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 2007-06-21

00:00:44 Quit Buschel ()
00:00:58 Quit bluebrother ("leaving")
00:05:45 Quit RoC_MM (Read error: 104 (Connection reset by peer))
00:05:46*linuxstb just realises petur also created a hard disk info wiki page...
00:06:27 Quit Arathis ("Bye, bye")
00:12:18 Quit Mouser_X (Read error: 110 (Connection timed out))
00:12:57 Quit aaronw ("ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502]")
00:14:14 Join JdGordon [0] (n=jonno@c220-237-57-32.smelb1.vic.optusnet.com.au)
00:15:01*DerPapst votes for linuxstb's hard disk info page and for peturs link collection.
00:16:37 Quit Febs (Read error: 104 (Connection reset by peer))
00:16:42DerPapstbtw iPod 3G there are 10GB 15GB and 20GB single plater and 30GB and 40GB dual plater.
00:19:17linuxstbDerPapst: It's a wiki ;)
00:19:20DerPapstand afaik there are ZIF connectors with 40 and with 18 "pins". maybe you want to mention that the iPod video ones are ZIF 40
00:19:34DerPapstwell i'm not registered..
00:19:44DerPapsti've tried once and failed ;)
00:19:47PaulJamlinuxstb: a little correction: the MK4004GAH is 8 mm high
00:20:04linuxstbPaulJam: Oops, thanks.
00:20:09linuxstbDerPapst: OK, I'll add that info
00:20:18DerPapstthanks ;)
00:22:02 Join fmoo [0] (n=fmoo|lt@static-74-41-12-115.dsl1.elk.ca.frontiernet.net)
00:22:13 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-cd31ca1a8f1c572a)
00:22:23linuxstbDerPapst: So I assume all the Toshiba drives listed are "ZIF 40" as well?
00:23:51DerPapsti guess so.
00:24:25 Join Soap [0] (n=Soap@rockbox/staff/soap)
00:26:35DerPapst0.85 Inch HDD = insane :D
00:27:29*linuxstb wonders what "Rockbox and iPodLinux offer open-source alternatives to the standard firmware and operating system, respectively." means on the ipod wikipedia page.
00:27:55 Quit jgarvey ("Leaving")
00:28:29 Quit lImbus (" zzZZzz")
00:30:06 Quit entheh ("^~")
00:32:21 Quit netmasta10bt (Read error: 104 (Connection reset by peer))
00:33:21*DerPapst doesn
00:33:33DerPapst't understand that statement
00:33:48DerPapst^ my return button is just too big :P
00:34:16 Join Mouser_X [0] (n=mouser_x@67.110.120.164.ptr.us.xo.net)
00:34:42 Quit ender1 (" Today's BOFH excuse: Zombie processes haunting the computer")
00:38:03 Join pengo [0] (i=pengo@60-242-139-27.tpgi.com.au)
00:38:23pengohow change the tempo of a track?
00:41:01linuxstbYou can only change the pitch in Rockbox.
00:41:06 Quit darkless ("Leaving")
00:41:34pengoi couldn't find that last time i looked either
00:41:58linuxstbSearch for "pitch" in the manual.
00:42:11pengothere's a manual?
00:42:36*Soap hopes that is an Australian sense of humor.
00:42:56pengo:]
00:43:08linuxstbThere's a website as well, and every page has a link called "manual" ;)
00:43:12 Quit luckz (Read error: 145 (Connection timed out))
00:43:32 Join jhulst_ [0] (n=jhulst@c-71-205-0-132.hsd1.mi.comcast.net)
00:43:45pengoall i'd need a manual for is to find out why "pitch" isn't with the rest of the audio settings, and what "party mode" is meant to mean
00:44:23 Quit jhulst (No route to host)
00:44:26linuxstbpitch isn't changed permanently, which is why it's not in the settings. But I can understand why you would expect it to be there...
00:44:58 Quit amiconn (" reboot...")
00:46:13pengoah i see.. i listen to large numbers of podcasts in a row so i want high-speed over many tracks
00:46:47pengothe manual only mentions pitch for the recorder?
00:47:20pixelmathere are manuals for every platform
00:47:49DerPapstpixelma: oh you're still awake?
00:48:10pixelmano... talking in my sleep ;)
00:48:15 Join amiconn [0] (n=jens@p54BD765E.dip.t-dialin.net)
00:48:20DerPapsthehe
00:48:40 Quit obo ("bye")
00:48:47DerPapstpixelma: is the flagcounting patch ok?
00:52:02pixelmahaven't tried yet, will do now (I'm not a big coder, only learned a bit TurboPascal years ago and reading a bit (about) C lately)
00:52:07 Quit Xerion (" ")
00:52:47DerPapstwhee Turbo Pascal... that is exactly how i started :-D
00:55:46 Join darkless [0] (n=darkless@62.79.44.48.adsl.vby.tiscali.dk)
00:56:18amiconnlinuxstb: Regarding ipodpatcher, would it be possible to detect whether the bootloader is already installed, and make the list of options depend on that?
00:57:40*linuxstb is also an ex-TP programmer
00:57:43DerPapstipodpatcher already detects if a loader is installed
00:58:10DerPapstit even can distinguish between the rb loaders and loadre 2 ;)
00:58:19amiconnoh, didn't notice...
00:58:35amiconnSo why does it offer to install the bootloader if it's already there?
00:58:42DerPapst−−list show all images that are in the fw partition
00:58:58DerPapstit updates the bootloader then
00:59:01linuxstb"install" replaces an existing bootloader.
00:59:28amiconnWhy would I need to replace a bootloader with its identical copy?
00:59:55DerPapstmaybe you'll have a newer version of ipodpatcher with a newer bootloader
00:59:55linuxstbIt doesn't check the version installed, just that there is any bootloader installed.
01:00
01:00:36linuxstbBut yes, it could be more intelligent, but I'm not sure it's worthwile.
01:00:41amiconnDerPapst: ipodpatcher offerent (i)nstall even though my mini g2 alraedy has the latest bootloader
01:00:41linuxstb^+h
01:00:55amiconn*offered
01:01:18DerPapstwould it be possible to get the loader version?
01:02:04linuxstbIt could read it from disk, and compare with the embedded version.
01:03:10DerPapstbut it wouldn't know if it is an older or newer loader. just that is is a different one?
01:03:50linuxstbTrue.
01:04:01DerPapstk... better than nothing ;)
01:04:35amiconnDetecting a difference might be useful, as that would also detect a damaged bootloader
01:04:46linuxstbThat kind of feature is probably more useful in rbutil though - to tell the user when they need to upgrade their bootloader.
01:04:53amiconnAlthough, that might be a rare condition...
01:05:44DerPapsti thought that rbutil would get that information from ipodpatcher
01:06:40linuxstbSort of. rbutil uses ipodpatcher as a library, and I think the functions are probably already there.
01:07:51linuxstbI just mean that the purpose of ipodpatcher is to install (or update) a bootloader, and no harm is done if the user installs the bootloader multiple times.
01:08:02 Join Llorean [0] (n=llorean@cpe-70-113-103-34.austin.res.rr.com)
01:08:05linuxstbSo it doesn't seem that important.
01:08:11DerPapsttrue
01:08:15 Quit matsl (Remote closed the connection)
01:09:05linuxstbWhereas one purpose of rbutil is to help users keep their install up to date.
01:12:45*linuxstb sleeps, goodnight.
01:12:47 Quit linuxstb ("CGI:IRC")
01:12:56amiconnHrrrm, pp5020 is nasty
01:13:15DerPapstheh pp5002 is even worse ;)
01:13:53amiconnThe mini g1 froze at 2 of 5 boot attempts
01:13:56 Quit joa (Read error: 110 (Connection timed out))
01:14:09***Saving seen data "./dancer.seen"
01:15:32DerPapstwhat are you trying to do?
01:15:43amiconnJust running rockbox...
01:17:24DerPapstoh? i thought a clean svn build is running fine on minis.
01:17:26pixelmaDerPapst: the newline doesn't seem to work this way - but maybe it's not even necessary. The splash also reads a bit cumbersome when you marked only 1 field, don't know if it is necessary to make it an extra case or if writing it "field(s)" would do...
01:17:33amiconnThis is unusable :(
01:17:43amiconnDerPapst: mini g2 works like a charm
01:18:04amiconnmini g1 freeze a lot while just browsing the db...
01:18:09amiconn*freezes
01:18:15 Join Tachikoma [0] (n=Tachikom@fl-71-53-129-3.dhcp.embarqhsd.net)
01:18:32amiconnANd only about 50% of the boot attempts succeed w/o freezing
01:18:39*amiconn disables CPU scaling
01:18:53*pixelma was about to suggest this
01:18:56Tachikomathe ipod photo with rockbox, can it play videos?
01:19:03DerPapstyupp
01:19:08Tachikomahmm
01:19:11Tachikomaok
01:21:41Tachikomahmm it wont play avi
01:22:09markunthere we go again.. avi :)
01:22:16DerPapstonly mpeg 1 and 2 and only in the resolution of you photo
01:22:16 Join Nico_P [0] (n=nicolas@jau31-3-82-239-20-145.fbx.proxad.net)
01:22:28 Quit ansivirus ("Leaving")
01:22:39Tachikomaok
01:22:52markunand audio needs to be mp1/2/3 at 44100Hz
01:22:58Tachikomahow come no other video formats?
01:23:12markunTachikoma: nobody implemented them
01:23:35DerPapstTachikoma: http://www.rockbox.org/twiki/bin/view/Main/PluginMpegplayer
01:24:15 Quit XavierGr ()
01:25:16saratogavideo might be a good GSOC project for next year
01:25:39Tachikomahmm
01:25:51Tachikomaso now i need a converter
01:26:05Lloreansaratoga: What about it?
01:26:10DerPapstsaratoga: are you volunteering? ;P
01:26:43DerPapstTachikoma: take a look at the link i gave you. it lists some encoders.
01:26:48LloreanYou'll always need a converter because at the very least you have to resize it to fit to the screen, so the argument for supporting other formats isn't terribly strong unless that format allows better quality at the same CPU load.
01:27:01saratogai have no interest in video
01:27:18saratogabut we'll probably get at least as many students next year, so we'll need new ideas
01:27:39saratogaand cleaning up our video support with some sort of real API would make sense
01:27:40Tachikomai did
01:28:05saratogabetter to do that before someone writes a second video format i think
01:28:28*DerPapst agrees
01:28:44 Join Thundercloud_ [0] (n=thunderc@82.152.203.15)
01:28:44 Quit Thundercloud_ (Read error: 104 (Connection reset by peer))
01:28:56 Join Thundercloud_ [0] (n=thunderc@82.152.203.15)
01:29:00 Join XavierGr [0] (n=xavier@ppp68-180.adsl.forthnet.gr)
01:29:42 Quit Soap ("Leaving")
01:30:00saratogamaybe once i get a better feel for how rockbox plugins work i'll write up someting in the GSOC wiki page about a generic videoplayer plugin
01:30:12*amiconn listens to aac tracks for the first time
01:30:26amiconnDoesn't sound very advanced, quality wise
01:30:52saratogaits just mp3 with different block sizes and better lossless coding
01:31:17saratogaand no hybrid filter bank
01:31:21amiconnThese are ~120kbps mp3... and they sound like 128kbps mp3 to me
01:31:23DerPapsti wonder why no one else had these problems though. maybe no other mini 1G users?
01:31:34amiconnmeh, ~120kbps aac I mean
01:31:48saratogayeah thats about right
01:32:02saratogathe HA listening tests never showed a statistical difference between the formats at 128k
01:32:36amiconnAt 128kbps, artefacts are clearly noticeable with a lot of tracks
01:32:47pixelmaDerPapst: there were, that's why the "no scaling" build exists in the unsupported builds forum...
01:33:10pixelmaapplies to 4g greyscale Ipods too
01:33:21pixelma(IIRC)
01:33:22DerPapstahh.. ok :)
01:33:52amiconnI would expect these problems on all the pp5020 targets
01:34:06 Quit PaulJam (".")
01:35:38pixelmayeah, but I _believe_ that scaling isn't enabled in the official H10 builds
01:35:56 Quit bdgraue (Remote closed the connection)
01:36:22pixelmaprobably for exactly this reason
01:38:23amiconnThat's a hackaround, not a solution...
01:38:34 Join Azraak [0] (n=jamie@oxcoda.safenetbox.biz)
01:39:10DerPapstpixelma: fixed :)
01:40:10*amiconn concludes that the first a of 'aac' can't mean 'advanced'
01:40:16AzraakHi. I'm trying to patch the menu code to talk in some places where it should but currently isn't; e.g. cancelling or accepting a setting change. It also speaks sometimes where it shouldn't.
01:41:17AzraakThe question: what is the general preference when having to call the same function many times? Normally, I'd make a boolean variable and just set it to true where required, but I'm nto sure which is pferred: making another bool var (I gues sthat's only 1 byte) or calling the same function with the same parameters in many different spots
01:41:43Azraakpreferred*
01:42:53 Quit Thundercloud (Read error: 110 (Connection timed out))
01:44:08amiconnCPU scaling is indeed disabled in cvs for H10 (both models)
01:44:14amiconns/cvs/svn/
01:45:24 Quit Nico_P (Remote closed the connection)
01:45:30 Quit Rincewind ("Cya")
01:46:25 Quit Tachikoma ("KVIrc 3.2.0 'Realia'")
01:49:04saratogaanyone here familar with how VLC coding works?
01:49:14 Quit chris_ (Remote closed the connection)
01:49:31 Join chris_ [0] (n=chris@S0106001346f5afbd.cg.shawcable.net)
01:55:38 Quit Nibbier (Read error: 113 (No route to host))
01:56:01 Quit Thundercloud_ (Remote closed the connection)
01:57:52DerPapstmaybe Antoine Cellerier is. mentor of GSoC at VLC http://code.google.com/soc/videolan/appinfo.html?csaid=9B7B6CA83B51E66D
01:57:55 Join Nibbier [0] (n=sven@port-212-202-177-141.dynamic.qsc.de)
01:59:15scorcheDerPapst: his nick here is dionoea
02:00
02:01:11DerPapstthanks ;)
02:14:14 Join Tim_Mitchell [0] (n=Tim@ool-44c63d6d.dyn.optonline.net)
02:19:00 Join Tostada [0] (n=chris@68-189-17-92.dhcp.rdng.ca.charter.com)
02:19:13 Part Tostada
02:19:17 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net)
02:19:40 Join dmccarthy [0] (n=dmccarth@74-33-123-178.br1.brz.pa.frontiernet.net)
02:20:07dmccarthyAnyone know what the status is on getting rockbox on the Gigabeat S?
02:21:11 Quit Tim_Mitchell ("Leaving.")
02:23:51DerPapstgood night at all
02:23:54 Quit DerPapst ("So Long And Thanks For All The Fish!")
02:28:34saratogadmccarthy: theres someone working on it
02:28:43saratogalast i heard they hadn't gottent he display driver working yet
02:29:26dmccarthysaratoga: thanks for the update...I saw some activity on the forums but wasn't clear if any more progress had been made
02:29:40 Part TrueJournals
02:31:05 Join webguest48 [0] (i=6152b561@gateway/web/cgi-irc/labb.contactor.se/x-61bc464101feaa59)
02:31:19webguest48jello
02:31:47 Quit webguest48 (Client Quit)
02:32:31 Join ansivirus [0] (n=ansiviru@adsl-76-225-180-38.dsl.rcsntx.sbcglobal.net)
02:32:59 Quit jhMikeS (Read error: 104 (Connection reset by peer))
02:33:16 Join dandin1 [0] (n=dandin1@bas12-ottawa23-1177806324.dsl.bell.ca)
02:34:39 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:37:00 Quit gtkspert (Remote closed the connection)
02:37:07pixelmadmccarthy: there's also a #gigabeat which is logged too - I guess people over there know more
02:37:32 Join gtkspert [0] (n=gtkspert@gateless.info)
02:37:40 Quit Mouser_X (Nick collision from services.)
02:37:51 Join Mouser_X [0] (n=mouser_x@67.110.120.164.ptr.us.xo.net)
02:41:26 Quit fmoo (Read error: 110 (Connection timed out))
02:42:46 Join fmoo [0] (n=fmoo|lt@74-37-78-20.dsl1.ekgv.ca.frontiernet.net)
02:42:47 Part pixelma
02:48:56 Join BHSPitLappy [0] (n=steve-o@adsl-66-139-196-107.dsl.rcsntx.swbell.net)
02:51:31 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net)
02:57:18 Quit dandin1 (Read error: 110 (Connection timed out))
02:57:32 Part TrueJournals
02:58:42 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net)
03:00
03:00:44 Join Febs [0] (n=chatzill@207-172-204-33.c3-0.rdl-ubr4.trpr-rdl.pa.cable.rcn.com)
03:02:53 Part TrueJournals
03:04:41saratogadmccarthy: check this out http://www.hack.id.au/files/gigabeat/irc/gigabeat/2007.06.20
03:04:49saratogaexplains the state of the project nicely
03:05:47pengocan i just say i want tempo for rockbox again :)
03:06:43scorchewell, you already said it, so i dont see a purpose of asking...
03:07:22scorcheignoring the fact that IRC is not the place for feature requests...
03:07:38pengook i wont say it then
03:07:46scorchebut you just did..
03:08:18pengowell...
03:08:55Lloreanpengo: Feel free to code up the feature and submit a patch.
03:09:07pengoLlorean: yes. yes i do feel free to do that.
03:09:23LloreanBut mentioning it again is just going to irritate people.
03:09:59pengoi did ask
03:10:19pengoif i could say it
03:10:27LloreanNo, you said it in the form of a question, there's difference.
03:10:38pengohrmm
03:11:06LloreanI assume you've actually followed the *proper* procedure?
03:12:09pengostep 1. bug people on irc.
03:12:25pengostep 2. assume someone else has already filed a feature request
03:12:31pengostep 3. repeat.
03:12:55Lloreanpengo: Step 1) Bug people on IRC, Step 2) Get banned for repeatedly violating channel guidelines.
03:12:55scorchea "no i havent" would have sufficed...
03:13:21pengoi can't click on things on the website because my mouse has stopped working
03:13:38LloreanSo what?
