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

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

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

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

#rockbox log for 2010-11-04

00:02:15 Join JesusFreak316 [0] (
00:02:22 Quit domonoky (Read error: Connection reset by peer)
00:04:29 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
00:06:12 Quit JdGord (Quit: Bye)
00:06:54 Join csu [0] (~csu@
00:07:17 Join csu2 [0] (~csu@
00:07:17 Quit csu (Read error: Connection reset by peer)
00:07:56 Quit csu2 (Client Quit)
00:10:20 Quit csu1 (Ping timeout: 240 seconds)
00:12:45 Join DSStrife89 [0] (~Strife89@
00:13:37 Join soap [0] (~soap@rockbox/staff/soap)
00:14:52 Quit Jerom (Quit: Leaving.)
00:20:16wodz'core' manual for MPIO is ready. 'Only' plugins left...
00:20:25 Quit hebz0rl (Ping timeout: 265 seconds)
00:21:52 Part MItaly
00:24:00 Quit wodz (Quit: Leaving)
00:25:03 Join fritschy [0] (
00:25:45 Quit petur (Quit: Leaving)
00:31:33 Part toffe82
00:34:26 Quit bertrik (Quit: :tiuQ)
00:36:58***Saving seen data "./dancer.seen"
00:37:01 Join dragon [0] (
00:37:04 Nick dragon is now known as Guest88514 (
00:37:50Guest88514hello world
00:39:36robin0800Guest88514: I guess it works
00:42:13 Quit JesusFreak316 (Ping timeout: 250 seconds)
00:47:18 Join drizztbsd [0] (~quassel@unaffiliated/drizztbsd)
00:51:59 Quit n1s (Quit: Lämnar)
00:53:52 Part lixus00
00:56:01 Quit Guest88514 (Quit: Leaving)
01:00:48 Quit JdGordon (Ping timeout: 276 seconds)
01:03:42 Quit balintx (Ping timeout: 264 seconds)
01:11:38 Quit fritschy (Quit: Lost terminal)
01:12:09 Join krabador [0] (
01:12:23 Quit DSStrife89 (Ping timeout: 240 seconds)
01:12:35 Quit anewuser ()
01:12:50 Join fritschy [0] (
01:13:05 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
01:14:28 Quit robin0800 (Remote host closed the connection)
01:15:47 Join anewuser [0] (anewuser@unaffiliated/anewuser)
01:20:49 Join balintx [0] (
01:27:59 Quit xxcv ()
01:28:45krabadorno rockbox prototype i can use for unbrick a Philips GoGear SA3115/02?
01:29:00 Quit anewuser (Ping timeout: 240 seconds)
01:29:22 Join anewuser [0] (anewuser@unaffiliated/anewuser)
01:31:58 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
01:38:33 Quit anewuser (Ping timeout: 255 seconds)
01:40:28LambdaCalculus37krabador: Nope.
01:41:09LambdaCalculus37krabador: The only GoGear devices Rockbox runs on are the HDD16x0, HDD63x0, and SA9200.
01:50:13 Quit niekie (Read error: Operation timed out)
01:50:53 Join markun_ [0] (~markun@rockbox/developer/markun)
01:53:00 Quit markun (Read error: Connection reset by peer)
01:57:18 Quit fritschy (Quit: bbye)
02:10:25 Join qwd [0] (~x@unaffiliated/qwd)
02:13:45qwdCan I use several genre entries in Rockbox? That is, say I have two albums, one has the genres "pop" and "rock" and the other one has "rock" and "folk". If I look at the tag "rock" will I see both? I just realized mp3 doesn't handle more than one entry so with the regular ipod software I only get one genre in there.
02:15:43JdGordon|try it and see...
02:15:54JdGordon|it depends how the tag is stored
02:16:13JdGordon|I'd guess though that it wont work how you are hoping
02:16:43JdGordon|ah, yes, if the ipod OF shows only one then very likely our db will only show one also
02:19:33qwdJdGordon|: well, the ipod shows one because that's all there is on an mp3. I was thinking if I encode my music to ogg vorbis instead where I can have several separate genre entries.
02:20:08JdGordon|I don't tihnk that would work
02:25:09 Quit LambdaCalculus37 (Quit: Fwump)
02:25:45 Quit krabador (Quit: Sto andando via)
02:25:46 Quit simonrvn (Quit: ZNC -
02:26:22 Join simonrvn [0] (
02:27:39 Join anewuser [0] (anewuser@unaffiliated/anewuser)
02:28:45S_a_i_n_tAdded: FS #11717 - RaaA (Or Similarly Large Target) Iconset ( )
02:28:53S_a_i_n_tJdGordon|: \0/
02:30:21S_a_i_n_tI was unsure of the naming convention, you know where they're going and what they should be called I assume.
02:31:06JdGordon|looking good :)
02:31:12JdGordon|wanna do a 32pt one also :)
02:31:37S_a_i_n_tfrom what I understand, unless it's now bullshit, the iconset is limited to 24px
02:32:10S_a_i_n_tIf that is indeed untrue, then I should be able to knock out a 32px iconset also.
02:33:30JdGordon|yes and no
02:33:43*JdGordon| can change anything.. *evil laugh*
02:34:15S_a_i_n_t"<snip>Each individual icon can be any size up to 24x24 pixels and in full 16-bit colour.<snip>" <−−
02:34:26JdGordon|yes, artificial limitation
02:35:02S_a_i_n_tGood, good...24 should be sufficient for now...though there's some pretty high res targets/potential targets now.
02:35:25S_a_i_n_tI'll start on a 32px set, but it won;t be right up the top of my list if 'ya know what I mean ;)
02:35:33JdGordon|yeah, and these big screen targets have even less memory issues (which is hwy its 24pt max currently)
02:35:45S_a_i_n_tAh...I see.
02:36:14JdGordon|we could probably drop hat down even, I doubt 24 is actually needed on any current non app targst
02:36:37S_a_i_n_tNo iconset I'm aware of is close to it.
02:36:55S_a_i_n_tI have an 18x18 set for my Video builds...that's it.
02:37:01***Saving seen data "./dancer.seen"
02:37:08S_a_i_n_tthemers don't seem to make iconsets anymore.
02:37:16S_a_i_n_tI guess it's boring.
02:40:31 Join CaptainKewl [0] (
02:41:59 Quit anewuser (Ping timeout: 240 seconds)
02:46:54 Quit Staphylo (Ping timeout: 276 seconds)
02:56:54 Join JesusFreak316 [0] (
03:05:21 Join anewuser [0] (anewuser@unaffiliated/anewuser)
03:16:49qwdI can't write to my ipod harddrive after installing Rockbox and doing mount -o remount,rw /media/X doesn't help! (This is on Debian Squeeze AMD64 and an Ipod 5.5 80gb white with FAT)
03:23:46krazykitdoes your user have permissions to write to the device? presumably you mount as root and tried to write as user
03:24:18qwdkrazykit: can't write to it as root either
03:30:22qwdI'll try to reset it with itunes on windows
03:30:32krazykitor check your mount options and make sure it's rw
03:30:37krazykitdo an fsck.vfat
03:31:17qwdkrazykit: I tried that but I got some error message. I'll just try the itunes solution and see if that works :)
03:31:21 Quit qwd (Quit: leaving)
03:37:21 Quit kazaik (Quit: Leaving)
03:55:13 Join bluebrother [0] (
03:55:14 Quit bluebrother (Changing host)
03:55:14 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
03:58:16 Quit bluebroth3r (Ping timeout: 276 seconds)
03:58:20 Quit DerPapst (Quit: Leaving.)
