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 2005-06-20

00:00:23muesli-i am as probably most of rbx-users a windows user..dont know anything about applying any patches ;)
00:00:52 Quit ep0ch (" HydraIRC -> <- IRC with a difference")
00:01:48Bagderthat leaves you with these options: learn how to patch, fix the patch or wait for someone else to complete the patch
00:02:06muesli-bohoo ;)
00:04:00 Join einhirn [0] (
00:06:12 Part hubbel
00:08:45 Join TCK [0] (
00:14:16 Join Strath [0] (
00:14:55 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
00:16:12 Join Tim [0] (
00:21:46amiconnSlasheri: There are major playback problems :(
00:22:23amiconnPlaying mp3s makes the elapsed time go negative (a few seconds, and then fluctuating)
00:22:38amiconn..and after another few seconds it stops completely
00:26:55 Quit TCK- (Read error: 110 (Connection timed out))
00:27:10linuxstbamiconn: I can't replicate that. My (small number) of MP3s seem fine.
00:30:07LinusNworks fine for me too
00:30:39amiconnHappens with all mp3 files I tried
00:30:42linuxstbI did read a similar complaint from someone on misticriver today though.
00:30:50LinusNme too
00:31:13amiconnThe funny thing is that ogg is working fine
00:31:19linuxstbamiconn: Is there anything special about your files?
00:31:26amiconnHowever, my mp3:ogg ratio is ~3000:16
00:31:49amiconnNothing special
00:32:01amiconncbr and vbr, usually completely tagged
00:32:12 Quit webguest78 ("CGI:IRC")
00:33:01linuxstbMy examples are 160kbit/s CBR with id3v2 tags, and 144kbit/s VBR with no tags. I also have untagged MP2 CBR files which play fine.
00:33:05LinusNi just built and tried the latest cvs with several mp3 files
00:33:09LinusNeverything works
00:33:23linuxstbI'm using the latest CVS as well.
00:33:32amiconnBah, got IllInstr now
00:33:47LinusNamiconn: are your codecs really up to date?
00:34:05amiconnI just updated them
00:34:12amiconnNow it seems to work...sometimes
00:34:24Timis ogg really working fine?
00:34:34amiconnThat is, it works if I am playing from stop (or resume)
00:34:51linuxstbTim: They play fine for me.
00:35:22amiconnIf I go back to the browser while already playing, and then start another file, the elapsed time display goes totally mad
00:35:26Timany limitations? pls. point me to a link if I missed one on the main site or forums
00:35:30LinusNamiconn: i see it too
00:35:34amiconn...and this was also when I got IllInstr
00:36:00LinusNTim: try it
00:36:19amiconnOgg doesn't support seek, and the resume point is wrong
00:36:40amiconn(not only the point within the file, which is expected with no seek, but also the file itself)
00:36:41Timhehe - haven't bought a iriver yet - my karma is about dead and I'm looking at alternatives that support Ogg
00:40:29 Quit ajt ("using sirc version 2.211+KSIRC/1.3.12")
00:41:29linuxstbamiconn: Whenever I play an MP3, then return to the browser, and play another, the current track elapsed time starts at 2:42:34
00:42:00amiconnYes, same thing I get
00:44:20linuxstbFLAC works fine, so it looks like an MP3 issue.
00:48:25 Join sockerteze [0] (
00:49:02 Quit Stryke` (Read error: 145 (Connection timed out))
00:53:38 Nick Febs_away is now known as Febs (
00:55:55FebsIf you click back to the beginning of the track, it corrects the problem and the elapsed time displays properly.
00:57:56 Quit Harpy (Read error: 145 (Connection timed out))
01:05:01 Join Coke_dead [0] (
01:08:10Coke_deadIs rockbox starting a H300 mod? :)
01:08:32amiconnYay, no more clipping when pumping up bass/treble :-)
01:15:37 Quit sockerteze ("Chatzilla 0.9.67 [Firefox 1.0.1/20050225]")
01:16:35 Quit daven ("CGI:IRC (Ping timeout)")
01:18:44 Quit elinenbe (Read error: 54 (Connection reset by peer))
01:19:19 Join elinenbe [0] (
01:19:41HClCoke_dead: no.
01:19:52HClnot till the bootloader gets written.
01:20:04Coke_deadBut the site hints work on it. :)
01:20:12HClyea, thats Bagder's fault.
01:20:23Coke_deadAnd rockbox arn't writing the bootloader, are they? :/
01:20:35HClthere's no work done on the h300 aside from the picture for the sim, i think.
01:20:41HCllinus has an h340
01:20:44HCland a bdm wiggler
01:20:51HCljust hasn't had time to work on a bootloader yet
01:20:51amiconnCoke_dead: We're not modding anything btw.
01:20:59amiconnWe're writing firmware from scratch
01:21:03HClthat too.
01:21:09Coke_deaderr. yeah.
01:21:16HClanyways. eventually. linus will write an h3x0 bootloader
01:21:20HClbut don't expect it to happen soon
01:21:26Coke_deadSadness. :/
01:21:35amiconnSomeone needs to identify the lcd
01:21:36Coke_deadiRiver is really crappy with their support of fireware. :/
01:22:30*HCl needs to think up good ways to flexibly allow sorts of "best rated and least recently played" for example.
01:23:16*HCl thought up a way.
01:23:33HClsortexpressions. what fun.
01:23:48HCltime to sleep now
01:33:09 Quit DMJC-L (Connection timed out)
01:35:33 Part DrMouss
01:40:45 Join BTKDaImMaikata [0] (~Bager@
01:40:46 Quit Bger (Read error: 104 (Connection reset by peer))
01:41:13***Saving seen data "./dancer.seen"
01:43:25 Join muesli__ [0] (
01:44:32 Nick Febs is now known as Febs_away (
01:51:42 Quit hicks ("Too lazy to change my quit message")
01:56:37 Part Tim
02:02:05 Quit muesli- (Read error: 113 (No route to host))
02:03:47 Join Bluechip [0] (
02:10:53 Quit Sucka ("a bird in the bush is worth two in your house")
02:13:06 Join xen` [0] (
02:31:08 Part Bluechip
02:32:19 Part LinusN
02:39:17 Quit Febs_away ("Chatzilla [Firefox 1.0.4/20050511]")
02:39:40 Quit kenshin (Read error: 60 (Operation timed out))
02:47:54 Join ashridah [0] (
03:00:36ashridahaah, man. bass sounds much better now.
03:00:53ashridahcan get a decent amount without hearing it clip :)
03:01:09*ashridah hands amiconn a cookie
03:06:53amiconnalmost sleeping though... 03:05 am
03:08:59 Quit muesli__ (Read error: 110 (Connection timed out))
03:36:00 Quit Strath (Read error: 110 (Connection timed out))
03:37:59 Quit elinenbe (" HydraIRC -> <- Leading Edge IRC")
03:40:49 Join kenshin [0] (
03:41:17***Saving seen data "./dancer.seen"
03:44:41 Quit xen` (Read error: 145 (Connection timed out))
03:45:43 Join webguest79 [0] (
03:46:39 Part webguest79
03:49:19 Join jakroo99 [0] (
03:49:23jakroo99whats up guys
03:50:02 Join amiconn_ [0] (
03:50:12jakroo99can anyone tell me what the rockbox team plants to implement in the h3xx firmware which i know is underway
03:50:25jakroo99well at least its wishful thinking that it will be out someday
03:50:34jakroo99im from misticriver
03:52:40kenshini don't believe it's been started, yet.
03:53:36jakroo99well according to the cvs its underway
03:53:46jakroo99i know linus hasnt even written the bootloader yet
03:53:53jakroo99but why is it in the cvs?
03:56:23 Quit jakroo99 ()
03:57:49 Quit amiconn (Read error: 145 (Connection timed out))
03:57:49 Nick amiconn_ is now known as amiconn (
04:05:57 Join QT [0] (as@area51.users.madwifi)
04:13:54 Join Zoom2 [0] (
04:17:36 Quit QT_ (Read error: 110 (Connection timed out))
04:19:06kenshincould someone help me with the simulator real quick?
04:19:45 Quit Zoom2 ("CGI:IRC (EOF)")
04:28:27 Join Dang [0] (
05:16:14 Quit Dang (Read error: 104 (Connection reset by peer))
05:24:23 Join Strath [0] (
05:24:47 Quit kenshin (Read error: 60 (Operation timed out))
05:41:18***Saving seen data "./dancer.seen"
06:05:05 Quit amx ("Client Exiting")
06:19:50 Join kenshin [0] (
07:08:49 Join webguest28 [0] (
07:09:37 Quit webguest28 (Client Quit)
07:17:40 Quit kenshin (Read error: 145 (Connection timed out))
07:22:53 Join Coldtoast [0] (
07:23:20Coldtoastjust updated to the newest bleeding edge and I see crossfading now works. excellent!
07:24:16 Join B4gder [0] (
07:31:19Coldtoastoh. err... for the iriver I mean
07:31:38ColdtoastOTF playlists soon hopefully
07:35:40B4gderrockbox supports that already
07:35:45Coldtoaston the fly playlists
07:35:50Coldtoastnot on the iriver. does it?
07:36:06BTKDaImMaikatamorning :)
07:36:12B4gderyes it does - the exact same way it works on the other players
07:36:19Coldtoasthmmmm. ok
07:38:29Coldtoastlooking at the manual now
07:39:03Coldtoasthmmm. you cna only queue dirs?
07:40:22 Nick BTKDaImMaikata is now known as Bger (~Bager@
07:40:40ashridahColdtoast: you can queue directories, individual files, whatever. just hold the joypad down and go to the playlist options
07:41:05Coldtoastok. looking at it now
07:41:10ColdtoastI did notice something just now with crossfading
07:41:21***Saving seen data "./dancer.seen"
07:41:28Coldtoastit stops working, apparently, once you get to the end of a playlist
07:42:01Coldtoastfirst time thru the playlist, works well but after that, doesn't work at all
07:56:18 Join DMJC-L [0] (
07:58:15Coldtoastcan't find the playlist submenu that has View, Insert, Insert Next, Insert Last, Queue, etc. Only the Playlist section in the options that has Create Playlist, View Current Playlist, Save Current Playlist, etc
08:05:19linuxstbColdtoast: It appears when you hold the joystick down in the file browser, and then select "playlist" from that menu.
08:06:00Coldtoastaaah there it is
08:06:28Coldtoastexcellent. thanx
08:07:41*B4gder is afraid to admit he _never_ used those features
08:11:20 Join Lost-ash [0] (
08:11:45 Quit ashridah (Nick collision from services.)
08:11:48 Nick Lost-ash is now known as ashridah (
08:18:40 Join XShocK [0] (
08:20:01 Join kenshin [0] (
08:22:05 Quit DMJC (Read error: 104 (Connection reset by peer))
08:25:49XShocKI want to congratulate you all! That is an awsome progress since i have been here last time! :)
08:27:45B4gderthe last few weeks sure have changed rockbox land
08:29:58 Quit DMJC-L (Read error: 110 (Connection timed out))
08:30:50Coldtoastrockbox in general?
08:31:47XShocKI am talking about sound framework + that highlight turning slowly.. :))))
08:32:22Coldtoastheh. the only rockbox experience I have is with the iriver
08:32:38Coldtoasthighlight turning slowly=fading backlight?
08:33:48XShocKI am russian with bad english. :)
08:34:03Coldtoastoh, no probs. I can still understand you
08:34:46XShocKHCl, btw, are there any updates on rockboy?
08:34:51ashridahheh, i imagine our russian is far worse than your english.
08:37:55Coldtoastthere are daily builds/bleedign edge builds XShocK
08:40:37ashridahColdtoast: he means has anyone worked on rockboy
08:40:45ashridahthe answer is not in quite some time
08:40:59ashridahit's still < 100% realtime
08:41:01Coldtoastoh. rockBOY
08:41:37ashridahand it's unlikely to ever be perfect, since input is impossible to get perfect, without using the remote as extras
08:41:43ashridahwhich would be unwieldy at best
08:41:44Coldtoastsorry. I'm Australian. my English ain't too great either.
08:42:36ashridahColdtoast: so am i, that's no excuse :)
08:43:05Coldtoastwhere are you?
08:43:21Coldtoastnetspace, eh?
08:43:26*ashridah nods
08:43:28Coldtoastsupposed to be good
08:43:39ashridahwell, their dsl is yeah, except i'm stuck on dialup here
08:43:46ashridahdialup's cheap tho, so i'm not complaining
08:43:55Coldtoastdialup should be free
08:44:06*ashridah is the unfortunate victim of telstra's RIM policy of yesteryear
08:44:33ColdtoastI'm with Internode
08:44:47ashridahdoesn't worry me much, i don't have much money to space to spend on a DSL bill, so 20/month suits me okay
08:45:08Coldtoastit's kinda funny really. one of my friends is with bigpond and his ADSL was down for a couple of days but mine wasn't
08:49:09 Quit ashridah (Remote closed the connection)
08:49:11 Join Lost-ash [0] (
08:49:34Slasheriamiconn: fixed the elapsed counter bug
08:50:54Slasherihi :)
08:51:21 Nick Lost-ash is now known as ashridah (
08:51:32*ashridah mutters
08:51:47BgerSlasheri: very good news ;)
08:51:47ashridahColdtoast: and thus, the price i pay for cheap dialup
08:51:57ashridah(at least there's no bandwidth limits)
08:53:24Coldtoastmind you tho... you are paying more than $20 a month for it
08:53:52Coldtoastyou'd be paying probably $35-40
08:54:59Coldtoastfunny that you were a victim of Talstra's RIM policy and here you are paying them probably $15-20 a month for internet access they're not even providing
09:05:39ashridahthat's still about the cheapest i'll get for a dialup service that doesn't bite
09:06:01ashridahbut it all goes to dad's CC anyway, so i think i can live with the phone bill.
09:07:47Coldtoastmy service provider has 24mbit/1mbit for $60 a month
09:08:27Coldtoastnot available in many areas yet tho
09:08:32Coldtoastcertainly not here in Tas
09:08:39Coldtoastnot even scheduled for this year
09:09:03ashridahi know, believe me, i know what kinds of dsl services are available. lord knows i've installed about 50% of them at various places i've worked at over the last few years
09:10:33Coldtoastyou should sneak into somebody's wifi and get free broadband. heh
09:11:06Coldtoastthere's an unencrypted access point right next to the local Blockbuster here where I am
09:11:27ashridahColdtoast: or i can just go into uni and plug in my mp3 player and leech as much as i want :)
09:11:28Coldtoastso when looking at movies to rent, I can take my PDA and see how they rate on IMDB
09:11:43Coldtoastyeah. good luck with THAT
09:11:52ashridahworks for me so far
09:11:58ashridahbeen over a year, they haven't told me off yet
09:12:04Coldtoastwho are the main ppl getting busted for illegal MP3s here in Australia?
09:12:15Coldtoastthey're idiots then :)
09:12:17ashridahi don't download illegal mp3s
09:12:24Coldtoastserves em right if they get busted
09:12:32Coldtoastfair enough
09:12:56ColdtoastI don't download MP3s at all 99% of the time
09:13:12 Quit Lynx_awy (" HydraIRC -> <- Leading Edge IRC")
09:13:12Coldtoastnot illegal ones anyway
09:13:28ColdtoastI don't buy em either. I really only download stuff ppl I know have written
09:13:56ashridahmostly i end up downloading linux distro isos and stuff
09:14:14Coldtoasthey. Any idea if decoding .ogg is less CPU intensive with Rockbox than it is with the iRiver firmware?
09:17:00ashridahno, although i suspect ogg's a little more intensive
09:17:25ColdtoastI know it is with the iriver firmware
09:17:47Coldtoastppl have tested and it looks like you get around 1/3 less playtime with .ogg than mp3
09:18:02 Join bobTHC [0] (
09:18:13B4gderogg uses more cpu with rockbox too
09:18:27B4gdercompared to mp3 that is
09:18:31Coldtoastany idea how much more?
