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 2005-10-13

00:01:25 Quit Bger (Read error: 110 (Connection timed out))
00:02:03*amiconn wonders how that works
00:02:30amiconnIs that fractional mode, or what?
00:02:48preglowamiconn: oh yes
00:02:52|Lupin|Pls: What is the "new:" attribute in .lang files used for ?
00:03:55preglowamiconn: you'd need frac mode for that, else it's impossible
00:04:29amiconnHmm... that looks like it would be even faster 64 bit mul... Any bit loss here?
00:04:55amiconn(I noticed that the shift counts aren't adding up to 32)
00:07:41 Join tvelocity [0] (
00:08:57 Quit ender` (Read error: 113 (No route to host))
00:10:34amiconnpreglow: Iiuc MULT31_SHIFT15() might deliver slightly off results
00:11:01preglowamiconn: how?
00:11:11amiconnIt looks like fractional mode doesn't give bit 63..32, but bit 62..31
00:11:16preglowamiconn: it'll work for signed 31 bit fixed point numbers, like i said, the top bit is lost :/
00:11:22amiconn(otherwise you would have to shift left by 17)
00:11:46amiconnAh, no
00:11:48 Join guillaumh [0] (
00:12:03amiconnI thought bit 31 would cause problems cause it overlaps
00:12:14amiconn...but it should be the same in both parts
00:12:23*amiconn will do some test
00:12:33amiconnDoes mpc.codec set the emac mode?
00:12:33preglowamiconn: but yeah, basically we'll just have to hope it never overflows
00:12:37preglowamiconn: it does
00:12:39preglowamiconn: frac mode
00:12:54preglowamiconn: saturate and round, as a matter of fact, and we'll need to disable the rounding mode for this
00:13:13amiconnIs rounding needed elsewhere?
00:13:18preglowwhich is no biggie, i already do that for tremor, because of MULT31_SHIFT15
00:13:42preglowit just rounds of the lsb when fetching the 32 bit answer from the accumulator
00:14:10 Quit matsl (Remote closed the connection)
00:17:27 Join matsl [0] (
00:17:39BagderI hate being stupid
00:20:27*amiconn too
00:20:37amiconnI mean myelf
00:20:44amiconn+s somewhere
00:20:48ashridahnice save
00:20:50 Quit Moos (Read error: 104 (Connection reset by peer))
00:22:58 Quit matsl (Remote closed the connection)
00:23:29 Quit ashridah ("Leaving")
00:24:03preglowamiconn: but like i said, using emac in frac mode should work great, as long as we don't get numbers large enough to overflow the top bit of the 64 bit mul
00:24:14preglowand that should never happen
00:24:15 Quit vmx ("Have fun!")
00:27:43 Join matsl [0] (
00:29:30 Part guillaumh ("bye")
00:43:37|Lupin|Hmm no body knows the meaning of "new:" lines in the .lang files, pls ?
00:43:51Bagderthey are for the new string
00:43:54Bagderthe translation
00:44:26|Lupin|ah !
00:44:58|Lupin|So each .lang file contains both the english string and the corresponding string in the foreign language ?
00:45:10|Lupin|ah ok
00:51:51amiconnpreglow: It works.... :))
00:52:30|Lupin|French TTS is ugly at the moment :)
00:58:16amiconnpreglow: Same old link :)
01:02:32|Lupin|bye everybody.
01:02:38 Quit |Lupin| ("leaving")
01:05:30amiconnpreglow: Another slight improvement. Now I even got a short unboost while playing iriver-xtreme.mpc
01:09:11preglowamiconn: hah, that's nice and short
01:09:19preglowamiconn: what's the boost factor now? ;)
01:09:25preglowsomewhat occupied with other coding atm, can't test
01:13:01amiconnAnother slight change (removing the zero shift special case) gave another slight boost
01:13:25amiconnboost factor for iriver-xtreme.mpc is now around 90%
01:14:27amiconnme commit :)
01:14:56dpassen1thanks and congratulations, amiconn
01:16:48amiconnMy first real codec work...
