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 2010-10-18

00:00:09HorschtVersion: r28070-100913
00:04:22 Quit leavittx_ (Ping timeout: 240 seconds)
00:22:56 Quit Zarggg (Ping timeout: 255 seconds)
00:24:45 Join Zarggg [0] (
00:27:02 Join S_a_i_n_t [0] (S_a_i_n_t@
00:37:54 Quit S_a_i_n_t (Disconnected by services)
00:37:56 Join S_a_i_n_t [0] (S_a_i_n_t@
00:49:49 Quit ender` (Quit: Today's BOFH excuse: broadcast packets on wrong frequency)
01:07:36 Quit Zarggg (Quit: Zarggg)
01:07:42***Saving seen data "./dancer.seen"
01:10:28 Quit s1gma_ (Quit: Leaving)
01:16:42 Join JdGordon| [0] (
01:16:43 Quit JdGordon| (Changing host)
01:16:43 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
01:19:58 Join Zarggg [0] (
01:25:55 Quit hebz0rl (Quit: Leaving)
02:05:52 Join chattr [0] (
02:11:14 Join lmjubuntu [0] (~lmjubuntu@2001:250:6801:4418:2ea:1ff:fe12:2816)
02:19:17 Quit crwl (Ping timeout: 276 seconds)
02:19:52 Join crwl [0] (
02:20:55 Part lmjubuntu
02:36:48 Quit robin0800 (Remote host closed the connection)
03:07:44***Saving seen data "./dancer.seen"
03:14:51saratogai need to talk to buschel but i think this mp3 idea will work
03:16:02saratogaand i think it will be a lot faster
03:18:02 Join robin0800 [0] (
03:21:05 Join krabador [0] (
03:38:54 Quit robin0800 (Remote host closed the connection)
03:39:18 Join Sudos [0] (
03:50:50 Quit chattr (Quit: gone)
03:52:43 Join chattr [0] (
03:56:05 Quit krabador (Quit: Sto andando via)
04:00:42scorchesoap: so as i said last night, there is a "Maximum time after posting to allow edit" option where one specifies a number of minutes
04:06:31 Quit TheSeven (Ping timeout: 276 seconds)
04:09:35 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:11:45JdGordon|nothing to disable editing after someone has replied though? :(
04:15:23 Quit kugel (Remote host closed the connection)
04:19:35 Join Barahir_ [0] (
04:23:30 Quit Barahir (Ping timeout: 272 seconds)
04:29:45 Join saratoga_ [0] (9803c22e@gateway/web/freenode/ip.
04:29:54saratoga_what the hell:
04:30:06saratoga_since when do we run on palm
04:30:29JdGordon|haha wow!
04:30:38JdGordon|Davide-nyc would be happy
04:31:55JdGordon|looks like we need a chinese speaker to get in touch with him
04:33:16saratoga_doesn't look like he posted the source
04:33:42JdGordon|google translate no worky:(
04:33:52JdGordon|how did you find that?
04:34:20saratoga_i was googling to see if people were actually using rockbox on android and that popped up
04:34:40saratoga_translate is down for chrome, but the search engine translate works
04:34:47saratoga_well works in that it gives you a nonsense translation
04:35:12saratoga_webos is apparently linux + SDL so I suppose it was not a difficult port
04:35:20saratoga_but i'm curious how its implemented
04:35:49saratoga_do we have any developers who speak chinese
04:36:17JdGordon|I think markun can (or was learning?)
04:40:00saratoga_bah its not a RAAA build
04:40:04scorchesaratoga_: my girlfriend can...
04:40:09saratoga_its a hacked up Gigabeat S build of 3.5.1
04:41:11saratoga_scorche: thanks, but we probably need someone familiar with rockbox development to talk to them
04:41:22saratoga_and hopefully encourage them to use RAAA now that its mostly ready
04:41:39scorchewell, she can translate what i say...
04:41:47JdGordon|if its a simple beast sim then raa would almost certainly "just work"
04:42:30JdGordon|hopefully this doesnt end up like the moto build :/
04:42:41saratoga_yeah might be nice to post there saying that we've now got the beginning of a framework for porting rockbox to phones
04:42:55saratoga_and encourage them to submit the source
04:43:25 Join dys` [0] (
04:45:40 Quit dys (Ping timeout: 276 seconds)
04:45:53scorchewrite something up, and i can have her try to translate a post
04:46:27JdGordon|well more than that... explain that very likely the only changes he would need are to configure to make the application build use the correct compiler and paths, and any hardware changes
04:47:05 Join bluebrother [0] (
04:47:05 Quit bluebrother (Changing host)
04:47:05 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
04:47:42saratoga_either jon or kugel should do it
04:47:54 Quit bluebroth3r (Read error: Operation timed out)
04:49:01saratoga_that reminds me, how close are we to getting rid of the grid interface touch devices?
04:49:53scorcheand they should try to stay away from really technical terms if they can...i doubt the gf would be able to translate extremely technical things - they generally are a very slang-ish thing in Chinese and unless being immersed in the tech culture there, it would be almost impossible to know the correct translation
04:50:30saratoga_do we really not have a single native chinese speaker doing rockbox development?
04:51:31scorchei dont think so - we could perhaps see about tracking down and asking the last person to do big edits to the lang files
04:53:29scorchei am sure there are a few names in the tracker if we look hard enough
04:53:52scorchethis person, perhaps?
04:56:12JdGordon|saratoga_: re getting rid of grid... that will never be gone, but as far as changing it from being the default, all we really need is a touch enabled sbs
04:56:20JdGordon|that would at least cover 90% of the ui
04:56:31saratoga_is that particularly hard to do?
04:56:47 Quit rvvs89_ (Ping timeout: 240 seconds)
04:56:52saratoga_sbs is the status bar?
04:56:56JdGordon|no, but it is "arty" which means argumentative
04:57:25JdGordon|the list skin... I've started one for 320x480 which is in svn and works pretty well
04:57:40saratoga_whats it actually stand for?
04:57:54saratoga_and how does it work? just a big font and you touch the item you want?
04:57:58 Join rvvs89 [0] (
04:58:06JdGordon|origionally "statusbar skin"
04:58:19 Quit amiconn (Disconnected by services)
04:58:20 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:58:20 Quit pixelma (Disconnected by services)
04:58:22 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:58:24 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:58:40 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:59:04JdGordon|the lists all work fine with a stylus, but things like going back to menus, or even up one level in the browser is hard with a finger without a sbs
04:59:47saratoga_i guess some kind of android style swipe left or right to go up through trees isn't possible?
05:00:04JdGordon|it should be
05:00:48saratoga_that actually sounds like something that might be worth taking directly from the android source code
05:01:52saratoga_but yes i think getting touch working reasonably well by default is pretty important, since its probably a prerequisite for getting people using rockbox on android
05:02:02saratoga_and getting users is a good first step to getting more developers
05:02:29 Quit TheSeven (Ping timeout: 265 seconds)
05:02:41saratoga_google suggests that most people playing around with android builds are baffled by the controls
05:02:43JdGordon|indeed, and yeah, it is entirely waiting for a sbs now, android is perectly useable now
05:02:50JdGordon|or no less than any other in development port
05:02:51saratoga_which neatly mirrors my own experience trying kugel's builds
05:03:28*JdGordon| wonders about drawing an overlay to show the grid positions
05:03:47saratoga_i'm hoping once the port becomes popular we'll see someone take interest in an icon based touch interface as well
05:04:17JdGordon|I'm not so sure that would work well for anything except the main menu
05:04:40saratoga_i was thinking folder icon (or album art) and title below it in a grid
05:04:44saratoga_gives you more room to touch with a finger
05:05:23saratoga_as the resolution of android devices creeps up, i think that'll become more and more attractive
05:05:28JdGordon|I have a really nice idea how to do it with the current skining system, but no clue how to acually code it so it works
05:07:43 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
05:07:47***Saving seen data "./dancer.seen"
05:12:10soapscorche, how about we just switch forum software to IP.Board? I'll pay for at least the first year.
05:12:39saratoga_whats wrong with what we have?
05:13:13soaplack of fine-grained control. Poor mobile interface.
05:13:25soapbut not really a whole lot.
05:13:35saratoga_if we're going to mess around with the site, I'd rather have some blog software built in :)
05:13:35scorchein its defense, it is a number of revisiona behind
05:13:50saratoga_so we could do news items like xbmc does
05:14:08soapor HA, yes
05:14:21JdGordon|copying HA cant be a good thing! :D
05:14:26scorchei was meaning to do an upgrade of the forums at DevCon, however i apparently was having too much fun and forgot about it
05:14:30saratoga_i've given up on ever having a new front page, but would be nice
05:15:10soapa front-page dictator needs to sieze power. Front-page-by-committee failed.
05:15:37scorcheit wouldnt be hard for me to set up one, however, I am a bit hesitant about Wordpress - there are so many security vulnerabilities that seem to get reported for it
05:15:51saratoga_what are they using?
05:15:55saratoga_it looks pretty awesome
05:16:41JdGordon|it would be really awesome if we could integrate the wiki+forum+blog+tracker into one user/pass combo and look/feel
05:17:05scorcheif we do do wordpress, i just need to be on my toes more about updating...
05:17:29saratoga_hmm ok
05:17:45scorcheJdGordon: it would be, but as it stands, they are hosted on multiple servers
05:17:46saratoga_its weird how sourceforge/google/etc don't provide something like this for open source projects
05:18:03JdGordon|isnt that what TRAC is?
05:18:05saratoga_i don't really care about common accounts
05:18:18JdGordon|and sourceforges crappy forum/tracker thing
05:18:24saratoga_hell we don't even really need comments on the blog, we could just link to the normal forums
05:18:46JdGordon|thats what do for their news items...
05:19:03JdGordon|news is a forum thread which the frontpage loads, comments link back into the thread
05:19:35JdGordon|if the forum db is accessable that potentially owuldnt be too difficult to add
05:20:38JdGordon|saratoga_: how do you know he used a beast sim?