09:18:49Coldtoastlooks like it's abut 1/3 more intensive with iriver's firmware
09:20:56 Join dator [0] (~irc@d092b07ee4b1a60e.session.tor)
09:22:13datorSlasheri: I think there is a bug where after the songs are loaded into the buffer, jumping back and forth between tracks would have a 2 seconds or so lag time
09:22:39datorI don't know if this is new or even a bug, if someone could test and confirm please
09:24:28Coldtoastwith the iriver port, I noticed that but I thought it might be cos crossfadign wasn't working so the pause is as long as the crossfade should be
09:28:10Coldtoasthmmm. ok. that isn't the case. just checked
09:35:45 Join dator_ [0] (~irc@22e27273181196ff.session.tor)
09:40:12 Quit dator (Remote closed the connection)
09:40:44 Nick dator_ is now known as dator (~irc@22e27273181196ff.session.tor)
09:41:24***Saving seen data "./dancer.seen"
09:48:30 Join Patr3ck [0] (
09:48:52 Join DMJC [0] (
09:53:54amiconnmorning :)
09:54:26Bgermorning, Jens;)
09:55:27 Quit DMJC ("Leaving")
09:58:06 Join DMJC [0] (
10:00:54XShocKgood morning
10:01:33 Quit XShocK (" HydraIRC -> <- 100,000+ downloads can't be wrong")
10:02:24amiconnB4gder: So far I got the impression that ogg actually uses less cpu than mp3 in rockbox
10:02:30datorgood morning too you too :)
10:03:11amiconnPlay some oggs and watch the cpu boost on/off ratio in the debug menu. Do the same with mp3
10:03:23B4gderok, it just surprises me
10:04:02Bgerlet's don't forget that iriver firmware probably uses motorola's mp3 decoder ....
10:04:03amiconnNah, I think preglow said this is due to the fact that libmad dithering is enabled
10:04:14amiconn...and tremor doesn't dither
10:04:29Bgeruf... forget me
10:05:14amiconnI wonder how easy it would be to disable dithering.
10:05:27B4gderfairly easy I believe
10:05:34amiconnSlasheri: Are you here?
10:08:48 Join Harpy [0] (
10:11:04 Join DaKi][er [0] (
10:14:35Slasheriamiconn: yes
10:14:40 Join preglow [0] (
10:14:46preglowremoving dithering is very easy
10:17:18preglowjust remove the scale() call in codecmpa
10:17:21amiconnSlasheri: Fading backlight looks cool, however I think the *fade in* is too slow
10:17:46amiconnIt simply takes too long until you see something after pressing a button
10:17:49Slasheriamiconn: Hmm, yep. I think that configuration options for fade in and fade out should be added
10:18:20amiconnYes. Fade in/out should be switchable, and fade in should imho be shortened to 0.5 s
10:19:07amiconnBetter still, the fade times could be made selectable, but we have to keep an eye on rockbox' number of options
10:22:51 Join Zoom2 [0] (
10:23:26preglowas long as the config menu is well structured, i think we can handle more options, though
10:23:45amiconnSlasheri: I just found that fading backlight causes a big problem on iriver - it leaves no free timer :-(
10:23:46ColdtoastI think the number of options is irrelevent
10:23:56Coldtoastjust so long as they're settings that MATTER we can't provide a timer to the plugins
10:24:17amiconnBybe bye metronome, grayscale lib, video, ... :-(
10:24:37Slasheriamiconn: Hmm.. doesn't kernel tick timer work for that?
10:24:56preglowwe could speed it up? :P
10:25:06amiconnwouldn't work
10:25:20Slasheribut maybe the backlight timer could be made switchable for different purposes
10:25:20amiconnThese plugins need exact timings that are not constand
10:25:45preglowthen let them have it
10:25:51Coldtoastcan't you just have it to the backlight checks to see if the timer is free tho? if not, no fading
10:26:00preglowas long as its done with some api function, the fading backlight can always be disabled then
10:26:08amiconnYes, that's what I thought. When a plugin registers a timer, backlight.c should get its hands off, and resort to simple on/off until the plugin releases the timer again
10:26:20Coldtoastthat'd make sense
10:26:22 Quit ashridah ("phone")
10:26:45amiconnSlasheri: Plugins uses plugin_register_timer() and plugin_unregister_timer() in plugin.c
10:26:58preglowthen it shouldn't be a problem
10:27:01Slasheriamiconn: Ah, good. I will take a look at that
10:27:09amiconnOn archos player where I'll implement fading backlight too we don't have this problem
10:27:15amiconnThe SH1 has 5 timers
10:27:47amiconnSO far we do use timers 0, 1 and 4 iirc
10:27:49Coldtoastany idea how many timers the h300 has?
10:28:00amiconnSame cpu as h100
10:28:10amiconnThese timers are a cpu feature
10:28:23Zoom2I think fading should be custom though, I like long fades, while i know many like short ones
10:28:45preglowno problem in having them custom either, as long as the pwm period is long enough to accomodate many steps
10:28:49amiconnZoom2: The fade out can be long, but fade in must be short for usability
10:29:13Zoom2Also, mabye enable a fade out, to the sleep timer. I use this option alot to try and helping me sleep, I find that it "wakes me up" to have a song cut off when im listening to it
10:29:23Zoom2could that be a possibility?
10:29:28ColdtoastI'd like no fade in personally
10:29:56Coldtoastfade out, yeah. but I like it just ON
10:29:59preglowand i'd like a short fade in and short fade out
10:30:00Slasheriamiconn: it looks an easy task to make the timer work with plugins. I think some timer.c should be written which can be used by both plugins and backlight
10:30:17preglowpreferences vary, obviously, so we should have it configable
10:30:38Zoom2what do you think about the fade out for sleep timer though?
10:30:51amiconnIiuc the current implementation allows setting 0.5s steps for a full fade
10:32:09amiconn...and it uses 2 secs currently
10:32:41Zoom2even for the sleep mode?
10:33:04Zoom2I find quite a few times my songs stop immediatly when I put a 15, or 30 minute timer
10:36:19amiconnZoom2: We're talking about backlight fading, not audio fading
10:37:49Coldtoasthey amiconn: with crossfade, it stopped working for me
10:37:51Zoom2ahh nice.
10:38:01Zoom2didnt mean to deviate topics
10:38:10amiconnSlasheri: backlight.c could have a function, backlight_steal_timer(bool lock), which would be called from plugin.c if the timer is needed
10:38:12Zoom22 cents: crossfading stopped working for me to
10:38:25Coldtoastbut it just started workign again
10:38:38Zoom2i think it still only works on a few tracks
10:38:57amiconnI never used it, and probably never will
10:39:00Zoom2but that is to be expected - still being worked on
10:39:13Coldtoastwell, it needs to be working properly tho
10:39:29Coldtoastone of the BIG things iriver users bitch about is no gapless
10:39:35Zoom2coldtoast: I think slasheri is working on it
10:39:48Coldtoastso, I think, getting that perfect is paramount
10:40:14Zoom2slasheri would fading out be a possibility for the sleep timer?
10:40:19Coldtoastgapless and OTF playlists are the 2 *BIG* things iriver users want
10:40:31Coldtoastthe way OTF is implemented is perfect, imho
10:41:14Zoom2coldtoast I am sure the devs are aware of this and working on it
10:42:02Zoom2the rockbox crew has already put the iriver team to shame, with the time/output ratio on the firmware
10:42:34B4gderwell, we aim at doing a good firmware, they aim at selling hw
10:42:39Coldtoastwell, the iriver firmware does what it does perfectly tho
10:43:03Coldtoastso it's not like the iriver team are completely imcompetent
10:43:23amiconnColdtoast: I agree 100% that gapless playback is a must, but I wonder what crossfading has to do with that
10:43:36amiconnImho crossfading destroys the gapless experience
10:43:42 Join cYmen [0] (
10:43:47Coldtoastwell that depends
10:44:08Coldtoastremoving gapes between 2 REALLY different tracks results in a horrible, abrupt change
10:44:22Coldtoast"removing" the gap by crossfading.. MUCH nicer
10:44:56Slasheriamiconn: Hmm, sounds good
10:45:18SlasheriZoom2: No, separate timer is required for that
10:45:21amiconnGapless playback doesn't mean removing gaps, but rather not adding gaps (like the iriver firmware does)
10:46:08datorSlasheri: have you read my comment above, or is that a known issue already?
10:46:19Coldtoastwhat I noticed with the crossfading not working is the player seemingly stopped buffering the next track
10:46:28Coldtoastso when it changes, it acces the HDD and loads the new track
10:46:42Coldtoastwhen crossfading works, there's the big buffering
10:46:59Coldtoastmy player seemed to stop buffering "properly"
10:47:01Slasheridator: that is a known issue. Track switching can be made a little faster by flushing the pcm buffer from a certain point so that playback would not stop
10:47:23Slasheridator: But i am afraid it can't be made 100% instant
10:48:03datorSlasheri: ah ok, I just started noticing it after it had already buffered, usually I switch tracks before the buffering begins
10:48:34amiconnSlasheri: Are you going to work on this stuff (configurable backlight fade & timer reservation)?
10:48:35 Join Lynx_ [0] (
10:49:13datorpoor Slasheri, he bit off more than he could chew :)
10:49:22Slasheridator: yep, that is caused by the pcm buffer that has buffered raw pcm data a few seconds. If crossfading is enabled, it should start fading at this point
10:49:33amiconnI'm asking because then I shouldn't try to implement backlight fading for archos player now, it could make a big mess interating the changes
10:49:40Zoom2I think gapless is great for "live tracks" but for random tunes I love the crossfading
10:50:09Slasheriamiconn: i can do the timer reservation and some stubs for configuration
10:50:14ColdtoastI agree Zoom2
10:50:41amiconnSlasheri: I could do that as well, that's why I'm asking
10:50:52datorSlasheri: appologies for all the constant nagging from all the iriver users :)
10:50:59Slasheriamiconn: ah, ok :) if you have time, please do so :)
10:51:00amiconnIronically the archos player is the only one that also allows backlight fading...
10:51:07Zoom2Slasheri: when you say a seperate timer is requried for the fading out, can you just make it "mimic" the option for the stop? If stop is enabled for fade out, can that be true for sleep timer and if disabled sleep timer would stop immediatly
10:51:24amiconnSlasheri: What timer is controlling the crossfade?
10:51:40Slasheriamiconn: no timer controls that
10:51:45amiconnAh yes
10:52:29amiconnHmm, I think audio fade out doesn't need a timer either, the tick should be enough
10:53:20Zoom2or if the fade out isn ta possibility, can we have a "sleep timer" by track count too, thus allowing the song to complete fully before ending?
10:53:22SlasheriZoom2: Hmm
10:53:51SlasheriZoom2: The backlight dimming requires a configurable very short timer periods that no other timer can handle
10:54:28Zoom2Slasheri: I was talking about an audio fade out
10:54:36Zoom2sorry to confuse the two
10:54:39Slasherioh :)
10:55:07Zoom2it may sound silly but it supsets me when im falling asleep when a song cuts immediatly while im zoning out
10:55:12Zoom2i find that it wakes me up
10:55:13SlasheriHmm, no separate timer should be required for audio stuff
10:55:14Coldtoastdoes Rockbox on anything support .mod?
10:55:26Zoom2and most of the time i hit play, and have to set it for another 15 minutes
10:56:10Coldtoastor is it a case of Rockbox only beign able to decode various formats, not generate the sounds used (as would need to be done with MOD)?
10:56:34Coldtoasta portable MOD player would be really cool
10:58:03HClmod playback codec is in the cvs but work hasn't started on it
10:58:24B4gderColdtoast: join in and make it happen!
10:59:03ColdtoastI would if I had any C experience. If it was written in Pascal... heh
10:59:12B4gdertime to learn!
10:59:30Zoom2anyway guys, im out - but on an ending note i would just like to say thanks for the hard work, you guys are doing a great job.
10:59:56ColdtoastB4gder: I'd rather work on getting iriver to opensource their firmware :)
10:59:59preglow.mod is coming
11:00:03preglowbut not for a little while
11:00:06Zoom2Slasheri: before I leave, can I get your opinion if the audio fadeout for the sleep timer is pausible and something I can look forward to?
11:00:08B4gderColdtoast: why?
11:00:36preglowi'd rather not have their sources :/
11:00:46Coldtoastnto to use their source
11:00:56B4gderseems like a waste of time and efforts
11:00:56Zoom2preglow: I agree
11:00:58Coldtoastbut to give a COMPLETE insight into the hardware
11:01:02 Join Aison [0] (
11:01:07SlasheriZoom2: yes it is possible
11:01:12preglowwe have pretty complete insight into the hardware
11:01:15Slasheriand i think it's very good idea to have it
11:01:24Coldtoast100% tho?
11:01:33B4gdernot the h300 lcd
11:01:36Zoom2thanks Slasheri, you made my night... I can now go to sleep with hope!
11:01:51preglowthey'll never open source their firmware anyway
11:01:59Zoom2nite all.
11:02:04Slasherinite, sleep well ;)
11:02:12preglowfor them, that's like giving away their secrets to iaudio and other people with similar platforms
11:02:12 Quit Zoom2 ("CGI:IRC (EOF)")
11:02:24preglowB4gder: i wonder if iaudio uses some rockbox code :)
11:03:34B4gderI bet they use our logo somewhere! :-P
11:04:59ColdtoastI really like the look of the M3
11:05:15ColdtoastI rarely ever look at the main LCD on my h140
11:05:34B4gderI do now, since my remote is filled with debug logs ;-)
11:05:45Coldtoastpretty much the only reason I still use the iriver firmware is the LCD on the remote
11:08:25preglowi only started using my remote when it showed debug logs ;)
11:08:47*linuxstb is still searching for his remote...
11:16:09HCli need to wake up...
11:17:48amiconnHmm, there's only one spare timer on archos... for debug builds
11:17:56amiconnOtherwise there are 2
11:18:11Coldtoastso not 5?
11:18:12preglowwhat are the timers used for on archos?
11:20:28amiconnTimer 0 is the kernel tick.
11:20:37Coldtoasthey. does the iriver firmware have some sort of compressor?
11:20:49preglowaudio compressor?
11:20:50preglowif so, no
11:20:55amiconnTimer 1 is the MAS poll interrupt on player, and used as a DMA delay on recorders & ondio
11:21:09amiconnTimer 2 is used for recording debug in debug builds
11:21:20amiconnTimer 4 is the plugin registerable timer
11:21:24preglowColdtoast: why'd you want that anyway? most music is compressed to death anyhow.
11:21:32amiconnTimer 3 is free so far
11:21:54ColdtoastI notice withthe bass settings in rockbox at the moment, anything >12 is a bit much for my earphones but <12 isn't quite enough. I get more bass without distortion with the iriver firmware
11:22:20preglowis this a very recent build?
11:22:38Coldtoastwas a few days ago
11:22:46amiconnpreglow: Timer 1 needs some more eplanations perhaps. The MAS polling on player is necessary because the demand pin can only interrupt on stop demand, not start
11:22:49ColdtoastI haven't tested wioth the latest, which I have installed the SH1 can only trigger on falling edge. This is solved on the recorders by an added inverter which connects to a second interrupt pin
11:23:29Coldtoastwill rockbox support SRS?
11:24:00preglowColdtoast: bass and treble handling was fixed yesterday
11:24:04amiconnThe DMA delay on recorder & ondio is necessary because the demand pin of the MAS goes crazy for layer2
11:24:06Coldtoastaah ok
11:24:25preglowColdtoast: you shouldn't comment on any features unless you've tried the newest build, things change quick these days
11:24:53Coldtoastany plans for parametric EQ?
