Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2005-06-25

00:00:00Slasherii think some people might want to have gapless switching (at least i heard that on this channel)
00:00:06t0masI guess 95% of the users would like to song to start when they hit the button... and don't care about a few ms. gap between manually switched tracks
00:00:17t0mas*the song
00:00:43Slasheritrue.. maybe you could enable that function until audiobuffer flushing is implemented
00:00:55LinusNSlasheri: rockbox had "gapless" switching a long time ago, but we removed it because the ui feedback sucked because of it
00:00:55Slasherii cannot change the code right now until i finnish my dsp stuff
00:01:06t0masLinusN: same problem now
00:01:23t0masSlasheri: I can... what should I check to see if we're crossfading?
00:01:55t0mas(this seems not very usefull:)
00:01:56t0masbool pcm_is_crossfade_enabled(void)
00:01:56DBUGEnqueued KICK t0mas
00:01:56t0mas return false;
00:02:02SlasheriLinusN: hmm, ok. however, i think with proper pcm buffer flushing the gapless switching could be made on iriver without any negative effects
00:02:45SlasheriHmm, it _should_ have return crossfade_enabled;
00:03:00Slasheriif it doesn't, please commit that fix :)
00:03:27CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:03:27*t0mas cvs updates... maybe I'm missing something
00:03:47t0mashm.. not changed
00:04:14t0masbool pcm_crossfade_init(void)
00:04:14***Alert Mode level 1
00:04:14t0mas return false;
00:04:14***Alert Mode level 2
00:04:27Slasheriweird.. i checked even cvs diff and it seems to be ok
00:04:41t0masI have 1.64
00:04:48t0massame as in webcvs...
00:05:00Slasherieh, you have something wrong in your sources..
00:05:10t0massomething wrong in CVS then...
00:05:11Slasheriare you sure you checked pcm_playback.c?
00:05:21Slasheriok, those are simulator stubs in playback.c :D
00:05:21*t0mas wakes up
00:06:51*t0mas starts looking at the right file :X
00:10:03amiconn+800 byte on iriver...
00:10:08 Join preglow [0] (
00:10:15preglowhi ho
00:10:35t0masoh... tnx amiconn, I remember a "Linus question"
00:10:44t0mashow big is the flash rom in the iriver?
00:10:53t0masand how big is the original firmware
00:11:13LinusNthe flash is 2mbyte
00:11:18t0mas<amiconn> +800 byte on iriver... <−− remembered it because of that
00:11:58t0masLinusN: and do you think it's possible to have both rockbox and iriver in flash?
00:12:42LinusNthere is about 64kbyte left iirc
00:13:01LinusNso the answer is no
00:13:22t0mashm... we need at least 200 kb right?
00:13:45amiconnThe iriver firmware is running directly from rom?
00:13:51 Quit TCK (Read error: 110 (Connection timed out))
00:14:11LinusNamiconn: no, but the iriver boot does
00:14:15***Alert Mode OFF
00:14:20LinusNit copies itself to ram
00:14:31LinusNbut it's not as simple as on the archos
00:14:52amiconnDoes that happen before the rockbox bootloader kicks in?
00:15:43 Join Jleagle [0] (
00:16:36t0masSlasheri? any idea what's causing it then? :)
00:17:02Slasherit0mas: hmm, causing what? :)
00:17:10t0masthe delay
00:17:32Slasheriah, it's because pcm buffer has buffered data from previous track
00:17:59Slasherithat should be flushed from right position after there are few chunks buffered from next track
00:18:09t0masbut it switched quicker before..
00:18:27Slasheriyes, because before it cleared the entire buffer
00:18:36*amiconn was lucky
00:18:37Slasherinow it doesn't do that without pcm_play_stop
00:18:40amiconnNo clash
00:18:42t0masok, maybe that is the desired behaviour atm?
00:19:22Moosamiconn : congratulations :D
00:19:41Slasheriand the initial delay before track starts playing is caused by disk spinup & codec load time and initial pcm buffering (4 chunks)
00:20:01t0masyes, but that was a short period...
00:20:18t0masfrom the filebrowser the disk is nomally spinning... so it was only loading a bit
00:20:43Slasherihmm, yes.. i can debug that later. maybe it could be made a little faster
00:23:30 Quit Jleagle ("Chatzilla [Firefox 1.0.4/20050511]")
00:24:25preglowLinusN: why does it copy itself to ram? i thought you said executing code from flash is actually faster
00:24:50t0masand if I change line 1010 in playback.c (not pcm_playback) from //pcm_play_stop() to "if (!pcm_is_crossfade_enabled()) pcm_play_stop();"
00:27:12Slasherit0mas: you can do that, it fixes the delay when crossfading is not enabled
00:27:21Slasherii can do a better fix later
00:27:27t0masyes, you should
00:27:29Slasherinow nights to all :)
00:27:35t0masit's better to do some sort of buffer flush
00:27:40t0maswith crossfading in mind...
00:27:59t0masI'll test this... and commit
00:28:05t0masto keep people happy about the speed :)
00:28:17t0masand you can change it back when there's a better solution
00:28:42LinusNpreglow: ask iriver
00:29:09LinusNpreglow: besides, the sdram is faster, since it can burst from sdram
00:29:59preglowyou can't burst from rom? :/
00:30:16preglowdoes that hold generally, or is it just a "feature" of that particuilar chip?
00:31:08LinusNfew flash roms have burst modes that are faster than consecutive reads
00:31:47LinusNflash roms have static addressing, not row/column
00:32:03 Quit TCK- (Read error: 145 (Connection timed out))
00:32:07preglowtoo long since i dabbled in this kind of thing
00:32:13amiconnOkay, buckle up...
00:33:22preglowdavid bryant's done some analyzing of how iriver does resampling
00:33:27preglowand they seem to do quite fancy stuff
00:33:37preglowmore than i would have expected
00:34:19amiconnlet's hope I didn't break anything
00:34:36LinusNpreglow: like what?
00:35:44preglowfor downsampling, they use a pretty high order (i guess) iir filtering, then linear interpolation decimation
00:35:57preglowfor upsampling they use a 64 tap ideal lowpasss fir filter
00:36:40preglowall of this is very possible, i just expected them to do some cheap ass stuff like they do with their playback speed handling, heh
00:37:22LinusNthe resampling seems to screw up the playtime calculations
00:37:34preglowour resampling?
00:37:36LinusNso 48khz tracks end prematurely
00:37:42preglowno surprise
00:37:51preglowi didn't test it much at all, i just intended for it to be put somewhere else
00:37:56preglowwhich is what slasheri is doing at the moment
00:38:32LinusNsome guy was so happy woth the resampling, because he ripped all his cd:s with 48khz :-)
00:38:49preglowkudos to him
00:38:54preglowi wonder how he manages it
00:39:06LinusNand he probably thought it gave him better quality :-)
00:39:30LinusNi know rippers that can resample
00:39:40LinusNgod knows why though
00:40:25HClodd people
00:40:34LinusNamiconn: so one should always set DRAWMODE_SOLID as some kind of default?
00:40:38amiconnLinusN: A part of the code size increase comes from the fact that I already implemented lcd_hline() and lcd_vline(). What I did not yet do is to actually use them in places where they could be used
00:40:55amiconnThey're optimised of course
00:41:08amiconnLinusN: Yes, DRMODE_SOLID is the default
00:41:32preglowbtw, i made speex compile under rockbox yesterday
00:41:47preglowdidn't do much, but enough to see it uses floating point functions, even in fixed point mode :/
00:41:48amiconnOnce I am done with that I'll update the wiki page to reflect current design
00:42:15amiconnpreglow: flixed point ;)
00:42:48preglowi think it's just for initialising sine tables and the like
00:42:55preglowit does a hell of a lot of allocing
00:44:48 Join kenshin [0] (
00:44:53LinusNamiconn: tLCDPixelFunc ???
00:45:07t0masSlasheri: the gap is now just a few miliseconds
00:45:11t0masbut the wps is still lagging
00:45:20 Join TCK [0] (
00:45:24t0mas(probably because the next track wasn't buffered yet)
00:45:32LinusNi don't like typedefs very much, especially not when they're mixed case
00:45:35amiconnI know that typedefs shouldn't be over-used, but these function pointers would be veeery cumbersome without
00:46:12amiconnAnd that's also why I used mixed case - to indicate that it's something unusual
00:46:22amiconnI did so in the grayscale lib as well
00:46:27LinusNand hungarian notation is forever banned in my world
00:46:43amiconnHmm, hungarian notation?
00:46:48t0mashm? the sString nInt thing?
00:46:52LinusNthe "t" prefix
00:46:59t0masor t_type
00:47:25amiconnLinusN: How would you suggest to call it then?
00:47:40t0masgood night
00:47:50t0mastime to enjoy some music and read a book... and sleep :)
00:48:24LinusNnite t0mas
00:48:43amiconnLinusN: That might be misleading, because there is also an array of those function pointers, and it will be exposed in the plugin api allow plugin library functions like lcd_filltriangle()
00:49:13LinusNwhy is it confusing?
00:49:26amiconnOne could think this is actually a function
00:49:33amiconn...or something like that
00:49:44LinusNthen he should read a book about c
00:49:56preglowi can't imagine a context where i can mistake i typedef for a function
00:50:15amiconnThere are some typedefs concerning the file functions, they're called like xyz_type
00:50:26LinusNi can live with that
00:51:30preglowdefinitely beats tLolFunc, yes
00:51:31amiconnsounds good
00:51:50amiconnI'll change that, and fix the fm builds
00:51:53LinusNbtw do you think a function pointer is faster than a switch?
