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 2004-09-01

00:01:33[IDC]Dragontalk_init() calls talk_buffer_steal(), just because that does reset most things
00:01:49[IDC]Dragonso I use that for my init as well
00:02:24amiconnIiuc talk_init does 2 different things: (1) It sets the talk language to the display language, and resets the "remember-whether-voice-file-is-present" flag. (2) It actually inits the talk engine, and therefore stops playback
00:02:28[IDC]Dragonbut talk_buffer_steal() calls mp3_play_stop(), which gives the undesired effect
00:02:43amiconnFor a solution, these 2 task would have to be separated
00:02:49[IDC]Dragonyes
00:03:50amiconnFor loading a new setting, what to call depends on whether the music is playing: if yes, do only task (1), otherwise do both
00:04:11*amiconn tests something
00:05:12[IDC]DragonI don't know if the reset part of talk_buffer_steal() is worth an extra function, of if I should just duplicate the code in talk_init()
00:05:35[IDC]Dragons/of/or
00:06:19amiconnThat odd behaviour does also occur when "playing" a .lng file or setting the language from the menu
00:07:08amiconnNot that I didn't expect that ;(
00:07:27[IDC]DragonI can move the code, no problem
00:12:45 Quit zeekoe ("quit...")
00:13:36[IDC]Dragontalk.c updated
00:13:48[IDC]Dragonplease try again from cvs
00:14:36*amiconn is checking...
00:17:57[IDC]Dragonworks for me
00:19:35amiconnWorks for me too. The tricky thing to check was: (1) enable voice menu, check if they're working, (2) start some music, (3) select a new language via the menu, (4) stop music, (5) check if voice menus use the new language
00:20:04amiconnBug solved :)
00:27:16[IDC]Dragonthanks for the excellent diagnostics!
00:27:30amiconnIt looks like I only discover bugs where I can delegate the fixing...
00:27:40amiconn;)
00:27:58[IDC]Dragondo you
00:28:00[IDC]Dragon?
00:28:15[IDC]DragonI don't remember...
00:28:34amiconn(1) The null pointer access in mpeg.c => Linus, (2) The talk init => you..
00:28:55[IDC]Dragonah, ok
00:28:59amiconnThere is another one which I'd like to ask hardeep about..
00:29:40amiconnUnfortunately he wasn't here for >3 weeks now
00:30:31[IDC]Dragonoops, just saw your PM about ebay
00:31:06[IDC]Dragonwe sould definitely try not to raise each other
00:31:13[IDC]Dragonshould
00:31:39*[IDC]Dragon bedtime
00:31:41amiconnOf course. Should I have known that?
00:31:50amiconnnite.
00:32:13[IDC]DragonI sent you an IRC msg, but you probably weren't looking
00:32:35amiconnI was looking, but only after the auction ended...
00:33:01[IDC]Dragonnext one is in ~2 days
00:33:07amiconnThere are a few more Ondios, yes
00:33:46[IDC]DragonI could mail-order one for a beit over 50 E
00:33:56[IDC]Dragons/beit/bit
00:34:03[IDC]Dragonwithout tuner
00:34:18[IDC]Dragonwell, later
00:34:23[IDC]Dragonnight!
00:34:31 Quit [IDC]Dragon ("CGI:IRC (EOF)")
00:34:42 Join Locoguy87 [0] (Locoguy87@Toronto-HSE-ppp3677866.sympatico.ca)
00:34:57 Quit Sergej ()
00:36:03 Quit Locoguy87 (Client Quit)
00:40:36***Saving seen data "./dancer.seen"
00:56:52 Quit Nibbler (Read error: 104 (Connection reset by peer))
01:00
01:03:25 Join BlueBandicoot [0] (~522215cc@labb.contactor.se)
01:04:01 Quit BlueBandicoot (Client Quit)
01:19:54 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
01:57:19 Join scott666_ [0] (~scott666@c-24-245-58-48.mn.client2.attbi.com)
01:57:20 Quit mecraw (Read error: 104 (Connection reset by peer))
01:58:11 Join mecraw_ [0] (~lmarlow@69.2.235.2)
01:58:15 Part scott666_
02:00
02:03:36 Join WatchMinister [0] (Anne@a20104.upc-a.chello.nl)
02:03:45WatchMinistergood day everyon4
02:03:48WatchMinistereveryone*
02:15:30 Quit _aLF ("Leaving")
02:32:27 Quit mecraw_ ("Trillian (http://www.ceruleanstudios.com)")
02:40:39***Saving seen data "./dancer.seen"
02:46:08 Part amiconn
03:00
03:10:24 Join Jedi86 [0] (~43a78162@labb.contactor.se)
03:16:16Jedi86does anyone here know anything about the possibility of porting / redesigning the rockbox firmware to be used on the iRiver h1xx series?
