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 2007-07-04

00:00:13amiconnOne could try and change the order of x and y in the function header, but I doubt that'll have much effect
00:00:39amiconnI also wonder whether it's possible to tell gcc that some function arguments are commutative
00:01:13preglowit is
00:01:21preglow%, or something
00:02:02Learsaratoga: Where did you add the calls?
00:02:05***Alert Mode OFF
00:02:12amiconnI am talking C level...
00:02:35 Join rotator [0] (n=e@rockbox/developer/rotator)
00:03:14preglowoh, right, function arguments
00:03:17preglowi doubt that
00:03:48amiconnI mean if you have int blah(int x, int y) whether you can tell that gcc might switch arguments without invalidating the result
00:03:59amiconnIt knows such things for operators
00:04:47 Quit ender` (" .......It might be written "Mindfuck", but it's spelt "L-A-I-N".......")
00:04:51amiconnHmm, the second version is sligthly faster than the first (90.72% realtime), but not as fast as svn (91.36% realtime) :(
00:05:03 Quit bdgraue (Remote closed the connection)
00:06:10 Part maffe
00:06:26preglowi don't get why it should be slower
00:06:33preglowhave you tried test_codec several times?
00:06:45 Quit jhulst_ (Read error: 104 (Connection reset by peer))
00:08:22saratogapreglow: I ran test codec then tried playing it back, neither did anything
00:08:31saratogait played at half speed on the gigabeat and the sansa
00:08:34 Join maffe [0] (
00:08:41saratogaerr, half of the speed reported by test codec i mean
00:08:56saratogai assumed this was because it was counting calls to functions
00:09:02preglowi've never tried profiling on an arm target
00:09:11preglowonce upon a time, it was unsupported, are you sure it's supposed to work now?
00:11:00saratogaI was able to do it last year
00:11:12saratogavery long time ago when I was first reading about rockbox
00:11:19saratogaon my ipod 3g actually
00:11:31saratogabut maybe its broken since then?
00:12:26preglowpossible, but who knows, i haven't tried profiling for a long time
00:12:58 Quit obo ("bye")
00:13:00 Quit desowin (Client Quit)
00:15:03 Join webguest26 [0] (i=44710737@gateway/web/cgi-irc/
00:15:25preglowgcc 4.3 has tons of coldfire changes, i wonder if we'll benefit
00:17:18mario_can I have a link to a download for iboy
00:17:26mario_can't find 1
00:17:28 Quit Bagder ("*plopp*")
00:19:06 Quit freqmod (Remote closed the connection)
00:19:13saratogamario_: ask the ipodlinux people
00:19:18LearTried a 4.3 snapshot. Don't remember exact result, but I don't think it was better than 3.4.6...
00:21:30preglow4.3 has completely new march and mcpu flags, did you change those3?
00:22:22LearSure, won't build otherwise.
00:22:37 Quit webguest26 ("CGI:IRC (EOF)")
00:22:42preglowhrmf :(
00:22:43amiconnpreglow: Multiple runs give exactly the same results (+/- 0.01%)
00:24:13preglowi wonder why...
00:24:21preglowi would certainly expect that to be faster
00:25:59 Join BHSPitLappy_ [0] (
00:26:30LearTo tell the truth, I actually don't remember if I got a complete build, so I might not have tested speed. Not better size at least. ;)
00:27:44amiconnpreglow: Perhaps cache aliasing changes. Might change when other parts of the codec change
00:28:14 Join jhulst [0] (
00:28:20 Quit saratoga ("CGI:IRC (Ping timeout)")
00:28:56CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:28:56*amiconn is running a runtime test on ipod video, btw
00:29:16amiconnIf the percentage is roughly calibrated, this looks promising...
00:29:39 Join atsea- [0] (i=ariel@gateway/tor/x-91679a7a53e6850d)
00:29:59LearNow I remember. Some code bloats, so IRAM gets full. Didn't actually fix that, since it might affect speed...
00:31:02 Quit mario_ ("Leaving")
00:31:32amiconnHmm, no seeking support yet... :|
00:32:56 Part maffe
00:33:01 Quit Thundercloud (Remote closed the connection)
00:33:26 Join maffe [0] (
00:38:48amiconnpreglow: Switching y and x makes this faster than svn...
00:39:22 Join webguest57 [0] (i=522cfa11@gateway/web/cgi-irc/
00:39:37preglowby how much?
00:39:52amiconnBy about the same amount as it is slower without switching
00:40:00 Join bb975347y [0] (i=522cfa11@gateway/web/cgi-irc/
00:40:42 Quit BHSPitLappy (Success)
00:40:47 Quit Quelsaruk ("KVIrc 3.2.6 Anomalies")
00:44:22 Quit webguest57 ("CGI:IRC (Ping timeout)")
00:45:00 Quit bb975347y ("CGI:IRC (Ping timeout)")
00:46:12peturanybody know a good tool to analyze FAT directories? (look at +/- raw dir entries)
00:51:02 Join CPrompt^ [0] (
00:51:58CPrompt^how would you install the patches to use something jBuild? does it just go in the .rockbox directory?
00:52:39pixelmapetur: btw. I just noticed today that the properties don't work a 100% on the virtual <MMC1> folder (the root of the MMC) - don't know if it's supposed to). It starts analysing but the result doesn't stay on the screen like on every other folder, subfolders work though even on the MMC
00:52:43stripwaxCPrompt^ : - first you install the build tools, then you obtain the rockbox source code, then you apply the patches according to the instructions in the wiki, then you rebuild rockbox
00:52:49stripwaxCPrompt^ : - did you read the wiki?
00:53:16CPrompt^yes I read the wiki. I have rockbox installed. I was wanting to up it to jBuild
00:53:34stripwaxCPrompt^ : - do you have the *build tools* and the *source code* installed
00:53:58CPrompt^I had jBuild installed once before, I don't remember rebuilding the source
00:54:11 Quit Lear ("ChatZilla [Firefox]")
00:54:19peturpixelma: it tries to use dircache, maybe that's the problem? Or does dircache do a passthrough for hotswap?
00:54:33stripwaxCPrompt^ : - do you just want to install JBuild on your device or do you want to compile it (i.e. are you a developer or just a user?)
00:54:41*scorche notices the usage "up it to jBuild"
00:54:42CPrompt^just install it.
00:54:47pixelmaperplexity: dircache on Archos?
00:54:56pixelmathat was to petur
00:55:22stripwaxCPrompt^ : - in that case you don't need to know or care about patches. just go install JBuild.
00:55:22peturpixelma: and I don't quite understand the last part of your remark
00:55:57stripwaxCPrompt^ : - note that JBuild is not a supported build of rockbox (i.e. don't ask questions about JBuild here, since the rockbox developers don't have any input in the JBuild build)
00:55:59CPrompt^stripwax : From the themes page this is under one of the themes I'd like to use "Notes:
00:56:00CPrompt^ *Recommended builds:* Rockbox jbuild or EvilG's build (Fusion flavor)
00:56:00CPrompt^Album Art, Customline, Multifont, Multifont-userfonts & BMP resize Patches required."
00:56:14CPrompt^good enough
00:56:17stripwaxCPrompt^ : - I don't understand what you are trying to ask.
00:56:17pixelmapetur: it works correctly for any subfolder of the MMC, just not the "root" of the MMC
00:56:27 Quit Febs (Read error: 110 (Connection timed out))
00:56:48CPrompt^don't worry about it. I'll find answers elsewhere sine I'm not supposed to ask here ;)
00:56:51 Part CPrompt^ ("Leaving")
00:57:23pixelmapetur: first it looks ok (analysing the content) but the result doesn't stay on the screen
00:57:52peturit just goes away?
00:58:06peturall the text?
00:59:05pixelmaumm... weird. Just tried with a different MMC and it works correctly
00:59:50peturit does an opendir and then recursively iterates through all subdirs, so I wouldn't expect any difference
00:59:53 Join Febs [0] (
01:00:33peturexcept for a stackoverflow in very deep nested trees (?)
01:00:58pixelmatime for a file system check, maybe. I know that the first MMC is a bit fragmented but was told that this doesn't matter much
01:01:58pixelmaI have lots of very small files on the non-working one (the svn manual folder)
01:02:43peturshouldn't matter
01:03:39peturwhen the screen goes empty, is it still scanning or did it hang/crash/exit?
01:04:22***Saving seen data "./dancer.seen"
01:05:20pixelmajust exits
01:05:37peturno error?
01:05:58pixelmano, it just returns to the file browser
01:07:00peturheh, I see it just exits on error (PLUGIN_OK)
01:07:15pixelmathere are a few files in quite deep nested folders
01:07:28peturmaybe a little splash or text would be good there
01:11:18pixelmaaha... I've moved the some folders 2 levels up and that helped
01:19:16peturpixelma: the version committed now should now print that it failed. Doesn't print why though...
01:19:34peturat least it's better than just exiting
01:20:36peturmaybe deeply nested trees on a target with little memory are the issue... is the main stack on archos smaller?
01:20:43 Quit Febs (Read error: 110 (Connection timed out))
01:22:21pixelmathe limit seems to be 8 levels deep - don't know about the main stack...
01:22:33*gregj gives up on trying to figureout serial ports on G4, without external help
01:22:40amiconnThe 8 levels is from the # of open dir handles
01:22:50peturah right...
01:22:55gregjamiconn: I was trying to view some serial port registers, but it doesn't ork
01:23:15linuxstbgregj: What kind of external help are you expecting?
01:23:34gregja hint, or give me some way to approach it
01:24:13amiconnIf it were easy, someone would already have done it
01:24:35linuxstbgregj: I would suggest disassembing the apple firmware and/or the diagnostics mode code.
01:25:38gregjyou want me to grab the first partition, and run through objdump ?
01:25:52amiconnpetur: Why not just a splash?
01:25:55 Join Olav [0] (
01:26:20linuxstbgregj: Not the entire partition, just the firmware.... ipodpatcher can extract the firmware for you.
01:26:58linuxstbgregj: Also, this page describes how to dump the diagnostics mode code from flash -
01:27:13gregjsee, that's the helping hand, thanks
01:27:16peturamiconn: because it can take some time to go through a big tree so I found it better to leave the message there because that's what the user is expecting anyway (info on the screen when it is done)
01:27:30amiconnThen you can still use a splash...
01:27:52amiconnJust splash with a timeout of 0 (immediate), then do the action_userabort thing
01:29:02 Join saratoga [0] (i=41becbf6@gateway/web/cgi-irc/
01:29:18 Quit Siltaar ("Aurevoir -")
01:29:39linuxstbgregj: Do you have any hardware you can attach to your ipod's serial port to see if it's doing anything?
01:29:55peturamiconn: are you feeling very strong about this or just a bit picky? ;)
01:29:58gregjI have the pilot
01:30:06gregjyou know, the one i am hoping to get running
01:30:22amiconnpetur: Why simple when you can do it complicated? ;)
01:30:30linuxstbgregj: What's that?
01:30:36amiconnReally, it's just about consistency
01:30:46peturalright alright
01:31:02amiconn(I know there are some places in the core which still don't use splash)
01:31:11saratogalinuxstb: when you get a chance, could you try profiling the WMA codec on coldfire?
01:31:36saratogai can't seem to do it on ARM
01:31:36gregjlinuxstb: ipod g4/photo's pilot
01:32:41linuxstbsaratoga: I don't think I'm going to have time for that in the next few days (I'm going away for four days on Thursday morning).
01:33:24linuxstbgregj: I still don't know what it is...
01:34:31gregjlinuxstb: next to the headphone jack, there's little slot, 4 pins
01:34:32preglowsaratoga: is it important anyway?
01:34:58*amiconn hates when things behave the opposite of what he thinks :(
01:35:20gregjlinuxstb: that's the one used by remote, someone mentioned here - this stuff is using serial port (I find it strange, because there's huuuudge slot on the bottom)
01:35:28amiconnTried replacing the bswap_* functions with rockbox' optimised swap*
01:35:35linuxstbgregj: OK. I was talking about something which would show you any signals on the serial port, not an actual accessory.
01:35:48amiconnThe darn codec becomes slower when doing this - on coldfire and arm...
01:36:04linuxstbSo our byte-swapping functions are slow?
01:36:29amiconnI don't think so
01:37:00amiconnThe difference isn't large, so it might be cache aliasing, or caching in general
01:37:08amiconnOur functions are inline
01:37:11 Quit lee-qid ("aufwiederbyebientotsayonara")
01:37:27gregjlinuxstb: the actuall acessory is what is usefull
01:37:44amiconnDifference is ~2% for arm, and ~0.5% for cf
01:38:05linuxstbgregj: Not when you're trying to make undocumented hardware do something....
01:38:22gregjshit, this mpeg player is fscked up
01:38:31 Quit jhulst (Read error: 104 (Connection reset by peer))
01:38:33gregjnow it produces funny noises
01:38:52amiconnOur swap16 is in fact the same C code (only on SH it's not), but swap32 is optimised for all architectures
01:39:01gregjmust be me, trying to encode sound into some strange format tht it doesn't recognize
01:41:13preglowi hate cache effects :/
01:41:16preglowdifficult to predict
01:41:31peturhmmm the previous commit on properties plugin saw the gigabeat bin size increase with 56 bytes...
01:41:31preglowpetur: tabs :-)
01:41:38amiconnpreglow: Btw, arm has mac instructions as well. Do you think those would help CMUL ?
01:41:51preglowamiconn: yes, i was intending to do that
01:42:00preglowpetur: your last commit has tabs
01:42:00gregjand it is funny, after a while - the mpeg player always stops sound playback (video still running smoothlee) - and doesn't respond to key events
01:42:27peturHow the hell... Anjuta is configured to use spaces... :(
01:42:34preglowamiconn: tremor pretty much already has the function we need, it's called XPROD31 or something
01:42:44preglowamiconn: which is the place i took the coldfire CMUL from, really
01:42:55*linuxstb can't see bswap_* being used
01:43:45*amiconn wonders whether he should commit the 1337 fixmul32() for coldfire, or rather fix the svn version to work for different PRECISION s
01:43:58preglowcommit the 1337 one
01:44:26preglowchanging PRECISION is not going to happen often, if ever
01:44:34gregjamiconn: there are some ports defined already there for ser0 and ser1
01:44:46amiconnYeah, and?
01:44:51gregjamiconn: which I presume - after what you guys have said here - is the serial port, and responsible for that remote
01:45:14gregjamiconn: so it should be "just" a matter of grabing firmware, disassembly, and finding places where it is being used, right ?
01:45:47amiconnEven though the port values are known (also from disassembling I presume) the ipl guys didn't manage to get it working
01:46:04 Quit Xerion (" ")
01:46:22amiconnIiuc there is some necessary init missing
01:47:11 Quit DataGhost (Nick collision from services.)
01:47:16gregjprobably, me guesses - irq mask ?
01:47:17 Join DataGhost [0] (
01:47:39*gregj reads all he can about 5020, it suppose to have jpeg decompressor, or idct stuff built in
01:47:54amiconnThat it doesn't
01:48:02gregjthat would explain why jpegs are opening up so fscking fast on apple's firmware, while here they are terrriiibbblllyyysllooow
01:48:32gregjit's a shame tho, that portaplayer site doesn't ork anymore
01:48:34linuxstbI didn't think Apple's firmware viewed jpegs, just uncompressed bitmaps converted by itunes.
01:48:40 Join DC1 [0] (
01:48:43*gregj goes for wayback machine
01:48:52gregjlinuxstb: it stores them as jpegs on my ipod
01:50:28preglowitunes does resize, that i'm almost certain of
01:50:47peturor at least creates thumbnails
01:51:59gregjyes, it does
01:52:06amiconnpreglow: Btw, is there any special reason why the arm fixmul32() is a macro while the cf version is a static inline fn?
01:52:11 Quit DataGhost (Nick collision from services.)
01:52:14 Join [Sur`DataGhost] [0] (
01:52:14*amiconn thinking about consistency again
01:52:52gregjI guess the reason PP didn't gave out any specs, is because probably even apple doesn't have them
01:53:08gregjPP delivers full RTOS with its chips, and SDK
01:53:12gregjso it says on their website
01:55:06preglowamiconn: no reason, no
01:55:19gregjoh, mpeg player hanged :) now there's reason for reboot
01:55:47 Quit petur ("Zzzzz")
01:58:34 Quit spiorf (Remote closed the connection)
01:59:46 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
02:00:52 Quit atsea- (Remote closed the connection)
02:01:44 Join stripwax_ [0] (
02:01:52 Quit bluebrother ("leaving")
02:03:39 Quit printfXh4 (Connection timed out)
02:04:26 Quit stripwax_ (Read error: 104 (Connection reset by peer))
02:04:50 Join atsea- [0] (i=ariel@gateway/tor/x-9c137c79b2c4247e)
02:06:37 Join sheenzy [0] (
02:07:56sheenzycan someone help me i loaded rockbox onto my iriver h320 and now it wont connect to my puter
02:08:02gregjhow certain you guys are PP5020 is unable to do any jpeg,xDCT,whatever things in hardware ?
02:08:24sheenzyany ideas?
02:08:36 Join andrew__ [0] (
02:11:35 Quit thegeek_ ()
02:11:39 Join addingmusic [0] (i=474872e4@gateway/web/cgi-irc/
02:13:41addingmusicif i convert my itunes library to a rockbox library will it just reorginize my files or actually convert the type of files they are?
02:14:32 Quit GodEater (Read error: 110 (Connection timed out))
02:14:45krazykiter, it'll just make a database. it won't do anything to the files themselves
02:15:09addingmusici meant reorginize sorry about that....wait did you mean reorginize?
02:15:49addingmusichow do i add other music can i drag and drop or something?
02:16:14krazykityes, drag and drop. but the original firmware won't recognize it dragged and dropped music
02:17:23addingmusicyeah i know but i have wma files and i dont want to convert them, so drag and drop into the /rockbox folder somewhere or into the ipodcontrol/music? And will it mess with the database i have already inialized?
02:17:36krazykitno, no
02:17:48krazykitdon't put it in .rockbox or ipodcontrol
02:17:49krazykitmake a new dir
02:17:58krazykitand it should add those files to the database
02:19:09addingmusicoh, sorry i just cant find where it explains how to do it.... so just maek a new folder in the root called music or something?...
02:19:23krazykithowever you want
02:19:33gregjlinuxstb: can you try it on your ipod please ?
02:19:37 Quit stripwax (Read error: 110 (Connection timed out))
02:19:59 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
02:20:01addingmusicwell i cant see it hold on what is it?
02:20:02gregjlinuxstb: its only the first 1.5m of offending mpeg, but it gives me a headache here
02:20:39gregj/tmp/ccfWYRAf.s:31: Error: invalid literal constant: pool needs to be closer
02:20:45addingmusicwell it wont load
02:21:15addingmusickrazykit: ok ill try it thanks for the help sorry if i have bothered you
02:21:42krazykitno bother
02:21:59krazykitaddingmusic, wma support is still young
02:22:07krazykita few days old young :P
02:22:26 Quit miepchen^schlaf ("Verlassend")
02:22:53 Quit andrew__ ("leaving")
02:22:59 Join andrew__ [0] (
02:23:09addingmusicyeah i know but still its worth a shot :)
02:25:03 Quit preglow (
02:25:05addingmusicnext i might try to do is add album art but i know howq to do it so it shouldnt be too hard
02:25:12*gregj goes to sleep
02:25:41NJoinpreglow [0] (n=thomj@rockbox/developer/preglow)
02:27:11addingmusicok well that doesnt seem to work it can identify the file just not play it right.... ill just have to convert i guess
02:27:23pixelmaJdGordon: morning... hope you'll have nice read ;) /me saying g'night though
02:27:54JdGordonjust read the comment
02:27:56JdGordonso the patch works?
02:29:28pixelmayes, after it's allowed to write a path with a dot in it (yes even the dot before the extension...) - wonder how you tested it then
02:29:37 Join Febs [0] (
02:30:02JdGordonI wonder how the bug got it...
02:30:29addingmusicwhen you convert the itunes database will the rockbox database be selected in the database option or in the files option?
02:31:02 Join pietro10 [0] (
02:32:23pietro10I installed Rockbox onto my 5G iPod, and I can't see the menus
02:32:27pietro10How can I make the font bigger?
02:32:41scorchechange the font
02:32:58pietro10Is there any way I can do that from the computer?
02:33:13 Quit borisyeltsin (Remote closed the connection)
02:33:15Febsaddingmusic: you don't "convert the itunes database." You initialize Rockbox's database, which will find the music iTunes has put on your computer.
02:33:28 Quit DC1 ("$4E75")
02:33:29 Join borisyeltsin [0] (
02:33:34Febsaddingmusic: you then browse that music using the database. But the process doesn't have anything to do with the iTunes database.
02:33:44scorchepietro10: you can do that through editing the config file, but it is mugh easier ont he device itself
02:33:58pietro10Thanks for the config file tip
02:34:07pietro10The thing is that I'm visually impaired and I can't see the menus on the device
02:34:21pietro10Another question: does Rockbox play iPod video files (.m4v)?
02:34:29Febspietro10: No.
02:34:30pixelmaJdGordon: happened to me ever since you told me to use the fat check too, other users were in here complainig their database couldn't be written or problems with the playlist control file (also in the forums) - the error messages contained that -116 too
02:34:32addingmusic...oh i did that on the device itself but i read in the manuel or faq that you can do it some other way, sorry about that
02:34:40pietro10ok, thanks
02:35:03pietro10Finally, is the menu skinnable like the player window?
02:35:34FebsYou can change the background image and the font (though the same font must be used for the menus and the WPS).
02:35:36scorchenot as much, but you can change the font, background and icons
02:35:45JdGordonpixelma: ok, i admit i didnt try createing files with every illegal letter, but just so you know, that bug was always there... I didnt change the illegal char list
02:35:58*Febs forgot about the icons.
02:36:35pixelmaJdGordon: huh? it was able to create a blah.mp3 before
02:36:50JdGordonbecause the filenames were never checked properly
02:37:20JdGordonI think only folder names were checked
02:37:33JdGordonbut even there a dot is legal
02:38:12*amiconn wonders whether JdGordon confuses longnames and shortnames
02:38:42JdGordonprobably... but meh.. its all good now
02:38:53JdGordonand the rec patch will go in a little later today
02:39:01JdGordonNico_P: hey man, you still round?
02:39:17Nico_PJdGordon: yes, and I have a fex changes available soon
02:39:23JdGordoncool :)
02:39:31JdGordonI was gonna say, we ran out of steam :p
02:39:35Nico_Pnot much though, don't get too excited ;)
02:39:44addingmusicdo i need to download another build of rockbox to get wma working?
02:39:46pietro10What is the config file?
02:39:53Nico_PI was quite busy today... I hope to do more work tomorrow
02:39:56pixelmajust saying something because you said everything works on your sansa - with the fat check you should have had the same problem
02:40:38Nico_PI've just written a basic bufgetdata (no use of guard buffer or anything) and I'm going to test it now
02:40:59JdGordonthe rec patch did work... i only tested creating a folder with one of the illegal chars.. the vkeyboard is too painful to test everyone seperatly
02:41:34JdGordonNico_P: did you try anything with the skipping the metadata blocks? and linking 2 items?
02:42:03pixelmathat wasn't creating a folder with an illegal char - just really recording where it has to write to a file called blah.mp3
02:43:41 Quit Soap (Read error: 110 (Connection timed out))
02:44:03pietro10what is the config file that edits the fonts?
02:44:36pietro10oir at least the theme so I can choose a theme whose default font size is big?
02:45:54addingmusicdo you know how to get wma working on the ipod?
02:47:26pietro10addingmusic: convert it in iTunes
02:47:36pietro10just add the .wma to the library and it will let you convert it
02:48:12 Join piroko [0] (
02:48:27pirokoHail rockbox etc
02:48:38 Part pixelma
02:48:53addingmusicno i dont want to convert i want to be able to drag and drop all my music.... but if i have to i guess i will
02:50:12pietro10addingmusic: just like the iPod with the Apple firmware, you have to use the iTunes to put the music on.
02:50:31scorchepietro10: what are you talking about?!?
02:50:47pietro10I'm talking about the fact that iTunes converts wma to mp3
02:51:01addingmusicok well i know your wrong
02:51:06scorcheitunes is not required with rockbox
02:51:17addingmusici can drag and drop and select from the file menu
02:51:42pietro10Sorry, I'm using a Mac
02:51:53scorchepietro10: that doesnt matter
02:52:00addingmusicfrom what i understand, and i also know wma support was implemented and i would like to learn how to use it.....
