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

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

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

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

#rockbox log for 2005-04-14

00:00:35rasheror maybe I'm totally missing something
00:00:39rasherwouldn't surprise me
00:02:26*rasher tentatively agrees not to break any United States laws on the rockbox forum
00:07:12rasheroh.. no, nevermind
00:08:43*rasher thinks
00:12:57rasheroh damn, this was entirely unsuccessful
00:19:19rasheralso, shouldn't the id3db be read when usb is unplugged?
00:21:49preglowwell, yes
00:22:24rasherthought so
00:22:28rasherthis doesn't appear to happen
00:22:53rasherat least, the tag browser gets VERY confused
00:23:18rasherhrm, someone in the wiki claiming RPSL is gpl compatible
00:23:25rasherI wish he'd get on irc
00:23:35rasher*waves* are you reading the logs?
00:24:12rasherand he forgot to unlock the topic ><
00:25:11 Join _aLEX [0] (
00:26:05 Quit _aLF (Read error: 54 (Connection reset by peer))
00:27:13preglowhelix stuff?
00:27:48rasherit definately is not gpl compatible
00:28:31rasherthey claim it's a "compatible" license
00:28:41rasherbut that's just to confuse matters I think
00:29:12rasherit may be compatible from the RPSL's point of view, but not the other way round
00:29:36rasherand they even have a note to that effect
00:31:05preglowwell, what do they have that we can use?
00:31:43preglowi'm hard pressed to come up with codecs we haven't covered or can easily cover
00:32:05rasherthey have an fixed aac
00:32:10preglowmaad2 has that
00:32:13preglowfaad2 <-
00:32:20preglowand it's gpl
00:32:24rasherI guess we don't care then
00:32:36preglowi've even got it semi-working with rockbox
00:32:46preglowbut stopped when i understood i already have my hands full
00:32:55preglowi'm too easily distracted, that's one of my key problems
00:33:09preglowi fiddle with a million things when i should be finishing mad optimisations
00:34:36rasher What kind of comment is "it is tried and tested - Motorola use it in their phones" anyway?!
00:34:42rasheroh, maybe he meant the codec
00:35:45 Quit _aLEX (Read error: 60 (Operation timed out))
00:37:30rasherhow did imdct_l fare?
00:37:36preglowi'm still working on it
00:37:42preglowplaying register allocator is not my forté
00:38:03rasherah, thought you had a working version back when you announced you were going to kill yourself
00:38:12preglowbesides, it's about 800 lines of arm assembler, don't expect it to be done in a jiffy
00:38:27preglowoh, no, far from it
00:38:46preglowi'm still spending time on doing stuff that should have been trivial
00:38:57preglowbut aren't, thanks to motorolas mac unit scheme
00:39:02rasherno pressure, just misunderstood you back then
00:39:29preglowno, that was just me ranting BEFORE i went to work, heh
00:39:42rasherah :()
00:40:39preglowto invert an accumulator
00:40:43preglow"a = -a" style
00:40:50preglowi have to write something around ten lines of assembler
00:41:40rasherthat's... rich
00:42:11preglowno, more like fifteen
00:45:34preglownow, that's if i want full precision, luckily, it's quite a bit less if i'm willing to sacrifice eight bits of precision
00:46:11preglowthree lines :-)
00:46:20preglowi also like talking too much, tell me to shut up
00:46:38rasherHrm, you'll have to imagine that
00:51:22*preglow pats HCl's back and murmurs "dct32... dct32..."
00:54:40*HCl hands preglow coffee
00:54:49HClganbatte yo
00:55:12preglowbaka yo
00:55:53HClwatashi wa baka da nai :/
00:56:10HCli told you to do your best :p
00:57:09preglowi know, ganbare and those guys ares one of the few words i can decipher the meaning of
00:57:29preglowi can't inflect them properly, but i get the gist of it, heh
00:57:41preglowinflect _it_
00:59:12 Join ashridah [0] (
01:01:01 Join DangerousDan [0] (
01:01:03 Join kramerica [0] (
01:02:50preglowmy brain hates playing register allocator
01:03:06 Join DMJC [0] (
01:15:17preglowDMJC: you got any idea why the debug sample playback code changes pitch with pll frequency?
01:15:17 Quit kramerica (Read error: 131 (Connection reset by peer))
01:17:02 Quit Sucka ("a bird in the bush is worth two in your house")
01:22:01*DMJC != soudn engineer
01:22:21preglowno, but you created the playback code :P
01:22:50preglowor am i too tired to think straight?
01:22:51preglowi am
01:23:31preglowi actually think i'll just go and remedy that
01:23:35preglowlater, all
01:23:45 Quit preglow ("fop")
01:24:33DMJCyou're too tired
01:25:45 Quit matsl ("Leaving")
01:26:18 Quit rasher ("CGI:IRC 0.5.4 (2004/01/29)")
01:34:10HClpreglow, with xshocks patch to get rockboy sound, we broke it so that it would only work at 120mhz
01:34:18HClthen linus fixed it up a bit
01:34:22HClso he'll prolly know about it
01:35:30 Join bagawk [0] (~Lee@bagawk.user)
01:35:35***Saving seen data "./dancer.seen"
01:36:34 Quit bagawk (Client Quit)
01:41:12 Join Bippy [0] (
01:48:52 Quit Bippy ("CGI:IRC (EOF)")
01:55:27 Nick tvelocity is now known as tvelocity[away] (
02:16:40 Quit ashridah ("Leaving")
02:26:32 Join stevenm [0] (
02:26:46 Quit asdsd_ (Read error: 54 (Connection reset by peer))
02:32:49 Join asdsd_ [0] (
03:30:07 Quit meshuga ("BitchX: the new hardcore, psycho, nitro client -- in a can")
03:35:37***Saving seen data "./dancer.seen"
04:04:57 Quit Aison (Read error: 54 (Connection reset by peer))
04:05:42 Join QT_ [0] (as@area51.users.madwifi)
04:08:36 Quit QT (Read error: 60 (Operation timed out))
05:22:05 Quit stevenm (Read error: 104 (Connection reset by peer))
05:22:07 Join stevenm [0] (
05:29:23 Join webguest13 [0] (
05:29:39webguest13anyone there?
05:29:44stevenmwhat's up ?
05:30:00webguest13I am trying to set up my environment and have a problem
05:30:10stevenmouch. what problem ? win or lin ?
05:30:26webguest13with CygWin
05:30:46 Quit asdsd_ (Read error: 145 (Connection timed out))
05:32:01webguest13When I try to make I get the error unrecognised architecture specification 5249
05:32:34stevenmsorry, I personally have very little experience with cygwin. Perhaps you can get someone to send you their environment
05:33:45webguest13okay, I will get in touch later
05:35:19 Part webguest13
05:35:40***Saving seen data "./dancer.seen"
06:11:10 Join DBygnes [0] (
06:11:34DBygnesis rockbox available on iRiver H140?
06:18:27DBygnesguess it's not..
06:24:34 Quit DBygnes ()
06:27:12 Quit stevenm (Read error: 110 (Connection timed out))
06:50:02 Join LinusN [0] (
07:35:42***Saving seen data "./dancer.seen"
07:45:15 Join rasher [0] (
07:46:17rashercurious about the ogg playback issue.. any idea what's happening?
07:46:30LinusNguess i have to work on the boot loader a little
07:46:48LinusNthe 1.65 firmware is a lot bigger than 1.63
07:47:25LinusNso i might have to move the loader a little
07:47:32rashersome things being overwritten/moved and the iriver firmware not coping?
07:48:06Rickhope that didn't break the "fixed" firmware's ability to update the firmware
07:48:19LinusNno it didn't
07:48:20rasherit didn't
07:48:39LinusNthat's one of the thing i *did* test
07:48:44rasherI've flashed back and forth to test it
07:53:40*rasher <- iRiver
07:53:43*rasher -> uni
07:53:55 Quit bobdbob ("Trillian (")
07:54:14*LinusN found a really weird bug
07:54:44LinusNif the disk spins down at the same time the dma starts playing, the iriver hangs
07:55:34rasheroh well, bye
07:55:36 Quit rasher ("CGI:IRC 0.5.4 (2004/01/29)")
07:56:35RickLinusN: the iriver firmware? "working one?
07:56:44 Join bobdbob [0] (
07:57:01LinusNRick: rockbox
07:58:11Rickwait, dma?
07:58:21LinusNaudio dma
07:59:00LinusNprobably an ata related issue
08:06:04RickOh, I was meaning to ask
08:06:27RickI was searching for example spi code in rockbox, not many things were obvious... know where I could look?
08:06:59Rickahh, that's probably why then
08:07:04RickI wasn't searching .S
08:07:48LinusNfmradio.c is probably the simplest
08:08:04RickI'm just looking to see if I can play with the remote
08:08:11LinusNgood luck
08:08:54LinusNdac.c is also an example
08:09:07LinusNno, sorry
08:10:03Rickis there currently a way in rockbox to tell if the remote is plugged i?
08:10:35RickI assume that would be one of the problems with working with the remote
08:11:25Rickand i'm guessing the hardware handles remote keys automatically
08:11:29LinusNi believe i found a way last time i looked, but i'm not sure
08:11:39LinusNwhy would you need to know?
08:11:56LinusNexcept of course to save cpu
08:11:56Rickwouldn't it be a waste to draw to the remote screen if it wasn't plugged in?
08:14:14LinusNi'm not sure if we can detect it, but i'll see if i can come up with something
08:14:25Rickthanks for the help
08:14:35amiconnIs the cpu port pin where the remote resistor network connects also a pio pin?
08:15:16amiconnIf yes, it should be possible to switch it to output, output a 1, switch back to input and check whether it goes to 0 within a short time
08:15:49amiconn(Same way as I detect the presence of the ata power switching circuit in the archos player bootloader)
08:20:25LinusNthe adc is external
08:20:32LinusNan spi device
08:20:58LinusNbut there are other pins we can do it with
08:21:10LinusNthe hold switch and on key sense
08:24:22 Quit bobdbob (Read error: 60 (Operation timed out))
08:24:22RickI noticed when you boot the original iriver firmware it takes a couple seconds for the remote lcd to initialize or whatever
08:24:36Rickmaybe not a couple seconds
08:24:40Rickbut there does seem to be a delay
08:25:38LinusNthat hardly means anything
08:25:47Rickno idea
08:25:51LinusNthey just do things in a specific order
08:25:52*Rick doesn't know hardware
08:45:29*LinusN finds a really old bug in ata_perform_sleep()
08:46:31LinusNwonder if this has any impact on RLD...
08:50:17LinusNbah, false alarm...
08:51:28LinusNstill, there is a possibility that the ata driver may hang if the ata thread isn't scheduled fast/often enough
08:52:14LinusNignore me
08:56:41 Join Harpy [0] (
09:09:27 Nick tvelocity[away] is now known as tvelocity (
09:16:35 Nick Lynx_awy is now known as Lynx_ (HydraIRC@
09:23:47 Join bobTHC [0] (
09:23:56bobTHCmornin' folks !
09:26:31 Join austriancoder [0] (
09:27:15austriancoderhi all
09:27:42austriancoderrick: i have here big parts of the remote control high-level driver finished
09:28:01austriancoderi need only a working low-level driver to test
09:29:10Rickaustriancoder: not to sound silly, but why/how would/could you develop a high level before the low level?
09:29:34austriancoderi asume some functions
09:29:58 Quit Hadaka (Read error: 54 (Connection reset by peer))
09:30:25austriancoderlike lcd_remote_write_command
09:30:36LinusNwriting is easy, testing is a little bit harder :-)
09:30:37Rickjust seems silly to do that
09:30:49LinusNRick: why?
09:30:56austriancoderrick: and why
09:31:02Rickdunno, just feels that way
09:31:50LinusNRick: i normally do it from the ground up, but in this case, it's nice to have the upper level ready to test with
09:32:12LinusNespecially since the low level isn't there, and austriancoder wanted something to do
09:32:24LinusNaustriancoder: did you have a look at the upside-down bug?
09:33:02austriancoderLinusN: yeah i have looked at it, but i didn't found the bug yet..
09:33:28*Rick goes to check
09:33:36Rickif I can find the link to the spec
09:33:46LinusNtomato lcd spec?
09:34:02 Join webguest54 [0] (
09:34:09Rickgot it
09:34:34RickI just noticed that the spec mentions 65x132
09:34:36 Join Naked [0] (
09:34:43Rickdoes that mean in the case of the iriver remote, that it's sitting on it's side?
09:34:47 Nick Naked is now known as Hadaka (
09:35:45***Saving seen data "./dancer.seen"
09:37:59austriancoderrick: could you try to write a lcd.s with remote controll functions?
09:38:00 Join rasher [0] (
09:38:10 Nick QT_ is now known as QT (as@area51.users.madwifi)
09:38:13Rickwell, I want to
09:38:29Rickbut I don't really have much experience in that field
09:38:46austriancoderthats also my problem
09:39:02Rickwhich is why I was asking for examples earlier
09:39:32austriancoderok ;)
09:41:45LinusNit isn't *that* hard
09:41:59LinusNonce you have set up the port pin configuration correctly
09:44:07 Quit Lynx_ (" reboot...")
09:50:08LinusNbut i guess it can be quite intimidating if you haven't done stuff like that before
09:51:40*Rick nods
09:55:02Rickit's mostly a "where do I begin" sortof thing
09:55:24Rickand just thinking about it
09:55:47Rickwould the backlight be part of the lcd?
09:56:12RickI thought that might be a good place to start
09:56:16Rickget that toggleable
09:56:21LinusNRick: do like this:
09:56:35LinusNcopy adc.c to lcd-h100-remote.c
09:57:58rasherThis ML business is wrecking my head
09:58:07LinusNthis will be C
09:58:18rasherYes.. I'm at uni :)
09:58:46LinusNah, ML, not Machine Language
09:59:02rasheroh ah.. indeed
09:59:34LinusNyeah, those functional languages are about as far from the real world you can get :-)
09:59:50rasherit's quite insane
09:59:59LinusNelegant, but insane
10:00:20RickLinusN: I'm reading the adc.c code now... is there anything else?