03:13:49Llorean"I can't do it the right way" is not an excuse for breaking the channel guidelines.
03:14:12***Saving seen data "./dancer.seen"
03:14:17pengoyou guys were friendlier last time
03:14:17LloreanNot to mention the internet is navigable by keyboard alone.
03:14:25pengowith difficulty
03:15:01LloreanStill, it negates your excuse entirely, since you can still do it the right way, you're just choosing not to.
03:15:31LloreanPeople were probably nicer the last time under the assumption that you didn't know or understand the rules yet. Now it's clear you don't have enough respect for the developers to follow their posted guidelines.
03:15:40LloreanIf you don't care enough about them, why should they care at all what feature you're interested in?
03:16:52pengo"go ask on tracker instead of here" would have done
03:17:18pengoi think http://www.rockbox.org/manual/ should redirect to http://www.rockbox.org/twiki/bin/view/Main/RockboxManual
03:17:45pengoor http://www.rockbox.org/manual.shtml
03:17:46LloreanWhere did you find a link to that page?
03:18:00pengoLlorean: google.com site:rockbox.org manual
03:18:02pengoor something like that
03:19:20pengosite:www.rockbox.org/ manual pitch
03:20:07LloreanHm, from that link I don't see it on the first page of search results
03:20:14LloreanMeanwhile I do see the official PDF manual on the first page.
03:20:30LloreanThough an older edition of it
03:20:31pengomust have gone to manual/wps.html and goen up one
03:20:42 Nick jhulst_ is now known as jhulst (n=jhulst@c-71-205-0-132.hsd1.mi.comcast.net)
03:20:52*Llorean didn't know the manual for 2.1 was still posted
03:23:10pengothe links from http://www.rockbox.org/manual.shtml are all broken 404 too
03:23:25pengomany are at least
03:23:57scorchenot for me
03:24:02LloreanSome are, some aren't
03:24:03pengoiriver 120
03:24:09scorchewait a bit for the DNS propogation
03:24:11LloreanThere's currently a problem with the daily manual build system
03:24:16pengoah k
03:24:21LloreanAt least for the HTML ones
03:24:59LloreanPDF ones, I imagine it's the DNS propagation as H120 works fine for me.
03:25:08pengohm k
03:25:09scorcheaye...as i said
03:25:16scorchewe just added a few mirrors
03:25:27LloreanSo I saw
03:26:00*scorche wished positive internet had a small banned instead of an avatar-size image
03:26:09scorches/banned/banner
03:26:51*pengo wishes when he installed a new bluetooth stack that it didn't require you to click on "connect mouse".
03:28:22 Quit spiorf (Remote closed the connection)
03:28:34 Join scorche` [0] (i=scorche@rockbox/administrator/scorche)
03:30:55pengoi'll shut up now
03:32:40 Join entheh [0] (n=purr@88-106-246-52.dynamic.dsl.as9105.com)
03:32:47 Quit darkless (Read error: 110 (Connection timed out))
03:34:56 Part fmoo
03:35:12*JdGordon finshed exams :D
03:37:33scorche`then what are you waiting for?...get to work...chop chop!
03:45:38 Quit scorche (Connection timed out)
03:45:38 Nick scorche` is now known as scorche (i=scorche@rockbox/administrator/scorche)
03:59:09 Quit saratoga ("CGI:IRC (Ping timeout)")
04:00
04:00:10 Join midgey [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
04:13:51 Quit ansivirus (Remote closed the connection)
04:14:36 Quit hcs (Read error: 110 (Connection timed out))
04:20:09pengoargh i hate acrobat's crapola integration with firefox
04:21:51 Quit JdGordon (Read error: 110 (Connection timed out))
04:27:02 Quit midgey ()
04:33:19 Join ansivirus [0] (n=ansiviru@adsl-76-225-180-38.dsl.rcsntx.sbcglobal.net)
04:34:44 Join midgey [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
04:39:14 Quit midgey (Client Quit)
04:41:59 Join midgey [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
04:42:03 Quit entheh ("^~")
04:42:52 Quit midgey (Client Quit)
04:45:11 Quit Azraak ("leaving")
04:55:50 Quit pengo ()
05:00
05:03:40 Join kkurbjun [0] (n=kkurbjun@c-71-56-227-141.hsd1.co.comcast.net)
05:05:08 Quit jhMikeS (Nick collision from services.)
05:05:14 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
05:10:10 Quit jhMikeS (Nick collision from services.)
05:10:16 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
05:14:16***Saving seen data "./dancer.seen"
05:42:20 Join Llorea1 [0] (n=llorean@cpe-70-113-103-34.austin.res.rr.com)
05:42:20 Quit Llorean (Read error: 104 (Connection reset by peer))
05:43:12 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
05:43:14 Nick Llorea1 is now known as Llorean (n=llorean@cpe-70-113-103-34.austin.res.rr.com)
05:55:38 Join toffe82 [0] (n=chatzill@65.198.26.227)
06:00
06:12:36 Join jdisco [0] (n=jdisco@pool-71-103-195-148.lsanca.dsl-w.verizon.net)
06:12:43jdiscohey guys
06:12:48jdiscoquick question
06:13:04jdiscodoes rockbox currently work with 5.5 macpods?
06:13:18scorcheafter they are converted to winpods, yes
06:13:24jdiscoah
06:13:32jdiscoi was hoping i wouldnt have to do that
06:13:35jdiscook ty
06:13:40 Quit jdisco (Client Quit)
06:41:12 Join Soap [0] (n=Soap@rockbox/staff/soap)
06:50:18 Quit Febs (Read error: 110 (Connection timed out))
06:53:41 Quit Mouser_X ()
06:56:07 Quit sando (Read error: 104 (Connection reset by peer))
06:59:03 Quit Jon-Kha (Read error: 145 (Connection timed out))
07:00
07:01:59 Join sando [0] (n=lolsteam@210.11.80.141)
07:02:19 Join ptw419 [0] (i=ptw419@216-188-249-122.dyn.grandenetworks.net)
07:02:51midkayat the front page: was someone so insanely lazy that they didn't bother capitalizing the first letter of the latest news item, or putting a period at the end?
07:03:34jhMikeSWhat is the intent of the "New File" trigger when it goes below the threshold. SVN splits and keeps recording but that just seems wierd. I'd think recording should pause, then split and then resume with the new file when the level goes back up.
07:10:00 Quit ptw419 ()
07:14:18***Saving seen data "./dancer.seen"
07:16:19Lloreanmidkay: Several don't have periods, and at least one other doesn't have a capital.
07:16:36LloreanThey're basically like commit messages but for the website. :-P
07:16:47midkayLlorean: yes, the news page is a total mess. but geez.
07:16:58midkayit'd take, what, half a second to proof read it?
07:17:16midkayLlorean: you mean commit messages written by jdgordon ;)
07:17:40scorchegrammar is overrated
07:17:51midkayscorche: you have server access, right? fix!
07:18:02scorcheno i dont
07:18:10LloreanI think only one or two people can add/change news.
07:18:39midkayLlorean: well, server access would mean you were able...
07:18:47scorcheit is a company server, so they dont feel quite free to give access out
07:18:59midkayscorche: did you not have it for a little while at leats?
07:19:01midkayleast*
07:19:05scorcheno
07:19:21midkayah, you said you were likely to get it or something..
07:19:31scorcherockbox.org/devconwest2007/ redirects to my page
07:24:37 Join ptw419 [0] (i=ptw419@216-188-249-122.dyn.grandenetworks.net)
07:25:22ptw419nothing
07:25:42ptw419think i did this adapter wrong somewhere b/c not even the led lighting up
07:25:45 Join RoC_MM [0] (n=Free@c-24-129-94-172.hsd1.fl.comcast.net)
07:26:13ptw419oops wrong chan
07:31:23amiconnjhMikeS: Did you see my ranting about how bad pp5020 behaves? Mini G1 svn build is unusable...
07:31:40 Join toffe82_ [0] (n=chatzill@65.198.26.227)
07:32:07 Join Mouser_X [0] (n=mouser_x@67.110.120.164.ptr.us.xo.net)
07:32:36amiconnPerhaps it varies between individual targets... at least this one is a perfect test candidate
07:34:04jhMikeSamiconn: sorry, I didn't. So it doesn't run at all then, huh? Maybe the other stuff will just work if that can be made to work. :\
07:34:22LloreanAnyone know if it's likely we'll have an iPod install method for pre-NT Windows versions, or does it seem that they'll remain unsupported for an indefinite period?
07:34:35*jhMikeS is just filling in the "doesn't run at all" bit
07:36:39amiconnIt freezes in ~50% of all boot attempts already (showing the rockbox logo)
07:37:17amiconnAnd if it doesn't freeze at boot, it will freeze after a few clicks for sure when browsing
07:37:56amiconnDisabling CPU scaling makes it work, including playback
07:39:22jhMikeSthere is a note about interrupts being wierd with scaling ... having to use |= TIMERx_MASK and all
07:40:07hcsthis all sounds very familiar
07:40:35*jhMikeS is also the lucky recipient of an iPod 3g without a cable and can't do anything with it right now. The e200 cable fits physically but doesn't actually work.
07:41:06*amiconn would like to help in solving the pp5020 mysteries
07:41:12LloreanThe e200 cable fits 3G iPods?
07:41:18amiconnI won't have this mini g1 available for long
07:41:35 Join darkless [0] (n=darkless@62.79.44.48.adsl.vby.tiscali.dk)
07:41:50jhMikeSIt fits but it'll get rather warm near the connector :)
07:42:18LloreanI was just wondering if maybe the dock was PP reference hardware too, originally
07:43:43jhMikeSamiconn: I'm not really sure what to do. Like getting the audio on e200 corrected it takes a zillion speculative code changes, pokes and animal sacrifices until it hits. :P
07:44:45Mouser_XAt least it doesn't require human sacrifices...
07:44:51Mouser_X(Though, perhaps they help?)
07:45:03*Mouser_X is *not* volunteering.
07:45:27 Nick blithe_ is now known as blithe (n=blithe@67.18.176.106)
07:45:36jhMikeSWell, the gods haven't demanded that _yet_
07:45:50Mouser_XAh.
07:45:52 Quit toffe82 (Read error: 110 (Connection timed out))
07:46:21amiconnI need some hints on how to grab the of's bare binary. Then I'll feed that through ida pro, and go hunting their clock changing routine...
07:46:27LloreanjhMikeS: We can see about that at DevCon '08 if all else fails
07:46:32jhMikeSamiconn: I guess I could provide the last patch and you could try with scaling off...I don't recall if linuxstb tested that
07:48:39*jhMikeS will volunteer to jump off the balcony ... hopefully the room will be on the ground floor ... with a balcony :)
07:49:03LloreanCan the CPU in the iPod be clocked to arbitrary speeds?
07:49:03*amiconn will ask linuxstb regarding the of binary
07:50:54LloreanMy understanding was that in the retail firmware there's some way to see something that seems to parallel load. It's a number that averages in the 29-34 range and seems to increase in value when an MP3 file is higher bitrate. This illustrated that the CPU didn't change fast enough, because with VBR files with bitrate spikes, the number would increase a larger amount right after the file skipped.
07:52:30 Quit BHSPitLappy (Read error: 110 (Connection timed out))
07:56:19jhMikeSida pro would be nice thing to have ... but a bit pricey atm
07:56:40Mouser_XWhich version...
07:56:59*Mouser_X has seen a few "around."
07:58:11jhMikeSI haven't looked into that...anything that can help organizing a disassebly to be easily readable and searchable would be nice
08:00
08:05:17 Join JdGordon [0] (n=jonno@c220-237-57-32.smelb1.vic.optusnet.com.au)
08:06:10 Join My_Sic [0] (n=MySic@mur31-1-82-237-204-133.fbx.proxad.net)
08:06:44 Quit amiconn (" reboot")
08:07:05 Quit My_Sic (Client Quit)
08:08:57 Quit GodEater (Read error: 110 (Connection timed out))
08:09:17 Join My_Sic [0] (n=MySic@mur31-1-82-237-204-133.fbx.proxad.net)
08:10:23 Quit My_Sic (Client Quit)
08:10:29 Join bdgraue [0] (n=bdgraue@dyndsl-085-016-168-052.ewe-ip-backbone.de)
08:12:35 Join rift_ [0] (n=opera@236.56.70-86.rev.gaoland.net)
08:18:19 Join amiconn [0] (n=jens@p54BD765E.dip.t-dialin.net)
08:22:05 Part toffe82_
08:29:24 Quit XavierGr ()
08:29:29*JdGordon hates the ipod scrollwheel!
08:29:54midkayand it hates *you*!
08:30:02JdGordononyl in soviet russia...
08:31:03midkayi see.
08:32:50 Join ender` [0] (i=krneki@84-255-206-8.static.dsl.t-2.net)
08:33:19 Join GodEater [0] (n=bryan@bb-87-80-121-64.ukonline.co.uk)
08:36:20daurnhey JdGordon
08:36:30JdGordonhey
08:36:41daurnsup
08:36:43GodEaterlast exam today JdGordon ? Or are you done already ?
08:36:48JdGordonnm
08:36:54JdGordonGodEater: finished! :D
08:36:57GodEateryay! :)
08:37:22JdGordoni just had a quick play with my sisters nano to fix the scroll bug... its gonna be a pita to fix :(
08:38:05*JdGordon would actually rather pass it off to one of the actual ipod guys :p
08:38:06GodEaterI can imagine - the scrolling issue *sounds* easy to fix, but as soon as you actually think about it... well, it makes my ears bleed.
08:39:17JdGordonI think the problem was that the wheel is too sensitive and is sending button_rel events too quickly
08:39:45jhMikeSis the scrollwheel nastiness the same nastiness the 3g has? it really should be made to work like OF which work as one would expect.
08:40:24JdGordonno idea
08:40:45midkaywhat scroll bug?
08:40:50JdGordonit would be nice if the piezo would sound like in the OF
08:41:00JdGordonaccelleration doesnt happen on ipods
08:41:03JdGordon... apparently ...
08:41:34midkaysure it does. it just depends if the screen updates quick enough.
08:41:43midkayyou're basing the accelleration on hz/10 now, right?
08:41:56amiconnJdGordon: Speaking about bugs - there is still a list bug related to font size change
08:42:00midkayif you boost the cpu it'll work fine. but at a lower boost rate, it doesn't update 10 times a second = accel won't take effect.
08:42:00jhMikeSIf it clicked along like retailos, it really shouldn't need it
08:42:52midkayJdGordon: good idea would be to make the "acceleration start speed" configurable. on ipods it'd need to be.. umm.. lower, so it would start sooner.
08:43:38JdGordonit waits byb default 500ms before starting to scrol
08:43:39JdGordonl
08:43:57jhMikeSif you rely on repeats and not velocity itself, it will always be weird feeling
08:44:02midkayi don't mean waits, but the actual .. well. i don't know how to describe it.
08:44:07midkayjhMikeS: but we don't rely on repeats anymore.
08:44:14midkaywe rely on a hz/10 threshold.
08:44:18midkaythat threshold needs to be configurable.
08:44:25midkayor a different default for ipods.
08:44:40amiconnAnd regarding the acceleration, it suffers from this problem because it's put in the wrong place
08:44:47JdGordonmidkay: hz/10 in the driver?
08:45:04midkayJdGordon: your own commit, 4th june.
08:45:05amiconnIt shouldn't matter at all how long the ui needs to update
08:45:07midkay"Fix FS #7242 - the action code now checks the time difference between events to decide if its a repeat or not. a repeat event is now if the previous action was the same and it occured < HZ/10 ticks ago"
08:45:31JdGordonah, hz/10 isnt fast enough?
08:45:47midkaythe problem is that ipods are slow, and at 30mhz (unboosted) some of them won't update quick enough to be _able_ to see scrolling at >hz/10.
08:46:21midkayat 75mhz (i.e. boosted) the screen/ui will update quick enough to fit at least 10 updates in a second, thus allowing acceleration to kick in.
08:46:48amiconnAnd the decision whether it's a repeat or not should only based on time difference for iput devices like the scroll wheel
08:47:09amiconnAll this should be in the button driver
08:47:13*jhMikeS may try something...even the sansa wheel can have a synthetic absolute position for velocity calculation
08:47:17amiconn...where it belongs
08:47:20midkayamiconn: i much agree on this, i'm a little annoyed in more than one place about the ipod's "if the finger's still in contact, it's a repeat" system..
08:47:36 Join kfazz [0] (n=ken@76-203-137-185.lightspeed.wlfrct.sbcglobal.net)
08:47:37GodEateryeah, that doesn't work
08:48:14amiconnmidkay: My statements are based on the fact that it seems this accel stuff is quite popular
08:48:14GodEaterif you've not let go, but paused your scrolling to check what's on screen, and then move your finger again, you expect the screen to start off slowly again, but it doesn't currently - it goes mental.
08:48:49jhMikeSit should be like a mouse drag...drag it far enough it's a repeat otherwise not
08:48:49midkayamiconn: i don't know what button_repeat has to do with acceleration - we no longer use it in calculation - but even before accel was introduced, it's always bothered me.
08:48:55amiconnPersonally I don't need it at all (but the repeat decision should be changed anyway)
08:49:04midkayGodEater: well, since 4 june it hasn't. it only did that for a few days.
08:49:22 Join Rob222241 [0] (n=Miranda@p54B16D4A.dip.t-dialin.net)
08:49:24 Join petur [0] (n=petur@rockbox/developer/petur)
08:49:40jhMikeSAcceleration should perhaps rely on the number of queue_posts skipped and send that data along with the message
08:49:46 Join kaaloo [0] (n=luis@rue92-3-82-232-48-241.fbx.proxad.net)
08:49:53 Part kaaloo
08:50:24amiconnjhMikeS: That would make it very ui dependent, which imo is a step in the wrong direction
08:50:28GodEatermidkay: you're right - I retract my statement. I must have only used it last around then and then turned it off.
08:50:45midkayGodEater: it's cool, it was changed.. but yes.. that was incredibly annoying. :)
08:50:56amiconnThis is what I don't like about the current acceleration code - it's *very* unpredictable how fast it will scroll
08:51:11GodEaterit still doesn't feel *quite* right - but it's a lot better than it was.
