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 2005-08-31

00:00:54amiconnLinusN: I had to add another item to the ReleaseTodo list :(
00:02:06XavierGrhmm why the put_cursorxy is removed and remained only for LCD_CHARcELLS?
00:05:54LinusNbecause it was only used for charcells
00:06:08 Quit midk ("Leaving")
00:06:19LinusNafter the arrow line selector was removed
00:08:14XavierGrwhat you removed the arrow line from the LCD_BITMAP?
00:08:52amiconnLinusN: 2 more items...
00:09:42XavierGrI am trying to adjust my remote build with the latest updates but I am a little lost...
00:10:01XavierGrYou removed the arrow from charcells or from the bitmap?
00:10:57XavierGroh so the bitmap has only the inverse cursor?
00:11:36Bagderbitmap models, yes
00:12:47XavierGrWhy did you do that? I mean, ok I didn't like the arrow too but maybe others...
00:13:23Bagderwe did it because we believe it was the right move
00:13:35Bagderto reduce option clutter and reduce memory footprint
00:13:45CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:13:45*linuxstb likes the arrow (liked...)
00:13:53Wetthm, sorry but I have a little question about lcd_blit : is there a difference between the buffer we have to give to on the archos and on the iriver ?
00:14:54amiconnlcd_blit() takes native lcd bitmaps
00:15:19Bagdertime to schleeep
00:15:27preglowmoving people around is tiring
00:15:31Wettso, for an archos I have 8 pixels in a byte, and on iriver only 4 pixels ?
00:16:32amiconnHowever, if you plan to do fast pixel flicking (like the grayscale library does) be prepared that using the 2 grey levels on iriver will look rather ugly
00:17:01amiconnThe fast flicking interferes with the internal greylevel generation of the lcd
00:17:34amiconnNo problem if you change data slowly, but then you should better not use lcd_blit() for compatibility reasons
00:17:59Wetterr, don't get what you mean
00:18:14amiconnEventually I'll change the behaviour of lcd_blit on iriver to take the same format as on archos. Requires a little preprocessing...
00:18:56Wettin fact i'm just trying to port the video.rock on iriver on my own, without sound for now
00:19:04amiconnAh, yes
00:19:20amiconnThen you may experience what I mean
00:19:51amiconnThe iriver lcd can generate 4 shades from the 2 bits per pixel: black, white, and 2 levels of grey
00:20:18amiconnHowever, better don't use the 2 grey levels if you're fast-flicking through multiple frames
00:20:29amiconnBoth the grayscale library and video.rock do this
00:20:47amiconnIf you use the 2 grey levels it will look very ugly
00:20:53Wettokok, got it :)
00:21:14Wettthe video format seems to be just in b/w, so it should be ok
00:21:15amiconnThe correct framerate on iriver is 70 Hz
00:21:28amiconnYes, but you can't take the data as-is
00:21:49amiconnThe current b&w video format is tailored to the archos lcd bitmap format
00:21:57Wettyes, i'll have to kind of convert 0 -> 00 and 1 -> 11
00:22:21amiconnI recommend a lookup table for this, like the one lcd-h100.c uses
00:22:26 Quit ansivirus (Read error: 104 (Connection reset by peer))
00:22:41amiconn(one lookup per nibble gives 2 bytes from one byte)
00:23:43amiconnMy idea was to do this in lcd_blit() one day. Removes the need to do this in a plugin, and lcd_blit() has that lookup table readily available as it is part of the lcd driver
00:23:53 Join hshah [0] (
00:24:02Wettlookup per nibble gives 2 bytes from one byte)
00:24:02Wett[00:24] <amiconn> My idea was to do this in lcd_blit() one day. Removes the need to do this in a plugin, and lcd_blit() has that lookup
00:24:12amiconnLinusN: Got an I0C: UserBrk while playback...
00:24:28Wettwhat the... well, gonna see that, thanks amiconn
00:24:39hshahwow - a lot of changes made today...
00:24:58amiconnLinusN: Just a sec, looking up the exact code spot...
00:28:04 Quit hshah (Client Quit)
00:28:33 Join ansivirus [0] (
00:29:10 Join ashridah [0] (
00:31:01amiconnLinusN: Different spot now, and it happened at the very end of the playlist
00:31:45amiconnThis time it was in playback_tick() (mpeg.c line 780)
00:31:49amiconnptd->id3.elapsed += (current_tick - last_dma_tick) * 1000 / HZ;
00:34:24 Quit Sucka ("a bird in the bush is worth two in your house")
00:42:37***Saving seen data "./dancer.seen"
00:43:55 Join TCK [0] (
00:51:10 Join Naked [0] (
00:52:07 Quit Hadaka (Read error: 110 (Connection timed out))
00:52:07 Nick Naked is now known as Hadaka (
00:52:15linuxstbI must be on my own with this opinion, but I've just tried the bar line selector for the first time, and really don't like it :(
00:54:21 Part LinusN
00:57:58 Join Sucka [0] (
00:58:34 Quit TCK- (Read error: 110 (Connection timed out))
00:59:33 Quit XavierGr ()
01:01:48Mooslinuxstb: just habit question :)
01:02:47Moostime to go, good night all
01:02:51linuxstbBut isn't the pointer the default option? So lots of people will have the same habit.
01:03:27Mooslot of people change this default option for reverse bar
01:03:50Moosyou was an ~exeption :)
01:04:07Moosgood night
01:04:11 Quit Moos (" HydraIRC -> <- Go on, try it!")