10:00:49LinusNyou only want to keep the iriver code in it
10:02:21LinusNthe things you want are from line 112 (#define CS_LO) up to line 210 (#elif CONFIG_CPU == TCC730)
10:04:12*austriancoder is away for a short time
10:04:33 Join Gibbed [0] (
10:04:35 Quit Rick (Nick collision from services.)
10:04:45 Nick Gibbed is now known as Rick (
10:04:47RickI curse you, router.
10:05:05Rick[01:00:01 AM] <LinusN> elegant, but insane
10:05:05Rick[01:00:21 AM] <Rick> LinusN: I'm reading the adc.c code now... is there anything else?
10:05:05Rick[01:01:48 AM] * Disconnected
10:05:31 Join ashridah [0] (
10:06:27LinusNyou only want to keep the iriver code in it
10:06:29LinusNthe things you want are from line 112 (#define CS_LO) up to line 210 (#elif CONFIG_CPU == TCC730)
10:06:59LinusNhang on
10:07:33Rickremoved :]
10:07:49Rickchanged the #ifdef accordingly
10:08:02amiconnRick: is very helpful in such cases :)
10:08:12LinusNok, so you have lcd-h100-remote.c?
10:08:13Rickamiconn: ah, thanks
10:08:17*Rick nods at LinusN
10:09:25Rickwith the non-h100 code stripped out
10:09:54LinusNnow rename the adc_scan function to:
10:10:00LinusNvoid lcd_remote_write_command(int cmd)
10:10:28Rickand i'm guessing I'll find the commands in the tomato spec
10:10:32*Rick opens that back up
10:10:44LinusNRick: don't worry about the commands
10:11:21*rasher pats Rick on the back
10:11:34*Rick grins
10:11:34LinusNthe interesting part is on page 17
10:12:31LinusNyou can remove adc_read(), adc_counter and adc_tick()
10:12:51LinusNnow, let's concentrate on adc_init()
10:13:08LinusNopen this page in your browser:
10:14:13*Rick is looking
10:14:40LinusNlook in the GPIO table
10:14:55Rickam, noticed 16 and 28
10:15:50LinusNand also GPIO1, 50
10:16:01*Rick nods
10:16:19LinusNso we need to configure these correctly
10:16:37Ricki'm guessing that occurs in (soon to be renamed
10:16:40Rick) adc_init
10:16:43Rickhit enter a little too early
10:17:16LinusNyes, GPIO_FUNCTION controls whether the pin is a general purpose I/O or not
10:17:44LinusNso, bit 16 and 28 should be set to 1 in GPIO_FUNCTION
10:18:20LinusNand bit 18 in GPIO1_FUNCTION
10:18:39LinusN(50-32 = 18)
10:19:08 Join Lynx_ [0] (HydraIRC@
10:19:45 Join B4gder [0] (
10:20:26rasherIt is still morning!
10:20:41LinusNB4gder: moo
10:20:49RickLinusN: in the comment for the existing GPIO_FUNCTION, are the numbers referring to the actual bits, or value?
10:21:20Ricknevermind, silly question
10:22:40LinusNthe comment for GPIO_FUNCTION should read "GPIO16: A0" and "GPIO28: CLK"
10:22:45Rickah, okay
10:23:22Rickif I converted correctly, should be 0x10010000?
10:23:38LinusNlooks fine
10:24:18LinusNnow do the same for GPIO1_FUNCTION (GPIO50: Data)
10:24:24Rickyeah, just did that
10:24:52LinusNGPIO_ENABLE controls the direction of the port pins
10:25:04LinusN1 = output
10:25:12Rickfrom what I read when I went through the tomato spec, it only goes one way anyway, right?
10:25:45LinusNso you should write the exact same bits to the xxx_ENABLE regs
10:25:56RickGPIO1 too?
10:26:13austriancodercool.. looks like the remote lcd is working in the next days ;)
10:26:16LinusNall pins are outputs
10:26:32LinusNA0, CLK and Data
10:27:02austriancoderwill switch to linux to prepear my high-level driver :)
10:27:04LinusNjust copy the GPIO_FUNCTION statements and rename GPIO_FUNCTION to GPIO_ENABLE
10:27:29 Quit austriancoder ("CGI:IRC 0.5.4 (2004/01/29)")
10:27:30LinusNand the same for GPIO1_FUNCTION
10:27:56LinusNnow, GPIO_OUT controls the actual value to drive on the port pins
10:28:03LinusNand GPIO1_OUT
10:29:28Rickhow does that work?
10:29:52LinusNyou set the corresponding bit to 0 or 1 in the register
10:30:39LinusNnow, we have one small problem
10:30:54Rickfinding what register?
10:31:37LinusNwe are missing one vital signal, the Chip Select
10:32:22LinusNi haven't investigated it fully, but it looks like CS is on GPIO34
10:33:11Ricker, wouldn't that be GPIO1 then?
10:33:25LinusNbit 2
10:33:27Rick(what's the difference between GPIO and GPIO1?)
10:33:58LinusNGPIO is i/o port 0-31, GPIO1 is 32-63
10:33:58 Join austriancoder [0] (~austrianc@
10:34:26LinusNthe registers are only 32-bit, but there are 64 i/o pins
10:34:49Rickmakes sense
10:35:04LinusNdid you add CS to the GPIO1 settings?
10:35:27LinusNbtw, better rename A0 to RS to match the data sheet better
10:35:28 Join rasher_ [0] (
10:35:43 Quit rasher ("CGI:IRC (Ping timeout)")
10:35:45Rickwell, assuming it needs to be added to GPIO1_FUNCTION, yes?
10:35:57LinusNjust like the other bits
10:36:01Rickah, okay
10:36:09Rickwhat exactly does the chip select do?
10:36:28LinusNit tells the lcd controller that we want to speak to it
10:36:45LinusNthere may be several spi devices on the same bus
10:36:59LinusNand you select the device to talk to by asserting the CS
10:37:33RickSo would I note that just as 'Chip Select' or is there a better name?
10:37:44LinusNCS is common
10:38:13LinusNso you now have CS, Clk, Data and RS
10:38:42LinusNwe want CS to be a 1 to begin with, in GPIO1_OUT
10:38:59LinusN(it is active low)
10:39:09Rickwhat does active low mean?
10:39:25LinusNit means that the chip is selected by setting CS to 0
10:39:38LinusNlook at page 17
10:39:58LinusNi *assume* that GPIO34 is connected to /CS1
10:40:33LinusNthe diagram shows that the communication starts by setting /CS1 low
10:40:49LinusNbut we want it to be high before starting
10:40:58LinusNso we set it to 1 in the init()
10:41:35LinusNGPIO1_OUT |= 0x00000004 iirc
10:42:27LinusNok, are you with me?
10:42:29*Rick nods
10:42:55LinusNhaving CLK low might be handy as well
10:44:28Rickwhat defines low?
10:44:59LinusNjust borrow the existing line in adc_init() and change the bit number
10:45:04*Rick nods
10:45:42LinusNremove the calls to adc_scan() and tick_add_task()
10:45:43RickGPIO_OUT &= ~0x00010000; /* CLK low */
10:45:47Rickyeah, did that already ;)
10:46:00 Quit edx (Read error: 145 (Connection timed out))
10:46:08LinusN0x00010000 is the RS bit
10:46:25*Rick read it backwar
10:47:01LinusNshould be 0x10000000
10:47:16LinusNok, not the init function is done
10:47:35LinusNlet's move to lcd_remote_write_command()
10:47:42RickI renamed it to lcd_remote_init, that's suitable?
10:47:44DMJCheh someone today accused me of writing rockbox's playback code
10:47:50LinusNRick: perfect
10:47:56DMJCI'm like.. WTF? I can't program worth a damn
10:48:04RickDMJC: hehe
10:48:25RickLinusN: I'm guessing it would be a good idea to update the macros?
10:48:29LinusNRick: we will now change the #defines
10:48:33*Rick nods
10:49:01LinusNwant to try yourself before i pat you on the head? :-)
10:49:22Rickalready done CLK and CS
10:49:26Rickwhat's DO?
10:49:29Rickand would I need it for RS?
10:49:38Rickand Data
10:49:46LinusNDO is Data Out
10:49:52RickAh, guessing that = Data
10:50:01LinusNactually no
10:50:12Rickwhat's the difference?
10:50:20LinusNthe macros are named from the ADC data sheet
10:50:33LinusNwhere Data Out is the data from the ADC
10:50:44LinusNyou want DI_xx
10:50:52LinusNrename them to DATA_xx
10:51:14LinusNand add similar macros for RS_xx
10:51:33LinusNi'll be away for 5 minutes, brb
10:52:59 Join preglow [0] (
10:53:28Rickgreets preglow
10:55:03rasher_STILL MORNING!
10:55:06 Nick rasher_ is now known as rasher (
10:56:37LinusNok, so now you have CS_LO, CS_HI, CLK_LO etc
10:56:41*Rick nods
10:56:52LinusNnow we can start with the actual communication
10:56:58LinusNpage 17 here we come
10:57:24Rickpage 17 is 'LCD DRIVER CIRCUIT', right?
10:57:24LinusNwith these preparations, all we need to do is follow the diagrab, bit by bit
10:58:09*LinusN is an idiot
10:58:24Rickguessing you were viewing the epson one?
10:58:26LinusNi looked in the wrong data sheet
10:58:39LinusNno, another tomato model
10:58:48LinusNpage 10 is what we want
10:59:06RickNo problem
10:59:43preglowgood day
10:59:48LinusNpreglow: hi
10:59:55LinusNwe start from the left
11:00:09Rickof table 5-3?
11:00:36LinusNno, figure 5-1, serial interface timing
11:00:42 Quit DMJC ("Leaving")
11:00:45Rickmy next guess ;P
11:01:08LinusNthe communication starts by asserting /CS1, i.e setting it low
11:01:40Rickyeah, starting to understand what the diagram means now that you said that
11:01:55LinusNthis translates to a CS_LO statement
11:02:01Rickwhich is already there
11:02:22LinusNbetter remove all the other statements btw
11:02:34*Rick nods
11:03:38Rickis SCLK CLK?
11:03:39rasherexciting, this is :)
11:03:44LinusNRick: yes
11:03:49Rickwhat is SID?
11:03:56 Quit amiconn (Read error: 110 (Connection timed out))
11:04:05Rick(maybe ID = input data?)
11:04:30Lynx_LinusN, Rick: what are you doing? And what the hell is a tomato spec? ;)
11:04:31LinusNyou catch on quickly :-)
11:04:42LinusNLynx_: remote lcd
11:04:43RickLynx_: tomato spec is for remote lcd
11:04:46Lynx_ah, ok
11:05:03Rickhe's teaching me about coding spi
11:05:15LinusNRick: when we do this, we also need to look at page 50 for the actual timing, but let's ignore that for now
11:05:38Rick(I'm guessing that's what the DELAY was for in the adc stuff?)
11:05:39LinusNso, we have set CS low, now what?
11:05:48LinusNRick: yup
11:06:02RickI'm not sure what two lines crossing each other mean
11:06:08LinusNsee the arrows on the SCLK graph?
11:06:40Rickwhat does it mean? setting to low then hi?
11:06:42LinusNthat means that the value on the SID line is sampled on the rising edge of SCLK
11:07:03RickI sortof understand that
11:07:19LinusNthe crossing lines means "value change"
11:08:33LinusNok, we know that CLK is already low
11:08:56LinusNso now we need to set the DATA pin correctly
11:08:56Rickso is the line for CLK 'upside down'?
11:09:38LinusNwe see from the diagram that bit7 is to be transferred first
11:09:44 Join amiconn [0] (
11:10:29preglowwhat's going on? remote lcd driver?
11:10:33LinusNthe classical way of doing this is to convert the bit vaue to a boolean expression
11:10:36LinusNpreglow: yes
11:10:58LinusNlike this:
11:11:08rasherI appear to be lagged from here to Siberia and back
11:11:17RickLinusN: er, so bit7 would be 0x1?
11:11:22LinusNif(cmd & 0x80)
11:11:33LinusN DO_HI;
11:11:33Ricknow I get it
11:12:18Ricker, don't you mean DATA_HI?
11:12:20LinusNthen we shift cmd one bit to the left and do it all over again
11:12:27LinusNRick: of course
11:12:56LinusNwe better do a loop, since there are 8 bits
11:13:07Rickthat's easy enough
11:13:31LinusNok, in the loop body, we first set the data bit accordingly
11:13:34 Quit webguest54 ("CGI:IRC")
11:13:50austriancoderi must go now.. if you have the basic lcd functions upload them and i will add them to my high-level driver
11:13:52*ferenczy is back
11:13:57austriancodersee you to night
11:13:59 Quit austriancoder ("using sirc version 2.211+KSIRC/1.3.11")
11:14:00LinusNthen we set CLK high to let the lcd clock the data
11:14:56Rickshould the loop permanently shift cmd?
11:15:26Rickdoes CLK set DATA back to lo?
11:15:28LinusNloop it 8 times, shifting cmd at the end of the loop
11:15:40 Join Aison [0] (
11:15:44LinusNCLK_xx only changes the CLK
11:15:53Rickso do we need to set DATA back to LO?
11:16:34LinusNthe DATA value is only used by the lcd on the rising edge of CLK
11:17:21LinusNit can be any value in between
11:17:47Rickso what if the cmd bit is false?
11:17:47LinusNso, we set the data pin, rise the clock, and now what?
11:18:00LinusNRick: we'll come to that
11:18:32LinusNso, to prepare for the next data bit, we should now set the clock low again
11:18:53LinusNthe loop should now look something like this:
11:19:01LinusNif(cmd & 0x80)
11:19:21*Rick nods
11:19:28Rickand i'm guessing a delay will go between the CLK's later
11:19:49*Rick adds a comment as a reminder
11:20:06LinusNso now we have transferred 8 bits
11:20:43Rickbut what i'm confused with is
11:20:58Rickonce you set DATA to hi it never goes back lo even if the bit is false
11:21:51Rickwell, assuming you start with DATA in lo
11:22:48LinusNmaybe you didn't get the idea with the if(cmd & 0x80)
11:23:01LinusNit should of course have an "else" clause
11:23:17LinusNif(cmd & 0x80)
11:23:21LinusN DATA_HI;
11:23:25LinusN DATA_LO;
11:23:30RickI wasn't sure of that
11:23:34Rickyou said we'd get to that, ;P
11:24:03Ricknow it makes sense
11:24:19LinusNok, my bad, i thought you understood that
11:24:35Rickit's ok :)
11:24:51RickI notice in the diagram, when it gets to the next byte, it does something with RS
11:25:38LinusNyes, this is the Register Select bit
11:26:06RickWhat does that symbol mean in the diagram? (the −−<==>−− thing)
11:26:32Rickeither or?
