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

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

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

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

#rockbox log for 2007-02-04

00:03:47 Join pseudoXh4 [0] (n=pseudo@
00:03:52 Quit Thundercloud__ (Connection timed out)
00:04:21 Quit Soul-Slayer (Read error: 131 (Connection reset by peer))
00:08:48pseudoXh4I have a *SOMEWHAT* unrelated question to Rockbox, but is anyone working on the Sansa E200 port here?
00:10:40Mikachui think they are
00:11:47 Quit dwihno ("It's actual proper DJ-juice!")
00:14:50 Quit hcs ("Download Gaim:")
00:15:04pseudoXh4Well.. would you happen to know what platform they use for the basis of their development?
00:17:12linuxstbRockbox is developed with gcc.
00:17:41pseudoXh4I mean.
00:17:47pseudoXh4What operating system?
00:17:59linuxstbRockbox is the operating system....
00:18:09pseudoXh4It's a PC operating system too?
00:18:16 Join hcs [0] (n=hcs@rockbox/contributor/hcs)
00:18:24linuxstbNo, but the PC operating system is irrelevant.
00:18:36linuxstbSome people use Windows, some Linux, some Mac OS X, some BSD...
00:19:11linuxstbBut under Windows you need Cygwin. i.e. any Unix-like environment.
00:20:17pseudoXh4I'm having problems with my Sansa on BSD, and wondered if any of the Rockbox developers happened to use BSD, as I get LUN0 locked on my screen when accessing my Sansa in recovery mode...
00:20:32 Part Llorean
00:21:32linuxstbSorry, I can't help with that (I don't have a Sansa, or BSD).
00:22:08 Quit fasmaie ()
00:23:02 Quit Sujdik ("Bersirc 2.2, for external use only. [ - Open Source IRC ]")
00:26:42markunI use FreeBSD, but no sansa
00:29:57 Quit lee-qid ("aufwiederbyebientotsayonara")
00:32:12 Join ben222222222222 [0] (i=4b510d4f@gateway/web/cgi-irc/
00:32:33 Join |AhIoRoS|_ [0] (n=ahioros@
00:32:41ben222222222222im having a problem with my sansa 200 player. can anyone help me
00:32:59pseudoXh4Same reason I came in. ;)
00:33:01 Quit |AhIoRoS| (Nick collision from services.)
00:33:08ben222222222222i think i might have bricked mine
00:33:13pseudoXh4:| :O :( :'(
00:33:19Mikachuyou're not the first
00:33:21pseudoXh4Checked the TroubleShooting page?
00:33:45 Quit |AhIoRoS|_ (Client Quit)
00:33:49ben222222222222i installed rockbox exactly how it says and now it says could not load image restart in recovery mode
00:33:58 Join bluebrother [0] (i=Zw7ajIXZ@rockbox/staff/bluebrother)
00:34:51preglowjhMikeS: any news?
00:34:51pseudoXh4Tried restoring your original firmware?
00:35:23ben222222222222i tried using a mi4 file from a sansa resource site
00:37:14ben222222222222i put it in in recovery mode and restarted it, but its still giving me the error and the file's no longer in the recovery mode partiton
00:38:47 Join webguest64erik [0] (i=51ea3d70@gateway/web/cgi-irc/ scared....
00:41:38 Join NanoUser [0] (i=3fc6232d@gateway/web/cgi-irc/
00:42:48NanoUserHi, Anyone know how to help me get Mpeg on my Nano (1st gen) ?? I'm kinda stuck, and my friend isn't a lot of help..
00:43:36NanoUserI mean I've read the manual, and didn't understand it much
00:43:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
00:43:50 Quit webguest64erik ("CGI:IRC (EOF)")
00:44:12 Quit linuxstb ("CGI:IRC (Ping timeout)")
00:44:14 Join ctaylorr [0] (
00:44:34SoapHave you read the mpegplayer wiki page?
00:46:19Soapok, what part specificly do you have a question on?
00:46:56 Quit Mouser_X (Read error: 110 (Connection timed out))
00:46:58 Quit ben222222222222 ("CGI:IRC (Ping timeout)")
00:47:06NanoUser*thinks* One second, because I don't know how to get Mpeg on there
00:47:26 Join linuxstb [0] (i=5343d4aa@gateway/web/cgi-irc/
00:47:35Soaphow do you get your music onto your Nano?
00:47:49NanoUserNormally yes
00:48:10Soapwell, you won't be able to use iTunes to transfer the mpeg files.
00:48:33SoapYou also won't be able to view mpeg movies from Database view, which I assume you are currently using?
00:48:39 Join linuxstb__ [0] (
00:49:00NanoUserWell I normally just use iTunes
00:49:11NanoUser(What is Database view?)
00:49:26MikachuNanoUser: do you actually use rockbox?
00:49:35 Join Mouser_X [0] (
00:49:36NanoUserYes it's on my nano
00:50:16 Join aliask [0] (
00:50:16 Quit NanoUser ("CGI:IRC (EOF)")
00:50:16linuxstbJust copy the .mpg file to your Nano as if your Nano was a hard disk (you may have to enable that option in itunes) and then select it from the Rockbox file browser.
00:50:28 Join NanoUser [0] (i=3fc6232d@gateway/web/cgi-irc/
00:50:45NanoUserBlasted Backspace key
00:51:17linuxstbJust copy the .mpg file to your Nano as if your Nano was a hard disk (you may have to enable that option in itunes) and then select it from the Rockbox file browser.
00:52:57NanoUserProblem: My Nano doesn't have the Mpeg player. Or does the Rockbox already install that?
00:53:20dionoeait installs it
00:53:24Soaphow old is your copy of rockbox, NanoUser ?
00:53:47DataGhostlinuxstb do you know if there is any page with information about what has been tried on the 2G nano (and failed)?
00:53:53DataGhostSo I'm not doing something that's already been tried
00:53:59NanoUserI got it off the current builds page
00:54:20Soapok, then you do have the mpegplayer, NanoUser
00:54:25Soapcd rockbox
00:54:25DataGhostI think I just found a valid FAT partition inside the nano 2G firmware
00:54:32DataGhostlooks like some resource thing
00:54:36DataGhostwhich does not give me much hope
00:54:51NanoUserIn terminal I would assume
00:55:06Soapno no, ignore my wrong window "cd rockbox"
00:55:23linuxstbDataGhost: The firmware partition looks very similar to the other ipods, apart from the main firmware (OSOS) image being encrypted. The IPL wiki has good descriptions of that.
00:55:40DataGhostyes I know that linuxstb :)
00:55:55DataGhostbut I need to know what's been figured out about that
00:56:00DataGhostlike that FAT partition
00:56:17SoapNanoUser: you do have the mpegplayer. It does not show up in the pluggins list, as is explained in the wiki page you said you read, because it is a viewer - simply "play" the mpeg file from the Rockbox file browser, making sure your "view" settings are set to "all" or "supported" not "music"
00:56:17DataGhostso I'm not doing useless work which has already been done :)
00:56:18linuxstbThat's what I mean - earlier ipods have a FAT image (5g and later versions of the Nano).
00:56:28 Join Thundercloud__ [0] (n=thunderc@
00:56:52DataGhostyes but I found a seemingly valid FAT16 partition (at least header) in the firmware image
00:56:55DataGhostnamed IPODRESOURC
00:57:09linuxstbYes... That's the RSRC image.
00:57:16NanoUser*looks for view page*
00:57:18DataGhostso that's been figured out already?
00:57:43SoapDataGhost: I just sold my nano 2nd gen today - don't go cracking it anytime soon! ;)
00:57:46linuxstbYes. It's the same as the 1st gen Nano - that's why I'm pointing you to existing documentation.
00:57:53DataGhostSoap I don't even own one :P
00:57:58DataGhostah ok
00:58:19DataGhostI'm probably looking at the wrong wikipage because it doesn't mention the rsrc at all
00:58:22 Quit Mouser_X (Read error: 104 (Connection reset by peer))
00:58:23linuxstbBut afaik, no-one has discovered anything interesting that's specific to the 2nd Gen Nano.
00:58:36 Join Mouser_X [0] (
00:58:37 Join dan_a [0] (n=dan-mirc@
00:59:19linuxstbThe site may be interesting to you though.
00:59:47DataGhostah well, I'm just quickly skimming through it, hoping to stumble across something
00:59:50NanoUserHow to I get to the View settings?
00:59:52 Quit Febs ("Chatzilla 0.9.77 [Firefox]")
01:00:00 Quit linuxstb_ (Read error: 110 (Connection timed out))
01:00:03DataGhostI also found a piece of crap which doesn't look encrypted but rather looks like actual code
01:00:11DataGhostbut I'm completely not sure about that
01:00:36DataGhostI'll just spend another 15 minutes looking at that firmware and then I'll stop, unless someone donates a nano :P
01:00:45SoapNanoUser: Menu−−>General Settings−−>File View−−>Show Files
01:01:00DataGhostit's not worth my money and time unless I have one, I'd better spend it working on the 5.5G
01:01:03SoapDataGhost: I sold mine for more than US retail....
01:01:12NanoUserThanks soap
01:01:22 Quit Mouser_X (Nick collision from services.)
01:01:31DataGhostthat smells like an investment
01:01:31 Join Mouser_X [0] (
01:01:35NanoUserOkay, now time to figure out how to put mpg files on rockbox....
01:01:56*linuxstb thought he mentioned that
01:02:11NanoUserOkay time to skim
01:02:28SoapDataGhost: the secret is to buy in the US - and sell on ebay in europe where they cost more.
01:03:39NanoUserOh btw how do u install skins?
01:04:05Soapthat is also covered in the manual NanoUser. They are called WPSs. While Playing Screens.
01:04:15DataGhostblame me for misreading
01:04:25DataGhostmisread a 0
01:04:34DataGhostand forgot to look at the address of that rsrc image
01:04:44NanoUserOkay, *galft WPS manual*
01:04:56NanoUsergalft = goes and looks for the
01:05:10DataGhostI thought the rsrc image was at 0x1016c0
01:05:16DataGhostit was at 0x161c000
01:05:20DataGhoststupid me
01:05:54NanoUserOkay dinner time, btw which should I look at WPS tags?
01:06:28DataGhostI just read it entirely backwards instead of bytewise :P
01:06:37***Saving seen data "./dancer.seen"
01:07:22 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
01:07:25 Join miepchen^schlaf [0] (
01:08:02DataGhostand that was aupd anyway. the partition is just named as if its the rsrc partition
01:10:38 Quit Thundercloud_ (Connection timed out)
01:11:33 Quit bluebrother ("Verlassend")
01:15:58 Quit pseudoXh4 (Remote closed the connection)
01:18:00 Join pseudoXhfour [0] (n=pseudo@
01:22:42preglowhrm, we really should have a fixed point lib or something
01:23:00preglowusing get_replaygain_int to perform exp10(x/20) doesn't feel all that... clean
01:23:52 Quit ender` (" If you're going to do something tonight that you'll be sorry for tomorrow morning, sleep late.")
01:27:06 Quit Bjoern-Erik (Read error: 104 (Connection reset by peer))
01:28:39 Quit ctaylorr (Read error: 110 (Connection timed out))
01:30:07 Join decayedcell [0] (
01:30:21 Quit hotwire_ ("Chatzilla 0.9.77 [Firefox]")
01:30:56*decayedcell suprised that Rockbox built on Linux is somewhat faster than Rockbox built on Cygwin
01:31:26preglowwhy surprised? cygwin is dog slow
01:31:39JdGordoncompiling? or running?
01:31:48decayedcellrunning on the actual iPod
01:31:59 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
01:32:00decayedcellI know its heaps faster in compiling, but its faster on the iPod !?
01:32:05preglowunlikely, unless you're running different gcc versions
01:32:31decayedcellcygwin GCC 4.0.2, Linux GCC 4.0.3
01:32:49preglowthen i think you're a victim of placebo :>
01:33:24decayedcellWell either that, or playlist scrolling suddenly wasn't as laggy
01:34:22preglowit really is quite unlikely that windows gcc should optimise better than linux gcc
01:34:36decayedcellno Linux GCC is faster
01:34:39preglowother way around, i mean
01:34:42 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
01:35:13NanoUserBack, yum Burgers
01:35:20*decayedcell shrugs
01:35:39NanoUserI have no idea how to do that. kewl
01:35:52decayedcelltype /me
01:36:09*NanoUser Runs across the chat room
01:36:20decayedcellignorance is bliss - if I think its faster, even if it might not be, then its all good :p
01:36:48linuxstbIt's not just the usual scroll speed varying with CPU speed?
01:36:55NanoUserStill tryin to figure out rockieboxie
01:37:27decayedcellWell, I remember on the last Cygwin compile that the scroll would lag and the CPU wouldn't boost to compensate
01:37:54decayedcellWith today's compile, the CPU seems to boost a bit and that makes it less laggy, it does still lag a bit but not as much
01:38:10linuxstbPlaying the same music?
01:38:40*decayedcell checks
01:39:31decayedcellAh nevermind still laggy... false alarm people :(
01:40:05 Join lini [0] (i=pugsley@
01:40:17decayedcellWhat I don't understand is why this doesn't happen with the 4G
01:40:22*Soap stables the calvary
01:40:52Soapstock 4G build?
01:41:12Soapeven if so the 4G doesn't have 1/2 the screen to update the 5G does.
01:41:17decayedcellIIRC yes
01:41:36decayedcellBut this is with exactly the same music
01:42:06linuxstbThe 5G's LCD needs 150KB of data to fill it.... The 4G greyscale is less than 10KB.
01:42:17decayedcellah okay that explains
01:43:20dan_apreglow: I saw in the logs a few days ago that you had some questions about the way I was planning to use the COP on the PortalPlayers
01:44:19decayedcelldan_a I can test the cop 8 patch on the iPod Nano, Mini 2G, 4G Color. Should be able to do it on Monday
01:44:28 Join shnee_ [0] (
01:45:08dan_adecayedcell: Thanks :) I don't think there's too much keeping it from being committed now
01:45:42 Join BigMac [0] (
01:45:49linuxstbIs it helping the CPU scaling problem on the PP5020?
01:45:58decayedcelldan_a Have there been any changes to iPodLoader2 to support Rockbox cop specifically?
01:46:22dan_alinuxstb: It appears to, on the 4G greyscale at least.
01:46:25linuxstbYes, I gave them a patch a month or so ago which was committed to the IPL SVN.
01:46:53linuxstbdan_a: No freezes, or just less freezes?
01:47:03 Quit Mouser_X (Nick collision from services.)
01:47:07dan_alinuxstb: Not a single one yet.
01:47:12 Join Mouser_X [0] (
01:47:28linuxstbIs that with any of the cpufreq patches, or simply the COP patch?
01:47:36dan_aCOP patch only
01:48:17linuxstbIs the CPU boosting frequently?
01:48:57dan_aI don't think that using the COP patch reduces boosting by much
01:51:01linuxstbSo is there anything holding you back from committing?
01:52:31dan_aIt breaks recording, there have been a few reports of the earlier versions not working, and I want to put in runtime detection of whether there is a kernel on the COP. I also want to try killing the main thread on the COP as per your suggestion.
01:52:39 Quit NanoUser ("CGI:IRC (EOF)")
01:53:33 Quit muesli__ ("ich will Kühe!!!")
01:56:49 Nick pseudoXhfour is now known as pseudoXh4 (n=pseudo@
01:57:18preglowdan_a: well, the question is pretty much answered
01:57:42preglowkernel on cop makes rockbox crash less with cpu freq scaling enabled?
01:59:39dan_aIt appears to - we're not sure why. Though MrH's clock handling document does mention checking to see if the COP is running or not, and sleeping it after changing the clock source if it was sleeping previously
02:00:37 Quit mattzz ("Leaving")
02:01:12 Join decayedcell_ [0] (
02:01:15 Join saratoga [0] (i=98039a62@gateway/web/cgi-irc/
02:01:33saratogado we have a theory about why the COP patch doesn't decrease boost much?
02:01:38preglowdan_a: really now
02:03:05Slasherisaratoga: audio decoding is not load balanced, so at least one core needs to be boosted as much as before
02:03:18Slasheribut ui can be run on separate core without lag
02:03:30saratogaso basically the COP just helps if you try and do more then pure audio decoding?
02:03:45decayedcell_Slasheri: would running the UI on the separate core help with scroll lag?
02:04:05saratogathe current setup is codec on one core, everything else on the other?
02:04:09Slasheridecayedcell_: with properly implemented COP support, it should totally eliminate that
02:04:10aliaskWould it be possible to split audio decoding to multiple threads to balance the load?
02:04:23*decayedcell_ jaw drops
02:04:26Slasheri(not not before i commit the patch to the gui list code)
02:05:47Mikachui don't think anyone has bothered to write a threaded mp3 decoder
02:05:59Slasherithat gui list code patch is almost commit ready, but i still have to figure out a few things if there is a better way to implement those
02:06:11BigMacAlright just checked the manual and still have no idea, what is cop, I haven't updaterd my build in a few months
02:06:30linuxstbSlasheri: Will that be separate to the ipod acceleration patch?
02:06:35saratogai don't think its possible to effectively thread an mp3 decoder since the spec is more or less serialized
02:06:45Slasherilinuxstb: yes, scrollwheel patch is a separate thing
02:06:47saratogaat least its not possible to thread an encoder, at least not without disableing various features
02:07:04pregloweven if threading decoding was possible: forget it
02:07:08Slasherigui list optimizations affects all targets
02:07:15preglowit would mean effectively rewriting our codecs from scratch
02:07:33saratogaso to be clear, we have the codec on one core, everything else on the other?
02:07:39saratogaor is it more complicated
02:07:46MikachuSlasheri: does that include the "dont redraw the whole screen when doing paged scrolling" patch?
02:08:01decayedcell_Mikachu which FS# is that?
02:08:09dan_asaratoga: At the moment, codec on one core
02:08:16Mikachudecayedcell_: don't know offhand
02:08:25BigMacSo I should proobably reinstall the bootloader on my 5g right?
02:08:27Slasheribut i am still not sure if that is a clean way to implement it: currently i have in scheduler a new routine queue_count that retuns the number of pending events in the queue. In gui list code i check if queue_count(&button_queue) > some_constant and then activate the optimizations
02:08:29Mikachuit doesn't apply to current svn anyway
02:08:38Slasherithat works fine, but might not be the best way to implement it
02:09:07SlasheriMikachu: yes, basically that is just that (improved version of it)
02:09:08preglowdoesn't sound like the cleanest thing ever, no
02:09:12preglowbut i don't have any better ideas
02:10:05 Quit decayedcell (Read error: 110 (Connection timed out))
02:10:13SlasheriMikachu: basically it just redraws the changed content of the screen, and when pending button events increases too much, we start dropping frames (not redraw screen every time)
02:10:19linuxstbBigMac: Does your current bootloader have white text on a black background?
02:10:36Mikachuah, i remember some talk about that
02:10:51Slasheripreglow: i also disabled the automatic paged scrolling, so now the cursor stays on top/bottom and it seems like the list is scrolling very fast
02:11:07Slasheriat those speeds one can't even notice that not every frame is not drawn
02:11:12decayedcell_ appears to be like Barry's common bootloader but for LCD scrolling?
02:12:05MikachuSlasheri: now you just have to draw that huge letter in the middle of the screen too and you're set
02:12:18preglowSlasheri: good, good
02:12:24preglowSlasheri: really looking forward to seeing this commited
02:12:43decayedcell_I think a lot of people are preglow :D
02:12:55Mikachui think just one person is preglow
02:13:08BigMaclinuxstb: nope
02:13:11Slasheripreglow: if you think that approach of calling queue_count from the gui code is ok, i could commit that quite soon
02:13:20Slasherii don't have either a better solution to do it
02:13:24linuxstbBigMac: Then yes, you should upgrade.
02:13:47preglowMikachu: there's always postglow, but he's not exactly trying to be preglow, i think
02:14:00BigMacOk, but is their a page that explains the changes and such and what has improved?
02:15:30BigMacOk well can someone explain why I would want a co processor boot if it works without it?
02:16:20dan_aBigMac: Because one day it won't
02:17:32 Join ctaylorr [0] (
02:17:37linuxstbThe latest bootloader is needed to initialise the cop properly. Without it, Rockbox won't be able to the COP.
02:18:03linuxstb^use the COP
02:18:31BigMacAh ok and will reinstalling the bootloader require deletion of all of my files on my ipod?
02:19:05linuxstbNo, but you should restore the backup of your firmware partition first, and then follow the installation instructions in the manual.
02:19:24DataGhostwell Soap, it's certainly working
02:19:48DataGhostiirc (forgot to note the time, stupid stupid... I know) I started the video playlist at around 6pm
02:19:55DataGhostnow it's 2:20am and it's still running
02:19:58BigMacI assume that is in the manual?
02:20:06BigMacthat was directed at linuxstb
02:20:09DataGhostand still charging
02:20:14 Join Thundercloud_ [0] (n=thunderc@
02:20:14linuxstbIs what in the manual?
02:20:35BigMacrestoring the firmware partition
02:21:41decayedcell_If you remove the bootloader that should be sufficient, thats covered in the manual
02:21:50linuxstbBigMac: I think so.
02:23:23 Quit jhMikeS (Nick collision from services.)
02:23:29 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:24:56 Join Soul-Slayer [0] (n=Jonno@
02:30:15preglowman, i almost wish this coef calculation was important, the way i'm shaving off stuff
02:30:18preglowkudos to me
02:30:18 Quit saratoga ("CGI:IRC (EOF)")
02:30:41BiptoNlinuxstb: have you tested the KoC8 patch on your 4g photo yet? i think it's about to hit svn
02:32:24 Quit wooo (Read error: 110 (Connection timed out))
02:32:42linuxstbBiptoN: No. Hasn't it already been tested on a Photo/Color though?
02:32:59dan_aBiptoN: I wouldn't say it was about to hit SVN... I want to be sure it won't hurt performance before I commit it, and one or two of the core devs have some questions about the way it works still
02:33:19BiptoNahh IC
02:33:43BiptoNhave the folks runnin it been using scaling do you know?
02:34:09dan_aI've been running it with scaling on a 4G greyscale
02:34:58BiptoNno problems ?
02:35:04BigMacbtw is the scroll acceleration part of svn now or still a no?
02:36:05dan_aBiptoN: In 4 sessions each of 45 minutes of playing, not a single problem.
02:36:34BiptoNnice, i havn't run it with the default 24/8 values
02:36:48BiptoNi ran it from 100% to death 6 times or more
02:37:01BiptoNlasted 6 hours each time
02:37:24BiptoN2 more hours than the battery was gettin prior
02:37:28 Join midkay__ [0] (
02:38:38dan_a2 hours better? Was that just with the KoCOP8 patch? Soap and Llorean (IIRC) did some tests with KoCOP6 and got about 15 minutes extra battery life
02:40:45 Quit appelsap ("( :: NoNameScript 4.02 :: )")
02:41:58BiptoNit was KoC6 and an svn a week ago
02:42:06BiptoNKoC8 yields the same also thoug
02:42:25dan_aOn which player?
02:42:44BiptoNdon't know how much of a difference it is with the 5,5g, i got 7hr 50min friday, runnin it non-stop all day long
02:42:53BiptoNi just bought the 5.5g
02:42:55 Quit My_Sic (Read error: 104 (Connection reset by peer))
02:43:25 Quit Thundercloud__ (Success)
02:44:16linuxstbdan_a: Have you thought about how rolo could be made to work with the cop? i.e. how to redirect it back to 0x0 at the same time as the main CPU?
02:45:09 Quit decayedcell_ (Read error: 110 (Connection timed out))
02:46:36preglowdoesn't have to be at the exact same time, does it?
02:46:56linuxstbI'm not sure.
02:47:00dan_alinuxstb: I've tried setting a flag in a variable held in IRAM to cause the COP to jump to 0x0, but that hasn't worked.
02:47:05Soul-SlayerI read in the FS for the KoCOP patch it was only supposed to provide a framework to allow for things to go onto the COP, what does it use on the COP now? Audio codecs?
02:47:30dan_aMy next test will be to use the mailboxes for passing a message that gets handled by an interrupt
02:48:22dan_aSoul-Slayer: There scope of that task has increased a bit! We're now running the audio codec thread on the COP (to prove that the framework is useful)
02:48:57Soul-SlayerAnd I understand it's been reasonably successful so far?
02:50:19dan_aNot as successful as we'd hoped, as the codec and audio threads seem to be lockstepped to some extent (i.e. moving one of them to its own processor doesn't make everything work faster.) But the concept works fine, and the MpegPlayer plugin successfully makes use of the COP
02:51:01Soul-SlayerProgress is progress though :).
