timing within a packet is still correct
11:20:16linuxstbpixelma: Do you still think that this feature is useful?
11:21:04gevaertsIf this happens near the end of a microframe, the packet collides with the SOF sent by the host. This causes the weird EHCI errors in my dmesg
11:21:23 Quit joshin (Read error: 110 (Connection timed out))
11:21:47ChristopherWlinuxstb: ok, I'll try to let Nico_P know about it tomorrow (well, later today, as it's 2:21 am here)
11:21:56linuxstbgevaerts: I thought you had tested with the buffers in IRAM?
11:22:16pixelmalinuxstb: I'm not sure
11:22:31gevaertsDue to cable delays and hub latencies, the probability of this microframe overflow happening is much higher with long cables (front port vs back port) or with hubs in between
11:22:50gevaertslinuxstb: I can't get it to work at all with IRAM
11:23:23linuxstbgevaerts: That's odd... Have you looked at the USB code in e200tool? I thought that used IRAM.
11:23:32linuxstbor is that doing things very differently?
11:24:09gevaertslinuxstb: I'll have another go this weekend. We'll see
11:26:54 Join tvelocity [0] (
11:27:12gevaertslinuxstb: but anyway, for decent performance we need at least 16kb and preferably 32kb of buffer. That's a lot of IRAM...
11:29:05pixelmalinuxstb: for example, it took me a while to get the iCatcher.diff there. It's nice that the WPS code gets shorter (especially the codec one) but it also hides things
11:29:07*gevaerts likes understanding things better, even if this doesn't make any difference to what the probable solution would be
11:30:06 Quit ol_schoola ()
11:34:04ChristopherWpixelma: a nice thing about FS #8844 is that you can take it or leave it. It doesn't change most uses of sub-images inside conditionals
11:34:45ChristopherWer, I mean most *existing* uses, so current WPS files should work fine
11:35:05linuxstbgevaerts: That shouldn't be a problem though - USB should be able to claim the plugin/codec IRAM (at least 48KB)
11:35:10pixelmayes, I see that
11:36:11linuxstbChristopherW: IMO, the problem is that it creates two ways to do the same thing. i.e. adds complication for only a little gain. I'm not saying I'm completely against it, just that it's not _needed_.
11:37:21ChristopherWI think the same can be argued of sub-images themselves too
11:37:51pixelmabut sub-images are a big gain IMO ;)
11:38:01ChristopherWbut I believe my patch is just a logical extension of sub-images
11:38:02gevaertslinuxstb: for UMS, yes. But because for MTP rockbox retains full control of the disk it would be nice to allow playback to continue. For USB Audio I haven't worked it out yet...
11:38:08amiconnlinuxstb: That would need a lot of precautions...
11:38:23linuxstbChristopherW: No, sub-images have the real advantage of significantly speeding up the loading time of WPSs.
11:38:54amiconngevaerts: It might very well be that the usb controller can't access the iram at all
11:39:01pixelmaChristopherW: that and you can use more images than before (if there were few at the same spot
11:39:24gevaertsamiconn: entirely possible
11:39:30linuxstbamiconn: What's the issue? I thought USB stopped everything running?
11:40:29*gevaerts suggests to first try to find out if this discussion is actually needed (i.e. if IRAM is actually usable for this)
11:41:00ChristopherWpixelma: with more changes to the parser, it should even be possible to have more than 52 sub-images per image, but all of the sub-images can be referenced only through my patch or through another incompatible different method
11:41:25 Join Mathiasdm [0] (
11:41:49linuxstbChristopherW: How would you do that? And why?
11:41:56pixelmaI don't get that
11:42:11amiconnlinuxstb: Not at all. USB stops rockbox from accessing the disk. A plugin might enter the usb screen, but it might still have iram in use -> crash after exiting the usb screen without precaution (i.e. swapping out & in)
11:42:19 Join MartinR [0] (
11:42:40LinusNgevaerts: i wonder if it would be beneficial to avoid the uncached-address alias and instead perform explicit cache flushes?
11:42:45ChristopherWfor example, you could have 100 sub-images (say, for a vertical or even diagonal! progress bar)
11:43:02linuxstbamiconn: OK. I thought plugins were forced to quit.
11:43:32MartinRgevaerts: with Windows I've never seen transfers larger that 4kb im my traces
11:43:42linuxstbChristopherW: Wouldn't that quickly fill up the available image buffer?
11:43:44ChristopherWand you can use "%?px<%xdp>" to reference all 100 sub-images at once
11:43:55pixelmaChristopherW: ah, understand now. But do we really need it? People tend to do crazy things in WPS though...
11:44:15ChristopherWwell, I'm using extreme examples
11:44:33LloreanWell, non-extreme examples can be done using existing means...
11:44:50MartinRgevaerts: Speed was at 5.5mb/s while reading though.
11:45:07gevaertsMartinR: linux uses up to 64k I think
11:45:10amiconnLinusN: No. cache flushes block the processor, while uncached access only has a small speed penalty for typical buffer copying on PP
11:45:53pixelmaChristopherW: and with current means you could combine say 2 bitmap strips for said 100 piece progress bar... you have 52x52 images
11:45:56MartinRThat's a bit more than with OF.
11:45:58LinusNamiconn: yes, it blocks the cpu, but i was curious if it in some way would affect the dma in a positive way
11:46:14gevaertsLinusN: that won't help as long as the problem is caused by non-usb-related access
11:46:39ChristopherWpixelma: and then you still have to specify all 100 sub-images in the %?px<> conditional
11:46:42LinusNi wonder what kind of access that would be
11:47:12pixelmaChristopherW: but I'm rather undecided on the issue myself
11:47:35LinusNwhat if we changed the burst mode in the dram controller?
11:48:19LinusNhmmm, when i think of it, long dram bursts could severely hamper the dma
11:49:22gevaertsThe delays that case problems seem to be (if I don't miscount anything) on the order of one microsecond
11:49:25LinusNlike if the cpu occupies the dram with a long burst write when the usb dma needs to read
11:49:54LinusNone microsecond is a long time
11:50:14ChristopherWignoring theoretical advantages (such as possibly more sub-images), here is a summary of the pro's and con's of my sub-image patch: PROS: smaller WPS files (optional, but can beneficial to the WPS author), CONS: very slightly larger code size.
11:50:32gevaertsI'll try to get an accurate measurement this weekend
11:50:42MartinRgevaerts: Ok about Linux. But just in case we had to use a smaller buffer, we would still achieve a decent speed, I believe.
11:50:53LinusNChristopherW: another pro: would be simpler wps authoring
11:51:11 Join desowin [0] (n=desowin@atheme/developer/desowin)
11:51:19*gevaerts doesn't know anything about dram burst settings
11:51:23LloreanLinusN: I'm not entirely sure it would be much simpler. Maybe faster, but making things less explicit makes debugging a bit harder.
11:52:46LloreanI imagine, among other things, it'll irritate new authors getting automatic subimage use working with ones such as Play/Pause/FF/RW since you now have to be aware of the exact order of the conditional when designing your images, rather than just picking each image individually.
11:53:09LinusNwell, i always thought that the long lines with bitmap id:s in conditionals were prone to typos
11:53:28gevaertsMartinR: on high speed it probably doesn't make much difference, but on full speed this kind of buffer size difference can easily cause a 20% slowdown
11:53:53LloreanAlso, and I don't know if you can do this yet (but it's something that should be added, imho, if you can't) it would conflict with any case where you have a strip that's mixed use (for example, Volume and Battery icons are the same size, so you make a single file containing both and split them between two conditionals)
11:53:59pixelmaLinusN: did you take a look at the iCatcher.diff in there and tell me if you understood the battery level line (the %xdA one). Though I have to admit that's quite complicated code in general to I believe get a charging animation
11:54:09ChristopherWLlorean: you still have to be aware of the exact order even with separate images
11:54:55LloreanChristopherW: I didn't say you didn't. I just said you would then need to be aware when designing the images, rather than simply be aware when creating the .wps file (which you're more likely to be doing with the .wps syntax page open)
11:55:06ChristopherWpixelma: the animation is a little spinning thing when playback is paused, not the bettery
11:55:28MartinRgevaerts: Ahh, yes there's higher latency with full speed. Didn't think of that.
11:55:39pixelmaLlorean: you could do that (and address the sub-images in a different order but it's of course a bit confusing)
11:56:11pixelmaChristopherW: there are two animation lines and I meant the first which is battery level (and charger)
11:56:26gevaertsMartinR: and full speed is exactly where the difference hurst most
11:56:38LinusNhmm, perhaps an animation wps tag would be useful?
11:56:52ChristopherWpixelma: oh, you're right (I didn't look at it too closely) :-/
11:57:29gevaertsLinusN: can we actually control burst mode on PP ?
11:57:30ChristopherWLinusN: yes, an animation tag is another thing that would be helpful
11:57:31Lloreanpixelma: It just doesn't seem to address a "real" problem to me (the patch thati s)
11:57:53LinusNgevaerts: i think we can, we just don't know how... :-(
11:57:55LloreanWith better WPS creation tools, which we may have after the summer, the author doesn't need to see the actual code anyway.
11:58:30 Quit kritter` (Read error: 113 (No route to host))
11:58:50 Join kritter` [0] (
11:59:15kritter`what is a good but inexpensive rockbox compatible mp3 player under $100 ?
11:59:19pixelmaChristopherW: I looked at the code very closely the last days ;)
11:59:46ChristopherWpixelma: are you the one who's been committing many changes to iCatcher recently?
12:00:08pixelmayes, moved it over to using bitmap strips
12:00:15 Join DerDome [0] (n=DerDome@
12:00:25GodEaterkritter`: firstly, what sort of $ ?
12:00:39ChristopherWyes, I saw those changes in the changelog a few days ago
12:00:41kritter`GodEater, about $100 would be ok
12:00:48kritter`hopefully it has good storage too
12:00:54LloreanLinusN: I just think the patch may not be the right solution to the problem. Especially with the potential for GUI tools on the horizon.
12:01:33GodEaterkritter`: I mean which currency ? USD, AUD, NZD, something else ?
12:01:36desowindoes rbutilqt show /Volumes in browsedirtree on mac os?
12:01:49kritter`sorry GodEater 100 USD
12:02:02GodEaterkritter`: <−− take a look there
12:02:05LinusNLlorean: maybe not
12:02:11kritter`thank you
12:02:30*LinusN goes to lunch
12:02:39GodEaterkritter`: we get so many people from different countries here it's hard to tell what '$' means
12:02:53LloreanLinusN: Then again, I'm in favour of "being able to do more things with images" rather than "more ways to do the same thing with images" if we're going to be adding things. ;)
12:04:48ChristopherWmaybe I shouldn't take it too personally if my patch doesn't make it into the official builds ;-)
12:04:56*pixelma turns to "boxes" with bitmap strips again
12:07:08 Join Xerion [0] (
12:07:13linuxstbChristopherW: No, please don't ;) The patch looks good, it's just that we're unsure about the feature...
