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

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

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

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

#rockbox log for 2004-10-18

00:17:18 Part scott666_
00:17:24 Quit _aLF ("Leaving")
00:20:06 Join Lurkski [0] (~Miranda@24.30.163.142)
00:22:41 Part Lurkski
00:58:30***Saving seen data "./dancer.seen"
01:00
01:03:51[IDC]Dragonmy Ondio tuner works!
01:04:03midkwoo! :)
01:04:13midkgreat work.. sounds like it's progressing quite well
01:04:22 Quit AciD (Remote closed the connection)
01:04:29midkcurious, what haven't you got to work yet?
01:04:49[IDC]Dragonwe got it pretty much complete now
01:05:44[IDC]Dragonthe keyboard screen is missing, tuner presets found no button
01:09:12midkgreat
01:24:40[IDC]Dragon'night!
01:24:45 Quit [IDC]Dragon ()
01:54:44 Join Bluechip [0] (~BlueChip@cpc3-colc1-3-0-cust61.colc.cable.ntl.com)
01:57:06midkhey bc
01:57:11Bluechiphey mk
01:57:32midkanything in the works lately?
01:57:47Bluechipbeen writing a book
01:58:14midkon what?
01:58:22Bluechiplinguistics
01:58:40Bluechipprimarily neuro linguisitc programming
01:58:52midkneuro what? :)
01:58:58Bluechipbut variations and adaptations based on it
01:59:02Bluechiplol ;)
01:59:47midkdo you plan to publish it or anything?
01:59:57Bluechipprobably, yes
02:00
02:00:11midkawesome
02:00:35Bluechipty
02:01:22midkappropriate("Good luck!")?say():shutUp();
02:01:32Bluechiplol
02:01:37Bluechiphe has mastered the ?
02:01:58midknot really sure if it would work or not.. last time i tried something like that i got an error
02:02:12midki stuck to switch(case) or if(x)
02:02:14midk:)
02:02:39Bluechipoften fixed by adding paren's everywhere
02:02:48Bluechip()?():();
02:03:46midkgood idea
02:04:54Bluechipespecially if you have them nested, when it becomes almost essential ...not sure why - poor compilation imho - but then I may be lacking the vital piece of information that makes everything clear
02:05:57midki never liked that style much anyhow.. i prefer something a little more readable like if-else
02:06:14Bluechipthey are good when used sensibly, i think
02:06:33midkif there's more than 3 or so if/else statements i use a switch
02:07:17Bluechiptheory has it that the compiler should do that anyway
02:07:38Bluechipyou should make two compiles, one with, one without, and see if there is any filesize difference on the .rock
02:07:55midksimpler to read is all
02:08:13midkmost of my programming now takes place at linav
02:08:55Bluechipprintf("%d file%s copied", n, (n==1)?"":"s");
02:09:24midklooks right to me
02:09:35midkto make it shorter:
02:09:41Bluechipi suggest that would be a beneficial way to use ?
02:09:43midkprintf("%d file%s copied", n, (n>1)?"s":"");
02:09:44midk:)
02:09:53Bluechip0 file copied ???
02:10:03midkoh.
02:10:05midk:p
02:10:06midkHAHA.
02:10:07Bluechiplol
02:10:08Bluechip;)
02:10:33midkrephrased: 'to make it shorter but less functional:' :)
02:10:40Bluechiplol
02:10:43Bluechip:D
02:10:51Bluechipnow.....
02:11:03Bluechiphow many times have *I* done that in my life?
02:11:07Bluechiplol
02:11:46midkmmm.. probably 0 :)
02:11:53Bluechipi wish
02:12:06Bluechipdid you read my post on the database idea yet?
02:12:21midknope
02:12:25Bluechipi used to do that for a living, you can be sure I cocked up more than one of those in my life
02:12:59midk'id3 database browsing'?
02:13:06Bluechipindeed
02:14:31midkaaaaaaaaaahhaha
02:14:34midk"Adding itunes support sounds like a clincher to me."
02:15:03BluechipI was actually referring to my most recent post
02:15:16Bluechipon database structuring
02:15:18midki'm looking
02:15:23midkthat just make me laugh
02:15:33Bluechipglad to be of service
02:16:12midk"PS. What is a "cluebie"?"
02:16:14midkHAHAHA funnier
02:16:34Bluechipstill don't know ...is it "have a vague clue" or "all clued up" or something else?
02:17:09midkno idea
02:17:56Bluechipit's creator did not reply :(
02:18:36midkhttp://static.userland.com/userLandDiscussArchive/msg015585.html
02:19:15Bluechipaha
02:19:39BluechipI'm a newbie - LMAO
02:20:16midk:))
02:20:50midkas far as iTunes, i think being a cluebie is worse ;)
02:21:20BluechipLOL
02:21:28*Bluechip giggling
02:22:51midk:)
02:23:53Bluechipwe need a util to pull the id3 tags out of an mp3 file
02:24:15midki need a util to gimme a c array from a 1bit bmp
02:25:03Bluechipyou can get the header data by copying my win32sim code and changing the fwrites to freads
02:25:50midkfread? HAHA
02:26:12midki just need to modify bmp2rb for it.. but whenever i run bmp2rb i get a segfault anyways
02:26:36Bluechipis your bmp saved in the "correct" format?
02:26:47midk1bit.. should be
02:26:55midki was thinking that may be it, i need to try again soon
02:27:17midkWINE can run notepad but not paint :(
02:27:45Bluechipweird
02:45:05 Join DJBaz [0] (~baz@80.229.144.229)
02:56:07midkbagder?
02:56:20midkBagder_, i mean :)
02:58:33***Saving seen data "./dancer.seen"
03:00
03:29:14midkbc?
03:29:24Bluechipmk!
03:29:31midk<3
03:30:09midkany idea what the problem is?
03:30:10midkicons.c:27: parse error before `linavLogo'
03:30:10midkicons.c:27: warning: excess elements in scalar initializer
03:30:10DBUGEnqueued KICK midk
03:30:10midkicons.c:27: warning: (near initialization for `linavLogo')
03:30:47Sebulba02Whats on that line?
03:31:16midkBITMAP linavLogo = {(unsigned int) linav_logo, 65, 10, 0, 0};
03:31:22midkconfusing, it works in a different .c
03:31:26midk.c file*
03:32:10midki think i got it.. just a bit of tweaking
03:32:27midki always do that.. i try to fix something for 20 minutes, then ask, then immediately figure it out
03:32:31midkapologies :)
03:32:43BluechipI do that in Tesco
03:35:30midktesco?
03:36:06midka store?
03:37:28Sebulba02your not the only one, sometimes you just need to talk to someone before you can figure it out, even if they have nfc of what your saying, heh
03:39:21midkhaha :)
03:39:32midkactually that was sort of it
03:40:16midki realized that you wouldn't know what BITMAP was −− i'm quite sure it's a standard C variable type −− and then i remembered i'd probably forgotten to include something
03:41:01midks/quite sure it's a/quite sure it's not a
03:41:56Sebulba02I was about to suggest that.
03:42:11Sebulba02But its always better if you figure it out on your own :)
03:42:46midknext time i've got a problem i'll just paste it in a text editor as if i were pasting it to you, then come up with the answer right away :)
03:43:24Sebulba02Or ask the wall your question & the answer will come to you.
03:43:47midkseems less convenient.
03:44:11Sebulba02Oh, you don't talk to the walls?
03:44:23CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
03:44:23*Sebulba02 must just be weird then.
03:44:31BluechipI talk to the trees, but they don't listen to me - LOL
03:44:35midki generally attempt to refrain from the practice...
03:44:38midkhaha :)
03:48:03*Sebulba02 goes back to star gazing.
03:49:00Sebulba02Bluechip: Well, trees have other things to worry about, like not losing leaves.. or producing fruit. A wall has nothing real to worry about, it'll listen, heh.
