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 2005-09-07

00:00:26iMarkmaybe a gigabit ethernet port in place of the firewire
00:00:51preglowbut blah, i need to get this macsr stuff commited before i vanish for a while
00:05:04***Saving seen data "./dancer.seen"
00:05:32preglowthis dsp in codec thread nastiness really destroyed the elegance of the macsr thread context
00:09:51 Nick iMark is now known as iMark`away (
00:11:57 Quit Wett (Read error: 104 (Connection reset by peer))
00:14:35 Nick paugh is now known as AliasCoffee (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:16:52 Join midk [0] (
00:18:16 Quit matsl (Remote closed the connection)
00:19:23 Join Wett [0] (
00:21:36rasherAnyone know how clever the rockbox firmware update is? How good is it at detecting that the file is damaged?
00:24:26]RowaN[h10 is up to 20gigs?
00:24:30]RowaN[ says 6 gigs
00:24:36]RowaN[but ...
00:25:03 Quit mirak ("Leaving")
00:25:31amiconnHrmpf. H140 just froze
00:34:07 Join edx [0] (
00:34:22 Quit Moos (" HydraIRC -> <-")
00:39:02 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
00:40:14rasherThe lacking section of the changelog annoys me
00:43:02 Join DangerousDan [0] (
00:43:57 Quit DangerousDan (Read error: 104 (Connection reset by peer))
00:45:30 Join TCK [0] (
00:52:13 Join stripwax_ [0] (
00:53:38stripwax_So, here's a question about playing a track when 'Move To Next Folder' is enabled. Am I right in thinking that selecting a given track creates a dynamic playlist consisting of all the tracks in that folder (not just the one you select)?
00:54:17rasherI believe so, yes
00:54:27amiconnstripwax_: yes, exactly
00:54:29rasheryou could always "view current playlist"
00:54:38rasherto check what's in there
00:54:55rasherWhat's a nice way to get a wav sine wave?
00:56:13rasherMaybe audacity will help me.
00:56:50rasherIt would.
00:57:33stripwax_amiconn - hm. So presumably when it gets to the end of the current playlist, it knows (somehow) that it should replace the dynamic playlist with all of the tracks from the next folder
00:57:36stripwax_How does it know?
00:57:55stripwax_(For example, it doesn't do this if I'm building a playlist one file at a time)
00:59:17preglowrasher: what you need that for?
00:59:36rasherJust to have something that's a constant volume
00:59:55rasherGoing to try and see if I can get a volume-level graph of rockbox/iriver
01:00:22amiconnrasher: Volume-level graph?
01:00:40stripwax_Ok, I'll stop beating around the bush; my question is really, if Move To Next Folder is enabled, why can't it do the same thing if we try and skip BACK a track from the FIRST track in the playlist.
01:00:44preglowi'll have to make coldfire_set_macsr use "r" as a constraint if i want it to be able to take variables as parameters as well
01:00:45amiconnThe volume mapping is dB-linear
01:00:59preglowbut then every coldfire_set_macsr will be two instructions instead of one, even when not needed
01:01:03rasheramiconn: I'd like to relate it to the way the iriver firmware does it
01:01:16rasherso I can see "25 in iriver = X in rockbox"
01:01:24*amiconn can't imagine what that should be good for
01:01:27stripwax_rasher - makes sense to me (as an ex iriver firmware user)
01:02:27preglowoh well, i guess that isn't the end of the world
01:03:02stripwax_amiconn - two reasons (one good one). So we can help ex iriver firmware users get to grips with rockbox. And potentially also so we can say 'rockbox is louder than iriver firmware' or 'rockbox is quieter than iriver firmware' at max volume levels (min too, i guess)
01:03:27rasherI'm pretty sure iriver is louder
01:03:30amiconnWell, zero is real zero in rockbox
01:03:57amiconnMax volume is the maximum possible without clipping risk
01:03:58rashernow how to measure this without having any measuring equipment..
01:03:59stripwax_amiconn - right but is '1' louder or quieter than iriver '1' ?
01:04:19stripwax_rasher - optical out surely
01:04:20amiconnIt's certainly quieter since it is much more fine-grained
01:04:36rasheramiconn: at max volume.
01:04:45 Quit chenco ("Leaving")
01:04:45 Quit ghostiger ("Leaving")
01:04:54stripwax_(or wait- does optical out bypass all volume changes?)
01:05:40preglowstripwax_: yes, it does
01:05:42rasherDoes the "Replay Gain > Prevent clipping" setting affect wav playback?
01:05:49preglowrockbox uses the codeec chip volume registers
01:05:58stripwax_rasher - sample line out into line in on your soundcard then, and don't change any of your soundcard settings in between
01:05:59rasherI don't have any equipment with optical in anyway.
01:06:04preglowno soft volume (yet)
01:06:42 Quit ender` (Read error: 113 (No route to host))
01:08:12rasherhuh.. this wav won't play.
01:08:45rashersine440hz.wav: RIFF (little-endian) data, WAVE audio, Microsoft PCM, 16 bit, mono 44000 Hz, says `file`
01:08:49rasherWhy wouldn't it play?
01:09:24stripwax_does RockBox support mono wavs?
01:09:38rashermaybe not
01:10:57rasherNow how to make audacity create a stereo file
01:12:19preglowwith the utmost difficulty
01:12:23preglowrubbish program
01:12:30rasherSure looks like it.
01:12:32stripwax_Google's probably quicker....
01:12:47preglowi can make you a nice stereo wav if you give me specs
01:13:20rasherI'm not sure specs are relevant
01:13:29amiconnAudacity surely isn't rubbish
01:13:31 Join zeekoe [0] (
01:13:36preglowamiconn: oh, but it is
01:13:43stripwax_rasher - check the link above and download "0_16.wav: Extreme pure sine (0.001% THD+N) with maximum level. Peak level meters must show exactly 0 dBFS, the corresponding RMS-value is exactly -3 dBFS."
01:14:01rasherstripwax_: yeah, I guess that'll do - thanks
01:14:16zeekoegenerating stereo wavs with audacity is easy
01:14:19rasherI wonder how long it is though
01:14:27 Quit _aLF ("Leaving")
01:14:34rasherzeekoe: do tell
01:14:35zeekoejust add a track, right click, make one of them left and the other one right
01:14:37preglowrasher: like i said, just give me desired frequency, amplitude and length, and i'll make you one
01:14:47*midk stares
01:14:55zeekoeyeah, i'm still alive :-P
01:15:05rasherpreglow: seems zeekoe came to the rescue
01:15:07midkomfg.. you weren't hit by a car yet or anything? :)
01:15:13stripwax_rasher - 224KB wav.
01:15:17zeekoenah, just been busy
01:15:27zeekoeor sleeping
01:15:44rasherstripwax_: that's nt very long.. I'll be fidgetting around and it'll be stopped or restarted and things will be horribly skewed
01:16:19stripwax_rasher - sorry, 224KB zip, 1MB wav. but yeah, true
01:16:44rasherActually. "Project > Add stereo track", "Generate > Tone"
01:16:45rasherthat works
01:16:53zeekoeah well
01:16:57zeekoeit's been a while
01:17:05zeekoedon't have it on my linux install yet
01:17:20rasherI guess what you said would've worked as well
01:18:11zeekoeyou can make one 100 Hz track and another 101 Hz
01:18:13zeekoeit's fun
01:18:21stripwax_Hm, wonder what RockBox does if presented with an 8-channel .wav file..
01:18:40zeekoestripwax_, just read through the source...
01:19:01rasherThe wav codec is lacking.
01:19:13stripwax_no biggie, I wouldn't use wavs anyway
01:19:31*preglow cries when he sees flac performance
01:19:33zeekoei remember replying to an email with something about wavs
01:19:42rasherThere's a patch that adds all sorts of stuff
01:20:06*zeekoe has got audacity now...
01:21:28zeekoe1245261proper WAV file handling :
01:21:28zeekoe This patch adds:
01:21:28zeekoe- correct parsing of WAV header, even if it's not the
01:21:28DBUGEnqueued KICK zeekoe
01:21:28zeekoecanonic 44-byte header
01:21:28zeekoe- handling of WAV files other than 44.1/16bits/stereo
01:21:51rasherwow, 100 101hz is upsetting.
01:23:22zeekoeone left, the other one right
01:23:24stripwax_zeekoe - including 8 channel? ;-)
01:23:26zeekoethat's cool
01:23:48preglowseems the macsr save fixed the vorbis problem
01:23:51zeekoestripwax_, what do you think?
01:24:05rasherOkay, I'm no longer convinced iriver is louder than Rockbox.
01:24:08stripwax_what's the vorbis problem
01:24:09rasherMy poor ears.
01:26:02preglowit would have surprised me considerably if iriver was louder than rockbox
01:26:47rasherI think it does allow for louder music if the bass is turned up
01:27:09preglowyes, maybe, but it will surely clip if it is so
01:27:16stripwax_preglow - or rather, what *was* the vorbis problem?