01:28:51***Saving seen data "./dancer.seen"
01:34:03preglowit always boosted before
01:34:06preglowthis is just remarkable
01:40:03preglowamiconn: is it realtime for all files you've tried?
01:40:30amiconnI can't easily try becuse of the playlist problem, and I only have rasher's files atm
01:41:15preglowwell, i'll update SoundCodecs saying it's barely realtime
01:42:32preglowamiconn: btw, can you use that same technique for the mandelbrot mul?
01:43:23amiconnUnfortunately not. Mandelbrot needs a full 64*64 multiply
01:43:30preglowyes, figured that
01:43:49preglowbut it sure does the trick nicely here
01:44:03preglowof course, it's still very slow compared to what it could be, but heaps better than it was
01:44:32amiconnI wonder why they claim musepack is faster than others...
01:45:25preglowbecause i bet it is, on arm...
01:45:31dpassen1and on x86
01:45:35preglowwhich has a full 64 bit accumulator
01:45:42preglowsame with x86, yes
01:46:03dpassen1encoding speed is the most impressive from my experience
01:46:06amiconnMaybe it is on fpu-equipped systems
01:46:36preglowamiconn: and, like i said, on arms, the fact that they don't use the entire accumulator width works to their advantage their
01:46:37 Quit odd (Read error: 104 (Connection reset by peer))
01:46:50preglowamiconn: for every bit in the multiplicands that are zero, they save some cycles on the mul
01:47:38preglowand they've more or less said themselves the only integer target platform libmpcdec has, is arm
01:50:39 Quit actionshrimp ("a bird in the bush is worth two in your house")
01:53:52preglowok, so only format not supporting seeking is ac3
01:53:54preglowtime to fix that
01:56:25 Quit matsl (Remote closed the connection)
01:57:03dpassen1is there any more room for significant optimizations for musepack?
01:58:09preglowdpassen1: tons, but the libmusepack people made our job very hard
01:58:16dpassen1its a shame really
01:58:27preglowi agree
01:58:31dpassen1to tether the format to specific architectures
01:58:32preglowi think musepack is a good format
01:58:54dpassen1quality at the standard profile is perfect for me
02:00:05preglowit's not directly tethered, but that was the target platform they had when they wrote it
02:00:09preglowit just so happens that it doesn't port well to ours
02:00:33preglowthat, and their use of a gigantic decoder state struct, which doesn't exactly make it easy for me to use iram
02:04:41preglowbtw, did anyone find out why builtin_alloca doesn't work on sims?
02:17:24 Quit JoeBorn ("")
02:25:57 Join Bger [0] (n=Bager@
02:46:20preglowamiconn: i've moved all codec stuff to codeclib.h now, btw
02:46:35preglowwhy is it that only the malloc functions needed to be protected with the codec_ prefix again?
02:51:07 Join hardeep [0] (i=hardeeps@norge.freeshell.ORG)
02:54:30amiconnThis is to prevent the sims (esp. cygwin) linking with the native functions - this will make it crash for sure
02:55:12amiconnBtw, I think it's in fact possible to use the emac for 64bit in mandelbrot.c,
02:55:31amiconnbut it requires unsigned fractional mode, and fiddling with accext
02:55:39preglowthat isn't available
02:55:41preglowthere's only signed
02:55:50amiconnmeh :(
02:56:20preglowbut yeah, only the malloc functions need to be protected?
03:01:12preglowbut yeah, i merged codec.h, codeclib.[ch] and xxx2wav.[ch] into just codeclib.[ch], now all codecs just #include "codeclib.h" and whatever specific headers they need, should i commit as is?
03:09:12*preglow cuts down on his "but yeah" quota
03:17:59 Quit dpassen1 ()
03:18:46 Quit hardeep ("BitchX: now with 38 percent fewer calories!")