03:34:05 Quit Jedi86 ("CGI:IRC (EOF)")
04:00
04:40:43***Saving seen data "./dancer.seen"
05:00
05:05:22 Quit Nibbler (Read error: 104 (Connection reset by peer))
05:17:33 Join maikeul [0] (~gromit@ALagny-151-1-32-220.w83-114.abo.wanadoo.fr)
05:19:00 Quit maikeul (Client Quit)
05:21:11 Join gromit`` [0] (~gromit@ALagny-151-1-32-220.w83-114.abo.wanadoo.fr)
05:25:36 Join webguest92 [0] (~43a751c2@labb.contactor.se)
05:26:00 Quit webguest92 (Client Quit)
05:27:24 Quit maikeul` (Read error: 110 (Connection timed out))
06:00
06:03:26 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
06:37:36 Quit Nibbler (Read error: 104 (Connection reset by peer))
06:40:47***Saving seen data "./dancer.seen"
06:42:49 Quit AciD ("Connection reset by beer")
07:00
07:37:56 Join LinusN [0] (~linus@labb.contactor.se)
07:38:34dwihnoWelcome.
07:40:18LinusNthx
07:48:50LinusNhehe, i found a USB memory stick in the water when i was on vacation
07:49:08LinusNrinsed it, dried it, and it works!
07:51:05gromit``lucky you !
07:51:08gromit``hello :)
07:51:24dwihnoCool!
07:51:28dwihnoLinusN: Jackpot! :)
07:51:35dwihnoLinusN: Did it contain anything snazzy?
07:51:55LinusNplenty of internal information from a major player in the telecom market on it :-)
07:54:14dwihnoWhoa!
07:54:24dwihnoSnazzy information \o/
07:54:48dwihnoNot too bright storing internal information unencrypted though.
07:57:05LinusNi think i've found out who is the owner
07:57:53dwihnoFinders fee for you \o/
07:58:14LinusNdamn! i know the guy!
07:59:05dwihnoNo finders fee then ;)
08:00
08:01:01dwihnoYou should teach him about using encryption on unsafe media :)
08:01:23gromit``was "water"maked \o/
08:01:33gromit``+r
08:01:38 Join midk [0] (~midk@c-24-18-39-204.client.comcast.net)
08:02:12LinusNhehe, it had no housing, only the naked PCB
08:02:30dwihnoLinusN: Well, he will be happy you got the data for him
08:02:45LinusNindeed, i'm emailing him now
08:02:47dwihnoPCB in the nude :)~
08:02:55midkhey LinusN.. did you hear my request about the function to clear the button status? is it possible?
08:03:12LinusNmidk, no i haven't heard
08:03:33midki asked last night... just something that would clear the queue to use before exiting a plugin
08:04:34 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
08:06:10midka function, or any kind of method would be fine
08:06:30LinusNmidk: i explained this to you
08:06:33midka function may be overkill but perhaps you know how?
08:06:42midki know, but.. it's not working
08:07:07LinusNyou still get stray OFF events?
08:07:23midkno, different keys too
08:07:31midklike if i start a game, hold left a second
08:07:38midkwait ten seconds
08:07:40midkand exit
08:07:46midkeven with a sleep(hz) in place
08:07:54midki still get a left keypress outside the plugin
08:08:06midk(or more)
08:08:16LinusNok, then there is a bug
08:09:14midkmust be
08:09:56LinusNfyi, there is a i see it
08:09:59LinusNi see it
08:10:06midkoh, good.
08:10:09midk:)
08:10:32*LinusN is a very silly person
08:10:47midkof course, of course.
08:13:01LinusNbtw, there is a button_clear_queue() call in the plugin API
08:13:24LinusN(but it doesn't work yet, due to my silliness)
08:13:32midkoh, calling that on exit should do instead of a sleep()?
08:13:35midkwhen you fix it. :)
08:13:51LinusNmidk: it is called by the plugin framework when the plugin exits
08:14:07LinusNas i explained the other day
08:14:11midkoh, so it's really all your fault
08:14:14midk:)
08:14:19LinusNas usual
08:16:06midki forgive you.
08:16:24LinusNi'll be forever grateful
08:16:45LinusNcd ..
08:16:50midkdahhaha
08:25:04LinusNmidk: fixed
08:25:59midkgood, other than that the only real bug is my faulty paddle code i conjured up at 12:30am last night...
08:26:06midkwelp, thanks
08:38:37 Join amiconn [0] (~jens@pD95D10DF.dip.t-dialin.net)
08:38:58amiconnhi every1
08:39:29midkhey amiconn
08:40:01dwihnoHi Jens!
08:40:41dwihnoVet du varför Jens är så bra på assembler? Han har så bra kodar-ARMar :)
08:40:46dwihnoHaha. Den var bra.