01:10:00 Quit ashridah ("Leaving")
01:10:15 Part dwihno
01:10:27 Join dwihno [0] (n=dw@
01:16:38preglowi think caching is a good thing, provided it doesn't clutter code too much
01:16:42preglowi'd use it for sure
01:16:44preglowi browse a lot
01:17:07preglowand the disk spinup does actually annoy slightly, if not tons
01:19:33 Quit ender` (Read error: 113 (No route to host))
01:19:35preglowsure, it'll have flaws, but disable it then, the code complexity argument is what decides the matter for me
01:23:28 Join ghostiger [0] (n=ghostige@tor/session/x-491c743d2d923aec)
01:24:20 Join Febs [0] (
01:25:07 Quit Febs (Client Quit)
01:25:17 Quit ghostiger (Client Quit)
01:25:52 Join ghostiger [0] (
01:26:03 Quit cYmen ("zZz")
01:26:05 Quit ghostiger (Read error: 104 (Connection reset by peer))
01:26:54 Quit lostlogic (Read error: 110 (Connection timed out))
01:27:20 Join lostlogic [0] (
01:27:57 Quit Sucka ("a bird in the bush is worth two in your house")
01:28:06 Join Sucka [0] (
01:30:51 Join solex_ [0] (
01:34:17 Quit Sucka ("a bird in the bush is worth two in your house")
01:34:24 Join Sucka [0] (
01:46:46 Quit solex (Read error: 110 (Connection timed out))
01:47:47 Quit Sucka ("a bird in the bush is worth two in your house")
01:58:11 Quit preglow ("leaving")
02:06:26 Quit lostlogic (Read error: 110 (Connection timed out))
02:07:00 Join lostlogic [0] (
02:10:59 Join Febs [0] (
02:13:16 Quit Wett ()
02:28:43 Join cYmen [0] (
02:42:40***Saving seen data "./dancer.seen"
03:03:59 Join JoeBorn [0] (
03:07:12 Quit HCl (
03:07:12NJoinHCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
03:12:26 Quit Slasher (
03:17:28NJoinSlasher [0] (
03:20:54 Join bagawk [0] (
03:41:16 Quit bagawk ("Leaving")
03:42:00 Join OnkelJonas [0] (
03:47:13 Join Mxm`Pas`Bien [0] (
03:47:27 Quit Maxime (Read error: 104 (Connection reset by peer))
03:50:54 Quit pike ()
03:51:35 Join pike [0] (
04:29:17 Quit JoeBorn ("Leaving")
04:42:41***Saving seen data "./dancer.seen"
04:44:16 Quit HCl (Read error: 60 (Operation timed out))
04:46:55 Join HCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
04:47:50 Part OnkelJonas
04:54:01 Join OnkelJonas [0] (
05:05:47 Part OnkelJonas
05:31:00 Join Maxime [0] (
05:31:00 Quit Mxm`Pas`Bien (Read error: 104 (Connection reset by peer))
06:00:04 Quit Slasher (
06:00:04 Quit solex_ (
06:00:04 Quit Hadaka (
06:00:04 Quit xen` (
06:00:04 Quit linuxstb (
06:00:04 Quit webguest07 (
06:00:04 Quit pill (
06:00:04 Quit pabs (
06:00:04 Quit HCl (
06:00:04 Quit Maxime (
06:00:04 Quit ansivirus (
06:00:04 Quit DangerousDan (
06:00:04 Quit cYmen (
06:00:04 Quit Febs (
06:00:04 Quit lostlogic (
06:00:04 Quit TCK (
06:00:04 Quit edx (
06:00:04 Quit amiconn (
06:00:04 Quit crwl (
06:00:04 Quit pike (
06:00:04 Quit Bagder (
06:00:04 Quit christel (
06:00:04 Quit Strath (
06:00:04 Quit dwihno (
06:00:04 Quit merbanan (
06:00:04 Quit ze (
06:00:04 Quit odd_ (
06:00:04 Quit thegeek (
06:00:04 Quit Nibbler (
06:00:04 Quit crash_ (
06:00:04 Quit CoCoLUS (
06:00:04 Quit Asku (
06:00:04 Quit AdNauseam (
06:00:04 Quit igor47 (
06:00:12DEBUGEOF from server (Success) (snapshot: netstuff.c line 545)
06:00:12***Saving seen data "./dancer.seen"
06:00:12***Started Dancer V4.16
06:00:12***Connected to on port 6667
06:00:12***Logfile for #rockbox started
06:00:16DEBUGEOF from server (Connection reset by peer) (snapshot: netstuff.c line 545)
06:00:16***No seen item changed, no save performed.
06:00:16***Started Dancer V4.16
06:00:16***Connected to on port 6667
06:00:16***Logfile for #rockbox started
06:00:26***Server message 501: 'logbot :Unknown MODE flag'
06:00:26Mode"logbot :+i" by logbot
06:00:26 Join logbot [0] (n=bjst@
06:00:26 Join godzirra [0] (i=shawn@
06:00:26 Join Maxime [0] (
06:00:26 Join HCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
06:00:26 Join ansivirus [0] (
06:00:26 Join DangerousDan [0] (
06:00:27 Join Seed [0] (
06:00:27Mode"#rockbox -s " by ChanServ (ChanServ@services.)
06:00:30 Join pabs [0] (n=pabs@
06:00:32 Join pill [0] (i=pill@
06:00:34 Join Slasher [0] (
06:00:45 Join CoCoLUS [0] (
06:00:57 Join christel_ [0] (i=christel@freenode/staff/christel)
06:00:59 Join lostlogic [0] (
06:01:01 Join Ismo [0] (
06:01:01 Join pike [0] (
06:01:01 Join dwihno [0] (n=dw@
06:01:01 Join Bagder [0] (
06:01:01 Join christel [0] (i=christel@freenode/staff/christel)
06:01:01 Join Strath [0] (
06:01:01 Join merbanan [0] (
06:01:01 Join ze [0] (
06:01:01 Join odd_ [0] (
06:01:01 Join thegeek [0] (
06:01:11 Join crash [0] (
06:01:11 Join AdNauseam [0] (
06:01:11 Join cYmen [0] (
06:01:11 Join Febs [0] (
06:01:11 Join TCK [0] (
06:01:11 Join edx [0] (
06:01:11 Join amiconn [0] (
06:01:11 Join crwl [0] (
06:01:13 Join pike| [0] (i=pike@
06:01:15 Join [1]Febs [0] (
06:01:21 Join Nibbler [0] (n=sven@
06:01:21 Quit Bagder (
06:01:21 Quit dwihno (
06:01:21 Quit Ismo (
06:01:21 Quit pike (
06:01:21 Quit thegeek (
06:01:21 Quit odd_ (
06:01:21 Quit Strath (
06:01:21 Quit ze (
06:01:21 Quit merbanan (
06:01:21 Quit christel (
06:01:23 Quit Nibbler (
06:01:23 Quit amiconn (
06:01:23 Quit Febs (
06:01:23 Quit AdNauseam (
06:01:23 Quit crwl (
06:01:23 Quit crash (
06:01:23 Quit cYmen (
06:01:23 Quit TCK (
06:01:23 Quit edx (
06:01:25 Nick [1]Febs is now known as Febs (
06:01:26 Join odd [0] (
06:01:28NJoinze [0] (
06:01:32NJoinBagder [0] (
06:01:39NJoinTCK [0] (
06:01:40NJoincrwl [0] (
06:01:40NJoinamiconn [0] (
06:01:41NJoindwihno [0] (n=dw@
06:01:43NJoinmerbanan [0] (
06:01:43 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
06:01:44 Join Rick [0] (i=rick@
06:01:48DEBUGEOF from server (Success) (snapshot: netstuff.c line 545)
06:01:48***Saving seen data "./dancer.seen"
06:01:48***Started Dancer V4.16
06:01:48***Connected to on port 6667
06:01:48***Logfile for #rockbox started
06:01:51***Server message 501: 'logbot :Unknown MODE flag'
06:01:51Mode"logbot :+i" by logbot
06:01:51 Join logbot [0] (
06:01:51 Join Strath [0] (
06:01:51 Join thegeek [0] (
06:01:51 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
06:01:51 Join merbanan [0] (
06:01:51 Join dwihno [0] (n=dw@
06:01:51 Join amiconn [0] (
06:01:51 Join crwl [0] (
06:01:51 Join TCK [0] (
06:01:51 Join Bagder [0] (
06:01:51 Join ze [0] (
06:01:51 Join odd [0] (
06:01:51 Join Febs [0] (
06:01:51 Join lostlogic [0] (
06:01:51 Join CoCoLUS [0] (
06:01:51 Join Slasher [0] (
06:01:51 Join Seed [0] (
06:02:00 Join Maxime [0] (
06:02:00 Join pill [0] (i=pill@
06:02:08 Join pike [0] (
06:02:13 Join ansivirus [0] (
06:02:15 Join edx [0] (
06:02:31 Join HCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
06:02:52 Join Rick [0] (
06:05:10 Join AdNauseam [0] (
06:05:11 Join crash [0] (
06:05:13 Join Ismo [0] (
06:05:15 Join godzirra [0] (
06:05:17 Join pabs [0] (
06:05:27 Join solex [0] (n=jrschulz@
06:05:28 Join Naked [0] (i=naked@
06:05:28 Nick Naked is now known as Hadaka (i=naked@
06:05:37 Join cYmen [0] (
06:05:39 Join igor47 [0] (
06:05:42 Join Asku [0] (
06:05:42 Join christel [0] (i=christel@freenode/staff/christel)
06:05:44 Join mbr [0] (
06:26:08 Join adiamas [0] (
07:00:11 Quit adiamas ("Chatzilla 0.9.68a [Firefox 1.0.6/20050716]")
07:25:40 Join B4gder [0] (
07:26:00 Join LinusN [0] (
07:39:16LinusNthe expected line selector roar:
07:39:48LinusNjust like when we changed the progress bar :-)
07:40:21B4gderI noticed too
07:40:27LinusNmorning amiconn
07:41:59amiconnIt seems the NULL pointer accesses while playback in mpeg.c are gone now
07:47:02amiconnLinusN: Do you think removing the peakmeter max option like kurzhaarrocker suggested will also cause such whine?
