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 2009-08-22

00:00:16MG_ManAm I correct in assuming that the h300's CPU is more powerful (140 mHz as compared to 80)
00:00:26AlexPSort of
00:00:38AlexPThe ipod is dual core, and architecture differences matter too
00:00:47AlexPIt isn't just a case of raw MHz
00:01:04linuxstbAnd the CPU in the h300 (and h100) is run at 124MHz.
00:01:25MG_ManHmm, I don't think the ability to buffer MP4s and AACs and such doesn't matter much to me, as I usually do MP3s and OGGs
00:01:39AlexPThen anything is fine
00:02:59MG_ManAn h320 for .99 on Ebay
00:03:04MG_ManThat's too fishy
00:06:49 Quit jfc (Read error: 104 (Connection reset by peer))
00:08:20BdN3504pixelma: i double checked, and it's really not working. please try to edit the patch i will be posting in a bit, but i doubt you'll succeed.
00:10:21AlexPBdN3504: What are you trying to add?
00:11:18MG_ManCan the H300 series charge through the USB port or only a DC plug?
00:13:10 Quit bluebrother ("leaving")
00:14:20***Saving seen data "./dancer.seen"
00:16:40BdN3504i'll post the patch in a minute, just let me check out. it's easier for me to explain, if you can look at the file yourself.
00:16:52eulerphisaratoga: will aac-lc work?
00:18:04saratogayes it will
00:18:09eulerphithanks :)
00:18:24eulerphiFAAD: compressed input bitrate missing, assuming 128kbit/s! ^_-
00:18:30eulerphiwonder how that happened
00:19:53eulerphii set it to 64kbps, but it's assuming 128kbps
00:20:00eulerphiin mplayer
00:21:15saratogawhy would you set a rate for a decoder?
00:21:25eulerphii mean when i encoded it
00:21:31eulerphiwith ffmpeg
00:21:57saratogaFAAD is a decoder, not an encoder
00:22:04eulerphii know
00:22:15eulerphiwhen i try to play it, i get that message
00:22:32saratogathats probably some kind of bug in whatever program you're using
00:25:14 Quit HellDragon (Read error: 104 (Connection reset by peer))
00:25:20 Join HellDragon [0] (
00:26:56 Quit robin0800 (Remote closed the connection)
00:31:52 Quit antil33t1 (Read error: 110 (Connection timed out))
00:33:33Mode"#rockbox -o scorche|sh " by ChanServ (ChanServ@services.)
00:36:37 Quit jgarvey ("Leaving")
00:37:30CIA-61New commit by mcuelenaere (r22459): Onda VX747/VX777 bootloader: make boot menu more robust + add reset Rockbox configuration option
00:38:07CIA-61New commit by mcuelenaere (r22460): Rockbox Utility: fix some typos in Dutch translation
00:38:58 Join JdGordon [0] (i=d80ca5a2@gateway/web/freenode/x-uyppyqybzcnkctnq)
00:47:35 Quit Hillshum (Read error: 110 (Connection timed out))
00:49:00saratogamcuelenaere: have you seen the red?
00:50:53CIA-61New commit by mcuelenaere (r22461): Add a BTN_MASK for the VX777 (aka fix red)
00:54:26CIA-61New commit by mcuelenaere (r22462): A new implementation of logf, logfdisplay and logfdump. ...
00:58:52 Quit JdGordon (Ping timeout: 180 seconds)
01:00:42 Part toffe82
01:10:25BdN3504how do i adress only monochrome targets in the manual? using \opt{player} doesn't seem to work, as the manual for "0" show up stuff which i actually opted out
01:12:29 Quit saratoga ("Page closed")
01:12:41pixelmaI`d invert the lcd_non-mono with an \nopt
01:14:06pixelmathe Player is only one very specific display (charcell) which is monochrome still, but the other monochrome screens are not charcell
01:14:28pixelmaand are not Player...
01:15:49pixelmaBdN3504: I don't understand the second part of your statement, so don't know what you want exactly
01:15:54BdN3504pixelma: displaying images can be done by which targets?
01:16:17 Quit captainkwel ("Page closed")
01:16:45BdN3504i meant i compiled a manual for "player" and got a manual for all archos players in which a nopt i defined simply is ignored
01:17:33BdN3504namely a \nopt{player} is displayed, although i selected player in the ../tools/config
01:18:10pixelmaall bitmap displays can show images so... \opt{lcd_bitmap}
01:18:29pixelmaist that in an example or so?
01:19:38BdN3504noi, it's in a normal paragraph like blalbalbabla.\nopt{player}{blablabla{ref} blabla.}
01:20:14pixelmamaybe you forgot a } somewhere before
01:21:22AlexPUsing an editor with some sort of syntax highlighting is a good idea - it'll show you matching {}s
01:22:39BdN3504i use a miktex disto
01:23:14AlexP.tex are just text files, you can use any editor you want
01:24:18BdN3504Using an editor with some sort of syntax highlighting is a good idea - it'll show you matching {}s
01:24:27BdN3504miktex -> texnic center
01:24:33BdN3504sorry if i was unclear
01:24:55AlexPThey are still just text files
01:25:18 Join Thundercloud [0] (
01:26:15BdN3504then i don't get your point
01:27:17BdN3504anyway, i think i'll post the patch. if anyone can fix it, fine. if not i don't think it's such a big deal. this is in a very comittable state, the only problems occuring refer to these opts
01:27:25AlexPYou can edit them in anything you like, miktex, notepad, whatever. If miktex does things like syntax highlighting, then great.
01:27:54AlexPIf \opts not working lead to the wrong information getting into the wrong manuals, then it isn't committable
01:28:34AlexPBut anyway, more eyes on a patch is always a good idea
01:28:42AlexPWhat are you trying to document?
01:30:45BdN3504and the only target which gets wrong is the player
01:31:10AlexPthat's great, but they all need to be right before it goes in :)
01:31:16AlexPGood work incidently :)
01:31:47BdN3504word on the streets is only 2 persons worldwide currently own and use a player...
01:31:55AlexPThat doesn't matter
01:32:01AlexPEven if it were true
01:35:37BdN3504i have to include some newlines, to make it more readable, you got another 5 minutes?
01:36:14AlexPI'm off to bed soon, tomorrow should be OK :)
01:36:47pixelmalearning how things work in general and how to get them right will help too. \opts are not a Player specific thing...
01:36:57 Quit ender` (" It always takes longer than you expect, even when you take Hofstadter's Law into account. -- Hofstadter's Law")
01:42:20 Quit fml ("CGI:IRC")
01:43:08 Quit GeekShad__ ("The cake is a lie !")
01:44:42 Join efyx [0] (
01:45:45pixelmadoes the themes site re-evaluate themes that were already uploaded when CheckWPS is updated? I'm asking because I came across a c200 theme on there when testing RBUtility today which is broken in current SVN due to mistakes that were ignored before but now aren't anymore
01:46:11BdN3504i fear i simply worked on the wrong file :(
01:46:33BdN3504main.tex of the build instead of the tree...
01:48:55 Quit gevaerts (Nick collision from services.)
01:49:04 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
01:55:51BdN3504k lol, my own f*ing stupidity: i really worked on the file from the build and not from the tree. update on the tracker
01:58:18mcuelenaerepixelma: I don't think so, at least not on a regular basis. I only see a cronjob for updating checkwps on the theme server..
01:59:12BdN3504ye, please check my patch.
01:59:23 Quit BdN3504 ("CGI:IRC")
02:01:41 Quit Rondom (Read error: 60 (Operation timed out))
02:03:07mcuelenaerepixelma: it seems like it runs checkwps on every theme upon adding a new one though
02:03:58 Join Rondom [0] (
02:07:25 Join toffe82 [0] (
02:08:33 Join evilnick_home [0] (n=evilnick@
02:10:58rashermcuelenaere: it shouldn't
02:11:53mcuelenaererasher: doesn't it call checkallthemes() on addtheme() in themesite.class.php?
02:13:11rashermcuelenaere: oh heh, checkallthemes() is the wrong name
02:13:22rasherit's called with a theme-id of the newly added theme
02:13:28rasherwhich means it only checks that theme
02:13:37rasherI can see where the confusion arises...
02:13:50mcuelenaereah ok :)
02:13:59rasherThe plan was to set up a cronjob that hits the ?runcheckwps link on the admin page, but I never got round to that..
02:14:18mcuelenaereis that the only thing that needs to be done? (ie ?runcheckwps is already implemented)
02:14:24***Saving seen data "./dancer.seen"
02:14:43rasherSee the bottom of the admin frontpage
02:15:48rasherJust the issue of requiring login, but I guess that's not really necessary
02:16:19rasherThere should probably be some sort of time-lock on it though so it doesn't get hammered. It takes a while to run
02:16:38mcuelenaereI guess that a separate php script in private/ calling the function should do it? (as a cronjob)
02:17:14 Quit martian67 (Connection reset by peer)
02:17:16rasheroh of course. That's much smarter :)
02:17:30rasherI'll fix that up right away, unless you're on it?
02:17:51mcuelenaereI'll leave it to you :)
02:18:04 Quit togetic ("WeeChat 0.3.0-rc2")
02:18:15mcuelenaerehmm all themes currently on the server passed checkwps
02:18:20mcuelenaerepixelma: ^^
02:18:32 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:19:08 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
02:21:43 Join tvelocity [0] (
02:24:28 Quit togetic (Read error: 104 (Connection reset by peer))
02:24:38 Join jboy_ [0] (
02:24:52 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:25:18 Quit evilnick_home (Read error: 60 (Operation timed out))
02:26:47 Join tvelocity[a] [0] (
02:26:54 Quit tvelocity[a] (Read error: 54 (Connection reset by peer))
02:28:14rashermcuelenaere: could mean two things..
02:28:24 Join evilnick_home [0] (n=evilnick@
02:29:08rashermcuelenaere: Ah, the server lacks php-cli
02:29:33rasherscorche|sh: mind installing php5-cli on the themes server?
02:29:46scorche|shif i must...
02:30:03rasherYou must!
