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 2002-12-05

00:08:27***Saving seen data "./dancer.seen"
00:12:36 Quit [keno] (Read error: 110 (Connection timed out))
00:38:48 Quit TotMacher ()
00:44:03 Join TotMacher [0] (
00:45:57 Part LinusN|away
00:53:17 Quit TotMacher ()
01:11:25 Join LinusN [0] (
01:35:18adi|sleephey linus
01:35:25adi|sleepi understand.. my keyboard design wasn't good enough
01:35:27*adi|sleep sniffles
01:35:30*adi|sleep smirks
01:37:28 Join TotMacher [0] (
01:47:49LinusNadi|sleep: no it's just that we don't like you :-)
02:08:31***Saving seen data "./dancer.seen"
02:10:45 Part LinusN
02:17:10 Nick _seb_ is now known as seb-away (
02:32:21 Quit TotMacher ()
02:32:27 Quit probonic ()
02:36:12 Quit datazone ("be back in a jiffy!")
03:44:02 Join games [0] (
03:44:43gameshey, has anyone ever heard of that gut, "Blueloop"?
03:45:28gamesor, does anyone know how to put on a patch?
03:46:20gamesrockbox sucks because thier isn't enough games
03:46:52 Part games
03:58:00 Nick seb-away is now known as _seb_ (
03:59:52 Quit Moof ("Client Exiting")
04:08:33***Saving seen data "./dancer.seen"
05:32:37 Join [keno] [0] (marklar2@
05:32:37 Quit k3no (Read error: 104 (Connection reset by peer))
05:53:39 Nick _seb_ is now known as seb-sleep (
06:08:35***Saving seen data "./dancer.seen"
07:07:25 Join Zagor [242] (
07:25:39 Quit mecraw1212 (Read error: 110 (Connection timed out))
07:25:48 Join mecraw1212 [0] (
08:08:38***Saving seen data "./dancer.seen"
08:13:20 Nick dw|gone is now known as dwihno (dwihno@Bald067.Baldakinen.Umea.SE)
08:13:21dwihnoHej gojs
08:16:13dwihnoHola seor, une cerveza grande, porrfarbror :)
08:33:07 Quit Zagor ("Client Exiting")
09:03:18 Join bobTHC [0] (
09:04:24 Join Zagor_ [242] (
09:04:56bobTHCgood morning!
09:05:04 Nick Zagor_ is now known as Zagor (
09:28:59 Join pyvasene [0] (~pyvasene@
09:29:14bobTHChi pyva!
09:29:38 Quit pyvasene (Client Quit)
09:29:41 Join pyvasene [0] (~pyvasene@
09:44:39 Join adi|dads [0] (
09:44:49adi|dadsZagor you around?
09:50:45adi|dadsjust looking for info...
09:51:00adi|dadswhat would have made my version of the keyboard better/more optimal?
09:51:37adi|dadsoh.. and have you played with the rockboxui sim lately?
09:51:51Zagori don't know. i looked at both versions, and just picked one. I felt Johns was a bit simpler.
09:51:57adi|dadsit seems to not fully flush the boot screen when it gets to the dir listing.
09:52:06Zagoryeah, i did the keyboard exclusively in the sims
09:52:47adi|dadsi have a completely empty archos dir in my sim directory
09:53:02adi|dadswhen i start the sim i don't get a full screen flush.. not sure why
09:54:07adi|dadsand not that im trying to be a pain ;)
09:54:13adi|dadson the buglist page
09:54:25Zagorok. i'll look at that
09:54:29adi|dadsany chance we could get 'won't fix' and 'fixed' to be diff table colors?
09:54:34adi|dadsjust so they stand out a bit more
09:54:52adi|dadsbackground wise
09:55:14Zagoruh? won't fix is red and fixed is green
09:55:38adi|dadsthe backgrounds are?
09:56:43adi|dadsnot on my version of netscape they aren't/
09:56:46adi|dadschekcing in opera right now
09:57:09adi|dadsill be damned
09:57:14adi|dadsin opera they are
09:57:19adi|dadsmy version of netscape they aren't
09:57:31adi|dadscommunicator 4.7
09:57:34adi|dadshow odd
09:57:34Zagorwhich version?
09:58:26Zagorthe colors are done in CSS. i guess NS4.7 didn't support it that well
09:58:50adi|dadshmmm... possibly
10:00:10 Join LinusN [0] (
10:00:38Zagorhi LinusN
10:00:45adi|dadsoh damn he's back
10:00:57Zagoreverybody quiet now
10:01:16LinusNremember that i can read the log :-)
10:02:22adi|dadsthats what you think
10:02:22Zagornot if I censor it :)
10:03:56bobTHC<<<<updated francais.lang commited!>>>>
10:05:54dwihnoC'est parfait!
10:06:22dwihnoTre's bien
10:08:41***Saving seen data "./dancer.seen"
10:09:11adi|dadsSo, what do you call an American with common sense and a good education?
10:09:21Zagorrare? ;)
10:09:24 Join Mine78 [0] (
10:09:27adi|dadsAn immigrant :)
10:09:39adi|dadsfigured you'd like that :)
10:09:59adi|dadsWhats the one thing Canada has that America doesn't?
10:10:07adi|dadssorry.. reverst that.
10:10:13Mine78Hi guys... do you remember the yesterday battery problem ? Well 1 of the 4 batteries was destroyed... with a tester it give oute 0.1 Volt... and I bought 4 new cells :(
10:10:20adi|dadsWhats the one thing America has that Canada doesn't?
10:10:42LinusNi dunno
10:10:44adi|dadsNice neighbors
10:11:03LinusNMine78: that's funny
10:11:15LinusNyou must have fried them good
10:11:27ZagorMine78: actually it's best to always use them in sets of 4. don't mix brands/models
10:11:47Zagordifferent brands have different charging characteristics
10:12:06adi|dadsSo what do you call and old britsh woman with inbreed kids that doesn't page taxes?
10:12:09Mine78new cell are 1600milliampere... can I use them with 1500 ?
10:12:32Zagordon't mix
10:12:56LinusNdon't even mix older with newer, even if they are the same kind
10:13:07Mine78but now I have 3 originale 1500 milliampere... almost new
10:13:33LinusNadi|dads: ?
10:13:58adi|dadsHer majesty
10:14:17ZagorMine78: use them for your TV remote control, bicycle lights or whatever. don't use them in your archos
10:14:29Mine78I risk to damage the archos ?
10:14:42ZagorMine78: if you mix batteries, you will damage the batteries again
10:15:08Mine78ok thank you... I am boring I know... really thanx
10:15:13Zagorno problem
10:15:30LinusNMine78: how do you charge them?
10:15:41LinusNi mean your charging habits
10:15:50adi|dadscatch them doing something illegal
10:15:53*adi|dads smirks
10:16:17 Quit [keno] (Read error: 54 (Connection reset by peer))
10:16:22LinusNadi|dads: what was the question?
10:16:23 Join [keno] [0] (marklar2@
10:16:29bobTHCfrench ppl : I want your suggestions for better translation in francais.lang
10:16:40Mine78Well in the 1st month I drain them each 24 hours uploading and playing mp3
10:17:38Mine78And i noticed that rockbox charging is more precise than archos charging... archos never says "charged", rockbox yes, after almost 4 hours
10:18:05Mine78Then at the beginning of 2nd moth they broke down
10:20:31 Quit Mine78 ("Shokowahhhhh dwahhhnowan!!")
10:23:06ZagorUwe's new charging code charges the batteries more (higher top voltage, plus some other changes). can anyone come up with a good way of testing the various charging algorithms?
10:23:31LinusNwe still need trickle charge imho
10:24:02ZagorI agree. We wouldn't need it if we could turn off the archos, but since we can't...
10:24:15Zagoralso I guess trickle is a good way to "top off" the charge
10:24:25LinusNi have read that too
10:25:20ZagorLinusN: are you planning to do some tests with dynamic watermark?
10:30:59LinusNno, at least not yet
10:31:23LinusNwell, i plan to, but not in the nearest future
10:31:45Zagorit seems hadaka was right and I was wrong. poweroff adds almost a second to my spinup time.
10:32:06LinusNdoes the mesurement work now?
10:32:15LinusNthe spinup time measurement
10:32:16Zagoryes, with my upcoming commit
10:32:27LinusNok, what time does it measure?
10:32:36LinusNfrom where to where?
