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 2005-06-24

00:00:10preglowas long as the aliens don't intercept them, sure
00:00:33bippleIm being serious...
00:01:02bippleDoes the hardware allow such frequency to dip to airplanes
00:01:03FlemmardLinusN, as I understand, lcd_framebuffer contains the "matrix" of the screen?
00:01:11amiconnThe radio chip can only handle the FM broadcast frequency range
00:01:19amiconn87.5..108 MHz
00:01:30crwlless than that, something like 7x...108
00:01:38bipplegod dammit
00:01:42amiconn(and iirc it could also handle the japanese frequency band)
00:01:55crwlyes, that's probably it
00:02:00Plugh_my FMR hard drive spews RFI in that whole frequency range
00:02:04bippleWhats the japanese frequency band ?
00:02:26Bagderyes, its from 78
00:02:33BagderI believe our wiki page is wrong
00:02:41Bagdersomeone checked before
00:02:49bipplehmmm what uses anything below 98 ?
00:03:06amiconn76..91 MHz
00:03:16amiconn(from the TEA5767 datasheet)
00:03:16 Quit webguest13 ("CGI:IRC")
00:03:33bippleWell, what would be broadcast from 76 to about 87... anything
00:03:42preglowbipple: fm radio stations
00:03:56Plugh_TV channel 3?
00:04:09bippleTHe current irver radio only goes to 87 doesnt it ?
00:04:11preglowjust forget being able to listen to other transmissions than commercial fm radio, i sincerely doubt that's possible
00:04:24preglowbipple: that's a firmware restriction
00:04:32Bagderbipple: not the japanese version
00:04:46bippleThat means any stations below 87 cant be recieved ?
00:05:06bipplewhich right handed blue tit did this :|
00:05:06n0bbywhat stations bellow 87?
00:05:07crwliriver radio goes down to 76
00:05:23crwlyou can set the tuner frequency range from the settings, at least in the US firmware version
00:05:24bippleThats what im asking whats below 87...
00:05:37crwlprobably no stations in the US or europe
00:05:45 Quit Nuxator ("(Neeeed Sleeeeep)")
00:05:56bippleawww cheese
00:06:13bipplei bought a cheap pair of wireless headphones in spain
00:06:33bippleand that cud pick up taxi radio sigansl and radio stations when cars went past
00:07:00bippleI even had somone say down the headphones please turn all mobiles and somthing somthing off..
00:07:17n0bbythose wouldnt be FM
00:07:55bipplecould i replace the iriver chip with the wirless headphone chip
00:08:07preglowyes, you could
00:08:11preglowbut it wouldn't work afterwards
00:08:15 Join Jleagle [0] (
00:08:38bipple Does MP3 work properly yet
00:08:51bipplelast time i had rockbox it nearly bricked my player
00:09:00preglowi sincerely doubt that
00:09:14bippleI dont know, took everyone ages in here to fix it...
00:09:15Bagderhow can you be near to brick it?
00:09:38Bagdersince you didn't, we all were as near ;-)
00:09:40bippleIt started when the MP3 wouldnt play...then it wouldnt boot up
00:09:56LinusNbipple: because the battery was low?
00:10:05FlemmardLinusN, a small question, why in lcd_framebuffer, we have LCD_HEIGHT/8 ? why /8 ?
00:10:12bippleNah, battery was full, even had the AC plug in
00:10:18BagderFlemmard: 8 bits per byte
00:10:24preglowbipple: after you have flashed it with the bootloader, all danger of bricking is over,
00:10:24LinusNFlemmard: each byte has 8 pixels
00:10:25Bagder8 pixels per byte
00:10:46LinusNpreglow: the v2 bootloader that is
00:10:48Flemmardbmp works with vertical lines?
00:10:49bippleI had to plug the USB in in the end to get it to boot, then i had to remove all the rockbox files before it would work agen
00:11:00BagderFlemmard: no, but the lcd does
00:11:04LinusNFlemmard: not bmp, but the iriver lcd
00:11:19Bagderecho in here? ;-)
00:11:26Flemmardok thx
00:11:33LinusNi have to go
00:11:41preglowme too
00:11:41LinusNcu around guys
00:11:42*Flemmard will study save_dump function
00:11:47Flemmardcu LinusN
00:11:48 Part LinusN
00:11:48 Part preglow
00:11:58Flemmardscreen_dump meme
00:12:41FlemmardBagder: want to try the basis of my draw plugin? :x lol
00:12:58amiconnBagder: As we are on that topic: Do you think it would be useful to add an option to switch FM bands?
00:13:07BagderFlemmard: not right now, busy with other stuff
00:13:12Flemmardnp ^^
00:13:19Flemmard(someone want to test? :x lol)
00:13:20Bagderamiconn: probably, yes
00:13:30amiconnI guess there'll be not many archoses (or even irivers) in Japan, but someone might travel to Japan...
00:14:25Jleagleit will be usefull for people abroad
00:14:25amiconnI don't know whether this is possible with the old samsung tuner though, but with the Philips its just flipping a config bit, and adapting the freqency display
00:15:19bippleHow far is radio finished ?
00:15:39bippleI consider radio more important than MP3, i shud of bought a cheap radio really
00:15:46 Quit Sucka ("a bird in the bush is worth two in your house")
00:15:48bippleRadio is like sex on a spoon
00:16:29Jleaglemp3 > radio :P
00:16:50Flemmardin rockbox will it be possible to record radio?
00:17:06amiconnShould be.
00:17:08bipplei thought that was a hardware limit recording it
00:17:11bippleReally ?
00:17:14amiconnThe hardware does allow it
00:17:26bipplewhat about harddrive noise
00:17:47Jleagleyou wont get any because there is no mike
00:17:57 Quit Bger (Read error: 104 (Connection reset by peer))
00:18:01bipplelol obviosuly
00:18:03amiconnThere could be interference from the HD
00:18:12bippleyeh thats what i meant
00:18:15amiconnCan't tell, never tried the iriver radio
00:18:23 Quit TCK (Read error: 145 (Connection timed out))
00:18:36bippleWill we be able to select stereo ?
00:18:38BagderI enabled it by accident once
00:18:47Bagdercouldn't figure out how to get out...
00:18:48 Join Bger [0] (~Bager@
00:19:02MoosIriver said it's HD interference, but this is the true ?
00:19:28amiconnDepends on the grounding
00:19:40amiconnCan be different between individual units
00:19:43MoosHD and radio can't play simultaneus for iriver
00:19:54MoosIriver Compagny
00:20:09Moosbut it's possibly lies
00:20:26bippletell me lies tell me sweet little lies (8)
00:20:36 Quit Aison|cinema ("( :: NoNameScript 3.72 :: )")
00:21:05Jleagleyer that would be excellent
00:21:15Jleaglewhy would they lie though lol?
00:21:31bippleWell it is iriver...
00:21:50bipplethey promised Firmware upgrades for about 2 years
00:21:50amiconnBagder: Did someone try AM reception with the samsung tuner back then?
00:21:54Moos:) it is a commercial company ;)
00:22:01Bagderamiconn: I can't recall any
00:22:06amiconn...or are there some parts missing for AM reception?
00:22:13amiconnThe chip is AM/FM
00:22:13bippleMost they did since is change the version number on the boot screen
00:22:14Jleagleyer true lol
00:22:37bippleIt can recieve AM ?
00:22:47bippleI was just told it could recieve FM
00:23:05bipplenext ill hear it can recieve MW and LW
00:23:46amiconnbipple: I'm talking about the samsung tuner chip used in archos fm recorder and old Ondio fm
00:23:52Bagderyou guys should pay attention
00:24:12bippleim a girl
00:24:17amiconnNewer Ondio FM has the same chip as the iriver, Philips TEA5767, and that can do FM only
00:24:38bipplechoddy old iriver shines thru agen
00:25:49bipplerockbox should make an MP3 player
00:25:56amiconnNah, the philips chip has way better FM reception than the old samsung
00:26:55 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
00:27:03 Join Stryke` [0] (
00:27:57Jleaglecan sumone explain to me how the wiki works, if i wanted to add something to the rockbox could i just download, edit and reupload my version???
00:29:04Bagderyou click edit
00:29:08Bagderedit the page
00:29:10Bagderthen save
00:29:15 Quit ep0ch (" HydraIRC -> <- State of the art IRC")
00:29:22Jleagleso whats to stop me deleteing eveything?
00:29:25n0bbytheres an in-browser editor
00:29:34n0bbyand an undo button
00:30:13Jleaglewow thats pretty trust worthy, i tink id have to add some sort of protection, it gets backed up all the time though right?
00:30:26n0bbyit keeps records of every change
00:30:35n0bbyit can be reverted as easily as changed
00:30:39Jleaglehmm thats good
00:30:51BagderJleagle: if you check the bottom of each page, you can check different versions and changes
00:30:53n0bbywikipedia survives quite well :)
00:31:57 Quit bipple ("CGI:IRC")
00:32:21Jleagleyer, ive never seen the admin side of a wiki, its great that you can just revert the changes
00:32:57Jleagle*goes off to install a wiki*
00:33:21Bagderthe most annoying part is people adding spam/links to pages
00:34:17Jleaglei can imagine, kinda inevitable though
00:34:50Bagderbut the pros outweights the cons
00:37:25 Quit Febs ("Chatzilla [Firefox 1.0.4/20050511]")
00:37:55 Quit t0mas (Read error: 60 (Operation timed out))
00:38:19Jleagledoes it require any knowledge of c to install patches?
00:38:47Bagdervery often, yes
00:38:50 Join lostlogic [0] (
00:39:00Bagderpatches are source code changes
00:39:16Bagdersource code is mainly C in Rockbox
00:39:53Jleaglei thought there might be instructions, like add * after *
00:40:13Bagderthey are "instructions" for the patch command
00:40:26 Join TCK [0] (
00:40:31Bagderbut sometimes they don't apply cleanly, then you need to do parts of it manually
00:41:09Jleaglehmm ok, i might give it a go
00:41:23n0bbyi'd rather whine for the code to get commited
00:41:39n0bbyi got the bitmapness in the daily doing that :)
00:42:11Bagderonly new rockbox coders fall for whine ;-)
00:42:21Jleaglelol badger
00:42:21n0bbybitmap in WPS
00:43:09 Nick Flemmard is now known as Maxime`Mrn (
00:43:13amiconnHummm. I should build new voice files...
00:43:50Jleagleooh! on the topic, i cant get mine to work... i put the english.voice in the lang dir, but nothing happens, you got any idears?
00:43:55 Quit Harpy (Read error: 145 (Connection timed out))
00:44:06amiconnJleagle: Doesn't work on iriver yet
00:45:22Jleagleheh, somebody told me it did, thanx ;)
00:45:47amiconnThat will take a while..
00:46:18amiconn...and it's one of the reasons why I still prefer my archoses for real usage
00:46:30amiconnThe voice UI is sooo handy in the car...
00:46:37HClwhy doesn't it work on iriver
00:47:52Jleaglehasnt been coded yet?
00:48:24HClwhats required for it?
00:48:53amiconn(1) A second codec, or at least a method to control a codec and feed it tiny snippets from ram
00:49:07amiconn(2) some changes in the talk code itself
00:49:14amiconn(3) Probably a format change
00:49:19HClits not my thing anyways...
00:49:34amiconnThe current voice files contain mp3, but pre-bitswapped for archos
00:49:35 Join yyz [0] (
00:50:13amiconnmp3 would be fine for a start, but the pre-swap has to go
00:50:41amiconnWe could either swap on load on archos, or simply make 2 sets of voice files
00:50:53amiconn(needed for other codec than mp3 anyway)
00:51:38Moosapear it needed a lot of work :(
00:52:10amiconnThe hardest (from my point of view) is the codec part
00:52:22amiconnThe other changes may be numerous, but simple
00:52:41Moospreglow know this?
00:53:09Moosfor the codec part
00:54:00amiconnI'd prefer a second libmad instance, statically linked
00:54:29amiconnIf we use mp3, it would make sense to use the same voice file for both archos and iriver
00:54:54 Join preglwo [0] (
00:54:58amiconnSince there is the size constraint for archos (RAM) the voice files are relatively small compared to iriver ram size
00:54:59preglwoi agree with amiconn on this
00:55:13preglwothe voice codec(s) should be statically linked
00:55:15Mooshi preglow we could have them loaded at all times when the voice ui is enabled
00:55:39preglwoi also believe we should use speex on the platforms where it's possible
00:55:47preglwobroadband mode gives excellent speech at pretty nice bitrates
00:55:51amiconn-> no reload delay, and also voice *while* playing
00:55:56preglwoi'm also hoping it will be a fast decode
00:55:56amiconn(impossible on archos)
00:56:04 Quit Chamois (" HydraIRC -> <- 100,000+ downloads can't be wrong")
00:56:20Bagderhaving the speach codec linked at another place would be the way to go, imho
00:56:27preglwolibspeex looks really optimisation friendly
00:56:35amiconnpreglwo: I know mp3 is not the best codec for speech, but using it would have advantages
00:56:37preglwoi would have started porting it right now, if i could
00:56:53preglwoamiconn: like? being able to use the same sets for archos?
00:57:02amiconnYes, that for one
00:57:05Bagderand the same scripts
00:57:10amiconn...and there is also the talkbox feature...
00:57:10preglwoamiconn: when you've made the recording, just doing two encodes should be fast
00:57:45amiconn...which plays mp3 file snippets announcing the directory names etc
00:58:03 Join ashridah [0] (
00:58:30amiconnI'm not worried about encoding twice, but rather about 2 sets of files to download for users
00:58:35preglwo don't see why a different codec option should invalidate all this
00:58:54preglwo2 sets of files to download should be no problem, when they can be marked as specifically as they can
00:59:02preglwoTHIS IF NOT
00:59:04amiconnWell, talkbox should always work with mp3 imho
00:59:14preglwobut anyway
00:59:17preglwoit's not a big deal to me
00:59:40preglwojust think it's weird not supporting the one open source codec out there that was made for this
00:59:47amiconn...and btw, uploading a full set of voice files already takes a while
01:00:26amiconnThey add up to ~10 MB, and ADSL upstream isn't quite fast...
