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

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

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

#rockbox log for 2009-06-07

00:00:55fmlYes, it's the same LANG_ID. Hence the meaning should be the same. Hence it's a bug.
00:02:40onlysoaaHmm! I'm wondering if I could use the _RGBPACK macros to write to an rgb888 framebuffer?
00:02:49notlisteningHow long do we search for?
00:03:56 Quit tomers ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042523]")
00:05:57 Join matsl [0] (
00:09:02fmlHrm... But this can be ambiguous... I.e. we probably have the global setting for the icons in the theme settings and if it's enabled then the viewer setting is relevant. But only then.
00:10:23 Quit slyyx (Read error: 60 (Operation timed out))
00:10:29 Join Lss [0] (
00:11:41LloreanThere probably shouldn't be a separate setting for the viewer.
00:12:20LloreanWe don't have a different setting for menus vs filetree vs database for icons, so playlist doesn't really need to be special, I think
00:13:13fmlI dont' like the icons in the file view as they only eat space, but in the playlist viewer, I'd like the song being played marked.
00:13:32LloreanThat shouldn't be treated as an icon.
00:14:51fmlBut it's an icon. Also, there is a marker if the item is moved in the playlist. It's also an icon.
00:15:46Llorean"Icons" should refer to the filetype icons.
00:16:07LloreanImages that are icon-sized but functional, like those, shouldn't be restricted by the option to display (or not) the other icons.
00:16:22 Quit stoffel ("leaving")
00:18:29fmlYes, that's also what I'd expect. But for a non-technical user, they are icons.
00:18:45LloreanI think for most non-technical users they're pointers, not icons.
00:18:53LloreanIcons, to non-technical people, are like Windows icons.
00:19:19Mikachuall images that are icon-sized are icons, that is more or less the definition of an icon :)
00:19:21 Join FlynDice [0] (n=FlynDice@
00:19:50FlynDicebertrik: ping
00:19:59LloreanMikachu: Where are you pulling this "definition" from?
00:20:26 Join fdinel [0] (
00:20:27Mikachufor example wiktionary says "A small picture which represents something"
00:20:34Mikachuand "An image, picture, or other representation."
00:20:50LloreanYes, and pointers don't represent something. They *are* something, whereas filetype icons represent the file type.
00:21:18LloreanAnd it really has nothing to do with being "icon sized"
00:21:28Mikachuto be clear we can call them 'filetype icons' in the settings then?
00:21:44LloreanIt's not just filetypes that get icons though
00:21:47LloreanIt's also setting types.
00:21:59LloreanThey're representative icons for list entries.
00:22:22LloreanWe don't hide the pointer in the normal list when icons are disabled, do we?
00:22:26LloreanThe little arrow to show which line is selected?
00:23:11Mikachui should hope not :)
00:23:23 Join Makuseru [0] (
00:23:31LloreanIt would seriously hinder usability
00:23:44 Quit Makuseru (Read error: 104 (Connection reset by peer))
00:23:46LloreanI'm kinda surprised we hide the images fml mentioned.
00:24:27 Join __lifeless [0] (n=lifeless@
00:24:31Mikachuon a nano screen, the note icon takes up maybe two letters space
00:25:03 Quit FlynDice (
00:25:03 Quit onlysoaa (
00:25:03 Quit _lifeless (
00:25:03 Quit goffa (
00:25:03 Quit J-23 (
00:25:37Mikachui don't remember, are the non-playing entries also indented to align with the entry that does have the now-playing icon?
00:26:04NJoingoffa [0] (n=goffa@
00:26:28Mikachuthey are
00:26:31NJoinFlynDice [0] (n=FlynDice@
00:26:31NJoinonlysoaa [0] (i=422480c4@gateway/web/ajax/
00:26:31NJoinJ-23 [0] (
00:26:38 Part onlysoaa
00:26:51 Join onlysoaa [0] (i=422480c4@gateway/web/ajax/
00:27:11LloreanThere should pretty much always be an now playing mark and move marker in the playlist
00:27:29fmlLlorean: just checked: if the icons are disabled but the cursor is set to "arrow" then the icon 'column' is still shown. In that column, only the pointer is shown. If the icons are shown too then two columns are reserved.
00:27:34Mikachui guess this is more of a tangent but it would be nice if the other entries weren't indented :)
00:27:52LloreanMikachu: I think it would look pretty bad if they weren't.
00:28:03Mikachusure, but you could see what you had in your playlist then
00:28:23LloreanOne or two characters isn't going to make a significant difference.
00:28:25Mikachui'm only bringing this up as a possible reason that icon is removed when icons are disabled
00:28:34MikachuLlorean: have you ever had an ipod nano? :)
00:28:43LloreanI was the first person to hear Rockbox on a Nano.
00:28:44LloreanSo yes.
00:28:59Mikachuwell, it's like 5% of the total characters on a line
00:29:23LloreanLess than that, at least with the default font
00:30:17Llorean3% or so, I'd say. Given that, if you have 30+ character filenames where you need that one extra character to identify what the file actually is, you might want a better naming scheme.
00:30:20Mikachui use the 6x13 one since i need somewhat readable cjk characters
00:30:26fmlOk, I think I'll file a bug report.
00:31:04LloreanMikachu: Do you really need *one* more character to identify your files? It happens to line up so perfectly?
00:31:23Mikachui'm just saying it could be a reason why the icon is removed
00:31:55Mikachuand you're right, i should probably use title-artist instead of artist-title on my ipod, but it's a little boring to rename files every time i copy them over :)
00:34:09Mikachui have a short attention span, i forgot mentioning the not indenting thing :)
00:34:56Mikachuit does happen every now and then that an artist or group name is more or less the width of the screen, yes
00:35:09 Join slyyx [0] (n=slyyx@
00:36:22 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
00:37:07***Saving seen data "./dancer.seen"
00:41:32slyyxonlysoaa: Ok, we know two things: On the AAT2550 pin 4 reports current charge, pin 14 and 15 are just "Is Charging"
00:41:55 Quit bluebrother ("leaving")
00:42:03 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
00:42:04slyyxI just need to figure out what pins go into the TCC...
00:42:33HillshumCould someone please delete for me?
00:44:08onlysoaaHey slyyx: good finds. I'm working on the fake 32-bit LCD driver. So far so good!
00:44:43onlysoaaColors are off at the moment but at least I can confirm that the controller can read from a 32bit framebuffer.
00:45:11HillshumLlorean: You really should just move into the "Other Utilities" board
00:45:11 Quit fml ("CGI:IRC (EOF)")
00:46:01slyyxonlysoaa: Did you know the P2 was designed to support an ext battery
00:46:16LloreanHillshum: No, it should stay locked.
00:46:20slyyxIt has a 24pin connector for one apparently
00:46:38LloreanHillshum: Notice the words "designed to work with Rockbox" in the description of the "Other Utilities" section?
00:46:43 Quit bertrik (Read error: 113 (No route to host))
00:46:48LloreanWhat parts of Rhapsody or the OF were designed to work with Rockbox, exactly?
00:47:06LloreanOr did you not read that far before deciding it was the right place for that thread?
00:47:13MikachuHillshum: done
00:47:26*Hillshum misunderstood the intent of the "other utilities" board
00:47:36HillshumMikachu: Thanks
00:47:44FlynDicebertrik: funman: (for the logs)Hey, I got radio working on my e200v2. I ran a battery bench with fs#10048 at fastbus 64/32(and got 12:38 by the way) and when I tried radio afterwards it worked. 64MHz was max for this setup. Checking back with higher freqs now.
00:47:58LloreanHillshum: Is there some way I could make it clearer than what's written there?
00:48:02onlysoaaslyyx: Whoa, I didn't know that! Maybe we could make a 50 hour battery after all this to match the D2. ;)
00:52:12 Quit DarkDefender (Remote closed the connection)
00:58:43 Join lee321987 [0] (
00:59:47lee321987I'm trying to build the Simulator in Ubuntu. When I configure, and select "S", I get this error: configure didn't find sdl-config, which indicates that you don't have SDL (properly) installed. Please correct and re-run configure!
01:00:52Hillshumlee321987: run "sudo apt-get install sdl-dev" (i think)
01:02:12 Join lee321987_ [0] (
01:02:32lee321987_sorry - lost my connection. Did I miss an answer? Anyone know what packages I need?
01:03:18Hillshumlee321987: run "sudo apt-get install sdl-dev" (i think)
01:04:34 Join funman [0] (n=fun@rockbox/developer/funman)
01:04:41lee321987_I'll download it. Thanks.
01:05:40 Quit kkurbjun (Read error: 110 (Connection timed out))
01:07:53Unhelpfulonlysoaa: so the data gets copied and converted into a temporary buffer before the LCD controller gets it?
01:08:49onlysoaaWell it's just like the D2's driver. The Rockbox framebuffer is copied into the driver framebuffer, which is what the LCD controller reads.
01:08:53 Quit lee321987 (Read error: 60 (Operation timed out))
01:09:10onlysoaaI'm having trouble though.
01:09:20onlysoaaunsigned long is 32 bits, right?
01:09:30onlysoaaOh shit, no, it's 16 bits.
01:09:44Mikachuer, on what arch would a long be 16 bits?
01:10:00onlysoaaWait what.
01:10:08onlysoaaI haven't a clue.
01:10:12onlysoaaHow long is a long?
01:10:23 Quit lee321987_ ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
01:10:33Torne32 bits on anything rockbox supports, afaik
01:11:04onlysoaaAlright, then, I'm confused as to why it seems to be cutting off the R channel.
01:14:29 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
01:15:58 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:17:25 Quit pixelma (Nick collision from services.)
01:17:25 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
01:17:39 Quit amiconn (Nick collision from services.)
01:17:39 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
01:17:43 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
01:17:59 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
01:18:13 Join robin0800 [0] (
01:19:03 Quit robin0800 (Remote closed the connection)
01:20:29 Quit ender` (" When I die, I want to go like my Grandfather who died peacefully in his sleep, not screaming like all of the passengers in")
01:22:55funmanFlynDice: do you have anything against committing FS #10048 ?
01:23:54FlynDiceNo it works very nicely for me besides radio and we can figure that out later. That's my opinion!
01:24:09 Quit HellDragon (Read error: 104 (Connection reset by peer))
01:24:19FlynDicedid you see my comment about radio?
01:24:19 Join HellDragon [0] (
01:25:07funmanyes i saw, but i can't help here since i don't have e200v2 :/
01:25:47FlynDiceit's a lead at least :)
01:29:35funmantoo late for commit: a blocking comment on the flyspray task
01:31:28pixelmaLlorean: I think I remember seeing that insert shuffled too again (but not always) and also on hwcodec. It's actually a quite old one which I thought was fixed... I also think there is a currently closed bug report for it on a Recorder, maybe time to reopen it...
01:31:29funmansomeone from the forums will send me a c200v2 and a fuzev2
01:31:42 Quit n1s ("Lämnar")
01:32:26funmandoes Dustin Skoracki (sko) come on irc ?
01:33:10funmanhe signed off 11 days ago :/
01:33:54onlysoaaHoly crap my LCD driver is slo~ow.
01:34:12onlysoaaI suppose it could be rewritten in asm later on.