10:32:52Zagorfrom poweron/soft-reset to completion of first read/write command (before buffer copy).
10:33:17LinusNok, from poweron to DRQ?
10:33:57LinusNdo you see any fluctuations in the timing?
10:34:05Zagorhaven't done many tests yet
10:35:35dwihnoIs there some easy way to split a byte into two parts (5, 3) ?
10:35:47LinusNin c? no.
10:35:49Zagorvery little fluctuation so far. 20 ms difference.
10:36:26LinusNa = x & 7;
10:36:35LinusNb = (x >> 3) & 0x1f;
10:36:50Zagordwihno: part1 = byte >> 3; part2 = byte & 7;
10:37:16dwihnobits and bones may break my bones, but pointers will never harm me :)
10:37:42ZagorLinusN: 100ms diff this time
10:37:59LinusNZagor: time for a histogram
10:38:49dwihnoLinusN: why do you & the shifted x?
10:38:56ZagorLinusN: could be an idea
10:39:07Zagordwihno: in case the input is larger than 8 bits
10:40:20LinusNdwihno: if x is unsigned char, it is not necessary
10:40:35dwihnoYou are like an educational centre :)
10:42:03LinusNactually, signed or unsigned does not matter
10:42:11LinusNas long as it is char
10:45:06LinusNZagor: do you have a rename function in fat.c?
10:45:37LinusNok, so a rename feature would be peanuts to implement?
10:45:52Zagoryes. I was thinking of making ON+PLAY in the browser bring up a quick-menu for file operations.
10:46:11LinusNgood idea
10:46:23LinusNcan you do mv?
10:46:35Zagornot yet
10:47:03Zagoralso the player keyboard is rather buggy still :-)
10:47:16LinusNoh, in what way?
10:47:42Zagorit simply needs more work
10:47:44LinusNbtw, 2.4.20 didn't solve my problems
10:48:53adi|dadsdefine buggy on the keyboard?
10:49:16Zagorthe left-right scroll is not limited properly, so you can scroll into the next page :-)
10:49:35Zagor...and, ultimately, scroll into all kinds of weird memory where you really shouldn't be :)
10:49:41*adi|dads wistles innocently
10:49:41LinusNZagor: have you removed logs from CVS?
10:49:50*adi|dads points Zagor at his version of the keyboard
10:49:53*adi|dads smirks
10:50:08Zagoryes, the old irc .logs. they are replaced by .txt logs
10:51:05Zagoradi|dads: uh, the only version I saw only works for recorder!?
10:51:41adi|dadsright, and?
10:52:14adi|dadsi didn't port to the player because i was told at the time it would be a while before we needed a keyboard and not to commit it because we didn't know where it would go :)
10:52:33adi|dadsthe keyboard you commited does not have a player version does it?
10:52:52Zagoryes it does. it is that version which bugs. the recorder version works fine.
10:54:06adi|dadssee.. you didn't say that ;)
10:55:25Zagor<Zagor> also the player keyboard is rather buggy still :-)
10:56:34 Join kargatron [0] (
10:56:47*adi|dads yells at his monitor for increasing his blind spots
10:56:50adi|dadssee.. my dad was right
10:56:55adi|dadsit _is_ making me go blind
10:56:59adi|dadsim sorry. ('
10:57:45kargatroni occasionally get what's termed an 'optical migraine' - painless, but i get this temporary huge blind spot. annoying.
10:58:04adi|dadsbtw.. 4.8 doesn't fix the css problem either
10:58:10adi|dadsgoing to try 6.* tomorrow
10:58:14adi|dadsi refuse to go to 7
10:58:15LinusNsara gets that too in some rare occasions
10:58:43Zagorthere's a 7?
10:58:46adi|dadsi get that to... generally.. when Ive been on the machine for hours..
10:58:51kargatronFeature req 646978 vs 645901 on char scrolling - are those dupes? they seem to ask for different implementations
10:59:02adi|dadsbut then again.. comes sunday night, and she leaves to go back home again.
10:59:05*adi|dads smirks
10:59:23kargatronone is for "bitmapped graphic demo" type, and one is for char-by-char
10:59:46adi|dadsthey are similar i think
10:59:55adi|dadsbtw.. what do we think of that char-by-char idea?
11:00:05kargatronshould i mark one dupe and just transfer a note then?
11:00:14adi|dadsi think thats good
11:00:18Zagorhe's approaching it all wrong. char-by-char is not the solution. lower update frequency is.
11:00:22adi|dadstransfer the 'bitmap' one...
11:00:26adi|dadsahh.. okay...
11:00:54adi|dadsbut with a lower update frequecy we get a slower scroll no?
11:00:55Zagori think we can reject the smooth request. we can't do fast smooth scrolls. they get unreadable.
11:00:56kargatronlower update freq - that different from scroll speed?
11:01:19Zagorkargatron: yes, we actually have two parameters: pixels to move the text per update, and updates per second
11:01:31kargatronwell, the point the earlier one makes is that he finds the archos scrolling easier to read. pure preference? or what?
11:01:43Zagorcurrently we can only change the pixel parameter, but we might want to add update frequency too
11:01:58Zagorit's easier to read because it's slower
11:02:03adi|dadshow is the bounce text demo diff from scolling in how it moves the chars?
11:02:15adi|dadsbecause i find that easy to read.. but file names scrolling tough
11:02:57Zagorcompletely different. also, it's easier because the font is bold
11:03:19adi|dadsdunno about that... i use a bold font while driving for file names...
11:03:26adi|dadsthat don't seem to help all that much
11:03:35adi|dadsbut i see what you're saying
11:04:05adi|dads621115 feature
11:04:09kargatroni myself am not crazy about the scrolling, but i don't know enough to think it's a legit complaint. I use uwe_prop, which is quite small anyway
11:04:19adi|dadsis that possible?
11:04:29adi|dadsWould it be possible to have the battery full percentage
11:04:29adi|dads shown during charging while off? Make it possilbe to
11:04:29adi|dads know when it is safe to take it for a full days use. thanks
11:04:35adi|dadssorry to spam
11:04:47adi|dadswe don't have control of the screen at that point do we?
11:05:17 Part Zagor
11:05:22adi|dadsi scared him off
11:05:56 Join Zagor_ [242] (
11:06:02 Nick Zagor_ is now known as Zagor (
11:07:40kargatronZagor,, if you want to make any comments on the scrolling thing.
11:07:51Zagoryeah, I saw. thanks
11:08:23adi|dadszagor we don't have control of the screen when the unit is off and charging right?
11:09:04kargatronaha, just figured out why i sometimes appear to get sf mail dupes.
11:09:29adi|dadswhys that?
11:09:30kargatronsf emails separately, then i also get list email of same thing.
11:10:05kargatronshouldn't anyone logged into sf see that?
11:10:55ZagorI guess maybe I get that too, but I have dupe elimination in my mail filter :-)
11:11:00kargatronreminder (not that it matters much), I will be offline dec 8-28, vacation
11:11:33ZagorThailand, was it?
11:11:55*Zagor is not envious at all :)
11:12:22kargatroni'll letcha know how not envious you should remain when i return. :)
11:12:34*adi|dads is an American and as such has been warned to _not_ visit that nation :)
11:13:01*kargatron is also, but thinks the warning is paranoid
11:13:21adi|dadshehe i agree...
11:13:33adi|dadsbut finances are what dictate my level of paranoia ;)
11:13:37kargatronor if not paranoid, overblown in interpreted severity
11:14:25dwihnoZagor: Good idea (on+play = file operations)
11:14:32adi|dadsbut we HAVE to go to war with Iraq... i mean.. they are letting in the inspectors and everything...
11:14:39adi|dadswhat i mean is.. um.. yeah...!
11:16:00adi|dads"Bush, the second. Half the votes and double the incompetence."
11:16:04bobTHCamerican always thing they are world cops...
11:16:23adi|dadstrue, but generally there is a good reasoon for that
11:16:40kargatrona very debatable statement. :)
11:16:52adi|dadswhich, his or mine :)
11:17:40bobTHCouch i'm not really agree with you, for me the only reason is business
11:17:51adi|dadsnah.. the reason is simple... it is (generally mind you) america that the UN and other nations (and their corporations) look to for monetary, humanitarian or military aide.
11:17:56*kargatron abstains from any further OT discussion :)
11:18:24Zagormonetary? uh, right. when did the US last pay their UN dues? 1970?