01:00:36preglwowhat bitrate?
01:00:47amiconnErr 1024/128
01:01:01preglwoi mean the files, not the connection
01:01:18amiconnAh, mp3. The snippets are currently 12 kHz vbr, limited to <= 64 kbps
01:01:37amiconnA full voice file is still ~1.3 .. 1.5 MB
01:01:58amiconnThe bitrate limit is because of the MAS limitations
01:02:53preglwospeex gives very nice quality at even 10kbps
01:03:16amiconnNow someone should write a speex decoder in MAS asm...
01:03:17preglwo'nice', at least :P
01:04:16preglwoso upload should at least be a lot smaller than for mp3
01:04:29Plugh_isn't telephone line quality something like 4k?
01:05:12amiconnpreglwo: The file generation shouldn't be a problem as long as there is a command line encoder for windows
01:05:24preglwoamiconn: sure is
01:05:32preglwoPlugh_: yes, but i don't like telephone quality
01:05:36preglwodecent voice needs 16khz sample rate
01:06:08Plugh_5.3k actually
01:06:08amiconnMy build system uses a vbscript controlling SAPI5 voices, lame, and 2 special tools written by [IDC]Dragon
01:06:27preglwoPlugh_: i'm listening to a 4kbps vbr stream as we speak
01:06:35preglwo4kbps cbr is also very possible
01:07:00preglwoanyhow, i'll port speex to rockbox anyway
01:07:01Plugh_g.723 is 5.3/6.4k
01:07:12preglwospeex isn't g.723 :)
01:07:15 Quit n0bby ()
01:07:23 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
01:07:55preglwounless someone beats me to it, that is
01:08:07yyzspeex will come in useful for audiobooks anyway, so worth having
01:08:14preglwoyes, ia gree
01:08:17preglwoi agree, even
01:08:23preglwoi think rockbox should have as many codecs as possible ;)
01:08:31amiconnRebuilding voice files...
01:09:02preglwolibspeex also supports encoding out of the box
01:09:05preglwonow that's a good application
01:09:26 Join Sucka [0] (
01:09:47preglwoBUT, speex does analysis by synthesis, which i think is pretty slow
01:09:54preglwobut we'll se
01:10:00preglwogod, i hate this web client
01:10:15amiconnpreglwo: lame -V 4 -B 64 −−resample 12 −−scale 0.6 −−vbr-new -t -S xx.wav xx.mp3
01:10:33amiconnThis is the most common line used in my voice script
01:10:50preglwowhy the scale?
01:11:41amiconn[IDC]Dragon used full scale first. Bagder said it was too loud compared to the music
01:12:06BagderI doubt I said that
01:12:14Bagderbut I guess that's not the point ;-)
01:12:16Jleagleis speax something that reads out what you type?
01:12:22amiconnWell, maybe it was Zagor
01:12:31amiconnI'm always confusing you ;)
01:12:56*Bagder runs off to bed
01:13:08Jleaglesounds like a good idear
01:13:15preglwodon't run into anything, now
01:13:18Jleagle"is speax something that reads out what you type?"?
01:13:56 Quit _aLF ("Leaving")
01:16:12ashridahJleagle: no, it's an audio codec oriented towards encoding speech
01:17:01Jleagleoh ok thnaks, it would be cool to have something that could speak out text like MS Sam
01:17:23Jleaglewould save alot of trouble
01:18:01ashridahit probably would. pity we can't find the source code to dr sbaitso, which is about the only text reading app i knew that ran on a 286. bet the code's a horrible horrible horrible mess tho :)
01:18:03yyztrouble? what are you? a hostage taker
01:18:25yyzi dont think ive ever once thought "damn i wish i had a portable speach synthesiser"
01:18:32ashridahyyz: if anything, he, or someone he knows, is probably blind
01:18:58yyzyes, I understand their usage on a computer, but on a portable device?
01:19:45ashridahcatering for reading impared users is something rockbox strives to be good at
01:19:51Jleaglei dont mean just for fun, you wouldnt need to have voice packs, you could just set it to say somethign in the code
01:20:57Jleagletheres a ~1MB program called ShitTalker which does that if anyone is interested, the owner might give away the source
01:21:12preglwowe'll probably try that some time, jleagle
01:21:17preglwobut it's pretty complicated
01:21:39ashridahJleagle: there's probably one or two that are already small. the problem is, it'll only support the one language, i'm betting.
01:21:43*amiconn wonders why people often want to do the second step before the first
01:21:51ashridahone that supports every language will be huge :(
01:21:59Jleagleyer i suppose
01:22:07preglwoamiconn: because the second step is more fun, of course
01:22:28Jleaglewe do have 20gB+ :D... yer thats right preglwo lol
01:22:35amiconnI don't like half-baked things for one
01:22:59Jleaglewhat would rockbox be without idears?
01:23:06amiconnI think what iriver rockbox needs is to get all current rockbox features to work
01:23:14Jleaglethats true
01:23:34ashridahi agree. get it to the same standard the archos port is
01:24:00amiconnBug-free playback with all options working, radio, recording...
01:24:08Jleaglethen start adding extras when its all 'done' maybe not looking good but no errors
01:24:21amiconn...voice UI
01:24:32amiconn...and probably *some* extended hardware features taken into account
01:24:40preglwospeex uses 2.6 seconds to encode 60 seconds of 8khz audio here the multi-codec support already does
01:24:56amiconn...and 4-grey mode will also be
01:26:05Jleagle4 grey?
01:26:12preglwo5.5 seconds to encode 60 seconds of 16khz audio ://
01:26:17preglwothis doesn't look too good
01:27:38amiconnfloating point?
01:28:18preglwoguess so
01:28:18Jleaglei gtg, tired... cya later
01:28:23preglwoand this is a pretty fast box
01:28:31preglwo2.2 ghz athlon 64
01:30:06preglwokilled powernowd, now runs at 4.6 seconds
01:30:13preglwoanywho, this is irrelevant, i'm still going to try
01:30:16 Join kenshin [0] (
01:30:39amiconnSomething to apply EMAC magic to ;)
01:30:56 Quit Jleagle ("Chatzilla [Firefox 1.0.4/20050511]")
01:31:37preglwoi can apply much faster filtering, per clock, on a coldfire than on this box
01:31:39preglwoof that i'm certain
01:32:08amiconnI really appreciate what you've been doing
01:32:21preglwowhat what? ;)
01:32:30preglwoi haven't been doing much lately
01:32:30amiconnI'm pretty certain I couldn't manage to do that kind of stuff
01:32:47preglwoi'm pretty sure you could, it's not advanced stuff
01:33:10preglwothanks anyhow, i'll take the compliment
01:33:19amiconnI would have to know way more about codecs... and audio filtering and the like
01:33:37amiconnI know some of the theory...
01:33:51amiconn...but the mathematics...
01:34:00preglwoi know a bunch of the theory, maths, and love to code it
01:34:06preglwoso it's all good for me
01:34:50preglwoeven though codecs weren't what i thought i'd be doing when i came aboard to do, i think it's fun, actually
01:34:59preglwo-"to do"
01:35:43amiconnI prefer to do that bitshifting stuff, that's what I know
01:36:13amiconn...and digging in datasheets how to control this and that piece of hardware even better
01:36:26preglwohaven't done too much of that
01:36:39preglwothe most low-level stuff i've done is writing an lcd screen driver, i guess
01:36:48preglwoin avr assembler
01:36:58preglwofor the fun of it, that's the sick part
01:37:00amiconnI did the MMC driver for ondio
01:37:45amiconnThere's also some nice asm in the grayscale lib
01:37:59preglwoi'm really starting to enjoy asm, that's what worries me...
01:38:17amiconnasm can be fun.
01:38:42preglwoeven though you'd think that imdct_l failure would have put me off asm forever
01:39:11amiconnStill no idea what would make that slower?
01:39:52amiconnWhen I didn't understand why a certain C construct was actually faster than my asm, I checked what gcc threw together
01:39:58preglwolinus said he'd have a look at it, but i don't think it's his top priority
01:40:13amiconnSometimes gcc generated asm is brilliant, and sometimes pure shit
01:40:35preglwomyeah, but i haven't got a c version of the algo i use
01:40:39preglwoonly an arm version
01:41:13amiconnAh, completely different algorithm.
01:41:28preglwoi thought it'd be tons faster, 'cause it's got a loop in it, will utilise the cache much better
01:41:32preglwoand it has all constants in tables
01:41:43preglwonot in the straight asm code, like the other one
01:41:45preglwobut still...
01:42:16preglwoan optimisation guide from motorola would have been appreciated
01:42:39markunabout text to speech, the ipodlinux guy that comes here said he had flite running.
01:42:54preglwoi don't know squat about how/if the internal pipelining works, how the emac unit is pipelined, how/when the parallel move is executed
01:43:04preglwomarkun: details?
01:43:30***Saving seen data "./dancer.seen"
01:43:55markunIt's a c reimplementation of festival:
01:44:11preglwosweet zombie jesus, the tar.bz2 is SEVEN megs
01:45:22markunMaybe it contains some voice data.
01:45:32preglwoi most certainly hope so
01:45:56markunThe output of festival is not so bad, but you have to get used to it.
01:46:20preglwoi think the output of festival is pretty good
01:46:25markunIt listened to quite a big part of the hitchhiker's guide with it :)
01:46:37preglwoyou're a bigger man than me
01:46:48preglwoi would have gone insane
01:46:57yyzshoulda just got h2g2 radio play, its loads better :)
01:47:07*amiconn doesn't know how festival sounds
01:47:29yyzi tried to listen to some book once, but it was Dark Elf and all the fantasy novel type made up works really screwed up my enjoyment
01:47:32markunYes, that's true. I have nice ogg versions but was thinking of buying the CD's.
01:48:01amiconnI know that Linus said MBROLA sounds like sh**, and I'm used to having a nice AT&T natural voice reading my menus...
01:49:14amiconnBut then I also know the good old narrator.device on AMiga
01:49:44yyztrying to get the right phonetic spelling was a challenge in itself
01:50:42thegeeknot too bad
01:51:04thegeekif I remember correctly at&t is not _that_ much better?
01:51:16thegeekit's been more than a year since I tested it though;)
01:52:44 Join preglow [0] (
01:52:44 Quit preglwo ("CGI:IRC (EOF)")
01:53:31amiconnthegeek: at&t is clearly better
01:53:58amiconnfestival sounds more like the microsoft voices, quality wise
01:54:02preglowhow i love pop-unders
01:54:20preglowwhere can i sample this at&t thing?
01:54:25 Join sockerteze [0] (
01:55:13thegeekyou are right
01:55:19thegeekI must have forgotten
01:55:20thegeekbut damn
01:55:32thegeekit takes like 2 full cd's ;)
01:55:39 Quit preglow (Client Quit)
01:55:54 Join preglow [0] (
01:56:03preglowwhy do i torture myself with this web interface...
01:56:09preglowanywho, the at&t thing sure sounded nice
01:56:19amiconnthegeek: 1 CD per voice
01:56:34thegeekfor each language?
01:56:40amiconnThe base package is 2 CDs, and includes 2 voices
01:56:40thegeekdamn, I remember I had the full set
01:56:42thegeekwas it that large?
01:57:03amiconnI have one additional voice, german
01:57:24preglowthough it handles some phonetic transitions awfully
01:57:36amiconn2 GB installation
01:57:36preglowsounds like they've just got a big database of sounds that they paste together
01:57:52thegeekI don't think they do
01:58:04preglowthen that's a bloody awful solution
01:58:05thegeekbecause it is still synthesis
01:58:10thegeekbut I agree
01:58:13preglowsome words sound great
01:58:16amiconnThey don't simply paste, but they do have a huge database
01:58:19thegeekthey probably have very detailed descriptions of some words
01:59:05 Quit Moos ("Bye ALL")
01:59:45 Quit sockerteze ("Chatzilla [Firefox 1.0.4/20050511]")
02:02:41preglowit surely sounds sweet sometimes
02:04:04preglowthe german one sounded really sweet
02:05:38 Quit markun ()
02:08:11 Quit ghostiger (Ping timeout: 14400 seconds)
02:09:56 Quit Sucka ("Leaving")
02:10:57 Join amx [0] (
02:11:54preglowmarvelous, seems like i'm getting a cold
02:13:49 Join amiconn_ [0] (
02:18:10preglowdo we have a software codec define?
02:18:36preglowi see the codecs arent compiled for h300 currently
02:25:02 Quit preglow ("CGI:IRC")
02:25:50 Join yyz- [0] (
02:31:26 Quit amiconn (Read error: 110 (Connection timed out))
02:31:26 Nick amiconn_ is now known as amiconn (
02:32:57 Quit yyz (Read error: 110 (Connection timed out))
02:33:56 Nick yyz- is now known as yyz (
02:42:00 Quit RotAtoR ()
03:21:06 Quit yyz ("Leaving")
03:38:26 Quit kenshin (Read error: 110 (Connection timed out))
03:43:33***Saving seen data "./dancer.seen"
03:53:41 Quit hardeep ("[BX] Amount of time you people were funny: --32 minutes -28 seconds")
03:55:15 Quit courtc (Read error: 110 (Connection timed out))
03:56:09 Join courtc [0] (
04:05:56 Join QT_ [0] (as@area51.users.madwifi)
04:09:58 Join kenshin [0] (
04:17:35 Quit QT (Read error: 110 (Connection timed out))
04:27:51 Quit amx ("Client Exiting")
04:40:53 Quit Heidelbaer (Read error: 110 (Connection timed out))
04:56:56 Join pittsburghZombie [0] (
04:57:18 Part pittsburghZombie
05:43:35***Saving seen data "./dancer.seen"
06:07:21 Join asdsd____ [0] (
06:07:43 Quit asdsd____ (Client Quit)
06:22:50 Join Heidelbaer [0] (
06:24:46 Quit kenshin (Read error: 110 (Connection timed out))
06:40:59 Join kenshin [0] (
06:42:19 Join hardeep [0] (hardeeps@SDF.LONESTAR.ORG)
07:02:00 Quit hicks (Client Quit)
07:06:29 Nick QT_ is now known as QT (as@area51.users.madwifi)
07:18:16 Quit kenshin (Read error: 110 (Connection timed out))
07:21:41 Quit ashridah (Read error: 110 (Connection timed out))
07:35:15 Quit hardeep ("BitchX Lite I said!")
