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

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

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

#rockbox log for 2007-02-13

00:00:13 Join Abcminiuser [0] (
00:02:07ildellait works :)
00:02:22 Quit bluebrother ("time for sleep")
00:04:07Nico_Pmarkun: i hadn't had a look at how the wps is made but it looks a bit strange to me
00:04:44 Quit entheh ("^~")
00:05:53markunNico_P: I will fix it, just need some time. Until now it was just copy paste :)
00:06:52 Quit linuxstb_ ("CGI:IRC")
00:07:00Nico_Pmarkun: what i find weird is the pg-220 bitmap
00:07:21Nico_Pwhy does it need to be this big with all this pink ?
00:07:36Nico_Pit makes the cuesheet markers ugly
00:08:19 Quit midkay ("Leaving")
00:09:01markunNico_P: that was what I was trying to ask (why are some of the pg- bitmaps taller than the pgln- bitmaps?)
00:09:46Nico_Pmarkun: well it might be worth trying to trim them down a bit :)
00:12:15 Join ricardo [0] (i=bd8766dd@gateway/web/cgi-irc/
00:12:25 Part ildella
00:13:40ricardomy ipod died
00:13:44markunNico_P: I guess the the gap between the status bar and the text will not look so good
00:13:49markunricardo: that sucks
00:14:03markunricardo: do you think you can still repair or are you looking for another player?
00:14:23Nico_Pmarkun: you're saying it's like that for alignment reasons ?
00:14:27 Part n1s
00:14:32markunI think so
00:14:50*Nico_P whispers "y-coord"...
00:15:11*markun wispers viewports back
00:15:29*Mikachu mumbles "millenium hand and shrimp"
00:15:38Nico_Pviewports are a different matter
00:16:00markunwhy would they be?
00:16:07Nico_Peven LinusN agreed we could add a few cosmetic patches whilst waiting for viewports
00:16:28Nico_Pfrom what i understood they are more of a split screen feature
00:16:32markunI think we should just push amiconn hard enough so he does it :)
00:16:43markunhi amiconn ;)
00:16:57 Quit Psilonaut ("Gone trippin")
00:17:49Nico_Phow would viewports change this particular problem ? by having a viewport just for the progressbar ?
00:18:08markunNico_P: what will happen if the y-coord for the pb is lower than the text that came before it and after the pb is more text?
00:18:54markunNico_P: no, only a viewport for the text in the center and above the progressbar I would think
00:19:00markunso 3
00:19:26Nico_Pmarkun: i've never tried that... in fact it isn't my patch, i just rewrote it in a simpler way
00:19:44Nico_Pi suppose it's up to the WPS creator
00:20:05 Join barrywardell [0] (
00:21:33ricardomy ipod doesn't turn on anymore
00:22:48 Quit ender` (" printk("autofs: Out of inode numbers -- what the heck did you do??\n"); -- /usr/src/linux/fs/autofs/root.c")
00:23:54Lloreanricardo: Usually this simply means that it needs charging and/or the reset trick
00:23:54 Quit ricardo ("CGI:IRC (EOF)")
00:29:37*Mikachu wonders how many perfectly fine ipods are thrown away every day
00:30:09peturbut so are many other perfectly fine things
00:30:28peturtoo many idiots in this world
00:32:30 Quit funky ("leaving")
00:32:49 Join toed [0] (
00:33:23toedjust wondering if there's a rockbox plugin that allows one to play go (the boardgame) or to work on go problems
00:33:25 Quit linuxstb (Read error: 110 (Connection timed out))
00:33:27toed(it's hard to search for go)
00:33:36 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
00:34:27 Join alkaline [0] (
00:35:15Lloreantoed: I don't believe so, no.
00:35:22 Quit Mouser_X (Nick collision from services.)
00:35:31 Join Mouser_X [0] (
00:35:39toedah, that's a real shame
00:35:56peturfeel free to contribute ;)
00:36:11 Quit mirak (Remote closed the connection)
00:36:14LloreanI'd certainly be interested in seeing someone make one. ;)
00:36:27markuntoed: are there any good GPL go engines?
00:36:47toedyeah, there's gnugo
00:36:55markunis it any good?
00:37:02LloreanIt's not bad.
00:37:05alkalineHey all i have a little prob with a ipod nano it won't start, but once plugged into a computer all it does is restart and restart, i tried the hold switch and menu and select trick but alas no good
00:37:07toedit's quite good, I think it's rated around 15k
00:37:26LloreanIsn't 7k assumed to be about the limit of computers anyway?
00:37:34toedI think so
00:37:51Lloreanalkaline: Select+Play for disk mode.
00:38:09toedbut AI isn't even necessary, if you had something that lets you play around on the board (thus allowing you to do problems) that would be wonderful as well
00:38:10 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
00:38:14toedI'd contribute if I could
00:38:35alkalineLlorean: use the hold switch as well? and should it be connected to computer for disk mode?
00:39:14Lloreanalkaline: No hold switch, force a reboot with Menu+Select (or while plugged into a computer) then hold Play+Select early on in the Apple logo
00:39:40 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
00:43:33 Join miepchen^schlaf [0] (
00:45:35barrywardellI can't go to page 2 of my watched tasks in fs.
00:46:14petursolution: stop watching so many tasks
00:46:25Lloreanbarrywardell: Resets to all tasks?
00:46:34 Quit Abcminiuser ("User pushed the X - because it's Xtra, baby")
00:46:40 Quit busa_blade ("Leaving")
00:46:46LloreanSame here
00:46:49barrywardelli can manually add tasks=watched to the url, but that's kinda annoying
00:46:57LloreanI've never used the watched tasks feature before.
00:47:05barrywardellwell, &tasks=watched
00:47:28alkalineSeems like the usb cable is shot, the disk mode launches, and sometimes rockbox loads but than it fails. and restarts i dont think it's giving the charge or it's cut
00:47:47 Quit Nico_P (Remote closed the connection)
00:48:17LloreanRockbox shouldn't launch if you're in disk mode...
00:49:23alkalinei try disk mode the bottom displays ok to disconnect, a check than switches to a no battery charge
00:49:25 Join midgey [0] (
00:50:33alkalineno battery charge
00:50:37alkalineat all
00:50:44alkalinebut it keeps restarting
00:50:45LloreanSo it won't charge in disk mode?
00:51:17LloreanThat suggests a hardware problem of some sort, yes.
00:51:23alkalinesigns change from a check, no battery, cross, than check,
00:51:25LloreanEither the cable, or one of the ports
00:52:13alkalineProbably, I just am lucky it isnt mine lol
00:52:20 Quit printfXh4 (Read error: 104 (Connection reset by peer))
00:52:33 Part toed
00:53:11 Quit lee-qid ("aufwiederbyebientotsayonara")
00:54:04alkalinealright now a huge battery is on screen i'll assume it's probably charging
00:54:14 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
00:55:08 Join Supachikn_ [0] (
00:56:15alkalinewell thanks Llorean for the help, I now know what to do once it does get a charge
00:56:36 Quit matsl (Remote closed the connection)
00:56:37Lloreanalkaline: Good luck
00:57:12 Quit alkaline ("Chatzilla 0.9.77 [Firefox]")
01:05:29 Join midkay [0] (n=midkay@rockbox/developer/midkay)
01:06:21 Join printfXh4 [0] (n=pseudo@
01:06:35 Join webguest75 [0] (i=478b1325@gateway/web/cgi-irc/
01:06:42 Quit obo ("bye")
01:07:03 Quit webguest75 (Client Quit)
01:07:07JdGordonanyone know how to checkout svn at a certain revision?
01:07:08 Join webguest75 [0] (i=478b1325@gateway/web/cgi-irc/
01:07:36 Quit webguest75 (Client Quit)
01:10:45 Quit Supachikn (Read error: 110 (Connection timed out))
01:10:53 Nick Brunellus__ is now known as Brunellus (
01:10:53safetydanJdGordon: svn update -r 1234
01:11:09safetydancheckout probably accepts the same switch
01:11:33JdGordongoogle said −−revision
01:11:39linuxstbOr (I think, reading the docs) svn co svn://
01:12:00***Saving seen data "./dancer.seen"
01:12:19linuxstb"svn help co"
01:13:07JdGordonbah, i was trying svn co help and getting errors
01:13:13 Join MonkeyTamer [0] (
01:13:35linuxstbI always do that first...
01:14:04Mikachuor if you use zsh you can just tabcomplete everything :)
01:16:07 Quit Arathis ("Bye, bye")
01:16:38JdGordonamiconn: I just checked out the revision before the menu code started going in, and we are only 1600 bytes above the rec build size it was then... and going by last nights commit, that should be easily got by the time im done :)
01:18:55safetydanAnd I think the lang cleanup savings around around 2000 bytes on the recorder. So we should be well under.
01:19:26*linuxstb goes to flyspray to look for pointless new features :)
01:20:01LloreanThe recent plugin patch is 'pointless'
01:20:22LloreanIt's only on H100 anyway, though. Heh
01:20:44linuxstbI would guess that most people only use about 5% of Rockbox anyway... Problem is that it's not the same 5%.
01:20:55JdGordonLlorean: are yoou talking about the one so you can shot and move in invadrox?
01:21:13LloreanJdGordon: No, the one that makes the SPDIF output flash in time with the VU Meter.
01:21:47JdGordonawesome :D
01:21:52*linuxstb misread the VU meter task as a bug... :)
01:22:16 Quit midgey ()
01:22:27JdGordonim fixing keywords now... anyone know any files that dont expand $Id$ anymore?
01:22:36JdGordonapart from the files ive added recently
01:22:50Lloreanlinuxstb: If it were a feature request, I would've just rejected it I think. As it stands now, I wish there was a reason for closing 'silly'
01:23:05 Join webguest37 [0] (i=ce74ba60@gateway/web/cgi-irc/
01:23:08linuxstbPlugins are the place for pointless features though.
01:23:32LloreanYeah, I guess it doesn't really hurt anything
01:24:10LloreanBut I guarantee, were that to get committed, someone would bring up "But you added so and so, why can't you add this?" and it'd have to be explained once more that plugins are 'free'
01:24:21 Nick webguest37 is now known as yun (i=ce74ba60@gateway/web/cgi-irc/
01:25:10 Join toed [0] (
01:25:54MonkeyTamerI personally just find it amazing Rockbox is as small as it is
01:26:08JdGordonits not as small as it should be...
01:26:14JdGordonwhich is casuing problems :p
01:26:26MonkeyTamerseems to be the case as I'm discovering
01:26:34MonkeyTamerfor certain platforms?
01:26:35toedis it possible to play or port ipodlinux games to rockbox? specifically, I'm looking at
01:27:23safetydantoed, it's possible. Someone just has to step up and do it.
01:27:37Mikachuit's not much easier than porting something else to rockbox
01:27:40linuxstbtoed: You can port them, which will basically mean adapting to use the Rockbox plugin API instead of the Linux/POSIX API. But the Rockbox plugin API attempts to follow POSIX.
01:28:16 Quit petur ("Zzzzz")
01:28:38toedI see, I imagine it's not a trivial thing to do though?
01:29:17JdGordonits pretty trivial
01:29:23safetydantoed, depending on your level skill with programming, it may or may not be.
01:29:24linuxstbNot trivial, but not that hard once you understand how Rockbox works.
01:29:26JdGordonthe worst part is any drawing code
01:29:40toedoh, I guess I'd have to port gnugo as well
01:29:47Mikachuit's trivial when you know how :)
01:30:27 Part barrywardell
01:30:27Mikachua math prof once spent 5 minutes in a lecture explaining why something was trivial
01:30:54MonkeyTamerthat's one of my math professor's favorite words...
01:31:32 Quit Mouser_X (Nick collision from services.)
01:31:41 Join Mouser_X [0] (
01:31:42toedhaha, I got something like that today; after a 20 minute proof the guy says the argument is elementary assuming measure theory
01:33:49 Quit yun ("CGI:IRC (EOF)")
01:34:35MonkeyTamernice, ipodpatcher saves me from typing the 2 associated with the ipod drive
01:35:02Mikachuhappy days are here again
01:35:32MonkeyTamerbut at a cost... what of the the need to make things smaller?
01:37:15 Join courtc [0] (
01:38:53MonkeyTameryay, happy days for me; an evident freezing bug in some earlier versions is no longer afflicting me
01:38:56 Quit jhMikeS (Nick collision from services.)
01:39:02 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:40:51LloreanMonkeyTamer: Which iPod?
01:41:27MonkeyTameripod color
01:41:52MonkeyTamerversions 70-80something were freeezing the ipod
01:42:09MonkeyTamerwith kop and without, but not anymore
01:42:29linuxstbI think that's just luck - some builds seem to (without obvious reason) work better than others.
01:42:43LloreanWell they shouldn't have been freezing _with_ KoC though
01:43:07MonkeyTamerthat's what was strange, but that was only for 1 day's worth of builds
01:43:34linuxstbLlorean: AFAIK, KoC doesn't completely fix the freezing.
01:44:09linuxstbAlthough possibly it's related to gcc version - the build server has just been upgraded to 4.0.3...
01:44:24MonkeyTamermy personal experience is for my ipod color, I've had zero freezes with kop with the exception of yesterday; the only exception is when I applied w1ll1am's patch to test it
01:44:57MonkeyTamerbut that of course is completely unrelated to kop
01:45:25 Quit [toffe] ("MegaIRC v3.27")
01:47:35 Quit wooo (Read error: 110 (Connection timed out))
01:47:39 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
01:48:05Lloreanlinuxstb: Ah, I'd heard the freezes had stopped with KoC6/8
01:48:16 Quit |Rincewind| ("Cya")
01:49:59BiptoNany new work on the recording issue with KoC yet? or the weird noise?
01:50:19LloreanWeird noise?
01:50:51 Join miepchen^schlaf [0] (
01:51:47 Quit Mouser_X (Nick collision from services.)
01:51:57 Join Mouser_X [0] (
01:53:15 Quit MonkeyTamer ("Chatzilla 0.9.77 [Firefox]")
01:53:48LloreanBiptoN: Weird noise?
01:55:39*Llorean sighs.
01:58:57 Join TrueJournals [0] (
01:59:00BiptoNrecording with the ipod
01:59:27BiptoNthere is a weird noise that comes through on one channel on the 5.5g an dalot more noise on the 4g
01:59:59LloreanIs there a bug report on this?
02:00:09BiptoNi dunno
02:00:18BiptoNi outta submit one
02:00:43LloreanIf there's not already one, then yes.
02:00:56LloreanDocument how to reproduce it, and maybe include a sample or two.
02:01:14BiptoNi can do that
02:01:27 Part TrueJournals
02:01:51linuxstbipod recording shouldn't really be called anything more than experimental at the moment...
02:02:18BiptoNthat's why i haven't submitted a bug report
02:02:22 Quit My_Sic (Read error: 104 (Connection reset by peer))
02:02:29BiptoNi know you and barry were just gettin it started
02:05:25 Part pixelma
02:15:53 Join Jsunu [0] (
02:22:57 Quit Mouser_X (Nick collision from services.)
02:23:06 Join Mouser_X [0] (
02:28:17 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
02:29:43 Join midgey [0] (
02:30:46 Join bonbonthejon [0] (n=jon@
02:37:05 Join [toffe] [0] (n=[toffe]@
02:37:41*Llorean thinks SPC should be committed just so it doesn't need resyncing daily. :-P
02:38:51Lloreanmidkay: What compiling problem were you having with VMWare?
02:39:02midkayLlorean: you're on the ball today. :)
02:39:16LloreanWell, that task was on my watch list.
02:39:17midkayi can't remember what it said.. some odd error, let me restart it.
02:39:35LloreanWere you compiling an ARM target?
02:39:38midkayi searched around to see if anybody had any siilar problem, couldn't find anything like it.
02:39:49LloreanIs your arm-elf-gcc 4.0.2 or 4.0.3?
02:39:57 Join miepchen^schlaf [0] (
02:40:12midkayi did look at the build log - everybody's running 4.0.3 now.
02:40:19LloreanIt's required now
02:40:25LloreanBut the new vmware image has it.
02:40:28midkaybut i didn't see any commit messages that introduced a requirement for it..
02:40:39LloreanOr you can jump through a couple really minor hoops to install it on your older one
02:41:00LloreanWell, if I recall, we didn't realize the commit required 4.0.3 (because many/most servers had it anyway) until after it went in.
02:41:20midkayremember which one it was?
02:42:05midkayno prob, just curious.. thanks for pointing it out. (i did check the vmware page when i got the error, too. it hadn't yet been updated..)
02:42:12midgeyi think it was jdgordon's initial menu commit
02:42:40midgey4.0.1 builds fine here on my mac
02:42:56LloreanSo it's just 4.0.2 that doesn't work?
02:43:10midgeyi believe so, well at least i have no problems
02:44:40LloreanI do remember some mention of it being a bug in 4.0.2, but I wasn't sure if that was the same thing
02:45:44*safetydan is all for comitting partially working codecs :)
02:46:08safetydanThe public embarassment of having it not work is even more motivation to fix it
02:49:06LloreanIt's not even 'partially working' in my experience.
02:49:33LloreanIt 'works fine except with unreliable metadata' I would say. Which is kinda the state of AAC anyway, except I think this works better than it. :-P
02:49:54 Join brenton_ [0] (
02:50:04 Part brenton_
02:50:07safetydanWell I'm redoing speex.c to use Tremor's ogg parser so hopefully that will fix a few things.
02:50:07LloreanThough I understand it could use a good deal of optimizing still
02:50:23midgeyi thought it wasnt fullspeed on quite a number of songs?
02:51:02Lloreanmidgey: Neither is AAC though, on most targets.
02:51:12midgeytrue enough
02:51:38LloreanI think SPC's at the point where putting it in SVN means less time ever spent simply resyncing it.
02:52:14Lloreansafetydan: I still haven't managed to reproduce the pops on any but that one file.
02:52:18LloreanIt really irritates me.
02:53:20LloreanCan speex files be cut easily? Anyone know?
02:53:42safetydanWell they're just ogg streams, so if you can find an ogg cutting tool
02:54:10LloreanI was thinking of cutting around one of the noises and attaching it to the patch if the cut file still showed the symptom.
02:54:15Lloreanerr to the tracker entry
02:55:02*midgey rocks out to Turtles in Time
02:55:16midgeyspc is working a lot nicer than the last time I tried it
02:55:41LloreanI've been using it a lot, though on a gigabeat where performance isn't really able to be judged.
02:55:52safetydanLlorean, That'd be good actually. Though have you tried playing it in the sim? There's some debug code in there that might be printing out errors if there's something up with the stream.
02:56:17bkon linux you can use ogmtools ( to split/merge Ogg streams
02:56:24Lloreansafetydan: I haven't tried it with the sim. I can't get audio to work in debian, and haven't got the tools to cross compile for windows set up.
03:08:09safetydanah well, if we can just get a stream snippet that reproduces the issue that'll be enough
03:08:31LloreanWell, next time Rasher's sims update I'll use one of those and try it.
03:11:55 Quit bonbonthejon (Remote closed the connection)
03:12:01***Saving seen data "./dancer.seen"
03:13:52 Quit lukaswayne9 (Read error: 104 (Connection reset by peer))
03:16:47 Join lukaswayne9 [0] (
03:19:41 Join bonbonthejon [0] (n=jon@
03:22:27 Join twisted` [0] (
03:24:18twisted`ey, question, is it absoletely impossible to get rockbox on a iPod that's HFS+ formatted?
03:24:58JdGordon well... you can install the bootloader... but thats it
03:25:53Lloreantwisted`: Is there some terribly compelling reason not to convert it?
03:28:34twisted`Llorean: yeah, no windows machine :)
03:28:54twisted`and fat32 sucks since I have files bigger then 4gb on my iPod stored sometimes too
03:29:51 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
03:31:25Lloreantwisted`: Converting it doesn't require a windows machine.
