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

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

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

#rockbox log for 2007-02-17

00:04:46 Quit Mouser_X (Read error: 110 (Connection timed out))
00:10:30preglowi see i can add more precision to the eq output
00:10:34preglowi wonder if i should bother
00:10:37 Part nls
00:16:10 Join whonicca [0] (
00:17:51 Quit Rondom ("Ex-Chat")
00:17:51whoniccaquestion guys, will rockbox render my ipod useless to my cars headunit and ipod adapter cdb-100
00:17:58toffewe have a strange problem on the gigabeat concerning the hard drive on the F10 and the F20, both hd are from toshiba and exactly same model mk2006gal but one is 10gb and the other one 20Gb, did you see this ont ipod 10gb and the 20gb ?
00:18:08Mikachuwhonicca: no, you can dualboot
00:18:51whoniccahmmm, interesting
00:19:07whoniccawill tracks i put via rockbox show up under ipod firmeware?
00:19:32Mikachuyou can play tracks in both if you continue to use itunes
00:20:10whoniccasucks, im on linunx
00:20:14whoniccathanks for ur help
00:20:22Mikachuwhonicca: um
00:20:29Mikachuwhonicca: how do you put tracks on then, gtkpod? that'll also work
00:20:40whoniccabeen trying gtkpod
00:20:55whoniccacant really get amarok working like i want it to, so yeah gtkpod
00:21:17Mikachuwhat i meant was that anything that plays in the ipod firmware will play in rockbox, except drm music
00:21:36 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
00:21:41whoniccai have no drm =)
00:21:48whoniccaand is that vise versa?
00:22:30whoniccanevermind u answered that a couple lines ago
00:22:46Mikachuno, rockbox supports more formats than ipods, so you can have music that only plays in rockbox
00:22:56Mikachuthis is all detailed on the web site
00:23:19whoniccaheh i only use mp3 format
00:23:50 Join combrains [0] (
00:25:14 Join blackness [0] (
00:26:17blacknessis there any way to get the output settings to display the screen through the ipod video cables
00:29:20 Part whonicca ("Leaving")
00:29:54 Join Wiwie [0] (
00:31:55 Join ender` [0] (
00:36:29 Quit lee-qid ("aufwiederbyebientotsayonara")
00:38:39 Join Winkleberry [0] (
00:40:06 Quit ender` (Read error: 104 (Connection reset by peer))
00:40:13Winkleberrymay i ask if formatting my iriver h140 will destroy the original firmware and rockbox?
00:40:30Mikachuyou may
00:40:43 Join ender` [0] (
00:40:46Winkleberryand might you answer?
00:41:17Mikachui don't know, but it sounds likely
00:41:37Winkleberryoh dear i don't know what to do then
00:42:34Winkleberrymy iriver shows a D symbol (with a line through it) with 0 folders and 0 files
00:42:55Winkleberryalthough on the pc it contains music
00:43:36 Join shodan- [0] (
00:44:00Winkleberryi thought formatting might help but I am afraid that if everything gets wiped the player will be dead.
00:45:14 Join printfXh4 [0] (
00:45:18shodan-hmm , this is more of a ipod question than a rockbox question but.. is it possible with the stock ipod firmware, that when you put the ipod in it's dock, prevent it to go into diskmode (and stopping the music) just like with rockbox when you put it in the dock you press "menu" and it doesn't go into diskmode ?
00:46:37Winkleberrythanks anyway mikachu bye bye
00:46:39 Quit Winkleberry ()
00:48:31Soapblackness: no
00:48:44Soapnot even Apple's firmware can do that.
00:51:22Mikachuwhat do you mean "not even", you can in rockbox?
00:51:50Mikachuand i think apple's firmware only goes into disk mode if the OS on the computer requests it (which both windows and linux does afaik)
00:54:12SoapMikachu: I was replying to blackness, not shodan-
00:54:23Mikachui'm sorry
00:54:35Soapsuch a bad bad boy! ;)
00:58:47shodan-Mikachu, my dock recharge via usb, is there a way I can tell the os (windows) not to request the ipod to go into diskmode then ? or do like rockbox does and press some key or key combo and it doesn't go into diskmode ?
00:59:10Mikachudon't think so, but if you do "safe eject" after, you can play music while it's still connected
00:59:15Mikachuit'll stop when you first insert it though
01:00:03Mikachui meant eject or safe removal
01:00:26 Quit Wiwie (Read error: 104 (Connection reset by peer))
01:02:08 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
01:02:21shodan-hmm, would have been great, it's annoying when I get home I'm listening to something, I put it in the dock, I get like 1 second of audio in my sound system, then it cuts and I have to start over :\
01:03:45 Quit barrywardell ("Konversation terminated!")
01:03:47 Join Landus [0] (
01:05:57blacknessI don't know what happened but all of my plugins dissapered, how do I get them back?
01:09:20 Join midgey [0] (
01:14:19***Saving seen data "./dancer.seen"
01:14:21 Join entheh [0] (
01:23:12 Quit matsl (Remote closed the connection)
01:24:37 Quit ender` (Read error: 104 (Connection reset by peer))
01:26:55 Join Brunellus [0] (n=luigi@unaffiliated/brunellus)
01:31:48 Quit toffe ("MegaIRC v3.27")
01:33:29 Quit midgey ()
01:34:25 Join Thundercloud [0] (n=thunderc@
01:37:29 Quit fasmaie (Remote closed the connection)
01:38:24 Join midgey [0] (
01:39:06 Join Mouser_X [0] (
01:39:23 Join barrywardell [0] (n=barrywar@
01:42:09 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
01:45:22 Quit Mouser_X (Nick collision from services.)
01:45:29 Join Mouser_X [0] (
01:45:32 Quit jhMikeS (Nick collision from services.)
01:45:38 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:47:42 Quit DataGhost ("NTOSKRNL.EXE caused a buffer overflow in System Idle Proce")
01:47:54linuxstb__barrywardell: Nice to see some usb commits, even if they don't do anything...
01:47:59 Nick linuxstb__ is now known as linuxstb (
01:50:17 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
01:51:10 Quit midgey ()
01:51:23 Quit ze (Read error: 104 (Connection reset by peer))
01:51:24 Quit redbreva (Read error: 110 (Connection timed out))
01:51:29 Quit Thundercloud__ (Read error: 110 (Connection timed out))
01:51:58 Join ze [0] (
01:52:07barrywardelllinuxstb: yeah, it's a start anyway. I've worked my way through a good bit of the code so far and could probably add some more stuff
01:52:26barrywardellyou might also want to try playing around with dr_controller_stop()
01:52:43barrywardellit stops the controller after you do dr_controller_run()
01:53:39 Quit Brunellus ("Ex-Chat")
01:56:05linuxstbYes, I tried that a week or so ago. Did you say you had problems with your Mac after dr_controller_setup/run ?
01:57:49barrywardelli remember having problems with the usb ports dieing until I rebooted
01:57:59barrywardellI've been without my mac for a couple of weeks though
01:58:59barrywardelllinuxstb: have you had any problems with that?
01:59:01linuxstbYes, the current USB detection seems to do that on my Mac as well. I tried adding a call to dr_controller_stop(), and that fixed the problem, but the detection also stopped working correctly... I never managed to work it out, but need to go back to it.
01:59:39barrywardelldr_controller_setup() only needs to be called once, on usb initialization
01:59:55barrywardell..._run() gets called on usb connect
02:00:03barrywardelland ..._stop() on disconnect
02:00:37 Quit Arathis ("Bye, bye")
02:01:15 Quit Mouser_X (Nick collision from services.)
02:01:24 Join Mouser_X [0] (
02:03:06linuxstbSo we could move dr_controller_setup() out of the usb detect function?
02:04:59barrywardellyeah, I think so. I haven't done so yet, but I think it could be moved into usb_init_device()
02:05:14linuxstbOK, I'll try that.
02:05:31linuxstbAnd call dr_controller_stop() when we detect a disconnect?
02:06:04barrywardellyes, I think so
02:08:50 Quit Xerion (Read error: 104 (Connection reset by peer))
02:08:56 Join Xerion_ [0] (
02:09:06 Nick Xerion_ is now known as Xerion (
02:11:06linuxstbOK, first test passed, USB connection to Linux detected successfully...
02:17:34 Join l0de [0] (
02:17:38l0deHail sirs
02:18:06l0deany news on the 80gb 5g? I heard there was some progress but there isn't much info in the thread
02:18:50Mikachunot done yet
02:19:07dan_al0de: The problem has been found, and now it's just a matter of solving it without breaking any other machines
02:20:11ThundercloudWhat exactly IS the problem anyway?
02:20:22l0deThat's what I wanted to know
02:20:32l0deWell I'm glad to hear there's some progress
02:21:43dan_aThe problem is something to do with the logical sector sizes on the 5.5G's 80Gb hard drive not matching the physical sectors, but I don't understand it properly
02:23:22linuxstbbarrywardell: That seems to have made my Mac happy...
02:23:30LandusOut of curiosity, does anyone know if the Rockbox devs are going to submit Rockbox as a Summer of Code mentor?
02:23:48barrywardelllinuxstb: great :)
02:24:12linuxstbbarrywardell: I connected and disconnected several times, and my Mac kept detecting it. I then installed a SVN build, and the first connect failed - it rebooted into disk mode and then immediately said "OK to disconnect".
02:25:13linuxstbAlthough I'm a little confused, because I can't see how the dr_controller_stop() function is being called... I forgot that we reboot to disk mode, so Rockbox will never detect a disconnect from USB.
02:25:44 Join [shodan] [0] (
02:25:53linuxstbSo unless I'm missing something, it seems the multiple calls to setup() were the problem.
02:26:05linuxstbLandus: Yes, there's been talk of it.
02:26:34linuxstbLandus: I think we need ideas for projects, and people to actually act as mentors...
02:28:13Mikachufixing the playback engine :)
02:28:17linuxstbThundercloud: IIUC, the problem is that the 80GB drive only allows reads (and presumably writes) of 2 sectors at a time, starting with an even-numbered sector. All other drives allow single sector reads/writes.
02:28:43Thundercloudlinuxstb: ... that sounds pretty stupid.
02:28:47barrywardelllinuxstb: I think it's hard to make much sense of the USB behaviour until we have a more complete driver
02:29:03linuxstbThundercloud: That was everyone's general reaction when it was discovered.
02:29:19ThundercloudWouldn't that like, severely impair function?
02:29:34barrywardellbut if it works, then great
02:30:01linuxstbbarrywardell: :) That's my general reaction with the ipods...
02:30:30ThundercloudWhat I both like and hate about the ipods is the amount of marketing and hype surrounding them.
02:30:38ThundercloudSamsung or any other company would be like:
02:30:45ThundercloudSleek... sensual... powerful... Z5
02:30:55ThundercloudWith lots of mysterious blackouts and fade-ins
02:31:00ThundercloudApple just goes
02:32:05 Quit Mouser_X (Nick collision from services.)
02:32:14 Join Mouser_X [0] (
02:32:35Shaid"Buy an iPod. All your friends will hate you and you'll be uncool and never get laid, ever, if you don't!"
02:34:21linuxstbbarrywardell: What are your plans for USB? Do you have much time to work on it?
02:35:07Landuslinuxstb: Pick out small, but useful projects, like battery optimizations, or a specific catergory of bugs.
02:35:34 Quit digen (Read error: 110 (Connection timed out))
02:36:36dan_alinuxstb: If I can shave a few bytes from the Archos builds, but have to make the code a bit uglier to do it, should I go for it?
02:37:34barrywardelllinuxstb: I have spent quite a bit of time looking through it
02:37:37barrywardelland done some work
02:38:37linuxstbdan_a: I guess it depends how ugly, and how many bytes...
02:38:38 Join dpassen1 [0] (
02:39:08barrywardellI was looking at this usb stack
02:39:50 Quit l0de ()
02:41:01dan_alinuxstb: I don't know how many bytes yet. Basically it's using a #define to define something (the processor number) either as a variable or as a literal number depending on whether it is being compiled for a dual core target or not.
02:41:31dan_a(I've had a rethink of the COP API)
02:42:57 Quit shodan- (Read error: 113 (No route to host))
02:44:46 Quit Mouser_X (Nick collision from services.)
02:44:55 Join Mouser_X [0] (
02:47:59linuxstbdan_a: Have you seen how the multi-volume support works (the IF_MV() macro) ? Maybe something like that could be used?
02:51:23barrywardelllinuxstb: do you know anything about the linux usb stack? If we could use that in Rockbox, then using the existing driver should be quite easy
02:51:54linuxstbNo, nothing at all.
02:51:55dan_alinuxstb: I'm using that in one place (I'm getting rid of create_thread_on_core, and changing create_thread() to take the core as an argument when we're compiled for dual core
02:52:32linuxstbdan_a: That makes sense.
02:52:47 Join Xerion_ [0] (
02:53:00 Quit Xerion (Read error: 104 (Connection reset by peer))
02:53:06 Nick Xerion_ is now known as Xerion (
02:53:41linuxstbdan_a: If you post a new patch, could you include the API change in mpegplayer and the plugin API?
02:54:40dan_alinuxstb: I will do. I probably won't get it done tonight - I only sat down to start it at 1:30
02:59:11 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
03:00:31 Quit Thundercloud (Remote closed the connection)
03:02:58 Quit wooo (Read error: 110 (Connection timed out))
03:10:33 Quit SaLoMoN (Excess Flood)
03:10:47 Join SaLoMoN [0] (
03:11:13 Quit SaLoMoN (Excess Flood)
03:13:47 Join SaLoMoN [0] (
03:14:24***Saving seen data "./dancer.seen"
03:20:47 Part barrywardell
03:22:28 Quit entheh ("^~")
03:22:28 Quit combrains (Read error: 104 (Connection reset by peer))
03:29:11linuxstbdan_a: I've just been reading the email of open patches in flyspray, and noticed your COP patch is 200 days old today :)
03:29:49dan_aIt's grown up a bit in that time
03:31:48 Join toffe [0] (n=toffe@
03:36:41 Join miepchen^schlaf [0] (
03:36:48 Quit Mouser_X (Nick collision from services.)
03:36:57 Join Mouser_X [0] (
03:37:02 Part pixelma
03:51:40 Quit Mouser_X (Read error: 60 (Operation timed out))
03:56:59Soapdan_a: thank you for the addition to IpodAccessories.
03:58:57dan_aSoap: No problem. Having the Bose might inspire me to try reverse engineering the protocol. Once dual core support is in SVN, of course!
03:59:19 Join fasmaie [0] (
03:59:52linuxstbAny ideas about the 4g bootloader problem?
04:01:10dan_aNone yet. I don't understand why it's hitting the greyscales and not the colours.
04:01:40 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:06:32 Quit Vyrus ("Leaving")
04:07:05*dan_a goes to bed
04:07:21*linuxstb goes to bed
04:14:04 Join saratoga [0] (i=98039a42@gateway/web/cgi-irc/
04:18:26 Join Mouser_X [0] (
04:24:00 Quit saratoga ("CGI:IRC (Ping timeout)")
04:24:23 Quit dpassen1 (Read error: 110 (Connection timed out))
04:35:27 Quit Mouser_X (Nick collision from services.)
