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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2006-09-15

00:00:27LandusI've even deleted .rockbox and rockbox.iriver, and reinstalled Rockbox.
00:00:33LandusNo change, apparently.
00:00:57LinusNbluebrother: that trademark seems to be registered in germany only. you need to register separately for a EU trademark.
00:01:17 Quit mirak_ (Client Quit)
00:01:48bluebrotherLinusN, ok.
00:02:04linuxstbWhy don't we register Rockbox?
00:02:16LinusNin which country?
00:02:24linuxstbIn the EU.
00:02:40linuxstbIf I'm understanding what you just said...
00:02:53bluebrotherdifferent thing, I just noticed your comment on FS #5990. I've made some mockups after discussion earlier today. If you're interested, http://www.stud.uni-karlsruhe.de/~uhcn/rockbox/
00:03:10 Quit mirak6 ("CGI:IRC (EOF)")
00:03:16LandusWhat's FS mean?
00:03:26LandusOh wait.
00:03:28LandusNevermind.
00:03:29LandusDuh.
00:06:48bluebrotherFlyspray ;-)
00:06:51LinusNhehe, there is a EU registration of the Rockbox trademark: Deutsche Rockwool Mineralwoll-GmbH
00:07:15bluebrotherRockwool?
00:07:24*bluebrother knows that name.
00:07:25LandusMan, I really hate how bad my coding skills are.
00:07:31LandusOtherwise, I'd be fixing this myself.
00:07:40GaloisI've seen products named rockbox in Future Shop
00:07:45Landushjkharadfhjjhadfhdas
00:13:26LinusNbluebrother: as far as i can tell, with my severly limited knowledge of german trademark law, rockbox-lounge should be clear if they stop offering rockbox builds
00:13:58 Part billodo
00:15:28bluebrotherLinusN, from what I know from the press there is still a chance they get trouble because of their site being "confusing" to customers.
00:15:44LinusNyeah right
00:15:55LinusNeat my shorts
00:15:57*preglow hates assholes
00:16:17preglowas in the two-legged kind
00:16:28*Bagder admits to being confusing to customers too at times
00:16:54*LinusN is even a confused customer at times
00:17:08preglowsomeone start lobbying samsung for docs
00:17:09bluebrotherhey, I thing I have a hacked version of the int slider working :)
00:17:41LinusNbluebrother: hacked? cool. did you break the encryption? ;-)
00:17:59Bagderhttp://www.engadget.com/2006/09/14/microsoft-launches-the-zune/
00:18:06Bagderlooks like a nice lcd on that thing
00:18:08bluebrotherI decrypted the mystics of rockbox coding :D
00:18:21 Quit midkay_ ("Leaving")
00:18:34 Quit mantono ("Don't worry, be happy")
00:19:21Bagder"you can share a full-length track with a friend, and they've got three times to listen to it over a three day period, after which they can flag the song for purchase on the Zune Marketplace"
00:20:38*LinusN sighs
00:21:02Bagderyeah
00:22:35preglowcan anyone tell me what's so grand about the ipod 5g hw eq?
00:22:46preglowsome people seem to think it's the second coming of christ
00:22:47LinusNless cpu?
00:23:19preglowthe sound is supposedely beyond grand
00:23:27LinusNoh?
00:24:01preglowhttp://www6.head-fi.org/forums/showthread.php?t=194390
00:24:01LinusNwell, it's probably because the software eq that this norwegian loser wrote is so lousy
00:24:12amiconnpreglow: The main difference is that you can use the hw eq without skipping, unlike its software twin
00:24:21preglowamiconn: read the posts
00:24:24 Quit bbroke ("Chatzilla 0.9.75 [Firefox 1.5.0.6/2006072814]")
00:25:06*amiconn checks the date
00:26:13preglowas far as i could see, the 5g hw eq was a really standard deal
00:26:13 Quit matsl (Read error: 60 (Operation timed out))
00:26:24preglowand i don't think i've done any major mistakes in coding the sw eq
00:27:35LinusNwell, could it be that the hw eq has different parameters, so it's only a matter of selecting a matching set of F/Q for the sw one?
00:27:57preglowshrugs
00:28:20preglowpeople only seem to be able to spout audiophile nonsense when they're talkinga bout it, so it's really hard to tell where the actual difference lies
00:28:28peturone guy says just enabling it makes it sound better - not even configure any band...
00:28:46preglowghargh
00:28:50preglowand i don't know anyone with a 5g
00:29:02belzepetur: lol
00:29:28*LinusN has a 5g
00:29:33peturhttp://www6.head-fi.org/forums/showpost.php?p=2360207&postcount=81
00:30:10 Quit aliasone_ (Read error: 110 (Connection timed out))
00:31:14amiconnpreglow: Really odd, especially the Q values
00:31:33amiconnThese values surely produce nasty peaks
00:32:16preglowLinusN: well, i'd appreciate some recordings of white noise passed through the hw eq :>
00:32:52 Quit [San] (Read error: 110 (Connection timed out))
00:33:00belzepreglow: i also noticed some problems with the sw eq on the h110
00:33:08preglowbelze: what kind of problems
00:33:09 Join [San] [0] (n=San@212.2.179.84)
00:33:14 Join Angyrmanh10 [0] (i=543891a5@gateway/web/cgi-irc/labb.contactor.se/session)
00:33:18belzeit produced some weird squeking noises on silent files
00:33:21amiconnpreglow: Speaking about passing signals through some filter - do you think recording the loudness characteristics of the MAS would be helpful?
00:33:23preglowbelze: ?
00:33:27belzebut only at sepcific volume-levels
00:33:31amiconnWould be a quite large set of recordings...
00:33:33belzei could record it, if you like
00:33:42preglowbelze: would be cool
00:33:46 Quit Angyrmanh10 (Client Quit)
00:33:51preglowamiconn: nah, three-four would be ok
00:34:01 Quit quiksilver ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
00:34:02preglowamiconn: they probably just interpolate between flat output and filtered output anyway
00:34:25preglowamiconn: well, it would be cool to see what kind of a filter they're using
00:34:27amiconnpreglow: loudness is by its very nature volume dependent, then loudness is adjustable itself (0 to 12 db), and the center frequency can be switched
00:34:36preglowamiconn: oh, right...
00:34:44preglowamiconn: then it's less helpful
00:34:49preglowand too much work
00:35:05amiconn1kHz is standard, and you can switch to 2kHz (Micronas calls that "superbass")
00:35:35 Join secleinteer [0] (n=scl@adsl-70-237-207-174.dsl.stlsmo.sbcglobal.net)
00:35:53*amiconn checks the datasheet
00:35:59preglowbelze: and you're absolutely sure it's the eq causing it? what volume levels?
00:36:19belzeyes, when i disabled it, it was all gone
00:36:27belzei am looking at it now
00:36:33amiconnOh, loudness is even adjustable from 0 to +17dB
00:36:36preglowthe only thing i can imagine happening that would make the eq produce noise at all with no input, is limit cycles
00:36:41preglowand i use far too much precision for those to be audible
00:36:52preglowplus, the filter structure i use isn't very susceptible to limit cycles
00:36:54amiconn...in 0.25dB steps
00:36:57belzeat first i noticed it on some classical music-files
00:37:06 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/labb.contactor.se/session)
00:37:12belzeand i thought it was some lousy encoding, but it wasnt
00:37:39preglowhmm
00:37:43amiconnHmmmm.................
00:37:49preglowperhaps i should do some rounding on the shelving filter coefficients
00:37:57preglowthey're only 25 bits...
00:38:04preglowaka. more than bloody enough
00:38:26amiconnAs it is, we're not doing true loudness on archos....
00:38:27linuxstb_petur: That comment about just enabling the hw eq improving the sound is amusing - the hw eq has always been "enabled" in Rockbox, you can't turn it off...
00:38:33amiconn(iiuc)
00:38:48LinusNoh my god!
00:38:50linuxstb_All that's changed is the ability to change the gains from 0db...
00:39:12Bagderhahaha
00:39:15LinusNi enabled the hardware EQ on my 5g and it sounds HORRIBLE with the default settings
00:39:24peturrofl
00:39:33Bagdernothing beats a good placebo
00:39:49preglowespecially if it's laced with enough audiophile gibberish
00:40:41amiconnThe loudness setting of the MAS doesn't adapt itself to changing volume
00:41:39LinusNsomething is fishy with the default hw eq settings
00:41:45belzepreglow: ok, do you want it as a wave-file?
00:41:51LinusNall bands default to +12dB
00:42:27 Quit webguest11 ("CGI:IRC")
00:42:44preglowbelze: whatever floats your boat
00:42:54 Part stripwax_
00:42:55preglowbelze: but not lossy compressed, if possible
00:43:10preglowLinusN: how sensible
00:43:18Landusadafadfadfaerwjhdha
00:43:22LandusStupid tagcache bug.
00:43:43LandusNothing seems to stop it.
00:43:55 Join belze_ [0] (i=nifty@dslb-088-073-194-211.pools.arcor-ip.net)
00:43:57 Join XavierGr [0] (n=xavier@ppp142-238.adsl.forthnet.gr)
00:44:05amiconnpreglow: The frequency responses of the MAS treble & bass filters are given in the datasheet
00:44:13amiconn(but not the loudness curve)
00:44:36preglowamiconn: didn't we agree we'd let the sw eq do the job of the treble/bass setting anyway?
00:44:36belzethat clicking comes from changing the volume, so there must be some very deep noise
00:44:47LinusNwith the hw eq gains at 0dB, there is absolutely no difference if you enable it or not
00:44:56preglowbelze: sounds like i've got a dc offset going
00:44:56LinusNhardly surprising
00:45:09amiconnpreglow: Yes
00:45:14synichow can I get to the WPS context menu from the rockbox menu?
00:45:15preglowLinusN: considering what linuxstb said, not at all
00:45:19synicer... sorry
00:45:22synicthe WPS
00:45:58preglowamiconn: which mas datasheet?
00:46:00synicthere doesn't seem to be a way to get there unless that's the screen you were on before you got to the menu
00:46:12 Part belze_
00:46:49belzebut what i found out is it doesnt appear when you play a silent file right after booting rockbox
00:46:51amiconnpreglow: http://www.rockbox.org/twiki/pub/Main/DataSheets/mas3587f_2pd.pdf page 9
00:46:59belzeit only happens when you play a music-track before
00:47:47preglowbelze: and you're still sure it's caused by the eq? sounds more like a value being left in the dac to me
00:47:53linuxstb_LinusN: Are you going to fix the default gain settings, or shall I?
00:48:08belzepreglow: if i disable the sw eq, its gone
00:48:52preglowamiconn: looks like second order shelving filters
00:48:53LinusNlinuxstb_: i'll do it
00:49:21preglowbelze: could you tell me exactly how you do it? i'll try to reproduce on h120
00:50:32belzeok, i just played back some random song and after that played back a file consisting of only silence
00:50:54belzeboth were ogg-files if that is important
00:51:19 Quit KlrSpz ()
00:51:29preglowhaha
00:51:39preglowyou know you're cool when you actually have an ogg file of only silence lying around
00:52:15belzei know its a lossy compression
00:53:07preglowbelze: no dice, no sound at all
00:53:32belzebut when i play back that ogg file consisting of silnece right after booting the device this effect doesnt occure
00:53:53preglowogg is good with silence, 1kbit/s
00:54:39belzei created the file with audition if thats what you mean
00:54:44belzeits no recording or sth like that
00:54:53preglowthis reminds me, the eq filter coefficients should be zeroed on forced track change
00:55:10scottderHmmm ilounge is saying the new nano is using the new chipsets....I assume this is bad news for Rockbox?
00:55:23preglowscottder: indeed
00:55:37preglowbelze: i've got an ogg track of pure silence anyway
00:55:44belzeok
00:55:55preglowbelze: but i can't hear a sound when playing it
00:56:00preglowno matter what i play before it
00:56:06preglowwhat are your eq settings
00:56:07preglow?
00:56:22belze+2db on the lowest and +4db on the highest band
00:56:27scottderthat sucks :(
00:56:32LinusNso, this "amazing sound" when enabling the hardware eq comes from the fact that all bands default to +12dB. how's that for an audiophile sound experience? :-)
00:56:35belzei have the latest cvs build installed
00:56:43*scottder wants 8Gb Flash ROckbox love! :)
00:56:58preglowoh
00:57:00preglowi've got it now
00:57:04preglowdc output
00:57:05preglowi knew it
00:57:08*petur wants 80GB
00:57:19amiconnflash??
00:57:25 Quit bluebrother ("Leaving")
00:57:26*scottder likes flash too much now....
00:57:30peturif possible ;)
00:57:33scottderHDs in mp3 players do not like me....at all
00:57:35scottder:)
00:58:13amiconnpetur: It sure is... the price is another matter ;)
00:58:38preglowbtw, settings acceleration via play button in eq screen is broken
00:58:39peturI saw some announcement for 64GB CF cards ;)
00:58:44preglowactions probably did it
00:59:30amiconnWhy doesn't that surprise me??
00:59:44preglowbelze: definitely getting it
00:59:56belzeok, nice :)
01:00
01:00:03preglowweird, weird
01:00:13preglowdc level is constant
01:00:23preglowbut at some volume levels, i get some other sound as well
01:01:15preglowargh!
01:02:06preglowbelze: well, thanks for reporting, i know what's wrong now
01:02:14preglowand i don't bloody like it at all
01:02:50LinusNthat sounds bad
01:02:55preglowprecision problems
01:03:01LinusNouch
01:03:03preglowi'm almost certain it is
01:03:15preglowand this is using 32 bit wide coefs
01:03:23LinusNouch ouch ouch
01:03:45preglowi bumped into it doing the eq grapher and wondered why it wasn't biting me in the filtering
01:03:48preglowturns out it is
01:03:54preglowfilter coefficients are very sensitive at lower cutoff frequencies
01:04:36LinusN8.24?
01:04:39preglowyes
01:04:52preglowi need a ton of mantissa for the shelf filters
01:05:08LinusNwould 4.28 help?
01:05:10preglowi'll see if seperating the gain out of the coef will help
01:05:20preglow4.28 isn't enough to represent 24 db gain
01:05:23preglowit'd overflow
01:05:28 Quit ender` (" Don't worry about the world coming to an end today. It's already tomorrow in Australia.")
01:05:35preglowthat's why i use 8.24 for shelving, and 4.28 for peaking
01:05:43LinusNah
01:05:51 Quit petur ("here today, gone tomorrow")
01:05:57 Join midkay [0] (n=midkay@rockbox/developer/midkay)
01:05:57preglowi'm checking out what the coefs are turning out as now
01:08:24preglowaccording to matlab, the coefficients are right
01:10:03 Quit Kohlriba ("Quit")
01:10:18preglowall the coefficients are really close 1 and 2
01:10:51amiconnSlasheri: I tried the scheduler patch on Ondio and X5 now. It seems to work properly. Power consumption on Ondio didn't change within measurement uncertainty (measured the current).
01:11:11preglowy[n] = 1.004311*x[n] - 1.991301*x[n-1] + 0.987135*x[n-2] - 1.991355*y[n-1] + 0.991392*y[n-2]
01:11:11amiconnThe priority switching feels really odd on X5 though
01:12:04***Saving seen data "./dancer.seen"
01:12:37preglowbelze: still, did you make a wav?
01:12:41LinusNamiconn: "odd"?
01:12:56amiconnAnd I don't like the code size increase it causes. ~+680 bytes on archos, ~+1200 bytes on coldfire
01:13:22LinusNwell, 1200 bytes isn't much to argue about, is it?
01:13:24amiconnLinusN: Play some music, the start playing jewels, or run oscilloscope, or plasma, and you'll see what I mean
01:13:46LinusNamiconn: stuttering gfx?
01:14:27amiconnYes, and with plasma, I have the impression that it doesn't even boost before switching priorities
01:14:51amiconn(but the impression can be wrong and its difficult to check)
01:14:55LinusNaha, but it's not the kernel responsibility to boost, is it?
01:15:13preglowshouldn't be
01:15:18amiconnNo, but it's also not the kernel responsibility to switch priorities
01:15:29LinusNah, of course
01:16:00belzepreglow: i havent tried a silent wave-file, if you mean that
01:16:02synicanyone have an iPod video? How can I prevent the current track from stopping when I'm scrolling?
01:16:03amiconnI think +1200 bytes, while not being that important on swcodec in general, are quite a lot for the kernel
01:16:15preglowbelze: just thinking about you saying you were going to record the sound
01:16:30belzethat is the wave-file i sent you :)
01:16:41preglowbelze: yes, so i see...
01:17:13 Join safetydan [0] (i=cbca159f@gateway/web/cgi-irc/labb.contactor.se/session)
01:17:56preglowbelze: analogue or digital recording?
01:18:01preglowlooks analogue
01:18:14belzeanalogue, i dont have a device with an optical input
01:18:58safetydanHas anyone actually tested that the peak filters on the HW EQ are actually enabled correctly? Febs was saying they didn't work when I did the first patch and I didn't really change anything after that.
01:19:10preglowhahaha
01:19:13preglowthis keeps getting better
01:19:46 Quit lucaferr ("a CBIRC user")
01:19:52preglowsafetydan: btw, any idea why fsincos isn't accurate to within 31 bits? shouldn't it be when you do 31 iterations?
01:20:55safetydanpreglow, IIRC it's only adds "roughly one bit" of additional precision per iteration. You may need to a few more iterations to get to 31 bit precision
01:21:05preglowright
01:21:15LinusNsafetydan: i can't hear any difference when changing the peak filter gain
01:21:26preglowin which case i'm laughing out loud
01:22:27safetydanLinusN, someone also suggested we were looking at the wrong data sheet and that maybe there aren't even peak filters to enable
01:22:56LinusNhaha
01:23:25LinusNeven more hilarious
01:23:56safetydanpreglow, have a look at this https://svn.sourceforge.net/svnroot/astroinfo/trunk/astroinfo/source/FixedPoint.c
01:24:11 Quit safetydan ("CGI:IRC")
01:24:21 Join safetydan [0] (i=cbca159f@gateway/web/cgi-irc/labb.contactor.se/session)
01:24:40safetydanarm and m68k optimised fixed point operations
01:24:55safetydannot coldfire though, but might be close enough
01:25:09preglowsafetydan: wtf
01:25:18safetydan?
01:25:22preglowsafetydan: it's also about a jillion times faster than cordic
01:25:33safetydanwell there you go then :)
01:26:17 Part synic
01:26:39preglowi wonder what's up with fpmul, though
01:26:44preglowif we have to use that, then it's slower
01:26:48preglowby a good margin
01:28:10amiconnLinusN: Most of the added binary size is contributed by thread.c. It's .text part is almost exactly doubled, both on archos and coldfire
01:28:40preglowman, oh man, some people do know what fixed point is all about
01:28:44*amiconn thinks it's caused by this linked list juggling
01:29:09preglowthis .c file depresses me
01:30:20amiconnThe m68k fpmul is significantly slower than what we can do with emac
01:30:29safetydanthe .h file says the number format is S1.30 for fpmul
01:30:30LinusNpreglow: could it be that m68k only has 16x16 mul, but coldfire has 32x32?
01:30:49preglowLinusN: no idea, never did code for m68k
01:30:59 Join BHSPitLappy [0] (n=steve-o@adsl-68-95-246-22.dsl.rcsntx.swbell.net)
01:31:20LinusNi believe it is so
01:31:28preglowthen huzzah!
01:31:33safetydanIt's for PalmPilot so it's a DragonBall IIRC. Not sure what flavour of 68k that was and whether there where any extension to normal 68k
01:32:11amiconnPlain 68000 can only do 16x16->32 bit multiplications
01:32:38*amiconn has an old book about 68000
01:33:20safetydanpreglow, is the arm stuff any good? I have no idea what arm the palm pilots use
01:33:31preglowdidn't look at the arm stuff, grew far too depressed to look at it
01:34:24*amiconn wonders what is depressing within that file
01:34:34preglowhaha
01:34:37preglowfixed point mastery
01:34:45*preglow kicks the eq
01:35:02 Quit KN|stiff (Read error: 104 (Connection reset by peer))
01:36:01preglowi don't see why these coefficients should make trouble
01:36:31preglowbut i guess that's why i haven't made captain yet
01:37:32amiconnTalking about coldfire - do you think the various "32x32->64bit + some shift" implementations we are using suffer from an emac stall?
01:37:52amiconnIt's a suspicious sequence - mac.l followed by mulu ...