07:47:22LinusNthere is always someone...
08:01:49***Saving seen data "./dancer.seen"
08:03:03LinusNok, so should we remove the xing header if the frame count is unknown?
08:07:52CoCoLUSanyone else having problems with freenode lately?
08:08:08LinusNseems shaky nowadays
08:08:34CoCoLUSseems like i have to re-identify myself to nickserv like every hour...
08:09:52amiconnLinusN: I don't think so. I have 2 ideas how to improve on this
08:10:26amiconn(1) I want to make the frame count known in more cases than it is now (when prerecording, and after file split)
08:10:44LinusNthat is possible
08:10:59amiconn(2) If we really can't get the frame count from the MAS because of the counter saturation, we can estimate it from the recording time
08:11:16amiconnShould be precise enough for those long recordings
08:11:44amiconnFrame lengths are fixed, only dependent on the sample frequency...
08:12:17LinusNthe only problem would be if the player expects the frame count to be exact
08:12:44LinusNthere is in fact a third option
08:13:09LinusNbut that is kind of tricky
08:13:22LinusNthe mas delivers the recorded data in bursts
08:13:36LinusNwith a pause until the next frame
08:13:39amiconn(1) requires storing the start frame number of a recording, and that also requires extending the prerecording buffer to store this in addition to the memory positions
08:14:14LinusNwe can measure the time between bursts and find the frame boundaries
08:14:44amiconnAre the bursts *always* continuous, even with heavy i2c traffic?
08:15:19LinusNno, i2c traffic seems to delay the bursts, as far as i have seen
08:15:56amiconnHmm, and the bursts don't correlate exactly with the frames, as they need to be a multiple of 30 bytes
08:16:11LinusNyup, so we won't find the exact boundaries
08:16:21LinusNbut we should be able to count the frames
08:16:26amiconnAnyway, our frame count probably isn't exact now
08:16:57LinusNperhaps not, it might be 1 off
08:17:04amiconnWe do frame boundary in some places search which may shift the count by a few frames
08:17:40LinusNmore than 1?
08:18:47LinusNah, we step back 1800 bytes and search forward from there...
08:23:20 Join ender` [0] (
08:23:32amiconnHehe, funny word order shuffling...
08:23:48amiconn*We do frame boundary search in some places...
08:24:02LinusNi managed to decipher it :-)
08:38:01amiconnIf we use the recorded seconds for a framecount estimate, this estimate should always be lower than the exact count
08:38:12amiconnI doubt that this would disturb a player
08:38:29LinusNprobably not
08:38:50amiconnIf a player would rely on the exact count, I can only imagine 2 minor things happening
08:38:50LinusNmp3directcut might have an issue, if it trusts the vbr header
08:38:54B4gdersounds like a fairly good first shot at least
08:39:33amiconn(1) It really trusts the number, and stops playing after this number of frames. Leaves us with a fraction of a second unplayed at the end
08:39:51amiconn(2) It only uses the number for playtime calculation, like rockbox
08:40:36amiconnMay lead to somewhat odd time display, like 4:00:02 of 4:00:00. Only cosmetic...
08:41:02amiconnHmm, and rockbox does this all the time on archos...
08:41:23amiconn...with lame encoded tracks, so the frame count should be exact
08:47:07LinusNrockbox uses 26ms per frame instead of 26.12
08:48:10LinusNso there will always be a small roundoff error
08:48:12amiconnHmm, there is one situation when calculating the frame count from the playtime doesn't work, and that's if recording from s/pdif and the s/pdif signal isn't present all the time...
08:48:33LinusNah yes
08:48:38amiconnThe 'small' error is often several seconds
08:49:10 Join banan_ [0] (
08:49:38amiconnHowever, we might be able to detect the presence of the s/pdif signal
08:50:08 Quit merbanan (Read error: 104 (Connection reset by peer))
08:51:25LinusNit should be about 1 second off per 4 minutes
08:56:31amiconnThen it matches my observations. I have quite a number of tracks > 10 minutes...
08:56:46 Join merbanan [0] (
08:57:01 Quit banan_ (Read error: 104 (Connection reset by peer))
08:58:39 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
08:58:44 Join einhirn [0] (
09:10:25 Join Paul_The_Nerd [0] (
09:23:14 Join leftright [0] (
09:23:57leftrightI only use the English Language file, is it ok to delete the other language files in .rockbox ?
09:25:03B4gderyou can delete the english too, if you use an english image
09:25:07B4gderas then the strings are built-in
09:28:01Paul_The_NerdQuestion: Does Midi2Wav actually do anything for anyone here? I actually looked at the source file... and the comments imply to me that it doesn't output sound on the hardware, just in the sim... so I'm a little confused...
09:34:59LinusNpaugh: midi2wav does what the name implies, it creates a wav file
09:38:21LinusNi meant Paul_The_Nerd :-)
09:38:34LinusNdamn nick completion
09:38:45*Paul_The_Nerd sighs.
