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

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

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

#rockbox log for 2004-08-18

00:00:02electronic50sorry is the lame decoder
00:00:13zeekoeyou don't need that either
00:00:15zeekoejust copy
00:00:44zeekoeyou should have an executable called avi2wav somewhere
00:00:49zeekoeyou can try that
00:00:53LinusNtime to sleep, nite all
00:00:59zeekoenite LinusN
00:01:10 Quit invistispy ("—I-n-v-i-s-i-o-n— 2.0 Build 2520")
00:01:12 Part LinusN
00:01:41electronic50ok thanks bye
00:01:42 Quit electronic50 ("  X-CRiPt 5.1   ")
00:02:26zeekoeyay, finally alone :-P
00:02:52s0cksis levin actually rob levin?
00:03:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:03:10*s0cks hates midk soo much.
00:04:35 Join scott666_ [0] (
00:17:50s0ckswith fame and glory...
00:29:02 Join midk [0] (
00:29:13 Quit uski (Remote closed the connection)
00:56:05 Quit Nibbler (Read error: 104 (Connection reset by peer))
00:56:10s0cksWhats up people?
01:00:43 Part s0cks ("X-Chat [2.0.10c] Quit.")
01:13:40 Quit zeekoe (Read error: 110 (Connection timed out))
01:24:03***Saving seen data "./dancer.seen"
01:51:47 Nick StrathAFK is now known as strath (
01:52:27midksup strath?
01:52:52strathhe's *everywhere*
01:52:59midkwhere else might you be
01:53:45midki only get to terrorize you in four channels?
01:53:46midknot fair!
01:54:10midknot even in #strath. :[
01:54:39strathi don't think that channel would be too popular :)
01:54:43midkstrath, check out #strath, i made you a surprise topic
02:26:28 Join Smooth [0] (
02:30:14 Quit scott666_ (Read error: 104 (Connection reset by peer))
02:30:32midkuh, HI
02:30:55 Join scott666_ [0] (
02:31:50jdeonhmm, when is Linus coming back from vacation? been two weeks now my patch has been stale
02:32:38 Quit Smooth ("leafChat IRC client:")
02:32:45 Join Nibbler [0] (
02:34:46 Join cabr [0] (
02:34:55cabrhello there
02:36:49cabrdoes anyone out here know if and how I can edit the .voice files? I'd like to create one myself...
02:37:24jdeonyeah, I think there is something in the docs on that
02:37:39cabrdidn't find something in the twiki
02:37:49amiconncabr: You can't edit them (easily), but if you have a tts engine you can build your own.
02:38:27amiconnUnfortunately there is no wiki entry describing how to do that :(
02:38:54cabrI'd like to do it the old way with a mic and a human voice ;)
02:39:25amiconnThis is also possible
02:39:56cabrare there any docs?
02:40:50*amiconn looks around
02:41:00amiconnSomeone should write them...
02:41:45cabrI know. open source ;)
02:41:57jdeon'my code is self-documenting'
02:42:16jdeon'if it was hard to write, it should be hard to understand'
02:43:31cabrthe - files are renamed mp3-files. but where's the key to the .voice files?
02:43:45 Quit mecraw_ ("Trillian (")
02:44:50amiconncabr: A voice file contains many mp3 clips (not in raw form, but bitswapped) and an index table for accessing the clips
02:46:00amiconnThere is a tool to "compile" a voice file from mp3 clips and the language description file available at [IDC]Dragon's homepage (who implemented the talking menus)
02:46:58 Part strath
02:47:32 Join yeft [0] (
02:47:41yefthello children
02:47:56 Quit edx ()
02:48:32cabrhmm... now I have to find his page
02:49:10yeftheh i wonder when midk wakes up and is all happy to see me
02:49:30cabrthank you
02:50:00yeftmaybe you guys could help me
02:50:21yeft...cause i want to go back to archos...but i'm scared
02:50:30 Join dstar5 [0] (
02:50:37yeftmy ajb6k and my fmr both died cause of shitty soldering
02:50:53yeftyou think i have to worry about the new av series?