08:40:51***Saving seen data "./dancer.seen"
08:42:00amiconn?:-|
08:42:24dwihnoSwedish joke about ARM and assembly
08:50:49amiconndwihno: Ah. Btw, "arm" means "poor" in german ;)
08:51:34amiconn(but it also means "arm")
08:56:10dwihnoarm = arm
08:56:34dwihnoWhy is Jens so good when it comes to assembly coding? Because he has such great coding-ARMs :)
08:56:37dwihnoOr something like that ;)
08:57:19amiconndwihno: Ok, so this joke would work in German too :)
09:00
09:00:35dwihnoamiconn: The joke is so universal :D
09:01:17 Join Zagor [242] (~bjst@labb.contactor.se)
09:06:16 Join amiconn_ [0] (~jens@pD9E7F393.dip.t-dialin.net)
09:06:35 Quit amiconn (Nick collision from services.)
09:06:36 Nick amiconn_ is now known as amiconn (~jens@pD9E7F393.dip.t-dialin.net)
09:29:59 Quit Nibbler (Read error: 104 (Connection reset by peer))
09:37:30 Join Bagder [0] (~daniel@1-1-5-26a.hud.sth.bostream.se)
09:37:40LinusNagnes asleep?
09:37:45Bagderyeps ;-)
09:37:59Bagderand a red build!
09:39:30LinusNyeah, i forgot to commit the .h file
09:41:41dwihnoBagder: You've got tillökning?
09:42:08Bagder11 months ago, yes, but now I'm taking care of her full-time while my wife has gone back to work
09:43:38dwihnoOh!
09:43:47dwihnoDaddy Bagdy ;)
09:44:07dwihnoKids are wonderful creatures, as long as they keep the noise levels down.
09:44:15*Bagder nods and tries to look like a parent
09:44:34dwihnoThe day I'll get children I'll probably get some kind of ear protection as well.
09:44:56LinusNas if it was the noise that was the problem...
09:45:12dwihnoThey also have too much of a leak problem
09:46:19 Join kurzhaarrocker [0] (~knoppix@p50876385.dip0.t-ipconnect.de)
09:46:31Bagderend of break
09:46:33 Quit Bagder ("Leaving")
10:00
10:14:33 Join [IDC]Dragon [0] (~d90a3255@labb.contactor.se)
10:14:48[IDC]Dragon'morning
10:15:02kurzhaarrockerMorgääähn
10:15:22[IDC]Dragon;-)
10:18:59*[IDC]Dragon just soldered in a new flash chip
10:20:03kurzhaarrockerDid the old one dissovle while flashing? :)
10:22:13[IDC]Dragonno, this was for a different board
10:24:05[IDC]DragonI didn't hit the 100000 cycle limit ;-)
10:25:11kurzhaarrockerYes, but who counted the tries Archos needed to get the original trash in? ;)
10:25:52[IDC]Dragonhmm, I had version 1.28
10:40:53***Saving seen data "./dancer.seen"
11:00
11:08:04kurzhaarrockerDo you do the 8 MB mod too?
11:09:33 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
11:13:17[IDC]Dragonkurzhaarrocker: yes
11:13:33[IDC]Dragonand 512KB flash instead of 256K
11:13:49[IDC]Dragonand white backlight
11:13:59[IDC]DragonOK, I'll stop bragging
11:14:03kurzhaarrockerDoes white backlight save energy?
11:14:14[IDC]Dragonnot at all
11:14:41kurzhaarrocker(I'd have been surprised if it did)
11:14:48[IDC]Dragonbut the backlight consumption in general is rather low
11:39:09 Join Lynx [0] (lynx@134.95.189.59)
11:47:42 Join lImbus [0] (~manuel@kernel.cycos.net)
11:51:49 Join projekt21 [0] (~ciphelp@host1913.igd.fhg.de)
12:00
12:16:26 Nick kurzhaarrocker is now known as kurzhaarrocker|l (~knoppix@p50876385.dip0.t-ipconnect.de)
12:16:26 Quit Nibbler (Read error: 104 (Connection reset by peer))
12:23:46 Part projekt21
12:40:55***Saving seen data "./dancer.seen"
13:00
13:23:48 Quit lImbus (Remote closed the connection)
13:31:07 Join lImbus [0] (~manuel@kernel.cycos.net)
13:37:41 Join edx [0] (edx@pD9EABD46.dip.t-dialin.net)
13:41:33[IDC]Dragonamiconn: r u there?
13:43:59dwihnoThere. Now I'm running Thunderbird.
13:44:04dwihnoBegone, Outlook!
13:46:08kurzhaarrocker|lI don't see much sense in porting any e-mail client onto rockbox.
14:00
14:00:51 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
14:40:59***Saving seen data "./dancer.seen"
14:49:40 Quit Nibbler (Read error: 104 (Connection reset by peer))
15:00
15:42:58 Part Zagor
15:51:47 Part LinusN
16:00
16:22:38 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
16:31:21 Join Le [0] (~lepoulpe@AMontpellier-251-2-2-166.w83-113.abo.wanadoo.fr)
16:31:41LeHi all !