11:24:58preglowyes, i'll write one
11:25:03preglowbut not for some time
11:25:05Coldtoastgood stuff
11:25:08preglowaugust, perhaps
11:25:14preglowso unless someone beats me to it
11:25:37Coldtoasthey cool. the bass/treble is good now. sorry
11:26:03preglowno worries
11:26:08Coldtoastall the way up to 24 with no distortion. yay
11:26:19amiconnColdtoast: tnx :)
11:26:22preglowamiconn added some clever magic yesterday
11:26:44Coldtoastat 24, there's anoticable change in the sound. But that could be my earphones
11:26:59 Quit DaKi][er (Read error: 110 (Connection timed out))
11:27:06Coldtoastbump the treble up to max and it fixes it
11:27:08preglowif your main volume is too low, the audio has to be scaled down to avoid clipping
11:28:00preglowrockbox handles a bit like the iriver firmware now
11:28:37amiconnpreglow: The clever magic was already there, it only needed to be adapted to uda (after realising that we need to prescale)
11:28:41preglowiriver firmware will just cap the bass boost if it's too loud to avoid clipping
11:28:57amiconnRockbox caps volume instead
11:29:07preglowlike that better
11:29:18preglowthen at least something happens when you adjust the eq
11:29:21Coldtoastcan REALLY hear the scaling when you adjust the bass
11:29:35preglowcoldtoast: not much we can do with that
11:29:44preglowwe can only adjust the bass in 2dB steps
11:29:59preglowso far, we do no processing in software, it's all hardware
11:29:59Coldtoastok. I think it's better anyway, to be honest
11:30:13amiconnColdtoast: For 24 dB of bass not reducing the main volume, you must have main volume at 71% or less
11:30:24Coldtoastat teh moment, rockbox doesn't "flavour" the sound
11:30:52preglowcoldtoast: as for srs, no, i don't believe anyone knows how to do srs processing
11:30:58preglowcoldtoast: and i'll bet it's patented
11:31:11Coldtoastthat's a point, yeah
11:31:24preglowbut we can do tons of dsp that we know how to do
11:32:19preglowas long as it doesn't involve too heavy spectral processing
11:32:32 Quit dator ("Remote closed the connection")
11:33:10Coldtoastthe builds I've checked otu over tha past couple of days have been a lot more stable too
11:33:23*HCl rubs his eyes
11:33:39ColdtoastI don't think I've had a lockup at all over the past 2 days
11:35:42ColdtoastI have my player charging right now. I might actually do a battery test once it finishes charging
11:36:56Bgerbtw, any news about iriver battery level?
11:37:54Bgerif there are not, is there any meaning in makin battery tests...
11:38:05Coldtoasterr... sure thing
11:39:05Coldtoastif I can get 16hrs, the batt level accracy is good enough for me
11:40:17crwlcould someone please fix rockbox/apps/lang/finnish.lang, line 2762? a quote sign is missing...
11:41:25***Saving seen data "./dancer.seen"
11:41:59preglowsomeone just needs to do some voltage measurements
11:43:02amiconnVoltage should be correct; only the runtime estimation needs to be calibrated
11:44:42 Join Lomaxx [0] (
11:45:48Coldtoastwith the same level of bass, the iriver firmware is a lot louder at the moment
11:46:32amiconnB4gder: What do you think; would it be wise to introduce another #define HAVE_* for dimmable backlight, or should I simply use BL_IRIVER || BL_PA14_LO
11:46:54 Join DaKi][er [0] (
11:47:20amiconnColdtoast: This is because iriver firmware caps the bass, so if you make it louder than a certain level, the bass is reduced
11:47:44Coldtoastalso, the Bass/Treble settings are forgotten when I power on again later
11:49:03Coldtoastjust checked. Put Bass and Treble all the way up, powered off then on again, it was louder and brighter and when I go into the settings, my old levels are as I set them but they're not affecting the sound
11:49:13Coldtoastso not forgotten, just ignored
11:49:31amiconnHmm, I'll check. Maybe an init call is missing
11:52:08amiconnHah, indeed
11:52:20preglowk, so how low can rockbox pull the voltage?
11:52:37amiconnuda1380_init() is called after settings_apply, but it always uses zero to init treble/bass
11:52:41Coldtoastactually.. just tested and when I max out Bass, min out treble then power off, I power on, the settings don't affect the sound. I go in, adjust the treble and only the treble settings kicks in
11:52:59Coldtoastso you'll have to go change the treble and bass setting every time you power on at the moment
11:53:01amiconnEither we need to move that up, or it has to use the actual values
11:55:56preglowwell, are the setting variables global?
11:56:21amiconnThe settings are global
11:56:56Lomaxxhi, which audioformats does rockbox support so far?
11:56:58amiconnHowever, we should avoid calling app functions like sound_settings_apply() from firmware code
11:57:40Coldtoastone of my friends found something REALLY irritating with his h300. If you have the USB mode set to DATA and the battery completely runs out, there's absolutely no way to charge without the AC
11:58:29Bgerpreglow: the voltage must never go below 3V for sure
11:59:33Coldtoasthe takes his h300 to work and plugs it into speakers. a few times it's run out and he's had a dead player til he got home. So eventually, when you start porting rockbox to the h300, it might be nice if once the battery gets to a certain level it switches the USB mode to Charge
12:00:10Bgerwhat do you think about accepting the same low battery level as in iriver fw ?
12:00:43HClbut doesn't it work fine with less than that?
12:01:17BgerHCl: the battery can be damaged if its voltage falls under 3V
12:01:23amiconnHCl: LiIon batteries should never be drained below 3 volts, that will shorten their lifetime
12:01:39HClare we draining it below 3 volts?
12:01:49amiconnHowever, iirc Linus said the hardware shuts down above that, so no problem
12:02:16amiconnHowever, we *do* drain below 3 volts on archos fm and v2 recorders, and that's bad imho
12:02:39Bgerdefinitely bad
12:02:56Lomaxxare there chances that rockbox will one day support formats like s3m,it,mod,xm,sid etc? or is that impossible due to technical reasons?
12:03:16HClLomaxx: nope, its possible.
12:03:21HClin fact. code to play them is in cvs
12:03:26HClit just hasn't been ported to fixed point yet
12:03:43LomaxxHCl, which formats exactly? where can i inform myself about it?
12:03:43Coldtoastcool! somebody else wants that!
12:04:04HClofcourse, i want mod playback too, and so does preglow and more people
12:04:15HClLomaxx: at the moment we have libdumb in cvs
12:04:21HClso any format that libdumb supports
12:04:30HClwhich is it, mod,xm at least
12:04:35HClsid is a different codec
12:04:36Lomaxxok, then i google for libdumb, thanks
12:04:39HCland i don't know s3m
12:04:57amiconnColdtoast: Sound settings on startup fixed; wait for the next bleeding edge :)
12:05:06Lomaxxwell it would be nice if it would support all the 180+ formats that deliplayer support ;) but ....YES...i know i am demanding ;)
12:05:14Coldtoastexcelle.t! thankx amiconn
12:05:15HClit still needs quite a lot of work before it will even work
12:05:25amiconnHCl: s3m is a mod format just like mod xm etc (ScreamTracker 3)
12:05:34HClthen that should be supported by dumb as well
12:05:38amiconnup to 32 channels
12:07:00Coldtoast32chans won't be enough for a few s3m's
12:07:08Coldtoaststuff by ppl like Purple Motion
12:07:22Lomaxxare you sure they use more?
12:07:33ColdtoastI think I've seen some of their tracks go to, like, 40+
12:07:33amiconnHuh? I meant s3m has a maximum of 32. Maybe I don't remember correctly :-/
12:07:36ColdtoastI can check actually
12:07:44amiconnI can check too
12:07:49Coldtoastthey might have been using xm
12:07:50preglows3m has a max of 32
12:07:52amiconnI happen to have a heapload of mods
12:07:53preglowxm too
12:07:56preglowit can use more
12:07:59Coldtoastok, that is
12:08:14ColdtoastPurple Motion are the ones that stand out as using a LOT of channels in my mind
12:08:29preglowi've listened to one or two mods in my lifetimes, yet
12:08:30HClgod i need coffee
12:08:40preglowColdtoast: i've seen people use more than him
12:08:42HCli used to have a small collection back in the days when mods were actually popular
12:08:57HClwhen there was no internet and there were bulletin board systems
12:09:10preglowwill be fun to see how well it performs on our cpu
12:09:18preglowit's quite memory intensive, what with all the samples
12:09:28HClpreglow: any idea how to get around dumbs icky floating points usage
12:09:45preglowyes, but i can't start coding that now
12:09:57HClhave time to explain it to me so i could try?
12:10:49preglownot really
12:10:57preglowi need to look at it myself first
12:10:59Lomaxxhm the purple motion s3m-songs i just testes (ca. 6 of them) all had rather low channel-usage. but i haven't got all of them
12:11:15Coldtoasthmmm. well, maybe PM uses nowhere NEAR 32. checking some out now
12:11:27preglowi don't remember him using that much, no
12:11:29preglownever 32, for sure
12:11:39preglowsecond reality s3ms use eight, i thnik
12:11:52ColdtoastScenial uses 13
12:12:08preglowbut don't you worry
12:12:13HCli'll just leave it for you then and work on database stuff
12:12:18preglowi've even got an improved libdumb waiting for rockbox inclusion
12:12:27preglowhcl: i wont work on dumb for a long time yet
12:12:39HClyea, i know, but i can't really manage it
12:13:30preglowseems we have someone on the forum that wants to do a wma fixed point port for us
12:13:43*HCl plans out the new sorting/clipping bit of the searchengine..
12:13:44preglowi'll give him a couple of pointers
12:24:01 Join hicks [0] (
12:28:37amiconnQuick survey: What values would make sense for configurable backlight fade in/out? My idea was Off/0.5s/1s/2s for fade in, and the same plus 3s/4s/5s/10s for fade out
12:28:57preglowoh, i want a quicker fadeout than that
12:29:11preglowi'd want 1 sec fade out
12:29:28amiconn... "the same _plus_..."
12:29:29Coldtoasthe had that in ther
12:29:54ColdtoastI think that sounds perfect but 10secs might be a TAD long
12:30:02Coldtoastbut it wouldn't hurt to have
12:30:16 Join t0mas [0] (
12:30:27amiconnThe 10 is an extra, to use up all settings values that can be stuffed in 3 bits
12:30:37Coldtoastheh. ok
12:30:44preglowwell, that sounds find then
12:31:37preglowfine, even
12:32:46preglowi still wonder about licensing issues, though
12:32:55preglowwill we get in trouble for bundling mp3, aac or wma?
12:33:18Coldtoastdon't bundle them then
12:33:36preglowi'm asking
12:33:52Coldtoastdo the whole "Which you can get from <the usual places>" heh
12:34:29preglowi think frauenhofer said they wouldn't require licenses from people with freely distributable decoders
12:34:40preglowaac is another matter entirely
12:35:26preglowit doesn't even play on the h1x0's original firmware
12:36:07SlasheriHmm, with the latest commit cpu is boosted two times in tree.c. i don't think this is the wanted behaviour?
12:37:48linuxstbRegarding WMA, I tried (but failed) to get the existing floating point WMA decoder integrated into Rockbox. gcc failed with internal compiler errors in the main decoding function in wmadec.c
12:39:03linuxstbm68k-elf-gcc (GCC) 3.4.2
12:40:00Slasherii fixed the cpu_boost thing from tree.c
12:40:08linuxstbI had to comment out 4 statements in order to get it to compile - of the form s->high_band_values[ch][i] = val
12:40:58preglowholy shit
12:41:09preglowffmpeg with only wma support is still a bunch of files
12:41:21linuxstbYes - I am sure it can be simplified a lot more.
12:41:35preglowanyone know if the wma implementation will read all wma files, btw?
12:41:38preglowdecode them, even
12:41:56linuxstbAccording to the source, it decodes wmav1 and wmav2
12:42:08preglowwell, what wma versions are about ?
12:42:17linuxstbBut I know nothing about the various versions...
12:42:28 Join ashridah [0] (
12:43:45preglowwe'll need to do a fast mdct library function some time
12:43:58preglowseems both liba52, aac, tremor and wma needs one
12:43:59linuxstbAll I can say is that it has worked on various live radio streams I've listened to using it (on PPC Linux - so mplayer must be using wmadec.c not the x86 DLL).
12:45:55Bgerit's mine turn to say "holy sh*t"
12:46:16Bgerone guy is trying to convince me in <quote>
12:46:26BgerI read about this problem, what kind of difference betwen US and EU version on rockbox componets chart. So what can i say: The US-Version doesnt have any OTG-Controller and US-Version cant use the USB-OTG-Function, as we can see on your photos, but i has an drm-controller inside of UDA 1380TT Stereo Audio Decoder/Encoder or may be in Motorola Coldfire Integrated Microprocessor SCF 5249VF140, like the new CPU for MACs, thats why H3** bought in the USA also can
12:46:57preglowthe last thing i see is 'USA also can'
12:47:32Bgerbought in the USA also can be used with Windows Media Player 10. The eurpopean version has the same, but just turned around. What would be really interesting, if some one be able to compare KR or JP Version with US and EU Version!!!
12:47:46preglowdrm controller? it's more likely they just decode the file while they load it
12:48:32Bgerpreglow this is absolute bullsh.t
12:49:39linuxstbIf there is a hardware "DRM controller", then I'm sure it's patented. Hence we can get info about it. It's not impossible - digital TV encryption has hardware components of that sort.
12:49:53linuxstbBut (of course) they can be emulated in software.
12:49:55 Quit DaKi][er (Read error: 145 (Connection timed out))
12:50:02Bgerguys, just see
12:51:06preglowthen i don't see why it stops supporting drmed files when you flash it
12:51:15 Join LinusN [0] (
12:51:49LinusNafaik, the DRM only needs a unique serial number for the hardware
12:52:06Bgerpreglow: to resume" this guy just doesn't know what he is talking about...
12:52:32BgerLinusN: afaik you're right
12:52:54preglowwell, he's not the first person without a clue that has opened his mouth
12:53:19Bgerthere was assumption that the serial number is stored in some block in the hex file ...
12:53:44Bgerand when you start upgrading, it just doesn't reflash this special block...
12:53:53preglowit's stored in the eeprom, afaik
12:54:03preglowand the eeprom is always erased when you upgrade
12:54:42Bgerthis is possible too
12:54:54Bgerhow then you can upgrade us version ?
12:55:22LinusNifaik, there haven't been any us firmware updated
12:55:28Bgersome guys have reflashed US players with US firmware and they didn't loose their DRM caps
12:55:35BgerLinusN: yes
12:55:43linuxstbThe flashing program could backup/restore the serial number if it wanted to.
12:55:52Bgerbut some of them were reflashing 1.02US with 1.02US
12:56:40preglowit's more than possible iriver just didn't think about this
12:56:45LinusNunfortunately, my 320 has already been "tainted" with korean firmware, so i can't investigate that
12:57:31BgerLinusN: you can always buy one new unit for this purpose...
12:57:45Bgerso, your unit is US model ?
12:58:15preglowoh, please
12:58:27preglowdoes anyone actually use drmed wma files?
12:58:36preglowif so, they deserve to be unable to play them ;)
12:58:47Bgerone q regarding USB mini connectors ... is it possible to plug in mini-B-male into mini-AB-female connector ?
12:58:47LinusNpreglow: full ack
12:59:29linuxstbBger: Were the people reflasthing 1.02US with 1.02US flashing a modded version of the firmware, or the original?
12:59:30amiconnI don't like digital restrictions management
12:59:41preglowit's not worth our time, if you ask me, and a matter like this might bloody well take quite a lot of time
12:59:42Bgerlinuxstb: original
13:01:01 Join Moos [0] (
13:01:05Bgeri suppose that this file is original ?
13:01:24Bgerdespite of the fact that you can't find link to it from the website
13:02:06preglowwould i be assuming too much if i say the linker always aligns functions by 2? anything else is illegal on colfire, so...
13:02:26LinusN2 or even 4
13:03:00BgerLinusN: any idea about usb connectors ?
13:03:13preglowi just need to do some rockbox before i continue boring myself through the day, so i'll fiddle with flac asm routines
13:03:33Bgergo, preglow, go :)
13:04:33preglowsince code is no longer in iram, splitting the lpc function into two functions doesn't matter too much
13:06:06HClyea, we should not support drm. heh..
13:06:19HClif we do, we're only helping commercial crap organisations like the riaa
13:07:39LinusNBger: no, i don't know
13:08:31*Bger is reading official OTG1_0a.pdf
13:08:57 Join muesli- [0] (
13:10:14Bger"The mini-A receptacle accepts either a Mini-A plug or a Mini-B plug"
13:10:32LinusNthere you go
13:10:33Bgerso if i can't put mini-B female in it, it's NOT a mini-AB, only a mini-A plug
13:11:03Bgerwhy then someone is trying to convince me that USB host plug is mini-AB ???
13:12:38 Quit edx (Read error: 110 (Connection timed out))
13:12:41Bgerhm, he is right...
13:14:12*t0mas starts saving money for a H3xx
13:14:16Bgerstupid me
13:14:24t0masthe thing sounds to intresting not to have it :P
13:14:59Bgermy previous try to put mini-B male in the host connector didn't succeed
13:15:09preglowi didn't know | was a comment character
13:15:13preglowfor asm
13:16:40LinusNpreglow: in some assemblers, any character is a comment character
13:16:56amiconnSlasheri: BL_PWM_INTERVAL is in s, not ms, correct?
