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

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

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

#rockbox log for 2007-03-10

00:01:08euz_Llorean, what does it mean when they say CF IDE SATA Ethernet, SD, usb, etc... have the same standard?
00:01:43euz_same standard in relation to controller?
00:02:40LloreanI have no idea. I don't even know where you're getting this from.
00:03:12 Join mattzz [0] (
00:06:12 Join gursikh [0] (
00:06:14 Quit K3nto ("Leaving.")
00:06:24gursikhOMG I LOVE ROCKBOX!
00:06:26 Part gursikh
00:06:28euz_I have no idea where I saw it, but I remember I thought it was weird
00:07:24LloreanIt could just be gibberish.
00:07:24deadheadLlorean, great news: new bootloader, latest build: everything is ok! The new bootloader installer really ROX. Kudos to chapman and all the rockbox team!!!
00:07:31euz_in the sense that hardware controllers follow a single standard
00:09:47bluebrothereuz_: I guess it simply was a list of supported connections by one chip / board
00:10:07muesli__new bootloader for what target?
00:10:13 Join decayedcell__ [0] (n=decayed_@
00:10:55deadheadmuesli__, ipod mini 2gen
00:10:56Lloreanmuesli__: The iPods
00:11:02muesli__ah ok
00:11:10muesli__not my cup
00:11:10 Join SirFunk [0] (
00:11:30 Quit Hoffmann ("Ciao")
00:11:30preglowdoes profiling work for arm at all?
00:11:47deadheadmuesli__, the one I used was very old , I installed it all by my hand with dd and instructione more than 6 months ago
00:12:08euz_I was wondering how hard it would be to connect SDHC to H120
00:12:14 Join cmptrgy412 [0] (
00:12:23muesli__wasnt the mini 2gen somehow encrypted?
00:12:31Lloreanmuesli__: Nano 2nd gen.
00:12:35muesli__ah ok
00:12:51muesli__i guess no solution was found so far?
00:13:13euz_if you swapped the IDE controller for example, would he rest be mostely software related, etc...
00:13:13LloreanNobody's even working on it here, as far as I know
00:13:34deadheadthank you guys for your help and your time ;) night
00:13:45 Quit deadhead ("Sto andando via")
00:14:11 Join Hoffmann [0] (
00:14:16muesli__i guess no good news since the upcoming ipods will feature the same
00:14:20 Quit Hoffmann (Read error: 104 (Connection reset by peer))
00:14:29 Part cmptrgy412
00:14:56euz_so much wasted time, money and cpu cicles with all these protections
00:15:09 Quit decayedcell__ (Remote closed the connection)
00:15:21 Quit midkay ("Leaving")
00:15:43 Join decayedcell__ [0] (n=decayed_@
00:16:00euz_with all the ipod logos on the case, how much money could they lose if you didn't see one on the screen?
00:17:17euz_look at what microsoft did with xbox360, bleeding kernel encrypted inside proprietary IBM CPU
00:17:53euz_I'm sure research will be used to screw more people later on, but it's absurd
00:18:26 Nick [Tesser-Away] is now known as [Tesser] (n=Tesser@unaffiliated/tesser/x-000001)
00:18:51 Quit obo ("bye")
00:19:03 Quit desowin ("use linux")
00:21:15Jonnexwhere can i download the rockbox gameboy plugin?
00:21:22decayedcell__its built into Rockbox
00:22:08 Quit webguest66 ("CGI:IRC")
00:22:44Jonnexwould i have to put the gb rom in a special folder?
00:23:03 Quit stamppot ("CGI:IRC")
00:23:26peturJonnex: just play it like a music file
00:23:27euz_for organizational purposes only :)
00:23:41Jonnexsweetness :D
00:24:22euz_would be if nintendo didn't held back on their consoles so much
00:25:03euz_no NES emulator?
00:25:31 Quit x2jmp (Remote closed the connection)
00:25:34euz_jonnex. drop by for a list with the best games for the system
00:25:43euz_it's usually helpful
00:25:44 Quit robin0800 ("An error? Impossible! My modem is error correcting.")
00:27:15Jonnexdoes it run any GBC game?
00:27:37euz_gb classic?
00:27:43linuxstbJonnex: Yes.
00:27:50euz_you want to play tetris? :)
00:28:31Jonnexhehe naaah, but perhaps some zelda!
00:30:09 Quit linuxstb ("Chatzilla [Firefox 1.0.2/20050317]")
00:34:08euz_which player has the best signal to noise ratio?
00:34:29 Part decayedcell__
00:34:50 Join decayedcell__ [0] (n=decayed_@
00:35:14 Join K3nto [0] (
00:35:21Lloreaneuz_: I don't think anyone here knows. SNR is hardly the only important aspect of sound quality, and may not even be a relevant piece of information for portable use.
00:35:23 Join web-taz [0] (
00:36:33euz_oh, which is deemed to have the best sound quality?
00:36:47peturH1x0 has digital in/out
00:37:07euz_I thought it was relevant, H120 SNR is 90
00:37:08K3ntohas anybody heard of plugging an ipod into the computer and the computer saying that it did not recognize it? this was in windows. in ubuntu nothing happens. Th ipod goes into disk mode, and says ok to connect after a breif "do not disconnect"
00:37:31 Quit web-taz (Client Quit)
00:37:33euz_dmesg doesn't say anything at all?
00:38:33Lloreaneuz_: Many people think it's relevant, but in many cases you're listening at volumes where it's not relevant in the slightest.
00:38:49 Quit lee-qid ("aufwiederbyebientotsayonara")
00:39:24 Join pearldiver [0] (
00:39:53 Quit decayedcell__ (Remote closed the connection)
00:40:03 Join decayedcell__ [0] (n=decayed_@
00:40:28euz_volume is a question of volts or amps?
00:40:44K3ntoalright, sweet
00:40:55euz_what does dmesg say?
00:40:58K3ntoi found a theme that works and is nice.
00:41:58 Join robin0800 [0] (
00:42:15K3ntoLlorean: rockbox is skipping to the next song out of no where.
00:42:33K3ntolike before the current song is over
00:42:57euz_heh, that for me meant disk problems
00:43:13K3ntoyeah i think my ipod is broken anyway
00:43:23K3ntohas anybody heard of plugging an ipod into the computer and the computer saying that it did not recognize it? this was in windows. in ubuntu nothing happens. Th ipod goes into disk mode, and says ok to connect after a breif "do not disconnect"
00:43:34K3ntothats happening also ^^
00:43:42euz_that's a problems that works nicely on pc for recorvering problematic disks
00:43:42 Join muesli- [0] (n=muesli_t@
00:44:05euz_HDD regenerator, but I could find a way to connect mine to the pc
00:44:19euz_couldn't make the connector
00:44:23 Nick datachild` is now known as datachild (
00:44:45euz_I recovered a HDD and used it for another 6 months without any glitches
00:44:45 Quit muesli__ (Read error: 104 (Connection reset by peer))
00:44:56K3ntohdd regenerator eh
00:45:03euz_then recovered it again, recovered data, and gave up on HDD
00:46:04euz_logically there's no such things as fixing a hardware problem through software, but who knows
00:47:02euz_today I might have say more than one exoteric thing, I'm sorry, not always like this
00:47:42euz_listening to Wagner, maybe motif is sipping through language barrier
00:48:37toffe82something interesting, can we put rockbox on this : :)
00:49:26toffe82"For the true geek, the electronics are "hackable," the case is removable"
00:49:34 Quit bluebrother ("leaving")
00:49:45euz_hdd regenerator, if you find a way to connect the ipod's hdd to your pc, you should give it a try
00:50:29euz_worked for some six months, didn't make a single of those horrible disk seeking on error noise
00:50:30K3ntoi can connect it. i just have to restart
00:50:41euz_you need to connect it to ide
00:50:50K3ntofor real
00:51:04euz_it's ipod connect proprietary?
00:51:19K3ntoi can go into disk mode lol
00:51:22decayedcell__the HDD connector is IDE
00:51:31euz_damned slang, is always beyond me
00:51:32decayedcell__but its like a 1.8 inch drive?
00:51:37decayedcell__needs an adapter
00:51:43scorche` is a ZIF connector
00:51:49decayedcell__ah okay
00:52:03euz_you can always put out the soldering iron :)
00:52:52K3ntoman, that gets annoying. not even hearing the whole song...
00:52:56euz_I remember soldering 42 wires modding a xbox
00:53:24euz_makes you want to cry
00:53:44LloreanK3nto: Try running a scandisk on your device.
00:54:03euz_I would recomend you to format the disk, but after I did that, all my space went to hades
00:54:12K3ntoim on linux
00:54:36K3ntoi formatted like 2 days ago
00:54:40euz_scandisk is useless
00:54:43Lloreanfsck then.
00:54:55Lloreaneuz_: There are errors it can detect.
00:55:25euz_file system errors
00:55:32euz_what's the fs on ipods anyway?
00:55:44decayedcell__FAT32 on the WinPod
00:56:02LloreanFAT32 on any iPod running Rockbox.
00:56:27euz_oh well
00:57:11euz_anyone ever heard about this player?
00:58:59euz_it would be nice to have Rockbox on a DIY network based player
00:59:20LloreanRockbox comes to players when people who own the player choose to port it, pretty much
00:59:25scorche`the port it to one and implement a network stack
00:59:34toffe82this one is better
01:00:29euz_that's prolly one of those one chip does all trinkets
01:00:41euz_usually horrible
01:00:56euz_s1mp3 or something
01:01:23 Join Joshizzle [0] (
01:01:30Joshizzlehi everyone
01:01:48euz_I saw one with organic led, pretty impressive
01:01:57 Quit Joshizzle (Read error: 104 (Connection reset by peer))
01:02:09euz_too bad pixels die with time, brighter the color the faster
01:03:05euz_scorche, what do you mean by network stack?
01:03:16 Join Joshizzle [0] (
01:03:30Joshizzledoes anyone know where I can get ipod_fw.exe?
01:03:30toffe82check this one and what they say : " Please note this is NOT the actual iPod Nano. It is a generic MP4 Player. "
01:04:03LloreanJoshizzle: Rockbox doesn't require that for installation.
01:04:12Joshizzlebut I do for the dual boot
01:04:24JoshizzleRockbox doesnt not support video well
01:04:29Joshizzlebut the iPod does
01:04:37LloreanYes, but Rockbox's bootloader dual boots.
01:04:45JoshizzleI dont thin kso
01:04:48LloreanI know it does.
01:04:59scorche`read the FAW
01:05:03JoshizzleI can just unplug my ipod from my usb
01:05:03scorche`FAQ even
01:05:05 Nick scorche` is now known as scorche (i=ScorchE@rockbox/administrator/scorche)
01:05:11Joshizzleit will reload to iPod
01:05:17Joshizzlei mean rockbox
01:05:29scorcheas i the FAQ
01:05:38LloreanJoshizzle: "I don't know how to dual boot" is not the same thing as "It's not able to dual boot"
01:05:39Joshizzlethe FAQ on rockbox's site?
01:05:39peturfrequently asked What?
01:05:41Joshizzlei am
01:05:57Joshizzlei cannot find the section on dual booting
01:06:01Joshizzlei searched it
01:06:10scorchethere is alink right in the topic...
01:06:15Joshizzleill check out what you just sent me
01:06:24euz_as mentioned above, turn on, and switch the hold button right afterwards :)
01:07:15euz_the ipod 5th gen is only 80mhz?
01:07:15euz_multiple cpus
01:07:31LloreanTwo cores, that we run at max 75mhz.
01:07:55Joshizzlethank you all
01:08:11Joshizzlenow I just got to unprotect all of my bought songs
01:08:24 Quit Joshizzle ("Chatzilla 0.9.77 [Firefox]")
01:08:29euz_when all else fails, read the manual :)
01:08:53scorcheLlorean: i will never understand why people think we lie...
01:09:08euz_motd wisdom is getting to me
01:09:11scorcheeuz_: should have already read the manual before using rockbbox ;)
01:09:32scorchein fact, we require that you have read that before asking a question
01:09:37 Quit mattzz ("Leaving")
01:09:42euz_just quoting motd
01:10:08***Saving seen data "./dancer.seen"
01:10:19euz_it's nice to read the manual, but not as nice as having someone else reading it for you :)
01:10:46euz_we could be like those people on the end of Farenheit 451
01:11:09euz_each knows a manual, I for example I'm pretty familiar with glftpd manual
01:11:35scorchebut this is #rockbox....where only rockbox related discussion is on topic
01:12:41euz_heh, have you ever heard about that ftpd?
01:12:56scorchelet me say that again....
01:12:58scorchebut this is #rockbox....where only rockbox related discussion is on topic
01:13:40euz_I'm just musing, very dully indeed, but still
01:15:02scorchethis isnt a social channel
01:15:29scorcheso we really dont care if it is "dully" or not
01:15:50euz_well, most of us wouldn't hang in a social channel for the world
01:16:18Lloreaneuz_: The point is, the channel is logged, and random musings make it harder for people to read through the log to see what's happened since they last joined the channel
01:16:59 Quit muesli- ("ich will KŘhe!!!")
01:17:25euz_the word dully actually exists :) ok, I'm really sorry
01:18:53euz_did you read about that s1mp3 player?
01:19:05LloreanYeah, they've been around for a little while.
01:19:16euz_it's quite common I hear, 1 chip does all
01:19:17LloreanAs far as I know they're not interested in working with Rockbox because we're GPL and they want to go BSD.
01:19:27 Quit robin0800 ("Famous last words")
01:19:32euz_original software is hell on wheels
01:20:04 Join midkay [0] (
01:20:18euz_those insidious chinese...
01:20:33 Join webguest56 [0] (i=cbdd34d5@gateway/web/cgi-irc/
01:20:47 Quit roolku ()
01:21:30euz_it's prolly custom made, should be trick to port
01:21:58euz_it's the only chip on the player other than memory and fm(optional) controller
01:22:00 Join robin0800 [0] (
01:22:47 Quit webguest56 (Client Quit)
01:22:57 Quit robin0800 (Client Quit)
01:23:51euz_open source needs some open hardware initiatives
01:24:45euz_like XY plotting routers for making pcbs
01:27:16euz_well, good night, I never acknowledge as sociable, but comparasion to such stoics as yourselves, props :)
01:27:18 Join robin0800 [0] (
01:27:31 Quit euz_ ("Leaving")
01:27:56 Part toffe82
01:27:59 Quit inversions (Read error: 110 (Connection timed out))
01:30:17 Quit robin0800 (Client Quit)
01:31:49 Join barrywardell [0] (
01:32:49 Join robin0800 [0] (
01:33:38scorchepetur: what i didnt tell him was that beer is highly on-topic with certain members of the communitity ;)
01:33:55 Quit robin0800 (Client Quit)
01:34:01peturshhhh you'll wake up preglow
01:34:34 Join Thundercloud_ [0] (n=thunderc@
01:36:21*dan_a_ thinks he deserves a drink
01:37:05dan_a_If I try to set the volume on one of the DAC channels on the Sansa... the screen goes black
01:37:53markunpreglow: how's the new resampler going?
01:40:35 Join Thundercloud__ [0] (n=thunderc@
01:41:22 Quit ender` (" The likelihood of Perl being involved in a system is directly proportional to the length of time the system has been in maint")
01:47:57 Quit Thundercloud_ (Read error: 145 (Connection timed out))
01:47:57Lloreandan_a_: That sounds like an unusual effect.
01:48:26 Quit jhMikeS (Read error: 104 (Connection reset by peer))
01:49:08dan_a_Llorean: I've never written an audio driver before - perhaps it happens all the time
01:49:38Lloreandan_a_: I suppose that's fair. Neither have I. :)
01:50:40preglowmarkun: i'm trying to tweak a few more cycles out of it
01:50:51preglowbut not as fast as i wish it would be
01:51:07markunpreglow: how's the quality?
01:51:15preglowah, it's not new like that
01:51:21preglowi'm just recoding the one we have in asm
01:51:33preglowlike jhMikeS did for coldfire
01:51:46*barrywardell remembers how reading the RTC on the Sansa caused the LCD to not work.
01:52:06Lloreanpreglow: Don't the ARM targets actually support a lot more in the way of real sample rates they can output at?
01:52:15 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:52:17dan_a_barrywardell: Can you remember why?
01:52:47preglowLlorean: they do indeeed
01:53:07barrywardellI think it was a timing issue. I never figured out the real reason, but reading it too soon seemed to cause the problem
01:53:10preglowbut that would mean more delving in playback.c than i'm qualified to do
01:53:48LloreanWell, there's also the fact that a good ARM resampler is a good thing to have anyway, there's no guarantee it'll always be true that ARM targets will have more real sample rate output options.
01:53:56barrywardelldan_a_: also, the lcd seems to not work some times when starting up with current svn.
01:54:35preglowLlorean: yeah, and there's the pitch screen which also uses it
01:54:56preglowLlorean: dynamically changing sample rates when possible will mean glitches you don't want when beatmatching
01:55:05LloreanAh, forgot all about the pitch screen.
01:55:09preglownot that i actually think anyone uses rockbox for that, but i like to think i'm useful...
01:55:15dan_a_barrywardell: I've noticed that (it gets an odd, organic pattern as it fades out, or it's plain white.) Unfortunately it's not often enough for me to be able to try to fix it. My guess is that some of the pauses need lengthing
01:55:21LloreanI've used the pitch screen a few times to speed up audiobooks.
01:55:36preglowwhich bloody reminds me of another thing i should work on
01:55:40preglowtime stretching
01:56:04barrywardelldan_a_: there may also possibly be a problem with udelay() - it's not wrap safe
01:56:37barrywardelldan_a_: but that shouldn't cause problems that are so frequent
01:57:34markunpreglow: any plans for a better resampler?
01:57:59 Join Shaid [0] (
01:58:14preglownot really, no
01:58:22markunpreglow: how's the one added to speex? Did you look at it?
01:58:28preglowhad a look, looks expensive
01:58:52preglowsomethings are struggling even with the one we currently have
01:59:02preglowand that's as low complexity as things get without sounding like complete shit
01:59:43preglowin a bag or not
02:00:43 Quit Thundercloud (Read error: 110 (Connection timed out))
02:01:02dan_a_I've got FIQs!
02:01:04 Quit pearldiver (Read error: 110 (Connection timed out))
02:01:29markunpreglow: we're using linear interpolation now, right?
02:01:33preglowdan_a_: working on sansa sound already?
02:01:35preglowmarkun: aye
02:01:50preglowmarkun: whatever scheme we use in the future will probably have to be used in conjunction with that
02:01:58dan_a_preglow: Why not? :D
02:03:22preglowgot your hands on the datasheet yet?
02:04:22dan_a_I have. It's so nice to have one for a change
02:04:48jhMikeSpreglow: you tried a codec from the latest svn?
