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 2004-04-16

00:00:06Quelsarukhe's the *father*
00:00:21lImbussomebody asked me why ONLY a archos has a cool opensource-fw
00:00:37Quelsaruknot only archos
00:00:40pvhThe Neo has one now too.
00:00:42lImbusso ?
00:00:57Quelsarukrockbox also supports Neo
00:01:03lImbusok, but it's till the same manufacturer, isn't it ?
00:01:11Quelsarukafaik, no
00:01:13pvhNope, different manufacturer.
00:01:40lImbusoh. cool. so rockbox starts to be A opensource firmware, not a archos os fw ?
00:02:12solarisamiconn: I did all that and still it doesn't load the > 15 days ago
00:02:12solaris[5:42pm] <Quelsaruk> more or less
00:02:12solaris[5:42pm] <Quelsaruk> all the time i've been away from tech :D
00:02:12DBUGEnqueued KICK solaris
00:02:12solaris[5:42pm] <Quelsaruk> hi Bagder
00:02:12solaris[5:43pm] <Bagder> hi
00:02:13***Alert Mode level 1
00:02:13solaris[5:43pm] <amiconn> Meanwhile I did post a grayscale framework, within a demo plugin (well, the demo itself is not very amazing).
00:02:15Quelsarukwell.. rockbox is an opensource firmware, but started supporting just archos player and recorder
00:02:16solaris[5:43pm] danhans224 ( has joined. «23 people»
00:02:18solaris[5:43pm] <Quelsaruk> cool
00:02:20solaris[5:44pm] <Quelsaruk> :)
00:02:28solarisoops, sorry
00:02:33solariswrong button
00:03:17Quelsaruksolaris, i was wondering what the hell was that copy&paste for ;)
00:03:56solarishehe, sorry....
00:04:01solarisi don't know what happend
00:04:18solarisi just pressed ctl-v and ....
00:04:30Quelsarukwindoze :P
00:04:30solarisi don't think i pressed ctl-c to copy anything...
00:05:04Quelsarukyou're using mirc?
00:05:15solarisnow that i have your full attention, ;) i need some advise to revive my jbr20
00:05:25Quelsarukjust marking text with the mouse will copy it
00:05:35Quelsaruksolaris, what happened?
00:05:38solarisi had firmware flashed w/2.1
00:05:48solarisand was trying to upgrade to 2.2
00:05:57solarisused rockbox.url
00:06:18solarisand after that, jbr20 doesn't respond with ON button
00:06:37solarisi can turn it on with F1+ON in archos 1.40a version
00:06:43solarisbut that's it
00:06:57solarisno matter what i try, i can not get rockbox to run
00:07:37Quelsarukbut, you can use archos firmware
00:07:39amiconnsolaris: Did you try the defrag way also?
00:07:51solarisyes, but i had some errors
00:07:52Quelsarukcan you plug it to the pc?
00:08:11solarisyes, with F1 pressed and plugging in USB
00:08:23Quelsarukwhy don't you copy an uptodate rockbox version (even 2.2) to the root folder
00:08:33solarislol, it doesn't work
00:08:38solaristried it so many times
00:08:49Quelsarukeven booting with F1??
00:09:13solarisbooting with F1 will start archos 1.40a
00:09:24solarisi can play mp3 and everything
00:09:34Quelsaruki mean, if the problem is in the flashed firmware, booting with archos and leting it *rolo* an error-free 2.2 must work
00:09:37solarisbut only in archos
00:09:49 Join facted [0] (
00:09:53Quelsaruki suppose you have tried that
00:10:18amiconnQuelsaruk: it doesn't, for whatever reason. We've been through it.
00:10:22Quelsarukif so, i have a stupid question, doesn't work, or maybe is a contrast error?
00:10:24solarisi had some problems with rolo
00:10:35Quelsaruki mean, loading a new version will change the settings
00:10:52Quelsaruk(it normally do)
00:11:00Quelsarukso maybe the contrast value is to low
00:11:05Quelsarukand you can't see the display
00:11:19Quelsarukjust an idea
00:12:05 Quit Galik (Remote closed the connection)
00:12:14***Alert Mode OFF
00:12:36Quelsarukit's possible that upgrading to a new rockbox version resets your settings
00:12:44Quelsarukthat includes contrast
00:12:48solarisdon't think so
00:12:48 Join top_bloke [0] (
00:12:54Quelsarukit's an option
00:12:59Quelsarukcheck this link
00:13:01solaris'cause i shuts off after 30sec
00:13:24solarisgood thought
00:13:40solarisbut how come i cant start it by using F1+on??
00:14:00CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:14:00*danhans224 has left the building
00:14:05 Quit danhans224 ()
00:15:19Quelsarukstart with normal boot
00:15:23Quelsaruki mean, rockbox boot
00:15:37Quelsarukand follow this key combo :)
00:15:46solaristhis will be run
00:15:49solarishold on
00:16:23 Part LinusN
00:16:44solarisit shuts off
00:16:56solarislet me see how long that is..
00:17:20solaris15 second
00:17:22Quelsarukwhile pressing this combo?
00:17:27 Part facted
00:17:32solarisand also without
00:17:33 Join facted [0] (
00:17:33Quelsarukor before?
00:17:43Quelsaruki mean, did you start pressing the keys?
00:17:47solarishow long should i wait b4 the code/
00:19:05Quelsaruki mean, when rockbox is loaded, you can press those keys
00:19:18solariswell, i don't think it's loading
00:19:34pvhQuelsaruk: Your nickname makes me crave diablo ii.
00:19:35solarisit'll shut off after that code if i'm fast
00:19:48pvhsolaris: the 'code' just takes you to the contrast menu
00:19:51solarisand if i'm slow, it shuts off in the middle
00:19:52Quelsarukpvh, i used it in diablo II ;)
00:20:02pvhsolaris: which isn't much help for you if the 'box isn't actually loaded
00:20:08pvhQuelsaruk: Sweet.
00:20:25Quelsarukthat code was just to enter contrast
00:20:38Quelsarukjust in case it was my first and quick thought
00:20:54solarisyes, i know but i don't think rockbox is loading
00:21:08solarisi can hear the disk spinning but that's about it
00:21:18Quelsarukand having a rockbox.ajz file in root folder, booting with ON+F1 doesn't work?
00:21:40solarisit doesn't work with that either
00:21:55solarisso, i'm thinking something is blocking the loading
00:22:12solarisshould i format the whole thing?
00:22:17solarisis that my next step/
00:22:25Quelsaruktry with the bleeding edge file
00:22:45Quelsaruki mean, try that but using the latest rockbox file, just in case
00:23:02solarisok, i'll do that
00:23:11Quelsarukand, i suppose the faster way to repair the box is sending a mail to the mail list
00:24:16solarisok, i'll send the mail
00:24:23lImbusdoes the plugin-framework (which btw is great) imply I#ve to prog all in one file or can I split it to multiple files ?
00:25:09amiconnlImbus: You have to use one file atm., but a multi-file solution is discussed.
00:25:11Quelsarukbtw, i'm gonna update my rockbox version and try that plugin :D
00:25:53solariswhat does rockbox_flash.rock do?/
00:26:06pvhFlashes the boot loader
00:26:19solarisso, can i try that with new script?
00:26:30solarisbut i can't see that file in archos
00:26:43amiconnpvh: This would be firmware_flash.rock; rockbox_flash.rock only flashes the second image (.ucl)
00:26:45Quelsaruksolaris, how did you try to update rockbox??
00:27:16solarisi used rock.ucl
00:28:00Quelsarukyou downloaded rockbox 2.2, copied to the box (including plugins and so on) and after loading that version... tried to flash it?
00:28:42solarisi tried to upgrade it by using rockbox.ucl
00:28:58solarisafter copying 2.2 files, over writing them
00:29:21solarisisn't that what manual said?
00:29:27Quelsaruki think so
00:29:49solarishow do you upgrade yours?
00:29:55Quelsarukroloing the 2.2 or just overwriting the files and playing the rockbox.ucl?
00:30:08solaristhat's the runny thing...
00:30:23 Join Pearl_Gate [0] (~Nickname@
00:30:54solarisafter copying arj... into the root, when i play the file, I had and error saying rolo error: and next line chesum error:
00:30:56Quelsarukmaybe, the only person that can answer your question, and help you is jorg
00:31:48Pearl_Gatechesum error=bad
00:32:04solarisok, i'll stop crying and wait for him and send him an -email
00:32:21Pearl_GateHas anyone had trouble with the battery covers on the JBR20?