01:34:19notlisteningFlynDice, what radio issues did you have i had some too last night
01:34:34funmanonlysoaa: usually only the f
01:35:03funmanonlysoaa: usually only the functions for drawing YUV (mpeg2 decoder) or greyscale (on mono/grayscale targets) are written in asm
01:35:46notlisteningdomonoky, you got a minute to talk through how you want the filenames outputted by the SAPI app?
01:35:49onlysoaafunman: The ARM framebuffer copy is also written in asm.
01:35:51Lloreanpixelma: Ah, I only searched open tasks before posting a new one.
01:35:52funmanthat said, I don't know if some LCD screens can output YUV
01:36:34onlysoaafunman: The P2's LCD controller can output YUV, yes.
01:36:43funmanonlysoaa: oh right, is this code only for rgb565 ?
01:36:51pixelmaLlorean: I also saw this weirdness again that when moving files in a playlist where songs had been inserted shuffled (I think that's the common factor, the wrong one will be shown on the marked line during the move (the right one has the "selected" icon and you also actually move the right one. Occured to me once recently, have to try out how to reproduce
01:36:54domonokynotlistening: it should output a wav into a given filename/path, like all other tts apps/scripts :-)
01:37:13Lloreanpixelma: I wonder why "Insert Shuffled" is a special case.
01:37:24onlysoaafunman: Yeah. The arm framebuffer copy copies from rgb565 to rgb565.
01:37:28onlysoaafunman: O
01:37:40pixelmaLlorean: could be that th bug report was about plain "shuffle", don't remember
01:37:42onlysoaafunman: I'm writing a function to copy from rgb565 to rgb888.
01:38:01funmancould finally get correct colors?
01:38:20onlysoaaI'm getting there. ;)
01:38:31notlisteningdomonoky, was not sure what pther processing you are doing on these files after generation
01:39:24onlysoaaShoot, for some reason, the two most significant bytes of each 16bit pixel are ignored.
01:39:34Lloreanpixelma: Any idea who'd be best to ask about playlist stuff?
01:39:40domonokynotlistening: after generating, they are encoded with rbspeex, and either combined to a .voice file or, places on the dap as .talk files.
01:39:58onlysoaaonlysoaa: So it happens that a black to red gradient I'm trying to display shows up as four dark gradients.
01:40:15funmanonlysoaa: how is 8+8+8 packed to 16 pixels ? Oo
01:40:24funman*16bits pixels
01:40:52notlisteningdomonoky, does your app do all that work or does mine tneed to do part or all of it?
01:41:04pixelmaLlorean: last time JdGordon gave me a "fix". He later said that he's surprised that it worked...
01:41:21pixelmamaybe it just covered the issue
01:41:29onlysoaafunman: 5+6+5 is though. It's the input to my function that's not reading properly for some reason.
01:41:51domonokynotlistening: rbutil does all that. a tts engine just needs to take a text and a filename, and generate a wavfile out of this.
01:42:00funmancouldn't some (slow) shifting per pixel just do the trick?
01:42:28onlysoaaThat's exactly what I'm doing!
01:42:40onlysoaaAnd it's slow, and for some reason, the R channel is missing two bits.
01:43:14notlisteningdomonoky, thanks, i am getting there quicker than i expected :D
01:43:36funmanperhaps a stupid idea : on the sansa ams the lcd controller is wired in a weird manner : there is a 16 bits fifo to transfer pixels, but only the 4 msb and 4 lsb are mapped so that's a 8 bits transfer
01:44:03funmani.e. transfer(char data) { REG = (data << 4)|(data >>4); }
01:44:06onlysoaaHuh? o_o
01:44:51funmanyou could try to shift again the 2 msb of R channel? (just an idea ..)
01:45:10 Join sko [0] (
01:45:49funmansko: hi! i just read the FS task you posted
01:46:24funmanhow similar are the e200v2 and fuze drivers with it ?
01:46:59notlisteningdomonoky, would it not be good to use the applications cache directory or am i just going to throw these files where rbutil runs from?
01:47:27onlysoaaShifting it by two
01:47:35domonokyrbutil will give you filenames some where in the OS specifc temp directory.
01:47:41skofunman: hi, uhm... I have to take a look... but i think they are nearly the same
01:47:44onlysoaaUhm... nevermind. I'll keep looking into it.
01:48:13funmansko: if the differences are only subtle, they can be merged into one file
01:48:33skoyes, i was thinking about that too
01:48:51funmankugel started unifying the drivers but it's paused now i think
01:49:20funmansee FS #9933
01:49:20slyyxonlysoaa: Would involve actually modding the player :P
01:49:20slyyxBut according to this diagram..the board is designed for it
01:49:29notlisteningbrilliant, so as an example -o c:/temp/a.wav will come from rbutil
01:49:30 Quit slyyx ("Reconnecting")
01:49:39 Join slyyx [0] (n=slyyx@
01:51:49 Quit Nico_P (Remote closed the connection)
01:52:06onlysoaaUgh... Why can't the dumb LCD controller shift the bits for me?
01:52:40slyyxonlysoaa: Which controller? the one in the TCC?
01:52:44notlisteningdomonoky, and as a nicety do you want differing audio formats?
01:53:10slyyxonlysoaa: Here, let me help ya
01:53:57domonokynotlistening: you mean different sample rates or alike ?
01:54:08onlysoaaslyyx: If you can get the R and B bits to shift left from an rgb565 framebuffer, it would be GREAT.
01:54:41notlisteningdomonoky, yeah sorry bad terminology used there
01:54:58domonokynotlistening: configurability is always nice :-)
01:55:39 Join intrados [0] (
01:55:43LloreanSome people are willing to spend the RAM on higher quality voices, while others would prefer them as small as possible and maintain understandability.
01:56:41notlisteningyou got it, just out of interest can i have a file name in rockbox with <> characters in it?
01:56:45skofunman: i took a look at both drivers, there are some differences, but i think they could be merged into one
01:57:15funmansko: if you can do that that would be nice ..
01:57:25pixelmanotlistening: don't think so
01:57:42Lloreannotlistening: It's still FAT32
01:58:00LloreanRockbox doesn't magically change the capabilities of the filesystem.
01:58:10funmansko: after that perhaps we can look at merging the LCD because they are very similar (approximate to 90° :)
01:58:16domonokywho do we speak the sd-card entry ?
01:58:48notlisteningright just funks up the xml parser in sapi but i can turn that on and off
01:58:50Lloreandomonoky: It probably falls back to spelling or numbering.
01:58:50skoi'll try it... but tomorrow, its really late here and i have to get some sleep (last night was very short too ^^)
01:59:07notlisteningyeah noticed that on my e200v2
01:59:08*pixelma remembers the false <MMC1> directory created when testing the recording directory setting patch and having a hard time getting it removed again
01:59:31 Quit intrados (Client Quit)
02:00:11funmansko: thanks ;) no hurry .. it's better to code when not tired. good night!
02:00:24slyyxonlysoaa: Why not just hand it rgb555?
02:00:35skofunman: thanks :) bye
02:00:43 Quit sko ("bye")
02:01:01 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
02:01:22slyyxonlysoaa: The lcdc supports 555 and 565 apparently..what is the issue?
02:06:56saratogais the TCC780 and TCC780X different?
02:10:47onlysoaaThe P2 uses the TCC7801. I think 780x refers to the whole family.
02:11:03 Join perfectdrug [0] (
02:11:16onlysoaaI tried setting it to rgb555, but it didn't turn out so well. :P
02:11:29onlysoaaI think the LCD needs rgb666 in.
02:11:30saratogais the 780 part of the 780x family?
02:11:42onlysoaaThere is no 780 I think.
02:11:45onlysoaaI'll be back in a bit.
02:11:52saratogathere is
02:11:55saratogai have a datasheet for it
02:12:18onlysoaaI think that's the 780x datasheet.
02:12:28 Quit onlysoaa (" ajax IRC Client")
02:13:17saratogaok good
02:15:08funmansaratoga: FlynDice do you happen to listen to flac files ? i just tested playing flac with fs#10048 and i can reproduce what mc2739 described; and now i'm playing some flac without the patch
02:15:50saratogafunman: no but i can make some
02:16:27saratogaonlysoaa: looking at the datasheet, have you tried checking the LCD Image0 Control Register (LI0C) to see which color format the controller is operating in?
02:17:44 Quit domonoky (Read error: 104 (Connection reset by peer))
02:22:36 Join onlysoaa [0] (i=422480c4@gateway/web/ajax/
02:23:14saratogaonlysoaa: did you see my remark?
02:23:26onlysoaaI don't think so.
02:31:58slyyxonlysoaa: The lcd is 6,6,6, but they are only giving 4,4,4 to the lcd
02:32:32slyyxSo, 454 makes no sense...
02:33:15onlysoaaHuh? Whhat's giving 444 to the LCD?
02:33:22slyyxsec, reading again
02:33:28onlysoaaThe LCD itself supports 888, but it's being fed 666.
02:33:48slyyxsorry, had the numbers wrong :P
02:33:53saratogachanging the controller to 565 might make sense then
02:34:09Unhelpfulyou need to know what the *controller* wants. whatever downconversion it's doing before sending to the LCD shouldn't matter.
02:34:13onlysoaaIndeed, but somehow the controller doesn't shift the R and B bits to compensate for the missing bits.
02:34:23slyyxUnhelpful: The controller has two options
02:34:40slyyxUnhelpful: 565, and 555
02:34:46onlysoaaThe controller does read rgb565.
02:34:54onlysoaaFeeding it 555 wouldn't hurt.
02:34:59onlysoaaWe'd lose a bit though.
02:35:22slyyxOh well
02:35:30slyyxOnly matters for album art
02:35:41Unhelpfulsince 555 still fits in 16 bits, if we *must* use 555 for the LCD, we may as well use 555 internally.
02:35:51saratogait supports a lot of modes
02:35:54onlysoaaWell we still need to feed 555 into the framebuffer. :|
02:36:46slyyxYou know more about the framebuffer then I do
02:36:50saratoga555, 565, 888, YUV
02:37:09***Saving seen data "./dancer.seen"
02:37:17slyyxsaratoga: You know things about the tcc?
02:39:15saratogai'm just skimming the LCD controller datasheet now
02:40:11slyyx..Why do I not see 888...
02:40:23saratogawhich datasheet are you looking at?
02:40:25onlysoaargb888 is in the framebuffer.
02:40:35slyyxOh..I see it
02:41:26slyyxSince the lcd is 666....We should use 555, no?
02:41:37saratoga565 would make the most sense
02:41:37slyyx888 would be pointless
02:41:47slyyxBut it is causing issues?
02:42:23slyyxIt is giving you issues, no?
02:42:25onlysoaa565 would indeed make the the most sense, but it's doesn't shift the R and B channels right!
02:42:40saratogaits probably not initialized correctly then
02:43:06slyyxonlysoaa: Are you reading the datasheet? Or basing it off of the D2 code?
02:43:32 Quit matsl (Remote closed the connection)
02:43:43onlysoaaslyyx: Both. :P
02:43:44saratogawhich mode is the control register in when you boot up?
02:44:08onlysoaaControl register of the LCD controller?
02:44:21onlysoaaI set it to output rgb666.
02:45:04saratogawhat is it set as the input?