08:51:42GodEaterJdGordon: Who's been saying this is broken on the ipods ?
08:51:48jhMikeSamiconn: it's queue dependent not UI dependent and that preserves the speed continuity...ticks can't pull more than 100 messages out per second so no more than 100 can be posted anyway
08:51:59peturimho, the button (wheel) driver should give an indication of speed (but I know nothing about the ipod wheel driver)
08:52:41 Join B4gder [0] (n=daniel@static-213-115-255-230.sme.bredbandsbolaget.se)
08:52:42JdGordonthe ipod wheel is the only driver iirc that actually does some accelleration..
08:52:51JdGordonbut its still not good enough for long lists
08:52:53jhMikeSnever mind some of what I just said :P ... some is bunk and I'm fixing the core patch atm too
08:54:16*JdGordon thought the ipod booted into OF for usb?
08:55:04GodEaterJdGordon: it changes into disk mode, not the OF
08:55:43amiconnjhMikeS: The problem is that the ui sometimes stops pulling events from the queue if it's busy
08:56:18amiconnJdGordon: Scroll speed on ipod should imo be solely based on finger speed, not on scrolling time
08:56:50 Quit DataGhost (Nick collision from services.)
08:56:56 Join DataGhost [0] (i=dataghos@ip3e832ea5.speed.planet.nl)
08:57:02JdGordoncan we add some firmware call to get a fingure speed on wheel targets?
08:57:26SlasheriJdGordon: it's easy to calculate the finger speed
08:57:46 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:58:02Slasherii could send you the old scrollwheel acceleration patch i did, i haven't had time to develop it now
08:58:09peturmaybe only post messages containing the speed and direction, and the ui reads them all and uses the last one to do its scrolling
08:58:48jhMikeSpersonally I'd like a certain cursor speed associated with the finger speed and the number of posts/second should be related to that
08:59:14jhMikeSHow long does it go between queue checks in the UI anyway?
08:59:17Slasherithe number of posts/seconds has a problem..
08:59:39 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
08:59:45peturI think speed measuring must be done earlier, not by the gui looking at posts/sec
08:59:52Slasheriabove a certain limit, UI cannot receive the posts fast enough and queue size would need to be increased
08:59:56jhMikeSit won't
09:00
09:00:13Slasheriso the queue length limits the maximum possible scrolling speed
09:00:21JdGordonif we could add wheel_getspeed() to the driver for both ipod and sansas it would make things much nicer
09:00:42Slasherihowever, it's quite fast even with the current queue length (with a patch)
09:00:46amiconnjhMikeS: The number of posts/second is not usable to find out the speed
09:00:49peturcan't we add that info to the messages?
09:00:56jhMikeSqueue_empty would remain as a block against overflow, but a variable incremented for every post skipped and reset to 0 when a message is posted
09:00:59amiconn...because repeat events are only posted if the queue is empty
09:01:40amiconnI would use the number of wheel position changes per second in the button driver itself
09:01:43Slasheriamiconn: the speed can be calculated in the driver before posts are generated (as my patch does)
09:02:03Slasheriand then the driver fills the queue with wheel events
09:02:07amiconnThis number needs to be smoothed of course (sliding average or something like that)
09:02:07jhMikeSthe sansa wheel driver decides when and when not to post events based on velocity
09:03:03amiconnFinger movement isn't perfectly even
09:03:19Slasherion fast speeds that doesn't matter
09:03:29jhMikeSnope...hence hysteresis
09:03:39Slasheriit feels good when the driver reacts immediately to speed changes
09:04:46amiconnSlasheri: I remember how your old patch felt... unusable for me
09:05:15Slasheriamiconn: well, there should be an option to adjust the acceleration amount
09:05:16amiconnScrolling behaviour was erratic
09:05:30Slasheriah, that has been fixed a long time ago
09:06:57Slasherii will see if i can make a patch.. atm i don't have the ipod to try but the patch should be still working
09:07:20 Quit Rob2222 (Read error: 110 (Connection timed out))
09:07:21jhMikeSwell, basically what I'm talking about is the UI is informed of the number of posts skipped by the driver, it can then use that to say, skip that number of lines since that would be where is was if it were able to keep up
09:08:07SlasherijhMikeS: yes, that would solve the problem with too small queues
09:08:16 Join Jon-Kha [0] (n=Jon-Kha@a91-152-87-243.elisa-laajakaista.fi)
09:08:26B4gderGodEater: did you check that zip is installed on debussy?
09:09:45JdGordonbah, battery is flat.. no more ipod fiddling for a while
09:10:18JdGordonjhMikeS: did you have any ideas why the sd patch might cause problems with the 2x1gb chips?
09:11:02 Join KTurkcemIRC09759 [0] (n=KTurkcem@88.254.178.190)
09:11:06 Part KTurkcemIRC09759
09:11:41linuxstbamiconn: To extract the raw binary of the OF, use "ipodpatcher -rfb apple_os.bin" (assuming it's still installed on your mini).
09:11:52SlasheriJdGordon: http://ihme.org/~miipekk/rockbox/scrollwheel_acc.diff please try that. The code isn't very pretty and the events_pending thing is unnecessary (it doesn't do its job) and should be removed
09:12:00 Quit rift_ (Read error: 110 (Connection timed out))
09:12:57 Join mbr [0] (n=mbr@p3EE059D1.dip.t-dialin.net)
09:13:02amiconnhrrmm
09:13:02JdGordonok, ive saved it, ill try later if im alowd to use the ipod again :p
09:13:04jhMikeSJdGordon: not yet...possibly timing again
09:13:31amiconnIt's annoying having to run ipodpatcher as administrator
09:14:04Slasheriamiconn: you could try that as well
09:14:17Slasheriwith the patch, scrollwheel should be very speedy and reactive
09:14:20***Saving seen data "./dancer.seen"
09:14:30jhMikeSJdGordon: I'm guess it's that ata_init is called in SVN which inits the card, and then there's some time that passes then the internal card is accessed
09:14:56GodEaterB4gder: no - didn't think to - I assume it's not then ? :(
09:15:02jhMikeSotherwise it's basically identical
09:15:08JdGordoncan we read the chip speed before its accessed and make up a sleep time based on that?
09:15:23B4gderGodEater: it's just my guess that the failed copying from there is due to that
09:15:38 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:16:08jhMikeSwe can't at first and chips are allowed almost 1second for their power up sequence by the standard
09:16:09GodEaterB4gder: correct - it's not installed - I'll email Stephen
09:18:17JdGordonjhMikeS: by the sounds of it, the timeout is way too long also
09:18:22jhMikeSSlasheri: I also limited the number of backlight/buttonlight calls to 4/second on e200 to not flood that too...the iPod driver just seem to pour it full of them without filtering
09:18:58JdGordoncan we have a much shorter timeout and have a few retries before panincing?
09:19:04jhMikeSJdGordon: the .1s?
09:19:20JdGordonthe hour he said he waited before the eventual panic
09:20:11SlasheriJdGordon: oh, that doesn't sound good either
09:20:12jhMikeSWhat would ever have a panic after an hour? I also don't think a panic should ever happen if the card isn't present anymore.
09:20:14Slasheriups, jhMikeS
09:21:04linuxstbjhMikeS: Did you read the discussion in the logs about the stkov in mpegplayer on Coldfire (reported on H100 and H300) ? Petur tracked it down to the cfg file being created, so it seems the stack just overflows by just a few bytes - if the cfg file is there, it works.
09:21:05jhMikeSI'm not aware of any 1hr timeouts in there ... but never really scrutized it
09:21:41jhMikeSlinuxstb: no...which thread?
09:21:51linuxstbThe main thread.
09:22:43linuxstbSeems to have been introduced a few weeks ago - see the logs on (I think) the 19th June, at around 9pm.
09:22:48jhMikeSnot suprising since the coldfires in those players use whichever stack is current for interrupts
09:23:20JdGordon0x80000ms is about 8min right?
09:23:30linuxstbSo the Cowon players are different?
09:23:33JdGordonsd_check_timeout(0x80000, id);
09:23:58jhMikeSit uses the microsecond timer ??
09:24:15JdGordonUSEC_TIMER
09:24:41jhMikeSlinuxstb: no...same thing. later coldfire revisions have interrupt stacks
09:26:06linuxstbI can only think of two possible solutions, neither of which are nice - 1) Removing the 4608 byte libmad buffer from the main stack (and hence from IRAM); 2) Increasing the size of the main stack.
09:26:40 Join davina_ [0] (n=dave@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com)
09:29:38GodEaterB4gder: I've actually downloaded the source to zip, built it, installed it in the user's bin directory (along with the compilers), and added the directory to the PATH in .bash_profile and acbuild.pl - will that work ?
09:29:50B4gdershould work fine, yes
09:30:06*GodEater crosses his fingers
09:30:20jhMikeSlinuxstb: does it change things much to take it out?
09:31:01linuxstbjhMikeS: I haven't tested - it needs testing on a Colour coldfire, which I don't have.
09:39:00B4gderthese very uninteresting "sansa newsletters" have popped up lately
09:39:12peturlinuxstb: I can't remember tracking down mpegplayer troubles, must have been somebody else
09:47:52 Quit sando (Read error: 104 (Connection reset by peer))
09:48:43linuxstbpetur: Ah, it was Lear...
09:48:57 Quit linuxstb ("CGI:IRC")
09:50:27 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:53:04 Join sando [0] (n=lolsteam@210.11.80.141)
09:54:56jhMikeSamiconn: if you'd like to test run the iPod without any f-scaling: http://www.rockbox.org/twiki/bin/viewfile/Main/MichaelSevakis?rev=1;filename=core_lock3.patch.txt
10:00
10:00:18 Quit linuxstb ("Leaving")
10:03:22 Quit sando (Read error: 104 (Connection reset by peer))
10:03:26 Join Lynx_ [0] (n=lynx@tina-10-4.genetik.uni-koeln.de)
10:05:24 Join sando [0] (n=lolsteam@210.11.80.141)
10:13:04 Join rift_ [0] (n=opera@236.56.70-86.rev.gaoland.net)
10:16:37B4gderouch, june 20th passed without a single commit!
10:17:07LloreanB4gder: I should've saved the fixes for my colours related mistakes and spread them out over a few days, 'eh?
10:17:31B4gderyes!!
10:25:11 Join obo [0] (n=obo@host217-41-62-170.in-addr.btopenworld.com)
10:35:26amiconnjhMikeS: Can't test now
10:36:00jhMikeSthat's ok, a lock parity bug got in there when resyncing it so I have to find that
10:41:13 Join Nico_P [0] (n=nicolas@jau31-3-82-239-20-145.fbx.proxad.net)
10:42:16*Nico_P is now officially free from java :)
10:42:55 Quit GodEater ("Leaving")
10:43:53 Part maffe
10:44:19 Join maffe [0] (n=maffe@barmen.interhost.no)
10:45:38B4gderdionoea: here?
10:49:31 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
10:50:03 Join FOAD_ [0] (n=dok@dinah.blub.net)
10:53:18 Join GodEater [0] (n=bryan@bb-87-80-121-64.ukonline.co.uk)
10:53:33 Join jas6180 [0] (n=jas6180@c-71-202-129-6.hsd1.ca.comcast.net)
10:54:13 Quit Nico_P (Remote closed the connection)
10:54:28jas6180Is there a list of guidelines for programming plugins and such? For example I believe I read somewhere that using malloc is 'bad', but I don't remember where or exactly what was meant by 'bad'.
10:54:51JdGordonmalloc is completly out....
10:54:57JdGordondocs/CONTRIBUTING
10:55:10jas6180ah, i guess that counts as bad...okay, that hurts. thanks
10:57:07 Join desowin [0] (n=desowin@avc146.internetdsl.tpnet.pl)
10:58:17GodEaterwell - using malloc would be bad if we supported it. But since we don't - it's not possible. :) :)
10:58:53 Join spiorf [0] (n=spiorf@host140-213-dynamic.0-79-r.retail.telecomitalia.it)
10:59:06 Quit ze (Read error: 104 (Connection reset by peer))
10:59:29 Join ze [0] (i=ze@cpe-76-175-22-254.socal.res.rr.com)
10:59:35jas6180yeah...i hadn't counted on that, i was planning on using an outside library that seems to use malloc rather heavily. ill have to see if i can change that soemhow
10:59:50GodEaterthat'll probably take quite a bit of work
10:59:53GodEaterwhich library was it ?
11:00
11:00:44jas6180it's part of gnugo, the SGF parsing library. actually from the looks of it it might only use malloc a few times, maybe i can hack something together
11:01:21 Join Nico_P [0] (n=nicolas@jau31-3-82-239-20-145.fbx.proxad.net)
11:01:31GodEaterdo you understand the reason for no malloc() ?
11:01:52GodEaterand therefore why you should avoid pretty much all methods for dynamic memory allocation ?
11:01:56jas6180well, i assume it's because most of the platforms aren't exactly swimming in memory
11:02:38Zagorjas6180: it is rather due to the undeterministic nature of malloc
11:02:47Llorean32mb is a fair amount of memory.
11:02:53GodEaterdon't we also lack MMUs on all targets ?
11:02:58 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
11:03:12B4gder... and the fact that we want all "unused" memory to be used ;-)
11:03:38Zagoryeah, that's reason #1
11:03:40B4gderGodEater: not on gigabeat afaik
11:03:43*GodEater remembers when his desktop PC had 32MB, and he thought it was a lot :)
11:03:55jas6180haha, maybe i'm missing something. i suppose i don't understand why malloc wouldn't be possible to implement
11:04:09B4gderpossible it would be, good it would be not
11:04:10Zagorjas6180: it is possible to implement. but we don't want it.
11:04:19*B4gder is yoda today
11:04:20GodEaterB4gder: are you saying the gigabeat actually possesses one ?
11:04:24B4gderyes
11:04:27GodEaterwow
11:04:27Zagorjas6180: static allocation is always better
11:04:59Zagor(for us)
11:05:09B4gderGodEater: arm9 cores come with it
11:05:14GodEaterneat
11:05:33B4gder(and that's also why all those linux-using DAPs use "real" linux and not uclinux...)
11:05:43linuxstbSo our friend the DM320 has one as well? (IIRC that's an arm9?)
11:05:47B4gderyeps
11:05:53B4gder"real" linux in there
11:06:02linuxstbAny idea of the clock speed of the DM320?
11:06:06linuxstb(the ARM part)
11:06:09B4gder~200MHz
11:06:13B4gderiirc
11:06:29linuxstbSo Rockbox would run well, even without a DSP...
11:06:35B4gderindeed
11:06:42Zagorhaven't we written a wiki page about "why no malloc" already?
11:06:43LloreanAnd they've stated they don't care about video.
11:07:02linuxstbAssuming the LCD is less than 320x240, then mpegplayer should be fine too.
11:07:04jas6180is there one? i didn't see it.
11:07:09*GodEater thought he'd summed up "why no malloc" quite well on the GoldenQuotes page
11:07:12B4gderZagor: I don't think so
11:07:17 Quit FOAD (Read error: 110 (Connection timed out))
11:07:34Zagorthen we should. it's quite a common question. understandably so.
11:07:48B4gderdm320 only requires a black blob to load to the DSP, and then we can pretend it is a gigabeat ;-)
11:07:56LloreanIt might actually be good to have the dm320, then.
11:08:05LloreanEncourage more people to hack at the DSP
11:08:07B4gderthe DAC is controlled by the dsp
11:08:26linuxstbLlorean: It's better to have a documented CPU...
11:08:38jas6180is there anything else that common that isn't available? i'm going to have to look through a lot of this and see how much needs to be reworked i guess.
11:08:57scorchejas6180: fixed point only
11:09:02Nico_PLlorean: I've finished school this morning so I'm now free for rockbox and my SoC :)
11:09:13Nico_Pbut this afternoon I have a plane for Paris
11:09:15Zagorjas6180: c++ ;)
11:09:16Lloreanlinuxstb: It always is, but if it's going to cost them more to switch CPUs, maybe that could be spent elsewhere.
11:09:21jas6180ah yes, thankfully this shouldn't be using floating point at least
11:09:30jas6180lol, yes i wish i could use c++ honestly, but i'll manage
11:09:39LloreanNico_P: Great news.
11:10:42linuxstbLlorean: Maybe they should look elsewhere for a reference design, rather than just looking as far as Neuros. IIUC, this isn't a Neuros project, but they've said they'll help.
11:10:45Nico_Pisn't there an arm-elf-g++ ?
11:11:15scorchethere is
11:11:16Nico_Por just c++ compilers for our targets ?
11:11:49linuxstbLlorean: It's just that designing an open player around proprietory hardware seems to be a pointless exercise.
11:11:54Nico_Pis c++ less efficient than C ?
11:11:56Lloreanlinuxstb: I definitely agree on that one.
11:12:07GodEaterlinuxstb: totally
11:12:14ZagorNico_P: c++ comes with a lot of baggage that would weigh down the binary size
11:12:15peturmessage to wiki admin: remove HarddiskInfo as another artivle made it obsolete
11:12:29Zagoralso c++ code typically makes heavy use of dynamic allocation
11:12:32B4gderNico_P: the way many people use it, sure, but it doesn't have to be
11:12:34linuxstbpetur: Sorry about that... ;)
11:13:00peturno problem. I added my links to your page, mine can go now...
11:13:04Nico_Pok
11:13:25jas6180well C tocde typically makes heavy use of dynamic allocation as well :)
11:13:38Nico_PI don't really see a reason why C++ would be appealing in this case, though, just being curious
11:14:01Nico_Pmaybe I have too much OOP in my system :)
11:14:05Zagorjas6180: yes, but much more explicitly. c++ hides it better (which imho makes it worse)
11:14:06jas6180in general it is somewhat more bloated size and memory wise, that's probably the main reason
11:14:19jas6180yes, that is somewhat true zagor
11:14:23***Saving seen data "./dancer.seen"
11:15:00RhinoBangabut look at the gains ... reference counting smart pointers!
11:15:11Zagoryay
11:15:15*GodEater readies his pitchfork
11:16:06jas6180so i assume if i implement a small scale malloc just so i can reuse a lot of this library, nobody is going to be too happy if i try to get it included in svn eventually, huh?