03:49:16Sebulba02Anyways.
03:49:17Bluechiphmmmmm
03:49:22Bluechipa good point
03:54:07*Sebulba02 thinks Cygnus is above him.
03:55:07BluechipIs it a split decision?
03:55:08Bluechiplol
03:56:19*Sebulba02 looks lost.
03:56:43BluechipDoesn't Cygnus link to the Gemini twins?
03:57:10BluechipMaybe my mythology and/or astrology is off
03:57:35Sebulba02I thought that was somehow a gemini thing, but I can't really tell based on this map.
04:00
04:01:04Sebulba02Ah, a square chart helps.
04:04:00Sebulba02Which oddly doesn't have gemini on it.
04:04:08BluechipLOL
04:04:24Bluechipevil twin wins again
04:04:55Sebulba02heh, hey.. I'm one of those :P
04:05:02BluechipLOL
04:05:18Sebulba023/4ths of my family is
04:05:22Bluechipweird
04:05:52Bluechipstatistically very unlikely
04:05:59Sebulba02Theres only 4 of us.
04:06:04Bluechipah
04:06:20Sebulba02But it is weird, yes.
04:06:33Bluechipthen what is the occasion celebrated by your parents three months after your birthday?
04:07:59Sebulba02Nothings in August, really.
04:09:00BluechipI was surprised by the number of people who realised they were "birthday presents"
04:09:57Sebulba02heh
04:10:08Sebulba02Not in my case.
04:10:16Bluechipme neither (fwiw)
04:10:36Sebulba02It was made clear a long time ago I was a mistake.
04:10:44Sebulba02but who isn't?
04:12:27BluechipMy mother was on the pill (although now knowing that the 21yr test results are in, it would seem ...not taking it as prescribed)
04:13:00Sebulba02heh
04:13:08Sebulba02Oops.
04:13:14Bluechiplol
04:19:42Sebulba02kick arse, Stargaze works in WINE
04:53:43 Quit DJBaz ("Leaving")
04:55:47 Quit midk (Remote closed the connection)
04:57:39 Join midk [0] (~midk@c66-235-14-120.sea2.cablespeed.com)
04:58:34***Saving seen data "./dancer.seen"
05:00
05:20:15 Quit Bluechip ()
06:00
06:58:37***Saving seen data "./dancer.seen"
07:00
07:02:25 Join LinusN [0] (~linus@labb.contactor.se)
07:11:13 Join AciD [0] (~gni@longchamp44-1-82-67-133-87.fbx.proxad.net)
07:19:46 Quit AciD (Read error: 104 (Connection reset by peer))
07:20:06 Join AciD [0] (~gni@longchamp44-1-82-67-133-87.fbx.proxad.net)
08:00
08:35:26 Quit einhirn ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
08:42:17 Join [IDC]Dragon [0] (~idc-drago@pD95123A5.dip.t-dialin.net)
08:43:05[IDC]Dragon'morning!
08:45:04[IDC]DragonLinusN: r u there?
08:48:43LinusNi'm here
08:49:05[IDC]Dragonthat's nice ;-)
08:49:20LinusNyeah, isn't it?
08:49:26[IDC]Dragonhow did you like my hacking?
08:49:37LinusNyou've been busy this weekend
08:49:50[IDC]Dragonmostly with our house
08:50:17 Join Zagor [242] (~bjst@labb.contactor.se)
08:51:07[IDC]Dragonanother good morning seem in order
08:51:17LinusNi haven't looked that much at your radio stuff, but so far it looks good
08:51:22[IDC]Dragonseems, hi Zagor
08:52:04[IDC]DragonI replicated the i2c for now, credited it back to you
08:53:13LinusNah, i see that
08:53:32[IDC]Dragondunno if multi-instance is beneficial, passing a struct to it, with the port addresses and bit masks
08:53:49LinusNit will be on the iriver
08:54:02LinusNit has separate i2c busses for each device (!!!)
08:54:42[IDC]Dragonthe port banging code would grow quite substantially, not using the short and_b() and or_b any more
08:55:03LinusNyup
08:56:09[IDC]Dragonthe iriver folks had either ports to waste, or want to independently do their i2c from various contexts
08:58:41***Saving seen data "./dancer.seen"
08:59:22LinusNtheir programmers probably didn't know about locking :-)
08:59:54[IDC]Dragon:-) ...or do I2c in interrupts
09:00
09:00:05LinusNmaybe
09:05:44 Join einhirn [0] (Miranda@bsod.rz.tu-clausthal.de)
09:06:57 Join kurzhaarrocker [0] (~knoppix@p50877AEF.dip0.t-ipconnect.de)
09:07:17kurzhaarrockerMoin.
09:07:18einhirnHello Kurzhaarrocker
09:07:29einhirnA feature Implementor?
09:07:48kurzhaarrocker:D That's what I crave for!
09:08:26einhirnAh... Less work for us Humans, if a plugin like that would exist... ;)
09:20:38 Join ey [0] (~knoppix@p508771E9.dip0.t-ipconnect.de)
09:21:28ey:( I stole my nick
09:21:28*ey slaps kurzhaarrocker
09:22:17Mode"#rockbox +o LinusN " by ChanServ (ChanServ@services.)
09:22:34LinusNwho shall i kick?
09:22:44eythat kurzhaarrocker - ghost
09:22:52Kick(#rockbox kurzhaarrocker :LinusN) by LinusN!~linus@labb.contactor.se
09:23:11LinusNthat felt good
09:23:17ey:D
09:23:28eyStill the nick is claimed to be in use
09:25:17eyLinusN: I updated the trigger code to the current cvs
09:25:37LinusNsaw that, great!
09:25:37einhirney: Can't Nickserv do something for you?
09:25:59LinusNyou should register kurzhaarrocker
09:26:08eyyes I should
09:26:59 Nick ey is now known as langhaarrocker (~knoppix@p508771E9.dip0.t-ipconnect.de)
09:27:32 Nick langhaarrocker is now known as kurzhaarrocker (~knoppix@p508771E9.dip0.t-ipconnect.de)
09:27:36Zagorbtw, kurz, is there a reason why splitedit uses the currently playing file instead of being invoked from the "open with" menu?
09:28:09kurzhaarrockerNot really but there are many reasons why the spliteditor should be reprogrammed from scratch :)
09:28:18Zagor:)
09:28:33LinusNfeel free to do it :-)
09:28:48kurzhaarrockerAnd I'm convinced that I started the spliteditor before there was the open with stuff
09:28:54LinusNyup
09:29:55Zagoryeah, I was just curious if you'd looked at it and dismissed it, or just hadn't looked at it
09:31:35kurzhaarrockerWell there in fact _is_ a reason: by pressing pause during playback you can preselect the coarse split position. Doing that within the split edtior is not very comfortable. It's merely for finetuning the split.
09:32:03Zagorok. we could have it both ways.
09:32:10kurzhaarrockeryes
09:32:49Zagordo you have any ambition to do more work on splitedit?
09:33:40kurzhaarrockerI cant guarantee to have enough time this century. Otherwise I have many ideas about it. Fell free to do whatever you like with it :)
09:34:04Zagorok, just checking
09:47:09 Nick Bagder_ is now known as Bagder (~daniel@1-1-5-26a.hud.sth.bostream.se)
09:55:08 Part kurzhaarrocker
10:00
10:01:18Bagdercurl release day
10:01:25LinusNwooo
10:01:43 Join kurzhaarrocker [0] (~knoppix@p508771E9.dip0.t-ipconnect.de)
10:02:54Bagderso now I deserve some coffee
10:03:21Zagorwhat do you guys say, should we release rockbox today too?
10:03:37kurzhaarrockerIs the recording bug squished?
10:03:40LinusNZagor: are you nuts?