02:05:13preglowjhMikeS: mentioned the results soon after you asked me
02:05:48jhMikeShuh...and I replaced my whole build, cleared settings and still get that
02:06:00preglowbuild from
02:06:28jhMikeSsvn, no mods
02:07:01preglowwell, it seemed perfectly ok here
02:07:02jhMikeSmaybe I should do that...not build myself
02:07:11preglowhaven't got any high bitrate oggs
02:07:22 Part pixelma
02:07:27preglowin a related note, downsample arm asm really doesn't seem to gain much :/
02:07:33jhMikeSIt's just that every single other thing is unaffected
02:07:40preglowi'm trying out a hack to shave two cycles off each mul now
02:07:42dan_a_It's really frustrating, the screen only working while music is playing...
02:08:11preglowdan_a_: cool, cool
02:08:18preglowdan_a_: looking forward to the gentlemen mail :>
02:08:21jhMikeSthe coldfire gain were just off the chart :\
02:08:39preglowjhMikeS: coldfire just seems to gain that much more from asm
02:08:39dan_a_preglow: There's no sound yet, but I know it's in there somewhere
02:09:19preglowjhMikeS: anyway, we'll see, now i've written it, i'm not just going to delete it
02:09:29Lloreanpreglow: The speex resampler is actually slightly broken, I think
02:09:33jhMikeSDo you test with the SPC codec?
02:10:09Lloreanpreglow: If I play a less-than-44.1 file, it plays back more quickly than the file does on PC, by a small amount
02:10:12jhMikeSI use that one for gauge it usually
02:10:55barrywardelldan_a_: there must be a reason why the i2c and lcd are interfering with each other
02:11:14preglowLlorean: oh?
02:11:31dan_a_barrywardell: The power for the LCD is built into the as3514 (or at least the backlight)
02:11:33preglowLlorean: i don't think that's resampler related
02:11:38Lloreanpreglow: Yes. I'd noticed speex played back slightly too quickly (listening to a file for about a minute, the DAP would get about a word ahead in the audiobook)
02:11:42preglowjhMikeS: upsampling is work in progress yet
02:11:47*barrywardell looks at lcd-e200.c
02:11:49jhMikeSthat speex interpolator has might familiar things about it to me :)
02:11:57Lloreanpreglow: Well, the problem happens in 16khz files, but not 44.1khz files, and it only affects speex.
02:12:26LloreanThe files play fine in foobar2000, but in Rockbox (simulator, H120, Nano, or Gigabeat, take your pick) the 16khz files play a bit too quickly.
02:12:37preglowis the resampler in speex even enabled?
02:12:40preglowi doubt it
02:12:45barrywardelldan_a_: good point
02:12:53*barrywardell also looks at backlight-e200.c
02:13:09Lloreanpreglow: Well, safetydan suggested the resampler might be the problem, he said Speex uses its own.
02:13:58preglowLlorean: i think it has its own, but i dobut it's used
02:14:00Lloreanpreglow: If speex uses our resampler, maybe I'm wrong. I could double check with a 16khz file in a non-speex format, maybe I didn't test properly with that.
02:14:18LloreanEither way, 16khz speex files play back too quickly, 44.1 don't.
02:15:17preglowLlorean: had a quick look at the source, and it does not look like it's used
02:15:27LloreanAlright, I'll double check with a non-speex 16khz file
02:15:31preglowdo that
02:15:45preglowour resampler being a bit off is not impossible at all
02:15:51preglowthere's some rounding being done here and there
02:16:39jhMikeSone part is 65536?
02:16:51Lloreanpreglow: Ah, yes, it definitely happens with our resampler.
02:17:18jhMikeSwhat's the ratio of the speed difference?
02:17:19preglowjhMikeS: delta calculation might also be a bit off
02:18:24LloreanjhMikeS: I don't really have a way to calculate that.
02:18:36 Quit bonbonthejon (Read error: 104 (Connection reset by peer))
02:18:47LloreanWithout just playing files through on each device, and seeing how much earlier one ends than the other.
02:18:54 Join AriX [0] (
02:19:10 Join mcscruff [0] (
02:19:32preglowratio of difference would be a nice figure
02:20:01LloreanOddly enough I'm suddenly having a hard time reproducing it with this .spx file.
02:20:30LloreanBy 'hard time' I mean usually if I get them exactly in sync, I can hear it after 60 seconds.
02:20:54LloreanAnd they seem to be exactly in sync this time... =/
02:21:06jhMikeSstill enough bits you'd have to wait a long time for 1s difference in progress
02:21:34jhMikeSand besides, rounding errors should make it play slower
02:21:35LloreanjhMikeS: Usually 60 seconds of playback results in about a whole word's difference between the reference PC and Rockbox
02:21:46LloreanjhMikeS: And yes, Rockbox is slower with the 16khz MP3.
02:22:18 Quit barrywardell ()
02:22:32jhMikeSthat's too much for it to be the resampler, I have no doubt about that
02:22:32 Part mcscruff ("Konversation terminated!")
02:23:09 Join toffe82 [0] (
02:23:25LloreanjhMikeS: Well, it didn't seem to happen with the 44.1 khz file. But then, It didn't happen with the later 16khz file, so apparently I haven't done enough testing.
02:23:58 Part Jonnex
02:24:01jhMikeSright, and it shouldn't change either, should be fixed in ratio
02:24:14LloreanYeah, I now can't get spx to do it at all.
02:24:43jhMikeSsee how talking about things helps? :)
02:25:13LloreanIt's really frustrating, because it _definitely_ happens.
02:25:22preglowi'm even using we believe you!
02:25:27preglowyes, that came out right
02:25:43 Part AriX
02:25:48preglownote to self: erase what you were about to write before leaving the keyboard before writing something else
02:25:54jhMikeSoverboosting vorbis, and elusive speex timeshifting...just one of those days
02:26:19preglowtesting with a 96khz flac to see if my downsampler has any effect now...
02:26:25LloreanjhMikeS: Well, MP3 seems to timeshift too.
02:26:31preglowit doesn't look too good
02:26:59preglowcache has far too much effect on pp for my tastes
02:27:54jhMikeSwhich model? 3g?
02:28:15LloreanjhMikeS: Okay, so despite being able to reproduce it not five minutes ago, I can't do it at all now.
02:28:23preglowwith a wee bit of luck i'm speaking one percent boost difference here
02:28:49 Join perldiver [0] (
02:29:17jhMikeS1%? Guess I got the C code down pretty well if that's it.
02:30:20jhMikeSActually, optimized C code got it most of the way there and ASM another 4 or 5% maybe from that on SPC test.
02:30:37preglowand that's even bloody utilising the arm early multiplier termination
02:30:48jhMikeSBut I've got another 2 or 3% waiting
02:30:52preglowwell, now downsample is done upsampling should be faster
02:30:53LloreanOkay, now the files are perfectly in sync, sound perfectly in sync, but the progress is ~4 seconds different.
02:31:00*Llorean mutters
02:31:26jhMikeShow can that be?
02:31:29preglowjhMikeS: we really, really need some bloody way of benching dsp/codecs
02:31:36perldiverlast post, good question really
02:31:43preglowi'm tired of this fucking boost counter
02:31:59preglowwith its varying around
02:32:14jhMikeSWe run them unregulated
02:32:17preglownow it looks like i've perhaps got 2% difference
02:32:41*Llorean gives up on reproducing this problem, for the moment, since everything seems to refuse to work right now.
02:32:50preglowjhMikeS: all that's needed is just a mode where we can let rockbox push data through as fast as it can with timing
02:33:06preglowthat would make me a happy camper
02:33:10 Join Obsys [0] (
02:33:11Lloreanperldiver: That's one of those features that will be negated entirely by viewports.
02:33:27jhMikeSright, so skip DMA transfers
02:33:28perldiverah, viewports
02:33:33JETC-rockboy is now available to X5?
02:33:41JETC-:-o ?!
02:34:33preglowJETC-: has been for quite some time
02:34:41preglowignore me
02:36:55preglowjhMikeS: i'm a bit confused now, will this work: result = last + (highpart << 16) + (lowpart >> 16); when all i want to do is add "last" and the 32 bit number i piece toghether from "low" and "high" ?
02:37:27preglowi figure it will
02:38:00jhMikeSwhy do you split it?
02:38:18preglowjhMikeS: arm works that way, it gives you the full 64 bit result of a mul in two regs
02:39:22preglowthe mul is slower, though, but it does early termination based on how many bytes in the second mul operand is zero, and seeing as how "frac" is only sixteen bits, i put them in the lower two bytes, but that means i need to do some shifting like that to recombine the lower and higher 32 bits
02:39:28jhMikeSso dont we shift right by 31?
02:39:48preglowwe do, but when frac is shifted fifteen bits lower than usual, that shift right becomes 16
02:40:33jhMikeSshift frac in advance so the whole result is in one reg
02:41:09 Join Strath [0] (
02:41:12preglowshift it up, you mean? if i do that, i miss the entire point of early termination in the multiplier
02:41:37preglowfor every byte in the second operand that is zero, one cycle is shaved off the mul, so if frac is contained only in the two lower bytes, i shave off two cycles
02:41:53jhMikeSoh, see I'm not up on all this yet...but hell, one of those thing I'd reality test and go with the faster one no matter what.
02:41:57 Part decayedcell__
02:42:07 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
02:42:18preglowhow do i test this? the savings are too small to be easily measureable
02:42:19jhMikeSbut how many go into recombining it?
02:42:36preglowinstruction count is the same no matter what
02:42:51 Quit eggy ("<null>")
02:43:09 Join webguest46 [0] (i=4741521c@gateway/web/cgi-irc/
02:43:16preglowthe one instruction i gain in recombine i shave off from not having to do the extre shift in phase << 16 >> 1
02:43:28preglowthe first shift is replaced by an "and 0xffff"
02:43:36 Quit petur ("Zzzzzz")
02:43:54webguest46hello im a newb to rockbox and im wondering if i can get any more codecs?
02:44:03preglowwebguest46: which codecs do you need?
02:44:10preglowwebguest46: codecs don't come flying in the window, someone needs to cod ethem
02:44:13preglowcode them
02:44:17webguest46non specific just wondering
02:44:24webguest46i cant code
02:44:25preglowwebguest46: all codecs rockbox supports are included by default
02:44:33 Part Domonoky_
02:45:15jhMikeSMy paypal account is open to those who wish to purchase rockbox extras :)
02:45:34webguest46also im using the gigabeat version and i know this questions been asked a million times but is ther anyway to remove sat encryption?
02:45:36preglowa case of beer gets you far in my case
02:46:36preglowespecially if you're from the us
02:46:48Lloreanwebguest46: You have to use the Gigabeat software to copy the files back off your device
02:47:17jhMikeSI so wish I were up on the ARM stuff like the CF ... I might actually have some idea of what to do with it.
02:47:35preglowwell, not much reading is required, but it takes a while to get the hang of it
02:48:22webguest46well i already took them off my player to a backup on my c drive. now when i go to the gbroom the files dont show?
02:48:36webguest46whoops didnt mean to put a ? there
02:48:45jhMikeSusually the read to get the basics of instructions, then a good lookup. I ususally proceed by looking at mnemonics that look like the do what I need.
02:49:26preglowit ended up fairly close to yours after all
02:49:44Lloreanwebguest46: This really isn't a support channel for problems you have with the original firmware and software. I've never used it, and never wanted to use it.
02:50:19webguest46oh well i gotta go anyway
02:50:23preglowjhMikeS: the arm instruction set is very easy to learn
02:50:27 Quit webguest46 ("CGI:IRC")
02:51:32preglowjhMikeS: btw, small opt in cf downsample, the move.l before .dsloop can hardcode a fetch from 0, since pos is zero
02:52:08preglowmight save you 1 cycle and zero bytes, heh
02:52:11jhMikeSsee...emac instructions looked cryptic too until the very minute I needed them, then it was all just clear as glass. So, after the first real bit of code I write it'll probably be another "how could I not have understood since birth?"
02:53:22jhMikeSI don't doubt I overlooked something in all that
02:53:56jhMikeSwhich function?
02:54:15jhMikeShehe...I'm out of it right now :p
02:55:13preglowbtw, perhaps it would be clever to use a 32 bit delta when upsampling, the upper word is always 0 anyway
02:55:18jhMikeSthat's the least executed branch at that
02:55:27preglowsure, didn't say it mattered much :>
02:56:17jhMikeScould do that, in cf it's basically swapped into that at the start
02:56:24preglowthat's what made me think of it
02:56:34 Join cmptrgy412 [0] (
02:56:35preglowmight not be too wise anyway, hmm
02:57:28jhMikeSmmmm...I see no problem to just do it when calcing the delta.
02:57:46jhMikeSwould save a swap instruction :)
02:58:39 Quit SimonSelki ("CGI:IRC (EOF)")
02:59:11 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
02:59:16jhMikeSActually, I'd like to tweak downsample to only read only sample when the advance moves it by only one position
03:00:26preglowwould make for a slow c implementation, though, don't have carry there
03:00:52safetydanLlorean, sorry to mislead you. I was actually thinking the speex codec wasn't telling the dsp the correct frequency for the speex files thus making them get resampled incorrectly.
03:01:13Lloreansafetydan: Ah well, at least it led me to discover that it's a global problem, not specific to speex.
03:01:42Lloreansafetydan: It also led me to discover the fact that as soon as I discovered it was global, it decided to no longer reproduce, at all. No matter what I try.
03:02:00safetydanAs all good bugs are wont to do.
03:02:42LloreanIt was 100% reproduceable until I reproduced it with an MP3 instead of a .spx. After that one MP3 showed the same symptom neither the .spx files or that .mp3 will do it any more. Including after rebooting.
03:02:50LloreanEntirely because they all hate me.
03:03:40LloreanMeanwhile, it's about the most minor bug ever as far as I'm concerned. I can't hear the difference, and can only identify it if I play the same audio elsewhere and listen to them side by side, hardly a 'normal' situation.
03:04:03*preglow does the ignorable bug dance
03:04:42jhMikeSyeah, not in c
03:04:50LloreanYeah, I'm perfectly happy with filing it in with "Our resampler isn't perfect, you dolt (referring to me)" and leaving it be.
03:05:02jhMikeSjust need bit 1 of the whole part to be the carry bit
03:05:34preglowanywho, i'm falling asleep again
03:05:37jhMikeSexacty same scheme
03:05:51preglowprobably back tomorrow :>
03:07:45 Join funky [0] (n=repulse@unaffiliated/funky)
03:10:10***Saving seen data "./dancer.seen"
03:14:29safetydanright, so that just leave speex stereo decoding to fix
03:21:50 Join Criamos [0] (
03:23:43 Join lurKer [0] (i=4a87b2ea@gateway/web/cgi-irc/
03:25:19 Join webguest56 [0] (i=9a14f589@gateway/web/cgi-irc/
03:29:31 Join Thundercloud [0] (n=thunderc@
03:30:47 Quit webguest56 ("CGI:IRC (Ping timeout)")
03:30:53toffe82jhMikeS: the first schematic of the gigabeat is on the wiki :) ok, it is just the remote control
03:30:56 Quit funky ("leaving")
03:32:26 Part nls
03:55:10 Quit Thundercloud__ (Read error: 110 (Connection timed out))
03:56:52 Quit dan_a (Remote closed the connection)
03:57:18 Join dan_a [0] (n=dan_a@
03:57:34 Quit cmptrgy412 ("Trillian (")
03:57:44 Join cmptrgy412 [0] (
04:05:05 Quit Criamos ("( :: NoNameScript 4.03 :: )")
04:18:47 Quit Nimdae ("all computers going down until tomorrow night or so")
04:22:32 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
04:22:53 Join miepchen^schlaf [0] (
04:24:37 Join webguest23 [0] (i=42c07542@gateway/web/cgi-irc/
04:30:05 Quit webguest23 ("CGI:IRC (Ping timeout)")
04:31:00 Join cmptrgy4122 [0] (
04:32:57 Join webguest23 [0] (i=42c07542@gateway/web/cgi-irc/
04:36:03 Join Thundercloud_ [0] (n=thunderc@
04:36:09 Quit Thundercloud (Read error: 104 (Connection reset by peer))
04:38:25 Quit webguest23 ("CGI:IRC (Ping timeout)")
04:41:26 Join webguest23 [0] (i=42c07542@gateway/web/cgi-irc/
04:42:47 Quit cmptrgy412 (Read error: 110 (Connection timed out))
04:45:01 Nick [Tesser] is now known as [Tesser-Away] (n=Tesser@unaffiliated/tesser/x-000001)
04:46:26 Nick cmptrgy4122 is now known as cmptrgy412 (
04:46:31 Quit cmptrgy412 ("Trillian (")
04:46:42 Join cmptrgy412 [0] (
04:46:54 Quit webguest23 ("CGI:IRC (Ping timeout)")
04:48:46 Join webguest23 [0] (i=42c07542@gateway/web/cgi-irc/
04:53:10 Quit K3nto (Read error: 104 (Connection reset by peer))
04:54:14 Quit webguest23 ("CGI:IRC (Ping timeout)")
04:54:39 Quit Bagder (
04:54:39 Quit YouCeyE (
04:54:39 Quit smably (
04:54:39 Quit tchan (
04:54:39 Quit bezlemin (
04:54:39 Quit spug (
04:54:39 Quit Overand (
04:54:39 Quit NuclearDog (
04:55:01 Join Bagder [0] (n=daniel@rockbox/developer/bagder)
04:55:01NJoinYouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
04:55:01NJoinsmably [0] (
04:55:01NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
04:55:01NJoinbezlemin [0] (
04:55:01NJoinspug [0] (
04:55:01NJoinNuclearDog [0] (
04:55:01NJoinOverand [0] (i=overand@pdpc/supporter/active/Overand)
04:56:10 Join webguest96 [0] (i=c1713011@gateway/web/cgi-irc/
04:56:12 Quit webguest96 (Client Quit)
04:56:38 Join datachild` [0] (
04:59:18 Join fejfighter [0] (
05:01:09 Quit lurKer ("CGI:IRC (EOF)")
05:04:08 Quit cmptrgy412 ("Trillian (")
05:04:52 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
05:05:13 Join miepchen^schlaf [0] (
05:07:01 Quit smably (
05:07:01 Quit NuclearDog (
05:07:01 Quit YouCeyE (
05:07:01 Quit spug (
05:07:01 Quit Overand (
05:07:01 Quit bezlemin (
05:07:01 Quit tchan (
05:07:01 Quit Bagder (
05:07:40 Join RoC_MM [0] (
05:08:32NJoinBagder [0] (n=daniel@rockbox/developer/bagder)
05:08:32NJoinYouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
05:08:32NJoinsmably [0] (
05:08:32NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
05:08:32NJoinbezlemin [0] (
05:08:32NJoinspug [0] (
05:08:32NJoinNuclearDog [0] (
05:08:32NJoinOverand [0] (i=overand@pdpc/supporter/active/Overand)
05:09:02 Quit Obsys (Read error: 131 (Connection reset by peer))
05:09:53 Quit Thundercloud_ (Remote closed the connection)
05:10:00 Join eggy [0] (n=eggy@pdpc/supporter/sustaining/yorkcc.eggy)
05:10:15***Saving seen data "./dancer.seen"
05:14:29 Quit datachild (Read error: 110 (Connection timed out))
05:15:11 Join Obsys [0] (
05:15:55 Join i6zl [0] (i=4bb425fc@gateway/web/cgi-irc/
05:18:15 Quit i6zl (Client Quit)
05:37:16 Join L30 [0] (
05:39:38 Join zorton [0] (i=zorton@unaffiliated/zorton)
05:40:13zortonI see that recording is now enabled on the nano. Does it just need to be tested with a line in connection via the dock connector?