02:51:00jdeonhard to say
02:51:08dstar5yeft: fix them
02:51:11jdeonwho knows how things are different these days
02:51:12yefti tried
02:51:15yefti suck at it
02:51:23jdeonthey could be the same old problems, or they could be better
02:51:26dstar5yeft: you could always send them to me though :)
02:51:30midki mean
02:51:32midkhi yeft
02:51:34dstar5i love jukeboxes
02:51:58yeftim definitly considering an AV480...cause the specs..are oh so beautiful
02:53:14yeftbut ive had both a fmr and a jb6k die..
02:53:31cabrI think I'm too stupid for that
02:54:01dstar5ill be more than happy to take them :)
02:54:15yefti think my jukebox is disassembled in a bag somewhere along with the FMR
02:55:49midkwas it yeft that called me grammar bot?
02:56:06midkyeah, probablu
02:56:19yeftit was
02:56:38yefthahah we can go back into the archives to find that late night..
02:56:51midkfeel free
02:56:53yeftheh remember when i almost tried to learn C
02:57:01midkwhat a concept
02:57:44yefti want this av480...but i also want to wait for this beast : to ever surface to the market..
02:58:02yeftbut we've been waiting for the taz for 2 years
02:58:40 Quit midk ("Leaving")
02:58:54cabrthx @ amiconn, I'm going to try the tools now
02:59:26 Part cabr
03:00:29 Join midk [0] (
03:00:33 Quit yeft ("jumping on the bandwagon")
03:00:55 Quit midk (Read error: 54 (Connection reset by peer))
03:00:56 Join midk [0] (
03:01:23 Join yeft [0] (
03:01:29yeftheh im back
03:01:36midkGRAMMAR BOT
03:06:04yeft...midk right now is using to find out if what i just said to him was gramatically incorrect
03:06:44yeftdont deny it
03:08:10midkand.. wtf, what did you say i need to correct?
03:10:03amiconnBig fat mega-commit done :)
03:12:00 Join Archos [0] (
03:12:31midkhi there
03:12:49yefti hate archos
03:13:16Archosi have i problem whit the synch of the VIDEO
03:13:22yeftwell its a love - hate relationship
03:13:54Archosthe video runs too fast
03:14:01Archosin the rockbox
03:14:13midkyeft, "it's"
03:14:44Archoswho can help me
03:14:45scott666_Archos: did you set the fps right when you started converting?
03:15:00 Nick scott666_ is now known as scott666 (
03:15:22Archosthe first time that i converted i forgoted
03:15:38Archosbut the second time i put and now is too slow
03:15:59Archosthe fps of the video is 23.97
03:16:47ArchosC:\Documents and Settings\LUIS CARLOS\Mis documentos\Archos\Rock Video>rvf_mux k
03:16:47Archosb2.rvf kb2.mp3 kb2_av2.rvf -play_fps 23.97
03:16:48 Quit Archos (Client Quit)
03:17:15 Join Archos [0] (
03:17:25 Quit Archos (Client Quit)
03:17:36 Join Archos [0] (
03:18:06kaboofai think i'm going to convert my massive movie collection to rvf
03:18:53Archoshelp me
03:19:16kaboofaWhat's wrong?
03:19:50kaboofayou may need to use external programs other than rvf
03:20:02kaboofalike some video mastering stuff
03:20:07kaboofai havn't worked with archos video
03:20:49 Part amiconn
03:21:08Archoswho is the creator of Rockbox, is here
03:21:18midknot here
03:21:18kaboofathere are a bunch of people that have contributed to rockbox
03:21:21yeftnot here
03:21:23kaboofalinus != here
03:21:29midknot linus, zagor
03:21:30kaboofawhat's bjorn's nick?
03:21:31Archosyes, yes
03:21:31yeftmah ze?
03:21:33kaboofawhatever ;\
03:21:35kaboofamy bad
03:21:41yeftim so confused
03:21:47kaboofai'm just listing people i see in the mailing list a lot ;\
03:21:51Archosbut who has the idea
03:21:55yeftsame here
03:22:05yeftim hungry
03:22:30kaboofaif you e-mail the mailing list, you may get a better responce
03:22:37kaboofabecause not all of the developers are here right now..
03:22:39yefti gotta go
03:22:48yeftlila tov
03:22:58kaboofaich liebe apfels.
03:23:00midklater yeft
03:23:00 Quit yeft ("CGI:IRC (EOF)")
03:23:23kaboofawhat is I see it often for hostmasks in here.