10:03:52Zagorhehe
10:04:05Zagorkurzhaarrocker: it appears to be a hardware bug
10:04:21 Join Headie [0] (~hehe@fsto6.sto.sema.se)
10:04:28LinusNBagder: i heard that wget is better than curl :-)
10:05:03LinusNZagor: do we know that? i haven't had any feedback from paul yet
10:05:25Zagor−−> appears <−−
10:05:26kurzhaarrocker:( No triggered recording in the next rockbox release
10:05:32LinusNand paul definitely has problems with the asm ATA optimizations
10:05:52BagderLinusN: :-P
10:06:14LinusNhe can't run the dailies without removing the ata optimizations
10:06:43Zagorneato. do we have his hardware info logged?
10:06:53LinusNthink so
10:08:20ZagorI recorded two hour-long sessions myself this weekend. the archos mic is quite impressive, for the price. a bit noisy, but very sensitive.
10:08:53kurzhaarrockerConcerning that recording bug: Why not turn off that crc check if it causes trouble?
10:09:08LinusNwe might add an option for it
10:09:25LinusNbut the stereo mode affects it as well
10:10:48kurzhaarrockerSensitivity without sacrificing the s/n ratio is the hard and expensive part.
10:11:07Zagoryeah
10:15:11Zagorwell I propose we disable crc, zip it up and release 2.3. all this waiting benefits noone.
10:15:25BagderI agree
10:15:35*kurzhaarrocker cries
10:15:51LinusNZagor: another release without a manual?
10:16:00Bagderthere is one
10:16:30LinusNyes, we have the preliminary that christi started writing, but it is far from finished
10:16:38Zagorif needs be, yes. we can't just sit around and wait forever.
10:16:50Bagderwell, no one writes docs then let's skip it
10:17:05kurzhaarrocker<- wrote docs for triggered recording :)
10:17:11LinusNshe specifically waited until we froze the features
10:17:21*Bagder slaps the "good contributor" sticker on kurzhaarrocker
10:18:26ZagorI don't want to freeze now, it just takes more time and halts current work. we've got a pretty good state right now, let's ship it.
10:18:55LinusNZagor: i assume you have tested the new fm radio code and found it allright?
10:19:45Bagderif it bugs terribly, we release a 2.3.1
10:19:52ZagorI'm confident jörg and jens have. and if it's not working well, I'll be happy to release a new version tomorrow
10:20:12LinusNi really don't like this
10:20:21ZagorI know. you never like to release :)
10:20:28LinusNthanks
10:20:33Bagderwe need a "release manager"
10:20:38Bagderhonestly
10:21:21ZagorLinusN: well speak up, what exactly is it you want to wait for?
10:21:50LinusNi want the recording issue resolved
10:22:04[IDC]DragonI want the keyboard for the ondio
10:22:05LinusNand i want a manual
10:22:21BagderI want world peace
10:22:26Bagder:-O
10:22:45[IDC]Dragonme too
10:22:54kurzhaarrockerAnd I want triggered recordings - but I know I'm the only one who wants that.
10:23:00Zagorhow many people are bitten by the recording issue?
10:23:11Zagorkurzhaarrocker: I want it too, but I don't want to delay the release for it
10:23:24LinusNwe don't know how many that suffer from the recording bug
10:23:25[IDC]Dragonthe ondio is almost ready...
10:23:26Bagderthis isn't the last release you know ;-)
10:23:33kurzhaarrockerI know, Zagor, just nagging.
10:23:49ZagorLinusN: how many have reported it?
10:24:12LinusNpaul is the only one who has put the finger on it
10:24:15*kurzhaarrocker has lost recordings
10:25:39Zagorso we're waiting for a bug that has bitten one or two persons? and the only clue we have is that crc introduced it.
10:25:56LinusNqoute from the forum:
10:25:57LinusNQuote from: pcspecialist on October 14, 2004, 09:38:34 PM
10:25:57LinusNI've had no luck getting stability out of any of the daily builds so you'll want to stay away from them.
10:26:22Zagorright, and how many bug reports have we seen from him?
10:26:38LinusNok, go ahead, release it
10:27:21LinusNhow many bug reports do you want to believe that there really is a bug?
10:27:27kurzhaarrockerWhy not first try to disable crc, let paul and me try and then release in two days or so?
10:27:58Zagorkurzhaarrocker: paul got a test build over a week ago. linus can you send the same build to kurz?
10:28:04LinusNi sent paul a test version weeks ago
10:29:03LinusNhttp://linus.haxx.se/paul_rectest1.ajz
10:29:07kurzhaarrockerI have a practice session tonight anyway. A good opportunity for a 4 h recording.
10:29:15LinusNthat is 2.2 with CRC check enabled
10:29:45LinusNif our theory is correct, the recording will be seriously corrupt
10:30:02kurzhaarrockerI'll try and report tomorrow
10:30:06Zagorok good, I'll wait
10:31:32Zagor[IDC]Dragon: how much work would you say is left on the ondio port?
10:31:52LinusNi think we should announce a feature freeze, like we used to do
10:33:27LinusNfor instance, we should make the recording file name generation algorithm an option
10:33:42Zagorwhy? that's a new feature.
10:33:42LinusN"regular" recorder users want that too
10:33:59LinusNwe've tons of feature requests for that one
10:34:03LinusNwe've had
10:34:04Zagormany people want a lot of new feaures
10:34:05[IDC]Dragondo we?
10:34:33[IDC]DragonZagor: a few days would be OK to brush it up, I'd say
10:34:35LinusNZagor: since the ondio does it, the regular one could do it as well
10:35:21[IDC]DragonI'm a big fan of releases, but not like announcing them for tomorrow
10:35:39*kurzhaarrocker wants directories generated by date and files named by time and my mind consists of many split personalities. Does that count extra?
10:36:28[IDC]Dragongive us some timline horizon to implement against, but no "snapshot release"
10:37:37[IDC]Dragonif I know a scheduled release is due, in, say a week, I can stabilize things well better
10:40:59[IDC]Dragonbbl
10:41:05 Quit [IDC]Dragon ()
10:46:30 Join ashridah [0] (ashridah@220-253-119-176.VIC.netspace.net.au)
10:48:36Bagderclamav uses libcurl now
10:48:48LinusNc00l
10:49:16kurzhaarrockerwhat a pity it doesn't use wget.
10:49:16*kurzhaarrocker hides
10:49:21Bagderhaha
10:49:32Bagderlibcurl is quite a different beast than wget
10:53:14*Bagder 's daughter is right now sitting on the floor and rips toilet paper into very very tiny pieces
10:53:15kurzhaarrockerThe only one time I wanted to use curl yet I had to realize my provider doesn't allow that.
10:53:54Bagder;-)
10:54:02Bagderyou mean the php module then?
10:54:07kurzhaarrockeryes
10:57:53 Join amiconn [0] (~jens@pD95D1099.dip.t-dialin.net)
10:58:07amiconnG'morning
10:58:16LinusNyo
10:58:34kurzhaarrockerghurt
10:58:44***Saving seen data "./dancer.seen"
10:59:21LinusNamiconn: we are discussing the recording issue
10:59:30amiconnYup. (log peeking)
10:59:42Bagderaha, a peeker!
11:00
11:00:02kurzhaarrockerThat reminds me of good old C64 basic
11:00:08amiconnI like to remind you that I cannot test the radio code; no box with a radio :(
11:00:32ZagorLinusN: which feature did you say ondio has that "regular" doesn't?
11:00:55LinusNgenerating recording file names, "file0001.mp3" etc
11:01:05LinusNor is it "rec0001.mp3"?