02:45:17saratogaand I guess also check the endianness too incase the OF is doing something silly with that
02:45:59onlysoaaRight now as rgb888, because I'm trying to write an rgb888 framebuffer.
02:46:13onlysoaaI had it set at rgb565 before though.
02:47:02 Quit perrikwp (" ajax IRC Client")
02:47:06saratogawhy 888?
02:47:10 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
02:47:14saratogai don't think rockbox supports that?
02:47:33onlysoaaBecause 565 was causing problems. D:
02:47:56slyyxonlysoaa: Why do we not just go 555?
02:48:12saratogawell 16 bits per pixel makes more sense then 15
02:48:20slyyx888 would make the most sense..then we lose nothing...
02:48:27saratogaexcept for rockbox support
02:48:34saratogawhich is kind of important
02:48:51onlysoaaWe can't just go 555 because the controller expects rgb565.
02:48:57onlysoaaThe bit layout is different.
02:49:03onlysoaaI'll be back in a bit.
02:49:03slyyxThe controller _has_ a 555 mode
02:49:46slyyxHmm..I need to lookup the proper shutdown for this...
02:51:38 Quit saratoga ("CGI:IRC (EOF)")
02:53:59 Quit perfectdrug ("Leaving.")
03:01:27 Join yashwa7 [0] (
03:01:55 Quit HellDragon (Read error: 60 (Operation timed out))
03:02:02slyyxonlysoaa: Any clue how to read the touchscreen
03:03:51 Join Trid [0] (
03:05:12 Quit funman ("leaving")
03:06:08TridHi. I've been using rockbox for a couple weeks on the V1 Sansa Fuze, and I wrote a step by step guide to putting rockbox onto unsupported sansa players. Could I have permission to edit the wiki? My username is NatStory
03:17:09LloreanTrid: Is there something wrong with existing instructions?
03:17:30TridI found them kind of confusing
03:17:39LloreanWe don't generally like random guides being added. There should be the developer instructions, and then when the port is complete, the manual.
03:17:40Tridand messed up a few times before I got it right
03:18:23LloreanWere they confusing in a poorly written way, or simply because you were unfamiliar with terms/tools/development?
03:19:01onlysoaaslyyx: I don't know how to access the touch screen. I guess you could look at the pins and see if you can get a reading from GPIO.
03:19:13TridI thought some steps were missing, and that other instructions were unclear
03:20:33 Join hillshum [0] (n=quassel@unaffiliated/hillshum)
03:20:36LloreanTrid: If you'd like to fix the instructions, I'll give you write access.
03:20:54TridThat would be great.
03:21:21slyyxonlysoaa: I duno which gpio it uses
03:21:37onlysoaaslyyx: The schematics don't show?
03:21:49slyyxAtleast, not named anything nice
03:21:57onlysoaaThat's odd. I'll look through it.
03:22:09yashwa7I applied for GSoC but failed to make it. Still i would like to work on my proposal. Was wondering if I will get an unofficial mentor.
03:22:51LloreanTrid: Done. As I said, please try to focus on making it developer information, rather than a simplified guide for end users.
03:23:07TridGot it. Thanks.
03:23:23 Join andrewbeveridge [0] (i=586d6810@gateway/web/ajax/
03:25:23slyyxonlysoaa: Connected vita "18bit parallell" (i2c or w/e its called I guess)
03:26:03onlysoaaslyyx: Do you think it could be UART?
03:26:50 Join HellDragon [0] (
03:26:54slyyxIt just says "18bit parallell" clue
03:27:07slyyxthat is the lcd
03:27:16onlysoaaslyyx: i2C_DAT1?
03:27:25slyyxLooking into it...
03:27:37onlysoaaIt'd be useful if we knew what model the touch panel is.
03:27:54slyyxSomeone does..the wiki does
03:28:14onlysoaaNo, that's just the LCD screen. Not the touch panel.
03:28:48slyyxah, yes
03:29:00slyyxmaybe the components list does
03:29:08onlysoaaOh right.
03:29:54Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
03:30:11slyyxAT300 AH59-01858A TOUCH/PANEL;87X45X0.15,CAPACITIVE,T1006
03:30:15Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
03:32:33onlysoaaGood find!
03:32:41onlysoaaSee if you an find a datasheet for it too.
03:32:49slyyxDoes bnot list a model number :P
03:32:59slyyxJust a samsung part number
03:32:59onlysoaaSorry, I'm a tad busy right now.
03:33:17slyyxyeah, np
03:34:18onlysoaaAh crap.
03:34:24 Quit Trid ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
03:46:38 Join metabsd [0] (
03:47:01 Part metabsd
03:48:19 Quit yashwa7 ("Easy as 3.14159265358979323846...")
03:49:11slyyxonlysoaa: How goes it?
03:49:30 Join saratoga [0] (i=9803c6dd@rockbox/developer/saratoga)
03:51:02saratogayashwa7: (for the logs) contact me when you get a chance
03:53:49 Quit chandoo ("Leaving")
03:56:54andrewbeveridgehi, can I request wiki write perms for AndrewBeveridge please?
03:57:39Lloreanandrewbeveridge: What were you going to edit?
03:58:03andrewbeveridgeLlorean: no edits, just creation of a new port page, for the Cowon O2
03:58:21andrewbeveridgeI'm compiling information at the moment
03:58:40saratogaLlorean: can we get GSOC applicants emails?
03:59:07LloreanI don't know. I think that may be under the purview of project admins, and I'm not one this year.
03:59:19saratogaah ok
03:59:42*andrewbeveridge says thanks quietly so as not to disturb =)
03:59:45*Llorean puts "contact email" on the list of things we need to get _before_ acceptance next year.
04:00:00hillshumsaratoga: ask Bagder?
04:00:05saratogaits surprising that the application doesn't give us their gmail account
04:00:42*slyyx throws arms up in the air and gives up....
04:00:58onlysoaaslyyx, where are you at?
04:01:44*Mikachu catches slyyx' arms and gives them back to him
04:02:39slyyxonlysoaa: Trying to find: Pin that connects from pins 4, and 14..that it states go out from the power chip..
04:02:59onlysoaaHmm... o:
04:03:05Unhelpfuljhmikes would still be the best person to ask about anything beast-specific, right?
04:03:41LloreanUnhelpful: Good starting point, at least.
04:04:11 Quit evilnick_home1 ("Leaving.")
04:04:23Unhelpfuli'd like to see if the hang check on it can be made to display a panic message... not even sure where that implementation is, i'm digging through the target dir now
04:04:34 Join evilnick_home [0] (
04:05:53 Quit evilnick_home (Client Quit)
04:06:08 Join evilnick_home [0] (
04:06:47*slyyx slaps head
04:06:59slyyx"The TCC780 has 8 channel general purpose low-power ADC for battery level detection,
04:07:02slyyxkey detection, remote control interface, touch screen interface, etc.
04:07:12slyyxRight in front me of
04:07:17*slyyx gets coat and leaves
04:07:51onlysoaaWell at least you found it!
04:08:06onlysoaaLooks like you'll end up doing most of the port in the end.
04:08:13slyyxgotta go read about adc now...
04:09:16onlysoaaHey, do you think there's a way to make the LCD controller shift the bits? I'd really rather not hack together an 32bit framebuffer if I don't have to. It's really slow.
04:10:42slyyxJust give it what it wants?
04:11:03onlysoaaThat's what I tried first!
04:11:18onlysoaargb565 in, rgb666 out.
04:12:04onlysoaaWell, it looks like what you have.
04:12:14onlysoaaUnless you modified code that I patched.
04:12:30slyyxlooks fine to me
04:12:47onlysoaaIt's greenish!
04:13:01slyyxlemme look again
04:13:03onlysoaaThe R and B channels need a shift to the left.
04:14:48slyyxoh yeah, I see
04:14:55slyyxlooks more like it is missing the red channel
04:15:20slyyxTest it, make it display three colour bars
04:15:31slyyxI bet you, the red one will not be red
04:15:51onlysoaaDo you think?
04:16:02slyyxyeah, try it
04:16:18onlysoaaI did the same thing with my rgb888 code, but the red channel got all weird.
04:16:29slyyxI duno how rockbox works for displaying stuff, but display a red blue and green bar
04:16:32onlysoaaIt'd be a real surprise if it does the same thing with rgb565.
04:16:43onlysoaaI wouldn't think so though.
04:16:51slyyxI am positive it is just missing red
04:18:08onlysoaaYou may be right. ;) I'm in Windows right now.
04:18:20onlysoaaI won't be able to recompile until I reboot into Ubuntu.
04:18:41slyyxI will code the test color bar thing then
04:19:11onlysoaaJust modify the Rockbox logo bitmap with some gradients and see what shows up.
04:19:13onlysoaaThat's what I did.
04:19:43slyyxI will
04:20:52slyyxonlysoaa: Which bitmap is it using?
04:21:09onlysoaaNative, 320xsomething.
04:24:19slyyxonlysoaa: yeah
04:24:20slyyxred is missing
04:24:23slyyxI win :P
04:24:44onlysoaaOh youuuu. You win indeed. :P
04:24:52onlysoaaNow what?
04:24:57onlysoaaOh wait!
04:25:01onlysoaaDoes red show AT ALL?
04:25:48slyyxLooks black
04:25:58onlysoaaNot even small speckles of red?
04:26:03*slyyx checks again
04:26:06onlysoaaThat's utterly strange.
04:26:14slyyxMy red was not perfectly red so it had some colour..but no red
04:26:30onlysoaaWhat to make of that now?
04:26:36slyyxred is just dark..
04:26:39slyyxvery dark
04:26:49onlysoaaAh hah!
04:27:05onlysoaaDid you put just a red box or did you do a black to red gradient?
04:27:11onlysoaaTry a gradient and see what it does.
04:27:35slyyxI just stuck a red, blue, and green line in with the gimp
04:27:44onlysoaaMake a red gradient.
04:27:52onlysoaaBlack to 0xff0000
04:28:13onlysoaaI think you'll get the same as I got with my rgb888 code!
04:29:09 Quit HellDragon (Read error: 104 (Connection reset by peer))
04:30:31onlysoaaWhat does it show?
04:30:33slyyxit displays
04:30:42onlysoaaFour gradients?
04:30:47slyyxthree or four..yeah
04:30:57onlysoaaSame as what I had.
04:31:05onlysoaaWhat's up with that now?
04:31:18slyyxLet me try it with the other channels to see if it is just red that is broken
04:31:33onlysoaaLooks like I knocked myself out making an rgb888 framebuffer for absolutely nothing. :P
04:31:37onlysoaaThe others are fine.'s the two most significant bits of the R channel that are cleared.
04:32:50onlysoaaDo you think it could have to do with timings or clock?
04:33:12slyyxLet me take a look at your code here for a bit..
04:33:37slyyxwhat did you change exactly?
04:34:26onlysoaaThe LCD controller configuration, mostly.
04:34:38onlysoaaJust compare it to the D2's and you'll see the differences.
04:34:39slyyxwhich file is that in?
04:34:45slyyxwhich file is that in?
04:34:53onlysoaalcd-ypp2.c and lcd-cowond2.c
04:36:30notlisteningnight all
04:36:44 Quit notlistening ("Leaving")
04:37:00slyyx...I do not understand what this thing does lcd_write_reg work?