02:51:38Soul-SlayerOn the topic of MpegPlayer - Does that still suffer from synchronization trouble?
02:51:46 Quit Mouser_X (Read error: 110 (Connection timed out))
02:52:36linuxstbIt depends what you mean by synchronization trouble. Video too fast or too slow, or the right speed, but out of sync?
02:53:02preglowdan_a: tried moving both?
02:53:15preglowwould probably need a sprinkling of proper mutexes to make it work
02:53:47pregloww1ll14m: there?
02:53:48Soul-SlayerI find it hard to tell whether the video is running at the right speed or not, but audio always tends to finish a lot faster than the video, so I imagine the video is running slightly slower.
02:53:49 Join decayedcell [0] (
02:54:14Soul-SlayerActually, now I think about it, I'm not sure which ones running faster, lol
02:54:15Mikachuand once the audio finishes, you can't open the menu to quit the plugin
02:54:32linuxstbSoul-Slayer: What player do you have?
02:54:37Soul-SlayerH10 6gb
02:54:58 Quit midkay_ (Read error: 110 (Connection timed out))
02:55:06dan_apreglow: I haven't yet. I'd need to work out what variables the audio thread has shared with other threads, because of the lack of any cache snooping between the cores
02:55:14linuxstbAre you running the kernel-on-cop patch with mpegplayer?
02:55:23Soul-SlayerI am, yes
02:55:34 Quit entheh ("^~")
02:55:39Soul-SlayerI've yet to try it without
02:56:23linuxstbThen the video is probably too fast - have you enabled the Limit FPS and Skip Frames options?
02:56:27Soul-SlayerI have an idea - I'll make a video counting up to 10 both graphically and using audio and see how they run with and without the patch
02:56:32Soul-SlayerI did, yes.
02:58:21BigMacWoah Woah Woah
02:58:33BigMacMpegplayer has audio now?
02:58:47linuxstbWhere have you been?
02:58:59BigMacI have been away for a while
02:59:00Soul-SlayerMid December it came out I believe
02:59:20BigMacSo wait what formats can be read?
02:59:25linuxstbBigMac: Read the PluginMpegplayer wiki page.
02:59:27BigMacI should prolly read the page first
02:59:33Soul-SlayerIndeed, lol.
02:59:35BigMacYou beat me to it
02:59:38linuxstbStill lots of work to do - but the status on that page is up to date.
02:59:57Soul-SlayerBut to answer your question - .mpg/.mpeg are the formats it can read
03:00:23 Part BigMac
03:00:29 Join TrueJournals [0] (
03:00:34 Quit linuxstb ("CGI:IRC")
03:00:49 Join BigMac [0] (
03:01:04BigMacWell congrats to all that made this possible
03:01:10tchan has anyone noticed the typo on the rockbox gigabeatfxport twiki page ? Namely, "Bootloader with bootsplpash", near the bottom of the page.
03:01:30tchanshouldn't it be "Bootloader with bootsplash"
03:01:43BigMacI remember a few months ago when everyone was proclaiming this as impossible
03:02:05 Nick linuxstb__ is now known as linuxstb (
03:02:14TrueJournalsBigMac: people were proclaiming what as impossible?
03:02:51BigMacVideo with audio
03:03:03BigMacNot exactly impossible but a while off
03:03:35TrueJournalsseems so long ago... :-p
03:03:41BigMacBecause we couldn't acsess the chip apple uses for h.264 (i believe the wolfenstein chip or something similar)
03:04:02linuxstbWe still don't, and mpegplayer isn't really for the 5g.
03:04:02BigMacwell any other major updates from about 3 months or so ago
03:04:22BigMacIt doesn't work on it?
03:04:45TrueJournalsno, mpegplayer works on it... but it doesn't use the iPod's chip
03:04:48decayedcellBigMac its a Broadcom BC something
03:04:52TrueJournalsit just uses the CPU to decode video
03:05:13*linuxstb thought BigMac was going to read the page...
03:05:14decayedcellWolfsen is the DAC, Wolfenstein is an uber old FPS :D
03:05:23BigMacWell why is it not really for the 5g
03:05:31linuxstbRead the page...
03:05:58linuxstb"NOTE: mpegplayer does all video and audio decoding using your device's main CPU. It does not use any special video decoding hardware such as the Broadcom Video Processor found in the ipod Video. mpegplayer therefore performs very badly on such devices in comparison with the manufacturer's original firmware."
03:06:42***Saving seen data "./dancer.seen"
03:06:52 Join Soul-Slaye1 [0] (n=Jonno@
03:07:41BigMacif FS #5755 allows for the fps to greatly increase why is it not included in the source
03:07:59dan_aBecause it's not ready yet
03:08:18dan_aBut I'm working on it
03:11:11Soul-Slaye1Does it currently increase performance at all?
03:11:38linuxstbYes, by a lot.
03:12:28TrueJournalsI think most people who use it see an increase
03:12:44TrueJournalsalthough, there are scattered reports of it actually giving the opposite effect
03:12:45linuxstbThe COP patch allows mpegplayer to run the video thread on the COP, and the audio thread on the main CPU. This gives playback speeds the same as when mpegplayer was just decoding video and no audio.
03:13:09linuxstbOr are we talking Rockbox in general?
03:13:16TrueJournalsI duno :-\
03:13:43TrueJournalsdan_a: Is it easy to move any thread to the COP, or is there a whole complicated coding process to go through?
03:13:46perl|perlwhat happened to the idea of Digger plugin?
03:13:55Soul-Slaye1Both, really, ANY increase in performance is a step forward in my opinion
03:14:00dan_aTrueJournals: It's sort of easy.
03:14:05TrueJournalsperl|perl: digger plugin?
03:14:14TrueJournalsdan_a: care to elaborate? :-p
03:14:17perl|perlyeah, the game Digger
03:14:22BigMacdoes elephants dream have sound in it because I tested it on the computer and it it playing with no sounds
03:14:24Soul-Slaye1Which threads would be of greatest benefit being on the COP?
03:14:46linuxstbBigMac: The old .m2v version had no sound, the new .mpg version does.
03:15:18BigMacOk I just downloaded it yesterday but I will again I guess
03:15:56dan_aTrueJournals: To move a thread over, you call create_thread_on_core(COP, ...) instead of create_thread(...) - but you need to make sure that any data that is shared across the two cores does not get cached by the processors, which either means careful flushing and invalidating of cache or using IRAM
03:16:29dan_aKoCOP8 doesn't work on iPod G3 :(
03:16:41TrueJournalsperl|perl: so digger is like digdug... but... like... not nintendo?
03:16:54TrueJournalsPretty fun...
03:17:04perl|perlyeah, it cries for rockbox :P
03:17:06BigMac <-that is what I downloaded...
03:17:07TrueJournalsanyway, I gotta go
03:17:10TrueJournalssee ya
03:17:12 Part TrueJournals
03:17:45linuxstbBigMac: Then yes, that has sound.
03:19:03Soul-Slaye1perl|perl: Don't forget Rockbox devs aren't interested in creating games from scratch... Adapting current GPL games is more whats aimed for I believe
03:19:29linuxstbLots of Rockbox games have been written from scratch.
03:19:42Soul-Slaye1In which case, ignore me.
03:19:44perl|perlthere are makefiles on that site
03:19:54linuxstbI think it depends on the dev - some people like porting, some like writing from scratch.
03:20:34BigMacI am playing in wmp and getting no audio let me try vlc
03:21:07decayedcellyou can also try Media Player Classic + ffdshow, that works quite well I believe
03:21:52BigMack it works in vlc so it just must be wmp issue
03:23:09BiptoNif i change "#define BOOTFILE "rockbox." BOOTFILE_EXT" to a different name and compile a new build will that new firmware look for the new filename i provide instead of rockbox. ?
03:23:11 Quit darkless (Read error: 113 (No route to host))
03:24:47 Quit Soul-Slayer (Read error: 110 (Connection timed out))
03:25:27BigMacvideo and audio are vastly out of sync is that supposed to be happening?
03:25:30 Join TrueJournals [0] (
03:31:11Soul-Slaye1Ok - To test MpegPlayer I've made a video that displays a number and 'beeps' when the number changes every 3 seconds... Should help me see which is out of sync
03:31:18linuxstbYes - the video is 24fps, but the 5g can't play it at that speed.
03:31:48linuxstbIf you enable the "Display FPS" option in the menu, it will tell you the speed.
03:32:07 Quit funky ("Changing server")
03:32:23 Quit rp- ("leaving")
03:33:03Soul-Slaye1Okay, audio DEFINITELY plays before video
03:33:50linuxstbSoul-Slaye1: What FPS does "Display FPS" display, and what FPS was your video encoded at?
03:34:01Soul-Slaye1That's with both Limit FPS and Skip Frames on
03:34:25BigMacAny builds out with dan_a's patch in it?
03:34:38linuxstbI think Senab's build includes it.
03:35:12Soul-Slaye1Encoded at 30, playing at 24 - I'll try at about 10 or so
03:35:25BigMacwoot 7.1 fps
03:36:20linuxstb320x176 should play at 14fps with the COP patch - so still too slow...
03:37:04BigMacAH well it was still cool to hear sudio and video
03:37:21BigMacguess I will have to check it out again in a few months
03:37:47linuxstbDon't expect much from mpegplayer on your 5g. But it works at full-speed on the Gigabeat...
03:37:49 Quit Brunellus (Remote closed the connection)
03:38:21decayedcellThats until someone figures out how to control the Broadcom chip right linuxstb :p
03:38:44SoapGPL games?
03:39:01linuxstbdecayedcell: Yes, but I haven't met anyone who has expressed any interest in doing that.
03:39:03BigMacOne day it will get up their I would hope
03:39:13SoapThere can be only one.
03:39:36linuxstbIt's easier to just spend $100 and buy a gigabeat...
03:39:46Soul-Slaye1Is there a way of getting VLC to display FPS?
03:40:49Soul-Slaye1Aha, found it
03:40:58BigMacNot for a student already in heaps of debt
03:41:12Soul-Slaye1Framerate is 10 before transcoding into an MPEG, I'll check on computer what it is after that
03:41:33SoapSell your iPod, buy a Gigabeat and 100 cases of Ramen noodles!
03:42:07BigMacI already thought of selling it but it is laser etched with my name so I don't think anyone would want it
03:42:14Soap10 is not a legal MPEG framerate. Most conversion programs will take 10fps sources and tripple up frames and encode to 30fps.
03:42:19BigMacAnd it is not in tip top shape anymoe
03:42:41Soul-Slaye1What framerate should I try then?
03:43:06Soaptry for what?
03:43:46BigMacI just noticed a bug in nsf and was wondering if anyone else has run across it, it like ceases and it does not change pbar
03:44:04Soul-Slaye1I'm testing to see the difference in time between Audio and Video on MpegPlayer on my H10 6gb, with and without KoCOP and to see whether Audio or Video run faster
03:44:05BigMacor time eft
03:44:09hcsBigMac: what game?
03:45:37linuxstbSoul-Slaye1: What is the FPS of the original video?
03:45:38BigMacyou meen the game title? because I am talking about nsf audio
03:45:52 Join kubiix [0] (
03:45:56Soul-Slaye1Whatever I like - I've made it from scratch purely to test with
03:46:08 Join funky [0] (n=repulse@unaffiliated/funky)
03:46:37Soul-Slaye1It just displays the numbers 1 through to 7 in three second intervals, and 'beeps' as it changes. I can set it to any FPS desired
03:47:09hcsBigMac: yes, I know, what game title specifically are you talking about? or do you notice this with any nsf?
03:47:47SoapSoul-Slaye1: set it to a legal mpeg framerate of your choosing then.
03:47:52 Part decayedcell
03:48:02Soul-Slaye1May I have an example?
03:48:31TrueJournals25 or 29.97 are legal mpeg framerates
03:48:46TrueJournalsno problem
03:48:57linuxstbmpegplayer supports 10, 12, 15, 23.9760, 24, 25, 29.9700 and 30
03:49:07linuxstbThe first 3 are unofficial.
03:49:15 Quit midkay__ ("Leaving")
03:49:51 Join decayedcell_ [0] (
03:50:18linuxstbSoul-Slaye1: Also, the bitrate you encode at will affect decoding speed. The lower the bitrate, the more FPS you can expect.
03:51:40TrueJournalsWhat would you say the best framerate to encode for rockbox would be?
03:52:13Soul-Slaye1TrueJournals: The GUI encoder you created - Do you know offhand what bitrate that transcodes to?
03:52:26TrueJournals400 I believe
03:52:38TrueJournals400 vb
03:52:44TrueJournalsaudio bitrate = 128 I think
03:52:50linuxstbTrueJournals: Depends on the target - see the table on the PluginMpegplayer page for the speed results with the elephants dream test video.
03:52:57TrueJournalslinuxstb: thanks
03:53:06Soapdepends on the source and your willingness to deal with reverse telecine. But I would go as low as possible - 24 if the source is progressive NTSC, 25 if PAL.
03:53:24TrueJournalsSoul-Slaye1: In some version sometime soon, I'm going to have an options dialog where you can set that
03:54:14TrueJournalsAlso, my GUI sets the framerate to 25 fps
03:54:25Soul-Slaye1If possible, make them generic to the player, that way people are less likely to get confused :p.
03:54:30 Quit phrozen77 (Read error: 110 (Connection timed out))
03:54:33linuxstbCan your GUI display the FPS of the input file?
03:55:06TrueJournalslinuxstb: not currently... I should keep track of feature request though...
03:55:33Soul-Slaye1My MpegPlayer does have a habit of crashing on 'Buffering...' also
03:56:04TrueJournalsI think that happens with everyone :-\
03:56:26Soapnever happened to me.
03:56:40TrueJournalsnevermind then
03:56:43Soul-Slaye1Happens after opening a few videos
03:56:53Soul-Slaye1First couple open fine, after that it just crashes
03:57:41TrueJournalssame here... what DAP do you have?
03:57:55Soul-Slaye1H10 6gb
03:58:12TrueJournalsmaybe it's H10-specific then...?
03:58:17TrueJournalsSoap: what DAP do you have?
03:58:25 Quit jhMikeS (Nick collision from services.)
03:58:25Soap5th gen iPod.
03:58:31 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:58:34TrueJournalsdo you use the KoCOP patch?
03:58:36Soapbut I never opened more than one in a row.
03:58:55TrueJournalsbut do you use the KoCOP patch?
03:59:09SoapI would open one, run it clear through, record the FPS, reset the player (to create a level playing ground) and test another encode.
03:59:24SoapMe and the KOP patch are best friends.
04:00:13TrueJournalsSo... one of two options: specific to KoCOP patch, or specific to opening one file after another
04:02:29 Join Mouser_X [0] (
04:02:52 Quit lubiix908 (Read error: 110 (Connection timed out))
04:04:08 Quit courtc (Read error: 104 (Connection reset by peer))
04:07:13Soap;topicseen#msg66913 is bordering on a golden quote.
04:07:37BigMachcs:any title
04:07:50 Join courtc [0] (
04:08:17hcsBigMac: so it just stops playing in the middle of a song?
04:08:36BigMacNo it plays but on the wps screen pbar does not move nor does the time
04:08:43dan_aSoap: LOL (or at least, grin a bit!)
04:09:19BigMacdan_a: do you have a 5g?
04:09:28 Quit spiorf (Remote closed the connection)
04:09:56SoapI think if he polished it a bit it would shine like gold. He has some great ideas in there - just not well developed. Ideas like Rockbox wanting to be like Microsoft. Ideas like ipodwizard being comporable to Rockbox. Great ideas, poor execution. ;)
04:09:58TrueJournalsSoap: I love how he compares iPodWizard to rockbox... seems like comparing apples to oranges there...
04:10:54dan_aBigMac: No. I only get iPods which have died off friends, and then fix them...
04:11:10BigMacWhos idea need polishing, mine?
04:11:36TrueJournalsI had to read that two or three times going... wait... wtf...
04:11:55BigMaceh lets call it over my head
04:12:19BigMacI am wiped like it is 3 am and it is only 10 pm
04:12:30linuxstbIt is 3am...
04:12:52BigMacno in EST
04:13:01dan_aWhen I get paid for working on Rockbox what a Microserf gets paid for working on MS software, I'll implement his teagames for him.
04:13:59TrueJournals[stupid question... linux n00b]How do I unmount... unmount command isn't working[/stupid question]
04:14:20TrueJournalsthat's it
04:14:23Soul-Slaye1umount /dev/hd##
04:14:27Soul-Slaye1Or whatever :p
04:14:47TrueJournalsthat worked
04:14:58Soul-Slaye1No problem.
04:15:14scorcheSoap: heh
04:16:10Soul-Slaye1"If someone at the IpodWizard forums ports you a Teagame to the IpodWizard OS, I will take up the challenge and port you TWO Teagames to the Rockbox OS." LOL
04:17:03 Quit funky ("leaving")
04:17:19BigMacSoap:Did you ever post that guide on mike a microphone throught the dock connector for the ipod / your preamp setup?
04:17:22*scorche would like to post something amusing in response, but nothing comes to mind
04:17:47TrueJournalsSoap: I think your quote is golden :-p
04:18:10SoapBigMac: Recording isn't 100% there yet, and I have been distracted, so I haven't finished my new "Uber" microphone preamp.
04:18:48SoapNot to mention the kernel-on-cop patch breaks recording and I love me some KoCOP more than recording on my iPod at this time.
04:19:27BigMacAh ok well make sure you post on the forums when you get around to it because that would be nice to get it working on mypod
04:20:06BigMacWhy do you "love you some KOCOP"?
04:20:53TrueJournalswho doesn't love KoCOP?
04:20:57Soapbecause it makes the UI much more responsive than it has been since the scheduler got reworked.
04:21:07*dan_a doesn't love KoCOP
04:21:19*TrueJournals thinks dan_a is in denial
04:21:24Soapbut we love dan_a FOR KoCOP.
04:21:46dan_aTrueJournals: Denial is an anagram of my name, so... yes!
04:21:53BigMacwhen did you last compile a build?
04:22:20Soaphours ago.
04:22:22BigMacor TrueJournals
04:22:40TrueJournalslike yesterday...
04:22:48BigMacAny chance you can upload it somewhere because I can never get cygwin to make zip
04:23:00TrueJournalsI'm trying to get a new build environment running perfectly (under VirtualBox)
04:23:03SoapAll my builds are too big for you.
04:23:14linuxstbdan_a: Does your Sansa ATA commit mean it now passes amiconn's test plugin?
04:23:17SoapTrueJournals: what is VirtualBox.
04:23:17dan_aBigMac: What do you need?
04:23:26dan_alinuxstb: Yes
04:23:40TrueJournalsSoap: VirtualBox is open-source virtualization software (like VMWare)
04:23:42BigMacJust a plain build with your KOCOP patch for my 5g
04:24:07TrueJournalsBigMac: too bad I have a H10 20GB...
04:24:08dan_alinuxstb: (though I'm just going to double check, 'cos it's 3am)
04:24:17dan_aBigMac: Give me 10 minutes
04:24:34SoapBigMac: forget cgywin - get the VMWare image.
04:24:36BigMacdan_a: Sweet deal. thank you
04:24:55BigMacSoap: Tried it before and never got it to work either
04:24:56TrueJournalsI agree with soap... definantly go for VMWare... WAY faster
04:25:03TrueJournalsthat sucks...
04:25:20Soapthen go coLinux. I think there are three of us who use it.
04:25:31BigMacIt usually ends up liike hours of work before all possible fixes are exhausted
04:25:46SoapI haven't updated it yet for building the manual - but I bet you don't care.
04:26:05BigMacNah I already have the pdf printed out
04:27:50BigMacSO colinux is good?
04:28:03BigMacI wish I could get ubuntu running on this new lappy
04:28:10Soaplike VMware - it is a way to run linux on your windows machine.
04:28:39SoapI have it set up as similar to the vmware image enviroment as possible.
04:29:31BigMacSo the KoCOP allows for like smoother menu scrolling or was I misintepreting you?
04:29:42 Quit toffe ("MegaIRC v3.27")
04:29:58 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
04:30:55Soul-Slaye1Which target is currently considered the best Rockbox-able target? Eg: Least problems, best performance etc
04:31:21TrueJournalsWhy does this question come up at least once a day?
04:31:30TrueJournalsWe should put that in the FAQ...
04:31:33Soul-Slaye1Lol :$
04:31:39Soul-Slaye1First time I've asked it or seen it :p
04:31:58SoapH300 in many regards.
04:31:58 Quit Mouser_X (Nick collision from services.)
04:32:05 Join Mouser_X [0] (
04:33:42BigMacSoap:That was directed towards you:)
04:34:44TrueJournalsBigMac: the KoCOP patch allows for a smother everythign
04:35:06TrueJournalsin general, rockbox seems to work better because some things (like audio codec) are run on the second core
04:35:11Soapyes, kernel on COP will help with menu scrolling - though it is no replacement for a proper scroll acceleration.
04:35:32*Soap 's dry skin has cleared up since KoCOP. Smooth!
04:37:04decayedcell_lol... what pH is KoCOP?
04:38:56BigMacmenu scrolling seems the same
04:39:12BigMacbut it probably is better in other aspects I guess
04:39:14 Join Thundercloud [0] (n=thunderc@
04:39:36decayedcell_BigMac to make it faster you'll need to apply a scaling patch
04:40:11BigMacdoh, wish I would have known that earlier
04:40:22dan_aSoap: That may true, but I've noticed that ever since I started work on dual core support the weather has been getting colder...
04:40:31BigMacI guess I will have to wait till next weekend when I can try compiling again
04:41:34Soapdecayedcell_: your patch which changes the basic processor speed?
04:41:35 Join Wiwie [0] (
04:41:50TrueJournalsgotta go... see ya...
04:41:51decayedcell_Soap I didn't make a patch, it's Williams
04:42:10 Part TrueJournals
04:42:22decayedcell_*nods*. Scrolling speed seems to be related to CPU frequency
04:42:38decayedcell_To prove this, you can try playing music then scroll and it *should* be faster
04:42:39Soapeverything is related to CPU frequency.
04:43:08decayedcell_So basically, Williams patch makes everything run faster, scrolling too. Nothing special ;p
04:43:12Soapbut that doesn't mean the solution to poor scrolling code is to clock the CPU faster. The solution is to improve the scrolling code.
04:43:55 Quit Wiwie (Client Quit)
04:44:01decayedcell_Yes I know, I wasn't talking about a complete fix, just a temporary thing to help alleviate the problem
04:44:16decayedcell_Hopefully Slasheri commits his GUI optimisations soon :)
04:45:14*decayedcell_ reboots into M$ windows
04:45:16 Part decayedcell_
04:45:59 Quit Mouser_X (Nick collision from services.)
04:46:08 Join Mouser_X [0] (
04:47:03BigMacalright I notice it now
04:47:12BigMacwhile a song is playing and I am scrolling
04:47:18BigMackudos dan_a
04:47:30 Join decayedcell [0] (
04:47:42*dan_a bows
04:52:30BigMacalright next task for you
04:53:09BigMacmake it so I can play doom, listen to music and switch between a window of a movie playing at full screen
04:53:13BigMacchop chip:P
04:54:25 Join tri170391 [0] (n=tri17039@
04:54:30dan_aOh, that's already done. Just edit firmware/export/config-ipodvideo.h and add the line "#define CAN_PERFORM_MIRACLES"
04:55:13BigMacANd now it works
04:55:22 Quit Thundercloud_ (Read error: 110 (Connection timed out))
04:55:44decayedcelldan_a lol......
04:56:27*dan_a goes to bed
04:56:48tri170391I'm using rockbox on ipod with COP 8 and Brightness 7 Patch, but when I play video with low Brightness, it usually turns off. Any solutions guys?
04:57:00decayedcellhow low? 4?
04:57:25 Nick BigMac is now known as Big_Nap (
04:57:37decayedcellbrightness 7 patch has problems with 4-1
04:58:02decayedcellbasically, you get flickering and/or the backlight doesn't stay as long and doesn't fade
04:58:02tri170391uhm I see
04:58:10 Quit Soul-Slaye1 (Read error: 104 (Connection reset by peer))
04:58:12decayedcellDoes 5 work?
