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

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

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

#rockbox log for 2006-11-03

00:01:08 Join daurn|laptop [0] (n=quae@
00:01:18Bagderrasher: I think I'm in favour of that patch
00:01:52Bagderwe need to make it easier for people to make voices
00:02:06preglowhow hard is it currently?
00:02:10rasherSounds good. Ideally it'd be using the regular path in the configure script and generate a Makefile at the same point, but I'm not sure how to add the needed stuff to the regular Makefile
00:03:02Bagderrasher: well, it could possibly just add a voice target unconditonally, and have that call the voice script that would do most magic
00:04:27Bagderbut now I'll go sleep
00:04:30rasher(that would enable language setting for one thing)
00:04:34rasherI'll give that a shot
00:05:30rasherpreglow: On windows you run a .vbs file and pray. On Linux you write your own script.
00:05:35rasherat least, that's how it is now
00:05:54Nimdaewell, i thought maybe the image was going out of range by a pixel so i shifted it left one pixel, still no show :(
00:06:07rasherMy patch adds it as a configure option along with manual and simulator (on Linux anyway)
00:08:31Nimdaei should probably build the uisimulator
00:08:57rasherSomehow I forgot that you cannot chose language in the configure script anymore, although that was removed ages ago.
00:09:06rasherNimdae: linux or windows?
00:09:26Nimdaei have both available to me ;)
00:09:37Nimdaebut doing my theming on linu
00:09:44rasherAh, then you probably won't need a prebuilt simulator.
00:13:00preglowhow hard would it be to write some gui .exe that does it? refuse to think it'd be very hard
00:13:19 Join scuba [0] (i=cb1c9fa7@gateway/web/cgi-irc/
00:13:47rasherVoice building? Probably not very
00:14:34rasherI guess having it as part of the build environment is not very useful for most users
00:17:34 Part tim66
00:17:42rasherBut would improve the chances of automatically built voicefiles
00:20:40 Quit qailer (Remote closed the connection)
00:24:14 Quit scuba ("CGI:IRC")
00:24:22 Join scuba [0] (i=cb1c9fa9@gateway/web/cgi-irc/
00:25:02preglowwhy a vbscript script anyway?
00:25:06preglowjust what happens to exist?
00:25:20markunrasher: and with lang v2 there can be different voice files for different targets
00:25:56rasherpreglow: yeah, just coincidence I think
00:26:59rashermarkun: Yeah, I
00:27:22rasherI'm adding the language question back into configure now, and using the general makefile
00:27:34rasherSo you'll just go ../tools/configure; make voice
00:28:09 Join JdGordon [0] (
00:29:05scubamarkun: can you propose to RogerBacon (as I cannot access #gigabeat from the CGI:IRC) that he purchase another S series from a large retailer with a good returns policy, keep the new unit and return his faulty unit?
00:29:48 Quit bluebrother ("Leaving")
00:30:55 Join [sellout] [0] (
00:33:21markunscuba: ok
00:33:48 Quit PaulJam_ (".")
00:35:21RogerBaconscuba :tanks for the solution
00:35:43 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
00:35:51RogerBaconthe retailer will maybe check the serial number ..
00:37:47 Join menosm__ [0] (
00:37:49rasherThink I'll try and adjust my script for windows voice building using that vbs script
00:38:03rasherFirst, install Cygwin.
00:38:12rasherNext, wait 800 hours for cygwin to install
00:40:37 Join webguest84 [0] (i=cfa2645c@gateway/web/cgi-irc/
00:41:49webguest84hi how do i revert back to apple firmware?
00:43:10rasherHang on
00:43:46rasherIt's in the manual, section 2.6
00:43:52rasherOr that.
00:45:00rasherI don't see uninstallation in the FAQ, actually
00:45:48webguest84yeah neither do i
00:46:12rasherAre you using an ipod Video, because I just picked a random Ipod
00:47:12Nimdaeok, got me a uisimulator now, got the same problem, so it looks like it's accurate between ipod and sim
00:47:16Nimdaenow to fix it :)
00:47:36Nimdaekinda cool that works like the ipod, including with sound
00:47:51rasherOnly thing missing is volume
00:48:06Nimdaewell, i guess i just need it to display
00:48:14Nimdaeonce ig et that far, i can try on ipod again
00:48:21rasherYou can change volume, but it doesn't affect the sound
00:48:23webguest84right, i don't want to uninstall it i just want to start the apple firmware
00:48:28webguest84but it's not working
00:48:28 Quit Kohlrabi (Read error: 60 (Operation timed out))
00:48:30rasherSo it's perfectly fine for testing
00:48:41webguest84i pressed play and shut it down
00:48:43Nimdaeah ok
00:48:50webguest84and then i held menu
00:48:54Nimdaethis image is 64 pixels wide
00:48:59webguest84but rockbox is still loading
00:49:19rasherwebguest84: Ah, there's a trick to the timing of that, someone with an Ipod should be able to explain.
00:49:36Nimdaewebguest84: as soon as you see the apple logo, with no light, press and hold that menu button...sometimes i have trouble with timing it
00:49:53oboholding menu right from the start is enough.... ie hold to turn on, and keep holding
00:50:03obobut you need to keep you finger still
00:50:10Nimdaei've run into odd issues doing that
00:50:19Nimdaelemme try it real quick
00:50:30Nimdaeyeah, works
00:50:36webguest84thanks now it works
00:50:48webguest84just one more thing
00:50:50 Quit menosm (Read error: 110 (Connection timed out))
00:50:59webguest84i don't see rockboy in the plugins list
00:51:03 Join Starrr [0] (
00:51:09webguest84how do i install it?
00:51:20rasherThat's because it's not a regular plugin, but a viewer. You simply "play" a .rb file to play that rom.
00:51:28webguest84oh ok
00:51:43StarrrQuick noobish question −− is Rockbox getting developed for iPod nano second gen?
00:51:44webguest84thanks guys
00:51:50webguest84really appreciate it
00:51:52 Quit webguest84 ("CGI:IRC")
00:52:06NimdaeStarrr: the hardware is new and being figured out
00:52:14rasherStarrr: Not as far as I know. I think at the moment the hope is that IPL will do the hard work
00:52:34StarrrI was thinking of getting one of the new nanos, but I hate iTunes ..
00:52:36StarrrAh well.
00:52:52NimdaeStarrr: you can get refurb 1g nanos at for a good price
00:52:57 Join webguest04 [0] (i=864c0302@gateway/web/cgi-irc/
00:52:59RogerBaconitunes is a piece of shit !
00:53:03StarrrYeah, I was looking at those.
00:53:06hcsStarrr: gtkpod, and a bunch of others, work fine
00:53:16StarrrAh, didn't know about those.
00:53:22Nimdaei'm finally independant from itunes
00:53:26***Saving seen data "./dancer.seen"
00:53:33StarrrI only stopped using iTunes a few months ago.
00:53:36Nimdaei was only using it for podcasts until i got juice :)
00:53:38StarrrFoobar2K <3
00:54:06JdGordoncan anyone remeber the name of the good linux mp3 tagger/renamer?
00:54:31 Join mechanori [0] (
00:54:33 Join einhirn [0] (
00:54:34 Quit einhirn (Client Quit)
00:54:43rasherEx Falso decent, but it does boss you around a bit
00:55:05 Join Paul_the_Nerd [0] (
00:55:07JdGordoneasytag :)
00:55:07Nimdaerasher: can you give me some hints as to why the vol images won't show so i know where to look?
00:55:09StarrrMeant to ask though, what is IPL?
00:55:34rasherStarrr: Ipodlinux
00:55:40webguest04Nimdae: have you tried "%xl|z|vol10.bmp|255|13|"? afaik the upper left corner is 0x0, so your bmp goes over the right side.
00:55:47Nimdaewebguest04: yeah
00:55:54StarrrThanks a lot. :)
00:56:10rasherNimdae: I don't really know much about the wps syntax to be honest
00:56:10TheShadowxIntelectual Property License? O_o?
00:56:34Nimdaewebguest04: yeah, it starts at 0x0, i have a bar starting at 0x13 and goes all the way accross to 319
00:57:13Nimdaeif it starts at 256 and is 64 pixels wide, it would go to 319
00:57:20 Part waltman
00:57:23 Join Soap [0] (n=Soap@unaffiliated/s0ap)
00:58:09webguest04Nimdae: oh you're right. my fault.
00:58:24Nimdaebut i did try bumping it over by 1 pixel
00:58:38Nimdaelet me check the image next to it
01:00:03webguest04Nimdar: does your wps use lots of graphics? there is limited memory to store the bitmaps (i think 2x screensize). does the image show when you use only %xdz ?
01:01:18Nimdaei'm thinking i'm hitting a size limit
01:01:37Nimdaei moved the images to a totally different location so it's not a range issue
01:01:42Nimdaei'm gonna try trimming up the images
01:03:28Nimdaefound an image i'm loading but not using that's not exactly small, heh
01:04:09Nimdaehad a stop image...there's no stop on the ipod
01:05:21webguest04the conditional for stop is shown on no target.
01:09:37 Quit mechanori (Read error: 104 (Connection reset by peer))
01:09:56 Join mechanori [0] (
01:10:46 Quit menosm__ (Read error: 104 (Connection reset by peer))
01:11:37 Join menosm__ [0] (
01:11:39 Nick menosm__ is now known as menosm (
01:13:29Nimdaeit shows
01:13:39Nimdaenow i gotta readjust the positioning of some images, heh
01:14:13 Quit mechanori (Read error: 104 (Connection reset by peer))
01:14:44 Join mechanori [0] (
01:16:45Nimdaeheh, if the uisimulator worked a bit better, i'd consider using it as my media player :P
01:16:56 Join coldsteal [0] (n=Owner@
01:16:59Paul_the_NerdA few people have said that. :-P
01:17:31coldsteali am trying to figure out what rockbox can all do before i install it
01:17:40Nimdaeplay music
01:17:54Paul_the_Nerdcoldsteal: Read the WhyRockbox page at
01:18:00coldstealwill it work with linux?
01:18:30Paul_the_NerdThat question doesn't really make sense...
01:18:36rasherAll current rockbox players are just USB Mass Storage devices
01:18:47hcsit'll work with anything with fat32 support...
01:19:00rasher(except one of the old Archos Jukeboxes which is Almost, But Not Quite UMS)
01:19:01hcs(as USB Mass Storage, of course)
01:19:38coldstealwell im asking how is the updaiting done
01:19:44rasherJust unzip a file
01:19:46*Nimdae installs his theme on his ipod to test
01:19:59rasherOn the drive. That's it
01:20:19coldstealokay well i have an ipod and right now im using it with amoroock
01:20:36rasherOh, you meant "updating" as in copying files to and from?
01:20:38coldsteali meant the music
01:20:45coldstealhow do i add music
01:20:57Nimdaelike copying files to any other usb drive
01:20:57rasherCopy it to the drive's harddisk
01:21:05rasherLike you would on any usb drive, yes
01:21:16coldstealbut in an ipod the music is all wierd
01:21:22rasherNot with Rockbox
01:21:27rasher(unless you insist)
01:21:30Nimdaerockbox != ipod
01:21:44 Quit petur (Read error: 110 (Connection timed out))
01:21:48Nimdaetrust me, i have an ipod 5g running rockbox
01:22:00rasherThat's something the Apple software imposes. When using Rockbox you can place files however you like
01:22:13coldstealso all i do is get the ziped file and extract it into thhe ipod
01:22:25coldstealwill it dule boot?
01:22:37rasherNot quite, the first install is a bit more involved. It's all described in the manual.
01:22:44rasherYes, it'll dual boot just fine
01:22:44 Join roolku [0] (i=522906c5@gateway/web/cgi-irc/
01:23:14rasherBut of course, if you boot the Apple firmware, you'll only see songs placed in that strange directory layout
01:23:18Nimdaesweet, my whole theme works now
01:23:35rasherNimdae: what was the problem? Size limit?
01:24:22Nimdaethere was a change i had planned making later to reduce the size, didn't think the bloat would affect it like this
01:24:39Nimdaething is, it doesn't look really graphic intense
01:24:45Nimdaeit just happens to be so, i guess
01:25:09Nimdaeis there a way to make a screenshot on the ipod?
01:25:12Nimdaeor rather rockbox
01:25:53coldstealdo i need to wipe my ipod b4 i do this/
01:26:03webguest04turn on screendump in teh debug menu and insert the usb cable to amke a screenshot
01:26:06Nimdaeyou don't need to, but it'd be a good idea to make a backup first
01:26:16 Join Davide-NYC [0] (
01:26:29coldsteali have like allot of files on it
01:26:29roolkuslasheri / pondlife: a recent change seems to have buggered (technical term) the runtime data display in the wps. It seems to be an issue with track_buffer_callback - it gets set correctly in tagtree_init, but when I check the value just before its execution it is somehow reset to 0 (and is not called as a result)
01:26:31Nimdaewebguest04: does this work in linux, windows, both?
01:26:37coldstealnot musc files
01:26:38 Quit Davide-NYC (Client Quit)
01:27:17 Quit leftright (" Try HydraIRC -> <-")
01:27:44rasherNimdae: In the sim, you can screendump with F5
01:27:50roolkuslasheri / pondlife: at least that is what I gathered from my debugf output in the sim - I could be wrong about the cause
01:27:51rasher(I think it is, one of the F buttons anyway)
01:28:23Nimdaeoh sweet, it detects the cable then triggers dump
01:28:42rasherYup, it's quite cunning
01:28:43Nimdaethat's one way to utilize extra functions without fudging with buttons
01:30:32coldstealcan someone pm me?
01:31:11 Quit obo ("bye")
01:32:34 Quit Paul_the_Nerd ("Leaving.")
01:32:45 Quit roolku ("CGI:IRC")
01:33:33Nimdaeneeds more work
01:33:38 Quit Criamos ("( :: NoNameScript 4.03 :: )")
01:33:42 Join MarcoPolo [0] (
01:34:04JdGordonNimdae: that looks cool
01:34:14coldstealNimdae: can u help me?
01:34:22Nimdaewhat with?
01:35:09rasherThat's not very specific
01:35:24coldstealwell getting started
01:35:42NimdaeJdGordon: i'm thinking of removing the playlist and time info on the left, and centering the track info on the bottom
01:36:03Nimdaecoldsteal: what i did to get started was do some reading ;)
01:36:20JdGordonNimdae: when its fiished, can you do the same theme for the h300?
01:36:37Nimdaei could try
01:36:41Nimdaebut i'd have to do it on the simulator
01:37:13Nimdaeany suggestions on how to improve it?
01:42:36 Quit linuxstb_ ("CGI:IRC")
01:46:29 Quit scuba ("CGI:IRC")
01:49:06 Quit MarcoPolo (Remote closed the connection)
01:53:42 Join Paul_the_Nerd [0] (
01:59:50rashercoldsteal: Just read the manual, it's fairly straight-forward if you follow it step by step.
02:00:06 Nick rasher is now known as rasher_ (
02:00:20 Nick rasher_ is now known as rasher (
02:00:31coldstealcan i keep the stuff i have on my ipod?
02:00:57rasherIf it's formatted as fat32, yes
02:01:08rasherUsual disclaimers apply.
02:01:17coldstealdoes it overwrite appls stuff
02:03:59 Join lostnihilist [0] (
02:05:30 Quit Nimdae (Read error: 110 (Connection timed out))
02:14:57 Join lost|X40 [0] (
02:17:44 Join mechanor [0] (
02:17:56 Quit mechanori (Read error: 104 (Connection reset by peer))
02:21:16 Quit mechanor (Read error: 104 (Connection reset by peer))
02:21:42 Join mechanori [0] (
02:23:29 Quit lostnihilist (Read error: 60 (Operation timed out))
02:25:16 Join NickDe [0] (
02:28:05 Join |AhIoRoS| [0] (n=ahioros@
02:36:04 Join AdamKili [0] (
02:37:24AdamKiliis there any way to get other MAME games working through PacBox besides pacman?