05:40:33L30Brasil mostra a sua cara!
05:42:51L30 '<
05:44:22 Part L30
05:50:45safetydanzorton, I think so but can't be certain as I don't know much about the iPod port
05:59:46 Join aliask [0] (
06:00:02 Quit atsea- (Remote closed the connection)
06:01:47 Join atsea- [0] (i=ariel@gateway/tor/x-6e8738d02ad51d9c)
06:09:45 Quit mirak (Remote closed the connection)
06:10:31 Join jhulst [0] (
06:14:54 Join RoC_MM_0w [0] (
06:22:07 Join RoC_MM_0x [0] (
06:27:14 Join Febs [0] (
06:30:23 Quit RoC_MM_0w (Read error: 145 (Connection timed out))
06:31:23 Quit RoC_MM_0x ("Leaving")
06:36:18 Quit RoC_MM (Read error: 110 (Connection timed out))
06:47:56 Part toffe82
06:48:52scorchezorton: yes
06:52:03 Join K3nto [0] (
06:54:05 Join perplexity [0] (n=joust@
07:01:32 Join aegray [0] (
07:01:44aegrayhas anyone worked with the i.mx31?
07:02:34aegray(or in terms of the i.mx31 relating to pp5020)
07:08:04 Quit printfXh4 (Read error: 113 (No route to host))
07:08:12 Join printfXh4 [0] (
07:08:54 Quit printfXh4 (Remote closed the connection)
07:09:33 Join printfXh4 [0] (
07:09:38 Quit printfXh4 (Remote closed the connection)
07:10:17***Saving seen data "./dancer.seen"
07:10:20 Join datachild [0] (
07:19:06 Quit datachild` (Read error: 110 (Connection timed out))
07:21:36perplexityYay.. H300 Bootloader USB mode works flawlessly now. Ta LinusN :)
07:22:57 Join printfXh4 [0] (
07:26:22 Quit siracs (Read error: 110 (Connection timed out))
07:37:13 Quit Febs (Read error: 110 (Connection timed out))
07:45:31 Quit K3nto ("Leaving.")
07:47:38 Quit atsea- (Read error: 104 (Connection reset by peer))
07:48:55 Join atsea- [0] (i=ariel@gateway/tor/x-5c769743118857f9)
07:51:43 Quit Daishi (Read error: 110 (Connection timed out))
07:52:16 Quit fejfighter ()
07:54:06 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
07:54:10 Join miepchen^schlaf [0] (
07:54:19 Quit jhulst (Read error: 110 (Connection timed out))
08:06:03 Join K3nto [0] (
08:06:15K3ntoi cant believe im about to put doom on my ipod
08:06:52K3ntomy dad was telling his IT guy about my recent interest in linux, and now i could have a summer job lined up. in an IT division!
08:07:14aliaskgotta love parent for hooking jobs up
08:07:37daurnimatorhey aliask
08:07:44aliasknooot much
08:07:47K3ntoheh ya
08:08:04juxtapK3nto, you played super mario?
08:08:25K3ntonot yet
08:08:30K3ntois it decent?
08:08:42juxtapneither. just heard about that yesterday.
08:08:49juxtapi've been slack with updating my rockbox
08:09:19*daurnimator waits for jdgordon & accompanying iriver
08:09:29daurnimatoraliask: what you doing for the long weekend
08:09:35aliaskwhat long weekend :(
08:09:42K3ntodid the creator(s) of linux antocipate its capabilities/addictiveness/usefulness?
08:10:30juxtaplinux was created for fun
08:10:35juxtapwasn't intended to be huge
08:10:40K3ntoi cant leave my computer
08:10:46K3ntonot since i installed ubuntu
08:10:57juxtapwell you're obviously not using bash shell :p
08:10:59daurnimatoraliask: you working monday?
08:11:35K3ntoi just imported my entire library over from my ntfs partition
08:11:36daurnimatorhaha. sucks to be you ;)
08:11:44K3ntonow i have no reason to go back to windows
08:12:33daurnimatorK3nto: why
08:12:39daurnimatorK3nto: linux supports ntfs fine
08:12:55daurnimatorinfact, i would use ntfs in linux over fat32
08:13:20zei thought write support was experimental and/or non-standard or something
08:14:34daurnimatorbeen stable for a good 9 months now
08:15:03K3ntodaurnimator: i thought there were plenty of ntfs issues with ubuntu
08:15:06K3ntobut nvm that now
08:15:15K3ntomy ubuntu universe is almost complete
08:15:22K3ntojust have to tweak my ipod a lil more
08:15:27daurnimatorwhat about multi-verse?
08:15:37K3ntoi think the HD is corrupt actually. anybody know of anything to fix that?
08:15:51daurnimatori pretty much left linux because there was no good audio player
08:16:03K3ntoAmarok dude
08:16:12K3ntoits really good. i think anyway
08:16:15daurnimatoramarok is a piece of steaming shit
08:16:34K3ntofor real?? why didnt you like it? so simple, so pretty
08:16:38*scorche coughs
08:16:56daurnimatorits so ugly, slow & hard to use
08:17:08daurnimatorscorche: got a whiff of amarok? :0
08:17:12K3ntomaybe its ur system? mine is fast like a abbit
08:17:25scorcheno...i got a whiff of an off-topic discussion
08:17:32 Join austriancoder [0] (n=austrian@
08:17:44K3ntoo sry
08:17:52 Join Rob2222 [0] (
08:17:53aliaskaustriancoder: Congrats on the great result of yesterday's meeting
08:18:36 Quit printfXh4 (Read error: 113 (No route to host))
08:21:09 Part K3nto
08:24:06 Join K3nto [0] (
08:25:32 Quit Rob222241 (Read error: 145 (Connection timed out))
08:29:09austriancoderaliask: thanks...
08:29:10K3ntoanybody here use doom? im having problems i thin
08:29:30*austriancoder plays with the developer board
08:44:49 Quit K3nto ("Leaving.")
08:45:49 Join kubiix [0] (
08:49:26 Join K3nto [0] (
08:51:21K3ntoif i could fix my ipod that would be awesome. then i wouldnt have to restart my pc everytime i wanted to gain access to it
09:04:37 Join Ribs [0] (n=ribs@
09:06:42 Join Babbelaar [0] (n=Jason@
09:06:58BabbelaarHello yall?
09:08:00BabbelaarUm.. I d/l the new h300 build a while back( i didn't d/l a build since like last year) and when i try to play songs it says "incompatiable"
09:08:17Babbelaarand when i look at version it sez "incompatiabl
09:08:31aliaskThat'll be because you only updated the rockbox.iriver file
09:08:38aliaskYou need to replace the whole .rockbox directory too.
09:08:48Babbelaarhow do i do that?
09:09:10BabbelaarCause i just read the manual about "updating" and it said d/l the new build and just hook it up
09:09:13aliaskDownload the new build zip, and extract the whole thing to the root of the harddrive
09:09:29aliaskMake sure you replace all the files it asks about
09:09:39BabbelaarIn Mac it doesn't ask about anything
09:09:46Babbelaarexcept rockbox.iriver
09:09:58*aliask sighs
09:10:12aliaskIs MacOS making your life easier? :P
09:10:19***Saving seen data "./dancer.seen"
09:10:24Babbelaarin this situation.. i'd have to say.. no :(
09:10:48aliaskI believe Macs hide folders starting with a .
09:11:07aliaskSo the .rockbox folder will be hidden - that might have something to do with it.
09:11:16BabbelaarBut say i extract the file there
09:11:19BabbelaarThe zip
09:11:27BabbelaarIm guessing all the other files will go
09:11:36aliaskWhich other files?
09:11:54Babbelaarno i mean that .rockbox file i can't see it on my mac
09:11:57Babbelaarbut like on my windows
09:12:29BabbelaarIf i extract the zip file in the Iriver, im guessing it will replace the .iriver folder
09:13:23aliaskIt should do it in macos too, but it looks like it's not. WinZip will definitely ask you if you want to overwrite the other files (which you do want)
09:13:42BabbelaarNa it's just harder to do it on windows
09:13:50BabbelaarCause my USB Data Port is bent..
09:14:10Babbelaarwhich makes me have to put pressure on the h300 when i want to d/l anything
09:14:12aliaskAh ok. There might be a way to do it on macs, but I don't have the first clue about them really.
09:14:20Babbelaarthanks though
09:14:27aliaskYou're welcome.
09:14:59 Join Presence [0] (
09:17:43amiconnSomething's wrong with the tree contexts
09:18:24amiconnIf you enter the file browser, move the cursor to e.g. the 5th item, then back out and enter the db browser, guess at what item you'll end up?
09:18:43amiconn5th item.
09:19:08LloreanI read file browser twice the first time, so I was thinking "It should be the fifth but apparently is not" rather than "It should be the first but apparently is not"
09:19:09amiconnBut the reverse is not true: the db cursor position doesn't "transfer" to the file browser
09:19:41safetydanBabbelaar, are you familiar with the terminal? You could always use unzip on the command line. At least I think OSX has unzip
09:19:46LloreanOkay, when a playlist ends, where exactly are you supposed to end up?
09:21:03BabbelaarI expanded it
09:21:07Babbelaarand it showed everything
09:21:18Babbelaarbut it put the thing into a folder on iriver
09:22:13Babbelaarlemme try one more th ing
09:22:35amiconnLlorean: If 'follow playlist' is disabled, you should end up wherever you were last. If you directly resumed after boot, hence didn't enter the file browser before, you should be in the root, first item
09:23:01amiconnIf 'follow playlist' is enabled, you should end up on the last file played
09:23:33K3ntookay, i love rockbox. im playing a 9-minute song and im gonna see if it plays it through without interruption
09:23:35amiconnAll this is only true for the file browser; the db browser can't do 'follow playlist'
09:23:39Lloreanamiconn: Okay, just making sure. All these people are complaining about playlists ending and finding themselves in the root menu, and my playlist just ended and I was where I launched it from, and I was just a bit uncertain how the behaviour worked from other people's statements. Thanks.
09:24:07 Quit Babbelaar ("Babbel v070113.19 Mac OS X Edition (Intel) -")
09:24:29 Part K3nto
09:24:53amiconnLlorean: I'm not sure how this is supposed to work wrt root menu. I was only talking about browser behaviour
09:26:58Lloreanamiconn: I think the problem is that if you launch a playlist from filetree, go to the menus to change a setting, then tap Play to return to the WPS, when the playlist ends you're in the menu, and that wasn't the previous behaviour so users find it unexpected.
09:27:16LloreanIt's an extension of how it worked before, making it more consistent, but users don't see it as that.
09:29:26 Join K3nto [0] (
09:29:57K3ntowoot! 8 minutes of music and no interruptions. My faith in rockbox is 100%
09:30:16aliaskRecent iPL convert?
09:30:38aliaskiPod linux
09:30:48aliaskHeh :0
09:30:51K3ntoand im neve going back!
09:31:02aliaskIt's pretty, but not very useful...
09:31:31K3ntowhat, ipod linux?
09:31:41aliaskYes, or at least last time I used it.
09:32:05K3ntoput it this way. apple's=broken. iPL=perfect
09:32:11K3ntoits very functional
09:32:28LloreanI would say iPL is very far from perfect.
09:32:43LloreanHow's the battery life?
09:32:56aliaskLlorean: How's our battery life...
09:33:03Lloreanaliask: I never claimed we were perfect either.
09:33:08K3nto50% = to 5 hours
09:33:11LloreanK3nto: You have a strange definition of good.
09:33:21LloreanIt's significantly less than the battery life in the Apple firmware.
09:33:21K3ntoand my ipod is about a year or more old
09:33:37LloreaniPL also doesn't have the best audio functionality in the world.
09:33:39K3ntorealistically though. when are you going to listen to your ipod for 10 hours of the day?
09:33:49Shaidsome people do
09:33:55K3ntoi dont lol
09:34:02LloreanSome people also need to use it across several days without the opportunity to charge.
09:34:04K3ntoso 10 is more than decent
09:34:40LloreanThank you for dismissing aspects of my lifestyle with but a single word.
09:35:09LloreanMy point is simply this: iPL is not perfect, nor will it be until there are no bugs and it surpasses _every_ aspect of both Rockbox and the original Apple firmware, and nobody can raise a complaint about it.
09:35:21LloreanThat's, to a lesser extent, the definition of perfect: Without flaw.
09:35:46scorcheit wont have a flaw in it when it can make me a latte
09:36:11K3ntoi see a fundamental flaw
09:36:11Lloreanscorche: I mean, simply enough, if iPL were perfect, why's he using Rockbox at all?
09:36:21scorchewith a shot of hazelnut and caramel
09:36:34K3ntoLlorean: what?
09:36:48LloreaniPL is not Rockbox.
09:36:53scorchenot even close..
09:36:53LloreaniPL is iPod Linux...
09:37:02K3ntowell now
09:37:19LloreanI mean, it'd be somewhat silly to have two different names for the same thing.
09:37:20K3ntoi like ROCKBOX. dont know about iPL
09:37:59*Llorean votes for a name change to something Recursive that end's with "isn't Linux"
09:38:50K3ntoits like an identification
09:39:17LloreanIt was a joke, based on the GNU name.
09:39:37HaSHwait rockbox isnt linux based?.
09:39:46LloreanNo, it's not.
09:40:07LloreanYes, really.
09:40:09scorchewe have our own faster, smaller kernel
09:40:24HaSHso not the linux kernel
09:40:29K3ntoit takes nothing from linux? or is it just not BASED
09:40:31K3ntoon linux
09:40:56 Join linuxstb [0] (i=d529f3ae@rockbox/developer/linuxstb)
09:40:59scorcheit was written from scratch
09:41:10LloreanThere are bits of code here and there, especially for the iPods, that were derived from iPodLinux
09:41:16LloreanBut the kernel itself was built from scratch.
09:41:20K3ntoi see
09:41:27scorcheLlorean: yeah...i was referring to the kernel
09:42:10LloreanK3nto: But usually when people say 'based on linux' they refer to it actually being a modified linux kernel.
09:42:15LloreanRockbox is very, very, very far from being that.
09:42:29HaSHand its not just a stripped down linux kernel?...its actualy your own(rockbox) kernel?
09:42:34LloreanHaSH: Yes.
09:42:45LloreanHaSH: That's kinda what "From scratch" means.
09:42:50scorchethat is what "from scratch" means ;)
09:42:55scorchebah you!
09:42:59HaSHLlorean, lol sorry its late....and im baked :-)
09:42:59scorchei am slow tonight
09:43:14K3ntoLlorean: but its still open sourced right
09:43:17scorcheLlorean: we do think alike though =/
09:43:23scorcheof course it is
09:43:23HaSHbut right on...thats cool. i always thought it was linux
09:43:25LloreanK3nto: Yes, it's released under the GPL license.
09:43:33LloreanI don't see how that's really related to being Linux or not.
09:44:12K3ntoHaSH: lol yea 7 coors and my keyboard dexterity is maintained lol
09:44:53 Part K3nto
09:45:03*linuxstb points people to
09:46:05HaSHah well its almost 4am im off to bed. cya
09:47:28 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
09:47:48 Join miepchen^schlaf [0] (
09:49:11amiconnlinuxstb: Didn't you say rolo'ing diskmode should work now?
09:49:20amiconnIt still doesn't on mini2g
09:50:09linuxstbRolo functionality seems to come and go...
09:50:42 Quit perplexity (Remote closed the connection)
09:50:44linuxstbBut I think it was working on both my Color and Video last time I checked.
09:50:55linuxstb(I can't check now, no DAPs with me)
09:53:08amiconnI only get the 'folder with exclamation mark' screen, followed by a reboot after ~10 seconds
09:53:24amiconnShutdown, rather
09:54:13amiconnHaha, and on second attempt (after rolo'ing rockbox), rolo hangs at "Waiting for coprocessor..."
09:54:54linuxstbAh yes, rolo'ing Rockbox is currently broken - if you check the threads after rolo'ing it, nothing is running on the COP...
09:55:50linuxstbI think dan_a has spent a lot of time investigating, but to no avail.
09:56:29amiconnHmm. When rolo'ing diskmode or diagmode (without rolo'ing rockbox before), Rolo briefly shows "Waiting for coprocessor..", however, when rolo'ing rockbox it does not.
09:57:29linuxstbThat sounds odd.
09:57:44amiconnHmm, now it did.....
09:59:21amiconnHmm, the thread debug screen now shows some odd thread behaviour...
09:59:49amiconnWith the codec thread running on the cop, the voice codec thread is constantly running (!)
10:01:06 Quit toer (Read error: 131 (Connection reset by peer))
10:01:29amiconnThis is with voice file present (of course), but voice disabled...
10:01:39 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:03:36 Join petur [0] (n=petur@rockbox/developer/petur)
10:04:04amiconnlinuxstb: Oh, and I think this also identifies our lockstepping threads!
10:04:29amiconnVoice and main code are running on different cores, but they're interlocked because of the buffer swaps!
10:04:46LloreanWould fixing it be as simple as dropping it on Core 1 as well?
10:05:07 Join perplexity [0] (n=joust@
10:05:17 Join toer [0] (
10:09:37amiconnHmm, these are 2 different problems. 'Voice codec' constantly running applies to all swcodec targets.
10:10:52*austriancoder has problems to include the new as3515 audio driver in the build system...
10:11:33 Quit linuxstb ("CGI:IRC (Ping timeout)")
10:14:58 Join linuxstb [0] (i=d529f3ae@rockbox/developer/linuxstb)
10:15:53*amiconn started a battery bench on his X5
10:15:56linuxstbaustriancoder: What's the problem? Have you added any new files to firmware/SOURCES ?
10:27:41 Quit Siku ("reboot")
10:35:10 Part Kitt0s
10:38:15 Join adam1302 [0] (i=cbab4195@gateway/web/cgi-irc/
10:39:26 Quit linuxstb ("CGI:IRC (Ping timeout)")
10:40:47adam1302hey how did it go with AMS
10:43:17austriancodervery good.. we got the as3514 datasheet and a as3514 testing board
10:43:19Bagderaustriancoder: 3515?