04:35:36 Join Mouser_X [0] (
04:39:12 Join dpassen1 [0] (
04:41:32 Join Atoma [0] (
04:41:50 Nick Atoma is now known as atoma12 (
04:42:18atoma12little question I had, is there any existing (or in development) nes emulators for rockbox&
04:44:21 Quit fasmaie (Remote closed the connection)
04:56:21Soapexisting? no.
04:56:36atoma12ok, and in development?
04:57:14SoapI just looked in the patch tracker and didn't see one.
04:57:22atoma12I see..
04:57:50atoma12I must look if there's any updates to rockboy
04:59:41atoma12well.. 2 processors? each running at 80mhz?
04:59:48atoma12wouldn't a nes emulator be possible?
05:00:05atoma12I mean I ran nes games perfectly on my old p1 running at 100mhz
05:00:15atoma12with 16mb of ram
05:00:40atoma12(i know I know, not the same hardware)
05:04:04LloreanNot even remotely.
05:04:21LloreanReally, it's hard to predict what will be possible.
05:04:27atoma12I see
05:04:31LloreanIn essence, someone would have to try it, spend time optimizing it, and see what happens.
05:04:56atoma12well first they should try optimising rockboy
05:05:06Llorean'They' who?
05:05:22LloreanIf someone made an NES emulator, it'd be some random person who knew how, and wanted to.
05:05:31atoma12... everyone :P else than me.. i'm no programmer
05:06:11atoma12because I really want a nes emulator for rockbox... ipodlinux is too much trouble
05:06:39atoma12I'll try to learn c++ and try to optimize rockboy :P
05:07:03atoma12and then distribute it to everyone, like any respectable open-source programmer
05:08:05atoma12.. anyways
05:14:28***Saving seen data "./dancer.seen"
05:14:36 Quit atoma12 ("Leaving")
05:16:28SoapIf I was dictator of the world, all registered on the wiki would be forced to fill out IrcNicks
05:17:19Soap(IRC and Forum and Filespray nicks would also be required fields to fill in on the user's wiki page as well)
05:29:58 Quit Mouser_X (Nick collision from services.)
05:30:07 Join Mouser_X [0] (
05:36:42 Quit relaxed ("")
05:46:00 Quit Mouser_X (Nick collision from services.)
05:46:09 Join Mouser_X [0] (
05:50:23 Join SirFunk [0] (
05:51:09 Join miepchen^schlaf [0] (
05:56:56LloreanThis... doesn't look right:
06:06:20 Join midgey [0] (
06:07:09rotatorLlorean: why not?
06:07:44Lloreanrotator: Because as far as I'm aware, Jeff has nothing to do with outside of hosting the forums?
06:07:54rotatorIIRC, Jeff registered the domain then gave it to us
06:08:15LloreanBut why would that make him the current contact info?
06:08:33rotatordunno, i guess it was never changed since
06:08:39LloreanAh, okay.
06:08:49 Quit Nibbier (Read error: 60 (Operation timed out))
06:08:56rotatormaybe it has to wait until the domain get renewed?
06:09:12rotatori don't know how these things work very well
06:10:16Lloreanrotator: As far as I can tell, all you have to do there is type in the new info, and then give them an email address to reply to to verify you actually control it.
06:11:01rotatorthen that sounds like something to take up with the swedes :p
06:11:47LloreanSpeaking of taking things up to the swedes, I was gonna suggest to them that we maybe set up a email address so maybe we can have a contact point for people writing articles.
06:12:21rotatorsounds like a good idea, then maybe articles could be corrected *before* they are published
06:12:46rotatori know you've sent your share of corrections :p
06:13:37LloreanI _think_ a letter from me is showing up in next month's Linux Magazine. Well, this month's, except it doesn't seem to be in any stores near me.
06:14:14LloreanOr maybe next months, I guess.
06:14:26rotatoryeah, that article was fun
06:14:37rotatormost of it was installation instructions which are now outdated
06:15:00LloreanI understand outdated instructions, especially since the article was actually older, and translated from the German edition of their magazine.
06:15:09rotatorah, okay
06:15:32LloreanBut the warning about how Rockbox removes part of the Apple firmware, and the bit about it flashing the ROM on the iPod, and various other things like that, which are both very untrue and I think somewhat dangerously misleading were an irritation.
06:15:59rotatoryeah definitely
06:16:22rotatorit also irks me how all these articles are just about the ipod
06:16:41rotatorwhich is probably the least well functioning of all the supported ports
06:16:42LloreanOr at least mainly concentrate on it, yes.
06:16:58LloreanIt makes Rockbox look a lot less complete than it is.
06:17:16 Join Nibbier [0] (
06:17:29rotatorbut the ipod is the most popular player, so what can you do? *sigh*...
06:18:12LloreanI also think we should set some 'goals' for 3.0 sometime soon. Rather than starting a freeze, we should set some targets for 3.0, and 'conditions' that we want to meet before starting a freeze.
06:18:38LloreanThat way even as we go about normal business, there's specific details hanging over everyone's head saying "Hey, these still need to get done if we ever want to have a 3.0 release"
06:19:51rotatori seem to recall some draft of 3.0 release goals on the wiki sometime last year
06:19:57LloreanDuring the freeze.
06:20:14LloreanIt mostly boiled down to "Solve the H300 battery issue, and fix the playback engine, and get TagCache working"
06:20:51LloreanBasically, a feature freeze with its primary goals including "solve an unknown problem", "rewrite a core piece of code" and "add a new feature", all really things that shouldn't have happened during the freeze.
06:21:01LloreanThey should've all been on a list of things to finish before we can enter the freeze.
06:21:52rotatorit seems like a release for the h1xx and h3xx at the very least could be done relativly easily
06:22:24rotatori don't think there are any major show stopping bugs or lacking features left there
06:22:33LloreanNow that recording seems stable again.
06:22:44LloreanThe only lacking features, I believe, are USBOTG for H300, and WMA for both.
06:22:52LloreanBoth of which I think aren't release-critical in this case.
06:23:04midgeywe still have some playback issues
06:23:12Lloreanmidgey: Yeah
06:23:26midgeyglad to see the voice boosting bug was fixed
06:23:32Lloreanmidgey: That's kinda what I had in mind for being in the pre-freeze list
06:23:50LloreanThings like the problems relating to skipping and seeking near the beginning or end of a song.
06:24:23LloreanI don't really think a feature freeze should involve any 'major' bugs.
06:24:28midgeypressing stop on resume causes issues, as oes deleting files that exist in the playlist
06:24:39LloreanSince those tend to be the sorts of bugs that are hard to fix.
06:24:42midgeynor do i, i think we made a mistake last time
06:24:58midgeyespecially since most devs just waited it out
06:25:06LloreanThat's why I'd really like to see a pre-freeze checklist get put together. Say now "We want to have a freeze, but we don't know when. This needs to be done first."
06:25:09LloreanThen it's hanging over everyone's heads.
06:25:16LloreanMaybe get a little self-conscious guilt working on our side. ;)
06:26:01midgeyso Mayday could come a year late :)
06:26:32LloreanI'd like to see a 3.0 happen though.
06:26:36LloreanThe big SWCODEC release.
06:26:48LloreanI think actually that with some polishing, Gigabeat's almost ready to be included in a release as well.
06:26:58midgeyhow's the x5 port?
06:27:04LloreanAnd if Dual Boot isn't deemed 'release critical' X5 could be as well
06:27:11*rotator ordered a gigabeat f40 recently :)
06:27:24*midgey is hoping to get one as a birthday gift
06:27:42LloreanThe person who was working on managed a hack to get the X5 dual booting and gave up on trying for a 'proper' solution (one that didn't involve bits of retail code that might change with firmware updates)
06:28:05midgeywas that eli?
06:28:10LloreanRaenye I think.
06:28:23LloreanI'm not wholly sure any more, it happened a decent amount of time ago.
06:28:39midgeythat sounds right
06:30:43LloreanBut I believe that dual boot is the only area in which X5 is behind H300. I think.
06:33:04LloreanAnd I think the Gigabeat needs to have its LCD Power Off function fixed, and the code for its Button LEDs fixed, and I think it's all working as intended.
06:34:53midgeyi know boosting was disabled since it was causing worse battery life, is it possible that the implemented method was wrong?
06:35:00scorcheisnt there some work on the cross that needs to be done still?
06:35:09Lloreanscorche: The cross really works fine.
06:35:09midgeyit seems strange to me that scaling could hurt battery
06:35:39Lloreanmidgey: Dunno, honestly. But the battery life is comparable to the retail firmware already I believe (I don't have the retail firmware installed to test with, so I only speak from hear-say)
06:36:13scorcheLlorean: something about sensitivity?
06:36:17midgeyaren't there touch features for the cross that aren't enabled
06:36:23Lloreanscorche: A better method for the cross could be implemented, but I personally can come up with no complaints on the current one.
06:36:46Lloreanmidgey: The retail firmware has gesture-like actions (moving your finger from left, across the center, to right) that perform specific tasks.
06:37:25Lloreanscorche: Right now, the cross works pretty much like a four-way directional pad. You press any of the four directions, like a button, and it does things. Plus the center is a 5th button. Sensitivity is about where I'd want it, personally
06:37:43Lloreanscorche: The improvement would be that it can be made pressure sensitive. Press harder on down, it scrolls faster, for example.
06:38:16 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
06:38:42 Quit Mouser_X (Nick collision from services.)
06:38:51 Join Mouser_X [0] (
06:38:53 Join miepchen^schlaf [0] (
07:03:38 Quit Mouser_X (Nick collision from services.)
07:03:47 Join Mouser_X [0] (
07:04:32 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
07:05:32 Join miepchen^schlaf [0] (
07:13:49 Quit rotator ()
07:14:31***Saving seen data "./dancer.seen"
07:18:07 Quit dpassen1 (Read error: 110 (Connection timed out))
07:20:50 Join tchan1 [0] (
07:22:57 Quit funky (Read error: 110 (Connection timed out))
07:33:17 Quit tchan (Read error: 110 (Connection timed out))
07:36:02 Quit Mouser_X (Read error: 110 (Connection timed out))
07:43:02 Join Mouser_X [0] (
08:04:22 Quit Mouser_X (Nick collision from services.)
08:04:31 Join Mouser_X [0] (
08:11:18 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
08:12:37 Join miepchen^schlaf [0] (
08:20:47 Quit tipi^ (Read error: 110 (Connection timed out))
08:20:57 Join redbreva [0] (
08:25:15 Quit toffe ("MegaIRC v3.27")
08:30:19 Join MonkeyTamer [0] (n=chatzill@
08:32:41 Quit Xerion (Read error: 104 (Connection reset by peer))
08:32:48 Join Xerion_ [0] (
08:38:10 Quit Mouser_X (Nick collision from services.)
08:38:19 Join Mouser_X [0] (
08:41:38 Quit MonkeyTamer ("Chatzilla 0.9.77 [Firefox]")
08:44:44 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
08:45:15 Quit thegeek ()
08:45:46 Join thegeek [0] (
08:49:25 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
08:49:45 Join miepchen^schlaf [0] (
08:50:29 Join Rob2222 [0] (
08:58:51 Quit linuxstb (Read error: 110 (Connection timed out))
09:07:59 Quit tchan1 (Remote closed the connection)
09:08:00 Quit Rob222241 (Read error: 110 (Connection timed out))
09:08:32hcsargh, all this &quote; in the wiki viewer is getting to me
09:08:50 Join tchan1 [0] (
09:14:32***Saving seen data "./dancer.seen"
09:18:49hcstime to start handling some entities
09:30:07 Part Llorean
09:43:08 Quit datachild (Read error: 60 (Operation timed out))
09:44:00 Join datachild [0] (
09:45:14 Join kubiix [0] (
09:55:24 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
09:57:15amiconnSlasheri: 153 red size points for you
10:00:28 Join desowin [0] (
10:04:31 Join linuxstb__ [0] (
10:08:31 Join relaxed [0] (n=relaxed@unaffiliated/relaxed)
10:17:45 Quit linuxstb_ (Read error: 110 (Connection timed out))
10:21:41 Quit linuxstb__ (Read error: 104 (Connection reset by peer))
10:21:51 Join linuxstb__ [0] (
10:22:10 Join kaaloo [0] (
10:24:14Slasheriamiconn: yeah, no doubt that would increase the code size
10:24:55 Quit Mouser_X (Read error: 110 (Connection timed out))
10:26:10 Join Rob222241 [0] (
10:28:39amiconnI'm looking for testers with arm targets (ipods, gigabeat, sansa, h10).
10:29:11 Quit Rob2222 (Read error: 54 (Connection reset by peer))
10:29:19hcsooh, what's up
10:29:54desowinamiconn: testing what ?
10:29:59amiconnMy patch gets rid of function sections, reducing binary sizes of core, plugins and codecs
10:30:36amiconnI want to know whether it breaks any currently working functionality (it shouldn't) and whether there are no build errors
10:30:49amiconnI tested building myself for ipod mini, video, gigabeat, and e200">
10:31:28amiconnReconfiguration is required
10:31:47amiconn(the toplevel Makefile needs a change)
10:32:03hcsok, I'll check it out
10:32:12desowinI'll test it on mini
10:32:36amiconnMini g2 seems to work, that's what I have
10:33:16hcsI'll use it for my subsequent ipod buildings
10:33:24amiconnOne special area to test would be the encoders, and the disk access on sansa
10:36:29amiconnBinary size will go down >10KB for the core
10:40:03 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
10:40:28 Join miepchen^schlaf [0] (
10:41:29midkaywow, *someone's* out to get the highscore for the binary size table :)
10:42:10pixelmamidkay: did you read the logs of yesterday? Re: chopper bugs?
10:42:48midkayi didn't read very much, let me have a look.
10:43:36midkayblack-on-black text?
10:44:26pixelmayes - and the H1x0-Problems and the too long 'splash' on Archos screens. The other seems to be a different bug
10:46:31 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
10:47:05midkaygot the one about the splash... h1x0 problems?
10:49:03pixelmalook at what BigBambi had to say
10:49:51pixelmastarting at 15:19
10:51:45midkayoh, alright. (i think these must have been existing problems, my color changes were only for color LCDs) but i'll try and do a quick fix.
10:52:42amiconnStrange character translation in linuxstb's commit msg...
10:53:03desowinoutside ascii
10:53:25pixelmamidkay: thought that was somehow related... but thanks for looking into it :)
10:53:56 Quit linuxstb__ (Read error: 113 (No route to host))
10:54:35midkaypixelma: no problem, it should be a pretty easy fix.. i found the code a little complicated though, tons of different color changes all over the place.. that's why i didn't bother trying to double-check grayscale stuff. :) i'll do a sim-test first.
