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 2006-03-08

00:00:44Mikachutry throwing out bits of ints on the ground
00:02:36 Join ScootScat [0] (
00:04:36 Quit bluey ("Leaving")
00:04:42 Join dn|gechu^aO [0] (
00:05:31 Quit midkay_ (Read error: 104 (Connection reset by peer))
00:08:51markunyoungcerea1: what are you working on?
00:09:37 Quit dn|gechu^aO ("WATCH ME DISAPPEAR INTO THE NIGHT!")
00:11:07youngcerea1top seacret
00:11:48markunsomething new or improving some things?
00:13:31***Saving seen data "./dancer.seen"
00:17:01 Quit gracchus (Read error: 110 (Connection timed out))
00:17:19 Quit ScootScat72 (Read error: 110 (Connection timed out))
00:26:24 Join TCK- [0] (
00:28:22BHSPitLappyyoungcerea1: do you know where Marburg is?
00:34:53*t0mas feels the need to rejoin the channel to showoff :)
00:34:57 Part t0mas
00:34:57 Join t0mas [0] (n=tomas@rockbox/developer/t0mas)
00:35:16t0mas00.34.57 Join t0mas [0] (n=tomas@rockbox/developer/t0mas)
00:35:19preglowhook me up as well
00:35:31t0massend me a mail...
00:35:38preglowdo i have to mail? :>
00:35:46t0masI can e-mail myself..
00:35:52t0masit's just that I need a list somewhere
00:35:55preglowmailing has to be involved? heh
00:35:59preglowi'll send you a mail
00:36:02t0masbecause the freenode people want them in groups of 10 or something
00:36:15 Join webguest44 [0] (
00:37:42webguest44what targets do transparent bitmaps work on? is it only 16bit?
00:38:49t0maspreglow: tnx, it's on my "get cloak" list :)
00:39:01preglownow, bedtime, later
00:39:07t0masgood idea
00:39:12Paul_The_Nerdwebguest44: It's only on color targets so far.
00:40:18 Quit TCK (Read error: 110 (Connection timed out))
00:40:49webguest44what targets do transparent bitmaps work on? is it only 16bit?
00:40:49 Quit webguest44 (Client Quit)
00:41:29amiconnt0mas ?
00:41:42t0mas*yawn* yes?
00:41:53amiconnWhat are you talking about??
00:41:57 Join hydrahead [0] (
00:42:07t0masthe cloak thing?
00:42:14t0masaren't you on the dev-mailinglist?
00:42:33t0masread the last 4 lines of the e-mail...
00:42:37t0masit's a hostmask like thing
00:42:40 Join webguest44 [0] (
00:42:50t0masfreenode uses it to "mark" people of a project
00:42:58amiconnUgh, I don't check mail *that* often.
00:42:59t0maskind of like showing "Hey, I'm a rockbox developer"
00:43:12webguest44Paul_The_Nerd: thanks. does "so far" mean amiconn is working on it for greyscale?
00:43:36t0masthat often? it's more than 1 hour ago... already had 3 replies :P
00:44:18amiconnI usually check once or twice per day...
00:44:24t0masah ok
00:44:48t0masmy pda starts beeping when I have new e-mail
00:45:03t0masand normally it's laying around my keyboard somewhere :)
00:45:39amiconnMy Amiga doesn't check permanently, only when requested. It could check regularly, but it wouldn't make sense
00:46:03amiconnCan't beep atm, need to repair audio out (already for the 3rd time) :(
00:46:14t0mashm... ok
00:46:23t0masbut if you want one... please send me an e-mail
00:46:26t0masI'm going to bed now
00:46:45Paul_The_Nerdwebguest44: "So far" means simply 'so far.' I don't believe anyone's working on it for grayscale, and it wouldn't really have much to do with the grayscale lib anyway I believe.
00:48:08*amiconn isn't exactly keen of reading emails within minutes after they arrived
00:48:39amiconnThat's the good thing with emails, compared to phone calls: they're asynchronous
00:48:47Mikachui have my thing set up to fetch every 5 minutes
00:49:12 Quit webguest44 ("CGI:IRC (EOF)")
00:50:29 Join sero [0] (
00:50:49seroi have a question about the ihp120
00:51:40seroi this always like this, that the line-out ist complete like the headphone
00:51:53serothat it is possible to regulate the volume??
00:52:22 Join midkay [0] (
00:52:35serobecause i normaly know it from other players that the line-out is an constant loud signal
00:52:45markunsero: the volume cannot be changed independatly on the irivers
00:52:46dpassen1If I understand you correctly, then yes, the line-out is affected by the volume.
00:53:12seroi thout it is a bad mistake in the device
00:53:33markunI also think it's a bit unfortunate
00:53:38dpassen1The volume is applied digitally, though, so a 0 db volume setting is a true lineout.
00:54:01seroa true line out?
00:54:14serowhat means that, the loudesd signal??
00:54:37dpassen1It happens to be the loudest on a H120, but that is not always the case.
00:55:16*amiconn squashed that nasty dependency problem :)
00:55:26seroyes it is the loudesd, but that means that with this configure the battery is faster at low voltage
00:56:11 Quit matsl (Remote closed the connection)
00:56:46amiconnvolume shouldn't influence battery lifetime if you're only using lineout, i.e. no headphone plugged
00:56:49seroa volume off -84db is mue
00:57:17seroreally should not?
00:57:34 Quit Thus0 ("Leaving")
00:58:15seroso to summerize this
00:58:59seroif iwant to make musik with the iriver on my amp with line-out i have to put volume to 0db for real line-out
00:59:42serobut when i want immediatly hear headphones after this i have to set volume to something human like -40db
00:59:55Paul_The_NerdYes. The iRiver firmware works the same way, it's part of how they designed the hardware.
01:00:39serothat isn't really cool, often i forget to set volume back and so i have a problem, after 3-4 times my ears are really bad shocked
01:00:41muesli-XavierGr ?
01:01:33 Quit hydrahead ("CGI:IRC (EOF)")
01:02:14Paul_The_Nerdsero: Unfortunately, again, it's the hardware that does this. It cannot be changed.
01:02:33seroi know that this is the hard-ware
01:02:51serobut in this way isn't it possible du reassamble it?
01:03:13seroa better solution would be fine
01:04:26Paul_The_NerdSince the volume is applied digitally, not really. The same signal is sent to both the line out and the earphone out, and there's not a way to get the audio before its volume is adjusted for just the line out, I believe.
01:05:27Mikachuyou could set the volume back to -40 every time you stop
01:05:35serocouldn't it be possible to write a small code to set the signal to max with one button
01:06:11seroin style of a panic button like when you hear it is too loud
01:06:17markunsero: I have headphone.cfg and lineout.cfg in my root..
01:06:34amiconnsero: Just press stop...
01:06:37Paul_The_Nerdsero: Yeah, if you create two config files, with just the volume lines, you could simply click on them.
01:07:12serohmm, i am not so familiar jet with rock box
01:07:18serobut it sounds good
01:07:20*amiconn has 3 .cfg files in his root to set parameters for different environments
01:07:24serook thx
01:08:51Paul_The_Nerd .cfg files are the answers to many, many questions.
01:10:01 Join virtualball2 [0] (
01:12:42 Join TiMiD[FD] [0] (n=TiMiD[
01:13:19*amiconn is just testing a vertical scrolling oscilloscope - doing the scrolling in software
01:13:47amiconn..on recorder, w/ vertical pixel packing, i.e. vertical scrolling involving bitshifting hell
01:14:13amiconnIt's working fast enough :)
01:16:15 Quit sero ("leaving")
01:29:34 Quit virtualball2 ("virtualball2 has no reason")
01:33:11 Quit quobl (Remote closed the connection)
01:33:33 Join quobl [0] (i=id@tor/session/x-a19bca881f6ad2f5)
01:35:11 Quit easy_east (Read error: 110 (Connection timed out))
01:36:28 Quit ender` (" Washing your car to make it rain doesn't work.")
01:38:10 Quit muesli- ("ich will Kühe!!!")
01:41:53 Quit nobelium ("Input/Output Error")
01:43:17 Join mikearthur [0] (
01:46:56 Quit herz42 ("Und wech")
02:13:34***Saving seen data "./dancer.seen"
02:13:50 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
02:21:09 Quit mikearthur ("Konversation terminated!")
02:21:51 Join mikearthur [0] (
02:28:02BHSPitLappyamiconn: as long as you're in the plugin (source), think you could move the "Speed: " text to a fixed location (corner?)? :)
02:28:12BHSPitLappyor would that not be a minor adjustment
02:28:33BHSPitLappyin oscilloscope
02:28:49BHSPitLappywait, are you not talking about that plugin I'm thinking of?
02:28:52amiconnI guess you mean oscillograph...
02:29:01BHSPitLappyoops, yeah :P
02:29:11amiconnOscilloscope is not yet ported to swcodec
02:29:16BHSPitLappyoscilloSCOPE... so it'd have jiggly lines
02:30:01amiconnOscilloscope does the same as oscillograp, except it draws horizontally instead of vertically
02:30:20BHSPitLappy"* amiconn is just testing a vertical scrolling oscilloscope"
02:30:20Shadowarrior13lol, the word scope implies squiggly lines?
02:30:27 Join fiftyfour123 [0] (
02:30:30amiconnOscillograph is supposed to scroll vertically so the position of the various messages makes sense
02:30:37BHSPitLappyShadowarrior13: based on every oscilloscope I've seen, yes
02:31:15amiconnHowever, it doesn't scroll on anything else that archoses and H1x0, because oscillograph uses a special hardware feature of the lcd that isn't available for the colour targets or greyscale ipods
02:31:17midkayoscilloscope is a misleading nane..
02:31:18BHSPitLappycan text not be drawn vertically?
02:31:41amiconnOscilloscope uses software scrolling
02:31:42midkayBHSPitLappy, you could if you wanted to, of course.. why?
02:32:02BHSPitLappy"[amiconn] Oscillograph is supposed to scroll vertically so the position of the various messages makes sense"
02:32:21midkaywhat does that have to do with drawing text vertically?
02:32:30 Quit fiftyfour123 (Client Quit)
02:32:42amiconnThe reason why oscillograph draws vertically because this hardware roll function is only available for that direction, and back when oscillograph was written, the rockbox lcd functions were too slow to do it in software
02:32:47BHSPitLappymidkay: umm...
02:33:01BHSPitLappyamiconn: ah
02:33:05midkayBHSPitLappy, back at you..
02:33:25amiconnSo, after our lcd_update() got fast enough, I wrote oscilloscope as a complement
02:33:43BHSPitLappymidkay: ... when the graph is vertical, the text goes horizontal
02:33:58midkayBHSPitLappy, what's wrong with that? much easier to read..
02:34:30amiconnI want to port oscilloscope to swcodec as well
02:35:04qwmBHSPitLappy: you're so incredibly annoying
02:35:20midkayamiconn, that'd be cool
02:35:21amiconnThere's one more problem with the scrolling though. Right now it wouldn't scroll on iPods. I'd need a working user timer...
02:35:56BHSPitLappymidkay: because, he's writing a horizontal one..
02:36:08BHSPitLappytherefore, I asked if text could just be drawn vertically
02:36:11midkayBHSPitLappy, what's the difference? why not draw the text horizontall each way?
02:36:14amiconnOscilloscope has no texts
02:36:23midkayyou'd rotate the ipod to view it or what?
02:36:24BHSPitLappyoscillograph does :P
02:36:31midkayamiconn, the "speed: xx" text i think he means.
02:36:45BHSPitLappyit lines up with the part of the graph that's drawing
02:36:50midkayi don't understand why he wants to put it in a harder-to-read location, though..
02:36:51amiconnYe. oscilloscope uses a constant speed
02:36:54BHSPitLappyit also gets garbled immediately
02:37:05midkayBHSPitLappy, not for me..
02:37:19BHSPitLappyhow is the corner harder to read?
02:37:24BHSPitLappyit's just like starfield
02:37:28midkayyou said you wanted it vertical.
02:37:32midkaythat's not vertical, that's horizontal.
02:37:42midkayhorizontal. horizon. across. standard.
02:37:46BHSPitLappyyou're combining two different points of a conversation.
02:37:54BHSPitLappypay. more. attention. context.
02:38:05midkayi'm. trying. you're. so. nonsensical.
02:38:25BHSPitLappylearn. english.
02:38:34qwmfuck. you. Lappy.
02:38:41Paul_The_Nerd" I asked if text could just be drawn vertically"
02:38:50midkayBHSPitLappy, vertically != sitting in a corner.
02:38:52midkaylast i checked,
02:38:56Paul_The_NerdThese are your words BHSPitLappy... sounds like vertically drawn text to me
02:38:56amiconnBHSPitLappy: We can't put the text in a fixed location in oscilograph, because of how it works (on targets where it works as it's supposed to, i.e. archos and H1x0)
02:39:06BHSPitLappythose were two DIFFERENT places in a conversation
02:39:11midkayBHSPitLappy, you keep mixing them then.
02:39:22BHSPitLappyI said the corner thing once, and it died.
02:39:27BHSPitLappyyou brought it back out of context.
02:39:31 Quit Aditya ("Trillian (")
02:39:33amiconnThe hardware LCD roll effects *all* the LCD content
02:39:35midkayso, back to vertical - why?
02:39:44BHSPitLappycorner thing == the vertical scrolling oscillograph
02:39:59BHSPitLappyhorizontal message == relevant to a horizontal scrolling oscilloscope
02:40:18midkaywhat about "vertical message"?
02:40:37BHSPitLappys/horizontal message/vertical message/
02:40:58midkayso. you want text that is drawn vertically when your ipod is horizontal?
02:41:04midkayhow is that easier to read than horizontal text?
02:41:33BHSPitLappy"[amiconn] Oscillograph is supposed to scroll vertically so the position of the various messages makes sense"
02:41:59BHSPitLappy"[BHSPitLappy] can text not be drawn vertically?"
02:42:30midkayBHSPitLappy, i still don't get why you want text vertically. what's the point?
02:42:56BHSPitLappymidkay: get a piece of paper, draw a line graph
02:43:18 Quit nave7693 (Connection timed out)
02:43:19BHSPitLappyif information needs to be placed pertaining to points on that line, wouldn't you have it vertically
02:43:31BHSPitLappyalong the x axis
02:43:42midkayvertical is y..?..
02:43:47midkayvertical = up and down.
02:43:52BHSPitLappyI know...
02:44:07BHSPitLappythe points of information are along the x axis..
02:44:07midkayas in:
02:44:10DBUGEnqueued KICK midkay
02:44:12 Join Aditya [0] (
02:44:19BHSPitLappythe text is shaped vertically.
02:44:22amiconnWe can't put text vertically, at least not without *major* work
02:44:35BHSPitLappya good example is the build chart on
02:44:50midkayBHSPitLappy, you mean layered vertically? i'd call that horizontal layers..
02:44:52BHSPitLappythose device names, along the x axis... the text is vertical, to make sense
02:44:59Adityaamiconn: yes you can
02:45:03Adityayou just do text.rotate(90)
02:45:06Adityaand bam
02:45:08BHSPitLappyI have no clue why you'd call that a "layer".
02:45:14midkayon top of each other?
02:45:28amiconnAditya: Haha, how would you do that in rockbox?
02:45:32midkayBHSPitLappy, i see what you mean on the build chart..
02:45:42BHSPitLappymidkay: if text was layered, it'd be rubbish.
02:45:45midkayi don't have ANY idea why you'd want to have it drawn like that on the LCD. you couldn't even read it.
02:45:45Adityaamiconn: I didn't say it was for rockbox =P
02:45:51midkaybefore it flew off the lcd..
02:46:03BHSPitLappymidkay: run the oscillograph plugin, with sound playing
02:46:08BHSPitLappyand make the Speed message show up
02:46:23BHSPitLappyimagine the same plugin, but rotated 90 degrees... the text'd be vertical.
02:46:35BHSPitLappyas long as it was following the drawing line, at least
02:46:39Paul_The_NerdBHSPitLappy: But it's *not* exactly the same. The only reason the Speed message *moves* is because of the hardware roll
02:46:53BHSPitLappyI know... I understand that
02:47:07BHSPitLappyI'm just trying to explain to midkay what I was saying in the first place
02:47:23BHSPitLappysince we have a little communication error, it seems...
02:47:24midkayBHSPitLappy, i see what you mean - in a way, about being garbled..
02:47:25Paul_The_NerdSo, why do you want to imitate that behaviour in a horizontal version, rather than just making the text stationary?
02:47:33midkaybut i don't see why making the text vertical would fix that?
02:47:46BHSPitLappymidkay: it wouldn't, placing the text in the corner would.
02:47:58BHSPitLappythe vertical text was referring to a horizontal oscillo
02:48:23Paul_The_NerdBHSPitLappy: The vertical one, the text *can't* be placed in the corner. The horizontal one, it can.