04:58:28tri1703915 sometimes work, sometimes not
05:06:46***Saving seen data "./dancer.seen"
05:13:31 Join midkay [0] (n=midkay@rockbox/developer/midkay)
05:14:33 Quit Thundercloud (Connection timed out)
05:19:29 Quit tri170391 (" HydraIRC -> <- IRC with a difference")
05:19:45 Join tri170391 [0] (n=tri17039@
05:21:24 Quit tanq (Read error: 110 (Connection timed out))
05:24:53 Quit tri170391 (" HydraIRC -> <- Try something fresh")
05:28:01 Quit barrywardel1 (Read error: 113 (No route to host))
05:34:34 Quit shnee_ (Connection timed out)
05:34:48 Join shnee_ [0] (
05:36:49 Quit Mouser_X (Read error: 110 (Connection timed out))
05:36:52 Quit midkay ("Leaving")
05:38:53 Join phrozen77 [0] (
05:44:05 Join midkay [0] (n=midkay@rockbox/developer/midkay)
05:49:37 Join |AhIoRoS| [0] (n=ahioros@
05:51:07 Quit Nihilanth ("Leaving")
05:57:12 Join homielowe [0] (n=chatzill@
06:02:01 Quit midkay ("Leaving")
06:04:37 Join webguest56 [0] (i=47cdc81c@gateway/web/cgi-irc/
06:05:23webguest56how do you get the video to sync with audio in mpegplayer?
06:07:25Soapthere is an option to limit frames. This will only help sync a movie which plays back faster than realtime.
06:07:25 Quit webguest56 (Client Quit)
06:09:59 Join midkay [0] (n=midkay@rockbox/developer/midkay)
06:17:34JdGordonanyone know a simple voice recorder for linux to test if its line-in works?
06:19:03 Quit midkay ("Leaving")
06:23:52 Quit shnee_ (Connection timed out)
06:24:22 Join midkay [0] (n=midkay@rockbox/developer/midkay)
06:26:51 Part decayedcell
06:30:13 Join decayedcell_ [0] (
06:40:03 Quit rotator ("zzzzzzzzzzzz")
06:41:18 Quit |AhIoRoS| (Connection timed out)
06:42:31 Join topbloke [0] (
06:45:46 Join kubiixaka [0] (
06:48:38 Quit midgey ()
06:49:43nixphoeniJdGordon: arecord
06:49:49nixphoeniand aplay
06:52:23JdGordoni cant get my mic working :(
06:57:47 Quit topbloke ("bye")
06:58:15 Join webguest19 [0] (i=3ce6691a@gateway/web/cgi-irc/
06:58:37webguest19can someone please help me with my ipod?
06:59:43webguest19I have read the manual and the forums, but I am still in trouble
07:00:17JdGordon... then ask your question....
07:00:45webguest19I have an ipod 4g, which I loaded rockbox onto
07:01:14webguest19but there were a million little distractions around me and I unzipped the main file into the wrong directory
07:01:41 Quit webguest19 (Client Quit)
07:02:30 Join webguest19 [0] (i=3ce6691a@gateway/web/cgi-irc/
07:03:03 Quit kubiix (Read error: 110 (Connection timed out))
07:03:53webguest19Same ipod guy again, I have this problem, I stupidly unzipped to the wrong directory and now I have an "error -1'" message and it says "loading original firmware", is has been like that for over an hour
07:04:07webguest19my computer won't recognise it
07:04:25webguest19so i'm kinda stuffed, I need my ipod for my work
07:04:38JdGordonhold menu and select untill it resets
07:04:56webguest19I did, but it just went back to the same error screen
07:04:56JdGordonthen connect it up to the computer again and extract the .zip to the correct place
07:05:26JdGordonreset it while its plugged into the comp
07:05:29JdGordonit should load the usb
07:06:09webguest19so hold down menu and the middle button while plugged into the comp?
07:06:41 Join midgey [0] (
07:06:48JdGordonno... disconnect it.... hold down menu after you reset it untill it loads back into the apple os
07:06:49***Saving seen data "./dancer.seen"
07:07:12webguest19oh, thanks
07:08:00webguest19so how long do i hold menu down for?
07:08:21decayedcell_until it says Loading original firmware....
07:08:38decayedcell_otherwise press Menu then flick it to hold then it should load into the original firmware
07:09:06webguest19it says that, but it has been saying that for over an hour
07:09:22webguest19does it normally take a long time?
07:10:49decayedcell_takes 30 seconds tops
07:10:56webguest19oh dear
07:11:05webguest19so i'm stuffed then?
07:12:29decayedcell_press menu + select to restart, press select + play to get it into disk mode
07:15:35webguest19now it won't restart
07:15:54webguest19ah got it'
07:16:35webguest19ah its in disk mode, now what do I do?
07:16:49webguest19there is a tick and it says ok to disconnect'
07:17:55decayedcell_open iTunes and press Restore
07:17:56 Join Mouser_X [0] (
07:20:37webguest19sorry for my stupidity, where is restore?
07:21:25decayedcell_if you open iTunes, on the left click on your iPod. Then on the right theres a huge page and u press the little restore button. iTunes 7 this is
07:22:25webguest19cool, thanks
07:24:04webguest19you;ve saved me a ton of grief, I was about to pawn off some stuff to buy a new one
07:26:35 Join perldiver [0] (
07:27:53webguest19and this restores the ipod to the way it was?
07:29:11 Quit homielowe ("Chatzilla 0.9.77 [Firefox]")
07:33:06webguest19ok, you seem to know your ipods, how stable is rockbox for the ipod? Should I put it on again?
07:33:37decayedcell_Its very stable for the 5G series, I have a 5.5G
07:33:42decayedcell_what iPod do you have?
07:34:26decayedcell_some useful info:
07:38:38webguest19I only have a 4g 20GB
07:39:00decayedcell_Well for that to work stably you need to have a patched Rockbox
07:39:29webguest19which is what I didn't do before
07:39:50 Join nonroot [0] (
07:39:52dewdudedecayedcell_, are you running one patched to put the codec thread on cop?
07:40:28webguest19no, I just followed the instuctions in the manual
07:40:50decayedcell_dewdude yeah just patch Rockbox with koCop 8 and it should be stable on the 4G ipods
07:41:36webguest19ok, how do I do that, is it on the site?
07:42:47dewdudeoh, i'm not real huge on compiling my own....the build mine is based off is well over a month old
07:43:10decayedcell_you need to download the source code, then the patch from Flyspray and patch it. Then compile mm
07:43:39*dewdude 's head explodes
07:43:44 Quit perl|perl (Read error: 110 (Connection timed out))
07:43:47webguest19ok, also I just wanted to know a bit about rockbox, does it have similar features to ipod firmware?
07:43:59dewdudei don't have access to a linux box right now..
07:44:11dewdudeand i don't feel like making compiling work in windows
07:45:52webguest19one of the main features the ipod has it an alarm clock with choice of music
07:46:17decayedcell_I guess there is a patch for an alarm clock
07:46:26webguest19that other mp3s dont have, does rockbox have one?
07:46:42 Quit Mouser_X (Nick collision from services.)
07:46:51 Join Mouser_X [0] (
07:46:54dewdudesupport for just about every format used on the internet
07:47:06webguest19so you have to put patches on for all the features?
07:47:11dewdudeat least the non propritary ones
07:48:50webguest19their all on the website though>
07:49:27dewdudewith rockbox, you don't need to use itunes
07:49:59dewdudethat right there sold me.
07:50:46webguest19you just put thing into the hard drive? awesome!
07:51:13dewdudeumm...the music..yeah
07:51:19dewdudeyou just copy it with explorer.
07:51:33webguest19this just keeps getting better and better
07:52:27dewdudeit's not that difficult to install provided you follow the right instructions
07:54:02 Part hcs
07:54:49webguest19so unzip the files to the same directory that you work from, and it will work?
07:56:38dewdudeheh, i'm not really the kind of person to ask for install help...
07:56:42 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
08:00:12 Quit Mouser_X (Nick collision from services.)
08:00:20 Join Mouser_X [0] (
08:03:10 Quit midkay ("Leaving")
08:05:00 Quit ctaylorr (Read error: 110 (Connection timed out))
08:05:07 Join Araelius [0] (
08:05:19 Join midkay [0] (n=midkay@rockbox/developer/midkay)
08:05:24Araeliushello all
08:05:37Araeliusi had a question regarding installation
08:05:50Araeliusmight anyone be able to help?
08:06:08OverandYou shoul probably just ask the question- that's a good way to find out.
08:06:34Araeliusok, well the "ipodpatcher −−scan" doesn't work when i attempt to install
08:06:44Araeliuscommand i should say
08:07:02Araeliusit says it's an unknown option
08:07:44OverandI haven't touched that in a while, and I don't remember options off the top of my head. It's possible that there's been a change in that ipodpatcher program- I seem to remember reading something about that
08:07:54Overandyou may have a newer or older version than the docs expect
08:08:11Araeliusi see
08:08:25Overandjust running "ipodpatcher" or "ipodpatcher −−help" should present you with a list of commandline options
08:09:26 Join Jakeworld [0] (i=cf3e9c55@gateway/web/cgi-irc/
08:10:01 Quit nonroot ("BitchX: its magically delicious!")
08:10:30BiptoNaraelious: you need to sun as root
08:10:33BiptoNor sudo
08:10:38Araeliusthe "−−help" command does not work and the when i run "ipodpatcher" it says that DISKNO is the number windows has assigned to my ipod
08:10:58Overandsounds like you've got a screwy version
08:11:00Araeliusi am doing this via the windows command prompt
08:11:05Overandto be honest
08:11:27Overandbut I haven't touched the ipodpatcher stuff since i installed rockbox a number of weeks ago- i'm a user, not a developer
08:11:50Araelius:( I guess I could give it a go in Ubuntu
08:12:30Araeliusneed to re-install, this is going to take a while
08:13:20Araeliusthanks for the help anyways
08:14:23Overandther eare probably people her who can help you
08:14:28 Quit Jakeworld ("CGI:IRC (Ping timeout)")
08:14:34Overandbut unfrotunately i'm not one
08:15:04Araeliusguess i will stay in here just in case, thanks again Overand
08:16:44BiptoNdoesn't ubuntu boot from the cd?
08:16:52BiptoNsave ya some time ;)
08:16:55Overandgood pint
08:17:00Overandthe latest ubuntu discs are live CDs
08:17:05Overand... good point, even.
08:17:11OverandI think trying to drink ubuntu could be bad
08:22:08 Quit Mouser_X (Read error: 145 (Connection timed out))
08:31:41webguest19my ipod is a good as new, I just wanna thank you guys again
08:37:38 Quit courtc (
08:37:38 Quit linuxstb (
08:37:38 Quit JdGordon (
08:37:38 Quit BHSPitMonkey (
08:37:38 Quit bun-bun (
08:37:38 Quit crwl (
08:37:38 Quit Rob2222 (
08:37:38 Quit BiptoN (
08:37:38 Quit Strath (
08:37:38 Quit akaias (
08:37:38 Quit Bagder (
08:37:38 Quit tchan (
08:37:38 Quit pabs (
08:37:38 Quit crashd (
08:39:22NJoinRob2222 [0] (
08:41:41 Quit midgey ()
08:41:51 Join midgey [0] (
08:42:14NJoinStrath [0] (
08:42:49NJoinpabs [0] (
08:42:53NJoincourtc [0] (
08:43:11NJoincrashd [0] (
08:43:30NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
08:44:10decayedcell_Araelius you have to navigate to the directory with ipodpatcher in it to run it
08:44:31Araeliusi did
08:44:58decayedcell_right then in Command prompt
08:45:00Araeliusit runs, but it says that the −−scan command is not an option
08:45:08decayedcell_wat version of ipodpatcher is this
08:45:15decayedcell_webguest19 I can send you a copy of my 4G build i you like
08:45:18Araelius1 sec
08:46:09NJoincrwl [0] (
08:46:29decayedcell_Thats very old
08:46:42decayedcell_if you go to the ipodinstallbeta page you should be able to get the new one 0 0.5
08:46:52Araeliushmm got it from install instructions
08:47:01Araeliusok, will try
08:48:50 Join Rob222241 [0] (
08:49:00Araeliusi did use those files
08:49:40Araeliussorry, these
08:50:05decayedcell_I'm pretty sure its 0.5...
08:51:17Araeliusok, will try again
08:52:38Araeliusnow it's .5, weird
08:53:38Araeliusso far so good, thanks decay
08:53:55Araeliusi still have no idea how i could've gotten the .4 build
08:54:45NJoinakaias [0] (
08:55:09*decayedcell_ is puzzled too
08:55:44Araeliuslmao, the wonders of the internet, maybe i travelled back in time for a brief moment
08:55:49 Join Bagder [0] (
08:55:50***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
08:55:58Araeliusanyway it seems to be progressing fine, thanks agaom
08:57:17Araeliusagain rather
08:57:27NJoinBHSPitMonkey [0] (
08:57:33NJoinbun-bun [0] (
08:58:22 Quit Bagder (Killed by (Nick collision))
08:58:23 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
08:58:23NJoinBagder [0] (n=daniel@rockbox/developer/bagder)
08:58:46 Join Bagder_ [0] (
08:59:09 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
08:59:37Araeliusrockbox officially installed :)
09:00:22NJoinBiptoN [0] (
09:00:48 Part decayedcell_
09:01:06 Join decayedcell_ [0] (
09:03:22 Quit linuxstb (Connection timed out)
09:06:35 Quit Rob2222 (Read error: 110 (Connection timed out))
09:06:51***Saving seen data "./dancer.seen"
09:08:40 Part decayedcell_
09:08:45 Join Mouser_X [0] (
09:09:59 Quit Bagder (Connection timed out)
09:10:32 Join kaaloo [0] (
09:10:38 Quit Araelius ("Chatzilla 0.9.77 [Firefox]")
09:10:40 Part kaaloo
09:11:44 Join decayedcell [0] (
09:23:28 Join _Veseliq_ [0] (n=veseliq@
09:24:39 Part decayedcell
09:26:24 Join decayedcell_ [0] (
09:27:23amiconndan_a: around?
09:27:56decayedcell_I think he's asleep
09:30:10 Join rp- [0] (n=rp@
09:38:21 Quit Mouser_X (Read error: 110 (Connection timed out))
09:39:55 Join Llorean [0] (
09:42:40 Quit webguest19 ("CGI:IRC")
09:44:39Lloreandan_a: KoC_6 worked absolutely fine on Nano. I haven't tested KoC_8 but I believe it's not functionally different anyway. KoC_6 is where I got my runtime improvement, so if it's valid as a 'test' you can consider Nano G1 as 'tested'
09:45:30 Join Genre9mp3 [0] (n=yngwiejo@
09:49:25 Join Mouser_X [0] (
10:07:50 Quit kubiixaka ("Miranda IM! Smaller, Faster, Easier.")
10:07:59 Quit Mouser_X (Nick collision from services.)
10:08:08 Join Mouser_X [0] (
10:09:30 Join ackbahr [0] (
10:15:17 Join ender` [0] (n=ender@
10:16:51 Join n1s [0] (
10:25:27 Quit juxtap ()
10:26:13 Part decayedcell_
10:28:02 Join decayedcell [0] (
10:28:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
10:29:42 Quit ackbahr ("Chatzilla 0.9.75 [Firefox]")
10:50:42 Quit aliask (Read error: 104 (Connection reset by peer))
11:03:28 Join bluey- [0] (
11:06:52***Saving seen data "./dancer.seen"
11:13:47 Join Bjoern-Erik [0] (
11:15:58 Join _ff [0] (
11:20:21 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:21:57Learamiconn: Is your build server 64-bit? Wonder if that's why my recent commit cause warnings when building the simulator. Or maybe it is gcc 4.1.2 that is more picky.
11:25:26decayedcellGCC 4.1.2 has been released?
11:25:54LearDon't think so. Probably a snapshot...
11:26:49amiconnLear: Yes, my build server (and some others) are 64 bit
11:27:01 Join mathgl [0] (
11:27:50amiconnAnd you can't fix the warnings from before your commit
11:28:21amiconnThey are caused by a bad snapshot of gcc 4.1.2 which is installed on lostlogic's build server
11:28:45amiconnBlah, ignore that
11:28:52*amiconn didn't see that there are 2 commits
11:29:00LearYou mean the ones in synth.c? I ignore those, this one was for code I actually changed.
11:29:26 Quit Mouser_X (Nick collision from services.)
11:29:35 Join Mouser_X [0] (
11:30:23amiconnLear: I guess that your fix won't fix the warnings
11:30:37amiconnI think it has to do with the ? : operator
11:31:12LearPossibly, but int* was used in other places (like mpa.c), so I thought it was worth a try...
11:31:23amiconnIt probably probably produces an 'int' result - and casting that to a pointer (of whatever type) on lp64 platforms will throw this warning
11:32:00amiconnYou'll either need to double cast, or cast the 2nd and 3rd operand of ? :
11:32:39LearHrm, rather not try that with only the build servers to test with. :)
11:34:00amiconnci->configure(DSP_SET_STEREO_MODE, (int *) (intptr_t)(fc.channels == 1 ? STEREO_MONO : STEREO_NONINTERLEAVED));
11:34:29amiconnci->configure(DSP_SET_STEREO_MODE, (int *)(fc.channels == 1 ? (intptr_t)STEREO_MONO : (intptr_t)STEREO_NONINTERLEAVED));
11:34:53amiconn...or even
11:35:22amiconnci->configure(DSP_SET_STEREO_MODE, (int *)(fc.channels == 1 ? (int *)STEREO_MONO : (int *)STEREO_NONINTERLEAVED));
11:35:24 Join My_Sic [0] (
11:35:44amiconnci->configure(DSP_SET_STEREO_MODE, fc.channels == 1 ? (int *)STEREO_MONO : (int *)STEREO_NONINTERLEAVED);
11:36:03amiconnI'll try that later in case your fix doesn't work
11:36:31 Quit amiconn (" HydraIRC -> <- Leading Edge IRC")
11:43:00rp-dan_a: do you have the same brightness changing issues as described in the sansa thread? i can't reproduce that
11:43:36dan_arp-: I'm just about to test
11:44:50Lloreandan_a: Did you see my message re the Nano / Kernel on Cop?
11:45:50 Quit JdGordon (Read error: 110 (Connection timed out))
11:47:04dan_aLlorean: I did, thank you. At the moment I can't see any major showstoppers for getting it into SVN
11:49:29 Join linuxstb [0] (i=5343d4aa@gateway/web/cgi-irc/
11:49:37dan_arp-: I can't reproduce it.
11:49:44 Join entheh [0] (
11:50:19 Quit bluey- ("Leaving")
11:50:37 Quit Bagder_ (
11:50:46rp-dan_a: ok, thank you
11:51:06NJoinBagder_ [0] (
11:51:24Lloreandan_a: The only one of the "Known bugs" I see as being a significant problem (especially if KoC plus the multiplier change, or KoC on its own fixes the scaling freeze) is the Recording breaking, but since it broke playback after anyway, I'd always kinda assumed recording wasn't quite right anyway.
11:52:27dan_aLlorean: I think that will just be a matter of making sure certain variables don't get cached.
11:52:34 Join bluebrother [0] (i=dx2zkZw3@rockbox/staff/bluebrother)
11:53:37linuxstbdan_a: How about committing everything apart from the change to create_thread() to run the codec on the COP? It can then be tested more widely before actually moving the codec thread.
11:54:12Lloreanlinuxstb: Is there something about the codec thread move specifically that's unreliable?
11:55:00dan_alinuxstb: Good plan. I'd held off from doing that before now beacuse I didn't know what the hit on battery life would be, but of course we know now that that won't be too much
11:55:24rp-dan_a: is the battery state read-out done in adc-e200.c adc_scan() ?
11:55:31linuxstbLlorean: There's a potential for problems due to the cache. I just think it's a big change, so it may be useful to do it in stages.
11:55:57Lloreanlinuxstb: Makes good sense.
11:57:05 Join funky [0] (n=repulse@unaffiliated/funky)
11:57:05Lloreandan_a: You have a PP5020 target right?
11:57:24linuxstbBTW, the recording broken after playback is (I think) due to either the DAC or I2S not being re-initialised correctly after playback. I would expect the COP patch to be causing a different problem, but I can't immediately think what it could be.
11:57:36dan_aI think I'll do one last test version, which takes the main thread off the COP and adds some mailbox messaging (to see if the COP is available, and to tell the COP to reset to 0x0)
11:58:07dan_aLlorean: Yes, I've got a 4G greyscale (and two Samsung YH-820s, but they don't count)
11:59:10Lloreandan_a: Does coprocessor support seem to fix the scaling bugs? I've got a very hard time figuring out which people are talking about what regarding that issue, these days.
11:59:40linuxstbdan_a: I can't remember if anyone has done this test before (I think they have though), but could you try using the old ipod_fw installation method to install the current bootloader.bin on your 4G and see if the original firmware loads?
12:00:20dan_aLlorean: It does - I've had no lockups using KoCOP8
12:00:25linuxstbI still haven't found the reason why the new ipodpatcher isn't working on the older ipods (wrt loading the original firmware), and want to confirm the bootloader itself isn't the issue.
12:00:50dan_alinuxstb: I'll get myself a coffee then try that
12:01:11linuxstbThanks. It would be nice to fix ipodpatcher before everyone is forced to upgrade their bootloader.
12:01:14decayedcelllinuxstb on the 4G grayscale I tried patching with ROLO, but that failed loading the original firmware. iPodLoader 2.4 worked however =/
12:01:29Lloreandan_a: Alright. With all the craziness on the "Fixing scaling" task, and the fact that people there seemed to see no problem mixing and matching patches and calling it 'testing' I wanted someone I actually know to tell me something about their experience. :)
12:01:53linuxstbdecayedcell: Are you using "ROLO" to mean the Rockbox bootloader?
12:02:07decayedcelllinuxstb yes thats the official name isn't it
12:02:13Lloreandecayedcell: No, ROLO is not the bootloader.
12:02:22linuxstbNo, ROLO is the feature in Rockbox itself to start other firmwares.
12:02:23dan_alinuxstb: I want to make COP builds work (but without using the COP) with old bootloaders
12:02:28decayedcellokay then bootloader not ROLO
12:03:05decayedcellI had the same problem with my 5.5G, but restoring the firmware partition and patching again fixed it
12:04:04linuxstbdan_a: As long as that's not hard. In those cases, it could be nice for Rockbox to display a splash telling people they need to upgrade their bootloader.
12:04:39decayedcellUsing the rockbox bootloader on my friends 4G iPod Color seemed to work too =/
12:06:45 Join mattzz [0] (
12:08:41 Join himitsu [0] (n=himitsu@
12:09:07himitsuhi all
12:09:20w1ll14mmorning all
12:10:18 Quit Mouser_X (Nick collision from services.)
12:10:26 Join Mouser_X [0] (
12:12:47 Join dwihno [0] (n=dw@
12:12:59 Part dwihno
12:13:40 Quit linuxstb ("CGI:IRC")
12:15:24dan_alinuxstb_: Have the instructions changed following Barry's bootloader unification? How do I create an apple_os.ipod?
12:15:31 Join perplexity [0] (
12:18:07linuxstb_ipodpatcher [device] -rf apple_os.ipod
12:18:08Lloreandan_a: ipodpatcher N -rf apple_os.ipod
12:19:14linuxstb_dan_a: Is that for testing rolo?
12:20:01w1ll14mlinuxstb: your patch for usb detection seems to work well so far and charges ipod is confirmed
12:20:18hachiwhat does the -f mean?
12:20:21dan_aNo, I was just going through the bootloader changes (as a bootloader build now creates bootloader.ipod instead of bootloader.bin)
12:20:31hachioh read firmware, nevermind
12:21:25 Join wooo [0] (
12:21:36linuxstb_The bootloader.bin is still there, in build-dir/bootloader/
12:24:05hachiI need a windows binary of ipodpatcher that supports the -rb syntax, anyone know where I can get one?
12:24:36hachier, someone might be sending it to me... nevermind
12:24:57Lloreanhachi: If not, you could just read the manual, as it tells you where to get the most recent official binary
12:25:12 Join spiorf [0] (
12:25:32hachiI thought I did that, it spits the usage at me when I try the command
12:26:07Lloreanhachi: Well, there is no -rb
12:26:34 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
12:26:35LloreanBut if you have one that supports −−scan, it should as well support -rf
12:26:38hachiwell great... I don't know how to load loader2 then
12:26:41 Join lee-qid [0] (
12:26:58linuxstb_You can't use IPL's install instructions?
12:27:01hachithe loader2 docs say that the latest version of ipodpatcher handles -rb, and can put .bin files in the right place then
12:27:17hachioh for christs sake, why do you assume that I'm not reading the docs?
12:27:30dan_alinuxstb_: The problem appears to be with the new bootloader, not with ipodpatcher
12:27:37Lloreanhachi: Because you're asking in #Rockbox about a function the official ipodpatcher doesn't have?
12:27:39linuxstb_Sorry - I didn't expect loader2's docs to be wrong...
12:28:10linuxstb_Or pointing to a Rockbox installation tool...