02:32:00kkurbjunmcuelenaere: awesome, work
02:32:07scorche|shapparently php will be getting upgraded to 5.2.10.dfsg.1-2 as well
02:32:18kkurbjun:) mrobe 500 works now for themes
02:32:31mcuelenaereso does Onda VX747 :)
02:32:39mcuelenaererasher: what two things?
02:33:05rashermcuelenaere: Either they are all flawless, or checkwps is too forgiving
02:33:12scorche|shwell, hopefully php stuffs wont break from this...i should update/upgrade packages on my server this weekend..
02:33:34*mcuelenaere guesses the second one is more likely
02:34:56scorche|shrasher: well, apparently it wants to nuke php5-sqlite{a} for this too...
02:35:18rasherscorche|sh: ?!
02:35:48 Join evilnick_home1 [0] (n=evilnick@
02:35:56 Quit togetic ("WeeChat 0.3.0-rc2")
02:36:18*scorche|sh directs rasher's glare towards aptitude
02:36:24rasherscorche|sh: Oh, you're running old packages and trying to install a single new one?
02:36:40scorche|shand it will be upgrading a bunch too
02:36:52scorche|shmaybe i should just upgrade them all this weekend
02:36:54rasherMaybe try apt-get
02:38:43 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:38:44scorche|shthat looks alright...
02:39:12 Quit togetic (Read error: 104 (Connection reset by peer))
02:39:47 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:40:03 Quit tvelocity (Read error: 110 (Connection timed out))
02:40:04 Quit togetic (Read error: 104 (Connection reset by peer))
02:40:46 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:42:00 Quit jboy (Read error: 101 (Network is unreachable))
02:47:33 Quit Thundercloud (Remote closed the connection)
02:51:09scorche|shrasher: try that
02:52:32 Quit evilnick_home (Read error: 110 (Connection timed out))
02:55:04 Quit evilnick_home1 ("Leaving.")
02:56:11 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
02:58:54 Join CaptainKwel [0] (
03:00:41MG_ManIs there a way to scroll an image across the screen for a progress bar other than using stacked BMPs?
03:00:47MG_ManNo I do not mean the current progress bar
03:01:11MG_ManBut say you have a box that you want to go across the screen
03:02:27goffa_hmm... having a wierd issue with my s120
03:02:42MG_ManIs there some way to use the playback percentage enumeration as the X coordinate or something? Or does wps script not do math functions like that
03:02:47goffa_sometimes when i delete 2 files in a row from the wps it locks and shuts down
03:04:16goffa_i guess my build is about a week old
03:04:24goffa_but didn't see anything relating to that problem
03:04:38 Quit togetic ("WeeChat 0.3.0-rc2")
03:04:51 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
03:04:56MG_ManLike for example, for the x coordinate of the image you want to move, have something like 176/%ps and it would update
03:05:02MG_ManOr is that not (currently) possible
03:05:39 Quit domonoky (Read error: 104 (Connection reset by peer))
03:06:04 Quit mcuelenaere ()
03:07:32MG_ManI guess I could fake it by using 2 progress bars
03:07:49MG_ManOne of them gold (the color of the box to move) and the other match the background
03:07:57MG_ManSo as the gold one goes, the background one also covers it up
03:18:37UnhelpfulMG_Man: in general, having any elements of your WPS overlap is not a good idea
03:19:12MG_ManI see where it can go wrong
03:19:22MG_ManI still want to try it out first, it might loo kreally cool of it works
03:20:17MG_ManI've had to do strange things to my theme to get it to wok right between album art and no album art, but not by overlapping
03:20:49MG_ManReason was the viewports kept eating through the text blocks I had and showing the purple background, so I thought I'd just put them on the background
03:21:01UnhelpfulMG_Man: i don't think there's any way to do what you're talking about. progress bar images are not slide over, they're always left-aligned and then truncated to indicate progress
03:21:27MG_ManYes, I know
03:21:37MG_ManI'm thinking of something though
03:21:48Unhelpfuli think that adding an offset-style progress bar would be a reasonable thing, though
03:23:32MG_ManWhere instead of chopping off the progress bar graphic according to how much is played, it drags it to the right
03:23:43MG_ManThen if you wanted rounded corners, you'd just place a tansparent image over it
03:23:49Unhelpfulthe obvious way would be just to make it 2x as wide as you want, and have the WPS draw <width> pixels starting at <offset>
03:23:59Unhelpfulyou can't "just place a transparent image over it" :)
03:24:57Unhelpfulthere is no support for overlapping viewports that are visible at the same time, as far as i know. no guarantees are made about their draw order, or even that all of them will be redrawn when one is.
03:26:22MG_ManSo far what I've seen that no matter what images are on top
03:26:33MG_ManI did this by having an image and text in the same place in one viewport
03:27:06MG_Manif the image is under the viewport, well then the viewport eats through it and shows the background
03:32:52UnhelpfulMG_Man: if you want an image behind your text, or behind some other dynamic item, what you need is to put that image in the background. :)
03:33:04MG_ManYes, that's what I ended up doing
03:33:18MG_ManIt got complicated when I wanted it to look different with album art though, but I worked it out
03:33:30MG_ManIt's just quite hacky and hard to explain, you'd have to see the background itself
03:34:25MG_ManHere is the background though
03:34:32 Join moos [0] (i=mustapha@rockbox/staff/moos)
03:34:37MG_ManWhy the black box? It's the backdrop for album art in case it's not square
03:35:05MG_ManWhen there's no album art it displays this over it
03:35:27MG_ManThen text is drawn over it, but it bleeds through
03:35:29UnhelpfulMG_Man: yes, that's probably the "best" way to do it. :)
03:35:31MG_ManIt shows black anyway
03:35:44MG_ManYou can't tell the difference in the actual thing though, so it's all good
03:36:17MG_Manonly real difference is that the purple/blue part seems to dither somewhat differently than the rest of the background, but you can't notice unless you're looking for it
03:36:27Unhelpfulbut, are you saying that when there *is* album art, it displays over the boxes for the title, album, and artist?
03:36:50MG_ManIt also adds ends to those boxes over to it's left
03:37:00 Quit moos (Read error: 131 (Connection reset by peer))
03:37:04MG_ManBy instead of displaying the titlesnoalbumart, it shows this
03:37:14Unhelpfulum, yes, well, bitmaps that are not already 15/16bpp are dithered on load, and the dither matrix aligns to the top right corner of the bitmap ;)
03:37:41MG_ManAnd there are alternate visports that have the text start next to the album art
03:37:46MG_ManOtherwise it's centered
03:38:01 Join moos [0] (i=mustapha@rockbox/staff/moos)
03:38:07MG_ManThat picture covers up the longer part of the bars
03:38:13Unhelpfulthe dither matrix is 16x16, so if you line up the top and left edges at 16*N you won't have a dithering problem
03:38:23MG_ManAnd the album art goes in the black box, whic his black no matter what because it's also on the background
03:38:32Unhelpfulalternately you could dither things yourself and save a 16-bit bmp :)
03:38:50MG_ManI'm pretty sure I did that, at least thats what photoshop says
03:38:58MG_Manit may be doing it wrong though, I'll re-save it in Pain
03:39:09MG_ManA typo? Or was it?
03:39:57Unhelpfulwell, if you did ordered dither when saving it, it's *still* probably aligned to the coordinates of the individual bitmap, and not the bitmap's on-screen coordinates
03:40:17MG_Manwell the thing is, it doesnt seem to dither the no-art overlay, but it does dither the background
03:40:23MG_ManProbably has something to do with permanent image or not
03:40:31Unhelpfulif you want the dithers to line up, then you need to dither each bitmap padded to full-screen size, and then clip it down.
03:41:34Unhelpfulrockbox will dither any bitmap on load if it's not 15 or 16bpp... unless the caller asks for dithering to be disabled. i'm not sure what WPS passes for the dither flag when loading images.
03:41:59MG_ManThere's a dither parameter for loading images?
03:41:59 Join evilnick [0] (n=evilnick@
03:42:18MG_ManLook around the title/album area
03:42:25MG_ManYou'll see it looks funky compared to the rest
03:44:02MG_ManIf there is a dithering flag, I don't see it on the CustomWPS page
03:44:03Unhelpful yes, that bitmap was either saved to 16bpp without dithering, or is saved as 24bpp and loaded with the don't-dither flag set
03:44:09MG_ManWhere is it? At the end?
03:44:22MG_ManIt's 0 or 1?
03:44:28MG_ManIt doesn't appear to be documented
03:44:28Unhelpfulit's not exposed via WPS, i think
03:44:49MG_ManYeah, I'll just have to fiddle with it
03:44:58Unhelpfulthe *function* that loads bitmaps takes a flag which can be used to tell it not to dither :)
03:45:12MG_ManI'll sort out all that once the actual functions of the skin are finished
03:45:29MG_Manwhich all that is left is a progress bar, no biggie
04:00:14MG_ManI dont know man
04:00:22MG_ManIt just seems the background is the only thing that dithers
04:00:29MG_ManANd I can't get Photoshop to do it on export
04:00:54MG_ManThey were all 24 bit to begin with
04:01:35 Quit efyx (Remote closed the connection)
04:05:24 Quit cg_ (Read error: 110 (Connection timed out))
04:06:49 Quit TheSeven (Nick collision from services.)
04:07:05 Join The_Seven [0] (
04:07:09 Nick The_Seven is now known as TheSeven (
04:08:58 Quit evilnick (Read error: 145 (Connection timed out))
04:09:11MG_ManThe only way I can get it to dither is sending it into index color 9256 colors)
04:09:16MG_Man(256 colors)*
04:12:58 Join evilnick [0] (n=evilnick@
04:14:20MG_Manwell, there's Save For Web, but there is no BMP support there...