02:52:01pietro10ok, I'm very confused
02:52:04pietro10just forget it
02:52:25scorcheaddingmusic: wma has just barely been is able to be played (although perhaps not in real time) as long it is not DRM encumbered
02:52:31scorcheplay it just like you would any other file
02:52:50pietro10good :-)
02:52:57pietro10so where is the config file?
02:53:05addingmusici tried i made a folder in the root called music and selected the file but it didnt do anyhting...
02:53:12pirokoaddingmusic: Non-DRM encumbered means you didn't buy it from the iTunes music store, in case you didn't know
02:53:38scorchepiroko: it means more than that and itunes doesnt sell WMAs in the first place
02:54:03addingmusici know what DRM is
02:54:08pirokoOh sorry. Shutting up now :-D
02:54:27scorcheat this rate, in a few hours, i can get everyone in this channel to shut up...
02:54:42addingmusicso can i just try to make a folder then select from the file menu?
02:54:49pirokoThat's not very nice :P
02:54:51addingmusicdo i need to install a new build?
02:54:59scorcheaddingmusic: yes
02:55:44addingmusicwhere do i find it just click current build and wonload the ipod 3ogb again and reinstall everything?
02:56:07OlavHow is it going with the zen vision:m firmware? Still stuck (as far as I know)?
02:56:13scorchejust extract teh .rockbox folder over everything
02:56:32scorcheOlav: the forum thread on it is as far as most of us know, likely
02:56:44addingmusicok thank you... i did that just a second ago and the wps is messed up
02:57:23addingmusicill just restore it and do it again no problem
02:57:35pietro10I'd like to know why the folder is .rockbox and not rockbox
02:57:40addingmusicc ya later
02:57:56pietro10It makes adding themes on Mac where it extracts to its own folder instead of .rockbox very very painful
02:58:36scorchepietro10: a variety of reasons that i likely cant name them all...i will differ that to the idle masses
02:59:22pietro10OK, I'm going to try working with voices now
02:59:44pirokoscorche: defer? ;)
03:00:14scorcheoh...i typed 2 fs
03:00:26scorchei make a lot of typos >_>
03:00:26pirokoAnd an I :-D It just confused me for a second
03:00:33piroko'Tis forgivable
03:02:52 Quit addingmusic ("CGI:IRC (Ping timeout)")
03:03:11 Quit piroko ("telinit 0")
03:03:18pietro10Am I supposed to hear a popping noise?
03:04:10pietro10The voice is being very annoying
03:04:23***Saving seen data "./dancer.seen"
03:04:42 Join piroko [0] (
03:05:28pietro10The voices are not saying the right thing and they only say certain things and it's very confusing!!! :-(
03:06:53 Join AdmiralJ [0] (
03:07:58pietro10And now Disk Mode is only saying it's OK to disconnect :-( :-
03:08:05pietro10( :-( :-( :-( :-( :-(
03:08:54pietro10please help!!!
03:09:51pietro10oh - it was never connected to begin with :-)
03:09:55pietro10but now I have to change the voice
03:11:06 Join jhulst [0] (
03:16:43Nico_PJdGordon: bufgetdata officially works :)
03:17:19pietro10OK, Rockbox is going berserk on me
03:17:21Nico_PI had a few headaches with the wrapping but it's all solved
03:17:32JdGordonusing the guard buffer now?
03:17:52Nico_Pnot yet. it doesn't take the minimum required size into account
03:18:02Nico_Pbut I'll commit as soon as I finish adding a few comments
03:18:24JdGordonok, ill see about fixing that today if i dont do anything socail :p
03:19:31Nico_Pshouldn't be too hard if we assume the required data size will always be smaller than the guard buffer size
03:21:19JdGordoni tihnk the reason the guard buffer was so big is that we wont need to worry that it will be too small
03:21:50Nico_Pyes, apparently 32K is the max size for audio chunks
03:22:07JdGordonyou know what is going to be painful to do?.
03:22:16JdGordonactually get this into the core....
03:22:18Nico_Pthing is we audio chunks isn't the only thing we'll be buffering
03:22:29JdGordonit seems that we have enough done tat we can start tinking about it
03:22:43Nico_PI think we need to add a parameter to bufopen(), to say whether a file is allowed to wrap
03:22:48Nico_PI agre
03:22:54Nico_P*agree even
03:23:04JdGordonalso what data_type it is
03:23:39JdGordonI dont think the wrap is so important just yet...
03:24:51Nico_Ptrue. I don't think we can accuratley simulate metadata-related stuff in the standalone app anyway
03:25:02 Quit jhMikeS (Nick collision from services.)
03:25:08 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:25:09Nico_Pand we need to start thinking about the threading :s
03:25:10 Quit atsea- (Read error: 104 (Connection reset by peer))
03:25:17pietro10forget it
03:25:26pietro10i'm done with something that isn't ready for the blind
03:25:28JdGordon_fun_ :'(
03:25:30 Quit pietro10 ()
03:26:21JdGordonNico_P: bufgetdata() doesnt move the m->buf_idx variable?
03:26:34Nico_Pno, you need to bufseek
03:27:06Nico_Pyou don't find that logical ?
03:27:14JdGordonnot really
03:27:33Nico_Pwell you might want to get a pointer to the data without actually wanting to consider it read
03:28:01JdGordonit also means that the data will stay valid untill a seek
03:28:04*JdGordon likes it
03:28:21Nico_Pto me, when you change h->buf_idx, it means the data before it has become disposable
03:28:34Nico_Pyes, that's it
03:28:40JdGordonyeah, na, i agree not moving it is the way to go
03:29:11JdGordonjust need to fix every codec to add the extra seek now :)
03:29:35Nico_Pthe playback engine will take care of that :)
03:29:47Nico_P.. hopefully
03:30:28JdGordonawake enough to shout down an idea I just came up with to keep track of handles?
03:30:33Nico_P(codec_advance_buffer_callback probably)
03:30:53Nico_Pwe'll see :)
03:32:04JdGordonevery track will be put in in this order |<handle array>|<codec>|MoB|<audio>|, the handle array will be a struct with each of the handles used by the track, the handle to that one then is put in the playlist array for the track
03:32:25JdGordonno need for MAX_HANDLES anywhere then
03:33:14JdGordonthe first 2 would be unbuffered quickly so the array would be copied locally once the track starts
03:34:04Nico_Pthe codec is copied to a static buffer when the track is loaded, right ?
03:34:36JdGordonumm, its copied into the codec buffer when its ready to be used
03:34:46JdGordon.. so yes :p
03:35:10*Nico_P is thinking...
03:36:02Nico_Pwhat's the "playlist array for the track" ?
03:36:37Nico_Pan existing struct I don't know of ?
03:37:12JdGordonsorry, I meant the playlist struct in playlist.c
03:37:22JdGordonthe thing playback is supposed to be working off
03:38:04JdGordonplayback would request the next track from playlist, it would then request the files be buffered, then it would assing the handle to the playlist and move onto hte next track
03:38:25JdGordonwhen the track starts playing it would first grab the handle from playlist and copy the array locally
03:38:55Nico_Psounds like a good idea
03:39:07 Quit krazykit ("leaving")
03:39:37JdGordoncrap, you know what else we havnt thought about? voice...
03:39:44Nico_Pwe allocate the space for the struct in the buffer and then fill it as we add the metadata handles
03:40:05Nico_Phmm yes, I don't really know how it currently works :(
03:40:23JdGordonvoice -> too hard basket :D
03:40:38Nico_Pdoesn't it only require a "voice codec" ?
03:41:40Nico_Panother thing that might add complexity is supporting next track metadata
03:41:50JdGordonno that should be fine...
03:42:13JdGordonmetadata is read when the track is read from disk... so just grab the array handle from the playlist like the current one
03:42:14Nico_Pit might require a second static MoB array
03:43:19JdGordonumm... should playback control the mob stuff? I would think wps would grab the info from te buffer without going through playback
03:43:27JdGordonit would have to get the handle from playlist, but thst easy
03:43:57Nico_Pideally playback should have nothing to do in the process
03:44:40JdGordonwe could grab the data from any track in the buffer
03:45:22JdGordonwe will need new wps tags for %NxT where x is the number of track into the future and T is the id3 tag :)
03:45:37Nico_PIIRC, currently the WPS has access to the struct mp3entry for the current and next track
03:47:09 Join atsea- [0] (i=ariel@gateway/tor/x-01e9b3c56f3c4249)
03:47:15Nico_Pthrough audio_current_track() and audio_next_track()
03:47:53Nico_Pthe struct mp3entry will be one of the metadata items we'll need to store
03:48:10Nico_Pbut that was pretty obvious :)
03:48:12 Join webguest83 [0] (i=40e585cb@gateway/web/cgi-irc/
03:49:43JdGordonNico_P: ok, it actually looks like we could look furter into the future with the current code by modifying those 2 functions... (/me wonders why there isnt any patches for this)
03:50:05 Join krazykit [0] (
03:50:22webguest83do any of you guys know where I can buy a new LCD for my F10?
03:50:29Nico_PI think it was suggested or discussed but ruled out as useless :p
03:50:42Nico_Pwebguest83: toffe82 would be your man
03:51:00Nico_Phe is the gigabeat guy here
03:51:06JdGordonah ok
03:51:15*JdGordon cant find those functions in the swcodec code
03:51:18JdGordononly mpeg.c
03:51:20webguest83ok thanks man. do you know around what time hes gonna be on?
03:51:38Nico_PJdGordon: playback.c:590
03:51:56JdGordonah, thanks
03:52:02Nico_Pwebguest83: he lives in the USA
03:52:14Nico_Phe's usually there quite often
03:52:39Nico_Potherwise you can try markun but he'll probably tell you to ask toffe :)
03:52:47krazykithe'll probably be around within a few hours
03:53:08webguest83hehe ok ill stick around
03:53:19krazykitwebguest83, probably find one on ebay. that'd be the best bet, to buy a dead-hard drive one with a good screen and swap them
03:53:37krazykitdon't spend too much, as the f40 goes for $140 nowadays
03:54:10Nico_PJdGordon: any plans on adding MoB simulation to the standalone app ?
03:54:33Nico_Peven basic, like with text files ans bitmaps of the same name
03:54:37JdGordonnot as such, we will need to test the handle skipping in it though
03:54:40webguest83yea true... but I saw other LCDs similar to the gigabeat one alone selling for $20max
03:54:55webguest83ill look on ebay though
03:55:34Nico_PJdGordon: maybe an interactive mode would be fun... the args list is the playlist and the user can enter commands to tell the app what to do
03:55:54JdGordonwaste of time imho
03:56:19Nico_Pmaybe but how do you want to handle the skipping then ?
03:56:48Nico_PI must admit I'm not very comfortable with the playback/buffering code in main()
03:56:50JdGordonprobably the first file is mob, 2nd is audio, 2rd is mob...
03:57:14Nico_Pyeah why not
03:57:14JdGordonits a test app... doesnt need to simulate it properly
03:57:24krazykitwebguest83, the problem is the connector. i'm not familiar with opening it up, but i'd assume the connector is gigabeat-only or something
03:57:35JdGordonevery line in the main() function is a waste of time unless its a specific test
03:57:38 Quit XavierGr ()
03:57:46Nico_Pwe still need to handle as much cases as possible, to ensure the implementation is as clean as possible
03:58:21Nico_Pthe graph view isn't a waste of time :D
03:58:28JdGordonno thats cool :)
03:59:31webguest83yea at first it was just the connector and the backlight was intermittent (thanks to my dropping it)... then it just died and I saw that in the top right of my LCD it looks like some crystals are leaking :\
03:59:35*Nico_P should go to bed now
03:59:49JdGordongnite :)
03:59:52Nico_Pheopfully tomorrow will be a long hacking day :)
04:00:09Nico_Pthanks, have fun today :)
04:00:16 Quit Nico_P (Remote closed the connection)
04:04:25 Join XavierGr [0] (
04:14:43krazykitoh, yes. is there a linux tool to submit .scrobbler.log?
04:16:31 Quit hcs (Read error: 110 (Connection timed out))
04:31:09 Quit piroko ("telinit 0")
04:38:01 Quit XavierGr (Read error: 110 (Connection timed out))
04:38:05 Quit scorche (Read error: 104 (Connection reset by peer))
04:41:37 Nick BHSPitLappy_ is now known as BHSPitLappy (
04:45:31 Quit jhulst (Read error: 110 (Connection timed out))
04:45:39 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
04:46:51 Join piroko [0] (
04:48:16piroko"Undefined instruction at 00057710" while playing an nsf file :-D
04:49:56pirokoOn the same file in the same place apparently
04:51:33pirokoAnd now it magically works. Strange
04:56:33 Join DC1 [0] (
05:01:20 Join DrPrune [0] (
05:04:12 Quit DC1 ("$4E75")
05:04:28***Saving seen data "./dancer.seen"
05:19:36 Join addingmusic [0] (i=474872e4@gateway/web/cgi-irc/
05:23:33 Quit addingmusic (Client Quit)
05:32:45 Quit webguest83 ("CGI:IRC (EOF)")
05:42:34 Join WMAdraganddrop [0] (i=474872e4@gateway/web/cgi-irc/
05:44:10WMAdraganddropok i just restored my ipod and downloaded another (hopefully newest) version of rockbox and put it on my player now all i ahve to do it drag all my music over and i can play it right?
05:44:47scorcheassuming you installed the bootloader too
05:45:21WMAdraganddropyes i did install the bootloader
05:45:33WMAdraganddropwell now all i ahve to do it wait and see if it will work
05:47:25WMAdraganddropim surprised WMA got implemented i didnt even know it was still being finished, but i am happy it did
05:47:37scorcheit isnt finished yet
05:47:58WMAdraganddropyes, but being where it is now is a good step right?
05:48:08scorchei suppose
05:48:41WMAdraganddropoh... that doesnt sound good is it not working well or something?
05:49:27scorchei didnt mean it in that way
05:50:16WMAdraganddropoh... well type is hard to see emotion lol
05:51:52WMAdraganddrophmmm the wma files just come up as "no file! ?artist? ?album? what does that mean?
05:54:16krazykitit means saratoga needs to do more coding
05:54:29WMAdraganddropand none of the mp3 files even play sound...
05:54:29saratogatell me more about this file
05:54:46saratogawell that sounds like you didn't install something if mp3 doesn't play
05:54:57WMAdraganddropabout the wma? what exactly do would you like to know?
05:55:26saratogaif it doesn't work once you get mp3 playing, let me know
05:55:47WMAdraganddrop.i dont i installed everything the same way.... i just draged and dropped them this time though instead of using itunes then initailizing the database
05:56:20WMAdraganddropshouldnt that have worked?
05:57:18saratogai'm guessing you screwed up some step when installing rockbox if you can't play music
05:57:57WMAdraganddrophmmmm..... damn? How do i fix it then just uninstall from ipodpatcher? or do i ahve to do another ipod restore?
05:58:49WMAdraganddropim happy the wma coder is here to help though perfect seeing as i want to play those files :)
05:59:23scorchedid you copy over the entire .rockbox folder?
05:59:52saratogaprobably just restart the install process from scratch
06:00:02WMAdraganddropyes just extract all to the ipod root which is drvie F on my pc
06:00:31saratogayou downloaded a normal build from
06:00:56WMAdraganddropthen i used ipodpatcher to install it, yeah i went to current builds and selected 30gb ipod video
06:01:04WMAdraganddropthat has the wma on it right?
06:01:30saratogayeah they all do now
06:02:00saratoganot that it matters if you're trying to decode mp3
06:02:10WMAdraganddropok well should i have let the ipod restart after i installed rockbox before i put anyhting else on it like fonts and WPS?
06:02:45WMAdraganddropwell yeah but i have wma in there too i just assumed the mp3's would work... bad assumption
06:03:08saratoga<WMAdraganddrop> and none of the mp3 files even play sound...
06:03:11krazykitthey do work. something's screwed up on your end.
06:03:12saratogatrue or false?
06:03:46WMAdraganddropno no sound the time doesnt even move it just displays the info acts as if all is well but then nothing happens
06:03:59WMAdraganddropso true sorry
06:04:21saratogaok then retry the install process and see if you can get an mp3 to work
06:04:43WMAdraganddropi know they work i just meant they would play and be fine with no fuss, why do people here get so jumpy?
06:05:35WMAdraganddropok ill just restore the ipod and redo it.... just to check everything just gets extracted to root right? and if it says "ok to overwrite" i press "yes to all?"
06:06:01WMAdraganddropok be back in a few to let you know
06:07:53WMAdraganddropwhen i drag the files over i can just make a folder called anything in the root of the player and tehy should play yes?
06:09:08krazykitshould, yes.
06:09:43WMAdraganddrop...hmm ok we shall see
06:11:59WMAdraganddropinstall rockbox then let it restart before addinf fonts and everything or just do it all in one go?
06:12:19saratogadoesn't matter
06:13:01WMAdraganddropim a little hesitant this time around, i dont know why so sorry for all the beginners questions
06:17:36AdmiralJhmm, one simple question: Is nclix still in progress or given up? I haven't seen it updated in a while and the links to the downloads are also broken
06:17:51WMAdraganddropok its working now but wma's are still not
06:18:07WMAdraganddropoh wait..... not you admiralJ
06:18:19 Join xNibbler [0] (
06:18:20AdmiralJoh its ok hehe
06:18:21WMAdraganddropi dont know about your question
06:18:35pirokoWmav1 at 64kbs is not working for me either
06:18:42AdmiralJI hope someone does though, meh
06:18:54scorcheAdmiralJ: what does that have to do with rockbox?
06:19:12AdmiralJAs I stated, some random question I decided to throw out heh
06:19:41 Quit Nibbier (Read error: 110 (Connection timed out))
06:19:46saratogapiroko: can you send me that 64k file?
06:19:50scorchewell, off-topicness is against teh guidelines
06:20:00AdmiralJeh, alrighty then
06:20:30pirokosaratoga: Sure. I'm on dialup though :-D
06:20:56saratogacan you email it?
06:21:01WMAdraganddropwhen i select the wma it just starts my mp3 over....
06:21:16pirokoOh sure. No dcc?
06:21:23saratogaWMAdraganddrop: how did you encode the file?
06:21:32saratogapiroko: I don't think the webclient can't do it
06:21:39pirokoAh right
06:21:40saratogacan do it
06:22:19saratogamgiacomelli is my gmail username
06:22:31pirokosaratoga: Alright :)
06:22:34WMAdraganddropwell, ill be honest with you i downloaded it from URGE (bought it mind you for the price) then used free4wma to strip the drm.... not gonna work huh?
06:22:53saratogaV1 is trivially different then V2, so if theres a problem with those it should be fixable pretty easily
06:22:59Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
06:23:06 Quit lazka_ (Remote closed the connection)
06:23:36saratogaWMAdraganddrop: i don't know what URGE or free4wma are, but assuming its a normal WMA file between 32 and 320k it should work
06:23:37Topic"Please read before speaking: | Always remember: Rockbox = BOS" by scorche (n=scorche@rockbox/administrator/scorche)
06:23:41Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
06:24:09*scorche coughs
06:24:29BHSPitLappyBoring, Optional Software
06:24:44BHSPitLappyBy Our Selves
06:24:50*scorche sighs
06:24:52WMAdraganddropwell URGE is a online music stro like napster and free4wma is like jhymn, it strips the drm with no audio quality loss, but other than that it is a wma at 192kbs....
06:24:59BHSPitLappyBanned Outside Sweden?
06:25:15pirokosaratoga: Sending now :)
06:25:40scorcheWMAdraganddrop: talk about the DRM issue in a PM with saratoga if you must
06:25:47scorchebut not here
06:26:19WMAdraganddropsorry i was trying to get away from it..... just not very well
06:26:20saratogaWMAdraganddrop: if you want, email me the file
06:26:30saratogano sense discussing how you got it anyway
06:26:45WMAdraganddropok where do i find your email address?
06:27:17 Quit borisyeltsin (Remote closed the connection)
06:27:18saratogamgiacomelli is my gmail
06:27:28saratogathought you saw it a minute ago
06:27:31 Join borisyeltsin [0] (
06:30:07WMAdraganddropok im sending it, but it is taking forever
06:30:21DogBoyso what does BOS mean
06:30:33scorchedont worry about it =P
06:30:53DogBoyI wasn't
06:31:36WMAdraganddropok it was sent it has this IRC name in the subject
06:33:28WMAdraganddropsure... will you let me know when you have checked it out?
06:33:55saratogayeah sure, no promise i'll have a fix for it any time soon though
06:34:35 Join ptw419 [0] (
06:35:07WMAdraganddropyeah thats fine im just curious
06:36:00pirokoscorche: You realize you come off as an ass a lot, right?
06:36:12*scorche shrugs
06:36:42WMAdraganddropscorche is the man what are you talking about?
06:37:07WMAdraganddropno im serious no laughs lol
06:37:16WMAdraganddropno he is the man
06:39:43pirokoMaybe I'm just over-critical
06:39:47WMAdraganddropsaratoga: will you be able to update me tonight or should i just go on lol?
06:40:21WMAdraganddropyou may be over-critical or he really is an ass, its anyones call really
06:40:48pirokoOr it's neither and he just *seems* like an ass to me ;)
06:40:53saratogaWMAdraganddrop: it works on my pc, trying it on rockbox now
06:41:02saratogahave to compile it first though
06:41:31WMAdraganddrop... what does compile it mean?
06:41:47*scorche whistles
06:42:10WMAdraganddropwhat does a whistle mean?
06:43:04pirokoWas the compile question serious?
06:44:49WMAdraganddrop...sadly as i am not really as knowledgable as i should be...
06:45:13WMAdraganddropim sorry lol.... or dont laugh casue this is not funny just sad
06:45:26pirokoNo no. You're fine :) Just a sec
06:46:02scorcheWMAdraganddrop: wikipedia is a great resource to find out about stuff
06:46:32pirokoGoing there now, actually ;)
06:46:36saratogaWMAdraganddrop: that file works fine . . .
06:47:27WMAdraganddrop..... really then what is wrong with me? lol
06:47:39 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
06:48:13WMAdraganddropwhat exactly did you do just drag and drop?
06:48:14saratogadon't feel bad, so far not a single user has managed to send me a file thats actually broken
06:48:33WMAdraganddroppiroko: thank you i understand now
06:48:51pirokoWMAdraganddrop: No problemo. Everyone's gotta start somewhere
06:49:10pirokoMy file works fine on my computer. It just doesn't run in rockbox. It sees it as a 64kbps file but just sits there
06:49:23WMAdraganddrophmmm well ill try to drag it again......
06:49:28pirokoAnd my email died at 90%........
06:49:36saratogadid you email me that 64k file?
06:49:53pirokoI'm gonna upload it to my server
06:50:10saratogaok then just post that link here or if I'm not around email it to me
06:51:02WMAdraganddropit doesnt work at all for me....
06:51:14pirokoWill do. The link will be when it finishes
06:51:28 Quit rotator ()
06:51:53WMAdraganddropdid i not install the right version of rockbox? can you provide a link to the page with the right one on it?
06:52:50WMAdraganddropang well thats where i just the one i used
06:53:08WMAdraganddrop....dang thats where i got the one i am using
06:53:20saratogawhat device are you using
06:53:41 Join davina [0] (
06:53:50WMAdraganddropipod 30gb video
06:54:25WMAdraganddropi feel stupid, kinda
06:55:20WMAdraganddropim gonna reinstall i guess
06:56:01 Join toffe82 [0] (n=chatzill@
06:56:58 Join The-Compiler [0] (
07:00:32 Join jhulst [0] (
07:02:39WMAdraganddrophaha it was the wrong version of rock box saratoga genius work on the wma schorce thank you so much for the link, proko thank you for info on the compiler im now leaving a happy rockboxer
07:04:24pirokoHaha. Cheers :)
07:04:29***Saving seen data "./dancer.seen"
07:05:01pirokoInteresting. After failing to play my wma, rockbox won't play anything else either until a reboot
07:05:02WMAdraganddropno no my friend.... Cheers all around! :)
07:05:09 Quit z35 ("Leaving")
07:08:43 Quit WMAdraganddrop ("CGI:IRC (EOF)")
07:18:33pirokosaratoga: Transfer is FINALLY complete!
07:23:35 Join piroko_ [0] (
07:23:44 Quit piroko ("telinit 0")
07:23:47 Nick piroko_ is now known as piroko (
07:25:00 Quit AdmiralJ ("Quitting")
07:25:51 Quit piroko (Client Quit)
07:27:26 Join piroko [0] (
07:29:49 Quit saratoga ("CGI:IRC (Ping timeout)")
07:33:28 Join saratoga [0] (i=41becbf6@gateway/web/cgi-irc/
07:33:54saratogapiroko: theres definately a problem
07:34:06pirokosaratoga: Yay! (I think)
07:35:50 Join eli [0] (
07:36:02pirokosaratoga: What seems to be the problem?
