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 2005-08-16

00:01:16 Join hshah [0] (
00:02:25hshahwho here has a machine without DevKit and cygwin on?
00:02:48BagderI do
00:02:53Bagderbut it runs Linux ;-)
00:03:09hshahi need a windows one :p
00:03:48hshahi need someone to test out all the steps at
00:06:55hshahwhere is everyone hiding 2nite?
00:14:59*Bagder points to a dusty corner of the channel
00:15:07Bagderthey're all there
00:15:23hshahwell if someone reads this while I am gone (i am going now) please test it for me and email me at rockbox [at] hirenshah [dot] co [dot] uk
00:21:05 Join webguest90 [0] (
00:21:10webguest90I'll give it a shot
00:22:06 Nick webguest90 is now known as rasher (
00:22:39rasherHas anyone else gotten "I41: PDIR2FULL" while playing?
00:22:53HCli have, actually
00:23:02HClit happens in a dir with over 700 songs
00:23:08HCli've been meaning to get that fixed someday
00:23:23HClnot saying in general
00:23:27HCljust my dir has over 700 songs
00:23:30HCland it happens in there
00:23:38rasherI'll see about mine
00:23:53hshahaha people
00:24:02hshahcan someone help me test out the steps at
00:24:15rasherYeah, I'm booting into windows as we speak
00:24:22hshahsomeone who hasn't got devkit or cygwin installed
00:24:23rashertaking its sweet time
00:24:32hshahthanks rasher :)
00:24:55hshahbah... its 11.30 and ive got to get up at 6 for work tomorrow
00:24:58rasherI've uploaded a fresh matrix.rock by the way
00:25:17hshahyeah i got ur email... and replied to it
00:25:21rasherit's still mysteriously broken
00:25:29rasherAh, so you did.
00:25:36hshahomg... i didn't realise it was u until u mentioned matrix.rock
00:25:51rasherThe bug is completely crazy. It's not there in the simulator
00:25:58rasherwhich is just wrong with a capital w
00:26:14hshahyou mean "Wrong"
00:26:30rasherThat I do
00:28:55*rasher ponders creating a grayscale matrix font
00:29:00rasherwould be nice
00:29:12hshah <−−- that is fastest growing thread so far :)
00:29:24rasheranother thing that'd be nice is getting the damn bug fixed
00:30:22hshahwhich bug... the matrix one?
00:31:21hshahi would love the matrox screen saver to come on 5 mins after no buttons have been touched...
00:31:44rasherScreen savers on a battery driven device is generally a silly idea
00:32:02hshahhmm... actually u have a point there...
00:32:10hshahbut it still would be cool to show people
00:32:15rasherBut yeah, seeing as I go days between charges
00:32:36rasherit would be cool
00:33:06*t0mas is back
00:33:23t0mashow would you want to modify it?
00:33:25hshahsame... i have never actually drained the battery during use... i accidently left it plugged into my pc once... and the battery died then... other than that... ive never drained it
00:33:39t0masIt does already use the dimming code in place
00:33:48t0masbut that isn't working the way I want it...
00:34:32hshahwhat u talking about t0mas?
00:36:52t0masmy patch
00:36:58 Quit bagawk ("Leaving")
00:37:20t0masnamed "light organ" patch by Slasher
00:37:29t0masbeat detectioni...
00:37:42t0masand flashing the backlight according to it :)
00:37:51hshahreally... WOW
00:37:52t0mas(eats cpu power)
00:37:59t0masbut it looks cool
00:38:03hshahi would still love to play around with it
00:38:13rashernow if this could somehow be exported to plugins..
00:38:13hshahi would sit in my room in the dark and have fun
00:38:16HClhows radio working nowadays?
00:38:32t0masrasher: what? bass peak data?
00:38:35rasherIt works. No presets due to missing keybindings
00:38:35t0masthat's easy....
00:38:43rashert0mas: yup
00:38:50t0masno problem...
00:38:52hshahrasher - im off to bed, but can u please email me after u test the guide
00:39:05rashernot really expecting any problems
00:39:07hshahcya all
00:39:16rasherit looks pretty sound
00:41:11hshahsome people had problems earlier
00:41:19hshahso i just want to make sure it works first time round
00:41:25hshahi can't really test it on mine, coz i have parts of cygwin etc etc
00:42:39rasheryou should probably use fixed-width for commands (=foo= in WikiMarkup)
00:44:10rasherinstead of italics
00:44:46rasherI'll show you what I mean, hang on
00:45:27rasheror not
00:45:36rasherGHOD is cygwin ever slow
00:46:25rasherthere we go
00:46:25hshahive changed them all
00:46:34hshahused = = for commands
00:46:37rasherI have, unsurprisingly, a now
00:47:49rasherBy the way, to avoid those invalid wiki links that happen for words that look like WikiNames, put <nop> in front of them
00:48:26rasherWeird, I don't see fixed-width commands
00:48:40rasherI don't even see an update
00:49:10hshahcoz i aint saved it yet :p
00:49:40rasherMakes sense.
00:49:47rasherThe guide worked
00:50:46rasherMaybe you could split up the "compiling the source code" into smaller groups ("configure", "build" or something)
00:50:50rasherLooks more readable to me
00:51:00rasherand more obvious that it's something to type
00:51:02 Join zezayer [0] (n=jake@
00:51:30hshahok... will work on that tomorrow
00:51:42hshahbed time now
00:51:55hshahi need more than 6 hours sleep :p
00:52:45Moosgood job hshah have a good night
00:53:02rasherI might try and change it a little
00:53:05***Saving seen data "./dancer.seen"
00:53:24rasherhshah: are you editing, or just forgot to release the lock?
00:53:46hshahi think i forgot
00:54:19hshahahh fuck this stupid machine...
00:55:34hshahfor some reason the page refuses to load... i can't unlock it :
00:55:52hshahoff to bed now... sorry rasher... will try and unlock it in the morning
00:56:16rasherit'll happen in 28 min. no biggie
00:56:32rasherless, now
01:05:14 Join cooleo [0] (
01:08:11zezayerhi cooleo
01:09:03tvelocityanyone know of a linux tool for declipping?
01:12:25cooleooh hey zezayer
01:12:33cooleoi'm guessing you have an iRiver
01:13:13cooleoi'm banging about here, H300 owner, hopeful of Rockbox but I don't wanna pressure anyone round here :D
01:13:41t0masLinus will get the bootloader done soon...
01:13:50cooleoi realised how annoying it was when I went on the bloody forums
01:13:52t0masthen we'll fixup some error's left
01:13:59cooleoits just people wanting lots of stuff
01:14:12t0masand then it can boot.... and we'll have a lot of new work todo
01:14:12cooleothats cool t0mas
01:14:32t0masafter the bootloader, booting isn't a problem
01:14:35cooleoi can't believe you guys do this stuff for free
01:14:37cooleoits awesome
01:14:41zezayeri remember a few months ago waiting for linus to make the h140 bootloader, was well worth the wait tho :)
01:14:43t0masand I think audio playback can be done with the H1x0 code
01:14:52cooleodo you get any of the donations?
01:15:11t0masit's used for the domain
01:15:12zezayerlots of the hardware is the same
01:15:17t0masand the BDM for linus
01:15:28rasherwould be fun to know the economics of the donations
01:15:29zezayerscreen will be the hardest but
01:15:42rasherAustriancoder bought a BDM as well
01:15:45t0mashi priced hardware to make the bootloader with
01:15:48rasheralso from the "Rockbox fund"
01:15:52cooleohave they still not confirmed the screen?
01:16:02t0masrasher: sure?
01:16:15rashercooleo: I think it's 99% sure
01:16:22Moosof course it's normal :)
01:16:22rashert0mas: pretty sure, not entirely
01:16:42cooleocheers rasher
01:17:09cooleoi was on a chase to see if I could get power from the screen, in parallel for some LEDS
01:17:19cooleoturned out to be incredibly hard! :D
01:17:24t0masyou can use the backlight pins
01:17:35t0masget the datasheet.... and use that pins?
01:17:45cooleoi'll have a look
01:17:46t0masor just lookup some VCC line?
01:17:54Moosgood night guys
01:17:55t0masthen you have a nice 5v
01:18:06 Part Moos
01:18:11cooleowhats a VCC line?
01:18:32t0masVCC is the "name" of the "input power" in most circuits
01:19:02t0masyou can check the datasheet of the LCD for example
01:19:05cooleowould that be on the diagram?
01:19:09t0massome pin will be named VCC
01:19:15cooleokk cheers
01:19:18t0masyes, it should be
01:19:23t0massome other will be named GND
01:19:42cooleook, well i'll try not to ruin my H320 heheh
01:19:57t0masok, try it with a multimeter first
01:20:06cooleothe HDDs got a bad rattle now, so I'm not that bothered
01:20:16t0masah, you can replace it...
01:20:17rasherDisassembling the h320 firmware seemed to reveal the LCD commands matching the ones in the datasheet for the one found
01:20:25rasherI believe.
01:20:27cooleoyeh i will replace the HDD
01:20:36t0masrasher: you know what controller the LCD has?
01:20:43rasherOr maybe that was the x5 firmware
01:21:10rashert0mas: Isn't it in the h3xxhardware.. something wikipage?
01:22:12t0masfound it
01:22:21t0masif that's true it isn't hard to get it working... afaik
01:23:13rasherIndeed, seems like the major problem is the bootloader
01:23:40t0masbut until now I've only worked with the HD44780 simple text lcd controller
01:23:44t0masthat was really easy...
01:24:06cooleothe colour screen is gonna be fun
01:24:16cooleowork for you guys
01:24:23t0masI don't think that's really hard...
01:24:26 Quit ashridah ("Leaving")
01:24:28rasherMost of the work is done, afaik
01:24:42rasherTo support colour
01:24:43t0masif it works... there are just extra color bits in the display buffer
01:24:54cooleoyeh i saw it on the builds thingy
01:25:03t0masamiconn has already done a lot for the 4 grayscale thing we have now
01:25:06cooleoadded more support for colour
01:25:09t0masit's just that, but with more color bits
01:25:28 Quit ender` (Connection timed out)
01:25:42cooleoapparently Jens did this: More preparations and conversions for colour LCD support.
01:25:59cooleowhich sounds good to me
01:26:05t0masJens == amiconn ;-)
01:26:22cooleoLinus sounded very negative about AVI support
01:26:28cooleothats upsetting
01:26:39t0masAVI the video format?
01:26:43cooleoi think everyone was expecting movies on the H300
01:26:48cooleoyeh the vide
01:26:50rasherI don't see the point in using avi
01:26:55t0maswell... movies might be possible
01:27:02t0masbot the avi format is a hard format to implement
01:27:05rasherYou're not going to be able to play ordinary files anyway
01:27:07cooleooh, i see
01:27:14t0masit can contain almost any type of video...
01:27:16rasherYou're going to be transcoding, might as well do it into something simpler
01:27:18cooleoso it will play movies?
01:27:28cooleothats good news
01:27:49cooleoi suppose i was just jumping to silly conclusions
01:27:56t0masan avi file can contain 1000s of codecs
01:28:09t0masso you would have to write a fast enough divx, xvid and may other codecs
01:28:12t0masthat would be mad...