03:23:49midkpeopel running the cgi script
03:24:04midkzagor hosts it, and he uses contactor
03:24:05***Saving seen data "./dancer.seen"
03:25:27 Quit Archos ("CGI:IRC")
03:25:52 Join Archos [0] (
03:26:13ArchosSorry how i can register my nickname
03:26:25midk /ns register <password>
03:26:28dstar5Archos: /msg nickserv register (pword)
03:27:36Archosi'll try it
03:27:42 Quit Archos (Client Quit)
03:28:00midki bet archos is track
03:28:03 Join Archos [0] (
03:28:35kaboofai need to code
03:28:45kaboofaI have a funny story, actually.
03:28:45midkso yeah... uh... code that's it
03:28:58Archosi'm now registered no???
03:29:00kaboofaThis Server company called Echo7 was going to hire me to program.
03:29:07kaboofaBut they will not draw up a contract for me.
03:29:16kaboofaAnd they want me to e-mail them my code, THEN they will pay me.
03:29:21kaboofaSo I told them 'hahaha. no.'
03:29:39kaboofathat was pretty much my reaction
03:29:42 Quit Archos (Client Quit)
03:29:57 Join Archos [0] (
03:30:03 Quit Archos (Client Quit)
03:30:06kaboofaon top of it all, he wanted me to code a huge server admin. software package, and UNDERCUT my friend's server company
03:30:17midkundercut? tell me more
03:30:45kaboofaas in, release my software before they released theirs
03:30:51kaboofafor cheaper
03:30:55kaboofawith more features
03:30:58kaboofai know
03:31:02kaboofafuck them
03:31:03midkwhat idiots
03:31:10kaboofaif any of you buy game servers from e7 i'll rip off your damn head.
03:31:19midkbut why?
03:31:24kaboofano idea
03:31:33midk/ i suck function
03:31:38midkvoid isuck(void)
03:32:35kaboofachomp($n = s/$USERNAME/<STDIN>/)
03:32:39kaboofathat's not right
03:32:44midks/ / / //
03:32:44kaboofafucking fuck fuck
03:32:45midkgod dammit
03:32:50kaboofas/ / /
03:33:29kaboofachomp($n = <STDIN>);
03:33:39kaboofais the best function ever
03:33:40kaboofai mean
03:33:42kaboofait's called CHOMP
03:33:44kaboofaas in
03:33:47kaboofamuch much much yummy
03:34:25kaboofaI think i'm going to make a perl script to send e-mails for me.
03:35:03midksuch random insults as "die, nerd" and "die! nerd
03:35:04kaboofawhy not?
03:35:23kaboofablue == transparent in the game
03:35:39midkwhat si it?
03:35:40kaboofaquestions? comments? concerns? mail it to /dev/null
03:35:49kaboofait's a decal for the game half life
03:35:54kaboofait's the FreeBSD Daemon
03:35:58kaboofathe mascot
03:36:05kaboofaand their slogan "Power to Serve!"
03:36:13midkoh, at this one bookstore, there was a sign "complaints go here" on the trash can lid
03:36:31kaboofaI think I did an OK job on that, esp. since I only had 254 colors to work with
03:36:42midkit's good!
03:36:48kaboofaThanks :)
03:36:59*midk mails
03:37:16kaboofais that a real address?
03:37:33kaboofano one has!!
