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 2007-10-11

00:00:00 Quit outside (Read error: 110 (Connection timed out))
00:00:21bgmrkwhat are the buttons u hold down to still use rockbox with it plugged into ur comp?
00:00:53zajacattacki think you hold down select while plugging it in
00:01:13bgmrki'll give it a try
00:01:14 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
00:01:16PaulJam__could also be menu
00:01:42bgmrkwasn't select
00:01:55zajacattackcan someone answer my original question?
00:03:15bgmrkmenu it was..ty
00:03:39Lloreanzajacattack: You should generally ask what player they're using, first.
00:03:53 Quit matsl (Remote closed the connection)
00:03:56zajacattackllorean: why?
00:04:07LloreanBecause it's different buttons on different players?
00:04:24LloreanIt's menu on some, select on others.
00:04:34LloreanAnd may even be something else on the H300, I'm not sure.
00:04:47pixelma'mode' on some others
00:04:56zajacattackoh, ok
00:05:19zajacattackbut can someone tell me how to fix the "bulk write error" in e200tool?
00:05:24PaulJam__rec on h300, but i think the plans are to use the hold button
00:06:20 Quit lee-qid ("aufwiederbyebientotsayonara")
00:06:23LloreanPaulJam__: I'd use A-B for that one.
00:06:46LloreanIt shouldn't be something you can do accidentally.
00:06:48PaulJam__Llorean: but a-b brings up the quickmenu
00:06:53LloreanPaulJam__: Yes, and?
00:07:01LloreanThe quickmenu doesn't do anything on its own without further button presses.
00:07:22LloreanHolding "Menu" on iPods brings up the quickmenu, and it serves the "charge instead of connecting" function when held on iPod.
00:07:31 Join Toadinator [0] (i=4cc58e66@gateway/web/cgi-irc/
00:08:19 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
00:09:02 Quit davina (Remote closed the connection)
00:09:16zajacattack"can someone tell me how to fix the "bulk write error" in e200tool?" still waiting
00:09:54Lloreanzajacattack: Please, quit repeating the question.
00:10:09LloreanTwo people have entered the room since you last asked it.
00:10:11zajacattackwell, no one is responding
00:10:26Toadinatoryeah, someone know the answer?
00:10:29LloreanYes, and asking the question again doesn't magically make people who didn't know before, know now.
00:10:52zajacattackwell, no one said anything, so i thought it was just missed by everyone
00:11:03LloreanDo you want all 141 other people to respond "No"?
00:11:27Toadinatorof course he doesn't/
00:11:36Kleviif the chat was jumping off thew wall with replies to toher things, then we mightve, but things are pretty slow at the moment.
00:11:44Klevi*to other
00:12:22ToadinatorDoes anyone know how far the iPod Nano 2G port is in development?
00:12:28LloreanToadinator: It's not, at all.
00:12:38Lloreanzajacattack: If nobody answers, assume it was read, and nobody knew, and ask again in maybe a half hour.
00:12:54Lloreanzajacattack: Don't assume it was ignored or missed and spam the channel by repeating it every five minutes, please.
00:13:00Klevi..theres a differnce in Ipods with varying hard drive sizes?
00:13:09LloreanKlevi: 2G refers to "2nd Generation"
00:13:26KleviIm used to seeing 2ndG
00:13:45 Quit obo ("bye")
00:14:02Lloreanzajacattack: Other options involve posting to the dev mailing list, or creating a post in the appropriate section of the forums, etc. Probably the mailing list is best.
00:14:21 Quit jhulst ("Konversation terminated!")
00:14:54Bagderzajacattack: I'd say it means there was a failure when sending usb data
00:15:12Bagderbulk write is the kind of sending it does
00:15:20 Nick Klevi is now known as Klevi_food (
00:16:35KiononOk, I've read the manual and the faq, does installing rockbox allow me to reinstall apple's software later if I need to?
00:16:46ToadinatorIt's still thre
00:17:46zajacattackis it best to just disconnect the sansa and try again later?
00:18:08Bagderyour guess is as good as mine
00:18:20 Quit bertrik (Read error: 104 (Connection reset by peer))
00:18:24Bagdertry later, try another port, try another computer I'd say
00:18:55 Join BigMac [0] (
00:19:17Bagderor possibly investigate libusb if you can get more details or try other approaches
00:19:52 Quit Zagor ("Client exiting")
00:20:09 Join JdGordon [0] (
00:21:52n1sKionon: if you read the manual you should have gotten to the section about uninstalling eventually ;)
00:23:53 Part zajacattack
00:24:13 Join andruk [0] (
00:26:42andrukis there a rockbox installer for linux (or does the installer run on wine) without having to install ipodlinux?
00:28:06 Quit n1s ()
00:29:12ashesandruk: im doing it now for the first time
00:29:46andrukty all
00:29:47 Join Rob2222 [0] (
00:29:52Bagderandruk: there's even a manual you could read...
00:30:25 Quit ddalton ("leaving")
00:30:30andrukcurrently doing so. i havent found my answer in the manual yet.
00:30:47ashesget rbutil (qt application), run as root, go to the installation menu and install the bootloader, 'firmware upgrade' the player (with iriver)
00:30:53 Join scorche|w [0] (
00:31:02ashesafter that i get segmentation faults from rbutil
00:31:10asheswhen installing rockbox
00:31:14andruklol, of course as soon as i say that i find what i was looking for...
00:31:24ashesi got the bootloader installed
00:31:29ashesand .rockbox is installed
00:31:34ashesbut it's booting iriver
00:32:29BigBambiCouldn't they hve looked Can someone remind me when the latest iPod bootloader required is from (roughly)
00:32:44BigBambiignore the first few words of that :)
00:33:17BigBambii.e. roughly how old doesn't work with current builds
00:33:37Bagderthe pp bootloader change that requires upgrade was done march 4
00:33:39ashesand now it doesn't boot on it's own, unless i plug it in and let it boot automatically to iriver's os
00:34:01BigBambiBagder: thanks
00:34:34 Quit Toxicity999 (Remote closed the connection)
00:35:44 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
00:36:32 Quit Kionon (Read error: 110 (Connection timed out))
00:36:33 Quit bgmrk ()
00:37:09ashesandruk: the rbutil application contains a link for the manual
00:37:10 Join Kionon [0] (
00:37:42 Part Klevi_food ("Leaving")
00:37:56andrukcool. i found the part in the manual, so i should be good to go. i just have homework to do instead, so itll be awhile before i get to it. :-/
00:38:45 Join Klevi [0] (
00:39:20 Nick PaulJam__ is now known as PaulJam (
00:43:00 Quit BigMac ("Leaving")
00:43:40 Quit Toadinator ("CGI:IRC (Ping timeout)")
00:44:17 Quit Toxicity999 (Read error: 104 (Connection reset by peer))
00:45:07spkyanyone use evilg's fusion?
00:45:48 Quit sin613 ("Leaving.")
00:47:10KiononHmm... did my question go through?
00:48:31PaulJamthe one about restoring the apple firmware?
00:49:03KleviSomeone mentioned your name awhile ago
00:49:13KleviIll see if i can find logs of mine
00:49:27KiononI got disconnected, unfortunately.
00:49:34KleviI see.
00:49:41 Quit ender` (" I went to the cinema, and the prices were: Adults $5.00, children $2.50. So I said, "Give me two boys and a girl."")
00:49:50Klevi>.< and I just noticed that since i reinstalled xchat logging got disabled..
00:50:00 Quit Jon-Kha (Read error: 131 (Connection reset by peer))
00:50:07 Join Jon-Kha [0] (
00:50:23pixelmathe channel is logged
00:50:39Kleviits updated on the wire?
00:50:44*pixelma prepares for a red delta
00:50:50KiononI just want to make sure that installing rockbox won't A) preclude the reinstallation of apple's firmware later B) delete my current files from my ipod.
00:52:39scorche|wKionon: rockbox doesnt mess with apple's firmware and is able to be uninstalled no problem; the installation will only delete files if you reformat
00:53:44PaulJambut it's always a good idea to have a backup...
00:54:06KiononI do.
00:54:11KiononBut just in case.
00:55:49 Join Reno [0] (i=4c10ea95@gateway/web/cgi-irc/
00:56:45 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
00:58:53 Join t0m4 [0] (
00:59:51KiononAnd I just froze it. :/
01:00:58 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
01:01:05Nico_Pwow 35k on sansa :p
01:01:41pixelmathe recording code is huge
01:02:58 Quit andruk ("ChatZilla [Firefox]")
01:03:15pixelmathat's why the OndioSP still has rombox and the FM doesn't (plus radio of course) :\
01:03:16Renohi i got a question i want to kow how i can install the GIFviewer on rockbox?
01:03:38krazykitReno, you need to compile your own build for that, patching the source
01:04:28 Quit jhMikeS (Nick collision from services.)
01:04:34 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:04:36parafin|awayReno, it's not very usable
01:04:50pixelmaof course hwcodec recording is a bit different from swcodec recording...
01:04:58KiononWhat kind of video can rockbox handle?
01:05:11Kiononjust mpeg?
01:05:26KleviI think.. *Is New*
01:05:44pixelmaread more in the manual and in the wiki (PluginMpegplayer)
01:06:09Renoum...i don't know how to compile. i read a page i fiound on the wiki about how to compile and i really don't understand it, (actually i'm not really even sure what compile means) sorry. =(
01:06:30KleviIm learning myself ^.^
01:06:41parafin|awayReno, what player do you have?
01:06:58Renoipod video 30gigs 5G.
01:07:09parafin|awayyou can try my custim build
01:07:32Renocustom build? where can i get it?
01:08:10Renoso i ccan just download it here roght?
01:08:20Renoi ment right
01:08:40***Saving seen data "./dancer.seen"
01:09:07parafin|awaydownload and unpack to root dir of player
01:09:17KiononOh bah.
01:09:28KiononI only bought my ipod for video. :/
01:09:45KiononOkay, thanks for the help, guys. Rockbox installed correctly.
01:10:17Renokk thank you. let me try it! :)
01:10:19PaulJamKionon: you can still boot into the apple firmware to watch videos
01:10:21 Quit scorche|w ("CGI:IRC (EOF)")
01:12:10Renook i got it, so all i need to do is click on the image and it will load right?
01:12:54KiononHow do I do that, Paul?
01:13:11Renosince it's a GIF image when the image loads will it move like its supposed to or will it stay still?
01:13:25JdGordonwow! 35k in recording!
01:13:43 Quit petur ("switch")
01:13:48 Join petur [0] (n=petur@rockbox/developer/petur)
01:13:58 Quit pixelma (Read error: 104 (Connection reset by peer))
01:14:44 Quit SirFunk (Remote closed the connection)
01:14:47parafin|awayReno, it won't move
01:15:01PaulJamKionon: i think by enabling hold while booting. shoult be explained in the manual
01:15:08parafin|awayyou'll be lucky if it loads at least normal image as expected
01:15:17parafin|awayas i said it's not very usefull
01:16:11 Join webguest38 [0] (i=c0dc8ac6@gateway/web/cgi-irc/
01:16:42KiononYep, thanks
01:17:00 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
01:17:10KleviWelcome back Pixelma
01:18:01Renois it possible to make a viewer that makes thos types of images move?
01:18:18KiononOnce rockbox incorporates most forms of video (h264/x264, XviD, DivX, etc), I'll probably be more interested in it.
01:18:36LloreanKionon: Why?
01:18:54KiononBecause I bought my video ipod specifically for video.
01:18:56KiononNot for audio.
01:18:58LloreanRockbox only needs one good video codec. As you have to pre-scale the videos in advance anyway, there's no good reason not to transcode them to the supported format.
01:19:11KiononNot strictly true.
01:19:35KiononI'm a video editor, those of us in my community use a variety of standards.
01:19:42*amiconn wonders why someone would buy a dap with video add-on instead of a real pmp for video
01:19:57KiononXvid/AVI and h264/MP4 being the most common, but not the only one.
01:20:08LloreanKionon: Yes, and those videos will need to at least be resized for use.
01:20:16 Join midgey [0] (
01:20:34LloreanKionon: If they're not 320x240 in advance, then yes
01:20:38KiononI just figured some program could just display "fullscreen"
01:20:40LloreanAnd very few videos start at 320x240
01:20:42KleviKionon *unrelated* Is there another player besides VLC that supports FLV file types?
01:20:47amiconnAnd afaiu h264 is completely out of question for the cpu power of any rockbox target, *perhaps* except the gigabeat S
01:20:50LloreanKlevi: ffdshow
01:20:55LloreanSo, anything in windows, just about.
01:21:00KiononYeah, that.
01:21:09Klevi*nods* thanks
01:21:15KiononGet amvapp, and you'll be good.
01:21:19LloreanWell, ffmpeg, so "Xbox media center, any linux player that incorporates it, etc"
01:21:28Kiononffmpeg is eh
01:21:31LloreanKionon: There's not enough CPU power available to automatically scale videos.
01:21:40KiononBoo. :/
01:21:40KleviVlc is fine on my desktop, but shoots up my laptops processor, making it heat up to almost burning my leg after awhile
01:22:06KiononIt's just that the transcoding I do now with either third party software or itunes makes it crappy.
01:22:16LloreanKionon: So, you have to convert them to display resolution in advance. MPEG2 was partially chosen because it'll get better framerates than the other formats. Would you rather 5-10 excellent looking FPS, or 30 "decent" looking FPS?
01:22:25LloreanNot that the ipod video even will get 30
01:22:33LloreanSince there's no way currently to use the broadcom chip
01:22:35*Isolinear <3 XBMC
01:22:37Kionon29.97 is fine with me :P
01:22:44KiononSince that's what I edit in.
01:22:49KiononI don't need the whole 30.
01:22:55LloreanKionon: So, you'd choose fluidity over individual frame quality?
01:23:01KiononYes, probably.
01:23:02 Nick Klevi is now known as Klevi_away (
01:23:06KiononDepending on the source.
01:23:13LloreanThen you've just admitted that the mpeg-2 player we use ought to be fine.
01:23:26 Part toffe82
01:23:31LloreanThen again, iPod Video is unlikely to ever have decent video playback in Rockbox anyway.
01:23:41LloreanSo it shouldn't be a big concern of you what format we chose, since you probably won't use it. :)
01:24:03KiononI'm not arguing that. All I'm saying is I wish there was a way that whatever I have on my computer (including skittles at 1024p) would play. >P
01:24:10KiononAs insane as that request may be.
01:24:31 Quit jhMikeS (Nick collision from services.)
01:24:37 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
01:24:38LloreanThere is no way.
01:24:44spkyanyone use EvilGs Fusion build?
01:24:59LloreanThe iPod is a dual core 75mhz processor.
01:25:01 Join chris_mt [0] (
01:25:05spkyif so, what are the requirements for album artwork?
01:25:27spkyalbumname.bmp in the same folder?
01:25:30KiononOh my.
01:25:33LloreanI've been corrected, 80mhz
01:25:39LloreanForgot that we finally are running them at spec.
01:25:48LloreanWell, some of them
01:25:48KiononThat's... bad.
01:25:54LloreanOthers can in theory go up to I believe 110.
01:26:02LloreanOr maybe it was 100
01:26:11Renoi have another qestion or questions. i installed the doom wad file thingy, and it work great but i have a different doom game wad is it possible to switch between both doom games?
01:26:25t0m4is it possible to install apache on an archos ?
01:26:26LloreanReno: See the instructions for addons
01:26:37Lloreant0m4: Rockbox isn't Linux.
01:27:02KiononThere is a linux for ipod, I think.
01:27:05Renomy other question is that when i save gameboy roms it wont save? it says data abort?
01:27:23Llorean's a bug then
01:27:50Renook? can i fix that? :)
01:28:20Renoa patch or something?
01:28:43Renocuz...i really don't know how to apply patches.
01:28:44LloreanWell, you'd need to figure out what's going wrong first.
01:28:58LloreanIf there were simply a patch that fixes the bug, we'd probably already include it.
01:29:18Renoi was just asking...
01:29:38 Nick Klevi_away is now known as Klevi (
01:29:56spkythanks alot =]
01:31:17Renowhat do you mean figure out what the problem is? All I know is that it says "Data abort at O1F852AC (0)" thats it.
01:33:06Nico_PJdGordon: seen my MoB update ? I've made good progres
01:33:40 Quit Toxicity999 (Read error: 104 (Connection reset by peer))
01:33:56Nico_Pand I have backwards skipping too now, although it fails in some cases for a reason I don't get
01:34:39Renothese are my final 2 questions, what is "Recording" on rockbox used for? If i download some video codecs can i paste them in the codecs directory, and will rockbox play that type of video format?
01:34:56LloreanRecording is used for recording audio from a microphone, and no.
01:35:59amiconnpetur: Why didn't you optimize the fd range checks (using an unsigned comparison)?
01:36:03*amiconn likes this trick
01:36:37peturfeel free to change
01:36:39amiconnif (fd < 0 || fd > MAX_OPEN_FILES-1) ==> if ((unsigned)fd > MAX_OPEN_FILES-1)
01:36:58peturI just copied from the other checks
01:37:00 Join qweru [0] (
01:37:02 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
01:37:05amiconnaha, hmm
01:37:07Renook thanks for the info and for that custom build! :)
01:37:15amiconnSome potential for green deltas :)
01:37:17Reno:) :) :) :)
01:38:22Nico_Pamiconn: casting to unsigned doesn't change the bits, right ? that's why it works ?
01:38:57 Quit Reno ("CGI:IRC (EOF)")
01:39:00Nico_Pok :) at first I thought the cast would give the absolute value
01:39:29amiconnI used this trick for clipping in the lcd drawing routines
01:39:49peturtime date calculation of the front page is funny: my commit is 'thursday' while the other ones before that are 'today' :)
01:40:05amiconnYes, due to the still unfixed 1 hour difference
01:40:42*petur enjoys committing in the future for another 20 minutes
01:43:51 Join GERM [0] (i=8ea5d033@gateway/web/cgi-irc/
01:45:32GERMQUESTION: rockbox now supports the c200 line of audio players, could the c100 line easily be ported using the existing c200 firmware?
01:45:37 Join Spiritsoulx [0] (n=eyes_of_@
01:46:03blitheDoes anyone ever find that wallpaper they manually set from their files don't stay when their player is turned on and off?
01:46:06krazykitGERM, if the c100 shares most of the same hardware, yes
01:46:25GERMthanks ill check on the hardware now
01:46:34PaulJamblithe: is the file in the .rockbox/backdrops directory?
01:46:50blithePaulJam: Nope, in my 'PICTURES' directory.
01:46:52jhMikeSpetur: I can think of a possibility why someone might experience lulls in flushing still. scheduler-related of course (mutexes in FAT driver). tbf anyway soon.
01:47:06 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:47:11amiconnpetur: Wow, it seems gcc already knows that trick, at least for SH
01:47:13PaulJamblithe: then copy it to the backdrops dir.
01:47:15blitheThe wallpaper can be set fine, but gets reset to the underlying color on restart.
01:48:06*jhMikeS at least has nicely working dual-core test plugins on H10
01:48:28amiconnblithe: You need to put it in /.rockbox/backdrops to make it permanent. This is intended behaviour
01:48:40blitheOdd, why would that be intended?