05:21:10saratoga_Target define: -DGIGABEAT_S
05:21:39saratoga_we should make rockbox-info.txt include the output of SVN diff
05:22:01JdGordon|that could be rather big
05:22:16saratoga_yeah but its just text, so it'll zip down to nothing
05:22:28saratoga_plus its not our problem ;)
05:24:12saratoga_scorche: so are you actually interested in setting up wordpress or is that not a good solution?
05:24:59scorchesaratoga: i suppose i could, however, i would rather other people do the template/css stuff
05:25:39saratoga_any idea who the other people are?
05:25:57saratoga_maybe ask on the dev list?
05:26:58scorchewell, likely the same people who worked on the theme site?
05:27:12JdGordon|which didnt go anywhere, so they may not be bothered
05:27:46*JdGordon| learns to read
05:28:05 Quit Horscht (Quit: Verlassend)
05:31:39 Quit ps-auxw (Read error: Operation timed out)
05:31:48JdGordon|there is probably a forum plugin to do this also
05:32:03saratoga_ok maybe send an email to the dev list and see if anyone is interested in helping?
05:44:58 Join ps-auxw [0] (
05:47:19*JdGordon| has a eureka moment in the dunny and fills a partial gap in his list skining plans
05:48:48S_a_i_n_tTMI, TMI!
05:50:05JdGordon|actually umm... :/
05:50:55JdGordon|arg, no, scrolling is still going to kill everything
05:51:30S_a_i_n_tCould you do forced scrolling?
05:51:51S_a_i_n_tlike the "hold X" to scroll business, and just not scroll automatically?
05:52:12JdGordon|that would suck... and no, maybe it isnt as bad as I think
05:53:02 Quit saratoga (Ping timeout: 265 seconds)
05:53:59saratoga_maybe having a second interface for touch targets would be better? (i don't know anything about the code though)
05:54:45JdGordon|I could see people wanting a grid or more control over the lists on non touch targets also
05:55:09 Join olll [0] (4ca6839b@gateway/web/freenode/ip.
05:55:58saratoga_i think having two interfaces would be nice in that it would enforce more abstraction of the UI from the core, but maybe its not worth it
05:56:30saratoga_eventually i want to have a way to issue UI actions via script for testing
05:56:34JdGordon|unless someone wants to go and make a headless RaaA port I dont tihnk that would happen :)
05:56:50JdGordon|nice timing there, that remark works a s a reply to both your last ones
05:57:37JdGordon|rasher did make a patch to issue commands to the sim through a named pipe, but a proper daemon would be awesome
05:57:57saratoga_really? is it on FS?
05:58:26saratoga_FS #10575
05:58:28saratoga_how did i miss that
05:59:47saratoga_how does that work for time sensitive actions?
05:59:57saratoga_can you tell it to pause for a second then do another action?
06:00:10JdGordon|wouldn't have a clue
06:00:23 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
06:00:26JdGordon|witha bit of tweaking it oculd probably be done
06:01:54saratoga_ugh this would have been amazingly useful last fall when debugging the clipv1 issues
06:02:26JdGordon|no, a proper way to run the playback engine outside the ui would be better :p
06:03:09 Quit S_a_i_n_t (Ping timeout: 240 seconds)
06:07:41saratoga_yes it would
06:09:26olllhas anyone had any sucess compiling RaaA with wince-gcc?
06:09:57JdGordon|I don't know if anyone has tried
06:11:27JdGordon|saratoga_: "load track -> play for 20 seconds -> skip backwards 5 seconds -> " the seeking could be interesting... if you use the seek setting that is easy, but if it actually needs you do hold |<< then you'd have to figure out how many SEEK|BUTTON_REPEAT actions would be needed to get the seek time correct
06:14:57olllI've spent a most of last sunday without success. With RaaA and openzdk, the zune seems much more approachable, maybe even a simpler windows mobile port
06:16:09JdGordon|what sort of problems are you hitting?
06:19:30ollli believe missing header/libraries, havent booted up my linux machine since last sunday :P
06:19:49JdGordon|are you just trying to build the RaaA sdl port?
06:20:03JdGordon|does openzdk have sdl?
06:21:29saratoga_when i played around with wince-gcc the problem i ran into was that many of the wince headers are weird
06:21:30olllno, i was starting with an existing sdl port for wince
06:22:01olllyeah, ^ i read the archives
06:22:16saratoga_although the recent android changes may have improved that, i'm not sure
06:22:24 Quit anewuser ()
06:22:58JdGordon|you should be able to build a binary without sdl dependancies at all if you stub a bunch of functions
06:23:12JdGordon|I thought kugel was going to write up how to port RaaA
06:23:51saratoga_he put a bit on the wiki about it IIRC
06:25:23 Join elcan [0] (
06:30:38JdGordon|it would be nice if those stubs were added as a "stub" RaaA port to help new people
06:31:17 Quit simonrvn (Ping timeout: 240 seconds)
06:31:18saratoga_yeah that would be cool
06:34:52 Join nostra [0] (53e96d9b@gateway/web/freenode/ip.
06:36:48nostraHi guys i found ur website few hours ago, cause i was gonna update the firmware for my sansa express (simple mp3-player)... but what was going throu my mind was "Can i make my own firmware to this device?" So now am asking you guys that knows alot about this. Is it possible for me to build my own firmware for this simple mp3-player?
06:37:59nostraIs there someway to decompile their rsc and take a look or something? All help would be greatful.
06:38:41JdGordon|simple answer is.. "depends on your skill level"
06:39:06JdGordon|I dont tihnk any work has been done on the express...
06:40:07nostraI do know some program a bit but i dont know where to start
06:43:31 Join S_a_i_n_t [0] (S_a_i_n_t@
06:45:37nostrasaratoga_: thank you, looks intressting
06:47:47 Join leavittx_ [0] (~lev@
07:07:49***Saving seen data "./dancer.seen"
07:13:10 Quit saratoga_ (Quit: Page closed)
07:13:32nostrahow am i able to look into them *.rsc? i found some text what they including but i want to see by myself
07:14:24nostraand *.sb
07:19:52 Quit MethoS- (Remote host closed the connection)
07:21:32S_a_i_n_tgoogle is your friend. seems to be MS Small Basic, and there are several .RSC editors I found from a singular search query
07:23:31JdGordon|I doubt .sb here is a MS file
07:23:46S_a_i_n_tI could also be a "Signed Byte" audio file, in which case Quicktime should be able to open it at least.
07:24:05S_a_i_n_tor some completely random other extension ;)
07:24:06JdGordon|isnt that the sigmatel firware image type?
07:29:04nostraMakes sense, "Sansa express, One of the Sigmatel based devices in the Sansa line"
07:29:54JdGordon| might be helpful
07:32:24 Join Buschel [0] (
07:33:27 Join einhirn [0] (
07:34:55 Join merbanan [0] (
07:35:20Buschelsaratoga: (for the logs) I have read all the mp3-discussions of last night. the dct32-asm stuff in libmad was deactivated as it is reasonably faster to use the C-version. the dct32-code wasn't optimized for cycles but for size −− the old cpu's seem to have handled those faster
07:35:41 Quit CaptainKewl (Ping timeout: 250 seconds)
07:35:45Buschelsaratoga: what is the idea you wanted to talk about?
07:37:39 Quit TheSeven (Ping timeout: 265 seconds)
07:41:44 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
07:49:24nostra"There actually aren't any separate radio chips because the FM tuner is built into the SigmaTel STMP3600 system-on-a-chip that runs the device. " wow so i could get FM if i get my own firmware into sansa express? :)
07:50:09ollldepends if your specific chip supports it, not just the product line
07:50:43nostrastill intressting, gotta learn this with firmwares for sure
07:55:07 Join stoffel [0] (
07:58:30 Join simonrvn [0] (
08:01:22nostraIv tryed ResEdit and few more programs still cant take a look into player.rsc
08:03:21JdGordon|the .sb is probably more important/interesting
08:05:02nostraYeah am gonna give 'em a try instead, see what i can get out of it
08:07:11 Quit simonrvn (Quit: ZNC -
08:07:36 Join simonrvn [0] (
08:07:51 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854])
08:13:30 Join S00row [0] (
08:21:57 Quit TheSeven (Remote host closed the connection)
08:30:01nostrais it possible to make ur own startup-LCD-animation on a device if you make ur own firmware?
08:31:36JdGordon|you can do anything if you make your own firmware
08:31:46nostraThat what i was thinking yeah
08:31:55tmzt_seems to be talking about repacking sandisk stuff
08:32:22nostraI just want to try to edit it somehow
08:33:33nostralearn more about devices
08:33:56nostraatleast firmwares
08:34:45tmzt_firmware is just the software that runs the device, this project is about an entirely new firmware for various audio players
08:34:58tmzt_you can read about it on the website referenced in the topic
08:35:40tmzt_just a hint, the animiation is probably either executable code or a series of bgr565 bitmaps in a compressed format of some sort, possibly a variant of rle
08:37:20 Quit JdGordon| (Quit: leaving)
08:37:59 Join bertrik [0] (
08:37:59 Quit bertrik (Changing host)
08:37:59 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:39:18 Join ender` [0] (
08:40:37 Quit olll (Quit: Page closed)
08:56:22nostratmzt_: Thank you for the hints, but it dosnt seems like it's that eazy to find where it is at in the firmware
08:58:20nostrayeah i know whats this project is about. Its a good project too. Maybe i will join when i got more skillz about this, but am sure its fun
08:59:45nostrai mean i will support when my skillz are increased
09:06:20 Join Rob2222 [0] (
09:07:51***Saving seen data "./dancer.seen"
09:09:43 Quit Rob2223 (Ping timeout: 245 seconds)
09:15:02 Quit jhMikeS (Ping timeout: 264 seconds)
09:23:13 Join Kitr88 [0] (
09:25:35 Quit Kitar|st (Ping timeout: 255 seconds)
09:25:46 Quit bertrik (Ping timeout: 250 seconds)
09:26:04 Nick dys` is now known as dys (
09:27:30 Quit Kitr88 (Ping timeout: 250 seconds)
09:28:07JdGordonsoap: scorche: can you ban dobson52 on the forums?