13:16:57LinusNas long as there is white space in between
13:17:18preglowLinusN: gas for 68k seems pretty picky
13:17:20Slasheriamiconn: yes, it's in s
13:17:37amiconnOkay, then the comment is wrong, I'll fix it
13:17:43Bgersomeone with US H3x0 here ?
13:17:44Slasherioh :)
13:17:52preglowLinusN: and i hate using /* */ for oneliners, it's too easy to make a mistake and do a block comment when you've got no syntax highlightingg
13:17:54HCli'm getting compile errors on the latest cvs
13:18:13*HCl checks whether its his fault
13:18:36Coke_deadi got a h320
13:18:43*Coke_dead attacks Bger
13:18:54B4gderHCl: 17 green zeroes say you're to blame :-)
13:18:59Bgercan u take a look at your USB connectors?
13:19:03HClB4gder: close :)
13:19:08Bgerare they *exactly* the same ?
13:19:13Coke_deadthey arn't
13:19:13HClcvs created a conflict out of my changes even though it wasn't a conflict
13:19:21Coke_deadand im saying that from memory
13:19:25BgerMEDIA and DATA i mean
13:19:28Coke_deadcause it isn't near me at this moment..
13:19:34Coke_deadi rememebr them being diffrent
13:19:37Bgerr u sure ?
13:19:48Coke_deadlike i said
13:19:52Coke_deadit's not here right now.
13:20:03*Bger goes to ask @ MR
13:20:04LinusNamiconn: will the timer_period calculation in TIMER1 be dynamic?
13:20:16amiconnLinusN: Hmm?
13:20:20LinusN(it is already today, since FREQ is a variable)
13:20:27amiconnAh, yes
13:20:33Coke_dead<3 MR
13:20:49amiconnI won't change that, only the count for increasing/decreasing the duty cycle will vary
13:20:56LinusNtimer_period = FREQ/20000 * BL_PWM_INTERVAL / 100 / 32;
13:21:05LinusNcould very well be precalculated
13:21:10amiconnCurrently it is 4, so 5ms * 4 * 100 steps == 2 seconds
13:21:37amiconn(That's why I suggested a minimum of 500ms)
13:22:13amiconnLinusN: How could that be precalculated? The cpu frequency may change between 2 interrupt calls
13:22:26amiconnOn player it is of course a constant
13:22:39LinusNyes, but the calculation can be done in set_cpu_frequency
13:23:19amiconnThat would be another interdependency. Is that good?
13:23:36LinusNnot sure, i just didn't like the calculations in the irq handler
13:23:43amiconnAnd, I can't commit system.c since it has those MFDR changes in it
13:24:11LinusNhave you tried them yourself?
13:24:11amiconnIt's working fine here, but it would be better to actually measure the clock
13:24:22amiconnI always have them in my builds
13:24:28LinusNi won't be able to measure it for at least a week
13:24:41HClhm.. make dependency bug somewhere.... got linking errors.. fixed after a make clean..
13:25:48preglowLinusN: seems gas only takes | or /* for 68k
13:26:25amiconnSlasheri: I'm tempted to rip out remote dimming. This code isn't used, and it just makes the functions harder to read
13:26:48LinusNwhy doesn't remote dimming work?
13:27:06SlasheriI think remote has EL-backlight, it flickers much if enabled
13:27:26amiconnEL isn't dimmable if the driver chip doesn't support it
13:27:28Slasheriamiconn: Hmm.. If you think we will never need to dim any remote, you could remove the code..
13:28:00*LinusN just learned something about EL backlight
13:28:14amiconnEL uses high voltages that aren't switchable fast enough from the low voltage side
13:28:41preglowhow high voltages?
13:28:42amiconnIt's something around 100 volts square wave 400 Hz
13:28:48Coldtoastsetting the main LCD is funky here
13:28:58Coldtoasterr. setting the main LCD to Inverse
13:29:05preglowwhy use an el backlight?
13:29:43LinusNColdtoast: how?
13:30:05ColdtoastI have vertical stripes of non-inversion
13:30:13Coldtoastlooks a bit like a bar code
13:30:15amiconnEL delivers even backlight distribution without diffusors
13:30:24Slasheriamiconn: I wouldn't remove the remote dimming code completely. What if somebody has a modded remote with led backlight and wants it to be dimmable? =)
13:30:34LinusNColdtoast: weird, it works for me
13:30:51LinusNSlasheri: remove it
13:30:51amiconnColdtoast: Push the contrast a bit
13:30:53ColdtoastI can take a quick photo if you like
13:31:08LinusNSlasheri: we can always get it back from the cvs history
13:31:10amiconnLinusN: Was that directed to me?
13:31:14SlasheriLinusN: ok
13:31:14preglowamiconn: so why didn't they use that for the main lcd?
13:31:25amiconnpreglow: I don't know.
13:31:32amiconnEL has disadvantages as well
13:31:36Bgerpreglow: to be dimmable :P
13:31:50Coldtoastno different amiconn
13:31:55amiconnThe inverter used to generate the high voltages may cause interference
13:32:14amiconnEL has limited lifetime, way lower than LED
13:32:22 Join einhirn [0] (
13:33:00amiconnpreglow: The ondio is prepared for EL backlight, and [IDC]Dragon actually modded his ondio to have EL backlight
13:33:16amiconnLED backlight would be impossible due to limited space
13:33:16Coldtoastusing an inverse bar as the selector doesn't have th elines
13:35:44amiconnpreglow: EL can have different colours, but with different efficiency and lifetime. The blueish variant in the iriver remote is the most common
13:40:54 Quit Patr3ck ()
13:41:21*Bger corrected the wrong information regarding USB connectors in the wiki
13:41:29***Saving seen data "./dancer.seen"
13:42:30preglowshould i make 16 bit versions of all the flac asm routines as well? :V
13:43:18BgerLinusN: it was mine mistake, so it's not goodie ...
13:46:32 Join RRRRRED_M_INFIN [0] (
13:46:44preglownick of the year
13:49:56B4gderLinusN: did you ever get "ownership" of this channel?
13:50:18LinusNforgot about that
13:52:50preglowafraid of takeovers? :P
13:53:15B4gderthose were the days...
13:53:36preglowhahaha, yes....
13:54:08B4gderwe wrote Dancer pretty much because of such stuff
13:54:14B4gderDancer == logbot
13:54:29*HCl goes to make coffee
13:54:42HClannoying, i should've brought my iriver, now i can't develop/test properly :/
13:54:45 Join bipak [0] (
13:55:32preglowwill we ever need to load to instances of a codec at once?
13:56:04linuxstbDo you mean two instances of the same codec, or two different codecs?
13:56:27linuxstbI can't think of a reason why we would.
13:56:41amiconnvoice ui
13:56:47preglowsomeone ought to slay, shoot then kill the libflac author for this c imitating piece of annoying c code that libflac is
13:56:49thegeekyou beat me to it
13:56:56preglowc PLUS PLUS
13:57:15*B4gder spots a cgi::irc user ;-)
13:57:21linuxstbIs that a realistic aim? The codecs are quite memory hungry.
13:57:30amiconnpreglow: However, the voice ui case is special, so probably we could get away with a single, statically linked codec for that
13:57:50preglowyes, and on iriver, we can probably get away by using a dedicated codec
13:57:52preglowlike speex
13:57:58linuxstbI've been thinking of looking at speex - that would seem ideal for.
13:58:06amiconnI'd prefer mp3 at least for a start
13:58:08preglowspeex is very nice code as well
13:58:17preglowshould be a breeze to optimise
13:58:21preglowi wish i had time to do it
13:58:24HClwhats bad about c plus plus imitating c code?
13:58:28preglowbut i will when i do have time
13:58:37BgerHCl: the reverse
13:58:40preglowhcl: well, have a look at it
13:58:51preglowhcl: it's even got private, protected, etc stuff
13:58:55HClnice :p
13:59:01preglowhcl: all the function names are 100 chars long
13:59:22preglowthere's nothing wrong with doing oo in c, but this guy has quite literally just ported a cplusplus prog to c
13:59:27preglowit even mimics constructors
13:59:58preglowamiconn: anyway, the codec used for the voice ui should be very tightly optimised
14:00:13B4gderbtw, we're now mentioned on the flac web page for player support
14:00:23Bgerwow :)
14:00:33*Bger goes to see
14:00:48B4gdernot entirely correct I see
14:01:26preglowbut yeah
14:01:38preglowi'll remove a malloc from libflac and stuff in a static decoder struct instead
14:01:47preglowthis will break multiple instances, but we shouldn't need it
14:01:48 Quit bipak_ (Read error: 60 (Operation timed out))
14:02:08Bgerdefinitely ... i didn't know that the MAS is capable of decoding FLAC :)))
14:02:49HCl :P
14:02:55HClthey'll have to fix that.
14:04:39preglowoh, almighty jehova, this source code is frustrating
14:04:53B4gderhe said jehova!
14:04:55amiconnBger: It might be, the problem is actually implementing it
14:05:10Bgeramiconn in 4kb dsp code ? :)
14:05:56preglowif you can implement mp3 decoding, then you can most certainly implement flac as well
14:05:58amiconnSurely not by imitating c++ :-P
14:06:18preglowbut the mas is a veritable nightmare to program
14:06:24amiconnpreglow: Yeah, but we don't know how much code the mp3 decoder uses
14:06:25Bgerbut afaik the mp3 decoding is in MAS's ROM ...
14:06:26preglowand that's assuming you have proper docs
14:06:42amiconnThe mp3 decoder is in ROM while all loadable codec have to reside in RAM
14:07:00preglowi will do like i said
14:07:05preglowand dont touch libflac
14:07:13Bgeramiconn: what's the size of this ROM ?
14:07:19preglowit is a malloc hell
14:07:45Bgerlinuxstb said that it's mostly malloc-ing for parsing metadata ?
14:09:09linuxstbThat's what I thought after a relatively quick look at it. We should also be able to disable/remove a lot of the metadata parsing code if we are implementing it in the core rockbox code.
14:09:47preglowdamn, the eq stepping is really noticable
14:09:55preglowBger: then he was wrong
14:10:04preglowBger: EVERYTHING is malloced
14:10:14Bgerblah ...
14:10:16preglowor more specifically: calloced
14:10:55 Quit DMJC (Read error: 110 (Connection timed out))
14:11:49 Join DMJC [0] (
14:14:02preglowc is pretty nice and all, but not for embedded platforms...
14:14:08preglowC PLUS PLUS, ARGH!
14:14:16preglowsomeone make this thing escape the pluses already
14:14:39 Join Chamois [0] (
14:15:13preglowoh, and btw, before someone starts asking
14:15:23preglowseeking for musepack files can pretty much be forgotten about
14:15:43Bgerwhy ?
14:15:53preglowbecause the bitstream is braindead
14:16:19preglowthey have no seek table, and all the frames seem to be heavily dependent on older frames
14:16:36Bgerhow much older ?
14:16:43preglowno idea
14:17:11preglowthe solution the libmusepack codec suggested was backup up the decoder state for all the places you might want to seek to...
14:17:15preglowwhich is a no-go for us
14:17:22LinusNhow lame
14:17:26preglowall in all a pretty difficult codec for embedded use
14:17:34preglowa pity, i had pretty high hopes for it
14:18:03preglowbut needless to say, i wont be ripping my cds to it when i can't bloody seek in them without having to jump through flaming hoops
14:18:04Bgerat least non-seeking support is a way better than no support at all
14:18:15preglowbut optimising it will also be a pain
14:18:25preglowas the codec is now, we need to do heavy prescaling of all coefs
14:18:34preglowthis might result in poor sound
14:18:53preglownot to mention that using iram will also be a pain, the decoder struct is huge
14:19:02preglowlike i said, a pretty difficult codec for embedded use
14:19:15LinusNtoo bad
14:19:25Bgerpreglow: very "good" news are coming from you today:)
14:20:10preglowmusepack is BARELY 100% realtime for medium bitrate files now
14:20:38preglownot really
14:20:43linuxstbpreglow: Have you looked any more at getting codecmpc to run?
14:21:10preglowthere's no point until the codec gets further optimising
14:21:17preglowi have it running, btw
14:21:21preglowbut it skips like hell
14:21:43preglowi can commit the little change that's needed, if you want it
14:22:15linuxstbI see no reason not to commit it. At least that part of musepack will be working.
14:27:57 Part LinusN
14:28:42 Quit Audiophi| (Read error: 60 (Operation timed out))
14:30:07amiconnConfigurable backlight fading now working nicely.
14:30:25amiconnCan't commit yet as compilation will break for player
14:30:44 Join RotAtoR [0] (
14:30:46amiconnImplementing it for player is scheduled next
14:31:02 Quit Bger (Read error: 131 (Connection reset by peer))
14:31:11 Join Bger [0] (~Bager@
14:36:50preglowseems the 16 bit lpc routines are the most common
14:37:02preglowmight be clever to implement them in asm using 16 bit multiplies, then
14:47:37 Quit lostlogic_ (
14:47:39 Quit Coldtoast (Read error: 131 (Connection reset by peer))
14:51:07 Join lostlogic [0] (
14:54:16 Join edx [0] (
14:57:00 Join DaKi][er [0] (
15:04:26 Join Coldtoast [0] (
15:08:53 Join Andrew179 [0] (
15:15:17Andrew179Would it be ok if I attempt to create IaudioPort and IaudioHardwareComponents pages on the wiki?
15:15:36B4gderplease do
15:16:11Andrew179great, I don't know if I can be much help with porting, but I do have access to an iAudio M3 that I could scan
15:16:21BgerAndrew179 hm
15:16:35Bgeri don't think it's necessary
15:17:02Coldtoasthow do you like the M3 Andrew179?
15:17:04bobTHCAndrew179 > u can scan the internals ?
15:17:12linuxstbBger: Why not?
15:17:34Bgerlinuxstb we already have a scan
15:17:43B4gderin the wiki?
15:17:46Bger[13:58:30] <Matt-UK> preglow: this is the link if you want it
15:17:54Andrew179My brother has one at the moment, I think it's great. I am planning on getting either a M3 or X5 for myself
15:17:54Bgersome days ago
15:18:18Bgernot in the wiki, but Andrew179 can just use these scans
15:18:22Andrew179I noticed that in the logs, but I was planning on compiling all of that info into wiki pages
15:18:32BgerAndrew179: just do it;)
15:18:33B4gderI think that is a fine indea
15:19:01 Quit Coke_dead (Read error: 54 (Connection reset by peer))
15:19:56Andrew179Alrighty, I'll get started on that
15:20:43Bgerhehe known drive Toshiba MK2004GAL
15:21:10preglowthere are x5 pictures at iaudiophile
15:21:16preglowtoo bad they didn't open it as much as they could
15:21:21preglowno sign of the audio codec, for example
15:21:25preglowor the cpu
15:21:36BgerAndrew179: btw this link is not so good
15:22:19B4gderits about time we got another player supported ;-)
15:22:42Andrew179ok, I'll start out with what's there. I should be able to get better internal scans of the M3 pretty soon
15:22:59 Join Sucka [0] (
15:23:36 Join Sucka` [0] (
15:23:38 Quit Sucka` (Remote closed the connection)
15:23:38 Quit Sucka (Client Quit)
15:23:43linuxstbB4gder: Or another broken column in the daily builds table....
15:23:49 Join Sucka [0] (
15:24:25Bgeri mean most of the pictures in it are unclear
15:24:25Bgerthere can be revision problems with iAudios ... i see V1.4 PCB
15:24:25Bgersome days ago i saw V1.3PCB ...
15:26:03Andrew179That could complicate things... :)
15:27:06 Quit muesli- (Read error: 113 (No route to host))
15:32:42 Join muesli- [0] (
15:35:06 Quit muesli- (Client Quit)
15:38:39 Join ec [0] (
15:39:36*Bger hopes that h3x0 port will happen before iAudio...