11:01:09amiconnOndio seriously lacks the keyboard :(
11:01:40Zagoryeah, because it doesn't have a clock
11:01:41amiconn(But I already have some idea how to do that with the limited keys)
11:02:03Zagoramiconn: if the crazy player people could do it, so can you ;)
11:02:11LinusNZagor: yes, but people have been bugging us for years about not doing that for the regular recorder :-)
11:02:30Zagoryes, and I agree it would be a nice option. but it's not a release-stopper.
11:02:58*Bagder is off
11:03:08LinusNsurely not, but i thought we could very well add it, since the code is there
11:03:43amiconnZagor: The player keyboard is rather unintuitive (imho, now that I could test it with my Studio)
11:03:58Zagoramiconn: i agree completely
11:04:12Zagorhowever I don't think it *can* be made intuitive
11:04:17LinusNthe player keyboard design is really silly
11:04:47amiconnI think it can, maybe that will become a by-product of the Ondio keyboard
11:05:03Zagorsounds good
11:05:07 Join Lurkski [0] (~Miranda@24.30.163.142)
11:06:40 Join edx [0] (edx@p54879108.dip.t-dialin.net)
11:09:46Zagorso is everyone happy with a week-long freeze instead then?
11:10:12 Join [IDC]Dragon [0] (~d90a3255@labb.contactor.se)
11:11:01LinusNyup
11:11:16amiconnZagor: I don't like freezes :( The plugins need quite some rework...
11:11:40LinusNamiconn: that's not adding features, is it?
11:12:46*[IDC]Dragon just steps back in, reading logs
11:13:04[IDC]Dragona week grace period sounds fine
11:13:22amiconnLinusN: Not quite, but it sometimes means to change how things work
11:13:47Zagorthat's why I wanted to do a snapshot release, to avoid hindering current work
11:13:57[IDC]Dragonamiconn: what else have we got left for the ondio?
11:14:10[IDC]Dragondid I forget something?
11:15:22amiconnMost lacking is keyboard. Then we need dual-volume and howswap
11:15:40[IDC]Dragonthe latter 2 can wait, I'd say
11:15:51[IDC]Dragonthose are big features
11:16:05Zagorwhy is keyboard essential?
11:16:18LinusNfile renaming etc
11:16:36[IDC]Dragonyou're prompted with that in various places
11:16:47[IDC]Dragone.g. .cfg saving
11:17:38[IDC]Dragonand I'm missing FM presets, found no button for that
11:17:53[IDC]Dragonmaybe Jens has an idea
11:18:35Zagorfile renaming is hardly essential on a flash player. config saving is nice, but could probably just as well use a couple of predefined names.
11:19:21amiconnZagor: I would have needed file renaming already several times
11:19:39Zagorbut if you think you can have keyboard code running in a short time, that's ok with me
11:20:05 Join mrelwood [0] (~d5f38ec0@labb.contactor.se)
11:20:11LinusNfm preset handling should work too imho
11:20:11Zagoramiconn: yes, it's a nice feature. but hardly essential. lots of players (most, I would bet) lack it
11:20:30[IDC]DragonI need to work a bit more on the tuner
11:20:34amiconn[IDC]Dragon: I can't test FM :( ... but, to overcome the small number of buttons, context menus would be rather handy...
11:20:54Zagorso now you're saying we should wait until the ondio is Release Quality
11:21:15[IDC]Dragonamiconn: but you can think about it ;-)
11:21:20ZagorI have massive respect for Jens and Jörg, but I doubt it will be done by next monday
11:21:26mrelwoodHey guys! Don't know if it's my english or understanding, but FAQ and forums didn't enlighten me. Is the Rockbox already working on a iRiver iHP-120?
11:21:33Zagormrelwood: no
11:21:41mrelwoodokay.
11:21:51[IDC]DragonZagor: it will be like 1.0 for rockbox
11:21:51mrelwoodis it still under developement?
11:21:53amiconnI wanted to ask whether it is desired to rewrite the text viewer a bit. Today, it uses a heapload of non-intuitive key combos to cycle through various settings. It should have a config menu instead, with save function.
11:22:16[IDC]DragonI wodn't worry about that
11:22:18LinusNmrelwood: we have only begun with the iriver port
11:22:23Zagor[IDC]Dragon: yes, that's how I feel it should be. and 1.0 didn't have file rename.
11:22:29Zagor(for example)
11:22:30[IDC]Dragonthe main thing is to scroll up and down
11:22:57kurzhaarrocker[IDC]Dragon: bookmarking would be nice
11:23:03mrelwoodcool! are the options and possibilities going to be the same as in the current for Archos?
11:23:11kurzhaarrocker(but that's something else)
11:23:13Zagormrelwood: yes
11:23:17LinusNmrelwood: that's the plan
11:23:17[IDC]Dragonkurzhaarrocker: it does
11:23:36mrelwoodso, input level metering for recording, and no recording time limits?
11:23:37*kurzhaarrocker is astonished
11:23:52LinusNmrelwood: yes
11:24:01[IDC]Dragonall Rockbox features are in Ondio as well, since it's the same code
11:24:04kurzhaarrockero-oh, the metering. That will be interesting stuff to port
11:24:22mrelwoodoh my! those are the only things that has kept me from getting the iHP!
11:24:31LinusNkurzhaarrocker: yes, we need a basic sound energy algorithm
11:24:39[IDC]DragonExcept for a few places where the buttons are too few and we had to explicitely cut features
11:25:02mrelwoodbut the recording formats and bitrates will be the same as they are in the iRiver originally?
11:25:14ZagorLinusN: i guess we need an rb lunch :)
11:25:32kurzhaarrockerBut once we have that metering code it will probably precise in contrast to the archos hardware crap.
11:25:51kurzhaarrocker+be
11:26:22Zagormrelwood: not necessarily. we will focus on mp3 first, and then add other formats after that.
11:26:37mrelwoodbut it will be able to do 320kbps mp3?
11:26:44amiconn[IDC]Dragon: What is the current button assignment in radio screen (both JB FMR and Ondio FMR)?
11:27:32Zagormrelwood: hopefully, yes
11:28:09LinusNmrelwood: we honestly don't know much about the limitations in the original iriver firmware
11:28:32Zagorlunch, see you later
11:28:33LinusNlunch time
11:28:35 Nick Zagor is now known as Zagor|lunch (~bjst@labb.contactor.se)
11:28:58mrelwoodThe original can rec up to 320kbps and 44.1kHz wav. But if the 320kbps is available, I'll surely wait instead of getting some other poor excuse for a DAP. Thanks guys!
11:29:51kurzhaarrockerWhat is DAP?
11:31:04[IDC]Dragonamiconn: FMR is tuning on left/right, volume on up/down, menu on F1, presets on F2, recording on F3, stop recording on Off, screen freeze on Play, exit on On
11:32:01mrelwoodDigital Audio Player
11:32:08kurzhaarrockerok
11:32:12kurzhaarrockerthx
11:32:21[IDC]DragonOndioFM currently is same for direction, menu on long Menu, recording on short Menu, stop recording on Off, screen exit on long Off
11:32:58mrelwoodamerican forums use it alot. like www.dapreview.com. mostly meaning mp3 players with a HD.
11:33:12[IDC]Dragonsee radio.c, line 55 ff.
11:33:46amiconn[IDC]Dragon: two qns: (1) "exit" (On) brings you where? (2) Presets is a menu, or a cycle-through?
11:34:21[IDC]Dragon(1) exit goes back into the main menu where you came from
11:35:16[IDC]Dragon(2) presets is a menu
11:39:13amiconnWhy do you exit from radio with "On" and not "Off"? Off would be much more logical...
11:40:22amiconnIt would be the same as with the recording screen: 1st "Off" stops the recording, 2nd "Off" exits
11:40:45amiconn(no long hold necessary)
11:44:13[IDC]Dragonyes, with some slight code modifications we may do so
11:45:04[IDC]Dragonbut how to call the presets menu?