03:31:40LloreanBut if you don't want to convert your iPod, you'll have to convert Rockbox, or live without it.
03:32:45 Join nickv111 [0] (
03:34:10nickv111Hey guys. I'm having trouble with Rockbox on my 3g iPod. While I can play music and all, every few seconds, it stops playing, and then resumes
03:34:14nickv111It's a little choppy
03:34:33LloreanYes, the 3G iPod has some performance issues inherent in the hardware design.
03:34:40LloreanWe haven't found a good way to solve that problem yet.
03:34:45nickv111I see.
03:35:07nickv111Is that also the reason for the bad scroll wheel response?
03:35:17LloreanSomewhat, yes.
03:38:43nickv111Overall, I'm pretty happy with Rockbox, though. I may get another iPod for running it. Any plans on getting the Nano 2g working? Last I checked, nobody was really dedicated to it. Same story now?
03:39:46LloreanNobody has one.
03:39:53nickv111I see.
03:39:56LloreanIt's also a fairly large amount of work.
03:40:04nickv111What needs to be done?
03:40:29LloreanFigure out how to decrypt the firmware, figure out how to load our own firmware, write drivers for a large number of new undocumented pieces of hardware...
03:40:31LloreanBasically, everything
03:40:35LloreanIt's a whole new port, from scratch
03:40:58nickv111If I had a Nano, I'd offer to help
03:41:05nickv111But, then again, I don't.
03:41:41*nickv111 ponders what he could do to help
03:42:01LloreanReally it needs probably several very dedicated people with the hardware.
03:42:21nickv111I know of one person with a 2g Nano, but I doubt he'd be willing to mess with it too much
03:42:21LloreanIf you want a modern flash based player for Rockbox, wait until the Sansa e200 has audio, probably
03:43:12safetydanCan the 3g nano play smoothly without the peak meters?
03:43:24Lloreansafetydan: I think _sometimes_
03:44:28nickv111So what's usually the process to start a new port?
03:44:32nickv111I mean, I wouldn't even know where to start
03:44:55nickv111Crack open the case and see what kind of components it has?
03:45:10nickv111Look up specifications?
03:45:16nickv111I'm just seeing where I'd even begin.
03:46:39LloreanWell, the first two things that need to be done (really in either order) is figure out exactly what all the hardware is in it, and gather datasheets for any that's documented, and figure out how to load your own firmware onto the device (what is necessary, are the files encrypted/signed, etc)
03:47:19nickv111To load your own firmware onto the device. . . Isn't that just overwriting your boot partition?
03:49:30LloreanThat just puts your firmware in place.
03:49:51LloreanBut for example on the 2nd generation Nano, the bootloader expects the firmware to be encrypted in a certain way
03:50:18LloreanSo, we need to either figure out how to encrypt our firmware in the right way for the bootloader to load it, or figure out if there's a way like on the Sansas for us to tell it to expect an unencrypted file
03:51:03nickv111Hmm. . .
03:52:09 Join TrueJournals [0] (
03:55:58nickv111Hey, I'm working on calculator.c. If I make any improvements, to whom to I give them so they can be committed?
03:56:27nickv111to whom do*
03:58:02TrueJournalsFlySpray... the patch tracker
03:58:04LloreanYou post a patch to the patch tracker
03:58:22nickv111Thanks for the help
03:58:25 Quit Mouser_X (Read error: 110 (Connection timed out))
04:02:11 Join maeck [0] (
04:03:18twisted`Llorean: so Rockbox doesn't support it? cause I couldn't really figure out what the cause of it is anyways...
04:09:31Lloreantwisted`: I don't know what you're asking.
04:09:41LloreanDo you mean "HFS+" by "it"?
04:11:29 Join scubacoles [0] (
04:13:05scubacolesIs there a reason why the .rockbox folder is built as a hidden file?
04:13:49TrueJournalsscubacoles: some operating systems hide folders with their first character being a .
04:13:50LloreanSo it doesn't show up as clutter during normal browsing, and so it's less likely to accidentally be modified on the device?
04:14:25TrueJournalsWhat Llorean said... :-p
04:15:03LloreanGenerally speaking, a user shouldn't be browsing into it during normal use.
04:18:23scubacolesMy annoyance with it is more due to MacOS Finder's insistence on overwriting the entire folder in an upgrade rather than just overwriting each component within the folder.
04:19:04LloreanWhat's the difference?
04:19:10LloreanDoes it delete files that aren't present in the new one, or something?
04:19:29scubacolesI have to view Hidden Folders and manuall drag across all but the wps, themes and config folders, otherwise I lose all data in those folders...
04:19:57midgeyfinder likes to delete the whole folder and then add the new one
04:20:01midgeyits quite annoying
04:20:02scubacolesOne thing that Windows Explorer does much better
04:20:31LloreanBeing unhidden would fix that for some reason?
04:20:45scubacolesand to view hidden folders, you need a 3rd party app
04:20:54midgeycommand line works the best here
04:21:09scubacoleswithout viewing hidden folders you can't browse to .rockbox
04:21:52midgeyyou can do "go to folder" in finder
04:21:59scubacolesyeah command line would work fine, but I'm very much a command line novice...
04:22:05midgeyits just not convenient
04:23:14scubacolesI'll try the "go to folder" method on my next update.
04:23:39 Part TrueJournals
04:23:39LloreanYou could also just extract directly to the device, couldn't you?
04:23:44 Join TrueJournals [0] (
04:24:32scubacolesExtracting to the device empties the wps, themes and config folders
04:25:00LloreanThat's a horrible... well... whatever does that.
04:25:25scubacoleswell using Stuffit to extract does anyway, I haven't tried any other archive utilities (if they exist)
04:25:56scubacolesmaybe .rockbox could not include those folders??
04:26:16TrueJournalsHey, I just wanted to let everyone know about my GUI program to encode videos for rockbox. Check it out at
04:26:22Llorean .rockbox doesn't have a config folder.
04:26:23TrueJournals(Just updated it...)
04:26:35LloreanAs for themes and wpses, how would you suggest we go about distributing the default ones?
04:26:42midgeyscubacoles: honestly your best bet is terminal (command line)
04:26:56LloreanIt seems that Rockbox shouldn't be blamed for bad tools. :-P
04:27:08midgeylets say is your build and your device is at /Volumes/H300
04:27:13midgeythe command is
04:27:24midgeyunzip -d /Volumes/H300
04:27:40scubacolesTrying to talk a world of Windows (and *nix) users to cater for a flaw in Finder isn't going to work is it!!!
04:28:00LloreanWhy not try to talk Apple into fixing the flaw?
04:28:05*midgey dislikes finder
04:28:06scubacolesmidgey: Thanks
04:28:12 Join tanq_ [0] (
04:28:43scubacolesLlorean: I wouldn't be the first, Finder is the worst thing about MacOS
04:29:57scubacolesconfig folder = backgrounds folder.... I knew there was a 3rd folder, just couldn;t remember what it was
04:33:20twisted`Llorean: well, why Rockbox cannot be installed and used on HFS+ :)
04:33:29 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:33:35 Join busa_blade [0] (
04:34:32Lloreantwisted`: Because there's no support for reading an HFS+ filesystem.
04:34:53 Part TrueJournals
04:36:18safetydanLlorean, it's not really a flaw. It's always been like that on Apple OS's
04:36:25busa_bladeanyone have a f40 gigabeat?
04:36:52Lloreansafetydan: It seems silly, at least
04:36:53Lloreanbusa_blade: Yes.
04:37:01*twisted` cries
04:37:29safetydanIf you think about it makes sense. It basically treats directories like files and allows you to overwrite them. Though this particular feature is a source of major holy wars between the Windows way and the Mac way.
04:37:38busa_bladeLlorean: what OS did you install rockbox from?
04:37:43Lloreantwisted`: Convert your iPod, keep a large secondary HFS+ partition for your giant files.
04:38:22Lloreansafetydan: It seems to me the simplest solution would be "Do you wish to overwrite the whole folder, merge contents giving the new folder priority, or cancel?"
04:38:24safetydanI'd also support having user configurable stuff outside of .rockbox. You should be able to blow away and restore .rockbox without affecting your settings.
04:38:26Lloreanbusa_blade: Windows.
04:38:32twisted`that would defeat the purpose of having space available on my iPod :)
04:38:48safetydanLlorean, that's been argued before. The Mac faithful like it as it is :)
04:39:22 Quit tanq (Connection timed out)
04:39:24Lloreansafetydan: No option to preserve the files?
04:41:27safetydanLlorean, nope. Part of the reason is that when you install applications on OS X, you just drag the application folder where you want it. If you merged the contents of the old and the new, you'd end up with a broken app.
04:41:44 Join MonkeyTamer [0] (
04:42:23*safetydan gets back on topic
04:43:07Lloreansafetydan: Honestly, it seems to make more sense to require a manual delete first than to require a much more tedious manual merge, but I come from a windows background so while I'm biased in favour of 'less work' I may not see it properly
04:46:27safetydanIt is less work for that one scenario, since it saves you from having to delete the folder first. If you actually want to merge the contents, then you're on your own in the Mac world.
04:47:00LloreanIs deleting a folder on mac as simple as clicking it, hitting a key, and choosing "Yes"?
04:48:00safetydanLlorean, I would assume so. Though I imagine most people would drag the folder to the trash rather than use the keyboard.
04:48:04safetydanmac people that is
04:48:20nickv111Hey, when a plugin does something like, "#include "math.h"", is that relative to firmware/include?
04:49:23safetydannickv111, it's relative to a few places. I forget the exact order, but current directory is one of them, along with firmware/include
04:49:33nickv111Okay. Thanks
04:49:47nickv111Well, calculator.c seems to define a few constants that are already defined in math.h. . .
04:50:55LloreanWith more or less precision?
04:51:35nickv111I also defined LN10 until I realized that there was M_LN10
04:51:41 Quit MonkeyTamer ("Chatzilla 0.9.77 [Firefox]")
04:51:42nickv111And PI is redundant because there's M_PI
04:53:25safetydannickv111, you can always try including math.h and find out
04:53:40nickv111#include "math.h" is already there
04:53:58nickv111So whoever coded this defined PI and such, instead of using math.h's constants
04:54:34SoapIs calculator older than those defines in math.h?
04:54:47nickv111Maybe. Whatever is the case, I'm fixing it
04:54:53nickv111I'm also coding in logs and such
04:55:02*safetydan is slightly dismayed to see that there's a cordic function in calculator.c
04:55:15safetydanIf I'd seen that I could have saved myself some googling
04:55:16safetydanoh well
04:55:33nickv111I don't quite understand the CORDIC function
04:55:45 Join miepchen^schlaf [0] (
04:55:54nickv111Why do they have a table of y=2^-x and y=atan(2^-x)?
04:55:59safetydannickv111, the calculator plugin hasn't had any calcuation related changes for nearly 2.5 years, so it's probably just bitrot resulting in PI being used instead of M_PI
04:56:16nickv111I _kind of_ understand CORDIC
04:56:24nickv111Not well, though, because my trig skills are seriously lacking
04:56:29nickv111I haven't even gotten to trig in school yet :-\
04:59:24nickv111Hmm. I understand a bit
04:59:43nickv111Good thing is, I understand stuff about complex phases
05:00:04nickv111When it says j, it means the imaginary unit, right?
05:00:08safetydanActually, Wikipedia's article is pretty good
05:00:14nickv111I'm used to it being i.
05:00:25safetydanyeah j tends to get used in engineering circles instead of i for the complex part
05:01:41safetydanIf you want to expand to other mathematical functions, cordic is a good base
05:02:12*safetydan goes back to his real work
05:10:11 Join Xerion_ [0] (
05:10:41nickv111Wow, I don't unerstand CORDIC much at all.
05:11:07 Quit Xerion (Read error: 104 (Connection reset by peer))
05:11:07 Nick Xerion_ is now known as Xerion (
05:12:06***Saving seen data "./dancer.seen"
05:12:46nickv111First of all, why is the rotation matrix what it is?
05:13:09hcs(because if it wasn't it'd be something else)
05:14:56nickv111No, what I mean is, how do you derive the rotation matrix?
05:15:03 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
05:19:46 Quit bun-bun ()
05:26:27 Join bun-bun [0] (
05:30:27safetydannickv111, maybe the original paper will help
05:32:44*safetydan doesn't really understand cordic that well but managed to get something running that works well enough for the equalizer coefficient generator in Rockbox
05:34:19 Quit BiptoN ("Leaving")
05:34:59 Quit dpassen1 ()
05:35:28 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
05:35:38 Part Llorean
05:37:33 Join Mouser_X [0] (
05:39:25 Quit Xerion (Read error: 104 (Connection reset by peer))
05:39:43 Join Xerion [0] (
05:44:25 Join markun_ [0] (n=markun@rockbox/developer/markun)
05:44:57 Quit scubacoles ("Leaving")
05:47:03 Join JerryLange [0] (
05:51:44 Quit nickv111 (Read error: 104 (Connection reset by peer))
05:52:36 Join nickv111 [0] (
05:53:00nickv111I think I'm starting to understand the CORDIC FAQ. Basically, to rotate by i is to rotate counterclockwise on the complex plane by 90 degrees, right?
05:59:17 Join arex [0] (
05:59:40*arex waves
05:59:49arexi need to know someting real quick
06:00:32arexim new to this rockbox firmware, and i need to know where i put the music on the mp3 player..
06:00:56arexdose it just go in a new folder named Music ??
06:03:36JerryLangewith the rockbox firmware you can put the music anywhere you want
06:04:27arexso it dosent have to be in its own folder ?
06:05:33JerryLangenope only if you want it to. its your choice.
06:05:56arexoh, alright
06:06:37*arex pats JerryLange on the back
06:06:43 Part arex
06:07:36 Quit markun (Read error: 110 (Connection timed out))
06:10:51 Quit nickv111 (Connection timed out)
06:18:30 Join BjoernErik [0] (
06:25:16 Quit Xerion (Read error: 113 (No route to host))
06:25:18 Join Xerion_ [0] (
06:25:32 Nick Xerion_ is now known as Xerion (
06:26:16 Quit markun_ (Read error: 113 (No route to host))
06:27:23 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
06:30:18 Join markun [0] (n=markun@rockbox/developer/markun)
06:30:33 Quit Bjoern-Erik (Connection timed out)
06:31:31 Part safetydan
06:40:38 Join miepchen^schlaf [0] (
06:46:08 Quit Mouser_X (Read error: 110 (Connection timed out))
07:04:41 Quit maeck ("Ex-Chat")
07:04:52 Quit thegeek ()
07:05:21 Join combrains [0] (
07:05:41JdGordonwoohoo :) another 200bytes off the rec build
07:06:31combrainswho was it that got gb cradle USB support?
07:06:33JdGordon1200bytes above where we were before the menu code started going in
07:06:41JdGordonmarkun probably
07:06:54combrainsnot by the rb front page
07:07:06combrainsi wanna shake their hand
07:07:44midgeykkjurbun i believe
07:08:37JdGordon:'( the system menu is a big one...
07:09:18combrainshmm... looks like its time for me to CVS up again - havnt bothered for about 2 weeks
07:09:59JdGordoncvs up wont do very much
07:10:06JdGordonwe have been on svn for over a month now
07:10:10combrainsdoh SVN
07:10:17combrainsmy badf
07:12:08***Saving seen data "./dancer.seen"
07:12:37hcshmm, trying to decide what to do with the SPC codec
07:13:01hcsdo I try to be done with it before initial commit?
07:13:02Lloreanhcs: Well, it's in need of syncing right now, isn't it?
07:13:23hcsfinding out, I'm svn'ing
07:13:35midgeysmall changes, 10 sec fix
07:13:36LloreanIt's id3.h again, if I recall. When they moved speex.
07:13:41hcsdag nabbit
07:13:47 Join Mouser_X [0] (
07:14:02 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
07:14:32LloreanCan you do anything about it reporting everything as 'Blues'? Maybe even just "Unknown" for the time being? That's the only thing obvious to me that doesn't sit well.
07:14:49LloreanFrom a user perspective, that is
07:14:52hcsok, when I sync it I'll throw something in
07:15:06midgeyoh and are all spc supposed to be 3:05 in length?
07:15:21Lloreanmidgey: That's bad tagging
07:15:37LloreanThey have to be tagged with a proper length, or they default to that.
07:15:46LloreanOr so I've been led to understand
07:15:52midgeyhmm can't say i've used spc before this is recommended for properly tagged spcs
07:16:13LloreanI've had no problems with the ones I got from the site hcs just recommended.
07:16:24midgeyah, i think i got mine from zophar
07:16:34LloreanYeah, the Zophar ones were fairly unreliable for me
07:17:03hcssnesmusic is more like HVSC in their attention to detail and completeness
07:17:45 Join miepchen^schlaf [0] (
07:18:31hcshow about the code layout I have now, with the main driver in spc.c and most of the emulation stuff in spc/ ?
07:18:37 Join thegeek [0] (
07:19:29midgeyhmm so rsn is just a renamed zip?
07:19:41hcsrenamed rar
07:19:57midgeyah makes sense
07:20:01LloreanDoes windows have a way to change the extension on a bunch of files easily?
07:20:07LloreanI ended up just using the command prompt.
07:20:25hcswell, yeah, and it's easy to with with the command prompt
07:20:35hcstougher in, say, bash
07:20:45Lloreanren *.rsn *.rar
07:21:08jhMikeShcs: if it turns out that it really needs to have globals for coldfire, it will need a header with the shared declarations outside of spc.c to compile
07:21:10hcswindows explorer has a lot of confidence in the extensions
07:21:12LloreanBut I realized I don't know a point and click way to accomplish the same thing. It seems like the kinda thing I'd have had to have talked someone through by now, but I haven't.
07:21:46hcsjhMikeS: well at the moment everything is just directly #include'd into spc.c
07:22:32jhMikeShcs: it needs it even then. you don't think spc_dsp.c and spc_cpu.c would be better to have?
07:23:08 Quit Mouser_X (Nick collision from services.)
07:23:11hcsjhMikeS: yes, blargg named the files that way "so no one would try to compile them by themselves"
07:23:17 Join Mouser_X [0] (
07:23:25jhMikeS?? lol
07:23:58hcsgenre = 255 is unknown?
07:25:19hcsI prefer to have everything in one huge file, like I do with nsf,c
07:25:21LloreanIs genre restricted to an int?
07:25:25LloreanOr char?
07:25:35bagawkthat would make sense to me, you are at the limit of a unsigned char
07:25:43hcschar, I think, and that's what the other metadata setters use by default
07:26:11LloreanI thought you could have arbitrary genres with some formats.
07:26:21LloreanI thought the char was just ID3v1
07:26:37LloreanNot that it matters in this case, since I guess there's not an Genre tag for these files. :)
07:26:50hcswell, I take it that if there's a genre string somewhere and it's unset it'll just use the id3 thingy, which we set to unknown
07:27:03hcsand I should probably do this for SID, NSF, and ADX, too
07:28:12Lloreanhcs: 36 is 'Game'
07:28:43midgeyspeaking of ADX is there some source to download adx files? all the ones I have didn't work
07:28:44LloreanWhich seems valid for NSF and SPC
07:29:07hcsmidgey: where did you get these adx files from?
07:29:09*midgey hopes he wasn't trying to play a movie file...
07:29:33midgeyone of my dreamcast games I attempted to back up to my hdd years ago
07:29:46jhMikeShcs: I'm not trying to complicate things...but to make it work all things need to be defined in spc.c before including either header since then Spc_Dsp.h has to know about spc_emu.
07:29:47hcshm, that ought to work
07:29:48midgeyno idea if I did it right
07:30:32jhMikeSif it ends up not really mattering, then it's all moot anyway :)
07:31:09hcsmidgey: there is another type of codec used in some files, that particular game might use it, and I don't know how to play it. If you could upload a representative file somewhere I could take a look.