04:00:07 Join togetic [0] (~togetic@unaffiliated/ibuffy)
04:09:27 Quit TheSeven (Ping timeout: 240 seconds)
04:13:46 Quit edboyer93 ()
04:14:00 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:18:51 Quit amiconn (Disconnected by services)
04:18:52 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:18:54 Quit pixelma (Disconnected by services)
04:18:56 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:18:58 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:19:12 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:28:18 Join fyrestorm [0] (
04:33:57 Join Barahir [0] (
04:35:18 Quit Barahir_ (Read error: Operation timed out)
04:36:53 Quit MethoS- (Read error: Connection reset by peer)
04:37:05***Saving seen data "./dancer.seen"
04:37:38 Join MethoS- [0] (~clemens@
04:38:46 Quit MethoS- (Remote host closed the connection)
04:47:06 Quit Judas_PhD (Quit: This is a quitting message)
05:04:43 Quit JesusFreak316 (Ping timeout: 245 seconds)
05:31:44 Quit ps-auxw (Read error: Operation timed out)
05:40:58 Join xxcv [0] (
05:41:24 Quit Horscht (Quit: Verlassend)
05:44:38 Quit JdGordon| (Ping timeout: 255 seconds)
05:44:44 Join JdGordon| [0] (
05:44:44 Quit JdGordon| (Changing host)
05:44:44 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
05:45:48 Join ps-auxw [0] (
05:46:48 Quit GodEater (Read error: Operation timed out)
05:46:54 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
05:54:48 Join monsterb [0] (~monsterb@
06:07:08 Quit anewuser ()
06:08:28 Join Judas_PhD [0] (
06:09:03 Quit xxcv ()
06:09:08 Join kugel [0] (
06:09:09 Quit kugel (Changing host)
06:09:09 Join kugel [0] (~kugel@rockbox/developer/kugel)
06:10:31 Quit scorche|sh (Ping timeout: 265 seconds)
06:10:53 Quit kugel (Read error: Connection reset by peer)
06:11:02 Join kugel [0] (
06:11:04 Quit kugel (Changing host)
06:11:04 Join kugel [0] (~kugel@rockbox/developer/kugel)
06:11:47kugelS_a_i_n_t: perhaps 36x36 rather tanned 32x32
06:12:12 Join scorche|sh [0] (
06:29:24 Quit kugel (Quit: Yaaic - Yet another Android IRC client -
06:35:06 Quit KiwiCam (Quit: Leaving)
06:37:07***Saving seen data "./dancer.seen"
06:44:08 Join soap__ [0] (
06:45:48 Quit soap (Ping timeout: 252 seconds)
06:49:10 Join KiwiCam [0] (
06:49:26 Quit rvvs89 (Ping timeout: 255 seconds)
06:50:55 Quit fyrestorm (Ping timeout: 260 seconds)
06:52:48 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
06:53:04KiwiCamHi peeps. Under Themes for Sansa Clip, my theme (uNbaLanCeD) is the only one where the pictures don't change, when you hover the mouse over them. My pic files end with .png - Should I use caps?
06:54:06 Join panni_ [0] (
06:54:52KiwiCamS_a_i_n_t: BTW. Thanks for your previous help. I've change my theme code. You probably would still have a fit at how it is ;) But, I'm happy with it.
06:57:53 Quit panni_ (Client Quit)
07:00:41 Join xxcv [0] (
07:10:50S_a_i_n_tkugel: (logs) Yes, I am indeed planning a second iconset now, much higher resolution than 24x24x16.
07:11:15S_a_i_n_tI was thinking along the lines of 36x36 also.
07:12:27JdGordon|36pt seems too big for the fontset we have.. isnt there only 2 or 3 fonts bigger than 35pt?
07:12:32JdGordon|32 sounds better
07:12:59 Join Buschel [0] (
07:13:41S_a_i_n_tWell, I was thinking about constructing a larger font also...but that is much, much further down the line ;)
07:15:46S_a_i_n_tThe problem with that will be *HUGE*
07:16:15S_a_i_n_tas, I'll be working with Unifont (for compatability), and that thing is already massive at 16pt
07:16:36S_a_i_n_tand to be of any real use, it will need to be about 40pt at least.
07:17:17S_a_i_n_tand there is *SO many chars in Unifont, it will take me a BLOODY long time to convert it by hand.
07:17:25LloreanAnd if antialiasing goes in, that'll multiply its size again.
07:17:37S_a_i_n_t(forgot to close [bold], oops.
07:18:13S_a_i_n_tLlorean: Yes, I had thought about that also...lets just say I have no desire for it to go in any time soon ;)
07:19:08S_a_i_n_tHowever, provisions exist to convert existing fonts to our format already. I am just not sure how good they are.
07:19:29LloreanWell, if you're converting from a mono bitmap font, they're perfect.
07:19:37S_a_i_n_tIt will almost certainly need cleaning up by hand but that is a lot easier than starting from scratch.
07:21:55 Quit CaptainKewl (Ping timeout: 240 seconds)
07:30:21JdGordon|we should just support vector fonts and be done with it! :p
07:31:03LloreanGet a decently fast vector rendering engine in place, then give people SVG to play with in the themes too. :-P
07:31:39S_a_i_n_tOh hush you!
07:31:54S_a_i_n_tMake me dream of cool things that no one wants to code ;p
07:38:16 Join esperegu [0] (~quassel@
07:47:29 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
07:51:02 Join stoffel [0] (
07:55:44 Quit BHSPitMonkey (Remote host closed the connection)
08:01:02 Quit JdGordon| (Quit: leaving)
08:02:07 Join kugel [0] (
08:02:09 Quit kugel (Changing host)
08:02:09 Join kugel [0] (~kugel@rockbox/developer/kugel)
08:04:48kugelS_a_i_n_t: I prefer 36 because a) icons commonly come multiple-of-12 sizes and b) even 35-Adobe seems to too small for some people so we might have a 38 or 40-adobe in the future
08:05:36 Quit merbanan (Ping timeout: 272 seconds)
08:05:39 Join JdGord [0] (~jd@
08:06:04 Join LinusN [0] (~linus@rockbox/developer/LinusN)
08:09:38 Quit stoffel (Remote host closed the connection)
08:14:59*S_a_i_n_t will see what happens
08:15:06 Join Drise [0] (
08:15:16S_a_i_n_t'tis in the future, after all.
08:16:03DriseHey S_a_i_n_t, I appreciate your help with compiling from source.
08:16:16DriseI got it working with the patch I wanted to give a go at.
08:16:30S_a_i_n_tDrise: No problems, were you wanting to compile SDL.dll also?
08:16:32DriseNow just to learn C and contribute, lol.
08:16:55kugelS_a_i_n_t: did you change the icons since you posted the last time?
08:17:02DriseHm... actually yes. I wanted to run the android sim, if there is one.
08:17:28S_a_i_n_tthere is an SDL app, but, not an "android sim"
08:17:52*JdGord wonders if dynamic screen size can reasonably be dependant on an application build?
08:18:49S_a_i_n_tDrise: Download the SDL source from here.
08:19:02kugelI would still find it better if its in sync with icons in svn, but I don't really care the direction of the sync (ie I'd be happy if the icons in svn are changed to look like your 24px version)
08:19:32S_a_i_n_tyes, that is more likely to happen that me changing the 24px version
08:19:51S_a_i_n_tthe current icons from the cabbie set represent their actions VERY poorly
08:20:05DriseS_a_i_n_t: How can i make an android sim, like the Fuze sim, for windows?
08:20:21S_a_i_n_tDrise: You can compile the SDL app
08:20:21 Join rvvs89 [0] (
08:20:41S_a_i_n_tDrise: Download the SDL source I linked.
08:21:01DriseI did not recieve a link..?
08:21:05kugelbut I think the rb symbol should be changed to show clef
08:21:38 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:21:40S_a_i_n_tkugel: I'd liek that too...but I didn;t feel like making it by hand.
08:22:23S_a_i_n_tthen extract it from the CygWin prompt whereever you'd like it (you can just delete the source afterward, you don't need it)
08:22:37kugelS_a_i_n_t: there is an svg for the clef. I think we even have a 24px png on svn
08:22:58S_a_i_n_tDrise: extract using "tar -xvf filename.tar"
08:23:01 Quit JdGord (Quit: Bye)
08:23:10S_a_i_n_tkugel: Ok, I'll look into it and post a revision
08:24:21S_a_i_n_tDriseL Let me know when you have downloaded the SDL source
08:24:30S_a_i_n_t* Drise: ^^
08:24:58Drisedonloaded and extracting
08:25:15S_a_i_n_tdon't extract it with a 7zip/whatever
08:25:23S_a_i_n_textract it with the commandline in cygwin
08:25:29Driseand done
08:25:33S_a_i_n_telse you'll get line ending troubles.
08:26:31S_a_i_n_tyou may need to change the gcc version you're using, can you paste the output from "gcc -v" for me please?
08:26:36S_a_i_n_tDrise: ^
08:27:06S_a_i_n_tOk, good.
08:27:28S_a_i_n_tcd to the directory "SDL-1.2.14"
08:27:39S_a_i_n_tthen do "./configure && make && make install"
08:28:13S_a_i_n_tafter this completes, you'll be able to compile the simulator and you can remove the SDL source.
08:28:31 Quit kugel (Quit: Yaaic - Yet another Android IRC client -
08:29:44DriseNot as long as the toolchains I hope, yes?
08:29:57DriseS_a_i_n_t: ^
08:30:19S_a_i_n_tto compile the SDL application, make a build folder in your rockbox source root, run "./tools/configure", then select 200 for "Application"
08:30:30S_a_i_n_tDrise: No, ~10mins or so to compile SDL
08:31:26 Join dys [0] (
08:32:46*Drise wishes the checking and such was done alphabetically to indicate progres...
08:34:27 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:35:11S_a_i_n_tDrise: What version of cygwin are you running?
08:35:45S_a_i_n_tit's interesting that gcc 3 came back from gcc -v, the default has been gcc 4 for some time now.
08:36:36S_a_i_n_tafter you've compiled this, its probably a good idea to update your CygWin installation by running setup again.