01:28:20cooleoiRiver did it somehow
01:28:26cooleoonly 10 fps though
01:28:28rasherWell you could just only support xvid-in-avi
01:28:28t0masand I don't think we can get it to realtime
01:28:44t0masthen still... iRiver didn't get it to realtime
01:28:49rasherI don't think we could do xvid better than 10fps
01:28:52t0masI don't think we will
01:29:03cooleoi don't think it would be worth the effort
01:29:15t0mastranscoding to rockbox format
01:29:22cooleogod knows why they didnt do something else
01:29:25t0masthat's the way to go
01:29:38cooleolike their own format with a piece of converting software
01:29:54t0masthat would be better...
01:29:58cooleoactually i bet they just stole it from the PMP players
01:30:00t0masbut it's not good for your commercials...
01:30:02cooleowhich play AVIs
01:30:03rasherwell, the rockbox format isn't compressed
01:30:21t0masrasher: true... we would need a new format for large video's
01:30:24rasherrockbox video on h320 would be huuuuge
01:30:42cooleoisn't it all worked out for the Archos players?
01:30:49cooleoi thought they played video?
01:31:21rasherThey do
01:31:33cooleoare the files massive on them?
01:31:40rasherNot too bad
01:31:44cooleohow would it be different on the H300
01:31:50rasherMore colours, larger screen
01:31:55cooleoif Rockbox is kinda universal
01:32:38rasherI suppose you could watch 34-greys 128x64 movies on the h320, but that'd be silly
01:33:09cooleoso just speculating here
01:33:18cooleowhat do you expect the fps to be like?
01:33:29t0mas> 20
01:33:32t0masif we do it right
01:33:35cooleoif you don't know then thats fine
01:33:39t0masbut that would cost a lot of diskspace
01:33:42rasherIt's 67fps on archos
01:33:56rasherBecause that's the refresh rate of the lcd
01:34:15t0masbut... with audio... in full color... 20 would be a good guess
01:34:17t0masmaybe 30
01:34:20cooleosurprisingly 10fps looks good on the H300 little screen
01:34:29cooleoso anything more would be great
01:34:29t0masbut not really much higher (wouldn't be any use)
01:34:41cooleonah 20 would be fine i'm sure
01:34:50cooleoaren't movies only 30fps?
01:35:03cooleoor is that something else that i've got confuse
01:35:04t0masdepending on the format
01:35:14cooleooh no, im not that mad :D
01:35:15rasherThat or 29,97 or something silly like that
01:35:25t0masNTSC is 29,97
01:35:30t0masPAL is less
01:35:34rasherwow, I remembered
01:35:34t0masand a normal number
01:35:41t0masNTSC is 29,9753.....
01:35:53rasherMan, ntsc is a riot
01:36:18t0masit's sometimes 23,97....
01:36:25t0masPAL is just 25
01:36:39cooleowe europeans are sensible
01:36:41t0masno weird numbers
01:36:59t0masNTSC = Never The Same Color :P
01:37:11t0masstupid system... weird framerate...
01:37:20cooleoPAL = Perfect and lovely
01:37:24rasherOf course the yanks would be using it!
01:39:02cooleowhats your specialties then?
01:39:20cooleodo you do specific parts of the work?
01:39:29cooleoor just anything that needs doing?
01:41:14t0masI'm not "the" programmer around here
01:41:24t0masI just do things I think I can do... and I like
01:41:37rasherI don't really do much. I'm not much of a coder. I try and accept simple patches and discuss higher level things
01:41:39t0masthe wps graphics was something I wanted to have personally... and it was doable
01:42:05t0masbut the codecs were to difficult for me
01:42:46cooleoit seems like Jens is the graphics expert
01:42:55rasherI try helping people so the ones with talents won't have to spend their time
01:43:09cooleothats what i'd do
01:43:14cooleoif i could even do it
01:44:00t0masJens is the graphics expert yes... he could've done the graphics wps too
01:44:02t0masif he had time
01:44:12cooleoah ok
01:44:34cooleoi love the custom WPS thing
01:44:34 Quit edx ()
01:44:37cooleoit really rules
01:45:03rasherI'm still hoping for someone to implement a more powerful conditional thing
01:45:23 Join Sucka [0] (
01:45:26cooleowhat will that do?
01:45:29rasherinstead of the simple %? in place now
01:45:54t0massomething like %if(codec = MP3)< something >
01:46:13rasherExcept I'd think a case-like thing would be more useful
01:46:53t0masI've been discussing a good syntax with someone from here
01:47:06t0masthe one who wrote the first conditional image hack
01:48:08t0masDave Wiard
01:48:15rasherHow about "%#xx<foo:bar|baz:qux|quux>", where xx is the field, and foo and bar are the values
01:48:24rasherquux is the "default case"
01:48:52t0mashe was thinking about a more readable syntax
01:49:09cooleolike real words! ; )
01:49:16rasherThat's a lost case on WPS already
01:49:21t0maslike real scripting languages
01:49:42 Part zezayer
01:49:51t0masbut it's time to go to bed
01:49:52rasherI was thinking simple to avoid having to extend the wps language too much
01:49:53t0masc u tomorrow
01:50:20cooleoya matey
01:50:56cooleoi should probably go as well
01:50:59cooleoits late
01:51:09cooleocya rasher
01:51:41 Quit cooleo ("CGI:IRC")
02:37:46 Quit hicks (Remote closed the connection)
02:37:47 Quit hardeep (Read error: 104 (Connection reset by peer))
02:53:08***Saving seen data "./dancer.seen"
02:59:35 Quit rasher ("CGI:IRC")
03:02:42 Join Craig__ [0] (
03:03:30 Join CyberDog [0] (i=UNKNOWN6@
03:08:51 Join bagawk [0] (
03:10:26 Join petelinux [0] (
03:10:48Craig__any committers out there care to take my patch for a spin?
03:10:58Craig__it's called "Warn on erasing modified dynamic playlist"
03:14:03 Quit elinenbe (" Want to be different? HydraIRC -> <-")
03:15:09 Nick petelinux is now known as petelinux2 (
03:15:19 Join petelinux [0] (n=petelinu@
03:15:43 Quit petelinux (Client Quit)
03:21:06 Join phaedrus96 [0] (
03:21:09 Quit phaedrus961 (Read error: 110 (Connection timed out))
03:21:26petelinux2can i change my server, example : ..?
03:22:35CyberDogwould it be possible to port rockbox to a z80 processor? (EG S1 Mp3 player)
03:28:51 Quit Craig__ ("Chatzilla [Firefox 1.0.6/20050716]")
03:31:18 Part petelinux2
03:48:16 Quit CyberDog ()
04:01:49 Quit Sucka ("a bird in the bush is worth two in your house")
04:51:48 Quit tvelocity ("Leaving")
04:53:10***Saving seen data "./dancer.seen"
05:08:56 Join solex_ [0] (
05:08:57 Quit phaedrus96 (Read error: 110 (Connection timed out))
05:14:32 Quit solex (Read error: 110 (Connection timed out))
05:14:47 Join phaedrus961 [0] (
05:25:50 Quit bagawk ("Leaving")
05:55:58 Quit crash_ (Read error: 110 (Connection timed out))
06:08:40 Join matsl [0] (
06:33:45 Quit hshah ("CGI:IRC (EOF)")
06:33:45 Quit t0mas ("CGI:IRC (EOF)")
06:53:12***Saving seen data "./dancer.seen"
06:59:31 Join courtc_ [0] (
06:59:58 Quit courtc (Read error: 110 (Connection timed out))
07:15:34 Join LinusN [0] (
07:20:45amiconnGood morning
07:22:06midkhooray! good morning rockboxers!
07:22:38LinusNyo d00dz
07:23:13LinusNmidk: i haven't seen you for quite a while
07:23:47midkLinusN, true.. i've not been around irc for the past few months. :)
07:24:24midkjust taking a break, doing other things.. etc. :)
07:25:04LinusNoh, i've heard of that...i believe it's called "have a life"
07:26:01midkoh, that! i've heard of it too.. unfortunately i didn't try anything that radical. :(
07:26:16LinusNyou scared me for a moment there
07:26:53midkit's okay.. i'm here now.. it's alright. if i ever try the elusive "life" thing, i'll be sure to let you all know and take a few people with me.. hooray! :)
07:30:40*amiconn is undusting his sh1 toolchain
07:31:04amiconnBuilding binutils 2.16 and gcc 3.3.6 under cygwin...
07:31:11midki should be doing such a thing, i haven't bloated clock.rock to the 32k limit yet. :)
07:32:14amiconnmidk: When rockbox starts working on iriver H3x0 get one, then try to fill plugin ram on it ;)
07:32:15LinusNmidk: you have 768k on the iriver :-)
07:32:25amiconnYou'll have 768KB to fill...
07:32:38amiconnLinusN: :)
07:32:39midkmaybe next i'll do fading or dissolving menus/screen transitions... or perhaps a method to design your own clock border!!! ;)
07:32:46midkhaha. hmm.
07:33:14midklet's see... desperate times call for desperate measures... :)
07:33:36midkmaybe a few selectable animated backgrounds based on some very complex algorithms. :)
07:33:54LinusNdamn, i just can't reproduce the AddrErr that so many people complain about
07:34:00midkincluding a physically accurate explosion that loops.. and perhaps some sort of plasma design.. hmm.. :)
07:35:38 Nick courtc_ is now known as courtc (
07:35:55 Quit matsl (Remote closed the connection)
07:36:36amiconnLinusN: I found that there already is gdb 6.3. Do you know whether newer gdb than 6.1.1 (the one recommended in the build instructions) will work for sh1?
07:37:59LinusNi believe i used 6.3 once
07:38:21LinusNtry it
07:38:53amiconnBah, compiling gcc 3.3.6 throws a heapload of warnings
07:43:13LinusNah, i finally made it crash
07:43:31*midk claps
07:44:25LinusN...but of course i couldn't repeat it :-(
07:46:35*midk boos and throws tomatoes
07:48:47*midk yells and throws rocks
07:49:11LinusNwho threw that stone?
07:49:20*midk hides
07:49:21LinusN"she, she"...."he, he"
08:04:49 Join webguest42 [0] (
08:08:40 Join ender` [0] (
08:10:10amiconnLinusN: Btw, what is the problem with the keymapping in the FM screen?
08:10:48amiconnI wouldn't think this would be hard, since it was possible to make it work on the Ondio, with significantly less buttons...
08:11:09LinusNyes, i could set it up to use it like the ondio
08:11:43LinusNthed problem is that i wanted to make use of the joystick puch, and that conflicts with the menu callback
08:12:04LinusNduplicate case values etc
08:12:15LinusNthe button mapping is a mess
08:12:47LinusNpossibly because i chose to use the menu code for the preset list
08:14:30LinusNbut i guess i could put the presets in the menu in the meantime
08:15:04LinusNbtw, i miss the radio simulation in the sims
08:15:15amiconnThe only thing not working on the Ondio is the screen freeze, which would be somewhat needed
08:15:34LinusNi think the peak meter should be removed
08:15:50amiconnThe radio chip is so close to the display that even the philips chip picks up some noise
08:16:03LinusNit generates interference on the archos and eats cpu on the iriver
08:16:20amiconnI'm thinking about a special slow-transfer mode in the lcd driver for use during radio operation
08:40:07LinusNah, fm radio simulation restored
08:41:02 Join hshah [0] (
08:42:31 Quit hshah (Client Quit)
08:53:16***Saving seen data "./dancer.seen"
09:02:19 Join einhirn [0] (
09:08:09 Join ashridah [0] (
09:08:39 Join t0mas [0] (
09:08:47t0masmorning :)
09:46:19 Quit webguest42 ("CGI:IRC")
09:47:35 Join crash_ [0] (
10:02:46 Join B4gder [0] (
10:14:30amiconnhi B(a|4)gder
10:19:48 Join webguest53 [0] (
10:21:00*B4gder is at work. Bagder is at home
10:23:15*B4gder goes for more coffee
10:23:33 Join hshah [0] (
10:24:36hshahhello all
10:24:39hshahwhose around then...