01:37:58preglowyeah, exactly
01:37:59preglowi think they are
01:38:07preglowthe mul instruction uses the mac unit
01:38:14preglowso it should stall
01:38:34amiconnThat's the big question - it all depends on how its implemented internally
01:38:34preglowthe freescale docs doesn't mention it, but it sounds logical
01:38:54amiconnMultiple mac instructions in sequence don't stall
01:39:31preglowsure no, it's pipelined to work like that
01:39:39preglowthey don't stall until you involve a moveclr or move.l %acc
01:39:59preglowbut i don't think mac/mul play that nicely together
01:40:14amiconnYes, and the muls/mulu include the move from the multiplier's output stage - that's why they take 4 cycles
01:40:23preglowexactly
01:40:33amiconnIf it's all pipelined mac.l/mulu.l shouldn't stall
01:40:44preglowi've got a feeling it stalls anyway
01:40:47preglowbut it should be possible to measure
01:40:55preglowwould be very interesting
01:41:25amiconnI did a (coarse) experiment where I could find no speed difference when moving the shift amount preloading between mac and mulu
01:41:39preglowhow coarse? the difference won't be huge
01:42:14amiconnI just ran mandelbrot and measured the time for zooming in 12 times in sequence
01:42:48amiconnHmm, not very scientific...
01:43:37preglowstill
01:46:00amiconnThe first zoom stages use low precision, which doesn't use emac...
01:46:28preglowsafetydan: i wonder what the 'remez' thing they mention is. the expressions look like standard mclaurin/taylor series expressions, and the only remez i know about is a fir filter design method
01:47:58 Join dan_a [0] (n=dan-mirc@217.23.173.156)
01:48:27preglownow, at a time like this, sound in the sim would be helpful indeed
01:50:17amiconnSound in the sim should work. It does here...
01:50:32amiconn(except mp3 on amd64...)
01:53:28preglowdoesn't work here
01:53:32preglowsome nice sdl glitch
01:53:36preglowwhich only happens in the rb sim
01:53:50preglowit used to work here before i reinstalled
01:54:17 Nick JoeyBorn is now known as JBonBike (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net)
01:58:14preglowtime to hit sack
01:58:17preglowlater
02:00
02:02:11 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
02:03:09 Part LinusN
02:10:34 Join ScoTTie_ [0] (n=scott@unaffiliated/scottie)
02:11:33 Join niskel [0] (n=niskel@fctnnbsc15w-156034073243.nb.aliant.net)
02:18:19 Quit spiorf (Read error: 60 (Operation timed out))
02:20:24 Quit ScoTTie (Read error: 110 (Connection timed out))
02:22:36 Quit lodesi ("leaving")
02:26:56 Join aegray [0] (n=aegray@74.135.163.80)
02:33:01amiconnWargh!
02:33:32*amiconn wonders what he did wrong with cvs
02:34:31*amiconn committed something not intended for commit, and with a wrong message
02:42:00 Join JdGordon [0] (n=jonno@c220-237-57-32.smelb1.vic.optusnet.com.au)
02:46:13 Quit ScoTTie_ (Read error: 110 (Connection timed out))
02:47:12 Quit dan_a (Read error: 110 (Connection timed out))
02:51:46LandusUh.
02:51:50LandusThe forums are dying.
02:52:27 Quit secleinteer (Remote closed the connection)
02:52:52scorchei heard the sky is falling as well
02:53:55LandusIt is?
02:54:08LandusHow fun.
02:54:16LandusWe can all die of oxygen deprivation!
02:55:09 Join secleinteer [0] (n=scl@adsl-70-237-207-174.dsl.stlsmo.sbcglobal.net)
02:58:17 Quit secleinteer (Read error: 104 (Connection reset by peer))
03:00
03:12:05***Saving seen data "./dancer.seen"
03:17:59 Join ScoTTie_ [0] (n=scott@unaffiliated/scottie)
03:26:47 Join |AhIoRoS| [0] (n=ahioros@201.224.122.227)
03:27:24 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
03:27:33 Join secleinteer [0] (n=scl@adsl-70-237-207-174.dsl.stlsmo.sbcglobal.net)
03:32:26 Join Kevinp [0] (i=JesterKi@d36-161-187.home1.cgocable.net)
03:32:30KevinpHey everyone
03:32:39Soap?
03:32:48KevinpI had a few pre-installation questions about Rockbox before I put it on my nano, was hoping you could help me out?
03:32:59Kevinpi searched the forum / documentation and coudln't find anything
03:33:23Galoisyes?
03:33:46Kevinpoh well, i was wondering (as ipodlinux doesn't do this)
03:33:59Kevinpwhen i begin playing a song and reverse through menus, will it still keep playing?
03:34:23Soapyes, you can access the menus while playing.
03:34:25 Join webguest32 [0] (i=c27d6d19@gateway/web/cgi-irc/labb.contactor.se/session)
03:34:36belzeyou can even run plugins while playing
03:34:37GaloisI assume you mean traverse. Yes you can traverse through menus while playing.
03:34:46 Join webguest08 [0] (i=7c5a93ae@gateway/web/cgi-irc/labb.contactor.se/x-c3ae20f831be749a)
03:34:49webguest32Is this the rockbox channel?
03:34:58webguest32This place is packed.
03:35:06belzeit is
03:35:07Kevinphm and .. i couldn't really find a _clear_ outline as to why rockbox is better than ipods actual firmware
03:35:22webguest08hi
03:35:34belzeKevinp: you have to decide that for yourself. a big advantage is the number of codecs it includes
03:35:42belzeso you can play ogg, flac, ...
03:35:58Kevinpperformance wise is it any better? in terms of things like ... battery life, sound quality etc?
03:35:59 Quit webguest08 (Client Quit)
03:36:13SoapKevinp - ogg / flac / other support. Crossfeed. GAPLESS. pluggins. customisable.
03:36:25belzei think battery life is not better, maybe even worse atm
03:36:29Soapworse
03:36:35Kevinphm
03:36:39Kevinpcrossfeed? gapless? ugh
03:36:41Kevinplemme google lol
03:36:48Soap60-70% of apple firmware. I'll have solid numbers this weekend.
03:37:18webguest32Sorry, but could someone answer one or two question I have please? I just deleted an old rockbox off my H140 (the old one was an experimental build and had a few errors)... anyway, I put the latest daily build on (sep 14) and it's working fine. Just two things really, how can I set it so when I hold left, it goes to the main menu, instead of having to press left a few times to get there. and another thing, ho
03:38:44 Quit Brock (Read error: 104 (Connection reset by peer))
03:38:57Soapgapless - Even on albums where track1 is supposed to blend seamlessly into track 2, the Apple firmware introduces a really frickin' annoying gap. Crossfeed is an audio effect where part of the right channel is added to the left, and vise versa. This is good as sunshine to use with headphones, as most (all?) modern albums are mastered for play through speakers where obviously your left ear can hear the right speaker's output.
03:39:12 Join ShadowdogMU [0] (n=Brock@cpe-65-28-252-205.woh.res.rr.com)
03:39:18Kevinphm
03:39:29belzewell, i dont really like the sound of crossfeed so far
03:39:36belzesounds a little blurry to me
03:39:41Soapwebguest32 - I don't believe you can set that at this time. Button assignments are not fully customizable.
03:39:52Soapbelze - not blurry at all.
03:40:34 Join Paul_The_Nerd [0] (n=Paul_The@cpe-70-112-165-230.austin.res.rr.com)
03:40:41belzenot blurry in means of quality, but i find it harder to locate the instruments
03:40:54Soapheadphones create a very unnatural sound. at a concert both ears hear the same thing, just at different times, and sometimes in different phases. With headphones your right ear can't hear the left channel at all. That is not only unnatural, it is not the situation taken into account when a CD is mastered.
03:41:19webguest32soap - are there patches available with these features because they were definitely on the one I had before I deleted it and I remember ages ago, before the one I justhad on, that I downloaded a patch for keeping the line selector in the middle while scrolling?
03:41:42Soapbelze - again, proper crossfeed can dramaticly improve soundstage and seperation as you are hearing the music the way the producer intended, not through the artificial sound of headphones.
03:42:19belzewell, then i have to play with the settings a little more
03:42:27Soapwebguest32 - I can't speak for certain, as button assignments aren't something that interest me. ;)
03:43:19Soapbelze - crossfeed is, and always will be an aproximation of how properly set up speakers will sound, but it is at least an aproximation.
03:43:55SoapOr you could seek out biaurial recordings, then headphones will be the right way.
03:44:18belzeyup
03:44:23Soap*biaural
03:45:33belzebut you are right on some recordings (keane) it sounds more natural
03:45:42SoapFor example, many live shows created by audience members are biaural recordings.
03:46:01 Join daurnimator [0] (i=quae@unaffiliated/daurnimator)
03:46:06Soapall my A Silver Mt Zion downloads from archive.org, for example.
03:46:06 Nick daurnimator is now known as daurn|afk (i=quae@unaffiliated/daurnimator)
03:47:06webguest32soap - ok no problem, thanks anyway.
03:50:02 Nick daurn|afk is now known as daurnimator (i=quae@unaffiliated/daurnimator)
03:51:03webguest32Does anyone know if roxkbox will be ported onto the new 80GB iPod? May finally be tempted to move on from my beloved H140 :( the 80GB hd is just too tempting. Is video available on the current ports to the 60GB iPod?
03:52:03belzevideo is only available without sound
03:52:38Paul_The_NerdWhy not get an 80gb drive for your H140 if you love it so?
03:54:23webguest32sure they are probably pretty much as expensive as the iPod itself?
03:54:57Paul_The_NerdWell, that somewhat remains to be seen.
03:55:16Paul_The_NerdIf the iPod's cheaper, you could always get the iPod, swap the drives, and use the iPod with the 40gig drive or something
03:55:56newbyx86So is there gonna be a RockBox for the Zune?
03:55:56newbyx86:D
03:56:04newbyx86I am gonna get them when they come out since my dad works for M$, so I can start working on one maybe. :P
03:56:11SoapIs there any reason to believe the 80GB ipod is any different than the 60GB (in ways which matter to RB)
03:56:17webguest32I don't really use the recording/optical/remote on the H140 anyway and if RockBox will be on the iPod, well a lot of gripes I've previously had with iPod (i tunes/taged broswing) would be gone. File Tree, for me, is the best way to browse.
03:56:28Paul_The_NerdSoap: There's no reason to believe it's any different other than possible a new LCD.
03:56:35belzenewbyx86: then start coding :)
03:56:39newbyx86;)
03:56:41newbyx86When it comes out. :)
03:56:43belzeand find out which chips are in there
03:56:45newbyx86hehe
03:56:47Paul_The_Nerdwebguest32: Well, the iPod battery life is still pretty bad.
03:56:50Soapwhereas the rumors (more than rumors) are that the Nano is samsung?
03:57:17webguest32Battery life on new 80GB is supposed to be 20 hours for music!
03:57:19webguest32not too bad
03:57:20belzefrom the facts i've heared i would get a zune right away
03:57:20Soap(should be a question mark after my second "rumors"
03:57:29Paul_The_NerdSoap: Well, Apple announced they'd be moving to Samsun. And the new Nano is both smaller, and sporting significantly improved battery life, so one has to assume *something* has changed internally
03:57:30belzethe wifi-module is so nice
03:57:40Soapwebguest32 - that is the same battery life as the 60 gig was quoted with.
03:57:41belzeand finally a player with an rds(!) receiver
03:57:51Paul_The_Nerdwebguest32: Not gonna be 20 hours in Rockbox. Expect maybe 8.
03:57:59Soap11
03:58:06Paul_The_NerdEleven now?
03:58:07Soapwith mp3 -aps
03:58:07Paul_The_NerdNice
03:58:23SoapI'll have the Apple Firmware benchmarks up this weekend for comparison.
03:58:31Soapand that's with a 32 meg build.
03:58:43SoapI'll add the 64 meg patch and try again next week.
04:00
04:00:20Soapwith lame -aps rips I don't expect 20 in apple firmware, either. Maybe 15-16 max.
04:02:24belzeSoap: do you know if the 5g ipod uses the same audio codec as the h10 and so also features that new hw eq?
04:02:31belzethe wiki says the h10 hs a Wolfson WM8731
04:02:44belzeand the one in the 5g ipod is also a wolfson afaik
04:02:47JdGordondoes anyone ellse have major problems with an irver h300 refusing to mount in linux?
04:03:00Paul_The_Nerd5G is a different wolfson
04:03:07Paul_The_NerdThe H10 uses the same one as the Mini I believe
04:03:10belzeah, ok
04:03:17belzeso no hw eq on that platform?
04:03:22Paul_The_NerdNope.
04:03:31Paul_The_NerdOnce everything's optimized the software EQ should be better anyway
04:03:50belzehehe, the default one in the h10 is quite bad
04:04:13Paul_The_NerdThe default one has nothing to do with Rockbox though
04:04:23webguest32gotta go, see ya
04:04:26 Quit webguest32 ("CGI:IRC")
04:04:26Paul_The_NerdRb has a quite nice software parametric EQ.
04:07:37belzedoes the h10 have any kind of hardware sound modification at all? like bass & treble?
04:08:09 Join Davide-NYC [0] (n=chatzill@user-12hdtm5.cable.mindspring.com)
04:08:43 Quit niskel (Remote closed the connection)
04:09:07 Part Kevinp
04:09:10Paul_The_NerdOut of curiosity, why the focus on HW sound modification?
04:09:29belzejust out of interest
04:10:08Paul_The_NerdI do believe it should have Bass and Treble.
04:10:13Paul_The_NerdBut I can't make a promise on that. ;)
04:10:21 Join dongs [0] (n=HPUX@i249222.ppp.asahi-net.or.jp)
04:10:53belzeok
04:11:12 Quit chromo41 (Read error: 110 (Connection timed out))
04:13:31Paul_The_NerdJdGordon: Could you please type out whole words on the forums. The posting guidelines do apply to developers as well.
04:13:46JdGordon:(
04:13:56JdGordonmeany!!
04:14:09Paul_The_NerdYes. Yes I am.
04:14:33dongsyea, 'u' is not a word
04:14:57 Join ScoTTie [0] (n=scott@unaffiliated/scottie)
04:15:04JdGordonI was actualy wondering which word i could have shortend... i guess i really am just lazy :D
04:16:52scorchedongs...is back?
04:16:52 Part pixelma
04:20:10 Join vertic23 [0] (i=bla@dslb-084-056-239-090.pools.arcor-ip.net)
04:22:34Soapis he, scorche? My /ignore list gets him.
04:23:08scorcheyeah...he is
04:23:12scorchesad eh?
04:23:29vertic23...me?
04:23:42scorchenope
04:23:47vertic23*phew*
04:24:17Soapahh, mandlebrot crashed.
04:24:20 Quit ScoTTie_ (Read error: 110 (Connection timed out))
04:24:42SoapMenu+Select to the rescue.
04:29:01Davide-NYCI saw a post on the forum about a manual make option to make a text only version of the manual for reading on the target.
04:29:09Davide-NYCI thought this was a great idea.
04:29:17Davide-NYCWhat does everyone think of this?
04:29:42Davide-NYClike a "make manual-text
04:31:15SoapBrowsing a 130 page (printed) document would be painful on the players.
04:31:26Soapesp. w/o hyperlinks.
04:32:33 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
04:33:42LandusIs anyone here familiar with C and the new patches for the new theme that Julius released?
04:33:58LandusGetting a compiling error and I don't possess the knowledge to fix it.
04:33:58 Join RoC_MM [0] (i=dragon@c-66-177-58-16.hsd1.fl.comcast.net)
04:35:23 Quit actionshrimp ("( www.nnscript.de :: NoNameScript 4.01 :: www.XLhost.de )")
04:35:46Davide-NYCSoap: true. maybe just the basics. No plugins. No games?
04:36:05Davide-NYCno gnu agreement
04:36:39Paul_The_NerdOr maybe improve the Text reader to be a basic HTML reader, so it can support the links. ;)
04:37:05Davide-NYCIt only 50 pages.
04:37:13Davide-NYCwithout screeshots I bet half that
04:37:31Davide-NYC25 pages doesn't sound terrible, even on the player.
04:37:32Paul_The_NerdSome of the most common questions are about plugins though.
04:37:33 Quit ScoTTie (Read error: 60 (Operation timed out))
04:37:42Soapmmm, HTML reader on rockbox.
04:37:44Paul_The_NerdAnd the gnu agreement really does need to be there.
04:37:50Davide-NYCurghh
04:37:56Paul_The_NerdPlus, there's been work toward an HTML reader in the form of the Rockipedia effort.
04:38:05Davide-NYCRockipedia?
04:38:09Davide-NYClinkee please
04:38:11Paul_The_NerdI don't know how far they got, but I remember reading some stuff about getting links working
04:38:22 Join ScoTTie_ [0] (n=scott@220-253-115-170.VIC.netspace.net.au)
04:38:28Paul_The_Nerdhttp://rockipedia.techmight.com/index.php?title=Main_Page
04:38:55Davide-NYCcrazy!
04:49:29SoapPlucker for rockbox would be just as nice, if not nicer than HTML support.
04:49:36 Quit amiconn (Read error: 110 (Connection timed out))
04:49:36 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
04:52:58 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox 1.5.0.6/2006072814]")
04:53:57goffais there sound on the gigabeat f40 yet?
04:54:06billytwowillyhow's the e200 rockbox support work going?
04:55:08goffahmm.. aparently not
04:59:24 Quit cdmackay ("bye all")
05:00
05:00:30Paul_The_NerdYou can expect any progress on "New Ports" to be reported in their official thread.
05:01:19 Join afruff [0] (n=afruff@pool-72-83-123-16.washdc.fios.verizon.net)
05:01:20 Join Rob2222 [0] (n=Miranda@ACB4735E.ipt.aol.com)
05:01:49afruffis the X5's reset button HW-based or SW-based?
05:02:03 Join ScoTTie [0] (n=scott@unaffiliated/scottie)
05:02:15afruffIF SW-based, does Rockbox modify any code controlling the reset button?
05:05:43 Quit netmasta10bt (Read error: 54 (Connection reset by peer))
05:07:11 Join webguest24 [0] (i=45dd7f9e@gateway/web/cgi-irc/labb.contactor.se/session)
05:10:37 Quit webguest24 (Client Quit)
05:11:49 Quit ScoTTie_ (Read error: 110 (Connection timed out))
05:12:06***Saving seen data "./dancer.seen"
05:12:37RaiderXhey i upgraded to the latest rockbox build for 5g ipod, when i hold down menu to load up the apple firmware, it says its booting and then it just restarts and wont load the original firmware :/ i didnt apply any patches or anything. maybe its just a bug in the build?
05:13:27Paul_The_NerdUpgrading to the newest Rockbox doesn't include changing the bootloader
05:13:41Paul_The_NerdThere's no reason that functionality should change as long as you didn't touch the bootloader.
05:13:54RaiderXthats wut i thought
05:13:54Paul_The_NerdI do assume you followed the instructions and *just* extracted a new rockbox.zip, and didn't redo the ipodpatcher stuff?
05:14:04RaiderXwell uno wtf happened then lol
05:14:21Paul_The_NerdI have no idea what "uno" means.
05:14:26RaiderXdunno
05:14:55RaiderXyea it prolly got messed up when i upgraded the apple firmware cuz then i had to uninstall rockbox and reinstall it. so ill just look into it more tomorrow
05:15:00BHSPitLappyyou shortened down an already shortened piece of slang?
05:15:05RaiderXyea lol
05:15:08RaiderXiuno
05:15:13Paul_The_NerdRaiderX: It sounds like you didn't reinstall the bootloader properly then
05:15:20RaiderXyea
05:15:36RaiderXwell im off to bed for now anyway, cya
05:15:45 Part RaiderX
05:15:48BHSPitLappyI wonder if the iPL Installer2 can install -only- the bootloader
05:15:55Paul_The_Nerdhttp://gallery.ilounge.com/ipod/thumbnails.php?album=61&page=1 <−− Dissected Nano 2G
05:16:03BHSPitLappythat would be easier for a lot of rockbox beginners
05:16:54Paul_The_NerdWell, Rockbox will eventually have a point 'n click installer of some sort, once it's actually in a release state
05:17:02Paul_The_NerdHrm... Don't see a chip labelled PP anything
05:17:19scorcheugh...site is loading slow
05:17:46 Quit Rob2222_ (Read error: 110 (Connection timed out))
05:17:57Paul_The_NerdThere's a chip with an Apple logo that says "Arm" on it.
05:18:02Paul_The_NerdAnd various numbers.
05:18:25scorchei see that also...
05:18:30 Quit |AhIoRoS| ("Abandonando, see you http://ahioros.homelinux.net")
05:18:52scorcheugh...no hi-res scans at all
05:18:57Paul_The_NerdYeah.
05:19:01Paul_The_NerdIt *is* iLounge.
05:19:13Paul_The_NerdIt's like iPod Porn for those people. Not really concerned about technical merit. :-P
05:19:28scorcheyeah...the pictures look nice and all...nice pose
05:19:35Paul_The_NerdAnyway, I'm gonna go out on a limb here and bet that Rockbox won't work on it. :)
05:19:50scorchefrom what it looks like
05:20:25BHSPitLappyyou know, I wonder why no 3rd party ever made replacement cases, for those willing to disassemble their iPods
05:20:41scorchecan you read the chip to the left of the apple chip?