00:32:50solarisis there anyway to see a file other than mp3 in archos?
00:33:30Quelsaruki have to go
00:33:37Quelsarukand good luck
00:33:39solaristhx for helping me
00:33:41Quelsarukjust a last advice
00:33:49Pearl_Gatei'll just sit in the TV and watch the couch
00:33:49Quelsaruktry sacrifying a goat
00:34:00Quelsaruki sell some cheap goats ;)
00:34:09Quelsarukonly 1000€ the unit :P
00:34:14Quelsarukand they work
00:34:21solarismarked free and clean goat?
00:34:39Quelsaruki'm an autoriced technopriest
00:34:47Quelsarukyeah, goat=cabra
00:34:59Quelsarukhave to go
00:35:04 Quit Quelsaruk ("Leaving")
00:35:06Pearl_Gatecu pra ti tb
00:35:56 Join Galik [0] (~galik@
00:37:08Pearl_Gateze+cabra= Zé Cabra
00:37:58zei'm a monkey though
00:38:59Pearl_Gatei like monkeys
00:39:02Pearl_Gatethey cute
00:39:57Pearl_GateSo, what's your opinion on Windows Longhorn?
00:40:44zethat's like uh the next ver that wont be out till 2005 or 2006 or somesuch isn't it?
00:41:13pvhMy opinion on Windows Longhorn is the same as my opinion on the afterlife.
00:41:18pvhI'll believe it when I see it.
00:41:27lImbuslonghorn looks sleek. nice with all those black surfaces+
00:41:36zepersonally, i think if microsoft is capable of making decent software/OS, they have no motivation to do so
00:41:49pvhze: Counterpoint −− Windows XP
00:41:51zebut that's a big if
00:41:55 Quit Galik (Remote closed the connection)
00:42:18pvhSay what you like about MS, but you have got to admit that Windows XP is a _very_ well designed OS.
00:42:23pvhMiles ahead of anything else they've done.
00:42:30zei do?
00:42:38zemy parents are both windows users
00:42:46zethey went from 98 to XP
00:43:00zeand then got fed up with it's crappyness and went to 2000
00:43:06*pvh shrugs.
00:43:08zewhich makes no sense to me actually
00:43:10zebut heh
00:43:30pvhThere are some transitional problems associated wtih upgrading from 98.
00:43:39pvhBut a clean, tuned install of XP is a thing of technological beauty.
00:43:48pvhParticularly once you hook it up with Litestep, Cygwin and the works.
00:43:51zeit was a clean install, not an upgrade
00:44:03pvhGet rid of WMP, Explorer, IE, Office...
00:44:06pvhNow you're really talking.
00:44:15zeheh things like litestep, cygwin and suck are beyond my parents
00:44:34zeand i think if you're gonna do all that then why bother with windows at all
00:44:50pvhStability, interoperability, gamability...
00:45:06pvhI use linux every day, I just don't use it on my desktop.
00:45:12zejust trying to be like unix, you might as well have a system that can actually handle unix stuff
00:45:22*pvh shrugs.
00:45:31pvhGood software design eliminates that barrier.
00:45:47zeyeah i guess i just don't game much any more
00:46:09zebut when i did i played mostly idsoftware games anyway
00:46:40zeand i'll tell ya what, q3 on linux totally outperformed q3 on windows on my system
00:47:09pvhAre they doing D3 for linux?
00:47:35zeyes, as far as i know they are
00:48:06 Join Galik [0] (~galik@
00:48:17pvhI never did get Wine to work very well under linux.
00:48:24pvhMaybe if I'd had a windows 98 cd it would have.
00:48:27zethey've said at some point that they were, i think they're doing the same parallel-development thing that they did with q3
00:48:27MTtheres not too much effort in porting games to nix
00:48:46MTwhen i say not much
00:48:49pvhMT: That all depends on your development environment.
00:48:58pvhI can tell you don't write games for a living.
00:49:26pvhIf the game is coded in SDL and wxWidgets and OpenGL, you're spot on though.
00:50:14MTokay, which bits of the new generation of games take the most time, and which bits are reasonabley straightforward
00:50:34MTworld engine, physics engine, textures/graphics
00:51:00MTswitching from win32 to nix wont make one arse of difference to that
00:51:28pvhMT: Right, which is why game companies find it so economical to develop cross-platform applications.
00:51:38pvhAnd there are so many Linux releases.
00:51:46MTno, they dont because theres very little demand for linux games
00:52:00MTlets take an example
00:52:02MTid games
00:52:03pvhIt's not about demand, it's about return on investment.
00:52:11MToh right
00:52:22MTyes, with zero demand theres a great chance of a decent roi
00:52:31zealso most companies don't bother to decently design the platform-specific elements or how they tie in with the rest
00:52:33pvhJohn Carmack has the luxury of knowing he's going to make a shit-pile of money on anything he puts out, so if he farts around in Linux, it's his perogative.
00:52:43 Quit facted ("Trillian (")
00:52:49pvhMost development teams don't have that kind of capital behind them.
00:53:23pvhIndie development also has that luxury.
00:53:45MThow does that contradict what i said
00:53:46pvhNeverWinterNights took _months_ of extra development to get their Linux versions out, IIRC.
00:54:00MT_months_ isnt that long
00:54:09MTim not surprised it took months
01:01:18Pearl_Gatei'm gay
01:01:23Pearl_Gateas in happy
01:06:00 Quit Pearl_Gate ()
01:13:41 Join mecraw__ [0] (~mecraw@
01:24:06 Join facted [0] (
01:24:35 Quit top_bloke ("The mind is a terrible thing to taste. Wasted 1 hour 12 minutes and 43 seconds online.")
01:30:21 Join top_bloke [0] (
01:43:00 Join BC [0] (
01:45:16 Join sleep21 [0] (
01:50:18***Saving seen data "./dancer.seen"
01:52:35 Quit Douche_Nutz (Read error: 60 (Operation timed out))
02:00:25 Quit silencer (Remote closed the connection)
02:01:31 Quit mecraw__ ("Trillian (")
02:07:04 Quit Galik ("Client exiting")
02:09:22 Join diddystar5 [0] (
02:10:19 Join sunpower [0] (
02:10:27 Quit solaris (Read error: 54 (Connection reset by peer))
02:11:23pvhHere's a thought: if the quick-menus are only going to allow one button at a time
02:11:31pvhDoesn't that mean the up button could be added?
02:11:45pvhI mean, the 'flipped' screen already uses up with the F-keys.
02:11:51pvhWhy not the un-flipped?
02:13:13amiconnBecause the quick screens are meant to be used in 2 ways: (1) press F key, then left/right/down, then the F key again to quit. This also works when flipped.
02:13:46lImbusup would work also ?!?
02:13:59lImbusmhmm. maybe just because I can't figure out what to set on that fourth button ? ok, recording *duck*
02:14:39amiconn(2) press and HOLD the F key (like a shift key), then left/right/down additionally. If you release the F key again, you leave the quick screen immediately. This is why they are called quick screens.
02:14:50amiconnThe latter doesn't work when flipped!
02:15:06pvhamiconn: Yes, but aren't two-button combos being outlawed?
02:15:19BCwe want user defined menus :)
02:15:52amiconnYes, they are (sort of) in favor of configurable / context sensitive menus.
02:16:00*pvh nods.
02:16:29BCi wish I could understand the problem with them
02:16:42BCapparently no-one can solve the recursion issue
02:16:49pvhWhich recursion issue is this?
02:16:49BCI'd love to, but I can't find it - LOL
02:17:22 Quit sunpower ("—I-n-v-i-s-i-o-n— 2.0 Build 3515")
02:18:12BCthe problem is entering a menu from a menu and getting into a loop
02:18:42BCbut I just don't understand why the code would allow that - or why it would be so difficult to defend against
02:18:59pvhAh, entering a customized menu from a customized menu?
02:19:06lImbusso it's rather a software recursion that a design one ?
02:19:20amiconnWhy should that get you into a loop? If you press "right" once, you can only go one step deeper, I think.