11:45:31[IDC]Dragonunless we do very unintuitive combos with the menu button
11:45:32amiconnYou can use some Menu+xxx combos
11:45:52amiconnYes, that is not very intuitive :(
11:47:14[IDC]Dragonwhat are your ideas for the keyboard screen?
11:48:04[IDC]Dragonlike, arrows move across the pickboard, menu arrows move within the string?
11:48:56[IDC]Dragonor, a 1-D pickboard instead of 2-D?
11:49:02kurzhaarrockerwith menu being a "toggle" or a "shift" (combos)
11:51:48amiconnNo, arrows move cursor around the screen (up/down/left/right), with a 4-row pickboard as it is now, but the cursor is also allowed to move outside the pickboard into the filename line, and move the insert position there. Menu+Left would be backspace there, and Menu+Right delete.
11:52:21amiconnMenu button inside pickbaord inserts character under cursor, long press accepts file name.
11:52:36kurzhaarrockerok, got understood.
11:53:22[IDC]Dragonsounds good
11:53:30amiconnMoving the cursor across the left or right border of the pickboard selects previous/next pickboard
11:53:55amiconn(Perhaps with an alias on Menu+Left/Right for direct pickboard selection)
11:54:19amiconnOff button = cancel (same as now)
11:54:34 Part Lurkski
11:55:35amiconnSome of these changes might be useful for recorders too, and the ideas could be re-used for a better player pickboard (1-line only, of course)
11:56:38[IDC]Dragond the nice thing is that tis code can be tested on the simulator, for a change
11:57:21[IDC]Dragon(oops, I fail with this PC keyboard already)
11:58:08kurzhaarrockerbtw: When I tried to build with target = recorder a dummy rombox is build. When I do the same with target = FMRecorder building rombox fails. Is that difference intensional?
11:58:52amiconnkurzhaarrocker: You need the updated uclpack for proper rombox. For FMR it fails because of too little flash space
11:58:58[IDC]Dragonyou probably have no up to date uclpack
11:59:28[IDC]Dragonamiconn: for OndioFM, it now fails again, too
11:59:32kurzhaarrockerok I'll test that tomorrow with current uclpack
12:00
12:01:09amiconn[IDC]Dragon: Radio is working properly now, or does it need some more polishing?
12:05:00amiconnI have to admit that I almost never use the simulator
12:34:14 Join webguest64 [0] (~5149bab3@labb.contactor.se)
12:43:19 Nick Zagor|lunch is now known as Zagor (~bjst@labb.contactor.se)
12:49:54 Join thedude02 [0] (thedude02@pm477-42.dialip.mich.net)
12:50:58 Part thedude02
12:51:31LinusNamiconn: u there
12:51:32LinusN?
12:58:48***Saving seen data "./dancer.seen"
13:00
13:00:54amiconnLinusN: Now I am.
13:01:02[IDC]Dragonme too
13:01:09[IDC]Dragon(just got back)
13:02:07[IDC]Dragonamiconn: the radio needs a better "tuned" detection for the search mode, ans I need to suspend it after use
13:04:17amiconnDoesn't the philips tuner have an autotune feature? Just use this...
13:04:47[IDC]Dragonit has, but the generalization would go overboard
13:05:29[IDC]Dragonright now, the radio screen doesn't know which tuner it is talking to.
13:06:23amiconnWhy? You could implement the autotune feature as a tuner api function. For philips, it would use the chip's autotune feature, and for samsung it would do it the same way as now
13:07:14[IDC]Dragonpossible, then the samsung implementation needs a task
13:07:43 Quit mrelwood ("CGI:IRC (EOF)")
13:08:09kurzhaarrockerThere's a bug in recording.c that can be fixed easily: When you enter quickscreens from the recording screen the blinking red led might stay switched on. Very irritating when there's neither disk access nor a recording in progress
13:08:19kurzhaarrockerWant a patch file?
13:14:15amiconn[IDC]Dragon: Why would it need a thread? The current system does the autotune in the main thread, right?
13:16:45[IDC]Dragonif we want a common interface, you either (1) say: tune to xx MHz, then tell me if this is a station. You iterate until a good one is found.
13:17:24[IDC]Dragonor (2) you say "search upwards" to the tuner interface, sit an poll until tuned
13:18:22[IDC]Dragonfor (2), and a tuner without a search feature, the tuner wrapper has to autonomously tune and test, to emulate it
13:19:02[IDC]Dragonso it would need a thread, unless you do something crappy like a tick function
13:19:29[IDC]Dragone.g. the polling from the top level could be used to advance
13:22:08amiconnIirc, the current tuning works that way, or can't you auto-search upward, but only tune in steps by hand?
13:23:25[IDC]Dragon?? The samsung tuner need to be set each time, the Philips has a search enable and a direction bit, plus a readback if the search stopped
13:23:28amiconnIn that case, I think a thread would be needed, for philips as well. Then the thread could report the "station found" back to the main thread via an event - no polling
13:23:48 Nick kurzhaarrocker is now known as kurzhaarrocker|l (~knoppix@p508771E9.dip0.t-ipconnect.de)
13:24:13amiconnMy question was: Can you auto-tune in the current radio screen, or not?
13:24:20[IDC]Dragonthe radio screen has nothing to do anyway
13:24:45[IDC]Dragonthe screen does have a search, if that is what you mean
13:25:18[IDC]Dragonit advances the tuner by 100kHz, checks for tuning, and so on
13:25:37amiconnNo I mean (like I would expect from it): short press of left/right would tue down/up by 50 kHz, while a longer press would start a search for the previous/next station
13:26:50[IDC]Dragonexactly, yes
13:27:33amiconnAnd the tune/check/tune.. is currently done within the main thread?
13:27:40[IDC]Dragonyes
13:28:15amiconnHmm. Then I think we should go for an additional thread.
13:28:43[IDC]DragonI try to keep it simple, if we can
13:31:06amiconnI think the auto-tune by the chip might work both faster and better than a hand-implementation. If this needs a thread for non autosearch capable tuners to be properly wrappable, then let's do it
13:31:49[IDC]DragonLinusN: reading?
13:35:20[IDC]DragonOr like I wrote above, I use the poll interval of the screen as a tick to try the next frequency
13:41:14amiconnThe samsung tuner probably needs some time until it tuned to a new frequency, right? Des it report back when done, or is there a fixed delay time?
13:42:24[IDC]Dragonsleep(1)
13:45:13LinusNwhat is the problem?
13:45:27[IDC]Dragonno problem
13:45:40LinusNthen why are we talking about separate threads for tuning?
13:46:38[IDC]Dragonok, (so far) the tuning is not detected, the search is failing
13:47:14[IDC]Dragonright now I don't remember if it always stops, or always carries on
13:47:25[IDC]DragonI think it stops
13:47:44[IDC]DragonJens raised the autosearch of the Philips
13:49:18[IDC]Dragonbut first I'd like to check what the IF counter really returns
13:52:52LinusNabsolutely
13:52:57 Join oxygen77 [0] (~Chris@pauguste-7-82-66-87-78.fbx.proxad.net)
13:53:16LinusNi prefer the "manual" tuning, because it makes the API simpler
13:58:40amiconnWhy not use a feature already present in the chip?
13:59:50Zagoruse a callback to avoid the need for a thread
14:00
14:00:00LinusNbecause it forces us to emulate it on the other chip and
14:00:33LinusNi think it's better to use the common feature set
14:00:47LinusNwhy is the autotuning feature better?
14:00:55LinusNis it more accurate?
14:01:13[IDC]Dragonif it works and the other doesn't, it is better ;-)
14:01:35Zagorwe already emulate it, don't we?