00:52:19amiconnYes, especially as they're called multiple times after being dereferenced
00:52:27LinusNof course
00:52:28preglowi think they most definitely are
00:53:59LinusNbut how often are they dereferenced directly?
00:54:25LinusNaren't you supposed to use the lcd_drawpixel() function?
00:54:55amiconnlcd_drawpixel() is the only function that dereferences it and then uses it only once
00:55:27LinusNi see
00:55:53amiconnThere are pixel functions for single pixels; these are used for pixels and free lines only
00:56:26amiconnFor all other primitives there are block functions which deal with whole pixel blocks equalling a framebuffer byte
00:57:19 Quit Aison ("( :: NoNameScript 3.72 :: )")
00:58:19amiconn...and lcd_fillrect does a special optimisation using memset() when applicable
00:58:30LinusNamiconn: yellow! yellow! yellow! :-)
00:58:37amiconnI know...
00:58:43amiconn[00:51:51] <amiconn> I'll change that, and fix the fm builds
00:59:14 Join muesli- [0] (
01:00:57muesli-re girls
01:01:25preglowk, k, something tells me the speex fixed point port isn't finished
01:06:54muesli-hai = (engl. shark) preglow ;)
01:08:22elinenbe|sleepinamiconn: nice graphics lib stuff... looking REALLY nice
01:08:33 Nick elinenbe|sleepin is now known as elinenbe|sleptto (
01:08:47 Nick elinenbe|sleptto is now known as elinenbe_ (
01:09:24amiconnThanks :)
01:10:30amiconnNext steps are (1) convert lcd_bitmap(), (2) apply all that stuff to the remote lcd driver, (3) Switch to 4-grey mode on iriver
01:11:02amiconn(4) adapt player graphics library, (5) rework greyscale lib
01:11:05muesli-yeah, gray rulez :D
01:11:21amiconn(6) add extra functions to the plugin lib
01:11:32Moosit's very promising amiconn :)
01:11:38amiconn(7) add solid mode to cube.c, like ipodlinux
01:11:57LinusN(8) get a life :-)
01:12:09muesli-(9) celebrating amiconn
01:12:53Moosall this needed lot of works good luck amiconn ;)
01:14:14LinusNamiconn: looks really good
01:14:17preglowmuesli-: i fail to see why you're educating me in my own language...
01:14:52muesli-so, you're german??
01:14:53LinusNmaybe he meant "hej"
01:15:19muesli-tak tak :D
01:15:23preglowi didn't get what you meant
01:15:25preglowno, i'm norwegian
01:15:28preglowand it's "hei"
01:15:29amiconnLinusN: Something must be wrong with the build error detection scripts
01:15:43amiconnThe FM builds were only yellow, yet they failed to link...
01:15:46preglowamiconn: that's well-known
01:16:04muesli-so hai its in norwegian shark as well?
01:16:16preglowmuesli: yup
01:17:01muesli-i have the feeling german and all scandinavian languages are quit similar
01:17:04preglownorwegian and german do have a ton of simiular words
01:17:05amiconnLinusN: There will be no more typedefs than these 2, but the definitions will be a little different for 4-grey mode
01:17:26LinusNi see
01:17:46preglownorwegian also has a ton of german words we inherited from trade with germans, heh
01:18:09muesli-some uk's here?
01:18:20muesli-or any other english natives?
01:18:37preglowi'm half english-native :>
01:18:47amiconnLinusN: Maybe I need to duplicate the defs for iriver, because the remote is b&w while the main lcd is not, once I get to that
01:18:55muesli-does "zukunftsangst" make sense to somebody?
01:19:25LinusNamiconn: probably
01:19:39preglowmuesli: well, i know what it means...
01:20:55muesli-thats one of three german words which were not possible to translate...
01:21:17elinenbe_amiconn: i expect you to get through that list by the end of the weekend :)
01:21:33amiconnelinenbe: Not really
01:21:35elinenbe_no, i kid... probably by the end of the week!
01:22:25 Join rasher [0] (
01:22:43muesli-cheers rasher
01:23:24amiconnLinusN: Forgot to remove some now-unused #defines...
01:23:33rasherSo.. the matrix plugin ("> behaves differently in the sim - correctly!
01:23:41rasherwhich confuses me
01:24:44rasherAnyone have any ideas what might be different? On the iriver the animation just scrolls a few lines, then goes blank (no lines appear), in the sim, it continues
01:25:21Plugh_On Tuesday, the 6th Circuit Court of Appeals in Cincinnati turned the playing field for digital sampling completely upside-down. The court ruled quite simply that any and all samples used must paid for, even if they sound completely different from their original. Previously, if a digital sample wasn't identifiable from it's original it was perfectly legal.
01:26:00rasherwell, if it isn't identifiable, good luck proving any use
01:26:07elinenbe_rasher: what's different with the rockbox on yout site?
01:26:31Plugh_rasher: the allegation alone can stop music from ever being released
01:26:42rasherelinenbe: nothing really.. a couple of extra fonts, and the matrix plugin - plus it's outdated
01:26:45Plugh_lawyers ain't cheap
01:26:49rasher(by a few hours)
01:26:56rasher(or half a day, more like)
01:27:23preglowamericans are already digging themselves deep
01:27:27preglowi'll just let them continue
01:28:43rasheroh, and I changed the icon of .rockbox files to an iriver-i!
01:29:04amiconnrasher: boooh!
01:29:22rasherwell having a 6x7 pixel archos jukebox didn't seem very appropriate either
01:29:47amiconnThen we need a rockbox icon..."> has an Rb
01:30:03rasherbut it's not very readable
01:30:45amiconnBtw, I never thought that something should symbolise an archos
01:31:04rasherIsn't that the idea?
01:31:09rasherlooks like one to me
01:31:48LinusNit's an X for eXecutable, iirc
01:32:18amiconnMight be an archos, but then it is a player or rec v1
01:32:28rasherThat's what I thought
01:32:32amiconnThat's not suited for the other archoses either
01:32:38LinusNfunny, i've never really looked that closely
01:33:36rasheranyway, the filename was .iriver, soooo
01:35:28amiconnUgh, speed mode in cube.rock looks really funny on iriver
01:36:18rasherit almost looks like a solid shape
01:36:24rasheror at least a wireframe shape
01:36:53amiconnIn speed mode you can't recognise much...
01:37:04amiconnStart cube, press select
01:37:20rasherif you have it rotated the right way, and only rotating along one axis
01:37:27rasherit looks like a wireframe shape
01:37:27 Join darthmall [0] (
01:38:03rasherelinenbe should know about the icon:
01:38:22rasherwell, except that it was 2,5 years ago
01:39:37muesli-will go 2 bed..g'night matey
01:42:06preglowthink i'll join you in that
01:42:12preglowmy head feels like it's bloody exploding
01:42:16preglowlater all
01:42:26 Part preglow
01:42:44rasherWhatever happened to the rockblox patch?
01:43:32LinusNi worked on it, but got other things to do, and forgot about it
01:44:03***Saving seen data "./dancer.seen"
01:45:25rasherhaha, the patch changes the entire file
01:48:08 Join Tomhab [0] (
01:48:13 Part Tomhab
01:48:27amiconnLinusN: Btw, my new routines have full clipping support
01:49:21 Quit Moos ("Bye ALL")
01:53:09 Quit muesli- (Read error: 60 (Operation timed out))
01:55:21LinusNtime to sleep
01:55:23LinusNnite all
01:55:35 Part LinusN
01:55:58 Quit Cassandra ("If I were actually witty, this quitline would be funny.")
01:56:20 Quit darthmall ()
02:03:37amiconnrasher: Yes, I have planned that for the plugin lib
02:03:48amiconnI already have a working implementation
02:03:57amiconn(For testing only)
02:05:38rasherfun, this mouse claims that its wheel and middle-button won't work outside of Windows
02:05:45rasheryet it works just dandy under Linux
02:06:15amiconnPerhaps it's an indoor-only mouse? ;-)
02:06:23rasherBog-standard HID mouse.. I wonder why they'd write something like that
02:14:31 Join amiconn_ [0] (
02:14:56TCKI would like to take this moment to say that Strawberries And Cream Kit Kats are the win.
02:14:59 Quit amiconn (Nick collision from services.)
02:15:00 Nick amiconn_ is now known as amiconn (
02:37:31 Quit Stryke` (Read error: 145 (Connection timed out))
02:49:38 Join Febs [0] (
02:54:55 Quit kenshin (Read error: 110 (Connection timed out))
03:23:25 Join kenshin [0] (
03:44:06***Saving seen data "./dancer.seen"
03:44:19 Quit _aLF ("Leaving")
03:53:39 Quit rasher ("CGI:IRC (EOF)")
04:00:40 Quit kenshin (Read error: 110 (Connection timed out))
04:01:52*tvelocity np GU 028 (Nick Warren)
04:05:38 Join QT_ [0] (as@area51.users.madwifi)
04:16:18 Quit QT (Read error: 110 (Connection timed out))
04:17:05 Quit hardeep ("[BX] Its not TV. Its BitchX.")
04:17:12 Join Stryke` [0] (
04:22:49 Join ashridah [0] (
05:01:15 Quit Stryke` (Read error: 110 (Connection timed out))
05:16:19 Quit elinenbe_ (" HydraIRC -> <- Go on, try it!")
