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

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

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

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

#rockbox log for 2007-08-13

00:03:45ddaltonI found the line snprintf(buf, 128, str(LANG_FM_STATION), freq / 100, freq % 100); Does that sound right. So now how do I find how the actual station frequency is displayed?
00:04:35bluebrotherthe frequency is in the variable freq.
00:06:13 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
00:06:45bluebrotherpixelma: tucoz noted that there is a description in the wiki on how to convert the svg images.
00:07:32pixelmathanks, will take a look - not sure yet if it's feasable with cygwin though...
00:07:53 Quit My_Sic (Read error: 113 (No route to host))
00:07:55 Join My_Sic [0] (
00:08:16DerPapstcould the svg be used for the sims too? i mean as background image
00:08:27 Quit davina ("xchat on Ubuntu 7.04")
00:08:28bluebrotherhmm, interesting idea
00:08:34 Quit ctaylorr (Read error: 110 (Connection timed out))
00:08:44bluebrotherpixelma: I could also look into it tomorrow
00:09:11*bluebrother laughs at the ccc frontpage ...
00:09:45pixelmathey would need some extra work if they wanted to be used in the sim (like a few side views maybe)
00:10:05pixelmaerm... if someone wants to use them for the sim
00:10:14DerPapstyeah... they have to be drawn..
00:10:25*pixelma obviously not completely home yet
00:11:25bluebrotherhmm, really? The few buttons that are on the side are named in the manual, so it should be possible for the sim as well
00:11:25pixelmaeven though svgs of the iriver and iaudio remotes are done (but not used yet)
00:11:26 Quit barrywardell ()
00:14:03 Join barrywardell [0] (
00:15:28pixelmayeah maybe that'll do (I drew them so that the buttons are visible even though they might not be in reality - to save a side view)
00:17:13bluebrotherlooks also somewhat more "natural" than an attached side view
00:18:20pixelmaI'll try if I can figure converting the svg out myself - if not, I'll ask again tomorrow ;)
00:18:22amiconnAn idea for the Iriver sims: simulate changing remotes by changing the image as well =)
00:18:58pixelmayou can do that for all: attaching and detaching...
00:19:25amiconnAll targets with remotes available, yes
00:19:29 Quit JdGordon ("Konversation terminated!")
00:20:58peturdamn... JdGordon just left :(
00:21:07 Quit Rondom ("Ex-Chat")
00:21:25 Quit robin_0800 (Read error: 104 (Connection reset by peer))
00:28:35 Join Robin0800 [0] (
00:28:52 Quit Entasis_ (Read error: 104 (Connection reset by peer))
00:28:52 Quit annulus_ (SendQ exceeded)
00:29:21 Quit Guile`` ("KVIrc 3.2.0 'Realia'")
00:30:17 Part n1s
00:31:55 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
00:37:59pixelmahehe, bluebrother - interesting ccc frontpage indeed :)
00:39:50 Quit Robin0800 (" HydraIRC -> <- \o/")
00:41:23 Quit My_Sic (Read error: 104 (Connection reset by peer))
00:41:33 Join My_Sic [0] (
00:45:32 Quit petur ("Zzzzz")
00:47:58 Quit spiorf (Remote closed the connection)
00:53:06 Quit Wofl (Remote closed the connection)
00:53:21 Join vmx [0] (
00:53:35 Quit Nibbier (Remote closed the connection)
00:55:17vmxhi, i try the rockbox simulator for sansa e200 atm. i built it, but the "wheel" doesn't work in the main menu (8 and 2 key). what have i done wrong? (current svn, linux, amd64)
00:55:35bluebrotherdid you use the num pad?
00:56:00vmxchangiong the volume with the same keys works
00:56:20 Join saratoga [0] (i=98039a34@gateway/web/cgi-irc/
00:56:25bluebrotherhmm. No idea −− perhaps some key bindings for the sim are missing.
00:56:33bluebrothertried the cursor keys?
00:56:55 Quit miepchen^schlaf ("Verlassend")
00:57:13vmxyes, the same problem. changing volume works, up/down in menus not
01:00:53DerPapsthmm... wasn'T there a wheel related commit a few days ago?
01:01:22*bluebrother goes building
01:02:21*DerPapst remembers that the wheel worked fine in the sim ~1 month ago
01:02:22 Quit ender` (" I spilled Spot Remover on my dog... Now he's gone.")
01:07:49bluebrotherhmm, wheel works fine here
01:07:57bluebrotherlinux, ia32
01:08:34vmxhm, i really need a 32 bit chroot
01:08:38bluebrotherPreviews for theme installation are working :)
01:08:55bluebrotherwell, if it's a 64bit related problem it should get fixed ...
01:09:52bluebrotherhmm, either I haven't understood it fully or QHttp::readAll() is a bit strange.
01:09:54vmxwith a 32bit chroot we could be sure that it is an 64bit issue
01:09:55 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
01:10:25*bluebrother has no experience with 32bit chroot
01:10:34 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
01:11:14vmxif i've some time left i'll take look. g'night
01:11:44 Quit vmx ("Have fun!")
01:12:25 Join webguest65 [0] (i=46f69b7a@gateway/web/cgi-irc/
01:12:28 Quit webguest65 (Client Quit)
01:18:42 Join advcomp2019- [0] (n=advcomp2@
01:20:09 Quit advcomp2019 (Nick collision from services.)
01:20:11 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
01:22:21 Quit bdgraue (Remote closed the connection)
01:23:14 Join nerochia1o [0] (
01:23:28DerPapstNight all
01:23:28 Join alien_ [0] (
01:23:30 Join BjoernErik [0] (
01:23:32 Quit DerPapst ("So Long And Thanks For All The Fish!")
01:24:24 Quit nerochiaro (Read error: 104 (Connection reset by peer))
01:24:48 Quit majeru (Read error: 104 (Connection reset by peer))
01:25:44 Join hannesd_ [0] (
01:34:36 Quit bluebrother ("zzz ...")
01:35:44 Quit Bjoern-Erik (Connection timed out)
01:40:12 Nick BjoernErik is now known as Bjoern-Erik (
01:40:44 Quit hannesd (Read error: 110 (Connection timed out))
01:40:45 Nick hannesd_ is now known as hannesd (
01:43:56saratogahas anyone seen an arm optimzied FFT implementation somewhere?
01:44:04saratogait seems like one must exist
01:44:10saratogagiven how popular FFTs and ARM processors each are
01:54:08 Quit jhulst (Remote closed the connection)
01:54:26Nico_Pamiconn: I'm reading some talk about scrolling code and viewports in the log from the 11th march... still not motivated ?
01:58:33***Saving seen data "./dancer.seen"
02:01:44 Part pixelma
02:15:49 Quit RaRe` (Read error: 110 (Connection timed out))
02:16:23 Join RaRe` [0] (
02:16:51 Join donutman25 [0] (n=chatzill@
02:22:04 Join jhulst [0] (
02:23:50rasherbarrywardell: About the `which gmake` thing. Wouldn't it make sense to use if [ -f `which gmake` ]? (and I should change this a bunch of places in the voice scripts also)
02:24:11rasherif [ -f "`which gmake`" ] actually
02:24:32 Join Wofl [0] (
02:24:38 Quit Wofl (Remote closed the connection)
02:26:34 Quit inversion (Read error: 110 (Connection timed out))
02:28:12 Quit Nico_P (Remote closed the connection)
02:28:18 Join Wofl [0] (
02:30:37 Join ctaylorr [0] (
02:32:27barrywardellrasher: sounds reasonable. I'll test that.
02:33:16rasherI really can't imagine why it wouldn't work. It seems the right way to check. Looks like which is not a POSIX mandated program by the way
02:35:37barrywardellyes, it should work. but OS X often behaves in weird ways
02:38:06rasherI wonder what the posix way to find an executable is
02:42:48barrywardellwell, that works just as expected now. thanks for pointing it out
02:43:28rasherAnd now we can remove Lenny Koepsell from the CREDITS file!
02:43:38rasher(not serious)
02:43:59scorchetrim the branches? =P
03:00:19 Quit jhulst (Remote closed the connection)
03:01:41saratogai just noticed the ffmpeg people are looking to create a standard fixed point MDCT library for their project
03:01:46saratogathat could be VERY useful for us
03:01:53saratogaparticularly for AAC and Tremor
03:13:45 Join elinenbe [0] (
03:26:32 Quit donutman25 ("ChatZilla [Firefox]")
03:29:59 Quit barrywardell ()
03:38:38 Quit reeyal ("CGI:IRC (EOF)")
03:58:37***Saving seen data "./dancer.seen"
04:03:17 Join tihoc4n [0] (
04:07:58 Quit Thundercloud (Remote closed the connection)
04:13:45 Join x2jmp [0] (
04:14:58 Quit tihoc4n ("Leaving.")
04:30:19 Quit x1jmp (Read error: 110 (Connection timed out))
04:36:14 Join Dr_nuby [0] (
04:36:44Dr_nubyany plans for firm ware for creative zen v plus?
04:37:19 Quit Dr_nuby (Client Quit)
04:37:23DogBoyinstalled to a sansa 8 gig I just bought
04:37:32DogBoythanks everybody
04:38:48 Quit x2jmp (Remote closed the connection)
04:40:44tchanIs it a known problem that upon updating a rockbox install from let's say 1.5 months ago with rockbox-svn to have the database wiped out and need to re-initialize it ?
04:41:11 Join webguest63 [0] (i=cfbfd7d3@gateway/web/cgi-irc/
04:41:57tchanthis happened on both my gigabeat-f and ipod-4g
04:43:44 Join J-_ [0] (n=justin@unaffiliated/j/x-388422)
04:44:04J-_Any headway on ZVM's yet?
04:44:31scorchethere is the forum thread for a reason..
04:47:00J-_where's the thread!?
04:47:26scorchesearch the forums and find out
04:55:27saratogaNew Ports
04:55:36saratogai don't think anyones seriously working on it though
04:55:41saratogajust a couple people poking around
04:59:00J-_How easy is it to replace a ZEN battery?
04:59:26scorcheplease read the link in the topic
05:00:08J-_someone piss in your coffee?
05:00:28scorchethis channel has guidelines, and they are to be followed...
05:00:42 Join TinoM| [0] (
05:01:08scorchewhat is the point of having guidelines and not enforcing them>
05:01:08saratogayeah we don't really talk about mp3 players in general
05:01:20saratogaso don't do that here
05:01:47scorcheour reasoning for having guidelines at all is written out in that link as well
05:02:00J-_So rendering a FCC compliment device is on topic I guess?
05:02:23scorchei am not quite getting you...
05:02:44J-_I guess not
05:04:50alienbiker99that didnt make sense
05:06:26scorcherendering can refer to a multitude of different things...
05:06:32alienbiker99rendering it ______
05:07:17scorcheregardless of what type of rendering you are talking about, unless it related to rockbox, yes it would be off-topic
05:11:09 Join authumalk [0] (
05:11:23authumalkanything cool happening with rockbox?
05:11:31J-_the first question was a yes or no answer, I didn't want to read a forum. I don't know any technical aspects of, "rockbox". but i guess if you're gonna be fucking lame, I won't use it, yes I understand it's completely voluntary work, but COME ON. Instead of being a prick, be nice for a fucking change.
05:11:39authumalkother than whats in the changelog?
05:12:00scorcheJ-_: have you read those guidelines yet?...they also talk about language
05:12:07scorcheauthumalk: there is the MajorChanges wiki page
05:12:51scorcheJ-_: i am very nice...nice doesnt mean that i let people walk all over the guidelines like they dont exist though
05:13:02alienbiker99pretty muhc you wont know the progress in thsi channel, maybe theres a zen cahnnel on the server. to know about the hardware, go to a ZVM forum, people here dont talk about hardware much
05:13:31saratogaJ-_: we really have no idea about the Zen
05:13:38saratogai mean, it has nothing to do with this channel
05:13:40scorchethe ZVM people dont come around here much, but they post in that thread, so that is why i told you to look there
05:14:28authumalkJ-_: you sir are a douchebag, if you want rockbox for your zen so bad go to the library and get coding for stupid fucktards who think the world revolves around them
05:14:48scorcheauthumalk: the guidelines apply to you as well...please read and honor them
05:14:55saratogai like how the developers end up being nicer then the users
05:14:55alienbiker99eh you dont need to be rude to him
05:15:02J-_authumalk: heh
05:15:04alienbiker99@ authumalk
05:15:43scorchesaratoga: i have never gotten the thing about not inserting emotion means we arent nice or are rude
05:15:55scorchethen again, that subject has been beaten to death by many..
05:17:01saratogayeah i hate smilies
05:17:11saratogaso if thats being rude, i'm fine with it
05:17:43scorchei use "=P" on lines where a hint is needed to potentially diffuse things...sarcasm can tend not to translate well without them
05:18:28authumalkI Guess I should have read this: "" * Before requesting help, make sure that you have read the FAQ and the manual. The next link is not required reading, but it does re-iterate our guidelines and is a very handy thing to have read; people will be much more willing to help you if you follow its advice .
05:18:28authumalk * Keep discussion Rockbox-related. Unsupported builds are classified as off-topic as well. If you wish to have off-topic chat, please go to #rockbox-community.
05:18:28authumalk * Demonstrate common courtesy to people. This includes such items as being polite to people, refraining from using vulgar language, treating people humanely, among other similar actions.
05:18:30authumalk * Use clear, grammatical, correctly-spelled english. "Words" such as "thx", "u", and "kewl" are not acceptable. This includes l337speak and odd usages of capital letters as well.
05:18:31 Quit authumalk (Excess Flood)
05:18:35 Quit TinoM (Read error: 110 (Connection timed out))
05:18:51 Join authumalk [0] (
05:18:59saratogai'd never actually looked at that, but they do seem like a pretty good idea
05:19:08scorcheauthumalk: read on a bit more, and you will see our guideline against long pastes =)
05:19:09authumalkmy point exactly
05:19:22authumalki kinda realized that
05:19:29authumalkman i'm a fag
05:19:40*scorche sighs
05:20:02*alienbiker99 hands scorche a beer
05:22:48 Quit Rick (Read error: 104 (Connection reset by peer))
05:23:36 Join Rick [0] (
05:25:22 Part J-_ ("thanks for the info saratoga")
05:35:48 Quit ptw419 ()
05:42:35 Quit Wofl (Read error: 104 (Connection reset by peer))
05:44:42 Join Wofl [0] (
05:45:29 Quit webguest63 ("CGI:IRC (EOF)")
05:52:33 Quit authumalk (" HydraIRC -> <-")
05:58:38***Saving seen data "./dancer.seen"
06:13:27 Join barrywardell [0] (
06:16:41 Quit BigMac ("Leaving")
06:16:58 Join BigMac [0] (
06:17:07 Join datasleep [0] (
06:17:37 Quit datachild (Nick collision from services.)
06:33:37 Quit barrywardell (Read error: 110 (Connection timed out))
06:41:24 Quit lazka (Remote closed the connection)
06:42:39 Join jhulst [0] (
06:45:37 Join datachild [0] (
06:45:46 Join webguest72 [0] (i=c8ccb0b5@gateway/web/cgi-irc/
06:46:13 Quit hcs (Read error: 110 (Connection timed out))
06:51:36 Join mael [0] (i=4b98c5de@gateway/web/cgi-irc/
06:51:50 Quit saratoga ("CGI:IRC (Ping timeout)")
06:53:22 Quit webguest72 ("CGI:IRC (EOF)")
06:54:06maelHello, question regarding rockbox on my ipod 5th gen - after playing a song for a few seconds while on hold, then releasing hold and pressing the menu button nothing happens unless I hit the button multiple times, any ideas why?