11:26:39Rickeither path, even
11:26:42LinusNdrawing the signal level in the middle means that the value if the signal is irrelevant
11:27:01Rickso i'm sorta right, I think
11:27:11LinusNand the <=> means that it can be either 0 or 1
11:27:14*Rick nods
11:27:20RickThat's what I figured
11:27:50Rickwhat is control data?
11:28:04RickI assume display data is for the actual lcd display contents
11:28:05LinusNcontrol data == cmd
11:28:24Rickso would we need to RS_LO at the beginning of write_cmd?
11:28:31Rickor it doesn't matter?
11:28:43LinusNthe RS value is ignored until we transfer bit number 6
11:28:48RickOh, ok
11:29:12Rickah, now I get it
11:29:41LinusNso if we follow the diagram by the book, we should transfer RS only when we're transferring data bit 6
11:30:24LinusNbut since it ignores the RS value the rest of the time, we can just set it to what we want before starting the transfer
11:30:46Rickwhich is what I thought, sorta
11:30:47LinusNand the lcd will read it when we come to bit 6 in the loop
11:30:52Rickbut, how are you reading it as bit6?
11:31:24LinusNwell, the 6th bit, rather :-)
11:31:43LinusNmaybe i should get better glasses
11:31:52Rickit looks like the 8th to me...
11:31:56LinusNit is
11:32:11LinusNsilly me
11:32:11Rickjust making sure
11:32:18 Quit Aison (Read error: 145 (Connection timed out))
11:32:39Rickso will we need RS_LO ?
11:32:40LinusNso, all we need is to set RS to the correct value right after setting CS
11:33:16Rickassuming write_command won't be used for data as well
11:33:26LinusNthe data sheet tells us that "The serial data input is display data when RS is high and control data when RS is low."
11:33:32*Rick nods
11:33:37Rickthat's where I got RS_LO from
11:33:42LinusNso we set it to 0, i.e RS_LO
11:34:05LinusNwe now end the transaction by setting CS high
11:34:11LinusNafter the loop
11:34:47Rickis that all we're sending?
11:34:49LinusNthis is not shown in the diagram, shame on them
11:34:54Rickbecause then it seems that the RS_LO doesn't matter
11:34:56Rickat all
11:35:02Rickbecause we never send an 8th bit
11:35:11LinusNwe do
11:35:21LinusN7..0 is eight bits
11:35:34Rickmisread the diagram
11:35:42Rickmaybe I need glasses :P
11:35:45LinusNthe loop should be 8 laps
11:35:47***Saving seen data "./dancer.seen"
11:35:57RickI just did a for
11:36:21LinusNthis function is used for sending single commands
11:36:24 Join Aison [0] (
11:36:34Rickthe original adc code returned a value, will that be needed?
11:36:36Rickdoesn't seem so
11:36:45LinusNno, this function returns void
11:36:49Rickas I don't think we can get any data back anyway
11:36:52*Rick nods
11:37:06LinusNnow we have the logic, let's look at the timing
11:37:10LinusNpage 50
11:37:59RickNo idea what this means, hehe
11:38:09LinusNwas about to say "intimidating"
11:38:15*Rick grins
11:38:45Ricki'm guessing the lines represent the stuff in the other diagram
11:38:50LinusNthe diagram might be confusing, but the table says most of it
11:39:02Rickwhat is SYMBOL?
11:39:10 Quit Strath (Read error: 110 (Connection timed out))
11:39:21LinusNthey are the names in the diagram above
11:39:41Rickyeah... but do they have any particular meaning?
11:39:47LinusNoh yes
11:39:50Rickoh, nevermind
11:39:53Rickit's just abbrvs
11:39:56Rickof 'item'
11:40:01LinusNlet's look at "Serial clock cycle"
11:40:09rasherBoy, am I ever hungry
11:40:13rasherlogbot: feed me
11:40:19LinusNwan't to continue after lunch?
11:40:34Rickwho, you?
11:40:40RickI had lunch say 14 hours ago
11:40:59LinusNnote to self: rasher != Rick
11:41:00Rickrasher, you're confusing Linus :(
11:41:23LinusNboth are blue in my irc client :-)
11:41:45LinusNok, serial clock cycle
11:42:17RickI don't think on the diagram
11:42:27LinusNone cycle is the transition from low to high and back again
11:42:28Rickunless CYS is a typo
11:42:42LinusNCYS is a typo
11:43:12Rickokay, so one cycle is transition from lo to hi in SCLK
11:43:25Rickwhat's Min mean?
11:43:29LinusNthe table tells us that the minimum time a cycle should take is 250ns
11:43:36Rickand what is ns?
11:43:41LinusNlo to hi to lo again
11:44:52Rickso how do we ensure that time?
11:44:59LinusNit also tells us that each phase of the clock must be longer than 100ns
11:45:35Rickphase = hi->lo?
11:46:03RickI think so
11:46:06Rickor lo->hi->lo
11:46:09LinusNthe clock has two phases, the low phase and the high phase
11:46:28LinusNor "period"
11:46:43LinusNand in this table "pulse width"
11:46:56Rickis 100ns, I think
11:47:30*Rick thinks
11:47:53Rickso, thinking about this, we want a delay after CLK_HI and CLK_LO
11:48:03LinusNso, when we set the CLK low, we must wait at least 100ns before setting it high again
11:48:13Rickof 100s, maybe 150s for CLK_LO?
11:48:45LinusNexactly, we must fulfill both requirements
11:48:57Rickwould it be wise to go over that a little bit?
11:49:00Rickto be sure?
11:49:05Ricklike 110 and 160
11:49:16LinusNor 130/130
11:49:22Rickyeah, that'd work
11:49:28Rickrasher: hehe
11:49:30rasher(nevermind me)
11:49:43Rickso how do we delay time in nanoseconds?
11:49:47*rasher watches some avid gentoo user complain
11:49:53LinusNok, one CPU cycle is 8ns in 120MHz
11:50:23Rick16.25 cycles
11:50:37rashergood luck waiting .25 cycle :)
11:50:50Ricki guess we could just chop that off
11:51:48LinusNeach CLK_xx, DATA_xx etc takes several cpu instructions
11:53:23Rickthat brings me back to the question of: how do you delay time in nanoseconds?
11:53:27LinusNmove #GPIO,%a0
11:53:35LinusNmove (%a0),%d0
11:53:45LinusNori.l #xxx,%d0
11:53:54LinusNmove %d0, (%a0)
11:54:18LinusNthat's a typical xxx_HI sequence
11:55:11LinusNthat's 8 instruction fetches
11:55:28LinusNplus 2 reads/writes to the port
11:55:41LinusN10 bus cycles
11:55:53 Quit Aison (Connection refused)
11:55:54LinusNat best
11:56:07LinusNwe could read the cpu timing diagrams as well to find out
11:56:31RickI wouldn't worry about that right now I guess
11:56:34Rickmaybe to optimize later
11:56:34LinusNthe point is that it will take *at least* 80ns to perform the DATA_xx statement
11:57:13LinusNand the if(cmd & 0x80) will take a few ns as well
11:57:53LinusNat least two instructions followed by a branch
11:58:17Rickso would that be around 100ns?
11:58:53LinusNso we have fulfilled the CLK low pulse width requirement
11:59:29LinusNnow let's see about the CLK high and cycle time requirements
11:59:37 Join Aison [0] (
11:59:38Rickdo I have my stuff backward?
11:59:44RickI have CLK_HI after the if stuff
11:59:55LinusNthat's correct
12:00:06Rickso are we fufilling CLK_HI or CLK_LO?
12:00:20Ricklogical would point to CLK_HI, I think
12:00:27LinusNclk is low when we begin the loop, we then set the data pin and then CLK_HI
12:00:33Rickoh, true
12:00:42LinusNthat's the "low pulse width"
12:00:55LinusNthe time before we set CLK high
12:01:04Rickah, okay
12:01:10Rickthat clears it up
12:01:29rasherthink I'll go home now.. and find something to eat on the way
12:01:30LinusNwe now set clk high and then low again right after that
12:01:39LinusNrasher: bye
12:01:43Rickwe also shift cmd
12:01:49 Quit rasher ("going home")
12:01:57Rickclk_hi, clk_lo, cmd <<= 1
12:02:32LinusNwe can see that we might set clk low again a little too soon
12:03:04LinusNthe high pulse width needs to be at least 150ns
12:03:05Rickit would be wiser to shift before CLK_LO
12:03:13LinusNRick: exactly
12:03:15Rickto include the shift as part of the hi delay
12:03:27LinusNRick: exactly
12:03:30Rickokay :)
12:03:37Rickso how long does a shift take?
12:03:52LinusNthe cmd value is likely to be in a cpu register
12:04:15Rickso fast?
12:04:24LinusNso it probably takes a single cycle
12:04:55LinusNi.e 8ns
12:05:17Rickand we need ~16cycles to get about 128ns
12:05:31LinusNso we need to insert a delay
12:05:52Rick15 cycles should be fine
12:05:55RickI think
12:06:07LinusNthis is something we probably would want to code in assembler later on
12:06:29LinusNbut for now, lets do a DELAY
12:06:31Rickthe delay or the entire function?
12:06:38LinusNentire function
12:06:46Rickwell, nothing like a good prototype ;)
12:07:00Rickfrom what I saw in the old DELAY
12:07:07LinusNthe DELAY macro does an empty loop of 10 laps
12:07:17Ricki assume 10 = cycles?
12:07:26LinusNmuch more, i'm afraid
12:07:39Rickso how much is a lap?
12:08:41LinusNthe compiler is not that clever
12:08:46Ricktaking from what you said earlier...
12:08:54Ricki'm guessing it's more than an if
12:09:13LinusNeach lap would take about 5 cycles
12:09:26Rickso we want 3 or 4 laps
12:10:55LinusNthe shift takes 8ns cycle, the CLK_LO takes 80ns, right?
12:11:47LinusNso we would like to delay about 60ns
12:11:50Rickyou can include CLK_LO in that too?
12:12:17LinusNthe actual transition takes place at the end of the CLK_LO instruction sequence
12:12:32LinusNthe final write to GPIOx
12:12:45Rickmakes sense
12:13:30LinusNso two laps would be adequate
12:13:35Rickfrom this, we only... yeah
12:14:24Rickis the CS1 delays taken care of too, because of the looping?
12:14:34LinusNlet's see
12:14:37Rickand RS
12:14:41Rickthey are 150
12:14:50LinusNthe /CS1 setup time is 150ns
12:15:11Ricki'm not sure how to read that
12:15:21Rickbecause of the big line at the end that goes down through everything
12:15:48LinusNit means that the lcd might take 150ns to recognize the /CS1 change
12:16:42RickI think it would
12:16:43Rickbecause we have
12:16:48Rickwell actually no...
12:16:52Rickwe should have like half of it
12:16:58Rickdue to RS_LO
12:17:04LinusNif we look at the diagram above, we see that the /CS1 setup time is defined as the time from CS1 low to the first CKL high
12:17:13Rickyeah, that's what confused me
12:17:24Rickyeah, I think we have it
12:18:03LinusNmakes sense, since all the lcd does is to wait for clk lo->hi transitions to sample the data pin
12:19:38LinusNthe logic inside the lcd is if(!CS1 && CLK_lo2hi) databit = DATA;
12:19:38Rickand RS is met too
12:19:42RickI think
12:20:34LinusNRS should be met by far
12:21:03LinusNthe only thing we have left is the CS1 hold time
12:21:50LinusNmeaning that CS1 should not go inactive until 150ns after the last CLK lo->hi transition
12:22:09preglowi wonder if imdct_l will frown upon me stealing eight bits of precision, using the accumulator extension bytes is major bother
12:22:29LinusNbut since we don't set CS1 inactive until after the CLK high pulse, we are home free
12:22:45LinusNpreglow: you're ears will tell you
12:23:20preglowoh, i expect my ears will be perfectly fine with it
12:23:25Rickwhat would be a good way to test this?
12:23:30LinusNRick: time to celebrate the completion of lcd_remote_write_cmd
12:23:33preglowthe default imdct_l seems to loose much more precision than that
12:23:35preglowbut we'll see
12:23:47Ricktake a look at the commands possible?
12:23:54Rickand choose something that does something obvious?
12:24:01LinusNRick: time to try to compile it
12:24:02preglowusing the extension bytes is going to make the coding painful on me, so i'll ignore them
12:24:07Rickthat too
12:24:20LinusNadd the file name to firmware/SOURCES
12:24:33*Rick nods
12:24:56LinusNi'll have to go to lunch now
12:25:16LinusNi'll be back in <30 mins
12:25:32LinusNin the meantime, your homework is to compile it
12:25:40RickI don't expect any issues, but yeah
12:25:49Rickhave fun
12:25:50 Quit Aison (Connection refused)
12:26:47Rickcompiles fine
12:29:59 Join rasher_mob [0] (
12:30:53rasher_mobExcellent. Irc on my phone :)
12:31:36 Join Aison [0] (
12:32:12Rickadded a simple #ifdef block in lcd.h
12:32:17Rickto define the remote functions
12:33:45*Rick goes to define the commands
12:34:39rasher_mobSounds like things are progressing nicely
12:35:46Lynx_rasher_mob: what kind of phone is it?
12:37:19 Join edx [0] (
12:37:29rasher_mobNokia 6610i.. Nothing fancy.. Just java and gprs
12:38:00rasher_mobTyping is sortof annoying, but it's fun
12:38:16preglowhowd you go about doing that?