10:24:48midkof course not i.
10:24:53midki'm just a bot.
10:24:55midkthe grammar-bot.
10:25:17midki'm a grammar nazi, and you meant "who's" instead of "whose". :p
10:25:41midki'm not a punctuation bot, but you also neglected a question mark at the end of that sentence...
10:26:18B4gderaha, it is one of _those_ days..., haha.
10:32:18hshaht0mas around?
10:33:51amiconnLinusN: gdb 6.3 doesn't build under cygwin :(
10:34:32 Quit midk (Read error: 104 (Connection reset by peer))
10:34:39 Join midkay [0] (
10:34:52 Quit webguest53 ("CGI:IRC (Ping timeout)")
10:35:02 Nick midkay is now known as midk (
10:39:03amiconnLinusN: Strange error message is displayed:
10:39:05amiconnmake[2]: *** No rule to make target `../../intl/libintl.a', needed by `run'. Stop.
10:39:12LinusNso how would we like the preset handling to work on the iriver?
10:39:26LinusNamiconn: interesting
10:39:43 Join bobTHC [0] (n=bobTHC@
10:39:50amiconnI'll incrementally go backward
10:39:58hshahanyone know how to do this:
10:40:01LinusNdecrementally? :-)
10:40:42hshahwhen playing a track and on WPS, you press the joystick down... and it opens up the last folder u were in... i want it to go to the file that is currently being played
10:40:58LinusNhshah: "Follow playlist"
10:41:19LinusNin the playback options
10:42:51hshahahh yes... cool... one problem solved... thanks LinusN
10:43:00LinusNhow would we like to bring up the presets in the fm screen?
10:43:12LinusNi suggest joystick-hold
10:43:29hshahi agree with that
10:43:33amiconnI did not yet try fm on iriver, so I have no opinion yet
10:43:40hshahit should bring them up as a list
10:43:49LinusNbut how to add a preset?
10:43:50hshahif possible
10:43:55LinusNit does
10:44:15bobTHChi folks !!!!
10:44:17LinusNhold play?
10:44:29hshahsounds good to me
10:44:59bluebrother^how about a "add new" in the preset list?
10:45:02LinusNi guess having them in the menu doesn't hurt either
10:45:18LinusNbluebrother^: as the first entry?
10:45:37 Join AEnertia [0] (n=aenertia@
10:45:39bluebrother^or as the last so one could reach it by simply going "up"
10:45:40AEnertiaHi all
10:45:45hshahone more question... when viewing the files after pressing the joystick down, lets say you are in 2 folders deep and you press the joystick left to get to the root, i want to be able to press left once more to get to wps...
10:45:54AEnertiaI just found a good game if you want to minimise procrastination:
10:46:06LinusNhshah: yuck!
10:46:25hshahLinusN: why?
10:46:50LinusNjust a gut feeling
10:47:06LinusNAEnertia: what does that have to do with rockbox?
10:47:15AEnertiaNot much admitantly.
10:47:35AEnertiaI do however have a question regarding iriver port
10:47:50hshahLinusN: I am just so used to that from the iRiver firmware that i always try and press the joystick left and wonder why it doesn't work, and then remember I have to press the play button
10:48:00AEnertiaI'm sorry this probably came up in the mailing list and I missed it.
10:48:12AEnertiabut the h10's are no go right because of the chipset?
10:48:23LinusNAEnertia: pretty much
10:48:49AEnertiaI just ask because I notice that the h10's are selling like hotcakes and are very 'fashionable'
10:49:06LinusNhshah: and i would probably be annoyed if one extra left would take me away from the browser
10:49:13AEnertiaHows the h3xx stuff comming btw?
10:49:48LinusNAEnertia: i got a dead h320 yesterday, i will slaughter it tomorrow and start making schematics
10:50:07AEnertiaWhy did it die?
10:50:20LinusNi have prepared my other h320 for my bdm
10:50:30LinusNAEnertia: i dunno
10:50:42AEnertiaso it is possible it's not dead?
10:50:49hshahLinusN: would that modification be hard to do? mebbe someone could make a patch for it... i would but i can't do c at all :s
10:50:56LinusNit's alive except that it can't talk to the hard drive
10:51:02amiconnLinusN: Trying gdb 6.2.1 now...
10:51:03AEnertiagoing on the number of 'Dead' archos units that come back to life by themselves
10:51:35LinusNAEnertia: i bought it for the sole purpose of slaughtering it
10:51:52amiconnhshah: I agree with Linus here. Also I think we shouldn't care about how the iriver firmware does this or that
10:51:54AEnertiathat's good , hopefully it's just a dead 1.8inch drive ehh =-)
10:52:08amiconnFor me, navigation in the iriver firmware is pretty illogical
10:52:19hshahok then guys... ill just have to get used to pressing play :p
10:52:23LinusNAEnertia: no, it's not the hard drive, i tried with another one too, just out of curiosity
10:52:40amiconnI always have to find my way to firmware update (that's the only thing I use the iriver firmware for)
10:53:20***Saving seen data "./dancer.seen"
10:54:33mbrhi there
10:54:53mbrI have a question regarding the feature freeze
10:55:14mbrHas anyone had a look at my multi align patch?
10:55:18mbrany comments?
10:56:16mbrany chance for inclusion?
10:56:45hshahhshah here
10:56:53hshahi just sent a reply to ur email
10:57:05mbrso did i
10:57:10hshahand LinusN - this patch is highly recommended
10:57:16mbrI attaches an updated patch yesterday
10:57:19mbrdoes it work?
10:57:33hshahoh rite... lemme see
10:57:41mbrI'm not sure if this patch is too "hacky"
10:57:55LinusNmbr: i haven't tried it, but the code looks pretty straightforward (although somewhat complicated)
10:57:56hshahits a great patch... it makes my WPS so much tidier
10:58:11hshahi can vouch for it... it works a treat :)
10:58:22LinusNmbr: it could probably benefit from a comment or two
10:58:25mbrI'll upload an more current patch in a minute
10:58:57hshahmbr - according to the sourceforge page, u haven't updated the patch
10:59:03LinusNlike commenting what the code that merges the strings is there for
10:59:03hshahit doesn't work with the current build
10:59:11hshah(the patch file doesn't work i mean)
10:59:17hshahthere is nothing wrong with the code
10:59:34mbrOK so I add some more comments
10:59:53mbrhshah: I meant the patch I attached to the mail yesterday
11:00:06hshaherm... there was a patch in the mail yesterday?
11:00:10hshahi didn't even see it...
11:00:14hshahand its at home now
11:00:27hshahcould u email it again... and ill check my mail again
11:02:01mbrhshah: sent
11:03:11*LinusN goes to lunch
11:03:35hshahreceived.... testing now
11:04:33hshahyep works
11:04:35hshahpatches fine
11:04:41hshahgreat work mbr
11:05:17mbrOK, then I'll add more comments and upload a new one to the patch tracker.
11:13:35amiconnLinusN: gdb 6.2.1 fails too, but at a different point :(
11:18:23hshaht0mas was making a patch that flashes the back light with the beat of the music
11:18:27hshahthat would be SOOOO cool
11:19:24amiconnImho this isn't something to include in the core
11:20:04hshahi know... but i could have so much fun with it :)
11:20:14amiconnIt eats a considerable amount of CPU power, and hence battery runtime
11:20:21amiconnIt would be ok as a plugin
11:21:54 Join webguest53 [0] (
11:24:09hshahyeah :)
11:24:31hshahi would hide myself in a dark cupboard for this :)
11:29:19amiconnt0mas should have a look at tsr plugins for perfecting it then ;)
11:39:32 Quit courtc (Remote closed the connection)
11:40:00 Join courtc [0] (
11:45:06 Quit webguest53 ("CGI:IRC")
11:59:00 Join zezayer [0] (n=jake@
12:01:13 Quit AEnertia (Remote closed the connection)
12:20:41 Join DarkkOne [0] (
12:21:44DarkkOneSo, I just read through the IRC log, and I figured I'd throw my "preference" vote for how I'd like preset handling to work. How 'bout holding the thumbstick pops up a list, as seems to have been suggested, and at the bottom of that list (reachable simply by hitting up once on the thumb) is a choice <Add>?
12:22:10*DarkkOne just didn't see the point of needing a second button.
12:23:44 Join rooom_wrk [0] (
12:25:46 Join Moos [0] (
12:28:39 Join hicks [0] (
12:37:28*rooom_wrk is away: Pracuji
12:46:41 Quit Rick (Read error: 104 (Connection reset by peer))
12:47:32 Join Rick [0] (
12:53:22***Saving seen data "./dancer.seen"
12:53:55 Quit midk (Read error: 131 (Connection reset by peer))
12:54:49 Join midk [0] (
12:57:43 Join Sucka [0] (
12:58:49 Quit Asku ("Reconnecting")
12:58:52 Join Asku [0] (
12:59:20 Quit Asku (Client Quit)
13:00:48 Join Asku [0] (
13:13:43 Join Febs [0] (
13:13:44 Quit hshah (Read error: 113 (No route to host))
13:13:49 Quit LinusN (Read error: 113 (No route to host))
13:21:37 Join LinusN [0] (
13:22:02amiconnLinusN: Anything above gdb 6.1.1 doesn't build under cygwin :(
13:22:09amiconn(for SH1 that is)
13:22:10LinusNhow fun
13:22:20 Join webguest88 [0] (
13:22:30amiconn6.2 and 6.2.1 fail at the same spot, 6.3 on another
13:22:51amiconnApart from that, I now have a shiny new SH1 toolchain
13:23:02amiconn(binutils 2.16 and gcc 3.3.6)
13:24:26 Quit ender` (Read error: 104 (Connection reset by peer))
13:24:46amiconnBuilding gcc took 2 hours...
13:25:56 Join XMaster-ShadowX [0] (
13:31:31 Join ender` [0] (
13:32:53 Quit midk (Read error: 104 (Connection reset by peer))
13:34:41 Part XMaster-ShadowX
13:41:52 Quit rooom_wrk ("UkonĨuji")
13:44:07 Join XavierGr [0] (
13:48:35 Join Coldtoast [0] (
13:49:15 Join hshah [0] (
13:51:45*ashridah watches a tumbleweed drift by
13:54:32XavierGrwhen I type cvs -z3 co rockbox-devel do I get the latest daily build or the latest bleeding edge?
13:55:02ashridahbleeding edge
13:55:03LinusNXavierGr: you get the very very latest
13:55:33amiconnLinusN: I just tried another thing: With gcc 3.3.6, it is obviously no longer necessary to build with newlib
13:55:36hshahwhy does everyone start talking when XG asked something... no one said anything to me
13:55:38LinusNhshah: your description on how to patch is not correct
13:55:58B4gderhshah: you didn't ask anything
13:56:01LinusNhshah: i seldom respond to "hi" or "hello"
13:57:15 Join ripnetuk [0] (
13:57:20 Part ripnetuk
13:59:20Coldtoasthey guys
13:59:26ColdtoastI revived a dead laptop
13:59:35 Quit ender` (Read error: 104 (Connection reset by peer))
13:59:45Coldtoastjust an old P2 333 tho. What Linux should I stick on there?