07:37:55elianyone using a nano?
07:38:53eliand with the current build does the backlight come on correctly?
07:41:38saratogapiroko: theres something different about that file
07:41:53saratogait causes everything to go wrong in the decoder
07:42:00saratogano idea at the moment what it is
07:42:24pirokosaratoga: Lol. I used ffmpeg
07:42:33saratogayou made it with ffmpeg?
07:42:59saratogacould you give me the settings you used to encode it?
07:43:21pirokoffmpeg -acodec wmav1 -i ../rip/finished/Coheed\ And\ Cambria/In\ Keeping\ Secrets\ Of\ Silent\ Earth/Coheed\ And\ Cambria-\ IKSOSE\ \[10-23\]\ A\ Favor\ House\ Atlantic.ogg test.wma
07:43:27pirokoThat was the exact line
07:44:14saratogaany reason you chose wmav1?
07:44:28pirokosaratoga: Because wmav2 didn't work. Lol
07:44:29DogBoy heh
07:44:59DogBoywhy wma in the first place
07:45:23pirokoTesting the new codec
07:45:28pirokoI'm an ogg guy myself
07:46:06saratogapiroko: I found a problem in the codec
07:46:11*Mouser_X prefers SPCs, ADXs, and such.
07:46:18saratoganot sure if its the only one, but its certainly going to prevent correct decoding
07:46:33Mouser_X(None of which are a popular format... SPC can't even be "made")
07:46:41pirokoSPC is a great format!
07:47:07Mouser_XI'm glad Rockbox supports it (thanks to HCS and Blargg).
07:47:11pirokoBrb, switching to aterm over xterm :P
07:47:18 Quit piroko ("leaving")
07:48:01Mouser_XSNES sound files.
07:48:15Mouser_XFrom Super Nintendo ROMs.
07:48:49Mouser_XIf Rockbox supported RARs (an archiving format, if you didn't know), I could fit 20,000 SPCs into 100 MB.
07:49:11Mouser_XOutside of a RAR, it'd be a few GB.
07:49:23hcsinside of a RAR, it's too dark to read
07:49:25 Join piroko [0] (
07:50:19pirokoWell crap. That didn't work
07:50:26pirokoStill in xterm :P
07:50:37 Quit piroko (Client Quit)
07:51:15 Join Wiwie [0] (
07:51:23 Quit Wiwie (Read error: 104 (Connection reset by peer))
07:51:34 Quit saratoga ("CGI:IRC (Ping timeout)")
07:51:42 Join piroko [0] (
07:52:08pirokoHoorah. Problem solved
08:00:59 Join Wiwie [0] (
08:04:26 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:08:12 Join bdgraue [0] (
08:15:13elihow do i remove the record option from the main menu of my nano?
08:15:23pirokoThis is amazing. Running rockbox for two hours continuously playing oggs and no lockups!
08:15:38 Join GodEater [0] (
08:16:07 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
08:19:21 Quit eli ("Ex-Chat")
08:22:14 Join Thundercloud [0] (
08:22:30 Quit piroko ("Zzzzzz......")
08:25:33 Join kaaloo [0] (
08:25:43 Part kaaloo
08:41:46GodEaterre Neuros N3 irc meeting : "may1937: they think they can get rockbox running on anything rather easily" <−− did we actually say that anywhere?
08:43:01 Join miepchen^schlaf [0] (
08:44:34 Join kretender [0] (
08:44:42LinusNGodEater: ehhe, not really
08:45:10scorchethere were a few comments that made me wish i was there for a few corrections and suggestions..
08:46:52GodEaterI think the closest we got was "someone will get rockbox running on it eventually"
08:47:14GodEaterand even then linuxstb caveated it with "but it'll likely suck if it's on a platform for which there are no docs"
08:47:26GodEaterI'm paraphrasing :)
08:47:27LinusNso true
08:47:38scorchethere were quite a few other comments too
08:47:56GodEaterI couldn't work out from reading it if they actually made any decisions either
08:48:16amiconnDoes the G5.5 have a different (higher capacity) battery than the "ordinary G5 ?
08:48:21GodEaterit mostly seemed to boil down to "woo yeah, it would be awesome if it had feature X"
08:48:59LinusNwe had this same discussion with joe born years ago
08:49:20LinusNeven then we told him that dm320 is a no-go
08:49:21GodEateramiconn: not to my knowledge. Interestingly enough I found out the hard way this morning that the USB ports on my monitor don't recharge the ipod battery very well. Saw a message I've not seen before on it when I finally got it into a fully charged USB port this morning.
08:49:45GodEaterI even took a picture ;)
08:50:05amiconnMonitor USB hubs are almost always bus powered, meaning they only provide 100mA downstream
08:50:24 Part toffe82
08:50:25 Join BK1 [0] (
08:50:30GodEateramiconn: which is a shame - considering all that lovely current going into the monitor that doesn't come from the computer.
08:51:13GodEaterI wonder if there's a usb gadget out there that you can plug in and get a mA rating from
08:51:15amiconnYe, but the USB hubs often sit in parts which aren't powred at all.
08:51:16 Quit maffe (Remote closed the connection)
08:51:54 Join ender` [0] (
08:52:11amiconnMy monitor also has an USB hub in the base, but the base of a flatscreen is detachable, so no power there
08:52:13scorcheGodEater: multimeter, but not USB gadget =P
08:52:17GodEaterslightly OT, but the PS3's USB ports *suck* for providing power.
08:52:42GodEaterscorche: I think I have one of those - but I got a bit carried away when I bought it, and went for one with LOTS of functions.
08:52:50GodEaterUp shot of which is I have no idea how to use it ;)
08:52:51scorcheGodEater: fluke?
08:52:53amiconnGodEater: We could display that in pp rockbox if we get an usb stack running
08:52:54 Nick BK1 is now known as BK (
08:53:16GodEateramiconn: what a great idea!
08:53:29GodEaterlets add that to austriancoder's todo list :)
08:53:38amiconnThe iAudio targets could also display it (a pin signalling 100/500 mA USB power from the usb-ata bridge is connected to a gpio)
08:53:46scorcheGodEater: as in the it a fluke?
08:53:54GodEaterscorche: oh - no idea
08:54:00scorcheoh...flukes are nice..
08:54:01GodEaterhaven't had it out of the cupboard for ages
08:54:22GodEaterit's bright yellow, has a huge LCD display on it, and a dial with about a thousand settings on it
08:54:33 Join pondlife [0] (
08:54:34amiconnHmm. According to there is no difference
08:54:39GodEaterthat's not very helpful is it ?
08:54:41scorchecould be, but there are a few copycats
08:54:57amiconnMeaning that my pp clock setup changes might have had a positive effect on runtime
08:54:58GodEaterI'll see if I can find it on the website of the place I bought it from
08:55:28GodEaterI got it originally to help me chip my Xbox, but then I found out I need to practise my soldering skills a lot before I got down and dirty with something so complex
08:56:25GodEaterall the soldering guides I got said "have the soldering iron one side of the pin, and the solder the other. But don't leave the soldering iron on there too long or you'll lift a trace!"
08:56:34GodEaterand I got scared
08:57:24scorcheGodEater: come to next devcon, and i am sure there are a lot of people willing to teach you to solder =P
08:57:41*scorche has a few easy projects to start off with
08:57:44*pondlife is annoyed by some recent stupid comments on Flyspray and might start deleting them
08:57:51GodEaterscorche: I have a guy in the office here who's an ex television engineer - I could probably get him to do it
08:57:57scorchepondlife: link so we can all enjoy them!
08:58:11*pondlife doesn't want to name names
08:58:29pondlifeBut if you look at the last 2 days ML you should spot them
08:58:36pondlifei.e. the Flyspray feed
09:00:32 Quit Febs (Read error: 110 (Connection timed out))
09:01:07GodEaterscorche: no - it's not a fluke - I just checked - I didn't spend *that* much on it :)
09:01:21 Quit sheenzy ()
09:02:31GodEaternice to see DVD Jon has managed to unlock and iPhone's music playing abilities without you having to sign your life away on a two year phone contract
09:03:18scorchethey got around the activate service screen?
09:04:04GodEateryou have no phone service (seems pretty obvious if you have no carrier) - but all the other features work
09:04:31***Saving seen data "./dancer.seen"
09:04:42scorchehaha...some of: is just rediculous
09:05:50GodEaterI just wish DVD-Jon hadn't gone over to the dark side (.NET) with all his latest toys
09:06:07GodEaterscorche: heh - yeah I read that yesterday
09:07:07 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:07:34*hcs takes this opportunity to plug #rockbox-community
09:07:52 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
09:07:53GodEateriphone hacking isn't necessarily off topic ;)
09:08:15 Join miepchen^schlaf [0] (
09:08:18 Join petur [0] (n=petur@rockbox/developer/petur)
09:08:51 Quit BK-- (Read error: 110 (Connection timed out))
09:09:08 Quit Thundercloud (Remote closed the connection)
09:09:59 Join DC1 [0] (
09:10:33*scorche shrugs
09:11:01scorcheapparently i am an ass now
09:11:35*scorche silences petur with some beer
09:11:51*petur receives two highlights in one line
09:11:59 Join webguest03 [0] (i=c023110b@gateway/web/cgi-irc/
09:13:27 Quit DC1 (Client Quit)
09:21:47linuxstbsaratoga: (If you read the logs) It seems ffmpeg encodes WMA files with multiple payloads per packet. All other WMA files I've seen just have a single payload (the entire superframe) per packet, but ffmpeg is splitting up the superframe into lots of small payloads. So the problem is with my asf parser, which assumes one payload per packet.
09:22:32*GodEater observes the conversation go sailing waaaaay over his head
09:22:57linuxstbThat's probably a good thing...
09:31:50 Join xei [0] (n=x@
09:31:54xeianyone up?
09:32:20*scorche pulls the covers back over his head
09:40:14 Quit The-Compiler (Read error: 104 (Connection reset by peer))
09:42:18 Quit jhulst (Read error: 104 (Connection reset by peer))
09:47:39 Join Wiwie [0] (
09:50:13 Join Siltaar [0] (
09:57:41peturI don't understand why some builds show these binsize fluctuations, buildlog shows the same compiler is used
09:58:30peturexcept for that of the gigabeat, where I see 4.0.3 and 4.0.4
09:59:12petur( = 4.0.4 and = 4.0.3)
10:03:37scorchea bit of discussion about the neuros player and other designs from the "audiophile community"
10:03:41 Join kaaloo [0] (
10:03:56xeirockbox rocks!
10:04:06 Part kaaloo
10:04:10markunxei: it does :)
10:04:18markunwhich player do you have?
10:06:41 Join spiorf [0] (n=spiorf@
10:08:00peturscorche: they also talk about the ISP1362 as USB (host) chip... better start working on that usbotg stack again :/
10:08:31scorchepetur: aye...i was wondering if you were going to catch that =P
10:08:39scorchealthough, these people would likely never build one
10:09:28peturAs it seems to be more of a home-player solution, I would recommend them gettting a SqueezeBox from SlimDevices.. Does need a server (PC)
10:09:48scorchei believe they would liek something transportable
10:10:18scorche(as would i, as i use my amp all the time as well)
10:12:17peturI only use the SqueezeBox at home, the sound is way superior than my h380. I really suffered when my SB was away for repair...
10:12:51 Quit Wiwie (Read error: 113 (No route to host))
10:15:07scorchepetur: what do you use as headphones?
10:15:16 Join obo [0] (
10:15:37peturbah... They call themselves "audiophile" and discuss using a PDA or N770 tablet as streaming device..
10:15:55 Quit bdgraue (Read error: 104 (Connection reset by peer))
10:16:02 Join bdgraue [0] (
10:16:11scorchepetur: there are wide ranges of people there
10:16:24scorcheread Patrick82's posts for some entertainment
10:16:52peturI use the ones that came with my h340, and at work I have a wireless one (welltech 40699) that sounds amazingly well (!)
10:16:52scorchehe is completely serious as well...
10:17:26*petur welcomes obo and wonders if he read his mails...
10:18:06*obo fires up his webmail
10:19:45*obo falls off his chair in surprise :)
10:19:59 Join printfXh4 [0] (
10:20:27*pondlife helps obo get up again
10:22:49 Quit xei (Read error: 113 (No route to host))
10:23:05peturscorche: I once had a projectleader that was just like that Patrick. He modified a room in his house for perfect accoustics and bought really expensive cables.... not to mention a CD burner specially made for audio because of more stable clocking...
10:23:18scorchepetur: he goes a bit far...
10:23:40petursome people just have way too much money to spend :)
10:25:20scorchecovers all of his equipment inside and out with ERS paper, claims that FLAC is brighter than WAV because FLAC uses 3% extra CPU power which is what makes the difference, using a round rock by his equipment makes a "HUGE" difference as a dampening stone....i could go on
10:25:42scorchehe is excessive to the point of insanity....and if you ahve seen his youtube videos..
10:26:01scorcheFebs is a regular voice of reason in his threads =P
10:27:32 Nick daurnimator is now known as daurn (i=daurn@unaffiliated/daurnimator)
10:32:26 Join desowin [0] (
10:46:58 Quit miepchen^schlaf ("Verlassend")
10:50:28 Quit Farp (Read error: 110 (Connection timed out))
10:51:55 Join norbusan [0] (
10:52:05 Part norbusan
10:54:03 Join homielowe [0] (n=chatzill@
10:56:40homieloweI put "WMA" as a codec on the wikipedia page with a reference that is a work in progress etc. Is this fine? or should I remove it entirely to stop the flooding and eventual clogging of the IRC channel of the uneducated(n00bs)
10:57:30scorchewe can take care of them =)
10:58:05markunand it's good to have some beta testers
10:58:16homieloweWell if anyone would like to clean it up,(the page that is) feel free
10:59:43homieloweI tested some random trance of the 192 kbps variety and it works very well (iPod 5G) for its stage of progress
11:00:38Zagorpetur: any luck with the fat dir move problem yet?
11:01:19peturnope... I tried changing the timestamps but I didn't see them change in dos, so I must be doing something wrong
11:01:31peturI did manage to get crosslinked files though ;)
11:01:48peturso I was probably changing the wrong entries
11:01:49 Quit crashd_ ("leaving")
11:01:55 Join crashd_ [0] (
11:02:00 Join bdgraue_ [0] (
11:02:03 Quit bdgraue (Read error: 104 (Connection reset by peer))
11:02:12*homielowe aus
11:02:21 Part homielowe
11:04:24 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
11:04:33***Saving seen data "./dancer.seen"
11:04:49peturZagor: I used update_short_entry on the . and .. entries right after the move, I would have guessed that would give correct timestamps.
11:05:50peturI forgot to update the patch as somebody bothered me to look at the properties plugin... *ducks*
11:06:21Zagorok, no worries
11:06:42peturany more ideas?
11:07:16ZagorI'm afraid I still haven't gotten time to take a good look at it yet, so no. :-(
11:07:46peturDo you know of a good tool to browse the dir entries lowlevel? (win or linux)
11:08:36amiconnpetur: winhex
11:08:48Zagorno. I only used my own fat test code when I wrote the driver
11:08:50amiconnIt's shareware though, the unregistered version doesn't let you edit
11:09:27peturanyway, the fault isn't really data corruption so the bug isn't overly critical
11:10:20peturwell winhex is really lowlevel or does it have templates to look in a more structured way?
11:11:21peturoh nevermind, I see we have it here at work, complete with the needed templates :)
11:13:38amiconnIt ships with templates for mbr, dir entry etc
11:14:36peturthat's nice... I'll investigate that way tonight...
11:18:22 Join Entasis [0] (
11:21:52 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:22:28 Quit kretender (Read error: 104 (Connection reset by peer))
11:23:35 Quit linuxstb_ (Client Quit)
11:30:50 Join kretender [0] (
11:33:31 Quit scorche (" HydraIRC -> <- Would you like to know more?")
11:35:38 Quit linuxstb (Read error: 110 (Connection timed out))
11:42:31 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
11:48:15 Quit kretender (Read error: 104 (Connection reset by peer))
11:59:28 Join Lear [0] (i=chatzill@rockbox/developer/lear)
12:09:54 Quit Siltaar (Remote closed the connection)
12:18:05 Join XavierGr [0] (
12:24:23 Quit DrPrune (Read error: 104 (Connection reset by peer))
12:33:02 Quit merbanan (Read error: 110 (Connection timed out))
12:34:16 Part Esp_
12:41:12 Join Shaid [0] (
12:43:01GodEateris everyone on holiday this week ?
12:43:24scorchedidnt you get the memo?
12:43:34GodEaterapparently not :(
12:45:11 Quit atsea- (Remote closed the connection)
12:46:36 Join maffe [0] (
12:46:53ShaidWhat does BOS mean?
12:47:00Shaid*points at topic*
12:47:12Shaidan acronym search for it found about 30 matches
12:48:26GodEater"Bag Of Shit"
12:48:42GodEaterat a guess
12:49:06GodEaterit may be necessary to capitalise it to be truly accurate it
12:49:17ShaidBalikpapan Orangutan Society?
12:49:30GodEaterless likely I feel
12:49:39 Join atsea- [0] (i=ariel@gateway/tor/x-a1288b509237bf9d)
12:49:43scorcheShaid: it is a joke that seems to have been reviatalized
12:49:47scorchewow, i cant type
12:49:52GodEater <−−see here Shaid
12:49:59scorchenumber 1
12:50:11Shaidduly noted.
12:50:47GodEaterwho decided to make it the topic ? ;)
12:50:55*scorche whistles
12:51:03*GodEater scrolled up to see
12:51:35midkayi thought he said rockbox sounded like a bag of shit. not that it _was_ a bag of shit.
12:51:58scorchemidkay: indeed...but it has been adopted
12:52:15Shaidthis gives me an idea for a wps...
12:52:20midkaywell, it's confusing and nonsensical and _wrong_, but whatever floats your boat.
12:52:21GodEateroh good lord no
12:52:42GodEaterI've always wondered what the bag of shit was supposed to be doing that made it make any sort of noise at all
12:53:19midkay12:51:17, midkay, haha. bag of shit. that's hilarious.. how can something sound like a bag of shit? bags of shit don't really make noise or sound like ...
12:53:58GodEaterI see I am not alone in pondering this midkay ;)
12:54:13midkayhaha. apparently not. i guess i pondered it too some months ago. although i forgot that i did. :)
12:54:24 Join guest84 [0] (i=864c034f@gateway/web/cgi-irc/
12:54:28Shaidthey go "squoosh", I guess.
12:54:45GodEaterif you throw one at something I guess so
12:55:56GodEaterhahaha - hadn't seen that quote from presence before
12:56:04scorchehaha...just going throught hat google search is amusing
12:56:31scorcheGodEater: i added it that sat night after he left =P
12:57:09scorche*Cassandra still wants a "bag of shit" plush toy with the Rockbox logo on it, like we discussed at Devcon.
12:57:12GodEaterdid his ipod glow in the dark after it's contact ?
12:58:20 Join merbanan [0] (n=banan@
12:59:14GodEaterwhat is Linus's "It's not entirely useless" comment in reference to ?
12:59:54 Join qwm [0] (
13:00:07qwmbattery time on ipod targets is like a big bag of shit.
13:00:17qwmfix it scorche, fix it now!
13:00:26*scorche salutes qwm
13:00:33*qwm hugs scorche
13:00:40qwmhow're you doing, you wanker
13:00:43scorcheqwm: GodEater is the one you need to yell at for that to be fixed
13:01:06*qwm yells at GodEater
13:01:20amiconnqwm: Is it still?
13:01:31*amiconn running 2 runtime tests on ipod atm
13:01:43*amiconn expects 13-ish hours from the G5.5
13:01:43scorcheamiconn: how are they coming?
13:01:48 Quit Anakin (Read error: 110 (Connection timed out))
13:02:16*GodEater hoards his secret "battery_life_extension_on_pp_targets.diff" patch with evil glee
13:02:39qwmamiconn: i might just be biased, but it still "feels" shorter. haven't timed or anything.
13:04:34***Saving seen data "./dancer.seen"
13:06:17 Join kretende1 [0] (
13:07:23 Quit xNibbler (Read error: 104 (Connection reset by peer))
13:07:58scorcheqwm: g'night sir....dont misbehave while i am asleep
13:08:27qwmdon't worry mate
13:08:33 Join Nibbier [0] (
13:08:46preglowamiconn: anything happened lately you'd expect to change the runtime?
13:09:03 Join Crash91 [0] (n=crash@
13:14:16peturre BOS −−>
13:15:33amiconnpreglow: At least *suspect*
13:17:30preglowamiconn: what would that be?
13:22:00 Join spiorf_ [0] (
13:22:01markunnice skin/wps combo:
13:22:54 Quit bdgraue_ (Remote closed the connection)
13:22:57 Quit Lear ("ChatZilla [Firefox]")
13:24:10GodEatermarkun: indeed ;)
13:24:29pondlifeamiconn: Enquiring minds want to know of your suspicions...
13:28:22GodEateramiconn: have you actually changed anything to affect battery life ?
13:30:39*pondlife would guess there is a reason for running *2* runtime tests...
13:31:10*GodEater assumes so too - but amiconn appears to have been kidnapped by aliens now
13:31:22pondlifeNah, he's just teasing us... or eating.
13:33:11*pixelma remembers amiconn talking about an idea to save battery yesterday, must be in the logs...
13:33:46 Quit borisyeltsin (Remote closed the connection)
13:33:54*petur wonders how austriancoder is doing...
13:34:03 Join borisyeltsin [0] (
13:34:38 Join bdgraue [0] (
13:39:38 Quit spiorf (Read error: 110 (Connection timed out))
13:40:34 Join thegeek [0] (
13:46:17 Quit atsea- (Read error: 104 (Connection reset by peer))
13:46:34peturoh crap... some spammer is using my email address as originator and my mailbox is getting flooded with bounces and spam alerts. Why the hell would a spamfilter send a mail back to the originator?
13:46:59LinusNthat is so lame
13:47:16GodEaterpetur: had that happen to me about 5 years ago
13:47:20GodEaterI had to abandon the account
13:47:21preglowah, people and the manner in which they are stupid
13:47:33LinusNjust as lame as the antivirus softwares that bounce virus mails back to the sender address
13:47:39preglowi'd just do a procmail filter or something
13:48:06 Quit austriancoder (Remote closed the connection)
13:48:35Lynx_is there a way to charge a h300 via usb if the battery is really flat? i forgot my adapter...
13:50:31peturhold REC while inserting the cable (and make sure USB charging is enabled in the settings)
13:51:25 Quit printfXh4 (Read error: 113 (No route to host))
13:51:34Lynx_hmm, battery is too flat i guess, i just reboots all the time when inserting usb
13:51:34peturhmmm that's in rockbox which you probably don't manage...
13:51:58Lynx_what does holding rec do? not go into iriver firmware?
13:52:25peturinside rockbox, it prevents entering USB mode (and thus spinning the disk)
13:52:26 Quit bdgraue (Remote closed the connection)
13:52:50Lynx_but the battery should never be that flat anyway i guess, it played for like 15 minutes with charge of 0% displayed yesterday. i guess it should have shut down, but didn't.
13:53:06peturat boot, it forces original firmware, maybe iriver can charge?
13:53:25Lynx_nah, iriver doesn't get to boot either
13:53:36Lynx_ah well, i'll have to find the charger at home then
13:53:52amiconnpondlife: One on G5.5 and one on mini G2
13:54:05dionoeadid you try leaving it plugged long enough ? (i sometimes have to wait 5 minutes before my iPod boots correctly when charging on USB)
13:54:17amiconnAnd I did indeed change something that I suspect to have an influence on battery runime
13:54:39 Join kaaloo [0] (
13:54:49 Part kaaloo
13:55:09Lynx_dionoea: i doubt it even starts charging, it just reboots after 1 second or so
13:56:02dionoeaah, the iPod firmware must have a special "charge if you can't spin the disk up" mode
13:57:30 Join bdgraue [0] (
13:58:01peturLinus should work a bit on that new H300 bootloader, it has charging incorporated... But it crashed on my h380
13:58:47 Join atsea- [0] (i=ariel@gateway/tor/x-715f9d1f069176d9)
13:59:45 Join ackbahr [0] (
14:00:48pondlifeI'm using the SVN bootloader fine on H340
14:00:58pondlifeComplete with USB charging
14:01:26peturWell here it would only boot 50% of the time, I fear it has to do with my 80GB disk (?)