04:37:11***Saving seen data "./dancer.seen"
04:37:19onlysoaaThat's specific to the D2 right now.
04:37:23onlysoaaIt doesn't do anything for the P2.
04:37:38onlysoaaI haven't figured out how to work it for the P2 yet, actually...
04:37:53onlysoaaIf you find out how, it could give us a clue as to how to get the rest to work too.
04:37:54slyyxBut you changed what you do with it...
04:38:23slyyxso for now, init_device is what is important..k
04:38:31slyyxah..I remember this
04:38:38onlysoaaI tried to mimick the suggested startup settings from the LCD data sheet, but obviously they aren't set correctly, because it's still specific to the D2 panel.
04:38:42andrewbeveridgehi, erm I seem to be redirected to "" whenever I try to edit my own wiki page... It was literally working fine a few minutes ago, any ideas?
04:39:37 Join Zom_ [0] (
04:39:39 Quit Zom_ (Client Quit)
04:39:43slyyxonlysoaa: brb, once I get back I will dive into the datasheet
04:43:37 Join HellDragon [0] (
04:43:45slyyxonlysoaa: back
04:45:30 Join chandoo [0] (
04:45:34 Join scott_67 [0] (i=3a6e9ee2@gateway/web/ajax/
04:46:15scott_67hi guys, is there a volume limit patch other than replaygain?
04:46:32scott_67I see
04:46:48LloreanWe've yet to figure out how to apply patches to humans to prevent them from pressing up continually, but word is there's a workaround in which you simply limit yourself.
04:46:59onlysoaaslyyx: welcome back. (:
04:47:02scott_67have found an old patch but looking for a precompile binary
04:47:41hillshumscott_67: to apply the patch you'll need the source code
04:47:46scott_67this one is old
04:47:54scott_67i know about patch and stuff
04:48:29hillshumand sync it to svn most likely
04:49:09scott_67podzilla doesnt have a volume limiting as well
04:49:27scott_67and i cant figure to load eu firmware on 4g ipod colour
04:50:42andrewbeveridgescott_67: may i ask why you need volume limiting so much?
04:50:49*andrewbeveridge hopes that isn't considered off topic
04:51:26scott_67@andrew, coz it will save the ears of my oldies
04:51:45scott_67i can only let them use psp bec of its alvs something
04:51:55scott_67but psp is bulky
04:52:04scott_67and i only have a spare 4g ipod
04:52:05LloreanPlease use real English in here, as per the channel guidelines.
04:52:10LloreanAlso, this is getting terribly off-topic.
04:53:01scott_67Sorry Llorean, my apologies. I was just gathering some information regarding my problem.
04:53:18slyyx...who wants volume limiting...anyway..I was going to rewrite the int lcd code..
04:53:55scott_67I want it slyyx
04:54:10slyyxJust do not turn it up that high?
04:54:49scott_67If you're telling that to a 50 year old that forgets about his memory, it is another story
04:55:18LloreanAs you mentioned earlier, replaygain is also an option.
04:55:30LloreanA volume limiter wouldn't do any good, since they could just as easily adjust the volume limit if they decided it needed to be louder
04:55:42slyyxonlysoaa: Look up what bits 2,6,8,25,and 30 are
04:56:03slyyxer..and 23
04:56:07slyyxnot 30
04:56:27scott_67If you locked the volume limit, it will take a bit of tinkering for them to adjust it
04:56:56perrikwpjust set a pre-cut in the equalizer
04:56:59scott_67anyway, it is a feature that I am looking for and there are reasons behind it.
04:57:05scott_67thanks perri
04:57:13slyyxThe p2 has it in an option...
04:57:19slyyxanyway, off topic
04:57:22Lloreanslyyx: And we don't care about original firmwares here.
04:57:53slyyxSomeone should code it as a passworded feature
04:58:00slyyxthen you give it to your kids, and not have to worry about it
04:58:18LloreanIt would always be possible to just reset the settings.
04:58:29hillshumstill security through obscurity
04:58:40 Nick hillshum is now known as Hillshum (n=quassel@unaffiliated/hillshum)
04:58:52slyyxHillshum: Password is not obscurity?
04:58:54andrewbeveridgeif it matters that much surely one could edit the source to remove the upper options and recompile?
04:58:57onlysoaaslyyx: the bits of what exactly?
04:59:03slyyxonlysoaa: the lcd cont
04:59:13slyyxandrewbeveridge: If your kids are that smart
04:59:14onlysoaaKk hold on.
04:59:21slyyxI am talking about these people with 8 year olds with mp3 players
04:59:34LloreanAs has been said, there's always replaygain or the EQ precut.
04:59:51scott_67I will try the EQ precut
05:00:03LloreanOr you could adjust the gain in the audio files when you encode them for your listeners, so that they're much quieter than average and impossible to get to an unsafe level even without a volume limiter.
05:00:41Hillshumslyyx: it might help to compile without the eq menu and use a .cfg file to set the pre-cut
05:01:22slyyxHillshum: could work, yeah
05:01:27 Join togetic [0] (n=ib@unaffiliated/ibuffy)
05:02:39slyyxonlysoaa: Those bits are set...I bet you it is something to do with those
05:04:19scott_67the eq precut works even at max volume, it is tolerable
05:04:28scott_67i just wish it can still be limited
05:04:33scott_67anyway, thanks guys
05:04:46andrewbeveridgescott_67: it can be, just edit and recompile.
05:05:29scott_67ah yeah, i can lower it further, the max now is -24db
05:07:47togeticyou rockbox experts, i'm needing some help
05:08:40onlysoaaslyyx: you have the TCC780x datasheet, right? It'd probably be quicker if you look it up yourself.
05:08:49togetici partitioned my ned 1stgen 4gb ipod nano that came in today
05:08:54 Quit HellDragon (Read error: 104 (Connection reset by peer))
05:08:55 Quit chandoo ("Leaving")
05:08:59togeticusing ^
05:09:00 Join HellDragon [0] (
05:09:06Hillshumslyyx: a password is still obscurity because there's other ways to get rid of the limit
05:09:11onlysoaaslyyx: In any case, I inverted the HSYNC and VSYNC clocks because that's how the OF did it.
05:09:39togeticdownloaded the mbr-nano4gb.bin, ran sudo dd if=mbr-nano4gb.bin of=/dev/sdb3
05:09:43onlysoaaslyyx: So don't reverse that. :P For the other bits... they made sense kinda so I just left them like that.
05:09:46togeticdownloaded the installer
05:09:54togeticwait, first i ejected it with the eject command
05:09:59togeticplugged it back in
05:10:05togeticdownloaded the installer and ran it
05:10:12krazykittogetic, you would dd the mbr to sdb, not sdb3
05:10:13slyyxonlysoaa: there is no 666 mode...
05:10:36togetickrazykit: why sdb?
05:10:36onlysoaaslyyx: It's 18pxwd.
05:10:48togeticsdb3 is the ipod, where it's mounted..
05:10:48slyyxoh, yeah :P
05:11:06andrewbeveridgetogetic: no, sdb3 is the partition
05:11:12andrewbeveridgesdb is the actual device
05:11:13slyyxonlysoaa: so it looks like it is somehow overflowing the red value..
05:11:28togetici'll try that again then
05:11:41andrewbeveridgeyou can't mount the device, you mount a filesystem on a certain partition
05:11:44 Quit HellDragon (Read error: 104 (Connection reset by peer))
05:11:57onlysoaaslyyx: Yeah.
05:12:28onlysoaaonlysoaa: I think it might have to do with the clock.
05:12:36onlysoaaslyyx* oops again.
05:12:56slyyxyeah, you did that timing math right?
05:13:00onlysoaaI'm thinking maybe it's shooting data too fast, so the red bits got clipped.
05:13:02onlysoaaI did not! :P it?
05:13:16onlysoaaI don't have my code in front of me. D:
05:13:25slyyxFine..I will do it
05:13:44onlysoaaThanks. :P We don't even know the refresh rate of this thing yet.
05:13:54onlysoaaAll that clockdiv stuff went right over my head.
05:14:04slyyxIts all in the datasheet
05:14:19onlysoaaI know, that's why it went right over my head. :P
05:14:34onlysoaaI'll go through it again though. I might figure out something out.
05:14:39onlysoaauhh, -out
05:15:02togeticack, now i can't mount it after unmounting it
05:15:15togetic$ sudo mount /dev/sdb3 /media/ipod
05:15:16togeticmount: special device /dev/sdb3 does not exist
05:17:02onlysoaaslyyx, what's the LCLK?
05:17:44andrewbeveridgetogetic: join the community channel please so i can try and help you (it is unrelated to rockbox)
05:18:23 Join HellDragon [0] (
05:18:53slyyxonlysoaa: onlysoaa clock...refresh..rate..or something
05:20:49onlysoaaGPIOD11 for LCLK
05:35:59onlysoaaslyx, found anything? o:
05:44:03slyyxonlysoaa: was afk
05:46:07slyyxHow do you know it is gpiod11?
05:47:12slyyxOh, yeah
05:47:17onlysoaaThe datasheet. ;D
05:48:06slyyxit is on the same pin as gpiod11, it isnt gpiod11 though...
05:49:18slyyxLCLKDIV is actually what you want
05:49:40onlysoaaLCLKDIV relates to LCLK.
05:49:54onlysoaaIt's important to know what LCLK is first.
05:50:51 Join Tristan [0] (
05:52:04slyyxI duno :P
05:53:00onlysoaaslyyx: shoot.
05:53:10slyyx programmed
05:53:10slyyxfrequency of the LCD/memory controller
05:53:23slyyxCheck the datsheet
05:53:36onlysoaaWait, it's set in system-tcc780x.c
05:53:37slyyxas in, the lcd datasheet
05:53:44onlysoaaI set it according to the OF.
05:53:59slyyxJust do the math and see if it fixes it :P
05:54:33onlysoaano no
05:54:44onlysoaaLCDSI CTRL0
05:54:48onlysoaaLook into that register.
05:54:58onlysoaaI found it.
05:55:14onlysoaaIt doesn't have to do with clocks I think.
05:55:16 Quit efyx_ (Remote closed the connection)
05:55:28slyyxJust tell me when you get it working :P
05:55:45onlysoaaAlright. Looks like I'm going back into ubuntu.
05:56:32onlysoaaI'll be right back. (:
05:56:34 Quit onlysoaa (" ajax IRC Client")
06:00:22 Join mib_fncp8x [0] (i=422480c4@gateway/web/ajax/
06:00:29mib_fncp8xHaha, I'm back. ;D
06:00:34 Nick mib_fncp8x is now known as onlysoaa (i=422480c4@gateway/web/ajax/
06:01:46slyyxonlysoaa: hey
06:02:52 Quit perrikwp (" ajax IRC Client")
06:02:57 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:03:15onlysoaaAlright, I'll be looking into the thing I found now.
06:10:58slyyxWell, sometime next week I will get the power stuff working, and make it so you can use the touchscreen
06:11:22slyyxgotta figure out acd
06:13:36 Join hd [0] (
06:13:38 Quit HellDragon (Read error: 54 (Connection reset by peer))
06:22:55 Quit timc (Read error: 110 (Connection timed out))
06:23:28 Quit Hillshum (Remote closed the connection)
06:25:57 Join timc [0] (n=aoeu@
06:28:43 Join ademille_ [0] (
06:29:32 Quit andrewbeveridge (" ZZZZzzzzzz.....")