02:19:32BCi can see no example where it would happen though
02:19:41lImbusI just figured out that for the quick screen f3, UP and DOWN both work. They do the same. has that to do with that recursion
02:20:54pvhNo, I don't believe so.
02:21:33BCthe cutom menu system is not associated with the quick menu system currently there
02:21:46amiconnBC: of course, if a configurable menu system is built, it has to use some stack-like arrangement for back-tracing.
02:22:11BCwhy does it have to be so complicated
02:22:30BCyou either return as normal or return to the custom menu
02:22:38BCthe end
02:23:05amiconnThis is actually very straight forward, I've seen that in many apps that allow for customizable menus.
02:24:02amiconnEvery internal operation (even calling a menu) is defined as a command. The menu definition itself is put in an ascii file.
02:24:20BCascii file - even better still :)
02:24:45 Quit sleep21 (Read error: 113 (No route to host))
02:24:47amiconnThis automatically allows for recursion if there is a stack, so that wouldn't give loops.
02:24:58BCWHAT recursion?
02:25:08BCthis is the bit that baffles me
02:25:38amiconnHaving a command within a menu definition that calls another menu, where there is a command that calls another etc.
02:25:52BCsurely the custom menu just starts the menu system at a given point
02:25:55pvhYes, but who is going to rewrite the code?
02:26:14BCthere are many capable - are you offering?
02:26:15pvhami: I think I understand the problem now. Each menu is hardcoded as to where it returns to.
02:26:25*pvh is working on his own ideas.
02:26:27BCyes, so just add ONE line of code
02:26:38pvhbc: Not that simple.
02:26:46pvhbc: Many, many, many lines of code would need rewriting.
02:26:50BCif (this was called by custom) return elsewhere
02:27:01BCfor the full implementation yes
02:27:09amiconnBC: This would mean that the custom menu is "flat", i.e. no sub menus. I thought of rewriting the _whole_ menu system to be configurable.
02:27:11BCbut for the engine - a few lines at most
02:27:22pvhbc: Functions don't 'return elsewhere' they simply 'return'.
02:27:47pvhbc: The rockbox menus return the selected item, I believe.
02:27:48BCpvh: lol
02:28:05pvhbc: There's a big difference.
02:28:19BCami: think purpose -people just want quick access to their favourite functions
02:28:56BCfer sure - a complete rewrite would be cool - but is it required for the purpose for which it is desired?
02:29:34BCyou're also modding core code - which the eschelons like to be kept to a bare minimum
02:29:55amiconnI think that could actually save a bunch of code. But if the custom menu is going to be flat, why could there be a loop condition?
02:29:57pvhBC: Have you done much coding?
02:31:06BCme? a coder
02:31:11BCyeah, one or two bits
02:31:40pvhWell, there you go. Better get a copy of the source and start patching it yourself!
02:31:41BCami: that's my point - I just don't get where the problem is! :)
02:31:51BCpvh: good idea
02:32:14amiconnIs there some conversation log about that problem?
02:32:20BCpvh: that's the nice thing about having the rockbox crew refuse to credit you for your work ;)
02:32:33BCami: woowww, way wayyy back when, yes
02:32:59BCpvh: ever played "othelo"?
02:33:07BCpvh: ever checked the patch log?
02:33:10pvhnot on the 'box.
02:33:15pvhare you credited as BC? ;)
02:33:31DBUGEnqueued KICK BC
02:33:31BCrockbox crew refuse to credit me
02:34:05pvhThere does seem to be a certain amount of 'us and them' attitude in the lists and such.
02:34:22BCyeah - it's a bit like that with open source
02:34:38BCopen source means "work for free for others" now days
02:34:52pvhHow much did you pay for Rockbox?
02:34:54BCstill, it's my choice, and I don't regret it
02:35:07BCdid you pay?
02:35:14BCor was that a sarcastic question?
02:35:17pvhNope, that was kinda my point.
02:35:27BCyep - work for free
02:35:30BCthe exact price I stated
02:35:42CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
02:35:42*pvh shrugs and goes back to the code.
02:35:44*BC look pleased with his clairboyance skills today
02:35:57BCLOL - clairBoyance
02:36:19BC"I'm floating off into the future" :)
02:36:28pvhOhhh, you must be BlueChip!
02:36:38 Nick BC is now known as Bluechip (
02:36:38DBUGEnqueued KICK Bluechip
02:36:42Bluechipsorry, yes
02:42:14amiconnGuess I will have to get at least some sleep.
02:42:48Bluechipneed to chat sometime with you about greyscale stuff
02:43:09 Part amiconn
02:44:17 Join WoodShop [0] (~GingerFat@
02:45:12pvhMmm, cheese and olives.
02:45:26pvhDo you have Vinta crackers where you live?
02:45:36Bluechipnot that i have ever seen
02:45:42pvhBad luck for you.
02:45:59Bluechipbut we do have "hovis digestives" which are damn nice as a sweet alternative
02:46:19pvhThose are good too.
02:46:30pvhWith olives though?
02:46:32pvhInteresting idea.
02:47:10pvhThe code for the play function in tree.c is suprisingly... klugey.
02:47:27top_bloketheres a food called digestives?
02:47:42Bluechipyes, it is also a generic name
02:47:42pvhI think the British invented it
02:47:51top_blokewow sounds yummy
02:48:03pvhSome people call them "baby cookies"
02:48:28pvhBut their general market appeal made them move from the 'under one' market into a more general audience.
02:49:03BluechipAre you thinking about "rusks"
02:49:10pvhRusks are good too.
02:49:18CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
02:49:18*pvh has a dutch family. :)
02:49:31Bluechipthey should NOT be in the "mother and baby" aisle
02:49:55pvhBut no, digestive cookies are these little sort of oblate flat things which are faintly sweet.
02:51:52pvhDoes anyone know if the file-types are going to be moved out of hard-code?
02:52:15Bluechipthere was talk of it
03:02:47scott666god i want to see kill bill
03:02:53scott666*volume 2
03:03:41lImbusomg, my brother doesn't stop quoting hattori hanzo
03:04:17scott666in english or japanese?
03:04:22lImbusno, things like buka buka and aligator
03:04:35lImbusok, he saw the german dubbed version
03:04:50scott666do they dub the japanese too?
03:06:52pvhYeah, I gotta get on that too.
03:07:06*pvh doesn't go to ShareReactor and doesn't see if he can't download it.
03:07:19Bluechipits dead
03:07:28BluechipSimon Moon is behind bars
03:08:31lImbuserr, no, they do'n dub that. iirc it was subtitled
03:10:29scott666ok, thats what i thought...thats how the american version is too
03:12:08 Quit WoodShop (""Will You Kill All Of Man Kind In Order To Save Your Sister, Or Will You Kill Your Innocent Sister To Save All Of Man Kind? W)
03:14:54 Quit top_bloke (Read error: 54 (Connection reset by peer))
03:24:14lImbuserr, has somebody else heard of an plain vanilla INT (not floating point) implementation of ogg vorbis lately ?
03:28:01scott666not me
03:28:26lImbuscombined with the souvenir somebody told me the MAS is reprogrammeable in a certain manner and with limits...
03:28:47Bluechipit is
03:28:57scott666whats the limit? 4k?
03:29:36 Join AciD [0] (
03:29:54lImbus""Tremor" integer-only codec now under BSD license"
03:29:55Bluechip4096 20bit instructions + same in register memory
03:30:13Bluechipif i were you - i would write a wav driver first
03:30:38lImbusoh, I'm not me either sometimes.
03:32:05lImbusis that tremor stuff still far away from what is possible with the archos ? (pssst, I got no clue, tell it nobody)
03:48:11 Quit pvh ()
03:48:11 Quit Nibbler (Read error: 54 (Connection reset by peer))
03:50:19***Saving seen data "./dancer.seen"
03:53:11scott666yay, overhand is back to just crashing the sim
04:07:09MTlImbus: not a hope in hell
04:08:37lImbusok, so probably not either if someday somebody writes a bool-implementation :=>
04:09:11Bluechiprockbox has bool
04:09:12lImbuserr, BOOL
04:09:16lImbusah, brainfuck
04:09:38lImbusI meant a bool-implementation of that ogg-thingy
04:11:24lImbusam I right that the most important cause for flashing an archos is to get rockbox running instead of that annoying charging screen if I plug in the external power ?