09:30:55nostrawow... i installed another firmware and got FMradio now :)
09:31:30JdGordongreat, but the only firmware which we allow discussion of here is rockbox
09:32:42 Join Kitar|st [0] (
09:38:48CIA-7New commit by jdgordon (r28303): Don't try to show a backdrop if the bmp wasnt loaded successfully
09:40:45CIA-7r28303 build result: All green
09:44:41 Join Guest14447 [0] (~bjst@rockbox/developer/Zagor)
09:46:04 Nick Guest14447 is now known as Zagor (~bjst@rockbox/developer/Zagor)
09:47:21 Join swilde [0] (
09:47:37 Quit stoffel (Remote host closed the connection)
09:59:02 Quit sasquatch (Quit: WeeChat 0.3.2)
09:59:27 Join sasquatch [0] (
10:08:41 Join hebz0rl [0] (
10:19:11 Join moos [0] (
10:19:11 Quit moos (Changing host)
10:19:12 Join moos [0] (moos@rockbox/staff/moos)
10:24:12 Quit simonrvn (Quit: ZNC -
10:26:01 Quit elcan (Read error: Connection reset by peer)
10:26:23 Join elcan [0] (
10:30:01 Quit nostra (*.net *.split)
10:34:02 Join DerPapst [0] (
10:34:17 Quit S00row (Read error: Connection reset by peer)
10:35:20 Join S00row [0] (
11:02:45 Join LinusN [0] (
11:02:46 Quit LinusN (Changing host)
11:02:46 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:07:55***Saving seen data "./dancer.seen"
11:08:06 Quit hebz0rl (Quit: Leaving)
11:09:33 Join simonrvn [0] (
11:09:36 Quit S00row (Read error: Connection reset by peer)
11:11:56 Join S00row [0] (
11:20:43 Quit timccc (Ping timeout: 252 seconds)
11:22:15 Join timccc [0] (~tim@
11:29:08 Quit Kitar|st (Ping timeout: 260 seconds)
11:36:54 Quit Rob2222 (Quit: Rob2222)
11:38:31 Join Rob2222 [0] (
11:40:50 Join Kitar|st [0] (
11:44:29 Join petur [0] (~petur@rockbox/developer/petur)
12:06:31 Quit chattr (Quit: gone)
12:07:13 Quit S00row (Ping timeout: 245 seconds)
12:11:06 Join chattr [0] (
12:12:45*JdGordon spots yet another path which wasn fixed for the get_user_file_path() work :/
12:14:02 Quit swilde (Ping timeout: 276 seconds)
12:16:38JdGordonwhy oh why wasnt it done silently in the open() call to just replace /.rockbox/ ? :'(
12:28:00 Join wodz [0] (
12:34:55 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
12:35:23wodzJdGordon: ping
12:41:15wodzis there a chance You will have access to Your archos recorder in not geological timeframe?
12:42:09JdGordonnot likely soon...
12:48:28wodzJdGordon: another question - currently only bmp graphic is used in themes - what would be needed to support e.g png beside decompressor in core?
12:49:09JdGordon2 lines changed from the skins point of view
12:50:10JdGordondoes png really give any benefits?
12:50:23wodzjust thinking about it
12:51:01wodzmost of my rockboxed targets are broken so I must hack something :-)
12:51:06JdGordonit would make more sense to use jpg instead of png, at least we dont need to add code for that
12:52:08JdGordonwell if you're bored.... want to play with buffering/playback? :D
12:52:35wodzjpg would have unpleasant visual effects with sharp objects like lines etc
12:56:09wodzSimply make black line on white background and save it as jpg. You will see that around edge there is 'noise' comming from the fundamental way jpg is compressed
13:01:25JdGordonjust for shits n giggles... could the image codecs be put into seperate files and swapped out as needed like the audio ones? using the same buffer even?
13:04:19 Quit leavittx_ (Ping timeout: 240 seconds)
13:05:33wodzthat's how it is proposed for imageviewer plugin by teru. He want to use overlay mechanism to load decoder when needed
13:07:57***Saving seen data "./dancer.seen"
13:08:46JdGordonnot as a plugin... in the core
13:10:46soapyes, wodz, jpg can be noisy - but that depends greatly on the situation and the compression settings. It need not be /visibly/ noisy.
13:13:50 Join teru [0] (
13:21:24 Quit S_a_i_n_t (Ping timeout: 245 seconds)
13:24:54 Quit benedikt93 (Ping timeout: 276 seconds)
13:25:12 Join S_a_i_n_t [0] (S_a_i_n_t@
13:25:19 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:27:38 Quit benedikt93 (Read error: Connection reset by peer)
13:28:05 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:38:40 Quit S_a_i_n_t (Quit: I'm only going to Heaven if it tastes like caramel...)
13:41:00JdGordondoes anyone remember any discussion about why we dont transparently open the right file in open() instead of getting the path before?
13:41:39JdGordon. /.rockbox is very very unlikely to be a real path on any OS
13:42:07wodzsomething is terribly broken - I just updated my MPIO to svn head and now it loads rb image ok, shows rockbox splash and turns off immediately without any notification.
13:48:25JdGordonok, open() goes straight to the OS for RaaA builds, slightly annoying
13:48:59gevaertsJdGordon: there was some discussion, and IIRC kugel wanted to do it this way. I don't remember why
13:49:33gevaertsI think to avoid having to play tricks with wrappers and linkers
13:50:21JdGordonwhich happens anyway unless file.h isnt being included at all?
13:55:15 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:56:23kugelJdGordon: why should we wrap open()?
13:56:31linuxstbJdGordon: It seems simpler to me for apps/ to know about the correct path, rather than relying on firmware/ translating it. And how would you handle files being in different places - i.e. not a single prefix?
13:56:37 Quit moos (Ping timeout: 250 seconds)
13:57:08JdGordonthis get_user_file_path() is umm... quirky to say the least
13:57:52JdGordonlinuxstb: same was as it is done now... ROCKBOX_LIB_PATH != ROCKBIX_DIR on RaaA
13:58:10JdGordonapps/ doesnt care where the files actually are, as long as they can be opened
13:58:26JdGordondoing this transparently in open() gives the individual port so much more flexibility
13:58:51JdGordonand 0 cost for normal builds
13:58:58kugelwe wanted access to the full filesystem in raaa
13:59:24JdGordonit has it
13:59:33JdGordon/.rockbox/ will never be a valid directory
14:00:16JdGordonin fact, we dont allow relative paths so ".rockbox/" would even be good enugh
14:00:26kugelyou want a virtual .rockbox dir in the root?
14:00:45JdGordonyes, but not visible to the user
14:01:00kugelwhy not visible?
14:01:24JdGordonbecause they would see it in /home/user/.config/rockbox/blaa like normal
14:01:59kugelyou can do this I guess, but does this work out in multi-user environments?
14:02:47JdGordonWhat I want is open(ROCKBOX_DIR "/somefile.blaa", O_CREAT); to be all the required apps/ code, open() would then see we are trying to open something in "/.rockbox" and see we need to have write access so it will open ~/.config./rockbox/somefile.blaa
14:02:49JdGordonof course
14:02:52kugelwe install into /usr/local so that it's initially multi user compatible; created files go under the user's home dir
14:03:34JdGordonthe actual file selection would be the same as svn, just that there would be no redundant call to get_user_file_path()
14:04:32kugelthat would probably be nice (if you can get it to work)
14:04:53kugelopen() is not wrapped because there's no need to, if you need to wrap it do it
14:05:12 Quit antil33t (Read error: Connection reset by peer)
14:05:21 Join antil33t [0] (
14:08:37JdGordonbtw, the sim does this already anyway.. sort of
14:08:39wodzwhat is the function that display rockbox screen on startup?
14:09:04wodzI mean the spash
14:09:05JdGordon"static char buffer[MAX_PATH]; /* sufficiently big */" <- umm.. no it isnt!
14:09:30kugelJdGordon: the sim wraps all opens to the current dir which we don't want on raaa
14:09:39JdGordonwodz: show_logo()
14:09:50JdGordonkugel: I know, thats why I said sort of :)
14:10:32kugelcan you do it without such a static buffer, which isn't very thread safe?
14:10:35JdGordonopendir() admitadly can cause problems...
14:10:47JdGordonthat paste was from svn code
14:13:00JdGordonopendir() I think needs to always load the writeable filesystem. I don't think it would ever make sense to load /.rockbox instead of actually loading a file except when we want to generate a filename which means we are writing
14:15:17kugeluhm no
14:15:38kugelif there's no themes installed in the user dir, then we load the themes from the installation dir which is not writable
14:15:41JdGordon*outside of the browser*
14:15:47kugeland browsing them is only an opendir
14:16:57kugelthe same applies to codecs and plugins which are always in the installation dir
14:20:46 Quit xnyhps (Ping timeout: 272 seconds)
14:21:05 Join xnyhps [0] (
14:23:37JdGordon works fine apart from "no .rockbox dir" splash because opendir() isnt done
14:26:59JdGordonget_user_file_path() needs to call real open() instead of sim_open() but yes that works perfectly
14:27:05 Quit elinenbe (Read error: Connection reset by peer)
14:27:06 Quit Loto (Quit: Loto)
14:27:08JdGordonopening: /.rockbox/backdrops/cabbiev2.bmp
14:27:08JdGordonopening: /home/jonno/.config/
14:27:22JdGordonopening: /.rockbox/database_idx.tcd
14:27:22JdGordonopening: /home/jonno/.config/
14:27:23JdGordonopening: /usr/local/share/rockbox/database_idx.tcd
14:27:33 Join elinenbe [0] (
14:27:36 Join einhirn_ [0] (
14:28:13kugelit opens database_idx.tcd twice?