12:39:23rasher_mobI.. Think that's the right url
12:40:43rasher_mobI have only got access to http though, so I have to use a proxy.. Which results in horrible lag
12:40:52Lynx_sounds expensive, gprs
12:41:41Rickthese instructions are confusing ;P
12:42:28Rickoh nice
12:42:31rasher_mobI pay per MB, so it's not expensive at all :)
12:42:34Rickthere's example setup things
12:43:04Lynx_rasher_mob: depends on how much i type ;)
12:43:24rasher_mobWould be even less if I could connect directly..
12:44:25rasher_mobLynx_: you'd have to type A LOT to cost of any considerable amount of money..
12:44:50Lynx_i know :)
12:44:54rasher_mobMy phone would probably explode before that happens..
12:45:10preglowhrmph, error
12:45:25Lynx_rasher_mob: connect directly?
12:46:21rasher_mobPlus I get all the great T9 errors.. Like me/of etc..
12:46:52rasher_mobLynx_: rather than through a http<->irc proxy
12:47:08Lynx_rasher_mob: ah, gprs does only allow http?
12:47:18preglowit's a provider thing
12:47:36 Join prethom [0] (
12:47:41preglowprethom: why, hello
12:47:43rasher_mobpreglow: trying wlirc?
12:48:03rasher_mobYeah.. Or in some cases a phone thing.
12:48:20preglowwhat a toy
12:48:53rasher_mobIt's much fun.
12:49:09Lynx_i assume t9 does not know w00t yet? ;)
12:49:25rasher_mobAnd if you don't pay per minute, pretty cheap
12:49:56rasher_mobT9 doesn't like words with numbers in them.. At all.
12:49:57Rickcan't figure out what VDD-VSS means
12:49:57prethomWorst thing is, its almost usable.
12:51:02rasher_mobYou're trapped, now!
12:51:15prethomEspecially with this nice tiny font.
12:51:17Rickoh, never mind I think
12:52:00 Quit prethom (Remote closed the connection)
12:52:15preglowbut whe the hell did i have to connect by proxy, my phone is listed as being supported in ordinary socket mode
12:52:15rasher_mobprethom: what kind of display do you have?
12:52:43rasher_mobProvider being stupid?
12:52:54Rickheya LinusN
12:53:02Rickbeen reading how the power on sequence goes
12:53:07 Join F1^Aison [0] (
12:53:10preglowbut i'd better not make it work, i don't like using mobile phones as it is
12:53:46Rickthe diagram is on page 40
12:55:57 Nick F1^Aison is now known as Aison` (
12:56:34LinusNRick: our job starts at "User application setup"
12:57:00LinusNbut we aren't done with lcd-h100-remote.c yet
12:57:37Rickyeah, I figured that
12:58:01rasher_mobNow the question is: am I lazy enough to wait 8 minutes for a bus to take me 750m?
12:58:25Lynx_rasher_mob: how sure are you it will really be 8 mins? ;)
12:58:37LinusNrasher_mob: walk, goddammit
12:58:42RickLinusN: what else do we need to do?
12:59:03LinusNwe need to implement lcd_remote_command_ex()
12:59:11Rickwhat will that do?
12:59:32LinusNand lcd_remote_write_data()
12:59:35rasher_mobBecause this is their starting point :) the only place where they're almost always on time
13:00:00Lynx_rasher_mob: where are you anway?
13:00:08LinusNlcd_write_data() is very simple
13:00:18RickI figured so, LinusN
13:00:32LinusNlet's begin with that
13:00:34rasher_mobLinusN: alright!
13:01:05LinusNrasher_mob: exercise is good for you
13:01:21preglowand it's good for me as well
13:01:22Ricki'm guessing we want page 11?
13:01:31 Join xen` [0] (
13:01:33preglowso i'll go drag out my bike and see what's up in trondheim
13:01:34preglowlater, all
13:01:48 Quit preglow ("leaving")
13:02:20LinusNthe prototype is void lcd_remote_write_data( const unsigned char* p_bytes, int count );
13:03:09LinusNcopy the lcd_remote_write_command function
13:03:37LinusNwhat it does is write an array of display data bytes to the lcd
13:03:48Rickobviously we need RS_HI now
13:04:04Rickfirst bit seems to be the count
13:04:10Rickbut what is /WR?
13:04:23LinusNare you at page 11?
13:04:32LinusNpage 10 is all we need
13:04:33Ricklooking at the top most diagram
13:04:51LinusNit's the exact same thing as the command write
13:04:57LinusNwith RS_HI
13:05:06LinusNand an outer loop counting the bytes
13:05:49LinusNso you create a local variable called, for example, "data"
13:05:57Rick(basically a cmd replacement?)
13:06:21LinusNand transfer each byte in the array
13:07:43RickCS_HI should be done after both loops right?
13:07:46 Quit rasher_mob (Read error: 54 (Connection reset by peer))
13:07:55Rickkay, just making sure
13:08:02LinusNCS surrounds the entire transaction
13:08:03 Quit Aison (Connection refused)
13:08:04Rickall set then
13:08:24LinusNok, next is lcd_remote_write_command_ex()
13:08:28LinusNwhich is a hybrid
13:08:48LinusNvoid lcd_write_command_ex( int cmd, int data1, int data2);
13:09:20LinusNi.e first send the command, followed by one or two data bytes
13:09:46Rickshouldn't the data1 and data2 be unsigned char?
13:09:57LinusNthe second byte will be -1 if it isn't used
13:11:05LinusNso you could just copy the transfer loop 3 times and set RS high after the first byte
13:11:16LinusNand the third transfer should be optional
13:11:20Rickwhy not just call the original functions?
13:11:29LinusNbecause they flip the CS
13:11:34Rickoh, true
13:11:42LinusNnot a problem per se
13:11:49LinusNbut non-optimal
13:12:13*LinusN gets some coffee
13:12:45Rickdo I really need a loop for the data ?
13:12:49Rickas i'm just writing two bytes
13:13:10RickI mean, the outer loop as in write
13:13:48Rickwell, I think I've got it
13:14:32Rickwith data2 optional
13:14:37 Join rasher [0] (
13:15:01LinusNRick: can i see your code now?
13:15:08Rickone sec
13:16:33rasherGot home before the bus left the station :)
13:17:14Rick(yes, I know the tabs are all messed up... I forgot to tab->spaces :P)
13:17:19*Rick fixes
13:19:33LinusNyou shouldn't edit the $id:$ field
13:19:34*rasher steals half of Rick's newlines
13:19:51RickLinusN: meh, it had the adc stuff in it -shrug-
13:20:05LinusNit is automatically updated by cvs
13:20:21LinusNit doesn
13:20:29Rickdoesn't what?
13:20:46LinusN't matter what you write in it, but the ending $ must be there
13:20:52RickI didn't know that
13:21:07Rickfixed :p
13:22:14Rickany comments? I gotta head to sleep now
13:22:15LinusNlcd_remote_command_ex() is not needed
13:22:23Rickoh, you told me to put that there
13:22:33LinusNtypo from my side
13:22:45*rasher applauds Rick
13:22:49Rickwell, removed
13:23:11RickI had started to make #defines of all the commands
13:23:15Rickwhile you were gone
13:23:17LinusNsaw that
13:23:33LinusNi /msg'd you
13:23:35Rickand I started doing the init part
13:24:16LinusNi'll commit this to cvs while you sleep
13:24:30LinusNhope it's been educational
13:24:35Rickyes, it has been
13:24:53RickI started a little snippet of what I could interpret from the diagram of the init sequence
13:25:04Ricklemme add that in a comment
13:25:46Rickdone, it's at the bottom
13:26:01LinusNsure, we must leave some work for austriancoder too :-)
13:26:59RickI *think* that's correct anyway
13:27:03Rickof what I had
13:27:12Rickwould that stuff be called in lcd_remote_init?
13:27:34LinusNsure, why not?
13:27:44RickI figured so
13:28:14Rickokay, that should be all the updates
13:28:17Rickgood night
13:29:08Rickthanks for the lesson
13:29:13LinusNyou're welcome
13:31:27 Join DMJC [0] (
13:35:49***Saving seen data "./dancer.seen"
13:40:59LinusNRick: committed
13:46:58rasherooh ooh
13:54:34B4gdermy cvs build fixes seem to work fine now
13:58:29B4gderbut my fb driver still doesn't
13:58:52B4gdergetting closer though
13:58:52LinusNthey should hire a real programmer instead
13:59:11B4gderwhy, you know any? ;-)
13:59:17LinusNme? no
14:05:29 Quit elinenbe (" Try HydraIRC -> <-")
14:06:05 Quit DangerousDan (Read error: 60 (Operation timed out))
14:09:15 Join austriancoder [0] (
14:09:52rasheraustriancoder: Hurry up, great things are afoot!
14:11:06austriancoderwill commit later the day the final lcd-h100-remote.c
14:11:13austriancoderat the moment i am at uni
14:11:38austriancoderin about 3-4 hours i can test all.. i forgot my mp3player at home :(
14:11:56rasherhow unfortunate :)
14:12:22austriancoderthats life...
14:13:33rasherlife... don't talk to me about life...
14:18:32austriancoderboring lesson
14:19:31 Quit austriancoder ("CGI:IRC 0.5.4 (2004/01/29)")
14:20:04 Join preglow [0] (
14:20:33preglowwow, the remote driver's working already?
14:21:28 Quit rasher ("CGI:IRC (EOF)")
14:22:14 Join rasher_ [0] (
14:23:05 Nick rasher_ is now known as rasher (
14:23:36 Join elinenbe [0] (~elinenbe_@
14:25:08LinusNpreglow: don't know if it works
14:27:28rasherpcm playback api?!
14:28:33LinusNrasher: just an extension of the existing one, to allow for chained buffers with watermarks
14:28:59LinusNjust another step towards mp3 playback
14:30:49LinusNsignal when it's time to refill the buffer
14:30:50ashridahpreglow: yeah. hit a low watermark, start filling. high a high watermark, stop filling
14:31:00preglowahh, THAT kind of watermark
14:31:53preglowonly watermark term i knew of that applied to audio, is subtly altering the audio to ease detection of copyright infringement, heh
14:32:01preglowand i didn't think that was what you had in mind, heh
14:38:05B4gder... did some subtle changes on the daily build page
14:39:29bobTHCyou have add the daily complete source
14:39:43B4gderI moved it into the table, yes
14:39:51 Join lolo-laptop [0] (~lostlogic@
14:40:26 Quit Nibbler (Read error: 113 (No route to host))
14:40:33 Quit Hadaka (Read error: 60 (Operation timed out))
14:41:24B4gderand reordered the columns
14:41:38B4gderand when checking older packages, all types now links to the daily changelogs
14:44:16B4gdercoffee break!
14:47:30*HCl yawns
14:48:33*preglow looks at the clock
14:48:37*preglow disagrees
14:48:51LinusNnot good? ;-)
14:49:17elinenbeit's morning for me.
14:49:37preglowgood it is
14:49:42preglowmorning it is not
14:51:45LinusNpreglow: i have some questions regarding the libmad api
14:51:54preglowLinusN: go ahead
14:52:20 Quit thegeek (Read error: 104 (Connection reset by peer))
14:52:21LinusNi can see the Stream type
14:53:23LinusNi'm looking at the decoding loop in mpa2wav.c
14:53:59 Join Naked [0] (
14:54:07preglowit too can see it :)
14:54:12 Nick Naked is now known as Hadaka (
14:54:15HClyes well its morning when i wake up :/
14:54:20LinusNi don't understand all the copying of data
14:54:27*HCl had a nosebleed at 4:30 am >.<
14:54:49LinusNHCl: girlfriend hit you? :-)
14:54:54HCleh, no.
14:55:00HClshe's not gonna hit me, heh.
14:55:06HCljust dryness stuff.
14:55:18HCli'm just vulnerable to them, had them since i was young
14:55:25preglowi sometimes do that as well
14:55:34preglowLinusN: can't say i get it either
14:55:58LinusNit *should* be enough to pass a pointer to the stream.buffer
14:56:02preglowi'm so used to nosebleeds i just automatically turn around and sleep on
14:56:07preglowyes, indeed
14:56:13preglowand i don't understand why he uses this interface
14:56:19preglowlibmad also has a file callback interface
14:56:24preglowwhich i really think we'll end up using
14:56:29preglowthat's what _all_ the codecs use
14:56:36LinusNany docs online somewhere?
14:57:40preglowmusepack uses callbacks, libflac uses callbacks, everything uses callbacks
14:57:47preglowand it's not a bit way of doing it
14:57:58preglows/bit/bad, i suck
14:58:53LinusNi'll rewrite the plugin then
14:59:21LinusNi bet all this copying takes a great deal of cpu
14:59:36LinusNyou might gain a few % by using the callback api instead :-)
14:59:47preglowwell, that obviously also has to do copying
14:59:49preglowbut not so much
14:59:54preglowi _think_
15:00:19 Join Bippy [0] (
15:01:42LinusNhmm, the callback api might not be what we want
15:02:10preglowthen we'll need to do a fair bit of coding :-)
15:02:25preglowif we could make the callback api do stuff in place
15:02:30preglowthen maybe it'd work better
15:03:21LinusNi wonder if mad_decoder_run() keeps all state information between the calls
15:06:41preglowlibmad had a funny interface
15:06:51preglowwould be nice if mad_decoder_run() returned once in a while
15:07:39LinusNwe can do that by returning MAD_FLOW_STOP in the output callback
15:08:07LinusNbut i want to know if we can continue seamlessly by calling mad_decoder_run() again
15:08:28 Quit bobTHC ("Smoke Weed Every Day !")
15:09:18LinusNno, it doesn't look that way :-(
15:09:42 Join Tang [0] (
15:09:55preglowhow nice
15:10:02Tanghi Preglow
15:10:07Tangwhat's up
15:10:07preglowwell, that's kick in the groin
15:10:08preglowTang: hi
15:10:16preglowTang: not much, libmad stuff
15:10:34 Join muz [0] (
15:10:37Tangi was talking about the 1.65 ogg bug
15:10:45Tangis it linked to Rbx or not finaly?
15:11:10muzum is it normal that my remote lcd still doesnt display nethin?