15:41:31***Saving seen data "./dancer.seen"
15:43:36 Quit Chamois (Read error: 131 (Connection reset by peer))
15:44:24 Part ec ("Leaving")
15:44:58preglowwell, linux has got h3x0
15:45:02preglowhe hasn't got an iaudio
15:47:12Coldtoastthe h3x0 are SO ugly
15:47:24Coldtoastthe designer must have had bad mushrooms or something
15:47:41Bgerthe designer is the same as the one of h1x0
15:47:46ColdtoastI know
15:47:48Bger(at least the same company)
15:47:59Coldtoasterr.. yeah. sorry
15:48:13ColdtoastI like the h1x0
15:48:18preglowh3x0 isn't exactly pretty, no
15:48:29preglowa designer on shrooms would have done better
15:48:30Coldtoastthe button layout is terrible too
15:48:58Coldtoastall smooshed together in the middle like that
15:49:21*Bger bought it because of the functions, not the appearance
15:49:37Coldtoastwhich functions?
15:49:44Coldtoastnot the so-called video playback I hope? heh
15:49:52Bgerneh :)
15:50:04ColdtoastUSBOTG would be ok I guess
15:50:06Bgerusb host, color screen
15:50:30Bgerand ... future rockbox
15:50:41ColdtoastI chose digital in/out over all that
15:51:24Bgerlet's be honest ... you don't have s/pdif phones, do you ?
15:51:38ColdtoastI have a mixer with digital in/out tho
15:52:12Slasherii also connect digital out often to HT system
15:53:22Bgerbut i must agree that h1x0 are looking more ... robust
15:53:35ColdtoastI should have bought a 20GB player tho instead of the 40GB
15:53:56Suckai should have bought a 60 instead of a 40 :D
15:54:04ColdtoastI've had mine since Nov last year and haven't put new stuff on there for maybe 4 months
15:54:15*Bger wants 80GB 1.8" HDD
15:54:17Coldtoastand it's only got ~8GB of stuff on there
15:54:50ColdtoastI actually like my PDA more
15:55:07Coldtoastit's a bloody good PDA tho
15:55:28thegeekthe sound from my axim x50v is not very good
15:55:35Coldtoastnot very good?
15:55:38Coldtoasthow do you mean?
15:55:42thegeekcompared to the h120 it's mediocre
15:55:42Coldtoastthat's what I have
15:55:48thegeekand compared to my home setup it sucks
15:55:53Bgerand the only one left is to say it's brand and model ... to complete the advertisment :P
15:55:53Coldtoastit's hard to compare tho
15:55:59 Join webguest51 [0] (
15:56:00Coldtoastthe h120 has all the DSP crap
15:56:04thegeekif you have decent headphones you'll notice it very fast
15:56:21thegeekthe sound is flat/dead, bad bass, bad treble, bad midrange;)
15:56:26Coldtoastthe x50v has EQ that's hidden from you tho
15:56:27thegeekit's not horrible
15:56:35thegeekbut when i have the h120 the choice is clear
15:56:42thegeekit's not about an eq
15:56:52thegeekyou can't fix everything with an eq
15:56:55thegeekfar fromit
15:57:05webguest51the progress bar fills up very quickly and the song elapsed time counts up rapidly in seconds continuously
15:57:08Coldtoastthe x50v has the same audio chip as the ipod
15:57:15thegeekI doubt that
15:57:20Coldtoastwanna bet?
15:57:25thegeekthe opamp is certainly not the same
15:57:38thegeekwhen comparing the sound on my h120 and the axim the difference is very obvious
15:57:42ColdtoastI REALLY liek the sound from mine
15:57:51thegeekwhat headphones do you use?
15:57:55ColdtoastI certainly don't find it flat
15:58:11thegeekI use the sennheieser 595, it's VERY transparent(as in very easy to hear problems )
15:58:16Coldtoastsame ones I use for my iriver; Philips HD250s
15:58:28Coldtoastwhich ARE very bass heavy tho
15:58:29webguest51is this hydrogen audio /
15:58:32thegeekwith my koss ur40's I can't hear much difference though
15:58:39thegeekso I guess it depends on the headphones
15:58:46preglowi think ipod has the dac on the portalplayer chip
15:59:37Coldtoastmaybe. I'd use my x50v exclusively if it was robust
15:59:55 Quit courtc (Read error: 113 (No route to host))
16:00:44thegeekget a magnesium case
16:00:44Coldtoastimagine screwing the touchscreen on that!
16:00:44Bgerwebguest51: with the last bleeding edge build ?
16:00:44thegeekI got a sena leather case and a magnesium case
16:00:44DBUGEnqueued KICK thegeek
16:00:44webguest51yes afew minutes ago
16:00:44Coldtoastif I could buy a remote for the x50v, I'd be in hog heaven
16:00:44thegeekit's impossible to make the magnesium case touch the screen
16:00:44Coldtoastdo you like the x50v much?
16:00:44Bgerwebguest51: i suggest to explain the problem
16:00:44***Alert Mode level 1
16:00:44***Alert Mode level 2
16:00:44thegeekread lots of books on it
16:00:45DBUGEnqueued KICK Coldtoast
16:00:45Coldtoastthey're pretty impressive thngs
16:00:45***Alert Mode level 3
16:00:45Bgerand Slasheri will take a look at it some time ...
16:00:47ColdtoastDell have announced there'll be an WM2k5 upgrade for the x50v too
16:00:48thegeekI know
16:00:48thegeekI don't care all that much
16:00:48webguest51when selecting play, the song loads, starts playing and the progrees bar then fills very rapidly
16:01:18Bgerwebguest51: are you sure that the file is 44.1kHz ?
16:01:25Coldtoastknow what I want tho? a passive packetsniffer for WM
16:01:26webguest51the elapsed time counter/display for the song then counts very in intervals of seconds continuously
16:01:38webguest51yes, 44100
16:02:02Coldtoastthe Zaurus runs Linux, so you can use Kismet on that
16:02:22Coldtoastlooked and looked tho and it seems there's nothing for WM
16:02:32Bgerwebguest51 is there any sound ?
16:02:54thegeeknever midn
16:02:58webguest51yes thjust a second
16:03:27webguest51Bger, sound plauback is normal
16:03:30thegeekmake a thread there Coldtoast
16:03:31thegeekand ask;)
16:04:21webguest51the procgress bar fills up within a few seconds of a new song being loaded
16:05:23Coldtoastyou could get picky and say "Well, yeah. they are the same speed; 624MHz. Just not the same performance"
16:08:13webguest51the wps track elapsed time just continues counting up from songs beggining until the song stops, weird
16:08:38Coldtoastwill the final rockbox look pretty much the same as it does now?
16:08:54Coldtoastas far as aesthetics goes
16:08:54BgerSlasheri are you here ?
16:08:56B4gderwe will never be done
16:09:04Coldtoastwell, for-mass-consumption. for iriver
16:09:19B4gderit won't be final either
16:09:27SlasheriBger: yes
16:09:27B4gderthere won't be any final one
16:09:36B4gderjust an endless series of versions
16:09:37Bgersee webguest51's problem
16:09:37Slasheriwebguest51: the progress bar problem should be fixed on latest bleeding edge builds
16:09:41Coldtoastbut the version for iriver finally judged to be layman-friendly
16:10:03webguest51ah I see, it displays hundreths of a second in the track elapsed time
16:10:06B4gderI guess our first public release will look pretty much like this
16:10:12webguest51Thanks Salsheri
16:10:15Coldtoasthm. ok
16:10:18Slasheriwebguest51: yes, that's fixed
16:10:44Slasheribetter crossfade fix coming soon
16:10:45ColdtoastI quite like the layout iriver came with in their firmware
16:10:46***Alert Mode OFF
16:10:58B4gderColdtoast: we are not iriver you know
16:11:05ColdtoastI realise this
16:11:14B4gderif someone improves our look, we will have an improved look
16:11:23B4gderif no one does it, it remains as it is
16:11:32B4gderopen source
16:12:53B4gderpeople like to think of a "Rockbox Team" that makes rockbox
16:13:02B4gderwe are not a team
16:13:12B4gderwe are everyone who likes to contribute
16:13:58Coldtoastbut you all communicate with eachother? discuss things, work out what needs improving/fixing/implementing?
16:14:08B4gdernot all of us
16:14:14B4gdermany people just post patches
16:14:18Coldtoastbe it via the web or actually talkign to eachother
16:14:30webguest51Slasheri, the progress bar still fills very quickly
16:14:44webguest51with latest build
16:14:53Slasheriwebguest51: Hmm.. What do you mean that it fills quickly?
16:14:57 Join courtc [0] (
16:15:48webguest51select new song, starts to play, the progress bar then fills within 5 seconds
16:16:13preglowwhat layout iriver came with? it's not much of a layout
16:16:19Slasheriinteresting.. i can't reproduce that :/
16:16:35webguest51mp3 lame aps
16:16:43preglowvery interesting
16:16:47preglowwhat about posting the file?
16:16:52Coldtoastdoesn't matter if it's "not much of a layout"
16:17:00Coldtoastsimple and good
16:17:08Coldtoastvery clear and mildly attractive
16:17:17webguest51er you want me to send you the mp3 ?
16:17:43preglowdon't look at me, heh
16:17:44preglowig totta go
16:17:50preglowi gotta go <-
16:17:53Slasheriwebguest51: please do that :)
16:18:45webguest51it does that with all the mp3 files that i have tries so far
16:19:02preglowColdtoast: personally, i hate the file menu, the font is bloody humongous, the playing screen just gives a moderate amount of ingo
16:19:06preglowinfo, even
16:19:12SlasheriHmm.. What the elapsed displays after it fills?
16:19:18preglowmy biggest annoyance overall is the huge font
16:19:26Slasheriwebguest51: are you sure you updated rockbox.iriver file also?
16:19:54webguest51it continues counting up, displaying 23:34;345
16:20:13Slasheriok, that problem really should be gone with bleeding edge builds :)
16:20:38webguest51right let me delete everything and reload
16:20:55B4gdernow where are all the wanna-be h300 hackers to fix all the keypad compile problems?
16:21:15webguest51strangely it doesnt do that with the 18ths daily
16:21:57Bgeri could try ... tomorrow...
16:22:04B4gdermost of the h300 build problems are key-related
16:22:07webguest51Slasheri, its normal with 18th's daily build
16:22:34Coldtoastpreglow: yeah. I definitely agree the font in the file tree is HORRIBLE. Especially after using Rockbox
16:22:49ColdtoastI like their VU meter and the way the rest of it fits around it
16:23:09B4gderyou _like_ that weird thing?
16:23:18Coldtoastyeah. Because it has curves
16:23:25preglowoh well
16:23:29preglowdo make a iriver wps one time
16:24:10preglowwps code will probably see a lot of updating once someone with the will to code a fancy gui comes around
16:24:16thegeekhas t0mas fixed the bug with the images in wps?
16:24:22HCli want to change browsing too
16:24:28amiconnHmrf, I don't believe this :-/
16:24:32HClwith two browse modes, normal, and alphabetically
16:24:37preglowamiconn: waht
16:24:47amiconnThe player backlight is not pwm dimmable yet it is a LED backlight
16:24:53HCllike, left jumps from A to B, down just moves down
16:24:55preglowHCl: two sorting modes, you mean?
16:25:05HCllike quick skipping to a letter in the alphabet
16:25:24HClpossibly with support for not just the first lettre
16:25:32amiconnMy code is correct; I tried dimming the red led instead which works perfectly
16:26:00preglowamiconn: sounds like the output pin goes through some weird circuitry
16:26:09amiconnBut the backlight is just on with some irregulat short off periods when I try to dim it
16:26:45amiconnUnfortunately the player schematics are incomplete and don't show the backlight hookup
16:26:56amiconnSo no archos is dimmable :(
16:27:02preglowit's got to be that
16:27:12preglowi've never seen a led that can't be pwm dimmed
16:27:15BgerB4gder in fact these problems should be solved by someone with CVS access ...
16:27:19preglownor is there a reason
16:27:30preglowbut i gotta go
16:27:34 Part preglow
16:27:46Bgerso i'm not the right person
16:27:56B4gderBger: I can hand out cvs commit access for it
16:28:14linuxstbBger: No excuse now :-)
16:28:27BgerB4gder hehe ok
16:29:05Bgerbut i guess someone will do it before me
16:29:25 Join aegray [0] (
16:30:12Bgerbye 2 all, nice evening
16:30:40 Quit tvelocity ("Leaving")
16:31:40Slasheribtw, i have noticed that divisions are really slow on coldfire so bit shifting should be used whenever possible
16:33:54HClbleh i'm bored.
16:34:07amiconnSlasheri: If you're dividing by a constant and this constant is a power of 2, gcc will take this optimisation automatically
16:34:16CtcpIgnored 2 channel CTCP requests in 1 second at the last flood
16:34:16*HCl goes to nap..
16:34:16Slasheriah, good
16:34:48webguest51I have installed a bedge build from the 18th nd the progress bar is normal as is the song elapsed time
16:35:55webguest51Slasheri, I have installed a Bledge build from the 18th and the progress bar is normal as is the songs elapsed time
16:36:12webguest51my apologies for the double post, damn lag
16:36:44Slasheriwebguest51: Hmm.. If you download the lastest build (2005-06-20 14:28
16:37:00Slasherido you still experience problems?
16:37:06webguest51I have twice already :)
16:37:17Slasheriyou can't have downloaded that build ;)
16:37:18webguest51with the problems as described
16:37:21Slasheriit's still compiling
16:37:32webguest51what time zone
16:37:49Slasherithe one displayed on the web page
16:38:14B4gderthe web page shows UTC/GMT time
16:38:31webguest51I'll try it again :)
16:38:37Slasherior if even it doesn't work, i think that the build system is broken
16:38:48B4gderI doubt that
16:38:56B4gderit does rm -rf * before all builds
16:39:23webguest51ok, right now everythings normal with the 18th's
16:40:42Slasheriand make sure you got the bleeding edge build, not daily build
16:43:08webguest51ok, bleeding edge build installed and wps perfect,
16:43:16Slasherigreat :)
16:43:29webguest51many thanks
16:43:38 Join DMJC-L [0] (
16:45:54webguest51just to be sure, it IS neccessary to reboot the player with new build files ?
16:46:21Slasheriyes it is
16:46:41Slasherior you should manually reload the rockbox.iriver
16:46:42*webguest51 blushes
16:47:13webguest51I'll buy the first round
16:58:00 Part webguest51
16:58:32 Quit hicks (Remote closed the connection)
16:58:46 Join hicks [0] (~hicks@e64b57abb06ddb9d.session.tor)
17:02:49 Join mikenz [0] (
17:02:57mikenzquick question
17:03:05mikenzanyone here using rockbox on iriver?
17:03:27mikenzwell I have it now runnig sweet but need a couple of pointers
17:03:56HClits prolly a good idea to read the manual
17:04:14mikenzhow to make a "on-the-fly" playlist, all I seem to manage is create playlists of a whole dir.
17:04:38mikenzok, the manual would be cool.. is that in the documents section?
17:05:00Coldtoastin the file tree, depress and hold the joystick button
17:05:10ColdtoastI asked the same thing a few hours ago
17:05:24mikenzcheers cold toast
17:05:39mikenzI had seen it when first used rockbox and forgot how I did it
17:05:55*Coldtoast hands cheers to whoever tol dhim
17:06:22mikenzit's probably 'in the manual' but I am too lazy to read those thigns
17:07:31mikenzcya thanks
17:07:36 Part mikenz
17:14:26 Quit Coldtoast (Read error: 131 (Connection reset by peer))
17:25:53 Join spiralout [0] (
17:28:19 Quit DaKi][er ()
17:29:56 Quit ashridah (Read error: 110 (Connection timed out))
17:30:00 Quit B4gder ("Lämnar")
17:30:47 Join ashridah [0] (
17:31:13t0masI've been in a server room the past 2 hours...
17:31:23t0masthen opened the door... and walked to my bike
17:31:36t0masbefore I reached my bike I was melted...
17:32:17*t0mas reserves the server room for tomorrow...
17:32:29t0masif there's nothing to do... I'll just make something todo there
17:39:49Slasherit0mas: hehe :)
17:40:52t0masdamn... how stupid can you be...
17:41:23t0massome manager: "Can't we buy new dell pc's? They've send me a great email offer... for 499 euro's per pc"
17:41:34***Saving seen data "./dancer.seen"
17:41:43t0mas(that's spam... and dell won't sell them for < 600 euros... or you have to drop warranty and get xp home)
17:42:51t0masreply from my boss: (cc to whole IT dep.) "We use those pc's downstairs as terminals for the unix system, they just run telnet... and don't need any fancy hardware. Plus you don't get those dells for < 600 euros."