04:11:38Bluechipquick boot imho
04:14:57lImbusok, thats comfort. my archos is not flasheable and I am planning about a small box that does following: patched between archos and power-supply with a link to the remot control plug: it first switches on the archos when dc-power arrives without giving the power to the archos, then, a few seconds later passes by the dc-power to the archos.l
04:15:07lImbusjust for my car use
04:15:37lImbusmay it interest other people ?
04:15:48Bluechipcould well do
04:16:28Bluechipiirc when my archos is off, oneof the remote keys makes the hdd spin
04:16:57scott666does car mode require a flashed box?
04:16:58Bluechipbefore you ask me to check, my remote needs a new battery :(
04:17:13Bluechipflashed is just normal without a disk file
04:17:52lImbuscar mode without flashed would result in a stuck in charging screen archos instead of playing
04:20:27 Join TheDude2 [0] (
04:20:45lImbuswith my circuit, when I start up the engine, first the archos would be booted, would load the rockbox-os from disk, then later start to charge. no need for a flashed archos
04:22:48 Quit Bluechip (Read error: 104 (Connection reset by peer))
04:25:21 Join BC [0] (
04:28:10lImbuswell, I think i've to oversleep that once or twice. better soon.
04:37:22 Join scott666_ [0] (
04:37:30 Quit jkerman (Read error: 110 (Connection timed out))
04:38:09 Quit scott666 ("i'll be back...eventually...")
04:47:41 Quit diddystar5 ("Leaving")
04:49:43 Nick scott666_ is now known as scott666 (
04:49:52scott666stupid ghost...
04:50:16BCghosty noise
04:52:59BCwhat range does ((PileCardCnt*6)+(((rb->rand()%5)-2)*PileCardCnt)/10) generate?
04:53:46scott66640% to 80% of PileCarcCnt
04:54:08BCcheers :)
04:54:50scott666the new alg is much shorter, but for some reason it explodes the sim
04:55:07BCaim it to me
04:55:07 Quit lImbus (Connection timed out)
04:56:13BCwhat is 25*/X
04:56:18BCsorry /*
04:56:29BCother than a gpf
04:56:50BCoh shit!
04:56:51scott66625 is a constant number to work with
04:56:54BCit's a C comment - LMAO
04:56:54scott666so is 13
04:57:20BCvery different from "25 / *5"
04:57:31BC(note space)
04:57:38scott666thatd be in interesting operation
04:57:47 Join Unknown [0] (
04:57:56scott666'25 divided times 5'
04:58:53BCthe * would be a pointer
04:59:09scott666stupid c...and its pointing
04:59:40BCor should I ask "what are you referring to"?
04:59:48BC& other stuff
05:02:52BCfor (y = 13/*((rb->rand()%5)+1)*(cut1/10)*/; y>0; y−−)
05:03:07BCyou realise you will get a different rnd each time around the loop?
05:03:31scott666that rand is the number of cards dropped
05:03:38BCoh hang on, I'm blind tonight
05:03:49BCyeah - sorry - me being stupid again
05:04:12scott666the rands are pretty much the only things im sure of
05:04:42scott666well...the type declarations are pretty good too i guess
05:05:05BCI adore your abuse of the for construct
05:05:15BChave to teach you "?"
05:05:45scott666i saw that used once
05:05:47BCif (y) then x=a else x=b
05:06:06scott666in this
05:06:11scott666i figured it was something like that
05:06:18scott666rockbox wps has that
05:06:19BCbool=bool?false:true is a favourite
05:06:52scott666*sometime similar to that
05:07:28BCignore brace mismatch
05:07:45BCor not, as there isn't one
05:07:50BCi thnk i need to go to bed
05:08:08BCdo you not think the latter is better?
05:08:46scott666yeah, it is
05:08:56scott666less possible division/rounding errors
05:09:11BCgeez - i got something right - lol
05:09:30scott666especially if cut < 10
05:09:58BCouch - yes
05:12:40scott666The instruction at "0x64b82424" referenced memory at "0x64b8c008". The memory could not be "written".
05:13:09scott666i like the quotes around "written"
05:13:36BClike it's a code word for something
05:13:45scott666heh, yeah
05:14:34scott666"oh yeah! i 'wrote' HER 'memory' last night!"
05:17:16 Quit c0utta{zz} (Read error: 110 (Connection timed out))
05:17:16 Nick Unknown is now known as c0utta{zz} (
05:19:20scott666IRC needs an equivlent to ghost busters
05:27:09zeit does
05:27:22ze/msg nickserv help ghost
05:29:10 Quit AciD ("'%20radi)
05:32:12 Join Nibbler [0] (
05:35:51 Join sleep21 [0] (
05:37:15sleep21i'm looking through the old mails to find conversations on cracking the usb encryption
05:38:41scott666any luck?
05:39:29sleep21i'm not seeing anything
05:39:38sleep21did they like do that first?
05:44:04scott666i dont think usb is encrypyed on the archos
05:44:23scott666its just a mass storage device
05:44:32scott666windows just thinks its a hard drive
05:44:40BCbog standard usb-ide device
05:45:14scott666does explorer know your nomad is a nomad? can explorer see it at all?
05:45:36sleep21the nomad has a proprietary FS
05:45:48sleep21that's one thing i'd like to fix ;-)
05:45:54scott666isnt there an open nomad project already?
05:46:03sleep21not to my knowledge
05:47:10scott666theres a linux driver for the nomad II
05:47:44sleep21yeah, that stuff
05:47:49sleep21that's different though
05:48:04sleep21plus im working with a Zen Xrea
05:50:20***Saving seen data "./dancer.seen"
05:52:13sleep21that stuff is OSS, but there isn't a firmware project afaik
05:53:21scott666you were asking about cracking the usb protocol though
06:05:17 Quit scott666 ("i'll be back...eventually...")
06:14:26 Join IRCMonkey___ [0] (
06:14:36IRCMonkey___hi is there anyone in here?
06:16:11IRCMonkey___it looks like my archos died today, and i havent even had it six months, i was curious if having the rockbox on it void my warranty?
06:18:10IRCMonkey___anyone know?
06:18:31BChmmm, probably
06:18:38BCdepends if they check
06:18:38IRCMonkey___damn it that effin sux
06:18:45BCflashed or just an update file?
06:18:54IRCMonkey___id unno
06:19:01BChow did you update to rockbox?
06:19:13IRCMonkey___copied a folder onto the drive
06:19:25BCthen you can pop the drive out and remove the folder :)
06:19:41IRCMonkey___pop the drive out?
06:20:10IRCMonkey___i can't access it anymore, it starts to load the OS and then stops at about halfway
06:20:21BCopen the box
06:20:23BCremove the hdd
06:20:27BCplug it into you pc
06:20:44BCyou will need a little-drive to big-connector adaptor
06:20:54IRCMonkey___i'll try i guess... thanks
06:21:34IRCMonkey___probably not going to do it though, i think the drive is effed... now i say that the archos is a P.O.S
06:21:54BCif u r in the states newmp3technology may help
06:22:27IRCMonkey___whats that?
06:22:46IRCMonkey___it just clicks and beeps while trying to load
06:22:55IRCMonkey___i think the drive is broken
06:23:05BCprolly just a dodgy connector
06:23:13IRCMonkey___well thanks
06:23:14BCits not unknown
06:35:46 Join matsl [0] (
06:39:54 Quit sleep21 ()
07:09:54 Quit matsl (Remote closed the connection)
07:15:14 Quit IRCMonkey___ ("ChatZilla 0.8.31 [Mozilla rv:1.4/20030624]")
07:32:15 Quit Nibbler (Read error: 54 (Connection reset by peer))
07:38:53 Join monkey [0] (
07:50:24***Saving seen data "./dancer.seen"
08:02:40 Quit c0utta{zz} (Read error: 110 (Connection timed out))
08:12:37 Join LinusN [200] (
08:52:34 Join amiconn [0] (
08:53:15 Part amiconn
08:54:02 Join mattzz [0] (
08:54:19mattzzg'morning everyone
08:54:36BC:) likewise :)
08:54:51 Join amiconn [0] (
09:15:20 Join Nibbler [0] (
09:18:42BCamiconn: free to chat?
09:30:17 Part amiconn
09:30:53BClol - that's a safe "no" ;)
09:33:18BCLinus, on the subject of "framework" code I have a question, if you are not too busy
09:37:33BCdue to a lack of malloc (no i don't fancy that job - thanks ;) )....
