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 2004-04-21

00:03:33 Quit mecraw__ ("Trillian (")
00:22:39 Join midk [0] (
00:26:37LaurentGtime to sleep
00:26:40LaurentGcu all
00:26:42 Quit LaurentG ("Download Gaim:")
00:56:16ricflairive noticed that the jukebox seems to be dead of power when the power indicater has it at about half full. Strange
01:27:19 Join top_bloke [0] (
01:37:12 Quit MT (Client Quit)
01:42:27top_blokeyo word
01:52:42***Saving seen data "./dancer.seen"
01:53:29 Part amiconn
02:22:50 Join omario [0] (
02:22:50 Quit bc|busy (Read error: 54 (Connection reset by peer))
02:23:00omariohey all
02:23:08midkyo yo yo
02:23:15omarioanyone own a 20gb archos here?
02:23:52omarioIT doesn't seem to have a radio?
02:24:24omariowhatever one that is..
02:24:33omario"Archos 20 GB USB 2.0 MP3 Jukebox Recorder
02:24:39omarioI can't seem to find a version number..
02:24:57top_blokeis it V2?
02:25:20omariohow can I tell?
02:26:15top_blokei guess it doesnt matter anyway
02:26:19omarioI mean.. I don't own it, but I'm thinking of buying one..
02:26:48top_blokewell theyre both good
02:27:07 Join MT [0] (mt@
02:27:11omariojust wondering.. the one I'm thinking of buying - doe sit have a radio?
02:27:24top_blokeunless it says FM
02:27:34midkor maybe if its a v3
02:28:28top_blokebut recent v2s dont have radio
02:28:30omariolol, I was kinda hoping there was some way to hack it up to have an FM radio.. like stick an antenna on it :P
02:29:19midkomario: it's a possibility
02:29:21omarioI had a few general questions though...
02:29:27omariolol, that would be cool :P
02:29:39omarioOne thing I was wondering is how it holds up to like.. runnign with it
02:30:58top_blokeprobably not recommended
02:31:20top_blokeit spins up like very 2 min
02:31:32top_blokegah i cant get sim to test the rock
02:32:15omariobut.. for something less bumpy.. like wlaking it would be ok?
02:32:16top_blokesays like E:\tetris.rock is not a valid windows image
02:37:27omarioat $129.. worth it?
02:37:52top_blokehell yeah
02:37:57top_blokeBUY IT NOW
02:38:09top_blokethats like free
02:38:15top_blokeyou can toss it if u dont like it
02:38:20omariolmao ok
02:38:27omarioso.. with rockboxx I can get games?
02:38:32midkno you can give it to ME if you dont like it
02:38:33top_blokehell yeah
02:38:40midkyes tetris, snake, wormlet, star
02:38:46top_blokewere working on improved teteis!
02:39:06midkvery hush hush
02:39:48omariosweet lol
02:39:51top_blokeoh crap
02:39:53top_blokei blew it
02:40:00top_blokei blew the cover
02:40:10omariodon't worry, I won't tell :P
02:40:12midk*pulls out gun
02:40:15midkmove along move along
02:41:09top_blokethis sim stuff takes so long
02:41:33top_blokei could have USB'd like 5 times by now
02:44:20 Quit elinenbe (" Like's GUI? Then try HydraIRC -> <-")
03:00:07 Quit top_bloke ("The mind is a terrible thing to taste. Wasted 0 seconds online.")
03:00:50 Join BC [0] (
03:00:57 Join top_bloke [0] (
03:02:05midkyo bc
03:04:33top_blokeok i got the score ++
03:06:56 Quit omario ()
03:25:29 Quit AciD ("")
03:25:29 Quit Nibbler (Read error: 104 (Connection reset by peer))
03:35:02 Quit midk ("ttyl")
03:52:44***Saving seen data "./dancer.seen"
04:28:38 Join thu [0] (
04:29:47BCOOPS THAT'LL BE HEX mode
05:03:36 Join silencer_ [0] (
05:03:36 Quit silencer (Read error: 54 (Connection reset by peer))
05:07:03 Join Nibbler [0] (
05:12:56 Quit top_bloke ("The mind is a terrible thing to taste. Wasted 2 hours 12 minutes and 41 seconds online.")