11:16:30*GodEater lights his torch and summons the villagers
11:16:39jas6180ya, that's what i thought...
11:16:39LloreanDoesn't Doom have a fake malloc?
11:16:53*Llorean seems to recall hearing mention of one somewhere, at least
11:16:55Zagorjas6180: plugins are free to do pretty much anything they want
11:17:23Zagorumm, what were you looking at again? :)
11:17:50Zagoryeah, plugins. then I wouldn't expect much opposition.
11:18:06jas6180i'm writing a game recorder/viewer for Go...mostly for myself but i figure eventually i'll probably share it around if possible in case magically someoen else wants it
11:18:11*GodEater douses the torch and sends the villagers back home grumbling.
11:18:13Zagorwe'll whine at you, but we'll do that anyway ;-)
11:18:18jas6180haha, okay cool
11:18:41*B4gder shuts off the villagers emulation (tm) he had running for GodEater
11:18:50GodEaterhehe
11:19:08 Quit toer ("leaving")
11:19:08B4gderthat plugin will be released soon
11:19:16 Join toer [0] (i=tore@skjeldal.com)
11:19:39jas6180yeah, maybe i'm too used to having dynamic alloc, but i can't even think how to do this algorithm without at least fake dynamic
11:19:54B4gderit takes a little getting-used to
11:21:34peturmost people are used to PC's eating more and more memory until they need to get rebooted
11:21:47 Join toer_ [0] (i=tore@skjeldal.com)
11:21:49 Quit toer_ (Client Quit)
11:21:56jas6180haha, only the windows users among us petur
11:22:07jas6180dynamic memory doesn't mean you have to leak everywhere, if you do it right
11:22:25peturit can happen anywhere
11:22:43peturbugs are not just written for windows
11:22:54jas6180...it was kind of a joke
11:23:31*petur uses both windows and linux and is fine with both
11:24:44RhinoBangaI use both but I prefer coding in windows ... the MS IDE/debugger is sooooo much better than anything available on unix
11:24:58B4gderI disagree
11:24:59GodEateraye - VisualStudio is pretty slick
11:25:21GodEaternot that I use it :)
11:25:22jas6180visual studio is rather nice, one of the few microsoft programs i like. still, programming on linux just seems nicer to me
11:25:43jas6180probably just more familiar with it
11:25:43*GodEater is one of those rare people who likes both vi and emacs.
11:25:47Zagorvisual studio is only nice because windows lack the proper tools to do multi-window programming
11:25:49RhinoBangaI'm in the middle of writing an app that will automagically ftp code from unix to my pc for editing in VS and whenever it changes ftp it back
11:26:16RhinoBangawhat do you mean Zagor?
11:26:17B4gdertry samba ;-)
11:26:35jas6180yeah, wouldn't some kind of network file system eliminate that need rhino?
11:26:47GodEateror sshfs :)
11:26:52*GodEater loves fuse
11:26:53RhinoBanganope ... when you have 1000+ machines with code on them you don't want to go around installing samba on every one ... ftpd on your client is much simplier
11:26:59B4gderI doubt fuse runs on windows
11:27:08B4gderbut sshfs is way cool indeed
11:27:11ZagorRhinoBanga: no shells, no external debuggers, no call trace tools etc.
11:27:16GodEaterI'm fairly confident it doesn't
11:27:29Zagorvisual studio is thus REQUIRED for windows programming
11:27:36*GodEater uses sshfs all over the place
11:27:37B4gderthey don't even have valgrind
11:27:39RhinoBangadefo stuff for call trace
11:27:53ZagorRhinoBanga: not all the way
11:28:03RhinoBangatrue about valgrind ... but then I use purify anyway
11:28:06B4gderalthough there are some fine expensive alternatives
11:28:30B4gderbut, to each their own
11:28:31RhinoBangaZagor: what do you mean? The call history stuff I've implemented works just fine
11:28:46RhinoBangaalthough adminitadly you have to ship a .PDB
11:29:08ZagorRhinoBanga: for your code, yes. what about the 3rd-party libraries? what about the windows kernel?
11:29:31RhinoBangastill get a traceback
11:29:43Zagoranyway, I didn't mean to start a "war". sorry, I'll stop...
11:29:46RhinoBangawhat are you using for tracbacks?
11:29:55B4gderstrace
11:29:58B4gder:-)
11:29:58RhinoBangalol ... it's not a war ... just a conversation
11:30:20RhinoBangamaybe I'm doing something different that will benefit you
11:30:31B4gderor just "gdb -p <pdi>" ctrl-c "bt" ;-)
11:30:47RhinoBangawe've found many a time gdb hangs
11:30:51jas6180oh good, from the looks of it fake mallocs are pretty common in plugins, i guess i won't be the first
11:32:18RhinoBangamy win32 stack dump code uses the dbghelp.dll and imagehlp.dll for info
11:32:38linuxstbjas6180: Apologies if you've been told this before, but one limitation on your plugin is that it only has 512KB available (for both code and data). If you want it to run on Archos devices as well, they only have 32KB for plugins. If you need more memory, you need to stop audio playback and grab the audio buffer.
11:33:12jas6180oh no i hadn't heard that. i assumed there was a limit, but wasn't sure what. thanks very much :)
11:33:22 Join The-Compiler [0] (n=florian@221.46.76.83.cust.bluewin.ch)
11:33:33jas6180i was wondering why some plugins stopped audio too, now it makes sense...thankfully i think that should be more than enough for me
11:34:51RhinoBangaLinusN: you around? Remember that id3 tag limitation I found? I changed the size of the id3 static buffer in id3.h from 300 to 600 and rockbox won't build the db properly anymore. Change it back to 300 and everything is fine
11:34:59linuxstbSome plugins also stop playback because they need the entire CPU - e.g. pacbox doesn't need much memory, but needs the CPU.
11:35:18linuxstb(apart from on the gigabeat)
11:35:43jas6180ah, yeah i'm sure playing audio takes up a good bit of the cpu
11:36:00Zagorlinuxstb: does it actaully refrain from stopping playback on gigabeat? just curious
11:37:49linuxstbZagor: It does.
11:37:56Zagorneat
11:38:03jas6180alright, thanks for the help everyone. i'm going to go hack on this for a bit and see how things go.
11:38:08 Quit jas6180 ("using sirc version 2.211+KSIRC/1.3.12")
11:38:23LinusNRhinoBanga: perhaps the 300 byte limit is hardcoded somewhere else?
11:40:15RhinoBangahmmm ... never thought of that
11:40:19 Quit rift_ (Read error: 110 (Connection timed out))
11:40:22*RhinoBanga goes off to do a grep
11:41:43*RhinoBanga cancels the grep after seeing how many 300's there are :O
11:42:23 Join Febs [0] (n=chatzill@207-172-204-33.c3-0.rdl-ubr4.trpr-rdl.pa.cable.rcn.com)
11:42:28JdGordonNico_P: so your all finished with school now?
11:42:32JdGordonRhinoBanga: did you try reinitialising the db or just updating?
11:42:43RhinoBangarebuilding every time
11:42:44JdGordonhaha
11:42:48JdGordonjhMikeS: you still round?
11:43:10RhinoBangaI went from 300 to 600 ... so now trying 400
11:44:01Nico_PJdGordon: yes :)
11:44:18JdGordoncongrats :)
11:44:49Nico_Pthis morning was the presentation of my group java project
11:45:05Nico_Pit went well and I'm pleased to be over with that project
11:45:39JdGordonso... MoB done in a week? :D
11:45:51Nico_Phaha maybe, who knows ?
11:46:29JdGordonI started my buffering code again yesterday but didnt spend much time on it... hopefully ill spend an hour or 2 on it tonight
11:46:33Nico_Pif you help me we could achieve almost 24h/day coding :)
11:46:41Nico_Pok
11:46:58scorcheat the average 8 lines per hour...
11:47:20scorche;)
11:47:51JdGordonNico_P: no, different time zones means 3odd hours/day possible
11:47:57JdGordon30 odd*
11:48:29Nico_Pwhat does "odd" mean ?
11:48:48RhinoBangaroughly
11:48:51JdGordonapprox.. roughly... some random figure around 30
11:49:01Nico_Poh ok
11:49:28Nico_Pyeah well that's good for productiveness :)
11:49:44 Join bluebrother [0] (i=v5gdWvka@rockbox/staff/bluebrother)
11:49:58JdGordonchrisjs169: you round?
11:50:10RhinoBangaexcept trying to schedule code releases ... hoping not to break someone's build in the other side of the world
11:50:53*Nico_P has to go for a bit... bbs
11:51:30jhMikeSJdGordon: yus
11:51:58JdGordonjust looking at the ata_init(), shouldnt the whole function be enclosed in the if (!initialized)?
11:52:23JdGordonwhy lock and unlock the spinlock if we dont actually do anything?
11:52:24jhMikeSpretty much
11:52:50jhMikeSwhat do you mean "don't do anything"?
11:53:24JdGordonif initialized == true nothing is done between the lock and unlock except turn the led off
11:54:59jhMikeSbecause one thread can be doing a transfer and another can call ata_init when the transfer is yielding
11:55:46JdGordonbut the only thing its doing is slowing the system down... once it gets the lock it immediatly unlocks it
11:55:55jhMikeSdid I add a sleep into ata_init there or was that later in my own patch
11:56:16JdGordonits there, HZ/10
11:56:37jhMikeSOh, ata_init does yield in several places, so the lock must be in effect
11:56:56jhMikeSI'm not sure what the led function does either
11:57:26jhMikeSthings will have to be more locked down if the COP is allowed disk access at some point
11:58:00JdGordonno, the locks are needed, fine, but in the case where its already intiialized, nothing is done in ata_init(), so waiting for the lock just blocks a process which doesnt really need to
11:58:34JdGordonhttp://pastebin.ca/580993
11:58:44jhMikeSat the moment, it's not critical but setting of initialized will be
12:00
12:00:08jhMikeSshould be fine for the time being
12:00:28JdGordonjust seems like a waste to me...
12:00:35 Part maffe
12:00:47 Join maffe [0] (n=maffe@barmen.interhost.no)
12:00:50JdGordonnot that it makes a difference, atm its only ever called once anywway
12:00:51jhMikeSif not initialized yet? no, it's needed to be thread safe
12:02:30JdGordonhmm?
12:02:41jhMikeSthat's why I want an ata_init that only gets called upon system boot and guaranteed to be accessed only once by the main thread and another for the usb otherwise I'll need more locks
12:05:04 Quit Siltaar (Remote closed the connection)
12:08:20Zagorfirst version of WhyNoMalloc. feel free to improve: http://www.rockbox.org/twiki/bin/view/Main/WhyNoMalloc
12:08:28jhMikeSwith usb connection I can't guarantee which thread gets there first really
12:12:36JdGordonZagor: no "5) Coz we said so!" ?
12:12:58Zagorhehe
12:13:38*petur fears his usbotg stack he's porting uses mallocs
12:17:25 Join Thundercloud [0] (n=thunderc@82.152.218.72)
12:17:26ZagorI would expect it does. not much non-embedded code uses static allocation.
12:17:26jhMikeSPoint 2 might need revision ... nothing keeps things from crashing if the audiobuffer available gets too small or even negative sized. I think a minimum playback reserve is in order as well as some loose management.
12:17:26 Quit LinusN (Read error: 60 (Operation timed out))
12:17:26markunjust when I'm editing the wiki rockbox.org stops responding here..
12:17:26 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
12:17:26peturmarkun: works fine here
12:17:26markunpetur: it works fine on my computer at home, so must be something here at work
12:17:26ZagorjhMikeS: yeah, but those are technical details. the principle of #2 is still valid, especially for segmented areas.
12:17:55JdGordonbah, my internet is fubar atm.... can someone with a voice file test a patch for me please?
12:19:40jhMikeSZagor: segmented areas = the various preallocated memory spaces like the plugin buffer?
12:20:34Zagoryes I mean the areas defined in the linker file
12:21:36 Join norbusan [0] (n=norbusan@dhcp24.dmg.tuwien.ac.at)
12:23:07markunpetur: my textmode browser did the trick :)
12:23:08*jhMikeS is still unable to delete outdated wiki page attachments from his own :\
12:23:21 Part norbusan
12:26:20Nico_PZagor: nice page
12:26:45markunLinusN: what is the reason that modifying attachments is not allowed for us?
12:27:31*linuxstb always thought of the wiki like csv/svn - you can't delete attachments as it destroys the history
12:27:45linuxstbs/csv/cvs/
12:28:29markunZagor: I think there are also cases where some very simple kind of dynamic memory would make sense
12:29:34markunfor example, if a jpeg is too big for the plugin ram, instead of taking the whole audio buffer it would make more sense to me to just take the audio buffer which has already been played back or which is last buffered
12:29:54markunbut yes, it increases complexity
12:30:26markunbut to take 30MB of buffer just because you need a little bit more RAM is also weird imo
12:30:26Zagormarkun: indeed it does. the audio system would have to be adjusted to account for a movable base pointer
12:30:27*jhMikeS was scheming something up the allow the audiobuffer to be partially taken and playback resumed and then freed again
12:31:21markunZagor: moving the pointer and then memmoving if needed?
12:32:55ZagorI question the need. pausing music while viewing Really Big Pictures (for a dap) isn't that much of a deal imho
12:33:16jhMikeSthe SWCODEC "base pointer" does move
12:34:14linuxstbjhMikeS: So your idea would be to stop playback, rearrange memory, then resume the last playing track?
12:34:33jhMikeSwe do it when enabling/disabling crossfade already
12:34:39*JdGordon thought about a ring buffer which could handle gaps a while ago which would work
12:34:45JdGordonbut its annoyingly complicated :p
12:35:26Zagorlunch
12:35:33jhMikeSnot really, just stop playback, tell it the buffer is trashed (which is automatic) and then restart it and it resets with the new pointers
12:35:42amiconnThe ability to reallocate buffers would indeed be nice
12:35:56JdGordon*cough* malloc
12:35:57JdGordon:D
12:36:01amiconnnope
12:36:03markunwithout malloc :)
12:36:21 Quit kfazz (Read error: 110 (Connection timed out))
12:36:27LloreanI never really understood the objection to stopping playback for certain thing as long as there was a prompt, and it resumed from where you were after.
12:36:31Lloreanthings
12:36:37amiconnJust what crossfade does, with restarting playback
12:36:44jhMikeSwhen recording starts, or IRAM is taken, things get moved around a lot already
12:37:33JdGordonthe buffering im playing with atm has a pointer where the currently valid data starts, and a pointer to where the last read was, so potentially everything between could be trashed at will without any problems
12:37:37jhMikeSif all we want is to say "give me 1MB"...well take the MB and restart playback...easy
12:37:48Nico_PMaybe allowing to take some free space from the main buffer could be one of my design goals when implementing MoB...
12:38:16amiconnThe buffer should never be fragmented imho
12:38:17Nico_PI don't see why it couldn't be possible
12:38:22markunLlorean: if you throws away the buffer and you have to rebuffer when resuming then I think it's a waste of battery
12:38:24 Join kfazz [0] (n=ken@76-203-137-185.lightspeed.wlfrct.sbcglobal.net)
12:38:34JdGordonamiconn: you just stopped me posting a different suggestion :p
12:38:54Nico_Pamiconn: but is it a problem when some data isn't needed anymore, to allow its space to be used ?
12:38:57Lloreanmarkun: Of course it's a waste of battery, that's why you prompt the user in advance so they know the rebuffer will happen and have the chance to just not enable the setting until next playback stop
12:39:28Lloreannext manual playback stop, rather
12:39:37Nico_PI think there must be some complexities involved that I don't fully understand yet...
12:39:41*jhMikeS apparently hasn't said enough about the details of already having made it trivial to do it all :P
12:39:43markunLlorean: why not keep the buffered data in memory as you are probably not going to use much of it anyway
12:40:31 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:40:36Lloreanmarkun: If you're not fragmenting the buffer, you need to either take from the beginning or end, right?
12:40:49JdGordonor the unused part
12:41:10JdGordonfragmentation is unavioidable if we want to allow this sort of thing
12:41:21markunLlorean: it is avoidable
12:41:38LloreanWouldn't that depend on the buffer's condition at the time?
12:41:40Nico_Pwhat's wrong with a bit of fragmentation ?
12:41:49markunyou can just move the part of the ringbuffer which you still need to the new smaller ringbuffer
12:43:20markunNico_P: what's the benefit of it?
12:46:15Nico_Pmaybe having to do less moving
12:46:58Nico_Pthe way I see it, if you strictly refuse a fragmented buffer it adds complexity... but maybe I don't see it right :)
12:47:23Nico_Pit adds complexity when you want to handle what we are talking about, that is
12:47:45markuna fragmented buffer adds complexity when the codecs are reading from the buffer
12:48:09JdGordonnot necasserily...
12:48:22markunwell, then it's ok :)
12:48:40*jhMikeS was just thinking of a single mem block at the buffer start that can be used by whatever and be expanded or shrunk if needed
12:48:41B4gdersay hi to http://tbrntech.rockbox.org/
12:49:02B4gder(yeah yeah you can't resolve it yet... :-))
12:49:35B4gderwill appear in the round-robin soonish
12:49:49markunB4gder: shows up fine here
12:49:56markunor what should I be seeing?
12:50:04B4gderjust the root dir of the download mirror
12:50:12markunyes, that's what I get
12:50:55The-CompilerWhere are the html-zip-manuals going?
12:51:00markunB4gder: will the elefants dream movies be moved here, or are they fine where they are?
12:51:05B4gderthey are there
12:51:16B4gdercheck the mpeg subdir
12:51:23markunah yes
12:51:29B4gderI put them into this "regular" system now
12:51:37B4gdersince we now have the downloads distributed to multiple mirrors
12:51:52B4gderThe-Compiler: /dev/null
12:52:02jhMikeSwhere's the list of X.rockbox.org's? seems to be growing
12:52:12 Quit chris_ (Remote closed the connection)
12:52:18B4gderI'll write up a wiki page about them soon
12:52:29 Join chris_ [0] (n=chris@S0106001346f5afbd.cg.shawcable.net)
12:52:34B4gderwe're all supposed to access them as download.rockbox.org
12:52:46 Quit linuxstb (Read error: 113 (No route to host))
12:53:06B4gderwell, and when one of them has problems we get mails like those sent to the ml recently
12:53:20B4gdersince one of the servers seem to have a bad vhost document root for download*
12:53:24 Part maffe
12:56:26jhMikeSamiconn: I updated the core_lock.patch.txt to get rid of the mistake, so whenever you feel up to it...it's on my page there. the other link given earlier isn't the one.