09:39:42Paul_The_NerdSo, it outputs a .wav file somewhere?
09:39:51LinusNyes, probably in the root
09:41:03Paul_The_NerdHave you used it succesfully?
09:41:17LinusNPaul_The_Nerd: hmm, it seems the wav creation code is inside a huge comment
09:41:24Paul_The_NerdThat's what I thought.
09:41:25LinusNi haven't used it
09:41:51Paul_The_NerdAlright, yeah, that's what I got from reading it, but I thought "hrm, I don't really understand a decent portion of what's going on here" so I thought maybe it was doing something that'd create output elsewhere.
09:42:04zei'd say you'd be better of using something like timidity with a patch set you know you like on a host and then encoding the wav to your preference and copying that to your player
09:42:26LinusNze: but where's the coolness in that? :-)
09:42:36*Paul_The_Nerd agrees with LinusN
09:43:00Paul_The_NerdYeah, I'd gotten a sorta "There's no output right now, because I don't want to make WAVs, but I haven't adapted it to work as a codec yet" kinda feeling from the comments.
09:43:09zeok, so if you just wanna have a nice rendition of a midi on your player, go with timidity... if you want coolness, go with midi2wav :p
09:43:50LinusNPaul_The_Nerd: and that translates to "nobody has cared enough to continue the development of the midi codec"
09:44:22LinusNand if it stays that way for the months to come, we will eventually remove it
09:45:09LinusNtime will tell
09:45:25LinusNi for sure won't work on it, i'm busy with other stuff
09:46:19Paul_The_NerdOkay, so, two questions then: Were I to work on it, what's the best way to determine how fast it runs? (Like I noticed the comments say it runs at approximately 78% speed, what's used to determine that sort of thing?) and secondly, what's a good "target" speed for a codec?
09:46:56LinusNthe xxx2wav api performs those calculations for you
09:47:17LinusNand 150-200% is decent imho
09:48:38Paul_The_NerdWell, we'll see if I manage to do this. I've been wanting to work on coding something again for a long while.
09:48:59Paul_The_NerdBut then, 2 hours ago I suddenly acquired a girlfriend, so this whole "time" thing may become less common.
09:49:31Maxime" I suddenly acquired a girlfriend" lol
09:50:05LinusNhmmm, seems like some of the xxx2wav functions are gone
09:50:38paughis there any kind of ETA for improved recording features? (in particular, recording to wav via the digital input)
09:50:49B4gdersome of them probably was scrapped when we converted to real codecs
09:50:55LinusNpaugh: no
09:51:31LinusNB4gder: seems so, how unfortunate
09:52:01B4gderit shouldn't be that hard to get is back, should anyone want to
09:52:12paughLinusN, ah.
09:52:25LinusNB4gder: true
09:52:35*paugh places an ad on ebay for a h140
09:52:57LinusNpaugh: selling your player?
09:53:27paughprobably. i bought it for field recording (before i knew about the glitch)
09:53:32paughnot much use otherwise
10:01:37 Join linuxstb [0] (
10:01:50***Saving seen data "./dancer.seen"
10:01:56paughLinusN, hmm, apparently it's recommended to use a cygwin based environment for development, are there any particularly cogent reasons why i shouldn't use linux instead?
10:02:15B4gderI'd recommend linux any day
10:02:26B4gderthe recommendation for cygwin is for windows people
10:02:36paughah ok
10:02:59B4gderok, the wording could be improved
10:04:02LinusNi changed it now
10:10:01 Part leftright
10:12:32 Join Bger [0] (n=Bager@
10:15:25 Quit Maxime ()
10:17:30 Join Maxime [0] (
10:20:46 Quit Maxime (Client Quit)
10:21:02 Join Maxime [0] (
10:26:35 Quit Maxime ()
10:27:20 Join tvelocity [0] (
10:28:11 Join Maxime [0] (
10:31:06 Join ashridah [0] (
10:51:46solexB4gder: do I need to build the whole toolchain myself even on linux (x86)?
10:52:07B4gderunless you can find a pre-built package somewhere
10:52:27solexpre-built especially for crosscompiling?
10:52:48B4gderwhich presumably isn't that common
10:52:54solexHm. Maybe I should start to learn C sometime
10:53:04solexWhich distro do you use?
10:53:04B4gderbuilding your own set really isn't hard
10:53:08B4gderI use debian
10:53:13*solex too
10:53:31B4gderbut I've never even thought of using a pre-built package for this ;-)
10:53:46solexWhere do you install it and how do you seperate it from the rest?
10:53:57B4gderI install it in a separate prefix
10:54:04B4gderlike in /usr/local/m68k
10:54:49solexDid you use the package toolchain-source or did you grab everything separately?
10:55:16B4gderI downloaded gcc and binutils tarballs
10:55:44B4gderI mean, for m68k that should be no problem
10:56:07B4gderwhy what?
10:56:16solexwhy is it no problem for m68k?
10:56:36solexas opposed to, er, something else
10:56:37B4gderbecause it works good enough
10:56:54B4gderI'm comparing to SH and calmrisc versions
10:57:08solexI'll give it a try in the next few weeks.
10:57:19solexThis project is really cool.
10:58:01*solex wishes for a python interpreter for iriver :)
10:58:22B4gderthere's a plugin task for you ;-)
10:59:03B4gdershouldn't even be very hard
10:59:06solex95% overhead, 5% fun!
11:00:29 Quit linuxstb ("Leaving")
11:02:35solexHm, if there is Debian for m68k...
11:03:21B4gderfor some weird reason ;-)
11:03:39solexSurprisingly, it is slow!
11:03:57solexWe don't have an MMU, do we?
11:04:25solexand no floats...
11:04:37LinusNand no malloc
11:04:51LinusNno console
11:04:55LinusNno keyboard
11:04:57LinusNno nothing
11:05:06B4gderwe have a cool logo!
11:05:13solexno networking :)
11:05:14LinusNah, forgot that
11:05:30BgerLinusN: there *is* keyboard (virtual):)
11:05:42B4gderand we could in fact provide a malloc for plugins
11:05:43CoCoLUSand we have a harddisk
11:05:55CoCoLUSbetter look at the positive side of things... :)
11:06:04solexever thought about swapping?
11:06:38B4gderswap what?
11:06:52solexto disk, when memory is full ;)
11:07:02B4gderfull of what? music?
11:07:29solexbut swapping music sounds interesting, too :)
11:08:03solexif we could get vmware running, there would be possibilities without a limit!
11:08:40 Join preglow [0] (
11:08:50preglownow there's a plan
11:08:50Maximeserioussam.rock :d
11:08:53Bgerdukenukem is enough :)
11:09:21Paul_The_NerdOh C'mon, they just released the Quake3 source. You *know* you want it. :-P
11:09:36Bgerat least doom 1 & 2 doesn't sound so unpossible :)
11:09:43Maximehm, already have Doom and Quake on SmartPhone :d
11:09:54B4gderlet's write a SH-emulator for iriver rockbox and make it a plugin so that we can run archos firmware on iriver!