02:37:40 Part coldsteal
02:41:12 Quit webguest04 ("CGI:IRC (EOF)")
02:43:44hcsAdamKili: Iit has nothing to do with MAME, it specifically emulates the Pacman arcade hardware. Possibly other games using the same hardware could be made to work.
02:44:50AdamKilihow can i find out if a game is compatible?
02:45:08 Quit RogerBacon (Remote closed the connection)
02:45:09AdamKiliand does anyone have firsthand experience getting a different game to work?
02:45:26hcsI don't think there are any compatible at the moment, it'd require modification to the pacbox.
02:48:15AdamKilithen why don't they work on modify it? then we could have access to a large MAME library of games. I'd help or do it myself if i knew how to do programming
02:48:43hcsIt probably wouldn't be very many games.
02:49:07Paul_the_NerdIt might be expandable to say, Ms. Pacman.
02:50:08AdamKiliit looks like has a big library of MAME games :)
02:50:13*Paul_the_Nerd sighs
02:50:18Paul_the_NerdAre you reading what's being said?
02:50:20 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
02:50:23Paul_the_NerdIt's NOT a MAME.
02:50:30Paul_the_NerdIt's a pacman hardware emulator.
02:50:31 Join aliask [0] (
02:50:43StarrrTo bring this up again; what are some other reccomended replacement firmware things, outside of rockbox?
02:50:52Paul_the_NerdMAME also emulates the pacman hardware, which is why it uses the same ROM files.
02:51:26JdGordonaliask: howd methods go?
02:51:30Paul_the_NerdStarrr: Google would serve you better. In general, there just aren't really many/any depending on what hardware you're talking about.
02:51:42aliaskJdGordon: Aced it. I think I dropped 1 mark.
02:51:48JdGordonwell done
02:51:52StarrrI was looking for something for a 2nd gen nano. :/
02:52:08aliaskBut the hard one is on monday. That will be the real test
02:52:14JdGordonStarrr: then your waiting for us and iPL
02:52:25Paul_the_NerdStarrr: Seriously, if someone breaks the encryption on an open source project, it's gonna show up in almost all of our projects soon after anyway
02:52:42 Join Nimdae [0] (
02:52:42AdamKiliok i get it. I just thought PacBox was a MAME emulator limited to pacman hardware. not visa versa
02:52:49Nimdaestupid internet provider
02:53:01Nimdaeso yeah, any suggestions on how to improve my wps?
02:53:19Paul_the_NerdAdamKili: MAME stands for Multiple Arcade Machine Emulator. If it were limited to pacman hardware, it wouldn't be multiple at all.
02:53:29***Saving seen data "./dancer.seen"
02:53:30Paul_the_NerdMAME is not released under a GPL compatible license, which is why we *cannot* use it.
02:54:13AdamKiliya it says that in the PacBox details
02:54:25 Quit mechanori (Read error: 145 (Connection timed out))
02:55:13aliaskCan anyone tell me a free ARM disassembler?
02:55:20Paul_the_Nerdaliask: objdump?
02:55:33aliaskDoesn't that only work with .o files?
02:56:30hcsit can work with raw binaries, too
02:56:44aliaskOh, great!
02:56:57*aliask needs to learn how to use it
02:56:58*hcs hopes he's right there...
02:57:08hcsI seem to remember doing so
02:58:48 Quit Arathis ("[rl_bot quit]")
02:59:09Paul_the_Nerdhcs: It's what seems to be recommended as the 'free' choice around here.
03:02:36 Quit Nimdae (Remote closed the connection)
03:03:45 Quit netmasta10bt (Read error: 60 (Operation timed out))
03:08:08 Join Nimdae [0] (
03:09:21 Join TheShadow [0] (n=bobdole@
03:10:04 Quit hardeep ("[BX] Mike Tyson says BitchX BITES! Do you HEAR what I'm saying?!")
03:10:37aliaskHeh, I finally found the answer on bagder's sansa bootloader page.
03:12:26 Quit AdamKili ("Ex-Chat")
03:16:04 Join JoeBlu [0] (
03:17:31 Join AdamKili [0] (
03:18:29AdamKilii just remembered Doom! can the Doom plugin play other games that end in .wad?
03:18:53 Quit Nimdae (Read error: 104 (Connection reset by peer))
03:19:09SoapThere is a whole wiki on doom.
03:19:54Soapsome of the addons for the original doom work, I believe the wiki mentions which ones are confirmed to work.
03:20:03 Quit TCK (Client Quit)
03:21:27AdamKiliso basically, you can only modify Doom, not play quite different games such as Super Mario Bros?
03:21:28 Join Nimdae [0] (
03:22:08scorche<scorche> i=linus@rockbox/developer/LinusN
03:22:08scorche<scorche> so instead of −−> JoeBlu ( has joined #rockbox
03:22:08scorche<scorche> it would be −−> JoeBlu (n=joeblu@rockbox/peon/JoeBlu) has joined #rockbox
03:22:08DBUGEnqueued KICK scorche
03:22:08scorche<JoeBlu> so what do i need to do to get it
03:22:10scorche<scorche> sorry, but you dont quite qualify
03:22:11scorche<scorche> unless for some odd reason i dont know you...
03:22:16scorche<JoeBlu> i don't think you know me... very new to rockbox
03:22:17scorche<scorche> then i will go back to the sorry statement =)
03:22:26CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
03:22:26*scorche curese x-chat
03:22:32Soap.wad was a generic extension by ID. It was nothing more than a renamed .zip with resources, so many things other than doom used WADs.
03:22:56Paul_the_NerdSoap: But most of them use the doom engine or a variant
03:22:59Paul_the_NerdSuch as ChexQuest
03:23:19Soapand most of those variants aren't supported by PrBoom, are they?
03:23:22Paul_the_NerdAdamKili: The Rockdoom plugin is only compatible with wads that run in PRBoom, essentially ones that will run with the original official Doom or Doom2 executables.
03:23:25JoeBluoh man! ChexQuest was such a ridiculous game
03:23:35Paul_the_NerdSoap: Chexquest is perhaps the only one that might be. ;)
03:25:36 Join scuba [0] (i=cb1c9fa7@gateway/web/cgi-irc/
03:25:38AdamKiliessentially, i want to know if there is any effort to create a NES, SNES, Atari, etc. emulator on RockBox? and where can i see if Chexquest will work on Rockbox?
03:25:58SoapI have't tried my x-men total conversion yet.
03:26:53 Part scuba
03:26:58Paul_the_NerdSoap: TCs that work with the official Doom or Doom2 executables *will* work. Or it's a bug we need to fix
03:27:26Paul_the_NerdAdamKili: See the patch tracker. And you can see if chexquest will work by trying it, seriously that's not such a confusing topic.
03:28:44SoapThe only addon wad a real man needs is the Evil Dead one. It's groovy.
03:29:04Paul_the_NerdActually, before Fox destroyed it, the AliensTC was REALLY good
03:29:14Paul_the_NerdBut it involved some executable hacking by way of a dehacked patch.
03:29:27Paul_the_NerdIt had eggs that sprouted facehuggers and everything, rebalanced weapons, etc.
03:29:43Soapnever saw that one, cool.
03:31:29Paul_the_NerdIt's hard to find these days
03:31:36Paul_the_NerdI have three 3 1/2 floppies with it *somewhere*
03:31:42 Quit Nimdae (Read error: 104 (Connection reset by peer))
03:34:34 Join Nimdae [0] (
03:34:59 Join RogerBacon [0] (
03:38:22 Part pixelma
03:38:25AdamKiliit works if anyone wants to know
03:42:17*Paul_the_Nerd figured it would
03:48:29 Join Nimdae_ [0] (
03:55:00 Quit Nimdae (Read error: 110 (Connection timed out))
03:55:45AdamKilihey on Doom, why isn't sidestepping left/right happen when scrolling on the clickwheel clockwise/counterclockwise.
03:55:51 Quit Nimdae_ (Remote closed the connection)
03:56:13JoeBluisn't sidestep when you click left/right?
03:56:57AdamKilioh ya and have turning left/right happen when scrolling. much better
03:58:02 Quit blue_lizard (Read error: 60 (Operation timed out))
03:58:29 Join blue_lizard [0] (
03:59:05 Quit NickDe ("Leaving")
04:04:58 Join JoeBorn [0] (n=jborn@
04:06:42 Join GreyFoux [0] (
04:13:41 Quit AdamKili ("Ex-Chat")
04:23:08 Join Nimdae_ [0] (
04:29:52 Quit RogerBacon ("Quitte")
04:33:30 Join GFoux [0] (
04:35:14 Quit GFoux (Client Quit)
04:38:43 Join actionshrimp [0] (
04:44:02 Quit Nimdae_ (Read error: 110 (Connection timed out))
04:51:30 Quit GreyFoux (Read error: 113 (No route to host))
04:53:30***Saving seen data "./dancer.seen"
04:56:10 Quit |AhIoRoS| (Read error: 110 (Connection timed out))
04:56:32 Quit sucka (Read error: 110 (Connection timed out))
05:31:36JdGordonmy sisters nano turned itself on into some diagnostic test? anyidea who to get there manually?
05:32:09Paul_the_NerdHold Left or Right while booting, if I recall
05:32:22JdGordonah ok
05:32:39Paul_the_NerdOkay, so left and select
05:33:38JdGordonyou learn something new every day :p
05:34:08 Join pacino [0] (
05:36:13pacinohi guys - what's the cheapest player that can run rockbox?
05:37:48Paul_the_NerdOndio SP maybe? Since a decent portion of them are discontinued, it depends on how much they're being sold for.
05:37:53Paul_the_NerdMaybe a 1 gig Nano 1st generation.
05:38:32 Quit JoeBlu ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
05:40:32pacinocool - thanks mate
05:40:54 Quit newbyx86 (Read error: 60 (Operation timed out))
05:42:14 Join |AhIoRoS| [0] (n=ahioros@
05:44:54 Quit mirak (Read error: 60 (Operation timed out))
05:51:35 Join newbyx86 [0] (
05:53:40 Join Davide-NYC [0] (
05:57:50 Quit funky ("leaving")
06:04:05 Quit pacino ()
06:04:07 Join Nimdae [0] (
06:04:24Nimdaei hate charter
06:08:41Davide-NYCany theories on that hfs+ business in the forum?
06:10:46Davide-NYCquestion: if I include the test_fps plugin in apps/plugins/SOURCES and make for h1x0 simulator I get errors.
06:10:51Davide-NYCIs this normal?
06:11:26Davide-NYCis the FPS test only for a certain types of display targets?
06:14:18 Join midgey34 [0] (
06:14:35Davide-NYCKnock, knock my rockbox brethren!
06:15:00 Join bobberty [0] (
06:15:34Paul_the_NerdDavide-NYC: There's no reason why it wouldn't be real, at the same time, I've seen absolutely no proof that it is. :)
06:15:56Davide-NYCI learned somethign watching the video
06:16:17Davide-NYCI learned that Wierd Al Yankovic is still making parody music
06:16:18Paul_the_NerdWhat'd you learn?
06:16:22Paul_the_NerdYes, he is.
06:16:32Paul_the_NerdI like the video claiming it was thought to be impossible
06:16:44Davide-NYCI messed that bit
06:16:55Davide-NYCopening credits?
06:16:58Paul_the_NerdSome of the text near the beginning, yeah
06:17:29Davide-NYCany idea about the test_fps with the simulators?
06:18:28Paul_the_NerdI don't think it's usable at all for sims.
06:18:38Paul_the_NerdBut i really don't know
06:18:53Davide-NYCwillit work for the player (real build)?
06:19:02Davide-NYCcharcell lcd
06:19:06Davide-NYC(I think)
06:22:09Paul_the_NerdI doubt it.
06:22:13Paul_the_NerdBut I really don't know
06:22:19Davide-NYCI ifdefed it out
06:22:46Davide-NYC!= sim and then HAVE_LCD_BITMAP inside
06:22:56Davide-NYC3 lines
06:23:02*Davide-NYC still learning
06:24:59bobbertyhey i'm looking for an old daily build for a h120...before mid august 2006. anyone know where i could get that? thanks.
06:26:16 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
06:27:02Paul_the_Nerdbobberty: What do you need an old build for?
06:27:28 Quit linuxstb__ (Read error: 110 (Connection timed out))
06:28:36bobbertystable recording. petur posted on the forum that recording is not stable. he suggested a build from before august.
06:30:13Davide-NYCI have not done long recordings as he has, but I use my h120 daily to record "conversations" and I have not noticed the dropouts he described
06:30:31Davide-NYCthese recordings are between 3 miuntes and 30minutes generally
06:30:45Davide-NYCbetween 3 and 10 of them daily
06:31:22Davide-NYCthis is not a scientific or conclusive test by any means, but I say that recording is quite stable
06:33:15 Quit |AhIoRoS| ("Abandonando, see you")
06:33:34bobbertyi just bought my h120 so the oldest build i managed to get was 09-30-2006. would you recommend that, or just go with the latest?
06:33:56Davide-NYCJust go with the latest.
06:35:00Davide-NYCBe aware that soon there will be a major update to the recording code. *At that point* I would wait to update rockbox until it had been tested for a while (like a month)
06:35:26bobbertyok...i'm taping friday and saturday night. saturday is the band i'm worried about losing. so i'll see how it goes, tonight.
06:36:05Davide-NYCIf you're really worried run a test before tonight.
06:36:43Davide-NYCThat way you can enjoy the show instead of worrying about the recording
06:36:55Davide-NYCPlease tell me about your mics etc...
06:37:14bobbertytrue...thanks for the help and advice!
06:37:45bobbertyi run with audio technica 853s (cards)...phantom power.
06:38:31bobbertydeneke ad20
06:38:42Davide-NYCthat's the pre I'm gonna get
06:38:52bobbertyit's a tank
06:39:20bobbertyi've been using it for a while...i've been happy with it
06:39:34Davide-NYCI'm gonna hit up the tradersdem for the AD-20 plus the stereo phantom power module.
06:39:39Davide-NYCsee what I can find
06:39:51bobbertyi bought mine from a shop in nyc
06:39:54Davide-NYCcan the AD-20 output 16bit?
06:39:59Davide-NYCYou in NYC?
06:40:21Davide-NYCMastodon played Conan, you saw?
06:40:37bobbertyno tv tonight
06:40:44Davide-NYCcouple nights ago.
06:40:50Davide-NYCAnyway this is way off topic.
06:41:14Davide-NYCGood luck and have fun.
06:41:22*JdGordon slaps Davide-NYC around with a large hammer
06:41:42bobbertymy first night out with the h120+rockbox so thanks!
06:41:46JdGordonyou know how u wanted someone else to check the bug with moving rockbox.iriver into .rockbox?
06:41:47*Davide-NYC suspect JdGordon likes Mastodon too
06:42:06JdGordonand i did... well it came time to bugfixing my custom build and i forgot it was there...
06:42:12JdGordonso i thought the bugs were all gone :'(
06:42:31Davide-NYCcrap! sorry
06:42:39*Davide-NYC disclaims all responsibility
06:42:57Paul_the_NerdWhat bug is this?
06:43:27Davide-NYCvoice files break if the firmware image is in the .rockbox folder
06:43:42JdGordonbut it doesnt...
06:43:52Davide-NYCeverything else (AFAIK) works
06:44:00Davide-NYCJdGordon: ?
06:44:14Davide-NYCit does not break?
06:44:20JdGordondidnt u decide it was your patched build?