07:43:36***Saving seen data "./dancer.seen"
07:49:23 Join Thasp [0] (
07:51:54 Join asdsd____ [0] (
07:51:55 Quit MrStaticVoid ("Lost terminal")
07:52:12 Quit asdsd____ (Client Quit)
08:05:35 Join ashridah [0] (
08:12:07 Join Coldtoast [0] (
08:20:25 Join DMJC [0] (
08:22:05Bgerwiki spam ...
08:26:36Bgerwho's Marcoen Hirschberg
08:27:01 Join markun [0] (
08:27:08markunBger: I'm marcoen :)
08:27:20Bgermarkun: u're very quick ;)
08:27:28markunYou too
08:28:09markunIs there an easyer way to place back the old revision than to just edit the page?
08:28:25Bgeri don't know
08:29:17markunWell, I have to go back to study for an exam now.. Bye.
08:29:25Bgerbye ;)
08:29:32 Quit markun (Client Quit)
08:37:29 Join Cassandra [0] (
08:44:57Cassandra*sighs* Another CD to return because of stupid copy protection.
08:46:17QTmoin. be happy cassandra. copy protection is the best excuse to return the CD after you have ripped it *gg*
08:46:53QTgo to the shop and act very dumb saying "it doesn't work in my car cd player"
08:47:56CassandraDamn thing won't rip properly.
08:47:57QTi always get my money back then
08:48:15CassandraOh, I've returned it before.
08:48:24QTi am usually very lucky using cdda2wav and putting the disc into a DVD writer drive
08:48:43CassandraIn the UK our consumer legislation says it has to be fit for the intended purpose.
08:48:54QTwhat does that mean?
08:49:14QTi mean if it does not work it does not work −−> faulty product
08:49:24CassandraIt means that since I can't put it on my MP3 player, it counts as faulty.
08:49:50QTone guy in the shop once wanted to argue with me about giving me a refund
08:49:57QTthen it is even better!
08:50:25CassandraYeah. If I could actually rip it. :(
08:50:40QTthe guy at the counter meant it would be possible that i copied it. but how should it be possible if the disc has copy protection? i then asked to speak his boss and a few mins later i have my refund
08:50:57QTtry different drives and different rippers
08:51:30QTdo you know the UnCD database?
08:51:41CassandraI once complained to Sony about a disc I'd returned and they send me a free copy. Which, oddly, ripped without a problem.
08:52:27 Join kenshin [0] (
08:53:14QTsearch on this site for your CD:
08:58:39 Join einhirn [0] (
09:03:40 Nick Lynx_awy is now known as Lynx_ (
09:09:47CassandraA couple of successful rips there, but not with drives I own.
09:10:13BgerCassandra: some friend with such drives ?
09:10:45CassandraMight as well just stick it in my housemate's CD system. That'll copy it.
09:11:04CassandraBut paying the RIAA tax to get a clean copy f***s me off.
09:12:30Bgernot only u ...............
09:13:42CassandraI will return it regardless. I'm not paying money for that crap puporting to be a CD.
09:14:52 Join bobTHC [0] (
09:16:32 Quit bobTHC (Client Quit)
09:26:33 Quit ashridah ("Leaving")
09:28:55 Join Bgr [0] (~Bager@
09:28:55 Quit Bger (Read error: 104 (Connection reset by peer))
09:29:45 Join Querty [0] (
09:35:07 Join oxygen77 [0] (
09:35:31 Quit oxygen77 (Client Quit)
09:43:39***Saving seen data "./dancer.seen"
09:46:53 Quit Maxime`Mrn ()
09:51:17CassandraRight, there's a letter to EMI written. (Well, editted from the one I sent Warner Music.)
09:53:05 Join Maxime`Mrn [0] (
09:53:16CassandraHey ami.
09:54:08CassandraI'm still wanting to change the iRiver keybindings. Need to make sure people are happy with them before I write the patch though.
09:54:37 Quit Sando (Read error: 104 (Connection reset by peer))
09:55:26CassandraI really can't see the point in binding play to the joystick button, since right on a music file is already bound to play in the WPS. (Or if it isn't, it should be.)
10:05:51 Join bobTHC [0] (
10:06:12 Quit linuxstb ("Leaving")
10:16:14 Join Sandoaway [0] (lolsteam@
10:22:37 Quit bobTHC ("Smoke Weed Every Day !")
10:26:09 Quit Cassandra ("If I were actually witty, this quitline would be funny.")
10:30:58 Join ashridah [0] (
10:40:43 Join Harpy [0] (
10:45:51 Join t0mas [0] (
10:47:34 Join Sando [0] (lolsteam@
10:47:34 Quit Sando (Remote closed the connection)
10:47:42 Join Sando [0] (lolsteam@
10:54:31 Quit Sandoaway (Read error: 60 (Operation timed out))
11:05:20 Join chuch [0] (
11:05:39chuchanyone alive ?
11:06:40ashridahmore or less
11:08:33chuchcool i got a 80gb travelstar with a user password on it and atapwd isnt unlocking it
11:09:03chuchanyone got any ideas i've used hddunlock and even went as far as trying some xbox utils
11:09:39ashridahfind the person who set the password and beat them until they reveal it?
11:10:43chuchbought it at a tradeshow
11:11:00ashridahhope you've got a receipt
11:12:43 Join Bger [0] (~Bager@
11:13:27chuchthat's it
11:15:24chuchanyone got some insight on this ?
11:18:13amiconnAhahaha! rotflmao, I found the fix for bug #1174245. Stooopid bug, that! Someone meant to do a bitwise not, but used a logical not...
11:20:08t0mashm? change from ! to ~
11:20:23amiconnyup... trivial, once the mistake is found.
11:20:44ashridahchuch: find vendor. beat vendor. get refund.
11:20:52amiconnhi t0mas
11:21:02t0maschuch: find vendor and ask for masterpassword is a better way ;-)
11:21:05t0masamiconn: hi
11:21:07ashridahor call manufacturer. explain problem. then find and beat vendor when manufacturer informs you the password cannot be reset
11:21:31t0masamiconn: and bye again... time for breakfast
11:21:37 Quit Bgr (Read error: 110 (Connection timed out))
11:29:06 Join LinusN [0] (
11:29:17 Part LinusN
11:30:10 Join linuxstb [0] (
11:31:42 Quit Bger (Nick collision from services.)
11:32:06 Join Bger [0] (~Bager@
11:33:00chuchso what's the use of this atapwd util that i found on
11:34:08Bgerchuch do you have relevant problem ?
11:42:23chuchyah i can't get it too unlock a drive
11:42:44ashridahthe question becomes, are you sure it's locked in the same fashion?
11:42:48chuchi got at user lock hd that im having a problem unlocking
11:43:18chuchashridah guess not
11:43:21 Join yyz [0] (
11:43:43***Saving seen data "./dancer.seen"
11:44:21ashridahbecause it sounds like you're facing a newer version of the problem that the archos units would have had
11:44:53ashridahi remember seeing some commentary about the annoying nature of newer versions of drive locking
11:45:18chuchyah i just read a 3 page forum post on expertexchange about it
11:59:03 Join Moos [0] (
12:00:14HCldrive locking is evil
12:00:17HClthe xbox does it too
12:04:11 Join n0bby [0] (
12:04:21n0bbyanyone here got CVS write access
12:04:33n0bbyreally minor change to a songDB file needed
12:04:36n0bbythe .bat file
12:04:39n0bbysince it doesnt work anymore
12:05:07n0bbyit'd be handy to tell the user to install perl and let them use the bat, but without perl.exe being called
12:05:12n0bbyi've tried it, it works
12:05:53n0bbyalso, does songdb support ogg files yet?
12:06:39ashridahthe java songdb does, yes
12:07:02n0bbytheres a java implementation?
12:07:58ashridahon the id3 database page, attached
12:08:31n0bbythe page didnt seem to mention it
12:08:50ashridahthe new songdb is currently still being developed and documented a bit
12:09:02ashridahmore of the technical side has been covered than user-oriented stuff
12:09:05n0bbyhow do i run a .jar on windows? winrar has stolen its assosiation
12:09:49Bgern0bby do you have sun java ?
12:09:55n0bbyi think so
12:10:03n0bbyi have azureus and it requires it AFAIK
12:10:07Bgerthen just java -jar ...
12:10:13Bger... = filename
12:11:05n0bbyit just output the usage
12:11:08n0bbyrather than run it
12:12:23n0bbyException in thread "main" java.lang.NullPointerException
12:12:23n0bby at TagDatabase.calcLimits(
12:12:23n0bby at TagDatabase.writeDatabase(
12:12:23DBUGEnqueued KICK n0bby
12:12:23n0bby at SongDB.main(
12:13:05n0bbygot it now
12:13:28Bgerafter that give directory name
12:13:37Bgeri mean java - jar songdb.jar dirname
12:13:44Bgerwith / slashes
12:13:57n0bby"java -jar songdb.jar .." for if its in .rockbox
12:14:13n0bbyshould add that to the wiki
12:14:23CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
12:14:23*n0bby does so
12:15:18n0bbysomeone else can do it
12:15:31*n0bby releases edit lock
12:15:34 Quit n0bby ()
12:19:35ashridahi honestly thought it already had
12:20:25HCli should fix it so it refuses to do anything if it can't find any tags
12:41:43Coldtoastin the menu, "Status-/Scrollbar"
12:41:54Coldtoastwhy's there a - after Status?
12:42:19Coldtoastseems unneeded to me
12:42:50amiconnIt's an abbrevition for Statusbar/Scrollbar
12:43:00Coldtoasthmmm. ok
12:43:34ColdtoastI don't think you need it personally
12:44:20Coldtoastmaybe Status/Scroll Bar
12:45:06Moosit' just a retail
12:51:05 Join Aison [0] (
12:51:16Coldtoastwhat's just a what?
12:51:33Moosyes scuse
12:51:53Coldtoastok. it is, yes. But why not mention it if I notice it is my theory?
12:53:01Moosfeel free to mention all you want
12:58:56 Quit kenshin (Read error: 110 (Connection timed out))
12:59:04ashridahthe problem is, status/scroll bar is an if-then-if-then-else conflict
12:59:22ashridahie, is it [status|scroll]bar or [status|scrollbar]
12:59:25t0mas<n0bby> anyone here got CVS write access <−− still needed?
13:01:17 Join Cassandra [0] (
13:03:05*amiconn sighs
13:03:20amiconnThat gfx api will be quite a bit of work...
13:03:42t0masyou have a vacation comming? ;-)
13:03:49amiconnNot the implementation itself, but the required changes in the parts that use it.
13:03:55amiconnI am on vacation
13:04:28amiconnI've just updated the api proposal
13:06:08amiconnOh, btw, the new gfx api will feature full clipping support
13:07:31Cassandraami: Any thoughts on changing the iRiver key mappings as per my mail to the list? I don't want to go ahead and do it without polling the dev team first.
13:08:07 Join Febs [0] (
13:09:39amiconnCassandra: What changes do you mean? There was quite some discussion...
13:11:13CassandraI think basically I'd want to shift play to the play/on key. Long press becomes context menu.
13:11:42CassandraNot quite sure what to do about the joystick button. Leaving it as another play key seems wasteful.
13:12:01CassandraEsp given that right ought to start playing while in WPS anyway.
13:12:04amiconnI think the current solution is better
13:12:48amiconn(1) Imho using select to play from the browser is intuitive, because I select something
13:12:49 Quit edx (Read error: 110 (Connection timed out))
13:13:13amiconn(2) There are problems related using PLAY for playing, since PLAY is also the poweron button
13:13:41amiconnThere is a reason why play is *not* BUTTON_PLAY, but BUTTON_ON ...
13:14:24amiconn..and doubling the play functionality on both SELECT and PLAY is wasteful
13:14:35CassandraThat'd explain why I always get ased if I want to resume, even when I was last at the end of a playlist.
13:14:36amiconn...and leaves the question how to resume...
13:14:49amiconnCassandra: No, that's a different story
13:15:06amiconnThe playback code is still buggy regarding playlists
13:15:24Cassandra*nods* The end of some tracks are getting clipped too.
13:17:36amiconnImho the current button assignment is quite logical:
13:17:59CassandraHmm. This might come down to a personal taste thing. For me it really isn't.
13:18:00amiconnIf you want to *select* a new track to play, use SELECT (or RIGHT)
13:18:17amiconnIf you just want to *play* something (=resume), press PLAY
13:18:41CassandraIt's certainly not logical that the play function and the pause function are bound to different keys.
13:19:07CassandraNor should pause be bound to select, since the play button has convenient pocket access.
13:19:20amiconn(1) On tape decks this was the plain standard
13:19:22yyzthe only thing I dont like about the current buttons is that pressing LEFT from root file menu doesnt take you back to WPS, you need to press PLAY
13:19:27yyzthat annoys me every single time
13:19:49yyzbut everything else seems perfectly logical to me, like amiconn says
13:20:10amiconn(2) It's just a coincidence that play and pause are identical on archos player& recorder
13:20:16amiconnOn ondio they are not
13:20:34Cassandra(2) Not really. Every commercial CD player ever works like this.