01:48:43blitheJust curious.
01:48:58SpiritsoulxSo I just installed RB on my friend's Sansa c2xx, and a few days later, the right output of the headphone jack goes dead. Anyone know why?
01:49:10SpiritsoulxNo, wait, left output. >_>
01:49:42GERMmine went dead a while back in both it probally just crapped out on you
01:49:44krazykitSpiritsoulx, bad hardware, most likely, but try using the latest build, and then clear the settings, to be sure. also check in the OF first
01:49:52GERMi use it for storage now...
01:50:00krazykitif it happens in rockbox and the OF, it's hardware.
01:50:08amiconnPartially it is a leftover from the old settings system. But it also allows testing backdrops without problems. As long as you *dont* put it into the place for permanent backdrop, you can even make your display completely unreadable with a backdrop. Reboot, and it's gone
01:50:22SpiritsoulxThat's what I thought.
01:50:29blitheamiconn: Ah, got it.
01:50:39blitheI can live with that. Just didn't know I had to do that.
01:51:30 Quit spiorf (Remote closed the connection)
01:51:38 Join Rob2222 [0] (
01:52:11*amiconn checks the other archs
01:52:56SpiritsoulxHow do you get into diag mode again?
01:54:18SpiritsoulxI meant for the Sansa Cxx.
01:55:01chris_mthey can anybody help me apply a patch to my cvs tree?
01:55:08chris_mtI can't get it to work
01:55:41scorchewhat diagnostic mode?...are you referring to either recovery or manufactuing mode?
01:56:04scorchechris_mt: what error are you ancountering? (and we use svn)
01:56:04SpiritsoulxThere's a diag mode on Sansa Cxx, isn't there?
01:56:11chris_mtsorry, I mean svn.
01:56:13scorcheif there is, i havent heard of it
01:56:22chris_mtjust a sec, let me copy/paste it...
01:57:26chris_mtcan't find file to patch at input line 5
01:57:41chris_mtI'm in the root and it's looking for apps/plugins/viewers.config
01:57:56chris_mtthis is the patch:
01:58:05chris_mtstarting with the first one...
01:58:31scorchefirst one?
01:58:45scorchethe updates are further and further down
01:58:55chris_mtthe .patch at the top of the list. or do I use the latest one?
01:58:59chris_mtbe right back...
01:59:29 Part webguest38
02:02:57amiconnReally interesting...
02:03:10GERMhaving trobles finding specs for c100/150 will the rockbox c200 firmware work at all (noob q)
02:03:11*amiconn just didn't expect that gcc knows this trick
02:04:14LloreanGERM: No, Rockbox only runs on the players it specifically says it runs on
02:04:30GERMk :P
02:04:50GERManyone know where to find spec for sansa players
02:05:27LloreanThe easiest method is probably google. Second easiest is to open up the player, identify the chips, then google.
02:05:58 Part GERM
02:06:55 Quit petur ("Zzzzz")
02:06:59 Quit t0m4 ()
02:07:29 Quit PaulJam (".")
02:10:34pixelmadidn't Soap say he had a C100 and wanted to look into it?
02:10:53 Quit Spiritsoulx ()
02:14:22SoapYes he did.
02:14:33*linuxstb passes Soap a screwdriver
02:14:46Soapno screws, the case is sealed tighter than Fort Knox.
02:14:56*linuxstb passes Soap a hammer
02:15:01SoapI'm out of town again, but debating physical destruction of the case.
02:15:25Soapand no, GERM, (who appears to have left) I can not find any specs.
02:15:34linuxstbSo we've no idea what CPU is inside?
02:15:40Soapnot yet
02:15:44linuxstbAre there any firmware updates?
02:16:09SoapI've read about some on the Anything But Ipod forums, but haven't use one yet.
02:17:28Soapperfect timing, pixelma. I had /just/ finished reading all the back-logs when you said my name.
02:18:59 Join iamben [0] (
02:19:57 Quit Toxicity999 (Remote closed the connection)
02:20:37linuxstbAh telechips ;)
02:20:38chris_mtok, so about that patch...
02:20:59Soapyou figured that out already?
02:21:19chris_mtscorche: you still here?
02:21:31scorcheyes, but what do you want me specifically for?
02:21:38SoapI had some assumptions, but wasn't going to voice them being ASSumptions.
02:21:53chris_mtany idea what I'm doing wrong applying that patch? I haven't done it before.
02:22:15chris_mtdo I take the first one from the thread, or the most recent update?
02:22:16*scorche gets out his crystal ball
02:22:36Soapchris_mt, Step back for a second. Why would you not want the most recent update?
02:22:38scorchewhy would you want to try an dapply an outdated version?
02:23:22chris_mtThe plugin files aren't in the SVN tree. So are they incremental, or does the last one contain all changes?
02:23:40Soapchris_mt, I say step back for a second, because I think you are getting overwhelmed, if not spooked. Take it slowly, think about each step, and I do believe between logic and the documentation provided you will find you can do it.
02:24:23krazykitchris_mt, they're not incrementel. that would be masochistic
02:24:38chris_mtI'm not afraid to screw it up - I can just re-checkout. But even the latest .patch file gives me similar errors.
02:24:45chris_mtThe paths all appear to be from the root.
02:24:53chris_mtand the files in question are there...
02:25:49krazykitchris_mt, well, maybe not recheckout, but certainly svn revert
02:26:22Soapchris_mt, no need to recheck out. "svn revert -R ." from the root of your svn (usually "rockbox")
02:27:10 Quit Nico_P (Remote closed the connection)
02:27:12Soaprevert as in "undo all changes" -R as in "recursive in all subdirectories, . as in "start HERE"
02:27:13 Part pixelma
02:27:17chris_mtok, using tortise SVN gave me some clues...
02:27:20 Quit squigtreiba (Read error: 104 (Connection reset by peer))
02:27:54chris_mtIt fetched the correct revision of the file and took me to tortisemerge. Do you have to do that manually when using the command line?
02:28:22*Soap can't speak to the tortise
02:28:43 Join RoC_MasterMind [0] (
02:30:12 Nick fxb is now known as fxb__ (
02:31:00 Nick fxb__ is now known as fxb (
02:32:46 Nick fxb is now known as fxb__ (
02:36:32 Join BigMac [0] (
02:36:45 Quit Arathis (Remote closed the connection)
02:37:30linuxstbSoap: Looks like the Telechips TCC770
02:37:53Soapyou want it?
02:38:50linuxstbSame as the M200
02:39:09linuxstbYou don't?
02:40:45 Quit chris_mt ("ChatZilla [Firefox]")
02:41:10SoapDo you think I will be able to do anything productive with it?
02:41:51KleviCygwin doesnt detect Either arm-elf-gcc or arm-elf-ld, I tried reinstalling it (downloaded from Rockbox) with no success. Any Ideas?
02:42:11SoapKlevi, it is a PATH issue I assume, step 4 IIRC
02:42:20KleviTried that as well.
02:42:34KleviMaybe I typoed something, though
02:44:03 Join toffe82 [0] (
02:44:33 Quit BigMac ("Leaving")
02:45:01KleviHmm.. even after using the alternate method (typing this in the build directory export PATH=/usr/local/arm-elf/bin:$PATH) I still get the same error when attempting to compile
02:48:26 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
02:57:15 Quit Klevi ("Quit messages. The unintentional channel spam of people closing IRC, realizing they have a life.")
03:06:03 Quit atsea-34 (Remote closed the connection)
03:08:43***Saving seen data "./dancer.seen"
03:16:50 Join jbm [0] (
03:18:57 Quit Toxicity999 (Remote closed the connection)
03:24:28 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
03:24:38 Join ArchFool [0] (
03:24:43ArchFoolW00t! :)
03:32:22ArchFool'aloo? Anybody 'ome? Anyone with a Sansa c200?
03:32:49scorchei see 130 users...and no question
03:33:34ArchFoolHeh... how do I mount my sansa c200 running rockbox? It connects and auto-mounts when running the native OS, but not rockbox.
03:34:00krazykityou connect in the original firmware, the "native os"
03:34:11krazykitUSB is currently not implemented.
03:34:21ArchFoolAahhhh... that explains it.
03:34:25ArchFoolFor curiosity, why not?
03:34:54krazykitbecause coding takes time, and the usb stack is still being written
03:35:06ArchFoolI just hate having to rebuild the database for the sansa firmware (slow!) and THEN for rockbox!
03:35:22 Join Klevi [0] (
03:35:25ArchFoolkrazy: great.. so it IS being worked on. :) :) Just need to wait.
03:35:33*ArchFool twiddles his three thumbs.
03:35:41krazykitArchFool, neat tip, if you use a format not supported by the OF, like Vorbis, the OF's refresh will be far faster ;-)
03:35:54Klevican someone explain how to remove the radio from the sansa c200s source?
03:36:14scorcheKlevi: ifdef out all the radio related code
03:36:24Klevifurther explain?
03:36:40scorchewell, that is assuming you are saying what i think you are saying with that..
03:37:04maxkelleyjust disable it in the config-c200.h file.
03:37:20scorchewhich would trigger the ifdefs ;)
03:37:22 Quit RoC_MasterMind ("Leaving")
03:37:23Kleviedit it?
03:37:37KleviHow so?
03:37:43maxkelleyscorche: yeah, but that makes it sound like you have to edit a billion files.
03:37:56scorcheKlevi: a text editor..
03:38:12 Join DGMurdockIII [0] (
03:38:24*ArchFool pulls up the text editor on his sansa c200. ;)
03:38:32maxkelleyheh, not quite.
03:38:43KleviI know that much.. i meant exactly what and how doi edit it?
03:39:01scorchecomment/uncomment it
03:39:22Kleviwith the # symbol, correct?
03:39:30 Quit midgey (Read error: 104 (Connection reset by peer))
03:39:33scorchethat is perl...not C
03:39:50scorchewell, other languages too..
03:39:51Kleviokay, sowhat does perl use
03:40:24scorchei am saying that the hash symbol is the wrong symbol for commenting code in C
03:40:27maxkelleyno, what you're editing is C :)
03:40:41maxkelleyC comments look like this /* comment */
03:41:22maxkelleyso, at the portion labeled, FM Tuner, put a /* at the beginning of the first line, and a */ at the end of the second.
03:41:28 Quit Kionon (Connection timed out)
03:41:45maxkelleywhy would you want to comment out radio, though?
03:42:02 Join Kionon [0] (
03:42:10KleviMy sansa hangs upon loading rockbox when the radio is enabled
03:42:19Klevigod knows why
03:42:35Klevithe OF can use it fine
03:42:36maxkelleywell, you should have told us.. are you latest svn?
03:42:54Kleviwhich, is the same as low_light's and he has no problem
03:43:02KleviNo, not right now
03:43:06KleviI will be
03:43:16maxkelleyare you rather closeish, though?
03:43:32KleviI tried it, without trying to edit it at alll, and it still didnt work
03:43:40scorcheKlevi: let us know if it does that with the current build
03:43:41maxkelleywell, you must, because the radio was enabled.
03:43:47maxkelleyanyhow, I'm going to bed.
03:44:36KleviI did tell you guys
03:44:43Klevimaybe it was missed.
03:50:20Klevi*testing newest build out now*
03:51:14Klevidoesnt hang
03:51:41Klevibut i get an Undefined Instruction at 01000014c (0)
03:52:00Klevithen it hangs
03:52:10Kleviit used to just stop at the Rockbox logo screen
03:52:50Kleviscorche ^
03:54:38Klevi<maxkelley> just disable it in the config-c200.h file. <−−−−−−- Where is that file located?
03:56:54 Quit hcs ("Leaving.")
04:05:15KleviNevermind. I found it, started compiling, all is well so far =)
04:06:55 Quit Klevi ("Quit messages. The unintentional channel spam of people closing IRC, realizing they have a life.")
04:20:06 Quit ArchFool ("p00f!")
04:23:46 Quit Bitter (Client Quit)
04:24:32 Join HellDragon_ [0] (n=Nocebo@unaffiliated/helldragon)
04:25:10 Quit HellDragon_ (Read error: 104 (Connection reset by peer))
04:25:37 Join HellDragon_ [0] (n=Nocebo@unaffiliated/helldragon)
04:27:54 Quit animeloe ("Leaving")
04:28:24 Quit DGMurdockIII ("get satisfied! • :: ««« (Gamers.IRC) »»» ::")
04:29:24 Nick HellDragon_ is now known as Bitter (n=Nocebo@unaffiliated/helldragon)
04:30:16 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
04:46:09 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
04:46:27 Join JasonF [0] (n=jay@no-sources/jasonf)
04:47:27JasonFI have the ipod bootloader thing installed on my ipod already, with the only options set to apple os, disk mode, and sleep. Is it possible (or is there a howto) on how to integrate rockbox with this?
04:48:07scorcheit is possible, but that is offtopic here, as that is not our bootloader
04:48:17scorchetry #ipodlinux
04:48:52JasonFWell then, question #2... can rockbox loader operate the same way, with options to boot apple os or rockbox on the fly?
04:49:00JasonFI'm basically hoping to do music with apple stuff, and games with rockbox stuff
04:49:29scorcherockbox's bootloader is dualboot by default
04:50:14JasonFi'll try that for now then
04:50:34 Quit Thundercloud (Read error: 104 (Connection reset by peer))
04:52:12 Join webguest13 [0] (i=4a469ecd@gateway/web/cgi-irc/
04:52:19 Quit webguest13 (Client Quit)
04:52:27 Join webguest13 [0] (i=4a469ecd@gateway/web/cgi-irc/
04:53:11 Quit webguest13 (Client Quit)
04:55:04 Join sneh1621 [0] (i=413cc33c@gateway/web/cgi-irc/
04:55:35sneh1621can someone direct me as to where i can find how to manually install rockbox on my 5G iPod video??
04:55:44LloreanI'd recommend the manual
04:57:03 Join alienbiker99 [0] (
04:57:03 Quit sneh1621 (Client Quit)
04:57:47 Join sneh1621 [0] (i=413cc33c@gateway/web/cgi-irc/
05:02:40 Join hannesd_ [0] (
05:03:44 Nick Gnu47 is now known as Gnu47|Asleep (
05:05:13 Join MBaller20 [0] (i=4a469ecd@gateway/web/cgi-irc/
05:05:32 Quit Toxicity999 (Read error: 113 (No route to host))
05:05:32 Quit sneh1621 ("CGI:IRC (EOF)")
05:06:24 Quit MBaller20 (Client Quit)
05:06:28 Join MBaller20 [0] (i=4a469ecd@gateway/web/cgi-irc/
05:08:00 Quit MBaller20 (Client Quit)
05:08:44***Saving seen data "./dancer.seen"
05:09:08 Quit hannesd (Read error: 110 (Connection timed out))
05:09:08 Nick hannesd_ is now known as hannesd (
05:10:18JasonFany way to set the apple firmware as the default
05:10:26JasonFso I have to put "hold" on to boot into rockbox?
05:10:47scorcheyou would have to recompile the bootloader
05:11:00 Join davina [0] (
05:12:54JasonFthanks scorche
05:12:58JasonFthis looks pretty cool
05:15:20 Quit bb (Nick collision from services.)
05:15:25 Join bb_ [0] (n=bb@unaffiliated/bb)
05:23:16JasonF /join #ipodlinux
05:26:36 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
05:33:02 Join JdGordon|uni [0] (i=82c20d67@gateway/web/cgi-irc/
05:33:22 Quit JdGordon|uni (Client Quit)
05:34:36 Join ilovegamestah [0] (
05:35:17ilovegamestahhi i installed rockbox last night. Everything was working fine. Turned it off and wene to turn it back on today and nothing happens screen just stays blank even when i plug the power in
05:35:20ilovegamestahwhats the go here?
05:36:44ilovegamestahok it just needed to be reset
05:36:52ilovegamestahmust have foze or something
05:44:06 Nick jhulst is now known as newb (n=jhulst@unaffiliated/jhulst)
05:44:37 Nick newb is now known as jhulst (
05:47:41ilovegamestahfuck the 5g is slow with rockbox
05:48:23scorchethe language isnt needed...
05:49:57 Join psycho_maniac [0] (
05:50:03ilovegamestahlittle kiddies i guess
05:50:19 Part JasonF
05:55:53 Quit ilovegamestah ()
06:02:56 Join Kionon_ [0] (
06:03:15 Quit Kionon (Read error: 104 (Connection reset by peer))
06:12:47 Join n17ikh [0] (
06:16:59 Quit psycho_maniac (" HydraIRC -> <- s0 d4Mn l33t |t'z 5c4rY!")
06:18:48 Quit n17ikh|Lappy (Connection timed out)
06:37:31 Quit Kionon_ (Connection timed out)
06:38:17 Join Kionon_ [0] (
06:42:59 Join Chronon [0] (
07:04:36 Quit qweru ("moo")
07:08:46***Saving seen data "./dancer.seen"
07:10:12 Nick fxb__ is now known as fxb (
07:12:21 Join kubiix [0] (
07:12:27 Quit jhulst ("Konversation terminated!")
07:12:41 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
07:17:05 Join petur [0] (n=petur@rockbox/developer/petur)
07:25:43 Quit Toxicity999 (Remote closed the connection)
07:27:06 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
07:29:46 Join blithen [0] (
07:30:04blithenhey everyone.
07:32:29 Quit Kionon_ (Connection timed out)
07:33:13 Join Kionon_ [0] (
07:33:19 Part blithen ("Leaving")
07:40:43 Part toffe82
07:43:28 Join blithen [0] (
07:43:47blithenHas anyone heard of the 2nd gen iPod nano firmware being worked on?
07:44:46 Nick fxb is now known as fxb__ (
07:45:54 Join ddalton [0] (
07:46:11 Quit ddalton (Client Quit)
07:46:16 Join ddalton [0] (
07:47:03 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
07:48:41ddaltondid my message come through?
07:49:07ddaltonok thanks I can't seem to svn up
07:49:59ddaltonmaybe I can now
07:51:46 Part blithen ("Leaving")
07:52:15ddaltonAny comments on p7910?
07:52:58 Nick rmf_ is now known as sslashes (n=rmf@
07:54:03ddaltonanyone looked at it?
08:01:35 Join haemmy [0] (n=stefan@
08:06:05 Quit petur ("home->work")
08:14:46 Join bertrik [0] (
08:14:53 Part Llorean
08:25:57ddaltonwhat variable should I send to my time function that holds the elapsed time in seconds for the wps?
08:29:09 Quit jhulst ("Konversation terminated!")
08:31:50 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
08:33:30 Join petur [0] (n=petur@rockbox/developer/petur)
08:33:45 Part safetydan
08:33:48 Quit Kionon_ (Connection timed out)
08:34:14 Join Kionon_ [0] (
08:42:29 Quit BigBambi (Read error: 113 (No route to host))
08:43:13 Join Rob222241 [0] (
08:45:49 Quit bertrik (Read error: 104 (Connection reset by peer))
08:46:26 Join cendres [0] (n=ashes@2001:5c0:8fff:ffff:0:0:0:23)
08:47:20 Join atsea-34 [0] (i=atsea-@gateway/tor/x-0c7c0bb67ee0108c)
08:48:43 Quit ashes (Read error: 113 (No route to host))
08:50:25spkyany input?