15:11:12Tangi tested with 1.65 rbx and the ogg skipped (without freeze)
15:11:19LinusNit's probably rockbox' fault
15:11:23preglowi don't know, i don't use 1.65
15:11:29TangAh okay
15:11:37TangHi Linus
15:11:37LinusNsee IriverBoot
15:11:42preglowsounds like a very subtle bug
15:11:46TangI'm going :)
15:11:54Tang(indeed very strange)
15:12:43muzyou know the low level remote lcd drivers for the iriver, do they display anything yet?
15:12:49preglowmuz: untested
15:13:06preglowmuz: it was written a couple of hours ago
15:13:17muzrighty ho
15:13:41preglowno need to be sorry
15:13:44muzanyway i tried it out, and nothing shows
15:14:10preglowunless you yourself have made rockbox use it, small wonder
15:14:17preglow'cause nothing uses it yet
15:14:23muzohhh righttttt
15:14:39ashridahhm. if i add *.png -k 'b' into the CVSROOT/cvswrappers file in a cvs repository (via checkout/in), that automatically makes it use -kb when adding .png files, right?
15:14:47 Join bobTHC [0] (
15:14:54Tangi wonder if it's not an iRiver intentioned "trap"
15:15:16muzdoubt it
15:15:49muzthey cud have done worse
15:16:37BippyIriver could of done worse? how so :|
15:16:42LinusNif iriver did add some protection against modifying the firmware, they would have added a "modified firmware" message
15:18:16LinusNlooks like we can't use the mad_decoder_run() interface...
15:18:30BippyIs MP3 running realtime yet ?
15:18:37preglowwhat a foolish design
15:18:55BippySo if i used the latest daily build will i be able to play MP3?
15:19:00muzdoes that mean we cant decode mad?
15:19:01LinusNBippy: no
15:19:05Bippyomg :'(
15:19:31preglowi can't imagine the libmad people couldn't guess people would like to do other stuff than just decode and mpeg file
15:20:05Tangokay Linus
15:20:14Tangwas just an hypothesis
15:22:19BippySo my little monkeys, if MP3 is running realtime why isnt it working :|
15:22:40B4gderwe're not that "little" ;-)
15:22:50preglowbecause there's more to it than just decoding an mp3
15:22:51LinusNBippy: "running realtime" means that it is fast enough to run in realtime
15:23:01muzwait i think i can answer that: its cos there is no audio api yet
15:23:01BippyWell im 6'3 so most people are small, sorry if your not
15:23:35BippyDoes that mean you still have to implement the code or somthing into rockbox ?
15:23:52B4gderI'm 6'4
15:23:54BippyEasy or hard job ?
15:23:56LinusNbut we're close
15:24:10muzhey linus what you workin on right now?
15:24:21BippyIm 6'4 actually aswell, im about 6'5 if you include the hair lol
15:24:23LinusNmuz: mp3 playback :-)
15:24:32muzooo yay my favourtie
15:24:47B4gderheh, growing fast I see :-)
15:24:47muzdo you reckon we will see it any time soon?
15:25:04BippyWill the code be comitted to daily builds after its finished ?
15:25:10LinusNof course
15:25:16preglowwhy, it's a lot more fun if it just comes flashing out of the blue, isn't it?
15:25:24preglowcheck daily, it'll be there when it's out
15:25:50BippyIve been checking since it was announced you were doing somthing last year in the iriver forum
15:27:40muzhey what are the chances of video playback on the iriver?
15:27:51preglowthat is
15:28:00preglowdepends on what kind of video you mean
15:28:05muzreally, i mean from a hardware perspective
15:28:06ashridahit's unlikely to look decent.
15:28:13muzthe same res as the screen
15:28:17LinusNwe have video playback on the archos
15:28:19preglowand i mean what codec?
15:28:21B4gderdoing similar as on the archos is quite possible
15:28:30preglowif you mean xvid, then i sure as hell am not doing it, heh
15:28:33preglowbut it can be done
15:28:37preglowit'll look like shit... but can be done
15:29:11preglowLinusN: are the archos displays more responsive than the iriver ones?
15:29:23preglowthe iriver display is slow as hell even on warm days
15:29:28LinusNthe hardware communication is *way* faster
15:29:32LinusNon the iriver
15:29:42B4gderyes, the archos LCD updates faster
15:29:44 Join Nibbler [0] (
15:29:46LinusNbut the iriver lcd is a lot slower
15:29:48preglowwell, yes, the lcd controller is on the memory bus, isn't it?
15:30:17LinusNthe lcd itself is very slow
15:30:57LinusNso video on the H100 will be blurry at best
15:31:25LinusNfull frame rate, but blurry
15:31:36LinusNlet's hope the 300 lcd is faster
15:31:45BippyWhy the hell would you want video on an *MP3* player , dont even start with video
15:31:59LinusNbecause it's fun?
15:32:08B4gderrockbox already has video
15:32:23BippyFun ? :| hm ok then
15:32:57LinusNBippy: that almost sounded like the famous "rubbish" statement
15:33:09muzwhats video like on the archos's? does it look good with the faster lcd?
15:33:21LinusNas good as you would expect
15:33:23BippyAlmost, but not quite
15:33:56preglowa gameboy emulator isn't anything i'd expect on an mp3 player either, but if you can do it, why the hell don't
15:34:22ashridah"why the hell don't"?
15:34:24*ashridah blinks
15:34:40BippyWell apart from the fact the sound doesnt work and it doesnt look so good
15:35:03BippyFM radio would be nice
15:35:09*preglow hastens to add that he has given up on correcting his language
15:35:10BippyId use rock box all the time then
15:35:25B4gderbippy: join in and make it happen
15:35:35BippyIf i could badger i would
15:35:42preglowBippy: the stuff that needs doing isn't always what you want to do
15:35:50BippyIm happy to test things for you though
15:35:51***Saving seen data "./dancer.seen"
15:36:04preglowBippy: part of the fun with projects like these is the fact you can work on what you want
15:37:21 Quit muz ("CGI:IRC")
15:38:35preglowcodec optimisations for one isn't the coolest thing in the world
15:39:10BippyReally ? Sounds quite exciting to me
15:39:31LinusNnot if you suck at it, like preglow :-)
15:39:52Bippypoor preggers
15:39:54*LinusN hides
15:40:05preglowi wont oppose that
15:40:31BippyOmg a robin just pooped on a chair i just cleaned
15:40:45LinusNgotta go home
15:40:50LinusNcu folx
15:40:56BippyHome ?
15:41:02BippyYou work on rockbox at work ?
15:41:03LinusNi'm at work
15:41:14 Part LinusN
15:41:26 Quit ashridah ("Leaving")
15:43:56preglowBippy: if your idea of fun is rewriting entire functions in assembler while utilizing a half-baked emac extension, then yes! it's a blast
15:44:22preglowit's not the most boring thing i've done, but not the funniest either
15:45:55BippyI feel for you
15:46:18BippyIm hungry :| Where you live then preggers?
15:49:55B4gderStockholm Sweden here
15:51:05preglowi'm both thirsty and hungry, wonder what i'll do about that
15:52:38*preglow removes bottles of beer before temptation strikes further
15:56:55Bippyi hope u drank the beer before removing them
15:57:49bobTHCu have deplete good beer like belgian or dutch beer or somthing totally undrinkable ?
15:58:15preglowenglish ale at the moment
15:58:21preglowi'm quite picky about what beer i drink
15:58:41BippyCarlling, stell, guiness, grolsch, corona, san miguel
15:58:44bobTHCand ure right
15:59:18preglowthey're all ok, but the three last ones are a bit bland, if you ask me
15:59:18BippyNot forgeting the best of them all, Black Sheep Special Brew
15:59:22bobTHCstella = good beer ? lol
15:59:28preglowblack sheep special is very nice
15:59:39preglowhaven't had it since the last time i was in england, though
15:59:43BippySan miguel is nice with a meal in my opinion
15:59:51preglowit certainly is light enough
16:00:04bobTHCi prefer drink wine with a meal
16:00:08preglowright now i've got a spitfire i look forward to having, and a nice ruddles county
16:00:11BippyStell rocks
16:00:18preglowplus some other stuff
16:00:20BippyStella# i think my A is broken
16:01:05BippyIm not into wine much, but a merlot is nice
16:01:17preglowwine's ok, beer is better
16:01:28bobTHCwine is a gift of god
16:01:36BippyGod sucks
16:01:37preglowsure, as is war ;)
16:01:56BippyWhiskeys good for gettin smashed fast
16:02:00bobTHClolo-laptop: i'm atheist
16:02:08bobTHClolo-laptop: i'm atheist
16:02:11preglowwhisky is very nice
16:02:14preglowbut enough of this
16:02:16preglowi'm out
16:02:17preglowsee you all later
16:02:22bobTHCc ya
16:02:24 Quit preglow ("pläpläplä")
16:02:36BippyAhhh how i love radio 1
16:02:50bobTHCthe best radio for d&b live
16:02:57lolo-laptopbobTHC: hrm? −− I mean I am too... but ...?
16:03:15BippyIm atheist...
16:03:57bobTHCsorry, i have a bug with my irc client and i want to wrote a" lol "a the begin of my sentence and it became "lolo-laptop"
16:04:17 Nick lolo-laptop is now known as lostlogic (~lostlogic@
16:04:32bobTHCsurely the autocomplete nickname feature who sucks
16:04:32 Join webguest71 [0] (
16:05:54BippyRight then my little and large monkeys, im going
16:06:07*Bippy *Waves*
16:06:15 Quit Bippy ("CGI:IRC")
16:09:02 Join thegeek [0] (
16:09:47bobTHClol test ;)
16:16:37*HCl wonders why he's building his own rockbox build when he can just get the daily build..
16:19:21 Quit webguest71 ("CGI:IRC")
16:19:23 Join webguest71 [0] (
16:19:30bobTHCbecause u're nit-picking
16:19:36 Quit thegeek (Read error: 104 (Connection reset by peer))
16:20:46HCli remember
16:20:51HClbecause i want grayscale support.
16:25:10 Quit B4gder ("CGI:IRC")
16:30:34Rickis there something special I need to do to checkout lcd-h100-remote.c?
16:30:39Rickbecause cvs isn't checking out a copy
16:30:44Rick(I removed the original)
16:32:03 Join Shagnar [0] (
16:32:44HClnope, worked for me
16:32:51Rickthat's bizaare
16:32:54HClmaybe cvs update -d
16:33:05HClis there any test code for the lcd driver yet?
16:33:05RickI always update with -Pd
16:33:15Rickwell, I was about to put some together
16:33:15HCltry a checkout, maybe.
16:33:18Rickbut I wanted to checkout
16:33:45*HCl updates his on his ftp
16:33:48Rickno, that doesn't work
16:34:07Ricki'll start with a new thing then
16:34:09HCljust check out again
16:35:01bobTHCHCl > when u thing the Grayscale lib will be added to the dailybuild ?
16:35:17HClwhen markun gets to adding the lacking functions, i think.
16:35:20HCli haven't seen him in ages
16:36:01 Join Stryke` [0] (~Chairman8@resnet-241-86.resnet.UMBC.EDU)
16:37:17 Join thegeek [0] (
16:37:33HClnice, lots of updates since i last updated my rockbox
16:37:54HCldoes anyone know whether the lcd screen is black and white or also 2bit grayscale?
16:38:06Rickthe remote lcd?
16:38:08Rickjust black/white
16:38:22Rickit also appears to be sitting on it's side
16:38:27Rick(spec says 65xsomething)
16:38:55HClis the interface fast enough to do temporal dithering?
16:39:03Rickthat's not something I would know
16:39:25Rick(Linus was teaching me how to work with spi, which is where lcd-h100-remote.c came from)
16:40:13*HCl sighs and swears he'll get proper sleep tonight.
16:40:15Ricknow, to do a little experiment on the iriver remote
16:40:48Rickyeah... should be visible
16:40:53Rickif I manager to initialize it properly
16:47:59Rickoh nice
16:48:05RickI could probably make the constrast really high
16:48:09Rickto make it painfully obvious
16:48:18 Join austriancoder [0] (
16:48:55austriancoderrick: i will commit tonight my high-level remote lcd driver with your basic functions
16:49:09austriancoderso you dont need to write it again
16:49:14austriancoderbecause i have done it
16:49:17Rickaustriancoder: okay
16:49:21austriancoderbut at the moment i am at uni
16:49:24HCli'm wondering, does rockbox support an mp3 id3 tag database ?
16:49:33austriancoderso i can't test or commit anything
16:49:34 Quit rasher ("CGI:IRC (EOF)")
16:50:29austriancoderbut you can think about an remote lcd plugin api
16:50:47austriancoderor how we can conntect the remote driver with rockbox's core
16:51:01HCli dunno
16:51:11HClit depends on whether we want to generally support remote lcds
16:51:13HCland i think we do
16:51:47austriancoderi think a generall support is not a bad thing
16:52:00bobTHCindeed too, imho
16:52:05HCland whether we'd want to support even more lcds
16:52:28austriancodermore then 2?
16:52:39Shagnarcan anyone help me? i just opened my h140 because the HDD isn't working anymore, and i dunno which screw i've to open now to get the HDD =>
16:53:17Rickno change in remote lcd
16:53:34HCldid you flash it with rockbox? or why didn't you just call iriver warranty..?
16:54:08RickI put it in the wrong section
16:54:11Rick(for simulator)
16:54:15HCl :P
16:54:42*Rick wonders if it will explode
16:54:47ShagnarHCl: yes, flashed with rockbox ;)
16:54:56HCljust checking
16:55:07HClthat makes sense then
16:55:14HClthe person to ask about that is linus though
16:55:16Rickstill no change
16:55:27austriancoderrick: what are you trying?
16:55:28HClrick, austriancoder got it to work, i think
16:55:42Ricktrying to initialize the lcd
16:56:18austriancoderrick: ok.. have you uncommented the code in lcd-h100-remote.c?
16:56:28Rickuncommented which code?