12:08:03LloreanChristopherW: We always like to see people trying to improve things. It just so happens that "improvement" is somewhat of a subjective idea.
12:08:49ChristopherWare you guys considering developing higher-level features in the wps that would make my patch irrelevant?
12:09:35LloreanChristopherW: There's the possibility of a graphical WPS creation tool, which makes the more significant benefit (easier to read .wps files) much less important since people will no longer be reading them in almost all cases.
12:11:06ChristopherWhmm, ok, but I still wonder what the downsides to my patch are even with the presence of graphical authoring tools
12:11:08pixelmabut that somehow reminds me of html creation tools which sometime lead to quite unreadable html code though...
12:11:43 Nick fxb__ is now known as fxb (
12:11:56ChristopherWbecause my patch is more of an "improvement" to the language than to the parser itself
12:12:47ChristopherWthe graphical wps authoring software can take advantage of the simpler language just as well as a person who manually edits his wps files
12:13:12 Part MartinR
12:13:12ChristopherWs/simpler/more flexible/
12:13:25LloreanI don't see it as being particularly more flexible.
12:13:54LloreanIt allows more images, but requires that they be in large subimage strips, which already average somewhat larger images that properly sized individual images (by necessity of sharing the size of the largest image) and there's a finite buffer for images already
12:14:08pixelmaand simpler is relative (see battery animation line)
12:14:28LloreanI think its added flexibility would be very, very limited, and the special cases you've described would be better solved with other solutions (progressively revealed bitmaps, like the progress bar, for volume, for example)
12:14:48pixelmain some places it's really more simple (e.g. the codec strip)
12:15:16Lloreanpixelma: But, in the case of the codec strip, it wouldn't be much more simple for a GUI tool anyway.
12:16:25ChristopherWeven progressively revealed bitmaps are too limited for me: the revealing of the bitmap is limited to a sharp vertical edge sliding horizontally
12:16:33LloreanChristopherW: Says who?
12:16:54LloreanThat's how it's currently done with the WPS, but that's not necessarily how it has to be done in the future.
12:16:54 Join cool_walking_ [0] (
12:16:58LloreanEr, with the progress bar.
12:17:07ChristopherWhow would you propose revealing a bitmap differently? using a different bitmap as a mask?
12:17:24LloreanRevealing it vertically or diagonally.
12:17:28ChristopherWwith an alpha channel?
12:18:24pixelmayour DAP is not a PC
12:18:41LloreanUsing a single bitmap, and just revealing more of it, will in a lot of cases save you a good deal of image buffer that can then be put to good use elsewhere.
12:18:44ChristopherWI know, that alpha channel bit is partly a joke
12:20:38pixelmasorry, didn't see Llorean's first mention of it
12:20:47pixelma(the revealing I mean)
12:21:12ChristopherWhow about a "palette-ized" bitmap, where ranges of indexes are shown depending on the current volume/playback position/battery level?
12:21:41ChristopherWthat would be complicated to create in a drawing program, to say the least
12:21:54*Bagder would like a to see alpha channel being seriously attempted
12:22:13Bagderjust ignore amiconn during the testing ;)
12:22:13LloreanBagder: There's an "antialiased font" patch floating in the wild somewhere, but it's not on our tracker.
12:22:24LloreanBagder: At least one of the unsupported builds out there makes use of it.
12:22:24*gevaerts proposes a gsoc project :)
12:22:36Bagderthere have been several alpha channel things in the past too
12:23:13Bagderespecially our fast-cpu targets should have little problems using it
12:23:15LloreanChristopherW: My point is, other than "allow one to use way too many images", I can't see what your patch actually adds.
12:23:32LloreanChristopherW: Under the assumption that the user will never see the .wps file, that is.
12:23:53pixelmaI'm still not convinced that an anti-aliased font looks really better (read: more readable) at the usual Rockbox font sizes (if I see results of PC graphic programs)
12:24:07ChristopherWtrue, my patch does not change anything visibly in the actual WPS
12:24:30Lloreanpixelma: I think at larger sizes (gigabeat / iPod Video screens) there are font sizes in use where antialiasing would help
12:25:13Bagderand I think an alpha channel's primary use would be for transparency things in the gui rather than "just" anti-alias
12:25:38LloreanChristopherW: So, if a GUI program comes about that can handle the complexities of the old tokens, your patch would at that point become "waste some binsize for a feature that does not affect users". I'm not saying it should be rejected now, but I think it may bear some waiting to see what happens.
12:26:02peturBagder: did you read ?
12:26:06ChristopherWI think that anti-aliased fonts are "fuzzier" on my Sansa e280 at least
12:26:16Bagderpetur: no!
12:26:29peturseems it backfired on creative ;)
12:26:35Bagderbig time
12:26:44ChristopherWplus the anti-aliasing looks very different depending on the viewing angle (read: bad at most angles)
12:26:50LloreanBagder: I'm trying to imagine what UI elements would make use of alpha channels.
12:27:10ChristopherWLlorean: a HUD perhaps?
12:27:31pixelmaChristopherW: a what?
12:27:32Bagderwe could have pop-up menus with transparent "backgrounds"
12:27:40Bagderand splashes as well
12:27:43ChristopherWyou can use a plugin but still see the WPS overlayed with beautiful translucency
12:28:14pixelmaI think I'd only find that confusing
12:28:27gevaertsAlbum art with alpha channels !
12:28:35LloreanBagder: Text on top of text is hard to read almost no matter what the translucencies are. :-P
12:28:41ChristopherWalpha-blend between the current and next album art!
12:28:48Bagdertranslucent album art across the entire screen! ;-)
12:29:03ChristopherW... just like crossfade for album art!
12:29:18BagderLlorean: it's a question of how translucent it would be
12:29:19*gevaerts thinks that one alpha channel is too restrictive. You need one per color channel
12:30:12ChristopherWwhy not two per color channel?
12:30:21ChristopherWor four?
12:30:38gevaertsHow would you use them ?
12:30:47ChristopherWmultiply them all together
12:30:59ChristopherWrather pointless, eh?
12:31:07*gevaerts thinks so too
12:31:18ChristopherWjust like alpha channels on a DAP
12:31:34 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:31:59pixelmacool_walking_: were you the one asking about building the manual under cygwin (reading the logs)?
12:32:00*ChristopherW wants smooth alpha blending on his '90's Walkman!
12:32:39*gevaerts offers ChristopherW blending of his '90's Walkman
12:33:33ChristopherWgevaerts: what, put it in a blender?
12:33:40gevaertsChristopherW: yes :)
12:33:59ChristopherWnah, I'll just take it apart and salvage the parts
12:34:29Bagderblending in a walkman, is that something like a mixed tape ?
12:34:54*Bagder ducks
12:34:57gevaertsprobably :)
12:35:05ChristopherWI think we need to rename this channel to #rockbox-community and vice versa
12:35:51linuxstbJust alpha-blend the two channels...
12:35:53 Join MethoS- [0] (
12:36:03ChristopherWwith what weighting?
12:36:26linuxstbProbably 0% 0% - I should be doing some work....
12:36:43gevaertsChristopherW: it's simple : if Llorean is awake you have to keep your forum posts on topic. If scorche|sh is awake you keep #rockbox on topit :)
12:40:03*ChristopherW needs sleep
12:40:16ChristopherW'tis almost 4 am
12:41:17 Join joshin [0] (n=joshin@VDSL-130-13-240-229.PHNX.QWEST.NET)
12:42:14 Join syn4pse [0] (
12:42:41syn4psei suppose there is no way to erase an erroneous double comment posting in flyspray :/
12:42:50*ChristopherW goes to bed and will dream that you guys will give his patch a chance :D
12:43:10gevaertssyn4pse: I can do it for you
12:43:29syn4psegevaerts: yahoo
12:44:05syn4pse#8647 - i put my updated .diff in twice
12:44:44syn4psefinally starting to get my plugin rolling...
12:44:54gevaertssyn4pse: done
12:45:04syn4pse:D thanks a lot.
12:45:30*gevaerts now edits the remaining comment to add some typos :-P
12:46:23syn4psei am pretty sure i messed up forward(s)/backward(s) −− i never get those right. further/farther −− i don't even touch those...
12:46:57syn4pseam waiting on btcreate to finish the .wwi for wikipedia
12:52:32Bagderworth a glance
12:52:42Bagderthat discussion always come up in new ports
12:54:45pixelmaBagder: you should go over to Err0r's place... ;)
12:55:14Bagderhah, yeah I was a bit surprised to see that
12:57:48syn4psehis argument doesn't hold water.
12:59:20*gevaerts still doesn't really get how money "speeds up" a port
12:59:35 Join OlivierBorowski [0] (
12:59:49Bagdergevaerts: only if there's a waiting army of poor skilled hackers without targets
13:00:14Bagderand that army is usually hard to find
13:00:41Bagderbut seriously, money sometimes help as then more people can buy targets
13:00:48Bagderand possibly sacrifice them for research
13:01:21Bagderit depends a lot on the specific targets
13:02:42Nico_Pwe need skilled low level hackers to buy gigabeasts! :)
13:03:21linuxstbNico_P: Good point - the Meizu people can't jump the queue ;)
13:03:56Bagdermrobe500, D2, Meizu, gigabeast ... we're certainly know how to focus!
13:03:56linuxstbWe should send Err0r a gigabeast...
13:04:40linuxstbWhich all have a different SoC...
13:04:42Nico_PBagder: you're even forgetting some: ZVM, v2 Sansas... :)
13:05:00Bagderoh indeed
13:05:08Bagderthe forum went down
13:06:51pixelmayes, it seems to do that around this time each day now
13:06:52Bagderand the view
13:07:06Bagderbut the v2 and the view are hardly "worked on" for real
13:09:06kritter`i ordered a h320 iriver. is that a decent player ?
13:09:16Bagderit is
13:09:20 Join MethoS-- [0] (
13:09:40Bagderdepending on what your definition of decent is
13:10:26kritter`i'll accept your definition for now. was hoping for something much better than my 2nd gen nano. going to put rockbox on the h320
13:12:51***Saving seen data "./dancer.seen"
13:13:53cool_walking_pixelma: yes
13:14:22cool_walking_pixelma: I got the same errors on andLinux and Linux though
13:15:32pixelmaFirst - do you build in a seperate directory? Second: do you use "make manual"? A simple "make" doesn't work anymore
13:15:44cool_walking_i was using "make"
13:15:48linuxstbkritter`: It's quite bulky, but is a "classic" Rockbox device, and Rockbox runs well on it.
13:16:21pixelmacool_walking_: (I've changed that in the wiki - but only after I saw your first question in the logs)
13:16:51kritter`linuxstb, ok. thank you
13:18:38 Quit OlivierBorowski (Remote closed the connection)
13:18:53cool_walking_ah.. a more sane error now. missing pdflatex.. anyone know what ubuntu package that's in?