12:28:26hachioops, I mean -ab
12:28:32hachiI've been typing it wrong
12:28:36hachi(to you)
12:28:56linuxstb_That's supported in the code in SVN, but that version of ipodpatcher hasn't been released as a binary yet.
12:30:01hachiright... okay
12:30:26linuxstb_dan_a: You installed the current bootloader.bin using the old ipod_fw method, and the original firmware doesn't start?
12:31:04dan_alinuxstb_: That's correct. I then put apple_os.ipod into the root of my 4G, and it still didn't start
12:31:52linuxstb_OK. The only common denominator seems to be a greyscale LCD - the bootloader works fine on the Color, Nano and Video. But people have reported problems on the 1st/2nd gen mini and 4g greyscale.
12:31:58Lloreandan_a: Did you extract it from an ipod that was in a condition that booted to the Apple OS?
12:32:17*n1s wonders why removing unused strings in enlish.lang sometimes increases binary size...
12:32:42JdGordonanyone know which file rockbox_browse() is in? this stupid ide doesnt have find in files!
12:32:47dan_aLlorean: I put the original bootpartition.bin back on, and confirmed that it booted straight into the OF before I started these tests
12:32:58Lloreandan_a: Okay, just making sure y'know. :)
12:33:05 Quit Lear ("Chatzilla 0.9.77 [Firefox]")
12:33:09*Mikachu doesn't understand how people can code without using ctags
12:33:09dan_aLlorean: Not a problem :D
12:33:19linuxstb_JdGordon: You're allowed to leave your IDE and use grep :)
12:33:30*dan_a googles for ctags
12:33:41MikachuJdGordon: ~line 1246 in tree.c
12:33:52*Mikachu just typed :tag rockbox_browse
12:33:59JdGordonlinuxstb_: grep wasnt co-operating :p and cheers Mikachu
12:34:57linuxstb_dan_a: I'm about to commit my work so far on ipodpatcher. I think it should work on the 3G now if you want to test.
12:35:28dan_alinuxstb_: I'll give it a try
12:36:02hachiif I try to compile ipodpatcher for windows from linux... is it gonna work? or should I just give up on that avenue now?
12:36:36hachicause I've gotta find disk space first... and I'd rather not be angry when I get to the end and find that doesn't actually work
12:36:37linuxstb_I compile it that way - the makefile is designed for it. Just type "make ipodpatcher.exe" (if you have the mingw compiler installed)
12:36:43Lloreanhachi: If you have access to a windows machine, why not compile on windows?
12:37:39hachibecause I've lost many many many hours trying to compile gaim for windows... I'm tired of trying to compile open source projects on windows... it never works
12:38:08hachiand then the developers just go 'oh well, works for me' and I'm right back where I started
12:38:25Mikachugaim developers are hardly representative of open source
12:38:37LloreanGaim compiling in windows is _not_ fun.
12:39:05linuxstb_hachi: ipodpatcher.exe is simple C with no external dependencies. You shouldn't have problems whereever you compile it.
12:39:10LloreanBut iPodPatcher is really easy to compile in windows. No cygwin or anything else is necessary.
12:40:30 Join robin0800 [0] (
12:42:28linuxstb_dan_a: New ipodpatcher committed.
12:42:36 Join anathema [0] (
12:42:54hachioh that is just plain nifty now
12:44:28linuxstb_What's nifty?
12:44:39hachirockbox booted :)
12:46:40linuxstb_w1ll14m: What charger were you using? USB or firewire?
12:48:56rp-is int 32 bit on arm architecture?
12:50:19rp-thank you
12:51:19decayedcellhmm its possible to write the kernel directly to the firmware partition?
12:51:22*decayedcell wow
12:52:07linuxstb_ipodpatcher [device] -wfb kernel.bin
12:52:46linuxstb_That wipes the Apple firmware and any bootloader you have installed, so you can only boot IPL.
12:53:09linuxstb_But should boot extremely fast...
12:53:31dan_alinuxstb_: Do you have a compiled Windows version of ipodpatcher? I don't have a windows compiler here
12:53:43linuxstb_One moment...
12:54:25dan_aThank you
12:55:55decayedcellThis fixes the overwriting of an existing bootpartition.bin too right linuxstb_?
12:56:43linuxstb_What do you mean?
12:56:56decayedcellSay you've done a -r bootpartition.bin
12:57:00linuxstb_Ah, the truncation thing?
12:57:12linuxstb_Yes, that fix was committed a few days ago.
12:59:02linuxstb_dan_a: One more test (on your 4G) could be useful - installing the old bootloader (convert to .ipod format with tools/scramble -add=ip4g bootloader.bin bootloader.ipod) with the latest ipodpatcher.
13:00:46 Quit wooo (Read error: 110 (Connection timed out))
13:00:50 Join amiconn [0] (
13:01:13 Quit Mouser_X (Read error: 110 (Connection timed out))
13:02:57dan_alinuxstb_: iPodPatcher and the newest bootloader work on the 3G. I'll try an old bootloader on the 4G now
13:04:16linuxstb_It loads both Rockbox and the original firmware OK?
13:05:37dan_alinuxstb_: Yes
13:06:54***Saving seen data "./dancer.seen"
13:12:37amiconnWhoa, x11 tunnelling is sloo-ooow
13:13:41dan_aamiconn: It depends what kind of network you've got. I've found it pretty quick as long as I'm going over a 10Mb or better link
13:14:13amiconnEffectively 384/192 kbps with quite some delay
13:14:27amiconn(2x adsl interleaved)
13:14:31dan_aOuch. You have my sympathy!
13:14:34Mikachuamiconn: you want to use either ssh -C or some sort of x compression program
13:15:03amiconn-C is enabled
13:15:14Mikachuwell, don't try without it then :)
13:15:43*amiconn 'll stop trying to tunnel x11
13:16:37amiconnIt's significantly worse than vnc, even worse than dameware
13:16:55Mikachuit's designed for low latency links
13:17:24amiconnIf I only knew how to make nx work :(
13:17:26perplexityamiconn, try NX if you really need to tunnel X. A bear to set up but I used it for ages on a 128k ISDN
13:17:30perplexitysnap :(
13:17:55amiconnNo packages available
13:18:13thegeek_compile from source :]
13:18:17perplexityBummer.. I was using Debian at the time so I compiled from source.. I recall it being a pain to get going tho
13:18:23 Join JustJohnny68 [0] (
13:18:56amiconnThis is debian-amd64
13:19:14thegeek_I did it on gentoo<3
13:19:24thegeek_and even then it was a bit of work to get working
13:20:01dan_alinuxstb_: New ipodpatcher / old bootloader does not boot the OF on a greyscale 4G. Rockbox boots correctly, though.
13:20:13amiconnI have to resort to a user-unfriendly console editor for now :(
13:20:18perplexityjust found my old nx server side script.. it's only 4 lines and not that ugly, but I don't recall how to start it on the remote machine now
13:20:49amiconnThe nx client is no problem, as this would be on a windows machine
13:21:41perplexity3 line paste coming..
13:21:45perplexityexport DISPLAY=nx/nx,link=modem,delta=1:8,accept=localhost
13:21:45perplexityexport LD_LIBRARY_PATH=/home/brad/nxcomp
13:21:45perplexitynxproxy/nxproxy -C
13:22:05perplexitythat was all I required to get it going on the linux end .. I never even installed the NX stuff, just put it in my home dir and compiled it up
13:22:17amiconnGah, source editing with vi - "fun" :(
13:22:31JdGordonanyone interested in discussing my menu patch which is ready for the first bit of its inclusion? or should i just hit the ml?
13:22:32Mikachusurely you have vim?
13:23:07JdGordonamiconn: use nano.. much easier
13:24:54 Join wooo [0] (
13:25:47amiconnJdGordon: Thanks... although, nano using utf-8 on a non-utf-8 terminal looks a bit strange
13:25:56 Join Wiwie [0] (
13:26:12amiconnBut the warning I wanted to fix is gone :)
13:26:23 Join inversions [0] (
13:30:28decayedcellnight all
13:30:29 Part decayedcell
13:32:09JdGordongrrr.. where is abs() hiding? :'(
13:32:36linuxstb_dan_a: So there seems to be two problems on the 4G? 1) The new bootloader not starting the OF with any install method, and the new ipodpatcher not starting the OF with either bootloader?
13:32:49 Quit anathema (Read error: 110 (Connection timed out))
13:34:15dan_alinuxstb_: That is correct.
13:36:06linuxstb_dan_a: Thanks for testing.
13:37:09dan_aI'll have a look at the code to the new bootloader later. Has caching been changed on the iPods at all?
13:37:12 Join muesli__ [0] (n=muesli_t@
13:38:03amiconnlinuxstb_: Hmm, what will I need to do in order to test this usb detection stuff?
13:38:27amiconnThe patch in fs says it's already a plugin (?)
13:38:52linuxstb_I made two patches - one a test plugin which you need to use with Barry's usb debug patch. The second is a full patch against SVN that changes the detection and reboots only with a PC connection.
13:39:03 Join pixelma [0] (
13:39:21linuxstb_The FS patch contains a general USB debug plugin. My plugin just attempted the detection.
13:39:29 Quit scorche (Read error: 54 (Connection reset by peer))
13:39:40linuxstb_(with code copied from the debug plugin)
13:39:49 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
13:43:32amiconnlinuxstb_: It seems your patch is not on fs...
13:44:15 Quit wooo (Read error: 110 (Connection timed out))
13:44:58linuxstb_amiconn: No, it isn't... (or usb_detect.c for my test plugin)
13:48:13amiconnOk, building...
13:48:38amiconnlinuxstb_: Why does the patch change the CONFIG_* identifiers for usbotg?
13:48:51amiconnJust adding one for USBOTG_ARC would be sufficient
13:49:02linuxstb_That was part of Barry's original patch.
13:49:10linuxstb_You mean the config.h changes?
13:49:47w1ll14many news about ipod nano 2g ?
13:49:47linuxstb_I noticed that, but it seemed sensible to me so I left it.
13:49:48amiconnThere's no functional difference, but I like those chip type related magic numbers...
13:50:29linuxstb_I don't feel strongly, so if you like them, I'm happy to leave them.
13:50:54linuxstb_w1ll14m: No change - i.e. no Rockbox developers are working on it.
13:50:55Lloreanlinuxstb_: Does your charger detection depend on an actual charger, or would it detect a powered USB hub not connected to a computer as a charger?
13:51:08Lloreanw1ll14m: Nobody is working on it.
13:51:14 Join TrueJournals [0] (
13:51:21linuxstb_Llorean: Good question. Can you test?
13:51:31Lloreanlinuxstb_: Point me to a patch, and I can.
13:51:40linuxstb_read back 7 minutes...
13:51:53LloreanGot it
13:52:00linuxstb_The usb_detect.diff should be applied against a clean SVN (or at least one without the USB debug patch applied).
13:52:41amiconnI can test this as well
13:53:16amiconnlinuxstb_: Hmm, did you also try it in the bootloader?
13:53:26Genre9mp3markun: thanks for the commit! :)
13:53:34markunGenre9mp3: np :)
13:54:07 Quit dpassen1 (Read error: 110 (Connection timed out))
13:55:07linuxstb_amiconn: Not yet, no.
13:55:26w1ll14mlinuxstb: ok thanx for the info
13:56:29markunw1ll14m: do you want to start a port to the nano 2g?
13:58:40Lloreanlinuxstb_: USB hubs will be detected as an actual USB detection. Makes sense, but good to know.
13:59:40linuxstb_I would have expected the opposite....
13:59:58linuxstb_Just to confirm, that's a USB hub not connected to a PC?
14:00:20LloreanIt's a very cheap little powered USB hub that I use to charge several devices.
14:00:32 Quit jhMikeS (Nick collision from services.)
14:00:33linuxstb_I guess we can live with that... It's the charger that's the painful bug.
14:00:37LloreanAnd I confirmed by plugging the iPod into a Griffin iPod charger, so the patch works.
14:00:38 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
14:00:46LloreanThe iPod charger didn't cause it to reboot
14:01:11 Join Arathis [0] (
14:03:44amiconnlinuxstb_: No reboot on charger, and reboot on usb connection, i.e. that works
14:04:01amiconnBut the charger isn't detected as usb power, and no charging animation either
14:04:52LloreanHm, but the Apple_OS sees the USB hub as a charger.
14:04:58LloreanSo that suggests something more could be done.
14:05:08linuxstb_Yes, I noticed that, and I can't understand why. Was it working correctly before (when you held MENU whilst inserting) ?
14:05:12linuxstb_(to amiconn)
14:05:39Lloreanlinuxstb_: Yes, at least on my Nano it did.
14:05:54 Join Thundercloud [0] (n=thunderc@
14:05:54amiconnThe usb power part still works when holding menu and connecting to real usb
14:06:03amiconnCharging anim doesn't, though
14:06:31*amiconn unpatches and rebuilds
14:06:48linuxstb_I noticed that on my Color, CONFIG_CHARGING wasn't defined.
14:07:41linuxstb_The charger_inserted() function is also only defined (in svn) for the Video (target/arm/ipod/power-ipod.c)
14:08:07amiconnHmm, perhaps an usb charger (and hopefully soon, firewire charger) should be counted as a 'real' charger
14:08:22amiconn(as opposed to usb power from a usb port)
14:08:29Lloreanamiconn: They should be, I think.
14:08:34LloreanIt makes sense, at least
14:08:48markundoes anyone object to me committing the rockboxed theme when it's ready?
14:08:48w1ll14mmarkun: i would like to, but i don't have a direct ipod nano 2g, but also i'm not very advanced in rockbox yet but i would like to give it a try :)
14:08:48linuxstb_There's no "real charger" for ipods, so yes.
14:08:49amiconnUsb chargers are usually built to deliver more than the standard 500mA
14:09:12amiconnMine does 1A
14:09:29markunw1ll14m: it will be a big challenge. A lot harder than the Gigabeat port I think.
14:09:30Lloreanamiconn: The one I have from Griffin is 500, but it was advertised for iPods but stated clearly it'd work for any USB device.
14:10:11pixelmamarkun: I don't like that he's switching off the status bar globally in the cfg-file...
14:11:07w1ll14mmarkun: are the chips marked with 'apple' confirmed which chips they supposed to be ?
14:11:15markunpixelma: but he doesn't have to, right?
14:11:40Genre9mp3pixelma: This can be changed quite easily if commited
14:11:51pixelmathat's what he wants
14:11:55JustJohnny68pixelma: I can take out that line for the SVN commited version
14:12:23markunis there a WPS tag to switch it off?
14:12:38pixelmayes, sure
14:13:00JustJohnny68I have the WPS tag set to off, so taking out the command in the .cfg isn't a problem
14:13:04amiconnlinuxstb_: The charging anim didn't work without the patch either. Strange, I never noticed that before...
14:13:23Genre9mp3markun: oops... a slight correction on the iCatcher commit... wps/UniCatcher/codec-wavpack-sc.bmp should be removed since there's no UniCatcher for H10 5GB
14:13:28linuxstb_It was the same on my Color - I never charged in Rockbox, so never noticed.
14:13:40pixelmamarkun, JustJohnny68: that's just my opinion though...
14:13:56amiconnSo the patch basically does everything it's supposed to do
14:14:03amiconnAny test targets left?
14:14:27markunpixelma: I agree. Maybe one day we can do some WPS like things with the menu and browser.
14:14:28linuxstb_It's been tested on the H10, Video, Color, and now the mini2g and Nano.
14:14:45markunGenre9mp3: just remove this file is enough?
14:14:46Genre9mp3JustJohnny68: You don't have to remove anything, if it gets commited the .cfg files are automatically generated by the wpsbuild script
14:14:46JustJohnny68The only complaints I've had so far are about that choice (about 5 different people so far)
14:14:51Genre9mp3markun: yes
14:15:11Genre9mp3markun: only from UniCatcher folder
14:15:23markunJustJohnny68: ok, don't worry about it then. Just send me the zip when you are ready.
14:15:37 Join Soul-Slayer [0] (n=Jonno@
14:15:37LloreanGenre9mp3: Status Bar in the menus isn't considered an aspect of the theme?
14:15:45Genre9mp3btw... speaking of the wpsbuild script, it should define the bg/fgcolours as well IMO
14:16:04linuxstb_Doesn't it define whatever is in the WPSLIST file?
14:16:09LloreanGenre9mp3: I've seen several themes that use a slightly different color area under the status bar, or some other form of highlight.
14:16:18Genre9mp3Llorean: what linuxstb_ just said
14:17:15markunGenre9mp3: done
14:17:30Genre9mp3The WPSLIST file should include bg/fgcolours as well IMO and the wpsbuild script should handle that
14:17:32*w1ll14m is thinking about buyin a nano2g 2gb
14:17:34Genre9mp3markun: thanks
14:18:21linuxstb_Genre9mp3: Yes, it should.
14:18:25LloreanGenre9mp3: Yes, FG and BG colors are quite important
14:18:42linuxstb_Also, afaik, it's not possible for a .cfg file to clear the backdrop.
14:18:51markunw1ll14m: very brave :)
14:18:56Genre9mp3linuxstb_: true
14:19:01 Join juxtap [0] (
14:19:07LloreanI still think, if a .cfg sets a backdground color, it should clear the backdrop image
14:19:38LloreanThen you'd just have to make sure that backdrop image comes after background color if a theme does both.
14:19:53Genre9mp3linuxstb_: IMO there should be a "backdrop: clear" or something setting for the .cfg
14:19:58JustJohnny68It would be nice to be able to clear the backdrop image. I've run into this problem switching themes. I only use bacground color.
14:20:05linuxstb_Or just "backdrop: "
14:20:21w1ll14mmarkun:lol 149 euro's ... i think i can mis it 15e of februari (when my work payed me:))
14:20:22LloreanBut what about themes that don't care about backdrop?
14:20:31linuxstb_They have to.
14:20:58Genre9mp3Llorean: Well, this can be handled by the WPSLIST file, too
14:21:00JustJohnny68How about "backdrop: false"
14:23:25 Join puppe [0] (
14:23:36Genre9mp3Llorean: I think though that not clearing the backdrop while setting a fgcolour may lead to problems even if the wps itself doesn't need that (eg. black fgcolour and black backdrop, even if the wps itself has its own background, the file browser/menu will be unreadable)
14:24:31 Quit thegeek_ (Remote closed the connection)
14:24:34 Quit DataGhost (Read error: 145 (Connection timed out))
14:25:09LloreanI'm trying to remember the exact situation I'd thought of where I didn't want it to clear, but it's not coming around any more
14:26:46Genre9mp3Llorean: wps that don't have background and the current backdrop doesn't mess with the fgcolour
14:27:03Genre9mp3but you can't always now if this will be the situation
14:27:40Genre9mp3JustJohnny68: Did you read my recommendation about using Nimbus-19 for big screens?
14:28:14Genre9mp3Llorean: btw, soon I'll be able to see how 6x12-13 looks like on Gigabeat :)
14:29:53LloreanGenre9mp3: It's readable by my standards, but I like tiny fonts.
14:29:54JustJohnny68Genre9mp3: Markun pointed this out to me. Nimbus 19 wasn't in last Mon/Tue daily font package, so I haven't tried it yet, but will shortly. I've updated to Nimbus for all but one device.
14:30:09LloreanGenre9mp3: I use 6x9 on the iPod Nano, if my memory serves on which I've selected.
14:30:38 Quit puppe ("Nettalk6 -")
14:31:05Genre9mp3Llorean: I like small fonts, too but I prefer nedore for that
14:31:21JdGordon(00:20:07) linuxstb_: Or just "backdrop: " <- this should work .. just looking at the code very breifly
14:32:01markunpixelma: I was trying to make a bold version of helvR12 with a script called 'mkbold' but unfortunately the result was not good enough.
14:32:30markunI hope I can find a better tool, if not maybe I should give it a go myself
14:34:51pixelmaI'd like to have some more fonts in "unusual" sizes... like 10 or 11 points. It just looks like it's going to be a lot of work
14:36:50JdGordonanyone good at drawing icons? We need one to show when the user in the a settings screen, in the menu or in the browser, my arty ability sux :p
14:37:13markunpixelma: dongs (yes, that anoying bastard) once wrote a program to generate bdf fonts from ttf which worked quite well.
14:37:13linuxstb_dan_a: Do you have a USB AC charger you can test with your 4G?
14:37:22 Part TrueJournals
14:37:31 Join TrueJournals [0] (
14:39:10 Join Mouser_X [0] (
14:40:54dan_alinuxstb_: I don't think so - I've got the 3G AC charger, which I think provides power over the firewire but not the USB
14:40:58Genre9mp3JdGordon: Just tested, "backdrop: " indeed clears the backdrop
14:42:11linuxstb_Genre9mp3: That's good. We should document that, and ask theme creators to use it - but only if they change the fg colour in their .cfg file.
14:42:24JdGordon... that means doing that for any of te filename settings will clear it
14:42:59Genre9mp3markun: I use otf2bdf for that
14:43:40linuxstb_Does anyone else get "M" when running "svn status" in the tools directory? I haven't modified the file...
14:43:58linuxstb_Seems to be something odd with the $Id:$ line
14:44:01JdGordonnot i
14:44:28JdGordon... but i did a svn up an hour or so ago
14:44:30markunGenre9mp3: I was not very impressed by the quality of the result, but perhaps it was just a problem with the fonts I tried.
14:44:30dan_aNot here
14:44:36linuxstb_I'm using svn 1.1.3 in Linux...
14:44:38Lloreanlinuxstb_: I'm of the opinion that all Theme authors should set FG color. Not setting it is asking for trouble, as that means you're assuming a default.
14:44:40 Quit w1ll14m (Read error: 104 (Connection reset by peer))
14:44:49amiconnlinuxstb_: 1.1.3 ??
14:44:58JdGordonlinuxstb_: 1.4.2 here
14:45:07*linuxstb_ should perhaps upgrade...
14:45:27Genre9mp3linuxstb_: I agree on this, We should add this to WPSLIST, as well along with the wps that specify colours
14:46:47Genre9mp3markun: Some fonts work well, others don't but generally the bigger the font size you set, the more chances are that the result will be good
14:46:58linuxstb_Problem fixed with 1.4.2...
14:47:16 Join w1ll14m [0] (
14:47:17markunGenre9mp3: also, his tool makes it possible to combine many different fonts to make one big unicode font
14:48:17Genre9mp3markun: What would you need more from Lucida Sans Unicode? :P
14:49:39markunno idea which ranges it includes
14:50:22 Quit JdGordon ("Leaving.")
14:53:57linuxstb_I've created a set of ipod bootloaders from current SVN - can people test them? There is also a new ipodpatcher win32 binary which should be used -
14:54:20linuxstb_Just upgrading the bootloader with "ipodpatcher [device] -a bootloader-ipodXXXX.ipod" should be working fine now.
14:54:47 Join DataGhost [0] (
14:55:35linuxstb_I don't think the new ipodpatcher fixes the 4g/mini Apple firmware issue, but it's possible...
14:56:30 Quit jaebird ("Leaving")
14:56:46 Join jaebird [0] (
15:03:31 Join decayedcell_ [0] (
15:05:36 Quit phrozen77 ("All generalizations are false.")
15:05:50 Quit Mouser_X (Nick collision from services.)
15:05:56 Join phrozen77 [0] (
15:05:59 Join Mouser_X [0] (
15:06:18 Quit robin0800 ("Never put off till tomorrow, what you can do the day after tomorrow")
15:06:56***Saving seen data "./dancer.seen"
15:07:15TrueJournalsIs the frequency scaling experiment patch still in flyspray? I can't find it..
15:07:26decayedcell_yeah its... lost...
15:07:41decayedcell_I was after the cpufreq3 patch =/
15:07:52decayedcell_i wonder where the closed FS go
15:07:57LloreanThey're still there.
15:08:16Lynx_Hi guys, did USB charging ever make it into CVS?
15:08:31decayedcell_ah change the display to All Statuses
15:08:51 Join robin0800 [0] (
15:12:26 Join pit_ [0] (
15:12:38 Part TrueJournals
15:14:14pit_sometimes my archos recorder v1 stops playing and changes from the wps into the browsing modus. i'd like to do an update but the website says thera are two errors in the daily builds.
15:14:31pit_which is the last good build?
15:14:32 Quit idnar (Connection timed out)
15:14:46 Quit DataGhost (Read error: 110 (Connection timed out))
15:15:09Lloreanpit_: Those two errors only relate to Rombox.
15:15:24pit_Llorean: ah, i'm gonna check it out ...
15:16:16pit_so it's better not putting it into the rom right now?
15:16:27pit_i mean, not flashing it?