14:28:14 Join Loto [0] (~ctrlproxy@xbmc/user/Loto)
14:28:54JdGordonno, each open() does a printf.. because get_user_file_path() calls file_exists() which calls open() it shows 3 printfs
14:29:01Stummimaybe it means "try to open..."?
14:29:08 Join anewuser [0] (kvirc@
14:29:10 Quit anewuser (Changing host)
14:29:10 Join anewuser [0] (kvirc@unaffiliated/anewuser)
14:29:21 Quit einhirn (Ping timeout: 255 seconds)
14:29:24JdGordonin the not-proof-of-concept patch it would call os open() not the open wrapper
14:29:35tmzt_could just say stating
14:29:55JdGordonthat was just to prove that the code worked... it wont last long
14:38:24JdGordonwriting to /.rockbox/themes/theme01.cfg fails if ~/.config/rockbox/themes doesnt exist :/ (problem in svn also)
14:39:15 Join Sudos|C840 [0] (~ZOOP@
14:39:39 Quit benedikt93 (Ping timeout: 264 seconds)
14:41:02 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:52:00 Quit benedikt93 (Ping timeout: 276 seconds)
14:53:39 Quit anewuser (Ping timeout: 255 seconds)
14:57:15Sudos|C840Just noticed something. I have a 3rd-gen iPod 10GB, Rockbox identifies it as an MK1504GAL, which is listed as a 15GB drive. Partition table also sees it as a 10GB drive with the usual 39MB of unallocated space in the front of the data partition.
14:58:23Sudos|C840I'm about to crack it open to confirm.
14:58:47gevaertsWhat does dmesg say?
14:59:30Sudos|C840not a thing- I have a Windows machine :/
15:00:32gevaertsright, then what do Windows looking-at-disks tools say?
15:01:13Sudos|C840exactly as I said - reports as a 10GB drive.
15:01:21Sudos|C840I cracked it open- it IS a 10GB drive.
15:01:31Sudos|C840and it's been replaced? oh wonderful.
15:02:45gevaertsThere being a 10GB partition doesn't mean it's a 10GB drive
15:03:05gevaertsAnd if rockbox says it's a MK1504GAL, it's because the drive says it's a MK1504GAL
15:03:58 Join anewuser [0] (kvirc@unaffiliated/anewuser)
15:04:04Sudos|C840yeah, but I wanted to make sure it wasn't a 15GB as the somewhat-unupdated Wiki states.
15:04:32Sudos|C840it's an actual 10GB drive as printed on the label, and the serial number checks out with Toshiba.
15:05:03Tornemodel numbers don't always iomply a single size
15:05:27Tornelots of drive models come in more than one capacity
15:05:45Sudos|C840yeah, but usually manufacturers will have at least one or two numbers different to derive sizes. was just a surprise to me is all.
15:06:32 Join CaptainKewl [0] (
15:07:58***Saving seen data "./dancer.seen"
15:08:36Sudos|C840I bought this thing used off eBay like last week solely to use Rockbox on it - already I've found out a great deal of information on the device's life.
15:08:52 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:09:23Sudos|C840another reason to check is because the stock firmware and the chrome shell give off two different serial numbers from both different revisions of this particular generation model.
15:09:34Sudos|C840so, thanks for that.
15:10:25 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
15:12:51CIA-7New commit by wodz (r28304): MPIO HD200 - fix stupid typo introduced in r28270 which forced device hard shutdown during boot.
15:14:05 Quit kugel (Read error: Operation timed out)
15:14:33CIA-7r28304 build result: All green
15:17:28 Join webguest60 [0] (
15:18:44webguest60im trying to get my lrc files in the right directory so that lrcplayer can play them automatically when i run lrcplayer
15:18:57webguest60the directory specified in the manual doesnt seem to be correct, however
15:19:57webguest60anybody know what the correct directory should be?
15:20:01teruwhere did you put the lrc file?
15:21:29webguest60ignore the first backslash...
15:23:35webguest60i tried /lyrics/musics/artist/song.lrc after that
15:23:37webguest60didnt work either
15:25:47terumusic/artist/song part must be repalaced by actual path of the music file.
15:27:37webguest60basically i have to have the structure as my actual music folder, but just with a lyrics folder above it?
15:27:47teruif the full path of music file is /ABBA/Greatest Hits/Gimme! Gimme! Gimme!.mp3, then relevant lrc file will be /Lyrics/ABBA/Greatest Hits/Gimme! Gimme! Gimme!.lrc
15:28:00teruyou are right.
15:29:31webguest60argh, ok. so the manual is wrong when it says that it searches for /Lyrics/Musics/Artist/Title.ext?
15:30:25gevaertsWell, the manually doesn't really say that
15:30:33gevaerts"If the audio file currently playing is /Music/Artist/Album/Title.mp3, then the following files will be searched for..."
15:30:44gevaertsThat's not unimportant :)
15:31:13gevaertsBut yes, it could probably be clarified
15:31:14 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:31:41webguest60alright. thanks.
15:33:29 Quit webguest60 (Quit: CGI:IRC (EOF))
15:38:33 Join nufsed [0] (
15:39:08 Quit nufsed (Client Quit)
15:40:28 Quit ThomasAH (Read error: Operation timed out)
15:40:29 Join jgarvey [0] (
15:41:13 Join nerdy_kid [0] (
15:42:24nerdy_kidhello, I have a sansa e200v2 that has the issue described here: How does one downgrade the bootloader? I downloaded the 3.5.1 rockbox zip but dont see it. Thanks.
15:44:05nerdy_kidis the bootloader rockbox.sansa?
15:44:12 Quit teru (Quit: Quit)
15:45:00gevaertsThe bootloader is a separate download, it's not packaged in the main
15:45:34gevaertsIt should be somewhere on
15:46:08gevaertsPossibly you just want an older mkamsboot, I'm not entirely sure
15:46:34nerdy_kidgevaerts thanks, i had found that but it seems to be all current versions. I need an old one (at least the thread said I need an old one)
15:48:04gevaertsnerdy_kid: I'm pretty sure now that you actually can use a current bootloader, you just need to use an older mkamsboot, and those are there
15:49:08 Quit antil33t (Read error: Connection reset by peer)
15:49:15 Join ThomasAH [0] (
15:49:18 Join antil33t [0] (
15:49:20nerdy_kidgevaerts ok, now I have mkamsboot 1.4? (so i should get 1.3)
15:50:09gevaertsI think so, yes
15:50:21 Quit timccc (Read error: No route to host)
15:50:38nerdy_kidgevaerts thanks :) I'll give it a shot
15:51:39gevaertsnerdy_kid: oh, did you use rbutil to install, or did you use mkamsboot by hand?
15:52:40nerdy_kidgevaerts I used the install utility.
15:52:55nerdy_kidyeah the newbie way ;)
15:52:57gevaertsI suspect mkamsboot 1.4 will work fine as well
15:53:09gevaertsIt's just that rockbox utility is even newer...
15:54:05 Quit CaptainKewl (Read error: Operation timed out)
15:55:21 Quit ThomasAH (Quit: brb)
15:55:47 Join ThomasAH [0] (
16:02:03nerdy_kidgevaerts sweet that worked thanks. It wont mount in Linux though −− usb 2-2: device descriptor read/64, error -110 in syslogs
16:02:16 Join Strife89TX [0] (~cstrife89@
16:02:48gevaertsnerdy_kid: you're using the 3.6 release?
16:03:17gevaertsThat should reboot to the original firmware on USB connect I think
16:04:16nerdy_kidthat seems to be the only way to get it to work, but I was hoping to have rockbox's usb support working. It supposidly works for the e200v2.
16:04:37gevaertsyes, but not in 3.6. You need a current build for that
16:05:37nerdy_kidgevaerts ah ok. Well I will just wait till 3.7 then, I tried a recent build and it was busted bad. Thanks for all your help :)
16:06:35gevaertsWhat was the problem? FS #11608?
16:06:48gevaertsIf it was something else, we'd like to know...
16:08:43 Quit sasquatch (Quit: kernel update happy dance a.k.a. reboot)
16:08:53nerdy_kidgevaerts I am new with rockbox so I dont know the current bugs, but with the recent build I tried the theme was mangled −− empty boxes where buttons should've been −− quite nasty looking.
16:09:15gevaertsnerdy_kid: with the default theme?
16:09:48nerdy_kidgevaerts not with the default theme actually, but with every other one that i tried. all the others work fine on 3.6.
16:10:15gevaertsThe theme format changed after 3.6
16:10:44gevaertsSo unless you also use updated themes, you'll have problems
16:11:07nerdy_kidok that would explain it. so not a bug then.
16:11:35nerdy_kidwell; maybe I will try fishing for updated themes and update rockbox then.
16:11:53nerdy_kidjust one theme that i really like anyway. thanks for all your help!
16:12:54 Quit TheSeven (Ping timeout: 265 seconds)
16:13:19 Join sasquatch [0] (
16:13:54 Join swilde [0] (
16:15:23 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
16:15:50 Quit antil33t (Read error: Connection reset by peer)
16:16:00 Join antil33t [0] (
16:17:40 Join MethoS- [0] (~clemens@
16:18:59wodzwhat (in theory) is needed to port C++ app to rockbox?
16:20:54 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
16:22:11 Quit einhirn_ (Quit: Miranda IM! Smaller, Faster, Easier.