07:31:26hcsthis site has a lot of ADXs to test:
07:31:54midgeyok, ill see if i can track down to problematic file; i may have since deleted it
07:32:18hcsjhMikeS: hmm, ok, well, guess we'll have to wait until we can find out
07:33:07hcsmeatdata.c :)
07:34:47jhMikeSI hope it doesn't cause it adds a bit of ugliness
07:35:31 Quit [toffe] ("MegaIRC v3.27")
07:36:07hcsregarding the pp5020 freq scaling issues, I couldn't help notice that frequency scaling was taken out of the gigabeat because it reduced battery life...
07:36:29hcsmight it be reasonable to do this for ipods as well, considering the more serious problems it causes?
07:36:46Lloreanhcs: There's already an unofficial build without frequency scaling
07:37:01JdGordonhehe /me has got rockbox on a diet :p the flab is just falling off!
07:37:03hcsI'm aware, and ditto for my builds
07:37:16LloreanBut wouldn't it be better to just solve the problem?
07:37:30 Quit atsea-196 (Remote closed the connection)
07:37:31hcssure, but which can we do right now?
07:37:32LloreanUsers have the option to use an unsupported build for the time being.
07:37:56LloreanBut if you give them a supported build without the problem, how many do you think will be willing to test potential fixes?
07:38:59LloreanAs it is, the majority prefer to look for workarounds than solutions, and don't like to try to reproduce bugs. Tell them we'll hand them the workaround on a silver platter... I dunno.
07:39:18 Join atsea-196 [0] (i=ariel@gateway/tor/x-b8743b2b7cc81c19)
07:39:22LloreanIt sounds cynical, but at least it provides incentive.
07:40:13hcswell, as I've said before, I feel it just harms the image of rockbox unnecessarily
07:40:44LloreanWell, then why not delist it as a supported target, categorize it as 'experimental'?
07:41:24*jhMikeS knows about the difficulty of getting serious testers for bug fixing
07:41:34LloreanI think a non-freezing version with even poorer battery life than the other PP targets isn't exactly good for the image either.
07:41:54LloreanIn the case of the Gigabeat, at 300mhz I understand it gets comparable battery life to the retail firmware.
07:42:17LloreanNot scaling is actually, apparently, 'doing it the right way'
07:42:35LloreanAnd there tends to be a lot of emphasis on 'doing it the right way' around here, it seems. ;)
07:43:48hcsok, well, just thought I'd bitch about it again
07:44:51LloreanHave you tried KoC8 (with or without the multiplier change?)
07:45:18hcsI think I did, but for some reason I don't recall the outcome
07:45:34hcsI switched to no boost for stability in profiling
07:46:29perldiverthe booting sequence on gigabeats is around 4 seconds faster now, we tested deleting (and putting just few bytes into needed files) the rest of the OF files
07:46:56Lloreanperldiver: This is assuming you never want to revert, right?
07:47:11perldiverbut you can always back it up
07:47:41LloreanConsidering I've never booted the original firmware, I don't think I'll need to worry _too_ much about it. ;)
07:47:45hcsjhMikeS, Llorean: just put up patch for genre and syncage
07:48:24perldiverthe whole sequence takes about 6 seconds for me now, from POWER button to WPS
07:48:55hcsjhMikeS: the SPC700 can have 256 samples in memory to look up, but certain games (Doom) change this mapping in realtime
07:49:04LloreanThat is the only thing uncomfortable to me about the Gigabeat. So much slower booting than my other players.
07:49:12 Join Abcminiuser [0] (
07:49:25perldiver6 seconds is still slow?
07:49:38perldiverbut yeah, flashing would help here
07:51:06perldiverOF boot was around 5 seconds
07:52:06LloreanWas that from a suspend-like state, or after a full battery poweroff?
07:52:31perldiverfull battery poweroff
07:52:34jhMikeShcs: ok, but a simple captured image like an .spc file doesn't do that. you need another cpu to send that data, right?
07:52:50perldiver"hibernation" mode gives around 3 to 4
07:53:14hcsjhMikeS: no, it can be changed by the SPC, there is a table for the address where each of the 256 samples starts
07:53:14amiconnperldiver: Then it's the first of I know of that actually boots quite fast...
07:53:20perldiverit was still kind of slow, as i remember it
07:53:53 Quit busa_blade (Remote closed the connection)
07:54:07hcsjhMikeS: this is very rarely done, though, Doom is the only one I've seen yet
07:54:08amiconnAll others are 10 sec or more, sometimes much more
07:54:37perldiverhmm why Llorean said he's uncomfortable with it then?
07:54:54Lloreanperldiver: Because I've been using a Nano. :-P
07:55:00perldiveroh :)
07:55:07LloreanWhich really has unbeatable boot time.
07:57:03perldiverim sure a flashed gigabeat might compete with that
07:57:18jhMikeShcs: I was asking about it regarding the spc file itself somehow changing the samples from one set to another completely different one (say, an spc file with 300 samples). it doesn't appear to operate that way. once it has cached the max 256 entries, there won't be any more in an spc file iiuc. sure, the game program can change them.
07:57:27amiconnYou cannot really compare flash with hdd based players wrt boot time
07:58:14amiconnThe Ondio also boots faster than the archos recorder or player, in spite of the rather slow mmc access
07:58:15 Quit scorche (Read error: 54 (Connection reset by peer))
07:58:43jhMikeShcs: never mind, I get it :P
07:58:44hcsjhMikeS: well, you're correct that no new sample data can get into memory, but the 256 entries you start off with might not be the same 256 entries you always have. Doom uses them more like channels than entries for sample pointers.
07:58:45 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
08:01:09jhMikeSok, so the SPC program can switch those around. It seems the cache continually increments though if it doesn't find it. Guess it would just overflow out of bounds.
08:01:37hcswell it switches between a limited set of actual samples
08:02:44 Quit Soap (Read error: 104 (Connection reset by peer))
08:02:50 Join Soap_ [0] (n=Soap@rockbox/staff/soap)
08:04:12hcsand I have it set up to just wrap around
08:04:28jhMikeSah, I don't believe it did that before iirc
08:05:01jhMikeSmaybe it did...I don't remember now :)
08:05:28hcsI think I've had that in there as long as I had the backup cache
08:06:45 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:06:46jhMikeSah, yes it's uint8_t so it wraps. I would've screwed it all up :)
08:08:47 Quit ptw419 ()
08:09:18hcsit isn't the best behavior, but I was just doing it to avoid overflow in pathological cases
08:10:07hcsspeaking of which, there are circumstances where the BRR cache doesn't work, such as when looping isn't just back in the sample, but somewhere else in ram
08:10:22hcsI haven't encountered these yet, but blargg warned about it.
08:11:02 Join PaulJam [0] (
08:12:03jhMikeShmmm, so the loop can loop back to another sample
08:12:58hcswhereas I assume at the moment that we'll have already encountered the loop point in the course of decoding the sample
08:15:19hcshave you done any comparisons with decode_brr inlined vs. not?
08:15:20jhMikeSI know I need to do more reading (and asking) but how is a loop address determined from the 9 byte block? It only seemed to have a flag that said to loop it.
08:16:11jhMikeSI inlined the head of it but not the more rarely executed decoder and it made no difference
08:16:13hcsthe loop address is sd [raw_voice->waveform].loop
08:18:57PaulJamhi, is it normal that when running a plugin (demystify) the device shuts down after a while (propably idle shutoff) even though music is playing? (i haven't tested with a clean SVN build)
08:20:51LloreanI don't believe I've ever seen it do that.
08:21:05LloreanClearly it _shouldn't_ do that.
08:21:13LinusNPaulJam: no, it should not turn off
08:21:36PaulJamoh, wait. forget that. i'm silly. i had set the shutdown timer earlier but forgot that.
08:23:43hcsjhMikeS: yeah, I think it is the massive decoder block that makes the difference, though I did notice an effect with the smaller keyon block
08:24:24 Quit Mouser_X (Read error: 110 (Connection timed out))
08:25:14JdGordonwhose amazing idea was LANG_BOOKMARK_SETTINGS_MAINTAIN_RECENT_BOOKMARKS ??
08:26:56 Part Llorean
08:29:06 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
08:29:46 Join ender` [0] (
08:30:11 Join Llorea1 [0] (
08:30:53 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
08:32:01 Join Vague_Rant [0] (i=opera@wikipedia/vague-rant)
08:33:17jhMikeShcs: what can even run the continuous decoder with gaussian besides a gigabeat? :) Wasn't a factor for anything I was looking into.
08:33:19Vague_RantIs there a way to import settings that I've exported from a previous Rockbox install?
08:34:06 Join tucoz [0] (i=528612c1@gateway/web/cgi-irc/
08:34:20JdGordonamiconn: 580 bytes to go before the rec build is under where it was before the start :D (the general settings menu saved 1kb and this is with display settings still on the old system ) :D
08:34:23Llorea1Vague_Rant: Simply click on the config file
08:34:24*JdGordon is excited :p
08:34:32Vague_RantSo obvious.
08:34:35Vague_RantThank you very much.
08:34:37hcsjhMikeS: nothing tht I know of, thus why I have that case
08:34:46 Join ackbahr [0] (
08:36:57 Join kaaloo [0] (
08:37:21tucozI have a suggestion i want to make. Now that we have two cross-platform themes, could we vote for/dictator decision, to make one of them the shipped default theme?
08:37:47hcsjhMikeS: are you nearing the point where you can put something up that I should wait to integrate or should I just continue plowing along with the arm-centric view?
08:37:49*JdGordon votes for rockboxed
08:37:58 Quit Supachikn_ (Read error: 110 (Connection timed out))
08:38:15LinusNi like rockboxed too
08:38:23jhMikeShcs: getting closer. have to make it compile with and without "this" first so I can verify it.
08:38:36tucozI like it too.
08:38:55 Quit midkay ("Leaving")
08:39:18JdGordonexcept it has too muuch wasted screen space on the bigger targets
08:39:29jhMikeShcs: For coldfire alot of those multiplies should be shift lefts. The mults are better on ARM?
08:39:38JdGordon(It should have the rockbox logo in the screens that it would fit)
08:39:52hcsjhMikeS: which mults?
08:39:53tucozI am thinking of giving the first timers a pleasant first boot, and it also means that we can start to upgrade the manual with new prettier screenshots of the gui/wps
08:40:04jhMikeSAnything like x*0x100 can be x << 8
08:40:28hcsjhMikeS: I think we sort of assume that gcc is smart enough to do that, is it not?
08:40:28midgeythe backdrop image could always be that yellow with a faded treble clef
08:40:58 Join xigxag [0] (
08:41:14hcsI'd like a default without peakmeters
08:41:40xigxagwow, wasn't expecting so many here
08:41:43jhMikeSI don't think it is...then again I want to be sure
08:41:58JerryLangei totally agree with hcs espically with the ipods
08:42:10 Join _Veseliq_ [0] (
08:42:15ackbahrHello! Anyone know about composite video signals, or a good place to talk/read about it? Thanks!
08:42:32Llorea1The orange everywhere isn't very pleasant with Rockboxed, to me.
08:42:38jhMikeSI've never seen it use a mult where a shift is specified though
08:42:46xigxaggood to see a strong community behind this project, I'm just about to install for the first time on my 5G
08:43:09 Nick Llorea1 is now known as Llorean (
08:43:16JerryLangewhat if you moved the tags around on the rockboxed theme? then it seems more spaced out.
08:43:24hcsjhMikeS: I'm pretty sure it'll use the shift instead of the mult when possible
08:43:48midgeyhmm, rockboxed seems off by one line on my h300, anyone else having the same issue?
08:44:58 Join decayedcell [0] (i=3ba74537@gateway/web/cgi-irc/
08:45:07 Quit decayedcell (Client Quit)
08:45:09LinusNackbahr: like this: ?
08:45:15 Quit Llorean (Read error: 104 (Connection reset by peer))
08:45:48 Join decayedcell [0] (i=3ba74537@gateway/web/cgi-irc/
08:46:00ackbahrWell, I *did* look this up, but I'd rather discuss the topic, because I'm up to something rather special....
08:46:13ackbahrThanks anyway
08:46:34LinusNackbahr: well, rockbox is about audio, you know ;-)
08:47:06xigxagYou guys wouldn't believe what a number iTunes did to my music library when I first installed it last year.
08:47:33ackbahrYep, I know.... But as audio and video are not so different, I thought maybe someone was qualified... Thank you anyway!
08:48:33jhMikeShcs: I try to leave as little up to it as possible cause it can be pretty braindead. It's not good at optimizing common subexpressions for example and putting those in a temp is almost always better.
08:49:20xigxagWhat do you guys use to get podcasts onto your rockbox?
08:49:21 Quit midgey (Read error: 131 (Connection reset by peer))
08:49:44jhMikeSbut if you're doing arm, you're using the newer version I assume? so it could be different now.
08:50:00xigxagI've been looking at myPodder, but there isn't much documentation available for it
08:50:01 Join Rob222241 [0] (
08:50:10hcsjhMikeS: yeah, 4.0.3
08:50:55hcsjhMikeS: I'm looking into the asm stuff in dsp.h now
08:51:06xigxagI really just need something that will download podcasts and something that will sync that folder with one on my rockbox - preferrably the same app
08:52:09jhMikeSI think I recall amiconn found the new one not so good on coldfire but don't recall why.
08:52:39amiconnIf you're referring to 4.0.3 it would be ok for coldfire
08:53:17amiconnBoost while playback doesn't change for most codec, vawpack would profit from 4.0.3
08:53:33amiconnWhat I found is that we cannot use 4.1.x for coldfire targets
08:54:12decayedcellwhy is that amiconn
08:54:23jhMikeSunless it's the defacto version for coldfire I guess I can't rely on any improvments in it
08:55:10amiconnGcc 4.1.x is stupid: it always passes the cpu type to gas. But gcc doesn't know the 5249 and 5250, so we compile with -m5206e
08:55:41jhMikeShcs: see anything that would matter for arm? I guess if it doesn't hurt it at least it's definitely good for cf.
08:55:43amiconnBut gas knows about the 5249, and we need to pass it the true cpu type
08:56:12amiconnThe 5206e lacks some registers the 5249 has, and which we need to use
08:56:49*jhMikeS can't believe after all this time that it shouldn't know about the 5249 and 5250
08:56:51hcsjhMikeS: you mean the asm? it all looks good to me, but I'm very inexperience with arm assembly. I'd like to test out what you have...
08:56:57 Join kubiix [0] (
08:57:17jhMikeShcs: will get it together asap as long as I can keep synced up :)
08:58:00amiconnjhMikeS: Try it... or believe me, gcc 4.1.x really doesn't know the 5249
08:58:01hcsjhMikeS: awesome, sorry to be bugging you about it but I'd rather not be duplicating effort
08:58:23hcsesp when my effort is not too great
08:58:24jhMikeSamiconn: I believe you. It just seems rather ludicrous
08:58:44 Quit kaaloo ("Leaving.")
08:58:58*amiconn thinks it would probably be a good idea to use the same gcc version for all targets
08:59:00safetydanamiconn, is there a bug raised against gcc about that?
08:59:10jhMikeShcs: It is. Was awsome to hear that and not have actually fire up the SNES :)
08:59:14 Join tvelocity [0] (n=tony@
08:59:14amiconnFor arm and sh we already do, so maybe we should switch to 4.0.3 for m68k as well?
08:59:32amiconnsafetydan: Not by me, and I won't do that
08:59:51amiconnThe gcc team isn't very cooperative :(
09:00:11Slasherihmm, i just found an easy solution to the tagcache endianess correction. Just create two functions ec_read and ec_write, which would act similar to read/write but take two extra parameters: structure alignment as a string, "lss" for "long,short,short" for example and the second parameter if endianess should be changed
09:00:17jhMikeSamiconn: If it makes better code for coldfire in 4.0.3 than 3.4.6 then I'm for it
09:00:20safetydanI remember you had some trouble with them before with a SH1 change they made.
09:01:21jhMikeSbtw, since I got the x5 back I've noticed wierd lcd behavior on x5 that booting an old build with a new bootloader doesn't have
09:01:41amiconnsafetydan: It wasn't an SH1 change, but a general change they introduced with gcc 4.0.0
09:02:38amiconn(but which hit rockbox on SH only; there's now a workaround - the problematic part is now completely hidden from gcc, i.e. pure asm)
09:03:44amiconnsafetydan: And that was not the only problem I had with gcc...
09:03:44 Join odb|fide1_ [0] (
09:03:49odb|fide1_good morning
09:04:22ackbahrGoodbye everyone!
09:04:33 Quit ackbahr ("Chatzilla 0.9.75 [Firefox]")
09:04:53odb|fide1_is there a list about all files on root level of an rockbox install on an ipod ? since my last asia trip i have strange autorun files and similar on my root-level. but i am not sure if i can delete all or not :D
09:05:02jhMikeSamiconn: so they don't really like to listen to us outsiders?
09:05:35Abcminiuserodb, all there should be in the base install is the .ipod firmware and the .rockbox dire
09:06:06 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:06:07odb|fide1_ok thank you
09:06:34 Quit Rob2222 (Read error: 110 (Connection timed out))
09:07:41 Quit daurnimator ("Cyas later...")
09:08:07 Join daurnimator [0] (i=daurn@
09:12:05*jhMikeS also notices that older recorded 88kHz wavpack file that used to play back fine cannot decode fast enough any longer on x5 :( The wavpack lib was updated iirc.
09:12:14***Saving seen data "./dancer.seen"
09:12:49 Join Bletch__ [0] (
09:20:28Bletch__hi, apologies if this question is slightly idiotic but I can't find the answer in the FAQs or in the docs - and google hasn't been much help either: How does one get out of the audio thread debug menu on an ipod nano?
09:23:17Mikachupress left or menu iirc
09:23:26Mikachusometimes it is a bit stubborn
09:23:36Bletch__I have tried both of those
09:23:45Bletch__that gets out of most of the debug menus
09:23:52Bletch__just not the audio one
09:24:03Bletch__same with my iriver, except I know how to get out of it on that
09:24:06Mikachutry a bit more
09:24:12Bletch__lol ok
09:24:16*Bletch__ tries some more
09:24:46JerryLangeuhh maybe try a long press of the play/pause button
09:24:57Bletch__ok, hammered it about 20 times and it finally exited
09:25:06Bletch__just takes a long, long time
09:25:13Bletch__thanks very much
09:25:21Bletch__maybe I have a faulty button on the player
09:25:22Mikachuyeah of corus eyou can just shut down too
09:25:42JerryLangebecause iirc the long press of the play/pause equals a stop on the ipod
09:26:06 Quit GodEater_ ("CGI:IRC")
09:26:13Bletch__yup but that wasn't what I wanted to do - on the iriver I check there to see what impack my eq/sound processing settings have
09:26:27Bletch__to see how long it takes the buffer to get back up to speed
09:26:41Bletch__anyway, it's all good now
09:26:49Bletch__thanks for answering my n00b question
09:27:17 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
09:28:05 Quit Jsunu ("Chatzilla 0.9.77 [Firefox]")
09:28:08 Part Bletch__ ("Thanks guys!")
09:28:50 Join Winchester345 [0] (
09:28:55hcsjhMikeS: just wondering, what is the time frame on "asap"?
09:29:06JdGordonbefore xmas :p
09:29:41safetydanWhich one?
09:30:23jhMikeShcs: yes, way before that. A day or so?
09:31:11hcsjhMikeS: oh, ok, awesome, just thinking back to the original question of whether to wait or commit what's there so far
09:32:31jhMikeSwell, if it were in cvs, it would be much easier to coordinate things
09:32:56hcshmm, want me to throw it up, then?
09:33:45DataGhostwow, the guys at TNT have superpowers
09:34:06GodEater_what sort of superpowers ?
09:34:12DataGhosttracktrace states that my package wasn't given to them (however it was said that it would be)
09:34:21DataGhostand now they are delivering it? :o
09:34:28GodEater_that's very clever of them
09:34:29hcshow 'bout the power to kill a yak from 200 yards away?
09:34:37hcswith mind bullets?