08:50:59GodEater_needs too many patches =/
08:55:17 Join ender` [0] (
09:00:31 Quit Rob2222 (Read error: 110 (Connection timed out))
09:06:57 Quit Kionon_ (Read error: 104 (Connection reset by peer))
09:08:47***Saving seen data "./dancer.seen"
09:08:49 Join CaptainSquid [0] (
09:11:16 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:20:19 Quit JdGordon ("Konversation terminated!")
09:21:48 Join JdGordon [0] (
09:25:57 Join n1s [0] (
09:39:47 Join barrywardell [0] (
09:41:08 Join qweru [0] (
09:41:41GodEater_seriously - someone needs to lock that patching thread and tell DavidRawson to stop wasting our (and his) time.
09:42:23ddaltonGodEater_: what pach thred?
09:43:00GodEater_in the "Getting Started and Compiling" forum
09:43:09GodEater_the thread is called "Patching"
09:46:53 Join homielowe [0] (
09:48:59 Quit haemmy ()
09:50:23Bagderwow, he has some minor details left to figure out ;-)
09:50:34petur"I did the stuff that does the thing and it did the install thing"
09:50:49peturready for GoldenQuotes
09:51:29 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
09:52:27 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
09:53:35 Quit qweru ("moo")
09:59:58 Nick fxb__ is now known as fxb (
10:01:20 Nick fxb is now known as fxb__ (
10:01:56 Join aliask [0] (
10:02:08ddaltonHey aliask
10:02:36 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
10:05:52Nico_Paliask: any news about the S ?
10:08:19aliaskNico_P: Not really, I haven't had much time to spend on it recently.
10:09:15GodEater_is the S code still in a branch, or has it been merged into trunk ?
10:09:27aliaskIt's been merged back in
10:11:44Nico_Pinterestin log I just read (yesterday's for #gigbeat)
10:12:14GodEater_ooh - doesn't even compile at the moment =/
10:12:34aliaskBootloader, or normal build?
10:14:37 Join spiorf [0] (
10:15:30 Join obo [0] (n=obo@rockbox/developer/obo)
10:18:09aliaskWell, it's an easy enough fix - pcm stuff hasn't been implemented so all those functions should be empty.
10:18:56JdGordonhey Nico_P
10:19:00JdGordonnice work on the mob update :)
10:19:39 Quit JdGordon ("rebooting")
10:20:56 Quit TMM (Read error: 110 (Connection timed out))
10:22:16 Join JdGordon [0] (
10:25:54pondlife"I put the player in manufacture mode and renamed rockboxpatch.c to BL_SDboardSupport.rom and patched it to my player with e200tool." lol
10:28:57Bagdernot _quite_ there yet ;-)
10:29:12pondlifeAt least he didn't use playback.c ;)
10:30:53aliaskGodEater_: Last commit should have got it compiling again
10:31:50pondlifeNico_P: I'm liking your recent playback.c mods (in your mob git), seems to be simplifying quite nicely now.
10:31:50n1saliask: what is the current state of the port? waiting for usb?
10:32:33*GodEater_ thinks aliask should update the port info page with details of yesterday's #gigabeat discussion
10:32:59n1sGodEater_ do you have a link to those logs?
10:33:00aliaskn1s: Sort of. Even when we have USB support in svn it likely won't work off the bat - we are still only running in the bootloader
10:33:28aliaskI'll get to editing the wiki right away
10:33:31*JdGordon wonders what i missed in #gigabeat?
10:34:35 Join thegeek_ [0] (
10:34:36GodEater_#gigabeat has a log link in the title....
10:42:07Jwhen usb support is in rockbox would it be possible for the DAP to sync it's clock a'la ntp? (obviously not out of the box, but would it be possible?)
10:42:19n1sGodEater_ hmm that's discouraging news about the brickage...
10:42:21GodEater_ <−− needs deleting...
10:43:10GodEater_n1s: yeah - I'm keen to see if I can get the firmware updater working on linux, but testing it is...not inviting currently
10:43:43n1sis there no known way to recover a bricked unit?
10:43:43BagderJ: everything is possible, but I would guess clock sync to be very low prio in the usb land
10:44:05GodEater_n1s: that's why I wanted to get hold of toffe, and ask if he'd made progress on the JTAG interface
10:44:22GodEater_he first posted about it back in March (in the thread), but there's precious little else been said about it
10:44:26JBagder: thanks. I was unsure as to whether the DAP could request the time signal or the host PC would have to tell the DAP
10:44:37aliaskToffe even suggested that it is possible to boot over the serial port (the imx31 datasheet indicates this also)
10:44:46JBagder: I don't mind it being a priority, just if it's possible :)
10:44:47n1sGodEater_: ah, yes, what kind of hardware would one need to use that?
10:45:06GodEater_aliask: but he never posted on whether he found the transmit part of the serial connector did her ?
10:45:21GodEater_n1s: no idea :)
10:45:37aliaskGodEater_: The transmit part?
10:46:04GodEater_aliask: his notes in the thread indicated he got data flowing from the S to his PC, but he couldn't go in the other direction (i.e. issue commands to it)
10:47:19GodEater_again though, I have no idea if that's the current state of affairs or not - he never mentioned anything more about it
10:48:23ddaltonI am on r15074. How do I go down to r15015?
10:49:00GodEater_svn co -r 15015
10:49:08GodEater_I think :)
10:49:15aliaskWell, it looks like he's found both tx and rx on the board. Maybe the software running on the S just wasn't listening to what he was outputting to it
10:49:19ddaltonok I will try that.
10:49:42GodEater_aliask: that's also possible I suppose - but would be unusual behaviour for a WinCE based device
10:49:50ddaltonnot enough arguments
10:50:11GodEater_ddalton: you still have to specify which repo you're getting the revision from
10:50:20n1sddalton: use "up" instead of "co"
10:50:26ddaltonI have checked out already
10:50:43ddalton"svn up r15015"?
10:50:43GodEater_ddalton: then do what n1s says, use "svn up -r 15015"
10:51:33GodEater_aliask: for example, I have a proper PDA that also uses WinCE, and when you hook the console up on that - you can issue commands to it
10:51:36ddaltonit just says at revision 15074 and does nothing.
10:51:47GodEater_aliask: which is useful for doing rom dumps
10:51:58pondlifeddalton: No space after the r
10:52:08 Quit thegeek (Read error: 110 (Connection timed out))
10:52:21 Join thegeek [0] (
10:52:31ddaltonstill no luck :-(
10:52:44peturGodEater: the WinCE debugger uses that interface too
10:53:00peturthe debugger on PC that is
10:53:33GodEater_ddalton: "man svn" then
10:55:05ddaltonI will just check out again at r15015 because can't be bother find it on google.
11:00:12GodEater_svn up -r should work
11:00:29ddalton"svn up -r15015"?
11:00:39GodEater_well svn help shows a space
11:00:45GodEater_but I've never used it
11:00:47GodEater_so I'm not sure
11:00:56 Quit thegeek_ (Read error: 110 (Connection timed out))
11:00:56*GodEater_ uses git svn
11:01:17Bagderwith or without space doesn't matter
11:01:34BagderI just tried both ways
11:01:35 Join thegeek_ [0] (
11:04:11pondlifeIndeed, I tried both under Cygwin too
11:07:24 Join juxtap [0] (n=juxtap@unaffiliated/juxtap)
11:08:21ddaltonPondlife: do you use "quilt" with cygwin?
11:08:49***Saving seen data "./dancer.seen"
11:09:04pondlifeJust patch and svn diff
11:09:38ddaltonApparently it is useful for handling multiple patches.
11:10:02ddaltonPondlife: what exactly did you type? "svn up -R 15015"?
11:10:21pondlifesvn up -r15015
11:10:35pondlifeNote lower-case r
11:11:05ddaltonOh sorry didn't pick that up with my screenreader. (Copy and paste is useful)
11:14:06 Quit barrywardell ()
11:16:46 Quit idnar (Nick collision from services.)
11:16:49 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
11:19:08 Quit thegeek (Read error: 110 (Connection timed out))
11:19:25 Quit ddalton ("leaving")
11:19:28 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
11:19:30 Join ddalton [0] (
11:19:37 Quit ddalton (Client Quit)
11:19:40 Join ddalton [0] (
11:19:52 Quit ddalton (Client Quit)
11:19:55 Join ddalton [0] (
11:20:32 Join Klevi [0] (
11:20:56ddaltonn1s: Are you nls?
11:21:06ddaltonin the tracker
11:22:02ddaltonoh you are
11:22:20ddaltondo you think the buttons should be voiced? (in the alarm screen)
11:22:41Klevirecorder/recording.c: In function 'recording_screen':
11:22:41Klevirecorder/recording.c:1870: error: 'SRCF_FMRADIO_PAUSED' undeclared (first use in
11:22:41Klevi this function)
11:22:41DBUGEnqueued KICK Klevi
11:22:41Klevirecorder/recording.c:1870: error: (Each undeclared identifier is reported only o
11:22:43Klevirecorder/recording.c:1870: error: for each function it appears in.)
11:22:45Klevirecorder/recording.c:1870: error: 'SRCF_FMRADIO_PLAYING' undeclared (first use i
11:22:47Klevin this function)
11:22:50Klevimake[1]: *** [/home/owner/rockbox/build/apps/recorder/recording.o] Error 1
11:22:51Klevimake: *** [build] Error 2
11:23:01peturKlevi: use pastebin please
11:23:04GodEater_Klevi: don't do that please
11:23:06KleviOh, wow I really have xchat for that right about now.
11:23:14KleviThat wasnt intentional
11:29:15n1sddalton: yes i am, and I think voicing those buttons could be helpful, how else should blind users know which button to press (well without looking in the manual)
11:29:49n1sand we display them in the screen so there seems to be a need for it
11:30:10 Quit Klevi ("Quit messages. The unintentional channel spam of people closing IRC, realizing they have a life.")
11:30:19pondlifeThe screen and voice UI should always match wherever possible IMHO
11:30:27pondlifeMakes support easier if nothing else
11:32:29 Join desowin [0] (
11:34:24aliaskGodEater_: ?
11:34:53GodEater_conf. call
11:35:17 Nick brm__ is now known as TMM (
11:35:21 Join Entasis [0] (
11:37:09 Join barrywardell [0] (
11:38:47ddaltonN1s: 1. How big will the voice file be? 2. Once you know the button presses then won't it just be really anoying?
11:38:57barrywardelllinuxstb: will I just make a tag for the bootloader? no need to add one for sansapatcher?
11:41:06barrywardelllinuxstb: also, should I change the version[] in bootloader/main-pp.c to 0.3 or leave it as APPSVERSION
11:41:08ddaltonPondlife: You really think all the button presses should be added to voice.
11:41:09n1sddalton: I don't feel strongly about it since I don't even have a player that supports alarm wake up but it seemed to cause problems for at least one user anyway I think you are a better judge of whether it's annoying or not than me
11:41:14ddaltonThat's going to be a big voice file
11:41:39pondlifeI just think the UI should be the same whether displayed or spoken
11:42:02ddaltonn1s: I think people can read the manual. But I can write up a patch if people want I will ask on the mailing list later Just need to subscribe.
11:42:03n1sddalton: I mean for _just_ this screen as it is non-standard, so it's just one added clip
11:42:25n1samd possibly other similar screens like set time and date
11:42:26peturI wouldn't voice buttons, we're not showing keys on the display too...
11:42:40n1spetur: we are, in some places
11:42:44ddaltonYes but it would be annoying when entering that screen (Especially if you use it a lot) and what about other screens like the delete file screen.
11:42:45pondlifeIdeally the voicing shouldn't even need special coding, it should come out from the lower-level code
11:42:49peturthen remove them ;)
11:43:35peturthose can be voiced because they promt the user for action
11:44:19ddaltonIt says "really delete" that's good. But should it say press select to delete stop to cancel
11:44:43ddaltonpondlife: what do you mean by "lower-level code"
11:44:46pondlifeIt should say whatever is displayed, and shouldn't need special coding
11:45:02ddaltonhow would that work? You need a lang file
11:45:22pondlifeThe displayed string should already be in a lang file
11:46:09ddaltoncan't you some times have it in a variable?
11:46:20ddaltonvariable ="something" display the variable
11:46:34ddaltonof type char I guess and I have another example hang on
11:46:46n1spondlife: afaik only splash() works like that automatically
11:47:02ddaltonID2P does that
11:47:39ddaltonPondlife: What about this? gui_syncsplash(HZ*2, "No .rockbox directory");
11:47:44 Nick bb_ is now known as bb (n=bb@unaffiliated/bb)
11:48:06pondlifeddalton: That couldn't be voiced or langed... No .rockbox dir = no .lang or .voice file
11:48:07JdGordonyou cant voice that splash
11:48:19JdGordonnot without TTS anyway
11:48:27ddaltongood point
11:48:34ddaltonbut something else like that?
11:48:34JdGordoninbuilt-TTS that is... having it in a plugin wouldnt work eother
11:49:00ddaltonAliask did you contact the author of espeak?
11:49:45 Quit linuxstb (Read error: 110 (Connection timed out))
11:50:16aliaskddalton: Yes, and he didn't want to change his license
11:50:59ddaltonAliask: So I guess we will just have to go back to an older version of espeak? Or do you have another idea?
11:52:08aliaskWell, the other option is to convert rockbox to GPLv3 - but that's a big issue
11:52:36n1sdid we ever reach a conclusion on which version we are currently under?
11:52:49n1sor rather are we "or later"?
11:52:49ddaltonSo we are just going to forget porting espeak to rb?
11:52:51ddaltonfor now?
11:53:24ddaltonn1s: isn't it v2?
11:53:41 Join PaulJam [0] (
11:54:14n1sddalton: there were some discussion going on but I don't know if a conclusion was arrived at
11:54:32ddaltonn1s: im pretty sure it is v2
11:55:05n1sbut yeah I think we are under v2 but the question is if we are under "v2 or later" or just plain v2
11:56:20ddaltonLine two of /docs/copying says " Version 2, June 1991"
11:56:28CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
11:56:28*JdGordon thought we were definatly v2 and most probably v2 or later
11:56:45GodEater_would anyone care if we (ever) manage to download all the files linked on the S Info page hosting them on the wiki itself ?
11:56:49GodEater_rapidshare is a PITA
11:56:50JdGordonor we would have no problems with the or later bit as far as imported code is concerened
11:57:14*ddalton Doubts a lot of players could even run espeak
11:57:29barrywardellBagder: can I send you the new sansapatcher versions for the download server?
11:57:31 Quit obo ("bye")
12:00:52*JdGordon wishes Karl would come online with some good news :(
12:07:41 Quit barrywardell ()
12:08:43 Quit aliask ("ChatZilla [Firefox]")
12:09:42 Quit ddalton ("leaving")
12:09:49 Join lee-qid [0] (
12:12:36 Join barrywardell [0] (
12:12:45GodEater_crap - it's impossible to get those files back from rapidshare
12:12:56Nico_Ppondlife: thanks :)
12:12:59Nico_PJdGordon: hi
12:16:27Nico_Pin case you didn't see last time, the wiki page report is already outdated... I got backwards skipping
12:18:28 Nick fxb__ is now known as fxb (
12:19:00Nico_PGodEater_: I think most of the files hosted on rapidhsare are too big for the wiki
12:19:16Nico_PI can host them on a webspace of mine though
12:19:40 Quit barrywardell ()
12:20:17 Join barrywardell [0] (
12:21:02 Quit Soap (Remote closed the connection)
12:21:28GodEater_also me
12:21:35GodEater_but you'd have to be able to download them in the first place
12:21:37GodEater_which I can't
12:21:41 Quit barrywardell (Client Quit)
12:22:18Nico_PI've just downloaded the firmware
12:29:15 Join ddalton [0] (
12:29:24 Part ddalton
12:30:07GodEater_ <−− interesting WinCE info. Wonder if we can do similar with the Gigabeat S
12:30:16PaulJamjust aout of couriosity: would it be possible to adjust the speed ot the RTC chip in the h300?
12:34:18LinusNyou mean calibrate it?
12:36:13TMMGodEater_, yeah! lets port rockbox to winCE :)
12:36:15LinusNno, you can't calibrate it, afaik
12:36:25PaulJamok, thank you.
12:36:52LinusNthe h300 uses the cheap solution for the clock source, so it isn't very accurate
12:37:15GodEater_TMM: I'm thinking more of using the WinCE bootloader to bootstrap Rockbox :)
12:37:54TMMGodEater_, don't the people have bootloaders for pretty much all ipaqs?
12:38:13 Join shodanX [0] (
12:39:08GodEater_they do
12:39:39GodEater_brb - work calls :(
12:52:10 Join obo [0] (n=obo@rockbox/developer/obo)
12:53:33JdGordonBagder: you round? can you delete an account on fs?
13:05:07 Join Thundercloud [0] (
13:08:51***Saving seen data "./dancer.seen"
13:14:09 Join pondlif1 [0] (
13:14:30 Part pondlif1 ("disconnected has pondlife")
13:17:25 Join mf0102 [0] (n=michi@
13:19:04 Join Arathis [0] (
13:22:30 Quit XavierGr (Nick collision from services.)
13:22:31 Join XavierGr_ [0] (
13:27:38 Quit pondlife (Read error: 110 (Connection timed out))
13:30:23 Join agm3nt [0] (
13:36:53 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:38:00 Quit thegeek_ (Read error: 110 (Connection timed out))
13:54:58 Quit iamben (Read error: 104 (Connection reset by peer))
13:57:37Nico_PGodEater_: I've uploaded files to the wiki or my webspace. Now none require rapidshare
13:59:38GodEater_Nico_P: thank you sir. You're a gentleman, and a scholar.
14:03:11 Join kubiix [0] (
14:06:06 Join jba [0] (
14:11:30 Quit kubiix (Read error: 104 (Connection reset by peer))
14:15:41LinusNJdGordon: which account? Nuri?
14:20:17 Quit jba ("Leaving")
14:23:15JdGordonhe's stopped though
14:26:39 Quit juxtap (Read error: 110 (Connection timed out))
14:28:10 Join leftright [0] (i=c41b5afa@gateway/web/cgi-irc/
14:29:58TMMGodEater_, 'scholar' what kind of compliment is that?
14:30:11GodEater_a very British one
14:30:30leftrightHi, could a dev please look at implementing Patch 7898, which fixes a Insert bug, and also 7911 which enables one to create a playlist with playlist stopped
14:32:37*leftright wonders who he can prod woth a few beers
14:34:14leftrightaha a taker
14:34:28JdGordondid someone say beer?
14:34:37leftrightit works
14:35:13*JdGordon opens those taks
14:35:16*petur has plenty, find something else :p
14:35:26JdGordonwas waiting for petur to speak up :D
14:35:52leftrightthe magic word summons all spirits
14:36:22LinusNleftright: regarding 7898, the current "Insert next" beaviour is the intended one
14:36:41LinusNfrom the manual: "Add track(s) immediately after current playing track, no matter what else has been inserted."
14:38:17leftrightthanks Linus, I will re-verify that behaviour when inserting more than one track
14:39:31JdGordonand 7911 is also the intended behaviour..