11:18:36*Zagor loves a good flame war :)
11:18:42bobTHCbusiness and world supremacy are the only two goal of USA
11:19:12adi|dadsZagor good point.. but what nations came to the aide of the Red Cross when it nearly went bankrupt in the 80's while helping other nations?
11:19:27adi|dadsit was canada and the us government that bailed out a US aide organization
11:20:01adi|dadsor how about those 'small' loans that were 'paid' back to help rebuild europe?
11:20:03adi|dadstwice ;)
11:20:07Zagori think you will find it difficult to convince the rest of the world that the US has a good foreign policy :-)
11:20:20adi|dadsim not arguing with you there
11:21:16adi|dadsbut you will have a tough time finding a country that doesn't want us aide when they are starving, broke, or being invaded.
11:21:17bobTHCwhat the only country who dont sign any ecological agreement ?
11:21:26adi|dads(well, a democratic country at least)
11:21:36bobTHCof course
11:21:46Zagoradi|dads: you're confusing US with NATO, imho.
11:21:57bobTHCi dont talk about liberia or afgani
11:22:02adi|dadswhy doyou say that Zagor?
11:22:33adi|dadsbobTHC I never said the country was perfect...
11:22:53bobTHCwhen ppl starving, usa send transgenic corn without any test on the health before
11:23:03Zagorbecause NATO is who helps UN when places need protection (bosnia etc). and NATO is a lot more than just the US
11:23:06adi|dadsmy personal opinion is.. frankly.. screw the rest of the world.. time we took care of our own first/only
11:23:07bobTHCstarving ppl are beta tester
11:23:12*kargatron is going to start sneaking in his fave unimplemented feature requests in this conversation. :)
11:23:17adi|dadsbobTHC thats crap and you know it...
11:23:27Zagorkargatron: hehe
11:23:27adi|dadsoh, i agree with you Zagor
11:23:46*dwihno is going to sneak in configurable scroll speed
11:23:53kargatronusa sux ID3 YEAR usa sux
11:24:04kargatrongo usa BOOKMARKING yea us
11:24:09adi|dadsbobTHC where are you from?
11:24:32LinusNhe's from CHIP8 France
11:24:49adi|dadssee.. he's got a prob from the start...
11:24:51kargatrondwihno, scroll speed is configurable. are you referring to something else?
11:24:55bobTHCadi: not, read more about the congo food help , congo take it but dont distribute it to his ppl only @ refugees
11:25:00adi|dadshe's from a country that can't even build a complete wall ;)
11:25:20dwihnokargatron: scroll speed isn't configurable, only the amounts of pixels being scrolled
11:25:25bobTHCi'm not glad of my country @ all
11:25:30adi|dadsbobTHC read more about America helping Mexico after their 10 major earthquakes
11:25:37adi|dadsor helping in S. America
11:25:47kargatronwell, it's *called* scroll speed! :)
11:26:00bobTHCdont talk with me about S america US politic
11:26:11LinusNcome on guys
11:26:11adi|dadsyour right.. lets go back to europe..
11:26:26bobTHCbecause it's the best exeample of imperialism
11:26:27dwihnokargatron: that's false advertising :) I tried to convince Linus to make it configurable too yesterday, and Hadaka agreed :D
11:26:32Zagorid3v2 is really stupid. no genre and no year tags.
11:26:32adi|dadsdid you guys ever finish paying us back for those small loans to help you guys out? (france i mean)?
11:26:53kargatronno year? what do you mean?
11:26:58LinusNthe genre tags are stupid imho
11:27:03adi|dadsengland held the world by the point of a gun for 300 years and your going to bitch to me aobut imperialism?
11:27:13Zagoradi|dads: drop it, ok?
11:27:22adi|dadshehehehe but this is fun ;)
11:27:27adi|dadsi like picking on frogs :)
11:28:08ZagorLinusN: yeah, I can agree about genre. but a year tag would be useful.
11:28:24LinusNi can't read
11:28:26kargatronpage not coming up
11:28:31adi|dadsZagor why would you feel a year tag to be useful?
11:28:32Zagornow there are two tags instead: TCOP and TPRO
11:28:34bobTHCduring 40years USA give money to gerilleros and to military goverment to be sure of having the partnership without attempt to make the region sure
11:28:44Zagorboth free-form text tags...
11:28:45kargatronhow cn a year tag not be useful?
11:28:57*bobTHC stop the USA troll
11:29:12LinusNyear is interesting imho
11:29:12Zagoradi|dads: i like to know what year the song was recorded/written
11:29:15adi|dadsbobTHC at Zagor's request im dropping it :) (you surrendering wussy frog you ;) )
11:29:37adi|dadsbut in the same vein, genre could be important to someone
11:29:41kargatronhm, can't get to, darnit
11:29:46Zagoradi|dads: i agree
11:29:46LinusNme neither
11:29:48*adi|dads doesn't care for either, but thats another story
11:29:57Zagortime to mirror it, i'd say
11:29:57bobTHChehe is down very frequently
11:30:19bobTHCsir,yes sir
11:30:24kargatronnot caring about year clearly labels you as a musical dabbler, adi|dads
11:30:32Zagoryes. i'll upload it to
11:30:42kargatronyear = history = context!
11:30:42adi|dadsdamnit.. for 2 weeks my dad bitched.. 'the cdrom isn't working' i tell him.. did you check the power cable going to it. "Yes im not stupid."
11:30:45adi|dadsso i check it...
11:30:56adi|dadsguess what was _not_ hooked up?
11:31:04LinusNthe floppy?
11:31:09LinusNthe hard disk?
11:31:12LinusNthe toaster?
11:31:14kargatronhis brain?
11:31:23adi|dadsbing bing bing
11:31:27bobTHCthe TV?
11:31:27adi|dadsgive kargatron a cigar
11:31:42LinusNoh, so it wasn't the cdrom after all
11:31:57adi|dadsi blame it on NATO
11:31:59*adi|dads ducks
11:32:31kargatronso i have mostly v2.3 tags - there's year there - 2.4 changes it? or is this an internal invisible thing?
11:32:35Zagoroh, this just gets better. id3v2.3.0 has TORY (Original release year). id3v2.4.0 doesn't!
11:32:48LinusNit's sooo silly
11:33:03kargatronand tcop/tpro are what?
11:33:06adi|dadswtf decides how the id3v2 is set up?
11:33:13kargatron(since i can't get to atm)
11:33:22Zagor*and* it has TYER, which is "The 'Year' frame is a numeric string with a year of the recording."
11:33:53LinusNso TYER is it?
11:34:55Zagortcop and tpro are Copyright and Producer free-format text strings
11:35:52Zagorfor 2.3 yes
11:38:29kargatroni encode with v1 and v2.3. i think all my mp3s have v1, and the newer both.
11:39:33 Nick kargatron is now known as karg|fud (
11:44:20Zagorid3 specs added to our documentation page
11:47:46 Join Bagder [0] (
11:48:45LinusNho Bagder
11:54:11adi|dadshmmm.. a hanging bug was reported
11:54:19adi|dadsive had it too recently, but i thought it was just me
11:54:24adi|dadsi get it when im in the quick menus
11:55:59 Part Zagor
11:56:35 Join Zagor_ [242] (
11:56:46 Nick Zagor_ is now known as Zagor (
11:56:58Zagori love this proxy :(
11:57:12LinusNZagor: how come you don't use muh?
11:57:18ZagorI do
11:57:26Bagderhe does, check his host
11:57:34LinusNthen why change nick?
11:57:45Zagorbecause I don't use muh from home
11:57:59LinusNw2hy not?
11:58:07Zagorno reason
11:58:49Bagderyou can have muh try different nicks anyway
11:59:22adi|dadskarg|fud you said sourceforge is sending you emails.. how did you turn that off?
12:00:09ZagorBagder: yes, but I don't want muh to hog 'Zagor', that's all
12:01:43LinusNZagor: i thought that was the point
12:02:21adi|dadsokay.. bed time.. night all.
12:02:23 Quit adi|dads ()
12:03:26ZagorLinusN: no, the only point for me is to access irc through a proxy.
12:06:00LinusNi use muh wherever i go
12:06:25Zagoryeah, I guess that would be simpler for me too
12:08:45***Saving seen data "./dancer.seen"
12:10:35LinusNi need a suggestion for a good dynamic watermark algo
12:10:52LinusNuse the bitrate infor from the header?