02:48:45amiconnI'll probably combine both oscillo* plugins into one, offering both aspects
02:48:45Paul_The_NerdWell, maybe can
02:48:53BHSPitLappybut the horizontal one wouldn't even run at all on the ipod, right
02:49:17amiconnI *hope* I can do it w/o the user timer, so it will work on all targets, and also in the sims
02:49:23BHSPitLappy(because of this rolling business)
02:49:39Paul_The_NerdBHSPitLappy: The Vertical one is the one that uses the hardware roll, right?
02:49:40amiconnNo, the vertical one doesn't run corretcly
02:50:02BHSPitLappythe hardware is top to bottom, and that's what the graph draws
02:50:09BHSPitLappy(if I understand all this correctly)
02:50:24Paul_The_NerdThe graph draws however it's told to. The graph is *moved* by the roll.
02:50:38amiconnOn iPod, it draws from top to bottom, and when it reaches the bottom, starts over at the toü
02:50:55amiconnSame on H300
02:50:57midkayBHSPitLappy, alright, the corner makes sense and always has. but a horiz oscillo - why not in the corner too?
02:51:07amiconnThat's however *not* how it's supposed to look
02:51:22Paul_The_NerdBHSPitLappy: On the H120 and others, the graph just scrolls, it doesn't reset at the top.
02:51:55amiconnIt should draw from top to bottom and then stay at the bottom and continue drawing, scrolling up the drawn graph
02:52:11amiconnIt does this correctly on archos and H1x0
02:52:39Paul_The_Nerdamiconn: Though, once per screen length there's about 4 or 5 horizontal rows of pixels that are not oscillograph data (they look random)
02:52:50Paul_The_NerdOn the h120 hardware
02:52:56BHSPitLappymidkay: it would make sense, too. I was just responding to amiconn saying "it's vertical because the messages wouldn't make sense otherwise"
02:53:28amiconnBHSPitLappy: I said it's vertical because of its history on the archos (the hardware rolling business)
02:53:31Paul_The_Nerdamiconn: Sorry, exactly 4 rows of random (or random looking) pixels, once per screen length
02:53:46BHSPitLappyamiconn: mmk
02:53:52amiconnPaul_The_Nerd: Ah, that's then a problem with the hardware roll on h1x0
02:53:58midkayBHSPitLappy, he wasn't saying it's vertical because the messages are horizontal. he said that SINCE it's vertical, the messages make sense being horizontal.
02:54:22amiconnSome parameter is probably off. I don't care much, as I plan to scrap hardware roll anyway
02:54:49BHSPitLappymidkay: either way, this conversation has been read way too much into
02:54:58Paul_The_Nerdamiconn: Okay, just thought I should report it, while I was noticing it. :)
02:55:05midkayBHSPitLappy, translation of your last message: "ah, i see"..
02:55:29BHSPitLappyactually, no.
02:55:31midkayyeah, ha.
02:55:46BHSPitLappy"ah, I see" means I wanted to know.
02:56:12midkayyour interpretation, perhaps..
02:56:39amiconnPaul_The_Nerd: If you set the speed very low, you'll also notice the the display 'jumps' whenever these 4 pixels reach the top...
02:57:36BHSPitLappy"BHSPitLappy: I have brown hair. midkay: Ah, I see." does it make sense if the information wasn't of interest to the person ;)
02:57:57midkayBHSPitLappy, yes - seeing != being interested by.. iirc..
02:58:20 Quit Daishi (Remote closed the connection)
02:59:01BHSPitLappyso it makes sense that you saw that I have brown hair, even when you never, in fact, saw it at all?
02:59:06CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
02:59:06*BHSPitLappy 's head explodes
02:59:12Paul_The_Nerdamiconn: What's interesting, is I don't remember seeing this behaviour on my previous build from shortly after the oscillograph was first implemented on H120 earlier. (Though I may have just not noticed it.
02:59:14 Quit mikearthur (Remote closed the connection)
02:59:33amiconnPaul_The_Nerd: I just checked the datasheet; this is because the LCD controller is 160x132 while the panel is 160x128
02:59:48midkayBHSPitLappy, "see" can also mean "understand".
02:59:51 Join mikearthur [0] (
03:00:01amiconnSo there are 4 extra pixel rows in the GRAM which don't correspond to actual pixels - normally
03:00:18amiconnThis means lcd_roll() can't be made work right on H1x0
03:00:24BHSPitLappy"I see the equation. I don't, however understand it."
03:00:25amiconnOne more reason to scrap it...
03:00:33midkayBHSPitLappy, that works. i said "sometimes".
03:00:38Paul_The_NerdWell, I never voted against the scrapping anyway. :)
03:01:33BHSPitLappymidkay: I don't recall ever saying "ah, I see"
03:01:52midkayBHSPitLappy, haha - nice memory, again - you didn't.
03:01:58BHSPitLappyomg wait, you're right about everything midkay
03:02:00BHSPitLappyI concede.
03:02:00midkayBHSPitLappy, remember, i translated it?
03:02:26midkayBHSPitLappy, thanks.
03:02:47 Quit mikearthur (Remote closed the connection)
03:05:14 Join mikearthur [0] (
03:06:20amiconnPaul_The_Nerd: Perhaps it could be made working correctly with the area scroll feature. This would however still leave us with the problem how to roll on iPod or H300
03:07:36Paul_The_Nerdamiconn: Well, I doubt a hardware scroll is particularly necessary anyway.
03:07:55amiconnI would like to have that on H300
03:08:23amiconnI did some calculation; software scrolling will be significantly slower on H300 than on H1x0 or archos,
03:08:37Paul_The_NerdWhy's that?
03:08:43amiconneven though vertical scrolling on archos and H1x0 involves bitshifting (!)
03:09:14amiconnWhy? Larger display resolution and depth, hence much more data to move
03:09:38amiconnAnd the framebuffer is in SDRAM
03:09:47 Quit mikearthur (Remote closed the connection)
03:10:03 Join mikearthur [0] (
03:10:06amiconnVertical full-screen scroll takes ~2ms on archos and ~1ms on H1x0 (not counting the lcd_update() )
03:10:16amiconnIt will takes ~13ms on H300...
03:10:25amiconnThat's at 45 MHz
03:11:21 Join gracchus [0] (
03:11:47Paul_The_NerdHow is the framebuffer copied to the LCD?
03:12:34Paul_The_NerdI mean, I don't really know how anything in that part of rockbox works, but it seems a roll could be emulated fairly efficiently by changing how that step works, unless there's something special there that prevent that.
03:13:00amiconnHow do you think this would work?
03:13:27amiconnThe framebuffer data needs to be copied into the GRAM of the LCD controller
03:14:23Paul_The_NerdWell, if you
03:15:51Paul_The_NerdI honestly don't know if this would be faster, as I really don't know what you're doing to move the data quickly, and I'm not to familiar with fast hardware programming at all.
03:16:38Paul_The_NerdBut, for this example, assume you've rolled half the height of the screen. Couldn't you simply copy the bottom half of the framebuffer into the top half of either another buffer or GRAM, and then the top half into the bottom? It's one more copy (or two more, if you need to make a second framebuffer). By "top half" and "bottom half" I mean the memory that represents the top and bottom half of the pixels, rather than specifically..
03:17:02Paul_The_NerdI mean, it's probably just a stupid idea, since I don't know how things work.
03:17:39amiconnThe GRAM can't be accessed directly. It is accessed by (memory-mapped) I/O
03:17:46 Quit mikearthur (Remote closed the connection)
03:17:57Paul_The_NerdAnd a second framebuffer would be way too slow?
03:18:01 Join mikearthur [0] (
03:18:24amiconnA second framebuffer would mean even more copying
03:18:37amiconnIt's the copying that is slow...
03:18:59amiconnThe 13 ms are just for moving the data in the framebuffer a few pixels in either direction
03:19:28amiconnIt would be somewhat faster when scrolling by an even amout of pixels
03:19:46 Join TCK [0] (
03:21:30Paul_The_NerdHow slow/fast is drawing to the framebuffer? Like, would it be possible to simply double the size of the framebuffer, and draw everything twice?
03:21:34 Join perldiver [0] (
03:32:57 Quit TCK- (Read error: 110 (Connection timed out))
03:47:20 Join Mikaelh [0] (
03:47:38 Quit Seed (Nick collision from services.)
03:47:46 Join Seed [0] (
03:48:49 Quit Mikachu (Read error: 104 (Connection reset by peer))
03:54:38gracchushello, does anyone have ideas/intentions on modularizing recording subsystems?
03:55:06gracchusseems like sometimes it's done by cpu, sometimes by platform, for instance
03:56:03gracchus(I'm taking a tentative swipe at my ipod's wolfson)
04:00:14gracchusI guess SWCODEC && HAVE_RECORDING would indicate the use of pcm_record.c
04:06:18 Quit mikearthur (Remote closed the connection)
04:13:35***Saving seen data "./dancer.seen"
04:35:30gracchusmmf, okay .. conditionalized, thread and queue initialize, files show up on disk .. I suppose at some point I need to turn on the adc
04:35:48gracchusI don't imagine a datasheet hsa turned up?
04:40:13 Join virtualball2 [0] (
04:42:36 Quit Maxime (Read error: 104 (Connection reset by peer))
04:47:17 Join nave7693 [0] (
04:47:29 Part nave7693
05:01:35 Join Rob2222_ [0] (
05:04:12 Join Maxime [0] (
05:08:01lostlogicpreglow? linuxstb? ipod LCD driver knowing people?
05:16:42 Join damaki_ [0] (
05:17:26 Quit damaki (Read error: 110 (Connection timed out))
05:19:52 Quit Rob2222 (Read error: 110 (Connection timed out))
05:34:51cooblostlogic: whats the question?
05:35:05 Part virtualball2
05:35:37lostlogiccoob: I forgot my question :(
05:35:44lostlogicoh wait, no I remember
05:36:02lostlogicthe yield in the while loop −− do we really need to do it for every loop iteration?
05:40:00 Join Sinbios [0] (
05:44:32 Quit RotAtoR ("zzzzzzz")
05:56:13 Quit Sinbios (Nick collision from services.)
05:57:06 Quit Shadowarrior13 ("( :: NoNameScript 4.01 :: )")
05:57:06 Join Sinbios [0] (
06:00:26gracchusmore noobishness .. IISFIFO in pcm_playback seems to be a memory-mapped fifo; what is it that 'IIS' stands for in this case?
06:01:32BHSPitLappyin this situation?
06:01:33 Join newnick [0] (
06:01:40 Quit newnick (Client Quit)
06:01:47 Join SlaytanicLemmy [0] (
06:01:50BHSPitLappy;) jk
06:02:02BHSPitLappyI have no clue.
06:02:21Paul_The_NerdMan, I only know one thing that abbreviates to IIS, and I'm pretty sure that's not it.
06:02:33BHSPitLappysome things aren't meant to be known.
06:02:43BHSPitLappyPaul_The_Nerd: you're an iPodder, right?
06:02:53Paul_The_NerdBHSPitLappy: Not originally.
06:02:55gracchusI'll settle for thinking of it as 'It... is... serial'
06:03:08Paul_The_NerdI have a 'Nano, but I also have a couple iRiver players.
06:03:57BHSPitLappyPaul_The_Nerd: I'm thinking of starting a general iPod channel, you interested in participating?
06:04:25SlaytanicLemmyI am getting a 5G for rockbox tomorrow. Is there a feature available that allows you to surround a playing song with the rest of the albumn fron which it came? If not, where in the code would this go? Would it be a plugin?
06:04:28*Paul_The_Nerd shrugs.
06:04:50Paul_The_NerdBHSPitLappy: I'll set my client to auto-join it, so I'll be in there whenever I'm here, and respond there when I remember to.
06:04:58SlaytanicLemmyThis was a feature I always wanted on my Karma.
06:05:13Paul_The_NerdSlaytanicLemmy: Surround it how?
06:05:26BHSPitLappyPaul_The_Nerd: it's #ipods by the way, #ipod is pointlessly registered
06:05:43SlaytanicLemmyAs in modify the current playlist so that the songs previous on the album are before and the songs later are after.
06:05:54SlaytanicLemmyThen you could listen to the whole album
06:06:03BHSPitLappyI've never heard of such an intuitive player.
06:06:17Paul_The_NerdSlaytanicLemmy: Well, if you just click on a song in a folder, if shuffle is disabled, it will create a playlist out of that folder, in the order the songs are in that folder, and start playing the one you clicked on.
06:06:26SlaytanicLemmyAlternatively, it could jump out of the existing playlist with a bookmark and create a new playlist with the album.
06:06:29BHSPitLappyisn't the point of a playlist that you decide what order you wish to hear songs in?
06:06:42Paul_The_NerdSo, if your album is in a folder on its own, then you would have a playlist of that album, and be listening to whichever song you clicked on.
06:06:52SlaytanicLemmyI usually just randomize my whole 5000 songs.
06:07:12SlaytanicLemmySometimes, I hear a song that makes me want to listen to the whole albumn - in order.
06:07:29SlaytanicLemmyThis would be a great (although esoteric) way to do it.
06:07:56BHSPitLappySlaytanicLemmy: well, can you code? :)
06:08:06 Join damaki__ [0] (
06:08:15Paul_The_NerdWell, until there's a database, it can't read tags until the songs are already loaded, so you're pretty much restricted to folder options. You could Queue that song's folder, I suppose
06:08:18BHSPitLappythe general way opensource works is something gets programmed because the programmer wanted it
06:08:21SlaytanicLemmySince the source is open, I figure it is easily implementable - get the albumn - save playlist with bookmark - create new playlist from the albumn with shuffle/random off .
06:08:40BHSPitLappygreat, you know JUST how to do it then! :D
06:08:53SlaytanicLemmyRight. Where should it go?
06:09:02SlaytanicLemmyIs Rockbox modular?
06:09:21SlaytanicLemmyWould it be a plugin or a direct feature?
06:09:48BHSPitLappyprobably need to write/submit a patch
06:09:53SlaytanicLemmyI have coded in the past - C, Cobol, JCL, Perl, etc.
06:09:56BHSPitLappythat's the way of the rock
06:10:24BHSPitLappygrab a cvs tree!
06:10:44SlaytanicLemmyI just need to know where to begin - I don't want or really need to get into all of th hardware stuff.
06:11:18BHSPitLappylike paul said... nobody has implemented the database yet
06:11:30SlaytanicLemmyWhat is the dev environment? C or C++? And can I dev in standard Linux.
06:11:36BHSPitLappyuntil then, it can't find other songs with similar album, etc
06:11:37Paul_The_NerdYeah, so if you want to use "Album "data from the tags, you'll need to probably wait on that.
06:11:51SlaytanicLemmyDatabase should not be required if things are organized into album directories.
06:11:54BHSPitLappyor write the database... ;)
06:12:10gracchusSlaytanicLemmy: ... yeah I was going to say that!
06:12:10SlaytanicLemmyThanks, Paul.
06:12:18BHSPitLappySlaytanicLemmy: don't submit a patch that's dependent on your method of file organization
06:12:25Paul_The_NerdBHSPitLappy: The database is *mostly* done, if I understand correctly.
06:12:30gracchusthe crosscompiler docs are very good
06:12:47Paul_The_NerdBHSPitLappy: He could do an "Queue surrounding files" as an additional context menu choice.
06:13:13SlaytanicLemmyRight - it would be a contextual menu from the play screen.
06:13:26BHSPitLappybut I don't see that being very useful (to be applied to CVS, anyway)
06:13:36BHSPitLappymore of an as-needed thing...
06:13:39***Saving seen data "./dancer.seen"
06:13:45BHSPitLappy(until there's a db :))
06:14:09SlaytanicLemmyIt would just queue all of the songs in the playing song's directory into a new playlist - the hard part would be to continue playing the playing song while the playlist is altered and switched.
06:14:10Paul_The_NerdSlaytanicLemmy: There's already a context menu (if you hold Navi, or Select, or whatever the archos equivalent is) with several options. One of them is Queue, so you'd just look at how that works, and add an expanded version of it, I imagine.
06:14:27SlaytanicLemmyThanks Paul.
06:14:48SlaytanicLemmyCan I test in the simulator and get accurate results?
06:15:06SlaytanicLemmySince this isn't iPod specific, it should work across players, no?
06:15:29BHSPitLappyjust filesystem polling and the existing queue function
06:15:34BHSPitLappyso it should
06:15:54Paul_The_NerdAnd there's usually people in here with various players willing to test a patch once it's at that point.
06:16:02SlaytanicLemmyWicked. I'll try and get started this weekend.
06:16:38BHSPitLappyI, however, am an iPod-specific person, so there could be some nuance I'm unaware of
06:18:18SlaytanicLemmyHow's voice menus coming along on the 'pod? Not a real hot topic, I imagine.
06:18:51BHSPitLappyI thought it worked, never tried.
06:19:01Paul_The_NerdDoes not work
06:19:03Paul_The_NerdDO NOT TRY IT.