09:37:45BCthe programmer needs to specify a few limits for the arrays
09:38:25BCif the framework were in say, ../frame/cards would they change these #defines on a plugin-to-plugin basis
09:38:38BCI can thin of solutions, but nothing clean
09:39:32LinusNwhat arrays need different limits for each plugin?
09:39:55BCmaxcards, maxpiles
09:40:35BCif structs are packed....
09:40:45BCa card is 11 bytes
09:41:03BCand exists twice
09:41:14BCand a pile is 6
09:41:35BCsorry, scrub that - the second occurence is 7 not 11
09:42:03BCso maxcards*(11+7) + maxpiles*6
09:42:07LinusNwhy must the cards/piles be defined in the framework?
09:42:25LinusNwhy not let the calling code define the piles?
09:42:38LinusNand send handles/pointers to the framework
09:42:55LinusNshuffle(struct pile *pile)
09:44:59BCI suppose I rely to heavily on globals - "old habits die hard"
09:45:59BCI fear you are right - the more i think about it - the more it makes sense
09:46:14LinusNif you work with pointers, the game can have ny number of piles, memory permitting
09:46:42LinusNand decks
09:46:59*BC nods in agreement
09:47:13BCthank you :)
09:47:19LinusNyou can use several decks, right?
09:47:24BCoh yess
09:47:30BChence MaxCards
09:47:31LinusNand shuffle them all together
09:47:54BCall cards start life on pile[STOCK]
09:48:50LinusNare there any game which needs two separate decks?
09:49:12BCthat would be easy enough, just create a new pile and move the cards there
09:49:37LinusNjust curious, seems like your framework handles it
09:50:27***Saving seen data "./dancer.seen"
09:50:32BCfair question,and yes, that is certainly easy enough to setup
09:50:40 Join amiconn [0] (
09:50:52BCI will upload the new rolling demo in a few mins
09:50:53 Nick amiconn is now known as amiconn|away (
09:50:56BCgrafix and all now
09:50:58LinusNcan i have the url to the src again?
09:52:02BCI will tell you once the latest is uploaded
09:53:43 Quit MT (Read error: 110 (Connection timed out))
09:56:38BCaha - sim bug - lol
09:57:07BCrb->lcd_clear_display(); does not clear the display unless you do an update immediately after
09:58:04mattzzand I thought that was a feature....
09:58:29BCclear display; draw stuff; update = new stuff overlayed ovre old stuff
09:59:33BCbut clear; update; display; update = new stuff only !
10:02:04LinusNweird, i can't see how the clear;draw;update could render that result...
10:03:14 Join MT [0] (
10:06:28LinusNthe core code uses that method all the time
10:06:40LinusNit must be a simulator plugin issue
10:09:38LinusNwould be interesting to try with the x11 sim and see if it behaves in the same way
10:13:14LinusNrmdir() is coming up
10:13:23LinusNwas a little harder than expected
10:13:32BCi must say i live in fear of it - lol
10:13:58BC"If you are sure press ON+F2+DOWN"
10:14:46BCi notice that someone has started work on the custom menus
10:15:04LinusNhe's been working on it for quite some time
10:15:35BCa guy had problems with the f3=playlist patch
10:15:50BCnoticed new f3 handling code
10:16:00LinusNup does the same as down?
10:16:36LinusNah, i know what you mean now
10:16:59BCoh, yes, and I just twigged what you mean - lol
10:17:30*LinusN tests the rmdir on his fm...
10:17:46BCLinusN is brave and likely has a bakcup
10:18:06BCdeltree /y c:\.
10:18:48BCdo we have mkdir and mv yet?
10:19:12LinusNi use my fm for dangerous experiments, never keeps anything valuable on it
10:19:20LinusNmkdir works
10:19:44LinusNi missed rmdir when i tested mkdir :-)
10:19:47BClast time i wanted that it was not there
10:20:01LinusNit has been there for ages
10:20:16BChaven't done any recording in months
10:20:50LinusNit is in 2.2
10:21:01BCwas likely 2.0 when I last tried it
10:21:06BCmaybe even before that
10:21:31BCi wanted to make a new dir and move my recording there
10:21:42LinusNthe recording now has an option to store the file in the current dir or in /recordings
10:22:02LinusNand it creates /recordings automatically if it doesn't exist
10:22:15BCwell done that man
10:23:37*LinusN runs fsck...
10:24:48BCwooo hoooooo
10:27:11LinusNnow to the boring work of adding it to the simulators...........
10:27:49BCThere are dun things to be done there too ;)
10:28:24LinusNthe win32 disk simulation for instance...
10:28:37BCwhat a fine example
10:28:49LinusNdid your patch solve that?
10:28:50 Quit monkey ()
10:29:38BCyes, but because I missed the #defines in dir.h for the x11 sim - the x11 sim overrides break it again
10:30:18BCI just added a ROCKBOX_ROOT or something and appended it in the browse routines iirc
10:30:41BCmy patch would probably work again if you removed all the x11 over-rides
10:32:10LinusNis there a way of enlarging the screen on the recorder win32 sim?
10:32:24BCyes, drag the edge of the window
10:32:28BC(not the corner)
10:32:33LinusNit's so tiny on my 1600x1200 screen
10:32:49BCmake sure you run at 200/400/800/etc %
10:33:29BCthere should be a compile switch so that only the little green box is displayed
10:33:47BCit's very pretty and all, but....
10:35:51LinusNi agree
10:36:00LinusNi like the x11 sim better in that respect
10:36:31BCsadly never seen it
10:37:25 Join bongo [0] (
10:38:10bongoHi guys I've got a problem with my Arhos Jukebox can anyone help?
10:38:22BCdepends on the problem
10:38:31LinusNthat's gonna cost... a LOT!!! :-)
10:38:49BCLOL - now THERE's a though
10:39:04LinusNbongo: what's the prob?
10:39:19BCyou've scared him off now
10:39:26bongowhen I start up I get HD Register eroor followed by a list of 'SC1' ETC
10:40:09LinusNbongo: player or recorder?
10:40:21bongorecorder 20Gb
10:40:28LinusNnot fm or v2?
10:40:44bongofirst line reads SC1 (85) 128
10:40:52LinusNnot fm or v2?
10:41:00bongono not fm or v2
10:41:18bongoJukebox Recorder 20
10:41:19LinusNand you can't hera the drive spin up, right?
10:41:35bongohang on I'll just listen for that
10:41:38LinusNor a "click click" sound?
10:42:54bongoYes I hear the click sound...also sometimes it just won't start up at now as soon as I hear the click sound it then turns its self off or crashers
10:43:12bongoIt was working fine two days ago?
10:43:58BCi fried mine and that is sounding very familiar
10:44:11bongoFresh batteries
10:44:21BCcharged externally?
10:44:23bongoHow did ya frie it
10:45:04BCi think ...trickle charge and lots of disk IO to the PC, let them drain too often
10:45:21BCi get the feeling I should have read the battery FAQ
10:45:56bongothis is going to sound really dumb
10:46:01BCnow they will charge, play for about an hour or two and die at about 75% (by the meter)
10:46:08bongoBut I just changed the batteries again
10:46:15bongoAnd it's just fired uo OK
10:46:15BCand wont hold a charge for more than a couple of days
10:46:23bongoI tried this sevral times?
10:46:31BCdodgy connector inside?
10:46:47BCtry squeezing and twisting the unit a little
10:47:43bongoDon't know .. it's like it's back to how it was now
10:48:01bongoI do charge the batteries externally though
10:48:01LinusNbongo: squeeze the bumpers together in the direction of the batteries
10:48:56LinusNit's very likely a broken solder joint
10:48:56bongosqueezing and slightly shaking and it's playing fine
10:49:11LinusNhandy with a soldeing iron?
10:49:29bongoYes I can manage to solidier ok
10:49:40LinusNthe green lines show where to solder
10:49:51bongook thanks
10:50:00LinusNit's a very fragile design, extremely silly
10:50:18BCwhats the red line
10:50:25LinusNand it breaks easily, especially if you replace the batteries frequently
10:50:39bongoYes I do replace the batteries alot
10:50:46LinusNthe red line shows where he removed an extra wire that holds the spring back
10:50:57LinusNa bad move imho
10:51:03bongoI'll see how it goes before I open it up first
10:51:09BCremoving the wire is bad?