11:10:12Maximea player on a player
11:10:21Maximefor nostalgic-one ^^
11:10:22BgerRockbox On Rockbox :)
11:10:38Maximeipod-os on iriver :d
11:11:16solexI have to leave. Bye.
11:11:48Maximewe should have tamagotchi on iriver :d
11:12:29 Quit paugh (Read error: 54 (Connection reset by peer))
11:13:00preglowLinusN: how's the h300 cooperating?
11:13:09Bgerstalled ? :)
11:13:28LinusNno time for h300 atm
11:18:48 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
11:19:07 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.6/20050716]")
11:25:09 Join webguest79 [0] (
11:26:13webguest79someone was/is working on Sudoku game, any firther news ?
11:26:50webguest79I think it was Linuxtb
11:28:08Bgerwhat's this "caption backlight" ?
11:28:23LinusNBger: it lights the backlight when changing tracks
11:31:10*Bger wonders how to translate this...
11:31:39t0maswhat language?
11:31:40preglowi translated it to something like 'backlight when changing tracks', no good equivalent
11:31:56Bgeri see it simillar :)
11:33:15 Part christel
11:40:42 Join Wett [0] (
11:41:35preglowhas anyone ever used a rockbox unit for beatmatching? :P
11:42:56ashridahit doesn't have pitch shifting yet does it?
11:43:44LinusNashridah: archos
11:44:06LinusNonly pitch change
11:44:18LinusNpreglow: yes
11:44:36*ashridah sniggers and pushes the guy down the stairs again to the beat
11:45:21LinusNone guy built a dj kit with two archoses and a mixer in a case
11:46:04preglowhah, extremely cool
11:46:16ashridahi'd have to say, the ipod's touchwheel thinggy would be perfect for that kind of thing, wouldn't it?
11:46:23preglowi think i'll enable pitch control on iriver once the freeze is over
11:46:46 Join Moos [0] (
11:46:56MoosMorning all !
11:47:05Bgerafternoon :)
11:47:21ashridahpity that's the only good thing about the ipod :)
11:47:47preglowashridah: yeah, i reckon it would
11:48:22preglowi reckon it would work great, as a matter of fact, depending on the resolution of the wheel
11:49:06ashridahit seems reasonably sensitive
11:53:27preglowdid anything come of the iram discussions?
11:53:31amiconnI'm quite sure I won't like such a trackwheel, as it makes navigating without looking at the display practically impossible
11:54:08preglowindeed, luckily you can also use it as a keypad
11:54:16ashridahamiconn: it still has up/down/forward/backward on it if you touch the axes
11:54:45ashridahand it tics (as in the unit itself tics, not thruogh the headphones) when you use it
11:54:53ashridahso there is feedback from it
11:55:51preglowit really is a nice way to navigate
11:55:55amiconnI just had a look at an ipod again this week. I still don't like the design at all
11:56:09preglowcan't say i _love_ it myself
11:56:10preglowit's decent
11:56:50 Join linuxstb [0] (
11:56:58amiconnThe archos is a bit bulky, agreed, but why the ??? do they paint such a device white!??? (or even more odd colours for the mini)
11:57:58linuxstbwebguest79: Yes, I'm still working on Sudoku. In fact, I was playing it on the way to work this morning.
11:58:18HClwhats sudoku?
11:58:20preglowit isn't painted :P
11:58:44preglowamiconn: some people like it, plain and simple
11:58:50linuxstbHCl: Use google :)
11:59:05preglowactually, most people do
11:59:28Bgerops, error
12:01:04 Join crashd [0] (
12:01:53***Saving seen data "./dancer.seen"
12:04:41HClnot really my kind of game
12:06:09Bgerwhat's this "Jump Scroll" ? :)
12:08:42 Nick paugh is now known as AliasCoffee (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
12:13:03 Join hicks [0] (
12:15:30 Join webguest68 [0] (
12:16:31B4gderBger: it moves the whole line a screen-width at a time, iirc
12:17:10B4gdermove, stop, move, stop
12:17:35 Join Nibbler [0] (
12:17:35Bgerhm, i must see it in the simulator
12:31:14preglowLinusN, amiconn: but yeah, did anything come of the iram discussions?
12:36:04Moosfor increase codec part?
12:36:39webguest79linuxstb: hah, good news indeed, sounds good :)
12:43:43amiconnBger: Jump scroll is player only atm, but I think this would be a good option for all other targets as well.
12:43:54amiconnOne day I'll probably implement it...
12:50:46 Quit tvelocity (Read error: 110 (Connection timed out))
12:54:24preglowMoos: yes, so i can optimise the codecs more
12:55:37Moosthat would be very cool :)
12:55:38Bgerpreglow: iirc, only the first of the two banks of SRAM in 5249 (the 64kb one) is capable of DMA
12:56:29preglowso? i don't believe we use dma with iram anyway
12:56:40Bgerwe dont ?
12:57:04preglowwell, i quite frankly don't care, the codec iram surely doesn't
12:57:07Bgeri doubt (except if we don't use DMA for IDE and for sound)
12:57:14preglowso that can use parts of the lasat 64k bank
12:57:16Bgerok, u're right for sure
12:57:33preglowand the last 64k bank is the one that doesn't support dma, i believe
12:57:40linuxstbIRAM is only used for the heavily accessed data needed by the codec - this is seperate to the main buffers, which probably do use DMA.
12:57:49Moospreglow=Mr. codec here :)
12:58:02Mooslinuxstb too :D
12:59:04BgerThe first SRAM, SRAM0 (32 KBytes) cannot be accessed by the on-chip DMAs of the MCF5249. The
12:59:04Bgersecond SRAM, SRAM1 (64 Kbytes), can be accessed by the on-chip DMAs.
12:59:20Bger(taken from MCF5249UM.pdf)
12:59:31preglowlinuxstb: i don't believe they use dma either
12:59:34Bgerpdf page 89
13:00:07linuxstbI'm guessing that only the IDE and Sound systems in Rockbox use DMA - and AFAIK those don't read/write to IRAM.
13:00:15preglowthe audio buffer, which is the only place we use dma for, is in ordinary ram
13:00:17Bgerpreglow: do you know that the 5249's DMA controller cannot access the SDRAM
13:00:21preglowthe ide system doesn't use dma
13:00:29LinusNthe coldfire can't do ide dma
13:00:39LinusNonly audio and rs232
13:00:39linuxstbIn which case it isn't an issue.
13:00:46Bgerwow :(
13:01:04LinusNwell, the 5249 that is
13:01:07LinusNthe 5250 can
13:01:49*Bger don't understand why iriver didn't use 5250 for h3x0 ...
