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 2006-01-22

00:00:06amiconnlinuxstb: (unrelated to the bmp patch itself) I am wondering why you use 3 different bitmaps for the 3 different states in sudoku, and why you're using native bitmaps
00:01:04linuxstbWhy not?
00:01:13amiconnThe inverse variant could be simply created with a draw mode, and the alternative colours for the pre-filled numbers with foreground / background colours
00:01:15 Quit BHSPitLappy2 (Connection timed out)
00:01:43amiconnOf course this requires mono bitmaps
00:01:50linuxstbThat's only because the current bitmaps are unimaginative. The colour version at least could be nicer.
00:05:08amiconnThe patch looks promising
00:05:40amiconnOne thing is that I would like the bitmap width and height macros as part of the build process instead of hardcoding them
00:06:07amiconn(e.g. in icon.h)
00:06:40preglowwhat is seamless seeking?
00:06:54lostlogicpreglow: seeking where the audio doesn't ever stop
00:06:57lostlogickinda like gapless
00:07:00lostlogicbut for seek
00:08:00linuxstbamiconn: Yes, so would I. But I think that can be added later. It would mean producing a new .h file, which is included by the files using libbitmaps.a and would introduce a nice dependency problem.
00:09:53amiconnThere are already some dynamic .h files in the build system
00:10:40linuxstbOK, it should be fine then.
00:11:20 Quit Pi___ (Remote closed the connection)
00:11:21amiconnOkay, not some, but one, lang.h
00:12:15amiconnYou might need a special workaround using touch, the same way as for lang.h
00:12:17preglowlinuxstb: it actually doesn't seem like retailos uses clickwheel acceleration
00:12:39BHSPitMonkeyhi linuxstb / preglow
00:12:42preglowBHSPitMonkey: hi
00:12:55linuxstbpreglow: I haven't really paid attention.
00:12:55BHSPitMonkeyanything new?
00:13:18preglowlinuxstb: just thinking of whether i should bother trying to implement it
00:13:31preglowif apple hasn't, i see no reason to do it myself, heh
00:14:14linuxstbNo, I'm not convinced it will be needed. I think the wheel works very nicely now.
00:14:38BHSPitMonkeyretailos's accel. is very handy with very large playlists :)
00:16:12preglowdoes it have accel, then?
00:16:22preglowif so, it doesn't apply everywhere
00:16:43preglowoh, it does indeed have it in list view
00:17:33preglowi really think we'll need some special handling of the wheel some day
00:17:51preglowor we'll need a really, really high event density for the biggest acceleration values
00:17:57 Quit FingerSoup ("Relax, its only ONES and ZEROS!")
00:21:23BHSPitMonkeyi'm pretty sure podzilla has wheel accel. too
00:21:27linuxstbamiconn: As I said, I agree the bitmap sizes should be in dynamically created .h files, but I would prefer to do that after "stage 1" is done. I don't like doing too many changes at once.
00:21:41BHSPitMonkeybut I don't currently have a working loader on my nano...grr
00:22:02linuxstbSo what's wrong with the Rockbox bootloader?
00:22:04 Join infamis [0] (
00:22:19Paul_The_NerdBHSPitMonkey: Have you tried the Rockbox ipod bootloader?
00:22:37BHSPitMonkeyPaul_The_Nerd: that's not the issue
00:22:52Paul_The_NerdWhat do you mean?
00:22:56Paul_The_NerdYou said you don't have a working loader...
00:23:23infamisWhat skills do you need in order to working on a bootloader? iAudio port
00:24:38Moosinfamis: Hello, better to re-ask when LinusN will be around
00:24:55Mooshe is the one working on it
00:26:12Moosdid you already take a look at iaudio components?
00:26:23infamisok. I just wanna do something; I can do x86 asm (though I haven't even touched it in years) & c++ ; it's hard to decipher the iAudio port status from the twiki
00:26:38infamisyea, been through those pages a lot
00:26:59linuxstbinfamis: You need to be able to find a way to compile and run your own code on your iaudio.
00:27:20Paul_The_NerdBHSPitMonkey: What's the issue then?
00:27:28BHSPitMonkeydon't worry about it...
00:28:17*Paul_The_Nerd is confused.
00:31:02 Join ashridah [0] (
00:31:12t0masinfamis: do you have any hardware / reverse engineering skills?
00:33:19 Join Paul_The_Ner1 [0] (
00:33:47infamisnot really. 5 years ago I started out trying to make my own (protected mode) os in all x86 asm, but then got a job as a php/asp developer and haven't even touched asm; I stll have aLOT to learn if I'm gonna do something; I just really wanna get into hardware programming
00:34:52t0masI've done the x86 OS too
00:35:01t0masbut with only a simple bootloader
00:35:04t0masand a C kernel
00:35:35t0masnever got past the virtual memory and tasks point :)
00:35:55 Quit darkless (Read error: 104 (Connection reset by peer))
00:36:08 Join darkless [0] (
00:36:10 Quit Strath (Read error: 104 (Connection reset by peer))
00:36:11infamisthat's cool; I never got passed step 1 though....building the os
00:36:19 Join Strath [0] (
00:36:26t0masyeah, but in asm it's a hell of a lot harder to do...
00:36:58t0masin C it's not that hard... and there are a lot of docs about it all over the net
00:37:22t0masbut for your iAudio bootloader... you really need some hardware skills to do that
00:37:40t0masat least you need a way to restore from failed attempts
00:38:09infamisyea; asm was my "first love"...haha. I was never really serious with it so I never really did anything good with it. that's why I'm trying to really do some stuff, or at least learn from what you guys are doing
00:39:01t0masghehe, don't count me as a core developer... I only did BMP loading and image support for the WPS
00:39:23t0masdon't have enough time (and skills) to really do the rockbox core stuff
00:39:46Moosand there aren't a lot of devs here touching or touched bootloader things :)
00:40:30infamisI don't even see how you would start messing with a bootloader!
00:40:36t0masit is Linus' think :)
00:40:49Mooshehe, true :)
00:40:59t0masinfamis: you can read the iriver bootloader...
00:41:08Paul_The_Ner1Well, unless you've got the appropriate means of recovering a bricked device, you tend to have to get it right on the first try.
00:41:09Moosfor irivers and iaudio at least
00:41:19Moosand linuxstb and preglow for ipods
00:41:39t0masPaul_The_Nerd: nobody is ever sure of getting these kind of things right in 1 shot
00:41:39infamisyea I've d/l'd the source but haven't really dwelved into yet
00:41:41 Join nave7693 [0] (
00:42:11Paul_The_Ner1t0mas: I know, I'm just pointing out that bootloader coding is riskier than the usual.
00:42:45t0masinfamis: I don't know anything about the iAudio... but have you checked if the original f/w is encrypted in some way?
00:42:47nave7693Can anyone tell me how to use the bootloader image I complied myself from cVS?
00:42:50linuxstbIt depends on the device though. On the ipod there was no risk at all.
00:42:57t0masnave7693: have you checked the md5?
00:43:15linuxstbnave7693: For what player?
00:43:18nave7693t0mas: it is not even the right size. it's 40K and firmware_flash will not accept it
00:43:21ashridahnave7693: for which platform did you compile the bootloader?
00:43:21t0maslinuxstb: I still don't know how that ipod thing works... can you restore the flash w/o software?
00:43:31nave7693linuxstb: I have an archos recorder V1, 2M
00:43:32t0masnave7693: then don't use it.
00:43:51infamist0mas: tell the truth, I wouldn't even know what to look for;
00:44:30linuxstbt0mas: The bootloader isn't in flash - it's just written to the first partition on the disk. Apple have their own bootloader (with an emergency usb disk mode) in flash - and we never touch the flash. So Apple's bootloader loads our bootloader which loads Rockbox.
00:44:43t0masah ok
00:44:55t0masand if it doesn't work, we just put a new loader on the disk
00:45:22nave7693it sort of make sense though, since bootbox has to me small to accomodate rombox, which is huge now
00:45:40nave7693I just don't know how I can flash bootbox + rombox to my recorder.
00:46:03lostlogicyay, AB-repeat is committed (no single button mode active), and I'm almost in sync with CVS!
00:46:16linuxstbnave7693: Have you found the instructions on the wiki?
00:46:41t0masinfamis: you can start with adding links to the firmware files to the wiki
00:46:47Mooslostlogic: congratulations
00:47:11BHSPitMonkeylinuxstb: does the rb bootloader boot linux and apple?
00:47:16nave7693and what the relationship is with those and the firmware_rec.bin image which is 256K
00:47:31linuxstbBHSPitMonkey: Yes.
00:47:37Moost0mas: the "problem" is we don't know at wich state Linus is for the iaudi bootloader
00:47:42linuxstb(if you compile it with gcc4)
00:47:53t0masah ok
00:47:59BHSPitMonkeylinuxstb: in that case, do you think you could tar it for me?
00:48:03*t0mas is still thinking about a port to Windows CE
00:48:11t0masbut it would be a really lot of work
00:48:19t0masand I suck at Windows CE programming
00:48:34BHSPitMonkeyt0mas: well, the sim already runs on windows desktop
00:48:58t0masyeah... it should be possible
00:49:04linuxstbBHSPitMonkey: Maybe preglow or Paul_The_Nerd can give you a compiled copy they have confirmed is working on their Nano.
00:50:00BHSPitMonkeyI rm'd my working image of loader2 (and all the rest) last night by mistake
00:50:03linuxstbnave7693: The Archos experts don't seem to be around at the moment. Have you read this page?
00:50:11BHSPitMonkeyso i'm on thin ice here
00:50:36nave7693linuxstb: you mean flashing bootbox is the same procedure as flashing rombox?
00:50:44 Quit darkless (Client Quit)
00:50:45 Join darkless [0] (
00:51:21linuxstbnave7693: I wouldn't like to give you any advice about that. It's been a year or so since I flashed my recorder. Read the wiki and follow the links.
00:51:38 Quit Paul_The_Nerd (Read error: 110 (Connection timed out))
00:51:45 Quit infamis ("CGI:IRC (EOF)")
00:53:45 Join thegeek [0] (
00:57:10BHSPitMonkeylinuxstb: will rockbox load SFonts?
00:59:20***Saving seen data "./dancer.seen"
00:59:28BHSPitMonkeyalso, what are the bootloader key combos?
01:01:59 Join JdGordon [0] (
01:02:17 Quit jelle-k|offline (Remote closed the connection)
01:02:37linuxstbI don't know about the fonts. MENU gives you the apple firmware, and PLAY gives you linux (you need a linux.bin in the root of your FAT32 partition)
01:03:04linuxstbThe button detection isn't great - you need to press and hold the button just before the apple logo appears.
01:03:34BHSPitMonkeyi tried << and >> but not the other two ;)
01:03:54BHSPitMonkeystarfield = pretty
01:05:07 Join Xerion_ [0] (
01:05:16preglowlinuxstb: btw, i've played around with ipodloader2 button driver
01:05:23preglowlinuxstb: and sometimes it works just great, sometimes it doesn't work at all
01:05:34preglowat least the crappy one we use now is consistent in being poor
01:07:05linuxstbI don't think it's that bad - as long as you press the button early enough.
01:08:21BHSPitMonkeycool, i can confirm that it boots linux and rockbox
01:08:35 Quit Paul_The_Ner1 (Read error: 110 (Connection timed out))
01:09:01BHSPitMonkeyand apple
01:09:22BHSPitMonkeyever think about making the bootloader a menu?
01:09:57*BHSPitMonkey waits for a defense of principle
01:11:25 Join infamis [0] (
01:11:55linuxstbIt's simply because I've got better things to do. I would prefer to spend the time improving Rockbox.
01:12:12 Quit ashridah ("Leaving")
01:12:15linuxstbBut the source is there for anyone to patch.
01:12:16BHSPitMonkeythat's a good reason
01:12:21 Quit NicoFR ()
01:12:25 Join ashridah [0] (
01:12:47BHSPitMonkeyI just assumed it was equally easy to use a menu in place of button detection, since a lot of elements seem to be there anyway
01:13:48 Quit nave7693 (Read error: 110 (Connection timed out))
01:14:30preglowby no means
01:14:49preglowwe'd need the menu code, ok, we'd also need a proper button driver without using interrupts
01:15:01preglowi wouldn't even want a menu there
01:15:16linuxstbWhat I would quite like to do is to add options in Rockbox to reboot into Linux or the retailos. All that's needed is to make Rockbox write a message into IRAM and then reboot. The bootloader could read that message and start the appropriate OS.
01:15:54 Quit Kohlrabi (Nick collision from services.)
01:15:59 Join Kohlrabi [0] (
01:21:37 Quit Thus0 ("Leaving")
01:23:38preglowsince we're already doing that, it's not a big deal
01:24:34BHSPitMonkeyI've already suggested that with loader2
01:24:43BHSPitMonkeybut, uh, loader2 isn't exactly being a good friend
01:24:55BHSPitMonkeyso that's why I'm on rbloader :D
01:26:39linuxstbYou just need to edit line 41 of config.c to read "const char* confname=NULL;" - that worked for me.
01:26:49linuxstb(in ipodloader2)
01:27:14BHSPitMonkeywell later after you said that, I asked josh what it was again, and he told me
01:27:21BHSPitMonkeydidn't wfm
01:27:30BHSPitMonkeyw/e, it's all g
01:27:43BHSPitMonkeythis bootloader seems to read a lot faster too
01:27:56 Join Kyl3 [0] (
01:28:12linuxstbYes, it uses the standard Rockbox FAT32/ATA drivers.
01:28:47preglowclearly a nice advantage of our bootloader
01:28:56preglowwe don't need to make new versions of everything just for the loader
01:28:59BHSPitMonkeyhmm, doesn't seem to recognize the SFont
01:29:13BHSPitMonkeyoh well
01:29:37*BHSPitMonkey adds aquafont.fnt to rockbox
01:29:45preglowhow's that look?
01:30:05BHSPitMonkeythe font, or my fnt conversion?
01:30:22preglowthe font
01:35:07BHSPitMonkeylooks pretty swell on the nano
01:35:57preglowooh, is it a kana only font, or does it have tons of characters?
01:36:06preglowi'd like to try that out
01:37:57 Quit wubbla (Remote closed the connection)
01:38:30 Join wubbla [0] (
01:40:36 Quit Moos ("Glory to Rockbox")
01:40:45 Join tvelocity [0] (
01:43:58infamisquick question: anyone know any good books regarding embedded systems programming...intermediate level?
01:45:17BHSPitMonkeypreglow: it was tons, the FNT I generated is just english though
01:45:28preglowBHSPitMonkey: could you put it up somewhere?
01:45:48BHSPitMonkeysure, what hosting do you recommend
01:45:56BHSPitMonkeyit's been a while since i've had to put anything up
01:46:00preglowyou could also just dcc it if possible
01:47:55BHSPitMonkeypreglow: not sure if my router's set up for this, but we'll try
01:48:01BHSPitMonkeyare you getting the request?
01:49:48 Join Lamed [0] (
01:50:08preglowBHSPitMonkey: yeah, but as you said, your router's not set up for this :)
01:51:27Lamedno way #define WPS_AB_SET_B_MARKER (BUTTON_REC | BUTTON_RIGHT) on gwps.h:73 is correct. this causes problems with the a-b repeat for the iriver's
01:51:56Lamedpreglow: long time no see
01:52:20Lamedyou can't really button_rec and button_right.
01:52:49 Quit linuxstb ("CGI:IRC")
01:53:15BHSPitMonkeypreglow: you have aim/yahoo/msn?
01:53:42preglowi have jabber/google chat
01:54:01muesli__google chat? never heard of it
01:54:05 Join linuxstb [0] (
01:54:08preglowgoogle talk, whatever
01:54:22muesli__same for this ;)
01:54:30preglowBHSPitMonkey: you could mail me it, i guess
01:55:03linuxstbOr upload it to somewhere like
01:55:15BHSPitMonkeyoh yeah, forgot about that little site :P
01:55:37BHSPitMonkeyI have google talk, don't think you can send files
01:55:52preglownah, you can't
01:57:15Lamedanyone else having his plugin.c crashing all over the place??
01:58:26BHSPitMonkeypreglow: pm
01:58:30linuxstbLamed: I don't think so. What's the problem.
01:59:08LamedCC plugin.c
01:59:10preglowBHSPitMonkey: you need to register with freenode, i think
01:59:15preglowBHSPitMonkey: at least i don't get no pm
01:59:16Lamedplugin.c: In function `plugin_load':
01:59:18Lamedplugin.c:440: error: parse error before '||' token
01:59:22Lamedplugin.c:377: warning: unused variable `rc'
01:59:23linuxstbRe-run ../tools/configure
01:59:40Lamedhuh. stupidity
01:59:48Lamedthank you!
01:59:55BHSPitMonkeypreglow: i was asking permission, like a good humanitarian :)
02:00:02preglowsure you can pm
02:00:12preglowuntil you fall over
02:02:06Lamedisn't anyone interested in fixing the iriver's ab bug?
02:02:21BHSPitMonkeypreglow: did you get it this time?
02:02:28t0masLamed? What's the problem?
02:02:36BHSPitMonkeybtw, linuxstb / preglow, about the reboot into diskmode thing
02:02:40Lamedhey t0mas.
02:02:50Lamedyou can't really button_rec and button_right.
02:02:52BHSPitMonkeyipodloader2 does that correctly (well, when you can get to the menu ;) )
02:03:01Lamedthat's happening over... wait
02:03:02BHSPitMonkeyso...maybe that could be a resource
02:03:06preglow_I'M_ not registered :PP
02:03:20BHSPitMonkeypreglow: will you give out your address in here then?