06:44:24JdGordonyeah, it worked fine here
06:44:36Davide-NYCOh man. You're right. I apologize
06:44:52Davide-NYCthat's why I didn't submit a bug report
06:45:06Davide-NYCOK, obviously time for bed
06:45:24 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
06:47:30midkaysomething that just kinda hit me that i'm near-certain we don't implement (i'd be very "wtf?" if we did) but i expect has been suggested before - whenever the user initiates a spinup and audio is playing, check the buffer level, and if it's under like 20% full then totally refill it so you don't have to spin-up again just a moment after the disk stops..
06:48:23JdGordonhaha, thats ingenious :D do we really not do that?
06:48:32midkaydoesn't sound too hard to do from a coding standpoint (though i couldn't, i don't know enough about playback) - a simple check, then either top off the buffer or don't..
06:48:46midkayi can pretty safely say that we don't, i've heard of/experienced/noticed it.
06:48:49JdGordonalthough it should be <70% or something
06:49:44Paul_the_NerdWhy not just, if the spinup was not called explicitly for the purpose of rebuffering, fill up the buffer immediately before a spindown happens.
06:50:07midkaythat's a good way to look at it, i thought after JdGordon said that - even if it's like 90% it'll only take half a second to refill..
06:50:07Paul_the_NerdReally, at 90%, a rebuffer will take so little time anyway, it's still not gonna be a bad thing, right?
06:50:36JdGordonyou have to be careful tho, it is possible, without good sanity checks, that refilling could never stop
06:50:51midkaybut that also means you'll clear it of the last played song or segment of the current song, so it's not so good for skippers.. maybe if an entire previous song is in the buffer, clear it and top that off (e.g. only clear completed songs) so you can still rewind in the current one and restart it or something.
06:51:33JdGordoni tihnk playback is pondlife or preglows domain...
06:52:32midkaymight be worth asking about/discussing, sounds nice (i was thinking about our delayed write system and suddenly wondered why i haven't ever thought of or heard about having the buffer refill itself on a disk spinup).
06:53:34***Saving seen data "./dancer.seen"
06:55:16 Part bobberty ("Leaving")
07:00:41 Quit midgey34 (Read error: 110 (Connection timed out))
07:04:51Nimdaeare there things i can change around other than just the wps?
07:05:29JdGordonNimdae: you can change absolutly everyting
07:06:09Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
07:07:16JdGordonmidkay: the problem with your idea tho is that i dont tinhk there is currently any way for the hdd driver to let the rest of rockbox know its been spun up
07:07:37midkaywhat's up with delayed writes then?
07:08:06JdGordondelayed writes is only for the config sector isnt it/
07:08:59midkayit's also used for logging and tagcache "gather runtime data" at least, as far as i know. either way - we have it and use it..
07:09:17JdGordoni have started something along these lines for a proper callback from ata in the menu recoding patch... but its doesnt work very well, maybe i should split that off and get it working seperaty
07:09:19midkayi'm not sure, though..
07:10:10midkaythat'd be cool..
07:22:39Nimdaelooks like some of the things i want to change require a patch, heh
07:25:21Nimdaewow, the patches needed for the jClix theme on ipod 5g look like something that should be considered, the level of customizability would be awesome
07:26:44 Quit [sellout] (Remote closed the connection)
07:29:36Paul_the_NerdNimdae: In the past I know some of them couldn't be disabled. As well, how many of them have actually been posted to the tracker?
07:30:22Nimdaei dunno, i don't really know what they all do other than what's presented
07:30:58 Quit aliask ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
07:31:00Nimdaejust like the ideas of variable font sizes, more color control, and some customization of the menus
07:31:46Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
07:32:56Nimdaeapparently the patches were designed to extend the current capabilities, not replace them
07:33:44Nimdaebah, who would have thought a simple image would cause confusion...had to change something in my theme because it made me second guess the interface, heh
07:37:47 Join dan_a_ [0] (i=dan_a@
07:39:44 Quit dan_a (Read error: 110 (Connection timed out))
07:41:09 Nick dan_a_ is now known as dan_a (i=dan_a@
07:42:02 Quit linuxstb_ (Read error: 110 (Connection timed out))
07:42:17 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
07:43:25Nimdaei dub my theme finished...and not really all that great :P
07:44:04Paul_the_NerdI shall proceed to laugh at it now, before I see it, so that I can get that part of the job done with maximum efficiency. :-P
07:44:49Nimdaelemme pull a screendump
07:48:15Paul_the_NerdActually, I kinda like it. Very simple. Not the sort I use, but still nice.
07:48:25Nimdaei like simple, heh
07:48:29Paul_the_NerdThe "Play" triangle is a little left of where I personally would've put it. But :-P
07:48:40Nimdaeit is exactly centered ;)
07:48:42Nimdaetrust me
07:50:35Nimdaeoh crap
07:50:40Nimdaei forgot to make a charge icon
07:51:18 Join bluebrother [0] (
07:51:32Nimdaei think i could probably improve the positioning of the status icons at the top
07:51:50Nimdaemaybe a bit closer together
07:51:52*Nimdae shrugs
07:52:08Nimdaei'm happy with it for it being my first theme for ANYTHING :P
07:55:48 Join nave7693 [0] (
07:56:02 Part nave7693
07:59:01 Quit pagefault ("Leaving")
07:59:11 Join MarcoPolo [0] (
08:01:03 Join sucka [0] (
08:04:46Paul_the_NerdNimdae: Exactly centered doesn't necessarily mean it looks best. I'd have actually tried to position it so the three points are equidistant from the circle, if I were going all mathematical. ;)
08:06:25daurn|laptopequidistant <−− thats a word that makes you feel all big inside, doesn't it? ;)
08:12:42 Join Rob2222_ [0] (
08:12:42 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:12:43midkayPaul_the_Nerd: i'm pretty sure centered means they are all equidistant from the circle - it's just that since it's much wider on one side and literally one pixel wide at the far end it looks off to the side...
08:14:27 Join linuxstb__ [0] (
08:15:43 Quit linuxstb_ (Read error: 110 (Connection timed out))
08:16:31Paul_the_Nerdmidkay: It's not an equilateral triangle, so centered could also mean that the rectangle the circle is inside of (with all three tips touching the rectangle) has a center point at the center of the circle, which is what it looks like to me.
08:17:18midkayyou lost me.
08:17:19Paul_the_NerdDoesn't really matter though, not my theme. :)
08:17:22 Join dickon [0] (
08:17:40midkayand i read that like 4 times.
08:17:55midkayyou officially suck at being understandable :)
08:18:21 Quit actionshrimp (Read error: 110 (Connection timed out))
08:19:06dickondumb new user question: how do I get from the file browser to the WPS (on an iRiver H340).
08:19:13Paul_the_NerdDraw a line through the right tip of the triangle, that splits the circle in an even half (so it passes through the middle of the left edge of the triangle).
08:19:19*dickon has searched the manual and forums and is still confused
08:19:29dickonrockbox, um, rocks, BTW :)
08:19:32Paul_the_NerdThat line goes the same length from the left edge of the circle to the left side of the triangle, as from the right tip of the tri to the right edge of the circle
08:19:42Paul_the_Nerddickon: Play a song.
08:19:47Paul_the_Nerddickon: If a song is already playing, tap play
08:19:50Paul_the_NerdPlay, not navi
08:20:23dickonah, my tapping was not subtle enough
08:20:36dickon(i was already playing)
08:21:07dickonnow i just need to find some time to look into writing a multiband compressor for rockbox :)
08:21:39dickoninstallation on my H340 went very smoothly
08:22:07NimdaePaul_the_Nerd: (re theme) maybe so, but meh, i didn't think it would have fit so well
08:22:10*Nimdae shrugs
08:22:12dickonshame about the state of the manual, but who wants to spend time documenting stuff which is (i guess) still evolving
08:22:23Nimdaeas in fit the rest of what's going on
08:23:01Paul_the_NerdNimdae: Hey, no worries. :) I'm a little bit OCD is all.
08:23:18 Join Star [0] (
08:23:34Nimdaei do see your point though
08:23:36Paul_the_Nerddickon: The manual is actually pretty good in most cases. The problem is that rarely when someone finds something actually missing do they post a proper bug report about it.
08:24:07Nimdaei'll keep that design tip in mind for future themes
08:24:23Nimdaebecause after doing this one, i think i enjoy it, and plan to do more ;)
08:24:38Paul_the_NerdNimdae: Well, my aesthetic sense surely isn't the world's. Almost every "popular" theme looks pretty bad to me. :)
08:25:07Nimdaesame with me
08:25:26Nimdaeas you will notice, there's not a lot of activity in my theme, tried to keep it simple
08:25:47Paul_the_NerdBut then I look at the screen so little that I don't actually use a theme, just a very slightly tweaked default WPS
08:26:11Nimdaei'd have to use a larger font on the default
08:26:57Paul_the_NerdNone of my Rockbox'd DAPs have particularly high resolutions
08:27:02Paul_the_NerdI think the Nano would be it.
08:27:37 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:28:04Nimdaeyeah, i figured the text size on the default was for compatability with other DAPs
08:31:23Nimdaei wanna make another wps that suits the styling of the black ipod a bit more (since that's what i have)
08:39:29 Quit Starrr (Read error: 110 (Connection timed out))
08:40:04 Join pondlife [0] (
08:42:09 Join myzar|away [0] (
08:43:05 Nick myzar|away is now known as myzar (
08:53:11Paul_the_NerdLinusN: The search problem is recurring.
08:53:36***Saving seen data "./dancer.seen"
08:54:41*Paul_the_Nerd wonders if something actually causes this.
08:59:25LinusNtry now
08:59:59Paul_the_NerdFixed, thanks
09:02:41 Join ender` [0] (i=null@
09:05:47 Join einhirn [0] (
09:05:47 Quit einhirn (Client Quit)
09:05:53JdGordonLinusN: do you remember the ata access later stuff from the menu recoding patch? midkay thought it would be good to get that out and use it to get the audio buffers to fill up every time the disk spins to save spinning it up manually... what do you rekon?
09:07:23midkaywell as i said it'd pose a problem for clearing already played buffer - you couldn't then rewind in the current track. i personally think that a nice medium would be to clear out all the already-played tracks and rebuffer that space (leave all of the current track buffered) - but then again with longer files it's a different story..
09:07:43midkayit needs some thinking out, but yeah - basically refilling at least sometimes on manual disk spinups (at least when buffer is low or something).
09:07:56midkaybut anyways - i'm actually heading off to sleep right now. so, night all.
09:08:16 Join nls [0] (
09:08:30JdGordonbah, sleep is for the week
09:09:00LinusNthat sounds like an interesting option
09:09:31*dan_a wouldn't mind sleep for a week!
09:09:39LinusNme neither
09:09:53nlsheh, feature request for gapless playback
09:09:54midkayhaha. i'm weak indeed.
09:10:38nlsand we are one request short of 500 (open that is)
09:10:58midkayan option might be a good idea.. if you aren't a rewinder/skipper, always purging everything that's already been played as an option would be good.. also one that keeps the current track data but nothing older.. and maybe one more that only refills when the buffer is like 50 or 75% empty.
09:11:23bluebrotherLinusN, do you have more knowledge of FS?
09:11:43bluebrotherwould be nice if FS could automatically link CamelCase words to the wiki.
09:12:12JdGordon[19:09] <LinusN> that sounds like an interesting option <- thats what i was after... i know now what im doing tomorow if my stupid friends want to study instead of going out :p
09:12:18JdGordoncyaz all
09:12:22 Quit JdGordon ("Konversation terminated!")
09:12:22LinusNbluebrother: cool idea
09:12:44bluebrotherafaik the next FS version has some kind of integration of dokuwiki
09:12:46LinusNwas about to say "stop!" to jdgordon ...
09:13:14bluebrother(which is a wiki that has great support for documenting code)
09:13:34bluebrothermaybe it would be as simple as upgrading and mod_rewrite-ing some urls?
09:14:30 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:15:57 Quit damaki_ (
09:16:29 Join tvelocity [0] (
09:16:48nlsLinusN, these can be closed now, right?
09:17:01 Join damaki [0] (
09:17:49LinusNnls: yes, i believe so
09:18:05nlsok, I'll do it
09:19:15 Join Gnelik [0] (n=Miranda@
09:21:27pondlifeLinusN: quick question? What sets the thread pointer in the queue_event structure? I only see it set to NULL.
09:23:06LinusNit is set when a thread waits for it
09:23:22 Join dan_a_ [0] (i=dan_a@
09:23:23 Quit dan_a (Read error: 104 (Connection reset by peer))
09:23:24pondlifeOK, and NULL
09:23:44pondlifeSorry...and NULLed after wait completes?
09:24:37 Nick dan_a_ is now known as dan_a (i=dan_a@
09:24:37LinusNthat is done by wakeup_thread()
09:24:58pondlifeThat makes sense. Thanks.
09:25:34pondlifeSo queue->thread->name is no way to identify a particular queue :(
09:25:35 Join ender1 [0] (i=null@
09:25:56*LinusN still waits for Slasheri to update RockboxKernel
09:26:02LinusNpondlife: nope
09:26:18LinusNbut you could add a name member
09:26:34pondlifeYes. I fear I will need to.
09:27:03*petur wonders if it's a big nono to show the setting on two different places
09:27:13LinusN"the setting"
09:27:21petura setting
09:27:26peturclip hold time
09:27:56peturwould be better if it was also shown in the recording settings next to cliplight
09:28:07LinusNi don't see a great problem with it, of it is obvious that it is the same setting
09:28:16Paul_the_Nerdpetur: Where else is it?
09:28:31peturGeneral Settings > Display > Peak Meter > Clip Hold Time
09:28:53Paul_the_NerdIs there a clip warning on playback clipping too or something?
09:29:01peturand I've seen too many questions about it already
09:29:04pondlifeWhy not just move it?
09:29:26peturbecause it's also used for playback clip indication?
09:29:37*petur wonders who uses it there
09:29:38pondlifeI didn't know that.
09:30:07pondlifePlayback is clipping. Yeah, so what - too late.
09:30:27pondlifeIt's much more a recording feature.
09:30:34peturit is
09:30:34GnelikBagder: here is the link to e200r firmware, upgrading proess is seems to be like normal e200
09:31:40LinusNi have kind of wondered if playback clip indication is that useful...
09:32:06BagderGnelik: you have a fdisk -l dump of the disk too?
09:33:23GnelikBagder: Nope. I have e200. Look at
09:33:56Bagderthat doesn't help, I'm curious on a "hidden" partition
09:34:07Bagderthe vanilla e200 has one
09:34:53Paul_the_NerdLinusN: I don't think it is, though if it's useful for making sure an EQ isn't making your music clip (assuming you can't hear it, though if you can't it isn't that big of a problem)
09:35:15GnelikIf I try to update my e200 with e200r firmware??
09:35:33BagderI don't think it'll work
09:35:41LinusNPaul_the_Nerd: removing silly options is always a good thing
09:36:08 Join dan_a_ [0] (i=dan_a@
09:36:18 Quit dan_a (Read error: 54 (Connection reset by peer))
09:36:39Paul_the_NerdLinusN: I agree. I'd actually be in favour of moving it to just the recording menu, myself.
09:36:49BagderGnelik: now I know it won't work
09:36:57LinusNdo we still have all those peakmeter options?
09:37:25BagderGnelik: it uses a different encryption key
09:37:46GnelikBagder: Or different bl?
09:38:03Bagderdon't know about the BL since that firmware zip only had a mi4
09:38:04LinusNyes, we have a shitload of peakmeter options
09:38:23Bagderbut the mi4 uses a different key than the normal e200 ones
09:38:26*LinusN joins amiconn in a code size crusade
09:38:44nlsand there we go 500 open requests yay! (and the 500th is yet another one for ipod remotes) :/
09:38:55*petur thinks of some features to add :p
09:39:22bluebrotherspeaking of the recording menu ... how about moving the recording menu one level up?