10:43:34 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
10:43:50JdGordonhey all
10:43:51austriancoder3514 ans 3515 are equal of the registers
10:43:52JdGordonhowd the meeting with ams go?
10:44:02Bagderaustriancoder: so why make a 3515 driver?
10:44:25Bagderand btw, are you allowed to show us the data sheet?
10:44:52austriancoderBagder... in my source tree it is called as3514...
10:47:45Bagderthat only makes calling it a 3515 driver even more weird in my eyes...
10:48:03 Join hannesd__ [0] (
10:49:01austriancoderBagder: was only a typo... sorry
10:51:31 Join iloric [0] (
10:52:00Bagderare you allowed to show the data sheet?
10:53:31 Quit adam1302 ("CGI:IRC (EOF)")
10:53:32austriancoderBagder: it will be not aviable for everybody..
10:53:52 Quit austriancoder (Remote closed the connection)
10:54:05Bagderthat's not what I asked though...
10:54:21 Join austriancoder [0] (n=austrian@
10:54:50austriancoderBagder: do want it?
10:54:55Bagderyes please
10:56:34 Join ender` [0] (
10:56:40 Join aegray_ [0] (
10:56:58dan_a_austriancoder: Have you fixed your build system problem yet?
10:57:11austriancoderBagder: will recheck it with rp- and our contact person.. but i think i can mail it to your later (you are not allowed to share it or show it others)
10:57:31austriancoderdan_a: yep... compiles...
10:57:32Bagderaustriancoder: it would make sense to show it to the core sansa hackers
10:57:41*austriancoder is back in an hour
10:57:42Bagderthat includes dan_a and barry too
10:57:53dan_a_I already got one off rp-
10:58:11dan_a_He thought that Barry, Bagder and myself should get them
10:58:16austriancoderokay.. than Bagder you will get one too
10:59:03austriancoderand they told us that we should not too much with the powermanager.. it can break the device very easily
11:00:50austriancoderBagder: check your mail
11:01:26*austriancoder hopes that he will commit a working dirver for playback and recording in some hours
11:02:11BagderI bet that
11:02:20 Join webguest06 [0] (i=543834e8@gateway/web/cgi-irc/
11:02:22Bagderwill boost the amount of private support mails I get ... :-)
11:02:29dan_a_austriancoder: Did you get sound yet?
11:03:09austriancoderdan_a: a klick like the headphone output was enabled
11:03:22austriancodernothing more,, at the moment
11:03:50austriancoderohh and will later make some pohtos of the devel board
11:03:54austriancoderso.. time to go
11:04:05austriancoderBagder: mail should be send
11:04:15 Quit hannesd (Read error: 110 (Connection timed out))
11:07:28 Join Siku [0] (
11:07:39 Join Jeton [0] (n=chatzill@
11:08:35 Quit aegray (Read error: 110 (Connection timed out))
11:08:56 Join tHEkIND [0] (
11:09:40 Quit webguest06 ("CGI:IRC (EOF)")
11:10:01tHEkINDI'd need to store a value
11:10:18tHEkINDHow can i know where i can store it on the memory ?
11:10:24***Saving seen data "./dancer.seen"
11:10:26 Join linuxstb [0] (i=d529f3ae@rockbox/developer/linuxstb)
11:11:56JdGordonusing an infra-red thermometer thingy to see which bits in the RAM chips are changed?
11:12:10tHEkINDThanks man
11:12:17LloreantHEkIND: Well you haven't asked a very specific question.
11:12:21LloreanMost people store values in variables.
11:12:28dan_a_tHEkIND: Are you using C or ASM?
11:12:30tHEkINDOkay okay
11:12:46tHEkINDI'm not very friendly with C
11:13:05tHEkINDBut i'll learn
11:13:45aliaskIf it's just a small value I think the best option would be declaring a variable, and setting it. That way you dont need to worry about memory etc
11:14:19tHEkINDokay i'll do that
11:14:20dan_a_Do you actually need to know where it is in memory? If you do, then you need to learn about pointers. Otherwise, as aliask says, declare it (like "int x;")
11:14:31tHEkINDNo i don't need pointer
11:16:03tHEkINDOkay i found it
11:20:13 Join funky [0] (
11:25:30 Join obo [0] (
11:25:35 Join Rob222241 [0] (
11:29:31 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
11:29:51 Join miepchen^schlaf [0] (
11:30:31 Join _pilled [0] (
11:35:42 Quit pilled (Read error: 60 (Operation timed out))
11:35:56 Quit Jeton ("Chatzilla 0.9.77 [Firefox]")
11:38:36 Join nls [0] (
11:42:45 Join decayedcell [0] (i=3ba79d47@gateway/web/cgi-irc/
11:42:47 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
11:42:57 Quit decayedcell (Client Quit)
11:43:00 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
11:43:10 Quit Rob2222 (Read error: 113 (No route to host))
11:43:25 Join decayedcell [0] (i=3ba79d47@gateway/web/cgi-irc/
11:44:03nlspixelma: I read your question in the logs, what do you have in mind?
11:44:03decayedcellthat single stage audio initialisation added 20 minutes~ of battery life to my iPod
11:48:04 Join Id2ndR [0] (
11:48:16w1ll14mmorning all
11:49:20pixelmanls: for example some settings for a default codepage on the player... some just aren't possible. I could exclude them in the lang file but IIUC amiconn said that also needs to be taken out somewhere else in the UI code
11:50:54 Join barrywardell [0] (
11:51:50 Join BHSPitLappy [0] (
11:52:32 Join BHSPitMonkey [0] (
11:52:43nlspixelma: if they aren'y used at all just removing them from the lang file for the player should be fine
11:55:27 Join voltagex [0] (
11:55:54voltagexanything interesting happening in rockbox land?
11:56:22markunvoltagex: did you see the new menu?
11:56:41dan_a_voltagex: Sound on the Sansa is hopefully on the way, we've got the iAudio M5 working
11:56:42pixelmanls: well... I've no idea. My suggestion is that I'll look into the lang file and post an update of it only with a lot comments. So that someone else could look up if it needs more changes - would that be ok?
11:56:43tHEkINDThe new menu look nice
11:56:48voltagexmarkun: yeah, now we need pretty colours as default
11:56:50preglowdan_a_: any progress?
11:57:24voltagexso at the moment, what's the best mp3 player to buy for running rockbox?
11:57:52nlspixelma: yes, I could do the other changes.
11:57:54dan_a_preglow: Clicks, but nothing recognisable as music yet. I don't know, though, if that's because the badly written flash or LCD drivers are slowing everything to a crawl (neither of them do any yielding yet)
11:58:41nlsI think I know what he meant that other code needs changing, the settings code isn't ifdefed very much for different targets, so that will be neccesary
11:59:30markunvoltagex: it depends on your needs
11:59:53pixelmanls: thanks :)
11:59:58markunvoltagex: if you don't care for radio and recording buy a Gigabeat F40
12:01:54markunif you care for those features but not for a nice screen, buy a iriver H1xx maybe
12:02:07Lloreanvoltagex: Well, Markun pretty much hit it on the head.
12:02:23markunLlorean: maybe i'm on his ignore list :)
12:02:37linuxstbvoltagex: What are your top three requirements for an mp3 player?
12:02:51Lloreanvoltagex: The H100 series runs really well with Rockbox, and has the most in the way of audio-specific features. The Gigabeat doesn't offer recording or radio, but has a big color screen and a ridiculously fast processor that is likely to be good for video.
12:03:55tHEkINDh10 is nice (no fm on rockbox now), but screen is little (128x128)
12:04:09LloreantHEkIND: It hardly qualifies as one of the best ones though.
12:04:33tHEkINDI like my h10 ^^
12:04:35LloreanAll of the PortalPlayer based targets are basically disqualified from the 'best' list because of battery and USB concerns, followed closely by performance limitations.
12:04:36preglowi'd say anything with portalplayer chips in it automatically disqualifies as the best choice
12:04:54tHEkINDI never tested rockbox on non-portalplayer devices
12:05:03LloreantHEkIND: It tends to work much better.
12:05:07preglowwell, it's usually a better experience
12:05:35tHEkINDSo my next player will be a non-portalplayer ^^
12:05:53preglowOR you could annoy portalplayer until they give us documentation on their chips
12:06:03preglowwe'd make a statue out of you
12:06:20tHEkINDThey never accepted ?
12:06:38LloreanPortalPlayer wouldn't even dignify me with a response to say "We won't help you"
12:06:41 Quit Ribs ("Client Exiting")
12:07:22tHEkINDMaybe they'll understand that rockbox is great for the sells ^^
12:07:40tHEkINDBut rockbox would have to be more famous
12:07:44voltagexI like radio, but I like video and rockboy better
12:08:06tHEkINDSomeone will implement re-buffer on mpegplayer soon ?
12:08:29LloreantHEkIND: Impossible to predict.
12:08:33voltagexNVIDIA® Acquires PortalPlayer®
12:08:33voltagexOn January 5th 2007, NVIDIA® Corporation completed its acquisition of PortalPlayer, Inc. a leading supplier of semiconductors, firmware, and software for personal media players (PMPs) and secondary display-enabled computers.
12:08:34markuntHEkIND: someone from #gigabeat wanted to work on it, but I haven't seen him for weeks
12:08:36voltagexthere's your problem
12:08:36Lloreanvoltagex: Then the Gigabeat is a good player for you.
12:08:42dan_a_I need to see if we can get any PP information out of nvidia
12:08:44 Join Ribs [0] (n=ribs@
12:08:46Lloreanvoltagex: This predates the NVidia acquisition.
12:08:46preglowvoltagex: we tried before that happened
12:09:00tHEkINDThat's bad cause mpegplayer is really a good thing
12:09:00LloreanNVidia just seems likely to have made it less likely, though.
12:09:12preglowLlorean: you seem to have whatever i plan to say today covered, so i'll just go have a shower ;)
12:09:12LloreantHEkIND: The primary focus of Rockbox is audio functionality.
12:09:13voltagexLlorean: well Nvidia support open source as far as they have Lunix card drivers :D
12:09:25Lloreanpreglow: It's spooky.
12:09:32Lloreanvoltagex: Closed source ones.
12:09:36aliaskvoltagex: But they're just binaries, not os
12:09:42markuntHEkIND: yes, I would love to be able to watch some videos on a long busride 2 months away..
12:09:51tHEkINDLlorean -> audio works very well already ! And video is most likely the functionnality that a lots of original firmwares don't have
12:10:00voltagexI don't hold high hopes
12:10:00voltagexso what docs do you need?
12:10:03LloreantHEkIND: Audio still has a lot of major bugs.
12:10:08voltagexmight try them on the phone tomorrow
12:10:12LloreantHEkIND: Especially in the voice aspect.
12:10:20tHEkINDYeah sometimes the mp3 decoders bugs
12:10:35Lloreanvoltagex: Datasheets for the PP5020 series, specifically 21 and 24 I believe, as well as the PP5002 would be nice as well.
12:11:01LloreantHEkIND: Besides, Rockbox's project focus is on audio functionality. Anyone is welcome to finish MPEGplayer, but it's not being focused on right now.
12:11:45decayedcellWhats the difference between the PP5024 and the PP5022?
12:11:46voltagexcan someone email me the questions to ask? I'll try them on the phone tomorrow
12:12:02dan_a_voltagex: Do they work on Sundays?
12:12:24tHEkINDif(lastbtn & (BUTTON_SCROLL_UP | BUTTON_SCROLL_DOWN)) -> is this meaning if lastbtn == BUTTON_SCROLL_UP || lastbtn == BUTTON_SCROLL_DOWN ?????
12:12:27Lloreandecayedcell: The PP5024 is designed for flash based players, and has some unique aspects for that. As well, onboard DAC I believe.
12:12:41Lloreanvoltagex: Yeah, might want to wait 'till Monday.
12:12:57*linuxstb may be motivated to work on mpegplayer again once his gigabeat arrives
12:13:08Lloreanlinuxstb: You've ordered one too?
12:13:11linuxstbtHEkIND: No.
12:13:13barrywardelltHEkIND: Yes
12:13:26barrywardelltHEkIND: well, kind of
12:13:40linuxstbThe two expressions aren't equivalent.
12:13:47barrywardelltHEkIND: linuxstb was right
12:13:50barrywardelli'm half asleep
12:14:00decayedcellvoltagex according to the iPL wiki, PP5002, PP5020, PP5021C-TDF, PP5022 are the iPod ones
12:14:36aliasktHEkIND: But in your head it's the same thing - it's just they way the reading function works
12:14:51tHEkINDI only know operators like == / > / < / <= / >= / !=
12:15:06Lloreandecayedcell: Yes, but Rockbox isn't only interested in the iPod ones.
12:15:18linuxstbLlorean: Yes, I won an F20 auction last week, so should be arriving in the next few days.
12:15:28nlsPP5024 is sansa, right?
12:15:32barrywardellsomeone who owns each pp target could check the hardware info screen to see exactly which pp chips they have...
12:15:43Lloreannls: Yes.
12:15:45decayedcellwhich ones are in the H1xx Llorean?
12:15:58Lloreandecayedcell: The H1xx is Coldfire. The H10 though isn't.
12:16:03aliasktHEkIND: Might help to read up about boolean logic
12:16:14Lloreandecayedcell: The H10 is the same as the iPod Mini I believe.
12:16:15voltagexhmm make that monday :P
12:16:16linuxstbtHEkIND: And bitwise operators
12:16:21decayedcellLlorean ah okay
12:16:32Lloreandecayedcell: MiniG2 I think
12:16:35linuxstbThe H10 has a PP5020 - same as 1st gen mini
12:16:38decayedcellLlorean yeah PP5020
12:16:42LloreanAh, I was wrong!
12:16:45 Join Lear [0] (i=chatzill@rockbox/developer/lear)
12:16:51linuxstb2nd gen mini has the PP5022 (the only ipod with it)
12:16:58voltagexholy shit I just lagged for 40+ seconds
12:16:59LloreanI should've remembered it was the 5020, since it suffered the 5020 problem.
12:17:04barrywardellh10 is PP5020D
12:17:17LloreanHas our 5020 problem let up? I haven't heard as many complaints about it.
12:17:31LloreanOr have they all just moved to a noscaling build finally.
12:17:35barrywardellthe cop changes seem to have done a lot to fix it
12:17:49linuxstbI think it's a lot better with the COP builds, but I've still experienced a couple of freezes since the commit.
12:17:58decayedcellso whats the difference between the PP5022 and PP5021C-TDF?
12:18:19tHEkINDOkay i understood the & meaning
12:18:29 Join desowin [0] (
12:18:57linuxstbdecayedcell: There's a public "product brief" summarising the features of the PP5022, but there isn't one the PP5021, so we don't know the differences.
12:19:29*aliask really wants to win this x60 auction
12:19:55barrywardellwhat's the difference between the PP5020 and the PP5022?
12:20:11preglowthey've probably promised apple not to make it public anyway
12:20:13preglowso good luck on this
12:20:17linuxstbThe PP5021/5022 both have 128KB of IRAM, compared to 96KB for the PP5020. I'm not sure of other differences.
12:20:27voltagexso, I ask for a shitload of datasheets :P
12:20:36linuxstbBut there must be some, as we experience PP5020-only problems...
12:20:50voltagexhmm, I'll think about it, I'm normally fairly good at talking
12:21:04voltagexjust got to work my way up the NVidia chain.
12:21:19decayedcellthe PP5021 lakes firewire
12:21:55amiconnpixelma, nls: The codepage problem on the player is that all possible settings are currently available for the player, but a number of them simply doesn't make sense. Setting them causes the unicode layer to correctly translate them, but there are no glyphs for these languages available
12:22:09amiconnThis applies, to chinese, korean, thai etc
12:22:37nlsamiconn: aha, so the setting is visible, never thought of that :-)
12:22:54decayedcellaccording to here: the PP5021 is just a PP5022 shrunken into TDF package =/
12:22:55nlsso it should be removed from the player builds?
12:23:02amiconnSo they should be taken out of the setting *and* the lang file
12:23:30amiconnBut the setting should not be removed completely; some values do make or will soon make sense
12:24:17nlsamiconn: as I don't understand what you mean, I'll leave it to you :-)
12:24:23linuxstbdecayedcell: What makes you think the PP5022 has firewire?
12:25:15amiconnnls: Latin1 and utf-8 make sense right now, because both can be used to represent latin text, which is what the charcell lcd is capable of
12:25:41nlsah, yes of course, I'm a bit hungover today...
12:25:46amiconnWith a bit of trickery, it will be possible to support latin2 and hopefully cyrillic and greek
12:26:09*preglow wonders if we have any greek player users
12:26:29nlsok, leave latin1, latin2, utf8 cryllic and greek, remove the rest
12:26:33amiconnpreglow: I don't know about any, but I do know that there are russian player users
12:26:45decayedcelllinuxstb: it says so on the page I linked above. Also, I do believe the Mini 2G can sync with firewire?
12:26:51amiconn...which already asked for basic cyrillic support
12:26:53linuxstbIsn't XavierGr Greek?
12:27:11nlsbut does he have a player? :-P
12:27:11pixelmalinuxstb: does he have a Player? ;)
12:27:12Lloreanlinuxstb: This is a Archos Player 'player'
12:27:25barrywardellPP5022 supports up to 100MHz, where PP5020 only supports 80MHz
12:27:28amiconndecayedcell: Yes. Mini 2G supports firewire
12:28:02 Join entheh [0] (
12:28:02LearCan you create a (long) FAT filename with a ':' (colon) in it on Linux? I've heard the FAT implementation is a bit relaxed, but is it that relaxed? :)
12:28:24amiconnpreglow: What I want to do is to bring the player lcd driver more in line with the bitmap targets, move relevant parts to target tree etc
12:28:38amiconnI also want to get rid of the extra conversion layer
12:29:00LearIt seems like a safe separator to use in the bookmark files, but...
12:29:06amiconnCurrently the driver does utf8->latin1->native lcd encoding (old or new detected at runtime)
12:29:37amiconnIt should do utf8->native lcd encoding
12:29:42nlsLear: you can create such filenames in rockbox tho :-)
12:29:46*preglow goes to finish asm resampling
12:29:48linuxstbdecayedcell: The 5022 product brief doesn't mention firewire (but the 5020 does).
12:30:10amiconnlinuxstb: Firewire definitely does work on pp5022
12:30:13amiconnI tried it
12:30:47decayedcellIt wouldn't make sense to exclude it from the product brief though
12:30:49Learnls: Ugh, that's not allowed, according to the FAT spec I've read...
12:30:50linuxstbamiconn: Yes, I'm just saying it's external to the pp5022
12:31:25nlsLear: correct, but we are relaxed :-) (there's a patch for restricting that in the tracker)
12:31:49 Quit linuxstb ("CGI:IRC")
12:32:00amiconnLear: The colon should be safe as a path separator, as long as we don't have a pc tool for bookmarks which might include a drive letter in the path
12:34:18LearWell, using DOS-style paths (with backslashes and drive letter) in the bookmark files isn't supported anyway, so it doesn't seem like something to worry about...