16:22:58Stummiwodz, hm. A big portion of masochism? ;)
16:23:34saratogac++ library stuff i guess
16:24:05saratogaBuschel: (for the logs) I want to ask you about the synth_full function in libmad which does the subband decoding
16:24:24 Quit nerdy_kid (Remote host closed the connection)
16:24:38saratogai have various questions about what its doing algorithmically, but what i want to do is rewrite it to use the ARMv5E and ARMv6 packed mul instructions to accelerate it
16:25:04saratogalooking at the pipelining on ARM11, i think we could probably get something like a factor of 2 speed up
16:28:26 Join swilde` [0] (
16:29:03kugelwodz: I guess it depends on how much the app depends on libstdc++
16:30:34 Quit chattr (Ping timeout: 245 seconds)
16:30:38linuxstbwodz: Converting to C is the best way... I did that with pacbox.
16:30:54 Quit swilde` (Client Quit)
16:31:23 Join swilde` [0] (
16:31:31 Quit swilde` (Remote host closed the connection)
16:32:27kugelsaratoga: that would be nice (2x speedup)
16:32:35wodzI was dreaming about scummvm on rockbox but I think it is out of my possibilities
16:32:52kugellibmad is dead slow compared to other codecs
16:33:09 Join belak [0] (~belak@subtle/user/belak)
16:33:19saratogamp3 is dead slow compared to other codecs
16:33:25saratogastupid mpeg jerks
16:34:11saratogabut i think it could be made a lot faster if ARM11 and Cortex versions of its functions were written
16:35:00kugelbtw, IIRC vlc has got some arm neon mdcts or something during gsoc
16:35:12kugelyou seem to be interested in neon
16:35:17 Quit wodz (Quit: Leaving)
16:35:18belakOk, I just flashed rockbox to my sansa fuze v1 and the main firmware will not boot up... so I can't connect with usb... any ideas? I can plug in the cord and it'll charge but then it gets stuck in rockbox rebooting over and over
16:36:19Stummibelak, do you the cable pluged in where the player is on or off?
16:36:32belakI've tried both
16:36:58gevaertsbelak: hold the appropriate button while booting to get to the original firmware
16:37:04Stummibelak, which build?
16:37:22belakStummi: 3.6
16:37:58saratogaeither update (so you get rockbox USB), or manually boot the sansa firmware before using USB
16:38:05Stummibelak, turn the player off, press and hold the left-button and turn it on while holding this button. Does this work?
16:38:17gevaertsbelak: the manual should tell you which button that is. Then either install a current build (instead of 3.7), or reinstall the bootloader following the manual installation procedure instead of rockbox utility
16:38:24belakTHANK YOU
16:38:27belakStummi: yes
16:38:43Stummihm, np
16:38:55gevaertssaratoga: I discovered now that the issue actually is that mkamsboot was updated, not the actual bootloader, and rbutil uses the new code while released mkamsboot doesn't...
16:38:58saratogakugel: yes neon looks cool, do you have a link for what they added?
16:39:05gevaertsSo it's a bit different than we assumed
16:39:11Stummihowever, this should not happen. USB-Support ist turned off in the build for the fuze
16:39:15saratogaheh weird
16:39:38Stummibut i only have sansa fuze v2.
16:39:47gevaertsStummi: it's enabled on current builds on the v1
16:40:02kugelno, I randomly stumbled upon this in a google result but I don't know what I searched for anymore
16:40:05belakgevaerts: well, not on 3.6
16:40:12gevaertsbelak: I know that :)
16:40:27saratogayeah 3.6 doesn't have it, mostly because we're too lazy to release 3.7 so the last release is quite old
16:41:06gevaertsWe should have 3.7 soon now though
16:41:20saratogadid anyone see the discussion last night about having a blog for the site
16:41:25Stummiyou know the magic sentence: "use current svn version" :)
16:41:38belakDidn't use the svn version...
16:41:45belakThought it would be too risky
16:41:56gevaertsStummi: yes, that's the one we want to get rid of
16:42:27Stummibelak, svn version is quite stable in most case
16:43:20saratogathe svn version is generally at least as good as the release version ;)
16:43:37belakUnless it doesn't work and bricks on firmware updae
16:43:39saratogasince we keep breaking the release versions!
16:47:10 Join toffe82 [0] (
16:47:12 Join _jhMikeS_ [0] (
16:47:13 Quit _jhMikeS_ (Changing host)
16:47:13 Join _jhMikeS_ [0] (~jethead71@rockbox/developer/jhMikeS)
16:47:13 Quit jhMikeS (Disconnected by services)
16:49:57 Quit MethoS- (Remote host closed the connection)
16:49:58 Join hebz0rl [0] (
16:50:30 Join chattr [0] (
16:54:34 Join kevku [0] (
16:55:02 Quit Sudos|C840 (Quit: class.)
16:58:16 Join MethoS- [0] (~clemens@
17:07:32 Quit kugel (Remote host closed the connection)
17:08:00***Saving seen data "./dancer.seen"
17:11:34 Join leavittx_ [0] (~lev@
17:12:20 Join _s1gma [0] (~d.d.derp@
17:17:35 Join timccc [0] (~tim@
17:19:34 Join fragilematter [0] (~fragilema@
17:22:41fragilematterhello everyone
17:22:56fragilemattersaratoga: ping
17:24:36 Quit belak (Quit: leaving)
17:24:56 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:27:36 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
17:28:53 Quit Topy44 (Quit: Leaving)
17:29:21 Join Topy44 [0] (
17:31:22saratogafragilematter: pong
17:31:48fragilematterI'm now rolling out a build with test_disk included
17:32:41fragilematterthe weird thing is:
17:33:16fragilematterwhen I run test_codec against an mp3 file, it appears to finish successfully, but I can't get the wav file to play
17:34:09saratogatry importing the wav file as a RAW file
17:34:13saratogait may just have a bad header
17:35:12 Join captainkewllllll [0] (2669ecc2@gateway/web/freenode/ip.
17:38:20fragilematterokay, I found no way to get vlc to import it as such, but I encoded it to ogg :))
17:38:51fragilematterand it does play back, with some minor clicking and skipping
17:39:21saratogabetter or worse then listening to it on the device?
17:39:39fragilematteron the device there's no skipping
17:40:03saratogaso its not the audio driver then
17:40:04fragilematterabout the clicking... I'll need to see how it performs on these more sensitive headphones
17:40:39fragilematterin the mean time I'm still compiling that build... slow processor :(
17:40:40saratogai usually just listen to the difference between a version decoded on PC and version decoded in rockbox
17:40:52saratogamakes it very easy to see if anything went wrong since you'll get silence normally
17:41:39fragilematterthe source mp3 is ok, I'll check it one more time when I hook up the sansa to dump the new build on it
17:41:51saratogadid test_disk work ok?
17:42:11 Part Zagor
17:42:35fragilematterit's still compiling :)
17:43:03saratogabe neat to know if the problem is the disk or memory
17:43:33saratogaif the codec loads at all some kind of memory corruption seems more likely
17:44:16fragilematterI still can't figure out why, at least on my player, the vorbis codec seems to be the only one triggering it
17:47:30 Quit thegeek_ (Ping timeout: 276 seconds)
17:50:08saratogamaybe some specific block of memory gets corrupted and that block is only really important for vorbis
17:51:19fragilematterokay, got the build on the player, also did an extra fsck which showed no errors
17:51:50fragilemattertest_disk -> write & verify?
17:52:40fragilematterstarted it, cpu unboosted
17:57:26fragilematternow comparing...
17:58:49 Join kevku [0] (
18:00:40fragilematterit passed
18:02:00fragilematterI've boosted the cpu and started another test run
18:05:43n1scould it have anything to do with the actual switching? vorbis is quite a bit slower than mp3 on amsv1 so it probably boosts more often than mp3 (if mp3 needs to boost at all)
18:07:11fragilemattern1s shouldn't a 48khz 320kbps mp3 make it more prone to misbehaving, should that be the case?
18:07:22saratogavorbis is faster then mp3 on amsv1
18:07:36saratogaand forcing boosting doesn't seem to make it better or worse
18:08:17fragilemattertest_disk passed when boosted too
18:08:18saratogaoh huh they're almost exactly the same speed
18:08:46n1sfragilematter: maybe, if that causes it to boost
18:08:51saratogathat looks wrong to me, something is probably slowing down vorbis since its faster on ARM9E and ARM7, it should be faster on ARM9 too
18:08:52 Join thegeek [0] (
18:09:13saratoganeither one will boost, the low clock on AMSv1 is way too high for that to ever happen
18:09:21n1ssaratoga: according to CodecPerformanceComparison mp3 is faster than vorbis but the mp3 numbers look weird since 320kbps is the fastest
18:09:34 Quit Strife89TX (Quit: I've been playing too much Dragon Quest IX.)
18:10:15n1sthe force boost point is very valid though
18:11:16saratogaall the mp3 results on the test codec page that are below 35mhz for AMSv1 are wrong
18:11:32saratogaits about 38-39MHz for both mp3 and vorbis
18:11:37saratoganot sure what happened with those results
18:12:28saratogatest_codec on amsv1 probably gives unrealistic results in general since the boosted performance will be much different then unboosted performance, and most codecs will never boost
18:12:43saratogato be accurate boosting should be disabled so that it tests the more realistic case
18:13:09n1syes, the beast has the same problem, it clocks up to the max freq when running test_codec now
18:13:26saratogaif it didn't take forever, i'd suggest decoding each file twice, once boosted and once unboosted
18:13:59saratogawe picked too long a test track
18:14:04saratoga30 seconds would have been a lot better
18:15:01n1salso we have too many different bitrates for most lossy codecs IMO
18:15:02saratogai guess test codec could be changed to have another option where it decodes just the first 30 seconds and does so unboosted
18:15:07saratogathat too
18:15:25saratogabut people are welcome to skip some bitrates if they like
18:15:59saratogai still think something is wrong with memory/clocking on AMSv1
18:16:05saratogamaybe this bug will shed some light on that problem
18:16:48fragilemattersaratoga: there may be something with switching, tho
18:16:59fragilematterI've played back an mp3, carefully listening
18:16:59saratogai don't think its related to switching
18:17:10saratogasince if you disable boosting the problem doesn't go away
18:17:19fragilematterduring the first few seconds, I can hear distinct clicking in the left channel
18:17:19n1ssaratoga: do you know if anyone disassembled the OF enough to know how it sets up clocks?