16:32:02amiconn[IDC]Dragon: Now I'm here
16:32:55 Part Le
16:34:49 Join Le [0] (~lepoulpe@AMontpellier-251-2-2-166.w83-113.abo.wanadoo.fr)
16:36:25 Part Le
16:37:41 Join LePoulpe303 [0] (~opera@AMontpellier-251-2-2-166.w83-113.abo.wanadoo.fr)
16:38:51[IDC]Dragonamiconn: welcome back
16:39:39LePoulpe303Hi all !
16:40:07WatchMinisterhey
16:40:13[IDC]Dragonamiconn: I read some reviews about the Ondio
16:41:01***Saving seen data "./dancer.seen"
16:41:08dwihno[IDC]Dragon: I'm interested to know as well.
16:41:45LePoulpe303Did anyone encounters problems with Archos JB S keys ?
16:41:54[IDC]Dragonsome remarks: a) it has no backlight, b) a plugged MMC disables the internal memory, c) doesn't charge batteries
16:42:37[IDC]Dragonwith b) the probably circumvent the same problem we would have with 2 "disks", how to browse that
16:42:41kurzhaarrocker|lLePoulpe303 they usually break when you have played to many games
16:43:34[IDC]Dragonc) it is meant to be operated on alklines
16:43:47[IDC]Dragonalkalines
16:45:13amiconnb) may as well be to only need one usb->mmc bridge (as you said the internal memory is accessed the same way as an mmc)
16:45:16LePoulpe303kurzhaarrocker|l: lol i prefer to play on my mobile, the archos JB Studio with its ugly LCD is not very gamer friendly
16:46:04dwihno[IDC]Dragon: You don't think it might be possible to use both internal and external memory at once?
16:46:24LePoulpe303kurzhaarrocker|l: : its' my + key that is almost broken ; logically its the most used key
16:46:25kurzhaarrocker|lThere are many Jukeboxes with broken buttons out there. But I haven't heared of any source of replacement buttons yet :(
16:47:30LePoulpe303fortunately i've bought archos ugly-and-expensive control to pilot it ; but i prefer repair it if possible
17:00
17:03:08 Join mecraw_ [0] (~lmarlow@69.2.235.2)
17:03:08 Quit Nibbler (Read error: 104 (Connection reset by peer))
17:03:46 Part LePoulpe303
17:32:33[IDC]Dragondwihno: if it's not a mechanical switch, it should be possible
17:33:23[IDC]Dragonbut Rockbox is not prepared for 2 disks
17:33:53[IDC]Dragonhow would you browse it? Add C:\ and D:\ to the path? ;-)
17:36:46dwihno[IDC]Dragon: yea :)
17:36:55dwihnoSimply selecting the partition
17:48:12 Join bagawk [0] (Lee@ACC31DA9.ipt.aol.com)
18:00
18:05:09[IDC]DragonHi Lee!
18:05:24[IDC]DragonI soldered the flash in today
18:07:32[IDC]Dragonleaving now for home, will try it out there
18:07:37[IDC]Dragonc u
18:07:52 Quit [IDC]Dragon ("CGI:IRC")
18:11:01 Part kurzhaarrocker|l
18:38:36 Quit Lynx (" HydraIRC -> http://www.hydrairc.com <- IRC has never been so cool")
18:41:04***Saving seen data "./dancer.seen"
18:48:13 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
19:00
19:03:39bagawkbrb
19:03:40 Quit Nibbler (Read error: 104 (Connection reset by peer))
19:03:43 Quit bagawk ("umount /dev/brain")
19:07:11 Join bagawk [0] (Lee@ACC61E2A.ipt.aol.com)
19:07:17bagawkwow
19:07:24bagawkhas anyone ever used litestep?
19:17:08 Join [IDC]Dragon [0] (~d9e3481c@labb.contactor.se)
19:18:29[IDC]Dragonbagawk: the new flash is programmed now
19:18:44bagawk[IDC]Dragon: cool :)
19:18:50[IDC]Dragonnext the PCB has to go into the box
19:18:59bagawkthats the easy part :)
19:19:09bagawkexcept i had a little trouble on the usb connections
19:19:14[IDC]Dragona virgin one, with warranty seal, etc
19:19:31bagawkyou can probably see the is a little burn on the board where those two holes are
19:19:40[IDC]Dragonyes, that has to be soldered out with a "strong" iron
19:20:16bagawk[IDC]Dragon: ever used litestep for windows?
19:20:27[IDC]Dragonwhat's that?
19:20:43bagawkit is a explorer replacement
19:20:53bagawkit is MUCh better
19:20:56[IDC]Dragonfile explorer?