06:30:15 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:32:46 Quit togetic (Read error: 104 (Connection reset by peer))
06:33:49 Quit ademille_ ("Leaving")
06:35:23 Join ademille [0] (
06:36:19 Nick ademille is now known as aaron2 (
06:37:13***Saving seen data "./dancer.seen"
06:43:30 Part aaron2
06:45:34slyyxonlysoaa: any luck?
07:05:47onlysoaaslyyx: Nope. ): Even after setting the LCDSI the colors are still the same.
07:05:55onlysoaaIs there a way to make the RB logo show longer?
07:08:49 Join NightMayor [0] (
07:10:46slyyxonlysoaa: Edit the code?
07:12:45 Part NightMayor
07:14:13onlysoaaslyyx: Duh, I was wondering where. :P
07:14:48onlysoaaIt turns out the LCDSI stuff is irrelevant, because the LCD controller talks to the LCD directly.
07:14:59onlysoaaLCDSI is for something like a different screen.
07:27:17 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
07:47:07 Quit __lifeless (Remote closed the connection)
07:47:23 Join __lifeless [0] (n=lifeless@
07:49:59 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
07:54:56 Quit slyyx (Read error: 110 (Connection timed out))
07:55:54 Join bertrik [0] (
08:06:48 Quit Horscht (Read error: 110 (Connection timed out))
08:20:52 Quit daurnimator (Read error: 110 (Connection timed out))
08:21:50 Join petur [0] (
08:22:15 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
08:22:36 Join Rob2223 [0] (
08:24:12 Quit jmillikin (Read error: 110 (Connection timed out))
08:37:15***Saving seen data "./dancer.seen"
08:40:17 Quit Rob2222 (Read error: 110 (Connection timed out))
08:44:07 Join HellDragon [0] (
08:44:35 Quit hd (Read error: 54 (Connection reset by peer))
08:53:21 Quit HellDragon (Read error: 104 (Connection reset by peer))
08:53:28 Join HellDragon [0] (
08:57:02bertrikdoes the sansa m200 use a center button to confirm things?
08:59:14bertrika bunch of ams sansa targets still define LANG_CONFIRM_WITH_BUTTON to "PLAY = Yes". I think it should be "SELECT = Yes"
09:01:12bertrikIf I change this in english.lang I suppose I also have to change it in all other languages, right?
09:03:10 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:09:16 Join stas [0] (
09:10:00stasHi guys, do you know if there is a pdf reader for it?
09:10:32bertrikNo, there is no pdf reader for rockbox
09:10:52stasnot even hacks?
09:22:28bertriknot that I know of
09:26:06 Join stoffel [0] (
09:27:23 Join hd [0] (
09:28:24 Join HellDragon_ [0] (
09:37:25 Join _lifeless [0] (n=lifeless@
09:39:56 Quit BHSPitLappy (Remote closed the connection)
09:40:54 Quit HellDragon (Read error: 110 (Connection timed out))
09:41:18 Join HellDragon [0] (
09:45:54 Quit hd (Read error: 110 (Connection timed out))
09:46:27 Quit HellDragon_ (Read error: 110 (Connection timed out))
09:50:42 Quit dmb ("Leaving")
09:51:20 Join hd [0] (
09:51:26 Join webguest9283 [0] (n=3e10f680@gateway/web/cgi-irc/
09:52:43 Quit __lifeless (Read error: 110 (Connection timed out))
09:56:02 Join n1s [0] (n=n1s@rockbox/developer/n1s)
09:56:26 Quit webguest9283 ("CGI:IRC (Ping timeout)")
10:00:36 Join dmb [0] (n=dmb@unaffiliated/dmb)
10:03:53 Quit HellDragon (Read error: 110 (Connection timed out))
10:04:46 Join flydutch [0] (
10:12:57 Join DarkDefender [0] (
10:13:30 Join antil33t [0] (n=Mudkips@
10:17:48 Quit antitrons (Read error: 110 (Connection timed out))
10:18:34 Join perfectdrug [0] (
10:37:19***Saving seen data "./dancer.seen"
10:43:05 Join DarkAnthey [0] (n=darkanth@
10:45:12 Join funman [0] (n=fun@rockbox/developer/funman)
10:45:19 Part perfectdrug
10:47:30 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
10:54:05 Join Jaykay [0] (
10:56:45*bertrik notices again that rockbox uses a "bouncy" progressbar to indicate file copy/move/delete progress
10:57:23 Join Jaykay_ [0] (
10:57:57bertrikIt's not really a bug but I'm tempted into making it a real progress bar that indicates a true percentage of kbytes moved/copied or # files deleted
10:58:56Jaykay_is gapless plyback supposed to work on an e200 with mp3-loops?
11:02:52 Quit n1s (Read error: 60 (Operation timed out))
11:04:44stasbertrik: thx
11:06:24linuxstbJaykay_: It should do, if the mp3 files have the gapless info stored. Do they play gapless in anything else (e.g. foobar) ?
11:09:25amiconnbertrik: Calculating the total beforehand would take a significant amount of time. Especially for delete opration, or when copying lots of small files
11:09:41Jaykay_linuxstb: i don't have any gapless info :) and i tried only in vlc-player, and vlc-player needs ~3seconds to play a track again
11:10:36bertrikfor delete we could just count the number of files instead of the number of bytes, since deletion time is probably not related to file size
11:11:06amiconnWell, counting the files would take almost as long as actually deleting them. Both are directory operations
11:11:42bertrikI don't know if this is indeed true
11:12:17amiconnWith recursive deirectory traversing it definitely is, at least without dircache
11:13:07 Quit Jaykay (Read error: 110 (Connection timed out))
11:13:14bertrikHmm, isn't there really anything better we can come up with than just a "bouncy" progress indicator?
11:13:37bertrikIt's now just a basic "hey I'm still alive" indicator but otherwise totally useless
11:16:07 Quit tessarakt (Read error: 104 (Connection reset by peer))
11:17:41 Quit AndyI (Read error: 113 (No route to host))
11:18:24 Join tessarakt [0] (
11:18:30Jaykay_linuxstb: Itunes plays it in a perfect loop without a gap
11:19:41bertrikFor file copy/move we could use a true progress indicator for each individual file. So you still see something to determine that it's still alive and for single big-ish files you see actual progress
11:20:38Jaykay_linuxstb: sorry, not perfect... it cuts the last ~0,5 sec of the track off
11:20:45 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
11:21:03linuxstbJaykay_: Do you know how the file was encoded?
11:21:06 Join n1s [0] (n=n1s@rockbox/developer/n1s)
11:21:15Jaykay_linuxstb: no
11:21:41Jaykay_tags says fraunhofer... not that accurate
11:23:02Jaykay_no, not the tags... only itunes
11:24:41pixelmamaybe it's the Itunes version of gapless? Did you try other media players?
11:26:13 Join q-rip [0] (n=92e6801d@gateway/web/cgi-irc/
11:26:46q-ripis there any chance that rockbox will run on a classic 80gb ipod soon?
11:27:18Jaykay_pixelma: only vlc-player, but it needs approx. 3sec to start the track again
11:27:20amiconnUnhelpful: ping
11:28:45n1sq-rip: no
11:29:21q-ripwhat do they need to encrypt the firmware?
11:30:08funmanApple knows, not us
11:30:18q-ripapple sux!
11:31:17pixelmaJaykay_: ah, now I understand what you meant with mp3 loop... to me gapless is for files that "belong together", so probably wouldn't call what you want (if I understood correctly) "gapless"
11:32:25Jaykay_pixelma: i have some tracks which are designed to be looped (without gaps) and while repeating one i hear gaps when the track is started again. thats my "problem" :)
11:33:49 Nick Jaykay_ is now known as Jaykay (
11:34:00 Quit funman ("worktime!")
11:34:15pixelmadon't have such tracks. But you should really try to find out how they were encoded (which type of gapless tags they have)
11:37:39Jaykaypixelma: should those tag be in normal id3-tags?
11:37:40 Quit daurnimator (Read error: 110 (Connection timed out))
11:38:46 Join AndyI [0] (i=AndyI@
11:45:26 Join robin0800 [0] (
11:47:46linuxstbJaykay: So it works for some of your files, but not all?
11:48:18Jaykaylinuxstb: no, on rockbox it works for none of those files
11:49:01 Quit q-rip ("CGI:IRC")
11:50:01linuxstbJaykay: Could you upload one of the files somewhere?
11:50:17Jaykaylinuxstb: sure... where?
11:50:29linuxstbWherever you can...
11:51:11linuxstbOr you could email it to me?
11:51:16 Join ender` [0] (
11:58:51Jaykaylinuxstb: net/download-1687797/205308_Final_Encounter.mp3.html">
11:58:52 Quit at0m (Remote closed the connection)
11:58:58 Part stas ("Leaving")
11:59:12Jaykay there it's played in a perfect loop
11:59:23Jaykayand you can download it there too argh...
12:01:10 Quit AndyI (Read error: 110 (Connection timed out))
12:04:18linuxstbJaykay: Thanks - I've downloaded it from I can't see anything that looks like gapless info in that file, which is why Rockbox can't play it gaplessless. I'm not an expert on mp3 gapless though - maybe someone else will read this and double-check.
12:13:19Mikachucan a file be encoded gapless with itself even?
12:16:18linuxstbMikachu: Why not? What's the difference?
12:16:50Mikachui don't know :)
12:17:07CIA-38New commit by bluebrother (r21201): Clean up configuration dialog debug output.
12:20:24 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:21:32 Join ender [0] (
12:23:37 Join moos [0] (i=mustapha@rockbox/staff/moos)
12:24:31 Quit scott_67 (" ajax IRC Client")
12:33:15 Part DarkAnthey
12:34:24 Quit ender` (Read error: 110 (Connection timed out))
12:35:58 Join fdinel [0] (
12:36:07 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
12:37:22***Saving seen data "./dancer.seen"
12:37:24 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
12:40:37 Join AndyI [0] (i=AndyI@
12:43:20 Quit AndyI (Client Quit)
12:45:33 Quit stoffel ("leaving")
12:49:22 Quit ender (Read error: 110 (Connection timed out))
12:50:06 Join AndyI [0] (i=AndyI@
12:51:39 Quit perrikwp (" ajax IRC Client")
12:52:02 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
12:53:52amiconnHmm. I wonder what would be the best place to put bit_n_table for plugins on SH1
12:54:13amiconnImho a whole new .c file for just this table would be overkill
12:55:10amiconnI could put it into gcc-support.c, but it isn't really gcc support. And other pluginlib .c files are even more specialised
12:58:09 Join at0m [0] (
12:58:40 Join marek__ [0] (
12:58:47 Nick marek__ is now known as IuDeX (
12:59:15IuDeXhey. Anyone got a new Clip BootloadeR?