16:56:52Ricknothing was commented except my little block at the bottom which was something I was working on before I went to sleep
16:57:20austriancoderat the botton
16:57:25austriancoderthe /*...*/ part
16:57:32Rickthat's what I was working on
16:57:52austriancoderah ok
16:57:58ShagnarHCl: kk. he'S who i'm waiting for... do you know when he'll be back?
16:58:07RickI am doing that - plus some other stuff
16:58:12Rickwhich is supposed to set the contrast to max
16:58:15HCli think he was saying he was gonna go home
16:58:19Rick(to make it obvious it works)
16:58:23HClso soonish
16:58:34austriancoderand what are you doing exactly?
16:58:43Ricksec, i'll show my additional lines
16:58:53Rick lcd_remote_write_command(0x81); // Set reference voltage lcd_remote_write_command(0x3F); // Max contrast
16:59:21Rickalthough it could be because I am missing the voltage set
16:59:25Rickbut I figured it had a default
16:59:33*Rick goes to find that
17:00:17 Quit webguest71 ("CGI:IRC")
17:01:24Rickstill no change
17:01:25*Rick hmms
17:01:34 Join XShocK [0] (
17:01:59austriancodermom = one moment
17:02:11austriancoderlook at page 40 of the lcd spec
17:02:22Rickthat's what I was following
17:02:32Rickwe start at 'user application'
17:03:06austriancoderare yu doning also some sleep?
17:03:29XShocKLinus, can you explain why that "PLLCR &= ~(1 << 22); /* Set AudioClk = FXTAL/2*/" makes an error? and why it makes double pitch
17:04:38XShocKwhat i though it should do is only set AudioClk period.
17:04:55austriancoderoh my god... only 5% left of my batteris... must log of now... sorr<
17:05:11 Quit austriancoder ("CGI:IRC")
17:05:41*Rick adds an LCD_REMOTE_DISPLAY_ON
17:05:57 Quit XShocK (Client Quit)
17:13:10Rickstill no change :/
17:25:35RickI think I found remote backlight
17:26:39 Quit tvelocity ("Leaving")
17:29:28Rickyup I did
17:35:52***Saving seen data "./dancer.seen"
17:42:12*Rick updated the portpinassignments page
17:44:22*Rick made it blink
17:47:46*HCl goes to play mario
17:47:56HCli'll look at dynarec when i manage to get some decent sleep
17:51:04Rickhow's dynarec coming?
17:52:53HClpartly working.
17:52:58HCli'm mostly worrying that it won't be any faster.
17:53:08HCldue to it not being able to work optimally without an assembly cpu core
17:53:23HClis the chip8 emulator working for iriver?
17:54:26RickI have a feeling it's running too fast
17:54:33Rickif you turn the contrast up
17:54:37Rickyou can see stuff happening
17:55:09 Join Sucka [0] (
17:56:44RickCould someone point LinusN to when he gets back?
17:56:54Rickoh wait
17:56:57RickI forgot my hosting is down
17:57:25HClyou can dcc it to me.
17:58:59bobTHCif u want to reach linus leave a memo with /memoserv
17:59:04RickbobTHC: good point
17:59:32HCldo we have some sort of id3 tag database at the moment?
17:59:55HCland how
17:59:56TangHI BOb
18:00:05HCland how does it get loaded?
18:00:06Rickgotta run
18:00:08TangHave some firefox matters
18:00:13Tangshould restart
18:00:15HClnicely in the background or crappily in the foreground like with the iriver firmware
18:00:39Tangnicely of course
18:00:45Tangwe're talking about Rbx
18:00:58Tanganyway it need the use of a small tool
18:01:06Tangbefore to create the DB
18:01:25 Quit Tang ("Chatzilla 0.9.66 [Mozilla rv:1.7.5/20041108]")
18:01:58RickHCl: taking a quick glance, the emulator doesn't even sleep() −− it yield()'s
18:02:13Rickrb->yield(); /* we should regulate the speed by timer query, sleep/yield */
18:03:49Ricki'm out
18:11:08 Nick Lynx_ is now known as Lynx_awy (HydraIRC@
18:11:36HCli don't understand the binary format stuff at all
18:19:03 Join t0mas [0] (
18:19:42t0masah... HCl? what university are you on?
18:20:05t0masoh ok, I was in Delft today
18:21:07*HCl looks around
18:21:23HCli dunno, i might be changing studies soonish. the math is killing me.
18:22:09t0masyou are doing math??
18:22:10elinenbemath killed me at uni... but I got through it...
18:22:18HClcomputer science
18:22:24t0masah ok
18:22:28elinenbeyou'll never need that stuff later though... pretty much no matter what you do.
18:22:32HClwell, i just get way too frustrated over math.
18:22:33*t0mas is thinking of aerospacde design
18:22:34HCli know that.
18:22:37HClbut i can't pass it.
18:22:47HCli tried this same course 3 times
18:22:49HClover 3 years time.
18:22:52HCland i still fail at it.
18:22:55HClso i'm giving up, :/
18:22:59t0masstatistics? :)
18:23:09HClcombinatorial math.
18:23:16t0masah.. I heard that was the worst...
18:23:19HClgenerating functions, recurrant equations etc.
18:23:23bobTHCas bob said :dont give up the fight !
18:23:30HCltoo late, heh.
18:23:43HCli'm just gonna refuse doing anything that stresses me *that* much.
18:23:52HClyou need to enjoy what you do.
18:23:54HClnot hate it.
18:24:30*HCl sighs.
18:24:47HClbleh. anyways.
18:25:11t0maslot of cvs activity today :)
18:25:13*HCl thinks he'll think a bit about the tag database stuff.
18:25:26 Quit Sucka ("a bird in the bush is worth two in your house")
18:25:27elinenbeHCl: enjoy rockbox....
18:25:41elinenbecombinational math is a waste of time anyway!
18:25:47elinenbedo you need the class to graduate?
18:25:56HClin fact
18:26:03HClif i can't pass it this year (which i can't)
18:26:08HCli'm not allowed to do any other courses whatsoever
18:26:28t0mashmz... time to change study :)
18:26:29bobTHCfucking system!
18:26:43HCli'm most likely gonna drop down to college stuff.
18:26:47HClnot 100% sure yet
18:28:05 Join Sucka [0] (
18:28:48bobTHCif u do that remember that and u'll never be trapped : "if you take care of a problem while it's small you won't have a bigger problem to deal with later"
18:28:54*t0mas is away: eating :)
18:29:33HCli guess
18:29:47HCli'm more a person of the mentality "whatever happens, happens."
18:30:02elinenbeHCl: time to change subjects... what are you doing with the tag database?
18:30:09HClnothing yet
18:30:14HClbut i don't understand the current binary layout
18:30:19HCland i'm gonna try to think of a better one.
18:31:18HClbbl, food
18:31:21ferenczysorry, for I'm discontinuing your discussion, but I have got a question> does anybody have the Xclef MT-500 player???
18:32:22bobTHCiirc some ppl start to work on this device
18:32:54elinenbeHCl: what about the default one that the iriver uses? is that superior or inferior? there is a description of that in the wiki.
18:35:18 Join LinusN [0] (
18:35:28bobTHCelinenbe >
18:35:42LinusNShagnar: any luck?
18:36:21 Quit xen` (Read error: 110 (Connection timed out))
18:41:50 Join preglow [0] (
18:45:58HClelinenbe: there is?
18:47:22HClelinenbe: where is the description of that?
18:48:39bobTHCbye all!
18:48:48 Quit bobTHC ("Smoke Weed Every Day !")
18:48:53HCl*looks at it*
18:49:02HCli think its safe to say the iriver format is even more crap o.o
18:49:59HClit just lists the id3 tags for every file
18:51:47 Part LinusN
18:53:20HClyea, i found it, somewhat.
18:53:50HClits safe to say the iriver format is even worse. it simply has no tables
18:54:05elinenbewhat about the ipod format?
18:54:13elinenbethat updates real time...
18:54:16elinenbeand quickly
18:54:33elinenbelike you can easily have a playlist that is "songs from the 90's over 3 minutes long"
18:54:40elinenbe"songs with eric in the name"
18:54:47HClsounds good.
18:54:49elinenbe"top 25 most played songs.
18:54:59elinenbe"top rated songs"
18:55:01HClsounds worth looking into
18:57:01elinenbethat's a good beginning source
18:57:23 Join Tang [0] (
18:57:41elinenbehere is some excellent documentation:
18:58:10CoCoLUSi bought a car radio, and it even has a usb plug
18:58:16CoCoLUSi tried rockbox with it :)
18:58:40HClusb plug?
18:58:42HClin a car radio?
18:58:57CoCoLUSyeah you can plug in any usb mass storage device and it plays the files
18:58:59CoCoLUSproblem is
18:59:08CoCoLUSthe radio only supports up to 1gb devices
18:59:15CoCoLUSsoooo... it won't play anything from the iriver
18:59:17CoCoLUSany thoughts?
18:59:19HCli have a pseudo-tape with an audio line in
18:59:23HClwhich i just plug into my iriver
18:59:29HCland i stick the remote on my car dashboard
18:59:36HClvoila, works like a charm
18:59:38HClfor 15 $
18:59:52Tanghi Cocolus
19:00:08HClit'd even get more battery lifetime cause usb connection keeps hdd spinning ;/
19:00:20Tangyeh cool stuff the USB port on yuor radio car
19:00:59CoCoLUSdoes rockbox recognize multiple partitions?
19:01:13CoCoLUSbecause if i would size the first partition down to < 1gb, maybe the radio would support the device
19:01:22HClno it doesn't.
19:01:50HClelinenbe: i'm looking at
19:02:11HCli'm thinking that it might be a good idea to completely adopt the itunesdb? though i'm not aware of any flaws the format has
19:03:12Tangwhat syou radiocar name?
19:03:34 Join asdsd_ [0] (
19:03:40 Part asdsd_
19:04:36CoCoLUSi don't understand
19:04:51HClthe only problem with itunesdb is that it seems fairly cpu intensive...
19:09:12amiconnCoCoLUS: Rockbox *can* support multiple partitions. This is just not enabled for all hard disk devices. It is used on Ondio, which has both internal flash and an external MMC option. Rockbox can access both :)
19:09:43CoCoLUSi take it the original firmware only uses the first partition?
19:10:36amiconnHCl: The rockbox id3db format has another severe constraint: it needs to have a fairly small code footprint. The current format needs about 4 KB of code - how much would supporting itunesdb take?
19:10:55amiconnCoCoLUS: Guess so; (still) don't have an iriver :(
19:10:56HClamiconn: too much. i was mostly considering having a different format available on iriver
19:11:09preglowwhat'd be the pros of doing that?
19:11:21HCla heck of a lot easier access to whatever you're wanting.
19:11:34HCl18:52 < elinenbe> and quickly
19:11:34HCl18:52 < elinenbe> like you can easily have a playlist that is "songs from the
19:11:34HCl 90's over 3 minutes long"
19:11:34HCl18:53 < elinenbe> "songs with eric in the name"
19:11:38HClstuff like that.
19:11:38preglowbear in mind i haven't tried the rockbox one
19:11:48HClneither have i, i'm just looking at our current format.
19:11:57amiconnTo be honest, I don't really need id3 browsing. Structuring your music collection properly is the key ;)
19:12:00HCland i know there's no way in hell it'll ever be able to do stuff like that.
19:12:43preglowamiconn: you can't organize properly for all needs
19:13:01elinenbeHCL: do you have a windows or mac box?
19:13:18amiconnI can organise properly for all *my* needs. That's obviously sufficient for me :)
19:13:18*t0mas is back :)
19:13:22elinenbetry it in itues...
19:13:26elinenbecreate a smart playlist.
19:13:38elinenbeplus there is a lot of open source code for this as well.
19:14:10 Nick elinenbe is now known as elinenbe|lunch (~elinenbe_@
19:15:09HCli'd really enjoy the flexibility of what elinenbe just said..
19:15:12amiconnThe only thing I would possibly use a database for is tracking how often I listen to what songs/albums. That's something the current db format isn't designed for
19:15:31preglowaudioscrobbler support!
19:15:39HClwhats audioscrobbler?
19:16:13HCltoo bad its not opensource
19:16:17HClat least
19:16:22HClnot in a glance that i looked over it
19:16:36preglowthink they don't want people to reverse the protocol
19:16:51preglowsince false submissions would be a problem then
19:16:51amiconnI wouldn't put such info on any device/server/PC that isn't under my private control for sure.
19:17:12preglowwhat info, what music you listen to?
19:17:51elinenbe|lunchaudioscrobbler is GPL :
19:18:45CoCoLUSThe protocol is HTTP based, simple & lightweight. You can implement it on any device that has web access. Read Protocol 1.1 (current) (New, RFC: Protocol 1.2) for the gory details. Existing plugins are all open-source, so you don't need to start from scratch.
19:18:53CoCoLUSsounds quite open-source to me
19:19:29 Quit einhirn_ ("Miranda IM! Smaller, Faster, Easier.")
19:23:06elinenbe|lunchHCl: you implement the ipod database format and I'll buy you one beer.
19:23:14elinenbe|lunchwell, make that a 6-pack
19:25:04 Join Bippy [0] (
19:25:09preglowyou better make that sake
19:25:20BippyMy river crashed
19:25:23Tangand for implementing RioDJ (karma DB)
19:25:36Tangyou offer him whisky?
19:25:47TangWhat's hapened Bippy?
19:26:08BippyIts stuck on Starting original firmware, tryin to find a pin now
19:26:38preglowbattery level is ok?
19:27:22amiconnStrange... Watching iriver H1XXs on ebay germany for quite a while. There were only a few ones in the past. All of a sudden theres an iriver H1xx flood...
19:28:20Bippyaw crap cant find a pin
19:28:24preglowamiconn: still haven't got yours?
19:28:30BippyWhy the hell did they make such a small reset hole
19:28:30amiconnNope :(
19:28:35Tangi thought Rbx should create iHP1xxmania
19:29:15amiconnUsually I have 2..4 iriver units on my watch list. Now there are almost 20...
19:30:02BippySome fallen off a back of a truck
19:33:42BippyOmg it wont let me start the iriver firmware
19:34:22pregloware you absolutely certain about the battery?
19:34:43preglowwhere does it hang?
