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

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

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

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

#rockbox log for 2006-08-03

00:01:55sharpeit's kind of funny seeing other people's desktops.
00:02:25 Quit hannesd (No route to host)
00:02:34dionoeaany idea what could cause a bmp->c file transformation to fail ?
00:02:42dionoeai get make[4]: *** No rule to make target `solitaire_numbers.8x10x16.bmp', needed by `/home/dionoea/rockbox/rockbox/builds/h3x0-sim/apps/plugins/bitmaps/native/solitaire_numbers.8x10x16.c'. Stop.
00:02:45midkaysharpe wants to see mine, right?
00:02:47sharpecorrupt bmp?
00:03:03dionoeais the gimp known to save corrupt bmps ?
00:03:10sharpebmp not actually there?
00:03:16dionoeait's there
00:03:28sharpemidkay: sure.
00:03:35*dionoea double checks the name
00:03:43Bagderthat's make complaining, it has no idea if the bmp is fine or not
00:03:43dionoeayep, filename seems ok
00:03:58dionoeawell i added the bmp in the SOURCES file
00:04:17Bagdertry removing the dep file
00:04:36 Quit petur ("here today, gone tomorrow")
00:04:36dionoeamake clean ?
00:04:56dionoeadid help :(
00:05:12 Join gursikh [0] (
00:05:15dionoeai also tried runing configure again
00:05:22 Part gursikh
00:05:27 Join Kalthare [0] (
00:05:46dionoeaor maybe there's a hidden typo in "#if defined( HAVE_LCD_COLOR )" ... but i can't find it
00:06:10dionoeait should read "didn't help". sorry :/
00:08:24KalthareIs the pwd correct?
00:08:45dionoeais that question for me ?
00:08:56KalthareIt looks like make can't find the file. Maybe it's in the wrong dir?
00:08:59Mikachuyou mean cwd, pwd is a command to print the cwd
00:09:07Kalthareokay, whichever
00:09:12dionoeawell it's in the same dir as all the other plugin bitmaps which work fine
00:10:04dionoeahmmm ... nevermind
00:10:13*dionoea should get some sleep
00:10:25linuxstbI'm sure it will fix itself tomorrow...
00:10:47Kalthareeh, sorry for being unhelpful.
00:10:48dionoeawell the numbers bitmap didn't have a color version ... rha
00:11:07 Part Vasu
00:11:25linuxstbdionoea: Did it need a colour version? i.e. can't you just draw the mono bitmaps in different colours?
00:12:27dionoeawell maybe you want elite graphics with some nice 3D color stuff ...
00:12:42dionoea(that's hypothetical)
00:12:48Bagderelite graphics in Rockbox? are you mad!
00:13:42 Quit bluebrother ("Leaving")
00:13:56KalthareIs it possible to make Rockbox use voice samples to announce song names when they start playing, or is it just for the menus?
00:14:53Mikachuyou would have to create a voice sample for every song...
00:15:20linuxstbYou mean you want a DJ introducing your songs?
00:15:48KalthareMore or less, yeah. Just when I'm using shuffle mode.
00:15:54Mikachuor possibly port a very simple tts engine like espeak
00:15:55linuxstb(btw, I think the answer's no).
00:16:26linuxstbYou can make Rockbox speak filenames in the file browser, but it can't announce tracks as they are playing.
00:16:44KalthareMmm. Maybe I'll try and add it.
00:16:44linuxstb(although it's possible, if someone coded it)
00:16:53Mikachulinuxstb: how does that work?
00:17:04Mikachupregenerated samples too?
00:17:23linuxstbI've never used it, but yes, they are pre-generated files.
00:17:37KalthareMikachu: yeah, they're MP3s. The instructions are on the wiki.
00:18:23*linuxstb wonders if blind users would like a talking WPS
00:19:00dionoeaor beeps when the battery is runing out
00:19:05Mikachu"you have paused the music"
00:19:08 Join freqmod_n [0] (
00:19:22linuxstbI know I've paused the music, I pressed pause...
00:19:35Mikachuyes, and the joke is that it is quite audible too
00:19:35Bagder"you have listened 2 minutes and 23 seconds.. .24 seconds... 25 seconds..."
00:19:54linuxstbtalking peakmeters...
00:20:12dionoeayay ! no build errors
00:20:20MikachuProgramming Without Thinking
00:20:39 Quit mikearthur (Remote closed the connection)
00:20:55Mikachuthat was in reference to talking peakmeters, not dionoea
00:21:04 Join mikearthur [0] (
00:21:42dionoeahum ... the CVS build didn't include my last commit ... might be the reason why there were no errors
00:22:03sharpejust maybe.
00:22:15Mikachulinuxstb: i have a keybind on my computer to make espeak read the current title out loud, but i've never used it.. :)
00:22:36 Quit mikearthur (Remote closed the connection)
00:22:54 Join mikearthur [0] (
00:22:59KalthareWell, gotta go.. busy day. I'll see if I can figure anything out for that DJ thing.
00:23:25BagderKalthare: that's not gonna be an easy task
00:23:34Bagderbut indeed a challange
00:23:44KalthareBagder: oh?
00:24:54 Join TCK [0] (
00:25:23 Quit freqmod (Read error: 113 (No route to host))
00:26:59 Join akaias [0] (
00:28:57KalthareHm, installing _another_ ARM toolchain. Woo.
00:29:10 Quit PyromancerX (Read error: 110 (Connection timed out))
00:29:15Bagdereveryone needs a handful!
00:29:22Mikachui think i have 2 or 3 too
00:29:58 Quit mikearthur (Remote closed the connection)
00:30:08KalthareI think I have, like.. 3 or 4. Plus this one.
00:32:02Bagderwhy stop at that when there's sh, m68k and more too! ;-)
00:33:22sharpei'll be like midkay. hahah.
00:33:34midkaystop that. stop ending your laughs with "h".
00:33:35 Join Pyromancer [0] (
00:33:41midkay"hah" or "heheheheh".
00:33:54sharpei never use "heheheheh"
00:34:07midkayexamples. now stop spamming the channel.
00:34:54 Quit freqmod_n (Remote closed the connection)
00:34:59KalthareHm, true. I _do_ have a Dreamcast and a pre-ARM Palm that I could theoretically develop for. ...But, no.
00:35:38Mikachumy other one is for gba
00:35:56 Quit pabs (Read error: 113 (No route to host))
00:36:55 Join pabs [0] (
00:37:00KalthareMine are for Palm, PSP (That's ARM, right? I haven't slept much lately), GP32 and GP2X.
00:37:47Mikachuwouldn't one be enough though? :)
00:37:54Mikachuor do they use slightly different binary formats?
00:37:56 Join ]RowaN[ [0] (
00:38:06Bagderif you build for linux you want a different one
00:38:11]RowaN[am i going mad or is there no longer a h120 build on the bleeding edge page?
00:38:27]RowaN[ah now its there
00:38:31*]RowaN[ scratches his head
00:38:34Bagderyou're mad
00:38:39sharpethe first one.
00:39:50 Part ]RowaN[
00:40:03KalthareMikachu: They all use different binary formats. Also, I used automated installers for all of them that pulled the devkits too.
00:40:47KalthareActually, I tell a lie. I think they mostly use ELF, but PSP and GP32 have them wrapped in custom formats.
00:41:00sharpelies aren't good.
00:41:00Mikachumy gba one is called arm-thumb-elf, not arm-elf
00:41:05Mikachuand i think i have one arm-agb-elf too
00:43:29Kaltharearm-thumb.. I think the GP32 used that too. Been a while since I used that one, and I don't even remember where it is now.
00:45:03 Quit klrspz ()
00:45:15KalthareHm, apparently PSP doesn't use ARM. Like I said −− not enough sleep.
00:45:56BHSPitMonkeyPSP is MIPS
00:46:52 Nick lds_ is now known as lodesi (
00:47:10Mikachudoes the ipod also support thumb code?
00:47:18KalthareAnyway, I gotta get going before I start spreading more misinformation. Seeyas.
00:47:30 Quit Kalthare ("Download Gaim:")
00:47:34Mikachumaybe not as much reason to use it on ipod as on gba
00:48:26sharpea cooler sounding closing line would be, "Anyway, I've got to go get information before i spread more misinformation."
00:48:38Mikachuno, not really
00:48:50linuxstbI don't think Apple uses it, but other PortalPlayer firmware seems to - e.g. the Sansa firmware has some thumb code, at least in the bootloader.
00:48:57sharpenot for what he was talking about, but it sounds cooler.
00:49:09Bagderand so does the h10
00:49:23linuxstbBagder: Just in the bootloader, or is it in the main firmware as well?
00:49:43BagderI don't know, I only read Barry's description about the bootloader
00:51:07linuxstbOK. I think I looked at the H10 bootloader as well, and saw a very early switch to thumb mode, the same as the Sansa.
00:51:21linuxstbs/I think//
00:51:47Bagderyeah, I suspect large parts of the bootloader is from PP and thus similar between players
00:52:19Bagdersaw my latest finding about where the mi4 file is on the Sansa? when I "upgrade" to a new mi4, it gets moved
00:52:32Bagderand is to be found on the second partition of the device
00:52:39Bagderwith a little fluff around
00:52:56linuxstbYes, I read that... At least it confirms it's not flashed.
00:53:11Bagderwell, not flashed in that sense anyway
00:53:30linuxstbNo... A bit similar to the ipods.
00:54:19linuxstbHave you had chance to experiement with the LCD framebuffer?
00:55:03Bagderthe bootloader doesn't even yet build properly, but when it does it should be swiftly done
00:58:58 Join philth|work [0] (
01:02:50 Join Febs [0] (
01:04:35***Saving seen data "./dancer.seen"
01:07:36 Quit obo ("bye")
01:14:54 Join shammy [0] (
01:16:28shammyHi, I just downloaded the ipod_fc source, tried compiling and instantly got a few hundred errors... any idea what I should do?
01:17:02desrtshammy; i'm guessing you should read directions more carefully
01:17:25desrtshammy; if you scroll up through the errors you will likely see something about a missing .h file. you're likely missing a dependency
01:17:49desrtshammy; but i know nothing about rockbox. just assuming using knowledge from unrelated projects
01:18:01 Join mikearthur [0] (
01:18:05 Quit mikearthur (Remote closed the connection)
01:18:08shammydesrt, perhaps i'm using an outdated wiki then?
01:18:15linuxstbThere are no dependencies - apart from the standard C library.
01:18:51linuxstbYou're probably just missing the base development package. Which Linux distro are you using?
01:18:51desrtshammy; what's the first error message thrown?
01:19:03 Join mikearthur [0] (
01:19:03desrtyou may need to install a -dev version of your libc
01:19:50Mikachuthe makefile only runs gcc -g ipod_fw.c -o ipod_fw
01:20:32shammydesrt, amongst the errors i did find a block of missing .h files
01:21:09shammylinuxstb, ubuntu
01:21:10Mikachuit only includes stdio, stdlib, string, errno and unistd
01:21:11desrtshammy; right. that's your real problem.
01:21:24Mikachuif you have gcc, you should have those...
01:21:39desrtMikachu; gcc usually ships in a different package from the libc-dev packages
01:21:50 Quit mikearthur (Remote closed the connection)
01:22:13shammyMikachu, i have gcc 4
01:22:51linuxstbI've forgotten the name of the generic development package you can install, but try apt-get install libc6-dev
01:23:56shammyit's installing, thank you
01:25:05 Quit dan_a (Read error: 110 (Connection timed out))
01:25:30shammycompiled without error, thanks very much everyone
01:25:39desrtshammy; you may want to install build-essential
01:25:51linuxstbAh, that's the name I was trying to remember...
01:26:05desrtshammy; it pulls in the compiler, libc dev files, make, etc... stuff you'll need to compile just about anything
01:26:14shammyalright I will do so, thanks again
01:27:19 Quit Kohlrabi ("Quit")
01:29:20 Join xoder [0] (
01:29:43 Quit xoder (Client Quit)
01:31:21 Quit mantono ("natti natti")
01:32:15 Quit lodesi ("Parti")
01:36:14 Join mikearthur [0] (
01:38:02 Quit mikearthur (Remote closed the connection)
01:38:09 Quit shammy ("Leaving")
01:38:21 Join mikearthur [0] (
01:38:24 Join aegray [0] (n=aegray@
01:40:26 Quit mikearthur (Remote closed the connection)
01:44:17 Join mikearthur [0] (
01:44:33 Quit mikearthur (Remote closed the connection)
01:44:50 Join BloodySorcerer [0] (
01:48:22 Join mikearthur [0] (
01:48:32 Quit mikearthur (Read error: 54 (Connection reset by peer))
01:50:10 Join mikearthur [0] (
01:50:11 Quit mikearthur (Remote closed the connection)
01:52:26*BHSPitMonkey parties
01:53:44*BloodySorcerer mopes
01:58:58 Join xmixahlx [0] (n=xmixahlx@
02:01:08 Join tournedo [0] (
02:01:46 Join [TCK] [0] (
02:02:17 Quit Daishi ("Client exiting...")
02:03:35 Quit akaias (Connection timed out)
02:05:10 Join akaias [0] (
02:06:32 Part pixelma
02:06:58 Quit TCK (Read error: 60 (Operation timed out))
02:18:50scorcheBagder: nice handwriting =)
02:22:58*BloodySorcerer twiddles his thumb(s)
02:23:00 Quit xmixahlx ("blah blah blah")
02:23:07Mikachuhow do you twiddle one thumb?
02:23:29tournedohi everybody !!!
02:23:36BloodySorcerer...i have no idea.
02:23:37BloodySorcererbut i am :D
02:23:53tournedohave a question about rockbox in ipod nano
02:23:59tournedocan play wma
02:24:03BloodySorcereractually it is possible to twiddle one thumb
02:24:17tournedoi play ogg
02:24:33tournedobut the wma shows like music but cant play
02:24:46linuxstbtournedo: No, Rockbox doesn't support wma yet.
02:25:05tournedothx man
02:25:27tournedoi read something in the codecs section of the forum
02:25:36BloodySorcererlinuxstb, you're a saint. my answer woulda been "RTFFAQ!"
02:25:40tournedobut thing maybe someone have codec
02:25:57tournedooks bloody
02:26:12BloodySorcererhence why i chose not to respond
02:26:18BloodySorcererbecause i'm just a rude asshat :)
02:26:37tournedoi understand
02:26:39Mikachui was going to answer but i was drinking tea
02:26:52linuxstbSee. we're a friendly bunch...
02:27:05tournedobut in some time maybe can play wma?
02:27:07tournedoor never
02:27:16tournedomaybe new versions?
02:27:29tournedonot have too many music in wma
02:27:33BloodySorcererlinuxstb, you refer to yourself in the plural? :)
02:28:01linuxstbtournedo: Yes, maybe in the future - someone has started work on it.
02:28:25tournedoso wait and update :)
02:28:49tournedoi use ipodlinux but music and mpd dont work fine for me
02:28:59tournedorockbox like a lot
02:29:07tournedosounds great
02:29:11tournedoand simple
02:29:13tournedois nice
02:29:24tournedowell thx for the answer
02:29:35tournedoand sorry my bad english im spanish
02:29:42BloodySorcereryou're english is quite good
02:29:47BloodySorcerermuch better than my spanish at least :D
02:30:04linuxstbYou mean "Your English is quite good."
02:30:14BloodySorcerer(spanish incapability is compounded by absence of accents)
02:30:19BloodySorcererlinuxstb, good catch O_o
02:30:27BloodySorcereri normally don't make mistakes like that
02:30:37*BloodySorcerer hangs self in effigy
02:35:27 Join ^jhMikeS^ [0] (
02:41:46 Join Paul_The_Nerd [0] (
02:43:03 Part Paul_The_Nerd
02:45:43 Join actionshrimp [0] (
02:48:44falconfoxfor some reason, i can't get my ipod color to use the ID3 database. i have "show files" = "ID3 Database" then in "tag cache" i first clicked "initialize now" then i clicked "update now"
02:48:49falconfoxand it doesn't seem to be doing anything
02:49:38linuxstbHave you shut down your ipod, and then restarted it?
02:50:20 Join spiorf [0] (
02:51:33falconfoxyeah i just did that
02:51:41 Quit jhMikeS (Read error: 110 (Connection timed out))
02:52:13falconfoxi know i have followed the manual exactly, i just can't figure this out
02:52:53falconfoxwhen i click "update now" shouldn't the hdd begin to spin?
02:56:40falconfoxit doens't matter if the ipod is plugged in or not, right?
02:56:51 Quit Nibbler ("Leaving")
03:01:19 Join BigMac [0] (n=dummeska@
03:01:28BigMachello poppets
03:04:37***Saving seen data "./dancer.seen"
03:06:20sharpethat reminds me of someone i know.
03:08:34 Join Nibbler [0] (
03:09:24 Join webguest59 [0] (
03:09:40 Join infamis [0] (
03:11:42infamisanyone....I want to help debugging playback-related code, but where do I start (source-wise?)