12:34:29pixelmanls: I think the M5 isn't taken into account in the patch yet - or did I miss it?
12:35:04 Join inversions [0] (
12:35:05nlspixelma: what do you mean?
12:35:18nlslike for using as a target in the lang file?
12:35:23 Join anathema [0] (
12:35:32Bagderaustriancoder: did you really send it? I haven't got anything yet...
12:36:32voltagexI'll say it again coz I only got a few model numbers... if I ring NVidia on monday, what exactly do I ask for?
12:36:53voltagexas far as portalplayer goes
12:37:26Lloreanvoltagex: Any of the iPod ones, or the PP5024 would be of interest.
12:37:42LloreanBut the most important thing would just be trying to get them to open up a line of communication with our developers at all
12:38:03pixelmanls: in the lang file there are sometimes target specific strings like "press x to do y" - and there I only see "x5: "
12:38:13LearHm, the shortname is checked for bad chars properly, but the long filename isn't checked at all, it seems.
12:38:17voltagexerr can I have a rockbox email address to give me some credibility? :D
12:38:51LloreanI don't believe we have Rockbox email addresses.
12:39:34nlspixelma: yes, correct, the m5 port came after that, but using m5 in such a place should work, it's just not done yet
12:39:46pixelmanls: maybe for the M5 it is already covered with it - it just has a misleading name currently (?)
12:40:09voltagexBadger Badger Badger Mushroom Mushroom! thanks.
12:40:15Bagdervoltagex: but I doubt rockbox gives any bonus points when talking to these guys
12:40:15LloreanBagder: Nifty
12:40:28LloreanBagder: Someone once brought up the idea of having a address someone could monitor to help those writing articles communicate with us, did you ever see that or have any interest in it?
12:40:35voltagextrue, but it's better than
12:40:50voltagexcan I have ?
12:40:55voltagexoh wait, there's the address
12:41:05BagderLlorean: I could easily fix an alias for it
12:41:08voltagexpage finally loaded
12:41:52LloreanBagder: I'd be more than happy to accept email forwarded from such an address, so that hopefully we can have a bit better relationship with future articles.
12:42:45aliaskHow would I go about getting one forwarded to my gmail account?
12:42:51 Join aegray [0] (
12:43:27voltagexconsidered doing a show for binrev radio? That'd garner the interest of a few more potential devs
12:46:25 Quit nls (Remote closed the connection)
12:46:55Bagderaliask: tell me your preferred email and forward address in a PM mail
12:47:05dan_a_aegray: You had a question earlier about the i.mx31... barrywardell has done a bit of work on the USB
12:47:10preglowBagder: can i too? :)
12:47:15Bagdergo ahead!
12:47:36barrywardellaegray: yes, I have done some work on it - started a driver
12:47:37voltagexergh, I'm lagging
12:47:44 Join nls [0] (
12:48:30voltagextesting 123!
12:51:09voltagexBagder: I sent an email to that email@rockbox address
12:54:12 Quit inversions (Read error: 110 (Connection timed out))
12:54:30 Quit aegray_ (Read error: 110 (Connection timed out))
12:56:55 Quit Shaid ("I will see you, in the end. And I will laugh at your pain...")
12:58:06 Join Febs [0] (
12:58:14dan_a_Gentlemen, I have a horribly distorted version of "My face, your ass" by Lambchop playing in my right ear
12:58:23*Llorean does the SOUND!!! dance.
12:58:46voltagexshoulda played the hamster dance for the first song
12:58:50Bagderdistortion is cool! ;-P
12:59:02*pixelma feels excluded from the fun ;)
12:59:18aliaskdan_a: I found that quite amusing as this is what I thought of:
12:59:45Lloreanaliask: Frighteningly enough, that was my first thought too.
13:00:05aliaskI had a nightmare about that damn animal thing the other week. Scary stuff.
13:00:37Bagderpixelma: time to buy another target then!
13:00:41dan_a_Now I just need to get sound in two ears, hopefully without distortion
13:01:07pixelmaBagder: I was referring to the "Gentlemen"...
13:01:16Lloreandan_a_: Did you see in the logs that it might be the voice thread that is lockstepping the audio thread on the coprocessor?
13:01:25dan_a_Ladies: I have sound!
13:01:34Bagderpixelma: I recall Christi saying something like that in the past too...
13:01:35pixelma:D thanks!
13:02:17*safetydan wonders what madman wrote vorbisfile.c
13:02:28dan_a_Llorean: I did. I had tried running with the voice turned off, which hadn't helped. Do people think that the best solution will be to move the voice thread over to the COP?
13:02:29safetydanand woo on sound :)
13:02:56dan_a_I'll hold off on committing, though - I think austriancoder should get first shot at that
13:03:34Lloreandan_a_: I dunno. But it seems that the voice thread is unusually active even with off, if a voice file is present.
13:03:38LloreanBut that's on all targets
13:04:18preglowdan_a_: looking forward to first gender neutral gentleman mail :P
13:04:55dan_a_I hope that logbot didn't feel excluded either
13:04:55 Join Sanitarium [0] (
13:05:28Sanitariumwow...just..WOW. Rockbox has changed alot since I last used it. Good work guys. And thank you
13:08:48 Quit safetydan ("Ex-Chat")
13:09:52barrywardellaustriancoder: can you send me the as3514 datasheet too, please? barry(dot)wardell(at)gmail(dot)com
13:10:25***Saving seen data "./dancer.seen"
13:11:28 Join pondlife [0] (
13:12:24barrywardellBagder: I have a question about TIME_AFTER. In what sense is it wrap-safe?
13:12:51 Join lee-qid [0] (
13:14:12 Join blind [0] (
13:14:13Bagderbecause of it counting the numbers as signed?
13:14:33barrywardellBagder: I'm thinking in terms of this fix for USEC_TIMER wrapping:
13:14:39barrywardellwould they be useful?
13:14:55JdGordonpondlife: hey
13:15:45JdGordonyou reverted my code this arvo?
13:15:50JdGordonor yesterday?
13:16:00pondlifeI didn't revert anything AFAIK
13:16:17JdGordonyeah, you unsimplified my simplified previous browser code
13:16:33pondlifeAh, it didn't work
13:16:33JdGordonit didnt?
13:16:33preglowBagder: do you think we could just check the Target: line in `gcc -v` for x86_64 to apply the patch in
13:16:33 Quit decayedcell ("CGI:IRC (EOF)")
13:16:49pondlifeNo. If you went into the Database from the menu, it didn't set previous_browser
13:17:20JdGordonoh, ok then
13:17:20pondlifeSo you return to the file browser (and could then get a nice crash)
13:17:29Bagderpreglow: yes I think we can, but that requires an existing (recent) gcc to invoke
13:17:36pondlifeThis was also why my DB auto-init stuff was stuffed
13:17:48pondlifeJdGordon: Please test what I've done though
13:17:52preglowBagder: well, you need that to compile the compilers anyway, so i don't see the problem
13:18:53JdGordonpondlife: tomorow if I get a chacne... I thought my code worked.. but i belive you if you say otherwise :p
13:19:10pondlife:p back atcha
13:19:16Bagderbarrywardell: I think you could simplify it by using the TIME_AFTER macro
13:20:35*JdGordon wants a bigger screen on his h300... some of the 5g wps's look awesome
13:21:17Bagderbarrywardell: something similar to
13:23:11barrywardellBagder: how does that compensate for USEC_TIMER wrapping around to 0 after 2^32?
13:23:50Bagderwhen counting them as signed, they wrap to negative after 2^31
13:24:04barrywardellahhh, I see now
13:24:23barrywardelli couldn't make sense of that before, thanks!
13:26:01JdGordondan_a_: bit late, but congrats on the sound :)
13:26:17dan_a_JdGordon: Thank you! Has your Sansa arrived yet?
13:26:20tHEkINDCan someone explain me how to only compile a rock ?
13:26:22JdGordonnot yet
13:26:24tHEkINDUnder cygwin
13:26:42JdGordonits somewhere between germany and aus as we speak :p
13:26:44BagdertHEkIND: it isn't possible unless you edit a few SOURCES files or similar
13:26:45nlstHEkIND: you cant
13:27:26 Quit anathema (Read error: 113 (No route to host))
13:27:38aliaskJdGordon: Ordered a sansa from overseas?
13:27:45tHEkINDIt means if i try to create something, each time i want to test it i'll need to spend 15 minutes compiling eaverything ?
13:28:15aliasktHEkIND: Just type make again, it detects the changes and builds only what it needs to
13:28:18BagdertHEkIND: get a (vmware) linux dev environment instead ;-)
13:28:36peturtHEkIND: make will only build changed files
13:30:18tHEkINDI can't use vmware -> my keyboard is an azerty and debian is in qwerty, and i didn't found how to change that
13:30:33tHEkINDanyway cygwin works nicely
13:30:44Bagderawerty is crazy
13:31:03JdGordontHEkIND: just change the keybaord setup in the vmware image?
13:31:05tHEkINDmake works well
13:31:14tHEkINDDidn't managed to do it
13:31:20tHEkINDCygwin works well now ^^
13:33:14barrywardellgrrr. should really have tested again before that commit!
13:33:59*Bagder feels a bit guilty but only a bit
13:35:14barrywardellthe mistake was my own-forgot to #include kernel.h
13:36:21preglownever even heard of that
13:36:23amiconnwow, 4269 points...
13:36:53JdGordonI could get that without even trying :p
13:36:55 Join elmargol [0] (
13:36:56nlsI was thinking about moving the playlist catalog directory configuration thing that no-one uses and that isn't documented into the regular settings but still not show it in the menu, what do people think about that?
13:37:43JdGordonalong similar lines, I wanted to make the recording setting actually choosable, and not just current or /recording .. any objections?
13:37:44amiconnJdGordon: Did you see my report regarding browser behaviour in the log?
13:38:06JdGordonis that what pondlife fixed?
13:38:12pondlifeOr broke?
13:38:18JdGordonon or the other :p
13:38:33barrywardellit's getting much easier to get high scores with all the new targets
13:38:34amiconnIt's behaviour I observed today
13:38:35pondlifeamiconn: What was it? Player icons?
13:38:54amiconnNo, cursor position when switching file browser <-> db browser
13:39:03pondlifeAh, that's been broken for a while
13:39:16JdGordonamiconn: it not remmebering in the db but it does in the tree?
13:39:18amiconnToday, starting 09:17 cet
13:39:24pondlifeJdGordon and I discussed it last week
13:39:43 Quit aliask ("G'night all")
13:40:09pondlifeJdGordon: Can you look into it perhaps? Don't think I've got much time (again).
13:40:17JdGordonSlasheri: hey, do you know if tc->selected_item is ever updated (or when?) in the db browser?
13:40:31*JdGordon doesnt think it is, which causes this problem
13:41:06 Part elmargol ("Ex-Chat")
13:41:32amiconnJdGordon: I think that neither browser should change the current position in the other
13:41:41JdGordonof course
13:41:44amiconnBoth browsers need their own current postion
13:42:07pondlifeAnyone know what the keymap is for the Player sim?
13:42:09JdGordonimo they should each have a context for their settings, not this silly shared one
13:42:22SlasheriJdGordon: hi, that should be updated outside of the db browser automatically
13:42:37JdGordondoes the db actually use that setting?
13:42:48JdGordonI didnt want to play with that before asking..
13:42:54Slasheriyes, i will check
13:43:20 Join mirak [0] (
13:43:38SlasheriJdGordon: for example tagtree_enter(), dptr = tagtree_get_entry(c, c->selected_item);
13:43:56Slasheritagtree_get_filename(), and so on
13:44:05Slasheriit's widely used to get the current selected item
13:45:42pondlifeSlasheri: Could you take a quick look at my 3 line patch -
13:45:43JdGordonAH, ok, just looking through tree.c now.. It seems I missed where it gets set.. Should be a relativly easy fix
13:45:58 Quit idnar (Nick collision from services.)
13:46:00 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
13:46:09preglowamiconn: do you know of a clever way to clear the top sixteen bits in a reg with arm?
13:46:42pondlifeSlasheri: It just does an occasional yield() during the database sort, to reduce the apparent lockup when committing.
13:47:01Slasheripondlife: yes, that looks good
13:47:14pondlifeArbitrary I know, but not too inefficient I hope.
13:47:25pondlifeSeems to make a nice difference
13:47:37Slasheribut code doesn't look very clear when you rely that the counter will wrap to zero
13:47:41Lloreandan_a_: You do realize telling _them_ that you got sound is just going to flood a series of "Why isn't it in SVN yet?" questions.
13:47:51pondlifeHmm, I was trying to keep code size down
13:48:01Slasheribetter to explicitly check some value of the counter and then reset it
13:48:11pondlifeWould it be better to make the source readable and hope that gcc optimises it
13:48:15Slasherior check current_tick to sleep at fixed points in time
13:48:18voltagexdan_a_: what have I missed? did you get stereo?
13:48:22pondlifeAh, that's a better idea
13:48:44dan_a_Llorean: I'm just going to tell them it's 'cos I'm a bastard!
13:48:46JdGordonSlasheri: any idea how to implement psuedo paths and/or set_current_file() for the db?
13:48:56dan_a_voltagex: No stereo yet, no.
13:49:07Lloreandan_a_: Ah, well if that's the plan, I'm perfectly happy with it. "Yeah, I got it working, but y'know... I just decided I'm not the sharing sort."
13:49:20Slasheripondlife: however, keep in mind also that __PCTOOL__ does not require yielding, so just ifdef it also
13:49:29voltagexLlorean: screw open source!
13:49:39SlasheriJdGordon: nope
13:49:39dan_a_I'll put a video up on Youtube, but then tell them that since they kept on asking for it I deleted it all...
13:50:01JdGordonSlasheri: :( it would make things a bit nicer... but oh well :p
13:50:20JdGordonpondlife: yeah, it was an easy fix...
13:50:28Slasheripseudo path probably needs to be a more complex path with the context for each entry
13:50:58Slasherito work with the db.. but you are free to implement it if you can find out a great solution :)
13:51:08 Part Llorean
13:51:26pondlifeHmm, where is HZ defined for each target?
13:51:42Slasheripondlife: sure
13:51:47preglowpondlife: kernel.h
13:51:52preglowpondlife: it's global for all targets
13:51:58JdGordonHZ isnt the same for all targets?
13:51:59Slasheriups, missed where :)
13:52:00pondlifeAh, ok. I thought it might vary
13:52:29Slasheripondlife: it might, but currently it does not
13:52:31Bagderwe should still work with the assumption that it may vary
13:52:33pixelmaSlasheri: easier and faster to ask you... do I remember correctly that "auto-update" only works with tagcache in RAM and/or dircache? (I don't know the wiki didn't tell me much)
13:53:25pondlifeWould "if ((current_tick % (HZ/4))== 0) yield()" result in a yield 4 times a second
13:53:39Slasheripixelma: at least dircache is required to detect deleted files automatically (i don't remember if tagcache in ram was necessary also)
13:54:12Slasheripondlife: about
13:54:17Slasheriunless we miss some ticks
13:54:27Slasheribut that is unlikely to happen
13:54:35pixelmaah thanks... so this lang string can be excluded for Archos...
13:54:54pondlifeOK, the compare is called very often.
13:55:28Slasheriand i am not sure how efficient the modulo operation is to be called very often
13:55:38pondlifeMe neither.
13:55:45obochanging the value of HZ leads to some "interesting" results :)
13:56:02JdGordonout of sync audio? what else?
13:56:07preglowshould work fine, shouldn't it?
13:56:28oboaudio was fine - but all the menu timings were out (this was before the rework)
13:56:37SlasheriJdGordon: audio is not going out of sync
13:56:48oboso trying to navigate with HZ=1000 meant some very short button taps
13:56:57barrywardellgrrr. that's what I get for only testing bootloader builds before committing
13:56:58Slasherithe codec has own clock for it
13:57:12JdGordonoh, cool
13:57:38SlasheriHZ is dependent of the configured interrupt timer tick interval
13:58:13Slasheriand the timer frequency should be a compromise between high resolution and performance
13:59:06obothere were quite a few places that do sleeps with hard coded values, IIRC things like the ata code
13:59:24Slasheriobo: but those are sleep(1), aren't they?
13:59:35Slasherithat is the smallest sleep possible
13:59:37JdGordonpondlife: did you fix 6763?
13:59:49 Quit voltagex ()
14:00:00oboSlasheri: mostly... but I thought sleep always added 1 anyway?
14:00:18pondlifeAnd now closed
14:00:30 Join webguest42 [0] (i=7c1dd65e@gateway/web/cgi-irc/
14:00:39Slasheriobo: and sleep(1) is better than yield() if we really are going to _sleep_ shortly and not just intend to switch threads
14:00:57Slasheriobo: true.. sleep(0) could be even better
14:01:15LearJdGordon: Regarding FS 6747, calling display->set_margins once would be a bit more efficient... :)
14:01:48LearThat's what the screen_set_?margin macros do, calling get?margin on the "other" value...
14:01:52Slasheriobo: of course, sleep() does not guarantee a sleep but makes it possible if there are no other threads in the list of running tasks
14:01:53JdGordonah yes... hmm
14:02:23 Quit funky (Read error: 131 (Connection reset by peer))
14:02:41LearOnly a few bytes, but still...
14:03:26pondlifeSlasheri: How about this:
14:03:53JdGordonLear: cheers, yeah a few bytes, and a couple times /s adds up
14:03:56pondlifeI know we'll miss a wrap round, but it's not critical
14:04:38 Quit webguest42 (Client Quit)
14:05:17Slasheripondlife: that looks better, but it still can be improved a bit :)
14:05:24pondlifeGo on..
14:05:32Slasherilast_tick can be wake_up_tick
14:05:46pondlifeAh, don't + so much
14:05:58Slasheriif (current_tick >= wake_up_tick) { wake_up_tick = current_tick + HZ/4; yield(); 0
14:06:03Slasheriyep :)
14:06:56barrywardellhmmm. why isn't HAVE_LIMITS_H defined for libmad?
14:07:01pondlifeOK... and I'll commit that once I've tested the timing
14:07:10Slasheripondlife: great
14:07:42JdGordonbuilds are slow tonight ? or am i just more imp[atient than normal
14:08:13Slasheripondlife: probably you should yield a bit more offen, like HZ/16
14:08:33Slasherithen priority scheduling should do rest
14:08:38pondlifeOK, I'll play
14:08:55pondlifeDon't want to overdo it and slow down the qsort
14:09:17pondlifeJust want to remove that "Rockbox has locked" experience
14:09:40Learjdgordon: a little perhaps. Used to be 6-7 minutes, but that was with fewer targets.
14:10:29pondlifeSlasheri: I'd be grateful if you could also check my root_menu.c tests that are used to decide if it would be a good idea to ask the user to init their db.
14:10:57pondlifeDon't want to cause any unnecessary inits.