13:02:45preglowprice, availability, etc
13:02:46B4gder"we already have so much experienc with the 5249" perhaps
13:02:59preglowwhat bagder said
13:03:06Bgermaybe :)
13:03:11Bgerwhat about iaudio then ?
13:03:13B4gderthat's happened more than once in the past in other projects
13:03:21amiconnAfaik the only place where we do use dma is for audio transfer
13:03:50*Bger thought that rb uses dma for ide too...
13:03:54MoosLinus: what's about the iram discussion?, is it possible to increase the codec part? :)
13:03:58amiconn...and no pcm buffer is in iram (?)
13:05:10preglowamiconn: correct, afaik
13:05:47LinusNyes, the pcm buffer is huge, isn't it?
13:05:47Bgerhm, it's very possible i'm wrong then, sorry
13:07:22preglowas far as i know, we never even got iram dma working
13:11:19 Quit Nibbler ("life is like a rental car, you fuck it up, and give it back.")
13:11:56amiconnpreglow: Nothing would change for the core and codecs regarding DMA capability of the iram
13:12:22amiconnCurrently, the core uses the DMA capable 64KB block, and the codecs use the 32KB non-dma-capable block
13:12:24preglowindeed not
13:12:50preglowbut yes, can i start assuming codecs will get more iram, so i can look at further opts?
13:12:58amiconnSwitching to 48/48 would just give 16KB of the dma capable block to the codecs, which still can't use dma of course
13:13:19amiconnI'm all for it
13:13:29preglowany thoughts on whether we should split layer 3 and layer1/2 into separate codecs?
13:14:42preglowthat way i can probably speed up layer1/2 very much
13:15:04preglowsince those layers are pretty small, code wise
13:16:28linuxstbpreglow: Why do you need to split it up? Could we just copy some tables to IRAM during initialisation?
13:16:31amiconnI don't know how much this would change the libmad code, making it harder to incorporate further libmad updates
13:16:54preglowwouldnt change it a bit
13:17:21preglowbut hmm
13:17:30linuxstbI don't think incorporating updates will be an issue - I would be very surprised if libmad ever changes.
13:17:30preglowi don't know how to exclude linking of the different parts, though
13:17:37preglowa couple of ifdefs
13:18:20preglowamiconn: idea would be that libmad and the codec plugin would be the same for both cases, just excluding layer12 code for layer3, and layer3 code for layer12
13:18:38preglowthink i'll put this idea on hold for now
13:20:41amiconnLinking of whole .o files from an .a automatically only links those that contain referenced symbols
13:21:48preglowamiconn: yes, indeed, we'll probably just have to add a couple of defines to the part of libmad that decides whether to call layer12 decode or layer3
13:21:59amiconnTo do this on a function level, you would need to either split the source file into many files, or use sectioned compilation and linkiŽng with garbage collection
13:22:04preglowamiconn: the codec plugin source will be the exact same
13:22:15amiconnThe latter is done in the plugin library
13:22:24amiconn...and bootbox
13:23:20amiconnHmm, but splitting would break playing e.g. layer 2 files with an .mp3 extension
13:24:01preglowyes, yes it would
13:24:05preglowgood point, actually
13:24:17pregloweven though rockbox will need another format detection mechanism in the future
13:25:07preglowlibmad uses a function table to decide which layer function to call
13:25:42preglowso if we neglect filling in those entries we don't need, the functions in question shouldnt be linked
13:32:39 Join _Wett [0] (
13:48:30 Join Sucka [0] (
13:49:34 Quit Wett (Read error: 110 (Connection timed out))
13:50:36Bgerwhat's difference between "insert next" & "queue next" ?
13:50:41Bgerwhat's the...
13:51:38LinusNinserted items stay in the playlist, but queued items disappear after being played
13:52:02Bgeraha, got it
13:52:43*Bger is discouraged from the way the bg translation started to look like...
14:00:40 Join edx__ [0] (
14:01:54***Saving seen data "./dancer.seen"
14:02:43 Join webguest19 [0] (
14:03:30 Quit webguest19 (Client Quit)
14:04:16 Join webguest37 [0] (
14:07:48 Quit edx (Read error: 110 (Connection timed out))
14:08:44 Join Nibbler [0] (
14:09:47 Join kurzhaarrocker [0] (
14:14:15preglowit doesn't have to be perfect
14:14:40Bgerit's not perfect AT ALL ...
14:17:38 Join webguest35 [0] (
14:17:59kurzhaarrockerThere's a little inconvenience with the trigger setup: When you enter the trigger setup from stop mode there's no signal so the meter doesn't do anything. With patch 1179046 it temporarily enters recording mode so that the user can adjust the settings according to the source.
14:18:32kurzhaarrockerDo you think that's worth applying? The patch is small, it doesn't really extend features, but fixes a ui problem.
14:18:49LinusNkurzhaarrocker: does it return to play mode when you leave the settings?
14:19:02kurzhaarrockerno it returns to stop mode of course :)
14:19:17webguest35hi, i found a bug on h100 w/ latest bleeding edge build: it freezes when loading a *.cfg while playing music.
14:19:34kurzhaarrocker(its probably a bit out of date but it should be very easy to update)
14:22:28webguest35goota move, ciao :)
14:22:34 Part webguest35
14:22:56_Wettamiconn : ehe, video works quite well now
14:28:06 Quit B4gder ("Lämnar")
14:28:10 Part LinusN
14:31:18 Part kurzhaarrocker
14:37:41 Nick _Wett is now known as Wett (
14:39:31 Join tvelocity [0] (
14:51:13 Join webguest66 [0] (
14:51:50webguest66anyone here
14:52:08webguest66ok then
14:52:44webguest66do u work with/for rockbox
14:53:11AliasCoffeethose idiots? no way.
14:53:20*AliasCoffee grins
14:53:39AliasCoffeenah i'm just a lurker
14:54:06AliasCoffeecept for now i guess what with the talking and stuff
14:54:20preglowwhat uć?
14:54:22preglowwhat up, even
14:55:14webguest66just back from fist day back in school, which was a half day, how are u
14:58:10 Quit webguest66 ("CGI:IRC")
15:08:03preglowwoop, gotta go
15:11:01 Part webguest79
15:16:43 Quit tvelocity (Read error: 110 (Connection timed out))
15:19:14 Quit Febs (Read error: 110 (Connection timed out))
15:29:07 Join Zagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
15:32:10 Quit webguest37 ("CGI:IRC (EOF)")
15:40:34 Quit AliasCoffee (Excess Flood)
15:43:56 Join AliasCoffee [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
15:55:18 Quit ashridah ("sleep")
15:56:23 Quit Wett ()
16:01:58***Saving seen data "./dancer.seen"
16:12:20 Join Liam [0] (
16:12:55LiamAnyone know anything about H320 Firmware?