08:36:43DriseI thought I had 4.4.4
08:37:06S_a_i_n_tthat's just the gcc version for the toolchain
08:37:09***Saving seen data "./dancer.seen"
08:37:10S_a_i_n_tnot the native version.
08:37:39 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:37:45DriseWell S_a_i_n_t I'm sorry, but clss in 6 hours is telling me I nee to sleep...
08:38:03S_a_i_n_tNo worries, anytime.
08:38:04DriseAnd I'm getting to the point of missing keys when typing.
08:38:33 Quit Drise (Quit: Thanks again.)
08:44:49 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
08:51:01 Quit xxcv (Ping timeout: 252 seconds)
08:52:43 Quit Gabe_G23 (Read error: Connection reset by peer)
08:52:43 Part LinusN
08:53:39 Part Zagor
08:54:07 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:54:40 Quit n1s (Quit: Lämnar)
08:59:36 Join ender` [0] (
09:00:22JdGordonTrying to convert the skin code to use host malloc instead of internal... it does a massive amount of small allocations (and maybe 1/3 of the tags do nonstandard allocing also).
09:00:43JdGordonwould it be acceptable to just keep a list of all alloc()ed objects so they can be easily freed?
09:00:51JdGordonor better to always just alloc a single massive block?
09:01:04JdGordon(this would be for applicaiton builds only of course)
09:03:54JdGordonby massive I mean probably up to 10 per tag(!)
09:04:07JdGordonmost would be 2/tag
09:04:26 Quit TheSeven (Ping timeout: 255 seconds)
09:05:10 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:08:17 Join merbanan [0] (
09:20:33 Join Rob2223 [0] (
09:21:12 Join petur [0] (d408b802@rockbox/developer/petur)
09:24:47 Quit Rob2222 (Ping timeout: 272 seconds)
09:29:32 Quit bertrik (Ping timeout: 252 seconds)
09:29:38JdGordoni.e this...
09:31:08ZagorJdGordon: what is the overall goal?
09:31:45JdGordonruntime screen size handling instead of indivual build
09:32:25JdGordonthe skin buffer is a big hurdle for it
09:33:07JdGordonI want to do this anyway though because 1/4 of the RaaA audio buffer is stolen for the skin buffer anyway
09:33:48JdGordoncloser to 1/6 even.. still crap
09:43:33 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
10:04:09 Join xxcv [0] (
10:15:23 Join casainho [0] (
10:17:37 Join PurlingNayuki [0] (
10:19:06Stummihow can i just check, if a file exists?
10:22:33 Quit PurlingNayuki (Quit: CGI:IRC (Ping timeout))
10:22:40KiwiCamHey Folks. Just bumping my previous post. How can I get my screenshots to change when you hover the mouse over them? I've tried .png .PNG but neither seems to make a difference. Any ideas?
10:23:41 Quit ajb (Read error: Connection reset by peer)
10:24:00 Quit Judas_PhD (Quit: This is a quitting message)
10:24:24 Join ajb [0] (
10:24:36gevaertsKiwiCam: did you include both menu and wps screenshots?
10:25:10KiwiCamNot the Menu. But additionals. It's the Unbalanced theme.
10:27:29KiwiCamI didn't include the menu shot as it's just the default. If I add that, then the others will be displayed in sequence on hover?
10:27:40 Quit ajb (Read error: Connection reset by peer)
10:27:48 Join Judas_PhD [0] (
10:28:20 Join swilde [0] (
10:30:04 Join ajb [0] (
10:32:15gevaertsno idea actually
10:32:52gevaertsHowever, there's really no such thing as "just the default". Cabbiev2 may be the default theme, but it's still explicitely a theme
10:32:57 Quit Judas_PhD (Quit: This is a quitting message)
10:33:44 Join Judas_PhD [0] (
10:34:27 Join einhirn [0] (
10:35:31 Quit factor (Read error: Connection reset by peer)
10:35:59 Quit xxcv (Ping timeout: 245 seconds)
10:37:13***Saving seen data "./dancer.seen"
10:38:25KiwiCamgevaerts: Your suggestion worked. Thanks. I don't really want the Menu showing. But, at least the screenshots are working now. Greatly appreciated.
10:39:00*gevaerts doesn't really understand why showing the menu is so bad
10:40:02 Join DerPapst [0] (
10:40:08Stummiwhere can i find the "rb"-field, which is used in each plugin for all system calls?
10:40:39Stummiwant to read the source-file for this think.. Just for interesset, which methods it has at all
10:40:44 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
10:41:07 Join einhirn [0] (
10:41:21 Quit timccc (Ping timeout: 276 seconds)
10:41:30gevaertsStummi: apps/plugin.h
10:45:29KiwiCamgevaerts: I suppose I just don't want people thinking they can't change it. I see that really, I can just upload any .png file in place of the menu, so I don't really have to have it showing anyway, as long as there's SOMETHING there. :)
10:45:45Stummigevaerts, ah, found where i was looking for. Thanks :)
10:46:38gevaertsKiwiCam: why would people think that based on a screenshot?
10:48:26 Join timccc [0] (~tim@
10:48:42gevaertsMost regular users won't change anything, and advanced users know they can change whatever they want. Not showing the menu really is a disservice to the first group and no help to the second
10:49:10KiwiCamgevaerts: I just happy it's working now. Thanks.
10:52:09KiwiCamgevaerts: But the Menu is set as to how I like it. Not how the first group would necessarily see it on their player.
10:52:37 Join factor [0] (
10:53:03gevaertsOh, so it will be set to whatever the previous theme had? That sounds bad...
10:53:09*S_a_i_n_t is in the "if you didn't do anything with the theme to change the look of the screen, there's no point in adding a screenshot of said screen" camp.
10:53:34S_a_i_n_tusers will get the wrong idea.
10:53:50S_a_i_n_t"hey, I downloaded your theme but my menu looks different"
10:54:09gevaertsYes. If that happens, IMO the theme is buggy
10:54:25S_a_i_n_tnot necessarily.
10:54:31KiwiCamYeah. I've ONLY changed the WPS, no other screens. I've customised the Menu to my liking with the menu settings, but no coding changes.
10:55:01S_a_i_n_tthen, IMO, you shouldn't have a menu screenshot.
10:55:33gevaertsIt means that if you switch to a random other theme, and then back to yours, it will look different. That only serves to confuse *exactly* those people you claim to be trying to help
10:55:39gevaertsHow is that not buggy?
10:55:49KiwiCamI agree. That's why I didn't want to show it. I can put something else in it's place, but I should have the option of not putting any image there.
10:56:05S_a_i_n_tgevaerts: if a theme only changes one screen, it is by design.
10:56:05KiwiCamThe webpage is perhaps requiring a change.
10:56:22gevaertsKiwiCam: what does showing the menu screenshot have to do with my argument?
10:56:32S_a_i_n_tchanging one screen shouldn;t revert changes from other themes.
10:56:39gevaertsI'm talking about what happens on the player
10:56:55S_a_i_n_tas am I.
10:57:14S_a_i_n_tif the theme is only a .wps, why should it change anything else previous themes have set?
10:57:39S_a_i_n_t(barring the .wps of course)
10:57:39gevaertsS_a_i_n_t: because doing otherwise will confuse exactly those people you claim you don't want to confuse
10:58:10S_a_i_n_tIt may be confusing, but I think it is less correct to revert changes other themes have made.
10:58:11KiwiCamThen uploaders of themes need to be forced to set or determine other screen options.
10:58:41S_a_i_n_ta theme is either full, or partial.
10:58:54S_a_i_n_ta partial theme shouldn't overwrite any other themes changes.
10:59:01S_a_i_n_tonly the changes it needs for itself.
10:59:27S_a_i_n_tthat being said, I think partial themes are annoying ;)
11:00:47KiwiCamI could set the config to default to the standard screens (correct term?), but isn't the idea to allow other to choose what they want, or to mix and match?
11:01:00S_a_i_n_tKiwiCam: Correct.
11:01:19S_a_i_n_tI'd be more pissed off personally if it reverted all my other settings.
11:01:58S_a_i_n_tI think what you are doing is correct, but only if you remove screenshots of screens you have not modified.
11:02:13gevaertsPeople can still change whatever they want if you specify settings
11:02:43S_a_i_n_tthe screenshots not alternating with mourover is not a big deal, they will still see your alternate screens when selecting the theme to download it.
11:03:28S_a_i_n_tgevaerts: I just don;t think it is correct for the theme to default setting from other themes, when the "theme" as a whole is only intended as a .wps
11:03:34KiwiCamI see that the other themes uploaded only have perhaps two screenshots. I'm guessing that they upload using the first two slots ie WPS screenshot and Menu screenshot, but put non menu pics in the Menu screnshot places, so that they get the screenshot change on Mouse Hover.
11:04:13KiwiCam"but only if you remove screenshots of screens you have not modified" Yes.