01:27:35preglowstripwax_: just some weird behaviour that dictatet the placement of a couple of emac init functions
01:28:06preglowamiconn: so, you reckon this counts as a bugfix?
01:28:43stripwax_oh ok. so nothing that actually affects anything, just the structure of the code
01:28:49preglowstripwax_: yeah
01:29:05rasherHrm. I've no idea how to measure the volume..
01:29:23preglowrasher: i can do a couple of quick measurements
01:29:30preglowi've got stuff hooked up anyway
01:29:53zeekoestripwax_: from quicly reading the source of the patch, it looks like 8 channel wavs don't generate an error. there are some (channels ==2) things inside, though so it probably won't play that well
01:30:09rasherJust do something like measure 5,10,15,20,25,30,35,40 with iriver
01:32:14preglowrasher: k, doing now
01:32:23stripwax_hm. am I right in thinking both headphone out and line out volume are controlled by the codec chip? why isn't line out fixed volume? apologies if that's a *REALLY* dumb question
01:32:43rasherstripwax_: You're right. And "it just isn't."
01:32:49preglowman, it feels awkward using the iriver firmware
01:33:15preglowstripwax_: why, because i for one happen to like it responding to volume
01:33:26preglowand i think i'm not alone
01:34:37stripwax_preglow - any advantage to using line-out vs headphone-out if you need volume control?
01:35:43stripwax_(hm, tho I guess it saves a lot of power if we're only listening on headphones). are lineout and headphone out driven at the same time?
01:35:45preglowi just don't see a reason for not having the option of keeping it volume controlled
01:35:54preglowyes, they are
01:35:55rasherIf I had to guess, I would have expected line out to be a fixed-volume without-effects output
01:36:00 Quit midk (Read error: 104 (Connection reset by peer))
01:36:07 Join midk [0] (
01:36:11rasherBut I'm not that bothered.
01:36:23rashernot at all, really
01:36:25stripwax_rasher - it's what I would have guessed to.
01:36:52preglowrasher: they seem to be the exact same volume
01:37:05preglowwhy would you expect that?
01:37:37rasherpreglow: What is the same volume?
01:37:51stripwax_preglow/rasher - iriver vs rockbox?
01:37:59stripwax_or lineout vs headphone?
01:38:31preglowrasher: looking at the rest of them now
01:38:31 Join zeekoe_ [0] (
01:38:35 Quit zeekoe (Read error: 104 (Connection reset by peer))
01:38:35preglowrasher: was talking about iriver vs rockbox
01:39:28rasherYeah, I'm most interested in the volume-distribution in iriver now
01:39:43rasherjust because, I guess
01:39:53stripwax_I would have guessed lineout is fixed volume so I could plug it into my amp and control the volume using the amp's volume control in exactly the same way I control the volume of my CD player. Having said that, my CD player also has a (totally redundant, imo) volume adjustment feature. so what do I know ;-)
01:40:29rasherYes. I expect the line-out to act like a "dumb" output device
01:40:39preglowrasher: i'll spam you the db measurements in query
01:40:58rasherif freenode will let you
01:41:13preglowlet's see
01:41:17preglowdid you get it?
01:41:28rasherDidnt' get anything
01:41:33preglowgimme a sec
01:41:37preglowtrying again
01:42:57 Quit ]RowaN[ ()
01:45:28preglowamiconn: looks like iriver too uses something db-linear-ish
01:45:54rasheryeah, it's not very interesting
01:46:28rasherso rockbox 50 ~ iriver 20, right?
01:48:31preglowrodckbox 50 ~ iriver 5
01:49:26rasherhow does that work then?
01:49:48preglowwhat iriver did was pretty sensible, if you ask me
01:49:58pregloweverything below 40 is pretty useless in the rockbox scale
01:50:08preglowso they compressed that entire area to under 5
01:50:40stripwax_I always thought iriver 20 was much louder than rockbox 50. now I know why.
01:50:41rasherAh, so there's a larger dip at the lower end than what I thought
01:50:59preglowthey've got a db linear scale at the top end
01:51:06preglowand a db log scale at the lower end
01:51:12preglowwe've got a db linear scale the entire way
01:51:32rasherSounds reasonable of them
01:52:04preglowyes, i think it's more sensible myself, a db linear scale is pretty nice and uniform, but it's not the way the ear works
01:52:52stripwax_preglow - where do the two scales join up? is it linear from 5-40 and log from 1-5?
01:53:07preglowkind of hard to tell, i just did a pretty coarse measurement
01:53:14preglowstripwax_: from 0 to 40, steps of five
01:53:22preglowstripwax_: first irregularity i saw was at 5
01:53:33preglowlinear down to 10, in other words
01:53:59preglowhmm, might seem it tends a bit more downward at 10 as well, btw
01:54:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:54:10*preglow does a graph
01:54:39rasherHard to tell from the points you gave me
01:55:33rashercan't see it for 10, at least
01:59:35preglowif i fit a curve through the data, it starts going nonlinear slightly before 10
02:00:30preglowbut who knows, we don't need to know exactly what they do anyway
02:00:39preglowat least we know they're not db linear
02:03:10preglowif you ask me, the usable rockbox range is 100 to 60
02:03:16preglowthe rest should be compressed to a smaller scale
02:05:08***Saving seen data "./dancer.seen"
02:06:17preglowbut ok, this macsr business seems to work, think i'll commit it
02:07:39rasherMy kingdom for a real BTS
02:08:35rasherbug tracking system
02:08:40rasherSourceforge is a joke
02:08:45preglowyes, yes it is
02:09:01preglowa bad one at that
02:09:13stripwax_gnight all
02:09:16 Part stripwax_
02:10:05rasherNot too keen on bojira either, but it is pretty widely used - maybe it's good enough afterall
02:10:25 Quit zeekoe_ ("Leaving")
02:13:10preglowbojira <-
02:13:23rasherAh yes
02:13:24preglowtook me a couple of secs before i caught on, hehe
02:13:34rasherPicked that up somewhere
02:15:46*rasher sets up a local copy of bugzilla
02:16:19preglowis it just me, or is max firmware size for some archoses exceeded again?
02:16:33preglowmy test build failed gloriously
02:17:28rasherDoesn't look like it from the build table
02:17:52preglowerror: max firmware size is 200KB!
02:17:52preglowmake[1]: *** [/home/thomj/rockbox-devel/build/ajbrec.ajz] Error 255
02:17:56preglowlooks like it from here :/
02:18:38rasherOh right, which gcc version are you using?
02:19:02rasherAh.. dunno then
02:19:17preglowi don't think any of my changes are compiled for archos
02:19:19preglowso can't be that
02:19:45rasherWow, that was easy.. "apt-get install bugzilla", answer a couple of questions, reload mysql, presto, instant bojira
02:20:40 Quit hicks (Remote closed the connection)
02:20:58preglowmight be the devel build option that tipped it
02:21:23preglowlooks like i didnt get that fm recorder anyway, so i'm probably going to stay archos agnostic
02:24:04preglowi dont know if this even counts as a bugfix, but amiconn seemed to want me to commit it, so here we go
02:24:09 Quit AliasCoffee ("Leaving")
02:24:32rashernow if I had a copy of the sourceforge xml export thing, I could have a go at creating an import script
02:25:00preglowisn't that in cvs?
02:25:44rasherNot that I know of
02:25:50rasherWhere would it be?
02:26:49rasherI very much doubt it
02:27:48preglowoh well, you'll just have to bug bagder
02:27:49rasherIt's this huge gob of xml that includes every bit of info in the sourceforge trackers
02:28:11rasherGood news is that bugzilla includes an import script
02:28:26rasherbad news is that it expects the input to be in bugzilla-xml-bug format
02:29:08preglowdoesn't sound like too big a job for a perl script
02:29:14preglowif they're both valid xml
02:29:15rasherIndeed not
02:29:21rasherI'd hope they are
02:33:57*preglow sighs, gets a barf bag and enters apps/codecs/libFLAC
02:34:57rasherhappy happy cplusplus land
02:40:39preglowseems like he just ran some evil c++ to c converter on a c++ project
02:41:41rasherI remember it being horrible just looking at it.
02:41:57rasherAnd I wasn't even trying to understand what it was doing
02:42:09preglowit's that kind of source code
02:43:10preglow50% of it is bullshit intended to make it look oo like
02:44:14coobyar libflac is ick
02:44:24 Quit massoud (Read error: 110 (Connection timed out))
02:45:05preglowcoob: word
02:45:20preglowok, latest build didn't break, just tell me if something soundwise isn't like it used to be
02:45:52coobi hate make oo in c :/
02:46:18rasherHm, bugzilla will need some tweaking to fit Rockbox.. "Platform: DEC, HP, Macintosh, ..."