14:01:46GodEaterdionoea: I believe you're correct, I experienced that mode this morning
14:03:24pondlifepetur: Or maybe I did mine at a good time and it got broken since?
14:03:25Lynx_pondlife: how long does usb charging take compared to the charger?
14:04:02 Join Rob2222 [0] (
14:04:22peturpondlife: I installed one that LinusN build for me just after devcon (I forgot to ask him there - we should have tested it there)
14:04:23pondlifeAbout 10 times as long?
14:04:24Lynx_heh, ok
14:04:51pondlifeThat's a piece-of-string guess, nothing measured ;)
14:12:30pondlifeI only just noticed the deprecated Releases page -
14:12:46pondlifeMaybe Releases should be removed from the left-hand menu?
14:13:04 Join Nico_P [0] (
14:13:25pondlifeAlso, should there be a quick link to download and run RBUtil for a quick install?
14:13:30 Quit guest84 ("CGI:IRC (Ping timeout)")
14:15:09petura direct link to rbutil would be nice
14:15:20pondlifeThinking of the n00bs
14:16:04pondlifeThere's not even a link on the current build page... is it included in the automated builds at all?
14:16:28pondlifeIIRC a rbutil commit still triggers a build.
14:16:35peturthe builds are made by hand
14:18:17GodEatera commit probably triggers a rockbox build - but I don't think RBUtil code is actually built across our targets
14:18:20Nico_PJdGordon: hi
14:18:36Crash91Guys, is there any chance of rockbox accepting animated GIFs for backdrops in the future?
14:18:49GodEaterwe have three target OSs which makes that a bit more complex to manage too
14:18:52Crash91Im thinking animated rockbox logo
14:18:58*petur bangs head agains desk
14:19:04GodEaterCrash91: very little chance
14:19:17Crash91Thought so :p
14:19:19pondlifeI think you'll find WMA will be supported first. Oh, wait...
14:19:30 Quit bdgraue (Read error: 110 (Connection timed out))
14:19:35GodEaterpondlife: Real ?
14:21:03Crash91Whos good at graphics here?
14:21:26GodEaterI don't think many of our "artists" hang out here much
14:21:41GodEateralthough I stand to be corrected if wrong
14:21:41Crash91forums then...........?
14:21:47GodEaterforums is your best bet yes :)
14:21:56Crash91well, anyway
14:22:24 Join norbusan [0] (
14:22:25Crash91i need help with it,check it out ;)
14:22:29 Part norbusan
14:22:43Crash91just started like 45mins ago
14:22:59*GodEater wonders if scorche is still awake...
14:23:05 Quit ackbahr (Read error: 110 (Connection timed out))
14:23:12 Join ackbahr [0] (
14:24:02Jhi guys - that raises a question; is it possible to change the splash screen from within RockBox?
14:24:42Crash91isnt the splash inside the rockbox file?
14:24:50Crash91.mi4 in my case
14:25:35GodEaterindeed it is
14:25:38Crash91to do that you would have to recompile the code to take from splash.bmp
14:25:41GodEaterand therefore not it's not ;)
14:25:46Crash91in /.rockbox
14:25:59Crash91or something, then only would it be possible
14:26:01GodEaterpeople keep talking about doing that
14:26:01Jis there any reason that it has to be?
14:26:10GodEaterJ: not really
14:26:19GodEaterat least - I can't think of a good one :)
14:26:20Crash91I think devs have more on their minds
14:26:27Crash91like WMA support lol
14:26:50GodEaterthere's only two of them doing that currently :)
14:26:52Crash91GodEater: Did you visit my site?
14:26:56GodEaterI just looked yes
14:27:20GodEaterCrash91: you've seen already ?
14:27:23markunJ: there were some plans to make it possible, but after that everyone lost interest again I guess
14:27:34Crash91Who hasnt?
14:27:36GodEatermarkun: it should be a pretty trivial change surely ?
14:27:53Crash91And where did you think i got the player pics for the gallery, ;)
14:27:56 Join Febs [0] (
14:29:25*amiconn wonders why people want to make rockbox booting slower...
14:29:48GodEateramiconn: takes all sorts =/
14:29:51GodEaterlive and let live I say
14:29:54Crash91Because they want more features
14:30:13Jamiconn: would changing the splashscreen drastically alter boot time?
14:30:17amiconnI don't see the splash screen as a feature
14:30:23GodEaterCrash91: amiconn is a purist, he likes keeping his DAP player for playing music and little else :)
14:30:32amiconnFast boot is a feature
14:30:40Jamiconn: it's not, but it you can rebrand everything else...
14:31:02GodEaterJ: no-one's stopping you downloading the code and implementing your own splash - that's a feature too
14:31:10peturyou can always compile a custom bootloader with your image in...
14:31:20GodEaterdidn't think the image was in the bootloader
14:31:33Crash91But what about noobs like me who dont know how to compile?
14:31:37JGodEater: oh, I know. I'm just trying to get a feel fro the codebase and this seems quite good...
14:31:43GodEaterCrash91: learn. It's not that tough.
14:31:51GodEaterwe have enough guides on the subject
14:32:38Crash91I need gcc for that right?
14:32:45Crash91or cygwin
14:32:54GodEaterdepends on your OS
14:32:58Jor vm and the image
14:33:00GodEateryou only need cygwin if you're runnign windows
14:33:05GodEaterthen you don't want cygwin
14:33:25Crash91Well, the problem is installing gcc
14:33:28Crash91im not root
14:33:37GodEaternot your PC ?
14:33:45Crash91Dad and mine shared
14:33:53GodEatershouldn't stop you
14:34:08GodEaterthe script will build all the compilers for you anyway
14:34:12Crash91But you see....he likes to keep me as a limited user
14:34:16GodEateryou just have to install them under your /home
14:34:20Crash91ahh ok
14:34:24Crash91good enough
14:34:31GodEaterkeeps them out of his PATH :)
14:34:44 Quit desowin (Read error: 113 (No route to host))
14:34:45Jesp since you need a cross-compiler :)
14:34:46Crash91haha :)
14:35:05GodEateralthough if there's no system gcc
14:35:11GodEaterthen building the cross-compilers will be impossible
14:35:22Crash91ok, i have gcc
14:35:33Crash91[crash@rt ~]$ gcc
14:35:34Crash91gcc: no input files
14:35:34Crash91[crash@rt ~]$ gcc
14:35:34DBUGEnqueued KICK Crash91
14:35:34Crash91gcc: no input files
14:35:35Crash91[crash@rt ~]$ gcc
14:35:36***Alert Mode level 1
14:35:36Crash91gcc: no input files
14:35:40GodEaterCrash91: do you have svn ?
14:35:49Crash91i dont think so
14:35:56Crash91how do i find out?
14:36:01GodEater"which svn"
14:36:05Jsvn −−help
14:36:09GodEateror svn −−version
14:37:19Crash91ok got it
14:37:32GodEateryou do ?
14:38:07GodEatermake yourself a directory
14:38:15GodEatercd into it
14:38:35GodEaterand then do "svn co svn:// rockbox"
14:41:11Crash91how long doe sit take
14:41:24dionoeaa few seconds
14:41:29 Part LinusN
14:41:32 Join desowin [0] (
14:41:39Jdepends on your internet speed
14:41:52GodEaterCrash91: you should get masses of output while it's working
14:42:04GodEaterouch, bit slow
14:42:05 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
14:42:07dionoeaah, then maybe a bit more than a few seconds
14:44:01CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
14:44:01*GodEater shuffles off to the gym
14:45:37***Alert Mode OFF
14:45:58Crash91does anyone know how big the total download is?
14:47:15Jmy rockbox directory is 178M - but some of that won't be downloaded...
14:47:22*Crash91 wishes he had a faster connection
14:47:52 Join Quippini [0] (
14:48:26Crash91so around 100MB?
14:50:01*J is unsure, could be 1/2 that
14:51:37Crash91Ill tell you when im finished :) on my connection, 100MB = ~1hour
14:52:06Jthe tarballs of the source are only 8M so I don't know where all the extra came from!
14:52:54peturobject files, temporary files, ...
14:52:55Crash91i dont think svn is getting tarballs
14:53:08Jno, no, but that's the source
14:54:05markunCrash91: so, how much was it?
14:54:18 Join Soap [0] (n=Soap@rockbox/staff/soap)
14:54:37Crash91i would guess around 15 meg
14:54:52Crash91but lemme check
14:55:31Crash91my directory is 67.2MB, but it couldnt have downloaded that in 20 minutes
14:55:52Crash91Im guessing 20MB of uncompressed tarballs?
14:56:27petursvn creates some files of its own
14:56:41peturbut not that much I think
14:57:38amiconnsvn needs twice the space of the actual source
14:57:50amiconn(in order to allow local svn diffing)
14:58:17*petur learns something new
14:59:40Crash91So the source totals 35M?
15:00:01 Quit ackbahr (Connection timed out)
15:00:36 Join Tamsou7 [0] (n=Happy_Ne@
15:01:53J*shrug* does it matter?
15:03:51J? nvm?
15:04:35***Saving seen data "./dancer.seen"
15:04:50FOADNo, go on, explain.
15:05:42Crash91Nevermind, it doesnt really matter now, I agree with J. Happy? :)
15:06:27Crash91If you own a Sansa, take a look at this, if not, take a look anyway
15:06:53Crash91I need people who are willing to help me make backdrops
15:07:05 Quit atsea- (Remote closed the connection)
15:07:53Crash91oops, refresh page if you are there
15:08:19JI like the beach pic
15:08:32Crash91Take it then :)
15:08:47Jdon't have a Sansa :) but thanks
15:08:51peturCrash91: what about copyrights?
15:09:04Crash91I got this from:
15:09:14gregjnow I am scared to use that software :D
15:09:15 Join bdgraue [0] (
15:09:40Crash91and i am just now putting the link on
15:09:49Crash91ive also made several myself
15:09:55Crash91i will upload them later
15:10:00peturgregj: that was before we drank any beer ;)
15:10:35gregjany whiskey lovers there ? I hate beer, so in case I might join you (maybe some day) - I would bring along something else :d
15:10:47gregjthat girl's british ?
15:10:50peturI like both ;)
15:11:14Crash91nice video :)
15:11:16peturCassandra is her nick
15:12:22Crash91who funded this?
15:13:03*gregj stops donations :P
15:13:27Crash91:) you must get a lot
15:13:44gregjsandisk is also a sponsor
15:13:53gregjdoes RB turn itself off, if idle for too long ?
15:14:02 Join atsea- [0] (i=ariel@gateway/tor/x-0bebcdebedd2807c)
15:14:13Jgregj: mine does - iPod
15:14:21gregjmine too
15:14:26gregjthat's why I ask
15:14:32gregjI have ipodg
15:14:35Jgregj: it's in the config somewhere
15:14:44Crash91I dont think mine does-Sansa
15:14:51 Quit maffe (Remote closed the connection)
15:15:09Crash91What is the point of the sleep timer? Doesnt work on my Sansa
15:15:11pondlifemarkun: ping
15:15:18Crash91Does it work on other targets?
15:16:12*gregj wishes the OS wouldn't be so #ifdefed, but structured like linux - arch/* stuff for specific bits
15:17:01*Crash91 wishes that rockbox had some sort of shell program
15:17:32pondlifeWhat for..?
15:17:58gregjCrash91: write shell plugin
15:18:15gregjand proper-usefull-everywhere keyboard
15:18:17gregjgood luck
15:18:22*Crash91 doesnt know C
15:18:22markunpondlife: pong
15:18:46Crash91gregj: Teach me? :)
15:19:00pondlifemarkun: Did you see the comment on ?
15:19:30pondlifeYou did a fix way back to clear the PCM pause status when PCM is stopped...
15:19:49markunI didn't see it no
15:20:25gregjCrash91: I am not the cheapest teacher, go grab a book, and learnyourself
15:20:44pondlife(It was to fix if that helps.)
15:22:15markunpondlife: I don't have time to look into it now
15:22:18markunmaybe tonight
15:22:40pondlifeNo worries, but I'll play with it..
15:23:08pondlifeand perhaps revert... ;)
15:23:10LinusNgregj: regarding the arch/* concept, that is what the firmware/target/* tree is for
15:23:20gregjLinusN: yes, but now it is a mess
15:23:29gregjLinusN: to be honest, too many #ifdefs inside files
15:23:35LinusNyes it is, but way better than before
15:23:41markunpondlife: why not just fix it for all 3 case?
15:23:50gregjsplit it into smaller functions, put it in arch/dir
15:23:55pondlife3 cases?
15:24:14LinusNgregj: you're new here, right?
15:24:25gregjis there a way to test mpeg player on PC, I thikn it would be much easier for me to look up bugs that interest me
15:24:28gregjLinusN: yes
15:24:29markunpondlife: 7187, 7257 and 7261
15:24:35LinusNgregj: i can see that
15:24:45pondlifeI missed one!
15:25:15 Part kretende1
15:25:23gregjLinusN: :]
15:25:50gregjmpeg player plays strangly on me, I hate it
15:25:54LinusNgregj: the target/* concept is new, and we are constantly moving functions to it
15:26:15markungregj: do you know the rockbox simulator?
15:26:16LinusNbut it only handles the target differences on the lowest level
15:26:51LinusNthere are still plenty of #ifdefs in the application code, and even more in the plugins
15:26:57*pondlife dreams of the day that #ifdef SIMULATOR is banned.... :)
15:27:13LinusNpondlife: you are surely a dreamer
15:27:14peturstubs for everybody!
15:27:45pondlifeLinusN: A dream, or a nightmare...??
15:28:02*pondlife goes back to work
15:28:06LinusNboth, i guess :-)
15:29:01peturLinusN: any progress on that new h3x0 bootloader? I can always try a new version (re crash problem on my h380)
15:29:21LinusNpetur: haven't worked on it at all
15:29:32peturLinusN: because we had a guy with flat battery and usb-only here some time ago - could have used it...
15:30:15gregjLinusN: ifdef's are evil, only way to show - "hey, I have no idea how to do it in a different way, but this code here for XXX should be YYY"
15:30:45LinusNgregj: that sounds extremely naive to me
15:30:54JdGordonNico_P: hey
15:31:00gregjmaybe it would in 80's
15:31:14JdGordoni am going to be useless untill monday
15:31:18 Quit bdgraue (Read error: 104 (Connection reset by peer))
15:31:18gregjbut I have 10 years of expierence in low level programming, and we're in 00's
15:31:21*petur bets gregj wants to introoduce c++ to use inheritance
15:31:25 Join bdgraue [0] (
15:31:27JdGordongot a friend coming to stay for the weekend tomorow...
15:32:10gregjpetur: that wouldn't be such a bad idea, maybe, if devices had more ram :) because it takes a bit more to store vectors to virtual functions, hence the sice of executable is bit bigger
15:32:19LinusNgregj: when was the last time you write a program to run on ~20 totally different embedded targets, with tough memory and cpu constraints?
15:32:27Nico_PJdGordon: ok :)
15:32:40gregjLinusN: never :)
15:32:47LinusNthere's rockbox for you
15:32:56gregjLinusN: but I wrote stuff that works on few platforms, not as many
15:33:00markungregj: just join in and help us improve it
15:33:20JdGordonNico_P: ... and i have to be up at 7am tomorow, so no 3am bed time tonight :p
15:33:23LinusNgregj: most of the #ifdefs can be removed, but it has a cost involved
15:33:30*gregj needs to watch itfor a bit, ask 10000 stuppid questions, look through the code milion times, bitch even more, and than try to do something usefull
15:33:37Nico_PJdGordon: hehe... any new ideas today ?
15:33:57LinusNand that cost must never be memory or cpu load
15:34:52Nico_PI was thinking that maybe a buffering thread isn't needed
15:34:58LinusNgregj: it is good that you look at the code, it is always good with a fresh view of things
15:35:08gregjLinusN: but sometimes you have to think would 2 cycles per this function is worth the mess
15:35:10JdGordonNico_P: ?
15:35:13 Nick spiorf_ is now known as spiorf (
15:35:16LinusNgregj: exactly
15:35:27gregjLinusN: and I can see the mess everywhere
15:35:34gregjwhich probably means, - it isn't worth
15:35:55JdGordongregj: not if that function usually takes 6 cycle to finish :p
15:36:10Nico_PJdGordon: the audio thread currently takes care of the buffering... maybe it wouldn't be too hard to keep it that way, and maybe it'd even be better
15:36:20Nico_P...haven't done too much thinking about it though
15:36:29pondlifeNico_P: I'd agree that less threads is better
15:36:43JdGordonwasnt the idea to remove the buffering handling from playback?
15:36:47gregjJdGordon: than it should be an inline function, per arch - unless it is called from milion different places - than a function makes sens, and 2 wasted cycles, around 10 wasted for function in/out - is nothing
15:36:54pondlifeNot necessarily, just to organise the code better
15:37:01LinusNgregj: we are constantly trying to improve the #ifdef mess, so we would appreciate your efforts if you decide to improve it
15:37:24gregjthanks for invitation, I'll try
15:37:45gregjI just have one box, ipod, so can't test it much myself - other that 'it compiles'
15:38:05 Join maffe [0] (
15:38:16LinusNgregj: and while you are investigating the code, have a look at
15:38:33LinusNand fill in the blanks when you find them
15:38:46pondlifeThat would be very useful.
15:40:02 Join MrKeuner [0] (n=kudo@unaffiliated/mrkeuner)
15:40:24 Quit Tamsou7 ()
15:40:51gregjLinusN: another thing, is a user expierence - you guys should get 1-2 ppl involved just doing that, testing user expierence
15:41:07LinusNgregj: amen to that
15:41:18JdGordonuser experience is definatly lacking...
15:41:22gregjLinusN: might be worth asking on kde/gnome lists, that are just about it,
15:41:26JdGordonbut devs usually find that stuff boring :p
15:41:39pondlifeI don't.
15:41:42JdGordondidnt someone offer to do it then vanish?
15:41:47gregjcos I am not myself an UE expert, but surely pissed sometimes with poor UE here
15:42:03 Join bdgraue_ [0] (
15:42:03gregjJdGordon: yes, there are ppl that are doing tjust that - working for KDE for instance
15:42:13gregjthey don't know as much about devels
15:42:25gregjusualy they are girls, and some pretty ones too :)
15:42:28 Quit bdgraue (Read error: 104 (Connection reset by peer))
15:42:37LinusNthe problem is that many people confuse "user experience" with "pretty graphics"
15:42:45 Quit RaRe ("Quit msgs should be longer.")
15:42:52gregjthan grab the right ones :)
15:43:06gregjdo you guys have any relese schedule ?
15:43:11gregjwith planned features, etc ?
15:43:20LinusNthe other problem is that people get used to how rockbox works and stop complaining :-)
15:43:26LinusNgregj: no and no
15:43:28gregjreleasing something , every 6 months is cruical
15:43:32gregjto spred the word
15:43:39gregjand let ppl test
15:43:41LinusNwe are not out to spread the word
15:43:41 Quit borisyeltsin (Remote closed the connection)
15:43:42pondlifeWe release every build
15:43:58 Join borisyeltsin [0] (
15:44:12Crash91guys, how do i run without root permissions? Do i HAVE to be root or is there some way around it?
15:44:17gregjI am tryign to help, so those are your weekneses
15:44:22JdGordongoing by the download stats, we havnt had a problem spreading the word
15:44:22gregjCrash91: sudo ?
15:44:29 Join Tim_Mitchell [0] (
15:44:34JdGordon... and who the heck wants more users anyway? they just complain
15:44:35Nico_PCrash91: edit it to change the paths where it installs things
15:44:42gregjJdGordon: you have to have something that's finished, and versionX, otherwise it is plain waste
15:44:50pondlifeReleases aren't needed - we tried that but it just slowed down development.
15:44:53Crash91WARNING: this script is set to install in /usr/local but has no
15:44:53Crash91WARNING: write permission to do so! Please fix and re-run this script
15:44:56Nico_PCrash91: all the paths are defines at the top
15:45:00pondlifeThis is just for fun, after all
15:45:06pondlifeNot work at all
15:45:12gregjpondlife:yes, slows down development, but makes the quality better
15:45:13Crash91so what do i change usr/local to?
15:45:28LinusNgregj: there is one fundamental thing that needs to be explained - we don't need more users for the sake of it
15:45:34 Join The-Compiler [0] (
15:45:34markungregj: I use rockbox right now and don't think it's a plain waste
15:45:37pondlifegregj: In theory, but in practise that didn't happen. Nobody bothered.
15:45:38Nico_PCrash91: /home/crash/ ?
15:45:38gregjpondlife: and creates deadlines, sort of
15:45:41LinusNwe don't need more popularity
15:45:48pondlifeDeadlines are baad
15:45:58pondlifeThis is NOT WORK
15:46:02pondlifeit is FUN
15:46:02markunit's just nice to have some new devs from time to time
15:46:15markunand they seem to find us anyway
15:46:18pondlifeMore users = more devs, so both good
15:46:30pondlifeMore ports = more users = more devs, too
15:46:33LinusNi agree that it would be nice to have a decent official release one day
15:46:35Crash91ok dling now :)
15:46:48gregjyes, but I would love to be able to put RB on my ipod, knowing that I can go on holiday, or on a train - and it won't crash 10 times
15:46:58pondlifeIt shouldn't
15:46:58gregjand tithat this _version_ is feature complette
15:46:58Nico_Pmaybe we need a "stable" branch
15:47:06LinusNgregj: rockbox on my iaudio never crashes
15:47:14JCrash91: expect it to take _ages_ - it did on my machine and I've a 4M connection
15:47:14gregjwell, nice
15:47:15markungregj: sure, but I prefer it even more to have the latest version :)
15:47:22JdGordonheck, it hasnt crashed on my sansa for ages now also
15:47:23amiconnLinusN: Really?
15:47:34amiconnI had a crash on H1x0 just yesterday
15:47:35Nico_PCrash91, J: it's the building that takes the most time
15:47:36pondlifeIt's pretty stable - I know that it sometimes displays the info fot the wrong track, but aside from that I don't have problems.
15:47:41gregjdid you guys every tried to learn something from other OS projects ?
15:47:42LinusNamiconn: wow
15:47:48peturgregj: then buy a decent DAP, not that ipoo
15:47:49amiconnAnd that was *not* the wma codec
15:48:00JdGordonamiconn: forgot to update the plugins with the rockbox. file?
15:48:02LinusNgregj: what a funny question
15:48:13gregjLinusN: play this on mpegplayer, and see if it works
15:48:17amiconnJdGordon: I *know* how rockbox works...
15:48:27markungregj: we learn from them by using their code :)
15:48:31gregjif so - show me diff between svn trunk and your tree :)
15:48:34amiconnThis was just before shutdown, when it was supposed to save the config
15:48:41Crash91Is there something wrong with the database? i keep getting an error when i try to initialize my database
15:48:47JdGordonsure... go ahead and blame _my_ code...
15:48:58gregjmarkun: code is not the only thing, other things - organization, releases, UE , etc - is as much important to what you do
15:49:02peturgregj: then stop chatting and do something about it
15:49:33gregjI am doing, but letting you guys realize that :)
15:49:55markungregj: so, the next release will be 6 months from now?
15:50:10pondlifeCrash91: Several people have reported that the database doesn't init properly... not seen it mentioned on Flyspray though.
15:50:12LinusNgregj: show me these other projects that create an OS for DAP:s, with 100% volunteers and no funding
15:50:35markunLinusN: zune-linux!
15:50:36gregjLinusN: every single OS project started like that
15:50:45gregjLinusN: from linux, to KDE
15:51:07Crash91EVEN WINDOWS
15:51:15gregjit was because they were releasing something usefull, and had releases - that they got founding
15:51:18JdGordonosx didnt...
15:51:25pondlifeFunding != Funthing
15:51:41Tim_Mitchellpondlife: I too have had problems with the database, getting it to update properly - sometimes it updates by removing everything (sigh)
15:52:01pondlifeNot update, initialise... that's even more fundamental.
15:52:15pondlifeSadly, it works here.
15:52:21gregjLinusN: just grab a piece of paper, and write down current features that are there, draw a line - and underneeth put all desired features are should be there in 6 months
15:52:26gregjLinusN: and you have release schedule
15:52:35Crash91im using yesterday's build
15:52:38JdGordonthat was done this time last year
15:52:46gregjLinusN: simple, just branch whatever is above the line - once it is 100% coded, and just let ppl test and fix the bugs
15:52:52JdGordonwe ended up freezing development for 3 months fixing bugs
15:52:58LinusNgregj: but the question remains: do we *want* a release schedule?