03:28:52***Saving seen data "./dancer.seen"
04:05:03 Join Vlad0man [0] (
04:05:35 Join QT_ [0] (i=as@madwifi/users/area51)
04:18:40 Quit QT (Read error: 110 (Connection timed out))
04:22:31 Quit Vladoman (Read error: 110 (Connection timed out))
04:43:52 Join Byron [0] (i=byron@
04:44:29ByronI have an ondio and I have Rockbox on it, is there any way to make the file names of the recordings different than rec0001 ect ect ect?
04:44:43ByronI'd like dates, or times, or something to prevent overwritting
04:47:51 Join Paul_The_Nerd [0] (
04:50:17Byronputting 2.5 on there to see if that fixes anything
04:50:38Byrondoes the ondio only work at usb 1? It's going really slow
05:28:54***Saving seen data "./dancer.seen"
05:38:36 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
06:01:58 Part Byron
06:02:16 Join Maxime [0] (
06:05:07 Join Paul_The_Nerd [0] (
06:09:30 Quit RotAtoR ()
06:35:30 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
06:39:31 Quit tvelocity (Read error: 110 (Connection timed out))
06:55:24 Join amiconn_ [0] (
07:13:02 Quit amiconn (Read error: 110 (Connection timed out))
07:13:03 Nick amiconn_ is now known as amiconn (
07:20:15 Quit Hadaka (
07:21:12NJoinHadaka [0] (
07:21:22 Join thegeek_ [0] (
07:21:45 Quit thegeek (Read error: 104 (Connection reset by peer))
07:22:30 Join pilophae_ [0] (
07:22:31 Quit pilophae (Read error: 104 (Connection reset by peer))
07:28:58***Saving seen data "./dancer.seen"
08:00:43 Join DangerousDan [0] (
08:02:42 Quit ryan_j_ ("leaving")
08:02:58 Join B4gder [0] (
08:05:07 Join ender` [0] (i=ychat@
08:05:34 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
08:23:23 Join phaedrus961 [0] (
08:44:58 Nick Vlad0man is now known as Vladoman (
08:46:37Bgermorning ;)
08:50:38 Join Zagor [0] (
09:20:38markunMorning Bger
09:29:01***Saving seen data "./dancer.seen"
09:31:06 Quit phaedrus961 ("Leaving")
09:35:58 Join phaedrus961 [0] (
09:45:24 Nick solexx_ is now known as solexx (
10:05:42 Join ]Maggie[_ [0] (
10:07:10 Quit ]Maggie[ (Read error: 104 (Connection reset by peer))
10:09:29 Quit Bger ("brb")
11:04:19 Quit linuxstb ("Leaving")
11:13:03 Quit thegeek_ (Read error: 104 (Connection reset by peer))
11:13:08 Join thegeek [0] (
11:14:22*HCl hugs his new firefox extension
11:23:22markunWhich extension is that?
11:26:55*Zagor just learned some new css: "overflow: hidden"
11:27:27Zagoryeah, it controls what to do with content that doesn't fit on the screen
11:28:08Zagorin my case, graphics in the header. setting it to "hidden" means that it doesn't cause scrollbars to appear when you shrink the window. effectively saying: "unimportant stuff"
11:28:24B4gderthat's a neat thing indeed
11:28:42HClmarkun: TBE
11:29:02***Saving seen data "./dancer.seen"
11:31:05Zagorsight. naturally MSIE doesn't support it, so 90% of visitors don't see it...
11:31:18B4gderwhat a surprise!
11:33:07 Join Febs [0] (
11:41:41 Join ashridah [0] (
11:45:28 Join seablue [0] (
11:53:26FebsThis is probably an issue for Linus, but I've seen three separate reports on Mistic River of Rockbox bricking players when the battery runs dead in bootloader USB mode.