12:11:00LinusNtune in as the song plays?
12:11:21Zagorbitrate info is usually correct
12:11:50Zagorthat plus spinup time should be good enough. with a 50-100% margin
12:12:14LinusNbut when do we know that the spinup time info is correct?
12:12:31LinusNin case of resume
12:12:34Zagorwhen it is != 0
12:12:56LinusNresume -> song starts playing immediately
12:13:03LinusNwith no spindown in between
12:14:04Zagorthe spindown timer only changes the value if the disk was sleeping
12:14:04Zagorso if no spinup, no spinup time
12:14:15LinusNok, use the max watermark until a measurement is made
12:14:55Zagoralso I guess we could fancy up the timing a bit and use average times
12:15:20LinusNyeah, a floating average
12:15:29LinusNor whatever it's called
12:16:17 Nick Zagor is now known as Zagor|lunch (
12:18:36 Nick karg|fud is now known as kargatron (
12:25:03bobTHCbye all
12:25:21 Quit bobTHC ("weekend!!!! rulez")
12:27:52 Join tot|away [0] (
12:27:58 Quit tot|away (Remote closed the connection)
12:28:47 Join TotMacher [0] (
12:30:21*Bagder is away: I'm busy
12:35:03*dwihno ate weird lunch: pasta con carne :)
12:49:06dwihnoLinusN: Wanna know something funny? The kebab-sng works flawlessly with the archos firmware :)
12:54:07LinusNdwihno: give it to me
12:55:23 Quit TotMacher ()
12:55:39 Join TotMacher [0] (
12:55:41 Quit pyvasene ()
12:59:21*Bagder is back (gone 00:29:00)
13:01:48 Nick Zagor|lunch is now known as Zagor (
13:03:50LinusNdynamic bitrate coming up rsn
13:04:55dwihnoYay! :D
13:04:58dwihnoYay for L
13:05:09 Join quelsaruk [0] (
13:05:20dwihnoLinus: You got file
13:05:31LinusNdwihno: i can't do dcc
13:05:58quelsarukLinusN: if you want use my ftp
13:05:58Zagorah, now there's a reason for not always using muh :-)
13:06:37Bagdernot really
13:06:42Bagdermuh deals with dcc
13:06:46Zagorit does?
13:07:02Zagorso why can't linus use it?
13:07:10BagderI don't know
13:07:18Bagderhis dlink perhaps
13:07:19dwihnoonly cowards use muh ;D
13:07:28dwihnohehu, muh
13:07:30LinusNBagder: i couldn't do it before dlink either
13:07:45Bagderthen it's because Linus is incompetent ;-)
13:07:50*Bagder hides
13:08:47dwihnoLinusN: You got /msg
13:09:04dwihnoOr perhaps muh re-routed them to /dev/null ;)
13:09:19Hadakaho hum
13:10:51dwihnoHello Hadaka
13:10:58dwihnoHadaka: Did you implement variable scroll speed? :)
13:11:37 Quit TotMacher ()
13:11:56LinusNHadaka: what watermark settings worked for you?
13:13:00Zagordwihno: it's already implemented. we just need to add a setting
13:13:30*dwihno gasps
13:14:57 Join TotMacher [0] (
13:16:12Hadakadwihno: no :)
13:16:38HadakaLinusN: nothing solved the spin-up-while-moving problem
13:16:51HadakaLinusN: but 60000/40000 was the best setting
13:17:05dwihnoI read the posts about battery in the mailing list.
13:17:25dwihnoWill a change make a notable difference?
13:17:26Hadakabut I think the low watermark needs to be increased even more to take into account the 5 second spin up
13:17:28LinusNhmmm, maybe the firmware in the drive performs some tests while powering up, and those fail when the drive is shaking?
13:18:43dwihnoI note no spin up speed difference with my IBM drive
13:19:11kargatronHadaka, summarize - poweroff=ON gives you poor results when moving? better when it's OFF?
13:19:29Hadakakargatron: yes.
13:19:41HadakaLinusN: could be
13:19:43Zagordwihno: really?
13:20:08Hadakaand the delay between the spinup and reading was weird
13:20:09dwihnoZagor: I'll test rightaway, just to double-check
13:20:13Zagordwihno: use the spinup timer. Debug->Disk info
13:20:19LinusNdwihno: wow!
13:20:35LinusNthe kebab song sounds really funky!
13:20:39Hadakaoh btw - how is the spinup initiated? just by reading from the disk or is it told to spin up?
13:20:54dwihnoLinusN: it sure does... fishy, huh? it sounds good in the archos firmware (*eww*)
13:20:56ZagorHadaka: it's explicit
13:22:53dwihnoWith power-off: 2340, without 2510
13:23:00dwihnoIsn't that funky? :O
13:23:21Zagorthat's suspiciously funky
13:23:53dwihnoGot 2180 now
13:23:56HadakaI think my 20ms is a little more funky ;)
13:24:26Hadakanow if only gmane would liven up so i could read the mailing list again
13:25:59Zagordwihno: that could be the reason IBM drives use more power on spinup. they simply do it faster.
13:26:33Hadakathere's no way to measure power expenditure inside rockbox is there?
13:26:56Zagorthat would be great
13:28:15ZagorLinusN: I'm not sure I want my watermark at 320kbit level for my 96 kbit vbr tracks :-)
13:28:32LinusNZagor: and how would you like it then?
13:28:44LinusNthere may still be 320kbit frames in it
13:28:45Zagori'm not sure. how about avg.bitrate * 1.5 ?
13:29:05 Quit TotMacher (Read error: 113 (No route to host))
13:29:06Zagoryes, but average rate is pretty accurate when you're talking about one meg chunks
13:29:12LinusNand if not?
13:29:16 Nick quelsaruk is now known as quel|out (
13:29:28LinusNZagor: better safe than sorry
13:29:34dwihnoZagor: Might be...
13:29:38Zagorthe point is, this is no good
13:29:52Zagor75% of my tracks are vbr. now rockbox uses *more
13:29:55Zagor* power than before
13:29:57Hadakabut if safe means that the buffer will not be used as fully, it means spinups more often which means more battery
13:31:03Hadakamake it dynamic - and allow for a 'safety zone' setting that can be big or small, depending on one's preferences
13:31:15Hadakaand automatically big when running with the charger
13:31:44Zagornot a bad idea, actually. that would be like the adjustable mode some people have asked for.
13:31:46 Join TotMacher [0] (
13:31:53 Part TotMacher
13:32:07Zagornormally "safe level" should be at 0, but if you are out running a lot or otherwise have problems, bump it to 2-4 seconds
13:32:09 Join TotMacher [0] (
13:32:23LinusNwe have only one problem left: headerless VBR's
13:32:41Zagorthose i'm fine with running as 320 kbit
13:32:50LinusNZagor: we can't detect them
13:32:51Zagorthose are bastard type files anyway
13:32:56Zagorah, right
13:33:12LinusNwell, actually we can, but it would be a mess
13:33:36Zagorwhat do they show up as today?
13:34:06LinusNCBR, with the bitrate of the first frame
13:35:41kargatronwow, i'm totally lost. what is watermarking, and what is it for? It impacts battery usage? can reduce jolt-skipping? sorry for the ignorance
13:36:10Zagorthe low watermark is how empty the buffer gets before new data is loaded from disk
13:36:16LinusNkargatron: the amount of bytes left in the buffer when we should start reading in data again
13:36:46kargatronthought it was some kind of digital signature in the file or something - no wonder i was so confused :)
13:36:51LinusNi'll create a buffer margin setting
13:37:22LinusNbut we're out of RTC space :-(
13:37:33Zagorugh, right
13:37:42Bagderwe need RTC compression! ;-)
13:37:56LinusNi am tempted to toss out some of the peak meter settings
13:38:13kargatronpresumably, some settings are accessed much less often, and can be moved to real files?
13:38:28Zagoror at least moved to the non-rtc are of the settings
13:38:40Zagorsuch as peack meter...
13:40:13ZagorLinusN: don
13:40:14LinusNZagor: why on earth do you use little endian for the runtime meters?
13:40:40Zagorit wasn't on purpose. I just split into bytes.