19:34:54HClsorry, responding to elinenbe
19:34:57BippyStarting Original Firmware
19:35:17preglowdoes the red led blink?
19:35:28preglowbefore it hangs, that is
19:35:31BippyLoads rockbox though, ill plug the mains cable in and make sure
19:35:54***Saving seen data "./dancer.seen"
19:36:15BippyIve got a low battery
19:36:21preglowcharge it
19:36:23preglowthen try again
19:36:26BippyI am now
19:36:28preglowiriver firmware is very picky about that
19:36:42BippyRockbox still says i have 25 left
19:36:55preglowwell, rockbox' notion of a good battery level differs from iriver's
19:37:27t0masa lot :)
19:37:56BippyFinally i can listen to my radio again
19:38:09preglowit starts now?
19:38:15*Bippy dances to music
19:38:22preglowyes, i expected that
19:38:45BippyYou so need the FM radio in the rockbox :'(
19:38:54preglowget us someone to code it
19:38:55BippyI wont need iriver then
19:39:12BippyOk ill go find wheller
19:39:40BippyWhat language does he need to know ?
19:42:52 Quit thegeek (Read error: 54 (Connection reset by peer))
19:46:17t0masbuild error's
19:46:20t0maslatest cvs
19:46:44BippyErrors dont sound good
19:46:49t0maswavpack.h:497: error: parse error before '->' token
19:46:50t0maswavpack.h:519: error: parse error before '->' token
19:46:50DBUGEnqueued KICK t0mas
19:46:50t0maswavpack.h:521: error: parse error before '->' token
19:46:50t0maswavpack.h:525: warning: type defaults to `int' in declaration of `value'
19:46:50***Alert Mode level 1
19:46:50t0maswavpack.h:525: error: `bs' undeclared here (not in a function)
19:46:50***Alert Mode level 2
19:46:50t0maswavpack.h:525: warning: data definition has no type or storage class
19:46:52t0maswavpack.h:527: error: parse error before '->' token
19:46:55t0maswavpack.h:529: error: parse error before '->' token
19:47:00Bippyerrors = bricks and bricks = me shit a brick
19:47:27preglowt0mas: daily builds claims there are no broken builds
19:47:36t0masjust checked
19:47:42preglowand i tend to believe it :P
19:47:50BippyI hate phones
19:49:37BippyOMFG phone agen !
19:50:01Shagnaryes! changed HDD, my player works again :-D
19:50:26t0maspreglow: in wavpack.h around that lines: #define getbit(bs)
19:52:27Bippysorry bout caps
19:52:42Shagnarpull out the cable ^^
19:52:54BippyI want have internet
19:53:35Shagnarif you don't have it, how are you online then?
19:53:54BippySome random guy knows small amount of C, would he be able to help with FM radio /
19:56:51***Alert Mode OFF
19:57:03preglowi don't think so
19:57:15preglowyou need to be pretty proficient in c
19:57:21preglowt0mas: i have no idea, works here
19:57:24preglowand now i need to reboot
19:57:54t0masc ya
19:58:33BippyYou could teach him preggers
19:58:58BippyHe knows VB and some others
20:00:31preglowin the time i would need for teaching him, i could have written it myself
20:00:34preglowno time anyway
20:00:36t0masI know VB too...
20:00:44t0masAnd some C and C++
20:00:53t0masBut I'm sure I can't just code radio support
20:01:19t0masthe problem with those things is that you need to know the device...
20:02:44TangBippy: cool it's solved
20:03:46Tangyour HD was dead?
20:03:53Tang(you're from MR?)
20:04:00Shagnaryes :)
20:04:07Tangokay you bought bigger HD?
20:04:36Shagnarno... I borrowed the broken player from a friend and changed the HDDs. now my H is running again:)
20:05:03CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
20:05:03*Shagnar is very happy ^^
20:05:06Tang(what juke he had?)
20:06:05Shagnarhad 140, brandnew but not working... he's too [faul] to bring it to the post so i asked him if i could change the HDDs
20:07:09HClthats a bit dumb of him...
20:07:16HClcause now you voided his warranty, most likely..
20:07:32Shagnardon't think so..
20:07:56HCli think your warranty is void when you open it up
20:08:38Shagnarthere are no secure-screws for that, so if you are carefully when opening it, that doens't matter
20:08:58Shagnari'm really lucky, else i would have got a H340 i think
20:11:11HClwell, you could just buy a replacement hdd.
20:11:12preglowi really don't think they can tell if you've opened it
20:11:24preglowbut yes, sells 30 gig hds now
20:11:27HClmaybe not, but technically, it would be voiding your warranty
20:11:46 Quit gromit` (Read error: 104 (Connection reset by peer))
20:13:03 Quit Bippy ("CGI:IRC")
20:13:14 Join gromit` [0] (
20:14:19 Join matsl [0] (
20:14:36 Join rasher [0] (
20:18:29Shagnaryes it would
20:19:10Shagnari asked someone who often opens Hs and he said me if i work good, the warranty shouldn't be void
20:19:46Shagnaryes i could buy a repl. hdd, but at this time, i didn't know if the HDD itself was broken or perhaps just another IC or something
20:20:16Shagnar@ HCl ;)
20:22:06Tangeh there is nothing to proove it have been opened
20:23:42 Join stevenm [0] (
20:23:49stevenmhello people
20:24:41HClokay, i'm looking at the itunesdb format
20:24:45HCland it has a few redundant fields
20:24:52HClso we might add something quite similar to it
20:27:58rasherwhy not copy it wholesale, to keep up with eventual updates easily?
20:31:02t0masit looks good..
20:31:03HClcould be possible.
20:31:11HClit would add *SO* much to iriver usability
20:31:30HCli'm definately gonna look at it a bit
20:31:42HClthough i somewhat need someone to explain how it works for rockbox at the moment
20:31:46HClregarding queues and playlists
20:31:50HCland browsing
20:31:53t0masif you have it working.. I can write a plugin for winamp.. to update the playcounts? :)
20:33:43Shagnarhmm i have to go now, but if linus returns, please let him know that my H works again :) thx!
20:33:46 Nick Shagnar is now known as Shagnar^afk (
20:35:06Tanghum i have a question
20:35:09Tangwhat mean CVS?
20:35:17HClconcurrent version... system?
20:35:18HCli think
20:35:21t0masConcurrent Versioning System
20:35:29Tanghum okay
20:35:49Tangsomeone told me "current version stable"
20:35:50t0masthe second one ;)
20:36:31Tangi was trusting this for a while
20:39:56 Join thegeek [0] (
20:40:54 Part Sebulba02 ("Why am I here?")
20:42:10HCllibitunesdb isn't very useful...
20:42:16HClit doesn't even allow searching.
20:48:14HCl*talks to ipodlinux people*
20:49:14HCl :P
20:49:25HCli just use whatever's handiest.
20:50:10HClpretty much, i'm gonna compile their x11 simulator
20:50:22HClthen pluck their podzilla apart till we have the ipod's ease of use
20:51:13preglowwont get that without the rotary controller thing
20:51:17preglowwhich i absolutely love
20:51:22preglowit's by far the coolest thing about the ipod
20:51:34*rasher tries updating the snap font again
20:52:18HClmaybe not.
20:52:25HClbut their database is gonna be nice either way
21:01:20rasherhurray, this font editor is actually responding to what I tell it to do
21:03:27 Join cYmen_ [0] (
21:03:54 Quit cYmen (Read error: 104 (Connection reset by peer))
21:04:16stevenmhaha any chance of us doing an IPOD port? :)
21:04:43rasherit'd certainly be possible, assuming people wanted to do the work
21:04:57stevenmyou're kidding.
21:05:05rasherAm I?
21:06:22elinenbe|lunchHCl: how long before the itunes format is on the iriver... I give you 1 week!
21:06:31elinenbe|lunchremember a 6-pack is on the line :-)
21:07:20preglowwhy shouldn't that be possible?
21:07:26preglowif ipodlinux guys can do it, so can we
21:07:40HCli don't like beer :P
21:07:40 Quit thegeek (Read error: 131 (Connection reset by peer))
21:07:41rasherThey even done a lot of work
21:07:48preglowelinenbe|lunch: i'll take that
21:08:03preglowelinenbe|lunch: what itunes format?
21:08:16preglowahh, db
21:08:19preglowthought you meant aac
21:08:22HClits a id3 tag db
21:08:36HClthere's opensource stuff available from the ipodlinux project
21:08:42preglowi could have aac running in a week, if i were to be payed :PP
21:08:53HCland the db is superior by far compared to iriver's db
21:08:58HCland our low-profile rockbox db
21:09:03pregloweverything is superiod to the iriver db
21:09:06 Nick elinenbe|lunch is now known as elinenbe (~elinenbe_@
21:09:07preglowwhich is a pile of shit
21:09:14stevenmipods are pretty popular....
21:09:18 Join thegeek [0] (
21:09:41elinenbepreglow: the apple lossless format has been open sourced...
21:09:50preglowwell, yes, the lossless one
21:09:59preglowwhich i couldn't care less about, heh
21:10:03 Join gromit`` [0] (~gromit`
21:10:14elinenbeHCl: what do you like?
21:10:22elinenbeyou need a nice woman for a night?
21:10:23preglowaac is mildly interesting
21:10:57elinenbepreglow: isn't there an open source aac decoder?
21:11:05preglowelinenbe: why, there are several
21:11:08stevenmhey preglow.. i am almost done converting the midi file io to posix... is there a function to convert string to int in rockbox.. atoi or whatever?
21:11:10preglowelinenbe: faad2 is what i'm looking at
21:11:20elinenbewell, you should implement it... if it is not hard.
21:11:22preglowstevenm: dunno, have a look in the headers
21:11:32stevenmer file loader
21:11:34preglowstevenm: why do you need that, btw?
21:11:39rasherthere is an atoi
21:11:48rasherin the plugin api
21:11:49stevenmparse the drum patch config file
21:12:05rasherstevenm: I told you docs/PLUGIN_API ;)
21:12:15stevenmvery dirty parser but for now..
21:12:20stevenmrasher, ah, right
21:12:31HClelinenbe: i have a gf, thank you very much :p
21:13:04 Quit gromit` ("Client exiting")
21:13:04 Quit gromit`` (Remote closed the connection)
21:13:26elinenbeHCl: so what'd you need?
21:13:42 Join gromit` [0] (~gromit`
21:13:51 Quit gromit` (Remote closed the connection)
21:14:50 Join gromit` [0] (~gromit`
21:14:53 Quit gromit` (Remote closed the connection)
21:15:10HClwell, i'd like a bucket of money :P
21:17:20 Join gromit` [0] (~gromit`
21:17:21 Quit gromit` (Remote closed the connection)
21:17:40 Join gromit` [0] (~gromit`
21:19:23 Quit gromit` (Remote closed the connection)
21:20:49 Join gromit` [0] (~gromit`
21:20:53 Quit gromit` (Remote closed the connection)
21:21:09 Join gromit` [0] (~gromit`
21:21:13 Quit gromit` (Remote closed the connection)
21:22:00HClthats annoying.
21:22:12HClthe ipodlinux project is saying the helix aac decoder is great
21:22:22HCljust in case anyone cared
21:22:30rasherpity we can't use it
21:23:32HClwhy can't we?
21:27:43rasherthe license isn't gpl compatible
21:30:22 Join gromit` [0] (~gromit`
21:31:03HClshouldn't ipodlinux have had trouble with it then.?
21:31:25rasherI'd imagine so
21:31:25preglownot all people care as much about licensing issues
21:31:45HClforgive me for sort of agreeing with ipodlinux a bit then..
21:32:08preglowi think that's far too important a thing to ignore
21:32:20preglowi'd be pissed if someone blatantly violated the license on something i wrote
21:32:49preglowfaad2 looks nice
21:33:01stevenmpreglow, okay, all IO converted to posix
21:33:16stevenmpreglow, how do you recommend I best try making this into a plugin ?
21:33:28preglowstevenm: just study the xxx2wav boys
21:33:33preglowi don't know how your api works
21:33:52stevenmI know that.. I was wondering, move all the files into apps/plugins or apps/plugins/midi or what
21:34:12preglowwell, no, the codec itself should be in apps/codecs/mymidithingy
21:34:19preglowand midi2wav should be in apps/plugins
21:34:43preglowHCl: why did they go with the helix decoder anyway?
21:34:51stevenmah.. then in my includes, I specify, include "../codecs/midi/synth.c" etc ?
21:35:04HClpreglow: i dunno, ask them
21:35:57***Saving seen data "./dancer.seen"
21:36:31rasherso this font editor will happily load bdf fonts and NOT SAVE AS BDF AGAIN
21:37:28*ferenczy is away: out of matrix
21:39:27HClthats a new one.
21:40:16rasherHurray! it exported to bdf
21:40:38rashernow to see if it actually fixed the errors I was trying to correct
21:42:07amiconnrasher: What strange font editor do you use?
21:42:49amiconnWhen I did 2 fonts for rockbox, I used pfaedit
21:43:44amiconn...under cygwin/x11. Was crashing pretty often with the old xfree86 on cygwin
21:44:17amiconncygwin switch-over to xorg-x11 magically fixed that
21:44:34rasherhah, small error in the font now - spaces are rendered as exclamation points
21:45:02stevenmrasher, off-by-1 with ASCII values? 32 vs 33 ?
21:45:15rasherheh, likely
21:45:25preglowthat helix decoder is swift
21:46:11 Join Tang_ [0] (
21:49:33 Quit Tang (Read error: 110 (Connection timed out))
21:50:06Tang_lol nice smiley rasher
21:50:12 Join Wingk [0] (
21:50:19Tang_i found the syntax
21:53:46rashercrazy.. now this I don't understand.. how did space turn into "!"
21:53:51HClgeeze, they can't even play mp3...?
21:54:03preglownah, think they're a bit more behind on the codec front than we
21:54:20HClhaven't they existed longer
21:54:37preglowyeah, but they slept for a good while, i think
21:54:51HClthats just what i'm gonna do too.
21:55:27*rasher punches various font editors
21:55:34HClah. they only know how to use one core, heh.
21:55:59Tang_good night
21:56:25Tang_HClgeeze, they can't even play mp3...?