09:34:56DataGhostit's been too long since I listened to that one :)
09:35:00jhMikeSWith karate I'll kick your ass, back to Tienenman square :)
09:35:39jhMikeSI'm for it throwing it up there
09:36:19hcsalrighty, how would you prefer the files to be structured? a subdir with some units or one massive file?
09:36:32jhMikeSnoone remembers that one from the album? :(
09:36:57hcssorry, I've just heard Tribute and Wonderboy
09:37:19jhMikeSoh, I've got the cd...*embarrased*
09:37:56 Quit darkless ("Leaving")
09:38:01 Join darkless [0] (
09:38:36*jhMikeS likes it unitized for easy scrolling
09:38:52 Join My_Sic [0] (
09:39:14hcssee... I'd like to have it in something like a final form before the commit, and I think your input may change things substantially
09:39:32Winchester345Heya, does the rb api have detection for the ipod scrollwheel touches (not scrolling or pressing)?
09:39:40 Join fejfighter [0] (
09:39:59linuxstbYes, rockboy uses it.
09:40:13Winchester345ah ok thanks linuxstb
09:40:33Winchester345and... is it generally not used to be more compatible with other players?
09:40:33 Quit decayedcell ("CGI:IRC (EOF)")
09:40:54linuxstbIt's generally not used because there's no reason to.
09:40:55jhMikeShcs: only if it actually matters. I'm trying to compile now for both cases and am having a little trouble. A quick test should be telling
09:41:30Winchester345fair enough
09:43:01linuxstbWinchester345: Is there anywhere in Rockbox you think it should be used/
09:43:24safetydanWell there goes that theory. Fixed-point version of speexdec works with my "broken" speex files. Now to find out why Rockbox is special.
09:44:12Winchester345well.... since it gives 8 directions..... could be used as a mouse..... but that's really unnessesary...... I was just wondering 'cause I never noticed it being used
09:44:24Mikachuactually it gives 96 directions
09:44:52Winchester345even better
09:45:03DataGhost96? :o
09:45:14DataGhostdiagnostic mode says 32, iirc...
09:45:24DataGhostbut outside there it's even more accurate?
09:45:28Mikachutrust me, it's 96
09:48:00*linuxstb wonders how Slasheri's scrolling changes are coming along
09:48:13Winchester345what changes are those?
09:48:30Winchester345oh btw.... just remembered..... wondering if it is an issue with my ipod or not.....
09:48:35jhMikeSfunctions that become void are troublesome
09:48:43Winchester345when a track has been playing for a while
09:48:44linuxstbOptimising the list drawing code (more noticable on the larger colour LCDs), plus implementing clickwheel acceleration.
09:48:55Winchester345turn off the hold switch.... then it seems to ignore key presses
09:49:01Winchester345until I do 2 or 3 in a row
09:49:13hcsWinchester345: this has been reported before
09:49:21linuxstbWinchester345: Yes, that's not uncommon.
09:49:22Winchester345oh..... so how's that going?
09:49:29Winchester345quiet annoying
09:50:58Winchester345is.... what causes it known?
09:52:17 Quit combrains (Read error: 110 (Connection timed out))
09:53:29xigxagdo I place the font folder directly in the .rockbox directory?
09:55:28Winchester345and yeah..... that's there it goes
09:55:47 Part tucoz
09:56:46 Join jba [0] (
09:58:05linuxstbJdGordon: Your commit message says your commit reduces the recorder build by 1200 bytes, but reading the IRC logs, you said it was 200 bytes...
09:58:52JdGordonlinuxstb: i was mentioning numbers in irc as i was goign through the menu... the number in the commit log is with the whole menu done
09:58:58*GodEater_ wonders what JdGordon is buttering amiconn up for....
09:59:47linuxstbJdGordon: So that commit really does reduce the build by 1200 bytes?
10:00:03linuxstbNice. How much is left for rombox to work now?
10:00:46JdGordoni dunno.. the number im going by is the one in the error log from the rec build
10:01:20JdGordon300bytes to lose before we go back to where it was before the first menu commit...
10:01:34JdGordonso after that its 4k or something iirc
10:02:49 Part toed
10:02:51JdGordonhmm... actually I dont know... that number is how much over the uncompressed max is... dunno how much rombox needs
10:05:59 Join Mouser_X [0] (
10:06:28linuxstbLooking at firmware/, it's 256KB - 0x7010 bytes (size of bootbox)
10:06:54Winchester345just out of curiosity....... what's rombox?
10:07:04linuxstb = 233456 bytes
10:07:26linuxstbWinchester345: Rockbox running directly from flash-ROM
10:07:50JdGordonso we are 3700 bytes over atm
10:07:53Winchester345wasn't.... something like that done using scrambler or am I way off?
10:08:22JdGordonwhich means once safetydan's lang fixes are in and the menus are done we should be well under
10:08:47JdGordon*giggles like a school girl*
10:08:53linuxstbPlus I think [IDC]Dragon has reduced Bootbox by about 10KB
10:09:28Winchester345I'll take that as I am way off
10:10:23JdGordonscrambler is for the iriver bootroms
10:10:39jhMikeShcs: neither using globals instead of "this" nor the larger buffer the make any difference as is. But it does play at 95% boost all in c where it would've needed about 105% - 110% before.
10:11:06 Join Nico_P [0] (
10:11:23 Quit BigBambi (Read error: 104 (Connection reset by peer))
10:11:25scorcheWinchester345: this is for the archos devices
10:11:33JdGordonI guess ill use this opportunity to dump another 150 bytes in a function which hasnt been used in _ages_ from menu.c
10:11:39Winchester345so.... that's irrelevant to the ipod?
10:11:44scorchewe are talking about "far off" beacuse currently, it is too big to fit in ROM
10:11:55hcsjhMikeS: not sure what you mean by "all in c", and could you describe what changes you have made to go from 105% to 95%?
10:12:07Winchester345ah I see
10:12:09hcsor, better yet, just post the patch and let me try itout :)
10:13:04 Join decayedcell_ [0] (
10:13:44 Quit Mouser_X (Nick collision from services.)
10:13:53 Join Mouser_X [0] (
10:14:38jhMikeShcs:'s the difference between patch 30 and 35 alone
10:15:37 Quit tvelocity ("Αποχώρησε")
10:15:38amiconnlinuxstb: 0x7010 is not only bootbox. It's bootbox + flash loader
10:15:46hcsjhMikeS: oh, ok, I thought you were doing some stuff with FRACMUL and friends
10:16:26linuxstbamiconn: OK. What do each of them do?
10:16:28hcshey, coincidence, we're at spc36.patch and the id3 genre number for Game is also 36
10:16:55amiconnJdGordon: I think your latest commit will bring back rombox for backlight-modded Ondio FM
10:17:11safetydanDoes rockbox use the low-mem branch of Tremor? The README.rockbox isn't clear on that.
10:17:45jhMikeSI'll post it for you though. I'll post those revisions after I get some rest. You can see if there's a difference any more going either way.
10:18:10hcsok, thanks
10:18:19jhMikeSUsing the pointers makes the code about .3k larger though
10:18:22amiconnlinuxstb: The flash loader is the first thing run (from the flash). It's rather tiny, about 3KB. The flash loader is what allows to choose between main & alternate image (i.e. rockbox and bootbox). It also contains the ucl decompressor
10:18:39hcsjhMikeS: I think I'll get a cleaned-up version of spc36 committed tonight
10:19:22amiconnBootbox is a cut-down rockbox, with only rolo, usb support (and charging on the rec v1)
10:19:50amiconnBoth flash loader and bootbox together are comparable to the swcodec bootloaders
10:20:04amiconn(but the swcodec bootloaders don't handle compressed images)
10:20:06jhMikeSgimme a couple secs I'll have it posted
10:20:17hcsok, no rush
10:23:12hcsI think I should turn up the gain on spc
10:23:28decayedcell_Has anyone got FS4984 to compile nicely against today's SVN
10:23:32hcsit is always far lower than other tracks...
10:24:24 Join jba_ [0] (
10:24:24 Quit jba (Read error: 110 (Connection timed out))
10:27:58 Quit YouCeyE (Read error: 110 (Connection timed out))
10:28:45safetydandecayedcell_, 4984 is probably a long way from being usable
10:29:03 Quit jhMikeS (Nick collision from services.)
10:29:03 Join JerryLange_ [0] (
10:29:08 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
10:30:51 Part decayedcell_
10:31:37 Quit PaulJam (".")
10:32:54jhMikeShcs: up
10:35:21 Join BigBambi [0] (
10:36:29AbcminiuserHCS: How far is the SPC codec off a commit?
10:36:42 Quit JerryLange (Read error: 145 (Connection timed out))
10:37:02 Nick JerryLange_ is now known as JerryLange (
10:37:23Mikachu<hcs> jhMikeS: I think I'll get a cleaned-up version of spc36 committed tonight
10:37:25hcsAbcminiuser: might get a commit tonight
10:38:23AbcminiuserAlso, is anyone else getting a freezup when USB is inserted in a 4G grayscale iPod when the COP patch is used?
10:38:41 Join decayedcell [0] (i=3ba74537@gateway/web/cgi-irc/
10:39:29hcsI've been getting that occasionally on a 4g color, without the cop patch
10:40:24AbcminiuserWell this is all the time
10:40:52AbcminiuserEvery USB insert either shows the little disk activity icon and the disk spins up (and nothing else), or just the USB insert image and nothing else
10:47:08decayedcellYeah got this problem a lot on the 5G
10:47:34decayedcellsame as the Nano, but it happens every single time
10:51:18 Join bluebrother [0] (i=XMi6Ljtt@rockbox/staff/bluebrother)
10:51:33AbcminiuserDecayed, with the COP patch?
10:51:41decayedcellyes sir with the cop patch
10:52:28AbcminiuserHmmm, at least it's seemed to have solved the frequency scaling bug for me (at the cost of slower scrolling)
10:53:05AbcminiuserAny devs have any ideas? I'd assume it would be a problem with the COP thread blocking the reboot into the disk mode, but I have no proof at all on that :P
10:55:48decayedcellI'm pretty sure its to do with the new config files with the USB charger detection update
10:56:27Winchester345Hey..... ..... how can does a 2 bit LCD show so many shades? O_o
10:58:20linuxstbtemporal dithering
10:59:05decayedcell33 shades thats quite impressive
10:59:09Winchester345is that something I can google or something you guys came up with?
10:59:14 Quit dan_a ()
10:59:34AbcminiuserI thought it was just PWMing the cells?
11:00:02AbcminiuserYou have a defined framerate, then the more each cell is turned on inside that framerate, the darker it appears?
11:00:28AbcminiuserSo if you are showing at 60Hz, and have a cell that's off 1/2 the time, it's 1/2 as dark as its neighbours?
11:00:29hcsthus, temporal dithering
11:00:34Winchester345and.... how come the plugin doesn't seem to compile for mini 2g while it says on the plugin page it suppports mini?
11:00:46Winchester345hmm.... clever.
11:01:13AbcminiuserMy AVR works have taught me something it seems :)
11:01:43hcsPWM = Pulse Width Modulation, I guess, but what's AVR?
11:02:15AbcminiuserYes, PWM = Pulse Width Modulation, and AVR is a type of Microcontroller made by Atmel
11:02:36AbcminiuserBasically, a non-crap PIC - better speeds, better architecture, etc
11:03:35Winchester345so.... how come I don't seem to have the greyscale demo on the 2g mini?
11:05:41Winchester345....... is it something I said? >.>
11:06:03amiconnWinchester345: grayscale.rock isn't compiled by default. It's more or less just a demo, intended for devs
11:06:23Winchester345so how would I get it to compile?
11:06:27amiconnThere are several plugins which use the grayscale lib: mandelbrot, jpeg viewer, doom, ...
11:06:42amiconnYou need to add it to apps/plugins/SOURCES
11:07:43Winchester345no such dir.... O_o
11:08:05Winchester345grayscale.c seems to be in the plugins folder though
11:08:20amiconnYes, and there's a SOURCES file in the same dir
11:08:43Winchester345oh >.> excuse my idiocy there
11:10:30Winchester345I want to code something useful for rockbox... just all this low level, cross-platform and foreign API is all new to me so I am kinda hopeless right now
11:12:17***Saving seen data "./dancer.seen"
11:15:00AveLD rockbox.elf
11:15:00Ave/data/cvs/rockbox/build/apps/settings_list.o:(.data+0x88): undefined reference to `__compound_literal.4'
11:15:15Avesince cvs-svn transition I cant build no more, any ideas?
11:15:23Winchester345are you using cygwin?
11:15:29Aveno, gcc 4.0.2 cross
11:15:52amiconnAve: I guess you're building for an arm target (e.g. ipod), and have arm-elf-gcc 4.0.2
11:15:58amiconnYou need 4.0.3
11:16:05Aveah ok thanks
11:17:21 Quit jba_ (Read error: 110 (Connection timed out))
11:19:57Aveis 4.0.4 ok?
11:20:22 Join ppeom [0] (i=oppa123@
11:20:48 Part Vague_Rant
11:21:01markunThe battery meter in the Rockboxed theme is completely wrong here: at 100% it shows up empty, at 90% it's 2/3 filled, at 60% 1/3 filled..
11:22:35 Join jba_ [0] (
11:22:38 Part Winchester345
11:22:47ppeomHi all
11:23:18directhex|workAve, can't you use the script, which downloads & builds the cross compiler you need?
11:23:46AveI could but I already have my own build thing set up
11:24:02AveI just need the new gcc toolchain in one path and thats it
11:24:32ppeomwho's know this tag?
11:24:37 Quit atsea-196 (Remote closed the connection)
11:24:48ppeomwhat's mean 70?
11:25:08directhex|workAve, so your cross compiler's fine except for the lack of gcc?
11:25:30Avedirecthex|work: no, I had too old version (4.0.2) for current svn code
11:26:28 Join atsea-196 [0] (i=ariel@gateway/tor/x-f7f36f73fdd95cf3)
11:28:37linuxstbppeom: That looks like an unofficial tag
11:29:37GodEater_yeah I was hunting for %e too - can't find it anywhere in the wiki
11:30:02ppeomI apply multi font patch..
11:30:17markunThose Gigabeat kids with their custom bootsplashes :)
11:30:33ppeomThat's tag multi font tag.. I don't know 70 mean..
11:30:52markunppeom: ask the multi font patch creator
11:30:54GodEater_markun: niceeeeeee ;)
11:31:12ppeomwho's make?
11:31:28markun'Amar' from
11:31:29directhex|workooh. multi font? sounds swish
11:31:51Avehm maybe I should change to afterall
11:31:53ppeomyes.. Jclix theme use it font..
11:32:12markunI wonder why the patch creators don't come here to talk about the future of rockbox but instead just work on their own things
11:32:32bkmultifont is a dead end
11:32:37bkbut it was fun to do
11:32:44markunin what way dead end?
11:32:59markunThe implementation is not good?
11:32:59bkbecause it uses a ton of RAM and ends up being fairly slow
11:33:06linuxstbmarkun: Do you know the font cache code?
11:33:15markunlinuxstb: a little. I know it's very slow :)
11:33:21bkthe current font code is very crufty
11:33:23ppeombk.. only ipod?
11:33:38linuxstb:( ... Do you think it could be extended to add multiple fonts?
11:33:51bklinuxstb: not easily
11:34:02linuxstbBut would that be the "right" approach?
11:34:05bkthe LRU stuff is painful to read through
11:34:11directhex|workbk, how much is a ton, assuming execution on a ridiculously overpowered player?
11:34:14markunbk, linuxstb: yes, I've though a lot about how to do a better implementation, but it's not so easy to do fast and not waste RAM
11:35:05bkdirecthex|work: (# of fonts) x (buffer size), buffer size with my patch was somewhere around 40kb
11:35:16markunlinuxstb: the biggest problem with glyph caching (especially with multiple fonts) is that they vary a lot in size
11:35:41linuxstbSo you get a fragmented cache?
11:36:02bkit's just not designed for variable size entries
11:36:02directhex|worki'd be happy enough if WPS font and menu font could differ
11:36:07markunlinuxstb: not right now, because for every glyph the space of the biggest one is used
11:36:10directhex|workwell. i'm happy enough now
11:36:19linuxstbmarkun: OK, I see the issue...
11:36:40markunlinuxstb: and if you load a 2nd font with even bigger glyphs..
11:38:09markunalso, for ever glyph printed, I think a linked list is traversed and some pointers changed to have a LRU implementation. I hope we can make the glyph lookup a bit more efficient.
11:39:08bkI tend to think glyph lookup should be pushed down to the loader code which probably involves totally rewriting the font code
11:39:36markunThe code was written by one of the chinese rockbox users, but they showed very little interest when I was working on unicode and used the code.
11:40:01markunbk: which loader code?
11:40:34bkmarkun: the disk font loading routines, instead of the higher level font cache stuff
11:41:09bkwith most fonts we don't have the memory to load the whole thing, then search for glyphs
11:41:13markunI still don't get what you want to do. Get rid of the font caching?
11:43:05 Part JerryLange
11:43:12 Quit Mouser_X (Read error: 110 (Connection timed out))
11:43:34markunperhaps you should just implement it to show me what you mean :)
11:44:13markunwill your implementation be faster and/or use less RAM?
11:44:45bkthat would be the goal
11:45:08bkI'm not interested in maintaining big patches indefinitely
11:45:21markunno, of course not
11:45:22bkthis is all in the planning stages right now though
11:45:24 Quit GodEater_ ("CGI:IRC (Ping timeout)")
11:45:36markunI remember it was quite difficult to maintain the unicode patch
11:46:47markunbk: can you maybe write your ideas on a wiki page or something?
11:47:21bkmarkun: yeah, I can sketch out what I'm thinking on the wiki
11:47:33bkI've got to run at the moment but I'll be around later today
11:47:40markunok, thanks
11:48:30directhex|workmarkun, did you finish your tale of how the gigabeat port came to be?
11:50:25markunno, not yet
11:50:51 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:51:26 Quit linuxstb (Nick collision from services.)
11:51:32 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
11:59:14 Join MonkeyTamer [0] (n=chatzill@
12:01:05DataGhostoh my god
12:01:11DataGhosttalking about n00b-proof instructions
12:01:20DataGhostI can't believe that there are actually people THAT stupid
12:01:33DataGhost'WARNING! Do NOT wash protector with PDA! Wash ONLY protector!'
12:01:36MonkeyTamerthere's no such thing as "n00b" proof
12:01:42DataGhostwell, more or less :P
12:01:50 Join jba___ [0] (
12:02:04MonkeyTamerfor every effort, there is a lack of effort to acknowledge said effort
12:02:26directhex|workthere's nothing wrong with "n00bs". the problem is with the stupid, not the ignorant. one of the two can be cured
12:02:31 Nick jba___ is now known as jba (
12:02:39MonkeyTamerthat is true
12:03:23MonkeyTamerwhile the ignorant may not know better, they may be learned; whereas stupid people insist on their ignorance, and refuse to be taught
12:04:32MonkeyTamerthere has to be warnings for everything; I'm honestly surprised that there aren't people making you sign waivers when you walk through a store
12:05:14xigxagspeaking of n00bs....I managed to hack together a script to sync my podcasts. Do you know if I can make it autostart when the drive is mounted?
12:05:16*safetydan wonders how the heck a member of structure can become null when passed to another function
12:05:53 Quit jba_ (Read error: 110 (Connection timed out))
12:06:03AbcminiuserYou know, the RockBlox iPod game would benefit from the scrollwheel touch API rather than the button API...
12:06:33 Quit My_Sic (Read error: 104 (Connection reset by peer))
12:08:38hcsI find its support for the ipod control scheme irritating
12:08:42hcstoo sensitive
12:08:52MonkeyTamerI assume you mean for moving the pieces?
12:08:55scorchei was about to say that
12:08:59scorchefor anything
12:09:09scorcheit would be too easy to screw up accidentally
12:09:22MonkeyTamerit already switches the pieces' position, but I agree: it's simply annoying
12:10:10AbcminiuserSweet bobble-headed jesus, so it does...