13:02:03 Join stripwax [0] (
13:03:34 Quit IuDeX (Remote closed the connection)
13:06:45 Join biengo [0] (
13:13:46 Quit perrikwp (" ajax IRC Client")
13:13:53 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
13:23:30 Join barrywardell_ [0] (n=barrywar@
13:24:11 Nick fxb__ is now known as fxb (
13:30:32 Nick fxb is now known as fxb__ (
13:31:04 Quit petur ("party time")
13:32:41 Nick fxb__ is now known as fxb (
13:40:43Unhelpfulamiconn: yes?
13:41:04amiconnSee my question ~50 minutes ago
13:42:04Unhelpfulmhm. i had created a support.c for it, with the expectation that there might be other things that could use it... it *does* seem a little silly in a file of its own, though.
13:49:25 Join HellDragon [0] (n=jd@
13:52:18 Join efyx_ [0] (
13:57:37Unhelpfulamiconn: is misaligned access the *only* trigger for data abort messages on ARM?
14:00:58 Quit perrikwp (" ajax IRC Client")
14:01:02Unhelpful...did somebody add new, and extremely verbose vorbis metadata debugging? :/
14:01:09 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
14:05:15 Join ender` [0] (
14:05:32 Quit crashd (Read error: 104 (Connection reset by peer))
14:05:47 Join crashd [0] (
14:06:22 Quit hd (Connection timed out)
14:10:11 Part Etu
14:12:10 Join ender [0] (
14:12:26stripwaxUnhelpful - ?
14:12:54Unhelpfulstripwax: the sim is suddenly dumping *all* comments to the console while building cache in pictureflow.
14:14:25stripwaxUnhelpful - oh .. I hadn't seen that, doesn't sound ideal
14:14:57Unhelpfulthe albumart "album art found" spam was bad enough, and still useful. but this is a bit much. :)
14:15:21 Join hd [0] (
14:15:32*stripwax tries to work out if profiling on arm targets is broken again
14:16:59Unhelpfuli didn't even know we had profiling on target :)
14:17:33 Join stoffel [0] (
14:18:18stripwax:) yep. need to turn it on in advanced build options, and also add profiling start/stop instructions to source code running in the thread you want to profile. whether it currently works or not is a matter for investigation.. certainly used to
14:19:04stripwaxwill see if H120 target profiling works .. arm seems to not register any function calls at all, so empty profile result
14:19:21Unhelpfulmakes me wonder if i *actually* need one-off benchmark plugins for jpeg, scaler, etc. ;)
14:20:50 Join HellDragon_ [0] (
14:20:56stripwaxUnhelpful - probably still useful, as the benchmark can drive the code in specific ways (e.g. like we still need test_codec) without needing to instrument the code first
14:21:04 Quit perrikwp (" ajax IRC Client")
14:21:13stripwaxuseful 'as well' rather than 'instead' that is :)
14:21:31 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
14:23:44Unhelpfulhrm, also re: beast hangcheck timer, it needs to be disabled on panic, or at least set longer before displaying panic message. beast goes off a few seconds after panic :/
14:24:22stripwaxyeah, getting zeros on H120 also. profiling seems borked
14:25:16 Quit hd (Read error: 60 (Operation timed out))
14:30:01 Quit ender` (Read error: 110 (Connection timed out))
14:30:28 Quit HellDragon (Read error: 110 (Connection timed out))
14:32:25r0b-i wish someone told me mpeg 1 was better then mpeg2 :P
14:34:06Unhelpfulit was? i don't see how, really.
14:37:23***Saving seen data "./dancer.seen"
14:38:23 Join gregzx [0] (
14:38:39*stripwax smacks forehead after realising $(PROFILE_OPTS) isn't included in most makefiles .. only some ..
14:39:31Unhelpfulhrm... i'll need to consult hardware docs to see if there's a way to set the action taken by the hardware watchdog :/
14:52:42 Join teru [0] (
14:53:11 Join perfectdrug [0] (
14:54:36 Quit Zarggg ()
14:55:07 Join HellDragon [0] (
15:05:22 Join Zarggg [0] (
15:06:59 Quit Nico_P (Remote closed the connection)
15:07:11r0b-mpeg1 for a mobile video isnt bad i guess
15:12:03 Quit HellDragon_ (Read error: 110 (Connection timed out))
15:13:37 Join hd [0] (
15:27:19 Join funman [0] (n=fun@rockbox/developer/funman)
15:27:39 Quit HellDragon (Read error: 110 (Connection timed out))
15:28:37funmanUnhelpful: permission, domain, and virtual address translation error causes data abort on ARM
15:29:14*Unhelpful suspects buflib_buffer_in/out
15:31:44 Quit DarkDefender (Remote closed the connection)
15:32:38funmanUnhelpful: you can get more information in UIE() by reading cp15 registers 5 & 6
15:33:08funmansee the technical reference manual for your cpu (i'm reading arm922t but i think it's similar)
15:33:10r0b-i cant believe mpeg1 is able to encode a 700MB .avi file to a 300MB.mpg file
15:33:23*funman points r0b- to #rockbox-community
15:33:41 Part r0b-
15:36:24teruWould someone please confirm FS #10249?
15:38:45funmansaratoga: ping
15:40:10n1steru: i think someone was working on a more complete fix for it or something but i don't' remember who
15:40:32n1shmm, my memor says Soap
15:41:20soapI'm not that smart.
15:43:57n1sah, it's AlexP, sorry
15:44:14perfectdrugis it possible to translate the manual in a smart way so that when something changes the changes are shown and you just have to update the missing strings? like the languagefiles? is this even considered?
15:44:56n1sperfectdrug: there is no such system in it currently at least
15:45:16perfectdrugwould it be possible?
15:45:53 Quit daurn (Read error: 110 (Connection timed out))
15:45:56 Join daurn| [0] (
15:47:19perfectdrugI just think when someone translates the manual of a stable rockboxversion if it would be wasted when you lost focus of the upcoming changes for later versions
15:57:24bertrikfunman, have you tried FS #10285 (alternative clip button driver) yet?
15:57:35 Quit stoffel ("leaving")
16:00:04funmanbertrik: no, i'm running a battery benchmark on my Clip (got 4:30 runtime with SVN and LCD always on)
16:03:09funmani'll enable logf on my fuze to try to see why flac files skip with fs#10048 (i never used logf before)
16:06:00 Join gartral [0] (
16:07:31terun1s: ok. thanks.
16:07:38 Join stripwax5443 [0] (
16:08:11gartralthe FeatureComparison page on twiki says the Sansa OF can't record, this isn't quite true, that should be "Yes -Partial/Badly", shouldn't it?
16:09:11n1sperfectdrug: it would probably be possible but i don't know how feasible it would really be, we are only barely keeping up with the manual as it is and translated variants would probably lag behind and confuse users even more
16:12:36gartralit also says Sansa Firmware records WAV only right under that
16:13:02 Quit moos (
16:13:02 Quit KBH (
16:13:02 Quit freqmod (
16:13:02 Quit FrankTM_ (
16:13:02 Quit DaCapn (
16:13:02 Quit fxb (
16:17:03funmangartral: it also says "Wrong facts? Feel free to edit this page."
16:17:43funmanthat said I don't know if "Sansa firmware" applies to both c200 and e200
16:22:53 Quit stripwax (Read error: 110 (Connection timed out))
16:26:36 Join robin0800_ [0] (
16:32:41 Quit perrikwp (" ajax IRC Client")
16:33:57 Quit robin0800_ (Remote closed the connection)
16:34:04 Quit lyngaas (Remote closed the connection)
16:34:21 Join robin0800_ [0] (
16:35:12 Quit robin0800_ (Remote closed the connection)
16:37:22 Join Juice^ [0] (
16:37:27***Saving seen data "./dancer.seen"
16:38:50 Join lyngaas [0] (
16:39:55NJoinmoos [0] (i=mustapha@rockbox/staff/moos)
16:39:55NJoinKBH [0] (
16:39:55NJoinfreqmod [0] (
16:39:55NJoinFrankTM_ [0] (
16:39:55NJoinDaCapn [0] (
16:39:55NJoinfxb [0] (
16:40:07 Quit robin0800 (Read error: 110 (Connection timed out))
16:42:24funmani'm trying to enable logf in pcmbuf.c but my fuze will lock on startup
16:43:19funmanlogf("pcmbuf len: %ld", seconds); will lock, logf("pcmbuf len: %ld"); will NOT, logf("pcmbuf len:", seconds); will -> I deduce that logf called with more than 1 argument will lock my fuze
16:47:42 Join stoffel [0] (
16:49:36 Quit onlysoaa (" ajax IRC Client")
16:50:44stripwax5443funman - does logf( "test %s", test ) lock? or something odd going on with the format string
16:50:48 Nick stripwax5443 is now known as stripwax (
16:51:44 Quit gartral ("leaving")
16:52:18stripwaxI mean logf( "test %s", "test" ) obviously
16:52:25funmanit fails as well
16:52:57funmanalso logf("bla", x); will lock and there is no format
16:56:05funmanhum my 'x' is a size_t perhaps that can harm
16:56:23 Join mazling [0] (
16:59:37funmanmaking it an int and using %d doesn't improve the situation - but using logf("bla %d", x); works fine >>
17:00:00funman... using logf(("bla %d", x); _somewhere else_ in pcmbuf.c
17:06:17 Quit teru ("Quit")
17:07:24 Join saratogahome [0] (n=41becb3b@gateway/web/cgi-irc/
17:07:31saratogahomefunman: pong
17:08:09 Quit mazling ("o/~ locked in silent monologue - in silent scream... o/~")
17:11:48funmansaratogahome: i was wondering if you used the latest patch on FS #10048 in the build you provided on ABI fuze forum, but i checked and this is really the last patch. I just posted a comment on the task : i think we still have problems with storage transfers, and this confirms what the people on the abi thread are seeing
17:18:26saratogahomefunman: i downloaded that patch just a minutes before posting the build, so it should be latest
17:18:37saratogahomei wonder why test_disk doesn't catch the problems though
17:18:50funmanyeah no doubt
17:19:04funmanperhaps the problems are not frequent enough for test_disk to catch them in one run
17:20:53funmanalso i don't know why we didn't see errors with other codecs than flac since I would say higher compression = less error tolerance
17:20:59funman(error in the bitstream)
17:23:52 Join togetic [0] (n=ib@unaffiliated/ibuffy)
17:24:02 Quit perfectdrug ("Leaving.")
17:24:19funmanand now my fuze doesn't boot >> perhaps r21200 which incidentally changed the FM code on Sansa ams
17:25:41saratogahomeperhaps lossy codecs just don't transfer enough from the disk to cause an error frequently
17:26:44funmansounds sensible. anyway i can't boot anymore with fs#10048, even r21200 applied. will do further tests later
17:31:51 Quit alexbobp (Read error: 110 (Connection timed out))
17:34:36 Join Lear [0] (i=chatzill@rockbox/developer/lear)
17:36:14 Quit stripwax (Read error: 110 (Connection timed out))
17:41:34 Quit funman ("leaving")
17:42:00 Join taylor_ [0] (
17:45:56 Join alexbobp [0] (
17:46:45 Join net [0] (i=c1ee8945@gateway/web/ajax/
17:49:14net<Kutaragi> The 'Walkman' of the 21st century
17:50:30saratogahomeprobably the wrong channel
17:52:47net<Kutaragi> is the iPod the walkman of the 21st century?