12:58:55chrisjs169JdGordon: I'm gonna be around later today
13:00
13:00:22*petur wonders why tbrntech.rockbox.org works immediatly and videolan.xxx and positive.xxx still not
13:00:43B4gderyeah, that's... weird
13:00:58B4gder... but they are CNAMEs and tbrntech is an A one
13:04:45B4gderJun 21 12:56:34 giant named[1639]: transfer of 'rockbox.org/IN' from 91.191.140.131#53: failed while receiving responses: REFUSED
13:04:55 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
13:04:55B4gderoops
13:10:00B4gder"present in too many channels/windows error"
13:10:34 Quit Febs (Read error: 110 (Connection timed out))
13:11:27B4gderthe download problem should now be fixed
13:14:25***Saving seen data "./dancer.seen"
13:25:47bluebrotherB4gder: have you noticed that the html manuals haven't been updated for a couple of days?
13:25:57 Join PaulJam [0] (i=Paul@vpn-3061.gwdg.de)
13:26:00bluebrotherthe ones I checked are dated 15th, while the pdf is up to date.
13:26:27B4gderyes, the builds fail
13:26:37B4gderwhich is why there are no zips either
13:26:43bluebrotherstrange −− they build fine for me.
13:26:53 Join the_conley [0] (n=conley@207.171.198.179)
13:26:57B4gderI'll try some manual builds later on
13:27:20B4gderI think I need to do some of the work they pay me for...
13:27:34bluebrotherbtw, I want to change the name of the output file −− manual-h300 instead of rockbox-build. Any objections?
13:28:03bluebrotheryou already use manual-h300 on the download page.
13:28:09 Part the_conley
13:28:19linuxstbbluebrother: Does that mean changing rbutil?
13:28:57bluebrotherlinuxstb: depends on. Currently the output file is named rockbox-build.pdf, while on the download page it's rockbox-h300.pdf (and similar)
13:28:58B4gderbluebrother: nah, just wave the flag before you commit that so that I can fix the server side script
13:29:10bluebrotherok.
13:29:24bluebrotherbut I could also use rockbox-h300.pdf
13:29:48linuxstbbluebrother: I was thinking about the html - would your name change affect the URLs that rbutil uses to browse the manual?
13:30:07bluebrotherI had an idea about fixing the dash issue (need to verify it once more though) which also gives us this as a neat side effect
13:30:18bluebrotherlinuxstb: yes, most likely.
13:30:54linuxstbbluebrother: What about just using printf - I posted a patch to flyspray, but never got around to committing it....
13:31:01bluebrotherbut the server could mod_rewrite the old urls (it's only the first part of the html file basename that'll change)
13:31:22linuxstbIt's easy to change rbutil - I just wanted to check if it needed to be done.
13:33:13 Join printfXh4 [0] (n=pseudo@ppp121-45-200-18.lns1.bne1.internode.on.net)
13:33:15B4gderbluebrother: with the new multi-mirror setup, we can't do much server-tweaks
13:34:48linuxstbHow does the mirroring work? Do you run rsync whenever new files are added, or is it synced via a cron job?
13:35:17B4gderthe mirrors rsync hourly from the master
13:35:49 Join Febs [0] (n=chatzill@38.98.196.75)
13:36:04bluebrotherlinuxstb: sorry, was interrupted shortly
13:36:06B4gderand to complicated matters, the master itself actually rsyncs from the daily build server...
13:36:31bluebrotherwell, you wrote that printf can be compiled into dash. What happens if it isn't compiled into it?
13:36:40B4gder... but every 10 mins during the "build period"
13:37:31 Quit spiorf (Read error: 110 (Connection timed out))
13:37:32bluebrotherI wanted to check that first. In the meantime I tried a different approach using sed, and I think it's easier in terms that it doesn't rely on features of the shell
13:37:54B4gderprintf is a regular stand-alone tool
13:37:56linuxstbbluebrother: Then it uses /bin/printf which appears to be ubiquitous.
13:38:28B4gdereven mentioned in "the single unix spec"
13:38:28 Join spiorf [0] (n=spiorf@host37-22-dynamic.21-87-r.retail.telecomitalia.it)
13:38:41linuxstbMy understanding is that printf is intended to replace echo.
13:38:52bluebrotherreally? /me goes checking
13:38:56B4gderyes, at least the "fancy" parts of it
13:39:12linuxstbOr not even that - echo seems to vary far too much, even for simple things.
13:39:30B4gderyes, echo can only be used with no flags and no backslash stuff
13:42:17linuxstbB4gder: So it's just the daily builds on the download server? The current builds are on build.rockbox.org, which is still a single server?
13:42:24bluebrotherhow's the state of printf on mac? I hope not to run into problems like with the xargs thing ...
13:42:27B4gderlinuxstb: yes
13:42:45*linuxstb goes to fetch ibook
13:42:55B4gderbuild.rockbox.org will remain a single server for now
13:43:11B4gderwe'll just move it to the other server (soon)
13:43:42 Quit gtkspert (Read error: 110 (Connection timed out))
13:44:17linuxstbbluebrother: Mac OS X uses bash, and that has a built-in printf will be used. There is no /bin/printf though.
13:45:10bluebrotherhmm. On my machine it's /usr/bin/printf ...
13:45:15linuxstbAh, there's a /usr/bin/printf...
13:45:20linuxstbBeat me too it.
13:45:26linuxstbto it even.
13:45:32bluebrotherok, then I guess it's safe to assume printf being present.
13:53:37markunFreeBSD does also have /usr/bin/printf
13:55:39bluebrotherhmm. Is "The single unix spec" available on the net somewhere? Google doesn't find it for me :/
13:56:24 Join gtkspert [0] (n=gtkspert@gateless.info)
14:00
14:03:06 Quit Nico_P (Remote closed the connection)
14:16:01 Join RaRe [0] (n=Laffin_B@202-89-187-101.static.dsl.amnet.net.au)
14:16:54linuxstbmarkun: Do you have a "disklabel.h" somewhere under /usr/include ? (I think maybe /usr/include/sys)
14:17:28markun/usr/include/sys/disklabel.h
14:17:59B4gderbluebrother: try opengroup.org, like => http://www.opengroup.org/onlinepubs/007908799/xcu/printf.html
14:18:23linuxstbOK. According to google, that should be in Mac OS X as well, but isn't... They must have removed it in recent versions. I need it for ipodpatcher to get the disk geometry.
14:18:48 Quit RaRe` (Read error: 110 (Connection timed out))
14:25:10 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net)
14:30:38JdGordonanyone else on ubuntu feisty?
14:30:51 Join Arathis [0] (n=doerk@p5484BE35.dip0.t-ipconnect.de)
14:32:28JdGordonyay, build servers get to do something again :)
14:33:20B4gderyay!
14:33:42*B4gder noticed that the dust had started to settle on them
14:35:54 Join midgey [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
14:37:52linuxstbApologies for the off-topic question, but I am trying to upgrade my kernel from 2.6.16 (Debian kernel) to 2.6.21 (built myself) and it's not finding the root filesystem. The kernel is finding the disk (displaying it as ata3) and I have the filesystem (ext3) compiled into the kernel. Anyone have any ideas or tips for debugging? e.g. can the kernel be made to display known devices? My current kernel is booting fine with the root=/dev/s
14:37:53linuxstbda5
14:38:58B4gderis it due to initrd stuff or other modules not present built-in?
14:40:29linuxstbI've built in all the modules I think are needed (i.e. the sata driver and ext3). But yes, it could be initrd related, I'm not sure how to deal with that.
14:40:50B4gderwith everything built-in it shouldn't be a problem...
14:41:03 Part TrueJournals
14:41:23linuxstbThe kernel messages display the model name of the disk, so I'm assuming the sata driver is OK.
14:42:57JdGordonamiconn: the list font problem is only is you go from a big font to a smaller one right?
14:45:59peturJdGordon: I'm using feisty on my laptop, but I'm new to linux so I'm probably not much of a help ;)
14:46:19*Llorean is in the exact same situation as petur.
14:46:36JdGordondo you get compile warnings when you do a target build?
14:46:38daurnok
14:46:43daurnthis may be a stupid question
14:46:54daurnbut, is there a player smaller than the 2nd gen ipod shuffle?
14:46:56peturJdGordon: none unless I write bugs ;)
14:47:17LloreanCompiles are fine for me, at east as of a couple revisions ago
14:47:22JdGordon:(
14:47:22*petur even manages to write bugs that do not give compile errors
14:47:24Lloreandaurn: Check out the MobiBlu cube
14:47:38B4gderlinuxstb: so what is the last thing you see on boot?
14:47:40daurnLlorean: isn't than an inch cubed?
14:48:21Lloreandaurn: A little under, yes.
14:48:41daurnshuffle is smaller :( - in 2 dimensions at least
14:48:54daurnand... a third of that is a belt clip
14:49:06PaulJamJdGordon: when you change to a bigger font it can happen that the selected line is below the visible area.
14:49:09linuxstbB4gder: Just something like "panic: can't find root filesystem", I forget the exact message.
14:49:26JdGordonPaulJam: ah yes, justs eeing that now
14:49:54linuxstbB4gder: Before that, it displays some info about disks (but not any /dev/ devices - just "ataN" and "scsiN")
14:50:10Lloreandaurn: The shuffle is greater than 1 inch in 2 of the 3 dimensions
14:50:14JdGordonlinuxstb: you compiled ext3 inbuilt, not as a module?
14:50:19linuxstbJdGordon: Yes.
14:50:27daurnLlorean: shuffle is 27x10x10mm
14:50:36daurnLlorean: mobiblue is 24x24x24mm
14:50:51JdGordonlinuxstb: you'll have much more luck in ##linux :p
14:50:53daurnoh
14:50:53*linuxstb hasn't compiled his own kernel for about 5 years...
14:50:54daurnwait
14:50:56daurnI'm an idiot
14:50:58Lloreandaurn: Not according to apple's ow site.
14:51:00Lloreanown
14:51:04*daurn gives Llorean a cooki
14:51:13daurns/cooki/biscuit
14:51:15daurndamn american
14:51:23JdGordonlinuxstb: oh wait.. did you do the full install? make install module-install etc?
14:51:28linuxstbJdGordon: Do you know if they are a friendly bunch in there?
14:51:33LloreanIn terms of volume, the shuffle's still smaller.
14:51:35JdGordonthey are not :)
14:51:43LloreanIn terms of wedging it into interesting places, the cube probably works better.
14:51:45JdGordonthey are typical linux ppl
14:51:53linuxstbJdGordon: I built it "the Debian way" using make-kpkg
14:52:07JdGordonhmm.. ok, no idea then
14:52:09daurnI just want something VERY small
14:52:23daurn - something I can hang from a shirt...
14:52:28daurnI'm very fussy
14:53:01B4gdernot that fussy, as you clearly can't live without rockbox ;-)
14:53:02peturdaurn: get one of those headphones with build-in mp3 player
14:53:05B4gdercan live
14:53:11LloreanJust get a shirt with a breast pocket, and put a Nano in it.
14:53:29peturyes, get some better clothes
14:54:02daurn... who has a brest pocket in a non-business shirt?
14:54:13*petur puts hand up
14:54:26*Llorean raises his own as well.
14:54:26JdGordondaurn: ipod flea ftw
14:54:31*linuxstb has trouser pockets, jacket pockets, sometimes a bag...
14:54:38LloreanI'm sure Apple will produce something smaller soon enough
14:54:45daurnlinuxstb: in shirt i said
14:54:48B4gderlots of pockets
14:54:50B4gderalways
14:54:58daurnI have plenty of pockets in my pants...
14:55:00linuxstbdaurn: You only wear a shirt?
14:55:32 Part LinusN
14:55:32daurnno, I'm just sick of have bulging pockets - I'm trying to go in the extreme opposite direction :P
14:55:48Lloreandaurn: Oh, so it's more anti-pants than pro-shirt.
14:55:51petur'the ipod-nothing' <−− very cheap, very light, very long battery life (many years): You just have to sing yourself
14:55:55B4gderbackpack!
14:55:56LloreanWhy not one of those wristwatch MP3 players?
14:56:02B4gderneed to fit all those rockbox targets ;-)
14:56:06LloreanI bet there's an s1mp3 based one, you could then get involved in that project.
14:56:18daurnLlorean: cause I don't want to plug headphones into a watch...
14:57:36daurnJdGordon: do you know what that ship off the nsw coast that got beached is called?
14:57:52JdGordonnup
14:58:02 Quit Thundercloud (Remote closed the connection)
14:58:35 Quit RhinoBanga (brown.freenode.net irc.freenode.net)
14:58:35NSplitbrown.freenode.net irc.freenode.net
14:58:35 Quit preglow (brown.freenode.net irc.freenode.net)
14:58:56NHealbrown.freenode.net irc.freenode.net
14:58:56NJoinRhinoBanga [0] (n=rhinoban@host-87-74-81-29.bulldogdsl.com)
14:58:56NJoinpreglow [0] (n=thomj@rockbox/developer/preglow)
14:59:15daurnJdGordon: http://youtube.com/watch?v=jfHvLAv0x-M
14:59:20daurnthe news told everyone to look it up :P
14:59:27daurnnot that good at all....
14:59:54*JdGordon is capped... no youtube for /me
15:00
15:00:04daurnhaha
15:00:13daurnsucks to be you/using your internet connection
15:00:18JdGordonam I dreaming or was there a function to call to update the number of liines in the display?
15:00:49daurnJdGordon: who you with?
15:01:19JdGordonoptarse
15:01:39daurnheh
15:01:45daurni haven't been with them in years ;)
15:02:06LloreanAnyone know a little bit about putsxy?
15:04:05*JdGordon knows a little bit
15:04:16JdGordonx,y are in pixels....
15:04:30LloreanI got that much. ;)
15:04:45LloreanVs the char position of so many of the others.
15:05:26LloreanI was just going to ask if there's an obvious (to someone not-me) reason why there isn't the option to style it right now, beyond 'it hasn't been needed yet'
15:05:48*JdGordon doesnt know
15:06:41*Llorean shrugs
15:06:58daurnlol
15:07:09daurnI thought you were making a pun on putty or something :S
15:07:44Lloreandaurn: Nah, lcd_putsxy which is the pixel positioned parallel to lcd_puts (which uses character pusitioning)
15:07:54LloreanExcept the XY variants don't seem to have an option for style
15:09:30 Join midgey_ [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
15:09:51 Quit midgey (Read error: 104 (Connection reset by peer))
15:12:07 Quit The-Compiler (Read error: 104 (Connection reset by peer))
15:12:52 Quit linuxstb (Read error: 113 (No route to host))
15:14:01 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
15:14:28***Saving seen data "./dancer.seen"
15:15:17linuxstbLlorean: What are you trying to do that needs putsxy?
15:16:16amiconnLlorean: There are other differences between lcd_putsxy and lcd_puts
15:16:19Lloreanlinuxstb: I was looking at hacking a separate WPS text color (separate from "Foreground") into my personal build, just to see what it'd take for WPS line colors eventually
15:16:37Lloreanamiconn: I'd noticed, though I can't say I understand them yet.
15:16:47amiconnlcd_puts pads to the end of the line, lcd_putsxy does not. That's one reason why there is no style variant
15:18:44JdGordonamiconn: PaulJam: wanna try out a patch to make sure its all fixed? or shall i just commit?
15:18:52linuxstbLlorean: Can't you just set the fg colour before calling lcd_putsxy?
15:18:54Lloreanamiconn: Why does that prevent it from being styled?
15:20:13Lloreanlinuxstb: I was pretty sure I could do that, but I was trying to find out if there were other ways.
15:21:13JdGordon3...2....1....
15:21:53JdGordonwas there a FS task with the list bug?
15:24:27 Join Xerion [0] (i=xerion@cp198589-d.landg1.lb.home.nl)
15:27:29daurnhaha
15:27:32daurnvery nice: http://www.shimafuji.co.jp/product/semc5701a01.html
15:27:52daurn - very small computer ;)
15:28:42JdGordonwoohoo, 2 green builds in a row :) should I go for three?
15:29:35JdGordonchrisjs169: ping?
15:29:40 Quit dmccarthy (Read error: 110 (Connection timed out))
15:30:28markundaurn: looks nice :)
15:30:54daurnevery port you could think of for something that size...
15:31:07daurnanyone able to understand what arch it is?
15:32:25markunhttp://www.am.necel.com/micro/product/vr/vr5500a.html
15:32:26ZagorVR5701 is a NEC cpu with MIPS core
15:33:05*B4gder worked with NEC chip with MIPS code
15:33:10B4gdercore
15:33:15B4gdernot that one though
15:33:17markunB4gder: nice?
15:33:38B4gderI worked with a dual-core one and it was nice enough, yes
15:33:57B4gderfor a DVB-T thing
15:34:13Zagornude pic: http://www.linuxdevices.com/news/NS3619879482.html
15:34:19B4gder... included mpeg2 decoders etc
15:34:53B4gderI believe NEC is one of those secret companies though
15:35:30B4gderI recall I googled for the chip I worked with and there was 0 (!) hits
15:36:38JdGordoncan someone have a quick look at #7222 please?
15:36:53*JdGordon wants to commit recording directory patch but needs that in first
15:37:08 Quit hcs ("Leaving.")
15:40:57daurnso... what do you guys think of the shuffle?
15:41:02daurnworth buying?
15:41:39daurni wish something that size had wavpack support :(
15:43:32B4gderthis is rockbox, we think rockbox-able players are better
15:44:29daurnwell... the smallest target isn't small enough :P
15:45:12 Quit Arathis ("Bye, bye")
15:45:13B4gderit's a matter of priority
15:45:22markundaurn: if you can find the specs of a small player perhaps a rockbox port is possible
15:45:33 Join Arathis [0] (n=doerk@p5484BE35.dip0.t-ipconnect.de)
15:45:48daurnmarkun: dont all the small ones just have a ROM?