05:20:45Paul_The_NerdNaah
05:20:50BHSPitLappysomeone could have made these aluminum cases for the 1G nanos already
05:20:51Paul_The_NerdI'm fairly certain there's no PP on it
05:21:12scorchealthough, that apple chip seem slike it has to be the processor with arm and all of the transistors
05:21:18Paul_The_NerdBHSPitLappy: I've seen re-cased iPods on sale, with various things like brushes steel casing, or gold, or wood.
05:21:31Paul_The_Nerds/brushes/brushed
05:22:08BHSPitLappycool
05:22:16BHSPitLappywell I did see the infamous wooden iPod mod
05:22:17Paul_The_NerdOf course, that could just be a rebranded PP chip to throw off hackers, or something
05:22:19BHSPitLappywhich was awesome
05:22:54Paul_The_NerdBut it seems more likely its the new Samsung one that was supposed to be happening.
05:23:30 Join u238 [0] (n=criminal@24-117-13-130.cpe.cableone.net)
05:23:38u238hey whats up
05:23:40u238im kinda drunk
05:23:41scorchei thought that it was the samsung chip, but Bagder said that is too major of a change
05:23:42u238but i got a couple questions
05:24:06Paul_The_NerdWhy would it be too major, I wonder?
05:24:09 Join Xadow [0] (i=c9f39883@gateway/web/cgi-irc/labb.contactor.se/session)
05:24:10scorchegoogle search on the numbers yielded no info (surprise!)
05:24:15Xadowhi ppl
05:24:15Paul_The_NerdYeah, tried that already. :)
05:24:56u238im sick and tired of the random folder names that itunes/ipod puts for the music
05:25:06Paul_The_NerdAh well, I suppose time will tell us what the Apple branded chip is.
05:25:09u238does rockbox does this?
05:25:12u238wow
05:25:17u238does rockbox do this?
05:25:25Paul_The_Nerdu238: Rockbox allows you to access your files in filetree mode if you want.
05:25:32u238well
05:25:38Paul_The_NerdWhich means you can just drag and drop folders of music onto it, and browse them with their real names
05:25:40u238i just bought a new car stereo
05:25:40scorchewell, from the sounds of things, PP doesnt even guve manufacturers datasheets, so im sure they had to use some proprietary tech to create bootloaders and such...they would have ot change for samsung, but i can see them having plenty of time for that
05:25:40BHSPitLappyPaul_The_Nerd, I thought samsung just made the storage chip.
05:25:43Xadowi doubt rockbox does it being a proyect with fully-customizable
05:25:47Xadowpurpose
05:26:10scorcheBHSPitLappy: the deal was for processors as well
05:26:13Paul_The_Nerdscorche: I doubt *that* much would have to change really. Just the drivers for the SoC stuff
05:26:17Xadow(@u238)
05:26:35Paul_The_NerdBHSPitLappy: There was an announcement at some point that the next Flash-based iPod would no longer be Portalplayer. I *think* they named Samsung, but I cant remember for sure.
05:26:35BHSPitLappywhat SoC stuff?
05:26:38scorchePaul_The_Nerd: which is why i was wondering about his too major of a change thing as well
05:26:40BHSPitLappyah.
05:26:51u238my new car stereo: http://www.bestbuy.com/site/olspage.jsp?skuId=7689129&type=product&id=1141762608953
05:26:58Paul_The_NerdBHSPitLappy: System on a Chip. The fact that USB is controlled by the PP chipset on the iPods, and so on.
05:27:01scorchePaul_The_Nerd: they did..i looked it up for midk recently because he hadnt heard about it
05:27:03u238and i can plug my ipod in via usb right in the front
05:27:06u238and it reads the folder names
05:27:10u238like it should
05:27:18u238but my ipod has a bunch of random folder names
05:27:19Paul_The_Nerdu238: As I said, Rockbox allows you to use folder names.
05:27:19BHSPitLappyoh.
05:27:37u238ok cool
05:27:40u238thanks
05:27:41u238one more
05:27:55u238can i see some damn screen shots? there are none on the site
05:27:58u238that i an see
05:28:01u238can*
05:28:02Paul_The_Nerdscorche: Okay. I remembered they named *someone* but I'm terrible with names.
05:28:05BHSPitLappyu238, it sounds like this doesn't even have to do with rockbox
05:28:10Paul_The_Nerdu238: Check the manual.
05:28:14u238ok thanks
05:28:16Paul_The_Nerdu238: Or the WPSGallery
05:28:17u238i will
05:28:20u238ok
05:28:21u238thanks
05:28:33Xadowi was really wondering about the video
05:28:34u238BHSPitLappy, read again
05:28:39scorchei hate playing the waiting game =(
05:28:42u238im inquiring
05:29:13Paul_The_NerdXadow: You mean the "5.5G" iPod?
05:29:16Xadowis the size on the converting instructions the only size the video can be?
05:30:18Paul_The_NerdNo, but it's the ideal size since the video won't be scaled.
05:30:31Paul_The_NerdWell, an ideal size
05:30:31u238does the FM radio work on ipods?
05:30:41Paul_The_Nerdu238: iPods don't have built in FM radios
05:30:47u238k
05:31:02u238do you have prebuilt responses?
05:31:04u238lol
05:31:11scorchehttp://www.tgdaily.com/2006/04/27/samsung_cliches_ipod_processor_contract/
05:31:17scorcheu238: he is a bot
05:31:21u238oh
05:31:23u238dangit
05:31:28u238i feel like a tool
05:31:32u238;
05:31:35u238;/
05:32:20XadowConvert your movie file to an AVI file, uncompressed, and with the size: 112x64.
05:32:24XadowT_T
05:32:36Paul_The_NerdXadow: What page are you looking at?
05:32:43Paul_The_NerdFor the RVF Player?
05:33:00Paul_The_NerdThat page is only valid for the Archos bitmap-screened players.
05:33:12Xadowi thought that rockbox would allow me to get higher video sizes then the ipod's default
05:33:20Xadowi see >_<
05:33:33Paul_The_Nerdhttp://www.rockbox.org/twiki/bin/view/Main/PluginMpegplayer
05:33:34scorcheXadow: the mpegplayer plugin is a work in progress
05:33:58Xadowany way an amateur can help with it?
05:34:08Paul_The_NerdAlso, Rockbox doesn't support output to TV, so at the moment, there's no reason to support resolutions higher than the screen size.
05:34:27scorchePaul_The_Nerd: apparently, according to that article, it isnt quite a straight processor changeout planned...this is because the portalplayer's chip influenced other design decisions as far as other components
05:34:45scorcheXadow: sadly, not really
05:35:10scorcheso it *could* be pretty big
05:35:46scorchebut it is still too early for all of these speculations...as i said...i hate playing the waiting game =P
05:35:50Paul_The_Nerdscorche: I saw that. Considering the size decrease and battery life increase, it seems *more* likely to me now. :)
05:36:05scorcheyeah
05:36:45Xadowactualy, suporting playback of a 1024x movie would be important, unless the size doesn't apply to the playback but just to the size it can be displayed
05:37:00scorcheoh yeah...i was watching a movie
05:37:06scorchePaul_The_Nerd: thanks for the distraction =)
05:37:26Paul_The_NerdXadow: It's rather unlikely the iPod will ever be able to do a 1024 by Anything movie
05:37:38Paul_The_NerdAlso, why is it beneficial or important at all, if you can resize the movies before putting them on it?
05:38:05Xadowbecause i could always keep the original size without using extra space
05:38:31 Quit u238 ()
05:38:55Paul_The_NerdHow does having the file take up at a minimum three times the space (on the iPod) as it has to save space, unless you're planning on keeping your only copy of the file on a portable device?
05:39:07BHSPitLappyXadow, you have a collectionof HD video? that you want to keep on your mp3 player??
05:39:43BHSPitLappyPaul_The_Nerd, I think he wants to have both accessible on the drive either way... and having one file serve both purposes is less space
05:40:09Paul_The_NerdTrue, but I wouldn't want my high definition video in MPEG2.
05:40:15XadowBHSPitLappy: could you fill your mp3 player with mp3 songs you will actualy listen?
05:40:17BHSPitLappynope...
05:40:29BHSPitLappyXadow, but you actually do have HD video?
05:40:47Xadowwhat do you mean with HD?
05:40:50XadowT_T
05:40:54BHSPitLappyhigh definition.
05:40:54Paul_The_NerdHigh Definition
05:41:04Xadow*ignorant*
05:41:04Paul_The_Nerd1280x720, 1960x1080
05:41:05BHSPitLappy1024x??? is high than DVD.
05:41:12BHSPitLappys/high/higher/
05:41:32Paul_The_NerdDVD is / should be 720x480 or 720x something lower if it's 16:9, right?
05:42:08Xadowi see
05:43:09Paul_The_NerdBut the iPod really isn't very powerful at all
05:43:19Xadowi bet
05:44:09Xadowwhat would you guys really recomend me for a high cuality player for video and lossless stuff
05:44:11Xadow?
05:44:42Paul_The_NerdIf you want to actually play video, one of the Archos PVRs
05:44:46Paul_The_NerdLike the 500 series maybe.
05:46:33Xadowfor the 112x64? doesn't realy sound good if i want to share it later
05:46:39Xadow*really
05:46:45Paul_The_NerdLossless on a portable is mostly a silly concept unless you're feeding optical out into a better quality DAC than most of them would come with.
05:46:57Xadowbut i want to do anything =P
05:46:58Paul_The_NerdThe Archos PVRs are not Rockboxable players
05:47:22Xadowdoesn't sound good =S
05:47:26Paul_The_NerdRockbox is focused on Audio Players, and so far most audio players can't really do high quality video.
05:47:33Xadow"rockboxable" ROFL
05:49:28Xadowbut what doesn't sound good is the "not" part of the sentence
05:49:56Xadowi have been looking to the proyect once in a while, i like it
05:50:26belzeits a great project
05:50:40belzeif your player is fully supported of course ;)
05:50:51 Quit sharpe ("Leaving")
05:51:32Xadowi was thinking on getting an 80 gig ipod to listen to whatever i want, whenever i want
05:51:33Paul_The_NerdWell Rockbox focuses on Audio. There's not a "good" video player that Rockbox runs on.
05:52:28belzeXadow: well, you think you can really fill that space with an efficient codec like ogg vorbis?
05:52:29Xadowand i have to admit i don't really want HD stuff, i just want to see the videos on the cuality they originaly are
05:52:43Xadowjust wondering the power of the devices you knew
05:53:36Xadowogg vorbis? meh... i love flac... since ape isn't suported tue to hardware limitations
05:53:44Xadow*due
05:54:21belzewell, as Paul_The_Nerd already said you wont need flac on a portable device
05:54:30Xadowi care about audio the most, hehe
05:54:30belzeor any other lossless codec :)
05:54:45LandusPaul.
05:54:54belzei also do, thats why i keep my music in flac on my pc
05:54:54Xadowso... it just won't be usefull, huh?
05:54:55Paul_The_NerdSeriously, unless your portable device has optical out, and you're running it into a *much* better DAC than the one in the device, you're losing plenty of quality just from the components anyway
05:55:06belzebut i convert it to ogg vorbis when i need it on my iriver h110
05:55:13LandusIt was picking up files in the /Recycled folder.
05:55:23LandusThe playlist trick showed as such.
05:55:24Paul_The_NerdLandus: So, checking the playlist solved it? :)
05:55:29LandusYes. Thanks.
05:55:40Xadowgratz
05:55:52Paul_The_NerdXadow: Basically, high quality Ogg/Vorbis or MPC shouldn't be distinguishable from lossless on a DAC except in the cases of artifacts.
05:56:21scorcheLandus: ha!...i thought you said that you knew for sure that there were no duplicates on the device =P
05:56:23Paul_The_NerdBut if you do want to use FLAC, you should go with the H140, that way you can at least hook it up to higher quality gear when/if it's available.
05:56:29LandusI did.
05:56:55LandusExcept that apparently, the way I checked didn't show the Recycled folder.
05:57:55Xadownow, when there is lost cuality of audio on the device, wouln't there be a larger cuality loss with lower cuality audio?
05:58:40belzeXadow: the thing is, that the components like the dac/amp/... cause more loss in quality than converting the files to a lossy format
05:59:11LandusWell. One problem solved.
05:59:14Xadowwould it be 99% unoticeable or would it just be the same thing anyway?
05:59:26belzehmm, does mp3 really work without boosting on the h1xx series?
05:59:38LandusOne more to go.
05:59:52XadowLandus: may i help you?
05:59:53Paul_The_NerdXadow: I will bet that whatever device you choose, if you were to attempt to ABX *good* lossy files vs. lossless files you would not be able to.
06:00
06:00:19LandusThough, it's more of a new build with Julius' new set of themes and patched.
06:00:32LandusI'm getting a compling error.
06:00:42Landushttp://www.rockbox.org/tracker/task/5900
06:00:45LandusLast comment.
06:00:51 Part afruff
06:01:06Paul_The_Nerdscorche: Almost definitely a samsung processor: http://forums.rockbox.org/index.php?topic=6518.msg50576;topicseen#msg50576
06:01:24Paul_The_NerdThere's going to be some work before the Nano 2G has a chance of being supported.
06:01:35scorcheyeah...i am writing a post in there sharing the pics you shared in here =)
06:02:10Xadowmeh... for helping here i'm a good SRO player...
06:02:18safetydanbelze, either no boosting or very close to it
06:02:50Paul_The_NerdLandus: Did any of the hunks fail while patching? I would hazard a guess that the part of the patch that modifies the function didn't work, which is why the function call isn't working.
06:03:50*Xadow thinks that it sounds coherent
06:03:54Xadowhahahaa
06:03:59safetydanLandus, the error tells you pretty much what you need to know. The setmargins function is out of sync with that line some how. You'll probably have to wait for someone to update the patches.
06:04:03belzesafetydan: i am also wondering about the performance speed-ups on mpc, eats less cpu than ogg vorbis
06:04:27belzealmost no boosting
06:05:39safetydanbelze, yeah musepack got a lot of optimisation work when it was added
06:05:46Paul_The_NerdLandus: In pijulius' version of the patch, that call has only two parameters. You're using the one modified by Max Weninger, which requires some other change apparently, though he doesn't actually tell you what
06:06:28Paul_The_Nerdsafetydan: It's actually that someone created a second version of the patch, that seems to have a prerequisite of some sort that calls that function with 3 parameters instead of 2, and so he's using the wrong patchfile from the looks of it.
06:06:47Xadowwhy would someone need less cpu usage? longer batery life?
06:06:54belzesafetydan: i read a while ago, that lossless codecs like flac would even work with 11mhz on the iriver hxxx. do you know if that will be introduced into cvs sometime soon?
06:06:59BHSPitLappyso do we have any tests of rockbox vs. 5.5g or nano2g?
06:07:20belzeXadow: exactly
06:07:25Paul_The_NerdBHSPitLappy: The Nano2G is almost definitely a no-go now.
06:07:35Xadowgood...
06:07:41Paul_The_NerdBHSPitLappy: It would require some amazing work of fate for the Samsung audio codec to work the same as the wolfson. ;)
06:07:50BHSPitLappyoh?
06:08:19Paul_The_NerdBHSPitLappy: Check the Nano2G thread in the forum, it seems someone's reasonably certain about the processor. The part number for it matches the labelling at least.
06:09:02BHSPitLappydoes that mean it won't boot, or the audio codecs will be messed up
06:09:11LandusSorry. Had to move away from the computer.
06:09:18Paul_The_NerdBHSPitLappy: It won't boot, and when it does boot, some work will have to be done to make it have sound.
06:09:36BHSPitLappysucks
06:09:53BHSPitLappyshould just put a 1g's internals into a 2g's casing, then
06:09:54LandusYes, a few hunks failed, but I manually put in the changes that didn't go through found in the .rej files.
06:10:19Paul_The_Nerdbelze: There's probably only minimal benefit in slowing the processor down that much. The user interface would probably get very slow.
06:10:32belzeah, ok, i see
06:10:38scorchePaul_The_Nerd: were you able to find any specs at all for the new processor, like pp's "press releases" or whatever?
06:10:47Paul_The_NerdLandus: As I said, the real problem is that you're using a modified version of that patch that doesn't modify everything it needs. Whoever created the patch had some changes he didn't include in it.
06:10:55safetydanbelze, no idea
06:11:07LandusJulius added the patch himself.
06:11:17XadowLandus: Translation: use another one
06:11:26Landussafetydan: My knowledge of C is rather.. minimal. Dunno what you were saying.
06:11:40Paul_The_Nerdscorche: Nope.
06:11:48Paul_The_NerdLandus: You're not using the Julius one.
06:11:58LandusHrm.
06:12:09Paul_The_NerdLandus: You downloaded the Max Weninger one
06:12:10 Quit vertic23 (Read error: 104 (Connection reset by peer))
06:12:12safetydanLandus, try following Paul's idea
06:12:20safetydanotherwise wait for someone to regenerate the patch properly
06:12:22LandusGuess I'll find out.
06:13:10 Join vertic23 [0] (i=bla@dslb-084-056-242-084.pools.arcor-ip.net)
06:13:27Paul_The_Nerdscorche: Samsung's site is *very* slow, S5L8701B05 came up with two google results.
06:13:41scorchePaul_The_Nerd: yeah...i am browsing their site now
06:14:19Paul_The_Nerdscorche: And the only page they have that google finds mentioning S5L8701B05 is their Eco-product compliance information from the look of it
06:15:02scorchePaul_The_Nerd: not listed on samsung's site under either of their arm7 or arm9 products
06:15:24Paul_The_NerdProbably not publically on sale yet, or something?
06:15:37scorchepossibly
06:16:22scorchehttp://www.samsung.com/Products/Semiconductor/Support/Label_CodeInfo/SystemLSIProductCodeConcept.htm
06:17:18scorcheinteresting..
06:17:51 Quit Nibbier (Read error: 110 (Connection timed out))
06:18:08Paul_The_NerdIt's a version B, BOM, Feature/serial 8701, small class 'L', mask option 05, apparently
06:18:15 Join Nibbier [0] (n=sven@e181098072.adsl.alicedsl.de)
06:18:19Paul_The_NerdErr MOS, not BOM
06:18:26scorcheyeah...
06:19:54*Paul_The_Nerd wonders how many hoops would be needed for datasheets
06:20:11scorchefor something that isnt even on their site yet
06:20:29Paul_The_NerdI suppose I'd have to wait until it's posted.
06:20:46Paul_The_NerdAsking before then would be like sending them an email saying "Hey, I want to reverse engineer the Nano 2G. Can you guys help?"
06:21:53XadowROFL
06:22:55Xadowmeh... do you guys send out any expresions?
06:23:13LandusAlright.
06:23:25LandusApplied all the patches from scratch and making now.
06:23:46LandusOne hunk error, but I was able to make the change manually.
06:25:12LandusDamn.
06:25:50XadowRelax, Breathe deep, Try again
06:26:17LandusDifferent error.
06:26:46LandusCan't copy/paste out of cygwin, so I'll have to link to a snapshot. Don't feel like typing the error out.
06:27:40Xadowtried highlight>ctrl+c ?
06:27:49LandusCan't highlight.
06:28:01Landushttp://www.frontiernet.net/~magic360/cygwinerror.jpg
06:28:51Paul_The_NerdAre those changes you made yourself, or relating to a specific patch?
06:29:04LandusNothing I made myself.
06:29:18Xadowdeclare the font stuff...
06:29:21Paul_The_NerdWhich patch?
06:29:21LandusThis is a new build from scratch. That must've been added by one of the patches.
06:29:40Paul_The_NerdIt looks like once more a patch had outside requirements.
06:30:52 Quit rotator ("zzzzzzzzzz")
06:31:15safetydanLandus, if you could find the include file that declares FONT_UI you could add a #include for it gui/statusbar.c
06:31:24safetydanI'd help but I don't have the source in front of me
06:31:28LandusHow would I do that?
06:31:37Landus#include then what?
06:31:49safetydanLandus, that's what you'll need to find
06:32:02safetydanit might be something like #include "font.h" but I could be wrong
06:35:25LandusThere doesn't appear to be a font.h file
06:36:29scorchePaul_The_Nerd: the samsung chips have quite a bit built in to them...
06:36:55scorchethe fastest ARM7 chip they have on their site is 66mhz as well...wonder how this chip is
06:37:04LandusWhere is font.h?
06:37:44safetydanLandus, it's under firmware/export
06:37:46Paul_The_Nerdscorche: 66mhz might even be enough for a Nano. Especially if it's dual core like the PP was.
06:37:50safetydanbut #include "font.h" should do it
06:38:11scorchePaul_The_Nerd: i didnt see any dual-core chips on there, but you never know
06:38:16Paul_The_NerdAh.
06:38:28LandusThat line is already in statusbar.c
06:38:42Paul_The_NerdWell, we have MP3 and everything already working at 75mhz, so 66 doesn't seem a huge stretch.