03:37:47kaboofai lied
03:38:13 Quit dstar5 ("umount /dev/brain")
03:42:20 Join Archos [0] (
03:42:48Archoshow i can compile the Rockbox Source Code with Windows
03:43:08midkit's a bit of work
03:44:33Archoshow much with a 633 mhz computer
03:44:53midkit's not really the speed, more the amount of work involced
03:45:01midki'd suggest you look at the site for docs
03:45:26Archosyes i looked but its only for cygwin and linux
03:45:59midkso.. you need cygwin or linux
03:46:26Archoswhere i can get the linux or cygwin
03:46:39Archosand its free or not
03:46:42midkview the docs, please
03:46:48midkcygwin is for windows
03:47:11 Quit Archos (Client Quit)
03:47:34 Quit midk (Read error: 54 (Connection reset by peer))
03:48:00 Join midk [0] (
03:51:45 Nick midk is now known as midk|bbiab (
04:26:06 Quit Nibbler (Read error: 104 (Connection reset by peer))
05:05:51 Quit Ka_ (
05:06:14NJoinKa_ [0] (
05:06:27 Join strath [0] (
05:06:42straththere ya are :)
05:22:15 Quit midk|bbiab (Read error: 104 (Connection reset by peer))
05:24:08***Saving seen data "./dancer.seen"
05:24:21 Join midk [0] (
05:53:46 Join vol [0] (
05:58:55 Join Nibbler [0] (
06:26:18 Quit midk (Remote closed the connection)
06:26:19 Quit Nibbler (Read error: 104 (Connection reset by peer))
06:26:21 Join midk [0] (
07:24:11***Saving seen data "./dancer.seen"
07:31:54 Quit lImbus (Remote closed the connection)
07:59:37 Join Nibbler [0] (
08:39:16 Quit Nibbler (Read error: 104 (Connection reset by peer))
08:46:57 Join Zagor [242] (
08:47:24midkhey zagor
08:51:17 Join Bagder [241] (
08:51:38midkhey bagder
08:53:01 Join amiconn [0] (
08:53:08midkhey amiconn
08:53:38amiconnhi everyone
08:53:50Bagderhi amiconn
09:00:11*Bagder gets sponsorship from the Rockbox paypal fund for a new Archos
09:02:15 Quit midk (Remote closed the connection)
09:03:04 Join midk [0] (
09:20:58 Quit midk ("just STOP it arspy")
09:21:10 Join midk [0] (
09:24:12***Saving seen data "./dancer.seen"
09:24:40 Quit midk (Client Quit)
09:24:42 Join midk [0] (
09:25:16 Nick midk is now known as midk|away (
09:25:24 Nick midk|away is now known as midk (
09:25:39 Quit midk (Remote closed the connection)
09:28:02 Join midk [0] (
09:29:07amiconnBah, the archos fw is crap.
09:29:21midkamiconn, you're just now realizing this? :]
09:29:26Zagorreally? :)
09:29:36midkare you SURE?
09:30:03amiconnI just wanted to stop my 6h test recording (to see if archos fw records correctly), and it told me harddisk error!
09:30:41midk[evil] hi [/evil] aaaaaaaaaaaaaaaaaaaaahhahahahahahahahahhhh......
09:30:48amiconnI'll see if the secornd try works....
09:31:13midkit won't of course
09:47:26 Nick gromit`` is now known as gromit`zzZZ (
09:51:02 Part scott666
10:17:44 Join Nibbler [0] (
10:18:57 Quit Nibbler (Read error: 104 (Connection reset by peer))
10:25:27 Join Nibbler [0] (
11:19:40 Quit Nibbler (Read error: 104 (Connection reset by peer))
11:24:14***Saving seen data "./dancer.seen"
12:02:52midkbed, nite
12:14:30 Nick midk is now known as midk|sleep (
12:24:27 Quit levin (Connection reset by peer)
12:24:32 Join levin [0] (levin@levin-pdpc.staff.freenode)
12:27:04 Join edx [0] (
12:31:05 Join lImbus [0] (
12:50:09 Join Nibbler [0] (
13:13:58 Join zeekoe [0] (
13:24:17***Saving seen data "./dancer.seen"
13:36:10 Quit zeekoe (
13:36:10 Quit amiconn (
13:36:10 Quit Hadaka (
13:36:10 Quit jdeon (
13:36:10 Quit Hes (
13:36:10 Quit Ka_ (
13:36:10 Quit strath (
13:36:10 Quit vol (
13:36:10 Quit lImbus (
13:36:10 Quit edx (
13:36:10 Quit midk|sleep (
13:36:10 Quit elinenbe (
13:36:10 Quit Bagder (
13:36:10 Quit ze (
13:36:10 Quit dwihno (
13:36:10 Quit Nibbler (
13:36:10 Quit webmind (
13:36:10 Quit gromit`zzZZ (
13:36:10 Quit Zagor (
13:36:10 Quit kaboofa (
13:36:10 Quit levin (
13:36:23NJoinzeekoe [0] (
13:36:23NJoinlImbus [0] (
13:36:23NJoinedx [0] (
13:36:23NJoinlevin [0] (levin@levin-pdpc.staff.freenode)
13:36:23NJoinmidk|sleep [0] (
13:36:23NJoinamiconn [0] (
13:36:23NJoinBagder [241] (
13:36:23NJoinZagor [242] (
13:36:23NJoinvol [0] (
13:36:23NJoinstrath [0] (
13:36:23NJoinKa_ [0] (
13:36:23NJoinelinenbe [0] (
13:36:23NJoingromit`zzZZ [0] (
13:36:23NJoinHadaka [0] (
13:36:23NJoinjdeon [0] (
13:36:23NJoinze [20] (
13:36:23NJoindwihno [0] (~dw@
13:36:23NJoinHes [0] (
13:36:24NJoinwebmind [0] (
13:36:24NJoinkaboofa [0] (
14:00:55Bagderemacs rules
14:01:38Bagderctrl-( control-s // backspace * space control-e */ contrl-) ctrl-u 1000 ctrl-x e
14:01:48Bagdercode-policed the solitaire ;-)
14:02:18Zagori've actually never thought about combining ^u and ^x-e
14:02:30Bagderits very handy
14:02:50zeekoewhat was the problem with solitaire?