05:34:48 Join kenshin [0] (
05:41:44 Join wacky_ [0] (
05:41:51wacky_does the talking menus work with iRivers ?
05:42:03wacky_I mean 'do'
05:44:09***Saving seen data "./dancer.seen"
05:49:38kenshini don't believe the voice menus are working with the h1xx players, yet
05:51:08wacky_k thanks
06:11:02 Quit Febs (
06:11:02 Quit amiconn (
06:11:02 Quit sanchny (
06:11:02 Quit t0mas (
06:11:02 Quit Harpy (
06:11:02 Quit Thasp (
06:11:02 Quit thegeek (
06:11:02 Quit crwl (
06:11:02 Quit RotAtoR (
06:11:02 Quit wacky_ (
06:11:02 Quit MO-Pantsu (
06:11:02 Quit hicks (
06:11:02 Quit edx (
06:11:02 Quit Seed (
06:11:02 Quit dwihno (
06:11:02 Quit crashd (
06:11:02 Quit gromit` (
06:11:02 Quit odd (
06:11:02 Quit ashridah (
06:11:02 Quit TCK (
06:11:02 Quit Coldtoast (
06:11:02 Quit xen` (
06:11:02 Quit silencer_ (
06:11:02 Quit Bagder (
06:11:02 Quit dionoea (
06:11:02 Quit Rick (
06:11:02 Quit mbr (
06:11:02 Quit Hadaka (
06:11:02 Quit tvelocity (
06:11:02 Quit Bger (
06:11:02 Quit bipak (
06:11:02 Quit CoCoLUS (
06:11:02 Quit elinenbe (
06:11:02 Quit pabs (
06:11:02 Quit Plugh_ (
06:11:02 Quit bill20r3 (
06:11:02 Quit ze (
06:11:02 Quit pike2 (
06:11:02 Quit Lynx_awy (
06:11:02 Quit HCl (
06:11:02 Quit crash_ (
06:11:02 Quit C-Keen (
06:13:23 Join hardeep [0] (
06:13:23NJoinwacky_ [0] (
06:13:23NJoinashridah [0] (
06:13:23NJoinFebs [0] (
06:13:23NJoinamiconn [0] (
06:13:23NJoinTCK [0] (
06:13:23NJoinRotAtoR [0] (
06:13:23NJoinColdtoast [0] (
06:13:23NJoinMO-Pantsu [0] (
06:13:23NJoinsanchny [0] (
06:13:23NJoinhicks [0] (~hicks@90ce51b15f4503ba.session.tor)
06:13:23NJointvelocity [0] (
06:13:23NJoinBger [0] (~Bager@
06:13:23NJoinbipak [0] (
06:13:23NJoinedx [0] (
06:13:23NJoint0mas [0] (
06:13:23NJoinHarpy [0] (
06:13:23NJoinThasp [0] (
06:13:23NJointhegeek [0] (
06:13:23NJoinpike2 [0] (
06:13:23NJoinSeed [0] (
06:13:23NJoindwihno [0] (~dw@
06:13:23NJoinxen` [0] (
06:13:23NJoinCoCoLUS [0] (
06:13:23NJoinelinenbe [0] (~elinenbe_@
06:13:23NJoincrashd [0] (
06:13:23NJoingromit` [0] (~gromit`
06:13:23NJoincrwl [0] (
06:13:23NJoinpabs [0] (
06:13:23NJoinLynx_awy [0] (
06:13:23NJoinodd [0] (
06:13:23NJoinHCl [0] (
06:13:23NJoindionoea [0] (
06:13:23NJoinsilencer_ [0] (
06:13:23NJoinBagder [0] (
06:13:23 Join Rick [0] (rick@Rick.user)
06:13:23NJoinbill20r3 [0] (
06:13:23NJoinC-Keen [0] (
06:13:23NJoinHadaka [0] (
06:13:23NJoinPlugh_ [0] (
06:13:23NJoinmbr [0] (
06:13:23NJoincrash_ [0] (
06:13:23NJoinze [0] (
06:14:18CtcpVersion from freenode-connect!freenode@connect.utility.freenode
06:14:18***Server message 477: 'logbot #rockbox :[freenode-info] please register your nickname...don't forget to auto-identify!'
06:21:38 Quit Thasp (
06:21:38 Quit sanchny (
06:21:38 Quit amiconn (
06:21:38 Quit Febs (
06:21:38 Quit hardeep (
06:21:38 Quit Harpy (
06:21:38 Quit t0mas (
06:21:38 Quit thegeek (
06:21:38 Quit crwl (
06:21:38 Quit odd (
06:21:38 Quit dwihno (
06:21:38 Quit hicks (
06:21:39 Quit MO-Pantsu (
06:21:39 Quit RotAtoR (
06:21:39 Quit Seed (
06:21:39 Quit wacky_ (
06:21:39 Quit gromit` (
06:21:39 Quit crashd (
06:21:39 Quit edx (
06:21:39 Quit mbr (
06:21:39 Quit Bagder (
06:21:39 Quit silencer_ (
06:21:39 Quit xen` (
06:21:39 Quit Coldtoast (
06:21:39 Quit TCK (
06:21:39 Quit dionoea (
06:21:39 Quit ashridah (
06:21:39 Quit Hadaka (
06:21:39 Quit Rick (
06:21:39 Quit Plugh_ (
06:21:39 Quit bill20r3 (
06:21:39 Quit elinenbe (
06:21:39 Quit CoCoLUS (
06:21:39 Quit Bger (
06:21:39 Quit tvelocity (
06:21:39 Quit pabs (
06:21:39 Quit bipak (
06:21:39 Quit ze (
06:21:39 Quit crash_ (
06:21:39 Quit HCl (
06:21:39 Quit pike2 (
06:21:39 Quit C-Keen (
06:21:39 Quit Lynx_awy (
06:23:23NJoinhardeep [0] (
06:23:23NJoinwacky_ [0] (
06:23:23NJoinashridah [0] (
06:23:23NJoinFebs [0] (
06:23:23NJoinamiconn [0] (
06:23:23NJoinTCK [0] (
06:23:23NJoinRotAtoR [0] (
06:23:23NJoinColdtoast [0] (
06:23:23NJoinMO-Pantsu [0] (
06:23:23NJoinsanchny [0] (
06:23:23NJoinBger [0] (~Bager@
06:23:23NJoinbipak [0] (
06:23:23NJoinedx [0] (
06:23:23NJoint0mas [0] (
06:23:23NJoinHarpy [0] (
06:23:23NJoinThasp [0] (
06:23:23NJointhegeek [0] (
06:23:23NJoinpike2 [0] (
06:23:23NJoinSeed [0] (
06:23:23NJoindwihno [0] (~dw@
06:23:23NJoinxen` [0] (
06:23:23NJoinCoCoLUS [0] (
06:23:23NJoinelinenbe [0] (~elinenbe_@
06:23:23NJoincrashd [0] (
06:23:23NJoingromit` [0] (~gromit`
06:23:23NJoincrwl [0] (
06:23:23NJoinpabs [0] (
06:23:23NJoinLynx_awy [0] (
06:23:23NJoinodd [0] (
06:23:23NJoinze [0] (
06:23:23NJoincrash_ [0] (
06:23:23NJoinmbr [0] (
06:23:23NJoinPlugh_ [0] (
06:23:23NJoinHadaka [0] (
06:23:23NJoinC-Keen [0] (
06:23:23NJoinbill20r3 [0] (
06:23:23NJoinRick [0] (rick@Rick.user)
06:23:23NJoinBagder [0] (
06:23:23NJoinsilencer_ [0] (
06:23:23NJoindionoea [0] (
06:23:23NJoinHCl [0] (
06:24:14 Join Stryke` [0] (
06:24:16CtcpVersion from freenode-connect!freenode@connect.utility.freenode
06:24:16***Server message 477: 'logbot #rockbox :[freenode-info] please register your nickname...don't forget to auto-identify!'
06:27:24 Quit kenshin (Read error: 113 (No route to host))
06:28:59 Quit wacky_ ("ahuh thanks")
06:35:55 Join kenshin [0] (
06:43:36kenshini'm lost in all these threads. i'll never get this simulator working without someone's help.
06:45:23 Quit RotAtoR ()
06:51:14 Join elinenbe_ [0] (
06:51:37elinenbe_hardeep: nice work on the new playlisting options... you should add them to the wiki!
06:52:22hardeepi figure the less the document the less chances someone reports bugs
06:56:40 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
06:56:55 Quit DMJC (Read error: 104 (Connection reset by peer))
06:57:42 Join DMJC-L [0] (
06:59:38 Join DMJC [0] (
07:19:25hardeepinteresting, my changes caused a compiler error on the gmini 120 build... woo
07:19:54hardeepinternal compiler error that is
07:27:58 Quit kenshin (Read error: 60 (Operation timed out))
07:44:10***Saving seen data "./dancer.seen"
07:45:30 Quit hardeep ("[BX] Bob Barker uses BitchX. Have your BitchX spayed or neutered.")
08:31:11 Quit DMJC-L ("Leaving")
08:45:25 Join kenshin [0] (
08:47:12 Quit t0mas (Read error: 110 (Connection timed out))
08:53:33 Join tvelocity [0] (
09:01:56 Quit tvelocity ("Leaving")
09:27:30Plugh_anyone here good with UNIX command line stuff? I'm looking for a way to capture internet radio feeds on a schedule
09:27:58Plugh_Some of my favorite shows come on when I'm at work or asleep, and all the stations I listen to have web feeds
09:28:31Plugh_I'd like to cron the downloads. I can start them, but I can't end them on a schedule
09:30:04SlasheriJust gracefully kill the parent process you have started
09:30:30Plugh_how can I do that within a cron?