02:03:40preglownah, i am registered here, just needed to ident
02:03:51BHSPitMonkeythere, preglow
02:03:56BHSPitMonkeyit's senteded
02:04:16BHSPitMonkeyheh ;)
02:04:27Lamedthe ab bug is because it's binded to button_rec | button_right over gwps.h. and you can't trigger rec+anything except ON ...
02:04:40t0masah ok
02:04:50t0masbut are there any buttons left on the iriver? :)
02:04:52linuxstbBHSPitMonkey: Rockbox already reboots into diskmode. I was talking about rebooting directly into Linux or the RetailOS.
02:05:20Lamedah... lemme think about it
02:05:50Lamedt0mas: anything happend with my screen scroll patch? was it refused by someone?
02:05:54preglowit does look kind of cool
02:05:56BHSPitMonkeylinuxstb: no, it doesn't, successfully
02:06:02BHSPitMonkeyand i know what you were talking about then
02:06:03 Quit mozetti (Read error: 110 (Connection timed out))
02:06:07t0masLamed: asked Bagder/Daniel to do it...
02:06:08preglowi think this thing would look nice with antialiasing
02:06:15t0masbut he has been very busy...
02:06:38BHSPitMonkeyand it even says this on the status page for the ipod
02:06:40lostlogicwhat's this about ab and bugs and buttons?
02:06:44Lamedvery as in _VERY_... it's been over then a mounth or two I believe...
02:06:59t0masyeah... I'll do it myself...
02:07:16t0mashave you checked if it breaks with anything done in the last 2 months?
02:07:44Lamedlostlogic: the ab bug is because it's binded to button_rec | button_right over gwps.h. and you can't trigger rec+anything except ON ...
02:07:44preglowi need to sleep
02:07:57t0masgood night
02:08:25lostlogicLamed: this being a hardware limitation that I overlooked?
02:08:27linuxstbBHSPitMonkey: That was fixed recently - I've just updated the wiki page. Have you tried it with the current CVS of Rockbox?
02:08:30t0masLamed: I'm updating my local copy now...
02:08:36t0masif that's done, I'll apply your patch
02:08:39BHSPitMonkeylinuxstb: bah.
02:08:40t0masand commit it
02:08:53Lamedt0mas: corrently it will just 'patch -p0' with some .rej ected. I will submit in a few minuts a newer for today's source.
02:09:13t0masyou have my e-mail address?
02:09:16BHSPitMonkeyyou sure have to keep up-to-date with this os
02:09:30lostlogicab repeat works fine with the current bindings here...?
02:09:39lostlogicis that a problem with the H1x0/
02:09:51Lamedt0mas: (i'm writing pretty slow) I was doing the same thing. I was just re building after re running configure, afterwards I can diff
02:10:12t0mascan you also do a quick build?
02:10:28t0masI'm at my laptop:
02:10:39t0mas[root@laptop2 build]# m68k-elf-gcc
02:10:40t0masbash: m68k-elf-gcc: command not found
02:11:10Lamedlostlogic: what player
02:11:37t0maslostlogic: he is right... it works on ipod etc. but it doesn't on Iriver H1x0
02:11:41Lamedlostlogic: i was just reffering to the daily build page that reports some errors. I'm still running make zip over here
02:11:45t0masyou can't use rec + left and rec + right
02:13:40Lamedg#$%dam cygwin. what the hell "BASH: fork: Resource temporarily unavailable" supposed to mean?
02:13:54t0masto much running processes
02:13:59t0masor out of memory
02:14:11Lamedon cygwin or under the os?
02:14:18t0masI think cygwin
02:14:47Lamedi was just opening it... probably ram
02:15:05t0masit's 2:15 here
02:15:12t0mascan we add 1 day to the 2 months?
02:15:15t0masand do it tomorrow?
02:16:22Lamedfirefox:32 mb of ram
02:16:22t0masI take that as a no ;)
02:16:32t0masah, yeah... it does that some times
02:16:41BHSPitMonkeylinuxstb: so are there any cool plugins that aren't included in the cvs?
02:16:45BHSPitMonkey(that work on the ipod)
02:16:51Lamedi'm on CHatzilla over here so i'll bback shortly
02:16:54t0masLamed: mine is at 69.924 KB now
02:16:54 Quit Lamed ("Chatzilla [Firefox 1.5/2005111116]")
02:17:55linuxstbBHSPitMonkey: No. There are some being developed for the iriver H300, but they will need adapting for the Nano's screen. They are here:
02:18:09t0mas69.704 KB now
02:19:44 Quit Xerion_ ("l8r...")
02:22:06BHSPitMonkeythose are cool
02:22:14BHSPitMonkeywouldn't RockNES be a good port ;)
02:23:43muesli__rockPSP a much better :D
02:24:01 Join lamed [0] (
02:24:09BHSPitMonkeyi meant NES for rockbox, not the other way around :P
02:24:43lamedt0mas: it's 03:25 over here... where do you what me to put it? sourceforge?
02:24:54linuxstbBHSPitLappy: You probably know about rockboy?
02:25:05BHSPitMonkeyi've heard of it
02:25:26t0masor copy it to your webspace or something
02:25:37linuxstbBHSPitLappy: It should be working on the Nano soon. But the ipod just doesn't have enough buttons...
02:26:38BHSPitMonkeydoes rb register staps?
02:27:02lamedI'm going through it to see it's good
02:28:47lamedit's sent
02:29:24t0mas(see pm)
02:33:31t0maswake up ;)
02:33:52 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
02:35:42*t0mas wonders where lamed has gone...
02:35:50 Quit einhirn (Read error: 104 (Connection reset by peer))
02:36:05t0masare you registered with freenode?
02:36:13t0masyou can't send pm-s if you're not
02:36:26 Join einhirn [0] (
02:36:28lamedhaven't you got my messages? I _am_ registered
02:36:36muesli__join #temp
02:36:39lamedmaybe i'm not logged in. but i tinkg i am
02:36:46lamed<lamed> it's the same patch, but not the same version. If you close the sourceforge patch, is it totally deleted? if so, delete it
02:36:53lameded> your third idea sound the most appropriate
02:37:00lamed<lamed> p.s shachar is my personal name.
02:37:03t0masok, then you're not logged in
02:37:36t0masand do you have any idea's why I haven't received your email yet?
02:37:41linuxstbBHSPitMonkey: No, we don't have "tap events" at the moment (apart from turning on the backlight), but we'll probably add them.
02:37:50lamed<lamed> identify ???????
02:37:54lamed-NickServ- You have already identified
02:38:00t0mas[02:37:45] -NickServ- You have already identified
02:38:28BHSPitMonkeylinuxstb: they don't seem to work for the nano in iPL, possibly due to the different wheel manufacturer (?)
02:38:38BHSPitMonkeybut diagmode recognizes them
02:38:46t0masthat's correct
02:39:01lamedsourceforge: 1 minute
02:39:05 Join webguest48 [0] (
02:39:21 Part infamis
02:39:54t0maslamed: I have it
02:39:54 Quit webguest48 (Client Quit)
02:39:59t0masjust took some time :|
02:40:55lamedand it's now on sourceforge as well :)
02:40:55t0masyou are in the credits file right?
02:40:57lamedt0mas do you get me?
02:41:12lamedt0mas: yes, for the hebrew translation :)
02:41:18t0mas(not receiving any privmsg's)
02:42:13 Join webguest48 [0] (
02:42:37lamedwhat? translating for a language that no one will probably ever use? -yes, I know,
02:42:44webguest48anyone know how to use the ab repeat on the h120?
02:43:06t0maswebguest48: it's not possible atm
02:43:11t0maskey bindings are wrong
02:43:19lamedam I the only one that's getting codec failiure for everything?
02:43:57lamedwebguest48: do you program?
02:44:36BHSPitMonkeylamed: nah, programmers have real IRC clients! :D
02:44:38 Quit lamed ("CGI:IRC")
02:44:41t0maswait 25 minutes, and you can download rockbox with your patch in it :)
02:44:58 Join lamed [0] (
02:45:06BHSPitMonkeyhaha lamed
02:45:22lamedBHSPitMonkey lol...
02:45:38lamedt0mas: nice. what was changed?
02:47:07lamedno warrning for the iriver. what else did you wanted me to see?
02:47:23t0masthe compile time
02:47:30lamedhoo.. I get it... it's still compiling... cool
02:47:32t0masit estimates to have a new build ready at 2:11 GMT
02:47:51t0masif you click on the date/time
02:47:59lamedhalf an houre...
02:48:16t0mas34 builds in 1743 seconds makes 51 seconds/build
02:48:24t0masit calculates that too ;)
02:48:29lamedCoooolll.....!!?!@#?>!@EW?W>A>?F yeyyyy?>FDF?G>#
02:48:43webguest48lamed: i dont program
02:48:47 Quit mikearthur ("Konversation terminated!")
02:49:02webguest48the bleeding edge build supposedly has ab repeat...
02:49:13webguest48so i thought i just needed to know the mapped keys
02:49:15t0maswebguest48: it has, but it isn't working for iriver
02:49:37t0masthe key definitions we have now are not supported on the iriver
02:49:41lamedt0mas: i've just remebered that it _might be_ that i've changed some order in one of the files so I won't always get Cvs errors... I'll check on that now
02:49:42webguest48i see
02:50:02t0maslamed: what order?
02:50:07webguest48and what's a seamless seeking?
02:50:22lamedt0mas... lemme just look for a sec
02:50:30t0maswebguest48: seeking to another position in the song, w/o stopping the sound
02:50:40t0masso that's basicly what ab-repeat does...
02:50:49webguest48ah nice
02:50:53t0masif it gets to point B, it seamlessly seeks back to A
02:51:31webguest48i see, thx
02:51:44 Quit webguest48 ("CGI:IRC")
02:51:46lamedt0mas- getting pm?
02:51:58t0masjoin #rockbox-temp
02:52:10t0masand use that as a PM replacement
02:55:46lamedt0mas: no luck.
02:55:46lamed*** Access to channel #rockbox-temp is blocked
02:56:04lamedrob, I get your messages
02:56:19t0masrob is a staffer
02:56:24t0masand he's not on the channel ;)
02:56:43linuxstbt0mas: You've broken all colour builds - the patch didn't add the new functions to lcd-16bit.c
02:57:05t0maslamed: someone from freenode
02:57:20lamedlemme check about that linuxstb, it should work!
02:57:55linuxstbYes, it should just be a copy and paste.
02:57:56lamedt0mas: I haven't replaced the order of anything.
02:58:11lamedlinuxstb: how do you know it's not working? it's still compiling?
02:58:25linuxstbI compiled it myself for the ipod.
02:58:26t0maslamed: local compile
02:59:23***Saving seen data "./dancer.seen"
02:59:27lamedyeah. he is right. it is somehow missing. sec.
02:59:53t0mastry to fix it...
02:59:58t0masand do a test build for the iriver H320
03:00:00t0masor the ipod
03:00:06t0masor some other color target
03:00:15linuxstbOr both.
03:00:21t0maseven better
03:00:24lamedi guess i had the file overwritten and so it's not in the diff.
03:00:29t0maslinuxstb? you have CVS access?
03:01:03t0mascan you test + commit?
03:01:13t0masI can't test it myself now..
03:01:19linuxstbCan't promise I'll be awake for long, but if I am, then I will.
03:01:31lamedlinuxstb: i've got it here.
03:01:32t0masif you're not... I'll do it tomorrow morning
03:01:39t0maslamed: sure it works?
03:01:42t0masthen just email it
03:01:50t0masI'll be around for 3 more minutes
03:02:33lameddoing it
03:03:22lamedit patches smooth, building & mailing now.
03:03:47lostlogichorizontal scrolling is for screen switches?
03:04:56lamedlostlogic: use ON+ RIGHT/LEFT on a filetree/playlist. you'll understand
03:05:38lamedt0mas: am building an h300 sim right now, Will work, but let me see
03:06:03lamedan old p3 850 :((((
03:06:16t0masghe :)
03:06:26t0masmy laptop is only 1.2 ghz too
03:06:51t0masand cygwin is way slower than native linux
03:07:20lamedyou're right. try race me
03:17:48 Quit muesli__ (Read error: 110 (Connection timed out))
03:19:12linuxstbt0mas, lamed: That's fine on the ipod now.
03:22:06t0mascheck that out in your part 3 too
03:22:18t0masyou should prob. just add some defines
03:22:52lamedt0mas: yes, i figured that this will happend. I cahnged those names at the newest version.
03:23:27t0masyou still have to define em away for the player I guess
03:23:36t0masas they aren't used in that build
03:23:39lamedt0mas, I have to have the older cvs version so i could cvs diff the source...
03:23:57linuxstblamed: Just provide a diff against current CVS
03:24:01t0masyou must diff it to the current one
03:24:13t0masbut I'm away now
03:24:21t0masI really need my sleep
03:24:33 Quit t0mas ("bye")
03:25:54lamedlinuxstb: that sounds a bit risky
03:26:10lamedbut i'm on it for now
03:27:13linuxstbWhy risky? Patches can only be applied against what is in CVS now.
03:27:46lamedlemme explain
03:29:00lamedlinuxstb: the first file i sent to t0mas was an older copy of the screen_scroll patch I had. the second file probably fixes most errors,
03:29:07lamedbut I had a new patch without the style-changes that were in the first (older) patch, and changed some integer names and stuff
03:30:20linuxstbThen you can check out a clean copy from CVS, reverse the first two patches yourself, and then apply the third patch, and then generate a diff against CVS.
03:31:03lamedlinuxstb: correct :) that's what I'm doing. but I don't like it because I prefer of manually checking my patchfiles
03:33:13linuxstbI'm sure you'll work something out :)
03:33:13 Join Jungti1234 [0] (n=jungti12@
03:33:39lamedI will... I feel so bad for sending him the wrong file :(((
03:37:55 Quit Kyl3 ("( :: NoNameScript 3.81 :: )")
03:50:35lamedlinuxstb: patch is ready. I'm compiling for h120 and simulators of right about everything. i have an oldish p3 850. care to check yourself?
03:50:40 Quit birdfish ()
03:54:06 Join birdfish [0] (
03:55:10 Quit tvelocity ("Leaving")
03:56:34 Join Paul_The_Nerd [0] (
04:04:53lamedanyone awake that can submit?
04:05:00 Quit einhirn (Read error: 104 (Connection reset by peer))
04:07:41Jungti1234I have p3 800
04:08:17lamed& u don't have cvs rights :)
04:09:43 Join RedLeg [0] (n=red@
04:12:17RedLegAnyone here?
04:16:05RedLegI'm getting an "ATA error: -11" message at boot all of a sudden.... any clues
04:17:16RedLegI should add: Jukebox recorder 20, has worked fine for YEARS, no changes...
04:17:41RedLegAnd, the drive works fine out of the Archos
04:20:07*BHSPitLappy is an iPod owner
04:21:30RedLegI may become one soon.... I'm not sure I want to invest in hardware this old....
04:21:55lamedredleg: what firmware version?
04:23:27RedLeglamed: I THINK it's 2.4.. I haven't upgraded in a while, and it's definitely not 2.5
04:24:00RedLegI get the error whether there's a drive in the unit or not
04:25:42RedLegIt won't boot into the Archos firmware either, message is: HD register error.....
04:26:03lamedcan you sync it with the computer?
04:26:30RedLegLemme try....
04:27:08Jungti1234hmm What is Horizontal scrolling?
04:27:50RedLeglamed: no, same error
04:27:51lamedI gtg. I don't know much but it sounds like an hard disk failing... I will try scandisk and defrag if you can sync the drive. there's a mith of freezing an h.d helps (google it) but i don't know if that will work for a player... jungti use ON+ LEFT/RIGHT on a playlist/fileview
04:27:58 Quit lamed ("CGI:IRC")
04:28:38 Quit kkurbjun (Read error: 110 (Connection timed out))
04:29:24 Quit webguest66 ("CGI:IRC (EOF)")
04:29:57RedLegWell, for what it's worth, the drive works fine out of the archos. The Archos doesn't even't seem to be trying to spin the drive up...
04:32:11Paul_The_NerdSo, maybe it's the ata controller itself
04:32:40Paul_The_NerdOr whatever
04:33:56RedLegYeah, that's what I'm afraid of... spontaneous hardware death
04:34:59Paul_The_NerdI live in fear of that every day. :-P
04:35:05RedLegStill, I'm curious... the rockbox error message is "ATA: -11"
04:35:07 Join lamed [0] (
04:36:05lamedjungti what do you think about the screen scoll?
04:36:28lostlogiclamed: is there still another part of that patch that you need to get committed?
04:38:22lamedlostlogic: I have a patch file that cleans up the code styling changed that I accidently had in the first file, solves the player problem, and changes some integer names for logic reasons
04:39:01lostlogicsince thomas went to bed, want to send it to me to clean up the build table?
04:40:07lamedI'm building for h120, simulators: h300 , player, recorder,gmini,ipod, x5. I only reached the h300. and old p3 850. I will be most greatful to you if you're willing to do it
04:41:20lostlogichehe, I can test build for most targets, and I have a faster box if you can either link or email it to me.
04:41:45lamedlostlogic, is the x5 already running? I guess not, right? - there are no keybinds for it because I wasn't sure if even has a simulator up and running
04:41:56lamedwhat's ur email?
04:42:04lostlogiclostlogic at lostlogicx dot com
04:42:26lostlogicx5 isn't running yet, but it'd be nice to support it with new things (not that some of my patches have supported it either)
04:42:28RedLegsolexx: what does the "ATA: -11" error message mean?