02:46:37preglowrasher: i'd expect that is easily tweakable
02:46:47rasherYeah, just need to find out *WHERE*
02:46:50rasherIt's useful
02:46:52preglowcoob: should be made a punishable offense
02:47:00rasherassuming we can make it say "Player" "OndioFM" etc
02:47:20preglowcoob: oo in c is ok, but just pretending you have c++ is not
02:48:22*coob from the stone age
02:48:24coobi hate oo
02:48:43preglowi think oo is completely ok
02:48:52preglowgot used to it a couple of years back
02:49:17 Quit Wett ("Visitez !")
02:52:18preglow * class FLAC__SeekableStreamDecoder : public FLAC__StreamDecoder
02:56:06preglowinclude/FLAC/seekable_stream_decoder.h: struct FLAC__SeekableStreamDecoderPrivate *private_; /* avoid the C++ keyword 'private' */
02:56:09preglowi have seen enough
02:56:28preglowthe jury may retire
02:57:55rasherNasty, the OS list is hardcoded into enter_bug.cgi
02:58:17rasherOh well
02:58:36rasherOr not.
02:59:17coobuse flyspray instead :D
02:59:29coobslim bugzilla
02:59:38coobit's not amazing, needs tweaking, but does the job.
03:00:11preglowi don't think anyone's ever been caught calling bugzilla 'amazing' anyway
03:00:44rasherIf it's too slim, we may as well just stay with SF
03:01:47rasherAlthough I have no real idea of which features are missing
03:01:52rasherapart from, you know, speed and stability
03:03:14rashercoob: where's your installation hiding?
03:03:46preglowgave libflac a nice little boost, i did
03:03:58rasherDoes it actually play now?
03:04:07preglowwent from 50% boost rate to 25¤
03:04:20coobrasher: flyspray?
03:04:23rasheriirc,, some files wouldn't even play realtime
03:04:35rasherWarning: mysql_connect(): Access denied for user 'ipod_req'@'' (using password: YES) in
03:04:36preglowrasher: you know of anyone that posted such files?
03:04:45rasherNo.. :-\
03:04:52coobour admins are lazy :/
03:05:32*rasher looks at flyspray's flyspray
03:05:43rasherhow meta
03:05:44coobwell, that was embarrasing.
03:06:08*coob makes note to self to kick lazy admins
03:07:16rasherdoes flyspray have a notion of duplicates?
03:07:19preglowrasher: you think codec opts count as bugfixes? :P
03:07:53rasherWell, considering that it wasn't even playing realtime for some files, I'm inclined to say yes, but I'm hardly an authority
03:08:02coobrasher: i just close entries as duplicates
03:08:04preglowi'll wait until tomorrow anyway
03:08:15rasherProbably a good idea
03:08:40preglowi'll be leaving for a couple of days tomorrow, but i think i'll have time to confer with the big boys before i leave
03:09:01rashercoob: Hmm.. I'd like to be able to link them like bugzilla does.. at least for Rockbox there are/were several items with a bunc of discussion on each
03:09:19rasherif course if you catch them at once it's no problem
03:10:14rasherpreglow: I'd think there's a bit more room for letting non-bugfixes get through in strictly iriver areas
03:10:34rasherit's not like there'll be any archos regressions
03:10:35preglowme too, just wanna have it verified to avoid embarassments
03:10:42rasherof course
03:11:09preglowand like you said, flac not bloody playing some files is most definitely a bug
03:11:22rasherI'm absolutely terrified of committing anything but language updates without checking it
03:12:15rasherI have this fat driver fix sitting
03:12:23rasherbut noone has had time to look at it :-\
03:12:47preglowwhat fix?
03:13:11rasherAdd a check to add_dir_entry to check for entries ending in "."
03:13:16rasher(which is not allowed)
03:13:24rasherApart from '.' and '..' of course.
03:13:37rasherAnd potentially other checks can be added in the same place
03:13:50preglowsounds pretty sane
03:13:52preglowyou need to nag more :)
03:14:05rasherGuess I do
03:14:14rasherThere's even a bug report about it
03:14:16preglowand thusly commenced the brushing of the teeth
03:19:02rasherIt's the only open bug under "drivers"
03:19:11preglowwith all this new iram, the future should be nice for the codecs
03:19:21rasherHow much did you get?
03:19:30coobdo you guys stick memcpy in iram?
03:20:18preglowcoob: afaik,yes
03:20:34preglowcoob: that is, i expect it will end up there, once amiconn writes an asm optimised version
03:20:40preglowi think we've got asm memset in iram
03:20:53coobany others?
03:21:17rashercoob: you'll be amused to find that I've ported matrix from podzilla to rockbox
03:21:29rasherHas a bug though :(
03:21:41coobhaha noway
03:21:55coobthat was like the second bit of c i wrote , it's horrid :)
03:22:10 Quit cYmen ("zZz")
03:22:25rasherHaven't touched a much of it
03:22:46rasherSimplified the init a great deal (not using alloc)
03:22:46cooboh well, i did nick your cube :)
03:23:04rasherAnd rewrote the blit_char of course
03:23:17coobah yes, i only used alloc because we have to deal with different screen sizes dynamically (not as a build option)
03:23:25rasherIt needs greyscale support badly though
03:23:38preglowcoob: not too much that i know of, lots of codec data in iram, suspect i'll stuff some codec code there as well soon"> is not so hot
03:24:28preglowrasher: bug a dev tomorrow, would be a nice bug to close before release
03:24:45rasherYes, indeed
03:24:54rasherI'm amazed noone has touched it
03:24:59rasherthe fix is really simple
03:25:02 Join amiconn_ [0] (
03:25:05preglowmysethink the patch looks fine and dandy, muself
03:25:10preglowtake 2:
03:25:16preglowthink the patch looks fine and dandy, myself
03:25:28rasherI wonder if amiconn's alive and wake
03:25:41coobdid you change the font?
03:25:43preglowi'd guess no on the grounds of him having a job
03:26:01rashercoob: Yeah - only had 1bit display at the time
03:26:14rasherNicked the font from cmatrix
03:26:46rasherPlus, I figured it'd be easier than convert from whatever image format you were using
03:26:56preglowrasher: i'd probably just commit this, can't see how it can go wrong
03:27:10preglowit's a small piece to revert in the off chance of it being borked anyway
03:27:20zecool somebody's started writing a linux usb-storage driver for linux and gotten so far as reading data from the disk and some elements of filesystem support
03:27:32zecool somebody's started writing a linux usb-storage driver for the RIO KARMA and gotten so far as reading data from the disk and some elements of filesystem support
03:28:09rashersounds like a dirty job
03:28:23coobyeah, gr_bitmaps are horrid
03:28:27coobi hate microwindows :/
03:28:52coobrockbox is written from the ground up, ipl stuff is written from the top down :)
03:29:12rasherYou should start a rockbox port to ipod
03:29:21coobno thanks
03:29:30coobi'm a lazy developer... the worst kind
03:29:40coobi like having linux there
03:29:46coobmakes porting stuff like doom easy :)
03:30:01preglowoooh, i'd love a rockbox port to ipod
03:30:11preglowchances are very good i'd buy an ipod for it
03:30:56coobwell, all the hard re is done for you
03:31:50preglowjust need someone to start the port
03:32:15preglowmight actually give it a shot if some money falls in my lap some day
03:32:27preglowyou got more than one core going yet, btw?
03:35:18coobhyyeah sort of
03:35:25*rasher starts watching the build table religiously
03:35:28coobwell, yeah
03:35:34coobcan assign a function to the cop
03:35:39coobthe video player uses it
03:35:54coobwe don't use it (really) for audio decoding yet
03:36:33coobthe PP502[2|0] is much nicer than the PP5002
03:37:31preglowcoob: do you know how the apple firmware does it?
03:38:33coobi assume it uses one for the interface/file buffering and one for decoding
03:38:58 Quit amiconn (Read error: 110 (Connection timed out))
03:38:59 Nick amiconn_ is now known as amiconn (
03:39:04coobbut i dont actually know
03:39:27coobi do know the PP5020 has some primitive sheduling stuff
03:39:57preglowyeah, that's kind of what i would have thought it did as well
03:40:08preglowboth cores are equally powerful and have got their own iram?
03:40:49coobnope, only one bank of iram..
03:41:09coobthey both run at the same speed so yeah
03:41:16coobwe run them at 75mhz
03:41:50preglowhrmph, i'd love to dip my hands into some arm coding
03:42:44rasherA rockbox port would be cool.. you could triple boot it.. that'd be a first
03:43:23coobhaha that wouldn't be too difficult at all :)
03:50:53preglowidata is half full for flac
03:51:03preglowi'm pretty sure i'll have this baby in good shape before long
03:52:36rasherI see green builds.
03:54:13preglowso it would seem
03:54:38rashercoob: What's the deal with podzilla using old rockbox font files?