15:45:51peturaren't those iFP players small?
15:46:54markundaurn: do they?
15:47:02daurnteh question is: have there been any firmware upgrades for shuffles?
15:48:01daurnanyway, you'd think that they'd use seperate mp3 decoding chips
15:48:50markundaurn: the new shuffle uses a samsung ARM and has firmware updates, but it probably has the same encryption problems as the nano 2g
15:49:03daurno.o
15:49:49daurnalso... it doesn't have much in the way of buttons
15:49:51B4gderdaurn: you better realize that whatever player you find (of that size), the firmware is gonna be major crap
15:49:51daurn - or a screen
15:50:16daurn:(
15:51:41markundaurn: http://www.dapreview.net/comment.php?comment.news.3618
15:52:35Lloreanlinuxstb: I think I'd actually have to convert putsxyofs to accept colors just because it's necessary for scrolling
15:53:12 Quit midgey_ (Read error: 113 (No route to host))
15:53:35 Join toffe82 [0] (n=chatzill@65.198.26.227)
15:54:25markundaurn: http://www.pmpz.net/picture_library/nano_s10.jpg
15:54:39*Llorean wonders if that suggests the bug he thinks might exist right now
15:54:48daurnmarkun: heh thanks
15:55:01daurndamn
15:55:02daurnexpensive
15:55:10markunhttp://prylportalen.mkf.se/ArticlePages/200611/07/20061107144404_MKF416/iriver_ipod-ihoppa.jpg
15:55:47daurnmarkun: anyway, its the depth
15:56:02daurnwithout the belt clip, the shuffle is tiny
15:56:12daurnhow do you hold onto the S10 :P
15:56:14Lloreanlinuxstb: I'm surprised nobody's reported this, but lines aren't colored while they're scrolling. Of course, it's only noticeable with the arrow line selector. =/
15:56:23markunthe S10 also uses a Samsung ARM http://www.iriverfans.com/news_detail.asp?ArticleID=888
15:57:24B4gderI find it amusing that Neuros is doing their new player without having a HW engineer
15:57:41B4gderor at least is considering it
15:57:45markunB4gder: is it realy Neuros who is doing the player?
15:58:03daurnits *meant* to be neuros
15:58:03B4gderwell, neuros would make it, but it is a bit community driven
15:58:10daurnbut, it was sort of abandoned a year ago
15:58:19 Join Thundercloud [0] (n=thunderc@82.152.218.72)
15:58:24markundaurn: not the one we are talking about
15:58:27daurn - may had the urge to make it happen
15:58:50daurnbut, its not as much neuros any more, as community
15:58:58daurnand... I don't see it going in the correct direction
15:59:06markundaurn: http://forums.rockbox.org/index.php?topic=11084.0
15:59:35daurnmarkun: I've read it a couple of times already
15:59:51daurnand the several mailing list discussions
15:59:53daurnand the irc logs
16:00
16:00:41B4gder"I expect it will take some involvement on our part to make this a reality" / Joe Born
16:00:59B4gderpretty vague, yes ;-)
16:01:01markundaurn: what made you think it was abondoned a year ago then?
16:01:35B4gderI think it was, this is just one of the community members dragging it up to the surface again
16:01:41markunah ok
16:01:45 Quit jhMikeS (Nick collision from services.)
16:01:47daurnmarkun: it was going to be a grand neuros venture - the neuros OSD was a quick test on the hardware - making a product at the same time
16:01:51 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
16:02:08daurnthe OSD becomes the flagship model - its all they make now
16:02:15daurnnot even finished properly
16:02:36daurnwas scheduled to be completed and on shelves - by september last year
16:02:55daurnN3 and a device called the 442v2 were meant to be created in the next couple of months
16:03:13 Quit PaulJam (".")
16:03:15 Quit Zagor ("Client exiting")
16:03:27daurnthe last dm320 product being the N3 - has everything, audiophile quality DAP - taking over where iriver h340 etc left
16:03:37daurnbut... it vapoured
16:03:51daurnnow in the last couple of weeks, there is this proposal from may
16:09:12 Join Rondom [0] (n=Rondom@p57A950EC.dip.t-dialin.net)
16:09:26 Quit JdGordon ("Konversation terminated!")
16:10:12daurnis it a problem that my aged, 3-4 year old dap is still more advanced that pretty much everything on the market in it's class?
16:10:53zedaurn: i know what you mean (i'm still using a rio karma :p)
16:11:18daurnI just keep waiting and waiting for something to replace it
16:11:24daurnbut... it never comes
16:11:24B4gderwell, nobody seem to be really intereted in making players in that class anymore
16:12:10daurnwhat I'd give for a good, 1.8" 8mm hdd player...
16:12:21B4gderor at least I consider the top segment to be somewhat emptied out, with a much bigger focus on smaller players and more "simple" hdd based ones
16:12:23daurnthats open ;)
16:12:26 Quit Arathis (Read error: 110 (Connection timed out))
16:12:51B4gderyou just said you wanted the world's smallest one
16:12:56B4gderyou won't fit a hdd in them
16:13:03daurnI know
16:13:13daurnI currently have the archos 402 - as you probably know
16:13:20 Join Arathis [0] (n=doerk@p5484BE35.dip0.t-ipconnect.de)
16:13:21daurnI need something to replace it
16:13:33daurnbut... in the meantime, I want something very small - and inexpensive
16:14:24daurnthere are 3 devices out there with 8mm drives
16:14:48daurniriver h340, iAudioX5 (60gb) and ipod.
16:14:58daurniriver isn't made anymore
16:15:33daurnand I'd like features and choices other than cowon and apple offerings
16:15:49markungigabeat F40, X60
16:16:27markunhave 8mm as well but also aren't produced anymore
16:17:12markunthe F40 can be found pretty cheap (around $100)
16:17:55daurnoh well
16:18:07daurnbefore I goto bed, one more topic
16:18:08daurnhow are bluetooth headphones?
16:18:20 Quit pearldiver (Read error: 110 (Connection timed out))
16:19:27 Quit linuxstb ("Leaving")
16:19:34B4gderhow?
16:19:35daurnno opinions?
16:19:47daurnwell, are they inferior quality?
16:19:52daurnare they too expensive?
16:19:59daurnany recommended models?
16:20:02B4gderI don't use any
16:20:08daurndoes anyone use them?
16:20:50markunnot me
16:21:13LloreanI can't imagine they can compare, price wise, with wired ones.
16:21:24daurnnow I think about it, I saw a guy using some at a tram stop a month or so ago
16:21:29LloreanI'd only ever consider them if I had a regular situation where a wire was impossible.
16:21:42daurnony time i've seen some i think
16:21:54daurnI'd just love to have nothing in my pockets
16:22:00 Join luckz [0] (n=luckz@luckz.de)
16:22:01daurnhave player in backpack or w/e
16:22:38B4gderwell, "everyone" uses them for phones here
16:22:51daurnjust some cord around the back of my neck with canal-phones and the reciever hanging down the back a bit (or like a medallion??)
16:23:18markundaurn: maybe you can find some review on head-fi.org
16:24:50daurnoh well
16:24:51daurnbed
16:24:53daurnnight all
16:25:00amiconnStandard bluetooth (1.0) isn't fast enough for hifi
16:25:11B4gderI believe they use 2.0(?) these days
16:25:12 Quit Mouser_X (Read error: 104 (Connection reset by peer))
16:25:23 Join Mouser_X [0] (n=mouser_x@67.110.120.164.ptr.us.xo.net)
16:27:51LloreanI think every set of bluetooth headphones I've seen intended for music use has been 2.0, at least
16:29:18 Part Llorean
16:30:48 Join pearldiver [0] (n=say@cpe-72-225-231-80.nyc.res.rr.com)
16:35:37 Join PaulJam [0] (n=pauljam@vpn-3148.gwdg.de)
16:35:52 Join n1s [0] (n=nils@h218n1fls35o293.telia.com)
16:38:30 Quit B4gder ("It is time to say MOOO")
16:41:57markunpreglow: my GSoC student also made a wiki: http://www.rockbox.org/twiki/bin/view/Main/TextToSpeech
16:41:58 Quit sando (Read error: 104 (Connection reset by peer))
16:44:10 Join chrisjs169|afk [0] (n=jack@c-68-33-114-53.hsd1.md.comcast.net)
16:45:02chrisjs169|afkThat seemed to do something....
16:45:10chrisjs169|afkWhat's Error SDCard: 1?
16:45:16GodEatermarkun: cool :)
16:46:11chrisjs169|afkSweet - I fixed my problem with 7134
16:47:10markunGodEater: comments and suggestions are welcome :)
16:47:16 Join sando [0] (n=lolsteam@210.11.80.141)
16:47:19alienbiker99rockbox is coded in c right
16:47:26markunyes, mostly
16:47:36markunthere is some assembly
16:48:00GodEatermarkun: think he/you have got all the bases covered - I can't think of anything else to suggest currently
16:48:48 Join Alonea [0] (n=chatzill@24-119-114-203.cpe.cableone.net)
16:49:26markunGodEater: I'm still interested in porting espeak as well
16:49:54markunfor that true 'bag of shit' sound which made rockbox so famous in the first place :)
16:50:09GodEaterhahahaha
16:50:10n1smarkun: it might be interesting to get it to read menues too, it seems only files/tags are planned...
16:50:12GodEatercan't wait
16:51:13markunn1s: yes, that way the 'voice file' would never be out of date
16:52:06n1smarkun: we might want to give users the option to use a voice file though...
16:52:24markunsure, I was not suggesting to remove support for it
16:52:54*n1s misread :)
16:53:14chrisjs169|afkjhMikeS: Looks like your away/idle/something, but the MicroSD patch works on my sansa (doesn't freeze) so the patch can be commited now
17:00
17:01:26markunn1s: maybe textviewer which is designed for ebooks would be nice, including tts support, chapter browsing etc
17:01:26 Quit sando (Read error: 104 (Connection reset by peer))
17:01:48 Join sando [0] (n=lolsteam@210.11.80.141)
17:01:50markun... a textviewer plugin ...
17:02:08n1smarkun: very
17:02:41 Part toffe82
17:07:37 Join lightmaster [0] (i=442f115f@gateway/web/cgi-irc/labb.contactor.se/x-3de587e368458b84)
17:07:49 Quit Alonea (Connection timed out)
17:08:37 Quit lightmaster (Client Quit)
17:13:55lostlogicsomeone tell jdgordon that he's my hero when he's around. I was hoping someone would pick up the colors patch and make it more useable.
17:14:30***Saving seen data "./dancer.seen"
17:15:31 Join WilliamBailey [0] (i=442f115f@gateway/web/cgi-irc/labb.contactor.se/x-4c32b79d868c1a1f)
17:16:33 Quit merbanan ("Leaving")
17:22:31 Quit sando ("Apply directly to the forehead")
17:27:40 Join XavierGr [0] (n=xavier@ppp37-226.adsl.forthnet.gr)
17:29:50WilliamBaileyi have been told that i have to be granted permission to contribute to rockbox and gain access to edit the pages
17:29:53 Join netmasta10bt [0] (n=torment@pool-71-251-102-2.tampfl.fios.verizon.net)
17:30:00markunWilliamBailey: one moment
17:30:33markunWilliamBailey: granted
17:30:59markuneh..
17:31:21WilliamBaileythanks
17:31:35 Join jgarvey [0] (n=jgarvey@cpe-066-057-231-236.nc.res.rr.com)
17:31:38markunshould work
17:35:50 Quit kubiix (Read error: 110 (Connection timed out))
17:41:36 Quit PaulJam (".")
17:42:00 Join Hammer89 [0] (n=soc_inte@static-host-24-149-229-197.patmedia.net)
17:42:58 Part Hammer89
17:44:34 Quit WilliamBailey ("CGI:IRC")
17:47:02 Quit XavierGr (Nick collision from services.)
17:47:06 Join XavierGr [0] (n=xavier@ppp227-179.adsl.forthnet.gr)
17:49:48markunI wonder if he's planning to actually update any of the wiki pages..
17:50:04 Join My_Sic [0] (n=MySic@mur31-1-82-237-204-133.fbx.proxad.net)
17:55:45 Quit XavierGr (Nick collision from services.)
17:55:48 Join XavierGr [0] (n=xavier@ppp6-88.adsl.forthnet.gr)
17:58:45 Join Nico_P [0] (n=nicolas@rob92-6-82-231-243-63.fbx.proxad.net)
18:00
18:00:36 Join Domonoky [0] (n=Domonoky@e180238030.adsl.alicedsl.de)
18:05:05 Quit Nibbier (Remote closed the connection)
18:05:14peturmarkun: now you know ;)
18:11:01 Join Arathis_ [0] (n=doerk@p5484BB5A.dip0.t-ipconnect.de)
18:16:33 Part chrisjs169|afk
18:16:45 Join chrisjs169|afk [0] (n=jack@c-68-33-114-53.hsd1.md.comcast.net)
18:17:54markunindeed
18:24:36 Quit XavierGr (Nick collision from services.)
18:24:40 Join XavierGr [0] (n=xavier@ppp6-88.adsl.forthnet.gr)
18:28:37 Quit Arathis (Read error: 113 (No route to host))
18:35:24 Join cruleworld [0] (i=4c100dd9@gateway/web/cgi-irc/labb.contactor.se/x-4ed8ca5a52f84a1e)
18:36:21cruleworldHello
18:36:35bluebrotherhi
18:37:50cruleworldCan anybody tell me why the MicroSD patch hasn't been acepted into the main build for the Sansa player?
18:38:18bluebrotherbecause it isn't finished yet?
18:38:33 Part Domonoky
18:38:43cruleworldit isn't?
18:38:47cruleworldHow so?
18:39:27desowinif it would work I'm almost sure it would get into trunk
18:39:37peturcruleworld: check today's log, there was some talk about it, it may be (almost) ready
18:39:38markuncruleworld: where's the patch?
18:39:57cruleworldIt works perfectly
18:40:02bluebrotherwell, if something hasn't been included into svn there are usually a couple of reasons: (1) it doesn't meet the code stanards, (2) it doesn't work properly or not on all players, (3) nobody had the time looking into it far enough (and fix possible problems)
18:40:23bluebrotheriirc that patch is still not working in conjunction with dircache.
18:40:36markuncruleworld: <chrisjs169|afk> jhMikeS: Looks like your away/idle/something, but the MicroSD patch works on my sansa (doesn't freeze) so the patch can be commited now
18:41:13cruleworldSo it's goign to be committed soon?
18:41:21markunprobably
18:41:29markunare you in a hurry? :)
18:42:07cruleworldCause thats what i am waiting for to have rockbox be my main firmware
18:42:24markunaha
18:42:31cruleworldi know there is no usb yet so not completly main but i will only to to OF to use the usb
18:44:26cruleworldusb support is supposed to be coming this summer right?
18:44:37peturwe hope
18:44:38 Quit obo ("KVIrc 3.2.6 Anomalies http://www.kvirc.net/")
18:44:56peturexpect more news 9 july
18:44:56bluebrotherlooking at the comments to that patch doesn't make me confident all problems are solved ...
18:45:19bluebrotherbut I haven't looked into it nor do I own a sansa, so ...
18:45:24cruleworldi have never had problems with it
18:45:36cruleworldand i own a sansa
18:46:25bluebrotherwell, something working on your player doesn't mean it's bugfree. That's the problem ...
18:46:47bluebrotherbut you can always compile yourself or use an unsupported build that includes the patch.,
18:48:24cruleworldi don't know how to cpmpile
18:48:33cruleworldi read the guide but i don't understand it
18:49:04jhMikeSchrisjs169: Ummm...say what? How does the same patch suddenly work? (and I didn't make any sacrifices to the gods today either).
18:49:35cruleworldhuh?
18:49:43bluebrotherwell, then you have two options: use an unsupported build that includes that patch or learn how to compile. You can always ask ;-)
18:49:52 Quit chris_ (Remote closed the connection)
18:49:53cruleworldhe's having problems with compiling though
18:50:06bluebrotherhe? Who?
18:50:10 Join chris_ [0] (n=chris@S0106001346f5afbd.cg.shawcable.net)
18:50:11cruleworldhow do i compile then?
18:50:33bluebrotherwell, in theory it's quite simple. Grab the vmware image and install vmware player.
18:50:53bluebrotherrun the image and navigate to the rockbox source folder in the image.
18:51:17bluebrotherupdate to the latest svn, patch, and then create a build folder
18:51:27cruleworldwait let me download these things and then you cna take me step by step
18:51:32bluebrotherrun ../tools/configure and "make" afterwards.
18:51:37*petur sees some progress from austriancoder :)
18:51:39 Join FOAD [0] (n=dok@dinah.blub.net)
18:51:56bluebrotheryou should check the SimpleGuideToCompiling wiki page.
18:52:10cruleworldi have
18:52:19bluebrother(and ignore the cygwin specific stuff, the vmware image has everything preinstalled)
18:52:36cruleworldit's really complicated for me cause i have a hard time reading long things cause i'm deslexic
18:54:42 Quit Rondom ("Ex-Chat")
18:55:35*jhMikeS reads the last comment in FS #7134 and goes "wtf?"
18:55:40bluebrothersorry, got to leave for like 15 minutes or so. Back shortly ...
18:56:14cruleworldok
18:57:28*petur connects cruleworld with jhMikeS
18:59:21*jhMikeS wonder's wtf again
18:59:55*jhMikeS apostrophised a verb :P
19:00
19:00:00markunjhMikeS: so it didn't work without a SDcard and after inserting one it now works even without a card?
19:00:40cruleworldi'm confused
19:01:39markuncruleworld: jhMikeS is the one who wrote the patch
19:01:40jhMikeSyes...and I want him to boot the OF and see if it reverts but he's away
19:01:50markuncruleworld: and the one who will commit it when he's done
19:01:57jhMikeSToni did...I just added some tweaks
19:02:56markunah, sorry
19:03:27cruleworldoh so why havben't you commited the patch yet?
19:03:55peturcruleworld: <jhMikeS>yes...and I want him to boot the OF and see if it reverts but he's away
19:04:08peturthat was for you
19:04:11markuncruleworld: more testing is needed
19:04:19cruleworldlike?