19:21:01bagawkjust discovered it a few minutes ago
19:21:06[IDC]DragonI use WinCommander
19:21:08bagawk[IDC]Dragon: not exactly
19:21:21[IDC]Dragonor nowadays called TotalCommander
19:21:26bagawkkinda like gnome instead of KDE
19:21:29[IDC]Dragon(don't like that name)
19:21:47[IDC]Dragonnever used that
19:22:08bagawkahh that is a file manger
19:22:14bagawkthis is a desktop replacment
19:22:27bagawk(btw i use directory opus as file manager)
19:22:28[IDC]Dragon#litestep
19:27:39bagawki cannot describe it, it is just so good
19:27:46bagawkand completely customozable
19:29:29 Part lImbus
19:38:38bagawksee you [IDC]Dragon
19:38:44bagawkill be around later
19:38:52 Quit bagawk ("umount /dev/brain")
19:43:30 Join hoxlund [0] (~none@co-colspgs-u5-c4d-235.clspco.adelphia.net)
19:43:31hoxlundneed some serious help
19:43:31hoxlundmy archos displays:
19:43:31hoxlund*PANIC*
19:43:31DBUGEnqueued KICK hoxlund
19:43:31hoxlundmount: -13
19:43:38hoxlundand can't restart it or do anything with it
19:44:28hoxlundis anyone actually hear or just away?
19:44:32hoxlundhere
19:48:52hoxlundemail me if someone gets these messages: hoxlund@adelphia.net
19:48:56 Quit hoxlund ()
19:50:00 Join hoxlund [0] (~none@co-colspgs-u5-c4d-235.clspco.adelphia.net)
19:50:05 Quit hoxlund (Client Quit)
19:50:30 Join bagawk [0] (Lee@ACC61E2A.ipt.aol.com)
19:54:48 Quit bagawk (Client Quit)
20:00
20:41:08***Saving seen data "./dancer.seen"
20:45:49 Join AciD [0] (~gni@longchamp44-1-82-67-133-87.fbx.proxad.net)
20:49:00 Join Nibbler [0] (~andrer@port-212-202-73-41.dynamic.qsc.de)
21:00
21:07:22 Join uski [0] (~uski@gandalf.digital-network.org)
21:08:55 Join _aLF [0] (~alex@d213-101-226-154.cust.tele2.fr)
21:10:07amiconn[IDC]Dragon: I'm currently trying to track down that .playlist_control corruption bug. It seems that this is not a file system bug, but a bug of the resume code.
21:12:15uskihi all
21:12:23uskiwazzup ? :)
21:12:23amiconnIf there are single tracks listed in the control file, this file is read from and written to at the same time. Argh!
21:17:04 Part uski
21:18:59amiconnIt seems that I am wrong on that point. There is a mutex...
21:21:52 Join webguest10 [0] (~d9f5b632@labb.contactor.se)
21:21:59webguest10hi
21:24:03webguest10i have a problem with my archos jukebox multimedia 20
21:25:29webguest10anybody here?
21:48:09 Join bagawk [0] (Lee@ACC2F413.ipt.aol.com)
21:48:17bagawkhey [IDC]Dragon
21:48:26bagawkgot the board in?
21:49:51webguest10hi
21:50:45bagawkhi
21:51:01webguest10i have a problem with my jbm 20
21:51:45bagawkno one here owns a jbmm that i know of, but shoot
21:53:20webguest10ok
21:53:55bagawkwhat is your problem?
21:54:26webguest10my av out is broken
21:55:47bagawkwebguest10: then you would need to find a replacment jack, and solder it in
21:56:06webguest10hmm. where can i find one?
21:56:17webguest10but the jack isn't totally broken
21:56:18bagawkhold on, i think i know a place
21:56:31bagawkis it jsut the solder then?
21:57:44bagawkif so, that should be easy
21:58:14webguest10i think the solder is ok. the jack is a bit broken. it cracks when i listen to the music
21:59:25bagawkcracks?
21:59:32bagawkas in it makes cracking sounds?
21:59:58 Quit gromit`` ("Client exiting")
22:00
22:00:08webguest10yes. i always have to push the earphones a little bit until it cracks no more
22:00:31webguest10but my headphones are ok
22:00:35bagawkok, then i would not know
22:00:56bagawkit may just be a bad contact, and you can try to clean off the headphone jack
22:01:01webguest10i can imagine that the electrical contacts are dirty
22:01:19bagawkyou can also go to the yahoo group
22:01:23webguest10what stuff should i use for this?
22:01:25bagawkthat woudl be a good place to go
22:01:29webguest10ok
22:01:46bagawkhold on, i will find you a link
22:01:52webguest10yeah
22:03:03 Join gromit`` [0] (~gromit@ALagny-151-1-32-220.w83-114.abo.wanadoo.fr)
22:03:07bagawkwebguest10: is this a part of your language:
22:03:07bagawkUtilisateurs d' ARCHOS, venez tous partager vos expériences, astuces, news... sur ce lecteur ainsi que sur ces accesoires.