21:56:31Tang_-> iPODlinus?
21:56:53Tang_seems it can play mp3 but using console mode
21:56:59Tang_not very ptractical
21:59:37 Join einhirn [0] (
22:02:19 Quit Wingk (Read error: 60 (Operation timed out))
22:04:39rashergot it.. I think
22:04:52rashercharacter 32 wasn't defined :X
22:05:06rasherGuess that'd be a problem
22:05:16 Join MoosCamaro [0] (
22:05:33Tang_hi Moos
22:05:56MoosCamarohi all
22:06:00MoosCamarohi Tang
22:06:19Tang_private for french discuss
22:07:13stevenmWoah, no COMPILER errors !!
22:07:24stevenmbut a like, 6 screens of linker errors
22:10:55amiconnreboot, brb
22:10:58 Part amiconn
22:14:50 Join asdsd_ [0] (
22:15:29rasherhurray,"> .. it's done!
22:15:47 Join belgarath [0] (
22:16:00 Join tvelocity [0] (
22:16:10belgarathhi all
22:16:13preglowarghh"> for the font
22:16:24stevenmpreglow, midi COMPILED
22:16:26preglowplease, god, don't let my lust for arm programming experience make me buy an ipod
22:16:59preglowstevenm: great, now make the midi2wav.c :-D
22:17:42Tang_Hum what's the PNG?
22:17:44stevenmpreglow, well.. the actual engine compiled. I now have to add the calls to write the wave file. Otherwise it all sits in a buffer in memory. And, I have no idea if this malloc() will work. I did get it from rockboy though..
22:17:55Tang_a new font?
22:18:04Tang_hi belgarath
22:18:06rasherTang_: a font I found and added some characters to
22:18:18Tang_okay thx rasher
22:18:56belgarathhi tang"> is a rockbox-version of it
22:19:15Tang_nice i'm gonna DL it
22:20:00rasher(place the bdf in fonts/ if you build rockbox yourself)
22:20:24Tang_mm i place it in font folder no?
22:20:36Tang_(i don't build it myself of course)
22:20:40rasheryeah, place the .fnt in .rockbox/fonts/
22:20:42Tang_i use the fnt?
22:20:49Tang_what i thought
22:20:55rasherthe bdf is of no use to the final rockbox
22:21:40Tang_just for devlopers?
22:22:01rasheror for anyone compiling rockbox themself
22:23:12 Nick Shagnar^afk is now known as Shagnar (
22:23:17Shagnarre :)
22:23:17stevenmwoah, midi loader ran on iriver.. at least got as far as loading the file properly
22:23:30Tang_hum okay so the fnt won't work with actual build
22:23:39Tang_since the bdf hadn't been implemented
22:24:00rasherTang_: oh no.. it's just that rockbox compiles .fnt from bdf files
22:24:34Tang_but if there is a fnt the bdf is useless no?
22:24:53rasherthey're both useful
22:25:23rasher.fnt is the final product
22:26:14rasherbdf is what the fnt is made from
22:26:16Tang_but the fnt work fby itself or not finaly?
22:26:17HCl (4:2) only where the floor is not type #,
22:26:23HCldarn mispaste
22:26:25HCli hate putty pasting
22:26:26Tang_ah okay i understood
22:26:26HClon click
22:26:37Tang_sorry if i was hard
22:26:52rasherthe fnt is what rockbox needs
22:27:41Tang_okay in fact before youir last post
22:27:51Tang_i didn't understood the fnt was made from the bdf
22:28:08stevenmHey guys.. does the iriver Simulator run as fast as possible, or does it try to run as fast as the actual target ?
22:28:13Tang_tht's why i was wondering about the bdf usage!
22:29:11rasherstevenm: I think it just runs as fast as possible
22:29:58stevenmrasher, ah, ok
22:31:02Shagnaris there a tool to easy make own fonts?
22:33:02rasherShagnar: linux or windows?
22:33:23Shagnarcould be both, but i'd prefer windooze
22:33:42rasherdon't know of any for windows
22:33:54Shagnarbut for linux?
22:34:10rasherFontforge, xmbdfed and pfaedit for linux
22:34:20 Join Strath [0] (
22:35:14rasherstevenm: definately runs as fast as possible.. vorbis2wav ran at 3000%
22:36:25stevenmwoah, midi works on simulator
22:36:40Shagnartry it in real, would be kinda nice ^^
22:36:46stevenmdoesnt write to a wav file (but rather a raw file, then opened with goldwave) but it worked apparently. file contains music
22:37:51 Quit t0mas ("good night")
22:42:01Shagnarvery fine :)
22:43:59 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
22:46:17belgaraththanks a lot for all you've done on teh project
22:46:25belgarathits great
22:48:16Tang_great althought it's not finished
22:48:19Tang_eh eh
22:48:44Tang_Rbx has great supporters
22:48:58Tang_what will it be when achieved
22:50:59CoCoLUSthey all will actually -use- it :D
22:51:58Tang_all but one
22:52:01 Quit asdsd_ (Read error: 110 (Connection timed out))
22:52:11Tang_someone i know
22:52:21Tang_i'll kick him if he use Rbx
22:53:23stevenmwhy is that?
22:54:19Tang_hum sort of private joke for Moos and some others french guys
22:54:37Tang_about a guy who make me a noise
22:54:47Tang_since i was supporting a prealpha
22:55:03Tang_(untested according to him)
22:55:34belgarathit hasnt bricked my player yet (touch wood)
22:56:57Tang_it didn't bricked any iHP till now
22:58:47rasherit's no less untested than the latest iriver firmware at the time of release
22:58:56rasherno less tested
22:59:15 Quit stevenm ("Leaving")
23:01:05 Join amiconn [0] (
23:07:02Tang_it's my opinion
23:07:13 Join Zombiejeff [0] (
23:07:46Tang_i don't avice noobs to upgrade with Rbx for no reason
23:07:47Bagder <= the swedish magazine article, now scanned
23:08:13belgarathwow cool
23:08:38belgarathis that a well known magazine?
23:08:40Tang_yeah you're famous
23:08:56 Join einhirn [0] (
23:08:59Bagderbelgarath: fairly I think, its about phones, mp3 players and stuff like that
23:09:51*preglow employs his swedish skillz
23:10:10*rasher tries to do the same
23:10:21rasherI hate when scanners do that trick
23:10:24rashermoire hell
23:11:12Bagderbut I didn't care that much about the quality of the photo
23:11:16Tang_anyone to translate?
23:12:29rasherisn't something missing?
23:12:43rasheroh nevermind
23:12:45rasherI'm an idiot
23:15:41preglowthese guys found you and asked for an interview, or?
23:15:59Bagderhe contacted us
23:16:42belgarathquite the celebrities now
23:17:05ZombiejeffHi all −− I've got a question about the code that I think should be pretty simple to answer...
23:17:27Tang_(when we begin like this
23:17:27preglownext time there'll be a ten-page exclusive, with the magazine shipping all of us two sweden on a two-day binge
23:17:33preglowZombiejeff: shoot
23:17:38Tang_then the question following will be very hard)
23:17:40preglowtwo = to
23:18:07ZombiejeffI'm trying to get my recorder to tell me what I was listening to when it resumes but I can't figure out is rockbox stores that info in a variable or not.
23:18:42Tang_hey it talk about the iHP 120 port too!
23:18:56preglowTang_: yes, it says that's their new platform
23:18:59ZombiejeffI am not much of a programmer −− except in old AtariBasic so I'm trying to stumble around code I don't know well.
23:19:45Tang_okay, strange but is tal about v2.3
23:19:53Tang_and you're at 2.4 version no?
23:20:06BagderZombiejeff: it doesn't actually store it so that it is available in the resume screen, no
23:20:12BagderZombiejeff: but you could fetch it
23:20:26preglowTang_: the article is a couple of months old
23:20:42Tang_ah okay
23:20:58Bagderthe interview was made before christmas
23:21:06ZombiejeffThe resume is in tree.c, right. So I have to pull it from another file?
23:21:55BagderZombiejeff: resume info is a playlist index number and a playlist name, so you need to get the actual file name from the playlist
23:22:09Bagderand then possibly extract id3 data or whatever
23:22:40ZombiejeffEven just the m3u filename would be useful as I almost always use playlists...
23:23:03Bagder_that_ is part of the resume info
23:24:39belgarathquestion: will the iriver rockbox support the .rvf files?
23:25:27rasherHeh, Linus's first comment seems totally over the head of the journalist
23:25:43pabsdid i miss an interview?
23:25:50ZombiejeffSo I have to pull the file from an include? Again, I've only programmed in basic before, so I'm asking some really basic questions.
23:25:55rasherIn a swedish magazine, Bagder scanned it
23:26:13rasher in case you read Swedish :)
23:26:22pabsnope, i'll bust out babelfish though :D
23:26:32Shagnaris there a solution yet for the bug in boot-loader enabled 1.65 FW? (oggs not played)
23:26:34Bagderbabel doesn't speak swedish
23:26:34pabserr no i won't
23:26:34rasherthat won't do you much good
23:26:44 Join muesli- [0] (
23:26:46pabsi just realized it's scanned, not OCRed
23:26:48rasherplus it's an image :)
23:26:48pabswell lemme see
23:26:51Bagder... and they are images
23:26:57preglowZombiejeff: include files has nothing to do with this
23:27:04preglowZombiejeff: you really should learn c :)
23:27:24Zombiejeffpreglow: yep, this may be my start!
23:27:30rasherevening muesli-
23:27:34preglowZombiejeff: the information you have when you see the resume screen, is a playlist name, and a entry number
23:27:37Tang_you have a mail
23:27:41muesli-hi rasher
23:27:47muesli-hossa tang
23:27:48preglowZombiejeff: so you need to open that file, and fish out the entry with the given number
23:27:50rasherso Bagder.. how about new fonts.. any criteria?
23:27:53pabsman everyone is famous his week
23:27:59preglowZombiejeff: then you'll have the filename of the track that is about to be resumed
23:28:16Bagderrasher: no limits really, just that the license allows us to use it
23:28:49belgarathsee you later
23:28:51 Quit belgarath ("CGI:IRC")
23:29:21ZombiejeffMaybe I should pick up a "C for dummies" book on the way home. Or is there a "Rockbox for dummies" book?
23:29:27rasherBagder: alright, this one's GPL (pulled from Debian and modified)
23:29:32preglowheh, 'fraid not
23:29:40preglowbut just learning c should do nicely
23:29:54Bagderrasher: then go for it!
23:31:00amiconnpabs: Let's see what ocr can do with this...
23:31:03rasherdammit... problems with my cvs again
23:31:12rasherI really should do a fresh checkout at some point
23:31:19muesli-btw wth are cvs!?
23:32:07Tang_lol i asked it before muesli
23:32:27preglowhaha, someone are operating searchlights somewhere, i'm getting a real batman vibe here
23:33:24rasherBagder: done... I think
23:33:28muesli-just ask robin 4 help :D
23:33:29rashernot quire sure if I got the adding right
23:34:13rasherLooks like I did
23:35:06preglowyou just add, then commit
23:35:21rashergot a bit confused by the message add gave me
23:35:31BagderI got a snap.bdf
23:35:47Shagnargn8 guys
23:35:54preglowgneight? :P"> < the font
23:36:02***Saving seen data "./dancer.seen"
23:36:10 Quit Shagnar ("rockBoxRocks")
23:36:17 Join stevenm [0] (
23:36:25rasherguess I should send this to Debian as well
23:36:35stevenmHello. Is there a way to write from the simulator raw to the computer's DSP ?
23:36:54rashersince they only included and a few others of the high-bit characters
23:37:01stevenmie, how do I tell the Sim to not try to open the sound device, then access it via symlink ?
23:37:54rasherspeaking of.. there was a patch that added sound support on linux.. I'm very curious about that
23:37:59stevenmah crap.. never mind. you'd need to somehow do IOCTL on it
23:37:59rasherdidn't apply though :(
23:38:14stevenmI made a symlink from /dev/dsp to ...../archos/dsp
23:38:26Bagderrasher: I think that needs to be remade to fit with the sound api anyway
23:38:38stevenmI hear NOISE because the DSP isnt initialized right
23:39:05stevenmis there a way to sneak an ioctl or two into a plugin ?
23:39:34Bagderioctl to do what?
23:39:58 Join StrathAFK [0] (
23:39:59Bagderthere's no such function in the plugin api
23:40:12preglowand you can't write to /dev/dsp anyway
23:40:19preglowyou're caged in the rockbox file system
23:40:19 Quit Zombiejeff ("CGI:IRC")
23:41:29stevenmpreglow, not if I make a symlink
23:41:46stevenmpreglow, I can hear sound.. it is just very very distorted due to wrong settings.
23:42:32stevenmpreglow, I am getting weirdness.. or I must have accidentally made a typo somewhere. Light clicking on rockbox version yet perfect sound on PC version
23:42:35preglowstevenm: anywho, why not just dump to a file?
23:42:43stevenmpreglow, faster, for testing
23:43:11preglownot faster if you have to jump through flaming hoops to add ioctl support :P
23:43:43stevenmpreglow, nope.. already working.
23:44:23stevenmjust added 2 linux includes to the plugin and a symlink. then pasted the init code into the plugin. works pretty well
23:44:24preglowthen hooray!
23:44:42stevenmwhy don't we add real sound support to sim ?
23:44:50preglowthe sound api isn't finalized yet
23:45:09Bagderfeel free to go ahead
23:45:12preglowit got changed today
23:45:27stevenmI wouldn't know where to even begin :(
23:45:35rasherI'm going to use rockboxui as my favourite musicplayer once it has sound support
23:45:40rasher(well maybe not quite)
23:47:05amiconnBagder: Pretty bad scans - reverse side shines through
23:47:16Bagderwhiner ;-)
23:51:32*rasher remembers his idea of downloading the patch from all patch-entries and see how many of them applies
23:51:43rasherand with how many warnings/errors
23:52:13rasherseems like a nice project
23:53:53 Quit Strath (Read error: 110 (Connection timed out))
23:55:47stevenmdangit where is that stupid clicking coming from ?
23:56:05*Rick is home

Previous day | Next day