09:30:58Slasheriyou have to schedule different entries for "start task" and "stop task"
09:30:58Plugh_by hand, I'd just ^C or pkill processname
09:31:08SlasheriOr maybe you could use at for that
09:31:28Plugh_the problem is passing the correct pid
09:31:47Slasheriah, hmm
09:32:06Slasheriyou have many processes with the same program name?
09:32:25Plugh_if two shows overlapped, yes
09:33:24Slasheriyes, that might require some scripting.. take a look at man at if that command could end the processes (i am not sure)
09:33:42Plugh_at would be clumsy
09:35:28Slasheriin fact i have an example C program that i wrote a long time ago. It kills a processes after selected timeout
09:39:05Coldtoastcan't you set each job up in order?
09:39:16Coldtoastthat way you know what belongs to what?
09:40:10SlasheriPlugh_:, it's just an "example code" so it's full of crap you might want to remove :) But it should work
09:40:27amiconnSlasheri: Did you have a look at hardeep's commit? It seems you will need to add something to playback.c to make it work on iriver
09:40:50Slasheriamiconn: Hmm, checking
09:41:31SlasheriHmm, the latest commit?
09:41:52amiconnThe 04:46 GMT
09:42:47Slasheriit doesn't work on iriver?
09:43:01SlasheriDoes it call audio_flush_and_reload_tracks?
09:43:16amiconnHe did something in mpeg.c, but mpeg.c is archos only
09:43:30 Quit kenshin (Read error: 110 (Connection timed out))
09:44:12***Saving seen data "./dancer.seen"
09:45:49SlasheriHmm.. We should find the problem, i can't say yet what might be wrong
09:46:35 Join DangerousDan [0] (~Miranda@
10:03:44 Quit ashridah ("Leaving")
10:05:42 Join t0mas [0] (
10:10:01SlasheriHmm, does somebody have any 22 kHz mp3/ogg file?
10:10:45t0masI have one
10:10:50Slasherigreat, could you send one? :)
10:10:58Slasherigood :)
10:12:18t0masmight take some time... bittorrent is uploading too :)
10:12:31Slasherihehe, no problem. almost here :)
10:13:29t0masdone :)
10:14:06t0mas(fyi, it's a dutch cabaret song :))
10:14:09Slasherii have almost the dsp stuff working, now testing resampling
10:14:35t0mashave you cvs updated lately?
10:14:46Slasherinope.. i might update today
10:14:47t0maspreglow made some upsampling work...
10:14:53t0masso it already plays this
10:14:58Slasheriyes, i am moving that to dsp
10:15:14Slasherimp3 files has no a better performance too
10:16:34amiconnSlasheri: Did you disable libmad dithering?
10:16:51Slasherii have moved it to dsp, it's configurable
10:17:07Slasherici->configure(DSP_DITHER, (bool *)true);
10:17:28amiconnOkay, nice
10:17:36amiconnGtg, cu
10:17:52 Part amiconn
10:25:56 Join Aison [0] (
10:35:43 Join jwalk [0] (~jwalk_edm@
10:37:08 Quit jwalk (Client Quit)
10:56:45 Join kenshin [0] (
11:22:21HCloh. bittorrent.
11:22:35HCl12 hours remaining :/
11:23:08 Join preglow [0] (
11:24:47preglowSlasheri: i think dithering should be a setting as well, so one can determine if the codecs that are asking for it should get it
11:24:57preglowi for one can't hear much of a difference
11:25:10preglowmy cold has escalated over the night
11:25:13preglowmy life is pure joy
11:25:30Coldtoastewwww. I hate that
11:25:59*HCl prods insomnia
11:26:07preglowwith good reason
11:26:36ColdtoastI'll be fine at work then wake up next day and it'll have taken hold
11:27:23Coldtoastdoesn't help that I have at least a couple of hundred people breathing on me over the course of the night
11:27:31preglowi'd guess not
11:27:40ColdtoastI work in nightclubs
11:27:44Coldtoastyou can imagine....
11:27:51preglowhahah, that's destined to help lots, yes
11:28:46Coldtoastbecause I know how flithy people are, I no longer touch anything in a public bathroom with my bare hands. Always get paper towel
11:28:51Slasheripreglow: Hmm, good idea
11:29:08Slasheripreglow: btw, can i use your resampler to 16 bit samples?
11:30:45Coldtoasthere's a question for you actually; in public toilets, who's messier? Men or women?
11:31:18t0masdrunken man for sure :P
11:31:29preglowslasheri: dithering should be the last to happen
11:31:32Coldtoastmen do more disgusting things but women are messier overall
11:32:05ColdtoastI swear women get toilet paper and have streamer parties
11:32:35preglowslasheri: we will very probably use more advanced resampling in the future, and if we do that directly on 16 bit data, we'll lose a lot of precision
11:33:06Slasheripreglow: hmm, ok. So 16 bit data should be scaled to 24/32 bits first?
11:34:14preglowyeah, it really should, i guess
11:34:19Slasherigood :)
11:34:21preglowwe might want to decide on some internal format
11:34:26preglowlike perhaps 24 bits
11:34:51Coldtoastactually, the most foul thing anybody's done at work would be getting REALLY drunk, going to the toilet, not makin git and taking a dump in your underpants with them halfway down then vomitting into them and leaving them on teh floor. Had a male AND a female do that one so maybe "most disgusting" is equal
11:35:54preglowmy god, you have won then
11:36:06preglowif you achieve that, you're humankinds finest
11:36:15Coldtoasthaha. definitely
11:36:43Coldtoastlast niht we had a 130kg fat guy refuse to walk so we had to drag him down the rap and out onto the footpath
11:36:51Coldtoastlocal bike gang member
11:37:29t0masColdtoast: why not just call the police?
11:37:32t0masand let them drag him out?
11:37:48Coldtoastnah. not major enough
11:38:07*t0mas can imagine how hard your job sucks sometimes...
11:38:09Coldtoastth eonly time we call the police is with big fights
11:38:33Coldtoastyeah. absolutely. 95% of the time it's sweet
11:38:44Coldtoastthe 5% tho... that's where we earn our money
11:38:49t0masyou're a bar guy? or security?
11:38:52 Quit kenshin (Read error: 110 (Connection timed out))
11:39:00t0masah... that's less sweet
11:39:06t0masdancing girls on the bar is ok ;)
11:39:18Coldtoastyeah. we're the least respected, most disliked
11:39:29*t0mas was in "the best club in the Netherlands" last week...
11:39:57t0masthey inveted something new for the heat...
11:39:59t0mashaha.. showers
11:40:24t0maschanged the whole place... beach party
11:40:39Coldtoastthis is REALLY bad to say, i know
11:40:50Coldtoastbut guess who the smelliest people we get are
11:41:07Coldtoastnothing to do with being racist. it's an observation
11:41:18t0masand added some showers and bay watch people to it
11:41:53t0masok... I don't know how it's in your club... but here they just refuse to let any black guy in...
11:42:07t0masand nobody does anything about it...
11:42:09Coldtoastit's weird. Africans are among the best dressed people who come in but they're also the smelliest
11:42:44Coldtoastman. we'd REALLY be in the shit with the law if we did that t0mas
11:43:06t0masyes, it's officially forbidden here
11:43:10t0masbut hey... they just do it
11:43:16Coldtoastheh. ok
11:43:35Coldtoastwe never EVER have African women in. ever
11:43:37Coldtoastit's always males
11:44:14***Saving seen data "./dancer.seen"
11:44:18t0mashm... well... you need a member card in the club here
11:44:40t0masso they just don't give them to black people... I guess that's the trick
11:48:32Coldtoastwill there be a VBRFix plug for iriver?
11:49:25Coldtoastfinding a few tracks on my h140 that need fixing , which I can do fine here, but if I'm out somewhere it'd be nice to be able to do it with the h140
11:56:55 Join webguest88 [0] (
11:58:09Slasheriargh, now i am rewriting the dsp.. this is really hard :)
12:05:41 Join RedBreva [0] (
12:06:10ColdtoastI can imagine :)
12:06:10preglowwhat went wrong?
12:07:15Slasherithe design was totally flawed ;) now i try to make it better scalable and different kind of optimizations to make code cleaner
12:09:42Slasherii found out that it's best to write the main dsp process function first and after that the other stuff ;)
12:11:00 Join cYmen [0] (
12:13:23preglowwhat stereo format do the codecs have to deliver? interleaved or noninterlaved?
12:13:25preglowor both? :)
12:13:32Slasheriboth :)
12:13:47Slasherior they can choose which one to deliver
12:14:15preglownoninterleaved is better for processing speed
12:15:25preglowwe'll probably have to hack tremor to make it deliver data in its internal format
12:15:39preglowi sorely wish i had time to work on this
12:30:53webguest88I thought that Linus had moved the 'playlist options' to first in the menu ?