04:14:26***Saving seen data "./dancer.seen"
04:17:05MG_ManI wonder if ImageReady can do it
04:18:53 Join FlynDice [0] (
04:22:32MG_ManHmm, I found a guide online for it, I'll try it out
04:22:38 Quit toffe82 ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
04:28:33 Quit chandoo ("Leaving")
04:31:44Unhelpfulperhaps the gimp or imagemagick. :P
04:34:21 Join Hillshum [0] (
04:36:50MG_ManYeah, I'll try GIMP
04:37:04MG_ManThe photoshop method still came out not-as-great looking
04:39:36Unhelpfuldithering should probably *always* be enabled when loading WPS images... i'm assuming it's *not* already, for some reason. also, we should probably have an option to offset the dither coordinates so that the dither matrix will line up correctly when images are overlayed on the background
04:39:51MG_ManYeah, that would work
04:40:01MG_ManBut from what I gather, preloaded images aren't dithered
04:40:08MG_ManAs evidenced by the screencap from earlier
04:40:37MG_ManThough this is r22050-090726 so its a bit old
04:41:21MG_Manotherwise the skin is done
04:41:27MG_ManI'll just sort that out manually then send it in
04:42:02MG_ManIt should be OK copyright wise, since it uses nothing from winamp itself
04:42:09MG_ManJust made to look like it
04:43:49MG_ManDoes Rockbox use R5G6B5 or X1R5G5B5?
04:44:01MG_ManOr should I say, which does the device displays use?
04:44:32 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
04:44:42MG_ManI guess I'll use R5G6B5
04:45:33MG_Manyeah, gimp didn't dither either
04:49:01 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
04:49:12Unhelpfuli believe rockbox will treat either as "native" on color targets, and force dithering off
04:50:42MG_ManTime to use good old DOS program CONVERT
04:53:00Unhelpfulhrm, actually, if you save them all *without* dithering as rgb565, they'll match that way, too.
04:53:38Unhelpfulif that's a DOS or windows version of imagemagick's convert program, then it runs on quite a bit more than DOS.
04:54:25MG_ManI'll try imagemagick then
04:57:22MG_ManSo imagemagick's dither makes it look roughly the same as if dosbox did it?
04:59:19Unhelpfulimagemagick has many dithering options. :)
05:00:24 Quit MG_Man ("Quitting")
05:01:27 Join MG_Man [0] (
05:17:11 Nick Hillshum is now known as Hilton (
05:17:27 Nick Hilton is now known as foodisyummy (
05:20:19 Nick foodisyummy is now known as batmanandrobin (
05:21:10 Quit MG_Man (Read error: 110 (Connection timed out))
05:21:36 Quit LambdaCalculus37 ("This computer has gone to sleep")
05:22:09 Quit batmanandrobin ("Ex-Chat")
05:34:49 Join MG_Man [0] (i=6165d00b@gateway/web/freenode/x-dxaixsvcdlnemplo)
05:34:56MG_ManI had another idea
05:35:11MG_ManSince ImageMagick doesnt work for me (Application failed to initialize properly ???)
05:35:31MG_ManIf I like the ROckbox dithering so much, I'll just use Rockbox's dithering
05:35:50MG_ManThat is, add the overlay picture to the background image, screenshot, then copy it back to its separate image
05:36:48MG_ManIt'll also ensure the matrix is aligned too
05:40:40 Join funman [0] (n=fun@rockbox/developer/funman)
05:41:59MG_ManShould I save a 16 bit version of the normal background too or let RockBox dither it?
05:42:28MG_ManDoes it make any difference performance-wise?
05:47:35Unhelpfulnot really, it's only loaded once. :)
05:48:04MG_ManI suppose that's cleared up then
05:48:29MG_ManRight, well
05:48:37MG_ManI'll change a couple more things then I'll upload it
05:50:11Unhelpfula 16-bit version will be *smaller* i'd think...
05:51:41 Quit BHSPitMonkey ("Ex-Chat")
05:52:26MG_ManPhotoshop still rips it apart if i save it as 16 bit
05:52:36MG_ManBut when I did the new overlay at 24 it looked perfect in rockbox
05:56:24MG_ManSee for yourself:
05:56:32MG_ManI don't notice a difference at all now
05:57:28MG_Man337KB for the image assets
05:57:32MG_ManSOund reasonable to you?
05:57:52MG_ManCOmpaed to cabbiev2 it seems a bit big
05:57:58MG_ManI can probably go and make all the buttons 16 bit
05:58:07MG_ManOr get imagemagick working
05:58:13MG_ManI probably grabbed the 64 bit version or something
05:58:15MG_ManSilly me
06:01:20MG_ManSO do I want windows-dll/dynamic or windows-static version
06:01:25MG_ManI don't understand what those really mean
06:02:12 Quit MG_Man ("Page closed")
06:02:27 Join MG_Man [0] (i=6165d00b@gateway/web/freenode/x-isvafwjdxmtuzlli)
06:14:30***Saving seen data "./dancer.seen"
06:21:11MG_ManOkay, seriously, why
06:21:19MG_ManI can NOT get ImageMagick working
06:21:50MG_ManThis application has failed to start because the application configuration is incorrect.
06:26:59 Join than1 [0] (
06:28:59MG_ManUnhelpful: Which ImageMagick interface (if any) do you use?
06:37:04MG_ManI'll try the zip version...
06:38:02MG_ManGeeze, it finally decided to go
06:38:09MG_ManRight, let's see what it can do
06:45:56MG_ManOkay, I can not figure convert out
06:46:18MG_Man-depth 16 does absoluteely nothing
06:46:25MG_ManI wonder if that means everything already IS 16 bit
06:49:44 Quit Horscht (Read error: 110 (Connection timed out))
06:51:09 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:56:54 Quit MG_Man ("Page closed")
06:59:58 Quit than1 ("Ex-Chat")
07:02:31 Join antil33t [0] (n=Mudkips@
07:08:55 Quit funman ("Lost terminal")
07:21:05 Join MG_Man [0] (i=6165d00b@gateway/web/freenode/x-zzpafjgmxmhvxopg)
07:21:23MG_ManANyone still alive in here
07:22:04 Quit MG_Man (Client Quit)
07:29:04 Quit martian67 (Read error: 104 (Connection reset by peer))
07:41:09 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
07:46:35 Quit z35 (Read error: 104 (Connection reset by peer))
07:47:09 Join MG_Man [0] (i=6165d00b@gateway/web/freenode/x-ycrxdosiaygiljnw)
07:47:14MG_ManHoly crap no way
07:47:50MG_ManI pretty much gave up on 16 bit images and tried 256 colors (or minimum for everything but the background) and it looks EXACTLY THE SAME in Rockbox as it did with everything in 24 bit
07:48:00MG_ManHowever, now all the art assets take up simply 82.1 KB
07:48:06MG_ManThat's less than CabbieV2.
07:48:17MG_ManI think I know what to stick with now
07:50:12MG_ManI guess I'll go to bed now and submit it tomorrow
07:50:24MG_ManNight, whoever is still alive here
07:50:29 Quit MG_Man (Client Quit)
07:58:54 Join jds2001 [0] (n=jds2001@fedora/jds2001)
07:59:42 Join z35 [0] (
08:14:33***Saving seen data "./dancer.seen"
08:43:03 Join evilnick1 [0] (n=evilnick@
08:48:30 Join Rob2223 [0] (
08:58:33 Join robin0800 [0] (
08:59:53 Join stoffel [0] (
09:01:05 Quit evilnick (Read error: 110 (Connection timed out))
09:06:45 Quit Rob2222 (Read error: 110 (Connection timed out))
09:09:08 Join cg_ [0] (
09:20:43 Join jchillerup [0] (
09:21:12jchillerupGood deal for people who would like a Rockbox-compatible ipod:
09:26:30 Quit stoffel (Remote closed the connection)
09:36:19 Join n1s [0] (n=n1s@rockbox/developer/n1s)
09:41:44 Quit robin0800 (Remote closed the connection)
09:42:13 Join robin0800 [0] (
09:44:05pixelmaI thought CheckWPS reuses the display code that the core would use - how could it be more forgiving than the parser on target?
09:48:56pixelmathe c200 "SansaPod" theme is definitely broken in SVN and I could find the errors - in the beginning there is "e%X|..." and later there is an %xdw where no image was preloaded with this identifier
09:50:07pixelmaso I'm surprised that a CheckWPS run the themes server wouldn't find that
09:55:44 Join flydutch [0] (
10:03:02pixelmathat was 1 theme of 1 I tried, it's probably unlikely that it was the only one with errors
10:06:30Unhelpfulhas anything changed in WPS lately? and does the theme server re-validate?
10:08:11pixelmayes, the parser got stricter with loading and preloading images, CheckWPS was broken and repaired yesterday and mcuelenaere said he ran a check again and all themes on the server had passed
10:14:37***Saving seen data "./dancer.seen"
10:19:35pixelmahehe, nice example "song" in the screenshot of the Ondas' themes subpage
10:20:24 Quit evilnick1 ("Leaving.")
10:49:40Unhelpfulpixelma: a practical use for AA fonts: sharp outlines make that scribble font look much less scribbly ;)
10:54:11pixelmanot practical for me though as I wouldn't use a scribbly font (and wouldn't use that theme, especially not this port (all portrait screen ports of this theme I've seen so far have that stretched background which looks disturbing and unnecessarily blurry))
10:54:54 Join BdN3504 [0] (n=5ce53941@gateway/web/cgi-irc/
10:54:57pixelmanot mentioning the fact that there is no port for one of my targets
10:57:13BdN3504pixelma: can you have a look and commit this?
11:00:15 Join ender` [0] (
11:00:45Unhelpfulpixelma: bah! i am again defeated by your cunning use of logic!
11:04:55pixelmaBdN3504: \nopt{player,lcd_color,lcd_mono}{\input{advanced_topics/viewports/grayscale-vp-syntax.tex}}
11:05:43 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
11:05:55Unhelpfulkugel: i think the glyph cache needs to be larger when using AA fonts - i have album titles that cause the disk to spin when they scroll :/
11:07:03BdN3504pixelma: what about it? that's for grayscale models
11:07:09pixelmacan't work for at least two reasons: you can't combine things in a \nopt (you can in an \opt), lcd_mono doesn't exist (and if it would, it would also be true for the Player, you wouldn't need the first one)
11:08:11BdN3504ah ok, why can't i use multiple things in nopt?