03:12:40infamisplayback.c & pcmbuf.c? but where?
03:12:46sharpeyeah, but a person i know who says it...
03:13:14BigMacHMM...MR T?
03:14:23Febsinfamis, probably the best person to ask is Slasheri.
03:14:45FebsOr lostlogic, though he hasn't been around much recently.
03:16:03BigMacFebs:are you a dev or just a forum mod?
03:16:32 Quit spiorf ("Read error: 104 (Connection reset by peer)")
03:16:36FebsI consider myself a contributor.
03:17:02BigMacfor what models?
03:17:20sharpenope. not mr. t.
03:17:22FebsI do not have the technical skills to be a developer, but the project is of great interest to me, so I have contributed my time writing documentation.
03:17:56BigMaccoool cool
03:18:07BigMacim trying to learn c so i can do my part
03:18:42infamisI just wanna jump into debugging but playback.c is so huuuuugeeeeeeeee!
03:18:56infamisand it's like, damn, where do I start??!!
03:20:23infamisot: anybody run gentoo? gentoo > ubuntu?
03:20:41FebsI'm not sure what time zone you're in, but many of the key developers are in Europe, so they're not usually around at this time of night.
03:21:00infamisyeah, I'm's usually busy around here at 4am (for me)
03:21:07Galoisare you looking for a distro flamewar or what?
03:21:10BigMacinfamis: im guessing linuxstb does
03:21:41infamisGalois, basically...just trying to decide getting back into linux after not dealing with it since '01
03:21:55infamisfeel lot's changed
03:22:04Galois in some ways it's a lot better and in some ways it has regressed
03:22:17webguest59regressed how?
03:22:32infamisdumb-downed?? ;)
03:22:34 Quit philth|work ("CGI:IRC (EOF)")
03:22:45GaloisI would say the technology has improved dramatically but the political situation is worse than before (and I don't mean internal politics here, I mean corporate politics)
03:23:39Galoisdumbed down is not quite the right word for it. Some of this new stuff is so dumbed down that I can't even figure it out
03:24:13Galoisfor example, any vaguely modern distribution will autodetect hardware on the fly for you now
03:24:40infamisthat's bad?
03:24:45Galoismost of the time it works great, but if you want to change anything in the autodetection, perhaps because it doesn't work or because you don't want it doing that, it's hard to change
03:25:20Galoisthe complexity of the underlying technology has grown even as the user visible part is supposed to be easier
03:25:58infamisyeah I remember when x11 took all day to build/install/set-up
03:26:19Galoisthe other day I popped in my external usb drive and the automounter conveniently mounted for me /media/disk /media/disk-1 /media/disk-2 /media/disk-3 etc. for each partition on the drive
03:26:41infamislol, damn those devs doing shit for you ;)
03:26:43Galoisand, to be honest, I don't know how to turn that off even if I wanted to (though, I haven't tried very hard)
03:27:30BigMacso is linux easy to get used to?
03:27:40BigMaclike going from windows to linux?
03:27:48GaloisI'm the wrong person to ask, I've had 11 years to get used to it
03:27:48infamisyeah I just feel getting back into linux will get me back into 'development mode'...messing with windows all these years has made me very lazy....
03:28:47Galoisthe legal situation is worse though. 10 years ago there was not really any such thing as illegal software categories
03:28:48BigMacbecause im building a computer and i feel that linux would be the way to go because it is free
03:29:12Galoisnow there are certain types of software that are illegal to make free, like dvd players
03:29:23 Quit chendo (Read error: 110 (Connection timed out))
03:29:33infamisyea I agree Galois
03:29:56 Join chendo [0] (
03:29:58infamisBigMac, be prepared to get frustrated like all hell when stuff doesn't work, even after you've tried hours to fix/install
03:30:03Galoisit's quite possible to build a computer for linux. You have to pay attention to make sure you buy hardware that has good drivers.
03:30:34BigMacthen not for me
03:30:48infamis...but it's a good experience though, that is if you're willing to learn
03:31:24BigMacill prolly hold out for the new windows and mac os' and see which better suit me
03:31:32infamisI don't know how old you are, but when looking for a job, employers find it attractive regarding your diversity with OSs & programming languages
03:31:57infamis...or even if you are 'in the field'
03:32:26infamisI mean; that is, if you're "in the field"
03:32:30BigMacyah im not old enough for the job yet...but i want to be a robotic engineer
03:32:36BigMacor some kind of programmer
03:33:09BigMacso once i get enough cash ill buy a ibook and vista desktop
03:33:10Galoisprogrammers basically live or die on free software
03:33:17BigMacand eventually build a linux pc
03:33:30infamisor you could go dual-boot
03:33:43Galoisbecause free software is written by programmers without customer pressure, so the authors tend to write the program to suit their own biases
03:33:53Galoiswhich, of course, are programmers biases
03:33:58BigMacyah i really want bootcamp for my mac
03:34:37infamisyeah, I tend to do things better (e.g., not the 'quick' or 'hacked' way of writing code) when it's not job-related
03:34:57Galoisit might even be fair to say that you should be concerned if you want to be a programmer and you find out that you hate free software
03:35:22GaloisI mean, even if it is frustrating, most of the time, a programmer type can figure it out
03:36:01infamisprogramming is all about compromise in my eyes...nothing ever works the way you want it to
03:36:20infamisor in other words, "there's always something"
03:37:07infamisjust testing color on the web client....
03:37:12BigMacyah except for simple flac conversionXD
03:37:21BigMaci cant find anything easy and good
03:37:35BigMaccant get eac to work
03:37:58Galoisuh: find -iname \*.flac -exec ogg123 {} \;
03:38:07Galoisby ogg123 I meant oggenc
03:38:31BigMacfor windows, and is this in the cmd
03:38:35infamiseac's worked for me, some drives it slows them down (a hard speed cap, not just because it's in secure mode, but a HARD cap)
03:38:51GaloisI'm confused, do you have a CD and want flac, or do you have flac and want something else?
03:39:02BigMaccd ripped to flac
03:39:12BigMacriptastic is awesome
03:39:21BigMacbut not free
03:39:22infamisfoobar with oggenc2.exe somewhere is what I do for transcodes
03:39:29Galoiswhat the hell is wrong with CDex?
03:39:43Galoisit's windows software, it's free, it does flac's just not EAC ;)
03:40:28BigMacill try CDex
03:40:31BigMacis it easy?
03:40:42BigMacand does it tag its songs buy itself?
03:40:58Galoisit used to use freedb for tags
03:40:59FebsI think that CDex's stock has risen now that a new version is out. It had been so long since it had been updated that its encoders had really fallen out of date.
03:43:40BigMacwow i really like this object dock program
03:43:49BigMaci wish their was a shortcut to show it
03:48:42BigMacanyone know of a good online c tutorial. one youve used and reccomend
03:49:01 Quit Sinbios (Read error: 104 (Connection reset by peer))
03:52:39 Quit webguest59 ("CGI:IRC (EOF)")
03:53:39 Quit tournedo ("Leaving")
03:54:45 Join niskel [0] (
03:54:50 Join Sinbios [0] (
03:56:42 Join JdGordon [0] (
03:57:27BigMacheya jd
03:57:35BigMacwatch this nights house
03:58:16 Quit tvelocity ("Ex-Chat")
03:58:38JdGordonBigMac: you mean last nights?
03:59:57 Join gursikh_ [0] (
04:00:14 Part gursikh_
04:00:16BigMacyou live in au i forgot
04:00:41BigMacthe one with the bubble girl?
04:01:31 Join ehour [0] (
04:03:47JdGordoni dont know, i taped it, ill watch it tonight
04:04:03 Join gursikh_ [0] (
04:04:06ehourthe jay leno tonigth show!
04:04:33 Nick gursikh_ is now known as khalsa (
04:04:37 Part khalsa
04:06:06 Join gursikh_ [0] (
04:07:12infamisanybody use an ide with the rb source?
04:07:48sharpemost people do use an ide.
04:07:53BigMacJdGordon: very good episode
04:07:56 Part gursikh_
04:08:24BigMacsharpe: accept me because i have no idea what it is of course:-)
04:08:32sharpean ide?
04:09:10BigMacide= indestructible deficating emu? otherwise i have no idea
04:09:24sharpeintegrated development environment
04:09:25scorchethen look it up..
04:09:38*scorche coughs
04:09:46BigMacscorche: i was just saying i don't use it
04:09:58BigMacsharpe: like cygwin and vmware
04:10:03scorcheand that you dont know what it is =)
04:10:17sharpethose are development environments
04:10:40BigMacdomo arrigato
04:10:55BigMacguys im picking up that book you told me to get tomorrow
04:11:15sharpethe c programming one?
04:11:49sharpeokay, good, because i have no recollection of reccomending any books.
04:11:57BigMacthen ill be on my way to accomplishing something
04:12:17BigMacyah the oreilly handbook you reccomended
04:12:38BigMacand i may pick up ther perl one as well because that is something else i wanted to get around to learning
04:13:11 Quit [TCK] (Client Quit)
04:13:13sharpeall i remember is talking about my c# handbook...
04:13:36scorchei think he is referring to the o'reilly quickguide or whatever
04:13:41BigMacthe pocket reference?
04:14:25sharpei bought the thing last year, haven't had a need for it yet.
04:14:45BigMaci heard it was very rich with info?
04:15:08sharpeit is, i just haven't done any c# related things lately.
04:15:24scorcheBigMac: keep in mind that it is a quick guide of will not teach you how to has a ton of info, but for whatever you dont understand, you can go online
04:15:53BigMacyh thats what ive heard. so im looking for a good free c tutorial
04:15:54 Quit BloodySorcerer (Read error: 104 (Connection reset by peer))
04:16:06sharpegood ole google...
04:16:19 Join BloodySorcerer [0] (
04:16:20 Join gursikh_ [0] (
04:16:22BigMaci know i searched google
04:16:28BloodySorcerergoogle ftw
04:16:44BigMaci wanted to ask around to seee if there were any you particularly reccomended
04:17:17BigMacis it good?
04:17:25sharpethat's the only one i recall.
04:18:23BloodySorcererwhat, ways to learn C?
04:18:35BloodySorcererlook at source code
04:18:40BloodySorcererthen google everything you don't know
04:19:12BigMacwell bllody, that would be meaning id be googling every line
04:19:18BigMaci know zero c
04:19:24 Part gursikh_
04:19:26BigMacthe only language i know is asm
04:19:31BloodySorcererdo you know *any* languages?
04:19:36BloodySorcererdamn O_O
04:19:46BloodySorcererthats hardcore.
04:20:14BigMaci am trying to learn languages
04:20:17BigMacim still a kid man
04:20:40*scorche coughs
04:20:45BloodySorcererdefine "kid"
04:20:53BloodySorcererscorche, want a cough drop?
04:20:54*sharpe stays quiet.
04:21:02BigMacunder 18
04:21:11BloodySorcererthen i'm a kid too
04:21:16BloodySorcererwhat were we discussing?
04:21:39BigMacwell im trying to take the initiative to learn. that is the first step to self imporvement
04:22:02BigMaci thought you were making fun of me for saying i learned asm
04:22:22BloodySorcereri'm impressed
04:22:34BloodySorcererasm makes me sad
04:22:58BloodySorcererits so ludicrously difficult compared to any other language i've learned/used looks like a good place to start
04:23:20BigMacoh well i don't know if its any different then regular, buts its assembly for the ti calculators
04:23:36BloodySorcererooh fun stuff :)
04:23:38BigMacyah i like that there is quizzes as well
04:24:03BloodySorcererwell imma go eat and watch TV and do other fat arse related activities
04:24:09BloodySorcererso have fun learning C! :)
04:24:13BHSPitMonkeyhow old are ya, BigMac ?
04:24:31BloodySorcerer(once you know C, C++, java, or php (to some extent) its a breeze to learn any other, fyi)
04:25:12BigMacyah im learning c and perl this summer
04:25:15BigMacthats my goal
04:25:34niskelThe kerrnigan & ritchie book is good
04:25:50BigMacbut i want to learn c++(which is supposedly close to c) and java and php during the year
04:26:00BigMacso then i know most of the core languages
04:26:06BHSPitMonkeyhey, I'm 16 too
04:26:07scorchenot java >_>
04:26:12BHSPitMonkeyfor 2.5 more hours!
04:26:40BigMacBHSPitMonkey: happy birthday in 2 hours!
04:26:52BigMacscorche: why not?
04:26:59 Join javkrei [0] (
04:28:10sharpeBigMac: he's have a very troubled past with java.
04:28:24sharpehad. and he doesn't like to speak about it...
04:28:36BigMacscorche:show me on the doll were java touched you:-)
04:30:16BigMacthere is only one WHERE and WHERE does the s go
04:31:29sharpethe where replaces the were.
04:31:39BigMacand the s?
04:32:00BigMacyour like my spell check.
04:32:05BigMacI like it:-)
04:32:23BigMacmunkey wolked to teh stor
04:33:02scorches/I am an idiot./munkey wolked to teh stor
04:33:31BigMacscorche:do you want me to get java up in here
04:33:52sharpescorche just doesn't like coffee.
04:34:44BigMacso i don't lik searing my flesh on a flame, but that doesn't mean i don't like scorche
04:36:37BigMacany of yall play yahoo pool
04:40:51sharpei would like to take this moment and note the dead silence, that i usually instill by saying something and would like everyone to know it wasn't me this time.
04:40:53BHSPitMonkey'course we do
04:40:56BHSPitMonkeyit's the sport of champions
04:42:04BigMacBHSPitMonkey: you do?
04:42:21BHSPitMonkeyhaven't tried it in linux I don't think
04:42:40BigMacoh ok
04:42:44javkreihi... somebody could explain why midi reproduction doesn't work in the rockbox simulator?
04:43:17BHSPitMonkeyI'm not sure that it even works in native rockbox.
04:43:44javkreithere is a plugin for it
04:43:57sharpedoes it work?
04:44:01javkreibut for some reason if it's a simulator build
04:44:10javkreiit doesn't work
04:44:23javkreisince I don't own yet a player with rockbox
04:44:45javkrei(although I have bought one, I'm waiting for it)
04:45:18javkreiI was wondering if it works in the players themselves
04:45:32javkreiI bought an iriver H120
04:46:19javkreihas anyone tride?
04:46:41BHSPitMonkeyI know there's a plugin, I'm just not confident that it works
04:48:05javkreimhhh, for what I've been reading, it worked... in the past at the least
04:48:26javkreibut for some reason it doesn't reproduce in the simulator
04:49:24javkrei#ifndef SIMULATOR
04:49:40javkrei rb->pcm_play_stop();
04:49:58javkreiand so omn
04:50:08javkreiso... it's disabled for the simulator
04:50:34javkreibut not for the players... I'm programming a few plugins
04:51:08javkreiand I'd like to make use of the midi library
04:51:47javkreiand it would be very handy if midi reproduction worked in the simulator
04:54:39 Quit sharpe ("Leaving")
04:55:45 Quit lukaswayne9 ("Ex-Chat")
04:58:33 Join lukaswayne9 [0] (
04:59:10 Quit niskel (Remote closed the connection)
05:00:06falconfoxwhy won't my ipod ID3 database work? i have "id3 database" enabled and i have forced update about 30 minutes ago, but still nothing when i select "artist"
05:01:22 Join Rob2222 [0] (
05:01:24 Quit BigMac (Read error: 110 (Connection timed out))
05:01:49falconfoxdoes my music on my ipod have to be in a specific location?
05:04:39***Saving seen data "./dancer.seen"
05:05:22 Join michaelconner [0] (
05:05:56michaelconnerhi... i noticed that rolo is supposedly working for ipods now
05:06:16falconfoxdoes music have to be inside a specific folder/
05:06:21 Quit lukaswayne9 (Remote closed the connection)
05:06:42michaelconnerhow would i get an apple.ipod firmware file to boot into via rolo?
05:07:54 Join lukaswayne9 [0] (
05:17:51 Quit Rob2222_ (Read error: 110 (Connection timed out))
05:17:57 Join safetydan [0] (
05:19:04 Quit michaelconner ("CGI:IRC")
05:22:08safetydanjavkrei: you may need to get the midi patch files for the midi playback to work in the sim
05:22:22safetydanthough if it's #ifdef'd out then you're stuck
05:23:01javkreiI've downloaded the patch...
05:23:09javkreiit's the #ifdefs...