06:39:06scorche...once we optimize the other codecs too
06:39:25Paul_The_NerdYeah
06:39:38safetydanLandus, not sure then
06:39:49safetydanLandus, bug the original patch creator and get them to up date it
06:40:17LandusI dunno which patch is causing this.
06:40:28safetydanLandus, try one at a time until it fails
06:40:43LandusI'll have to do that after school.
06:40:52LandusI should've been asleep several hours ago.
06:45:16Xadowi could say that me too, but i don't use to sleep early
06:45:26Xadowbut i'm out right now, have fun!
06:46:32Xadowif you like to...
06:49:00 Join ScoTTie_ [0] (n=scott@220-253-62-88.VIC.netspace.net.au)
06:58:24 Quit ScoTTie (Read error: 110 (Connection timed out))
07:00
07:02:43 Join netmasta10bt [0] (n=torment@pool-71-251-99-29.tampfl.fios.verizon.net)
07:05:21 Part Paul_The_Nerd
07:06:04 Quit Xadow ("CGI:IRC (EOF)")
07:06:04 Quit safetydan ("CGI:IRC (EOF)")
07:06:04 Quit webguest63 ("CGI:IRC (EOF)")
07:12:11***Saving seen data "./dancer.seen"
07:22:42 Nick daurnimator is now known as daurn|afk (i=quae@unaffiliated/daurnimator)
07:31:04 Quit ScoTTie_ (Read error: 110 (Connection timed out))
07:33:14 Quit netmasta10bt (Read error: 60 (Operation timed out))
07:41:15 Join ScoTTie [0] (i=skotty@220-253-111-112.VIC.netspace.net.au)
07:46:33 Quit Lear ("Chatzilla 0.9.74 [Firefox 1.5.0.5/2006071912]")
07:47:01 Join Davide-NYC [0] (n=chatzill@user-12hdtm5.cable.mindspring.com)
07:47:12 Quit Davide-NYC (Client Quit)
07:48:11 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
07:48:33 Quit RoC_MM ("Leaving")
07:49:01 Join theli_ua [0] (n=theli@mail.ukrcard.com.ua)
07:52:18 Join ScoTTie_ [0] (n=scott@220-253-111-112.VIC.netspace.net.au)
07:55:17 Join mantono [0] (n=mantono@c83-250-204-173.bredband.comhem.se)
07:57:37 Join BHSPitLappy [0] (n=steve-o@adsl-68-95-246-22.dsl.rcsntx.swbell.net)
08:00
08:02:14 Quit ScoTTie (Read error: 110 (Connection timed out))
08:02:45 Join hcs [0] (n=agashlin@nat01-silvers-ext.Rutgers.EDU)
08:20:29amiconnbelze: WM8731 has no hardware tone control at all
08:20:39 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
08:25:27 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
08:26:33 Join BHSPitLappy [0] (n=steve-o@adsl-68-95-246-22.dsl.rcsntx.swbell.net)
08:27:18 Join ender` [0] (i=null@84.52.165.220)
08:29:36 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
08:30:57belzeamiconn: ok, thx for the answer
08:31:57 Join BHSPitLappy [0] (n=steve-o@adsl-68-95-246-22.dsl.rcsntx.swbell.net)
08:33:19 Quit BHSPitLappy (Remote closed the connection)
08:33:51 Join sando [0] (i=lolsteam@144.135.255.155)
08:35:22 Join BHSPitLappy [0] (n=steve-o@adsl-68-95-246-22.dsl.rcsntx.swbell.net)
08:35:57 Join chendo_ [0] (n=chendo@203-206-53-130.dyn.iinet.net.au)
08:38:05theli_uaanyone knows of some fast scaling technique ...? so that can be used in emulators to fit different targets' screens ....
08:38:14 Quit chendo (Read error: 60 (Operation timed out))
08:38:47BagderI believe Rockboy simply skips every N line
08:39:37theli_uayes i know ... it won't fit any screen ...
08:41:06theli_uai now use smth like precalculated step to skip lines .... it works very fine on my mini but is not fast enough on color targets like h300
08:51:36 Nick Nibbier is now known as Nibbler (n=sven@e181098072.adsl.alicedsl.de)
08:52:44BHSPitMonkeyit'll fit the 5G
08:57:51theli_uaha ha
08:59:57Zagorsooo, hands up all you who are going to preorder a brown zune?
09:00
09:00:16ZagorI always thought there's too little brown plastic in consumer electronics
09:00:44hcsI do love my old woodgrain 8-track stereo receiver...
09:01:53Zagorwoodgrain is for losers. this is plain, even-colored, classy, brown plastic: http://www.microsoft.com/presspass/presskits/zune/images/image002_low.jpg
09:04:51 Join dan_a [0] (n=dan-mirc@217.23.173.156)
09:05:48Bagderso the Nano 2G is Samsung based after all
09:05:49 Quit Spida (Read error: 104 (Connection reset by peer))
09:05:56scorcheyup
09:06:18Bagderit makes me wonder what the video 5.5 is
09:06:19 Join Spida [0] (n=timo@spinnennetz.org)
09:06:34 Quit ShadowdogMU ("You know you'll miss me :P")
09:07:02scorcheconsidering it was just the firmware, lcd, and hd capacity changed, i am thinking it is still PP
09:07:41Bagderyes, me too but I thought the Nano was still on PP too ;-)
09:07:48scorchebut i didnt ;)
09:07:54Bagderhahaha
09:08:23scorchenano size and battery life changed as well...
09:08:40scorcheif the 5.5 isnt PP i would expect at least *some* battery usage differences
09:09:03Bagderwell perhaps there are, but not big enough to mention
09:09:19scorchethe nano change was pretty big
09:09:22scorchei dont know..
09:09:38Bagderthe nano change was indeed major
09:12:14***Saving seen data "./dancer.seen"
09:14:47amiconnbrb
09:14:48 Part amiconn
09:15:59 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
09:16:10JdGordonthe nano isnt pp anymore? they did end up moving to samsung did they?
09:16:53daurn|afkdoes rockbox work on 5.5G?
09:17:33BagderJdGordon: samsung it is
09:17:38Bagderdaurn|afk: nobody knows
09:17:40scorcheJdGordon: see forum thread/IRC backlog conversation between paul and i imtermittently from about 4 hours ago
09:17:52daurn|afkand i guess not the nano 2gf
09:17:53JdGordonk
09:18:04JdGordonso tis a race between us and ipl to get it going :D
09:18:23scorchedaurn|afk: we know enough about that to know that rockbox will not work on it without some tweaking
09:19:13 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
09:19:16scorcheJdGordon: i would think that sharing information would be more efficient, but think of how it will be as you will =P
09:19:54Bagderdo we have any interested developers for these new beasts?
09:20:35JdGordonif i knew anything about reverse engineering and had $250 spare id volanteer :p
09:22:17 Quit aegray (Remote closed the connection)
09:22:28hcsshould there be any difference between loading rockbox with its own loader or loader2?
09:22:40scorcheyes
09:23:15hcsok, I don't want to jump to conclusions, but it was very crashy on my ipod photo with the rockbox loader, I'm using loader2 now and no crash yet
09:23:18scorcherockbox's loader relies on button presses (or lack therof)....loader2 has a menu to choose which os to load
09:23:32hcsoh, right, I know about the interfaces
09:23:37scorcheah...as in that
09:23:54scorcheclarifying questions goes a long way =P
09:24:10hcssorry, I was looking for "it loads the exact same image into RAM"
09:24:16hcsmaybe with an insult thrown in
09:24:41hcsbut I guess there's some differences in hardware init, or I'm just getting lucky
09:25:08BagderJdGordon: I was more thinking of a known developer we could fund a player for
09:25:36hcsthanks to all involved in this, its a great program, if I knew ARM I'd help out
09:25:49Bagderhcl: you'll learn quickly!
09:25:53Bagder;-)
09:26:24hcsI may, I'm fluent in MIPS, done a lot of N64 development.
09:26:34Bagderaha
09:26:43BagderMIPS is... weird asm
09:26:59BagderARM is dead easy in comparison I'd say
09:27:14hcscool, I was just going to say how easy MIPS is
09:27:18Bagderhaha
09:27:37Bagderit _could_ also be a sign of me not too accustomed to MIPS
09:27:55 Join aegray [0] (n=aegray@74.135.163.80)
09:27:56hcsits one the the most reduced of the RISC ISAs
09:28:18Bagderindeed
09:28:46Bagderbut reverse engineering MIPS is harder since there's no 32bit constants anywhere
09:29:06Bagderimho
09:29:07hcsmost of rockbox is in C, though, so I shouldn't even need to get much into the specifics
09:29:11Bagderindeed
09:29:21Bagderonly small parts are asm
09:29:28Bagderand not many people need to fiddle with those
09:29:57hcsI want to port the ipod parachute game as closely as possible, I love that thing
09:30:45Bagderstarting with a plugin like that is a perfect entry to Rockbox hacking
09:31:25hcsmaybe see if there's something to fix up in the loader... still no crashes playing OGG for about a half hour
09:32:13Bagderpoke linuxstb when he pops in, he knows those areas
09:33:17hcsdo you know about the graphics? it seems like everything in rockbox and ipodlinux updates fairly slowly, I can see things partially drawn, are there commands we're missing or do we not wait to commit the whole screen at once, or...?
09:33:51Bagderwe have a framebuffer in ram that is sent to the LCD on lcd_update()
09:34:23Bagderbut it should not do anything partially drawn
09:34:45 Join theli_ [0] (n=theli@mail.ukrcard.com.ua)
09:34:51hcsespecially with the selection highlight, I see diagonal lines in it where the drawing hasn't finished
09:35:12hcsmaybe its using a hardware inversion thing carelessly?
09:35:15Bagderand how can you tell it "hasn't finished" ?
09:35:40Bagderhttp://www.rockbox.org/twiki/bin/view/Main/LcdFrameRate
09:36:09Bagder(that mostly shows how fast lcd_update() is)
09:38:06hcswhen I scroll the highlight up and down I can clearly see a shallow diagonal line where the highlight ends and the normal background color starts, I have absolutely no idea what to blame it on but I figured I'd ask
09:38:19BagderI don't know either
09:38:44amiconnhcl: Ipod color/photo?
09:38:52hcsphoto, yeah
09:38:52amiconnErm, hcs
09:39:01amiconnI have a suspicion what that is...
09:41:01amiconnAre you able to build yourself?
09:41:04 Quit JdGordon ("Konversation terminated!")
09:41:05 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
09:41:17hcsyes
09:41:27hcsI'm running the one I built now
09:41:38 Join stripwax [0] (n=stripwax@i-83-67-214-206.freedom2surf.net)
09:41:39amiconnOkay, open firmware/lcd-ipod.c
09:42:02amiconn...then change the following:
09:42:16amiconnline 515f:
09:42:28amiconnif (pixels_to_write > 0x10000) {
09:42:35amiconncrap
09:42:42amiconnif (pixels_to_write > 0xffff) {
09:42:55amiconnh = (0xffff/2) / width;
09:43:14amiconnand line 712f to the same
09:43:26hcsok, change from 0x10000 on those two lines to 0xffff?
09:43:35amiconn(i.e. limit at 65535 instead at 65536)
09:43:49amiconnYes, but it's 4 lines
09:43:53hcsok
09:46:52hcsmight I say how lovely it is to have a source distribution that builds with minimal effort...
09:48:16Bagder:-)
09:48:31*Bagder thinks our build system is pretty good
09:48:38 Quit theli_ua (Read error: 110 (Connection timed out))
09:48:47*hcs agrees
09:49:57Slasheriamiconn: hmm, that seems to fix the issues with cpu boosting with plasma plugin for example.. i just added very simple feature of cpu auto boosting to thread.c. If it detects that core has been last in sleep long ago (CPU_AUTOBOOST_TRIGGER interval, for example 100ms), cpu will be boosted by kernel. And then unboosted immediately when core goes in sleep. cpu_boost still keeps cpu boosted if requested, so kernel wont override anything. It ...
09:50:03Slasheri... just makes playback code simpler, since it's enough just to trigger the cpu boosting manually if buffer goes low without the need to worry about unboosting the cpu
09:50:06Slasheriamiconn: what do you think about that? :) i am just testing it
09:50:31hcsok, rebuilt with the modification, still seems to behave the same way
09:50:33LinusNautoboosting????
09:50:43SlasheriLinusN: just experimenting with it..
09:50:59*LinusN whispers......KISS
09:51:03 Join webguest71 [0] (i=c27f0812@gateway/web/cgi-irc/labb.contactor.se/session)
09:51:03amiconnSlasheri: There seems to be a rather ugly bug with the scheduler bug. Idle poweroff may hang rockbox, observed on X5
09:51:08Slasherii am not sure if that is a good idea, but currently i haven't yet seen bad things to happen with it
09:51:28Slasheriamiconn: hmm, interesting
09:51:52Slasherii will try if it can be reproduced with iriver
09:52:30 Join bluebrother [0] (i=E47LWOEf@nat-wh-1.rz.uni-karlsruhe.de)
09:52:58 Part stripwax
09:53:40Slasheriamiconn: with the current playback code and cpu boosting, the problem is that playback engine wont know if other threads require more cpu time, so it unboosts cpu immediately when buffer is full. That might then stutter soem plugins for example
09:53:43LinusNSlasheri: i think it's a good idea to do the scheduler changes in small iterations
09:54:03SlasheriLinusN: ok, i will leave that feature out from the first commit at least
09:54:09LinusNthanks
09:54:11Slasheri:)
09:56:05LinusNSlasheri: url to the latest scheduler patch?
09:57:45SlasheriLinusN: http://ihme.org/~miipekk/rockbox/scheduler9_rc.patch
09:57:52Slasherii think that should be the latest
09:58:14Slasheri(without that new cpu boosting experiment of course)
09:58:26webguest71@FlySpray admins: could you please correct the summary text of the task FS #5994. The first word should be 'Mark', not 'Park'
10:00
10:00:11Bagderdone
10:00:13LinusNwebguest71: done
10:00:16LinusN:-)
10:00:16Bagderhaha
10:00:23Bagder"This task was edited by someone else before you saved. "
10:00:32LinusNi won!
10:00:42*Bagder hangs head
10:00:50hcsamiconn: any other suggestions?
10:01:00 Join spiorf [0] (n=spiorf@host61-203.pool8248.interbusiness.it)
10:01:34LinusNwebguest71: i like that patch btw
10:02:05webguest71LinusN: thanks. Twice.
10:02:54 Nick hcl is now known as HCl (n=hcl@hecate.student.utwente.nl)
10:03:05 Join theli__ [0] (n=theli@mail.ukrcard.com.ua)
10:03:11hcswhoops, didn't know I was so close to another name
10:05:56scorchehe has an acidic personality anyway....
10:06:10LinusNwebguest71: one minor quirk, your patch might break the "Follow playlist" functionality
10:07:50 Join Mmmm [0] (n=mscarrat@cpc4-hem13-0-0-cust438.lutn.cable.ntl.com)
10:08:21scorchenot even a giggle =(
10:08:44hcs(ba dum cha)
10:14:01webguest71LinusN: ehm... how? Does 'Follow playlist' use rockbox_browse?
10:14:35LinusNit uses lastfile
10:14:44LinusNor rather set_current_file()
10:15:18webguest71LinusN: another note: file name extensions should be defined somewhere as they are used in many places
10:15:52 Join mirak6 [0] (i=56da71cd@gateway/web/cgi-irc/labb.contactor.se/session)
10:16:22webguest71LinusN: well, my first thought was to introduce another "global" var (since dirbrowse has no params) but then I saw lastfile...
10:16:43 Quit theli_ (Read error: 110 (Connection timed out))
10:18:23 Quit dan_a (Read error: 110 (Connection timed out))
10:19:28 Quit mantono (Remote closed the connection)
10:21:01 Join theli_ua [0] (n=theli@mail.ukrcard.com.ua)
10:22:03LinusNwebguest71: maybe you should consider using snprintf() instead of strcpy/strcat
10:23:49 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au)
10:25:10LinusNSlasheri: it would be nice if you stick to the relevant changes in your patch
10:25:26LinusNexample:
10:25:28LinusN- if (state != boost_state) {
10:25:28LinusN+ if (state != boost_state)
10:25:28LinusN+ {
10:25:52LinusNthis is 1) not necessary and 2) breaks the coding style in that particular file
10:26:46LinusNthere are also a few unnecessary whitespace changes
10:27:36LinusNall these unnecessary diffs make it harder to review the patch
10:28:22 Join dan_a [0] (n=dan-mirc@wormhole.domicilium.com)
10:28:27mirak6hi
10:30:25aliaskNew nanos look like they have the new chips in them.
10:30:42aliaskhttp://gallery.ilounge.com/ipod/displayimage.php?album=61&pos=29
10:31:19markungood day mirak6
10:31:19Bagderaliask: yaps, been mentioned in the rockbox forum too
10:31:40aliaskStill arm core though :)
10:31:50LinusNapple chips, yummy
10:32:32scorchealiask: gotta be faster than that =P
10:32:41aliaskHey come on, I've been at school...
10:33:10aliaskDoes anyone know the specs of those chips?
10:33:47 Join webguest92 [0] (i=c0647cda@gateway/web/cgi-irc/labb.contactor.se/session)
10:34:02scorchealiask: see the forum post....it is all we know atm
10:34:29aliask*sigh* Fine.
10:34:54scorchedont sigh me mister!....i have been doing research all night! =P
10:35:09aliaskFINE...
10:35:14aliask;)
10:35:27 Quit theli__ (Read error: 110 (Connection timed out))
10:36:01hcsis anyone working on iTunesDB support? I'm thinking of picking up iPlay
10:36:23 Quit Mmmm (Remote closed the connection)
10:38:44 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/labb.contactor.se/session)
10:39:05 Join Mmmm [0] (n=mscarrat@cpc4-hem13-0-0-cust438.lutn.cable.ntl.com)
10:39:45 Quit aegray (Remote closed the connection)
10:40:03markunhcs: what kind of iTunesDB support?
10:40:09linuxstb_hcs: No, I don't know of anyone working on the itunesdb.
10:40:40hcsI don't know exactly, either, just a handy way of accessing it
10:40:55hcsright now I have a linux program to read the db and generate playlists
10:40:55linuxstb_Do you know about tagcache?
10:41:06linuxstb_Ah, the playlist part?
10:41:07hcsyeah, but my files are horribly undertagged
10:41:26hcsthe only organization I have is through the itunes db
10:41:53linuxstb_So itunes doesn't change the tags if you edit the information in itunes?
10:42:23hcsI don't know, I've been using gtkpod for a while
10:42:46 Join bbroke [0] (n=bbroke@p54BD390F.dip0.t-ipconnect.de)
10:42:51linuxstb_I'm just wondering how you can organise them in the itunesdb without them being properly tagged.
10:43:19hcsI import whole albums at a time, it assigns default titles based on file names, then I mass tag the newly imported files with the album name
10:43:47 Join Jungti1234 [0] (n=jungti12@124.60.9.28)
10:44:21hcsoh, I was supposed to ask you about loader2 vs. the rockbox loader
10:44:32linuxstb_I just think it would be a lot easier to properly tag your files and use tagcache (or just browse by filename) than attempt to support the constantly moving target that is the itunesdb.
10:44:45Jungti1234hi all
10:45:02hcsyeah, I just want a lazy solution :)
10:45:19linuxstb_Adding support for the itunesdb won't be that...
10:45:45hcsiPlay already has it, working, I just need to sure it up a bit
10:45:45 Join aegray [0] (n=aegray@74.135.163.80)
10:46:32hcsanyway, the playlist directory works fine for the moment
10:47:24linuxstb_back in a few minutes...
10:47:25 Quit linuxstb_ ("CGI:IRC")
10:52:42 Quit webguest71 ("CGI:IRC (EOF)")
10:53:39 Join webguest94 [0] (i=cb3bbc3d@gateway/web/cgi-irc/labb.contactor.se/session)
10:53:40 Quit solarflare (Read error: 104 (Connection reset by peer))
10:53:56 Join solarflare [0] (n=supernov@82-203-170-134.dsl.gohome.fi)
10:54:39webguest94hey guys im wondering if some one can tell me how to install the album art patch
10:55:37bluebrotherwebguest94, there are some wiki pages about that.
10:55:49webguest94oh really
10:55:55bluebrotherSimpleGuideToCompiling and WorkingWithPatches pops in my mind.
10:56:17bluebrotheryou don't "install" a patch, you apply it to the source. Then you need to build rockbox yourself.
10:56:25webguest94errr do you have to compile and stuff
10:56:30bluebrotheryes.
10:56:38webguest94oh
10:56:45webguest94well il give it a shot
10:56:47bluebrotherpatches can't be "installed"
10:56:56webguest94im not much of a programer
10:56:59bluebrotherthey only contain differences to the sources.