17:53:36krazykitnet, this is an on-topic channel about rockbox. please read the guidelines linked in the channel topic
17:56:21 Part net
17:56:24 Join net [0] (i=c1ee8945@gateway/web/ajax/
17:57:10 Part net
17:57:11 Join net [0] (i=c1ee8945@gateway/web/ajax/
17:57:29 Part net
17:59:59 Quit stoffel ("leaving")
18:01:17 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
18:10:50 Quit taylor_ ("Leaving")
18:19:24 Quit n1s ("Lämnar")
18:21:00 Quit saratogahome ("CGI:IRC (EOF)")
18:28:04amiconnUnhelpful: Btw, BIT_N() in plugins is not just for speed - if may cause the shift-by-n subroutine to be no longer needed, and then it makes the plugin smaller
18:28:13amiconnHappens e.g. for alpine_cdc
18:30:34*amiconn thinks that might also be due to a related optimisation though (replacing /4 by >>2 and %4 by &3)
18:31:52 Join robin0800 [0] (
18:35:24 Quit fyrestorm ("lamers envy me like they envy bill g -- main boot xp, just the way it should be!")
18:37:18 Join n00b81 [0] (
18:37:30***Saving seen data "./dancer.seen"
18:43:52 Quit itcheg (" ajax IRC Client")
18:51:20 Join fyrestorm [0] (
18:54:05 Join togetic1 [0] (
18:54:30togetic1i restored ipod using itunes on my bro's windows computer
18:54:41togetic1so i have the firmware
18:54:45togetic1and then i installed rockbox
18:55:01togetic1and in .rockbox, created folders for my audio
18:55:06togetic1how do i start rockbox?
18:55:30togetic1i stopped the device using itunes
18:55:38togetic1removed it from the usb outlet
18:55:43 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
18:55:46togetic1couldn't figure out how to start rockbox
18:55:58togetic1then i plugged it back in and created the folders and put the audio there
18:56:09togetic1however, now i don't know how to start rockbox
18:56:15togetic1could someone please help?
18:57:28togetic1i only have about 5 more mins that i can stay here before we leave on this mini-vacation
18:57:49 Nick n00b81 is now known as taylor_ (
18:57:50krazykiti suspect you have to reboot the ipod
18:57:56togetic1how do you do that?
18:58:24amiconnUnhelpful: jpeg viewer is sped up bit BIT_N() on recorder (+8% for large images when decoding at 1/8 - the speedup is in the huffman decoding)
18:59:16 Quit taylor_ ("Leaving")
18:59:36pixelmacould you also please try to put more of your thoughts on one line? Makes it a lot easier to read...
18:59:37Lloreantogetic1: Also, you don't need to put your audio in .rockbox, in fact of all the places it can be put (which is, anywhere) that's probably the worst since it's a hidden folder
18:59:40amiconnUnhelpful: (1) What about the jpeg code in apps/recorder/ - is it related in any way to apps/plugins/jpeg? Also, is it used on SH1 (pictureflow?)?
18:59:40togetic1running rockbox now :)
19:01:24Unhelpfulamiconn: apps/plugins/jpeg oes not presently use the core-import decoder. there are defines for an in-memory version of it, and that is built in pluginlib but not used anywhere yet. we *could* use that decoder in pluginjpeg. pictureflow uses the core decoder on targets that have it, and a pluginlib import of it (without the in-memory code) on targets that don't.
19:02:33amiconnSo the code in apps/recorder/ does get used on SH1?
19:02:50Unhelpfulin pictureflow, yes, it does.
19:03:39 Join Chesteta [0] (
19:04:20Unhelpfulthe in-memory version of the core decoder is part of a longer-term plan to get the jpeg plugin unified with the core decoder, so if that gets finished, BIT_N in jpeg_load.c would help there, too.
19:05:03Chestetahello, I just did an svn update and recieved a message saying "conflict discovered in 'firmware/drivers/generic_i2c.c' I can select postpone, diff-full, edit, mine-conflict, theirs conflict or show all options... any tips?
19:05:35UnhelpfulChesteta: you need to merge the changes you've made with the ones from svn.
19:05:49 Quit flydutch ("/* empty */")
19:05:57ChestetaI believe this is due to the fm patch I have applied... would 'theirs conflict' basically unpatch the file and get the latest file?
19:06:23 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
19:06:39Unhelpfuli'm not sure, i don't work with svn that much. just using the svn version of that file with the rest of your files patched might be bad, though.
19:06:47 Join soycamo [0] (
19:07:12Chestetaunhelpful: thanks, I did not make the changes, (well I did patch it but did not write the code) so not sure how to merge it...
19:11:26 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
19:11:42Chestetahmm I think im just going to use 'postpone' then...
19:12:49 Join merbanan [0] (
19:13:13Chestetathen ill 'unpatch' then update and repatch
19:17:08 Quit Chesteta ()
19:18:08 Join Xerion_ [0] (
19:26:39 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
19:27:16 Quit Llorean (Read error: 54 (Connection reset by peer))
19:27:41 Join Llorean [0] (
19:33:53 Quit dmb ("Leaving")
19:34:26 Quit Xerion (Read error: 110 (Connection timed out))
19:34:26 Nick Xerion_ is now known as Xerion (
19:40:32 Quit soycamo (Read error: 110 (Connection timed out))
19:41:07 Quit itcheg (" ajax IRC Client")
19:41:28 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
19:43:35 Quit biengo (Remote closed the connection)
19:51:02 Quit togetic1 ("Leaving.")
19:53:23 Join stripwax [0] (
19:54:11 Join wincent [0] (
20:10:29 Join bluebrother [0] (
20:35:20 Quit itcheg (" ajax IRC Client")
20:37:35***Saving seen data "./dancer.seen"
20:44:52 Join Chesteta [0] (
20:45:17Chestetahello, is flyndice or funman around?
20:46:43GodEaterdoesn't look like it
20:46:48Chestetahe mentioned I needed to use -p1 for funman patches... I had done this; the rest of the patch did apply correctly however the 'sources' part did give that error
20:46:51GodEaterah, so you're set away, but pretending ? :)
20:47:36Chestetafigured it would be better to try and talk to you in person rather than through the thread ;)
20:48:07Chestetaany advice?
20:48:34ChestetaI am sure I used -p1 because it did patch, just gave that error in the begining
20:49:14FlynDiceChesteta: I tried applying the latest patch and it applied fine to svn. You could try reverting that 1 file I guess but I'm not really an svn expert
20:49:36Chestetaoh; ill just re-checkout everything i suppose
20:49:54 Join balug_ [0] (
20:49:57Chestetai just had which is why i posted the comment since i had no other patches applied
20:50:01 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
20:50:14Chestetamaybe i just got a bad checkout?>
20:50:14 Join petur [0] (
20:52:36FlynDiceChesteta: can you just look at the diff and make the correction yourself, it's probably just 1 line in sources?
20:53:00Chestetaumm I can try... not much of a coder :/
20:53:22Chestetaim willing to give it a shot if you can answer questions :)
20:53:38FlynDiceits not really coding it's cut & pating...
20:53:43Chestetaoh hah
20:53:44FlynDiceer pasting
20:53:52Chestetaim a pro at cutting and pasting ;)
20:54:30Chestetacan i use notepad to edit the diff or does that not support all of the characters? (i know sometimes notepad messes sourcecode up)
20:54:46FlynDicedid it leave you a sources.rej file?
20:55:52Chestetawhat directory would that be at? it said something about apps\(cant remember)\sources in the error
20:56:37Chestetai couldnt find the directory it was in
20:56:48FlynDicetry apps/firmware/sources I think
20:57:19Chestetahmm wierd its not there :/
20:57:32FlynDicewait, was it in apps?
20:57:33Chestetathats probably why it didnt work (doh)
20:57:52Chestetaim missing the directory for some reason ??
20:58:09Chestetathats why it didnt patch properly, sorry for the confusion
20:58:12stripwaxthere isn't an apps/firmware directory.. just apps/ or firmware/ ..
20:58:40FlynDiceyes, my faulty memory I'm finding the dir right now
21:00:20saratogahas anyone setup an ARM emulator for testing code?
21:01:00bertriksaratoga, funman did this IIRC when experimenting with the ams sansa dualboot loader
21:01:34Chestetaflynndice: the checkout finished, it patched just fine this time
21:01:55FlynDiceok, after scrolling up to take a look it seems the generic_i2c.c file was the one creating the problem, for some reason I thought it was one of the SOURCES files
21:02:22 Join jmillikin [0] (
21:02:22FlynDiceWhy is that always quicker... ;)
21:02:23bertrikuh oh I last worked on that file
21:03:09FlynDicebertrik: I think your file is fine, it was a patching issue I think
21:04:22Chestetaactually when i try to apply generic_i2c_fix.patch to r21201, it says reversed or previously applied patch detected
21:04:49Chestetaso was your last work doing the same thing as generic_i2c_fix.patch?
21:05:15FlynDiceChesteta: Ithink that's the patch he committed
21:05:57 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
21:07:09Chestetaohh, it doesnt say its closed in fs#10267 so thats why i tried applyin it
21:13:49CIA-38New commit by Domonoky (r21202): pdbox: new patch by wincent. make s_print work on sim.
21:18:32 Quit itcheg (" ajax IRC Client")
21:20:19domonokyhi wincent, just commited you latest patch :-) (there where still a few instances of "x=x" instead of "(void)x", but i changed it before commiting.)
21:21:43wincentThanks :-)
21:24:52 Join Lars_web [0] (n=Lars_web@
21:24:56Lars_webHey all
21:25:14Lars_webI come for one second using an ugly irc applet, to make a question
21:26:02Lars_webI have a Sansa with rockbox, it's currently totally discharged, my lattest pull was compiled afaik with usb support
21:26:25Lars_webstill when I hook it into the pc to recharge and listen to music, rockbox launches and then launches the OF (it's rockbox doing it, not the bootloader)
21:26:30 Quit barrywardell_ ()
21:26:45saratogaboot first, then plug in the USB
21:26:48Lars_webis there any way I can launch rockbox telling it to use usb in rechage mode only and stay on rockbox, so I can play music from rockbox while recharching?
21:26:59Lars_websaratoga: too little juice to boot without connecting
21:27:13LloreanThen you need to let it charge a little first.
21:27:19saratogarecharge it then
21:27:26Lars_webAnd once I do it
21:27:37Lars_webHow can I tell rockbox to recharge only? not to enter disk mde
21:27:39LloreanLars_web: Hold "select" while inserting the USB cable in rockbox, exactly like it tells you in the manual
21:27:57Lars_webthanks llorean, I looked but missed it on the manual
21:28:54bertrikhm, I'm getting immediate data aborts on my clip with funman's latest patch when trying to start playback
21:30:19Lars_webok I'm off to continue fighting the router and net setup
21:30:22Lars_webthanks guys
21:33:34saratogaapt-get on an emulated arm box is really really slow
21:34:38Lars_webhav you tought about negative playlist insertion?
21:34:45Lars_webI have 3 audiobooks on my rockbox
21:35:11Lars_webit'd be nice to be able to insert all artists on the db shuffled, and then de-insert the two audiobook artists..
21:35:48LloreanYou could use comments, genres, or simply make use of the filetree to avoid accidentally inserting audio you don't want in your playlist.