05:34:11 Quit Nibbler (Read error: 104 (Connection reset by peer))
05:52:48***Saving seen data "./dancer.seen"
05:55:32 Quit BC ("Trillian (")
06:16:24 Quit scott666 ("i'll be back...eventually...")
06:34:15 Join midk [0] (
07:06:36 Join LinusN [200] (
07:10:12dwihnoWelcome back \o/
07:10:23midkhey linusn
07:13:03 Join Nibbler [0] (
07:26:30midkwoo woo
07:33:55 Quit jkerman (Read error: 104 (Connection reset by peer))
07:34:46 Join jkerman [0] (
07:38:16 Join [IDC]Dragon [0] (
07:38:38[IDC]Dragongood morning earlybirds!
07:40:35[IDC]DragonLinusN: I've finally fixed my mp3 clips switch problem.
07:41:54[IDC]Dragonmp3_play_pause() is evil, the pause via SCR0 eats data.
07:52:49***Saving seen data "./dancer.seen"
08:01:16 Quit Nibbler (Read error: 54 (Connection reset by peer))
08:35:36LinusN[IDC]Dragon: like we suspected
08:36:49dwihnoAre there any good ways to make automated backups (revision-based would be nice too) in a *nix environment?
08:40:01LinusNof course
08:40:12[IDC]DragonLinusN: but I haven't touched mp3_playback.c to change the pauseplay, instead I kicked mp3_play_pause() out of shutup() and instead disable/enable the DMA.
08:40:49midkshutup() LOL
08:41:28LinusN[IDC]Dragon: nasty
08:41:42[IDC]DragonI know
08:41:59dwihnoLinusN: I wouldn't mind a pointer or a tip :)
08:42:28LinusNdwihno: just to back up your personal files, or?
08:43:13dwihnoLinusN: Nah, complete company backups.
08:44:00 Join mattzz [0] (
08:45:23dwihnoLinusN: it has revision support?
08:49:31LinusNi dunno
08:49:47LinusNwhy is date not good enough?
08:49:54mattzz str_vrc_fcode_for,
08:50:58mattzz[IDC]Dragon: check out"> :o)
08:52:25LinusNcommit! commit! commit!
08:52:26midkooh lemme see
08:52:29midkwhat what what
08:52:32midkwhat is it
08:53:38midkcan this be downloaded?
08:53:49mattzzLinusN: you can find the source at"> (I dont have access to CVS)
08:54:12mattzzLinusN: or I can commit it to patchtracker
08:54:26LinusNmattzz: why is is a separate plugin?
08:55:01[IDC]Dragonmattzz: I saw it yesterday, have it running already, nice.
08:55:14mattzzLinusN: because of the grayscale framework overhead. It is about three times slower than the b&w version
08:55:31LinusNcan't they be in the same source file, with some options?
08:55:33[IDC]DragonI still would replace the b/w
08:55:46LinusNgrayscale: on/off
08:56:29mattzzI agree on replacing the b/w version. I dont have a key left for options ;-)
08:57:14midktrying it now
08:57:21[IDC]Dragonmattzz: I saw you and amiconn in IRC, but could not participate. I was behind a network split. (Whatever that is)
08:57:30midknice job mattzz
08:57:38midkvery cool
08:58:00mattzzamicon and IDC where a great help
08:58:34midknite all
08:58:37 Quit midk ()
08:58:53*mattzz does some self marketing today ;-)
09:00:26 Join amiconn [0] (
09:00:50mattzzmorning jens
09:01:16amiconnMorning all
09:06:12LinusNyo amiconn
09:15:13LinusNmattzz: you want to replace mandelbrot.c with graubrot.c?
09:15:39mattzzLinusN: makes sense to me
09:16:01[IDC]Dragonyes, but call it mandelbrot still, I'd suggest
09:16:04LinusNi'll do it, and i'll change the attempt to empty the keyboard queue
09:16:09LinusN[IDC]Dragon: yup
09:16:25[IDC]Dragongraubrot was a german joke
09:16:34mattzzyup, better call it mandelbrot
09:16:40[IDC]Dragonbecause "grau" means "gray"
09:17:01mattzzand graubrot is sooo healthy my ma told me.
09:17:22[IDC]Dragonnot schwarzbrot?
09:17:40mattzzschwarzbrot was even worse!