13:21:23Nico_Pone of the texlive-latex-* ones probabably
13:21:41pixelmacool_walking_: I don't know anything about ubuntu packages, but I know the packages listed in the wiki are still the old ones (I've been told that they were linked though)
13:22:16linuxstbMaybe texlive-latex-extra
13:22:21cool_walking_yep, tetex* are linked to texlive*
13:23:40cool_walking_ah.. it's downloading ~300MB
13:23:55Nico_PChristopherW: just saw your patch... I'll try to review it later today
13:24:20cool_walking_thanks for the help, pixelma.
13:25:28linuxstbNico_P: Did you read the discussion in IRC this morning about it?
13:26:05 Quit MethoS- (Read error: 110 (Connection timed out))
13:26:36Nico_Plinuxstb: I'm in the process of reading it
13:29:28 Quit DavidSG (Read error: 110 (Connection timed out))
13:30:09 Quit MethoS-- (Remote closed the connection)
13:30:33 Join mikus [0] (n=5316c5a0@gateway/web/cgi-irc/
13:33:15linuxstbNico_P: I'm unsure about it - my initial reaction is that it's not needed any more, and makes WPSs harder to read (because you don't know immediately that sub-images are being used - you have to go back and check the %xl)
13:33:52pixelmahmm... I guess even current "boxes" looks a bit wrong on the Ipod Video? The "rating" box lies over the playlist and playtime lines (at least in my sim)
13:34:35pixelmaif you can read it that is
13:35:04 Join der-ftz [0] (
13:35:41Nico_Plinuxstb: I'm unsure too
13:37:13pixelmathe used graphics for it are also black and white whereas there are colours in the background picture
13:41:17Nico_Plinuxstb: after some thought, I'd say I agree it doesn't solve any real problem
13:41:55 Quit mikus ("CGI:IRC (EOF)")
13:43:03 Join FergoFrog [0] (
13:47:20 Quit FergoFrog (Client Quit)
13:47:22*gevaerts doesn't understand the usb problem anymore
13:49:27*Nico_P leaves for class
13:53:49 Part der-ftz
13:54:27 Join mcflow [0] (
13:54:55LloreanBagder: I've attempted to respond in that m6 money post thing, hoping to clarify a little bit. If you wouldn't mind reading it and seeing if there's anything you don't like?
13:55:38Bagdersure, I can give it quickie before I need to rush
13:56:08LloreanJust responded to the thread, hopefully to try to help clarify why "giving money to the whole" is actually going to have a greater net benefit for the M6 even.
13:57:06gevaertsThere are broken packets now and then. They always seem to have a multiple of 16 correct bytes, followed by a wrong crc.
13:57:28BagderLlorean: well put!
13:58:11LloreanBagder: I'm always slightly worried that the misunderstanding just comes from people thinking Rockbox is like... 30 different firmwares with the same name, rather than one. :)
13:58:49Bagderyes, we must remember to try to explain these basics to newbies as they really can't be expected to know it all from the start
13:59:03BagderI just tend to forget to take that into account...
13:59:42 Join n1s [0] (
14:00:30*Bagder runs off
14:03:10kritter`does rockbox support both wma and m4a ?
14:03:45linuxstbkritter`: Yes, as long as they're not encrypted (DRM'd)
14:03:56kritter`oh good. thats helpful.
14:04:22linuxstbFor m4a (which is just a container), Rockbox supports AAC (the most common codec), and Apple Lossless (ALAC).
14:05:30 Join luckz [0] (
14:05:49dionoeaHowdy. Looks like the forums are broken
14:05:59dionoea"Database error"
14:06:00kritter`i have an ipod nano and i was devastated that after adding some new files i could not get a listing of anything on ipod. it had a good number on them previously never thought there would be a problem.
14:06:51kritter`so i decided to do the best thing and get an iriver h320 which i hope is much better than the nano with rockbox on it (even without)
14:07:06kritter`yes i heard that maybe 1.5 hours ago dionoea not sure when they will be back up
14:07:22pixelmaI guess is the WPS backdrop image is also stored in the WPS image buffer, meaning I can't save anything by making a bmp a backdrop instead of a loaded image?
14:07:23Lloreandionoea: They're not totally broken. Refreshing sometimes works. Unfortunately, nothing I can do.
14:07:42dionoeaAh ok. I'l just have to wait then :)
14:07:47linuxstbpixelma: No, that's a separate buffer IIRC...
14:07:55Lloreanpixelma: I think the backdrop gets a sep... yeah, what linuxstb said
14:08:21 Quit corevette (Read error: 110 (Connection timed out))
14:08:37linuxstbpixelma: Yes, it does - there are two backdrop buffers declared (main and wps)
14:10:03 Quit davina (Read error: 104 (Connection reset by peer))
14:10:04pixelmagood, so I should try - if I make the rating box in "Boxes" a colour bitmap, I get "read_bmp_fd: Bitmap too large for buffer: 21120 bytes." on the Ipod Video. But it doesn't use a backdrop yet but a large bitmap across the whole screen using a lot of transparent magic magenta
14:10:59pixelmathat would be solved by making it a backdrop... I already wondered because besides that one the bitmaps are quite small
14:11:42 Join davina [0] (
14:11:54LloreanWell I mean, a fullscreen color bitmap would be something like 1/3 of the bitmap buffer right there, isn't it?
14:15:19pixelmahow large is the bitmap buffer?
14:16:40LloreanThere's a formula for it somewhere. It's defined in terms of screen dimensions, and bit depth.
14:16:42 Join Spex [0] (
14:17:15LloreanSo actually a fullscreen image would be a more significant portion of it than I thought.
14:17:45*gevaerts thinks that this should be a malloc()/realloc() system :-P
14:17:49*gevaerts runs away
14:17:52linuxstbgevaerts: What a great idea!
14:18:19*linuxstb doesn't understand the point of the second part of that formula - ignoring depth...
14:19:07Lloreanlinuxstb: Enough for one screen's worth of color images, and two screen's worth of mono-images.
14:19:20LloreanSince mono images are handled separately (and were more used back when we didn't have a transparent color)
14:21:11 Part Spex
14:25:41amiconnlinuxstb: That formula reserves space for 1 fullscreen native bitmap + 2 fullscreen mono bitmaps
14:25:55amiconnBah, too late
14:26:51krazykitoh yikes. database error on the forums.
14:27:26 Join midgey [0] (
14:27:31linuxstbamiconn, Llorean: Thanks. I think I'll comment that in the code...
14:28:25pixelmabtw. I was a bit surprised that mono bitmaps in the WPS are drawn as I would expect it - black as foreground and white as background colours. Wasn't that a problem with the icons in the menus?
14:29:11 Join ali_as [0] (
14:32:45 Join Spex [0] (
14:32:47 Quit Spex (Client Quit)
14:33:29 Join Spex [0] (
14:34:07 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
14:34:13 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:34:28 Quit ch4os_ (Read error: 104 (Connection reset by peer))
14:40:36 Quit thegeek (Read error: 104 (Connection reset by peer))
14:41:56Lloreanpixelma: I thought mono bitmaps were supposed to display as "fgcolor" and "clear"
14:46:44 Quit knittl (Remote closed the connection)
14:49:21pixelmawell, yeah maybe. But it looks like it doesn't work currently (you could try the current "Boxes" theme on an Ipod Video or sim) but that's a different thing than what I meant. If I remember correctly then the old mono bitmaps were always drawn with black as foreground colour (maybe white was "clear" there) and people reported problems who used a bright on very dark colour scheme...
14:49:55pixelma...which is why JdGordon invented "magic cyan" for them to be interpreted as foreground colour
14:50:46Lloreanpixelma: Ah, no.
14:50:48pixelmaI thought...
14:50:56LloreanMono bitmaps worked right (well, were working right at one point)
14:51:13LloreanMagic Cyan was for the icons, because those weren't actually treated the same way as mono bitmaps
14:51:46*Llorean can't remember why though
14:52:13pixelmayes, that's why it surprised me that they work somewhat correctly in WPSs (at least without magic colours)
14:52:44LloreanWell, afaik, mono bitmaps have worked in the WPS for quite some time.
14:53:09pixelmaor the other way round: it surprised me that it didn't work correctly in menus...
14:53:20LloreanYeah, I can't remember why they didn't work in the menus.
14:53:43linuxstbIIUC, there is only one type of icon - "native format", but in the WPS, the two types of bitmaps are distinguished when read from the .bmp file.
14:56:40pixelmaI just noticed that my longer response misses an important word... I was speaking about the "old mono icon bitmaps"
14:57:55pixelmaa backdrop works nicely, thanks
14:59:08*Llorean is glad to hear it.
15:04:28 Quit gromit` ("Coyote finally caught me")
15:04:46 Join gromit` [0] (
15:04:58 Quit LambdaCalculus37 ("CGI:IRC (EOF)")
15:05:26 Join thegeek [0] (
15:05:26 Join Davide-NYC [0] (
15:06:09Davide-NYCIs anyone else getting build errors in Cygwin?
15:08:29markunBagder: glad to see it's not only us who are against a port specific donation fund
15:09:14 Join fehmicans [0] (n=canavar@
15:11:30pixelmaDavide-NYC: like?
15:12:06Davide-NYCmake[1]: *** [rocks] Error 2 followed by make: *** [build] Error 2
15:12:52pixelmaI've build a sim (but a hwcodec) sim earlier today with a checkot of yesterday though
15:12:53***Saving seen data "./dancer.seen"
15:13:22pixelmadid you try a "make clean" and running configure again?
15:14:21 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:17:06Davide-NYCI did both of those things a few times
15:17:16 Join Thundercloud [0] (
15:17:17Davide-NYCI guess I'll try a fresh checkout
15:17:33Davide-NYCI also tried updating the rockbox portion of cygwin
15:21:09 Join ali_as_ [0] (
15:23:29 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
15:23:38 Quit shotofadds (Read error: 104 (Connection reset by peer))
15:24:19 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
15:25:00pixelmaDavide-NYC: is there an error before that? I'm currently trying an Ipod Video sim now after an "svn up"
15:25:30 Quit ali_as (Read error: 110 (Connection timed out))
15:25:57 Quit Thundercloud (Read error: 104 (Connection reset by peer))
15:26:30Davide-NYCpixelma: not really, just the usual build output. I've just updated cygwin (I believe) and I have deleted all of the RB folders. Currently DLing a fresh trunk
15:27:02 Join Thundercloud [0] (
15:27:03*Davide-NYC has exactly 12 minutes
15:27:33shotofaddsdoes anybody object to splitting out the PCF5060x register defines, as in ? I don't really know whose area this is.
15:32:12pixelmaDavide-NYC: the sim built fine here but I haven't changed something in the setup for a looong time...
15:33:41Davide-NYCcan you build a sanca c2x0 sim for me?
15:33:42 Quit DataGhost (Nick collision from services.)
15:33:50 Join DataGhost [0] (n=dataghos@unaffiliated/dataghost) 4 minutes!
15:34:27 Join perrikwp|class [0] (i=98213a67@gateway/web/ajax/
15:34:42pixelmaI can't built that in 4 minutes...