14:11:04Slasheripondlife: i will soon
14:12:02pondlifeMaybe the init now/update now options should be moved from settings into debug...?
14:12:30pondlife(And I could update rather than init if that's detectable.)
14:13:47 Join tHEkIND_ [0] (
14:14:32 Quit tHEkIND (Read error: 131 (Connection reset by peer))
14:15:14Slasheripondlife: hmm, how would you update the db (manually) then?
14:15:35Slasheriis user does not want to use the auto-update
14:16:00pondlifeGood point...They don't want to be prompted do they?
14:16:24Slasheriprobably not prompted _every_ time they enter to the db browser =)
14:17:42nlspondlife: I think the initialize now function should be moved to debug menu now as it is easier to do it the new way and many people are confused by it and first click initialize and the try to update and the wonder why it doesn't work
14:18:26pondlifeYes, I spotted this confusion, hence the new approach.
14:18:28Slasheriand accidentally clicking "initialize now" isn't good thing either as it erases the entire db if exist
14:19:11 Join robin0800 [0] (
14:19:12nlsyes, and it should only be necessary once unless the database gets corrupted by some bug ;-)
14:19:31dan_a_Does anyone know of an easy way to graph which functions get called by other functions in Rockbox? Either with a Windows tool or a Linux one?
14:20:20Slasheridan_a_: hmm, not sure but you could check some C source code analyzer utilities
14:20:46pondlifeJdGordon: root_menu.c has this previous_music variable, so that PLAY can go back to either WPS or radio. I thought it was decided that PLAY would not go to radio?
14:20:59peturdoesn't doxygen do this?
14:21:00pondlifei.e. PLAY always goes to WPS
14:21:26JdGordonpondlife: no, the now playing item always goes to wps.. play goes to previous tho
14:21:45pondlifeAh, ok. Should perhaps recording be covered by this variable too then?
14:22:34pondlifeThe browsers seem to remember better now by the way, thanks
14:22:37Slasheripondlife: i will remove that readyvalid and move initialized to the correct place
14:23:00 Quit perplexity (Read error: 110 (Connection timed out))
14:23:02pondlifeSlasheri: I tried that, it slowed down bootup
14:23:25Slasheriah, yes
14:23:28pondlifeAlthough you could probably remove that check now the root_menu does it anyway
14:23:30Slasherii will think that
14:23:39JdGordondont yield at all if its updating because of a bootup
14:24:07pondlifeIf you select Database whilst it's initing, it will now just display "Database is not ready" until it becomes ready..
14:24:21pondlifeWell, that's the intention.
14:24:38pondlifeSo no need to wait as part of general bootup
14:24:40Slasheripondlife: so it no longer displayes the current progress?
14:25:07pondlifeDepends. It does that if it's initializing the db. I meant initialising the thread
14:25:17Slasherithen that's ok
14:25:26pondlifeThe two meanings of initialise that had me so confused.
14:25:50nlspondlife: does that work ok when setting the db browser as start screen?
14:26:00nlsnice :-)
14:26:08pondlifeIt all drops out in the wash, so to speak
14:26:27pondlifeAssuming my tests on the database stats are correct, of course!
14:27:07nlsthere was a bug with it when starting and the database thread was initializing and it would say not ready and drop you in the main menu
14:27:30pondlifeIt will now come up "Not ready" and wait until it is ready (or the user presses STOP)
14:27:41pondlifeBut it's pretty quick
14:27:58nlsaha, that is much nicer :-)
14:31:01 Quit pixelma (" bb")
14:31:55pondlifeSlasheri: When committing, what else locks in tagcache.c for a while (aside from qsort)?
14:32:00pondlifeFile writing?
14:32:23Slasheripondlife: no other things should lock, there should be already yields in other loops
14:32:32pondlifeIt's better with the sort yields, but still not ideal
14:32:45pondlifeHZ/20 and HZ/4 feel the same, btw.
14:33:25preglowdon't people fucking check what features we have before requesting anymore?
14:33:35pondlifeDisk writes have an implicit yield, right?
14:33:39Slasheripondlife: try HZ/100
14:33:53Slasherithat shouldn't be too often to affect the performance
14:33:58pondlifeEvery tick
14:34:02pondlife(at the moment)
14:34:10Slasherihmm, yes
14:34:35Slasheriaudio codecs yield/sleep also very often
14:35:13pondlifeI'm just navigating the main menu for my test. If it locks for more than 10 seconds that is regarded as bad
14:35:53amiconnJdGordon: Hmm, it seems an old action bug is back now :(
14:36:02 Join Thundercloud [0] (n=thunderc@
14:37:00amiconnIf you are in a plugin that has a menu with a 'Quit' option, and you invoke this option with 'Right', the plugin exits - only to make the plugin browser immediately call it again
14:37:19amiconnJust observed on H1x0
14:37:36pondlifeThat was working yesterday, is it a brand new bug?
14:37:39amiconnStrangely enough that doesn't happen when invoking Quit with 'Select'
14:37:55pondlifeAh, I might have used Select
14:38:06JdGordonjust needs an action_signalscreenchange() then
14:38:22amiconnI often use Right instead of Select to avoid the loud (mechanical) click
14:38:43pondlifeAh, yes H1x0
14:38:46JdGordonamiconn: which plugin?
14:38:58pondlifeAlso try brickmania
14:39:17 Join inversions [0] (
14:39:20JdGordonhmm.. yeah, its here alos
14:39:23JdGordonalso even
14:39:40Slasherii always press the select using both thumbs and cloves to prevent that annoying click :)
14:39:42*JdGordon hates plugins with splash screens
14:39:55 Join anathema [0] (
14:40:14JdGordonshould plugin_load() do the signalscreenchange call or the plugin browser?
14:40:21pondlifeSlasheri: HZ/100 is just the same
14:40:32 Part Strath
14:40:37Slasheripondlife: hmm, then it must be something else..
14:40:37pondlifeI need to do some profiling I guess
14:40:51Slasheripondlife: the disk is not active when the locking happens?
14:41:05JdGordonis the cpu being boosted?
14:41:16pondlifeSlasheri: I think so
14:41:24pondlifei.e. I think the disk is active
14:41:29Slasheripondlife: then there are really not many places to cause that
14:41:53Slasheripondlife: try putting much yields to the loops between that qsort to see if it helps
14:47:04pondlifeHmm, if I initialise the database, there's a period where it can be browsed before it gets invalidated.
14:47:30Slasheripondlife: it is valid until the committing begins
14:47:53Slasheriduring that time, database can't be accessed
14:48:12pondlifeSo it writes to temp files?
14:48:40pondlifeEven better is there a wiki to explain the process, and how RAM/disk options affect it.
14:48:47Slasheriyep. at first it generates a temporary db about the files to be added to the db and then that file is committed
14:49:01preglowamiconn: do you know of a clever way to clear the top sixteen bits in a reg with arm?
14:49:11Slasheripondlife: the diagram at beginning of tagcache.c explains it a bit
14:49:12preglowpreferably one instruction :>
14:49:17pondlifeSo the commit is a delete/rename?
14:49:38Slasherino, it's much more
14:50:15pondlifeWhy? We could reduce database downtime maybe?
14:50:32Slasheriduring commit all referred string indexes are resorted, and lookup id's re-assigned
14:51:12Slasherihmm, the downtime is only a few seconds if just a few files are added..
14:51:32Slasheribut indeed, it _could_ be reduced to zero
14:52:14Slasheribut that needs copying all database files for the commit to operate with them, then remove the old ones and rename
14:52:16 Join Arathis [0] (
14:52:33Slasheripondlife: not sure if that really is necessary
14:52:39pondlifeProbably not
14:52:46Slasherimakes just code more complex and slower
14:53:08JdGordonamiconn: ok fixed
14:53:19 Join Stefan [0] (n=chatzill@
14:53:21JdGordonthis fix should have been done back in august tho :p
14:55:15pondlifeOK, the lockup is in the "Generate reverse lookup entries" loop
14:55:34pondlifePutting a timed yield in there fixes it.
14:55:52Slasheripondlife: build_index() is the most important function during the commit, and probably the largest and complex one in the tagcache :)
14:56:11pondlifeWell, I'm peepholing it...
14:56:21Slasheripondlife: ah, sounds good. so probably yield in both compare and that loop would be good
14:56:28Slasheriit might be good idea to create a function of that timed yield
14:56:35pondlifeYep, just did that!
14:56:39Slasherinice :)
14:56:59Slasheripondlife: then it could be called in other placed too!
14:57:11Slasherito replace almost all yields in tagcache.c with that one
14:57:32 Join voltagex [0] (
14:57:51pondlifePut the #ifndef __PCTOOL__ outside the function call I assume...
14:58:27Slasheripondlife: just surround the whole function with that ifdef
14:58:55Slasheriand ifdef it as empty define if __PCTOO__ is defined
14:58:57pondlifeYes, I meant not to bother with any yields at all on the __PCTOOL__
14:59:08 Join dukeman [0] (
14:59:08pondlifeThere are some that are done
14:59:09Slasheriok :)
14:59:11 Quit Stefan ("Chatzilla 0.9.74 [Firefox]")
14:59:50pondlifeAh, I see - they're already #defined out
15:00:18 Quit inversions (Read error: 113 (No route to host))
15:00:20JdGordonwhy not put that timed yield in misc.c so they whole apps can use it?
15:00:28JdGordonor even redefined yield to it?
15:00:40pondlifeUrgh, side effects?
15:00:51JdGordonhmm.. yeah, scratch thats 2nd line
15:00:55Slasheriyield should not be redefined
15:00:58pondlifeTiming needs lots of testing.
15:01:21Slasheribut for some other apps it might be useful as well
15:02:05 Quit XavierGr ()
15:02:38Slasheripondlife: for example find_entry_ram() that counter could be removed too and replaced with the timed yield
15:02:45pondlifeJust did that
15:02:48Slasherigreat :)
15:03:03pondlifeMade my fn return true if it actually yielded...
15:04:17 Join rp- [0] (i=rp@
15:04:54rp-dan_a: where is the patch? :)
15:05:45dan_a_rp-: There's still lots to do - I seem to keep losing I2S sync.
15:07:38rp-i2s isn't covered very mutch in the datasheet
15:07:58 Join XavierGr [0] (
15:08:04 Quit barrywardell ()
15:08:25pondlifeSlasheri: Nothing to do with my mods, but the first Initialize Now I do always completes really quickly (<10 seconds). Sound like an Update Now...?
15:09:11pondlifeIt just counts up from 0-100% quickly, with no committing.
15:09:39 Join K3nto [0] (
15:09:46Slasherihmm, then it does just the updating instead of initialization.. weird
15:10:09pondlifeFirst time after I boot up....
15:10:32***Saving seen data "./dancer.seen"
15:10:32Slasheripondlife: initialize now should call tagcache_rebuild
15:10:41Slasherifirst thing that function does is to remove all db files
15:10:54pondlifeI have load to ram and auto update enabled.
15:11:10Slasherihmm, it should also disable the db
15:11:37Slasheripondlife: or maybe you didn't wait long enough for the auto update to run?
15:11:46Slasheriand the initialize command got queued
15:11:59pondlifeThat's it
15:12:13pondlifeMaybe the initialise should clear any updates from the head of the queue?
15:13:11Slasherihmm, maybe. it should somehow stop the current operation first
15:13:44tHEkIND_What is "GPIOD_OUTPUT_VAL" ?
15:14:40Slasherisounds like GPIO port D output buffer register
15:15:04Slasheriwhich contains the value of the port D
15:15:30tHEkIND_what's on port d ^^
15:15:31Slasheriups, which is used to put the output value to the buffer
15:15:39Slasheriinput register would contain the value
15:15:47tHEkIND_okay i see
15:15:53 Quit JdGordon ("Konversation terminated!")
15:16:00SlasheritHEkIND_: depends on the target
15:16:14tHEkIND_I'vee seen on i/o ports
15:16:24Slasheriprocessor data sheet should contain information how the I/O port D is connected to the pins
15:16:34tHEkIND_If you touch the scrollpad on an H10, port D changes from FF to df
15:16:36Slasheriwell, that is an I/O port
15:16:49tHEkIND_FD actually
15:16:55tHEkIND_DD to FD
15:16:58Slasherithen scrollpad is connected to port D
15:17:30tHEkIND_What's that difference beetween |= and =
15:17:30tHEkIND_ ?
15:17:34pondlifeSlasheri: OK, yielding commit patch is now on
15:17:36tHEkIND_and &=~
15:17:56Slasheri|= is same as value = value | something;
15:18:06Slasherithose are binary or, and, not operations
15:18:07pondlifeHmm, I would seriously learn C before you play with I/O ports.
15:18:31tHEkIND_I'm only modifying how rockbox decodes
15:18:36tHEkIND_Nothing more
15:18:44Slasheridecodes what?
15:18:51tHEkIND_The scrollpad
15:19:20tHEkIND_Now, rockbox goes up if you touch the top of the scrollpad
15:19:30Slasheripondlife: hmm, compare() does not use the timed yield function, is that intended?
15:19:30tHEkIND_And goes down if you touch the bottom
15:19:37tHEkIND_And is buggy on the middle
15:20:00Slasheripondlife: otherwise it looks ok to commit :)
15:20:24pondlifeCan you test that the PC tool builds?
15:20:28pondlifeOr can I?
15:20:36Slasherijust cd tools && make database
15:22:10pondlifeI get "cc1: error: unrecognized command line option "-iquote""
15:22:17pondlifeLots of times
15:22:32Slasherihmm, you have a _very_ old host compiler? =)
15:23:00pondlifegcc version 3.4.4 (cygming special, gdc 0.12, using dmd 0.125)
15:23:42 Quit Id2ndR ("Parti")
15:23:53Slasheri"-I- This option has been deprecated. Please use -iquote instead ..."
15:23:57Slasheriman gcc says that
15:24:08Slasherimaybe cygwin is something weird then
15:24:25Slasheribut that can be fixed later when pctool is more usable
15:25:05pondlifeOK, I'll commit and let you test it compiles ;)
15:25:12Slasheriand indeed, you have an old gcc
15:26:33preglowjhMikeS: awake?
15:26:35amiconnpreglow: All I know is that gcc clears the top 16 bit by a left-shift-right-shift sequence (2 insns). I can't think of another way
15:26:49 Join muesli__ [0] (n=muesli_t@
15:26:52preglowamiconn: nah, me neither, so i'll just stick to wasting a reg on a mask
15:28:13amiconnUnfortunately arm doesn't have partial register ops (like the coldfire)
15:28:23amiconnWith those you can do some neat tricks
15:28:36preglowarm doesn't really even have a notion of a partial reg, apart from when writing/reading to memoruy
15:29:14preglowoh well, it's risc
15:29:25 Quit Stalwart (Read error: 104 (Connection reset by peer))
15:29:57 Join Stalwart [0] (
15:30:46pondlifeHmm, I wonder if the yielding should be skipped if we're doing a delayed commit...?
15:31:17pondlifeSlasheri: Any way to know if we're doing a commit during startup?
15:32:37Slasheripondlife: if TAGCACHE_FILE_TEMP exists, that is pretty good sign for a commit during startup
15:32:46 Join webguest32 [0] (i=42c07542@gateway/web/cgi-irc/
15:33:07 Nick webguest32 is now known as SoapOffice (i=42c07542@gateway/web/cgi-irc/
15:33:07Slasherior is commit_delayed flag is set
15:33:28 Join perplexity [0] (
15:35:19 Join lion2k7 [0] (i=4db38812@gateway/web/cgi-irc/
15:35:28 Quit Obsys ("Chatzilla 0.9.77 [Firefox]")
15:35:45pondlifeSlasheri: OK, committed that, will check back after lunch
15:35:58dan_a_Hi lion2k7
15:37:17CtcpPing on #rockbox from lion2k7!i=4db38812@gateway/web/cgi-irc/
15:37:17***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! ( )'
15:37:29 Quit Thundercloud (Remote closed the connection)
15:39:21 Quit voltagex (Read error: 110 (Connection timed out))
15:39:55 Join Thundercloud [0] (n=thunderc@
15:40:21 Quit lion2k7 ("CGI:IRC")
15:42:45 Join DerZerleger [0] (
15:44:10 Join fasmaie [0] (
15:45:20tHEkIND_I can write that: up = data + 0xAA; ??
15:45:59tHEkIND_will see
15:46:30tHEkIND_oka it works
15:47:07 Quit DerZerleger (Client Quit)
15:49:25dan_a_Arrrgghh! Are there any linker script gurus about?
15:51:49 Quit lee-qid (Read error: 110 (Connection timed out))
15:54:09 Join muesli- [0] (n=muesli_t@
15:54:58 Join eggie [0] (
15:56:00 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
16:00:30 Join funky [0] (n=repulse@unaffiliated/funky)
16:01:29 Join Criamos [0] (
16:01:58 Quit eggy (Read error: 145 (Connection timed out))
16:02:38 Quit muesli__ (Read error: 110 (Connection timed out))
16:05:11 Join habana [0] (i=58a10615@gateway/web/cgi-irc/
16:06:07habanaCONGRATULATIONS HACKERS FOR SOUND (on the sansa) !!!!
16:06:51habanawe dumb users love your valuable work ^^
16:06:58dan_a_habana: Thank you! It's not working correctly yet, though
16:06:58 Join kruegge [0] (
16:08:10habanabêêêêê You just did it. The proof of concept is just beautiful :)
16:10:07 Quit lini (Read error: 110 (Connection timed out))
16:11:06 Join lini [0] (i=pugsley@
16:13:53 Join lee-qid [0] (
16:13:57 Join Thundercloud_ [0] (n=thunderc@
16:29:34 Quit Thundercloud (Read error: 110 (Connection timed out))
16:30:01 Join toffe82 [0] (
16:35:24 Join muesli__ [0] (n=muesli_t@
16:37:11 Join FelixKarl [0] (
16:37:58 Part FelixKarl ("Ex-Chat")
16:39:08 Quit kruegge ("KVIrc 3.2.5 Anomalies")
16:41:07SoapOfficeRe: the people suffereing from button "bounces" on the iAudio X5:
16:41:54SoapOfficeHow hard (how needed) would a button calibration plugin be? Or are the button press values embeded so deep in the code that such an idea would be impratical?
16:42:31 Join FOAD_ [0] (
16:43:01SoapOfficeJust reading the mailing list and seeing how different people have different "stable" values for button presses, and I can imagine that values for button presses could drift over time as the physical contacts first break in, then wear and get even dirty.