11:04:42gevaertsS_a_i_n_t: rockbox doesn't really support having both "partial" themes and "full" themes. Unless you find a way to change that, I don't see any better solution than discouraging "partial" themes
11:05:10S_a_i_n_tgevaerts: very few themes as a whole are "full" themes.
11:05:14KiwiCamYeah. It's not a biggie. I was just jealous that the other themes could do that and I couldn't :(
11:06:25S_a_i_n_tif RB "doesn't support" partial themes, then those themes should be removed from the themesite that do not provide the full set of files, or a theme .cfg that reverts other screens to default.
11:06:32S_a_i_n_twhich, would be crap IMO
11:06:48KiwiCamI just think a webpage tweak would solve that menu screenshot requirement. ie If you don't upload a pic to it, it doesn't show it, but still rotates the other screenshots.
11:08:18gevaertsI find it *really* annoying to check out themes one by one, find one I like, use that, and then three weeks later when I see a new theme, try it out, decide I still prefer the old one, and then find I have to remember the exact order I was trying out themes in last time to recreate the desing I want
11:08:31S_a_i_n_tTo put it this way, if I have a theme set on my device and I knowingly download a "theme" that just consists of a .wps, then I expect only the .wps to be changed.
11:09:04S_a_i_n_tI would NOT expect to lose all my other changes, when it is so easy to revert the changes manually if it is needed.
11:11:09KiwiCamS_a_i_n_t: I agree.
11:12:21*gevaerts doesn't see how that helps anyone except those people who read the theme config files before loading them
11:12:57 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:13:01S_a_i_n_tgevaerts: It is the responsibility of the themer to provide accurate shots of the screens changed.
11:13:03KiwiCamThe easy work around is just to put something else in the Menu Screenshot place so that no menu shot is displayed. Anyway. I have to catch some Z's. Goodnight all.
11:13:16S_a_i_n_tif that is not done, it is the fault of the themer, not the theme.
11:14:04KiwiCamS_a_i_n_t: I agree again. Darn. Go to bed Cam!
11:14:09S_a_i_n_tpeople (if the themer has rightly included images) can see what they are getting at the time of download, I also include the changes to be expected in the themes description.
11:14:44S_a_i_n_tAny confusion that arises is a result of the themer, and not the theme IMO
11:15:36CIA-7New commit by 03jdgordon (r28480): RaaA: Use the host's malloc() for the skin engine.
11:17:25S_a_i_n_tanyone know where this is?
11:17:27S_a_i_n_t"[20:23] <kugel> S_a_i_n_t: there is an svg for the clef. I think we even have a 24px png on svn"
11:17:36CIA-7r28480 build result: All green
11:17:51gevaertssasquatch: People will assume that whatever they see after installing the theme is what the theme provides
11:17:58gevaertsS_a_i_n_t: ^
11:18:03*gevaerts apologises to sasquatch
11:18:13pixelmaS_a_i_n_t: I wouldn't know of such an SVG
11:18:21S_a_i_n_tgevaerts: well...this a flawed assumption.
11:18:24gevaertsS_a_i_n_t: moreover, in such cases they'll assume that the theme author forgot to include a menu screenshot
11:18:34gevaertsS_a_i_n_t: is it? Or is yours?
11:18:37S_a_i_n_tpeople should assume that the theme provides what it displays at the time of download.
11:19:13S_a_i_n_tI could pose the same "why?" to your assumption no?
11:20:01gevaertsAre you asking why people would assume that, or why I think they would?
11:20:21S_a_i_n_tIf I installed a theme, and afterward there was still parts of the previous it not ridiculous to assume that everything you see now is a part of the current theme?
11:20:31S_a_i_n_ts/current theme/theme you just loaded/
11:20:47S_a_i_n_tit would be more natural to assume that changes simply were not made to all screens.
11:20:53AlexPArguing about what people will assume is a bit silly, I can see people thinking both things
11:20:56gevaertsWhy would it be? People make derivative themes all the time
11:21:23AlexPS_a_i_n_t: e.g. the theme I mainly use (and ade) is just cabbiev2 with bigger fonts
11:21:26pixelmaS_a_i_n_t: there's an SVG of the complete logo (sans the metal structure on the "box" part). There should be an .ico file of the clef for the fav icon and it could contain vector data but I don't know and would have to look up where it is, my guess would be on the www or themes module
11:21:37gevaerts"Oh, it uses the same font as the previous theme. It must not set the font then"
11:21:44 Join n1s [0] (
11:21:44 Quit n1s (Changing host)
11:21:44 Join n1s [0] (~n1s@rockbox/developer/n1s)'s pointless arguing about it. it's not like anything is going to change unless you "outlaw" partial themes.
11:22:44S_a_i_n_tand I don;t see that happening.
11:22:51AlexPwhy not?
11:23:04S_a_i_n_tWhy would it now all of a sudden?
11:23:12S_a_i_n_tit;s been fine up until now.
11:23:12AlexPI'm not saying it should (I can see the argument both ways), but anything could happen depending on what the devs want
11:23:21AlexPS_a_i_n_t: Because nobody noticed?
11:23:38S_a_i_n_tI think its expected bahaviour personally.
11:23:43AlexPRight, personally
11:23:44S_a_i_n_tnot a case of no one noticing.
11:23:49gevaertsS_a_i_n_t: you've just been convincingly arguing that the current model is flawed. Why wouldn't we change it?
11:24:03AlexPAnd if it is you vs a dev who thinks otherwise (and cares enough), then it'll change
11:24:17S_a_i_n_tyou've been arguing it's flawed....I argue it is expected.
11:24:47gevaertsYou've *tried* to argue that, but the message I got from what you said is "it's flawed"
11:25:30sasquatch11:18:25 * | gevaerts apologises to sasquatch <<< no problem... and it didn't hurt... :-)
11:25:39AlexPsasquatch: You just wait :)
11:26:28sasquatchok −−- we'll see...
11:27:57KiwiCamAs a new user, I like the way that I can just create a WPS and leave the other screens alone, AND still be able to share what I've done. If I had to go through the hassle of creating all the other screens as well, then I probably wouldn't have gone as far as I have. I like the partial theme aspect.
11:28:23KiwiCamI thought I was going to bed?
11:29:15gevaertsYou don't have to *create* anything. There's this concept called "copying"
11:29:29AlexPMaybe it'd be worth if a screen hasn't been themed having the image for it on the themesite say something like "Menu screen will not be changed" or some such
11:30:09KiwiCamAnd the good news is, that now that I've had some experience and pleasure creating a WPS, I'm now enthused about changing the others. So, it's good for me as a beginner. It draws me in...
11:30:27S_a_i_n_tAlexP: That's kinda what my thoughts are, but IMO the "image" you speak of is implied by the fact that there isn't an image of screen X, so screen X does not change.
11:30:39AlexPImplying things like that is not good
11:30:47AlexPYou never know what people will think
11:30:48S_a_i_n_tperhaps making it ABSOLUTELY clear the themers must supply screenshots of changes?
11:30:56AlexPI would think that they forgot to upload the image
11:31:14AlexPIt needs to be explicit one way or the other
11:31:26S_a_i_n_tit should be quite easy to get the themesite to force this.
11:31:43S_a_i_n_tif an .fms is present then an FMS screenshot is NEEDED to upload, etc.
11:32:01S_a_i_n_t.wps == WPS screenshot, etc.
11:32:42gevaertsAnd .cfg == menu screenshot?
11:32:45AlexPThey should all be forced, if you haven't changed anything say so
11:32:56AlexPIt has to be explicit, not implied
11:33:06AlexPThe way one person thinks is different to another
11:34:27 Join soap [0] (~soap@rockbox/staff/soap)
11:34:33AlexPI tend to think that a theme should supply all files - if you haven't changed anything copy the default one in. If someone wants only the wps, they download the theme and set wps not theme through the menus
11:34:37S_a_i_n_tgevaerts: iconset == <anything other than default>, or font == <anything other than default> in the .cfg could quite easily == Menu Screenshot, yes.
11:34:38KiwiCamPerhaps just a bold note on the upload page stating that screenshots should be uploaded only if they've been changed? (And of course tweaking the hover thing ;) )
11:34:41AlexPBut I'm not that bothered
11:35:08AlexPKiwiCam: People don't read stuff no matter how bold it is
11:35:11*gevaerts still disagrees with the basic premise that a theme should be allowed to be incomplete
11:35:30S_a_i_n_tbut, they are incomplete by design.
11:35:32JdGordonwithout paying any attention to the topic, I agree, sort of
11:35:33S_a_i_n_tthat is my point.
11:35:42AlexPAnd I think uploading a random menu shot is not a good thing - it is innacurate, if you don't set it it could be anything depending on the previous theme they were using
11:35:44JdGordona theme is just a .cfg though
11:36:34KiwiCamAs a noob, I thought the Theme help files were very clear. I picked up on the partial's straight away.