15:16:31decayedcell_Build bootloader file
15:16:32decayedcell_find: apps: No such file or directory - is that meant to happen whilst compiling bootloader
15:16:47 Join idnar [0] (n=mithrand@unaffiliated/idnar)
15:17:02Lloreanpit_: You can still flash Rockbox, just not the Rombox version of it. It just means that th Rombox file isn't available at all.
15:17:18pit_now i got it. thx!
15:18:29pit_is the stop playing thing a common error with archos recorder?
15:18:50pit_got cvs20061210
15:19:38LloreanI haven't heard anyone report it before, but I don't know how many people regularly use an Archos Recorder still, let alone update very often.
15:20:20pit_poor old recorder ... :-(
15:20:36amiconnpit_: What build are you using
15:21:35 Quit Mouser_X (Nick collision from services.)
15:21:42 Join Mouser_X [0] (
15:22:17amiconnHmm, that already has the extra OFF button filtering.
15:22:22*mattzz loves his recorder!
15:22:38pit_mattzz: me too!
15:22:40 Quit phrozen77 ("All generalizations are false.")
15:22:47amiconnMaybe your batteries are worn out or the battery contacts have problems
15:23:10 Join phrozen77 [0] (
15:23:11pit_amiconn: i got 2300mah batteries - not so old
15:23:14amiconnThe v1 recorder have a special logic which is supposed to shut down the recorder on low battery voltage
15:23:36pit_it doesn't shut down, it just stops playing, leaves the wps
15:24:01amiconnThis logic is hooked in parallel to the OFF button, hence short voltage "ditches" below some threshold produce an OFF button event
15:24:10mattzzdepending on a very special mp3 file?
15:24:12pit_pushing the on button continues playing
15:24:23pit_mattzz: no
15:24:58amiconnThe button driver tries to prevent these events by extra filtering, but if you have bad battery contacts or bad batteries this won't help
15:25:21pit_ok, i'm going to check out other batteries ...
15:25:27amiconnI have 2700mAh cells which already have problems...
15:26:15pit_hopefully i haven't got problems with the contacts ...
15:32:10 Part decayedcell_
15:34:17 Join darkless [0] (
15:36:28 Join decayedcell [0] (
15:40:16 Part decayedcell
15:42:31 Quit pit_ ("Verlassend")
15:46:00 Join datachild [0] (
15:46:22 Join webguest20 [0] (i=485e425f@gateway/web/cgi-irc/
15:46:28preglowamiconn: you haven't seen a fast fixed point 1/x around, have you?
15:47:04preglowi'd like to avoid the 64 bit divides altogether, hrmph
15:47:39webguest20i tried to put rockbox on my sansa but it said no partition with white screen rockbox bootloader ?
15:47:57 Join Criamos [0] (
15:48:14 Part pixelma
15:48:30dan_awebguest20: Have you formatted your Sansa at some point? That happens if it has a FAT (not FAT32) partition on it
15:49:16amiconndan_a: Would it make sense to build the sansa build with fat16 support?
15:49:28webguest20no i just did the step to install rock after restart it said
15:49:59dan_aamiconn: I think it would - it seems that some of them are FAT16 formatted. I don't know if this is from the factory
15:50:27dan_awebguest20: I'll update it to support your Sansa
15:50:38webguest20how do u install rockbox on sansa e250 ?
15:51:03amiconnShould be simple, just #define HAVE_FAT16SUPPORT in config-e200.h
15:52:24 Part JustJohnny68
15:52:34 Quit Mouser_X (Nick collision from services.)
15:52:36webguest20 tried that doesnt work
15:52:37markunpreglow: what is the speed of a division compared to a multiplication?
15:52:42 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
15:52:43 Join Mouser_X [0] (
15:53:05Lloreanwebguest20: Well, remember, it is in early development. There are still plenty of bugs, which is why we don't recommend non-developers use it.
15:53:11webguest20dan_a work for sandisk ?
15:53:23preglowmarkun: depend son the platform, on arm, a mul is waaaay faster since there is no hardware divider
15:53:24Lloreanwebguest20: In your case, it's likely that the filesystem on your Sansa is not the one expected by Rockbox.
15:53:31dan_awebguest20: No, I'm a Rockbox developer
15:54:01preglowmarkun: and on coldfire a mul is still way faster, since we're talking 64 bit operations
15:54:02webguest20fat 32
15:54:23Lloreanwebguest20: You're 100% certain it's Fat32 and not Fat16? As in, you've gone and checked just now?
15:54:56markunpreglow: ouch, didn't know divisions were done in software on ARM
15:55:08webguest20when i put rockbox on my play it restarted my player then a white screen loading bootloader ?
15:56:22Lloreanwebguest20: Yes, you've already told us that. Though this time you left out mention of the "No Partition" message.
15:58:45markunpreglow: which part of the code are you looking at anyway?
15:58:47jhMikeSpreglow: What I have is this: if I use the actual progress of the src pointer to decrement length instead of input_size, no more clicks on a 1kHz sine wave...or anything else.
15:58:52preglowmarkun: eq.c right now
15:59:05preglowmarkun: nothing much, mainly just me honing my fixed point skills
15:59:18preglowalthough it will give faster eq coef calculations, which i doubt matters much
16:00:02 Join thegeek [0] (
16:00:23 Quit Arathis ("Bye, bye")
16:00:27 Quit webguest20 ("CGI:IRC (Ping timeout)")
16:00:36preglowsave a bit of space, too
16:01:06*markun enjoys the SPC codec
16:02:03preglowyou should :)
16:02:08preglowi think i'll stuff all these fixed point macros into dsp.h
16:02:14preglowthey're duplicated some places :/
16:04:04jhMikeSpreglow: ok, so you don't give a damn now? ;) I saw that you asked earlier.
16:04:25 Join webguest36 [0] (i=485e425f@gateway/web/cgi-irc/
16:04:35webguest36it said fat
16:06:20webguest36need help to install rockbox sansa e250
16:06:35preglowjhMikeS: bout what?
16:06:40Lloreanwebguest36: So it didn't say exactly "FAT32"?
16:06:54webguest36no just fat
16:07:12preglowjhMikeS: if about resampling, then hell yes i give a damn, but i can't be bothered to go playback.c diving right now :/
16:07:15Lloreanwebguest36: Then Rockbox won't work on yours yet.
16:07:28webguest36it needs fat32 ?
16:07:32jhMikeSpreglow: my getting rid of clicks...but don't know why it should work yet. see a few messages above
16:07:45Lloreanwebguest36: Yes.
16:08:08preglowjhMikeS: haha, i didn't see that :) reading now
16:08:09webguest36how to do that with out loose my things
16:08:22 Join webguest29 [0] (i=47cdc81c@gateway/web/cgi-irc/
16:08:33Lloreanwebguest36: Just wait for a version that works.
16:08:44preglowjhMikeS: well, it doesn't sound too strange to me
16:08:55preglowjhMikeS: sounds like an altogether safer way of doing things anyway
16:08:57webguest29how do i get mpegplayer video and audio to sync?
16:09:04Lloreanwebguest36: Rockbox on Sansa isn't ready for users yet anyway
16:09:05webguest36cnt wait any more
16:09:22Lloreanwebguest36: I'm 100% certain that you in fact _can_ wait, you just don't want to.
16:09:43Lloreanwebguest29: Turn on the framerate limiting option, and frameskipping. That's the best you can do right now
16:09:58webguest36is there a way to get the rockbox firmware in sana updater ?
16:10:30Lloreanwebguest36: No.
16:10:50webguest29Llorean: im suprised it got that far that you can actually play mpeg :)
16:11:01webguest36do you any other way to play games on sansa e250 ?
16:11:37Lloreanwebguest36: This channel isn't about the Sansa, it's about Rockbox. And Rockbox is about Audio far more than it's about games.
16:11:57Lloreanwebguest36: But I wouldn't be surprised if there's no other way to play games on it right now.
16:12:22jhMikeSpreglow: true, but it's odd since upsamling should be able to process exactly as many samples as are handed to it and at least one output for any input length.
16:12:41webguest36why is it so hard install rockbox ?
16:13:00Lloreanwebguest36: It's not released yet!
16:13:21webguest36when will it be released
16:13:37Lloreanwebguest36: Nobody knows, and nobody can guess.
16:13:58preglowjhMikeS: yes, it is weird that doesn't work too
16:13:59LloreanI'm not joking.
16:14:14preglowbut shrug, the other way does sound easier
16:14:39webguest36is there sounf on rockbox yet ?
16:14:50 Part Soul-Slayer
16:15:16Lloreanwebguest36: No.
16:15:20LloreanNot on the Sansa
16:15:34LloreanThere is on many other players, but we still don't have the knowledge needed for the Sansa.
16:15:36webguest364 get this im getin psp
16:15:40 Quit webguest36 ("CGI:IRC")
16:16:05preglowgood riddance
16:16:34webguest29Llorean: it still lags a good 10 secs behind
16:16:39LloreanThe problem is right now the *only* sort of person installing Rockbox on the Sansa are game players, because we don't have audio yet so the people who don't care about games haven't installed it yet.
16:16:50Lloreanwebguest29: Well then, that's the best you can hope for, for now.
16:17:22webguest29Llorean: have you gotten it to sync? im using a video ipod
16:18:27jhMikeSpreglow: I guess the next step is: wtf is going on? And the DSP should return what it really ate.
16:19:20Lloreanwebguest29: It syncs fine on my Nano, with the coprocessor patch, and low bitrate videos.
16:20:29webguest29Llorean: is there a coprocessor patch for ipod video?
16:20:40linuxstb_Why do we make Sansa builds available anyway?
16:20:56preglowfor the same reason we did with h1x0 back in the days?
16:21:44linuxstb_But Rockbox is much higher profile nowadays. Most recent ports haven't had daily builds until sound is working IIRC.
16:22:01preglowtrue enough, i guess
16:22:23Lloreanlinuxstb_: I think the Sansa was made available because other than sound, everything pretty much works and including it on the build table made it easy to see if something broke it.
16:22:32LloreanThough I don't think downloadable binaries should've been linked.
16:22:41preglowthat can be changed :)
16:22:41 Quit Mouser_X (Nick collision from services.)
16:22:52 Join Mouser_X [0] (
16:22:54linuxstb_Exactly - things can be in the build table without being downloaded (e.g. the bootloaders, sims,...)
16:22:59LloreanIf I understand correctly, there's even _one_ person out there who has sound, but due to the NDAs he can't share the code.
16:23:39Lloreanlinuxstb_: Which's what should've happened with the Sansa. The Gigabeat was like that for a while, wasn't it?
16:24:31bluebrotherthe download link to go to a page that tells it isn't ready yet and anyone who wants to try needs to compile himself
16:24:34rp-Llorean: is that one maybe Mrh? :)
16:24:44Lloreanrp-: I actually don't think it is.
16:24:48LloreanI'm not really sure though.
16:24:55LloreanI just remember once seeing/hearing mention of it.
16:25:17jhMikeSAny restrictions on sharing a compiled binary?
16:25:26rp-damn NDA's :/
16:25:45 Quit phrozen77 (Remote closed the connection)
16:25:52preglowis this the austrianroflsystems chip?
16:26:11 Join phrozen77 [0] (
16:26:11preglowisn't that getting solved anyway?
16:26:14LloreanYeah, it is.
16:26:14linuxstb_Yes, well their chip embedded inside the PP5024.
16:26:16tchanLlorean, linuxstb_: I took your excellent advice and if the ebayer comes through, I should have my new(used) gigabeat F40 soon. The ipod 4g grayscale was just not working very well for me.
16:26:39Lloreanpreglow: It's supposed to be being solved, but you never know how that turns out until it's done.
16:26:45jhMikeSI'm asking if that person can distribute their compiled version of rockbox with sound...noone knows?
16:26:45preglowLlorean: deed
16:26:55webguest29any hope of the hd elephants dream file playing on my video ipod? hehe
16:26:56LloreanjhMikeS: They can't, because it's GPLed.
16:27:10preglowalso, it depends on the nda
16:27:12LloreanjhMikeS: If they distributed the binary, they'd have to provide source upon request. They can't provide source, so they can't distribute the binary.
16:27:16preglowbut the gpl issue is valid, of course
16:27:58*jhMikeS is looking for a loophole
16:28:02Lloreanwebguest29: Video still needs an awful lot of optimization.
16:28:23LloreanjhMikeS: Well, we still have Austrian whatever offering to help us.
16:28:32linuxstb_That's assuming MrH even got audio playback working within Rockbox - maybe he just did his own tests, outside the playback code.
16:28:35LloreanjhMikeS: Not to mention the Sansa _emulator_ that could potentially let us learn a few things
16:28:39Soapso how "nudge-nudge, wink-wink" can you be and still avoid violating a NDA?
16:28:54webguest29Llorean: well before they said it wouldnt even be possible so i was suprised to see there is actually something that doesnt work too bad considering
16:29:11Lloreanwebguest29: I'm surprised anyone said Video in Rockbox wouldn't be possible
16:29:21 Quit phrozen77 (Remote closed the connection)
16:29:30jhMikeSLlorean: That's an approach I was thinking of taking with the Ali chip on x5 but maybe was overkill for that one.
16:29:39 Join phrozen77 [0] (
16:30:04webguest29Llorean: I think more specifically this model, ipod video
16:30:15 Quit lee-qid (Read error: 145 (Connection timed out))
16:30:40Lloreanwebguest29: Well the general consensus for the 5G has pretty much been "It's gonna be slower than the others, because the screen is big and we can't use the broadcom chip"
16:30:48jhMikeSparadoxically free software is blocking free software
16:31:14 Join tucoz [0] (i=480ec035@gateway/web/cgi-irc/
16:31:35LloreanjhMikeS: Well, AMS seems amenable to an alternate arrangement to their standard NDA, though I don't believe they've told us what it is yet.
16:32:07webguest29Llorean: why cant they use the chip? Couldnt they use a wrapper?
16:32:35*linuxstb_ repeats his request from about 90 minutes ago for people to try out the new ipod bootloaders and ipodpatcher before he makes a new release.
16:32:40Lloreanwebguest29: I'm not even sure how you mean that in this context. The hardware is completely undocumented so we can't use any of its decoding features.
16:32:43rp-i'm still waiting of a message of austriancoder that he got the datasheet :>
16:32:43Soapthe "alternate arrangement" is to lure all the unsuspecting Rockbox devs to their mountain castle lair, and lock them in the dungeon before the port sees the light of day!!!!
16:32:54LloreanSoap: That seems likely.
16:33:09Lloreanlinuxstb_: Do you need it tested on Nano, since the old everything worked fine anyway?
16:33:26linuxstb_Llorean: Yes please - just to confirm everything is still fine.
16:33:36tucozIsn't it possible to distribute a compiled version of rockbox with the sansa sound driver, and provide the rockbox source on request. Isn't this compliant with the GPL?
16:33:39webguest29Llorean: it seems like someone would figure out how it works though eventually
16:34:04bluebrothertucoz, that is ... but you need to distribute the sources for the audio stuff too
16:34:10Lloreanlinuxstb_: I should be able to just -wf apple_os.ipod then -a the bootloader, right?
16:34:17bluebrotherand that in turn is forbidden by the NDA.
16:34:33Lloreanwebguest29: Eventually that may be true, but it's not going to be for a long time. Since that only affects ONE device, nobody's focusing on it specifically
16:34:49tucozhmm. i thought you only had to distribute the parts that are gpl, not your own work
16:34:52Lloreanbluebrother: I think he's referring to the "binary blob" method
16:35:04bluebrotherbut how should that work?
16:35:22bluebrotherit would need a binary blob that's not part of Rockbox but gets loaded dynamically
16:35:27tucozsimilar to the gigabeat for instance
16:35:28linuxstb_Llorean: The most important is just to try "-a bootloader" with your existing bootloader installed, to test if the upgrading is working. If you have time, any other tests you can think of would also be useful.
16:35:41bluebrotherhow does it work on the GB?
16:35:43tucozthey use linux which is gpl, but some parts are not
16:35:46Lloreanlinuxstb_: Well, right now my bootloader was -wf'ed
16:35:58linuxstb_runtime linking doesn't remove the need for all code to be GPL'd.
16:36:06jhMikeSChange the rockbox licensing subltely for certain situation like this?
16:36:08webguest29Llorean: i was thinking like a ndiswrapper for linux
16:36:12tucozand they only have to provide the linux sources
16:36:26tucoz(i think)
16:36:29Lloreanwebguest29: Someone would have to at least figure out how to use their code then, still
16:36:29linuxstb_tucoz: That's because Linus doesn't enforce that aspect of the GPL.
16:36:37linuxstb_(the other Linus...)
16:36:39tucozaha. ok
16:37:10tucozthanks for clearing that out. so linux is a special case then.
16:37:26linuxstb_But that assumes that both a) Mr H has a working audio driver (not just test code); and b) He wants to release it in any form, which AFAIK he doesn't.
16:38:39preglowlinuxstb_: svn ipodpatcher is what i should use?
16:38:45linuxstb_Yes please.
16:38:45tucozit's a bad solution anyway. let's keep our fingers crossed that the NDA issues gets ironed out then.
16:39:37linuxstb_The new bootloaders include the changes to the error messages, so no more "what does error -1 mean?" questions...
16:39:42Lloreanlinuxstb_: Everything seems to work fine
16:40:38webguest29ooo i just found i good combination for cleaning my ipod
16:40:46preglowlinuxstb_: looks like all's good
16:41:02Soapunless the error messages are hyperlinks to magic genies which fix things, I have a feeling they won't entirely stop the questions.
16:41:07 Quit bluebrother ("reboot time")
16:41:12linuxstb_Llorean: I guess you've read the commit messages, but ipodpatcher should now be more reliable when upgrading an existing ipod - the previous version didn't work when I upgraded an old ipod_fw installed bootloader on my 5g, but the new version passes that test.
16:41:20webguest29about 50% isopropyl alcohol and 50% distilled water
16:41:50preglowany eq users around keen on checking out new eq code?
16:41:53Lloreanlinuxstb_: Yeah, I saw. I tried putting a bootloader on (the first Hold-switch version) and then updating it to the one in your .zip, and had no problems.
16:42:05jhMikeSpreglow: sure, I always use it
16:42:30preglowjhMikeS: patch'll be fine with you? i'll fix you up soon, just need to go heat this place
16:42:47jhMikeSpreglow: yeah, sounds good
16:43:22linuxstb_Soap: I agree, but at least "file not found" is a better clue than "error: -1".
16:43:47linuxstb_(it also shows the name of the file it hasn't found)
16:45:13 Quit tucoz ("CGI:IRC (EOF)")
16:45:49webguest29errors should be organized based on the dewey decimal system . . . like error 918.04 would mean that it is an error due from traveling for too long in the US
16:45:53linuxstb_preglow: Did you test on 64-bit linux?
16:46:05linuxstb_(test ipodpatcher)
16:46:43 Join mattzz_ [0] (
16:51:08 Join CriamosAndy [0] (
16:51:18dan_alinuxstb_: New bootloader/ipodpatcher works on G4 - although booting the OF still doesn't work. I'll test the G3 after I've done a bit more housework
16:53:15linuxstb_dan_a: Thanks.
16:54:11 Quit webguest29 ("CGI:IRC (EOF)")
16:58:09 Quit Criamos (Read error: 145 (Connection timed out))
17:00:50linuxstb_dan_a: I'm just about to test kernel_on_cop_8 on my Color, without the codec thread on the COP, to see if it prevents freezing.
17:01:43dan_alinuxstb_: My suspicion is that it won't, because the COP will be sleeping most of the time.
17:02:28Big_Napalright anyone recently start coding for rockbox or remembers what book's / tutorials they used to learn c and arm
17:03:21 Quit mattzz (Read error: 110 (Connection timed out))
17:05:22linuxstb_dan_a: It boots at least... I'll let you know.
17:05:33jhMikeSBig_Nap: my tutorial is usually to stare at it 'till I get it :)
17:05:44dan_aBig_Nap: I used other people's code. Take what works already, and modify it to work slightly differently. Also, a copy of the ARM ASM instruction set is handy
17:06:07*linuxstb_ uses dan_a's approach.
17:06:47linuxstb_I would recommend reading some of the Rockbox code (maybe start with some of the plugins), and see if you can follow what's going on.
17:07:00***Saving seen data "./dancer.seen"
17:07:13 Nick Big_Nap is now known as Big_Mac (
17:07:30Big_MacOk so I I have to get the source first right?
17:07:37 Join lee-qid [0] (
17:08:02*jhMikeS just changed the x5 display driver around for the first go at changing things
17:08:34Big_MacYou mean just like a line that is displayed
17:08:35 Join mattzz__ [0] (
17:08:51Big_Macor the driver that is used to connect to the display
17:08:57 Join toffe [0] (
17:09:03preglowlinuxstb_: yes
17:09:05jhMikeSBig_Mac: the code that writes data to the display
17:09:06Big_Macor neither
17:09:38dan_aBig_Mac: It helps if you know what you want to do to start with.
17:09:59Big_MacDo you think changing the button you press to go into apple os would be to large?
17:10:14jhMikeSindeed, I did know what I wanted to do
17:10:24Big_MacI assume it would just be from BUTTON_MENU to BUTTON_SELECT
17:10:47amiconnlinuxstb_: What's the deal with the new bootloaders? The new usb detection?
17:11:30Big_Macalright well I already have a c compiler on here, so should I try vmware again, or is there something else someone reccomends for retireving the source and patching
17:12:16*amiconn thinks the archos recorder ata speed is pretty impressive given the cpu clock
17:12:25amiconn <= new wiki page
17:12:28jhMikeSBig_Mac: vmware works nicely for me
17:12:28XavierGrBig_Mac: doesn't vmware works for you?
17:12:48 Join Lear [0] (i=chatzill@rockbox/developer/lear)
17:13:10Big_MacI tried it before and could never get it working
17:13:20Big_Macit wouldn't make zip
17:13:22XavierGrwhat was your main issue?
17:13:36XavierGrwhich version was it?
17:14:07Big_Macand I was trying to learn how to patch and it is very scary when you start get all of these "hash fails" and other errors that want me to locate files
17:14:36XavierGrwell if the patch is outdated then you have to go through it (or wrongly applied)
17:15:07amiconnjhMikeS: What are you doing wrt lcd driver?
17:15:14Big_MacI would go with august to september version
17:15:27Big_Macdon't remember the version number
17:15:46jhMikeSamiconn: nothing atm, was just telling Big_Mac about the first rockbox thing I got into
17:15:46XavierGryou can always try the new version of vmware image, although I think it is a little messed up (not rockbox related though)
17:15:49LloreanXavierGr: Messed up?
17:15:52amiconnAh ok
17:15:58XavierGrapt-get doesn't work I think
17:16:15LloreanXavierGr: I used it to install the package needed for manual building now.
17:16:25jhMikeSI couldn't be doing x5 anything since "they" still have it
17:16:42 Quit Mouser_X (Read error: 110 (Connection timed out))
17:16:43XavierGrLlorean: the latex packages were installed on the previous image too iirc
17:16:51XavierGrbut anyway I might be wrong
17:16:51LloreanXavierGr: For UTF-8
17:16:57Big_Mac which should I take for the tutorial, the how to compile in windows or the other one
17:16:57 Nick mattzz__ is now known as mattzz (
17:16:57Soapthe secret, Big_Mac, is small steps.
17:16:59XavierGryes this was out back then
17:17:05Big_Macor a different one at that
17:17:21mattzzamiconn: should I run test_disk on the sansa?
17:17:35LloreanXavierGr: I apt-get installed the UCS package for latex.
17:17:57LloreanXavierGr: So, it at least works for me. :)
17:17:58SoapDownload VMWare, since it is a complete and working enviroment, do some stock builds, get comfortable with svn and building, THEN start applying simple patches, then worry about modifying the source code yourself. IMHO.
17:18:00linuxstb_amiconn: Mainly the new textual error messages. I haven't added the USB detection.
17:18:17Big_MacOk soap
17:18:35Big_Macand where is the image I should dl?
17:18:36amiconnlinuxstb_: Hmm, does the ipod bootloader even try to detect usb?
17:18:36XavierGrLlorean: then I am misinformed
17:18:53*amiconn thinks it doesn't
17:19:03LloreanXavierGr: Maybe the other person had broken it. People seem adept at breaking things. :)
17:19:13SoapBig_Mac: wiki −−> docs index −−> vmware
17:19:25Lloreanamiconn: From a user's perspective, I've never seen any sign of it doing things different if booted with USB connected
17:19:37XavierGrLlorean: I will have to try myself then, though the old image still works fine for me
17:19:58amiconnIt _might_ be useful to add usb detection to the ipod bootloaders
17:20:25amiconnIt would avoid loading rockbox first when powering up with usb connected
17:20:36Big_MacBTW doesn't win rar handle extraction of .7z?