14:39:42JdGordonalthough, im not sure i agree with both of them
14:41:46 Quit lee-qid ("aufwiederbyebientotsayonara")
14:42:50*amiconn thinks that neither svn nor the fix in 7898 is correct behaviour
14:42:51leftrightthere is a change of playlist creating philosophy, so I guess it needs to be discussed, although I like the idea of being able to create a playlist without it starting playback immediately
14:43:12TMMwe could import sqlite :)
14:43:12amiconnI would expect 1 6 7 8 9 5 4 2 3
14:43:24LinusNamiconn: me too, actuallt
14:43:32 Join Benoitb [0] (
14:43:39pixelmatry again ;)
14:43:54TMMno sqlite? :)
14:44:14GodEater_ <−− implement on the forums please
14:44:34*JdGordon wonders how 5 could be before 4?
14:44:47JdGordon1 4 7 8 9 5 6 2 3 makes sense to me
14:45:02amiconnBecause "insert next" always inserts after the current (1 in this case)
14:45:39amiconnOr is intended to...
14:48:06 Nick parafin|away is now known as parafin (
14:48:07preglowthis ipod guy on flyspray is cracking me up
14:50:10JdGordonim surpised he didnt reopen it..
14:50:23JdGordoni sure would have if someone did that to me :p
14:50:30TMMhow come no 3 hours can go by in this channel without someone mentioning the i word? ;)
14:50:41JdGordonwhich i word?
14:50:46GodEater_oh dear
14:50:54*GodEater_ chides JdGordon
14:51:08JdGordondid /me miss something?
14:51:35GodEater_read preglow's last line ?
14:51:50*GodEater_ claps
14:52:05*JdGordon wonders how we could discuss rockbox without mentioning its biggest target?
14:52:08n1sI would expect the same order as amiconn...
14:52:21JdGordonthat order doesnt make sense to me...
14:52:31GodEater_I also agree with amiconn's ordering
14:52:46 Quit leftright ("CGI:IRC (Ping timeout)")
14:52:53JdGordonmust be a northern hemisphere thing then?
14:53:11GodEater_write it down on a bit of paper JdGordon
14:53:42n1sJdGordon: must be :-) however the order that tasks says is in svn makes no sense at all
14:53:45JdGordoninsert next insters after that last thing inserted, and should do it in a block if its a playlist/dir
14:54:04GodEater_no insert next inserts after what's PLAYING
14:54:19GodEater_insert inserts after what was inserted last
14:54:26n1sJdGordon: I see insert nex as inserting after the current track and "insert" as inserting after last inserted
14:54:41n1sapparently I agree with godeater too :_)
14:54:42JdGordonoh. misread LinusN's quote from the manual... ok that order makes more sense
14:54:48GodEater_the only one I don't understand is insert last
14:54:54JdGordonalthough adding in blocks does make sense if its a dir
14:55:19n1sGodEater_: append to the playlist
14:55:36GodEater_so why is it not called "Append", since you're not inserting ? :)
14:55:57LinusNGodEater_: good point
14:55:57n1sbecause of the Queue/Insert terminology perhaps
14:55:57LinusNit should definitely be append
14:56:04 Quit agm3nt (Read error: 104 (Connection reset by peer))
14:56:15GodEater_n1s: that terminolody is *clearly* bad, since it's confusing
14:56:21n1swhat should "queue last" be called then?
14:56:29LinusN"append queued"
14:56:34LinusNor something
14:56:48n1sok, I can live with that :-)
14:57:56GodEater_are the insert points independent for queue and insert ?
14:58:03 Join barrywardell [0] (i=c101acbc@gateway/web/cgi-irc/
15:00:32preglowi never remember how the insert stuff works
15:00:40preglowseems kind of unintuitive
15:00:46GodEater_I never use it myself
15:00:48preglowbut then again, it's not the only part of rockbox that is :>
15:01:01JdGordonunfortunatly its not
15:02:47JdGordonwhat about if one song is being added, we open the playlist viewer with it already selected to be moved around?
15:03:01JdGordonand possibly do that for dirplay/playlist also?
15:04:26*JdGordon wonders if we can get a low-mem version of the playlist viewer going if the plugin buffer cant be stolen
15:06:16*amiconn thinks the insert/queue stuff is pretty intuitive
15:06:54 Part Benoitb ("Kopete 0.12.5 :")
15:07:06preglowwhat's the difference between insert and queue?
15:07:15JdGordonqueue removes t after being played
15:07:19preglowwhy, that's not intuitive at all
15:07:32preglowit's impossible to figure that out from
15:07:36preglowfrom the menu entries alone
15:07:39pixelmathat's the same as a lot other media player do
15:07:51preglowi guess i just don't use playlists enough, then
15:07:55preglowwhich is true, since i don't use them
15:07:57GodEater_but even they don't tell you that
15:08:03GodEater_you have to work it out on your own
15:08:50crwlis there a way to make one file (or more) come next even if playmode is shuffle?
15:08:53***Saving seen data "./dancer.seen"
15:09:40n1spreglow: I implemented your idea about changin midiplayer to synth one voice to a buffer instead of one sample, it got a bit faster but i now have to memcpy that buffer to the output buffer so that dma will work
15:09:59preglown1s: and now it's more friendly to asm optimize as well
15:11:51 Join rvvs89_ [0] (n=rvvs89@unaffiliated/rvvs89)
15:12:14n1spreglow: I have'n done much with that but the fact that the same voice is several times made it possible to break out some stuff of one the synthing loops
15:13:31preglown1s: btw, why don't you work directly in the end buffer?
15:14:10 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
15:14:11n1spreglow: because end buffer is accessed by dma and on coldfire the dma can't access iram
15:14:15 Join kubiix [0] (
15:14:28n1sthe pcm_play_data function uses dma
15:14:28preglown1s: right, i keep forgetting this isn't a codec :)
15:14:51preglowhmm, it can't access dma? i thought the first 64kb could be accessed by dma
15:15:02n1swhen it's turned into a codec I'll make sure to fix that
15:15:07preglowyou can't
15:15:18preglowwhen it's a codec, you work directly in iram and let playback.c take care of the rest
15:15:40n1spreglow: well, that's true but 48 of those are used by the core and there is no way to tell the linker where to put the buffer
15:16:21preglown1s: i wonder if core needs dmaable iram
15:16:31n1sI actually got _one_ build working with a 4K buffer in iram but then i cahnged something small and it wouldn't work anymore
15:16:58preglowcould any core devs comment on this?
15:19:20 Quit petur ("reboot")
15:19:50 Join bluebrother [0] (i=w27PB21u@rockbox/staff/bluebrother)
15:20:25preglowbtw: why don't we support embedded cue sheetS?
15:20:59 Join webguest61 [0] (i=29e1f27c@gateway/web/cgi-irc/
15:21:09 Quit webguest61 (Client Quit)
15:21:35 Join Bene^da [0] (n=bla@
15:21:55|Rain|has anyone else had problems with the tag database rebuilding when it shouldn't, ultimately resulting in duplicate entries for every song?
15:22:01 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
15:22:55 Join kubiix [0] (
15:24:00pondlife|Rain|: Do you have both dircache enabled and database in RAM enabled?
15:24:09 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
15:24:39 Quit barrywardell ("CGI:IRC (Ping timeout)")
15:25:36 Join petur [0] (n=petur@rockbox/developer/petur)
15:25:47LinusNpreglow: we currently map the iram as 64k dma-capable ram followed by 32k non-dma ram, which means that the linker will use dma-capable ram for non-dma stuff
15:26:03preglowLinusN: sure, but we could change that, right?
15:26:13LinusNwe could map it the other way, so that the stacks and core data structures end up in non-dma ram
15:26:29preglowdoes core use any dma on coldfire?
15:27:27linuxstbpreglow: I looked at embedded cuesheets (in FLAC), and couldn't decide which of the two standards to support - foobar helpfully both embeds the entire cuesheet as a vorbis comment, and also creates the native FLAC cuesheet metadata structure, with the individual track metadata as vorbis comments. There's a forum thread and/or flyspray task where I discussed this with an interested user.
15:28:20LinusNpreglow: the pcm playback is in the core
15:29:12JdGordonlinuxstb: isnt our thing to only support the standard?
15:29:19LinusNthe entire dsp engine is in the core, isn't it?
15:29:24linuxstbJdGordon: Yes, but which one? ;)
15:29:37JdGordonnative FLAC cuesheet metadata structure sounds more standard...
15:29:39|Rain|pondlife: I'm using an e200, which I don't believe has the 'database in RAM' option
15:29:49|Rain|(because it's flash-based)
15:29:50pondlifeAh, ok
15:30:17preglowlinuxstb: yeah, foobar has retarded support for that
15:30:26preglowLinusN: it is
15:30:42preglowlinuxstb: i'd use the native flac one
15:31:14preglowLinusN: so, all we use dma for is audio? no ata?
15:31:27LinusNpreglow: true
15:31:28linuxstbpreglow: So would I, but I seem to recall there being some reason why that wasn't obviously the best thing to do. I forget why though...
15:32:58linuxstbHere's the forum discussion -
15:33:59JdGordongo with whichever is easier to implement?
15:35:19linuxstbThat would be the embedded .cue file - we just use the existing parser and point it to the section of the .flac file with the cuesheet.
15:35:51JdGordonso add that one, then add the other?
15:36:01 Join ArchFool [0] (
15:36:09*ArchFool says, |}{|
15:36:19JdGordonits 2 faces...
15:36:24JdGordonno a vase...
15:36:31ArchFoolOr... an hourglass on its side. :)
15:36:39linuxstbI would prefer not to support multiple ways of doing the same thing - then you get problems when a file has both (such as anything foobar creates...)
15:37:10JdGordonflip a coin then?
15:37:24linuxstbOr not do anything, as I never embed cuesheets...
15:37:34TMMworks for me (r)
15:37:37preglowembedding cue sheets is really sweet
15:38:00TMMI always failed to see the point to be honest
15:38:09TMMwhy not just split the damn file up into tracks?
15:38:12LinusNpreglow: well, we do use dma for the lcd on the h300, but it uses regular DRAM
15:38:45linuxstbTMM: I like the idea that anyone can take any audio file, no matter what format, drop it onto a Rockbox player, and play it.
15:38:49 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
15:38:51preglowTMM: some people prefer one file to several
15:39:04TMMwell, yeah, but I just wondered what the point is
15:39:04preglowTMM: i used to, but i just stopped doing it because if issues like this
15:39:38TMMI'd agree that rockbox should just play whatever you throw at it, if it's not DRM'd :)
15:40:04*ArchFool tweaks his RockBox settings... I hope this works!
15:40:21preglowit seems foobar's behaviour is to convert the binary flac cuesheet to a cue text file, then store that in vorbis tags
15:40:26preglowsweet god, how bloody redundant
15:40:51linuxstbpreglow: Are you playing with foobar now?
15:41:11preglowTMM: also, you need cue sheets to retain sub-indices
15:41:14linuxstbI'm curious to know what it does if you edit a track name...
15:41:14preglowlinuxstb: no
15:41:19preglowlinuxstb: not even in windows
15:41:46linuxstb... or give it a file with conflicting cuesheets
15:41:56 Join webguest9975 [0] (i=29e1f27c@gateway/web/cgi-irc/
15:42:53webguest9975Can I record on an iPod 3g?
15:44:25 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
15:44:54rvvs89_webguest9975: Why don't you ask Apple?
15:45:19 Quit petur ("moving...")
15:45:48 Nick Tanuva|off is now known as Tanuva (n=tanuva@
15:45:52 Join _kch_ [0] (
15:46:27preglowlinuxstb: bah, sorry i asked, how forgot how tedious this stuff becomes :)
15:47:40Nico_Plinuxstb: do you intend on implementing code to extract an embedded cuesheet ?
15:47:55 Quit webguest9975 ("CGI:IRC (Ping timeout)")
15:48:33 Quit _kch_ (Client Quit)
15:48:41 Join petur [0] (n=petur@rockbox/developer/petur)
15:49:02 Quit petur (Client Quit)
15:49:25 Nick Gnu47|Asleep is now known as Gnu47 (
15:51:38 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
15:51:55preglowi wonder how many lines of code are needed to decode flac cue to our cue struct
15:52:04preglowsupporting both methods might work nicely
15:52:16preglowparsing the vorbis tag one should be done with existing code
15:52:31 Quit jhulst ("Konversation terminated!")
15:53:03Nico_PI plan on modifying the cue code to store cues as MoB handles once playback is stable
15:53:27Nico_Pbest to to is probably to store the cue struct once it's been parsed
15:54:01Nico_Pthat way the setting and buffer_alloc can go away :)
15:54:03ArchFoolNico! I have a son called Nico! :)
15:54:37Nico_PArchFool: is he just called "Nico" ? My real first name is Nicolas
15:54:48 Join PaulJam_ [0] (
15:56:00ArchFoolNico's real name is Nicholas. :) He's 1/2 Mexican.
15:56:40ArchFoolMmmm... FLAC on my Sansa!
15:57:27Bene^dahi , I just tried the FS #5960 patch , has anyone tested this because JdGordon's patch doesn' t seem to work
15:58:36JdGordondoesnt work? or needs resync?
15:58:46*ArchFool wonders how many people here have had nocturnal transmissions.
15:59:26Bene^dafolders with the ROCKBOX_DATABASE_IGNORE_FOLDER file are added to the database
15:59:40JdGordonoh ok..
15:59:49 Quit CaptainSquid ("Miranda IM!")
16:00:20Nico_Pare there any volunteer devs to test my MoB code on a not too powerul SWCODEC target ?
16:00:22 Join rasher [0] (n=rasher@rockbox/developer/rasher)
16:00:37*Nico_P only has a gigabeat... not a very good indication
16:00:56*JdGordon 's sansa is wiling to play guienea pig
16:00:56ArchFoolDepends. What's a MoB? What's a SWCODEC? ;)
16:01:23Nico_PArchFool: I said "devs" because I don't want to explain it all now ;)
16:01:33*ArchFool nods his head.
16:01:37*ArchFool bobs his head happily.
16:01:52Nico_Palso there is no benefit to users currently
16:01:55n1sNico_P: i could test
16:02:04 Join scorche|w [0] (
16:02:05JdGordonNico_P: can yo do a e20 build for me? or pass me a diff?
16:02:06ArchFoolI'm a developer. I just usually develop in Perl. ;)
16:02:07 Join barrywardell [0] (i=c101acbc@gateway/web/cgi-irc/
16:02:20Nico_PJdGordon, n1s: the gitweb interface allows downloading a snapshot
16:02:38Nico_Pclick "snapshot"
16:03:00preglowNico_P: sure, is nano good enough?
16:03:10Nico_Ppreglow: of course
16:03:12preglowoh, forget that, windows :/
16:03:15preglowcompiler is in linux
16:03:39*JdGordon can do builds for people once the snapshot is downloaded
16:04:10Nico_PJdGordon: if the download is too slow I can provide a patch
16:04:16JdGordonits done
16:05:14preglowJdGordon: whip me up a quick nano build and i'll checkie out when i've had a shower
16:05:40JdGordonanyone else want a build?
16:05:45n1sJdGordon: I'll take a h330 please :-)
16:06:00preglown1s: btw, how much faster did it go?
16:06:44*JdGordon loves his quad core
16:07:00*ArchFool only has two cores. =snif=
16:07:06*preglow has one :/
16:07:11n1spreglow: with the loop optimizations it made possible the number of skips on my primary test file dropped from 12 to 2 but that also makes it a pretty lousy test file now :-)
16:07:14*ArchFool dumps core.
16:07:30preglown1s: wow, that's pretty significant
16:07:41*preglow concludes he's still good when it comes to dsp code :>
16:08:01 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
16:08:23JdGordonunless of course it doesnt want to work :(
16:08:42n1spreglow: yep, it' sa nice speed boost, I guess it will show better results with heavier files thoug
16:08:57 Join kubiix [0] (
16:09:39JdGordonpreglow: n1s:">
16:09:51Nico_Phmm looks like I recently caused a regression somewhere in the skipping
16:09:52n1sJdGordon: got it thanks
16:10:08ArchFoolpreglow... is that like a glow-plug on a diesel? =blink=
16:10:09preglowJdGordon: cheers
16:10:11pixelmaJdGordon: ok, I would try an M5 build...
16:10:25*JdGordon adds e200 and m5 in a min
16:10:52*Nico_P feels he is going to get flooded with "it doesn't work" or "it crashed" comments
16:10:55pixelmanice compile service here ;)
16:11:26n1sNico_P: playback resuming works, it just started playing :-)
16:11:31preglowArchFool: i really didn't know of anything that meant "preglow" when i started using this nick, i just used to call myself "glow" and i liked the sound of "preglow"
16:11:38ArchFoolgetcher red hot rockbox builds! FResh off the press!
16:11:44preglowbut yeah, preglow is the kind of thing you apply on a diesiel engine car to make it start :P
16:11:48pixelmaNico_P: easily... should I try on my Ondio too? :P
16:12:03JdGordonpixelma: ok, m5 is up also
16:12:21Nico_Ppixelma: hehe, no need :)
16:12:53 Quit PaulJam (Read error: 110 (Connection timed out))
16:13:23JdGordonbah, my sansa has gone walk-abouts.. ill test it out tomorow and let you know (prob in the logs)
16:14:01 Quit ArchFool ("p00f!")
16:14:41 Part LinusN
16:15:44n1sNico_P: some light skipping and seeking backwards and forwards works :-)
16:16:01Nico_Pn1s: does it feel more or less responsive than the SVN code ?
16:17:01n1sNico_P: well it is bloody instant on trackskip when it doesn't need to access the disk
16:17:11 Quit tchan ("WeeChat 0.2.7-dev")
16:17:19 Join Arathis_ [0] (
16:17:26 Join przemhb [0] (
16:17:30JdGordongnite all
16:17:33 Quit JdGordon ("Konversation terminated!")
16:17:37n1sbut I haven't noticed any delay in svn either so I would say as good as or better
16:17:49 Quit Arathis (Read error: 110 (Connection timed out))
16:17:52pixelmaNico_P: are there restrictions in the format?
16:17:56Nico_Pcool :) I was afraid it would be super slow on H300
16:18:09*barrywardell can test on sansa
16:18:12Nico_Ppixelma: no, but switching codecs will most probably fail
16:18:20barrywardelldata abort :(
16:18:31pixelmawell... tried aac first and it hung instantly
16:18:50Nico_Pah... well I got a segfault with WAV... I think I know why
16:19:06Nico_Pn1s: what codec are you using ?
16:19:14 Quit Zagor ("Client exiting")
16:19:16pixelmamp3 works but it rebuffers on trackchange even if the next track should be still in the buffer
16:19:24pixelma(skipping forward)
16:19:33n1sNico_P: so far vorbis but I put some other stuff in the playlist
16:19:51 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
16:19:59pixelmaI can feel the hd spin and it also shows disk activity in my wps
16:20:07Nico_Ppixelma: it rebuffers if the track wasn't buffered. it can happen because the buffering loop isn't too good right now
16:20:39Nico_Ppixelma: so far I haven't really tried to optimise disk spins
16:21:30przemhbHi all
16:21:30pixelmarebuffers on every trackchange while skipping forward with just standard length mp3s (I know it wouldn't happen that often with svn)
16:22:10n1shmm, I got some really weird playlist viewer/ what's actually playing inconsistency...