11:08:21 Join DarkDefender [0] (
11:08:55pixelmayou would have to nest this like e.g. \nopt{lcd_color}{\opt{lcd_non-mono}{your inclusion}}
11:09:05 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
11:10:57pixelmathe lcd_color etc. comes from the languages' features.txt (in apps), there you can look up what features exist to use for automatic in- or exclusion)
11:13:29pixelmabluebrother explained the \nopt two days ago here
11:14:42pixelmaBdN3504: I also think that you don't need a full WPS like example to explain the viewport syntax
11:17:57 Join Topy44 [0] (
11:18:10Topy44strange problem with my h340:
11:18:21Topy44if i turn it on i get an "ATA error: -11"
11:18:35Topy44but if i connect it to usb it works fine, and if i disconnect it it boots normally
11:18:46Topy44turn it off, turn it back on, ata error -11 again
11:19:13Topy44i removed the hd for a surface scan and just put it back in again, it does that since
11:22:18BdN3504 right, that has excaped me. i took that example to show what i wanted to show there... how would you explain that you don't have to use individual identofoers for VPs?
11:25:20 Quit Horscht (Read error: 110 (Connection timed out))
11:27:20BdN3504would this give me grayscale models only \opt{lcd_non-mono}{\nopt{lcd_color}{\input{}}}? and has it correct syntax now?
11:27:50Topy44btw, some people suggest -11 would be "hd not spinning up" and means that the hd is broken, though it quite obviously does spin up and works just fine when connected to the pc
11:28:01 Join decayedcell [0] (
11:31:04pixelmaBdN3504: yes, should work. By the way - your clock examples also have the problem that there are targets without RTC (and no, not only the Player=
11:32:49pixelmaTopy44: did you check that all the connections/connectors are fit a 100% again?
11:32:50BdN3504ok, so then i'll just copy the example off the wiki. btw, i could also use lcd_charcell for the player, right?
11:33:05Topy44pixelma: yeah, only one i disconnected was the hd anyway
11:33:13Topy44and as i said - if i connect it by usb it goes on just fine
11:33:24Topy44also after disconnecting the usb cable it boots up and works normally
11:34:22Topy44its completely reproduceable
11:35:20pixelmaBdN3504: yes, and it would even be more appropriate as this is the feature it really depends on. This way it would be correct if something else comes along with a charcell display (unlikely for a DAP itself but maybe a remote, or at some point someone showed interest in those USB displays as an additional display, some of which are charcell)
11:39:09 Quit amiconn (Nick collision from services.)
11:39:13 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
11:39:22 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
11:39:56 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
11:39:56 Quit pixelma (Nick collision from services.)
11:40:15 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
11:44:57 Quit ender` (" Washing your car to make it rain doesn't work.")
11:48:45BdN3504is my assumption, that all colour models have rtc right? at least it says so in the devicechart, but i might be wrong
11:51:06 Join ender` [0] (
11:56:58 Join webguest51 [0] (n=47c339b2@gateway/web/cgi-irc/
12:00:57 Quit webguest51 (Client Quit)
12:11:17 Join pjm0616 [0] (n=user@
12:12:02 Join webguest13 [0] (n=3dfa7162@gateway/web/cgi-irc/
12:13:58 Quit webguest13 (Client Quit)
12:14:39***Saving seen data "./dancer.seen"
12:30:38 Join Turniphead [0] (
12:32:19 Part Turniphead
12:32:22 Join Turniphead [0] (
12:36:23TurnipheadI have a minor problem with my second-hand Cowon M3. I installed Rockbox 22446-090821. I hear a clicking sound about half the volume of a normal keyclick while I'm playing audio. The same click occurs when I muce the cursor. All this while my Key Click setting of set to Off. Can anybody shed some light on this?
12:43:09 Quit BdN3504 ("CGI:IRC (EOF)")
12:43:49n1sBdN3504: it might be tru *now* but nothing says it will stay true with future targets, it's better to make stuff depend on the things they really need
12:44:36pixelmaTurniphead: does this click while playing audio happen at regular intervals (without you touching the controls), and/or does it happen more often when there is a lot going on on the screen, like scrolling track title etc. (provided you have the remote connected)
12:45:15 Quit robin0800 (Remote closed the connection)
12:45:41TurnipheadThe remote is connected. And I'm not moving the controls. The sound occurs at regular intervals of about half a second or less.
12:46:24TurnipheadIt does seem to occur more often if there is scrolling text involved
12:47:13pixelmaI assume it doesn't happen without the remote?
12:47:34TurnipheadNope, no clicks at all
12:48:15n1sdoes the iaudios have that remote tick reduction setting?
12:48:43pixelmasounds a bit like the "remote ticking" problem that's there on other targts with remotes (it also depends on your actual device)
12:49:07TurnipheadWhere would I find that specific setting?
12:49:39pixelman1s: I don't think so but I believe seeing reports that some Iaudios have that problem - I could check if the setting is there on my M5 but don't have a remote to test
12:50:32TurnipheadIn my quest around the menu's of Rockbox I did not find a setting like that. As it is likely I would have tested that option :)
12:51:48 Join robin0800 [0] (
12:52:46pixelmahmm... it's also a bit different as the remote LCD is treated as main screen on the M3 and as a remote screen on the M5
12:53:28 Nick YpsyZNC is now known as Ypsy (
12:54:40TurnipheadIndeed. That would probably be the option is absent. Though it still suffers the problems of the remote.
12:56:39 Quit n1s ("Lämnar")
12:57:29pixelmait's not present here as well (in the Remote-LCD settings) too, but if it was prepared for the Iaudio remotes it would need special treatment for the M3
12:59:39TurnipheadSo the chances that a simple patch would do are low. I would love to have a look though.
12:59:50pixelmaTurniphead: maybe you can add a comment along the lines of "happens on M3 too, would need special care in the menu"?
12:59:58 Join BdN3504 [0] (n=5ce53941@gateway/web/cgi-irc/
13:00:38BdN3504if i want to add a file to svn and set the keywords properly, what determines the id?
13:00:47TurnipheadThanks, will do!
13:01:05pixelmaTurniphead: the last comments say it's out-of-sync though and the comment itself is quote old already too
13:01:45pixelmawhich is not a big surprise as the patch is 2 years old
13:03:08BdN3504in other words, how to use that command? can you give an example line?
13:04:48TurnipheadHrmm, added the comment, but I see your point.
13:05:38*pixelma wonders where jhMikeS went
13:05:48 Join stripwax [0] (
13:12:51pixelmaTurniphead: not sure if it helps but if you say that it's worse with more scrolling - maybe you could use (or create) a more static WPS as a band aid fix to lower the ticking?
13:14:25pixelmaor stay in the menu
13:17:35TurnipheadI noticed that last remark being helpfull (the menu thing) I'll try a WPS with as few 'live' or scrolling information fields as possible
13:21:12pixelmayou could disable scrolling lines (get rid of %s), don't show playback time etc., things that need regular updates. Have a look at the CustomWPS page in the wiki for possibilities. It's not a nice solution but if it makes listening more enjoyable, maybe worth it for now
13:21:44 Quit stripwax (Read error: 104 (Connection reset by peer))
13:24:20 Quit BdN3504 ("CGI:IRC (EOF)")
13:24:28 Join stripwax [0] (
13:24:29 Join bdn3505 [0] (n=5ce53941@gateway/web/cgi-irc/
13:25:28bdn3505ok, i updated the patch with what has been suggeested, thanks for your support. please commit asap.
13:33:02 Quit bdn3505 ("CGI:IRC")
13:34:53 Join Thundercloud [0] (i=thunderc@
13:50:31 Join AndyI [0] (i=AndyI@
13:54:57 Join kugel [0] (n=kugel@rockbox/developer/kugel)
13:59:50Turnipheadpixelma: the clean WPS thing doesnt comfortably help the ticking sound. I'll stick with hanging around in the menu's for now.
14:00:37Turnipheadthis way, I'll have all the info ready when I need it, and can listen without so much as a blip.
14:02:49 Quit AndyIL (Read error: 110 (Connection timed out))
14:04:47soapI was going to reply to a recent post to the user mailing list with a polite "Do not talk about removing DRM on this mailing list, please." But do not see an explicit rule against this. Is there an implicit rule against this?
14:05:43 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
14:06:34TurnipheadIts implicitly written in the IrcGuidelines. But I don't know it would apply to the mailing list. It is however common sense that is would.
14:07:30pixelmaimplicit would also be "it doesn't have to do with Rockbox", not sure if that's enough... ;)
14:09:35pixelmaand if there is a note somewhere you can point to about respecting copyright, that would fit too I think
14:09:44 Join teru [0] (
14:12:39kugeldoes the Player have wps-viewports?
14:12:57AlexPsoap: I'm sure that is covered by the general we don't want to get involved in that
14:13:09pixelmakugel: I don't think so
14:13:10AlexPBest all round to keep it off the lists
14:13:57kugelBdN3504 made a viewport section for it
14:14:44***Saving seen data "./dancer.seen"
14:17:56kugelhm no, he rather mixed up mono and charcell
14:21:56pixelmathe whole patch is still too complicated and puts too many things in there - e.g. the examples wouldn't have to be that many if they wouldn't deal with something that relies on RTC presence
14:22:25pixelmathe exampleas are also too "complete" IMO
14:23:35pixelmathe examples too
14:23:49kugelI don't think complete examples are bad
14:24:18 Join PaulJam [0] (
14:24:27pixelmapeople won't read too long text
14:25:24kugelbut people like examples better than plain text
14:29:53 Quit faemir ("Leaving")
14:30:13pixelmabut probably not 100 long examples for every second WPS feature, and this case the examples contain unrelated things that are distracting again - explain conditional viewports with album art/no album art cases but why take the RTC things into it?
14:30:27pixelmasorry, bad paste
14:31:20pixelmamaybe it would be a possibility to have a short section on the features and the syntax and then an example WPS, well commented
14:32:40pixelmaI'd suggest cabbiev2 (or whatever default WPS there is), though I think some of the cabbiev2 ports would need some cleanup first for this
14:33:30terudionoea: FS #10555 - Rockpaint: improve browsing font.