04:42:31Paul_The_Nerdlamed: Probably not worth bothering with gmini.
04:42:50lostlogiclamed: you have the calmrisc16 target building!?
04:42:58RedLeglamed: what does the "ATA: -11" error message mean?
04:43:22lamedredleg: u really should refer to lostlogic/paul on that, beforehand asking me :D
04:43:40RedLeglamed: OK, thanx....
04:43:40lamedit's an h.d issue
04:43:59lamedlostlogic: calmrisc16?
04:44:24lamedlostlogic: mailbox?
04:44:27lostlogiclamed: that's the compiler target for gmini, right... and I haven't gotten a toolchain working for it yet
04:44:34RedLegi know it's an HD issue, but the HD works fine out of the unit.....
04:45:24Paul_The_NerdRedleg: It means that it can't access the HD. For example, if the batteries were too low to spin it up.
04:45:36lamedlostlogic: I lost you. anyways. what's ur mailbox so I could mail you the latest diff?
04:46:02lostlogiclamed: see my message @ 21:42 with my email spelled out ;)
04:47:29RedLegPaul_The_Nerd: Yep.... this happened to me a couple of ears ago.... a drive got to the point that the archos couldn't spin it up. This is the replacement for that drive, CAREFULLY specc'd to be within the archos capabilities. WOrst thing: it worked fine last night, dead this morning.
04:47:42lostlogicRedLeg: it is an error having to do with MASTER_SLAVE detection
04:48:12lostlogicmost likely returned from ata.c line 1210
04:48:14RedLeglostlogic: so maybe a jumper would help?
04:48:45lostlogicRedLeg: not sure, I am not intimately familiar with what this code does, I just tracked down that error code to the line it was thrown from.
04:49:20lamedlostlogic: sent
04:49:48RedLeglostlogic: understood, thanks.... I'll go see if I have a jumper that will fit this drive and try. I have the unit apart anyway....
04:49:54Paul_The_Nerdlostlogic: I took a completely different approach, and searched the mailing list archives. Hehehe.
04:50:51lostlogicPaul_The_Nerd: hehe, I'm all about reading code, gives me excuse to learn something new
04:51:46Paul_The_Nerdlostlogic: I have ata.c open in an editor, but I searched the site first in case the question was already answered.
04:52:15lamedlostlogic: x5 should work with any problem what so ever. all we have to do is make a keybind. I'm not familiar wiith their keys so i gave up on it for now.
04:52:20lameddid ya got my file?
04:52:30RedLegI'm looking at the drive, and no jumpers is supposed to be MASTER.....
04:52:57RedLegAnother data point: I get the same error with NO drive in the chassis
04:53:31Paul_The_NerdRedLeg: How is your battery charge?
04:54:09lamedRedleg: I acctually got this message with my h120 a little while ago. it worked the second time i tryied, but my h.d is getting pretty curropted lately
04:54:20RedLegPaul_The_Nerd: Yah... I've tried several sets, and charged them specifically
04:54:43lostlogiclamed: is AB repeat still a problem on H1x0, so I need a different keybinding for it?
04:55:12 Join kkurbjun [0] (
04:55:25Paul_The_NerdRedLeg: Hrm. I dunno, but it sounds like some other hardware's the problem then, since you say the HD works outside the unit.
04:55:34lostlogiclamed: and if so, do you have any suggested keybindings that are free and functional on WPS
04:57:01RedLegOK, tried a different drive.... at least got the version: Rockbox 2.3
04:58:00lamedlostlogic: don't pm ok? - I don't think anyone commited anything about the ab keybinds, but I'll check
04:58:05Paul_The_NerdWait, so a different drive boots?
04:59:27***Saving seen data "./dancer.seen"
04:59:30RedLegNot quite.... but I got the RB boot splash, not an error message.... trying ANOTHER now
04:59:35lostlogiclamed: ok −− do you know of any free keys in the WPS on H1x0?
05:00:18lamedahhhh... that was the last pm
05:00:22Paul_The_NerdIsn't the record one free?
05:00:37lamedpaul... you need two keys for a-b repleat
05:00:52Paul_The_NerdYou do?
05:00:58Paul_The_NerdWhy exactly?
05:01:12 Join Rob2222 [0] (
05:01:25 Join j [0] (
05:01:42Paul_The_NerdWhat is the second key needed for?
05:01:44lamedI'm not sure. i haven't used a-b repeat... it was only implemented today remember.... I went trhough the code and found out there are two bindings.
05:01:54Paul_The_NerdWhat does each one do?
05:02:01Paul_The_NerdBecause you could use long and short presses of Record
05:02:03lamedprobably first que and second que.
05:02:19Paul_The_NerdBut why does that need two different keys?
05:02:28Paul_The_NerdI think one would be enter/exit A-B mode, and the other would be setting the points.
05:02:38Paul_The_NerdAgain, long for the first, short for the latter.
05:02:57Jungti1234What is horizontal scrolling~?
05:03:26lamedJungti. I've already replied you about that before
05:03:43Jungti1234I don't understand.
05:04:10jhi, i plugged in the wrong voltage charger to my jukebox studio 10, and now it won't charge internally. does anyone have this problem or know how to fix it?
05:04:10lamedjungti: what player?
05:04:23lamedj is it working?
05:05:00jit works fine, but does not charge the batteries. it says charging but it is only running from batteries
05:05:07RedLegOK, with the 3d HD, I get the same ATA: -11 error, and the same HD Register Error out of the Archos firmware... looks like I may have an Archos hardware issue....
05:05:20Jungti1234lamed: H300
05:05:40lamedis it updated with the bleeding build?
05:05:48 Join Sacro [0] (n=root@
05:05:54Jungti1234not yety
05:06:01 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
05:06:14Sacrodoes the iriver H300 support midi?
05:06:36lamedupdate, and then use ON+RIGHT/LEFT on the fileviewer, and you'll understand
05:06:45Jungti1234ok :)
05:07:40lamedjungti: i will update the wikimanual soon :)
05:08:05lamedsacro: i think you can use midi2wav for that
05:08:37Sacrolamed: yeah, i can use timidity++ to convert, but i selected one on the mp3 player, it got as far as "selecting patches" and died
05:09:27Jungti1234you know?
05:10:17lamedlostlogic: are you on it?
05:10:22lostlogiclamed: am on it
05:10:47Sacrohmm, it plays mpc, and i converted them to mp3 eventually
05:11:59 Quit RedLeg ("Leaving")
05:12:37lostlogic12 targets to build on the wall, 12 targets to build... take one down, run it around, 11 targets to build on the wall
05:12:49lostlogiclamed: can you build the simulator targets for me? I'm lazy. :-P
05:12:50lamedgreat, then i'm on the a-b patch
05:13:07lamedh300 sim is working
05:13:07lostlogicor we can just assume the sims are OK :-P
05:13:30lamedand now i'm all out of memory again :((( i guess i'm seriously infected with something.
05:13:38lostlogicyou run windows?
05:13:55lamedghe,,, you don't have to build everything!!!
05:14:20lostlogichaha, I need to setup a build script like the server has locally... my box is fast enough to do all of the builds in reasonable time
05:14:27 Join BHSPitLappy [0] (
05:14:58lamedsome color builds (h300, ipod), the recorder, player, maybe an ondio, that's it! (should make sure that there are no warnings for the player, should be ok)
05:15:16lostlogicI'm building with -Werror, so no worries on that.
05:15:28lamedlogic: what's your cvs name?
05:15:38lostlogicusername is lostlogic, _name_ is Brandon
05:15:54lostlogicI'm teh newb
05:16:09 Quit Rob2222_ (Read error: 104 (Connection reset by peer))
05:16:10*lamed newere
05:16:12Sacroh300 sim?
05:16:44lostlogiclamed: meant I'm the newest committer.
05:18:44lamedyeah I knew what you ment
05:19:13lamedand yeah winxp. I've installed ubuntu linux but i can't get my pptp connection working :(
05:19:25lostlogicok, building iPod and then committing your patch, looks good to me.
05:19:37lamedshould be good.
05:19:46lamed-> THIS TIME ...
05:19:55lamedwhat do you think of the usability though?
05:20:50lostlogicit's neat −− needs to turn off the autoscrolling of long titles, IMO
05:21:25lamedwhat autoscrolling are you talking about?
05:21:31lostlogicoh crap, something I did today has totally messed software codecs up.
05:22:16lamedwhy not using a catch click on the record button, so the first click=set A point, and the second click =set B point?
05:22:51lostlogiclamed: that's actually how I'd originally designed it for H[13]x0, but other developers thought that giving a whole button to this feature was not OK
05:23:27lostlogiclamed: there's a WPS_SINGLE_BUTTON_PRE and WPS_SINGLE_BUTTON that you can define for a given target, and then you can define AB_REPEAT_ENABLE to 2 instead of 1 to take it out of the menu and have it more like the original iRiver feature
05:25:12lamedwhooo man you reallly f#$Fcked up the codecs....
05:25:17lostlogiclamed: your patch (lightly modified) is committed, you'll probably have conflicts when you update
05:25:28lostlogiclamed: you trigger the same weird bug I did where they skip all around weird?
05:25:32lostlogiclamed: it doesn't happen all the time
05:25:38lamedyey! i love conflicts
05:25:51lamedyes... it just happend the first
05:26:02lostlogiclamed: gah, I'll have to hunt that down now before I can sleep
05:26:17Jungti1234hmm lamed
05:26:39Jungti1234I pressed ON+Right/Left, but there is no change.
05:26:54lamedlisten, i agree a-b isn't important enough to be used as a button. but surely isn't a good idea for two buttons! what about having ON+SELECT as the only button for it? (similar to iriver, different button)
05:27:45lamedlet the bleeding edge finish building
05:27:53lamedand then download
05:28:24lostlogiclamed: that seems reasonable if it's free, just define WPS_AB_SINGLE and the AB_REPEAT_ENABLE to 2...
05:28:25Jungti1234Does it become scroll from left to right?
05:28:45lostlogicprobably also define WPS_AB_A and WPS_AB_B as well so that people can move markers easily
05:29:01lamedit _is_ working now btw. jungti, find a reallllly long file names folder and then mess around on+right/left
05:30:13Jungti1234I witnessed it.
05:30:19lamedlostlogic: you've got a point. acctually, iriver sucked so bad about a-b mode, that I haven't thought what might be better with a button for every marker
05:30:31Jungti1234Is it 'Screen Scrolls Out Of View'?
05:31:08lamedI say destroy the WPS_NEXT_DIR button... hate it.
05:31:28Jungti1234OK! I understood it.
05:31:35Jungti1234thanks lamed!
05:34:04lamedsure... it was nothing... just learning how to program in C after being away from programing for several years... nothing really
05:35:17lamedjungti1234: are you an english native speaker?
05:35:40Jungti1234no i'm korean
05:35:59lostlogiclamed: any idea how to trigger the playback bug exactly?
05:36:27 Quit Sacro (Remote closed the connection)
05:37:02lamedlostlogic: took me several weeks for learning the main thread... :(
05:37:33lostlogiclamed: I meant in using the player do you know what you do to make it do that weird skippy thing?
05:38:12lamedlostlogic: I think ON+REC / ON+MODE is a pretty good substitude for REC+LEFT/RIGHT. each one of those buttons are on the other side of the player.
05:38:41lostlogiclamed: I don't have a H1x0 so you know better than me
05:38:49lamedlostlogic: i only know it happends when you listen to beatles - all you need is love
05:39:17lostlogiclamed: :-P
05:40:29lostlogicit has to do with skip functionality, I think
05:41:57lamedlostlogic: but it's really not comfortable. this bind that is. we only have REC, ON + REC, ON+SELECT, ON+MODE available. maybe, a nice idea would have to move the WPS_NEXT_DIR to long clicks, and have a short ON+left/right do ab repeat. you probably should consult everyone first :)
05:42:29 Quit j (Remote closed the connection)
05:43:09lostlogiclamed: for now, let's just give it single button functionality only, and we can give it more after more thinking.
05:45:00lamedsounds decent, but it's up to you... it's 6:42 AM over here!!! I'm supposed to wake up in two hours! and I can't even compile because of low memory all the time. -thanks for your help, good day
05:45:15lostlogiclamed: hehe, g'night
05:46:24lamedwooo, so nice to see my name over rockbox's main page :D I will save myself a copy of the page
05:47:01lostlogicand I don't think it was me who broke things
05:47:39lostlogicin _strange_ news, I think somehow the fix for the popping noise caused the weird skippiness
05:48:07lamedbtw needs some work. goodluck with that problem :)
05:50:27 Quit lamed ("CGI:IRC (EOF)")
05:52:59 Join midkay [0] (n=midkay@
05:53:58 Quit midkay (Client Quit)
05:54:22 Join midkay [0] (n=midkay@
06:11:21 Quit ashridah (Read error: 110 (Connection timed out))
06:11:26 Join ashridah [0] (
06:14:46lostlogicSlasheri: you don't by any chance happen to be still around?
06:20:39 Join kernel_sensei [0] (
06:20:40 Quit kernelsensei (Read error: 104 (Connection reset by peer))
06:20:50 Nick kernel_sensei is now known as kernelsensei (
06:25:53 Quit ashridah ("Leaving")
06:26:14 Join ashridah [0] (
06:37:43 Quit stamppot ("CGI:IRC (EOF)")
06:38:23 Quit RotAtoR ("zzzzzz")
06:45:27 Quit Jungti1234 ()
06:59:28***Saving seen data "./dancer.seen"
07:26:19 Part Paul_The_Nerd
07:39:44JdGordonhey, does any1 use rockcalander?
07:56:05JdGordonany1 else tryed compiling the latest sources and have them fail on plugins.c?
08:00:13BHSPitLappyany1 here know how to spell?
08:00:23JdGordonnot i
08:02:06midkayoh! me! PICK ME!!!
08:07:19JdGordon...wierd... reconfing cygwin fixed the problem!
08:12:52 Join Jungti1234 [0] (n=jungti12@
08:17:59 Quit DJDD__ ("Trillian (")
08:20:51 Join DJDD__ [0] (
08:24:20 Quit midkay (Read error: 104 (Connection reset by peer))
08:24:42 Join midkay [0] (n=midkay@
08:39:18 Join EvilDude [0] (i=EvilDude@
08:43:10 Quit Jungti1234 ()
08:56:17 Quit DreamTactix291 (Read error: 110 (Connection timed out))
08:59:29***Saving seen data "./dancer.seen"
09:13:14 Join mikearthur [0] (
09:17:37Slasherilostlogic: hehe, now i am.. it's still only 1017 am here :)
09:30:21 Join SnokeekonS [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
09:30:51 Nick SnokeekonS is now known as AliasCoffee (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
09:39:05 Join Kohlrabi [0] (
09:41:17Slasheriah, that horizontal scrolling is great, especially with the "screen scrolls out of view" setting turned on
09:42:29ashridaho.O didn't things already scroll horizontally?
09:43:11Slasherinot the whole screen at same time :)
09:43:31ashridahi see
09:43:42Slasherilostlogic: Hmm, i really think that screen scrolls out of view should be enabled as default.. the scrolling is quite ugly without it enable
09:45:28amiconnThere is another bug. It is enabled for Ondio, but this button mapping can't working in most lists. While BUTTON_MENU can technically be used as "shift" (it is in a couple of plugins), in most lists a long press of MENU calls the menu
09:45:44amiconnThis simply can't work and should be deactivated
09:52:56Slasheriamiconn: hmm, do i remember correctly that i shouldn't put any system calls to pcm_play_pause in pcm_playback.c?
09:53:20Slasherii cannot see any reason directly for that at the moment
09:53:42Slasheribut i think it was an issue long ago
09:55:12Slasheriif it's safe to mute audio there, we would get rid of one little pop
09:55:26amiconnI don't know...
09:55:35 Join SteL [0] (
09:55:57Slasherihmm, maybe i will try that..
10:02:03 Quit mikearthur (Remote closed the connection)
10:03:05amiconnSlasheri: Hmm, I checked the cvs revisions for pcm_playback.c. Maybe it is rev. 1.67 (by Linus) what you remembered
10:03:27Slasheriah, i will check that revision
10:03:49amiconnYou must not call any uda_ functions (which do i2c) from functions which may be called in interrupt context
10:04:27Slasheriyep, true. But that pcm_play_pause() shouldn't get called within an interrupt
10:05:33amiconnUsually not, but double-check the metronome plugin.
10:07:53amiconnAny news on the tagcache front?
10:08:22Slasherii haven't had much time for that in the past few weeks..
10:08:38Slasherimaybe at next week if it's less busy
10:10:31 Join needleboy [0] (
10:10:50 Join Matze [0] (
10:11:16Slasheriargh, it seems metronome plugin calls pcm_play_data within a timer interrupt.. and i have already added sleep and mute trick there :/
10:12:06 Quit needleboy (Client Quit)
10:15:12Slasherimaybe i will just modify the metronome plugin little
10:15:34amiconnImho the low level functions shouldn't do too many trciks
10:16:52Slasheriamiconn: Hmm, in fact.. i think i can move that single trick to pcmbuf.c
10:17:02Slasherithen it would be a problem anymore
10:17:05 Join amiconn_ [0] (
10:31:02Slasheriamiconn: in fact there seems to be no need to put any mute calls on the pcm level.. all of them can be done on higher level :)
10:31:11Slasheriso i am going to remove them
10:31:40 Quit EvilDude ()
10:35:23 Quit amiconn (Read error: 110 (Connection timed out))
10:35:24 Nick amiconn_ is now known as amiconn (
10:35:52 Join ender` [0] (i=ychat@
10:36:07 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
10:36:26 Join DreamTactix291 [0] (
10:44:19 Join Pi___ [0] (
10:56:46Rob2222good morning
10:57:06midkaywhy hello there mr. Rob. a good morning to you as well.