15:35:51 Join kugel [0] (n=kugel@unaffiliated/kugel)
15:36:04pixelmaDavide-NYC: I have on lying around here from 1st of April (no joke) but it also some changes (keymap, mod codec and sokoban)
15:36:50Davide-NYCpixelma: thanks, I will have to revisit this problem later on tonight.,
15:36:56 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
15:37:52 Quit LambdaCalculus37 (Client Quit)
15:38:30 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
15:38:44 Quit LambdaCalculus37 (Client Quit)
15:39:07 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
15:39:44*LambdaCalculus37 slaps CGI:IRC
15:39:49 Join mcuelenaere [0] (
15:42:50 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
15:43:53 Quit Davide-NYC ("ChatZilla 0.9.81 [Firefox]")
15:47:23 Join qwedsa [0] (
15:48:08 Join mcuelenaere [0] (
15:52:57 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
15:57:18 Join FOAD_ [0] (
15:57:20 Part Spex
15:59:03 Quit qwedsa_ (Read error: 110 (Connection timed out))
15:59:49n1sshotofadds: is that new pcf506060.c file d2 specific ore just some general interface stuff?
16:01:08shotofaddsn1s: it's intended to be generic, the existing 50606 driver is very coldfire-specific.
16:01:19shotofadds(ie. in target tree)
16:01:33 Join Xezzy [0] (
16:02:04 Quit cool_walking_ (Read error: 110 (Connection timed out))
16:02:26n1sshotofadds: yeah was just wondering if the generic could be used as an interface to that so that the interface is similar across ports (and the code slightly less confusing)
16:03:01amiconnThe coldfire pcf driver is like that now because of optimisation.
16:03:10 Join Christopher1 [0] (
16:03:21shotofaddsthe interface (.h) is shared between the two, at least for those functions implemented.
16:05:19domonokyare the forums dead again ?
16:05:28LinusNdomonoky: looks like it
16:05:33n1sah yes, but the code in the new pcf506060.c is actually only used for some of the targets with that chip so that's why I thought it would be a bit confusing but, feel free to ignore me :)
16:05:36LinusNand there is nothing we can do...
16:05:54 Join jgarvey [0] (
16:05:54 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
16:05:58*domonoky thinks we should really move to forums to our own servers..
16:06:11domonokys/to/the :-)
16:06:14 Join corevette [0] (
16:06:44shotofaddsn1s: I think it certainly deserves a comment to that effect, at least...
16:06:50LambdaCalculus37domonoky: I agree. Having the forums bork on us often is annoying.
16:07:36LinusNdomonoky: i agree completely
16:13:03 Quit FOAD (Read error: 110 (Connection timed out))
16:13:04 Nick FOAD_ is now known as FOAD (
16:13:12linuxstbLinusN: Would you want to host them at haxx?
16:13:30LinusNlooks like we don't have much choice
16:13:44linuxstbscorche has shown an interest in hosting them.
16:13:50LinusNah yes
16:13:53 Join mcuelenaere [0] (
16:14:31LinusNlooks like jeff has lost interest anyway, since he doesn't fix them, and won't reply to my emails
16:15:24linuxstbYes, it's annoying when misticriver is working fine, and we're forgotten about...
16:15:40LinusNthe only problem is that we might not be able to migrate the existing data...
16:15:58LambdaCalculus37Crap, what does that mean for all of the posts?
16:16:07linuxstbI thought Llorean has been taking backups?
16:16:14LinusNhe did?
16:16:29dionoeayou usually can get a full db dump through the admin interface
16:16:30linuxstbHe's been saying he has...
16:16:35dionoea(at least you can in phpbb)
16:16:37LinusNaha, good
16:16:51LinusNthen we won't lose that much
16:17:01LinusNare the attachments stored in the db too?
16:17:22 Join Spex [0] (
16:17:27linuxstbNo idea, but I don't think they matter - the limit on filesize made the attachments feature useless.
16:17:48LinusNthen i say we move the forums asap
16:18:05GodEaterdo we have control over the DNS entry too ?
16:18:18 Quit Xezzy (Read error: 104 (Connection reset by peer))
16:18:26 Quit ChristopherW (Read error: 110 (Connection timed out))
16:18:53*linuxstb pings scorche and Llorean to get things moving...
16:19:06 Join Xezzy [0] (
16:19:27LinusNi wonder why he doesn't respond though...
16:19:45n1sLinusN: any chance of getting an updated h300 bootloader released?
16:19:47LinusNhe is a busy man, but i emailed him several times
16:20:02Xezzywill rockbox display custom tags (txxx) in the future?
16:20:08linuxstbDo you have a private email address for him, or do you just use the jeff at mistic river address?
16:20:10LinusNn1s: are all the ata issues resolved?
16:20:31LinusNlinuxstb: i only have the misticriver address
16:20:49n1sLinusN: I have no idea petur was playing with it last I heard it was crashing somewhere in thread.c so i guess not...
16:21:13*LambdaCalculus37 wonders if Jeff is blatantly ignoring everyone, or really is too busy to respond
16:21:40peturyes, my bootloader will either hang or crash in some thread code
16:22:28peturit seems to be some subtle timing thing, adding a sleep() at the right spot can make it crash reliably :(
16:22:40LinusNLambdaCalculus37: still, if he is that busy, he shouldn't run our server anyway
16:22:54LambdaCalculus37LinusN: Very true.
16:23:08LinusNespecially since he locked us out
16:23:25LambdaCalculus37What the deuce!?
16:23:35LambdaCalculus37Why the hell would he do that for?
16:23:50LinusNi dunno, i can't log in to the server anymore
16:23:59peturLinusN: if only I had known, I would have waited with releasing my CF patch - Jeff is also using CF ;)
16:24:34LambdaCalculus37LinusN: What about Zagor and Llorean? They're the only other admins besides yourself and Jeff.
16:24:35peturmaybe I should have a chat with him ;)
16:24:41LinusNstill, i think he just changed the password for some other reason, and forgot that we depended on it
16:26:18LinusNLambdaCalculus37: what about them?
16:27:13 Join domonok1 [0] (
16:28:13LambdaCalculus37LinusN: Never mind. I just remembered you wrote "locked us out". :)
16:28:30*LambdaCalculus37 should get new glasses :)
16:28:32LinusNwell, all he did was change the password for the rockbox account
16:28:58LinusNwhich isn't the same as the forum admin interface
16:28:59dionoeaor removed it...
16:29:23 Quit Xezzy ("ChatZilla 0.9.81 [Firefox]")
16:30:35LloreanLinusN: He'd also given me an ftp account for uploading files (like the badges) for the forums, and changed the password on that too.
16:30:45LloreanSo at least two accounts were changed.
16:30:50LinusNthere you go
16:31:02*linuxstb looks at "today's posts" on misticriver and concludes the Rockbox forums are far busier
16:31:02 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
16:31:19gevaertslinuxstb: not right now :(
16:31:44linuxstbgevaerts: So that's the reason... ;)
16:31:56 Join Rincewind [0] (
16:32:17Lloreanlinuxstb: I gave scorche a backup of our server, but when he's running the script sql is reporting problems. It may be though that the backup comes from while we've got broken tables.
16:33:01linuxstbIs scorche using the same version of mysql?
16:33:22LloreanNo, much newer.
16:33:27LloreanOur servers are MySQL 4
16:33:32linuxstbThat could be an issue as well.
16:33:43LloreanBut I'd successfully run the backup script on an MySQL 5 server about six-ish months back.
16:34:02LloreanWhen duplicating the forums to my personal machine to play with a few things safely
16:34:16 Join dabujo [0] (
16:34:53linuxstbLlorean: I don't know mysql (I use postgresql), but I'm happy to take a look and see if I can see any problems.
16:35:47LloreanI don't really know mysql either.
16:36:06 Join toffe82 [0] (
16:36:49LloreanThe forum backup .sql file is about 90mb, and i don't remember which line was reporting an error for scorche.
16:36:57 Quit perrikwp|class (" ajax IRC Client")
16:37:00LloreanIt gzs down to about 27 or so though
16:37:17linuxstbI can try importing it into mysql and see what happens.
16:37:49LloreanCan you accept dcc?
16:37:57markunBagder, Llorean: nice reply from kgb2008 at meizume
16:38:13peturbut if we move we need a recent backup anyway? Can we get that?
16:38:34Lloreanpetur: I have one less than 2 days old
16:39:05 Join mcuelenaere [0] (
16:39:10LloreanI also have one that's 15 days old, but comes before the tables were broken again, so may be the last "good" one depending on what's preventing the import.
16:39:32linuxstbLlorean: Yes, DCC should work
16:42:06LambdaCalculus37markun: Do you have a link to the reply?
16:42:12gevaertsIf you can't figure it out, I'm willing to have a look as well (same as linuxstb though, no real mysql experience)
16:43:03LambdaCalculus37I could take a look as well, even though I have very little MySQL experience myself.
16:43:33*gevaerts thinks that a mysql port to rockbox is not very probable. Noone seems to know it...
16:43:44 Quit domonoky (Read error: 110 (Connection timed out))
16:44:01*Llorean thinks sqlite is vastly more suitable anyway
16:44:04 Join Dark_Apostrophe [0] (n=darkapos@supporter/monthlybyte/DarkApostrophe)
16:44:18linuxstbI'll try importing it, and then post any error messages here if I can't figure them out.
16:44:18Dark_ApostropheDoes RB support .m4b files?
16:44:36Nico_PLlorean: would sqlite seriously be suitable for roxkbox?
16:44:49Nico_PDark_Apostrophe: yes it does
16:44:50gevaertsNico_P: why not ?
16:44:51markunDark_Apostrophe: long time no see I think
16:45:00Nico_Pgevaerts: I don't know the requirements
16:45:05Dark_ApostropheNico_P: Thanks
16:45:12pixelmaNico_P: I don't know, I don't know roxkbox :P
16:45:14Dark_Apostrophemarkun: Hehe, been having computer troubles lately :P
16:45:19LambdaCalculus37Dark_Apostrophe: As long as they're DRM-free, of course. ;)
16:45:28LloreanNico_P: SQLite is designed as an embedded database solution. If any SQL is suitable, it's the one. I'm really not sure though. :-P
16:45:30*Nico_P has troubel typing
16:45:33Nico_Psee? :)
16:45:34Dark_ApostropheLambdaCalculus37: Hmm... I don't know if it is or not. I guess I can give it a try.
16:46:20LloreanNico_P: Requires 16 KB of stack, and 100 KB of heap. the library itself can apparently be 250KB with all features enabled. So, sounds "small"
16:47:09gevaertsNico_P: From the website : 'SQLite is a good fit for use in cellphones, PDAs, set-top boxes, and/or appliances.'
16:47:22Lloreangevaerts: My toaster could use a good database...