06:19:03SlaytanicLemmyI was looking at a remote control (Nyko was the manuf., I think). That would be great for keeping your eyes on the road.
06:19:14 Quit imphasing (Connection timed out)
06:19:39SlaytanicLemmyYea, I thought I read that it locks up reeeeel good.
06:19:47Paul_The_NerdThe voice menus cause all kinds of fun problems, as of last time I tried, up to and including removing your ability to listen to music until you disable them *and* delete the .voice file (which somehow screws things up even when voice is disabled and the settings have been reset)
06:19:50 Quit perldiver (Read error: 110 (Connection timed out))
06:19:55Paul_The_NerdThey work fine on the H120 nowadays though.
06:20:01BHSPitLappysomeone was very interested in writing a TTS module for iPodLinux the other day, though
06:22:13gracchusIIS := Inter-IC Sound
06:22:36 Quit damaki_ (Read error: 110 (Connection timed out))
06:22:36gracchushow sensible
06:23:08SlaytanicLemmyI read on the status page for the iPod that there were thoughts on using the 2nd processor for some functions.
06:23:26SlaytanicLemmyDo all iPods have a 2nd processor capable of GP code execution?
06:23:38Paul_The_NerdSlaytanicLemmy: Well, it seems the current ideology is "Let's see how far we can go on just one" but yeah, I *believe* all ipods have two cores.
06:23:39SlaytanicLemmyOr is it just the Broadcom on the 5G
06:24:04Paul_The_NerdIt's definitely not *just* the 5g. The PP chip has two cores, if I understand.
06:24:38BHSPitLappywhy is the bootloader so stubborn :P
06:25:37SlaytanicLemmyIs it possible to make the bootloader configurable as to which OS to load by default? Or can the bootloader boot the original iPod OS?
06:26:26Paul_The_NerdIt can boot the iPod OS. You can configure it by building your own. Because Rockbox is intended as a *replacement* firmware, it'll probably never be configurable beyond "Hold Menu to start Retail"
06:26:35Paul_The_NerdBut the iPodLinux's Loader2 can be configured much more freely
06:26:59SlaytanicLemmyThanks P.
06:27:17SlaytanicLemmyDoes the CVS tree include the bootloader code, or is that separate?
06:27:32Paul_The_NerdIf you get rockbox-devel it's included
06:27:40Paul_The_NerdIf you just get rockbox you'll have to co bootloader as well
06:27:46Paul_The_Nerdand probably simulator
06:28:40SlaytanicLemmyI think I'll stick with just rockbox for now. I don't want to mess with the bootloader.
06:29:03Paul_The_NerdWell, as I said, you can use Loader 2.
06:29:09Paul_The_NerdAnd holding down Menu while booting will go into Retail
06:29:54SlaytanicLemmyDoes a reboot clear everything out? Is there state information saved to the HD that could create a cyclic crash?
06:30:49SlaytanicLemmyThat is, in the regular rockbox - not something that I may put there.
06:31:15Paul_The_NerdThere is a config block written to a sector on the HD, that can be cleared in the case of it causing a problem by turning on hold immediately after turning on your ipod.
06:31:44BHSPitLappySlaytanicLemmy: you should get rockbox-devel
06:31:45SlaytanicLemmyThanks. That would've taken me a while to figure out from the code.
06:31:48qwmi thank god for every minute BHSPitLappy is quiet.
06:32:30*BHSPitLappy comes in and out of conversations, sorry if I'm being random
06:32:50gracchusarmfans .. can someone suggest a good intro that covers irq/fiq handling?
06:34:35SlaytanicLemmyI can't find a good description of the album art feature. Where do you put the pictures? What format? Any proggies that can generate them from a bunch of audio files organized into albumn directories?
06:35:16Paul_The_NerdThere's not an album art feature yet
06:35:53Paul_The_NerdIt's just a patch, so the description will be with wherever the patch is kept. I'm assuming there's an entry for it at in the tracker, but I could be wrong. THey don't always make it there
06:35:55SlaytanicLemmyHuh? I've seen screen shots of WPS's with AA.
06:37:40Paul_The_NerdThe patch has changed its album art formatting requirements a few times already
06:42:26 Quit SlaytanicLemmy ("CGI:IRC (EOF)")
06:42:26 Quit c0utta ("CGI:IRC (EOF)")
06:43:17 Join aaronfg [0] (
06:43:50aaronfgwith the most recent CVS update for the ipod video, what's the new frame rate for the LCD?
06:45:32BHSPitLappyeffing fast.
06:45:40midkayaaronfg, it was 54fps iirc... so probably about 56.
06:46:38aaronfgbaby steps... :)
06:46:40aaronfgnice, thanks
06:46:47midkayaaronfg :)
06:47:27Paul_The_NerdWell, the framerate isn't going to mean so much, other than "more efficient means less time spent updating the screen, and more decoding audio"
06:49:32aaronfgany increase is fine in my book
06:49:47aaronfgwould this also help in the buffer issue on long/fast scrolling?
06:51:26*BHSPitLappy guessed that.
06:51:32Paul_The_NerdProbably not
06:51:41Paul_The_NerdMaybe minimally
06:52:21aaronfgis there anyway to see battery status in the sim?
06:54:50BHSPitLappyis it a laptop?
06:55:17aaronfgno, desktop
06:55:25gracchuspcm_playback's callback_for_more seems easy to stomp...maybe that code is in transition
06:55:33aaronfgon a laptop the sim takes the laptop battery? that's pretty neat
06:56:45 Join TCK- [0] (
06:57:00Paul_The_Nerdaaronfg: I think he was just joking
06:57:57BHSPitLappyaaronfg: yeah, I dunno what you were expecting...
06:58:57aaronfghey, i don't know. would be a cool thing to see
06:59:32aaronfgit's just sucky to test out WPS battery icons if you have to keep copying stuff over, ejecting the player, etc
06:59:42aaronfgi was just wondering if you could simulate battery status in the sim
07:00:03Paul_The_NerdYeah, it seems the sim should count from 1-100 on battery level with about 1/2 second for each one, or something.
07:00:12Paul_The_NerdOr 0-100 rather
07:00:14 Join DSGASDGASDGHASDH [0] (
07:02:38midkay= 50 seconds to see the whole loop? :)
07:02:47Paul_The_NerdRoughly, yeah
07:02:55midkayi'd say 0-100 in steps of 2 or 3, 1/2 or 1/3s each..
07:03:04 Quit aaronfg ("User pushed the X - because it's Xtra, baby")
07:03:11Paul_The_NerdI did say 1/2 second each. :-P
07:03:27 Quit Sinbios (Connection timed out)
07:03:42 Nick DSGASDGASDGHASDH is now known as Sinbios (
07:03:51Paul_The_NerdOr what it could do, is for the battery status conditional in WPS, it could just cycle through it at 2 seconds per condition, instead of cycling charge values
07:05:41midkayit takes you that long to see that it's working? :) i can hardly imagine even needing that.. if your bitmaps are in order and looking correct, and the wps code looks right, it should be fine.. i'd say 1/2s for each at the most
07:06:34JdGordonany1 do windows programming? iv got a tray app and for some reason when u right click it to get the menu the cursor changes to the egg-timer icon... but onyl the first time the menu loads.. any idea why?
07:07:19Paul_The_Nerdmidkay: 1/2 second might be good enough. I'm more in favor of slightly too much time, than too little for some people.
07:07:33midkayJdGordon, no idea
07:07:47midkayPaul_The_Nerd, i guess longer is better than complaints, true..
07:07:55JdGordon:( iv resorted to manually setting the cursor back to what it should be..
07:08:03JdGordonbut i dont know if thats so good
07:08:06JdGordonanyway.. back to idling
07:08:20Paul_The_NerdJdGordon: No clue here either. Windows is funky.
07:08:40gracchuson iPod, pcm playback and rec will have to share knowledge of FIQ state, so as not to abort eachothers' DMA transfers, and the FIQ handler should deal with both read and write FIFOs
07:08:47gracchushope I'm not way off track
07:09:46 Quit TCK (Read error: 110 (Connection timed out))
07:10:54 Nick Aditya is now known as Aditya|Sleep (
07:11:05Aditya|Sleephave a good one guys.. I am out for the night.. er.. morning
07:21:30 Join DJ_Dooms_Day [0] (
07:24:54 Quit midkay (Read error: 104 (Connection reset by peer))
07:25:03 Part gracchus
07:27:06JdGordondoes any1 know how to add conext menu's in explorer without ATL ?
07:41:09 Nick BHSPitMonkey_ is now known as BHSPitMonkey (
07:41:15 Join Moos [0] (
07:41:49 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:42:46 Quit josh_ (Remote closed the connection)
07:42:48 Join josh_ [0] (
07:43:28MoosMorning all !
07:51:55Bg3rmorning/evening :))
07:52:47 Quit needleboy ()
07:59:03Bg3rJdGordon : with reg win32 api ? :)
07:59:11Bg3rreg = registry
07:59:24Bg3rHKEY_CLASSESS_ROOT ...
07:59:52JdGordonthat doesnt let u choose to add a menu depending on the individual file tho does it?
08:00:07JdGordonlooks like im gonna havta learn atl quick..
08:00:20Bg3ri don't think u need ATL for this...
08:00:32 Join DSGASDGASDGHASDH [0] (
08:01:10JdGordon.. the only tutes ive found on it are all atl
08:04:08 Quit DSGASDGASDGHASDH (Client Quit)
08:09:02 Join amiconn_ [0] (
08:09:39 Quit TCK- (Read error: 104 (Connection reset by peer))
08:13:43***Saving seen data "./dancer.seen"
08:17:31 Quit amiconn (Nick collision from services.)
08:17:31 Nick amiconn_ is now known as amiconn (
08:20:19 Quit Sinbios (Success)
08:22:26 Join ender` [0] (i=ychat@
08:24:36JdGordondid any1 here try out my text editor?
08:24:44Bg3rJdGordon link ?
08:40:52 Quit JdGordon ("Miranda IM! Smaller, Faster, Easier.")
08:40:57 Join needleboy [0] (
08:41:20 Quit needleboy (Client Quit)
08:42:02 Join needleboy [0] (
08:53:49 Quit needleboy ()
08:55:33 Join needleboy [0] (
09:00:47 Join petur [0] (
09:01:49 Quit needleboy (Read error: 104 (Connection reset by peer))
09:02:25 Join needleboy [0] (
09:05:52 Quit needleboy (Client Quit)
09:06:37LinusNBg3r: t0mas said "off-list please" :-)
09:08:46*Bg3r stupid
09:09:22*petur wonders why the mail wasn't sent to the committers mailinglist instead
09:09:36t0mascan I send mail to that?
09:09:50*t0mas expected that one to be 'CVS' only
09:09:57t0masso I used the dev list...
09:10:28 Join needleboy [0] (
09:10:57 Quit petur ("CGI:IRC (EOF)")
09:10:59amiconnPaul_The_Nerd: I've fixed lcd_roll() on H1x0, even though it'll probably be scrapped soon
09:11:04 Quit needleboy (Client Quit)
09:11:07 Join petur [0] (
09:11:40 Join needleboy [0] (
09:14:40Paul_The_Nerdamiconn: I like that. Even when it's probably going to be thrown away, you make it work. Heh.
09:15:19amiconnWell, it's a 4 line fix
09:15:31amiconnWe just need to set up area scroll properly
09:16:07amiconnReading & understanding the datasheet was more work than the actual fix
09:16:13 Join damaki_ [0] (
09:16:48LinusNamiconn: that's often the case with lcd controllers
09:16:51amiconn..ut I'm like that: I want to know why things don't work as they should...
09:17:24amiconnLinusN: The 'spurious' OFF events on recorder v1 aren't all that spurious...
09:17:54amiconnPlease take a look at lower left corner
09:18:15amiconn(after rotating 90 degrees to landscape)
09:18:41amiconnU17B "presses" Off if the voltage drops below a certain level
09:18:41LinusNa zener
09:19:21LinusNyeah, a low-voltage protection
09:19:39LinusNthe one that turns it off when the voltage gets too low
09:19:45t0masan el cheapo brown out detector...
09:19:55LinusNel cheapo is the word
09:20:12t0masaren't those normally connected to the reset pin of the cpu?
09:20:26LinusNthank god it isn't in this case
09:20:36t0maskicks in to early?
09:20:45LinusNwell, not exactly
09:20:48amiconnSo... pulling high couldn't have worked of course.
09:23:01LinusNit's been a long time since i looked at those schematics :-)
09:23:34LinusNso, we have a short voltage drop when the drive spins up
09:23:44LinusNand we need to filter it somehow
09:24:10LinusNi guess a more aggressive filter on the OFF button should be enough
09:24:19 Quit needleboy ()
09:25:02 Join needleboy [0] (
09:27:46 Join safetydan [0] (n=dan@
09:30:14 Quit damaki__ (Read error: 110 (Connection timed out))
09:30:50markunneedleboy: do you still have this problem?
09:32:17 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
09:36:55 Join damaki__ [0] (
09:38:28nudelRe the GUI thread, I think it would be great to allow the user to add shortcuts (to folders (or locations in the tag database) as well as to items and places in the menus), and hide existing items from, the top level menu.
09:39:25nudeland I agree with Play (or some key) always going to the WPS... after using Rockbox for a few days I still find myself wondering how the f*** I get back to the WPS :-)
09:40:19nudelIt's always good to have a "take me home" key to push when you're lost or want to quickly go back to normal, or pause the track or whatever
09:40:36 Quit kernelsensei (Read error: 110 (Connection timed out))
09:41:14LinusNnudel: i agree about the "take me home" button, but what exactly do you find hard to understand about how to get to the wps?
09:41:38nudelsometimes play does take me to the wps but sometimes not
09:41:52LinusNit is so essy
09:41:57nudeli guess it's a distinction between file browser and menu, but i'm not used to that distinction because the list of files (tags) in other players *was* a menu
09:42:06 Quit tenzing (Read error: 110 (Connection timed out))
09:42:21LinusNthe browser is the "home"
09:42:23 Join linuxstb_ [0] (
09:42:48LinusNfrom there you get to the wps by playing a music file, or by pressing Play
09:43:08nudelpresumably to get to WPS from <anywhere> i keep pressing menu or left until it won't go any further, then press play
09:43:16LinusNin the wps you can open the menu, end you get back to the wps by leaving the menu
09:43:54LinusNand if you enter the menu from the browser, you get back to the browser when leaving the menu
09:44:10LinusNthe menu always returns to the place from where you opened it
09:44:22 Quit linuxstb (Read error: 104 (Connection reset by peer))
09:44:27LinusNit really shouldn't be that hard to understand
09:44:30nudelit makes sense after an explanation, but IMO it's an artificial "implementation detail" type distinction between the menus and the browser... i'm not sure many users will expect it, so it'll seem like sometimes the keys do one thing and sometimes another
09:45:09nudelhmm so the menu is sort of like a windows pop-up menu in terms of functionality?
09:45:59nudelthat makes sense... it would be totally obvious if it displayed as an overlay on top of the previous screen (like a pop-up menu) on targets with the screen space... of course that kind of GUI wishing is easier said than done :)
09:46:23LinusNyes, we are working in that direction
09:46:41 Join kernelsensei [0] (n=boris@gentoo/developer/kernelsensei)
09:47:04peturoh... with window open/close animation? :D
09:47:29nudelcool, that's great to know. maybe i can help once i get up to speed. I doubt I'll ever be able to help with ARM optimizing the MP3 codec or working out how bits of hardware work and that kind of stuff, but I'm sure I could help with GUI programming
09:49:38nudelunless I win the lottery and don't have to do my day job anymore, in which case the first thing i'll do is buy a book on ARM assembler :-)
09:50:07nudel(or give the money to apple on the condition they add gapless to their firmware and make it a standard feature in the DAP industry, for the love of god!)
09:50:34 Join c0utta [0] (
09:51:04 Join JdGordon [0] (
09:51:07 Quit damaki_ (Read error: 110 (Connection timed out))
09:51:23*Paul_The_Nerd always imagined the menu as a layer on top of things anyway.
09:51:42*LinusN too
09:51:51Paul_The_NerdThe problem is that a lot of people think of the WPS as home still. Though I don't really understand why there are people who want to see the WPS even when playback is halted.
09:52:12LinusNPaul_The_Nerd: that's beyond me too
09:52:51nudelso you can see the status of the playback... is it paused? did it stop at the end of the album? etc.
09:53:03LinusNyeah, if rockbox implements a "take-me-home" button, it would probably take you to the browser
09:53:13peturone word: GUI - people want to see eye candy the whole time, not a file browser...
09:53:34nudeland it's what you'r eused to seeing while listening to music... if i want to change albums i'm used to pressing some keys to get to the album list and selecting it. plus what petur said is absolutely true
09:53:41LinusNnudel: most people would hear that the music stopped
09:54:13Paul_The_Nerdnudel: As well, you only leave the WPS during playback if you stop the music, or if you press something to go to the filetree/menu. People have requested that stopping the music not leave the WPS.