10:55:16amiconndesowin: Yes, but that's the translation of one letter!?
10:55:44desowinunciode chars can take up to 4 bytes
10:55:52amiconn(and iirc cvs had no problem with extended chars)
10:56:51amiconnTrue, maybe linuxstb used utf-8
10:57:00amiconnThen it would be 2 bytes
10:57:29 Join Lear [0] (i=chatzill@rockbox/developer/lear)
10:58:33pixelmamidkay: I guess the reason why it crashes on Archos (at least on my Ondio) when starting a new game is a bit more tricky...
10:58:42 Join Ribs [0] (n=ribs@
10:58:56midkaypixelma: ah, that's not good :)
10:59:49pixelmastrange thing: it works in the sim although it's so fast that it's almost unplayable
11:00:03 Quit bk ("Leaving")
11:00:19midkayif i can reproduce that on my Recorder then i can have a poke around and see if there's an obvious fix.. (h1x0 problems reproducible here, so i'll fix that now.)
11:00:53pixelmanice :)
11:04:34amiconnReproducable on recorder here
11:05:08 Join Llorean [0] (
11:06:50LearInteresting... -fstrength-reduce causes the same kind of problems as does newer versions of GCC (on an H1x0)...
11:12:45midkaypixelma: alright, i fixed the color bug (game over/press select text, and invisible menu) as well as the string-too-long-for-recorder bug (using a splash now)..
11:13:14amiconnOn X5 the choice of buttons in chopper is weird
11:13:33amiconnI'd expect Up would move the heli up, but instead it's Select...
11:13:49midkayh1x0 uses up *and* select, i guess i can do that on the x5 too..
11:14:02amiconnAh, yes
11:14:14midkayanything else?
11:14:34***Saving seen data "./dancer.seen"
11:15:15amiconnHmm. The speed is very target dependent
11:15:22pixelmathe Ondio button mapping is weird too
11:15:33amiconnOn ipod mini it's much harder than on H300
11:15:45amiconnVery fast (and the mini also suffers from invisible text)
11:15:52midkaypixelma: suggestions?
11:16:04amiconnI guess that applies to all greyscale targets
11:16:20midkayamiconn: invisible text should now be totally fixed (all gray targets, right).
11:16:24pixelmamy suggestion: "up" (and/or "Mode") for action, "off" to quit
11:17:36 Quit bun-bun ()
11:17:40 Join Thundercloud [0] (n=thunderc@
11:18:56midkaypixelma: also done.
11:19:38pixelmathanks! :)
11:20:50midkayno problem, *commits* :)
11:25:45 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
11:29:26 Join petur [0] (n=petur@rockbox/developer/petur)
11:31:05 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
11:34:01midkayhey JdGordon?
11:34:49midkayoh, just a sec.
11:34:50 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
11:35:17midkayk, never mind. :)
11:35:23JdGordonhehe ok :()
11:35:27linuxstbpixelma: Does Chopper still crash on the Ondio?
11:35:30JdGordonthat was meant to be :)
11:35:31midkay[convert file view menu!]
11:36:54JdGordonhmm? i thought iv done that one
11:37:50 Join perpleXa [0] (i=perplexa@unaffiliated/perplexa)
11:37:59pixelmalinuxstb: I tried your suggested change and it didn't help if that's what you mean. Haven't tried the new version but I think the commit doesn
11:38:14midkayJdGordon: it doesn't seem like you have..
11:38:45linuxstbpixelma: Do you know if anyone (amiconn?) has tried it on other Archos targets?
11:38:57LearAny opinions about adding the -fstrength-reduce optimization (at least on ColdFire)?
11:38:59amiconnYes. Same crash on recorder
11:39:16LearOn ColdFire, the code generated for FOR_NB_SCREENS isn't exactly pretty... :)
11:39:17linuxstbOK, I'll find some batteries and try it on my recorder.
11:39:38 Quit perpleXa (Client Quit)
11:40:35pixelmaJdGordon: did you see my complaint about the "Rockbox info" and "Runtime" screens - I think it has to do with your icon changes
11:40:49linuxstbSpeaking about Chopper, what do ipod users think of making a touch on the clickwheel act as fly, instead of pressing SELECT?
11:41:07*petur just bought an MK8007GAH for his h340
11:41:24Mikachulinuxstb: i thought that too yesterday
11:41:34JdGordonpixelma: I did, and rpleied... unless i go out tonight ill fix it
11:41:46linuxstbMikachu: I would have expected you to :)
11:41:54Mikachunow you know
11:41:57amiconnI wonder whether it's worth going through all the source and put ICODE_ATTR to all function definitions where the declarations have it, and put it on all declarations in .h files when it's an external function
11:42:10amiconnThis would allow us to get rid of -mlong-call on arm
11:42:43amiconnICODE_ATTR would then be __attribute__((section(".icode"),long_call)) , making only iram functions long call
11:42:59pixelmaJdGordon: ok, saw the reply now :)
11:43:13amiconnBut this can't be done without mentioned changes, because then gcc complains about conflicting types
11:43:17Mikachuyou don't have to make a long call when returning from an iram function?
11:43:46amiconn...seems gcc sees the __attributte__((long_call)) as part of the function type
11:43:56amiconnMikachu: return != call
11:44:13Mikachuwhat i meant was if an iram function calls a non-iram function
11:44:29amiconnThe only problem would arise if an iram function calls a non-iram function, but that wouldn't make sense
11:44:40*JdGordon 's lang strings are all fubar!
11:44:44Mikachuah, that's true
11:45:07 Join bun-bun [0] (
11:45:13dan_aamiconn: In that case, lets do it and if anything breaks then it was a bug in the first place
11:45:23Mikachuare long calls slower or just bigger code?
11:45:40 Quit Ribs ("eh eh ehhhh!")
11:45:45LearIn codecs it might...
11:45:46dan_aamiconn: Is there anything specific you want testing with your function-sections patch?
11:45:49amiconndan_a: I don't expect anything to break, the question is whether it's worth it. Potentially hundreds of places to change
11:45:53 Join Ribs [0] (n=ribs@
11:46:05amiconnMikachu: Bigger and slower
11:46:39amiconndan_a: Especially e200 disk access, and recording aiff, wav, wavpack, mp3
11:46:41LearE.g., Tremor has a few cases where inline functions calls a (bigger) function once in a while. Don't know if that inlined function might end up in IRAM:ed code.
11:46:45linuxstbIf it gets rid of unneeded long-calls, then I think it's worth it.
11:47:14amiconn(these are the files I had to change wrt 'static' qualifier without being able to test myself - mini g2 cannot record)
11:47:54dan_aSet up a Wiki page, and each volunteer can do a few changes - unless it can be sed scripted
11:48:32amiconnlinuxstb: My arm-sections patch is a start. It gets rid of long calls to static functions (unstatic'ing a few ones where necessary), and also decreases binary size due to getting rid of all those tiny sections which need alignment
11:48:49amiconnWould be nice if you could test on your video (iirc the video can record?)
11:48:55JdGordondoing puts(0,y,str); seems to put the text not at x=0... is this probably becuase of the xmargin or offset setting for the screen?
11:49:34amiconndan_a: Automatic editing impossible, I'd say
11:49:37linuxstbamiconn: Sure. dcc it to linuxstb_ What needs testing?
11:50:11amiconn[10:31:03] <amiconn>">
11:50:22amiconn[11:46:26] <amiconn> dan_a: Especially e200 disk access, and recording aiff, wav, wavpack, mp3
11:50:32midkayJdGordon: i'd try setmargins to 0,0... but you will need to set them back after you leave (oldx=get_xmargin and oldy=get_ymargin, i think, and then setmargins(oldx,oldy) when you leave).
11:50:35amiconnThe latter is true for all arm targets which can record
11:51:03midkaywell good luck with that. i'm going to bed.. back in a number of hours :)
11:51:17hcsamiconn: no problem with flac or spc decoding or wiki vieweing
11:51:24JdGordonthanks midkay
11:51:47hcsamiconn: though I know that's not what you need tested...
11:51:57JdGordonthe list sets the margin correctly, so i dont need to reset them :)
11:52:09midkaynp... hey, check screens.c's function for the quickscreen if you get stuck, it has to set and re-set margins itself.
11:52:16midkayoh, nice, alright :)
11:52:29*midkay is off, night.
11:54:25amiconnhcs: It's also good to know that general behaviour seems ok
11:54:28JdGordonpixelma: all fixed :)
11:54:37 Quit linuxstb ("CGI:IRC")
11:54:43amiconnlinuxstb: Btw, you need to reconfigure in order to get the desired effect
11:56:32pixelmaJdGordon: sounds good :)
11:56:45 Quit Tom0473 ("[G]Script - v3.3 daGroove - -")
11:56:49 Join Tom0473 [0] (
11:57:56dan_aamiconn: Disk access works on the Sansa - I'm just adding the test_disk plugin to make sure transfer times haven't changed
11:58:28JdGordondoh! that commit is going to break the player :p
11:59:14pixelmawill you ever learn :P
11:59:36JdGordonneva :D
12:01:14JdGordonBagder: another idea for the binary deltas... could we have a delta for the past 2 weeks of commits (or less..)?
12:04:13pixelmaJdGordon: who's the payer? ;)
12:04:51JdGordon? did i typo in the commit message?
12:04:58 Join Mouser_X [0] (
12:05:15*JdGordon blames it on the 40C temp today and the air-con dieing
12:08:06 Join lee-qid [0] (
12:08:26LearYou should move here, only 6C here. :)
12:08:50JdGordonna, way to cold
12:09:12 Quit kaaloo (Read error: 110 (Connection timed out))
12:09:56JdGordonanyone want to test the first lot of display menu conversion?
12:10:08crwlit's very hot here considering the time of the year, over 1 C...
12:10:23JdGordonhaha.. oh noes! heat wave!
12:11:01pixelmaJdGordon: I'd like to test
12:12:11linuxstb_amiconn: I've just tried recording with all four encoders with your arm-sections.diff (on my 5g) and they all worked fine.
12:12:42amiconngoodie :)
12:12:52*amiconn wants to commit that stuff
12:13:10*amiconn wants a green high score in the size table ;)
12:13:29dan_aamiconn: Sansa disk access is fine
12:14:17SlasheriJdGordon: +1C here and not cold at all :)
12:14:59Slasheriah, crwl was faster ;)
12:16:19JdGordonwe should get a svn highscore page going in the wiki for both tables :)
12:17:06pixelmahmm... 2 hunks failed
12:18:01JdGordonwhich? probably no big deal
12:18:19pixelmaapps/menus/settings_menu.c at 214; apps/menus/main_menu.c at 95
12:19:00pixelma241 I mean
12:20:16JdGordonumm... I think you can ignore them
12:20:45JdGordonunless they stop the compile
12:21:14JdGordonthey didnt come us as conflict when i did a svn up.. so they should be ok
12:21:41Mikachuyou probably already committed them :)
12:22:18JdGordonthen they shouldnt have come up as fails for the patch? no?
12:22:32Mikachuwell, patch will say the hunk is reverted
12:22:34JdGordonand I dont tihnk so... working from 2 different trees here
12:22:50Mikachui recognize the one in settings_menu, it's the sleep timer icon thing
12:23:11amiconnBagder: seems to be down
12:23:22Mikachuit works here
12:23:22 Quit JdGordon (Remote closed the connection)
12:24:15 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
12:24:25JdGordonbah, stupid gaim crashed
12:24:30amiconnHmm, looks like a momentary glitch looked really strange. No images and no css
12:24:58dan_aamiconn: MP3 recording is OK on the 4G greyscale - do you need the others testing since linuxstb has already done them on the video?
12:25:15amiconnNo. If it works on one, it should work for all
12:25:52amiconnIt's just that I wanted to make sure the unstatic'ing didn't break the encoders somehow
12:26:08pixelmareverted the patch, did a "svn up" again and patched again, now getting only the second fail (though I did the up before the first try too)... oh well
12:26:25 Join digen [0] (n=digen@unaffiliated/digen)
12:26:52amiconnSometimes there are strange effects one wouldn't expect. E.g. why does an rockbox mini2g binary crash with stkov main when compiled with -O2 ?
12:27:39amiconnHah, nice green values in the size table :)
12:28:54Mikachuthat's a lot of function calls
12:28:54JdGordonwhat did they do to rombox? or nothing?
12:29:06amiconnNothing. This is arm only
12:29:06Learamiconn: Maybe less keen on cleaning up stack used for function calls (and batch them up in one big "restore")?
12:29:56*JdGordon thinks someone should tell amiconn we only really care about bin size on the arcos targets.... :D
12:30:36amiconnNah, this commit doesn't only improve binary size. It also makes things go (a bit) faster
12:30:50preglowooooh, nice
12:30:58dan_aJdGordon: Yeah, lets see if we can get the core size on ARM all the way up to 32Mb!
12:30:58amiconnHmm, and it should even allow the profiling parser to work for arm
12:31:05amiconn(no function sections anymore)
12:31:24Mikachuamiconn: how much do you think no -mlong-calls would save?
12:31:26pixelmaJdGordon: don't know if it's important but I got a warning "settings_menu.c:682: warning: 'lcd_settings_menu' defined but not used". Maybe this is because of the failed hunk?
12:31:27amiconndan_a: The iFP is also arm - and it only has 1MB RAM (!)
12:31:50 Join Rondom [0] (
12:32:06amiconn..where core, plugins, codecs and the buffer have to fit
12:32:16 Join entheh [0] (
12:32:37dan_aThat's... a challenge!
12:33:10amiconnMikachu: Not sure, maybe another 10KB
12:33:21JdGordonpixelma: its because im not finished yet....
12:33:23preglowi think it'd save even more
12:33:34 Join BHSPitMonkey_ [0] (
12:33:41Mikachuwould it compile if you just removed it? just to see how much it saves
12:33:54amiconndan_a: tomal took that challenge, and iiuc he already managed to get some codecs running
12:33:58pixelmaJdGordon: ok, I guess I can ignore it then
12:34:06amiconnCodecs are limited to 128KB and plugins to 64KB on iFP
12:34:20*JdGordon is anoyed, he could have ot a ifp7... for $37 but didnt
12:34:22 Quit Ribs ("eh eh ehhhh!")
12:34:40amiconnMikachu: It doesn't compile, or rather, it doesn't link
12:35:33amiconnHmm, one could get an upper limit estimation for the size decrease when compiling with ICODE_ATTR being empty, i.e. no functions going into iram
12:38:29preglowi think linuxstb did that a while ago, and i'm pretty sure the saving were more than 25kb combined
12:42:52Shaidlooks on the wiki for an abbreviations page
12:42:59JdGordoniriver flash player
12:43:15 Join bluey- [0] (
12:43:54preglowprobably our currently most limited platform
12:43:59preglowthat is, swcodec platform
12:45:32*amiconn tests
12:45:54 Join funky [0] (n=repulse@unaffiliated/funky)
12:47:34pixelmaJdGordon: mmm, the LCD settings submenu is empty but I guess that is also because it isn't done yet...