05:23:38javkreiI'm building it now... with the ifdefs removed... I'll see how that goes
05:23:45javkreiI doubt it'll work
05:24:52safetydanThe #ifdef's that I can see probably shouldn't be there as I can't think of a reason they won't work in the sim
05:24:59safetydanI could be wrong though as I don't have the source in front of me
05:25:15falconfoxjavkrei: when i transfer music to my ipod, does it have to go to a specific location/
05:25:15javkreimhh, that's what I thought
05:25:16safetydan(well, not all of it anyway)
05:25:24javkreiI'll tell you in a few minutes
05:25:44falconfoxjavkrei: what is the point of waiting
05:25:44safetydanfalconfox, put it wherever you want. Rockbox doesn't care
05:26:02falconfoxsafetydan: the reason i ask is that my ID3 database is not working correctly
05:26:12falconfoxand i am using yesterday's rockbox build
05:26:17javkreifalconfox: sorry, I was talking to safetydan
05:26:23falconfoxjavkrei: oh
05:26:37safetydanfalconfox, sorry, tagcache isn't something I use so I'm probably not going to be much help
05:27:01javkreifalconfox: I can't be of help either, sorry... I don't have a player...
05:27:08safetydanfalconfox, try the forums
05:27:14javkreiI'm waiting for it to arrive
05:27:34safetydanor at least say in what way that the id3 database doesn't work
05:27:56falconfoxsafetydan: what do you use instead?
05:28:18safetydanthe file view
05:28:33falconfoxsafetydan: well i select "ID3 database" for "show files" . then i force tag cache update
05:28:53falconfoxthen the hdd spins for about 30 seconds , and then nothing, even if i restart
05:29:09 Quit ehour ()
05:29:29safetydanfalconfox, have you read through this?
05:30:11falconfoxif that is similiar to the PDF manual, then yes
05:30:19falconfoxi have followed the directions word for word
05:31:02falconfoxalright ... i just hit "initialize now" then it said "updating in background"
05:31:16falconfoxthen in the upper right it had a hdd icon which lasted for about 15 seconds.
05:31:58falconfoxthen i clicked "update now" and it said "data abort at 00037E0"
05:32:18 Quit joe2 (Remote closed the connection)
05:33:03safetydanfalconfox, if it's crashing you'll probably have to talk to Slasheri as he's the tagcache guy
05:33:18 Join placebo_lu [0] (
05:33:43falconfoxoh ok, well that crash was the first i got, usually it just doesn't work
05:34:13javkreiI've just tried to build rockbox without the #ifndefs...
05:34:29javkreiand... well, as expected it doesn't work
05:34:47falconfoxok, well now all i am getting is that error
05:35:02falconfoxhow do i do a fresh install of rockbox, i mean getting rid of all the stored settings
05:35:18javkreiit seems that the simulator doesn't have the functions pcm_play_stop(..)
05:35:25safetydanfalconfox, what player?
05:35:34falconfoxipod color
05:35:45javkreipcm_set_frequency and pcm_play_data
05:35:53placebo_lucould anyone help me out? I've just put rockbox on my new iRiver H340 and it won't play my .ogg files. here is a link to what the wps screen says when I try:
05:35:54javkreiI wonder why is that
05:36:03Mikachufalconfox: turn the hold switch on as you start rockbox, it will clear the config
05:36:27falconfoxMikachu: alright ill try it
05:37:24safetydanplacebo_lu, what program are you using to rip the oggs?
05:37:48placebo_luthey are 160kb .oggs
05:37:54safetydanHow old is the ogg encoder?
05:38:16safetydanRockbox used to have problems with oggs encoded with the pre-1.0 encoder (though that might have been fixed at some point).
05:38:26javkreinevertheless there is mp3 playback...
05:38:59placebo_lui downloaded EAC about 2 days ago
05:39:08safetydanjavkrei, I think those functions are too low level to be of use in the sim which is why they're not simulated. You could raise a flyspray issue and someone might try and fix it
05:39:09placebo_luso i'm pretty sure its the latest version
05:39:43safetydanplacebo_lu, if you have the ogginfo tool, try running that over one of the oggs
05:40:30placebo_lui don't, i'll get it now and try it out
05:41:32placebo_lucan you give me a link?
05:42:03safetydanplacebo_lu, I only know where it is for Linux
05:42:12safetydanYou didn't add ID3 tags to the ogg files by chance?
05:42:32falconfoxsafetydan: how come you love safety so much
05:43:01placebo_luI thought I did - EAC does it autmotically, doesn't it?
05:43:10 Quit _Lucretia (Read error: 113 (No route to host))
05:43:20javkreimmh, I've now seen that they are implemented
05:43:30safetydanplacebo_lu, oggs can't work with ID3 tags, it basically makes them corrupt
05:43:32javkreiin /uisimulator/sound.c
05:43:42safetydanfalconfox, it's just a name
05:43:49 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
05:44:00falconfoxsafetydan: well you picked it for some reason, i am just wondering
05:44:12javkreibut there is another #ifdef which apparently avoids them being built
05:44:46safetydanplacebo_lu, try finding a tool to strip the ID3 tags off the files and make sure you don't encode any more oggs with ID3 tags enabled
05:44:52javkreiI've commented that too...
05:45:11placebo_luany recomendations for the tool?
05:45:14safetydanfalconfox, I was working on flight safety stuff at the time I came up with the nick
05:45:23javkreiit still doesn't work...
05:45:29BloodySorcerermeep, bigass lightning.
05:45:32BloodySorcereri should probably go now.
05:45:37safetydanplacebo_lu, ask around on misticriver, I don't use Windows so I can't really offer any tools
05:45:45placebo_luis there another form of tagging I should use?
05:45:51placebo_lucool, will check back at mr
05:45:52Mikachuhow do you put id3 tags on an ogg file? i thought they were saved in extra mpeg frames
05:46:08 Join _Lucretia_ [0] (
05:46:11placebo_luits odd, because they work when i use the iriver firmware :/
05:46:30safetydanplacebo_lu, oggs have their on tag format called vorbis comments
05:46:39BloodySorcereraka metadata
05:46:54safetydanMikacu, some software just slaps the ID3 block on the front or end of the file regardless of whether its actually an mpeg stream
05:47:11placebo_luok, thanks for all your help guys :)
05:47:22placebo_luesp. safetydan!
05:47:34Mikachusafetydan: heh, okay
05:47:35BloodySorcerertheres a nifty tool for windoze called "multi id3 tag editor"
05:47:38BloodySorcererit'll probably help you
05:47:46BloodySorcererdoes batch processing
05:49:36javkreiwell... if someone knows how to enable midi playback in the simulator, or can guide me how to go about it, please help me
05:50:12safetydanjavkrei, it sounds like it'll need some coding work to fix, so maybe just raise an issue in the bug tracker
06:02:18javkreiok, I'll do that
06:04:21javkreimm, safetydan: I don't have a flyspray account yet
06:04:32javkreicould you raise it for me?
06:07:21javkreisorry, I've just registered
06:07:27javkreiI'll do it
06:08:43 Join ehour [0] (
06:15:11 Join JoeBorn [0] (n=jborn@
06:15:25 Quit placebo_lu ("CGI:IRC")
06:19:13javkreiwell... mp3 playback doesn't work either in my build of the sim
06:29:10 Quit lukaswayne9 ("Ex-Chat")
06:31:07safetydanjavkrei, it should work even if it's a little flakey sometimes
06:36:38javkreimhh, yes, rasher's simulator builds do have mp3 playback
06:36:41goffahmm... got an x5... when i mount it its read only.. checked kern.log.. got this:
06:36:50goffaAug 2 22:35:30 delubuntu kernel: [17352716.428000] fat_get_cluster: invalid cluster chain (i_pos 0)
06:37:00goffaran fsck.vfat
06:37:10goffait seems to leave stuff untouched
06:37:17Mikachuyou have to say -a or -r
06:37:29Mikachuthe default is to only check
06:37:36goffawhat's the dif?
06:37:46Mikachu-r asks, -a fixes everything
06:37:50goffaoh ok
06:37:54Mikachuwhere fix means tries to fix
06:39:34goffahad me worried.. just got my h140 back from the factory today... thought my x5 was crappin out on me
06:50:00 Quit Sinbios (Connection timed out)
06:50:09 Join Sinbios [0] (
06:51:19 Join Paul_The_Nerd [0] (n=Main@
06:51:29goffastupid question.. but do i need to run the h140 fwpatcher in wine? or is there a linux ver out there?
06:52:08Paul_The_NerdThere are linux instructions at the IriverBoot page.
06:52:17Paul_The_NerdBut they're not as simple as just running fwpatcher.exe in wine.
06:58:39 Quit aegray (Remote closed the connection)
06:59:07 Join aegray [0] (n=aegray@
06:59:21 Join scott666_ [0] (
07:01:19 Quit falconfox ()
07:02:49goffathink i got it
07:02:53goffajust have to copy the new hex
07:03:02 Part scott666_
07:03:19goffawaiting for these oggs to finish encoding so i can copy them to my x5.. then it'll be full..
07:03:27goffathen i can start filling the 140
07:04:43***Saving seen data "./dancer.seen"
07:09:28 Quit aegray ("i'm out")
07:10:22 Quit ehour ()
07:21:43 Quit Paul_The_Nerd ("Leaving.")
07:29:51Bg3rmorning ;)
07:29:57 Part desrt
07:32:27 Join agabus_ [0] (
07:48:54 Join ehour [0] (
07:49:45 Part ehour
07:50:29 Quit agabus (Read error: 110 (Connection timed out))
07:51:54 Join ScoTTie [0] (n=scott@unaffiliated/scottie)
07:56:34 Part safetydan
08:05:03 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
08:05:12 Quit amiconn (Nick collision from services.)
08:05:12 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
08:06:04 Join aliask [0] (
08:07:04 Quit ScoTTie (Read error: 60 (Operation timed out))
08:21:00 Nick BHSPitMonkey is now known as BHSPit[Birfday (
08:21:06 Nick BHSPit[Birfday is now known as BHSPit[Birfday] (
08:29:43 Join dan_a [0] (n=dan-mirc@
08:37:16Slasherihmm, it seems fat_seek is unoptimized for performance
08:37:46Slashericreating a cluster history table holding a few last accessed clusters, would make small seeks fast
08:39:19amiconnAs someone put it recently: do not overengineer the world
08:39:48Slasheritrue, but with very simple change that could boost tagcache performance
08:39:57Slasheriat least i will try if it has any real effect
08:40:24amiconnYeah, and it will increase code size, and ram usage, and complexity
08:40:41amiconnAbove all the fat code has to be reliable
08:40:47 Quit vadim (Remote closed the connection)
08:41:31Slasheriamiconn: sure, but that was meant to be very simple change only in fat_seek()
08:41:45 Quit JoeBorn ("Leaving")
08:44:21 Quit BHSPit[Birfday] ("Leaving")
08:47:26JdGordonhow big is a cluster?
08:47:56amiconnDepends on the disk size
08:48:28 Join BHSPitMonkey [0] (
08:48:43 Join midgey34 [0] (
08:48:45amiconnFAT/FAT32 allows clusters from 1 to 64 (and inofficially, 128) sectors in powers of 2
08:48:48Slasheriamiconn: ah, in fact never mind, there was already a check for "clusternum >= file->clusternum" that i didn't notice
08:48:52 Join ScoTTie_ [0] (n=scott@unaffiliated/scottie)
08:48:58Slasheriso not much to do there then
08:49:39 Join BHSPitLappy [0] (
08:52:14 Join scott666 [0] (
08:54:15amiconndionoea: around?
08:54:49 Quit BHSPitMonkey ("Leaving")
09:04:33 Quit Hulex ()
09:04:45***Saving seen data "./dancer.seen"
09:05:22 Join Hulex [0] (
09:12:11 Join petur [0] (
09:13:07*petur hugs his rockbox shirt
09:13:31*scorche hugs his own
09:13:42peturthe combination with an orange safety vest makes people think I'm police (when riding my bike)
09:13:59scorchei havent worn mine yet =O
09:14:09scorcheBagder: came =D
09:14:45scorcheand, if you didnt read the logs, i commented on your nice handwriting
09:14:49Bagderwhat a timing
09:15:00scorcheno kidding
09:15:13scorchethe tape was a bit difficult though ;)
09:16:22scorchei got a bit scared when i came home and didnt see it on the porch..
09:16:32scorche...then i realized they stuffed it in my mailbox
09:18:43amiconnPorting the grayscale lib to ipod is more work that I thought
09:19:23amiconnIt needs a different greyscale framebuffer format for buffered mode too in order to be efficient
09:19:47amiconnSo most drawing functions need an alternative implementation...
09:22:47amiconnNot especially difficult work, just tedious
09:22:56 Join _FireFly_ [0] (
09:28:03 Nick BHSPitLappy is now known as BHSPitL[B-day] (
09:43:17 Quit BHSPitL[B-day] ("Leaving")
09:55:54 Quit dan_a (Read error: 110 (Connection timed out))
09:57:15 Join scott666_ [0] (
09:57:46 Join Farpenoodle [0] (n=solo84@
09:59:54 Quit scott666 (Read error: 104 (Connection reset by peer))
10:11:18 Join mikearthur [0] (
10:12:18 Join dan_a [0] (
10:16:20 Join safetydan [0] (
10:21:05scorcheand i am out...
10:21:25scorchepacking up comp for the ride to vegas, and then home =)
10:22:37 Quit scorche ("Leaving")
10:26:17BagderINPUT(crt0.o) is needed for all bootloaders except the ARM ones
10:26:27Bagderbut my sansa bootloader fails without it...
10:26:53dan_aBagder: why isn't it needed on the ARM ones?
10:27:05BagderI don't understand it
10:27:19Bagderbecause the file is needed and used
10:27:24markunBagder: we need it for the gigabeat too
10:28:15Bagderthat's some ld magic that I don't get
10:29:40 Join pondlife [0] (
10:30:30Bagdermarkun: you build bootloader from the CVS version?
10:31:21pondlifeHi Flyspray admin types!
10:31:36linuxstbBagder: It's needed for all apart from ipod.
10:31:43Bagderpondlife: how can we help you good sir?
10:31:53pondlifeCould someone please close
10:32:02markunBagder: the code is on in cvs yet, but will be soon
10:32:09pondlifeMiika already did a commit which covers this
10:32:31 Join spiorf [0] (
10:32:36linuxstbBagder: For some reason, there's a linker error if you use it. Possibly because ipod.c refers to a data structure in crt0.S (the header from the original Apple firmware that's been put there by ipod_fw.c).
10:32:40JdGordonpondlife: done
10:32:43pondlifeAlso can be closed
10:33:04pondlifeI think that's about it for now! Many thanks JDG
10:33:09Bagderpondlife: I've made you a developer in flyspray, you can now "admin" tasks
10:33:14pondlifeWooh, thanks
10:33:37pondlifeI'm away for the next week, so probably won't do much, but that should save me pestering
10:34:01Bagderand there are many more weeks to come ;-)
10:34:10pondlifeHope so!
10:34:21pondlifeWhat, you mean it won't be finished soon???
10:34:33Bagder"when is the final Rockbox version coming?"
10:35:18pondlifeActually, while I'm here and before I go. Has anyone had a look at ?
10:35:52pondlifeI can say for sure it fixes a crash. But it may be a hack,. hardeep/lostlogic will probably need to confirm..
10:37:33pondlifeI'm not sure what is meant to provide the final Q_VOICE_STOP, but I suspect something is!
10:38:07amiconnImo voice is too dependent on playback state right now
10:38:15pondlifeI agree
10:38:22pondlifeIt should be a level above that completely
10:38:51amiconnI guess it's necessary to do the mixing. That should also fix things like voice not working during pause
10:38:58pondlifevoice > playback > pcmbuf
10:39:25amiconn...and clip preemtion, and latency, and....
10:39:35pondlifeI understand talk.c and a bit of pcmbuf.c, but playback.c is still mysterious.
10:40:34pondlifeI can say for sure that pressing stop during voice playback and audio playback (e.g. while filename is still being spelt) will end up in that yield() loop forever
10:40:38 Join ehour [0] (
10:40:49pondlifeThat's about all I know for sure
10:41:56Bagdermarkun: I added crt0.o now for non-ipods
10:42:19BagderI can now build a full PP5024.mi4 file
10:42:24amiconnBagder: Why isn't the mi4 creation logic added to scramble.c ?
10:42:51Bagderbecause we don't have MrH's real name and the code is not GPL'ed
10:42:52pondlifeMy suspicion is that hardeep was trying to catch the situation where no Q_VOICE_STOP was expected (i.e. queue_empty), but the logic is slightly wrong for this. And my suspicion is probably wrong too.
10:43:10amiconnBagder: oh.
10:43:44Bagderand the mi4code is quite a lot more advanced than "just" scramble
10:44:11merbananBagder: did you get the new mi4code version ?
10:44:22Bagderyeps, its on the site already
10:44:30*merbanan checks
10:44:53amiconnBagder: So it's not possible to build for sansa just from the source tree?