09:55:02nudeli think it's a matter of personal preference rather than a wrong/right thing, to be honest... making the home button configurable makes everyone happy
09:55:15markunpetur: not all people (me for example)
09:56:06nudelPersonally, unless i'm *in the process of* selecting a file|menu, i don't want to see the file|menu list.
09:56:37nudelIt winds me up that on my Sony player to get back to the WPS I have to press the menu button twice then scroll to an item and select it. I've always wished it had a button I could just push to go there like the Karma has.
09:56:41JdGordoni agree /\ but unless im listening to the music i dont want to se the wps
09:59:07nudelA GUI which changes screens because the music stopped, rather than just displaying a Stop icon on the current screen/status bar (if any), doesn't seem right to me. I don't think the screen should ever change by itself without the user pushing a button.
09:59:41Paul_The_NerdI think the very name "While Playing Screen" gives away the whole screen's intention...
10:00:13nudelI don't think name should dictate function :-) Call it Playback Status Screen if you like. hehe
10:01:01Paul_The_NerdAt the moment it's more like "Name describes function"
10:01:08nudelFor me it's a bit like if Foobar2000 reached end of the last song in the playlist and then decided to open a file browser. Well, not that bad but kinda...
10:01:33Paul_The_Nerdnudel: It's like if foobar2000 closed itself at the end of a playlist, returning you to the OS.
10:01:47Paul_The_NerdIt wouldn't be opening anything, it would just be removing itself.
10:02:07nudelthat would be really annoying because i might want to continue using foobar
10:02:39 Quit JdGordon (Read error: 104 (Connection reset by peer))
10:02:58Paul_The_NerdYeah, but the big difference is, if you want to continue running rockbox, you just launch another song. It's still the same steps whether the WPS was up or not, it's just one less button press before you can browse.
10:03:04nudelI think to me the WPS *is* the player. Everything else is a menu I go into temporarily to do stuff.
10:04:06Paul_The_NerdAnd see, that's the difference. To me the WPS is information about what's playing. I mean, I could be playing a game instead. The WPS for all practical purpose is to me a popup letting me know what track is on, how far I am into it, and what's next.
10:04:08nudelTrue, but what if i want to skip back a track in the old playlist? Or just see waht the last track played was called? I guess I can press play to do that but I'd prefer not to have the screen change without me doing anything.
10:04:30nudelI guess we see things differently then. I see the WPS as the main screen and everything else as a popup. :)
10:04:34 Join JdGordon [0] (
10:04:57peturalso: when I switch on RB, I can press play to start playback, but I don't see what it will start playing until it plays.
10:05:13nudelIMO games are a bonus feature and shouldn't affect the music playback functionality
10:05:25Paul_The_Nerdnudel: They don't affect the music playback functionality.
10:06:03markunnudel: maybe you should write down you UI ideas
10:06:20nudelI plan to, but want to use RB as-is for a bit longer.
10:06:20Paul_The_Nerdnudel: The filetree is the "home" screen, because all the functionality sprouts from the filetree. You can't do anything until you either select a file, or resume the current playlist.
10:06:55nudelYou can resume the current playlist from the WPS
10:07:09peturPaul_The_Nerd: not exactly: the fact that radio and recording are only menu items isn't logical
10:07:14LinusNnudel: but i can't select a file
10:07:58peturone would think that the main usage scenarios are on the same level
10:07:59LinusNpetur: why isn't it logical?
10:08:15markunLinusN: you could if the filebrowser was a pop-up like the menu (not that I say it should be)
10:08:26peturbrowsing, playback, radio, recording, plugins
10:08:44nudelTrue, but I'm used to being at the WPS and if I want to select a file I open the menu (or browser in this case, but it's all just menus in my mind). If I want to open a file I'm used to, without thinking about it, pushing Menu, down, right and I'm in the file list... I don't want to think about whether or not music is currently playing and I'm on the WPS screen or the device, withotu me doing anything, changing screens
10:09:22peturmaybe use REC to toggle between 'modes'?
10:09:46peturmaybe put this on the devcon agenda? ;)
10:09:54Paul_The_NerdMaybe just don't have the WPS leave when you reach the end of a playlist.
10:10:09Paul_The_NerdWhen you hit stop though, it seems like it should definitely go back to the browser
10:10:21 Quit quobl (Remote closed the connection)
10:10:41nudelI don't have a stop button, only play/pause... And pausing definitely shouldn't go to the browser as it may be temporary.
10:10:46Paul_The_NerdI personally thing though that if no playlist is 'active' (if you aren't in play or pause state) then it doesn't make sense for the WPS to be available.
10:10:59Paul_The_Nerdnudel: Hold play/pause for about a second, and it stops.
10:11:05nudelah ok
10:11:59nudelPersonally, I'd prefer it if the screens only changed when I pushed a button (i.e. Menu), but I don't care very strongly what happens when Stop is pushed as I'll never do that. :-)
10:12:31nudelAnd if the WPS isn't the "home" screen but I can always quickly get tot he WPS by going "home" and then pressing play (or whatever), then that's fine as well.
10:12:37LinusNpersonally, i don't think it's a big deal to have the wps as the "main" screen, but at the same time i can't understand why the rockbox concept would be so hard to grasp for some people
10:13:05Paul_The_Nerdnudel: I agree the screen changing without user input is... questionable.
10:13:06nudelI'd prefer the WPS was home, or it's configurable or whatever, but I can live with pressing Play twice to get to the WPS, rather than just once.
10:13:44***Saving seen data "./dancer.seen"
10:14:03nudelSo long as there's some very simple action I can commit to muscle memory which will take me to the WPS from anywhere, I'm happy enough.
10:14:21Paul_The_NerdI still think it makes more sense for the file tree to be home, since that's the lowest level. The playback screen is dependent upon the choice of files, not the other way around...
10:14:22LinusNiirc, very early versions of rockbox didn't leave the wps
10:14:58LinusNand people felt it was so strange that it stayed, and didn't return to the browser
10:15:05LinusNso we changed it
10:15:33JdGordonumm... so why not just add the config optin for it?
10:15:51LinusNJdGordon: because it totally turns the rockbox internals upside down?
10:15:52nudelI see the WPS as a status screen... Mine will also display the clock and so on (once I install that patch :-)). If no file is selected it'll show a lot of blank space, which is fine as that's the current status.
10:16:24LinusNnudel: that's not how rockbox sees it
10:16:28JdGordonLinusN: why? wouldnt the wps just think music is paused/stopped? or is there omre to is?
10:18:02 Join Jungti1234 [0] (n=jungti12@
10:19:31LinusNJdGordon: well, maybe i misunderstood your statement
10:19:56LinusNi think making wps the main screen would affect a large portion of the rockbox code
10:20:07JdGordoni meant, just add an option to stay in the wps at the end of the playlist.. isnt that what ppl want?
10:20:17JdGordon..not make it the "home" screen
10:20:18nudelIs it a known thing that going into disc mode (to transfer some music) resets the theme when you come back to RB? Maybe it's the same issue with other settings like bass boost also not being saved (or being reset) on the 5G iPod. I saw someone mention that.
10:20:59LinusNi think the backdrop handling has a few flaws
10:21:14Paul_The_Nerdnudel: When it goes into disk mode, it doesn't shutdown cleanly, so you lose any settings that you've changed since the last proper shutdown or save.
10:21:34LinusNJdGordon: ah, i see, well that should be possible
10:21:48LinusNnudel: ah, you have an ipod
10:21:50nudelPaul_The_Nerd: ah okay, i'll shutdown properly in future
10:22:00nudelI do, but I only just bought it for rockbox
10:22:07nudelI don't *do* non-gapless players. :-)
10:22:27LinusNso what did you have before the ipod then?
10:22:30nudelfinally I have a 60gig gapless music player \o/
10:22:30Paul_The_Nerdnudel: All you really need to do is shutdown properly once to make sure the config block gets written. Then, as long as you don't change anything, the USB inserts shouldn't reset things, just reset them to whatever the last proper shutdown had set.
10:22:33 Join quobl [0] (n=quobl@tor/session/x-6745873937ad08ed)
10:22:58nudelrio karma, which was great but a bit thick and 20gig wasn't enough, then i got a sony 30gig which as a music player was great (apart from lack of analogue control for scrolling large lists) but sonicstage... argh
10:23:05Paul_The_NerdTomal was the one working on the iFP-7xx port right? Has he been seen lately?
10:23:38nudeli started writing a program to convert my FLAC albums to .iso images, mount them, then send button presses to the SonicStage window and monitor its list control to wait for them to rip, so i could get my music on to the sony gaplessly...
10:24:08LinusNoh my
10:24:09nudelbut now that rockbox is ported to ipod I figure it'll be a lot less effort to ditch that sonicstage project and put my spare time effort into RB instead
10:24:31nudelin the end my 30gig sony player had about 5gig of music because it was just too much hassle, haha
10:24:44nudelmy 60gig ipod is now full :)
10:25:23 Join Thus0 [0] (
10:26:02nudelBTW I wrote this page which I really must update to mention RB. Just don't have enough time to do all my spare-time projects *and* play Zuma on the xbox. :-)
10:26:23LinusNbtw, here's some reading regarding ui:
10:26:43Paul_The_NerdJust a thought: Could recording / radio be offloaded into plugins to reduce binary size?
10:26:50nudelthanks LinusN
10:27:21LinusNPaul_The_Nerd: that has been discussed, but i don't think it's a good idea
10:29:27Paul_The_NerdOn second though, I think I can see how it might have some disadvantages.
10:30:00LinusNregarding popup-style menus, we need a proper "viewport" gfx functionality, something we don't have today
10:30:23 Nick gromit` is now known as gromit_ (
10:30:30nudelI was thinking that while making my theme. Being able to say, draw this string within this rectangle using this font...
10:30:38 Nick gromit_ is now known as gromit` (
10:30:44 Join Kohlrabi [0] (
10:30:59LinusNwe are moving in this direction, it's just a matter of time (and hard work)
10:31:53nudelit's great to know the direction is there. and the current setup makes sense based on where the firmware came from (line based devices with small screens)
10:32:39nudelof course those devices still need to be supported... is the GUI currently abstracted enough from everything else to make that a clean switch? or is a lot of restructuring needed?
10:33:06LinusNi believe we will need quite some restructuring
10:35:51 Quit needleboy (Read error: 110 (Connection timed out))
10:39:02 Quit safetydan ("Leaving")
10:43:42linuxstb_lostlogic: Are you around?
10:43:49 Nick linuxstb_ is now known as linuxstb (
10:44:39 Quit quobl (Remote closed the connection)
10:44:44 Join needleboy [0] (
10:45:33 Join safetydan [0] (n=dan@
10:48:26 Join tucoz [0] (
10:48:29 Join damaki_ [0] (
10:48:39 Join _FireFly_ [0] (
10:50:03tucozhello. Do you mind if I commit a very minor fix to the ipodvol wps?
10:51:11tucozto fix the statusbar issue. (it should be visible in the browser)
10:51:33LinusNgo ahead, i don't mind
10:51:47LinusNi don't use ipodvol anyway :-)
10:52:27 Part Paul_The_Nerd
10:52:57tucozI actually like it
10:53:25LinusNmainly because it uses that lousy 7-bit chicago font
10:54:01tucozLinusN, that is true. I think someone should adapt it to use nedore, or whatever it is called
10:54:15LinusNyes, chicago12 should be exterminated
10:54:25LinusNthat someone could be you :-)
10:55:13LinusNhmm, hey to you too
10:55:22tucozYes, I know. I have to figure out how to use wps then. Never written a single letter of wps code though.
10:56:30Jungti1234Rockbox does make H300 dangerously?
10:56:51LinusNJungti1234: huh?
10:57:01peturyou got to love translation software
10:57:25Jungti1234H300 of two Korean Rockbox users burnt.
10:57:42Jungti1234battery and mainboard
10:57:57markunDon't think it's RB's fault, but who knows
10:58:18Jungti1234We disordered.
10:59:08LinusNJungti1234: are you sure that it was rockbox that destroyed the h300's?
10:59:26Jungti1234They used Rockbox.
10:59:58peturthey also used a h300
11:00:03youngcerea1oh yes in german we sad it's Bild a newspaper ;)
11:00:04LinusNi use windows, but i don't blame microsoft when my pc breaks
11:00:13Jungti1234I'm not sure. I'm asking possibility to you.
11:00:27LinusNit's not impossible
11:00:35youngcerea1mother kills kids and bild speaks first with the burgers
11:00:38LinusNbut i very much doubt it
11:00:46youngcerea1no chance
11:01:04youngcerea1only they have other battery put in
11:01:10markunLinusN: if windows would for example overclock your CPU you could blame MS
11:01:28LinusNwell, yes
11:01:50Jungti1234They are using iriver original battery.
11:02:24LinusNit's just so frustrating to see all there "rockbox killed my player" in various forums, when it turns out that it wasn't rockbox at all
11:02:26youngcerea1than it not realy do
11:02:37 Quit damaki__ (Read error: 110 (Connection timed out))
11:03:02youngcerea1yes i know the code and the chips there is no chance
11:03:06Jungti1234While one of them uses Rockbox's record function, it(mainboard) burnt.
11:03:26youngcerea1rockbox not iverwriten the conrol routins of pcf chips
11:04:03youngcerea1only what can kill is bad speackers and on long time the battery faster then the orginal firmware
11:04:33LinusNJungti1234: what part of the mainboard was broken?
11:05:28youngcerea1LinusN: the most people how say rockbox kill the player only have too many mony too think the proken the boot code and think its kill the hardware
11:05:28Jungti1234I don't know.
11:07:03youngcerea1meybe the use an hammer and a knif to put out the usb cable ;)
11:07:08Jungti1234iriver spoke so..
11:07:34LinusNof course iriver said that it is rockbox' fault
11:07:46tucozAnother weird thing about IpodVol is the fact the the volume is displayed as db with a percent sign after the value.
11:07:47LinusNthey would be stupid if they didn't
11:07:49Jungti1234'Because you used Rockbox, main board burnt!'
11:08:08LinusNtucoz: change that as well
11:08:11tucozI think I fix that wps to use nimbus12 and remove that % sign
11:08:17LinusNdo that
11:08:25 Join imphasing [0] (
11:08:37nudelI think there's a patch which changes the volume WPS code to a percentage instead of a db level, maybe it assumes the patch is installed.
11:08:56LinusNnudel: no, that wps was made before we changed it to db
11:08:57tucoznudel, but CVS wps's shouldn't rely on patches
11:08:59 Quit nudel ("At Argon, we're working to keep your money.")
11:09:03Jungti1234Another one.
11:09:09 Join nudel [0] (
11:09:17nudeloops closed wrong window :)
11:09:34nudelthe dangers of binding alt-f4 to a spare mouse button :)
11:09:53Jungti1234He was charging H300.
11:10:10Jungti1234oh nono
11:10:27Jungti1234:) I misunderstood.
11:13:19*amiconn wouldn't like the wps to become the main screen at all
11:14:03*amiconn also wouldn't like "browsing modes" the way the iriver firmware does it
11:14:13tucozamiconn, why? that is how iriver do it. It must be better ;-)
11:14:35amiconnnudel: You don't need the wps to check playback status. That's what the statusbar is for
11:15:13amiconnYou can even stop playback while being in the file browser (I'm assuming the button to do this is defined for iPod)
11:15:22nudeltrue. though there's no (last) song title in there. (though if the status bar was skinnable there coudl be :-))
11:16:14 Join KN|stiff [0] (
11:16:31nudelI think it sums up as I'm used to 3 different devices where the WPS screen is home and everything's launched from there, but in RB the WPS screen is more like an application launched from the browser.
11:16:50nudelSo long as there's a quick way I can get to the WPS I'm happy, either way...
11:17:30nudelThough it's weird not having the WPS available at times, I'm sure I'll get used to it.
11:18:12Jungti1234ok, I sayed to them that it may not be Rockbox's problem.
11:18:45linuxstbnudel: I don't know what DAPs you have used, but it's not standard for the WPS to be the "home screen". The ipod doesn't work like that for example.
11:20:44nudelyou're right, it's been a long time since i properly used the ipod firmware (years)
11:20:57nudelthe karma's like that
11:21:24nudeland the sony sort of is (but once in the menus it's a pain to get back tot he WPS, but it's always available and never goes away unless you press Menu)
11:25:30nudelgood day all, i'm off to work (late as always :))
11:30:18linuxstbBTW, I was reading about the itunesdb database format, and apparently there is a limit of 56 utf-16 characters (112 bytes) in the filename field - so maybe that is the reason why some people are having problems with the foo_pod plugin.
11:36:09 Join ashridah [0] (
11:40:07 Join quobl [0] (n=quobl@tor/session/x-3a004f3b6011e78e)
11:44:24tucozLinusN, I fixed the IpodVOL wps. But, it seems that I can't align the text to to the bottom of the screen.