15:52:59pondlifeBut nobody tests/fixes
15:53:01Nico_Pthen there was a feature freeze.... dark times of rockbox history :D
15:53:18gregjthat's why you have thing called branch
15:53:29pondlifeSo the branch withers.
15:53:30gregjevery developer can create branch of their stuff
15:53:43gregjwork on something, and later put it toughether for next release
15:53:45Nico_PI agree a branch would've been better... but we'd all have done work in the experimental branch
15:53:56pondlifeYou are free to make a branch now and work from that if you want..
15:53:57peturgregj: you're oversimplifying so much I almost wonder if you aren't a manager in disguise :p
15:54:04Nico_Pgregj: merging is a pain with svn
15:54:16JdGordonNico_P: spoken like a tru git convert :D
15:54:19gregjNico_P: yeah? try it with cvs!
15:54:24Crash91guys, how do i compile .patches?
15:54:35peturCrash91: see wiki
15:54:41Nico_PJdGordon: after trying it with git I never want to branch nin svn
15:54:49pondlifeCrash91: wiki for WorkingWithPatches
15:54:53SlasheriTim_Mitchell: you are using sansa?
15:55:00gregjpetur: I am not manager
15:55:05gregjI am developer, 100%
15:55:14amiconngregj: There is one fundamental problem that doesn't exist for other projects: rockbox runs on many different devices, and there is no single developer who has access to all of them
15:55:24Nico_Pgregj: you said you were involved in a project, but I can't remember which
15:55:27gregjyou can guys develop till the end of world, for fun. and it will work stable only on your boxes
15:55:28markunTim_Mitchell: rocks?
15:55:38amiconnSo each feature needs to be tested *and often fixed* for other targets by other devs
15:55:40Tim_Mitchellmarkun: rocks?
15:55:42SlasheriTim_Mitchell: ok, no idea about that then. but at least sansa is known to have some sort of issues
15:55:43gregjNico_P: KDE ?
15:55:51amiconnAnd that often doesn't happen :(
15:56:01pondlifegregj: So what's not stable...?
15:56:01Nico_Pgregj: right
15:56:17JdGordonSlasheri: what sort of issues? I just did an init and it worked fine
15:56:19Nico_Pgregj: you're not at aKademy ?
15:56:25peturgregj: another problem: we often have no datasheets, so if your ipod crashes a lot, go blame apple/pp
15:56:28gregjamiconn: so, instead of trying find an excuse for not doing stuff, try find excuse for doing it
15:56:31gregjamiconn: so, instead of trying find an excuse for not doing stuff, try find excuse for doi:)
15:56:40Crash91pondlife: thansk
15:56:56SlasheriJdGordon: i have seen database files from one people with byte drifts inside the file.. i doubt it could have something to do with the fs drivers
15:57:03gregjpetur: and yyou think linux doesn't have that prpblem ?
15:57:15Tim_Mitchellmarkun: for clarification, I didn't understand what you were asking by "rocks" - I'm not on IRC here all that often
15:57:15gregjagain, excuses excuses
15:57:17amiconngregj: The current problem is that many targets are in a state where a release doesn't make sense. Unstable/unfinished features etc
15:57:27markunTim_Mitchell: I guess I missed a part of your conversation and just suddenly saw "gigabeat" out of nowhere :)
15:57:29gregjNico_P: nope, I couldn't makeit to glasgow this year
15:57:40amiconnAnd I just mentioned why a single dev can't solve this problem if others aren't willing to help
15:57:42Tim_Mitchellmarkun: oh
15:57:48Nico_Ppity, it looks like they're having lots of fun :)
15:58:11Nico_Pgregj: what kind of work did/do you do in KDE ?
15:58:12peturgregj: then again, hibernate of linux on my laptop is horribly broken, so they have the same problems too
15:58:12LinusNlast time i checked, we don't have as many paid developers as linux does
15:58:14amiconnAnd if you check the # of commits during our feature freeze last summer, you'll see that it didn't work out...
15:58:40gregjamiconn: well, than write a chart, what's working, what's not, what's the status, get someone involved in updating it, release only for devices that you are able to test, ask ppl to test on others
15:58:48gregjNico_P: kopete
15:59:08gregjpetur: but imagine it would be developed like rockbox, you wouldn't even use it now
15:59:22JdGordongregj: well if we did that it would waste time when we could actually be fixing bugs/ doing other usefull stuff
15:59:26pondlifegregj: Flyspray records all known issues. You are welcome to fix some.
15:59:48LinusNgregj: "get someone involved in updating it" - why didn't we think of that?
15:59:53gregjbecause no one has a laptop that you have, and it works for them - the wouldn't care much
15:59:55Tim_Mitchellguys, do any of the flash based rockboxed players play ogg well? my gigabeat rockboxed is great but not for running
16:00:03LinusNgregj: exactly
16:00:08peturgregj: I don't think I've had RB crash more on my h380 then I've seen Linux go bad on my laptop. So good for us then I would say
16:00:18markungregj: but people *are* useing rockbox
16:00:18JdGordonTim_Mitchell: both the nano and sansa play them fine
16:00:53pondlifeThe number one point is that we develop it for ourselves and for fun.. Not for mindshare or anything much else.
16:00:56gregjmarkun: good :) but there are still bugs, or is it just me that notices 100 bugs - within last 3 days of using it only
16:01:09pondlifegregj: Make sure you report them then.
16:01:10markunTim_Mitchell: The Sansa E200 and 1st generation ipod Nano
16:01:11gregjstuff that's broken, half working, not working, mpegplayer is 80% fo that
16:01:21Tim_Mitchelljdgordon: really, I didn't think the portalplayer chip had the processing power, as the regular ipods haven't been too great with ogg from what I've read
16:01:29LinusNgregj: remember that ipod is one of the worst rockbox targets
16:01:31 Join austriancoder [0] (n=austrian@
16:01:39markungregj: I can make a release for you with mpegplayer removed, no problem :)
16:01:42JdGordonTim_Mitchell: they both decode at over 100% realtime
16:01:42pondlifempegplayer isn't core either, it's a recent plugin.
16:01:53Slasherigregj: you should try rockbox on iriver H140
16:02:06Slasheriit's from other world than the ipod
16:02:15pondlifeOR anything not PowerPortal based.
16:02:21Nico_Pa world with datasheets
16:02:22markungregj: but I don't think it's bad you bring up this topic
16:02:28LinusNpondlife: portalplayer :-)
16:02:38pondlifeoops, paging Freud
16:03:15*amiconn would like to see gregj's 100-bugs list
16:03:32Tim_Mitchelljdgordon: cool, I assume the sansa is the preferred player even though usb isn't yet supported?
16:03:32Febsgregj: how about if you prepare the chart and be the "someone" who updates it?
16:03:36peturprobably copy/paste from flyspray
16:03:59pondlifeFlyspray *is* the chart... we just need to compel people to fix stuff.
16:04:04markunamiconn: but you would probably say "I wouldn't use that feature anyway" :)
16:04:20*pondlife wouldn't use that feature anyway, for sure
16:04:21 Quit webguest03 ("CGI:IRC")
16:04:42 Join RaRe [0] (
16:04:49LinusNand i couldn't care less about mpegplayer - rockbox is an audio player firmware
16:05:06markunLinusN: but it was nice on the plane to sweden :)
16:05:11amiconnmarkun: In case you didn't notice: I already fixed fieatures I don't use myself several times
16:05:31gregjFebs: I'm a developer, this would be boring for me
16:05:37Mouser_XIndeed. I like watching videos during my lunch break at work.
16:05:40pondlifeI broke features I don't use myself several times
16:05:45gregjSlasheri: I don't have iriver, I have only ipod g
16:05:53gregjSlasheri: g4/photo, how much is that iriver ?
16:05:55Mouser_Xlol @ pondlife
16:06:03Jgregj: I get the feeling that most people on this channel are the same.
16:06:06*petur has an opinion on people who buy ipoo
16:06:09gregjI can get something that's less than 50 quid, and test it myself/send some developer to test it
16:06:19gregjpetur: keep it to yourself :P
16:06:32pondlifegregj: Start by reporting your bugs on Flyspray.
16:06:35Tim_Mitchellguys, is there a place on the forums for classifieds - I don't see it and I have a nearly new iriver H140 that I'm considering selling - figured rockboxers would be the most interested
16:06:36JdGordonTim_Mitchell: the nano is slightly thinner/smaller, has a tiny screen compared to the sansa, is hard to buy, and is apple...
16:06:38Jpetur: some have them given to them as gifts
16:06:53gregjpetur: I thought about buying something that can natively play oggs before, but rockbox is an answer for me - to continue using ipod, and have phun with oggs
16:06:59pondlifeTim_Mitchell: The user ML is the best place
16:07:02Slasherigregj: not much in any way :P
16:07:10gregjpetur: if I could only make mpegplayer more stable, and fix few smaller issues - full happines
16:07:37peturgregj: post in flyspray and attach patches if you can ;)
16:07:40 Quit JdGordon ("Konversation terminated!")
16:08:01Tim_Mitchelljdgordon: I really was just referring to status of rockbox support - I realize a first gen nano might be hard to get
16:08:05LinusNgregj: you could probably do some good with rockbox as a developer, but i don't think trolling about bad project management is the best way to start
16:08:08Tim_Mitchellpondlife - ok thanks
16:08:25austriancoderpetur: I got the storage gadget driver compiling, but it dosent work.. after some ours of debuging i found the problem... i get no valid usb packets from the controller. So i looked at e200tool ( which uses the usb controller and I did a merge of the usb stuff from e200tool in arcotg driver.. its very hackish.. polling no interrupts, but to demonstrate the problem its okay. I do the same as e200
16:08:25austriancoder/soc/merge_of_e200tool_usb_stuff_and_i_dont_get_valid_setup_packets.diff - but it doent work... maybe you can help me to find out what I do wrong?
16:09:14LinusNgregj: and comparing rockbox to <insert pc based project here> is very much an apples-to-oranges comparison
16:09:20peturaustriancoder: how 'bout tonight?
16:09:56gregjLinusN: see, that's the point. I don't troll :) I am trying to give you my opinion, as someone expierenced in project managment and develokemt
16:09:58 Quit crashd_ ("leaving")
16:10:02 Join crashd_ [0] (
16:10:09austriancoderpetur: okay.. tell me a time and I will be on irc
16:10:42gregjLinusN: some things are common between PC and embeded projects, you guys are refusing what I said - as ranting, sad...
16:10:46peturaustriancoder: I guess it will be around 21:00 :/
16:10:50gregjmy wasted hour than
16:11:02 Join bnakibbho [0] (i=0@
16:11:18LinusNgregj: not wasted at all, but i don't think we have the same goals as kde
16:11:27austriancoderpetur: so in... 5 hours?
16:11:39peturok for you?
16:11:39LinusNgregj: and i do agree that releases are good
16:12:10markunamiconn: yes, I know, it was my way of trying to be funny
16:12:16gregjI know you don't have the same goas as KDe
16:12:18austriancoderpetur: yup
16:12:37gregjevery single suggestion was rockbox specific, not a copy of what KDFE is doing, etc
16:12:58gregjevery single project has different goasl, different ppl, and different <anything>
16:13:28gregjbut some things are are always the same, everyone one wants its code to be good, fast, feature complette, and have happy users
16:13:49*Mouser_X is a happy user on the Gigabeat.
16:13:49gregjand above all, everyone would love to have hardware producers to support them
16:14:01gregjevery embeded project that is
16:14:03 Quit desowin ("use linux")
16:14:12markungregj: you forget the fun part
16:14:29Nico_PI'm here for the fun :)
16:14:32gregjit is always fun, because no one forces you to do it
16:14:41gregjyou want - yu do, you don't want - you don't - someone else will
16:15:05LinusNgregj: ...and there goes the release schedule
16:15:17Febsgregj: how then would you propose to accomplish the chart that you suggested, which is−−by your own admission−−not something that a developer wants to do?
16:15:23gregjby release schedule, I mean - some goal
16:15:50gregjdo you guys have project leader, or someone ?
16:16:11*LinusN points to himself, björn and daniel
16:16:39markunthe last two are also known as Zagor and Bagder
16:16:44gregjthat would be the man usualy, who knows everything about the project - and should at least suggest what should be on that chart, and what has to change
16:16:58FebsBut who is going to do the chart?
16:17:16FebsI mean, the chart is great in theory, but if it is not going to ACTUALLY get done, then it is not useful.
16:17:22gregjchart is just an example, anything that would work this way - show what's in what's out
16:17:35LinusNthis is the chart we used last time we tried:
16:17:44gregjI've seen it on wiki actually, for ipod - but I konw already for a fact that it is outdated, some develops here told me
16:18:05markungregj: things like this?
16:18:18gregjso you guys have such things
16:18:32Nico_Pamiconn: by any chance, don't your recent changes fix FS #6040 ?
16:18:43gregjr38 - 29 May 2007 - 21:40:22 - PeterOlson
16:18:47gregjupdated even not so long ago
16:19:31FebsNote the proposed release dates.
16:19:37gregjI would suggest that someone working on the stuff that's on the list, to update specific part of that site
16:19:40gregjyeah :)
16:19:50Febsgregj: again, who will work on that?
16:19:53gregjcool, see, so you guys are telling me - I am ranting, and you have something going on
16:20:07markungregj: wasn't it the leader of the project who should suggest such things?
16:20:31gregjFebs: lets say you want to do "voice support" 100%, once you're done - you modify that page - to reflect what you just did
16:20:57amiconnNico_P: That's a somewhat strange bug report. Boosting is now enabled on H10, but I didn't get any feedback yet as to whether it's stable (I expect it to be stable though)
16:21:31amiconnBut most details in that report aren't related to boosting at all...
16:21:40gregjmarkun: yes, but if - say - developer X wants to do something, he can put it in a flyspray, or whatever - discuss the thing on mailing list, and than put it there himself, or let project leader speak about it - and devvide, at the end of the day - this is why he is called 'leader'
16:22:14Nico_Pamiconn: I wish I had thought about this before... my brother has a (non-rockboxed) H10 I could test on, but he's leaving for a trip today
16:22:44gregjso the release date was may 2006, and so far - year later, some features are not finished - but I bet, there's 1000 more features - half done :)
16:22:55LinusNgregj: absolutely
16:23:09markunalso some which could be called 100% done
16:23:15gregjso, still, no one executed that plan, no leader
16:23:18markuneven if they contain some bugs
16:23:25gregjbugs are normal
16:23:28Nico_PI'm under the impression that the dv ML isn't used much
16:23:29gregjsort of
16:23:41 Quit Zagor ("Client exiting")
16:24:24Tim_Mitchelldoes anyone know how complicated adding remote control functionality for the gigabeat when HOLD is activated? i.e. would it be too complicated for someone's first rockbox patch?
16:24:32LinusNgregj: we had a feature-freeze period back then
16:24:34gregjmarkun: bugs happend, but if I was a user, without much knowledge about developing - I would give up testing after first three things didn't worked
16:24:35pondlifegregj: So how should a leader make someone fix a particular bug? Do they ban them from working on any other part of Rockbox until they have completed their allocated work?
16:24:50markungregj: here you can see what has been done since then:
16:24:55LinusNgregj: but it turned out that most developers vanished
16:25:04gregjin other projects (TM) - you are responsible for the feature, you fix bugs
16:25:11pondlifegregj: If the first 3 things didn't work, read the manual!
16:25:18GodEaterTim_Mitchell: do you mean when hold on the main unit is switched on ?
16:25:19gregjthe only thing leader can do, is he can remove your feture from release - if it ain't fixed/done 100%
16:25:28markunTim_Mitchell: the remove doesn't work for you with HOLD enabled?
16:25:34markunthe remote..
16:25:37gregjLinusN: so they were not serious about it
16:25:45LinusNgregj: exactly
16:26:11Febsgregj: OK, so when the feature is "Software codec playback," how does one remove that from the release?
16:26:15gregjLinusN: I know 10000 ppl who can do stuff, half finished, because they have no longer interest in , or are not man enough to finish something they started
16:26:18*GodEater laughs at seeing the same conversation we always have when a newbie keen developer comes along
16:26:33pondlifegregj: Well send them our way
16:26:36Tim_MitchellI haven't bothered using the remote because the wiki page says that the remote won't work if hold is enabled
16:26:45GodEaterTim_Mitchell: the wiki is wrong then :)
16:26:56*gregj laughs at ppl laughing
16:26:58Mouser_XTim_Mitchell: Works fine for me. It's how I use my Gigabeat most of the time.
16:27:20markunMouser_X: did you notice any improvements lately?
16:27:26LinusNgregj: but that doesn't mean we shouldn't try again
16:27:28gregjpondlife: wwhat for, to only leave things half done - and no one else to fix them later ?
16:27:30Tim_Mitchellsee, as a software developer, I should have known better not to bother reading the documentation :P
16:27:37markunMouser_X: well, I did some commits not too long ago
16:27:39GodEaterTim_Mitchell: very true :)
16:27:39markuna few weeks
16:27:56pondlifegregj: They get finished eventually. And certainly shouldn't break anything else.
16:27:58Mouser_Xmarkun: I'm using a build that perldiver made for me. It has MOD support. It's at least a month old...
16:28:01gregjLinusN: true, but if you see thign that is almost working - what is exactly purpose of adding it to release
16:28:18gregjFebs: you are not serious asking such simple questions, are you
16:28:18pondlifeSo someone else can run with it, get inspired and complete it?
16:28:21markungregj: I did a half-ass implementation of a crossfeed DSP effect and later someone came and fixed it
16:28:48LinusNgregj: well, in the case of the Mayday release, we coudln't really remove the Playback feature :-)
16:28:54markunbut features need to be broken enough for people wanting to fix them I guss :)
16:29:08gregjokay, but ould you give it to someone who is end user, to use it ?
16:29:15GodEateror for someone to want them in the first place
16:29:20 Part Tim_Mitchell
16:29:24 Join desowin [0] (
16:29:25gregjLinusN: the question was silly, obviously you cannot remove it - it either works or not
16:29:40LinusNso we did have to add it to the release
16:29:43gregjLinusN: it should be 'finished, with some known bugs'
16:29:49pondlifeWe are all users. Most of us have a very limniteed amount of dev time to donate - in my case about an hour per month.
16:30:05LinusNgregj: ...which brings us back to "user experience"
16:30:17gregjlike I said , very important part
16:30:28LinusNwe can't really release a buggy playback engine
16:30:46gregjLinusN: than make first priority fixing it
16:30:51gregjusualy bugs have priority
16:30:55LinusNit has always been prio #1
16:30:58Febsgregj: You suggested that if features are incomplete, they be removed from the release. My question was no more "silly" than your suggestion.
16:31:06gregjlike in my company, P1 bugs are show stoppers,
16:31:14gregjp5 and p6 don't matter
16:31:16pondlifeThey are here too, hence no 3.0
16:31:36gregjif the release branch doesn't have known p1 bugs, and all features are completted (the chart for the releasre) -we release
16:31:49LinusN...and that has yet to happen here
16:31:52FebsOK, and that brings us back to exactly where we are now.
16:31:53gregjin next 6 months, we release fix version for that release, etc
16:32:15pondlifeYes, but you probably pay people so you can say what they should fix.
16:32:31gregjthat's why I give them deadlines
16:32:40pondlifeSo what if they ignore you?
16:32:50pondlifeDo you just lose a developer?
16:32:52gregjsee, 99% of KDE developers don't get paid for their work, it is for fun only
16:33:02gregjpondlife: that mens bugger just got laid off
16:33:08 Quit austriancoder ("Kopete 0.12.5 :")
16:33:25*pondlife wants his "I was fired from Rockbox" T-shirt
16:33:46*LinusN prepares the "i fired them all - haha" t-shirt
16:33:47gregjschedules just keep things organized,
16:33:48peturit certainly doesn't sound like fun to code for KDE then...
16:33:56gregjit is fun
16:34:03GodEaternot my idea of fun
16:34:09gregjbut, - makes developers responsible for what they do
16:34:09peturme neither
16:34:27gregjif user complains, there is bug in bugzilla - I feel responsible for fixing it
16:34:28markungregj: here you just get yelled at if you broke the builds
16:34:34GodEaterwell last time someone broke rombox amiconn just went round and beat them up
16:34:47peturgregj: contrary to the KDE devs, I think most RB devs have a real life and enjoy it too
16:34:54GodEateryes - we like beer
16:35:05markunI'm drinking beer as we speak :)
16:35:09markunat work
16:35:13gregjGodEater: well, he broke RB on ipod twice when I used it - and got beaten here by me
16:35:13GodEaterthat's so cool
16:35:16*Mouser_X doesn't drink alcohol.
16:35:28markunMouser_X: that's also cool :)
16:35:28*gregj is at home, working for work - and trying to help you here
16:35:30pondlifeThere are alcohol free beers
16:35:49pondlifeWell, you're helping amuse me during work... :p
16:35:53*petur gets reminded he should be working...
16:35:57Mouser_Xpondlife: Which taste absolutely nasty.
16:36:02LinusNstill, i think we could do with straightening out the ReleaseTodo
16:36:13*petur wipes the page
16:36:16peturall done
16:36:24GodEaterI'm waiting for the, "and you should include malloc()" argument to start now
16:36:34 Join and [0] (
16:36:35pondlifeMust we...?
16:36:36LinusNGodEater: hahaha
16:36:47*GodEater readies his pitchfork and torch.
16:36:49*petur runs away *fast*
16:36:50gregjI hate to say, you need some leader's to organize the thing bit, tiny bit more :) - and more users are happy, and stuff is even more RockBoxSolid
16:36:56*gregj quits the managers talk
16:37:16*petur repeats: gregj: you're oversimplifying so much I almost wonder if you aren't a manager in disguise :p
16:37:22markungregj: so, are you joining rockbox?
16:37:24pondlifegregj: Again, I suggest you start by logging the bugs you've found.
16:37:39GodEaterpetur: he already admitted to "project management experience" :)
16:37:44pondlife(Assuming they're not already logged, of course.)
16:38:14gregjmarkun: I'll think about it, but I like to work with serious ppl - no offence, but lauging at some things I just said - wasn't quite that
16:38:25gregjI have them written down
16:38:36LinusNwell, if the goal is to make a release and make rockbox more popular - then perhaps we need a leader to organize it a little bit more
16:38:54pondlifeI think stability is a good aim, not popularity,.
16:38:56gregjsome of them might just be 'broken' UE, from your perspective something you know about, but you don't use stuff in that way
16:39:08markunwhat's UE?
16:39:12gregjuser expierence
16:39:13LinusNuser experience
16:39:28pondlifeWell if the UI is inconsistent and/or the manual is wrong, then it should be fixed.
16:39:36markungregj: it's nice to have first time users report problems
16:39:40gregjlike text appearing from under the bitmap, in user interface, etc
16:40:25markungregj: sometimes you just need to fight a bit here to make people see you are right :)
16:40:25pondlifeBitmap? A backdrop?
16:40:30gregjsomeone said somewhere that he gets 5h battery life for instance, I get 11 on apple's firmware, 3h on rockbox
16:40:53markungregj: the navigation used to be a bit chaotic to the first time user but most of us were fine with it
16:40:55gregjmarkun: this is always the case, if you have someting, and you can prove it - I am trying to prove sometihng, I know I'm right
16:41:02LinusNgregj: it's interesting that you brought the battery issue up
16:41:12markunit took a while discussing it until we changed it to the current layout
16:41:28LinusNgregj: there you have another fundamental difference between kde and rockbox
16:41:38gregjhaving customizeable layouts is cool, why not do the same thing for main menu too ?
16:41:39dionoeakde is bloated rockbox isn't ?
16:41:46dionoeaoops, sorry about that
16:41:57LinusNgregj: kde involves a lot less reverse engineering of hardware
16:42:11pondlifegregj: Because we should be fixing the audio playback first, not adding new eye candy...?
16:42:13GodEatergregj: support nightmare when we allow users to do that
16:42:22markungregj: the proof I have is just from experience here. We all thought a realse was a good idea until we figured out it was not going to work.
16:42:35gregjalso, I would suggest - going either databse ddirection (my favorite), or playlists direction (both can be merged)
16:42:52LinusNwe want both
16:42:54*GodEater will quit using rockbox the day it goes Database only
16:43:02dionoeasame here
16:43:04*petur too
16:43:14peturor rather, we'll fork ;)
16:43:26GodEaterI stopped using the Apple firmware for a reason
16:43:27gregjwhy not make it - playlist gets parsed - hops into database (for faster access)
16:43:29 Quit desowin (Read error: 113 (No route to host))
16:43:52gregjand you have db only backend , but you can still use playlists as you wish
16:44:13peturgregj: because some of us don't use the database and rather use the RAM it eats for something else
16:44:17*dionoea loves how people who've known some software for less than a day always have great ideas about what should change to make it better :)
16:44:33markundionoea: it's normal
16:44:41[Ray]it's called a fresh look at thing
16:44:45dionoeayeah, we get that all the time on the VLC forums :)
16:44:53LinusN[Ray]: good point
16:45:00gregjabout db, when RB advance from one song toother - it shows a file name for 2-3s - shouldn't that be updated from db, not from file's ID3 ?