22:03:33webguest10no. i'm from germany ;)
22:03:37bagawkohh
22:03:54bagawkone second
22:03:59webguest10ok
22:04:02bagawkthere are a few groups
22:04:05gromit``archos users, come and share your experiences, tips, news... about this player and its accessories
22:04:22bagawkhttp://groups.yahoo.com/group/archosmultimedia/
22:04:27bagawkthere you go :)
22:04:41webguest10thx
22:04:45bagawkthere are lots of members, you are sure to get what you want there :)
22:05:18webguest10i need to have an account... ?!
22:05:27bagawkyes :(
22:05:52webguest10ops. i forgot. i have an account *gg*
22:05:57bagawk:)
22:08:13webguest10are there any non archos firmwares? rockbox hasn't one
22:09:05bagawkthere is one being made for the gmini, and there is one for the AV300's
22:09:24bagawk(and ofcourse rockbox for the player/recorders)
22:10:27[IDC]Dragonamiconn: I'm back
22:10:45bagawkhey [IDC]Dragon
22:10:51webguest10but i can't find any for jbm 20
22:10:51bagawkgot the board in now>
22:10:54webguest10hi
22:11:11[IDC]Dragonbagawk: not yet, didn't try
22:11:25 Quit webguest10 ("CGI:IRC (EOF)")
22:11:27amiconn[IDC]Dragon: That .playlist_control issue becomes more and more strange :(
22:11:40[IDC]Dragon:(
22:12:12amiconnDo you have any clue what happens if an open() ed file is open()ed again? What would be the correct posix behaviour?
22:12:50bagawkwebguest10: go to http://groups.yahoo.com/group/archosmultimedia/post and describe your problem
22:13:15amiconn[IDC]Dragon: Strangely, the resume of even very long control files works if I resume on power up, but not if I resume from ON. It works on the sim, though
22:14:24amiconnI inserted a splash() telling me the filesize that the resume function got from the fd = open() ... filesize(fd) calls. Strange, this yields almost always 273.
22:15:34amiconnEven more strange, 273 is one more than PLAYLIST_COMMAND_SIZE >:-||
22:17:28gromit``for accounts : check www.bugmenot.com
22:18:42bagawkgromit``: is that the mozilla firefox extension?
22:19:18gromit``maybe
22:19:25gromit``there is extensions for everything :)
22:19:25bagawkit is neat :)
22:19:33gromit``are
22:20:39gromit``a shame firefox takes too much time to load
22:20:45gromit``i could be a great browser
22:21:05bagawkit loads fairly fast for me
22:21:10gromit``on linux ?
22:21:12bagawk~4 secs or so
22:21:35bagawki have not usesd firefox on linux, but have use dthe standard mozilla
22:21:45bagawkbut it seems most things take a bit onger to start in linux
22:21:50bagawk*longer
22:21:53gromit``on linux i htink it is twice as long as on windows
22:22:31bagawkgromit``: what do you use? opera?
22:22:52gromit``konqueror
22:23:03gromit``opera isn't free
22:23:20bagawkheh i cannot stand konqueror
22:23:25gromit``me too
22:23:30gromit``too many bugs
22:23:36gromit``though interesting features
22:23:43bagawki assume you are using KDE also....
22:23:53gromit``bugs with javascript/mimetypes...
22:23:57gromit``since version 3.2
22:24:04gromit``enlightenment before
22:24:13bagawkypu should try xfce
22:24:33gromit``i really enjoy specific windows options
22:24:39bagawkhttp://xfce.org/index.php?page=screenshots&lang=en
22:24:53gromit``"you open this size on desktop number 3, with no border... and don't move !!"
22:25:01gromit``i look
22:25:28gromit``kde & konqueror has known great improvements
22:25:39gromit``it is worth having a new glance at it now
22:25:54bagawki hate those frigeen icons that bounce in KDE when start things
22:26:04gromit``not mandatory :)
22:26:49gromit``am i here ?
22:26:58bagawkyes
22:27:23gromit``weird it seemed i had no net anymore :)
22:27:23bagawkthere is also a desktop manager for windows i just discovered today
22:27:27bagawklitestep
22:27:31gromit``too buggy
22:27:34gromit``:)
22:27:41bagawki see no bugs so far
22:27:42bagawklol
22:27:52gromit``i used it last year
22:27:56bagawk(.24.7)
22:28:15gromit``xfce has a nice look
22:28:29bagawkit does, and it uses very little ram
22:28:35bagawkand very customizable
22:29:56gromit``maybe i'll put it on my server :)
22:30:13gromit``i'll have to grab every xf* application !