09:39:29LinusN"BTW, how can I express my gratitude for such a great program?"
09:39:37LinusNopen your wallet, dammit! :-)
09:39:51bluebrotherIt can get accessed from the recording screen and it seems a bit strange to go to another submenu to actually start recording
09:40:04bluebrotherthat's different to e.g. the radio screen.
09:40:14*Paul_the_Nerd has file a few feature requests.
09:40:42LinusNbluebrother: makes sense, i think
09:40:54rasherBagder: I modified my voice building patch. It's much less hackish now.
09:41:21LinusNspeaking about voice building, maybe it's time to start building archos-specific voice files?
09:41:22*Paul_the_Nerd raises the "Why not use the Record button" topic.
09:41:40*LinusN raises the "why not ban Paul_the_Nerd" topic
09:41:45 Quit ender` (Read error: 110 (Connection timed out))
09:41:49rasherNot all targets have one
09:41:59Paul_the_Nerdrasher: This is true.
09:42:08LinusNthose who do should use it imho
09:42:23Paul_the_NerdHow many recording targets that don't have a recording button, have that F3 button (or equivalent) the Recorder has?
09:42:29bluebrotherit could also get used as a generic quick button. But maybe that's option bloat.
09:42:39rasherYeah, it's an unfortunate consequence of gradual development
09:42:49LinusNi'm in favour of using hold+repeat to get to the recording screen
09:43:15*bluebrother agrees
09:43:18peturbluebrother: the moment you enter recording screen, prerecording starts, maybe you'd like to change some settings before it does that?
09:43:44bluebrothermaybe, but if you change settings mustn't the prerecording start again?
09:43:49LinusNunfortunately, rec+repeat is the quick menu on x5 :-(
09:44:32Paul_the_NerdLinusN: My suggestion was that Long-Record (Rec+Repeat) go to the recording screen from WPS and Filetree. In Filetree, Rec on its own is Queue Next (or just Queue), Play+Record is insert. In WPS, Rec shows current playlist, and maybe Rec+Play is playlist catalog (or some other function, that last combo leaves me baffled)
09:44:33bluebrotherI don't see any problem there. Also, the recording settings could get their own submenu e.g. below the general settings.
09:44:36peturbluebrother: for encoder settings I think it does, for all other settings it keeps on buffering
09:44:47Paul_the_NerdLinusN: But then someone pointed out that the X5 uses Rec for useful stuff already
09:44:57 Quit bawb2 (Remote closed the connection)
09:45:01rasherPaul_the_Nerd: too bad for the x5 I say
09:45:06Bagderyeah, the x5 has a shortage of useful buttons
09:45:13rasherbutton assignments don't have to be the same
09:45:14bluebrotherhmm. But I don't see any reason why not to start prerecording over when settings change.
09:45:18Paul_the_NerdBut I think it'd overall cover *most* of what people want from the button on iRiver
09:45:29LinusNPaul_the_Nerd: it's funny how the manufacturer labels the buttons, and then afterwards find out that they have too few of them
09:45:40peturbluebrother: not for all settings changes *please*
09:45:49Paul_the_NerdI'm not a fan of that second quickmenu the soft button on Recorder gets to, the screen related one.
09:46:07bluebrotherhehe ... I won't do that, I'm only suggesting.
09:46:56 Nick myzar is now known as myzar|away (
09:47:11dan_a_I've thought in the past that maybe rec+repeat for the recording screen, play+repeat for the WPS and menu+repeat for the main menu might be logical
09:47:31bluebrotherwhat does rec on the x5? Main menu?
09:47:57peturbluebrother: I was just thinking of this: you want to record line-in but source is still set to mic. If you start to record too soon you'll have a part of mic included in your recording...
09:48:27bluebrothersure, but when you restart prerecording this shouldn't happen.
09:48:37peturso only restart prerec for source change and format change
09:48:41GnelikBagder: Where can i read about changing keymaps
09:48:56bluebrotherand I guess you'll start recording only when you set things right.
09:49:05bluebrotheror am I wrong with that assumption?
09:49:15BagderGnelik: I'm not sure there's much else than source code to read
09:49:37peturbluebrother: yes
09:49:43bluebrotherbut if the recording settings are below general settings this problem also shouldn't arise.
09:50:04bluebrothercurrently it's the same issue if you access the recording settings from the recording menu.
09:50:48bluebrotherpetur, yes to my assumption being wrong?
09:50:54peturbluebrother: would you dupleicate the settings menu then?
09:50:57GnelikBagder: than what is "Prereq button code" my english is not very good
09:51:34petur[09:48] <bluebrother> and I guess you'll start recording only when you set things right. <−− yes
09:52:01bluebrotherduplicate? no.
09:52:31peturwould you remove recording settings from recording screen then?
09:52:36bluebrotherIt's the same as now, you can access the recording settings from the recording screen and from some menu. I only would put that menu somewhere else
09:52:37dan_a_Gnelik: prereq is short for prerequesite - the button that must be pressed before
09:53:07bluebrotheradd a "Recording Settings" entry to the main menu, or put it below the "General Setting"
09:53:33Gnelikdan_a_: thnx
09:53:54LinusNthe extra level in the recording menu is often confusing for the blind
09:54:28peturbluebrother: I'd rather keep the recording settings menu accessible from within the recording screen
09:54:50LinusNpetur: nobody is talking about removing the settings from the rec screen
09:54:54bluebrotherpetur, I never intended removing that.
09:55:22bluebrotherbut I think it would be more logical to have a "Recording Settings" entry like the other settings in the main menu.
09:55:30 Join bawb2 [0] (n=bawb2@
09:55:38petur[09:50] <petur> bluebrother: would you dupleicate the settings menu then?
09:55:44petur[09:52] <bluebrother> duplicate? no.
09:56:00 Quit tvelocity (Read error: 110 (Connection timed out))
09:56:15peturbah... much lag
09:56:42bluebrotherbtw, the main menu could get a bit resorted. Put all stuff like recording and fm radio to the top and all settings entries to the bottom or similar
09:57:24peturwould make sense
09:57:27 Quit EspeonEefi ("Leaving")
09:57:41bluebrothermy idea is only to have the "Recording Settings" in a different position −− the main menu.
09:58:08petur[09:54] <petur> bluebrother: I'd rather keep the recording settings menu accessible from within the recording screen
09:58:10bluebrotherand accessing that directly from the recording screen −− I thought this is the way it works now.
09:58:31*petur is confused
09:58:50bluebrotherok, I'm trying ...
09:59:04bluebrother(a) have a "Recording Settings" entry in the main menu
09:59:20bluebrother(b) have a "Recording" entry in the main menu that takes you directly to the recording screen
09:59:52bluebrother(c) from the recording screen some button (A-B on my h120) takes you directly to the recording settings like it already does.
10:00:56peturwhich shows the same settings menu on two different places - I hope this won't confuse some people, I wouldn't mind
10:01:15pondlifeAha, the backlight queue is the overflowing one
10:01:59bluebrothersure, but it does so already. It's only those two places are below one "Recording" entry which I thik would better be removed so this entry they are below afterwards is the main menu
10:02:03*Bagder goes for a refill of that black stuff
10:02:24peturbluebrother: ok, fine for me then...
10:02:44bluebrothermaybe I was talking in a somewhat confusing way ;-)
10:03:17peturI'm not much of a 'morning person' ;)
10:03:40bluebrotherme neither
10:03:45*petur discovers his brain was not switched on yet *kicks it*
10:04:01pondlifeAnyone mind if I commit named queues - only enabled in LOGF builds?
10:04:49dan_a_Bagder: Guinness? At this time of day?
10:05:21pondliferefill, too!
10:06:41peturbah... the only pub that served Murphy's has closed this summer
10:07:08*bluebrother only heard of "Murphy's law" before
10:09:39*petur goes away to do some stuff his boss can appreciate
10:09:47 Quit petur ("worrrk")
10:10:42pondlifeWhy would the backlight thread overflow so easily?
10:11:06*pondlife wonders if his named queues stuff is telling the truth.
10:14:30pondlifeNope, it *is* the backlight
10:14:33*scorche pings amiconn
10:14:49GnelikBagder: do we need new keymaping or i am doing unused work???
10:15:05BagderGnelik: build a simulator and see
10:15:29LinusNpondlife: because the backlight thread gets an event for every button event
10:16:01GnelikBagder: i have allready buil it.
10:16:20pondlifeIt also overflows on shutdown, if playing back.
10:16:24BagderGnelik: and how are the keys working?
10:16:53GnelikBagder: bad
10:17:05LinusNpondlife: probably because holding OFF generates a stream of button events
10:17:37BagderGnelik: then you should patch it to become better! ;-)
10:18:19pondlifeHmm, but how to solve...?
10:18:32pondlifeAnyway, time for work.
10:18:54LinusNpondlife: i can imagine one simple fix: only send the backlight_on event if the backlight is off
10:19:45pondlifeCould that not risk a backlight on being missed?
10:19:57LinusNbut that opens up a can of worms with races
10:19:59pondlifeI assume the queue reader is quite low priority here.
10:20:05pondlifeYep, I don't like races.
10:20:21 Join EspeonEefi [0] (i=espeonee@CERULEANCITY.MIT.EDU)
10:20:31pondlifeMaybe have a minimum interval between postings?
10:20:55pondlifeWe have a tick timer, don't we...?
10:21:08LinusNor let the backlight thread use queue_remove_from_head()
10:21:23pondlifeWhat does that do?
10:21:37LinusNremove all identical events from the head of the queue
10:21:51pondlifeThat would be safe.
10:22:00rasherBagder: any time to look at my updated voice building patch?
10:22:05GnelikBagder: how can i only rebuild keymap-e200.c not runing make for all project?
10:22:16LinusNyes it is used in the hw playback engine for the buffer refill events
10:22:43pondlifeUse it after queue_wait? I'll have a search....
10:23:11pondlifeAh, no - before queue_post
10:24:11dan_a_Gnelik: make should only rebuild files that have changed.
10:25:02Gnelikdan_a_: it do so but much time gone while it find that keymap-e200.c is changed.
10:25:28 Join roolku [0] (i=522906c5@gateway/web/cgi-irc/
10:25:49pondlifeThat works for me
10:26:47roolku<LinusN> i have kind of wondered if playback clip indication is that useful... <−− extremly useful when adjusting the precut for the equalizer
10:27:06Paul_the_NerdGnelik: Are you running configure again, or are you just going into the build directory and typing make?
10:27:09roolkupondlife: did you by any chance my comment from this morning?
10:27:12LinusNroolku: ok
10:27:39roolkupondlife: slasheri / pondlife: a recent change seems to have buggered (technical term) the runtime data display in the wps. It seems to be an issue with track_buffer_callback - it gets set correctly in tagtree_init, but when I check the value just before its execution it is somehow reset to 0 (and is not called as a result)
10:27:51roolkupondlife: any idea how this can happen?
10:28:52Paul_the_NerdGnelik: That's the proper way. How long does it take?
10:29:52GnelikPaul_the_Nerd: a minute to find that keymap changed, than it upgrades exe file and i interrupt it witn ctrl+
10:30:26BagderGnelik: try 'make bin'
10:30:41Bagderit builds the main bin only
10:32:21GnelikBagder: thanks, now i dont need to interrupt it
10:32:56 Quit roolku ("CGI:IRC (Ping timeout)")
10:33:37GnelikLittle question: may i write redeclarete cancel button for quick screen?
10:36:58pondliferoolku: Sorry, no
10:37:12pondlifeDo you mean the progress bar?
10:37:53 Join roolku [0] (i=522906c5@gateway/web/cgi-irc/
10:38:02roolkustupid web client
10:38:29roolkupondlife: sorry, no the playcount etc
10:38:54pondlifeNo, sorry. I do know of a problem where the progress bar will read 0:00 near the end of a track.
10:39:27pondlifeIs it the track_buffer or _unbuffer that is at fault?
10:39:54pondlifelostlogic redid the buffering a short while ago, that is most likely when it changed...
10:39:58roolkupondlife: (wps tag %rp for example)
10:40:28pondlifeIs this data from tagcache runtime stats?
10:40:40roolkupondlife: if printing the address of the call back function immideatly after it is set it is correct
10:41:20roolkupondlife: but when checking it just before it is supposed to be called it is reset to NULL (and obviously doesn't execute)
10:41:54roolkupondlife: looks like a buffer overflow or something overwriting the value with 0
10:42:10roolkupondlife: yes
10:42:50pondlifeCan you build a sim and put a breakpoint on audio_set_track_buffer_event?
10:42:58roolkupondlife: (but I could be wrong)
10:42:58pondlifeThat might be useful.
10:43:23pondlifeI'm not able to help today, I'm afraid. Work...
10:43:36pondlifeOf course, make sure this is on Flyspray too...
10:45:25roolkupondlife: ah, I can set write-access breakpoints - that sounds like a good idea
10:45:40roolkupondlife: got to go to work myself though...
10:45:57pondlifeYou can try it, but I'd start with the simple flow stuff here.
10:51:03 Quit roolku ("CGI:IRC (Ping timeout)")
10:53:39***Saving seen data "./dancer.seen"
10:54:38GnelikBagder: can i remap cancel action only for quick screen
10:55:18BagderI'm not the most fluent person in the action code
10:55:22Bagderbut I would think not
10:57:12dan_a_I think that technically you can, by using the quickscreen context. I'm not sure it's a good idea though.
10:58:25Bagderright, as the point is to use the similar mappings all over
11:00:10GnelikIn the main mapping left button is cancel but in the quick screen shufle options if at the left side
11:03:05pondlifeOops, red ink
11:03:13amiconnLinusN: You really think we have too many peakmeter options?
11:03:29amiconnI only see one option that might be superfluous, and that is linear mode
11:04:18 Join Deef [0] (
11:04:57LinusNi think the min peak meter might be superfluous too
11:06:55 Join spiorf [0] (
11:09:46amiconnI don't think so. The min setting allows to select either precision or overview
11:10:09*amiconn always sets min peakmeter to -50dB instead of the default -60dB
11:10:44 Quit spiorf (Read error: 54 (Connection reset by peer))
11:11:36 Quit damaki (Read error: 104 (Connection reset by peer))
11:12:17LinusNok then
11:12:50dan_a_Gnelik: What do the other ports do? I think the Iriver H10 has similar keys to the Sansa
11:15:20Gnelikdan_a_: they use another buttons
11:15:31 Join skwad [0] (
11:19:00 Quit idnar (Read error: 110 (Connection timed out))
11:19:15scorcheamiconn: should i be searching for "gm71vs65163clt5" for the 8MB chip, or is there an easier to find chip?
11:19:54dan_a_Gnelik: Then I guess use another button. I don't think that's very user-friendly, though.
11:21:11Gnelikyep, i'l do so
11:21:14amiconnscorche: I have no idea. Never did the 8MB mod, and don't intend to do it either
11:22:08scorcheany particular reason, or just not worth the bother?
11:23:13 Join aliask [0] (
11:23:20amiconnI don't want to risk my archos
11:23:49 Join damaki [0] (
11:27:23Gnelikdan_a_:i thought i would be much easier
11:27:26 Join spiorf [0] (
11:28:17 Join Kohlrabi [0] (
11:29:13 Quit spiorf (Read error: 54 (Connection reset by peer))
11:30:20 Join jhMikeS [0] (
11:30:56 Quit Deef (Read error: 60 (Operation timed out))
11:31:03pondlifeSlasheri: Do you think bug could be closed now? No response from anyone in quite a while.