11:45:22tucozIf I push the text one line further down, it will be cut by half by the screen border. But, it's better with nimbus-12 anyway.
11:45:37 Join damaki [0] (
11:49:11tucozI think I just commit this, and close as it seems that this patch is committed by lenox.
11:50:11 Quit needleboy ()
11:51:53 Join needleboy [0] (
11:57:33 Quit Moos ("Glory to Rockbox!")
11:59:00 Quit damaki_ (Read error: 110 (Connection timed out))
12:06:13markunneedleboy: are you here?
12:06:55markunneedleboy, Paprica: could you check if this bug still exists?
12:09:17 Join tvelocity [0] (
12:09:23 Join linuxstb_ [0] (
12:09:53 Quit linuxstb (Read error: 110 (Connection timed out))
12:12:14 Nick linuxstb_ is now known as linuxstb (
12:13:47***Saving seen data "./dancer.seen"
12:14:48 Quit petur ("time to say foo(d)")
12:15:17 Quit tvelocity ("Ex-Chat")
12:17:41 Join mantob [0] (
12:18:40mantobhi there, does someone know, what kind of hardware is located in the gmini xs 202 ?
12:22:07Bagderyes we do
12:22:13Bagdercheck the rockbox wiki
12:24:13 Join Paul_The_Nerd [0] (
12:28:13 Join Nico_P [0] (
12:31:18mantobbadger: sorry, i only found stuff about gmini 220 and xs 200, not xs 202
12:32:15Bagderhm, right
12:32:17Bagderthen no
12:32:34 Join Moos [0] (
12:32:46 Quit c0utta ("CGI:IRC")
12:33:42mantob:) thx
12:34:13Bagdergiven the model name, I'd guess it is similar
12:34:18Bagderbut of course we can't know for sure
12:34:28Bagderuntil someone figures it out
12:40:15 Join damaki_ [0] (
12:40:52 Join muesli__ [0] (n=muesli_t@
12:44:41 Quit ashridah ("Leaving")
12:46:27lostlogiclinuxstb: hey, I'm guessing that you're the guy to ask about the yield in a while loop in lcd_update −− how many times does that loop loop? How long is an individual loop iteration approximately?
12:51:37 Join petur [0] (
12:54:34 Quit damaki (Read error: 110 (Connection timed out))
12:57:28linuxstblostlogic: The reason that I constantly yield() is that the loop is just a delay to wait until the Broadcom chip has finished updating the LCD.
12:57:33 Join c0utta [0] (
12:57:47linuxstbBut yes, maybe yielding in every iteration isn't necessary.
12:58:17linuxstbBut if we don't yield, then we are just wasting CPU cycles - cycles that can be used for audio decoding.
13:02:04LinusNhow many cycles are we talking about?
13:02:18linuxstbThe total delay can be up to 14ms.
13:02:45linuxstbSo we can't perform two lcd_updates closer than 14ms apart.
13:02:46lostlogiclinuxstb: ahh, ok.
13:02:52amiconnlinuxstb: That'll be "fun" for oscilloscope...
13:03:20LinusNlinuxstb: maybe keep track of the time and only delay if the update is too early?
13:03:48lostlogicLinusN: it's a poll
13:03:54linuxstbThat's more or less what we do. The start of the lcd_update_rect() function polls the broadcom chip to see if it is done.
13:04:07LinusNaha, so it isn't a "blind" delay
13:05:15linuxstblostlogic: Also, looking at your changes to lcd-ipodvideo.c, was there a reason you moved the variable declarations from the start of the function to the inner-blocks? I thought the Rockbox style was to declare variables at the start of a function.
13:05:45 Join quobl_ [0] (n=quobl@tor/session/x-3ed2751219beae1e)
13:06:05amiconnRockbox style doesn't define that. It just says variables need to be declared at the top of blocks, i.e. no C99-style declarations
13:06:18lostlogiclinuxstb: I wasn't aware of that −− the only reason is that it helps me keep track of register usage without going to the assembly.
13:06:35linuxstbI know it's not in CONTRIBUTING, but I'm just going by the existing code.
13:07:04LinusNalso, the non-c99 recommendations aren't that critical anymore, now that all targets use fairly recent gcc versions
13:07:42LinusNbut i personally don't like in-code declarations anyway
13:07:46amiconnlinuxstb: Look at the grayscale lib or xlcd. I'm declaring many variables within inner blocks
13:07:51lostlogicI guess it's also because m68k-gcc isn't smart enough to optimize register usage for variables that aren't explicitly falling out of scope but aren't used any more in a function all the time −− arm appears to be
13:10:44lostlogiclinuxstb: I just realized however that there may be some more tweaking available to me in there that will involve moving some variables back out of blocks :-P
13:11:50Nico_Phi guys
13:12:03Nico_Phas someone had a look to my updated album art patch ?
13:12:05linuxstblostlogic: :) I don't feel strongly about it anyway - I don't mind either way. It's just that I hadn't seen it very much in Rockbox (I haven't looked at amiconn's libs).
13:13:39linuxstbNico_P: I haven't looked at it, but will try to. I'm hoping that you have only temporarily move your code out of the get_metadata() function...
13:14:28Nico_Pwell it doesn't look like i cana access the file buffer from there
13:14:51linuxstbNo - you will need to pass extra parameters to the get_metadata() function.
13:15:11Nico_Pand it works the same, as i load album art from the same function that calls get_metadata()
13:15:32Nico_Pit's done at the same moment
13:16:05Nico_Pwhy is it so important to have it in get_metadata() ?
13:17:08linuxstba) Because it comes under the category of getting metadata; b) To keep playback.c as small as possible; c) To enable support for reading cover art from id3v2 tags or similar.
13:17:49Nico_Pgood points
13:17:52linuxstbBut it's a minor point - it's more important to get the code working first, and then we can try and move the code back afterwards.
13:18:05Nico_Pbut the get_metadata() will have to be moved and/or changed
13:18:19 Quit Aditya|Sleep ("Trillian (")
13:18:57Nico_Put i think that's one of the topics that will be discussed at the devcon
13:19:15 Quit quobl (Remote closed the connection)
13:20:01Nico_Panyway it seems to work fine
13:20:13Nico_Pi expected to see some problems but they aren't coming
13:20:52muesli__XavierGr ?
13:24:58 Quit quobl_ (Remote closed the connection)
13:26:15 Quit KN|stiff (Read error: 110 (Connection timed out))
13:30:34lostlogicman, i2c on the ipod is still brutally slow −− the peak meter is very responsive, but gets paused for like 1/10 of every second while the RTC updates.
13:31:40linuxstbNico_P: I've just looked briefly at your patch, and there are two potential bugs (unless I've missed something). 1) The audio buffer is a ring-buffer, so what happens when you load an image at the buffer-wraparound point? e.g. if the current pointer to the free space in the audio buffer is 50KB from the end, but you try and load a 100KB image? 2) What happens if the album art is overwritten by a subsequent track?
13:32:03 Join mikearthur [0] (
13:32:47 Join quobl_ [0] (n=quobl@tor/session/x-88dcd456a6c94c61)
13:32:58linuxstbI also don't think it will be a problem to move it back to get_metadata - you just need to pass a pointer to the free space in the audio buffer (and a max size), and return an integer saying how much space you have used.
13:33:28 Join Rondom [0] (
13:34:12preglowlostlogic: someone really should ask portalplayer for some specs :>
13:34:48lostlogicpreglow: :-P
13:35:07Bg3r"see, guys, rb will raise the number of the sold iPods, thus will raise your incomings, too..." :P
13:35:33lostlogicBg3r: I guess people like me are the exception? I wouldn't buy an ipod until there was sound on IPV.
13:36:50linuxstbSupporting Apple's rumoured touch-screen ipod in Rockbox will be interesting...
13:38:07preglowi'm just clinging to the hope that it's just rumours, heh
13:38:27lostlogicI think I might switch the battery read down to a 8 bit ADC instead of 10, because 10 really isn't necessary and it's 1 byte less of i2c.
13:40:46Nico_Plinuxstb: i don't know for both of your questions
13:41:12Nico_Pand it crashes when the next song in playlist starts
13:41:20t0maspreglow, safetydan, Slasheri, petur, Bg3r, tucoz: Cloaks are done for you
13:41:32Nico_Pi hadn't noticed that because i didn't even listen to an entire song yesterday evening :p
13:41:32t0masI think you'll have to rejoin to see them in some clients
13:41:42t0masbut they alreay work in /whois [your_nick]
13:42:34tucozt0mas, thanks
13:42:36 Part tucoz ("Leaving")
13:42:49safetydant0mas, cheers
13:43:32 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
13:43:33preglowt0mas: cloak names are only five characters?
13:43:35Nico_Pgtg... i'll look at it a little more this evening
13:43:41t0maspreglow: hm?
13:43:44linuxstbNico_P: I'll try and find time to look at it, but one thing you will need to do is to copy the image somewhere else just before the track itself starts to get decoded.
13:43:45preglowahh, right
13:43:50t0masyours was wrong
13:43:54preglowi got a message from lilo saying my cloack name was pregw
13:43:55Nico_Plinuxstb: ok
13:43:57linuxstbNico_P: The WPS image buffer is probably the best place to copy it to.
13:44:02t0maspreglow: I already fixed that ;)
13:44:16t0mas[13:40:07] * [preglow] (n=thomjoha@rockbox/developer/pregw): Thom Johansen
13:44:17t0mas[13:41:23] * [preglow] (n=thomjoha@rockbox/developer/preglow): Thom Johansen
13:44:53Nico_Pthere's some things i corrected but i don't have time to upload the new patch... plus i only changed two lines and it doesn't add much stability so i'll do that later
13:45:15 Quit Nico_P ()
13:48:23*amiconn doesn't fully understand the purpose of the cloaks
13:48:48t0masit shows that you're part of the rockbox project...
13:48:55amiconnDo people care that much about the hostname?
13:49:01t0massome do
13:49:15amiconnIt's only shown when you log on, and that may be once per day or even less
13:49:16t0masand freenode wants to use it in the future to mark project participants
13:49:17 Join leftright [0] (
13:49:36t0masthey want a ~ in front of names of people who're not identified
13:49:38leftrightLostlogic, do you get pops/artifacts when skipping tracks
13:49:50t0masand they're thinking about adding some char (channel status?) to project participants
13:50:00t0masdo newbies asking questions know who to ask
13:52:03leftrightLostlogic, I have tried rockbox on two seperate players, no xfade, and i get pops/artifacts when skippping tracks and during powerup, and when playing from pause
13:53:03leftrightusing yesterdays daily build H140 ^^
13:53:06amiconnt0mas: How does this hostmask thing work, i.e. how does the server know I am Me when I log on?
13:53:14t0masnickserv identify
13:53:37amiconnDoesn't identify happen after joining the channel?
13:53:43t0masdepends on the client
13:53:47t0masmine does it before...
13:55:03amiconnAh, yes.
13:55:17amiconnI could do both in HydraIRC, rather flexible
13:55:39t0masyeah, I've used HydraIRC on windows...
13:55:43t0masbut it was a beta iirc
13:55:58amiconnSo, it's just another setting that is associated with the identify info...
13:56:21lostlogicleftright: on ipod?
13:56:50leftrightiHP140 and H140
13:57:01safetydanleftright, do you have the EQ enabled?
13:57:16safetydanTry disabling it and see if the pops go away
13:57:30lostlogicpops on powerup are 'normal' for both iRiver and rockbox firmware, IIRC
13:58:05safetydanCurrently the EQ history is not reset on track skip so it may cause pops
13:58:24safetydanpowerup pops are "normal" as lostlogic says
14:00:42 Nick Bg3r is now known as Bger (n=Bager@
14:01:09leftrightLostlogic, safetydan, I have disabled the EQ and it the artifacts when skipping tracks are still evident
14:02:03leftrighton both headphones and optical out
14:02:40 Nick Bger is now known as Bg3r (n=Bager@rockbox/developer/Bger)
14:02:59 Join perldiver [0] (
14:03:32 Quit DJ_Dooms_Day (Read error: 104 (Connection reset by peer))
14:04:14 Join DJ_Dooms_Day [0] (
14:05:23safetydanleftright, that's good (for me anyway as it means I don't have to do anything in the EQ code)
14:06:40leftrightheh, you're off the hook then :-)
14:09:29*safetydan bats the popping audio ball back to lostlogic and Slasheri
14:12:42*leftright makes sure it hits them on the head
14:13:48***Saving seen data "./dancer.seen"
14:14:39Slasheriinteresting, i haven't had that issue yet.. but i will try with fresh cvs soon
14:15:04leftrightHi there slasheri
14:15:16XavierGrI can confirm popping on audio when I change tracks too.
14:15:22leftrighthow's the army treating you ?
14:15:34XavierGrIt was fixed at some time, but then it appeared again.
14:15:40Slasherileftright: hi :)
14:15:55lostlogicIs there anything particular onthe ipod that runs every second (other than the RTC update)?
14:16:09leftrightare they making you run up and down hills for fun
14:16:10muesli__XavierGr so tell me how many strophes has greeks hymn!?
14:16:12Slasherileftright: well, it's fine in general but i am quite busy with everything right now
14:16:24lostlogicwatching the audio thread there is a definite "buffer grow for 1s, slip, grow 1s, slip" thrend
14:16:32XavierGrmuesli: many!!!!
14:16:45XavierGr100 something I think, let me check
14:16:58muesli__they said 158 on tv yday!
14:17:44XavierGryup they are right I just checked
14:18:06muesli__how comes? nothin to do=?
14:18:20 Quit DJ_Dooms_Day ("Trillian (")
14:18:21XavierGrbut only the first 4-6 are mainly known
14:20:12XavierGrthe national anthem is a poem written back on 1821. Then it got a melody which is used for all the verses, but of course not all verses are played/sang.
14:20:52XavierGrmost people know the first 2.
14:21:27XavierGrit would be just crazy to know all of it....
14:21:50XavierGrand meanigless IMHO
14:23:21Bg3rXavierGr what about the Iliad etc ...
14:25:49 Quit Paul_The_Nerd ("Leaving.")
14:26:26XavierGrWell that is a great epic poem, it doesn't have strict verses. :p
14:27:09leftrightthanks folks
14:27:13 Part leftright
14:27:53Bg3ri mean that Homer has known it by heart (at least that's what people tell...)
14:30:27XavierGrThe Iliad comprises roughly 16,000 lines of verse, I remember when I was doing this at school, of course we didn't cover it all.
14:31:10Bg3rhehe i remember this too..
14:33:04XavierGryou learn Homer too? I wasn't sure that about it, though I had heard of it.
14:34:22Bg3ryes, of course ..
14:34:48muesli__homer? dooh! ;-)
14:36:51 Nick Mikaelh is now known as Mikachu (
14:37:44 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:43:37lostlogicI think that (grumble) by (grumble) further throttling the amount of I2C per second vorbis can be made back to mostly realtime on IPV (grumble)
14:44:28lostlogicit also look like charge detection by USB works so we can show the cute lil' charging icon.
14:46:10Jungti1234bye all
14:46:27 Quit Jungti1234 ()
14:46:59amiconnlostlogic: Wouldn't be the correct solution to get i2c up to speed?
14:47:31 Nick Rondom is now known as Rondom_ (
14:47:44lostlogicamiconn: yes, I agree completely, but without specs, I'm unsure of how to proceed in that area :-\
14:48:19 Nick Rondom_ is now known as LINK (
14:48:53 Nick LINK is now known as Rondom (
14:49:01 Part Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
14:49:41lostlogicI'm guessing that if the PP I2C timings were known, we could do away with polling loops and use either sleeps or short timing loops to time the I2C transactions, but how can we figure that out w/o specs?
14:54:45 Quit c0utta ("CGI:IRC (EOF)")
15:00:15 Join Rondom_ [0] (
15:00:20 Part youngcerea1
15:01:48 Quit needleboy ()
15:02:14 Nick Rondom_ is now known as Rondom (
15:07:10 Quit Moos (Read error: 104 (Connection reset by peer))
15:07:52 Join KN|stiff [0] (
15:10:04 Join Moos [0] (
15:16:37Papricahave you try rockcalendar?
15:16:52 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
15:21:04 Join youngcereal [0] (
15:22:00 Join midkay [0] (n=midkay@
15:22:14amiconnlostlogic: Experimentation...
15:24:49 Join Kohlrabi [0] (
15:26:54 Quit linuxstb (Read error: 110 (Connection timed out))
15:30:01lostlogicamiconn: hmm, yes, I suppose I could make some brash timing guesses and binary searchish for values that work. but not today.