13:41:04ZagorLinusN: what kind of margin do you have for CBR data? I can't see from the patch
13:41:14Zagori.e error margin, if the spinup takes too long etc
13:41:47LinusNi multiply the spinup time by 3, where 2 is adequate
13:41:51 Quit quel|out ("KVIrc 3.0.0-beta1 "Eve's Avatar"")
13:42:18Zagorso, basically, the disk powers up 10 seconds before data runs out?
13:42:21LinusNlater to be transformed into the buffer margin setting
13:42:35Zagorsounds good
13:42:47LinusN10 secs if the spinup time is 3.33 secs
13:43:11Zagori'll pull and test instead of askin a lot :-)
13:43:12 Nick dwihno is now known as dw|gone (dwihno@Bald067.Baldakinen.Umea.SE)
13:43:55LinusNi found some room in RTC
13:44:19Zagorif you only need a few bits, there's some space available
13:44:22LinusN0x25 0x39 <disk_spindown flag>
13:44:30LinusN6 bits free
13:45:29Zagor7 even. spindown is just one bit. (and it's wrong. poweroff is at 0x25. spindown is @ 0x1a)
13:45:45ZagorI wrote the comment wrong
13:46:09LinusNZagor: must leave one bit free, so the byte value doesn't accidentally become 0xff
13:46:57Zagorstill buffer margin should only be like 3 bits right?
13:47:06LinusN0x25 0x39 <disk poweroff flag, MP3 buffer margin>
13:47:41LinusN0-7 seconds?
13:47:57Zagoryeah, isn't that good?
13:48:00LinusNor indices in a list of values?
13:48:29Zagor0-7 is a good start. we can always make a list later if we want
13:48:32LinusNlet's start with 0-7 secs
14:08:49***Saving seen data "./dancer.seen"
14:10:44Zagoryikes. 'lowest' during a 128kbit CBR is ~110 KB
14:11:21Zagortesting 320
14:11:57LinusNremember, "lowest" doesn't tell the entire truth
14:12:23LinusNit seems like it never goes below C000 or something
14:12:27Zagor240 KB during 320 kbit
14:12:47Zagormeseems it's being a bit over-conservative :-)
14:13:11LinusNZagor: i tested with *2 and it wasn't enough
14:13:18Zagori have test files @
14:13:29LinusNwell, it was exactly on the margin
14:13:49Zagori'll test some too
14:13:52LinusNZagor: you mean the giana files? i use them for testing.
14:13:58Zagorok. good.
14:19:43LinusNwhen we have fine-tuned this function, we really can claim optimal battery usage :-)
14:20:35Zagorcan you look into the latest value, if it is incorrect. it's a good value to look at
14:22:37Zagoroh crapola. the id3v1 loader lseek()s to every little piece of info. that hurts!
14:23:34LinusN"lowest" should be correct now
14:23:41Zagoryou found somehting?
14:24:04LinusNit wasn't updated when it became 0
14:24:12LinusNso it never reached 0
14:24:30LinusNotherwise it is correct
14:25:59LinusNwe may want to account for the read() time as well
14:26:09LinusNand swap()
14:26:53LinusNthe most sensitive part is when data has been read, and the swapping hasn't begun yet
14:27:28LinusNah, i have an idea
14:27:38Zagorwe need a margin, for sure. i just want to make it as optimal as we can
14:28:04Zagorthe swapping time should be pretty linear with the length of the read data, no?
14:28:39LinusNyes it is, but there is no swapping taking place when the disk is reading data
14:29:29Zagorright. what I mean is the swap time could be easily included in the watermark calculation
14:30:01LinusNah, true
14:30:10LinusNbut the swap time is transparent
14:30:22LinusNit can play and swap simultaneously
14:30:43Zagorexcept for the first chunk
14:30:45LinusNit just need to be ahead of the playback
14:30:55LinusNwait for my fix...
14:30:58Zagorok :-)
14:31:03LinusNthing is:
14:31:25LinusNwhen the buffer wraps, it needs to do 2 read() calls
14:31:29Zagorwhat's the emacs command for setting a particular line's indentation?
14:31:35LinusNit will swap between the calls
14:31:44LinusNbut only 8Kbytes
14:31:48Zagorah, ok
14:32:11LinusNso i'll add a check for "playable space"
14:32:27LinusNif there isn't enough playable space, it will swap in larger chunks
14:32:38Zagorsounds good
14:33:24BagderZagor: control-x .
14:34:05Bagderor did you mean a c indent level?
14:34:35Bagdercontrol-c control-o on that particular line
14:36:15LinusNor TAB
14:36:38LinusNor am i missing something?
14:37:06Bagderif you wanna change the level
14:37:12Bagdertab uses the level
14:37:13ZagorLinusN: i wanted to change the indentation rule for a specific line ('case' in
14:39:04LinusNcan you guys repeat the "skips 2 songs" bug reported in the mailing list?
14:39:07LinusNi can't
14:40:06LinusNimproved swap committed
14:40:16Zagori haven't tried :)
14:44:05LinusNhehe, the watermark gets lower when i disable disk poweroff
14:44:31LinusNi wonder if we should call the buffer margin something else?
14:44:37LinusNin the settings
14:44:50LinusNlike "skip buffer" or the like
14:44:59LinusNwhat is it called in cd players?
14:46:26ZagorI think they usually call it "Skip protection"
14:46:59ZagorAnti-skip even :-)
14:47:26Zagornah, differs. "Skip protection" is good IMHO
14:48:43LinusNbut 0 seconds makes it sound like we have none at all...
14:49:04LinusNwell, maybe it's true
14:49:07ZagorCall it "Extra anti-skip" then :-)
14:49:27LinusN"Anti-skip buffer"
14:49:43LinusNa tough one to translate :-)
14:50:19LinusNis "Anti-skip buffer" ok?
14:51:00Zagoryes, I think so
14:51:17BagderAn tisk IP ?
14:51:57LinusNgo back to sleep, Bagder
14:52:11Bagderme not sleep
14:52:14Bagderme debug
14:52:25LinusNgo back to debugging, Bagder
14:52:42*Bagder goes back to debugging
14:52:57LinusN...and stay there! :-)
14:53:32*Bagder watches the thick doors close and hears the lock click behind him
14:54:44BagderSKIP = Sick Kids need Involved People, Inc. is nice ;-)
14:55:48Bagderso, anti-skip must be people who are not involved
14:56:09 Quit kargatron ("Suckers!")
14:58:28 Nick TotMacher is now known as tot|away (
15:01:16 Quit tot|away ()
15:01:32 Join TotMacher [0] (
15:11:15ZagorBagder: did you add the "min frame size" check to setid3v2title()? the TYER frame is 4 bytes (8 with header)
15:12:16BagderI don't see it matters who did it, but I don't think it was me
15:12:35Zagorit doesn't matter, I just wondered if you knew something I don't :-)
15:12:54Bagderbut I don't ;-)
15:13:04ZagorI noticed ;-P
15:14:35Zagorthe id3v2 header is supposed to be 10 bytes. I just found a file saying just "ID3" and then start stacking frames...
15:15:04Zagorit's nice having the spec. too bad people don't give a rats about it... :-)
15:15:23Bagderthat's the nice part with specs
15:15:33Bagderthey're fine to read but people don't care about them ;-)
15:15:59Bagderbeen there a few times with the http spec
15:19:41 Join ken0 [0] (marklar2@
15:19:41 Quit [keno] (Read error: 54 (Connection reset by peer))
15:30:42Bagdertime to run home
15:30:46 Quit Bagder ("Client Exiting")
16:03:31 Join Bagder [241] (
16:05:26LinusNthat's what i call running
16:05:52BagderI run fast
16:08:50***Saving seen data "./dancer.seen"
16:18:59 Join edx [0] (
16:21:29Zagorboom. 1KB text added :)
16:22:24BagderI can't understand people who want that
16:22:33Bagdercan't they hear the music?
16:22:52Bagderor are they thinking like this "oh, what's this genre of music called?"
16:23:06Zagoralso it's always wrong on stuff you download
16:23:37Zagorstill, a complete id3v1 implementation has it, so...
16:23:43BagderI agree
16:23:49BagderI just won't use it
16:24:02LinusNand it's also a matter of taste
16:24:17LinusNis it trance or is it dance?
16:24:20BagderI don't know anyone with taste ;-)
16:25:13Bagderright, I can never find a fitting genre when I've tried to fine one
16:25:22LinusNand what's the difference between Ballad and Power Ballad?
16:25:40LinusNand who dares to tag it Porn Groove?