16:25:07n1sdunno if it has to do with this
16:25:35Nico_Pn1s: did you change the playlist ?
16:25:43Nico_Pif so it's expected
16:25:48n1sah, ok
16:26:11n1sthe only shorten file I have results in "paperclip mode" :-)
16:26:25pixelmaNico_P: same happens after a seek and skipping backwards even if it's just to the beginning of the track. Don't know if it's got to do with the M5 only having 16MB of RAM (only playing an album of mp3s currently)
16:26:40Nico_PI get a segfault with a WAV file on sim... strange because it used to work fine
16:27:08Nico_Ppixelma: yes, it's normal. the disk usage isn't very smart yet.
16:28:06pixelmawell it looks to me as if it just starts buffering again, thought that is exactly what it should not do
16:28:11przemhbbarrywardell: have you tried finding out H10's tuner control signals in hardware or have you disassembled OF?
16:28:50barrywardelli tried a little but couldn't get it to work
16:29:07barrywardellit's the same hw as the h300
16:30:02przemhbbut you have tried via OF disassembly, is that right?
16:30:05barrywardellso we should just be able to adapt that driver
16:30:26barrywardellno, no OF disassm for radio
16:31:11przemhbfrom what I have founf out we've got the same chip, the same driver, but different GPIO pin which drives BUS_ENABLE of the TEA5767
16:31:40przemhband different switching of playback source
16:32:33barrywardelland different i2c
16:32:45przemhbso if I'm right we have to find out where signals: BUS_ENABLE for radio and A,B for multiplexer are located, right?
16:33:16przemhbthe i2c already has it's driver, so that's not a problem I believe
16:33:20barrywardelland make adapt the i2c(easy)
16:33:59przemhbI am thinking about hardware approach
16:34:23pixelmaNico_P: I have the impression that the beginning of the track isn't played correctly (while buffering) ... yes ... listening to a live album confirms that because it isn't gapless. I don't hear a pause but the next track starts not at the beginning but a tad bit later
16:34:32barrywardelli think we need to find bus_enable first
16:34:55barrywardellmultiplexer can come later
16:35:00Nico_Ppixelma: yeah I think I noticed that too, but only very recently...
16:35:09Nico_Pit's new I think
16:35:22przemhbis there a plug that enables setting individual I/O's of the PP5020 for the H10?
16:35:36pondlifeNico_P: I'm about to test using JdGordon's H300 build
16:35:50przemhbrockbox plug-in
16:36:01barrywardellmemmon plugin is one option
16:36:02pixelmaNico_P: well it's the first time I try :)
16:36:14barrywardellyou can change any memory address
16:37:15przemhbhow can I do this?
16:37:44*Nico_P tries git-bisect to find the cause of all these recent regressions
16:39:06przemhbI would have to have it working on H10; I would disassemble hardware and measure BUS_ENABLE pin of the tuner chip when setting I/O's
16:39:38pondlifeNico_P: It seems to be keeping the audio buffer completely full. Or are the debug screens not being updated?
16:40:03Nico_Ppondlife: what's the audio buffer ?
16:40:18Nico_Pis it the file/main buffer ?
16:40:35pondlifeGo into Debug > View audio thread
16:40:38Nico_Ppondlife: the debug screen is probably wrong
16:40:45preglowNico_P: do you still use a max file number, btw?
16:40:47przemhbthank you - it may be helpful
16:41:06barrywardellprzembh: there's no specific gpio testing plugin
16:41:07Nico_Ppreglow: the "tracks" array now has 128 elements
16:41:16preglowsounds pretty reasonable
16:41:38barrywardellbut it would be helpful to have one
16:41:51preglowNico_P: what size is that struct?
16:41:59przemhbone more question: where can I find memmory addresses of the H10's PP5020 GPIO ports?
16:42:04pondlifeNico_P: Do you want bug reports yet?
16:42:11pondlifeIf so, how do you want them!
16:42:45Nico_Ppondlife: I think I'll wait a bit before asking for bug reports... I have enough in my head :)
16:42:46barrywardellcheck firmware/export/pp5020.h
16:42:54pondlifeOK, thought you might
16:43:03pondlifeIt's working ok mainly
16:43:10barrywardellthere's a list of what i've found so far in the wiki too
16:43:15Nico_Ppreglow: struct track_info is 24 bytes
16:43:20preglownot bad
16:43:47pondlifeI've got 38 tracks buffered here...
16:44:38przemhbI think I have all the necessary info at the moment; I will try to find out the signals mapping and let you know
16:44:41preglowit just sped through five tracks here...
16:45:02pondlifeYes, here too
16:45:13preglownow it "codec failure"s its way through tons of spcs, so no, it's not bug free :)
16:45:15przemhb(but it may take some time because I'm quite busy)
16:45:43przemhbThank you barrywardel for your help.
16:45:45 Join outside [0] (i=nat-021a@
16:45:49Nico_Ppreglow: if you make it change codecs it will fail
16:45:52Nico_Pdid you ?
16:46:21pixelmabut meta data is displayed correctly (also the next track's), and yes I have it skip half of a track too (still the live album, all mp3)
16:46:51barrywardellprzemhb: good luck.i hope you figure it out
16:46:55pixelmawithout doing something myself just listening for a while, didn't touch any controls
16:47:00preglowNico_P: i did
16:49:41przemhbthank you, bye, bye
16:49:55 Quit przemhb ("ChatZilla [Firefox]")
16:52:52 Join petur [0] (n=petur@rockbox/developer/petur)
16:53:45 Quit pondlife (Read error: 104 (Connection reset by peer))
16:57:37Nico_Pgit bisect is a wonderful tool
16:59:53*Nico_P feels stupid for having handed out a version with regressions
17:02:21Nico_Pbarrywardell: you said you had a data abort with the mob code you tried ?
17:03:51pixelmaNico_P: actually on track change with that life album something weird happens: at the end of the first track it starts playing from the same position in time of the next track basically I can only listen to every other song, depending on which is longer that can lead to other effects and sometimes playback just stops. Don't know if it's important but these are vbr mp3s
17:04:25Nico_Ppixelma: would you mind getting the latest snapshot and seeing if the problem persists ?
17:04:55 Quit barrywardell ("CGI:IRC (Ping timeout)")
17:05:05pixelmado I need git to checkout?
17:05:14Nico_Ppixelma:;h=mob, then click snapshot to get a zip file
17:05:56Nico_Ptake the one from the latest commit of course
17:06:48desowin"Bump version for new release with fixed OF loading on e200." <- it still doesn't work here with e280
17:07:00 Join toffe82 [0] (
17:08:19 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
17:08:57***Saving seen data "./dancer.seen"
17:09:38 Nick Arathis_ is now known as Arathis (
17:13:45 Quit TMM ("Ik ga weg")
17:14:01 Join freqmod [0] (
17:14:46Nico_Ppixelma: any better ?
17:15:05pixelmajust started compiling, downloading took a while
17:16:53Nico_Pthe awkward buffering will probably still be there but you shouldn't have the start of track problems and gapless should be fine
17:17:26pixelmaget some warnings in playback.c
17:18:31Nico_Poh ? I'll try a coldfire build
17:18:49 Quit cendres (Read error: 104 (Connection reset by peer))
17:18:49 Join miepchen^schlaf [0] (
17:18:55linuxstbdesowin: The SVN bootloader still doesn't work for you?
17:19:04 Join ashes [0] (n=ashes@2001:5c0:8fff:ffff:0:0:0:23)
17:19:06desowinoops, sorry, didn't update PP5022.mi4
17:19:29linuxstbdesowin: Are you compiling yourself? I don't think the download server has been updated with the new files yet.
17:19:48desowinyes, and I found problem, it works now, thanks anyway
17:20:08linuxstbnp. Thanks for the confirmation it works ;)
17:20:28desowinbut on amd64 there's bunch of warning
17:22:11 Join seablue [0] (
17:22:56 Join barrywardell [0] (
17:24:28 Join Frazz [0] (
17:25:47linuxstbdesowin: Yes, barrywardell mentioned those yesterday.
17:26:20 Quit tedrock (Client Quit)
17:26:56 Quit linuxstb ("Client Exiting")
17:27:35barrywardelldesowin: so the new sansapatcher works OK for you now?
17:28:28 Join tedrock [0] (
17:28:47desowinbarrywardell: yes, I forgot to update PP5022.mi4 in sansapatcher dir first (then I told here that it doesn't work), but when I updated PP5022.mi4 it works like a charm
17:29:16barrywardellI have emailed binaries for windows, linux, osx and linux amd64 to Bagder, so hopefully he'll have a chance to put them up on the download server soon
17:29:37 Quit PaulJam_ (Read error: 104 (Connection reset by peer))
17:30:16 Join PaulJam_ [0] (
17:32:39 Join adam__ [0] (
17:33:08adam__hey guys, I've got an iPod Nano and i'm trying to figure out if it is compatible with RockBox
17:33:19adam__it's the red one that came with AIDs
17:34:05bluebrotherthat's a 2nd G
17:34:07GodEater_it's not
17:36:41scorche|wto clarify, it is a second den that is not compatible :)
17:37:11bluebrotherhmm, we should really extend IdentifyPlayer and link it from the front page.
17:40:31 Quit Arathis ("Bye, bye")
17:43:32 Join kugel [0] (
17:46:13 Join Arathis [0] (
17:47:04pixelmaNico_P: yes, it's better. I can now listen to all songs and gaplessly
17:47:38Nico_Pcool :) something was very wrong with the previous version
17:48:26Nico_PI think I'll soon have track transitions ready
17:48:32Nico_Pcodec transitions I mean
17:48:56 Quit pondlife ("disconnected has pondlife")
17:49:35pixelmastill rebuffering on every track change though (at the end of the current track) even if the next should be completely in buffer
17:49:52Nico_Pit should but it isn't
17:50:15Nico_Psometimes the next few tracks end up being in the buffer and skipping will be instantaneous
17:50:18pixelmaok, then keep up the work :)
17:50:23Nico_PI am :)
17:52:35 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
17:52:40adam__i want my ipod to work
17:52:50adam__is there anything at all that i can do to make it work with rockbox
17:53:19 Quit barrywardell (Remote closed the connection)
17:53:31bluebrotherunless you find a way to run custom code on that player, no
17:53:41scorche|wyou can work on porting rockbox to your device, but it requires a lot of time and effort
17:53:48scorche|w(and quite a bit of a "clue"
17:54:58pixelmascorche|w: by the way... nice collection of pics. I found most of them earlier but didn't notice until today that you three were on quite a few... :)
17:55:17 Join XavierGr [0] (n=xavier@
17:55:41scorche|wpixelma: yeah...all the ones with a group photo, we are in somewhere :)
17:56:01scorche|wi should post some of the ones where we are hiding behind people so you can have more of a challenge ;)
17:56:38pixelmamakes a nice "where are the rockbox devs?" game
17:56:54 Nick adam__ is now known as eclecticus (
17:57:28scorche|wi was tempted to circle them in red, but that would ruin the fun
17:57:44scorche|wperhaps i should have it as the "answers"
17:58:08eclecticusscorche|w = work i assume?
17:58:11 Quit Gnu47 ("You know you'll miss me :P")
17:58:14delYsidAM I completely missing something or does autofire_delay_settings in fireworks.c have one element too much?
17:58:25Nico_Pscorche|w: could you give the link to your photos ?
17:58:59delYsidI suggest reducing the array size to 15, removing the duplicate "300ms" entry and reducing autofire_delay_values to 15 as well (as well as the call to set_int().
17:59:10pixelmaNico_P: same as yesterday">
17:59:15scorche|wNico_P: none of markun's are up there (not that he took many), but it is at scorche.cleansoap...bah
18:00:09Nico_Phmm this time even the page doesn't load :(
18:01:38pixelmaweird, no problem here (but it was also ok yesterday)
18:03:31eclecticusquestion - what irc client do you all use
18:03:53 Quit freqmod (Remote closed the connection)
18:04:30Nico_Pcodec transitions ! :D
18:04:59pondlifeNico_P: But I only just got the last one built...:/
18:05:18Nico_Ppondlife: don't worry, it's still buggy :p
18:05:42 Quit XavierGr_ (Read error: 110 (Connection timed out))
18:06:53 Quit Isolinear (Read error: 110 (Connection timed out))
18:07:18 Quit ivan` ("Coyote finally caught me")
18:07:26GodEater_sweet jesus that David-NYC bloke looks scarey!
18:07:47delYsidCan someone verify P#7933 and commit it please?
18:07:56pondlifeGodEater_: Where?
18:08:27GodEater_the NYC meeting thread
18:08:34 Join ivan` [0] (n=ivan`
18:08:35GodEater_PS - you coming on the 18th ?
18:08:58GodEater_lowest turnout yet then
18:09:03pondlifeJust you?
18:09:04GodEater_looks like it's just me and linuxstb so far
18:09:22pixelmaNico_P: should I paste the warnings in playback.c somewhere or did you already take a look yourself/plan to look into it later?
18:09:44Nico_Ppixelma: I've looked but didn't find an easy way to fix them
18:10:05pondlifeAt least it's playing through an album ok-ish now (so far)
18:10:38pondlife37 tracks buffered at the mo too
18:11:00Nico_Ppondlife: how do you know ? the debug screen ?
18:11:12Nico_PI'm not sure if it can be trusted
18:11:24pondlifeIt's about right for the test album I'm using
18:11:42pondlifeThe buffer usage indicator seems to be going down properly now too
18:12:14pondlifeAt track transitions, the number buffered drops by one too... so I'm fairly sure it's right
18:12:52Nico_Pyeah if the debug screen reads the tracks array it can be close from being right
18:13:24Nico_Pbut the data might not be buffered on disk yet, that's the buffering API's work and the debug screen doesn't know about it
18:13:26pixelmapondlife: in "view audio thread"?
18:13:34pondlifeAh, ok
18:13:47pondlifeNico_P: That's good, isn't it
18:13:54 Quit eclecticus ("ChatZilla [Firefox]")
18:13:58Nico_PIMO it is
18:14:05Nico_PI should so a buffering debug screen
18:14:10pondlifeWe could have a view buffering thread option too
18:14:27Nico_PI'm not sure how it would access the data though
18:14:43pondlifeYou'd need to add some bufdebug_... API entries
18:14:48Nico_P(it's meant to be private to buffering.c)
18:15:01 Quit petur ("work->home")
18:15:03Nico_Pmakes sense
18:15:11pondlifeThe vars would definitely remain private
18:15:23pondlifePubCon??? It's DevConPub!
18:17:12scorche|wfixed :)
18:17:32GodEater_and how is it you're not attending next week then pondlife ?
18:17:36GodEater_it was YOU that picked the date
18:17:44pondlifeI know, but I never promised
18:17:55pondlifemrs pondlife came up with a better offer..
18:17:56pixelmapondlife: for me it still rebuffering on every track change
18:18:18pondlifepixelma: I've not checked that yet... just want to see it working
18:18:25*GodEater_ wants to see the excuse note from mrs pondlife
18:18:55pondlifepondlife can't come out today because....
18:19:04GodEater_he's forgotten his gym kit
18:19:14pondlifethe dog ate his Rockbox
18:20:09pondlifeNico_P: That annoying code to keep the progress bar running all the way to the end of a track... has it been removed?
18:20:14pixelmapondlife: I thought you were saying earlier that it works correctly for you, maybe misread then
18:20:29Nico_Ppondlife: yes, and you can see it's missing... I don't really know what do do
18:20:30pondlifeWell, it plays through an album without doing weird things like before
18:20:49pondlifeNico_P: As long as that's intentional, we can find a way to fix somehow.
18:20:49Nico_Pactually I simply disabled it
18:21:23pondlifeHmm, why can't we maintain it? It needed a global variable to hold the last track length, right?
18:21:32Nico_Pok, codec change now appears to be safe
18:21:55pixelmapondlife: ah, ok - you meant that
18:21:58pondlifepixelma: It's not buffering on every track change, but it does buffer much earlier than before
18:22:07Nico_Ppondlife: actually it accessed the prev track's mp3entry directly. this isn't good with mob so it could be replaced with global vars
18:22:31pondlifeYes, we were going to have a single variable just for this purpose I thought.
18:22:56Nico_Ptwo actually, because it needs prev_track_elapsed and prev_track_len
18:23:09 Quit obo ("bye")
18:23:23pondlifeOK, but only one track's worth
18:23:24 Join bluey- [0] (
18:23:51Nico_Pit's for when the playback engine has changed tracks but not the WPS
18:23:57pondlifepixelma: Are you using H300 too?
18:24:39pondlifeI'm not seeing much more disk activity here... assuming the status bar indicator is reliable
18:25:10pondlifeMind you, I'm not seeking or skipping at all
18:25:50 Join Gnu47 [0] (i=Gnu47@
18:26:15pixelmapondlife: no, I'm testing on M5, but looks like you are right. Now I'm watching the audio thread it was just by chance at a track change but disk is spinning more often (overall). No skipping or seeking here too - will pay a bit more attention
18:26:58pondlifeNico_P: Is there an easy way an SVN user (i.e. me) could update from your git "mob" head? Or will I need git?
18:27:27Nico_Ppondlife: you'd need git
18:27:41pondlifeOK, I actually have it... do I start with a git clone?
18:27:52pondlifeOr checkout
18:28:51Nico_Pthis should set you up with a working git-svn repo on which you can use git svn rebase
18:29:11Nico_Pif you only want to keep track of the mob branch you can simply clone though
18:29:41pondlifeI just want a low-bandwidth way to get your changes here
18:29:43*Nico_P pushes his changes for codec ransitions
18:30:15Nico_Ppondlife: you have no choice but to do the initial clone which will download the entire repo :/
18:30:23pondlifeNo problem
18:30:28pondlifeAs long as it's only once
18:30:58Nico_Pyes it's only once... then "git pull origin mob" (I think) should give you the latest updates
18:31:28Nico_Pproblem is I have to rebase mob on svn from time to time
18:31:29pondlifeorigin/mob or origin mob ?
18:31:50 Quit homielowe (Read error: 110 (Connection timed out))
18:31:56Nico_Porigin mob I think... "pull the mob branch from the origin remote"
18:32:10pixelmaI thought I could save the diff and patch that into my local mob tree
18:32:12Nico_Pbut maybe origin/mob works too
18:32:39 Join Crash91 [0] (n=evil91@
18:32:41Nico_Ppixelma: you can do that too but IMHO it's less practical
18:33:00*Nico_P tries cloning his own repo
18:33:14 Join Stupid0 [0] (
18:34:00pixelmawell... but I don't need to get git for it
18:34:31Nico_Ppondlife: once you have your local git repo you can make changes to it, commit them and push them to me
18:34:41Nico_Por I can pull from you
18:35:00pondlifeI don't think I'll be doing more than testing
18:35:11pondlifeMight put some use cases on the wiki
18:35:24Nico_Pgood idea :)
18:35:35Nico_PI need to start looking at the bugs in flyspray
18:35:46pondlifeEven better idea :)
18:36:10Nico_PI'm pretty confident at least some of them solved themselves
18:37:28*pondlife awaits the big commit before updating Flyspray ;)
18:37:46Nico_Pthe big SVN commit ?