10:57:39 Quit DreamTactix291 (Read error: 110 (Connection timed out))
10:59:23 Join Jungti1234 [0] (n=jungti12@
10:59:33***Saving seen data "./dancer.seen"
11:02:54*BHSPitLappy yawns.
11:20:45 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
11:22:38BHSPitLappyyou don't want to be here
11:22:42BHSPitLappyit's damn boring.
11:24:17t0masnobody around?
11:24:29BHSPitLappynobody with fingers apparently
11:24:33midkayi'm.. kind of here.
11:24:40t0masI'm going to leave you too ;)
11:24:43 Part amiconn
11:24:44t0mastime for breakfast
11:25:44Jungti1234what is 'Descramble'?
11:26:32t0mas(if you're talking about a firmware file)
11:26:52Pi___what is the minimum things to #include for a new .c file?
11:27:08t0maswhat file?
11:27:16t0masor non rockbox?
11:27:22Pi___no, menu thing
11:27:35Pi___in rockbox
11:27:36t0mashm... depends on what functions etc. you need
11:27:58Jungti1234how do I make voice file?
11:28:14BHSPitLappyPi___, what are the digits in you?
11:28:24Pi___ATM, i get the error:
11:28:24Pi___undefined refference to 'color_menu'
11:28:49t0masah ok
11:30:28t0mashm... I don't know where those are...
11:33:12 Join mozetti [0] (
11:33:37Pi___my code -
11:33:53t0maserror on line?
11:34:00 Quit ender` (Read error: 113 (No route to host))
11:34:30Pi___error is in settings_menu when it calls color_menu
11:35:09t0maswhere is the settings_menu function?
11:36:03Jungti1234what is 'Scramble'?
11:37:07Pi___??, Settings_menu, line 1510 is
11:37:08Pi___ result = color_menu(&color);
11:38:04t0masJungti1234: yes
11:38:09Jungti1234thanks.. :)
11:38:10t0masah ok
11:39:11t0masPi___: have you added that function?
11:39:19t0masas I don't really get what you're doing...
11:40:33Pi___i have a .h file withih declares:
11:40:34Pi___#define MAX_COLORS 64
11:40:39Pi___int color_menu (unsigned *ref);
11:40:51Pi___struct color
11:40:51DBUGEnqueued KICK Pi___
11:40:51Pi___ unsigned rgbref;
11:40:51Pi___ char name[28];
11:40:51***Alert Mode level 1
11:42:10t0maswell... then you have to include that .h in the file with the error...
11:42:42Pi___i want to create a menu that allows the user to pick a color, this will get more complicated when costom collors are added, so i think it should have its own file and not clutter up the settings_menu.c
11:42:53t0masah ok
11:43:21Pi___#include color_menu.h is allready there
11:43:32t0masand the error you get...
11:43:39t0masgive that including filename and line numbers etc...
11:46:40t0masbbl... time to eat
11:49:04Pi___ /rockbox/apps/settings_menu.hc:1510: undefined refference to 'color_menu'
11:49:10Pi___ /rockbox/apps/settings_menu.hc:1510: undefined refference to 'color_menu'
11:50:52***Alert Mode OFF
11:53:16 Quit mirak (Success)
11:59:58 Quit ashridah (Remote closed the connection)
12:00:01 Join Lost-ash [0] (
12:00:31 Nick Lost-ash is now known as ashridah (
12:00:50 Join Moos [0] (
12:03:39 Join SteL_ [0] (
12:04:42 Quit SteL_ (Client Quit)
12:06:35 Nick Pi___ is now known as Pi (
12:06:35DBUGEnqueued KICK Pi
12:07:10 Part SteL
12:08:27 Join SteL [0] (
12:10:01 Join Skydemon [0] (
12:22:02 Quit Skydemon ("bye bye")
12:25:48 Join DangerousDan [0] (
12:27:29 Join ender` [0] (i=ychat@
12:32:19 Quit perplexity (Read error: 110 (Connection timed out))
12:37:08 Join DreamTactix291 [0] (
12:38:50 Join NicoFR [0] (
12:39:55 Part SteL
12:41:27 Join [1]SteL [0] (
12:42:41 Join Hansmaulwurf [0] (
12:44:20 Join San||Away [0] (
12:46:19 Quit mozetti ()
12:51:50 Quit Rob2222 ()
12:52:27 Join Rob2222 [0] (
12:56:30 Quit [1]SteL (" HydraIRC -> <- The professional IRC Client")
12:56:46 Join tiennou [0] (
12:56:53 Join Matze [0] (
12:57:56 Join XavierGr [0] (
12:59:37***Saving seen data "./dancer.seen"
13:02:09 Join amiconn [0] (
13:02:48 Join koseski [0] (
13:04:19 Quit goa ("Client suicide")
13:05:34Rob2222hi XavierGr
13:08:48BHSPitLappyhi, kirsten dunst
13:09:35 Quit tiennou ("CGI:IRC")
13:10:28 Quit koseski ("CGI:IRC (EOF)")
13:12:03 Join Membrillo [0] (
13:12:14amiconnHmm, no lamed here
13:15:15 Join goa [0] (
13:15:52t0masamiconn: looking for him?
13:16:08amiconnI had an idea for the horizontal scrolling patch
13:16:14t0mastell me...
13:16:46amiconnInstead of using the cumbersome, and not possible on all targets button combo...
13:16:59amiconn(mentioned that earlier today - it's aproblem on Ondio)
13:17:02 Join muesli__ [0] (n=muesli_t@
13:17:29t0mas(thinking of that... we need some button thing... as the AB-repeat doesn't work with the iriver buttons)
13:17:49amiconn...we could use BUTTON_(LEFT|RIGHT) | BUTTON_REPEAT instead
13:18:14amiconnThat's usually unused in lists, entering / leaving sub-lists is done with short presses
13:18:48Slasheriamiconn: that sounds great
13:19:14amiconnI don't know how much this will affect browsing experience,
13:19:44amiconnbecause the sublist entering / leaving will then trigger on the release instead of the press
13:20:00amiconn...i.e. a bit later
13:20:15t0maswe have to change more for that?
13:20:26t0masor just edit the key-bindings for the scrolling patch for ondio?
13:20:33amiconnNot just edit
13:20:55amiconnIt needs this _PRE thingy to distinguish short <-> long press
13:22:12t0masok, so we have to change the menu button assignments?
13:23:35 Quit ender` (" I was trying to daydream, but my mind kept wandering.")
13:25:53 Quit AliasCoffee ("Leaving")
13:32:53 Join ender` [0] (i=ychat@
13:36:08 Quit ender` (Client Quit)
13:37:08JdGordondoes any1 use the calendar plugin?
13:39:29Hansmaulwurfi would, but i have only the H120 JdGordon
13:39:47Hansmaulwurfor is there a patch for the H100 series?
13:39:51JdGordondo u know if it is ideal to show a weekly timetable?
13:39:58JdGordondoubt there is
13:40:30Hansmaulwurfwell, i have only seen some pictures
13:44:21 Join SteL [0] (
13:50:32 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
13:51:03Jungti1234I use RockCalendar
13:51:19JdGordonis it useable to show a weekly timetable?
13:51:48JdGordonto the coding cave batman....
13:51:57JdGordonsorry... ive gone loony
13:52:46CtcpIgnored 2 channel CTCP requests in 25 minutes and 44 seconds at the last flood
13:52:46*JdGordon really needs to be able to have the timetable on the iriv instead of in my damn wallet :p , too much hassle when its printed
13:53:40Rob2222has someone battery_benched his H300, yet?
13:54:20Jungti1234JdGordon: I want it too.
13:55:00JdGordonive only got a h300 so thats my main target... ideally it wont be hard to move to other targets...
13:55:48Jungti1234I have H120, H320. :)
13:56:13JdGordonhaha, lifes tough aye? :D
13:57:32*JdGordon needs to set up his 3rd monitor to code properly :'(
13:57:43JdGordonbut its too damn hot for 3 crts
13:58:01 Join ep0ch [0] (n=ep0ch@
13:58:55ep0chhi, where can i find out more about what the horizontal scrolling patch does? cant find anything about it...
13:59:59*JdGordon is curios about that also
14:00:27Jungti123405.06.36 # <lamed> update, and then use ON+RIGHT/LEFT on the fileviewer, and you'll understand
14:01:14Jungti1234I didn't know it too.
14:01:34Jungti1234But, I know now.
14:01:54Jungti123405.07.40 # <lamed> jungti: i will update the wikimanual soon :)
14:05:44 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
14:06:27JdGordonhows rockboy going? useable yet?
14:07:24JdGordonlast time i tried it it was too slow to actually use
14:07:57Jungti1234It supports frameskip automatically.
14:08:20Jungti1234I have much interesting game.
14:08:25ep0chhmm what are like the top 10 gameboy games?
14:09:08Jungti1234top 10? :)
14:09:09ashridah doesn't everyone own tetris?
14:09:39ep0chi know nothing about gameboy games with the exception of mario and tetris
14:10:56 Join ST [0] (
14:11:36JdGordonep0ch: same.. i never had a gameboy
14:12:06Jungti1234hey all
14:12:19 Join Febs [0] (
14:12:22JdGordonthanx... u prob shouldnt post that here tho
14:12:33JdGordonfreenode doesnt like that sorta thing
14:12:37Jungti1234Tetris is there!
14:12:50 Join ender` [0] (i=ychat@
14:13:49ep0chhmm its probably not a good idea posting links to roms in here
14:15:52Jungti1234 - GB
14:16:27 Join Rondom [0] (
14:16:29Jungti1234You don't know Korean... ^^;
14:18:25 Join mikearthur [0] (
14:18:28JdGordonlinuxstb: is it possible to do something like press stop on the h300 when its plugged into usb mode so it automatically unmounts and turns off instead of doing it with windows which is a pain?
14:19:39ashridahhrm, how is the device going to tell windows you want to disconnect it?
14:19:55JdGordonit cant auto-unmount itself?
14:20:02*JdGordon thought it coudl
14:20:22ashridahhow is it goign to know if windows has committed everything to disk yet?
14:20:31 Part ep0ch ("Kopete 0.11 :")
14:20:40JdGordonhmm.. good point...
14:20:44*JdGordon is a dill
14:21:01JdGordonactually no.. i blame my stupidity on the 2 43C days weve had here
14:21:04Rondomit's theorethically possible by installing a driver for windows ;-)
14:21:05JdGordonin a row
14:21:12 Quit Membrillo ()
14:21:17ashridahi hear you on that
14:21:20ashridahfricking roasting today :(
14:21:34 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
14:21:47Rondomthe driver would listen and the player sends an request to unmount the device to the pc
14:22:00ashridahRondom: how?
14:22:11ashridahthe player uses a standard ums-storage chip
14:22:14RondomI'm not sure if it is possible
14:22:33Rondom^^ that's what has just come into my mind
14:22:34ashridahso far as i know (i could be mistaken) there's no way for the player to add usb packets when it's active.
14:23:17ashridahit'd be nice if it could, but then you'd want to just build it into the usb-storage chip itself, and signal that
14:24:21Rondomashridah: maybe putting a file on the disk
14:24:29Rondombut polling isn't good
14:24:43ashridahif the usb-storage chip is active, rockbox does not have disk access
14:25:14*Rondom should better not talk about things he doesn't know about
14:27:46 Quit ST ()
14:27:57JdGordon(0,0) is top left correct? not bottom left?
14:28:03JdGordon(in rb)
14:28:06JdGordonon the lcd
14:29:30Rob2222if i have volume at -45dB do i know how much power the earphones consume?
14:30:15preglowdepends on the earphones
14:30:22preglowand the amp
14:33:28Rob2222MDR-ED31 headphones with a 16 Ohm / HP impedance, 108 dB / mW
14:33:32Rob2222whats the amp?
14:34:32preglowthe headphone amp in the iriver
14:34:54Rob2222so thats a point we didnt knew, right?
14:35:09 Join mirak [0] (
14:35:32XavierGrpreglow do you think that battery time can get very high if we make a test without earphones connected?
14:35:40 Quit Jungti1234 ()
14:35:44preglowi don't think it'll matter too much
14:36:04XavierGrbecause I did some tests without phones.
14:36:37XavierGrI did the test twice with volume 50 (old system) and volume 0 (new system) and there results were the same.
14:38:58 Quit linuxstb (Read error: 110 (Connection timed out))
14:39:29 Quit San||Away (Read error: 110 (Connection timed out))
14:39:39Rob2222Ill make some power consumption measurements with Multimeter now.
14:40:28Rob2222Do you think I can use the AC plug when I check if the charger is disabled (charging pin at GPIO) ?
14:41:57preglowpractical tests should always be done with earphones
14:42:21preglowsince people are going to refer to your results for practical use
14:42:58Rob2222i made my benches with phones at -45dB
14:43:17Rob2222now ill check the power consumption with and without
14:43:50JdGordonhow do i draw text to the lcd? just rb->lcd_puts(); or is there more that needs to eb done?
14:46:25preglowyou need to update the screen too
14:46:37JdGordonye, but nothing else? it doesnt work...
14:46:51preglowthat should be enough
14:47:47 Join Sacro [0] (n=root@adsl-83-100-174-192.karoo.KCOM.COM)
14:48:09JdGordonnop, blank screen....
14:49:40XavierGrSlasheri: Nice work with the "pops", "clicks", "tst" (call it what you want). I couldn't get and sort of them while seeking. Also changing folders I think went smooth. I got a small one when I was changing tracks while the buffer was filling.
14:49:49XavierGrI will do some more tests.
14:52:23JdGordongrr.... ok, my maths must be off...
14:55:03 Quit Hansmaulwurf (Read error: 104 (Connection reset by peer))
14:55:21JdGordonpreglow: soz, whats the diff between puts and putsxy?
14:55:28XavierGrWhere is the A-B option in iriver?
14:55:41XavierGrops sorry just found it
14:56:12JdGordonah, ok, dw.. badly named funcs
14:57:38XavierGrok I found the A-B option but which button triggers it?
14:59:41***Saving seen data "./dancer.seen"
15:03:24NicoFRrec+left : set A point
15:03:30NicoFRrec+right : set B point
15:03:40NicoFRrec+stop : clear points
15:03:49NicoFRthat's on H300
15:03:51XavierGrSlasheri: While your fix works if someone skips one folder (Hold PLAY + RIGHT/LEFT), If I press this combo twice I will get the same odd behaviour. (The first track of the folder will start for 0.5s and then it will restart)
15:04:00 Join tucoz [0] (
15:04:20tucozXavierGr, where did you find the option?
15:04:45tucozThe A-B option that is
15:04:49 Join lamed [0] (
15:04:57lamedhello all.
15:05:00XavierGrit is in the reapet modes
15:05:03lamedAmiconn: around?
15:05:07XavierGrhi lamed
15:05:17tucozok, thanks
15:05:20lamedhi man!
15:07:40tucozDo that work on the h1x0 ? I can not get it to work
15:07:42XavierGrtucoz for some reason I can't get it to work.
15:07:43lamedamiconn: I agree with what you said about |we could use BUTTON_(LEFT|RIGHT) | BUTTON_REPEAT instead| for the screen_scroll patch. I was thinking the same, but i didn't wanted to mess around already-existed keybinds. btw the ipoders might like using long clicks on the forward and backwards buttons.
15:07:54XavierGrif you find out how please tell me
15:08:21tucoz:) Seems like it gets stuck on ffw and rew.
15:08:24lamedSlasheri thanks for the compliments :D I Was trying to tell you guys it's the greatest patch for a loongg time!!
15:08:50tucozlamed: you wrote the scroll patch?
15:09:19NicoFRXavierGr: rec+left : set A point
15:09:23NicoFRrec+right : set B point
15:09:28NicoFRrec+stop : clear points
15:09:34amiconnRec+Anything won't work on H1x0
15:09:38lamedNicoFR you can't REC+ anything except ON
15:09:43NicoFRok sorry
15:09:49XavierGrNicoFR: Thanks but it wont work
15:10:04tucozlamed: good work. I never imagined that I it would be this usefull, but I am convinced now :)
15:10:05amiconnThe only possible 'shift' button on H1x0 is Play (BUTTON_ON)
15:10:10 Join matsl [0] (
15:10:12XavierGrso amiconn, will A-B work on H100?
15:10:45amiconnNot with the current keybinding
15:10:59XavierGroh okay, thanks
15:11:30lamedt0mas: last night me and lostlogic were talking about short ON+LEFT/RIGHT for AB repeat, and long ON+LEFT/RIGHT for folder change. that will cure the keybinding problem (btw with the same trick as the one that will be used at the fileview)
15:11:45lamedtucoz thanksss
15:12:03 Join webguest63 [0] (
15:12:04XavierGrSlasheri did quite a nice job with pops. Only one I can trace is with the TWICE (or more) change folder behaviour, that I said earlier.
15:12:22webguest63how does the horizontal scrolling work on the h120?
15:12:27lamed(and I hope you all guys are reading the log)
15:12:37webguest63and what's "screen scrolls out of view"?
15:12:43lamed14.00.27 # <Jungti1234> 05.06.36 # <lamed> update, and then use ON+RIGHT/LEFT on the fileviewer, and you'll understand
15:12:45tucozwebguest63: in the file menu hold on + left/right
15:13:00preglowi wonder if using a software enveloping function would help reduce pops completely
15:13:06 Quit paugh ("Leaving")
15:13:31tucozdid Slasheri use the trick from the uda docs?