10:45:37Bagderamiconn: nope, you need to build the separate tool as well for now
10:47:14Bagderwell, its not a major hurdle
10:47:25Bagderwe required uclpack for years
10:47:34*amiconn wonders what made Bagder deviate from the realname policy
10:47:44BagderI don't
10:47:48Bagderit isn't in Rockbox
10:48:27amiconnThere is a difference though. ucl is GPLd
10:48:37Bagderbut we used it as an external tool
10:48:41Bagderas we now use mi4code
10:48:54BagderI'm not saying I think this is a great way
10:49:03Bagderbut these ports are *early*
10:49:05amiconnHmm. Probably not too much of a problem
10:49:07Bagderso it doesn't matter much
10:49:24amiconnSome other things will also use external tools which are not GPLd
10:49:43 Join jn [0] ( (hopefully soon) voice building will use SAPI5
10:51:23*amiconn wonders what's the license of mi4code
10:51:31amiconnThe header doesn't state any
10:52:20merbananBagder: did you try the new option that shouldn't need the TEA encryption step ?
10:52:48Bagderare you sure it works like that?
10:53:15merbananBagder: it's my idea :), it works in theory
10:53:27jnac charging seems to not give much of a charge :|
10:54:19jnanything big happen in the ipod-photo daily build in the last week?
10:54:38 Join kidd3ckz [0] (
10:54:51merbananBagder: in theory it would be possible to boot firmwares on hardware for which we don't have the TEA keys
10:55:35 Join Kohlrabi [0] (
10:55:55Bagderyeah, if they accept those lengths of the plain text value
10:58:41 Quit sando (Success)
10:59:01 Join sando [0] (i=lolsteam@
10:59:34 Quit mikearthur ("Konversation terminated!")
11:04:49***Saving seen data "./dancer.seen"
11:06:45 Join TeaSea [0] (
11:09:02 Quit eGen_ ("mdlo kon ! ... runk kon !")
11:10:38 Quit cal_rifken (Read error: 110 (Connection timed out))
11:14:30 Join eGen_ [0] (
11:16:32 Join Acksaw [0] (
11:16:40Acksawhey guys
11:16:58Acksawim in need of some help setting up my EQ for my Sennheiser MX550's
11:17:08Acksawis anyone proficent in this?
11:21:06safetydanAcksaw, you might want to check the various forums and see if anyone's done settings for similar headphones
11:21:41Acksawi've looked over MR
11:22:20safetydannothing close?
11:23:11safetydanWhat part isn't working for you?
11:24:11Acksawi've just never set up an EQ before
11:26:50safetydanHave you read through the manual?
11:27:21Acksawsince there isnt one as far as i know
11:28:17Acksawand this is for the graphical eq?
11:28:23safetydanIt's for the whole of Rockbox
11:28:29safetydanthere's a section on the equazlier in there
11:33:48*amiconn wonders who did the ipod button assignment in grayscale.c
11:33:59amiconnIt's impossible to leave the plugin
11:34:32 Quit jn ("Lost terminal")
11:34:32dan_aThat sounds like something of an oversight...
11:35:12amiconnThe button is defined, but can't work that way
11:35:29amiconnIt's a dev-only plugin anyway
11:35:44*amiconn is working on the grayscale lib port for greyscale ipods
11:37:13*dan_a has a greyscale ipod...
11:37:20*amiconn knows
11:37:34dan_aThat's for things like Doom and Rockboy, isn't it?
11:37:54amiconnDoom, mandelbrot, fire, plasma, and the jpeg viewer
11:39:26amiconnRockboy doesn't need it, as a gameboy only has 4 levels of grey anyway
11:39:39dan_aIf you want me to test anything, let me know - although I can't do anything while I'm at work
11:40:13amiconnThis stuff should work on 3g the same way as on 4g and mini
11:40:17*amiconn has a mini2g
11:40:49amiconnThe only thing that should be tested explicitly on 3g is the lcd's internal frame rate
11:40:55amiconnI have a test plugin for this
11:42:41dan_aOK - I'll get that tested when the lib is ready
11:44:36 Join excitatory_ [0] (
11:45:44amiconnsim is already working...
11:45:46 Join einhirn [0] (
11:45:46 Quit einhirn (Client Quit)
11:47:04 Quit excitatory (Connection timed out)
11:47:06dan_aWhere can I get the test plugin then?
11:47:21amiconnNow I need to try getting my head around arm asm...
11:48:25amiconnI can send it to you, but it will probably commit it to cvs along with the ported lib
11:49:02dan_aIt's probably easiest if I take it from CVS then
11:49:05amiconnI still don't know whether the lib will work as intended on ipod
11:49:34amiconnThe lcd hardware might cause unforeseeable effects
11:49:56dan_aWhy would that be?
11:50:02amiconnNo other way to find out that port and see
11:50:25amiconnIt's because of the way the grayscale lib works
11:50:54amiconnIt does so-called temporal dithering, i.e. fast flipping through a number of monochrome frames
11:52:10 Quit safetydan (Read error: 110 (Connection timed out))
11:53:29amiconnFor one, this flipping must be in sync with the lcd internal frame frequency (or nearly in sync, as all LCD controllers I've seen so far use an RC oscillator which isn't very stable frequency-wise, and it's impossible to read back the current frame state)
11:53:33 Join safetydan [0] (
11:55:04amiconnThere can be 2 caveats: (1) It might be impossible to do the flipping fast enough due to bottleneck(s). (2) The LCD may act up (showing artefacts) when doing the fast flipping
11:55:05 Join freqmod [0] (
11:55:36 Quit ehour ()
11:55:55dan_aI suspect that I might be affected by (1)!
11:56:08amiconnNo more than 4g and mini
11:56:16amiconn(I guess
11:57:26 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
11:57:32 Quit amiconn (Nick collision from services.)
11:57:34 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
11:57:36dan_aOnly if everything is done in IRAM (if I understand everything correctly)
11:57:38amiconnYou can build & run test_fps.c for 3g to see how fast the 3g can update the lcd
11:57:59amiconnThe bottleneck isn't the RAM in this case
11:58:08amiconnIt's the PP LCD interface
11:59:10 Quit thegeek (Read error: 113 (No route to host))
11:59:11dan_aDo I just need to add test_fps.c to apps/plugins/SOURCES to do that?
11:59:32amiconn(or rather, the lack of knowledge on our side how to change the PP LCD interface timing)
12:00:35amiconnYes, that should work
12:01:28safetydanwhat's the current recommended version of gcc for m68k?
12:02:14safetydanis it worth upgrading from 3.4.5?
12:02:24amiconnprobably not
12:02:42 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:02:47SlasheriHmm, i just got an idea that it would be possible to define more complex searches for tagcache when it's loaded in ram.. For example referencing to other tags. And when it's not in ram, an error message could be displayed that the search is too complex, please load tagcache to ram.
12:04:34 Join sIRwa2 [0] (
12:06:14sIRwa2i just got my new lcd for the jukebox recoreder, but i cant find a guide to disasamble the jukebox on the website anymore
12:06:41 Join BigMac [0] (n=dummeska@
12:06:54 Part BigMac
12:07:19*amiconn points sIRwa2 towards
12:13:08sIRwa2may i thank thee
12:13:14 Join obo [0] (
12:14:15safetydanamiconn, I think I've asked this before... are the grayscale lib displays supposed to "shimmer" periodicaly?
12:14:39*dan_a now has test_fps.rock to put on his iPod
12:14:39amiconnThat happens because of 2 effects
12:15:16amiconn(1) Slight difference between internal lcd frame rate and refresh rate
12:15:36 Quit aliask (Remote closed the connection)
12:16:17amiconn(2) Whenever data is written to the LCD GRAM position the LCD controller is displaying at that very moment, there is a collision which causes artifacts
12:16:37amiconnThese artifacts are black on the archos LCD, and white (and less visible) on the H1x0 LCD
12:23:01 Join lodesi [0] (
12:40:17 Join A_M [0] (
12:44:40 Join barrywardell_ [0] (
12:45:14barrywardell_Hey Bagder
12:45:18Bagderhey ho
12:45:29barrywardell_i see you have a new mi4code version. What is updated?
12:45:30Bagdernoticed my
12:45:41barrywardell_noticed that too. very useful
12:45:46Bagderbarrywardell_: the ability to set the plain text part
12:46:00Bagderpossibly we can skip the TEA encryption by setting the whole file as plain text
12:46:43Bagderset the plain text *length* rather
12:46:53barrywardell_i see
12:47:18barrywardell_not too critical to getting it to work on h10 or sansa though
12:47:34Bagdernot really, no
12:47:46barrywardell_I also noticed your new cvs commits
12:48:02barrywardell_Specifically having crt0.o
12:48:05Bagderyeps, now I get a PP5022.mi4 out of my sansa bootloader build
12:48:14Bagderready to copy to target
12:48:26barrywardell_do you think it is necessary to have that for the H10 too?
12:48:49Bagderconvenient at least, and I think you just need to edit configure to have it work
12:48:57barrywardell_have you tested your new build on your sansa yet?
12:49:06Bagderbut nothing
12:49:17BagderI mean, I have no means to see if it works
12:49:33Bagderdoing LCD tests but no sucess yet
12:49:41barrywardell_i suppose without the lcd or disk it's hard to test
12:50:26barrywardell_i've edited my configure version to use
12:50:36barrywardell_much easier
12:51:10 Join thegeek [0] (
12:52:01barrywardell_am I right in understanding that previously crt0.S wasn't used by the arm bootloader?
12:52:10barrywardell_but now it is?
12:52:54Bagderyou tried CVS up and rebuilt your bootloader?
12:53:21Bagderthe problem before was that crt0.S was not properly "pulled in" when I build the sansa loader, so I had to do this trick
12:53:31Bagderthe ipod loader fixes it by itself
12:53:56barrywardell_ah, i see
12:54:02merbananbarrywardell_: what happend to your H10 player ? did you fix it ?
12:55:21barrywardell_just got a build error (multiple definition of 'start') after CVS up and rebuild bootloader. that is what you are referring to bagder?
12:55:35Bagderyou're using the ipod way then I guess
12:55:53barrywardell_I have left lots of the ipod stuff there for the moment
12:56:10Bagderprevent from pulling in that file and it should work
12:56:10barrywardell_merbanan: the hdd failed. I sent it back to iriver for repair
12:56:34merbananbarrywardell_: your fault or hardware failure ?
12:57:57sIRwa2damn, my jukebox usb is solderd to both motherboards
12:58:31barrywardell_merbanan: not totally sure. i was running code from rockbox on it at the time, but nothing I thought was dangerous so I suspect hardware failure
13:00:39 Quit A_M ("CGI:IRC")
13:00:53 Quit ^jhMikeS^ (Read error: 110 (Connection timed out))
13:03:15 Join jhMikeS [0] (
13:03:29 Join rconan [0] (n=richard@
13:03:43 Quit rconan (Read error: 104 (Connection reset by peer))
13:04:53***Saving seen data "./dancer.seen"
13:06:19 Join tvelocity [0] (
13:12:28Bagderdebugging a linux kernel certainly gets easier once the serial port starts working
13:12:38 Quit t0mas (Remote closed the connection)
13:12:40 Join t0mas [0] (
13:13:15 Join webguest79 [0] (
13:13:28 Join oiup [0] (
13:13:47oiuphello, anyone from rockbox here?
13:13:54 Join ax_ [0] (
13:14:04oiupknow about the rd2780?
13:14:14Bagderwhat's there to know?
13:14:47oiupjust wondering if it would be possible to get people interested in porting rockbox for the device
13:15:00Bagderothers have mentioned interest before
13:15:17 Quit ax (Read error: 104 (Connection reset by peer))
13:15:34oiupprecisely but they don't seem very enthusiastic, at least they are at the started but then nothign really moves on
13:16:11oiupwe got the boot sequence and all
13:16:26Bagderbut the best way to get interested people is to hunt them up in forums where you find owners of that player
13:17:11oiupyeah but you know, people signup post a topic saying "im interested in making a new firmware for this thing" and then disapear
13:18:00oiupso whats so different with the archos? i mean you guys have made rockbox for it, why not the rd2780?
13:18:15Bagderwe've made rockbox for some ~15 players by now
13:18:41oiupright, i haven't seen any screens of rockbox running on archos, know where i could find soem?
13:18:48Bagderoiup: its simple, no one who owns a rd2780 has had enough interest or skill
13:19:15oiupyeah i guess
13:20:16Bagderoiup: so stop waiting, start acting on your own
13:20:50Bagderthere's a huge community of rockbox hackers around to assist with help on generic hacking and rockbox internals
13:21:03 Quit webguest79 ("CGI:IRC")
13:21:04 Join webguest79 [0] (
13:21:33 Join oipp [0] (
13:21:44oippgot deconnected, could you give me that link again?
13:21:54Bagderthere's a huge community of rockbox hackers around to assist with help on generic hacking and rockbox internals
13:22:17Bagderjust take the plunge and never turn back! .-)
13:22:24markunand don't give up
13:22:30markunothers will join you
13:22:34oippyeah but i don't know a thing about programming unfortunately
13:22:41Bagdermore people are likely to join if any progress is shown
13:23:18 Join ripnetuk [0] (
13:24:06ripnetukhi... ive just got myself a Archos AV500... i realise that Rockbox isnt even started to be supported on it, but was wondering if anyone knows /any/ tech details about it (like processor, chipset etc)??
13:24:15ripnetukjust out of geeky interest ?
13:24:37BagderI think that's a TSM320 or similar
13:24:57oipptheres no screens of a unit running rockbox playing video?
13:25:16ripnetukcool - thanks bagder :)
13:25:31markunoipp: only the old archos units can play video (in grayscale)
13:25:40ripnetukive never come across archopen - geek heaven
13:25:43Bagderyeps, TMS320 based
13:26:06 Quit oiup (Read error: 104 (Connection reset by peer))
13:26:34oippis there any other archos ports?
13:26:42oipplike not rockbox
13:27:17Bagderbut I'm not sure how that works or exactly what they can do
13:27:34oippk thanks
13:28:07Bagderwhy care about Archos ports?
13:28:16 Join ^jhMikeS^ [0] (
13:28:23 Part barrywardell_
13:28:23BagderLD2780 is hardly an archos
13:28:40oippbecause the rd2780 is quite similar to the archos
13:28:46Bagderin what aspects?
13:28:54oippi think it has the same procesor
13:28:55Bagderand similar to what archos?
13:29:14markunoipp: the archos players we support are probably not the ones you are talking about
13:29:36oippprobably not yeah i realised from the screens
13:29:38Bagderright, archos gmini 400 and AV400 have the TMS320DM270
13:29:58oippo ok, rd2780 has Ti DM270 processor
13:30:10Bagderyes, that's the one
13:30:20BagderTMS320 DM270
13:30:31ripnetuklooks like it needs a propritery tool chain :( althought wikipedia says there is some gcc support
13:30:43Bagderits an ARM core
13:30:55Bagderso its full gcc support for that
13:30:55oippyeah, don't some pda's have similar processors?
13:31:27oippthe easieast thing would be to just upgrade the current firmware
13:31:33oippi think
13:31:38Bagderyeah there might be a PDA somewhere but TMS320 is more portable player oriented
13:31:50Bagder"upgrade" ?
13:31:57 Quit ScoTTie_ (Read error: 110 (Connection timed out))
13:31:57Bagderbut you don't have the sources
13:32:05oippyeah like if i could upgrade the codecs or something
13:32:10Bagderso that upgrade would be what you should write
13:32:17oipppart of the source code has been released
13:32:24Bagderyes, the public linux sources
13:32:31Bagderyou can get them from everywhere
13:32:36oippoh ok so theres nothing usable?
13:32:38Bagderthe interesting parts are not in there
13:32:53Bagderwell, some parts might be interesting of course
13:32:53oippdamn would have to bribe an rca dev or something
13:33:13Bagderoipp: welcome to reverse engineering. this is rockbox ;-)
13:34:09oippwell the device will only play mpeg4-sp, it refuses to play b-frames (damn thing just skips all b-frames), would it be because the procesor isn't powerful enough? or is it just a software issue?
13:34:36BagderI would guess it is a sw issue
13:34:49Bagderbut that sw is in the DSP parts
13:35:14oipptheres a new rca x3000 out that plays mpeg4-asp might try and find what hardware is in it
13:35:46 Join aliask [0] (
13:37:09oippowell g2g, cyas, will look into this some other time, thanks for all your help
13:37:13 Quit oipp ()
13:37:13 Quit webguest79 ("CGI:IRC (EOF)")
13:40:39JdGordonreally ot.. does anyone know how to convert numbers between any two bases recursivly?
13:40:59JdGordonhell even iterativly would be fine.. i cant get it working :'(
13:43:00 Quit jhMikeS (Read error: 110 (Connection timed out))
13:43:08sIRwa2WOOT i did it! i replaced my archos recorder screen.. 1:40 min
13:43:19sIRwa2damn that was hard. and very messy
13:46:51safetydanYay, I can now calculate the apparent right ascension and declination of the sun on my h120...
13:46:57safetydannow if only the damn thing had an rtc
13:47:10pondlifeAre e-mail addresses used in Flyspray likely to be spam-proof? I can't see any way to view them, but wanted to be certain.