16:45:09gregjdionoea: you must be very young than
16:45:26markungregj: It's not clear to me what you are sugesting, maybe you could make a wiki page or patch or something?
16:45:30dionoeaand why ?
16:45:48*GodEater also fails to spot the link
16:45:51gregjthe worse enemy of every progress, is 'nah, it is enough for me'
16:45:54GodEaterand isn't by any definition young
16:46:23LinusNgregj: why is progress important when it is good enough?
16:46:24Crash91guys, to install a patch, do i just cd to /home/crash/rockbox and then do patch < music_led.patch
16:46:25pondlifegregj: You should certainly log that last one - there is development of buffered metadata going on.
16:46:32markungregj: and it *is* enough for me!
16:46:44GodEaterCrash91: usually yes, if it's a well written patch
16:46:59pondlifeCrash91: You may need to do patch -p0 < ...patch
16:47:00gregjI think that someone else looking and making suggestions, from just user perspective - not knowing the code - is very good indeeed
16:47:04LinusNCrash91: patch -p0 < music_led.patch
16:47:12pondlifegregj: Absolutely
16:47:13Crash91and what about *.c files? How do i make a *.rock
16:47:14gregjif you know the code, and how it works - you are blinded by some of that knowledge
16:47:14LinusNgregj: definitely
16:47:20peturcontinuous progress is where bloatware starts, most small usefull programs I know ended like that
16:47:24LinusNgregj: amen to that
16:47:25GodEaterCrash91: one step at a time tiger
16:47:30gregjnot realising that, and laughing at it - is someone very young would do
16:47:36gregjhence my comment
16:47:38Crash91GodEater: :D
16:47:56gregjs/someone/something someone/
16:48:23pondlifeyoung, or flippant
16:48:23*GodEater agrees with petur
16:48:30GodEaterlook at mozilla / firefox
16:48:34markunthe only thing I'm really worried about it that some day rockbox will not be able to keep up with the DAPs that are still for sale
16:48:38Mouser_XJust to note, I very much prefer to not use the DB as well.
16:48:47LinusNpetur: rockbox is well on its way to bloatware if you ask me
16:48:57Mouser_XIt has its positives, but not enough of my files are tagged properly to make it worth using.
16:49:27markunLinusN: yes, I think so too. Do you know a good strategy to prevent it?
16:49:37LinusNnot without ruining the fun
16:49:40Crash91Skipping patch.
16:49:40Crash911 out of 1 hunk ignored < LOL
16:49:41peturLinusN: not that far yet, or give an example
16:49:44 Join linuxstb [0] (i=d556da1b@rockbox/developer/linuxstb)
16:49:59pondlifepetur: Icons? The DB?
16:50:05LinusN mpegplayer
16:50:12GodEaterDB can definitely go ;)
16:50:18peturplugins don't count
16:50:24Mouser_XI don't consider mpegplayer as bloat.
16:50:41Mouser_XI agree with petur as well.
16:50:56 Quit Quippini (Connection timed out)
16:51:02markunLinusN: the archos recorder had video playback long before the mpegplayer
16:51:02Mouser_X(Not that what I think makes much of a difference, I'm just adding my 2 cents.)
16:51:05 Join toffe82 [0] (
16:51:05peturGUI is far from where it should be to look nice, viewports will help there
16:51:18gregjmarkun: there's so many DAPs on the market, covering all of them is a wet dream
16:51:19LinusNmarkun: yes, and i consider that bloat as well :-)
16:51:29 Quit Entasis (Read error: 104 (Connection reset by peer))
16:51:38gregjI like the fact that there's mpeg player in rockbx, I just wish it would be much better in terms of quality
16:51:49*GodEater makes a mental note to buy LinusN a shuffle for xmas
16:51:50Mouser_XIt works fine for me.
16:51:54gregjso far only one short movie worked on it 100% :)
16:51:56Crash91GodEater: Patched, now what about *.c files :)
16:51:57LinusNGodEater: haha
16:51:59markungregj: you completely missed what I was trying to say
16:52:00 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
16:52:21gregjmarkun: sorry, I only read your last 2-3 lines, had to jump on work emails
16:52:22GodEaterCrash91: make a sub-directory of your /home/crash/rockbox directory
16:52:27GodEatercall it something like "build"
16:52:30LinusNmpegplayer will always have low prio for me
16:52:31GodEaterand cd into it
16:52:55GodEaternow type "../tools/configure"
16:53:08GodEaterthe menu you get should be pretty easy to follow
16:53:09markungregj: I meant more that new players are becoming more and more complex and have more security features
16:53:13gregjand read devels wiki about compilation
16:53:23gregjmarkun: hmm
16:53:36gregjmarkun: I remember, once , I had philips HDsomething player
16:53:44gregjmarkun: the firmware there is encrypted
16:53:45GodEateryeah, apart from the ZuneLinux geniuses, no-ones got code running on the Zune for example
16:53:59markungregj: yes, I guess it has always been a problem not just now :)
16:54:12gregjmarkun: and I saw only one post on the web, regarding its CPU - from the philips developer, asking onlkml, - if they had any luck running linux on it
16:54:34Crash91GodEater: Does arm-elf-gcc have to be in the same directory?
16:54:40gregjmarkun: and I could swear, the reason they encryt the firmware is probably because they use linux or derrived work on it
16:54:41markungregj: there are some portalplayer based players which we could also support, not sure if it was one of them
16:54:42GodEaterno - it has to be in your PATH
16:54:50gregjCrash91: on path, read the wiki
16:55:07gregjmarkun: they used philips arm CPU
16:55:17gregjmarkun: so I doubt it was portalplayer
16:55:23linuxstbgregj: mpegplayer is a work-in-progress plugin. Are you suggesting works in progress like that shouldn't be in the official builds?
16:55:39GodEaterthat's what it read like to me...
16:55:49 Join Farp [0] (i=Farp@
16:55:51gregjlinuxstb: everything here is a work-in-progress thing
16:55:53gregjit seems
16:56:03gregjlinuxstb: bt, where about in UK you are ?
16:56:05Febsgregj: Sorry, someone came into my office so I've been away for a few minutes so forgive me if I backtrack for a minute.
16:56:11*gregj lives in Cambridge
16:56:13GodEaterdon't tell him it's a trick!
16:56:18FebsPlease do not think that I am not taking your suggestions seriously.
16:56:20 Quit amiconn (Nick collision from services.)
16:56:20 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
16:56:28gregjGodEater: :P
16:56:31*linuxstb hides and hopes gregj doesn't find rasher's map
16:56:42GodEateroh damn - I'm on that too
16:56:50pondlifelol - oh me too.
16:56:54FebsIt's just that (a) much of what you are suggesting has been implemented already with varying degrees of success, or (b) I don't know that there is a mechanism would allow us to do what you are suggesting.
16:56:55gregjyes, I will invade your area and kill all DAPs in neck of woods
16:57:59GodEatertalking of location, I completely failed to check out Holborn last night
16:58:00 Quit bnakibbho (Read error: 104 (Connection reset by peer))
16:58:04GodEaterI blame the thunderstorms
16:58:08GodEaterwill try again this evening
16:58:15GodEaterDevConPub WILL be a roaring success
16:58:34pondlifeHah, that's (c) pondlife... and I'm not even going
16:58:49GodEatergo on - you know you want to
16:58:54 Join lnakioahd [0] (i=0@
16:58:54 Quit maffe (Remote closed the connection)
16:58:55pondlifeIt's true
16:59:01pondlifeBut I doubt I'll get time
16:59:08gregjLondon, nice
16:59:13FebsHey, if I can make it there from Philadelphia, you have no excuse ...
16:59:15GodEatercurses - he found us
16:59:17gregjI used to live there for 3 years, being active in GLLUG
16:59:24gregjGodEater: :P
16:59:33pondlifeGLLUG - Sounds like a drinking club...
16:59:39 Join maffe [0] (
16:59:44gregjgo and tell em
16:59:50gregjthey will like it :)
16:59:53LinusNwe are all members of RUG
17:00:12pondlifeYou mean they don't have regular GLLUG glugs?
17:00:36markungregj: some of the guys you have been talking to:
17:00:45GodEaterLinusN: does that mean we have to wear wigs ?
17:01:07Crash91GodEater: you dont seem to be responding...
17:01:13LinusNGodEater: oh yes
17:01:24GodEaterCrash91: are you registered with irc services ?
17:01:27Nico_Plinuxstb: wrt the 32K codec reads, what heppens when the end of the file is reached ? it's not possible to provide 32K in that case...
17:01:34GodEaterthat'll be why then
17:01:48Crash91ok join #crash
17:01:59linuxstbNico_P: No, but you'll get an entire frame (assuming the file isn't corrupt, in which case I'm not sure what will happen...)
17:02:29gregjwine sounds so much better than beer
17:02:43linuxstbNico_P: I think most codecs won't request data beyond the end of the file - they know how many frames are in the file, and stop at the last one.
17:02:45Nico_Plinuxstb: what does that mean ? still 32K or less than that but clearly indicated ?
17:02:56LinusNi have to go - nice chat gregj, i hope we haven't scared you off :-)
17:03:09gregjI tough
17:03:25 Part LinusN
17:03:56Nico_Plinuxstb: is a frame alsways 32KB ?
17:03:59linuxstbNico_P: The read() function will return however many bytes are left in the file. So if a codec requested 32KB of data, but only 18KB is left in the file, it will only receive 18KB.
17:04:16linuxstbNico_P: No, frames are typically a few hundred bytes. 32KB is a worst-case maximum.
17:04:30linuxstb(size depends on codec and bitrate)
17:04:33markunpetur: we need to start writing some kind of progress report for google soon, right?
17:04:37***Saving seen data "./dancer.seen"
17:04:58Nico_Pand will the codec accept to get less than 32KB in the middle of a file ?
17:05:21peturmarkun: yes
17:05:25gregjmarkun: who's doing wma thing ? should it play wmv's sound ?
17:06:03peturgregj: it is barely one week old, do be patient
17:06:09markungregj: saratoga is working on it
17:06:13linuxstbIt all depends on the codec. Some codecs will error-out if they don't get a complete frame, so decoding will stop. Others might be able to recover and continue playing. It may make it clearer if you look at some of the codecs (the .c files in apps/codecs)
17:06:17markungoogle summer of code project
17:06:50 Join freqmod [0] (
17:06:54Nico_Plinuxstb: OK, so the only case where the codec will accept less than what it requested is twoards the end of the file ?
17:06:55markungregj: I'm not sure what it should play. wmav1 and v2 I think
17:06:57linuxstbgregj: I wrote the ASF parser used for WMA. It plays the first wma audio stream it finds in the file, ignoring all other packets. I've only tested it with a few wmv files, but it plays them fine.
17:08:20linuxstbNico_P: Im not sure what you're asking - a codec should always receive what it requests (apart from the expected end-of-file condition), but codecs are written not to request more than 32KB.
17:08:22gregjI am patient, I know it is barely started - I am just asking, don't eat me alive
17:08:28gregjplease plase please, oh noooooo....
17:08:42markundrama queen :)
17:08:53*gregj uploads wmv to server
17:09:22Nico_Plinuxstb: I just wanted to make sure I undertanded what the codec expects wrt to the size it requests
17:09:33 Join bluebrother [0] (i=V380D3Sm@rockbox/staff/bluebrother)
17:09:42 Quit and (Read error: 104 (Connection reset by peer))
17:10:08Nico_P*understand even
17:11:14 Join thepenguin [0] (
17:11:51 Quit bospaadje ("great minds run in great circles")
17:12:53 Join bospaadje [0] (
17:14:01thepenguinHello. I have an issue with rockbox on my 5.5g 30gig iPod.
17:14:27thepenguinI am using the database to play my songs...
17:14:43thepenguinWhen I try to play a song from the track menu
17:14:58thepenguinThere is an ubearably long delay while it searches through all the tracks
17:15:01thepenguinand then it plays it
17:15:21thepenguinI have load to ram enabled
17:15:26thepenguinand dircache enabled too
17:15:29thepenguinany ideas?
17:16:09linuxstbgregj: 403: Forbidden
17:16:15gregjlemme check
17:16:23Mouser_Xthepenguin: Don't use an iPod? (Only somewhat serious there)
17:16:37thepenguinwasn't my decision :P
17:16:58Mouser_XI figured. Thus my partial lack of seriousness.
17:17:03thepenguinsadly, it is the only thing I have
17:17:22thepenguinI'd like to make the switch over to rockbox
17:17:27thepenguinbut this isn't helping much
17:17:40gregjlinuxstb: try now
17:17:40Mouser_XI don't use the DB, so I'm no help. Sorry.
17:17:46GodEaternor I
17:17:47thepenguinahh OK
17:17:49 Join miepchen^schlaf [0] (
17:17:53linuxstbgregj: Downloading now...
17:17:55GodEaterI deleted all the iTunes loaded music
17:18:13Crash91GodEater: Lastly.....what about .c files...
17:18:15gregjlinuxstb: it's not uploaded as whole yet, but you should get the idea from what's there
17:18:26GodEaterCrash91: that's what "make" is doing - it's building them
17:18:41Crash91so i just place the *.c s in plugins?
17:18:48GodEateroh - where did you get it from ?
17:19:04GodEaterwhat is it ?
17:19:22gregjlinuxstb: I know nothing about wmv and wma , other than it is crappy quality
17:19:29linuxstbgregj: Do you know how it was encoded?
17:19:37gregjno idea
17:20:02gregjthe guy who plays there, he encoded that on windows
17:20:29GodEaterCrash91: no - it's a bit more involved - you have to put the .c file in the plugins dir, but also need to edit the SOURCES file
17:20:37GodEaterand I've no clue where the contents of the file go
17:20:54linuxstbHmm, it contains the same troublesome "frame split into multiple payloads within one packet" that ffmpeg generates... Looks like I'm going to have to deal with such cases, which is a pain.
17:21:18 Join printfXh4 [0] (
17:21:35gregjlinuxstb: :(
17:21:41thepenguinwhile I'm there a way to sort playlists by artist?
17:21:47gregjso, can I use the mpeg player on symulator ?
17:22:10linuxstbIt worked in the past, but I haven't tried it recently...
17:22:14dionoeai believe that it segfaults in the simulator (might be wrong though)
17:22:42GodEaterthepenguin: to be honest, usage of the database is best asked in the forums - bascule is the expert, and he doesn't hang out here much
17:23:15Crash91GodEater: /apps/plugins/bitmaps/native
17:23:28*Mouser_X only visits the forums when they're linked to in here...
17:23:36Mouser_X(And even that's rare.)
17:23:54GodEaterMouser_X: and you, like me, have no idea on anything but basic usage of the database
17:24:04GodEaterhence asking here is pretty much a waste of time :)
17:24:40linuxstbI'm not sure of any devs apart from Slasheri who regularly use the database...
17:25:05thepenguinI may end up using a ID3 tagger to rename all the iTunes imported songs and just make playlists
17:25:14thepenguinif that will work
17:25:38Mouser_XIt's worth the try, IMO.
17:26:08thepenguinthanks guys, will get started =)
17:26:23gregjlinuxstb: this one is also borked on current builds, sound is funny
17:26:25GodEatersorry we weren't more help thepenguin
17:26:36gregjlinuxstb: but I'll try to see on simulator, if I can fix it
17:26:41thepenguinno problem...
17:26:50thepenguinit's not your fault :P...Apple *shudders*
17:27:44gregjlinuxstb: sound makes loads of problems here, usualy after one minute it stops sound playback, and than even keys are unresponsive - which makes me wonder
17:27:53GodEaterI expect Bascule would probably know the answer - it just depends on how impatient you are for your answer
17:28:45thepenguinI replied to a 2 page thread that described the exact same problem
17:28:52GodEateryes I saw
17:28:53thepenguinhopefully that will bring it to someones attention
17:29:00gregjlinuxstb: sorry, not trying to rant here- but I wish someone said - "yeah, that is x, and y", no time to fix it, but this can be done z/e
17:30:04 Join XavierGr_ [0] (
17:30:08 Quit XavierGr (Nick collision from services.)
17:30:12GodEaterthat would require someone who was an expect with wma
17:30:15GodEaterwhich none of us
17:33:08GodEaterwoo! DataAbort at 00038FC0!
17:33:13gregjexpert you mean?
17:33:24GodEateryes - typo sorry
17:33:44gregjI just wasn't sure
17:34:13 Quit petur ("gone shopping")
17:35:39Crash91GodEater: I seem to have to do source ~/.bash_profile every time, is this normal?
17:36:28GodEaterno - if you logout and back in again it should pick it up every time
17:36:35 Join desowin [0] (
17:38:59pixelmathepenguin: are you just "clicking" on the song you want to play from the "all track" in the database by any chance?
17:39:02*GodEater notes Crash91 doesn't disappear and so hasn't followed the above advice
17:39:28*Crash91 feels like logging out later
17:39:54thepenguinpixelma: yes
17:40:02 Join n1s [0] (
17:40:23*Crash91 is at learning C
17:40:58GodEaterCrash91: we'll expect a new codec from you by Friday then :)
17:41:22linuxstbCrash91: If you change .bashrc instead of .bash_profile, you shouldn't need to logout, just open a new terminal window. .bash_profile is for login shells.
17:41:39pixelmathepenguin: that will make a playlist with all songs it can find in there, starting with the one you chose... that can take a while depending on how many are in there
17:41:50Crash91ok, thanks
17:41:58thepenguinpixelma: it happens with every song one after another
17:42:11thepenguinpixelma: if i select a song, let it load and then select another song
17:42:15thepenguinit does the same thing
17:42:28thepenguinhowever, if i click the next button, it plays it instantly
17:42:29Crash91GodEater:Ill expect 5 plugins from you, as youre more experienced :)
17:42:43*GodEater has a day job
17:43:14pixelmayes. And in your while playing screen is there a number that says "playing song #1 of xx"? If so what number is xx?
17:43:54Crash91Fair enough, im on holiday for 2 more months :)
17:43:56pixelmathepenguin: or something in that line
17:44:16*GodEater 's last holiday was May last year
17:44:18thepenguinthere is a [37:1194] which I assume is what you're looking for
17:44:23GodEaterfor 10 whole days
17:44:26thepenguinas I have 1194 songs loaded
17:44:41thepenguinI selected around the 37'th song on the list
17:45:14pixelmathepenguin: I think that's what I mean. Collecting 1194 songs in the playlist will obviously take a while
17:45:35Crash91/crash91 thinks GodEater needs to "Have a break... have a Kit Kat" before being kicked for being off-topic
17:45:35thepenguinshouldn't it only be a one time thing though?
17:45:42Crash91oops, i meant /me
17:45:56 Join guest41 [0] (i=864c0346@gateway/web/cgi-irc/
17:46:24pixelmathepenguin: if you select it from the the database again, it'll start creating the playlist again
17:47:05thepenguinwhere is this playlist and how do I play songs from it
17:47:08GodEaterso don't select from the "All tracks" section again :)
17:47:13guest41thepenguin: does it display "37:1194" directly after you have selected the song or after you have skipped/listened a few tracks?
17:47:57pixelmathepenguin: you can view the playlist from the while playing screen but if you put your songs on using itunes, that won't help you much
17:48:07thepenguinthat's why
17:48:15thepenguinI was confused at first...the names were all crypted
17:48:28thepenguinanother reason I despise Apple =)
17:48:59thepenguinthanks a lot, that explains it =)
17:49:36 Join My_Sic [0] (
17:49:43pixelmayou could also insert the one track you want to hear only. "Long select" on the song and then "insert", choose one of the insert options
17:50:38*Crash91 thinks GodEater has gone to sleep
17:50:59*GodEater still has a day job and is talking to his boss
17:51:01thepenguincould a feature be added that will use the track's tag instead of file name when generating a playlist?
17:51:09pixelmathepenguin: or don't use the "all tracks" section but the more organised
17:51:48pixelmathepenguin: there is a patch for it in the tracker, and I guess that will be implemented officially... *one day*
17:51:53 Join saratoga [0] (i=41becbf6@gateway/web/cgi-irc/
17:52:12thepenguinthen I think I want that patch =)
17:56:09pixelmaI'm not too sure if that's synced and how good it works, also don't know the flyspray number as I don't use the database myself in everyday use
17:57:12markunmy god, look what happened to
17:57:32markunjust when the project was doing so well..
17:58:40Crash91why would their own team do that??
17:59:55saratogalinuxstb: the problem you mentioned with superframes, which file is that in reference to?
18:00:30 Quit guest41 ("CGI:IRC (Ping timeout)")
18:01:57saratogaso far the problems i'm aware of are 1) WMAV1 files do not work because I left a few FP multiplies in that code path, and 2) line spectral pair stuff isn't finished so low bitrate files fail
18:02:59linuxstbThe ffmpeg encoded sample someone posted about 12 hours ago
18:03:00Crash91GodEater:[crash@rt rockbox]$ svn revert -R
18:03:00Crash91svn: Not enough arguments provided; try 'svn help' for more info
18:03:13linuxstb(the wmav1 file)
18:03:28saratogalinuxstb: ok thanks
18:03:51 Join Wiwie [0] (
18:03:52saratogai only noticed the problems in my part of the code, so I didn't realize there was a problem with parsing too
18:04:25linuxstbI'm thinking that the handling of multiple payloads per packet should be done in the wma decoder itself, but I'll wait until the decoder has been cleaned up a little more before investigating that.
18:04:35 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
18:05:43 Join Domonoky_ [0] (
18:05:43linuxstbIt's possible that each payload is an individual 2048 sample block, in which case it could be feasible for the codec to ask the decoder to decode one block at a time.
18:07:31saratogacould you elaborate some more about how asf packets work? How big are they exactly?
18:07:54saratogado they always have to contain at least one superframe or can they contain fractions?
18:09:16toffe82markun: they have a myspace page now :)
18:09:31linuxstbsaratoga: I only know what my example files (which I've gathered from various corners of the internet) look like.
18:11:00linuxstbASF files consist of objects, and objects can contain sub-objects. IIUC, a typical ASF file contains a header object, followed by a data object. The data object contains packets (typically around 3-10KB), which in turn contain sub-packets (called "payloads").
18:11:19linuxstbUntil I saw the ffmpeg file this morning, I assumed that (for audio streams), one packet = one payload = one superframe.
18:11:41linuxstbIn the video streams I've seen, the packets always have multiple payloads.
18:13:06markuntoffe82: I wonder who this insider was which removed the database. Maybe he did it himself? :)
18:13:55linuxstbBut the ffmpeg file I saw (and the example from gregj), the audio stream packets are split into multiple payloads. They look like the packets still contain a whole superframe, but they're fragmented (they have small payload headers between the parts of the data) into payloads.
18:14:07toffe82markun: or you under a pseudo ;)
18:14:25markuntoffe82: nah, I was having too much fun with the old site
18:14:46linuxstbBut I want to change that anyway - a whole superframe is too much to decode in one go, I think it would be better if we could decode 2048 samples at a time.
18:14:58toffe82markun: look the blog on the myspace page, it is the same
18:15:07linuxstb(reducing memory requirements, and hopefully allowing more IRAM usage on Coldfire)
18:15:59 Join ackbahr [0] (
18:18:11 Join rotator [0] (n=e@rockbox/developer/rotator)
18:18:28preglowlinuxstb: word
18:18:36preglowdecoding a block at a time is the way to go
18:20:01preglowbtw, the clipping part in the fixed32 -> short conversion is kinda redundant
18:20:17preglowthe fixed point format is 16 bit anyway, so it'll overflow the limits at which clipping happens
18:20:34saratogapreglow: not to mention I haven't found any files that clip anyway
18:20:41saratogaexcept for broken ones anyway
18:20:43preglowsaratoga: that would be my next point, yes
18:20:53preglowi'd be very surprised if 16.16 gives enough headroom
18:21:18saratogai think thats left over from the original codec which overflowed constantly
18:21:38preglowtry encoding an almost full scale square wave, for example, then encode that
18:21:43preglowif that doesn't overflow, then it never will
18:21:56saratogai'll take a look at it now
18:22:28linuxstbsaratoga: Is your SVN access sorted out yet?