11:36:37 Quit soap__ (Ping timeout: 240 seconds)
11:36:39AlexPyeah, so if someone only wants part of a theme they set that part through the menus, not the whole theme
11:37:14JdGordonwhy isnt the sdl app being built with -std=gnu99 ?
11:37:43n1shmm, scaling seems to be off
11:38:17*gevaerts is happy to know that JdGordon agrees. He only wishes he knew what with
11:38:21S_a_i_n_tpixelma: Thanks, sorry I missed your post...I was wondering why I was still being highlighted.
11:38:42S_a_i_n_tI really need (preferably an SVG) of the Rb Clef, not the "Rb" logo.
11:38:48JdGordongevaerts: that was to you, although i sort of take it back.. i'm staying out of this!
11:38:53S_a_i_n_tBut it looks like I'll have to do it by hand...uuurgh.
11:38:58gevaertsJdGordon: ok :)
11:39:24gevaertsS_a_i_n_t: then grab the svn logo and remove anything you don't want
11:39:28KiwiCamMust sleep. Thanks for the fun. Have a good one.
11:40:43gevaertsS_a_i_n_t: or do something advanced like e.g. look for svg files in the source tree
11:45:02pixelmagevaerts: it won't be a simple "remove everything you don't need" because in the logo the clef is positive/negative and you need to combine it to get one form again... possible of course
11:45:17pixelmaS_a_i_n_t: the logo SVG is somewhere in the manual part
11:45:34JdGordondoes anyone know the sdl lcd code?
11:46:55JdGordonwhere is the sdl window size setup? fiddling with hosted/sdl/lcd-bitmaps.c doesnt seem to fix it
11:47:59pixelmaaha, rbutil. Didn't know that. AFAIR this clef is a bit different though
11:48:55S_a_i_n_tpixelma: there is an .ico of the clef in RButil, and also the "Rb" logo (small) but the .ico won't scale well I don't think.
11:49:24bluebrotherS_a_i_n_t: Rockbox Utility also has an svg version of the "clef" icon
11:49:37pixelmaIIRC can contain vector data but don't have to
11:50:17pixelmabluebrother: didn't you squeeze it a bit?
11:52:12bluebrothersqueeze in what way? I've created an svg version of the clef, mostly based on the svg version of the logo itself with a few adjustments (small rotation IIRC) to make it fit better for an icon
11:53:06Stummihm, how can i just print something out to the console in simulator mode?
11:53:23bluebrotherStummi: DEBUGF()
11:54:54Stummii tried this already. Do i need a special compile flag or so, to get working this?
11:55:55*bluebrother isn't too familiar with that and thought it should just work
11:56:05 Quit casainho (Remote host closed the connection)
11:56:47Stummioh, yes it works. Seems to was my failure. Sorry :)
11:58:31pixelmabluebrother: it looks a bit rotated and wider (in aspect ratio) than in the icon but it's possible that it matches the favicon better. That's not very important though
11:59:23JdGordonboot time screen sizing mostly works :)
11:59:31JdGordon./rockbox −−display 640x480
11:59:38 Join Staphylo [0] (
11:59:42ZagorJdGordon: yay
11:59:58JdGordonneed to do some tricks to make the skins load the correct images though
12:00:24JdGordonthen someone else can fix the :)
12:01:32bluebrotherS_a_i_n_t, pixelma: I've also put some other variants online when I did that icon. See the svg files at
12:01:48S_a_i_n_tbluebrother: Thankyou.
12:02:02bluebrotherrockbox-clef.svg seems to be the unrotated version
12:02:17S_a_i_n_tThis is for the rockbox icon for the RaaA/large screen target iconset.
12:02:43bluebrotherif the screen is large enough you might even be able to use rockbox-icon.svg :)
12:03:03S_a_i_n_tI was previously using the "Rb" icon from RButil
12:03:30S_a_i_n_tbluebrother: Errr...I mean the iconset, 24x24 ;)
12:06:40bluebrotherthe svg should give you a much better quality ;-)
12:06:43Stummiwhat is the simplest way to check, if the volume is turned to null? (the entry at top of the volume-menu, not "0db")
12:08:53 Join casainho [0] (
12:10:22 Join lmjubuntu [0] (~lmjubuntu@2001:250:6801:4418:2ea:1ff:fe12:2816)
12:11:36 Join JG [0] (
12:12:35 Quit JG (Client Quit)
12:13:04 Join JG [0] (
12:14:06JGHi Just wondering if anyone can think of a way to get rockbox to use only lower half of the ipod screen?
12:14:14JGThe top half of my ipod screen doesn't work
12:14:35gevaertsYou could be creative with the UI viewport
12:14:51JGcan you advise how?
12:15:04JGI've only been a rockbox user not developer
12:16:03rasherWas the 3.7 release tagged?
12:16:09*gevaerts doesn't remember the details. He's sure someone else here does though
12:16:14gevaertsrasher: yes
12:17:07rasherSo it was.
12:17:49S_a_i_n_tJG: the manual can advise you of how to do this.
12:18:12S_a_i_n_tyou will still have to create a custom theme, as well as altering the UI viewport if you want all screens to work.
12:18:58 Part lmjubuntu
12:19:09JGI think I've found it
12:19:32JGCan someone point me how to install a theme I edit? I've only used the rockbox util
12:29:04JdGordonscroll engine is totally screwed up with this dynamic lcd size patch
12:35:23JGOk so I've set the view port options on every theme installed to ui viewport: 0,66,176,66,- no change in picture when I reboot, any tips?
12:35:57 Quit soap (Ping timeout: 255 seconds)
12:36:23JdGordonare you actually loading that theme?
12:36:50JGHow do I load a theme? Does this not happen on startup?
12:37:14***Saving seen data "./dancer.seen"
12:37:26JdGordonnot if you havnt got it configured to
12:37:42JdGordonhow are you setting the ui viewport?
12:37:43JGI changed all themes in the themes folder
12:39:10JGplugged in the ipod to a mac went to /Volumes/IPOD/.dropbox/themes and edited the text files there
12:39:41 Join PurlingNayuki [0] (
12:39:55JdGordonwhat did you change in which file
12:40:02PurlingNayukiDropbox cannot be visited in China.
12:41:35JG.dropbox appears to be where all the dropbox install goes onto the ipod.
12:42:27pixelmait's called rockbox though...
12:42:50JGno it has a . in the name to hide it from view in finder on an apple
12:43:23pixelmabut the project is called Rockbox :)
12:44:08JGoh I've reloaded the theme on the ipod itself. Hard without being able to see but it is now working...
12:44:20JGThanks so much! I can use the ipod again!!!
12:44:31pixelmayou could use a voice file
12:44:35JGRockbox ROCKS!!!
12:45:06JdGordonwhats about dropbox?
12:45:21*JdGordon got theme loading working on dynamic screen size builds :)
12:46:51S_a_i_n_twas "dropbox" just a mistake, or is there some alternate version of RB with .dropbox as the main folder?
12:47:46JGsorry yes mental aberation .rockbox not .dropbox
12:47:49JGha ha ha
12:47:54JGstupid typo
12:47:59JGone prob still though
12:48:16JGthe theme goes back to full screen on reboot and I have to go through the menus and load it.
12:48:29JGHow can I get it to hold my theme selection?
12:49:08S_a_i_n_t[00:49] <S_a_i_n_t> how are you rebooting?
12:49:09S_a_i_n_t[00:49] <S_a_i_n_t> are you using menu+select, or pressing play until it shuts down?
12:49:17 Quit PurlingNayuki (Quit: CGI:IRC (Ping timeout))
12:49:55S_a_i_n_tif shut down correctly, rockbox should store those settings in the theme .cfg
12:50:05S_a_i_n_ts/theme .cfg/.cfg/
12:50:09JGpressing play until it shuts down
12:50:23S_a_i_n_tthat is extremely odd then.
12:50:48S_a_i_n_tyour settings should be saved when you shut down.
12:51:36 Join xxcv [0] (
12:52:11JGmaybe a bug
12:52:31pixelmawhere exactly did you put your theme?
12:52:51JGI was on cabbiev2 and loaded cabbiev2 after changing the viewport. It did not hold after reboot
12:53:01JGChanged to rockboxed and rebooted. it stayed
12:53:13JGback to cabbiev2 and rebooted and now it holds
12:53:36pixelmacabbiev2 is made for the full screen
12:54:01JGit seems to work perfect with half screen after the viewport options are set.
12:54:10 Join lixus00 [0] (
12:54:21JGis there a theme you'd recommend for half screen?
12:54:35S_a_i_n_tone you create specifically
12:54:52S_a_i_n_tall themes are made for fullscreen.
12:55:49pixelmathe failsafe one probably works without trouble but doesn't look as nice
12:56:33 Quit xxcv (Ping timeout: 276 seconds)
12:57:40JGHave to say Cabbiev2 seems very usable in this config. Almost perfect.
12:58:04JGmust go do some work but thanks so so much!
12:59:03 Quit lixus00 (Quit: Leaving.)