07:01:40 Quit datasleep (Read error: 110 (Connection timed out))
07:09:46 Quit perrikwp (Read error: 104 (Connection reset by peer))
07:10:12 Quit blithe (Remote closed the connection)
07:10:23 Join blithe [0] (
07:11:27 Quit sarixe ("Peace")
07:15:23 Quit jhMikeS (Nick collision from services.)
07:15:29 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
07:16:15 Part toffe82
07:18:51 Join perrikwp [0] (n=chatzill@
07:24:28 Join BobShiel1 [0] (
07:24:28 Quit BobShield (Read error: 104 (Connection reset by peer))
07:28:45amiconnjhMikeS: Is there a special reason for the sysevent "classes"?
07:33:06alienbiker99sansa e250 is on sale for $50 on woot it is
07:34:00alienbiker99i kinda wish i had money to get it
07:34:13 Quit Wofl (Remote closed the connection)
07:34:23alienbiker99maybe if its there in the morning i will buy it
07:34:31scorchei doubt it will last long
07:34:47 Join Wofl [0] (
07:35:17alienbiker99this is very tempting
07:35:19scorchethat is half price, basically
07:35:31 Quit jhulst (Remote closed the connection)
07:36:03scorchei kind of wish you hadnt told me...
07:36:25scorchealthough i would have checked woot later anyway..
07:36:43 Quit Wofl (Remote closed the connection)
07:36:57alienbiker99hahaha, its on my google homepage so i look at it everytime i click the home button
07:38:24alienbiker99hmm i should pick one up for my dad
07:40:07jhMikeSamiconn: one moment...brb
07:47:30alienbiker99hmm i wonder if the insignia has the same chips as the sansa
07:48:23jhMikeSamiconn: no special reason atm other than to keep sys bits most significant and the particular message code least though wanting to watch for a particular event range could be more efficient that way. there's no binsize penalty that came up.
07:50:19*jhMikeS must say though that a cache tester plugin shows odd behavior from the flush/invalidate functions on e200.
07:51:51Tempishow up to date is the Sansa info on the rockbox site?
07:54:58 Quit XavierGr (Read error: 110 (Connection timed out))
07:55:53DogBoyI just used it to install rockbox Tempis
07:56:01DogBoyhow up to date does it need to be?
07:56:16TempisI don't mean rockbox itself
07:56:50TempisI was reading the faq and I was just wondering if it still has the problems they list there
07:56:55 Join ddalton [0] (
07:57:07amiconnjhMikeS: The sysevent classes seem a bit arbitrary to me. Other than that, I think it's enough to indicate a sysevent by the msb being set
07:57:23amiconnIt's not terribly important, however, I need to add some. Hence my question
07:57:42ddaltonwhere is the best place for a function? at the bottom of the file?
07:58:08advcomp2019did the rockbox site die
07:58:15TempisI'm on it right now
07:58:30amiconnjhMikeS: Btw, the way the pcmrec thread handles contains a potential race condition
07:58:37amiconn*handles usb
07:58:39***Saving seen data "./dancer.seen"
07:58:44advcomp2019it is only the forums that is not working right
07:58:49Tempisjust an fyi, woot has the sansa e250's for $50
07:59:06Tempisseems pretty good to replace my apparently dead archos
07:59:10amiconnjhMikeS: Check lines 1632ff in pcm_record.c
07:59:16Tempisstupid solder points
07:59:48alienbiker99ifi bought this, it would be the 2nd mp3 player ive bought from woot =\
07:59:52scorcheadvcomp2019: it is just slow
08:00:08Tempisare there any screenshots of the sansa version of rockbox?
08:00:33advcomp2019i just got "Sorry, SMF was unable to connect to the database. This may be caused by the server being busy. Please try again later."
08:01:01amiconnjhMikeS: When receiving the usb connected event, it does some stuff, then flushes the queue, then acks and waits for disconnect. However, the flush might have flushed an usb disconnect event in 2 situations:
08:01:33Tempisalienbiker99, you can't beat that for 50 bucks
08:01:49alienbiker99i know, i need to get $50 first haha
08:01:49Tempisand 2gig microsd cards are only like 20
08:01:53amiconn(1) if either pcmrec_close() or reset_hardware() yield - didn't check that. (2) if the disconnect event was already in the queue as well when the thread got a chance to run
08:02:11Tempisnow the 4gig microsd cards need to come down in price
08:02:30alienbiker99idk if it supports 4gb microsd cards
08:02:53Tempiswell that would suck
08:03:24advcomp2019it does not support 4gb or higher yet
08:03:25scorcheTempis: what good would screenshots do?...rockbox is rockbox, generally
08:03:50TempisI'm just curious as to what the ui looks like is all
08:03:58Tempisit's not terribly important
08:04:05alienbiker99tempis check your pm
08:04:10TempisI'm going to put it on there no matter what
08:04:46scorcheTempis: have you ever run rockbox on any other target?
08:05:00Tempisyeah, an archos
08:05:07TempisStudio and Recorder
08:05:12Tempisboth now dead
08:06:07jhMikeSamiconn: hmmm....probably should check that more closely
08:06:22alienbiker99i think i might buy it to have a flash player
08:06:47amiconnIt's not very likely, because it would require these events to appear in quick succession
08:07:20amiconnDetection is delayed for debouncing, so it might in fact never happen
08:09:20jhMikeSit can never be in a recording state in the first place so the thread would be free to respond quickly
08:12:41jhMikeSI guess the usb_wait_for_disconnect function will do the clearing job anyway :)
08:15:40 Join ender` [0] (
08:16:19amiconnNo, it doesn't
08:16:41amiconnIt checks every single event whether it's a disconnect event. It doesn't use queue_clear()
08:17:04amiconnHmm, or maybe that's what you mean....
08:19:24jhMikeSbasically, but there is a problem with a thread blocked waiting for reply. it doesn't call queue_reply.
08:20:30 Join GodEater [0] (n=bryan@rockbox/staff/GodEater)
08:21:51jhMikeSqueues will auto reply in the core kernel update though so queue_reply would only be needed if a return value other than 0 is desired.
08:22:24*amiconn still doesn't really understand what this synchronous queue mechanism is for and how it works
08:23:53amiconnI mean, if 2 threads have to communicate and their actions must be serialized, why not just send a message from thread 1 to thread 2, and then wait for thread 2 to send another message back?
08:23:56jhMikeSto hold a thread for a reply to a specific message
08:24:04jhMikeSthat's what it does
08:24:11amiconnThread 1 could continue processing other messages in the meantime, not blocking it
08:24:53jhMikeShow would I write a message loop for a thread that has unknown function?
08:25:00amiconnBlocking the thread until that specific message is replied to might cause processing of other messages to stall
08:25:48amiconn...and processing of system-wide messages must be relied upon for usb (storage) connection atm, and, hopefully soon, for poweroff/suspend
08:25:51jhMikeSprecisely the idea
08:26:02 Join Rob222241 [0] (
08:26:28jhMikeSit's written to reply as soon as the state change requested is official and any data sent is safely transfer
08:27:01 Quit Siku ()
08:27:20jhMikeSsome messages there could probabaly just be posts, others must block the source thread
08:27:40amiconnjhMikeS: Yes, but using 2 separate messages would allow the requesting thread to react to other messages, which might be important
08:28:04jhMikeSwhere? sent to whom?
08:28:08amiconnImagine thread #1 sending a message to thread #2, hence being blocked.
08:28:41amiconnNow the user plugs USB, SYS_USB_CONNECTED being broadcast to all public queues
08:29:09amiconnRockbox won't go into usb mode until all those threads acknowledged the connection - but thread #1 cannot ack because it's blocked...
08:30:16jhMikeS1) recording won't be happening 2) if recording isn't happening the reply will be almost instant and the thread won't be blocked for but milliseconds at best
08:31:32amiconnI'm not referring to recording, but to the mechanism in general
08:32:44amiconnThis scenario will work if thread #2 ensures to releases the sender if it receives such a public event - but that's only possible if its own queue is also public...
08:33:07jhMikeSif recordin was blasted with messages to process, it will process them all before the USB connect is acked anyway so that thread alone will hold the connection up
08:33:17 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
08:34:10jhMikeSpublic events are posted and do not block any thread
08:34:58amiconnNo but a thread blocked by sending a synchronous message before won't be able to acknowledge a public message
08:35:02jhMikeSthe sender waits for a reply on the message pulled from a particular slot
08:35:08amiconn...until it is unblocked
08:35:42jhMikeSneither will recording ack the event until it gets it and the final acks will be delayed exactly the same
08:36:43jhMikeSthe thread that posted a message to recording may ack since it was never blocked, but recording won't ack until it processes the posted message first
08:38:16amiconnA thread that *sends* a message to recording will be blocked until recording replies, correct?
08:38:28jhMikeSyes, of course
08:38:50amiconnAnd that thread can't ack another message until recording replies
08:39:15jhMikeSand recording couldn't even receive the SYS_USB_CONNECTED itself any sooner than that
08:39:24amiconnUh, why?
08:39:39jhMikeSnot till it waits on it's queue again
08:39:44markunamiconn: are you working on ipod 2nd gen support?
08:40:02amiconnmarkun: Of course, why?
08:40:17 Join datasleep [0] (
08:40:34 Quit datachild (Nick collision from services.)
08:40:36 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:40:37amiconnjhMikeS: hmm...
08:40:42markunamiconn: sorry, I wanted to paste a link to the post, but I guess you can find it
08:41:09GodEaternot with the forums in their usual unresponsive state at this time of day
08:41:47markunGodEater: they also stopped working for me now :(
08:42:11GodEaterah - they've woken up again
08:42:31GodEater <−− that message I'm guessing
08:43:30amiconn2gig second gen???
08:43:36amiconnThat's nonsense
08:43:53amiconnThe 2nd gen came in 5, 10 and 20GB
08:43:57 Quit Rob2222 (Read error: 110 (Connection timed out))
08:44:14GodEaterI'm guessing he's talking about a Nano in the thread then
08:44:27GodEaterwhich comes in 2, 4 and 8 gig
08:44:42jhMikeSin fact, thread #1 can't go ack until pcmrec is ready to yield anyway
08:45:15amiconnjhMikeS: So, does this synchronous mechanism require the receiving thread to reply asap and never yield inbetween?
08:45:31amiconnI guess not
08:45:58jhMikeSno, it can yield all it wants in between and securely receive data from the posting thread
08:46:42 Join Gibbed [0] (
08:46:46 Quit Rick (Nick collision from services.)
08:47:00 Nick Gibbed is now known as Rick (
08:48:23 Quit mael ("CGI:IRC (Ping timeout)")
08:48:27jhMikeSPCMREC_STOP could simply be posted. in fact it really should be, no reason to block that one.
08:49:46jhMikeSsame goes for PAUSE/RESUME
08:55:03 Join B4gder [0] (
08:55:05 Join davina [0] (
08:55:19 Join desowin [0] (n=desowin@
08:55:25ddaltonhow can I check in my code what button was pressed. For example up on the iriver.
08:55:28jhMikeSamiconn: if usb_wait_for_disconnect calls queue_reply(0) in the loop all should be fine and that race condition gone and that should work as I intended in the first place.
08:57:40ddaltonhow do I voice a decimal number? for example 87.5
09:01:30*jhMikeS already wanted to voice demicals correctly but that's really not implemented for numbers in general
09:02:14amiconnWell, since we don't use float/double, you can't do that directly anyway
09:02:47amiconnBut output_dyn_value does voice decimals
09:04:05jhMikeSiirc the function is over-specialized and was of no help in trying to voice decimals in the gigabeat's tone controls.
09:05:12 Join petur [0] (n=petur@rockbox/developer/petur)
09:05:17ddaltonWhat arguments does output_dyn_value require?
09:05:20amiconnThat function is specialised for outputting high range values with prefixes
09:05:43amiconnThere were no other places where decimals were neccary back when coding it
09:05:44ddaltonbut what infomation do I need to give it.
09:06:03ddaltondo I have to tell it what to voice
09:06:33amiconnIt can be used either for printing or for voicing. If you want both, you need to call it twice
09:06:47jhMikeSpersonally, I'd just replace the darn thing with a generic number voicing/formatting function
09:07:08ddaltonto voice a variable that is 2.1 how would I tell it to say it. and the variable is called i
09:07:40amiconnIn rockbox we don't use floating point variables, so no variable can have a value of 2.1
09:07:53ddaltonwhat about the radio?
09:08:05amiconnThe radio uses plain Hz internally
09:08:27amiconnSo it doesn't store 89.7MHz, it stores 89700000
09:08:36ddaltonso how would I speak that. I got it to voice the station but it didn't say the decimal
09:11:10ddaltonso how does it display 89700000
09:11:18ddaltonas 89.7
09:12:30jhMikeSsomething like snprintf(buf, len, "%s.%s", freq/1000000, freq/100000 %10)
09:13:12B4gderbut %d and not %s ;-)
09:13:20jhMikeS:P hehe
09:15:13ddaltonwhat is len?
09:15:44 Join webguest52 [0] (i=c023111d@gateway/web/cgi-irc/
09:15:50GodEaterlen is the size of buf
09:16:03ddaltonso then how would I voice this?
09:16:15*petur feels it would be more time-efficient if somebody were to write this for him :/
09:16:39jhMikeS )
09:16:57*GodEater has never done any work with the voice code and is therefore unable to answer
09:17:20 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
09:17:26 Join ddalton [0] (
09:18:10ddaltonok so I will then just try a couple of things.
09:18:20jhMikeSoutput_dyn_value really is just not a good function for having control over numbers with decimals. it's really geared towards file sizes and things like it.
09:18:24ddaltondoes anyone know how to check what button has being pressed?
09:21:06jhMikeShave a look at firmware/drivers/button.c and it should give you all that info see apps/action.c too
09:21:42ddaltonI have found the line: output_dyn_value(NULL, 0, free, kbyte_units, true); what does free do?
09:23:30peturddalton: I think you need to examine more code and try things out. Programming is way more than knowing the language.
09:23:58ddaltonso you can't tell me what free does in that function?
09:24:01 Join sarixe [0] (
09:25:14peturddalton: because I would have to look it up myself too
09:25:28 Quit sarixe (Client Quit)
09:25:31scorcheddalton: sorry if this is too blunt, but we cant always be around here to answer every single question you need to find things for yourself
09:26:11 Quit webguest52 ("CGI:IRC")
09:26:15 Join webguest52 [0] (i=c023111d@gateway/web/cgi-irc/
09:28:15 Join aliask [0] (
09:29:30 Quit dan_a (Remote closed the connection)
09:31:10 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
09:31:13 Join sergey [0] (
09:31:15 Join ddalton [0] (
09:32:14 Join Siku [0] (
09:34:11 Join fm2 [0] (i=c27f0814@gateway/web/cgi-irc/
09:34:19fm2GodEater: ping
09:35:35fm2GodEater: have you seen my questions/suggestions about the shortcuts plugin (on Fri and Sun)?
09:35:44GodEaterno - where did you put them ?
09:36:11jhMikeSmarkun: where's the info available for the gigabeat mmu (the mcr/mrc stuff)?
09:36:19fm2GodEater: on the irc
09:36:37fm2But you were not there at that moment
09:38:58GodEaterfm2: I'm terrible at reading the logs
09:39:07GodEateryou'd be better off posting them to the forum thread
09:39:43fm2Or should I repost them now?
09:40:20GodEaternothing stopping you...
09:41:54fm2GodEater: 1) Why do you call sc_init() (and hence load the default file) even if the plugin was called with a parameter, i.e. via the context menu?
09:42:12fm2(2) Why are there two different funcs for loading the default and a custom file? A copy-paste rest?
09:42:43fm2(3) In the plugin, the extension ".link" is hard coded. OTOH we want the association of file exts to viewers be configurable in the config file.