14:00:30bluebrother^I'd recommend Fedora or Ubuntu
14:00:46ColdtoastFedora on such a low end machine?
14:01:01ColdtoastUbuntu seems REALLY popular. maybe that'll do
14:01:10hshahthen u wouldn't want SLES 9 :p
14:01:23bluebrother^why not? But you should try something different than KDE or Gnome.
14:01:57bluebrother^I have Fedora Core 2 running on a server machine. P2-400
14:02:08bluebrother^runs nicely even with freevo :)
14:02:21Coldtoasthow much RAM?
14:02:39bluebrother^64 MB, recently upgraded to 192MB
14:02:43 Join ender` [0] (
14:03:13bluebrother^it's a bit sluggy, but with fluxbox or similar I this this would be pretty usable for daily stuff.
14:03:59bluebrother^my old machine was a K6-2 450 and I had KDE3 running on it. ~300MB RAM.
14:05:33amiconnLinusN: At least under cygwin I can tell for sure that newlib is no longer necessary for building sh-elf-gcc 3.3.6. This cuts down build time and size of installed files dramatically
14:06:47amiconn(gcc build time: 2h −−> ~0.5h)
14:07:15hshahwhats gcc?
14:08:11B4gdergcc is the compiler
14:12:03Mooshshah: your wiki page is still loked
14:15:43amiconnLinusN: I'll try the same under linux
14:19:30 Quit ender` (Read error: 104 (Connection reset by peer))
14:22:30LinusNwill the rockbox world scream and send bomb threats if i remove the peak meter from the fm radio screen?
14:22:47B4gdertry it and see!
14:23:01Moosprobably no
14:23:04*B4gder calls in the bodyguards
14:24:20 Join ender` [0] (
14:24:21CoCoLUSi think you should even remove them from the main unit wps...
14:24:32amiconnLinusN: Don't dare! ;)
14:25:06 Join webguest59 [0] (
14:25:20LinusNthe peak meter is a nuisance in the fm screen because it generates radio interference on the archos models, and will eat lots of cpu on the iriver
14:25:30LinusNand thus lots of battery
14:25:54CoCoLUSwhy does it eat more cpu than those on the wps?
14:26:02amiconnI'd rather fix the interference (I told you my idea this morning) than remove the peakmeter
14:27:11DarkkOneIsn't the peakmeter optional anyway?
14:27:22 Part webguest59
14:28:32CoCoLUSit's enabled by default
14:28:33LinusNCoCoLUS: it doesn't eat more than on the wps, but we can run the cpu on a much lower frequency if we lose the peak meter, and saving lots of battery
14:29:15CoCoLUSok... i'm still in favour of a per-default wps without peak meters :)
14:31:30DarkkOneWait, so you're talking about removing it as an option in the WPS wholly Linus?
14:31:42 Quit ender` (Read error: 104 (Connection reset by peer))
14:32:54 Quit zezayer ("Download Gaim:")
14:33:41CoCoLUS<fn~LinusN> because...?
14:33:53CoCoLUSyou said yourself, it does cost a lot of cpu/battery power
14:33:58amiconnLinusN: Binary size reduced from 192 MB to 82 MB without newlib (/opt/sh1, containing binutils 2.16 + gcc 3.3.6 + gdb 6.1.1)
14:34:02 Join zezayer [0] (n=jake@
14:34:03CoCoLUSand most people won't know that, and leave it enabled
14:34:21amiconnI doubt that it does cost a lot of battery
14:35:34XavierGris there an option to disable peakmeter in WPS?
14:35:35hshahanyone got any suggestions to the errors these guys are getting:
14:36:05 Join ender` [0] (
14:36:46CoCoLUSwell if the drain is negligible, by all means, leave it on per default, but i was under the impression that it does indeed cost noticeable cpu cycles
14:37:19webguest88I vote for a peakmeter-less-tuner-wps. I don't like "nervous" things on the screen
14:39:37DarkkOneI'm kinda a fan of the tuner screen as it is right now, anyway.
14:39:58zezayersimple is always better
14:41:19 Join leftright [0] (
14:41:40leftrightheh, I've just discovered an amusing bug
14:42:11leftrightthe player repwars the very las song in the last file
14:43:01 Quit ender` (Read error: 104 (Connection reset by peer))
14:43:57leftrightplay the very last song the the last file and it just repeats it endlessly
14:45:13LinusNthe peakmeter in the tuner screen will eat considerably more cpu than it would without it
14:45:43LinusNbecause in the normal wps, most cpu power is used for decoding anyway
14:45:53LinusNbut the tuner is idle
14:45:56bluebrother^why not make it optional?
14:46:10 Nick Sucka is now known as Sucka`away (
14:46:21 Quit phaedrus961 (Read error: 110 (Connection timed out))
14:46:38 Join phaedrus961 [0] (
14:47:55leftrightmy spelling is pathetic today
14:48:10*leftright scolds fingers and thumbs
14:26:07***Started Dancer V4.16
14:26:07***Connected to on port 6667
14:26:07***Logfile for #rockbox started
14:26:08***Server message 501: 'logbot :Unknown MODE flag'
14:26:08Mode"logbot :+i" by logbot
14:26:08 Join logbot [0] (
14:26:08 Join Chamois [0] (
14:26:08 Join ender` [0] (
14:26:08 Join phaedrus961 [0] (
14:26:08 Join zezayer [0] (n=jake@
14:26:08 Join Coldtoast [0] (
14:26:08 Join XavierGr [0] (
14:26:08 Join Asku [0] (
14:26:08 Join Sucka`away [0] (
14:26:08 Join Rick [0] (i=rick@unaffiliated/Rick)
14:26:08 Join hicks [0] (
14:26:08 Join Moos [0] (
14:26:08 Join DarkkOne [0] (
14:26:08 Join courtc [0] (
14:26:08 Join bobTHC [0] (n=bobTHC@
14:26:08 Join B4gder [0] (
14:26:08 Join crash_ [0] (
14:26:08 Join ashridah [0] (
14:26:08 Join einhirn [0] (
14:26:08 Join solex_ [0] (
14:26:08 Join gromit`` [0] (n=gromit`
14:26:08 Join thegeek [0] (
14:26:08 Join pike [0] (
14:26:08 Join amiconn [0] (
14:26:08 Join Nibbler [0] (
14:26:08 Join lostlogic [0] (
14:26:08 Join QT [0] (i=as@madwifi/users/area51)
14:26:08 Join TCK- [0] (
14:26:08 Join ansivirus [0] (
14:26:08 Join silencer_ [0] (
14:26:08 Join crashd [0] (
14:26:08 Join Seed [0] (
14:26:08 Join merbanan [0] (
14:26:08 Join pabs [0] (
14:26:08 Join Ismo [0] (
14:26:08 Join bluebrother^ [0] (
14:26:08 Join dwihno [0] (n=dw@
14:26:08 Join mbr [0] (
14:26:08 Join crwl [0] (
14:26:08 Join ze [0] (
14:26:08 Join HCl [0] (
14:26:08 Join igor47 [0] (
14:26:08 Join Bagder [0] (
14:26:08 Join odd_ [0] (
14:26:08 Join CoCoLUS [0] (
14:26:08 Join dionoea [0] (
14:26:08 Join pill [0] (
14:26:08 Join Slasher [0] (
14:26:08 Join Hadaka [0] (
14:27:34 Join LinusN [0] (
14:27:47Chamoisserver is back
14:27:49Chamoiswith linus
14:28:15B4gderhe didn't dare to come back here until the server was up again ;-)
14:29:11 Join hshah [0] (
14:29:20hshahso what happened to the website just then?
14:29:30LinusNmajor server crash
14:29:44LinusNneeded physical attention
14:29:44B4gderit used its legal right to say nothing for a while
14:30:02LinusNhe took the fifth
14:30:13hshahso where are the forums situated then?
14:30:19hshahcoz they were fine
14:30:30B4gderat misticriver-Jeff's
14:30:47hshahahh rite
14:30:52hshahthat would explain it then :)
14:36:53XavierGrOk I would need some help over this. Code is here:
14:37:13XavierGrThe first time code runs fine then the big char array is not working as intended
14:37:42XavierGrThe program opens a folder and filters only jpg files
14:37:45zezayerXavierGr: your latest remote build doesnt work
14:38:23XavierGrIt builds the names of the files in contents and has the pointer array to point the right position for the filename.
14:39:11zezayerits ok redownloaded it and works now
14:39:26XavierGrOnce the file loop is finished I extract the whole path to 2 arrays previous and next to load the files.
14:40:03XavierGrMy main concern is why if it is running right the first time then the big array is confused with other data.
14:40:38B4gderperhaps you overflow the array
14:40:43LinusNXavierGr: you shouldn't advance the length before you assign the pointer
14:40:59B4gderyou don't check if you go beyond the buffer size
14:41:42XavierGrB4gder: I will correct this later once I made this running for 4 or 5 files
14:42:02DarkkOneBut that could be causing your problem.
14:42:32B4gderyou also assume that contents[] is zero filled at start
14:42:45B4gderwhich it isn't
14:43:11B4gder(at least that's how I interpret your code)
14:43:21XavierGrDarkone: I think not. I use 4 files for test so I don't think that I run out of buffer.
14:43:26LinusNbut the most obvious bug is that "length" is advanced *before* the pointer is assigned, so each pointer points to the byte right after the string
14:43:50LinusNline 50 and 53 should switch paces
14:44:22XavierGrso is this posible that it runs fine the first time and then not?
14:44:36B4gderpure luck
14:47:15LinusNXavierGr: make the contents array global or static
14:47:20 Join XMaster-ShadowX [0] (
14:47:31LinusNin fact, make them all static
14:47:40B4gderand use strcpy() instead of strcat()
14:48:54XavierGrB4gder: But with strcpy the whole array will be replaced to the new value. I want to add strings to contents not replace.
14:49:32XavierGrstrcat contcatenates strcpy copies
14:49:46B4gderbut you don't even use the concat feature of strcat
14:49:53B4gderyou add length manually
14:50:22XavierGrahh so if I define the legth in strcpy it shouldn't get replaced?
14:50:25B4gderstrcpy() would fix your no-trailing-zero bug
14:50:35XavierGrokay got it.
14:50:45XavierGrah and LinusN why make the global?
14:51:03XavierGrI need only previous and next to keep their values
14:51:09LinusN1) it is automatically zero-filled
14:51:18LinusN2) it doesn't eat up all stack
14:51:32B4gdernext and previous keep pointers to that data
14:51:35XavierGrmake all the char arrays global then?
14:51:42B4gderso they point out into oblivion afterwards
14:51:56LinusNand 4) it is often easier to debug
14:52:21B4gderhm, no I was wrong
14:52:38B4gderthe smaller buffer sizes confused me
14:52:42DarkkOneMy usual debugging process is to work out what all the variables' values should be by hand in advance as far as I can, then output all over the place until I get a wrong number or string. ><
14:53:19DarkkOneI'm still missing something though... the +1 on line 50 is for what?
14:53:19LinusNlogf is your friend
14:53:36LinusNDarkkOne: the null byte
14:53:48XavierGrB4gder: the null character.