10:51:36BCor archos made a bad move when they put it there?
10:51:39LinusNthe spring is too strong
10:51:54LinusNthat's why the solder joint breaks
10:52:21LinusNthat isn't the design error, though
10:52:53LinusNthe design mistake is to let a simple solder joint take the force from the battery springs in the first place
10:53:40LinusNthe wire does two things: it holds back the spring, and it adds an extra ground connection
10:54:22BCmakes sense, perhaps a small cap would be a better choice
10:55:24bongoI guess it was'nt originally design for exchanging the batteries frequently
10:56:19BCstill a little careless of the enfineer though
10:57:09amiconn|awayLinusN: This morning I noticed a bug in rockbox: There are entries which belong to the localization shown in the credits scroller (way down, at pos. 43..49)
10:57:52amiconn|awayThese are the entries to localize the voice menu, and that very menu doesn't get localized. This occurs at least with english.lng and deutsch.lng
10:58:11 Nick amiconn|away is now known as amiconn|work (
10:58:41BCSo all i need to do is create a menu entry called "Bluechip" and my name will appear in the credits :)
11:01:28LinusNamiconn|work: i don't get it
11:04:51LinusNamiconn|work: i see it now
11:19:08amiconn|workLinusN: For a different matter, I would like to do what the comment in lcd.c says: "Ultimately, all calls to lcd_write(false, xxx) should be substituted by this"
11:19:26 Quit bongo ("Leaving")
11:19:53amiconn|workUnfortunately, all calls of this type that are still there are from lcd-player.c, but as I don't own a player, I couldn't test it.
11:20:00amiconn|workAny suggestions?
11:23:27LinusNlet me check that
11:23:40LinusNit doesn't really matter for the player
11:23:58LinusNsince it only transfers a handful of bytes for each transfer
11:26:20 Join Galik [0] (~galik@
11:29:30BChere's something I've never had to consider before ....what does return(i++) return, given i=5
11:34:54BCglad I didn't have to write the compiler for that line of code!
11:37:17amiconn|workLinusN: Yes, but this way we could get rid of the second unrolled lcd transfer loop to save space.
11:38:00LinusNah, you mean to not ever use lcd_write_data() for the player
11:38:07LinusNand get rid of it
11:39:12amiconn|workNo, I thoguh of doing it the other way round: Having the unrooled loop in lcd_write_data() only, and change lcd_write() to lcd_write_command()
11:39:59amiconn|workI even think of a way to use the unrolled loop for transferring both commands and data.
11:45:49LinusNnot if it gets any slower
11:48:40amiconn|workYes, this is might introduce more overhead, so maybe I won't do that.
11:49:40amiconn|workHowever, my first idea (changing lcd_write() into lcd_write_command() ) should not impact performance.
11:50:31***Saving seen data "./dancer.seen"
11:50:42amiconn|workIt is very straightforward to replace lcd_write(false, var) with lcd_write_data(&var, 1) _if_ var is of type char.
11:51:47amiconn|workUnfortunately, this is not the case with all variables in lcd-player.c that are used that way. And if I change that, I couldn't test myself, since this cannot be tested with the sim.
12:00:56LinusNi think we should try to remove one of the three functions in lcd.c
12:02:25amiconn|work? I see only 2 functions: lcd_write() and lcd_write_data().
12:27:23LinusNi mean, we could just remove the player version of lcd_write_data()
12:37:58amiconn|workThis way, the unrolled loop for "data" bytes in lcd_write() would stay in for the recorder.
12:40:33LinusNyup, and we save a few bytes in iram on the player
12:44:11amiconn|work...but not on the recorder...
12:45:26amiconn|workIf we throw out that loop from lcd_write() instead and limit it to only perform lcd_write_command(), this would save space both for player and recorder.
12:47:13LinusNamiconn|work: whichever is best
12:47:39LinusNwe want to save iram space without affecting recorder preformance
12:49:43amiconn|workYep, that is why I suggested it to do it that way. But this requires to replace all calls of lcd_write(false, ...) (only used for the player) with lcd_write_data(..., 1). But if I do this change, I cannot test it, since it has to be tested in a player.
12:49:51amiconn|workThis was my initial question.
12:50:54amiconn|workAs it is now, the unrolled data-writing loop in lcd_write() is there for both player and recorder, but completely unused for the latter.
12:52:19LinusNi see
12:55:33amiconn|workAnd the unrolled loop within lcd_write_data() is vice-versa not used on the player (if I did not overlook something).
13:00:12amiconn|workSo my idea would unify data writing on player & recorder (both would use write_lcd_data(), with unrolled loop). Command writing would then be called lcd_write_command(), also for both player & recorder.
13:01:25LinusNi'm all for it
13:01:35LinusNyou do the code, i do the testing
13:03:52amiconn|workOk then, will be my next task.
13:04:18LinusNi'll work on your recording code later
13:04:19amiconn|workAnother one: did you find some time to give my recording patch a try?
13:04:28amiconn|workOoops, this crossed.
13:04:30LinusNgotta have lunch first
13:08:14 Quit Galik (Remote closed the connection)
13:20:08 Join c0utta{zz} [0] (
13:21:07 Join Galik [0] (~galik@
13:24:35 Quit Galik (Client Quit)
13:24:41 Join Galik [0] (~galik@
13:25:30 Quit Galik (Client Quit)
13:44:47 Nick c0utta{zz} is now known as c0utta (
13:44:49 Nick c0utta is now known as c0utta{zz} (
13:45:44BChit it firmly near the control panel
13:45:54 Nick c0utta{zz} is now known as c0utta (
13:46:12c0uttahydrairc sometimes gets confused...
13:46:41BCi hear you are working on the custom menu code?
13:47:03c0uttai await linus's comments...
13:47:06c0uttabut yes
13:47:19BCdid you solve "the recursion problem"?
13:48:07c0uttai don't think it's a problem the way i've implemented the functionality
13:48:31BCcos i never understood how it could be a problem
13:50:33***Saving seen data "./dancer.seen"
13:50:39LinusNhow come you don't understand it?
13:53:03 Join [1]c0utta [0] (
13:53:07BCI just cant see where it would matter - maybe just what I have in mind
13:53:16[1]c0utta#$^#$% wireless
13:53:24[1]c0uttanot having a good night here
13:54:59LinusNexample: you create a custom menu for the F2 key for the radio screen
13:55:28LinusNthis menu contains "FM radio", which calls the radio_screen() function
13:55:57LinusNthat way you can call radio_screen() over and iver until you run out of stack
13:57:12[1]c0uttawell, i say "silly you" :)
13:57:45LinusNthat is the "recursion problem"
13:58:02BCbut if you f2->radio ...then f2 will take you back UP to the f2 menu
13:58:13BCnot DOWN into another one
13:58:47BCit would solve the problem
13:58:56LinusNcase BUTTON_F2:
13:59:09 Quit c0utta (Read error: 60 (Operation timed out))
13:59:10 Nick [1]c0utta is now known as c0utta (
13:59:11BCif (f2_active) ....
13:59:33LinusNi didn't say it was unsolvable, just that you have to take care
13:59:47BCi thought it was the big stumbling block
14:00:01LinusNit is one of them
14:00:29LinusNyou need to prevent recursion somehow
14:00:40BCalthough it is not a problem now (pending your approval) I was just interested to see what I had missed/presumed-obvious
14:00:56LinusNit is still a problem in c0utta's code
14:01:09LinusNnothing prevents recirsion
14:01:12BCoh! poo
14:01:35LinusNsomehow, we need to invent a way of "rewinding" the stack
14:01:42LinusNnothing in his code
14:02:09BCif (f2_active) return else f2_menu()
14:02:22LinusNsomething like that
14:03:34LinusNbut you want, for example, to be able to enter the FM radio screen from the dir browser custom f2 menu
14:03:50LinusNand the fm radio f2 custom menu must still work
14:03:53 Join nevs [0] (
14:04:02LinusNwe just don't want it to recurse
14:04:35nevscould i ask a question about rockbox
14:04:51nevsis it possible to have a function to play back songs backwards?
14:04:52LinusNabout rockbox? in this channel? naaaaaaah!