09:42:43GodEaterthe answer to both is the same: I didn't want it to do custom files when I started writing it
09:42:53fm2This is a contradiction IMHO, and it can be resolved by splitting the plugin into two. One would be only responsible for adding files to the default shortcuts file and would be called via the context menu (add). The other would load the files and show the links there. It would be configured as the viewer for .link files.
09:43:07fm2Then we could just remove the code that checks whether the file is a .link file from the plugin.
09:43:55GodEaterfm2: feel free to submit a patch
09:44:18fm2GodEater: I knew that :-))) But does this sound reasonable to you?
09:44:39GodEaterfm2: doesn't matter to me one way or the other
09:45:08GodEaterbesides, it doesn't matter if I think it's reasonable or not. If you want it commited - you'll have to convince a dev
09:45:56peturfm2: will you continue FS #7346 ?
09:47:21fm2petur: not at the moment. I've ruined my dev env, so can't code now
09:47:39GodEatertools/ ...
09:48:17fm2GodEater: to the dev env belongs also the HW, so the .sh won't help :-/
09:49:32fm2petur: but the change is really simple and can be done at any time, it doesn't depend on any other RB internals. And it's not that critical
09:50:01peturfm2: I'll do it then.... (if time is found)
09:50:46fm2petur: ok. I'll do it also if I repair the comp by then. We'll see :-)
09:50:59GodEaterwhat's wrong with the computer you're using now ?
09:51:10fm2It's not mine
09:51:19GodEaterdetails details...
09:51:32 Join miepchen^schlaf [0] (
09:52:03fm2Yes, that's where the devil is
09:53:32 Quit sergey (Read error: 110 (Connection timed out))
09:53:35GodEaterI'm not sure end users would be so keen on your two plugin approach you know
09:55:06fm2GodEater: they wouldn't see the viewer. And actually they wouldn't need to see the "adder" either
09:55:36GodEaterhmm - I guess not
09:58:07fm2Ok, they could use the "lister" standalone, it would then show the default link file. But in most cases that part would be called by playing the .link file anyway (IMHO)
09:58:43***Saving seen data "./dancer.seen"
10:01:00 Part fm2 ("bbl")
10:10:59 Join My_Sic [0] (
10:11:23B4gdergosh, the e200 thread in the forum is still viewed twice as much as the second most viewed thread
10:11:52 Quit EbErT ()
10:13:02 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
10:18:20 Join pondlife [0] (
10:19:35*GodEater looks at the views for Creative and Nano 2G...
10:21:08 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
10:21:37Tempiswhat category is that thread in B4gder
10:21:49 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
10:21:53scorcheTempis: New Ports...
10:22:10B4gdera bit misleading at this point
10:22:24TempisI just bought an e250
10:23:00GodEaterwonder why it's still so popular
10:23:22 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
10:23:34Tempisit seems to be a pretty decent player at a low price
10:23:43 Join miepchen^schlaf [0] (
10:24:04B4gderit is, yeah - I think he meant the thread
10:24:15GodEaterI did
10:24:21TempisI misunderstood, sorry
10:24:32Tempisit's late
10:25:25TempisI just bought an e250 off woot tonight, which comes at a pretty good time, since my archos seems to have snuffed it
10:25:28*GodEater thinks it should possibly be archived off somewhere - I can't imagine what information there is in it that people still need to read
10:25:57Tempisguess I can pull my 60GB drive out of it now
10:26:55 Quit webguest52 ("CGI:IRC")
10:27:00 Join webguest52 [0] (i=c023111d@gateway/web/cgi-irc/
10:27:19Tempisanyone here have a Sansa?
10:27:39*GodEater wishes woot would ship outside the USA
10:28:04B4gderTempis: i do
10:28:10Tempishow do you like it?
10:28:22B4gderit's a neat little thing
10:28:51B4gdercurrently I use my x5 most of the time
10:28:58Tempiswhen you plug it in does it show up as just an external drive?
10:29:09B4gderyes, if I would boot the OF
10:29:22Tempisso you can just copy files to it and play them
10:29:59Tempisdo you have rockbox on it?
10:30:02B4gderthat's the way to go for all players rockbox runs on
10:30:07B4gderof course
10:30:14Tempisworks well?
10:30:23B4gderyeps, runs fine
10:30:36*GodEater suspects Tempis doesn't know who Bagder is...
10:31:05TempisI suspect you're right
10:31:40GodEaterput it this way, if Bagder said he was running anything OTHER than rockbox on a target, we'd accuse him of heresy
10:32:10 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:32:25Tempisso he's the main man, so to speak
10:32:31GodEaterone of the three yes
10:33:03B4gderI'm mostly blabbing about it, not poking on the lowlevel bits as much as some of the others
10:33:27GodEaterbut you do have MrH in your hip pocket too ;)
10:33:38B4gderhaha, yeah
10:33:44Tempiswell, I've been quite satisfied with the versions I've had my my two Archos players
10:33:45GodEaterwithout whom, the sansa port would have been a great deal more tricky
10:34:00*B4gder bows in MrH's general direction, whereever that is
10:34:28B4gderTempis: rockbox really is a lot more fun on more modern players than the archoses
10:34:29Tempisalthough I haven't been alltogether satisfied with the archos players themselves
10:34:32*GodEater still envisions a park bench looking at a lake somewhere, and two shady figures in trench coats...
10:34:55 Join nick89 [0] (
10:35:52TempisI'm sure I'll be more than happy with it on my new Sansa
10:36:05Tempisonce I get it that is
10:36:13*pondlife notices that the bork bork bork patch was rejected :/
10:36:31pondlifeResulting in a loss of silliness.
10:36:38scorchewe can make bork.lang
10:36:45B4gderTempis: the biggest downside is that rockbox runs slightly worse run-time wise on it
10:36:47Tempismy archos recorder bombed out this afternoon
10:36:53Tempisyeah, I read that
10:37:12TempisI don't generally listen for long periods of time, so it's not that big a deal
10:37:21Tempisan hour or so at the most
10:37:42Tempismostly when I'm outside doing the yard
10:38:07 Join bluebrother [0] (i=USzsHdat@rockbox/staff/bluebrother)
10:38:12Tempisthe runtime thing is being worked on though correct?
10:38:17TempisI thought I read that
10:38:23B4gderon and off, sure
10:38:33B4gderit's not an easy issue to address
10:38:37TempisI'm sure it'll get worked out in the end
10:39:15Tempisand I'm sure videos and all those games drain the battery
10:39:31Tempisat least faster than just listening to some music
10:40:04B4gderyeah, but when we compare run-time we try to compare using the same usage
10:40:27Tempiscan rockbox access the microds slot in the sansa?
10:40:32Tempiserr microsd
10:41:04Tempishow's the access time when the sansa is reading from the slot
10:41:18Tempisthe same as from it's internal memory?
10:41:30Tempisor can you tell a difference?
10:41:39*B4gder doesn't know
10:41:39 Join n1s [0] (
10:41:48Tempisoh well
10:41:49B4gderI haven't tried the microsd slot
10:42:08TempisI'll probably get a 2gig card soon
10:42:32Tempisperhaps one day the 4gig cards will be usable
10:42:47B4gderI'm not sure about that
10:42:51pondlifen1s: Thanks for the "sim uses the same voicefile" update btw.
10:43:00B4gderbut then I don't know the details of "sdhc"
10:43:24n1spondlife: have you tried it with voice?
10:43:25bluebrotherthey mostly only use a different addressing of the sectors
10:43:47B4gderso its only a software/driver modification needed?
10:43:59bluebrotheras far as I understand, yes.
10:44:01pondlifen1s: I'm about to..
10:44:13B4gderthat's what I've been learning too, but I've not got it 100% verified
10:44:17bluebrotherbut iirc the sdhc cards aren't required to support spi mode any more
10:44:42bluebrotherso that could be an issue. But this information also could be wrong −− I haven't verified that.
10:45:00bluebrotherI just looked it up about half a year ago, when I worked with sd cards.
10:50:47 Join My_Sic [0] (
10:52:08pixelmabluebrother: do you remember the one who reported the scroll wheel not working in a sansa sim? Seems to be related to downloaded source (not checked out) - I just compiled a sansa sim from it to try to find out something else and got the same result (checked out one works fine); probably some file missing from the tarball...
10:52:57n1spixelma: I could reproduce it here just now with a checked out source...
10:54:26n1sYes, I broke it yesterday, enabling the wheel acceleration for the sim broke it, now to find out why...
10:54:27bluebrotherpixelma: the problem seems to be 64 bit related
10:54:28pixelmahmm... maybe my checkout is not the newest then...
10:54:47bluebrotherunfortunately that tracker entry is _really_ worse. *hmpf*
10:54:59pixelmaah, n1s seems likely
10:55:09advcomp2019one thing i know is the sansa connect is going to be able use sdhc cards
10:55:13bluebrotherhe asked yesterday evening and I tried on ia32 and told him it works on that
10:55:46B4gderare there really microsdhc cards available?
10:55:47bluebrotherand he also told he uses a chroot32 environment. Also not part of the task :(
10:55:59n1shmm, is a 'long long' supposed to be 64 bits on our targets?
10:56:03bluebrotheryes, they are.
10:56:19advcomp2019yea there is
10:56:36B4gderlong long should be fine for all targets and sims afaik
10:56:54 Quit My_Sic (Read error: 104 (Connection reset by peer))
10:57:09pixelmabluebrother: but for me it's not working in a cygwin built sim, winXP (32)
10:57:36pixelmaI'll up my checkout and try again
10:57:47*bluebrother does the same
10:58:21pixelman1s: btw... I found "apps/sound_menu.c" is missing in the tarball (if I didn't miss something)
10:59:03n1spixelma: that file is not in svn
10:59:30 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
10:59:50pixelmahmm... then that's a relict in my checked out tree...
10:59:52 Join miepchen^schlaf [0] (
11:01:38markunjhMikeS: in the ARM9 documentation, let me check which one it is
11:02:56bluebrotherhmm, now it seems broken for me too. Strange.
11:03:10bluebrotherI'm absolutely sure I did a svn up yesterday before trying.
11:04:06 Quit miepchen^schlaf (Client Quit)
11:04:44 Quit nick89 ("Gotta Go")
11:05:11 Part HellDragon
11:05:18amiconnn1s: Maybe 'long long' is 128 bit on a 64 bit system?
11:06:44B4gder"at least 64 bits" it seems
11:06:50n1samiconn: I tried replacing the suspect with uint64_t and it made nodifference, but do you know of an easy way to find out?
11:07:19B4gdern1s: make a quicky check for sizeof(variable)
11:07:32 Join norbusan [0] (
11:07:36B4gderI'm sure it is 64bit...
11:07:56 Part norbusan
11:08:08B4gderI don't know of any system that feature a >64 bit integer data type
11:11:17 Join My_Sic [0] (
11:11:26amiconnLinusN: Any news regarding the oldlcd player?
11:12:11n1sB4gder: you were right sizeof(long long) returns 8 -> 64 bits
11:14:10pixelmaI can also confirm that the scroll wheel is not working when the sim is compiled from the checkout
11:18:42 Join BigBambi_ [0] (n=Alex@
11:22:42jhMikeSmarkun: thanks so much
11:23:20n1sjhMikeS: any ide why wheel acceleration would kill the wheel in simulators?
11:23:49 Nick datasleep is now known as datachild (
11:27:28jhMikeSn1s: it does?
11:28:42n1sjhMikeS: yes, after I enabled it for the sim yesterday, sansa sim has no up/down movement in lists, I've gotten as far as that button_apply_acceleration() always returns 0...
11:29:10jhMikeSintptr_t is 64 bits for a 64-bit system but only the wheel driver itself should set bit 31 in the data. unless the sim posts something with that set, the code shouldn't execute.
11:30:05n1sin the data variable passed to that function?
11:30:36jhMikeSyes. the driver should post a delta of 1 and not set bit 31.
11:30:56 Nick TinoM| is now known as TinoM (
11:30:57jhMikeSso data = 1 << 24
11:31:31n1sjhMikeS: ok, thanks, I will test it
11:31:56jhMikeSdoes the sim button driver set the data for the last message and implement button_get_data?
11:33:04n1sjhMikeS: I'm not too familiar with the button code but yes, button_get_data is there
11:33:36*jhMikeS now seems to remember having to add that or something just to get it to compile
11:37:35 Quit BigBambi (Read error: 110 (Connection timed out))
11:38:41 Nick nerochia1o is now known as nerochiaro (
11:40:40 Quit perrikwp ("ChatZilla [Firefox]")
11:41:15 Join Meridiana [0] (
11:44:02 Join miepchen^schlaf [0] (
11:44:15 Join Thundercloud [0] (
11:45:12 Join Entasis [0] (
11:50:19 Part Meridiana
11:50:45ddaltonis the simulator working with voice?
11:51:04n1sddalton: yes, it should if you have a 32 bit computer
11:52:54amiconnOnly for swcodec targets though
11:52:57ddaltonok because it wasn't before.
11:54:53ddaltondoes the radio work on the simulator? Can I test it on it?
11:55:27n1sddalton: I'm not sure how that would be possible, you can enter the screen but it doesn't do much...
11:58:46***Saving seen data "./dancer.seen"
12:00:35ddaltonhow do you make rockbox say a certain thing every 3 seconds?
12:01:28 Join spiorf [0] (
12:01:58peturddalton: I think something similar is done in the yesno screen (yesno.c)
12:02:05n1sddalton: I guess you make an infinite loop with a sleep(3*HZ); in it somwhere and preferrably a way to exit it :-)
12:04:28 Quit TinoM (Nick collision from services.)
12:05:42ddaltonn1s: Can you give me an example?
12:06:32ddaltonLike this: sleep(3HZ); or is that wrong?
12:09:22bluebrotheryou need to multiply, i.e. 3 * HZ
12:09:38bluebrotherlike n1s mentioned
12:09:49peturthat would however mean it is unresponsive for 3 seconds.... not what you want probably
12:10:55bluebrotherindeed ... better use shorter delays and a counter
12:11:11peturddalton: look at the function gui_syncyesno_run() in apps/gui/yesno.c
12:14:15ddaltonso how do I tell the voice to shut up for 3 seconds?
12:14:23ddaltonand then talk again?
12:16:48 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
12:17:09 Join miepchen^schlaf [0] (
12:18:12bluebrotherhmm. Does someone know where I can find information about this windows rc files?
12:23:47 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
12:29:07 Join x1jmp [0] (
12:29:15 Quit ctaylorr (Read error: 110 (Connection timed out))
12:39:30 Join Febs [0] (
12:43:10DMJCthis is a little OT
12:43:22DMJCbut does anyone know if it's possible to hack the ipod's itunesDB
12:43:36DMJCto make it store the music in the toplevel folders?
12:44:54B4gdereverything is possible given enough time and resources
12:45:16DMJCheh cool
12:45:24*amiconn wonders why people using rockbox would even care about the itunesdb
12:45:26desowiniTunesDB has limit for filename (directories+filename have fixed maximum size)
12:45:38 Join x2jmp [0] (
12:46:01DMJCamiconn: because rockbox isn't really ready for production use yet (low battery life)
12:46:44*bluebrother suspiciously looks at B4gder
12:47:11GodEaterAFAIAA it's possible to change the iTunesDB to do this already
12:47:40GodEaterthe problem is the the Apple Firmware ignores any entries in it which are referenced outside of the hidden ipod directories
12:47:40 Join rusty208 [0] (i=dcfd58d2@gateway/web/cgi-irc/
12:47:48GodEater</off topic>
12:47:49desowinthere was foobar plugin to recreate iTunesDB, you might want to try it
12:48:09GodEaterit doesn't work anymore
12:48:12desowin<GodEater> the problem is the the Apple Firmware ignores any entries in it which are referenced outside of the hidden ipod directories <- wow, it worked for me, only if total filepath was small
12:48:13GodEaterfor reasons stated above
12:48:13DMJCGodEater: so I need to find a hack for the firmware :)
12:48:25desowinDMJC: grab rockbox instead
12:48:30DMJCalready installed
12:48:36GodEaterdesowin: presumably you're on some very old version of the ipod firmware then
12:48:41DMJCbetter codec support definately in rockbox
12:49:03DMJCappleos can do .ogg and wma but they won't enable it
12:49:05 Join My_Sic [0] (
12:49:19B4gderhow do you know it can?