14:02:58Zagorc++ comments
14:03:14zeekoeyeah, they say that
14:03:22zeekoewhat are c++ comments?
14:03:23zeekoe/ comment
14:03:28Bagderand a few veeeeery long lines
14:03:32zeekoelike that?
14:03:59zeekoeneed to get icecream
14:08:26 Join StrathAFK [0] (
14:08:30 Quit strath (Read error: 104 (Connection reset by peer))
14:19:23 Part Bagder
14:21:33zeekoeis solitaire in the daily builds yet?
14:21:44zeekoehow is it called?
14:21:58zeekoeyesterday i installed the newest
14:22:32Zagorbrowse plugins
14:23:11zeekoei'll try again
14:23:55zeekoehmm... it's not in
14:24:16zeekoei guess it was compiled a bit too early
14:25:01zeekoei'll compile it myself
14:29:20Zagorthe builds are created in the morning, and thus don't contain the changes added during the day. get 0818.
14:30:24zeekoei already compiled the latest
14:34:22zeekoeminesweeper is nice...
14:45:17 Join sethians [0] (
14:59:14 Join mbr [0] (
15:14:46 Join Nibbler [0] (
15:19:06 Quit sethians ("CGI:IRC (EOF)")
15:23:11 Nick gromit`zzZZ is now known as gromit` (
15:24:21***Saving seen data "./dancer.seen"
15:58:22lImbusanyone here can tell me if in rockbox sleep(0) is the same than yield() ?
15:58:22 Quit Nibbler (Read error: 104 (Connection reset by peer))
15:59:16 Join Strath [0] (
16:18:15 Quit StrathAFK (Read error: 110 (Connection timed out))
16:19:13 Nick Strath is now known as StrathAFK (
16:20:47 Nick StrathAFK is now known as Strath (
16:25:51amiconnlImbus: No, it isn't.
16:34:07 Nick Strath is now known as StrathAFK (
16:34:57lImbusthanks. so probably sleep(0) does simply NOT sleep, and yield() yields
16:35:29 Nick StrathAFK is now known as Strath (
16:36:06amiconnlImbus: sleep(0) does sleep until the next tick (so it sleeps 5 ms on average)
16:37:14 Nick Strath is now known as StrathAFK (
16:37:45StrathAFKsorry for nick hopping.... sorting something out with nickserv
16:37:57StrathAFKok, now really afk
16:58:10 Join [IDC]Dragon [0] (
16:59:04amiconnhi Jörg
17:00:06 Join webguest11 [0] (
17:00:27 Join mecraw_ [0] (~lmarlow@
17:00:48 Quit webguest11 (Client Quit)
17:00:48 Quit [IDC]Dragon (Client Quit)
17:01:03 Join [IDC]Dragon [0] (
17:01:14[IDC]Dragonhi again
17:01:22[IDC]DragonI forgot to use Firefox
17:07:27[IDC]DragonJens, what an impressive const-job!
17:08:02amiconnIt's still not complete, the last part will get done this evening (at least I hope so)
17:08:24amiconnConst'ing the apps was a bit of a nightmare...
17:10:10 Nick gromit` is now known as gromit`aw (
17:10:29 Join dstar5 [0] (
17:12:02dstar5hi [IDC]Dragon
17:19:47dstar5did that cpu board come in?