16:25:50BagderBallad and Power Ballad are overly stupid
16:26:42LinusNand Humour vs Satire
16:26:53LinusNi can go on for ages
16:27:13LinusNthe whole genre thing is pretty silly imho
16:27:27ZagorI think we've established that, yes :-)
16:28:12LinusNZagor: try the newest dynamic buffer handling
16:48:10ZagorLinusN: it seems playback start is delayed longer now
16:49:26LinusNyes, the default watermark is higher now
16:49:54LinusNwait until the HD has stopped and see if that makes any difference
16:50:21Zagorit does. no sound until disk reading is completed.
16:51:09Zagorgotta go now. see you later!
16:51:14 Part Zagor
16:51:25Bagderuh, me too
16:52:07 Nick LinusN is now known as LinusN|away (
16:55:44 Quit laotan (Read error: 104 (Connection reset by peer))
17:01:00 Join laotan [0] (
17:46:48 Quit TotMacher ()
17:47:47 Join TotMacher [0] (
18:02:06 Nick LinusN|away is now known as LinusN (
18:08:54***Saving seen data "./dancer.seen"
18:20:13 Join kargatron [0] (
18:41:08kargatronwhoa, wps tags - cooool.
18:41:09 Nick seb-sleep is now known as _seb_ (
18:41:42kargatronhow was genre implemented - a read-only file of a number-name table?
18:48:18 Join Stone [0] (
18:48:37 Part Stone
18:48:56 Join [keno] [0] (~marklar2@
19:00:22 Quit ken0 (Read error: 110 (Connection timed out))
19:17:37 Quit TotMacher ()
19:17:58 Join TotMacher [0] (
19:22:53Mode"#rockbox +o Bagder " by ChanServ (ChanServ@services.)
19:23:08Topic"Zagor appeared on swedish TV -" by Bagder (
19:26:25 Join pyvasene [0] (
19:26:48 Join Triple-z [0] (
19:28:49Triple-zzagor on tv? nice, when?
19:28:56Bagderjust 10 minutes ago
19:29:16Bagderthey taped it yesterday
19:29:42Bagderit was regarding his "swap appartment" site
19:30:14LinusNand they gave the wrong URL! :-)
19:30:26Triple-zwhats the site?
19:30:28BagderI've mailed Arne about it to set one up
19:30:46BagderTriple-z: its in swedish:
19:47:34 Join Zagor [242] (
19:47:44Bagderthe tv-star arrives!
19:47:58*Bagder bows
19:48:05*Triple-z too
19:48:13Zagorhey. too bad my vcr didn't record anything, so I haven't seen it myself :-)
19:48:30BagderI recorded it
19:48:34Zagoryou did? cool!
19:48:41BagderI got home 2 minutes before it started ;-)
19:49:02Bagderas you saw, they showed a bad url
19:49:25Zagorthey did?
19:49:38LinusNsilly people
19:49:58LinusNwhy do they always assume that all addressesm begin with www?
19:50:01BagderZagor: see my mail I mailed you
19:50:26LinusNi recorded it too
19:50:43ZagorBagder: can we fix a CNAME for www.bjorn ?
19:50:43LinusNand they displayed the url in all capitals :-)
19:50:55BagderZagor: but the URL of your browser was also seen on TV, with the correct url
19:50:57LinusNwe have mailed the admin already
19:51:05BagderZagor: read my mail ;-)
19:51:13*Zagor goes reading his mail
19:54:43ZagorLinusN: can you divx-encode the video and put it somewhere for me?
19:58:01LinusNyou'll have to give me some time for that
19:58:43Zagorif you could just digitize it, I could do the conversion of course. but I guess the conversion is not what takes time :-)
20:02:13LinusNno, it's booting into w2k and firing up the matrox card
20:03:48Bagderwould be neat to get docs on how to run on the fm
20:04:21Bagdercorrect address there
20:04:46Zagoryup. i think the reporter had it right, but the graphics people screwed up
20:07:31Bagderthey showed it around 6 too
20:08:27Zagorwow, in the 5-minute news? maybe it will air in the 21.40-news too then.
20:08:55***Saving seen data "./dancer.seen"
20:09:25LinusNcan you guys try the latest BE and see if it skips, and if the start-of-playback speed has improved?
20:09:52Zagorwill do. the one I fetched just before I left skipped a lot :-)
20:11:31LinusNtry with and without poweroff
20:12:12LinusNand experiment with the anti-skip setting
20:20:13Zagori'm getting a lot of mail from people who saw me on TV :-) apparently it aired nation-wide, not just in Stockholm
20:22:00 Join johnny [0] (
20:29:27edxZagor: you've been on TV? for what reason? :)
20:29:38Zagormy apartment swapping site
20:30:05edxnice :) - a well known channel?
20:30:37Zagoryeah, SVT2, on of the two main state channel. on their regional news show.
20:30:59 Quit [keno] (Read error: 104 (Connection reset by peer))
20:33:45johnnyi just installed rockbox on my ipod. now there's smoke coming out of it
20:34:03Bagderthat's our self-destruct mechanism ;-)
20:34:31Bagder"this disk will self-destruct in 5 seconds"
20:37:49 Join [keno] [0] (~marklar2@
20:41:30 Quit pyvasene ("Leaving...")
20:47:20Zagorthere's constantly 6-8 searches running. nice having with dual Athlons :)
20:48:38Zagorabout 200 new ads the last hour
21:00:29LinusNtried the dynamic watermarking yet, Zagor?
21:00:49Zagortesting now
21:01:26Bagderok, time to test
21:01:49 Join Mine78 [0] (
21:01:57LinusNespecially the playback start time
21:02:54Zagorwhat's the new wm: values?
21:03:05Mine78Linus I am able to partially reproduce the bug I told you some days ago... about the blank screen exiting and re-entering a song
21:03:45Mine78It happens when you play a song than exit and go around in direcotries
21:04:03Mine78when the song change and you re-press on to see which one it is, sometime the screen is blank
21:04:34LinusNZagor: WaterMark
21:04:37Mine78It doesn't happen always. It often happens if I use the 5x7 font... dunno if it's related
21:04:41LinusNcalculated - lowest
21:05:01LinusNMine78: interesting
21:05:02Zagorok. got lowest A2FE now. that's a nice low margin.
21:05:18LinusNantiskip = 0?
21:06:12Bagderskip at first attempt
21:06:25Bagder128kbit cbr
21:06:29Zagorstartup is much quicker now
21:06:45Bagder3e80 - 0
21:07:05kargatronis that bug recently noted on the list still undiagnosed (audio dropouts during disk read)?
21:07:23Bagderbut its crap
21:07:28LinusNkargatron: it is being worked on
21:07:46LinusNkargatron: it's only been a few hours since we introduced it, be patient :-)
21:07:46Zagorkargatron: it's the new dynamic watermark code. being tuned right now.
21:07:51Bagderand another skip
21:08:02Bagder5dc0 - 0
21:08:03LinusNBagder: poweroff = yes?
21:08:16ZagorBagder: what is your spinup time?
21:08:27kargatronjust asking to judge if i should grab latest, is all. i've never deliberately rushed developement. :)
21:08:38Bagderpoweroff = yes
21:08:49LinusNkargatron: please use the latest, to test
21:09:06LinusNBagder: the spinup measurement gives very short times
21:09:09LinusNfor you
21:09:37LinusNthat is loooong
21:09:42Zagorvery very strange
21:09:42Bagderand 600ms now
21:09:50Zagorthat's the bug
21:09:55Zagorhow the heck can it be 600?
21:10:39LinusNBagder: do you use a playlist?
21:10:44ZagorLinusN: until that is fixed, I suggest you treat any spinup time < 2500 as 2500
21:10:52BagderLinusN: playlist, yes
21:10:56LinusNtry without
21:11:00LinusNdir play
21:11:11LinusNZagor: can you put that in the ata code?
21:11:27Zagorthat makes it much harder to debug
21:11:41LinusNZagor: i don't get it, but ok
21:12:11Bagderfound another 128kbit now, but this time it says bb80 -
21:12:23LinusNbb80 is way low
21:12:28ZagorI mean if I change it in ata.c, the bug will be invisible and we can't get third party reports about it
21:12:44Zagorsince Disk info will also report 2500 as lowest
21:12:44Bagderspinup now says 1910 ms
21:12:49LinusNthird party will not know if i fix it either
21:13:00LinusNZagor: aha, now i get it
21:13:04Zagorthey will when we ask then to check disk info
21:13:34Bagderskips just as bad
21:13:36ZagorBagder: what disk do you have?