12:10:21AbcminiuserBut if RockBoy can use it, why not RockBlox?
12:10:26AbcminiuserAnd yes, to move the pieces
12:10:31MonkeyTamerI can't stand the blasted scroll wheel for the brickbuster game
12:11:33MonkeyTamerI can see the benefit at higher speeds for that game, but the scroll wheel can be inaccurate
12:12:21MonkeyTamertry playing that brickbuster game and see what I mean
12:13:02 Quit fejfighter ()
12:15:26MonkeyTamerdoes it seem to anyone else like the Rockbox menu, as in the main menu scrolls slower?
12:15:39MonkeyTamerand only the Rockbox menu
12:17:18 Quit MonkeyTamer ("Chatzilla 0.9.77 [Firefox]")
12:18:35w1ll14mmorning all
12:18:57safetydanCan someone give me some hints on how bufferpool becomes null at the end of this gdb session?
12:23:14safetydanor actually ignore that altogether
12:23:17*safetydan slaps forehead
12:24:04 Nick Soap_ is now known as Soap (n=Soap@rockbox/staff/soap)
12:28:36 Join wooo [0] (
12:34:02DataGhostgod, your parents must be assholes... what I just saw :D
12:34:08DataGhostI got an e-mail from 'Jesus C.'
12:35:05 Quit lukaswayne9 (Read error: 104 (Connection reset by peer))
12:35:55markunsounds like a normal latino name to me :)
12:41:58 Join lee-qid [0] (
12:42:56AbcminiuserSounds like the sort of identity I'd set up for myself ;)
12:43:28AbcminiuserOh, is the chopper patch ready to be comitted? It's cross-platform I believe
12:43:35AbcminiuserAnd works well in the tests I've performed
12:43:58AbcminiuserOr what about the pulsing starfield? That's neat, and a very simple patch
12:44:20AbcminiuserI hate to see all the little patches languishing forever in the tracker until they become out of date :(
12:44:34safetydanAbcminiuser, copper with or without the menus? It would crash after you opened and closed the menu a few times for me.
12:44:39safetydanchopper even
12:45:04AbcminiuserWithout is the one I've used
12:45:18AbcminiuserI can't test the with, since I'd have to update Cygwin
12:45:44*Abcminiuser really has to set up a dual boot Ubuntu environment so he can compile rockbox quickly and easily
12:46:09scorchewhat is wrong with vmware or colinux?
12:46:27 Join kurzhaarrocker [0] (
12:46:31safetydanWell I've had enough coding for one night. A question for the logs, why does Tremor have oggmalloc.c which is awfully similar to codeclib.c?
12:46:33 Quit safetydan ("Ex-Chat")
12:46:40AbcminiuserI've tied virtualization, and it Ubuntu didn't like it one bit
12:46:46AbcminiuserNo display driver, I guess
12:47:03directhex|workdisplay drivers is the curse of virtualization
12:47:03scorchedefine "didn't like it"
12:47:19*kurzhaarrocker is impressed: My iRiver ihp120 did record at a sample rate of 96 kHz. Wow!
12:47:39AbcminiuserWell, it would show the boot screen, then go black and do nothing
12:47:55scorchehave you tried the rockbox image?
12:48:33 Join lukaswayne9 [0] (
12:48:42AbcminiuserRockbox image
12:49:00directhex|workrockbox vmware machine
12:49:21AbcminiuserWasn't aware of it
12:49:22scorchehe means the image of
12:49:32AbcminiuserWoah, how'd I manage to do that?
12:49:37 Quit kurzhaarrocker (Remote closed the connection)
12:49:58scorchego to VMwareDevelopmentPlatform
12:51:44 Quit Abcminiuser (Read error: 104 (Connection reset by peer))
12:52:30 Join Abcminiuser [0] (
12:52:30 Quit decayedcell ("CGI:IRC (EOF)")
12:52:58 Quit ze (Read error: 104 (Connection reset by peer))
12:53:32 Join ze [0] (
12:54:34 Join decayedcell_ [0] (
12:56:50 Quit Nico_P (Remote closed the connection)
12:57:53 Join barrywardell [0] (
12:57:59 Join anathema [0] (
12:59:34 Quit decayedcell_ (Remote closed the connection)
12:59:47 Join decayedcell_ [0] (
13:00:17xigxagwhen I plug in my ipod, explorer kind of freaks/freezes for about 8 seconds - is this normal?
13:02:01jbais it scanning the disc for auto play?
13:02:27xigxagyeah I have it enabled
13:02:41xigxagI'm trying to get my autorun.inf to load an exe
13:03:11xigxagbut I just tried it with autorun.inf removed, and explorer still had the same trouble
13:07:40 Join Nico_P [0] (
13:12:18***Saving seen data "./dancer.seen"
13:15:17 Join Thundercloud [0] (n=thunderc@
13:16:04 Quit wooo (Read error: 110 (Connection timed out))
13:17:26 Part decayedcell_
13:18:08 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
13:18:14 Join miepchen^schlaf [0] (
13:18:36 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
13:18:55 Quit Abcminiuser (Read error: 110 (Connection timed out))
13:19:31 Join Criamos [0] (
13:21:11 Quit jba (Read error: 104 (Connection reset by peer))
13:25:43 Join Mouser_X [0] (
13:36:43 Quit Mouser_X (Nick collision from services.)
13:36:52 Join Mouser_X [0] (
13:45:06 Join wooo [0] (
13:45:32 Quit jhMikeS (Nick collision from services.)
13:45:32 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
13:45:36 Join miepchen^schlaf [0] (
13:45:38 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
13:46:23*petur considers upgrading his h340 to h380.... $260 without shipping
13:46:36 Quit anathema (Read error: 110 (Connection timed out))
13:46:36 Quit Nico_P (Remote closed the connection)
13:46:54 Join anathema [0] (
13:52:04 Part odb|fide1_
13:52:45 Join Rondom [0] (
13:58:45 Quit himitsu (Read error: 111 (Connection refused))
13:59:21 Quit Rondom ("Ex-Chat")
14:02:11 Join nick89 [0] (
14:02:22 Quit nick89 (Read error: 104 (Connection reset by peer))
14:04:06 Join OgMaciel [0] (n=omaciel@ubuntu/member/gnukemist)
14:04:17 Quit wooo (Read error: 110 (Connection timed out))
14:05:12 Quit directhex|work ("Leaving")
14:13:34 Quit OgMaciel (Read error: 104 (Connection reset by peer))
14:14:43 Join OgMaciel [0] (n=omaciel@ubuntu/member/gnukemist)
14:21:06 Quit OgMaciel ("Ex-Chat")
14:27:30 Join himitsu [0] (n=himitsu@
14:27:39 Quit Mouser_X (Read error: 110 (Connection timed out))
14:28:32 Join markun_ [0] (i=c2fe7883@rockbox/developer/markun)
14:29:25markun_petur: Do you think any of this project could be useful to us?
14:32:25JdGordonBagder: I want to change the english of a translatable string, Do i change the source: or only the dest in english.lang?
14:40:35LinusNJdGordon: read the comment in the beginning of english.lang
14:42:10 Quit markun_ ("CGI:IRC 0.5.7 (2005/06/19)")
14:42:43JdGordonso umm? Do I change both or need to create a new lang?
14:43:08LinusNchange both
14:43:14LinusNwhich phrase btw
14:43:24JdGordonplugin browser to browse plugins
14:44:24markunwhy not just "plugins"?
14:44:37LinusNyou mean browse plugins -> plugin browser?
14:45:15LinusNit should work if you change both src and dest
14:45:21JdGordonmarkun: i dont really care what it says.. but everyone seems to agree that browse plugins is bad
14:45:28scorcheBagder: was there a reason that you changed the approximation of the VMware image zip file back to "~217"?
14:46:01LinusNJdGordon: inconsistency is bad
14:46:22JdGordonalso, will I be allowed to hide the recent bookmarks item if the option is disabled? (which is default)
14:46:32scorcheoh...sorry...ignore me...i was looking at the raw bytes...
14:46:40LinusNJdGordon: i think you should
14:46:50*scorche should sleep more so he doesnt do this type of thing...
14:48:54JdGordonLinusN: yay :)
14:55:13JdGordonwhat are we calling the plugins? just plugins? or browse plugins?
14:55:27*JdGordon is slowly getting fed up with this patch :p
14:56:05 Join Arathis [0] (
14:56:07 Join |Rincewind| [0] (
14:57:09LinusNJdGordon: i *think* just "Plugins" is fine
14:57:13markunJdGordon: to me "file, database, plugins" looks better than "browse files, database, browse plugins"
14:57:23markunfileS for the first one
14:57:46LinusNmarkun: sounds ok to me
14:58:25markunLinusN: anything rockbox related you are working on?
14:58:50LinusNnope, unfortunately not...
14:59:15markunI just keep adding things to my list, but spend very little time actually implementing anything..
14:59:23LinusNtell me about it
15:00:26LinusNdamn, i was about to try the root menu patch, but it didn
15:00:39JdGordonill do a new one if you want
15:01:01Lynx_what's a root menu?
15:01:04|Rincewind|I wasn't able to compile with root menu yesterday, too">
15:01:40 Join webguest15 [0] (i=4671586f@gateway/web/cgi-irc/
15:01:41markunLynx_: a main menu to navigate to all the different rockbox parts
15:02:00markunLynx_: file browser, settings, WPS, recording screen etc
15:02:10*LinusN compiles...
15:02:36webguest15markun: got a question for you re: gigabeat
15:02:44JdGordoncodewise its bassically done now, just need to agree on the items, their order and the english
15:02:45markunwebguest15: ok, just ask
15:03:13markunJdGordon: and there was also some redundant code in the tree browser code, right?
15:03:17LinusNi'm wondering if we should commit the scroll margins patch...
15:03:22webguest15markun: down at the bottom of, there's a file labelled with the comment "after backing up your GBSYSTEM folder, you can impove boot time extracting this."
15:03:35markunwebguest15: yes
15:03:41JdGordonmarkun: yeah, there is a fair biit of dead code because of this.. but that can wait till after the commit imo
15:03:52webguest15markun: they seem to be dummy files. any danger of bricking my gigabeat putting them on there?
15:04:09JdGordoneach of the screens really should be checked individually for exess code now
15:04:28markunwebguest15: that chance always exists, but you can always recover.
15:04:29 Join perplexity [0] (
15:04:44markunwebguest15: a few of us have tried it and it worked really well
15:04:49 Quit anathema (Read error: 110 (Connection timed out))
15:04:59webguest15markun: cool... will give it a shot
15:05:12webguest15markun: thanks for all the work you and the other gigabeat devs are doing, btw
15:05:20JdGordonLinusN: if the only reason its not in is because we ar waiting for viewports then I rekon it should go in
15:05:31 Join OgMaciel [0] (n=omaciel@ubuntu/member/gnukemist)
15:05:34markunwebguest15: and the other rockbox devs have done too of course
15:05:42webguest15markun: indeed! :)
15:05:46 Quit webguest15 (Client Quit)
15:06:49*JdGordon isnt sure if not allowing ACTION_STD_CANCEL to exit the browsers is such a good idea
15:07:42markunJdGordon: I'm now also thinking the 'root' menu shouldn't be at the root :)
15:08:04markunbut just be accessed by the menu button
15:08:12markundon't know what the rest of you think
15:08:12|Rincewind|is there a svn command to update all files and automatically remove all local code changes?
15:08:17Mikachusvn revert
15:08:26JdGordonthat doesnt update tho
15:10:08|Rincewind|hm, I really don't know if everything is broken now, I better start completely fresh
15:10:25markun|Rincewind|: revert will work, no need to download everything again.
15:10:59|Rincewind|it doesn't matter, I have a fast internet connection.
15:11:22markun|Rincewind|: try "svn revert -R ./" and then "svn up"
15:12:04|Rincewind|I did that, but i didn't trust it and did a "rm -rf *" afterwards ^^
15:12:12Mikachujust check the output of svn diff
15:12:21***Saving seen data "./dancer.seen"
15:14:29|Rincewind|it would make more sense to have the root menu be the root (espesially if we want database to behave the same as the browser)
15:15:04markun|Rincewind|: why?
15:15:06|Rincewind|but I don't know that code very well, so don't listen to me
15:15:40markunif they don't go to the menu if you press left it will still be the same behaviour
15:15:52|Rincewind|then the root menu would be call the other screens and is the central point in rockbox
15:16:21markunit will be the central point anyway, even if you have to go there with the menu button
15:16:45markunI just think it's more consistant because it's the only way to go there from the WPS
15:17:19 Quit GodEater_ ("CGI:IRC (Ping timeout)")
15:17:19markunIt's more like a teleporter screen that way :)
15:17:35|Rincewind|I was thinking code wise, currently the wps and the menu exit to the file browser, if I rember correctly
15:17:43JdGordonLinusN: had a play with it yet?
15:17:55LinusNJdGordon: waiting for that damn cygwin crap
15:18:05JdGordonstill wow :p
15:18:10JdGordonhavnt you got ccache installed?
15:18:12markun|Rincewind|: no, the menu either exits to the file browser or the WPS, depending where you called it from
15:18:36LinusNJdGordon: nope, maybe i should try it
15:18:37amiconnJdGordon: ccache makes things worse on cygwin
15:18:48LinusNamiconn: really? wow!
15:18:50JdGordonoh, ouch
15:18:51markunLinusN: switch to vmware like the rest :)
15:18:56JdGordonor colinux
15:19:03amiconnThe problem is the windows filesystem layer, not cygwin itself
15:19:31Mikachuso compiling things in say msvc is slow too?
15:19:31*JdGordon hopes it was ok to crosspost a message on both ml's
15:19:44|Rincewind|markun, ok, I don't know what I am talking about and possibly it doesn't matter anyway
15:19:58amiconnmarkun: I am still using cygwin for building rockbox most of the time, even though I have both vmware and a real linux box available
15:20:11JdGordonamiconn: isnt there a gcc option to use pipes for all the intermediate data instead of files? would that help on cygwin?
15:20:31Mikachusurely everyone has -pipe in their cflags by default?
15:20:34markunamiconn: but you are more conservative than the rest so I didn't expect anything else :)
15:20:43amiconnIt's simply more convenient to use cygwin when using a windows editor
15:20:55Mikachu(does windows support pipes though?)
15:20:59amiconnOf course it does
15:21:21Mikachui don't know why it's of course, but okay
15:21:43amiconnmarkun: This has nothing to do with being conservative, but with being convenient
15:21:58Mikachudoes rockbox use -pipe in cygwin then?
15:22:00markunok ok :)
15:22:35*JdGordon doesnt see CFLAGS being set in the makefile
15:23:10markunamiconn: I've never used either vmware nor cygwin so I shouldn't give any advice anyway
15:23:20MikachuJdGordon: just run make V=1
15:23:29amiconnPerhaps I'll change to a compile-only mini vm soon
15:23:41amiconn(without x11)
15:24:38JdGordonamiconn: strip x out of the colinux image and have it running in the background, the use it exaclty like you would a remote linux box (samba and ssh) and your good
15:24:43barrywardellJdGordon: The H10 OF uses a fairly intuitive root menu type system. It calls them Music, FM Radio, Recording, Photo, Text, Browser, Settings.
15:24:50barrywardellMusic is the database mode
15:25:17JdGordonI dont like Music for database.. wont that confuse ppl looking for the wps?
15:25:24amiconnJdGordon: I was thinking about doing that, just with vmware instead of colinux
15:25:48SlasheriLibrary would be nice
15:25:49markunbarrywardell: and how do you navigate from the different items to this menu?
15:25:58markunSlasheri: yes, it would
15:25:59JdGordonthe reason I said colinux is because it can run it as a windows service, and not worry about it if you dont need it, its always ready for you
15:26:08barrywardellpushing left enough times always brings you back to the root
15:26:32markunbarrywardell: and how do you get to and from the WPS screen?
15:26:52barrywardellget to the WPS screen by pressing play
15:27:13barrywardellpressing left in the wps screen brings you back to the Music/Browser screen
15:27:54*LinusN likes the root menu
15:27:59markunin that case it's pretty good yes. But I don't think we want to change left to do the same in our WPS
15:28:15barrywardellif you go to the WPS by pressing play, left brings you to the screen you were in before the WPS
15:28:32|Rincewind|I think left reverts the track in wps?
15:28:40amiconnurgh, contextual buttons :(
15:28:58markunamiconn: which we want to get rid of..
15:29:23 Quit ppeom (Read error: 104 (Connection reset by peer))
15:29:34|Rincewind|a rockbox where every button does the same in every screen wouldn't be very usable...
15:29:34JdGordondoes CFLAGS=-pipe enable or disable pipes?
15:29:42amiconnI think the root menu should be called with the menu button, from all screens
15:29:50markunamiconn: yes, me too
15:30:21amiconnLikewise, the menu button should be usable for quickly leaving the whole menu tree, returning where you came from
15:30:25LinusNhmmm, somehow selecting "Files" now always takes me to the wps...
15:30:31barrywardellyes, agreed.
15:31:22JdGordonLinusN: it does? wierd :p
15:31:23barrywardellwhat should happen when you "back out" of the WPS? back to database/file browser?
15:31:40markunbarrywardell: you can't back out :)
15:31:44amiconnbarrywardell: Back out how?
15:31:45markunonly to the root menu
15:31:51markunwith the menu button
15:31:52JdGordonbarrywardell: if you press select, it will put you in the last browser, if you press menu it will put you in the root menu
15:32:06barrywardelli mean, where do you go from the wps?
15:32:14|Rincewind|JdGordon: recording screen is now duplicated in the main menu and the root menu
15:32:22JdGordoni know
15:32:44JdGordonim going to leave that temporarily, because its more hassle to remove now than when i get to that menu in the reworking...
15:32:49barrywardellJdGordon's answer is what I was thinking. "back out" was the wrong term
15:33:10markunwell, I don't think select should do that in the WPS
15:33:26JdGordonthats what it does currently... im open to suggestions tho
15:33:32JdGordondo we need an extra button in the wps?
15:33:48barrywardelli think it should be possible to go from the wps to either the root menu or the current play track in the database/file browser
15:33:58|Rincewind|jdGordon: I like it that you like my suggestion I made in the flyspray comment. That would be the first idea of mine that makes it into rockbox :)
15:34:44*bluebrother just read the last tracker comments on the root menu
15:35:01bluebrotherlooks like it becomes something I thought of first, but I somewhat even confused myself :(
15:35:10 Join midkay [0] (n=midkay@rockbox/developer/midkay)
15:35:11bluebrotherI should get more sleep those days
15:35:38bluebrotherJdGordon: but how should the browser behave if you press left from the browser's root?
15:35:47JdGordonim not sure
15:36:07JdGordonim split between doing nothing and going to the root menu
15:36:07bluebrotherrequire you to press the menu button to call the menu? If you consider the "root" menu as a root it should (logically) go back to that menu
15:36:23bluebrotherOTOH I don't like that, as it breaks hold-left-to-root
15:36:45bluebrotherand will make it harder for the blind people, as they might find themselves in some menu instead of the browser
15:36:47JdGordonI can get around that
15:36:59JdGordonI can force it to stop at / even if left is held down
15:37:06*JdGordon praises the action system :D
15:37:15|Rincewind|if holding left puts you back to root of file browser and a second press of left goes to root menu then I think that blind users could navigate with this, too
15:37:18LinusNi'd like the root menu to have a different look than the other menus
15:37:25bluebrotherbut what if you press left a couple of times quickly? Me does this sometimes ...
15:37:33LinusNmaybe that could be made with a different backdrop
15:37:53bluebrotherput the RB logo in it ;-)
15:38:03bluebrother(as backdrop)
15:38:08JdGordonI want to give that menu the clef icon...
15:38:15JdGordoncan anyone draw one at 6x8 size?
15:38:19bluebrotherhmm, nice idea.