13:47:56safetydanah well, guess I'll just have to rely on users to enter the current time on non-RTC targets
13:48:20safetydanpondlife, I haven't received any spam because of Flyspray. I think your email address isn't visible to non-admin users
13:48:36Bagderonly in mails flyspray sends
13:48:36pondlifeJust added it properly
13:48:49Bagderif you subscribe to a notification
13:49:24markunsafetydan: someone did a RTC mode for the iriver H1xx. I think it was Slasheri.
13:49:31markunRTC mod
13:50:23safetydanI'm not sure if my skills with a soldering iron (i.e I have no soldering skills) make it a wise move to start tinkering with the insides of my h120
13:50:57safetydanbesides, I'm not that desperate to have the time on there. it'll be quick enough to enter todays date in to the app I'm slowly building
13:51:18Bg3rsafetydan: :D
13:52:52Bg3rsafetydan: what app are you building ?
13:53:30safetydanBg3r, basically porting this to Rockbox
13:53:43safetydanbut when I say porting I mean rewriting from scratch and in the Rockbox manner :)
13:53:53safetydanso more... taking the concept or being inspired by
13:54:26Bg3ryay :)
13:55:00Bg3rvery good ;)
13:55:41 Quit petur ("more worrrk")
13:56:12 Join ScoTTie [0] (n=scott@unaffiliated/scottie)
13:56:32 Quit Sinbios ("If the definition of a klutz is someone who doesn't have eyes on their ass, then yes, I suppose I am a klutz.")
13:57:48ripnetuk:( looks like Archos no longer have their firmware on the hdd (==safe hacking) - looks like the firmware is flashed. Having said that, when I upgraded, it happened WAY fast, so i guess it could be on a hidden sector on the hdd... anyone know? (AV500)
13:58:25Bagderjust dd the whole disk before and after upgrade! ;-)
13:58:43 Quit _Lucretia_ (Read error: 110 (Connection timed out))
13:59:07ripnetukbagder - yes, thats what I plan to do (apart from the fact ive already upgraded and you cant downgrade - ive gotta wait for new firmware from Archos)
13:59:19ripnetukbut I could look in a dd for strings
13:59:50ripnetuki always likes the way Archos just loaded the fw from disk - i was very scared when I first flashed my iRiver with the rockbox loader :)
14:00:06ripnetukgotta go... thanks for help
14:00:08 Quit ripnetuk ("Ninja IRC v1.5.8.1(#1) exiting after 36m52s of use")
14:01:07 Quit Kohlrabi (Nick collision from services.)
14:01:17 Join Kohlriba [0] (
14:04:46 Join Criamos [0] (
14:05:54safetydannight all
14:05:55 Quit safetydan ("Ex-Chat")
14:07:04 Join Luckydevil41 [0] (n=19555779@
14:07:24Luckydevil41who to load into apple firmware on nano?
14:10:49 Join _Lucretia_ [0] (
14:25:33 Quit Febs (Read error: 110 (Connection timed out))
14:27:55 Quit aliask ("Chatzilla 0.9.72 [Firefox]")
14:35:19 Join barrywardell_ [0] (
14:36:57 Join Moos [0] (
14:37:05MoosHi all !
14:37:23JdGordonits mooooss!!!!
14:39:55barrywardell_iriver have received my h10 today. "Our technicians will keenly check and test the device."
14:40:00Moosis there someone with CVS access here for a one line bugfix?
14:40:02 Join lightyear [0] (
14:40:04barrywardell_hopefully they don't check it too keenly!
14:40:57MoosBagder: hi, there?
14:41:09Bagdera bit
14:42:07Moosjust one missing tag in francais.lang
14:42:30Moosthat made the "no" string not working properly
14:43:27MoosI don't have synced the francais.lang yet, and I thought no need to do a patch for just one missing tag
14:44:06Bagdernope, but I'd need to fix things to get the unicode to work and I don't have that in order
14:44:25Moosah ok
14:46:17MoosI'll see if there are a lot of new strings to tanslate and make a patch
14:46:24markunhi Moos!
14:46:39Moosheya markun !
14:46:39 Join ehour [0] (
14:47:19 Part ehour
14:49:11amiconn[14:44:04] <Bagder> nope, but I'd need to fix things to get the unicode to work <== What things?
14:49:29Mooseditor thing?
14:50:24 Join sharpe [0] (
14:52:44dan_aCan anyone tell me if there are any major conceptual errors with this: - the idea is that if the COP on PP-based machines is available, then the codec thread will run on it
14:53:10dan_aI know that there will be major problems in practice with doing that (like cache concurrency)
14:53:17dan_abut is the basic idea sound?
14:53:41linuxstb_dan_a: I haven't looked at your patch, but if you want to test your ideas, don't forget that you can disable the cache completely.
14:54:00dan_alinuxstb: Just by not calling ipod_init_cache()?
14:54:21Bagderhaving the codec thread there sounds like a plan
14:55:24linuxstb_dan_a: Yes. Your ipod will be dog-slow, but it will still work fine.
14:55:53dan_as/dog-slow/even more dog-slow/
14:56:09linuxstb_:) dog-with-a-limp-carrying-a-heavy-bag-slow
14:56:33BagderI hear dogs with heavy bags are reaaaally slow
14:56:58dan_aAlthough it will be a 2 dog team now...
14:57:24Bagdertwo dogs carry bags better
14:57:56dan_aSomehow I can't believe it'll work, it all seems too easy.
14:58:48Mikachui thought the official stance was not to use the cop if it wasn't absolutely necessary
14:59:03linuxstb_Mikachu: It seems pretty necessary on the 3g.
15:02:07dan_aI've got a vague plan to only use the COP if the CPU boost is very high, but that's for when I understand the code a bit better
15:02:38Mikachui don't think you can measure boostedness, the playback code will only set the boost flag once iiuc
15:02:39dan_a(and when I understand C a bit better...)
15:03:57 Quit Luckydevil41 ("ThisRealityIsJustAnitherDream")
15:04:09 Quit Moos ("CGI:IRC (EOF)")
15:04:52Mikachuonly two more commits until playback.c 1.337
15:04:56***Saving seen data "./dancer.seen"
15:05:06*freqmod wonders what that would have to be done to get speex into cvs (now that the feature freeze is over)
15:05:52 Part scott666_
15:05:55 Quit spiorf (Read error: 104 (Connection reset by peer))
15:06:33JdGordonMikachu: both my random folder patches would do it!
15:07:35 Join Moos [0] (
15:08:09 Nick Kohlriba is now known as Kohlrabi (
15:09:33 Quit crashd_ (Remote closed the connection)
15:12:13*amiconn is looking at a working grayscale.rock on his ipod mini
15:12:18linuxstb_freqmod: My thoughts are that it would be nice to have our own Ogg parser optimised for Rockbox - for use by speex, vorbis and possibly the FLAC decoder. IIUC, your patch adds a second Ogg parser (libOgg) to Rockbox.
15:12:31linuxstb_\o/ How does it look?
15:12:43amiconnGrainy (the lcd internal frame frequency is way lower than on archos or iriver)
15:13:00freqmodi have moved the framing.c into libspeex, but I am positive to a ogg parser specific to rockbox
15:13:07 Join spiorf [0] (
15:13:15amiconnI also suspect some controller setting to be sub-optimal
15:13:31linuxstb_That's no surprise...
15:13:33freqmodI tried to port speex to use the tremor parser, but it won't work because the buffers it delivers are non linear
15:14:00freqmoddoes flac support OggFLACs?
15:14:03javkreiHi, someone knows why midi playback doesn't work in the simulator?
15:14:22amiconnThe ipod mini backlight is very nice for greyscale
15:14:26javkreiand if it would be too difficult to fix?
15:14:48linuxstb_freqmod: No. Currently the FLAC decoder only supports native FLAC. But I'm not sure if OggFLAC is very popular - I can't recall anyone asking about Rockbox support for it.
15:15:06 Quit Moos ("CGI:IRC (Ping timeout)")
15:15:34freqmodno i don't think so, but then i cant steal flac's ogg decoder :(
15:16:37freqmodand tremor's will not work unless i reprogram the speex library, which leaves me with libogg, or a library written from scratch
15:16:38linuxstb_freqmod: I did start work on a from-scratch Ogg parser, but didn't spend much time, and didn't get very far. The aim would be to implement it without any malloc, and with the very minimal amount of copying of data.
15:18:29merbananlinuxstb_: as usual look in ffmpeg, the ogg demuxer should be quite ok
15:18:46linuxstb_But ffmpeg isn't afraid to use malloc...
15:18:50freqmodspeex would only need the framing.c routines (not the bitwise routines).
15:19:54freqmodwould api compatibillity (with libogg) or effectivenes be most valued?
15:20:46linuxstb_I didn't go as far as looking at the current APIs, so I don't know. Obviously maintaining the same (or very similar) API will mean no (or less) changes in speex and Tremor.
15:21:57freqmodthe biggest help in preformance was if the data could be keept in rockbox's file memory, not copied.
15:22:27*freqmod looks at (ogg framing specification)
15:22:50 Join webguest05 [0] (
15:23:53 Quit webguest05 (Client Quit)
15:24:11 Join webguest55 [0] (
15:25:13 Quit webguest55 (Client Quit)
15:27:49markunlinuxstb_: tremor doesn't use the libogg api. a bit towards libogg2
15:28:57freqmodTremor uses some kind of linked pages etc, but the api is quite similar.
15:31:12 Join Poka64 [0] (
15:36:00 Quit philth (Read error: 113 (No route to host))
15:38:02 Join ftugrul [0] (n=tugrul@
15:39:24ftugruli'm using rockboox h300 experimental build, i think i've found a bug
15:40:14ftugrul"keypress effects when backlite is off" is selected for lcd remote, but it's not working. is this common?
15:43:52infamisno h300 here
15:44:18linuxstb_ftugrul: Have you tried an official Rockbox build? We can't help with experimental builds.
15:44:31linuxstb_(but if the bug happens in the official build, we would like to know)
15:45:00ftugrulok, i'll try it with latest cvs and report to you
15:45:36infamiswait....what's an expiremental build? cvs + patches?
15:46:40ftugrulhmm, i have to try latest non-cvs build?
15:48:17infamisprobably a daily build or cvs build
15:48:57ftugruli'll try latest "cvs build"
15:49:57 Quit sIRwa2 (Read error: 110 (Connection timed out))
15:50:48ftugrulis "" ok?
15:53:56infamisdaily builds run at a specific time-of-day. there may be be commits after the daily build has run, so that's why the cvs-build is the most-up-to-date build.
15:53:58linuxstb_ftugrul: When you change between an official Rockbox and an experimental version, it's probably best to force a settings reset when you first start the new version - hold REC as Rockbox is booting.
15:54:54ftugruli'll just rename .rockbox and rockbox.iriver then install new build.
15:55:41freqmoddoes anybody know the rockbox buffering code. Would a 2xpage buffer, or up to a bage seeking backwars be cheapest?
15:56:04linuxstb_How big is a page?
15:56:45freqmodwhen ogg_sync_buffer reallocates pages it adds
15:56:49freqmodforget it
15:57:01freqmodit is the remaining of the buffer when all pages is removed
15:57:36freqmodi think it depends on the encoder, for a normal speex file 2,5 kb.
15:58:35freqmodthe question is if should implement a ringbuffer or tell the codec to seek(unused bytes) back and return data from there
15:59:12preglowi thought the pp5024 was very similar to pp5020. not so?
16:00:00freqmodpreglow:AFAIK the only difference is a broken cache
16:00:12preglow_pp5024_ has a broken cache???
16:00:26preglowsure you're not mixing it up with pp5002?
16:00:32ftugrulok, it's fixed on official build.
16:01:41Mikachuftugrul: the settings aren't saved in .rockbox
16:02:00infamisisn't it in memory?
16:02:25ftugrulMikachu, will it remove my original h300 firmware's radio presets and rockbox radio presets?
16:02:37ftugrulif i reset settings?
16:02:42Mikachuno idea, i don't think so
16:02:54Mikachuthey are probably too large to be saved as settings
16:03:35freqmoddoes request buffer advance the buffer?
16:04:53freqmod(from a codec)
16:08:22linuxstb_freqmod: No. request buffer just gives you a pointer to the start of the buffer. You manually advance it. The buffer_read() function will both copy the data and advance the buffer.
16:08:53javkreicould someone tell me if midi playing is working in rockbox?
16:09:04linuxstb_But the idea is that you should advance it as soon as possible - to give the audio thread the ability to fill it with future tracks.
16:09:12ftugruljavkrei: yes
16:09:23ftugrulbut you need to get patchsets first.
16:09:33Bg3rftugrul: it shouldn't, but u must open the preset list manually
16:09:45javkreiok, thanks, so it's just in the simulator that it doesn't right?
16:10:34freqmodlinuxstb_:But i want to keep it unadvanced to advance it just as much as the codec has processed (right before the next request_buffer).
16:10:51freqmodwould that be worse than a ringbuffer?
16:10:55linuxstb_That's fine - most codecs work like that.
16:11:30linuxstb_i.e. they request some buffer (enough to guarantee it contains at least one compressed frame), decode it, and then advance the buffer "consumed" number of bytes.
16:11:49freqmodok, then ogg_sync_insert_reference takes the pointer from request_buffer, and then ogg_sync_get_advance returns the number of bytes processed
16:12:12javkreiftugrul: the uisimulator, that's a simulation of rockbox that runs on pc's
16:12:29 Join Febs [0] (
16:12:32ftugruloh, no
16:12:47ftugruli'm using it with my h340, and it works.
16:13:02javkreiok, thanks a lot
16:13:47ftugrulnot perfect, it sometimes can't catch buffer, but plays :D
16:14:13 Join klrspz [0] (n=klrSpz@
16:17:37freqmodA c question, would "ogg_sync_insert_reference(state,request_buffer(<chunk size>,&realsize),realsize)" send the real size of request_buffer to ogg_sync_insert_reference?
16:20:50klrspzassuming it's the 3rd parameter to that function, then yes it appears so
16:26:02 Part ftugrul ("Konversation terminated!")
16:28:22linuxstb_freqmod: Personally, I would split that into two lines - just to make it clearer.
16:30:51freqmodok, it is currently just a comment on how to input data, i haven't converted the whole file nor started compiling it yet.
16:31:54freqmodit does seem like continued packet could be a problem
16:33:01*amiconn found the problem of the heavy grescale flickering
16:33:26amiconnThe lcd internal frame frequency is significantly higher than I thought
16:33:29*freqmod congratulates amiconn high that we can't send one source frame for every internal frame :(
16:34:05JdGordondidnt you say it was too low before?
16:34:22amiconnYes, that was a measuring error
16:35:59amiconnWe can do every second frame, and that eats 75% CPU time - when boosted, on mini! :(
16:36:32linuxstb_Ouch... You do have a COP to use though...
16:37:00amiconnThat won<#t help
16:37:22amiconnIt'd still be impossible to drive every single frame
16:37:47amiconnThe lcd controller access is the bottleneck
16:38:02linuxstb_Yes, but it frees the CPU.
16:38:10amiconnWe can do ~60fps on mini when boosted, but the internal frame frequency is 88Hz
16:38:58linuxstb_So how does 44fps look?
16:39:30amiconnOn H1x0, we can do 360fps unboosted, and only need 70 for greyscale... :/
16:39:31 Quit infamis ("CGI:IRC (EOF)")
16:39:41linuxstb_Ouch again.
16:40:34dan_aIs that a hardware design problem or is it the iPod LCD driver?
16:41:25amiconnIt's a lack-of-knowledge problem
16:41:45*amiconn wants PP docs, damnit!
16:42:27amiconnlinuxstb: 44Hz is bearable, comparable to the grayscale lib on archos when the archos got somewhat warm
16:43:13 Quit barrywardell_ ()
16:43:17amiconnThe gamma is weird.... need to measure
16:44:06amiconnThe natural gamma (ignoring nonlinearity) seems to be < 1
16:47:05 Join Spida [0] (
16:53:09*freqmod thinks he has to keep the implementation of a per page ringbuffer to support continue packages
16:55:36 Join ehour [0] (
16:55:52 Quit ehour (Client Quit)
17:04:59***Saving seen data "./dancer.seen"
17:12:41 Quit JdGordon ("Konversation terminated!")