22:32:34bagawkthere is another project that adds some plugins to xfce that are nice
22:32:38bagawklike cpu monitor etc
22:33:04bagawkhttp://xfce-goodies.berlios.de/
22:33:27gromit``i want the smallest launch bar
22:33:35bagawk:)
22:33:39bagawkyou can do that
22:33:41gromit``to benefit the fact i have a 17" screen
22:33:41gromit``:)
22:33:45gromit``so no goodies :)
22:33:57[IDC]Dragonamiconn: (files) I don't know what happens if you open again
22:34:04bagawkone way to save space is to move the workspace switch to the top
22:34:09bagawkwhere the taskabr is
22:34:14[IDC]Dragonin windows, it depends on the mode
22:34:30[IDC]Dragonbut certainly no 2 can write
22:34:40gromit``it has a gnome-like look it seems
22:34:43WatchMinisterhey
22:34:56bagawkit is nothing liek gnome
22:35:08bagawkgnome is almost like kde
22:35:12gromit``sure :)
22:36:40[IDC]Dragonbed time
22:36:44gromit``gn
22:36:45bagawksee you [IDC]Dragon
22:36:49 Quit [IDC]Dragon ("CGI:IRC (EOF)")
22:37:04gromit``when i have some time i'll install xfce on my server
22:37:21gromit``kde is a little too greedy for a PII233
22:37:24gromit``:
22:37:27gromit``:)
22:37:29bagawki am going to the library see you in about 15-20 mins :)
22:37:34gromit``cya
22:37:41gromit``maybe i'll be sleeping
22:37:48gromit``i haven't slept last night :)
22:37:50bagawkok
22:37:52bagawklol
22:38:00 Quit bagawk ("umount /dev/brain")
22:41:12***Saving seen data "./dancer.seen"
22:52:16 Join LinusN [0] (~linus@labb.contactor.se)
22:52:28amiconnhi LinusN
22:52:36LinusNamiconn: which mode are you opening the files in?
22:53:28amiconnI'm not opening them myself, I try to analyze the .playlist_control corruption bug
22:53:57amiconnI make some slowly progress, though.
22:54:00LinusNthe same file can be opened as many times you wish, as long as it's read-only
22:54:06 Join bagawk [0] (lee@IC62.library.oregonstate.edu)
22:54:18amiconnThe resume function opens it with O_RDWR
22:54:33amiconn...but closes it before, as I just found out
22:54:36LinusNtwice?
22:54:43LinusNok
22:54:55 Quit WatchMinister ("Those who would give up essential Liberty, to obtain a little temporary Safety, deserve neither Liberty nor Safety.")
22:55:03amiconnThe problem is likely created earlier:
22:56:05amiconnI just found (via splash debugging), that the file size at the start of the resume function (before it closes the old handle) is still correct (some 21000 bytes in my test case), but the file pointer points to pos. 273
22:56:35amiconnI'm almsot sure the old handle was created with O_TRUNC...
22:56:46amiconn*almost
22:57:12amiconnSo this seems to be no file system bug, but a bug in the playlist handling code...
22:57:37LinusNnice
22:58:00amiconnErm, not really. I still have to spot that d*** bug
22:58:02 Join moormaster [0] (~trillian@port-212-202-174-153.dynamic.qsc.de)
22:58:08 Join zeekoe [0] (zeekoe@zeekoe.kabel.utwente.nl)
22:58:16 Part moormaster
23:00
23:04:36 Quit Ka_ ("* poof *")
23:06:49LinusNok, so we insert a directory into an empty playlist, creating a new control file in the process, right?
23:07:41amiconnThe playlist mechanism is rather complex, and works different than one could think of in many places:
23:07:50LinusNthen we do playlist_start()
23:08:08LinusNthis does a seek() to the beginning of the control file
23:08:08amiconnIf we play a file within a dir, a dynamic playlist is created, containing that dir itself.
23:08:18LinusNwhich is opened with O_TRUNC
23:08:51amiconnContrary, if we on+play->playlist->insert a directory, the individual tracks are written to the control file.
23:08:53LinusNwhen we later close it, it will be truncated
23:09:44amiconnIf I didn't overlook something, the control file is _never_ opened with O_TRUNC...
23:10:45LinusNcreate_control() calls creat()
23:11:13amiconnargh
23:11:19LinusNcreat() implies O_TRUNC
23:12:10amiconnSo perhaps the file should be created with creat(), closed immediately, and the reopened without O_TRUNC
23:13:01LinusNsounds like a working solution
23:13:19amiconnOr (maybe easier): seek to end before closing
23:13:42LinusNbut we want the control file to be truncated
23:13:50LinusNwhen we create it
23:14:44amiconnOk, right
23:14:46LinusNreopening it seems to be the best solution
23:15:06dwihnoWups
23:15:24dwihnoI just realized I lost my GUI code when my disk crashed
23:15:32dwihnoA good reason to re-write it.