18:23:00 Quit borisyeltsin (Remote closed the connection)
18:23:16 Join borisyeltsin [0] (
18:25:48 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
18:26:17 Join Wiwie [0] (
18:27:21 Join petur [0] (n=petur@rockbox/developer/petur)
18:29:16 Join lazka [0] (
18:31:37preglowlinuxstb: btw, there might be some problems when trying to make dsp convert from 32 bit samples for wma
18:31:45saratogalinuxstb: linus says so, but I haven't tried it yet
18:32:21preglowlinuxstb: wma is our first codec which has absolutely no headroom, so there might be problems with the generated clipping constants
18:33:01saratogaheh, square wave does clip a little bit
18:33:45preglowclip or overflow?
18:33:59saratogaoverflow I assume, though it sounds like clipping
18:34:05saratogai don't think it can actually clip
18:34:07 Join My_Sic [0] (
18:34:08preglowhmm, could you post a sample of some kind?
18:34:14preglowi'd be more surprised by clipping than overflow
18:34:24saratogai'm sure its not clipping
18:35:30preglowoverflow isn't good at all
18:35:40 Join seeee [0] (
18:35:45preglowall codecs are capable of it, but most of our codecs have at least three bits of headroom to prevent it
18:35:50saratogaactaully, I suppose its probably both, since that last step before the clamping left shifts one bit
18:35:58seeeehi all
18:36:00 Join ackbahr_ [0] (
18:36:04saratogathough i'd guess the clipping is just due to overflows earlier on
18:36:37saratogapreglow: its not a big deal, I probably just need to renormalize by one or two bits more in the FFT
18:36:48saratogaright now I just added bits one by one until my test files sounded right
18:36:50The-CompilerHi seeee :)
18:38:23preglowit has to be eliminated anyway
18:38:28preglowoverflow is nasty business
18:39:01saratogait doesn't happen with my peak normalized test file, so its probably not a huge deal for real files
18:39:07saratogai'll try and have a fix tomorrow
18:39:20saratogatoday is independence day, so i probably won't have much sober time
18:39:37preglowthought that was today :>
18:40:06preglowit'll probably be a problem for heavily compressed and limited music
18:40:19preglowlike most modern music, heh
18:41:01saratogai want to submitt that patch i made yesterday that moves the fixed stuff out of wmadeci.c
18:41:06preglowtime for food
18:41:08preglowsure, commit away
18:41:13saratogashould I check out a fresh svn copy and test it first you think?
18:41:26saratogaor is that paranoia
18:42:22pondlifeNo, check it first..
18:42:54markuncould also be paranoia :)
18:43:05pondlifewho says :(
18:44:37Crash91guys, how do i revert my svn? svn revert -R doesnt work
18:44:53linuxstbAdd a "." - svn revert -R .
18:45:36Crash91Thanks :)
18:45:53 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
18:46:52 Quit scorche (Connection timed out)
18:47:51 Quit obo ("KVIrc 3.2.6 Anomalies")
18:52:20 Quit ackbahr (Read error: 110 (Connection timed out))
18:56:16 Join DC1 [0] (
18:56:33saratogais there a coldfire optimized fixmul32b?
18:56:43saratogaor did that get folded into CMUL()?
18:56:46 Quit ackbahr_ ("ChatZilla [Firefox]")
18:56:56 Quit Olav ()
18:57:28Crash91guys, i think the music_led patch is redundant
18:57:38Crash91nothing seems to be working on my sansa
19:00:09 Join XavierGr [0] (
19:01:46 Part MrKeuner ("Ex-Chat")
19:04:38***Saving seen data "./dancer.seen"
19:06:03 Join My_Sic [0] (
19:06:18saratogaanyone going to be around to see if I'm about to break coldfirm playback?
19:06:18 Quit My_Sic (Client Quit)
19:06:23saratogai don't have one of those to test on
19:07:26amiconnipod G5.5/80 runtime: 12h 15 min
19:08:48amiconnipod mini G2 runtime: 8h 40min
19:09:06 Quit XavierGr_ (Read error: 110 (Connection timed out))
19:09:18 Join obo [0] (
19:09:34amiconnThe latter is practically no change from what I got before my clock setup changes
19:10:18dionoeadid you figure out the log function?
19:10:39amiconnIt turned out to be a "simple" integer division
19:10:59dionoeaah :)
19:11:48dionoeaIs that going to be useful to lower CPU voltage ? (and raise battery life)
19:12:36amiconnI added proper setup for the register where this division was involved in calculating the setting
19:12:57 Join jhulst [0] (
19:13:15amiconnBut now that I know the value is proportional to cpu clock, I don't think it's a voltage setup
19:13:43amiconnIt's more like a timing setup (dram refresh timing?)
19:13:50 Quit jhulst (Read error: 104 (Connection reset by peer))
19:15:25dionoeaah ok. It's less likely to be related to the lower than OF battery life issue?
19:22:16 Join BiptoN [0] (i=4ca6cb5d@gateway/web/cgi-irc/
19:23:00BiptoNHello everyone, happy 4th of july to all the american folks
19:24:45bluebrotherhmm. Does it really make sense to have a feature request about mpeg4 video open?
19:25:08 Quit DC1 (Remote closed the connection)
19:27:36pondlifebluebrother: Have we normally just closed them? Maybe the Gigabeat is capable...?
19:27:52pondlifempeg4player ? ;)
19:28:05bluebrothernot exactly sure ... but I think it's kinda pointless as about all targets would be to slow.
19:28:16bluebrotherand you need to transcode the video anyway ;-)
19:28:40pondlifeI'm not so fussy about FRs, more about bug reports...
19:28:44 Quit thegeek ()
19:28:46pondlifeGo ahead and close if you want though
19:28:59pondlifesaratoga: I'm here with an H300
19:30:04pondlifeBut I haven't copied any WMAs onto it yet :/
19:30:43*n1s loves large structs with no comments about what each value is for...
19:31:16linuxstbbluebrother: I don't think it's an unachievable feature - a lot of OFs play MPEG4. If we don't keep one request for it open, people will just keep requesting it...
19:31:54pondlifeTrue, it'll save repeat effort.
19:32:33*amiconn agrees with linuxstb
19:33:50amiconnsaratoga: The static inline function needs to be in a .h, *not* in aa .c
19:34:11amiconnThe way you did it it will no longer be used inline
19:34:39amiconn-> severe performance drop
19:35:21preglowsaratoga: it got folded into CMUL
19:35:34preglowsaratoga: will fold in for arm too, if no one beats me to it
19:35:45 Part Crash91
19:36:24amiconnsaratoga: Oh, and there are yellow builds
19:37:58pondlifeOh, and it's "Oops", not "Opps", ;)
19:39:48BiptoNdoes anyone know how to boot to iriver firmware
19:39:56linuxstbNew kids, not understanding the oops tradition...
19:39:59pondlifeH300, hold REC
19:40:08BiptoNi hold the rec button and hit play and it flashes that it's gonna load the irver firmware then turns off
19:40:21amiconnYou need to hold Play longer
19:40:21pondlifeHold play for longer
19:40:23perplexityyou need to keep the play button down longer then
19:40:33amiconnThe OF takes a while until it checks the power button
19:40:37perplexityGee, think that's a known issue ?
19:41:24BiptoNthanks i'll give it a try
19:42:02BiptoNthank you guys, I appreciate ya givin me the time :)
19:42:17pondlifeNo problem
19:42:28saratogaamiconn: the warnings in the yellow builds aren't from the wma decoder
19:42:33saratogaany idea about that?
19:42:47pondlifeSOURCES missing a CR/LF?
19:43:20pondlifeWhatever it is in *nixland
19:45:40lazkai have: "<stdin>:3:11: warning: no newline at end of file"
19:46:33linuxstbI'm guess SOURCES is missing a newline at end of file
19:46:36linuxstb(in libwma/)
19:47:39saratogalinuxstb: correct, thanks
19:47:39 Quit BiptoN ("CGI:IRC (EOF)")
19:47:56saratogai moved the coldfirm fixmul32 to the header file and made it static
19:48:03saratogaanything else before I commit?
19:50:10pondlifeAny more paranoia?
19:55:08 Join austriancoder [0] (n=austrian@
19:58:33saratogayes, all green
19:58:41 Quit saratoga ("CGI:IRC")
20:00:27linuxstbHas anyone tried WMA on a 64-bit sim? (I'm not confident...)
20:01:25 Join Rondom [0] (
20:04:00n1slinuxstb: I could do it if you point me to a wma file somewhere
20:05:03obolinuxstb: seems to work
20:05:20 Join My_Sic [0] (
20:06:02obolinuxstb: spoke a bit too soon - seg faults on track change
20:09:44*n1s listens to "Bangles - Walk like an egyptian" :)
20:10:06n1slinuxstb: metadata doesn't show up...
20:10:36peturn1s: known...
20:11:17n1sThe same bug for Tremor was because of nasty assumption that long is 32 bits...
20:11:31n1sas so many other fun 64 bit sim crashes....
20:12:18n1slinuxstb: track change wma->ogg works fine
20:12:59n1sand ogg->wma too
20:13:39 Quit freqmod (Remote closed the connection)
20:13:54 Join PaulJam [0] (
20:14:38 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
20:14:43peturaustriancoder: your patches don't apply cleanly, could you first update to current svn please?
20:15:22linuxstbobo: So it crashes at the end of the track for you?
20:15:25austriancoderpetur: one moment
20:15:35obolinuxstb: no, that was on a manual track change
20:16:43merbananlinuxstb: do you know why the Tremor mdct isn't used in wma ?
20:18:25austriancoderpetur: patch -p0 < mer.... works for me (tm)
20:19:12 Quit printfXh4 (Read error: 113 (No route to host))
20:19:36peturhmmm either my revert failed or something is wrong....
20:20:09linuxstbmerbanan: No. I've only been involved with the ASF parsing, preglow and saratoga know about the internals of the codec.
20:20:39peturaustriancoder: ok, something was wrong here, works now
20:21:15austriancoderpetur: also download e200tool source...
20:22:07 Quit ptw419 ()
20:22:36obolinuxstb: crashes on the end of tracks for me:
20:23:03preglowmerbanan: any reason it should be?
20:23:50amiconnjhMikeS: ping
20:23:52preglowonly efficiency-wise thing i know about the tremor mdct is that some people have been replacing it with an fft based one, which is what the ffmpeg one is
20:24:47n1slinuxstb: hmm, maybe my sim is not the latest svn... will retry
20:26:40linuxstbobo: Hmm., that's an odd place for it to segfault...
20:27:41austriancoderpetur: soo.. the problem is that i dont get a valid setup packet from the usb controller... I have tried it for some hours... with the two ways descriped in MCIMX31RM.pdf. Then I got the hint that e200tool uses usb and so I merged the e200tool usb stuff in the arcotg driver...
20:28:52*linuxstb goes home
20:29:03 Quit linuxstb ("CGI:IRC")
20:29:12preglowyou can't just take the e200tool stuff, it isn't licensed, afaik
20:29:49 Join RoC_MasterMind [0] (
20:30:01preglowwhich means everything is copyright mrh and you can't use any of it
20:30:16XavierGrsaratoga I have an H300 and H100 if you need me to test any wma progress on coldfire
20:30:22austriancoderpreglow: I know.. i used it only to find why I dont get setup packets... but i also dont get setup packets the way e200tool does it
20:30:26preglowbut i bet that might be solved if you make bagder ask him
20:30:51preglowwell, if you haven't taken it verbatim, there isn't a problem anyway
20:32:42n1slinuxstb: (for the logs) still no crash at end-of-track with latest svn...
20:32:50*austriancoder is not the best englisch speaker on this planet
20:33:05merbananpreglow: isn't the tremor mdct more optimized on embeded archs ?
20:33:06 Join Xerion [0] (
20:33:43preglowmerbanan: perhaps
20:34:03preglowworth a shot anyway, but not a priority
20:34:10preglowoptimising can always be done by other people
20:34:19 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
20:34:27merbananpreglow: anyway, this should be applied to the source
20:35:15merbananpreglow: and this
20:35:16preglowmerbanan: i think saratoga is in the process of merging ffmpeg svn to our source now
20:35:28preglowbut these are important?
20:35:58merbananwell the first one affects quality the second one affects speed quite much
20:36:23preglowi'll remind saratoga when he comes back
20:38:18merbananare prebuilt vlc and window tables on the TODO list ?
20:39:00preglowdon't think so
20:39:10preglowdoes that matter, though?
20:39:31preglowi think it's generated quite fast, but i don't know
20:40:16merbananwell if the tables are rebuilt on every played file I guess it could matter
20:40:26XavierGrwhere do we put aliases in linux? in .bashrc or .bash_profile? (on debian that is)
20:40:32preglowyeah, it is, so i guess we should fix that
20:41:05preglowbut again, stuff like fixing low-bitrate files and merging ffmpeg svn is more important
20:41:11merbananIMO thats the right thing to do on embeded platforms
20:41:13preglowperformance stuff actually comes last
20:41:17preglowyeah, i agree
20:41:46merbananbut functionality first
20:42:03preglowwell, at least know that it is almost realtime anyway
20:42:34merbanananother thing, the Tremor mdct might be more resistant to overflows, but I could be wrong
20:42:37amiconnpreglow: Will you do the arm CMUL soon?
20:43:15preglowamiconn: i really doubt it, i'm pretty busy the coming days
20:43:25preglowand i don't have my nano here to test
20:43:49preglowamiconn: btw, it's pretty much already done, like i said, XPROD31 in tremor requires only minor modification to work
20:44:05preglowperhaps XNPROD31 works as is
20:44:16preglowmerbanan: that might be interesting indeed
20:44:25amiconnXPROD31 loses one bit of precision
20:45:03amiconnNot sure whether that's important
20:45:08preglowmerbanan: anyway, i'll keep the tip in mind and investigate when i have time, if no one beats me to it (unlikely)
20:45:12preglowamiconn: it does?
20:45:19preglowamiconn: oh, yeah,
20:45:26 Quit borisyeltsin (Remote closed the connection)
20:45:31preglowwell, at a cost of two cycles you can fix that quite easily
20:45:37preglowbut you need to move the shifts to asm
20:45:46preglowprobably doesn't matter in the least
20:45:59 Join borisyeltsin [0] (
21:00:29 Join AceNik_ [0] (n=AceNik@
21:01:09AceNik_hey guys i tried syncing my .log scrobbler.log files wth o many links n softwares all fail to work , any suggestions?
21:01:49peturAceNik_: submission servers are down probably
21:02:10peturthey have been up and down like a jojo for almost 2 weeks
21:02:38AceNik_so should i preserve my .log files then n try again after a few days
21:02:47peturthey have a status page somewhere
21:03:41peturAceNik_: yes, just let it grow until it works again. use latest QTScrobbler, it uses protocol 1.2
21:03:50 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
21:04:08pondlifeobo: I commented on ...
21:04:22AceNik_can you provide me link to 1.2, i got only 0.5
21:04:32pondlifeobo: .. but if it works then it's surely an improvement.
21:04:42***Saving seen data "./dancer.seen"
21:04:53peturAceNik_: 0.5 is latest, I meant the protocol it uses is 1.2
21:06:00 Join Thundercloud [0] (
21:06:03AceNik_secondly i tried an "svn up" after a long time n it shows an error couldnt add sokoban.levels n stops, so i deleted the file n just made a ummy one , now it shows svn lockup
21:06:12AceNik_after cleanup it doesnt work svn up just doesnt respond
21:06:42obopondlife: I think it could be changed to pcm_paused = audio_status() & (AUDIO_STATUS_PLAY | AUDIO_STATUS_PAUSE)
21:12:32amiconnNow, let's see....
21:12:52 Quit RoC_MasterMind ("Leaving")
21:14:09preglowsee what?
21:14:22peturaustriancoder: do you have an idea which part is 100% working ok? I'm looking at dr_controller_setup() and I see MrH putting usleep() between stop/restart and other changes, where the arcotg code doesn't. If you take over his code and it doesn't work, it must be in the part that you didn't take...
21:14:48amiconnpreglow: Decoding speed with arm asm CMUL...
21:15:12austriancoderpetur: okay.. will try to add the usleeps
21:15:46 Join My_Sic [0] (
21:16:04 Join Siltaar [0] (
21:16:11amiconnpreglow: My test track went 122% -> 146% realtime :D
21:16:22amiconn(on mini G2)
21:16:48preglowholy shiznit
21:16:49amiconnThis is the 31bit precision version
21:16:57preglowthat's way, way above what i would have expected
21:17:24preglowno wonder CMUL asm helped so much for coldfire
21:18:11 Quit AceNik_ ("bye guys have fun , enjoy !!!!!")
21:18:17amiconnSo, do we want that extra bit or not?
21:18:59preglowi'd say ignore it
21:19:13preglowif tremor survives it, so will wma
21:21:14 Quit The-Compiler (Read error: 104 (Connection reset by peer))
21:21:25amiconnI'll put a comment in
21:26:07 Join The-Compiler [0] (
21:26:41 Quit Thundercloud (Remote closed the connection)
21:27:01 Join Thundercloud [0] (
21:28:48 Join webguest34 [0] (i=8e039814@gateway/web/cgi-irc/
21:29:05webguest34Can you start your ipod back into the normal os once you install Rockbox?
21:29:37XavierGrnotepad2's feature to keep file's original line endings (either windows or linux) is a double edged knife
21:29:40Mouser_XIt's called Dual Booting.
21:29:42 Quit BK (Read error: 104 (Connection reset by peer))
21:29:51Mouser_XRead the manual.
21:29:57Mouser_XI don't have an iPod.
21:30:00webguest34What page.
21:30:12Mouser_XIt would say... It's one of the basic features.
21:30:25webguest34I know I have a sansa, and it works great but im having difficulties with the ipod
21:30:30XavierGrI moved my alias from cygwin to linux and I wasted 30 minutes trying to figure what was to blame
21:31:05pixelmaI think the IpodFAQ in the wiki is quite nice too
21:31:09amiconnXavierGr: Oh, do you have cygwin set to dos line endings?
21:31:35XavierGrit seems so :\
21:31:56amiconnDefault is *nix line endings
21:32:02XavierGror else I don't know how on earth those dos line endings appeared
21:32:14Mouser_Xwebguest34: Check here -
21:32:37 Quit The-Compiler ("Connection reseted by god")
21:36:16webguest34nevermind, hold switch, dur.
21:36:17 Quit webguest34 ("CGI:IRC")
21:37:18peturXavierGr: did you use tortoisesvn with cygwin?
21:40:30 Join kismet [0] (
21:44:27XavierGrpetur: nope, strange thing is that these aliase were created in cygwin then imported on the wiki vmware image working fine. After that took them from there to import them to the new sdl-cross-compiled image, only to discover dos line endings
21:46:06Mouser_XSo, when booting, what's the "rescue mode" on the Gigabeat?
21:46:36austriancoderpetur: 100% works the detection if there was a setup packet recived on an endpoint
21:47:38austriancoderpetur: one moment.. I try something
21:53:09Mouser_XSo, in the last month, was the bootloader updated for the Gigabeat?
21:53:26 Join Wiwie [0] (
21:53:28Mouser_XExcluding installing a new bootloader, I think I've tried everything...
21:54:15bluebrothereasy way to check: check the timestamp of the bootloader download file ;-)
21:54:57gregjrockbox = BOS ?
21:54:59gregjwtf is BOS
21:55:26 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
21:55:31austriancoderpetur: soo.. the problem seems the memcpy. I can detect if an setup packet (usb request) was recived at an endpoint - via UDC_ENDPTSETUPSTAT. Then i copy setup_buffer from queue head to my local buffer. And here seems the problem.. usb controller does not update data in queue head
21:55:32XavierGr"bag of shit"
21:55:38Mouser_Xgregj: Check here:
21:56:00XavierGrcopyrighted by alexshigo :P
21:56:19linuxstbaustriancoder: cache issues?
21:56:37 Join Javacat [0] (n=ghsgh@unaffiliated/javacat)
21:56:50JavacatWhen can I expect a port of rockbox for the iphone?
21:56:52*Javacat hides
21:56:58Mouser_Xbluebrother: How about since April 1st? (My bootloader "FWIMG01.DAT" is dated as April 1st, 2007)
21:57:05gregjI like this one
21:57:29bluebrotherMouser_X: well, the download server shows 21st april as file date.
21:57:32austriancoderlinuxstb: MrH. does in his e200tool nothing with the cache - but should i do a cache flush?
21:57:42bluebrothercheck the listing here:
21:57:43jhMikeSamiconn: gnip
21:57:55linuxstbaustriancoder: Does he use IRAM? IRAM isn't cached.
21:58:08gregjand yes, compared to iriver FW, rockbox is BOS - quality wise, I am sorry - but that is true. on the other hand, feature wise (or whats the - bee-kul-when-feeneishead set of plugins)
21:58:22Mouser_Xbluebrother: It won't hurt to try... (Since it's currently not getting into Rockbox at all, it's certainly better than where I'm at now.)
21:58:25gregjiriver sux feature wise, and so does apple
21:58:32amiconnjhMikeS: Will you update your core locking patch? The one in the wiki produces failed hunks
21:58:35gregj(reason why i installed rockbox on my ipod)
21:58:58austriancoderlinuxstb: nope... no use of IRAM at the moment
21:59:03*bluebrother disagrees Rockbox being BOS quality-wise
21:59:08jhMikeSamiconn: with current svn? alot?
22:00:10gregjbluebrother: quality - battery lifetime (2-3x less than on orginal firmware0, quality - borked mpeg player, quality!
22:00:13jhMikeSI'll have a quick look. it's not really the approach I want to take but if it runs it's a step in the right direction anyway.
22:00:16gregjbluebrother: sound quality is okay
22:00:38bluebrotherbattery life is great on coldfire devices.
22:00:49 Join crop [0] (i=51d1d75d@gateway/web/cgi-irc/
22:01:15bluebrotherand even for that players it hasn't been released ...
22:01:59croppetur: hey
22:02:59croppetur: I think it was you who committed the patch for the context menu for FM in the main menu, right? I have a question
22:03:34*linuxstb should delete mpegplayer from gregj's builds...
22:03:45cropIf I'm not in the FM mode yet and call up the context menu there is an item "Add preset". It's not clear what would be added then
22:04:35cropCould we disable that item if not called from FM? Or would it be too much hassle?
22:04:50 Join TrueJournals [0] (
22:05:15Mouser_Xmpegplayer works fine for me.
22:05:16peturcrop: yes the context menu isn't 100% ok, maybe some items should be disabled, I agree
22:06:34croppetur: only two are not quite OK IMHO: "add preset" and "save preset list". Others are ok
22:06:43XavierGrgrepj: what do you actually like on the iriver FW? Except from the nice graphics on some screens it is brain dead and stupid
22:06:54XavierGrpetur are you sure you fixed the full array error?
22:06:59XavierGrI still get it on my h100
22:07:50peturgot maybe a modified viewer.config?
22:07:57 Part TrueJournals
22:08:05croppetur: but I assume the feature isn't used that often and is here just for consistency (since other menu entries have a context menu) −−> a good thing (consistency)
22:08:07bluebrotherI noticed that problem too
22:08:23cropehm... what problem? May I ask?
22:08:25 Quit Neovanglist (Connection reset by peer)
22:08:42linuxstbaustriancoder: Looking at e200tool, it looks to me as if it's run from IRAM - it's linked to run from address 0x40004000
22:08:48*n1s used h300firmware for about a year before rockbox was available and had to resort to tral-and-error everyime he had to fiddle with any of the menus...
22:08:52linuxstb(the arm_code.c contents)
22:09:27jhMikeSamiconn: test building ... should have it in a few minutes
22:09:40austriancoderlinuxstb: ah... will try something
22:10:29XavierGroh and gregj: iriver fw delivers less battery power than rockbox does on H300 and H100
22:11:29*crop goes back in the hole
22:11:31XavierGrquality wise rockbox is far better on all aspects of coldfire original fw (except USB OTG and WMA (which will be ready in a couple of weeks)
22:11:31 Part crop
22:12:24n1sXavierGr: although we have a fair bit of bugs and interface stuff to improve ;-)
22:12:24peturcrop: I also had doubts about the recording entries there...
22:12:55austriancoderlinuxstb: I only need to change e.g. int i; to int i IBSS_ATTR to have i in IRAM?
22:13:29XavierGrn1s: nonsense :P
22:13:42linuxstbIf it's a global variable, yes. If it's on the stack, then you can't control where it is (only where your stack is). If you're running code in the main thread, the stack is in IRAM.
22:13:51XavierGrmost bugs are on features that iriver doesn't provide anyway...