16:48:20LambdaCalculus37Llorean: For a variety of bagels and breads? :)
16:48:45*Nico_P wonders whether any OF uses sqlite as a backend
16:49:30 Quit corevette (Remote closed the connection)
16:50:15 Quit syn4pse ()
16:50:35linuxstbLlorean: It seemed to import fine...
16:50:46gevaertsLlorean: you need
16:51:35linuxstb114541 rows in the main "smf_messages" table.
16:51:40LambdaCalculus37gevaerts: Let's get one for DevConEast and DevConWest... we'll make toast for breakfast with a Rockbox logo on each slice. :P
16:52:12gevaertsDevConToast, the toast for those occasions when you don't have a date !
16:52:33Lloreanlinuxstb: Well that's a slightly newer backup than scorche tried. So either the previous one was corrupted, or it's something up at the server he's using.
16:52:45markunWhat about DevConDownUnder?
16:52:49LloreanI guess we'll see next time he's around
16:53:01linuxstbLlorean: I just installed what's in Debian unstable - "mysql Ver 14.12 Distrib 5.0.51a, for debian-linux-gnu (i486) using readline 5.2"
16:53:20gevaertsmarkun: let them eat cake :)
16:55:10linuxstbLlorean: Ah no, there was an error in the search table...
16:56:07*gevaerts expects a search table to be disposable
16:56:21linuxstbSeems there was an error dumping that table.
16:56:21 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
16:56:40Lloreanlinuxstb: Well, probably because the table needs repairing on the server, I'd imagine.
16:56:43linuxstbgevaerts: Yes, it doesn't seem important.
16:56:54linuxstbsmf_log_search_results is the exact name of the table.
16:57:20LinusNyou can ditch that table
16:57:49linuxstbLinusN: You mean drop it completely?
16:58:12 Join in-ferno [0] (
16:58:15LinusNwell, you should probably create it, but you won't have to populate it
16:59:17 Part LinusN
17:00:43 Join MethoS- [0] (
17:02:47 Quit wisp_sh (Read error: 110 (Connection timed out))
17:03:28 Join mf0102 [0] (n=michi@
17:04:17 Join mcuelenaere [0] (
17:05:26in-fernohi there, just a message to say good luck to sansa e200v2 port !
17:06:22markunin-ferno: do you know who's working on it?
17:06:56in-fernoI saw some members on Rockbox thread that were working on it yeah
17:07:09in-ferno*Rockbox board sorry
17:09:05kugellinuxstb: did you look at v9 of my custom list patch?
17:10:33linuxstbkugel: No, not yet.
17:11:37n1sanyone have an idea why plugin and codec building is working differently plugins using SUBDIRS etc?
17:11:47kugellinuxstb: if you have time
17:12:20linuxstbn1s: They all have their own Makefiles, so there are very likely inconsistencies...
17:12:55***Saving seen data "./dancer.seen"
17:13:41n1slinuxstb: yes but the plugins/Makefile uses SUBDIRS and lets the dir plugins do all the linking etc themselves while the codecs/Makefile does all the linking of codes and the individual makefiles just make libs
17:14:16amiconnThis is because some subdir plugins *need* their own linking (for overlaying on archos)
17:14:58n1samiconn: IHMO the plugin way seems simpler and more self contained, any reason it would not work for codecs?
17:15:34 Quit petur ("connection reset by belgian beer")
17:15:54 Join lee-qid [0] (
17:16:09linuxstbLlorean: Ignore what I said earlier about no errors - the way I was doing the import skipped them and carried on... I've now found another error - an illegal duplicate entry in the smf_log_search_subjects table.
17:17:35GodEaterlinuxstb: I don't think you need the content of that table - just arrange for an empty copy of it
17:18:06Lloreanlinuxstb: Well, hopefully the forums will be up long enough at some point, with unbroken tables, for us to get a clean export.
17:18:22Lloreanlinuxstb: Otherwise, I have one from two weeks back, as I said, that *ought* to not have the currently broken tables in it.
17:18:43linuxstbIt seems to be a bug in the export feature somewhere - it's exporting a text string without quotes if the content is numeric - i.e. "00" is being output as plain 00, and getting confused with "0" (output as 0)
17:19:28 Quit midgey ()
17:20:53 Quit Nevtus ("Gone")
17:20:53 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
17:21:15 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
17:22:03gevaertslinuxstb: do you have mysql4 available ? Those sound like the kind of errors that may be caused by a stricter database
17:24:39linuxstbgevaerts: No, but I've fixed them now...
17:26:04linuxstbNow it's imported correctly.
17:26:56 Quit DerDome (Read error: 110 (Connection timed out))
17:29:58 Join DerDome [0] (
17:42:54 Join tomcat_ha [0] (n=F14_fana@
17:43:07 Join MethoS-- [0] (
17:43:08tomcat_hamy ipod fell
17:43:21tomcat_hait no longer wants to boot
17:43:30tomcat_hait says connect to itunes to restore
17:43:54tomcat_haany advices?
17:44:28 Quit lee-qid (Read error: 110 (Connection timed out))
17:44:29LambdaCalculus37Hard drive based or flash based? (Silly question)
17:45:37tomcat_haipod gen 5.5
17:45:52 Quit GodEater (" ajax IRC Client")
17:46:28LambdaCalculus37Can you get into Emergency Disk Mode?
17:46:47tomcat_hahow do i do that?
17:47:10LambdaCalculus37MENU+SELECT to reboot, then immediately press SELECT+PLAY at the Apple logo.
17:47:15LambdaCalculus37EDM should come right up.
17:48:51 Join mcuelenaere [0] (
17:49:57tomcat_hai dont need to hold it down then right?
17:50:15 Part Spex
17:50:24LambdaCalculus37You have to hold down SELECT+PLAY until EDM comes up.
17:51:01 Quit gevaerts ("work->home")
17:51:11tomcat_hait just says ok to disconnect now
17:51:26LambdaCalculus37Simple black and white screen?
17:51:58tomcat_hawith disk mode at top
17:52:18LambdaCalculus37That's Emergency Disk Mode. Get your USB cable out now and connect it to your computer.
17:52:40LambdaCalculus37See if you can at least read the disk.
17:53:10 Join thundr2 [0] (
17:54:08tomcat_hadoesnt look like it
17:54:15tomcat_hawin explorer seems to have frozen up
17:56:46tomcat_hacant access it
17:58:51LambdaCalculus37Give me a bit to think of a solution.
17:59:15tomcat_hathanks anyway ^^
17:59:40LambdaCalculus37Sorry about that.
17:59:50LambdaCalculus37See if a restore and a reinstallation will help out, though.
18:00:04LambdaCalculus37If not, then you may have damaged the hard drive.
18:00:09 Quit MethoS- (Read error: 110 (Connection timed out))
18:00:31pixelmarasher: is the red caused by your server?
18:01:49tomcat_harestore with itunes right?
18:02:16LambdaCalculus37Yes. See if it helps.
18:02:20pixelmarasher: I meant in contrast to some coding error, but looks like it
18:03:29tomcat_hadoes it matter what verison of itunes i download?
18:03:43 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
18:03:45 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
18:03:47LambdaCalculus37No, but best to just use the lastest version.
18:04:46tomcat_hamanaged to keep this bloatware of my laptop till now :|
18:05:13 Quit in-ferno ("Quitte")
18:05:46rasherpixelma: I don't understand why that happens.. I'll get a segfault every now and then, but no pattern at all :\
18:05:49tomcat_hawouldnt even have gotten the ipod if there was decent competition memory wise
18:06:55 Join webguest55 [0] (n=c6984602@gateway/web/cgi-irc/
18:08:40LambdaCalculus37tomcat_ha: I understand what you're saying, but please try to keep it on-topic here in the channel.
18:10:18 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
18:11:44 Join mcuelenaere [0] (
18:13:12 Quit webguest55 ("CGI:IRC")
18:16:07amiconntomcat_ha, LambdaCalculus37: No need to install full-bloat^h^hwn itunes for restore. The ipod updater is sufficient
18:17:24thundr2The forums appear down. Are they simply too busy right now?
18:18:08 Join gevaerts [0] (
18:18:10LambdaCalculus37amiconn: Ahh, that's right. Thanks for pointing that out. :)
18:18:38Rincewindpixelma: you worked on the sansa c240 keymap, right?
18:19:14pixelmamade a patch (for discussion and also for idea)
18:19:16 Join argumentD [0] (
18:19:40pixelmaFS #8824 if you want to have a look
18:19:59Rincewindi just installed rockbox on my new sansa c240 and I really miss the button to go from the menu or file browser back to wps
18:20:42pixelmayes, one part of the patch... because I missed it too
18:21:19RincewindI see, first part of the patch, exactly my idea
18:21:45LambdaCalculus37pixelma: I've gotten pretty used to your new keymapping for the c200 now.
18:21:55LambdaCalculus37I guess I needed a couple of days to get into the feel of it.
18:22:16pixelmaand I still hope for some ideas (especially on the virtual keyboard issue
18:22:31 Join miepchen^schlaf [0] (
18:23:22*LambdaCalculus37 needs to use the virtual keyboard on the c200 a little more, it seems :)
18:24:49tomcat_hawhich ipod updater would that be?
18:24:52tomcat_haofficial apple?
18:25:05LambdaCalculus37The same!
18:25:30Rincewindpixelma: and i see we have the short rec button question again (in your patch), but I won't discuss anything about that now
18:25:57n17ikhhas it ever been considered to make keymappings modular and easily user-configurable?
18:26:11gevaertsn17ikh: considered and rejected
18:26:17LambdaCalculus37tomcat_ha: Here's one:
18:26:28n17ikhI see that it's some of the biggest arguments in the forums over keymappings and keymap changes
18:26:35 Join bluebrother [0] (
18:26:42Rincewindn17ikh: you don't want to sugesst anything about that issue, trust me on that
18:27:01 Quit XavierGr (Nick collision from services.)
18:27:13 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
18:27:38n17ikhI guess even the issue of configurability was contentious
18:29:50 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
18:29:53pixelmaRincewind: but please feel free to comment in the tracker, I really put it there to have something to work with (and wasn't satisfired with everything)
18:30:14Rincewindpixelma: I'm typing as you speak
18:31:03pixelmanice, thanks
18:31:18Rincewindbut I have used rockbox with the c200 for maybe 5 minutes now, comparing it to my e200 and my h120
18:31:37Rincewindand I see the problems we have without dedicated up/down buttons
18:32:11Horschtheh, pegbox is a nice game, but the controls on ipod are not realy working, I think
18:32:44pixelmaRincewind: yeah, it's kind of weird...
18:32:59 Part tomcat_ha
18:33:17Horschthaving "previous level" and "next level" mapped to the scrollwheel.... makes accidental level skips very likely
18:35:53 Join mcuelenaere [0] (
18:38:44 Quit n1s ()
18:40:41 Quit desowin ()
18:41:37 Quit Thundercloud (Remote closed the connection)
18:41:45 Join tomcat_ha [0] (n=F14_fana@
18:42:54 Join desowin [0] (n=desowin@atheme/developer/desowin)
18:43:39 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
18:44:16kritter`is it possible in rockbox to −− instead of having a playlist where you know whats behind and ahead of you, play all of the songs in random order ?