17:14:43 Join joe2 [0] (
17:15:13FebsHmmmm. What's going on with this? ../tools/configure: line 472: syntax error near unexpected token `<<<'
17:16:57freqmodcvs difference error
17:17:22freqmodgrep for the <<< token and decide how to merge, the data before === is the original, and after is cvs
17:19:16 Join agabus__ [0] (
17:25:51 Quit YouCeyE (Read error: 110 (Connection timed out))
17:27:03 Quit BloodySorcerer (Read error: 104 (Connection reset by peer))
17:32:12 Quit agabus_ (Read error: 110 (Connection timed out))
17:32:34 Join pi [0] (
17:37:07pican someone with commit access have a look at fixing francais.lang as highlighted in FS #5741
17:39:27 Join cismo [0] (
17:44:20 Quit freqmod (Read error: 104 (Connection reset by peer))
17:47:06markunpi: committed
17:48:42 Join |Neekeetos| [0] (
17:54:22 Quit |Neekeetos| ("CU")
17:54:35 Join freqmod [0] (
17:55:27 Quit Criamos (Read error: 110 (Connection timed out))
17:59:02 Join moixa [0] (
18:00:41moixahiall. quick question about upgrading. i want to upgrade my existing .rockbox (on my x5) with the current cvs. can I just overwrite everything or should i backup my settings (where are those?), delete everything and do a clean install?
18:01:17pioverwriting should be fine
18:01:22linuxstb_I always upgrade by just overwriting.
18:01:37moixaok, thanks. i was just wondering.
18:01:43moixawhere are the settings saved anyway?
18:01:50linuxstb_Sometimes a new version of Rockbox will reset your settings. So before you upgrade, export your settings to a text file using the "Manage Settings" option.
18:02:08moixaah. good point.
18:02:11linuxstb_In an otherwise unused disk sector between the partition table and the first partition.
18:02:36moixaI see. ok, I will backup and then overwrite it. cheers.
18:04:34 Quit Galois ("Leaving")
18:05:30 Part linuxstb_ ("Leaving")
18:06:02 Quit cismo_ (Read error: 110 (Connection timed out))
18:08:05moixabtw. at the iaudio list item shoud link to
18:08:12 Join Galois [0] (
18:08:16moixa(or give me, TobiasSager, write access... ;-)
18:10:36obomoixa: you now have write access
18:10:38 Join bluebrother [0] (
18:11:00moixaobo: Thx
18:12:42moixabye everybody
18:12:44 Part moixa ("/me says bye bye.")
18:27:02 Part freqmod ("Konversation terminated!")
18:27:16 Join webguest38 [0] (
18:27:36 Join Daishi [0] (
18:29:22 Quit webguest38 (Client Quit)
18:40:51*dan_a reminds himself to check builds on all architectures, not just the ones he thinks he's making changes to.
18:45:11*dan_a repeats himself, and then runs and hides
18:49:21amiconndan_a: Did you already run test_fps?
18:50:14dan_aamiconn: Not yet, but I'll do it tonight
18:50:27Acksawanyone good with setting up a EQ on my H320?
18:50:33Acksawwith rockbox.. of course
18:50:36Mikachudid you say the old bootloader will start current rockboxes correctly and the new won't start old, or the other way around?
18:50:51amiconnIf you get ~50fps boosted then it'll be sufficient for the grayscale lib
18:50:55markunAcksaw: still didn't figure it out?
18:51:13markunwhy do you want to setup a EQ?
18:51:32Acksawto make my music sound a lil better
18:51:50dan_aMikachu: Old rockboxes won't work with the new BL. New rockboxes will work with the old.
18:52:01markunAcksaw: what's lacking right now with your headphones?
18:52:08Mikachudan_a: that is good then
18:52:16Acksawlil bit of treble wouldnt go a miss
18:52:26dan_aMikachu: There's no point in upgrading until something is actually *done* with the COP
18:52:38markunjust move slider 1 and 5 until it's right
18:52:44Mikachuyeah i just wanted to be sure it wasn't that old bootloader can't load new rockbox
18:53:16Bagderdan_a: just note that PP is now used on more (possible) platforms than ipod
18:53:35BagderI meant like for commenting etc
18:54:04dan_aBagder: I'll remember that.
18:54:23dan_aIt's only really the Sansa and H10, though, isn't it?
18:54:29Bagderright now, yes
18:54:44FebsAcksaw: I have a post on Mistic River with some suggestions on how to use the parametric EQ. Have you seen that?
18:55:38AcksawI haven't, no
18:55:45Acksawcare to point me in the right direction?
18:56:13Acksawcurrently i have 80Hz on 2.0 and 12000Hz on 2.0
18:56:16 Quit pondlife ("byebye")
18:56:16Acksawrest are 0
18:56:26Acksawbut it still sounds a little "dull"
18:57:12AcksawFebs I have mx550's if thats anyhelp
18:57:56 Join pixelma [0] (i=pixelma@
18:59:15FebsAcksaw: here's a good place to start. Band 0 is similar to the bass boost function, except that you have more control. Likewise, Band 5 is similar to the treble boost function.
18:59:55Acksawi just want a little more spark
19:00:01Acksawinmstead of a dull tone
19:00:39FebsIf you set the frequency for Band 0 to 300hZ, and the frequency for Band 5 to 1.5kHz (1500Hz), you will have an approximation of the bass and treble boost functions on bands 0 and 5.
19:01:03Acksawi actually have a EQ on foobar
19:01:09Acksawthat i'd like to port someone across
19:01:25FebsOK. That should be easy.
19:02:05Febs(though you may be more limited in Rockbox because you have only 5 eq filters)
19:02:18Acksawwould you like to see the setup?
19:05:02***Saving seen data "./dancer.seen"
19:05:06FebsOk, try this:
19:05:14FebsBand 0 gain=0
19:06:25FebsBand 1 Freq=80Hz, gain=5
19:07:05FebsBand 2 Freq=330Hz, gain=-5
19:07:12 Join xmixahlx [0] (n=xmixahlx@
19:07:37FebsBand 3 Freq=1.8kHz (1800Hz), gain=1
19:08:58FebsBand 4 Freq=5kHz, gain=3
19:09:31Acksawhow do i change the first number on the freq?
19:09:31FebsPlay with the Q settings to suit your taste.
19:09:46Acksawas its on 3992 atm
19:09:54Acksawand i dont know how change the 2 to 0
19:10:11 Join jaczehack [0] (
19:10:45FebsDo you mean change 3992 to 3990? Or change from band 2 to band 0?
19:10:56Acksaw3992 to 3990
19:11:17FebsDon't worry about it. It's a .02% difference, which is probably completely inaudible.
19:11:26Acksawwhat about band 5?
19:11:33FebsActually it's less than that.
19:11:52FebsThere is no band 5. It had a brain freeze when I wrote that above. It should be bands 0 through 4.
19:12:10Acksawi have 5 bars
19:12:26Acksawand you gave me up to band 4
19:12:28Febs0,1,2,3,4 ==> 5 bars
19:12:39Acksawwhat freqeuncy is 0?
19:13:06FebsConsidering that I suggested that you set the gain to 0, the frequency doesn't matter because you aren't changing anything.
19:15:04FebsNow, all that said, I suggest that you do this: save the EQ settings that I just gave you in case you want to go back to them.
19:15:53FebsSet the gain for all channels at 0
19:16:32FebsTake channel 1, 2, or 3 (NOT 0 or 4). Turn the gain up fairly high (all the way to 24 if you want).
19:16:37FebsSet Q to about 5.
19:18:04FebsSet frequency at 50Hz. Gradually raise the frequency from 50Hz through the 18kHz (or whatever the top frequency is). Stop from time to time and listen to how the sound of your music changes.
19:19:02Acksawdoing that now
19:19:26Acksawsounds like its echoing
19:19:27Febs(Alternately, set all of the slides in foobar to 0. Raise the slides one at a time. Then lower them one at a time. Listen to how the sound changes.)
19:19:53Acksawthat sounds reallly bad
19:20:03Acksawits like whistling
19:20:07FebsThe point is that by doing this, you will learn how the various frequencies affect the sound of your music, and you can then use that knowledge to adjust the EQ on your own.
19:20:39FebsBecause when it comes right down to it, there really is no substitute for listening.
19:20:58Acksawi just saved that eq
19:21:02Acksawnow i cant find it in the list =\
19:21:54FebsOne more hint: as you sweep from the bottom frequency all the way to the top, if you hear the music get suddenly louder, that is a frequency you should cut in your EQ. If you hear the music suddenly get quieter, that is a frequency you should boost.
19:22:07 Join BHSPitMonkey [0] (
19:23:25 Join mikearthur [0] (
19:24:05FebsThus endeth the lesson.
19:24:11 Nick BHSPitMonkey is now known as BHSPit[Birfday] (
19:24:28preglowFebs: for that to work reliably, the sweep amplitude should be adjusted by an equal perceived amplitude curve
19:24:48preglowif you just do a flat sweep, it'll sound like it does dips even if nothing is wrong
19:25:26preglowunluckily, such a curve is dependent on sound pressure level as well
19:25:40preglowmight perhaps just use an a weighting curve or something
19:25:54 Quit javkrei ("CGI:IRC (Ping timeout)")
19:26:04preglowas in what replaygain uses
19:26:13 Part jaczehack
19:26:19*Febs seems to recall that he once he had idea for a plugin to help with that ...
19:26:32preglowi did start coding one
19:26:38preglowall it currently does is a linear sine sweep
19:28:49 Quit dan_a ()
19:30:12 Join barrywardell [0] (
19:31:09 Join petur [0] (n=petur@rockbox/developer/petur)
19:32:19lex hah my real desktop ;D
19:33:30 Join hannesd_ [0] (
19:33:57 Quit tvelocity ("Ex-Chat")
19:34:57 Join hannesd__ [0] (
19:36:11BHSPit[Birfday]that's terrifying, lex
19:37:09BHSPit[Birfday]strange, that we both run and support linux, and we both have Microsoft mice and keyboards :D
19:40:31 Quit hannesd_ ("Client suicide")
19:41:51markunpetur: I also had my iriver skipping when I was running one time
19:41:59markunbut ot
19:42:17markunbut it's a bout 1.5 years ago and the hd is still fine
19:43:42peturstill, if you're moving it so that it starts to skip music, the hdd won't like it
19:44:11markunI don't take it with me anymore when I'm running
19:44:23markun(and don't run much anymore :)
19:46:48 Quit lodesi ("Parti")
19:47:18 Join lodesi [0] (
19:51:15 Quit spiorf (Remote closed the connection)
19:52:47 Quit obo ("bye")
19:56:18 Quit pi ("Trillian (")
19:56:49 Join pi [0] (
20:07:02 Quit lodesi ("Parti")
20:13:38 Join mkey [0] (
20:14:34 Join Terinjokes [0] (
20:16:00Terinjokesbhspit, today your birthday?
20:19:14 Join pondlife [0] (
20:20:23 Join menollo [0] (
20:20:43pondlifeA little heads up to another talk bug fix.
20:21:53pondlifeIt allows use of .talk clips with no voice file loaded (previously this crashed).
20:22:13 Join spiorf [0] (
20:22:56pondlifeThere's a very minor issue remaining, only caused if you have no voice file and had .talk clips disabled, and then you enable .talk clips wirhout rebooting. ButI'm off on holiday now, so won't be able to go any further with it.
20:22:58Terinjokesis .talk for voice control, or having rockbox announce song title?
20:23:08pondlifeSong titles
20:23:16pondlifeWell dir and file names
20:23:48pondlifePersonally I like to dir .talk clips when driving, and have no need for a voice file
20:24:01pondlifeto *use* dir
20:24:30Terinjokesi think i have that option turned on, maybe it doesn't work on the ipod
20:24:31 Join jaczehack [0] (
20:24:45pondlifeYou need to generate the .talk files too.
20:24:56 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
20:25:13pondlife for details
20:26:15pondlifeMuch nicer than having filenames spelt out IMHO
20:26:32pondlifeAnyway, gotta pack now. Bye all
20:26:43 Quit pondlife ("gone!")
20:27:16 Join lodesi [0] (
20:31:49pilli have an idea
20:32:06pillthere is a mod to do on the ihp1x0 to allow charging via usb
20:32:46pillwould it be possible for rockbox to have a "Allow USB charging when plugged ON/OFF" so that when you plug your unit for like 3 minutes it wont start a charging cycle?
20:33:30Terinjokesi don't understand, you DON'T want it to charge?
20:33:39pilli dont want it to charge eveytime
20:33:49pillit would kill my battery
20:33:54bluebrotherpill, I don't think this would work
20:34:07bluebrothercharging is done in HW on the h100 series.
20:34:30bluebrotherI don't know the circuits but I'm pretty sure there isn't a way to prevent charging.
20:34:51pillso if i do the mod
20:34:55sharpebluebrother: disconnecting the cable?
20:34:58pillit will charge everytime
20:35:08pilli plug the usb cable
20:35:23bluebrothersharpe, hehe. Nice option.
20:35:37sharpewell, it's pretty universal. :)
20:36:12bluebrotherbut maybe a bit ... manual ;-)
20:36:57pillthis is a leet hack
20:39:38bluebrotherpill, you could build a usb cable that doesn't carry power.
20:39:58bluebrotheror add a power switch to a usb cable ;-)
20:40:11Terinjokesor you don't plug the cable in!
20:40:31pillusb that doesn't carry power
20:40:42pillthat's a brilliant idea
20:40:55Terinjokesok, i have a question about voice files
20:40:57bluebrotherjust break the 5V line in a cable with a switch.
20:41:15bluebrotherbut I don't know how much usb itself will like that. At least you could try.
20:41:40Terinjokesi wish to use but their is no link to download it
20:41:43 Join javkrei [0] (
20:42:27 Join webguest71 [0] (
20:42:41 Quit webguest71 (Client Quit)
20:42:53 Join hotwire_ [0] (
20:44:06 Join dan_a [0] (n=dan-mirc@
20:44:17pillmaybe there could be a way to have the ihp drain power from the usb
20:44:23pillnot charge
20:44:30pilljust be powered by the usb host
20:44:46bluebrotherTerinjokes, AT&T files have been removed due to copyright issues.
20:44:51amiconndionoea: Around now?
20:44:59bluebrothersee the note on the top of the VoiceFiles page.
20:45:14Terinjokesbluebrother, you guys did a horrible job at "removing" it
20:45:30Terinjokesbut yes, i see the warning now
20:46:57hotwire_hey folks, i'm trying to write a plugin, but would like to talk to somebody about the right data structures to use for it
20:48:27Terinjokesah! the voice file is broken!
20:48:45 Quit hannesd__ (Read error: 104 (Connection reset by peer))
20:49:03Terinjokesit is reading off the wrong menus!
20:50:24 Join elinenbe_work [0] (
20:50:55dionoeaamiconn: ?
20:51:39amiconndionoea: I found 2 graphical bugs in solitaire only affecting non-archos targets
20:52:18dionoeabig color screens ? :)
20:52:22dionoeaand the numbers ?
20:52:30amiconn(1) The bitmaps for the empty final stacks are completely messed up
20:52:47dionoeaok, that's the big color screens thing (or maybe not so big)
20:52:59amiconnNo, also 4-gray
20:53:19amiconn(2) (colour only) The number bitmaps have a green background. Looks odd on a white card....
20:53:20dionoeawell bigger than 64 pixels high screens :D
20:53:21Terinjokesbluebrother, i'm a little confused about voice. what do i do if i want it to read off the name of the song
20:53:35dionoeayeah, i noticed those two bugs ... i"ll fix them :)
20:54:02bluebrotheryou need to generate .talk clips or let the voice spell the filenames to you.
20:54:04amiconnpixelma found 2 more bugs which affect archos, and possibly other targets
20:54:16dionoeaamiconn: what are they ?
20:54:34 Quit barrywardell ()
20:54:49Terinjokesok, i have it set up for reading filenames, but it kinda crazy, going like "j4xx56jjxpercentbattery"
20:55:13 Quit menollo ("CGI:IRC (EOF)")
20:55:16amiconn(3) The suits bitmaps for the 2 lower cards in "draw three" are "falling out"
20:55:44dionoeawell they're on the left so we can see them. Do they overflow the card ?
20:55:58dionoeahum ... ok
20:56:24dionoeabtw, anyone here know how to change a bitmap's palette in the gimp without converting to 24bit colors and back to another palette ?
20:56:53bluebrotherTerinjokes, have you installed the correct voice file? You can't use a cvs build with a non-cvs voice file.
20:57:18Terinjokesyeah, i'm on a CVS build with a CVS voice file
20:57:29dionoeaamiconn: what was the 4th bug ?
20:58:00 Join hannesd_ [0] (
20:58:46dan_aamiconn: I've just done test_fps on a 3g. 92.5fps when the CPU is at 75MHz
20:59:08amiconn(4) has to do with "draw three" as well. When putting 2 of the three cards elsewhere, the 3rd card vanishes
20:59:36dionoeahum... was that after or before my 2nd solitaire.c commit ?
20:59:57pixelmaI thought it was after.. but I'll try again
21:00:00 Join pepie34 [0] (
21:00:29dionoeai though that i had it fixed ... but since i was really tired i might have been wrong
21:00:40amiconnA maybe related observation by me is that on recorder, if you push play, even if the card can't be put where the cursor is, one card vanishes from the drawn ones
21:00:57 Quit ze (Remote closed the connection)
21:01:11amiconnIf all three have vanished and you press play again, all 3 cards reappear
21:01:19*dionoea knew that there were going to be problems after that commit :)
21:01:53pepie34in the last cvs commit, it sid to have some work done on coprocessor on ipod
21:02:01pepie34what has been done ?