19:04:26cruleworldi'm not away
19:04:45 Quit Thundercloud (Remote closed the connection)
19:05:00 Join Thundercloud [0] (n=thunderc@82.152.218.72)
19:05:24peturmaybe I understood that incorrect - sorry
19:05:31markuncruleworld: chrisjs169 is away, not you
19:06:00*petur slaps forehead and hides before forehead logs in to slap back
19:06:25cruleworldwell i gtg
19:06:27 Quit cruleworld ("CGI:IRC")
19:06:53 Join forehead [0] (i=810dba04@gateway/web/cgi-irc/labb.contactor.se/x-f7a0aeb3c22c771a)
19:06:54 Quit miepchen^schlaf (Read error: 113 (No route to host))
19:07:00*forehead looks around
19:07:26markun:)
19:07:45*forehead spots petur and wonders what to do now ...
19:07:59peturha ha ha
19:08:15*forehead is tired and leaves ... this time.
19:08:23 Quit forehead (Client Quit)
19:08:40 Quit FOAD_ (Read error: 110 (Connection timed out))
19:09:58jhMikeSpetur: what your op about the triggered split? in SVN it doesn't pause but splits and keeps recording but it _feels_ like it should split and then pause. think I should change it?
19:10:12 Join obo [0] (n=obo@82-46-82-224.cable.ubr02.trow.blueyonder.co.uk)
19:10:46peturI have never used it, wouldn't know how it should work
19:10:57peturis that in the manual?
19:11:15bluebrotherwell, if you want to create marks by splitting the recording, wouldn't it be bad to pause?
19:11:26jhMikeSI looked but I don't think it is actually but I didn't check every manual :)
19:11:40 Quit XavierGr (Nick collision from services.)
19:11:43 Join XavierGr [0] (n=xavier@ppp41-125.adsl.forthnet.gr)
19:12:07peturchecking... but getting page not found - some new download server not synced yet?
19:12:31amiconnbluebrother: Btw, I think fixing dircache + hotswap is not a trivial task. And I don't think dircache makes sense on the sansa...
19:12:42peturgood to know the names of the individual servers :)
19:12:45 Quit Nico_P (Remote closed the connection)
19:12:53jhMikeSIt could be if that's the intent. Stop of course stops, pause just pauses, but split just keeps recording so it feels lopsided to not have split+pause
19:13:45amiconnSplit should just split and keep recording. That's what split is for - splitting a recording into several tracks
19:13:47jhMikeSamiconn: I agree, dircache is pointless since there's no delay for access
19:14:21petursplit has two options: start a new file or stop recording (according to the manual)
19:14:31amiconnIf you want to end a track and pause the recording at the same time, just use stop, then start later on
19:14:32***Saving seen data "./dancer.seen"
19:14:36bluebrothercan dircache easily get removed from the sansa build?
19:15:08peturbah... "Add description of triggered recording."
19:15:09jhMikeSisn't there a HAVE_DIRCACHE or something?
19:15:24 Nick Arathis_ is now known as Arathis (n=doerk@p5484BB5A.dip0.t-ipconnect.de)
19:15:52markunyes, there is
19:15:58 Join wesmo [0] (i=cb625b9e@gateway/web/cgi-irc/labb.contactor.se/x-f1299c604cefea4f)
19:15:59markuncurrently in config.h
19:16:03amiconnbluebrother: Shouldn't be too difficult, although the definition isn't in the config-*.h file
19:16:19markunbut maybe it should just be defined in every config-*.h file?
19:16:34markun(where it applies)
19:16:34wesmohi anychance i could get editing permission for th wiki???
19:16:37jhMikeS#ifndef DONT_HAVE_TAGCACHE|#define HAVE_TAGCACHE #endif ?? :P
19:16:41amiconnBut removing dircache from a build has a drawback - it cuts some features from the database
19:16:49bluebrotherwesmo: only if you give us your wiki name ;-)
19:16:50amiconn(like being able to commit w/o reboot)
19:16:58wesmoWessamBaghdadi
19:17:00markunjhMikeS: you want to disable tagcache as well?
19:17:14jhMikeSwoops ... no.... type
19:17:17wesmothnaks btw who do i speak to about the creative port
19:17:17jhMikeStypo
19:17:18markun:)
19:17:27amiconnNot tagcache, but maybe tc_ramcache
19:17:48markunwesmo: talk to jhulst about the creative port
19:17:53amiconnNot sure whether there is an advantage of having tc_ramcache without dircache
19:18:22amiconnThe proper solution for dircache would be to limit it to the inbuilt media (non-hotswap volume)
19:18:24bluebrotherwesmo: done.
19:18:44jhMikeSwhat's the point in any feature designed to deal with disk spinups when there are none on a flash device?
19:18:46markunamiconn: even if it's flash?
19:18:50amiconnThis will be needed with otg, or if that elio port will ever take off
19:19:08amiconn(the elio is hdd based _and_ has an sd slot)
19:19:11 Join maffe [0] (n=maffe@barmen.interhost.no)
19:20:27amiconnmarkun: Well, I'm not sure why these features are connected... Slasheri should know
19:22:51markunmaybe just to combine the scanning
19:23:04wesmodoesnt look like he is responding
19:23:23markunwesmo: what did you want to talk about?
19:23:34wesmohave any steps been taken with the port to zen touch/micro/xtra
19:24:02markunwesmo: I think only what is discussed in the forum thread
19:24:40wesmoon epizenter?
19:24:43bluebrotherwesmo: btw, you can't send PM's on freenode unless you are identified at nickserv
19:25:00markunwesmo: http://forums.rockbox.org/index.php?topic=3320.0
19:25:08wesmooh k thnks ill set tht up now
19:30:12 Join Zeze21 [0] (n=usamah@catv54031C19.pool.t-online.hu)
19:30:25Zeze21hey
19:30:49Zeze21what happened to the ipodbootloader?
19:31:17jhMikeStagcache doesn't seem to exclude any commits if dircache doesn't exist :\ It just steals the dircache buffer.
19:31:34 Quit Thundercloud (Remote closed the connection)
19:31:44wesmothere's no prob with me editin the creative zen touch port page
19:31:49wesmo???
19:32:03markunwesmo: no, go ahead
19:32:33Zeze21could someone tell me where i could find the ipod bootloader the url for it does not work
19:33:06Zeze21please :)
19:33:39markunZeze21: sorry, don't know
19:34:02Zeze21hm well thank you is there a way to install rockbox without the loader?
19:34:26markunyes,with rbutil
19:34:41 Join Thundercloud [0] (n=thunderc@82.152.198.174)
19:34:42Zeze21ok where could i get that?
19:34:44peturZeze21: what link is not working?
19:34:47Zeze21http://download.rockbox.org/bootloader/ipod/ipodpatcher/win32/ipodpatcher.exe
19:34:50Zeze21this one
19:34:54markunhttp://www.rockbox.org/twiki/bin/view/Main/RockboxUtility
19:35:13peturZeze21: try http://haxx.rockbox.org/bootloader/ipod/ipodpatcher/win32/ipodpatcher.exe
19:35:26peturBagder?
19:35:59Zeze21thank you both
19:36:04Zeze21have a nice day
19:36:09 Quit Zeze21 ()
19:37:07peturthere seems to be a problem with the download server distribution it seems, rbutil might actually fail when downloading
19:40:45peturBagder: the TBRN server is not ready for use as download server
19:41:53 Quit desowin ("use linux")
19:42:18jhMikeSpetur: I take it no more lost sample reports have come up? (though I expect I would have been notified if so)
19:42:37peturno, everything's fine, thanks
19:42:48 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
19:44:08jhMikeSI know you weren't having trouble but I was wondering about on tapersection.com and whatnot
19:45:03peturno, I keep a close eye on taperssection and our own recording subforum ;)
19:46:07 Quit wesmo ("CGI:IRC (Ping timeout)")
19:46:24peturand many seem to have upgraded to current versions (they were still using august 2006 builds but I think they still have them in case something would come up)
19:47:31SlasherijhMikeS: it can steal the tagcache ramcache buffer as well, but that might not be large enough
19:48:07 Join gletob [0] (i=4c03782c@gateway/web/cgi-irc/labb.contactor.se/x-f6b8c2316ebbfbcc)
19:48:21 Join bluey- [0] (n=bluey@dslb-088-074-023-091.pools.arcor-ip.net)
19:48:57gletobcan anyone tell me if rockbox will work with this http://www.walmart.com/catalog/product.do?product_id=4910672
19:49:50 Join mpeccorini [0] (n=carchila@66.249.198.18)
19:50:51 Part mpeccorini
19:50:51 Quit gletob (Client Quit)
19:53:32 Join The-Compiler [0] (n=florian@46.19.76.83.cust.bluewin.ch)
19:54:44jhMikeSSlasheri: but commit requires reboot without dircache? ??
19:55:11 Join SaLoMoN_ [0] (n=SaLo@xglusers.de)
19:56:51 Quit SaLoMoN (Read error: 110 (Connection timed out))
19:58:26 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
20:00
20:04:27SlasherijhMikeS: nope, if loaded ramcache is big enough
20:05:49 Quit GodEater (Read error: 110 (Connection timed out))
20:05:54jhMikeSok. one thing I wondered is if tags are stored packed in RAM or field width (I just don't know)
20:06:31SlasherijhMikeS: the same format is used as in tagcache files
20:06:53Slasheriso first comes a header telling the tag length, and then the tag
20:08:37Slasheripractically tags are loaded to ram by almost copying the files to ram
20:09:00Slasherihowever, the tag containing filenames has a special handling and is never loaded to ram
20:09:40Slasherithat's because then dircache already contains filenames and that's used instead
20:11:03jhMikeSso why does amiconn say some functionality is lost if dircache is not built? It's just an issue of available buffer size and it just potentially has less mem available?
20:11:03 Quit netmasta10bt (Read error: 104 (Connection reset by peer))
20:12:18SlasherijhMikeS: no functionality should be lost if autoupdate is enabled for flash targets after disabling dircache (currently that's just #if 0 in tagcache.c)
20:12:39Slasherior autoupdating deleted files
20:14:01amiconnI thought at least auto-updating deleted files won't be possible w/o dircache
20:14:32amiconnAnd regarding the commit, I didn't actually try it. I just noticed that the db commits right away if dircache is enabled
20:14:36Slasheriit's possible but currently disabled in code because it makes no sense for hdd targets (would be very slow)
20:14:49jhMikeSthe _only_ #if 0 around tagcache_modify()?
20:15:06SlasherijhMikeS: line 3513
20:15:20Slasherinot the only one :)
20:15:53Slasheribut i am worried how much operating tagcache on flash targets wears the flash memory..
20:15:56jhMikeSI searched for "#if 0" not "# if 0" :)
20:16:03Slasheri:)
20:17:37jhMikeSwear leveling should minimize it ... recording should be no better for it
20:18:59amiconnSlasheri: Btw, do you know what causes the nasty behaviour of dircache in conjunction with recording?
20:19:19Slasheriamiconn: hmm, i haven't tried that yet..
20:20:15Slasheriamiconn: what weird happened with that (can't remember exactly)
20:21:10amiconnWell, if you enter the recording screen while dircache is still scanning in the background, the dircache scan becomes extremely slow
20:21:49amiconnI would understand that happening which prerecording enabled and a demanding encoder, but it even happens with no prerecording and recording format set to wav
20:21:57amiconns/which/with/
20:22:16Slasheribtw, some Enrico wrote me about tagcache generating broken files on sansa.. I analyzed the files and verified there were byte shifts (missing bytes in the file). I have never seen tagcache behaving that way before. I strongly doubt there might be a bug in the sansa FS low level code
20:22:39amiconnjhMikeS: Do you know why recording only starts after spinning up the disk on targets with no rtc?
20:23:20amiconnSlasheri: If this behaviour is only observed on the sansa, I suspect the sd driver
20:23:51Slasheriamiconn: at least i haven't heard anybody else reporting that
20:24:11jhMikeSamiconn: because it has to hit the disk to get the first filename
20:24:12 Join freedomlinux2 [0] (n=Judy_Wen@pool-71-175-186-240.phlapa.east.verizon.net)
20:25:11amiconnjhMikeS: Why does it need the filename before even starting recording?
20:25:30amiconnIt should only hit the disk when the first flush becomes necessary
20:25:51Slasheriamiconn: hmm, interesting. it sounds like recording screen eats much cpu then.. and recording is prioritized over dircache
20:26:13jhMikeSit obtains a filename through create_numbered_filname which is the first file of the same form with the highest index number
20:26:15 Part freedomlinux2
20:26:49Slasheriamiconn: could be the peak meters for example
20:27:01amiconnjhMikeS: Yes I know... but why would this be needed at the very start of the recording?
20:27:06 Join PaulJam [0] (i=Paul@vpn-3074.gwdg.de)
20:27:36 Join webguest60 [0] (i=46a77451@gateway/web/cgi-irc/labb.contactor.se/x-af737bd63160cdcf)
20:27:49jhMikeSit's not...if you mean upon entering the screen and not actually hitting record, then this is new behavior
20:28:05jhMikeSother than loading the encoder, but all should do that
20:28:08amiconnThis is annoying for 2 reasons: (1) I expect the recording to start immediately when pressing start. I missed the proper start due to that several times. (2) Spinning up the disk at the beginning of the recording is bad for built-in mic recording
20:28:34amiconnjhMikeS: The delay happens when I press start
20:28:57amiconnThe encoder should be loaded already, I would think
20:29:07jhMikeSIt always shows the file that will be written. it always did and I never changed that aspect of it.
20:29:16 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
20:29:19amiconn(loaded when entering the recording screen)
20:29:28 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
20:29:34jhMikeSit is...perhaps that's also a good time to get the first filename to be used
20:29:43amiconnImo there is no need to show the filename until it is known.. just show nothing meanwhile
20:29:44SlasherijhMikeS: maybe it could check the filename immediately when entering recording screen?
20:29:58Slasherior even on bootup
20:30:13amiconnYes, getting the filename right away when entering the recording screen would be another option
20:30:30amiconnOn bootup might cause problems
20:30:40jhMikeSwhen loading the encoder, also hit the disk for the first filename. when stopping and flushing, get the next filename for the next start
20:30:47webguest60hello all......
20:30:54amiconn(like deleting/adding files in usb mode, or even deleting from within rockbox)
20:30:55webguest60hello all......
20:30:57DBUGEnqueued KICK webguest60
20:30:57webguest60hello all......
20:31:08webguest60hello all......
20:31:26markunwebguest60: please..
20:31:30jhMikeShello, hello, hello hello
20:31:44webguest60I was hoping someone might be able to help me out a little.. installing cygwin (i've done it before, but this time, running into an issue .. probably easy)
20:31:48webguest60whoa.
20:31:49 Quit petur ("work->home")
20:31:52Slasherihmm, true and encoder needs to be loaded anyway.. so it sounds good to do that when entering to the screen
20:31:57webguest60sorry
20:32:11markunwebguest60: what issue are you running into?
20:32:15webguest60I didn't mean to sent that msg so much... must be this webclient
20:33:00jhMikeSthat's really just a function of the screen, not the recording system. it just writes to whatever path it's told
20:33:02amiconnSlasheri: On hwcodec, no encoder needs to be loaded... but then this is only for non-rtc target
20:33:06webguest60so.. um.. ya, my responses are lagging pretty hard here... =\ maybe I'll go get mIRC
20:33:28webguest60 I was just installing cygwin and such, again.. and I was having trouble, last time I used the walkthrough with screenshots, not a problem.. now, though, I wasn't able to dload rockbox. the second setup portion.. when I put in http:/download.rockbox.com in the user URL box, and try to dload it says that rockbox.com doesnt' have the 'setup.ini' file
20:33:44amiconn..and the only non-rtc hwcodec target which can record is the Ondio FM, where it doesn't matter when the filename is checked
20:33:45jhMikeSyeah, for RTC targets, it only needs to get the current time, so no hit
20:34:30pixelmaisn't checking for filename when entering the recording screen the suggestion for the recording directory patch, anyways?
20:34:36amiconnSo that sounds like a plan: on non-rtc targets the recording screen should determine the first filename on entering
20:34:44jhMikeSthis assumes your clock is set correctly of course ... so no guarantees :)
20:35:03amiconnjhMikeS: Hmm, btw that reminds me:
20:35:30amiconnIirc you added a function to delay newfile creation on rtc targets in order to make the names unique
20:35:31jhMikeSI'd just make them both do it that way anyhow
20:35:35amiconn(same-second problem)
20:35:37jhMikeSyes
20:35:45amiconnBut what happens when the rtc doesn't tick?
20:35:59amiconn(i.e. when it's not yet set after a battery change?)
20:37:00jhMikeSwell, I guess you've had it then ... what should the filename obtained in get_datetime_filename be then?
20:37:38 Quit webguest60 ("CGI:IRC")
20:37:49 Join BlahHumbug [0] (n=blahhumb@wsip-70-167-116-81.sd.sd.cox.net)
20:37:52BlahHumbugaha!
20:38:24 Join webguest60 [0] (i=46a77451@gateway/web/cgi-irc/labb.contactor.se/x-0f8f8ee62ac6e560)
20:38:31amiconnjhMikeS: I think it will hang forever with current svn if you start a new recording... maybe it could fall back to numbered filenames?
20:40:23BlahHumbughello all... (only once this time) =)
20:40:24jhMikeSSure...can it just check if the clock is set
20:40:24 Quit webguest60 (Client Quit)
20:41:00amiconnNot sure how much complexity this would add, as the strategy for obtaining filenames is different
20:41:30amiconnH,, it could just check if the clock is set and if it's not, just splash that the clock needs to be set first
20:41:39amiconns/H,/Hmm/
20:41:47 Join Siltaar [0] (n=Siltaar@reverse-52.fdn.fr)
20:42:31jhMikeSyes if that's not too annoying ... or rockbox could ensure the clock is ticking at startup (set it to 00:00 hrs but this risks overwriting files)
20:42:59jhMikeSa __tiny__ risk though
20:43:03BlahHumbugSo, I'm hoping someone might be able to help me out. I was just trying to compile a build with a few patches that i'd like to see, I haven't in a while, and I was just installing cygwin again. but on the second portion of 'setup' where you dload the rockbox specifics, it says that rockbox.com doesn't have the 'setup.ini' file.