12:49:20DMJCitunes on mac has the icons for both
12:49:26DMJChas had them for years
12:49:29JdGordonhow do i get the mem usage for a process in linux?
12:49:30DMJCapple got asked about it
12:49:41DMJCthey said they wouldn't until either format got 50% market share
12:49:41B4gderI don't believe them
12:49:52DMJCbecause apple are jerks
12:49:55B4gderI say they speak bs
12:50:48DMJCwell I have ganked out the icons before
12:50:50*B4gder is the most reliable apple insider source on the internet... or something ;-)
12:50:54DMJCthey do exist :)
12:50:59DMJCk heh
12:51:03B4gdericons sure
12:51:10B4gdercodecs surely not
12:51:22desowinB4gder is damned hacker :P
12:51:31*pondlife spotted are planning "Understand iTunesDB internals" :)
12:51:59B4gderyeah we put our hope to them now when the zune linux site is gone :-O
12:52:08*amiconn thinks it's kinda funny that a paper exists that explains why PP5002 is unable to play ogg
12:52:15scorchethey at least have some remote sense of what they are doing...
12:52:22B4gderamiconn: haha, yeah that is amusing
12:52:27amiconnNow rockbox does exactly that, even with only using one of the cores
12:52:34GodEateramiconn: is that the one Hugo Fiennes wrote ?
12:52:47amiconnUmm, I don't remember the name
12:53:10*GodEater thinks it probably was
12:53:12markunamiconn: rockbox is no proof
12:53:22pondlifeActually, they seem to be at least collecting info... and I like their section on Entropy.. ;)
12:53:27GodEaterwe're all about the smoke and mirrors us
12:53:34DMJCis the 75mb file in the hidden directory on the ipod the firmware?
12:53:41scorchethe chess player!
12:53:49desowinDMJC: firmware has its own partition
12:53:50rusty208is there any way to change from rockbox and the default firmware without the use of a pc?
12:53:59DMJCso what would be the 75mb thing?
12:54:08scorchewe have a mini turk rearranging the bits into aac data =)
12:54:19GodEaterrusty208: on what player ?
12:54:38rusty208fith gen ipod video (80gb)
12:54:45GodEaterrusty208: then yes - read the manual
12:54:59rusty208lol cheers. bbs :P
12:55:38GodEateryes, that's the one Hugo wrote
12:56:17scorchewell, there is that broken cache, so it isnt like they were lying
12:56:35GodEaterhe was and is well placed to say
12:56:58B4gderhe was right on some parts, just did the wrong conclusion it seems
12:56:59GodEatergiven his place did the Rio Karma, and he's now Mr. iPhone.
12:57:04 Quit My_Sic (Read error: 104 (Connection reset by peer))
12:57:11GodEaterhe did say "may" not have the horsepower ;)
12:57:37scorche"in theory possible"
12:57:39 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
12:57:41B4gder"The Karma plays OGG, though it's still a resource hog - you get about 25% less battery life - about 11-12 hours compared to 15+ for MP3 due to the extra cycles and memory requirements when compared to the more svelte codecs."
12:57:43GodEaterthey used an a completely unoptimised version of the ogg codec for the karma
12:57:43scorcheit seemes we proved his theory..
12:58:05GodEatertremor straight out of the box as I recall
12:58:12DMJCeew heh
12:58:14GodEaterone of the tremor authors got quite cross about it as I recall
12:58:19DMJCI remember iriver back in the day
12:58:25DMJCbefore it was optimised
12:58:29DMJCit was pretty funny
12:58:38GodEaterhe said "this was supposed to be a sort of proof of concept - you weren't supposed to just implement it without some tidying up!"
12:58:45DMJC(first non rockbox developer to flash their player)
12:59:14*GodEater is still waiting for Hugo to approve his friend request on facebook
12:59:28DMJCwhat is the cpu in the ipod video?
12:59:37DMJCnot motorolla is it?
12:59:55GodEaterARM7 - dual core
13:00:09DMJCI would laugh my ass off if there was a 68040 player running amigaos out there
13:00:45 Quit x1jmp (Read error: 110 (Connection timed out))
13:01:44B4gderbtw, it would be cool to see more rockbox committers reguster on ohloh and submit their project, cast kudos and stack rockbox
13:02:01B4gderonly 24 stacks so far
13:02:13*desowin already stacked
13:02:20*GodEater will do that as soon as he gets commit access
13:02:22DMJCheh now here's an idea
13:02:36B4gderscorche: as in a software piece you use
13:02:42DMJCyou said the database ignores dirs not inside off the hidden folders..
13:02:45 Join w1ll14m [0] (
13:02:46DMJCwhat about a symlink?
13:02:48desowinthat stacking is for users, not only for commiters, or I am wrong?
13:02:55B4gderdesowin: right
13:03:26GodEaterDMJC: there's no such thing on FAT32
13:03:34 Join haemmy [0] (n=stefan@
13:04:20DMJCspeaking of, is anyone considering HFS+ support in rockbox?
13:04:25DMJCis it unfeasible?
13:04:28GodEaterit's on our NoDo list
13:04:56DMJCfat32 is annoying
13:05:00DMJCdamn 4gb file size limit
13:05:20*scorche removes the soapbox from under DMJC's feet
13:05:24*GodEater doesn't find that annoying on a DAP
13:05:44DMJC80gb player :) portable hdd not just a DAP
13:06:14*GodEater also has an 80GB player, and hasn't ever found a >4GB file he wants to carry round
13:06:17pixelmaFebs: you thing (ehm think) he'll understand the thing better this way? ;)
13:07:06DMJCGodEater: vmware virtual discs and dvd isos :)
13:07:17desowinthen 80gb is small
13:07:18n1sGodEater: what about all those week-long mp3s...
13:07:30GodEatern1s: of course, how could I have forgotten
13:07:33DMJClol week long music file
13:07:40Febspixelma: just trying to speak his language. :)
13:07:46scorchedesowin: that is what my 160 gig HD in my archos is for =)
13:08:01DMJCI'd like to see somone fill an 80gb drive with midis
13:08:12n1syay, feature requests for unimplemented features... _)
13:08:14GodEaterwhat a weird thing to wish for
13:08:45DMJCwell it'd be interesting to see if there actually exist 80gb of midis
13:09:25pixelmadon't forget all the +4GB midis... *g*
13:09:46DMJChow long would a 4gb midi run for?
13:10:00DMJC200 years later..
13:10:17desowin"I missed the centre part, rewind"
13:10:27*scorche readies the offtopic stamp
13:11:16markunDMJC: feel free to work on it
13:11:18DMJCso more on topic, with power saving, is that likely to get worked on anytime soon?
13:11:26DMJCand yeah heh how would you go about doing it?
13:11:31markun(that was about NFS+)
13:11:36DMJCI figured heh
13:11:56markunrockbox is open source, you can change it any way you like
13:12:06scorcheDMJC: please search through the forum/wiki/irc history...these questions have been answered many times int he past
13:12:23scorcheon that note, if you figure out how to do it, let us know =)
13:12:43scorchepower onsumption
13:12:50DMJCah k
13:13:01DMJCbtw that reminds me
13:13:13DMJCwhere is the top level menu defined? I might have a play around with that code
13:13:32DMJCthemeing tools were a bit limited for messing with that
13:14:25scorcheB4gder: how does one register as a developer on this?...i am not seeing proper documentation =P
13:14:56B4gderscorche: register first, then go to a project, view contributors, click "this is me"
13:15:09scorcheah...i shall wait till i actually commit then =P
13:15:33B4gderwell, you can still register you just can't get listed as a contributor until you actually are ;-)
13:15:48scorchei have contributed many times! =P
13:15:54markunDMJC: apps/menus/main_menu.c I think
13:16:08B4gderscorche: on ohloh, only commits count
13:16:12*GodEater has contributed a whole three times
13:16:45pixelmaB4gder: including all the "oops" commits? ;)
13:16:57B4gdera good reason to do many oops commits!
13:17:08JdGordondoh! you've figured out my plans!
13:17:09*pondlife must commit lots of comments :)
13:17:33B4gderafter all, linus is #1 committer and he does lots of oops ;-P
13:17:48B4gderhe's a great example to learn from
13:17:52*B4gder bows
13:17:57DMJCLinusN rocks :)
13:18:14LinusNdamn, you're on to me
13:19:25 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
13:19:46 Join miepchen^schlaf [0] (
13:23:17pondlifeHmm, could we have the Categories in Flyspray put into alphabetical order please?
13:24:59*B4gder pokes the list
13:25:56*amiconn still doesn't really understand the purpose of ohloh
13:26:13pondlifeB4gder: Thanks
13:26:18pondlifejhMikeS: ping
13:26:29B4gderamiconn: keep track of open source, the people who write it and what we use
13:26:42*pondlife would rather not be tracked.
13:27:02*B4gder temporarily disables pondlife's tracking device
13:27:05amiconnB4gder: Yeah, but what for? I can't imagine anyone needing that...
13:27:20DMJCcopyright stuff?
13:27:23B4gderit's not about "need" to me, but about fun and interesting
13:27:37scorcheisnt that why we do rockbox? =)
13:27:43pondlifeI'm failry sure and are the same issue... jhMikeS might want to confirm as he understands it...
13:28:24*amiconn thinks that rockbox svn is enough for tracking rockbox
13:28:42B4gderamiconn: not if you want to see historic development and comparisons against other projects etc
13:29:00*petur has used ohloh to find a particular commit of somebody...
13:29:07B4gderand if you're interested in your fellow committer's other projects etc
13:29:30Aveamiconn: do you have a suggestion for debugging the current nano playback problem (
13:29:41AveI have failing device and I could test out
13:32:46jhMikeSpondlife: yups
13:32:53*GodEater laughs at Febs ipod forum post
13:33:32FebsThanks, I'll be here at week. Be sure to tip you bartenders and waitress.
13:34:44amiconnAve: I have a suspicion what might go wrong, but I don't have a patch yet
13:34:52jhMikeSpondlife: yeah, probably the same and probably the r12 getting clobbered
13:35:18amiconnThe last comment in the tracker entry suggests that it might be related to the ipod HW revision
13:35:23GodEaterFebs: how's the veal ?
13:35:29amiconnLlorean: What HW rev does your nano have?
13:35:51Aveamiconn: "version 1.0" MA099FB as reported by apple-side
13:36:41*GodEater double takes at Llorean's amazing ventriloquist performance
13:36:45amiconnThe ipod HW rev, not the disk firmware
13:36:54amiconn...and not the apple version either
13:37:01Avehow di I find out
13:37:19Aveof course I dont have the device with me now
13:37:26amiconnSystem->Debug->View HW Info
13:38:13Aveok I'll check it later and compare to the entries in the bug tracker
13:40:59 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
13:41:05 Quit pondlife ("disconnected has pondlife")
13:42:37 Join webguest57 [0] (i=4a841ac2@gateway/web/cgi-irc/
13:43:02 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
13:43:23 Join miepchen^schlaf [0] (
13:43:25webguest57Could a wps with album art be used on a player without Album art
13:43:56DMJCwhich players have album art support?
13:44:08webguest57The patch.
13:44:16DMJCah heh
13:44:17webguest57i dont have it
13:44:28peturwebguest52: no
13:44:43webguest57that limits me alot...
13:44:47 Quit webguest57 (Client Quit)
13:44:54GodEateryou're welcome
13:44:55peturwebguest52: the wps parser will reject it and load the default one
13:45:06B4gderhey, that was 57 not 52 ;-)
13:45:12*B4gder giggles
13:45:26*petur loves autocomplete
13:46:49*pixelma remembers bluebrother's suggestion regarding the default nick provided by the webclient...
13:47:25B4gderwe should make it pick a random stupid name and append the number to it
13:48:09B4gderpossibly prepend the number too
13:48:24GodEaterB4gder: 12lazy_bastard12
13:48:26B4gderfor tab completion friendlyness
13:48:29GodEaterfor example
13:48:34B4gderlike that
13:49:13pixelmahe suggested a nick that would make them want to change it before joining (like "iamanidiot")
13:49:39B4gderI don't think that will work, hence why I suggest multiple such words
13:49:52*petur suggests donttalktome
13:50:13peturor ignore_me
13:50:27ZagorI think a list of near-synonyms for "guest" is the best
13:50:33pixelmaand for everyone to get rid of their aggression "forehead" :P
13:50:58B4gderZagor: yes I think so too
13:53:35 Quit desowin ("use linux")
13:55:11LinusNit might be easier if the number is prepended rather than appended
13:55:52B4gderalthough possibly somewhat harder to read
13:56:40 Join Rondom [0] (
13:56:44*GodEater proposes hex numbers
13:56:57GodEatercoedbabe being a personal favourite :)
13:57:17*elinenbe thinks all the popular patches should be looked at and committed... this includes the (never happening) viewports...
13:57:59scorcheGodEater: when c0edbabe came up, i thought it was a joke at first...
13:58:10*n1s thinks all devs will flee in terror if elinenbe gets what (s)he wishes for...
13:58:34GodEateror turn in their graves
13:58:37pixelman1s: look at the IrcNicks page...
13:58:47***Saving seen data "./dancer.seen"
13:59:09*elinenbe is talking about the AA / margin / etc (all WPS) patches... those seem to be the most popular ones.
13:59:19n1spixelma: ah, s/(s)//
14:00:02n1selinenbe: there's actually a reason for most of these patches not being comitted
14:01:41DMJChas anyone put in a request for themeable menus? I tried to find it in the requests
14:01:57elinenbeDMJC: there are themable icons in the menus...
14:02:19DMJCI mean like editing the menu's placement on the screen
14:02:26DMJCnot just the icons
14:02:41DMJCalso a seperate font for menu and directory listings
14:03:13DMJCand the icons at the top of the menu screen
14:03:14scorcheit would be a support nightmare...that as well has been suggested many times...please do search before making suggestions like that
14:03:16DMJCbattery meter etc
14:03:52DMJCwhy would it be a support nightmare if all default themes had it turned off?
14:04:00DMJCyou can just tell people to reset to default settings
14:04:18DMJCthat dumps the themes too
14:04:35 Join kubiix [0] (
14:04:48DMJCmenu/select not our problem
14:05:02DMJCthat's how ISPs deal with stuff they don't want to support ;)
14:05:35scorchewe arent an ISP
14:05:46scorcheISPs also pay their employees
14:06:18scorcheit really is a hassle that we would rather not go into...though a patch on the tracker would be fine
14:07:17DMJCgood enough, patches are easy enough to make
14:07:41scorchethen go ahead and add the functionality and submit to the tracker...
14:08:07DMJCheh I'll look into it but I can't guarantee I can do it, my coding skills are crap to say the least heh
14:08:18rusty208does any one have any idea why i cant get the "flow" theme for ipod vidoe 80gb to work?
14:08:31DMJCtried it in a simulator?
14:08:39scorcherusty208: it likely requires a patched build
14:08:54rusty208nope... how do i try it in a sim?
14:08:57scorcheDMJC: well, you said it was "easy enough"...
14:09:10DMJCyeah easy enough to apply a patch to the source and build it :)
14:09:23DMJCplus it gets rid of people who don't know howto do that :)
14:09:32scorchein this context, that is not what is meant by making a patch...