12:48:00 Quit BHSPitLappy (Connection timed out)
12:48:35 Quit BHSPitMonkey (Connection timed out)
12:48:41JdGordonpixelma: yes, and peak meter... but everything else should work
12:51:15pixelmathe "show icons" changes aren't behaving right - how can I say - they are applied only when I return to the display settings menu
12:51:58JdGordonyeah, I know... not 100% sure how to fix that
12:52:27JdGordonI might have to put a special case check for that one
12:53:41amiconnpreglow: -50720 bytes for ipod video
12:53:51 Quit blithe (SendQ exceeded)
12:54:09 Join perpleXa [0] (i=perpleXa@unaffiliated/perplexa)
12:54:17dan_aI'd say that's worth doing then!
12:54:28JdGordonare the units in the set_int screen supposed to be translatable?
12:54:44pixelmaJdGordon: I'll test further during the day
12:54:45 Join Obsys [0] (
12:54:47 Quit perpleXa (Client Quit)
12:55:07JdGordonpixelma: cheers :)
12:57:17amiconnThe real gain will be lower; this test build has all functions short_call (and some features wouldn't work correctly, obviously rolo)
12:59:19amiconn...50KB just for long function calls...
13:01:48preglowamiconn: yeah, that's what i thought i remembered
13:02:01preglowamiconn: arm code isn't so notorisly uncompact after all
13:02:44dan_aI'd still like to do some work on getting some of the arm code to compile as thumb code
13:03:28preglowwell, idealistically, all code apart from performance sensitive stuff should be thumb
13:03:31preglowthat's how it's meant to be used
13:04:31preglowat least everything but the dsp code, that needs smull/smlal here and there
13:04:41preglowi don't really think it'll matter much for us
13:05:18preglowand i think we can expect plenty of resistance from gcc in the matter
13:05:57dan_aI've found trying to interwork thumb and non thumb to be painful with GCC
13:07:01amiconnpreglow: For sure.
13:08:25 Quit Mouser_X (Nick collision from services.)
13:08:33 Join Mouser_X [0] (
13:08:43JdGordonas someone done a naughty commit to english.lang recently? my langs are all wierd
13:09:06 Join jba [0] (
13:10:27JdGordon.. it would also help if i add the correct menu item :p
13:11:40 Quit jba (Client Quit)
13:13:46LearHm.. USB charging string was incorrectly added, but that was a few weeks ago.
13:14:37***Saving seen data "./dancer.seen"
13:15:41 Join wooo [0] (
13:15:49preglowi've tried doing thumb builds myself, and gave up quickly
13:15:55pregloweven after writing some thumb code
13:16:01preglowbut gcc was just plain unreasonable
13:19:42dan_aYay! I've got a way to improve the dual-core API which saves nearly 100 bytes on the Archos Player!
13:22:56Shaidwhat's the thumb?
13:23:39 Quit Mouser_X (Nick collision from services.)
13:23:50 Join Mouser_X [0] (
13:24:38dan_aShaid: It's an instruction set for ARM processors (like in the iPod.) It can make smaller, though possibly slower, code
13:28:08Shaidoh, right.
13:29:48dan_aSmaller code means that there's more memory available to cache the disk contents, which means the hard drive doesn't need to spin up as much, which means better battery life. It also means that less code needs to be transferred from memory, so it might be faster
13:35:36 Quit jhMikeS (Nick collision from services.)
13:35:42 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
13:37:40 Join thegeek_ [0] (
13:39:28 Quit Mouser_X (Nick collision from services.)
13:39:37 Join Mouser_X [0] (
13:48:07preglowi've been able to boost the precision of the eq, and it comes at a cost of around 3 extra cycles per sample for arm, and around 5-6 for coldfire
13:48:18preglowworth commiting, or?
13:49:53LearHow much is that, relatively speaking?
13:53:03 Quit thegeek (Read error: 110 (Connection timed out))
13:55:36preglowfor the arm, almost nothing
13:55:40preglowfor coldfire, a little bit
13:56:11preglowi think the coldfire one is a bit over 20 cycles in total now
13:56:34 Join Winchester [0] (
13:59:00Learyou mean, after the changes?
14:03:45JdGordonanyone know where HAS_REMOTE_BUTTON_HOLD is defined?
14:04:55LearHrm, fairly big increase then. And how much is the precision improved?
14:05:16Learjdgordon: target config header?
14:05:58JdGordonits not in config-h300.h like i would expect.. but the code is putting up an error which is only possible if that is defined
14:06:17JdGordonmissing initializer :'(
14:06:39 Quit bluey- ("Leaving")
14:06:51preglowLear: depends
14:08:46preglowLear: the lower bits aren't just thrown away as before, with shelving filters you'll get around six bits extra precision
14:08:51preglowwith peaking, around three
14:09:26linuxstb_JdGordon: I expect it's in button-target.h for the h300.
14:10:01LearSounds like a big gain, unless it only affects 1 or 2 bits in the final result... :)
14:10:02preglowso you might not get full 24 bits of audio if we ever support that :)
14:10:34JdGordonlinuxstb_: hmm... so I have to include button.h to get it?
14:11:05JdGordonah :D cheers
14:12:12Learpreglow: you mean that for 16-bit output you don't gain anything at all?
14:12:19preglowLear: i doubt it
14:13:23LearThen I'm not sure if it's worth it... ;)
14:13:57preglowi'll do some tests one day, right now i doubt it's worth it myself
14:14:59linuxstb_Could you include it with a #ifdef - in case we do ever support 24-bit output?
14:15:19preglowlinuxstb_: yeah, eaasily
14:15:23preglowi was planning on that
14:16:58 Join jellisii [0] (i=4ae6865c@gateway/web/cgi-irc/
14:27:01 Join kaaloo [0] (
14:27:35 Join tipi^ [0] (
14:33:10 Quit petur ("later")
14:39:20 Quit Mouser_X (Read error: 110 (Connection timed out))
14:41:02*linuxstb_ shouldn't try to run an fmrecorder build on a v1 recorder...
14:42:50 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
14:43:14Mikachuthere is also #pragma no_long_calls, long_calls_off and long_calls
14:51:12 Join fasmaie [0] (
14:55:32 Quit scorche (Read error: 60 (Operation timed out))
14:56:48 Quit pixelma (Read error: 110 (Connection timed out))
14:58:21 Quit himitsu_ (Read error: 111 (Connection refused))
14:59:09*JdGordon once again wished for the player build to die!
15:00:23preglowLear: not bad, not bad
15:05:26w1ll14mgoodmorning all
15:05:57 Join l33ch [0] (i=9ec3654a@gateway/web/cgi-irc/
15:06:24l33chhi all
15:06:55 Join Corsac [0] (
15:06:58Corsachi there
15:07:46w1ll14ml33xh, corsac: hi
15:07:53 Quit kaaloo ("Leaving.")
15:08:07w1ll14mfinaly it's weekend
15:08:23l33chfinaly after yesterday party
15:08:26JdGordonits been the weekend here for 25 hours already... damn wierdos!
15:08:31preglowhow important is updating copyright years in files?
15:08:45w1ll14mJdGordon: lol :)
15:08:47Corsaci've got a question about rockbox playback. is there a way to have "by album" random ?
15:08:51JdGordonless important that fixing bugs.. :p
15:08:59Corsacsomething like a random by folder but using album tags ?
15:09:07JdGordonCorsac: not yet
15:09:14 Join himitsu [0] (n=himitsu@
15:09:21JdGordoni want to do that using the database, but dont really know how to
15:09:31Corsacis it tricky ?
15:10:05l33chConvert the Display menu to the new system. :)
15:10:05JdGordonnot if you know how to use the database in code
15:10:31l33chI can test it soon ;)
15:10:48JdGordonyou can test it now.. all the builds are finished compiling
15:11:03Corsacme ?
15:11:08l33chme ;)
15:11:12Corsacha :)
15:11:36preglowwhat's up with the huge size differences among the ipod targets?
15:12:23SlasheriJdGordon: using the tagcache api should be very simple, at least doing basic searches and stuff like that
15:12:29preglowBagder: good point
15:12:29JdGordonamiconn: check the build page... we have green again :D
15:12:42preglowoh, ondio fm
15:12:44JdGordonSlasheri: yeah, but its not documented at all is it?
15:12:44preglowit's back
15:13:03*preglow pats JdGordon on the back
15:13:08SlasheriJdGordon: not yet, but tagcache.h and tagtree.c should give some examples how to use it
15:13:09*JdGordon kicks everyones ass (except amiconns) in the delta table :D
15:13:16amiconnJdGordon: Yeah, one less red build, but rombox isn't back yet for this target
15:13:37JdGordonhaha just say
15:13:42JdGordonstill 2500 bytes over
15:13:42BagderJdGordon: well, you managed to get a gapless green there, amiconn failed that ;-)
15:13:42amiconnIt's just no longer so much oversize that the linker complains
15:14:01preglowit's the colour that coutns!
15:14:09JdGordonyeah :D
15:14:20JdGordonwhatever coutns is :D
15:14:21amiconnMove in the right direction I'd say
15:14:36preglowBagder: but yeah, about updating copyright years when updating files, important?
15:14:39***Saving seen data "./dancer.seen"
15:14:42preglowi tend to forget it
15:14:56amiconnpreglow: What size differences you're referring to? Between models?
15:14:57BagderI think we should do that, but we basically never do it
15:15:09preglowamiconn: yes, and it's probably images
15:15:33Bagder(add-hook 'write-file-hooks 'copyright-update)
15:15:36amiconnYes, for sure it's the images, plus some feature differences
15:15:47amiconn(e.g. recording)
15:15:47Bagderemacs just rocks ;-)
15:16:13JdGordonis there a reason the (c)'s are always to a person and not Rockbox ?
15:16:21BagderJdGordon: several
15:16:30Bagder1) Rockbox is not an org that _can_ have a copyright
15:16:58Bagder2) we don't want to bother with signing over copyright from persons to something else
15:17:09JdGordonoh ok
15:17:48JdGordonbut its silly having the copyright messages.. after a few months the origional coder doesnt really have much code that is untouched by someone else anyway
15:18:14Bagderwell, the original coder still has the copyright if the work is his/hers and only modified
15:18:15 Join powr-toc [0] (
15:18:22Bagderuntil of course some magic limit is crossed
15:22:30Corsachmhm, sometimes themes are displayed kind of weirdly (like not aligned with the screen or something like that). but i'm using an experimental/patched build so maybe it's that :)
15:27:15linuxstbamiconn: Can you try chopper on your mini at some point? I'm wondering if that will also have problems.
15:28:00amiconnI already did, and it works
15:28:17amiconnIt's just rather hard to play, because it's faaast
15:28:43linuxstbI think that's easily fixed - there's a "cycletime" constant, that we just need to make depend on LCD_WIDTH.
15:29:52 Join dan_a_ [0] (i=dan_a@
15:30:41pregloware people allergic to automatic frame rate limiting?
15:30:42JdGordonBagder: whats the last column in the deltas table? average?
15:33:10 Join thegeek [0] (
15:33:11JdGordonwoot,... 2420 on first go in chooper :)
15:34:35 Join Arathis [0] (
15:36:35 Nick tchan1 is now known as tchan (
15:36:36l33chi have scrore 1273 :-(
15:41:23 Quit dan_a_ ()
15:43:00 Join redwood [0] (
15:45:19Bagdersomeone should wash Linus Torvald's mouth with soap
15:45:25 Quit miepchen^schlaf ("Verlassend")
15:45:41 Join miepchen^schlaf [0] (
15:46:11Mikachuor s's?
15:46:40Bagderhm, s' I believe isn't it?
15:46:56Bagdertricky english
15:47:19linuxstbWhat's he said?
15:47:44BagderI just think he's getting worse and worse, this time I read his recent gnome attack
15:48:17LearHe did answer with some code though. :)
15:48:22Bagderyes he did
15:48:55 Quit thegeek_ (Read error: 110 (Connection timed out))
15:48:57BagderLear: I just think his language is unnecessary offensive
15:49:37JdGordondid he submit patches he knew full-well wouldnt be accepted just to "proove" how right he was about them?
15:49:59Bagderthe latter
15:50:10BagderI mean, it was to prove a point I think
15:50:27JdGordonthe point being gnome thinks users are idiots?
15:50:42Bagderno, I think it was to show that the fix was wasy
15:50:46Bagderor something
15:50:49JdGordonah ok
15:51:42BagderI'm not really raising my eyebrows because of the actual topic
15:52:13JdGordonim reading the artivle now
15:53:36JdGordonhehe kvetching :) someone watched the nanny :p
15:54:10Corsacwould you recommend a them, btw ?
15:55:13linuxstbAnyone know what I09:CPUAdrEr means on the Archos?
15:57:24 Join thegeek_ [0] (
15:59:51 Quit mathgl (Read error: 104 (Connection reset by peer))
16:05:11Juice^what is the big differences on the new menu system?
16:06:41JdGordon(depends who you ask), cleaner and simpler code
16:06:52JdGordonor horrible messy and unintelligable code
16:06:57XavierGrwell, I have to agree with Linus Torvalds that GNOME is very limiting, but yeah I find his way of expressing it a bit too aggressive
16:07:35Corsacmenu system in rockbox or in gnome ?
16:07:58JdGordonCorsac: I was talking about rockbox.. dunno about Juice^ :p
16:08:03Corsacok :p
16:08:49Juice^JdGordon: yes, talking about rockbox.. is there any visual changes except for the new icons? (which is great)
16:09:23 Join ``0_0 [0] (
16:09:40``0_0hi is the 5g ipod 80gb avaliable?
16:10:02dan_aBagder: IIRC, Linus Torvalds has written a management document which includes the advice that being controversial often gets very good results.
16:11:15Corsac"Sim" builds are for emulators ?
16:11:23 Quit funky ("leaving")
16:11:35JdGordonJuice^: nope
16:11:36 Join funky [0] (n=repulse@unaffiliated/funky)
16:11:53dan_aCorsac: Simulators, not emulators (they look like the target, but don't emulate the hardware)
16:12:16Juice^great for making wps and backdrops etc
16:12:33Juice^to quickly test it
16:12:50 Quit thegeek (Read error: 110 (Connection timed out))
16:13:21Corsacmpf, even with a fresh installed rockbox current build I have those weird themes not aligned
16:14:01ShaidMy only device which rockbox has a practically dead screen, atleast you see something
16:14:19ShaidActually, the speaking menus is the reason I installed rockbox originally.
16:14:36linuxstbCorsac: Which theme, and which version of Rockbox (an official or unofficial build)?
16:14:51Corsaci just installed
16:15:07w1ll14mhow is this possible, i can run my ipod 5g 30gb at 105 Mhz (for testing) and my ipod 5g 60Gb only max 90Mhz (also for testing)
16:15:11Corsacwith theme Unicatcher, for instance
16:15:23linuxstbCorsac: Did you install the fonts zip file?