17:43:42*t0mas didn't dare to email that... I'm just a simple it helper... lucky to have this job... not having to work in shop or something
17:44:13 Join courtc_ [0] (
17:44:36 Quit courtc (Read error: 110 (Connection timed out))
17:44:54 Nick courtc_ is now known as courtc (
17:44:58t0mashm... some people really don't get it... P4's as a telnet terminal.....
17:45:47spiralouthi is it possible to copy or move files from one folder to an other?
17:46:06t0masspiralout: yes
17:46:13t0masjust type the full path
17:46:29bill20r3my H340 arrives today.
17:46:31*bill20r3 waits
17:46:43spiralouttype the full path where?
17:46:57spiraloutah rename...
17:46:59t0masspiralout: when you chose to move or copy a file, it asks for a new filename
17:47:05t0masrename = move
17:47:12spiraloutand copy?
17:47:29spiraloutah yeah sry
17:47:43spiralouttry it
17:49:55 Quit crwl (
17:50:34ashridahaah, duct tape, is there anything it can't do?
17:50:45t0masyes there is
17:50:48t0mascool you down
17:51:05ashridaht0mas: not so. cover mouth and nose, suffocate, then start cooling
17:51:30ashridahor, use it to tape over a vent that's giving out heat
17:51:40*t0mas tapes his pc
17:51:50t0mastemp alert :P
17:52:02t0mastape the speaker too?
17:52:58spiralouthmm doesnt work for me...:´(
17:54:25spiraloutrename failed
17:54:40t0masdon't use \
17:54:46t0masit's linux like... so use /
17:55:40spiraloutyes i know there isn´t \ there is only / so i have to use /
17:55:42 Join pabs [0] (
17:58:14spiraloutcan´t get it ...everytime i try it "rename failed...
18:00:40spiraloutwhen i write .cfg file it works
18:01:02NJoincrwl [0] (
18:05:56spiraloutok when i write the config file the path is /.rockbox/config00.cfg press enter and it works but when i remove the / with an selftyped / than it failed so maybe not the right / ?? could it be? can´t find another
18:13:44spiraloutstill keep trying
18:18:12spiraloutok forget this with the / that seems not to be the problem the problem is write/save .cfg files work ...rename .mp3 don´t work don´t know why
18:30:21 Quit spiralout (Read error: 104 (Connection reset by peer))
18:30:59 Join Chamois [0] (
18:31:00 Join spiralout [0] (
18:34:15 Join Coldtoast [0] (
18:34:42Coldtoastso somebody found why inverse mode had non-inversed sections in it?
18:34:54Coldtoastjust updated adn it's fixed
18:38:27 Join N-Mi [0] (~nicolas@N-Mi.staff.calvix)
18:38:35N-Mi'lo there
18:39:43 Join TCK- [0] (
18:41:57amiconnSlasheri: There are 2 playlist bugs, possibly related:
18:43:01amiconn(1) When a playlist is playing, you enter the playlist viewer and change track from there, the playlist viewer bails out because it thinks the playlist has ended
18:44:26amiconn(2) When a playlist ends normally, it is not correctly marked as such, so you can still press PLAY and get the resume request. Of course answering 'yes' leads to and immediately end again, but this can be repeated over and over
18:44:48amiconnIt should not react on PLAY at all in this state
18:45:37Slasheriamiconn: Hmm, i can't reproduce the first problem
18:46:10amiconnPlay some music, enter the pl viwer and press select on a different track than is currently playing
18:46:20amiconn-> bails out
18:46:20Slasheriah, trying that
18:46:28Slasherii just removed some tracks
18:46:35amiconnIt is supposed to stay in the viewer
18:46:47amiconnIt does on archos ;)
18:46:49Slasherihmm, it stays on the viewer..
18:46:55Slasheristill can't reproduce it :D
18:47:09Slasheriokay, now i found the problem
18:48:21Slasherihmm.. when i select a new song from playlist, wps calls audio stop
18:48:25Slasherithat is a problem
18:48:27amiconnAbout to commit configurable backlight dimming (just squashed the last bug)
18:48:35Slasherigreat :)
18:49:00bill20r3with the nifty fade-out?
18:49:00Slasheriamiconn: if possible, try add cpu_boost(true) while the backlight is being dimmed
18:49:02amiconnThis also has backlight_allow_timer(bool on) in preparation for plugin timer
18:50:10amiconnSlasheri: While in the pl viewer, wps code shouldn't be able to send anything, 'cause it isn't executed
18:50:25Slasheriamiconn: Hmm, interesting..
18:50:29Slasherii try again
18:50:57Slasherii got audio_stop and immediately after that audio_play
18:51:03amiconnIf you get audio_stop, this must either come from the pl viewer itself, or indeed from wps, but only because the pl viewer is exited before
18:51:11Slasherii can't do much about that if wps really calls audio_stop
18:51:15amiconn...which shouldn't happen, and doesn't on archos
18:52:11Slasheriyep, audio_stop -> playlist viewer think it's end of playlist and exits..
18:52:44 Quit gromit` (Read error: 60 (Operation timed out))
18:54:57amiconnHmm, it's the playlist viewer itself who calls audio_stop() when selecting a new track. Still this doesn't make it leave on archos....
18:54:59 Join Philip_0729 [0] (
18:56:30SlasheriHmm, maybe the audio_stop should be removed from playlist viewer when compiled for iriver target
18:56:40Slasherithat should fix the problem
18:57:41amiconnHmm, I don't know whether this is the best idea. There must be a reason why it works on archos, perhaps the playback message handling.
18:57:52 Quit TCK (Read error: 110 (Connection timed out))
18:58:13amiconnImho we shouldn't differentiate too much between targets for no reason
18:59:37Harpycould someone send me a ihp_120.hex with bootloader v2?
19:00:01Coldtoastmake it
19:00:04bobTHCall is on the site
19:00:23Coldtoastthere's a nahdy dandy tool to berge the bootloader
19:00:30SlasheriHarpy: i have one with some graphics mods to iriver firmware
19:00:30Coldtoasterrr. nahdy=handy
19:00:31Harpyi guess i'll have to..
19:00:48spiraloutwould it be possible or is there any kind of plugin which can copy all files in a playlist to a defined folder...?
19:00:52HarpySlasheri, what's that?
19:01:12 Join Stryke` [0] (
19:01:27Slasheri1.65E with modified graphics and rockbox bootloader. But maybe you don't want better
19:01:30bobTHCHarpy >
19:01:37Slasheribetter if you build one yourself
19:02:17Harpythere were some missing dependancies, so i couldn't compile the tools from the cvs.. i didn't try to fix them, though
19:02:33Harpyand the wine trick didn't work
19:02:46Slasherihmm, what was the error message?
19:03:08Harpysomething about not being able to write the file
19:03:33Slasheriplease paste the error
19:03:47Harpyok, i'll try it again
19:04:57 Quit ashridah ("Leaving")
19:05:43spiraloutwould it be hard to code those plugin...? or is there allready a way to do that
19:07:12Harpyfwpatcher.exe pops a error dialog that says: Couldn't modify existing file. Check if file is write protected, then try again.
19:07:59Coldtoastthat's what I use
19:08:55amiconnspiralout: There is no copy plugin yet
19:09:11amiconnThis would indeed be great, especially on the ondio
19:09:29amiconnThe functions itself aren't hard, but the user interface needs some thoughts
19:10:22spiraloutmhh i think open with...and then only the destination folder
19:13:24 Quit bobTHC ("Smoke Weed Every Day !")
19:13:29 Quit Philip_0729 ()
19:16:48HarpyColdtoast, the md5 doesn't match with the one on the site. this is v2, right?
19:18:05Coldtoastas far as i know
19:18:14Coldtoastfeel free to not use it tho, obviously
19:19:10Coldtoastwhat's teh m45 you have?
19:20:17Coldtoastactually, that's probably yhe firmware I did on Windows using fwpatcher
19:24:58 Quit RRRRRED_M_INFIN ("( :: NoNameScript 3.81 :: )")
19:25:45 Join gromit` [0] (~gromit`
19:28:31 Join TCK [0] (
19:29:29 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:31:39 Join ep0ch [0] (
19:32:36ep0chSlasheri, good job on crossfading :)
19:33:30 Quit Stryke` (Read error: 60 (Operation timed out))
19:33:47ep0chyou know about the slight pop after fade in?
19:34:27Slasheriep0ch: yes i know that (currently i don't use fade in), but that will be fixed
19:34:55Slasheriamiconn: the backlight settings works great :)
19:35:01ep0chgood stuff, can i ask for a possible feature too? :)
19:35:14Slasheriof course you can ;)
19:35:52Slasheriamiconn: don't you mind if i put few cpu_boosts to backlight.c?
19:36:21amiconnbacklight fading doesn't draw that much cpu power
19:36:24ep0chSlasheri: an option to only crossfade when previous/next is selected, and not when a track automatically finishes (if that makes sense)
19:36:29Slasherithat should fix the flickering completely when pcm tries to change cpu frequency
19:36:45Slasheriamiconn: yep, but frequency switching still causes undesired effects..
19:37:05amiconnHmm. I can't say that I did notice flickering
19:37:25Slasheriamiconn: try enable crossfading (that keeps pcm watermark much higher)
19:37:32Slasherithen you will notice flickering
19:38:03Slasheriep0ch: hmm, that could be possible
19:38:25Slasheribut crossfading doesn't work always even when you change tracks manually
19:38:44ep0chwhy is that?
19:39:34Slashericrossfade can work only when pcm buffer is almost full and next track is not loaded from disk
19:41:13Coldtoastthe new crossfading settings are perfect
19:41:35***Saving seen data "./dancer.seen"
19:42:44ColdtoastHarpy: according to the rockbox site, fwpatcher is V2
19:43:26ep0chideally (for me), the crossfading should not kick in if the tracks are contiguous (i.e. from the same album and played in order)
19:43:30ep0chbut that is probably a bit OTT
19:44:11amiconnSlasheri: That explains it. I never enabled crossfade so far
19:46:44Slasheriamiconn :) so what do you think about the cpu_boosting? that should guarantee flicker-free dimming
19:46:55Coldtoastany chance of having FWD/RWD play the audio as it seeks along?
19:47:16 Quit TCK- (Read error: 110 (Connection timed out))
19:47:36SlasheriColdtoast: you mean that rb plays some samples from audio while it seeks?
19:48:02Slasheriok, that requires some re-design to wps and other code
19:48:21Slasherinot so easy to do
19:48:21Coldtoastjust cos that's what I'm used to with most devices I've ever had
19:48:33amiconnIt'd require a ton of redesign
19:49:26HarpyColdtoast, hmm.. maybe the md5s on the site are not up to date with v2 then
19:50:27 Join |ep0ch [0] (
19:50:34 Join [1]ep0ch [0] (
19:50:54Coldtoastwhen you DO seek and the position bar fills up, it's sort of odd that when you get to the point you want then release, the bar empties, fills then plays
19:52:13 Quit ep0ch (Read error: 145 (Connection timed out))
19:52:13 Nick [1]ep0ch is now known as ep0ch (
19:52:38Coldtoastis that cos when you seek, it doesn't move to the position as you're actually doing it?
19:53:28 Join Zoom2 [0] (
19:54:55|ep0chsomething odd, changing volume when paused causes ticking noises to be heard, but only when volume < 95
19:55:52Coldtoastwow! it does too
19:56:02Coldtoastfrom about 90 to 50
19:56:41|ep0chand you sometimes get wierd quiet short waveforms played back too at some volumes
19:56:52ColdtoastI start JUST hearing it around 42 then it stops at 91
19:56:57 Join west-acre [0] (
19:57:02|ep0ch95 for me
19:57:22west-acrehey hey. wots bin happnin s
19:57:30west-acreover the last few days?
19:57:33Coldtoastactually, stops at 89 for me
19:57:44west-acreany big steps int he H1XX port?
19:57:53ColdtoastI have mine plugged into an amp and pumped up REALLY loud to see where the noise starts
19:58:01|ep0chpoor speakers
19:58:17Coldtoastnah. I have decent speakers
19:58:28|ep0chstill, speakers probably won't like it
19:58:29ColdtoastLogitech z-560
19:59:24Coldtoastwon't like what?
19:59:59|ep0chrandom impulses
20:00:30|ep0chlike playing pure noise through a speaker, not good
20:00:34Coldtoastit's not random tho. the only noise I get is the ticking
20:01:01Coldtoastjust changed to the iriver firmware and there's no ticking at any volume
20:04:55 Join Stryke` [0] (
20:06:23 Join tvelocity [0] (
20:08:46 Quit Zoom2 ("CGI:IRC (EOF)")
20:13:24Coldtoastwhen in Paused mode, can you make the progress bar flash or something?
20:14:54 Join Zoom2 [0] (
20:15:13|ep0chanyway i'll leave you all in peace
20:15:21 Quit |ep0ch (" HydraIRC -> <- IRC has never been so good")
20:15:21 Quit ep0ch (" Want to be different? HydraIRC -> <-")
20:24:22west-acrehey. the fading backlight in rockbox, how is it configured?
20:28:44Slasheriwest-acre: general settings -> display -> lcd settings
20:29:02west-acreis it caption backlight?
20:29:14Slasherino, backlight fade in & out
20:29:26west-acrei had to firmware upgrade.
20:29:50Slasheritry the bleeding edge builds
20:30:07west-acre10 second fade out awesome
20:30:10west-acreyeh i got that
20:30:23west-acrebut i mean firmware upgrade in the iriver firmware. is this necessary?
20:30:45west-acrei thought not.
20:31:03west-acrewot is the deal when upgrading the firmware to bleeding edge?
20:31:37Zoom2you get the latest bugfixs
20:31:48west-acrerite. url?
20:31:49Bagderand the latest bugs too! ;-)
20:32:00Zoom2heh yeah
20:32:24Zoom2it basically keeps everyone on the same page, bugwize, plus additinal features
20:32:43west-acrewhere are the latest bugfixs?
20:32:54Coldtoastwith .wps files, it'd be nice if you could have things like bold, italics, maybe even inverse
20:33:00Slasheriwest-acre: you dont need to update the iriver firmware, just the .rockbox -directory and rockbox.iriver file from root
20:33:09Zoom2there is a log on the front page that displays the recent cvs changes
20:33:11BagderColdtoast: yes it would be nice, send us a patch! ;-)
20:33:33west-acreyeh that'd be awesome with the .wps
20:33:33Coldtoastwould if I could
20:33:46Zoom2Slasheri: did you get a chance to check out the sleep timer audio fader?
20:34:11SlasheriZoom2: not yet but i will look that
20:34:33Zoom2ok thanks
20:35:09Slasherihow long fade out do you prefer, something pretty long like a minute?
20:36:01Zoom2that would be good. since it is a sleep timer i think a long fade would do the trick
20:36:22Zoom2it will help doze off with the music
20:37:06west-acrehey. wot about the comment tag? can it scroll accross the screen or isnt it supported?
20:37:11Slasheriok, hmm. Maybe good configurable duration could be between 30s to a few minutes
20:37:12west-acreA line beginning with #rockbox is a comment/
20:37:20 Nick Lynx_ is now known as Lynx_awy (
20:38:23Zoom2sounds good to me
20:38:29Slasherigreat :)
20:39:04Zoom2thanks again
20:39:42Coldtoastknow what would be cool? if the sleep timer would ladd tracks randomly from progressively quieter music. hehe
20:40:10Coldtoaststart out with death metal, end with Enya :)
20:40:29amiconnBagder: I already thought about font attributes a bit. For the scrolling lines, this would basically require some kind of atrribute memory, storing the attributes of each letter
20:41:14Zoom2coldtoast, that or a volume fader would be nice - but seems to be quite technical
20:41:30ColdtoastI'm joking
20:41:36Coldtoastit'd be funny tho
20:41:36Zoom2i know =)
20:41:44Bagderamiconn: attributes? won't it just be font?
20:41:45Zoom2thats basically what i do in my playlist
20:42:10amiconnBagder: Hmm? I mean bold, italic, underline, inverse, colour...
20:42:28amiconnDifferent fonts are even harder, since they might have different heights
20:42:38BagderI meant we'd support those by letting them be different fonts
20:42:53Bagderbut perhaps that's not too smart
20:43:30 Join [IDC]Dragon [0] (
20:44:16Coldtoasthey! it'd be really awesome if you could create ASCII art on the screen
20:44:34amiconnOoh, rare visitor
20:44:38amiconnHi jrg :-)
20:44:42amiconnJrg even
20:44:48[IDC]Dragonhey, somebody remembers me
20:44:56Bagderwho are you?