14:01:39amiconnI think it will be both faster and more accurate, because the chip decides when to advance the frequency
14:01:41LinusNi doubt that the semi-manual methos doesn't work
14:01:45[IDC]Dragonlet's rather call it autosearch
14:02:20LinusNZagor: yes, but we'd have to emulate in on the API level, if we want the interface "clean"
14:02:46[IDC]Dragonthen please change the subject until I found out
14:03:31ZagorLinusN: is that so bad? if both use a callback for screen updates, the api would be nice and clean
14:03:41[IDC]Dragon(the discussion is premature)
14:04:02Zagorwell we can still discuss the principle
14:07:14[IDC]Dragonok, my $0.02 then: in terms of code size, it probably doesn't matter much if the stepping is done in the screen or in a tuner thread. The latter architecture pays off if you have a search-capable tuner, then you can drop that ( little bit of) code
14:07:42[IDC]Dragonright now, I tried not to turn everything upside down
14:07:45 Part oxygen77 ("Cho")
14:07:45Zagorwhy a thread instead of a callback?
14:08:01[IDC]Dragonwho does the stepping?
14:08:31Zagorsince we are talking in the context of a single thread, the answer is: the code
14:08:54[IDC]Dragonso you tick the lower tuner layer?
14:09:03Zagori.e. tuner_autosearch(up, &search_callback)
14:09:28[IDC]Dragonas a blocking call, you mean, now I get it
14:09:31Zagoryes
14:09:59[IDC]Dragonhmm, might not be so friendly to all the other button handling
14:10:05 Join edooo [0] (~edooo_188@82.129.244.5)
14:10:32Zagorsure it would. the callback can check the button queue and order the search function to abort
14:10:54 Quit edooo (Remote closed the connection)
14:11:07Zagor(via its' return code)
14:11:09[IDC]Dragonthen the callback ticks the app?
14:11:17 Join edooo [0] (~edooo_188@82.129.244.5)
14:11:21 Quit edooo (Remote closed the connection)
14:11:43Zagorthe callback handles screen updates and maybe button checking to see if user wants to stop searching
14:11:56Zagorapp-level stuff
14:12:08[IDC]Dragonthis is very upside down now
14:12:20Zagorif the callback returns false, tuner_search aborts and returns
14:13:41[IDC]DragonI don't like to overuse callbacks, they revert the normal layer hierarchy
14:14:23Zagori don't think this is overuse. it's a simple way to add user feedback to a long blocking function without having to add a dedicated thread
14:15:01Zagorsort of like a progress indicator callback
14:15:32[IDC]Dragonbut a big one, which contains all the event-driven code of the screen
14:16:25Zagorwhat it does is up to the application. the tuner code just contains the option.
14:16:36Zagorhow do you mean all event-driven code?
14:16:59Zagorthe callback would not start new functions
14:17:19Zagor(other than lcd_drawstring etc)
14:17:50[IDC]Dragonit has to do all the work the screen usually does. Check buttons, invoke actions
14:18:31 Join edooo [0] (~edooo_188@82.129.244.5)
14:18:33Zagorno, the only button check is to allow the user to abort searching.
14:18:49 Quit edooo (Remote closed the connection)
14:19:03 Join edooo [0] (~edooo_188@82.129.244.5)
14:19:06[IDC]Dragonand if we simply want to adjust volume during the search?
14:19:11Zagoryou can't
14:19:26 Quit edooo (Remote closed the connection)
14:19:41 Join edooo [0] (~edooo_188@82.129.244.5)
14:19:53Zagormaybe i'm misunderstanding everything, but surely while searching there is no actual sound?
14:19:53 Quit edooo (Read error: 54 (Connection reset by peer))
14:20:20[IDC]Dragonnot nice. Then I have to exit the search, remember the condition, later adjust the volume and kick off a new search
14:20:49Zagorexit the search? doesn't the search exit automatically when you find a channel?
14:20:54[IDC]Dragonwhat piece of code were we trying to save, btw?
14:21:36Zagorwe want to be able to use the hardware search
14:22:03[IDC]DragonI know. At quite some cost, it seems now.
14:22:04 Join edooo [0] (~edooo_188@82.129.244.5)
14:22:30Zagor?
14:22:47 Quit edooo (Remote closed the connection)
14:23:02 Join edooo [0] (~edooo_188@82.129.244.5)
14:23:37[IDC]DragonI mean, if we start to workaroud setting up an interrupted search, to allow things like volume set
14:23:41 Quit edooo (Remote closed the connection)
14:24:01Zagorgranted, I don't have an FM unit so I'm probably in the blue here. but doesn't it work just like a normal car radio: press SEARCH+ to search for a channel at a higher frequency. the display shows the increasing frequency and the sound is muted. when a channel is found, the display stops and sound is turned on again
14:24:14Zagorwhy would you want to adjust volume while searching. there is no sound!
14:24:15 Join elinenbe [0] (~elinenbe_@65.115.46.225)
14:24:25elinenbebonjour
14:24:42 Join edooo [0] (~edooo_188@82.129.244.5)
14:24:46 Quit edooo (Read error: 54 (Connection reset by peer))
14:24:58[IDC]Dragonthere is sound, and it may be interesting information to you, because you e.g. just went over that weak station you're looking for
14:25:18Zagorit's like adjusting volume while fast-forwarding an mp3 file
14:25:39Zagorbut, sure, we can adjust volume in the callback if that is a desired function
14:26:22[IDC]Dragonplease let me try to use "Samsung mode" first
14:26:42Zagori.e. not use the hardware feature?
14:26:55 Join edooo [0] (~edooo_188@82.129.244.5)
14:26:58[IDC]Dragonfor the time being, yes
14:27:29[IDC]Dragonthe Ondio needs to support both tuners anyway, decided at runtime
14:27:56 Quit edooo (Remote closed the connection)
14:28:34[IDC]Dragonand you want to release in a week ;-)
14:29:02Zagori wanted to release today...
14:29:05amiconnFor getting it to work, I agree (like not using DMA was useful for me to get MMC to work at all), while once it works, using the chip's autotune would be desired.
14:29:40Zagoramiconn: agreed
14:29:52amiconnIf there is a feature in one chip and not in the other, it is imho better to implement the feature in software for the chip lacking it, instead not using the feature just because the other chip doesn't offer it
14:30:04Zagorif practical, yes
14:30:26[IDC]Dragonbut imagine you have to use the non-DMA for the other anyhow, then it's tempting to not utilize it
14:31:13[IDC]Dragonassumed if there is no practical drawback
14:32:02[IDC]Dragonguys, I'm happy the tuner works at all today
14:32:14amiconnImho the auto-tune in software should not need much code, but it needs either a thread, or a tick-task
14:33:28Zagorwhy?
14:34:07[IDC]DragonI gree that I prefer the tick over the callback.
14:34:13amiconnIn order to get the same behaviour as for the autotune-capable chip, without adding extended callback mechanisms
14:34:52Zagora callback is MUCH more simple than adding tick tasks or threads
14:35:24[IDC]Dragonthis callback loop would be an emulation for the autotuning, because naturally it could return right away
14:36:05Zagorthe callback has nothing to do with tuning, it only handles the user feedback
14:36:17[IDC]Dragonelse you would fire off the search, then query it once in a while how it is doing
14:37:39Zagor...which requires a state machine in the user interface code, PLUS a whole thread just to create user feedback
14:38:03[IDC]Dragonno
14:38:28[IDC]Dragoncase 1, autosearching tuner:
14:38:58[IDC]Dragonthe user holds e.g. Right to start search
14:39:22[IDC]Dragonthe app call search(up, 95.5 MHz)
14:39:50[IDC]Dragonwhich returns after setting the frequnecy and search dir
14:40:22[IDC]Dragonthe app periodically updates the screen, so it calls get_search_state()
14:40:57[IDC]Dragonwhich returns current frequency, stopped, strenght, whatever
14:41:17Zagorwhat happens when the user holds right again, during searching?