15:30:35 Part youngcereal
15:32:36 Join youngcereal [0] (
15:42:08LinusNi normally use a logic analyzer
15:42:20LinusNto verify the timing
15:42:25 Quit quobl_ (Remote closed the connection)
15:42:39preglowi bet you also are one of the only ones in here right now that's got a logic analyzer
15:42:48preglowthe ones i've used cost a ton
15:44:54LinusNthis one was roughly $1200
15:45:32LinusNwhich could be "a ton" to a lot of people
15:45:47LinusNbut not my employer :-)
15:54:57lostlogicI'd also have to take apart my ipod which I'm not prepared to do...
15:56:35lostlogicIt looks like i2c read multiple of 4 bytes at a time should be entirely doable, but without knowing what command to send to the i2c control register I'm not sure what to start with.
16:07:49 Quit XavierGr (Read error: 110 (Connection timed out))
16:09:18tucozIs there an up-to-date wiki-page on the config-file format?
16:09:34 Join gracchus [0] (
16:13:51***Saving seen data "./dancer.seen"
16:19:07preglowwhat's with people and this anonymous thing
16:19:17peturwas that the last mail of gl ?
16:20:15LinusNpreglow: i don't know
16:20:55pregloware they afraid someone are going to dig up their address and kill them or something?
16:21:28preglowwe won't accept people who are not willing to die for rockbox, and that's our official standing!
16:21:30 Join SereR0kR [0] (
16:21:37LinusNand we are "arrogant and single minded" according to him
16:22:08LinusNsorry, "closed minded"
16:22:13 Join adiamas [0] (n=adiamas@
16:24:21safetydanWell I'll be. I didn't think anyone had problems giving their name for the cause.
16:24:30preglowwe've had some
16:24:36preglowi just don't see the problem
16:24:40LinusNme neither
16:25:27safetydanThe only time I can see it being a problem is when it would cause a problem for Rockbox, e.g. MP3 player manufacturer employee writing Rockbox code on the side not wanting their name in the credits
16:25:30 Join XavierGr [0] (
16:26:01peturmaybe they fear tying their name to an open source project is bad for their image? Or like safetydan says: doesn't want employer to know
16:26:49XavierGrso contibuting code under a nickname isn't accepted?
16:26:56LinusNperhaps, but from my experience it's more like some kind of principle about being anonymous on the net
16:27:03safetydanI'm sure it's been debated to death before so I'll say no more :)
16:27:15peturhe could have given any fake name
16:27:30gracchusquick question aside... to record on an ipod, I would probably have to modify the FIQ handler to service both read and write IIS FIFOs?
16:28:28gracchusgetting orientated so pardon if the question is 'off'
16:28:38*preglow sees the demise of his asm fiq handler coming
16:28:53gracchusit's real nice so I didn't want to hurt it!
16:29:43XavierGrand who is gl exactly? Does he come in IRC?
16:30:01preglowsometimes, i think
16:30:40LinusNXavierGr: well, gl calls himself "gl.tter" in the rockbox forum
16:31:47peturand probably some anonymous email address...
16:32:57XavierGrI don't know if this was discussed before, but he has a point, why not accept a nickname? The contribution is that matters...
16:33:45Moossame with wiki account: exemple 2 hours ago
16:34:18markunLinusN: did you ask why he wanted to remain anonymous?
16:34:47XavierGrnow that you said it: Can someone delete my previous wiki account "AlexSpy"? I made a new one with my full name...
16:35:17Moosthere are a lot of Wiki account with nick names only
16:35:41MoosXavierGr: LinusN is your man
16:36:53XavierGrLinus could you delete it? (it is embarassing too. I mean AlexSpy? What was I thinking :D)
16:38:37Mooshehe :)
16:39:13 Join linuxstb [0] (
16:40:34 Quit youngcereal (Read error: 110 (Connection timed out))
16:41:11 Join youngcereal [0] (
16:41:17 Join Siku [0] (
16:43:21tucozRegarding the peakmeter settings. Do we still use percent there?
16:43:30tucozor is it plain db these days?
16:43:35t0masLinusN had some trouble with the name thing again?
16:44:26*t0mas still doesn't understand
16:44:41t0masworking with real people... on a real project... why not use real names and real e-mailadresses?
16:44:44preglowpeak meter does both
16:44:58LinusNt0mas: we do
16:45:02preglowdb is default, but i always set it to linear
16:45:13t0masLinusN: yeah, I didn't get his problem ;)
16:45:33LinusNah, now i understand what you mean
16:45:46 Join TCK- [0] (
16:45:57LinusNwell, some people have this thing about privacy
16:46:10t0masI have that too...
16:46:29peturbah, just find out who registered and
16:46:37t0masnever give my real name to registrations etc... and you don't have my phone number (I don't like the commercial calls)
16:46:41t0masbut my name... come on...
16:47:22 Join bluey [0] (
16:48:17tucozpreglow, ok. Thanks
16:48:59tucozI didn't know if we still used a scale other than dB (overall).
16:50:00XavierGrt0mas: Still, if he doesn't wnat to give it out, it is his right... I don't understand why would someone want to hide his real name, but even then I respect his choice.
16:50:38LinusNwe respect his choice
16:50:49XavierGrMy question is why Rockbox doesn't allow nicknames under a contribution
16:51:02t0maswell... I think it's primarly a legal issue...
16:51:20 Join needleboy [0] (
16:51:27LinusNyes, it's partly a legal issue
16:51:46 Quit needleboy (Client Quit)
16:52:11XavierGrif it is a legal issue then I pass...
16:52:13 Join needleboy [0] (
16:52:59t0masI'm not sure what the real problem is (legally)
16:53:26t0masbut I think we need to be able to tell others who exactly did something, especially regarding patent issues...
16:54:38tucozBut I must say it's also a nice policy. Like Bagder said in the mail, real contributions by real people.
16:55:02LinusNgotta go, cu
16:55:17t0masLinusN, Bagder and Zagor are responsible for the project
16:55:26t0masif we ever get patent trouble etc
16:55:44t0masit's not a registered company or something... legally it's just them using code send to them by others
16:55:53 Join IcyStorM [0] (
16:56:00 Part LinusN
16:56:48 Quit needleboy (Client Quit)
17:01:16 Join needleboy [0] (
17:01:35 Join Rondom_ [0] (
17:02:51 Quit petur ("work")
17:04:40 Quit needleboy (Client Quit)
17:05:11IcyStorMWhere can I get the new faster LCD drivers for iPL running the iPod 5G
17:05:55t0masI would say, ask in #ipodlinux
17:05:58t0masbut you already did
17:06:06t0masso wait for someone there to answere...
17:06:11IcyStorMHahahaa :)
17:06:36 Quit TCK- (Read error: 110 (Connection timed out))
17:07:19preglowtucoz: building the manual now, gotta say it looks sweet
17:07:21 Quit Rondom (Nick collision from services.)
17:07:26 Nick Rondom_ is now known as Rondom (
17:07:40tucozpreglow, nice. But, it's a long way to go :)
17:08:25preglowdon't think the chapter headings should be on a page of their own, though
17:08:27tucozpreglow, autogenerated latex code is not the most pleasant thing to work with
17:08:57tucoztrue, that is the fancy header stuff in the preamble. We can remove those later on.
17:09:56tucozpreglow, Ah, I see. I know what you mean. I think it was made like that in th 2.4 docs.
17:10:53 Quit bluey ("Leaving")
17:11:02preglowthe fancyheader stuff is actually quite nice, if you ask me
17:11:10preglowit's just the blank page on chapter start i don't like
17:11:25preglowmost manuals, unless you count the huge ones, don't do that
17:11:27tucozyes, I know what you mean
17:11:36preglowthen again, this might end up pretty huge as well...
17:11:41preglowstill, i like it
17:11:42preglowit looks nice
17:11:53preglowwhat font is it?
17:12:00t0masis there a pdf export function for the manual?
17:12:09 Join asraniel [0] (
17:12:17tucozI defined the font with \renewcommand{\familydefault}{\sfdefault}
17:12:30tucozt0mas, if you run make, you get a pdf
17:12:34t0masah ok
17:12:52tucozI guess it's some standard sans serif font.
17:12:54amiconntucoz: I get several 'image ignored' lines. Is that expected?
17:13:15tucozamiconn,hmm. What target do you build for?
17:13:20amiconnrecorder v1
17:13:29tucozor do you mean in the actual pdf?
17:13:40amiconnIn the pdf
17:13:47preglowi do tha tas well
17:13:49preglowfor h1x0
17:13:59tucozOk, that is expected. Those are images not includeded.
17:14:10preglowtucoz: so, you're still pretty much working on the writer->latex source?
17:14:28tucozAnd that line serves as a reminder to what needs to be added.
17:14:48preglowperhaps i'll pitch in a bit, i miss working with typography
17:14:55 Join bluey [0] (
17:15:03tucozpreglow, yes. But, I normally remove that code and just keep the section layout
17:15:45tucozJust using the writer2latex source as a template.
17:16:10*amiconn highly appreeciates the documentation effort
17:17:01preglowme too
17:17:07preglowgood documentation is a good thing
17:17:12amiconnWriting documentation is a tedious task...
17:18:49tucozbut with this approach, the docs will hopefully keep theirselves up to date by just committing the docs to new features along with the feature.
17:18:51preglowyes, but i do find it entartaining at times
17:19:59*amiconn usually reads documentations, unlike many others.
17:20:08tucozLike, simply writing up a small doc for a plugin and commit it along with the plugins.
17:20:29amiconnBut I'm usually too lazy to write docs
17:20:54tucozI noticed when I made screenshots to some of the old plugins that a lot of those need to be adapted properly to the larger screen sizes.
17:20:59 Quit Zagor ("Client exiting")
17:21:29amiconntucoz: Indeed. But many devs prefer working on new ones
17:22:12tucozso I noticed. I got tempted to start working on some of those, but I think I should get the manual at least somewhat up to par with the wiki first.
17:22:17amiconnNew features ar usually more interesting, but this way we'll never get H1x0 and newer targets into release state
17:22:26safetydanSeems like one of the easier ways to get in to development is to make the old plugins work
17:22:27DBUGEnqueued KICK Paprica
17:22:50Papricai Attach a bitmap on the wrong place
17:23:08safetydanI was happy to see that poking at oscilloscope triggered some lcd driver work
17:23:30tucozsafetydan, yes. Totally. And, probably quite rewarding to make rockbox look better and not that half baked on the newer targets.
17:23:49Papricalol what an idiot
17:23:53amiconnsafetydan: My current plan is to rewrite oscilloscope to not use a timer, but do some more intelligent frame rate adaption, then port it to swcodec
17:24:13amiconnAfter that, my idea is to combine both oscillo* plugins into one, offering both aspects
17:24:25safetydanthat sounds good to me
17:24:26amiconn..and finally scrap lcd_roll()
17:24:37safetydanI understand the original reason for two plugins, but it will be nicer to combine them
17:24:50safetydanAnd we can now say Rockbox has visualisations! :)
17:25:03amiconnIt already had, for quite some time
17:25:06CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:25:06*safetydan waits for the requests to put plugin output in to the WPS...
17:25:12safetydanamiconn, for swcodec I mean
17:25:25amiconnsafetydan: There's also vu_meter waiting to be adapted...
17:26:53amiconnsafetydan: This scrolling stuff is something I was planning to do for I don't remember how long
17:26:55safetydanamiconn, well it works, but scaling it up was something for another time
17:27:27amiconnIirc these were the only functions missing for completion of the GraphicsAPI list
17:29:25amiconnRegarding visualisations in wps... the scrolling routines always scroll full screen
17:30:02amiconnPartial scrolling would be really hairy on lcd depths < 8
17:30:35preglowtucoz: seem's we're using cmr sans
17:31:51preglowtucoz: btw, any reason at all we're using a sans font for our main font?
17:33:20safetydangotta go
17:33:21 Quit safetydan ("Leaving")
17:42:12tucoztucoz, Not really. Just tried to reflect what the 2.4 docs made
17:42:47Mikachutalking to yourself?
17:43:09tucozpreglow, Not really. Just tried to reflect what the 2.4 docs made
17:43:27 Quit KN|stiff (Read error: 110 (Connection timed out))
17:48:09tucozpreglow, in some ways the serif font looks better. But, it does have that LaTeX feel to it. Wonder if we could find some other nice font
17:49:01preglowthe 2.4 manual font looked like crap, let's not duplicate that
17:50:15tucozI just noticed that I the font is defined twice in the preamble.tex. If you comment out both of those lines (line 4 and the last line) you get a serif font.
17:52:24tucozMaybe we should use that instead. I think it looks prettier
17:56:21 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
18:01:37preglowknow, done it already
18:01:39preglowtried a couple
18:01:49preglowand unluckily, times does seem to be what looks the best
18:02:10tucozIs that the standard font?
18:02:26preglowcmr is the standard
18:02:27preglowgod, i hate cmr
18:02:45tucozHow do I define times?
18:03:29preglowtry \usepackage{times}, \usepackage{charter}, \usepackage{palatino}, etc
18:03:40 Join damaki [0] (
18:03:47preglowdifferent fonts in latex is a bloody nightmare
18:03:56preglowbut i don't think we'll be able to use anything but the default bundled ones anyway
18:04:07preglowfonts costing money and such
18:05:03tucozwell, sure. But, if Times looks ok then we could use that further on.
18:05:42tucozI am fixing the tables now, to use booktabs instead of the standard crap
18:07:06amiconnI think sans serif looks better for headlines
18:08:08tucozThat we could fix later on. We could wrap those inside a sans serif environment. But, I think we should use Times as the main font anyway.
18:13:52***Saving seen data "./dancer.seen"
18:16:49 Join damaki__ [0] (
18:18:40 Quit damaki_ (Read error: 110 (Connection timed out))
18:23:31 Join Genre9mp3 [0] (
18:25:48 Quit _FireFly_ ("Leaving")
18:27:39 Quit Thus0 ("Leaving")
18:31:06 Quit damaki (Read error: 110 (Connection timed out))
18:34:47Mikachuwhat does the circle with the dot in the upper right mean?
18:35:26tucozzero ?
18:35:45Mikachuto the right of the clock, it only appeared briefly now, never noticed it before
18:45:58 Join quobl [0] (i=id@tor/session/x-b480e10edafdaca5)
18:57:51 Quit adiamas ("Chatzilla 0.9.71 [Firefox]")
18:58:35ender`Mikachu: i think you're looking at the disk activity indicator
19:08:19 Join damaki_ [0] (
19:18:15 Join damaki [0] (
19:22:23 Quit damaki__ (Read error: 110 (Connection timed out))
19:22:59 Join damaki__ [0] (
19:24:34Mikachuender`: oh, okay
19:24:38Mikachumaybe it's new for the nano
19:24:40 Join KN|stiff [0] (
19:25:09 Join coolst [0] (
19:25:55 Quit bluey ("Leaving")
19:27:09coolstWill Rockboy work on my ipod video? i dont want to test it on my ipod so i tried it in simulator and it looks like rockboy is not ported to this resolution. will that be different when i use it on my ipod?
19:27:28 Join bluey [0] (
19:28:45 Join molest [0] (
19:28:45 Quit coolst (Client Quit)
19:31:35 Quit damaki_ (Read error: 110 (Connection timed out))
19:37:03 Nick mantob is now known as mantob|away (
19:37:07 Quit damaki (Read error: 110 (Connection timed out))
19:37:10 Part mantob|away
19:48:39 Quit Kohlrabi (Nick collision from services.)
19:48:49 Join Kohlrabi [0] (
19:49:10 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
19:49:52XavierGrsafetydan: what happened with the copy paste patch?
19:51:50safetydanXavierGr, got held up by debate about code size increase for the core
19:52:00safetydanmight be better as a file manager plugin
19:52:04XavierGris it so much?
19:53:09safetydanIIRC it was 1200 bytes for coldfire, 1400 for arm and 1600 for SH
19:53:13 Part IcyStorM
19:53:22safetydannot much for the iPod/iRiver but a reasonable amount for archos
19:54:55Moosmaybe this feature justify this litle increasment
19:55:03*Moos vote for it :)
19:55:08amiconnsafetydan: Did you exclude the retry-with-copy if move fails for non-multivolume targets?
19:55:19amiconnIirc the version I tried didn't have that
19:55:28 Quit perldiver (Read error: 110 (Connection timed out))
19:55:43safetydanamiconn, no I haven't looked at the code size since I added that exclusion
19:56:40amiconnarchos could profit from that. The only multivolume archoses are the Ondios, and they don't suffer from the codesize problem as much as the recorders
19:56:53XavierGrsafetydan: I assume that you use the audio buffer if there is no playback right?
19:57:22amiconnNo, that's not necessary
19:57:24Mikachuamiconn: if you are mr oscillograph, did you notice the text printing when changing speed looks really weird?
19:57:31safetydanXavierGr, plugin buffer only, larger buffer doesn't help that much
19:58:09XavierGrah okay
19:58:23XavierGrand what about archos that has32 kb in plugin buffer?
20:00:56amiconnsafetydan: It seems the code in flyspray isn't the latest?