13:20:48 Join dave__ [0] (
13:21:02 Quit dave__ (Client Quit)
13:21:13 Join dave__ [0] (
13:22:14 Part dave__ ("Leaving")
13:22:21amiconnApart from that, play and pause aren't different, if you look at it the way I see it
13:22:51 Join Sucka [0] (
13:25:07amiconnThinking like: To just play something, press play. This resumes on iriver, and does a similar thing on a cd player, since it can only play the currently inserted disk. The same button is pause.
13:25:43amiconnTo play something new, you currently have to use select. However, this also (roughly) equals a cd player:
13:25:59amiconnTo play something new on a cd player, you have to change disks.
13:26:11amiconn...and that's not done with the play button, right?
13:26:41Cassandra*nods* OK. That's quite convincing. Still feels wrong to me, but ...
13:27:21CassandraAlthough of course resume is currently PLAY, SELECT.
13:27:35CassandraPLAY, PLAY would be more logical, don't you think?
13:27:41amiconnIf you have it set to ask, then yes
13:27:59amiconnThe text is still wrong, you probably know why
13:28:22amiconnI agree that accepting *both* select and play as positive answer would be logical
13:28:35CassandraWe don't have platform dependent language strings.
13:29:52amiconnWhat do you think?
13:30:02amiconn(about selct and play counting as 'yes')
13:30:24Cassandra*nods* Can't think of a good reason why not.
13:31:06CassandraI wonder what I'm going to call the play button in the manual. Calling it play will confuse people, I think.
13:31:17CassandraIt confused me, certainly.
13:31:24amiconnAh ,yes
13:31:38amiconnFor me, it is still the PLAY button
13:33:38CassandraI see it as a sort of hybrid RESUME/PAUSE/ON.
13:33:47CassandraThe problem is, you can't call it that.
13:34:04 Quit yyz (Read error: 145 (Connection timed out))
13:35:00ashridahi wouldn't mind having play act as 'resume' if the player's just started, even if resume is set to no (or perhaps 'only on play' or something)
13:35:58amiconnashridah: It does that
13:36:08*Cassandra nods. There's a lot of sense in that. A request to resume by a specific button shouldn't require confirmation.
13:36:20amiconnYou can always resume with PLAY regardless of the resume setting
13:36:41amiconnThe first press will bring up the resume request if there is something to resume
13:36:48CassandraIf resume is set to "no" do you get prompted, or does it just work?
13:37:02amiconnAh umm, good question
13:37:11*amiconn tries
13:37:55CassandraI'm beginning to think we should map the on button's resume functionality to behaviour as per resume = yes.
13:38:50amiconnHumm, it really doesn't work
13:39:13amiconnI think the resume setting should be changed to be a 'resume on startup' setting
13:39:32CassandraI think you're right.
13:39:35amiconnThis means that the 'ask once' functionality has to go though
13:39:46CassandraDoes it?
13:40:04amiconnAsk once destroys the resume info if you answer no
13:40:29 Join edx [0] (
13:40:29CassandraHmmm. Maybe it shouldn't.
13:40:58amiconnI think the resume on startup should obey the resume setting, but while running, the PLAY button (= ON on archos recorder etc) should always resume
13:41:23amiconn...and for this to work, the resume info must not be destroyed
13:41:32CassandraI agree, although that's potentially confusing for new users.
13:42:31CassandraBecause they may expect the resume (PLAY) button to play the current track (like I did).
13:42:59amiconnThat can be explained in the manual...
13:43:21CassandraIf you made it act like the SELECT button if there was no resume info, would that be more or less confusing? some other rockbox subtleties need explanation
13:43:32*ashridah notes that the current button assignments in rockbox for the iriver were just thrown in haphazardly when it couldn't play music
13:43:45***Saving seen data "./dancer.seen"
13:43:59Cassandra*nods* That's why we're trying to rethink them.
13:44:04amiconn..e.g. the volume versus treble/bass problem. I *love* that discussion
13:44:19t0masamiconn: saw my idea on that?
13:44:22amiconn...and I definitely prefer the rockbox way
13:44:25ashridahamiconn: to be honest, i kinda side with you. i'd just love better feedback that it's started volume limiting
13:44:34ashridahie, the counter can't progress if it's hit the limit
13:44:43t0masthat was my idea
13:45:06t0maswith an option to pass that point (with a notice) and start having iriver-fw behaviour from that point
13:45:06amiconnashridah: Hmm. That won't be easy, because the limiting happens in low-level code, not in high level
13:45:25amiconnt0mas: Why????
13:45:28ashridahbeing able to choose which kind of anti-clipping limiting happens might be nice, but *shrug*
13:45:34t0masbecause everybody is happy then?
13:45:42amiconnYou can always increase the volume further. Just reduce bass/treble
13:45:46CassandraShouldn't be too difficult to programatically calculate the max from the bass and treble settings.
13:45:52t0massome people want to stop at that point (you), and some people want to continue... and reduce bass/treble
13:46:26CassandraI seem to get distortion if I put the bass setting up much.
13:46:32amiconnCassandra: Yes, but that requires more knowledge of the low-level limits in the application layer. Not good...
13:46:37t0masthen you might have an old version Cassandra
13:47:02Cassandra*nods* Maybe I'll try it again.
13:47:12amiconnNo, that distortion results from too much filtering. This is not clipping
13:47:29CassandraToo much filtering, what's that?
13:48:23amiconnThe filter curve gets so steep that the waveform overshoots
13:48:56amiconnAt least I think that's what causes the distortion
13:49:24amiconnThere can't be clipping because the limiter is working very strict
13:49:26CassandraRight. That doesn't mean anything to me really. Is there a potential solution?
13:49:41ashridahcould i get that in plain english and simple diagrams? :)
13:49:44amiconn...reducing the mixer gain by MAX(treble_boost, bass_boost)
13:49:52*ashridah was never very good at audio/electronic engineering
13:50:43CassandraProbably ought to be done by someone who knows what they're doing at some point.
13:50:49amiconnCassandra: I'm not sure, but perhaps this can be solved by implemeting software eq, depending on the filter type
13:51:11amiconnAnyway, 24 dB bass boost sounds like crap for almost everything
13:51:30CassandraI was getting this as low as 11.
13:51:41CassandraMaybe I need to delete my old config file.
13:51:46amiconnThere is no 11...
13:52:05amiconnBass enhancement a la MDB might do a better job for earphones
13:52:14CassandraThere is on my iRiver. Possibly because I saved my original config when the scale was wrong.
13:52:37amiconn11 is impossible hardware-wise, the hardware works in 2 dB steps
13:52:52amiconnI think you'll get 10 when you feed it 11 from a config file
13:52:52t0masoh lol
13:53:33amiconnThe bass setting just shifts right by 1 to get the hardware value
13:53:41CassandraI'm going to go fix up resume, I think.
13:54:00 Quit Sando (Read error: 104 (Connection reset by peer))
13:54:09Cassandraami - can you think of any reason I shouldn't make the resume button play the current track if there's no resume info?
13:54:22CassandraThis is certainly behaviour I'd like.
13:54:35amiconnHmm, I would think that's confusing
13:55:54 Join bipak [0] (
13:56:24 Join RotAtoR [0] (
13:56:29CassandraIt's one of those UI design philosophy things. Do we try to guess what the user "wanted" to happen and risk being confusing, or do we make it rigid and risk having it be confusing in a different way?
13:57:13amiconnImho too much guesswork is more confusing, but this is me.
13:57:34ashridahand no matter which one you choose, you either end up confusing previous users of rockbox, previous users of iriver's firmware, or both :)
13:58:24CassandraI feel a config file option coming on.
13:58:40amiconnOh no! No more config options... ;-)
13:58:59CassandraWell, you know. They were looking a little thin on the ground.
13:59:45 Quit Febs ("Chatzilla [Firefox 1.0.4/20050511]")
14:02:21CassandraArgh. I've just remembered how much I hate the resume logic.
14:07:17CassandraAha! I thought that'd be the case.
14:07:37CassandraControls are inconsistent between the remote and the main unit.
14:07:44CassandraThis definitely *is* confusing.
14:07:59CassandraOn the remote, PLAY = SELECT.
14:09:51amiconnAaargh! No more broken plugins pls!
14:11:03 Quit bipak_ (Read error: 110 (Connection timed out))
14:13:12CassandraHoly christ, this is just wrong. Buttons labelled the same on the remote and the unit should do the same thing.
14:14:26CassandraThis is not helped by the remote having three more buttons than the main unit.
14:14:26amiconnIf so, the assignment is probably wrong
14:14:48CassandraYes and no. The menu button *ought* to open the config menu.
14:15:12CassandraHowever, for consistency A-B ought to.
14:15:37 Join kenshin [0] (
14:15:45CassandraBut then what do we map SELECT to?
14:16:29CassandraThis is actually bad design on iRiver's behalf, but we're stuck with trying to turn it into something sensible.
14:18:10CassandraI think we have to ignore the MENU text on the remote, and map SELECT to NAVI/MENU
14:19:06amiconnIt does the same in iriver fw
14:19:21amiconnNAVI/MENU opens the file browser as does SELECT on the main unit
14:19:32CassandraOK. That's cool.
14:19:55CassandraThere's no mapping for the RC_MODE button in button.h.
14:20:52CassandraOh, it's labelled VOL.
14:20:54CassandraNever mind.
14:21:45amiconnIt's also labelled A-B/MODE (below the control)
14:22:01CassandraI meant in the source it's labelled VOL.
14:22:22CassandraNot the best name for it, but ...
14:22:33amiconnImho it should be renamed to BUTTON_RC_MODE
14:22:44CassandraOK will do that.
14:23:28 Join cYmen [0] (
14:27:44CassandraOh dammit. Now I need to install the 68k compiler.
14:35:09 Quit Sucka ("Leaving")
14:44:55 Join Sucka [0] (
14:49:54amiconnI wonder whether I should commit parts of the gfx api while it is work in progress, when I have a working state
14:50:24amiconnI would need to break plugin api compatibility almost every time to avoid crashing plugins
14:50:44Bgeramiconn: if you don't commit, please, back-up ;)
14:54:52CassandraYou know I keep thinking it ought to be possible for plugins to be compiled standalone.
14:55:35CassandraWell so plugins need not necessarily be distributed with rockbox.
14:55:54amiconnI still don't understand...
14:55:55CassandraMake them more like executables for the Rockbox platform.
14:56:10 Join west-acre [0] (
14:56:14amiconnThey are...
14:56:17west-acrehey there
14:56:53CassandraOh, maybe I'm just mad then. Any plugin with the same API no. can run on any Rockbox?
14:56:57amiconn...that's why the plugin api version exists
14:58:11amiconnA plugin can run on rockbox if it's own api version number is less or equal than that of the core *and* equal or greater than the min api version number of the core
14:58:25amiconn...and of course the device model matches
14:59:05amiconnWhat I mean when talking about breaking api compatibility is that I have to bump the min api version number of the core
14:59:17amiconn...because of incompatible changes to the api
14:59:18CassandraI wonder why more people don't distribute plugins independently. older plugins don't run
15:00:04amiconnI would have to do that almost everytime I commit a new stage of my gfx rework
15:00:34CassandraAren't you supposed to be able to create a new branch for stuff like this?
15:00:41amiconn...because functions get ripped out and replaced with fewer, more flexible functions
15:00:52amiconnRockbox doesn't use braches
15:01:21CassandraCan't see why you couldn't create one to work on then merge back in later.
15:01:35CassandraAlthough I have to admit branches confuse me.
15:01:36amiconnI already have the single pixel stuff implemeted
15:02:16amiconnNah, at least I would have to talk to one of the 3 Swedes before doing so, and of course learn how to work with branches in cvs...
15:02:49CassandraCan we start calling them the Three Wise Swedes?
15:03:17amiconnHmm, are you sure?
15:03:28amiconnI am not... ;)
15:03:48CassandraSsh. Or they'll take away your CVS access.
15:05:06CassandraPah, why does compiling a little applet like gcc take so damn long?
15:05:39 Join Jleagle [0] (
15:05:56DBUGEnqueued KICK west-acre
15:05:56west-acrehey wots bin happening with the rockbox port over the past gew days, i got fading backlight, any more?
15:06:40Jleagleyou could see the change logs, i cant see a link to any other than todays though
15:07:11west-acreyeh, i could only see todays, that's why i asked :D
15:07:41amiconnDaily builds page, under 'source' there is a link 'older'
15:08:00amiconnOn that page, there is a 'changelog' link for every daily build listed
15:08:23amiconn(one month worth)
15:08:34Cassandraami: Was it you that fixed the sleep time while powered on? Idle timer needs fixing too.
15:12:44amiconnI left the idle timer as-is on purpose
15:13:16amiconnI don't want the unit to idle poweroff while connected to the charger
15:13:31amiconnYou can still poweroff with OFF, that was already fixed
15:13:40CassandraIt always has with the Archos.
15:13:40gromit`hrm :)
15:13:47t0masgromit`: wrong window? ;)
15:13:50gromit`ya :)
15:14:09west-acreImage support for the WPS. ?
15:14:13t0masah better than wget http://some.pronsite.xx ;-)
15:14:15amiconnCassandra: The archso can't be powered off while the charger is connected, that's a hardware limitation
15:14:22gromit`heh :)
15:14:32t0maswest-acre: hm?
15:14:33amiconnThe iriver can, and you can do that, with the OFF button
15:14:34CassandraMaybe I'm just going mad then.
15:14:38t0maswest-acre: you can display bmp images on the wps
15:14:53 Join ilikedirt [0] (
15:14:54 Quit cYmen (Read error: 110 (Connection timed out))
15:14:58gromit`browsing pornsites with wget is quite unefficient
15:15:18t0massome people really are stupid...
15:15:27amiconnCassandra: A couple of people agreed with me that idle poweroff should be inhibited while the charger is connected
15:15:40t0masI asked someone to send me some mp3's... but re-encode them... as he uses 64 bits and I don't like that sound :P
15:15:55t0masnow he re-encoded the 64 kbit mp3s as 320... not the original CD
15:16:25t0masso now I have 320 kbits mp3s wich still sound bad
15:17:37CassandraWhat was the reason, just out of curiosity.