14:09:42DMJCI know heh
14:09:59DMJCmaking a patch isn't the difficult bit it's making the code work in the first place
14:10:12DMJCstill noone's done it yet? I find that a little surprising
14:10:31scorchethen do it yourself...
14:10:47 Quit sneakums (Remote closed the connection)
14:10:56scorcherusty208: that theme does require a patched build, and will not work with the official version of rockbox
14:11:21DMJCspeaking of patched builds what is the location in the wiki of the build instructions?
14:11:32DMJCit seems to have moved from when I was building my iriver firmware
14:11:58scorchei thought that was easy enough too =)
14:12:07GodEaterDMJC: has the search function also moved ?
14:12:19rasherBagder: ping
14:12:23DMJCheh found the svn page :)
14:12:33B4gderrasher: you ok with me? ;-)
14:12:40DMJChaven't bothered o build it in like 2-3 years
14:12:43rusty208scorche: were do i find a patched build?
14:12:54GodEaterrusty208: in the unsupported builds forum
14:13:05 Join sneakums [0] (
14:13:11amiconnRegarding default names - I like the default in AmIRC: Lazy Bone, Badly configured clients Inc.
14:13:15scorcheeither use a different theme, or find a suitable theme on the unsupported builds section of the forums....keep in mind those builds are just that...unsupported
14:13:21 Join rogelio [0] (n=rogelio@
14:13:37rasherB4gder: That'll do. I have a voice-building commit on the table. Allows to build multiple languages at once, and lets the user configure the tts/encoder. Naturally you'll have to change your scripts
14:13:39rusty208Cheers fellas. ill b back when i scew up :P
14:13:47 Join desowin [0] (n=desowin@
14:13:50scorcheamiconn: i prefer something a bit more neutral
14:14:05B4gderrasher: fire it away and I'll adapt
14:15:28 Quit Febs (Read error: 110 (Connection timed out))
14:15:35GodEaterrusty208: if you screw up, we'll just tell you to install the supported version again
14:15:55DMJCah automation.. man's greatest gift to man
14:16:11rasherB4gder: It also allows using the name in the language-selection
14:16:22DMJCcan you brick a player with bad rockbox firmware?
14:16:29scorchedepends ont he player
14:16:31GodEaterDMJC: no
14:16:36scorchethe no
14:16:42GodEateryou can brick it with a bad bootloader, and even then that's only on certain players
14:16:48GodEaternot the ipod though
14:17:27DMJCI remember the iriver was like... ok now if this bones up... you'll need a serial adapter heh
14:17:44scorchewhat are you talking about?
14:17:57DMJCfirst time flashing of the iriver
14:18:16 Quit haemmy ()
14:18:17DMJCit was back ages ago when it wasn't 100% stable
14:19:51 Part rogelio ("Kopete 0.12.4 :")
14:21:17DMJCsorry not a serial interface I'm referring to the bdm interface on the iriver
14:24:24 Join misarm [0] (i=5546a921@gateway/web/cgi-irc/
14:26:28 Quit misarm (Client Quit)
14:28:02 Join My_Sic [0] (
14:30:53 Quit My_Sic (Client Quit)
14:31:24B4gderrasher: there's a bug
14:31:40rasherB4gder: I can imagine. What's up?
14:31:43B4gderin the voice: section in the generate Makefile
14:31:51B4gderthere's "@APPS@" left in my file
14:32:24rasherDidn't I just move that.. hang on
14:33:00DMJCthere's already a menu patch
14:33:46rasherB4gder: ah, right, of course. I only moved the "features" part (because it's convenient to be able to make features on its own)
14:33:58scorcheDMJC: please direct all of your social chatter towards someplace such as #rockbox-community
14:40:43*bluebrother wonders if we should use the archos device as icon for rbutil or the website favicon
14:41:17B4gderor just a bag... :-)
14:41:31aliaskbut what kind of bag?
14:41:34bluebrothergreat idea −− does someone have an ico file for that?
14:41:41scorchea bag full of something...
14:42:14bluebrotheruse a stop sign?
14:42:30aliaski vote the favicon
14:42:33bluebrotherand add a "please don't install our software" notice upon rbutil startup? ;-)
14:42:44bluebrotherrasher has a nice ico file of that ...
14:42:52bluebrother(the favicon I mean)
14:43:06scorcheor we could use ...
14:43:09rasherSame as is in use for fwpatcher
14:43:26B4gder"Please note that our money-back guarantee period is now limited to two years."
14:43:43bluebrotherscorche: that would look rather strange scaled-down unfortunately
14:44:07scorchebluebrother: sadly, yes, but it could be like Clippy for the rbutil!
14:44:09preglowicon of the year
14:44:20scorchethe new clippy!
14:44:25bluebrotherindeed. Add a clippy to rbutil!
14:44:43bluebrother"It looks like you want to install Rockbox. Do you want to read the manual now?"
14:44:50rasherB4gder: should be fixed now
14:44:51preglowHi, I'm Badgy, the helpful cretin. It look's like you're trying to install Rockbox!
14:44:51scorchei definitely think that would be a legendary clippy...
14:45:02scorchepreglow: bagdy =P
14:45:53B4gdernow how did that private photo leak? ;-P
14:46:24scorcheB4gder: hey now...i spent a good 10 min making that photo!
14:46:24 Quit rusty208 ("CGI:IRC (EOF)")
14:46:29bluebrotherscorche: btw, I changed rbutil_themes.php a bit again. I'll email it to you once it stabilized :)
14:46:48bluebrotherand on that topic: I would like to have some version information about themes ...
14:46:49scorchebluebrother: great...i almost have the site back to its previous status =)
14:47:31bluebrothermy idea was to save the installed version number for everything, so you can easily check if there's an update
14:47:57bluebrotherincluding themes −− but then we'd need some version string. Could be a date as well
14:48:46scorcheversion would be simple enough...but is this something that needs to be done before it goes live, or something to be added to the list after it goes into svn?
14:49:15bluebrotherwell, I'm not sure −− we could always add things later
14:49:33bluebrothermy idea was to have some cfg-like format for the txt files
14:49:37 Quit webguest52 ("CGI:IRC (Ping timeout)")
14:49:40bluebrotherAuthor: Some Name
14:49:48bluebrotherVersion: 20070813
14:50:02bluebrotherDescription: some text ...
14:50:09scorchethey are going to be a standard format...the user will not touch them, and they will be created by the php script
14:50:45bluebrotheryeah −− but how is the format defined right now?
14:51:00*amiconn spots green deltas :)
14:51:02scorchethat is in my to do list before it goes live
14:51:12bluebrothercurrently I'm simply displaying the txt file in rbutil.
14:51:45scorchewell, we will have 2 different text inside the zip for users, one for the site
14:51:51scorche/for us
14:52:11B4gderrasher: yeps, seems to work fine now
14:52:16bluebrotherhmm. And what information does which file hold?
14:52:55 Part austriancoder ("Kopete 0.12.4 :")
14:53:11bluebrotherI guess you're interested:
14:53:16rasherB4gder: Debian packages a few different festival voices. I had luck with the italian voice, althought it seems it's voicing things a bit wrong (I get a lot of " at the end of strings)
14:53:29 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
14:53:36 Join miepchen^schlaf [0] (
14:54:03B4gderrasher: I'll settle with the english ones for now until voice gets solid again
14:54:20scorchebluebrother: name, description, CC data, email, copyright data, basically anything we want
14:54:27scorcheas i said, i still have to finalize that
14:55:44bluebrotherok. Then I need to adjust the display in rbutil later
14:55:45scorcheobviously, the one outside the zip would be best to use for rbutil, but it would be best to split it off into vars and then show what you need
14:56:06scorchei can always implement a third text file if needed, but this seems like overkill
14:56:12bluebrotherI agree.
14:56:30bluebrotherif we have some key: value format it will be sufficient
14:57:00scorchewould lines of data be sufficient without the key?
14:57:23bluebrotherif we have them defined properly I guess so.
14:57:50scorcheunless you wish to show me how in php to have it deal with that format correctly =)
14:58:25 Join Febs [0] (n=chatzill@
14:58:36DMJChow long should a rockbox build take?
14:58:46scorchethat depends on a number of things...
14:58:47GodEaterhow long is a piece of string
14:58:58DMJCipod build 2gb ram athlon 64 x2 4400+
14:59:02rasherscorche: list($key, $value) = explode(":", trim($line), 2);?
14:59:06scorchedepends on more things..
14:59:07GodEaterDMJC: OS ?
14:59:13bluebrotherin fact, as we open the description txt with the php script anyway I could also put that into the output generated by rbutil_themes.php
14:59:16n1sDMJC: ccache?
14:59:21DMJCnot afaik
14:59:34aliaskShould take 1-2mins
14:59:43scorchebluebrother: we also generate those files by php
14:59:46DMJCok so something is quite broken
14:59:54n1saliask: may be faster if he's using make -j
15:00:00bluebrotherthe description files?
15:00:09scorcheboth .txts
15:00:12bluebrotherwell, as those are static files it isn't an issue
15:00:33n1sDMJC: does it tell you anything?
15:00:34DMJCcan I paste 9 lines here?
15:00:41B4gderuse a pastebin
15:00:42bluebrotherDMJC: use a pastebin
15:00:45 Quit Zagor ("Client exiting")
15:00:49n1shah :-)
15:01:10scorcherasher: thanks...i shall look more into it...i dont deal much with php as you can tell =)
15:01:18*n1s thinks we're a pretty paste-unfriendly channel :-)
15:01:28scorchewell, there are guidelines...
15:01:36bluebrotherhmm, putting the information in the generated output might be a good idea ... less files to grab from the server
15:01:52DMJCit keeps repeating that but bumping the number up each time
15:02:51scorchebluebrother: can we continue this tomorrow night, perhaps?....i am seeing light through my blinds =)
15:03:02n1sDMJC: is your checkout current?
15:03:13DMJCfrom about 30-60 mins ago
15:03:18n1srasher has been messing with the confugure script
15:03:26bluebrotherI'll try to finish theme installation, and we can adjust it to the site later.
15:03:30n1sDMJC: update again rerun configure
15:04:28DMJCthat looks more normal
15:04:41DMJChmm nope, same error
15:05:04n1sDMJC: have builds worked ok before?
15:05:19DMJCno this is my first time attempting to build it
15:05:34DMJCthey have worked on this machine before but the target was iriver
15:05:37DMJCand it was AGES ago
15:05:41n1sDMJC: how did you install your build environment?
15:05:51*GodEater pokes Febs
15:06:08GodEaterreply to your ipod thread thing
15:06:11GodEaterbit shirty
15:06:15GodEaterneeds your love and attention
15:06:19FebsAlready done.
15:06:36GodEaterso efficient
15:06:46bluebrotherDMJC: is that line correct? Starting with "./"?
15:06:54n1sDMJC: what does arm-elf-gcc -v tell you
15:07:32DMJCUsing built-in specs.
15:07:33DMJCTarget: arm-elf
15:07:33DMJCConfigured with: ../gcc-4.0.3/configure −−target=arm-elf −−prefix=/home/james/rockbox/arm-elf −−enable-languages=c
15:07:33DBUGEnqueued KICK DMJC
15:07:33DMJCThread model: single
15:07:34GodEaterI keep getting tempted to wipe most of the responses to that thread, and then just lock it
15:07:34DMJCgcc version 4.0.3
15:08:23FebsGodEater: same guy just posted in the "Rockbox KILLED my ipod thread." I think he's earned himself a ban.
15:09:01B4gderDMJC: are you using gnu make?
15:09:04rasherAnd apparently, we're all inbred morons.
15:09:04n1sDMJC: how about make -v
15:09:06GodEateroooh - you're tolerant today then hey ? :)
15:09:22DMJCyes 3.81
15:09:28GodEaterand a thread all of his own I see
15:09:33DMJCThis program built for x86_64-pc-linux-gnu
15:09:56 Join webguest52 [0] (i=c023111d@gateway/web/cgi-irc/
15:09:57GodEaterhaha - yeah - go get him Febs
15:09:58bluebrotherDMJC: have you called the configure script the way you posted earlier starting with "./"?
15:10:29rasherbluebrother: Really doesn't matter.
15:10:50DMJCI think I figured it out
15:10:57DMJCnope I didn't
15:11:06DMJCyeah I didn't have a build directory i do now
15:11:25bluebrotherrasher: sure? We always recommend to use a build directory ...
15:11:50rasherbluebrother: but rockboxdev builds in a separate directory all by itself
15:11:51DMJCyup fixed it
15:11:59DMJCran the configure script from the sourcecode directory
15:12:06rasherbluebrother: ${HOME}/build-rbdev
15:12:07DMJCwhich rewrote the makefile
15:12:14DMJCwhich totally boned the build of course heh
15:12:23bluebrotherrasher: I was referring to building Rockbox, not the compiler
15:12:31DMJCblew it away, svn update and made a target dir
15:12:33B4gderDMJC: you really should pay attention to the instructions...
15:12:43GodEaterHurry up with that ban febs
15:12:48GodEaterhe's posting all sorts of crap now
15:13:03B4gderyeah get him off that forum
15:13:45CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
15:13:45*GodEater wishes experts had a proper "delete" button
15:13:49rasherbluebrother: but he only said he built the compiler with that line
15:14:24w1ll14mwhich dac is in the ipod video ?
15:14:42bluebrotherindeed. I misread that −− but it looks like I accidentially hit :)
15:14:53FebsI already banned him. He must have snuck that last post in.
15:15:03n1sw1ll14m: WM8758
15:15:07DMJCok assuming this worked
15:15:13DMJCwhere should rockbox get spat out?
15:15:19w1ll14mn1s thanx!
15:15:45scorcheGodEater: ah, but having them in to be deleted makes for future enjoyment
15:15:46rasherDMJC: in the dir where you're building. make zip will create a handy zipfile
15:16:06DMJCah that's better
15:16:30DMJCvery nice
15:17:15GodEaterscorche: then a quicker way to move one post there
15:17:25GodEaterrather than all that split / move nonsense
15:18:04DMJCpatches are unsupported right?
15:18:26DMJCbut they generally work
15:18:27scorchepatches are just patches
15:18:38rasherDMJC: they might be supported by their author
15:18:44scorcheif they are used in a build, then that build is an unsupported build
15:18:49DMJCthat's fine
15:19:00bluebrotherthere is never a guarantee that a patch works.
15:19:22DMJCbut they usually do? they usually fail?
15:19:29scorcheFebs: didnt want to keep the porn post around for future enjoyment?
15:19:48B4gderit wasn't fun anyway
15:20:10bluebrotherDMJC: there is no usually
15:20:15DMJCpatches apply to your build directory?
15:20:27FebsI think I hit delete before I realized that it had been moved to the "to be deleted" forum.
15:20:34DMJCO'll figure it out I guess
15:20:53bluebrotherif a patch just got posted they usually work. Depending on the patch it might take a long or really short time until it doesn't work anymore
15:20:55scorchewell, while i am up, might as well do the raffle of the shirts and mug
15:21:13bluebrotherwe usually ask patches to get created from the source root
15:21:22bluebrotherso you can apply them with patch -p0
15:22:37bluebrotherunfortunately not all submitted patches follow this. The most do.
15:22:59DMJChmm failed hunks.. bad
15:24:59DMJCone last thing
15:25:09DMJCif a patch fails to apply with failed hunks, don't bother building it?
15:25:38B4gderthat's left for you to decide/work out
15:25:39bluebrotherwell, if it fails you need to fix it. Building most likely will fail, depending on what failed upon patching.