20:45:07*[IDC]Dragon doesn't know
20:45:17amiconnBagder: Different fonts would still create the problem that we'd need to remember them per lett.
20:45:32 Part aegray ("Leaving")
20:45:47amiconnbold, italic, underline can all be created algorithmically
20:46:03*[IDC]Dragon updated from cvs, everything's so different...
20:46:28Bagderamiconn: I know, I've just been thinking that it would be smarter to just generate new fonts setups for them
20:46:39Bagderseparate font files
20:46:53Bagderthen italic is no different than another font
20:47:01amiconnWhy? Imho this would waste memory
20:47:13Bagderhow come?
20:47:21amiconn..and remembering different fonts per letter creates more problems
20:47:22Bagderyou mean you'd only generate the used letters?
20:47:36amiconnNo, I mean they're generated on the fly while blitting
20:47:44 Join asdsd____ [0] (
20:47:48 Part asdsd____
20:47:55Bagderwell, we still want multi-font support
20:47:58amiconnItalic is certainly the hardest
20:48:28[IDC]Dragontypographically speaking, italic is a different font, not just skewed
20:48:37amiconnMulti-font doesn't go well with the concept of scrolling lines
20:48:41amiconn[IDC]Dragon: I know
20:49:03amiconnStill this is what AmigaOS uses for bitmapped fonts, and it's looking fairly good
20:49:49Coldtoastwhat about basic line formatting? Left align, Center, Right align?
20:50:29amiconnLeft align is already implemented ;)
20:51:12 Join joshn_454 [0] (~joshn@
20:51:22 Join Audiophil [0] (
20:51:46*amiconn is nitpicking
20:51:56Audiophildoes the current release have speed issues?
20:52:10Audiophilskipping to the next song is rather slow
20:53:40joshn_454Slasheri, are you miipekk on cvs commits?
20:54:50amiconnSlasheri: The #ifdefs shouldn't be necessary if we define cpu_boost() as an empty macro on fixed frequency platforms
20:55:06amiconnThat would help to reduce #if clutter
20:55:17Bagdergetting rid of ifdefs is a Good Thing
20:55:45amiconn#define cpu_boost(on)
20:55:48amiconnin system.h
20:55:52joshn_454also, the last #ifdef commit switched the cpu_boost parm around
20:55:54*[IDC]Dragon tries to understand autoconf.h
20:55:58joshn_454from false to true
20:56:10[IDC]Dragonin the case of bootloader
20:56:45Bagder[IDC]Dragon: what's the problem?
20:57:00amiconnjoshn_454: That's definitely not good
20:57:04Chamoisthre is a problem in the last cvs commit
20:57:07Chamoisoh sorry
20:57:18[IDC]DragonI'm picking up my old subject of "Bootbox", a minimalistic Rockbox
20:57:51[IDC]DragonI've been bending the configure script to make this the bootloader for Archos models
20:58:08 Join cooleo [0] (
20:58:34[IDC]Dragonconfig.h expects to find "autoconf.h"
20:59:00Bagderautoconf.h is a corner stone of the build process now
20:59:34[IDC]Dragonso why doesn't it look into the build dir?
20:59:54Bagderwhich doesn't?
21:00:02cooleoi've got a hardware question about screens, does anyone know about screens and their power sources?
21:00:16[IDC]Dragoninstead, it's sufficient if I copy it into my newly apps dir
21:00:38cooleofair enough
21:00:57Bagder[IDC]Dragon: you should have -I$(BUILDDIR) in INCLUDES
21:01:04Bagderin the Makefile
21:01:08[IDC]Dragoncooleo: wrong group
21:01:15[IDC]DragonBagder: thanks
21:01:24cooleowhere should I ask Dragon?
21:01:25Coldtoastperhaps you should wait a bit longer for an answer cooleo?
21:02:04cooleoi'm a complete newbie here
21:02:39cooleoi'm working on modifying my iRiver H300
21:02:52Coldtoastwhat are you doing to it?
21:03:10 Join yyz [0] (
21:03:17 Join TCK- [0] (
21:03:18cooleoi need to tap off the power source from the screen but I'm not sure where to tap in
21:03:40cooleoim trying to tap into the battery through the screens connection
21:03:49Coldtoastwhat are you adding?
21:04:05yyzanyone tell me how I am supposed to be able to run rockboy, after downloading the prebuilt rockbox with it + greyscale patch? I dont see it under plugin menu/
21:04:06cooleothis is kinda embarrasing
21:04:09cooleobut LEDs
21:04:15Coldtoastdoesn't sound like such a good idea to me, but hey
21:04:20cooleoheheh i know
21:04:32Coldtoastis there much room in there for resistors?
21:04:34cooleobut i don't want them on 24/7
21:04:44cooleoyeh i could get a resistor in
21:05:27crwlyyz, just browse to a .gb file in the filebrowser
21:05:41yyzmust they be in root dir?
21:05:49yyzi tried that, but in a subfolder and didnt see em
21:05:56cooleoas you've probably assumed its powered and controlled by one of those orange flexy boards
21:06:04amiconnSlasheri: Buuug! When you plug in usb while the music is playing, it correctly detects usb, stops playing and accepts usb. But after leaving usb, rockbox starts loading from disk like mad...
21:06:05crwlyyz, go to preferences and set the filebrowser to show all supported files
21:06:14yyzthatll be it
21:06:17yyzthanks crwl
21:06:41Coldtoastcooleo: just a suggestion, but onc eyou completely screw your player, you should donate it to Rockbox
21:06:48cooleoits got very fine connections on it
21:06:55cooleohehe I probably would
21:06:58yyzyep, works a treat
21:07:01yyzthanks again :)
21:07:08cooleoim aiming not to do that though! :D
21:07:22Coldtoastaim low
21:07:32Coldtoastand a bit in front
21:07:37cooleoanyway, it would only break the screen
21:07:40cooleoor battery
21:07:57cooleoand they cost nothing compared to the rest of the stuff in there
21:08:10cooleoi'll give you guys whatever breaks :P
21:08:36Coldtoastis the screen AC or DC?
21:08:44cooleoi couldnt tell you
21:08:46Zoom2cooleo, Linus might know
21:08:51Zoom2but he comes later in the day
21:08:52cooleoyeh i know hes got one
21:08:52Slasheriamiconn: interesting, investigating that :)
21:09:04cooleohow late does he normally comeon
21:09:17Slasherijoshn_454: almost, Miika on cvs :)
21:09:20yyzeww. locked up :( *goes to find paper clip*
21:09:25Zoom2depends where you are, this is a woldwide chat
21:09:30Slasheriyyz: what did you do?
21:09:35amiconnSlasheri: I'll commit a fix for the mixed up cpu_boost
21:09:35cooleoim in UK so its 8 here
21:09:45cooleoim guessing Linus isnt!
21:09:47Zoom2EST i think he comes on at like 2-3 AM
21:09:54Slasheriamiconn: good
21:09:56cooleohmmm ok
21:09:59yyztrying to load a rom
21:10:08yyzprobably just one that aint compatible
21:10:12Slasherithere are plenty those #ifdefs in other sources too
21:10:26cooleoshouldnt that be soon then?
21:10:37cooleoseeing as EST = GMT -5
21:10:38Zoom2cooleo you can also try to submit a post in the forums
21:10:42Zoom2he reads those
21:10:46Zoom2no it will be a while still
21:10:48Zoom212 hours or so
21:10:50cooleogood point Zoom
21:11:14Slasheriyyz: ah :)
21:11:15cooleowhat time is it EST now?
21:11:22Zoom23:11 PM
21:11:27Slasherifortunately that has nothing to do with music playback ;)
21:11:55amiconnSlasheri: There are? Hmm, they could be removed now
21:11:55crwlor maybe yyz was playing music while trying to load a rom :)
21:11:59yyzoh, is playback so good that a lockup should suprise you?
21:12:05yyznah i wasnt
21:12:09cooleodamn he's on in the morning here then
21:12:15Slasheriamiconn: just do grep ADJUSTABLE firmware/* apps/*
21:12:24cooleoi'll try and catch him on thursday morn
21:12:40cooleothanks alot Zoom, I appreciate it
21:13:02yyzeww. you forget how shitty they feel out of iskins
21:13:49amiconnSlasheri: Another bug: The time and track display in wps seems to stall around a track change
21:14:28Slasheriamiconn: yes, that's more a known issue than a bug.. Will be fixed if it's possible
21:15:22amiconnMeh, I should work on asm memset() now
21:16:15amiconnm68k abi is cumbersome. I'm practically required to push registers, soo little scratch registers....
21:17:34Zoom2no problem cooleo
21:19:11Slasheriamiconn: fixed the first bug
21:21:00 Quit TCK (Read error: 110 (Connection timed out))
21:21:05 Join Andrew__ [0] (
21:22:45cooleook i posted it Zoom
21:22:57cooleoi'll still try and catch him on IRC later
21:23:02cooleopin him down
21:23:22cooleoyou can avoid Forums but you can't avoid IRC!! : D
21:23:32yyzanyway gotta go study, thank for help folks
21:23:40 Quit yyz ("later")
21:24:17 Quit cooleo ("CGI:IRC (EOF)")
21:24:19Zoom2heh yeah, i brb though gotta restart
21:25:03 Quit Zoom2 ("CGI:IRC (EOF)")
21:28:08 Quit Andrew179 (Read error: 145 (Connection timed out))
21:33:26 Quit rob- ("ircII2.8.2-EPIC3.004 --- Bloatware at its finest.")
21:36:49 Quit spiralout ("ChatZilla [Firefox 1.0.4/20050511]")
21:41:38***Saving seen data "./dancer.seen"
21:41:58 Quit Stryke` (Read error: 60 (Operation timed out))
21:46:27 Join webguest33 [0] (
21:46:29 Quit webguest33 (Remote closed the connection)
21:46:39 Join preglow [0] (
21:47:11preglowdoes musepack playback actually work without skipping?
21:47:43preglowsomeone obviously felt so and edited the wiki
21:53:45preglowbut it skips like mad here
21:55:04 Quit Audiophil (Read error: 145 (Connection timed out))
21:57:19 Join Psy-Dead [0] (
21:57:28 Nick Psy-Dead is now known as n0bby (
21:57:33n0bbyany devs in?
21:57:48n0bbywho did the lcd fade stuff?
21:58:10n0bbyi was wondering if fade-in could be added to the bootloader, and to the shutdown sequence
21:58:21n0bbyfade out for shutdown obviously :)
21:58:35preglowi really think we should keep stuff like that out of the bootloader
21:58:39preglowbut then that's me
21:58:48n0bbyyeah, i suppose it should be kept small
21:58:52preglowsmall and safe
21:58:53n0bbydidnt think of that
21:59:05n0bbya nice long fadeout at shutdown is needed though :)
21:59:18preglowthat should be more doable
21:59:27n0bbysame length as the audio fade
21:59:42n0bbythat reminds me, fade length needs to be adjustible
21:59:49n0bbyfor audio that is
21:59:59n0bbyor has it been added in the last week or so?
22:00:05n0bbyi just updated there now
22:00:15*n0bby plays around a bit more
22:01:03n0bbyi know thats a needless relic of the archos
22:01:06 Quit west-acre ("I-n-v-i-s-i-o-n 2.0 Build 3515")
22:01:43n0bbyahah, audio fade isnt adjustible yet
22:01:43preglowwhy, that's not needed by archos?
22:01:50preglowsounds more like a design decision to me
22:01:57n0bbyarchos shuts down too fast
22:02:01n0bbyhardware controlled poweroff
22:02:06preglownow really
22:02:13n0bbythey couldnt save fast enough
22:02:47 Quit RotAtoR ()
22:02:57n0bbyi gtg, be back soon. gonna leave mirc running so i can see what anyone says :)
22:03:10 Nick n0bby is now known as n0bby-brb (
22:03:24preglowman, the fade is pretty smooth
22:03:26pregloweven at 10 secsx
22:13:30bipakmay i make a feature request? :)
22:16:32Coldtoasthey. in .wps, how do you create an unbroken line?
22:16:51Coldtoastlike, _ but not at the base
22:17:06[IDC]Dragonbagder: what's the new "paradigma", build the stuff in a subdir?
22:17:12Coldtoast_ but located -
22:17:47[IDC]Dragonwhen to use $(OBJDIR), when $(BUILDDIR) ?
22:19:31amiconnOBJDIR is different for the different parts, it points to the appropriate sub-dir of BUILDDIR
22:19:31 Join asdsd____ [0] (
22:19:39 Part asdsd____
22:24:49[IDC]Dragonnow it at least builds, with the .ucl in the build dir, ajbrec.ajz in a subdir
22:25:04[IDC]Dragonis this what it should be like?
22:25:16*[IDC]Dragon should try a regular build
22:27:45amiconnajbrec.ajz and ro(ck|m)box.ucl should be in the build dir
22:27:48linuxstbpreglow: I've corrected the Musepack status in the WIki. It's not real-time for me either on any of my MPC files.
22:28:46 Nick n0bby-brb is now known as n0bby (
22:29:03amiconnPlugins end up in apps/plugins
22:29:36amiconnAll .o files are in various sub-dirs. rockbox.(elf|bin|map) are in apps/
22:29:49 Join Zoom2 [0] (
22:30:31*[IDC]Dragon figures
22:35:53n0bbyanyone know where linus is at with h300 and h110 bootloaders?
22:37:27preglowit's just barely realtime here
22:37:31preglowplays with small gaps
22:37:54preglowi'll let the musepack people have ample chance to fix their fixed point weirdness before i hack more on it
22:38:22preglowi can't do the prescaling trick, since they use variable shift sizes all the bloody time
22:38:43preglowfor now, every mul in musepack ends up as six-seven muls and some adds in the code
22:38:44[IDC]Dragonphew, bootbox builds again (makefiles are not my thing)
22:40:11n0bbyCVS/WIKI QUESTION!
22:40:17n0bbycvs login
22:40:17n0bbycvs -z3 co rockbox
22:40:25n0bbysay instead of haxx
22:40:37 Join RotAtoR [0] (
22:40:47n0bbythats not a wikipage, so a webdev needs to fix it if i'm right
22:40:51 Join Rori [0] (
22:41:29RoriHey guys. Was the bug whereby you set the treble but when you switch it back on the treble is back to 0 even though it says it's not fixed?
22:41:39n0bbyi believe so
22:41:52n0bbydownload a daily and test for yourself
22:42:01Coldtoastit's fixed
22:42:08Roriok. any idea about the playback suddenly stopping and dropping back to the main menu yet?
22:42:21ColdtoastI've not even seen that bug
22:42:25n0bbyme either
22:42:33Coldtoastso prolly fixed also
22:42:53n0bbyas i said, get a daily and report back if its not :)
22:43:02Coldtoastget a bleeding edge
22:43:07RoriI had is happen in 2 different ways the other day using a build from a couple of days ago. It stopped playback for 2 seconds then carried on. Then it stopped playback and went to the file menu then shut down the player
22:43:40Rorikinda odd
22:44:29 Quit Lomaxx ("Leaving Teramis")
22:45:53 Join Stryke` [0] (
22:46:44 Join ep0ch [0] (
22:48:15ep0chpreglow: which .rock/codec file is the mp3 codec compiled to?
22:49:09ep0chdon't suppose you have a version of it with the dithering disabled? :)
22:49:29preglowcorrect, i don't
22:49:42ep0chwould like to build me one?
22:49:47ep0ch^ you
22:49:51preglowno, not really
22:50:01n0bbypreglow, when compiling, whats the difference between "normal" and "devel" builds?
22:50:02ep0chfair enough
22:50:10n0bbysim and bootloader are obvious
22:50:14ep0chi'll get cygwin setup then...
22:50:18preglown0bby: devel has some logging options
22:50:30n0bbyany reason not to build it?
22:50:31preglown0bby: like the very useful Logf option
22:50:45RoriLame gapless playback works so seamless it's sexy
22:50:51n0bbyok, devel it is :)
22:50:56preglowep0ch: if you catch me while i've actually got my development environment going, chances are better, i'm doing something completely different now
22:51:30n0bby"enter developer options" do i want debug and logf or just logf?