03:54:44rasher(if you know)
03:55:30rasherSeems weird since a) it uses pcf as well b) it uses the old format
03:57:13coobno idea..
03:57:27rasherfair enough
03:57:31coobi didnt do any of the font stuff
03:57:41coobi think microwindows can cope with fnt/pcf
03:58:07rasherWonder why fnt is used at all
04:01:55rasherI mean there's bdf2pcf..
04:05:11***Saving seen data "./dancer.seen"
04:05:14 Join QT [0] (i=as@madwifi/users/area51)
04:05:26preglowflac.c is such a nice mix of different code styles
04:05:36preglowoh well, might as well end the night with some relaxing code cleaning
04:13:20*rasher ponders nitpicking dansk.lang
04:18:06 Quit QT_ (Read error: 110 (Connection timed out))
04:24:38preglowthink i'll use what's left of the day for sleep
04:24:40preglowsee you later
04:24:48 Quit preglow ("leaving")
04:47:24 Quit rasher ("CGI:IRC")
05:19:59 Quit dpassen1 ()
05:23:47 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
06:05:15***Saving seen data "./dancer.seen"
06:35:42 Join LinusN [0] (
06:41:52zeso the 1st-3rd gen ipod has the portalplayer PP5002, the karma has the PP5003, the 1st/2nd gen ipods have the wolfson WM8721 DAC as does the karma, the 3rd gen ipod has the cypress CY7C68013 USB chip as does the karma
06:42:09zecan't tell what LCD controller/driver the karma has though
07:31:15 Join B4gder [0] (
07:55:47amiconnI've done a >12 hour test recording yesterday; got not even a signle frame error. Was recording from line in, q=6, 44.1kHz, stereo
08:05:17***No seen item changed, no save performed.
08:12:27 Join ender` [0] (
08:14:15amiconnLinusN: Did you read about my finding concerning the inclusion of the whole IRAM area into the rockbox binary, and the suggestion to introduce .ibss (and .iconst) sections in the logs?
08:14:37LinusNhaven't read the logs yet
08:15:07 Quit midk ("Leaving")
08:15:33amiconn starting 08:17
08:17:49amiconnYes, for constants
08:17:55LinusNwhat would be the difference?
08:18:00LinusNfrom idata
08:18:28amiconnThe difference is only for the compiler
08:18:48amiconnHandling in the iram copy routine would be same (even both at once)
08:19:19amiconnHave you tried declaring something like const int test = 32 IDATA_ATTR; ?
08:19:26LinusNah, yes
08:19:43amiconnThat doesn't work -> section attribute mismatch iirc
08:22:09 Quit paugh ("ah caynt get no.. serotonin..")
08:36:33LinusNLANG_END_PLAYLIST_RECORDER is bad imho, it says "End of song list" instead of "End of playlist"
08:37:13B4gderisn't that to better cover for directories as well?
08:37:34LinusNit's the word "song" that bothers me the most
08:38:09LinusNand rockbox is all about playlist nowadays
08:38:18B4gdersure, but people don't know that
08:38:24LinusNdirectory play is just creating the playlist for you
08:38:55LinusNi think "end of song list" isn't any better than "end of playlist" in this context
08:39:33LinusNespecially since there is no such thing as a "song list" anywhere else in rockbox
08:40:24LinusNand an audio book isn't a "song"
08:41:14B4gderI read you
08:41:22B4gderI still think playlist is a bit too restrictive
08:41:32B4gderfor the simple minds of mere users
08:42:33LinusNi *might* accept "track list" or even just "list"
08:42:55LinusNor even "no more tracks"
08:42:59 Join webguest82 [0] (
08:43:52LinusNi just think "song list" sound so lame
08:44:06B4gderI have no real opinion on either choice
08:49:36amiconnHehe, deutsch.lang uses "Playlist zu Ende". Reminds me that I wanted to check the translation...
08:49:51amiconnIn other places 'playlist' is translated...
08:50:50LinusNi think "end of playlist" is the best choice
08:50:58amiconnI agree
08:54:20 Quit webguest82 ("CGI:IRC (EOF)")
08:55:26LinusNdamn, gotta go
08:55:28 Part LinusN
09:13:39iMark`awaycould someone make me a logf build of something fairly recent?
09:13:45iMark`awayor if they have one handy
09:13:55iMark`awaylogf is useful for the remote :)
09:14:04iMark`awayand my cygwin is fucked
09:18:31 Quit merbanan ("Leaving")
09:19:51 Join merbanan [0] (
09:22:41 Join Zagor [0] (
09:23:37 Join pilot000 [0] (
09:27:21zelastlog -regex '5002'
09:27:31zeerg wtf
09:32:46 Quit pilot000 ("CGI:IRC")
09:32:55 Join pilot000 [0] (
09:57:53k0obamornin all as new i will ask my noob's question : is there a plug in to pitch mp3 with rockbox maybe stupid question...
09:59:01Zagorno. but on the archos version, pitch is built-in
10:00:06k0obaon my jukebox 6000 i cud pitch mp3 ? zagor ?
10:01:45Zagorsorry, recorder only
10:01:57k0obathx a lot zagor
10:02:29*k0oba reminds me zagora beautiful place under the sun
10:05:06HClanyone ever heard of htcpcp/1.0 ?
10:05:20***Saving seen data "./dancer.seen"
10:05:24Zagoryeah. you run it?
10:05:32HClno, not yet
10:05:52HCli think i should find an implementation of it though.
10:06:01Zagorultimate geek thing, along with udp over pigeon
10:07:34 Join linuxstb_ [0] (
10:18:29 Join thomjoha [0] (
10:19:35thomjohaamiconn: cool, i didn't know using constraints like that was possible
10:19:53 Nick thomjoha is now known as preglow (
10:20:10 Join DangerousDan [0] (
10:24:33 Quit linuxstb (Read error: 110 (Connection timed out))
10:43:11 Join Wett [0] (
10:43:30 Quit linuxstb_ (Remote closed the connection)
10:53:24preglowamiconn: and in case you haven't read the logs, i've got flac running a bit faster, you think that counts as a bugfix?
10:59:13amiconnpreglow: I'd think so
10:59:32amiconnEven if it's not (only) a bug fix, it's not a new feature either
10:59:50amiconnFeature freeze does mean just that - no new features allowed
11:00:20amiconnIn fact a number of things I've done recently aren't strictly bugfixes, but optimisations
11:03:28preglowi've got the boost ratio down from 50% to 25% on one file
11:03:47preglowdo you know of any files that didn't play?
11:03:52*solex_ is copyying 2.5gb of flacs to his irver atm
11:04:02 Join LinusN [0] (
11:04:44amiconnpreglow: I have a number of files that cause boost ratio of 100%, and stutter
11:05:12amiconn(a whole album encoded with -8)
11:06:02amiconnI wonder why you didn't think about the alternative constraint thingy - see my latest commit...
11:06:18preglowi didn't, because i didn't know it was possible with several constraints
11:06:28preglowi thought the inline assembler wasn't smart enoug
11:09:11amiconnThe gcc manual is really helpful here...
11:10:31 Join webguest75 [0] (
11:10:59preglowyeah, i've read it lots, still didn't know of that
11:11:08 Quit webguest75 (Client Quit)
11:11:10 Join webguest75 [0] (
11:12:29 Quit webguest75 (Client Quit)
11:12:45CoCoLUSthank god for faster flac... almost all my files stuttered last time i tried
11:12:58 Join kurzhaarrocker [0] (
11:13:48kurzhaarrockerLinusN: I think I won't find the time to fix split editor any further before V2.5. I'll be on vacation soon.
11:13:59LinusNsaw that
11:17:24 Join ashridah [0] (
11:19:27 Quit kurzhaarrocker (Remote closed the connection)
11:23:00solex_has the flac patch been already committed?
11:29:30ashridahflac patch?
11:29:37preglowwhat patch?
11:29:55preglowi can commit it, yet
11:31:15preglowi take it iram isn't zeroed each time a codec is loaded?
11:31:35preglowshould probably memset it to avoid surprises, since flac originally used calloc for this memory
11:32:30amiconnI think it does zero IRAM
11:33:09amiconnThe codecs include the whole IRAM segment if they use IRAM, as do the plugins. This includes ".ibss" space
11:34:14preglowbut yeah, code that used to be needed, but now isn't, should i take it out, or keep it to somehow easy merging of library updates?