17:23:48 Join Nibbler [0] (
17:24:25***Saving seen data "./dancer.seen"
17:31:33 Nick StrathAFK is now known as Strath (
17:47:44 Join maikeul [0] (
17:48:38 Quit gromit`aw (Read error: 110 (Connection timed out))
17:52:53[IDC]Dragondstar5: not yet
17:53:21[IDC]Dragonbut as you may have seen, I already offered the board being replaced to uski
17:53:32dstar5yes lol
17:53:46[IDC]DragonI hope it all works out...
17:55:42dstar5good for two people
17:56:14dstar5i think the US is doing new inspetion of mail now
17:56:29dstar5i think it is takinglonger for thuings to be sent
17:56:48dstar5i had to sign this little customs slip saying what was in it, and etc
18:05:28 Join mecraw [0] (~lmarlow@
18:05:28 Quit mecraw_ (Read error: 104 (Connection reset by peer))
18:06:38[IDC]Dragonand what did you say it is? A bomb timer?
18:08:18 Join omes [0] (
18:18:20 Part Zagor
18:18:56[IDC]Dragonamiconn: any plans for font rotation?
18:19:16[IDC]Dragon(I guess you're busy with recording tests)
18:24:31amiconn[IDC]Dragon: Font rotation is on my todo list...
18:24:43dstar5see you
18:24:48 Quit dstar5 ("umount /dev/brain")
18:25:25[IDC]Dragonsorry for being pushy (and passive myself)
18:25:32 Part lImbus
18:25:46amiconn[IDC]Dragon: Any news on the bit corrupting recorder?
18:28:31 Join scott666 [0] (
18:29:53 Quit omes ("leaving")
18:29:56[IDC]Dragonamiconn: no, I sortof have given up
18:35:02amiconn[IDC]Dragon: I found a problem possibly related to your flash bootloader...
18:35:25amiconn...a way how I can reproducibly trigger an ata error -31:
18:36:07amiconn(1) (when box is off): Plug in the charger with F1 pressed - Archos charging screen comes up
18:36:39amiconn(2) Try to start firmware with ON, not holding F1 this time - rockbox tries to boot, then ata error -31
18:39:29zeekoeif we're talking about bugs anyway... i found this one: play a song, press on, wait a few seconds, press off [playing stops], press on, press play [playing resumes from the moment when you pressed on]
18:43:12[IDC]Dragonamiconn: I'm aware of that one, Linus pointed me to it 2 weeks ago or so
18:43:44[IDC]Dragonbefore, I didn't know you can exit the Archos charging screen
18:43:54[IDC]Dragonor, I forgot about it
18:43:57amiconnDo you already have an idea what may cause this?
18:44:15[IDC]Dragonit's the coldstart detect
18:44:46[IDC]Dragonwe're not coldstarting in that case, but still would need to do the full ATA init
18:44:52amiconnThat means starting rockbox from archos charging doesn't get detected as coldstart?
18:45:12[IDC]Dragoncorrect in one way:
18:45:39[IDC]Dragonit's not a coldstart to main init, so the charging screen is skipped
18:45:47[IDC]Dragon(which is OK)
18:46:21[IDC]Dragonbut, on the other hand, parts of the ATA init are skipped as well, which is no good
18:46:38[IDC]Dragonso, I need a better detection there
18:47:02[IDC]Dragonhaven't researched if and how this is possible
18:48:55amiconnIs it possible to check the hd itself if it is initialized?
18:49:51amiconnOr another possibility: simply do the ata init on every boot. Could that break something else?
18:51:16[IDC]Dragonthere's different possibilities, I guess.
18:51:41[IDC]DragonI've just been too lazy to tackle this minor issue.
18:52:05[IDC]DragonOr, in other words, I have no time for Rockbox any more.
18:52:59[IDC]DragonBut I read the group and join here to encourage as much as I can. ;-)
18:54:23[IDC]Dragonto better answer the question: the ATA cold start resets the disk, this is no good for regular start
18:57:21[IDC]Dragonamiconn: I'm looking at the font code now
18:58:04[IDC]Dragonin the generated sysfont.c, _sysfont_offset is a 32 bit array?!
18:58:25[IDC]Dragonshoudn't a short be sufficient?