11:53:52FebsHere's one of them:
11:55:43SlasheriIf the charge light doesn't come up, there has to be something wrong with the hardware (for example battery protection circuit has permanentely broken)
11:56:15SlasheriBut yes, we really need to force a power off when the voltage drops below 3.0V
11:58:16SlasheriHmm, the musepack finally works realtime.. That is great :)
12:02:51 Join linuxstb [0] (
12:09:27amiconnSlasheri: :) But unfortunately .mpc isn't correctly working in playlists :/
12:09:47Slasherioh, i am just testing it.. :)
12:18:22preglowcan't exactly figure out why either
12:29:46linuxstbDoes playback.c rely on get_metadata() setting any particular variables? Very little is done for Musepack files in get_metadata.c()
12:42:10preglownothing is done but getting ape tags
12:42:51preglowlinuxstb: i've merged all the codec header stuff into codeclib now
12:48:49linuxstbpreglow: nice. There's only one more box to tick on the SoundCodecs page now - seeking in AC3 files...
12:50:45preglowlinuxstb: yup, i plan on fixing that today
12:50:56preglowlinuxstb: should be a quickie, all the code i need is in metadata.c already
12:51:35linuxstbDoes anyone know what's happened with the AAC work? Did fuzzie give anyone a copy of his source?
12:53:36preglowand as far as i know, no
12:53:45preglowi didn't even remember who worked on it ;)
12:54:06*linuxstb apologies to fuzzie
12:54:26linuxstbIt would be nice to get it into CVS, even if it's not working.
12:54:34preglowindeed, too
12:57:44preglowi'd love aac playback
12:58:02preglowand speex :>
13:02:10preglowhmm, speex seems to require libogg
13:07:04Seedpreglow: what speed is speex running at?
13:07:11Seedon your PC
13:08:51preglowi have no idea
13:08:54preglownever measured it there
13:11:58markunpreglow: most off the things speexdec needs from libogg are also in Tremor.
13:13:12preglowmarkun: yeah, but i believe not all were
13:13:21markunI can't find ogg_stream_init
13:14:03markunBut googling for ogg_stream_init and Tremor gives me some hits.
13:16:11preglowbut ok, i'll commit this codeclib change and hope nothing breaks
13:17:53 Quit Febs (Read error: 110 (Connection timed out))
13:25:16preglowmarkun: i think libogg should compile very cleanly on target anyway
13:29:06***Saving seen data "./dancer.seen"
13:33:06preglowthere we go
13:39:29markunI wonder when libogg2 will be ready
13:43:57 Quit B4gder ("time to say moo")
13:45:07preglowgod knows, why's that being developed?
13:45:17preglownew container format or just new implementation?
13:52:08markunI think it's just a new implementation. Tremor uses ideas from libogg2, I think that's why we can't use it with speexdec as it is.
13:52:52*preglow wonders why these sim errors didn't show up in his build...
13:54:23preglowriiiight, gcc 4
13:56:09markunJust a different API.
14:03:55preglowis request_buffer supposed to take a 32 bit or 64 bit number?
14:04:20preglowi wonder why some codecs use size_t for it
14:04:41preglowbut of course, size_t is never 64 bit on our platforms
14:07:47preglowi wonder why the hell these errors haven't occured before now
14:07:52preglowwarnings, that is
14:22:09 Join Febs [0] (
14:25:51 Quit linuxstb ("CGI:IRC (EOF)")
14:28:47 Join linuxstb [0] (
14:29:07linuxstbpreglow: Did you forget to remove xxx2wav.[ch] from CVS after your codeclib update?
14:41:45preglowlinuxstb: no, i'll do it soon
14:43:21preglowwhat the hell is happening with the builds
14:49:13preglowwhy the bloody hell didn't these warnings appear before now? the code in question always had type mismatches
14:50:39linuxstbIt must be related to the new order of the header file includes.
14:51:10linuxstbBut I agree it's strange.
14:51:19preglowi wonder if the long parameter in request_buffer shouldn't be unsigned
14:51:24preglowdoesn't make much sense for it to be signed
14:51:39preglowunless i misunderstand completely, that is
14:51:58linuxstbI don't think size_t was a bad choice of type. It's defined in firmware/sys/types.h as "unsigned long".