15:18:14CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
15:18:14*t0mas is intrested too
15:18:43linuxstbIMO, idle poweroff is there to save power - which is meaningless when you are plugged into the mains.
15:18:58linuxstbIf I want to turn it off, I know where the off button is.
15:19:15t0masI don't when I'm sleeping...
15:19:22t0masand I've repeat off...
15:19:25CassandraWhat about saving electricity.
15:19:35CassandraI often put an album on when I go to bed.
15:19:39t0masso after playing some album... it should stop and shutdown... even when the charger is plugged in
15:19:53linuxstbAnother config option?
15:19:54*linuxstb ducks
15:20:18amiconnCassandra: Use the sleep timer, then it does power off
15:20:27amiconnThat's why I fixed the sleep timer behaviour
15:20:37amiconnOtherwise I agree with linuxstb
15:21:04CassandraI guess. But setting the sleep timer is not convenient.
15:21:20CassandraHmm thats a candidate for addition to the iRiver F2 menu analogue.
15:22:02 Quit DMJC (Connection timed out)
15:22:08CassandraWhich by the way there seem to be two main candidates for. Long press on MODE or short press on REC.
15:22:19CassandraI think I favour long press on MODE.
15:22:41Cassandra(although that means we then can't put an A-B repeat function on that key)
15:23:14 Quit ilikedirt ("Chatzilla 0.9.68a [Firefox 1.0/20050524]")
15:23:24amiconnI don't need an F2 equivalent
15:23:36CassandraI don't need one. I want one though.
15:23:45amiconnI don't want one either
15:23:45CassandraJust makes life easier.
15:23:58CassandraI do. :P
15:23:58 Join DMJC [0] (
15:25:38CassandraHeya HCl.
15:26:03CassandraAnyone else strongly against a quick menu for iRiver?
15:26:29amiconnCassandra: A long press can't be called quick...
15:26:32HClwhat do you mean quick menu?
15:26:42CassandraF2 on the Archos.
15:26:46HCla menu you can quickly access while playing?
15:27:06Cassandraami - it's still quicker than navigating through the settings.
15:28:13amiconnImho this can't be done comfortable on iriver. MODE is on the side while the joystick is on top
15:28:27HCl?, you can't press another button at the same time you press MODE
15:28:36*HCl missed something..
15:28:46HCli favor long press on mode
15:29:07amiconnHCl: The quick menus are supposed to work really quick.
15:29:29CassandraAm I the only person to hold my Archos in the palm of my hand and use the joystick with my thumb?
15:29:41HClamiconn: why does it matter?
15:29:47CassandraOr other hand.
15:29:59amiconnE.g. the F2 menu is brought up by pressing and *holding* F2, then selecting a direction for the desired option
15:30:00HClbetter than not having quick access to a menu at all?
15:30:13amiconnLeaving off F2 will immediately close the quickmenu again
15:30:18HClwhat i need for the runtime database is a menu you can bring up while playing
15:30:38Cassandraami: Nope. Just checked.
15:30:38HCli don't need to have a quickmenu in the archos sense, then.
15:30:40amiconnHCl: Add it to the wps context menu
15:30:49HClamiconn: okay, wasn't sure whether we had that :)
15:30:58amiconnWe have that since y'day
15:31:05HClah :P
15:31:13amiconnLinus committed an old patch of Henrik Backe
15:31:18HCl :)
15:31:22HCli'll check out and update my player
15:31:28HCland work on the runtime database
15:32:10CassandraThat's very cool.
15:32:30Cassandraami: I think you're misremembering how the quick menu works.
15:32:48amiconnThey work in 2 ways, I know
15:33:01amiconnBut the way I described is the *really* quick way
15:33:21amiconnI still have some working archoses to check...
15:33:25CassandraOh. Didn't know about that. Better add it to the manual at some point.
15:34:14CassandraIt really is incredibly hard to use one handed though. I'm still not convinced by your objection.
15:34:27amiconnYou can press F2 and leave off again, and the menu will stay. Then you can select your option, and have to press F2 again to get rid of the menu
15:34:34amiconnNot really quick I'd say...
15:35:15amiconnIf you mean that way, I wouldn't call it quickmenu, it'd just be yet another menu...
15:35:25CassandraWhy can't you press another button at the same time as MODE? Generic limitation of the iRiver keys?
15:36:02HClcurse this forum xD why is the spell check button at the spot where you expect the button for post to be
15:36:05amiconnmode/left/right/up/down/select are all hooked up to the same adc channel
15:36:25amiconn+ record
15:37:07amiconnThe only independent buttons are PLAY and STOP
15:37:14CassandraHowever, I'd still argue it's quicker than using mode to change them.
15:37:16amiconnSame goes for the remote
15:37:48amiconnAll remote buttons except PLAY and STOP are hookep up to a single adc channel
15:37:49CassandraLong press on ON is currently unbound.
15:41:33HClgood work on mp3 playback
15:41:38HClfirst time i'm playing an mp3 with rockbox
15:41:39HCl :)
15:41:59HCli need a better wps though
15:42:09CassandraIt's a beautiful thing, isn't it?
15:42:30HCltesting my runtime db hack i added
15:42:34HClshould trigger at the end of the song
15:43:00HCllcd works nicely as debug output
15:43:03HCli mean, remote
15:43:20HCli do notice tons of disk activity though
15:43:22HClis that normal
15:43:49***Saving seen data "./dancer.seen"
15:43:52CassandraDoes seem to be, yes.
15:43:54ashridahHCl: currently, rockbox caches around a minute of the first song, then waits for a minute, and caches as much of the playlist as it can
15:44:02*HCl checks logs
15:44:05HClashridah: i see.
15:44:18CassandraI think it should start by caching 5 secs, so that playback starts immediately.
15:44:22ashridahthis is a distinct contrast to iriver's firmware, which only caches the current song, and even then, possibly not much of it
15:44:23CassandraThen a minute.
15:44:27CassandraThen the playlist.
15:44:33HClashridah: *nods*
15:44:42HClwe run at higher cpu frequencies though.
15:44:48ashridahCassandra: 5 seconds is longer than the hard drive can spin down and back up and settle, i suspect
15:44:55HCli don't suppose anyone has made any comparisment in battery usage yet?
15:45:10ashridahhrm, a few people have.
15:45:13CassandraCurrently the two seconds or so it takes to start is quite disconcerion.
15:45:20ashridahi don't know the results. it's not horrible tho
15:45:32amiconnThe playback should simply start if there is buffered enough to not run out while buffering more
15:45:42amiconnThis is how it's done on archos
15:45:54ashridahCassandra: it's entirely possible that the player could start playing as soon as the codec has its first frame, it'd be a lot more work tho, i imagine
15:46:18*HCl 's debug message didn't show up
15:46:22CassandraWe should do it though. Even iRiver does.
15:46:39amiconnashridah: Not the first frame, but enough to not run out of data immediately
15:46:46*HCl wonders whether the distortion he hears is caused by their crappy livingroom speakers or rockbox..
15:46:50ashridahamiconn: hm, true.
15:46:57amiconnAnd it's not that hard, if the buffering is properly implemented
15:47:12HClSlasheri: around ? :/
15:47:37 Quit Jleagle ("Chatzilla [Firefox 1.0.4/20050511]")
15:47:47*ashridah notes that mp3 playback still mysteriously cuts out and the player immediately shuts down for him on a particular playlist he uses
15:48:03ashridahi need to get around to poking at it to see if it's a particular song, or a length-of-time issue or what
15:48:08CassandraI don't think our codec expert is on at the moment, sadly.
15:48:51CassandraWhat the hell is "Ask Once" as a resume option for exactly?
15:49:10*HCl 's hack to add runtime db stuff didn't work.. must wait for slasheri :/
15:49:22amiconnWell, it does ask only once...
15:49:40*HCl goes to work on school stuff instead
15:49:41CassandraYes, but why would anyone want the unit to behave in this way?
15:50:07amiconnSome people might be bothered about the resume? request when it does appear every time
15:50:36amiconnIf we change the resume setting to only define the resume behaviour on startup, we don't need it, as I already stated
15:50:55CassandraWe don't?
15:51:17amiconnThose people should then set 'resume on startup' to 'no' and if they occasionally want to resume,
15:51:18 Quit Bger (Read error: 54 (Connection reset by peer))
15:51:29amiconnthey could still do that using PLAY
15:52:13amiconnBtw, these resume changes practically forbid using PLAY as a synonym to SELECT
15:52:15 Join Bger [0] (~Bager@
15:52:23amiconn..if there is no resume info
15:52:40CassandraDo they?
15:52:51CassandraMy brain is full of cotton wool today.
15:52:51amiconnThe user often won't know whether there is resume info or not, and so using PLAY can be very surprising
15:53:13 Join fogcat [0] (
15:53:17Cassandra*nods* Yeah. I'd thought about that. It's relying on the user keeping state in their head.
15:53:28amiconnYes... and that's bad
15:53:44CassandraI think we should make play flash up "No playlist" if there's no playlist loaded.
15:53:55amiconnI would certainly run into big trouble, having 4 different units...
15:54:27amiconnYes, that might be a good idea
15:54:58CassandraStill trying to work out if there's an easy way to leave ask once in without zeroing the playlist.
15:55:29amiconnDrop that, I'd say
15:55:56amiconnI think the changes should be as follows:
15:55:56CassandraI may do that, yes.
15:56:17amiconn(1) Resume info should always be stored, regardless of the resume setting
15:56:35amiconn(2) The resume setting should be renamed 'resume on startup'
15:56:40amiconn...and behave that way
15:56:59amiconnResume from the browser pushing PLAY should always be done, without a request
15:57:06 Quit fogcat (Client Quit)
15:57:14Cassandra*nods* All three of those I agree with.
15:57:41amiconn...and if there is no resume info, it should splash that message you suggested
15:58:04CassandraOr maybe "No resume info" would be better.
15:59:42 Join webguest38 [0] (
16:00:55 Join fogcat [0] (
16:02:41fogcatwith the WPS getting more graphical are there any plans to change the screen definition "language"?
16:03:03t0masyes, I've been thinking about it
16:03:21t0masbut couldn't find a smart way to do it... without losing functionaliy we have now
16:03:29fogcati was having a Friday afternoon muse about that - current system is v line orientated
16:04:02fogcatwas wondering about a VERY cut down version of HTML style div placing
16:04:17t0masdifficult to code...
16:04:28amiconnThere is no need to think about no-line oriented
16:04:32t0masbut feel free to post a proposal to the mailinglist, wiki or forum
16:04:42*t0mas is away again :)
16:04:43fogcatahh yeah... well that's where not knowing the code frees my imagination ;-)
16:04:53amiconnIt's a non-no until at least two things are sorted, one of them being veery hard
16:05:07HClman its nice to finally be close to being 100% again
16:05:14*HCl codes for school
16:05:44fogcatI did have a peek at the source (but I last coded any C about 12 years ago)
16:07:04 Join tvelocity [0] (
16:12:44 Join preglow [0] (
16:13:26*preglow FINALLY got a cold
16:13:49ashridahyou say that like it's some kind of acheivement
16:14:15ashridahthat said, that means you can stay home and work on rockbox. \o/
16:14:19preglowi haven't had one for a couple of years, and i've been praising the almighty lord for every day
16:14:36ashridahheh, i went through a few years like that. was great
16:14:39preglowno i can't, i need to hand in my master's thesis in a week
16:14:47 Quit webguest38 ("CGI:IRC (Ping timeout)")
16:14:50zecouple days ago i got a sore throat for no apparent reason
16:14:52zestill have it :/
16:15:04preglowthere's no mistaking this
16:15:04HCl :/
16:15:08HClokay thats not funny
16:15:09preglowsore throat, stuffy nose
16:15:13*HCl just recovered from being ill :/
16:15:25HClsounds like more than a cold preglow :/
16:15:29HClisn't a cold just a stuffy nose?
16:15:31zeits weird, i had no contact with anyone sick
16:15:32preglowdiminshing sense of taste ://////
16:15:41preglowno, i always get a sore throat
16:15:44zeHCl: often sore throat and cough too
16:15:55preglowi chatted with two germans with a cold two days ago
16:15:58preglowi wonder where i got it from...
16:16:00ashridahHCl: it is, i've been through the same thing. i got a cold, then a few weeks later, i developed a nasty cough that lasted for a few weeks. finally shook it off, and then my partner for a large project got mono
16:16:37preglowi bloody HATE having a cold
16:16:42 Quit fogcat ("CGI:IRC (EOF)")
16:17:44HCl :/
16:18:07HClmarkun: added the function stuff to bla.g, its on my ftp in the eindopdr dir
16:18:16 Join hicks [0] (~hicks@90ce51b15f4503ba.session.tor)
16:19:15HClour cat has it so hot that he's not sleeping on his pillow anymore, cause it gets too warm
16:20:01*ashridah stabs steam in it's crappy, non-threaded ass
16:20:31zeyour cat has it that way? he paw around with the thermostat?
16:21:09HClits just really warm
16:21:11zeoh heh
16:21:12ashridahze: uh. cats are generally furrier than people
16:21:13HCltropical temperatures
16:21:23zeashridah: usually
16:21:29HClalso cuter.
16:21:32HCldon't forget cuter.
16:21:33zeHCl: often
16:21:41HCl :d
16:22:09zeever notice the more feline characteristics a woman has the hotter she tends to be? or is that just me... heh
16:22:54*HCl stares at his computer.
16:23:12HClhm, something is going definately wrong.
16:23:23ashridah<HCl> it's fogging up
16:23:34t0masI've found a very irritating bug...
16:23:53t0masif you stop playback, and insert 1 song into an empty playlist, it starts playing...
16:24:04t0masthen you stay in filebrowser to add some more songs
16:24:11t0masand you keep adding them
16:24:22*HCl goes to reboot, the magical fix thing for windows computers.