15:14:01tucozfor pop-prevention
15:14:29lamedbtw the reason why scroll out of view is turned off by default, is that you can get the text totally out of view and I imagin some users won't Like it.
15:14:54webguest63i see
15:15:00webguest63how about horizontal scrolling?
15:15:11amiconnlamed: Btw, I think the horizontal scrolling looks really odd (sometimes confusing) with 'scroll out of view' switched off
15:15:37tucozwebgues63, isn't that what you do with on+left/right?
15:15:56lamedFIXME: the Right boundery for the screen_scroll is fake. (just set to 1000 pixels.) The real way would be to check what's the longest item in list by pixels and set this as the maximum marker.
15:16:15lamedamiconn I was just refering to that.
15:16:21webguest63oh tat's what it is?>
15:16:30lamed<lamed> amiconn: I agree with what you said about |we could use BUTTON_(LEFT|RIGHT) | BUTTON_REPEAT instead| for the screen_scroll patch. I was thinking the same, but i didn't wanted to mess around already-existed keybinds. btw the ipoders might like using long clicks on the forward and backwards buttons.
15:16:36webguest63oh right, nvmd i got confused with "vertical scrolling" for text
15:16:42webguest63which i thought was the one that was added
15:17:05 Part tucoz
15:17:29lamed<lamed> t0mas: last night me and lostlogic were talking about short ON+LEFT/RIGHT for AB repeat, and long ON+LEFT/RIGHT for folder change. that will cure the keybinding problem (btw with the same trick as the one that will be used at the fileview) (amiconn)
15:18:23 Quit Rondom (Read error: 113 (No route to host))
15:18:33 Quit JdGordon (Read error: 104 (Connection reset by peer))
15:18:38 Join tucoz [0] (
15:18:56tucozlinuxstb, and all other ipod 5G owners.
15:19:20tucozSeems like the headphones could be damaged by the 5g's
15:20:06preglowi wonder how that's possible
15:21:29tucoz(just testing)
15:21:45tucozsomeone had a suggestion that the tv-out somehow manage to send the video signal to the headphones aswell.
15:22:00tucozhehe, I sent the last message before the (just testing)
15:22:40tucozanyway. beware all you video-ipodders. see you
15:22:44 Part tucoz
15:23:18preglowi doubt the video output is enabled until you use it
15:23:20preglowand besides
15:23:34preglowi doubt a high freq signal can fuck up an earbud
15:24:28muesli__i've seeb horses puke
15:24:32muesli__(as we say)
15:25:11preglowi've seen horses do A Lot, but not puke
15:25:39muesli__thats just an idion that says, nothin is impossible
15:25:54 Quit DreamTactix291 (Read error: 110 (Connection timed out))
15:26:12preglowhaha, new one
15:31:56 Quit kernelsensei (Remote closed the connection)
15:31:59 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
15:35:54 Quit matsl (Remote closed the connection)
15:37:19 Quit ashridah ("Leaving")
15:49:17lamedBuG: NO usb reconazation while using rockboy!
15:50:38preglowyou mean recognition?
15:50:50preglowdo you need usb in rockboy? :>
15:50:53 Join thegeek_ [0] (
15:53:20 Join linuxstb [0] (
15:53:20 Quit webguest63 ("CGI:IRC (EOF)")
15:54:38linuxstblamed: USB-detection patch for Rockboy would be welcome :)
15:54:49mirakhey, can anyone acces to this video ?
15:55:02mirakit says it's not accessible to my country
15:55:24Sacromirak: i got it here
15:55:32Sacrothe video, not the message
15:55:40 Join perplexity [0] (i=heh31099@
15:55:43 Join tvelocity [0] (
15:56:25mirakSacro: where do you live ?
15:56:46Sacromirak: England
15:56:52mirakI am in france
15:56:58Sacroim sorry
15:57:15mirakI though it would have been a US stuff
15:57:25lamedI might find the time this week, with the ab repeat new keybinds for iriver & screen_scrolll new keybinds idea. cya all
15:57:27 Quit lamed ("CGI:IRC")
15:57:32mirakSacro: what's the size of the video Sacro ?
15:58:11mirakI will download it out of emule
15:58:14Sacrodoesnt say
15:58:33 Quit Febs (Read error: 110 (Connection timed out))
16:01:53mirak"shake hands with beef" is that an english expression ?
16:02:05mirakwhat's the meaning of that ?
16:02:20preglowmy nano does indeed usa a noticable amount of power when sleeping
16:03:01linuxstbIs that Apple's sleep or Rockbox sleep?
16:04:08preglowapple sleep
16:04:34preglowi just kept it sleeping for around twelve hours, and the battery meter was noticably slimmer when i turned it on
16:05:28Rob2222I need help
16:05:47Sacromirak: never heard that expression
16:06:41Rob2222Im doing power consumption measurements on my H300
16:06:51 Join Mmmm [0] (
16:07:09Rob2222atm im checking the disc consumption
16:07:26 Quit thegeek (Read error: 110 (Connection timed out))
16:07:41Rob2222I have NO PHONES / VOLUME0 / BL always ON
16:07:57Rob2222And i am in PAUSE mode.
16:07:58 Quit mikearthur (Remote closed the connection)
16:08:14Rob2222consumption: 159mA
16:08:26Rob2222PRESSING RIGHT
16:08:42Rob2222=> START SPINNING => 490mA max
16:08:56Rob2222READING => 410-420mA max
16:09:05Rob2222SPINNING => 260mA
16:09:07Rob2222OK so far
16:09:18Rob2222but now, after disc stops spinning
16:09:28Rob2222I have a power consumption of 190mA
16:09:32Rob2222not 150mA
16:09:49Rob2222Charger is disabled (checked GPIO pin)
16:10:00Rob2222NOW! I press PLAY AND PAUSE
16:10:20Rob2222and im back at 150mA
16:10:51Rob2222WHY the unit stays after loading a new song at 190mA in PAUSE mode _UNTIL_ I pressed PLAY/PAUSE
16:10:59XavierGrstrange! And what's the consumption with headphones?
16:11:14Rob2222ill up my txt in some minutes
16:11:27XavierGrokay nice
16:13:44Rob2222I know why
16:14:05Rob2222after loading the new track the CPU stays at 124MHz !
16:14:10Rob2222in PAUSE mode
16:14:28 Quit thegeek_ ("( :: NoNameScript 3.81 :: )")
16:16:21MmmmAmiconn...are you there? I found something that you broke! :)
16:18:32 Join mikearthur [0] (
16:19:00 Quit Sacro (Remote closed the connection)
16:19:21linuxstbMmmm: What's broken/
16:19:57 Quit uwe_ (Read error: 110 (Connection timed out))
16:20:01 Join Sacro [0] (n=root@adsl-83-100-174-192.karoo.KCOM.COM)
16:20:03Mmmmwhen you turn off the player (on h140) the sound doesnt switch off anymore until the player is off
16:20:20 Join uwe_ [0] (
16:20:21Mmmmit used to swithch off when the turning off prompt came off
16:20:36Mmmmwhich was very useful when not looking!
16:20:53linuxstbWhy do you say it's amiconn who broke it?
16:21:05Mmmmjens committed it
16:21:07linuxstbHave you found the offending commit?
16:21:16 Join San||Away [0] (
16:21:18Mmmmeasy fix: #define WPS_RC_EXIT (BUTTON_RC_STOP | BUTTON_REL)
16:21:18Mmmm#define WPS_RC_EXIT_PRE BUTTON_RC_STOP
16:21:30Mmmmin hwps.h
16:21:36linuxstbAh, so does it just affect the remote?
16:21:59Mmmmnope, also: #define WPS_EXIT BUTTON_OFF
16:21:59Mmmm#define WPS_EXIT_PRE (BUTTON_OFF | BUTTON_REL)
16:22:50 Quit mikearthur (Remote closed the connection)
16:22:56Mmmmcommit was on 13th jan....I've only just updated my build so was slow to spot it! :)
16:23:59linuxstbOK. I'm sure amiconn will read this in the logs.
16:26:30Rob2222XavierGr: Got it?
16:26:40MmmmActually i reprted it wrong! :D the sound now switches off immediately instead of waiting!
16:27:49linuxstbThat was the intention of the patch.
16:28:07Rob2222all add USB Mode power consumption...
16:28:40linuxstbMmmm: So is there a problem?
16:29:11Mmmmyeah... When you're not looking you cant tell whether it's switched off yet!
16:29:40Mmmmand I'm never looking! :D
16:30:25linuxstbAh, so when you shut down your player, you want the audio to keep playing until the unit actually powers off?
16:30:39linuxstbBut the new behaviour is for playback to stop as soon as stop is pressed?
16:30:52Mmmmno, just untill the "shutting down" screen comes up
16:30:56 Join mikearthur [0] (
16:31:06mirakwow I am again sseing alive arrestation
16:31:12mirakfrom by room window
16:31:14preglowwhy does this matter at all?
16:32:00MmmmLinuxstb: then you know when you can release the stop button, safe in the knowledge it will switch off
16:32:13 Quit ender` (Read error: 113 (No route to host))
16:33:25MmmmThe 4 lines of code I mentioned fix this perfectly!
16:33:27 Join ender` [0] (i=ychat@
16:37:07 Quit Sacro ("using sirc version 2.211+KSIRC/1.3.12")
16:37:12MmmmPreglow: were you talking to me or mirak?
16:37:20lostlogicmorning it is.
16:39:36mirakMmmm: no
16:39:48mirakMmmm: just I seen an arrest in my street
16:40:13solexxMmmm: sounds like a really useful feature. Most of the time I am shutting down my unit without looking, too
16:40:20XavierGrRob2222: Quite interesting results.
16:40:33XavierGrCould you upload them to the wiki.
16:40:46MmmmMirak: ooh, preglow: solex says it all! :)
16:40:58linuxstbMmmm: IIUC, the problem was that before it took a while for the WPS to react to the stop button being pressed - so the user didn't know if it was registered or not. Does your patch change that?
16:41:20XavierGrAlso, it seems that testing without phones didn't showed a spoiled result.
16:42:58Piaggh, i want to create a menu that allows the user to pick a color, this will get more complicated when costom collors are added, so i think it should have its own file and not clutter up the settings_menu.c
16:43:05Mmmmlinuxstb: hmmm it seems it needs two presses now! Now I see what amiconn was up to!
16:43:18Pii cant get the other file to work properly
16:43:32 Join _FireFly_ [0] (
16:44:12mirakwow man there is also a building on fire
16:45:38Mmmmlinuxstb: ahh I got the lines the wrong way around! tsk :) Testing again....
16:45:45preglowyou're living in downtown brooklyn or what?
16:46:34XavierGrmirak: Firefly did, he just entered the channel! ;)
16:46:40XavierGradd ^ it
16:49:07MmmmLinuxstb: it seems to work fine for me!
16:49:23amiconnMmmm: There is a reason why I did these changes. The old way (triggering on the release) was very unresponsive.
16:49:51lostlogichmm, there's gotta be another way than a lock to keep playback sane on weird actions. can't afford to have a lock involved in functions that may be called several times per decde callback.
16:49:55amiconnIt often happened that I thought the unit doesn't react to STOP because I pressed it too long
16:50:04amiconn...and there is no reason to trigger
16:50:18amiconnon the release, because we don't use long presses
16:50:37Mmmmamiconn:(except to turn off) I've been pressing play and stop now for a while and it seems ok!
16:51:17amiconnI don't really understand what this change may have broken
16:51:54Mmmmwhen you switch the unit off, now you cant tell when to let go of stop (unless you are looking at the screen)
16:53:27amiconnHmm, I won't have noticed this. I *always* stop playback before shutting down
16:54:12amiconnHowever, if this small mapping change causes the change in behaviour, then something is wrong with the code doing that does the stop-on-shutdown
16:54:37amiconns/that does//
16:56:17Mmmmaah... I havent looked at that, I just undid your change and noticed the functionality came back
16:57:11 Join musahi [0] (
16:57:21amiconnMy guess: Instead of properly checking for a shutdown event and stopping playback in response to that, the code might check the button events
16:58:47Mmmmline 623 gwps.c - # ifdef WPS_EXIT_PRE what's that space for after #??
16:58:48_FireFly_XavierGr: what have i done ??
16:59:35 Part musahi
16:59:44 Join musahi [0] (
16:59:45***Saving seen data "./dancer.seen"
16:59:47lostlogicMmmm: just indenting, has no function
17:00:00Mmmmso like commenting out?
17:00:20lostlogicno, the space changes nothing, as is standard in C
17:00:43Mmmmoh, ok.. my text editor doesn't like it! :)
17:01:12 Quit musahi ()
17:01:48Rob2222can it be, that the iriver FW has better MP3 decoding?
17:01:55Rob2222== less power?
17:02:14lostlogicRob2222: sure, the mp3 encoder is very new
17:02:17XavierGrFirefly: Only joking (though it was a bad one) mirak said (just before you enter) that a building next to him was on fire.
17:02:33 Quit mikearthur (Connection timed out)
17:02:36lostlogicI'm going to do something _very_ strange... and try to put the swcodec engine into UML.
17:02:36 Quit San||Away (Read error: 110 (Connection timed out))
17:02:43 Quit akaidiot (Read error: 110 (Connection timed out))
17:02:49 Join mikearthur [0] (
17:02:49XavierGrRob2222: Dod you have the power consumption from iriver firmware?
17:03:09Piwhat is the minimum things to #include for a new .c file?
17:03:20lostlogicpi: there isn't one
17:03:30Rob2222XavierGr: doing iriver measurements atm
17:04:06Pihmm, i keep getting - undefined refference to 'color_menu'
17:04:20XavierGrI mean good....
17:04:22amiconnMmmm: Wanna try something?
17:04:34Mmmmof course...
17:04:47amiconnIn gwps.c:
17:04:49lostlogicPi: you neet to include any particular header that your program uses...
17:05:06amiconn- Change lines 650ff (case SYS_POWEROFF) to read as follows:
17:05:14amiconncase SYS_POWEROFF:
17:05:22amiconnexit = true;
17:05:54Mmmmwhat about the bookmark?
17:06:00 Join DreamTactix291 [0] (
17:06:10Piin the main program, if i write include color_menu.c it works
17:06:26amiconnThen go further down to line 694 (just above the #ifdef AB_REPEAT_ENABLE) and insert the following lines:
17:06:51amiconnif (button == SYS_POWEROFF)
17:07:02amiconnThat should do the trick.
17:07:19amiconnThe bookmarks will be handled by the standard exit path
17:08:05Mmmmok....I'm on it...
17:09:32_FireFly_Pi why do you want ro include a source-file(*.c) >>
17:09:45amiconnThe idea is that a shutdown event will trigger a stop in exactly the same way as WPS_QUIT, let the standard exit path do some of its work (including audio stop), and then call the standard poweroff code
17:10:12Piit works when i include the source file but not when i include only the header
17:10:50_FireFly_Pi do you programm for rb or a standalone app ??
17:10:55Mmmmamiconn:So that last bit of code is right after audio_stop(); ?
17:11:26Mmmmok, and I'll redo your original fix....
17:11:33amiconnPi: _Never_ include a .c file, that's baad style
17:11:34_FireFly_Pi is color_menu.c from you ??
17:12:04 Join Matze [0] (
17:12:17Pi _FireFly_ yes i wrote it
17:12:33_FireFly_color_menu.c must also be included in the compile-stage
17:12:38Piamiconn: that is why i am trying to fix
17:12:41_FireFly_then you can simply use the header
17:13:03 Quit NicoFR ()
17:13:47 Quit SteL (Connection timed out)
17:14:08Pii keep getting the error undefined refference to 'color_menu'
17:14:40 Join SteL [0] (
17:14:49_FireFly_can you paste the whole error-msg-line
17:14:55Mmmmamiconn: do you want WPS_EXIT_PRE defined too with that?
17:15:07amiconnThere is no _PRE
17:15:26Pi /rockbox/build/apps/settings_menu.o: In function 'background_color_menu':
17:15:26Pi /rockbox/apps/settings_menu.hc:1510: undefined refference to 'color_menu'
17:15:54_FireFly_?? settings_menu.hc ??
17:15:58Mmmmamiconn: ahhh....that's what i thought....sound still stops the second you press stop! :(
17:16:25amiconnHmm, I think that's what you wanted...
17:16:34Pii cant type
17:16:34_FireFly_Pi is color_menu only defined in color_menu.c
17:16:48_FireFly_Pi why not copy and paste it :)
17:16:55amiconnPi: Why type? There is copy-n-paste...
17:16:55Pidefined there and in the header
17:17:15_FireFly_Pi how is it defined ??
17:17:21Pii cany coppy from the cmd window
17:17:40_FireFly_you can
17:17:41Piint color_menu (unsigned *ref);
17:17:50amiconnMmmm: It seems I don't undertsand the problem.
17:18:44Mmmmamiconn: sound should stop instantly if you press stop but it should wait if you hold stop until "shutting down" is shown and then sound should stop
17:19:16Mmmmdo you see what i mean?
17:19:27Pihere is color_menu.c
17:20:17 Join Kohlrabi [0] (
17:20:42Slasherihmm, i fixed that problem a long time ago.. or do you mean it will stop immediately when you are on the file tree?
17:20:48Slasheriit has still that issue
17:20:48amiconnMmmm: Hmm, there are confliciting requirements. I also want it to stop immediately, but if you trigger the audio stop on release, it's not instant
17:21:07amiconn...and since stop isn't even immediate, I ended up pressing too long
17:21:34amiconn...causing it not to stop, and me thinking the unit doesn't react to button presses
17:21:49Mmmmahhh your problem then is you dont release immediately when stopping!