20:43:30amiconnThere is no rockbox.com
20:43:38amiconnIt's always been rockbox.org
20:43:50BlahHumbugthat's what I meant.
20:43:51BlahHumbugsorry.
20:43:52BlahHumbug=0
20:44:53jhMikeSand what's this about dircache updates slowing to a crawl? they should during a disk write and with other higher priority thread activity since dircache is background priority
20:45:05BlahHumbug"Unable to get setup.ini from <http://rockbox.org>
20:45:07BlahHumbug"
20:45:44jhMikeSbefore, the priority inversion would dominate the pcmrec thread trying to flush data
20:45:47amiconnjhMikeS: I'm not sure. Just entering the recording screen triggers this behaviour
20:46:10amiconnNot even prerecording needs to be running
20:50:47 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
20:51:45jhMikeShmmm...then the codec is doing anything but yielding the normal way
20:51:55bluebrotherBlahHumbug: well, as a different approach: how about using the vmware image?
20:57:54BlahHumbugoh... I'm not familiar with that..
20:58:03BlahHumbugI'll go through the wiki etc..
20:59:12BlahHumbug" VMware Development Platform for Rockbox?
20:59:16bluebrotherwell, it's basically similar, but it's a linux running in a virtual machine
20:59:29BlahHumbugscore... I'll go through, try this method.. thanks.
20:59:30BlahHumbug=)
20:59:33bluebrotherso the image can get redistributed with everything installed.
20:59:40bluebrotheryes, that platform :)
21:00
21:00:00bluebrotherlinux is different to cygwin but it's similar. IMO it's easier to setup
21:00:29BlahHumbugok, thanks much,.. i'll take a swing at ti
21:00:32BlahHumbug*it
21:00:55 Quit BlahHumbug ()
21:01:30 Quit Thundercloud (Remote closed the connection)
21:04:55 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
21:05:17 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
21:09:54 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
21:10:16 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
21:11:41 Join webguest17 [0] (i=5b84f61a@gateway/web/cgi-irc/labb.contactor.se/x-5cccbafd9b94ecea)
21:11:53webguest17hello there
21:11:53 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
21:12:14 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
21:12:26webguest17i need help with my iriver h10 20gb
21:12:57 Quit Noah0504 ("Leaving")
21:13:45webguest17um... hello?
21:13:54FebsJust ask your question.
21:13:57GodEater_just ask your question
21:14:00webguest17oh, sorry
21:14:01CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:14:01*GodEater_ high fives Febs
21:14:15 Join cruleworld [0] (i=4c100dd9@gateway/web/cgi-irc/labb.contactor.se/x-61c917c265370b59)
21:14:17webguest17i just think my iriver needs battery replacement
21:14:26webguest17but i don't know how to replace it
21:14:34***Saving seen data "./dancer.seen"
21:14:41webguest17and where to find a battery to replace it with
21:15:07cruleworldprobaly can't
21:15:21webguest17that's absurd
21:15:34cruleworldwhat?
21:15:42webguest17what you just throw your iriver away after 2 years?
21:15:56 Join [sellout] [0] (n=sellout@016-000-039.area7.spcsdns.net)
21:16:01cruleworldno i don't have one
21:16:05 Quit RhinoBanga (Read error: 110 (Connection timed out))
21:16:12cruleworldi have a player that i can change it myself easily
21:16:37webguest17that's convenient
21:16:53cruleworldyep Sansa e250 is what i got
21:17:06cruleworldif you ever get a sansa get the e280
21:17:08webguest17i am really attached to my player and it only needs battery replacement. otherwise it's ok
21:17:10 Join RhinoBanga [0] (n=rhinoban@host-87-74-81-49.bulldogdsl.com)
21:17:12 Join low_light [0] (i=c730180b@gateway/web/cgi-irc/ircatwork.com/x-237342cf394a75d4)
21:17:28cruleworldoh i see
21:17:32webguest17i don't think i'm getting a new one
21:17:39webguest17not for now at least
21:18:33cruleworldthat's fine
21:18:39webguest17really, how does someone replace the battery in ipods? it must be harder than with iriver..
21:18:47pixelmawebguest17: tried searching in the misticriver forums? (don't have an iriver, so can't help any further)
21:18:52Febswebguest17, I would look on Mistic River.
21:19:03*Febs high-fives pixelma.
21:20:15bluebrothermisticriver has a wiki entry about replacing the battery for h100 / h300
21:20:45 Quit cruleworld ("CGI:IRC (EOF)")
21:21:49webguest17that's what i'll do thanks
21:25:19 Join webguest03 [0] (i=425cb8cb@gateway/web/cgi-irc/labb.contactor.se/x-bedf6402cc05e4ac)
21:25:25 Quit webguest03 (Client Quit)
21:25:35 Join curious [0] (i=425cb8cb@gateway/web/cgi-irc/labb.contactor.se/x-58e53bf07076ec2f)
21:26:58 Join chewbacca [0] (n=chatzill@dsl092-184-203.sfo1.dsl.speakeasy.net)
21:27:21 Quit webguest17 ("CGI:IRC")
21:28:52chewbaccaHello. Does anyone know if all of the pdf/online manuals are supposed to be available (or if 404 is the intended result) ?
21:29:04chewbaccafrom http://www.rockbox.org/manual.shtml
21:29:45chrisjs169|afkthey work for me
21:30:14chewbaccahttp://download.rockbox.org/manual/rockbox-player.pdf is a 404 for me (for example)
21:30:24 Join PaulJam_ [0] (i=Paul@vpn-3074.gwdg.de)
21:30:49chrisjs169|afkdisplays for me
21:31:10chrisjs169|afkwhich one are you looking for? I'll temp upload it to another site
21:31:13low_lightit's a problem with the new download mirrors...try http://videolan.rockbox.org/manual/
21:31:27chewbaccai will try the videolan site.
21:31:55 Quit bluey- ("Leaving")
21:32:00chewbaccahttp://videolan.rockbox.org/manual/rockbox-ipod3g/ looks promising! Thank you lowlight
21:32:04 Quit Febs (Read error: 104 (Connection reset by peer))
21:32:32chewbaccachrisjs... i'm just looking for "how involved is it" so anyone would do (i do have two ipods here for experimentation)
21:34:26 Join mirak [0] (n=mirak@m60.net81-64-221.noos.fr)
21:37:04 Quit curious ("CGI:IRC (EOF)")
21:41:18 Quit RhinoBanga (Read error: 110 (Connection timed out))
21:41:32 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
21:41:37 Join RhinoBanga [0] (n=rhinoban@host-87-74-81-65.bulldogdsl.com)
21:41:53 Join miepchen^schlaf [0] (n=hihi@dslb-088-072-214-247.pools.arcor-ip.net)
21:42:05 Join petur [0] (n=petur@rockbox/developer/petur)
21:47:00 Quit PaulJam (Read error: 110 (Connection timed out))
21:50:02 Join entheh [0] (n=purr@88-106-242-42.dynamic.dsl.as9105.com)
21:51:38 Join creative [0] (i=59f0e7df@gateway/web/cgi-irc/ircatwork.com/x-810b19779ce078eb)
21:53:20creativeHi. I just came across Rockbox recently and it sounds brilliant! Is there any chance of developing a firmware for the Creative players (particularly the Zen Vision:M)?
21:53:45bluebrothercheck the New Ports forums
21:54:04bluebrotheriirc there is a thread about that player.
21:54:22creativeThanks for the advice bluebrother
21:58:20 Quit [sellout] (Read error: 113 (No route to host))
22:00
22:00:02 Quit chewbacca ("ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502]")
22:00:38 Join RhinoBanga2 [0] (n=rhinoban@host-87-74-81-77.bulldogdsl.com)
22:01:35 Quit RhinoBanga (Read error: 110 (Connection timed out))
22:08:30 Quit creative ("CGI:IRC (Ping timeout)")
22:09:33 Quit ompaul (Client Quit)
22:11:56 Quit low_light ("CGI:IRC")
22:12:34 Join DerPapst [0] (n=DerPapst@pD9EB3C3A.dip0.t-ipconnect.de)
22:12:57DerPapstgood evening :)
22:15:13DerPapstlol. i though voting for non devs was disabled on FS.
22:15:35 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net)
22:15:36*DerPapst accidently voted for a task
22:15:52peturheh, I did too
22:16:43 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
22:16:43 Part TrueJournals
22:17:14DerPapstluckily it wan't one of my own tasks ;)
22:23:04 Join RhinoBanga [0] (n=rhinoban@host-87-74-81-90.bulldogdsl.com)
22:24:06 Quit RhinoBanga2 (Read error: 110 (Connection timed out))
22:25:45bluebrotherit's not voting. Setting the severity is disabled for non devs ;-)
22:26:17DerPapstah
22:26:20DerPapst:)
22:36:12GodEater_linuxstb: http://www.gossamer-threads.com/lists/mythtv/users/275682
22:36:22GodEater_(when/if you get round to reading the logs!)
22:39:44 Join webguest55 [0] (i=548c9069@gateway/web/cgi-irc/labb.contactor.se/x-27efe7918acef688)
22:39:49DerPapstteh ps3 haxx0rz
22:40:17webguest55could anybody help me? I just upgraded my rockbox after about 4 months and now I get a Data error whenever I try to play a song
22:40:21webguest55pls =)
22:40:27Bagderipod?
22:40:32webguest55yes
22:40:33webguest55^^
22:40:37DerPapsti bet you have to update the bootloader
22:40:40Bagderthen upgrade bootloader
22:40:45webguest55ok
22:44:18webguest55the pre-compiled build has been last updated august 2006 oO
22:44:47peturBagder: did somebody already tell you that something is borked with download.rockbox.org?
22:45:07Bagderas in right now you mean?
22:45:16GodEater_webguest55: what on earth are you talking about ?
22:45:23petursince at least this afternoon
22:45:29GodEater_Bagder: yes - I'm seeing it too - lots of 404 errors
22:46:01Bagderon the root or specific pages/files ?
22:46:06peturit fails if the tbrn server is selected
22:46:50Bagderwhat if you use tbrntech.rockbox.org instead of download.rockbox.org, is that too a bad url?
22:47:12 Join PaulJam [0] (i=Paul@vpn-3057.gwdg.de)
22:47:15 Join RhinoBanga2 [0] (n=rhinoban@host-87-74-81-114.bulldogdsl.com)
22:47:21peturduh... now it is working
22:47:36BagderI bet you got another server
22:47:41petur2 minutes ago it was serviong an error page
22:47:52peturserving even
22:48:14peturthe errormessage was from tbrn
22:48:15 Quit RhinoBanga (Read error: 110 (Connection timed out))
22:48:31Bagderyes, I can see that tbrn has wrong server for download.*
22:48:31peturah well, it's working now
22:48:39Bagdercurl -H "Host: download.rockbox.org" tbrntech.rockbox.org
22:48:55Bagder... is not our downloads...
22:49:06 Join loswillios [0] (n=hbt@unaffiliated/loswillios)
22:49:14loswillioshi
22:49:29loswilliossome links on rockbox.org are b0rked
22:49:36peturhahaha
22:49:39GodEater_it's amazing...
22:49:51GodEater_the timing was just perfect
22:49:52loswilliosie the docs
22:50:01 Join miepchen^schlaf [0] (n=hihi@p54BF6A57.dip.t-dialin.net)
22:50:02loswillioswhat? did I miss something ;-)
22:50:05GodEater_loswillios: we're aware, and dealing :)
22:50:14GodEater_you turned up in the middle of our discussion
22:50:22*DerPapst wants a cookie too
22:50:26loswilliosheh
22:50:35GodEater_well - by "we" I mean "Bagder" ;)
22:50:40*loswillios is about to free his ipod
22:50:51loswilliosbut not without docs :P
22:50:52GodEater_welcome to the fold :)
22:51:08GodEater_loswillios: what OS you run ?
22:51:08 Quit webguest55 ("CGI:IRC (EOF)")
22:51:15Bagdertry (for example) videolan.rockbox.org instead of download.rockbox.org for now
22:51:27loswilliosGodEater_: winXP
22:51:42GodEater_loswillios: try out RockboxUtility
22:51:47DerPapstheh.. the test codec plugin just wrote a 240MB file :-D
22:51:51GodEater_there's not much you need to read to use it
22:52:02GodEater_DerPapst: good god
22:52:21peturlol... god, pope, ...
22:52:25DerPapstit took even pretty long in the sim
22:52:36DerPapsthahaha
22:52:39GodEater_hehehe
22:52:51Mode"#rockbox +o Bagder " by ChanServ (ChanServ@services.)
22:52:53GodEater_we just need an ayatollah or something now
22:52:56DerPapstuh ohh
22:53:08*DerPapst hides
22:53:08loswilliosGodEater_: where can I get it? I can't find rockboxutility on the site
22:53:09Topic"if download.rockbox.org acts funny, use videolan.rockbox.org instead until things settle" by Bagder (n=daniel@rockbox/developer/bagder)
22:53:09GodEater_topic change I'm guessing
22:53:28GodEater_http://www.rockbox.org/twiki/bin/view/Main/RockboxUtility
22:53:36Mode"#rockbox -o Bagder " by Bagder (n=daniel@rockbox/developer/bagder)
22:53:49*DerPapst sets mode +o on DerPapst
22:53:50GodEater_Bagder: is positive.rockbox.org broken still ?
22:53:57DerPapstmeh.. not working -.-
22:54:10Bagderno, it should be working fine
22:54:25GodEater_ah cool
22:54:25 Join webguest55 [0] (i=548c9069@gateway/web/cgi-irc/labb.contactor.se/x-475591a28b3caa9a)
22:55:11Bagderhm, or perhaps not
22:55:24 Join Domonoky [0] (n=Domonoky@e180238030.adsl.alicedsl.de)
22:55:25*Bagder composes another mail...
22:56:17GodEater_looks ok to me =.
22:56:27Bagderit doesn't seem to sync
22:56:31GodEater_ah
22:56:33Bagdersee mpeg/
22:56:37GodEater_k
22:56:59GodEater_thoroughly empty =/
22:57:07*petur finds the saved devcon webcam pics :)
22:58:14loswilliosGodEater_: thx
22:58:24webguest55and ive got another weird problem... a problem which ive had since I first installed rockbox ... while playback is active the screen sometimes freezes (but the music doesnt stop) and the only possibility to unfreez it is to restart the ipod... how can fix that? =(
22:58:29peturhttp://videolan.rockbox.org/devcon2007/saved/20070520-134025.jpg
22:58:41loswilliosdamn you itunes. keep disconnecting my ipod after sync :/
22:59:24DerPapstmonty python wasn't it?
22:59:30peturyes
22:59:37DerPapsthehe
22:59:48 Join webguest30 [0] (i=480ec220@gateway/web/cgi-irc/labb.contactor.se/x-d531f7e398c5208b)
23:00
23:00:18scorchepetur: red > blue chimay...agree?
23:00:27peturhttp://videolan.rockbox.org/devcon2007/saved/20070519-092611.jpg
23:00:32webguest55nobody? ^^
23:00:55loswilliosGodEater_: heh. rockboxutility woun't work b/c of the b0rked links
23:01:11loswilliosah i can change them
23:01:12scorchewebguest55: is this using the latest build?
23:01:24webguest55yep
23:01:53webguest55and it has been for any other build I used until today
23:02:45webguest30Hi, um... Rockbox has been running slowly on my h10 reccently. While playing music, Bubble plugin is sluggish. I am using the latest build. Is there a problem with my player, or is rockbox suddenly sounding like a letdown?
23:03:11scorchewebguest30: on that device, plugins have always been sluggish while playing music
23:03:19pixelmawebguest55: do you have an 80GB 5.5g Ipod?
23:03:41webguest55nope... 5g 30gb
23:04:09 Quit The-Compiler (Remote closed the connection)
23:04:32 Quit webguest30 (Client Quit)
23:04:34 Quit PaulJam_ (Connection timed out)
23:05:54Bagderok, server fixed
23:11:11 Join kubiix [0] (n=Miranda@mos-81-27-201-28.karneval.cz)
23:14:35***Saving seen data "./dancer.seen"
23:18:20 Join midgey [0] (n=tjross@c-71-205-31-207.hsd1.mi.comcast.net)
23:18:32 Part loswillios
23:20:17 Quit webguest55 ("CGI:IRC (EOF)")
23:21:55 Part chrisjs169|afk
23:24:56 Join saratoga [0] (i=980398fe@gateway/web/cgi-irc/labb.contactor.se/x-2ca6dfd62a39f25f)
23:25:12 Join RhinoBanga [0] (n=rhinoban@host-87-74-81-150.bulldogdsl.com)
23:26:15 Quit RhinoBanga2 (Read error: 110 (Connection timed out))
23:28:00 Quit Domonoky (Read error: 110 (Connection timed out))
23:34:20PaulJamis there a trick to use < and > in the forum?
23:34:45peturuse the html escape codes?
23:37:55PaulJamnm, i just used different symbols.
23:40:52 Join Lars_G [0] (n=lars@unaffiliated/lars-g/x-000001)
23:41:08Lars_GSlight offtopic question for big brains, any idea why my pod is not showing up in dcop kded mediamanager fullList ??
23:45:10 Quit entheh ("^~")
23:48:00Lars_Gany tips please? :(
23:48:57*DerPapst 's brain isn't big enough
23:49:25 Quit jhMikeS (Nick collision from services.)
23:49:29 Join Alonea [0] (n=chatzill@24-119-114-203.cpe.cableone.net)
23:49:31 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
23:50:14 Join FOAD_ [0] (n=dok@dinah.blub.net)
23:50:53 Quit printfXh4 (Read error: 113 (No route to host))
23:51:15 Quit davina_ ("xchat on Ubuntu 7.04")
23:52:27 Quit midgey ()
23:54:14 Quit PaulJam (".")
23:56:10 Quit idnar (Nick collision from services.)
23:56:12 Join idnar_ [0] (n=mithrand@unaffiliated/idnar)
23:58:38 Join hannesd_ [0] (n=light@gate-hannes-tdsl.imos.net)

Previous day | Next day