15:28:41Lloreanamiconn: 0x000C0005
15:31:26DMJCheh the scrollwheel patch is boned, the menu theming one is more promising
15:32:21scorchewe have winners =)
15:32:35*GodEater wants to know how fair and impartial the raffle was
15:32:35bluebrothernice :)
15:32:52scorcheGodEater: my methods are in the thread
15:32:53*bluebrother goes checking the results
15:33:03scorchebluebrother: very nice...for you =)
15:37:11*GodEater likes that we used a RB plugin for the decision
15:37:14GodEaterwho wrote it though ? ;)
15:37:49scorchewe? =P
15:38:25B4gderAdded Sun Mar 5 22:42:50 2006 UTC (17 months, 1 week ago) by lostlogic
15:39:21 Quit desowin ("use linux")
15:41:52scorchebluebrother: are those "//" a separation, or an official use?
15:42:06bluebrothersomething in-between.
15:42:43bluebrotherofficially (AFAIK) they are a separator, so I'm using them. As long as the part before and after them is present it doesn't matter
15:43:02*scorche heads off to
15:43:21bluebrother(c28 is the room number −− still living in a student hall :) i guess
15:44:01scorcheno it is com...odd
15:54:44 Join Tanja [0] (n=tanja14a@77-105-52-131.adsl-1.sezampro.yu)
15:58:49***Saving seen data "./dancer.seen"
15:59:15 Part Tanja
15:59:55 Join desowin [0] (n=desowin@
16:00:18 Join bobbyguzik [0] (
16:02:53Aveamiconn: I'm running hw rev 0xc0005 like the other non-working one
16:04:13 Quit bobbyguzik (Client Quit)
16:07:04 Join bobbyguzik [0] (
16:10:29amiconnWhat HW revision does your nano have?
16:14:29Lloreanamiconn: 0x000C0005
16:15:52 Part bobbyguzik
16:16:10DMJCnice I got one patch applied
16:24:39LloreanYeah, things would just make sense if mine wasn't working right
16:24:55scorcheand/or mine
16:26:11scorcheas well, my hardware revision is 0x000C0006
16:30:07amiconnscorche: But yours is working, right?
16:30:50amiconnFrom the comments in flyspray it looked like the bug hits on rev. 0xc0005
16:31:24*amiconn would like to have access to such a problematic nano
16:31:51amiconnremote debugging is cumbersome and slow
16:32:22 Join norbusan [0] (
16:32:33 Quit idnar (Nick collision from services.)
16:32:35 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
16:32:46 Part norbusan
16:36:43 Part Llorean
16:36:51 Quit B4gder ("It is time to say MOOO")
16:37:33Avedamn he left
16:37:45Avesee, it sporadically works
16:38:07Avethen suddendly starts to garble up, I was suspecting heat related hw problem before I started to dig flyspray
16:38:41 Quit pixie (Remote closed the connection)
16:39:37 Quit webguest52 ("CGI:IRC")
16:43:00 Part LinusN
16:44:41 Quit x2jmp (Remote closed the connection)
16:45:33 Join david [0] (n=chatzill@
16:46:02davidis it possible to view pictures or set a backdrop on an ipod 4g grayscale?
16:47:19davidi put a bmp file with the same dimensions (128x160) in the backdrops folder, but i can't view it. however, the grayscale is a 2 bit color scheme, so i guess this would prevent any image viewing?
16:47:48amiconnYou can
16:47:59amiconnContext menu->set as backdrop
16:48:41davidbut i wasn't able to view it from the File menu
16:49:20davidsry, where is the context menu?
16:49:27amiconnYou can set it as backdrop, but not view it directly
16:49:34amiconnThese are 2 different things
16:52:34dionoearockpaint (the only bmp viewer atm) only works on color targets
16:52:38dionoeaI should fix that :)
16:52:55davidwhat do you mean by Context menu?
16:53:16 Quit JdGordon (Remote closed the connection)
16:53:28peturdavid: now would be a good time to read that manual ;)
16:53:52davidthat's what i was hoping to avoid :(
16:58:08 Join Wofl [0] (
16:58:27 Join toffe82 [0] (
17:04:06*Nico_P just spotted the rockbox facebook group
17:04:58 Quit Wofl (Remote closed the connection)
17:06:23GodEaterNico_P: which one ?
17:06:34 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
17:06:46Nico_Pthe one with only 15 members (now 16)
17:06:54Nico_Pbut actually I saw both
17:09:30GodEaterI think I joined both
17:09:41*aliask joins the small one
17:12:04 Quit david (Remote closed the connection)
17:16:45 Join shriphani [0] (
17:16:55 Join saratoga [0] (i=98039a34@gateway/web/cgi-irc/
17:17:01shriphanihello. i am trying to install rockbox on the 80 gb 5.5g ipod.
17:17:15shriphaniwhere do i find a rockbox.ipod file ?
17:17:49 Join yjoe [0] (n=jj@
17:18:03GodEatershriphani: which bit of the installation instructions in the manual did you not understand ?
17:18:04FebsIt's contained in the Rockbox current build. See section 2.2.3 of the manual.
17:18:04peturshriphani: please read the manual and/or use rockboxutility to install
17:18:26GodEaterI got in first here....
17:18:51shriphanii am installing from
17:19:21peturu n s u p p o r t e d
17:20:02peturshriphani: first get the official version working
17:20:49shriphaniokay. i unzipped it wrongly the 1st time.
17:21:21peturoh my those builds are a bit old....
17:21:30shriphanioh oh.
17:21:50GodEaterpetur: careful, you're straying into support there....
17:22:02*petur backs off quickly
17:22:26shriphaniyou are prohibited from supposrting ?
17:22:31shriphanisupporting ^
17:23:11GodEaterwe *can't* support them
17:23:19GodEaterwe have no idea what they're supposed to do
17:24:26 Quit yjoe ()
17:25:09shriphaniso the new build is the 64mb ipod video build ?
17:32:55alienbiker99i just bought the sansa from woot =)
17:36:32 Quit spiorf ("Read error: 110 (Connection timed out)")
17:36:51 Join spiorf [0] (
17:36:52dionoeawow, that's cheap
17:39:12Nico_Plinuxstb: here ?
17:40:12linuxstbNico_P: Yep.
17:40:41saratogalinuxstb: i have lots of asf questions
17:40:57linuxstbsaratoga: I'll try...
17:41:15Nico_Plinuxstb: what do you think about making get_metadata() take a struct mp3entry as parameter instead of a struct trackinfo ?
17:42:23saratogalinuxstb: i wrote seeking code that seems to work pretty well
17:42:25Nico_Plinuxstb: sadly the public git hosting site to which I could have pointed you for a patch is currently unavailable, but I can produce one if you want to see
17:42:55 Join EbErT [0] (
17:43:54saratogahowever, i occasionally get glitches were it just sort of dies on me and skips to the next song
17:44:34saratogai'm wondering if theres ever a time when the packet length info in the header isn't enough
17:44:50saratogaalso, i'm wondering if packets are always the same size
17:45:42Nico_Plinuxstb: ... quite a simple change actually
17:49:42 Join jgarvey [0] (
17:51:45 Quit petur ("*plop*")
17:52:07 Quit aliask ("ChatZilla [Firefox]")
17:54:45*linuxstb back
17:56:11linuxstbsaratoga: One problem you'll have with seeking (which I forgot to mention earlier..) is that the ASF parser moves data in the audio buffer (when a packet contains multiple payloads, which seems to be a minority of streams). I've since realised that's a bad idea, but haven't got around to fixing it. But afaik, packets are always the same size in bytes.
17:57:15 Quit tumu ("~")
17:58:52 Join Wofl [0] (
17:58:54***Saving seen data "./dancer.seen"
17:59:47linuxstbNico_P: That would seem to make sense to me.
18:04:49nerochiarolinuxstb: speaking of the asf parser, it seems you added a couple extra fields at the top, that doesn't seem to be "standard" for waveformatext (if there's any standard)
18:06:18linuxstbnerochiaro: That struct should probably be renamed - it started out as something called waveformatext, but is used to store all information read from the asf headers which is later needed by the decoder.
18:06:40nerochiarolinuxstb: it's mostly waveformatext
18:06:51nerochiarolinuxstb: except for the 2 extra fields nad the fixed extra data buffer
18:07:11nerochiarolinuxstb: at last as far as i can see
18:08:55saratoganerochiaro: i'm wondering if that zip i sent you the other day actually works
18:09:09saratogathinking about it i may have sent you a version i was testing something out on
18:10:45nerochiarosaratoga: no, it doesn't. for some reason all blocks always fail
18:10:52 Quit Wofl (Remote closed the connection)
18:10:59 Join yjoe [0] (n=jj@
18:11:19nerochiarosaratoga: i didn't notice initially, because it didn't stop decoding
18:13:14yjoeI upgraded from a January build to a recent build on a iPod 5g and now the car adapter mode does not seem to work. Is this a known issue? I did not see a bug for it.
18:16:31saratoganerochiaro: try this
18:16:47saratogai just compiled it and decoded a file but if you could do the same now to make sure
18:16:52nerochiarosaratoga: thanks, i'll try in a few minutes
18:17:51saratogalinuxstb: how is that a problem for seeking?
18:18:06saratogadoesn't the audio buffer get flushed when theres a seek?
18:18:33preglownot if you don't seek outside buffer boundaries
18:22:26 Join Domonoky [0] (
18:23:23linuxstbsaratoga: What preglow said - IIUC, if you seek back, and the data hasn't yet been overwritten, the playback engine is smart enough to reuse it.
18:23:43saratogaso the audio buffer is the large (20-30MB) buffer?
18:24:12linuxstbYes. You can tell when that happens - if you seek backwards and the disk doesn't spin up.
18:24:32saratogai didn't realize that was the audio buffer
18:24:41saratogathat will certainly make backwards seeks complicated
18:24:49 Join Lear [0] (i=chatzill@rockbox/developer/lear)
18:25:11nerochiarosaratoga: it always segfaults when closing asf, for some reason. but aside of that, it works
18:25:23saratogayes it tries to double free something
18:25:38linuxstbsaratoga: No, I know how to fix it, and will have a go now. You shouldn't need to worry about it. We need to make a rule that codecs can't modify the audio buffer though...
18:26:11saratogalinuxstb: what about some of the other issues that have cropped up, like playing ffmpeg
18:26:15saratogas asf files
18:26:21saratogai'd like to look into them
18:26:33nerochiarosaratoga: good
18:26:36linuxstbsaratoga: I haven't looked at those. BTW, do you have a patch for seeking?
18:26:58linuxstbAnd how accurate is it? i.e. after a seek, do you know for certain where in the stream (in terms of samples from the start of the file) you are?
18:27:37nerochiarosaratoga: linuxstb: i have almost finished with the decoder port here, and it would be interesting to have some sample files that use more than 1 payload per packet, for example.
18:27:41saratogalinuxstb: an ugly one
18:30:02saratogalinuxstb: its accurate
18:30:22saratogaat least on my files i couldn't get it to lose track
18:30:33saratogapreglow may have had more luck then i since he got more wma files
18:30:57saratogabut basically i just read the packet lengths off and skip ahead that much, etc until i find the packet i want
18:34:13 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
18:35:33linuxstbsaratoga: Looking at your patch, am I right in thinking you just read (forwards or backwards) through the whole file, parsing the packet headers as you go?
18:38:01 Join ze_ [0] (
18:38:12preglowonly forwards
18:39:21linuxstbAh yes, a "TODO" exists for backwards...
18:39:54preglowseveral ways to seek backwards exists, i think
18:40:16preglowyou could just seek back to the start, then seek forwards, or step backwards, but i don't know how that sits with the playback engine
18:40:26linuxstbIIUC, all the asf packet headers contain timestamps, so it should be possible to seek properly - i.e. estimate the target packet based on average bitrate, then adjust.
18:40:44linuxstb(all packets are a fixed size in bytes, so seeking to packet N is trivial).
18:40:44preglowthat would be optimal
18:40:48preglowbut would need a resync
18:41:09linuxstbWhat do you mean by resync?
18:42:03preglowwell, if you just seek blindly around based on avg bitrate, you're gonna need to resync to a packed boundary before looking for data
18:43:05Nico_Plinuxstb: "We need to make a rule that codecs can't modify the audio buffer though": as long as the codecs only modifiy the parts of the buffer where the compressed audio is stored, it's OK, isn't it ?
18:43:49linuxstbNico_P: Not if the user then seeks backwards. It's desirable to not have to re-read that data from disk if it hasn't been overwritten by later tracks.
18:44:12linuxstbpreglow: All ASF packets are a fixed size in bytes, so we can seek directly to a packet boundary.
18:44:26 Quit ze (Read error: 110 (Connection timed out))
18:44:27 Quit chrisjs169 (Remote closed the connection)
18:44:27 Nick ze_ is now known as ze (
18:44:28Nico_Poh so the codec isn't capable of rereading the parts it changed, is that it ?
18:44:45preglowlinuxstb: what about vbr?
18:44:55preglowlinuxstb: anyway, that assumes no errors in the stream
18:45:35 Join chrisjs169 [0] (
18:46:43linuxstbpreglow: Packets are still a fixed size (in bytes). But you're right that a resync should be added, just as a safety measure. I think most codecs do that anyway.
18:49:04 Join robin0800 [0] (
18:52:33pixelmabluebrother: do I remember correctly that the font used in the svg was one of the palatino font family?
18:53:37preglowlinuxstb: yeah, they're a fixed size, but that doesn't help you far when you're skipping several of them, unless vbr is really uncommon
18:53:55preglowanyway, i don't think walking all of them is that bad
18:53:59preglowisn't that what ogg does anyway?
18:56:20linuxstbpreglow: I would be very surprised if any codec framwalked the file to seek - it wouldn't be good when you're seeking in a long file (larger than the buffer).
18:57:39 Quit shriphani ("Lost terminal")
18:57:48preglowindeed not
18:58:24linuxstbpreglow: You know the following about ASF files: 1) The size of the packets (always a fixed number of bytes); 2) The filesize of the file, and hence the number of packets in the file; 3) The duration of the file in ms. So assuming we can work out how to interpret the timestamps in the packet headers, seeking should work by calculating the target packet #, seeking to it, (making sure we are at a packet boundary - only an issue for corrupt
18:58:24linuxstbed files), and then adjust.
18:59:35preglowbut yeah, i'll be surprised if we don't need some kind of resync
18:59:39preglowshouldn't be a big deal anyway
18:59:45preglowassuming it's documented...
18:59:50linuxstbBut why should you (if the stream isn't corrupt) ?
19:00:34preglowwell, unless i've misunderstood something, you'll need to when you estimate your seek position and hit between packets thanks to vbr
19:00:54linuxstbpackets are always a fixed size in bytes.
19:01:03linuxstbIt's the number of samples they contain that varies.
19:01:58preglowthat'll simplify stuff
19:02:10preglowis that a common arrangement elsewhere?
19:02:34linuxstbI don't think so.
19:02:38preglowno, me neither
19:02:51linuxstbMost formats are fixed in terms of samples, and hence variable in terms of bytes.
19:03:53preglowyeah, or like mp3, fixed in both
19:04:14 Join webguest85 [0] (i=bd9cea8a@gateway/web/cgi-irc/
19:04:22w1ll14mcan some one explain a little or a webpage that does about how to write a specific bit of options to a hardware register in C?
19:05:33linuxstbw1ll14m: Assuming the hardware register is memory mapped, then look at a file like firmware/export/pp5020.h in the Rockbox source.
19:06:04webguest85Hi,I've been registered on the TWiki and I want access to edit make nor modify
19:06:53Febswebguest85: what is your TWiki user name?
19:07:14linuxstbw1ll14m: And then look at something like firmware/target/arm/system-pp502x.c for examples of how to use those definitions.
19:07:38 Join perrikwp [0] (n=chatzill@
19:07:51 Join Guile`` [0] (n=Guile@
19:09:30w1ll14mlinuxstb: thanx!