16:15:24 Join thegeek [0] (
16:15:32Corsacmaybe not
16:15:41Corsacno, guess not, as I've purged my previous install
16:15:45Corsacthanks, i'll try that
16:16:21linuxstbw1ll14m: They're different devices... The 60GB possibly has less space inside for cooling, or a warmer hard disk, or a warmer battery, or ....
16:16:48w1ll14mlinuxstb: i guess you are right
16:17:02w1ll14mbut then, the ipod 5g 30GB is a lot smaller
16:17:13w1ll14mi ca't get it, but whatever :)
16:17:16linuxstbBut has a smaller battery and smaller drive...
16:17:25w1ll14mlinuxstb:that's true
16:18:04Corsaclinuxstb: thanks, a lot better :)
16:18:06linuxstbOr maybe it's simply that the PP chip in your 60GB is lower quality than the one in your 30GB...
16:18:20Corsaclinuxstb: would you recommend a theme (and maybe an experimental build if the theme needs some additional patches) ?
16:18:29w1ll14mlinuxstb: why would they do that ?
16:19:27linuxstbSome batches will be higher quality than others IIUC - just how chip manufacturing works. We don't know at what speed the PP5021 is officially rated, which could just be 90MHz anyway.
16:20:00 Join Brunellus [0] (n=luigi@unaffiliated/brunellus)
16:20:26 Quit JdGordon (Remote closed the connection)
16:20:40*Mikachu notices one of the red staples on the build table is shorter
16:23:43w1ll14mlinuxstb: i thought it was 80MHz damn, i would love to have a pp5022 in my ipod 5g :) woohoo
16:24:26 Join nls [0] (
16:25:36linuxstbYou could probably sell it, buy a gigabeat, and still have money left...
16:26:07w1ll14mlinuxstb: hmmmm ... sounds intresting ...
16:26:36 Quit Obsys (Read error: 60 (Operation timed out))
16:27:19w1ll14mnice those menu icons!!!!
16:30:09Corsaci'm quite lost with the themes :)
16:31:46 Quit thegeek_ (Read error: 110 (Connection timed out))
16:32:20SoapCorsac: are you still having text alignment issues?
16:32:33CorsacSoap: not with pure themes, it seems
16:32:45Soap"pure themes" means what?
16:32:59Corsacthose shipped with standard builds
16:33:26SoapI strongly suspect you haven't downloaded the font package.
16:34:05Corsaci have
16:34:12Soaphave you installed it?
16:34:19*Corsac shrugs
16:34:37 Join Domonoky [0] (
16:34:42Corsacbut i'm trying themes which I'm not sure need patchs or not :)
16:34:54CorsacHipod CA for exemple, I guess it requires an album art patch or something like that :)
16:34:57Soapwhere are you getting your themes?
16:35:24Soapwell, it should be relatively easy to determine from the screenshot if a theme needs the album art patch.
16:35:26Corsachmhm, I downloaded a theme pack from mysticriver
16:35:49Corsaci'm currently looking for a good looking theme, that's all :)
16:36:22Soapthen all bets are off because all the ones I've seen there assume you are running their latest "Super-Duper-Patch-AllButTheKitchenSink" builds.
16:36:57Corsacyeah :/
16:37:11 Join Obsys [0] (
16:37:14Corsac there
16:39:10*dan_a makes a recording with the codec thread running on the COP
16:39:10 Quit desowin (Read error: 104 (Connection reset by peer))
16:39:39 Join desowin [0] (
16:40:08CorsacSoap: that's why I asked for a nice looking theme :)
16:40:13 Join thegeek_ [0] (
16:40:28Corsaci'm not against experimental builds, but I don't really know about theme
16:40:30SoapAll the one on that page /should/ work except FrostBoxAA
16:41:02 Join redbreva_ [0] (
16:43:22Shaidthat arctic desert one is nice
16:47:48 Quit redbreva (Read error: 110 (Connection timed out))
16:47:49Corsaci just tried it
16:47:56Corsacbut it selects a white font ?
16:48:00Corsacnot really readable
16:48:04Corsacand on the shot it's a black one
16:48:08Corsaci'm confused :)
16:50:02Shaidyou're sure you installed the fonts? :P
16:50:09Corsaci have
16:50:13Corsacthe correct font is selected
16:50:17Corsacbut not the right color
16:50:57Corsacwith a reset colors it's better :)
16:51:04Shaidyou don't have some kind of invert screen colour thing turned on?
16:53:44linuxstbdan_a: What was the recording bug with the COP patch?
16:53:49jellisiiis vorbis fixed on 3g? I just tried it with IpodLinux and once it appears to work fine until it has to hit the drive directly.
16:54:47linuxstbjellisii: The only problem should be that it's slow - meaning pauses in playback due to buffer underruns, and I don't think any optimisations have been done recently. Or are you getting different problems?
16:55:03dan_alinuxstb: That was lostlogic's check that we really are running on the COP. Recording requires the codec thread to yeild. I don't understand why it didn't work, though
16:55:31linuxstbBut it's working now?
16:55:44dan_aIt is - at least, for recording to MP3
16:55:48 Quit thegeek (Read error: 110 (Connection timed out))
16:56:00jellisiilinuxstb: Yes, that's the case.
16:56:09linuxstbWhen I tried the last COP patch, it froze immediately when I entered the recording screen.
16:56:33dan_aKoCOP 9 won't :D
16:57:28linuxstbI look forward to testing. Is it showing any signs of the hoped-for speedup?
16:58:17dan_aNot yet. I think that that needs the playback engine to be rewritten.
16:58:30linuxstbDoesn't everything need that?
16:59:23dan_aAt the moment, they're at the UN saying "World peace? I'm pretty sure that will be easy once the Rockbox playback engine is rewritten"
16:59:45l33chwho is working on KoCOP 9 ? :D
17:00:15dan_al33ch: Me. Getting rid of some of the bugs which are keeping it out of SVN, too
17:00:54jellisiiallright then: How about FLAC? I have the buffer underrun problem in IpodLinux too.
17:01:38dan_ajellisii: Flac should work fine (and any other lossless codec). I've had good results with Musepack too
17:01:53jellisiihmm.... must be my encoder then.
17:03:04HardDisk_WPsee ya
17:03:29l33chdan_a: I have H10 port and I used this sample .mpeg video file but there were bugs and audio was faster as video ;-) I used the version 8 patch :)
17:04:31dan_al33ch: Audio being faster than video will be something in the MpegPlayer plugin. What bugs did you see that aren't already on the Flyspray thread?
17:06:31l33chit were in decoding video ... there were fragments .. but as you mentioned it is maybe in this MpegPlayer plugin.
17:06:53 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
17:07:23*dan_a blames linuxstb for that ;)
17:08:01 Quit linuxstb (Nick collision from services.)
17:08:03 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
17:08:50 Quit darkless ("Leaving")
17:09:36linuxstb"audio faster than video" will be caused by you trying to play a video which your player isn't able to decode in realtime. The benchmarks table in the PluginMpegplayer will tell you the kind of framerates you can expect.
17:10:21l33chI go read them
17:10:32linuxstbThose are the average speeds, so if you want to avoid any frame skipping, you should encode at less than those speeds. Bitrate and encoder also affect the decoding speed.
17:11:49 Quit BigMac (Read error: 104 (Connection reset by peer))
17:12:56 Join rotator [0] (n=e@rockbox/developer/rotator)
17:13:05l33chlinuxstb: for iriver H10 5/6GB 128x128 there is no value :-(
17:13:30linuxstbThen you need to test yourself...
17:14:00l33chlinuxstb: I know that you'll say that :)
17:14:42 Part nls
17:14:44***Saving seen data "./dancer.seen"
17:15:15l33chlinuxstb: I knew that you'll say that :P
17:19:41Soapl33ch: all those test results (supposedly) are with the Kernel on COP patch (for the ARM targets).
17:19:56SoapPlease don't add numbers to the table which aren't on such a patched build.
17:20:00Corsachmhm, rockbox only support fat32 filesystems ?
17:20:05 Join BigMac [0] (
17:22:13 Quit Obsys ("Chatzilla 0.9.77 [Firefox]")
17:22:22 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
17:22:28 Quit linuxstb (Nick collision from services.)
17:22:32 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
17:24:57 Join Asura [0] (
17:25:00AsuraCan anyone help me with a song im looking for?
17:25:05 Nick Asura is now known as Asira9 (
17:25:47 Part redwood
17:29:34dan_aHmm... music is playing even if the kernel hasn't been initialised on the COP. That's not very logical.
17:29:40Soaplinuxstb: after two rounds of testing - six total full rundowns, 11h 15min stock build, 11h 30min COP, 11h 45min COP+64MB
17:31:05Domonokylinuxstb: i am looking at your comments for the rbUtil patch. (I dont know how to solve 1 and 5, but rest is fixed)
17:31:46Domonokyi think 1) wrong messagebox (error instead of info) is a wxwidget bug..
17:32:32jellisiihmm..... lemme just grab the rockbox install and see if it's any better... apparently the overhead for ipodlinux is too much for the 3g
17:33:12Asira9Can anyone help me with a ROCKsong im looking for
17:33:39dan_aAsira9: That's off topic for this channel, so no
17:33:55Asira9whats a rockbox then?
17:34:11Asira9a software
17:34:21 Quit Asira9 ()
17:36:50l33chhehe :)
17:37:46l33chlinuxstb: Can you please tell me how to apply your patch ?
17:38:14l33chi just did patch < kernel_on_cop_8.diff
17:38:20l33chand it says
17:38:22l33chcan't find file to patch at input line 32
17:38:34 Quit SirFunk (Remote closed the connection)
17:38:40l33chPerhaps you should have used the -p or −−strip option?
17:38:47dan_al33ch: Which directory are you in?
17:38:53Domonokyuse patch -p0 < ...
17:38:58Domonokyor -p1
17:39:14l33ch /root/rockbox
17:40:23dan_aI think patch -p0 < kernel_on_cop_8.diff should do it, then
17:40:27l33chwith -p0 it was working :)
17:41:22l33chI go test on what fps is this MpegPlayer is good for me :)
17:41:29Soapyea, KoCOP8 needs a -p0
17:41:37 Quit linuxstb (Read error: 113 (No route to host))
17:42:40 Join NeXus [0] (
17:43:12 Nick NeXus is now known as Ne-X-us (
17:44:22 Quit Lear ("Chatzilla 0.9.77 [Firefox]")
17:45:28l33chbtw...stupid question COP == KoCOP8 ? or what is the difference ?
17:46:01SoapKernel on COP is the task. They are up to patch #8.
17:46:30dan_al33ch: COP is the Co-Processor (the iPods, h10 and Sansa have 2 processors)
17:46:53 Part Winchester
17:47:29 Join Nico_P [0] (
17:47:49 Quit gtkspert (Read error: 110 (Connection timed out))
17:48:19 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
17:51:01Ne-X-userr... has anyone worked out whether its actually possible to get code running on the 2g nano?
17:52:13 Join _Veseliq_ [0] (
17:52:17dan_aNe-X-us: Apple have managed it, but nobody else. We need to decrypt either the firmware or the bootloader so we can learn how to put our own software on, and how to control the hardware
17:54:45Ne-X-usyeah i have read quite a bit about it and I was thinking even if we do decrypt the firmware we won't necessarily be able to ENcrypt anything else - ie. apple may have the only encryption key if it uses public key encryption
17:57:00linuxstbDomonoky: Regarding bug 1), it looks like rbutil is using wxLogMessage and wxLogInfo - wouldn't it be better to use wxMessageDialog with the appropriate icon (error, information, etc) ?
17:57:17Ne-X-usdoes anyone know what encryption it uses?
17:58:31 Join petur [0] (n=petur@rockbox/developer/petur)
17:58:32Domonokyprobably, is just need to read a bit about wxWidgets Log and Message facilites..
18:02:05LloreanNe-X-us: I don't think anyone's really put a lot of work into it, yet. Or at least, nobody's talked about it here.
18:03:21preglowperhaps slasheri's id3 fix has broken the info tag reading somehow?
18:03:30preglowanyone tried if mp3 gapless still works?
18:03:41SoapLlorean: any problem with me moving (4 months old) to the trash?
18:04:22Nico_PSoap: why do you want to trash it ?
18:04:37 Quit jellisii ("CGI:IRC (Ping timeout)")
18:04:41SoapNico_P: What good is an unsupported build from october 2006?
18:04:54 Join gtkspert [0] (
18:05:03LloreanSoap: Go ahead.
18:05:07Nico_Pah sorry i thought the thread was about the acceleration patch
18:05:13Nico_Pnot the build
18:05:30LloreanSoap: In about a week (I'm going on a trip soon, when I return from that) I plan to go through every thread in the Unsupported Builds forum.
18:05:45SoapI've emailed / pm
18:06:02preglowMikachu: that's what i'm talking about
18:06:03Soap'd the owners of old threads (30-90) days and given them a heads up to show up or...
18:06:05LloreanI'm going to make a slight change in the rules for the forum, and then notify everyone that isn't in compliance with the new rule, give them 3 days, and then move their threads to the trash.
18:06:36LloreanBasically, I'm going to make it more clear that they need to be more compliant with the GPL in several cases.
18:07:28LloreanPlus I'm going to put an official rule into effect regarding when a build is considered 'outdated' and that it will be removed entirely (not just hidden in the trash) if it becomes so (from that point on).
18:08:15 Part Ne-X-us
18:09:06 Join jellisii [0] (i=4ae6865c@gateway/web/cgi-irc/
18:09:31linuxstbpreglow: I don't have any gapless MP3 files to test...
18:13:26Nico_Ppreglow: gapless mp3 seems to work for me
18:13:39linuxstbOn which target, h300?
18:14:05linuxstbEveryone has a gigabeat nowadays...
18:14:18Nico_Plinuxstb: that's because it rocks :)
18:14:56Nico_Pbut i still have my H300
18:15:05dan_aUpdated Kernel on COP patch posted in FS #5755
18:15:29 Part kopf
18:20:08 Join webguest51 [0] (i=3fe66c50@gateway/web/cgi-irc/
18:20:33webguest51i have an error
18:20:47webguest51is anyone here to help?
18:21:18Lloreandan_a: Okay, I'm not sure what you're saying about frequency scaling there. Specifically, you said "compiled with no frequency scaling support"... Does that mean with scaling disabled?
18:21:35Lloreanwebguest51: There are many people here. Generally it's best to ask the question, and then wait and see if anyone knows the answer.
18:22:03dan_aLlorean: Yes - I tested with and without frequency scaling
18:22:28webguest51my error is displayed on the Ipod and it says "file not found" but i went through all the steps
18:22:57Lloreanwebguest51: I assume you mean that the file it's not finding is Rockbox.ipod?
18:23:05LloreanWhere, *exactly* is your Rockbox.ipod file?
18:23:21webguest51i put it in the Cdrive
18:24:18LloreanFirstly, that's not an exact location. Secondly, that's not where the instructions tell you to put it.