15:38:23LinusNJdGordon: lol
15:38:38bluebrothershouldn't we consider using icons in different sizes?
15:38:46|Rincewind|well, if you moved too far, then you can push right once and you are back in the file browser root
15:38:54markunLinusN: and maybe some icons?
15:38:57bluebrotherwhile 6x8 is nice on my h120 its kindof tiny on ipod video etc.
15:39:01LinusNbluebrother: we've been considering that for years :-)
15:39:06barrywardellthe H10 OF has a different backdrop for each item in the root menu. probably a bit excessive
15:39:16LinusNmarkun: sounds nice
15:39:18bluebrotheruse a clef as backdrop ;-)
15:39:18 Join n1s [0] (
15:39:34bluebrotheryou could make it as light as possible for the 2bpp devices
15:39:41JdGordonno, I dont think we need to make it any different
15:39:43|Rincewind|a backdrop on h120 would make the text very hard to read
15:40:00bluebrothernot if you design the backdrop carefully
15:40:51JdGordonLinusN: also, the root menu code is the same stuff that drives every other menu.. so hacking its looks will be annoying and bloaty...
15:40:52markunJdGordon: would be nice if the whole menu could be centered (if the text fits of course)
15:40:52bluebrotherJdGordon: how about this: if the startup screen is set to file browser, left from the file browser root doesn't leave it but requires you to use the menu button
15:41:11|Rincewind|even black text on light gray is difficult to read, The only possible backdrop that i can imagine is a border, but that's not easy with different font sizes
15:41:21bluebrotherand if the Rockbox menu is set as startup screen allow left to go "up" to the Rockbox menu
15:41:36bluebrothercould work the same way for the database browser
15:41:43JdGordonmarkun: that would need work in the list widget (which probably should happen), but wouldnt work so well atm
15:41:49LinusNmarkun: i was just about to say that
15:42:05|Rincewind|bluebrother: with that you are mixing setttings that don't have do do with each other
15:42:06JdGordonbluebrother: that could work...
15:42:08markunwould the viewports make this easier?
15:42:15JdGordonprobably not
15:42:24bluebrother|Rincewind|: why am I mixing settings?
15:42:59|Rincewind|startup screen and left from file browser are two different things, imho
15:43:02LinusNbluebrother: "startup screen" is only supposed to set which screen is displayed when you start the player, not to set a virtual root
15:43:08markunamiconn: how are the viewports supposed to work? Do you give them a fixed size and position?
15:43:23JdGordonits a bit premature asking that isnt it?
15:43:30bluebrotherI know, but that could be somewhat abused to set that behaviour too
15:43:41LinusNbluebrother: "abused" being the work
15:44:09JdGordonso are we agreeing that the only way out of the browsers is with the menu button?
15:44:31JdGordoni mean.... uh... damn.. need sleep
15:44:37|Rincewind|and play to go out of the browser and enter wps
15:44:40markunJdGordon: I do
15:44:56bluebrotherJdGordon: I think this will be confusing to users that prefer the root menu as real screen root.
15:45:13bluebrotherunless we consider the root menu being the new main menu, and not a root for all screens
15:45:16JdGordonhow about, if we get complaints we change it then? ;p
15:45:25|Rincewind|would it hurt to make this an option?
15:45:27bluebrotherhehe ... sounds good
15:45:28peturindeed, that way it's no longer a root but you moved browser into the old menu
15:45:37markunbluebrother: new main menu would be my vote
15:46:02bluebrotherIMO we shouldn't make a new root menu that is root of all screens but instead have a new main menu
15:46:03n1sI'd like the root menu to be the root menu so to speak
15:46:18*JdGordon i getting so confused
15:46:19bluebrother(which is why I find this "root" naming a bit confusing)
15:46:46|Rincewind|if the root menu is the new main menu then it's not a change, just a reorder of things and naming it different
15:47:16bluebrother|Rincewind|: no, as it now also allows to access other screens it didn't allow before (because of design reasons)
15:47:33bluebrotherbut it's still a usual menu, similar as before.
15:47:49n1snow _that_ would be confusing IMO
15:48:19 Part LinusN
15:48:29markunn1s: why?
15:48:57bluebrotherI think I'd prefer simply a new menu like the current "root" menu
15:49:01n1sif you are in the menu, press left to fileborwser, right doesn't go back
15:49:21markunn1s: no, left in the menu will not do anything
15:49:29bluebrother... that new main menu shouldn't go back to something with left
15:49:49bluebrotherbut instead it should be possible leaving that menu from any point using the menu button
15:49:53n1sIMO if it is the _root_ then left should do nothing, just as left in the filebrowser root does nothing currently
15:49:57markunbut of course there will be a patch to call the WPS when you press left :)
15:50:10n1sof course ;-)
15:50:27 Quit linuxstb (Read error: 110 (Connection timed out))
15:50:31|Rincewind|I think n1s wants to say that you enter the browser with LEFT on the first entry of the menu (so you think you are one level deeper) but you can't go back with right then
15:50:32 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
15:50:55|Rincewind|sorry, i confused left with right...
15:50:58bluebrotherleft? Isn't selecting right?
15:51:04n1s|Rincewind|: yes I mixed left and right up :-)
15:51:22bluebrotherquick poll: anyone not confused? ;-)
15:51:49n1sto be clear, IMO what you enter with right should be possible to exit with left.
15:51:50Mikachui'm not, but i haven't read anything
15:53:20JdGordonso we are back where we started? except everyone is confused
15:53:33n1sJdGordon: seems so, yes :-)
15:53:51Mikachudo NOT pass go, do NOT collect $200
15:54:13 Join Phalangees [0] (
15:54:26|Rincewind|we only agree that we don't agree with each other ^^
15:54:43bluebrotheris there any point I can pass and collect free time?
15:55:14*petur reads about free time and follows bluebrother
15:55:37*JdGordon has plenty
15:55:44JdGordonah the life of a uni bum :D
15:55:49*|Rincewind| goes to repair his bike
15:56:08markunJdGordon: so, 1 menu (which might not be called root) from which to go to everything else, and only menu brings you to this screen and nothing else changes screens?
15:57:18JdGordonyes, except there are shortcuts between screens
15:57:24markunare there?
15:57:35JdGordonfiles <-> wps
15:57:41markunyou want to keep that?
15:57:45JdGordonof course
15:58:05JdGordonwps > recording (_maybe_ )
15:58:05 Join perl|work [0] (
15:58:09bluebrotherpressing select on a file in the browser and not starting playback would be quite weird
15:58:20*JdGordon agrees
15:58:20markunbluebrother: true :)
15:58:38markunbut it could start in the background :)
15:58:40bluebrother_but_ you could just create a new playlist, start playback (like it currently does) and _not_ switch to wps
15:58:47|Rincewind|having buttons on the player and not using them to navigate quickly would be a waste
15:58:54bluebrotherthat might make pretty much sense as well.
15:59:16JdGordonway too many points to argue on in the patch :p
15:59:29markun|Rincewind|: I just think that navigation should be more unifform
15:59:52bluebrotherso, the screens could only change by play (always to wps), menu (always to the menu)
16:00:14bluebrotherselect always selects, and all other screens are accessible through the new Rockbox menu
16:00:25|Rincewind|I'm ok with uniformism: PLAY -> WPS, MENU -> root menu, SELECT -> browser/database −− from everywhere
16:00:32markunbluebrother: and play in the WPS goes back to the previous screen?
16:00:44JdGordonno play would pause then
16:00:46bluebrotherwith perhaps adding an additioinal shortcut to the recording screen for devices that have a dedicated recording button
16:01:01bluebrotherno, as play is also pause we would need to make an exception here and use select for that
16:01:10markunand the Gigabeat select pauses
16:01:39amiconnThere are more exceptions on certain targets
16:01:54bluebrotherhow do you go back to the browser on the gigabeat then?
16:02:25markunbluebrother: with the same button as you use to enter it (power button)
16:02:45markunwhich might be changed to be the 'a' button
16:03:33bluebrotherhmm. Makes sense, but not for targets that label the play button play/pause
16:04:39markunyes, the labels force you to use the buttons in some ways
16:04:56|Rincewind|I don't think we should stick to the labels, because the Hxx doesn't have a menu button, so we would have no way to enter the root menu...
16:05:37bluebrotherthe h100 has a menu button −− the one we currently use
16:05:43bluebrotherlabelled "mode" iirc
16:06:14bluebrotherwe should stick to the buttons where it makes sense, and if the play button is labelled play/pause we should use that button to pause playback
16:06:21|Rincewind|yes, but mode isn't the same as menu, and in the OF the menu was on long select if I remember correctly
16:06:25bluebrothereverything else would just confuse the users
16:06:49peturh3x0 OF uses REC longpress for its 'root' menu
16:07:08bluebrotherwe don't have any "mode" function, so I think it's perfectly ok to reuse that button for menu
16:07:09Mikachui don't think you should follow OF keybinds blindly
16:07:19*bluebrother agrees
16:07:44 Join Mouser_X [0] (
16:07:54bluebrotherwe should follow it where it makes sense for users. The most important will be play / pause / stop
16:08:24Phalangeeswhy is stop really needed??
16:08:35|Rincewind|Not using the play/pause button for something more useful in a non-WPS screen just because it is labeled play/pause is a case of "following blindly" for me
16:09:12bluebrotherI'm not speaking of a screen where it makes sense to reuse that button.
16:09:17 Quit perplexity ("Check out my rear window")
16:09:32bluebrotherI'm speaking of screens where we currently don't use it otherwise (like the menu) and wps
16:10:22JdGordoncan we change the rockbox logo to use a different not so it is easier to make a icon out of it? :D
16:10:46peturhaving playback control when _not_ in a WPS is a must... I even miss volume control :/
16:11:01Mikachui moved volume control to the quick screen, yay me
16:11:24*JdGordon thinks the quickscreen is next to get overhauled after this
16:12:00markunpetur: the Gigabeat has dedicated volume buttons which we would like to be able to use from all screens to change the volume
16:12:19peturI can understand that
16:12:41markunbut for some plugins it's nice to reuse them
16:12:50Mikachuthat seems sort of hard unless you change the volume from the button driver
16:12:53JdGordonok, im going to bed... if anything is agreed on while im sleeping put it in the fs page pls
16:13:05barrywardellthe root menu should be designed to work with a minimal set of buttons common to all targets. shortcuts could be added for each target depending on button availability
16:13:21JdGordon^ speaks the truth
16:13:39*|Rincewind| agrees with this
16:13:45 Quit JdGordon (Remote closed the connection)
16:13:57barrywardelli'm not sure what those minimal set would be. maybe left, right, menu, up, down?
16:14:08n1sI think we need a UI dictator, that listens to people and the does exactly what (s)he wants :-)
16:14:24Mikachuthe problem is no buttons are common to all targets :)
16:14:41peturn1s: a dictator doesn't listen to people :p
16:14:43markunLinusN is a good dictator I think
16:14:54|Rincewind|left, right, up, down and one menu button should be possible on every target
16:14:56Mikachufor example ipods only have left, right, menu, play, and the wheel which no one else has
16:15:01markunpetur: what's "verlicht despoot" in english?
16:15:11*n1s agrees with barrywardell too and think what you enter with right should be exitable with left...
16:15:13peturno idea ;)
16:15:31n1spetur: ok, but I want's one that pretends to listen! :-P
16:16:05markunn1s: what about the WPS?
16:16:15barrywardellMikachu: but scroll=up/down, left=left, right=right, menu=menu
16:17:47n1smarkun: well, that is a special case, left and right have some very logical functions there, but entering the filebrowser with right, makes me want to leave with left :-)
16:18:21markunn1s: I know, but because of the WPS I thought maybe we shouldn't do that
16:19:31n1smarkun: i don't like having left doing nothing in what will appear as menus to most users...
16:19:53n1sI mean it will exit the regular menus but not browsers..
16:20:20markunit will not exit in from the settings menu either
16:20:33*petur gets the feeling we can get rid of the A-B menu button on h3x0 and re-use it for something else :)
16:21:10peturbut then you can't access settings from the filebrowser as they're mapped next to each other
16:21:34n1smarkun so the menu is enter with right and your trapped (evil laughter)? yes I know you get back with the menu button but you can't go up one lvel
16:21:52 Quit Mouser_X (Nick collision from services.)
16:22:02 Join Mouser_X [0] (
16:22:23markunpetur: it will require some more navigation yes
16:22:32markunn1s: yes, it's a trap :)
16:23:10markunbut so is the WPS
16:23:48n1sIf we leave left doing nothing, I personally think right shouldn't do anything either and all 'enter' actions will be done with the select button, (if we have one)
16:24:03bluebrotherpetur: who do you want to go to the menu without the mode button?
16:24:16Phalangeescouldn't right insert a song into playlist and select just play it?
16:24:18markunn1s: also in the settings menu?
16:24:47n1sANd then we could support skipping songs when not in the WPS on targets with loads of buttons (insane laughter)
16:24:55markunI think we are trying to change too many things at the same time :) or is it a good time to do so?
16:25:45n1smarkun: it seems strange to me having right entering and left doing nothing, that is my problem, and I think it would be confusing
16:27:44markunn1s: maybe we should ask JdGordon to implement it this way and try it
16:29:37n1sIt is a difficult decition to make in a large group with lots of oppinions involved, so I think that we will not really get anywhere unless someone just decides something, maybe Jdgordon should make the call...
16:31:47XavierGrnow that you mention of buttons... we should utilize the record button on iriver H100/H300
16:32:13XavierGrand oh please remove that awfull splash message when you press left on a setting screen
16:32:56amiconnLeft should accept instead of cancel
16:33:00n1syes we probably should but it is a similarly difficult thing to decide anything with buttons involved is just ewww
16:33:02amiconnThen the splash would go
16:33:07XavierGrI agree with amiconn
16:33:16XavierGrit was way better with the old behaviour
16:33:16n1samiconn: I agree with you there
16:33:21XavierGrfaster and more consistent
16:33:23markunand I would like splashes to go away when a button is pressed
16:33:24amiconnIt was like this for ages
16:34:09XavierGrcancel should be the stop button not the left
16:34:22amiconnyes, exactly
16:34:49XavierGrthe most annoying thing is when I just go to a setting to see a value, then I press left to leave and get the annoying cancel message
16:34:58Slasherihehe, now tagcache is endianess-independent :)
16:35:08XavierGrand no, it is not consistent to press right to leave a setting screen
16:35:27amiconnIt doesn't hurt though
16:35:50XavierGrboth can be there
16:36:33amiconnOn the recorder there used to be 3 'aceept' buttons for settings
16:36:39amiconnLeft, Right and Play
16:37:00XavierGrwell it was something like that on H100 too, right? (left, right, navi (click))
16:38:18markunSlasheri: good!
16:38:52amiconnXavierGr: Yes. Play on the recorder is where Navi is on the irivers, in the center of the cursor cross
16:39:44 Join mathgl [0] (
16:44:58 Join theone [0] (i=180c4cca@gateway/web/cgi-irc/
16:50:15 Quit Mouser_X (Nick collision from services.)
16:50:24 Join Mouser_X [0] (
16:57:19 Join Paprica [0] (n=Paprica@
17:01:30 Quit petur ("stkov")
17:03:22 Join ackbahr_ [0] (
17:04:17 Quit ackbahr_ (Client Quit)
17:07:37 Join ackbahr_ [0] (
17:07:59theonehow do you dispose of li-poly batteries?
17:11:48 Join [toffe] [0] (n=[toffe]
17:12:28***Saving seen data "./dancer.seen"
17:15:28 Quit ^BeN^ (Read error: 110 (Connection timed out))
17:16:09 Quit printfXh4 (No route to host)
17:25:28 Join printfXh4 [0] (n=pseudo@
17:27:13 Quit Seed (Nick collision from services.)
17:27:20 Join Seed [0] (
17:27:37 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
17:27:43theonehow do you dispose of li-poly batteries?
17:28:16Phalangeesa fire ;D
17:28:50Phalangeesthat was a joke. don't actually throw it in a fire
17:31:42Phalangeesi'm almost tempted to open up my gigabeat hard drive and check how many read heads it has
17:32:04Phalangeesoops wrong channel
17:32:10 Part Phalangees ("Leaving")
17:33:08 Quit YouCeyE (Read error: 104 (Connection reset by peer))
17:35:51theonefire it is
17:41:34 Quit theone ("CGI:IRC (EOF)")
17:41:34 Quit ackbahr_ (Read error: 104 (Connection reset by peer))
17:41:36 Quit mathgl ("Quitte")
17:42:37 Join fasmaie [0] (
18:01:01 Quit lee-qid (Read error: 104 (Connection reset by peer))
18:03:53 Join Rondom [0] (
18:07:32bluebrotherhmm. Rolo-ing seems not to save the settings
18:07:37bluebrotheris that intended?
18:08:12 Join dune2 [0] (
18:10:34 Quit Mouser_X (Read error: 110 (Connection timed out))
18:11:24 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
18:12:18 Quit BigBambi ("Leaving")
18:17:18 Part fasmaie
18:17:42 Quit YouCeyE (Remote closed the connection)
18:25:14 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
18:28:13 Quit YouCeyE (Client Quit)
18:39:16 Join RedLeg [0] (n=waca-wac@
18:39:21 Quit dpassen1 (Read error: 145 (Connection timed out))
18:40:39RedLegCan anyone answer a simple question about the ipod bootloader for rockbox?
18:41:44bluebrothernot as long as you didn't asked
18:42:04bluebrothersimple rule: just ask, if someone can answer your question you'll get an answer.
18:42:42RedLegOK, thanks...
18:43:41 Join Llorean [0] (
18:43:44RedLegCurrently the bootloader defaults to loading RockBox, and falls back to the original firmware of RockBox is not found, or if (in my case) menu is held while booting. Is tere a way to bame the bootloader load the original firmware by default?
18:44:08GodEaterand thrice no
18:44:09RedLegWell, that was simple.... 8-)
18:44:14bluebrotherit's the rockbox bootloader and thus not wanted
18:44:25LloreanIt becomes more complicated if you're willing to compile you own bootloader. ;)
18:44:43bluebrotheris compiling that complicated? ;-)
18:45:08bluebrotherbut, if you compile the bootloader yourself you could change that behaviour
18:45:13LloreanNo, but you do have to make a small change to a single source file, and some people feel that 'programming', no matter how minute an amount, is 'complicate' :)
18:45:28LloreanIf I recall you can change the behaviour from setting a == to a !=
18:45:35bluebrotherbut programming is fun ... :D
18:45:43RedLegAH... well, I have had the cross-compile environment going in the past, but that was for my original Archos..... I don't suppose the iPod would be that different?
18:45:53LloreanNot really, no.
18:46:03RedLegOK.. I may look into that.
18:46:12LloreanThe only difference is that instead of the sh GCC, you set up an arm GCC.
18:46:48RedLegDoes anyone have any experience with the ipodlinux bootloader and loading RB? It purports to support a boot menu....
18:48:10LloreanGenerally speaking, support for iPodLinux's loader comes from the iPL guys.
18:48:34LloreanBut yes, it is configurable and has a menu. You'll have to check their documentation for how, though.
18:49:27RedLegUnderstood.... but I'm wondering if it causes problems with loading RB. It claims to, but I really don't want to reload all this blasted music AGAIN due to problems....
18:49:57DataGhostiPL bootloader loads rockbox just fine
18:50:00LloreanYou... shouldn't have needed to RE-load your music in the first place.
18:50:08DataGhostrockbox just doesn't load fine on my 80GB yet
18:50:08GodEaterwhichever bootloader you go for is entirely independent of your music
18:50:28LloreanExcept on HFS+ iPods, Rockbox's install is non-destructive to your music.
18:51:11RedLegI had some pretty severe HD corruption that I'm pretty sure was caused by a very weak battery. I've replaced it, reformatted everything and things _seem_ OK now....
18:52:00RedLegThat's a troubleshooting FAQ that I wish were more emphasized: If you're having what _seems_ to be HD problems, try replacing the battery....