18:58:28 Quit vol (
18:59:12amiconnGrr, my box is _very_ sluggish atm. Not surprising though, since I'm checking a 650 MB mp3 with vbrfix, which needs > 1 GB virtual mem
19:00:13NJoinvol [0] (
19:00:49[IDC]Dragonmy rvf_mux also contains an mp3 parser, you could use that to verify some apects
19:01:16amiconnI already verified with mp3utility (which was _much_ faster)
19:02:19amiconnInteresting news: my second test recording with Archos fw worked (almost 9 hours at 44.1 kHz stereo, q=7) - no error, not even a corrupt frame...
19:04:16[IDC]Dragonso we can improve...
19:05:23amiconn(_sysfont_offset), yes, a short should be sufficient. I simply don't know if we should deviate from the current font format, which is based on the Microwindows Project
19:05:29[IDC]Dragonamiconn: have you seen my question about _sysfont_offset ?
19:05:39[IDC]Dragonoops, crossing message
19:06:20[IDC]Dragonthat would break the font file format, right?
19:07:47amiconnYes, it would break compatibilty if the .fnt would be generated the same way...
19:08:48amiconnBut we could leave the on-disk format as-is (with long), and still use short values in memory. This would only require a small change in loading.
19:08:54[IDC]Dragonto use it as shord, we'd have to change the font struct, which probably also defines the file format
19:09:57amiconnWe could use 2 different structs: font (as now) and font_compact (with short)
19:11:37amiconnFont loading should then check for offset values too large for short, and report an error
19:24:29***Saving seen data "./dancer.seen"
19:28:37amiconn(but this shouldn't occur anyway, since the font buffer is < 64KB anyway)
19:30:21[IDC]Dragonlet's not make the code too complicated, else we save nothing :-)
19:31:07amiconnAnyway, implementing bit pattern rotaion and byte-packed storing within convbdf will free more memory
19:57:21[IDC]Dragonleaving, cu!
19:57:32 Part [IDC]Dragon
20:15:01 Quit zeekoe (Read error: 110 (Connection timed out))
21:24:33***Saving seen data "./dancer.seen"
21:46:40 Nick midk|sleep is now known as midk (
21:46:40 Quit Nibbler (Read error: 54 (Connection reset by peer))
22:21:44 Nick midk is now known as midk|shower (
22:22:11 Join dstar5 [0] (
22:36:06 Join invistispy [0] (
22:36:31 Nick invistispy is now known as Havena (
22:38:06 Nick midk|shower is now known as midk (
22:39:29Havenahey anyone awake
22:39:48Havenaanyone try out that avos yet
22:39:55dstar5midk has
22:40:00midki surely have
22:40:02dstar5i dont have a av300
22:40:08Havenawut u got
22:40:19midkme? 320
22:40:21dstar5ok grammer police.
22:41:08midkit's spelling, by the way
22:41:16Havenatrying to decide to get fm recorder v2 then i saw avos now imthinking about the av120 or multimedia 20
22:41:31Havenaor v2*
22:41:49Havenadunno how avos is running tho
22:41:53dstar5not the fm
22:41:56midkavos doesn't run on the multimedia or 120/140
22:42:02dstar5they have problems
22:42:04Havenau serious
22:42:36Havenathe should say that on the site i coulda been screwed
22:42:39dstar5get a plain old recorder, or a v2, there have been less reported problems with them :)
22:43:13dstar5and i have heard the multimedia released at about the same time as the fm was a bit flaky also
22:43:24midkHavena, all the pictures show an av series..
22:43:55Havenabut they say all av300 series and on archos its listed as av300
22:44:01Havenau know what im just retardere
22:44:25midkyou're right, it could be clearer though
22:44:35Havenahow much would a 320 cost?
22:44:56midkhm, probably
22:45:09Havenatoo rich for my blood rite now
22:45:27dstar5you can get a recorder on ebay for 100$ or less
22:45:46dstar5or a gmini for a bit more i think
22:46:16dstar5but no rockbox for gmini
22:46:24dstar5but some people are working on it
22:46:33Havenadoesnt gmini firmware kinda suck
22:46:52dstar5better than the builtin firmware for the recorders
22:46:58dstar5but from what i hear, it does
22:47:04Havenayeh but theres rockbox:D
22:47:32dstar5looks like a gmini on ebay will be from 180-200$
22:47:41dstar5(or more)
22:49:12dstar5humm the 220 gmini is what costs more
22:49:21dstar5the 120 will be from 180-200$
23:00:16 Quit midk (Read error: 104 (Connection reset by peer))
23:02:53 Join midk [0] (
23:10:48amiconndstar5: Your new vu_meter does show a slightly odd behaviour in some places...