16:15:25BgerLiam: just sit down & relax :)
16:15:51LiamSorry :)
16:16:36Bgerno problem, just linus doesn't have time atm for making bootloader for h300
16:17:18LiamYeah, I know, I was just wondering
16:21:57Bgeram i right with the fact that i cannot translate F[123] button bars to bulgarian (at least not with cyrillic letters, because the system font doesn't have cyr. chars?
16:22:07Bger*about the fact*
16:23:57 Quit Liam ("CGI:IRC (EOF)")
16:25:32Bgeris there any way to change the sys font ?
16:29:51 Quit thegeek (Read error: 104 (Connection reset by peer))
16:33:50 Quit webguest68 ("CGI:IRC (Ping timeout)")
16:34:12 Join Lamed [0] (
16:36:37 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
16:41:40Bgeri mean user way
16:42:16ZagorBger: no
16:43:28Bgerthe other way is ln -s rockbox_default.bdf to any cyr font ?
16:45:21Zagoryeah, but lots of things will break since they depend on the size of sysfont
16:45:44Bgerwhat must be it ?
16:46:26Zagori don't remember, check the bdf file
16:46:32Bgerits not proportional, am i right?
16:47:06amiconn6x8 fixed width
16:47:26amiconnThis should be changed for iriver, imho, as the lcd has finer pixels
16:47:37Zagori agree
16:48:10amiconnPerhaps 6x10, that would account for the changed pixel aspect
16:49:28Bgersorry for the silly question, but what's the way to get the font size from .bdf file ?
16:49:41Bgeri'm looking into one and there is line "SIZE 8 96 96"
16:49:45Bgerwhat does that mean ?
16:50:41 Join thegeek [0] (
16:56:10 Join einhirn [0] (
16:57:03 Join bagawk [0] (
16:58:05 Join webguest61 [0] (
17:13:14solexmaybe height an dpi?
17:20:17 Quit AliasCoffee ("Leaving")
17:25:48 Quit Nibbler (Read error: 110 (Connection timed out))
17:26:01 Join Nibbler [0] (
17:52:31 Join DangerousDan [0] (
17:56:53 Join muesli- [0] (
17:57:56Bgerbye rb
17:58:02 Quit Bger ("bye")
17:59:25 Join amiconn_ [0] (
18:00:11 Join muesli__ [0] (
18:00:34SlasherHmm, this is weird.. I am using oggenc with -q 5 and rockbox can barely play those files (slight skips are possible)
18:00:44Slasheram i still using too high quality?
18:00:52 Quit TCK (Read error: 104 (Connection reset by peer))
18:02:00***Saving seen data "./dancer.seen"
18:12:44Slashernever mind, there was some other problems (still investigating)
18:15:52 Quit ender` (Read error: 113 (No route to host))
18:16:04Slasherah, it seems like my directory caching triggered some bug that caused the codecs to behave unexpected slowly
18:16:28 Quit amiconn (Read error: 110 (Connection timed out))
18:16:28 Nick amiconn_ is now known as amiconn (
18:23:04 Quit muesli- (Read error: 110 (Connection timed out))
18:30:48 Part linuxstb
18:37:52 Quit muesli__ (Read error: 110 (Connection timed out))
18:49:20 Quit bagawk ("Leaving")
18:49:33 Join bagawk [0] (
18:50:13 Quit bagawk (Client Quit)
18:57:53 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
18:59:35 Join kurzhaarrocker [0] (
18:59:53 Nick edx__ is now known as edx (
19:01:43kurzhaarrockerin convbdf.c in line 1050 there's a lazy initialization of the variable int ofr. That doesn't compile with old compilers as the declaration of ofr must be at the beginning of a block.
19:02:03kurzhaarrockerI'd correct that myself but I don't have cvs write access...
19:04:54 Quit cYmen ("plonk")
19:09:52 Quit Bagder ("Off to search for that connect-resetting peer guy!")
19:15:03 Join cYmen [0] (
19:34:09 Join Wett [0] (
19:36:08 Quit paugh (Remote closed the connection)
19:39:41 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
19:40:41 Join linuxstb [0] (
20:02:04***Saving seen data "./dancer.seen"
20:15:48 Join IcantFindANick [0] (
20:15:57IcantFindANickwho do I report this
20:16:03IcantFindANickshould be erased?
20:18:14 Nick Lamed is now known as LameD (
20:18:38kurzhaarrockerI don't know anything about the iRiver development. Is that feature already implemented in the iriver, IcantFindANick?
20:18:53 Quit Rick (Read error: 104 (Connection reset by peer))
20:19:05LameDah, yeah, on RockBox. (huh?_
20:19:47 Join FingerSoup [0] (
20:19:51LameDHo it was a joke! "iRiver development"... Funny.
20:20:03 Join Rick [0] (
20:20:34LameDwho has admin rights exept the one who have opend the tracker?
20:21:56kurzhaarrockerThen mentioning that this feature request can be closed here in the irc might be sufficiant. Some of the core developers actually read irc logs (they're sick!). When one of them stumbles across this hint and is in a good mood he might close that feature request.
20:25:05LameDyeah, figures. (they're sick)
20:28:50kurzhaarrockerAnd when you tell them "You need help!" The might say: "Yes, you're right. And You could help with translations, write the manual, ..."
20:38:24 Quit kurzhaarrocker (Read error: 104 (Connection reset by peer))
20:40:58 Join linuxstb_ [0] (
20:42:33FingerSoupjust curious, what are the memory restrictions on building plugins - ie: how much do I have to worry about storage efficiency?
20:44:03linuxstb_It depends if you are on Archos or iRiver - but I'm not sure of the exact numbers.
20:44:25linuxstb_But the principle is always to be as efficient as you can be.
20:47:20FingerSoupOK, I'm fooling around with the idea of doing a connect 4 like game to learn the rockbox API.... what's going to hurt more - processor timing, or memory effieciency... Or am I being too paranoid (My first time developing anything for an embedded device)
20:47:28linuxstb_Checking the source, I think the Archos units have 32K for plugins (code and data), and the iriver has 768K. But if you need a large amount of RAM, you can grab the audio buffer - which obviously stops playback.
20:47:47FingerSoupI don't want that... :)
20:48:06linuxstb_Are you planning a computerised player? That could involve higher RAM requirements for searching.
20:49:27FingerSoupI was... my idea was to store a 2 dimensional 7x6 array of ints, and a seperate array of 7 to keep count of each column...
20:50:03 Quit linuxstb (Read error: 110 (Connection timed out))
20:50:15 Nick linuxstb_ is now known as linuxstb (
20:51:01FingerSoupplayer 1=1, player2=10. to check for win, etc, add in groups of 4...
20:51:41FingerSoupif 4 pieces add up to 4, p1 wins. if 4 pieces add up to 40, p2 wins...
20:52:22FingerSoupcomputer player is just a modification of the win-checking algorythm
20:52:44FingerSoupdo you think this is doable?
20:53:29FingerSoupI actually have an iRiver, but if I can make it work on an archos as well, I will...
20:53:33linuxstbShould be no problem - even on the slower Archos devices.