21:02:27dionoeabtw, is it still safe to use CVS builds with the old bootloader on ipods ?
21:03:19pepie34dionoea, is that what you are alking about: ipod coprocessor commit?
21:03:29dan_adionoea: Yes, just not the new bootloader with the old builds
21:04:46dan_aBasically, the new bootloader doesn't turn the coprocessor off - it leaves it up to the code to do that. If you use an old build with that then the coprocessor runs around doing all sorts of random stuff to data
21:04:51Terinjokeswhat does the ipod coprocessor commit do?
21:05:05***Saving seen data "./dancer.seen"
21:05:07amiconndan_a: 92fps? Wow! That's faster than 4g and mini!
21:05:22dan_aTerinjokes: It just gives developers a way to start putting code on it
21:05:23amiconnThat's the full screen value (1/1)?
21:05:31pepie34dan_a what is the purpose to do that on the bootloader ?
21:05:45Terinjokes[/sarcasm] does it fix rockboy?
21:05:52dan_aamiconn: Yep: the 1/4 was something like 300fps!
21:05:55pepie34I 'have ssen some code to wake the cop from the cup in ipodinux
21:06:42 Quit Farpenoodle ("Large sacks of cheese")
21:06:44dionoeadan_a: so it'll take more battery ?
21:06:44amiconndan_a: Then I don't expect problems for the grayscale lib on 3g. Maybe we can even go full-rate!
21:06:52dionoeaerr ... nevermind
21:06:59dan_apepie34: The loader for rockbox code in both IPL's loader and ours was putting the COP to sleep, which meant that we couldn't use it from inside the program
21:07:04amiconn(but that would probably eat almost all CPU)
21:07:18pepie34yes i understad that
21:07:22dan_aamiconn: Oh well, we've got 2 of them!
21:07:22amiconnI expect the lcd frame frequency to be around 78Hz like on 4g
21:07:34midgey34dionoea: why does the suitsi bitmap look so bad?
21:07:54dionoeabecause 1/ the bitmaps are horrible and 2/ stride is wrong in the ocde
21:07:55pepie34but you can wake it up then for the program
21:07:56amiconndan_a: Now if you can make up an api to use the cop from within plugins...
21:08:03dionoeamidgey34: but i'm fixing that :)
21:08:21midgey34I thought I was using the same array in my blackjack code and it looks fine there
21:08:35dionoeait's using bitmaps since yesterday
21:08:36dan_aamiconn: 2 weeks ago I barely knew about APIs, and just wanted the titles to scroll in my iPod.
21:08:44dan_aIt was harder than I thought...
21:10:28pixelmadionoea: so you did find it?
21:10:38pepie34you need to write 0x0 to 0x60007004 to wake it up from cpu
21:10:46dionoeapixelma: i just finished fixing the bitmaps ... i'll start looking now :)
21:10:59amiconndionoea: Btw, what was the problem with the 16bit bitmaps again?
21:11:18dionoeathat compile problem i had yesterday ?
21:12:07dan_apepie34: The problem is that it then starts running from where it had stopped. In the meantime memory has been remapped so we have no control over what the coprocessor is doing.
21:12:08dionoeawell the file didn't exist and wasn't supposed to exist ... but i was too tired to notice that :/
21:12:55dionoeakind of silly
21:13:08amiconnYeah, don't tell me about silly bugs...
21:13:13*amiconn made several today
21:13:30pepie34dan_a for now in ipodlinux it is even not pu to sleep
21:13:50amiconnE.g., what is wrong here: for(i = 7; i >= 0; i++) ?
21:13:52*dionoea notices that numbers are always black now on color screen targets ... rha :(
21:13:58pepie34it is used as a kind of dma
21:14:06dionoeaamiconn: hehe, this is one of my favorites :)
21:14:07Mikachuamiconn: nothing if you want it to loop 2^32 times
21:14:16elinenbe_workamiconn: how far away are you from anti-aliased fonts? ;-)
21:14:26pepie34it enters a loop that copy the audio buffer to the codec device
21:14:35*amiconn slaps elinenbe_work around a bit
21:15:20pixelmanow I tried with a bleeding edge... if that's a hint: every time I put one card from the "draw 3 heap" away with "long left" (on Ondio) the last of the three vanishes
21:15:36*dionoea hopes that JdGordon's action keymap patch will be commited soon so i can clean plugins code a bit
21:16:22dionoeai'm wondering if i should go back to the old layout for the small archos screens
21:16:25*amiconn starts tackling unbuffered grayscale lib mode
21:16:26midgey34dionoea: are these of any use?
21:16:36dionoeamidgey34: i already fixed them :)
21:16:40dionoeanot commited yet
21:16:41midgey34well then
21:16:44elinenbe_workamiconn: hey... watch that! I just think that is resonable to ask of you :-)
21:16:45midgey34good job
21:16:46dionoeathanks anyway
21:16:48*elinenbe_work hides
21:17:01pixelmaand I couldn't access the very last card...
21:18:15Terinjokesi know this is kinda unsupported, but can anyone tell me how to add album art to a skin?
21:18:19dionoeawell i only changed display code ... i'll double check the diff
21:18:21amiconndionoea: Just a hint: the bitmap build system creates an appropriate .h file for every bmp it converts
21:19:15pixelmamaybe that bug was in before - I can't tell because I wasn't able to play that for a while
21:19:30dionoeayeah, might be
21:21:14dionoeahum numbers+suit height is just bigger than the card's height ... that's problematic
21:21:28dionoea(for archos)
21:22:51pepie34dan_a, may be i have an idea: we can set the interrupts table before wakig up the COP, trigerring a software interupt then put the cop up
21:23:31bluebrotherwould be nice if solitaire would use greyscales ... it's pretty hard to distinguish the colors here.
21:23:50pepie34I don't know where the pc is store before going into the interrupts on arm
21:23:52dionoeapixelma: hum, the play button thing is indeed weird. I think that it was already in the code before my commits yesterday... i'll try to fix it
21:23:58dan_apepie34: the interrupts table on the COP?
21:24:42pepie34I have a discussion with some guy involved on ipodliux
21:25:14dan_aIt's easy enough to sleep and wake the COP from the main thread now: you just set COP_CTL = PROC_WAKE
21:25:21pepie34and i told me we can either affect each interupt on cop or cpu
21:25:23 Join barrywardell [0] (
21:25:33 Quit hannesd_ (Read error: 60 (Operation timed out))
21:26:21pepie34Can we still use the ipl?
21:27:00dan_apepie34: Of course!
21:27:04 Part jaczehack
21:27:44 Join muesli__ [0] (i=muesli_t@
21:27:51Terinjokesi'm bored!
21:27:56dan_aI'm not exactly sure what we would want to do with interrupts on the COP - I'm fairly new to all this. Feel free to write some code though!
21:28:09 Join lee-qid [0] (
21:28:16pepie34Well i do't really know arm asm
21:28:22FebsBagder: around?
21:28:38dan_apepie34: It's easy enough to learn
21:28:50 Quit javkrei ("CGI:IRC")
21:29:15pepie34my ida was to put the cop on interrupts as soon as it wakes up
21:29:27 Quit Terinjokes ()
21:29:31pepie34and as we can set the interupts table
21:29:33dan_a(I say this as someone who'd not looked at ARM ASM before a few weeks ago, and who hadn't done any programming for 5 years)
21:30:05pepie34we can make execute the code we want as soon as it wakes up
21:30:21 Quit mr_cheez (Read error: 110 (Connection timed out))
21:31:46pepie34the only dirty thing is to modify the pc that is saved somewhere (on the stack i assumed) for the interupt return
21:33:01dan_apepie34: The COP has its own PC
21:33:15dan_aIt's a complete second processor
21:33:30pepie34yes but when a cpu goes into an interupts
21:33:41pepie34it push the currnt pc somewhere
21:34:09pepie34to return to correct normal code after the interrupt s cleared
21:34:31pepie34on 68k it is on the stack
21:34:48pepie34don't know for the arm
21:35:50Mikachuthat is a clever trick
21:35:51pepie34usually all the proc register are push on the stack on interupts
21:36:25 Join tvelocity [0] (
21:37:18 Quit elinenbe_work ("CGI:IRC (EOF)")
21:38:18pepie34the simpliest way to make the cop execute what we want after a wake up is to never return from interupt
21:38:22linuxstbAFAIK, the COP and main CPU share the same interrupt handlers - there is only one set of vectors.
21:39:03linuxstbInterrupts can be independently enabled for the two processors though.
21:39:15dionoeapixelma: i really can't find a pretty way to display both number+suit on the two lower cards from the remains stack on small screens
21:39:17pepie34and with the interupts enable mask we can enable one interupt only to the cop
21:39:21 Join lodesi_ [0] (
21:40:21linuxstbpepie34: Yes, but what interrupt, and why? What's wrong with just setting a function and making the COP run it, which I think is what dan_a is doing.
21:41:00pepie34linuxstb may be it is power consuming
21:41:22pepie34no i just find the interupt thing a good trick
21:41:28dan_apepie34: Not if you do COP_CTL = PROC_SLEEP when you don't need it!
21:42:09linuxstbpepie34: It just sounds like a trick without a purpose...
21:42:18pepie34linuxstb thank you
21:42:35amiconnpepie34: arm uses shadow registers for interrupts
21:42:42Mikachucouldn't it be used to wake up the cop and gain control over it if the pc was unknown?
21:42:43amiconnDunno whether the pc is one of them
21:43:42pepie34linuxstb the ipl put the cop to sleep ?
21:43:50linuxstbIsn't it just FIQ mode that has the banked registers? AFAIK, a normal interrupt just uses the standard registers.
21:43:59pepie34how linux put the cop up ?
21:45:01pepie34Mikachu, yes it is what it was intend for
21:45:03pixelmadionoea: I try to think of something too... btw. there's also another small graphical glitch: there is no space between 1st and 2nd column, as well as 4th and 5th column
21:45:04linuxstbpepie34: Yes, IPL puts the COP to sleep. I assume it's woken automatically to handle the FIQs, and then sleeps again automatically when the FIQ returns.
21:45:33pepie34but i read the linux boot code
21:45:38linuxstb(but I haven't looked at IPL's FIQ code in detail - Rockbox uses the main CPU to handle the FIQ).
21:45:45pepie34and both proc are running at boot
21:45:45dionoeapixelma: yep, i fixed that :)
21:45:50linuxstb(I mean IPL's COP code)
21:46:03dan_aI'm off out now - having had a data abort trying to run a thread on the COP - back later!
21:46:13 Nick dan_a is now known as dan_a_away (n=dan-mirc@
21:46:13linuxstbpepie34: Yes, they're both running at boot, and then the COP puts itself to sleep.
21:46:43pepie34by linux boot code i was alking about the actual linux boot not the ipl
21:46:53 Join nls [0] (
21:47:06nlsbluebrother, here?
21:47:07*amiconn also encountered data aborts today
21:47:24linuxstbpepie34: Yes, I know.
21:47:48linuxstbThe bootloader wakes the COP up again before running the kernel.
21:48:03bluebrothernls, yup.
21:48:08pepie34that is what i did not understand
21:49:02pepie34so ipl can set the PC to (do a jump on) both CPU and COP from the CPU
21:49:30nlsbluebrother, I saw that the "Manage Settings" menu is under system options in the configure chapter while it's actually in the eop level of the menu... (it's in the right place in the "main menu" chapther tho...)
21:49:32 Quit Poka64 ("nite all")
21:49:47linuxstbNo, the way it works is that the COP puts itself to sleep. But the next instruction the COP executes when it wakes up is to load an address from a memory location, and put that address in the PC.
21:49:59nlstop level even
21:50:00 Quit lodesi (Read error: 110 (Connection timed out))
21:50:10pepie34:) kool
21:50:12 Nick lodesi_ is now known as lodesi (
21:50:33pepie34thanx for the information linuxstb
21:50:34 Quit Daishi ("Client exiting...")
21:50:42bluebrothercorrect, that's wrong. Hadn't noted that before.
21:50:48nlsand I think we should move it but where?
21:51:57bluebrotherbtw, have you seen the ManualStructureTalk page I've set up in the wiki?
21:52:21bluebrotherI'm somewhat unhappy with the separation of the configuration in two chapters.
21:52:27dionoeapixelma: i might have a solution. I'll commit so you can give it a look
21:52:39pixelmanice :)
21:52:49dionoeait's a bit ugly ... but who cares ? :)
21:52:51nlshad a quilck glance but haven't read through it all yet...
21:53:19nlsI'll post ant comments in the wiki
21:53:35pixelmadionoea: it cannot be uglier than before, can it? :)
21:53:41dionoeai guess not :)
21:53:50bluebrotherhow about moving it to the main menu (where it's already mentioned) but separating the cfg file format out and put that into the advanced topics?
21:54:02dionoeaif that still doesn't work well ... i'll revert to the old (pre 1.20) layout for small screens
21:54:24bluebrotherI think we should make the wps section in advanced topics a bit more general ... cover cfg files, themes and (as a subset of that) wps.
21:54:28dionoeabtw, did you try doing a stack with like 6 unknown + 12 known cards on the last column to see how it looks ?
21:55:30nlsbluebrother, sounds like good ideas, I had that thought about themes and wps too.
21:56:58bluebrotherit's a bit the direction I tried to point out in the new structure proposal. Start as easy as possible with increasing difficulty level towards the end of the manual ;-)
21:56:59pixelmadionoea: I didn't count the cards but I had some "longer" columns - there was nothing odd that I would remember
21:57:01 Part pi
21:57:13Febsbluebrother: the configuration section was pretty much unworkable as a single chapter. We ended up with too many sub-sub-sub sections.
21:57:58bluebrotherFebs, I know. But I also don't like the current solution. Time to find a better one ;-)
21:58:04dionoeapixelma: commit done. You can give it a try :)
21:58:25amiconndionoea: Why not make the cards _a bit_ higher for archos (to fix the 'falling out' graphics)?
21:58:27bluebrotherBut when we do a restructuring so that the basic topics are covered in the first chapters this might become better.
21:58:31 Quit lodesi ("leaving")
21:58:36amiconnCurrently the cards are wider than tall, which looks odd
21:59:08dionoeaamiconn: well making them higher means that you really won't see much if you have lengthy stacks
22:00:07amiconnI don't think it makes a big difference. Only the top card of each stack, and the top row of the screen (the draw stack and the final stacks) will become higher
22:00:21Febsbluebrother: I like the proposal to change the chapter names to remove "Configuring Rockbox" from the Sound Settings and General Settings chapters.
22:00:37dionoeaamiconn: well i'll give it a try if the current solution doesn't look good enough
22:03:08pepie34do you have any code to deals with ipod radio remote?
22:03:13pepie34I can't find it
22:03:26 Join Hansmaulwurf [0] (
22:03:32pepie34and the volume buttons work out of the box
22:04:59Hansmaulwurfjust a fast question, should the H120 disc continue spinning in USB mode with rockbox or should it stop spinning after a time not using?
22:05:39amiconnH1x0 continuously spins the disk in usb mode
22:05:55amiconnThis has nothing to do with rockbox, but with how the USB bridge is set up
22:06:08amiconnAlso happens in retailos
22:06:10Hansmaulwurfhm, did this change in the past, i think i remember a time when the dis stopped spining
22:09:12nlsbluebrother, I think the proposed new manual structure looks goood. ;-)
22:10:12bluebrotherdamn, opt is quite tricky to get all spaces right.
22:10:57nlsyep, the % is quite useful but it's kind of unpredictable...
22:12:01nlsanyway time to sleep, g'night
22:12:03dionoeapixelma: ?
22:12:09bluebrotheri just tried it a different way. Still looking what's the best.
22:12:12 Quit nls (" HydraIRC -> <- The dawn of a new IRC era")
22:13:51pixelmadionoea: just finished (it takes a while for the Ondio)... that doesn't look very well (too crowded)
22:13:51bluebrotheranyway, bbl.
22:13:55 Quit bluebrother ("Leaving")
22:14:00dionoeapixelma: hehe :)
22:14:14 Join lodesi [0] (
22:14:40sharpeit's great seeing improper grammer in ads. "Who is you favorite designer?"
22:15:06dionoeapixelma: the only other solution would be to do like amiconn suggested. But i'm not 100% sure that it'll look better once you've drawn most of the cards
22:15:48amiconndionoea: How many pixels are missing to fix numbers and suits vertically?
22:16:13amiconnIf you make the cards 14x12 instead of 14x10, they will already be taller than wide on target
22:16:16dionoeai'd say like 2 to make it look acceptable, 4 or 5 to make it look good
22:16:26amiconn(due to the non-square pixels)
22:18:16pixelmaI'll try to find something the "graphic way" too, maybe I could suggest something later
22:19:05 Join freqmod [0] (
22:20:06 Join apo` [0] (
22:20:26dionoeapixelma, amiconn : does this look better ?