22:51:35ep0chok cheers
22:51:38preglowwhy don't you want dithering anyway?
22:51:46preglowjust logf if fine
22:51:52preglowi believe DEBUGF is a sim thing
22:51:56ep0chwant to know a) how it sounds and b) performance
22:52:14preglowperformance is quite a lot better, and it sounds just the same to my ears
22:52:29ep0chmp3 atm uses about 50% boost
22:52:30amiconnDEBUGF() is for simulator (don't need to define DEBUG there) and for debug builds
22:52:36ep0chvorbis is much less
22:52:46amiconnSince there is no gdb stub for iriver, it's not usable there
22:53:02n0bbythanks :)
22:54:02n0bby seems to lack iriver info
22:54:08preglowep0ch: theoretically, mp3 should be faster than vorbis
22:54:16Roriis there a way to start rockbox from the remote?
22:54:17ep0chi know :(
22:54:58HClRori: just press play
22:55:01ChamoisRori : very short press on the remote play button
22:55:06ep0chanyway i want to get a development environment up, about time i learned some C
22:55:14preglowh, indeed
22:55:26ChamoisRori : very very short
22:55:40Rorihas anyone figured how the display on the remote lcd will look/work
22:55:54HClthe same way it works/looks now?
22:55:55Roriin regards to use
22:55:57HClwith maybe some changes.
22:56:13Roriall I see if Rockbox logo right now on it
22:56:28HClthen your firmware is malfunctioning.
22:56:44DBUGEnqueued KICK HCl
22:56:47HCli'm not quite awake :X
22:56:54HClthought you said main..
22:57:02CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:57:02*HCl shuts up and crawls to a corner
22:57:03Zoom2thats normal for the remote i thnk
22:57:12Zoom2HCL good morning =)
22:57:20HCli'm about to go back to sleep ;p
22:57:26Zoom2err 1
22:57:29Zoom2wow i messed that up
22:57:30Roriyeah what I mean is has anyone figured out how it will look on the remote lcd for file browsing etc
22:58:42n0bbybugger. error during compile
22:58:43Zoom2i think they will have a customizable wps
22:58:49Zoom2it should mimic the main lcd
22:58:57RoriI can't make use of the remote at present without something to look at and having the play on your hip is kinda well you catch my drift :)
22:59:15n0bbywhat is "Error 2" when compiling?
22:59:38Rorinot so fussed for in the car. I need a better mobile phone holder though ;)
23:00:06n0bbyi got error 1 in build/apps/codecs/dumb/itread.o
23:00:14n0bbythen [all] error2
23:00:16preglowgcc 3.3, yes?
23:00:30ep0chwhat do you guys use Perl for?
23:00:34preglowyou need 3.4 or dumb will vomit
23:00:37n0bbyi thought i'd used the one that the wiki told me to
23:00:40preglowep0ch: misc scripts and database generator
23:00:41Bagderep0ch: various scripts
23:00:45n0bbymust be (or have been) out of date
23:00:51preglown0bby: wiki is broken, i thought someone had fixed it
23:01:06RoriI wish more people would rip to Lame or Ogg. Too many rips out there in non-lame mp3 format :P
23:01:44n0bby" WARNING It is not clear whether these versions work correctly to build current rockbox code. You are strongly advised to build your own CrossCompiler instead.." :S
23:02:08preglowyes, that'd be the fix
23:02:15preglowcygwin doesn't like us these days
23:02:17n0bbyis there a command i can use in cygwin to check my GCC version?
23:02:24preglowgcc -v
23:02:43n0bby3.4.4 (cyming special)
23:02:55preglowwell ok
23:03:01preglowyou need to compile yourself, then
23:03:04 Quit Chamois (" HydraIRC -> <-")
23:03:12n0bbyright, more fun ;)
23:04:05n0bbyim changing the wiki to explicitly state that it'll vomit on dumb
23:04:09Bagdercodecmpc wins the size compo
23:04:09 Join Benacool [0] (
23:04:21preglowmusepack wins most compos
23:04:37preglowthe internal struct is humongous
23:04:57BagderI'm fiddling on moving codecs to load in a separate buffer
23:05:32preglowyou mean separate from audio buffer?
23:05:34preglowif so, why?
23:05:35Bagder0x3C000 bytes required atm
23:05:44Bagderseparate from the plugin buffer
23:06:37preglowmusepack is less than optimal in most ways i can think of
23:06:57preglowconceptually it's nice, but libmusepack is a bitch to port
23:07:49n0bbywiki fixed.
23:08:06ep0chwhat IDE do you guys use (if any)?
23:08:24n0bbyif i'm building my own cross-compiler, can i still use cygwin?
23:08:59Bagdersure you can
23:09:09n0bby"Experiments have shown that gcc-3.4.x for cygwin does not build binutils correctly! (at least not for m68k - not sure about other versions) For now, when building on cygwin, get/downgrade to gcc-3.3.3 first"
23:09:15n0bbyWHAT THE ***
23:09:26n0bbythis is hella annoying
23:10:58ep0chi look forward to that then...
23:11:34n0bbyep0ch, are you trying to compile too?
23:11:44ep0chjust started to install cygwin
23:11:56ep0chinstalled gcc-3.4.1
23:12:31n0bby3.4.4 is newest
23:12:52n0bbyand you need 3.3.3 to build binutils
23:14:33ep0chcorrection 3.4.4 is what i've installed
23:15:17ep0chso uninstall 3.4.4, install 3.3.3, build binutils, then uninstall 3.3.3 and install 3.4.4...?
23:16:00n0bby"gcc-3.4.4.tar.bz2" why isnt that tar.gz?
23:16:18preglowbecause tar.bz2 is smaller?
23:16:19Bagderbecause bz2 compresses a lot better
23:16:29n0bbydoes it make any difference to me?
23:16:36n0bbywhen trying to use it?
23:16:40Bagderyou download a smaller file ;-)
23:16:47n0bbyno further headaches?
23:16:53n0bbyi'm paranoid now
23:16:59n0bbyOPENSOURCE IS EVIL _
23:17:28Bagderand Rockbox is satan
23:17:37n0bbyquite possibly
23:17:46n0bbya rather deicious choclolate coated satan
23:18:06Zoom2... *is scared
23:18:30ep0chi wish shuffle mode was more accessible...
23:18:39n0bbydo i need to bother with "gdb-6.1.1.tar.gz" or is that for linus with his expensive debugger toy that he bought?
23:18:48HClthe latter.
23:19:13Bagderalmost at least
23:19:28n0bbyhow do i install/remove gcc?
23:20:20Bagderit explains how to build/install a cross compiler
23:21:39n0bbybut what about the fact i need to use 2 different compilers just to make a cross complier? >_>
23:21:54n0bbyi dont see the switch from 3.3.3 to 3.4.4 mentioned
23:21:58Bagderthat's cygwin craziness
23:22:26Bagderto encourage usage of real operating systems ;-P
23:22:59n0bbyi'll consider dualboot next time i reinstall windows
23:23:13n0bbyso for now i'm asking how to switch between 2 versions of the GCC
23:23:26Bagderinstall them in different paths
23:23:27ep0chgcc -v ?
23:23:31Bagderthen set the PATH differently
23:23:36ep0chgcc -V ?
23:23:47Bagdergcc −−version
23:25:21n0bbywhen i type that i get a "sustem shared memory mismatch dected" error and it kills my children
23:25:45n0bbyoh fsk this
23:25:51n0bbywhats a good linus distro?
23:26:04HClubuntu is marching upwards too.
23:26:09HClwhich is based on debian
23:26:13HClbut more aimed for desktop
23:26:23Stryke`(k)ubuntu is very useable
23:26:34n0bbythat'll do me then :)
23:26:51Stryke`kubuntu if you like KDE
23:27:06n0bbydoes it mind my using an ATI card? i know theres issues with propriatry drivers and debian peoples
23:29:20Seedn0bby: just be lucky you don't have the ATi stuff I'm getting.. No Linux distro works with it ;)
23:29:54Bagderperfect opportunity to start writing linux drivers!
23:29:57Seeda one month-old chipset for notebooks. barely any Windows support
23:30:03n0bbya radeon 9600XT will be ok then?
23:30:08*Bagder ducks
23:31:17 Quit hicks (Connection reset by peer)
23:31:23n0bbylinux scares me, the amount of circular depenancys, broken RPMs and driver issues i've had to deal with (usually i deal with them by going back to windows and throwing a hissyfit :P)
23:31:58Coldtoastmy x800 worked perfectly fine on Mandriva
23:32:36Coldtoastjust wish the card hadn't died
23:32:39ep0chmy compile just died on DUMB
23:33:01ep0chu star :)
23:34:02HCln0bby: wow, then you really got the wrong distros.
23:34:10HCllinux versions with rpm are a big no no
23:34:24HClyou need debian.
23:34:44 Join cooleo [0] (
23:34:59preglowyou need something taking care of dependecies for you, that's for sure
23:35:02preglowi hate rpm based system
23:35:04preglowcooleo: hi?
23:35:27Stryke`rpms are the worst
23:35:31Coldtoastthere are tools to help with dependencies tho with rpm
23:35:42cooleolast time i came in here i didnt get another hello for about 2 mins
23:35:42linuxstbAnyone know if Fedora Core 4 has the equivalent to "apt-get install ..." ?
23:35:46Coldtoastyum for example
23:36:11 Join hubbel [0] (
23:36:39 Join muesli- [0] (
23:37:17cooleono luck yet ColdToast
23:37:23Bagderbut no RPM system I know of can do the never-ending upgrade story of Debian
23:37:24ep0chpreglow: from looking at the cpu boost, i reckon without dithering its about 2.5 times quicker
23:37:25cooleoim still looking about though
23:37:49Stryke`hows the sound without dithering?
23:37:59ep0chcant tell the difference
23:38:13ep0chwouldn't mind grabbing the output and doing a diff
23:38:15Stryke`interesting, perhaps this should be the normal codecmpa then
23:38:16preglowep0ch: are you certain?
23:38:16ep0chor abx
23:38:35preglowi would never ever have expectet that big a boost
23:38:44ep0chlet me try somemore tests
23:39:17hubbelTried recording several hours today, just to see how long the battery would last.. The results was over 4.5 hours.. (44.1khz, stereo wav)..
23:39:44ep0chits definetly quicker, but i can't accuratly say by how much
23:39:49 Quit TCK- (Read error: 60 (Operation timed out))
23:40:03preglowhubbel: that's using the entire mp3 buffer for recording data?
23:40:29Suckaget an audiophile to see if they can tell the difference :D
23:40:44n0bbypost it on hydrogen audio then
23:40:48Stryke`do the other codecs dither?
23:40:50n0bbytheyre complete freaks
23:40:55hubbelpreglow: not really..
23:41:04Rori!anime karas
23:41:07hubbelpreglow: just 1.5 MB or something
23:41:09preglowwell, then the recording time can surely be extended
23:41:42***Saving seen data "./dancer.seen"
23:41:42hubbelpreglow: and probably way longer using speex, one only needs to spin up the HD once in an hour or so.. =)
23:41:54preglowspeex looks like a really nice codec
23:41:57preglowi want to port it now :/
23:42:04n0bbyholy cow, thats efficient
23:42:23preglowthat is, i know it's a cool codec, i've used it before, but looks like porting it'll be fun
23:42:24hubbelpreglow: yp.. it's been ported to arm and "sharkfin", so fixed-point should be there already
23:42:26preglowlots of asm ;)
23:42:32preglowfixed point is there
23:42:44preglowblackfin, btw :P
23:43:05preglowsharc is their other dsp platform, heh
23:43:14hubbelpreglow: ok =)..
23:43:21n0bbyi thought asm didnt need changed for different targets since it is so low level? O_o
23:43:32Coldtoasthey does rockbox now shut off the player before voltage gets dangerously close to 3v? Probably been fixed ages ago, right?
23:43:33preglowthat's exactly why it needs to be changed
23:43:49hubbelpreglow: wonder if we can mail the speex people and the will port it automaticly =)
23:43:51preglowColdtoast: hardware doesn't allow us to get too low
23:44:02preglowhubbel: nononono, i wanna do it :)
23:44:12Coldtoastok. cos I was just skimming the forums on misticriver
23:44:13hubbelpreglow: hehe =)
23:44:22n0bbyyou nerds :P
23:44:44Coldtoastwhere they were saying with rockbox, had some funkiness when voltage got low
23:45:03preglowyes, and i believe linus said the hardware doesn't let us get it too low
23:45:52 Part n0bby
23:45:55 Join n0bby [0] (
23:46:20preglowdoes anyone know if gas has a set of local label names that are reset for each non-local label?
23:46:23n0bbytrying to change the font in mIRC to something a little more easy on the eyes
23:47:06 Join iRiverowner [0] (
23:47:13ep0chok twice as fast
23:47:21iRiverowneris the backlight fading on the iRiver intentional?
23:47:43preglowiRiverowner: well, it would hardly fade nonintentionally
23:47:49n0bbyits optional
23:47:51iRiverownerlooks good though
23:47:54n0bbyi love it
23:47:59iRiverownerIs it possible to fix it on a dimmer level?
23:48:08iRiverowneri'e "brightness" control?
23:48:16preglowbut not as of right now
23:48:20n0bbygood idea
23:48:23preglowbut it's a possibility we're looking into
23:48:27iRiverowneryea I just installed the latest version of iRiverbox
23:48:31n0bbyits a bit harsh in total darkness
23:48:35iRiverownerit looks good the way it fades gracefully in
23:48:36n0bbywould save battery too
23:48:43iRiverownerwould it work on the remotes' backlight as well?
23:48:50preglown0bby: not necessarily, that's why it isn't in yet
23:48:52iRiverowneroh ok
23:48:59preglowiRiverowner: no, the remote backlight isn't possible to fade
23:49:08iRiverowneri know the remote uses an EL backlight, and I believe it will wear out over time
23:49:18preglowiRiverowner: you can't fade EL backlights
23:49:18n0bbythere you
23:49:23n0bbyit does fade then :P
23:49:43iRiverownerwell yea, its some sort of polymer that glows, but wears out
23:49:59Coldtoastsee this post?
23:50:04ep0chcan i rename the codec files via rockbox? or best to switch off and rename the files via usb?
23:50:08iRiverownerAnyway im happy with iRiverbox for the time being, since the shuffle is much better
23:50:30ColdtoastNarf's last post on that page is interesting
23:50:38ep0chahh what the heck lets see what happens :)
23:50:47iRiverowneri can't believe how good you lot are
23:52:06preglowep0ch: rename? yes of course you can
23:52:25preglowep0ch: the codecs are reloaded from disk each time they're needed
23:52:51ep0chyeah just tried cool :)
23:53:05preglowbut beware that the codec might be cached, so you might need to switch to a file of another type to reload the codec
23:53:09Coldtoastyay. got a .wps I like
23:53:09Coldtoastcouple of fomatting things that would be solved if you could justify stuff but otherwise, nice
23:53:43ep0chi just rename the file in the codecs dir? or rocks dir? or both?
23:53:46 Join ripnetuk [0] (
23:53:53ep0chcos i just did the codecs dir
23:53:55Coldtoastman! 14 bleeding edge builds jsut today alone!
23:54:05Bagdermy first take
23:54:13Bagderno iriver around atm though ;-)
23:54:35iRiverownerstill dunno why the iriver f/w is so poor at shuffle
23:54:58preglowdon't try to understand it
23:55:00n0bbythe random generator is poor
23:55:02preglowBagder: untested, i hope? :P
23:55:13Bagderpreglow: untested, yes indeed
23:55:31Bagderbut builds fine
23:55:43 Quit iRiverowner ("CGI:IRC (EOF)")
23:55:49Bagderoh, incomplete...
23:56:49Bagderthere's a CODECS file in there, specifying which plugins that are actually codecs
23:57:25preglowdo we want the codec files to be in plugins/ ?
23:57:28BagderI'm not very happy with this solution, but it is better than now
23:57:34preglowthe bovious choice of codecs/ is taken, though
23:57:47preglowbovious <- word of the day
23:57:57 Quit Harpy (Read error: 145 (Connection timed out))
23:57:59Bagderyes, and it re-uses the same Makefile etc
23:58:26Bagderso it was conveniant to leave them there
23:58:43hubbelthe iriver has a wonderful random source, just sample from the internal mic and use the last significant bits in each sample or so =)

Previous day | Next day