17:20:38 Join ender [0] (n=ender@
17:20:55XavierGryou have to download 7zip
17:21:18amiconnBut if we do that, it should be able to distinguish between usb and charger (i.e. use the new usb code not yet in cvs)
17:21:23jhMikeSpreglow: where's the patch? aren't you burning up by now?
17:21:55preglowjhMikeS: was exceedingly hard making a decent fire today, making a patch now
17:21:58Genre9mp3Llorean: I had some problems with the apt-get install on Debian-3 image... I was getting some dependencies errors
17:22:00Learbig_mac: I'm pretty sure it does...
17:22:06Big_MacXavierGr: It works on here??
17:22:21Genre9mp3Llorean: Maybe the image is updated again?
17:22:25 Quit mattzz_ (Read error: 110 (Connection timed out))
17:22:37LloreanGenre9mp3: No, I'm using the -3 image.
17:22:55 Quit Bjoern-Erik (Connection timed out)
17:23:13Genre9mp3hmm... weird.. .I couldn't install almost anything... the first thing I tried to apt-get was gedit
17:23:14jhMikeSI'd better get a source tree clean first
17:23:27LloreanGenre9mp3: Well, the only thing I installed was the UCS thing.
17:23:46Genre9mp3after many tries I quited and switched back to Debian-2 image
17:24:08 Join mm2002 [0] (
17:24:09Genre9mp3though this costs me the fact that I have to use svn 1.1
17:24:14preglowjhMikeS: found a bug, i'll let you know when things are up to snuff
17:24:14 Nick mm2002 is now known as mm2003 (
17:24:15Genre9mp3and make 3.8
17:24:53mm2003how i use the album art patch??
17:25:13 Quit ender` (Connection timed out)
17:25:13Genre9mp3mm2003: you have to know how to patch the source and then compile it
17:25:31mm2003so how i patch the source?
17:26:20Genre9mp3WorkingWithPatches and SimpleGuideToCompiling at the wiki
17:27:04LloreanGenre9mp3: gedit won't install though.
17:27:32Genre9mp3Llorean: you get dependencies errors as well?
17:27:33 Join michaelconner [0] (i=4671586f@gateway/web/cgi-irc/
17:27:35LloreanGenre9mp3: Yup
17:27:45michaelconnerhi all
17:28:06LloreanGenre9mp3: But is it really a problem with the image itself, or just the fact that it's using the 'unstable' source for the packages?
17:28:07 Join desowin [0] (n=desowin@unaffiliated/desowin)
17:28:17michaelconneram trying to put original iriver firmware back on h120 (had flashed rockbox to ROM)... instructions say "Convert an original firmware provided by iRiver from ihp_120.hex to ihp_120.bin"
17:28:31michaelconnerhow does one do that?
17:28:51Genre9mp3Llorean: To tell you the truth I'm not sure, but the dependencies errors themselves seem pretty messed up to me
17:29:06n1smichaelconner: with the scramble tool found in the tools dir
17:29:20Lloreanmichaelconner: You didn't make a backup copy as the instructions told you to?
17:29:45LloreanGenre9mp3: All it says to me is a list of things that it depends on but "it is not going to be installed"
17:29:58michaelconnerllorean: apparently not
17:30:16Genre9mp3Llorean: yes, that shouldn't be the case I think
17:30:26michaelconneri do have an iriver firmware on my computer's HD, though
17:30:57Lloreanmichaelconner: For future reference, it's generally a good idea to follow _all_ of the instructions. But yes, as n1s mentioned, I believe you use the scramble tool.
17:31:13amiconnVery convenient - on mini2g GPIO_C bit 1 is firewire detection, and GPIO_C bit 2 is firewire power :)
17:31:27michaelconnern1s: thanks
17:31:38michaelconnerllorean: yeah, I really *should* know better
17:31:57LloreanIt's especially bad to ignore instructions when doing something that can easily brick your player
17:34:30 Quit perplexity ("* follow the yellow skin toad *")
17:35:17 Quit linuxstb_ (Read error: 110 (Connection timed out))
17:35:28 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
17:37:07amiconnGenre9mp3, Llorean: The problem with the current debian vmware images is that they are based on debian stable. It would be nice to build new images based on testing (or unstable)
17:37:34amiconnDebian stable is simply too outdated to be useful in our case
17:39:33linuxstb_amiconn: You're right, there's no USB detection at all in the ipod bootloaders at the moment.
17:40:17 Quit BiptoN (Remote closed the connection)
17:40:32*amiconn would like to know the correct gpio pins for firewire detection on other ipods
17:40:58linuxstb_It could be interesting to see what happens by just branching back to 0x0 in the Rockbox bootloader (to transfer control back to the flash bootloader and hence usb disk mode), rather than a real reset.
17:41:30linuxstb_I just realised today that my ipod dock has a firewire socket on the back... I need to see if I've got a cable though.
17:41:58 Quit ender (Read error: 104 (Connection reset by peer))
17:42:25*linuxstb_ locates a firewire cable
17:42:58preglowjhMikeS: i have tested briefly on targets myself, but do mind your ears
17:43:01amiconnlinuxstb_: I like the "dynamic" disk mode more than the one that can be entered manually
17:43:02w1ll14mlinuxstb: i have a firewire cable for ipods i use to charge my ipod video
17:43:17amiconn...because it automatically reboots after disconnect
17:43:21jhMikeSpreglow: Anything in particular you want me to do?
17:43:58Big_Macany ipod users here who can reccommend some patches they fin useful outside of dan_a's KOCOP if their are any
17:44:38w1ll14mbig_mac: usb_detection from linuxstb
17:45:06w1ll14mfor using usb for charging or data
17:45:06preglowjhMikeS: just regular usage, listening to differences, tweaking, etc
17:45:17preglowjhMikeS: there really should be no discernable difference, though
17:45:19amiconnlinuxstb_: says that GPIO_C bit 2 is firewire power on the G4 as well, so I guess that bit 1 (marked unknown) will also be firewire detect
17:45:46jhMikeSpreglow: you moved the fracmul stuff for general use? :)
17:45:56preglowjhMikeS: well, i figured why not
17:46:04preglowjhMikeS: probably should use a better place than dsp.h, though
17:46:13Big_Macwhat is linux's real name
17:46:16w1ll14mamiconn: if there's a plugin or debug availible i can test, i have a firewire charger and firewire connection to pc
17:46:16preglowjhMikeS: it was duplicated it dsp.c and eq.c anyway
17:46:23linuxstb_amiconn: I get lots of GPIO activity with a firewire connection between my 5g and a PC. B goes from 0xf9 to 0xf8, E goes from 0x21 to 0x20 and L goes from 0xA8 to 0x80
17:46:23jhMikeSHow bout the fixed point library stuff?
17:46:39linuxstb_w1ll14m: Go into the main menu, then Info -> Debug -> View IO ports and see what changes there.
17:46:40jhMikeSbuilding now...
17:46:50w1ll14mok just a minute
17:46:56amiconnlinuxstb_: Port B is charger status according to ipl
17:47:02amiconnPort B bit 0 even
17:47:25amiconnNow which of the other 3 is fw power? hmm...
17:47:56 Quit michaelconner ("CGI:IRC (EOF)")
17:48:00linuxstb_At least with USB, Port B bit 0 doesn't indicate charging on my Color though. Let me try firewire.
17:48:09w1ll14mlinuxstb: on GPIO_E changes from 21 to 20 when firewire charger is connected
17:48:17jhMikeSpreglow: i get a "fsqrt defined but not used".
17:48:38jhMikeSIgnore that?
17:48:43preglowyeah, for now
17:48:51preglowi removed use of it through being a bit more clever
17:49:01linuxstb_w1ll14m: Do B and L change?
17:49:17amiconnlinuxstb_: Color is different...
17:49:58linuxstb_amiconn: Yes, I can't spot any GPIO that indicates charging status on my Color.
17:50:04linuxstb_(I mean with my experiments)
17:50:38amiconnI am mainly interested in GPIO C changes with firewire connection on color/4g
17:51:06dewdudewhy is it i can't get anything i need on flyspray...heen trying to find cop patch for hours
17:51:13w1ll14mlinuxstb: L changes as well, when normal 28 or A8, charger connected:20 or A0, when pc connected:00 or 80
17:51:24amiconnGPIO C bit 0 toggles as long as firewire is connected on my mini G2
17:51:31 Quit efyx ("Quitte")
17:51:34Lloreandewdude: Just search for coprocessor
17:51:40amiconnBit 1 goes low with both 4-pin and 6-pin firewire
17:51:45w1ll14mlinuxstb: B is always F1 or F9
17:51:52dewdudei tried that
17:52:02dan_adewdude: Or go straight to patch 5755
17:52:15amiconnBit 2 goes low on a 6-pin socket only, and it does so about one second before bit 1 goes low
17:52:21w1ll14mlinuxstb: so only GPIO_E and GPIO_L changes
17:52:21linuxstb_When connecting my Color to my PC firewire, I get C goes from 0x36 to 0x30, but flashes 0x32 as I'm inserting.
17:52:26dewdudei thought that was just the workframe to make cop work
17:52:42amiconnSo I am pretty sure bit 2 is firewire power, and will go low even with a fw charger (which I do not have)
17:52:46dewdudehence the summery "add beginnings...."
17:52:46dan_aIt started off as that
17:52:59rp-dan_a: did you try anything on the sansa battery?
17:53:02amiconnlinuxstb_: 6-pin firewire?
17:53:05dewdudeyeah, cuz my current rocobox cop build is...OLD
17:53:13dewdudei guess i should try to learn to build my own
17:53:18jhMikeSpreglow: seems to be working just fine. so you have optimizations to the eq code itself. need I check boost?
17:53:30Lloreandewdude: Flyspray is just patches, not builds, anyway
17:53:31dan_arp-: Anything I tried made the screen go funny
17:53:35linuxstb_amiconn: Yes.
17:53:39preglowjhMikeS: just coef calculations, so no
17:53:41dewdudei know it's just patches
17:53:52dewdudei wanna get some patches so i can patch some source and maybe build my own rockbox
17:54:02amiconnOkay, so 4G (colour and greyscale) fw detection is the same as mini G2
17:54:04linuxstb_w1ll14m: Is your ipod fully charged?
17:54:23rp-dan_a: yeah, same problem here, after the first i2c send it hangs
17:54:39Big_MacWhat file stores the image dispayed on boot up?
17:55:08rp-dan_a: and i can't find some init code in the BL, before that sequence starts, very odd
17:55:19jhMikeSpreglow: well I'm cranking the values all over the place and my ears are ok :)
17:55:28linuxstb_Big_Mac: apps/bitmaps/native/rockboxlogo*.bmp
17:55:51Big_Macthen if I change that and make zip it will be changed?
17:56:09LloreanBig_Mac: You have to do a make first. Then make zip.
17:56:44linuxstb_dan_a: My Color is still happilly playing back MP3s without freezing...
17:57:19w1ll14mlinuxstb: yes my ipod is charged fully
17:57:46amiconnlinuxstb_: It looks like fw power counts as "external power" on the G5. Check GPIO L bit 3 on
17:57:48preglowjhMikeS: at 24 db gain for lowshelf, i think i can pick up some wraparound noise, but that's not something i introduced now, for sure
17:58:00preglowjhMikeS: goodie, thanks. this was x5?
17:58:07linuxstb_w1ll14m: Bit 0 of GPIO_B indicates charging, it changes from 0xB9 to 0xB8 when it's charging. You could try again after using up some of your battery.
17:58:17jhMikeSno, H120...x5's not back yet
17:58:22preglowcoldfire anyway
17:58:32w1ll14mlinuxstb, ok i will be testing now
17:58:59jhMikeSall I get is ditortion with low shelf cranked
18:00:36Soapeven with a matching dB precut?
18:00:44jhMikeSCan't hear any wraparound noise
18:00:58jhMikeSSoap: Was testing it deliberatly to saturation
18:01:41 Quit pseudoXh4 (Remote closed the connection)
18:02:00lexlinuxstb_: hey... does the loader 2 enable the cop support in rockbox?
18:02:09lexor is there a new loader what does that
18:02:21w1ll14mlex: only rockbox patched bootloaders enable COP
18:02:44lexdoes it speed rockbox a lot
18:03:25w1ll14mlinuxstb: GPIO_B is constantly changing from F1 to F9 and from F9 to F1 doesn't change when connecting charger
18:03:57w1ll14mlinuxstb: only GPIO_E and GPIO_L seems to change here
18:04:26 Join fasmaie [0] (
18:04:29w1ll14mlex: at this time not very much audio thread is pretty heavy which consumes most power of ipods when using DSP
18:04:43w1ll14mof devices*
18:04:48dan_alex: The latest (SVN) Loader2 works with the COP. Also the latest (SVN) Rockbox bootloader works
18:04:53 Join Thundercloud_ [0] (n=thunderc@
18:05:08w1ll14mdan_a: loader2 supports now too ? that's interesting...
18:05:33lexdan_a: is there a built version of the svn loader somewhere
18:05:44lexhmm anyway
18:05:46dan_alex: I don't know
18:05:58lexdo you know what causes that i can't play music on rockbox? :p
18:06:01lexand i'm using 5.5G
18:06:07lexit loads but it doesn't play the music
18:06:10Lloreanw1ll14m: You don't need any patched bootloaders, as long as you're using the newest code.
18:06:21Lloreanw1ll14m: The Rockbox one doesn't require patching either.
18:06:37lexso my old ipl loader 2 works with it? :o
18:06:48dan_alex: That sounds like you're using a bootloader that doesn't work
18:06:56w1ll14mLlorean: then i missed some information i think....
18:06:58lexyeah, it's pretty old...
18:07:11lexabout the first one what worked on 5.5G
18:08:10linuxstb_lex: Easiest way is to try it and see if it works...
18:08:23Lloreanw1ll14m: The old patch required a patched bootloader, but it hasn't required one for the new version for a while. I've told people this in many places.
18:08:28linuxstb_But it sounds like you need to upgrade to a newer loader2 or Rockbox bootloader.
18:08:52lexlinuxstb_: but where can i get a built new loader 2
18:08:59w1ll14mLlorean: ok, haven't seen it yet, i'm noet very active at forums
18:09:00lexlinuxstb_: could you build it for me please? :)
18:09:30lexbecause i'm on windows and i don't like to mess with cygwins and so on
18:09:34Lloreanw1ll14m: It also says that in the comments on the patch entry as well.
18:09:43Lloreanlex: This is #Rockbox, Loader2 isn't our software.
18:09:48linuxstb_lex: You should ask in IPL-land...
18:10:14lexis rockbox's loader still working the way the first ipodloader worked
18:10:57linuxstb_It doesn't stop to ask you questions, if that's what you want. But you now start the Apple firmware by turning the hold switch on.
18:11:20lexi mean a menu would be cool
18:11:39linuxstb_Not everyone agrees with that.
18:11:43w1ll14mlex: hehe i don't like a menu :) jsut roxkbox is fine for me ;)
18:11:44 Join Criamos [0] (
18:12:03lexand apple os launching with hold off, for me
18:12:28 Join ender` [0] (n=ender@
18:13:01SoapI don't believe you'd need any coding experience to make a custom bootloader which does that.
18:13:15w1ll14msoap: lol you are definitly true
18:13:31SoapBut expecting the _Rockbox_ bootloader to boot stock firmware by default....just seems silly.
18:13:32 Join decatense [0] (
18:13:54*w1ll14m hates apple's stock firmware
18:14:33SoapRockbox isn't targeted at people who love the stock firmware and occasionally want to play...
18:14:44decatensehi can someone help me?
18:15:10decatensei want to log plays for but the option isn't in my general settings>playback menu
18:16:28Lloreandecatense: Are you sure you're using a recent enough build?
18:16:43decatenseyes, version 2.5 has it right?
18:16:47LloreanNo, it doesn't.
18:16:55LloreanVersion 2.5 is over a year old.
18:17:02decatenseoh? i read somewhere it did. oh wow
18:17:06decatenseso what version do i need?
18:17:23LloreanJust the current build.
18:17:47decatenseso is 2.5 just the last stable release?
18:18:16Soapif you define "stable" the same way Debian does...
18:18:24linuxstb_amiconn: Adding this code - - into main() in bootloader/ipod.c will cause your ipod to enter disk mode directly from the Rockbox bootloader, but at least on my 5g, I get a nasty sounding click from the hard disk. (my test was just the code to enter disk mode, without any detection).
18:18:27Llorean2.5 is the last version-numbered release.
18:18:42decatenseah i see
18:19:58LloreanThere shouldn't be any documentation that suggests is part of 2.5. Do you remember where you read it?
18:20:26amiconnlinuxstb_: The nasty click also happens when entering disk mode from within rockbox and the disk is still running
18:20:38amiconn(i.e. when powering up by connecting usb)
18:20:51decatenseit was a forum somewhere
18:20:57decatensenot the actual site
18:21:01linuxstb_Llorean: Well, the "manual" link on the website doesn't state that it's not for 2.5.
18:21:05 Quit Thundercloud (Read error: 110 (Connection timed out))
18:21:16decatenseyeah it doesn't
18:21:24Lloreanlinuxstb_: Those manuals do have a "Draft" watermark on every page though.
18:21:39LloreanWell, the PDF ones did, at least.
18:21:53linuxstb_Yes, but everyone still told to RTFM...
18:22:17LloreanIt also says "Daily Built Manuals", though I guess that doesn't tell people that the manuals belong to the daily builds.
18:22:38 Quit Kittt0s (Read error: 110 (Connection timed out))
18:22:58decatenseokay so, is it worth it to get every daily build or..?
18:23:24linuxstb_decatense: It is if you want to use the new features, like the Last.FM logging...
18:23:38linuxstb_Sorry, not _every_ daily build.
18:23:44Lloreandecatense: You should at least update as new features become available.
18:24:08linuxstb_amiconn: I guess we'll need to shut the disk down cleanly, which will delay the reboot?
18:24:12decatensewell that's sort of what i mean, obviously the feature is there and it isn't documented on the site when it became part of it so how am i supposed to know?
18:24:48amiconnlinuxstb_: Hmm, in fact this doesn't happen anymore, but it used to
18:25:03mm2003i try to open the cygwin and it doesn't open what's the problem?
18:25:14Lloreandecatense: It's in the changelog, you can see in the changelog the date it became part of Rockbox.
18:25:32Lloreandecatense: The changelog contains every change since 2.5, I believe.
18:25:38w1ll14mamiconn, linuxstb: it sounds a little scary when the disk clicks SOD :)
18:25:41linuxstb_amiconn: Yes, there's now a call to ata_sleepnow() followed by a sleep(HZ*2) before Rockbox reboots to disk mode.
18:26:37w1ll14msorry, wrong screen***
18:27:45 Part decatense
18:29:55 Quit CriamosAndy (Connection timed out)
18:30:58 Join bluebrother [0] (i=gPNe61Cw@rockbox/staff/bluebrother)
18:31:06mm2003guys i need help the cygwin doesn't open he said bash is not recognized
18:31:15mm2003what i need to de?
18:32:11 Join barrywardell [0] (
18:32:37w1ll14mmm2003: i think you are missing bash .....
18:33:03w1ll14mand seems to me that cygwin is calling for bash
18:33:14mm2003so what i have to de?
18:33:43*amiconn found asuspicious thing...
18:34:12w1ll14mmm2003: well i've never used cygwin because i have a full featured linux system that is capable of doing anything that i would need cygwin for.
18:34:24Lloreanmm2003: It sounds like it hasn't been installed properly
18:34:25w1ll14mmaybe someone else know it, or google :)
18:35:34*Soap likes the idea of using a known-good-image build enviroment to start - makes troubleshooting easier.
18:35:39 Join directhex [0] (
18:35:51 Quit Big_Mac (Read error: 110 (Connection timed out))
18:36:46*w1ll14m thinks soap is right, maybe we should have a fresh and new vmware image or zo
18:37:08bluebrothermm2003, run the cygwin setup and tell it to install bash
18:37:19bluebrotherbut why not using the vmware image? Setup is way easier
18:37:22Soapthe current one should be fresh-enough for all but manual building, eh?
18:37:34bluebrotherit should.
18:37:49w1ll14mSoap: i think so, never used the vmware image
18:37:58bluebrotherand manually installing unicode support for latex shouldn't be a problem.
18:38:04Lloreanbluebrother: It isn't.
18:38:26bluebrotherit isn't? Ah, the image is stable and unicode packages are only in unstable?
18:38:31Lloreanbluebrother: I can't remember what the package was named now, but apparently it's one of the few packages the current vmware image will install without a hassle.
18:38:35Soap(though I don't suspect manual-building is at the top of many new-to-compiling user's lists) ;)
18:38:41Lloreanbluebrother: No, I meant it isn't a problem.
18:38:58bluebrotherah. Got it too late ;-)
18:41:15 Join webguest19 [0] (i=485c0855@gateway/web/cgi-irc/
18:41:18webguest19Copy OF.bin to the SYSTEM folder on your Sansa ?
18:41:43webguest19where do u get that from ?
18:42:54Lloreanwebguest19: It's what you rename your original firmware file to.
18:43:11rp-after you have decrypted and stripped the header
18:45:18jhMikeSpreglow: it appears the flac codec has been giving me flac by not being truthful about the data size. A wave file from the same source gives no clicking up or down if I use the formula from a couple days ago for up and the original for down.
18:45:28webguest19how to decryted
18:45:36preglowjhMikeS: well, how nice
18:45:51Lloreanwebguest19: Look, simply put, Rockbox on Sansa isn't ready for normal users yet.
18:45:51rp-webguest19: read
18:45:51preglowargh, another bug in the eq code
18:46:04 Quit desowin ("use linux")
18:46:04jhMikeSyeah... :P
18:46:11rp-and listen to what Llorean said :)
18:46:37preglowwhy oh why must emac be so stupid
18:46:50preglowsome kind of accumulator shifting instructions would be lovely
18:47:31preglowa full 64 bit accumulator wouldn't be too ba either
18:47:38 Join Arathis [0] (
18:47:43 Quit Thundercloud_ (Read error: 104 (Connection reset by peer))
18:47:50jhMikeSI know, no other arithmetic on accumulators. Was ticking me off but I found wacky ways around it.
18:47:59 Join Thundercloud [0] (n=thunderc@
18:48:59preglowall i need to is bloody shift the accumulator some bits up before retrieving it, but no, that needs to be done with line after line of code
18:49:12 Join desowin [0] (
18:49:52linuxstb_amiconn: Do you know how easy it would be to make Rockbox do a clean shutdown before rebooting into disk mode?
18:50:05preglowno, wait... i think i'm having a brain moment here
18:50:25 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
18:51:20 Quit Thundercloud (Read error: 104 (Connection reset by peer))
18:51:45 Join Mouser_X [0] (
18:51:45 Join Thundercloud [0] (n=thunderc@
18:52:26Learpreglow: like shift done via mul/div?
18:53:03 Quit webguest19 ("CGI:IRC (EOF)")
18:53:06amiconnlinuxstb_: afaik it does now, that's something added by Slasheri
18:53:08jhMikeSpreglow: What's in the brain moment? I had to trick it to always round towards zero.
18:53:55preglowjhMikeS: the brain moment is just me being an idiot when i should be reading how the accext registers are arranged
18:54:21jhMikeSor misarranged ;)
18:54:23linuxstb_amiconn: It does? I don't see a "shutting down" splash, and it seems to be too fast to save the settings. Let me test....
18:54:45preglowjhMikeS: isn't all you need to do to make it do that turn of the rounding flag in macsr?
18:55:56linuxstb_Hmm. If I change a setting, and then insert USB, it freezes on the USB screen with the disk activity icon on, but no disk activity....
18:57:04linuxstb_Unattaching makes Rockbox leave the USB screen, and then a reattach works. Rebooting into Rockbox shows my setting change was saved.
18:58:18jhMikeSpreglow: then it rounds towards -inf
18:58:23jhMikeSjust like >> 31
18:58:40linuxstb_Trying the same thing again worked fine (no freeze, setting change saved).
18:59:01mattzzwhat does a blinking red led indicate on an AJR (playing music, running a plugin)
18:59:57preglowjhMikeS: ehm, that's truncation, which is the same as round towards zero
19:00:31preglowfor rounding towards -inf, it would have to do a floor()
19:00:33 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
19:02:46jhMikeSno...shift rights are toward -inf. Whats -1 >> 1? It's still -1.
19:04:23 Quit Mouser_X (Nick collision from services.)
19:04:32 Join Mouser_X [0] (
19:04:33 Join Thundercloud_ [0] (n=thunderc@
19:04:35 Quit mm2003 ()
19:06:01mattzzcan somebody help me with the blinking red led on my ajr?