18:17:22fragilematterand then it stops
18:17:40fragilematterI'll try boosted to see if it makes any difference
18:17:42saratogan1s: funman looked at it in some detail, and another person on the mailing list dumped many of its registers via jtag
18:18:18saratogabut we could easily be missing something like the ipod nano 1G's "don't behave strangely above 64MHz bit"
18:19:11fragilematternope,same thing
18:19:13saratogai suspect something is wrong with the memory bus, since both the 1T internal DRAM and the external SDRAM have nearly identical performance, which suggests that something in between the memory and the L1 is stalling
18:22:07 Quit Dreamxtreme (Quit: Don't follow me)
18:23:20 Join Dreamxtreme [0] (~Dre@
18:24:28 Join stoffel [0] (
18:26:06 Quit n1s (Read error: Connection reset by peer)
18:26:10 Quit anewuser ()
18:26:13fragilemattersaratoga: mp3, as coped from device:
18:26:37fragilematterplayed back and recorded from the output:
18:26:50fragilematternotice the subtle clicking on the left channel at the beginning
18:27:16fragilemattertest codec on the same file, wav encoded to ogg q7 -
18:27:21fragilemattersame clicking
18:27:45fragilematterand test_codec on the ogg still results in a divide by zero at 30687778
18:28:48fragilematterand I've listened to that track so many times that I'm fed up with it, even though I downloaded it from jamendo two days ago
18:30:26 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:30:58 Join knittl [0] (~knittl@unaffiliated/knittl)
18:31:09knittlany server/repo admins here? you broke git-svn gateway again :D
18:31:45 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:40:38 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:40:38 Part LinusN
18:40:38kugelsaratoga: I think I tested iram&ram with test_mem.rock, iram was ~50% faster when boosted but the same when unboosted
18:50:06 Quit kugel (Quit: exit(0);)
18:50:17 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:50:40saratogathat doesn't really make sense since IRAM runs at PCLK and thus doesn't boost
18:50:46 Quit swilde (Remote host closed the connection)
18:51:19knittlhey saratoga, kugel: do you guys know a serveradmin with access to the svn repo?
18:51:33kugelknittl: what's wrong?
18:52:22kugelgit-svn is working for me
18:53:34kugelsaratoga: I'm fairly sure that were the results
18:53:58saratogayeah i don't doubt that
18:54:03saratogai just don't think it makes sense :)
18:54:14knittlkugel: yeah, but the git mirror is wrong
18:54:30knittlso trunk != origin/master
18:54:52kugelwhat is it then?
18:54:58knittlyou had that problem once already
18:55:10knittlit looks like an svn commit was edited
18:56:03kugelyou need to git svn reset to that edited commit locally (IIRC)
18:56:21knittli've reset master to remotes/git-svn
18:56:31knittlthe next pull will create a merge commit again
18:56:36knittl(trees are the same though)
18:56:50kugelit does fast-forward here so the mirror is not wrong I'd say
18:57:21knittlhm. strange.
18:58:23knittlgimme a sec
18:58:23kugelgit cherry master origin/master shows you the first different commit, you need to hard reset to that and svn reset to the corresponding svn revision
18:59:07knittli used merge-base ;) but it should be the same commit
18:59:30*kugel doesn't know about merge-base
18:59:45knittlbut i think master compares trees rather than commit ids
19:00:40knittl$ g rev-parse origin/master remotes/git-svn
19:02:20knittli have "Merge branch 'startfolder'" (64d420e83) in origin/master which is not in remotes/git-svn
19:02:51kugelthat commit has been edited yes
19:03:02knittlit's still in the git mirror
19:03:11knittlorigin/master does not lie ;)
19:03:28kugeland why does git svn rebase work just fine for me?
19:03:31knittlcan you run rev-parse on your side?
19:03:40knittlgit svn rebase works fine
19:03:52knittlbut git pull does not
19:04:00knittlwell, it work fine, but it creates bogus history
19:04:00kugelboth work fine for me
19:04:17knittlthe mirror is not a mirror but an alternate reality ^^
19:04:31kugelmy update command is "git pull −−ff-only && git svn rebase" which doesn't allow messing up the history
19:04:53knittli use git svn rebase first, coz otherwise you will pull history twice
19:05:08knittlcan you run git rev-parse origin/master remotes/git-svn
19:05:36kugel$ git rev-parse origin/master remotes/git-svn
19:05:42knittlwooot oO
19:06:06knittlha. i know why yours show both the same
19:06:17knittlyou rebased when the commit still was not edited
19:06:27knittlso you have the same version as the git mirror
19:06:30knittli only rebase from time to time
19:06:34kugelbut I do have the "Merge branch 'startfolder'" one in my master
19:06:43knittlkugel: yes, and that commit is wrong
19:07:00kugelI see
19:07:02knittlit should be "New setting to control the file browser start location."
19:07:50knittltry it. git svn reset 28200 && git svn rebase
19:07:57knittlyou will then have diverging heads
19:08:01***Saving seen data "./dancer.seen"
19:08:12kugelno thanks, I'll with that until Zagor confirms that he fixed it :)
19:08:30knittlah yeah zagor. i've walked him through fixing it the last time
19:08:34kugel(and that he hopefully disables post-editing ci messages again)
19:08:42knittlalthough i did not know about git svn reset. that will make everything _very_ easy
19:08:55knittlhow is that even possible with svn?
19:09:13knittli thought you needed svn-dump/admin/filter the repo for that?
19:09:37 Join benjamin__ [0] (
19:09:56knittlah ok. i'm happy i don't have to use 'real' svn any longer
19:10:20knittlwho needs repo integrity after all? :)
19:11:05 Join earcar [0] (
19:12:19knittlwhen is zargor usually around?
19:18:28AlexPHe's in Sweden, so on and off european daytime/evening
19:18:28knittl'zagor'. sorry. my timezone, good
19:18:28AlexPI only mention the spelling so that he sees that you want him :)
19:18:28knittlhe's not online, so hilighting won't work ^^ thanks anyway
19:18:53AlexPindeed, but there are logs
19:18:58AlexPAnd lots of devs scan them
19:19:10saratogajust because you don't see someone doesn't mean they aren't watching
19:24:18 Join Jerom [0] (~jerome@
19:24:25kugeldo we accept ruby files in our repo?
19:24:46AlexPI can't see why not if it is related :)
19:24:57AlexPNot if you just want to add random stuff you found :)
19:25:11kugelI've coded a somewhat better parse_testcodec than the one in svn but I've written it in ruby (which isn't as established as perl or python)
19:25:52 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:25:56knittldiaspora uses ruby xD
19:26:59knittljust saying
19:27:21rasherIf we have python already, I'd say ruby is acceptable as well. As long as it's just a peripheral script like that
19:30:03 Join Strife89TX [0] (~cstrife89@
19:35:46saratogabetter then my script ? ;)
19:35:49 Quit earcar (Quit: bye)
19:38:15pixelmahow should that work? ;)
19:38:53gevaertssaratoga: you mean you're *really* a bot? :)
19:41:16 Join drizztbsd_ [0] (~quassel@unaffiliated/drizztbsd)
19:42:26 Quit drizztbsd (Ping timeout: 260 seconds)
19:42:44 Join bmbl [0] (~bmbl@
19:42:44 Quit bmbl (Changing host)
19:42:44 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
19:43:25 Nick drizztbsd_ is now known as drizztbsd (~quassel@unaffiliated/drizztbsd)
19:46:57 Quit timccc (Ping timeout: 272 seconds)
19:49:11 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:51:47 Join earcar [0] (
19:53:49 Quit earcar (Client Quit)
19:55:54 Join s1gma_ [0] (~d.d.derp@
19:57:40 Quit _s1gma (Ping timeout: 245 seconds)
20:11:42 Join drizztbsd_ [0] (~quassel@unaffiliated/drizztbsd)
20:12:46 Quit drizztbsd (Ping timeout: 260 seconds)
20:12:49 Quit Xerion (Ping timeout: 252 seconds)
20:13:04 Nick drizztbsd_ is now known as drizztbsd (~quassel@unaffiliated/drizztbsd)
20:16:16 Quit stoffel (Ping timeout: 264 seconds)
20:16:55 Quit Jerom (Remote host closed the connection)
20:17:06 Quit Strife89TX (Quit: Blazer.)
20:20:04 Join Jerom [0] (~jerome@
20:20:17 Join Xerion [0] (
20:22:40 Quit DerPapst (Quit: Leaving.)
20:33:51 Quit Jerom (Quit: Leaving.)
20:34:57 Join wodz [0] (
20:36:26saratogasome kind of automatic release system might be neat, just have a script branch every 3 months, then we try and backport any major bug fixes, and then after a week or two, the branch becomes a release
20:36:51saratogaskip the feature freeze, do point releases after each release when bugs are fixed
20:38:41wodzare there more detiled instructions how to profile something in rockbox than
20:39:12saratogai don't think so
20:39:25saratogabasically everytime you want to profile you have to figure out what broke profiling this time
20:39:43saratogastripwax has done it recently though, so you could ask him
20:41:12wodzbut what is general scheme? 1) I have to have profile build and than what?
20:41:27 Join anewuser [0] (kvirc@unaffiliated/anewuser)
20:41:54saratogayou compile a profile build, then you add those two functions around the code you want to profile
20:42:13saratogaadditionally you'll probably want to disabling at least some inlines, because those will make it harder to profile
20:42:57wodzyou mean profile_thread() <some block of code> profstop() ?
20:43:13wodzdoes it work in sim too?