12:35:39webguest88ah the 'view current playlist' is added to a loong wps play push, nice
12:39:03HCli really need to get to coding
12:39:35*webguest88 gives HCl a push
12:49:14 Join Moos [0] (
12:49:35MoosHello all
12:56:10 Quit RedBreva ()
12:56:20 Join wacky_ [0] (
12:56:36wacky_even with bleeding edge, I have a disc that doesn't play with Rockbox, a bunuch of .ogg files
12:56:50wacky_ and 15.ogg and 16.ogg at the same place
12:57:31wacky_they start playing, and then hang there and come back to the file tree
12:57:48wacky_but they send out no sound
12:58:14 Join edx__ [0] (
13:03:29 Quit edx (Read error: 60 (Operation timed out))
13:03:39Coldtoastnot very good quality .oggs tho. heh
13:03:54Coldtoastthere's ticking and stuff thru the one I just listened to
13:04:06Coldtoastis it from vynil?
13:05:21 Join Nuxator [0] (
13:05:52thegeekthat's the obvious answer
13:06:00Nuxatorjust a question about the upsampling in mp3:
13:06:06thegeekwrong window again
13:06:31Nuxatorit plays the file fine but time elapsed is wrong so the track is cut before its end
13:06:35preglowyes, we know
13:07:01Nuxatorbut it sounds ok
13:07:23Nuxatorok thanks for the answer preglow
13:07:28Coldtoastlooking at he detaiils for one of the .oggs wacky_ and looks like they were encoded with a REALLY old version of libVorbis
13:07:53preglowNuxator: resampling and stuff is bein reimplemented as we speak
13:08:05Nuxatorwow great work
13:08:11 Join kenshin [0] (
13:08:44preglowhow old version?
13:10:44preglowohh, nice soundtrack
13:13:52Nuxatorheve to go cu
13:14:01 Quit Nuxator ("Chatzilla [Firefox 1.0.4/20050511]")
13:15:25 Nick edx__ is now known as edx (
13:16:47preglowtremor should handle even older files
13:18:02Slasheripreglow: Hmm, can i assume that MAD_F_FRACBITS is the output sample depth from libmad?
13:21:12preglowi think you can
13:21:15Slasherigood :)
13:21:35preglowif it's greater than /-1 in their own fixed point format, it should be clipped
13:22:53preglowog great, plus/minus
13:30:35 Join ep0ch [0] (~ep0ch@
13:31:53ep0chi have an obeservation, not sure if it's known or not...
13:32:11ep0chmono mp3 playback = good, mono Vorbis playback = 2x playback :)
13:32:47preglowhmm, i didn't know that
13:33:00preglowi'll let slasheri fix it, so i don't conflict his codecs
13:33:15Slasheriyes, that will be fixed by dsp
13:33:21ep0chthought so
13:33:43ep0chwill the dsp be responsible for replaygain as well?
13:33:52ep0chcan't wait :)
13:34:02preglowdon't expect replaygain to be implemented yet, btw
13:34:09preglowbut that's where it'll be done
13:34:14Slasheriep0ch: yes, that's dsp's job :) it can be implemented later
13:35:20*t0mas yawns... "Emerging package 8 of 126"
13:44:17***Saving seen data "./dancer.seen"
13:48:10 Join Chamois [0] (
13:48:25thegeek(ccache + distcc*3) = powah.
13:55:55 Join bipak_ [0] (
14:06:52 Join yyz [0] (
14:06:52 Quit Bger (Read error: 104 (Connection reset by peer))
14:07:30 Join Bger [0] (~Bager@
14:09:42 Quit DangerousDan (Read error: 104 (Connection reset by peer))
14:11:30 Quit bipak (Read error: 110 (Connection timed out))
14:25:19 Quit yyz (Remote closed the connection)
14:55:16 Join Jleagle [0] (
14:57:24 Join shank [0] (
15:00:01 Quit Jleagle (Client Quit)
15:00:15shankcan anyone tell me the status of the h120 port?
15:00:23shankor point me to some docs about it?
15:12:42 Quit kenshin (Read error: 110 (Connection timed out))
15:15:32shankpreglow, thanks
15:18:59webguest88Thank you for re-arranging the menu's and making the navigating more user friendly
15:19:07 Join |ep0ch [0] (~ep0ch@
15:22:19 Part webguest88
15:32:22*HCl finally got the military harrier jet fighter :)
15:39:11 Quit ep0ch (Read error: 110 (Connection timed out))
15:39:11 Nick |ep0ch is now known as ep0ch (~ep0ch@
15:40:39SlasheriHmm, < 60% boost ratio with 128 kbit/s mp3 file with the new dsp system without resampling
15:41:48 Join Teslacuted [0] (
15:42:37crwlis the current version of rockbox more usable than version from about three days ago? :)
15:43:10Slasherii doubt that, dsp stuff is still not committed :)
15:43:28Slasheribut it might be, improved menu system at least
15:43:43crwlwell, mainly, are there any new problems? :)
15:44:16Slasherii don't know any new issues ;)
15:44:18***Saving seen data "./dancer.seen"
15:44:48crwlheh, i'll just try it then
15:45:00MoosSlasheri:the new dsp stuff coming soon?
15:45:01crwlwelll, it didn't compile
15:45:17SlasheriMoos: yes, maybe tomorrow
15:45:24Slasherinow it works much better
15:45:32Moosw00t very cool
15:47:18crwlnothing, compiled after a fresh checkout
15:52:26 Join Yokalosh [0] (
15:52:52MO-Pantsuto me Rockbox is not far from what I want now. All I am really interested in are the remote LCD, accurate FF/RW in MP3 and FF/RW in Ogg and eventually some form of gapless hack for MP3's that were not ripped with Lame to be gapless
15:52:56YokaloshDoes anyone know what to do about an I/O error when trying to view files on a jbr10 on a pc?
15:53:24MO-Pantsuoh and recording of course with meters
16:05:56 Join kenshin [0] (
16:07:34Yokaloshis it that nobody knows anything about this problem or you can't be bothered?
16:07:46HClwhat problem?
16:08:41kenshinis there any way to see the WPS in the simulator?
16:08:57Yokaloshwhen i try to browse the files on my jukebox via my pc i get a hardware I/O error
16:09:06HClbad harddisk.
16:09:14HCltry scandisk
16:09:30Yokaloshi'll give it a go if it lets me
16:09:32kenshineverytime i use "play" to see it, the sim crashes somewhere after yield_codec in audio_load_track.
16:10:12kenshinand i'm so lost in where the threads are at when it dies.
16:11:22preglowslasheri: so, how does it work currently? codecs send it audio, it then optionally resamples and dithers, then sends to pcm buffering?
16:11:39preglowslasheri: you are aware the dithering needs to be aware of the resolution of the data it gets, btw?
16:16:50Yokaloshhey HCI as soon as i try to right click on my jukebox in my computer it just crashes explorer.exe
16:17:27HCltry linux?
16:17:46Yokaloshi don't have a computer with linux on it
16:17:48HCli dunno, i don't have an archos
16:19:11Yokaloshall i get is 'G:\ is not accessible. The request could not be performed because of an I/O device error.'
16:20:32preglowsounds nice
16:20:56Yokalosh:( got any idead preglow?
16:21:18preglowcan't say anything based on just that information
16:21:33Yokaloshthat is all it gives me :(
16:21:50preglowdoes the unit work fine?
16:22:06Yokaloshyes by itself
16:22:25Yokaloshjust not once plugged into any of the computers in my house, all of which it worked fine with up untill a few weeks ago
16:22:25kenshinYokalosh: can you run scandisk on it in a console?
16:22:41Yokalosherm i can try
16:23:10kenshindisclaimer: i run linux so i can't tell you how to do it in Windows
16:23:27ep0chhow is it connected USB?
16:23:51ep0chtried another USB port?
16:24:03Yokaloshi have tried it on about 5 different pcs
16:24:41Yokaloshi could try linux i guess, i have the rest of the weekend before school and a couple of 4 gig hard drives lying about
16:25:12ep0chor you could try a version of linux that runs off CD...
16:25:21Yokaloshahhh yes
16:25:27Yokaloshi know nothing about linux yet
16:25:31ep0chdunno, i use FreeBSD
16:25:44crwlknoppix is probably the most well known one
16:27:06Yokaloshokay, so what am i doing?
16:27:40ep0chdownload knoppix, burn it and boot it
16:27:42kenshindownload knoppix, burn it, boot it
16:28:04Yokaloshokay lol
16:28:42ep0chand learn how to mount stuff in linux
16:29:41kenshinYokalosh: when you plug in the USB cable, look at /var/log/messages to see what device name linux wants to use
16:31:09Yokaloshi need to find a cd to burn it onto now....
16:31:18 Quit Teslacuted (Read error: 113 (No route to host))
16:31:53Slasheripreglow: Hmm, yes. First codec configures and enables dsp. Then codec calls audiobuffer_insert and if dsp is enabled, dsp catches that call. Then stereo mode is converted to noninterleaved if it's not already, sample will be scaled up if necessary (not working yet). After that dsp chain is applied (resampling, bias, dithering, scaling) and data copied to pcm buffer.
16:32:50ep0ch"<Slasheri> Hmm, < 60% boost ratio with 128 kbit/s mp3 file with the new dsp system without resampling"
16:32:53ep0chthat good?
16:33:14Slasheriep0ch: i think it's little better than currently
16:33:58ep0chwierd, because vorbis is about 20% at a similar bitrate
16:34:31ep0choh i have a question regarding cpu boost...
16:35:20ep0chwhat's more effecient (power wise), have a lower speed CPU boost for a longer time, or higher speed for a shorter time?
16:36:12preglowwe need to optimise libmad further :/
16:38:38preglow1L << (MAD_F_FRACBITS 1 - SAMPLE_DEPTH - 1)
16:38:46preglowcan anyone explain the plus 1 and the minus 1 ?