11:31:54 Join idnar [0] (i=mithrand@unaffiliated/idnar)
11:31:55Bagderimho, entries that await feedback can be closed without action after about 2 weeks wait or so
11:31:57 Quit skwad (Read error: 104 (Connection reset by peer))
11:32:27Paul_the_NerdI think two weeks is a perfect waiting period
11:32:31Paul_the_NerdIt's not like they can't be reopened
11:32:34 Join skwad [0] (
11:33:14 Part LinusN
11:34:13pondlifeOK - that's another 3.0 bug fixed!
11:34:16jhMikeSpondlife: one thing that floods the backlight queue is the Caption Backlight option being on. but I think for that the most recent state requested should be the one used and those message should never be queued at all. Just something I thought about some time ago but never tried out.
11:34:54pondlifeJust scrolling flooded the backlight queue, or shutting down.
11:35:25pondlifeTesting the end condition before queuing could introduce a race, so best not IMHO.
11:35:26Slasheripondlife: yep, there has been a lot of fixes to tagcache during that period
11:35:34pondlifeIt's closed.
11:36:38 Quit pondlife ("disconnected has pondlife")
11:36:39Slasherigood, in fact i don't even currently know if there exists any serious bugs in tagcache at the moment
11:37:23amiconnNo serious bug, but there's still that weird track number guessing
11:39:31*jhMikeS isn't sure what sort of race condition pondlife could be referring to.
11:40:50 Join qailer [0] (
11:43:02Slasheriamiconn: yes, but in fact that's not so bad anymore because now there is also a (hidden) flag when track number has been guessed
11:43:31Slasheriin future it should be possible to detect that flag in tagnavi configuration and ignore guessed track numbers if user likes it that way
11:43:49linuxstb__Is the generation of tagcache now robust? i.e. are we still getting reports of failures to initialise?
11:43:54 Nick linuxstb__ is now known as linuxstb (
11:44:19amiconnThe guessed track number is always wrong here
11:44:19Slasherilinuxstb: currently there is no that kind of bugs i would know about of
11:44:52amiconnThe only tracks where it guesses are those which _deliberately_ have no track number, and then it uses some random number (from artist or so)
11:47:52Slasherihmm, i wonder if that guessing causes more troble then than it's worth of
11:48:26*amiconn has lots of tracks listed as having a track number of 66
11:49:22Slasheriit would be easy to remove that guessing but i still wonder would it be better for user to be able to disable it (through a custom tagnavi configuration) or completely remove the guessing
11:49:39Paul_the_Nerdlinuxstb: There's still at least one active thread in the forum where someone is having it fail to initialize
11:49:55Paul_the_NerdSlasheri:;topicseen#msg56547 <−− But it may be a bad file.
11:50:23Paul_the_NerdI really *really* think that "Initialize" should not be a background function, but rather provide progress and maybe even some sort of failure report
11:50:53jhMikeSis it the sort of guessing where not even a human would have a reasonably high chance of being right?
11:51:02Paul_the_NerdI think the vast majority just set their player down and wait for initialize to finish anyway
11:51:39SlasheriPaul_the_Nerd: ah, that kind of reports/problems are hard to solve because there is no way to know what has really happened
11:52:02Paul_the_NerdSlasheri: Indeed.
11:53:33Slasherilogf output of the problem would be more useful.. it would be useful to implement a good error reporting system indeed
11:53:57 Quit skwad (Read error: 113 (No route to host))
11:54:04Paul_the_NerdI think a lot of users would appreciate a non-background "Initialize" anyway, just to they can more clearly know when it's done, and so on.
11:54:12Slasheria syslog that could open a "pop-up" message or something like that when an error has occurred
11:55:02Slasherihmm, true. Maybe it could be an option so that when starting initialisation, it would start in a foreground. And pressing some button, would bring it to background
11:55:22 Join roolku [0] (
11:55:25Slasheriand when selecting "initialize/update" again, it would bring back the process from background to foreground
11:56:04Slasheriwould would you think about that?
11:56:41roolkuSlasheri: Hi :)
11:56:48Paul_the_NerdSlasheri: I'd be fine with that.
11:56:48Slasherihi roolku :)
11:56:59roolkuSlasheri: did you see my comments re: runtime stats gathering
11:57:08Paul_the_NerdMaybe even first press of "Update" stays in the background, and a second press is foreground (but Initialize defaults to foreground)?
11:57:09SlasheriPaul_the_Nerd: at least that would be possible to implement with a reasonable effort
11:57:21roolkuSlasheri: I now know the problem, but not the solution (don't understand playback)
11:58:13roolkuSlasheri: audio_preinit is called *after* audio_set_track_buffer_event and overwrites the vector for the callback with NULL :(
11:58:13Slasheriroolku: yep, but i haven't got a time to investigate that yet.. I am running a somewhat older (a few days) build so I don't have that problem.. It might have something to do with pondlife's recent commits
11:58:43Slasheriroolku: oh, just a moment.. i will have a look
11:58:52roolkuSlasheri: I think it might - I had a look at the diffs, but nothing obvious stuck out :(
11:59:03SlasheriPaul_the_Nerd: yes, that sounds great
11:59:17roolkuSlashheri: thanks
11:59:34Paul_the_NerdSlasheri: As well, that'll help people report more useful information maybe
12:00:01jhMikeShmmm...thanks actually. should test a cvs build and see if that's what's haunting me.
12:00:55Slasheriroolku: hmm, i wonder how that is possible.. audio_preinit() is called from main() before tagtree_init()
12:03:08roolkuSlasheri: this is the debug patch I am using:
12:03:32roolkuSlasheri: and it shows: ROOLKU: SET 4305099
12:03:32roolkuROOLKU: RESET 0
12:04:35 Quit Gnelik (Read error: 131 (Connection reset by peer))
12:05:07roolkuSlasheri: also when playing a file, I get: ROOLKU: TEST 0
12:05:24Slasherihmm, do you try it with a real unit or with simulator?
12:05:38Slasheriit seems that simulator initializes tagcache too early
12:05:50roolkuI noticed it on the real unit first (no display of %rp anymore)
12:05:57roolkuthen I experimented with the simulatior
12:06:04Slasheriah, hmm..
12:06:11roolkusimulator to find out what is wrong
12:06:22Slasheribut at least simulator has that issue, but shouldn't be the same with real unit
12:06:53 Join fejfighter [0] (
12:07:11Slasheriso please try logf if you could find the cause by using the real unit :)
12:07:55roolkuSlasheri: there is so much output on the logf it scrolls past and I can't see a thing :(
12:08:08roolkuSlasheri: I have tried
12:09:28Slasheriroolku: can you use the logfdump feature?
12:09:42Slasherithat will create a /.rockbox/logf.txt file
12:09:48roolkuroolku: I think I might remove "track_buffer_callback = NULL;" and see if I get my %rp back
12:09:48Slashericontaining the full logf output
12:10:04roolkuSlasheri: ah, good idea, forgot about the dump
12:12:36 Quit damaki (Read error: 104 (Connection reset by peer))
12:12:47 Join damaki [0] (
12:21:01 Join pixelma [0] (i=pixelma@
12:25:50 Join zok [0] (
12:29:38 Join `axion [0] (
12:42:15 Quit sucka ("( :: NoNameScript 4.01 :: )")
12:45:49 Join TeaSea [0] (n=thunderc@
12:47:28roolkuSlasheri: you are right - it only seems to be doing it in the sim - back to the drawing board :(
12:48:34 Join pondlife [0] (
12:53:41***Saving seen data "./dancer.seen"
12:59:38 Join mirak [0] (
13:02:30 Quit nudelyn ("At Argon, we're working to keep your money.")
13:07:56 Join nudelyn [0] (
13:09:45 Join Criamos [0] (
13:11:47 Join Stargazey [0] (
13:18:46 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
13:19:11 Part Paul_the_Nerd
13:21:04amiconnpondlife: Two questions arise from your commit:
13:21:33amiconn(1) It would be interesting to know _what_ is flooding the backlight queue instead of just fixing the symptom
13:21:57amiconn(2) Does backlight need a queue at all?
13:22:07 Join PaulJam [0] (
13:27:00 Join `axion_ [0] (
13:28:00jhMikeSamiconn: (2) heh :D Possibly but I don't think on/off messages have to be FIFOed but rather placed aside yet still logically dequeued by the thread much like WM_PAINTs in windows.
13:29:01pondlifeamiconn: (1) The button presses call backlight_on() faster than the queue is cleared.
13:29:18pondlife(2) Not sure
13:29:35 Quit `axion_ (Client Quit)
13:29:38 Quit Star (Read error: 110 (Connection timed out))
13:29:46pondlifeI just wanted to fix that so I could see if I get any other queue overflows.
13:29:54 Join lucas42 [0] (
13:30:01 Quit `axion (Read error: 145 (Connection timed out))
13:30:31jhMikeSpondlife: why would a race condition be inevitable from using the final state anyway?
13:32:35 Quit Siku ()
13:33:09 Join dan_a [0] (i=dan_a@
13:33:16 Quit dan_a_ (Read error: 131 (Connection reset by peer))
13:34:34pondlifeI was thinking if the backlight_on call checked to see if the backlight was already on, then it wouldn't reset the timer correctly, or somesuch.
13:35:15pondlifeHaven't put much thought into it, but if you test outside of the queue, then you are surely risking a race situation. That's assuming the queue is there for a good reason...
13:35:24jhMikeSoh, no. that wasn't where I was going with it. that would occur that way.
13:35:35pondlifeAh, I misunderstood then
13:36:11jhMikeSThe message would exist but only in its final state. It would be "dequeued" by the thread still. Just in a variable on the side that says it exists and what the final message was.
13:36:37pondlifeA seperate state variable on the queue posting side?
13:37:48 Join GreyFoux [0] (
13:37:58amiconnImho the backlight could do with bool current_state; bool last_state and a last_update timestamp
13:38:11amiconnThen it wouldn't need a queue at all
13:38:12pondlifeYes, the whole thing needs simplifying.
13:38:26jhMikeSIt's get set in queue_post to the final state. It gets cleared in queue_wait and the correct on/off message returned. A three state variable.
13:38:42 Join netmasta10bt [0] (
13:38:51pondlifeWell, I'll leave that to the backlight experts.
13:38:51jhMikeSBut I'm not saying to modify the queue functions, just whatever the backlight would implement.
13:44:59 Join yaca [0] (
13:45:29 Quit MarcoPolo (Remote closed the connection)
13:47:00 Join the_dan [0] (i=cb3b7570@gateway/web/cgi-irc/
13:47:57the_danhey i have a quick Q about rockbox for the iriver H20
13:48:53the_dando i need to remove all my music from the iriver and reload it
13:49:56the_danand all i need to do is drag and drop the music, and rockbox reads tge ID3s?
13:50:12amiconnH20 ?
13:50:27markunBagder: for the gigabeat owners the answer to this question will be 'yes' :(
13:50:33Bagdersome people call the h10 with 20GB disk h20
13:50:50Bagdermarkun: :-/
13:50:51the_dani was just on mistic river and they call it H20
13:50:59amiconnmarkun: Why is that?
13:51:01the_danthought i had the lingo wrong or something :S
13:51:04 Quit fejfighter ()
13:51:12markunamiconn: toshiba encrypts all the files you put on the device
13:51:41the_danwell, done deal, rockbox goes on tonight!
13:51:46the_danthanks for the help
13:51:49 Part the_dan
13:54:33linuxstbmarkun: What will installing Rockbox on a Gigabeat involve? Is it just a matter of copying some files to the FAT partition?
13:56:19markunoverwrite the kernel file FWIMG01.DAT with our rockbox bootloader and unzip the rockbox zip.
13:57:17peturso there will be an extra file in the root?
13:57:31markunno, the extra file is in GBSYSTEM/FWIMG
13:57:34 Quit lucas42 (Read error: 104 (Connection reset by peer))
13:57:46peturextra directory then
13:57:49Bagdernice names ;-)
13:58:03peturand if it gets deleted?
13:58:20markunthem you need to do a little surgery :)
13:58:35Bagderreally? no recovery mode?
13:58:57markunit enters recovery mode when you start the gigabeat with the hdd removed..
13:59:16Bagderoh, but how do you restore it without hdd?
13:59:21markunafter that you can connect the hdd while it's on and connect the USB cable.
13:59:34Bagderpretty awkward
13:59:49markunBefore we had a 'rescue mode' in our bootloader that was how we had to develop..
14:00:22linuxstbDoes the disk show up as a UMS device?
14:00:32markunToshiba still is not sure if they want to give me the datasheet for the LCD driver IC
14:00:45markununtil then I don't think we will work on a bootloader to be flashed.
14:01:47markunWe can copy the LCD init code from the original firmware, but that would be copyright violation and I would rather know exactly what the commands do.
14:02:17 Join Blackfish [0] (
14:03:48markunwe noticed there are some interesting things we can do with the gigabeat dock connector. It has i2s in and out and i2c.
14:03:53 Join `axion [0] (
14:05:43markunlinuxstb: yes
14:06:26 Quit pondlife ("disconnected has pondlife")
14:07:14linuxstbDoes the encryption happen on the device when files are transferred via MTP? Or does the host computer encrypt them?
14:07:37markunyou can do either
14:10:58markunI have a disassembly of the file kernel module that does the encryption
14:11:26markunbut I prefer not to use it and just use unencrypted files
14:11:56 Join [sellout] [0] (
14:19:07linuxstbIs there anything special the Toshiba firmware does that would mean people would like to dual-boot?
14:20:27 Join TCK [0] (
14:25:03markunlinuxstb: wma maybe
14:25:06 Quit petur ("worrrk")
14:25:24markunand maybe UI rotation
14:26:00 Join Arathis [0] (
14:26:23blue_lizardUI rotation seems to be very important in this channel for the last few days
14:27:34markunreally? Didn't notice
14:28:05Bagderui rotation?
14:28:13aliaskTurn display 90 degrees
14:28:17aliaskLandscape view
14:29:44markunmovies can just be encoded rotated, so it's not really a problem
14:30:02linuxstbmarkun: Hopefully WMA will come. I wonder if the Gigabeat could handle the original floating-point version of the codec...
14:31:28linuxstbWhats the current status of the gigabeat port? Is the filebrowser working yet?
14:32:20markunNo, still not. I traced the problem down to somewhere in gui_list_draw
14:33:13markunneeds more debugging
14:33:46markunIt completes init() at least
14:34:09 Join Deef [0] (
14:34:45bluebrotherhas there been any work on wma lately?
14:37:36linuxstbNot afaik. It's on my todo list (cleaning up the patch, mainly the ASF parsing side of things), but no time/motivation yet.
14:40:15markunlinuxstb, amiconn: what is your opinion on a seperate gigabeat branch?
14:42:23amiconnWhy a separate branch?
14:42:53markunbecause it would make development for us faster and we don't want to break the other targets all the time
14:43:02*amiconn doesn't see the point in branches. They just mean more work without benefit afaics
14:43:26markunWell, if you can convice us there is no benefit I will not ask about it anymore
14:43:56markunIt's just that we have a lot of tetscode and it's hard to keep it in sync with patches
14:44:31markunwhen rockbox is working a bit better we can all work on different parts and don't need to branch anymore
14:46:29 Join the_dan [0] (i=cb3b7570@gateway/web/cgi-irc/
14:46:38amiconnThere is firmware/test/ for testcode
14:46:56linuxstbBut you'll still have the problem of merging the branch back into Rockbox eventually.
14:47:09markunyes, I know that
14:47:48amiconnWell, I never really worked with branchen, so maybe I don't get the whole purpose of them
14:48:06markunwhat would be the problem for the non gigabeat devs if we have this branch?