10:57:08webguest94hmm
10:57:10webguest94ok
10:58:23SlasheriLinusN: well, it's not possible to avoid all kind of those "not necessary" changes, because usually one doesn't even think doing such a change. And btw regarding the coding style of that file.. it's a full mess of (...) { and (...) \n { styles. So it's rather hard to stick on any of these
10:58:27bluebrotherit's different to the patches e.g. Microsoft releases. Those are binary patches, but this only works if you know what exactly version those binary patches are build against.
10:58:59LinusNSlasheri: it's very possible to avoid those changes
10:59:12webguest94so will the patch, patch up the rockbox.iriver file to work with album art
10:59:16LinusNSlasheri: just review your own patch and undo the unnecessary changes
10:59:27Slasherihmm, true. that would do it
10:59:47 Quit XavierGr (Read error: 110 (Connection timed out))
10:59:51bluebrotherwebguest94, no.
11:00
11:00:22scorchewebguest94: you will apply the patch, then re-compile...please read the pages that he referenced
11:00:26bluebrotherThe patch will change the source. Then you can recompile the source which gives you the rockbox.iriver file (and the other files that rockbox needs)
11:00:43webguest94ok
11:01:00Slasheribut i am not sure if it's for the effort to undo some minimal changes that would just improve the code readability
11:01:04LinusNSlasheri: i'm trying to figure out why idle poweroff would hang
11:01:16SlasheriLinusN: sounds great
11:01:31Slasherii haven't yet seen that to happen with my iriver :/
11:03:51dongsany new things on ipods wrt power management?
11:04:32dongsLinusN: hey wahts this? Set the iPod hardware EQ gain default to 0dB <
11:04:41scorchedongs: the site lists all CVS changes...
11:05:11LinusNdongs: it defaulted to +12dB before
11:05:20dongshrm
11:05:24dongsthat makes it sound bad?
11:05:30LinusNincredibly bad
11:05:40dongsthis is only if EQ is on?
11:05:43dongsor all the time
11:05:44LinusNthe eq is supposed to default to flat
11:05:49dongsah
11:08:34linuxstbhcs: Going back to your loader2 question - are you running exactly the same rockbox build as you did with the Rockbox bootloader? If not, could you test your currently installed build with the Rockbox bootloader? i.e. try and prove conclusively that loader2 helps?
11:12:18***Saving seen data "./dancer.seen"
11:20:36mirak6linuxstb have you reviewed the patch ?
11:21:08 Quit aliask (Read error: 54 (Connection reset by peer))
11:21:48 Join _FireFly_ [0] (n=FireFly@p54A4701E.dip.t-dialin.net)
11:22:05 Join Arathis [0] (n=doerk@p508A7C48.dip.t-dialin.net)
11:24:00linuxstbmirak6: Yes, I've got a few comments :). 1) It's full of TAB characters; 2) You make a few unnecessary whitespace changes to existing code; 3) The mpeg2 decoder struct can be made static for all targets, but just the Coldfire version in IRAM (but I'll test that change on the ipods and see if it helps) 4) Can you be more specific about the speed improvement it gives? Preferably using one of the Elephants Dream videos so we're all
11:24:00linuxstbtesting the same thing. 5) Would it be possible to put your Coldfire idct it in a separate file? - e.g. idct_coldfire.c
11:27:01 Quit aegray (Remote closed the connection)
11:27:41 Join aegray [0] (n=aegray@74.135.163.80)
11:27:52 Quit Jungti1234 ()
11:30:53 Quit bbroke ("Chatzilla 0.9.75 [Firefox 1.5.0.6/2006072814]")
11:34:37 Quit webguest94 ("CGI:IRC (EOF)")
11:37:26 Quit aegray (Remote closed the connection)
11:38:24 Join aegray [0] (n=aegray@74.135.163.80)
11:38:30 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/labb.contactor.se/session)
11:40:34 Join theli_ [0] (n=theli@mail.ukrcard.com.ua)
11:41:00 Quit theli_ua (Nick collision from services.)
11:41:08 Nick theli_ is now known as theli_ua (n=theli@mail.ukrcard.com.ua)
11:46:00 Quit bluebrother ("Leaving")
11:46:09 Join bluebrother [0] (i=duq0nY3B@nat-wh-1.rz.uni-karlsruhe.de)
11:46:45preglowapple marked chips just don't really bode any well for a port
11:47:44LinusNnope
11:48:10LinusNpreglow: any new regarding the eq problems last night?
11:48:16preglowno time right nowe
11:48:18LinusNnews even
11:49:07preglowhmm, think i'll check out if it happens on ipod as well
11:49:09preglowit should
11:54:08preglowbut i can't make it happen
11:54:45preglowyes, there i did
11:55:26mirak6linuxstb I want to do idct_add and idct_copy in asm. for elephant dream in 196*126 I go from 20 fps to 33
11:56:15amiconnpreglow: If the coefficients are always close to 1.0 or 2.0, you can improve precision without needing more bits...
11:56:53linuxstbmirak6: What about without the IRAM change? i.e. can you give figures for the two changes individually? I would prefer to commit them separately, with different commit messages.
11:56:55preglowamiconn: they are always close to that for low cutoffs
11:57:04preglowamiconn: not by any means in other cases
11:59:24mirak6the performance gain is really obvious when you test it. profiling would help though. What I need to test carefully is how memory acces can be optimised, for exemple, is it preferable to acces 4 consecutive bytes in sdram with 4 move.b or is it better to load that long word to iram then acces the bytes with move.b from that iram place. Same for writting..to
11:59:43 Join MarcoPolo [0] (n=MarcoPol@che78-2-82-227-240-106.fbx.proxad.net)
12:00
12:01:28linuxstbCouldn't you just do it in asm? i.e. read the four bytes into a register, do the processing, then write the register back?
12:02:10 Quit aegray (Remote closed the connection)
12:02:19linuxstbOr preferably, read/write more than 4 bytes at once.
12:03:02 Join aegray [0] (n=aegray@74.135.163.80)
12:08:09mirak6linuxstb the problem is that it's not easy to get the bytes indivdiually. if you look in the asm functions that do clipping and add dest+block, I managed to copy back to sdram in one shot by moving bytes to register then shiffting to left. but getting the bytes from a registery loaded to ram seems harder because you get the bytes and in the wrong order and can't do a circular shift
12:08:12 Quit aegray (Remote closed the connection)
12:08:32preglowreading a byte from memory to a register is one cycle faster for long words than bytes
12:08:47 Join aegray [0] (n=aegray@74.135.163.80)
12:08:58preglowyou should always burst read if possible
12:11:18mirak6I can't find an efficient way to get bytes from a register without putting them each in individual register
12:11:34preglowyou just shift them out each time you need one
12:11:49preglowyou'll need a shift and an and
12:11:50mirak6in fact I don't see how to do that with a low cost
12:12:03preglowcost won't be low, no
12:12:15preglowit might just be cheaper to fetch bytes one by one if the source is in iram
12:12:38scorchelinuxstb: maybe i should have made "some" in caps to match how i said it to myself in my mind =P
12:13:20mirak6that's why I was wondering if using a little iram buffer of a long word wouldn't be easier.
12:13:50vertic23good morning everybody
12:14:54 Quit aegray (Remote closed the connection)
12:15:30 Join lee-qid [0] (n=liqid@p54966AC3.dip.t-dialin.net)
12:16:09 Quit lee-qid (Client Quit)
12:17:01 Quit petur ("lunch")
12:17:22LinusNSlasheri: looks like it's not only the idle shutdown that hangs on the x5, the regular shutdown does too
12:17:26 Join aegray [0] (n=aegray@74.135.163.80)
12:19:28vertic23does anyone have any experience or opinion about the "Cowon iAudio X5V 20GB"?
12:22:22LinusNvertic23: that's the one without the FM radio, right?
12:22:55LinusNi have the regular iaudio x5 and i like it
12:23:33 Quit aegray (Remote closed the connection)
12:24:08 Join aegray [0] (n=aegray@74.135.163.80)
12:24:48vertic23yes, it is - atm I have a creative zen microphoto.. (with FM which I actually have used once...)
12:25:12vertic23...and it sucks ;P
12:26:43 Join barrywardell [0] (n=barrywar@194.46.178.162)
12:33:18vertic23uh.. but the remotecontrol is +50
12:33:29vertic23is it worth it?
12:42:01 Join RaiderX [0] (n=Snake@71.196.36.53)
12:44:51 Quit aegray (Remote closed the connection)
12:45:35 Join aegray [0] (n=aegray@74.135.163.80)
12:47:20 Quit RaiderX ("Anacna "^_^"")
12:48:52 Join ShadowdogMU [0] (n=Brock@cpe-65-28-252-205.woh.res.rr.com)
12:49:34jhMikeSLinusN: You were saying the x5 is hanging at power off?
12:50:38 Quit aegray (Remote closed the connection)
12:50:57*jhMikeS points out that the x5 calls remove_thread on the backlight thread when shutting down.
12:51:22 Join aegray [0] (n=aegray@74.135.163.80)
12:51:23preglowwhy???
12:51:36jhMikeSpreglow: Looooooong story...
12:52:34 Quit barrywardell ()
12:52:34jhMikeSperglow: there's no pullup resistor apparently for the backlight power converter so if the backlight is off when KEEPACT goes low you get a flash
12:52:58preglowsounds like a hack anyway...
12:53:38preglowhow do you know the backlight is on when shutdown occurs?
12:53:46jhMikeSpreglow: so...the backlight and lcd are turned on at powerdown at the bl thread is set to remove itself.
12:54:02jhMikeSpreglow: cause I wrote the hack to hide the flash
12:54:21preglowright, you answered me there
12:54:54 Join barrywardell [0] (n=barrywar@194.46.178.162)
12:55:14jhMikeSIf Shlasheri's thread code is hanging there I suggested a place to look, namely remove_thread([self])
12:57:27SlasheriLinusN: hmm, weird.. do you know why that happens?
12:58:13Bagderanyone knows how to make an area interpreted as thumb when using IDA?
12:58:39preglowshould be very easy
12:58:44preglowi THINK i did it once
12:59:07 Part LinusN
12:59:42BagderI thought so too, but I've failed so far
13:00
13:00:42SlasherijhMikeS: hmm, maybe remove_thread is the problem then.. i need to test that more, it might contain some bugs
13:01:24preglowwell, a thread removing itself is a weird case...
13:01:43jhMikeSinit_voice uses remove_thread but not on itself
13:02:07jhMikeSpreglow: why...what if you want to start a thread, have it do something and then terminate without waiting for it?
13:02:19jhMikeSthe case was handled in the old code btw
13:04:24Slasheriyep, remove_thread might crash in the patched scheduler when removing the current running thread because it doesn't do the proper preparations for the next thread
13:05:43jhMikeSSlasheri: How bout adding a get_current_thread() function? Should be there IMHO.
13:06:18amiconnpreglow: A thread removing itself isn't weid at all. It's done in other places as well (mostly plugins working with background threads)
13:06:54 Quit aegray (Remote closed the connection)
13:07:34 Join aegray [0] (n=aegray@74.135.163.80)
13:10:34SlasherijhMikeS: that shouldn't be any problem. what would you do with such a function?
13:11:04 Quit idnar (Read error: 60 (Operation timed out))
13:11:23 Join idnar [0] (i=mithrand@unaffiliated/idnar)
13:11:35jhMikeSSlasheri: It's good if you need the thread but don't happen to have the variable at hand
13:11:55jhMikeS*Need the current thread rather
13:12:06Slasherihmm, true. but why would you need that?
13:12:21***Saving seen data "./dancer.seen"
13:12:33Slasherinew scheduler would return just an entry to thread_entry struct
13:12:49 Quit aegray (Remote closed the connection)
13:12:54Slasheri(cores[CURRENT_CORE]->running contains the current thread)
13:13:07jhMikeSSlasheri: An index or a pointer?
13:13:08Slasherihmm, .running even
13:13:11Slasheripointer
13:13:28 Join aegray [0] (n=aegray@74.135.163.80)
13:13:37jhMikeSIt should return whatever parameter remove_thread would take
13:13:55jhMikeSand whatever create_thread returns
13:14:07Slasheriah.. but better solution could be to pass NULL or -1 to the remove_thread
13:14:13Slasheriso that would indicate the current thread
13:14:55jhMikeScould do that. Why would it be a problem to return the value create_thread returns?
13:15:32Slasheriwith new scheduler we would need to search and find that value
13:15:58Slasheriunless code is changed to return thread_entry pointer directly instead the number
13:16:00jhMikeSIt's not indexed?
13:16:06dan_aBagder: What could you possibly want to disassemble that would contain mixed arm and thumb code?!
13:16:14Slasherinot directly
13:16:15Bagderthe sansa bootloader
13:16:19 Join amsys [0] (n=uruz@home.amsys.cz)
13:16:23dan_a:D
13:16:24jhMikeSSlasheri: windows returns a pointer to the thread block as the thread id btw
13:16:45Slasherithe static list is still used, but we don't automatically know the in what index the thread belongs
13:17:06Slasherimaybe we should change that also
13:17:10jhMikeSSlasheri: So it wouldn't be wierd...could cast the pointer to an unsigned long value for convenience
13:17:29Slasheriyep
13:18:12Slasheribut i will try now if fixing the remove_thread fixes that bug on shutdown.. trying to simulate it on iriver
13:19:03preglowdan_a: lots and lots of stuff contain mixed code
13:19:40 Quit aegray (Remote closed the connection)
13:20:16 Join aegray [0] (n=aegray@74.135.163.80)
13:20:19jhMikeSSlasheri: do tell when you know! Just add the define to use x5_backlight_shutdown
13:23:26 Quit daurn|afk (Read error: 54 (Connection reset by peer))
13:25:30dan_apreglow: I'd guessed what it might be, having been staring at the same thing myself quite often
13:26:02 Quit barrywardell ()
13:27:20SlasherijhMikeS: good, i was able to simulate the crash.. now fixing the issue :)
13:27:31 Join daurnimator [0] (i=quae@unaffiliated/daurnimator)
13:27:37 Nick daurnimator is now known as daurn|afk (i=quae@unaffiliated/daurnimator)
13:28:12 Quit Arathis ("[rl_bot quit]")
13:31:31 Quit aegray (Remote closed the connection)
13:31:33jhMikeSSlasheri: did you get rid of that constant checking of the start address?
13:31:43linuxstbBagder: I was just about to commit FILES... :)
13:32:11Bagderhaha
13:32:28Bagderwell, then you can check if I missed anything
13:32:32 Join aegray [0] (n=aegray@74.135.163.80)
13:33:25linuxstbYou missed COPYING, but maybe that shouldn't have been committed anyway...
13:33:51BagderI didn't miss that...
13:34:11SlasherijhMikeS: Hmm, i haven't touched that (yet) :)
13:34:21linuxstbOK.. I must have missed it then :) Just did a diff between my FILES and your version...
13:34:29Bagderbut yes, as that COPYING is identical we should probably not include it
13:34:34SlasherijhMikeS: but now the thread removing is fixed.. creating a new patch soon
13:34:52 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/labb.contactor.se/session)
13:36:28 Join ally_ [0] (n=chatzill@80-77-241-35.dsl.xifos.net)
13:36:46ally_hello
13:38:09ally_I want to edit the H10 WPS page, just modified an iAudio WPS to work on the H10, could I get access to the edit wiki function?
13:39:25ally_my user name is AlastairCassell
13:39:29 Quit ally_ (Client Quit)
13:41:11linuxstbBagder: OK, I've removed COPYING now (along with a few other files that theli_ua told me shouldn't have been there).
13:42:47theli_uai hope i haven't included anything needed in that list ;)
13:42:55*petur added ally_ to the wiki users
13:43:28linuxstbtheli_ua: It still compiled for the ipod, so it should be OK...
13:43:44Slasheriok, now i am going to remove the threadnum and convert that to thread_entry.. should reduce code size also, that amiconn especially likes ;)
13:43:51 Join jadezhong [0] (i=de483967@gateway/web/cgi-irc/labb.contactor.se/session)
13:44:08theli_ualinuxstb,yeah .. i've made a test build without that files before messaging you :)
13:45:17linuxstbtheli_ua: It would be useful if you could create a wiki page for zxbox and add some documentation.
13:46:02theli_uayes i thought about creating a wiki and explaining basic usage
13:46:02 Quit jadezhong (Client Quit)
13:46:26 Quit aegray (Remote closed the connection)
13:46:29amiconnSlasheri: What about the idea of using the single global list w/ flags instead that linked list fiddling?
13:47:09amiconnThe size of thread.c's .text section almost doubles with the new scheduler, both on archos and coldfire
13:47:13dan_aBagder: "Change segment register value" of register T to switch IDA between thumb and arm mode for a certain instruction - or press Alt-G
13:47:18 Join aegray [0] (n=aegray@74.135.163.80)
13:47:21amiconnDidn't check arm yet, but I expect similar results
13:47:26Bagderdan_a: thanks!
13:49:24jhMikeSSlasheri: get_current_thread is useful if say, the current thread doesn't have it's id available but wants to pass it to another thread for referral to the first. A simple #define CURRENT_THREAD -1 won't work in that case. A get_current_thread api should be "instantaneous" (constant time, no search) too.
13:50:41Slasheriamiconn: i think the both ways to do it has some benefits
13:52:00 Join Magda [0] (n=chatzill@203-214-38-36.dyn.iinet.net.au)
13:52:58Magdahey people, i have a potentially dumb question
13:53:17*dan_a wishes that the demo version of IDA Pro would work with raw files
13:53:25vertic23hehe
13:53:26 Quit aegray (Remote closed the connection)
13:53:45preglowthe demo version of ida pro only works for intel, no?
13:53:47Magdai'm contemplating changing my collection to ogg files
13:53:55preglowMagda: congratulations
13:54:01Magdaso i'd need rockbox to play them on zee ipod
13:54:17Magdabut do ogg files display album covers?
13:54:23dan_apreglow: x86 and Arm, but only PE or Elf files
13:54:28preglowdan_a: roit
13:54:58preglowadd a cute little elf header to it :>
13:55:33linuxstbMagda: Rockbox doesn't officially support displaying album art. There's a patch (included in most of the unofficial Rockbox builds) that adds that capability. It needs the artwork to be stored in separate .bmp files - so it will work with any audio format.
13:55:51Magdaaah
13:56:06Magdaso it works, but with extra work >.<
13:56:16 Join Kohlrabi [0] (n=Kohlrabi@dslb-082-083-137-081.pools.arcor-ip.net)
13:57:31Magdaok, but how about on any music player on my computer? will they display album covers? do ogg files have that tag capability?
13:57:40Magda(noob question, i know)
13:57:55theli_uai've tried to create a new page http://www.rockbox.org/twiki/bin/view/Main/PluginZXBox but how do i make it appear at PluginIndex page?
14:00
14:01:33Magdahmm, i've just answered my own question and it seems that they dont
14:01:35Magdawell, crap
14:01:54Magdawhat music format will give me good sound and album images then >.<
14:02:11Magda(without having to make seperate files)
14:02:48peturseparate files are more efficient imho
14:03:00 Nick daurn|afk is now known as daurnimator (i=quae@unaffiliated/daurnimator)
14:03:10preglowthey are indeed
14:03:21amiconnlinuxstb: zxbox isn't slower on archos than rockboy. And unlike rockboy, zxbox also runs on Ondio...
14:03:24 Join aegray [0] (n=aegray@74.135.163.80)
14:03:45Magdabut it kinda kills the point of tagging programs like musicmatch picard
14:04:07 Join Criamos [0] (n=Criamos@p54930785.dip0.t-ipconnect.de)
14:04:11Magdaand it involves extra work looking for the image or scanning it myself
14:04:33amiconnPutting "album" art into each track's tag is a weird concept
14:04:53Zagortheli_ua: you need to refresh the cache of PluginIndex. I did that now.
14:05:30Magdaamiconn: why do you say that?
14:05:36theli_uaZagor, thanks
14:05:50*jhMikeS thinks plugins/codecs should use import tables instead of the api stucture pointers.
14:05:51peturMagda: 'album' <−−> 'track'
14:06:02ZagorjhMikeS: why?
14:06:07preglowjhMikeS: doesn't go too well with the binary format
14:06:47Magdapetur: you have the album or release that the track comes from in text, so why not an image?
14:06:55jhMikeSto keep binary compatibility so plugins/codecs can be precompiled
14:06:58Magdaits nicer to look at then squinting to read small text
14:07:15BagderMagda: because it is the same image for all songs in an album
14:07:17jhMikeSGive each api or exported data an id that is never reused
14:07:21dongsi got this awesome idea why dont they store ALBUM ART inside CD-TEXT blocks as BASE64 ENCODED binary??
14:07:50ZagorjhMikeS: why/when would you want to install new plugins without the corrresponding rockbox binary?
14:07:51linuxstbamiconn: So you think we should include zxbox in the Archos builds?
14:07:54jhMikeSThe ids should be in the pointer table and sorted so binary search can be used
14:08:07MagdaBagder: if there were individual images for individual tracks, i'd use that instead.