18:52:54RedLegIn any case, thanks for the help, and for all the work on the project.
18:53:08 Join robin0800 [0] (
18:53:12 Join BigBambi [0] (
18:53:26 Quit RedLeg ("Leaving")
18:54:16 Quit linuxstb (Read error: 110 (Connection timed out))
18:54:30 Quit barrywardell (Remote closed the connection)
18:54:31 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:04:52amiconnlibmad0 and libmad0-dev packages do exist on debian-amd64
19:05:07*amiconn wonders whether those packages have the 64 bit issues fixed
19:05:28*amiconn installs libmad0-dev
19:05:57 Quit lost|X40 ("Leaving")
19:07:13 Quit crwl ("WeeChat 0.2.2")
19:11:16 Part Llorean
19:11:29 Join crwl [0] (
19:12:30***Saving seen data "./dancer.seen"
19:13:47 Quit Siku (Read error: 131 (Connection reset by peer))
19:15:40 Join My_Sic [0] (
19:16:44 Join Febs [0] (
19:17:29amiconnnothing :(
19:18:32 Join Soul-Slayer [0] (n=jonno@
19:22:17 Join Siku [0] (
19:27:36 Join Thundercloud_ [0] (n=thunderc@
19:34:28 Quit Arathis ("Bye, bye")
19:36:41 Join directhex [0] (
19:36:51 Join CriamosAndy [0] (
19:37:03directhexmarkun, is it me, or does the remote on the gigabeat behave rather oddly when the main player hold switch is enabled?
19:38:03 Join rotator [0] (n=e@rockbox/developer/rotator)
19:44:01robin0800 on this page does not seem to work does it need updating for svn?
19:48:31amiconnThe bitstream functions in libmad are written in a way that makes them break if 'long' isn't a 32 bit datatype
19:48:52 Quit dune2 ("Leaving")
19:50:35 Join Mouser_X [0] (
19:51:37 Quit Thundercloud (Connection timed out)
19:55:08 Quit Criamos (Read error: 110 (Connection timed out))
19:57:12 Join madduck [0] (n=madduck@debian/developer/madduck)
19:57:44madducki enabled the database index auto-update and i also initialised it by hand on my iAudio X5
19:57:58madduckyet when i go from an album to the artist dir, it takes between 2 and 5 seconds
19:58:04madduckany idea why?
19:58:36madduckalso, when in the playback window and i tap "next", it takes about 1-2 seconds for the next song to start, the current one will still be playing during that time.
19:59:54 Quit Mouser_X (Read error: 145 (Connection timed out))
20:00:19 Quit darkless ("Leaving")
20:00:19Slasherimadduck: you need to load database to disk, dircache is good to enable also
20:00:20madduckand entering the recoding screen actually freezes the device after the microphone has been turned on.
20:00:27Slasheri*to ram
20:00:32n1smadduck: sounds like it takes some time to spin up the disk, try enabling the "Load to RAM" setting for the database, or set the disk spindown higher, (will kill batterytime)
20:00:36madducki have enabled it to be loaded into ram
20:00:50*amiconn blames preglow for the non-working libmad on amd64
20:01:13 Join sbeh [0] (
20:01:16 Join darkless [0] (
20:01:28amiconnpreglow: You committed Antonius' libmad optimisations - which are breaking libmad on amd64
20:01:33madduckhow do i turn on dirindex?
20:01:39madducksorry, dircache
20:02:25madduckah ,found it
20:03:17madduckmuch better!
20:03:28 Join ender1 [0] (
20:03:42madducknow how can i get rid of the 1 second delay when tapping next?
20:04:27amiconnpreglow: FS #5219, namely the huffman decoder optimisations
20:07:37 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
20:13:45Slasherimadduck: put the anti-skip buffer to its maximum
20:13:56Slasherito 10min. or something like that
20:14:55madduckokay, that fixed it, but why???
20:15:12Slasherithat delay is caused because disk has to spin up
20:16:07Slasherisetting anti skip buffer to 10 minutes keeps always 10 minutes worth of data in ram, very nice with crossfade especially
20:19:39madduckso for my last question of the day: how can i coerce the thing into recording? right now, it just freezes...
20:19:51Slasherii have no idea
20:19:58Slasheriit should work fine
20:20:00 Join mattzz [0] (
20:20:31amiconnOtoh a high anti-skip buffer setting wastes battery, because rockbox is forced to rebuffer earlier
20:20:35 Quit ender` (Read error: 110 (Connection timed out))
20:20:41 Join phil_munich [0] (i=55b51144@gateway/web/cgi-irc/
20:21:45Slasheriamiconn: that depends, skipping often reduces that effect
20:22:36Slasheriit would be nice to measure that how much buffer size affects to battery time
20:22:36madduckso if i never skip, does it drain the battery?
20:22:42Slasherithere is a limit somewhere
20:22:53Slasheriit could drain it little faster
20:22:54madducki would think that if i never skip, battery is actually conserved, cause the disk only has to spin up once every, say 9 minutes.
20:22:58phil_munichhi! is there anyone, wo could tell me how to change the range of scanning on an rockbox iriver h320?
20:23:20Slasherimadduck: nope, increasing that setting causes disk to spin more often
20:23:25phil_munichehm.. i mean the range of the fm radio
20:23:33madduckSlasheri: why?
20:23:42madduckthat makes no sense!
20:23:46 Quit YouCeyE (Remote closed the connection)
20:23:49Slasherimadduck: because not the full buffer capacity is used
20:24:01Slasherionly the capacity - 10min. is used for example
20:24:09BigBambimadduck, say there is space in the buffer for 20 mins - setting antiskip to 10 mins means a spin up every 10 mins. Setting it to 5 mins givs a spin up every 15 mins
20:24:10 Quit Rondom (Read error: 110 (Connection timed out))
20:24:29 Join Rondom [0] (
20:24:32linuxstbphil_munich: You change the FM region setting in the radio settings.
20:24:36madduckah, i thought antiskip == buffer size
20:24:51BigBambias the buffer is refilled when there is the amount set by the anti skip left in it
20:25:09hcsjhMikeS: happy mornin'
20:25:15BigBambiso longer anti-skip = more left unused in the buffer when it is refilled
20:25:27Slasherimadduck: buffer size is fixed to the available amount of free memory
20:25:32madduckmakes sense now
20:25:47Slasheriantiskip sets the buffer margins
20:26:27*linuxstb likes the latest root menu
20:27:53*BigBambi too
20:29:04amiconnSlasheri: Skipping often may reduce the effect, but it will never get negative
20:29:05hcsis the root menu stuff still only on FS or have I missed it?
20:29:27linuxstbhcs: Still on flyspray.
20:29:40amiconnIf (and only if) you *always* skip before the buffer gets below margin, the effect is zero
20:29:58*amiconn skips perhaps every 50th track or even less often
20:30:17Slasheriamiconn: true, but probably that effect is quite small with there large buffers
20:30:33jhMikeShcs: hello :) I'm putting the coldfire asm in now.
20:31:31Slasherihehe, i skip more often, everytime the database shuffles a bad track ;)
20:31:46amiconn10 minutes is almost the complete buffer size with vbr −−preset standard mp3 on X5 and H100/H110. (half the buffer size on H120+ / H300)
20:31:55 Join dan_a [0] (n=dan-mirc@
20:32:05amiconnI wouldn't say that's a small effect
20:32:12Slasheriamiconn: there is a limit that at the maximum half of the buffer capacity is used by the antiskip buffer
20:32:32 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
20:33:46Slasherii still dont see why it would have a significant effect
20:34:17hcsjhMikeS: coldfire asm or the general purpose FRACMUL stuff?
20:34:18amiconnHalf the buffer size -> twice as many spinups over time
20:34:47 Quit thegeek ()
20:35:08jhMikeSthat too
20:37:27 Join thegeek [0] (
20:39:40madduckwill rockbox ever support USB mode while continuing to play?
20:40:35GodEaterUSB mode implies there are changes happening to the filesystem
20:42:15madducksure, but why would that disturb rockbox?
20:42:28 Join Thundercloud__ [0] (n=thunderc@
20:42:28madduckif i delete the song it's playing, fair enough.
20:42:33madduckbut if i just add more music??
20:42:56desowinit can destory file system
20:42:58hcsyou can't tell, when something else is writing to the hard drive you must abandon all hope
20:43:38madduckthis is VFAT, no? even windows can deal with it... or is the USB layer not a part of rockbox?
20:44:17 Quit phil_munich ("CGI:IRC (EOF)")
20:44:28desowinI have no idea what you mean by ' even windows can deal with it.'
20:44:34Slasheriamiconn: lets think that iriver could have something like a 6000 mWh capacity battery and one spinup (2A, 2s) consumes 1 mWh of energy at maximum while full buffering takes something like 5 mWh (30s) and half buffering 3 mWh (15s) with spin-up energy included. So increasing that anti-skip buffer to its maximum adds something like 1 mWh/spinup at maximum
20:44:37madduckis rockbox an OS of its own or based on a *nix system?
20:44:48desowinOS of its own
20:45:00madduckah, okay.
20:45:14madduckthen it's probably a design "limitation" i guess.
20:45:24desowinone question
20:45:39madduckdon't get me wrong, i am not dizzing it, it saves my day all the time! :)
20:45:52desowinis there any firmware to any player that allows you to modify file system on computer when you're playing music ?
20:46:56n1smadduck: what you are asking is impossible on most of our targets where USB communication is handeled in hardware
20:48:21n1son some targets where USB (probably will be) handeled in software, it could theoretically be possible but would be a major pain to create, and USB in rockbox is not supported at all on any of those targets yet
20:48:59Slasheriamiconn: ups, just forget those previous calculations, there was a mistake.
20:51:59Slasheriamiconn: one spinup 4mWh, 30s buffering 15+4, 15s buffering 7.5+4
20:52:17jhMikeSadded env_rates to IRAM as well...makes 7% speedup
20:52:57Slasheriand battery capacity 7 Wh (1900 mAh battery)
20:55:54Slasheriamiconn: so if nothing else would consume energy, one could do raw buffering over 3h with normal buffer size compared to over 2.5h with half of the buffer size
20:59:05madduckn1s: yes, it makes sense if USB and rockbox are in the same ring
21:01:53madduckthanks guys for a very fruitful half an hour! :)
21:01:56 Part madduck
21:04:03 Quit robin0800 ("Take my advise. I don't use it anyway")
21:04:53 Quit Thundercloud_ (Connection timed out)
21:05:53 Join l33ch [0] (i=9ec3654a@gateway/web/cgi-irc/
21:06:02l33chhi all :)
21:06:43 Join mathgl [0] (
21:06:46l33chi have question ... how is working this dual booting on iRiver H10 ? :D
21:08:20l33chI did everything what was written on web
21:08:54l33chbut after i turn on player i don't see choice which firmware to load :)
21:09:32l33chis it in preparing state or is anybody whom is this great oportunity working ? :P
21:11:37Febsl33ch: did you read this?
21:11:50GodEaterFebs: darn it - I was just typing that too!
21:12:32***Saving seen data "./dancer.seen"
21:13:07Soul-SlayerI'll just make it easy :P Hold down the 'Cancel' button after turning it on
21:13:14l33chthx guys
21:13:14Soul-SlayerWill boot into OF
21:13:20l33chi gonna check it :)
21:13:23 Join mirak [0] (
21:14:10Soul-SlayerThen you need to reset the player to get back to rockbox (Either remove the battery and replace it on the 5/6gB, or press the reset button on the 20gB)
21:14:34l33chit's working
21:14:46l33chthis cancel button :)
21:15:05l33chguys you are awesome :) today mp3 h10 6gb arrived
21:15:21l33chand now i'm happy :)
21:15:29Soul-SlayerGreat :).
21:18:36l33chnow only thing i wanna is to implement or just patch for .cuesheet files and then i'll be in heaven ;-) it's all what i miss on another mp3 players
21:19:27Soul-SlayerI have just learnt how to patch myself, could make you a build if you wish :p./
21:21:43l33chi tried it before
21:21:50l33chwith this debian image
21:22:19Soul-SlayerAny luck?
21:22:30l33chi wanted to build normal version
21:22:37l33chbut there were errors
21:22:43Soul-SlayerNormal builds are downloadable from the site itself
21:22:47l33chi then compiled new gcc
21:23:07l33chbut not with this cuesheet support
21:23:07Mikachuthere's a new vmware image with 4.0.3
21:23:16l33chyeah ?
21:23:29l33chi go check ... is there a link ?
21:23:35Mikachusame place as before iirc
21:24:28Mikachubut if you already compiled 4.0.3 yourself, not much point
21:25:55Mikachuplease try not to use "multi-segmented" downloading...
21:26:18l33ch33 seconds left
21:26:57 Join midgey [0] (
21:27:15l33chsorry did it problem to you ?
21:27:23Mikachuno, it's just bad in general
21:27:29 Quit Kingsqueak (Remote closed the connection)
21:27:40Mikachudoing my part to help the internet
21:27:54l33chi notice it for next time
21:28:45Soul-Slayerdan_a: Any news on the KoCOP :)? I can't imagine it's too far from SVN now (After testing on all targets and repairing the 'Recording' function)
21:28:58 Join decayedcell_ [0] (
21:30:45dan_aSoul-Slayer: No news - I've had lots of real life to deal with lately. I have a suspicion as to what is breaking recording (if it's not cache synchronisation with the codec thread, which it isn't, then it's probably IRAM use) but I've not done a test for that. I'm having a couple of nights off, but I'll get back to it later in the week.
21:30:58 Join Thundercloud_ [0] (n=thunderc@
21:32:33l33chSoul-Slayer: Can I ask you to patch the latest source with this cuesheet patch and build it ? :D
21:32:53Soul-Slayerl33ch: Certainly, any other patches?
21:34:33l33chdo you recomend any others ?
21:35:34Soul-SlayerWell, as I was just discussing, the KoCOP patch makes use of the H10s coprocessor, which can increase performance. If you use Rockbox's recording feature however it will make things harder for you, as it doesn't work with the patch implemented
21:35:49Soul-Slayerhowever, as you have dual booting set up, there shouldn't be a problem as you can use the recording from the OF if you need to
21:37:15l33chtomorrow i'll go to security conference so i use this original recording feature
21:37:32l33chto have sure to be recorded :)
21:37:48l33chbut if there exists some performace patch
21:37:53l33chso why not use them :)
21:38:36Soul-SlayerCurrently the recording feature is a little tempremental on Rockbox in my eyes, after recording you have to reset the player in order to listen to songs and so forth, may be worth waiting for recording to get fixed up properly before you rely on Rockboxs recording feature anyhow ^^
21:38:49Soul-SlayerI'll put the KoCOP patch in and if you don't like it, tell me and I can make one without for you ^^
21:39:47l33chplease don't forget on this cuesheet patch :)
21:39:50l33chthanks in advance
21:39:56Soul-SlayerAlready in there ^^
21:41:34 Part decayedcell_
21:42:01l33chthen let me know from where I can download :)
21:42:19Soul-SlayerWill do - Just compiling as we speak.
21:42:24l33chif there is anything with i can help you with tell me :)
21:42:41Soul-SlayerThe satisfaction of making someones life a little easier is all I need :p
21:43:28Soul-SlayerSeeing as I don't have the knowledge to actively help in developing Rockbox, I may aswell use the knowledge I do have ^^
21:45:58perl|workare you using the cue sheet patch with the album art one?
21:46:16Soul-SlayerI haven't included the album art patch, no.
21:46:32Soul-SlayerShould I?
21:46:45perl|workthey dont compile together
21:46:54Soul-SlayerAh, okay.
21:46:58Soul-SlayerI shall take that as a No then :p.
21:47:30l33chtake it as No :)
21:48:48Soul-SlayerI don't have any cuesheets to test with, I'll upload it and hopefully it will work :p.
21:49:05l33chi can check it right now :)
21:49:11Soul-SlayerOkay :).
21:51:13 Join Mouser_X [0] (
21:51:47Soul-SlayerDirect link :$
21:52:11 Quit bluebrother ("leaving")
21:54:48 Quit Thundercloud__ (Read error: 110 (Connection timed out))
21:55:07l33chnow i copy t player
21:55:11Soul-SlayerI believe you need to enable cuesheet support in 'General Settings -> Playback -> Cuesheet Support -> On' once you have it copied on
21:57:18l33chplease reboot to enable
21:57:32Soul-Slayer for more information - The patch implementation uses "the amarok way"
21:57:35w1ll14mmikachu: what do you think about a delay when going above 0db volume to use the +6 as a 'boost' so that the user wouldn't accidently blow up their ears
21:57:39Soul-SlayerJust turn the player off and on again
21:57:42w1ll14mmaybe a useless idea ....
21:58:26Soul-Slayerw1ll14m: Maybe even before that, even 0 is pretty loud
21:58:56w1ll14msoul-slayer indeed.... but you can choose where 'boost' starts
21:59:18w1ll14msome sort of are you shure you wan't to go louder ?
21:59:49Soul-SlayerPerhaps not quite like that, for that could get a little annoying... Perhaps something in options to cap the volume?
21:59:54l33chthis cuesheet is working :)
21:59:58Soul-SlayerGreat :).
22:00:14w1ll14msoul-slayer: it can be set as an option which you can enable or not....
22:00:17l33chthx you are my man of the day :)
22:00:29 Join ackbahr [0] (
22:00:29Soul-SlayerGlad I could be of assistance.
22:01:02Soul-Slayerw1ll14m: Fair enough, but rather than a warning, why not an explicit cap?
22:01:27 Join Thundercloud__ [0] (n=thunderc@
22:02:17 Quit mathgl ("Quitte")
22:02:23l33chi use koss stereophones so i can relugate volume in wire :)
22:02:24 Quit _Veseliq_ (Remote closed the connection)
22:02:52l33chwhere can i notice better performance now ? :D
22:02:58 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
22:03:03l33chis there any plugin on which i can test it :)
22:03:12w1ll14msoul-slayer: hmm because sometimes there is a nice part in a song... that's where the cap is unwanted
22:03:38w1ll14mi mean a nice part, that is a lot less loud than the rest of the son
22:03:52w1ll14msorry for the english ;)
22:04:02Soul-SlayerWell, for instance, playing games whilst listening to music without the patch could cause the game to slow down, it shouldn't do so quite so much with the patch implemented
22:04:04l33chfor mine too :)
22:04:33w1ll14ml33ch: what is your target?
22:04:35Soul-Slayerw1ll14m: In which case, you could turn the cap off and have unrestricted control to volume still
22:04:47w1ll14msoul-slayer: true
22:05:07w1ll14mbut then you would have to go to menu's
22:05:10w1ll14mand change it.
22:05:30w1ll14mmy idea is to let the user scroll for 1 sec more, then it would enter 'boost' mode
22:05:34l33chw1ll14m: if you mean my mp3 player ? I own iRiver H10 :)
22:05:37linuxstb_Why can't you simply just not turn the volume up?
22:05:40w1ll14ml33ch: ok
22:05:51w1ll14mlinuxstb_: i agree ;)
22:06:03w1ll14mthis was probably a stupid idea....
22:06:06Soul-SlayerThirded. How did the debate start? :p.
22:07:31 Quit Thundercloud_ (Read error: 145 (Connection timed out))
22:07:38 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
22:07:38Soul-SlayerAnyhow. Glad I could help l33ch, I shall be back soon.
22:07:51Mikachuw1ll14m: why're you asking me?
22:08:26 Join Winchester345 [0] (
22:09:07w1ll14mmikachu: i thougt 'll just ask mikachu no spectaculair reason
22:09:54w1ll14msorry for the english ;)
22:10:00Mikachuwhat english? :P
22:10:14w1ll14mspectaculair i think i wrote it wrong ;)
22:10:42Mikachui know
22:10:46 Quit l33ch ("CGI:IRC (Ping timeout)")
22:11:20Winchester345if I writ a plug in..... do I put the c file in SOURCES or is it possible to compile the .rock file by itself ?