14:52:39linuxstbsorry, firmware/include/sys/types.h
14:53:35preglowno, more or less a good choice, but it's unsigned, when request_buffer takes a signed parameter
14:53:53linuxstbMaybe ci->advance_buffer() should be changed to use size_t
14:54:12linuxstbsize_t seems to be used a lot in Rockbox - e.g the string.h functions.
14:54:56preglowrequest_buffer needs a change too
14:55:03preglowi wonder what's up with all the (int) casts in that one
14:55:38linuxstbI'm guessing that size_t is the POSIX type for referring to sizes of chunks of memory.
14:57:28preglowit's meant for use as an indice
14:57:42preglowa size_t is guaranteed to be large enough to access all memory in a system
14:57:56preglowit's also meant for storing sizes, of course
14:59:51linuxstbSo we should be using size_t then...
15:01:25preglowyes, probably
15:04:51 Join einhirn [0] (
15:05:39XavierGrwow 9 codecs so far in Rockbox!
15:05:55XavierGrThat's indeed a multi codec jukebox.
15:06:04preglowi count only eight :>
15:06:13XavierGrwith Wav and mp4
15:06:39preglowmp4 as in alac or aac?
15:06:43preglowaac isn't supported yet
15:07:19 Join actionshrimp [0] (
15:07:29XavierGrI don't know codecs are too complicated.... :P
15:08:57XavierGrsay preglow, sometimes rockbox playback timer is wrong. (or the remaining timer is wrong) It says 10 minutes remaining but the playback timer shows 12 minutes and counting.
15:09:28XavierGrIs this normal? I have many vbr files though...
15:16:41markunI have a CBR mp3 file and when you seek to almost the end of the file (50 minutes and 9 seconds long) it keeps playing till 50:13
15:17:55XavierGrmine does that sometimes for more than 1 minute.
15:18:07XavierGrit isn't much of a problem just wondering.
15:18:20XavierGrStrnage is that backlight caption works correct.
15:18:59XavierGrIt will turn on x seconds before the song (truly) ends and then turn off x seconds after the new song has started.
15:21:50preglowdepends on the file, i guess, if the frame count is off, the calculated estimated time will also be off
15:24:32linuxstbXavierGr: Have you run your VBR files through vbrfix ?
15:29:08***Saving seen data "./dancer.seen"
15:43:13preglowok, i'm just reverting the wav.c thing and cast to long * in request_buffer call
15:44:06amiconnmp3 vbr seeking is always somewhat off, because the position handling mixes file postions and stream positions
15:45:23markunamiconn: I have a cbr mp3 file where seeking is off.
15:47:02preglowlinuxstb: do you want to fix the alac.c type warnings? :P
15:48:08preglowor else i'll just put in a couple of casts there as well
15:48:34preglowSlasheri: any thoughts on changing the advance_buffer and request_buffer size parameters from long to size_t? doesn't really make sense for them to be signed
15:48:39amiconnCasting to int is a bad bad idea...
15:48:58preglowamiconn: where?
15:49:46preglowi haven't got time to go through the codecs and make sure all the types have correct sign right now, so just opting for casting size_t* to long* when calling request_buffer, it's not any worse than the implicit cast that happened before anyway
15:49:57Seedpreglow: *ping*
15:50:04amiconnlong is ok, but int isn't for the buffer positions
15:50:08preglowSeed: *pong* ?
15:50:16preglowamiconn: agreed
15:50:21amiconn(assuming that someone will pick up the gmini port at some point)
15:50:21Seedprivate message
15:50:34preglowSeed: riight, i forgot i can't send without identing first, gimme a sec
15:51:21linuxstbI can't do anything at the moment (I'm doing my day job), but I think we should change ci->advance_buffer (and the other functions) to use size_t instead of long. We will probably then get a lot of warnings in the codecs that need fixing.
15:54:10amiconnIs this also used for negative 'advance'?
15:54:31preglowgotta go for a sec
15:54:33 Join samwichse [0] (
15:54:39linuxstbamiconn: Good point. I'm not sure.