19:06:34LearIt's not disc access?
19:06:56mattzzsure, and when it starts to blink?
19:07:04***Saving seen data "./dancer.seen"
19:08:53 Join DarkCow [0] (
19:09:20 Nick mattzz is now known as mattzz_away (
19:09:28DarkCowhi, crwl, you don't know me, but did you ever figure out how to get xchat's ctrl-w to stop being so crap?
19:09:43LloreanDarkCow: This is #Rockbox.
19:10:12jhMikeSpreglow: it was important to round towards zero in the blit synth so all the quantization error when the angles were small would be above the ideal curve and could be easily clipped to lim
19:10:23DarkCowyeah, but he isn't in any other channels, and PMing is rude
19:10:46LloreanDarkCow: Coming into support and development channel and bringing up something entirely off topic isn't?
19:11:01DarkCows'pose :-/
19:12:12 Part mattzz_away ("Leaving")
19:14:08linuxstb_dan_a: kernel_on_cop_8 is still running fine on my Color after more than 2 hours... (with the codec thread moved back to the main CPU).
19:14:54linuxstb_... And I've allegedly got 68% CPU left.
19:15:29Lloreanlinuxstb_: Did your 4G freeze fairly consistently in the past?
19:15:47 Join Wiwie [0] (
19:15:51amiconnBlarg, gcc :(
19:16:06amiconnWhy does gcc optimise inconsistently?
19:16:22directhexyou can get an academic license for ICC if you prefer
19:16:24linuxstb_Llorean: It has been recently.
19:16:44amiconndirecthex: I doubt that icc does support the sh architecture...
19:16:57directhexamiconn, pfft, minor detail
19:16:59Lloreanlinuxstb_: So it seems to possibly be fixed, then?
19:17:55linuxstb_It could just be luck, the stability of PP5020 builds always seems to be random. But it seems good so far.
19:19:09directhexis accurate? i thought the karma used some custom jobby that's now used by sigmatel & deployed in systems like the trekstor vibez
19:19:12jhMikeSpreglow: The wav and flac codecs both put out > 16 bits/sample to the dsp right?
19:19:44linuxstb_directhex: Yes, the Karma had a PP chip.
19:20:07amiconnHmm. struct user_settings could be reordered to minimise padding
19:21:51amiconnNot sure whether it's worth it; it would collide with feature conditionals
19:25:16 Quit DarkCow (""Input/output error" - the only way to quit IRC in style")
19:25:17jhMikeSuser_settings could lose some padding by #ifdef-ing the unused stuff on a target.
19:25:35preglowjhMikeS: well, i think so
19:25:40preglowjhMikeS: most of them output 32 bits now
19:25:50 Quit Thundercloud_ (Read error: 104 (Connection reset by peer))
19:26:04 Join TrueJournals [0] (
19:26:19 Join Thundercloud_ [0] (n=thunderc@
19:26:40jhMikeSpreglow: ok, good. then the flac codec is lying and the wav not
19:27:24jhMikeSnot that that's really good :P
19:28:07 Quit Thundercloud (Read error: 110 (Connection timed out))
19:29:39 Quit mathgl (Remote closed the connection)
19:32:30preglownow, if something as simple as the wav codec can't keep track of the number of samples it does, then hooray for that
19:33:13 Join Genre9mp3 [0] (
19:36:46 Part TrueJournals
19:37:14preglowso it better be flac, yes
19:37:22preglowi want the ultimate instruction set, please
19:37:34jhMikeSyippie! guess I can simplify the heck out of stuff
19:38:04preglowyippie indeed!
19:38:23jhMikeSyeah, they couldn't even have byteswap and bitswap on our cpu's.
19:40:13jhMikeSanother thing is the conspicuous lack of a single cycle integer abs instruction on stuff
19:40:44preglowguess they felt moveq and sub did the trick
19:40:46preglowthey were wrong!
19:40:48 Part Llorean
19:41:05preglowhell no, you'll need to branch too
19:41:08preglowthey're on crack
19:43:11preglowemac is nice in its own way, though
19:43:23preglowfaster than anything else we can do on other platforms
19:44:35jhMikeSyes. I would have liked a double register mult/div like intel cpus. makes a muldiv really simple. and they were so on crack they didn't have rems/u return both the quotient and remainder so you have to divide twice. :P
19:44:46preglowheh, yeah
19:45:07w1ll14mpreglow: do you have some information about the EQ optimizations ?
19:45:23 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
19:46:05jhMikeSw1ll14m: I'm not sure. He might know something. ;)
19:46:30pregloww1ll14m: not much, it's really not something to care too deeply about, people probably won't know much of a difference
19:46:57w1ll14mjhMikeS: lol
19:47:07preglowit certainly won't give you any extra battery, i've already made the filtering itself as fast as i can
19:47:10w1ll14mpreglow: ok, but then it's an optimization :)
19:47:53w1ll14mpreglow: we should be proud of it when it works :)
19:48:19jhMikeSdoes it save any bytes? that's always a good thing.
19:48:20 Join mattzz [0] (
19:48:38w1ll14mjhMikeS: i definitly agree :)
19:49:09preglowjhMikeS: saves plenty of bytes
19:49:26jhMikeSpreglow: Is the fsqrt a general purpose one? Should it be in fixedpoint.c?
19:49:39preglowjhMikeS: sure, should work for anything
19:49:46preglowit's dog slow, but nice enough
19:49:59preglowjhMikeS: yeah, i'll probably move it there just you be able to stow it away some place
19:50:17 Join Bjoern-Erik [0] (
19:50:25jhMikeSgood idea, don't expunge it if it possibly useful
19:50:48preglowtoo bad it's a plugin lib, though
19:51:02preglowwould be nice to be able to make a lib which both core and plugins could use
19:51:14preglowbut i'm not skilled enough with the build system to do that
19:51:20 Join Genre9mp3 [0] (
19:51:55 Join mathgl [0] (
19:53:04 Quit directhex (Remote closed the connection)
19:53:29jhMikeSMaybe I should add my emac-ed fsin function to the lib if you need super fast trig.
19:53:41preglowjhMikeS: sure would be cool to see it
19:53:47preglowjhMikeS: how accurate is it?
19:54:21 Quit Mouser_X (Read error: 110 (Connection timed out))
19:54:38jhMikeSlargest harmonic is about -97db
19:55:03preglowdoesn't say too much about its applicability for the eq, though
19:55:15jhMikeSgood enough for a blit synth to have a noise floor below -120 db
19:55:16preglowthe old sin i had had first harmonic at about -80 db and was completely useless for the eq
19:55:26preglowfor blit it's probably nice, yes
19:55:31preglowbut i'd still love to see it
19:55:36preglowgot it hanging around on the web somewhere?
19:55:56jhMikeSno, I can pastebin
19:56:02 Join directhex [0] (
19:56:17jhMikeSit is based on the one from your signalgen
19:57:12amiconnjhMikeS: rems / remu *do* return both quotient and remainder
19:58:13jhMikeSamiconn: Where? I tried it any it didn't seem to be doing it.
19:58:46amiconnrems / remu are in fact just divs / divu where the remainder goes into an extra register instead of being discarded
19:59:46preglowjhMikeS: ah, yes, that was the one i first used for the eq
19:59:59preglowa nice oscillator sine, but not so nice for coef calc, i discovered
20:00:03amiconnThe quotient should be in Dx
20:00:06 Join TrueJournals [0] (
20:00:35jhMikeSpreglow: well I used double the entries and more bits which made all the difference
20:01:12*directhex wonders whether the doom plugin could be rotated for screens that are taller than they are wide
20:01:58n1sisn't it rotated on the GB?
20:01:58jhMikeSamiconn: seemed to not have been changing Dx and it says if you want the quotient call divs/u. Maybe I did something goofy. :\
20:06:02jhMikeSn1s: ??
20:06:52n1sjhMikeS: ah, yes the original doom for GB patch had rotation but it seems kkurbjun didn't use that patch, so no rotation
20:08:51 Join Thundercloud__ [0] (n=thunderc@
20:09:18preglowargh, passing 8 to FRACMUL_SHL generates an instruction which isn't valid but is correctly discarded, but with a warning :/
20:09:43jhMikeSn1s: mean screen rotation? The gigabeat's a portraint display?
20:10:14n1syes it is, 240x320
20:10:33n1sbut it does have rotation in mpegplayer
20:11:22 Join Rondom [0] (
20:12:13jhMikeSand it looks good viewing it 90deg? The x5 lcd looks bad if you're not looking at it just the right way.
20:12:37markunDoes anyone know how to use /ignore to ignore join and quit messages of someone?
20:13:27markunn1s: doom rotates fine here..
20:13:37 Quit _Veseliq_ (Remote closed the connection)
20:14:34n1smarkun: oh, sorry for the confusion, it wasn't mentioned in the commit message or in the patch that kkurbjun posted...
20:14:34markundirecthex: doesn't it give the option to rotate on your player?
20:14:55markunn1s: it doesn't rotate by default but it's optional
20:15:04markunI think it shouldn't be optional though..
20:15:37markunmapping a 320x200 game to 240x320 is a bit weird
20:17:21directhexso there is. clever!
20:19:03markundirecthex: but remap your keys :)
20:19:15directhexyeah, so i see :x
20:22:32preglowanything nice i need to know if i am to include preprocessor code in a macro? :>
20:23:19jhMikeSpreglow: like #define x() #define something_else ??
20:24:20preglowjhMikeS: stuff like that, yes, basically i'm putting an #if in a #define macro lol(x, y)
20:24:28jhMikeScan you even do that? =|
20:24:32preglowi don't know :>
20:24:50preglowi'm trying to get rid of the offending line of asm if one of the const parameters to the macro is 0
20:24:58jhMikeSWhen I have those questions I just try it out
20:25:00preglowbut i'm not a wiz at the preprocessor
20:25:08preglowi have, but it doesn't work
20:25:08jhMikeSWhy can't those be inline functions?
20:25:36preglowthey can, i guess, but it wouldn't rid me of my problem
20:25:42preglowand the code would be slower
20:25:56preglowhmm, or possibly not, but anyway, my problem would be the same
20:25:59jhMikeSI haven't noticed a compromise with small inlines
20:26:05 Join rUiSu [0] (
20:26:13jhMikeSyou can ifdef inside it then :)
20:27:13 Quit rUiSu (Client Quit)
20:27:14jhMikeSI have seen a compiler that will allow that sort of thing but can't recall which.
20:27:36preglowvomits then too
20:27:49preglowi really have no idea if this can even work
20:28:36preglow'#' is not followed by a macro parameter
20:28:42 Quit mattzz ("Chatzilla 0.9.77 [Firefox]")
20:28:46preglowi get that whenever i try to insert preprocessor code in a macro
20:28:48amiconnjhMikeS: Hmm, I just checked and the programming manual says you need divs / divu to get the quotient
20:28:53preglowso it looks it tries to use the concat operator
20:28:57preglowor stringify, i mean
20:29:50amiconnBut I've seen 3-register forms of divs / divu in disassemblies of rockbox for our cf targets...
20:29:57preglowamiconn: FRACMUL_SHL(x, y, z) is a multiline asm() macro, and i need to omit one of the lines of asm based on the (compile-time constant) value of z. possible?
20:30:09 Quit _ff ("Konversation terminated!")
20:30:57 Join mattzz [0] (
20:30:57 Join Alonea [0] (
20:31:09amiconnUmm, not sure, but I guess that's not possible
20:31:10jhMikeSpreglow: I think you should google the obfuscated c code contest
20:31:20 Quit amiconn (" Like's GUI? Then try HydraIRC -> <-")
20:31:44preglowjhMikeS: possibly :/
20:31:58preglowproblem is that gas vomits on lsr.l #0, %dx
20:32:02jhMikeSThey do that there but I don't know the comiler that accepts it
20:32:04preglowand rightly so, since it's not legal
20:32:26 Quit Thundercloud_ (Connection timed out)
20:32:30preglowbut i need it
20:32:39preglowthat is, it results from a use case i have
20:32:44 Quit Paprica (Read error: 110 (Connection timed out))
20:32:48jhMikeShmmm...cant use the .word thing?
20:33:11preglowi can use loads of things, but it'll end up in the code
20:33:22preglowi'd rather the code just didn't go in, since it's a nop
20:33:23idnarpreglow: maybe just have two functions?
20:33:52preglowidnar: i don't think macros bend that way, it would need to be called the same thing
20:33:55jhMikeSI know, use a stringizer for the parameter?
20:34:14idnarpreglow: what I mean is, FRACMUL_SHL(x, y, z) and FRACMUL_SHL_ZERO(x, y) or something
20:34:30preglowidnar: it would have to be FRACMUL_SHL_EIGHT in that case
20:34:44preglowthe parameter value is actually 8, but 8 - 8 is 0
20:34:48idnarah ok
20:34:59preglowjhMikeS: hmmm
20:35:08idnarok, I guess that's not so great then
20:35:41preglowi guess i could just try to use a smaller shift
20:37:00preglowor, *gasp*, not use FRACMUL_SHL
20:37:05preglowi was just hoping to learn something here :)
20:37:34hcsdan_a: I'm going to be testing out kernel_on_cop_8 on my ipod color today
20:38:08w1ll14mkernel_on_cop_8 works as expected on 2 5g ipods 30 and 60GB
20:38:37hcsso I've heard, but he mentioned that it is untested on 4g color
20:38:52 Quit Alonea (Remote closed the connection)
20:39:35 Join Juice^ [0] (n=juice@
20:40:17*Soap is heartbroken the EQ work won't reduce boost.
20:40:37*Soap is heartwarmed that the current EQ filters are so efficient. ;)
20:41:08dan_ahcs: Thanks. linuxstb_ tested without running the codec thread on the COP and confirmed that it works OK
20:41:54hcsdan_a: so would my testing be redundant? I don't\ know which versions do exactly what
20:42:09*w1ll14m loves eq so much that he even changed the CPU frequencies instead of just the EQ frequencies
20:42:23preglowi'll just bloody well go ahead and use moveq.l, that way the 0 won't matter
20:42:56 Quit mathgl (Remote closed the connection)
20:43:09dan_ahcs: It would be redundant - although wider testing never hurts. The patch also appears to do Good Things with the frequency scaling
20:43:25preglowGood Things as in it will never crash anything again?
20:43:34jhMikeSpreglow: you're using constant shifts?
20:43:39preglowjhMikeS: yes
20:43:44hcsdan_a: ok, the freq scaling freeze issue is mostly what I'm looking for
20:44:01preglowjhMikeS: compile-time constant and everything, the code gets spat out like lsr.l #7, %d0
20:44:54dan_apreglow: I won't guarantee that, but I've not had a crash since I started using it
20:45:21preglow\o7 everythings (8 - shift)? and shift is 1 - 8?
20:45:29preglowouch, that's one broken arm
20:45:44preglowjhMikeS: i need two shifts, one z and one 8 - z
20:46:14preglowjhMikeS: i've got the z shift covered, but 8 - z sometimes becomes 0, so the code ends up as lsr.l #0, %dx
20:46:37preglowand then gas complains
20:46:54preglowit does the right thing, ignores the instruction, but i don't want a spotty build table from the warnings
20:47:25 Quit Bjoern-Erik (Connection timed out)
20:48:03hcsdan_a: it isn't booting for me (stuck at end of bootloader), should I have not rebuilt the bootloader?
20:48:33dan_aYou should not need to... how recent is your bootloader?
20:48:43 Quit spiorf (Remote closed the connection)
20:49:02barrywardelldan_a: I'm doing some testing on cop 8 with my H10 at the moment
20:49:03hcswell, I just built one now, from current svn, after applying the patch, figured it'd be best to have the same version
20:49:03jhMikeSIn macro: if (8 - z == 0) {} else { asm (""); }
20:49:26barrywardellwith frequency scaling disabled, i still don't get a main thread on the cop
20:49:43rp-dan_a: when you tried to read the battery level on the sansa, did you do it in the adc_scan() function?
20:49:47barrywardelland disabled scaling is the default in svn
20:50:01Juice^RTC support for Sansa. - whats this RTC thing? :)
20:50:12rp-Juice^: real time clock
20:50:14barrywardellJuice^: Real Time Clock
20:50:40barrywardelldan_a: testing with frequency scaling enabled now and haven't had a freeze yet in over two hours
20:50:55rp-or barrywardell, did you some tests with the battery on the sansa?
20:50:56 Quit Wiwie (Read error: 110 (Connection timed out))
20:51:16Juice^so the clock is now working for the sansa
20:51:20barrywardellrp-: I tried reading the battery level in adc_scan and just got a corrupt lcd display and rockbox locked up
20:51:26Juice^can it be adjusted within rockbox too?
20:51:30barrywardellJuice^: yes and yes
20:51:32dan_abarrywardell: The same happens on the Sansa. I've got that on my list of things to fix before committing
20:51:51dan_arp-: I did the same as barrywardell with the same results
20:52:16barrywardelldan_a: i had a version of ipod_set_cpu_frequency() which gave me a main thread on the cop with frequency scaling disabled
20:52:16preglowjhMikeS: i don't quite get how that'd work
20:52:29preglowjhMikeS: you mean like two completely different versions of the code, then?
20:52:31barrywardelldan_a: based off MrH's docs
20:52:40jhMikeSI think if the if is a constant expression, it only compiles the reachable branch
20:52:54rp-ok, i tried the same and had the same problem, now i moved the code to somewhere else and there it seems to work, no corrupt lcd
20:53:21rp-maybe the i2c isn't initiliazed before calling adc_scan?
20:53:44preglowjhMikeS: it does, that's right
20:53:56jhMikeSpreglow: could probably split the block
20:54:05dan_abarrywardell: I think I remember testing it over Christmas and calling ipod_set_cpu_frequency later worked
20:54:56preglowjhMikeS: since this is becoming wildly inefficient anyway, i'm just going with the moveq.l
20:55:00Juice^barrywardell: sweet mary ... :)
20:55:09dan_arp-: I think that other targets use i2c for the battery levels, so that doesn't sound right, but maybe it does cause problems
20:55:11barrywardellrp-: i2c_init() is called before adc_init() in init() in apps/main.c. where did you put it when you got it to work?
20:55:44preglowi really could just go back to it being the way it was before with shifting in zeroes, but for arm the high precision shifting is free, and i want all platforms to create the same coefs
20:55:49barrywardelldan_a: I wonder if calling it while the cop is still sleeping is the problem.
20:56:02rp-barrywardell: i but it in the debug_menu.c
20:57:37barrywardellJuice^: Just don't try set the date to anything before 2000. That won't work.
20:57:43barrywardellrp-: i'll try that
20:57:45Juice^When inside the setting time/date for the sansa, what button is supposed to be the "save" button?
20:57:46w1ll14mmaybe the lockup of the pp5020 are caused by a sleeping COP and a changing CPU frequency. Possible solution in this case would be wake the COP then change frequency and sleep the COP again
20:57:54Juice^barrywardell: ok ok
20:57:57dan_abarrywardell: That sounds plausible. Perhaps we should wait for the COP to wake up in the main thread
20:58:23dan_aw1ll14m: That would sort of tie in with what MrH found
20:58:30rp-barrywardell: and what i saw from the disassembled BL, the readout sequence is done 3 times, but i don't know why
20:58:46barrywardellJuice^: I think it's select. I also think the buttons for that screen should be changed
20:58:55jhMikeSpreglow: As long as the compilation is efficient. It shouldn't reload anything in the asm
20:59:27barrywardelldan_a: yes, I think we should. I'll do some testing on that
20:59:29preglowjhMikeS: correct, but i don't want to add a hack of that magnitude just to save sixteen bits of code and one cycle
20:59:52w1ll14mdan_a this just came by in my thoughts a few days ago, and i wasn't reading irc but i understood that it was about COP and lockups on PP5020. so i thought 'll just day it :)
20:59:54Juice^barrywardell: yea.. they should, kinda unlogical right now
21:00:02*jhMikeS would go right ahead if he gets what he wants from it =:)
21:00:24Juice^barrywardell: select button did not save
21:03:45preglowthis is going to end up with me just starting to write everything in asm
21:07:07***Saving seen data "./dancer.seen"
21:07:55Juice^oh my... i just clicked on "rockbox default" theme in the settings, and i cant read any of the text in the menus.. its all garbled
21:08:21Juice^never happend before
21:09:48 Quit Lear ("Chatzilla 0.9.77 [Firefox]")
21:09:56barrywardelldan_a: do I just check bit 31 of COP_STATUS to see if the COP is sleeping?
21:12:05 Quit fasmaie (Remote closed the connection)
21:12:06dan_abarrywardell: MrH checked bits 30 and 31 of COP_CTL (i.e. cop_status = COP_CTL >> 30; if cop_status == 0 then cop is sleeping)
21:12:08 Join fasmaie [0] (
21:12:18 Quit Juice^ ("brb")
21:13:15preglowrockbox.* is really starting to grow big
21:13:57 Join Juice^ [0] (n=Juice@
21:14:03n1sdynamically loaded modules!
21:14:10*n1s runs
21:15:04 Part TrueJournals
21:15:08 Join linuxstb [0] (i=5343d4aa@gateway/web/cgi-irc/
21:16:01linuxstbdan_a: Bad news, I've just returned to see my Color froze...
21:16:40barrywardelldan_a: thanks. maybe we should do it that way in crt0-pp.S then
21:17:10 Join pseudoXh4 [0] (n=pseudo@
21:17:23 Join decayedcell [0] (
21:18:38preglowalmost all the size gain on coldfire was eaten by FRACMUL_SHL, bleh
21:18:39preglowoh well
21:21:35 Join BiptoN [0] (
21:25:09 Join rotator [0] (n=e@rockbox/developer/rotator)
21:25:52markunDoes anyone here know how power efficient FPGAs are? Would they be an option for a DIY DAP project?
21:26:35 Join topbloke [0] (
21:26:55 Quit BiptoN ("Leaving")
21:28:02 Join Llorean [0] (
21:29:23 Join ff [0] (
21:29:32barrywardellJuice^: SELECT works for saving the new time for me
21:30:04mattzzbarrywardell: did you have a chance to check out the memory monitor plugin?
21:31:02Juice^barrywardell: well.. maby it just wont let me set the date back to 4-1-07 then..
21:31:13Juice^beginning with month there
21:32:18barrywardellJuice^: applying rockbox default_theme didn't garble the screen for me. I did just experience a a weird garbled screen when the backlight tried to turn off though
21:32:49barrywardellJuice^: I changed to April 1st 07 ok...kinda - it changed to April 2nd instead
21:32:49Juice^hehe.. there's lots of strange bugs :)
21:33:07barrywardellmattzz: yes, I tried it out. could be quite useful in certain situations
21:33:15Lloreanlinuxstb: Does other PP targets include the Sansa, or just the PP5002 targets, because as far as I can tell that's all your list is missing? (Regarding USB patch)
21:33:18 Join freqmod [0] (
21:33:36mattzzbarrywardell: ok, if you find anything or need anything pls let me know
21:34:01 Join Mouser_X [0] (
21:34:04barrywardellmattzz: I was going to suggest an option to select a starting address, but you already did that :)
21:34:22barrywardellJuice^: it's likely that the rtc code is a bit buggy
21:34:43Juice^barrywardell: yes indeed.. too bad im not a debug hero
21:35:34 Part decayedcell
21:36:34 Join Thundercloud_ [0] (n=thunderc@
21:39:46preglowdivull %d1,%d2,%d2
21:39:57preglowcan anyone tell me what kind of coldfire instruction that is? i've never seen it...
21:40:38preglowyet it's used in my disassembly here
21:41:01 Join ctaylorr [0] (
21:41:47Lloreanlinuxstb: The patch seems to work for the e200 as well
21:42:16Lloreanpreglow: Clearly the disassembler's making things up to see if you're paying attention.
21:43:56barrywardellJuice^: I just committed one fix for the date setting. Does it work for you at all?
21:44:24Juice^barrywardell: i cant set the current date at all no..
21:44:36preglowLlorean: it succeeded, i'm wasting my time reading pdfs now...
21:44:49 Join BiptoN [0] (
21:46:02barrywardellJuice^: weird. works for me here.
21:46:43Juice^i'll try to reload my rockbox on the sansa
21:47:01barrywardellwait till the newest build is done, then try
21:47:51Juice^i'll wait then. how long does it take prox?
21:48:07barrywardellabout 7 minutes
21:48:25Juice^ok so every 10mins
21:48:33 Join senab [0] (
21:48:56BiptoNhey barry i've been recording with a 5.5g ipod, do you know why there is noise in one channel?