14:41:21[IDC]Dragoncase 2, Samsung tuner:
14:41:51[IDC]Dragonit may start a new search at the current frequency
14:42:35[IDC]Dragonwhich is probably no harm and no difference
14:42:50[IDC]Dragonit definitely has to start anew if the user holds Left
14:43:01[IDC]Dragonready for case 2?
14:43:15Zagori don't need it
14:43:42Zagori still think adding a thread is much more unneeded complexity than just running a callback each .25 MHz or whatever
14:44:24Zagorplus that we don't have these types of asynchronous calls anywhere else in rockbox
14:44:38[IDC]Dragonmy dummy approach for Samsung wound be to do the first tuning on search(up, 95.5 MHz)
14:45:07[IDC]Dragonthe next on each get_search_state() call
14:46:52 Nick kurzhaarrocker|l is now known as kurzhaarrocker (~knoppix@p508771E9.dip0.t-ipconnect.de)
14:49:54Zagorquestion: how is tuner_search different from jpeg_decode?
14:50:12*Bagder gets bored by the db thread on the list
14:51:04[IDC]DragonZagor: because you found a callback there?
14:51:08BagderI actually tried to make a db according to the TagDatabase wiki page
14:51:53Zagor[IDC]Dragon: it appears to be a callback that does pretty much want I'd like the callback in tuner_search to do
14:53:30[IDC]Dragonthat callback doesn't run the whole UI
14:53:48Zagorneither would the one in tuner_search. i don't know where you got that from.
14:54:32Zagorlet me explain my view:
14:55:21Zagorif user holds right, app calls tuner_search(up, 95.5, &callback)
14:56:09Zagortuner_search() starts searching upwards from 95.5, either using hardware search or software search. each 0.25 MHz it calls callback(some,info)
14:56:51[IDC]Dragonfor HW, I can't assure a certain frequency interval
14:57:03Zagorok, at some interval then
14:57:53Zagorcallback() redraws the screen with the new info, and checks if user pressed an abort button. if so, it returns false
14:58:12Zagortuner_search() checks the return code from callback(). it it is false, it stops searching and returns to the app
14:58:43Zagorif we want to be able to adjust volume during search, callback() can check for those buttons too
14:58:52***Saving seen data "./dancer.seen"
14:59:13[IDC]DragonI was afraid this redraw and button handler would replicate a lot of the normal UI
14:59:31Zagorit can call the same draw function as the normal ui loop
15:00
15:00:15Zagorbut the abort search button is a search-specific ui, so it has to be done separately
15:01:00 Join pillo [0] (~trillian@navlab03.dei.unipd.it)
15:03:01[IDC]Dragonmy thinking was that the search runs more or less in the background, decoupled from the screen
15:03:55Zagoryes, but I don't think the user sees it that way so there is no need for the code to do that either. searching is a pretty active action.
15:09:59amiconnUsing a thread or tick task would enable setting the volume etc. independently whether a seach is running or not, with no additional code
15:10:20Zagora new thread is no additional code? :)
15:10:55amiconnPlus, the thread/task would only needed to be started if a samsung tuner is found.
15:11:59Zagorwhy do you want to use a thread when it's not needed?
15:12:08LinusNi didn't follow the discussion
15:12:21LinusNdo we know for a fact that we need to use the auto-tune funcion?
15:13:08Zagorit's not a question of "needs", we're discussion on the assumption that autotune will bring a benefit
15:14:03LinusNin any case, i fail to see the need for a separate thread
15:14:30amiconnWith the callback approach you would have to account for whether a search is running or not in the ui code. With a thread/task you don't
15:14:41LinusNyes you do
15:14:47LinusNthe user wants the feedback
15:15:34amiconnYes, but that is much easier (imho) than different button checking in a large callback function
15:15:34Zagoramiconn: quite the opposite. with a callback you *don't* need such state information, since everything is a single thread and thus call tree
15:15:49Zagorfor the 98th time, it's not a large callback function
15:16:26LinusNit could be the same function that the radio code uses for the screen update
15:16:41amiconnYou need to distiguish whether the ui code is called via callback or directly, or you have to duplicate the ui code
15:17:01Zagorno we don't. both functions can call radio_drawscreen()
15:17:34LinusNbut the button polling might be slightly different in the callback context
15:18:00Zagoryes, because it is different functions. when not searching, you naturally don't have a button to abort it...
15:18:06amiconnyup, that's what I mean
15:18:12LinusN(btw, this is exactly what i want to avoid by not using the autotune function)
15:18:17Zagorsort of like you can't stop ffwd in pause mode
15:19:03LinusNbut the extra complexity in the button handler is nothing compared to adding a new thread
15:20:10Zagorjust imagine all the nice cleanup code. "oh, do we have a search running? we need to stop it before exiting radio screen"
15:21:11Zagorhow do you stop a search? are some buttons conditional depending on if you are in search state or not?
15:22:43LinusNcurrently, you stop the search by clicking left or right
15:22:53LinusNi.e tune manually
15:23:00LinusNiirc
15:32:50elinenbeLinusN: how is the iriver progress coming? anything on the drawing board right now?
15:34:34Zagorwhich is the better win32 program: eac or cdex?
15:35:47Zagorbtw, Bagder: i tried upgrading my plextor firmware but it still freaks out on bad disks.
15:35:52kurzhaarrockerDepends on wether you prefer emacs or vi? :)
15:35:56BagderZagor: ok
15:36:29Zagorkurzhaarrocker: emacs. doesn't everyone? ;)
15:37:39kurzhaarrockerapropos thread: I'd still favor a thread to read out the peak meter registers :)
15:38:28kurzhaarrockerconcerning eac vs cdex: I used both and both worked to my satisfaction. I don't prefer either over the other.
15:38:39Zagorso they're pretty much the same?
15:38:52kurzhaarrockerAt least the result is pretty much the same.
15:39:54LinusNelinenbe: will probably work on it this evening
15:40:08LinusNtimers and LCD driver is next
15:40:26kurzhaarrockerLinusN is our lonely worrier at the iriver front.
15:40:54Zagori'm looking at the mp3 encoder right now
15:41:18LinusNShine
15:41:34Zagoryeah
15:41:43 Join JoyOfDuck [0] (~d98680b3@labb.contactor.se)
15:43:12JoyOfDuckHello - Can anyone spare a minute to shed light on a mis-behaving Jukebox Studio
15:43:23LinusNshoot
15:43:55JoyOfDuckthankyou - My little box of tricks just stopped working
15:44:07JoyOfDuckIt only powers on when the USB cable is conmnected
15:44:25JoyOfDuckAppears to work fine in USB mode, but not at all in play, or battery charge mode
15:44:34JoyOfDuckThought it might be a hardware issue
15:44:54kurzhaarrockerAre the batteries really fully charged? Might they be old and broken?
15:45:14JoyOfDuckI thought it might be the batteries, so changed them for new ones
15:45:42LinusNwhat happens when you try to turn it on?
15:46:07JoyOfDuckThe green LED flicks on once - no disc motor, no backlight
15:46:26LinusNdoes it charge the batteries?
15:46:35JoyOfDuckAppears not
15:46:46LinusNmy guess is that the IRF7416 is broken
15:46:49JoyOfDuckI took it apart and checked voltages
15:47:01JoyOfDuckGot les than three volts across the battery terminals
15:47:06LinusNwhat heppened when it stopped working?
15:47:26LinusN3 volts
15:47:34kurzhaarrockerthat's nearly nothing
15:47:38LinusNhave you checked the battery terminals themselves?
15:47:42JoyOfDuckI left it charging for a few hours and noticed that it wasn't warming up
15:47:49JoyOfDuckTried switching it on and it was dead
15:47:51LinusNthe connections to the small PCB:s
15:48:02kurzhaarrockerOr maybe you checked the wrong terminals and measured only two batteries?