17:22:12amiconnYes, probably for two reasons
17:22:16_FireFly_Pi: have you added color_menu.c to the SOURCE-file and does it get compiled ??
17:22:30Slasherithen i just think it should not stop if pressed longer but long enough..
17:22:32amiconn(1) The H1x0 buttons give almost no mechanical feedback
17:22:44Slasheriso user could "cancel" that button press
17:23:10 Join guituser47 [0] (
17:23:11Pi_FireFly_: what do you mean?
17:23:14Mmmmamiconn: that's why I never noticed a problem before....I always just tap stop to sop playback!
17:23:14amiconn(2) Stop isn't immediate even now.
17:23:36guituser47chi italiano?
17:24:15amiconnMmmm: With the old behaviour, sometimes it took me several (3..5) presses of Stop until it really stopped. This was annoying
17:24:41amiconnEither I didn't press hard enough, so the button didn't register, or I pressed too long
17:24:54_FireFly_Pi: in apps there is a SOURCES-file this file holds all source-files in the apps-dir which should be compiled, depending on the target
17:25:05Mmmmreally? I've been sitting here pressing play and stop with the old code for 15 mins and never a prob! :)
17:25:15Piahh thanks will try there
17:25:32_FireFly_you must add new source-files to this file so that they get compiled
17:25:36Slasherithis is interesting.. i have never had that problem
17:25:40_FireFly_maybe a reconfigure is needed
17:25:42amiconnMmmm: I think I have a solution....
17:25:50 Join Philip_0729 [0] (
17:26:18*Mmmm waits eagerly for amiconns solution....
17:26:25Slasheriand i found it very annoying when the playback before stopped immediately when i pressed stop and i couldn't tell when the unit has been shutdown unless i took it out of pocket
17:26:53amiconnRevert to old behaviour, but do *not* define WPS_EXIT_PRE
17:26:56MmmmSlasheri: that's exactly my point :)
17:27:21amiconnThis way, both short and long presses will trigger the exit on release
17:27:22Mmmmamiconn: ahhh...I'll give it a go...
17:27:33 Part guituser47
17:27:51amiconnIt might be that this causes unwanted wps_exit when leaving a menu
17:28:45amiconnIf this is the case, we will need WPS_EXIT_PRE and WPS_EXIT_PRE2, both being a valid precondition for WPS_EXIT
17:29:26amiconn#define WPS_EXIT_PRE BUTTON_OFF
17:29:46amiconn#define WPS_EXIT_PRE2 (BUTTON_OFF | BUTTON_REPEAT)
17:29:59amiconn#define WPS_XIT (BUTTON_OFF | BUTTON_REL)
17:30:25amiconnand then change line 624 of gwps.c to
17:30:51amiconnif (lastbutton != WPS_EXIT_PRE && lastbutton != WPS_EXIT_PRE2)
17:31:35amiconnEquivalent changes will be needed for remote, of course
17:33:48Mmmmwhen do you think there may be unwanted wps_exit ? when you press stop in menu? do people really do that? :D
17:34:44amiconnStop is one way of leaving menus
17:34:54MmmmYep, pressing stop at bottom of menu tree exits wps. i'll try PRE2
17:35:29preglowlinuxstb: btw, wiki says tick timer isn't accurate, and i'm pretty sure it is
17:37:12linuxstbYou wrote that :)
17:37:18linuxstb(I think...)
17:37:35Rob2222and all others who are interested
17:38:02linuxstbWe should probably test it by comparing against the ipod's microsecond timer. A plugin could do that.
17:38:27Rob2222The interesting thing is that it seems, the iriver firmware doesnt need to boost the CPU.
17:38:38linuxstbOr a debug option - it could be useful for other targets with RTC
17:38:51Rob22223 minutes i checked no boost while playing 128kbit ABR
17:39:00preglow128kbps shouldn't boost
17:39:05 Join einhirn [0] (
17:40:34Rob2222my unit did
17:40:36amiconnRob2222: Afaik the iriver firmware uses a different concept. Instead of having a base cpu clock and temporarily switching to a high clock (like rockbox does), the iriver firmware uses fixed cpu clocks depending on the status & codec
17:40:51Rob2222yeah, looks like
17:40:58preglowlinuxstb: doesn't panicf() work on rockbox yet?
17:41:02amiconnThis is less flexible
17:41:05preglowehh, ipod
17:41:20Rob2222but iriver needs less pover then RB-nonboost
17:41:47linuxstbpreglow: I don't know. I've never done anything specifically to make it work.
17:42:41Mmmmamiconn: IT WORKS!!!
17:43:10 Join gtkspert [0] (
17:43:16Mmmmnow to see to that remote...
17:43:24Rob2222what works?
17:43:46amiconnRob2222: These measurements are rather imprecise. I suspect the charger continues to consume power even in indicated-full state
17:44:26amiconnA precise method would be to open up the unit, disconnect the battery, and connect a regulated lab power supply set to 4.00V instead of the battery
17:45:13amiconnThen the lab power supply (or a separate amperemeter) will give precise readings
17:45:16Rob2222why should the charger charge without indicate it on the charge pin?
17:45:26amiconnI will do this sooner or later, at least with my H140
17:45:33XavierGrAlso the H300 is known to have less battery time that default firmware, right.
17:45:34Rob2222yes, sure. a lab power supply would be more precise
17:45:39amiconn(for evaluating deep-sleep mde)
17:45:45XavierGrI should do the same test on my H100.
17:45:48Rob2222but i did all measurements more than 1 time
17:46:08XavierGrBut I don't think that my multimeter will do a good job....
17:46:28amiconnXavierGr: Yes, that's known. I have plans to make the gfx drawing significantly faster, as this is consuming a lot of CPU time right now
17:47:37amiconnThat needs changes in font format & handling, and it practically requires linuxstb's bmp-in-build-system patch
17:48:22 Join petur [0] (
17:49:04amiconnThe gfx drawing speed will hopefully increase by factors ...
17:49:17preglowwill this be h3x0 specific or 16 bit specific code? :>
17:51:24Mmmmamiconn: here's a patch then for the off/stop issue...:
17:51:26amiconnA bit of both
17:55:05 Quit DreamTactix291 (Read error: 110 (Connection timed out))
17:56:36Mmmmoops..small error there, I've re-uploaded a correct one...
17:58:28 Join ender1 [0] (i=ychat@
18:04:14 Join DreamTactix291 [0] (
18:06:31 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
18:07:15 Join tulkas [0] (
18:07:24 Join Kohlrabi [0] (
18:07:38 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
18:08:01tulkasdo you know how to open .mi4 file?
18:08:36 Quit tulkas (Client Quit)
18:11:25 Quit ender` (Connection timed out)
18:13:35 Quit SteL (Connection timed out)
18:14:35 Join SteL [0] (
18:17:34lostlogicSlasheri: adding the mutex to pcmbuf cost ~4% boost on my test track
18:17:55Slasherilostlogic: oh :/
18:18:12Slasheriso then we should find other we to protect those functions
18:18:30lostlogicSlasheri: I'm working on graphing out playback.c, pcmbuf.c and pcm_playback.c to see what's goin' on with it.
18:18:49lostlogicand I think I've got another tweak for the pcmbuf_callback which will get back the 4% for now ;)
18:19:07Slasherihehe, sounds cool ;)
18:19:27lostlogicif only I could quit messing up my test case
18:22:54 Quit DrumRBoy320|Away (Read error: 104 (Connection reset by peer))
18:23:01 Join Nanji [0] (i=Nanji@spe-95-120.Reshall.Berkeley.EDU)
18:26:04 Quit DreamTactix291 (Read error: 110 (Connection timed out))
18:26:07lostlogicI cannot for the life of me figure out why inlining pcmbuf_num_used_buffers hurts performance by multiple %.
18:26:39lostlogicit's a 4 instruction function, should be an ideal inlining candidate.
18:27:13t0masI've seen weird things in that cases too...
18:27:54t0masrecursive function calling (calculation gcd iwth euclids algorithm) is sometimes faster than using a while loop
18:27:57t0masreally weird
18:29:25lostlogicI guess JSR is fairly cheap if the called function only uses scratch registers and takes no arguments (as in this case)
18:31:02Nanjihi....i am a newbie to rockbox, i just installed Rockbox on my iriver h340 and it plays all fine, but....
18:31:09Nanjihow do i charge my player through USB now?
18:31:42t0maslostlogic: I guess so...
18:31:52t0masmine was on x86, with 2 arguments...
18:31:59preglowlostlogic: how often is it called?
18:32:09Nanjican anyone help me?...
18:32:22 Join ender` [0] (i=ychat@
18:32:26t0masNanji: I have no idea... can't you just plug it in?
18:32:28Philip_0729Nanji: iirc hold record while using bootloader USB mode
18:32:35DBUGEnqueued KICK Nanji
18:32:35Nanjiit says USB mode...that' 143t0mas14
18:32:44Nanjiit fails 1414Philip_072914
18:33:14***Alert Mode level 1
18:33:14Nanjio nvm, now it works...i was trying to boot to iriver firmware at first 147Philip_072914
18:33:41Nanjithx all
18:34:05 Join slimx [0] (
18:34:29slimxlo all
18:34:29lostlogicpreglow: dma interrupt handler
18:35:34slimxnow i've got a compiled ipod3g bootloader to test
18:36:19slimxcan someone tell me how-to try it ;-) ?
18:36:49Philip_0729Nanji: lol having told you how to do it... it won't work for me ;)
18:36:50 Quit manhattan ()
18:36:58***Alert Mode level 2
18:36:58Nanjillol 148Philip_072914
18:37:00 Join Luthion [0] (
18:37:13***Alert Mode level 3
18:37:13Nanjimy battery is dying though :( 1412Philip_072914
18:37:16linuxstbslimx: Do the same as if you were installing the ipodlinux bootloader - but replace their "loader.bin" with your "bootloader.bin"
18:37:44slimxk thx
18:37:51linuxstbOr follow these instructions:
18:38:04linuxstb(which are basically the same as ipodlinux)
18:38:38 Join Chamois [0] (
18:38:39slimxi've seen some code that load a linux.bin
18:39:02slimxin bootloader/ipod.c
18:39:37 Quit Seed (Nick collision from services.)
18:39:43linuxstbslimx: The first step is just to get your bootloader displaying something on the lcd - that's very a big step.
18:39:46 Join Seed [0] (
18:39:59slimxhey hey :p
18:40:10slimxi guess you're right
18:40:41 Quit dpassen1 (Read error: 110 (Connection timed out))
18:40:51 Quit _FireFly_ ("Leaving")
18:41:11 Quit Seed (Nick collision from services.)
18:41:20 Join Seed [0] (
18:41:48slimxso let's install it
18:42:30 Join LinusN [0] (
18:42:45LinusNlostlogic: inlining code that reside in SDRAM is usually a bad idea
18:43:03LinusNbecause you don't utilize the cache
18:43:20preglowi've had excellent results doing it too
18:43:21lostlogicLinusN: hmm... there are two places it will be called, one is IRAM'd, so the performance penalty must come in the other.
18:43:25LinusNthat's probably why your inlining gets slower
18:43:27preglowit all depends on how often it's called
18:43:43 Join San||Away [0] (
18:43:55LinusNgotta run, cu
18:43:58 Part LinusN
18:44:08 Quit ender1 (Connection timed out)
18:47:14***Alert Mode OFF
18:51:18 Join Paul_The_Nerd [0] (
18:52:52linuxstbslimx: Any luck?
18:53:22lostlogicSmart man that Linus −− inlining the IRAM'd version, but not the SDRAM'd version gives the best of both worlds.
18:54:34mirak<LinusN> lostlogic: inlining code that reside in SDRAM is usually a bad idea
18:54:35mirak<LinusN> because you don't utilize the cache
18:54:41mirakwhat does he mean precisily ?
18:55:11preglowdo you know what a cache is, then?
18:55:40preglowif you inline the function, every instance of it will be moved to where it's called, so it'll have to be loaded into the cache again
18:55:53preglowif the function isn't inlined, it's always located at the same address, so the code can be pulled from the cache
18:56:15mirakalso about that I have a question
18:56:29mirakhow can you inline a function and declare it to IRAM at the same time ?
18:56:43lostlogicmirak: if it's inline, it will be in RAM where the caller is
18:57:04lostlogicmirak: I just copied the code out into the one I wanted it inlined, and then left the function for the other instance
18:57:44mirakI don't understand your phrase
18:58:33Matzewhich sentence?
18:58:45lostlogicIf a function is inline, it is copied to where it's called. Therefore it will occupy the same RAM section as the calling function. What I did was physically copy the code into the calling function in one instance, but not in another.
18:58:47miraklostlogic: "I just copied the code out into the one I wanted it inlined, and then left the function for the other instance"
18:59:07Matzehe has function abc() and function czx();
18:59:11Matzehe wanted to use abc in czx..
18:59:20linuxstbYou basically have to have two copies of the function (e.g. with different names) - one declared inline, and one declared as ICODE_ATTR.
18:59:28Matzeso he put in INLINE there. means: the contenct of the function is there, not only a call to the function
18:59:28mirakhow can you do it in "one instance but not the other" ?
18:59:43Matzehe left the original function as it is
18:59:49***Saving seen data "./dancer.seen"
18:59:51Matzejust made it inline aditionally
19:00:36slimxi've got it displaying something
19:00:55linuxstbCongratulations. The port's well on its way then.
19:01:13slimxbut not really beautifull :p
19:01:16 Join ender1 [0] (i=ychat@
19:01:19Matzeslimx: what are u trying?
19:01:24slimxpixels ...
19:01:32 Quit ender1 (Remote closed the connection)
19:01:36slimxrockbox bootloader on iPod3g
19:01:44Matzenice :)
19:01:47slimxbacklight is good
19:02:16slimxbut the lcd driver print just bad things ...
19:02:26linuxstbThat's not surprising for a first attempt.
19:02:29Matzewell so maybe u have to change row/col order?
19:02:34Matzeor something like that...
19:02:37slimxin fact it print something
19:02:55slimxthat's nice
19:03:04Matzehardest step :)
19:03:48slimxwhat about making some diff stuff
19:04:47slimxso i think that the h100 lcd driver can't be used with the ipod :p
19:04:55 Quit Rob2222 (Read error: 104 (Connection reset by peer))
19:05:31 Join Rob2222 [0] (
19:06:09 Quit ender` (Read error: 110 (Connection timed out))
19:06:16slimxhow can i run a debuger with the bootloader ?
19:06:53 Quit Mmmm ()
19:07:16linuxstbslimx: You can't.
19:08:13Rob2222Has any dev recognized, that in PAUSE mode the CPU stays at full speed after switching to the next title? You anyone test this?
19:10:48 Join akaidiot [0] (
19:11:33 Join thegeek [0] (
19:15:08 Join Lear [0] (
19:15:38linuxstbamiconn: Does the h1x0's LCD remote uses horizontal packing for the pixels?
19:17:34linuxstbslimx: The first thing you need to work out is the format of the data that you have to write to the LCD - you should be able to find that by reading the IPL source and/or trial and error.
19:17:59 Join Rondom [0] (
19:18:57preglowlinuxstb: looks like there should be plenty of optimising potential in ipods display driver
19:19:18linuxstbYou mean the lcd_update() function and such-like ?
19:19:36coobpreglow: does the compiler make a real mess of it?
19:21:09preglowcoob: haven't had a look at disassembly yet
19:21:54 Quit gtkspert ()
19:21:59linuxstbBut are you talking about the code in lcd-ipod.c or lcd-16bit.c (or both) ?
19:22:02slimxok linuxstb i'm looking for that
19:23:10slimxbut i know they are unsigned char * in ipl source
19:24:12preglowlcd-ipod.c, mainly
19:25:57slimxipl writes it by rows so am i
19:27:05preglowyeah, don't complicate it yet
19:28:54 Part Philip_0729
19:30:36 Part Paul_The_Nerd
19:30:49 Quit darkless (Remote closed the connection)
19:32:23 Join darkless [0] (
19:37:25amiconnlinuxstb: No, the H1x0 and H3x0 remotes use vertical pixel packing like the archos and H1x0 main LCDs
19:38:03amiconnSo the H3x0 will need to support both mono bitmap formats, which requires an extended font format
19:40:11Pican you add a file to the .rockbox folder at the compiling stage?
19:44:48linuxstbamiconn: OK, thanks. I'm working on my bmp-in-build-system patch again now, and am hoping to commit something this evening.
19:51:00 Join infamis [0] (
19:54:03Papricamm its good?
19:55:21t0masPi: what kind of file?
19:55:52Pia data file that contains the defalt colors
19:56:11t0masI don't know
19:56:17t0masI assume it's possible
19:56:32t0masI think Bagder knows, he has done most of the makefiles iirc
19:57:01XavierGrmy god!!!! The world record for the longest continuous bout of hiccups goes to Charles Osborne (18941991) from Anthon, Iowa. The hiccups started in 1922 at a rate of 40 times per minute, slowing to 20 and eventually stopping in February 1990 a total of 68 years. [2]
19:58:56slimxpatch commited
19:59:00XavierGrImagine to live the rest of your life with hiccups! :X
19:59:07slimxna submited :p
19:59:07t0masit must be in his head...
20:01:16linuxstbPi: I think that's controlled by the tools/ script
20:02:53linuxstbslimx: There doesn't seem to be any attachment to your patch.