16:39:17preglowseems kind of very pointless
16:40:17 Quit shank (Read error: 110 (Connection timed out))
16:40:43Slasheripreglow: yes, i think there is no need for that (i have removed the 1 - 1 from code)
16:41:32preglowi also wonder why the bias is there
16:41:36preglowsome kind of rounding?
16:42:52Slasherii really don't exactly know what it does :) i think it's needed to convert the mad's fixed point format to some other form:
16:42:56Slasheri| * Fixed-point format: 0xABBBBBBB
16:42:58Slasheri| * A == whole part (sign + 3 bits)
16:43:01Slasheri| * B == fractional part (28 bits)
16:43:21preglowi think it does highpass triangular dithering with second order noise shaping
16:43:36preglowbut i'm just wondering about the bis
16:43:44preglowi think it rounds the lowest bit
16:43:45Slasheriah, interesting. so in fact it's dithering and not bias at all?
16:44:17preglowthe entire thing does dithering and quantisation, yes
16:44:27preglowi just don't get the bias line
16:45:00preglowbut i think it's just for rounding the lowest bit for even more accuracy
16:45:05 Join yngwi [0] (
16:45:32Slasherihmm, ok. then i should disable that "bias" also when disabling dithering
16:45:54preglowwell, depends, the rounding is probably separate from the dither
16:46:11preglowhmm, no, it's not highpass triangular in its current form, it's rectangular dithering
16:46:27preglowthe dither->random assignment is commented out
16:47:23Slasherimaybe i will leave the dithering in dsp as it's now and let somebody fix it :)
16:48:21preglowas it currently is, it doesn't dither at all
16:48:24preglowit just adds a constant bias
16:48:36preglowi wonder why the hell that line has been commented out
16:48:39preglowslasheri: yeah, do that
16:48:40*HCl bites his game.
16:48:49preglowhcl: you're supposed to be biting code
16:49:00HClnot till i finished this part of my game
16:52:56 Quit xen` ()
16:53:41preglowi don't get the bias line
16:54:06 Quit Yokalosh ()
16:56:54preglowi suspect it might be to remove dc offset
16:56:59preglowin which case, slasheri, it should be left out, yes
16:57:03preglowif not dithering
16:57:17preglowi don't think there's anything in that routine that should be running if dithering isn't enabled
16:57:20Slasheripreglow: ok, i can try without it :)
16:57:33preglowi've been running without with no ill effects, at least
16:57:39Slasheriah, good
17:00:46preglowclipping, though, you might want to add, now that i think about it
17:01:05preglowjust try
17:01:09preglowyou'll see ;)
17:03:37 Join webguest24 [0] (
17:04:34Slashericlipping is already implemented :)
17:05:04preglowmost transforms codecs can't guarantee in any way that the outputs stay within the bounds
17:05:20preglowif you look at most mp3 and vorbis files, you'll see that clipping isn't rare at all
17:06:14Slasheri| ci->configure(DSP_SET_CLIP_MIN, (int *)-MAD_F_ONE);
17:06:15Slasheri| ci->configure(DSP_SET_CLIP_MAX, (int *)(MAD_F_ONE - 1));
17:06:24Slasheriit's configured that way by codecs
17:06:53preglowoh, lovely
17:07:01preglowthat's grand
17:07:03Slasheriother options are:
17:07:04Slasheri| ci->configure(DSP_SET_FRACBITS, (int *)(MAD_F_FRACBITS));
17:07:04Slasheri| ci->configure(DSP_DITHER, (bool *)true);
17:07:04DBUGEnqueued KICK Slasheri
17:07:04Slasheri| ci->configure(DSP_SET_STEREO_MODE, (int *)STEREO_NONINTERLEAVED);
17:07:20preglowand it wont clip if those aren't set, i assume?
17:07:27Slasheriah, and DSP_SET_FREQUENCY of course
17:07:36Slasheriit wont
17:09:35elinenbe_Slasheri: what's the new DAP code that is about to be committed?
17:10:16Slasherielinenbe_: it enables some common dsp functions for all codecs such as resampling
17:11:00Slasherii think he meant DSP =)
17:11:03preglowyou mean DSP ? ;)
17:11:23elinenbe_yeah −− sorry.
17:11:32elinenbe_i was just reading!
17:11:51elinenbe_preglow: does that mean your resampling code is no longer necessary>
17:12:14Slasherino it's used, that code is only moved elsewhere
17:12:34Slasheriso that all codecs can use it
17:13:05elinenbe_so then resampling will work with everything...
17:13:33preglowthe only reason i put it in mpa.c, was for someone else to move it elsewhere
17:14:25ep0chand there's no resampling if the if the input is 44.1Khz this time?
17:14:42elinenbe_so are you doing a set of dsp functions? what about replaygain support?
17:14:51Slasheriep0ch: no
17:15:21Slasherireplaygain can be implemented later. It should be quite trivial after that
17:16:55preglowoh yes
17:17:01preglowa single instruction extra in the dsp path
17:19:23webguest24wait, this dsp.... will it have nifty effects like reverb and chorus?
17:20:24preglowperhaps in the future
17:20:33preglowi'd like to implement something like that
17:20:35Slasherithe first implementation of dsp will not have any effects but it's possible to add them in future
17:20:54preglowbut don't expect too much, we're going to be banging against the mhz wall pretty often
17:22:43webguest24wow, that would be really cool
17:24:46preglowreally? :P
17:24:54preglowhow often do you apply chorus to the music you listen to?
17:25:07preglowi really like delay effects, but i still don't use them when i'm listening to music
17:26:20webguest24reverb would definitely be the coolest thing... hehe... really make those ipod users jealous
17:29:05preglowbut yeah
17:29:13preglowi will implement stuff like that if there's enough cpu left
17:29:20preglowi will probably implement them anyway, for real time use
17:29:33preglowso the h120 can be used as en effect box
17:35:05ep0chpreglow: you ever use foobar2000?
17:35:11preglowyes, all the time
17:35:14preglowwhen i'm in windows
17:35:19preglowit's my preferred player
17:35:28ep0chhave you seen the noise sharpening dsp?
17:35:41preglownoise _sharpening_?
17:35:46preglowwhy would you want to sharpen noise?
17:35:58ep0chit's actually quite good
17:36:16pregloware you sure you don't mean noise shaping?
17:36:21SeedDSPs usually shape noise, not sharpen it :)
17:36:44ep0chit adds the differential of the wave
17:36:48preglowif you mean noise shaping, then i don't see how you can shape noise outside of the quantisation process
17:36:55ep0chi'll find the thread...
17:36:57preglowdo so
17:38:25preglowthat's dirt simple
17:38:39preglowdifferential = simple subtraction, integration = 1 pole iir
17:38:53ep0chand a multiple
17:39:21preglowbut this does pretty much what an eq does, really, just much more extreme
17:39:37preglowbut i gotta go make dinner
17:39:45ep0chworth bearing in mind :)
17:44:20***Saving seen data "./dancer.seen"
17:52:39 Join Sucka [0] (
17:54:56 Join Road_Runner [0] (
17:56:17Road_RunnerQuestion: I have a thought about how to implement right-to left laguage... where do I post it in order of people to read and review it?
18:00:12Road_Runnerand yet another question... this irc channel is said to be nice and friendly... is it?
18:05:46Road_Runnerthank you.
18:11:35preglowyou could post it on the mailing list, i guess
18:11:54preglowbut wasn't someone already working on right to left languages?
18:12:47 Join hardeep [0] (
18:17:09hardeepSlasheri: was just scanning the logs, you don't have to do anything for the change I submitted, it works correctly on the iRiver
18:17:28Slasherihardeep: ah, good :) i haven't tested it yet
18:20:40 Join Jleagle [0] (
18:21:58 Quit webguest24 ("CGI:IRC (Ping timeout)")
18:26:17 Quit Jleagle ("Chatzilla [Firefox 1.0.4/20050511]")
18:31:20 Join Bgr [0] (~Bager@
18:31:20 Quit Bger (Read error: 104 (Connection reset by peer))
18:32:24 Quit Chamois (" HydraIRC -> <- The professional IRC Client")
18:47:52 Quit preglow ("CGI:IRC")
19:01:41 Quit Road_Runner ("CGI:IRC (EOF)")
19:12:15 Quit elinenbe_ (" HydraIRC -> <-")
19:14:12 Quit ep0ch (" HydraIRC -> <- Leading Edge IRC")
19:18:24 Quit Sucka ("Leaving")
19:21:18 Join Stryke` [0] (
19:22:11 Join Sucka [0] (
19:26:28t0masok... why the hell is this in the bash history of a root account: cat /boot/kernel-2.6.11 > /dev/dsp
19:28:27 Join Bruno [0] (
19:28:48Bruno"cat /boot/kernel-2.6.11 > /dev/dsp" Sound of kernel??? Should'nt be nice to listen
19:30:45Brunoshould be on
19:32:02 Quit Bruno (Client Quit)
19:37:44 Join muesli- [0] (
19:38:32CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
19:38:32*t0mas emails everybody with the rootpassword :)
19:38:52t0masit must be one of the 4 admins... and it's not me...
19:39:07t0masso one of the 3 others was listening to the kernel sound... haha...
19:43:06muesli-the other 2 are watching soccer :D
19:44:03HCl9 hours remaining...