09:18:12mattzztoday I mostly eat toastbrot - as a result
09:18:26mattzzbut that's boring stuff ;-)
09:19:10LinusNi think apfelmännchen is a cute name for it :-)
09:19:42[IDC]Dragonis that international?
09:20:16LinusNnot at all
09:20:33LinusNjust german iirc
09:22:08[IDC]Dragongotta eat my toastbrot
09:25:36LinusNmattzz: committed
09:26:40amiconnLinusN: Anything new about the lcd code on the player, or the recording code?
09:27:08LinusNwas planning to work on that today
09:28:32[IDC]DragonLCD consolidation?
09:28:56[IDC]DragonI didn't dare to do that, for a lack of a Player
09:29:25amiconnIf my first attempt works on the player (it does on the recorder), I have some further optimization in mind.
09:29:34[IDC]DragonBut I wrote that comment which seemed ta have inspired you ;)
09:29:55[IDC]DragonI would have only the short loop on the player
09:30:05[IDC]Dragonit's not speed critical
09:30:09amiconnI'm also lacking a player, therefore asked Linus to test.
09:30:18[IDC]Dragonthe unrolled for recorder then
09:32:04amiconnI wanted to leave the unrolled loop for both player & recorder (the IRAM space isn't that tight on the player as Linus told me).
09:34:04 Join amiconn_ [0] (
09:34:30 Quit amiconn (Nick collision from services.)
09:34:30 Nick amiconn_ is now known as amiconn (
09:34:38amiconnMy next optimization would be to use the unrolled loop also for command writing (without including it a second time). I estimated that this could give another 3% speedup for lcd_update() / lcd_write_data()
09:35:04[IDC]DragonMr. 3%, haha
09:35:56[IDC]Dragonthe player needs a different unrolled loop
09:36:11[IDC]Dragonbecause my carry bit trick is not working there
09:36:34[IDC]Dragonso I thought you may just leave the C loop in there
09:36:43[IDC]Dragon(easier to read)
09:38:12amiconnThere is no pure C loop in the current version for the player either
09:39:54[IDC]Dragonah, sorry
09:40:38amiconn(the transfer is done in assembler for both models)
09:41:04[IDC]Dragonjust saw my comment about tilting the LCD buffer, that's outdated, long done.
09:41:26amiconnBut there are different loops for command and data transfer, the command loop is not unrolled currently.
09:41:39amiconnYes, I will take out that comment on the way.
09:41:47[IDC]DragonI know, could be unified.
09:42:04 Join Nibbler [0] (
09:42:16[IDC]DragonGotta move my butt to work, see you later!
09:42:44 Quit [IDC]Dragon ()
09:46:22 Part thu ("Client exiting")
09:46:31 Nick amiconn is now known as amiconn|away (
09:52:53***Saving seen data "./dancer.seen"
10:41:52 Nick mattzz is now known as mattzz|work (
10:42:38LinusNamiconn|away: OK to commit your LCD stuff, remove Jörgs old comment about tilting the frame buffer as well
10:46:38 Nick amiconn|away is now known as amiconn|work (
11:36:19 Quit Nibbler (Read error: 54 (Connection reset by peer))
11:49:40ricflairlinusn, do you know any idea why my jukebox recorder 20 would turn off when pwer gets down to around half way on the battery monitor, and also it seems to go from 100 percent to around 60 percent withing about half an hour to an hour.
11:52:54***Saving seen data "./dancer.seen"
11:53:53amiconn|workricflair: I think one of the battery cells may not be ok. Either you could check this with an external charger that is able to do tests, or simply put in another set of batteries.
11:55:08LinusNricflair: have you checked your battery connector yet?
11:56:14ricflairnot yet linusn, i guess thats going to be the problem isnt it.
11:56:30ricflairamiconn, i dont have the tools to test , so i may just pick up another battery set
11:56:51ricflairi left it charging from ll last night, so its been on for near 12 hours charge now
11:57:02ricflairand it doesnt say charged yet
11:59:14amiconn|workLinusN: LCD code committed, as well as a minor fix for the sim (lcd-common.c was missing lf at the end).
12:05:00LinusNsaw that, gr8
12:11:48amiconn|workLinusN: Btw, very strange commit message for mandelbrot.c ...