15:56:31linuxstbLooking at codec_advance_buffer_callback() in playback.c, it looks like it is expecting a positive number, but it's not obvious.
15:58:08linuxstbBut ci->request_buffer() should definitely only be a positive number.
16:00:17 Quit einhirn (Read error: 113 (No route to host))
16:00:57Slasheripreglow: sorry, i am on the train now, vacation this weekend :) please feel free to change them :)
16:00:58 Quit linuxstb ("CGI:IRC (EOF)")
16:14:24 Quit ashridah ("Leaving")
16:21:01 Join _FireFly_ [0] (
16:27:44 Join ashridah [0] (
16:35:10 Join _coder` [0] (i=shani@Ariel.Atlantica.US)
16:35:11 Quit coder` (Read error: 104 (Connection reset by peer))
16:35:31 Part pilophae_
16:42:44preglowno, it's not really apparent
16:43:38preglowbut ok, i'll delete codec.h and xxx2wav stuff
16:48:17preglowhmm, seems there's still a lot of codec.h references
16:49:58 Join linuxstb [0] (
17:05:42 Quit Zagor ("Client exiting")
17:09:32 Quit _coder` (Remote closed the connection)
17:18:39 Join solexx_ [0] (
17:27:36 Quit Febs ("CGI:IRC (EOF)")
17:29:10***Saving seen data "./dancer.seen"
17:29:31 Quit solexx (Read error: 110 (Connection timed out))
17:34:07 Quit ashridah ("Leaving")
17:45:48 Quit Maxime (Read error: 104 (Connection reset by peer))
17:45:59 Join Maxime [0] (
17:49:57 Join dj_salo [0] (
17:50:00dj_salohello people
17:50:16dj_saloi have a question regarding the iHP 120 version of Rockbox
17:50:54dj_saloi am trying to set up the development environment on windows (cygwin distribution of source code)
17:51:12dj_saloi was able to setup Cygwin, but I am unable to add the compiler
17:51:20dj_salofor the processor in the iHP 120
17:51:21preglowahh, i have no idea about cygwin
17:51:28dj_salomk68 (if i rem correctly)
17:52:05dj_saloaha, i see
17:52:22dj_salowell i'll keep on trying then :) thanks!
17:59:23 Quit dj_salo ("CGI:IRC (EOF)")
18:14:20 Nick ]Maggie[_ is now known as ]Maggie[ (
18:17:49 Join dpassen1 [0] (
18:23:18 Join Moos [0] (
18:24:05 Join Lear [0] (
18:49:00 Quit _FireFly_ ("Leaving")
19:05:21 Nick Lynx_ is now known as Lynx_awy (
19:15:32 Quit linuxstb ("CGI:IRC (EOF)")
19:23:33 Quit Moos (Read error: 110 (Connection timed out))
19:29:12***Saving seen data "./dancer.seen"
20:00:13 Quit jeddy3 ("Lost terminal")
20:10:35 Join webguest89 [0] (
20:11:09 Quit webguest89 (Client Quit)
20:28:50 Join linuxstb [0] (
20:34:19 Quit phaedrus961 ("Leaving")
20:36:30 Join phaedrus961 [0] (
20:38:17 Quit XavierGr (Read error: 110 (Connection timed out))
21:00:52 Join muesli- [0] (
21:02:27 Quit seablue ("0")
21:29:15***Saving seen data "./dancer.seen"
21:32:36 Quit muesli- (Read error: 104 (Connection reset by peer))
21:33:56 Join webguest00 [0] (
21:34:20 Join Paul_The_Nerd [0] (
21:34:37 Quit Lear ("Chatzilla [Firefox 1.4.1/undefined]")
21:40:08 Join muesli- [0] (
21:41:03 Quit webguest00 ("CGI:IRC (EOF)")
21:41:06Paul_The_NerdAnyone here familiar with WPS? I'm having a relatively minor issue, but it's kinda bugging me.
21:42:52dpassen1development-wise or user-wise?