21:13:38Bagderon every load
21:13:51Zagoryes, with those buggy spinup times it will skip
21:14:01Bagderfujitsu MHN2200AT
21:14:32Zagorouch, I have lowest 0 now. didn't notice the skip though
21:15:05 Join jh101 [0] (
21:15:10Bagderit not only skips
21:15:16Bagderit plays a part of the previous song
21:15:24Bagderwhen it had moved on
21:16:14Bagderwhat's the write and swap valueS?
21:16:32LinusNindexes in the ring buffer
21:16:37LinusNwrite = disk write
21:16:41LinusNswap = swap
21:16:49Zagorreally? ;)
21:17:04 Quit Mine78 ("Shokowahhhhh dwahhhnowan!!")
21:17:31Zagorsound start is nice and fast
21:17:54LinusNi have spinup=1630ms
21:18:22Zagorany ideas how that can happen?
21:18:26BagderI try a 192kbit now
21:18:36Zagorinternal disk prefetch or something?
21:18:43Bagderbut still a bad spinup at 620
21:18:54Zagorbut then it should be like 20ms, not those odd half-low values
21:19:22Bagderno skip!
21:19:28Zagorgot a skip now too, at 128 kbit. spinup 3550ms
21:19:39Zagorwm: 1B580 - 0
21:20:15Zagoradding spinup time to the mpeg debug screen might be an idea
21:20:24Bagderall in one
21:21:10Zagorskip again. 3650 ms
21:21:46Zagor1B580 should be plenty of time @128kbit
21:21:48Bagdermy 192kbit seems to do fine now
21:25:18Bagderany theories on the spinup time bug?
21:25:50Bagderno sound
21:25:54Bagderbut it runs
21:26:02Bagderthe playable moves
21:26:12Bagderthat's what I call a skip
21:26:25Zagorit's very strange. how can it spin up, and not calculate the time correctly?
21:26:45Bagderand it came back when the whole buffer was through one extra lap
21:26:54 Quit edx ("good night, guys")
21:27:07LinusNBagder: that may be a double swap
21:27:19Bagderwould it be silent then?
21:28:19Bagderif a skip occurs, shouldn't the wm value change?
21:28:23Zagor160 kbit seems to work too
21:28:33LinusNBagder: if it swaps, and then swaps back, yes
21:28:54LinusNBagder: no it shouldn't change
21:29:08 Join elinenbe [0] (
21:29:11elinenbehi there :)
21:29:14elinenbelong time no see!
21:29:19Bagderwhy not? shouldn't it compensate somehow?
21:29:22Bagderhey elinenbe
21:29:27elinenbeZagor appeared on TV???
21:29:28elinenbefor what?
21:29:39elinenbeyou have been making some awesome progress!
21:29:41elinenbeway to go!
21:29:53Zagorit was not about rockbox :-)
21:30:17LinusNBagder: the whole idea is that we should never skip
21:30:22elinenbeZagor: what was it for?
21:30:45Zagorelinenbe: i run an apartment swapping web site, and it made the news
21:30:46elinenbeLinusN: I don't know what you are talking about, but I would rather have 100 hour batterlife with random skips every 5 seconds.
21:30:48Triple-zthe new keyboard will be used for search also?
21:30:52elinenbeZagor: nice :)
21:31:00LinusNTriple-z: for all text input
21:31:15Triple-zcant wait
21:31:37Bagderfunny enough my 192kbit ones run fine
21:31:53ZagorBagder: are you getting any mail? i haven't gotten one the last 30 minutes
21:31:58LinusNBagder: it compensates for high bitrate
21:32:06Triple-z3 thing i really lookin for: 1. quque 2. search 3. bidi scroll
21:32:19LinusNZagor: labb os down due to byta-overload
21:32:26BagderZagor: I'm zero mails since 30 mins too
21:32:50ZagorLinusN: could be, if the mail passed through it :-)
21:33:22LinusNi thought the rockbox mailing list did that
21:33:37Zagoryes it does. but I get *no* mail.
21:34:33Zagornice. a 56kbit song gets a very low watermark
21:34:37Bagderno, there's a prob somewhere
21:35:01Zagorsendmail is responding, at least
21:35:33Zagoruh, no it doesn't. not on :-(
21:35:48elinenbewhat is bidi scroll?
21:36:27 Nick LinusN is now known as LinusN|fairytale (
21:37:09BagderZagor: right, its a problem
21:37:27BagderI could mail myself to from home
21:37:52Bagder"Domain of sender address does not resolve"
21:38:14Zagorboo. bad night for a breakdown :-(
21:38:47Zagor25 sendmail processes running
21:38:51Bagderbut its weird since the dns seems to work fine
21:39:00Zagorbut no cpu load
21:40:04Zagorelit dns is busted
21:43:03elinenbeyou guys ROCK the BOX!!!
21:44:40elinenbeI think that you should add .ogg playing.
21:44:46elinenbeeither that or .divx playing.
21:44:56elinenbeI would love to watch divx movies on my recorder
21:45:06elinenbehehe −− I love giving you a hard time :)
21:45:30Triple-zdoes ipod used special software to transfer files? or is it like the ABJR?
21:45:54elinenbeTriple-z: special software
21:46:09Triple-zu sure?
21:46:15elinenbeTriple-z: the ipod wishes it was as cool as the Archos Recorder (with rockbox)
21:46:20elinenbeTriple-z: yup, I have one!
21:46:31Triple-zyou can put only mp3 on it?
21:46:33elinenbeTriple-z: I NEVER use it... I really only use my Archos Recorder.
21:47:00elinenbeTriple-z: no, you can mount is as a hard drive, but when it is a hard drive you can not see the music files on it.
21:47:04Triple-zyou got the windows version of ipod?
21:48:00Triple-zdude you should mail you your ipod if you dont need it..
21:48:25elinenbemail it where? China?
21:48:40elinenbeI have the mac version, but I converted it to the windows version.
21:48:57Triple-zwhere you live?
21:49:18Triple-zwhere do you live that is
21:49:36 Quit jh101 ()
21:50:46Triple-zanyway hows the ipod frimware? with the jog and all that, is it good?
21:51:42Zagori'd also like to know if it really does no caching during playback
21:53:07elinenbeIT is okay.
21:53:27elinenbethe new version is way worse (non moving jog dial!)
21:53:34elinenbeI hate the way it works/feels
21:54:57elinenbeit is like a touchpad on a laptop!
21:55:17elinenbeZagor: I am unsure about that...
21:55:54Triple-zhows the screen? design? size?
21:56:39 Join BoD[] [0] (
21:56:53BoD[]hi !
21:57:30elinenbehi there.
21:57:37elinenbeare you an archos jukebox user?
21:58:17johnnyi'm using an ipod ;-)
21:58:34johnnyjust in case you someone wanted to know *g*
21:59:08Triple-zyeah i wanna know
21:59:26Triple-zits smaller than archos?
21:59:31y0mhi there
21:59:46johnnyi think so
22:00:07johnnyit's like a card game
22:00:19y0mit looks smaller
22:00:21Bagderhow's the playback time with it on fully charged batteries?
22:00:34Triple-zwell it dosent got open source code frimware right?
22:00:39johnny10 hours
22:00:42BoD[]does it have tetris ? ;))
22:00:53Triple-zor sokoban?
22:00:54johnnyBoD[]: only break-out ;)
22:01:08Bagderwe need breakout
22:01:11BoD[].. really ??
22:01:18BoD[]I thought it had no game
22:01:23Triple-zyou like it generally?
22:01:32johnnyi'm in love with it
22:01:46Triple-zu got also archos?
22:02:03BoD[]how many ms between 2 tracks when clicking "next"
22:02:11Triple-zso what are you doin here dude?
22:02:28johnnyTriple-z: i'm finding your project quite intresting
22:02:30Zagor10+ concurrent byta-processes now. the cstrike server is starting to see increased pings...
22:03:28Triple-zwell it not mine buy it is quite intresting
22:03:31johnnyTriple-z: and i found the rockbox-website already, when i was still unsure, which player to buy
22:03:38johnnyok ;)
22:04:01Triple-zipod does look cool
22:04:22Triple-zbut i got my archos before the windows ver of ipod
22:04:27Bagder32 MB ram makes the programmers spoiled ;-)
22:04:52Triple-zyeah that this is da bomb, 32mb jeez
22:05:02johnnybut there aren't a lot of ipod-programmers, i'm afraid
22:05:18Bagderstill, it has gaps in the playback they say is any good?