12:29:37LinusNbuah, the shell didn't like my exclamation signs :_)
12:54:13amiconn|workGrr, I have caused a player sim warning...
12:56:01LinusNah, forgot to test that
12:56:41amiconn|workWill fix it asap.
13:02:05 Join AciD [0] (
13:13:58 Join Nibbler [0] (
13:24:25 Join pfavr [0] (
13:26:09mattzz|workseen BC
13:26:30mattzz|worklogbot_: seen BC
13:45:09 Quit pfavr (Remote closed the connection)
13:52:56***Saving seen data "./dancer.seen"
13:57:38c0uttau there linus ?
14:11:19 Join pfavr [0] (
14:24:50 Quit pfavr ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]")
14:28:52 Quit silencer_ ("leaving")
14:28:59 Join silencer [0] (
14:30:19 Join [IDC]Dragon [0] (
14:45:45 Quit Nibbler (Read error: 104 (Connection reset by peer))
14:47:23 Join pfavr [0] (
14:49:07 Join Nibbler [0] (
14:50:23 Quit mattzz|work ("CGI:IRC")
14:52:51 Join Robb1eVox [0] (
14:53:21 Part [IDC]Dragon
14:59:31 Quit AciD (Read error: 54 (Connection reset by peer))
15:20:28 Quit Nibbler (Read error: 104 (Connection reset by peer))
15:36:37 Quit pfavr ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]")
15:47:38 Join pfavr [0] (
15:52:59***Saving seen data "./dancer.seen"
16:00:54ricflaircan the unit be damaged if it powers off whilst your copying files to it
16:01:23dwihnoWell, the file system might get a minor corruption
16:01:45dwihnobut the hardware should be OK, unless you make it a habit :)
16:02:22ricflaircool, its just with me having battery problems this happend yesterday. all seems fine, but i just wanted to make sure
16:05:11dwihnoWhen it happens to me, I always check the file system.
16:09:29 Join mecraw__ [0] (~mecraw@
16:10:35ricflairhow can you check it
16:13:03Robb1eVoxne1 here tried the plugin for connection to car as a cd changer ?
16:14:34 Join solaris [0] (
16:16:09 Quit pfavr (Read error: 104 (Connection reset by peer))
16:16:32 Join mattzz [0] (
16:17:42 Join pfavr [0] (
16:22:37dwihnoricflair: well, if you use windows, scandisk :)
16:24:18Robb1eVoxfinding the cd changer protocol looks like the hardest part, didnt realise lot of the archos plugin sw was already done
16:31:31pfavrjust a stupid idea: how about ide-smart support in rockbox (wonder if it is possible to query the smart info through usb?)
16:54:23 Join methangas [0] (
16:59:00 Join AciD [0] (
17:02:55 Quit solaris (Read error: 60 (Operation timed out))
17:03:31 Join Nibbler [0] (
17:04:00 Quit mattzz ("CGI:IRC")
17:04:57 Join mecraw_ [0] (~mecraw@
17:04:57 Quit mecraw__ (Read error: 104 (Connection reset by peer))
17:26:01 Join Ka__ [0] (~tkirk@
17:37:04 Join [IDC]Dragon [0] (
17:38:00[IDC]DragonRobb1eVox: from logpeeking I saw you asked about CD changer emu?
17:53:00***Saving seen data "./dancer.seen"
18:53:34 Quit pfavr ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]")
18:54:40Robb1eVox:) lo
18:55:20Robb1eVox[IDC]Dragon u still there ?
18:55:51[IDC]Dragonyes, seldom looking.
18:56:53[IDC]DragonI made the Alpine plugin, dunno if you know.
18:56:53 Nick amiconn|work is now known as amiconn|away (
18:57:14Robb1eVoxim lookin at linking my ajb to a renault head unit which i thing is a spin off of the sony system
18:57:28Robb1eVoxye only started lookin today and found out quite a lot
18:58:05Robb1eVoxtrouble is i dont not found any renualt info at all
18:58:16 Quit AciD (Read error: 54 (Connection reset by peer))
18:58:30[IDC]Dragonyes, if they just buy it...
18:58:34Robb1eVoxhow did you go about learning how to do the alpine setup ?