18:38:35pondlifeYes... at this rate we'll have MOB in there in no time..
18:38:42pondlife(no pressure, again) :)
18:39:37pondlifep.s. I'm getting build warnings, do you know about them?
18:39:55 Join remi__ [0] (
18:39:56Nico_Pthe type-punned pointer dereferencing business ?
18:41:08Nico_Pyeah well I have no idea how to make these disappear
18:41:19Nico_Pother than by changing a compile flag
18:41:26remi__will it be long till rockbox supports ipod nano 2g?
18:43:20 Join petur [0] (n=petur@rockbox/developer/petur)
18:43:22pondlifeNico_P: Did you push your codec stuff yet?
18:43:45pondlifeI'm still getting "Already up-to-date."..
18:43:55Nico_Pthe three commits from the last 20 minutes
18:44:13Nico_Pmaybe you are... the latest is "Rename getptr to bufgetcodec and make it ensure the ... "
18:46:03pondlifeGuess I am, git log includes that
18:46:06 Join Pyromancer [0] (
18:46:23 Join The-Compiler [0] (i=534d0ed7@gateway/web/cgi-irc/
18:47:44The-CompilerI have some question about the c200: What does work, what not? What's the cpu of the c200?
18:48:38pondlifeNico_P: MOB seems much more responsive to me... might be my imagination though
18:48:47Nico_Pone of my M4B podcasts makes the sim segfault, both SVN and MoB
18:49:03Nico_Ppondlife: to me too, but I was afraid it was just my imagination :)
18:49:08pondlifegdb rockboxui.exe
18:49:18Nico_Pyeah it's in the mp4 metadata parser
18:49:24pondlifeOr, pop it on Flyspray for now and ignore
18:50:02 Quit The-Compiler (Client Quit)
18:50:29pondlifeOops, crashed by skipping back 2 tracks
18:50:39pixelmaand I just wanted to answer him...
18:51:47Nico_Ppondlife: really ? I've been skipping back plenty more than that without any problems ? anything peculiar with the tracks ?
18:51:54pondlifeShort tracks?
18:52:01pondlifeAbout 10 seconds each
18:52:09pondlifeIt span up the disk, then locked
18:52:29Nico_Pcould you send the tracks to me ? i don't have such short ones
18:52:54Nico_P...but I just got a lockup :/
18:53:00pondlifeMe too (again)
18:54:18 Quit Stupid0 ("Bersirc 2.2, for external use only. [ - Open Source IRC ]")
18:54:27Nico_Phmm mine wasn't my fault, it was the previously mentioned M4B podcast crash
18:54:48 Quit mf0102 ("Verlassend")
18:57:47 Part pondlife ("disconnected has pondlife")
19:02:00 Quit pixelma (Read error: 104 (Connection reset by peer))
19:08:02 Part Crash91
19:09:00***Saving seen data "./dancer.seen"
19:09:21n1sNico_P: the type-punned pointer warnings comes from casting to different types, for example this cast (unsigned char **)&id3
19:09:53 Join ilgufo [0] (
19:09:53Nico_Pn1s: yes, but I've tried t solve it without success
19:10:59 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
19:15:22*amiconn wonders whether he should torture-test MoB together with the .mod patch on a 64MB iPod video
19:15:28n1sNico_P: you can also do it the less clean way and simply add -fno-strict-aliasing to the compiler flags, we do that for gcc 4.x builds anyway
19:15:48amiconnMight be that it would have to buffer all 80 .mods I have in that folder...
19:16:05Nico_Pamiconn: please do :)
19:16:56amiconnThe code throws type-punned pointer warnigns even with gcc 3.x?
19:17:03amiconnThat's usually a bad sign iirc
19:17:50*amiconn wonders what would be necessary to test the thing
19:17:55rasherJust test it with the .sid codec..
19:18:18 Join Lear [0] (i=chatzill@rockbox/developer/lear)
19:19:39pixelmaI let my M5 just play now and hold it in the hand, the disk spins up every 4..5 minutes when in the wps, doesn't have to do with track changes. That seems a bit too often to me (currently 128kbps mp3s, got 13,3 MB of free buffer) and yes - typed that one-handedly
19:20:18 Join bertrik [0] (
19:21:02Nico_Ppixelma: I have done nothing to optimise disk spins yet
19:21:08amiconnI mean how to fetch the source tree in order to test it
19:21:34Nico_Pamiconn: you can get a zip snapshot or use git to clone the repo
19:21:36pixelmayou can download a snapshot as I did
19:21:50*amiconn wants to stay far away from git
19:22:54 Quit Entasis (Read error: 104 (Connection reset by peer))
19:22:58 Quit rvvs89_ (Read error: 104 (Connection reset by peer))
19:23:04 Join Entasis [0] (
19:23:24 Join rvvs89_ [0] (n=rvvs89@unaffiliated/rvvs89)
19:23:27 Join Fraser [0] (
19:23:28 Join RaRe` [0] (
19:24:23 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
19:24:23 Join TMM [0] (
19:24:59n1sNico_P: if I add a lot of sids to the playlist then stop and resume again it will not start playing, in the audio debug thread track count stops at 128
19:25:22Nico_Pyeah 128 is the max number
19:25:26n1sbtw I'm using the version jdgordon built earlier
19:25:51n1sbut it seems like it won't start playing if more than 128 tracks fit in the buffer
19:26:00Nico_Pn1s: maybe update, the version JdGordon built had some regressions but I don't know if they would affect what you're doing
19:26:00amiconnWhy is there a maximum number?
19:26:23Nico_Pamiconn: because I need to store handle IDs somewhere
19:26:32preglowso that the player doesn't try to buffer 5000 sid files?
19:26:33rasherIt's a language course bug waiting to happen!
19:26:34preglowthat would take ages
19:26:48amiconnI thought that MoB would use a linked list within the buffer...
19:26:53Nico_Pplayback.c can't access the linked list directly
19:27:36rasherpreglow: but after that, you could practically remove the harddisk
19:29:11n1smeh the neverending looping of sid tracks is anoyying
19:29:19*amiconn thought we would get rid of that static limit stuff...
19:29:39pixelmaNico_P: my comment about disk spinning was more directed to pondlife (though I see now that he left) because he said it wouldn't spin up so often for him - ok the H300 has more RAM but it should also be more often than svn for him... or there must be another difference
19:29:48preglowrasher: well, yeah, but i don't WANT it to buffer 5000 sids, that's overkill
19:30:06preglowthe disk would be spinning for a pretty long time
19:30:42preglowso the static limit is a good thing, if you ask me, not that i think it should necessarily be enforced with a static array
19:30:53TMMyou could do it in the background while it was doing other stuff anyway, if the disk spins up, just keep it spinning for a bit longer?
19:31:03preglowTMM: battery life is an issue
19:31:10remi__I was just thinking.. is it possible to remove the nano 2g firmware and then replace it with 1g?
19:31:10preglowthe drive sucks down currenmt
19:31:11preglowcurrent <-
19:31:18preglowremi__: hell no
19:31:22TMMif the lists' in memory, after that you'll save battery
19:31:59preglowTMM: assuming i want to listen to all of them, the fact that i keep songs in the same dir doesn't automatically imply that i always want to listen to as many of them as possible
19:32:18preglowTMM: if i want to listen to one, then another one in another dir, 4999 sids just got buffered during a thirty second spinup for no reason
19:32:46TMMI suppose that's true
19:32:47preglowand thirty seconds is probably an optimistic estimate
19:33:06preglowthanks to metadata parsin
19:33:30preglowNico_P: when does metadata parsin happen with mob? during disk load or afteR?
19:33:32n1spreglow: you should use that "Insert" function we talkt about earlier
19:33:43Nico_Ppreglow: during
19:33:51preglown1s: i don't like playlists
19:33:53rasherWhich reminds me.. someone with a clue and a heart for sids should have a look at FS #7891
19:34:19preglowrasher: that database is a text file :/
19:34:32preglowrasher: parsing a bloody text file for every sid load sounds "interesting"
19:34:43rasherthere could be a conversion script to a more friendly layout
19:35:10rasheralso, the md5s are sorted, so it'd "only" be a binary search :)
19:35:14preglowor the bloody idiots could start tagging their files, like the rest of the world
19:35:17amiconnNico_P: I get no warnings during build (core and codecs)...
19:35:27rasherpreglow: I suggest ogg/sid
19:35:34Nico_Pamiconn: the warnings only appear on coldfire builds
19:35:41preglowi really wouldn't use ogg for a non-streaming codec
19:35:51*amiconn is building for ipod video
19:36:08preglowat least not without asking someone skilled with ogg
19:36:14Nico_PI didn't know about them until pixelma mentioned them
19:36:20rasherMy point was just that sids by design can't be tagged, afaiu
19:36:22n1samiconn: strict aliasing is disabled for gcc => 4 in configure
19:36:42 Part remi__ ("Ex-Chat")
19:36:58 Quit Frazz (Read error: 110 (Connection timed out))
19:37:00amiconnn1s: Yeah, probably due to some bogus warnings... if 3.x gives those, I think they're not bogus....
19:37:39preglowrasher: i really hate any scheme that needs to modify files
19:38:19 Join lee-qid [0] (
19:38:34preglowfor pure text data base we'd need an md5 routine too, that kinda sucks...
19:38:52amiconnWhy is that?
19:39:03amiconnAre those text database pre-made using md5?
19:39:09rasherThe best part is that the fingerprint isn't simply an md5sum of the .sid.. but something else
19:39:13preglow19:35 < rasher> also, the md5s are sorted, so it'd "only" be a binary search :)
19:39:23 Join Daenyth [0] (
19:39:32rasherActually, I'm not even entirely sure it's md5, just has the same size
19:40:00DaenythI just updated my ipod video to the latest build, from an older one, and now it won't start correctly. It just stalls at the splash screen with the rockbox logo
19:40:18*amiconn thinks that if a format can't be tagged, then leave it that way
19:40:32 Quit RaRe (Read error: 110 (Connection timed out))
19:40:40 Quit bluey- ("Leaving")
19:40:44Daenythhas anyone heard of anything that might cause that?
19:41:12 Join Domonoky [0] (
19:41:51 Join merbanan [0] (n=banan@
19:43:11DaenythI just redid the bootloader with ipodpatcher, let me see if that helps...
19:44:39n1sDaenyth: is it a 5/5.5G/video?
19:44:55n1syup, that has been reported
19:45:07Daenythdowngrade the firmware then?
19:45:09Daenythwhich version is safe?
19:45:51n1sI don't know but jhMikeS who reworked the pcm code last week (?) suspected it might be the cause so go back about a week or so
19:45:55DaenythI used
19:46:21 Quit ilgufo (Connection timed out)
19:46:30n1sthat would be the current build
19:46:36amiconnNico_P: Did you try casting to (void**) in order to avoid the warnings?
19:46:45Daenythok, lemme try one from september
19:46:54Nico_Pamiconn: yes, but maybe I didn't do it quite right
19:46:56amiconnBlah, I'll just try
19:48:10rasherpreglow: ah yes, it is md5, but it's... "special":
19:48:55Daenyththe 9/30 build locks also
19:49:30amiconnNico_P: How far from svn is that tree, apart from your changes?
19:50:03Nico_Pamiconn: it's based on r15048
19:50:35*Nico_P should rebase
19:50:57*amiconn wanted to check the binary size changes
19:51:23 Join webguest98 [0] (i=55e3986a@gateway/web/cgi-irc/
19:51:24amiconnAtm it looks like it becomes a bit smaller, but there were size-increasing changes in svn after 15048
19:51:27 Quit webguest98 (Client Quit)
19:51:37 Join zicho [0] (
19:51:51 Join webguest97 [0] (i=55e3986a@gateway/web/cgi-irc/
19:51:54 Quit webguest97 (Client Quit)
19:52:49Nico_Pamiconn: what are you comparing ? svn versus the build from the snapshot ?
19:53:17Nico_Pso there might be a slight increase
19:53:41Nico_PI can compare quite easily but I have uncommited changes right now
19:54:48 Join ilgufo [0] (
19:56:08Daenythn1s: the 9/11 build is still locking it.. I'll try again tonight but I don't have more time right now
19:58:01Nico_Pamiconn: the mob build is 1688 less for me
19:58:07 Join Crash91 [0] (n=evil91@
19:58:08 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
19:58:39*amiconn silenced the "type punned pointer..." warnings locally
19:58:56pixelmaDaenyth: do you know how old your bootloader is=
19:58:58Nico_Pnice :) how did you do ?
19:59:06amiconnJust cast to (void *) instead of (unsigned char **)
19:59:15amiconnNote the single asterisk
19:59:56amiconnbufgetdata() knows what it expects after all, and (void*) is the fit-all universal pointer
20:00:27pixelmaDaenyth: sorry, overlooked your statement that you updated it too, ognore
20:00:40amiconnBut there's one more warning in playback.c for coldfire
20:00:50amiconnplayback.c:2751: warning: 'fd' might be used uninitialized in this function
20:01:51Nico_Pamiconn: so for example in bufgetid3 it should be (void *)id3 ?
20:02:22amiconnNo, (void *)&id3
20:03:19 Join agm3nt [0] (
20:03:24Nico_Pso a void * can be an ** ?
20:03:40Nico_Pyeah I guess
20:04:07amiconnvoid* is just a pointer that can point to anything
20:04:19 Quit flynux (
20:04:19 Quit Ave (
20:04:23amiconnSo in this case it points to something that is a pointer itself
20:05:06 Quit agm3nt (Client Quit)
20:05:34 Join agm3nt [0] (
20:05:41Nico_Pamiconn: should I declare bufgetdata as requiring a void ** ?
20:05:56 Join thegeek [0] (
20:05:59amiconnLeave it as it is
20:06:08Nico_Pit's just that unsigned char ** feels dirty
20:09:06NJoinflynux [0] (n=flynux@2a01:38:0:0:0:0:0:1)
20:09:12amiconnThen make it require a void* (again, one asterisk), and change the assignment to *data = &buffer[h->ridx];
20:09:37amiconnHmm, that might not work...
20:10:25Nico_Pand it feels even more dirty IMHO
20:10:26amiconn(void**) might in fact work, with the same assignment
20:12:43 Join Llorean [0] (
20:12:59 Quit ilgufo (Nick collision from services.)
20:13:02 Join gufo [0] (
20:13:34 Quit gufo (Connection reset by peer)
20:14:26preglowamiconn: if you're interC[C[Cested, my idea for rendering midis turned out to be faster than the current one, according to n1s
20:14:37preglowwhat the hell happened in "interested" there...
20:15:04amiconnNico_P: This compiles, and after thinking a bit more, looks like it makes sense
20:15:18amiconnThe resulting binaries have the same size to the byte
20:15:33Nico_Pamiconn: what's "this" ?
20:16:00amiconnssize_t bufgetdata(int handle_id, size_t size, void **data);
20:16:19amiconn...and then doing *data = &buffer[h->ridx]; in bufgetdata()
20:16:30amiconn...and calling it with a cast to (void*)
20:16:36preglown1s: btw, why don't you commit it ?
20:17:01NJoinAve [0] (
20:17:35 Join elinenbe [0] (n=elinenbe@
20:18:14amiconnThe void **data reflects that the returned pointer can point to anything, depending on how the caller interprets it
20:19:06preglown1s: is there a wiki page for the midi plug?
20:19:22*amiconn would guess PluginMidiplay
20:19:24elinenbeunrelated, but does anyone have have experience using a bash script to ssh to a gateway machine and then again ssh'ing from there?
20:20:02amiconnNah, slightly wrong: PluginMidiPlay
20:20:23scorche|welinenbe: yes, it is unrelated...
20:20:46elinenbejust looking for some advice.
20:21:03 Join barrywardell [0] (
20:21:05Lloreanelinenbe: Try reading the channel topic. It might have a suggestion.
20:21:16preglown1s: are there any fancier patch sets around than this?
20:21:18 Join linuxstb [0] (
20:21:39elinenbeLlorean: I don't understand :-)
20:22:18Lloreanelinenbe: 1) There are guidelines. 2) There's a part of the topic that says "Please direct offtopic..."
20:23:40Nico_Pthis compiles fine
20:24:35amiconnLooks okay to me
20:24:54amiconnJust keep in mind that you can't do pointer arithmetics with void* in standard C
20:25:15amiconnIn gcc you can, but that's a gnu extension, and should be avoided imho
20:25:19Nico_Psounds logical
20:25:54amiconn(gcc treats void* as char* for pointer arithmetics)
20:26:13preglowisn't that ansi c spec?
20:26:50 Join Jchord [0] (
20:27:23Nico_Pamiconn: btw, why do you want to stay far from git ?
20:27:43amiconn(under "5 Extensions to the C Language Family")
20:27:51 Join ilgufo [0] (
20:28:33preglowis that so
20:29:30 Quit ilgufo (Remote closed the connection)
20:32:33amiconn(1) I prefer one central repository. (2) git is (reportetly) slow on cygwin
20:33:14scorche|wamiconn: you can have one central repository with git..
20:33:18Nico_Pah yes you're on cygwin
20:33:30scorche|win fact, that is how we would set it up...
20:33:45Nico_PI hear a windows native implementation is in the works and already quite advanced
20:34:07scorche|wNico_P: i could always ask the git guys i met at google
20:34:14amiconnYes, you can. With svn, you must
20:34:44Nico_Pscorche|w: who did you meet ?
20:35:04amiconnThe 128 track limit is too low
20:35:12scorche|wthere were 2 git guys that i remember...talked more with a guy named "shawn"...would have to see the wiki to find out more
20:35:28amiconnI now have 84 MODs buffered, which just need around 1/3 of the whole buffer
20:35:43amiconn...and it took very few seconds to buffer them
20:35:58Nico_Pamiconn: struct track_info is only 24 bytes so we can probably set the limit to way more than 128
20:36:17Nico_Pamiconn: they might not actually be buffered though
20:36:41amiconnI would prefer no imposed limit, but if that's unavoidable, the code must be able to handle the case that the IDs are used up before the whole ram is filled
20:36:46Nico_Pscorche|w: Shawn O. Pearce and Johannes Schindelin (from the google git page)
20:37:16Nico_Pamiconn: doesn't it ?
20:37:21scorche|wamiconn: it would basically be the same, but you would have the additional benefits of being able to pull a branch from another dev, and other stuff like that...but you actually dont *have* to use talk there was about git, and one project said they had a lot of older devs that were very leery, but they set it up exactly like it was with svn, and it worked great...they gradually started using the other features
20:37:45scorche|wnotes are on the jot wiki that i can look up later..
20:38:00Nico_Pscorche|w: is that wiki accessible ?
20:39:16amiconnscorche: I'm not sure whether I would like this ability being present, even if I wouldn't use it...
20:39:26 Join ilgufo [0] (
20:39:45scorche|wNico_P: by a few of us...we have been using Bagder's credentials...when markun gets back, we will likely talk a bit about what we learned, what can be improved, etc...i may write a wiki page about it
20:39:54amiconnNico_P: I dunno; I don't have that many small tracks, but iirc pondlife reported a problem with resume not working
20:39:55scorche|wamiconn: if you didnt use it, what would it matter?