14:08:25BagderMagda: I was only saying why embedded the image in songs is silly
14:08:28Magdaut's also the same text for all tracks on an album.
14:08:33 Join Genre9mp3 [0] (n=yngwiejo@88.218.17.158)
14:08:35Bagderyou'll have the image N times isntead of 1
14:08:42jhMikeSZagor: So people can distribute plugins an binaries for use on rb like program...but I think relocatability would have to be threre too.
14:08:44MagdaBagder: i know. and i cant see why it's silly
14:09:00MagdaBagder: aah ok, now i get you
14:09:02Bagdera matter of opinion of course
14:09:17amiconnlinuxstb: Perhaps. The display resolution speaks against it though. The lack of resolution is even worse than for rockboy
14:09:36BagderjhMikeS: I don't see how we gain much by such a move
14:09:40amiconn(Spectrum: 256x192, Gameboy: 160x144, Archos: 112x64)
14:09:44linuxstbMagda: A bigger reason IMO against Rockbox using album art in tags is that it is normally the wrong size for your LCD - so it needs rescaling. It's also normally JPEG, which is harder for Rockbox to decode, and would mean adding JPEG decoding to the Rockbox core code.
14:10:05ZagorjhMikeS: it's like with linux kernel modules. if it's good it should be in the distribution. making a messy API with dynamic lookup etc is far too much trouble than it's worth
14:10:38linuxstbSo it's a lot more efficient to do the JPEG decoding and rescaling once on your PC, and then just store them as uncompressed .bmp files. itunes does the equivalent job when it creates the album art database on your ipod.
14:10:51jhMikeSBagder: I think it's wierd plugins (which to me are like apps) have to be built with the core.
14:11:07Bagderweird perhaps, but very effective
14:11:14Bagderand makes an easy installation
14:11:25preglowdoing a relocatable format would be very nice
14:11:32preglownot having to do the bloody codec swapping anymore, for instance
14:11:38dongsoh so zune is out
14:11:42preglowalso, we don't have to put the codec at the end of ram anymore...
14:11:42dongsleet
14:11:46Zagordongs: no, just announced
14:11:51dongsoh
14:11:54Zagoravailable around christmas
14:11:56dongslooks fuckign awesome
14:11:57jhMikeSWhat if (when I'm done on other tasks) I prove some simplicity. Can the linker generate an import table?
14:12:03Zagordongs: yeah, if you like brown ;)
14:12:03dongsi hope it runs XP Embedded or osmethign
14:12:07dongsbrown?
14:12:13dongsi see white version
14:12:17Zagorit's available in white, black and ... brown
14:12:21dongsya so?
14:12:27Magdalinuxstb: good point. *sigh* i'm going to have to work on this more
14:12:27Zagorjust funny
14:12:29dongsand ipod is available in HOT PINK
14:12:37jhMikeSpreglow: I was thinking last night about that regarding transcoding of files to have two codecs loaded...decoder and encoder.
14:12:47ZagorI just find it an odd third color to choose.
14:12:48Magdacheers everyone for your help
14:12:57 Part Magda
14:13:17Zagordongs: http://www.microsoft.com/presspass/presskits/zune/images/image002_low.jpg
14:13:29dongshrm
14:13:35dongsis that a clickwheel ipod style?
14:13:42dongsor just updownleftright
14:13:49Zagorit's a joystick afaik
14:13:52dongsmeh
14:14:10Slasheriamiconn: with my new version, binary size increase on archos only 356 bytes
14:14:15Slasherii wouldn't say that is much :)
14:14:21Slasheriand with linked lists of course
14:15:26amiconnRoughly half of what the rc9 added
14:15:57Slasheriyep
14:20:59SlasherijhMikeS: http://ihme.org/~miipekk/rockbox/scheduler10_rc.patch, please try if that fixes the problem
14:21:25preglowjhMikeS: yes, exactly
14:21:31jhMikeSSlasheri: I never tried it before to know it was broken :) but okay...
14:21:34preglowi'd love to have a simple relocatable plugin format
14:21:47linuxstbIs there such a thing?
14:22:02preglowcoff is simple enough
14:22:18preglowanyway, it's not hard to homebrew a format
14:22:24preglowi remember nasm coming with something custom once
14:22:34 Quit aegray (Remote closed the connection)
14:23:13preglowwe could always use elf :PPP
14:23:34amiconnurgh
14:23:41 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au)
14:23:47preglowi just don't like all the ugly warts the current system has
14:24:13amiconnThe current system is plain & simple
14:24:14dongshuhu, are the plugins now relocatable?
14:24:22preglowamiconn: i didn't say it wasn't, did i?
14:24:25preglowit's too plain and simple
14:24:32dongsor do they have to be compiled as position independent code
14:24:34preglowhaving to do codec swapping to even use two codecs at the same time
14:24:56linuxstbdongs: They are linked to execute from a fixed address near the end of RAM.
14:25:00preglowa problem we have to fix with multiple linking
14:25:26jhMikeSSlasheri: have to update my patch tester checkout...
14:26:18preglowthe plugin format works nice for old limited archos stuff, but it's going to become too limited for newer platforms sooner or later
14:26:25preglowin some ways, it already is
14:26:37preglowwhat with the codec swapping...
14:28:20 Join aegray [0] (n=aegray@74.135.163.80)
14:28:30jhMikeSbuilding...
14:29:01amiconnpreglow: With relocatable plugins we'd need to introduce some kind of dynamic memory management. Certainly not wanted...
14:29:03jhMikeSSlasheri: just want me to checkout the shutdown part?
14:29:11preglowamiconn: why?
14:29:55amiconnWith dynamic memory management every allocation might fail, and has to be checked
14:30:10amiconnI don't want to see out-of-memory errors in rockbox
14:30:11preglowamiconn: why would we need memory management?
14:30:20preglowi know we don't want it, you don't need to tell me that
14:30:29Bagderwe don't need malloc to do relocatable load
14:30:31amiconnWhere would you load the relocatable plugins?
14:30:36preglowamiconn: in the plugin buffer...
14:30:47amiconnHow would that work?
14:30:54Bagderjust that the plugin buffer don't have to be fixed at an absolute position
14:30:57pregloweh
14:30:59preglowyou just load it
14:31:02preglowhow hard can it be?
14:31:07preglowplugin buffer is the same old thing
14:31:10preglowit just floats around
14:31:11amiconnYeah, but where in the buffer?
14:31:22preglowit's wherever we bloody want to put it
14:31:28preglowstraight after all the other rockbox data
14:31:38amiconnHuh?
14:31:50preglowi really don't get what your point is
14:31:55amiconnWhat if the codec changes, and the new codec is larger than the old one?
14:32:06preglowthe plugin buffer is still fixed size
14:32:09preglowjust like now
14:32:19amiconnNot plugins, codecs...
14:32:26preglowwell, plugins, codecs, everything
14:32:29preglowthat part won't change
14:32:31amiconnThat's related to your 2-codec scenario
14:32:48amiconnMost of the time we don't need 2 codecs
14:33:13preglowi don't even know how that's handled now. where do we stuff the extra codec?
14:33:25amiconnhmm....
14:33:49amiconnMaybe that's a point...
14:33:49jhMikeSSlasheri: almost there...
14:33:58preglowthe extra codec buffer would be allocated on boot if voice is enabled
14:34:00preglowsimple as that
14:34:13preglowthrough perfectly ordinary rockbox means
14:34:29amiconnhmm again...
14:34:36 Quit aegray (Remote closed the connection)
14:35:03amiconnWhat about the iram parts?
14:35:12preglowthat will still need extra handling
14:35:22preglowthe loader will always have to pretend iram is used by the loaded object alone
14:35:28preglowand the playback thread needs to swap it out
14:35:30preglowwe can't avoid that
14:35:53 Join aegray [0] (n=aegray@74.135.163.80)
14:36:04preglowbut that copy is a lot faster than the ram to ram copy we need for codec swapping
14:36:09preglowhalf a megabyte of copying...
14:36:34amiconnNot that slow with burst-mode memcpy
14:36:44preglownah, not that slow, but unecessary still
14:36:59amiconnSo relocatable plugins and codecs would solve a few problems, but leave the majority of them
14:37:32preglowthe majority which is what? the iram problem? the need for static buffers? both of those can't be avoided, unless we start doing malloc, which i don't think we should
14:37:58amiconnThe only things it would solve are (1) Codec swap only needs to swap iram, not dram. Swapping and interlocking is still necessary. Plugins using iram also still need to stop playback and mute voice
14:38:30preglowit would solve the need for multiple builds for targets with differing amounts of memory too
14:38:56amiconn(2) It would make it easier to adapt to varying memory sizes (no separate builds for 64MB ipods anymore, and also for 8MB-modded archos)
14:39:23preglowanyway, i'm quite certain this feature would be allowed in rockbox if implemented cleverly
14:39:26amiconnBut then it complicates the loader
14:39:29preglowbut that needs to be done first :>
14:39:43amiconn...and makes the plugins larger, as they need a relocation table
14:39:46Zagorit also complicates the plugins/codecs
14:39:46linuxstbCouldn't we just have a (smaller) voice codec buffer, and compile the voice codecs to run from there?
14:40:03amiconnZagor: Not the codec/plugin code though
14:40:05preglowlinuxstb: i'd rather we just compiled in a custom voice codec
14:40:39Zagoramiconn: well, depending on what form of cosmetics we add
14:40:40amiconn...unless we can convince gcc to produce truly position independent code
14:41:09preglowalso, i don't really know if pic code is any slower than ordinary code
14:41:17preglowso i might of course be overlooking tons of things
14:41:49preglowarm code is pretty pic by default, at least
14:41:50jhMikeSSlasheri: things seem to be working fine...something seems "crisper" in the way its running. My OGG (Aerosmith - What It Takes) went from 85% boost with EQ to 71% !!! wtf
14:42:06pregloww00t!
14:42:23Bagdergosh
14:42:30preglowhow's that possible?
14:42:31jhMikeSI guess it was a good idea not to schedule waiting threads
14:42:47preglowi'll have to do some performance measurements myself now
14:42:53Bagderhow does this perform on the arm targets?
14:42:58preglowi intend to find out
14:43:09jhMikeSI always use the same file for testing the OGG.
14:43:40 Quit aegray (Remote closed the connection)
14:44:24 Join aegray [0] (n=aegray@74.135.163.80)
14:44:36preglowjhMikeS: btw, what bitrate is that?
14:45:08jhMikeS224
14:45:15preglowvorbis is slow...
14:45:28jhMikeSvorbis is good punishment
14:45:51preglowcompiling rockbox takes bloody ages with all these plugins!
14:45:57preglowi'm starting to see your point, jhMikeS
14:46:04jhMikeSpreglow: that's why I want a plugin SDK
14:46:47Bagderor just a make target that skips plugins ;-)
14:46:48jhMikeS76% boost with EQ and Crossfeed
14:47:48jhMikeSBagder: That would help things a bit...if the change only affect fw I just copy the rockbox.iaudio to my player when its done.
14:47:51SlasherijhMikeS: hehe, sounds great :)
14:47:59preglowbenching nano and mp3 now
14:48:14 Quit Genre9mp3 (Read error: 110 (Connection timed out))
14:48:17jhMikeSI want to test it with no minimal dsp
14:48:20amiconnpreglow: If you're building on linux, a full build is blazingly fast imho
14:48:24jhMikeS*with minimal rather
14:48:43preglowamiconn: i am, and it's not slow, but it could be faster without all the big plugins like doom, rockboy and pacbox
14:48:52 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
14:48:55preglowanyway
14:48:58*amiconn is building on cygwin most of the time
14:48:59preglowi don''t feel very strongly about it
14:49:28preglowi can just disable them manually anyway
14:49:29preglowwhich i think i will
14:49:36amiconnBuild times for swcodec easily exceed 10 minutes, but I wouldn't exclude plugins...
14:49:40 Quit aegray (Remote closed the connection)
14:49:41preglowholy shit
14:49:47 Join webguest84 [0] (i=54bcd4b1@gateway/web/cgi-irc/labb.contactor.se/session)
14:49:50preglowwhy, oh why, is it that slow
14:50:12jhMikeSI wish we had any easy way to keep assembly listings with binary and source when building and an easy way to make individual files
14:50:18Slasheritry with make -j2 or -j3
14:50:23LinusNjhMikeS: make bin
14:50:23Slasherimight speed up it a bit
14:50:26 Join aegray [0] (n=aegray@74.135.163.80)
14:50:31Slasheriit's quite fast with my dual xeon server :)
14:50:32 Join CriamosAndy [0] (n=Criamos@p54933B26.dip0.t-ipconnect.de)
14:50:37amiconnFrom my observation the windows file system performance is the culprit
14:50:38preglowhaha
14:50:44preglowamiconn: wow
14:50:51amiconnBuilding on microsoft sfu isn't much faster
14:51:13LinusNyou can type "make bin" to exclude the plugins
14:51:17jhMikeSGet about 38% without effects
14:51:46theli_ualinuxstb, in current (committed) state zxbox uses grayscale rendering on gs lcds with VERTICAL_PACKING and 4 colors when horizaontal packing (it is much faster) ... i wonder is this ok? (or should we change both to greyscale(just change one define)) .... or i can write output driver for vertical packing ...
14:51:47jhMikeSLinusN: don't know how I missed that one...errgh
14:52:16LinusNjhMikeS: it's a pretty new feature
14:52:27amiconn'make hex'
14:52:29amiconn;)
14:52:34LinusNmake sense
14:52:38jhMikeSLinusN: make asm -hex -source
14:52:46jhMikeS!!
14:53:10preglowdoesn't look like it helps out too much on the nano
14:53:15preglowbut there's a bit to go still
14:53:50jhMikeSI wish create thread would just return an unsigned long or something...cleaner to me
14:54:44jhMikeSLinusN: so it didn't exist before and therefore I can keep my sanity and not self-loathe for being dense? ;)
14:55:26preglowokay, 225avg mp3 went from 50Â% boost to 47%, so there's some merit in it. let's commit this mutha!
14:55:39LinusNjhMikeS: something like that :-)
14:56:03preglowgui does feel... different... though
14:56:53 Join XavierGr [0] (n=xavier@ppp142-171.adsl.forthnet.gr)
14:57:14jhMikeSWant to try long directory scrolls during playback...
14:57:23preglowhrm
14:57:33preglowclickwheel handling seems to suffer
14:57:35XavierGrYooppie! My internet connection was just doubled! :D
14:57:45linuxstbYou can access two internets?
14:57:46 Quit mirak6 ("CGI:IRC (EOF)")
14:57:55preglowtwo uplinks! hooray!
14:57:58XavierGrI mean the speed! :D
14:58:03Bagderyay, both internets at once1
14:58:13jhMikeSaudio doesn't break anymore it seems
14:58:21preglowhrm
14:58:44preglowcould someone else on ipod try the scheduler patch out with some ~200kbit mp3s?
14:58:53preglowdo some navigating
14:59:00preglowthe clickwheel seems to just stop responding some times
14:59:12XavierGrfrom 386kbps downstream it became 768.
14:59:37linuxstbpreglow: I'll give it a go. Should 192kbit CBR mp2 be similar?
14:59:47preglowdoubt it
14:59:47LinusNSlasheri: did you fix the shutdown issue?
15:00
15:00:14jhMikeSpreglow: most of my mp3's are greater >= 200
15:00:35preglowlinuxstb: nah, 192kbit mp2 handles much better
15:01:06*linuxstb searches around for some lossy files
15:01:25preglowi don't think the priority thing goes too well with the clickwheel on ipod
15:01:34jhMikeSIMO this is a nice improvement
15:02:07jhMikeSpreglow: Maybe then clickwheel need priority boost at certain times or something
15:02:20preglowjhMikeS: the clickwheel handling is in the ui
15:03:03preglowjhMikeS: btw, have you done the new peak reading code?
15:03:23jhMikeSScrollers on the WPS get jumpy though
15:03:26amiconnIsn't the clickwheel handled in the button driver?
15:03:40ZagorXavierGr: I guess you don't want to know about my 17 Mbit downstream then? ;)
15:03:44amiconnjhMikeS: Try jewels...
15:03:44preglowamiconn: well yeah, but it's the gui reading the wheel status that's the problem
15:03:49jhMikeSpreglow: yes...but it's quite intertwined with what I've done with recording in general
15:04:00amiconnIt gets _really_ jumpy, especially on X5
15:04:08 Quit aegray (Remote closed the connection)
15:04:09jhMikeSamiconn: with music going?
15:04:13amiconnyes
15:04:26preglowjhMikeS: sure, did you do playback peaks as well?
15:04:43*jhMikeS is happy he's actually listening to his x5 for a bit :)
15:04:51jhMikeSpreglow: no playback changes
15:04:52 Quit CriamosAndy ("( www.nnscript.de :: NoNameScript 4.03 :: www.XLhost.de )")
15:05:04preglowjhMikeS: wouldn't playback code benefit from the same kind of peak handling as well?
15:05:08 Join aegray [0] (n=aegray@74.135.163.80)
15:06:11SlasheriLinusN: yes, i think so
15:06:17LinusNwhat was it?
15:06:32SlasheriLinusN: you could try the latest patch, http://ihme.org/~miipekk/rockbox/scheduler10_rc.patch
15:06:34jhMikeSpreglow: maybe...I also don't do negation in the loop
15:06:41LinusNSlasheri: will do
15:06:42Slasherithere was a bug in remove_thread function
15:06:47LinusNaha
15:06:52LinusNthe backlight thread removal
15:06:53linuxstbpreglow: Yes, the clickwheel does feel very "sticky" with the scheduler patch.
15:06:56Slasheriyep
15:07:08preglownot a good thing
15:07:17preglowmost people will probably listen to high bitrate mp3s all the time
15:07:29*amiconn wonders how sticky it will become on 3G
15:07:33 Quit Siku ()
15:07:49 Quit daurnimator ("I wonder where that fish could be")
15:08:00jhMikeSI'm sure priorities will need tweaking
15:08:03 Quit Criamos (Read error: 110 (Connection timed out))
15:08:14preglowSlasheri: what's up with the #elif -> # elif changes?
15:08:23preglowi've never seen that style used in rockbox
15:08:32LinusNbacklight.c:543: warning: passing arg 1 of `remove_thread' makes pointer from integer without a cast
15:08:40linuxstbBut I can't make my ipod skip with my 210kbps MP3 file...
15:08:47 Join daurnimator [0] (i=quae@unaffiliated/daurnimator)
15:09:23dongslinuxstb: lol oic, thats kinda bad lorf.
15:09:39dongslinuxstb: so theres a limit on how big a plugin can be?
15:09:46dongsand you cant load more than one at a time?
15:09:50preglow512kb
15:09:58preglowand no, you can't
15:10:03amiconnOn swcodec
15:10:18jhMikeSLinusN: I think the thread id should be unsigned's not structure pointers
15:10:28preglowhah
15:10:35preglow230kbps mp3 and five band eq doesn't skip
15:10:37Slasheripreglow: just some editors automatic identations.. but there is same code also other places
15:10:42preglowgod, how it struggles, but it doesn't skip
15:10:50dongsjust add peakmeters
15:10:52dongsif you want skip
15:11:00preglowsome day i'll tell the world what i think about automatic indentations
15:11:04jhMikeSjewels was running just fine with 220+k mp3s
15:11:05SlasheriLinusN: ah.. of course i forget to try compiling the X5 target.. :) but i think it should work even with the warnings
15:11:12dongspreglow: i think they are awesome
15:11:15dongspreglow: indent -kr = <3
15:11:19 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
15:11:23Bagderin real editors YOU control the indents it makes... :-)
15:11:31*preglow hugs vim
15:11:36*LinusN hugs emacs
15:11:39dongslol emacs.
15:11:50*Bagder votes emacs for president
15:11:52SlasheriBagder: well, i pressed tab on the lines to correct identation of a block..
15:12:05XavierGrZagor: Have I told you that I hate most of the peple here because of their internet speeds? :p
15:12:08Slasherijed is a very similar than emacs, but more advanced for coding ;)
15:12:20ZagorXavierGr: :)
15:12:23***Saving seen data "./dancer.seen"
15:12:24dongsin vim ctl+d / ctl+d controls indent
15:12:24Bagdermore advanced? really?