21:43:20Paul_The_NerdIt's more user-wise but it might be an indication of a minor bug in the software.
21:43:28Paul_The_NerdActually, lemme update to today's build and verify that it's still there.
21:47:09 Join Febs [0] (
21:47:17Paul_The_NerdNope, problem's still there
21:47:39Paul_The_NerdMy second and third lines merge in my WPS (into one line) unless I put an extra line between them.
21:47:39dpassen1what is it
21:47:55dpassen1can i see the code
21:48:09dpassen1i'm no expert
21:49:00Paul_The_NerdOne second
21:49:34 Quit ender` (Read error: 104 (Connection reset by peer))
21:50:11 Join RotAtoR [0] (
21:51:51Paul_The_NerdSee the completely blank line as the third line? If I remove it, it draws line 2 and 3 on the same line.
21:52:25dpassen1as is, does it leave a blank line between lines 2 and 3?
21:53:07Paul_The_NerdOn the screen the contents of line 3 are concatenated onto line 2.
21:53:25Paul_The_NerdSo if I leave it blank, it works well enough, but it seems an odd thing to happen.
21:54:52 Join ender` [0] (i=ychat@
21:55:01dpassen1whats with the odd indentation of the first few lines
21:55:02Paul_The_NerdAs it is right now, on the unit, it displays Song on line one, Artist on line 2, and Album on line 3. If I remove that blank line, it displays Song on one, and Artist *and* Album on line 2.
21:55:33Paul_The_NerdThey're indented because I have an image that uses that part of the screen
21:56:01dpassen1all your ifs are closed, so its nothing with that, you're going to have to find someone with greater knowledge than i
21:57:28Paul_The_NerdI checked all the ifs first. It's a modification of someone else's WPS
21:57:40Paul_The_NerdAnd one day it just stopped working. Sadly, I don't really remember when
21:58:02dpassen1That's quite odd.
21:58:09Paul_The_NerdWell, it was with some daily build
21:58:25Paul_The_NerdI just don't remember when. At the time I was in a rush, so I just added the line, it worked, I left.
21:58:35Paul_The_NerdI was looking at it and thinking about changing it, and remembered the problem.
22:12:45 Quit muesli- (Read error: 113 (No route to host))
22:24:24 Nick gromit` is now known as weregromit (n=gromit`
22:24:27 Nick weregromit is now known as were-gromit (n=gromit`
22:29:24 Join OnkelJonas [0] (
22:31:36 Join muesli- [0] (
22:42:58 Join toolmanwill [0] (
22:44:29toolmanwillhello everyone, has anyone looked into why the firmware/bootloader becomes corrupted if the battery runs down while in bootloader mode?
22:44:56Paul_The_NerdIt was being discussed earlier.
22:44:58toolmanwilljust curious cause i seen others have had a similar problem
22:45:07toolmanwillmine is on its way to iriver america now for even exchange
22:45:57muesli-TiMiD ;)
22:48:01 Quit toolmanwill ("CGI:IRC (EOF)")
22:48:03 Join toolmanwill [0] (
22:50:54toolmanwillheh not much said about it, but then again linus hasnt seen Febs comment about it either
22:51:25toolmanwilli'm sorta afraid to use rockbox on my replacement now if thats not fixed, dont wanna brick another player
22:52:54Paul_The_NerdWell if it *only* happens when the battery dies, isn't it pretty easy to avoid?
22:53:43toolmanwillsometimes shit happens
22:54:10toolmanwillso just in case there should be some kind of shutdown
22:54:27Paul_The_NerdIt sounds like they were discussing that as a solution
22:55:02toolmanwillgood, i'd hate for many others to have this happen as an accident
22:55:54toolmanwilli guess rule 3 1 should be never EVER leave ur rockbox in bootloader usb mode if not plugged in and you have checked that fact
22:56:02toolmanwilloops rule
22:57:27OnkelJonaspretty ironic that its the feature that can save your butt from a messedup player that ends up bricking one
22:58:52Paul_The_Nerdtoolmanwill: Was yours one of the posts on misticriver?