18:59:02[IDC]Dragonscope, logic analyzer, uC board, PC
18:59:03Robb1eVoxbuy it ?? how u mean
18:59:17Robb1eVoxahh ok
19:00:36 Quit [IDC]Dragon ("no fate but what we make (EOF)")
19:01:42 Join [IDC]Dragon [0] (
19:02:01[IDC]Dragonsorry, unstable connection.
19:02:53Robb1eVoxdid i read somewhere about a system for oems as well ?
19:05:02[IDC]Dragonyes, some other brands use M-Bus, too.
19:06:22Robb1eVoxim only guessing but gather its a unilink derivitave as my connectors are the same as this
19:07:15Robb1eVoxthink my best way forward is to build a gnunilink interface and see what i can get it to do
19:07:46Robb1eVoxi believe the system used in vw's is the same too
19:08:58 Quit [IDC]Dragon ("no fate but what we make (EOF)")
19:12:38 Join [IDC]Dragon [0] (
19:12:56[IDC]Dragonsorry, this is pointless
19:14:47[IDC]Dragonyou're right, before thinking more about Archos you need to master the protocol.
19:14:59[IDC]Dragonbest done with a PC
19:15:35Robb1eVoxi wasnt expecting to be able to interface it tbh, just looking at audio inputs untill i found out that archos dev had already been done
19:33:13 Part [IDC]Dragon
19:47:39 Nick amiconn|away is now known as amiconn (
19:53:03***Saving seen data "./dancer.seen"
19:54:50 Join ruumis [0] (
19:55:36ruumisthinking about picking up that archos 20GB that's down to ~$130 after rebate on amazon
19:55:47ruumisI take it you're all owners?
20:04:39Robb1eVoxnice price
20:06:18ruumisI've heard so many conflicting things about quality control and support...
20:08:32ruumisI'm also trying to figure out of the one on amazon has NiMH or Lithium Ion
20:08:47ruumisthe product description says NiMH but at least one of the reviews says Li-Ion
20:13:31Robb1eVoxprettu sure they minh
20:13:37Robb1eVoxnimh even
20:18:23Robb1eVox1500 nimh are standard (just looked in mine)
20:19:43 Join mattzz [0] (
20:22:10amiconnHi Matthias
20:22:35 Join Guest [0] (
20:22:42mattzzHi Jens
20:22:46 Quit Guest (Client Quit)
20:24:45mattzzIs there currently anybody in charge of the user manual?
20:26:36mattzzOK, everybody is shouting loud enough - please stop it ;-)
20:38:04mattzzI'd suggest to update the manual for the upcoming release and skip the manual for version 2.2
20:40:41 Join [IDC]Dragon [0] (
20:41:04[IDC]DragonJens, r u there?
20:42:48[IDC]Dragongot a question about your grayscale framework
20:42:50amiconnI've just received the AT&T natural voices along with the german voice. So expect me to build a german.voice this evening.
20:43:20[IDC]DragonI've just updated the .lang file yesterday
20:43:33amiconnHave seen it.
20:43:36[IDC]Dragonshould be a piece of cake then.
20:44:06amiconnI first have to figure out the process
20:44:38[IDC]Dragonthat's easy
20:45:02[IDC]Dragonabout gray_drawgraymap():
20:45:21[IDC]Dragoncan you introduce a "stride" parameter?
20:45:49[IDC]Dragonso the line distance of the source is independent of the width?
20:45:59[IDC]Dragon(or am I missing something?)
20:46:20amiconnDo you want to show partial bitmaps?
20:46:50[IDC]DragonSo I can scroll along a larger one.
20:47:24amiconnThere is not yet a parameter to do that (I didn't think about it), but of course it would be easy to integrate.
20:48:13[IDC]Dragonyes, sure
20:49:08[IDC]Dragoninstead of x and y the stride would be sufficient
20:49:32[IDC]Dragonsrc can point to whereever in the image
20:49:41amiconnI should add that to gray_drawbitmap then as well, for symmetry.
20:50:03[IDC]Dragonah, yes
20:50:37amiconnNo, stride has to be an additional parameter, since x and y define the upper left corner of the destination (on the display)
20:51:04[IDC]Dragonoops, yes
20:53:10amiconnI wonder if I should commit grayscale.c itself, since it is not intended (primarily) for direct use.
20:55:59[IDC]Dragonyou can make an #if 0 around it to prevent it in the distibution.