15:12:28dongswhich could be differnet from tabsize
15:12:32dongs(which should always be 8 anyway)
15:12:34 Quit aegray (Remote closed the connection)
15:12:41dongsi hate shit that thinks tabsize is liek 4 or osmethign
15:12:41dongsugh
15:12:42SlasheriBagder: at least it works better with less configuration :)
15:12:52LinusNtabsize is irrelevant, since you don't use tabs
15:12:57SlasheriBagder: and it does its job without the need to program some lisp
15:13:01peturhahaha
15:13:13Bagderyeah, you use "S-Lang" instead
15:13:16Bagdermuuuuch better
15:13:19Slasheriyep
15:13:23Bagderhahaha
15:13:26Zagoreveryone knows S-lang already :-)
15:13:29preglowhaha
15:13:33preglowgot, i loathe emacs
15:13:37preglowgod too
15:13:50preglowtake some emacs lovers .emacs from him, and he's helpless
15:13:54 Join aegray [0] (n=aegray@74.135.163.80)
15:14:01preglowbut enough about that
15:14:03preglowbbl
15:14:06Bagderwell, that only show the powers of it
15:14:20BagderI have an advanced setup I like
15:14:26Bagderso of course I'd miss it
15:14:31linuxstbI'm playing a 215kbps MP3 file with the default WPS, and no skipping... (ipod Photo) I don't know if it would have skipped normally though.
15:14:34Zagortake away my .procmailrc and I'll sob like a baby. is that a negative thing for procmail?
15:14:50*Zagor fans the flames
15:15:03Topic"emacs emacs emacs!" by Bagder (n=daniel@rockbox/developer/bagder)
15:15:07*Bagder runs
15:15:09LinusNlol
15:15:20idnar/topic vim vim vim!
15:15:21Bagdertopic war
15:15:49Bagderhow we love debating the really big issues in the world
15:15:51dan_ased sed sed!
15:15:58LinusNedlin edlin edlin
15:16:25peturMicrosoft Visual Studio Microsoft Visual Studio Microsoft Visual Studio
15:16:27*petur ducks
15:16:33bluebrothercopy con copy con copy con
15:16:39LinusN"Windows - from the guys who brought you Edlin"
15:17:04dongsya, bashing windows is "in"
15:17:06*petur actually used edlin a lot in the past :(
15:17:10dongsand linux has been 'ready for desktop' for hte last 8+ years
15:17:12Bagderdon't wake dongs
15:17:15LinusNhaha
15:17:58bluebrotherhow long has windows been "ready for desktop"?
15:18:04vertic23LinusN: are you able to watch videos?
15:18:04dongssince day 1
15:18:07dongsbecause its on desktop
15:18:18Zagorbluebrother: it's all fixed in the next version
15:18:20Bagderwhy would anyone care about "ready for desktop" ?
15:18:39bluebrotherhehe.
15:18:50Bagderits been on mine for years
15:20:35LinusNSlasheri: shutdown works fine
15:21:07*jhMikeS has to go help his sister move...cu later
15:21:08 Quit Mmmm (Remote closed the connection)
15:21:11 Quit jhMikeS ("Y? Because it's X-tra")
15:23:38LinusNSlasheri: seems to work fine, but we might want to tune the scheduling a bit. the ui suffers quite a lot
15:25:01 Quit webguest84 ("CGI:IRC")
15:26:16vertic23LinusN: I am about to order my X5V for 270.. is that ok? :) something to mention before I click on "submit"?
15:27:03SlasheriLinusN: hmm, then it probably needs to be tuned.. i dont how how i could try that with iriver..
15:27:11LinusNvertic23: is 20gb enough for you, and are you sure that you don't want the fm radio?
15:27:19vertic23yeah
15:27:26Slasherior maybe the button thread needs a higher priority on ipod
15:27:31vertic23btw - how do you charge the X5V?
15:27:35vertic23the battery
15:27:42amiconnSlasheri: There is no such thing as a button thread
15:27:49LinusNyou use a charger, via the subpack
15:27:54vertic23hopefully not like mine over USB everytime you plug it in
15:27:54 Join actionshrimp [0] (n=nn@host86-140-236-161.range86-140.btcentralplus.com)
15:28:08vertic23ok, very good
15:28:12vertic23creative is so wrack..
15:28:14Slasheriah, hmm
15:28:32nudelyn/j #
15:28:32nudelyn#riovolution
15:28:33nudelynoops
15:28:35LinusNvertic23: that's my main complaint about the x5, you need to carry that subpack around
15:28:47vertic23is it big?
15:29:05LinusNno, but you have one extra item to keep track of
15:29:09amiconnEl weirdo subpack.... and the odd placement of the joystick
15:29:47vertic23indeed.. but mostly I have a bagpack ...like women their handbag ;)
15:30:10vertic23btw - what's a subpack? is it only the charger?
15:30:41vertic23...ow - and I need to buy it extra?
15:32:27LinusNit's included
15:32:55LinusNthe subpack is needed for USB, charger and Line in/out
15:33:36vertic23oh, ok - because I see you can buy it seperately
15:33:52LinusN...if you lose it
15:34:12 Join synic [0] (n=squish@pdpc/supporter/student/synic)
15:34:56vertic23aight.. then it's almost ordered
15:35:05 Quit HCl (Remote closed the connection)
15:35:15peturthat subpack was the one reason I got the h340 and not the x5
15:35:48 Join tvelocity [0] (n=tony@ipa246.6.tellas.gr)
15:35:50Bagderthe upside with it is of course that the player is somewhat smaller when you carry it around
15:36:43LinusNwell, they could have integrated the subpack by making the player slightly thicker
15:37:14vertic23...so there is no real "USB" plugin
15:37:19vertic23you always have to use subpack
15:37:31Bagderyes
15:37:41vertic23I see..
15:38:06vertic23well - if this is the only disadvantage... in comparison to my crappy creative player it's nothing ;)
15:38:32vertic23never ever creative.. no more.. they suck big time
15:39:53Bagderhey a wma decoder
15:40:20Bagderor at least almost
15:40:34vertic23btw - do you think that it'll be possible to watch videos over X5V with rockbox in the (near) future?
15:41:09LinusNyes
15:41:41vertic23...but not at the moment, right?
15:41:55LinusNonly without sound
15:42:30vertic23hehe, so ...a minor problem which can be fixed..
15:43:17LinusNi wouldn't call it minor, but anyway...
15:43:37vertic23okok...
15:43:47vertic23and what about DOOM I+II on the X5?
15:43:56vertic23possible?
15:44:15*LinusN starts doom I on his X5
15:44:37vertic23hehe - is it fun with that "tiny" display?
15:44:46vertic23and.. the buttons
15:44:47LinusNdefinitely playable
15:45:05vertic23you think that the remotecontrol is essential..?
15:45:11LinusNnot really
15:45:18vertic23okay
15:46:59 Join Criamos [0] (n=Criamos@p549336B0.dip0.t-ipconnect.de)
15:47:27vertic23thank you for your help - I guess I'll be here more often from now on ;)
15:47:48amiconnYeah, doom's playability on the colour targets is okay
15:48:02LinusNsurprisingly fun imho
15:48:10amiconnEven on greyscale targets it's playable, but you need good eyesight
15:48:21vertic23hehe
15:48:48amiconnHard to distinguish red, yellow and blue keys and the matching doors
15:49:01LinusNhehe
15:49:10vertic23hehehe
15:50:15*LinusN remembers the old april fools joke
15:50:15 Join monkeydog [0] (i=james@monkesys.net)
15:50:19LinusNhttp://www.rockbox.org/doom/
15:51:02 Part monkeydog ("Leaving")
15:51:25 Quit aliask (Read error: 104 (Connection reset by peer))
15:51:47 Join mantono [0] (n=mantono@c83-250-204-173.bredband.comhem.se)
15:53:03Topic"How hard can it be?" by Bagder (n=daniel@rockbox/developer/bagder)
15:54:56 Quit belze (Read error: 110 (Connection timed out))
15:55:52*linuxstb spots a wma decoder attachment
15:56:00*LinusN too
15:56:21Bagderwoo
15:57:15amiconnLinusN: That doom on archos one?
15:57:15linuxstbOuch, it's an entire rockbox source tree...
15:57:42amiconnIn fact it might even be possible to get it to work at least on 8MB-modded archos
15:58:00amiconnThe speed is another thing...
15:59:54preglowwoo, wma
15:59:57*LinusN notices that paul jones has never heard of patches
16:00
16:00:14Bagderso that's why it was so big...
16:01:15peturmailinglist?
16:01:26Bagderhttp://www.rockbox.org/tracker/task/4984
16:01:30peturah
16:01:47 Join dave [0] (n=dave@router-three.whitehorse.co.uk)
16:01:54preglowsomeone make a patch!
16:03:27preglowdamn, that's alot of code
16:04:31preglowahahha, dAYM
16:04:40preglowit does 64x64->64 multiplies
16:05:02preglowthat just can't be necessary
16:05:17davehi i wonder if anyone can shed any light on a problem I am having.. I have a iriver h10, I've loaded the rockbox software and the bootloader, rockbox starts and I can browse the text and playlist folder.. but the rockbox folder has an X next to it and i cant see any files? can anyone help?
16:05:26LinusNpreglow: read his comments in the fs task
16:05:34 Join hcl [0] (n=hcl@hecate.student.ipv6.utwente.nl)
16:05:43bluebrotherdave, there is no rockbox folder
16:05:43LinusNdave: that is not a folder
16:05:47 Nick hcl is now known as HCl (n=hcl@hecate.student.ipv6.utwente.nl)
16:05:50daveahh
16:05:51preglowLinusN: what about it?
16:05:51bluebrotherthat's the system itself.
16:06:01LinusNpreglow: he mentions the 64 bit muls
16:06:03davei am sure i should be able to see my music files tho?
16:06:04preglowLinusN: he went from 32x32->32 muls
16:06:15bluebrotherwhere is your music located?
16:06:18preglowLinusN: going from that to 32x32->64 should be a little less overkill
16:06:39davei can get into setting etc, but cant see any music? /Music as standard
16:06:55 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au)
16:07:36bluebrotherthen why can't you just change into /Music?
16:07:47LinusNbluebrother: because he doesn't see it?
16:07:50davecos i cant see it!
16:08:01bluebrotheroh, is this hidden on x5 per default?
16:08:11LinusNbluebrother: ehum, h10
16:08:21bluebrothertry setting the show files to all.
16:08:27bluebrotheroops, h10
16:08:34*bluebrother cleans his glasses
16:09:03davehow do i show all files? surely the /music folder isnt hidden?
16:10:22bluebrotherGeneral settings / File view / Show files / all
16:10:46davei can see the music folder on my PC... but not on rockbox
16:10:57davei use kubuntu dapper btw, not windoze
16:11:26daveor was that a command i run on my iriver?
16:11:35davei am such a noob
16:11:43LinusNdave: and you unmount the h10 properly before you unplug?
16:11:47bluebrotherI guess the folder is marked hidden but Linux hides files differently.
16:12:51amiconnYes, on unix-like systems .dotfiles are hidden
16:13:11amiconnRockbox knows both methods of hiding files
16:13:25bluebrotherdos / Windows has a flag in the file to mark it hidden.
16:13:40bluebrotherafaik when mounted on linux those hidden flags get ignored
16:13:53 Part LinusN
16:14:20 Quit dave (Read error: 104 (Connection reset by peer))
16:14:35*bluebrother notes to mention hidden files in the manual
16:15:17 Join rretzbach [0] (n=robert@dslb-084-062-001-141.pools.arcor-ip.net)
16:15:56 Join dave [0] (n=dave@router-three.whitehorse.co.uk)
16:16:46davesorry my pc hung before
16:17:06bluebrothernp
16:17:08davei am now showing all files and can see my files + music...so thankyou :0
16:17:46bluebrotheron linux I think you can change the hidden flag with "mattrib"
16:18:02davehow far down the line to an actual release is the rockbox h10 version? is this still in beta or the like?
16:18:22bluebrotherbut I don't know how you tell which file to use ... mtools works with "drive letters"
16:18:34peturimho, that's what attributes are for, not hiding based on the name
16:18:58bluebrotherthe h10 has started recently so it's pretty early.
16:19:24bluebrotherI like the hiding based on filename, makes a lot of things easier.
16:20:21amiconnImho hiding based on filename is an odd concept
16:20:54peturmaybe you can squeeze the write protect in the filename too? ;)
16:21:01daveone more question, does USB mode work atm? as when i connect i get a USB icon on iriver, but nothing on the PC?
16:22:02amiconnFunnily enough, there's one place where *nix and windows behave the opposite way: smb/samba shares. With samba you can set the hidden flag independent of the name, in winodws the name has to end with $ in order to be hidden
16:22:36peturyes, those $name shares are weird
16:23:07peturmost people don't know that $c is available by default unless you disable it ;)
16:23:15 Quit aegray (Remote closed the connection)
16:24:12 Join aegray [0] (n=aegray@74.135.163.80)
16:24:13amiconnIt's C$, and it's only accessible by administrators
16:25:11amiconn...and if simple sharing is enabled (default in WinXP, and the only option in WinXP home), the administrative shares don't exist
16:25:45*petur is dyslectic
16:26:06peturI've never made it it xp - w2k runs just fine and stable for me
16:26:22peturcrap - can't type too
16:29:39 Join KlrSpz [0] (n=klrSpz@69.15.248.2)
16:30:31 Quit aegray (Remote closed the connection)
16:30:48pregloww2k is the nicest windows, yes
16:30:50preglowdefinitely
16:31:09*linuxstb thinks there is still a lot of WMA work to do...
16:31:16preglowlinuxstb: no shit
16:31:19linuxstbI can't notice any IRAM usage either yet...
16:31:22 Join aegray [0] (n=aegray@74.135.163.80)
16:31:37linuxstbpreglow: I'm not even looking at the decoder itself, just the higher-level stuff.
16:31:47preglowlinuxstb: that was what i was talking about
16:31:50preglowbut first, someone make a patch :>
16:32:08preglowi think i'll have a look at it afterwards
16:32:17davewow just experienced gapless playback on iriver.. AT LAST! thankyou rockbox people
16:32:28preglowdave: sweet, innit
16:32:49davewas sooooo annoying when listening to mixes split up into tracks
16:32:55linuxstbSomeone in Apple must have tried Rockbox....
16:34:07davefastforwarding rewinding so much better.. this rocks! if this had the radio function I wouldn't go back to original f/w ever
16:34:12preglow"guys! you won't believe it! turns out we suck rod!"
16:36:41 Quit aegray (Remote closed the connection)
16:37:16 Join aegray [0] (n=aegray@74.135.163.80)
16:37:56aliaskUuuughhh. Just compiled an english.voice using some TTS from loquendo.com - they're so good.
16:38:40aliaskStupid microsoft sam.
16:38:45 Quit daurnimator (Read error: 104 (Connection reset by peer))
16:38:50vertic23lol - sucks
16:39:13 Join daurnimator [0] (i=quae@unaffiliated/daurnimator)
16:40:13daurnimator- like (eg) archos saying "Some people have made their own codecs for our earlier model player" etc
16:42:19 Quit aegray (Remote closed the connection)
16:42:55 Quit daurnimator (Read error: 104 (Connection reset by peer))
16:43:04 Join daurn [0] (n=quae@unaffiliated/daurnimator)
16:43:09 Join aegray [0] (n=aegray@74.135.163.80)
16:43:20 Nick daurn is now known as daurnimator (n=quae@unaffiliated/daurnimator)
16:43:42daurnimatorarg
16:44:10*daurnimator is VERY angry: THIS COMPUTER CANNOT UPLOAD ANYTHING OVER 50kb
16:44:35 Quit bluebrother ("Leaving")
16:49:49spiorfhi, any chance to use the scrollwheel with doom on the ipod?
16:51:26spiorfthe patch on flyspry don't apply anymore
16:51:48 Quit Zagor ("Client exiting")
16:55:57 Join Mmmm [0] (n=mscarrat@cpc4-hem13-0-0-cust438.lutn.cable.ntl.com)
16:56:46theli_ualinuxstb, i've created a page with some primary documentation here http://www.rockbox.org/twiki/bin/view/Main/PluginZXBox , i'll add more later
16:58:01*petur runs off to have some beers ;)
16:58:08 Quit petur ("burps")
16:58:55 Join Genre9mp3 [0] (n=yngwiejo@88.218.17.158)
16:59:00preglowfu!
16:59:20 Join theprodukkt [0] (n=Vito@O2d19.o.pppool.de)
16:59:29theprodukkthelp
16:59:39preglowok
17:00
17:00:03 Quit thegeek (Read error: 104 (Connection reset by peer))
17:00:10theprodukktwhats wrong with my rockbox when i get a "I05 at 3130ED9E" message while initializing my tagcache? (im using rb 060910)
17:01:11theprodukktsry, the adress is 3103ED9E
17:02:38*theli_ua runs to play soccer and drink beer ... bye all
17:02:45 Quit theli_ua ("Leaving")
17:05:10 Join leftright [0] (n=leftrigh@p549963CC.dip.t-dialin.net)
17:05:51 Join Edventure [0] (n=no@c-71-206-101-68.hsd1.mi.comcast.net)
17:08:11theprodukktthen rockbox turns off
17:08:15davecan i boot to original firmware? i tried holding the <- button during boot and it said "loading iriver firmware" but just hung and did nothing?
17:08:38theprodukkteven the tagcache.tmp is cleared to 0bytes
17:08:38preglowyou should hold record to boot retailfirmware
17:09:20theprodukktis there a site which explains all posibilities to access special functions (like that one) by holding keys pressed?
17:09:22leftrightany manual editors around ?
17:09:59daverecord? what button is that? i am on a h10?
17:10:15 Quit theprodukkt ()
17:11:14 Part leftright
17:11:34 Quit billytwowilly (Connection timed out)
17:11:41preglowdave: right, right, dunno then
17:12:25***Saving seen data "./dancer.seen"
17:12:45 Quit aliask ("Chatzilla 0.9.75 [Firefox 1.5.0.6/2006072814]")
17:13:04davenot to worry... i just wanted to check if i could get it to work, which i can...so reverting back to orig as i need to FM record tonight
17:13:09davethanks for all your help :)
17:13:12preglowheh
17:13:13preglownp
17:13:24preglowh10 coder seems to be away now
17:13:27preglowso no help to be had
17:13:44preglowi think i remember that the boot to original firmware feature doesn't work yet, though
17:13:57preglowthe h10 port is still very new
17:14:24daveyea understand that, am very grateful to everyone involved and all who helped me today
17:14:58daveit was only a few weeks ago that there was no workable release for h10, so things seem to be moving quickly
17:15:19davelets hope for a full release soon so i can finally say goodbye to the original fw
17:15:32preglowit's progressing fast, that's for sure
17:16:30 Quit JoeBorn ("Leaving")
17:18:35 Join webguest83 [0] (i=82cfedef@gateway/web/cgi-irc/labb.contactor.se/session)
17:19:34EdventureI keep getting the "tagcache is not ready" error on my ipod 5g - I just installed rockbox, is there something I'm doing wrong? or do I just have to wait for the cache to load?
17:20:10 Quit webguest83 (Client Quit)
17:20:18SlasheriEdventure: you need to initialize it first from the file view menu
17:20:30 Join ecE260 [0] (i=82cfedef@gateway/web/cgi-irc/labb.contactor.se/session)
17:20:48EdventureI did that, and it shows me a little circle up by the clock
17:20:51Slasheriand when disk has stopped spinning, reboot and it should work
17:21:03Slasheriok, then wait until that icon disappears and reboot
17:21:23Edventurek, i did that, but i'll try again
17:22:14 Quit dave (Remote closed the connection)
17:22:31Edventurewhen it starts up, it says "Committing tagcache 1/13" - it goes up to 3/13 then goes the artist, album, etc page
17:22:46Edventurei click on artist, and it says tagcache is not ready
17:23:32 Quit webguest92 ("CGI:IRC")
17:24:46Slasherihmm.. if it stops at 3/13, then something is really wrong
17:25:14Edventurewell thats no good
17:25:18Slasheriit would be great if you could try it with a logf enabled build to get the actual error message, or with a simulator
17:25:23Slasheribut now i need to go ->
17:27:58 Join billodo [0] (i=billodo@nat/sgi/x-593e22740f32ea06)
17:28:37 Join Rob2222_ [0] (n=Miranda@ACB4735E.ipt.aol.com)
17:28:47 Join Arathis [0] (n=doerk@p508A7C48.dip.t-dialin.net)
17:29:35 Quit Kohlrabi (Nick collision from services.)
17:29:45 Join Kohlriba [0] (n=Kohlrabi@dslc-082-082-129-236.pools.arcor-ip.net)
17:29:45 Quit Rob2222 (Read error: 104 (Connection reset by peer))
17:35:32 Join thegeek [0] (n=thegeek@s026b.studby.ntnu.no)
17:37:52 Join netmasta10bt [0] (n=torment@pool-71-251-101-72.tampfl.fios.verizon.net)
17:38:29 Quit [San] ()
17:40:54 Join KN|stiff [0] (n=phhome@p54B7F178.dip.t-dialin.net)
17:57:27 Quit ecE260 ("CGI:IRC")
17:57:38 Join MarcoPolo [0] (n=MarcoPol@che78-2-82-227-240-106.fbx.proxad.net)
17:58:33 Join daurn [0] (i=quae@unaffiliated/daurnimator)
17:59:48 Quit daurnimator (Nick collision from services.)