22:21:41amiconnHard to say on the PC screen... but there are some graphical glitches in it
22:22:20pixelmabetter than it is now... yes
22:22:33amiconn(1) The 4th and 5th column overlap by 1 pixel (one pixel border instead of 2)
22:22:40barrywardelljust looking at dan_a's latest commit of crt0.S, line 337
22:22:44dionoeahum ... the three draw cards overlap the spades stack
22:22:55barrywardellit looks a bit like the #ifdef's are mixed up
22:22:56pixelmabetween 1st and 2nd too (still)
22:23:15dionoeawell that's automaticaly computed at the moment (column spacing)
22:23:24amiconn(2) The final stacks are spaced by 1 pixel, but the three draw cards overlap the stack
22:23:42Bagderbarrywardell: I agree that it looks... wrong
22:24:04pixelmadionoea: you could push the final stacks together too
22:24:10barrywardellnot sure what he intended
22:24:25dionoeai'll push the 3 draw cards to the left
22:24:38 Quit pepie34 ("Leaving")
22:28:39Bagderbarrywardell: I'm afraid your patch to firmware/drivers/ata.c is not good
22:28:51 Quit tvelocity (Remote closed the connection)
22:29:29barrywardelli was a bit confused by that file. can you elaborate
22:30:16Bagderyes, that file is/can be used by target_tree targets too
22:30:31Mikachuisn't it a bit counterproductive to add h10 keybinds everywhere when the action thinger is about to be added?
22:30:34Bagderyou should rather just make sure it builds for you
22:31:16BagderMikachu: hm, maybe
22:31:32Bagderbut "just about to" can be very long at times
22:31:45barrywardelljust looking at the latest patch i'm updating, and I have that bit in ata.c removed
22:32:11barrywardelljust depending on the standard PP stuff
22:32:44barrywardellMikachu: I don't know much about the new action thinger
22:33:03barrywardellbut made the keybindings to get the sim and plugins to build
22:33:49 Join linuxstb_ [0] (
22:34:29FebsBagder: I noticed that memory size for a number of Archos platforms can be selected in configure. Just out of curiosity, how is that selection handled for the automatic builds?
22:34:56BagderFebs: just like the other selections, the script sets a value
22:35:08 Quit xmixahlx ("blah blah blah")
22:37:05linuxstb_Also, isn't some of the H10 code now in target-tree (e.g. ATA and USB) identical to other PP502x targets?
22:37:33Bagderyeah, it shouldn't be in target/
22:37:48pixelmadionoea, amiconn: what do you think of this: <- that's with the current card size and I changed the hearts and diamonds
22:37:50Bagderonly target-specific adjustments should be there
22:38:29Mikachuthat sort of looks like standing mahjong tiles
22:38:58Mikachuhow come only the bottom-most card has a rounded corner?
22:39:43dionoeapixelma: i like the hearts/diamonds change ... makes it more readable
22:40:05dionoeapixelma: did you like change the card width back to 14 px ?
22:41:15barrywardellBagder: just noticed your new cvs commits there
22:41:32pixelmadionoea: sorry I didn't get you?
22:41:39barrywardellthanks. makes my life much easier keeping up to date with changes you make
22:41:59dionoeawell the card width is now 15px so that the suit doesn't touch the card's border
22:42:15dionoeawhich makes it impossible to have 1px between each column
22:43:02barrywardelllinuxstb_: It is identical to other PP targets, but that stuff was already put into target tree by the iaudio port
22:43:20 Join webguest60 [0] (
22:43:25pixelmadionoea: card is 15px width (with the border).. I took that of a screenshot
22:43:50dionoeawell it still looks like before my 2nd commit today (or 1st ... don't remember)
22:43:55webguest60hellop anyone who knows how it works with rockbox on a nano?
22:44:29barrywardellso I had to either put duplicate code in target or edit the target_tree #ifdefs in ATA.h
22:44:37barrywardelli mean ata.c
22:45:01*freqmod has finished a better libogg implementation for speex
22:45:16pixelmadionoea: yes - but there are 2 pixels space between suit and numbers... maybe you can save 1 pixel there?
22:45:29*freqmod don't like Tremor's non standard libogg
22:45:38linuxstb_I haven't really looked at the target-tree stuff yet, but can't you just create a pp5020 directory for the pp5020-specific stuff, and move the relevant parts of the ipod code there?
22:45:54linuxstb_(and then also use them for the h10)?
22:46:28Bagderthey should probably be in target/arm/
22:46:32pixelmadionoea: btw. I actually drew it that way for the "draw 3" part.. :)
22:46:40dionoeayeah, i noticed
22:46:54amiconnlinuxstb: _move_ is an important word here, btw
22:47:13linuxstb_target/arm/pp5020 ?
22:47:22barrywardellBagder: I had thought about that, but would that mess up non-PP arm stuff?
22:47:28amiconncvs and move don't mix well
22:47:36barrywardellmaybe target/arm/ata-pp5020.c?
22:47:41Bagderpossibly, but currently the arm/* dir is named after the company that makes the device, not the one who make chips
22:47:58Bagderbarrywardell: yes, I like that
22:48:04 Join Kolya [0] (
22:48:05dionoeapixelma: but that would be too complicated to handle.
22:48:16dionoeamy last commit should fix it
22:48:18amiconnBagder: It's not (always) named after the company, and imho it would be a better idea to stick with the product line instead of company
22:48:20webguest60help with rockbox on a nano anyone?
22:48:28amiconn(like it is the case with iaudio)
22:48:38pixelmadionoea: didn't think of this :/
22:48:45Bagderamiconn: that's hardly the point here
22:49:16Bagderbecause PP is not the name of neither Apple nor ipod
22:49:33KolyaHi there, just installed rockbox on an iriver h300 and the music plays sloooow. What's up with this?
22:49:50pixelmadionoea: would it be possible if you would have different bitmaps there?
22:49:58KolyaApart from that it works great and looks even better.
22:50:04linuxstb_Where does code like FM tuner drivers, or audio codec drivers which might be used by two completely different targets live? Will that stay in firmware/drivers/ ?
22:50:14Bagderlinuxstb: yes
22:50:18dionoeapixelma: doesn't the current solution already look good enough ? :) (see my last commit)
22:50:27Bagderfirmware/drivers is for chip drivers
22:50:31barrywardellso stuff that is common to ipod/h10/sansa should go in target/*-pp5020.c?
22:51:00 Quit webguest60 ("CGI:IRC")
22:51:00pixelmalet me see...
22:51:06barrywardellI mean target/arm/*-pp5020.c
22:51:29linuxstb_Another complication is the pp5002 - a lot code is the same, but using differnt #defines from pp5002.h/pp5020.h.
22:51:34 Join Doomed [0] (
22:51:41Bagderbarrywardell: *-pp50xx.c then I guess, or perhaps just *-pp.c
22:53:53Bagderit also struck me that functions like in i2c-pp5020.c probably shouldn't have ipod_* prefixes...
22:54:25linuxstb_No, that should be generic.
22:54:45Bagderwe should probably have a generic front end so that we can have the same function call
22:55:09Bagdernot that it is very important
22:55:36Doomedcan somebody help me with updating source with the CVS
22:55:47dionoeapixelma: ok, i'll stop solitaire changes/commits for today :)
22:56:00dionoeagood night
22:56:17pixelmahehe.. ok - thanks and good night :)
22:56:29KolyaI found a pitch control but that doesn't help with mp3 files playing at half /speed/. HElp anyone?
22:56:59 Quit spiorf (Read error: 110 (Connection timed out))
22:57:52dionoeapixelma: hum ... my solitaire.c commit didn't make it
22:57:59dionoeai'll commit it again i guess
22:58:30pixelmaI was already wondering...
22:58:43dionoeait should be fine now
22:58:54dionoeasee you tomorrow
22:59:40barrywardellmost of the stuff in the h10 target tree is just dummy functions taken from the ipod code with the contents of the functions #if 0'ed out
23:00:02Doomedwhere do i get the .cvspass ??
23:00:25Bagderbarrywardell: yeah, I think this early in the port work a build that links is prio 1, cleaning it up to be nice is prio 2 once functionality actually starts to show
23:00:44BagderDoomed: you "cvs login"
23:01:07barrywardellok, so we can probably just leave things are for the moment. even if there is some code duplication
23:03:29 Quit freqmod ("Konversation terminated!")
23:03:41Doomedthen do i cd to where my source is and enter : cvs up -dP ?
23:04:40 Join freqmod [0] (
23:04:41KolyaSorry people for bothering you but I can't find this in the forums nor in the wiki or google: On a fresh rockbox h30 install the audio plays much too slow. Can someone help me please?
23:05:07***Saving seen data "./dancer.seen"
23:05:25*amiconn spots an odd commit msg
23:06:29Doomedlooks like H10 is coming along
23:07:25FebsKolya: what type of file are you trying to play?
23:07:39KolyaFebs: MP3
23:07:58FebsAnd does this happen with all of your MP3s?
23:08:05Kolyayes all of them
23:08:10*barrywardell hopes his H10 gets repaired soon so he can get back to it!
23:08:27FebsHave you somehow changed the playback speed setting?
23:08:33*amiconn expects cvs conflicts :(
23:08:51KolyaFebs: I wouldn't know. But it's a fresh install
23:08:57Doomedbarrywardell did u have to pay to get it fixed?
23:09:03KolyaIt was like that from the start
23:09:36KolyaOtherwise the software runs fine
23:09:52FebsTry clearing the settings by doing the following: Turn the player off. Press Play/On to turn it back on and IMMEDIATELY press the record button until you see the message "Settings Cleared"
23:10:01barrywardellDoomed: don't know yet. I got a mail from iriver today saying they had received it and "Our technicians will keenly check and test the device"
23:10:10Doomedah ok
23:10:12Doomedgl man
23:10:53KolyaFebs: Thanks man, that did the trick. :)
23:11:15KolyaI wonder whyt I changed there...
23:11:52 Nick dan_a_away is now known as dan_a (n=dan-mirc@
23:12:06markunfreqmod: don't you like tremor's libogg because it's a bad implementation or just because it's non standard?
23:12:40KolyaWell anyway, that's some great piece of software you have written there. Thanks to all involved. :)
23:14:02freqmodbecause it is non standard
23:14:33linuxstb_Are you going to try to get Tremor working with your new libogg?
23:14:38freqmodand linuxstb wanted a unified libogg library before committing speex
23:15:22freqmodspeex works without a malloc (but it requires one ringbuffer :( )
23:15:42freqmods/speex/ framing in speex/
23:15:54preglowdoesn't tremor use a libogg2 variant?
23:16:24preglowin any case i think the tremor libogg is actually more efficient than the standard one
23:16:39freqmodi heard something about that... I don't know if it would be any better to port tremor to use libogg...
23:16:51freqmodpersonally i would think it is just a waste of time
23:17:37freqmodwhat does linuxstb_ think?
23:18:02linuxstb_I don't feel that strongly about it.
23:18:16preglowwhat would the benefits be?
23:18:39freqmodthen i wont, we are talking about changing 1500 lines of code + to remove 1000 lines of code
23:18:41 Join safetydan [0] (
23:19:16freqmodpreglow: one less memcopy per data, but a tremor codec totally different from the original code
23:19:43freqmodand 1000 lines of codes less in rockbox (if speex uses the same framing.c
23:20:51preglowi say forget it for now
23:21:01preglowthey're gonna roll out libogg2 after a while anyway
23:21:11preglowin which case it'll need to be done again
23:21:19preglowit's supposed to be better suited for embedded work
23:21:55*freqmod wonders if libogg2 supports whole raw packet contents linearly in memory
23:23:33freqmodThen, back to the reason for the libogg reimplementation, what needs to be done to make speex considerable for cvs?
23:24:07preglowdoes it crash anymore?
23:24:14preglowwhat about seeking?
23:24:34freqmodno, not as far as i know, and i have been using it a lot to play audiobooks
23:26:22 Join Alladin [0] (
23:26:47Alladinanyone who is able to help me with the installation?
23:27:13markunAlladin: which player?
23:27:56markunDid you try the installation howto in the wiki?
23:28:32freqmodworks, but takes some time. Does not work on files <1min.
23:28:45preglowfreqmod: why not?
23:28:53preglowfreqmod: also, didn't it have gcc 4 issues?
23:29:10freqmodyes, but that was a problem with passing a variable to sqrt for sterio
23:29:43pregloware there many stereo speex files around? :>
23:29:53Alladini just wonder how to do with the bootloader because i have installed already installed the "loader2" which makes me choose between ipodlinux and apple´s original os
23:30:31Alladinit shall also be able to choose rockbox if installed...
23:30:45freqmodsterio still does not work quite right, but i have made a table for the most important cases, and a intSqrt function wich is farly accurate (but not enough for perfect playback)
23:31:07dan_aAlladin: I think if you have a "rockbox.ipod" file in your iPod then the loader2 will be able to boot it
23:32:12freqmodMy loader.conf, works with rockbox.ipod, however i have a farly new ipodloader 2 (cvs)
23:32:27Alladindan_a: so i just skip the bootloader part of the installation and juast runs the .exe file?
23:32:50freqmodjust download a .zip file and place it in the root directory of your ipod
23:32:50 Quit muesli__ ("ich will KEe!!!")
23:33:15dan_aAlladin: What freqmod said. But unzip it first! And get the fonts too
23:34:03freqmodspeex stereo are perfectly listenable. (and as jmspeex said stereo voice is not really much help)
23:34:11preglowheh, no
23:34:23preglowfreqmod: but anyway, i'm quite eager to commit it, and will look at it when i get time
23:34:30preglowwhich might be some time after the weekend
23:34:45Mikachucould be helpful if it's a dialog and the voice have their own channel maybe
23:34:48Alladinahaa. and what about the rockbox.ipod-file?
23:35:25freqmodI will make another patch (hopefully tomorrow) to import the new framing code, if it works when i listen to some audiobooks.
23:35:53freqmodhowever it is not tested on any players except ipods
23:36:00preglowfreqmod: has it been tested on something other than ipod?
23:36:03preglowah, right
23:36:08preglowi can test it on coldfire
23:36:23freqmodgood, would probably be quite slow
23:36:25dan_aAlladin: That goes in the top directory of the Nano
23:36:27 Quit Acksaw ("( :: NoNameScript 4.03 :: )")
23:36:32freqmodbtw. is coldfire big edian?
23:36:39Mikachudan_a, Alladin: you can put it in .rockbox too if you want
23:37:04Bagderfreqmod: yes
23:37:49 Join hotwire__ [0] (
23:37:50freqmodgood, because I haven't looked for edian problems :S
23:38:18amiconnThen there probably will be some
23:38:22amiconnarm is little endian
23:38:28freqmodyes i know
23:39:15freqmoddoes anybody know a way to test for edian problems on a x86?
23:40:47Alladinim sorry but im not so good at these things but here is what i deal with: I hve installed the loader2, downloaded the rockbox installer.exe and what im a bit worried for is if i just runs the installer the loader2 will not be able to boot rockbox... puuh
23:41:49freqmodalladin: download: and unzip that in the ipod root folder (d:\ or whatever)
23:42:25freqmodyou'll need this too:
23:44:58 Quit Rob2222 ()
23:45:31Alladinok.. now the first zip fil is on the nano
23:46:28freqmodthe second should be unzipped the same way
23:47:07Doomedermm im getting a error when i try to make , about lang files
23:47:20secleinteerhey, is there any way to move the scroll bar to the right side of the screen?
23:47:37BagderDoomed: have patches applied?
23:47:58Bagderthen I blame them
23:48:57Doomed*starts over
23:48:58Alladinok now im done!
23:49:09secleinteeris there any way to move the scroll bar to the right side of the screen?
23:49:27secleinteersong list
23:49:44Doomedwould that depend on the theme, thats what im thinking but dont take my answer
23:49:46Alladinfreqmod: do you think the loader will recognice rockbox automatically?
23:49:59secleinteerk thx
23:50:45freqmodi think so, you could try (and when you do, check your version of loader2)
23:51:19Alladinits 2.4
23:51:24*freqmod does not like the data abort the new framing library gives on ipod
23:51:38freqmodand rockbox does not appear?
23:52:03Alladini shall try...
23:53:12 Quit hotwire_ (Read error: 110 (Connection timed out))
23:54:24preglowthink i'll call it a night
23:54:56freqmodgnight (good idea
23:55:40Doomedermm how do u no the version of the loader
23:56:08 Quit Kolya (Read error: 104 (Connection reset by peer))
23:56:19freqmodit is written at the top of the screen when you boot
23:56:22 Quit Hansmaulwurf ("( :: NoNameScript 3.81 :: )")
23:58:37Doomedfor all versions of RB?
23:58:45 Join Kolya [0] (
23:58:50freqmodno the loader2 loader from ipodlinux

Previous day | Next day