16:44:19 Join FelixKarl [0] (
16:46:11 Quit muesli- (Read error: 110 (Connection timed out))
16:50:31 Quit zylche (Remote closed the connection)
16:50:31 Quit FOAD (Read error: 145 (Connection timed out))
16:51:41 Quit fasmaie ()
16:53:36 Join fasmaie [0] (
16:59:41tHEkIND_Can someone help me
16:59:46 Join zylche [0] (
16:59:47tHEkIND_I'd need to create a rock
16:59:53tHEkIND_To show values
17:01:48peturtHEkIND_: look at helloworld.c in /apps/plugins
17:01:59tHEkIND_Just saw that thanks
17:02:33peturtHEkIND_: you'll need to add your c file to SOURCES and FILES to have it build and added to the zip
17:03:02tHEkIND_On the folder apps
17:03:13tHEkIND_Inside my folder builds or the other ?
17:03:29petur /apps
17:03:39tHEkIND_okay thanks
17:03:58 Quit ompaul ("init 6")
17:04:10peturerr... FILES is not needed as it already adds all c files
17:04:21 Quit fasmaie ()
17:04:23tHEkIND_only c yes
17:04:41peturc and h ;)
17:06:28 Quit robin0800 ("IceChat - Its what Cool People use")
17:06:38 Join robin0800 [0] (
17:06:56 Join robin0800_ [0] (
17:07:36 Quit FelixKarl ("Ex-Chat")
17:07:45 Join MrSun [0] (
17:08:00austriancoderBagder: did you recevie it?
17:08:03 Quit Stalwart ("room network restructurization, will be back soon")
17:09:11austriancoderBagder: i send it to the mail addi listed in
17:09:11 Quit robin0800 (Client Quit)
17:09:33 Quit robin0800_ (Client Quit)
17:09:37 Join robin0800 [0] (
17:10:14SoapOfficeaustriancoder: so, did I miss an earlier IRC conversation about your meeting on Friday? (or am I thinking the wrong Friday?)
17:10:35***Saving seen data "./dancer.seen"
17:10:47peturSoapOffice: yes you did
17:11:50 Quit mirak (Read error: 104 (Connection reset by peer))
17:12:41*austriancoder will write a wiki page about it soon
17:13:12rp-hi austriancoder
17:13:24rp-how was your trip back? :)
17:16:41 Quit MrSun ("Ex-Chat")
17:16:51 Join MrSun [0] (
17:18:07austriancoderrp-: hi
17:19:16austriancoderrp-: i am very very tired but i played with the developer board for some time and the drivers is getting..
17:21:18 Join Id2ndR [0] (
17:22:20rp-i'm playing with the sound, but i2s seems to make trouble
17:23:25*petur gets the feeling several people are doing the same work in parallel
17:23:37preglowdan_a has trouble with i2s too
17:23:57austriancoderevery body wants to the man who...
17:24:11rp-i know, we are talking about it
17:24:16dan_a_petur: We are - I don't think that's always a bad thing
17:24:40austriancoderdan_a: hmmm
17:25:08*SoapOffice is happily amazed that even though the datasheet can not be made public they are willing to allow code based upon the datasheet to be made public.
17:25:57preglowdon't ask...
17:26:42dan_a_SoapOffice: As I understand it, they want to avoid people changing the dc-dc converter to send lots of volts through the system, but they're happy for music to be played
17:27:20dan_a_austriancoder: Have you got any results with sound yet?
17:27:44 Join Nico_P [0] (
17:28:32 Join roolku [0] (
17:28:53 Join mattzz [0] (
17:28:56 Part K3nto
17:29:30SoapOfficedan_a: That's cool as water-to-wine. I'm always happy when the cynic in me is proven wrong! Bring on the new Flash-player!
17:29:30 Quit habana ("CGI:IRC (EOF)")
17:30:42 Join Menollo [0] (i=53740bec@gateway/web/cgi-irc/
17:34:02 Join mirak [0] (
17:36:10 Quit Menollo ("CGI:IRC (Ping timeout)")
17:36:33tHEkIND_is a way to write something on the screen ?
17:37:51dan_a_tHEkIND_: Have a look at docs/PLUGIN_API
17:39:43mattzzin a plugin you can use rb->lcd_putsxy(0, 15, "Something");
17:40:08mattzz[ ]
17:40:35 Part nls
17:40:51 Quit MrSun ("Ex-Chat")
17:41:36 Join bluebrother [0] (i=Jheu8JLH@rockbox/staff/bluebrother)
17:41:54toffe82SoapOffice: hello are yoy Soap ?
17:43:05tHEkIND_And withoud plugin ?
17:45:40 Quit Nibbier (Read error: 104 (Connection reset by peer))
17:45:40 Quit iloric ("Miranda IM! Smaller, Faster, Easier.")
17:45:51 Join iloric [0] (
17:48:21SoapOfficehey toffe82
17:48:28SoapOfficelistening to it as we speak.
17:48:43toffe82is every thing ok ?
17:48:47 Join twenty6ix [0] (
17:48:57twenty6ixhi all, is anyone able to answer a couple Rockbox questions for me?
17:49:20twenty6ixi checked the FAQ and i couldn't find what i'm looking for
17:49:24thegeekit's generally better if you just ask
17:49:29SoapOffice'tis all good. There is that broken snap in the case-joint upper-left corner - but that will be covered by the case I ordered - so I couldn't be happier.
17:50:02twenty6ixhaha, alright then, just making sure people were around... anyway, i'm getting an iPod next week, and i want to install Rockbox cause i hate iTunes, but i'm wondering if Rockbox will still play videos, and if so, will it only play .MP4's?
17:50:42SoapOfficemp4s will not happen on the ipods until if the Broadcom chip gets reverse engineered.
17:51:01 Join jhulst [0] (
17:51:09SoapOfficethere is not enough horsepower in the main processor to do H.264.
17:51:34twenty6ixwell, i meant iPod-format video, i thought it was .mp4
17:51:34SoapOfficeBut MPEG1/2 works now. See the wiki PluginMpegplayer
17:51:50twenty6ixwill divx or xvid work with Rockbox?
17:52:12tHEkIND_No way
17:52:16twenty6ixhm ok
17:52:25tHEkIND_Xvid and Divx requires too much CPU
17:52:30twenty6ixso to play video with Rockbox i have to convert it to .mpg?
17:52:42toffe82Soap : I will get some more F40 this week, if you have any problem with the case, I will send you another one (case)
17:52:49twenty6ixok, i appreciate the help guys, thanks a lot
17:53:41 Quit twenty6ix (Client Quit)
17:53:59SoapOfficethat's what I'm talking about also. ("iPod-format video") The iPod main CPU does not decode those videos on the 5/5.5G. Processing is offloaded to a Broadcomm dedicated chip which is not supported by rockbox, Rockbox will decode MPEG1/2 videos, but does so on the main processor, so performance is reliant on the processor speed:screen size ratio. The ipod video models fall on the low-FPS side of that rati
17:54:10SoapOfficetoffe82: fooey.
17:54:16 Join kvader [0] (
17:54:30SoapOfficetoffe82: you sold me a fine product at a fine price - I have no complaints.
17:54:53SoapOffice(oops, my long rant to twenty6ix was too late)
17:54:56toffe82I want every body happy, so I take care ;)
17:56:19SoapOfficeI installed an updated version of the Gigabeat firmware first thing - gah! That was a mistake. What a silly firmware.
17:56:31toffe82I try to find a hd interface with zif connector and it is impossible to find it inthe us for cheap (120$ here , 35$ in europ) :9
17:58:05toffe82need somebody to buy one in eyrop and send it to me
17:58:08JETC-is rockboy finally ported to X5?
17:58:17toffe82europ ^^
17:58:28JETC-i don't see it in the plugins
17:59:30tHEkIND_Try to upload a gb ron on your player
17:59:44tHEkIND_Rockboy don't show up on plugins i thought
18:00:28 Quit kvader ("No windows for this server")
18:00:55 Join kvader [0] (
18:01:15preglowamiconn: do you know how much pp gains from burst transfers? does it do them at all?
18:04:03 Join mattzz_ [0] (
18:05:50 Quit mattzz (Read error: 110 (Connection timed out))
18:07:40JETC-found it
18:07:44JETC-under viewers
18:12:31 Join efyx [0] (
18:13:10 Join Nibbier [0] (
18:13:18 Quit mirak (Remote closed the connection)
18:15:39 Join barrywardell [0] (
18:21:19 Quit mattzz_ (Read error: 104 (Connection reset by peer))
18:24:08 Join mattzz_ [0] (
18:25:23 Part austriancoder ("Kopete 0.12.4 :")
18:27:26 Join Hoffmann [0] (
18:28:40 Quit robin0800 ("IceChat - Its what Cool People use")
18:28:42 Quit BigMac (Read error: 110 (Connection timed out))
18:32:54 Quit Lear ("Chatzilla 0.9.77 [Firefox]")
18:34:28 Part kvader
18:34:35preglowjhMikeS: damned clever trick, that swapping phase trick
18:35:03 Join lion2k7 [0] (i=4db38812@gateway/web/cgi-irc/
18:35:39dan_a_Hi lion2k7
18:36:26 Quit lion2k7 (Client Quit)
18:36:46 Join lion2k6 [0] (i=4db38812@gateway/web/cgi-irc/
18:39:19 Join _Amblin [0] (
18:40:46 Quit pondlife ("disconnected has pondlife")
18:40:46 Quit lion2k6 (Client Quit)
18:45:00 Quit muesli__ (Read error: 110 (Connection timed out))
18:47:45 Quit petur ("sssssssssss---------PLOP!")
18:55:21 Join relaxed [0] (n=relaxed@unaffiliated/relaxed)
18:58:43 Join BigMac [0] (
19:03:35 Nick mattzz_ is now known as mattzz (
19:05:12*mattzz wonders how the maze plugin will make its way into svn
19:05:54 Join robin0800 [0] (
19:07:53preglowdivine intervention
19:09:34 Join webguest22 [0] (i=502acc61@gateway/web/cgi-irc/
19:09:43 Part webguest22
19:10:41***Saving seen data "./dancer.seen"
19:14:13 Join Ice8lue [0] (
19:14:47Ice8luegood evening
19:16:20Ice8lueis one of u working on the sansa port?
19:16:40barrywardellyes, a few people here are
19:16:54dan_a_Ice8lue: rp-, austriancoder and myself are right now
19:17:03 Join mirak [0] (
19:17:29Ice8lueso how was the meering with AMS? i read there are first beeps from the device?
19:17:43Ice8lueoh, meeting
19:18:46 Part tHEkIND_
19:18:55dan_a_I didn't go, but I hear it was very good. And, yes, there is (poor quality) sound coming out now
19:18:58 Nick eggie is now known as eggy (
19:19:50mattzzdan_a: great news!
19:20:16dan_a_mattzz: I know - apart from the poor quality bit!
19:20:22Ice8lueyay very good news =) cant wait to play quake with sound^^
19:20:33Ice8lueis there any possibility how i could help?
19:20:52mattzzdan_a: how poor is it?
19:20:57barrywardelldan_a_: how far have you got with the sound?
19:21:15barrywardelldan_a_: is it the same missing bits problem MrH had?
19:21:54dan_a_Distorted sound in the right channel, very quiet distorted sound in the left channel - maybe as if one bit from the left channel was going into the right
19:22:20_Amblinbut it beats the hell out of no sound at all
19:22:25_Amblinonly a matter of time
19:22:44dan_a_And it keeps dying to just blips - I think we lose sync on the I2S bus
19:23:22 Join pearldiver [0] (
19:23:25Ice8luedo u still need ppl for the team?
19:24:05dan_a_Ice8lue: Always!
19:24:44Ice8lueso what can i do? am learning C atm....aybe som graphics design?
19:25:49barrywardellIce8lue: I think the plugins are a good way to get into Rockbox programming
19:26:17barrywardellmaybe try writing your own plugin, or adapt a one that doesn't work on the sansa yet
19:26:35dan_a_Ice8lue: Whatever interests you :D interested in programming but i think my skills are too low. graphics would be nice...maybe theming or such things
19:28:58barrywardellWPS's will hopefully be useful in the not too distant future ;)
19:29:40Ice8lue^^ well, i think now itsa only a matter of time
19:32:32rp-dan_a_: did you figure out anything?
19:32:44dan_a_rp-: Not yet
19:34:13 Nick Kape is now known as Akitochi (
19:35:02rp-i thought with the datasheet it would be easier, damn i2s
19:35:19 Nick Akitochi is now known as Kape (
19:35:32dan_a_rp-: We've got further than we would have done without the datasheet
19:35:38 Quit perldiver (Read error: 110 (Connection timed out))
19:36:31rp-thats right, who is gona request some datasheet for the pp from nvidia? :)
19:36:32 Quit Hornet_ (Read error: 104 (Connection reset by peer))
19:36:41 Join Hornet_ [0] (
19:39:05 Part Ice8lue
19:40:49 Join skyfaller [0] (n=skyfalle@wikipedia/Skyfaller)
19:41:04rp-success no more distortion
19:41:33skyfallerhey folks... I have a 1st gen iPod nano, the 1GB model... but on the page there is no partition table file for the 1GB model
19:42:05skyfallercould anybody help me reformat my iPod, or do I have to find a Windows computer?
19:44:51barrywardellrp-: great!
19:46:43 Join Lurker [0] (i=4a87b2ea@gateway/web/cgi-irc/
19:47:48 Quit lee-qid ("aufwiederbyebientotsayonara")
19:49:52 Quit _Amblin (Read error: 110 (Connection timed out))
19:51:02 Quit Nico_P (Remote closed the connection)
19:52:06 Join lee-qid [0] (
19:52:22 Join Jon-Kha [0] (
19:53:53 Quit darkless ("Leaving")
19:57:13 Join webguest14 [0] (i=425dac13@gateway/web/cgi-irc/
20:07:16 Quit juxtap (Read error: 104 (Connection reset by peer))
20:14:15 Join nls [0] (
20:15:59barrywardellrp-, dan_a_: is there a patch I can try?
20:17:41 Join midgey [0] (
20:18:21 Quit iloric (Read error: 104 (Connection reset by peer))
20:18:44rp-i will make one for you
20:19:56 Quit pabs (Remote closed the connection)
20:20:07 Join pabs [0] (
20:20:21 Quit SoapOffice ("CGI:IRC (EOF)")
20:27:24 Join mike__ [0] (
20:31:42 Quit BigMac (Read error: 110 (Connection timed out))
20:35:01 Quit Febs (Read error: 110 (Connection timed out))
20:45:54 Join aegray_ [0] (
20:49:40 Nick mike__ is now known as BigMac (
20:50:33 Quit funky ("leaving")
20:54:27 Quit webguest14 ("CGI:IRC")
20:55:00 Join Rondom [0] (
21:01:53 Quit aegray (Read error: 110 (Connection timed out))
21:03:40 Quit Criamos (Read error: 110 (Connection timed out))
21:10:45***Saving seen data "./dancer.seen"
21:11:15 Nick w1ll14m is now known as w1ll14m_ (
21:12:04 Quit Lurker ("CGI:IRC (EOF)")
21:14:06 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:15:33 Join muesli__ [0] (n=muesli_t@
21:22:41 Join webguest51 [0] (i=502acc61@gateway/web/cgi-irc/
21:25:33 Quit webguest51 (Client Quit)
21:29:08 Join mcscruff [0] (
21:30:04mcscruffi had a look through the manual, and am initizalising my files but its saying found 0 (i have a 30gb ipod video) will it read my ipod files?
21:31:07nlsmcscruff: it should read any non-drm files...
21:31:33mcscruffi just restarted it, i dont think i have installed it correctly
21:31:38nlsis the little disk icon in the upper left corner visible?
21:31:55nlsright corner even
21:31:56mcscruffsaid something about no directoryand installation incomplete
21:32:07mcscruffonly got the time in the right
21:32:32nlsmcscruff: ah, yes you need to install the .rockbox directory directly on the root of your player
21:32:43 Join x1jmp [0] (
21:32:44mcscrufftheres my problem :P
21:32:46 Quit XavierGr (Read error: 113 (No route to host))
21:32:52mcscruffi just put the rockbox.ipod on there
21:33:03nlsget the directory too
21:33:09 Quit mirak (Remote closed the connection)
21:35:13mcscruffcoz it had a . it was hidden
21:36:08mcscruffif i plug it into the pc, why does it not come up as usb, do i need to put the ipod into disk mode
21:37:53 Quit midgey ()
21:43:52mcscruffnls: i put it in the root dir and get the same problem
21:44:41 Join K3nto [0] (
21:46:23 Join ross [0] (
21:47:03K3ntohey guys, i have a broken set of headphones im trying to fix
21:47:09K3ntosomewhat on-topic
21:47:21rossHey everyone. I'm going to buy a Video iPod. I know RockBox currently does NOT support the 80 gig model, but I would like the space (for FLAC files). Will there be support for this model in the near future?
21:48:33dan_a_ross: It's impossible to predict. That said, we know what the problem is, so at some point in the future there will be support
21:49:04 Quit barrywardell ()
21:49:05thegeekthere is a very good chance it will happen
21:49:15thegeeksince linusn seems to know exactly what is wrong
21:50:01rossexcellent. Thanks dan_a_ and thegeek
21:50:46preglowit's just a matter of time, really
21:50:51preglowbut we can't tell you how much of it
21:52:24rossOK, thanks!
21:52:29 Quit ross ("Leaving")
21:53:16 Join merlin2049er [0] (
21:54:16 Quit skyfaller ()
21:54:49 Quit SirFunk (Remote closed the connection)
21:54:54 Join midgey [0] (
21:56:03 Join linuxstb [0] (i=d529f3ae@rockbox/developer/linuxstb)
21:56:56*linuxstb reads the logs and wonders if there is anyone not writing a Sansa audio driver today...
21:58:13dan_a_linuxstb: There's a small community of nomads wandering the desert who said they couldn't start until Monday
21:58:23 Quit jhulst (Remote closed the connection)
21:59:11linuxstbdan_a: hehe. Hopefully they'll be too late.
22:00:18linuxstbCan the Sansa also record?
22:01:01dan_a_But not yet
22:01:19nlsmcscruff: do you have a .rockbox directory in the root of your ipod and in it a few folders called codecs, rocks etc?
22:03:17*linuxstb sees the 8GB e280 with FM is available for $185 USD on ebay
22:03:34*preglow wonders what the dollar is worth these days
22:03:45linuxstbAbout 3 sheckles
22:03:52preglow6.2 nok
22:03:54preglowvery nice
22:04:18preglownot much at all, that
22:04:43 Quit efyx (Read error: 110 (Connection timed out))
22:04:53nlsthe 8gb sansa with radio is a lot cheaper than the 8gb nano here :-)
22:05:07 Join efyx [0] (
22:05:07linuxstbAnd expandable with MicroSD, so 10GB of flash...
22:05:11preglowbut i don't think i want another pp dap
22:05:31Bagder"unbrick" of course being a bit of a contradiction
22:05:57nlsYou can only unbrick players that aren't really bricked :-P
22:06:19preglowintroducing the Apple Lazarus
22:07:35Soap$185 is a fine price for a 8GB sansa in good condition. $200 is my "benchmark" price.