15:48:15JoyOfDuckI checked the battery termansl at the top
15:48:26kurzhaarrockerok
15:48:27JoyOfDuckAlso checked the two solder tabs at the top end of the board
15:48:34LinusNthen your batteries are discharged
15:48:48JoyOfDuckThey are brand new, so I expect that they are discharged
15:49:35JoyOfDuckIs the IRF7416 the battery chargin component?
15:50:21JoyOfDuckI was prepared to lose my favourate toy, but then noticed it worked fine with a USB cable plugged in
15:50:32JoyOfDuckThought that might indicate a simple hardware problem
15:52:02JoyOfDuckOh well, that looks like it. Thanks for your time - I'll start saving for an iRiver
15:52:04kurzhaarrockerTo start charging you have to leave them for serveral minutes in the jukebox
15:52:26JoyOfDuckOh yes - I have left them variously for several hours
15:52:45kurzhaarrockerIf you have an external charger you might want to try charging them separately.
15:53:02JoyOfDuckGood idea - I'll try and borrow one
15:53:24JoyOfDuckMight be able to limp along until Christmas :-)
15:54:24JoyOfDuckIf anyone thinks of anything, my addy is "joy at beyond2000.co.uk"
15:54:27JoyOfDuckThanks again
15:54:59LinusNyou're welcome
15:55:53LinusNgotta go now, cu guys
15:55:57 Quit webguest64 ("CGI:IRC (EOF)")
15:55:59 Part LinusN
15:57:20amiconnkurzhaarrocker: Iirc he said he has a Studio - Players charging is hardware
16:00
16:02:56 Join Lynx_ [0] (lynx@134.95.189.59)
16:05:31kurzhaarrockerright, amiconn, stupid me.
16:05:50JoyOfDuckHelo are we still thinking about my Studio 20 issue?
16:15:40amiconnJoyOfDuck: I'd try the following things:
16:15:52JoyOfDuckHello
16:16:18amiconn(1) Take a set of 4 new batteries, charge them externally and put them into the box. Does it start?
16:16:40amiconnIf yes, the charging circuit in the box is most likely broken.
16:17:50JoyOfDuckThanks - I plan doing this first point following the earllier suggestion
16:18:19amiconnIf it does not start, (2) measure the voltage across the battery contacts, both at zero load (box off) and while trying to start.
16:19:12amiconnIf the voltage drops below ~4.5 V while trying to start (with fully charged cells) the battery contacts are broken
16:20:26amiconnAnyone here who knows which of the 4 contacts are the ends (can't check myself atm) of the 4-cell circuit?
16:21:53 Quit Zagor (burroughs.freenode.net irc.freenode.net)
16:21:53NSplitburroughs.freenode.net irc.freenode.net
16:23:15NHealburroughs.freenode.net irc.freenode.net
16:23:15NJoinZagor [242] (~bjst@labb.contactor.se)
16:24:14JoyOfDuckI see a photo on the Rockbox site at http://www.rockbox.org/docs/solderjoints.jpg
16:26:05JoyOfDuckI also had the second symptom of broken contacts: Reboot when the bumpers are compressed
16:26:10Zagorhmm, the "fixed point" implementation of Shine uses double, log and sqrt...
16:27:06amiconnJoyOfDuck: The pictures are for a JB recorder, but the problem most likely also exists for players.
16:27:20JoyOfDuckAhah!
16:27:37JoyOfDuckThanks again for the help - I'll report back once I have a working player :-)
16:47:11kurzhaarrockerWell the peak meter code contains some log too, Zagor :)
16:53:36Zagor:)
16:54:14Zagorsee you
16:54:15 Part Zagor
16:55:22 Quit ashridah ("sleep")
16:58:53***Saving seen data "./dancer.seen"
17:00
17:01:09 Join gromit``` [0] (~gromit@ALagny-151-2-2-244.w82-121.abo.wanadoo.fr)
17:06:27 Join methangas [0] (methangas@0x50a476bc.virnxx10.adsl-dhcp.tele.dk)
17:18:22 Quit gromit`` (Read error: 110 (Connection timed out))
17:22:03 Join mecraw__ [0] (~lmarlow@69.2.235.2)
17:25:59 Quit [IDC]Dragon ("CGI:IRC")
17:34:20 Part pillo
17:36:26 Part kurzhaarrocker
18:00
18:03:52 Quit einhirn (Read error: 54 (Connection reset by peer))
18:15:44 Quit edx (Read error: 54 (Connection reset by peer))
18:21:48 Join elinenbe_ [0] (~elinenbe_@65.115.46.225)
18:39:39 Quit elinenbe (Read error: 110 (Connection timed out))
18:39:39 Nick elinenbe_ is now known as elinenbe (~elinenbe_@65.115.46.225)
18:42:18 Join _aLF [0] (Alexandre@mutualite-3-82-67-66-128.fbx.proxad.net)
18:58:57***Saving seen data "./dancer.seen"
19:00
19:54:44 Join schnauz [0] (zazz@dh051-118.chem.sunysb.edu)
20:00
20:13:04 Quit Headie (Read error: 110 (Connection timed out))
20:29:17 Quit _Lucretia (Connection timed out)
20:29:49 Join [av]bani [0] (~goemon@washuu.anime.net)
20:29:59 Part [av]bani
20:29:59 Join _Lucretia [0] (~munkee@abyss2.demon.co.uk)
20:42:27 Join Headie [0] (~hehe@fsto6.sto.sema.se)
20:59:00***Saving seen data "./dancer.seen"
21:00
21:01:06 Quit methangas (" HydraIRC -> http://www.hydrairc.com <- IRC has never been so cool")
21:07:56 Join edooo [0] (~edooo_188@82.129.244.5)
21:11:24 Quit JoyOfDuck ("CGI:IRC (Ping timeout)")
21:13:17 Quit edooo (Read error: 54 (Connection reset by peer))
21:15:47 Nick _Lucretia is now known as _Lucretia_ (~munkee@abyss2.demon.co.uk)
21:28:37 Join edooo [0] (~edooo_188@82.129.244.5)
21:32:10 Quit edooo (Read error: 54 (Connection reset by peer))
21:34:55 Join edooo [0] (~edooo_188@82.129.244.5)
21:35:47 Quit edooo (Read error: 54 (Connection reset by peer))
21:49:51 Join scott666_ [0] (~scott666@c-24-245-58-48.mn.client2.attbi.com)
21:54:17 Join gmb [0] (~441ae028@labb.contactor.se)
21:55:34 Quit gmb (Client Quit)
22:00
22:23:42 Join dropandhop [0] (~43646aab@labb.contactor.se)
22:23:47dropandhophey all!
22:23:53dropandhopiriver user here
22:23:57dropandhopwas wondering how i could help
22:24:01dropandhopi can't code tho!
22:34:20 Quit elinenbe (" Try HydraIRC -> http://www.hydrairc.com <-")
22:41:14 Join [av]bani [0] (~goemon@washuu.anime.net)
22:41:22 Part [av]bani
22:46:53 Join Bluechip [0] (~BlueChip@cpc3-colc1-3-0-cust61.colc.cable.ntl.com)
22:59:01***Saving seen data "./dancer.seen"
23:00
23:26:25 Nick scott666_ is now known as scott666 (~scott666@c-24-245-58-48.mn.client2.attbi.com)
23:36:56 Quit dropandhop ("CGI:IRC")
23:38:57 Join gromit` [0] (~gromit@ALagny-151-2-2-244.w82-121.abo.wanadoo.fr)
23:38:57 Quit gromit``` (Read error: 104 (Connection reset by peer))
23:52:20 Join iRiverMan [0] (~acca5915@labb.contactor.se)
23:52:26iRiverManhey people
23:52:59iRiverMancan someone help me regarding an iriver please?
23:53:12 Quit Hadaka (No route to host)

Previous day | Next day