23:15:39dwihnoThat will be my christmas gift to the project
23:15:53LinusN2004 or 2005? :-)
23:16:34dwihnoDepends
23:16:40dwihno;)
23:16:54dwihnoI hope I'll get the sound menu done
23:17:54dwihnoTime to curl to bed... :)
23:18:02dwihno(yes, a slight pun intended)
23:18:06LinusNdwihno: nite
23:18:20 Join SmoothOperator [0] (~acaf80f5@labb.contactor.se)
23:20:42LinusNamiconn: btw, we can boost the playlist handling slightly by not doing fprintf()
23:21:26LinusNit calls write() for every byte
23:24:47amiconnWe'd need an extra snprintf buffer for that
23:24:55LinusNyes
23:26:13LinusNor we could add a small buffer to fprintf()
23:26:58amiconnIn fact I think that's the better idea - it will boost fprintf in all places
23:27:03LinusNlet's say, 64 bytes
23:27:40LinusNlet's do that after we have fixed the playlist bug
23:28:24amiconnThat beast is fixed now. yeah!
23:28:29LinusNwow!
23:29:04LinusNi'm doing the fprintf buffering now
23:35:05amiconnMaybe the gain from this is less than we would expect. The file system does sector buffering anyway
23:35:38amiconnBtw: Any news on the NULL pointer access in mpeg.c?
23:35:42LinusNperhaps, but there is a lot of code to execute for each byte
23:35:51LinusNi haven't looked at it yet
23:36:08 Join silencer [0] (~silencer@dyn-213-36-104-214.ppp.tiscali.fr)
23:36:24LinusNi'll leave the fprintf() optimization for now, until i can benchmark it correctly
23:37:34amiconnLinusN: I'd like to optimize the screendump function. This uses 2 buffers with the size of lcd_screenbuffer (imho with no reason), and the bitmap rotation is rather slow...
23:38:20LinusNi didn't know there was a performance problem...
23:38:52amiconnNot really a performance problem for it's current occasional use...
23:39:29amiconnbut (1) considered it's occasional use, it should take less space (both code & data)
23:39:40 Join AciD` [0] (~AciD@longchamp44-1-82-67-133-87.fbx.proxad.net)
23:39:51LinusNi agree
23:40:13amiconnand (2) after improving it, it could be put into the api, for using it in e.g. etch-a-sketch
23:40:32LinusNsure
23:40:52LinusNyou have my blessing :-)
23:40:57amiconnI'm thinking about implementing the reverse function (.bmp loading) as a plugin lib function
23:41:16LinusNgood idea
23:41:31amiconnThe optimized screendump will use a very little assembler macro, the rest can be done in C
23:41:58LinusNremember that it is used in the simulator too
23:42:27amiconnThen the macro can be defined in real C (much slower, but working everywhere)
23:42:41LinusNis the speed an issue?
23:43:13amiconnThe assembler macro will do the following: Shift one variable right by 1 bit, and another variable left by 1 bit, shifting in the lsb of variable (1) into (2)
23:43:23amiconnThis will be both faster and smaller
23:43:45LinusNi mean, we talk a lot about portability nowadays
23:44:17LinusNbut on the other hand, the next device won't have the same lcd anyway...
23:44:54amiconnyes
23:44:59zeekoe!!!yay for amiconn!!!
23:45:02zeekoe:)
23:46:32LinusNgotta reboot, cu later
23:46:44 Part LinusN
23:50:30amiconnzeekoe: That "beast" you are unable to decipher is actually a part of the bitmap rotation I was talking about...
23:50:56zeekoeyeah, i already thought it would be some rotating thing
23:51:07zeekoeshifting bits, and then adding stuff to it
23:51:32zeekoeyou _do_ understand it?
23:51:46amiconnThe rockbox native bitmap format is a bit strange. That's because it is the same format as used by the lcd controller
23:51:58amiconnYes, I understand it.
23:52:28zeekoenice :)
23:52:43zeekoeoh... i thought the bmp-native format was strange
23:52:43 Quit SmoothOperator ("CGI:IRC (EOF)")
23:52:51 Nick gromit`` is now known as gromit[DLM] (~gromit@ALagny-151-1-32-220.w83-114.abo.wanadoo.fr)
23:54:00amiconnzeekoe: Actually, I already dealt with it a lot, as it is used in a number of places I touched: the lcd_bitmap() function and the grayscale framework are examples
23:54:13 Join Elemeno [0] (~4givn@dpc691997050.direcpc.com)
23:54:21ElemenoHi folks
23:54:27bagawkhi
23:54:57Elemenohey, has anyone tried to replace/reposition the internal mic on the jukebox?
23:55:04zeekoeme
23:55:09zeekoenot seriously though
23:55:36zeekoecouldnt get the cpu board removed without breaking
23:56:05Elemenohmm.. my mic was attached to the board by two inch-long wires - red and black
23:56:10zeekoecool
23:56:20zeekoedid you try it?
23:56:47Elemenowell, I recorded with the mic about an inch away from the jukebox, and there is hardly *any* background noise
23:56:53Elemenosuch as clicking, whirring, etc
23:57:03zeekoenice....
23:57:23Elemenoyeah. so now I'm trying to figure out a permanent fix
23:57:58zeekoei thought about leading it out through the digital in/out
23:58:33Elemenothe microphone itself is about the size of a pencil eraser. I just wonder where I can mount that and still have it be safe

Previous day | Next day