14:53:49DarkkOneWell most recently I've been doing alot of PHP stuff, so it's just echoes all over the place. Heh.
14:54:10DarkkOneGotcha, null char.
14:54:17LinusNi use a debugger whenever i can
14:54:22XavierGropps I thought B4gder said that
14:54:28 Join edx [0] (
14:54:31LinusNplugins are a little bit tricky to debug though
14:54:37 Quit ashridah ("Leaving")
14:54:39hshahLinusN: any way of modifying the text that comes up when you turn on the iriver?
14:54:45XavierGrI can't use logf into plugins right?
14:54:50LinusNhshah: which text?
14:55:04LinusNXavierGr: rb->logf()
14:55:19LinusNbut you need to select logf in the configure script
14:55:21hshahas soon as u turn the iriver on... texts appears ... i think it says rockbox vX etc etc
14:55:29XavierGrI thought I did that and I got a compiler error not knowing the function I will try again.
14:55:33LinusNhshah: why do you want to change that?
14:55:43B4gderXavierGr: in plugins you need to use LOGF()
14:55:51LinusNB4gder: ah, right
14:56:03XavierGrso I just call it LOGF()?
14:56:10 Join b0bTHC [0] (n=bobTHC@
14:56:13hshahLinusN: for the hell of it really... im trying to personalise my iriver as much as possible
14:56:22LinusNLOGF("My variable: %d", variable);
14:56:35XavierGrgot it, like printf
14:56:40LinusNhshah: i wouldn't recommend messing with the boot loader
14:56:51hshahok sure... advice noted
14:57:03hshahwouldn't want to break my iriver would we
14:57:17LinusNquite unnecessary
14:58:17 Part LinusN
14:58:28 Join LinusN [0] (
14:58:59 Quit ender` (Read error: 104 (Connection reset by peer))
15:01:24 Quit edx ()
15:01:53XavierGrYou guys are life savers! Problems solved!
15:04:02 Join ender` [0] (
15:04:21XavierGralso what do you think about the stripping methods I use for path, name, file (lines 20- 31)?
15:05:19XavierGrCould it get a little more optimized?
15:05:32hshahzezayer's idea in the forums is pretty cool
15:05:43XavierGrwhich idea?
15:06:27B4gdertimeshifting radio
15:06:52amiconnLinusN: That's strange: building sh-elf-gcc 3.3.6 without newlib under linux does *not* work
15:06:52zezayerpausing / rewinding live radio, idea nicked from MR forum, prob not poss but woudl b good
15:07:17XavierGrwouldn't be that a little difficult to implement?
15:07:32XavierGrEspecially when the record routines are not finished for iriver yet.
15:07:47zezayerpausing mite b poss [ tho hard ] rewinding will almost deff b impos
15:08:41hshahcool idea none the less
15:09:12 Quit bobTHC (Read error: 110 (Connection timed out))
15:09:29B4gderover two years old ;-)
15:10:14hshahahh well its the first time i read about it, so its new for me :p
15:11:03amiconnLinusN: Trying that gives a long list of syntax error output, then make errors ending with
15:11:05amiconnmake[2]: *** [libgcc/./embed-bb.o] Fehler 1
15:11:06amiconnmake[2]: Leaving directory `/home/jens/build/gcc/gcc'
15:11:06amiconnmake[1]: *** [stmp-multilib] Fehler 2
15:11:06DBUGEnqueued KICK amiconn
15:11:06amiconnmake[1]: Leaving directory `/home/jens/build/gcc/gcc'
15:11:06amiconnmake: *** [all-gcc] Fehler 2
15:12:06amiconnThe errors all occur in ../../../gcc-3.3.6/gcc/gcov-io.h:222
15:12:15amiconnDrop the :222 ...
15:12:20 Part XMaster-ShadowX
15:16:20 Join XMaster-ShadowX [0] (
15:16:43XavierGrYay scrolling in jpg viewer is working (though alphabetical sorting is not implemented)
15:17:16zezayerbye all
15:17:18 Part zezayer
15:18:12XavierGrLinusN: What do you think that should happen whe scrolling if someone gets to a pic that is unsupported?
15:18:27B4gderadvance to the next
15:18:41XavierGrCurrently an error message pops up and quits. Do you think that it must continue to the next jpg?
15:19:22XavierGrand what if there is only 1 not supported jpg?
15:19:38XavierGr(1 jpg in the whole folder)
15:19:40LinusNit should show the pics it supports and skip the rest
15:19:58 Quit pill ("changing servers")
15:19:58XavierGrso no error messages?
15:20:05 Join pill [0] (
15:20:15LinusNwell, you could splash() a short message if you like
15:20:50XavierGrDoes the HZ parameter in splash() is target specific?
15:20:59LinusNamiconn: and if you try with newlib?
15:21:04XavierGrBecause HZ*2 is very long time for iriver
15:21:16XavierGrbut maybe is short for archos?
15:21:18LinusNHZ is one second
15:21:26LinusNon all targets
15:21:31XavierGrthats good.
15:21:40LinusNthat's the whole point
15:22:16XavierGrI will reduce all warnings to HZ/2 I think. I remember that It was kinda anoying to have to wait 2 seconds on a not supported jpg.
15:23:05LinusN0.5 second is a little too short imho
15:23:28B4gderimho, you should only show a message if no files at all is supported
15:23:30hshahguys, is there a problem playing wav files?
15:23:32LinusNmaybe 1s for the first message and 0.5s for the rest
15:23:46LinusNhshah: yes and no
15:23:49hshahcoz i have a 15 second wav and it plays around 2 secs of it b4 moving onto the next track
15:24:10LinusNhshah: interesting
15:24:20hshahwant me to upload the wav?
15:24:24LinusNyes please
15:24:37XavierGrI say 0.5 because these warnings (in the jpg.c) aren't of any iportance for the user.
15:29:47hshahgtg now... LinusN... please email me an answer or PM me on the forums - email address is
15:29:51 Quit hshah ("CGI:IRC")
15:31:09 Quit Chamois (" Want to be different? HydraIRC -> <-")
15:34:23 Quit DarkkOne ("Chatzilla 0.9.68a [Firefox 1.0.6/20050716]")
15:34:35 Part LinusN
15:35:58 Quit b0bTHC ("Smoke Weed Every Day !")
15:41:31 Join Aison [0] (
16:11:45 Quit XavierGr (Read error: 104 (Connection reset by peer))
16:26:11***Saving seen data "./dancer.seen"
16:30:46 Quit XMaster-ShadowX (Read error: 110 (Connection timed out))
16:35:38 Join Craig [0] (
16:38:29 Join ender1 [0] (
16:38:31 Quit ender` (Read error: 104 (Connection reset by peer))
16:38:41 Nick ender1 is now known as ender` (
16:38:57 Quit B4gder ("Lämnar")
16:46:34 Join amiconn_ [0] (
16:56:07 Quit Craig (Nick collision from services.)
16:56:40 Join Craig_ [0] (
17:03:04 Join Lear [0] (
17:11:05 Quit amiconn (Read error: 110 (Connection timed out))
17:11:05 Nick amiconn_ is now known as amiconn (
17:38:32 Join phaedrus96 [0] (
17:43:36CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:43:36*HCl feels productive today
17:43:38*HCl prods amiconn
17:46:27*HCl stares at some pretty outdated commented code he wrote a while ago
17:47:16 Quit phaedrus961 (Read error: 113 (No route to host))
17:48:29HCli don't like c.. no functions with the same name but different arguments
17:49:49*HCl should've documented his code
17:49:49Slasheri like c much more than c++ for example
17:50:06Slasheri have found that c code is almost always more readable and easier to maintain than c++
17:50:11HCli dunno
17:50:21HCli don't like having getfentrybyfilename(struct id3 *blah)
17:50:27HCland getfentrybyfname(char *path)
17:50:34HClwhen they're doing the same.
17:51:16Slasherhmm.. maybe you don't need the first function at all
17:51:36HCli dunno, i need to read my own code again
17:51:57Slasheryou could always get_file_entry(blah->path)
17:53:24HCli'll change that later
17:53:31HClfirst i got to figure out what this outdated code is supposed to do.
17:53:35HCli think i'll just rewrite it..
17:56:13 Join rooomish [0] (
18:00:40*HCl wonders whether he could do &(long)0
18:02:33 Nick Sucka`away is now known as Sucka (
18:17:21 Join alienhuladancer [0] (
18:17:56alienhuladancerwhy re all the plugins not in the .rockbox folder in the daily builds anymore?
18:18:17Learthere were zero before?
18:18:25Lear(in size, that is...)
18:18:32alienhuladancerthey were all organized in .rockbox
18:18:46alienhuladancernow they are all free in the .zip file
18:19:01Learand you didn't extract it incorrectly? :)
18:19:21alienhuladanceri just opened it with winzip
18:20:15Learlooks fine to me. check your winzip config. :)
18:20:31alienhuladancerok thanx...I think I got it
18:21:50alienhuladancergot it.....the .rocks used to be in .rockbox in the zip file.....just out of curiosity....oh well
18:22:01alienhuladancerdo you write rockbox?
18:26:13***Saving seen data "./dancer.seen"
18:35:55 Join MentholMoose [0] (
18:38:34 Quit MentholMoose (Client Quit)
18:39:28 Join XMaster-ShadowX [0] (
18:39:50 Join webguest16 [0] (
18:40:15 Join XavierGr [0] (
18:40:35XavierGrhello, I have a quick question about filenames.
18:41:14XavierGrI know that currently the filename of a file (path and name) should be up to 260 characters.
18:41:43XavierGrwhat about the name of the file alone (not the path). Is there a limitation, or it shares the same with the path.
18:44:01alienhuladancero sorry I was gone....
18:44:05alienhuladancerdo you write code?
18:45:56 Join bagawk [0] (n=lee@unaffiliated/bagawk)
18:49:52XavierGrerror: variable-size type declared outside of any function
18:50:01XavierGrwhy is that? Ihave set a define for that.
18:50:30Learxaviergr: don't think I've seen that error before...
18:50:50XavierGrwell I have a static global char array.
18:51:19XavierGrand I want to have as many elements as global_settings.max_file_in_dir
18:51:44XavierGrI did a define for it but again it ouputs the same error.
18:51:56Learah, well, that must be a static size, because the compiler needs to have that information. the setting is only available during run time.
18:52:16Learstatic as opposed to dynamic...
18:53:41XavierGrany solutions for it?
18:54:24 Join edx [0] (
18:55:07Learnot really. Rockbox uses a hack to get that in the "core", by using a piece of the audio buffer as the directory cache. This is only possible to do during startup though (so the file buffering code doesn't need to deal with changing buffer sizes).
18:56:02Learif it is a plugin, use part of the remaining plugin buffer (perhaps in a similar manner).
18:56:24XavierGrI don't see any alternatives, thanks.
18:56:58 Join hshah [0] (
18:57:52 Join matsl [0] (
18:57:52 Quit webguest16 ("CGI:IRC (EOF)")
18:58:00 Join edx__ [0] (
18:58:43hshahanyone here who can code in C write me a small patch?
18:58:49hshahit shouldn't be too hard...
18:59:42Craig_what are you after?
19:00:29*HCl prods amiconn, Bagder..
19:00:32HClwake up you two
19:00:52*HCl is in front of another decision about the tagdb and needs to know opinions.