14:39:07 Join SUSaiyan [0] (
14:39:17 Join bs66_ [0] (
14:41:23 Join mcuelenaere [0] (
14:41:47mcuelenaerepixelma: if you look on the theme server now, several c200 themes don't have a 'Works in the current build' line
14:42:04mcuelenaereperhaps that should be a red 'Doesn't work in the current build' text though
14:43:29pixelmaah, true. That would be better... plus a way for themers to update their own themes
14:44:08kugelanyone should be able to imo
14:46:24kugelmcuelenaere: did you intentionally ignore my comments about the name of "vfnprintf" (w.r.t. the new logf)
14:47:03mcuelenaerepixelma: done
14:47:11mcuelenaerekugel: hmm no, I must've missed them
14:49:00mcuelenaerekugel: what name do you suggest?
14:50:41kugelmaybe vuserprintf or vcbprintf? the "fn" stands for function since you pass one to it
14:50:46rashermcuelenaere: If you check the output of private/runcheckwps.php, it seems there's something wrong
14:50:53pixelmamcuelenaere: would it be possible that the RBUtility would check that line and doesn't "offer" the broken themes? Hmm... I can see a potential problem though - the ones just currently broken will still work with the release
14:51:47mcuelenaerepixelma: I think that's a question for bluebrother or domonoky, and when you're at it could you ask them to include the menu image too? ;)
14:51:54mcuelenaererasher: I'll take a look
14:52:45rasherpixelma: The theme site already tells rbutility about this
14:52:46mcuelenaerekugel: vuserprintf sounds better than vcbprintf
14:52:52 Nick Horschti is now known as Horscht (n=Horscht2@xbmc/user/horscht)
14:53:19rasherOr should..
14:56:00pixelmarasher: the last RBUtil release still offers me currently broken themes, just tried (of course this could be a problem of the Utility not the themes site and might be fixed in SVN)
14:57:10domonokyhm, the theme site tells rbutil, if a theme passed a specific version/release test. but rbutil doesnt use this at moment.
14:57:16rasherpixelma: yes, there appears to be an issue
14:57:46rasher <−− Notice that the broken themes are listed as having passed
14:59:06domonokyso its both ? the passed checks are wrong and rbutil doesnt use them ? :-)
14:59:17rasherSo it would seem
15:00:47rasherah, got it
15:00:57domonokyalso i find it difficult to think of a good way to check this things in rbutil. if the installed rb version is 3.3 and the theme site tells me pass_release="3.2" what should it do ?
15:02:06rasherWait for 3.4 :)
15:02:43rasherdomonoky: What's better to do if it doesn't pass. [error]
15:04:31mcuelenaererasher: runcheckwps.php doesn't give any output? Or does it just takes a while to do its job?
15:05:05 Quit rasher (Read error: 131 (Connection reset by peer))
15:05:48domonokyrasher: telling since whith version it doesnt pass would be helpfull. so if fail_release = 3.3 we now that for versions before its ok, and afterwards its broken so we hide it. (same for revisions)
15:05:48 Join n1s [0] (n=n1s@rockbox/developer/n1s)
15:07:28pixelmacurrently the parser "only" got stricter, so if the WPSs were written correctly they will work with current build and the releases. This is not true with changes that break old WPS because of a new syntax, these changes don't happen very seldomly and are avoided. I can only think of the progressbar syntax as an example and the scrolling margin which didn't exist for long in SVN though
15:08:19mcuelenaererasher: nvm, it's done now
15:08:28pixelmaso unless something of the latter happens, the Utility could stay on the safe side and just offer themes that work for the current build
15:10:14domonokypixelma: i woud think the pass_current="..." will be often behind the current revision rbutil can install. if we stay on the safe side, then no themes would be installable.
15:10:34mcuelenaererasher: you're right, there's definitely something odd with that script; it seems to run checkwps several times for the same WPS.. (and it doesn't run it for RWPS'es)
15:11:30 Join rasher [0] (
15:13:49rasherdomonoky: that would requires us to save all checkwps results we ever had :\
15:14:06domonokypixelma: if you look at you will see that pass_release is still at 3.2, so if we use this value we wouldnt have any theme for 3.3 and if we ignore it we show all, even when many themes are broken because of stricter parser.
15:14:15 Quit stripwax (Read error: 104 (Connection reset by peer))
15:14:31rashermcuelenaere: Yeah, I disabled checking rwps because it was causing the themes to not show up at all. Although I forget the exact reason
15:14:52rasherdomonoky: is pass_current="" better than no pass_current line?
15:15:17domonokyrasher: no, you just need to store the first failed revison and not update it on new checks, unless it passes again. (the clear the result to empty)
15:16:17domonokyrasher: for rbutil pass_current="" is the same as no pass_current line (Qt ini parser gives a empty string for non-existant values)
15:17:13pixelmaif you look for pass_current, you will have themes working for 3.3 and 3.2 as well
15:17:19*domonoky wants a "fail_current" and a "fail_release" line.
15:18:16domonokyso we can hide them if the current installed version is <= fail_release :-)
15:18:27domonokyäh ">="
15:19:42pixelmathis is true as long as some syntax doesn't change dramatically
15:19:49domonokyhm, i think this is still not enough.. it may fail again in even older releases/revisions.. so we need to know the working "window".
15:23:04domonokyhm, looks like this is very difficult to solve properly.
15:28:07rasherOr maybe just not hand-hold the user *this* much
15:29:26rasherWe can tell in some cases that it *won't* work. I think we should leave it at that
15:29:33domonokyyes, maybe the best way would be to just use passed_xxx line to display a "last successfully test on version x" line
15:30:24domonokyrasher: with the current info, how can rbutil know for sure it wont work ?
15:30:51rasherIf the user has a current release, and pass_current is empty
15:31:14rasherOr (in the future) if the user has 3.4 and pass_release is empty
15:32:37 Join Thundercloud_ [0] (
15:32:42 Quit Thundercloud (Read error: 131 (Connection reset by peer))
15:32:51kkurbjundomonoky: could you provide a pointer on what I should be looking at to add support for a new target in rbutil?
15:32:56domonokyso if pass_current/pass_release is empty we just hide it. (even if it might work if the user has a older version installed)
15:33:12rasherdomonoky: the theme site doesn't show the theme in that case either
15:34:14domonokykkurbjun: add the new target to rbutilqt.ini (just look at the other targets there for help). If it needs a new way to install the bootloader, you need to provide a new base/bootloaderXXX class.
15:34:45kkurbjuncool, and then I just have to add something for autodetection?
15:36:04domonokykkurbjun: autodetection based on USB-IDs is automatic (usb-ids are also in rbutilqt.ini) otherways to autodetect are harder to implement, (autodetection.cpp/h needs a rework to easily add new detection methods)
15:36:28kkurbjunoh, ok, I was looking at the special case stuff based on folder names
15:37:06domonokykkurbjun: you can add this in base/autodetection.cpp if you want.
15:37:42kkurbjunwhat is the reason for the special case detection?
15:37:53domonokybut autodetection.cpp really needs a rework to make it easier to extend, and also more reliable.
15:38:22kkurbjunif it does it by usb id on the m:robe then I think that would be best
15:38:40domonokykkurbjun: the problem with autodetection is: usb-ids arent always unique, and there is no crossplatform way to map USB-IDs -> mountpoints (at least at moment)
15:40:02 Quit Rondom (Nick collision from services.)
15:40:12 Join Rondom [0] (n=Rondom@
15:42:16kkurbjundomonoky: what is the target id string per target.. it doens't seem to match up to the configure script or the platform number
15:44:11domonokykkurbjun: target_id should be the target_id from configure. But i am unsure if it is still used.
15:45:38domonokyits still used for voice generation. so if it doesnt match configure. its a bug!
15:45:53kkurbjunok, I'll change the mrobe 500 and 100
15:46:12 Join webguest43 [0] (n=586e5ed6@gateway/web/cgi-irc/
15:47:19 Quit n1s ("Lämnar")
15:47:56domonokymaybe rbutil should use the target_id from rockbox-info.txt, the we wouldnt need it in the ini-file anymore.
15:49:23mcuelenaererasher: the problem is probably because $output isn't cleared
15:49:49 Quit webguest43 (Client Quit)
15:50:03kkurbjundomonoky: that was really easy to add support for a new target :-D
15:50:23kkurbjunand it is working fine too withthe themes and all
15:50:43domonokythats the way it should be :-)
15:52:01 Quit PaulJam (".")
15:52:53 Part decayedcell
15:54:24 Quit robin0800 (Remote closed the connection)
15:57:34 Join robin0800 [0] (
15:58:06 Quit Rondom (Nick collision from services.)
15:58:16 Join Rondom [0] (
15:59:21rashermcuelenaere: That sure sounds likely
15:59:51rasherBut $ret should still be valid
16:00:15kkurbjundomonoky: so the bootloader on the m:robe 500 needs to do a couple of things... right now it is enclosed in a zip file that has a complete image of the original firmware with a patch applied to it. You have to unzip that and then copy a file (rockbox.mrboot) to the root of the player. It then also needs a zip extraction of an exploited svg into another folder of the player. I could package it all as one..
16:00:20kkurbjunOn the H300 it looks like we require the user to download a copy of the firmware themselves... if we follow that method it will limit the users to only be able to install the bootloader in windows.. If we provide a pre-made zip they should be able to use it on any platform, but I don't know what the legal reasons are for making them download it themselves.
16:00:54mcuelenaererasher: hmm I hadn't actually looked at $ret nor $result['pass']..
16:01:17domonokythe legal reason to download himself, is that then we dont distribute software we dont own.