22:08:15linuxstbThere seem to be a whole load of them at that price, new.
22:08:28linuxstb(and worldwide shipping)
22:08:34Soapmaybe it is time to update my watch-price list
22:08:54Soaphow much is shipping?
22:10:11linuxstbTo the UK is $30 -
22:10:43jhMikeSpreglow: did you find swapping phase helped arm too?
22:10:50preglowjhMikeS: sure did
22:11:26Soapdang - my price list is out of date. I'm seeing them for 160-180 shipped to me.
22:12:15*jhMikeS hopes the audio hardware delay can be move to a single function for all chips
22:12:28preglowjhMikeS: i think i've got a bug in your upsampler, btw
22:12:48jhMikeSpreglow: do tell
22:13:24preglowjhMikeS: line 173 dsp_cf.S, shouldn't that be cmp.l %d2, not %d3?
22:13:49jhMikeShold on...I've got the modded one open...
22:14:46 Join jhoney [0] (n=jhoney@
22:14:57 Join barrywardell [0] (
22:15:19jhMikeSpreglow: oops
22:15:34mcscruffnls: i was putting it on my ext2 partition , not my ipod one :P
22:15:53preglowjhMikeS: not that it's ever likely to trigger anyway...
22:15:58*Bagder gets sansa support questions from people not even using Rockbox...
22:15:59jhMikeSwait a second
22:16:04nlsah, glad it worked out now :-)
22:16:16mcscruffme too, i just saw the themes!!!!!
22:16:18linuxstbBagder: That was Sandisk's evil plan...
22:16:34mcscruffbut i need to download a build with album art or something
22:16:36Bagderthey sure fooled me
22:16:58jhMikeSit should be right no matter
22:17:11nlsBagder: the "Do Not Email Daniel" section on the unbrick page is nice! :-)
22:17:33BagderI'm quite fed up with the constant stream of "help!" mails I get
22:17:58jhMikeSit would only affect a single frame after downsampling
22:18:57preglowjhMikeS: that's true, didn't think about combining the two
22:19:22mcscruffwho's builds are best and most up to date with the album art and margins?
22:19:41preglowjhMikeS: running a frame of upsampling after a frame of downsampling
22:20:18mattzzBagder: nice and helpful page. What do you think about md5sums for the mi4 images on your page?
22:20:30jhMikeSwill happen if downsampling then switching immediately to upsampling
22:20:48Bagdermattzz: do you really think that matters much?
22:21:06BagderI had md5sums there in the past actually
22:21:21mattzzBagder: I personally have a better feeling with md5sums whenever I download something like a firmware image
22:21:50Bagderyou're right, I'll add md5sums
22:21:55mattzzBagder: thanks
22:23:20 Join XavierGr [0] (
22:24:21jhMikeSpreglow: another thing I'm starting to prefer is numbered labels: 1: .... bge.b 1b
22:24:25 Join pabs_ [0] (
22:24:36 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
22:24:39preglowjhMikeS: yeah, started prefering them a while ago myself
22:25:27 Quit amigan (Remote closed the connection)
22:25:36preglowno reason to come up with silly label names
22:27:00*linuxstb reads the Rockbox wikipedia article and is impressed by how up to date it is
22:27:15preglowyeah, we've got someone working on that somewhere
22:28:23mcscruffso rockbox can play gameboy colour games?
22:29:13nlsmcscruff: yes
22:29:17Soapwith varied degrees of success
22:29:18 Join amigan [0] (i=dcp1990@unaffiliated/amigan)
22:29:22mcscruffso why did i install ipod linux!!!!!
22:29:50DataGhostbecause it's superior
22:29:52*DataGhost runs
22:30:07mcscruffif mpd gets better it will be
22:30:10scorchethey really cant be compared....
22:30:32mcscruffomg im torn between the 2, they can both stay for now
22:30:35DataGhoststarting all the way from the source code even :)
22:30:39nlshow can it be superior when it doesn't even support my player? :-P
22:30:50scorchewell, they have 2 completely different goals ;)
22:30:59DataGhostbecause it supports the 80GB >:)
22:31:10linuxstbIPL has a goal? :)
22:31:22DataGhostdoes it? or are you being sarcastic?
22:31:24scorchelast i heard it was for specific testers only and no one else had access to it =P
22:31:33Soapyes it does. A website with 60% uptime.
22:31:33scorcheor is that just what you tell the people in IRC? =P
22:31:38Soapand they are getting close.
22:31:40DataGhostheh scorche
22:31:50DataGhostI think it was linked on almost 40 blogs
22:31:56DataGhostiPL supports the 80GB since 6-3
22:32:04DataGhostfor the public time to check the vorbis decoder from and see if mine is somehow not compiling right
22:32:12scorchemeh...i dont attempt to find out much about ipl
22:32:21linuxstbDataGhost: I just mean that IPL doesn't seem very focused as a project, whereas Rockbox has a clear focus on the best possible audio playback.
22:32:21scorchedoesnt interest me ;)
22:32:34DataGhostoh... ah well
22:32:41 Join darkless [0] (
22:32:58DataGhostit's actually just focused on 'running Linux on an iPod' to my knowledge
22:33:02DataGhostand it's doing a pretty good job :P
22:33:16mcscruffwhats the difference between
22:33:45mcscruffwhat is it
22:34:33DataGhostit does look awfully similar to the word 'processor' :o
22:35:22mcscruffbut how do i know which one to use
22:35:47 Join |Rincewind| [0] (
22:36:02linuxstbmcscruff: They're both over a month out of date - a lot of changes have happened in the last month.
22:36:05DataGhostI don't know, it's probably explained in a wiki somewhere... last time I saw mention of 'cop' in the RB source, it looked quite experimental to me
22:36:12linuxstbSorry, almost a month...
22:36:28DataGhostoh yeah and I missed those numbers being a date :P
22:36:49 Quit Sanitarium (Read error: 110 (Connection timed out))
22:36:54linuxstbCo-processor support is now in the official Rockbox builds, as well as a completely redesigned menu layout.
22:37:07 Join Sanitarium [0] (
22:37:25Presencehurray for co-processor support! :)
22:37:36Soapmcscruff: all of the iPod unsupported builds with Album Art and such are woefully out of date.
22:38:09linuxstbDataGhost: Are you still working on 80GB support for Rockbox?
22:38:22SoapMay I strongly suggest you become familiar with Rockbox on a stock "current" build, as much of the documentation and help you have avaliable to you assumes you are running the latest version./
22:38:31DataGhostlinuxstb no not really
22:38:41DataGhostlinusN is working at that, right?
22:38:59DataGhostI got to the point where it's 'working', it can read my FAT32 partition
22:39:08Soapmcscruff: Those builds are called "unsupported" for a reason - all help queries regarding them must be relegated to the fourm thread from which you got them.
22:39:09DataGhostonly 2 folders deep, after that it'll show garbage and crash
22:39:17DataGhostit can't find the .rockbox folder and after some attempts it'll crash
22:39:25DataGhostand when you don't do anything fancy with it, it'll crash anyway
22:39:28 Quit pabs (Read error: 110 (Connection timed out))
22:39:32mcscruffSoap, i'll just get one coz im too lazy to patch my own :P
22:41:09DataGhostas far as I can see, there's "just a buffer problem" which doesn't look that easy to fix
22:41:52DataGhostanyway I shared my 80GB knowledge and iPL patch with LinusN, hoping it'll be useful
22:42:02 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
22:42:07DataGhostbut I guess he'll finish 80GB support before I do
22:42:23DataGhostand I have to study for some exams in a few weeks :) so I don't really have the time either
22:46:24 Quit barrywardell ()
22:46:29 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:46:42preglowfind the time!
22:46:48preglowexams can wait too
22:49:15jhMikeSok, disabling priority_x has no effect :\ and it's the only change to the scheduling loop
22:49:59 Quit x1jmp (Read error: 110 (Connection timed out))
22:52:30preglowjhMikeS: what about the vorbis boosting conondrum, btw?
22:52:36 Join juxtap [0] (
22:52:57preglowor is that what you're talking about?
22:53:03jhMikeSwhat about it? I was getting 19% now I'm seeing 78% on the same file...vorbis only
22:53:16preglowi tried again, behaved fine
22:53:16jhMikeSit is running slow for some reason
22:53:55 Nick pabs_ is now known as pabs (
22:54:20jhMikeSdeleted my config.cfg, turned off the new scheduler change...ahhh
22:55:02jhMikeSIs this "conondrum" something others have seem?
22:55:05 Join skyfaller [0] (n=skyfalle@wikipedia/Skyfaller)
22:55:16preglowhaven't heard anything
22:56:54 Part K3nto
22:59:00nlsjhMikeS: with the latest current build vorbis boosts normally on my h320
22:59:01jhMikeSI mean it's clearly running very slow
22:59:19nlsabout 20% wit q6
22:59:30jhMikeSnls: that's what's giving me seems unique to my player
22:59:36jhMikeSsound about right
23:00:00jhMikeSvoice file present?
23:00:13jhMikeSwill check that out then
23:04:57 Quit amiconn (Read error: 110 (Connection timed out))
23:05:29 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
23:05:30jhMikeSgetting the voice file out of the way cures it...and the slowdown was real and visible...
23:05:30preglowno voice file and boost is back?
23:05:36preglowto normal, that is
23:05:44jhMikeSvorbis turns into a pig when a voice file is loaded
23:05:54jhMikeSok now
23:05:59 Part skyfaller
23:06:07amiconnpreglow: How should I know whether the pp does burst transfers, without datasheet?
23:06:52preglowamiconn: just asking, you've written assembler routines for it and always do a thorough job, so thought you might have benchmD[D[Darked
23:06:56 Quit lee-qid ("aufwiederbyebientotsayonara")
23:07:06 Quit JETC- (Read error: 110 (Connection timed out))
23:07:09preglowdon't ask me what happened in "benchmarked" there
23:08:27mcscruffim compiling my own :) , what patch's would you say are good to have?
23:08:45amiconnjhMikeS: Perhaps that has to do with the voice thread always in running state in current svn?
23:09:13 Quit Galois ("Leaving")
23:09:17amiconnI first observed it on ipod and thought it might have to do with dual core operation, then found it also happens on coldfire targets
23:10:05 Join kubiix [0] (
23:10:11jhMikeSamiconn: why should it be and only affect a single codec? the mutex should sleep since nothing is different. :\
23:10:26 Quit kubiix (Client Quit)
23:10:27amiconnDunno, different yield frequency perhaps?
23:10:50Soapmcscruff: 5495
23:10:51***Saving seen data "./dancer.seen"
23:11:28Soapmaybe 6616 if that is your thing.
23:12:49mcscruffisnt 5495 gonna be in the svn?
23:12:50amiconnI also suspect the fact that voice runs on core 0 and main codec on core 1 to be the cause for the practically unchanged performance with cop support enabled on pp. While both threads are running on different cores, they cannot run in parallel because of the buffer swapping
23:13:16 Quit |Rincewind| ("Cya")
23:13:28 Join terry_ [0] (
23:13:46Soapdid it get commited while I wasn't looking?
23:13:48preglowamiconn: well, shouldn't people with no voice files see an improvement?
23:14:04amiconnpreglow: Yes they should
23:14:09preglowwe don't :/
23:14:12amiconnI never tested without voice file
23:14:37Soapnope, mcscruff, it is still open.
23:15:16amiconnBut something is definitely fishy wrt voice thread
23:15:21*linuxstb sees a fixed-point patch for the ffmpeg RealAudio (cook) decoder..
23:15:34 Join Galois [0] (
23:16:08linuxstbApparently the patch gave a 25x speedup on ARM...
23:16:21preglowcompared with what, fpu emulation?
23:16:48linuxstbI assume so, let me
23:17:03preglowsounds like a small speedup factor, then
23:17:05mcscruffSoap: man that was
23:17:20 Quit terry_ (Client Quit)
23:17:31mcscruffhit paste :P, was gonna say 3045 is a must have
23:18:04jhMikeSthe same yield should be executed now as before ... Will actually change the mutex sleep to be identical to before but I don't expect any change really.
23:18:50 Join barrywardell [0] (
23:19:21linuxstbpreglow: Here's the original post -
23:19:38dan_a_I'd like to move the voice thread to the COP, but run out of IRAM - I need some help to write the linker script to use uncached DRAM
23:19:43preglowoff topic: would anyone happen to know a generic way to deliver mails to specific mbox decided by the to: address? procmail? .forward?
23:20:06 Join x1jmp [0] (
23:20:22Bagderpreglow: procmail
23:20:56preglowi see the server runs exim and dovecot, does any of that matter at all?
23:20:59preglowah, ok
23:21:19preglowi had a mail server myself a couple of years ago, but i've forgotten practically everything, i see
23:21:24Bagdernormally you can invoke procmail from your forward unless it runs it by itself
23:21:38 Join rotator [0] (n=e@rockbox/developer/rotator)
23:21:57*Bagder has a >400 lines procmail file
23:22:32Kasperlethe host i receive mail on uses courier and maildrop. maildrop filters are pretty nice, too
23:23:23preglowlinuxstb: i don't like the part in the reply about people not maintaining fixed point code
23:23:44 Join pondlife [0] (
23:23:59preglowsounds pretty bad to allow people to break fixed point code while updating float code
23:24:10preglowespecially in these days of embedded multimedia
23:25:53jhMikeSidentical code to before and the problem persists with a voice file present
23:26:00linuxstbpreglow: Yes, but that's just one person's reply - the rest of the thread seems reasonable.
23:26:06preglowlinuxstb: yeah, see so
23:26:40*linuxstb needs to invest more time understanding ffmpeg's architecture
23:27:14 Quit ompaul ("night all")
23:27:22preglowyeah, i have the tree checked out here myself with the intent of doing so
23:27:41preglowbut i guess i'll never get around to it
23:28:20BigMacWhat timezone are flyspray times based off of
23:29:33amiconnOur runtime estimation on X5 is _way_ off
23:30:09amiconn>13 hours now and still going. Estimation is only 8 hours iirc
23:30:13jhMikeSI revert everything to the old system and it persists :\
23:30:16linuxstbBigMac: I don't know, but most Rockbox things are on Central European Time (GMT+1)
23:30:35BigMacah ok thank ya linuxstb
23:30:37mcscruffi ran the sh file that buils the gcc-elf thing but whats the command to make that in my path for make?
23:30:42jhMikeSExcept disk access goes back to slowing down everything
23:30:42amiconnjhMikeS: In such cases a binary search through the dailies might help
23:31:01amiconnIf you found the day, then continue with checking this day's commits
23:31:03jhMikeScould be
23:31:25*amiconn did that several times
23:31:45jhMikeSIt could have been there already and I just discovered it...can't say
23:32:22jhMikeSA somewhat tedious undertaking though .... argh
23:33:25preglowtedious and conclusive...
23:34:40amiconnYou should be down to the specific day with 6 checks...
23:35:09amiconn(checking the full month back)
23:35:17jhMikeSstill I find that tedious when I want to work on new stuff :)
23:36:11*jhMikeS will just shut up and get this over with
23:37:32pondlifeMy hint: Keep a notepad handy with your binary search points noted.
23:37:35 Join K3nto [0] (
23:37:48pondlifeIt gets even more tedious if you lose track :)
23:38:19jhMikeSYes...I usually do that
23:38:31K3ntohey i have a slightly off-topic question
23:38:36K3ntohey guys, i have a set of stereo headphones im trying to fix. in the middle of the cord, there was a volume adjustment dial. the dial was busted, and now im trying to fix the headphones. There was a copper wire per strand and a green/red one in each. I removed the enamel and twirled them and finally, soldered them
23:38:49K3ntoi out it in another chan too tho
23:38:52jhMikeSstill hate binary searching like this >:P
23:39:02scorcheK3nto: and what does that have to do with rockbox?
23:39:34*amiconn thinks adding new features without having a stable base is worthless, and not much fun either
23:40:14K3nto*sigh. nothing.
23:41:55KasperlejhMikeS: heh. i once did that searching through a couple of months worth of commits to the full FreeBSD source tree just to find out that somebody flipped the default value of a sysctl :/
23:42:38preglowK3nto: is this a success story or is there a question buried in there?
23:44:32jhMikeSI can say this...the 2007-02-08 build has it
23:44:43preglowperhaps it's always been there, then?
23:44:50 Join chromoXdor [0] (
23:45:26jhMikeSI have a feeling about that ... wonder why though. You sure nothing in mpa.codec would clobber the malloc buffer?
23:45:30 Quit efyx (Remote closed the connection)
23:45:54preglowjhMikeS: why would clobbering the malloc buffer slow stuff down? i don't even know where the malloc buffer is located these days
23:46:09preglowis it still a separate 512 kb buffer?
23:46:21K3ntopreglow: nvm, im in a dif chan with this
23:46:24jhMikeSit is not swapped like the other buffers
23:46:43preglowjhMikeS: fair enough, but why would clobbering slow stuff down?
23:46:53jhMikeSI have no friggin' idea :)
23:46:54preglowi would expect horrible distortion, sure, but not slowing down
23:46:57jhMikeSme too
23:47:15preglowthe chances of slowdown being an effect from a clobbered malloc buf are really minute
23:47:17jhMikeSjust reaching for anything for an answer atm
23:48:34mcscruffwhen you compile rockbox do you get alot of warnings?
23:48:37jhMikeSstarting ridulous usually leads to getting somewhere
23:48:49jhMikeSmcscruff: never
23:48:55 Join Soap_ [0] (n=Soap@rockbox/staff/soap)
23:49:03scorchemcscruff: you can see the warnings and errors on the build page in the chart
23:49:20 Join RogerBacon [0] (
23:49:23mcscruffit didnt error out :S is that good or bad :P
23:49:33RogerBacondid anyone know if the creative Zen M have a disc mode ?
23:51:06preglowif it hasn't got a rockbox port, we probably don't know
23:51:19 Quit Soap (Nick collision from services.)
23:51:24 Nick Soap_ is now known as Soap (n=Soap@rockbox/staff/soap)
23:51:48RogerBaconany trick to put it in UMS ?
23:51:53RogerBaconthat's really suck
23:52:02RogerBaconmtp is really a piece of crap
23:53:42linuxstbAccording to this page, it has a "configurable MSC partition (512MB to 16GB)"...
23:53:42preglowbeer time!
23:53:46jhMikeSalingnment? issue?
23:53:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
23:54:08RogerBaconthis device need WMP10 ONLY, im with WMP11, so if i understand,microsoft want to use old software ...
23:54:31 Quit desowin ("use linux")
23:55:10z0dein brickmainia the game if you pause it while you have the glue power up and then unpause the ball stops moving
23:55:46z0dealso the random folder plugin does not start a new folder while playing brickmanina
23:56:03z0deand theres no way to access music control without losing your game
23:56:16z0deit also needs more levels
23:57:53SoapOn the second point, you kind of answered yourself. Snake won't run at the same time brickmania does either. :)

Previous day | Next day