21:37:38 Quit Chesteta ()
21:39:27 Quit bmbl ("Woah!")
21:43:00 Quit Lars_web (Read error: 60 (Operation timed out))
21:45:37 Quit kkurbjun ("Leaving.")
21:46:41 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.5pre/20090605044036]")
21:46:43 Quit hd (Client Quit)
21:56:44 Quit petur (Read error: 104 (Connection reset by peer))
21:57:14 Join HellDragon [0] (
21:57:24 Join petur [50] (n=petur@rockbox/developer/petur)
21:58:02 Quit moos ("Rockbox rules the DAP world")
22:00:56 Quit bluebrother (No route to host)
22:04:14 Quit HellDragon (Read error: 104 (Connection reset by peer))
22:07:36 Join HellDragon [0] (
22:09:11 Join mcuelenaere [0] (
22:11:57 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
22:12:27 Join rasher [0] (
22:12:48 Join mcuelenaere [0] (
22:16:43 Quit merbanan ("Leaving")
22:27:16 Join bluebrother [0] (
22:29:52CIA-38New commit by mcuelenaere (r21203): Onda VX747: 'touchscreenize' the play button on the cabbiev2 theme
22:30:12bluebrotherlinuxstb / domonoky: any comments about FS #10220? I'd like to get that ready for committing the next couple of days.
22:32:55 Quit wincent (Read error: 110 (Connection timed out))
22:37:39***Saving seen data "./dancer.seen"
22:40:06 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:41:05kugelbertrik: what speaks against using my patch? Speed up is speed up
22:44:55 Quit balug_ ("Ex-Chat")
22:47:16 Join arcticfang [0] (n=4b5904fe@gateway/web/cgi-irc/
22:48:24domonokybluebrother: no comment. if it works, put it in :-)
22:49:57arcticfangHey everyone. Another question. I've got XCode for Mac. I'm wondering, 1) How can I make the .rockbox folder unhidden when viewing my iPod on the Mac? and 2) How can I insert patches to Rockbox using XCode?
22:52:30Unhelpfulhas any attempt ever been made at simulating the rockbox scheduler in the sim?
22:53:00saratogai think we currently use SDL for that right?
22:53:27arcticfangCan anyone answer my 2 questions?
22:53:45krazykitarcticfang, please have patience in waiting for an answer
22:54:37arcticfangSorry, just trying to hurry. ^_^;; My apologies.
22:55:18Unhelpfulsaratoga: as i understand it, we use SDL threads, which are not far removed from system threads. we could simulate scheduling perhaps by having yields set a per-thread lock, and wait on it, and then adding a scheduler thread that unblocks RB threads appropriately.
22:56:12bluebrotherok :)
22:56:13 Quit FlynDice (Read error: 54 (Connection reset by peer))
23:00:47Lloreanarcticfang: How to view hidden folders is more of a thing specific to your operating system, not really Rockbox-specific. Try google
23:00:55LloreanAs to "inserting patches", there's a guide to compiling on our wiki that should help\
23:04:20 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
23:07:43 Quit arcticfang ("CGI:IRC (EOF)")
23:08:36 Join arcticfang [0] (n=4b5904fe@gateway/web/cgi-irc/
23:09:48arcticfangThanks Llorean. I figured it out. Now I have 2 more questions - 1) How can I install patches to rockbox on my Mac, and 2) Are .cpp files compatible with Rockbox, or is it just .c files?
23:10:30 Join Insilic0 [0] (n=4a45af4d@gateway/web/cgi-irc/
23:10:37Lloreanarcticfang: I already told you what to look for regarding patches.
23:11:33LloreanAnd .cpp and .c files are just source code. Rockbox is written in C so most files with the extension .cpp probably aren't usable, but then they wouldn't be a patch for Rockbox anyway.
23:12:50arcticfangLlorean: Sorry, Firefox quit on me so I didn't see your last post. Where should I look regarding patches? And also, is there any way to make a .cpp file into a patch?
23:13:08Mikachuarcticfang: the chat is logged fyi
23:13:16bertrikkugel, what patch? you mean the DBOP FIFO with interrupt patch?
23:13:24Lloreanarcticfang: The developers portion of the wiki. And, if you don't know C and C++ a .cpp file won't do you any good.
23:13:35bertrikI think for the clip it's not worth the complexity over the simpler patch and clip is already plenty fast
23:13:41Lloreanarcticfang: A patch is just a list of source code changes. To make one you have to know programming.
23:14:34 Join ademille [0] (
23:14:40kugelbertrik: I can't really understand prefering a slower way of doing in a driver
23:15:20kugelalso, it's not really complex at all, imo
23:15:29 Quit Insilic0 ("CGI:IRC (Ping timeout)")
23:15:32ademilleMaurus, I lied about my vx777 sd card. It's actually a 4gb SDHC.
23:15:47arcticfangLlorean: I'll check that out. ^_^ I've been getting into C++ recently. I know what a .cpp file is, and I somewhat know the basics of C++. Would I need to change to C instead for my work to be compatible with Rockbox?
23:16:00bertrikkugel, do you even know why it is faster, then?
23:16:07kugeland I wouldn't want different code doing the same thing on our AMSes (I still have unified drivers in my mind)
23:16:17*bluebrother points out that he can write programs using .txt files. The file extension doesn't matter ...
23:16:34mcuelenaereademille: ah, that could explain some things
23:16:36Lloreanarcticfang: As I said, Rockbox is written in C. You should spend some time practicing writing plugins for Rockbox and learning about it.
23:16:38bluebrotherarcticfang: Rockbox is C, with a bit of asm.
23:16:43kugelbertrik: probably because polling hardware pins is slower than reading a variaible
23:16:49mcuelenaereademille: have you tried editing those files as I asked you?
23:17:16bertrikkugel, and do you also plan to have the same fix in the asm greyscale driver?
23:17:49bluebrotherwhile C and C++ are usually treated equally the languages are in fact rather different. Given that you use the specific language features. And while C++ does understand _most_ C it is not simply a superset of C.
23:17:51arcticfangThanks. :) I will go check out that developer's section.. thanks again. ^_^
23:17:51bertrikI still think it's unnecessary complexity for clip
23:18:32kugelwhere is that complex?
23:18:34 Quit arcticfang ("CGI:IRC")
23:19:08ademillemcuelenaere: I'll have to setup the development environment. I don't have a cross compiler setup yet. I just pulled down the latest from svn.
23:19:31mcuelenaereademille: ok, you know of tools/
23:20:00ademilleI think I read about that couple months ago. Will that setup my cross compiler?
23:20:22mcuelenaerejust select mips and add the toolchain to your path afterwards
23:20:27ademilleGood deal. It can be a real pain to get those right.
23:20:32ademilleWill do thanks.
23:21:49bertrikkugel, should be obvious
23:22:06kugelit's not obvious to me
23:23:43kugelit's a tad bit more complex, yes, but far from too complex to justify a slower solution
23:24:49 Join solexx [0] (
23:24:57kugelare your numbers based off SVN or with the MMU patch (I seem to remember the difference was higher with the patch)
23:25:42solexxcurrent SVN doesn't build for h120 on my system
23:25:47kugelsvn+my patch vs mmu-patch + my patch that is
23:25:52solexxThe first error is "rockbox/apps/settings_list.c:796: error: `LANG_SORT_INTERPRET_NUMBERS' undeclared here (not in a function)"
23:26:13CIA-38New commit by learman (r21204): Fix missed text change in beep volume, and adjust related strings to match (good for the keyclick setting too).
23:27:09CIA-38New commit by amiconn (r21205): Convert a number of places in core and plugins to use the BIT_N() macro instead of 1<<n. Speeds up things on SH1, and also reduces core binsize. Most ...
23:27:22solexxBut the dev page on says current SVN is ok. What could be wrong on my side?
23:27:41Bagdersolexx: patches applied?
23:28:11solexxBagder: none, apart from my build directories the working copy is clean
23:28:25Lloreansolexx: Tried removing the build folder and creating a fresh one?
23:28:56solexxLlorean: not yet, I only tried make reconf. But I'll try now.
23:31:31bertrikkugel, how much faster is it with the interrupt? it's not exactly slow right now.
23:31:45bertrikand it annoys me that we don't know why it is faster
23:32:12bertrikthere is such a thing as trying to over-optimise things
23:32:29kugeluhm, polling hardware pins vs a in-ram variable doesn't sound right for you?
23:32:30solexxLlorean: strange, it works when I use a new build directory
23:33:14kugelbertrik: you call that over-optimising?
23:33:22bertrikwe're not polling hardware pins
23:33:57linuxstbbluebrother: My comments on your beastpatcher patch are the same as domonoky ;) You may want to point mcuelenaere to it though (FS #10220).
23:34:00solexxInterestingly, when I make clean, I still have a few files left: lang.{c,h,o} and max_language_size.h. They don't exist in the new build dir after building & cleaning.
23:34:06kugelbut a register that's not in RAM, but in a seperate unit
23:34:31kugelI can imagine that this is slower
23:34:46Lloreansolexx: There's "make veryclean" too
23:34:52mcuelenaerelinuxstb: ?
23:35:21linuxstbmcuelenaere: See FS #10220 - bluebrother asked if I had any comments on it.
23:36:16kugelcalling it over-optimizing seems hilarious to me. It's not really complex at all (also, comments can explain it nicely). I really think we should get the maximum peformance we can get out of our drivers
23:36:47mcuelenaereah ok. Well I don't currently use Windows anymore so I don't know whether I'm needed here (and I believe bluebrother is more familiar with Win32 than me)..
23:37:11linuxstbmcuelenaere: Then I guess we'll all leave bluebrother to it ;)
23:37:38*bertrik misses the hilarity somehow
23:37:47solexxLlorean: oh, didn't know that. Thanks. Apparently those files were to blame. make veryclean doesn't remove them either, though.
23:37:51kugeland I miss the complexity
23:38:03bertrikok, start coding every driver in asm then
23:38:25bertrikhow much faster is it anyway? do you know?
23:41:53kugelit was noticeably faster on the fuze unboosted vs your patch (65 or so to 90)
23:41:57 Quit jordan` (Read error: 104 (Connection reset by peer))
23:42:15 Join jordan` [0] (i=gromit@
23:42:37kugelthat's with the mmu patch
23:42:39amiconnsolexx: Those files changed position at some point. New 'make clean' won't delete them in their old places
23:43:05kugelcan this be fixed? that's annoying
23:43:08amiconnDelete them by hand, or wipe the whole build dir and configure a fresh one
23:43:25 Join moos [0] (i=mustapha@rockbox/staff/moos)
23:43:57amiconn'make veryclean' is rarely needed. The difference to 'make clean' is that 'make veryclean' cleans the tools as well
23:46:31bertrikI get 2156 fps unboosted and 3562 fps boosted now, I think the display refreshes the screen at 80 fps
23:48:28kugelso my patch is 700fps faster unboosted
23:50:02kugel(with the mmu patch)
23:55:39*amiconn is a bit puzzled regarding the deltas
23:56:00amiconnThere shouldn't have been any delta for arm/ coldfire/ mpis
23:56:06*amiconn blames gcc
23:56:35Bagdernah, surely gcc is innocent! ;-)

Previous day | Next day