17:59:50 Nick daurn is now known as daurnimator (i=quae@unaffiliated/daurnimator)
18:00
18:00:43 Quit Pyromancer ("Ex-Chat")
18:01:27 Join Pyromancer [0] (n=pyromanc@dsl092-069-150.bos1.dsl.speakeasy.net)
18:02:54 Quit daurnimator (Client Quit)
18:07:14 Quit synic (Remote closed the connection)
18:08:01 Quit Criamos ("( www.nnscript.de :: NoNameScript 4.03 :: www.XLhost.de )")
18:26:03 Quit dan_a (Read error: 110 (Connection timed out))
18:36:27 Quit Pyromancer (Read error: 110 (Connection timed out))
18:36:40 Join Pyromancer [0] (n=pyromanc@dsl092-069-150.bos1.dsl.speakeasy.net)
18:37:51 Quit Landus (Read error: 110 (Connection timed out))
18:41:11 Quit Pyromancer ("Ex-Chat")
18:47:38 Join chromoXdor [0] (n=Miranda@p54B3EC18.dip.t-dialin.net)
18:49:22 Join Pyromancer [0] (n=pyromanc@dsl092-069-150.bos1.dsl.speakeasy.net)
18:50:10 Quit mantono ("Don't worry, be happy")
18:51:21 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
18:58:25 Quit dpassen1 ()
18:59:45 Quit Arathis ("[rl_bot quit]")
19:00
19:00:15 Join PaulJam [0] (n=pauljam@p54BCF2B6.dip.t-dialin.net)
19:00:24 Join dan_a [0] (n=dan-mirc@217.23.173.156)
19:05:57 Join Arathis [0] (n=doerk@p508A7C48.dip.t-dialin.net)
19:08:40 Quit tvelocity ("Αποχώρησε")
19:08:59 Join pixelma [0] (i=pixelma@82.193.235.34)
19:12:26***Saving seen data "./dancer.seen"
19:17:09PaulJama little off topic, but can anyone tell me if there are restrictions in filesize or length of video playback on the X5 in the retail firmware?
19:20:38 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
19:21:35 Quit Arathis ("[rl_bot quit]")
19:25:57 Join bytie [0] (n=anonym@Q2e54.q.pppool.de)
19:30:20 Join Shomakom-ink169 [0] (n=Shomkom@212.117.141.138.static.012.net.il)
19:30:41 Nick Kohlriba is now known as Kohlrabi (n=Kohlrabi@dslc-082-082-129-236.pools.arcor-ip.net)
19:31:39 Quit Shomakom-ink169 (Client Quit)
19:33:16 Join webguest97 [0] (i=c3ec4b1b@gateway/web/cgi-irc/labb.contactor.se/session)
19:37:34 Quit bytie ("Dana")
19:45:23 Join lee-qid [0] (n=liqid@p54967B39.dip.t-dialin.net)
19:55:07 Quit lee-qid (Read error: 104 (Connection reset by peer))
19:55:55 Join lee-qid [0] (n=liqid@p54967B39.dip.t-dialin.net)
19:56:26 Join merbanan [0] (n=banan@c80-216-155-218.cm-upc.chello.se)
19:57:48 Quit Nibbler (Read error: 110 (Connection timed out))
20:00
20:01:54 Join Nibbier [0] (n=sven@e181104237.adsl.alicedsl.de)
20:04:56 Quit lee-qid (Read error: 54 (Connection reset by peer))
20:08:17 Quit PaulJam (".")
20:08:48 Join Tazsie [0] (n=jmezach@cc7304-b.deven1.ov.home.nl)
20:08:54 Part chromoXdor
20:10:59 Quit Edventure (Read error: 104 (Connection reset by peer))
20:14:42 Nick ScoTTie_ is now known as ScoTTie (n=scott@unaffiliated/scottie)
20:15:48 Quit Nibbier (Read error: 60 (Operation timed out))
20:19:12 Join lee-qid [0] (n=liqid@p54964F6C.dip.t-dialin.net)
20:21:57 Quit Taz^ (Read error: 110 (Connection timed out))
20:22:49 Join damaki [0] (n=Chocolat@ALille-253-1-3-77.w90-7.abo.wanadoo.fr)
20:24:26 Quit Mmmm (Remote closed the connection)
20:24:38 Join Nibbier [0] (n=sven@e181107167.adsl.alicedsl.de)
20:25:01 Quit lee-qid (Read error: 104 (Connection reset by peer))
20:26:49 Join lee-qid [0] (n=liqid@p54964F6C.dip.t-dialin.net)
20:29:09 Quit HCl ("Lost terminal")
20:34:30 Quit damaki (Read error: 104 (Connection reset by peer))
20:34:31 Quit lee-qid (Read error: 104 (Connection reset by peer))
20:39:02 Join lee-qid [0] (n=liqid@p54964F6C.dip.t-dialin.net)
20:42:56 Part ScoTTie
20:45:54 Join barrywardell [0] (n=barrywar@194.46.178.162)
20:46:37 Join damaki [0] (n=Chocolat@ALille-253-1-3-77.w90-7.abo.wanadoo.fr)
20:47:22 Quit lee-qid (Read error: 104 (Connection reset by peer))
20:49:47 Join lee-qid [0] (n=liqid@p5496537E.dip.t-dialin.net)
21:00
21:03:23 Join karim [0] (n=karim@ip-7.net-82-216-143.rev.numericable.fr)
21:09:50 Join lee-qid_ [0] (n=liqid@p54965848.dip.t-dialin.net)
21:12:30***Saving seen data "./dancer.seen"
21:17:48 Quit lee-qid (Read error: 110 (Connection timed out))
21:18:19 Quit lee-qid_ (Read error: 104 (Connection reset by peer))
21:19:02 Join lee-qid [0] (n=liqid@p54965848.dip.t-dialin.net)
21:27:55 Quit lee-qid (Read error: 104 (Connection reset by peer))
21:31:41 Join donut [0] (i=502958ac@gateway/web/cgi-irc/labb.contactor.se/session)
21:31:48donuthey all
21:31:56 Quit billodo ("leaving")
21:33:18 Quit donut (Client Quit)
21:35:46 Join lee-qid [0] (n=liqid@p549667EC.dip.t-dialin.net)
21:36:50 Quit barrywardell ()
21:40:25Slasheripreglow: btw, about the scheduler patch. Did you find that the UI was always more laggy or just when buffer goes to critical?
21:40:54Slasheribecause with iriver i can't test on same environment
21:41:15 Quit lee-qid (Read error: 104 (Connection reset by peer))
21:46:34 Quit ender` (" How many AOL users does it take to change a light bulb? Eleven. One to ask to be on the light bulb gif mailing list, nine to")
21:50:27 Join EV|Toshiba [0] (n=Testing@pool-72-64-146-8.tampfl.fios.verizon.net)
21:53:54 Join ender` [0] (i=null@84.52.165.220)
21:54:41 Quit aegray_ (Read error: 110 (Connection timed out))
21:56:35 Quit _Veseliq_ (Remote closed the connection)
21:56:42 Join |AhIoRoS| [0] (n=ahioros@201.224.122.227)
22:00
22:00:47 Join Febs [0] (i=2662c44b@gateway/web/cgi-irc/labb.contactor.se/session)
22:01:09preglowSlasheri: it's always a bit sticky feeling at high boost rates (~50%), but sometimes it just stopped responding altogether
22:01:18preglowSlasheri: in periods, so i assume it has to do when the buffer went critical
22:08:40 Join XxBigP123xX [0] (i=182e1507@gateway/web/cgi-irc/labb.contactor.se/session)
22:10:52XxBigP123xXhas anyone here bought the ipod games?
22:12:22 Quit XxBigP123xX (Client Quit)
22:14:08 Quit dongs (Remote closed the connection)
22:14:12 Join dongs [0] (n=HPUX@i249222.ppp.asahi-net.or.jp)
22:20:25Slasheripreglow: what if you boost cpu manually, can you feel any difference then?
22:21:31Slasheriif you find it different, then it could be that buffer goes critical too fast
22:22:04dionoeahi
22:22:43 Join Davide-NYC [0] (n=chatzill@user-12hdtm5.cable.mindspring.com)
22:23:56*Febs reads the logs from last night and is highly entertained by preglow's comments on the hardware EQ.
22:24:56*dionoea noticed that enabling the EQ just made it sound louder ... not especially better
22:27:10Davide-NYCY'all saw that tarfile with the WMA decoder buried in it?
22:27:45Davide-NYChttp://www.rockbox.org/tracker/task/4984
22:28:16Davide-NYC(not that I care about WMA, but it would be another feather in the RB cap)
22:29:59 Join hardeep [0] (n=hardeeps@c-71-202-85-199.hsd1.ca.comcast.net)
22:30:55preglowFebs: yes, a prime moment
22:32:16hardeephmmmm, forums down?
22:32:23preglowyup
22:34:44Febspreglow: I never noticed the default to +12db. I'm thinking that was introduced at some point after I first tried the hardware EQ, because I surely would have noticed.
22:34:50Davide-NYCthey work for me. Albeit slow
22:35:04Davide-NYC*the forums
22:35:09 Join matsl [0] (n=matsl@1-1-4-2a.mal.sth.bostream.se)
22:35:11FebsBut I assumed, wrongfully apparently, that when people were talking about how the hardware EQ sounded so great, it was because they were actually adjusting the EQ.
22:35:43FebsNot just turning it on and enjoying super bass boost.
22:35:58FebsNo wonder some people complained of distortion.
22:37:10linuxstbFebs: Yes, I think the gain (-12db to +12db) was originally stored as an unsigned number from 0 to 24 (with 12 as the default, representing 0dB). It was then changed to a signed number (-12 to +12), but the default was kept as 12, which then meant +12dB.
22:37:33webguest97Slasheri: Can you explain how does the timeout part of waiting-with-timeout work in your new scheduler? I don't quite understand how the timeout wakes the waiter.
22:38:04Febslinuxstb, that makes sense.
22:38:45Soapand the golden ear club couldn't tell that this was an artificial sound, not a "truer" sound?
22:38:49 Join Landus [0] (i=Landus@70-100-181-192.dsl1-erie.roc.ny.frontiernet.net)
22:39:07linuxstbSoap: Don't spoil the head-fi discussions with facts...
22:39:54SoapI gotta, I can't speak their language.
22:40:29 Join webguest91 [0] (i=d53dc019@gateway/web/cgi-irc/labb.contactor.se/session)
22:41:21Slasheriwebguest97: it works similar way than sleeping tasks
22:41:40Slasherisame code wakes up both sleeping and blocked with timeout tasks
22:42:07Davide-NYCQuestions: What is 'zxbox', how do I access it and does it exist for the Sim?
22:42:24webguest91Hi, I have a process related question. In the bug/patch tracker there are many patches that are just there −− with little chances to be committed. Many are very useful. But if a patch is not committed quickly chances are good that it will become outdated.
22:42:39linuxstbDavide-NYC: http://www.rockbox.org/twiki/bin/view/Main/PluginZXBox
22:42:52webguest97Slasheri: But isn't it only iterating through the sleep list? I must be reading something wrong...
22:42:54Davide-NYClinuxstb: THanks
22:43:21webguest91So wouldn't the following be useful: before committing a patch, it's discussed e.g. on IRC. And if it's found useful it's committed within one or two days.
22:43:28Slasheriwebguest97: yep, and blocked with timeout tasks are stored in sleeping list :)
22:43:54webguest97Ah ... that explains quite a bit :-)
22:44:00Slasheriwebguest97: true, it has already discussed quite a lot here..
22:44:16Slasherithat's why i have already made 10 revisions of it :)
22:44:46webguest91Slasheri: do you mean webguest91? (patch handling?)
22:45:15Slasherioh, yes.. those webguest nicks are just confusing :)
22:45:25webguest91Slasheri: sorry, I haven't followed the previous discussion
22:45:43*dionoea whispers something about /nick
22:45:57Davide-NYCLOL I thought "spectrum emulator" could have soehow been related to "spectrum analyser" and that zxbox was some sort of audio tool extraordinaire.
22:46:07markunpreglow: I'm trying to create a patch from the wma codec dir but cvsdo is being a bitch
22:46:22Davide-NYCOh well, gets added to the 'nofun' patch and removed. :-(
22:46:32linuxstbDavide-NYC: I guess the Spectrum wasn't popular in your part of the world?
22:46:37webguest91Slasheri: "10 revisions" −− isn't it frusrating? It's a waste of energy and time! Commit quickly or don't implement. Or just outline the solution in words.
22:47:21Davide-NYCI don't know, but unless it has serious uses I'm not into it.
22:47:21Slasheriwebguest91: well, there has been a lot of improvement
22:47:29Slasheriit's not a good idea to commit a broken patch
22:49:14webguest91Slasheri: yes. But there are quite a few useful and non-breaking patches (e.g. 5907) that don't get committed and stale
22:50:35linuxstbwebguest91: The author of #5907 has cvs commit access himself, so I don't know why he hasn't just committed it...
22:50:47webguest91Slasheri: ...and fonts, and ... But the devs are working on their own things (which is more fun). They should dedicate, say, 20% of their time to patch monitoring/committing
22:51:29webguest91Slasheri: ehm... didn't know that
22:53:35webguest91But in any case: RockBox is a nice piece of software!
22:54:17 Quit aegray ("Lost terminal")
22:57:09 Part webguest91
22:57:54preglowlinuxstb: have someone posted the new revelations to them?
22:57:58preglowi've lost the url to it
22:59:36linuxstbNo, no-one seems to have mentioned it: http://www6.head-fi.org/forums/showthread.php?t=194390&page=5
23:00
23:01:12preglowAll I did was "turn on" the Hardware EQ and nothing else. The sound is amazing!
23:01:15preglowi cabn't read this
23:01:16 Join bytie [0] (n=anonym@p548CAD3C.dip0.t-ipconnect.de)
23:01:27preglowi'm so thorougly annoyed by audiophile people
23:01:27preglowhahaha
23:02:34 Quit bytie (Client Quit)
23:05:10 Join bytie [0] (n=anonym@p548CAD3C.dip0.t-ipconnect.de)
23:07:45 Join rotator [0] (n=e@rockbox/developer/rotator)
23:08:32 Quit maddog012 (Client Quit)
23:09:24 Quit Davide-NYC (Remote closed the connection)
23:09:37dionoeadoes (-1)<<1 equal 2 or -2 ? or is that architecture dependant ?
23:10:18 Join skymt0 [0] (n=skymt0@pool-71-241-29-60.norf.east.verizon.net)
23:10:41Bagderits a weird thing to do to start with
23:10:50preglowwhy?
23:10:57preglowi do stuff like that all the time
23:11:11Bagderand -1 then equals ~0 ?
23:11:11preglowand it's -2, should always be
23:11:21BagderI don't think ansi c defines that
23:11:22dionoeaso bitshifts keep the sign ?
23:11:31preglowdionoea: left shift always does
23:11:35preglowdionoea: not so with right shift
23:11:40dionoeahum ... ok
23:11:52preglowleft shift doesn't have to keep anything
23:12:03preglowan arithmetic left shift is the same as a logical left shift
23:12:09dionoearight ?
23:12:09preglowbut with a right shift, you need to copy the sign bit
23:12:16dionoeaoh ok :)
23:12:18preglowand that isn't defined by c
23:12:25preglowhowever, it almost always works like you want
23:12:34***Saving seen data "./dancer.seen"
23:12:35dionoeaalmost ...
23:12:46preglowit's compiler dependent
23:12:55preglowlike i said, it's not defined in ansi c
23:12:57dionoeawhat does gcc do ? :)
23:13:01preglowthe correct thing
23:13:05dionoeaok, great
23:13:11preglowit used an arithmetic shift for signed numbers, and logical shift for unsigned numbers
23:13:23preglowBagder: but why is that a weird thing to do?
23:13:28Bagder"The result of signed right shift of a negative number is implementation defined"
23:13:33preglowBagder: i need to do stuff like that all the time to correct fixed point formats
23:13:46BagderI just find it weird
23:13:56Bagdernegatives and shifts are not combined in my brain
23:14:13skymt0is this a good place to ask for help, or should I head to the forums?
23:14:32skymt0I ask because it seems dev-focused right now
23:15:00preglowBagder: it's good old two's complement math :)
23:15:15preglowskymt0: just ask ahead and see what happens
23:15:22preglowskymt0: not only developers hang here
23:15:49Bagderwell, I think I was more thinking about the problems with >> if the left operand is negative
23:16:02skymt0Ok. I just installed rb on my 1g mini, now it refuses to mount in ubuntu linux. Help?
23:16:03preglowyup, and that's a real thing to be cautious about
23:16:08 Join webguest37 [0] (i=917408a3@gateway/web/cgi-irc/labb.contactor.se/session)
23:16:34preglowskymt0: what does the ipod screen say?
23:16:51skymt0it just has the normal rb interface
23:16:51 Join safetydan [0] (n=dan@cor8-ppp2595.for.dsl.connect.net.au)
23:17:00skymt0with the fs menu, settings and all
23:17:08dionoeawhen plugged to usb ?
23:17:13skymt0fw
23:17:17preglowskymt0: ah, ok, then it's not registering usb, i don't know if that's normal for 1g mini
23:17:29preglowskymt0: you can always just shut down rockbox and insert the usb, i think
23:17:37preglowskymt0: either that, or you need to manually enter disk mode/retailos
23:17:48dionoeapreglow: he's using firewire it seems ...
23:17:53linuxstbskymt0: Rockbox doesn't detect firewire - you'll need to manually reboot to disk mode or retailos.
23:17:58preglowright, forget me
23:18:00skymt0ok, thanks
23:18:01dionoea:)
23:18:04skymt0bye
23:18:07 Quit skymt0 ("Leaving")
23:20:08 Quit dan_a (Read error: 110 (Connection timed out))
23:21:21safetydanHas someone made a patch from that WMA tree yet?
23:21:31dionoeamarkun was making one
23:23:11safetydanlooks like a very old rockbox tree
23:23:18safetydangoing to take a while to pull out the changes
23:25:45 Join solarflare_ [0] (n=supernov@MKCMLVI.gprs.saunalahti.fi)
23:27:49preglownot too many changes to the main three, i'd expect
23:27:54 Quit Febs ("CGI:IRC (EOF)")
23:27:55preglowtree, even
23:28:15safetydanyou'd hope not
23:29:33 Quit solarflare (Connection timed out)
23:29:44 Quit Galois (Remote closed the connection)
23:30:13 Join Galois [0] (i=djao@efnet-math.org)
23:33:21 Join stevenm [0] (n=stevenm@infranelson.student.umd.edu)
23:33:48preglowsafetydan: the lower eq band is buggy, btw
23:33:59preglowit seems precision problems live to annoy me also outside of the grapher
23:34:13stevenmHello. I may have the iriver coming to me again, if my lazy friend remembers.
23:34:35stevenmI am unsure if I commited this, but.. does the midi player skip to the first note in the file automatically, or is there usually a delay?
23:34:45safetydanpreglow, how come no one noticed until now? and is there a solution?
23:35:41merbanansafetydan: what wma tree ?
23:35:57stevenmeq, eh? How is that implemented? fft, or filter, or some other way?
23:36:02safetydanmerbanan, the one that was posted to flyspray
23:36:06preglowstevenm: iir filters
23:36:15preglowsafetydan: because it's hard to trigger
23:36:24merbanansafetydan: do you have a link ?
23:36:28stevenmpreglow, aah cool! I'm finally in the second part of the DSP class, actually know what that is :)
23:36:28preglowsafetydan: you need to play a file, then play a file with complete silence, and sometimes it'll trigger and make a dc offset
23:36:34preglowstevenm: heh
23:36:39Bagdermerbanan: http://www.rockbox.org/tracker/task/4984
23:36:47safetydanmerbanan, http://www.rockbox.org/tracker/task/4984 but it's not usable yet
23:36:49preglowstevenm: i implemented the eq as a usual cascade combination of five second order iir filters
23:36:54preglowstevenm: nothing fancy
23:37:00stevenmpreglow, aah I see
23:37:06 Quit solarflare_ (Read error: 104 (Connection reset by peer))
23:41:21 Join hcl [0] (n=hcl@hecate.student.ipv6.utwente.nl)
23:43:23 Join Ally [0] (i=504df123@gateway/web/cgi-irc/labb.contactor.se/session)
23:44:16webguest97Has there been any progress on the Sansa port lately?
23:44:16 Quit Ally (Client Quit)
23:44:25preglownope
23:45:36 Join ally [0] (n=chatzill@80-77-241-35.dsl.xifos.net)
23:45:39 Quit KlrSpz ()
23:45:44allyevening.
23:47:17safetydanbah, lousy X11
23:47:18 Quit safetydan ("Ex-Chat")
23:47:31allyjust updated the H10 WPS wiki, not sure if I've got it quite right, if someone one wants to take a look and edit feel free :p
23:48:01 Quit _FireFly_ ("Leaving")
23:48:29 Quit ally (Client Quit)

Previous day | Next day