23:11:09dstar5amiconn: where?
23:11:53amiconnIt's with the digital meter. I'll try to explain:
23:12:53amiconn(1) For the analog meter, you calculate the decay by taking a weighted average of the last value and the current one.
23:14:07amiconnThis works because the x resolution of the analog meters is relatively high, and properly simulates the inertia of a real needle
23:15:44amiconnHowever, the digital meter has a much lower x resolution, so this causes a problem:
23:16:03dstar5i see
23:16:10dstar5what is a good way to do it?
23:16:14amiconnLets suppose the last_value is 6, and the decay is set > 1
23:17:18amiconnThe new value is 5. What happens is that the displayed value is calculated to be 6 (because the weight of last_value is >1)
23:17:33amiconnThis may be repeated to eternity...
23:18:10amiconn(2) Furthermore, the weighted averaging doesn't exactly behave like a real digital meter would
23:19:22amiconnA digital meter has no inertia, so usually they follow a level increase immediately, and delay only the decrease
23:20:12dstar5ok, then only do the weighting if the new level is lower than the last?
23:20:54amiconnBtw: the weighting problem also exists for the analog meter, but usually isn't visible
23:21:21dstar5is their a good solution on that side?
23:22:26amiconnA simple solution would be to not calculate the number of leds directly, but a higher value (maybe "1/10 leds"), do the averaging with that, and finally divide by e.g. 10
23:23:08amiconnFor the proper digital behaviour you could check if the new value is higher than the last and then skip the averaging, yes
23:24:14dstar5also, i tried to play with the calc_db() in peakmewter.c, but the values it spits out seem to be way higher than they should any ideas on something to make it work better?
23:24:36***Saving seen data "./dancer.seen"
23:25:13dstar5i amde the db scale for the digital look good by mutiplying the number by 4, so it looks good
23:25:33dstar5the analogi s another sotory, it skips about 10 pixles for the original first 2
23:25:51midkdigital is pretty useless, especially in db mode
23:25:56amiconnThe "full" solution would require to not only store the last value (containing all averaged values before), but maintaining a ring buffer of the last n values, then calculate the display value by weighted average of that
23:27:47dstar5ohh makes sense
23:27:59dstar5very much sense
23:28:13dstar5and the decay value will just be the n
23:28:46midkaka "i don't get it but i'll do it so you'll accept the path"
23:29:49dstar5midk: no i actaully do understadn, and have vu_meter.c open and working now
23:29:50amiconnOr you could use a constant n, and determine the decay by using different weights for the averaging
23:30:08midkdstar5, i agree that you have it open, but i still doubt you fully get it
23:30:26dstar5midk: why do you pester every thing now?
23:31:10dstar5look it up i thought you were the grammer police :P
23:31:17amiconnmidk: If *you* get it *you* could improve it too. That's the niceness of open source software...
23:31:21 Join Nibbler [0] (
23:31:44midkdstar5, i know what pester means, i don't understand why you choose to use the word
23:31:57midkamiconn, i sure could, i sure don't want to
23:32:10dstar5midk: because it fits
23:32:24midki guess one could look at it that way.. that one would be "dstar5",
23:36:23dstar5whoops i need a #ifndef SIMULATOR
23:39:52dstar5amiconn: are there any more features you could want in it?
23:42:12amiconnIt's not about features *I* want. The vu_meter plugin is your code, so you decide what features it should include. I just pointed you to some problems with the current code
23:43:26dstar5i like ideas, 2 people can come up with more ideas than 1
23:47:02 Nick midk is now known as midk|gone (
23:47:17amiconnYour next version could perhaps extend the digital meter to include peak hold
23:48:29dstar5sure could...
23:48:48dstar5thanks for that idea :)
23:51:40dstar5ohh me have another idea
23:52:20dstar5have a option for minim eter to show a % value instead
23:53:46amiconnYes. Maybe it should display dB instead if the main meter is set to dbfs
23:54:03dstar5ohh :)
23:57:58dstar5time to go ill get those changes done soon
23:58:09dstar5bte, i have checkers working a little
23:58:23dstar5it can draw the initial board, and you can move around
23:58:29amiconnNice :)
23:58:39dstar5see you
23:58:44 Quit dstar5 ("umount /dev/brain")

Previous day | Next day