22:11:38linuxstb_You need to add it to SOURCES.
22:11:42Mikachuit would be nice if you could write "make foobar.rock" like in linux, but i don't think you can
22:12:18linuxstb_Feel free to add it... ;)
22:12:33Winchester345ok.... and... does that add it to the menu or does the menu scan all the .rock files int he plugin folder?
22:12:37Winchester345*in the
22:12:46linuxstb_The "menu" is a file browser.
22:13:09XavierGrRockbox lacks an internal scripting language! LOL
22:13:23Winchester345ah..... didn't think of it that way... but evidently it is
22:13:26XavierGrdo stuff like the TI calculators
22:13:41hcsI've been considering putting something like that together
22:13:57XavierGrscripting capabilities?
22:13:57hcsI've even got a pun on BASIC for the name: LITHIC
22:14:07 Join shayx [0] (n=shay@unaffiliated/shay)
22:14:09XavierGrwell it would be sure nice to have one
22:14:20shayxhello folks, I think that I screwed up an iPod
22:14:35XavierGrlike making sort programs that can do series of stuff on the fly
22:14:39shayxtrying to install a firmware with ipodpatcher, I guet this error
22:14:40shayx[ERR] Firmware partition doesn't contain Apple copyright, aborting.
22:14:41shayx[ERR] Failed to read firmware directory - nimages=0
22:14:54linuxstb_shayx: What kind of ipod?
22:14:57shayxI just dd'ed and MBR image
22:15:13shayxlinuxstb_: hey there, remember me from yesterday?
22:15:33shayxlinuxstb_: it seems to be a 4th generation, 20G iPod (graysclale)
22:16:01linuxstb_Of course, is your MBR being written now?
22:16:19shayxlinuxstb_: just wrote it
22:16:24shayxshay@virginia:~/temp/rockbox$ sudo dd if=mbr-4g-20gb.bin of=/dev/sda1
22:16:39linuxstb_So fdisk shows the partition table correctly?
22:17:04shayxoh wait
22:17:10Mikachushayx: not sda1
22:17:12shayxI think that I got the error
22:17:19shayxI should dd to the 'root` partition
22:17:22shayxthanks Mikachu
22:17:36mattzzQuestions to ipod users: what is the PLA_QUIT button in the sim?
22:17:38Mikachuit's too bad you can't write the absence of a 1 with a large font
22:17:39w1ll14mshayx, i think you have damaged the firmware partition
22:17:51 Join fasmaie [0] (
22:17:51linuxstb_Michacku: Read from about 13:41 if you want Chapter One of shayx's problems.
22:18:08Mikachuthat's a new spelling
22:18:16 Part fasmaie
22:18:26shayxw1ll14m: sod I do
22:18:28linuxstb_:) Web client's tab completion sucks...
22:18:31shayx /clear
22:18:40shayxw1ll14m: s/sod I do/So do I
22:18:47shayxforget my lapsus mentis
22:18:56linuxstb_shayx: I thought you determined your HDD was read-only and you couldn't write the MBR to it.
22:19:14shayxlinuxstb_, Mikachu: I still get the same error
22:19:26Mikachudid you copy the firmware partition to sda1 again?
22:19:29shayxIs there any way to fix a damaged firmware partition?
22:19:46linuxstb_shayx: Is your partition table correct? You can't do anything without that...
22:19:52w1ll14mdd a image from ipod firmware
22:20:07shayxMikachu: tried to, got the same error
22:20:44w1ll14mshayx: can you use itunes to recover your ipod ?
22:20:45shayx"fdisk -l" shows the followin:
22:20:46shayx/dev/sda1 1 5 40131 0 Empty
22:20:47shayx/dev/sda2 6 2431 19486845 b W95 FAT32
22:20:59linuxstb_Then that's perfect - just do what I suggested yesterday.
22:21:03shayxnot really, I don't have any windows computer
22:21:12w1ll14mjust a sec.....
22:21:20*shayx checks the log
22:21:27w1ll14mget your firmware here:
22:21:35w1ll14mit a zipped file, unzip it
22:22:00 Quit ackbahr (Read error: 104 (Connection reset by peer))
22:22:09shayxgreat, I'll check it out
22:22:26w1ll14mshayx: then dd Firmware- to /dev/sda1
22:22:37w1ll14mi thougt that was the way to restore you fw partition
22:22:40shayxit's a binary file
22:22:49w1ll14mit's a binary zip
22:23:01w1ll14mif you extract it you will get two files
22:23:07shayxw1ll14m: okay, I'll try, thanks
22:23:23shayxthen I'll be able to install Rockbox's firmware?
22:23:31w1ll14miPod_10.3.1.1.ipsw contains 2 files: manifest.plist and Firmware-
22:23:38w1ll14mthen dd Firmware- to sda1
22:23:57w1ll14mthan you would have the apple copyright again
22:24:12w1ll14mit it has written well to sda1
22:24:27w1ll14mand indeed then you should beable to use ipod patcher again
22:24:32 Join ackbahr [0] (
22:25:08shayxI am also interested on join the development of Rockbox
22:25:14shayxI have a device that is not supported
22:25:28w1ll14mwhich device is it ?
22:25:31shayxa Creative Nomad Jukebox Xtra (60 Gigas)
22:25:50w1ll14mcreative, i think you have to forget....... but you can try :)
22:26:12w1ll14mcreative devices are difficould to port... as far as i know
22:26:14shayxit has a rare processor, I have some experience with cross-compiling and embbeded development
22:26:31linuxstb_shayx: Here's our "guide" for new ports -
22:26:33w1ll14myou will need a cross compiler fot that processor
22:26:46shayxw1ll14m: guessed so
22:26:47shayxlinuxstb_: thanks
22:26:55w1ll14mshayx: :)
22:26:59linuxstb_What processor has it got?
22:28:42shayxgive me a sec, I'll finish installing this firmware to the iPod and I'll tell you all the research I've done so far
22:28:58shayxI'll upload all the links I found on the device to my account
22:29:13Bagdercould be a tms320 family
22:29:31*Bagder guesses
22:29:32shayxBagder: in fact, it is
22:29:45Bagderwhat do I win? ;-)
22:29:49shayxIt's a tms[something]
22:29:53linuxstb_Then is your friend.
22:30:04linuxstb_(or at least, their source code)
22:30:07shayxw1ll14m: the tried the 4.* and the 10.* firmwares, either of them work
22:30:17shayxlinuxstb_: I'm writing that down
22:30:21n1sBagder: you win an opportunity to be the first to port rockbox to it :-P
22:30:59w1ll14mshayx: did you unzipped those files ?
22:31:01 Join petur [0] (n=petur@rockbox/developer/petur)
22:31:06directhexi know a guy at TI. in the calculator software division, but he might know a guy who could be useful
22:31:09shayxw1ll14m: sure I ded
22:31:19 Join Kingsqueak [0] (n=who@
22:31:21 Quit ackbahr (Remote closed the connection)
22:31:24w1ll14mshayx: then i'm out of ideas ;)
22:31:35 Join ackbahr [0] (
22:31:36w1ll14mtry a friends computer to restore
22:31:44linuxstb_shayx: Did you also format the FAT32 partition?
22:31:46shayxI'm willing to give access to a machine that is connected to the Nomad player, if he wants to join me on the development
22:31:46w1ll14mwith mac or windows
22:31:57shayxlinuxstb_: no, on the firmware partition?
22:32:05 Part Kingsqueak
22:32:14linuxstb_No, "mkfs.vfat -F 32 /dev/sda2"
22:32:15shayxw1ll14m: I think that that's what I'll do, but I'll have to wait until tomorrow :(
22:32:24 Join Kingsqueak [0] (n=who@
22:32:36w1ll14mshayx: that sucks :(
22:33:09shayxlinuxstb_: same error icon :/
22:33:56linuxstb_What does ipodpatcher say now?
22:34:09linuxstb_It should tell you if your firmware image is OK.
22:34:23shayxlinuxstb_: should I −−scan or try to patch the rockbox firmware?
22:34:43linuxstb_Try "ipodpatcher /dev/sda −−list"
22:34:58shayxgreat, now it found the device
22:34:59shayx[INFO] Scanning disk devices...
22:34:59shayx[INFO] Ipod found - 4th Generation ("winpod") - /dev/sda
22:35:08shayxeven if the firmware doesn't really works
22:35:38linuxstb_That means that (at least as far as ipodpatcher is concerned), the firmware partition looks OK. You could try installing the Rockbox bootloader and see what happens.
22:36:02shayxI should use the /dev/sda or the /dev/sda1 device node?
22:36:24linuxstb_The device is optional. Just do "ipodpatcher −−install"
22:37:02shayx[INFO] Wrote 3563520 bytes to firmware partition
22:37:02shayx[INFO] Bootloader installed successfully.
22:37:07shayxI like those messages. :-)
22:37:19shayxI started to forget how fun is to work with embedded
22:37:39shayxsucks, same error icon
22:37:46shayx(used ipodpatcher −−install)
22:38:00linuxstb_It's definitely a greyscale ipod?
22:38:29shayxfor sure
22:38:49shayxwhat about the mbr-4g-20gb.bin file?
22:38:58linuxstb_ipodpatcher identifies an ipod from the firmware image - so that means you've installed the 4g greyscale firmware image.
22:39:15shayxI understand
22:39:15linuxstb_And yes, that MBR is from a 20GB greyscale.
22:39:30shayxwhat might be the error?
22:39:33 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
22:39:39shayxit doesn't work with both rockbox and apple's firmware
22:40:36linuxstb_What exactly is the error screen?
22:40:53shayxI'll find a screenshot on google, sec.
22:40:57 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:41:06linuxstb_After you install a new Firmware image, it should ask you to connect your ipod to a wall charger so it can update the flash contents.
22:41:16 Quit amiconn (Nick collision from services.)
22:41:17 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:41:56mattzzAny plugins that need verification? −−>
22:42:09mattzz(just a proposal)
22:45:25 Part n1s
22:46:23linuxstb_shayx: Try writing the Firmware-X.Y.Z file to your ipod again, and check the Linux error logs for problems.
22:46:52shayxI'm still not sure if I should use the 10.* or the 4.* firmware version.
22:47:01shayxalthought I tried both of them
22:47:34linuxstb_Are they different?
22:48:17shayxI dont't know, I think that it deppends on the generation of the iPod
22:48:57 Quit mattzz ("Leaving")
22:49:20linuxstb_Just do a diff between them, or calculate their md5 checksums.
22:49:45shayxthe dd'ed partition of the firmware, (the firmware partition) is /dev/sda1, right?
22:49:55shayxlinuxstb_: diferent sizes, they're surely different.
22:50:08shayx(didn't know that dd worked on binary files)
22:50:50 Join JustJohnny68 [0] (
22:51:35 Quit Rondom ("Ex-Chat")
22:51:52 Join RogerBacon [0] (
22:52:01shayxoh wait, there's no difference between the Firmwares (10.* and 4.*)
22:53:48 Quit OgMaciel ("Ex-Chat")
22:54:14shayxnothing seems to work... any other clues?
22:57:34linuxstb_Sorry, no. Does the ipod pass the diagnostics tests?
22:59:10shayxlast time I checked (yesterday) it did, let me check again.
22:59:26 Quit dpassen1 (Read error: 145 (Connection timed out))
23:02:18shayxsighs, doesnt want to enter diagnostics mode.. it jumps directly to the error icon...
23:02:26shayxI need a cig, brb
23:02:30 Quit Mouser_X (Read error: 145 (Connection timed out))
23:02:31 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
23:02:37Mikachusometimes it just wants some charging
23:02:55 Quit ackbahr (Remote closed the connection)
23:03:26perl|work"Make database endianess independent."
23:03:35perl|workisn't "endianness" ?
23:03:39 Join ooAlexoo [0] (
23:03:56Mikachumaybe it's a female endian
23:04:12Mikachuor should we call them native emericans?
23:12:24shayxokay, it entered on diagnostics mode, any specific diagnotic I should do?
23:12:36***Saving seen data "./dancer.seen"
23:13:10linuxstb_Whatever you fancy...
23:16:22 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
23:17:46 Join p0lak [0] (
23:20:44 Join funky [0] (n=repulse@unaffiliated/funky)
23:21:12 Join lostnihilist [0] (
23:23:31Winchester345if I have modified files in the rockbox folder.... running svn update replaces them with the latest files right?
23:23:54Bagderit'll merge your changes with the new ones
23:24:19 Join webguest55 [0] (i=43bbe7a0@gateway/web/cgi-irc/
23:24:28webguest55@find rockbox
23:24:34webguest55I know that don't work
23:24:46Bagderit did
23:24:49BagderI saw it
23:25:04Bagderyou found #rockbox!
23:25:07webguest55I wanted to know why rockbox freezes when certain songs are playing
23:25:41Soul-SlayerWhich codec? What bitrate? Do they work ok on your computer?
23:26:08webguest55yea the songs work fine anywhere else and even with the reg, ipod firmare
23:26:15webguest55they are mp3
23:26:39webguest55I don't know the bitrate, my guess would be 128 bit
23:26:52Soul-SlayerSo this happens with more than one song?
23:27:47webguest55yea, I don't recall which actual songs, but it does happen. nothing will work but the next song will come one and so on untill the folder is done
23:28:03Soul-SlayerSo it skips the song?
23:28:12Soul-SlayerWhich iPod do you have?
23:28:49webguest55nope, plays right through until the folder has no more songs. No functions work though and the screen is frozen. I use the 30Gig Video
23:29:12 Part perl|work
23:29:14Soul-SlayerAhh I see, so it locks up your controls but still plays the songs?
23:29:46webguest55i should have just said that huh?
23:30:05Soul-SlayerHave you tried using a different WPS?
23:30:12shayxhm, on the Flash memory diagnostics, it gets stucks on one of the checksums, can be this a pointer to my problem?
23:30:45shayxignore my comment..
23:30:50webguest55I use the default rockbox theme with wps, just the backdrop is custom
23:31:09Soul-SlayerCould you try using the iCatcher theme, and playing a song again?
23:31:28webguest55i haven't seen this happen with anyother wps or theme.
23:32:11webguest55I'll change the theme right now and see what happens, give me a minute.
23:32:23Soul-SlayerOkay, to my knowledge (I haven't used it in some time) the Rockbox default theme uses peak meters, which always used to make my iriver H10 either stutter on music or lock up... I'm not sure if this got fixed or not, but it could be something to look into
23:33:10 Join lee-qid [0] (
23:33:32hcsthere seemed to be agreement on a default without the peakmeter when it was brought up yesterday
23:34:11Soul-SlayerI too would agree with that
23:34:27webguest55I think that was it. nothing locks up, but now nothing shows in the wps. no music info or volume indicator.
23:34:35Winchester345how does bmp2rb work for grayscale screens? do the images come out nicely shaded or does the bmp need to be made with the 3 shades to look right?
23:34:54Soul-SlayerI suggest you look around for themes people have made for your ipod and find a WPS that you like the look of :).
23:35:01Soul-SlayerOr even make your own to your needs
23:35:41webguest55yea I guess thats what I am going to have to do. thanks for your help. Also do you know why the battery dies really fast?
23:36:12safetydanwebguest55, either your battery is old, or your using Rockbox on one of the targets that we haven't really got to grips with power saving on yet
23:36:40Soul-SlayerNumber of things really, Rockbox isn't as optimised as it could be at the moment, also it could be equalisers, pitch settings, high volumes, all sorts of things have a factor
23:36:49webguest55the latter might be the problem because it's a new iPod
23:36:59Soul-SlayerConstantly using high consuming plugins like mpegplayer or doom and so forth
23:37:12Soul-SlayerThat would be it then
23:37:32safetydanwebguest55, well how fast is fast? At the moment I think battery life on iPod's under Rockbox is about 50% of what you would get under Apple's firmware
23:37:38webguest55i don't play games, i did see a setting in rockbox but didn't understand. batter limit or something like that.
23:37:58Soul-SlayerThat is merely for approximating how much battery life remaining.
23:38:31webguest55batter capacity, what does this setting do if changed. right now it's on 1300
23:38:35webguest55that was default
23:38:40Soul-SlayerIf you go into the Info screen, it will give you a % and an approximate time remaining, changing your battery size in the menu will just modify how it calculates that.
23:38:50directhexwebguest55, changes how long rockbox predicts your remaining battery life for
23:39:11Soul-SlayerLeave it on it's default, otherwise you will start getting erratic battery approximations
23:39:37 Join Domonoky [0] (
23:39:52LloreanNot that they aren't erratic now.
23:39:53webguest55ohh. thanks for your help soul-slayer and safetydan
23:39:59Soul-SlayerLol, true enough.
23:40:21Soul-SlayerBest thing you can do is to sit it out, keep an eye on new builds and see if any optimisations have been made for your build
23:40:22directhexLlorean, t'is remarkably accurate with "6 minutes" for the other half's h320
23:40:27webguest55is this something new, or has this happened before
23:40:49webguest55the battery thing
23:40:52Lloreandirecthex: Yeah, but it's uncalibrated for iPods.
23:41:12Soul-SlayerBattery life for new targets always starts out poor
23:41:16Lloreanwebguest55: Rockbox has poor battery life on the iPods right now. It's not a 'thing' it's just the state of the software.
23:41:31webguest55ohh. well cya later
23:41:35 Quit webguest55 ("CGI:IRC")
23:42:02 Quit My_Sic (Read error: 131 (Connection reset by peer))
23:43:25 Quit midgey ()
23:43:54LloreanjhMikeS: SVN.
23:44:38safetydanhrmm... so it's the iFP port that added oggmalloc
23:44:44jhMikeSLlorean: say again?
23:44:47safetydanI wonder why he did that rather than just add the functions to codeclib
23:45:00LloreanjhMikeS: Your most recent comment on the SPC codec. ;)
23:45:28Lloreansafetydan: Could it have some benefit for the fact that it's an extremely low memory target?
23:45:32jhMikeSoops...hrm...old habits. my fingers just type it by themselves. ;)
23:45:36LloreanI know the feeling
23:46:05safetydanLlorean, afaics the malloc stuff in oggmalloc is identical to the codeclib code.
23:46:14safetydanThe only change is the tmpmalloc code
23:46:32*safetydan has been fighting with Tremor and Speex ogg parser reuse
23:47:43 Join dan_a_ [0] (n=dan-mirc@
23:47:48 Quit dan_a (Read error: 104 (Connection reset by peer))
23:47:48safetydanAt the moment, Tremor has its own allocator which works well enough when it's the only thing using the ogg parser. It dies in weird ways when you try and reuse it with code that use the codeclib allocator.
23:48:47 Join sdfdsa [0] (
23:48:55sdfdsaanyone know where i can find chip8 games?
23:49:57Soul-SlayerTheres a link on the rockbox site somewhere I think
23:50:01 Nick logiclost is now known as lostlogic (
23:50:06Lloreansdfdsa: Googling Chip8 games seems to work well too
23:50:41sdfdsaive been googling it
23:50:47sdfdsabut i keep finding amulators and thats it
23:50:50Soul-SlayerLlorean: Problem with that is the ones on the rockbox site have keymaps set up for them I think
23:51:39Soul-SlayerI'll see if I can find the page
23:52:04Soul-Slayer3 links at the bottom of that page.
23:52:24 Quit dpassen1 (Read error: 145 (Connection timed out))
23:56:33 Part Llorean
23:57:02sdfdsawhat about gameboy games?
23:57:07sdfdsawhere do you find those?
23:57:13sdfdsaare those easy to run with rockbox?
23:57:15hcsjhMikeS: ooh, 37.5
23:57:22Soul-SlayerUnable to disclose that
23:57:34Soul-SlayerBut they run simply by opening it in file browser
23:57:49 Quit petur ("Zzzzz")
23:58:51DataGhost <- short but exciting... anyway, that's a HFS+ disk read on a 80GB iPod :)

Previous day | Next day