12:59:32 Join csu [0] (~csu@
13:01:13 Join lixus00 [0] (
13:03:43JdGordonkugel: arg... you made dynamic screen sizing a bit painful on android.... FS #11615 ... I did say c should NOT be setting the dimensions..... :/
13:03:43 Quit JG (Quit: CGI:IRC (EOF))
13:15:47 Join PurlingNayuki [0] (
13:15:56PurlingNayukigiovanni_: I've changed the wiki. Please read and tell me if you're satisfied.
13:16:08 Quit casainho (Ping timeout: 252 seconds)
13:27:45stacker55I have updated the VirtualMachine for rockbox development from debian sarge to debian lenny. Is there someone i should send it to, or should i put it on the wiki?
13:30:18 Join LambdaCalculus37 [0] (~3f74f70d@rockbox/staff/LambdaCalculus37)
13:31:44LambdaCalculus37bluebrother: Ping
13:41:52 Part lixus00
13:48:01PurlingNayukiPatch is really unlovely...
13:49:15 Join teru [0] (
13:49:50 Join hebz0rl [0] (
13:50:21S_a_i_n_tPurlingNayuki: ?
13:56:23 Join Alchimysta [0] (
13:56:42 Quit Alchimysta (Client Quit)
13:56:56 Join Alchimysta [0] (
13:57:06AlchimystaCan you explain in more simple words this phrase? 2: add "ines" to $ ROCKBOX/apps/plugins/Subdirs2: add "ines" to $ Rockbox / apps / plugins / SUBDIRS
13:58:35S_a_i_n_tAlchimysta: edit those two files to include "ines".
14:02:39AlchimystaSUBDIRS What is it?
14:03:45S_a_i_n_tAlchimysta: If you don't have a development environment set up, with a working toolchain, and a copy of the rockbox source code checked out...don't bother.
14:03:46 Quit antil33t (Read error: Connection reset by peer)
14:03:53 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
14:03:57 Join antil33t [0] (
14:06:09AlchimystaI can compile the plugin and send it infones (MegaUpload)?
14:06:57AlchimystaYou tell me you can compile the plugin and send it infones (MegaUpload)?
14:09:37Alchimystayou can compile the plugin and send it infones (MegaUpload)?
14:10:46bluebrotherLambdaCalculus37: pong
14:12:14AlchimystaHelp me!!!!
14:13:46LambdaCalculus37bluebrother: Question about rbutil... I want to start adding support to install the builds and bootloaders for the GoGear devices to it. The install method is the same as most of the other PP targets (H10, m:robe, YH-9x0).
14:14:22S_a_i_n_tAlchimysta: No one will help you with that attitude
14:14:35S_a_i_n_tyou have already been told what you need to do, and what you need to do it.
14:15:18bluebrotherLambdaCalculus37: ok. You basically only need to edit rbutil.ini in that case.
14:16:06 Join drizztbsd_ [0] (~quassel@unaffiliated/drizztbsd)
14:16:11LambdaCalculus37bluebrother: Okay. Basically add the devices, URLs for download, and then test to make sure it installs correctly, right?
14:16:20LambdaCalculus37bluebrother: Or is there anything I need to know about?
14:16:46 Quit drizztbsd (Ping timeout: 272 seconds)
14:17:02bluebrotheryou need to change two places: the [platforms] section needs a new one that holds the section name of your device
14:17:33bluebrotherthen add a new section named after your device (i.e. the short name accepted by configure) the same way the other sections are.
14:18:06bluebrotherthe [platforms] section is used for sorting and for recognizing the available platforms so it's necessary.
14:18:31AlchimystaS_a_i_n_t: But I'm not able ... I was just wondering if anyone could do for me ...
14:18:54S_a_i_n_tAlchimysta: You *are* able, you are just not willing to learn.
14:19:16S_a_i_n_tthe resources are available, everyone else learnt the same way.
14:19:27 Join CaptainKewl [0] (
14:19:37LambdaCalculus37bluebrother: Ahh, I see how the [platforms] section is laid out... it's just platformXXX=<name>.
14:20:00LambdaCalculus37bluebrother: This looks pretty easy. Thanks for the rundown. :)
14:20:08AlchimystaI can not learn just because I understand little English and the translator sucks ... (It's not that I want)
14:22:18bluebrotherLambdaCalculus37: oh, if you have any pre / post installation hints you need to add it to the source code of the bootloader install method class. I.e. if bootloadermethod=mi4 you need to adjust base/bootloaderinstallmi4.cpp
14:22:45LambdaCalculus37bluebrother: Okay.
14:22:50bluebrotherthat text unfortunately is only handled in the code itself. I'm not too happy with that but I'm missing a better idea :)
14:23:10bluebrotherbut that should be it
14:24:00n1sdoes anyone know where i can find a 48kHz flac sample or something i can easily turn into one, wav perhaps?
14:24:22LambdaCalculus37Oh, SNAP... no one ever did device detection for either the HDD6330 or HDD1630:
14:26:01*LambdaCalculus37 has to remember to do that today
14:26:18rasherHm, bdf2bmp is giving me bdf2bmp.c:100: error: conflicting types for ‘getline’
14:26:25rasherBecause of stdio.h
14:26:37rasherIs there some proper way to handle this, or should we just rename getline?
14:28:00 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
14:29:19bluebrotherLambdaCalculus37: you can leave that out. Of course detecting it won't work then but that doesn't affect installation itself fortunately :)
14:30:49LambdaCalculus37bluebrother: Nah, I'd rather have them in. I can at least do the HDD6330 USB IDs now as I have it with me.
14:31:25LambdaCalculus37I still have to find my HDD1630 and get that one brought up to speed.
14:35:24StummiWhat is the best way, when I want a patch addet to the svn?
14:35:37n1sStummi: post in on FS and nag people here
14:36:11Stummiok, first step is already done :)
14:36:52 Join anewuser [0] (anewuser@unaffiliated/anewuser)
14:37:17***Saving seen data "./dancer.seen"
14:37:56LambdaCalculus37Stummi: Have you tested your patch to see if it builds okay on other platforms besides yours?
14:38:23StummiI testet the patch with the simulator
14:38:43 Join komputes [0] (~komputes@ubuntu/member/komputes)
14:39:13LambdaCalculus37Stummi: I mean besides the Fuzev2.
14:39:20StummiLambdaCalculus37, and i didn't changed something critical which could cause an crash on some devices
14:39:54StummiLambdaCalculus37, hm, does ist suffice to test the patch with other sim-builds?
14:40:27 Quit JdGordon (Ping timeout: 255 seconds)
14:43:35 Quit PurlingNayuki (Quit: CGI:IRC (Ping timeout))
14:51:27AlchimystaSomeone else who speaks Italian? "Cè qualcuno che parla italiano?"
14:51:44 Part LinusN
14:52:14 Quit antil33t (Read error: Connection reset by peer)
14:52:26 Join antil33t [0] (
14:53:08S_a_i_n_tAlchimysta: This channel is ENGLISH only
14:54:47LambdaCalculus37Stummi: That does help, but since the patch is in Flyspray, that should also give others a chance to test it.
14:55:10 Quit CaptainKewl (Ping timeout: 240 seconds)
14:55:41Stummiok. Would it be good to send a mail to rockbox-dev mailinglist to call for testers?
14:57:51Alchimystanot very good, if one does not speak English that you do not know how to talk ...
14:58:55Stummiyour english seems to be good enought to be unterstanded :)
15:00:15 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
15:02:28AlchimystaI use the translator ... : (
15:03:36 Join soap [0] (~soap@rockbox/staff/soap)
15:05:12stacker55the following location is where the zipped updated VM can be found:
15:05:33stacker55scorche: that might be interesting for you to look at
15:06:08LambdaCalculus37gevaerts: Regarding your post to the -dev ML, have you tried the most recent patch for FS #8806?
15:06:26stacker55the file itself is rather big: 8GB; so downloader beware
15:07:53LambdaCalculus37stacker55: Why is it so big? How much stuff did you add to the VM?
15:07:59gevaertsLambdaCalculus37: no. I'm only willing to spend time on it if it's worth the effort :)
15:08:45stacker55LambdaCalculus37: it's regular zip, not 7zip, and i had to increase the vm disk size. The updates needed more room than the original VM provided
15:09:10stacker55The VM itself has plenty of room inside
15:11:28 Quit csu (Ping timeout: 264 seconds)
15:14:32 Join LinusN [0] (~linus@rockbox/developer/LinusN)
15:14:33LambdaCalculus37gevaerts: I've yet to try the most recent patch out, but I think that we should get the task committed already. The most recent version is a codec version.
15:15:34LambdaCalculus37I'm going to try it out later on.