20:58:24 Join AciD [0] (
21:06:03Robb1eVoxdo all archos models have the remove feature in the jack ?
21:08:22[IDC]Dragonsome FMs lack it
21:08:37Robb1eVoxhow bout on a studio ?
21:08:49[IDC]Dragonbut just because it's not connected internally
21:08:49Robb1eVox<−− got a studio 10
21:09:09Robb1eVoxit on the earphone socket or line ?
21:10:45Robb1eVoxi got absolutely no idea what protocol this head unit uses, could be 1 wire mbus type or maybe 2 wire
21:36:59lurwas/me's former Archos firmware Jukebox 6000 is now a Rockbox 20000
21:37:15*lurwas 's former Archos firmware Jukebox 6000 is now a Rockbox 20000
21:37:20lurwaseven. :)
21:49:47 Join scott666 [0] (
21:53:04***Saving seen data "./dancer.seen"
21:55:37 Quit [IDC]Dragon (Read error: 104 (Connection reset by peer))
21:57:03 Join solaris [0] (
22:00:05mattzzc0utta: r u there?
22:00:40 Quit Ka__ ("Leaving")
22:16:22 Quit solaris ("—I-n-v-i-s-i-o-n— 2.0 Build 3515")
22:18:30 Join pfavr [0] (
22:21:52 Join [IDC]Dragon [0] (
22:22:03amiconnre Jörg
22:22:09[IDC]Dragontoday is disconnection day...
22:22:31[IDC]Dragonvoice file done?
22:23:14amiconnI wonder which way is better to generate a .voice - lang2textaloud or the .cmd file that uses lang2wav and voicefont
22:23:22 Join Bagder [0] (
22:23:31 Quit pfavr ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]")
22:23:36[IDC]Dragonlang2wav is recommended
22:23:53[IDC]Dragonmauch quicker and less hassle
22:23:57amiconn(I've already done my first attempt with the .cmd, but the voice file is too big - 2.5 MB
22:24:16amiconn(tried 32 kHz, though)
22:24:23[IDC]Dragonthen you need to increase the -v value
22:24:37[IDC]Dragon32 kHz is definitely too much
22:24:49[IDC]Dragontry −−resample 16
22:25:07[IDC]Dragonand we don't need −−nores any more
22:25:16[IDC]Dragonthis also saves space
22:25:29amiconnIf -V9 is not enough to fit it with 32kHz, I will first try to use −−lowpass nn rather than −−resample nn
22:25:42amiconnYes, −−nores is already thrown out
22:25:55[IDC]Dragon-v does lowpass, this is how it works, iirc
22:26:16[IDC]Dragon32 kHz is way out
22:26:49amiconnused -V 7 now, down to 1.8 MB
22:27:21[IDC]Dragonwhy are you so keen on 32 kHz?
22:27:29[IDC]Dragondigital out user?
22:28:17amiconnNo, I even don't possess any device (other than the Archos) that has an spdif connector. But there may be others that want to use it.
22:28:43[IDC]DragonI've given that up for english long time ago.
22:29:04[IDC]Dragongerman has larger "sentences", so it's worse.
22:29:23amiconnit fits with -V 8 and 32 kHz.
22:29:41[IDC]Dragonit should be <1.5 MB
22:29:56[IDC]Dragondoes it?
22:30:00amiconndown to 1.4 MB
22:30:10[IDC]Dragonwhat did you do?
22:30:24amiconnlame -V 8 −−scale 0.6 -t −−vbr-new −−nohist %%f
22:30:42[IDC]Dragonnothing unusual.
22:31:36[IDC]Dragonpersonally, I'd prefer a lower sample rate and less lossy compression, but well.
22:31:51amiconnBtw: The size is almost equal to your mary.voice
22:32:15[IDC]Dragonmary is 22 kHz
22:32:31[IDC]Dragoncrystal is 16
22:32:31amiconnI will generate several versions with different sample frequencies and compare them.
22:33:35amiconnThis sounds really nice!
22:33:50[IDC]Dragoncan you upload one, please?
22:34:28amiconnHmm, the compression is a bit too tight, I get ringing artefacts. Will try with 22 kHz first.
22:45:07amiconnJust detected a funny error in your script (because the Windows shell complained about it) - in line 4 you use as C++-style comment to comment out the "rem" (sic!)