16:01:24Topy44why not release the patch as a binary patch file
16:01:58kkurbjunthe problem is that the download from the olympus site is in an executable - I'm not sure how the files are stored in it
16:02:01Topy44that can be applied with an (included) windows tool, but also with diff or the like on linux and mac
16:02:09domonokyso if we host this ZIP somewhere else where nobody cares and tell the user to download it from there, we should be fine :-)
16:02:27kkurbjunand they would need to run the exe and have the know how of which files to pull and when from it
16:02:28domonokyor we write a tool to extract the of from olympus exe file :-)
16:03:16Topy44well, try the known extraction methods... for example, check if universal extractor can find the file
16:03:25mcuelenaerekkurbjun: does WinRAR/WinZIP/... show an 'Extract to here' quick menu when right clicking on it in Explorer? If so, it should be a known algorithm
16:03:36 Quit robin0800 (Remote closed the connection)
16:03:40domonokythe technical side of this shouldnt be a problem. rbutil can download/unzip etc just fine. we just need to clear the legal side.
16:03:41kkurbjunmcuelenaere: that's what I'm looking for now..
16:03:43 Join fyrestorm [0] (
16:03:56mcuelenaerekkurbjun: Topy44's suggestion of Universal Extractor is even better
16:04:02 Join robin0800 [0] (
16:04:15 Quit teru ("Quit")
16:04:55*domonoky thinks sombody should just try to open it with 7zip, if we are lucky it works :-) (7zip can open nearly everything)
16:05:26CIA-61New commit by Domonoky (r22463): rbutil: use target-id from rockbox-info.txt and remove it from rbutil.ini
16:05:48rashermcuelenaere: okay, I've really no idea what's going on with the checkwps run :\
16:06:52Topy44btw, i still have a problem with my iriver h340
16:07:13Topy44since i removed the hd for a scan and replaced it, i get an ata error -11 on start
16:07:28Topy44but if i connect it to a usb cable it works fine, and if i disconnect the cable it boots normally
16:07:32 Quit patmulchrone (Remote closed the connection)
16:07:35Topy44turn it off, back on again, same error.
16:07:59mcuelenaererasher: I'm also seeing several ZIP extraction failures on my local copy, but that could be my fault
16:08:06Topy44apparently -11 means "no hd spin up", but thats obviously wrong (i can hear it clearly spinning up and doing its self check normally)
16:08:15kkurbjun:), ok, looks like the exe can be extracted :)
16:09:59rashermcuelenaere: okay, taking my widecabbie as an example, look at this:
16:10:03rasherI can't make sense of that
16:10:32 Join kushalone [0] (n=kushal@
16:11:27mcuelenaerehmm that's pretty messed up
16:12:16rasherThere may be other issues, but that's not right. Either it's an error or it's not. I also don't understand why it'd fail to load that image, but who knows
16:13:38pixelmaTopy44: maybe it's a timing problem (for some reason it doesn't spin up as expected), this would explain why it works when just disconnected from USB and if the needed timing for USB is different that could explain why it worls there. I'm just guessing though
16:14:04Topy44yeah... i would like to know from some coder what EXACTLY ata error -11 means
16:14:15Topy44so how does it "detect" that error
16:14:45***Saving seen data "./dancer.seen"
16:14:59Topy44i've been wondering if meaby a reformat of the partition or the like could have any influence - because of course having the hd removed and checked and treated by a stack of different utilities can mess up things a bit
16:15:08kkurbjundomonoky: do we expect that the user will have 7z installed on their system when using rbutil?
16:15:27Topy44it could have messed with the filesystem in a way that linux/dos/windows dont care about but rockbox doesnt like
16:15:32 Quit CaptainKwel (Remote closed the connection)
16:15:37 Join stripwax [0] (
16:15:53domonokykkurbjun: no. rbutil has built-in zlib for zip. but thats probably not enogh for that. We should exchange it with a 7zip library.
16:16:10 Join andrewrb [0] (n=586e5ed6@gateway/web/cgi-irc/
16:16:22domonokykkurbjun: we should try to make rbutil not relay on anything installed on the users-system.
16:18:05 Join andrewRB_ [0] (
16:18:29 Nick andrewRB_ is now known as andrewbeveridge (
16:21:10 Quit andrewrb ("CGI:IRC (Ping timeout)")
16:22:39 Nick andrewbeveridge is now known as andrewrb (
16:22:47mcuelenaererasher: it seems like the skin engine doesn't think failed images should return a failure on skin_data_load() - not much I can do about
16:23:26 Join CaptainKewl [0] (
16:23:57rasherI wonder why that image fails in the first place
16:24:12rasherDid JdGordon's recent rework make the image buffer smaller?
16:24:47scorche|shrasher: he says "technically yes"
16:24:51andrewrbhi, when was jpeg support enabled for album art?
16:25:00rasherscorche|sh: :(
16:25:08rasherHe broke my wps!
16:25:27Torneandrewrb: 2009-05-01: Rockbox can now read JPEGs for Album Art. (JPEG decoder added to core)
16:25:31Torneandrewrb: included in 3.3
16:25:38scorche|shhe says to increase the skin buffer define in skin_buffer.c
16:26:36mcuelenaererasher: read_bmp_file() returns -6 on that image
16:26:49andrewrbthanks. which revision would that be? It's just, the resize method seems to be horrible (at least on my D2), I was hoping I could have a fiddle
16:27:14Tornethere's been a lot of changes to the jpeg code since then
16:27:27Torneif you're not using a current build try that first
16:28:31mcuelenaererasher, scorche: increasing the skin buffer define helps
16:28:58andrewrbbuilt last night :/
16:29:11*rasher finds it mildly bad manners to decrease buffer sizes
16:30:15scorche|shhe didnt decrease the size...he just changed the usage (apparently)
16:30:44andrewrbTorne: to clarify, even a 100px cover.jpg file which should not need resizing at all looks horrible in the WPS
16:30:50rasherWhich means the available size was smaller. Which means it was in fact, decreased.
16:30:58Torneandrewrb: works for me..
16:31:08andrewrbTorne: target?
16:31:17rasher"src_dim.width > BM_MAX_WIDTH" tests true
16:31:47Tornei mean i've not looked particularly closely but all my album art is jpeg at arbitrary random sizes and it's always looked fine ;)
16:31:50scorche|shrasher: he is especially begrieved that he cant actually chat on here about this today =)
16:32:03kugelrasher: the buffer wasn't decreased, but it's shared with other wps stuff now
16:32:11andrewrbit may be a d2-specific thing.
16:32:22andrewrbi'll have to have a chat to shotofadds
16:32:28 Quit stripwax (Read error: 104 (Connection reset by peer))
16:32:58scorche|shkugel: not stuff
16:33:20kugelwhich is only the wps as of now
16:34:00kugelI'm surprised that a theme actually hits the limit, all the themes I use don't even use half of the buffer
16:34:21kugelscorche|sh: but yea, my custom statusbar work also uses the same buffer
16:34:34Torneandrewrb: post on the forums with an example of the album art you have and the size your wps is rendering it at?
16:34:41scorche|shskin engine isnt wps...up the buffer size by 1000 x skin_token
16:35:35pixelmarasher: did you check what buffer your WPS actually uses (in the Rockbox Info screen)
16:35:39kugelthe wps uses the skin engine, but nothing else right now. the skin engine was derived from the wps
16:36:38andrewrbTorne: yeah, well myself and around five other D2 users have confirmed there is a problem over on the iAudiophile forums, with various sample images. I may post on the rockbox forums but as shotofadds is the only person really working on the D2 port it seems pointless (i have personal contact details for him)
16:37:22Torneyah, but it's not going to be specific to the D2 entirely, probably; it'll be whatever uses the same processor (probably some optimised code is dodgy on that proc?)
16:37:44andrewrbhmm, ok. i shall go post.
16:41:17mcuelenaeredomonoky: you seem to've forgotten to remove the targetid's from the Onda's in rbutil.ini
16:42:21domonokymcuelenaere: true, will remove them now.
16:42:29scorche|sh1000 x skin_token is a bad fix, actually MAX_TOKENS
16:43:04CIA-61New commit by Domonoky (r22464): rbutil: also remove the target-id from the onda targets.
16:46:51 Join stoffel [0] (
16:48:43 Quit mcuelenaere ()
16:49:57 Quit andrewrb ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
16:53:13 Quit bs66_ (Read error: 113 (No route to host))
16:55:55 Quit DarkDefender (Remote closed the connection)
17:02:24 Join Fanuc98 [0] (
17:03:06Fanuc98What is Rockbox?
17:05:16domonokyFanuc98: see :-)
17:05:53 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
17:07:06Fanuc98I looked. I do not have a mp3 player, but I know what it is and how to convert mp3 / 4 from and to .wav
17:10:53 Quit robin0800 (Remote closed the connection)
17:11:20 Join robin0800 [0] (n=robin080@
17:12:06rasherkugel: my theme was using the _entire_ bitmap buffer for the giant volume display
17:15:07rasherIt's currently using 18.4KB / 85.0KB
17:15:15rasherWhich leaves not enough RAM for my volume bar.
17:15:46kugelhow big is the bar?
17:16:05rasherAnyway, the effect is that there is no less room for a WPS to use. Regardless of how you try to weasel out of it
17:16:33rasherThe file is 103KB
17:18:13kugellooking at it, it seems that you could make it smaller
17:18:59 Join andrewbeveridge [0] (
17:19:06 Nick andrewbeveridge is now known as andrewrb (
17:19:53scorche|shrasher: he agrees woith you that it is an issue, but you are the only person who has really come up against it, and with such a huge buffer, you will probably be better off making your own build
17:20:09rasherscorche|sh: Me and whoever else is using my skin
17:20:29rasherAnyway, even if it would be possible to optimise it, that doesn't change the fact that this change has lowered the available room for WPSes
17:20:49*gevaerts votes for MEMSIZE/2 for the wps buffer
17:21:00andrewrbany D2 owners here?
17:21:31kugelwell, you could make small strips and display X strips depending on the volume
17:21:53jchillerupGood deal for people who would like a Rockbox-compatible ipod:
17:22:24rasherkugel: that sounds gross. Also, back to my previous point
17:23:01rasherThis change has lowered the limits of what's possible.
17:23:19kugelyes I noticed that
17:23:58 Quit Horscht (Read error: 110 (Connection timed out))
17:24:11kugelbut I'd rather optimize this single theme instead of pushing the buffer for everyone. the big volume bar is done in a pretty wasteful way anyway
17:24:36rasherCan't you only do strips vertically?