19:44:22HCli wonder how it manages that, since this morning at 11 am it was 12 hours remaining
19:44:23***Saving seen data "./dancer.seen"
19:49:56 Join darthmall [0] (
19:52:02t0maswhat? 9 hours?
19:52:14HClfor my download, yea..
19:52:25HClat least the progress changed from 50% to 69%
19:52:28t0masI tought rockbox battery life.
19:52:44t0masthat's a pretty good estimate...
19:52:51t0masit was at 3:45 this morning
19:52:59t0masand 3:53 later it stopped
19:53:06muesli-hcl, could you please offer a new graypatched rbx on your website?
19:54:35t0maswe really should make a cygwin/windows wizzard
19:54:43t0masso everybody can compile rockbox :)
19:54:56HClmuesli-: mmm, in a while..
19:55:35muesli-krauts were just fried by brazil
19:57:24 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
19:57:29 Join Stryke` [0] (
20:02:04t0masmuesli-: and the young dutch guys lost after 12 penaltys...
20:02:17t0masstupid nigerians...
20:02:35muesli-we are loser today :D
20:03:11t0maswhat was the end for krauts?
20:03:22t0maspenaltys too? or normal losing? :)
20:04:18Moosviva Braziul :D
20:04:20kenshinyay brazil!
20:04:42muesli-normal losing by 2:3
20:04:47Mooscurently the best team in the world
20:05:14Moosdeutsch team is good too
20:05:33Moosbut not enough ;-)
20:05:38kenshini wish we got more futbol here in the US
20:05:57muesli-futbol *g*
20:06:01muesli-nice spelling
20:06:15kenshinwould you rather i said "soccer"?
20:06:20Moosin US not is a girl sport?
20:06:40muesli-soccer sounds not that better..i expected football
20:06:55muesli-that would be the direct translation
20:07:12t0masit's soccer
20:07:15t0masfootball is something else
20:07:18kenshini use futbol to distinguish it from american football
20:07:31t0masyes, football is american football...
20:07:38t0masthe game we play here is called soccer in english
20:07:43t0masdon't know how they got that name
20:08:05 Nick Sucka is now known as Sucka`away (
20:08:06Moosfutbal is not the portuges prononciation?
20:08:30muesli-no hablar portuges
20:08:37Moosme too
20:08:47Moosbut in Brazil yes ;)
20:09:35kenshindunno but at least the other americans around me (all 280 million of them) know i'm not talking about american football
20:10:11kenshincause "soccer" is like "miles" or "inches". only the two or three holdouts use them... :/
20:10:42t0masthat's possible... as we learn US english in schools here... not american
20:10:50Moosyou are rare ;)
20:11:34t0masUS english -> GB English
20:11:58t0mas<t0mas> that's possible... as we learn US english in schools here... not american <−− there
20:12:07t0masUS English == american...
20:12:10t0masI ment GB English
20:12:46kenshini think i'm just more in touch with the rest of the world
20:12:51Moosin all schools in the world, it's the GB classical english
20:13:12kenshini don't mind metric. not at all.
20:13:23*t0mas neither...
20:13:53muesli-quarter pounder with cheese :D
20:13:59Mooskenshin: San Antonio Spurs US bascketbal champion :)
20:14:14kenshinroyale with cheese?
20:14:32muesli-yes, 4 sure :d
20:14:44kenshinMoos: don't remind me. my Sonics lost because they beat up our players to knock them out.
20:14:52 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
20:15:09t0maskenshin: your bug is fixed :)
20:15:23t0mas(just recognized the name... haha)
20:15:40kenshint0mas: i noticed. already been using it. :)
20:16:30kenshinfirst noticed the cvs commit notice so had to try.
20:16:41kenshinis anyone working on a Japanese translation, yet?
20:16:51t0masdo you speak japanese?
20:16:52kenshinRomaji or UTF-8?
20:16:57t0mas(and we don't have unicode)
20:17:26kenshinnot well but enough to get me by a translation
20:17:31muesli-Amnicon was working on some gfx..didnt see any..where are they?
20:18:28kenshini'll be going through a full year of Japanese starting in the fall
20:19:25Moosvery good you're lucky
20:19:44kenshinwhich will include both Romaji and Kanji so if there's ever UTF-8 i'll do Kanji translation, too
20:21:29 Join shank [0] (
20:22:29Mooshardeep: have you see the red build generated by your very nice playlist modifs?
20:23:05hardeepMoos: yeah, it appears to be some bug with the compiler
20:24:01Moosyes for gmini have you any idea what should it be?
20:25:18Moosit's never very good to leave one warning ;)
20:26:13Moosi like your modifications :)
20:26:50hardeepMoos: unfortunately, it's a little more difficult since this is a compiler error... it doesn't appear to be any particular line that's causing it, it's the entire case statement
20:27:39Moosi don't know anyone here working in gmini port :(
20:27:53hardeepi think the only one was JyP
20:27:59 Quit muesli- (Read error: 104 (Connection reset by peer))
20:29:52Mooshave you still others ideas for playlists?
20:30:05Moos*great ideas :)
20:31:46hardeepnot currently, everything i've wanted is pretty much in now
20:33:40Moosyes it's true it's already very elaborated, it's fine :)
20:36:04 Quit darthmall ()
21:04:46 Join preglow [0] (
21:09:28 Join JoeBorn [0] (
21:10:22JoeBornhey guys, we're having a discussion on next gen neuros hardware and OS
21:10:38JoeBornand we'd love to get your advice from experience.
21:10:53JoeBornits on #neuros if you're interested.
21:11:21dwihnoMake it the first "out of the box" rockbox player! :)
21:11:27dwihnoThat's my advice ;)
21:11:34JoeBornas you are well aware, we've hit a number of dead ends, adn we're looking to avoid that goign forward.
21:11:42JoeBornwe're considering that, we really are.
21:12:24JoeBornsorry for not giving more notice, I just thought of swinging by while we were having the discussion
21:13:12 Quit kenshin (Read error: 145 (Connection timed out))
21:13:45dwihnoTalk to LinusN, Bagder or Zagor
21:14:26JoeBornwill do, again I appoligize for the short notice, I just didn't think of it sooner.
21:26:22preglowaren't you guys stuck with a texas instrument dsp chip of some sort?
21:26:50JoeBornwe're designing a next gen player, we're not "stuck" with anything.
21:27:10JoeBornthat's actually what we're talking about. swing over and join the discussion f you can.
21:28:37dwihnoSounds like great news.
21:29:05dwihnoTalk to the core team. They are more full with wisdom and ideas than the Fort knox is loaded with gold! :)
21:29:28JoeBornyou guys have always been very nice to us, it's a real shame we haven't been able to get together on a device yet.
21:29:28preglowahh, i'm mistaking you
21:29:34 Join rasher [0] (
21:30:14preglowmight be fun to see what people want hardware-wise
21:30:24rasherjump on in
21:30:30preglowa dsp MIGHT be nice, but it wouldn't exactly make development easier
21:31:08preglowcan't join anything with this web client
21:31:17preglowi should get back to my writing :>
21:31:44preglowoh, why the hell not
21:31:44 Join TCK- [0] (
21:31:46 Part preglow
21:32:12 Join preglow [0] (
21:32:33JoeBornyou can use
21:32:41JoeBornIRC webclient.
21:32:52JoeBornanyway we're winding down. I'll give more notice next time.
21:32:54preglownah, i just booted into irssi
21:34:37dwihnoThere is one downside with chosing the rockbox way. People will stick to their units for aeons of time :) I still use my first-gen recorder since 3 years back
21:34:52 Join webguest44 [0] (
21:35:37preglowthere's a risk i might want to hold on to my h120 for a while, yes
21:35:52JoeBornlonger lifecycles of the product is the other postive edge of that sword.
21:38:32dwihnoOn the other side, using rockbox-compliant hardware, there WILL be software updates
21:39:15dwihnoMost manufacturers DO provide flashing for their units... But what good is a phone call, if you're unable to speak?
21:39:26JoeBornwhich keeps the device current, without expensive hardware upgrades from the manufacturer
21:39:55 Quit TCK (Read error: 145 (Connection timed out))
21:39:56dwihnoUse some kind of standardardized battery, or make SURE there will be available cells in the future as well.
21:40:11dwihnoAnd mass storage compliance of course
21:40:28dwihnoNo stupid "windows media" this, or "synchronization manager" that
21:42:14SlasheriHmm, i would like something like iriver h1xx (software codec platform, optical out)
21:42:29preglowyou've got it!
21:42:41preglowi think most parts of the h1x0 is nice, actually
21:42:43preglowapart from the coldfire
21:43:15JoeBornwhat's wrong with coldfire?
21:44:15preglowit's a bit awkward and buggy in places
21:44:22preglowand really need assembly for fast dsp processing
21:44:25***Saving seen data "./dancer.seen"
21:44:35preglowwhereas on an arm, for example, you're very likely to get away with c
21:44:44rasherOr a better compiler than gcc..
21:44:54preglowwell, yeah
21:45:02preglowgcc could do it fine
21:45:07preglowbut there's not much interest
21:45:20preglowand i don't want to kill myself just yet, so i wont have a look at it ;)
21:47:49 Nick Sucka`away is now known as Sucka (
21:53:56 Join kenshin [0] (
21:55:05preglowcan't say i see why people is so fascinated with putting linux on protables
21:55:58rasherin the end, you end up building something like podzilla (in the case of ipl) on top
21:56:09rasherin which case all the linuxness is just dead-weight
21:57:05preglowyes, exactly
21:57:10rasherThat's my take on it, anyway
21:57:24preglowunless you're going for a linux system, i don't see much of a point
21:57:33preglowbut i don't see much point in a linux system on a portable :/
22:00:43*rasher ponders compiling a list of changes since 2.4
22:00:57rasher(for archos)
22:02:23*HCl got shot down in his airplane on san andreas, jumped out, and parashuted to safety just before hitting the ground o.o.