18:24:30LloreanYou didn't go through the steps exactly as they are in section 2.2.2 of the manual
18:25:15 Join barrywardell [0] (n=barrywar@
18:25:25 Quit Arathis ("Bye, bye")
18:28:32webguest51im still confused, could you tell me exactly what to do?
18:29:13Lloreanwebguest51: Are you reading the actual manual? What part of the instructions in it is confusing you?
18:29:26LloreanIf they're confusing, we need to improve them, so it would be very helpful if you could be specific.
18:29:56 Part Corsac
18:30:24 Join dpassen1 [0] (
18:31:19 Join toffe [0] (
18:32:02 Join mirak [0] (
18:32:39jellisiihey, that's cool: Vorbis files encoded at -q 6 seem to play nicely on my 3g.
18:33:23webguest51so i extracted the files to a folder in the Cdrive named rockbox, when i went through the rest of the steps i didnt have any problems, i guess i just dont know where i can put the files so the Ipod nano can find them
18:34:59 Join Gigabeater [0] (i=464e0a82@gateway/web/cgi-irc/
18:35:56GigabeaterSomeone please exlain to me how to get Doom to work on my Gigabeat. I've read the manual and still don't understand
18:36:10Lloreanwebguest51: That's not what the instructions say to do though. I told you specifically, read section 2.2.2 of the manual.
18:36:13Lloreanwebguest51: Have you done that yet?
18:36:27LloreanGigabeater: What part don't you get?
18:37:02l33chl8r guy
18:37:14 Part l33ch
18:37:23 Quit jellisii ("CGI:IRC (EOF)")
18:37:32LloreanGigabeater: Specifically, how far did you get with the instructions?
18:37:42webguest51ok, where is my "playe's drive"
18:37:48GigabeaterI'm not sure exactly.. I made a doom directory and put the rockdoom.wad in there. With the is another .wad file. I don't know where to put it, so I threw it in .rockbox/doom/addons. When I try to play doom I get a missing wad error
18:38:46LloreanGigabeater: The instructions tell you that you need to put the game wad in "that" directory, meaning the same one you put rockdoom in.
18:38:57GigabeaterI tried that too
18:38:59GigabeaterAnd it didnt work
18:39:03Lloreanwebguest51: If you look in "My Computer", which drive is your player?
18:39:23LloreanGigabeater: You have /.rockbox/doom/rockdoom.wad and what?
18:39:39linuxstbdan_a: Why is "fallback" needed in create_thread? Is that for people running incompatible bootloaders?
18:39:48GigabeaterI just too doomf.wad out fo addons and put it in doom folder
18:40:06 Join Mouser_X [0] (
18:40:52LloreanGigabeater: Try renaming it doom2.wad
18:40:53webguest51opps my bad, i needed to rebot the nano, thanks though
18:40:54 Join Arathis [0] (
18:41:32dan_alinuxstb: Yes, or so that if there are MAX_THREADS threads on one core, the thread can be created on the other
18:43:03GigabeaterIt wokrs! That's Llorean
18:43:11 Join bro2baseball [0] (
18:43:30*Llorean wonders when doom stopped allowing doomf.wad to work
18:44:36LloreanAnyone mind if I commit a small change that makes Rockbox Doom show "Doomf.wad" as "Freedoom" and treat it as a supported base wad, since that's how we're distributing it at the moment?
18:45:03 Quit Mouser_X (Remote closed the connection)
18:45:11 Join shodan- [0] (
18:45:14 Join Mouser_X [0] (
18:45:40linuxstbLlorean: It's about time you started committing things ;)
18:46:25Lloreanlinuxstb: I've made two whole changes to the manual, and fixed one whole warning. So there. :-P
18:46:39LloreanLemme just make sure I don't need to change anything but these two strings, first.
18:47:08GigabeaterWhat button opens doors in doom for gigabeat
18:48:03Gigabeaterand is there a way to enable sound
18:48:03LloreanGigabeater: If it's not in our manual, once you find out which buttons do what, could you post them to the PluginDoom wiki page?
18:49:18Gigabeaterahh got it
18:49:27GigabeaterYeah I can.. how do I post it though
18:49:33GigabeaterCould I just tell you?
18:50:32LloreanIt'd be best if you just filled in the keys thing on the PluginDoom wiki page.
18:51:12 Join PaulJam [0] (
18:52:31 Quit gtkspert (Read error: 110 (Connection timed out))
18:53:01LloreanThe "rotate screen" option in Doom doesn't rotate the controls...
18:53:52GigabeaterI can't figure out how to fill in the keys thing in the PluginDoom page
18:54:13 Quit [shodan] (Read error: 60 (Operation timed out))
18:55:00barrywardelldan_a: I now get a main thread on the cop with v9 :)
18:55:12barrywardellwhat did you change that could have caused that?
18:56:45PaulJamOn my h320 with a non SVN build, the options for backlight and backlight while charging are both called "backlight". does this problem also happen with a svn build?
18:57:58 Quit Gigabeater ("CGI:IRC (EOF)")
18:58:11 Quit Mouser_X (Nick collision from services.)
18:58:15Nico_Pwhat do you guys think of my progressbar slider patch ? should i commit it ? (it adds 420 bytes to the gigabeat bin)
18:58:20 Join Mouser_X [0] (
18:58:27barrywardellPaulJam: I'm getting that here on my H10 with SVN build
19:03:02 Join Thundercloud_ [0] (n=thunderc@
19:07:03 Nick BHSPitMonkey_ is now known as BHSPitMonkey (
19:07:12barrywardellPaulJam: think I see the problem. fix on the way...
19:08:08 Quit Mouser_X (Nick collision from services.)
19:08:17 Join Mouser_X [0] (
19:08:19 Join jellisii [0] (i=4ae6865c@gateway/web/cgi-irc/
19:09:49PaulJamanother thing: in the id3 priority selection it seems as if the entrys are interchanged
19:10:11 Join gtkspert [0] (
19:10:25barrywardellPaulJam: I just committed a fix for the first problem.
19:10:34PaulJamif i select "v2 then v1" it doesn't see the replaygain info for the files
19:10:51 Quit Nico_P (Remote closed the connection)
19:10:57PaulJamthank you
19:11:20webguest51im having trouble listeing to my music, i get it from i tunes
19:12:02barrywardellwhere is the id3 priority setting?
19:12:03PaulJamwebguest51: do you mean songs from the itunes music store?
19:12:36webguest51yes, from the music store, and i dont know wut the id3 priority setting is
19:12:47barrywardellmy question was for PaulJam
19:12:52PaulJambarrywardell: general settings => playback => id3 tag priority
19:13:42PaulJamwebguest51: rockbox doen't play DRMed content
19:13:57Lloreanwebguest51: Music Store songs are encrypted, we will not support them.
19:14:24webguest51so the music i bought off of itunes, and the music i have stored on itunes i can not listen to?
19:14:46***Saving seen data "./dancer.seen"
19:14:48LloreanJust bought.
19:15:25 Join Thundercloud__ [0] (n=thunderc@
19:15:26LloreanWe've made no secret of this, it's even in the iPodFaq. You knew already, I'm sure, that you couldn't listen to it with any other MP3 player too. iPods with Rockbox are basically 'another MP3 player', at least while running Rockbox.
19:17:50mirakwebguest51: I must download illegaly the music you bought legally to listen it
19:18:01mirakwebguest51: you must download illegaly the music you bought legally to listen it
19:18:14jellisiiThank apple for that.
19:18:28 Quit Thundercloud (Read error: 110 (Connection timed out))
19:18:36mirakI think it's not even bound to apple
19:18:46Soapwhoa mirak. There's no need to suggest illegal activity.
19:18:51LloreanWell, the problem with iTunes is that you aren't buying music, you're renting it. They just hide that.
19:19:02LloreanBut it's neither here nor there.
19:19:08mirakSoap: what is illegal ? the guy as the right for is songs or not ?
19:19:12LloreanIt's not supported, and if you want to use the iTunes Music Store, you can't use it with Rockbox.
19:19:16Lloreanmirak: It is.
19:19:40SoapHe doesn't have a right to the song, he has a right for limited use of said song.
19:19:42Lloreanmirak: At least in the United States. You're not getting an all-inclusive right to that work, just that one particular copy of that work.
19:19:55mirakapple says now that DRM are bad, but they used them to keep people around ipods and itune I guess
19:20:15SoapBut this is not the place to split such hairs, not argue the merits/costs of DRM.
19:21:28webguest51its just economics, keeping the consumer trapped in the company, and they are doing a fantastic job of keeping thousands of people within the company
19:21:30mirakrockbox is part of that revolution anyway, the right to use your hardware as you want
19:21:59webguest51it just a shame we suffer, due to it
19:22:34mirakthey should have stick to vinyls
19:22:55SoapCan we talk Rockbox in the #rockbox channel?
19:22:59jellisiiugh... that's somethign I don't miss.
19:23:13jellisiivinyl, that is.
19:23:20mirakis there a scratch plugin for rockbox ?
19:23:39mirakI guess it could work well with ipods wheel
19:23:41Mikachujust put some butter in the headphones slot
19:24:10jellisiithe "hiss and pop" synthesizer is in the works, I'm shure
19:24:23 Quit mirak ("Ex-Chat")
19:25:14 Join mirak [0] (
19:25:26webguest51okay im havin trouble listening to music that i bought
19:25:46Mikachudid you just forget the whole last 10 minutes of conversation?
19:26:05webguest51no music that i bought off of a CD
19:26:57LloreanWhat problem are you having, then?
19:27:58barrywardelllinuxstb: I think with your usb detection code, it should report charger connected when it detects the charger rather than usb
19:28:01webguest51Well when im in the music player, there is nothing but question marks on everything
19:29:08LloreanWhat 'music player' are you referring to?
19:29:33jellisiiis there any obvious reason that Rockbox should hang my player when I hit the play/pause button on my 3g ipod?
19:29:33peturwebguest51: how did you convert the music from CD to files on PC, and what format?
19:30:01Lloreanjellisii: Every time you press it?
19:30:16 Join ender` [0] (
19:30:34Lloreanjellisii: And by 'hang' do you mean 'freeze and require a reboot' or just 'take a few seconds to do anything'?
19:31:05jellisiiLlorean: hangs with "loading. . ." displayed. and it seems to be every time.. lemme reboot and check
19:31:18webguest51i have the CD on both Itunes and microsoft media player
19:31:26Lloreanjellisii: So it's freezing while resuming?
19:31:46jellisiioh hey, it's intermittant... hmmm
19:32:03jellisiiLlorean: resuming or starting playback after reboot.
19:32:47jellisiilemme try with just MP3 files... the vorbis overhead may be killing the processor.
19:33:56 Nick Mouser_X is now known as front (
19:34:37perldiverwhat's "Freedom" game?
19:35:00peturwebguest51: so what format did it produce? Give the extension of the music file you copied on your player
19:35:07 Nick front is now known as The_Master (
19:35:55 Nick The_Master is now known as The_Mouser (
19:36:14 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
19:36:53webguest51how can i load music on the ipod, and listen to it
19:37:04Lloreanperldiver: Freedoom, you mean?
19:37:04 Nick The_Mouser is now known as Mouser_X (
19:37:25perldiverLlorean ha, yes, sorry, Freedoom
19:37:27Lloreanperldiver: It's a full Doom2-compatible IWAD produced with entirely GPL licensed material.
19:37:33LloreanWhich means we can distribute it.
19:37:42LloreanIt's the doomf.wad that we provide on the Extras page.
19:37:42perldiveraha, thanks
19:37:59LloreanYou can you any Doom2 compatible add-ons with it, too. Or should be able to.
19:38:41peturwebguest51: firt you mst convert the CD to music files that are supported by RockBox: MP3, Ogg Vorbis, Musepack, FLAC, AAC, ALAC, AC3 and WavPack (pick one ;) )
19:38:55PaulJamI have a question about the new "play next" feature. i used this feature with a Database entry, but the playlist that was created was not shuffeled even though shuffle was enabled. is this intended?
19:38:57Lloreanwebguest51: Have you read the section in the manual on playing music?
19:39:00 Quit Thundercloud_ (Success)
19:39:10jellisiiwebguest51: Then you'll need to boot the iPod into diskmode to copy files to it.
19:39:22 Join Thundercloud_ [0] (n=thunderc@
19:39:53LloreanPaulJam: It works like "Insert" which means there should be a Play Next Shuffled.
19:40:42linuxstb_petur: You forgot Shorten, WAV, AIFF, MP1 and MP2 ...
19:41:02*petur copy-pasted from the wiki :p
19:41:07PaulJamLlorean: thanks for the information.
19:41:21Mouser_XYou also forgot ADX.
19:41:29Mouser_XThere is an ADX encoder available.
19:41:30hcsheh, beat me to it
19:41:36perldiverwhat's the difference between Play Next and Queue?
19:41:37LloreanAnd Speex, if you're really crazy.
19:41:45Lloreanperldiver: There is a manual. :-P
19:41:52Lloreanperldiver: Play Next is the opposite of Queue in every way.
19:42:11perldiveris Play Next already documented?
19:42:27LloreanPlay Next clears the current playlist, makes the current song "queued", and replaces the existing playlist with whatever you chose to play next.
19:43:01LloreanIt "plays" the selection next.
19:43:02Mouser_XAh. That could be useful, in some rare instances.
19:43:43LloreanIt was a very frequently requested option.
19:43:51jellisiiLlorean: My "hanging" problem seems to be just switching between songs.
19:44:03jellisiiand intermittant.
19:44:32 Quit funky (Read error: 113 (No route to host))
19:45:14webguest51so there is not a way to convert the music from itunes to the rockbox, legally?
19:45:40Mikachudepends on where you live
19:45:48 Join funky [0] (n=repulse@unaffiliated/funky)
19:45:59Mikachuthen no chance
19:46:13Mouser_XCouldn't you just play the audio, and record it back to something else?
19:46:26Mikachutechnically it's not hard at all
19:46:33Lloreanwebguest51: Apple suggests that you burn a CD, and then convert that CD if you must use iTunes songs with an unsupported MP3 player.
19:46:37Mikachubut it's illegal to bypass copy protection in their country
19:46:37Mouser_XSo, while it's playing in iTunes, record the audio with something else, and convert the audio to some other format?
19:46:39jellisiiThe DMCA prevents that being done legally. Taking the analog output is technically circumventing copy controls.
19:46:59Mikachuyou can strip the drm digitally too i heard
19:47:13Mikachusome program can swipe the decrypted data from memory while playing in itunes
19:47:15linuxstb_Isn't there an interoperability clause in the DMCA?
19:47:16jellisiiAmerican legislattos == stupid.
19:47:25jellisiier.. legislators...
19:47:32jellisiilemme get my glasses
19:47:33Mouser_Xjellisii: Yes, that's a better way of saying it.
19:47:34barrywardelllinuxstb_: you see my comment about ipod charger detection?
19:47:46perldiverMikachu it involves a lot of messing around and perl scripts hehe
19:48:08webguest51is converting a copy cd easy?