16:24:25t0masthen after some time (before the end of the first song) the music stops, and it goes to stop mode
16:24:41amiconnHmrf. This bit masking, xoring and inverting drives me crazy
16:24:46t0masand it loses the playlist you've been editing... because it stops
16:25:34*amiconn needs to imagine what the individual draw modes are *really* supposed to do
16:29:08HClhow odd.
16:29:22HCla program i use daily just suddenly started crashing without a reason
16:29:31HCland now it just crashes when i try to start it..
16:32:02CassandraHmm - am I right in thinking that global.settings.resume.index isn't currently getting reset at the end of an iRiver playlist?
16:32:37amiconnglobal_settings.resume_index it is. But I think you're right, and it's a bug
16:32:57Cassandra*nods* Any idea where I should look to fix it?
16:33:28amiconnI would ask Slasheri
16:33:45amiconnYou could also try to compare how this is handled on archos
16:33:59amiconnarchos code is in firmware/mpeg.c
16:34:07amiconniriver code is in apps/playback.c
16:39:36preglowcats are cool
16:39:41preglowi might be getting one when i move
16:39:50*Cassandra has 4.
16:39:57CassandraBut only 1 brain between them.
16:40:02preglowof course
16:40:25 Join rasher [0] ("> <−− Matrix plugin available
16:41:17preglowbut i gotta go buy something for my throat
16:41:30rasherexcept it's somehow broken, and I can't figure out why
16:41:40rasherCassandra: I know!
16:41:48CassandraI wish people'd stop writing damn plugins. I keep having to document the bastard things.
16:41:59rasherWell this doesn't do anything
16:42:03rasher"press stop to exit"
16:42:05rasherthat's it
16:42:08preglowit's testable?
16:42:09CassandraNo offense to you, rasher. Sounds cool.
16:42:20rasherthe animation just dies after a few seconds
16:42:24CassandraI still have to do a screenshot. Well, per platform.
16:42:50preglowrasher: this'll be cool with grayscale!
16:43:00rasherpreglow: not with the current font it won't, but yes
16:43:27rasherShould use the font from the xmatrix screensaver I suppose
16:43:30preglowrasher: why not? the font doesn't have to be grayscale to be able to scale it to other shades of gray
16:43:38rasherOh, true
16:43:49rasherbut it's so thin and pretty aliased
16:44:27rasherwell, noticably line-ish
16:45:35preglowoh, well
16:45:42preglowi gotta go back to writing
16:45:50preglowand possiblt killing myself
16:45:54 Quit preglow ("CGI:IRC 0.5.4 (2004/01/29)")
16:46:08 Quit rasher ("CGI:IRC 0.5.4 (2004/01/29)")
16:54:47 Join Rori [0] (
16:55:00Roriwhich lossless codec is playing back best at the moment?
16:55:18amiconnHow do you mean?
16:55:28Roriwhich lossless compression codec
16:55:28CassandraIs it just me, or is seeking completely failing to work in oggs?
16:55:29amiconnBoth plain wav and wavpack are playing nicely
16:55:39Roriape, flac, wv?
16:55:41amiconnSeeking doesn't work in ogg yetr
16:55:58amiconnRori: ape isn't supported, and flac has performance problems
16:56:19amiconnSo wavpack is your best option
16:58:04Rorican you ff/rw wavpack?
17:01:58 Join hardeep [0] (
17:07:26 Join webguest02 [0] (
17:08:13 Quit webguest02 (Client Quit)
17:16:00linuxstbRori: Yes, seeking in wavpack works.
17:17:19linuxstbI agree that wavpack is the best lossless codec on the iRiver. It's probably the most worked-on codec apart from MP3, and the work has been done by the wavpack author himself (David Bryant).
17:20:55 Part courtc ("Leaving")
17:22:56 Join Strath [0] (
17:27:07Rorimaybe record direct to wavpack? :)
17:27:28 Quit Querty ("CGI:IRC")
17:29:01Roriwhat is a .m4a file?
17:31:11linuxstbApple use .m4a for both AAC and ALAC (Apple Lossless Audio Codec). But it's probably AAC.
17:31:56linuxstbRecording to wavpack should be possible - wavpack encoding and decoding have a similar complexity.
17:33:28Rorithat would rule
17:33:42Rorilossless recording with nearly half the space used
17:35:12RoriI love it when I dig into my CD collection and discover something I forgot I had
17:35:16amiconnThat depends on waveform complexity. My test album got reduced by merely ~33%
17:35:27amiconnStill a bit better than FLAC though
17:35:28Roriit all helps
17:36:41Rorifound a CD not on freedb
17:37:53 Quit Godeater ("CGI:IRC")
17:37:58*amiconn is going to try the new gfx stuff for the first time :holds breath:
17:38:58*ashridah lights prayer candle
17:39:37amiconnThe part I've already done seems to work...
17:41:22 Quit StrathAFK (Read error: 110 (Connection timed out))
17:43:11Mooshi Slasheri
17:43:52Mooshave you got news about bugfix for playback?
17:43:53***Saving seen data "./dancer.seen"
17:44:15Slasherino, i am currently working on dsp support for playback system
17:44:22Slasheriafter that i will try to do more bugfixes
17:45:13amiconnOkay, seems all went well up to now
17:45:20amiconnNext parts...
17:45:44Slasherisounds great :)
17:45:59RoriMP3 still needs a fix for FF/RW accuracy and 22khz playback
17:46:09linuxstbDo we have a list of known bugs for the iRiver playback code?
17:46:26amiconnCurrently I'm doing this in the recorder lcd driver only, meaning in current state iriver won't even compile
17:46:27Slasherilinuxstb: i have some list, maybe i could put that online
17:46:49amiconnBut porting is easy, the routines in question just need to be copied verbatim
17:46:53linuxstbWhat about here?:
17:47:09linuxstbI started that page, but haven't linked to it from anywhere yet...
17:48:51Slasherilinuxstb:, there are something in form of irc logs :)
17:50:00linuxstbSlasheri: OK, I'll try and tidy it up and add it to the IriverStatus page.
17:50:06Slasherigood :)
17:50:29linuxstbI'll try and replicate the bugs before listing them...
17:51:23Slasherithere is also one bug related to buffer filling while player is on motion and can't access really well to the hdd
17:51:33Slasherii don't yet know exactly what causes that
17:51:46 Quit Maxime`Mrn ()
17:52:03Slasheribut it will cause skipping and eventually may crash the system if the hdd can't be accessed
17:57:00amiconnThat shouldn't happen. It should try to recover, stopping the sound meanwhile if the buffer is running out
17:57:17amiconnThat's what the archos implementation does
17:57:47Slasheriyep, but there is some bug somewhere. I have managed to crash the system only two times so i don't know yet what causes it
17:57:58SlasheriThe playback will normally stop and resume
18:07:18 Join sanchny [0] (
18:09:14 Quit ashridah ("Leaving")
18:09:37CassandraSpeaking of which, the resume info isn't being correctly overwritten once a playlist finishes.
18:10:53CassandraAnd the keyboard driver hates me, but I don't think that's anyone's fault.
18:11:29CassandraWhat button event would you expect the keyboard driver to return when you called a blocking read while the ON button was held down?
18:12:17CassandraI was expecting it not to return until I let go of the key, but apparently not.
18:12:30amiconnWhy should it return nothing?
18:12:44CassandraCause my brain isn't working?
18:14:05amiconnIf it would not do that, it would be impossible to receive button_repeat events with blocking reads
18:14:12amiconn...and that would be plain wrong
18:15:08CassandraWhat about if you called a non-blocking read?
18:15:41amiconnThe read method doesn't influence the return value
18:15:52amiconn...doing otherwise would be silly
18:16:05Cassandra while (button_get(false) & TREE_POWER_BTN) sleep(HZ/10);
18:16:23amiconnWhy would you do that????
18:16:24CassandraI would expect that to wait until the ON button is released. It doesn't seem to.
18:18:23amiconnIt should, but it will also eat the first event after the power button was released
18:19:01amiconn(if that event follows within the sleep() time)
18:19:12CassandraWhich is not likely.
18:19:15amiconnJust eating button event is often a rather bad idea
18:19:33CassandraWell it doesn't work anyway. Falls straight through that bit of code.
18:19:45amiconnIt is very likely if the next event is already in the queue
18:20:05amiconnThe button driver is queued, don't forget that
18:20:13amiconn...and that is a good thing imho
18:20:26CassandraWish I knew why it wasn't doing what I expected though.
18:20:40amiconnI don't get what you're trying to do
18:21:05amiconnThere already is code in the resume request to handle the case of a still held down power button
18:21:21CassandraThere was a good reason why I ripped it out.
18:21:31CassandraUnfortunately I can't remember it now.
18:21:46CassandraHowever, I think I'll have a rest and come back at it fresh.
18:21:58CassandraAnd start by putting that code back in.
18:22:30amiconnThe gfx code is also somewhat difficult
18:22:38amiconnSo many cases to handle...
18:22:56amiconnI *want* full clipping support
18:23:01CassandraYeah. A lot of rockbox is like that. It's the downside of such feature rich code.
18:23:09amiconn...with as little additional code as possible
18:23:38RoriI notice you have to be REALLY quick when pushing the remote button to start Rockbox instead of iRiver
18:23:41amiconnFortunately I already have some code I can reference - the grayscale library
18:23:47Roriany way to increase the time?
18:24:02amiconnI can't use it verbativ - too many tradeoffs in that code.
18:24:27amiconnRori: That is a bootloader feature. Removing it requires updating the bootloader
18:25:05Roriit's too darned quick you have to literally flick the button to get it to load Rockbox
18:25:05amiconn...and that should be done by someone knowledgeable equipped with a bdm
18:25:08 Join Febs [0] (
18:25:09amiconn..i.e. Linus
18:26:09CassandraAFAIK you can't start Rockbox from the remote.
18:26:20RoriI'd say a count of 1 second for Rockbox and 3 seconds for iRiver would be best
18:26:21CassandraIt always starts the original firmware.
18:26:24RoriYes you can
18:26:38Roriyou have to press the play button very quickly
18:26:40amiconnCassandra: You have to be really quick
18:26:57*t0mas is reading a pointer orgy from hell
18:27:20CassandraWell *I* can't.
18:27:57t0masI tried it
18:28:03t0masyou have to be _very_ quick
18:28:07t0masbut it's possible
18:28:34 Quit hardeep ("BitchX: ribbed for her pleasure!")
18:28:35CassandraFrom what I remember of the CVS, that's actually a bug.
18:28:37t0masmaybe Linus can add the new idea (< 2 sec = rockbox > 2 sec = iriver) to bootloader v3 ?
18:28:53t0masCassandra: it was... but in that version you didn't have to be so quick
18:28:58amiconnI'd say drop iriver boot from the remote
18:29:01CassandraFrankly I'd rather see him check the record button like with the main unit.
18:29:12t0masthat would be ok to me to
18:29:16amiconnThe record button check is sufficient, imho
18:29:22amiconn...and only on the main unit
18:29:59Roriwhen MP3 at lower Hz is sorted I can safely dump iRiver fw boot anyhow
18:30:00CassandraDunno. It might be kind of useful to be able to start the iRiver firmware from the remote.
18:30:11RoriI need 22Khz etc playback
18:30:19Rorifor audio books
18:31:01t0masRori: update to latest cvs
18:31:05t0masit's working now :)
18:31:14amiconnRori: It should already work (for mp3)
18:31:47Roribut I still want to boot to Rockbox easily from remote (can't wait for LCD browser etc to work on remote I use the remote a lot)
18:32:02amiconnCassandra: As soon as rockbox does all things it should do for a first release, I'd even dump iriver fw completely
18:32:22t0masamiconn: then it'll need a fw update function too...
18:32:28amiconnOf course
18:32:41amiconn...and rockbox should be flashable
18:32:49t0masthat would be nice... yes
18:32:53amiconn-> less boot delay
18:33:01t0masbut I think thats something for linus again? bdm needed?
18:33:34CassandraDoesn't really need to boot much quicker, to be honest.
18:33:40amiconnAt least the first stages...
18:34:17t0mashow big is the iriver fw?
18:34:33t0masand how big is the flash rom?
18:36:43Rorihmmm. bookmarking works great. I wish there was a way for it to distinguish between an audio book and a music file though
18:37:03Rorifrom a tag for instance
18:37:40RoriLike. Have an option to only bookmark files with an audiobook genre tag
18:38:13RoriI always have ideas for Rockbox lol
18:38:27Rorishame I can't code shit
18:38:43RoriI'm an idea's man! ;)
18:44:36 Join _aLF [0] (
19:03:19 Quit linuxstb ("Leaving")
19:03:47Rorihas anyone figured out why sometimes I get 'codec failure' during playback and it drops back to the file list? it's only occasionally
19:05:09 Nick Lynx_ is now known as Lynx_awy (
19:07:18MoosRori: look if you don't have any songs unsuported (i think at WMA), hapened here with WMA the codec it's not currently implemented in lib
19:07:57RoriI don't understand that statement
19:08:18Rorithis is playing back MP3
19:08:36Mooshave you only MP3
19:08:47Rorino Ogg too
19:09:38Rorilightning...big thunder clap
19:09:42Roriscary :)
19:10:18Moosdon't remenber, playback it don't curently bug free ;)
19:10:38MoosSlasheri work on it
19:10:47 Quit Febs ("Chatzilla [Firefox 1.0.4/20050511]")
19:11:09Rorino worries
19:11:11Moosbut here the only files with this panic are for my rare WMA files
19:11:51 Quit kenshin (Read error: 60 (Operation timed out))
19:13:46Plugh_anyone know of a text to mp3 conversion util that's free?
19:14:00Plugh_something that uses the MS voice would be fine
19:14:44RoriI know of one but not it's name
19:17:04Plugh_I have some books in txt format and would prefer to listen to them than read them
19:17:14amiconnThat will become a monster commit for sure :-/
19:17:16Plugh_and I don't mind computer voice
19:17:27amiconnExpect some graphics glitches...