14:48:07 Quit pleXa ("Lost terminal")
14:48:32amiconnThe only case where branches make sense for me is when there is a defined release version which should still receive bugfixes but no new features
14:48:51linuxstbI guess the problem is that you are more than one person hacking away at the same "unofficial tree" - so you want to keep your local copies in sync with each other...
14:49:05amiconnEven then it means additional work
14:49:25markunlinuxstb: exactly
14:49:32the_danright so ive gotten rockbox onto my H10, but the tracks are 'skipping'. Ive swapped to a non default WPS and its still doing it
14:50:28Arathisthe_dan: a non default wps without peakmeter?
14:50:31linuxstbSo does CVS allow you to resync your branch to changes made in HEAD after the branch?
14:50:38amiconnDoesn't git allow to set up sub-repositories?
14:51:16the_danArathis: didnt know i had to take away the peek meter, can i just disable it in settings or does it need to be written out of the wps?
14:51:21amiconnNote that I never worked with git, but it was mentioned at devcon and I've read a bit about git afterwards
14:52:30Arathisthe_dan: it has to be written out of the wps
14:52:40Arathisbut that isn't difficult at all.
14:53:04the_danim guessing there are some without it on the rockbox forums?
14:53:40the_danah in the wiki
14:53:46***Saving seen data "./dancer.seen"
14:54:33DeefIs there any kind of nifty visualisation thing for Rockbox?
14:54:46DeefLike the things you sit and stare at in Winamp
14:54:58ArathisiCatcher is without and it's with rockbox by default, but in the forum (@misticriver) and at and in the wps gallery in the rockbox wiki there are too. I think all H10 exclusive wps' are without cause it wouldn't make sense to implement it anyway
14:55:30Arathisthe_dan: do you prefer icandy or simple wps?
14:56:04the_danright now, anything that will work
14:56:07the_dansimple is fine
14:56:36the_danalso none of the WPS's fit on my screen properly
14:56:37the_danwhich is odd
14:56:59Arathisthe_dan: do you have a 5/6gb or 20gb h10?
14:57:18the_danand its skipping with iCatcher
14:57:46the_danit seams to depend on what im doing
14:58:02the_danif i am changing the volume or something itll happen regualry
14:58:16bluebrotherthe_dan, have you enabled eq / crossfeed / other sound changing functions?
14:58:46the_danall i have done is swap the WPS
15:00:00 Join MarcoPolo [0] (
15:00:01DBUGEnqueued KICK Arathis
15:00:01Arathisthese are direct links to WPS galleries for the H20 (H10/20GB)
15:01:30the_dannow whats the bext way to install these? dump in the directory and run the WPS?
15:01:44the_danor take them apart and add to the .rockbox folder?
15:03:18Arathisextract them and put them on the player. they should have the same filestructure as the rockbox folder
15:04:06Arathisand then choose one through the options/theme menu
15:04:47ArathisIf your layer still skipps afterall look at the RB forum and I think somebody allready had the same problem
15:05:34 Quit mirak (Remote closed the connection)
15:05:57Arathisthe_dan: and perhaps reset your settings by switching hold while booting
15:07:57ArathisEQ, peakmeter in WPS should be disabled at all. crossfeed won't affect playback that hard that it skips
15:08:46the_danjust reinstalled rockbox
15:08:49the_danetc etc
15:08:55Arathisit can also skip if you play a game or other CPU intesive plugin. but barrywardell works on activating the second CPU core so this won't happen anymore
15:08:58the_danand itll only skip with an EQ preset on
15:09:09the_dandual core? haha thats nuts
15:09:57Arathisthe_dan: so playback doesn't skip anymore for you (at least without EQ and pekameter)
15:10:27the_danyeah it good now
15:11:18the_danalthough, eq presets would have been nice :(
15:12:17Arathisif they aren't too hard it should work though
15:13:37the_dancould the amount of files on disk effect the cpu load?
15:13:48the_danenough to push it over the limit with the eq on....
15:14:10 Join spiorf [0] (
15:14:47 Quit spiorf (Read error: 54 (Connection reset by peer))
15:15:16the_danalthough ive gotta say, ti sounds better than with the iriver gear
15:15:56 Join NickDe [0] (n=nicholas@
15:16:02Arathisthe_dan: you could activate tag-cache to ram and dir cache to speed things up. don't know if it effects eq and cpu usage
15:17:38the_danmm where is the tag cache menu?
15:18:04 Quit rconan ("Leaving")
15:19:35Arathisgeneral options > file view
15:19:42Arathisthere is a manual btw ;)
15:20:17the_danit mentions the menu....but not where it is
15:20:20Arathisdir cache is at general settings (not options as said above..) > system > disk
15:20:41Arathissry, don't have english as language a rb
15:20:52the_danno worries
15:25:00 Join rconan [0] (
15:25:22 Join _FireFly_ [0] (
15:35:14 Join barrywardell [0] (i=892b7a86@gateway/web/cgi-irc/
15:36:53 Quit roolku ()
15:39:03 Quit aliask ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
15:43:34 Part zok ("Verlassend")
15:43:50 Quit amigan (Read error: 54 (Connection reset by peer))
15:45:06 Join MadDog011 [0] (
15:46:07 Part MadDog011
16:00:15 Join funky [0] (n=repulse@unaffiliated/funky)
16:00:25 Join damaki_ [0] (
16:02:42 Quit Deef ()
16:09:34the_danthanks for the help Arathis , its running sweet now
16:09:36 Quit the_dan ("CGI:IRC")
16:12:32 Quit thegeek_ ("( :: NoNameScript 4.03 :: )")
16:14:06 Quit damaki (Read error: 110 (Connection timed out))
16:17:43 Join spiorf [0] (
16:18:24 Quit spiorf (Read error: 104 (Connection reset by peer))
16:19:48 Join spiorf [0] (
16:20:14 Quit spiorf (Read error: 104 (Connection reset by peer))
16:23:15 Join mirak [0] (
16:25:17 Quit barrywardell ("CGI:IRC (EOF)")
16:25:20 Join spiorf [0] (
16:25:39 Quit spiorf (Read error: 104 (Connection reset by peer))
16:33:02 Quit GreyFoux ("Le vrai danger, c'est quand les hommes penseront comme les ordinateurs")
16:34:53 Quit secleinteer (Read error: 60 (Operation timed out))
16:35:07 Join secleinteer [0] (
16:36:26 Join lee-qid [0] (
16:39:43 Quit qailer ("Leaving")
16:42:12 Join qailer [0] (
16:43:13 Quit mirak (Read error: 104 (Connection reset by peer))
16:50:26 Join spiorf [0] (
16:51:27 Quit spiorf (Read error: 104 (Connection reset by peer))
16:51:39 Join spiorf [0] (
16:52:03 Quit spiorf (Read error: 104 (Connection reset by peer))
16:53:49***Saving seen data "./dancer.seen"
16:54:02 Join spiorf [0] (
16:59:18 Quit spiorf (Remote closed the connection)
16:59:52 Join thegeek [0] (
17:01:21 Join spiorf [0] (
17:01:30 Quit spiorf (Read error: 104 (Connection reset by peer))
17:02:29 Join spiorf [0] (
17:02:45 Quit spiorf (Read error: 104 (Connection reset by peer))
17:03:00 Join spiorf [0] (
17:03:32 Quit spiorf (Read error: 104 (Connection reset by peer))
17:03:41 Join Rincewind [0] (
17:06:23 Quit Rincewind (Client Quit)
17:07:05 Quit PaulJam (".")
17:07:42 Join spiorf [0] (
17:07:56 Quit spiorf (Read error: 104 (Connection reset by peer))
17:08:31 Join spiorf [0] (
17:10:48Bagderspiorf: fix it
17:12:58spiorfi'm bored enough of this by myself, don't worry
17:13:25 Join Blackfish` [0] (
17:13:32 Quit Blackfish (Read error: 145 (Connection timed out))
17:16:15amiconnBagder: Could you change to exclude the eq folder for hwcodec?
17:16:56Bagderyes, I have that on my todo
17:23:54amiconnThe backdrops folder should also be removed for charcell targets and b&w bitmap targets
17:24:54amiconn(charcell probably already is)
17:25:20 Join webguest74 [0] (i=5493be33@gateway/web/cgi-irc/
17:28:28webguest74hi, I just got my 320 and want to put rockbox on it. i'd like to enable video, so I guess I need the 1.29k firmware. but I can't find it. the link in the rockbox install page seems to lead to a chenged page. i managed to find the H300kor-v129-zip file there but it doesn't contain a .hex file but a iriverFirmwareVer1_29Korea.exe. what should I do now?
17:30:10rasherI believe I have it on">
17:30:33rasherAnyway, that file is a self-extracting exe.
17:32:30 Quit webguest74 ("CGI:IRC (EOF)")
17:33:00 Join webguest74 [0] (i=5493be33@gateway/web/cgi-irc/
17:33:09 Join |AhIoRoS| [0] (n=ahioros@
17:34:03 Quit webguest74 (Client Quit)
17:34:30 Quit akaias (Read error: 60 (Operation timed out))
17:36:49 Join bundik [0] (
17:45:26 Join akaias [0] (n=akaias@
17:47:30 Join Ribs [0] (n=freenode@
17:51:14 Join mirak [0] (
17:55:32 Nick ender1 is now known as ender` (i=null@
17:59:13 Quit dan_a (Read error: 104 (Connection reset by peer))
18:00:35 Quit _FireFly_ ("Leaving")
18:00:59 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
18:01:18 Join bluey- [0] (
18:02:08 Quit linuxstb (Read error: 110 (Connection timed out))
18:02:23 Quit mirak (Remote closed the connection)
18:06:00 Join mirak [0] (
18:17:51 Quit bluey- (Read error: 110 (Connection timed out))
18:24:20 Quit TCK (Read error: 148 (No route to host))
18:29:00 Nick daurn|laptop is now known as daurn|afk (n=quae@unaffiliated/daurnimator)
18:40:45 Join Siku [0] (
18:40:56 Quit bagawk (Read error: 110 (Connection timed out))
18:41:31 Join Larsie [0] (i=53538563@gateway/web/cgi-irc/
18:43:48 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
18:45:09LarsieI'm trying to get a 'normal' sansa e200 build
18:45:19bluebrotherthere is none
18:45:45bluebrotherwhile the sansa port is work in progress there isn't a working build suitable for users.
18:45:51Larsiehuh, so how do they use the simulator?
18:46:05LarsieI have build the simulator already
18:46:16bluebrotherthe sim is different −− simply as it doesn't simulate the complete hardware.
18:46:23bluebrotherit's a simulator, not an emulator.
18:46:23 Join TeaSeaLancs [0] (n=thunderc@
18:46:27Larsieand I have experimentated with the bootloader
18:46:47Larsiehow can I get the simulator to work then?
18:47:16bluebrotherwhen building the sim Rockbox gets compiled for your PC, not for the target. Low-level stuff gets wrapped by usual PC OS functions.
18:47:35bluebrotherI never tried. Just building as sim?
18:47:56LarsieI did, then it say's it misses the rockbox files
18:48:27Larsieno .rockbox directory, Installation incomplete
18:48:39bluebrothermake install
18:49:07Larsiehehe, that might just be it :)
18:49:33Larsienope, well.. I'll try to rebuild the sim once more
18:50:58Larsiedoesn't work... :(
18:51:01 Join hardeep [0] (
18:52:24 Join bluey- [0] (
18:52:34 Join tim66 [0] (
18:53:51bluebrotherLarsie, just tried. Works fine for me −− just the usual sim building. configure, make, make install.
18:53:52***Saving seen data "./dancer.seen"
18:54:02bluebrotherI only have no idea which keys to use :)
18:54:05 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
18:54:20bluebrotherbut I got to go now ... back later.
18:55:03Larsiehmmz... I'll download the latest source then
18:56:33 Quit yaca ("leaving")
18:57:05 Join yaca [0] (
18:58:39 Quit TeaSea (Read error: 110 (Connection timed out))
19:00:30 Join PaulJam [0] (
19:01:28 Quit daurn|afk (Read error: 104 (Connection reset by peer))
19:01:59 Join daurn|afk [0] (n=quae@
19:04:22Larsiedebian:/home/lars/rockbox-bleeding/sim# make installinstalling a full setup in your archos dircp: cannot stat `/home/lars/rockbox-bleeding/apps/eqs/*.cfg': No such file or directory
19:04:44Larsiedebian:/home/lars/rockbox-bleeding/sim# make install installing a full setup in your archos dircp: cannot stat `/home/lars/rockbox-bleeding/apps/eqs/*.cfg': No such file or directory
19:05:17Larsieand yes I did do make before make install
19:05:26Larsieand that just compiles fine
19:05:41 Quit daven ("leaving")
19:07:29 Quit qailer ("Leaving")
19:08:12rasherLarsie: the tarballs are broken in various ways. Try getting from cvs
19:09:35Larsieyou mean with the cvs command?
19:09:42rasherYeah, cvs checkout
19:13:04 Join barrywardell [0] (i=892b7a86@gateway/web/cgi-irc/
19:13:25Larsiemake: *** /home/lars/rockbox/uisimulator/sdl: No such file or directory. Stop.make: *** [all] Fout 2
19:13:47Larsiethe tarball worked better :P
19:14:37Larsiethere insn't a uisimulator folder at all :X
19:15:00rashercvs co uisimulator
19:15:05 Quit funky ("leaving")
19:15:09rasherin rockbox/
19:15:24Larsiehehe ty
19:15:33 Quit bluey- (Read error: 104 (Connection reset by peer))
19:15:34rasherThe rockbox module only includes what's necessary to build a regular build
19:15:46rasherthere's rockbox-devel or something like that which includes everything
19:16:15LarsieI was just reading about it :)
19:16:42rasherBut getting rockbox, then getting whatever else you need works
19:16:49 Quit mirak (Remote closed the connection)
19:18:35 Join RogerBacon [0] (
19:18:40rasherI'm trying the tarball to try and figure out what's missing
19:18:51Larsiesome config file
19:19:44rasherthe entire "eqs" dir is missing, yeah
19:19:59rasherSomeone forgot to update apps/FILES
19:21:01Larsiestill nothing :(
19:21:29Larsieno .rockbox directory, Installation incomplete
19:21:37rasherdid you make install?
19:21:38 Join obo [0] (
19:22:26 Join mirak [0] (
19:22:38rasherThat's odd
19:22:42Larsiedebian:/home/lars/rockbox/sim# make installinstalling a full setup in your archos dirdebian:/home/lars/rockbox/sim#
19:23:00rasherLet me try, sansa?
19:23:13Larsiethe SDL build
19:23:19 Join Ribs2 [0] (n=freenode@
19:23:56lowlightLarsie: now run rockboxui
19:24:34LarsieI'm using a rockbox patched gcc version 4.1.2 20060901
19:24:45rasherNot for the simulator
19:24:50rasherThat uses your host gcc
19:24:51 Join menosm___ [0] (
19:25:08Larsieno but for the compiling I meant
19:25:18Larsiethere's noting in the archos dir
19:25:28Larsieand nothing in the zip
19:25:33rasherSomething went horribly wrong then
19:25:38rasherDo you have zip and unzip installed?
19:25:43amiconnDo you have zip installed
19:25:48 Quit mirak (Remote closed the connection)
19:25:51Larsieah I don't think so
19:26:03rasherThat'd be it then
19:26:20Larsiehuh, yes I do have zip
19:26:25rasherAnd unzip?
19:26:39LarsieThis is Zip 2.32 (June 19th 2006), by Info-ZIP.
19:26:52rasherSame as I have
19:27:01LarsieUnZip 5.52 of 28 February 2005, by Debian. Original by Info-ZIP.
19:27:27rasherNow it's getting interesting.