19:10:06Febswebguest85: you should now be able to edit the wiki.
19:10:27 Join petur [0] (n=petur@rockbox/developer/petur)
19:10:56webguest85Febs,how do I log on in IRC?,I dont have an IRC client,do you know about one wich is free?
19:11:25FebsI use Chatzilla, but there are plenty out there.
19:11:59webguest85Thanks a lot Febs
19:12:21FebsIf you just want to use a different nickname, you can type /nick TheNickNameYouWant
19:12:46 Quit webguest85 ("CGI:IRC")
19:19:26amiconnlinuxstb: Did you already try to understand the timestamp formats? Are there example files somewhere?
19:22:31amiconnperrikwp: Do you wanna test something on mini G1? Rockbox on mini G1 doesn't do backlight fading yet 'cause nobody tested it, but the min G1 backlight is almost the same as the mini G2 backlight
19:22:56amiconnSo the test would just mean to enable the feature in the platform config, compile and test
19:23:09amiconnIf it works, it can be committed
19:24:07 Join personal [0] (n=chatzill@
19:24:08bluebrotherpixelma: URW Palladio L according to inkscape
19:24:38perrikwpamiconn: sure , but I have to go check something first, I'll be back in about 10 mins
19:24:44 Part personal
19:26:09 Join personal [0] (n=chatzill@
19:26:18pixelmawell not to mine (it took a sans-serif one) - looks like it took one without notifying that it didn't find the one specified in the file... but I found "Palatino linotype" on my box which looks close similar enough
19:26:20 Quit personal (Client Quit)
19:27:28pixelmabluebrother: now to find out why I get a rotated pdf when running epstopdf...
19:27:45bluebrotherI guess that Palladio L is simply a replacement font for Palatino
19:27:47 Join X86__ [0] (n=chatzill@
19:28:12pixelmaI believe tucoz told me so too
19:28:28bluebrotherhave you checked the page orientation? It should be portrait.
19:28:45 Quit Nico_P (Remote closed the connection)
19:30:58X86__how is going the progress of the sansa rahpsody port?,I'm looking for how to get rockbox on it
19:31:10bluebrotherthere is no progess.
19:31:25bluebrotherread the thread in the new ports forums, that's all that's going on.
19:33:15 Quit Rick (Read error: 104 (Connection reset by peer))
19:33:45 Join Rick [0] (
19:33:58 Quit X86__ ("ChatZilla [Firefox]")
19:35:36pixelmabluebrother: in the svg properties?
19:36:00bluebrotherI checked File / Document properties in inkscape
19:37:08pixelmabluebrother: yes it's correct there - and the eps also looks right
19:37:30perrikwpamiconn: ok, I'm back
19:40:46 Quit Rick (Read error: 104 (Connection reset by peer))
19:40:58 Join Wofl [0] (
19:41:21amiconnperrikwp: Open firmware/export/config-ipodmini2g.h and firmware/export/config-ipodmini.h in a text editor. Mark lines 75..77 in the former (they contain #define HAVE_BACKLIGHT_PWM_FADING), copy them, and paste them at line 75 into the latter.
19:41:38amiconnBuild for mini G1, 'make zip', and install the zip (not just rockbox.ipod)
19:41:49 Join Rick [0] (
19:42:19amiconnThen go to Settings->General Settings->Display->LCD Settings
19:43:08amiconnYou'll find 2 new options for fade-in time and fade-out time. Set one of the larger timeouts for each, and check whether the backlight fades in/out as intended
19:44:06amiconn(the reason for installing the full zip is that the .lng file will change, and if you just install the binary, the menu item text may become mixed up)
19:45:01amiconnVoice *will* be mixed up. You'll either need to build your own voice file, or fetch a new one tomorrow
19:46:41perrikwpi don't use voice
19:48:47perrikwpbut I just got a compiler error "settings_list.c:614 error: 'LANG_BACKLIGHT_FADE_IN' undeclared here"
19:49:12amiconnOh, hmm, then you need to make clean as well
19:49:41amiconnlang.h isn't regenerated automatically
19:49:56*amiconn thinks he knows why backlight fading might not have been enabled earlier
19:50:01perrikwpnow it's compiling
19:50:40amiconnIt uses lots of interrupts, and interrupts stopping to work was the effect of the PP5020 bug, which is fixed for a while now
19:55:10perrikwpit works
19:55:31perrikwpfades in and out perfectly
19:55:40amiconnAs expected :)
19:55:55amiconnI'll commit then
19:56:34 Join XavierGr [0] (
19:57:06 Quit Rick (Read error: 104 (Connection reset by peer))
19:57:28pixelmabluebrother: even the data about the bounding box seems right, it seems to just decide that the final pdf should be portrait instead of landscape (because it's wider than high with the text) - wonder if I can suppress that somehow...
19:57:45amiconnperrikwp: Thanks for testing :)
19:57:54 Quit w1ll14m (Read error: 110 (Connection timed out))
19:58:13perrikwpamiconn: your welcome, glad to help anytime!
19:58:18bluebrotherhmm −− sounds strange. I guess it's not related the viewer?
19:58:20pixelmabluebrother: maybe then the document's properties should be set to landscape too =)
19:58:29bluebrothermight help ;-)
19:58:57***Saving seen data "./dancer.seen"
19:59:08 Join Rick [0] (
20:00:37pixelmahmm, it didn't help
20:00:51bluebrothertoo bad :(
20:08:58 Join Gibbed [0] (
20:09:02 Quit Rick (Nick collision from services.)
20:09:16 Nick Gibbed is now known as Rick (
20:10:45 Quit Febs ("ChatZilla [Firefox]")
20:14:22nerochiarosaratoga: still around ?
20:15:29 Quit perrikwp ("ChatZilla [Firefox]")
20:15:43 Join w1ll14m [0] (
20:28:39 Join Sivart0 [0] (
20:31:05 Quit Sivart0 (Client Quit)
20:32:45saratoganerochiaro: am now, whats up
20:34:09nerochiarosaratoga: it's something probably very silly, but at the moment can't figure it out. i've separated the decoder in a stand alone library, using libasf as parser, and i manage to decode wma files. i dump the raw wave data to disk, then try to play it, but it's always at the wrong speed
20:34:40nerochiaroi'm surely calculating wrong the amount of decoded data, or writing it down wrong
20:34:49*petur watches the calendar and realizes GSoC ends in a week...
20:35:54nerochiarosaratoga: when i decode with your standalone decoder, the file is ok
20:36:15Domonokyin a week ? how are our students ?
20:37:05Domonokyit seem the TTS Project wasnt a success ..
20:38:50 Join robin_0800 [0] (
20:39:08 Join perrikwp [0] (n=chatzill@
20:40:47peturusb isn't running well too :/
20:42:19saratoganerochiaro: sounds like you're assuming a certain sampling rate
20:42:30saratogayour software will just use whatever is in the wave header
20:42:37saratogamake sure you're writing the correct value
20:43:48nerochiarosaratoga: i'm not even writing the header for this test, just using aplay and forcing the sample rate that i see in the waveformatext (which i know is correct because it's the one you display in your standalone decoder for the same file)
20:44:01nerochiarosaratoga: just dumping raw pcm data
20:45:01saratogaare you forcing 16 bit and stereo?
20:45:12nerochiaroaplay -r 44100 -c 2 -f S16_LE
20:45:43saratogais it too fast or too slow
20:45:43nerochiarothe test file is the g_128k.wma that i found on your site
20:45:49nerochiaromuch too fast
20:46:20 Quit robin0800 (Read error: 110 (Connection timed out))
20:46:21 Nick robin_0800 is now known as robin0800 (
20:47:10saratogai have no idea then
20:47:24nerochiarosaratoga: well, ok, i'll figure it out. thanks anyway
20:47:42 Join saratoga5 [0] (i=98039a34@gateway/web/cgi-irc/
20:47:42 Quit saratoga ("CGI:IRC (EOF)")
20:47:49preglowhow's MoB doing?
20:49:14saratoga5i always just used the wav write function in main.c to wrap things in a wav container
20:49:19saratoga5then i played them with foobar
20:51:28 Quit perrikwp ("ChatZilla [Firefox]")
20:54:35 Quit BigMac (Remote closed the connection)
20:54:35 Quit feisar (Read error: 104 (Connection reset by peer))
20:57:27 Join feisar [0] (
20:59:04LearRegarding wma, any suggestions for small and simple encoder program?
20:59:04 Quit EbErT ()
21:00:21saratoga5Lear: on Windows?
21:00:23 Join Febs [0] (n=chatzill@
21:00:55saratoga5i just use the windows media encoder program
21:03:20LearSeems a bit overkill for my needs (just a small test file or two), but I guess that'll do.
21:04:03saratoga5theres a command line version of it too i guess
21:04:51 Quit XavierGr ()
21:05:07 Join XavierGr [0] (
21:09:39 Quit Wofl (Remote closed the connection)
21:10:30 Join Wofl [0] (
21:11:38linuxstbamiconn: I added some code to parse the asf timestamps and some DEBUGF lines to display them, but didn't get much further. I think I gave saratoga that patch.
21:12:59 Join BigMac [0] (
21:13:10 Join tumu [0] (n=tumu@
21:22:19 Join tucoz [0] (
21:22:48tucozpixelma, are you struggling with the svg->pdf conversion?
21:23:15amiconnSeems cygwin's epstopdf is too old
21:23:46amiconnJust replaced it with the latest from debian (it's a perl script)
21:23:56GodEater_linuxstb: had a nice holiday (assuming that's where you've been) ?
21:24:00 Join tihoc4n [0] (
21:24:30tucozamiconn, i see.
21:24:45nerochiarosaratoga5: i've added the wave header and also managed to find out how to write the correct amount of data so it has the correct speed. so it mostly works
21:25:02amiconn...which doesn't work because gs was renamed
21:25:04linuxstbGodEater_: Yes thanks ;)
21:25:15GodEater_go anywhere nice ?
21:25:30pixelmatucoz: at least I already got the png :] - but I really want to able to do those myself in the future
21:25:39linuxstbGodEater_: Egypt - Cairo and a cruise up and down the Nile.
21:25:53GodEater_oooh - very ice
21:25:56GodEater_nice even
21:26:03tucozpixelma, yes of course. so you got it working now?
21:26:16*pixelma 's been there too - seen no ice ;)
21:26:17nerochiarosaratoga5: last problem i'm having, is that it has an annoying "clipping" sound and sounds a lot muddier than the output of your decoder, and lower volume
21:26:24nerochiarosaratoga5: something is surely up
21:26:26pixelmatucoz: not yet
21:26:48tucoznot even with the correct epstopdf?
21:26:51linuxstbpixelma: Yes, "ice" is the last word I would use to describe Egypt - temperatures around 40 degrees...
21:27:25*GodEater_ really must get a new keyboard at home
21:27:28pixelmatucoz: well the newer epstopdf doesn't run for the reason amiconn stated above
21:27:28linuxstb(which is apparently a cold summer)
21:27:59tucozpixelma, ah. i thought that was the problem with the older version..
21:28:36pixelmatucoz: no the problem with the older version is, that it has a "nice" auto-rotate "feature"
21:29:35tucozhow nice of them to add such a gem
21:29:36saratoga5nerochiaro: what did you change to make that happen?
21:29:52 Quit tihoc4n ("Leaving.")
21:30:42nerochiarosaratoga5: i was calculating wrong the decoded data size. i was forgetting to take into account that it was 16 bits per sample, so i was copying only half of it
21:31:03saratoga5the clipping thing though
21:31:29saratoga5what exactly have you done?
21:31:37nerochiarosaratoga5: (if you want i can upload the wave somewhere)
21:31:45nerochiaroand the source
21:32:14saratoga5well is it much different then my test program?
21:32:23saratoga5not really sure what you've been working on doing
21:33:10nerochiarosaratoga5: well, it's pretty similar to yours, but it uses the same wma decoder that is in rockbox, thus it decodes frame by frame, not superframe by superframe like yours
21:33:52saratoga5so you've just taken the wmadeci.c from rockbox and combined it with the asf parser from my test program?
21:34:08pixelmatucoz: got it now - not rotated :)
21:34:12nerochiaroyes, but i combined it with libasf instead
21:34:25tucozpixelma, great. what did you do?
21:34:26nerochiarosaratoga5: just to give you the idea
21:34:50nerochiarosaratoga5: well, given that the asf parser in your program is libasf too, the answer is "yes" :)
21:35:01amiconntucoz: Newer epstopdf versions want to use 'gswin32c' under cygwin, but the correct name is just 'gs' like under *nix
21:35:28pixelmatucoz: amiconn helped me and can probably explain that better - maybe I should document that somehow in the wiki? Worth it?
21:35:30amiconnSo the line that sets the binary name for cygwin needs to be commented out in newer epstopdf versions
21:36:11tucozpixelma, why not? like a note for cygwin users. Just to get it written down somewhere.
21:36:36amiconnCygwin is really in need of updating of some packages
21:36:45amiconnThe packaged epstopdf is from 2001....
21:37:02tucozamiconn, the same is mine for ubuntu
21:37:18tucozEPSTOPDF 2.7, 2001/03/05
21:37:39Learlinuxstb: Any idea what the ASF parser is reading as bitrate? Doesn't seem to be bitrate at least...
21:38:02saratoga5Lear: isn't it bits/sample?
21:38:07amiconntucoz: And that one has autorotate. That 'feature' was removed in 2004
21:38:12saratoga5i know thats what the wma decoder uses
21:38:28LearNo, that comes later.
21:38:32pixelmatucoz: it's just that I don't know if the average cygwin user is able to get hands on a newer epstopdf (I got one from amiconn's linux box)
21:38:38Lear(After blockalign.)
21:38:39linuxstbLear: I think it's bytes/second.
21:38:43amiconnDebian unstable has epstopdf from 2.9.5gw from 2006
21:39:14linuxstbLear: Hence the "bitrate *=8;" line.
21:39:32tucozamiconn, weird. i converted lots of pixelma's svg's, and they were not autorotated in any way. maybe it only rotates them under certain conditions
21:39:58LearTried a parser of my own (based on asf.c), that returns 64 kbps, but Foobar (and Explorer) says 40 kbps...
21:40:06LearThere's a separate bitrate object, it seems.
21:40:07peturBagder: the build page was acting up a bit: after it was 3 minutes over time, it started reloading the page again and again
21:40:11tucozpixelma, i see.
21:40:33pixelmatucoz: probably when the width > height
21:40:43pixelmawhich was the case here
21:41:07tucozbut e.g. the rockbox logo worked fine
21:41:22pixelmahmm... very weird then
21:41:37amiconnpixelma: Just attach it to the wiki page. It's bsd licensed afaics
21:41:51linuxstbLear: That's odd... Presumably Rockbox displays 64kbps for that file?
21:42:03tucozyes. maybe it is some windows specific problem. i see there are some windows checks in the source for epstopdf
21:42:30 Join bdgraue [0] (
21:43:04amiconntucoz: The win32 checks don't trigger on cygwin
21:43:15amiconnThe cygwin checks do, however
21:43:17LearI'll try to get a look. It is only 1 sec long; maybe that's the problem, the overhead is large, and asf.c includes the overhead, whereas Foobar does not.
21:43:50nerochiarosaratoga5: that's how the decoded file sounds here:
21:44:07linuxstbLear: What do you mean by overhead? You mean it calculates the bitrate based on filesize and file duration, rather than reading it from the header?
21:44:15LearAnd MP3Tag says 46. Figures.
21:44:38LearOverhead, as in tag headers and whatnot, compared to actual audio data.
21:45:46pixelmaamiconn, tucoz: I will update the wiki later, first I want to know if a 2nd gen Ipod manual will have the drawing now (and how it'll look) and then commit :)
21:46:33tucozcool. look forward to seing that
21:46:44pixelmatucoz: do I need to enable something else?