19:01:13hshahim looking for a patch that will allow me to go to WPS from the root directory when pressing left
19:04:49hshahi don't think it will be too hard to do
19:05:16Craig_the problem is not such much coding the patch as getting consensus on interface changes. If you've got that first, someone (me?) might look at coding it.
19:07:12Craig_and when I say consensus, I mean from those with commit access.
19:07:54 Quit alienhuladancer ()
19:08:00hshahnah... im just looking for one of those patches that probs won't be commited... but one that people can still use if they feel like it
19:08:35 Quit XavierGr ()
19:12:35 Quit bagawk (Remote closed the connection)
19:16:38 Join bagawk [0] (n=lee@unaffiliated/bagawk)
19:17:18 Join Aison` [0] (
19:17:19 Quit edx (Read error: 110 (Connection timed out))
19:17:41 Join edx [0] (
19:17:44 Quit Aison` (Read error: 104 (Connection reset by peer))
19:22:17 Quit edx__ (Read error: 110 (Connection timed out))
19:29:19 Quit rooomish ("game is over")
19:34:13 Quit Asku ("Changing server")
19:39:39 Quit hshah ("CGI:IRC")
19:39:39 Quit gromit`` (Read error: 104 (Connection reset by peer))
19:42:00 Join NibbIer [0] (
19:43:04 Quit ansivirus ("Leaving")
19:50:31SlasherHmm, dual codec support almost done but i don't know yet if it would work at all ;)
19:51:34bagawkSlasher, lets see oyu do cross-fading using both codecs!
19:52:07 Join muesli- [0] (
19:52:19Slasherhmm.. you mean i could crossfade between voice-ui samples? :D
19:52:52Leardual, as in one mp3 and one other? or can voice be in any format?
19:53:17Slashervoice can be any codec, but it should be fast enough
19:53:28Slasherso now we can load two dynamic codec at same time
19:53:41muesli-mmh, who did the last gapless-fixes from yesterdays build?
19:53:46amiconnI think we can start with mp3, then swithc to speex when it gets implemented
19:53:59Slasherthen both are decoding the stream in parallel and using the same code space
19:54:14LearI did.
19:54:24Slasherno, interleaved i mean..
19:58:33 Quit Nibbler (Read error: 110 (Connection timed out))
19:59:03 Nick NibbIer is now known as Nibbler (
20:02:02 Join NibbIer [0] (
20:03:55 Join BoD[] [0] (
20:03:58BoD[]Hello !
20:04:00Slasherat least codec swapping _seems_ not to crash the playback and it's fast enough at 120 MHz
20:04:31BoD[]I'm currently testing a creative zen touch
20:04:37BoD[]and ... well it sucks :)
20:04:40Slasheri will try soon some real voice ui playback
20:05:16BoD[]it is not possible to browse by folder
20:05:30BoD[]and it is not seen as a usb mass storage
20:05:38BoD[]you need a special driver
20:05:50BoD[]and then you can only copy music files... no data files
20:07:57Learslasheri: two things:
20:08:00BoD[]i'm desperate to find a good player
20:08:34muesli-l8er mates
20:09:00Lear1) I've run a while with 3 * chunk_size as watermark a while now, seems to be enough - except when playing high bitrate oggs with 95% cpu boost, in which case file buffering starts to interfere with playback... :)
20:09:23Learso the pcmbuf watermark can probably be reduced a bit...
20:10:17Lear2) the gapless fixes I committed apparantly interferes a little with the crossfade; the "old" track stops a bit earlier now (but more correct, based on the id3->length information at least).
20:11:07LearI don't really understand how the crossfade works (when to start and all that), so I'm not sure how that should be fixed...
20:12:07 Join xNibbler [0] (
20:12:39 Quit Nibbler (Read error: 104 (Connection reset by peer))
20:15:33BoD[]what's a good mp3 player ???
20:16:09BoD[]i mean
20:16:19BoD[]i had an archos a long time ago
20:16:58BoD[]i had a iriver for 2 days
20:17:04BoD[]i didn't really like it
20:17:17BoD[]no way to scroll
20:18:49 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
20:19:16BoD[]when you have a lots of artists
20:19:30BoD[]how do you go to the letter M
20:19:40BoD[]you press on down for 2 minutes :)
20:19:59BoD[]with a wheel, or a ipod-style-pad-thinggy, it's fast
20:21:01 Quit bagawk ("Leaving")
20:22:26 Join Nibbler [0] (
20:22:59 Quit NibbIer (Read error: 104 (Connection reset by peer))
20:26:15***Saving seen data "./dancer.seen"
20:26:16ColdtoastI have no probs getting to M fast
20:26:46 Join [1]Moos [0] (
20:26:48Coldtoastas my dirs are A-E, F-J, K-O, P-T, U-Z
20:27:44 Quit muesli- (Read error: 110 (Connection timed out))
20:28:16BoD[]well that's a solution
20:28:17 Quit Moos (Read error: 104 (Connection reset by peer))
20:28:19 Nick [1]Moos is now known as Moos (
20:28:25SlasherLear: Hmm, yes. Maybe that watermark could be lowered a bit
20:28:27BoD[]but I prefer a wheel or something like that
20:28:58SlasherLear: If the files really end where they should, that shouldn't be a problem for the crossfader (at least when set to mix-mode)
20:29:04amiconnNarrgh! Again one of those things that keep me from using linux over and over :(
20:29:24Slasheramiconn: Hmm?
20:29:26amiconnI installed gcc 4.0.1 in my debian vm. Now I can't logon as user
20:30:35Slasherhmm.. gcc shouldn't have anything to do with system logon process..
20:30:40amiconnIf I try, gdm pops up an error window with an ok button, then goes back to the logon screen
20:31:08amiconnIt is possible to log on as root (I first needed to allow this in the gdm configuration)
20:31:25amiconnThe error message is in german, here's a rough translation
20:31:54Slashertry using kdm/wdm instead of gdm.. maybe one of those would work better
20:32:01Slasheror try apt-get update && apt-get dist-upgrade
20:32:35 Quit xNibbler (Read error: 104 (Connection reset by peer))
20:32:52HClamiconn: would you mind if i adjust the database design so that filelen is MAX_PATH by default?
20:32:56amiconn"GDM was unable to write into your legitimation(?) file. This may mean that no disk space is available or your home directory could not be opened for writing. Anayway, you can't log on. Contact your system administrator"
20:33:03HClthat way i can be guaranteed that i can always add new files
20:33:14amiconnIt used to work before
20:34:18crwlgcc 4.0.1 only seems to be found on debian unstable/testing, which is not meant for people who don't know how to fix the problems that will occasionally come when using unstable
20:34:28SlasherHmm, interesting.. try rm -rf your ~/.xauth/ dir
20:35:05Slasheri have unstable on all computers
20:35:14Slasherand it works really well
20:35:32amiconncrwl: I installed gcc 4.0.1 from the graphical package manager (synaptic). Afaik I don't have debian unstable among my installation sources
20:35:54Slasheramiconn: i think that broke your library dependencies
20:36:03*HCl stares at amiconn
20:36:05amiconnThat's what I hate about Linux - it's just not that you install it and it just works
20:36:18Slasheryou shou upgrade to unstable first..
20:36:26crwlamiconn, well, says otherwise
20:36:39amiconnI have to keep kernal at 2.4.x in all cases
20:36:52crwlSlasher, synaptic shouldn't (and doesn't, afaik) break dependencies
20:36:53Slasheryou can keep it.. but why?
20:37:08amiconnkernel 2.6.x doesn't work correctly under VMware
20:37:21crwlthe gcc4 from unstable just probably downloaded something that else it depended on, something that happened to be broken right now
20:37:24Slashercrwl: hmm, it shouldn't.. but that's always possible when updating to an unstable package with stable system
20:37:31Slasheramiconn: ah, ok
20:37:32crwlwell, true
20:38:23crwlSlasher, i currently run ubunty breezy (development branch) on my desktop (and do dist-upgrade almost daily), this has been broken in many exciting ways recently :9
20:38:25 Quit Aison (Read error: 104 (Connection reset by peer))
20:39:37Slashercrwl: oh, i haven't tried ubuntu yet :)
20:40:13ColdtoastI'm installing Ubuntu right now
20:40:22Coldtoaston an old P2 333
20:40:28crwlmy server has some strange combination of debian woody, sarge and unstable - i've never done a full dist-upgrade to it, i've only installed security upgrades and occasionally got newer versions of programs from unstable
20:41:50crwlSlasher, ubuntu is pretty nice, it mostly seems like debian with a release schedule to me ;)
20:42:29crwlthough they seem to be busy adding all kinds of new features not present on debian, too
20:42:35HClum. amiconn..?
20:42:38HClcan you read my question? o.o.
20:43:23Slashercrwl: hehe, nice
20:43:36amiconnHCl: Do you mean the runtime database or the tag database?
20:43:42HCltag database
20:43:47HClruntime database has no stored path
20:43:55HClit'd mean that entries in the tag database for the file path
20:43:57HClare always the max length
20:44:10HClonly thing that would need changes are the database generators
20:44:10amiconnThe tag database isn't changed on the player, right?
20:44:20HClnot yet. i wrote code just now to do that.
20:44:26HClso the runtime database will work
20:44:35HClregardless of it being in the tagdatabase or not
20:45:02amiconnHmm, I don't think I would want that
20:45:17amiconnUsing MAX_PATH all then time would be a waste of space
20:45:23HCli know.
20:45:39HClbut its the only way i can guarantee that i can add new files
20:46:56HCli guess the other way could be possible.
20:47:14HClan extra file with files that should be added, dunno how much overhead that would give
20:47:25HClhave you had any time to check whether your archos implementation works?
20:49:19amiconnNot yet. I need to build current databases first, then try it out on iriver, check what runtimedb it produces, then try the same on archos after implementing the hooks
20:49:28 Join NekoNoNiaow [0] (
20:50:55amiconnHmm, it seems gdm is right after all. My disk is full... tststs
20:51:50amiconnI wonder why no other program complained though...
20:52:19 Join Asku [0] (
20:52:24amiconn...and what took away all that space
20:52:31crwlthey perhaps didn't write anything to the disk that was full (maybe you have /home on different partition?), or didn't write at all
20:52:52amiconnNo, it's only one partition
20:53:30NekoNoNiaowamiconn: if you're under linux logs can fill a partition if you don't limit their size
20:53:33 Join webguest92 [0] (
20:53:48crwldebian should come with a working logrotate configuration, though
20:54:02webguest92amiconn: I'm betting on apt cache (/var/cache/apt/archives). Try running apt-get autoclean
20:54:06NekoNoNiaowcrwl: in theory it should indeed
20:54:11crwlpossibly non-working if you run unstable
20:54:22crwli've had issues with it
20:58:58 Quit Nibbler (Read error: 110 (Connection timed out))
20:59:22 Join Nibbler [0] (
21:00:05amiconnSpace issue solved now, logon working again
21:00:36amiconnThe update to newest installed gcc 4.0.1 by default...
21:01:22webguest92Didn't you say you were running stable?
21:02:34amiconnI just used the debian web installer to set up the vm. I checked now, package sources are all 'testing'
21:03:35amiconnI wonder why debian has so many confusing version names
21:04:07amiconnstable/testing/unstable vs. woody/sarge/? vs. version numbers...