19:27:46rasherls -l archos/
19:27:52rasherls -a archos/
19:28:06Larsiedebian:/home/lars/rockbox/sim# ls -a archos/. .. .rockbox
19:28:10rasherThat's correct
19:28:27 Quit Ribs (Nick collision from services.)
19:28:28 Nick Ribs2 is now known as Ribs (n=freenode@
19:28:44rasherIt really should work then.
19:29:01rasherWhat happens when you run rockboxui? (./rockboxui)
19:29:30Larsiewait a second I'm trying make install again :P
19:29:44rasherSidenote, why is the dir still called archos? Won't "disk" or something make more sense? amiconn?
19:29:57amiconnDoes that really matter?
19:30:06amiconnIt's just a sign of rockbox' history
19:30:36Larsiedebian:/home/lars/rockbox/sim# ./rockboxui −−−−−−−−−−−−−−−−−−−−−− DirectFB v0.9.25 −−−−−−−−−−−−−−−−−−−−- (c) 2000-2002 convergence integrated media GmbH (c) 2002-2004 convergence GmbH −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−-(*) DirectFB/Core: Single Application Core. (2006-07-31 16:59)(*) Direct/Memcpy: Using MMXEXT optimized memcpy()(!) Direct/Util: openi
19:31:11rasherNot awfully, I'm mostly thinking about "my" users (precompiled sims). They might not get it (although I've heard no questions or complaints, so maybe I'm misunderestimating them)
19:31:27rasherLarsie: Oh ah - you're using directfb?
19:31:38Larsieehhrr I guess so
19:32:04rasherWait.. what kind of machine are you doing this on?
19:32:23LarsieAthlon XP 2600+ running Debian etch
19:32:24rasherAnd how.. through ssh by any chance?
19:33:06Larsiethrough a terminal with root acces
19:33:16rasherHrm.. that might actually be a problem
19:33:29rasherWait.. are you not using X?
19:33:30Larsiethe root or terminal?
19:33:35Larsieyes I am
19:33:35rasherThe combination
19:33:41Larsieah ok
19:33:43rasherOkay, are you logged in with root user?
19:33:52 Quit RogerBacon (Remote closed the connection)
19:33:56rasherOr did you switch in the terminal?
19:34:12rasherBecause if you did, the root user won't use the X display
19:34:20rasherTry this...
19:34:30rasherDISPLAY=":0.0" ./rockboxui
19:34:50rasherBut really, there's no need for being root for this
19:35:00Larsiedebian:/home/lars/rockbox/sim# DISPLAY=":0.0" ./rockboxuiXlib: connection to ":0.0" refused by serverXlib: No protocol specifiedXlib: connection to ":0.0" refused by serverXlib: No protocol specified
19:35:00 Join actionshrimp [0] (
19:35:17Larsiewell ok, i'll try my own useraccount then
19:35:29rasherThat's probably because it's the wrong user then, I thought it might be enough to specify a display
19:35:35rashersince you're root and all
19:36:07rasherBut X is picky with permissions. You might have luck convincing it to let root draw to the display, but it will be more hassle than switching user
19:36:19 Join mirak [0] (
19:36:49 Quit mirak (Remote closed the connection)
19:37:11Larsiearg I can't make directories in the rockbox dir
19:37:18Larsieso it was the root thing
19:37:26Larsiethanks rasher
19:37:52rasherchown -R lars:lars rockbox
19:37:53rasheras root
19:37:56rasherShould fix it
19:38:08Larsieah too late, I already rm'ed it
19:38:46LarsieI'm using linux for nearly 2 months now
19:38:54Larsieso I'm a bit noobish :P
19:39:21 Quit menosm (Read error: 110 (Connection timed out))
19:41:58zeso according to an ipodlinux forum post from last month, somebody managed to get rockbox on an ipod 5.5G and it worked
19:42:21zejust a matter of a bit of (rather ugly) partition/filesystem hacking
19:42:40rasher30 or 80gb?
19:43:35 Quit daurn|afk (Read error: 104 (Connection reset by peer))
19:43:51zei think it was an 80GB that they did it on
19:44:07 Join daurn|afk [0] (n=quae@
19:44:30zeer wait
19:44:30rasherMight be interesting, I believe someone already has 30gb sortof working
19:44:39zeit just quotes a rockbox thread
19:44:40zedoh :p
19:45:06rasherAnd that'd probably be the person, I believe Kalthare is using a 30gb ipod
19:45:13CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
19:45:13*amiconn is finally getting somewhere with his emacified lcd_yuv_blit() :-)
19:46:04rasheryuv, psh. Who needs colours to play music?
19:47:00rasherLooks like my fix for the eqs worked.
19:47:18Larsiethere should be more players out there which focus on music instead of fancy plastic cases and GUI's
19:47:22rasherLet's see if anything else is missing
19:47:43jhMikeSamiconn: modified system memory guard to watch the voice swap buffer and bingo there something writing to it while voice is swapped but haven't gotten to narrowing down the fault :)
19:48:03amiconnHmm, now I'm somewhat confoozled about emac, latency and pipeline stalls :/
19:48:20amiconnThe speedup from shuffling instructions around isn't deterministic
19:48:37zei watched a movie off my mom's ipod the other day, via its tv out. surprisingly, it worked and looked great
19:48:52zelikewise when i watched one on its little screen, it was still surprisingly good looking and perfectly watchable
19:49:17jhMikeSamiconn: shouldn't it be doing something useful while waiting for results? :P
19:49:30zebut as far as music goes, come on... fancy themable GUI's are an mp3 tradition :p
19:50:11amiconnAnyway, current speedup is ~12% on H300 (vs. CVS)
19:50:12rasherJust kidding, I wouldn't mind a colourscreen. But I don't want it enough to ditch my trusty ihp120
19:50:25jhMikeSwe need a suite of gradient fill functions :D
19:50:34zerasher: yeah i know how you feel
19:50:50zerasher: i think i'd probably still take my karma over my mom's ipod...
19:51:18zealthough i dunno, i might change my mind on that once rockbox is 100% functional and optimized for ipods
19:51:41LarsieI hate the sansa actually, I just choose it for rockbox. The only other rockbox player I could get was an Ipod but I hate those even more :P
19:51:56Larsiethough I didn't hate the sansa before I bought it
19:52:30nlsWhat's wrong with it?
19:52:43zei just wish there were to be karma-descendants of the video + higher-capacity class and onward
19:52:46Larsiestupid stock firmware
19:52:51zeand rockbox support for them :p
19:53:12Larsiebut I'll just have to wait for rockbox :)
19:53:13jhMikeSamiconn: think the x5 could benefit from the DMA lcd work? Doing 32-bit pixels was easy and really didn't hurt it anyway. What about DMA for rect fills?
19:53:15rasherLarsie: I believe "stupid" is redundant in that sentence
19:53:54LarsieIt's just laggy and missing a lot of features
19:54:00amiconnjhMikeS: I don't think it'd help much. However, memcpy() will be improved further :)
19:54:13 Join Rob2222 [0] (
19:54:25amiconn32bit pixels are a waste of memory
19:54:40Larsieargh sim still doesn'twork
19:54:44LarsieI give up :P
19:54:51rasherWhat's wrong?
19:55:02amiconnLCD update would possibly be a bit faster but drawing would be slower
19:55:09nlsDoes rockboy use lcd_update or some custom black magic (like doom)?
19:55:11Larsieagain missing .rockbox
19:55:17jhMikeSamiconn: Why wouldn't it help it much? If there's a good speedup the memory cost isn't that horrible relative to RAM size.
19:55:24rasherLarsie: Did you make install?
19:55:40rasherHow are you running rockboxui?
19:55:51Larsiein X
19:56:13rasherI mean, how do you run it? Should be ./rockbox
19:56:21Larsieoh ok
19:56:25 Join Lear [0] (
19:56:26rasherWhat were you doing
19:56:39Larsiewhee, that works
19:56:48amiconnjhMikeS: Double the framebuffer memory and you double filling times. Update won't see such a big speedup as on H300. The minimum cycle time of the HD66773 is higher, so DMA needs more waitstates
19:57:14amiconn(Okay, updates at 11MHz would see roughly the same speedup as on H300)
19:57:36 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
19:59:16 Quit TheShadowx ("leaving")
20:00:11 Quit Siku ()
20:01:17jhMikeSDefinitely didn't see a doubling of fill times that's for sure. I implemented memcpy32 for filling blocks and pixels were stored in lcd format. All the graphical plugins ran about the same speed. couldn't really see any slowdown actually. I think pacbox lost about 1-2fps but then full screen tests with test_fps lost about 3fps but no DMA.
20:01:17 Quit Larsie ("CGI:IRC (EOF)")
20:04:10amiconnConverting rgb to pixel format would be rather odd
20:04:27 Join Siku [0] (
20:04:45amiconnThe pixels would have to be in 0000000rrrrrrggg0000000gggbbbbbb format
20:05:01jhMikeSyup but I just stored the green twice to extract it easier
20:07:07rasheramiconn: do you know how the tarballs are built?
20:07:19linuxstb_You mean the source tarballs?
20:07:32rasherYeah. Is it part of the build system, or just some script on the build server?
20:07:55linuxstb_I'm not sure, but I know it uses the apps/FILES and firmware/FILES files, which everyone forgets to update...
20:08:08rasherYeah, that's what reminded me
20:08:11linuxstb_I would expect it to be in tools/ though.
20:09:20rasherMight just be tools/release
20:10:24 Quit Rob2222_ (Read error: 110 (Connection timed out))
20:10:27 Join bagawk [0] (n=lee@unaffiliated/bagawk)
20:13:09jhMikeSI've still got the whole thing for 18-bit in a separate checkout. Could still be built but probably would clash with CVS now.
20:14:32*amiconn wonders how that strange format would be called for LCD_PIXELFORMAT
20:14:48jhMikeSI called RGB6666 =:)
20:15:04amiconnRGB63x36 even
20:15:30jhMikeSBut like I said it had green twice so the unpacking for green would be easier
20:16:13 Quit pixelma (" reboot")
20:18:50amiconnI guess that's 0000.000r.rrrr.rggg.ggg0.gggg.ggbb.bbbb then?
20:19:36amiconnRGGB6666 ;)
20:19:39 Join pixelma [0] (i=pixelma@
20:19:50jhMikeSyup >:)
20:20:14 Quit daurn|afk (Read error: 104 (Connection reset by peer))
20:20:40 Join daurn|afk [0] (n=quae@
20:22:47jhMikeSof course the 0000.000r.rrrr.rggg went in the low word to eliminate a swap
20:23:57amiconnYou mean a swap within lcd_write_data()?
20:25:23jhMikeSit became very tiny
20:25:29amiconnFor DMA operation 0000.000r.rrrr.rggg would need to be the high word
20:26:03jhMikeSeasy to reverse really, just change the packing macros
20:31:02 Join mirak [0] (
20:31:05 Join theprodukkt [0] (
20:31:48 Quit Criamos (Read error: 104 (Connection reset by peer))
20:33:54 Quit daurn|afk (Read error: 104 (Connection reset by peer))
20:34:25 Join daurn|afk [0] (n=quae@
20:34:34 Join qailer [0] (
20:39:46 Join Criamos [0] (
20:39:52amiconnHmm, no firefox2 yet in debian unstable...
20:41:48rasherWon't be.
20:42:30rasherThere'll be IceWeasel at some point, however.
20:42:49theprodukktis it hard to make an own linux distribution containing just some webservises like apache and a firewall in an uml (like ipcop)?
20:43:59 Join perpleXa [0] (n=perpleXa@unaffiliated/perplexa)
20:53:54***Saving seen data "./dancer.seen"
20:54:46Nimdaei'm working on yet another theme :P
20:54:53Nimdaei don't think it'll interest anyone here though
21:03:56 Join funky [0] (n=repulse@unaffiliated/funky)
21:07:53 Join pagefault [0] (n=pagefaul@pdpc/supporter/active/pagefault)
21:10:03 Join CriamosAndy [0] (
21:10:49jhMikeScrap, all this swapping stuff is allocated out the audio buffer and that is the problem.
21:11:28 Quit daurn|afk (Read error: 131 (Connection reset by peer))
21:11:53 Join daurn|afk [0] (n=quae@
21:16:25jhMikeSI can't use audiobufend to get my available memory to allocate chunks cause it scribbles all over the swapped codecs. Why not just declare static uninitialized data for that?
21:16:45 Quit MarcoPolo (Remote closed the connection)
21:20:41LearWhat swapping do you mean?
21:21:01jhMikeSwhen voice is swapped out
21:21:01 Quit lowlight ("CGI:IRC (EOF)")
21:22:12jhMikeSaudiobufend is not the real end of available buffer since audio_reset_buffer has taken some away for swapping and other things
21:22:38 Quit pagefault (Read error: 104 (Connection reset by peer))
21:23:02LearThe voice codec? It needs to be runtime dynamic, doesn't it...
21:23:08 Join Paul_the_Nerd [0] (
21:23:47 Join RogerBacon [0] (
21:23:55jhMikeSIt's swapped out to load the encoder
21:24:59lex_hmm hey
21:25:11lex_can i get a patched ipodpatcher.exe for 5.5G somewhere?
21:25:23jhMikeSOld recording must have just been missing it or something was changed and I missed the commit. But it still misses it by chance
21:25:47lex_i don't feel like to boot a livecd or reinstall my ubuntu :(
21:27:02jhMikeSThe current CVS recording misses it that is
21:27:28LearWould looking at the ringbuf macros help? I.e., base everything on filebuf and filebuflen instead?
21:28:43jhMikeSI don't know. Will have to look at them. Maybe just declare a global that is set to the real end of audio buffer.
21:28:53 Quit Criamos (Read error: 110 (Connection timed out))
21:29:20LearBtw, the last part of the audiobuf is used for the PCM playback buffer (plus the guard buffer).
21:29:47LearAh, codecs before that. :)
21:29:47 Quit linuxstb_ ("CGI:IRC (EOF)")
21:30:05 Quit XavierGr ()
21:30:32jhMikeSWere the codec swap buffers originally just static arrays? I thought I saw that at some point but could be mistaken.
21:31:16 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
21:31:22linuxstb_lex: I _think_ this works:
21:31:25LearHm.. That rings a bell, yes. But maybe only for the IRAM buffer...
21:31:44lex_linuxstb_: thanks :p
21:31:44jhMikeSThe DRAM buffer is the one of concern
21:32:20amiconnjhMikeS: The recording codec should be loaded in place of the main playback codec, shouldn't it?
21:32:35amiconnThe voice codec could stay where it is afaiu
21:33:27jhMikeSI need as much of that as I can grab. I can use the audio codec swap buffer safely.
21:34:11LearChecked the initial voice commit. That used the space after filebuflen too.
21:34:12 Quit daurn|afk (Read error: 104 (Connection reset by peer))
21:34:21lex_linuxstb_: what about ipod_fw.exe? :)
21:34:37amiconnWhat's so different between swapping playback<->voice and recording<->voice?
21:34:38 Join daurn|afk [0] (n=quae@
21:35:58jhMikeSnothing except recording was assuming everything up to audiobufend was available and it sure isn't. even current CVS basically does except for 8000 bytes.
21:36:10 Join pagefault [0] (n=pagefaul@pdpc/supporter/active/pagefault)
21:37:25 Quit daurn|afk (Read error: 131 (Connection reset by peer))
21:37:59 Join daurn|afk [0] (n=quae@
21:37:59 Join gtkspert_ [0] (
21:38:42linuxstb_lex_: In the Rockbox wiki - the IpodInstallation page.
21:38:46 Quit daurn|afk (Read error: 131 (Connection reset by peer))
21:39:19amiconnjhMikeS: The swap buffers are located before the audio buffer afaik
21:39:19 Join daurn|afk [0] (n=quae@
21:39:25lex_linuxstb_: does that work with my 5.5G?