11:34:29 Join Moos [0] (
11:34:35preglowtook out a bunch of mallocs and accompanying pointer checking
11:35:32 Join cYmen [0] (
11:43:32preglowaight, i'll just plain delete it then, to avoid having commented code spread all over the place
11:48:57preglowflac is such a malloc hell
11:52:05 Quit cYmen (
11:52:05 Quit ashridah (
11:52:05 Quit amiconn (
11:52:05 Quit TCK (
11:52:05 Quit dwihno (
11:52:05 Quit Asku (
11:52:05 Quit bluebrother^ (
11:52:05 Quit godzirra (
11:52:05 Quit igor47 (
11:52:05 Quit Rob- (
11:52:05 Quit coob (
11:52:05 Quit ze (
11:52:05 Quit crwl (
11:52:05 Quit odd (
11:52:05 Quit Strath (
11:52:05 Quit Zagor (
11:52:05 Quit ender` (
11:52:05 Quit QT (
11:52:05 Quit Bagder (
11:52:05 Quit pike (
11:52:05 Quit Nibbler (
11:52:05 Quit Slasheri (
11:52:05 Quit HCl (
11:52:05 Quit Wy (
11:52:05 Quit pabs (
11:52:05 Quit Ismo (
11:52:05 Quit crash (
11:52:05 Quit preglow (
11:52:05 Quit B4gder (
11:52:05 Quit Hadaka (
11:52:05 Quit Moos (
11:52:05 Quit DangerousDan (
11:52:05 Quit pilot000 (
11:52:05 Quit merbanan (
11:52:05 Quit edx (
11:52:05 Quit Mxm`Pas`Bien (
11:52:05 Quit solex_ (
11:52:05 Quit Rick (
11:52:05 Quit thegeek (
11:52:05 Quit iMark`away (
11:52:05 Quit gromit` (
11:52:05 Quit silencer_ (
11:52:05 Quit Seed (
11:52:41 Quit k0oba ()
11:53:46NJoinBagder [0] (
11:53:46NJoincYmen [0] (
11:53:46NJoinMoos [0] (
11:53:46NJoinashridah [0] (
11:53:46NJoinDangerousDan [0] (
11:53:46NJoinpreglow [0] (
11:53:46NJoinpilot000 [0] (
11:53:46 Join Zagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
11:53:46NJoinmerbanan [0] (
11:53:46NJoinender` [0] (
11:53:46NJoinB4gder [0] (
11:53:46NJoinQT [0] (i=as@madwifi/users/area51)
11:53:46NJoinamiconn [0] (
11:53:46NJoinTCK [0] (
11:53:46NJoinedx [0] (
11:53:46NJoinMxm`Pas`Bien [0] (
11:53:46NJoinsolex_ [0] (
11:53:46NJoinodd [0] (
11:53:46NJoinze [0] (
11:53:46NJoincrwl [0] (
11:53:46NJoindwihno [0] (n=dw@
11:53:46NJoingodzirra [0] (
11:53:46NJoinigor47 [0] (
11:53:46NJoinAsku [0] (
11:53:46NJoincoob [0] (
11:53:46NJoinStrath [0] (
11:53:46NJoinbluebrother^ [0] (
11:53:46NJoinpike [0] (
11:53:46NJoinRob- [0] (
11:53:46NJoinNibbler [0] (
11:53:46 Join Rick [0] (i=rick@unaffiliated/Rick)
11:53:46NJointhegeek [0] (
11:53:46NJoinHCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
11:53:46NJoiniMark`away [0] (
11:53:46NJoinHadaka [0] (
11:53:46NJoinSlasheri [0] (
11:53:46NJoingromit` [0] (n=gromit`
11:53:46NJoinsilencer_ [0] (
11:53:46NJoinWy [0] (
11:53:46NJoinpabs [0] (
11:53:46NJoinIsmo [0] (
11:53:46NJoincrash [0] (
11:53:46NJoinSeed [0] (
11:53:46Mode"#RockBox +o Bagder " by
11:53:47***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
11:53:47***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
11:55:33solex_preglow: i was talking about your flac speedup you were talking about
12:02:05 Join Febs [0] (
12:05:23***Saving seen data "./dancer.seen"
12:11:12preglowcan't promise your files will work now either, but i'd think they should
12:11:33preglowi'll just ignore all the comments and commit it for now, so you can try it
12:11:39preglowi've got to leave for a couple of days soon anyway
12:12:21solex_preglow: until now, all my files played well
12:12:30preglowuntil now?
12:12:33solex_but reducing cpu load is always a good idea
12:12:46solex_and still :)
12:12:51preglowyou weren't the one with the glitchy files
12:13:43preglowthere, commited
12:13:57preglowamiconn: you got a chance to test if your files work now?
12:14:40solex_i won't be able to test it until this evening
12:14:54solex_but i'll report if i find an error
12:15:16preglowi wont be around, so lets hope you wont ;)
12:18:00 Join webguest05 [0] (
12:18:49 Part webguest05
12:44:23amiconnpreglow: Not right now, but in the evening
12:55:19pikewanna see a cool little notebook ?
13:22:52 Join hshah [0] (
13:26:38 Join hicks [0] (
13:27:20 Quit pilot000 ("CGI:IRC (EOF)")
13:44:53 Join banan_ [0] (
13:46:05preglowwoop, gotta go
13:46:06 Quit preglow ("leaving")
13:49:47 Quit hshah ("CGI:IRC")
13:58:44 Join banan__ [0] (
14:00:17 Quit merbanan (Read error: 110 (Connection timed out))
14:02:13 Join webguest10 [0] (
14:03:00 Join Sucka`away [0] (
14:03:08 Nick Sucka`away is now known as Sucka (
14:05:25***Saving seen data "./dancer.seen"
14:16:01 Quit webguest10 ("CGI:IRC (EOF)")
14:17:06 Quit banan_ (Read error: 110 (Connection timed out))
14:26:14 Quit Febs (Read error: 110 (Connection timed out))
14:28:54 Quit B4gder ("Lämnar")
14:59:26amiconnLinusN: Regarding your latest wps commit - I'm already thinking about a fundamental wps problem for a while.
14:59:56amiconnCurrently, unconditional images are drawn with DRMODE_FG, in order to blend them with the wps content
15:00:54amiconnHowever, a number of users already requested 4-grey support, but for 4-grey bitmaps there is no draw mode, they're always drawn opaque
15:01:23LinusNyes, that will be a problem
15:01:29amiconnI did not yet find a good solution...
15:01:49amiconnExtending the bmp loader will be easy; that's a thing I want to do anyway
15:01:58LinusNthat's the easy part
15:02:23amiconnIt will be able to load any .bmp (except 32bit space-wasters) and convert to either mono or native format
15:03:05 Quit hicks ("Too lazy to change my quit message")
15:03:32amiconnMeaning, it will load 1, 4, 8, 16 and 24 bit uncompressed bmps, similar to bmp2rb, just with way less memory consumption
15:05:08amiconnThe only solution for the bg image problem I can think of is drawing unconditional images first, then text content, then conditional images
15:05:31LinusNthe problem is that the text is drawn opaque
15:05:51amiconnYes, and always up to the end of line
15:06:04amiconn...except with multi-alignment
15:08:49amiconnHmm, text isn't necessarily drawn opaque. lcd_puts() does this, but lcd_putsxy() uses the draw mode like any other drawing function
15:09:57LinusNwe could try DRMODE_FG
15:10:47amiconnIt will require to do a full redraw if any line of text changes though
15:11:21amiconn...because there is no way to remove the old text from the background
15:13:15amiconnSome parts will always need to use DRMODE_SOLID, e.g. scrolling text, peakmeter
15:13:48amiconnWith a new design, we could allow to set foreground and background shade for these elements
15:31:07 Quit ashridah ("Leaving")
16:00:52LinusNamiconn: what's left in the "Xing header generation" issue in ReleaseTodo?
16:03:41amiconnLong recordings (or shorter recordings started after a long time of prerecording etc.) get a Xing header without framecount
16:04:33amiconnThat's the problem I intend to solve with estimation from the recording time; I'm working on it
16:04:47amiconnYou know, the 32bit overflow problem
16:05:30***Saving seen data "./dancer.seen"
16:05:33amiconnCalculating playtime from frametime in msec and framecount (and the reverse calculation) wil never overflow for mp3s stored on a FAT(32) partition,
16:06:18amiconnbut the same calculation based on framecount, sample frequency and frame size in samples may overflow
16:07:26amiconnThe (rather synthetic) worst case would be a 2GB-1 file with 8kbps and 24 kHz sample rate
16:07:40 Join hicks [0] (
16:08:42amiconnThe frame length in such a file would be 24 bytes, so framecount = (2GB-1) / 24B . Multiply by frametime (24 msec) does just not overflow...
16:09:58amiconnI'll try with long long, but I expect the binary to grow significantly...
16:10:48amiconnThe bad thing with frame_time is that it's somewhat off for 44.1, 22.05 (and 11.025) kHz
16:14:18LinusNgotta go now, cu later
16:14:22 Part LinusN
16:17:13*amiconn found a solution...
16:18:00amiconnFortunately 44100 and 1152 have a number of factors in common, so 44100 / 1152 can be reduced to 1225 / 32
16:20:46 Join rasher [0] (
16:20:58rasherBagder: you around?