22:45:38[IDC]Dragonoops, in which one?
22:46:18mattzzamiconn: the c++ commenting style is hard wired in joergs brain ;-)
22:46:30*mattzz ducks and covers
22:47:32amiconnI should have known, since there are some even in rockbox sources where they shouldn't exist. ;-)
22:47:54[IDC]Dragonit is, I can barely stand the Rockbox coding style
22:48:18[IDC]Dragonvery restrictive, imho
22:48:31[IDC]Dragon(so unnecessary)
22:49:22 Quit AciD (Read error: 104 (Connection reset by peer))
22:49:46LinusN[IDC]Dragon: we have received a lot of praise for our simple code
22:50:11[IDC]Dragonhow do you mean?
22:50:47LinusNpeople have praised us because the source code is easy to understand
22:51:09[IDC]DragonI'm in no way criticising that
22:51:21[IDC]DragonI was after the // comments
22:51:24LinusN"it is, I can barely stand the Rockbox coding style"
22:52:11[IDC]Dragonhehe, no offence
22:52:41[IDC]Dragonthe // is just in my fingers, mattzz is right
22:53:33[IDC]Dragonand it allows me 3 more precious chars for my comments before they hit the 80 char barrier ;-)
22:54:17amiconn[IDC]Dragon: Is lang2wav unable to generate 16 kHz output directly?
22:54:32[IDC]Dragonamiconn: yes
22:55:09amiconnWould it be possible to integrate 16 kHz (i.e. if the sapi allows that)?
22:55:09[IDC]Dragonthe SAPI output should always run at the "native" output rate of the voice
22:55:22[IDC]DragonMS does a bad job of resampling
22:55:40[IDC]DragonLame is better for that
22:56:27[IDC]Dragonbut what is native, you're right, only experiments can tell where it sounds OK.
22:57:09[IDC]DragonI found 32 kHz for AT+T and 22 kHz for MS best.
22:57:14amiconnI wonder why the high-bandwidth natural voices are called "16 kHz"?
22:57:39[IDC]DragonI think I tried 16, 32 was better.
22:58:17amiconnSo my choice will be (for now) "lame -V 6 −−resample 16 −−scale 0.6 -t −−vbr-new −−nohist"
22:58:34amiconnThis sounds best for german voice "Klara"
22:58:54[IDC]DragonURL? (drool)
22:59:31[IDC]DragonKlara is my grandma's name ...
23:00:42mattzzwhat is the maximum size of an lcd_bitmap?
23:01:06[IDC]Dragonwhat do you mean?
23:01:14[IDC]Dragon112*64, you know"> (and the typo I pm'd is corrected)
23:04:01mattzz[IDC]Dragon: I was just wondering how to create a 9x9 bitmap, how the src has to look like
23:04:27 Quit methangas (" HydraIRC -> <- IRC has never been so cool")
23:04:36Bagdertry bmp2rb
23:05:27[IDC]DragonI'm not sure if the display function works for >8 pixel height, the logo is "striped"
23:05:37Bagderit works for 16
23:05:46Bagderbut no more, afair
23:06:21mattzzI want to create a "checkbox" widget
23:06:29mattzz8x8 is fine though
23:06:50Bagderthen use the bmp2rb tool
23:07:27[IDC]Dragonor think binary
23:07:43[IDC]Dragon(that's how I did icons)
23:08:59mattzzalready have thought binary - but just 8 bits wide ;-) like in the good ol' days
23:09:38amiconnAs I get some pronounciation problems for German, how is this supposed to be corrected - should I change the spelling in the .lang file to get it correct?
23:10:00amiconn(May look strange sometimes)
23:10:12[IDC]Dragonchange it like what?
23:10:27[IDC]Dragonreal misspellings, or just hyphens?
23:10:56amiconnSometimes deliberate misspellings would be necessary.
23:11:17[IDC]DragonI haven't done such yet
23:11:49amiconnI would do that, I'm just asking if this would be "the right way" (tm)
23:11:49[IDC]Dragonthat's why resume is like resumé
23:12:16 Quit Bagder ("Leaving")
23:12:29[IDC]Dragonthere's no right way yet
23:12:57[IDC]Dragonhowever, I'd say that such is specific to a certain TTS engine
23:13:06 Join Bagder [0] (
23:13:18[IDC]Dragonso, non-portable
23:14:43amiconnYes, that's why I ask. Some mispronounciations are engine dependent, while others come from using foreign words that are spoken german-ish, like "byte", "ampere" etc.