18:44:32 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:45:21 Quit Mathiasdm ("Yuuw!")
18:45:53domonok1kritter`: just make a playlist of all files, and play it with shuffle enabled.. for details see manual :-)
18:46:04 Quit argumentD (Read error: 113 (No route to host))
18:46:40 Part thundr2
18:46:45 Quit davina (Remote closed the connection)
18:47:19 Join davina [0] (
18:48:24 Part mcflow
18:50:29bluebrotherkritter`: Rockbox is playlist based, so you'll always have at least a dynamic playlist. But you can add all songs to this playlist and shuffle it.
18:50:50 Quit davina (Remote closed the connection)
18:51:06*bluebrother notices that there already was an answer :o
18:51:47Rincewindpixelma: what do you think of a short press of rec to resume in file browser on the c200?
18:52:17Rincewindthis would leave the volume buttons functional, which is nice, too
18:53:01 Join davina [0] (
18:54:12 Quit pixelma (Read error: 110 (Connection timed out))
18:57:00 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
19:02:28 Join mcuelenaere [0] (
19:03:45 Quit tvelocity ("Αποχώρησε")
19:03:59amiconnmarkun: ping
19:05:52 Join ol_schoola [0] (
19:06:48markunamiconn: pong
19:07:03amiconnWill you do the splitting?
19:08:14amiconnjhMikeS: around?
19:12:56***Saving seen data "./dancer.seen"
19:13:59 Join waldo [0] (
19:15:49markunamiconn: I was planning to, but linuxstb told me he thought the splitting could be avoided by just making some header files with target specific defines
19:16:04markunso I put it on hold
19:16:19amiconnImo splitting the .lds files is better, because it makes things much more readable
19:16:34amiconnEven with #defines, you'd still have all this ifdefing
19:17:34amiconnAnd I am asking for a specific reason - I want to introduce another section for portalplayer: the uncached alias addresses, so that it's possible to put variables there without performing steps at runtime
19:17:34 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
19:18:05markunamiconn: are you planning to do the portalplayer shuffle in the target tree?
19:18:17amiconnI need this for the greylib on COP
19:19:07amiconnIf the plugin is allowed to stop playback I can use IRAM, otherwise that's forbidden
19:19:35amiconnBut if the greylib runs on cop, data that is accessed by both cores must be uncached
19:20:04amiconnPerhaps I need to do that :\
19:20:36markunI wouldn't mind doing that as well, but I trust you more than myself for making less mistakes :)
19:20:40 Join bughunter2 [0] (
19:21:03amiconnRemember my erroneous commits... :(
19:21:28markunI've got some pretty big scores in the build table as well
19:21:43 Join midgey [0] (
19:21:44markunshall I give it a go and show you a patch to review?
19:21:49 Nick Christopher1 is now known as ChristopherW (
19:22:07*amiconn needs to un-mess his tree :/
19:22:28markunI use git now which is really nice
19:22:51markunyou just start a new branch from your 'master' branch and work on some sub-project
19:22:59markunit's really a pitty that it doesn't work well in windows yet
19:23:17*amiconn still doesn't understand how git could help
19:23:52markunyou can also keep a clean svn tree and make copies of that, or what else are you suggesting?
19:24:09amiconnNo, I need to finish my half-done stuff
19:24:38amiconnI do want all changes in one tree, because that's what I want to use for my own builds
19:24:39markunit doesn't help you to get the work done of course :)
19:24:51 Join mcuelenaere [0] (
19:25:11markunin fact I now have quite a number of branches with the beginnings of something which just lay around and I might never finish
19:26:01*linuxstb now mostly uses a single tree, and keeps patches of things that aren't finished
19:26:15linuxstbWhich I guess is similar to git.
19:26:29markunlinuxstb: but git updates the patches as you go along
19:26:57markunI used to use your system, but sometimes the patches need some work before they can be applied after a "svn up"
19:27:13amiconnI keep all my changes in a single tree
19:27:32amiconnOtherwise I would loose track of what I changed very quickly
19:28:54linuxstbmarkun: How is that different to "svn up" generating conflicts?
19:30:48 Join sio [0] (n=54ca3917@gateway/web/cgi-irc/
19:30:48markunlinuxstb: with conflicts you can just edit the source file and remove the conflict. With a patch I never figured out how to really do that.
19:31:13markunwhat I did was revert the tree, apply the patch, update, fix the conflic and then make a new patch
19:31:44sioAre iPod Classic going to be suported?
19:31:57markunsio: doesn't look like it
19:32:23sioOK :(
19:32:41markunI haven't seen any progress at getting code running on it or dumping and decrypting the firmware or something like that
19:32:58 Quit bluebrother ("leaving")
19:33:22markunsio: usually rockbox only starts working on a player after it's out of production ;)
19:33:50 Quit MethoS-- (Remote closed the connection)
19:33:58markunbut seriously, I don't think anyone know where to start hacking the ipod classic
19:34:45LambdaCalculus37markun: What it's going to take is decrypting the original firmware.
19:35:09markunI have no idea. Didn't really look into the ipod classic.
19:37:00 Quit Nevtus ("Gone")
19:37:35LambdaCalculus37I look at one. Even the drive layout doesn't look the same as previous iPods.
19:38:17amiconnMaybe they've just hidden the layout via usb
19:38:17 Quit sio ("CGI:IRC (EOF)")
19:38:37amiconnOne would need to remove the hdd and hook it up to a pc
19:39:23 Quit ol_schoola ()
19:39:24 Quit mf0102 ("Verlassend")
19:39:26amiconnShould be doable with the 80GB model, but the 160GB will need further thinking because of the CE-ATA interface
19:39:29 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
19:39:51linuxstbThat would be my guess as well. But that probably wouldn't help - it could still be encrypted. I think you can download a Classic firmware via Apple's website in the same was as all the other firmware partitions.
19:42:04Horschtpeople have succesfully flashed an usigned Firmware to ipod touches
19:42:54LambdaCalculus37Horscht: Do you have a link to a site that shows this?
19:43:10Horschtdate says march 30th
19:43:16*amiconn would only be interested in the classic 160GB - and only because of its capacity
19:43:38desowinit says iphone
19:43:54amiconnI'm not interested in the touch/iphone/nano at all
19:44:35desowinoh, there's "(and ipod touch)" there
19:46:10 Part linuxstb ("Leaving")
19:46:25*gevaerts wonders if maybe amiconn doesn't like touchscreens ;)
19:46:40Horschtwell, desowin. aren't ipod touch and the iphone actualy the same hardware (minus some phone stuff on the itouch of course)
19:46:52amiconnCertainly not - and I also don't like devices which are too big for their capacity
19:47:28amiconnOf the more recent flash-based rockbox targets, I deem the c200 the most interesting (because it's _small_)
19:48:15HorschtI myself would be interested in an ipod touch if it had at least 80GB
19:49:41*gevaerts thinks that touchscreens are very useful in general, but doubts if that carries over to an audio player. After all, you want to operate those by touch alone
19:50:17Rincewindamiconn: I noticed that the c200 has less electrical noise in the headphones than the e200
19:50:33 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:51:12 Quit Xerion (Read error: 110 (Connection timed out))
19:51:12*amiconn thinks that touchscreens are nice for *stationary* devices which *require* you to look at the screen while operating them
19:51:35amiconnBut for mobile devices they don't make sense to me
19:52:11amiconnIdeally I want to be able to operate the thing without looking at it - that's why I much prefer mechanical buttons over touchy stuff
19:52:40gevaertsexactly. Unless you want this media player for video _only_ (and even then...), you want real buttons
19:53:39amiconnAnd those mechanical buttons should be placed on one face of the device (ideally the front), because that makes one-handed operation easy
19:53:46 Quit XavierGr (Nick collision from services.)
19:53:57 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
19:54:11amiconnEver tried to operate a mobile touchscreen device one-handed?
19:54:18*gevaerts doesn't want to try
19:54:26 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
19:54:50*LambdaCalculus37 tried once... it wasn't easy at all
19:54:54amiconnThat is, without placing the thing on some surface, but holding it.
19:55:15 Join roolku [0] (
19:57:10toffe82this because the interface has a bad design :) I have an idea on this one
20:04:19ChristopherWtoffe82: are we now saying that Apple designed a bad UI? :o
20:09:56 Join Xerion [0] (
20:13:18toffe82I am saying that ;)
20:13:48toffe82because they didn't think about people with one hand for example
20:14:57ChristopherWor no hands
20:17:15toffe82you can use a pda with one hand, because of the button, so what is missing on the itouch ? just a simulation of the buttons on the lower left or right corner of the screen so you can use it with you thumb
20:17:31Horschtso? they don't think about visualy impaired people either
20:18:47Horschtor people without hands
20:19:20 Join mcuelenaere [0] (
20:20:37 Join j8048188 [0] (n=a3f8c271@gateway/web/cgi-irc/
20:20:40toffe82when you hold a dap , you use it with your thumb because of the position of the buttons, why not implement this on a touch screen
20:21:04 Quit j8048188 (Client Quit)
20:21:08 Join j8048188 [0] (n=a3f8c271@gateway/web/cgi-irc/
20:22:36Horschti use my nose
20:22:46Horschtor other pointy body parts
20:23:06LambdaCalculus37I use my toes. :)
20:23:36toffe82not joking but I can use my pda with one hand with no problem, so why they can't do it
20:23:48amiconnroolku: Btw, for the M3 help texts, you don't need to explicitly mention that the REMote buttons are used
20:24:01amiconnThis is implied for graphical plugins on M3, imho
20:24:27amiconnAT least that's how I did it for the other plugins
20:24:41roolkuamiconn: I am using to (less important) buttons on the main unit though
20:24:51roolkuto = two
20:25:09midgeyroolku: glad you took an interest in pegbox :)
20:25:36roolkumidgey: did know how much effort it would be :)
20:25:39amiconnPlugins should be converted to use the action api, for more flexibility without too much ifdefing
20:25:43Horschtspeaking of which, how hard would it be to change the button layout in pegbox?