21:04:39webguest92Not more confusing than windows
21:04:47webguest92Not by much anyway
21:04:58webguest92longhorn, vista,
21:05:14amiconnlonghorn isn't official, so it doesn't count
21:05:52webguest92When the company uses the name, it's official to me
21:05:53amiconnI've seen no place that uses the ms build numbers to indicate a windows version to use
21:06:13webguest92Well, I've seen no place using the debian version numbers alone
21:07:01amiconnI did
21:07:23crwlstable is always the newest release (currently named sarge, previously woody), testing is the release that'll someday be the next stable (it'll be named etch) and unstable is always named sid, i don't think this is really difficult at all
21:07:38crwland the releases always have a version number with the name, too
21:08:18webguest92"Results 1 - 10 of about 197,000 from for longhorn" that's official enough for me
21:08:43webguest92Also, holy crap, 197,000!?
21:09:07crwlhmh, the newest release currently is sarge, the previous one was woody, i don't know if i sounded clear enough :)
21:09:21crwlanyway, sleep ->
21:20:39NekoNoNiaowseen sox
21:20:55 Join tvelocity [0] (
21:20:57NekoNoNiaowwell tried :)
21:25:43 Join Jato [0] (
21:26:02 Part Jato
21:29:41 Join muesli- [0] (
21:30:39bluebrother^amiconn: maybe you should try a different linux distribution ...
21:31:07bluebrother^I tried Debian once and didn't liked it at all. Went back to Redhat and now I'm pretty happy with Fedora.
21:31:23bluebrother^(as workstation)
21:35:35webguest92No way. Debian reigns supreme!
21:35:36 Quit muesli- (Read error: 104 (Connection reset by peer))
21:35:55amiconnSo far I didn't really like any distribution
21:36:07webguest92You should try ubuntu some day
21:36:18webguest92It sounds like you'd expect something like that
21:36:27amiconnDebian is what worked best so far, but even then not really smooth
21:36:40bluebrother^have you tried Suse? Another one I didn't liked really.
21:36:40webguest92Ubuntu is Debian with added smooth, pretty much
21:37:06bluebrother^I've seen Ubuntu once or twice, looked pretty nice.
21:37:17bluebrother^but I prefer KDE instead of Gnome ;-)
21:37:30tvelocityubuntu roxxx
21:37:38amiconnThere are always things that don't work in the default installation, and getting them to work takes hours, sometimes days, or even no way at all
21:37:41bluebrother^but it's a distro that's on my should-try-list.
21:37:50amiconnThat'as why I use linux only if I have to
21:38:05webguest92Well, the default installation of Windows infuriates me
21:38:11bluebrother^what kind of are these things you came upon?
21:38:13webguest92I can't even burn cds!?
21:38:20webguest92(very well)
21:38:35webguest92It doesn't even come with a capable web browser :(
21:38:36amiconnI can install windows in under 2 hours with all hardware working. I did so on quite a number of boxes
21:38:37bluebrother^on Windows I need usually ~3 days until I can use it productively :)
21:38:55amiconnI didn't manage to do that on a single linux box
21:38:57webguest92Well, I installed Linux on my laptop in an hour
21:39:18amiconnwebguest92: wlan working? gfx chip working? sound working? modem workin?
21:39:30bluebrother^I reinstalled my laptop ~1/2 year ago. Nearly everything works.
21:39:30webguest92Yes, yes, yes, maybe
21:39:50webguest92I don't have any desire to test my modem
21:39:57webguest92Frankly I wish it'd go away
21:40:11amiconnI always prefer a modem over no internet connection at all
21:40:23webguest92Modem != internet connection
21:40:33bluebrother^in my case: no (knew it before), yes, yes, no (winmodem but easy to add)
21:40:47webguest92Surely not for me, since I don't even have a phone line.
21:41:02bluebrother^I once installed a laptop with a real modem (no winmodem) and it worked ootb.
21:41:28amiconnI'm talking about the integrated modem (99% of them are winmodems)
21:41:51webguest92Well, I don't know as I said, but I saw something about it.
21:42:30bluebrother^this _can_ be tricky. My laptop has a winmodem and I can use it without any kernel modules thanks to alsa. I need only a daemon :)
21:42:43webguest92Anyway, winmodems not working is hardly the fault of linux
21:43:01amiconnbluebrother^: Btw, SuSE is a distro I know of. Wouldn't try it on any of my boxes...
21:43:23amiconnI know suse since kernel 0.9.something
21:44:00bluebrother^amiconn: I tried various distros over the past years. Always I got into touch with Suse I switched back to something else :)
21:44:09webguest92I tried installing redhat5.2 the other day using qemu
21:44:17webguest92Was surprisingly easy
21:44:21webguest92wouldn't run though
21:44:30bluebrother^but they're great with their hardware recognition. Until Yast trashes all config files.
21:44:32webguest92Or rather, the graphics was messed up
21:44:37amiconnwebguest92: Yes, but that's not the point. I have a certain piece of hardware, my Laptop. I want to work with it, and use all of its features that seem useful to me
21:44:58*bluebrother^ too :)
21:45:23amiconnI can do that with WinXP, with little configuration time
21:45:47amiconnWinXP even was preinstalled, and I didn't need to reinstall a single time
21:46:02amiconnI can't do that with Linux even with days of configuration
21:46:07bluebrother^sure. Your vendor preinstalles WXP some way it workes with all hardware.
21:46:11webguest92Well, you still shoudln't blame Linux for that. Blame your shitty hardware.
21:46:51bluebrother^I didn't wanted to convince you. But esp. on laptops debian may be the wrong choice imo.
21:46:57amiconnThe thing is that even hardware that is said to be supported in Linux is hard to get working sometimes, if at all possible
21:47:29bluebrother^I know. But this is really dependent on the exact type of hardware.
21:47:45amiconnAnother thing is that I would switch to linux immediately, if I would find a way to make it work without needing days and days
21:47:51amiconnThat's what really annoys me
21:47:53bluebrother^I have a preinstalled WXP and it runs pretty crappy. I seldomly use it.
21:48:28bluebrother^so in my case linux runs better than Windows on this box (ok, there are some minor glitches)
21:48:56amiconnE.g., I tried to get wlan running on another box, Ralink RT2500 chipset
21:49:07amiconnThere is even a driver from the chipset manufacturer
21:49:23webguest92Only one thing didn't work out of the box on my laptop with ubuntu: cpu frequency scaling. Took "modprobe p4-clockmod" to get that working.
21:49:24amiconnI was able to compile and load the module w/o problems
21:49:30bluebrother^never heard of this chipset.
21:49:40amiconnHowever, I don't manage to get a connection to the AP, whatever I try
21:49:48*NekoNoNiaow is Laurent by the way (for those who remember ;) )
21:50:17amiconnIt works without problems in windows, on the very same box, to the same AP
21:50:27bluebrother^eeks. I really don't like manufacturer drivers under linux. Had some bad experience with that.
21:50:35amiconnIt's open source
21:50:40webguest92Maybe the driver is using some non-standard configuration
21:51:01webguest92instead of wireless-wotsit
21:51:09bluebrother^I tried getting a realtek-wlan to work. They used pretty strange configuration stuff.
21:51:22webguest92As does linux-wlan-ng
21:51:32webguest92Why they do that is baffling
21:51:37bluebrother^but the driver (binary :( was unusable, reproduceable lockup of the whole machine :(
21:51:59webguest92Good thing there's a different driver for that wlan card
21:52:10webguest92Which uses the standard configuration things
21:52:21webguest92And works out of the box on ubuntu \o/
21:52:37bluebrother^nice :)
21:52:48webguest92On pcmcia even
21:53:33 Join webguest16 [0] (
21:59:53BoD[]does anybody know if the gmini 400 is good ?
22:00:48 Part webguest16
22:04:01BoD[]and/or gmini xs202
22:11:00Bagderrockbox doesn't run on them
22:11:07Bagderwhich makes them bad ;-)
22:11:41 Join the_winch [0] (n=winch@
22:12:16NekoNoNiaowbut other projects are working on them (google)
22:12:27Bagderthere are?
22:12:54NekoNoNiaowit's a guess, i imagine they are not much different from other archos multimedia models
22:13:18 Quit edx ()
22:13:24Bagdernew firmwares are really rare
22:13:35Bagderso my guess is that there is no other project
22:13:48BoD[]what about rockbox on ipod ?
22:14:09Bagdernobody seems to want it enough
22:14:15Bagdernobody with an ipod that is
22:14:19 Join hshah [0] (
22:14:21BoD[]well :)
22:14:24BoD[]too bad for them
22:14:38BoD[]i know I would probably buy an ipod if rockbox ran on it
22:14:39NekoNoNiaow (from rockbox' site)
22:15:38NekoNoNiaowBoD[]: an ipod is already very usable it seems so the need for a better firmware is less obvious
22:16:00Bagderyou find a firmware project on that site?
22:16:11BagderNekoNoNiaow: I very much doubt that
22:16:51NekoNoNiaowBagder: i never said that, but the fact that the site exists, proves there's at least a bit of work done
22:17:01BoD[]well browsing by folders is a must have feature for me
22:17:05Bagderthey scanned the pcbs, yes
22:17:11Bagderquite far from a firmware project
22:17:37BagderI'd love to see more firmware projects
22:17:52Bagderbut people like to talk a lot more than they hack
22:18:06 Join Strath [0] (
22:18:08BoD[]what about linux for ipod
22:18:12BoD[]maybe that's good
22:18:59NekoNoNiaowah shit, my problem's still here
22:19:17BagderBoD[]: possibly, just that linux seems so out of place on a music player
22:19:40webguest92If people wanted to port rockbox to ipod, it'd be so much easier than any of the other undocumented players
22:20:02webguest92(due to the work of ipodlinux)
22:20:11BoD[]it's too bad that 1/ I suck, 2/ I don't have an ipod :)
22:20:19 Join einhirn [0] (
22:20:23BoD[]otehrewise i'd do it myself ;)
22:20:35NekoNoNiaowdoes anyone remember what the dependency ": Makefile" means in a makefile ?
22:20:57Bagderit means the left side file depends on the Makefile
22:21:29 Quit Lear ("Chatzilla [Firefox 1.0.6/20050720]")
22:22:10NekoNoNiaowso i'm gonna hack that config.texi stuff instead of trying to fix it, bad bad GNU guys
22:23:42 Part XMaster-ShadowX
22:24:10bluebrother^why is the ID3 info screen still called "ID3 Info"?
22:24:34Bagderbecause no one fixed it?
22:24:47BagderI don't care very much
22:24:53bluebrother^As rb plays ogg, wavpack etc. wouldn't it be better to call this "Metadata Tag" or similar?
22:24:59bluebrother^ok, thats a point :)
22:25:14webguest92One thing that bothers me, while we're at it
22:25:18bluebrother^are the strings only in the language files?
22:25:30NekoNoNiaowthere's a strange problem when building the binutils : a VERSION variable is not set which causes a file not to be created, then the build to fail
22:25:42bluebrother^or is there a hardcoded "C" "locale"?
22:25:54webguest92is that under LCD Settings is "LCD Mode", where the options are "inverse" and "normal"
22:26:08Bagderbluebrother^: the strings are in the lang files, but one lang file is the "hardcoded" one at build-time
22:26:08webguest92This should be "Inverted" "yes/no"
22:26:16***Saving seen data "./dancer.seen"
22:27:30bluebrother^so I need only to change the language files? So this sounds to be an easy one.