15:16:34stacker55i'm trying to get a smaller file by zero-filling unused clusters; see how well that works
15:17:51 Nick drizztbsd_ is now known as drizztbsd (~quassel@unaffiliated/drizztbsd)
15:18:35 Join csu [0] (~csu@
15:20:31 Quit Alchimysta (Quit: CGI:IRC (EOF))
15:22:55 Quit csu (Ping timeout: 255 seconds)
15:23:06 Join csu [0] (
15:28:17 Quit factor (Read error: Connection reset by peer)
15:29:36 Join Alchimysta [0] (
15:29:50 Quit Zarggg (Ping timeout: 240 seconds)
15:32:01 Join lixus00 [0] (
15:32:11 Part lixus00
15:32:58 Join factor [0] (
15:32:58 Quit Alchimysta (Client Quit)
15:44:46pixelmagevaerts: Rockbox already has a MOD codec and IIRC it can somehow deal with files larger than codec buffer but only plays the simple format modules.I once tried the MikMod patch (quite some time ago) and it wasn't able to play some of the ones that already worked, I'm willing to try again
15:45:08AlexPS_a_i_n_t: I think it is acceptable for him to ask if there are any Italian speakers around who can help him. If he finds one, then they can take it to a PM
15:45:34pixelmagevaerts: on the other hand it was able to play .it files etc. but I don't have many of these
15:47:41LambdaCalculus37pixelma: I've tried the mikmod codec patch several times and have been able to play all MOD files I have in my collection without issue.
15:48:06pixelmaand one "minus" about the current codec version - it doesn't display playback time correctly and IIRC you need to skip manually to advance in the playlist (need to try that out myself again)
15:48:39LambdaCalculus37pixelma: IIRC the current MOD codec doesn't display playback time correctly either.
15:48:42pixelmaI don't know how other chiptunes codec do, I believe there are similar problems with SID
15:48:51LambdaCalculus37pixelma: And NSF.
15:49:13pixelmaLambdaCalculus37: yes, I meant the current SVN codec version, sorry for not being clear
15:51:39 Join toffe82 [0] (
15:52:28stacker55LambdaCalculus37: replace .zip with .7z (in about 20 minutes it'll be online) and the file size will be 1/10th; so about 820MB for the VM.
15:53:56 Nick YPSY is now known as Ypsy (
15:54:04LambdaCalculus37pixelma: I can confirm that you do have to skip manually to advance in the playlist when playing MOD files.
15:55:09LambdaCalculus37pixelma: The playback time behavior is a little harder to explain, but basically it's behaving like it's displaying a "block" of the current MOD file.
15:55:21*LambdaCalculus37 doesn't have a good explanation for it
15:55:53StummiLambdaCalculus37, i sent a mail calling for testers for my patch to the dev-list now
15:55:54stacker55sratch that, it is online now.
15:56:06Stummilets see if i get some feedback :)
15:58:08 Join notsoap [0] (
15:59:18pixelmaLambdaCalculus37: yes, it displays the pattern number on place of the playback time. It's nice if you have "Caption Backlight" enabled
15:59:20 Quit soap (Ping timeout: 245 seconds)
15:59:40pixelmais that one seems to take the numbers there as seconds
16:00:42 Quit teru (Quit: Quit)
16:00:59LambdaCalculus37pixelma: And IIRC the mikmod patch has the same behavior.
16:04:25 Quit notsoap (Ping timeout: 276 seconds)
16:04:51evilnick_BStummi: It wasn't clear from the flyspray description, did you fix the bug in Pokemon?
16:05:30Stummievilnick_B, no
16:06:01Stummi(and i dont know something about an bug in polemon)
16:07:28Stummievilnick_B, can you point me to the bug, please?
16:09:57 Join lixus00 [0] (
16:17:19Stummiah, you mean the thing that it freezes, when you try to heal pokemon without sound?
16:17:45Stummihm, is there a reason for turning the sound off?
16:19:11Stummihow i said: Is there a special reason to turn the sound off (performance or so?). If not, you could turn sound on and set volume to null
16:22:36TorneStummi: Yes, it's faster if sound is disabled
16:22:51TorneStummi: but it breaks games which rely on sound interrupts
16:23:04Tornesince if you don't emulate the sound hardware you don't know when to trigger interrupts :)
16:26:28Tornesound is expensive to emulate on lots of consoles
16:31:39Stummiwow, didn't had thunk that
16:32:34Tornenot necessarily more expensive than, say, video, but you can't play most gmaes without video :)
16:33:20 Quit petur (Quit: reboot)
16:37:20***Saving seen data "./dancer.seen"
16:38:17 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
16:43:56LambdaCalculus37Something strikes my curiosity... since the Toshiba Gigabeat V is by and large the same hardware as the Gigabeast, has anyone tried seeing if the Gigabeast port runs on the V?
16:44:39toffe82I did it and it didn't work , but I should retry, it was a long time ago
16:45:37*Stummi just subscriped for cvs and sf-list too \o/
16:48:53 Part LinusN
16:50:23 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
16:51:48LambdaCalculus37toffe82: How much of the hardware is different between the S and the V?
16:52:55toffe82the lcd
16:53:21toffe82and no FM if I remember
16:53:22Tornedoes the V have the same signature checking bug?
16:53:37LambdaCalculus37Torne: IIRC it should. But toffe82's the man to ask.
16:53:59toffe82I don't know
16:54:03LambdaCalculus37Torne: And I wanted to find out if the T has the same bug as well, since the S, V, and T are very similar hardware-wise.
16:55:49pamaurydoes the pcm.h interface requires the data to be aligned in some way or did something changed in the last 6 months that could be considered as a non-minor change ?
16:55:58toffe82how can we look for the bug looking in the code ?
16:57:01LambdaCalculus37toffe82: I was trying to disassemble the Gigabeat T firmware to look for an exploit, but I'm no damn good at reading ARM assembly.
16:57:49 Quit user890104 ()
16:57:51LambdaCalculus37toffe82: But IIRC you, me, and saratoga had tried an experiment last year where I pushed a Gigabeat S firmware to my T via sendfirm, and it asked to restore.
16:58:21LambdaCalculus37toffe82: I think I also tried an nk.bin patched with the beast bootloader and that just left it in a reboot loop.
16:58:40gevaertsLambdaCalculus37: does that prove anything?
16:59:14Tornetoffe82: it's not very obvious
16:59:18gevaertsI mean, I flashed my bluetooth adapter with a meizu test bootloader, and it didn't ask for a restore, so it must have worked!
16:59:40LambdaCalculus37gevaerts: Nothing concrete, but I think that the digital signature bug may be on the T as well.
17:00:14gevaertsLambdaCalculus37: it may accept the thing for installing but not for booting
17:00:59LambdaCalculus37gevaerts: Yes, and IIRC someone had tried the same with the V as well.
17:10:04LambdaCalculus37Torne: If you want to take a look at the Gigabeat V and T firmwares, there are updaters on the wiki.
17:10:24Tornenot really :)
17:10:27Tornei have enough firmwares
17:12:47 Quit TheSeven (Ping timeout: 260 seconds)
17:14:55LambdaCalculus37Actually, the Gigabeat V updater isn't even on its wiki page.
17:16:26toffe82I have it but you can find it on the toshiba site
17:17:28 Join lixus001 [0] (
17:18:22LambdaCalculus37toffe82: Can you post a link to it on the wiki?
17:19:47toffe82LambdaCalculus37: I have no time now, can you do it :
17:19:58 Quit lixus00 (Ping timeout: 240 seconds)
17:20:04 Part Zagor
17:20:13LambdaCalculus37toffe82: Will do. :)
17:22:36 Join Strife89 [0] (a80dbf53@gateway/web/freenode/ip.
17:24:34LambdaCalculus37toffe82: Done.
17:25:20lixus001does beastpatcher work with v1.3 of tohsiba gigabeat s
17:28:08 Quit antil33t (Read error: Connection reset by peer)
17:28:17 Join antil33t [0] (
17:37:57 Quit ajb (Read error: Connection reset by peer)
17:46:25 Quit antil33t (Read error: Connection reset by peer)
17:46:34 Join antil33t [0] (
17:48:47 Join Alchimysta [0] (
17:49:14 Quit Alchimysta (Client Quit)
17:50:50 Quit Kitar|st (Ping timeout: 240 seconds)
17:55:58 Nick Ypsy is now known as YPSY (
17:56:15 Join Kitar|st [0] (~Kitarist@
18:04:09 Join kugel [0] (
18:04:10 Quit kugel (Changing host)
18:04:10 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:10:41 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
18:13:00 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:16:31 Part lixus001
18:19:34 Join panni_ [0] (
18:21:03Strife89 <−−- I would also like to see the plugin be commited soon, if and when it's patched up once more.
18:23:59 Join bertrik [0] (
18:24:00 Quit bertrik (Changing host)
18:24:00 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:24:14 Join Alchimysta [0] (
18:29:28 Quit Alchimysta (Quit: CGI:IRC (EOF))
18:30:34 Join Sarcastic_Dude [0] (
18:31:09Sarcastic_DudeWas some new functionality added to WPS
18:31:22Sarcastic_Dude's in the last release?*
18:33:43Sarcastic_DudeI just downloaded and installed 3.7 and my WPS no longer works...I assume something was changed in the layout of the WPS file?