20:25:58amiconnNote: I'M *not* talking about pluginlib actions
20:26:03 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
20:26:05midgeyroolku: i've been letting it rot on the tracker since no one seemed very interested
20:26:07roolkuamiconn: I would agree and ideally that would somehow linked to the manual generation
20:26:13 Quit j8048188 ("CGI:IRC (Ping timeout)")
20:26:26Horschtnetx/prev level on the scrollwheel is not good in my oppionion
20:26:55roolkumidgey: I always had it in my build and kind of forgot about it until I got new targets (mrobe100 c200) where it didn't work :)
20:26:56amiconnRight now, you need to ifdef the code whenever you want to assign multiple buttons to the same action, need to use _PRE buttons for short/long decisions etc
20:26:56Horschthave it on a button combo with select would be better (talking about ipod video, btw)
20:27:30amiconnWith actions, all this is just an array. Assigning the same event to multiple buttons for some targets won't require any ifdef in the code
20:27:37roolkuHorscht: I have changed that in the last commit (not sure if for the better)
20:27:58*amiconn wants all plugins on M3 to at least be exitable without the remote
20:28:09HorschtI am off, partying
20:28:11roolkumidgey: BTW: if you feel like making some nice graphics for the small colour targets (c200) as the are currently b/w
20:28:49midgeyroolku: i didn't make those graphics. they're from someone on the forums (or maybe MR) and i never got the psd for them
20:30:03midgeyi wonder if a comment should be added to say the game idea is from Detached Solutions
20:30:14roolkuHorscht: that is how it is done now
20:30:37 Join asdrubal [0] (n=dribbles@unaffiliated/asdrubal)
20:30:55asdrubalWhen I go to Files in rockbox...I can browse folders, but I don't see any files in them. c200
20:31:07asdrubalanyone know what might be wrong?
20:31:55linuxstbasdrubal: Have you changed the "file view" setting?
20:31:59 Join austrian_guest [0] (n=5079497a@gateway/web/cgi-irc/
20:32:08asdruballinuxstb, what's that?
20:32:18linuxstbThe setting that controls the file view...
20:32:22asdrubalwhere is it
20:32:27*BigBambi points at the manual
20:32:47asdrubaloh I see
20:33:02asdrubalit was set on SHow Files -> playlists instead of all
20:33:47asdrubalsorry to bother you
20:34:46roolkumidgey: there is one thing I don't quite understand (could be intention): when you start the plugin again, it remembers but the levels you have played before, but you can't resume - is this correct?
20:34:46 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
20:35:01roolkumidgey: i.e. resume only works until you exit the plugin?
20:35:55 Quit austrian_guest (Client Quit)
20:36:51midgeyroolku: i think that was intentional but i dont remember. i thought you could resume a saved game or play a level you had already completed
20:38:56roolkumidgey: I think it is sufficient the way it is, but couldn't understand why it saves two datafiles.
20:39:17linuxstbasdrubal: No problem.
20:39:28midgeyi'll look it over and see if i remember why i did it that way
20:39:58*roolku goes to close the flyspray task
20:40:04asdruballinuxstb, is there a way to make song switching faster?
20:40:09asdrubalsometimes there's like a 10 second delay
20:41:23BigBambiasdrubal: What version of rockbox do you have?
20:41:28asdrubaltoday's build
20:41:34 Join DavidS1 [0] (
20:41:38BigBambiThat doesn't mean much
20:41:44asdrubalfor sansa c200
20:41:47BigBambiThere are many updates a day usually
20:41:50BigBambiWhat version?
20:41:59BigBambiYou can find it in system −−> Rockbox info
20:42:41 Join mcuelenaere [0] (
20:42:59BigBambiheh, the latest :)
20:43:15BigBambiCoolio, just wanted to check
20:43:37BigBambiWhen does the slowness occur? i.e. how often roughly?
20:43:55asdruballike when I click forward multiple times or hit forward than backward
20:44:02asdrubaland then maybe forward again
20:44:44asdrubalquestion: INT: 862MB/957MB means that's how much is used or is free?
20:44:52BigBambiasdrubal: It is when you skip out of the end of buffered data, it needs to refill the memory from 'disk'
20:45:01ChristopherW862 is how much is free
20:45:25BigBambiIt is expected to be slower than skipping within buffered tracks, but 10 seconds sounds a bit long
20:45:30asdrubalBigBambi, shouldn't reading from flash be low latency?
20:45:44BigBambilatency isn't the only issue
20:45:59BigBambiIt also has to read the data from flash to memory
20:46:13BigBambiBut yes, 10 seconds sounds a bit slow
20:46:15asdrubalI guess microsd is only what, 10MB /sec?
20:46:39BigBambiI wonder if it is filling the entire buffer before starting the next track
20:46:44asdrubalBigBambi, well thanks for the tips.. I guess I'll go on my walk now. if I get problems I'll let ya know
20:46:52BigBambiI don't have my c240 with me, so I can't check
20:46:57amiconnBigBambi: It's not, or at least it should not
20:46:59 Quit ColinT (Remote closed the connection)
20:47:18 Join ColinT [0] (
20:47:24BigBambiamiconn: I know it should not, but 10 econds to skip to a non-buffered track sounds a bit slow
20:47:37asdrubalBigBambi, it's only after I do a bunch of forwards and backwards
20:47:41amiconnReading the microsd is certainly slower than 10MB/sec in rockbox
20:47:42asdrubalin combination
20:47:47asdrubalamiconn, why's that?
20:49:17gevaertsasdrubal: microsd is not very fast, and the rockbox driver is not optimal
20:49:44BigBambiasdrubal: What format music?
20:51:31BigBambiSo say a file of 5 MB, it shouldn't take 10 seconds to read that, no matter how slow the disk is is my point
20:54:45 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
20:56:29 Join austrian_guest [0] (n=5079497a@gateway/web/cgi-irc/
20:57:28austrian_guestWhat does "The sample buffer should be 32-bit aligned" mean? Unsigned int? Unsigned long?
20:57:49ChristopherWon ARM, an int is 32 bits
20:58:12amiconnIt means it should be 32 bit aligned. Use int32_t
20:58:29ChristopherWI'm guessing that longs are also 32 bits, but as amiconn said, use int32_t
20:58:44amiconn'int' and 'long' aren't the same width across architectures
20:59:10 Quit perrikwp (" ajax IRC Client")
20:59:41austrian_guestwhat does "aligned" mean?
20:59:42amiconnFor our current targets, both int and long are 32 bits, but (a) there once was an attempt to port to calmrisc, where 'int' is 16 bit, and (b) the assumption also breaks when building sims on e.g. amd64
21:00:21 Join mcuelenaere [0] (
21:00:48amiconn'long' is 64 bits on amd64
21:01:39ChristopherWI know about the C standards and porting issues; I just said on the ARM an int is 32 bits
21:02:28ChristopherWhow big is "long long" on am64?
21:02:34ChristopherW128 bits?
21:02:48 Quit shotofadds (Read error: 110 (Connection timed out))
21:02:49amiconnAfaik also 64 bits
21:02:50 Join csc` [0] (n=csc@archlinux/user/csc)
21:02:56 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
21:03:32ChristopherWthen how big is a short? 16 bits maybe?
21:03:40ChristopherWor is it 32 bits?
21:03:45amiconnshort is always 16 bits afaik
21:03:51amiconnint is 32 bits on amd64
21:04:37ChristopherWI guess it doesn't really matter how big the sizes are, as long as they meet the constraints given in the ANSI/ISO C standard
21:04:49amiconnThat's why amd is called an 'lp64' platform, as Longs and Pointers are 64 bits. There are also ilp64 platforms
21:06:58ChristopherWbut a C compiler writer can choose almost any size for each type, regardless of the platform (on am64, the compiler author could use 64-bit shorts, ints, longs, and long longs if he wanted)
21:09:14ChristopherWgranted, it's somewhat wasteful to do that
21:09:43gevaertsYou forgot char...
21:10:09amiconnAfaik there are platforms where a char isn't 8 bits...
21:10:22ChristopherWlike VAX, where it's 9 bits (or is it 18?)
21:10:42shotofaddsamiconn: (re: touchscreens) the D2 is the most pocket-friendly DAP I've ever used. The 3 buttons are used for vol up/down & menu, or skip forward/back & pause if hold is enabled.
21:10:56shotofaddsthat use of the hold buttons sounds horrific on paper, but actually works really well.
21:11:35shotofaddsI would like to see Rockbox implement the same
21:12:26 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
21:12:57***Saving seen data "./dancer.seen"
21:15:36 Join Hellz [0] (n=xaxaxas@
21:16:00 Quit Hellz (Client Quit)
21:17:29amiconnshotofadds: How would you select new tracks without looking at the display? Or adjust bass/treble?
21:19:30amiconnThat's something I can do with buttons (and then the voice ui makes it much easier)
21:19:51 Join mcuelenaere [0] (
21:20:07shotofaddsI'm intrigued how buttons left you choose a new track without looking at the display?
21:21:02amiconnWith the voice UI it's really easy. Without it, one needs to remember the button sequences to get through the menu(s) and the file structure
21:21:40amiconnThat's not at all easy, but reportedly possible (blind users using OFs have to do that)
21:22:48 Quit waldo (Remote closed the connection)
21:23:54shotofaddssure. that's already possible with the D2's current keymappings (in Rockbox, not possible with the OF).
21:24:06shotofaddsI guess having those few buttons is a lot better than a pure touchscreen...
21:27:08 Quit phi (Read error: 110 (Connection timed out))
21:27:44 Join DavidSG [0] (
21:35:34 Join kugel [0] (n=kugel@unaffiliated/kugel)
21:37:53 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
21:37:53 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
21:40:55 Quit DavidS1 (Read error: 110 (Connection timed out))
21:43:50 Join mcuelenaere [0] (
21:44:32 Join in-ferno [0] (
21:48:54 Join Llorea1 [0] (
21:49:05 Quit Llorean (Nick collision from services.)
21:49:07 Nick Llorea1 is now known as Llorean (
21:50:10 Join kushal_12_27_200 [0] (n=kushal@
21:52:44 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
21:53:37 Quit austrian_guest ("CGI:IRC (EOF)")
21:53:50 Join philoware [0] (n=canavar@
21:58:36 Join kuge1 [0] (
21:58:36 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
21:58:59 Part kuge1
22:00:06 Join austrian_guest [0] (n=5079497a@gateway/web/cgi-irc/
22:00:53austrian_guestDoes anyone know how I can print a 'long' value with debugf? If I write "%d" I get a warning...
22:01:06 Quit fehmicans (Read error: 110 (Connection timed out))
22:01:11Nico_Paustrian_guest: %ld
22:02:14austrian_guestI still get the warning:
22:02:36austrian_guest'warning: int format, long int arg'
22:03:16austrian_guestrb->debugf("%ld\n", indata[i]); where indata is an array of int32_t -> long
22:04:11 Join mcuelenaere [0] (
22:04:23Nico_Pmaybe you need to cast it to long then
22:05:42austrian_guestnope, no luck..
22:06:17Nico_PI'm surprised
22:06:38austrian_guestI'm out of my wits...
22:06:46Bagderso what's the warning with %d ?
22:06:53austrian_guest'warning: int format, long int arg'
22:07:04Bagderand with %ld?
22:07:20 Join mcflow [0] (
22:07:24austrian_guestthe same ???
22:07:44Bagdervery strange
22:08:06austrian_guestThat's what I thought..
22:08:12Nico_Paustrian_guest: are you positive you're editing the same file you're compiling?
22:08:32austrian_guestOf course I am
22:08:49austrian_guestThat was one of the first things I checked ;)
22:08:52amiconnAnd you use a lowercase L ?
22:11:24kugelWhatI's about the forum? I've read the server will change?