20:03:48linuxstbAlso, I don't think you should use lcd-ipod.c - even though it's got a generic name, I think it would be cleaner to keep that for just the colour LCDs.
20:04:07linuxstb(if CVS supported renaming, I would rename it to lcd-ipodcolor.c)
20:04:13CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
20:04:13*linuxstb waits for preglow...
20:05:32*preglow stops by and mentions svn
20:07:03San||Awayhey, does work for anyone?
20:07:12 Join DreamTactix291 [0] (
20:07:18PiI am having trouble getting the patch to include my new files
20:07:20Pitryed cvs diff -uN > color.patch
20:08:47t0mashm... that is the right command...
20:09:08t0maswhat does it output? (don't paste > 5 lines here please)
20:09:13Piat the top it says:? apps/color_menu.c
20:09:13Pi? apps/color_menu.h
20:09:27linuxstbYou can't do that unless you've done "cvs add" - which you can't do unless you have cvs write access...
20:09:47t0masPi: what does the patch do?
20:10:33Piit sets the fore and background colors of all the menues etc and saves them in general settings
20:11:02t0mashave you done cvs update in the past hours/day?
20:11:15Pithis is the older version
20:11:48Piupdated about an hour ago
20:12:06t0masand have you tested if it compiles for all targets?
20:12:30Pinot yet
20:13:37Piit has #ifdef HAVE_LCD_COLOR round most of the code
20:13:41 Quit SteL (Connection timed out)
20:14:05Piwhich should i test it compiles on?
20:14:48t0masiriver H1x0, archos player and some color target?
20:15:11t0masbecause 1. has bitmap LCD but no color and 2. has no bitmap LCD
20:15:20t0mascheck carefully for warnings....
20:17:22Piok there are errors with iriver h1x0
20:17:24linuxstbI would check an archos recorder (mono lcd) as well.
20:17:39t0maslinuxstb: Iriver H1x0 is mono lcd...
20:17:46linuxstbNo, it's greyscale.
20:17:50t0masooh wait...
20:18:10t0masPi: errors? or warnings?
20:18:29 Join SteL [0] (
20:19:21Pierrors, i changed LCD_DEFAULT for the global setting which is not deffined by non color targets
20:22:35slimxsorry for delay
20:22:54slimxthe patch is in sf now
20:23:05slimxi had to bzip it though
20:26:04linuxstbslimx: Your patch contains a binary file at the end - tools/codepages
20:26:07 Join Kaggen [0] (
20:26:21 Quit mirak ("adieu jolie fille madrileneeeeeeeeee")
20:26:46lostlogicasdf;lasjkfas;ldkjf I hate computers.
20:27:48 Part infamis
20:29:13Piif(screen->depth > ??) what should go here to differenciate between colour and non colour?
20:31:09linuxstbUse HAVE_LCD_COLOR if you want to differentiate between colour and non-colour
20:32:59 Join webguest83 [0] (
20:36:13Pihmm, lcd-h100.o and lcd-16bit.o are fighting, i persume i should have deleter the build folder before compiling
20:36:21lostlogicSlasheri: you around?
20:36:54Slasherilostlogic: yes, but going soon
20:37:11 Join mirak [0] (
20:37:29linuxstbPi: Yes. It's best to create new directories for each build - e.g. "build-h100", "build-h300" etc
20:37:47lostlogicSlasheri: quickly then... I've got some changes that get back about 4.5% boost, but on skip, I get one pcmbufdesc worth of playback from the track skipped from
20:38:42lostlogicSlasheri: Any idea what could possibly cause that? My change is the keep the 'desc' around as a static instead of loading it from the array on every interrupt, which should be OK, because it should have always been set by the previous interrupt call.
20:40:02Slasherihave you changed any behaviour how the playback pauses, clears the buffer or something like that?
20:40:19Pithat is the h1x0 compiled and i didnt see any warnings
20:40:33lostlogicnope, no changes in playback.c...
20:40:51 Quit SteL (" HydraIRC -> <- IRC for those that like to be different")
20:41:05 Quit akaidiot (Client Quit)
20:41:14 Join akaidiot [0] (
20:41:21 Quit Luthion (Read error: 110 (Connection timed out))
20:42:00 Join Mmmm [0] (
20:43:02MmmmSo amiconn, are you going to commit that fix for turning off/stopping or are you not convinced?
20:44:59Pithe archos player has failed, just after 'descramble'
20:45:35 Join RotAtoR [0] (
20:46:18Pii have no idea what is wrong with it
20:48:29Pigcc -O -ansi -g -g sh2d.c -o sh2d
20:48:29Pigcc -DHAVE_LCD_CHARCELLS -D__CONFIG_H__ -I../firmware/export generate_rocklatin.c
20:51:12KaggenPI: I don't really know what you're doing so I'm sorry if it's a nonsense question, but have you run configure in a fresh build folder and compiled with make?
20:51:56Pii think so, will try again
20:51:59KaggenIf the error apperade at a recompilation you can also try 'make clean'
20:52:19Kaggenapperade -> appeared
20:52:41Pino, i think i make a new folder for the arcos
20:53:39Kaggenchanging target without running make clean or a starting from an empty build-folder is probably bad
20:57:31Pisame again, though this time it made sh2d and rocklatin before it complained
20:57:32Pilimits.h:12275: limits.h: No such file or directory
20:59:52***Saving seen data "./dancer.seen"
21:00:01 Quit amiconn (" HydraIRC -> <- Try something fresh")
21:00:22Kaggensounds like a build enviroment error, but not sure. Isn't there a how-to on the rockbox wiki somewhere?
21:00:27Pisame again, though this time it made sh2d and rocklatin before it complained
21:01:34Piyea, it is hapening so fast and it doesent look like other errors that i have seen before
21:02:50KaggenHas it worked before? if so, have you changed any code?
21:03:49Pii have never tryed to compile for the arcos player before, but i can compile for iriver targets
21:05:06KaggenI haven't tried to archos target myself so I don't know any details but if the release you have is good I would presume it to work if iRiver compiles.
21:05:51linuxstbPi: Do you have an sh1 cross-compiler installed?
21:06:10Piarcor recorger has same effect
21:06:31Piarchos recorder has same effect
21:06:32 Join JoeBorn [0] (
21:06:33linuxstbWhat does typing "sh-elf-gcc -v" tell you ?
21:06:46 Quit Mmmm ()
21:07:22Kaggenahh, of course, the compiler...
21:07:57Pireading specs from /opt/sh1/lib/gcc-lib/sh-elf/3.3.1/specs
21:09:22linuxstbThat's fine then - you have the sh1 cross-compiler.
21:09:35PiConfigured with: ../../gcc-3.3.1-3/configure −−target=sh-elf −−prefix=/opt/sh1 −−enable-language=c −−disable-nls
21:09:51linuxstbAre you using the devkit?
21:10:00Piyes (i think)
21:10:40linuxstbThe devkit can't compile the Archos targets any more - as you've just found. It's missing limits.h (and other possibly files)
21:11:18Piok that explaines it, what should i do?
21:11:26linuxstbInstall Linux or Cygwin...
21:12:30linuxstbThe devkit isn't an "official" part of Rockbox - so it's not always up to date with current Rockbox requirements.
21:13:57 Join Xerion_ [0] (
21:16:06 Quit Chamois (" HydraIRC -> <- State of the art IRC")
21:18:00 Quit Rondom (No route to host)
21:18:07 Join ender` [0] (i=ychat@
21:19:02 Quit akaidiot ("( :: NoNameScript 3.81 :: )")
21:20:26Piupdated the source forge submition with all the new code if someone else wants to try
21:20:34 Join Bger [0] (n=Bager@
21:20:46Bgerevening :)
21:22:11Bgerwhat's this "horizontal scrolling patch" ?
21:22:42lostlogicbger manual look to the side to view long filenames in the dir browser
21:23:03lostlogicBger: it needs work (ie deactivating the automatic scrolling of the long name under the curser when it's active)
21:23:12Bgerah :)
21:23:57 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
21:24:04Bgeri should think about the remote keymappings of this ...
21:24:25 Join TCK [0] (
21:26:09 Nick San||Away is now known as San| (
21:27:51peturMmmm: are you around?
21:28:46Bgerpetur suggestions for h300 remote buttons for this horizonal scroll ? :)
21:29:04*petur doesn't have any remote :(
21:29:06Bgerthere are no free buttons :(
21:29:28peturcould have used one this afternoon (stealth recording)
21:29:43 Join Kohlrabi [0] (
21:30:04peturfield testing recording gain patch :D
21:30:09Bgerthe only possible combination is MENU (GPI) + some other
21:39:51preglowi like that newest patch
21:41:41 Quit thegeek (Read error: 104 (Connection reset by peer))
21:42:24 Join Nuxator [0] (
21:43:16 Join vger_ [0] (
21:44:36 Join pinkutank [0] (n=ddd@
21:44:48Pihas anyone had a look at the color settings2 patch/ does it work with the archos targets?
21:45:15 Join Toni1 [0] (
21:45:24Toni1Hi there
21:45:47Nuxatori've got a problem with directory cache
21:45:53preglowToni1: hello
21:45:59preglowToni1: see you've been busy
21:46:13Toni1Hi preglow, already seen my new patch?
21:46:20preglowyeah, i'm looking at it as we speak
21:46:44Toni1fine, I have updated the mp3encoder a little bit too
21:46:49Nuxatori enabled it but it isn't able to finish scan
21:47:02XavierGrHi Bger! Wazzup? I thought about the circular battery elements but it will need an element reordering to work pleperly
21:47:06Toni1now high bitrates should perform better
21:47:15linuxstbToni1: Cool patch. :)
21:47:26linuxstbWhere have you been the past year? :)
21:47:29Nuxatoris there any way i can erase my settings without entering rockbox?
21:47:37preglowToni1: why have you reintroduced IDATA_ATTR and LINE_ATTR defines?
21:47:41preglowthey don't belong there
21:48:05preglowthey're not even used, heh
21:48:08Toni1oh sorry, this was not intentionally
21:48:34linuxstbNuxator: If you are on the iriver, then hold down record as you are booting.
21:48:34 Quit darkless (Read error: 104 (Connection reset by peer))
21:48:49 Join darkless [0] (
21:48:57Nuxatorplease help rockbox can't start any more with dir cahe enabled
21:49:48XavierGrhold rec while booting.
21:49:59preglowand yeah, the mp3 encoder was originally several files, yes?
21:50:20Xerion_record during booting loads into iRiver firmware
21:50:32Nuxatoryes but it boots irver's firmware
21:50:36Xerion_it doesn't clear rockbox settings
21:50:49Toni1preglow.: yes
21:50:51peturstart holding REC as soon as you see the bootloader
21:50:56Nuxatorbut i want to erase the cache dir setting to use rockbox again
21:50:59linuxstbNuxator: Hold on, and then when the bootloader starts, release on, and start pressing (and holding) record.
21:51:03Bgerpress it after the bootloader starts reading the rockbox.iriver
21:51:07Xerion_ok i see :)
21:51:19Xerion_just good to know :p
21:51:27preglowi'd like to split it to several files again
21:51:30XavierGroops sorry for that.
21:51:34preglowi'm not a big fan of huge c files
21:51:38Toni1preglow: still the ind2idx table is not complete, but only 320kBit seems to make use of the high values (bigger than 255)
21:51:39linuxstbpreglow: I agree.
21:51:40 Join ender1 [0] (i=ychat@
21:51:48 Quit TCK (Read error: 110 (Connection timed out))
21:51:51Nuxatormany thanks
21:51:58preglowToni1: ok, so 320kbps still wont work?
21:52:41Toni1preglow: yes it works, but unfortunately I have no good ears, and no tools to measure the quality difference
21:53:01Nuxatorok it worked
21:53:03Nuxatormany thanks
21:53:19linuxstbToni1: Do you handle configuration? i.e. the options for encoding can be different for each codec (kbps doesn't make any sense for WAV or wavpack for example)
21:53:25preglowToni1: is the current iram partitioning completely random or have you checked to see those arrays are the most important one?
21:53:35preglowlinuxstb: yeah, i was about to mention that, we need a way for codecs to export settings
21:54:12Toni1linuxstb: yes that's true: so I use the quality settings: 0-5 for mp3 6 for wavpack and 7 for wav
21:54:27mirakis lossless compression like flac not usable realtime ?
21:54:42Toni1preglow: not in detail, more like a feeling
21:55:03Toni1mirak: wavpack is :)
21:55:03LearNote that Shine was used in a recent listening test - as low anchor...
21:55:39preglowLear: well, it's crap quality, but we don't exactly have many choices
21:56:52 Quit DreamTactix291 (Read error: 110 (Connection timed out))
21:56:59Toni1mirak: but wavpack seems around 900kBit against 1440 kBit raw wav
21:57:05linuxstbmirak: As Toni1 says, wavpack lossless encoding is realtime (and just as good, if not better than FLAC). I don't think we'll be able to get the libFLAC encoder working realtime.
21:57:14 Join solexx_ [0] (
21:57:49linuxstbToni1: So you don't let the user choose the codec?
21:57:50preglowwavpack compressed better than flac
21:57:59 Join akaidiot [0] (
21:58:24 Quit akaidiot (Client Quit)
21:58:25linuxstbIs that also the case for the Rockbox version of the encoder though?
21:58:27Toni1linuxstb: the user can choose between raw wav / mp3 / wavpack by the quality setting
21:58:30preglowlinuxstb: no idea
21:58:33 Quit Nuxator ("CGI:IRC (Ping timeout)")
21:59:09linuxstbToni1: Yes, I understand that. But I still think we should make it more explicit than that - and give the options for configuration of each codec. But that's a future enhancement.
21:59:13Toni1linuxstb: also no idea
21:59:38 Join SteL [0] (
22:00:11 Join Benacool [0] (
22:00:12linuxstbBut anyway, we only really need one lossless encoder. I've certainly got no intention of looking at libFLAC again.
22:00:29Toni1linuxstb: ok, but if quality settings would saved to disk, is there still need for?
22:02:46 Join Mmmm [0] (
22:02:51preglowbtw, does the encoder plugin memory currently share space with the decoder?
22:03:45Toni1preglow: yes, but with audio_stop() it should not matter
22:04:43Toni1and for the voice decoder I created a voice_stop()
22:05:02 Join linuxstb_ [0] (
22:06:12linuxstb_Toni1: I'm just think about the future and if we added other encoders, such as speex. But as I said, it's something that can be changed in the future, I don't think we should worry about it yet.
22:06:25 Quit linuxstb (Nick collision from services.)
22:06:31 Nick linuxstb_ is now known as linuxstb (
22:06:41preglowwhat naming scheme should we use for encoder pluginms?
22:07:13Toni1linuxstb: ok, and quality setting is really not the very correct place for encoder selection
22:07:16 Join thegeek [0] (
22:07:16linuxstbWhere are they currently stored? Is there something like an apps/encoders/ directory?
22:07:25linuxstb(I mean in Toni1's patch)
22:07:27preglowi think they should reside in codecs/ too
22:07:51Toni1preglow: I think so too
22:08:06preglowbut yeah, we need a naming scheme
22:08:37preglowperhaps we should change extension to .dec and .enc
22:08:47pinkutanklol I patched the 1.29 j verison with rockbox
22:08:56pinkutankit still loads iriver
22:10:53Toni1have to go, bye all.
22:10:57 Part Toni1
22:12:33 Quit Lear ("Chatzilla [Firefox 1.5/2005111116]")
22:13:08 Quit solexx (Read error: 110 (Connection timed out))
22:13:44 Join matsl [0] (
22:15:10MmmmPetur: I'm here now, how about you?
22:15:37Mmmmhave you had a look at the new patch on SF?
22:15:51peturdoing that now
22:17:08peturI think we can keep mic recording mono, and use that combined L/R selection thing for the other latforms as well.
22:18:17MmmmYou mean get rid of all that amazing mic code? :-0
22:18:56peturno, always handle mic as mono (L=R)
22:19:37Mmmmbut it works well, no? it could be useful to have balance control
22:19:53 Quit ender` (Connection timed out)
22:20:26Mmmmdo you think it's too complex?
22:20:55peturnot for mic I think. I think I even saw that Tony1's new patch only stores left channel if mic
22:21:34peturmy alternative turned out to be even more complex... ;)
22:21:55Mmmmheh heh...its difficult isnt it? you dont know till youve tried eh?
22:22:20linuxstbAnyone know about committing binary files to CVS (from Linux) ?
22:22:44MmmmPetur:what do you think of my anti-literal technique
22:23:06 Join thegeek_ [0] (
22:24:11peturbetter than using literals. Only works if they are multiples... then again, this code is quite dedicated atm
22:25:47Mmmmthey'd have to be mad to make hardware where the dig gain didnt go exactly into one analogue gain! but stranger things have happened i suppose :)
22:26:49preglowlinuxstb: there's a flag for it
22:27:22MmmmPetur: what does Tony1's patch do?
22:27:34peturmultiple encoder support
22:27:49Mmmmahh.. Thats been committed right?
22:28:19preglowno, and it's not quite ready yet
22:28:21linuxstbpreglow: Yes I've found it
22:28:23preglowno build system support
22:28:23 Join amiconn [0] (
22:28:40 Quit slimx (Remote closed the connection)
22:29:13 Quit vger_ ("gtkBitchX-1.1-final -- just do it.")
22:29:18 Quit Pi (Remote closed the connection)
22:29:20 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
22:29:26 Join Kohlrabi [0] (
22:30:18MmmmSo if it only stores left channel from mic in then we have to have mono control right?