20:43:18saratogaaround then you realize it probably doesn't compile :)
20:43:31saratogai think so, but profiling on PC is probably not so useful
20:43:47wodzI mean only to get familiar with this
20:44:08saratogai can't think of any reason it shouldn't, but you'd have to try
20:44:56 Join retroj [0] (~retroj@unaffiliated/retroj)
20:45:42retrojhello. does rockbox provide any way to emulate symlinks on fat32? this would really make my day :)
20:46:04 Quit xnyhps (Ping timeout: 240 seconds)
20:46:53wodzfat32 does not support symlinks so no
20:47:18retrojhence my saying emulate.. a file type which rockbox would recognize as a redirect
20:47:59gevaertsretroj: depends on what exactly you want
20:48:33gevaertsWe have a shortcut feature that may be helpful. See the manual for details (I never used it...)
20:48:35*TheSeven wonders if there are any big differences between usb2 and usb1.1
20:48:54 Quit krazykit (Ping timeout: 265 seconds)
20:48:54gevaertsTheSeven: what sort of differences?
20:50:05retrojgevaerts: okay, thanks for the tip.. i'll see if that answers my question
20:50:05gevaertsretroj: it should at least answer your question. Whether they do what you want is something else though :)
20:50:05TheSevenno idea... my post-mortem memory dumper works on usb1.1 but not on usb2 when running from a panic handler, even though i wrote it for usb2-only
20:50:05retrojgevaerts: ha
20:50:05gevaertsTheSeven: packet size?
20:50:05TheSevenso i used 0x200 maxpackets everywhere and transfer sizes of up to 256K
20:50:18TheSevenexcept for the control ep of course which is at 0x40
20:50:27gevaerts512 byte packets shouldn't work on full speed...
20:50:32gevaertstransfer sizes don't matter
20:50:32TheSevenon usb1.1, everything works fine, even the 256K transfers
20:50:37retrojlooks like it is what i hoped for. thank you
20:50:45TheSeventransfer sizes do matter because i might overflow a register in the usb core
20:50:56TheSevenon usb2, enumerations fails
20:51:02gevaertswell, maybe, but usb doesn't care
20:51:22gevaertsah, so the host doesn't like your descriptors?
20:51:27 Join p3tur [0] (~petur@rockbox/developer/petur)
20:52:04gevaertsDo you have the output of lsusb -v of the device?
20:52:55TheSevenon usb1?
20:53:05TheSevenlet me obtain that...
20:53:06gevaertswell, both, if possible
20:53:28TheSevenwell, usb2 doesn't enumerate when booted through a panic. it works flawlessly when just uploaded as a firmware
20:53:51gevaertsWhat does "doesn't enumerate" mean though?
20:53:57 Join krazykit [0] (
20:55:34CIA-7New commit by kugel (r28305): Replace with a bit more powerful script. ...
20:56:28TheSevenusb1.1 box, booted through a panic:
20:57:13TheSevengevaerts: "doesn't enumerate" = windows doesn't even detect that something is plugged, linux fails with -110 for the first device descriptor and -71 for the retries
20:57:25CIA-7r28305 build result: All green
20:57:36TheSeventhe device in question is Bus 002 Device 004: ID ffff:a112
20:58:26TheSeventhe kernel doesn't complain about the maxpacket. dmesg output:
20:58:27TheSeven[ 86.924042] usb 2-2: new full speed USB device using uhci_hcd and address 4
20:58:27TheSeven[ 87.093154] usb 2-2: configuration #1 chosen from 1 choice
20:58:46gevaertsThe endpoint actually works?
20:59:06TheSeveni haven't verified the data read, but i can do a quick check
21:00:32wodzsaratoga: you ware right - profiled build doesn't compile :/
21:01:01gevaertsI'm surprised you can actually get 512 byte packets over a uhci controller
21:01:28TheSeventhe dumped data appears to be sane
21:03:02gevaertsAnyway, if you get a timeout on descriptor read, it would seem that your controller isn't properly initialised or something like that
21:03:42TheSeventhe fun fact is that this only happens if it both was booted through a panic and is connected to a usb2 device
21:04:26TheSevenif it's either connected to usb1 or booted directly without running rockbox before, it works just fine
21:04:46TheSevenand the usb initialization code is very similar to the code rockbox uses
21:05:48wodzhmm profiled target build do compile. Compiling profiled simulator fails
21:07:32TheSevendmesg from a usb2 linux box:
21:07:33TheSevenusb 1-10: new high speed USB device using ehci_hcd and address 6
21:07:34TheSevenusb 1-10: device descriptor read/all, error -110
21:07:34DBUGEnqueued KICK TheSeven
21:07:34TheSevenusb 1-10: new high speed USB device using ehci_hcd and address 7
21:07:34TheSevenusb 1-10: device descriptor read/64, error -71
21:08:04***Saving seen data "./dancer.seen"
21:08:44TheSevenuh oh
21:09:05TheSeveni just did a coldboot dump of the state after the stub was connected to usb2 windows or usb1 linux
21:09:25TheSevenin both dumps, the very kilobyte where the stub resides was trashed to all-0xff
21:09:38TheSeventhe stub is running with all caches disabled
21:09:51TheSevenso i wonder how it can work on usb1, and who dares to overwrite it
21:09:59gevaertsI didn't do it!
21:16:11TheSevenhm, the 0xff thing might be a coldboot problem, it seems to appear randomly, independently of the pc it was connected to
21:18:33 Part retroj ("Channel buffer killed")
21:22:33 Join DerPapst [0] (
21:24:28 Join b0hoon [0] (
21:26:59 Part fragilematter ("Leaving.")
21:30:56CIA-7New commit by b0hoon (r28306): HDD6330: Try to adapt the keys for the new touchpad code, remove the non existing buttons, enable morse input.
21:32:43CIA-7r28306 build result: All green
21:34:10saratogawhats the deal with the hdd6330
21:34:18saratogais it almost unstable?
21:35:18b0hoonsaratoga: i think it have to be in unusable state yet because of the lcd driver for example
21:35:30saratogawhats wrong with it?
21:36:20b0hoonit doesn't have power management at all and no lcd blit funcions
21:37:18kugelcan it play music without regular crashes or risk to brick it?
21:37:45b0hoonyes. there is no problem with the playback
21:37:55kugelthen it should be unstable IMO
21:38:07b0hoonand no risk of bricking it cause it's PP
21:38:26saratogamight as well bump it to unstable in hopes of getting other people interested in it
21:38:41b0hoonand there is no plugin keymaps
21:38:55kugelbad battery life or unimplemented features are a reason to not be stable but not for unusable
21:41:08b0hoonif so then maybe it is worth to jump with it to the unstable
21:42:45 Join slooopy [0] (
21:43:20saratogamaybe someone who wants to use plugins will add keymaps
21:44:04b0hoonyes, i must add the simulator for it...
21:46:11saratogai wonder about the Onda too
21:46:24saratogaas I understand it, the main thing missing is the internal memory, but its working with SD
21:47:06b0hoonI don't know much about Onda but i would like to have one :)
21:47:28b0hoonA lots of toys! :)
21:54:23CIA-7New commit by b0hoon (r28307): HDD6330: Patch to enable/disable backlight. ...
21:55:59CIA-7r28307 build result: All green
21:57:49 Join liar [0] (
22:00:33saratogab0hoon: whenever you want to make it unstable, i think you just need to update the front page via SVN, then ping one of the swedes with a bootloader and tell them to update the front page
22:00:44saratogai guess install directions on the wiki too if you haven't already
22:05:18 Quit bmbl (Quit: Verlassend)
22:06:01b0hoonsaratoga: ok thanks :)
22:06:31b0hoonaaa right - the bootloader... something is wrong with the OF loading yet
22:16:45kugelb0hoon: perhaps the backlight functions should be implemented in a seperate file if they completely ifdeffed
22:18:55 Join _s1gma [0] (~d.d.derp@
22:21:02b0hoonkugel: yes, i thought about separating HDD1630 from HDD6330 code, in fact these devices are more different than similar
22:21:15linuxstbsaratoga: Shouldn't that be in the reverse order? i.e. make the bootloader (and manual, if possible) available, then update the front page?
22:21:36wodzis it possible to derive .map file from .elf for plugin?
22:21:49 Quit s1gma_ (Ping timeout: 276 seconds)
22:22:11b0hoonlinuxstb: yes, that would be the right order i think
22:26:30linuxstbb0hoon: When you release a bootloader, you should build it with a proper version number (e.g. 1.0) and then tag it in svn. See
22:27:30b0hoonlinuxstb: yep, i remember it from the vibe port :)
22:28:10 Quit komputes (Remote host closed the connection)
22:28:29 Quit slooopy (Ping timeout: 265 seconds)
22:40:48wodzwell, this is definitely a bug - simulator doesn't compile with profiling 1) thread.c is not compiled where profile_thread() function is defined but thread-sdl.c instead 2) *.map files are not created when building simulator 3) something else is still wrong
22:51:21knittlhmm. zagor is still not here
22:51:40kugel1 is not a bug (except that profile_thread() should be defined in thread-sdl.c or in a seperate file maybe)
22:53:30 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
22:54:27 Quit kevku (Read error: Operation timed out)
22:56:00wodzkugel: I just described the symptoms
22:57:47 Quit ender` (Quit: drug, n: A substance that, injected into a rat, produces a scientific paper.)