22:02:37HClpreglow: yup, its odd.
22:07:31rasherhrm, can I make cvs spit out a list of commits?
22:08:08HClcvs log
22:09:20Mooslike that ?
22:09:29Moosmade in Bagder
22:09:39Moosnot updated
22:10:16rashera matter of cvs log >2004-12-23
22:11:08Moosthis is it was all after the last Rockbox release
22:12:31preglowupdating digest? ;)
22:15:28rasherno.. I'd need someone to look at the scripts for that first
22:15:37rasherprobably Bagder
22:16:03Moosyes probably
22:16:07rashereach time I've tried to update it, thescripts fall over
22:16:30rasherI was more thinking of a release note for 2.5
22:18:33shankwill 2.5 have h120 support?
22:19:08rasherThat's not the plan, no
22:19:40rasherProblem is, h120 is still not quite there
22:19:56preglowwont be for a while either
22:20:04preglowthough all the major parts are in place now
22:20:06rasherAnd archos users are missing out on a lot of new features
22:20:17preglowit's only right to release 2.5 now
22:20:33preglowand introduce iriver support with a big 3.0 bang!
22:20:42shankno problem :-)
22:21:35rasherit seems the right path
22:23:54shanki'll give the latest cvs code a whirl
22:37:34 Join TCK [0] (
22:45:11Bagdergood evening crowd
22:49:29rasherBagder: I assume was created by a script - could you update it?
22:49:56rasherAttempting to write release notes for a future 2.5 release
22:50:35rasherI guess this should would be perfect for a wikipage
22:50:48 Join ]RowaN[ [0] (
22:50:59rasherI'll make one once I have something
22:51:37]RowaN[hey guys.. in the last week or so ive noticed that max volume on rockbox seems a lot lower than usual.. could there be some setting ive accidenly enabled or is this on purpose?
22:51:50]RowaN[(on my 120iriver)
22:52:20rasherThe bass and treble settings will affect how high the max volume will go
22:52:24rasherto avoid clipping
22:52:35]RowaN[ah thanks
22:52:39Bagderwe should add that to the iriver faq
22:52:54Bagderunless its already there...
22:53:19preglow]RowaN[: if you use really loud volume and eq, the volume will be turned down, but shouldn't be a problem with lower volume
22:53:20 Join jborn_ [0] (
22:53:27 Quit JoeBorn (Read error: 131 (Connection reset by peer))
22:53:45 Nick jborn_ is now known as JoeBorn (
22:56:03]RowaN[skip to previous track doesnt seem to work for me in latest build.. whats the deal with that
22:56:22]RowaN[unless i press it a ton of times
22:56:26Bagder2664 commits
22:56:54Moosamicoon n1 i assume :)
22:57:12]RowaN[ah i think its when im trying to play files in the fooked area of my hd =]
22:57:28Bagder 729 amiconn
22:57:28Bagder 570 bagder
22:57:28Bagder 469 linus
22:57:37Bagder 269 hcl
22:57:37Bagder 265 dave
23:01:28 Quit TCK- (Read error: 110 (Connection timed out))
23:03:17 Quit JoeBorn ("Leaving")
23:09:29Bagder24 different committers
23:10:15preglowwhat does that count, btw? number of commit sessions? files commited?
23:10:41Bagderthat commits per user is per file
23:11:04Bagderwhich really is the reason why I have a such a high number ;-)
23:14:43Coldtoastthe WPS refresh bugfix
23:15:13Coldtoastworks nicely. how about the subdirs issue?
23:15:30Coldtoastis that an easy fix?
23:15:51Coldtoastok. with my .wps and the images it uses in .rockbox, it works fine
23:16:16Coldtoastlooks liek that
23:16:40Coldtoastif I put images in a subdir, say .rockbox/wps/edan and adjust the path to the images in teh .wps, it looks like this
23:17:30Coldtoastthe little rockbox logo isn't there and you can see the "current track" image has some dots missing
23:18:19Bagderthe image name must not be longer than 31 letters
23:18:20ColdtoastI think the missing dots is the logo
23:18:31Coldtoastnot including the path surely?
23:18:38Bagderthe whole
23:18:49Bagderbut I think we should extend that
23:19:01Bagderseems a bit too restrictive in my eyes
23:19:05Coldtoastthat'll be the reason then
23:19:20Coldtoastit definitely is if you want to use subdirs
23:19:37Coldtoastthe logo is called "teeny_logo.bmp" which isn't unusually long
23:20:47Coldtoastthe whole image name with the lpath is "wps/edan/teeny_logo.bpm"
23:20:53Coldtoastwhich isn't over 31
23:21:04Bagderthen that's not the reason
23:21:31Coldtoastjust remembered; ep0ch tried my wps and just had one subdir and it did the same thing
23:25:30hardeepBagder: i have no idea how to fix the red gmini 120 build. It's a weird compiler bug
23:25:37BagderI noticed
23:25:47hardeepit doesn't like more than 4 case statements in that particular code
23:25:49Bagderthe gmini build is on a decay
23:27:09BagderI fear we will see more problem in there soon enough and have to remove it from the table
23:27:16Bagderunless someone from the gmini camp shows up
23:27:28hardeepwas anyone besides jyp working on it?
23:27:37Bagdernot really
23:28:03Bagderhe had some support with the hw-level, but afaik jyp wrote all the code
23:28:14Bagderand that includes the gcc port
23:29:22Bagderyeps ;-)
23:30:03Coldtoastjust did a test here
23:30:13Coldtoastdeleting a char from teh image name
23:30:29Coldtoast%x1|wps/edan/te_logo.bmp|116|117| works
23:31:02Coldtoast%x1|wps/edan/teeny_logo.bmp|116|117| doesn't
23:31:15Bagderyes, it seems it wants more than just the file name within 32
23:31:21preglowany plans to make the wps code be able to hardcode a font?
23:31:49Bagderwould be good for the ones sharing WPS codes
23:31:57preglowthe font i want for my wps isn't necessarily the font i want elsewhere
23:32:51BagderColdtoast: I believe this is actually due to a bug
23:33:06rasherper-line font-config in wps would be even nicer
23:33:22Bagdereven per-text
23:33:56Bagderthe scrolling is what makes multifonts tricky
23:33:58preglowthat's taking it a bit too far for my needs, but i can imagine it being useful
23:34:34Coldtoastyeah. letting the wps set the font would be good. So other ppl don't have to go changing fonts all the time
23:34:47preglowshouldn't making scrolling font-aware be semi-easily doable?
23:35:06Coldtoastand it'd be cool to have access to the rockbox icons
23:35:13Bagderwhat icons?
23:35:19Coldtoastso I have have an icon of a musical note in front ot the track name
23:35:25Coldtoasti the file tree
23:35:32Coldtoastyou have little musical note icons
23:35:34Bagderthat would be rather easy
23:35:45Coldtoastyeah. t0mas said t'd be easy
23:35:53ColdtoastI think it'd look neat
23:42:10 Join yyz [0] (
23:42:46 Quit webguest44 ("CGI:IRC")
23:43:54Bagdert0mas: booo
23:44:26***Saving seen data "./dancer.seen"
23:48:09Coldtoastdid you find out why that guy's .ogg wouldn't play? few hours ago
23:48:50Coldtoastwacky_ it was
23:49:01Coldtoasthad an album that wouldn't play on iriver
23:52:56yyzodd, my oggs play fine
23:53:05Bagderit would be interesting to get such a problematic song
23:53:14Coldtoasthe posted them
23:53:29ColdtoastI tried that one and yeah. doesn't work but does with the iriver firmware
23:53:44yyzi gotta download that, interesting
23:54:10BagderColdtoast: so that particular one fails?
23:54:20Bagderok, getting it
23:54:28Coldtoasthe said all the others did too
23:54:37ColdtoastI just tried that one tho
23:54:40BagderI bet one is enough to use for debugging
23:55:09Coldtoastit was encoded with an old version of libvirbis I noticed
23:55:09yyzhad he ripped the album himself, we know what kinda settings he used?
23:55:12Bagdert0mas: for the log: I mailed you a suggested patch
23:55:16Coldtoastfrom 2001
23:55:38Coldtoasthe'll hopefully wake up in a few hours
23:55:48crwllibvorbis 1.0 rc3 isn't terribly old, i've got older stuff somewhere
23:55:48Coldtoastit was wacky_
23:56:53crwli think it should have produced files that are compatible with all vorbis decoders, even with those that don't support floor0
23:57:38crwlNegative granulepos on vorbis stream outside of headers. This file was created by a buggy encoder
23:57:41Coldtoastslow server when I downloaded it
23:57:43crwlogginfo says stuff like that
23:57:51Coldtoastaah ok
23:57:53crwlso there probably is something wrong with it
23:58:03BagderI got it at 89kb/s
23:58:12yyz4.6 all the way.
23:58:27Coldtoasthaha! good ol' dialuo?
23:58:31yyzoh yes
23:58:46yyzthe country has some advantages. but connectivity isnt one of them

Previous day | Next day