16:22:54 Quit bluebrother^ (
16:22:54 Quit Asku (
16:22:54 Quit dwihno (
16:22:54 Quit amiconn (
16:22:54 Quit cYmen (
16:22:54 Quit godzirra (
16:22:54 Quit Rob- (
16:22:54 Quit igor47 (
16:22:54 Quit coob (
16:22:54 Quit TCK (
16:22:54 Quit ze (
16:22:54 Quit Sucka (
16:22:54 Quit crwl (
16:22:54 Quit odd (
16:22:54 Quit Strath (
16:22:54 Quit hicks (
16:22:54 Quit Bagder (
16:22:54 Quit Zagor (
16:22:54 Quit pike (
16:22:54 Quit Nibbler (
16:22:54 Quit ender` (
16:22:54 Quit QT (
16:23:36 Quit Slasheri (
16:23:36 Quit HCl (
16:23:36 Quit Wy (
16:23:36 Quit crash (
16:23:36 Quit Ismo (
16:23:36 Quit pabs (
16:23:36 Quit Hadaka (
16:23:36 Quit rasher (
16:23:36 Quit banan__ (
16:23:36 Quit silencer_ (
16:23:36 Quit iMark`away (
16:23:36 Quit thegeek (
16:23:36 Quit solex_ (
16:23:36 Quit Mxm`Pas`Bien (
16:23:36 Quit DangerousDan (
16:23:36 Quit Moos (
16:23:36 Quit Seed (
16:23:36 Quit Rick (
16:23:36 Quit gromit` (
16:23:36 Quit edx (
16:26:07NJoinamiconn [0] (
16:26:35 Join solex [0] (
16:26:50NJoinSlasheri [0] (
16:26:52NJoinBagder [0] (
16:26:52 Join igor47 [0] (n=igor47@
16:26:52NJoinTCK [0] (
16:26:52NJoinpike [0] (
16:26:52NJoinNibbler [0] (
16:26:52NJoinQT [0] (i=as@madwifi/users/area51)
16:26:52NJoinender` [0] (
16:26:52NJoinZagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
16:26:52NJoinhicks [0] (
16:26:52NJoinze [0] (
16:26:52NJoingodzirra [0] (
16:26:52NJoinStrath [0] (
16:26:52NJoinodd [0] (
16:26:52NJoincrwl [0] (
16:26:52NJoinSucka [0] (
16:26:52Mode"#RockBox +o Bagder " by
16:26:53***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:26:54NJoinHadaka [0] (
16:26:54 Quit Hadaka (
16:27:00 Join silencer [0] (
16:27:00NJoinHCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
16:27:00NJoinWy [0] (
16:27:00NJoinpabs [0] (
16:27:00NJoinIsmo [0] (
16:27:00NJoincrash [0] (
16:27:05 Join Maxime [0] (
16:27:09NJoinSeed [0] (
16:27:09 Join gromit` [0] (n=gromit`@
16:27:10***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:27:21 Join iMark`away [0] (n=Mark@
16:27:21NJoinedx [0] (
16:27:22***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:27:28NJoinbanan__ [0] (
16:28:01 Quit Strath (
16:28:01 Quit ze (
16:28:17NJoinMoos [0] (
16:28:37NJointhegeek [0] (
16:28:42 Join Rick [0] (
16:28:43***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:28:58NJoinStrath [0] (
16:29:30NJoinrasher [0] (
16:29:32NJoinAsku [0] (
16:29:34NJoinbluebrother^ [0] (
16:29:36NJoincoob [0] (
16:29:54NJoincYmen [0] (
16:30:04NJoindwihno [0] (n=dw@
16:31:16NJoinze [0] (
16:31:42 Join Naked [0] (
16:31:57 Nick Naked is now known as Hadaka (
16:45:12 Quit TCK (Read error: 104 (Connection reset by peer))
16:45:53 Quit edx (Read error: 110 (Connection timed out))
16:52:24NJoinMxm`Pas`Bien [0] (
16:52:24 Quit Maxime (Read error: 104 (Connection reset by peer))
16:52:39 Join OnkelJonas [0] (
16:58:06 Join tvelocity [0] (
17:17:49 Join webguest21 [0] (
17:18:18 Nick iMark`away is now known as `away (n=Mark@
17:19:32 Part webguest21
17:26:47 Quit Nibbler ("life is like a rental car, you fuck it up, and give it back.")
17:27:26 Join Nibbler [0] (
18:05:34***Saving seen data "./dancer.seen"
18:07:05 Quit `away ("Leaving")
18:07:35 Join bagawk [0] (i=1000@unaffiliated/bagawk)
18:16:50 Quit thegeek (Read error: 104 (Connection reset by peer))
18:17:08 Join webguest79 [0] (
18:18:08webguest79in wps can I write, %?bt >20 and <30
18:19:46 Join thegeek [0] (
18:23:12rasherwebguest79: No.
18:23:22rasherThere's nothing like that.
18:26:16webguest79thanks, pity, I wanted to add several icon bmp's to indidcate battery level,
18:27:44 Join edx [0] (
18:28:53 Part webguest79
18:31:17 Join zezayer [0] (n=zezayer@
18:34:12 Nick banan__ is now known as merbanan (
19:01:44 Quit bagawk ("Leaving")
19:11:27 Quit Seed (
19:11:27 Quit Sucka (
19:11:27 Quit godzirra (
19:11:27 Quit crwl (
19:11:27 Quit odd (
19:11:27 Quit thegeek (
19:11:27 Quit Nibbler (
19:11:27 Quit ze (
19:11:27 Quit merbanan (
19:11:27 Quit gromit` (
19:11:27 Quit Bagder (
19:11:27 Quit hicks (
19:11:27 Quit Zagor (
19:11:27 Quit igor47 (
19:11:27 Quit pike (
19:11:27 Quit QT (
19:11:27 Quit ender` (
19:11:28 Quit HCl (
19:11:28NJoinodd [0] (
19:11:30 Join gromit` [0] (n=gromit`
19:11:31***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:11:33NJoinQT [0] (i=as@madwifi/users/area51)
19:11:36 Join ze [0] (i=ze@
19:11:37***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:11:40 Join Seed [0] (i=ben@
19:11:41***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:11:42 Join godzirra [0] (i=shawn@
19:11:43***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:12:02 Join Sucka [0] (n=NNSCRIPT@
19:12:03***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:12:03NJoinigor47 [0] (n=igor47@
19:12:15 Join ender` [0] (i=ychat@
19:12:16 Join hicks [0] (n=hicks@
19:12:16NJoinmerbanan [0] (
19:12:16***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:12:16***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:12:21NJoinpike [0] (
19:12:23 Join Bagder [0] (n=daniel@
19:12:24NJointhegeek [0] (
19:12:24***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:12:34NJoinZagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
19:12:34 Join NibbIer [0] (n=sven@
19:16:24NJoinHCl [0] (i=hcl@2001:610:1908:8000:290:27ff:feca:8029)
19:22:37 Join crwl [0] (
19:22:38***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
19:23:38 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
19:25:33 Quit OnkelJonas (Read error: 104 (Connection reset by peer))
19:29:17 Part hicks ("Leaving")
20:01:59 Join LinusN [0] (
20:04:30amiconnre LinusN
20:05:37***Saving seen data "./dancer.seen"
20:05:46 Join preglow [0] (
20:05:52amiconnI can't say I like the current wps image solution, especially with your latest image inflation commit
20:06:13preglowamiconn: tried any flacs?
20:06:31amiconnA really sophisticated wps uses so many .bmps that it's an utter mess
20:06:59LinusNamiconn: what is wrong with it?
20:07:12amiconnI'd want to reduce image loading support to one single image, and the display tags should allow clipping
20:07:33LinusNone image?
20:08:02amiconnThis way, any graphical wps design would consist of exactly 2 files, the .wps defining the layout and the .bmp containing the graphics
20:08:51LinusNyou mean it would cut out the various images from one large bmp?
20:09:02amiconnMany skinning systems do that
20:09:05LinusNwhy would that be better?
20:09:31amiconn- less sophisticated loading
20:09:41amiconn- faster loading of the whole wps
20:09:54LinusNbut instead much more sophisticated cutting from the bmp
20:10:05amiconn- .wps file will probably get smaller, because of only one filename
20:10:19LinusNbut instead a lot of cutting instructions
20:10:38amiconnThe cutting is simple, just pass the 6 parameters to lcd_mono_bitmap_part(). It will do the rest
20:11:15LinusNsounds like it would be a lot more difficult to make a wps
20:11:26amiconnEvery image display tag would have 6 parameters instead of 5, but you will save <n-1> image loading tags for <n> images
20:11:55LinusNbut you would have to use a more advanced image editor to create the wps graphics
20:12:15LinusNms paint won't do anymore
20:12:30amiconnThe images for all wps's in the gallery are tailor-made anyway
20:12:40amiconnms paint would still suffice
20:13:06LinusNof course they are, but merging all images and calculating the cutting coordinates will be a pain with ms paint
20:13:11amiconnIt does display coordinates
20:13:14preglowms paint isnt exactly good at the pixel level
20:14:03amiconnEver tried View->Zoom->Magnify (or whatever it is called in english ms paint) ?