17:25:19kugelyes I think so
17:25:27rasherI don't see how you'd do it then
17:25:37Fanuc98can Rockbox run in DSL? (damn small linux?)
17:25:42kugelwell, I didn't mean to have those strips in a single file
17:26:14rasherThen I'd just end up with n files, totalling the same size as this single one
17:27:52rasherThen you'll have to explain yourself better
17:27:58rasherUnless I can do partial bitmap drawing in the horizontal direction, I just don't see how this'd work
17:29:19rasherAlso, this doesn't really help whoever else will be hitting this new, lower limit
17:29:53kugelwell, you need strips of the full volume and "no" volume. Split those into 6 or so strips. now if you're going from low volume to high, just replace the next strip (which is a strip of the no volume pic) with the corresponding strip of the full volume
17:30:05andrewrbwhy would album art appear differently in the simulator to on the actual target?
17:30:07kugelI don't think anyone else will hit the limit soon
17:30:29rasherkugel: "strips of the full volume and "no" volume." = twice the size of the current strip!
17:30:35 Join paulk_ [0] (
17:30:40rasherkugel: you don't think..
17:30:52kugelwell, it seems I can't explain it properly
17:31:20kugellet's see if I can just implement it so that you see what I mean
17:32:05 Join toffe82 [0] (
17:32:45 Quit BlakeJohnson86 ("Leaving.")
17:35:39pixelmakugel: I think I understand what you mean, I guess that'll work but leads to quite unreadable WPS code (the rockboxed.wps uses that technique if I remember correctly - and I found that hard to understand)
17:36:09 Join BlakeJohnson86 [0] (
17:37:01andrewrbthe flyspray is just for supported targets, right?
17:39:08kugelandrewrb: No, we generally accept patches for unsupported targets. Bugs are different, we don't like bug reports that are specific to an unsupported target
17:39:41andrewrbwhat if i'm unsure if the bug is specific to the target or not =/
17:44:15rasherandrewrb: It sounds likely that this is some sort of driver issue. What's the difference?
17:44:52andrewrbrasher: sorry, the difference between what?
17:45:04rasherandrewrb: Between how album art appears
17:46:20andrewrbwell the bitmap art looks as it does in the image viewer. the jpeg art has, uh, vertical ripples? Not entirely sure how to describe it
17:46:43andrewrband the distortion is related to the input file size
17:50:00 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
17:50:32 Join evilnick [0] (n=evilnick@
17:51:38andrewrbthe smaller the image is
17:51:44andrewrbthe more distorted
17:52:59andrewrbrasher: can you suggest a way for me to determine if it is a target-specific issue?
17:53:38 Join evilnick1 [0] (n=evilnick@
17:54:18 Quit paulk_ ("Ex-Chat")
18:01:41 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
18:04:07 Join evilnick2 [0] (n=evilnick@
18:06:57 Join evilnick3 [0] (n=evilnick@
18:06:57***Alert Mode level 1
18:06:57DBUGEnqueued KICK evilnick
18:06:57DBUGEnqueued KICK evilnick1
18:06:57***Alert Mode level 2
18:06:57DBUGEnqueued KICK evilnick2
18:06:57DBUGEnqueued KICK evilnick3
18:06:57***Alert Mode level 3
18:10:56 Quit evilnick (Read error: 110 (Connection timed out))
18:14:48***Saving seen data "./dancer.seen"
18:16:48 Quit andrewrb ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
18:16:58***Alert Mode OFF
18:20:04 Join bertrik [0] (n=5a809af7@gateway/web/cgi-irc/
18:20:06 Quit evilnick1 (Read error: 110 (Connection timed out))
18:21:05 Quit LambdaCalculus37 (Read error: 145 (Connection timed out))
18:21:48 Quit bertrik (Client Quit)
18:22:00 Join bertrik [0] (n=5a809af7@gateway/web/cgi-irc/
18:23:38 Quit evilnick2 (Read error: 110 (Connection timed out))
18:27:11 Join Strife89 [0] (n=michael@
18:27:28 Quit evilnick3 (Read error: 110 (Connection timed out))
18:29:17 Quit JdGordon (Read error: 110 (Connection timed out))
18:31:57 Quit bertrik ("CGI:IRC (Ping timeout)")
18:32:45 Join domonoky1 [0] (
18:34:18 Join guest001 [0] (
18:35:10Fanuc98Wow, I'm drunk...
18:37:24 Part Fanuc98 ("Leaving")
18:37:49 Quit guest001 (Client Quit)
18:43:23 Quit Strife89 ("Leaving")
18:48:54 Join Miehas [0] (n=428db9eb@gateway/web/cgi-irc/
18:49:37MiehasA quick question. Is there a build for an Iriver T5?
18:50:25 Quit domonoky (Read error: 113 (No route to host))
18:50:45pixelmaMiehas: is it on the list on the front page?
18:51:14MiehasI did not see it there or in the Develpoer resources
18:52:37 Quit robin0800 (Remote closed the connection)
18:52:54 Join evilnick [0] (n=evilnick@
18:53:12 Join robin0800 [0] (
18:53:24 Quit kushalone (Client Quit)
18:53:29 Quit Miehas (Client Quit)
18:55:03 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
18:59:18 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
19:01:04 Quit robin0800 (Remote closed the connection)
19:04:22 Join DarkDefender [0] (
19:08:27 Quit LambdaCalculus37 ("This computer has gone to sleep")
19:21:52 Quit flydutch ("/* empty */")
19:25:03 Join PSPdemon [0] (
19:37:03 Join madao-hasegawa [0] (
19:37:47madao-hasegawaHi. Litle offtoppic question: is it possible to read micro SD card plugged into sandisk fuze, without explicit card reader, by just using the player?
19:40:20madao-hasegawauh oh I got it as /dev/sdb
19:40:23 Part madao-hasegawa ("bye")
19:41:53 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
19:44:32 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:47:26 Quit JdGordon (Read error: 110 (Connection timed out))
19:49:39 Quit evilnick (Read error: 110 (Connection timed out))
19:51:13 Join kushalone [0] (n=kushal@
19:53:41 Join mcuelenaere [0] (
19:54:33 Join BlakeJohnson86 [0] (
20:03:36 Join gammy [0] (
20:05:54gammyhey guys. I recently blew the output segment of my iRiver iHP-120 and I was wondering if anyone has one I could trade? I don't need the hdd, and I'd even be interested in it if it was missing hdd and battery. I have a 2nd gen ipod nano 8GiB as incentive.
20:07:28 Quit BHSPitMonkey (Read error: 113 (No route to host))
20:08:11 Quit stoffel (Read error: 60 (Operation timed out))
20:09:03 Join aevin [0] (i=eivindsy@unaffiliated/aevin)
20:10:39 Join mmmmna [0] (
20:12:59mmmmnaAnyone home?
20:14:22gammyerm mm
20:14:49***Saving seen data "./dancer.seen"
20:15:10mmmmnammmm... na.
20:15:43 Nick blithe_ is now known as blithe (
20:16:56 Quit TheSeven ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
20:17:03 Nick Ypsy is now known as YpsyZNC (
20:17:12mmmmnamanual upgrade question regarding Sansa e260 not R?
20:21:25AlexPgammy: I doubt anyone would swap a H140 for a 2nd gen nano
20:21:33mmmmnaok, I'll come back later.
20:21:36 Part mmmmna ("Leaving")
20:21:42AlexPmmmmna: For a manual upgrade, just overwrite
20:22:15gammyAlexP: I only want a H120 without hdd
20:22:31gammyAlexP: but I have to try.
20:22:48AlexPgammy: Yeah, and I would be very surprised if anyone here with a H1x0 would swap it instead of just buy a HD
20:22:53AlexPThey are very sought after
20:23:26AlexPYou could try broken ones on ebay, but even those tend to go for quite some cash
20:23:35AlexPIf they appear at all
20:24:08AlexPStill, you never know :)
20:24:31gammyyeah I noticed
20:24:46gammywhat I need is some non-techie who has one covered in dust on a shelf
20:25:01AlexPThis isn't the best place to look then :)
20:25:40gammyyeah I know
20:25:46gammyStill worth a shot though
20:27:39gammyI've had that iriver for so long and have repaired it so many times
20:27:46gammyit's no longer a music player, it's a companion :>
20:34:02PSPdemonsad day.... my e260 doesnt work correctly anymore ( hardware issue ) :(
20:34:07PSPdemonnow i have to find a new rockbox player!
20:34:23PSPdemonout of curiosity does rockbox work on the onda 777+?
20:35:10PSPdemonive been through the forums and saw that no one mentioned it even though it is jz4740 based
20:39:57AlexPPSPdemon: All supported players are listed on the front page of
20:40:17PSPdemonAlexP, i know this
20:40:24AlexPThen why ask?
20:41:00 Join Grahack [0] (
20:41:05PSPdemonhowever in the new ports wise i know that the 747, 747+, 777, 767, and most Jz4740 ingenic chip players run rockbox
20:41:18AlexPTo some extent
20:41:24PSPdemoni was curious though if it works on 777+
20:41:43AlexPIf they aren't in the New Ports forum or on the wiki, then no
20:46:52 Join stoffel [0] (
20:54:26mcuelenaerePSPdemon: doesn't the VX777+ have another chipset?
20:55:28mcuelenaereah no, it seems it doesn't
20:55:35mcuelenaerethen it should run the VX777 builds
20:56:56kugelwhat's the difference between those ondas?
20:59:39mcuelenaerekugel: the VX777 and the VX777+ ?
20:59:56kugeland 747(+),767 :)
21:01:17mcuelenaere747: my DAP (you've seen it?), 747+: 747 + camera, VX777: newer version of VX747 + TV-out AFAIK, VX777+: same as VX777 + ?, VX767: non-touchscreen, bigger screen and more physical buttons-version of VX777
21:01:52mcuelenaerewpyh has a V767
21:02:08mcuelenaerebut LCD never worked on it and I don't have any players to test it on
21:02:22mcuelenaereother players should work fine
21:02:56mcuelenaereand the builds should also run on VX757, VX9xx (or VX8xx) and possibly other ChinaChip players (Ainol V2000SE, Dingoo A320, ...)