20:54:17linuxstbI found the minesweeper.c source file very useful when starting my Sudoku game - it also works on all targets.
20:55:06FingerSoupOooh, I was thinking of doing sudoku, but putting in a solver is a bitch.... checking for a win is easy tho... :)
20:55:24FingerSoupI will look into the minesweeper...
20:55:34linuxstbI'm personally not interested in an automatic solver - that's my job :)
20:56:27linuxstbAll I do is three checks (row, column, 3x3 grid) each time a number is added (or changed). The game ends when the board is full.
20:56:29FingerSoupAuto-generators are like auto-solvers: you need to make sure there's only 1 answer....
20:56:48FingerSoupor do you create the puzzles externally?
20:57:11linuxstbYes, an auto-generator is needed - but I am just planning to have some pre-generated levels for the first version.
20:57:27linuxstbA solver/generator can be added later.
20:57:37FingerSoupcool. how long before it's ready for inclusion?
20:57:40amiconnIcantFindANick: Changed the resolution of that FR item to fixed.
20:57:59linuxstbI'm not rushing it - but I hope to be able to commit something when the feature freeze is over.
21:15:10 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
21:20:55FingerSoupI also noticed - rockbox supports bool. is bool in this case a true 1 bit bool, or is it an int? will it save me memory to use bools?
21:21:16 Join matsl [0] (
21:21:20FingerSoup(I am used to C without bool)
21:24:10linuxstbI don't know, but I would guess it's just an int.
21:24:12 Join Musicmad [0] (
21:25:25Musicmadhi all - is anybody else getting no sounds when booting iriver every now and then?
21:26:09dpassen1only if i mess with recording in the debug menu
21:26:10FingerSoupenum would still leave it as an int. I'll just stick to int's
21:26:24linuxstb<stdbool.h> seems to be included everywhere, but I can't find it in the Rockbox source. So I'm confused.
21:27:30FingerSoupMusicmad - nope. however sometimes songs skip to the next one for no apparent readon, and when I try and go back, the song stops... I can then play back the "problem" song without a problem. It's an unreproducable bug....
21:28:05Musicmadok - I get no sound at all when starting the radio for instance. But only sometimes.
21:28:14FingerSoup<> indicte they are standard library things...
21:28:36dpassen1enumeration according to that file
21:29:00linuxstbdpassen1: Yes, but that's only used for the win32 simulator (as the location suggests). I've now found it in my C library includes (/usr/local/coldfire/....)
21:29:10Musicmadfound another thing I think. Play a song 'til the end. Then when I press play again (to well play it again) it resumes the last 2-3 seconds of the song instead?
21:29:19dpassen1linuxstb: is it the same?
21:30:14linuxstbIt #defines "bool" as "_Bool", and true/false as 1/0.
21:30:37FingerSoupso int?
21:30:45FingerSoupor bit?
21:31:32FingerSoupbah, never mind it won't kill my program to use int's...
21:31:55FingerSoup(I'll try and make it as cross-platforma s I can...
21:34:11 Join Lear [0] (
21:55:26 Join XavierGr [0] (
21:56:48*HCl reads the dev mail and is happy about finally someone who appreciates the java tagdb generator ^^
22:02:05***Saving seen data "./dancer.seen"
22:05:57 Join ender` [0] (
22:08:12XavierGrHCl: Is the tagdb finished?
22:12:04HClpretty much
22:12:09HClthe searchengine needs some work
22:12:18HClbut the database is working fine, as far as i know
22:18:57 Quit FingerSoup ("OUCH!!!")
22:20:56linuxstbThis may be a silly question, but does (or could) the tag database work with no tags - e.g. if you have an Artist/Album/Track.mp3 directory structure?
22:25:12dpassen1i think the browser with the tagdb needs to be tweaked a bit
22:25:40 Join Bagder [0] (
22:26:04 Quit Zagor ("Client exiting")
22:34:53 Join bluebrother^ [0] (
22:36:12dpassen1heres how id like it
22:45:36 Quit Lear ("Chatzilla [Firefox 1.0+/undefined]")
22:53:17 Join OnkelJonas [0] (
22:56:40OnkelJonasweeee.... rockbox rocks!
22:57:51OnkelJonasyea... just thought i'd mention it
22:58:30 Quit XavierGr ()
22:58:36BagderI actually used the iriver firmware today
22:58:54Bagderthat boot time really is slooooow
22:59:31Maximeeven if your player is fragmented :x
22:59:34dpassen1i could live with a slow boot
22:59:36OnkelJonasi have *1* wma album, but it is really really good, so I had to delete it from the iriver to stop booting the org. firmware
22:59:36 Join webguest63 [0] (
22:59:36Maximeit's worse
23:00:21linuxstbOnkelJonas: Just convert it to a lossless format (e.g. Wavpack lossless).
23:00:37webguest63Speaking of the iriver firmware. I'd forgotten about it. The other day when Rockbox wouldn't play (my own fault, I had mixed versions of codecs/Rockbox), I left my player in my pocket, instead of using the iriver firmware. Only thought of using it after I got home.
23:00:49OnkelJonas /OT - opera [the browser company] is having a birthday party and giving away free registration keys... its quite worth a try
23:01:22OnkelJonaslol... neer thought og that linuxstb ... and I even use wavpack for all my own rips :D
23:03:28HCllinuxstb: yes, easily.
23:03:46HCllinuxstb: in fact, it already supports this for album/title.mp3/ogg files
23:03:46linuxstbHCl: It _does_ work or _could_ work?
23:04:03HClit doesn't support the artist thing yet, but it shouldn't be hard to add in the java version
23:07:18linuxstbOne thing I do like is the idea of the runtime database - I'm assuming I need a tag database for the runtime database to work?
23:07:24Bagderspeaking of that, you know the meaning of those two undocumented command line switches int he perl version HCl ?
23:08:17Bagder−−dirisalbum and −−dirisalbumname
23:09:39 Quit dpassen1 ()
23:10:54 Quit Musicmad ("Trillian (")
23:22:55 Join ashridah [0] (
23:38:17Wetthm, i guess i can't use mp3_play_data-like functions when using an ihp... Please, can someone tell me what should I use to play sound from mp3 in an ihp plugin ?
23:45:57HCllinuxstb: yes
23:46:24HCllinuxstb: i wrote code that should allow it to generate a tag database on the fly if it didn't have a runtime database, but it was partly rejected, and i can somewhat see the point anyways
23:46:32HClBagder: yes
23:46:47HClBagder: dirisalbum means all files in a directory is a single album
23:47:06HCldirisalbumname means that if it can't find the album tag for a file, it'll revert to the name of the directory the file was in
23:57:34linuxstbWett: Search for "SWCODEC" in apps/plugin.h - that defines the functions.
23:58:02linuxstbI'm not sure how they work though. I think they were added for Rockboy, but I could be wrong.

Previous day | Next day