20:40:15amiconnscorche: Others who would use it could mess up something
20:40:34amiconnThis git feature sounds too complex to me to be trustworthy
20:41:08rasherIt works for the Linux kernel.. surely it won't mess up Rockbox
20:42:10zichoDoes anyone know if the Gigabeat F runs Linux in its default firmware?
20:42:28linuxstbzicho: Yes, it does.
20:42:37zichoAh, nice.
20:42:48linuxstbNot really...
20:43:21linuxstbBy all accounts (from people who have used it), it's one of the worst firmwares known to mankind.
20:43:24amiconnNico_P: In fact it was n1s who reported that problem
20:43:30 Quit barrywardell ("Konversation terminated!")
20:44:06zicholinuxstb: Haha, I didnt use it. I changed to rockbox immediately.
20:44:16zichoBut it was ugly as hell as far as i saw.
20:46:06amiconnHmm, I'll have to check something when switching back to an svn build... *possibly* the MoB code doesn't yield enough/properly/whatever
20:46:41amiconnScrolling in the wps is slow as hell on G5.5 (*might* be due to peakmeters - that's what I have to compare)
20:47:26 Quit ilgufo (Nick collision from services.)
20:47:29 Join gufo [0] (
20:47:32Nico_Pamiconn: it's very possible
20:47:38amiconnHmm, skipping back spins up the disk
20:47:57amiconnI know that svn also does this
20:48:01preglownow always
20:48:04preglownot <-
20:48:08Nico_Pyes, that's because track handles are released immediately after the track is finished playing
20:48:22preglowand i'd love to retain that
20:48:29Nico_PI should probably be a bit more conservative about that
20:48:31preglowif data is still there, it should be used
20:48:56amiconnNico_P: The ymight need one more possible state
20:49:35Nico_Phandles ?
20:49:52amiconnIf they're available for overwriting but not yet actually overwritten
20:50:05Nico_Pyes I see what you mean
20:51:01 Join Redbreva [0] (
20:52:09 Quit gufo (Read error: 104 (Connection reset by peer))
20:52:25 Join Zagor [0] (
20:53:04 Quit alienbiker99 (Read error: 110 (Connection timed out))
20:53:36preglown1s: the patch set from the wiki page is not very nice...
20:54:21preglowalso, are you sure the linear interpolation is done properly? kind of doesn't sound like it is
20:54:36 Quit petur ("switch")
20:54:55 Join petur [0] (n=petur@rockbox/developer/petur)
20:56:14 Join gufo [0] (
20:57:00preglowthere's some weird kind of high frequency ringing here
20:57:18amiconnNico_P: SVN has the same slow-scroll problem, so the MoB code seems to be okay
20:57:56Nico_Pamiconn: btw, have you noticed a change in responsiveness for track skipping and seeking ?
20:58:43amiconnI didn't try seeking at all though
20:59:25amiconn...only skipping
21:01:21n1spreglow: I am actually not very familiar with the techical details so it might very well be broken, also I haven't found any other patch sets
21:01:32n1snot that I have been looking...
21:02:09preglowsomething is definitely up with the sound
21:02:33 Join gufoo [0] (
21:02:44preglowalso, we need some reverb :P
21:03:40n1spreglow: I actually introduced a bug with my last commit but I can't hear it and it only affects the last fraction of a tick that doesn't fit the buffer so I don't think that's it
21:04:11preglownah, surely not
21:04:15n1sI'd say we need to be realtime on pp before reverb :-)
21:04:15preglowwhy don't you commit the new stuff, btw?
21:05:15n1sdunno, the memcpy feels hacky and it's probably not needed for others than coldfire so I have to #ifdef a bit
21:05:48n1sI think I'll do it tomorrow
21:06:12preglowlike linusn said, making core use non-dma memory before dma memory should only be a matter of hacking plugin scripts
21:06:18preglowlinker scripts
21:06:28preglowbesides, this is eventually going to be a codec where the output buffer is in iram anyway
21:06:52preglowthe memcpy is hacky, but only needed for coldfire
21:06:56preglowand not needed for long, hopefully
21:08:13n1sand as always I will need to find someone to test it for me as I still don't have any other target than my h300 but that is for tomorrow
21:08:42 Join tomtom [0] (
21:09:02***Saving seen data "./dancer.seen"
21:10:34 Join ilgufo [0] (
21:11:24preglowi've got a nano
21:11:39tomtomI've just heard of rockbox for the first time. I'm having some questions and I hope someone has a second for me
21:11:41preglowlooking at the interp function i can't see anything obviously wrong with it
21:11:48n1sI'll nag you when I have a patch then ;)
21:11:48preglowtomtom: just ask
21:11:58amiconnpreglow: codecs can't use iram for dma either. Only the core can
21:12:17amiconn...unless you mean that the buffer will reside in core iram
21:13:36 Join Soap [0] (n=Soap@rockbox/staff/soap)
21:13:37tomtomwhat formats of music can be played with an ipod "updated" with rockbox? can it really be filled with drm protected songs?
21:13:40amiconnI wonder whether swapping would be a good idea. It would mean that codecs and plugins could use dma, but the core in turn couldn't
21:13:49preglowtomtom: no drm
21:14:01preglowtomtom: we'll probably never support drm, we don't like it
21:14:09tomtombut wma itself is possible?
21:14:14 Quit zicho ("Lämnar")
21:14:18preglowtomtom: yup, but it's a work in progress
21:14:20 Quit ilgufo (Nick collision from services.)
21:14:22 Join gufetto [0] (
21:14:25amiconnThere's no way around that; only one of the 2 iram blocks of the coldfires supports dma
21:14:29 Join zicho [0] (
21:14:38scorche|wtomtom: and see SoundCodecs (wikipage) for supported codecs
21:14:56n1samiconn: do we use dma for iram in the core?
21:14:57amiconn5249 has 64KB of dma capable iram, and 32KB of non-dma capable iram. 5250 has 64KB of each type
21:14:58preglowamiconn: codecs have no need for dma since it passes the buffer on to the core anyway
21:15:09preglowamiconn: midiplay is a plugin, which is why it's an issue
21:15:29amiconnPlugins and codecs have the same issue... unless codecs also memcpy()
21:16:55tomtomis it - as a matter of principle - possible that apple's ipod will be made compatible with drm protected wma files ever or is this rather a pipe dream?
21:17:02n1sor maybe we could device some linkerscript hack to let us decide what it puts inside the 16kB of dma capable plugin/codec iram
21:17:31Lloreantomtom: If it ever happens, it won't be by way of Rockbox.
21:18:10amiconnn1s: That won't help the iaudios
21:18:12n1stomtom: that would probably be illegal in parts of the world etc.
21:18:29n1samiconn: ah do they have 64/64 split?
21:18:34preglowamiconn: codecs don't care, if any memcpy happens, it's the core doing it, and probably in the form of src point begin iram and dst pointer being ram in some process
21:18:51tomtomyes you're right, stupid idea...
21:19:19preglowLlorean: keep in mind that rockbox would probably end up in legal trouble if we ever DID implement drm support
21:19:25 Quit rvvs89_ (Client Quit)
21:19:25preglowtomtom: that was for you
21:19:35amiconnn1s: I stumbled upon that when digging for the reason why rockboy hung on X5...
21:19:43preglowdrm isn't very open source friendly
21:19:57n1sI wonder if any plugin or codec uses that extra 16 KB
21:20:01amiconnIt worked *by chance* on H1x0/H300, because the audio output buffer was in the first 16KB of plugin iram
21:20:21amiconnn1s: Not yet, but the core does neither
21:20:37amiconnIt just feels logical to do a fair split
21:21:04tomtomthat's propably right, preglow
21:21:04amiconn(unless it doesn't make sense like on archos)
21:22:20 Join webguest46 [0] (i=9fbf0c18@gateway/web/cgi-irc/
21:22:59amiconnn1s: How big buffers are we talking, btw?
21:23:08 Quit Jchord (Read error: 104 (Connection reset by peer))
21:23:23amiconnPerhaps they could be put on the stack. plugins run in the main thread, and the main stack is in core iram
21:23:35 Quit gufetto ("So Long, and Thanks For All the Fish -")
21:23:43 Join ilgufo [0] (
21:24:04 Quit gufo (Connection timed out)
21:24:18 Quit gufoo (Success)
21:27:40n1samiconn: we can choose pretty freely, currently the output buffer is 32kB but before that it was just 1
21:28:15n1sHow big is the stack?
21:29:06tomtomwhat do you think about everyone having his own codecs and restrictions for his players today? shoudln't every customer be able to decide what player he wants to fill with what music from what supplier?
21:30:00tomtomor do you support the politics of the big companies?
21:30:35n1stomtom: what are you getting at?
21:30:41 Part agm3nt
21:30:43BigBambitomtom: I don't really understand the question
21:30:53amiconnMain stack is 8KB but a fair bit is needed for other stuff
21:31:05amiconnI think 2KB buffer on stack would be affordable
21:31:34preglowtomtom: that is what most of us think
21:31:48preglowtomtom: and that is why we support a very great number of codecs
21:31:53n1samiconn: that will probably increase skipping, but it's definitely worth testing
21:32:06preglowtomtom: i can stuff all files i have on my computer straight onto a dap with rockbox, and it will all play as it is
21:32:21preglowtomtom: that includes mp3 files, aac files, vorbis files, flac wiles, wavpack files, wav files, spc files
21:32:25 Join massiveH [0] (
21:33:49 Join fxb_ [0] (
21:34:46tomtomyes of yourse rockbos supports as much as formats as possible but what I'm talking about is the approach of the industry
21:35:02preglowthe industry isn't something we can do anything about
21:35:23 Quit fxb (Nick collision from services.)
21:35:27 Nick fxb_ is now known as fxb (
21:35:35 Join _fxb [0] (
21:35:56 Quit fxb (Client Quit)
21:36:00 Nick _fxb is now known as fxb (
21:36:29BigBambitomtom: We just do our own thing
21:36:40 Join xnyhps [0] (
21:36:53*BigBambi is using the we losely, personally I do very little :)
21:37:00*BigBambi also can't spell
21:38:01tomtomyes that's right, of course we can't and your project including your ambitions are great but what I wanted to know is what you think about those harassments.
21:38:40amiconnpetur: funny bug...
21:39:10BigBambiWhich harassments? It'd be nice if the world were open and DRM less, but it isn't but rockbox exists
21:39:23Lloreantomtom: But, this channel is about #rockbox, both support and technical details. That sort of thing is off topic here.
21:39:44TMMtomtom: in my opinion: in the end, it's not going to be viable for most companies to build their own proprietary firmwares. Some companies will start shipping rockbox as their default firmware and their players will sell for a lot less, DRM free.
21:40:08*BigBambi points tomtom at #rockbox-community
21:40:21TMMtomtom: people keep forgetting that DRM is fucking EXPENSIVE :)
21:40:27tomtomthat would be great
21:40:57TMMbut yeah -community for this
21:41:12tomtomyou're right, let's stop here
21:41:29tomtom... and wait till rockbox is available for 6th generation^^
21:42:05BigBambiKeep on waiting
21:42:09Lloreantomtom: That discussion will never be on-topic for this channel.
21:42:34TMMtomtom: start hacking! or stop buying hardware with eaten apples on it
21:43:17TMMtomtom: that bite from the apple is a symbol for the part of the soul you lost when you bought it :)
21:45:10amiconnlinuxstb: around?
21:46:46linuxstbamiconn: Yes.
21:47:20amiconnlinuxstb: Just checking; maybe I'll soon have something that would need testing on a color...
21:47:55linuxstbOK. I should be around for a few hours.
21:49:09*amiconn wants to clean up the PP colour lcd drivers (color/nano and H10), and port the fast update to the H10
21:49:46amiconnIf it doesn't work on H10 out of the box, a register dump from a color might be helpful
21:49:50 Join Entasis_ [0] (
21:53:07 Quit Entasis (Read error: 104 (Connection reset by peer))
21:53:38 Quit webguest46 ("CGI:IRC (EOF)")
21:55:12 Join mistermuckle [0] (n=choffman@
21:56:38 Quit krazykit ("Connection reset by beer")
21:57:00 Part mistermuckle
21:57:24*petur wishes people would stop using his friday quit message
21:57:56*BigBambi changes his quit messqge...
21:58:17*petur takes an empty beer bottle...
21:58:25Nico_Ppetur: what is it ?
21:58:42*BigBambi hides
21:58:56petur"Connection reset by beer"
22:00:26 Quit desowin ("use linux")
22:00:35 Nick Tanuva is now known as Tanuva|off (n=tanuva@
22:00:38 Join barrywardell [0] (
22:00:42 Quit Bene^da ("get satisfied! • :: ««« (Gamers.IRC) »»» ::")
22:05:31 Join Rob2222 [0] (
22:05:34 Quit Rob222241 (Read error: 104 (Connection reset by peer))
22:07:08*amiconn fetches a beer
22:07:25 Quit spky (Read error: 110 (Connection timed out))
22:07:25 Quit lee-qid ("aufwiederbyebientotsayonara")
22:07:54*BigBambi envies amiconn, for I am beerless
22:09:01 Quit ilgufo (Read error: 110 (Connection timed out))
22:10:02 Quit BrianHV (Remote closed the connection)
22:10:43 Quit Crash91 ("Bye Bye!")
22:11:18 Join BrianHV [0] (
22:11:52 Quit amiconn (Nick collision from services.)
22:11:58 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
22:12:17 Quit barrywardell ()
22:13:19 Quit merbanan (Remote closed the connection)
22:15:26 Quit tomtom ()
22:19:45Bagdersansapatcher time
22:20:58tumuwould there be any idea in implementing zip/rar filter in front of codecs?
22:21:11Bagderin front of?
22:21:33tumuso that codecs itself don't need to have the support for archives
22:21:39*Domonoky sees there is a new release of sansapatcher.. is it only a new bootloader ?
22:22:13tumumainly for computer generated music files which usually are archived for space reasons
22:22:48 Quit hannesd ("Client suicide")
22:23:14linuxstbDomonoky: Yes. Nothing to do for rbutil.
22:23:21 Quit Lear ("ChatZilla [Firefox]")
22:23:54Domonokyonly the bootloader on has to be updated..
22:25:08Bagderno sansapatcher too
22:25:12tumui think the current codec api would work nicely even when dealing with archives
22:25:14n1stumu: yes there are some plans for a separate loader for some formats such as midi and xm
22:25:57 Join krazykit [0] (n=krazykit@
22:26:06linuxstbtumu: That would mean adding zip support to the file browser and other parts of Rockbox... (rar is closed and horrible...)
22:26:32tumulinuxstb, there is open source unrar available
22:26:41Bagderthat doesn't make the format open though
22:27:09TMMtumu: the license of the unrar utility is quite horrendous
22:27:09tumuwell, i'm not planning on having rockbox making the archives, so.. :)
22:27:22TMMtumu: very, very GPL incompatible
22:28:05Bagderrar is evil
22:28:17tumuanyways, licenses aside
22:28:18rasherIt's probably also prohibitively slow
22:28:31rasherIt'd make more sense to unpack on your computer
22:28:35rasherspace issues aside
22:28:50TMMthere is a pure GPL unrar utility as well afaik
22:28:54tumurar can be replaced with other as capable ones
22:29:02TMMreverse engineered, but it won't unpack everything made the last 3 years
22:29:07tumuso i don't consider it as an issue
22:29:25 Join sin613 [0] (
22:29:41Bagderlinuxstb: I got two mi4 files from barry for this update, do you happen to know what firmware.mi4 is in this context as compared to PP5022.mi4 ?
22:30:02linuxstbIs that the c200 bootloader?
22:30:18Bagderah, could be
22:30:21BagderI better ask him
22:30:26Zagorfirmware is c200 and PP5022 is e200 iirc
22:30:27linuxstbI'm guessing barry didn't build an e200r?
22:30:44Bagderit doesn't look like that
22:32:10amiconnlinuxstb: Care to test a patch for me on color?
22:32:26amiconnShould probably also be tested on nano...
22:33:06 Quit Fraser (Read error: 104 (Connection reset by peer))
22:33:11linuxstbamiconn: Sure.
22:33:18 Join Klevi [0] (
22:33:43amiconnI'm mainly interested in whether lcd works stable
22:33:51linuxstbIf you want to DCC it, send it to linuxstb_
22:33:52amiconnThere should also be a slight speedup
22:34:19linuxstbShould I build test_fps then?
22:34:24KleviIs there a way to unbrick a sansa that does nothing when you try to turn it on?
22:34:29jhMikeSamiconn: It appears that clock skipping when scaling is obligatory when using the COP at least on 5020. Otherwise it's super happy dual core there. :) I just need to add that ingredient. No freezes ever from any sort of core sync now.
22:35:47amiconnjhMikeS: Then why does the ipod firmware not need clock skipping when scaling?
22:36:21jhMikeSMaybe it's device-dependent? The H10 firmware does.
22:36:38 Quit Rob2222 (Read error: 104 (Connection reset by peer))
22:37:07jhMikeSI can say for sure it's that on H10 and it corrupts the RAM if you don't.
22:37:08amiconnDoesn't make much sense imho. Scaling is purely PP internal
22:37:49preglowclock skipping?
22:37:53jhMikeSIt's the only thing that ever causes a problem. If it stays boosted, there are no problems.
22:37:56amiconnlinuxstb: It would be great if you could run test_fps before & after
22:38:07krazykitKlevi, depends on what "nothing" means. have you tried putting it in recovery mode?
22:38:21amiconn(just to check whether my optimisations don't cause the opposite of what they should)
22:38:26KleviNothing, meaning that only the menu and button lights come on =/
22:38:27jhMikeSpreglow: the processor control registers let the cores lay idle for certain amounts of time
22:38:33linuxstbamiconn: OK, will do.
22:38:37KleviI cant get into diag mode
22:38:54 Join matsl [0] (
22:39:12preglowjhMikeS: and how does that relate to scaling?
22:39:46KleviI formatted my sansa completly by accident in class today, it was in my pocket.
22:39:58linuxstbKlevi: What do you mean by "diag mode" ?
22:40:05jhMikeSpreglow: H10 crashes if the COP is in use during switches. We don't see it because the COP is normally in idle mode. mpegplayer doesn't crash out because it only runs boosted.
22:40:29preglowwhat a piece of shit portalplayer is starting to sound like
22:40:38preglowthey probably don't even _have_ specs
22:40:51preglowthey just tweak around bugs as they go and force everyone to use their libs
22:41:04 Join vosque [0] (
22:41:19jhMikeShehe...they did correct this trouble. I thought it was the concurrent write in Peterson's algorthim so I tried Dekker's (which only writes turn upon unlock) but no change.
22:42:31KleviAm i SOL?
22:42:31krazykitKlevi, if it screwed up like that just in your pocked, my guess would be that it's hardware rather than software
22:43:02Klevino, it said format complete, right on the screen whe nI looked at it
22:43:10Klevithen it shut off.
22:43:15preglowjhMikeS: so you're getting closer with the dual core stuff?
22:43:30ZagorKlevi: I formatted my c200 just the other day. what's the problem?
22:43:33preglowjhMikeS: you have any special stress tests you use to see if things are up to par?
22:43:38jhMikeSpreglow: it works...period. I just need to make the scaling behave.
22:43:44KleviIt wont turn back on after formatting it.