21:47:35tucoziirc, it's just eps->pdf and it should be fine
21:47:36amiconnpetur: I thought some more about the events which need a reply.
21:47:57pixelmaI'll just try
21:48:22amiconnSince there are a few, not just one, but not may either, what do you think about putting a function to register for a certain event into the respective module that processes the replies?
21:48:47tucozpixelma, just name it like ipod2g-front.pdf
21:49:23tucozand ipod2g-front.png (or whatever the "environment" variable for ipod2g is)
21:49:24amiconnE.g. for usb, usb.c would have a function usb_register_reply(bool enable), which makes the usb module increment or decrement a counter how many replies to expect
21:49:58peturso no more broadcasting
21:49:59amiconnEach thread that needs to react and reply would call that function immediately after creating its queue
21:50:12pixelmatucoz: I thought it should be named "ipod1g2g"
21:50:14amiconnNo, the events would still be broadcast to all public queues
21:50:37tucozpixelma, then that is probably right.
21:50:40peturah right
21:50:43 Join Wofl_ [0] (
21:50:49amiconnBut threads with a public queue which don't need to do something special wouldn't register. They would still receive the event, but not need to reply
21:51:55saratoga5nerochiaro: that file is just static
21:51:58amiconnE.g. the backlight thread needs to register for power events (SYS_POWEROFF/SYS_SUSPEND/SYS_WAKEUP), but not usb
21:52:09 Join homielowe [0] (n=chatzill@
21:52:23petursounds like a way to do it...
21:52:28amiconnIt could still react to usb if wanted and e.g. make the backlight light up like if a button was pressed
21:52:46 Quit miepchen^schlaf ("Verlassend")
21:53:12homielowedionoea: around?
21:53:18amiconnThe question how to handle things in ata.c remains though...
21:53:46pixelmatucoz: yes, everything's in place now :)
21:54:06nerochiarosaratoga5: something went wrong with the upload then, it definitely plays here
21:54:08peturamiconn: but the threads involved will need carefull coding to get the registering and notifying in line. Could turn out ugly if many threads need this code
21:54:29peturmaybe something can be put in the default handler?
21:54:46amiconnThe default handler is for the main thread only
21:55:35amiconnEach thread that does need to reply to such events needs 2 additional lines of code (one for registering one for deregistering)
21:56:17peturwell if the registering is per module, you'd need a call per module, right?
21:56:22nerochiarosaratoga5: actually, it's probably something either with your download or with the player you use. i downloaded it here again and it plays
21:56:28nerochiarosaratoga5: definitely not static
21:56:33amiconnIn turn, all threads which don't need special handling can have two case:s from their switch removed
21:57:51amiconnpetur: It's per public queue, and one thread never has more than one queue. It might also have none, or one that isn't public
21:58:42amiconnRight now there are 16 threads (there can't be any target where all of them are present) with public queues
21:58:58amiconnAll of them have to reply to USB events in the svn version
21:59:00***Saving seen data "./dancer.seen"
21:59:01peturwell you would have to register and notify both usb and powermanagement or am I missing something
21:59:13amiconnHowever, most of them just say 'got it, thanks', and go on
21:59:16GodEater_linuxstb: you feeling generous enough to point this person at some info on how to set up OSX to develop with properly? Even if the stuff is already installed?
21:59:29petur(in case both are needed of course)
21:59:44amiconnThese threads would neither have to register nor handle them anymore
22:00:13amiconnOnly those threads which really need to do some housekeeping would need to register
22:00:28 Quit Wofl (Read error: 110 (Connection timed out))
22:00:48amiconnBtw, there's a bug in svn related to usb handling and voiced splashes
22:00:48peturand threads that should (for example) no longer try to access the disk...
22:01:13 Quit BigMac (Remote closed the connection)
22:01:17 Join Wofl [0] (
22:01:40 Quit Wofl (Remote closed the connection)
22:01:48amiconnIf you plug usb and nothing was voiced before, the voice thread tries to load mpa.codec... after the ata thread released ata
22:02:16peturregistration would solve that...
22:02:37amiconn-> A (non-voiced) error splash says 'Couldn't load /..../mpa.codec'
22:03:09amiconnNot on its own, that's my other problem
22:03:41 Join Wofl [0] (
22:03:45amiconnThe ata thread can't release the disk before other threads which need the disk for their housekeeping are done
22:04:19amiconnRight now I have no real idea how to handle this
22:05:17peturusb knows when all threads are notified?
22:05:30 Join robin_0800 [0] (n=robin080@
22:05:35amiconnI.e. the ata thread is special. It is not the event initiator (that's either the usb thread, or in future also the power management thread), but it needs to know when all others are ready
22:05:36*petur doesn't know how this internal part works
22:06:01peturamiconn: usb/power could call ata?
22:06:34amiconnWhen the usb thread detects a connection, it broadcasts SYS_USB_CONNECTED. queue_broadcast() returns the number of queues it posted to
22:07:02amiconnThen the USB thread sits and waits for (n-1) replies to come in before enabling the usb bridge
22:07:23amiconnSame thing on disconnect, just reverse
22:07:48amiconn...and that the usb thread only expects as many disconnect acks as it got connect acks before
22:07:59 Join BigMac [0] (
22:08:33GodEater_why not just get the usb thread to only message ata
22:08:38peturso usb can call ata when all replies are in, no?
22:08:39amiconnHmm, that also means disconnect might have a similar problem. ATA must be back in business before all others do
22:08:44GodEater_and let ata issue the SYS_USB_CONNECTED
22:08:50GodEater_and wait for the n-1 replies
22:09:19GodEater_and then tell usb it's done ?
22:09:43GodEater_only I assume it would be n-2 replies
22:10:05*GodEater_ caveats this may be completely naive
22:10:13GodEater_it's just what I thought of
22:10:25amiconnThe n would change with the registration system
22:10:50*amiconn needs to think this over. It might be the solution :)
22:12:13amiconnThis could perhaps be implemented by making the ata queue private, and provide hook functions to post to it
22:12:46amiconnThe USB would first broadcast SYS_USB_CONNECTED, collect the replies, the post privately to ata, and when that replied as well finally enable usb
22:12:51 Quit Wofl_ (Read error: 110 (Connection timed out))
22:13:25amiconnAnd the reverse on disconnect: disable usb, post privately to ata, wait for reply, then broadcast, and wait for the others to reply
22:13:50GodEater_are there any situations where ata has to be aware of other global events
22:14:36amiconnThe same wiould then be needed for powermgmt
22:14:51peturor add functions to ata to let usb control it?
22:15:23GodEater_that would require less fundamental changes as I understand it yes ?
22:15:56GodEater_but perhaps not the ideal solution
22:17:24 Quit BigMac (Remote closed the connection)
22:18:09 Join BigMac [0] (
22:19:30chrisjs169what's button_rel do?
22:20:10rasherBagder: does user: have any meaning in lang files, and is it safe to add one if it's not there already?
22:20:18 Quit robin0800 (Read error: 110 (Connection timed out))
22:20:18 Nick robin_0800 is now known as robin0800 (n=robin080@
22:20:21Bagderit has no current meaning, no
22:20:37Bagderso yes, you can add one or just ignore that until we actually start using it
22:21:11rasherI'm working on the web-translating thing again, and to avoid more special-cases, I'm just going to add an empty user line when writing out the result
22:22:05 Join advcomp2019- [0] (n=advcomp2@
22:22:07rasherMost strings have one anyway.
22:22:11 Quit advcomp2019 (Nick collision from services.)
22:22:15 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
22:23:14amiconnrasher: user: is meant for when localisation/voicing of plugin gets added
22:23:41rasherAh, right
22:24:12amiconnEmpty means core
22:25:47 Quit homielowe (Read error: 110 (Connection timed out))
22:25:53amiconnHmm, the ata thread doesn't need to do much on usb connect, but it will need to do a different thing in case of poweroff/suspend
22:26:17amiconnOn usb it needs to power on the hdd, on suspend it needs to power off the hdd
22:27:11*rasher wonders why LANG_DECAY isn't voiced
22:27:29 Join IA4004 [0] (n=chatzill@
22:28:51 Join dan_a [0] (n=dan_a@
22:29:31 Quit dan_a (Client Quit)
22:29:41pixelmaisn't that Archos (masf) only?
22:30:14rasherBut there's no english voice string for it
22:30:25rasherJust masf: ""
22:30:35pixelmait doesn't make much sense to voice so much on hwcodec, since it can't be played during music playback anyways
22:30:35chrisjs169what's button_rel do?
22:31:32rasherpixelma: can't the setting be changed while playback is stopped?
22:33:24pixelmait probably can but I usually don't change sound settings without listening to them at the same time, don't know if it's worth it
22:33:36 Part tucoz ("Leaving")
22:34:12rasherpixelma: but it looks to me like the rest of the mas settings are voiced
22:34:58amiconnThis ID seems to be unused
22:35:05amiconnBefore JdGordon'
22:35:23amiconns settings rework, some settings had different titles than their menu item
22:36:16amiconnI.e. 'Auto Volume' had 'AV Decay Time' as its title. Now it's also 'Auto Volume'
22:36:23pixelmarasher: maybe you are right
22:36:26rasherAh, so this is an ID that was used only as a title?
22:36:28amiconnThat should be fixed, but the string doesn't need to be voiced
22:36:49Bagdertime to add a comment explaing that? ;-)
22:37:28amiconnThe possible settings are Off, 20ms, 2, 4 and 8 (the latter are seconds)
22:37:45amiconnHence 'time'
22:38:55amiconnHmm. In the USB case, the ata thread shouldn't block others from accessing the disk (since it only powers on the disk)
22:39:38*amiconn doesn't want to get a multi-threading headache again ;(
22:39:38pixelmapetur: are you going to explain the new setting in the manual? ;)
22:41:27 Quit Domonoky (Read error: 104 (Connection reset by peer))
22:43:12 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:43:26 Quit amiconn (Nick collision from services.)
22:43:26 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:43:45*petur returns
22:43:51amiconnBah, now I have to adapt 2 settings instead of one when changing file view....
22:44:15preglowamiconn: why hasn't pwm backlight been enabled before now?
22:44:19preglowon minig1, that is
22:44:21rasherMan, wasn't it possible to change all these language-strings automatically when adding the "features" thing?
22:44:42amiconnpreglow: Nobody tested it...
22:45:01amiconnAnd I think it would have made the freezing on PP5020 a lot more frequent
22:45:11preglowvery likely
22:45:21amiconnNow that it doesn't freeze anymore, it works like a charm
22:47:18 Join robin_0800 [0] (
22:47:40amiconnMy theory is that the freezing happened because some operations for the clock change happened in the wrong order, causing some invalid intermediate states. Then interrupts can make it worse because the can happen anytime, even during such an invalid state, making it last longer
22:48:14peturamiconn: you want a 4th file extension option (being act like it did in the past)?
22:48:43preglowamiconn: doesn't some nanos still have trouble?
22:49:23*petur reads a tex file for the first time in his life
22:50:50bluebrotherpetur: and, got scared?
22:51:32peturjust scared of editing without building... I'll have to see if I can build a manual first
22:52:00 Quit saratoga5 ("CGI:IRC")
22:55:15 Join DerPapst [0] (
22:56:02*bluebrother still has that tex file around that scared JdGordon :P
22:56:27*petur already failed applying the patch for ubuntu :(
22:56:37bluebrotherwhat patch?
22:57:28bluebrotherpetur: that isn't needed anymore. Will update the wiki ...
22:57:54 Join donutman25 [0] (n=chatzill@
22:57:57 Quit merbanan (Remote closed the connection)
22:58:45 Quit desowin ("use linux")
22:58:50bluebrothersection edits in the wiki would be really a nice addon.
22:59:20*DerPapst loves mediawiki
23:00:25pixelmabluebrother: what settings did you use for the scaled down pngs that are now used in the build table and so on?
23:00:30bluebrotherso, let's see how much of the installation I broke in rbutil ...
23:01:08*pixelma should probably just look at their properties
23:01:30bluebrotherpixelma: I used the default size and scaled them down with convert -scale 80x80 <file>
23:02:01pixelmaI see, thanks
23:02:31 Quit robin0800 (Read error: 110 (Connection timed out))
23:02:32 Nick robin_0800 is now known as robin0800 (
23:07:17rasherIs anyone working on converting screens to user-font to get rid of the nasty LANG_SYSFONT hack?
23:07:53*petur does his first manual commit dance
23:08:17Learlinuxstb: I think asf.c reads send_time rather than play_time for id3->length...
23:08:53 Quit spiorf (Remote closed the connection)
23:09:22bluebrotherpetur: congrats :)
23:09:39pixelmaBagder: you around?
23:09:42LearEven then, Foobar seems to get better values somehow...
23:09:47peturI kept it simple, let me know if you're not happy with it
23:09:51*Bagder goes beep
23:10:19linuxstbLear: There's also a "pre-roll" value that needs to be subtracted. That helps...
23:11:07LearAs in lead_trim?
23:12:23linuxstbMaybe - it's the field after send_duration. But I don't think asf.c is reading send_duration...
23:13:13LearAccording to ffmpeg's asf.c, the order is send_duration, play_time, preroll, and RB seems to read send_time.
23:13:39pixelmapetur: many thanks! (we have a few undocumented features at the moment)
23:13:56peturwent fairly smooth
23:14:41linuxstbLear: According to libasf (my reference), the order is play_duration, send_duration, preroll
23:15:11linuxstbMSN supports that -
23:15:44linuxstbSorry, MSDN...
23:16:40LearAnd ffmpeg only reads preroll as 32 bits.
23:17:19peturlinuxstb: any movement in ?
23:18:04linuxstbLear: But ffmpeg is using send_duration to calculate the file duration - so it's using the first value anyway.
23:20:13linuxstbpetur: Not from me... I think amiconn's fixes (which IIUC have removed most long calls) probably make it obsolete.
23:20:37LearHrmph, that test file I sent you has a preroll larger than play duration...
23:21:19peturlinuxstb: should I close it?
23:22:55 Quit davina (Read error: 104 (Connection reset by peer))
23:23:09LearWait a sec, didn't accound for change of order. Read send duration rather than pre-roll... :)
23:23:47 Join davina [0] (
23:33:23 Join Keripo [0] (
23:40:20 Join skazi [0] (
23:40:25rasherCan diffs (-u) have comments?
23:40:54skaziis there some site listing all available plugins for rockbox or is there just the bluechip's rockbox plugins site?
23:41:07Bagderrasher: yes, patch ignores everything before what looks like a patch
23:41:28rasherBagder: Ah, so I can just put my comments in cleartext at the top?
23:41:31bluebrotherskazi: the manual holds descriptions for the plugins ...
23:41:36Bagderrasher: exactly
23:41:49skazirasher: thanks!
23:41:54 Quit skazi (Client Quit)
23:42:16 Part Keripo
23:42:41 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
23:43:11bluebrotherseems like exact the theme zips I'm testing with are broken :(
23:44:56 Quit Lear ("ChatZilla [Firefox]")
23:49:07 Join nick89 [0] (
23:50:13 Quit davina ("xchat on Ubuntu 7.04")
23:55:04rasherHrm, genlang -u is a bit picky about whitespace, it seems
23:55:15rasherit doesn't think " *: none" is the same as " *: none"
23:55:42rasherAt least, I'm pretty sure that's what it's telling me
23:55:51Bagderbut yeah, I think that's because it simply compares two strings basically
23:57:10rasherI thought it would be a bit more clever. At least split it at the first ":", trim whitespace off each side and THEN compare.
23:57:34Bagderlame author clearly...
23:57:58 Join Llorean [0] (
23:58:32 Join tihoc4n [0] (

Previous day | Next day