20:02:14amiconnOh, and another hint: You should round down the buffer to a multiple of 512 bytes. Using a few extra bytes may slow down copying as it ends up handling partial sectors
20:03:35safetydanamiconn, no it's not the latest
20:03:43safetydanbyte size for coldfire is now 1100
20:04:01safetydanand I'll look in to rounding the buffer size as well
20:05:23amiconnbuffersize &= ~0x1ff;
20:05:33safetydanah, well that saves me thinking about it then :)
20:10:23 Join ep0ch [0] (
20:11:37 Quit quobl ("Leaving")
20:13:54***Saving seen data "./dancer.seen"
20:13:58tucozamiconn, do you know how to make screenshots in the player sim?
20:14:37tucozf5 didn't work. (cool that the cube works for the player btw)
20:14:39amiconnYou can't...
20:14:48 Join Nico_P [0] (
20:14:48solexxHi! I am trying to compile the current latex manual from cvs
20:15:05tucozphew. Well, then someone else will have to make those
20:15:07Mikachuif you're in windows you can probably press alt+printscreen and paste in paint?
20:15:08 Quit KN|stiff (Read error: 110 (Connection timed out))
20:15:08solexxbut I don't know anything about makefiles and such stuff
20:15:13tucozsolexx, does it work?
20:15:13amiconnAtm the only way is making screenshots with methods provided by the OS
20:15:24solexxtucoz: I have no idea :)
20:15:35solexxI have checked out rockbox-all
20:15:36tucozsolexx, you make a directory called build-man for instance
20:15:46tucozand then you run ../tools/configure
20:16:02solexxah, thx!
20:16:17tucozselect your player, manual build, etc and run make
20:16:21solexxdoes the target platform matter for the manual?
20:16:42tucozsolexx, the manual is very much work in progress, but that is the idea
20:17:26tucozSo yes, if you select h300 you get a manual with h300 specific details and screenshots (although lot's of info and screenshots are missing)
20:17:43safetydando the other sim targets support screenshots?
20:17:50tucozsafetydan, yes
20:17:51solexxthanks a lot. latex is complaining about undefined references but I have a 67 page pdf file
20:18:23tucozsolexx, that is expected. We (I) haven't looked at the references yet.
20:18:38solexxmaybe I will find some time to fix some things...
20:18:42safetydanodd, I wonder what's different about the player sim
20:18:45solexx...after my exams next week
20:18:49amiconntucoz: Perhaps we could add the screenshot capability to the sim code for player, as (1) the sim code holds the image as a bitmap and (2) we're focusing on the SDL sim now
20:19:11*solexx thinks there's just too much cool stuff around to do
20:19:17amiconnFor the bitmap targets, the screenshot code in the sims is the same as on the targets
20:19:22tucozamiconn, that sounds like a good idea. It is a tedious task to make screenshots anyway.
20:20:01safetydanah, that makes sense
20:20:24tucozsolexx, If you would like to start with something, I would suggest taking a look at one of the now included plugins and follow that approach to add the not-yet-added plugins to the docs
20:20:40tucozfor instance bejeweled
20:21:38solexxi'll keep that in mind
20:23:44tucozI read that nils (or what is nick is) was working on something with the docs. I would like to know what that is, so that I do not duplicate his work.
20:27:02 Join xmixahlx [0] (n=xmixahlx@
20:28:30 Quit damaki__ (Remote closed the connection)
20:29:27 Join damaki [0] (
20:30:06 Quit bluey ("Leaving")
20:32:32tucozAnother issue I've been thinking of is what default font we should use (and wps). I think the current default is a bit too tiny for the larger screens.
20:33:39solexxtucoz: agree
20:33:54 Join petur [0] (i=petur@rockbox/developer/petur)
20:34:11solexxi am using xtal-14 since it supports iso-8859-1
20:34:16tucozAlthough, imho, there should be a unified look of Rockbox on all targets. However, I think the default Rockbox font and wps is not really a show-off on the video ;-)
20:34:44amiconnI agree about the wps.
20:35:11amiconn(although I would be very hesitant to make a bitmapped wps the default)
20:35:33tucozAnd as we add screenshots to the manual, we need to use a default setting.
20:35:53amiconnThe default font also needs some thoughts. It should be a fixed width font
20:36:20 Quit damaki (Remote closed the connection)
20:36:51amiconnIn order to make it approximately the same *physical* size as on archos, it would need to be an 11px font on H1x0, and 14px on H300
20:37:22tucozAre there any good looking fixed width fonts around?
20:37:39amiconnDon't know about the iPods; the important value is the dpi of the lcd
20:38:09 Join _asraniel [0] (
20:38:11Mikachuthe nano's display is Very Small [tm]
20:38:25solexxtucoz: xtal-14 is fixed
20:39:00tucozok, but we would need a font that looks the same on all targets if I have understood this correctlu
20:39:13amiconnsolexx: xtal-14 is proportional
20:39:14tucozor similar
20:40:08tucozI think the default Rockbox UI should at least look similar on all bitmap targets
20:40:18 Join damaki [0] (
20:40:41Mikachui think the current default blue background and tiny text is a bit boring
20:41:18amiconntucoz: rockbox_default is originally called clR6x8, check fonts/rockbox_default.bdf
20:41:31amiconnDoes that exist in other sizes?
20:41:44 Join kclaf [0] (
20:41:55tucozI have no idea :-)
20:43:36tucozamiconn, looks like it
20:44:33tucozI can select other sizes in fontsel at least
20:47:10 Quit damaki (Remote closed the connection)
20:50:57 Join damaki [0] (
20:51:47 Quit asraniel (Connection timed out)
20:53:21 Join thegeek_ [0] (
20:55:15 Quit thegeek (Read error: 113 (No route to host))
20:55:42 Join Rondom [0] (
21:02:27molestanybody knows if rockboy works on ipod 5g?
21:04:02Nico_Plinuxstb: are you here ?
21:05:20safetydanactually I was wrong, it's 1336 bytes for cut/copy/paste on coldfire (with all changes)
21:05:31safetydanI should probably update the patch
21:07:26preglowmolest: NO
21:07:30preglowcapslock, yes
21:08:54 Quit Genre9mp3 ()
21:09:45 Join linuxstb_ [0] (
21:10:47linuxstb_Rockboy does (sort of) run on the 5g. It doesn't work in full-screen mode, and like all ipods, the button mapping is inadequate.
21:11:14Mikachusafetydan: just one more byte
21:12:01amiconn;-) That's not possible, instructions are word sized...
21:12:41peturamiconn: I'm gonna trigger clipping on the first max value for swcodecs, no matter what you say. I'm listening to a recent recording and there's much more clipping than I got warned about during recording. Better to be on the sensitive side.
21:13:07Nico_Plinuxstb: you told me it would be better to copy the album art to the wps image buffer
21:13:32Nico_Pbut then do i need to load them in the audio buffer ?
21:13:43Nico_Pi'm getting a little confused
21:14:22linuxstb_I'm saying that as soon as a track starts to play, the buffer space used by the album art is freed. This means that if Rockbox fills the buffer with a new track, the album art will be overwritten.
21:14:53linuxstb_So the album art image for the current track (and only the current track) needs to be copied elsewhere.
21:15:04Nico_Pthat does happen... at least i think so because just before the end of the track, the album art is replaced by random garbage
21:15:08linuxstb_And I suggested the wps image buffer as a suitable place to copy it to.
21:15:33 Join bluey [0] (
21:15:50linuxstb_The alternative is to have a buffer specifically for album art - which people who don't use album art may not like.
21:16:05Nico_Pusing the wps image buffer sounds like a good idea
21:16:43Nico_Pcurrently i load the wps in audio_load_track(), almost at the same time as get_metadata()
21:17:17Nico_Pin fact, just after get_metadata() and just before the audio data is buffered
21:18:08Nico_P"as soon as a track starts to play, the buffer space used by the album art is freed" −−-> what function does that ?
21:18:58 Join Rondom_ [0] (
21:23:44linuxstb_Nico_P: I don't know. Maybe lostlogic or Slasheri can point you in the right direction.
21:29:42Nico_Pok then i'm going to look for it, they don't seem to be here atm
21:31:08Nico_Palso ther's some thing i notice
21:31:24Nico_Pwith a big bitmap, the audio is delayed
21:32:25 Quit Rondom_ (Nick collision from services.)
21:33:28preglownot so strange
21:33:41Nico_Pwhy's that ?
21:34:44petureh.. don't break gapless...
21:34:56Nico_Ppetur: i'll try ;)
21:35:41Nico_Pthere's no reason it should be broken... acutal album art loading happens when the track is buffered
21:40:22safetydanIs the %acc0 register 64-bit?
21:40:27safetydanon the emac I mean
21:44:55safetydanah, okay, 48-bit
21:45:18 Join Aditya|Sleep [0] (
21:46:21 Join Paul_The_Nerd [0] (
21:46:31linuxstb_Nico_P: Which audio formats have you been testing with?
21:47:06 Join actionshrimp [0] (
21:48:30kclafhey guys
21:48:35kclafif i would like to boot to apple firmware by default, would it be better to edit bootloader module and compile my own
21:48:41Nico_Plinuxstb: only MP3
21:48:51kclafor use ipodlinux bootloaderv2 ?
21:49:21Paul_The_Nerdkclaf: Either way works, though loader2 might be easier
21:49:41kclafok thanks
21:50:16kclafmy purpose is to be able to use ipod with alpine car audio still
21:50:31kclafi fear it might not like rockox boot
21:55:08linuxstb_Nico_P: Looking at your patch, you don't appear to be adjusting buf_ridx to skip the album art in the buffer.
21:56:10Nico_Pi don't
21:56:11tucozAha, top reveals some interesting info. I wondered why the fan on my laptop starts cooling down my computer ever so often, and look what I found. 'rockboxui' using 99.6% of the CPU. naughty process.
21:56:41 Join KN|stiff [0] (
21:56:58linuxstb_Nico_P: You may want to look at the audio_update_trackinfo() function, plus the places in playback.c where that function is called.
21:58:30safetydantucoz, it should only do that when you're playing sound
21:59:26tucozsafetydan, probably. But the window was long gone. Only the process was remaining :)
22:00:13tucozOne of the simulators I started today crashed somehow and I forgot to see if the process was still running :-)
22:00:25 Join Spida [0] (
22:01:38SpidaI want to put rockbox on my iriver h340
22:01:44Spidaanything I should read?
22:02:03tucozyes. there is a page in the wiki. hold on.
22:02:18SpidaI have found some information, but none of it seems complete...
22:03:48 Join solexx_ [0] (
22:04:16tucozIn simple terms, download original fw, patch using fwpatcher.exe, copy patched hex to iriver, run fw-upgrade from iriver fw, download a from and unzip that to the root of your iriver. And then your done
22:05:56safetydanpreglow, is it just me or do you not use the d4 register in coldfire eq?
22:09:53Nico_Plinuxstb: adjusting buf_ridx seems to cut down (if not eliminate) the delay on play
22:10:24Nico_Pand i suppose you meant audio_update_trackinfo() would be a good place to copy the album art over to the wps image buffer... ?
22:11:39Nico_Pbut it's not called for the first played track
22:13:55***Saving seen data "./dancer.seen"
22:14:48 Quit Moos (Read error: 110 (Connection timed out))
22:18:18*petur takes so much time to write a (big) answer in the forum that he's logged out before posting and his post is lost :(
22:19:08kclafUnable to get setup.ini from <>
22:19:11 Quit solexx (Read error: 113 (No route to host))
22:19:12 Part Paul_The_Nerd
22:19:22kclafis that normal ?
22:19:43kclafwhen i want to add rockbox as cygwin mirror
22:20:52kclafan1 could try it please? i tried like 10 times
22:21:32*lostlogic beats portalplayer over the head with a manifesto on information sharing
22:26:37 Quit gracchus (Read error: 110 (Connection timed out))
22:27:38*petur hopes lostlogic has hit them hard enough
22:30:06safetydanbah, every attempt I make to port the eq filter routines to C results in lovely, ear ruining, white noise
22:33:06 Quit molest ("CGI:IRC (EOF)")
22:33:35preglowahh, right
22:33:40preglowyeah, i've tried that as well :-)
22:34:09preglowthough, if the arm optimised one works, then so should a c version
22:34:39safetydanI'm not familiar with arm syntax, so I started with the coldfire one since I at least vaguely understand m68k
22:35:01preglowi could do a fast attempt at having it work
22:35:06preglowat least i know how it should work
22:35:45 Join Daishi [0] (
22:36:02lostlogichmph, never tried to go _from_ ASM to C, interesting task.
22:36:03safetydanYes it's pretty much a straight copy of the assembler
22:36:39preglowerror 1, acc should be signed, not unsigned
22:37:40preglowand i think there's some serious shifting you're not doing here, that needs doing
22:37:58preglowremember that the coldfire emac routines always return the top bits of the result, not the bottom ones
22:38:10safetydanah, now that's a pretty fundamental thing to miss
22:38:20safetydanshould probably have read up more on the emac
22:38:50safetydanso every add involving acc0 I need to discard the top 16 bits
22:39:52preglowmno, i'd just multiply and accumulate straight forward, then at the last step, shift the entire things 'shift' bits to the left, then save the top 32 bits of the result
22:39:58lostlogicbut the C won't even use the EMAC
22:40:14preglowlostlogic: no, it'll have to use a long long accumulator, like the arm version does
22:40:44lostlogicright, so talking about what the EMAC does effecting his C code... doesn't make sense
22:40:45preglowsafetydan: and btw, i'm not convinced the c compiler does a proper 64 bit multiply there
22:41:13lostlogicprobably have to cast to a long long on one of the vars first
22:43:11safetydanwell, now I get silence instead of white noise
22:43:13 Quit KN|stiff (Read error: 104 (Connection reset by peer))
22:43:17safetydanI guess that's an improvement :)
22:44:59preglowi'm doing a quick recode myself right now
22:49:42preglowi don't get sound now
22:49:45 Quit imphasing (Read error: 104 (Connection reset by peer))
22:49:52preglowdoes the sim still have 64 bit issues?
22:49:55 Join jocke1s [0] (
22:50:25safetydanpreglow, I thought amiconn had found them all, but it's possible he didn't
22:50:31 Part ep0ch ("Kopete 0.11.1 :")
22:50:34 Quit SereR0kR ("XChat Aqua")
22:50:42 Part jocke1s
22:51:05 Join jocke1s [0] (
22:51:09amiconnpreglow: I'm quite sure it has. There are still 10 warnings for swcodec, and I don't get it to play music
22:51:26preglowme neither
22:51:32preglowthen i can't test my routine
22:51:34jocke1shi, anyone got experience with getting an ipod video mounting properly on linux (prreferably ubuntu)
22:51:35amiconnThat might be a problem with my installation though.
22:51:42preglowit crashes on codec load here
22:51:46amiconnI don't know what method sdl uses to play sound
22:51:59preglowWe open the real file 'archos/.rockbox/codecs/mpa.codec'
22:51:59preglowFatal signal: Segmentation Fault (SDL Parachute Deployed)
22:51:59preglowXlib: unexpected async reply (sequence 0x63)!
22:51:59DBUGEnqueued KICK preglow
22:51:59preglow*** glibc detected *** corrupted double-linked list: 0x00002aaaab09b278 ***
22:52:00safetydanamiconn, by default I think it's OSS
22:52:20amiconnThis is one of the drawbacks of linux, too many different methods for the same thing...
22:52:23Nico_Pi've got to establish a link between the playback code and the wps code
22:52:27Nico_Phow can i do that
22:52:30amiconnOSS, ALSA, ESD, ...
22:52:48safetydanpreglow, I've seen that happen if audio play back fails
22:52:48preglowthe situation in windows is much better, directsound, mme, asio...
22:53:12amiconnAt least the wavemapper *always* works
22:53:28preglowheh, that depends on the sound card
22:53:35 Join DrumRBoy320 [0] (
22:53:36preglowit didn't on the fancy rig i used in trondheim
22:53:42lostlogicjocke1s: I have IPV and I use it in linux just fine
22:53:42preglowbecause of sample rate issues
22:54:16DrumRBoy320is it possible to tripple boot rockbox, iPL, and retail f/wares on the 5g 60 gig ipod?
22:54:49preglowwell, oss works
22:54:54preglowso it's something else
22:55:02preglowDrumRBoy320: yes
22:55:21safetydanpreglow, if you attach gdb to it and do a back trace I expect to see it crashing somewhere in the pcmbuf code
22:55:21jocke1slostlogic: what linux U use?
22:55:31 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
22:55:39DrumRBoy320do, im assuming you need a custo loader.bin in the iPL installer, is that availble somewhee?
22:55:49lostlogicjocke1s: it doesn't matter what distro −− it's a USB Mass Storage device, the only thing that possibly matters is what kernel you are using
22:55:56lostlogicI'm running 2.6.16-rc5-mm3 currently.