14:05:08BCLinus: errr I thought f2 was a re-organised menu, and f3 was <thing>specific functions?
14:05:12LinusNnevs: the answer is no
14:05:26BCcould you not play all the packets in reverse order?
14:05:39LinusNyou will be able to have different f2 menus in different contexts
14:06:00nevsim using the player version
14:06:02LinusNno, the samples in each frame will still be played forward
14:06:21BCbut how long is a frame is reality?
14:06:31nevslets say i got an mp3 file, can they make it so if i go to like sound options, mode, and then like a "backwards" mode
14:06:37BCif it's short enough, you would get a reasonable approximation no?
14:06:49LinusNroughly 1100 samples
14:06:59nevswhy isnt is possible?
14:07:01LinusNand the frames are not independent
14:07:02BCnevs: backwards is NOT supported atm
14:07:19nevsbut ppl have discussed it?
14:07:28BC1100 - that's quite bit - oh well
14:07:36BCquite A bit
14:07:38LinusNthey need to be played forwards, because of the bit reservoir
14:07:54LinusNa 44100 khz frame is 26.12ms
14:08:03BCi looked for specs on mp3 data the other day but found NOTHING!
14:08:26LinusNthey are not public, you need to cough up some dough to get hold of it
14:08:32nevswill it ever possible,? or does it need programming or is it that the hardware can not handle it inside the mp3 player?
14:08:35LinusNas far as i know
14:08:41BCoh! that'll explain that then
14:08:45LinusNnevs: here's the deal
14:08:58LinusNthe decoder is in hardware
14:09:16LinusNall we do is send it mp3 frames
14:09:17BCDid you see my post on relatime mixing (in mono)
14:10:07LinusNwe can't reverse the order of the frames, nor make it play the mp3 data backwards
14:10:30LinusNBC: yes i saw it, it *can* work, if you encode the data properly
14:10:47BChmmmmm - interesting
14:10:48LinusNyou need a true stereo separation (eg not "joint stereo")
14:10:55nevsso its impossible yeah
14:11:01LinusNnevs: for us, yes
14:11:17nevsoh well
14:11:26LinusNbut i know too little about the mp3 encoding to say for sure
14:11:34LinusNnevs: sorry about that
14:11:38nevsif i post it on feature request, do u reckon someone would know
14:11:58LinusNnevs: it's not a matter of "knowing how to do it"
14:12:07c0uttahere's my angle - rockbox has been written in a very modular fashion, but still not modular enough to support "events"
14:12:23c0uttathat's why (i believe) recursion is an issue
14:13:09LinusNc0utta: you mean design it as a huge FSM?
14:13:37c0uttathat's an option!
14:13:55c0uttanot one that i'd like to do though
14:14:34LinusNthen what is an "event" in your preferred design?
14:15:19c0uttaan event is classed as events performed by a user
14:15:29BCany recursive routine can be written iteratively - it's just a matter of approach
14:15:33LinusNyes, but how are they supposed to be handles
14:16:25 Quit nevs ()
14:16:36LinusNc0utta: each "screen" is a separate loop
14:16:45c0uttathe problem is that functions call functions that call functions - common in a procedural language like C
14:17:03c0uttawe can carefully program recursion out, but it's a pain
14:17:04BCif (active) return
14:17:15c0uttaBC: right
14:17:34 Join AciD [0] (
14:17:39LinusNBC: you're on to something
14:17:45BConly appears after if (button==f2) easy to grep for
14:18:13LinusNeach "screen" has an "active" flag
14:18:26LinusNlike "bool radio_screen_active"
14:18:29BCno, just one flag
14:18:40BCbool f2_menu_system_active
14:19:28LinusNthen you wouldn't be able to have more than one custom menu
14:19:39c0uttaand we have more than i custom menu
14:20:01BCf2 is your generic custom menu
14:20:08LinusNi want to have a custom dir browser f2 menu, and call radio_screen() which has another custom f2 menu
14:20:13BCf2 is <active_thing> specific
14:20:21LinusNf2 is context sensitive
14:20:28DBUGEnqueued KICK BC
14:20:28BCf3 is <active_thing> specific
14:20:51BCi though f3 was context sensitive and f2 was my new personal menu layout
14:21:12c0uttathey're both context sensitive BC
14:21:29c0uttawe'll hardcode the F3 options though
14:22:30c0uttalinus, have you looked at me code ?
14:22:39LinusNnot yet
14:22:52BCi think you have found the source of your problem
14:23:16LinusNbc, not at all
14:23:23c0uttai'm waiting :)
14:23:36LinusNor you want the f2 menu only in the dir browser?
14:26:15c0uttai've spent hours looking at rockbox code and to make it work properly we have to moduralise code even further so only one defined action is done at any one time
14:26:25c0uttais modularise a word ??
14:26:33BCit is today :)
14:26:46BCalthough I THINK it was yesterday as well
14:27:25LinusNc0utta: so define an "action"
14:27:27c0uttabookmarks are a classic example
14:27:44c0uttaadd bookmark to favourites
14:28:16LinusNyou don't add bookmarks to favourites, you add mp3 files
14:28:40c0uttaforget i said that
14:28:45c0uttaadd mp3 to favourites
14:29:10c0uttashould we be able to do this from wps ? i think so
14:29:26c0uttauser selects "add mp3 to bookmark"
14:29:42c0utta"action" code decides whether we're in wps or tree mode
14:30:09c0uttaif in wps mode write out the current played file
14:30:19c0uttaif in tree write out the current selected file
14:30:50c0uttathat's an event
14:30:56LinusNin my world, they would be two different actions
14:31:51c0uttabut ultimately you run the same code to save a chr* to a file ?
14:32:10c0uttajust the method to get the char changes
14:33:16c0uttathis might sound stupid..but
14:33:38c0uttai would like to see every state (wps/tree/recording/fm) represented by an object
14:34:15c0uttayou could then perform events on those objects - i think this would avoid recursion
14:34:46amiconn|workc0utta: Couldn't you just limit the stack for nested calls to a fixed depth?
14:35:13c0uttaamiconn: yes, that's another way and much easier too
14:35:31c0uttawe could do that now (in fact, i'll do it my action code)
14:36:15CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
14:36:15*LinusN goes to eat something
14:36:17amiconn|workThis way, at least rockbox wouldn't crash if someone builds a recursive custom menu. At the maximum depth it will simply stop entering submenus.
14:36:21*LinusN is away: late lunch
14:36:23BCand what happens when it is fill
14:36:37c0uttai suppose nothing
14:36:44BCami: from the system declared it will be nigh impossible to avoid
14:37:05BC"why does my f2 button stop working sometimes?" hmmmm
14:37:48amiconn|workIt could give you a splash screen message "maximum menu nesting reached!"
14:38:03BCi can't see that getting past them in charge
14:38:59amiconn|workSomeone who doesn't understand that would probably not make its own custom menu. At it could be explained in the manual.
14:39:05amiconn|work*And it..
14:40:01c0uttado you guys understand rockbox threading ? i don't by the way
14:40:34amiconn|workI've only a rough understanding of it.
14:40:57c0uttahow/where are the threads cycled ?
14:41:04BCat a guess - a bunch of interrupt drive stack swaps - but I've never really examined it too closely
14:41:25LinusNthey are cooperative
14:41:49c0uttawhere's that code linus ?
14:41:51LinusNwhich is called indirectly by sleep() etc
14:41:53amiconn|worksee firmware/thread.c
14:41:57LinusNthread.c and kernel.c
14:44:09c0uttahow many threads are there at the moment ?
14:47:21 Join methangas [0] (
14:48:37amiconn|workc0utta: 7 if I counted correctly
14:48:37 Join lImbus [0] (
14:49:14c0uttathanks amiconn
14:51:34lImbusBC: I do unterstand now the problem you wanted to tell me about my trick switching trhough the power to the archos later
14:55:15lImbusI just got my remote, and then I noticed it hasn't an ON-button. THEN, after another cup of tea, I know why.
14:56:32lImbusThis only means I have to switch my archos on with a wire connected to the button of my archos, like the RTC-Alarm hack.
14:56:49BCdid you find that one of the buttons spins up the hdd if you hold it for a second or two?
14:57:43lImbusahh, not yet, forgot that one. hold on
14:58:27 Quit mattzz ("CGI:IRC (Session timeout)")
14:58:53lImbusapparently not on mine
14:59:12BCI will have to get a new battery and try it again sometime
14:59:22BCyou have Archos f/w?