22:28:03BagderYes, I believe that should be enough
22:31:53amiconn'make' in tools/ throws a ton of warnings with gcc 4.0.1
22:32:09Bagdersigned/unsigned ones?
22:32:26hshahamiconn - i have a patch now that takes me to wps on left from root
22:32:29webguest92Why's bdf2bmp not compiled, by the way?
22:32:33Bagderthey're only disabled elsewhere ;-)
22:32:37hshahCraig_ was nice to make it for me
22:35:17Bagderwebguest92: probably because very few ever use it
22:35:26bluebrother^amiconn: have you heard of colinux? Maybe this is of interest for you ... running linux on windows without vmware. But I haven't tried it myself.
22:41:47 Join hardeep [0] (i=hardeeps@norge.freeshell.ORG)
22:44:53 Quit matsl (Remote closed the connection)
22:45:42NekoNoNiaowi hate it when configure takes more time than the build itself
22:45:42 Quit Moos (Read error: 104 (Connection reset by peer))
22:45:47 Join Moos [0] (
22:45:51 Quit Craig_ (Read error: 104 (Connection reset by peer))
22:45:57NekoNoNiaow(i know that no one cares, but it relieves me ;) )
22:46:28Bagderstill, configure as a concept is a pretty good thing
22:46:41NekoNoNiaowi agree
22:47:12NekoNoNiaowbut it's really a shame that you need 12 years of experience in Jedi training to be able to understand if fully
22:48:12Bagdernah, it only takes 6 years ;-)
22:48:14bluebrother^where can I get that Jedi training to understand configure?
22:48:44NekoNoNiaowbluebrother^: actually, the first 6 years are spent looking for the Jedi master
22:49:17Bagderm4 and sh in a lovely mess
22:49:28bluebrother^eeks. I have that much other things to look for ... I guess this will have to wait for another 6 years :)
22:49:34hshahanyone know t0mas's email address?
22:50:09bluebrother^hshah: iirc it's on his wiki page.
22:51:16NekoNoNiaowand i hate it even more when you forgot to appropriately clean the project of previous configure stuff, and have to redo everything from scratch
22:51:23NekoNoNiaowwhich i just have to
22:51:45Bagderget a faster box ;-)
22:52:01NekoNoNiaowi can't, i want to test it on that special box :)
22:52:08hshahbluebrother^:where abouts would this page be?
22:52:38bluebrother^the rockbox users page. Look for his real name at the irc nick page :)
22:52:49BagderNekoNoNiaow: I know that, I sometimes debug my configure scripts on ancient machines and it usually makes go mad while waiting for slow configure scripts
22:52:52NekoNoNiaowi'm trying to build the gmini's binutils on mac os x
22:52:58BagderI meant I know that feeling
22:53:02hshahthe irc nick page?
22:53:25hshahk found it
22:53:31NekoNoNiaowbadger : you're my friend ;)
22:53:59Bagdermy program builds and runs on 40+ OS+CPU combos
22:54:23Bagderand you have it installed ;-)
22:54:36NekoNoNiaowI guess there's one optimization that configure needs : being able to minimize itself so it tests the minimum set of things to test for a particular program. I know it's supposed to be the case, but i can't help thinking it doesn't work
22:55:12Bagdereach configure script is setup by an author to do the tests it should
22:55:13NekoNoNiaowBagder: what is it ?
22:55:27NekoNoNiaowlet me check
22:56:27NekoNoNiaowit is indeed :) and i didn't knew you were the Daniel Stenberg :)
22:56:36Bagderthat's me
22:57:43NekoNoNiaowconfigure/autoconf would be real wonders if they also detected by themselves all the tests to be done, then they would fully deserve their "auto" prefix :)
22:58:39Bagderyes, that'll happen just after source code is generated automatically when you think about a cool program
22:58:43NekoNoNiaowbut that's a bit OT with this channel :)
22:59:00NekoNoNiaowBagder: i'd like that too :)
22:59:28webguest92I like how many programs check for a FORTRAN77 compiler
22:59:31NekoNoNiaowbut i think it may be done to some extent (not the source code generation ;) )
22:59:35webguest92Just because autoconf seems to like doing that
22:59:58Bagderrecent autoconf defaults have gone bananas if you ask me
22:59:58 Quit hshah ("CGI:IRC (EOF)")
23:00:03NekoNoNiaowwebguest92: hey, how true, i see that one all the time
23:00:08Bagderlike the fortran crap
23:00:40NekoNoNiaow"one never knows if a good old fortran77 compiler could not compile the c sources in case there was no c compiler" says the old adage
23:02:04NekoNoNiaowremind me to test this interesting hypothesis soon ;)
23:04:15NekoNoNiaowit might be that gcc is a subset of g77
23:04:22 Join Craig_ [0] (
23:04:25 Nick Craig_ is now known as Craig (
23:04:34NekoNoNiaowargh, still not compiling
23:07:25 Join LinusN [0] (
23:07:37amiconnhi again LinusN
23:07:50LinusNhi, was curious about Lear's latest commit
23:08:30amiconnI tested building sh-elf-gcc 3.3.6 with & without newlib on debian, both with the old native gcc 3.3.5 and the new 4.0.1.
23:08:32LinusNis he sure that using a 64 bit int solves the problem?
23:08:46amiconnResult: newlib workaround still works, without newlib it does not
23:09:07amiconnI wonder why it works without newlib on cygwin...
23:10:18LinusNyeah, that would be nice to know
23:10:43LinusNaccording to my calculations, a 32-bit int could count 27 hours worth of samples
23:11:05amiconnIt's nice that it does though. I suspected that it might have to do with the native gcc version, that's why I tried again after updating my debian
23:11:37amiconnIirc the newlib trick was necessary on cygwin too, but that was back when cygwin had gcc 3.3.3
23:11:44amiconnCurrent cygwin uses gcc 3.4.4
23:12:02 Join midk [0] (
23:12:23midkwelcome to zombocom!
23:12:28NekoNoNiaowok, i renounce, i think the gimni binutils miss a few changes from the gnu source tree, let's hack the makefile for good
23:12:55BagderNekoNoNiaow: I never liked the way they did the gmini binutils dev
23:13:11Bagderthey should've just made patches against public binutils versions
23:13:16NekoNoNiaowBagder: what do you mean ?
23:13:37Bagderthey way the added the whole binutils tree into their own cvs
23:13:51Bagdergee I type crappy
23:13:55amiconnHmm, it seems there is a problem with the peakmeter on iriver
23:14:03NekoNoNiaowthat said, they had to deal with the samsung source tree, merging with the official gnu one was not so obvious to do
23:14:30amiconnI have configured my peakmeters the same way across my units. Logarithmic, min = -50 dB, max = 0 dB
23:14:44NekoNoNiaowi can see why they added it : the binutils for calmrisc are not officially supported by gnu
23:14:53BagderI know
23:15:12Bagderthat's the whole point with their binutils work afaics
23:15:25amiconnHowever, the peakmeter on iriver is almost at 0 dB most of the time, and often indicates clipping, while the archos peakmeter does not, with the same mp3 files
23:15:33NekoNoNiaowwhat does gnu need to officially support it ? just a maintainer who will have a look at possible bugs and test new versions ?
23:15:36Bagdernow the gmini cvs can't keep up with the gnu one easily
23:15:53amiconnAfaik gnu says they don't support 16 bit archs
23:16:06midkamiconn, unrelated, but.. i don't know about the specs of the iriver or anything, but i wonder if it's capable of a spectrum analyzer? as a plugin or something.
23:17:12NekoNoNiaowamiconn: huh ? there's still plenty of them, most notably in the embedded market
23:17:31Bagderyes, but not with gcc support
23:19:09NekoNoNiaowtrue, because they made a stupid architecture decision long ago
23:19:21LinusNi don't think it's *that* stupid
23:19:28BagderI agree with LinusN
23:19:34Bagderthey had to draw the line somewhere
23:19:55BagderI'd say Archos was the stupid party here ;-)
23:20:07NekoNoNiaowif they had made gcc machine producing code independent of the underlying hardware, things would be much easier now for them
23:20:23BagderI doubt that
23:20:34NekoNoNiaowless efficient but more flexible for development
23:20:42Bagderfor <32 bits, yes
23:20:46LinusNefficiency is important
23:21:09NekoNoNiaowLinusN: i don't know, i spend more time writing code than compiling it :)
23:21:32 Quit Coldtoast ("Peace and Protection 4.22")
23:21:32Bagderyes, but even more time should be spent _running_ the code
23:22:02NekoNoNiaowsure, but this should not impact the generated code
23:22:30Bagderwell yes, and Rockbox should work on all players
23:22:56NekoNoNiaowbut of course, i'm just talking, unless i fork gcc to prove i'm right that will just be words ;)
23:23:00Bagderwe all need to make decisions to get the right focus
23:24:19NekoNoNiaowsometimes, the thought surfaces, but then i remember how hard it is to do much simpler things and i forget it instantly :)
23:26:14 Join ashridah [0] (
23:26:38amiconnHmm, it isn't that clear. I think I've read somewhere that gcc doesn't support <32 bit architectures, but there are some officially supported 16 bit archs...
23:27:38amiconnH8 is one example
23:29:14 Nick webguest92 is now known as rasher (
23:29:26rasherI'm confused by the islenska language update
23:29:46rasherRunning uplang on it generates a gazillion ### comments
23:30:20BagderI was puzzled by his comment
23:30:26Bagder"updating with the english file available on the site."
23:30:36Bagderwhat file is that?
23:30:53 Quit HCl (Read error: 104 (Connection reset by peer))
23:31:10rasherHm, there's a link to on the HowtoUpdateLangfile wiki page
23:31:28rasherShould be a checkout link
23:31:33rasherI fix.
23:32:55NekoNoNiaownth recompilation of the evenig
23:33:07NekoNoNiaowand last one before going to bed
23:33:18Bagdermmmmm bed
23:33:30NekoNoNiaowgive me the email of that stallman guy, we need to talk
23:33:51BagderI don't think he's that involved with those things these days
23:34:17Bagderbut he seems to like to talk ;-)
23:34:26NekoNoNiaowBagder: he was involved in the architecture decision, that's makes him a target of my anger ;)
23:34:37 Part LinusN
23:36:06NekoNoNiaowof course, i'm joking, i globally share his ideas on software
23:37:20Bagderrasher: good work on the lang files front
23:37:39 Join muesli- [0] (
23:38:08NekoNoNiaowis japanese supported by the way ?
23:38:35rasherBagder: I see Language patches are assigned to quelsaruk - is this automatic?
23:38:50rasherAh, thought so
23:38:54Bagderuse the admin link
23:40:24NekoNoNiaowomg, i screwed everything
23:40:28*rasher stabs sf
23:41:32 Join HCl [0] (
23:41:45rasherIt seems that randomly, it'll forget who I am and deny access
23:44:32*NekoNoNiaow just realizes he has done everything for nothing
23:44:52NekoNoNiaow"Note : there's currently a bug with the binutils (tex)info install for CalmRISC16. It can safely be skipped with make -k install"
23:44:55rasherCan I change the auto-assignment from quelsaruk to me, or should I leave him?
23:44:59NekoNoNiaowi'm going to kill myself
23:45:09rasherOh dear
23:45:31NekoNoNiaowno, i'm going to kill someone else
23:45:35NekoNoNiaowmuch better ;)

Previous day | Next day