22:55:57amiconnpreglow: (1) How would I tell SDL to use a different method than OSS? (2) Is there some SDL sound test code?
22:55:57preglowsafetydan: mad_frame_init...
22:56:31preglowit's possible our codecs aren't 64 bit safe
22:56:52jocke1slostlogic: ok I am running gnome at the moment and it tries to automount. But it doesn't work very stable. Any hints on how to connect?
22:57:11preglowit works
22:57:13preglowit's libmad
22:57:21preglowtremor works, but DAMN, it's close to noise
22:57:27Mikachuamiconn: i think there's a SDL_AUDIO_OUTPUT variable you can set, not sure about the name
22:57:27preglowand it's two times as fast as it should be
22:57:47Bagder64 bits is two times the bits you should have ;-)
22:57:51amiconnpreglow: I I tried ogg back then. It didn't play either..
22:57:53CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:57:53*Bagder ducks
22:57:58safetydanamiconn, you export an SDL_AUDIODRIVER environment variable before running ./rockboxui
22:58:18safetydanthe values are 'artsd', 'esd', 'alsa', 'oss', and 'dma'
22:58:20safetydanI think
22:58:24Mikachuthat's the one
22:58:30amiconnI'll try.
22:59:00lostlogicjocke1s: well for one, don't use automount if you're having trouble
22:59:04lostlogicjocke1s: define 'stable'
22:59:36lostlogicjocke1s: when I insert the device, I get device nodes /dev/sda, /dev/sda1 and /dev/sda2 sda2 is the ipod's main partition containing all data, sda1 is the boot partition
22:59:40preglowalsa didn't work
22:59:53preglowjust got silence
23:00:00 Join Rondom_ [0] (
23:00:14amiconnpreglow: libmad isn't 64bit safe??
23:00:31preglowamiconn: doesn't exactly seem like it, but i should think so
23:00:43jocke1slostlogic: I have no idea how to disable automount.
23:00:43jocke1sSometimes it shows up on the desktop and under /media/ipod (/dev/sde2 I think). But writing to it doesn't allways work. And when it sometimes apperantly does work I am not sure If the changes are actually performed on the Ipod itself
23:00:53preglowcompiling libmad is pretty fast
23:00:57preglowcould try out
23:00:58safetydanpreglow, do the peak meters move in the WPS?
23:01:24amiconnCompilation of the whole sim is pretty fast on my amd64...
23:01:28lostlogicjocke1s: do you unmount it before removing it?
23:01:57safetydanalso, do either of you get a message about "Unable to open audio driver" on stderr?
23:01:58jocke1sI can not umount it. It says its busy
23:02:22preglowsafetydan: not with ALSA
23:02:24Mikachuyou should eject in addition to umounting
23:02:29preglowsafetydan: with oss it works, but output is garbled
23:02:45lostlogicjocke1s: well maybe it's busy and you shouldn't remove it until the pending writes are done... make sure you have no windows open accessing the device, and try running a sync before unplugging it even if you can't safely eject it.
23:03:05amiconnI can't get it to play with either of oss, alsa , esd
23:03:10lostlogicand removing it without unmounting it is probably what's causing the trouble, because then the device nodes have a hard time getting uncreated sometimes
23:03:29amiconnPlayback seems to start and counts up to 0:03 (silent), then stops
23:03:53amiconnThe sim doesn't hang though, Stop brings me back to the browser
23:04:00preglowamiconn: it does that here as well if i set SDL_AUDIODRIVER
23:04:06preglowamiconn: i had to unset it to make it work again
23:04:21jocke1slostlogic. Sounds fine. What do you mean by "sync"
23:04:31Nico_Phow can i acces the gwps data from playback.c ?
23:04:40lostlogicjocke1s: the linux command sync −− writes all pending data to devices
23:04:58*safetydan somehow feels that saying "sound works for me" isn't going to help
23:05:08lostlogicNico_P: that's backward, gwps can access some playback data, but playback should not access gui data
23:05:17amiconnpreglow: Same effect. (if I didn't unset SDL_AUDIODRIVER the wrong way)
23:05:24 Join asraniel_ [0] (
23:05:35preglowuse 'unset'
23:05:38 Quit Bagder ("Off to search for that connect-resetting peer guy!")
23:05:48preglowjust setting it to "" doesn't work
23:05:49amiconnI used 'export' to set it?
23:05:51jocke1slostlogic: k
23:05:51jocke1sI'll try it again, thanks
23:06:04Nico_Pand can i call a function which is in the wps code from playback.c ? or is it the same ?
23:06:30 Part jocke1s
23:06:30Nico_Pit's to copy the album art bitmap over to the wps image buffer
23:08:17 Join imphasing [0] (
23:09:13 Quit linuxstb (Read error: 110 (Connection timed out))
23:09:29 Join linuxstb [0] (
23:10:28Nico_Pi'm going to bed... i'll look at that later when i'm not sleepy and i have some time
23:10:36Nico_Pgood night
23:10:47 Quit Nico_P ()
23:12:27preglowit's a bit hard to test if 64 bit multiplication works properly on a 64 bit machine..
23:12:47 Quit imphasing ("Reboot..")
23:13:41 Join JdGordon [0] (
23:14:25 Join ashridah [0] (
23:14:49 Join RotAtoR [0] (
23:15:05safetydansuddenly the arm version of the eq routine makes sense... if only I'd actually read the comment saying r10 and r11 are the 64-bit result
23:15:16 Quit Rondom (Nick collision from services.)
23:15:18 Nick Rondom_ is now known as Rondom (
23:15:24preglowmy attempt
23:15:40pregloweven though i can't test it
23:16:39safetydanwell I get some warnings about pointer sizes
23:17:13preglowon mine?
23:17:38preglowpointer size of all things
23:17:47safetydanon yours yes
23:17:51preglowwhich lines?
23:17:52safetydanbut it seems to work
23:18:03safetydaneq.c:229: warning: cast from pointer to integer of different size
23:18:04safetydaneq.c:230: warning: cast from pointer to integer of different size
23:18:04DBUGEnqueued KICK safetydan
23:18:04safetydaneq.c:231: warning: cast from pointer to integer of different size
23:18:04***Alert Mode level 1
23:18:04safetydaneq.c:232: warning: cast from pointer to integer of different size
23:18:11 Join TCK [0] (
23:18:16 Quit _asraniel (Connection timed out)
23:18:28preglowi don't get it...
23:18:40safetydanno, neither did I
23:18:51preglowperhaps it thinks i'm trying to cast the 'f' ...
23:19:05preglowstuff some parantheses around the entire deal
23:19:18preglowaround f->history[lol][hehe] part
23:19:34Mikachu(long)f->foo would cast the f i bet
23:19:43 Join imphasing [0] (
23:19:44DrumRBoy320hmm, is repartitioning the drive going to make my iPod un-returnable?
23:19:57preglowDrumRBoy320: what do you mean?
23:20:12 Nick StrathAFK is now known as Strath (
23:20:17DrumRBoy320well, i want to install IPL (which seams to allow for rockbox also)
23:20:26DrumRBoy320and it says you must repartition
23:20:29vcaany modification of firmware technically voids the warranty afaik
23:20:40JdGordonsomeone elignten me.. what is the argument on the rb dev ml?? the guy doesnt want his name assosiaed with rb?? wtf?
23:20:46preglowyou should assume that any tampering with the ipod will make apple throw a tantrum
23:20:59DrumRBoy320well, yea, but i know rockbox is completely changable, its only one command to replace the standard bootloader
23:21:13preglowJdGordon: he wants to remain anonynmous
23:21:25preglowJdGordon: and rockbox doesn't want to deal with anonymous people
23:21:27safetydanpreglow, ... ah you already saw that
23:21:39preglowsafetydan: saw what?
23:21:46preglowahh, yeah
23:21:46safetydanisn't history 2d?
23:21:50JdGordondamn wierdo! i like seeing my name on the web :D good for the ego!
23:21:51preglowit is
23:22:18preglowi made a glitch there, yes
23:22:22preglowfix that :>
23:22:24amiconnpreglow: Sound is working now. Turned out that my system doesn't like ESD running
23:22:27preglowgcc didn't even squeek here
23:22:41preglowwhat _is_ esd?
23:22:46CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:22:46*amiconn wonders why on some systems ESD improves things while on others it does the opposite
23:22:57amiconnEnlightened Sound Daemon
23:23:01Mikachuesd is a software mixing daemon
23:23:06preglowwhy the hell do we need another sound daemon?
23:23:07Mikachuif your sound card only allows one client
23:23:14preglowpeople should just use bloody alsa
23:23:15Mikachui do believe esd was one of the first ones
23:23:16safetydanwhich seems unnecessary if alsa is working for you
23:23:25preglowbetter yet
23:23:29preglowpeople should just use jack
23:23:43amiconnNoisy, and double-speed
23:23:51preglowamiconn: exactly the same as here
23:24:12safetydanpreglow, seems to work, but I think the extra processing load causes buffer underruns for the sound layer in the sim and it flakes out
23:24:20amiconnNow if I only could get samba to work...
23:24:29amiconnWould make things way easier for me
23:25:27preglowsafetydan: it's not that much more processing...
23:25:39safetydanpreglow, it definitely works
23:25:46preglowthen hooray for me
23:26:00safetydanActually I think the extra processing comes from me holding up on the gain button thus updating the coefficients constantly
23:26:14preglowstill, this is a bloody pc we're talking about
23:26:16 Join ark_112 [0] (
23:26:18preglowit should friggin' work
23:26:23ark_112hey all
23:26:44ark_112could comeone tell me if they could send me a compiled version of the bootloader for 3g ipods? plz
23:26:45preglowsafetydan: does it sound the same?
23:27:01safetydanpreglow, I'll check
23:27:46preglowi don't think 3g rockbox is quite ready for use
23:27:57ark_112i just wanna test it
23:27:59 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
23:28:05***Alert Mode OFF
23:28:11safetydanpreglow, pretty close or identical in a quick check
23:28:24preglowark_112: it's on the cvs build page
23:28:27preglowark_112: just get one there
23:28:28safetydan60 hz LS +10db and the clipping sounds around the same
23:28:35preglowsafetydan: should be identical
23:28:44preglowit basically does what the arm version does
23:30:04preglowwell, anywho
23:30:08preglowgo ahead and commit it
23:31:50preglowit sure as hell doesn't work on amd64, though
23:32:00preglowi bet the eq coef calc is off
23:32:05 Join nls [0] (
23:32:15*lostlogic pats himself on the back for sticking to a 32bit system on his athlon64 for now
23:32:15ark_112but i have to build it dont i?
23:33:04preglowark_112: well, no, not if you don't want to
23:33:12preglowark_112: the cvs build page has precompiled daily binaries for ipod 3g
23:33:26preglowark_112: if you truly want to build it yourself, then you are most certainly welcome to do so
23:33:27ark_112o ok, wats the address for that?, click cvs builds on the left
23:34:41ark_112thanx for your help guys :)
23:35:27nlstucoz: I was adding the images from the 2.4 manual back to the beginning of chapter 2 but couldnt find sutable ones for the newer targets
23:35:47tucoznls, ok. Nice.
23:36:02tucoznls, what targets are that? ipodcolor and nano?
23:36:18nlswell and the irivers :)
23:36:26tucozoh, of course :)
23:36:47tucozare that pictures _of_ the targets your thinking of?
23:37:16nlsyep the "Your Jukebox" (or whatever) in chap 2
23:37:58nlsthe pics in the simulator already have labels on them
23:37:58tucozHope I didn't break anything for you with my changes today
23:38:11ark_112the build doesnt include the bootloader
23:38:14nlsno not really
23:38:53tucozI see, maybe you could just add some image from google images for completeness and we could change the actual image later on
23:39:05tucozor is that illegal perhaps?
23:39:12amiconnpreglow: Without ESD sound works in SDL, and also in Rhythmbox - but only one of them at a time
23:39:18ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:39:25amiconnIf I start both, the second one doesn't play
23:39:45nlsmaybe I could check cvs history to see who added the pics
23:39:50preglowamiconn: what sound card?
23:39:55nlsto the sim
23:39:55tucoznls, so you miss images for the h300, h120/h140, ipod color and the nano?
23:40:15amiconnpreglow: Chipset integrated sound (ATI XS200)
23:40:22ark_112ipod 3g boootloader
23:40:23preglowahh, no idea, then
23:40:24tucozOk. I know there has been posted images earlier on. I go look in the wiki
23:40:32preglowa lot of those onboard sound cards are really shoddy
23:40:48nlsand the recorder v1 and v2 share image but maybe thats ok?
23:41:02tucozI am pretty sure I've seen high-res images of those players.
23:41:19 Join damaki_ [0] (
23:41:33ark_112does someone hav a compiled version of a 3g ipod bootloader they could email to me?
23:41:44 Join webguest06 [0] (
23:41:46safetydanpreglow, I'll look at comitting tomorrow, must sleep
23:41:48 Part safetydan ("Leaving")
23:41:55 Quit imphasing ("Kernels!")
23:42:11ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:42:21webguest06tucoz: has for iriver h1x0
23:42:26 Quit webguest06 (Client Quit)
23:42:28tucoznls, I wouldn't know of that. I think it's ok as a starter at least. If the actual units differ (in terms of how they look) we could always change that later on
23:42:43ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:42:45 Quit mikearthur (Remote closed the connection)
23:43:07 Join mikearthur [0] (
23:43:10 Join Henrico [0] (
23:43:25tucoznls, yes. Check out that page. Looks like quite ok pictures of most of the targets
23:44:11tucozgrrr, except for the ones you were looking for
23:44:28ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:44:29ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:44:30DBUGEnqueued KICK ark_112
23:44:30ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:44:31***Alert Mode level 1
23:44:31ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:44:33***Alert Mode level 2
23:44:33ark_112does someone hav a compiled version of a 3g bootloader they could email to me?
23:44:51youngcerealyour mail
23:45:16tucozark_112, patience is a nice thing
23:45:17 Join herz42 [0] (
23:45:28ark_112sorry :( just no one was answering
23:45:50youngcerealno prob
23:45:57youngcerealsek please
23:46:06youngcerealmy baby is not the fastest
23:46:09preglowark_112: i have just told you it's on the cvs build page
23:46:12youngcereal;) but i love it
23:46:24nlsIt's got one for h1xx at least
23:46:30preglowor can't you get it?
23:46:48ark_112i downloaded the build but i doesnt include the bootloader
23:46:51nlsany preference on wich one we should use
23:46:58preglowwhy aren't the bootloaders for download?
23:47:12ark_112i dunno
23:47:28ark_112i cant see them
23:47:34preglowno, they're not there
23:47:40preglowi just figured they'd be since they're build
23:47:45nlsgooglesearch for images on
23:48:24ark_112o ok, they say u hav to build it ur self if u want it but i was hoping someone could email it to me
23:48:45tucoznls, not really. I think the main focus is (at this stage) to get the manual somewhat up to par with the current state of rockbox. That is, get the text right (key-presses, get rid of autogenerated crap etc)
23:50:10nlstook the h140 one 'cause it was on
23:50:14nlsin the oic
23:50:35nlsOk so keypress labels then...
23:50:36youngcerealark_112: i have sent
23:50:40CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:50:40*nls ducks
23:50:49tucozAh, labels also
23:50:55BHSPitLappynls: aren't you the guy that was working on video in iPL?
23:51:20nlsI'm gonna put 'em on there but what should I call 'em
23:51:29nlsBHSPitLappy: no
23:51:37nlsmirak i think?
23:51:47 Quit ark_112 ("CGI:IRC (EOF)")
23:52:13tucoznls, hmm. As it isn't a screenshot, simply call it what it is. Like ipodcolor-front.png or something like that?
23:53:02tucozthen we could refer to it by the \playertype macro
23:53:45tucozor maybe not that macro, but something else.
23:54:04nlsok, and what to call the buttons like ON versus Play
23:54:20nlsor the juystick select
23:54:34***Alert Mode OFF
23:54:56tucozThat is another issue. I wonder if we should use the terms used in the source. Like SELECT etc.
23:55:14 Join imphasing [0] (
23:55:24preglowtucoz: you agree with me that we can ditch all the \newpage commands spread around chapter beginnings?
23:55:31tucozpreglow, yes
23:55:38tucozI ditched some of those
23:55:42tucozforgot about plugins
23:55:56preglowahh, that's the one i'm looking at, yes :-)
23:56:15nlsbut on h300 should we go with NAVI or SELECT
23:56:43XavierGrcool! Now I have a way to mod pcb materials and smd componetnts
23:57:12 Quit damaki (Read error: 110 (Connection timed out))
23:57:32tucozIn my opinion, if the button is labeld NAVI, I think we should go for NAVI but otherwise I am not sure
23:57:58tucozAs the recorder etc, use F1 and so on
23:58:49 Join webguest39 [0] (
23:58:55nlsI dare not change the old labels...

Previous day | Next day