23:00:20 Join krabador [0] (
23:08:05***Saving seen data "./dancer.seen"
23:08:19 Join Llorean1 [0] (
23:08:21 Join Zarggg_ [0] (
23:08:29 Quit p3tur (Quit: here today, gone tomorrow)
23:08:38 Quit Llorean1 (Client Quit)
23:09:41 Quit petur (Quit: Leaving)
23:09:50 Join drizztbsd_ [0] (~quassel@unaffiliated/drizztbsd)
23:09:53 Join kazaik [0] (
23:10:44 Quit wodz (Quit: Leaving)
23:11:54 Join tah [0] (
23:12:01 Join Barahir [0] (
23:12:06saratogathe theme site needs a download button
23:12:16saratogait took me too long to realize i need to click the picture
23:12:33 Join rasher_ [0] (
23:12:34 Quit rasher_ (Changing host)
23:12:34 Join rasher_ [0] (~rasher@rockbox/developer/rasher)
23:12:35 Join zu [0] (
23:13:29 Quit parafin (Disconnected by services)
23:13:30 Join CIA-107 [0] (~CIA@
23:13:35 Join parafin|away [0] (
23:14:04 Nick parafin|away is now known as parafin (
23:14:30 Join Jerom [0] (~jerome@
23:14:31 Join S_a_i_n_t [0] (S_a_i_n_t@
23:14:48saratogai grabbed a theme off the theme site, but it doesn't actually load with the latest build
23:15:01saratogais this a problem with the theme or is that expected if i'm somehow doing something wrong
23:15:42kugelsaratoga: there's a download button right below the picture
23:16:13saratogakugel: not for me, at least on this theme:
23:16:49kugelthe download button is on the overview, the page you linked is after you clicked on a picture the first time
23:16:51 Quit drizztbsd (*.net *.split)
23:16:51 Quit ThomasAH (*.net *.split)
23:16:51 Quit Barahir_ (*.net *.split)
23:16:51 Quit Zarggg (*.net *.split)
23:16:51 Quit JdGordon (*.net *.split)
23:16:51 Quit xavieran (*.net *.split)
23:16:51 Quit Llorean (*.net *.split)
23:16:52 Quit zu_ (*.net *.split)
23:16:52 Quit rasher (*.net *.split)
23:16:52 Quit CIA-7 (*.net *.split)
23:16:52 Nick rasher_ is now known as rasher (~rasher@rockbox/developer/rasher)
23:17:20saratogaah you're right
23:17:24 Quit domonoky (Read error: Connection reset by peer)
23:17:26saratogastill a download button there would be nice
23:20:31 Quit Jerom (Quit: Leaving.)
23:20:50kugelsaratoga: so you clicked on the picture ones already, clicking it another time can't be that unnatural :P
23:21:06saratogasomeone linked it to me asking why it doesn't work, so i didn't see the first page
23:21:37 Join Jerom [0] (~jerome@
23:22:21saratogahuh loading the theme in the sim does the same thing as the fuze, but it gives no debug output saying it failed
23:22:24saratogais that normal?
23:22:56 Quit b0hoon (Quit: GTG, bye.)
23:24:04 Join xavieran [0] (
23:24:40kugelit shouldn't behave differently
23:25:33saratogai thought if it can't load a theme you get an error explaining what didn't parse?
23:25:44S_a_i_n_tsaratoga: Debug output from the sim has been absent since the syntax/parser/skin-engine changeover.
23:25:53saratogaah ok
23:26:07S_a_i_n_tI've mentioned it once or thrice to JdGordon.
23:27:09S_a_i_n_tHe wants to get on-device debugging ifo going, and, what would indeed be cool but I've made it pretty clear I think simulator debug output is pretty important.
23:28:36saratogai wonder why the theme site says the file works
23:28:41saratogai thought we had automated checking
23:28:48 Quit Jerom (Quit: Leaving.)
23:29:00S_a_i_n_twhat's the issue with the theme...I could have a look if you want.
23:29:10saratogaits the one i linked above
23:29:11S_a_i_n_tAnd, the themesite is pretty braindead really.
23:29:14saratogait just doesn't load
23:29:22saratogano idea whats wrong, someone reported it
23:29:26S_a_i_n_tThe error checking is quite flawed, but works for the most part.
23:29:32 Nick drizztbsd_ is now known as drizztbsd (~quassel@unaffiliated/drizztbsd)
23:29:54S_a_i_n_tsaratoga: Which theme, you have a linky?
23:29:56 Join Jerom [0] (~jerome@
23:31:03S_a_i_n_tThanks, I'll have a look.
23:31:33 Quit Jerom (Client Quit)
23:32:28 Join Jerom [0] (~jerome@
23:34:19S_a_i_n_tsaratoga: Are you loading it via the theme .cfg?
23:34:30S_a_i_n_tif you "run" the .wps, it should load.
23:34:40saratogawhats wrong with the cfg?
23:34:47S_a_i_n_tthe theme .cfg is just warped...garbage.
23:35:01S_a_i_n_tthe entire thing is "ÿþ#"
23:35:15S_a_i_n_tso, it's not exactly going to load.....anything ;)
23:35:16saratogaits fine for me
23:35:20knittlhey guys. one thing: my clock is broken
23:35:25knittlfor a long time already
23:35:25S_a_i_n_tthe .cfg?
23:35:34knittlit does not display the last digit
23:35:45saratogayes, SIMPLE.cfg
23:35:54knittli only see 20:3
23:36:02S_a_i_n_tknittl: We need more info than that, what theme, what device, where is the clock you're talking about...etc.
23:36:02knittland then the symbol for "disk-access"?
23:36:07saratogaprobably a broken theme
23:36:15saratogaor a theme meant for a different font size
23:36:18knittlsansa e280v2
23:36:25knittlthe black and white theme
23:36:30S_a_i_n_tknittl: what theme?
23:36:55knittljust a sec
23:37:13gevaertshm checkwps segfaults on that theme for me...
23:37:26S_a_i_n_tsaratoga: You see the .cfg correctly? The theme editor and SCiTE are telling me it only consists of "ÿþ#"
23:37:39gevaertsand on others
23:37:45knittlgrml, now i switched themes and cannot find it again …
23:37:48saratogaprobably a busted zip program or download
23:37:53saratogaits just a normal text file
23:37:55S_a_i_n_tperhaps it downloaded incorrectly, but there's no errors in the archive. :/
23:38:08CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:38:08*S_a_i_n_t tries again.
23:38:23gevaertsS_a_i_n_t: it seems to start with a BOM I think
23:38:30gevaertsoh, right, it's UTF16
23:38:34gevaertsThat won't work well
23:39:01saratogaso just the file encoding is wrong?
23:39:57gevaertsS_a_i_n_t: yes, that seems to be it
23:40:23S_a_i_n_tsaratoga: Does the .wps load on device if you run the .wps file from the filebrowser?
23:40:41S_a_i_n_tthemeeditor says she's all good.
23:41:06saratogafixing the encoding didn't fix the problem
23:41:17knittlok, it's a cabbiev2 without backdrop
23:41:38knittland classic_statusbar breaks it
23:42:03gevaertscheckwps seems to be broken
23:42:08saratogayes the WPS seems to work fine
23:42:13saratogabut the cfg file does not
23:42:18S_a_i_n_tknittl: a screenshot would be nice, so I can see how far out the alignment is.
23:42:27S_a_i_n_tknittl: Do you know how to use screendump?
23:42:35knittlS_a_i_n_t: it's exactly one digit
23:42:46knittland now my wps is different :-/
23:42:50knittlhahaha, i fail
23:43:17knittli'll use rockbox_none as base skin, that seems to fix it
23:43:42saratogabah changing the encodings some more fixed it
23:43:45S_a_i_n_tknittl: I shall have a look at the problem later in the day, I'm a little busy right now but thanks for reporting it.
23:43:59knittldon't hurry :)
23:44:49saratogai wonder how this config file ever worked if the encoding wasn't supported
23:46:18S_a_i_n_tPass...maybe it works if the default codepage is changed?
23:46:41S_a_i_n_tthe themeeditor sure as hell doesn't know what to do with it, that's for certain.
23:46:54CIA-107New commit by gevaerts (r28308): checkwps: properly zero the wps struct to avoid interesting issues
23:48:03saratogaok reported a link to the fixed theme
23:48:07saratogahopefully someone will update that
23:48:32CIA-107r28308 build result: All green
23:48:47 Quit jgarvey (Quit: Leaving)
23:49:22gevaertsI guess the theme site code can handle utf16...
23:49:50S_a_i_n_twhwn it says "Simple" really means it, I was looking for the wpf folder with the additional images.
23:49:56S_a_i_n_tlols, facepalm. ;)
23:50:22S_a_i_n_t*when, *wps
23:50:54 Quit benedikt93 (Quit: Bye ;))
23:53:22SudosOut of curiousity, is there a way to add a battery capacity to the list of such in the settings menu in Rockbox?
23:53:34saratogayeah its in your devices's config file IIRC
23:53:49saratogaso /firmware/export/config, add it, and then compile
23:54:23Sudosheheh, it's been a while since I've had to compile- lemme see.
23:54:24S_a_i_n_tbattery capacity should already be in the settings, shouldn't it?
23:54:29Sudosit is
23:54:31S_a_i_n_tOr is it not enabled for all targets?
23:54:38Sudosbut for the iPod 3G it only goes up to 1200mAh
23:54:45SudosI have a 1300mAh I dropped in.
23:54:54 Quit captainkewllllll (Quit: Page closed)
23:54:59saratogato be clear, you realize that only changes runtime estimation right?
23:55:02gevaertsI wouldn't be surprised if the runtime calculation on 3G is off anyway :)
23:55:21Sudosyeah, it's a buggy yet relatively stable build, gevaerts
23:55:26saratogaand if you change the battery, the runtime estimation doesn't work anyway until you recalibrate it
23:55:35saratogaoh wait no that last part is wrong
23:55:41saratogai'm thinking of the battery meter, not runtime
23:56:02 Quit _s1gma (Quit: Leaving)
23:56:07S_a_i_n_tinstall the dynamic runtime estimation patch ;)
23:56:19gevaertsSudos: to be honest, for a an 8% difference, I wouldn't bother
23:56:31Sudosyeah, true
23:56:38gevaertsThe estimate is going to be off by more than that
23:56:54saratogayou can't do dynamic estimation on old ipods
23:57:06SudosI'm planning on doing a battery bench with it fully charged anyways, when it gets there
23:57:22Sudosjust popped the battery in, but of a snug fit but it fit in

Previous day | Next day