22:12:26 Join J-23 [0] (
22:13:50Lloreankugel: Nothing more than rumor at this point. The idea has merely been suggested.
22:14:04kugelah ok
22:16:12 Join wisp_sh [0] (
22:16:43markunLlorean: we probably shouldn't argue too much with the meizu guys. Let's just do the port, then they can either use it or not.
22:16:56kugelHeh, I didn't knwo that X5(L) belong to iPod series ;)
22:17:28kugelAccording to the table at . I'm gonna fix it
22:17:57Nico_Paustrian_guest: rb->debugf("%ld\n", (long)indata[i]); should work fine
22:18:02amiconnEh, that should have gone 1 column right...
22:18:20Lloreanmarkun: Just was pointing out some inaccuracies. There's also a thread there that claims Rockbox is Linux, but I lost it by the time I registered.
22:21:02markunLlorean: I corrected him in another thread already I think :)
22:21:02 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
22:22:30markunLlorean: it really pisses me off that you have to wait 45 seconds between searches
22:26:02LloreanAh, good.
22:27:32 Join mcuelenaere [0] (
22:32:16 Join seren [0] (n=4fd3fcb1@gateway/web/cgi-irc/
22:32:28serenlinuxstb: ping?
22:36:16 Quit in-ferno ("Quitte")
22:36:18serenlinuxstb: you placed a comment about how the buffer size is calculated. But could you explain _why_ we need a buffer of that size, i.e. exactly one native and two mono bitmaps. Why not two native or three mono?
22:38:03Lloreanseren: It's a total size. If you use no mono bitmaps, you can use more than one fullscreen worth of native ones.
22:38:53linuxstbseren: No - I can't remember what discussion went into deciding that formula.
22:41:36serenlinuxstb: I mean, once it's known that we need one native and two mono bitmaps the formula is clear. But how was that decided? Just a feeling? That this size shoud be enough for most cases?
22:41:52 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
22:42:09Lloreanseren: More or less the "it should be enough".
22:43:20 Join chronon [0] (
22:43:50serenLlorean: or even so: you should arrange it so that it's enough. Otherwise you lose :-)
22:43:53 Quit midgey ()
22:44:29 Join m0f0x [0] (
22:46:21 Part J-23
22:46:55gevaertsseren: if it's not enough, it's the WPS authors who lose ;-)
22:50:12jhMikeSamiconn: here (though not for long)
22:51:46pixelmaI guess not many people care but thought I could ask. The H300 port of the "Boxes" theme currently achieves a viewport like thing for the main box by overlaying a bitmap as some kind of frame (the center is transparent). I guess this doesn't even look nice when the line has to scroll (if it shows up at all currently). Any objections to making it a real viewport now (gets rid of the bitmap)?
22:52:03 Quit austrian_guest ("CGI:IRC")
22:52:07Lloreanpixelma: Sounds like a good idea to me.
22:53:18pixelmaI'd like to not make it too complicated and would only use 1 viewport which means that the peakmeter is also not the width of the screen (anly the viewport's) and the time and date info at the top would be a bit shifted to the left
22:53:52pixelmadoesn't hurt IMO
22:55:21 Join mcuelenaere [0] (
22:55:22pixelmaok, that could always be changed later if someone thinks it isn't appropriate I guess
22:55:55amiconnjhMikeS: How costly is a corelock?
22:57:42amiconnI need to introduce core locking in lcd-gray.c for certain functions, and I'm not sure whether it's better to let the other core *occasionally* wait a bit longer, or rather lock/unlock quite often
22:58:02jhMikeSamiconn: practically free
22:58:38amiconnOkay, so you think I could lock/unlock for each pixel line transferred?
22:58:44*Llorean thinks "practically free" is almost as dangerous a phrase as "trust me" or "it'll be easy"
22:58:55amiconn(in lcd_blit_mono() and lcd_blit_grey_phase(), to be precise)
22:58:59jhMikeSit's definitely more expensive if using peterson's rather than swp
22:59:13amiconnThis will only be used on 1st/2nd Gen, so PP5002
23:00:08amiconnI need to lock out the CPU from flipping the invert bit (due to backlight change) while the cop is doing grey blits
23:00:25jhMikeSI'd do some tests if using it for every line. The call version could perhaps become measureable.
23:01:20 Quit thegeek (Read error: 110 (Connection timed out))
23:01:28amiconnEvery line woul mean ~12300 lock/unlock pairs per second
23:01:40Nico_Pmay I request a quick explanation of what a core lock is?
23:02:31jhMikeSNico_P: to serialize processor core access in a block of code
23:02:46 Join appleipodder [0] (
23:03:26appleipodderForum isnt working
23:03:33BigBambiwe know
23:03:55Nico_PjhMikeS: accross several cores? when does a block of code interact with several cores?
23:04:21jhMikeSNico_P: google about spinlocks (I've munged some terminology :)
23:04:21 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
23:04:28amiconnjhMikeS: Another question: If I have some data that is just filled in once from the CPU, and only used by the COP afterwards, I don't need UNCACHED_ADDR(), correct?
23:04:51Nico_PjhMikeS: ok. I tried core lock in wikipedia but assumed it wasn't the right result :)
23:04:53jhMikeSNico_P: whenever COP/CPU both attempt to use a piece of code at the same time
23:04:53amiconnAll I need to do is flush the cache after the initial filling?
23:05:19jhMikeSamiconn: flush the writing core, invalidate the reading core
23:05:27Nico_PjhMikeS: ah, so it's mutual exclusion between cores instead of threads?
23:05:28BigBambiappleipodder: There isn't much we can do at the moment, just be a bit patient :)
23:06:10jhMikeSNico_P: it's a simpler lock for sections that complete quickly and does no blocking
23:06:16amiconnHmmm. How would I invalidate on the reading core if I have no thread running there?
23:06:53Nico_PjhMikeS: it uses busy waiting instead, is that it?
23:07:06jhMikeSNico_P: yes
23:07:46amiconnBtw, greyscale-on-cop speeds up doom by ~50%, and test_fps for greyscale by ~42%
23:07:49jhMikeSamiconn: that's always the rub - the core itself has to do it
23:08:38amiconnOnly very little load remains on CPU (probably due to RAM access by the COP) - measured 2% on PP5002, and 1% on PP5022
23:11:19jhMikeSI was working on actually making NOCACHE_*ATTR real sections - the .map files look ok but something is fighting me
23:11:52*amiconn wants to do that too
23:11:58jhMikeSbasically it works but there's some minor trouble I haven't tracked down yet
23:12:02amiconnThat is if you're not beating me :)
23:12:25amiconnI need it in order to put grey on COP in plugins which must not use IRAM
23:12:33 Join mcuelenaere [0] (
23:12:40amiconn(in order to keep music playing)
23:12:59***Saving seen data "./dancer.seen"
23:13:16amiconnRight now I can only do it if struct grey_info is in IRAM, i.e. in test_fps, doom, mpegplayer and zxbox
23:13:46amiconnGrey on COP will be a feature bit - so one can still run it on CPU if desired
23:14:34amiconnThe bit will be simply ignored on single-core targets, so no ifdefing at all in the plugin :)
23:14:34 Quit markun (Read error: 104 (Connection reset by peer))
23:15:04jhMikeSthe SPC codec was working like that, but mpegplayer fails to init and playback locks after some time (this might be due to having some trouble in the frequency scaling)
23:16:00amiconnBtw, is the SPC codec dynamically adjusting its precision, or is it just a coincidence that it is *just even* realtime on PP5002?
23:16:32 Join markun [50] (n=markun@rockbox/developer/markun)
23:16:58jhMikeSit works the same on all PP but the PP5002 uses more IRAM
23:18:56 Part mcflow
23:19:11jhMikeSthe boost wasn't much higher on my 3g than for H10 or e200 and I could play every single track on 3g
23:19:20amiconnHmm, maybe that was unclear....
23:20:32 Join thegeek [0] (
23:20:44amiconnI mean whether the SPC codec adjusts its output precision (emulation precision, whatever) dynamically to get the best possible quality using all available power
23:21:18amiconnThe tracks I tried produce ~92% boost ratio on my 2nd Gen
23:24:46 Quit seren ("CGI:IRC (EOF)")
23:25:56jhMikeSit's has no dynamic precision. 92% is about normal for the heaviest tracks.
23:25:57 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
23:26:11jhMikeSe200 does it them about 89%
23:26:30amiconnHmm. I wonder what happens if I enable sw tone controls...
23:27:06jhMikeSit won't affect it since most of the load is on COP
23:28:22jhMikeSthe codec thread spends a good amount of time waiting for it to produce at least one buffer full
23:31:07*jhMikeS has to go clean up and do some boring non-rb-related business...bbl
23:33:03 Join mcuelenaere [0] (
23:33:51*amiconn has a working corelock in lcd-gray.c :)
23:42:48BagderI noticed the fuze firmware is as big as the e200v2 ones
23:43:01Bagderwhile all other v2 firmwares are ~5MB
23:44:50amiconnjhMikeS: I now verified that swp(b) *is* broken on PP5002 as well
23:45:00Bagder0x500400 vs 0xF00000 actually
23:45:34LloreanBagder: The other v2s don't do video, do they?
23:45:46amiconnWhen switching to CORELOCK_SWAP, my lcd-gray corelock just acts as if it wouldn't exist...
23:45:51Bagderquite possibly not
23:46:03LloreanThough I wouldn't expect MJPEG alone to account for quite that much...
23:46:16Lloreanor whatever new video codec they may have
23:46:32Bagderno, but I think they've just gone to the next aligned size
23:46:38Bagderit compresses very good still
23:46:44Bagderindicating lots of "waste"
23:47:03 Quit desowin ()
23:47:04Bagderls -l fuze/
23:47:09Bagder-rw-rw-r−− 1 dast dast 15728640 May 31 2005 FUZEA.BIN
23:47:09Bagder-rw-r−−r−− 1 dast dast 7700624 Apr 4 23:15
23:47:54amiconnMulti-line paster ;)
23:48:02*Llorean gets out the pitchforks and torches
23:48:45Bagderin my world, 2 or 3 lines are fine to paste
23:49:03 Quit kushal_12_27_200 ("This computer has gone to sleep")
23:51:09 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
23:51:48 Join qwedsa_ [0] (
23:52:11 Quit jgarvey ("Leaving")
23:53:37 Join seren [0] (n=4fd3fcb1@gateway/web/cgi-irc/
23:54:05serenpixelma: what tool do you use to glue the images together (for strips)?
23:54:48Nico_Pseren: the montage tool from imagemagick IIRC
23:54:52pixelmafollowing linuxstb's advice to use imagemagick's montage
23:55:00Bagderanyway my v2 page is now updated with the fuze bin file and some other minor updates. now bedtime
23:56:35serenAha, nice to know.
23:57:21 Join mcuelenaere [0] (
23:58:36pixelmathat's why the last one still wasn't committed (placed the " wrong...) :\
23:58:52*seren goes the Bagder way.

Previous day | Next day