19:18:40Rorilook there
19:20:32*amiconn is currently contemplating whether it would be easier to *not* have a global draw mode, but make the draw mode an additional parameter...
19:22:53RoriAT&T Natural Voices is one of the best text to speech engines around. Uses real voices and splits the vowels and consonants up into little slices then uses them to reconstruct speech that sounds more realistic
19:24:01Plugh_looking, Rori, but it's 20 bucks
19:24:02 Quit Coldtoast (Read error: 104 (Connection reset by peer))
19:24:28Roriyou could always try P2P ;)
19:24:35Roricheapskate :P
19:24:56Plugh_no, just P-whipped
19:25:06Plugh_my wife won't spring for anything like that
19:26:46Roriwell get it from P2P and then pay later if you like it
19:29:18Plugh_trying it
19:36:44 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:38:08Thasp[13:14:42] <Plugh_> anyone know of a text to mp3 conversion util that's free? <−−- ? o_O
19:38:24Thaspthey have those nowadays? :o
19:39:42RoriI was listening to the sample MP3's and AT&T Ray sounds the best. Does not seem to have those synthetic glitches between the phonemes
19:40:09Roribut it's a short sample so hard to tell. Still a bit too pausey for my liking but very understandable
19:40:49RoriA real person reading changes the tone of their voice depending on the meaning of the phrase
19:40:57zesomebody should do a physical modelling speech synthesizer
19:41:27RoriPeople read something adding their own 'flair' to give a phrases, sentence or paragraph meaning
19:42:07Plugh_Thasp: the voice menu tagger for directories is a free text to speech converter
19:42:09Rorialthough these text to speech engines are getting better and better they still lack that quality
19:42:28Plugh_I could probably hack it up for text files
19:42:30 Join godzirra [0] (
19:43:00godzirraI see that you guys have ihp-120 support and I'm looking at changing to your firmware, but I'm on a linux box and had a question or two. ;)
19:43:27Plugh_don't ask permission to ask a question
19:43:29HClthats not really a but, since windows is more unpractical with rockbox (when developing)
19:43:30Plugh_fire away
19:43:54***Saving seen data "./dancer.seen"
19:43:59godzirraI wasn't asking permission. I was stating something. ;)
19:44:07HClwhat're the questions?
19:44:25godzirraI was curious where I get the tools to update the firmware the "hard way"... specifically descramble, mkboot, etc.
19:44:32HClthey're in cvs
19:44:40godzirra(Sorry.. a bit laggy... compiling a ton of stuff.)
19:44:43HClin the tools sub directory
19:44:56hicksgodzirra, The fwpatcher works with wine.
19:44:57hicksThats the way I did it on linux.
19:44:57godzirraAh.. I havent checked them out from CVS yet.
19:45:05godzirrahicks: I'm installing wine now to see if it works for me.
19:45:19*HCl yawns
19:45:29godzirrathats whats compiling.
19:45:36hicksah lol :)
19:46:46HClits less trouble to just check out cvs.. but okay
19:47:04HClcold showers rule
19:47:51godzirraI'm reading the cvs docs now
19:48:28godzirracrap. havent installed cvs on this box yet.
19:52:03 Join hardeep [0] (
19:52:44 Join hubbel [0] (
19:54:25 Quit hubbel (Client Quit)
19:54:33*amiconn is still undecided about the drawmode implementation
19:54:40 Quit Strath ("Client closed")
19:55:41amiconnThere are many places that require a different drawmode everytime...
19:56:56amiconn...but otoh I remember that implementing a global drawmode reduced code size when I did the grayscale lib...
19:58:15amiconn...and if I go for an additional parameter, the question arises how to handle brightness / colour as soon as I start implementing 2-bit gfx...
19:58:30amiconnMore parameters???
19:59:23 Join Stryke` [0] (
20:00:13godzirraWhen I configure the tools, do I want normal or bootloader?
20:00:19godzirrafor my first install onto an IHP-120
20:12:12godzirraAll the sudden wine seems much easier when I Realize I have to compile the cross compiler still. :)
20:12:35 Join Febs [0] (
20:17:19 Join MO-Pantsu [0] (
20:19:18 Join Coldtoast [0] (
20:24:27 Join Redbreva [0] (
20:25:16 Quit Febs (Read error: 104 (Connection reset by peer))
20:29:56ColdtoastI found a situation where backlight fading isn't smooth
20:30:19Coldtoastnothing very important tho
20:30:22 Join kenshin [0] (
20:30:31Coldtoastjust when you run a plugin like Cube
20:31:15 Join _Redbreva [0] (
20:31:19ColdtoastI imagine you'd never get it smooth in that situation anyway
20:32:38amiconnPerfectly smooth here
20:33:03Coldtoastreally? I ran it today and it wasn't
20:33:15amiconnThere will be some plugins that will disable backlight fading while they're running, but these aren't implemented for iriver yet
20:33:32amiconn(metronome, video etc)
20:34:14 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
20:34:19 Join Stryke` [0] (
20:34:55Coldtoastah. sorry. it was Logo
20:35:29 Quit Rori (Read error: 110 (Connection timed out))
20:35:50amiconnHmm, you're right
20:36:00amiconnI wonder why that is - shouldn't happen
20:36:10amiconnThe backlight fading is interrupt driven
20:36:25Coldtoastthat's interesting
20:36:49amiconnSeems you found a bug
20:37:03 Quit Redbreva ("CGI:IRC (EOF)")
20:37:13SlasheriI think the plugin does not keep cpu_boost in sync
20:37:28HClhey Slasheri
20:37:35Slasheriso it calls cpu_boost(false) more often than cpu_boost(true)
20:37:36HClany progress on that end of song event thing?
20:37:37Slasherihi HCl :)
20:37:39amiconnThe plugin doesn't touch cpu_boost
20:37:49amiconn...and it also happens with snow.rock
20:38:03SlasheriHCl: do you really need information about previous song or is current/next song information enough?
20:38:05godzirraI keep watching what you guys are saying and I understand every third or fourth line. ;) Usually the ones where you're saying hi to someone else.
20:38:32HClSlasheri: in theory, next is enough, but then i also need an event on the start of a song
20:38:39HClat least
20:38:41HClif i can assume
20:38:44HClthat the next song
20:38:47HClalso means that the last has finished
20:38:57SlasheriHCl: yes, you will get that event always with the information about next song :)
20:39:09HClalso if there is no next song
20:39:15HClsay it reaches end of playlist
20:39:18HClwill i still get the event?
20:39:24Slasherithen you will get an event with empty value
20:39:34HClokay, then that should be fine
20:39:39Slasherigood :)
20:39:39HClwill i get one before playing any song?
20:40:02HClif the first song gets played
20:40:03SlasheriHmm, no? You get it when first song starts playing
20:40:06HClwill i get one before it plays?
20:40:40Slasheriyou get the event at exact time when real audio output begins
20:40:48HClshould be fine :)
20:40:52Slasherigreat :)
20:41:52ColdtoastL ike that you re-added the playl;ist options as well as leaving them in the Playlist menu, btw
20:42:24SlasheriHCl: I can't patch the code right now because i have lots of changes and the code doesn't work at all but you could do the change: in playback.c uncomment queue_post(&audio_queue, AUDIO_TRACK_CHANGED, 0); Then in case AUDIO_TRACK_CHANGED: add your function call there
20:43:02HCli'll try
20:44:00Slasherii think you can pass the id3 structure to your function
20:44:11Slasheriyou can read the pathname from that struct
20:44:18HCli'll look at it in a bit
20:44:20HCl :)
20:50:03 Quit Sucka ("a bird in the bush is worth two in your house")
20:51:45 Join RedBreva [0] (
20:51:57*amiconn strongly hopes that Linus' commit doesn't clash with my changes :-/
20:56:21Mooswe hope too
20:56:41Mooshave you solved your complicated puzzle?
20:57:11amiconnYes, at least I hope so
20:57:46amiconnI'll keep the drawmode concept, and set some conventions
20:57:57amiconnSee my latest wiki update
20:58:29 Quit _Redbreva (Read error: 145 (Connection timed out))
21:01:05Moosi congrats you, apear for me very complicated
21:01:24amiconnIt's tedious
21:01:25Moosi'm not very competent for this things
21:01:42amiconnThat will become a monster commit - so many places to change
21:02:03godzirraWow... the rockbox on my ihp120 looks ... weird.
21:02:12godzirraIts got an odd line underneat the progress of a song, for instance.
21:02:22Moosamiconn: it was laborious?
21:02:36godzirraIs that normal?
21:02:52amiconngodzirra: That's the not-yet-functional peakmeter
21:03:12godzirraAhh, gotcha.
21:03:23t0masgodzirra: take a look at the CustomWps page in the wiki
21:03:54Mooshi t0mas
21:03:57t0mas(or get one here:
21:04:00t0mashi Moos
21:04:09godzirraThanks. Reading.
21:04:11amiconnHaha: #> makemake
21:04:16amiconn-> command not found
21:05:52godzirraIS the 'default' supposed to only use half the screen?
21:07:49Stryke`with the default font, you can fit 26 lines on the iRiver, i don't think anyone has a 26 line wps
21:07:59godzirraYeah.. that seems a bit extreme.
21:08:10godzirraI'll play with a few.. just trying to get it to look.... normal. :)
21:08:21Stryke`the font i use Chicago 12, you get 9 lines
21:08:31godzirraAnd I guess my A-B button doesnt change the playmode anymore.
21:08:41t0masit doesn't
21:09:08amiconnStryke`: The default font allows 15 lines (16 without status bar)
21:09:13godzirraWas that a statement or a question?
21:09:49amiconnFor 26 lines you'll need a 5-pixel font
21:10:08CassandraI'd like to see an example of a WPS using images.
21:10:44Stryke`my mistake
21:10:50ColdtoastI have on Cassandra
21:10:55Coldtoastsimpl etho
21:12:29CassandraLooks good though.
21:12:55godzirrat0mas: Was that a question or a statement?
21:13:04Coldtoastthx. it's nice and clearn
21:13:25godzirraColdtoast: I love that.. its nice looking.
21:13:45ColdtoastI went back to the iriver firmware and it's SO hard to look at now
21:13:54godzirraColdtoast: I can imagine. Where can I get your WPS? :)
21:13:58Moosespecially with the soon commiting of amiconn ;-)
21:14:04Coldtoasteverything's SO massive
21:14:24CassandraAlways said amicon should be commited. :)
21:14:42ColdtoasthmmI think I might zip it up and upload it
21:14:49godzirraColdtoast: Would you please? I'd like to try it out.
21:14:56amiconnMoos: My commit is supposed to not change the appearance, but there might be bugs
21:15:17hardeepdon't you mean undocumented features?
21:15:17amiconn...and it is a big step forward towards more flexible graphics
21:15:50Moosa ok, and about greyscale?
21:16:02Moosan oyher step?
21:16:06amiconnThat is the next step when the gfx api is done
21:16:32 Quit RedBreva (Read error: 110 (Connection timed out))
21:16:36amiconnI'm testing concepts with that api as well
21:16:50t0mas<godzirra> t0mas: Was that a question or a statement? <−− statement
21:16:54t0masbut I don't like that ;)
21:17:10Coldtoastwith the .wps, it'd be cool if you had avvess to the icons rockbox has inbuilt
21:17:15t0masI only haven't dared to toch it because of Cassandra and amiconn being busy with buttons things
21:17:29Coldtoastso you could have a little icon of a musical note in front of a trackname
21:17:33t0masColdtoast: acces to the icons isn't hard...
21:17:46t0massounds like a good idea...
21:17:57t0masjust add an image tag for build in images
21:18:16t0mas%xf image from file
21:18:22t0mas%xb buildin image
21:18:28t0mas%xd display the image
21:18:37t0mas(because we wanted to separate that)
21:18:52west-acre Rockbox is an Open Source replacement firmware for the Archos Jukebox 5000, 6000, Studio, Recorder, FM Recorder, Recorder V2 and Ondio MP3 players.
21:18:52west-acreRockbox is a complete rewrite and uses no fragments of the original firmware. It boasts numerous features while remaining easy to use.
21:18:52west-acreInstalling Rockbox does not harm your Archos, it works the same way you upgrade your ordinary Archos firmware.
21:19:10west-acreno iriver ?
21:19:28HCliriver is still testing.
21:19:35t0masthe site is way behind development
21:19:45t0maswe need some people to update it when we're about to release for iriver
21:19:46godzirrat0mas: Ack. its set to infinitely loop my directory I'm in... thats ... bad.
21:19:47Moospre alpha version
21:19:56godzirrat0mas: is there no current way to set your playback mode then?
21:19:58t0masgodzirra: you can change it in the settings
21:20:17t0massettings -> general -> playback or settings -> playback
21:20:20t0masI'm not sure
21:20:49t0massettings -> general settings -> playback -> resume
21:21:32godzirraah.. ok.
21:21:46godzirrathis may be a silly question... but pressing my nav button doesnt get me to settings anymore...
21:22:01godzirranor do I know how to go up a directory.
21:22:49godzirrathat makes sense.
21:22:58godzirraSorry.. not used to my iriver making sense.
21:24:23Coldtoastgodzirra: put all the files in that zip in .rockbox. If you want to be neater and put images in a subdir, you'll have to edit the .wps yourself.
21:24:37godzirraHrm... it says if you put your .wps file in .rockbox, you have to "play" the wps to load it?
21:24:52godzirrabut when I try to go into the .rockbox directory, its telling me "Play resume, yes or no?
21:26:04godzirraColdtoast: If I make the images a directory, do I need to use \ or / 's?
21:26:14godzirramv the images into a directory I mean
21:26:23CassandraYeah. You need to use the SELECT button (centre of the joystick)
21:26:51godzirraCassandra: thats what I was using
21:27:09godzirraColdtoast: Thanks
21:28:14godzirraI hit "select", "left", "up" (until I get to the 'X rockbox' directory) and hit "select" again and it says this:
21:28:17godzirraRolo Loading
21:28:26west-acrehey hoo designed the rockbox logo?