22:43:52ZagorI even gave it fat32 :)
22:43:57preglowjhMikeS: cool, then we need to make people start talking about preemptive multitasking soon :P
22:44:03ZagorKlevi: are you on linux?
22:44:13KleviNo, I can be, but then i lack an internet connection
22:44:13jhMikeSpreglow: the dual-core queue_test plugins and a version of the dual-core SPC codec (which uses sems and events)
22:44:27KleviBut, your saying there is a way to fix this?
22:44:35jhMikeSpreglow:'s not far off from that
22:45:06preglowjhMikeS: i would imagine not, making your current code go preemptive is probably mostly a question of making rockbox apps/ code do threading properly
22:45:12jhMikeSI mean I can lock it by killing the spc thread from View OS Stacks, but that's it.
22:45:12ZagorKlevi: I don't know what state your player is in, so I can't tell. but I'd like to know what "fdisk -l /dev/sda" says in linux
22:45:24*amiconn is going to extend test_fps
22:45:40KleviIll try getting my laptop online, one miute
22:45:47 Join fm2 [0] (n=chatzill@
22:46:09Zagorthe sansa OF has a formatting option, but one would assume it doesn't brick the player...
22:46:21jhMikeSpreglow: I think most concern there would hardly be a concern with a "narrow" interface that dispatches and serializes things.
22:46:40fm2petur: hello petur. Why the need to explicitly declare the size of the array in fireworks.c?
22:47:14preglowand making people write proper multi-threading code, of course
22:47:16fm2Wouldn't a declaration with just [] do the same?
22:47:22preglowi've got things to learn in that department myself...
22:47:25*jhMikeS suggests int myarray[strlen(s)]; <== totally legal :)
22:47:46amiconnjhMikeS: Urgh, that would be runtime...
22:48:05jhMikeSwell, it's an alternative to _alloca anyway
22:48:38amiconnYes, it's even standard C99
22:48:50amiconnBut I would avoid it if not absolutely necessary
22:48:50preglowc99 supports dynamically sized local storage?
22:49:01preglowi love that feature
22:49:09preglowand it has so very little runtime overhead
22:49:11jhMikeSis a subtract from the sp a big deal?
22:49:33*linuxstb switches computers
22:49:38 Quit linuxstb ("ChatZilla [Firefox]")
22:49:46amiconnjhMikeS: The strlen() is the overhead...
22:50:15jhMikeSwell, if you use it anyhow it's not more. It does depend of course.
22:50:19 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
22:50:53amiconnThe subtract from SP is not, of course. It's just variable instead of constant
22:51:03fm2What do the => and ===> mean in the list box with the forums?
22:51:04 Quit seablue ("life, death, life, death")
22:51:18 Join Rob2222 [0] (
22:51:23 Join p3tur [0] (n=petur@rockbox/developer/petur)
22:52:44jhMikeSit would probably need sp+n for locals and fp=sp-size for the dynamic stuff. It could have a tiny bit more than the sp change.
22:52:53 Quit Klevi ("Quit messages. The unintentional channel spam of people closing IRC, realizing they have a life.")
22:52:59 Join Klevi [0] (
22:53:07Klevialright laptops on
22:53:12amiconnWe always comiple using -fomit-frame-pointer
22:53:24 Nick parafin is now known as parafin|away (
22:53:40jhMikeSprobably won't work here then unless an explicit switch off is possible per function
22:53:47*p3tur kicks his router
22:54:02fm2bluebrother: you're a forum moderator, right? Any comments about the purpose of => and ===>? Why the different length?
22:54:15 Quit petur (Nick collision from services.)
22:54:20jhMikeSpetur: A network router or a woodworking tool??
22:54:20 Nick p3tur is now known as petur (n=petur@rockbox/developer/petur)
22:54:28fm2p3tur: are you the same person as petur?
22:54:31linuxstbamiconn: Patch seems to work fine. test_fps with your patch (30MHz) - 19.5fps fullscreen and 75.5fps 1/4 screen. @80MHz - 51.5fps fullscreen and 202.0fps 1/4 screen.
22:54:47jhMikeSoh, yes..."p3tur" today...woops
22:55:17peturthis dlink router crashes when somebody downloads a big file from my pc :(
22:55:18jhMikeSnow he switched back again
22:55:53 Part Redbreva ("Out for the day...!")
22:55:54 Join haemmy [0] (n=stefan@
22:55:58amiconnlinuxstb: Ok, so basically no speed change
22:56:09ZagorKlevi: what happens when you turn on your sansa?
22:56:21amiconn(assuming the values in LcdFrameRate are still valid for the Color)
22:56:34KleviZagor, just a blank screen
22:56:45amiconnpetur: Is one of the PCs on wireless?
22:57:12amiconnlinuxstb: No crashes/freezes/hangs?
22:57:16 Part fm2 ("Nobody seems to like to talk to me today. 'll try next time.")
22:57:28Kleviby the way, Zagor what command did you want me to run with linux?
22:57:38linuxstbamiconn: I've just reverted your patch, and will run test_fps again.
22:58:31peturamiconn: yes, but not the one that hosts, and the client is from the internet, so no wireless involved except my laptop
22:58:34linuxstbamiconn: SVN is identical...
22:58:47ZagorKlevi: try connecting your sansa to usb, and run "tail -f /var/log/kern.log" to see if it registers
22:58:53 Join PhinnFort [0] (n=martin@
22:59:08Kleviwindows came up with USB device
22:59:14amiconnpetur: I had the same issue with a netgear (copying large files between wireless and wired clients crashed the router)
22:59:28amiconnI now replaced it... no crashes so far
22:59:42ZagorKlevi: how big?
22:59:49 Join _aLF [0] (
22:59:56 Quit xnyhps ("I had a question, but I can read faqs...")
22:59:57peturamiconn: different issue here - found a newer firmware, will disconnect soon ;)
22:59:57Kleviwouldnt recognize it.
23:00:09jhMikeSpreglow: I did do something to single core that's making it behave oddly after adding the alternate locking. Do doubt just a dumb error. It cropped up last time I majorly changed the patch around. I hate boring to track even if simple.
23:00:14Zagorwhat does kern.log say?
23:00:19jhMikeS*No doubt
23:00:47 Join barrywardell [0] (
23:00:52KleviZagor only wireless devices came up under that command.
23:01:10Klevinothing dealing with USB
23:01:22linuxstbamiconn: I've installed your patch again, and it seems stable.
23:01:36 Quit petur ("upgrading")
23:01:37Kleviill brb
23:02:05amiconnI got gfx glitches from applying the same idea to H10.. but there's a bug in svn already (I think)
23:02:27 Quit BrianHV (Remote closed the connection)
23:03:39preglowshut up!!"¤
23:03:50linuxstbbarrywardell: Can you build an e200r bootloader to go with the other two? It's not used by sansapatcher, and has a different mi4 encoding to the vanilla e200.
23:04:46 Join alienbiker99 [0] (
23:05:41 Join petur [0] (n=petur@rockbox/developer/petur)
23:06:37jhMikeSbeer beer beer beer beer!!
23:06:43 Quit Domonoky (Read error: 104 (Connection reset by peer))
23:07:34KleviZagor, you mentioned another command earlier, what was it?
23:07:40*jhMikeS shouldn't have done that...for his own sake too
23:07:41*preglow covers ears and sings
23:07:47ZagorKlevi: it assumed you got a usb connection
23:08:13Klevii do.
23:08:25ZagorKlevi: I thought you said it wasn't recognized?
23:08:27 Join BrianHV [0] (
23:08:41Kleviunder windows it isnr
23:09:02Kleviit comes up as USB Device. from there it does nothing
23:09:04***Saving seen data "./dancer.seen"
23:09:14ZagorKlevi: so you can't browse it?
23:09:28preglowZagor: how's usb going, by the way?
23:09:32*jhMikeS may put b-word refences in the kernel now :P
23:09:36KleviNothing happens, the backlight for the lcd doesnt come on either
23:09:47Zagorpreglow: pretty good. I'm starting with the ata/scsi translation code now.
23:09:57preglowZagor: really looking forward to this
23:10:03Zagorme too :)
23:10:12barrywardelllinuxstb: will do. I didn't realise it was different
23:10:23ZagorKlevi: do the buttons light up?
23:10:43Klevilsusb brings up Sansa Corp
23:10:53Klevi"Sandisk corp
23:10:59jhMikeSZagor: you took over the USB project?
23:11:22ZagorjhMikeS: well I sort of restarted it. I'm writing a new implementation from scratch.
23:12:26KleviTheres nothing I can really do here is there?
23:12:34*amiconn found the glitching problem
23:13:21ZagorKlevi: run "lsusb -v | less". look for the "endpoints" below the sandisk entry. you should have two "bulk" endpoints.
23:13:33Zagorif so, you should be able to mount it in linux too
23:13:52amiconn...or not :\
23:13:55barrywardelllinuxstb: sent to Bagder
23:14:01jhMikeSI take it any paragraphs-length writups on threading should probably be on a wiki page and not in comments in thread.c? Maybe a reference "See xxxx for details?"
23:14:16 Quit BigBambi ("Please insert girder")
23:14:23Bagdergot them
23:14:32 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
23:15:11ZagorjhMikeS: how about both? I know I would prefer an explanation in the code rather than hunting around on the web for it.
23:15:27Zagor...if I were digging around in the code, that is
23:15:43preglowi'd rather have it in the code
23:15:52Kleviso then, how would I mount it?
23:15:55jhMikeSZagor: That's why I wanted to point to it there. It really should have some writup on how it operates.
23:16:12Klevimount /dev/sda?
23:16:29ZagorKlevi: well do you have a /dev/sda device?
23:16:30linuxstbBut if it's too long, maybe just a URL in the code?
23:16:54jhMikeSZagor: there's also explanations on _how_ to implement certain routines so any target gets all the low-lag benefits and such.
23:17:01KleviYeah, thats what it used to come up as
23:17:41ZagorKlevi: then try "fdisk -l /dev/sda" and post the result on
23:18:00Klevibut if it isnt booting anything it wont be in fstab
23:18:12Zagorfstab has nothing to do with this
23:18:29KleviWell, if it isnt in there i cant mount it, thats why i said
23:19:18Klevi? no result..
23:19:30Zagor_no_ result? not even an error message?
23:19:35Klevinothing at all
23:22:21preglowjhMikeS: anyway, if you're feeling suicidal, there's no reason NOT to fix up rockbox code to do thread communication properly. after that's done, a quick preemptive threading showcase would be cool (and probably needed to convince some people) :P
23:22:23advcomp2019does anyone knows why rockbox is picking up the fm radio frequencies off frequency now
23:22:23Zagorah right, fdisk doesn't say anything
23:22:23ZagorKlevi: so what does "ls /dev/sd*" say?
23:22:23Bagdersansapatcher + sansa bootloaders updated
23:22:46 Join [IDC]Dragon [0] (i=0c182f0a@gateway/web/cgi-irc/
23:22:59KleviUnable to mount /dev/sda
23:22:59barrywardellyay :)
23:23:36zichoKlevi: Shouldnt it be like "/dev/sda1"?
23:23:39[IDC]Dragonhi guys!
23:23:43 Quit haemmy ()
23:23:52KleviI told you it isnt mounting at all
23:24:03Bagderlook an [IDC]Dragon! ;-)
23:24:04ZagorKlevi: that was not the output from 'ls /dev/sd*'
23:24:06KleviNormally, yes.... oh wait misread nick...
23:24:32[IDC]Dragondoes somebody know if the Archos 605 is somewhat of an open platform?
23:24:49BagderI believe the archos guys use linux
23:24:49[IDC]Dragon(it runs linux, so...)
23:24:57 Join homielowe [0] (
23:25:00Klevino, theres no such directory as "/dev/sda"
23:25:00Bagderbut I doubt they release their modules' source code
23:25:20 Quit petur ("*plop*")
23:25:24[IDC]Dragonit would be OK if I can compile for it
23:25:25Bagder[IDC]Dragon: that a dm320 target?
23:25:32ZagorKlevi: did you find any bulk endpoints in lsusb -v?
23:25:37[IDC]DragonI'm not sure
23:26:26[IDC]DragonI stubled over that device yesterday, it could be a nice portable browser
23:26:30 Nick Klevi is now known as Klevi-Dinner (
23:26:37Klevi-DinnerIll be back soon
23:26:39[IDC]Dragonsmaller than the Nokia 800
23:27:20[IDC]Dragonwith the same screen
23:27:38advcomp2019some fm frequencies are 0.05MHz and some are 0.1MHz off
23:27:38linuxstb[IDC]Dragon: You could try archopen, although their website isn't the clearest.
23:27:41rasher Wow, they really want your wallet.
23:28:05BagderI find no proper dissection on it
23:28:29amiconnlinuxstb: is your Color LCD type 0 or type 1?
23:28:45linuxstbamiconn: Good question. It's a type 1.
23:28:47jhMikeSpreglow: I think playback is the first code that will get extensive treatment there.
23:28:56preglowjhMikeS: it's probably what needs it most
23:29:25linuxstbamiconn: I don't know of anyone with a type 0, but if their ipod was sold as a "Photo" then the chances are high.
23:29:25[IDC]Dragonit's almost as small as an ipod touch, but has way better screen resolution
23:29:29jhMikeSit hardly survives as it is. _after_ the MoB patch of course which should ease things.
23:30:06amiconnlinuxstb: For type 1, it looks like a slight speedup is possible (gleaned from H10 - the wait_write is only necessary every other write for 8-bit width as it seems)
23:30:31amiconnI'll make a patch
23:30:31webmindrasher, sounds like something you want to run some oss on
23:30:55amiconnIs the nano always type1?
23:32:44pixelmaadvcomp2019: something isn't stable with the radio (maybe an initialisation problem?). But I sometimes got that too - then most of the times a reboot helps, low_light reported that he had to set the frequency region setting in the OF again (on c200 though but it has the same tuner chip). I had one occasion where the radio wasn't detected by Rockbox after starting it in the OF once it was back in Rockbox too
23:33:50*[IDC]Dragon will buy a Sansa, you may see me again then
23:34:08Bagder[IDC]Dragon: old timers revival then!
23:34:09*[IDC]Dragon is in gadget-buying mode
23:34:14Bagdereven zagor is back ;-)
23:34:21[IDC]Dragonthe dollar is so cheap!
23:34:27Bagderhaha, indeed
23:34:29 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
23:34:33ZagorBagder: :)
23:34:48[IDC]DragonI hope customs won't dissect me
23:35:20advcomp2019pixelma, i seen something in the middle part of the fm frequencies.. it is like flowing up and down now
23:35:22 Quit Toxicity999 ("Leaving")
23:35:32[IDC]Dragonthe US sansa has an FM radio?
23:35:52Bagderbut there are euro ones with fm too
23:36:05Bagderif we're talking e200
23:36:13pixelmathere are radio Sansas available in Europe too
23:36:16[IDC]Dragone280, yes
23:36:17amiconnlinuxstb: New patch up:">
23:36:22Bagderthe c200 seem to all have fm, even if not said to ;-)
23:37:44preglowso we're not the only ones having a good dollar now
23:38:24Bagderright, it is the dollar that is low so that goes for plenty people
23:38:45Bagdereven canadians go US to shop now I hear...
23:39:13sin613they've always come here. i live an hour from the border, and my town is always swamped with people from canada.
23:39:15[IDC]Dragoneuropean consumer electronic retailes still convert 1:1 and cash the rest
23:39:29sin613every weekend. they drive campers down and park in the walmart parking lot
23:40:07amiconn@Nano owners: This patch should ideally also be tested on nano
23:40:17linuxstbamiconn: It doesn't link - undefined reference to lcd_send_lo
23:40:54rasher[IDC]Dragon: even worse for the brits
23:41:06 Nick Klevi-Dinner is now known as Klevi (
23:41:12pixelma[IDC]Dragon: but you're currently not in New York City by any chance?
23:41:25amiconnlinuxstb: oops - I'll check
23:42:12 Quit Klevi ("Leaving")
23:42:38 Join Klevi [0] (
23:42:38amiconnAh, lcd_update_rect() and lcd_yuv_blit() try to send commands directly that way - need to fix
23:42:52[IDC]Dragonpixelma: no, I'm in silicon valley
23:44:29 Quit outside (Read error: 110 (Connection timed out))
23:49:08linuxstb[IDC]Dragon: You're not interested in porting to a new device?
23:51:05[IDC]Dragonwhich would that be?
23:51:16*[IDC]Dragon has already done that...
23:51:22linuxstbGood question...
23:51:33amiconnlinuxstb: New patch up:">
23:51:38linuxstbThere's a definite lack of nice DAPs at the moment...
23:51:41amiconn^^ verified to build this time :)
23:51:52[IDC]DragonI thought you're gonna advertise one
23:52:16[IDC]Dragonthe nano 3g is nice...
23:52:18bluebrotherlinuxstb: did you have had any chance checking FS #7913?
23:52:45*amiconn thinks the nano 3g is ugly
23:52:50preglowit is
23:52:58preglowand it really isn't nice to use either
23:52:59[IDC]Dragondid that guy return with the apple datasheets? ;-)
23:53:00preglowfits poorly in the hand
23:53:01linuxstbbluebrother: Didn't I already write a patch for that?
23:53:05bluebrotherthe nano 3g has the nickname "fatty" for a reason ;-)
23:53:06amiconnToo big for a flash based dap, even though it's thin
23:53:17bluebrotherlinuxstb: did you write a patch? I only remember you trying it out
23:53:34bluebrotherbut I might have missed that patch
23:53:46*amiconn would like to see a port to the shuffle (2nd or 3rd gen), or the classic
23:53:52[IDC]Dragonhave you held it? I found it amazingly small
23:54:08preglowit's too short
23:54:08[IDC]Dragonalmost like an after eight wafer
23:54:14preglowit kept slipping around in my hand when i tried to navigate
23:54:16linuxstbbluebrother: I'll try and look at it tomorrow - I need to do something else this evening, and my ibook is elsewhere.
23:54:17preglowthe nano1g fits perfectly
23:54:25[IDC]Dragonhot-glue it to a stick!
23:54:29amiconnyesBut even those aren't *that* nice, as they have no radio, as all ipods, and no recording either
23:54:31bluebrotherno hurries.
23:54:42*preglow thinks he's done with ipods
23:54:52linuxstbbluebrother: I think I'll need to get libusb working as well... I'm not sure how that works with univeral binaries...
23:54:54amiconn...and that touch stuff
23:54:57[IDC]Dragonso this all sounds like Sansa then
23:55:09bluebrotherah, ok.
23:55:19amiconnAlthough, the Nano (1st gen) and video have the best (so far) LCDs of all our colour targets
23:55:43linuxstbThere are lots of telechips devices out there, but they all seem flash based.
23:55:43amiconnI'm talking about the actual panels
23:56:49*linuxstb agrees with preglow about ipods
23:58:12 Join spky [0] (
23:58:41 Quit Klevi ("Leaving")
23:58:47 Join zajacattack [0] (i=426172ec@gateway/web/cgi-irc/
23:58:53 Join Kayasha-kun [0] (i=46518fe6@gateway/web/cgi-irc/
23:58:58Kayasha-kunHello people, got a quick question

Previous day | Next day