19:48:13Dunkelschubalthough getting it off the ipod is extremly simple
19:48:15 Quit Thundercloud__ (Read error: 110 (Connection timed out))
19:48:20jellisiiperldiver: which also is against the DMCA
19:48:23MikachuDunkelschub: it's still encrypted then
19:48:26linuxstb_barrywardell: Yes. I think we can just detect chargers by GPIO though, at least on the ipods.
19:48:27Lloreanwebguest51: Look, if you _really_ want to use iTunes, you probably shouldn't be using Rockbox.
19:48:51webguest51yeah i think i have come to that
19:49:29*Mouser_X *really* wants to avoid anything apple based...
19:49:32jellisiiwhich is a bummer, because iTunes is a real pain to use, in my opinion.
19:49:40Mouser_XAs such, I use Rockbox on a Gigabeat.
19:49:46Lloreanwebguest51: What was it about Rockbox that interested you in it, anyway?
19:49:54*linuxstb_ likes Apple hardware, but hates Apple software.
19:50:03MikachuLlorean: you sound like it's strange that someone would like rockbox :)
19:50:32*jellisii agrees with linuxstb
19:50:43webguest51games, more music, and popular science said it would be an upgrade
19:50:50LloreanMikachu: I'm curious if there's something specific he was looking for, since he seems pretty attached to iTunes.
19:51:24barrywardelllinuxstb: I mean, currently when we connect the charger we just ignore it as if nothing happened. we should report it being connected
19:51:27LloreanMikachu: I'd gotten the impression it wasn't an 'informed' opinion.
19:51:33webguest51and the custum stuff
19:51:52 Quit bro2baseball ("Download Gaim:")
19:52:09linuxstb_barrywardell: I think that's just because the GPIO code hasn't been implemented. We also need to detect charging on the 12V (firewire) pins.
19:54:56barrywardellah, ok
19:54:58 Join webguest14 [0] (i=d90a1a7e@gateway/web/cgi-irc/
19:56:16ArathisIs there a reason why the YEAR tag of ogg files isn't read correctly? When showing ID3Tags of the current playing track while in WPS it shows the year, but when using tag-cache and browsing by year the files don't show up there
19:56:35Mikachui think it's because it's a string in vorbis tags
19:56:54Mikachubut i'm only vaguely sure
19:57:06linuxstb_I think YEAR/DATE is a string in most tags.
19:57:46linuxstb_Arathis: Are your tags "YEAR" or "DATE" ? I thought DATE was standard in vorbis comments?
19:58:13*Domonoky updated the rbutil patch. new Windows binary: domonoky/">
19:58:21 Quit webguest51 ("CGI:IRC (EOF)")
19:59:14Arathislinuxstb_: date=2006
19:59:41ArathisI just looked at tag-cache menu and what he showed in the wps context
20:00:05Arathisso DATE is used (regarding to the header of the file ^^)
20:02:36 Quit webguest14 ("CGI:IRC (EOF)")
20:02:46 Join SirFunk [0] (
20:03:47linuxstb_Looking at the vorbis parsing code, it stores the DATE string in the "year_string" field, and doesn't attempt to populate the "year" with an integer. I'm guessing tagcache is using the year integer, and not falling back on the string, whereas the WPS checks the string first, and if there is no string, uses the integer.
20:05:13Mikachui was close then
20:05:56linuxstb_Exactly, so...? Vorbis DATE can be anything, how can you convert it to a year?
20:06:40Arathiswith "anything" you mean string, integer, etc?
20:06:52Mikachuextracting \<[0-9]{4}\> is a good bet :)
20:07:07linuxstb_0204 isn't going to be a year...
20:07:55hcsI'm pretty sure we've had this discussion before. While it may be hard to impossible to get everything right, it is still possible to make an effort that works most of the time.
20:08:04Mikachuit's not a date either
20:08:12linuxstb_I'm sure we must have done.
20:08:34hcsI specifically remember coming up with a bunch of stupid ways to give valid dates
20:09:37LloreanAnd populating the list with 'valid' years that make no sense is, in my mind, worse than not populating it with years.
20:09:39 Quit midgey ()
20:10:54LloreanIt seems to me we could look for just a year, or YYYY-MM-DD and in either of those cases extract the year, all others discard?
20:11:01Mikachuwhy not just put the whole date string in? it's up to the user to tag consistently
20:11:11 Quit Mouser_X (Nick collision from services.)
20:11:20 Join Mouser_X [0] (
20:12:10LloreanMikachu: How could you put the whole date string in as a single int?
20:12:43Mikachui mean in year_string then
20:13:09LloreanThat's or less what I was suggesting though. Assume that they're either just going to provide a year, or a standard date.
20:13:52linuxstb_12 876§12Q7YUJIOPK[L;'.?A-[|
20:14:10LloreanHaving computer issues?
20:14:11 Quit wooo ()
20:15:04 Join wooo [0] (
20:15:08linuxstb_Ooops, just cleaning the keyboard... :)
20:15:23jellisiiand we were on our way to a bash quote...
20:17:05linuxstb_Mikachu: That's what happens now - the DATE string goes into year_string. But I guess tagcache wants to sort and filter by year...
20:18:29 Quit jellisii ("CGI:IRC")
20:18:34 Join jellisii [0] (i=4ae6865c@gateway/web/cgi-irc/
20:21:24 Join robin0800 [0] (
20:21:48peturwhat's the default wps on H1x0?
20:22:14linuxstb_Same as every other player afaik - rockbox-default (it's hardcoded in the wps code).
20:22:28SoapThat's really a philosophical question, isn't it? How far should Rockbox go to accommodate inconsistant (if not outright poor) tagging?
20:22:29peturnot iCatcher?
20:22:50linuxstb_No, although I wrote a patch to make iCatcher the default theme on all players.
20:22:53*petur is trying to help a h120 newbie on MR
20:23:16Soaptell him it is hopeless, and he should sell me his h120 for $100
20:23:20jellisiiI would think that the tagging is the responsibility of the encoder: The spec for every format is documented.
20:23:36robin0800How do you get cygwin to use arm-elf-gcc 4.0.3
20:23:49Soapyes - that was directed at you petur
20:23:57peturyou're looking for an h120?
20:24:03SoapI'd love one.
20:24:54robin0800any experts here with cygwin?
20:25:01blackness I don't know what happened but all of my plugins dissapered on my 5g ipod, how do I get them back?
20:25:21Lloreanrobin0800: The official install directions do that for you...
20:25:32Soapblackness: did you mount your ipod and check to see if the .rock files are actually still on the hard drive?
20:26:37robin0800 are you you sure this is not broken?
20:26:55 Join Thundercloud__ [0] (n=thunderc@
20:27:21perldiverLlorean cant find anything about Play Next in the manuals
20:27:38PaulJamcould someone with a svn build look if the statusbar overlaps the text in the "Running Time" screen?
20:28:06Lloreanrobin0800: It does look like it's outdated, apparently
20:28:21Lloreanrobin0800: Just use or the manual CrossCompiler install instructions then
20:28:28Lloreanperldiver: It appears to be undocumented for the moment.
20:28:56perldiver worked for me for the 4.0.3 update
20:29:12blacknesswhere exaclty are the plugins located?
20:29:23robin0800how do you use
20:30:01LloreanYeah, I misread
20:30:06Lloreanrobin0800: The .ini is fine, it's not outdated.
20:30:33Lloreanperldiver: I opened it up in a text editor that didn't support unix style line endings and misread the previous version line as the current one.
20:30:49robin0800when I run setup it produces errors
20:31:17blacknessnm got it
20:32:49jellisiiI wonder if I just have a hardware problem... It's still hanging intermittantly. I'm trying to figure out a way of consistantly replicating it.
20:33:18miraklinuxstb: you run linux on your mac ?
20:33:19robin0800it says can't get setup.ini etc and fails
20:33:50linuxstb_mirak: I did on my previous ibook, but don't on this one.
20:33:58Lloreanrobin0800: Error messages tend to actually *mean* something. Saying "etc" and dismissing them makes it hard for anyone to have any hope of helping you.
20:35:22miraklinuxstb: ah ! you run windows !
20:35:49linuxstb_I run from windows...
20:36:15perldiverdownload that and overwriting actually works too
20:36:30perldiverbut it doesnt give me errors with regular setup as well
20:37:14miraklinuxstb: I have two macs my neigbourgh gave me, I gave up running os x on them. it looks good etcetera but applications are slower than on linux, for exemple firefox crawls
20:37:28jellisiiLinux for work, windows for play.
20:37:41 Join midgey [0] (
20:38:42linuxstb_mirak: Unless a breakthrough has been made recently, I can't use my wireless network on my ibook with Linux, so I have to use Mac OS X for that. I can live with it.
20:39:01barrywardellPaulJam: It doesn't overlap for me
20:40:01Mikachui'd expect wlan to be an area with fast progress
20:40:32miraklinuxstb: on good hardware sure.
20:41:00 Quit YouCeyE ("Leaving")
20:41:12PaulJambarrywardell: ah, thank you. then it is propably caused by one of the patches.
20:42:22miraklinuxstb: does linux supports well energy saving on laptops ?
20:42:48ender`mirak: depends on the laptop, but usually works fine
20:42:54Mikachulinux does, gnome/kde isn't so good
20:44:14Mikachuthey have tons of little processes that wake up very often, doing nothing, but they still interrupt the power saving mode of the cpu
20:44:40robin0800downloaded and extracted where do Icopy it to?
20:44:44 Quit fasmaie (Read error: 110 (Connection timed out))
20:44:56barrywardellanyone objections to committing FS #6662?
20:45:45Mikachulooks pretty sane
20:45:50linuxstb_Does Rockbox have a math.h then?
20:46:44Mikachuthree of them
20:46:45*linuxstb_ finds math.h answers his own question
20:46:49Mikachuapps/codecs/libmusepack/math.happs/codecs/libspeex/math.h firmware/include/math.h
20:46:58barrywardellmath.h is #include'd in calculator.c already
20:47:15Mikachui'm assuming the latter is the used one :)
20:48:31barrywardellOK. I'll commit it. M_PI, etc are defined in the latter one
20:49:05Mikachusort of strange to define 2*pi and pi/2
20:49:20Mikachui'm assuming even gcc will evaluate those at compile time
20:50:07blacknesshow long does it usually take to load a zxbox plugin
20:50:55 Quit Thundercloud_ (Connection timed out)
20:53:17 Quit robin0800 ("Make it idiot proof and someone will make a better idiot.")
20:55:38desowinjust wondering - is there 'database creator' running on computer not on device ?
20:55:50 Join blithe [0] (
20:55:50 Quit Brunellus ("Ex-Chat")
20:56:23 Join mattzz [0] (
21:04:45 Join perl|perl [0] (
21:04:45 Quit jellisii ("CGI:IRC (EOF)")
21:07:16perl|perlis Dean Murray here by any chance?
21:13:11 Quit Mouser_X (Read error: 110 (Connection timed out))
21:14:49***Saving seen data "./dancer.seen"
21:15:42 Quit perldiver (Read error: 110 (Connection timed out))
21:19:04 Part barrywardell
21:20:03 Quit ``0_0 (Remote closed the connection)
21:22:32 Join barrywardell [0] (n=barrywar@
21:35:45 Quit Kitt0s ("( :: NoNameScript 4.02 :: )")
21:38:10 Join Kitt0s [0] (
21:39:21linuxstb_dan_a: My ipod color has been running almost 3 hours now with koCOP9, so it's looking good.
21:42:33linuxstb_Although it does feel slower in normal use...
21:43:44barrywardellslower than without using the COP?
21:44:49linuxstb_Yes. The menus seem more sluggish.
21:45:47barrywardellthat's weird
21:46:47linuxstb_I'll try an svn build later to compare (I want to see how long it runs without freezing).
21:47:29 Quit _Veseliq_ (Remote closed the connection)
21:53:49preglownice bug, that...
21:54:19Bagderwe should add some gcc option to warn when doing #if on undefined defines
21:54:47preglowwell,isn't that a quite common usage scenario?
21:54:58Bagderit shouldn't anyway
21:55:06Bagderyou should #ifdef then
21:55:11Bagderor #if defined(blabla)
21:55:15preglowahhh, yes, yes
21:55:17preglowof course
21:56:09Bagder"Warn if an undefined identifier is evaluated in an #if directive."
21:57:04hcsthat'll throw some warnings in the spc, I know
21:57:25preglowi think it's a good idea anyway
21:57:50hcsI agree, just letting you know ahead of time, blargg liked to use the 0 assumption
22:00:01Bagdermaaaaany warnings
22:00:55Bagderno wrong, make that maaaaaaaaaany
22:01:44linuxstb_In spc or everywhere?
22:01:49Bagderall over
22:02:03Bagdermostly the same defines involved
22:02:06linuxstb_But we're nice programmers writing clean code...
22:02:44BagderI guess most of the warnings are in the same header files actually
22:02:56Bagderjust included from everywhere
22:02:56 Quit printfXh4 (Read error: 113 (No route to host))
22:02:58Bagderlike export/config.h
22:03:33hcsgrumble grumble people not setting file pointers to null after closing
22:07:27 Join printfXh4 [0] (
22:08:58 Join tick [0] (i=95e13cf6@gateway/web/cgi-irc/
22:10:39tickAnybode here?
22:11:18 Quit powr-toc ("Leaving")
22:11:19 Nick printfXh4 is now known as Xh4 (
22:13:46tickOk, then I'll ask. Under cygwin, when I start make, the CPU goes to ~50% and nothing happens. In the process list, I see svnversion.exe. If I kill it it appears again. If I kill it again make does its work very quickly. What could it be?
22:14:04 Join bullmoon [0] (i=549957fa@gateway/web/cgi-irc/
22:14:22Bagdertick: you using source checked out from svn?
22:14:40tickBagder: yes
22:14:59bullmoonBagder, do you mind if I ask a server related question
22:15:33Bagdertick: did you check it out with cygwin?
22:15:36Bagderbullmoon: shoot
22:15:55bullmoonis it possible to run p2p software like DC++ on a NAS device
22:16:14tickBagder: ehm... I think no. I checked it out with TortoiseSVN.
22:16:45Bagdertick: I bet that is the reason then, the cygwin svnversion not liking it or something
22:16:52tickBagder: does the file format differ between tortoise and cygwin svn?
22:16:53*petur votes for the offtopic question of the day
22:17:17Bagdertick: I don't know but they could differ on line endings or so
22:17:36tickBagder: I'll try to execute svn update under cygwin
22:17:59Bagderbullmoon: I thought it was about a Rockbox server. But you can run software on a computer yes, as a NAS is a computer I guess
22:18:48bullmoonthanks, and i guess i should've been more specific, sorry
22:19:03Bagder-Wundef also breaks the sansa build...
22:19:20Mikachuhow can it break? you have -Werror?
22:19:50tickhmm... svn under cygwin brought just a few lines, not the whole source tree. And I couldn't find a file svnversion.exe on my hard disc. Weird.