23:14:46[IDC]Dragonwhen I did the german for the L&H voice, I was also adjusting
23:15:04[IDC]DragonEverything had to be very german
23:16:48[IDC]Dragonthat's why I couldn't use established foreighn words like: playlist, frame
23:17:01mattzzBagder: bmp2rb does a nice job!
23:18:57amiconn[IDC]Dragon: So what do you think would be best? There are some words in that are not easily replacable, like "loudness".
23:20:16[IDC]DragonI wouldn't try to be perfect
23:20:26scott666could you just use english strings inside of voice files of other languages?
23:20:41scott666err, not strings, clips
23:20:58mattzzBagder: <whisper> but it prints one comma too much in the last line </whisper>
23:21:04[IDC]Dragonand have different speakers?
23:21:29Bagdermattzz: that's left to not take away all the fun from you ;-)
23:21:56mattzzBagder: I know, I know
23:21:57amiconnscott666: This would be possible if the whole clip contains words of the same language (although it would sound strange with different voices)
23:22:27[IDC]Dragonyou can mix manually before running "voicefont"
23:23:53[IDC]DragonKlara doesn't soundlike my grandma
23:24:19[IDC]Dragonway better than L&H
23:25:19[IDC]Dragonwe've got something for Bagder
23:25:40*Bagder pays attention
23:26:07[IDC]Dragonplacing Jens' voice file to the haxx page
23:26:39amiconn[IDC]Dragon: I would rather like to do some corrections (some things should be possible with hyphenation, e.g. Klara sometimes gets "au" wrong)
23:27:07amiconnThis would be spoken a-ua
23:33:06amiconnBagder: I found a type on the web page, and have a complementing info as well. Unfortunately I'm not familiar with the file format used for the web page in CVS.
23:34:28amiconn[IDC]Dragon: What may cause that the month in "set time and date" is not voiced?
23:35:02 Quit Bagder ("Leaving")
23:35:05 Join Bagder [0] (
23:35:28Bagderamiconn: so tell me and I'll fix
23:37:02LinusNamiconn: the month issue may be my fault
23:37:31amiconnBagder: (1) In /cvs.html, first <h2> the "repository" misses on "o"
23:38:48amiconn(2) In /docs/features.html: The Archos firmware supports ID3v1, but not v2, at least on the recorder v1 (this is stated unknown)
23:41:00LinusNamiconn: fix for month issue committed
23:41:02Bagderthe format is simply html that is processed with a C preprocessor
23:41:46Bagderamiconn: fixes committed
23:41:53scott666while you're at it, see:
23:41:57amiconnHmm, I though it is something strange since it has a .t extension
23:42:19scott666unless did that already and i missed it
23:42:22Bagderamiconn: that's simply for easy Makefile creation, all .t are processed into .html ones
23:42:33[IDC]Dragonamiconn: line 864 of screens.c is wrong
23:42:50[IDC]Dragontalk_id(LANG_MONTH_JANUARY + value - 1, false);
23:42:56[IDC]Dragonshould rather be
23:43:53[IDC]Dragontalk_id(LANG_MONTH_JANUARY + tm->tm_mon, false);
23:44:09amiconn?? In my copy of screens.c (current cvs) line 864 reads "int julianday;" ??
23:44:28[IDC]Dragonthen it's not from today
23:44:53[IDC]Dragonline 846, sorry
23:45:08[IDC]Dragon(finger twister)
23:48:26amiconnI have now 10 problematic words / clips on my list after a complete walk-through (I think)
23:50:42LinusNamiconn: just update from cvs, i have already fixed that bug
23:51:56[IDC]Dragonoh, I'm too slow, was just about to commit
23:53:07***Saving seen data "./dancer.seen"
23:53:42amiconnLinusN: Thanks
23:57:06amiconn[IDC]Dragon: I think I will fix the pronounciation where it is possible with hyphenation, or where it is simply a matter of words, and commit that.
23:57:30amiconnFurther corrections with deliberate misspelling I will do locally.

Previous day | Next day