22:14:17bluebrothern1s: bugs? What is a bug? ;-)
22:16:12austriancoderlinuxstb: wow... it works now ;) Man.. you will get a beer if we meet us in real life
22:17:05XavierGrso what is working now?
22:17:42austriancoderI need to put the queue heads in IRAM.. seems to work.. will do some stress tests..
22:17:49peturI had read the remark about relocating to 0x40004000 but didn't know this was iram
22:18:11linuxstbpetur: Yes, IRAM is at 0x40000000 on portalplayer.
22:18:57peturaustriancoder: so now do the same with the original arcotg code? (irq based)
22:19:32*austriancoder also read the realocating, but did not connect it with IRAM
22:20:04*petur knows nothing about portalplayer :/
22:21:11austriancoderpetur: yep.. but will check now if i really recive what usb host in my pc sends..
22:21:58peturagreed, you can use this code to check everyting works and then move back to the original code
22:22:19gregjon symulator, where is it looking for 'filesystem' of the device ?
22:22:32linuxstbIn the "archos" directory.
22:22:53peturgregj: make install <−− will create it
22:25:40 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
22:27:18 Join Neovanglist [0] (
22:27:30gregjSegmentation fault (core dumped)
22:29:54amiconnlinuxstb: There is a problem with test_codec and voice
22:30:22gregjbut that might be because Eve Online is using sound exclusively on linux
22:31:40 Quit linuxstb ("CGI:IRC (Ping timeout)")
22:32:15austriancoderwith the help of usbmon on linux I can say that I get what usb host sends :)
22:32:28amiconnOn the h180 i get an exception or a freeze when returning from test_codec, depending on the format tested
22:32:50amiconnwma gave an IllInstr exception, ogg gave a freeze
22:32:53 Quit Xerion (" ")
22:32:57amiconn(nasty on h1x0...)
22:33:02linuxstb_amiconn: Hmm...
22:33:14n1samiconn: is it enough to just have a voice file present?
22:33:48amiconnI had voice enabled...
22:34:00linuxstb_Does the voice thread have its own stack?
22:34:29linuxstb_I mean, does it swap the codec's IRAM stack ?
22:34:36amiconnI guess this is a codec swap problem, i.e. the core expects the voice code to be there, but test_codec replaced it
22:34:52amiconnvoice has its own stack
22:34:54n1samiconn: I remember I got Illinstr errors after trying to resume playback after using test_codec, but they magically went away... maybe I had a voice file and deleted it...
22:35:12linuxstb_Ah, so voice was in the codec buffer before test_codec started?
22:35:23amiconnMost probably
22:35:26linuxstb_Meaning test_codec should save and restore the codec buffer?
22:35:38peturaustriancoder: congrats!
22:35:45 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
22:36:03amiconnlinuxstb_: Either that, or tell the core to do it
22:36:11linuxstb_Do you know how to fix?
22:36:12jhMikeStest_codec has to swap voice away just like playback does
22:36:28*linuxstb_ doesn't know anything about voice, and doesn't have voice files installed
22:36:46gregjvoice - the thing that comes out of your mouth when you spek
22:36:49jhMikeSif a voice file is present the voice codec is current in memory
22:36:50amiconnn1s: This could also happen without a voice file, but talk clips enabled & present
22:37:05amiconnOr does that not work?
22:40:38 Join ptw419 [0] (
22:41:27desowinaustriancoder: nice to see you're getting usb working :-)
22:41:48amiconnjhMikeS: Core locking patch still freezes at boot or a fraction of a second later :(
22:43:11*jhMikeS is just weirded out by that
22:43:40amiconnHmm, at the 8th try it managed to get the browser to display without freezing, and I was able to browse a bit, use the menu etc
22:43:46 Join stripwax [0] (
22:44:00amiconnBut then it froze again (in the thread display)
22:44:52jhMikeSThe current_core routines? maybe something's actually different there?
22:46:20amiconnBut there is a possible optimisation for the core check
22:47:08jhMikeSfor current_core?
22:47:19amiconnInstead of reading (*(volatile unsigned long *)(0x60000000)) and the converting to unsigned char, the OF just reads (*(volatile unsigned char *)(0x60000000))
22:47:55jhMikeSthe asm routine already just uses ldrb
22:47:56 Join TrueJournals [0] (
22:48:01 Join BK-- [0] (
22:52:02Mouser_XOkay. So I got Rockbox back to normal. Could anyone tell me what files I need if I want to restore my playlist? (I had to delete "/.rockbox" since I wasn't even booting for a bit there...)
22:52:27 Join scorche [0] (i=HydraIRC@rockbox/administrator/scorche)
22:52:27Mouser_X(I copied it though, so I could just dump it back on, but I'm cautious of doing so.)
22:52:48Mouser_XHowdy scorche. Busy?
22:53:04scorchedepends on your question =P
22:53:36Mouser_XMainly, what files do I need in order to restore my playlist after deleting /.rockbox?
22:53:41jhMikeSamiconn: is there any chance on these 5020s that any memory remaps are overlapping one another?
22:53:43Mouser_X(I have a copy of it.)
22:54:05Mouser_X(And restoring the playlist position would be great as well.)
22:54:11scorcheyou mean resume your playlist?
22:55:05 Quit borisyeltsin (Remote closed the connection)
22:55:20 Join borisyeltsin [0] (
22:55:35Mouser_XYes, I know how to resume it, but since I deleted .rockbox, I figured I'd need to restore something.
22:55:50bluebrotherhmm, .playlist_control maybe?
22:56:18scorchebluebrother: i seem to remember something about it being stored in a different i crazy?
22:56:21Mouser_XIs that all I need? I wasn't sure if there was anything else (I guess I could have tried it first, then asked...)
22:56:32bluebrotherdifferent partition? I don't think so.
22:56:36amiconnjhMikeS: I don't think so, but I can't be sure
22:56:47Mouser_XWell, settings and stuff were hidden, somehow, in an older version.
22:56:54bluebrothermaybe you're confusing it with the settings block we had earlier?
22:56:54Mouser_XThey changed that though.
22:57:08scorchebluebrother: perhaps
22:57:26*scorche just woke up to his computer being reset from power outage =/
22:58:07jhMikeSamiconn: linuxstb posted a memory map setup and it looked to me like it might. that remapping of flash ROM in system_init also bugs me a bit
22:58:22Mouser_Xscorche: I have a UPS, but it*very* rarely helps...
22:58:36scorcheMouser_X: it was out for a few hours
22:58:57 Quit n1s ("Leaving.")
22:59:10 Join DerPapst [0] (
22:59:34Mouser_XMy brother has/had a UPS that could last for 2 hours, including his router and internet stuff. I think it was 2.5 or 3 using bare minimum stuff...
22:59:59Mouser_X(I could also be remembering that wrong. It's been quite awhile since I saw it.)
23:00:09Mouser_XIt lasted at least an hour though.
23:02:02amiconnjhMikeS: why? It's the same for all PP502x targets...
23:02:24amiconnBut speaking about remapping - it looks like the OF does *not* remap sdram to 0x0
23:02:33jhMikeSis it? it seems to remap to already mapped regions on some devices
23:02:58amiconnBut it maps *something* to 0x0 - maybe a small part of sdram or iram (??)
23:04:43***Saving seen data "./dancer.seen"
23:04:59preglowunderstanding what the hell those mmap regs do would please me
23:05:28jhMikeSI don't know. I haven't analyzed it much but something looks suspicious enough to dig into to me.
23:06:09 Quit davina (Remote closed the connection)
23:06:19Mouser_Xscorche: Blast. Copying .playlist_control only restores the playlist. I was hoping to restore the point in the playlist last played as well...
23:06:40Mouser_XAlso, any idea why the WPS "Pixel" isn't working?
23:07:06bluebrothermaybe the position is in nvram.bin? (just guessing)
23:07:12Mouser_X(Do I need the album art patch, or something?)
23:07:42 Join stripwax_ [0] (
23:07:55amiconnbluebrother: You're guessing right (except on archos recorders where it's in the actual nvram)
23:09:16amiconnlinuxstb_: Strange that only coldfire targets had that warning...
23:09:25gregjlinuxstb_: heh, so the video works okay on simulator - must be problem somewhere else outside mpegplayer than
23:09:36gregjlinuxstb_: which makes it harder ..
23:10:23gregjlinuxstb_: but it s funny, simulator uses 30% of my 1.5ghz cpu :) compared to 2x80mhz in ipod...
23:12:04pixelmawell it's a simulator not an emulator...
23:12:21gregjI know how it owrks under the hood, I just said that it is funny
23:12:35 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
23:12:50linuxstb_amiconn: Indeed. I'm guessing it's a difference between gcc 3.4 and 4.0
23:12:51amiconndisplay output in the sim isn't optimised much, because it's not needed
23:12:56gregjit clearly shows tho, that some other parts of OS must be broken , or it doesn't simply work correctly on arm
23:13:47linuxstb_mpegplayer is the only part of Rockbox using both cores on PP targets, and that's still in its infancy.
23:13:52gregjor doesn't interract with mpegplayer
23:14:09gregjlinuxstb_: one for sound one for video ?
23:14:14gregjlinuxstb_: if so, it would explain a lot
23:14:36 Part TrueJournals
23:15:00 Quit stripwax (Read error: 110 (Connection timed out))
23:16:11 Quit merbanan (Remote closed the connection)
23:16:25linuxstb_gregj: One for video, one for everything else.
23:16:49 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
23:17:01gregjand because it is Preemptive OS - there are troubles
23:17:07gregjif you don't sleep often enough
23:17:24 Join stripwax [0] (
23:17:27desowin2x80mhz in ipod?
23:17:52gregjthat must be the shit with sound disapearing, and not being able to use controls anymore
23:17:59gregjso rockboxes OS sucks on SMP
23:18:33gregjdesowin: 2x78 on rockos, because guys have no idea(yet) how to set it to 80
23:18:40amiconnrockbox isn't a preemptive kernel
23:18:54amiconnAnd we know how to set 80MHz
23:18:56gregjcooperative - I'm sorry
23:19:03gregjamiconn: you set you don't know on the other day
23:19:21desowinthe other day cop wasn't used at all ...
23:19:22amiconnIt's just easier to program the pll with a fixed denominator
23:19:30XavierGrpetur: it was test_codec plugin that added one more line on viewers.config that I compiled to test wma
23:19:45Mouser_XIf you're using a WPS that has album art (I can see it in the *.wps file), but you're using the "current" build (which of course doesn't have album art), would that part of the WPS be ignored?
23:19:53amiconnAnd it's not SMP either
23:19:55Mouser_X(This WPS isn't loading for me.)
23:20:13gregjamiconn: there's so many things I don't understand than
23:20:20linuxstb_Mouser_X: The WPS parser would probably reject the WPS as having invalid tags.
23:20:25bluebrotherMouser_X: the wps will be ignored if it contains errors
23:20:26peturXavierGr: it wouldn't hurt to increase the array more with 1 or more entries...
23:20:26linuxstb_(which it sounds like it is doing)
23:20:29gregjamiconn: every time I think I understand something, you guys are crashing me to bone
23:20:40 Quit pondlife (Read error: 104 (Connection reset by peer))
23:20:42Mouser_Xlinuxstb_ and bluebrother: Thanks. I'll edit it out then.
23:20:43desowingregj: I got used to it :D
23:20:54gregjamiconn: tell me the truth than, or die :P
23:21:02gregjamiconn: so we have 2 cpus on ipod, nice
23:21:09gregjamiconn: and how the hell non smp system can use them :>
23:21:10amiconnSMP would mean that rockbox schedules threads to balance load
23:21:15peturgregj: every part of rockbox is being described by you as crap...
23:21:16*JdGordon ates 7am
23:21:18JdGordonmorning all
23:21:21dionoeathen not than ... bah
23:21:25dionoeahi JdGordon
23:21:30amiconn...which would kinda require a preemptive kernel afaiu
23:21:32gregjpetur: I never said crap
23:21:38stripwaxgregj - are you being constructive
23:21:44*dionoea understood it like petur
23:21:56scorchepetur: see topic
23:21:57amiconnBut in rockbox, threads are bound to one of the cores
23:21:57gregjdionoea: quote
23:21:59stripwaxrockbox sets the cpu affinity for each thread
23:21:59 Join ckani4002 [0] (
23:22:05 Quit ckani4002 (Client Quit)
23:22:19gregjamiconn: SMP is ability to run on more than one CPU, and handle them
23:22:25amiconnHence it's not symmetric
23:22:29gregjamiconn: up to which point, it depends on CPU
23:22:38gregjtfu, depends on OS
23:22:39amiconnNo, that's just MP (without the S)
23:22:59desowingregj: keep in mind you're talking with rockbox hackers
23:23:09gregjthere's problem somewhere, which causes mpegplayer to behave not quite alright
23:23:16stripwax'anyways'. so is gregj going to rewrite rockbox kernel to make it not crap?
23:23:28gregjeither sounds get funny, or silence - and no controls
23:23:41jhMikeSif we get this stupid multicore weirdness solved, I'll let threads hop cores if desired
23:23:43gregjand as linuxstb pointed out, sound runs on 'other' cpu/thread
23:24:11gregjstripwax: I never said it is a crap, god damnit!
23:24:14gregjquit it
23:24:26gregjI say, there's certainly a bug somewhere
23:24:26scorchegregj: it isnt is a bag of shit!
23:24:33dionoeawell you've been critising close to every aspect of rockbox since you joined the chan
23:24:34gregjand mpeg player exposes it
23:24:40gregjscorche: UE , quality wise, yes
23:24:58stripwaxgood, so maybe there's a bug and rockbox sucks at SMP. next step?
23:24:59bluebrothergregj: your wording isn't the nicest. Maybe you should take care to not get misunderstood here?
23:25:07gregjI always give a reason , it is not plain ranting
23:25:20preglowrockbox does currently suck at multicore, i think
23:25:25preglowunless slasheri has commited his stuff
23:25:52gregjbluebrother: true, but when I as trying to have a point - and worded it nicely, no one give a damn
23:26:18bluebrothergregj: definitely not.
23:26:36gregjbluebrother: I did at the begining
23:27:03bluebrotherwell, you got attention. If you rant-like go on people might start to ignore you
23:27:15gregjbluebrother: I think I just found a way to get you guys interested, and have conversation - I have to use that kind of wording to get anyone involced
23:27:22gregjdamn, I am knackered
23:27:23bluebrotherand you shouldn't forget that lots of the hardware isn't documented at all.
23:27:27desowingregj: have you read info what portalplayer is like?
23:27:44gregjdesowin: what you mean
23:27:57stripwaxgregj - there is NO developer documentation for the cpu
23:28:07gregjI know that
23:28:10desowinportalplayer is CPU used in some rockbox targets
23:28:13bluebrotheryou're mostly talking about some software being broken, but are you really aware how different it is programming a hardware you don't have documentation?
23:28:20gregjI know it is hard work
23:28:33bluebrotherit's a complete different thing than programming x86 for example
23:28:41gregjcan we get over that, I had to write few plugins for IM protocols, that have no docs
23:28:55stripwaxIM protocols can be sniffed on the wire
23:28:56linuxstb_gregj: What problems do you have with score.mpeg ? I'm watching it now on my ipod Photo, and it played to the end without any noticable problems.
23:29:00gregjit wasn't hardware, but I know what it is like
23:29:22gregjlinuxstb_: it must get exposed on my one than
23:29:31gregjlinuxstb_: want me to jump on a bus tomorrow, and show you ? :>
23:29:43bluebrotherwell, then I'm kinda surprised about your general tone in here. Devs are doing their best to make Rockbox better as you might imagine yourself
23:29:44stripwaxis it ipod *photo* only
23:30:03 Quit BK-- ("You can't fire me! I quit.")
23:30:03linuxstb_gregj: Just tell me exactly what to look for...
23:30:45gregjbluebrother: just pay more attentiion to what I say, and don't get me as 'another ranter', and I will be polite as lamb
23:30:52 Part Javacat
23:31:15jhMikeStwo things need doing for nice multicore support: 1) 5020 weirdness discovered 2) flushing specific cache ranges so keeping coherency is efficient
23:31:27gregjlinuxstb_: I am playing it, and sounds gets funny after 1-2 mins
23:31:28bluebrotherwell, it's not me doing loads of programming work.
23:31:48linuxstb_gregj: Is it always the same place?
23:31:49 Quit stripwax_ (Read error: 110 (Connection timed out))
23:31:50gregjlinuxstb_: than - silence - random timing (threading issue)
23:31:57linuxstb_I never get silence...
23:32:05*stripwax wonders who stripwax_ was
23:32:08bluebrotherbut you should really consider hackers starting to ignore you if you talk the way you did recently.
23:32:16gregjlinuxstb_: and when it gets silenced , I can only do menu+action (hard reset)
23:32:28gregjbluebrother: okay, thanks for pointing it out - I appriciate
23:32:35linuxstb_Have you changed the limit fps/skip frames settings from their defaults?
23:32:39scorchegregj: we call it select rather than action
23:33:16gregjlinuxstb_: I tried different settings
23:33:35gregjlinuxstb_: by default it limits fps and frame skipping is on - I belive
23:33:53gregjlinuxstb_: so right now I am on no fps limits, and frame skipping on
23:34:03gregjI'll try with fprs limiting, hold on
23:34:49linuxstb_Maybe I don't have enough of the file - did you upload it all?
23:35:10linuxstb_(I have just over 6MB)
23:35:12gregjwhole file is 400MB
23:35:17gregjit is 2h long concert
23:35:24linuxstb_Do you get the same problem playing the 6MB file you uploaded?
23:35:25gregjthat I would love to watch from time to time
23:35:41gregjonly the funny sound tho, no silence problem
23:35:43linuxstb_All I can say is that it works perfectly for me...
23:35:50gregjI can upload more, if you want
23:36:28gregjcan you try setting 'limit fps' to no than, and try again
23:36:32linuxstb_I doubt that would help, if you get the same problem with the 6MB sample.
23:36:44gregjit is random
23:36:45linuxstb_That's just a debugging option - it should really be removed.
23:36:51gregjie - doesn't alwyays happend in the same spot
23:37:13gregjso I'll run it with fps limit on
23:37:28gregjand will see, hold on - will give you feedback in few mins
23:37:38gregjall I can say, it works briliant on sim
23:37:50linuxstb_Yes, unless you want to benchmark video decoding for example, there's no reason to change the options.
23:37:55gregjalright, already hppend
23:38:02gregjwith fps limit on
23:38:23linuxstb_Is your filesystem clean?
23:38:45gregjI don't have camera here to show you, playing video - no sound, completly uinresponsive keys
23:38:51gregjwhat do you mean by clen ?
23:38:57linuxstb_Free of errors
23:39:00gregjI have about 15GB of music on it
23:39:09gregjI'll try fscking it
23:39:11gregjhold on
23:39:35gregjdoesn't respond to usb events either
23:39:47linuxstb_USB is software...
23:39:52gregjI know
23:39:59gregjhence I mentioned
23:40:16gregjokay, hard reset - will do fsck
23:40:39gregjanother question - how far are you guys from providing native usb disc thingie on ipod ?
23:40:50gregjwithout need to actually reboot it into diagnostic mode
23:41:00 Quit rotator ()
23:41:03scorchewe dont reboot into diagnostic...
23:41:04linuxstb_There's a google SoC project to implement it
23:41:12linuxstb_(I think...)
23:41:23*linuxstb_ isn't actually sure what the USB SoC project is aiming to do...
23:41:29*Nico_P neither
23:41:47scorchepetur is around, i believe...
23:42:33scorchethat should get him here =P
23:42:47gregjlinuxstb_: so it works the same way for every other player ?
23:44:16gregjlinuxstb_: I restarted, and while RB is still updating DB in background, playback was bit greenish (video)
23:44:35gregjlinuxstb_: it happends sometimes with broken mpeg, but next time I started mpeg player on the same file - it is alright
23:44:54gregjfilesystem clean, same problem happening again
23:44:56Nico_Pthe application doesn't say much...
23:45:12bluebrotherthere is a wiki page describing the goal of the USB project
23:45:19linuxstb_gregj: Does what work the same for every other player/
23:45:19bluebrotherdamn, too slow.
23:46:16gregjlinuxstb_: other player on rockbox ?
23:46:25bluebrotherwell, doesn't the application name the goal? It's kinda hidden in the "About" section
23:46:36linuxstb_gregj: I don't have the DB enabled, so that could be a difference. What happens if you wait for the DB update to finish before starting mpegplayer?
23:46:43 Join entheh [0] (
23:46:50gregjlinuxstb_: I always did before
23:47:11gregjlinuxstb_: it looks like mpegplayer didn't got all packets - but still is eager to play - with half empty buffers
23:47:13linuxstb_gregj: 23.42.47 # <gregj> linuxstb_: so it works the same way for every other player ? <−−- What does the "it" refer to?
23:47:17gregjlinuxstb_: (the green video thing)
23:47:31gregjlinuxstb_: usb, other player - a device
23:47:43gregjlinuxstb_: other boxes, targets, hardware, - mp3 players
23:47:49linuxstb_I know what "other player" means, I didn't know what "it" meant...
23:47:57gregjUSB stack
23:48:07linuxstb_And no, USB disk mode is in hardware on most targets.
23:48:17linuxstb_Only PP has a software USB disk mode.
23:48:23gregjI see
23:48:36linuxstb_And hardware USB is fully supported in Rockbox.
23:48:43bluebrotheradd USBOTG on h300 to the list of software implementation ;-)
23:48:47 Join My_Sic [0] (
23:49:05gregjcan I turn off DB altoughether ?
23:49:09 Quit My_Sic (Client Quit)
23:49:12linuxstb_On many devices (all Archos, iriver h100) Rockbox can run from flash and you can fully remove all traces of the original firmware.
23:49:18bluebrotherbut, strictly speaking, I wouldn't count OTG as USB mode :)
23:49:38linuxstb_(sorry, not all Archoses are flashable...)
23:50:47markungregj: you can comment out the #define HAVE_TAGCACHE in firmware/export/config-ipod??.h
23:50:56markun(tagcache is the old name of the database)
23:51:28*linuxstb_ thinks it's unfortunate that gregj's first impression of Rockbox was using the buggiest plugin (mpegplayer) on the buggiest variant (PP5020) of the buggiest series of players (ipods) that Rockbox runs on.
23:51:31markungregj: but you don't like ifdefs :)
23:51:43bluebrothermarkun: s/old/internal/ ?
23:51:51gregjmarkun: :P
23:52:00markungregj: what would have been a better way?
23:52:15gregjlinuxstb_: hey mate, I am willing to help
23:52:24gregjlinuxstb_: since not so many devs have ipod
23:52:28*petur returns
23:52:51gregjlinuxstb_: I just have to get familiar with the source, and this is hard - #ifdefs hell is one of those things that puts me off :)
23:52:53linuxstb_There are quite a few devs with ipods, but what's needed is people willing to do low-level hardware hacking.
23:53:06preglowand reverse engineering...
23:53:10*linuxstb_ likes #ifdefs
23:53:13JdGordongregj: let me tell you.... its alot better than it used to be
23:53:21gregjI belive you
23:53:26*linuxstb_ clarifies that with "a reasonable number"...
23:53:30stripwaxgregj/linuxstb - have you both tried running using the same settings file as each other? just a thought
23:53:45gregjstripwax: good point
23:53:49gregjwhere's the settings ?
23:53:53linuxstb_My ipod was recently reformatted, and is almost empty, with the default settings.
23:53:55gregjsomewhere in .rockbox/
23:54:19stripwaxgregj in that case could you just reset your settings to default, same as linuxstb
23:55:02linuxstb_petur: We were wondering what the aim of the USB SoC project was...
23:55:06*petur has been busy at work merging some parallel development trees using ifdefs and was quite happy with it
23:55:31peturfirst aim is providing a usb stack for device mode
23:55:55linuxstb_So a disk mode for PP targets?
23:56:09peturdisk and other
23:56:20gregjmy config
23:56:22linuxstb_perplexity: What others are there?
23:56:30peturaustriancoder was looking at serial
23:56:36linuxstb_s/perplexity/petur/ (sorry for waking you perplexity)
23:56:39*gregj removes .rockbox, and gets completly new one - instead of just replace
23:57:10stripwaxbootloader version?
23:57:36gregjme ?
23:57:49 Quit desowin ("use linux")
23:58:04stripwaxyep, is that same as linuxstb's bootloader version?
23:58:12linuxstb_I'm using v1.1. gregj: If you hold the "right" button when booting, you'll see the version
23:58:38 Part Domonoky_
23:58:44 Join TrueJournals [0] (
23:58:48gregjwhen booting rockbox ?
23:58:55gregjI just removed it, I have to reinstall - hold on

Previous day | Next day