22:19:57 Part bullmoon
22:20:10BagderMikachu: it makes stuff like "name = DEFINE;" fail if DEFINE isn't defined
22:20:28Bagdercommon/dircache.c:55: error: 'DIRCACHE_LIMIT' undeclared here
22:20:55BagderI haven't worked it out completely
22:21:06Bagderthat file shouldn't even compile on sansa
22:21:21tickSo svn under cygwin is 'compatible' with tortoise.
22:21:40peturtortoisesvn, that is
22:22:01tickpetur: that was my conclusion, not a question :-)
22:23:45Bagderhm, reeeeed
22:24:16dan_alinuxstb_: Good to hear it's working to some extent. I'm sure there's still weird things going on: why should recording not work if the codec thread doesn't yield, when the codec thread is on a different core?
22:24:28tick Bagder: everything ok with you keyboard? ;-)
22:25:39peturWHOA monster score :p
22:25:54petur37826 points
22:26:14barrywardelldan_a: I'm getting some weird behavior. A couple of times I have had it reboot straight away on startup, then it boots the second time and sometimes i get an error ("Un...") flashes on the screen over the Rockbox logo for a split-second.
22:26:29midgeyoh man, i believe thats the second highest score ever
22:26:46dan_aI'm getting errors on the 3G too
22:26:57Bagderyeah, all over
22:27:22 Quit relaxed (Remote closed the connection)
22:27:39tickAha! I've found it. svnversion is a program that I have under cygwin. But I don't understand why it's executed. @Bagder: any ideas?
22:27:54Bagderthe build procedure runs it
22:28:00preglownow that's a score
22:28:00Bagderto figure out the rev number
22:28:27Bagderits the config.h change
22:28:51Bagderfix coming up
22:29:03tick Bagder: is it necessary? It takes quite a bit of time.
22:29:28barrywardelldan_a: unfortunately, it's not reliably reproducable. Only about 1 in 10 times
22:29:53barrywardelldan_a: do I need to update my bootloader too?
22:30:13tickBagder: and it's very annoying if you want to have many quick edit-compile cycles
22:31:30dan_abarrywardell: Only if you haven't since about August
22:32:31Bagdertick: create a docs/VERSION file with your version number in
22:33:20barrywardelldan_a: in that case, then I don't.
22:35:25tickBagder: just one line with my current svn revision?
22:35:52 Part digen ("Leaving")
22:35:58Bagderyes, or whatever you want it to show in the version screen
22:36:38tickBagder: Yessss! Thanks a lot!
22:36:47dan_abarrywardell: Does it behave any better if you put "#if CURRENT_CORE==COP" around the call to ipod_set_cpu_frequency in system_init()?
22:37:01perl|perlis Dean Murray here by any chance?
22:37:10barrywardelldan_a: I'll test that
22:37:41tickBagder: does it have to be a number? Or just an arbitrary string?
22:37:43 Quit Tom0473 (Read error: 131 (Connection reset by peer))
22:37:54Bagdertick: any string I believe
22:38:31barrywardelldan_a: do you mean COP, or CPU?
22:40:43dan_abarrywardell: i mean COP. KoCOP8 had it done by the CPU, and that didn't work.
22:41:02dan_aWe might just need to call that function at a different time
22:41:49tickBagder: just as an info: not every string in doc/VERSION is ok. I got compile error in main.c:110 after I entered 'doc/VERSION' there. 'doc_VERSION' is ok.
22:43:07peturtick: try two slashes
22:43:53tick petur: how is it used? Does it get passed as an option to the compiler? Or is it used as a define in the compiler command line?
22:44:29peturprobably in some string function, and / is an escape char,// is the / char
22:44:51barrywardelldan_a: or maybe use a more advanced function. one like I had before based off how the H10 OF does it
22:44:56barrywardellI'll try that too
22:45:01 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:45:03tick petur: / is an escape??? Not \?
22:48:31 Part tick ("gtg")
22:52:52Bagder-Wundef coming up
22:54:08 Join Mouser_X [0] (
22:55:58Bagderthere are a few warnings left
22:56:14mattzz need some help?
22:56:23 Part blackness
22:58:21BagderI think fixing the remaining ones should be fairly quick and easy
22:58:53 Quit amiconn (Nick collision from services.)
22:58:53 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:59:31Bagderbut thanks anyway
23:02:13*amiconn wonders what Bagder is doing...
23:02:30*Soap wonders what a Wundef is
23:02:48Bagder"Warn if an undefined identifier is evaluated in an #if directive."
23:05:57Bagderamiconn: just breaking a few builds so that less builds will break in the future
23:06:01*amiconn wonders whether this -Wundef stuff is really necessary
23:06:15Bagderit would've caught the bug Linus just fixed
23:06:21Bagderand we've had such bugs in the past too
23:06:22amiconnWe went for years without, and it makes the expressions longer
23:06:30Bagderthat are rather hard to catch
23:06:54Bagderit just makes a few expressions longer
23:07:54amiconn-Wundef: Protects from programmers who cannot type properly ;)
23:08:05Bagdernot only
23:08:21Bagderit also protects authors who don't include the correct include file
23:08:24amiconnYellow and red all over
23:08:39Bagderlike for #if PAD == ANOTHER_PAD
23:09:31amiconnShouldn't this change only produce yellow, but no red?
23:10:02barrywardelldan_a: With only running the ipod_set_cpu_frequency() on the COP I haven't been able to reproduce the reboot problem yet :)
23:10:10BagderI broke that when I "fixed" dsp.c
23:10:32amiconnHmm, the log colouring script seems to be a little confused
23:11:31amiconnThe 'In file blah' lines for warnings should be yellow, not red, shouldn't they?
23:12:17Bagderwhy does dsp.c do this?
23:12:29Bagdersound.h seems to be a bit odd
23:12:37Bagderit breaks if config.h is included before it
23:14:18 Join Lear [0] (i=chatzill@rockbox/developer/lear)
23:14:43Bagderthat's the change that causes the reds anyway
23:14:46LearOuch, ugly build table...
23:14:52***Saving seen data "./dancer.seen"
23:14:57Bagderxmas-tree look!
23:15:20LearAnyone working on it?
23:15:40Juice^ouch :o
23:16:14 Join x1jmp [0] (
23:16:33 Quit Mouser_X (Read error: 110 (Connection timed out))
23:17:11dan_abarrywardell: Excellent!
23:18:07LearHuh? Complaints about "bool"? Don't really see why...
23:19:20dionoeahowdy. same error here
23:19:35preglowamiconn: i've seen plenty of errors/mistakes that would never have happened if this was enabled, so i welcome it, at least
23:19:37dionoeaC doesn't define bool
23:19:44dionoea(or at least old versions didn't)
23:20:32dan_aIMHO, anything that makes bugs easier to catch is good, even if it does make for longer lines
23:22:59Learbool is a newer addition (like C98 or C99 or something). Worked before, but the move of a config.h include somehow breaks it...
23:24:06linuxstb_Shouldn't you include <stdbool.h> to use bool?
23:24:32Bagderyeah, that fixes it
23:25:45Bagdercommitted now
23:27:28Learuda1380.h still fails here...
23:28:48Bagderfor what target?
23:28:51 Join Wiwie [0] (
23:29:05Learuda1380.h:27: error: syntax error before "extern"
23:29:43LearMakes no sense, but...
23:29:44peturlinuxstb: re FS #6648 (the similar names fat bug): was it essential that one file was a long filename and the other a valid shortname (8.3)?
23:30:04BagderLear: not for me
23:30:24peturlinuxstb_ too
23:30:31Bagderrev 12362?
23:30:36LearHm.., maybe deps aren't cleaned enough...
23:30:49LearYep, double-checked that.
23:31:26linuxstb_petur: I didn't do any more tests - all I know is that those particular names caused the problem.
23:31:32Juice^the gameboy emu on sansa is quite laggy when playing a game like mickey mouse i just tested, when there is enemies.. so its the calculation of their behaviour that is taking up alot of cpu, or is it unoptimized code? maby a combination
23:32:00LearInclude from mp3_playback.c this time; stdbool.h should perhaps go into the header actually using it...
23:32:42LearHuh? mp3_playback includes it...
23:32:49 Quit ender` (" Logic is a systematic method of coming to the wrong conclusion with confidence.")
23:33:15 Join kubiix [0] (
23:34:42Bagderthe h1x0 gets quite a few warnings in ata.h
23:34:43amiconnQuite obscure bug Linus found...
23:35:02*amiconn would probably never have found it
23:35:11Bagder... that -Wundef would've helped ;-)
23:35:47amiconnAll my gapless mp3 files use the lame -nogap method, still playing gapless even with this bug
23:35:52 Join ender` [0] (
23:35:57Bagdernow friends, feel free to nuke a few warnings
23:36:31LearGah! A stray character sneaked it's way into a file. :(
23:36:59Bagderdid I add one?
23:37:40LearNo, I did...
23:37:56linuxstb_What should CONFIG_CHARGING be defined as for the ipods? The 5g defines it as CHARGING_MONITOR, but it's undefined everywhere else.
23:38:59LearI'm fixing the CONFIG_LED thing, if no-one beats me to it.
23:39:44amiconnCONFIG_LED is intentionally undefined in one case well as several other CONFIG_* definitions
23:40:30Mikachuyes, and when checking if they are you use #ifdef?
23:40:33amiconnAdding the #if defined() check probably is not what we want
23:40:44 Quit ender` (Read error: 131 (Connection reset by peer))
23:40:47Bagderamiconn: in what case ?
23:41:21amiconnThe CONFIG_* definitions are the 'multi-value' definitions
23:41:26LearWhat else to do then?
23:41:34linuxstb_CONFIG_RTC is the same IIRC.
23:41:45amiconnCONFIG_TUNER too
23:41:51Bagderamiconn: and they are even in fact one of the reasons -Wundef is very good for us
23:42:02Mikachuthen you have to define it to 0 in the undefined case
23:42:08Bagdersince we do a lot of code #if CONFIG_BLABLA == BLABLA
23:42:12amiconnIt can be set for the Samsung or the Philips, and it can be undefined to indicate no tuner
23:42:14preglowanyone care deeply about the decibel range for software bass/treble boost controls? if no one does, it'll end up as -24/24 in 1 db increments
23:42:41Bagderamiconn: then the first #if would be #if !defined(CONFIG_TUNER), right?
23:42:48amiconnIn fact, CONFIG_TUNER is slighly different as it's even a bitmask
23:43:09 Join ender` [0] (
23:43:20amiconnBagder: That's what I mean... lots of extra checks in the #if's
23:43:28Bagderyes, but good checks
23:43:43amiconnWe could also set them to an explicit 'absent' value
23:43:45Bagderwe have had more than one bug due to mistakes with this
23:43:48preglowi'm willing to do some extra checkery to not have to care about bugs like this
23:43:57amiconnThat'd require less extra checks
23:44:20Bagderamiconn: yes, that's another option that in my view is just as fine
23:44:23preglowok, -24 to 24 decibels it is
23:44:36amiconn+/- 24 dB is quite a lot
23:44:39preglowindeed it is
23:44:57*amiconn has seldom seen larger ranges that +/-12 or +/-15dB
23:45:06preglowthe eq has -24 to 24
23:45:14dan_apreglow: My mixer does +/-18dB in 0.5dB steps
23:45:30preglowdan_a: 0.5 db steps are somewhat unnecessary for this application, if you ask me
23:45:47preglowanyway, i'll just tack it on so people can try it out, then we can see
23:46:07amiconnHehe, the (newer) MAS can even do 1/8 dB steps for treble & bass
23:46:24preglowhow... useful
23:46:53 Quit Wiwie (Read error: 113 (No route to host))
23:47:28preglowamiconn: oh, and the cutoff frequencies for the controls should be fixed, yes? like it already is in hardware
23:47:36preglowthat'll simplify things a lot
23:47:46amiconnI think so
23:48:12preglowi don't see the point with anything else, if you want more control, use the eq
23:48:14 Join Mmmm [0] (
23:48:20preglowthat's built for more control and has better filters
23:48:55Mikachuand the point of these is they're faster?
23:49:04preglowyeah, slightly
23:49:18preglowtwo of these will be as fast as one eq filter
23:49:29preglowwhich means both bass and treble controls at the cost of just bass control with the eq
23:49:55preglowalthough i see it more as a familiarity thing. people who can't be bothered to learn how the eq works can still use these
23:50:32amiconnImho it's rather useful on ipods without hardware tone control, e.g. the mini
23:50:35preglowi might be able to make them even lower cost in the future, but for now, i'll take the route that doesn't need me to code a new filtering routine in asm
23:51:06BigBambiHurray! I just bought a F40 as backup to my H160 for £60 including shipping from the States!
23:51:25*BigBambi had to get on the Gigabeat bandwagon
23:53:17preglowthat's nothing
23:53:28BigBambii know :)
23:53:40Mikachuheh, i only find one store in sweden on a quick search, 180£
23:53:58BigBambiwell, this was from ebay
23:54:05BigBambibut for £60 I'll take the punt
23:54:08*amiconn wonders what currency this is
23:54:14BigBambigb pound
23:54:28BigBambi£60 is roughly 90 euroi
23:54:35amiconnI can only see strange character pairs here
23:54:44Mikachui don't even know what charset i have set for this channel now :)
23:55:01dan_aBigBambi: Nice price! Most of the ones I've seen have been about 90GBP
23:55:09preglowi see £
23:55:11preglowso i guess it's utf8
23:55:30pregloweverybody should use utf8, so change your settings!
23:55:48BigBambiYeah, I was pleased - I use my H1x0 all the time, but have been half looking for a gigabeat since everyone here seems to like them - for 60 gbp it is a nice backup
23:55:48LearHrm, I got complaints when I used that...
23:56:00preglowLear: you need to ignore the complaints
23:56:18BigBambipreglow, is the problem at my end or others?
23:56:24LearWell, it was from Linus, so it wasn't just anyone...
23:56:33BigBambiI have no idea what my system is set to
23:56:37*petur bought an 80GB disk and new battery for his h340 today
23:56:37preglowi told the few people i still speak to in norwegian to either switch or just read my gibberish for as long as they want, and lo and behold, now they use utf8 :>
23:56:40amiconnpreglow: Can you tell me how to set windows to utf-8? ;)
23:56:47preglowBigBambi: not your end, that's for sure
23:56:52preglowamiconn: your irc client should be able to do it
23:57:06amiconnThe client uses what the os uses. No separate setting
23:57:07BigBambiOK, cool - does this mean WinXP is correct by default :)
23:57:10dionoeahum, anyone with access to the svn accounts list ? (do i still have an account?)
23:57:13peturNettalk does it correctly
23:57:17preglowamiconn: really now, most clients can do it, though
23:57:31preglowamiconn: windows is unicode internally, just not utf8, as you probably now
23:57:41amiconnYes I know
23:58:50dionoeaBagder: ?
23:58:53Mikachuminus points for hydra irc, irssi and xchat can set charsets per channel :)

Previous day | Next day