22:59:50OnkelJonaswoa... really funky soundcorruption after turning id3 database on/off (with no db file present) ... ouch my ears
23:00:57Paul_The_NerdYeah, from my perspective the id3 database doesn't seem too reliable/consistent
23:01:17Paul_The_NerdIt's caused that problem for me as well
23:01:18BagderI agree
23:01:25Bagderbut then hardly anyone works on it
23:01:36Bagderso that's no surprise
23:02:16 Join DangerousDan [0] (
23:02:18Paul_The_NerdWell it *generally* works on Archos, no?
23:02:20OnkelJonasi have fanatically well sorted (and tagged) files, so I don't need it - especially now with directory caching (the lack of that has been my biggest annoyance with rb...)
23:02:29BagderPaul_The_Nerd: it works exactly the same on archos
23:02:51OnkelJonasdoes archos have the modular codec system as well?
23:02:52BagderArchos users are not used to use a db
23:02:59BagderOnkelJonas: no
23:03:12OnkelJonaskkk... just curious
23:03:23 Quit DangerousDan (Client Quit)
23:05:30Paul_The_NerdI think a lot of other things qualify as 'more important' than the DB anyway
23:05:43Bagderme too
23:05:55toolmanwill@Paul the nerd, yes i post on those threads
23:06:00BagderI don't use the db anyway, I just wrote the initial perl version
23:06:19Paul_The_NerdThe perl and the java both produce corrupted tagdb files for me.
23:06:36Bagderyes, they have bugs
23:06:53Paul_The_NerdThey produce databases corrupted in the same manner at least, so it's consistent behaviour. :)
23:07:08Paul_The_NerdI'm not much for databases either
23:07:15Bagderif you know a specific file that messes it up, I could consider doing some debugging
23:07:33OnkelJonassure... i dont care about it either - i was just playing around and was pretty surprised by the nasty sound corruption...
23:07:40Paul_The_NerdI think it's just because my files have Ape and ID3 tags
23:08:14Paul_The_NerdI merged the tags and stripped the Ape ones, but I haven't retried generating the database. (The merge shouldn't have been necessary, as the data was identical, but I decided to play it safe.)
23:08:37OnkelJonasbtw - I have an album that seems to get the start of its files cut out - is there a bug / feature ths could be?
23:09:43BagderPaul_The_Nerd: right, those things have no APE knowledge what so ever
23:10:11BagderOnkelJonas: mp3?
23:11:27Paul_The_NerdBut it's not a big deal for me. The DB would be neat, but I think I used the db feature... maybe 4 times total on the iRiver firmware...
23:12:09OnkelJonasBagder: yes
23:12:42BagderOnkelJonas: then I'd suspect that the files either have weird VBR headers, weird id3 tags or that there is a rockbox id3 parsing bug
23:15:01OnkelJonaskk... it doesn't look funky, but who knows
23:15:11 Quit muesli- (Read error: 113 (No route to host))
23:19:32Paul_The_NerdYou could run vbrfix and see if that does anything, OnkelJonas.
23:20:50OnkelJonasi dont think im gonna bother... its just a single album i dont listen to often... it was just incase there was an easy fix :P
23:22:38 Quit RotAtoR ()
23:23:35 Quit toolmanwill ("CGI:IRC (EOF)")
23:29:18***Saving seen data "./dancer.seen"
23:32:50 Join DangerousDan [0] (
23:32:50 Quit CoCoLUS (Read error: 104 (Connection reset by peer))
23:33:35OnkelJonasbtw... is there any chance album art will ever be implemented?
23:35:13 Join CoCoLUS [0] (
23:35:30OnkelJonasheh .... let me refrase - are there any plans to implement album art?
23:35:40Bagdernot by me
23:39:53 Join Bondold [0] (
23:41:39 Quit Bondold (Client Quit)
23:41:41 Join webguest51 [0] (
23:42:54 Quit webguest51 (Client Quit)
23:45:06 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))

Previous day | Next day