21:39:38amiconnbuffer_alloc() works from the start of buffer
21:40:53jhMikeSno, they're right at the end. I've got it logged and initialzing the chunks was causing debug exception cause I set a guard on dram_buf[1] after the voice was swapped.
21:41:41 Quit pagefault ("I hate ati")
21:41:45*linuxstb_ doesn't like the fact someone added a patched ipod_fw.exe for the 5.5g to the Rockbox wiki, without the patched source or any contact details...
21:42:19rasherLawsuit time!
21:42:59linuxstb_Anyone know JoelSnyder?
21:43:13Kasperlethe patch for ipod_fw is in the bugtracker and referenced in the ipod 5.5g thread in the "new ports" forum
21:43:28*jhMikeS will be the bounty hunter :)
21:44:44pixelmaI think he was on IRC as Joely but I don't know anything else
21:45:00lex_thumbs up!
21:45:16lex_yay it wants to restore itself
21:45:17 Part Stargazey ("Leaving")
21:45:24amiconnUmm, the buffers are indeed allocated at the end...
21:45:25 Quit Blackfish` (Read error: 110 (Connection timed out))
21:45:44amiconnjhMikeS: playback.c lines 3088..3130
21:46:19lex_linuxstb_: i guess it needs a patched ipod_fw joo
21:46:26amiconnI wonder why it is done in this cumbersome way instead of using buffer_alloc()
21:46:58amiconnHmm, maybe that's because it might want to reinit on the fly
21:47:06amiconnCurrently buffer_alloc() is one time
21:47:27*amiconn wonders what happened to the global realloc idea
21:47:45jhMikeSsomehow these have to cooperate :)
21:49:24jhMikeSyeah, 3088-3130 was what I was talking about.
21:49:45linuxstb_lex_: There's a special ipod_fw_5.5g.exe file on the wiki page - that's patched to work with the 5g.
21:50:03 Join MarcoPolo [0] (
21:51:01 Quit gtkspert (Read error: 110 (Connection timed out))
21:52:34 Join nudelyn2 [0] (
21:52:54 Join shattered [0] (
21:52:57jhMikeShmmm...things should be arranged to be able to take as much as possible for recording and just leave the voice codec iram and dram at the end.
21:53:59 Quit bundik ("Quit")
21:54:21jhMikeSlooks to be arranged nicely already.
21:54:34 Quit mirak (Remote closed the connection)
21:56:49Bagderrasher: eqs change for FILES looks fine
21:57:36amiconnjhMikeS: Reusing the main codec iram & dram for the recording codec might allow voice _while_ recording
21:58:18amiconn(a feature blind people would really appreciate I guess)
21:59:30 Join mirak [0] (
22:00:02jhMikeSIt's something I'm aiming at eventually
22:01:06 Quit nudelyn (Read error: 60 (Operation timed out))
22:01:54jhMikeSI should but the IRAM and dram buffer together for each codec then I can write over the audio decoder swap space since it isn't needed.
22:02:38amiconnThat swapping business wouldn't be necessary when linking the voice codec into the main binary
22:02:54amiconn(well, for iram it would still be necessary)
22:03:24jhMikeSWhat about the just have _one_ swap buffer idea?
22:03:46jhMikeSIf no voice is needed then there'd be no swap buffer.
22:04:13rasherBagder: Thanks, I checked the tarball afterwards which had the eqs, so I assumed it was fine
22:05:14amiconnjhMikeS: I wonder why audio_reset_buffer() allocates one buffer when talk_voice_required() returns false...
22:05:56amiconnNow I get 16.5% speedup in lcd_yuv_blit() :-)
22:05:57jhMikeSit should allocate none :)
22:05:58Learamiconn: no voice codec needed?
22:06:17amiconnLear: Yeah, but then _no_ swapbuffer should be necessary
22:06:25LearAh, see your point now... :)
22:06:32amiconnNo swap, no buffer...
22:07:25jhMikeS...and one when voice is required since one is always in running state.
22:07:45amiconnYeah, but then we need a special memswap() function
22:08:24jhMikeSwhy should that be a big deal? just add it to the lib. :\
22:08:29 Quit theprodukkt (Read error: 145 (Connection timed out))
22:09:17amiconnIndeed no big deal. If we require the swapbuffer to be line aligned (position and size), the swap could be done really fast on coldfire...
22:10:34 Quit bawb2 ("seacrest out!")
22:10:37jhMikeSdoesn't it have to be aligned already? so you'd need 16-byte alignment. If we have a memswap function it should be a generally useful one imo. Anything overlapping is undefined behavior of course.
22:11:10amiconnThe codec ram is line aligned. The buffer isn't (yet)
22:11:44Learamiconn: Originally, there was no buffer when no voice. So it can be done. :)
22:11:50amiconnGeneral use memswap() is way more difficult to optimise for speed
22:13:40jhMikeSwouldn't the initial/final byte/word/longword copies be skipped
22:13:49jhMikeSerr swaps
22:14:09amiconnIt's more complex than that
22:14:23amiconnmemset() is simple, as there is only one address, the destination
22:14:41amiconnSo align, do line fills, and do the remainder
22:15:03amiconnmemswap() has the same difficulties as memcpy() and memmove() - there are 2 addresses
22:15:24jhMikeSwas gonna say it's more like that
22:15:30amiconnOnly one can be aligned, the other has to be accounted for all the time
22:15:52amiconnThat's why memcpy() and memmove() are currently monsters for coldfire
22:16:01mirakamiconn that's on h300 the 15% boost ?
22:16:28 Quit menosm___ (Read error: 145 (Connection timed out))
22:16:59 Join bawb2 [0] (n=bawb2@
22:17:07amiconn(but I have an idea how to get them _significantly_ smaller without loosing much speed for unaligned destinations, even make it faster for odd (off by one byte) destination addresses and special cases like dram->iram with bad alignment)
22:17:22 Join kaspar [0] (
22:17:49amiconnmirak: Yes. Note that it is the *synthetic* speedup of lcd_yuv_blit() alone, overall speedup of mpegplayer is smaller
22:17:52jhMikeSI haven't looked at them myself so am not aware of their monterness
22:18:09amiconnI'll also port it to X5
22:18:15 Quit barrywardell ("CGI:IRC")
22:18:25amiconn(even wrote it with the X5 LCD in mind as well)
22:18:34jhMikeSWhat's it worth for the player?
22:18:51amiconnjhMikeS: Take a look. Total size of memcpy() and memmove() is almost 2KB, all asm
22:19:15amiconnSpecial cases for numerous src-dest offsets
22:19:18 Join menosm___ [0] (
22:19:20 Nick menosm___ is now known as menosm (
22:19:39amiconnBut I'll drop that, in favor of copying as long blocks as possible
22:20:01mirakamiconn: you used mac this time ?
22:20:18mirakyou said you had smaller perfs
22:20:34amiconn48 bytes per loop instead of 16, and the algnment/remainders will be done with movem as well (to avoid the move-to-memory pipeline stall)
22:20:54jhMikeS2KB as assembled instuctions or source? :P
22:21:17amiconnmirak: Yes, I use emac. I came up with a clever idea how to reduce the number of mac operations, and also how to fill the emac latency
22:21:26amiconnjhMikeS: as assembled
22:21:56mirakamiconn: what is it ?
22:22:00amiconnSource is almost 54KB
22:23:00 Join lostnihilist_ [0] (
22:24:16jhMikeSguess we don't wanna go there _just_ to swap codecs since it will cost more than the aligment itself.
22:25:09 Join pagefault [0] (n=pagefaul@pdpc/supporter/active/pagefault)
22:25:29 Quit CriamosAndy (Read error: 104 (Connection reset by peer))
22:29:23 Quit lost|X40 (Read error: 131 (Connection reset by peer))
22:34:06LandusIs the daily source from today having problems building the h3x0 firmware, or is just me?
22:34:13lex_linuxstb_: oh, now i found it :)
22:34:27BagderLandus: it has, but I believe its been corrected during the day
22:35:24lex_a hot ipod smells so good
22:35:27LandusHas the scrobbler patch already been applied to the sources?
22:35:37BagderLandus: yes
22:36:30LandusFirst time −− damn it.
22:36:33rasherBagder: any thoughts on my updated voice-building patch?
22:36:42Bagderrasher: looks fine to me
22:37:22rasherBuilding for other languages is tricky. People pretty much have to set their own settings for the TTS by editing the settings file I write
22:37:37Landus 2006.11.03 4-36-51 PM.png
22:37:41LandusCan someone look at that.
22:37:48rasherThey might want to edit that anyway, so I don't think it's a huge problem. Okay to commit?
22:37:50LandusI got the same error message from the build today.
22:37:56LandusI tried using the build from yesterday.
22:38:09amiconnMy idea for building voice files on windows is to have a configure-voice, which reads the available voices and languages from SAPI
22:38:18Landus 4-36-51%20PM.png
22:38:25LandusGah. Forgot a space.
22:38:47amiconn...and then creates a Makefile for those voices the user wants to build
22:39:21lex_rockbox booted
22:39:49lex_ty :p
22:39:56lex_at last
22:39:58LandusYou there?
22:40:09lex_so, good night
22:40:11LandusGuess not.
22:40:20rasheramiconn: Yeah, I was going the route of creating a custom makefile first, but currently I'm just adding a "voice" target to the regular Makefile
22:40:25lex_i'll try to do something with rockbox on my 5.5G :)
22:40:32rasherThat can be changed pretty easily though, so I'll just go ahead and commit
22:40:59amiconnrasher: That voice target should call the created Makefile
22:41:27rasherLandus: get a fresh checkout from CVS or a bleeding edge source tarball
22:41:41LandusThey're compiling right?
22:41:43rasheramiconn: Ah, that can be arranged as well
22:41:49rasherLandus: Yes
22:42:33rasherFairly easily anyway
22:42:37LandusOkay. Hopefully, this'll work.
22:42:52rasherI'll just double-check my voice-building tree and see if I missed something, then commit if not
22:44:48jhMikeSnot sure why we should have any recording related stubs at all if no recording is available :\
22:45:31bluebrotherhmm, someone is facing the "Filetype string buffer empty" message due to viewers.config
22:45:52bluebrotherI had this also some time ago, I think the message is confusing.
22:46:54LandusI love you in a non-romantical way.
22:47:26Learbluebrother: so I changed the Swedish translation. ;)
22:48:20bluebrotherLear, I'm not sure if this is the only reason that causes that message.
22:48:27 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:49:09LandusIs the scrobbler patch applied to the bleeding edge source tarball?
22:49:15bluebrotherbut I hadn't found the motivation to check yet. When I got into that problem I simply removed some other lines I didn't need then.
22:49:32bluebrotherLandus, it has been commited a week or so ago.
22:49:42LandusI don't keep up with the updates.
22:50:00LandusI think the build I'm using is based off the source from two months ago.
22:51:28bluebrotherurgh. That's old :)
22:52:09 Join bundik [0] (
22:53:30LearI think it is. Though the use in get_string isn't the most obvious one, so I might've got that one wrong. :)
22:53:59***Saving seen data "./dancer.seen"
22:54:05 Quit amiconn (Nick collision from services.)
22:54:05 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:55:05*bluebrother is reading that funny logo discussion on the forums since a couple of minutes
22:56:12Bagderrasher: tools/FILES needs attention as well
22:56:18BagderI don't think it includes .sh files
22:56:44 Join Blackfish [0] (
22:56:48*jhMikeS wonders what damage will ensue if codecs go at the end of audiobuf instead
22:57:06 Join XavierGr [0] (
22:57:08 Join Blackfish` [0] (
22:57:41BagderjhMikeS: on ARM, we want the codecs near the kernel since there are 64MB ram targets and we want to be able to avoid the long jump
22:58:14Bagderbut we don't even support the 64MB ram properly yet...
22:58:16jhMikeSfor the swap space?
22:58:43 Quit kaspar (Read error: 110 (Connection timed out))
22:58:52jhMikeSsorry if I wan't clear :)
22:58:59rasherBagder: oh shoot, I checked it for wavtrim.c, but forgot those
22:59:36rasherBagder: should I include *.sh or just the specific files?
22:59:58 Join sucka [0] (
23:00:00Bagderrasher: use *.sh, as then you can remove the two specific ones
23:00:59 Join Criamos [0] (
23:01:51 Quit daurn|afk (Read error: 131 (Connection reset by peer))
23:02:46 Join daurn|afk [0] (n=quae@
23:05:50jhMikeSIt's just that right now it's filebuf->codec swap->other stuff and I want filebuf->other stuff->codec swap
23:06:43Bagderah, you talk about the swap area
23:06:58Bagderwell that can be whereever, afaik
23:07:36jhMikeSyeah. would like as much buffer as possible for recording so anything up to the voice swap buffer should be available
23:08:13 Quit daurn|afk (Read error: 104 (Connection reset by peer))
23:08:30 Join daurn|afk [0] (n=quae@
23:12:18 Quit daurn|afk (Read error: 104 (Connection reset by peer))
23:12:44 Join daurn|afk [0] (n=quae@
23:16:19LandusSo. Last two months.
23:16:25LandusAny new features I might want to know about?
23:16:47Paul_the_NerdLandus: The changelog is public information, y'know.
23:17:07 Quit MarcoPolo (Remote closed the connection)
23:17:09 Quit actionshrimp (Read error: 110 (Connection timed out))
23:17:19LandusWhich is two months of CVS notes to look at.
23:17:36LandusIf there was anything mentionable, at least one person here would know.
23:17:41bluebrotherhmm, the MajorChanges wiki page is pretty much outdated :(
23:18:15Paul_the_NerdLandus: And imagine if everyone came in here and asked for what's new on a frequent basis. It's better if we just get you in the habit of doing it yourself, like we have the others.
23:18:16 Join MarcoPolo [0] (n=MarcoPol@
23:18:32LandusBut I don't.
23:18:49LandusUnless two or three months is frequent.
23:18:53Paul_the_NerdIt doesn't matter whether you do it regularly or not, it's more or less a widespread policy.
23:19:06Paul_the_Nerd"Those who help themselves are served by the best suitable."
23:19:18 Join petur [0] (n=petur@rockbox/developer/petur)
23:19:30bluebrotherLandus, it's the same thing as this "how to ask questions a smart way".
23:22:25jhMikeSman all this allocation is spread out everywhere and not contained in one spot. must be careful.
23:24:08bluebrotherLandus, at least quite a bunch of bugs has been fixed, so upgrading should be worth anyway.
23:26:07 Quit RogerBacon (Remote closed the connection)
23:27:03 Quit MarcoPolo (Remote closed the connection)
23:27:19 Join dau [0] (n=quae@
23:28:08 Join MarcoPolo [0] (
23:28:19 Quit dau (Read error: 104 (Connection reset by peer))
23:29:19 Quit nls (" HydraIRC -> <- IRC with a difference")
23:29:56 Join dau [0] (n=quae@
23:30:53 Quit qailer ("Leaving")
23:36:09 Quit TeaSeaLancs (Read error: 110 (Connection timed out))
23:36:44 Join TeaSeaLancs [0] (n=thunderc@
23:40:19jhMikeShmm guess most audiobuffer usage isn't sensitive to changes
23:43:09 Quit daurn|afk (Connection timed out)
23:43:40 Quit bundik ("Quit")
23:44:52amiconnHmm, only 8% speedup on X5 :/
23:46:07 Quit MarcoPolo (Remote closed the connection)
23:46:20 Quit Lear ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
23:47:05 Join MarcoPolo [0] (
23:51:23jhMikeS8% is marginally better than <=7.9999...% :D
23:53:52 Part Paul_the_Nerd
23:55:51 Join menosm___ [0] (

Previous day | Next day