20:14:11LinusNwe will revamp the entire wps anyway, won't we?
20:14:28dpassen1in paint, you can even show the grid of pixels
20:14:33amiconnLinusN: yes
20:14:49amiconn..but that shouldn't stop us from doing it incrementally
20:14:49 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
20:15:59LinusNall i want is to make it work well in this release
20:16:00preglowbut has anyone checked to see if problem flacs play? feedback would be most interesting
20:16:30amiconnpreglow: I just started my problematic album and let it stay in audio thread debug screen
20:16:54amiconnBoost ratio 50..60 % so far
20:17:11preglowmy files were at around 25%
20:17:15*rasher catches the digital flow
20:17:33rasherLet's see if any of these exhibit problems without your opts
20:17:55preglowtheres tons more to be done for flac, i just did the simplest of the simplest stuff
20:18:22amiconnpreglow: This album was locked at 100% before...
20:19:09preglowthen success
20:19:13preglowflac can be saved indeed
20:19:53LinusNgotta go now, i hope i'll be back later this evening
20:19:55 Part LinusN
20:21:33rasherwent from 100% to ~60% for me as well
20:21:47paughpreglow, sorry, i missed the start of this conversation.. are the flac improvements in the latest build?
20:22:04rasherpaugh: Latest bleeding edge build, yes
20:22:24amiconnpreglow: Now up to 70%, but still playing fine
20:22:39preglowlike i said
20:22:44preglowive hardly touched it
20:22:49preglowiram is still only half-full
20:23:06preglowas soon as ive got a reliable method of performance measurement, ill start with some more stuff
20:23:26Moosmaybe MPC :)?
20:23:44preglowMoos: sorry, but no, heh, not unless i get immensly bored
20:24:04Moosdon't worries
20:24:35Moosany mpa improvments with this free space iram?
20:25:41preglowive got mp1 and mp2 running faster
20:25:56Moosoh wonderfull :)
20:26:00preglowand ill see about mp3 as well
20:26:44preglowMoos: mpc support requires a full rewrite of the libmusepack math part, which im not exactly keen to do myself
20:27:35rasherSeems the musepack devs started hiding..
20:31:42CoCoLUSmy flacs play fine now
20:31:42preglowCoCoLUS: yo
20:31:47CoCoLUSno stuttering
20:31:54CoCoLUSalthough i just have two albums :)
20:32:03CoCoLUSbut those played fine.
20:32:04preglowhehe, i have none
20:32:29rasherBagder: could I have the xml blob from SF? (unless it contains sekrit details)
20:32:40rasherGoing to see about importing into bugzilla
20:32:46rasherHave a local installation running now
20:32:52Bagdersure you can, hang on...
20:34:55rasherDebian package of bugzilla works nicely btw
20:36:09Bagderthe xml is ~6.5MB
20:36:25rasherI guess it needs a bit of tweaking if you want it to run out of /bugzilla/ instead of /cgi-bin/bugzilla/
20:36:40Bagder830K compressed
20:37:06Bagderdidn't see my privmsg?
20:37:37rasherThere we go
20:45:19 Quit Wett (Read error: 104 (Connection reset by peer))
20:51:48 Quit thegeek (Read error: 104 (Connection reset by peer))
20:55:12 Join thegeek [0] (
21:04:03Zagorseen this?
21:04:18Zagorapparently people did want a screen on their players after all...
21:08:28dpassen1who would've thought
21:08:35dpassen14 GBs of flash is pretty sweet though
21:08:52 Join Maxime [0] (
21:09:06dpassen1i wonder if people will be ripping out the flash like the micro drives with the minis
21:09:24preglowthats small
21:09:28preglowhardware info, please ;)
21:09:48Zagorportalplayer for sure
21:10:29HCli don't see what people see in ipods..
21:10:56coobits app502x
21:10:58dpassen1its a status thing, walk around a college campus, tons
21:11:00coobits a PP502x
21:11:04coobnot sure if 20 or 22
21:11:05HCldpassen1: haha. yea. the cost of a 4gb flash is more than the cost of that thing
21:11:23HClin compactflash anyways :p
21:11:28dpassen1could use 4 GBs of flash for my dads D70
21:11:38coobit wont be too quick (the flash) :)
21:12:08dpassen1it'll be a long time before i upgrade my iHP-120, hopefully ;-)
21:12:20HClsame, only iHP-140
21:12:49HCliriver rocks :)
21:13:00dpassen1hardware, maybe
21:13:49dpassen1you were a fan of the iHP firmware?
21:13:54HClyou can just upgrade the disk if you really want to..
21:13:59HCloh, no, nm, i get it :)
21:14:03HClyes, the iriver hardware rocks
21:14:09dpassen1i wouldn't buy a new mp3 player unless i was certain of an impending rockbox port
21:14:23HCljust today a classmate saw my iriver and told me iriver owns :p
21:14:27HCland that its a fun toy, heh :p
21:14:42HCli wonder if he knows rockbox o.o.
21:14:45dpassen1heh, i've seen 2 other iHP-1x0s in more than 2 years at my school
21:14:57HCli know markun, obviously
21:15:01HCland then someone else i worked with
21:15:05HCltold me he had an 120
21:15:21HClhe was watching my computer remotely and saw my h140 connected
21:16:08HClso i know 2 other people who have one, and one person who knows about them..
21:16:40 Join ]RowaN[ [0] (
21:16:52dpassen1one of my roommates in my apartment has an iAudio X5 30 GB, its not as impressive as i thought it'd be
21:16:54*HCl stares at a soccer match.
21:17:02HClgeeze. 3-0 for the netherlands..
21:17:13HCli never saw an iaudio x5
21:17:27HClhow is it not as impressive as you thought it to be?
21:17:45dpassen1its significantly smaller than my 120, but the firmware is very reminiscent of the iHPs and i hate waiting for it to play its little intro video on bootup
21:17:58HClthats nasty
21:18:16dpassen1also, it currently has an odd sorting behavior A-Z, then a-z
21:18:18]RowaN[right, ive got my souncard going into my amp now via digital coax (not optical).. is it normal that volume/treble/bass settings on my pc affect the output? i thought it would just send a clean signal, like the irivers optical output does (correct me if im wrong)
21:18:40coobnorthern ireland/england is a boring game :/
21:18:53HCldpassen1: thats a quite common sorting flaw
21:19:00HCldpassen1: iriver original has that too
21:19:17dpassen1yes, the firmware is very similar to irivers
21:19:37HCli never understood why irivers didn't support on the fly playlists.
21:19:47dpassen1also, i can't imagine who decided to put the USB 1.1 OTG port on the player, but the 2.0 port on a separate subpack
21:19:47HClhence i made the searchengine
21:21:21rasherokay, that ipod nano is slick looking
21:21:45HCli still don't get attracted by it...
21:23:52rashernow why they'd put a colourscreen on it...
21:24:20Maximebecause "colour is great!! even if it's useless!"
21:24:20dpassen1'wow factor'
21:24:24cooberm rasher
21:24:29coobit does have a colour screen
21:24:36rasherYes.. which is what I don't get
21:24:40coobi read why don't...
21:25:02coobbecause some people have pictures, and like to look at them.
21:25:17cooband some people like watching family guy on their ipods (i.e. me)
21:25:20rasherIf this was around a year ago I wouldn't have bought my h120
21:25:30rasherI never really *wanted* 20gb
21:25:36dpassen1rasher: a 4 GB player would have swayed you?
21:25:37 Part dpassen1
21:25:38amiconnpreglow: The album played almost up to the end; average boost ratio 66%
21:25:39rasherAnd smaller is better.
21:25:40 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
21:25:56amiconnThe hang near the end was caused by low bat...
21:26:19rasherOh wait, I keep forgetting the horrible itunes deal
21:26:30rasherThere's still no way in hell I'd be buying an ipod.
21:27:01coobi said
21:27:11coobnot buying an ipod because it's an ipod?
21:27:11rasherI want to be able to use my player as a UMS device. Purely.
21:27:21coobyou can.
21:27:31coobif you use linux :)
21:27:33rasherNo, because I can't just put music on it wwithout jumping through hoops
21:27:41dpassen1you can't play music you've loaded on via UMS
21:27:56coobrunning an open source db generator is jumping through hoops? ok.
21:28:05coobdpassen1: you can with linux :)
21:28:20rasherWell yeah, it is.
21:28:34dpassen1still no gapless or replaygain, which are big selling points for me
21:28:57 Join webguest77 [0] (
21:29:05rasherDoes ipods even allow anything like browse-by-filesystem?