22:30:30Mmmmas in HAVE to
22:31:13linuxstbOK, first attempt at bmp2rb support in the build system is commited to CVS - don't forget to re-run configure after updating.
22:31:42peturMmmm: that's what I said. There's really no use in allowing separate L/R gain when the source is mono
22:33:11Mmmmon the fly balance control though? I like that! but if the encoders only sample one channel, then there is no point
22:34:17peturbalance for what? it's the build-in mic...
22:35:23MmmmI'm really racking my brains here....there MUST be a use for it... :D
22:36:51 Join ansiVirus [0] (
22:37:06 Quit SteL ("Chatzilla 0.9.67+ [SeaMonkey 1.0b/2005121900]")
22:38:07Bgermaybe i'm really dumb, but what does this " error: invalid operands to binary !=" means here:
22:38:20Bgerif(i<buffelements && ((i>0 && (bat[i-1]!=rb->battery_voltage())) || !i))
22:38:42MmmmPetur:Ok...I give up :D So we should change the display too to just one level indicator.
22:38:54Mmmmfor mic in
22:38:54 Quit JoeBorn ("Leaving")
22:39:49linuxstbBger: Try adding more brackets
22:40:03 Quit thegeek (Read error: 113 (No route to host))
22:40:10linuxstbIMO, you can't have too many brackets.
22:40:24Bgerah, i'm dumb :)
22:40:34Bgerbat[] is struct :)
22:40:39*Bger hides
22:40:43linuxstbOr that... :)
22:40:57 Quit uwe_ (Client Quit)
22:41:10Bgerbrackets were ok (long live for "%" in vim:))
22:41:30 Join Chamois [0] (
22:41:49linuxstbRotAtoR: Do you still have the original bmp files you used for bejewelled?
22:41:59linuxstb(the ones you used bmp2rb with)
22:45:35amiconnlinuxstb (and Toni1 if you read the logs): I have some encoder remarks:
22:45:54RotAtoRlinuxstb: yes, i have them here somewhere
22:46:08amiconn- The recording settings (i.e. quality etc) are saved to disk
22:47:21amiconn- It would probably very helpful for our blind users (and I'm sure we *will* get requests for it) if the voice can be made to work during recording
22:47:45peturisn't it a hardware limitation?
22:47:52linuxstbRotAtoR: I would like to add the original bmp files to CVS (and remove them from the .c file) now that I've committed bmp2rb build support.
22:48:15peturunless you pause input (for peakmeters) to output voice
22:49:05RotAtoRlinuxstb: yes, definitely, your new patch looks like it will clean up the source files a lot
22:50:01RotAtoRdo you want me to send you the bitmaps or should I make the changes?
22:50:29linuxstbI'm very happy to let you do the changes.
22:50:29 Quit pinkutank ()
22:50:59RotAtoR:) ok, i'll take a look at how to implement the changes then
22:51:21linuxstbThanks. Let me know if you have any queries - but the Sudoku changes should give you a good idea.
22:51:40RotAtoRnow it should be much easier to add graphics to the side bar too...
22:51:41amiconnpetur: Afaik the hardware is able to do full duplex. The peakmeters are 100% software
22:51:47RotAtoRok, thanks
22:51:58amiconn(calculating the maximum across a part of the pcm buffer)
22:52:44preglowhardware can do full duplex
22:53:26peturDidn't know about the full duplex
22:54:18preglowstill, i don't know when we'll need to use the full duplex caps
22:54:27preglowapart from my effect box project, of course
22:54:45linuxstbPaprica: How is brickmania coming along? I'm keen to commit it to CVS now we can move the bitmaps out of the .c file(s).
22:55:35amiconnpreglow: Voice UI during recording
22:58:24preglowamiconn: that'll be hell
22:58:49linuxstbOuch, nasty build table...
22:59:03preglowanother case of swapping memory like hell while something intensive is going on
22:59:30 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
22:59:53***Saving seen data "./dancer.seen"
23:01:49 Join Kohlrabi [0] (
23:02:18 Quit Nanji ("ۣ˯оԼԣ")
23:02:37BHSPitLappyhey #rockbox
23:02:41 Quit Chamois (Read error: 104 (Connection reset by peer))
23:02:55peturMmmm: are you going to adapt your patch or shall I?
23:03:03 Join Chamois [0] (
23:03:13MmmmI'm doing it right now! :)
23:03:32MmmmIt hurts to get rid of all that hard work! :)
23:03:35 Join Coolhawk [0] (
23:04:00MmmmNow it looks so simple..:D
23:04:07peturI saw you still had some tabs and 80+ lines... while you're editing
23:04:40*petur remembers how Linus gave him a hard time with his bad patch
23:05:23Mmmmheh heh...I'll look out for them ;)
23:07:52 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
23:08:31*linuxstb commits fix for simulator builds.
23:10:44Bgerpetur haha :)
23:10:58CoolhawkOk, for everyone interested I've finally finished the schematics over the charge unit of the iAusio X5, you may locate that here : If anyone needs further help on this issue, just post a reply then :)
23:11:59 Quit Chamois (" Want to be different? HydraIRC -> <-")
23:13:07peturhehe, I like hand-drawn schematics :D
23:13:44Coolhawkhehe well second draft is better, the first one was on the fly as I measured
23:14:17MoosCoolhawk: you could put it in the wiki
23:15:36CoolhawkSure I could do that, maybe even Linus could add to his pages under the charging section
23:15:56CoolhawkI would think this could come in handy in the future for some
23:16:33Moostemporary issue
23:17:52Moosyou could put them here in waiting Linus publish the final version one
23:18:24 Join TCK [0] (
23:18:27Moosor here
23:20:52 Quit Xerion_ (Read error: 104 (Connection reset by peer))
23:21:37CoolhawkYeah, I think I'll do that, maybe I'll get the time to make it proper in my Protel99 :)
23:22:01MmmmPetur: OK, its working... I'll test it, tidy up and post it on SF tomorrow, its bed for me now though, Gnight...
23:22:09 Quit Bger ("[BX] Reserve your copy of BitchX-1.1-final for the BeOS today!")
23:22:52amiconnThe build table is looking stained these days :/
23:24:41peturamiconn: good sign, it means changes are being made and work is being done ;)
23:24:51amiconnIt's not a good sign
23:25:01peturtoo many targets to test them all, he?
23:25:14amiconnLast week we had an all-green build table
23:25:42amiconnThat means 20 consecutive builds didn't cause a single build error
23:26:16peturbut did you see the amount of commits done recently?
23:27:46 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
23:29:06*t0mas apoligizes to amiconn
23:29:11amiconnpetur: 20 builds are 20 builds
23:29:15t0mas1 line of red+yellow is mine
23:29:50XavierGrI think that's why cvs is for, to have multiple builds in case of a bug.
23:30:11 Join infamis [0] (
23:30:19XavierGrOfcourse developers should be more carefull but whats the deal for some bugs.
23:30:28XavierGrIf they are corrected no harm in the end.
23:30:42XavierGrThough they may confuse others if unsolved.
23:31:18t0maswe should have some testing system
23:31:38t0masa more general one... because I don't have an ondio compiler for example
23:31:44*amiconn always test-builds before committing anything
23:31:52t0masand I can't keep track of all the new color targets added in the last few months
23:32:10 Quit Coolhawk ()
23:32:12t0masamiconn: I try to do that too... but only for player, recorder and some color target
23:32:18t0masusually Iriver...
23:32:32t0masI don't even have a build environment for ipod... or ondio
23:32:46infamisQuestion: iRiver has the 10k max file limit & Rockbox defeats the limit? I read the docs/TECH file & it says because it stores indexes to the files instead of the actual file names....for playlists. Does this mean that the actual _number of files_ on the device is increaesed to?
23:32:47 Quit ender1 (Read error: 104 (Connection reset by peer))
23:32:51muesli__24bit Color BMP Support for WPS!!!
23:32:53muesli__wohoo :D
23:32:54 Join ender` [0] (i=ychat@
23:32:57amiconnHuh? How do you build for player/recorder then?
23:33:18linuxstbWe now also have three different sims - my mistake was only testing on the SDL sim...
23:33:28t0masoh wait...
23:33:31t0masthose are sim builds too
23:33:33linuxstbBut I agree people should test as much as practical before committing.
23:33:39amiconnHehe, in fact that's the sim I completely ignore for now
23:34:21linuxstbBut it could be nice if Bagder published his server scripts so we could run the server build system locally.
23:34:25t0masbut still I think we need some way to test build for all targets... is it possible to do that with normal gcc?
23:34:27amiconnt0mas: If you can compile for player/recorder target, then you can compile for ondio target too
23:34:44t0masamiconn: sure? because then I should setup a build dir for that
23:34:50amiconnOf course
23:34:51t0masgmini is the other one then?
23:35:08amiconnThe gmini port is frozen, so I don't care either
23:35:18 Part Kaggen
23:35:22 Join DreamTactix291 [0] (
23:35:36amiconnOf course I don't always build each and every target and sim before commit, that would take too long
23:36:00t0masI assume it's possible to try to compile all targets with a standard gcc?
23:36:01amiconnI choose the ones to test-build based on what kind of change I made
23:36:11t0masthat way we can spot the warnings and error's easily
23:36:27amiconnNope. You'll need the appropriate cross compiler for the respective target
23:36:57t0masisn't it possible to make a normal gcc process the source? and spit out errors and warnings?
23:37:06amiconnBuilding the cross compiler is no biggie, it just takes quite some time to do it on cygwin
23:37:29 Join Mongey|homework [0] (
23:37:48preglowwhat is missing before the sdl sim matches the other sims?
23:37:51preglowi'm quite eager to be rid of them
23:38:00 Quit Mmmm ()
23:38:03linuxstbpreglow: It doesn't compile on windows, and the sound is buggy
23:38:14amiconnWell, if the SDL sim looks like the X11 one, I'll never drop win32
23:38:38linuxstbI was thinking about that - it could have been better to base the SDL sim on win32, rather than x11.
23:39:06 Quit TCK (Read error: 110 (Connection timed out))
23:39:14preglowamiconn: looks like? what do you mean?
23:39:18amiconnWhile others may disagree, I think the picture of the unit is often useful
23:39:38linuxstbThat should be relatively easy to add to the SDL sim now though.
23:39:39preglowyou mean the bloody bitmap?
23:39:43preglowi couldn't care less about that
23:39:49preglowbut sure, if it's wanted
23:40:21preglowshould be, yes
23:40:25preglowwhy doesn't it compile on windows?
23:40:29preglowshould be a really trivial thing to fix
23:40:46 Join RedLeg [0] (n=red@
23:41:14preglowlinuxstb: sounds is buggy on the other sims as well
23:41:20amiconnSelected sdl simulator
23:41:20amiconnconfigure didn't find sdl-config, which indicates that you
23:41:20amiconndon't have SDL (properly) installed. Please correct and
23:41:20DBUGEnqueued KICK amiconn
23:41:20amiconnre-run configure!
23:41:34RedLegI'm looking for help with what seems to be a hardware problem.... can anyone help?
23:41:42amiconnThe win32 sim doesn't require extra software
23:41:44preglowusing sdl-config is really unecessary
23:42:06linuxstbpreglow: The sound playback just stops after a couple of seconds for me. It worked quite well under x11.
23:42:11preglowall you need is -lSDL
23:42:16peturpreglow: I once posted a small patch for the sound of the w32 sim that made it play ok
23:42:18preglowi could never get x11 sound to work
23:42:24preglowi guess it relies on oss
23:42:39linuxstbOr it makes assumptions about your soundcard.
23:42:48linuxstbBut yes, I think it was oss-based.
23:42:53amiconnlinuxstb: Sound in sims is fishy in general. It works for me in win32, but stutters a bit from time to time. The X11 sim plays for half a second, then stops
23:43:02preglowlinuxstb: sdl sound works fine here, the first time... after you stop audio, it'll never work properly for more than a few seconds
23:43:12RedLegAnyone ever seen "ATA: -11" as an error message immediately at poweron?
23:43:24preglowbut anyway
23:43:36amiconnI remember that it worked in X11 once, but can't remember when it brooke
23:43:39preglowthis is a golden opportunity to have a single codebase for sims, we should make it work
23:43:54preglowSDL is a really lovely api compared to both winapi and x api
23:44:27amiconnThe win32 api is rather well documented
23:44:31 Join Bagder [0] (
23:44:37preglowyes, but hell to use
23:44:42preglowsdl api is also well documented
23:44:45preglowand not hell to use
23:45:08amiconnImho x11 really is hell. Win32 is okay
23:45:33preglowif you think it's ok, i bet you really haven't used it yet :P
23:45:34amiconnDunno about SDL yet
23:45:38Rob2222Dont buy a cheap ipod battery for an iRiver H300. It's almost wasted money. :(
23:45:40preglowx11 is a greater hell, yes, agreed
23:45:55preglowand that doesn't even do everything winapi does
23:47:17Rob2222You get 10,5h out of the stock 1300mA battery and I get now 12,5h (exact same settings) out of my "2200"mA battery. :(
23:47:50Rob2222preglow: You mean me?
23:47:51muesli__only 12h? which codec?
23:48:07Rob2222Mp3 @ 128 ABR
23:48:10Bagderooo bmp build
23:48:13Rob2222Volume -45dB
23:48:26muesli__once i had 29h
23:48:33muesli__with a 2200mah
23:48:40muesli__iriver lab mode
23:48:54preglowRob2222: yeah, you
23:48:57Rob2222my 2200mA only costs 9,99 Euro, maybe its relabeled ...
23:48:59CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:49:04muesli__it was the same
23:49:05preglowonly gaining 2 hours with that setup is realle unlikely
23:49:08Rob2222the thing is, its not my unit ...
23:49:18Rob2222let me tell...
23:49:30preglow10.5 hours from standard battery with that setup is also very poor
23:49:33preglowthis is h3x0, yeah?
23:49:38Rob2222my old (1 year in use) stock battery gets 9,5h.
23:49:54Rob2222so i tool the H320 of my friend thats 1 weeks old
23:49:58muesli__preglow h3xx contains a 1700mah
23:50:05Rob2222anf from his unit i got 10,5 h
23:50:23Rob2222udes same rockbox, playlost and same mp3s
23:50:36Rob2222dam what am i typing
23:50:44Rob2222muesli__: sure?
23:51:01muesli__what is sure
23:51:17infamisdoes rockbox have a 10k limit on the number of files able to be on the player for the iriver port?
23:51:17Rob2222stock: LPCS 355385/KPPJDLJI
23:51:20infamis I know it doesn't for the playlist but does it for the actual number of all files on the hard drive?
23:51:21XavierGrNo, I think not. H300 comes with the same battery as H100 AFAIC.
23:51:30preglowinfamis: no
23:51:39muesli__XavierGr NO
23:51:46muesli__il have a look
23:51:51XavierGrProof link?
23:52:13Rob2222LPCS355385/KPPJDLJI <= on my stock battery
23:52:29RedLegAnyone ever seen "ATA: -11" as an error message immediately at poweron?
23:52:42RedLegHardware is a Jukebox 20 recorder
23:53:03infamisdoes rockbox load all files' info on the hard drive into memory on startup? or what file in firmware/ should I be looking for?
23:53:16Bagderinfamis: no it doesn't
23:53:32Rob2222http:// <= here it is. the stock battery
23:53:46Rob22221350 average 1400mA notes on the stock battery..but i am sure if read somewhere 1700mah
23:54:30Rob2222muesli__: I gave you the number of my H340 stock battery
23:54:33infamisso startup is practically instant since it doesn't read the whole hdd?
23:54:43XavierGrso it is rated at 1400mah?
23:54:48Bagderinfamis: you never started it?
23:54:51t0masinfamis: it does the dircache in the background
23:54:54Rob22221400max, 1350 average
23:55:13infamisI have an iaudio x5, but it has the 10k limit also, so I'm trying to see what you guys actually did to defeat it
23:55:17infamisno iriver
23:55:17Rob2222but my 2200er is bullshit :(
23:55:23linuxstbOK, build table green again....
23:55:31Bagderinfamis: we have no such silly limit
23:55:37XavierGrDoesn't H300 has a sticker with the battery rating?
23:55:37Moos infamis: start take around 7sec here on h140
23:55:48XavierGrH100 has one and says 1300 mah
23:55:53Bagderinfamis: we do have a configurable max number files in a playlist and in a dir
23:55:56muesli__XavierGr no
23:56:03muesli__i thought there was one
23:56:07muesli__but i was wrong
23:56:37muesli__and there battery is lying just in front of me
23:56:44XavierGrFuck, that meas that now that I will upgrade my H300 will have far lowerbattery time than H100.
23:56:48muesli__nothin to see
23:57:01XavierGrI should have rethink for that
23:57:03infamisBagder: but no limit really on the actual number of files you can have on the hard drive, correct?
23:57:10Bagderinfamis: correct
23:57:23amiconnXavierGr: That drawback will only be temporary, I hope
23:57:49Bagderinfamis: we support 120GB disks on the Achos players... people are likely to have much more than 10K files on those
23:58:18infamisright, that's when I bought my X5 60gb & found out the 10k limit, I was astounded!
23:58:24Rob2222XavierGr: With a good 2200 you should get at least 17h ^^ :)
23:58:40infamisI had bought the 60gb just cause my rio karma was only a 20gb
23:58:40XavierGramiconn: Even if you cacth iriver in battery times, the internal battery is rated at 1400mah, (and iriver sasy 16 hours for both H100 and H300) (H100 has 1300mah internal) that means that H300 needs more battery.
23:58:45*amiconn wonders how this sdl stuff is supposed to work

Previous day | Next day