14:59:30BCin ROM
14:59:44lImbuserr. how to compare if we got the same remote ? mine hasn't a battery as far as I can see ?
14:59:52BCit has
15:00:00BCyou take the tiny screws out of the back
15:00:06lImbusyes, my archos isn't flasheable
15:00:06BCand there's a button cell or two inside
15:00:38lImbusok. mine works if the rockbox is running
15:03:48lImbusHORRAY: I found out something: If I give DC-Power for about one second, then unplug it again, the archos-os is loading rockbox. This is what I want: I want rockbox on a dc-event.
15:03:55 Quit adi|home (Client Quit)
15:04:34BCyou devious b^std
15:05:04lImbusI have to mount a small thing that, IF power comes along, passes it by for a second, then cutting it again. wait until rockbox is up, then pass through power to be charged.
15:05:18BCput it all though smoothing caps
15:05:42*lImbus is happy
15:06:26BCeasy enough to build
15:07:37lImbusdidn't anybody find that out yet ? *surprised*
15:09:28 Nick BC is now known as BC|coding (
15:09:28DBUGEnqueued KICK BC|coding
15:14:44amiconn|worklImbus: This was mentioned several times by [IDC]Dragon, on the ml and in the channel.
15:15:36lImbusand has ever somebody build that ? would be easier to order a part list @reichelt for me
15:25:17 Nick lImbus is now known as neuer (
15:25:17 Nick neuer is now known as lImbus (
15:26:20BC|codingIf you "open the box and upgrade the firmware chip physically" you can have speech menus on your Neo
15:28:54 Part LinusN
15:34:07BC|codingcan i do #if TYPE==STRING ?
15:34:21BC|coding#define TYPE STRING
15:50:26c0uttaBC: according to K&R - yes
15:50:36***Saving seen data "./dancer.seen"
15:50:43BC|codingthanks dude :)
15:50:58BC|codingcan't believe it's not in my book
15:52:01c0uttapage 77 in mine
15:52:39c0uttaassuming "my book" means K&R
15:52:55BC|codingi don't have the K&R "bible" :(
15:53:43c0uttawant a pdf ?
15:53:53BC|codingOhhh ohhh ohhhh :)
15:53:55BC|codingyes please
15:54:06BC|codingText Searchable :)
15:54:27c0uttayes, it's text searchable
15:54:46c0uttais dcc ok ?
15:54:54lImbusme to
15:54:56BC|codingpaper is good for lots of things, but not for word searching
15:55:09BC|codingyes, dcc is fine :)
15:55:16c0uttabe aware my firewall might stop me from doing this
15:55:29BC|codingi have all other methods avilable to me also
15:56:22c0uttahey, it's working.
15:57:02BC|codingIt's my lucky day
15:57:07c0uttait's 975kb compressed
15:57:38c0uttait has taken me AGES to find the PDF
15:58:16BC|codingyou know, it never crossed my mind to go looking for it, but I am mighty pleased it exists
15:58:36c0uttait apeears to have been scanned, but then the text converted
15:58:55BC|codinggotta love OCR
16:02:19 Join solaris [0] (
16:02:40BC|codingfantastic :) ....I owe you one :)
16:03:03 Quit solaris (Client Quit)
16:03:16lImbusme too. it's great. no matter
16:03:30lImbuserr.. no more whining I forgot the bible at work
16:04:37BC|codingand there it is 77 :)
16:05:15 Join solaris [0] (
16:05:55lImbusyeah, with a bad and ugly example :-/
16:06:02 Join amansinger [0] (
16:06:14BC|codingback to coding now - thanks guys
16:06:56 Quit amansinger (Client Quit)
16:09:07 Nick c0utta is now known as c0utta{zzZZ} (
16:10:31solarishi, quick question, how can reply back to the mailing list?
16:22:31 Quit AciD (Read error: 104 (Connection reset by peer))
16:29:41 Join Galik [0] (~galik@
16:31:54lImbussolaris: afaik, you must be subscribed. then it should be ok to reply to the address everybody sends his messages to (can't tell you, i am not subscribed either)
16:48:30 Join Ralle[DK] [0] (
16:50:06Ralle[DK]any1 wanna help me with the flash thing?
16:51:01Ralle[DK]im very noob in the rockbox
16:53:30lImbussorry, I can't. never done and have to go.
16:53:33 Quit lImbus (" HydraIRC -> <- Nine out of ten l33t h4x0rz prefer it")
16:53:48Ralle[DK]who can help me?
16:58:40methangasRalle[DK], go to
16:58:52methangascheck the section called "installing the firmware"
17:03:39 Join AciD [0] (
17:27:16Ralle[DK]what do i do if it fails and its the right rom and i havent turned off?
17:27:21Ralle[DK]just for security reasons
17:33:25Ralle[DK]are you there?
17:33:53methangasthink there's something in the manual about it
17:34:19Ralle[DK]hmm 2 secs
17:39:11 Quit AciD (Read error: 104 (Connection reset by peer))
17:40:23 Nick amiconn|work is now known as amiconn|away (
17:45:35 Join AciD [0] (
17:46:28 Join mecraw__ [0] (~mecraw@
17:50:38***Saving seen data "./dancer.seen"
18:10:01 Quit Ralle[DK] ()
18:19:59 Join lImbus [0] (
18:29:13 Nick amiconn|away is now known as amiconn (
18:55:10 Join Ralle[DK] [0] (
18:57:17lImbusralle: hey, how's the flashing
18:58:52 Join cjnr11 [0] (
18:58:55 Part cjnr11
19:02:39 Quit lImbus (" Like's GUI? Then try HydraIRC -> <-")
19:03:19Ralle[DK]hey! i've read all the manual and cant get my archos to flash! i need help!
19:03:42Ralle[DK]any1 who can help me?
19:04:33Ralle[DK]why are you all afk?
19:05:10Ralle[DK]i cant make my flashthingy work
19:06:26Ralle[DK]are you there?
19:06:42Ralle[DK]any1 there?
19:06:50Ralle[DK]i'll go get some food
19:13:05 Nick facted is now known as facted|away (
19:31:20 Quit Nibbler (Read error: 104 (Connection reset by peer))
19:33:29Ralle[DK]is any1 here?
19:38:51Ralle[DK]can any1 here me?
19:50:39***Saving seen data "./dancer.seen"
19:51:32 Quit facted|away ("Trillian (")
19:53:31 Join amansinger [0] (
19:54:15Ralle[DK]is any1 here?
19:55:06amansingerat least one person
19:55:20amansingeractually at least two
19:59:08amansingerWould you like something?
20:01:30 Quit amansinger ()
20:03:44Ralle[DK]are you here?
20:25:56 Join facted|away [0] (
20:34:27 Quit Ralle[DK] (Read error: 54 (Connection reset by peer))
21:01:19 Quit AciD (Read error: 104 (Connection reset by peer))
21:04:21 Join deadite66 [0] (
21:16:25 Join Nibbler [0] (
21:36:10 Quit deadite66 (Read error: 110 (Connection timed out))
21:44:04 Part amiconn
21:46:28 Join scott666 [0] (
21:50:40***Saving seen data "./dancer.seen"
22:12:03 Quit Nibbler (Read error: 54 (Connection reset by peer))
22:14:52 Quit Galik ("Client exiting")
22:36:00 Join Ralle[DK] [0] (
22:36:12Ralle[DK]can any1 give me some info?
22:37:09 Quit Ralle[DK] (Client Quit)
22:46:57 Quit facted|away ("Trillian (")
23:01:05 Join AciD [0] (
23:07:54 Nick c0utta{zzZZ} is now known as c0utta{afk} (
23:10:30 Join Ralle[DK] [0] (
23:10:56Ralle[DK]is anybody here?
23:15:04 Part Ralle[DK]
23:21:58 Join amansinger [0] (
23:23:25 Quit amansinger (Client Quit)
23:36:19 Join amiconn [0] (
23:46:52 Quit methangas (" HydraIRC -> <- Nine out of ten l33t h4x0rz prefer it")
23:47:15 Join lImbus [0] (
23:50:42***Saving seen data "./dancer.seen"
23:51:26 Join Nibbler [0] (

Previous day | Next day