21:29:17dpassen1nope, only their db
21:29:31rasherWell that's another deal-breaker to me
21:29:38Bagderipodlinux does I guess ;-)
21:29:48rasherAbout that Rockbox port to ipod...
21:30:05 Quit Mxm`Pas`Bien (Read error: 110 (Connection timed out))
21:30:18webguest77the ipodlounge was great for doing reasearch on the pod, was put off by the windows users problems
21:30:35coobipodlounge is horrid :/
21:30:44dpassen1ilounge, nowadays
21:30:48coobipodlinux does.
21:31:02coobwhatever lounge, it's a pile of turd.
21:32:07webguest77but the defects are real, forums are great tools for assessing units failures
21:35:14webguest77coob: does ipodlinux playback mp3 gapless ?
21:36:42coobi'm the only person working on the music backend
21:36:45cooband i'm terribly lazy
21:37:13webguest77so it'll be a while then :)
21:37:21coobwebguest77: forums are filled with clueless tools, i wouldn't treat anything on any forum worth shit.
21:37:47coobwebguest77: probably not, when people actually bother to work, developing under linux is actualy really quick
21:37:58rasheripl does seem sort of immature - the code (or distro more like), not the coders
21:38:00coobas you can really easily port stuff.
21:38:07webguest77yes and no, I can differentiate between a teenagers rant and a real problem
21:38:08coobit's not very efficient, but meh, it works.
21:38:28coobwebguest77: i just cant be bothered to wade through 839021830921 posts of OMG L@@K
21:38:42webguest77I share that sentiment
21:43:00 Part webguest77
21:49:26 Join solex_ [0] (
21:51:58 Join _aLF [0] (
21:56:55 Join matsl [0] (
22:05:22 Quit solex (Read error: 110 (Connection timed out))
22:05:41***Saving seen data "./dancer.seen"
22:13:07preglowamiconn: you exited before the end, or it started skipping?
22:14:25amiconnNone of the two
22:14:40amiconnIt started disk spinup/down/up/down...
22:15:05amiconnThen I connected the charger, but it didn't recover.
22:15:53amiconnI was able to return from the debug menu to the wps. Playback was stopped, wps showed last position
22:16:14amiconnAnother STOP press returned to the browser as normal
22:18:08 Quit Moos (Read error: 104 (Connection reset by peer))
22:18:15 Join Moos [0] (
22:19:30amiconnThere are still many quirks in the sw playback engine :(
22:27:08 Quit Zagor ("Client exiting")
22:43:27 Quit dpassen1 ()
22:52:08 Quit edx (Read error: 110 (Connection timed out))
22:52:17preglowyes, yes there are
22:52:32preglowwell, well
22:52:38preglowat least flac behaves better now
22:54:54 Nick Sucka is now known as Sucka`away (n=NNSCRIPT@
23:03:03 Quit zezayer (Remote closed the connection)
23:08:51rasher" is reporting that Microsoft is hauling the European Commission into court" ... what the...
23:14:37 Join San [0] (n=Test@
23:20:15Sananyone know the name of the main prosessing chip in the H300?
23:20:21Sanis it moterola?
23:20:33Bagdertried the wiki?
23:20:45Sannot yet
23:20:47Sanis it in there?
23:21:17Sanah, thanks
23:21:26rasherBagder: I'm going to give up on the bugzilla import I think
23:21:40]RowaN[in rockbox when you delete a file it asks u to presss PLAY to confirm, but play does not confirm the deletion.. you have to click the joystick to confirm (iriver)
23:22:39rasher]RowaN[: correct.
23:22:40Bagder]RowaN[: yes that is an old and known issue
23:23:31Bagderit'll get fixed in the language overhaul
23:23:59rasherComments *must* include a mention of which buttons are used on different units
23:24:08rasherotherwise translators would have to own every device
23:24:44Bagderwell yes, but soon the english strings will differ for different targets
23:24:54Bagderfor this very reason
23:24:58rasherOh right.
23:25:09rasherI forgot that there are English strings to go by
23:25:59amiconnI'm thinking about icons for the buttons for quite some time now
23:26:03rasher13 languages fully done.. quite good
23:26:16amiconnI'm still not sure whether this would be a good idea, but... has some advantages
23:26:51amiconn(1) Icons don't need to be translated :P
23:27:02amiconn(2) They take up less space on the display
23:28:04preglowand what would an icon for the joystick look like? ;)
23:28:29rasherI know!
23:28:29amiconnThat's one reason why I am not sure it would be a good idea"> < has a joystick icon
23:29:02amiconnThe F buttons on the recorders is another case
23:29:16preglowrasher: where is it?
23:29:18amiconn...but I doubt these will be used verbatim in a request
23:29:30rasherpreglow: search for joystick in the page
23:29:32amiconnThey are intended to be used as soft buttons, so...
23:29:47preglowwell, its not exactly that kind of a joystick, hehe
23:29:56rasherTrue, but it'd get the point across
23:29:59amiconnAn icon for the Ondio MODE button could also be difficult
23:31:23preglowamiconn: what do you think of compressing the lower end of the volume scale, btw? i dont know if you read yesterday nights logs
23:31:48amiconnI don't think this is a good idea
23:32:13preglowi think its more sensible, everything from 0 to 60 could easily be compressed to a smaller range, since its a range that really doesnt need as much detail as 0-60 give
23:32:21 Join DangerousDan [0] (
23:32:41preglowlinear db doesnt give linear loudness perception, at least for me
23:32:51preglowand linear loudness perception is what we really want, imho
23:32:59amiconnThe current scale is a linear-to-dB mappinq (well, except 0% of course) and I think this is the most meaningful mapping
23:33:41rasher50-60 is more or less the same "difference" as 0-50 to me
23:33:47amiconnCompressing 0 to 60 to a smaller scale would actually compress the range I use most when using earphones, so I wouldn't like that at all
23:33:50preglowrasher; yes, exactly
23:34:18preglowamiconn: at those volumes, i hear practically nothing with my setup
23:34:33Mooshere too :(
23:34:47amiconnWhen I'm listening in a quiet environment, I can easily go down to 45 and still enjoy my music
23:35:09preglowthat i cant
23:35:10amiconn(on H1x0)
23:35:28amiconnSennheiser MX450
23:36:34preglowthe step from 0 to 60 db is about as big a step in loudness as 60-80 for me
23:37:15rasherHm, is MX500 the successor to 450, I wonder
23:37:29amiconnI bet it's not
23:38:05amiconnIirc MX500 and MX550 have a builtin volume control
23:38:31rasherSo they do
23:38:37rasherAnd "bass boost".. argh!
23:38:38preglownoticed that iriver firmware uses a linear db scale down to the last ten points, where a log scale is used
23:38:41*amiconn is confused about mpeg versions, layers and frame sizes
23:38:51preglowand in use it seems a much better scale, to me
23:38:53rasherOh well, I don't own any "real" headphones anyway.
23:39:01preglowamiconn: how?
23:39:03amiconnDifferent sources give contradicting info....
23:39:19preglowah,i notice that too
23:39:33amiconnI'm looking for the # of sample in each of version 1/2/2.5 and layer 1/2/3
23:39:35preglowi bet libmad can give you the correct answers
23:39:56amiconnLayer 3 is easy, and consistent
23:40:25amiconnV1L3 has 1152 samples/frame, V2L3 and V2.5L3 have 576 samples/frame
23:41:09amiconnBut it seems that this "halve-size rule" for V2 and V2.5 doesn't hold for layer 1 and 2
23:41:25amiconnThis also means our code in mp3data.c is wrong...
23:42:10pillis there a reason why some file genres dont get retrieved by %ig?
23:42:53pillthe genre is there, it just doesn't get displayed by the WPS
23:43:25rasherCould you supply a file? (unless of course someone pops in and give the answer)
23:44:02pillwell i could
23:44:38rashermy name at-sign my name .dk
23:45:02rasheractually no
23:45:11rasheruse jfh at-sign adr .dk
23:45:26rasherThe other one does greylisting and may end up delaying the mail half an hour
23:45:35rasherWhich seems broken
23:46:08rasherwith jfh first
23:46:18*rasher punches the spambots
23:49:04pillon its way
23:49:07pillit's 4MB
23:50:15pillit does it on several files
23:50:19pillfor no apparent reason
23:50:40rasherHere we go.. downloading..
23:52:30 Join ashridah [0] (
23:53:29preglowmake.bat in liwavpack is pretty redundant....
23:57:04rasherHm.. doesn't seem to be in id3.c
23:57:09rasher129 - Hardcore
23:57:43pillthough i have many Hardcore tracks and it's displayed
23:57:48rasheranything > 126
23:58:01rasherThat's probably because it's a id3v2 tag
23:58:06rasherThose just include the text
23:58:14rasherBut this is v1 only
23:58:28rasherAnd Rockbox doesn't seem to know about v1 genre numbers >126
23:58:29pilllet me test
23:58:55pillyou seem to be right

Previous day | Next day