21:28:30godzirrathen: "Resume? PLAY = Yes Any Other = No
21:28:45Coldtoastyou're selecting rockbox.iriver. not .rockbox
21:29:01godzirrahrm. I don't see .rockbox then? :/
21:30:33Coldtoastdo you have General Settings>File View>Show Files set to All?
21:30:53godzirranot if it didnt come that way by default. Lemme try that.
21:31:03Coldtoastok. it's set to Supported then
21:31:06godzirraon a side note, after flashing, do I need ihp_120.hex on my irivers hard drive?
21:33:52godzirranow I'm just trying to find a "here's what the keys do now" in the documentation
21:34:34godzirraahh found it
21:35:30amiconnI think I should commit what I have as soon as it compiles cleanly, but I may not be around to fix quirks during the weekend
21:36:01amiconnIf I don't commit, chances are that I'll get a whole bunch of clashes next week :-/
21:36:34Mooscommit ;)
21:36:47 Quit ghostiger2 (Remote closed the connection)
21:37:30amiconnIt will take a while to get it to compile cleanly
21:40:34godzirraok. how the heck do I get into my settings menu :/
21:42:35godzirraDarnit.. the pdf settings don't have info on what each key does for the iriver
21:43:58***Saving seen data "./dancer.seen"
21:44:49Cassandrathe MODE (A-B) button.
21:47:16godzirraCassandra: thank you
21:47:20godzirraColdtoast: your wps looks awesome.
21:47:35godzirraok.. one last question.. I appreciate all your help, and rockbox seriously kicks some ass.
21:47:49godzirraIs it normal to take about 2 seconds between when you hit pause and when it actually pauses?
21:47:50 Join ghostiger2 [0] (debian-tor@5df48ea5ace8dedf.session.tor)
21:49:34CassandraIt fades out.
21:49:59godzirraok.. thats a cool feature, was just curious if that was by design or not.
21:50:13godzirraIs there anything that a non-Cish type programmer can do to contribute to the project?
21:50:26godzirraI'm a web programmer and do perl stuff and database stuff a lot, but C was never my strong point.
21:51:07 Quit west-acre (Read error: 60 (Operation timed out))
21:52:12CassandraI don't know of anything specific.
21:52:28Coldtoastthe little "rockbox" in the bottom right looks cute, huh? :)
21:52:40godzirrahrm.. my littel rockbox isnt showing up :/
21:53:22godzirrathat was the teeny_logo.bmp right?
21:53:36Coldtoastoh? hmmmm. dunno why that'd be. I am using the rockbox_default.fnt tho
21:55:14godzirranot sure why its not showing up
21:55:36godzirraThis is the right first line righT?
21:55:44godzirraAssuming I copied all the images into edan_img/
21:56:22Coldtoastthis is odd. I rearranged mine so I have all teh images in .rockbox/wps/edan
21:56:54godzirrathat line looks right doesnt it?
21:57:07Coldtoastand yeah. using %x1|wps/edan/teeny_logo.bmp|116|117| doesn't load it
21:57:25godzirraall the other images load though.
21:57:29Coldtoasthey godzirra: see the "Current Track" image?
21:57:36Coldtoastwith the tiny dots after it?
21:57:40godzirraI see the other ones all just fine.
21:57:48godzirraartist/album and next track too
21:57:48Coldtoastok. are you missing any dots in that one?
21:57:55Coldtoastme too
21:58:21Coldtoastbut if you put all the images in .rockbox and use no ptahs in teh wps to load them, it all works
22:00:36godzirrahrm. how weird.
22:00:40godzirrathat only that one image doesnt load though
22:00:44godzirrathe other ones load fine in the paths.
22:00:58godzirraColdtoast: Perhaps try adding another image (any image) before the teenylogo... maybe it just isnt loadingf the first image for some reason?
22:01:21Coldtoastthere's something else odd happening tho is how it looks with all files in .rockbox is how it looks with the images in a subdir
22:02:45 Quit RotAtoR ()
22:02:48 Quit ghostiger2 (Remote closed the connection)
22:03:06Coldtoastthe little logo isn't there and there's a bit missing from the "Current Track" image
22:03:10godzirraThats weird
22:03:12godzirraI have no idea.
22:04:16Coldtoastseems liek something that'l be fixed in an upcoming build. Not something wrogn with my .wps
22:04:46amiconnIt seems I have to touch every friggin plugin
22:05:30CassandraHow annoying.
22:05:55godzirrahow do you tell what plugins at.. or better, where can I read about plugins?
22:07:26CassandraThe manual's a good place to start. Or the wiki.
22:08:09godzirrathanks.. I'm off to read both. Thanks for all the info. :)
22:08:20Coldtoastgodzirra: my wps works correctly as I sent it tho
22:08:26Coldtoastjust dump it all in .rockbox
22:08:45ColdtoastI can deal with that. hehe. Interesting to find a little niggly tho
22:12:47godzirraAll good. I still love the wps.
22:14:15Coldtoastcool, make one ;)\
22:14:28godzirraI will when I have more time to read the docs. ;)
22:15:18godzirraI moved everything into .rockbox/wps
22:15:24godzirraincluding edan.wps and all the images
22:15:28godzirraand still get the same results.. almost
22:15:40godzirrathe teeny_logo pops up for a split second on the top RIGHT corner and then disappears.
22:16:07Coldtoastyeah. the status bar probably destroys it
22:16:17godzirrain your image though it was in the bottom left corner?
22:17:11Coldtoastsupposed to load it at 116, 117
22:17:45godzirrathats where its loading it
22:17:51godzirrabut 116,117 is the bottom right corner isnt it?
22:18:36godzirrathere we go
22:18:39godzirrait works now :)
22:18:45godzirraI'm going home guys.. I'll chat with everyone later. :)
22:19:03Coldtoastthe .wps has to be in .rockbox for it to load next boot tho
22:19:08godzirraI saw
22:19:10godzirraI changed mine
22:21:09godzirraCassandra: Let me know if there's anything I can do to help the project. Like I said, I'm not much of a C programmer, but I'll do anything I can to help you.
22:21:59CassandraBest ask someone like LinusN.
22:22:07CassandraI just write the manual.
22:22:09godzirraokee dokee.
22:23:38Cassandra(and the occasional hack)
22:23:55*amiconn is having a hard time with the clock plugin
22:24:50CassandraWhat's up?
22:25:17amiconnDozens, erm, hundreds of calls to lcd_clearrect(), lcd_invertrect() ...
22:25:56CassandraGlobal search/replace?
22:26:07amiconnUnfortunately not
22:26:13Coldtoastnoticed something else. Screendump ignores inverted :CD
22:26:24amiconnI have to place calls to lcd_set_drawmode() before & after
22:26:42amiconnColdtoast: Inverted lcd is a hardware feature
22:27:13Coldtoastthere's something odd with mine
22:27:33Coldtoastwhen I switch to inverted, I have vertical stripes thru it
22:27:55Coldtoastnot evenly spaced. Looks a bit like a bar code
22:28:54Coldtoastbut one of the builds from teh past two days didn't do it
22:30:39CassandraWorks fine for me.
22:31:47Coldtoastyeah. somebody else tried and it was fine for them too
22:32:05Coldtoastit'll be my player I guess
22:33:31Coldtoastif I adjust the contrast so the screen is black there are no stripes tho
22:50:57CassandraDamn this stupid code. I *knew* there was a reason I changed it. It doesn't actually work properly. The first resume press is *always* ignored, which on iRiver we do not want.
22:58:10 Join RotAtoR [0] (
22:59:58amiconnMrf, I do not get this :-/
23:00:32amiconnWhy the **** is a #define unknown in a source file although the corresponding header *is* included?
23:00:58Bagderrun the same gcc line manually with -E
23:01:27Bagderusually gives some clues
23:01:34amiconnHmm. Seems it is not included because it isn't found, but the compiler doesn't complain???
23:05:54amiconnHow do I set verbose mode so I can see the gcc line?
23:06:01Bagdermake V=1
23:06:57 Quit kenshin (Read error: 110 (Connection timed out))
23:11:17hardeephmmm, the ondio fm is pretty cool
23:12:17amiconnBagder: Hmm, called the line with added -E. Nothing happens?!
23:13:01amiconnAh no
23:13:12amiconnIt wrote the text output to the .o file....
23:13:21Bagderright, if you kept -o
23:19:27amiconnI found the reason
23:19:47amiconnThere are 2 files called lcd.h, one in rockbox and one in rockboy
23:20:06amiconnNow what to do about that???
23:20:15amiconnI need my drawmode #defines...
23:20:17Bagderrename the rockboy one
23:22:47hardeepany complaints if I add a feature to reshuffle when the playlist repeats? i think it would be great for the ondio
23:23:20Bagderhardeep: there's a feature request for that since... eons back ;-)
23:23:38Bagderiow: would be great
23:23:46hardeepi had written something a while back... just found it again heh
23:24:23Bagderyou an ondio user now?
23:24:54hardeepnot really, just saw it on ebay and thought why not
23:27:19amiconnI'm getting weird compilation problems now :-(
23:28:17amiconnhw.c: In function `hw_dma':
23:28:18amiconnhw.c:55: error: `lcd' undeclared (first use in this function)
23:28:40amiconnlcd is declared in the (renamed) lcd.h, which is properly included in hw.c ...
23:28:51amiconnThis rockboy stuff is all messy :(
23:30:03 Quit Mx|afk ("leaving")
23:30:43 Join LinusN [0] (
23:31:56amiconnMeh, I think I'll fix that the quick way
23:32:16hardeeprm -rf ?
23:32:22amiconnI want to get this stuff up and running, delaying the commit for too long won't do any good
23:38:18MoosHola Linus
23:38:25Moosque tal?
23:38:53LinusNfine, thx
23:39:17LinusNor is it "mui bien" or something?
23:39:43Moossi muy bien, yo tambien :-)
23:40:16Moosça va les vacances?
23:40:27LinusNtrès bien
23:40:38LinusNbeacoup de soleil
23:40:59t0masyeah, I hope it stays this way for a few more weeks :)
23:41:14LinusNgoing home tomorrow
23:41:16amiconnIch versteh' nur Bahnhof
23:41:22Moosyes t0mas here in france is too hot
23:41:35t0maslol @ amiconn
23:41:39amiconnYay! It compiled without warnings for recorder :-) Now checking other archos platforms, testing on recorder in parallel
23:41:48t0masIch spreche ein sehr klein bischen Deutsch...
23:41:53Moosalélouia :)
23:42:18 Part chuch
23:42:19t0masbut... I've been thinking about this delayed playing...
23:42:26t0massome people complained about
23:42:45t0massomeone removed the call to audio_stop before the next function was called in wps.c
23:43:06amiconnLinusN: The recorder binary got ~600 bytes larger due to my gfx changes. Is that acceptable?
23:43:19amiconnbitmap drawing is not yet converted
23:44:00***Saving seen data "./dancer.seen"
23:44:16LinusNamiconn: of course
23:44:44amiconnI hope that it will get a bit smaller again once I applied all my ideas
23:45:45LinusNis it "paint-the-build-table-red" commit time then? :-)
23:46:04amiconnIt shouldn't
23:46:05Bagderred is a nice color ;-)
23:46:43t0mas3 knowledgable devs around... has any of you looked at the next song delay on iriver?
23:47:00LinusNi have second thoughts about the playlist menu in the browser context menu
23:47:48LinusNt0mas: i guess it takes a few secs for the decoder to reach the pcm low watermark
23:48:16t0mashm.. but it was fast before
23:48:17BagderI'm not that into the playback code to be able to tell
23:48:31t0masmaybe it has something todo with the crossfade code
23:48:41t0masbut I was unable to find the cause...
23:49:27amiconnBoooog... :-/
23:49:57amiconnLinusN: I hope your changes don't clash with my gfx api conversion
23:50:11LinusNi doubt it
23:50:24amiconnI had to touch many many source files
23:50:40LinusN"cvs update"
23:53:08 Quit Heidelbaer (Read error: 104 (Connection reset by peer))
23:53:33t0masLinusN? playback.c line 1.010
23:54:02t0massomeone removed that stop...
23:54:38t0mascommented out in v1.55
23:54:45t0mas"Some re-desing to buffer filling and track loading code and a few bugfixes."
23:55:01Bagderstopping it doesn't sound right to me
23:55:03hardeepLinusN: do you mean the view/save etc. options when in the file browser?
23:55:10hardeepLinusN: re: second thoughts
23:55:23LinusNthe "create" option
23:55:35hardeepLinusN: 'cause I don't like those either... i use insert, etc. more then the others when browsing files
23:55:37Bagdert0mas: this is Slasheri's territory
23:56:04LinusNin my view, the context is "the file/dir marked by the cursor"
23:56:17Bagderand now, sleeep
23:56:22hardeepyeah, that makes more sense
23:56:25t0mashm... nice idea
23:56:26LinusNbut "create playlist" uses the current dir
23:56:27t0masbut it's to hot to sleep
23:56:34amiconn100 KB diff, almost 50 files touched
23:56:35t0masSlasheri? are you around?
23:56:46hardeepand view/save don't even refer to a file
23:56:54t0masthat audio_stop() call doesn't break crossfading... so why is it gone? :)
23:57:01LinusNhardeep: true
23:57:26 Join TCK- [0] (
23:57:42LinusNhardeep: the only playlist option that should be in the browser context menu should be "insert"
23:57:58LinusNand friends
23:58:06hardeepLinusN: i agree
23:58:18Slasherit0mas: yes, that is required for cross fading. Some buffer flush could be implemented to make song change faster when crossfader is not available
23:58:53t0masSlasheri: what about checking if crossfading is on, and if it's off: audio_stop()
23:59:01SlasheriThat prevents also having gaps between tracks when selecting a new track to play

Previous day | Next day