21:03:14mcuelenaereadjusted to their respective LCD and physical buttons of course
21:06:20Unhelpfulhrm! valgrind says that we write some font data that's uninitialized... might explain the garbage chars i'm seeing. :P
21:06:37 Quit antil33t ()
21:07:47kugelUnhelpful: I was having many problems with the monofur font: the normal one is mostly utterly slow with many disk accesses, the italic version is fine
21:08:14kugelthe font has only 600 glyphs or so, and even with height=10 it's horrible. I can't imagine the glyphs are so big
21:08:38Unhelpfulkugel: this sounds like the trouble i'm having. i tried increasing MAX_FONT_SIZE *4... no luck.
21:09:56 Join T44 [0] (
21:11:02Unhelpfulif i go to the bottom of my album list where all of the japanese stuff collects, scrolling the list becomes *terribly* slow
21:11:14Unhelpfulalso i sometimes see junk drawn out of place :/
21:12:45 Join Hillshum [0] (
21:12:47kugelI cant remember having these problems with the initial version of the patch
21:13:44kugelmaybe our resyncs removed code that was actually used. I'm also curious about the "#define ALPHA_COLOR_FONT_DEPTH 2", it should really be 4 I think
21:14:01Unhelpfuli can't either... how long ago was the change 2->4 bits?
21:14:15Unhelpfulwell, the other values calculated *from* it are all correct
21:14:34kugelthe initial patch already has this
21:15:01kugelbut I've seen 2 bit fonts, they look terrible compared to what the patch currently does
21:16:32 Quit Topy44 (Read error: 60 (Operation timed out))
21:17:27 Quit martian67 (Read error: 104 (Connection reset by peer))
21:19:38T44sorry for repeating myself, but i seriously need to get this fixed...
21:20:21T44any ideas why my h340 shows "ata error: -11" when booting, but runs fine when connected to usb and boots into rockbox fine after disconnecting it?
21:20:54T44i had the hd removed to run a surface check and put it back in now, nothing else done to it
21:21:27 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
21:21:41Unhelpfulkugel: well, i don't remember the versions i've tried before looking terrible
21:21:42kugelUnhelpful: (correct channel this time) is there a way to optimise the size of the fonts? the one 12px font is 520k while 12px helvetica is only 35
21:21:50 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
21:22:10T44same problem with the original firmware btw ("check hdd connection")
21:22:16kugelUnhelpful: the patch doesn't do the 2bit fonts, I saw them when the the author worked on it
21:24:39 Nick Hillshum is now known as hasd (
21:24:55kugelUnhelpful: oh, I just notice a comment I've left there
21:26:40Unhelpfuloh, so, older versions were dropping lots of glyphs?
21:27:28Unhelpful16-Mix.fnt is ~5MB, perhaps this is my trouble ;)
21:28:25kugelI don't know, I never noticed missing glyphs
21:28:59kugelbut I also don't really have stuff where I need anything "above" iso8859-1
21:29:49Unhelpfuli thought glyphs were loaded and cached as needed... should the memory requirement really go up if i'm still displaying the same text?
21:31:54kugelI don't think so. The cached glyphs are already the bitmaps, right? They shouldnt be much bigger that mono fonts IIUC
21:32:59Unhelpfulum, 4x bigger... 2bpp vs 1bpp. but i meant because of the extra characters. it shouldn't matter how many characters the *font* has if the number i display stays the same, right?
21:33:16Unhelpfuls/2/4 :)
21:33:42kugelI meant bitmaps in the native format, but I might be wrong
21:34:52gevaertsJdGordon says you're probably wrong
21:35:19kugelgevaerts: slap him
21:35:19Unhelpfulkugel: oh, no, they're *smaller* than native-format bitmaps, they're 4bpp vs 16bpp
21:36:02kugelUnhelpful: I thought they were in native format in the font buffer
21:36:05Unhelpfulkugel: did you take a look at my convttf changes? i'm still seeing some broken bits... i can't findanything obviously wrong, but i didn't write this originally ;)
21:36:37kugelI didn't either ;)
21:36:45kugelwhat sort of changes did you make?
21:36:48Unhelpfulkugel: how would they be? obviously we're passing them to the draw routine as 4bpp, and i don't see any code added to convert down
21:39:03Unhelpfulbasically, the final line height is supposed to be the -p you requested. there will be -r pixels of padding added above and below characters, with -r defaulting to 1, and the magic "add a row to line height" gone
21:39:11rasherscorche|sh: wrong link
21:39:43rasherBut I don't see it..
21:39:45scorche|shoh, right
21:39:51kugelUnhelpful: that's what my version also does
21:40:03kugelfor the -p thing at least
21:40:10Unhelpfulkugel: i'm also doing it without rerunning the height test. :)
21:40:20T44the tower of rockbox is such a crazy idea considering most of those players are hd based and have very slippery surfaces :)
21:40:24scorche|shrasher: try it now
21:41:25UnhelpfulT44: we're such such wiiiiiild an ca-razy guys!
21:41:26kugelUnhelpful: I do that because fonts are generally too big without rerunning (they look giant compared to our mono fonts even if the height is the same), and you cannot see all common chars entirely sometimes. rerunning scales the font down bug gives better results should look at how i did it, perhaps? ;)
21:44:22 Quit martian67 (Connection timed out)
21:44:35 Join evilnick1 [0] (
21:44:44Unhelpfulbasically the height test is reworked from being a measuring of the font at a specific size to a way to calculate the correct size
21:46:34 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
21:46:43kugelUnhelpful: I see, won't that give smaller fonts?
21:47:14 Quit kushalone (Client Quit)
21:47:26Unhelpfulit will give you fonts that, including the vertical padding, are the height you requested. isn't that what we want?
21:47:34kugel(some of) our mono fonts are also cut off, and some glyphs aren't displayed correctly in favor of slightly bigger common glyphs
21:48:07 Join Strife89 [0] (
21:52:13Unhelpfulhrm, i'm pretty much satisfied with the results @ 16px. the only real problem is the garbage charters... wish i knew where those came from.
21:54:07kugelUnhelpful: see FS #9931
21:56:19 Join niekie [0] (
21:56:48Unhelpfulalso, if we're going to go a few steps larger or smaller to try to fit, and then just truncate, i can't see how that's much better than scale-to-fit. the amount truncated ends up depending on how correct the fonts ascent/descent values are.
21:57:11Unhelpfuladding a truncate option, or an option to align glyphs up or down, would be a better way to get larger glyphs
21:58:35 Quit hasd (Read error: 145 (Connection timed out))
21:59:56niekieGreetings everyone. It seemed my MP3 player has died on me, and I'd like my next one to be RockBox compatible. Are there any players anyone would specifically recommend?
22:00:37 Quit bubsy ("I'll be back somewhere in time...")
22:00:42niekieI see Archos and iPods are supported. I saw an Archos player at the store today, though I'm not sure if it was one of the supported ones.
22:01:09gevaertsmost probably not
22:01:27niekieAlso, as Archos is listed as MP3 and WAV only, that's not what I'd really want.
22:01:41niekieSince most of my music collection is in free formats like .ogg and .flac
22:01:48Unhelpfulniekie: that's because the supported archos devices are *very* old ones with hardware decoding
22:03:11niekieHardware decoding is bad.
22:03:14CtcpIgnored 4 channel CTCP requests in 0 seconds at the last flood
22:03:14*bluebrother wonders if we could sendfirm the bootloader on the H10 MTP
22:04:03pixelmathe "best" player for you is depending on what you want/need like recording, radio, screen etc. The BuyersGuide in the wiki gives an overview and might be a good start
22:04:21pixelmaor what you don't need ;)
22:04:59niekiepixelma: ah, many thanks, looking at it right now.
22:08:50 Join TheSeven [0] (
22:11:44T44i have a h340 and i think it supports just about everything that rockbox can do pretty well. its a big old chunky piece of hardware though. kids will laugh at you in the subway. :)
22:13:22niekieBetter than having your laptop out there and listening music that way ;-)
22:14:53***Saving seen data "./dancer.seen"
22:16:47 Join Sajber^ [0] (
22:22:14 Join hasd [0] (
22:28:03T44niekie: nowadays with netbooks and nerds being considered "cool", i guess thats perfectly acceptable :)
22:28:17T44anyway, FAR off topic
22:30:46LambdaCalculus37T44: Yes, that was quite off topic.
22:33:14 Join petur [0] (n=peter@rockbox/developer/petur)
22:37:15 Join bubsy [0] (n=Bubsy@
22:42:47 Quit petur ("Zzzzz")
22:43:06 Quit FOAD ("I'll be back")
22:43:19 Join FOAD [0] (
22:45:10 Nick hasd is now known as hillshum (
22:45:23 Quit CaptainKewl (Read error: 54 (Connection reset by peer))
22:48:26 Join Ubuntuxer [0] (
22:48:28 Quit T44 ()
22:49:16 Join CaptainKwel [0] (
23:04:06 Quit rphillips ("leaving")
23:04:45 Join rphillips [0] (
23:05:49 Quit Turniphead (Read error: 104 (Connection reset by peer))
23:07:37 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
23:21:50 Join faemir [0] (n=faemir@
23:22:41 Quit Rondom (Nick collision from services.)
23:22:51 Join Rondom [0] (
23:43:16 Quit rphillips (Client Quit)
23:43:59 Quit Rondom (Nick collision from services.)
23:44:09 Join Rondom [0] (
23:45:52 Quit stoffel (Remote closed the connection)
23:46:05CIA-61New commit by mcuelenaere (r22465): Touchscreen list handling: fix issue where scrolling with the scrollbar could bring up the context menu for an item
23:50:21 Join rphillips [0] (
23:52:37 Join MG_Man [0] (
23:53:05 Quit Ubuntuxer ("Leaving.")
23:54:27 Quit Rondom (Nick collision from services.)
23:54:37 Join Rondom [0] (

Previous day | Next day