22:05:33johnnyTriple-z: no, too high level ;)
22:05:46Triple-ztoo high?
22:05:57johnnyTriple-z: this is more intresting:
22:06:30BoD[]you can modify the firmware in an ipod ??
22:06:39johnnynot yet, i think
22:07:04Triple-zanyone ever saw rioriot?
22:07:19 Nick LinusN|fairytale is now known as LinusN (
22:07:20johnnyonly pictures of it ;)
22:07:27Triple-zor used even better
22:07:29BoD[]what is it
22:07:29elinenbeI think the rioriot is a piece of shiz...
22:07:32BagderLinusN: any good tails tonight? ;-)
22:07:36Triple-zpics i also saw
22:07:45elinenbedid anyone ever hear back from the company that is making an "open source" player?
22:08:13Triple-zi wanna see some good pic of ipod near archos to see the size diffrence
22:08:21Triple-zipod on your plam
22:08:24LinusNMamma Mu bygger en koja
22:08:28Triple-zsomething like it
22:08:38 Quit elinenbe ()
22:08:56***Saving seen data "./dancer.seen"
22:09:15Triple-zanyone got that kinda pics?
22:09:33johnnythere are lots of pics with hands
22:10:11Triple-zgot more good sites?
22:10:23johnnyyes, but not ipod-related ;-)
22:10:48johnnythat was a joke (sort of)
22:10:57Triple-zguessed that
22:11:02johnny;-) <=- blonds doin it with ipods on them :D
22:12:07BoD[]i actually clicked on that
22:12:25Triple-zits working?
22:12:40BoD[]how sad am i :)
22:14:05Zagoroh god. 30 processes now... this is worse than slashdot :-)
22:14:36Bagder~600 new ones since this afternoon ;-)
22:14:47BoD[]by the way.. any idea about the HITACHI_DK23CA-20 problems I experience?
22:15:03johnnysooo. i'm going to a pub now ...
22:15:04BagderBoD[]: we have heaps of new problems right now
22:15:09ZagorBoD[]: can't say I do
22:15:21BoD[]ok :)
22:15:50BoD[]i'll stay with the 2002 11 30 version
22:15:55 Join ]ack[ [0] (~a@
22:16:01BoD[]and try new ones as they come
22:16:33Triple-zjo: you got the 10gb windows version?
22:16:56Bagderah, right those ipods have tiny disks
22:17:21Triple-zthey got 20ver i think
22:17:43Bagderfor an arm and a leg if so
22:17:45Triple-zhow fast USB 2.0 actually works?
22:17:48johnnyTriple-z: exactly. 10 gig, win
22:17:53BoD[]i'd buy one if they'd come in 40G .... and cheaper;))
22:18:06johnnyand it's a 1.8-inch-disk
22:18:36BagderTriple-z: the 2.0 is MUCH faster than the slowmo 1.1
22:18:38 Join ken0 [0] (marklar2@
22:18:38johnnyusb 2.0 is a little bit slower than firewire i think
22:18:51ZagorUSB2 is actually a little bit faster than firewire
22:19:07Zagorbut the disks are always the limiting factor, so it doesn't differ
22:19:09johnnybut i was right with the "little bit"-part
22:19:22Triple-zbagder: but how fast its working with archos?
22:19:45BagderTriple-z: I don't remember exactly, but I think I copied over my first 15GB in just a little over an hour
22:20:05Bagderit was a while ago
22:21:15]ack[sorry to bother, does anyone here have a 0.7a drive working in their archos? I know the old hitachi drives were 0.5a...
22:21:37Bagdera as in ampere?
22:22:01]ack[all the drives I see in stores are 0.7a
22:23:06BagderZagor: did ya watch the late abc?
22:24:14]ack[is there a difference I should be worried about?
22:24:35*Bagder still runs with the original 20GB
22:24:42Zagorno, naturally I missed that... but they have it published on I'm installing realplayer now...
22:25:06Bagderaha, neat. Put a link on the site
22:25:21Zagorah, good idea
22:25:23]ack[hmmm....anyone else here have a non-factory drive installed in their JB?
22:25:34Bagderbtw, should we point to the archos user forums from the rockbox site?
22:28:10Bagderabc is the news show Zagor appeared in
22:28:42BoD[]oh :) what what is about
22:28:48BoD[]i mean
22:29:01BoD[]what was it about
22:29:03BoD[]? mails coming through
22:29:12BagderBoD[]: about his apartment site
22:29:21LinusNBoD[]: he has made a site for swapping appartments
22:29:30BoD[]oh really :)
22:29:48BoD[]what's the url
22:29:50ZagorBagder: i noticed :-)
22:31:13Bagderwith exponential back-off times, you never know ;-)
22:32:54ZagorBoD[]: (it's all in swedish)
22:35:21BoD[]perl :) youhouu
22:35:48BoD[]ah ahha :) I like the simcity logo :))
22:36:20 Quit [keno] (Read error: 110 (Connection timed out))
22:42:45BoD[]going to watch tv
22:42:51 Nick BoD[] is now known as BoD[tv] (
22:45:34LinusNok, the current watermark calculations are a bit optimistic
22:48:02 Join ken0_ [0] (marklar2@
22:50:06LinusNZagor: finally, you can put "as seen on TV" on anything you've made :-)
22:50:25BagderYES! ;-)
22:50:48Bagderbut I didn't see any rockbox
22:51:47Zagorwe actually touched on that during the interview, but of course it didn't make it to the final cut
22:53:17 Quit ken0 (Read error: 104 (Connection reset by peer))
22:57:01LinusNnew mpeg.c committed
22:57:56Bagderah, and a safety check for really bad spinup times ;-)
22:58:58LinusNnow it wastes space, but is safer
22:59:24Triple-zanyone got avatar of abjr?
22:59:28LinusNi'd like you giys to try it
22:59:40BagderI'm about to
22:59:51Triple-zlike this one for ipod
23:00:20Bagderwhat's avatar?
23:00:54Triple-zmall pic
23:01:07Triple-zcheck that one of ipod
23:02:52BagderLinusN: does playlist or dir matter?
23:03:41Bagderkicking off with a 192
23:04:00Bagder2bf20 seems big enough
23:04:17Zagorgosh! I just looked at their 5-minute news spot from 18.55. A whole minute was spent about my site!
23:05:06Bagder128kbit now
23:05:18LinusN2bf20 may be close to the margin for a 192kbit/s file
23:05:39Bagderdid fine
23:06:29Bagder1d4c0 on 128kbit
23:08:37Bagderskips multiple times too
23:09:01LinusNok, that means that the 2.5s safety margin isn't enough
23:09:49LinusNok, as long as the spinup measurement fails like this, this isn't gonna work
23:09:50Bagder192kbit skipped too
23:10:19LinusNthat is way low
23:10:58LinusNwhat is the shortest real life spinup time we've seen?
23:12:07BagderI dunno
23:12:31LinusNi see a potential measurement problem
23:13:49LinusNoh, false alarm
23:20:27LinusNZagor: have you tried?
23:20:59Zagorno, i've been a bit busy :)
23:21:13Zagoranyone know a way to rip realplayer streams?
23:21:27LinusNno i don't
23:22:23LinusNit seems like we have real problems measuring Bagder's spinup times
23:24:35Zagoryeah. i can't figure out how that happens :-(
23:24:50LinusNBagder: what spinup times seem realistic to you?
23:25:39LinusNZagor: is there a chance that the sleeping flag can be set without the drive being asleep?
23:26:20LinusNi guess not
23:26:37Zagori can't see how, anyway
23:27:00LinusNit works 100% for me
23:27:24Zagoryeah, me too
23:27:46LinusNi guess the toshiba is nice to us
23:31:23BagderI find it really hard to estimate the spinup times
23:31:52Zagoryes it is
23:32:03Triple-zanyone got a good case for the jukebox?
23:35:43Bagdertime to hit it
23:35:49Bagdernight guys
23:35:51 Quit Bagder ("")
23:44:26Zagorsame here
23:44:27 Quit Zagor ("Client Exiting")
23:49:16 Quit TotMacher ()

Previous day | Next day