21:49:08BiptoNmy 4g had alot of noise in both channels
21:49:09barrywardellJuice^: no, about 7 minutes after a svn commit
21:49:10BiptoNsame noise
21:49:21senabi was just syncing a patch that uses the old settings.c unfortunately i keep getting a warning while compiling:
21:49:37Juice^barrywardell: ok
21:49:37senabsettings_list.c:673: warning: missing initializer
21:49:37senabsettings_list.c:673: warning: (near initialization for ‘settings[160].<anonymous>’)
21:49:42senabany ideas?
21:50:03barrywardellBiptoN: I dunno. I didn't notice any noise on my H10. Maybe linuxstb would know. He got it working on the iPod
21:50:43LloreanBiptoN: Are you using the same dock adapter for both?
21:50:46 Quit pseudoXh4 (Remote closed the connection)
21:51:06senabhere's the line 673: {F_T_INT,GS(ipod_scroll_wheel_acceleration_fast),LANG_IPOD_SCROLL_WHEEL_FAST,INT(135),"ipod scroll wheel fast threshold in clicks/sec", NULL},
21:52:05 Join spiorf [0] (
21:53:45BiptoNsame adapter
21:53:59BiptoNon the 5.5 the noise is alot lower and only on one channel
21:54:33LloreanBiptoN: Since I haven't heard anyone else mention noise while recording yet, it may just be a hardware-related issue.
21:54:43Juice^barrywardell: that fix you just commited sounds like the exact thing happening on my rockbox right now. i'll update it in a few mins and report back
21:55:12 Join TrueJournals [0] (
21:55:51 Quit fasmaie (Remote closed the connection)
21:55:58BiptoNhuh, very strange
21:56:14BiptoNsame setup i used on my x5
21:56:29BiptoNi resoldered the adapter yesterday also
21:56:35barrywardellwhich channel was the noise on?
21:56:47BiptoNright side
21:56:51BiptoNlemme double check
21:57:09BiptoNboth channels on the 4g
21:57:26BiptoNright channel barry
21:57:39BiptoNit registers noise just sittin there on the recording screen
21:57:46BiptoNmaybe the hard drive?
21:58:32 Join Hans-Martin [0] (
21:59:21rp-hm strange, now i put the battery code at the end of i2c_init() and it doesn't freeze, so something bad must happen in adc_scan()
21:59:39 Nick Everybody|away is now known as Everybody (
22:00:01Hans-Martinhi folks, I've got a very specific iriver iHP hardware question: how do I disconnect the battery connector? I don't want to break it...
22:00:17 Quit Thundercloud__ (Read error: 110 (Connection timed out))
22:01:25 Quit senab ()
22:01:25 Quit topbloke ("bye")
22:01:53Hans-Martinmy primary problem is that the reset switch does not work, probably I pressed it with too much force or an unsuitable tool - I suspect that it needs a little re-soldering
22:01:57barrywardellrp-: I tried an empty adc_scan apart from the i2c code and it still didn't work
22:03:23 Join Thundercloud__ [0] (n=thunderc@
22:04:01Genre9mp3Hans-Martin: maybe this is helpful:
22:04:38 Join DataGhost [0] (
22:05:30rp-how often is adc_scan() called? or from where?
22:05:31preglowsweet lord, gcc isn't very clever is it
22:05:42jhMikeSpreglow: divull? wtf. That's an m68k instruction but not one for coldfire.
22:05:50 Join rUISu [0] (
22:05:56preglowjhMikeS: well, m68k-elf-objdump spat what you see there at me
22:06:14preglowjhMikeS: perhaps i need to be more specific as to what arch i'm using
22:06:16 Quit freqmod (Remote closed the connection)
22:06:47 Quit rUISu (Client Quit)
22:06:48 Quit Mouser_X (Nick collision from services.)
22:06:57 Join Mouser_X [0] (
22:07:00barrywardellrp-: it's called by the adc_tick function
22:07:26jhMikeSI'd say so...:) That's 68020 +
22:07:27barrywardellonce per second
22:07:51barrywardelland once in adc_init()
22:07:54rp-hm could this be a problem?
22:08:45markunHans-Martin: you can just unplug it, but it's a bit tricky because there is not a lot of space
22:09:17markununplugging was a lot easier than reinserting from what I remember from my battery upgrade
22:09:29preglowjhMikeS: now it suggests remul %d1,%d2,%d2 ...
22:09:50preglowusing -mm68k:5249 as a flag to objdump
22:10:03Hans-Martinmarkun: ok, I'll try
22:10:17markunHans-Martin: are you having difficulties with it or didn't you try?
22:10:31barrywardellBiptoN: I don't see anything jumping out at me in the recording code. And I haven't heard any other reports of your problem so it could be a hardware issue. I think Soap did some testing - maybe he can report if he has the same problems
22:10:53BiptoNdon't you have an ipod barry?
22:11:04Hans-Martinmarkun: it seemed to need quite a bit of force, and as I said I don't want to break anything else
22:11:15BiptoNjust loadin up the recording screen shows signal hittin on the channel/s
22:11:27BiptoNthat's with no adapter plugged in
22:11:27*jhMikeS was hoping preglow found secret coldfire instuction for 64 bit arithmetic
22:11:57BiptoNi can take some pics for ya, if it'd help. I have tried it on 2 4g grayscale ipods and my new 5.5g
22:11:58barrywardellBiptoN: no, only H10. linuxstb did the work on ipod recording
22:12:12BiptoNi'll have to ask him when he gets in
22:12:19BiptoNthanks barry
22:12:23barrywardellyes, probably best to do that
22:12:46*linuxstb is in
22:13:05preglowjhMikeS: so you don't know what remul is?
22:13:30BiptoNlinuxstb: do you get signal on either channel or both when you bark off the recording screen on your ipod?
22:13:51 Join Paprica-Birthday [0] (n=Paprica@
22:14:05Hans-Martinmarkun: yay it worked - indeed very little space there
22:14:32preglowjhMikeS: god, gcc is braindead. when i do *c++ = somevalue; it doesn't use the post-increment addressing mode at all, it splits it up into a couple of movs and an addq
22:14:40jhMikeSpreglow: no, can't find much on that one
22:15:16jhMikeSheh...that's bad
22:15:16BiptoNbarry i have a 4g grayscale with no HD, would it benefit you at all in any way?
22:15:34linuxstbBipton: What do you mean by bark off?
22:15:42BiptoNfire it up
22:15:46BiptoNturn it on
22:16:02barrywardellBiptoN: not right now, thanks anyway.
22:16:40linuxstbYes, I get very low-level signals on both channels, fluctuating slightly. Both on my Color and Video IIRC.
22:17:52BiptoNmy 4g's had a good little amount coming through on both channels
22:18:03BiptoNmy new 5 only gets a little on the right channel
22:18:11BiptoNenough to be annoying though
22:18:25 Quit w1ll14m ()
22:18:47BiptoNdo you think it maybe from hd noise or some other interference?
22:21:10 Join pseudoXh4 [0] (n=pseudo@
22:21:31BiptoNis there any acoustic management in the toshiba drives or does rockbox support that
22:22:47jhMikeSpreglow: I get my assembly listings from adding -save-temps to GCCOPTS in the makefile. Maybe that's stupid but at least that's the code that actually gets used.
22:23:50preglowjhMikeS: i've been using -S for that, but -save-temps is far less intrusive indeed...
22:23:54*preglow tries out
22:24:14linuxstbBipton: Sorry, I don't know.
22:24:51jhMikeSYou'll have a mess of crap in your source tree though :P
22:25:25preglowjhMikeS: well, it's easy to delete
22:25:56preglowfind|grep \.tmp\.|xargs rm should do the trick
22:26:15 Join CpuWhiz [0] (
22:26:22 Quit Thundercloud_ (Connection timed out)
22:26:41preglowi'm looking forward to gcc 4.3, that's for sure
22:26:51Juice^barrywardell: super. the update for the time is now working on the sansa
22:27:54jhMikeSbe sure to delete the ".s" but not ".S" :)
22:28:35preglowjhMikeS: if i've named a file something.tmp.something.S, i get what i deserve
22:29:12jhMikeSwhat about lcd-as-x5.S?
22:29:13 Join Bjoern-Erik [0] (
22:29:31preglowsweet, sweet lord, tell me i am not seeing this
22:29:39jhMikeSwhat's that?
22:29:48preglowjhMikeS: the line i pasted should delete stuff with .tmp. in the name, it's not based on extension
22:30:03 Quit Bjoern-Erik (Read error: 104 (Connection reset by peer))
22:30:09preglowjhMikeS: it seems gcc didn't remember the result of the multiple identical DIV64 calls in eq_pk_coefs
22:30:10Lloreanpreglow: You're not seeing it. But, assuming you were seeing it, what would you be seeing?
22:30:21preglowthat's just... nasty
22:30:55preglowi should stop reading disassemblies
22:30:57jhMikeSoy, but as long as you do I suppose that's good enough for it
22:31:10preglowespecially for non-critical code
22:31:17 Join efyx [0] (
22:31:48jhMikeSit's done deliberately to silently encourage the use of human written assembly ;)
22:31:51 Join Kitt0s [0] (
22:32:31preglowi don't get this, gcc is also stuffing my eq coef results away two places, but only one is needed
22:33:44jhMikeSI find gcc to be a bad common subexpression optimizer. it usually needs doing explicitly and makes a huge different. That includes multiple references to a particular array element with the same variable.
22:34:14 Join Bjoern-Erik [0] (
22:34:33jhMikeSthat or a flag needs turning on
22:34:45 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
22:35:13preglowjhMikeS: i always thought it did that decently, but i see i'm wrong
22:35:18preglowi'll rectifiy it in the code for sure
22:35:46jhMikeSfortunately it seems good at optimizing away the temp variables
22:37:06Mikachuwhy do all cs teachers always say that compilers are better at optimizing than people?
22:37:30preglowMikachu: i guess they're not using gcc or doing code that really needs to run fast
22:37:47preglowgcc does a pretty nice job for x86, that i know
22:38:16preglowi can write better code for both coldfire and arm than gcc does any day
22:38:48linuxstbpreglow: Volunteering to act as build server?
22:39:07preglowi might pull the average build time down somewhat...
22:40:29jhMikeSbut I guess it all comes down to how good human are at optimizing compilers
22:40:57 Join amiconn [0] (
22:41:36mattzzA question regarding plugins: If rb->yield() is used, playing music in parallel should be no problem, right?
22:41:43barrywardellJuice^: great. let me know if you find any more bugs with it. i'm sure there are still some in there!
22:41:44preglowfrom 342 bytes to 262 bytes with just that....
22:41:50*preglow kicks gcc
22:42:08jhMikeSpreglow: deliberately putting common results in variables?
22:43:03rp-barrywardell: i'm sure that the freeze with the adc_scan() battery code, has something todo because the adc_tick runs in a seperate thread
22:43:28rp-but i can't imagine what
22:44:03preglowjhMikeS: yes
22:44:27preglowweirder still: it still contains the same amount of calls to the div routine
22:44:29jhMikeSmattzz: the plugin must yield explicitely or call functions that cause a thread to yield or sleep implicitely, music or not.
22:44:47jhMikeSpreglow: but you saved the div result?
22:45:21Juice^barrywardell: I will :)
22:45:59mattzzjhMikeS: Everything is fine until my ajr accesses the disk to reload the mp3 data buffer
22:45:59jhMikeSnow that's extra braindead
22:46:00preglowjhMikeS: and the size gain is there, so god knows what happened
22:47:00jhMikeSmattzz: does it try loading everything in one shot? I'm not clear what you're up to.
22:48:06mattzzjhMikeS: I am running my rocklife plugin which uses yield() while playing music. When the disc gets accessed, music stops afer a while and the red LED starts to blink
22:48:35barrywardellrp-: hmmm. you could get rid of all that adc_tick and adc_scan stuff and just put your code in adc_read(). adc_read() is the only function called by rockbox iirc
22:48:38jhMikeSmattzz: might not be yielding enough to let it rebuffer
22:49:16mattzzjhMikeS: how do I yield enough in a loop, do I need usleep additionally?
22:49:23 Join Wiwie [0] (
22:49:38pregloww00t, usb code
22:49:46rp-barrywardell: ok thx, i will try that
22:50:11linuxstbpreglow: Yes, it looks nice... It's mostly barry's work, but I like the commit stats :)
22:50:21jhMikeSmattzz: could call yield more than once. disperse the calls.
22:50:45mattzzjhMikeS: OK, will try that.
22:50:50amiconndan_a: Your sansa disk fix is .... strange
22:51:13preglowjhMikeS: this really does boggle the mind, every single FRACMUL that is the same gets calculates several times by gcc
22:51:50 Quit Mouser_X (Nick collision from services.)
22:52:01 Join Mouser_X [0] (
22:52:02jhMikeSthose are assembly blocks though...somehow I'm not surprised
22:52:16preglowjhMikeS: also, they're marked volatile. that might be it
22:53:03jhMikeScould be but not sure if that'll get rid of them. probably should just condense the same and save it.
22:53:06Juice^would be nice if the properties "window" of a jpg file showed info about the imagesize and bits
22:53:21Hans-Martinmarkun: wow - I disassembled that 2x4 mm reset switch, repaired it (I *had* used the wrong tool to press it - a needle), reassembled it, and it seems to work again :-)
22:53:41preglowjhMikeS: nope, wasn't that
22:54:03markunHans-Martin: fhew! No keep a paper clip in your wallet :)
22:55:28amiconnpreglow: gcc bein g braindead is nothing new nor is it limited to coldfire; I've come across a similar stupidity when checking the sansa disk transfer loop
22:55:40Hans-Martinmarkun: yup. but first I'll reassemble the iriver. shouldn't be a problem, all other parts are really clunky in comparison :-)
22:56:30markunHans-Martin: which one do you have?
22:57:22preglowamiconn: not new, no, but the depths of its idiocy will never stop boggling me, it seems
22:57:26jhMikeSgcc needs it's hand held a little bit :)
23:00:04 Join miepchen^schlaf [0] (
23:00:19Hans-Martinmarkun: iHP-120
23:00:24markunme too
23:00:31amiconndan_a: I think some timing isn't set correctly in the pp5024 flash sd interface, and that causes the occasional write errors in the aligned loop (which you cut away now)
23:00:49amiconnDid you try putting a nop *between* the memory load and the sd write?
23:00:52Hans-Martinmarkun: I got it from a coworked who wanted an ipod because of the bigger disk
23:01:02Hans-Martinoops coworker
23:01:17amiconnAlso, did you try defining DATA_REG as *(unsigned short *) ?
23:01:30Hans-Martinmarkun: an running it with rockbox really rocks :-)
23:01:31amiconn*(volatile unsigned short *) even
23:02:20barrywardelllinuxstb: nice to see that usb code helped a bit. I have an updated version on my work machine, so I'll have a look at updating SVN if that's OK with you?
23:02:37barrywardell Also, I think I rememer setting the USB controller to run without doing anything else caused the USB stack on my mac to crash
23:02:54barrywardellso we'll have to watch out for reports of that
23:03:07SoapBiptoN: I have noticed no discernable noise with iPod recording - tomorrow when I am done with my battery bench tests I will record "silence" with and without my line-in dongle and report on the dB level of the noise floor.
23:03:22markunHans-Martin: now that you know how to disconnect the battery you could upgrade it. The 2200mAh ipod 1/2g batteries fit and are very cheap on ebay.
23:03:28dan_aamiconn: I've tried the *(unsigned short *), but not tried with the nop in a different place.
23:03:37 Part TrueJournals
23:03:49rp-where can i find the formular thats used to display the numerical battery level?
23:04:13Hans-Martinmarkun: yup but until now I'm very happy with the battery life
23:04:23barrywardelldan_a: waiting for the cop to be awake still doesn't give me a main thread on the cop on my H10
23:04:45barrywardellrp-: in apps/powermgmt.c
23:05:06linuxstbbarrywardell: Sure. You also mentioned something about running a stop() function?
23:05:37rp-barrywardell: powermgmnt is in firmeware/, but thank you :)
23:05:45Hans-Martinmarkun: btw, which size torx do you use for the screws? I used a T5 but it feels like it's a little small. but T6 is definitely too big
23:05:53barrywardellrp-: oops, yes. that's what I meant :P
23:07:08barrywardelllinuxstb: it's in the mx31 driver linked from the mx31.h file
23:07:09***Saving seen data "./dancer.seen"
23:07:16 Quit Thundercloud__ (Connection timed out)
23:07:42 Join Soul-Slaye1 [0] (n=Jonno@
23:07:45 Quit Mouser_X (Nick collision from services.)
23:07:51barrywardellUOG_USBINTR =0; UOG_USBCMD &=~USB_CMD_RUN_STOP; should be all that's needed
23:07:54 Join Mouser_X [0] (
23:08:14barrywardelli haven't really tested that though
23:09:03markunHans-Martin: I had a small flat screwdriver which also did the trick
23:10:26 Join wooo [0] (
23:10:54Hans-Martinmarkun: yup should work as long as the screws don't fit too tightly. - I had the torx tools anyway since I needed them for my old Powerbook 5300 (RIP) :-)
23:11:26BiptoNsoap: thanks, that'd be a good thing
23:11:51BiptoNhas anyone noticed that when you turn the X5 off, the battery seems to 1-800-DIE?
23:12:33Soul-Slaye1Same on the H10 I find :$
23:12:43LloreanBiptoN: Is there any chance you could be talked into just speaking clearly without strange things like that?
23:12:54SoapBiptoN: could you do the same on your 4 and 5.5? I'm really curious.
23:13:18BiptoNsoap: i recorded on two different 4g grayscales' and a 5.5g
23:13:43Soapok - I'm just curious what the noise floor number is in dB.
23:13:47BiptoNthe 4g's both had quite a bit of signal being picked up on both channels with and without the dongle
23:14:07BiptoNthe 5.5g only had a little on the right channel, but it's the same noise
23:14:22Hans-Martinmarkun: ok, player reassembled, everything's cool - just haven't got enough time tonight to continue the project which requires a working reset button :-(
23:14:29BiptoNllorean: no problem, I'll try for ya;)
23:14:34Soapok - if you have a chance I'll really dig quant over qual
23:18:09BiptoNdo you want me to pull it up in audacity Soap?
23:18:25BiptoNi don't have a working 4g anymore, but i can get dB on the 5
23:18:31Soapwhatever you are most comfortable with.
23:19:13Juice^barrywardell: did you see the latest post the forum about the screen going black on the sansa? with the strange "fiber" looking colors then going to complete black
23:19:18SoapI'm more curious than anything - I guess if we had enough data we could determine with reasonable certainty if your problem is unusual.
23:19:48preglowjhMikeS: the call to divsdi did reduce by one, i was looking at the wrong listing...
23:22:00 Nick Everybody is now known as Everybody|away (
23:22:13barrywardellJuice^: yes I saw it. under what conditions did that happen?
23:23:47jhMikeSpreglow: yes, the right listing is usually the correct one :)
23:27:17Juice^barrywardell: it happend a few times when booting up a rockbox from 25-01-2007, not 100% sure if thats the date, but i think it has to be
23:29:38barrywardelli've had problems with that myself on occasion too. It seems to bootloader doesn't fully load sometimes. Probably a problem with the LCD driver
23:30:48 Quit inversions (Read error: 110 (Connection timed out))
23:32:10Juice^barrywardell: that can be the problem yes..
23:33:04 Join shnee_ [0] (
23:33:06barrywardelli think it has been around since the LCD driver was written.
23:33:30 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
23:33:48Juice^ok. nice to know
23:34:35 Join w1ll14m [0] (
23:35:45 Quit funky ("leaving")
23:36:43n1sheh, bringing up the context menu in an empty folder and selecting properties segfaults the sim :-) on target it just plain don't work
23:37:07*jhMikeS is finding it tempting to go optimizing the DSP scheme but should probably just put in the resampler clicking fix to keep the change focused
23:37:19preglowjhMikeS: i'm planning on looking at that myself one day
23:37:24preglowsharing ideas might be nice
23:37:47preglowthere's quite a bit of unnecessary stuff in there for sure
23:38:04jhMikeSI'm full of them so it's tempting to just go rewriting everything
23:38:47 Quit Mouser_X (Read error: 110 (Connection timed out))
23:38:50jhMikeSSo instead I'll go changing playback.c, pcmbuf.c, and dsp.c and headers and not get the actual bugfix done :P
23:41:30 Quit My_Sic (Read error: 104 (Connection reset by peer))
23:41:30markunHans-Martin: what are you working on?
23:41:51jhMikeSI'd like to kill all the repeated conversion bytes<=>samples
23:42:41 Join Mouser_X [0] (
23:42:42Hans-Martinmarkun: processing audio input and playing it on output, hopefully resulting in some kind of digital guitar effects thingy
23:43:07markunHans-Martin: I think preglow was thinking about something similar
23:43:27barrywardelldoes anyone else get the recording screen stuff stuck on the statusbar, even in the wps?
23:43:59preglowjhMikeS: oh yes, oh yes
23:44:05preglowjhMikeS: confusing it is too
23:44:09Hans-Martinmarkun: yeah I've heard that, too. but since this is a fun rpoject, I just want to do it - because I want to have the fun :-)
23:44:15preglowjhMikeS: but i think killing the resampler bug first would be wise
23:45:05jhMikeSpreglow: agreed
23:45:28markunHans-Martin: he has too much fun with gcc anyway ;)
23:45:52 Part n1s
23:45:58Hans-Martinmarkun: and right now I can put the newly-working reset switch to good use since I am not able to get recording to work without locking up the player...
23:45:58jhMikeSnot to say there won't still be a bug went downsampling small data but I haven't seen it come up and won't kill anything.
23:46:12 Quit subson ()
23:46:43preglowjhMikeS: downsampling small data?
23:47:06 Join Administrateur [0] (
23:47:11 Join My_Sic [0] (
23:47:11AdministrateurYo evry1
23:47:19 Nick Administrateur is now known as Ar53n (
23:47:30Ar53nI need help , could someone help me?
23:47:39preglowyou need to ask
23:47:40preglowthen see if help comes
23:47:51Ar53nis there GBA emu for rockbox
23:48:01Ar53n? =)
23:48:15Ar53ngosh =(
23:48:23Ar53nthx for answer
23:48:23jhMikeSIf you pass just a few samples (or one), how can you guarantee output. I mean how could you put one sample into a downsampler and expect to get a sample back every time. It seems the situation of tiny remnants on the last loop could occurr.
23:48:33 Quit Juice^ ("Leaving")
23:48:33 Join inversions [0] (
23:48:56 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
23:49:21Ar53nanother question , i've installer many addons on in my doom folder such as counterpod ( doom/addons ) after i've chosen the addon and lunch the game , it crash and goes back to the plugin menu
23:49:51LloreanDoes it say something right before it exits?
23:50:26Ar53n=/ it just goes back to the plugin menu =(
23:50:47LloreanSo, when you tell it to launch the game, you don't get *any* text on the screen?
23:51:00LloreanBecause usually Doom has quite a bit of text while it starts up.
23:51:36Ar53nyeah i does wait i'll look
23:51:52jhMikeSpreglow: I mean, it's a contradiction if you downsample and get as many samples back as you put in, no?
23:51:55Ar53n*it ddoes
23:52:19LloreanAr53n: Which main wad are you using?
23:52:28LinusNamiconn: i'm investigating the weird outletm3 player
23:52:47Ar53n*you cannot - file with the shareware version
23:52:55preglowjhMikeS: most of the time
23:52:58Ar53nthat wut it says
23:53:09LinusNno matter what i do, the cpu frequency is doubled
23:53:09LloreanAr53n: Yes, you cannot use addons with the shareware Doom. You have to have a full Doom wad.
23:53:09 Quit miepchen^schlaf ("Verlassend")
23:53:10preglowjhMikeS: but with a small downsample ratio, yes, it is possible in the short term
23:53:19Ar53nmmmm ok i see!
23:53:19LloreanAr53n: And "wut" is not a word.
23:53:26Ar53nwell lthx Llorean! =)(
23:53:33Ar53nbig hug for u <3 (k)
23:53:36LloreanAr53n: Try using Freedoom.
23:53:54LloreanName it doomf.wad or doom2.wad and it should work with the addons.
23:54:25Ar53nok ! ill try it n tell u if it works!
23:54:34mattzzjhMikeS: fyi - adding another yield() did the trick. thanks
23:55:08jhMikeSmattzz: good to hear it worked :)
23:55:21*mattzz wasn't aware of a friendly yielding policy
23:55:43barrywardellhmmm. the recscreen_on setting doesn't seem to be getting saved properly
23:56:34 Quit lee-qid ("aufwiederbyebientotsayonara")
23:58:10 Quit bluebrother ("sleep")
23:58:58Ar53nLorean i downloaded freedoom , i renamed it doomf n when i lunch it says missing base wad

Previous day | Next day