18:33:55 Quit FOAD (Ping timeout: 265 seconds)
18:36:07 Join lixus001 [0] (
18:36:21 Part lixus001
18:37:03 Join tcm1998 [0] (~tcm1998@
18:37:24***Saving seen data "./dancer.seen"
18:37:53tcm1998I'm quite impressed by the rb3.7 stability on my nano 1G, compared to the 3.6 release :D
18:39:25tcm1998on the 3.6 release, I kept getting "undefined instructions", "data abort" or sometimes it just hang itself up
18:39:38tcm1998none of that so far on 3.7 (knocks on wood)
18:40:40tcm1998Very nice indeed, because stability is the most important thing on a player (IMHO)
18:40:55gevaertsoh, definitely
18:41:11gevaertsI can't remember any major issues in 3.6 though
18:41:17tcm1998I want to load it up with a playlist and not have to give it another thought until the workday is over.
18:41:40tcm1998Well, I never reported them (I've been using it for only a week now)
18:41:49Sarcastic_Dudeapparently the people who can answer my question are idle or not on IRC atm :(
18:41:52tcm1998well, maybe about 10 days
18:41:54AlexPSarcastic_Dude: The syntax changed]
18:42:04Sarcastic_Dudein what way?
18:42:15AlexPSee the manual for how the tags work now]
18:42:39Sarcastic_Dudeyou mean the CustomWPS page?
18:42:40AlexPand then check which of yours need changing
18:42:48AlexPno, the manual
18:42:49tcm1998well, I guess they weren't MAJOR issues, but just pretty annoying
18:42:56gevaertsI believe there's a wiki page describing the changes too
18:42:59Sarcastic_Dudeok thanks
18:43:23tcm1998and I was going to ask about it, but then I saw the new release and decided to go for that, first
18:43:38AlexPSarcastic_Dude: if it was from the theme site, most have been updated
18:43:50tcm1998I worked my way through the subversion log, but it didn't list anything to drastic
18:43:52Sarcastic_Dudeit was a personal one I wrote ages ago
18:44:12gevaertsSarcastic_Dude: there's also a conversion tool somewhere I think
18:44:19tcm1998unless it has something to do with threading and locking
18:44:48gevaertstcm1998: depends. Reproducible data aborts on a common player are pretty big issues, especially if you're talking about common codecs
18:45:14tcm1998unfortunately, it wasn't reproducable in any way
18:45:24tcm1998just happend every few hours
18:45:39tcm1998most of the time, the playlist was screwed up to, but not always
18:45:54AlexPSarcastic_Dude: I did mean the manual, but that wiki page should have been updated too
18:46:25tcm1998my personal suspicion was that it was because the player was doing more than one thing at the time.
18:46:34tcm1998e.g. updating the database while playing
18:46:38Sarcastic_DudeIs the syntax drastically different? It took the greater part of a day to write it the first time, I don't want to have to do it all over again...
18:46:40tcm1998or changing theme
18:46:47 Quit anewuser ()
18:46:54AlexPSarcastic_Dude: not hugely
18:47:04AlexPupdating should be pretty trivial
18:47:40Sarcastic_DudeOk, so I'll do that later. For the moment, I'll stick to 3.6...thanks AlexP
18:48:01rasherThe release is for the stable ports, right?
18:48:06gevaertsSarcastic_Dude: has a summary
18:48:56pixelmaSarcastic_Dude: there is a conversion tool in the source you could compile, somewhere in the forums there was a precompiled binary for Windows
18:49:44Sarcastic_Dudethat would be useful - you guys are very helpful, I appreciate it :)
18:50:00pixelmacalled skinupdater I think, maybe you can find it if you search for that
18:50:16 Join jgarvey [0] (
18:51:01 Join anewuser [0] (anewuser@unaffiliated/anewuser)
18:52:30Sarcastic_Dudegot to go...Rockbox is the greatest!
18:52:32 Quit Sarcastic_Dude (Quit: Sarcastic_Dude)
18:57:59 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
18:59:08tcm1998I'm off for now
18:59:11 Quit tcm1998 (Quit: Leaving)
19:02:23 Join user890104 [0] (
19:12:17kugelgevaerts: isn't the audio buffer used for playback in the mikmod patch=
19:12:51 Join Alchimysta [0] (
19:13:34 Quit Alchimysta (Client Quit)
19:19:53gevaertsWell, it calls plugin_get_audio_buffer()
19:20:22bertrikLambdaCalculus37, what was the deal again with the gogear players w.r.t. the tuner chip?
19:20:24 Quit DerPapst (Quit: Leaving.)
19:20:49bertrikThere is a single rockbox target that in some cases has an fm tuner and in other cases it has not?
19:21:19 Quit markun_ (Read error: Connection reset by peer)
19:21:32gevaertsIf you're talking about *real* single targets, there's pixelma's M5L
19:21:42LambdaCalculus37bertrik: There are several models of the 16x0 and the 63x0 players. There's the 1620 and 6320 which lack FM radio, and the 1630 and 6330 which have FM radio.
19:22:16pixelmabertrik: e200s and X5s
19:22:41LambdaCalculus37bertrik: And Gigabeast.
19:22:45pixelmahmm, maybe I misunderstood
19:23:02bertrikWhat I meant to ask is, do we really need runtime tuner detection for these targets, or can you just enable/disable it compile-time for those 16x0 and 63x0 players?
19:23:14 Join moos [0] (moos@rockbox/staff/moos)
19:23:18pixelmaI was just drawn into the discussion as gevaerts pinged me and I didn't read too much of backlog :\
19:23:28*gevaerts apologises :)
19:23:48bertrikpixelma, I wasn't very clear, it meant specifically the 16x0 and 63x0 targets
19:23:50LambdaCalculus37bertrik: Runtime tuner detection would be much easier because all we would have to offer is one unified build that will run on all of the same series.
19:24:48kugelgevaerts: I read that one comment, the real problem seems to be that mikmod needs to decode the entire file at loading time
19:24:48bertrikLambdaCalculus37, thanks for testing so far, btw :)
19:25:00kugelor it at least needs to load the complete file
19:25:09gevaertsyes, exactly
19:25:21pixelmaruntime tuner detection is there in general but probably has to be enabled for the GoGears then
19:25:30kugelthat is the real problem, not the memory limitations :)
19:26:21LambdaCalculus37bertrik: No problem. :)
19:26:26bertrikLambdaCalculus37, do you have a 63x0/16x0 target that could have a tea5767, but doesn't?
19:26:32gevaertskugel: The way I see it those are just two different ways to state the same problem
19:26:39kugelit does so apparently to convert 8bit to 16bit samples
19:26:57kugelI don't think loading the entire file is needed just for that
19:26:58LambdaCalculus37bertrik: I just have the one 6330 with me right now, and toffe82 is sending me a 1630 in the post soon.
19:27:17gevaertskugel: it needs to have those samples available at all times
19:27:21LambdaCalculus37bertrik: The only other GoGear I have with me right now is the SA9200, and that doesn't have a radio AFAIK.
19:27:32pixelmaI'm currently testing the latest mikmod codec version patch on my M5, works well so far if the file isn't larger than codec buffer which is rare at the moment with the buffer which got slighly larger when combining things (it's at 1MB or so IIRC?)
19:27:40 Join Buschel [0] (
19:27:57pixelmaI only found one .it file which was larger and it caused a freeze
19:28:00kugelgevaerts: ok, I haven't found the comment yet explaining that
19:28:05Buscheln1s: you there?
19:28:49LambdaCalculus37bertrik: I do have one question, though... how can you determine the FM radio hardware without opening the device up?
19:30:08Buscheln1s: based on your measurements the precision of the mpc patch looks fine. I have a question regarding the saturation of "D"-coefs though
19:30:16gevaertskugel: it's basically a bit similar to midi, where you have a (potentially huge) soundfont and a relatively small "tune description", except that with mod files the "soundfont" (which has a different name here, but I don't know it) is embedded in the file
19:30:38pixelmawhat I seem to remember is that the current SVN MOD codec could handle larger files somehow (I think back when I tried it before inclusion, codec buffer was half its current size and I had like 2...3 MODs that were larger and it played them. Unfortunately I can't remember for sure and if it is what was included)
19:31:23pixelmagevaerts: there are small samples in wave format IIRC, if you refer to the "instruments"
19:31:48kugelso perhaps adding audio_get_buffer() to the codec api could "solve" this?
19:31:53bertrikLambdaCalculus37, I guess you can't really be sure without opening up. We could probe for it in software, but that can only prove presence, not absence.
19:31:58gevaertskugel: audio_get_buffer() stops playback
19:32:29kugelbut the could can restart it, no?
19:32:41gevaertsthen buffering gets the buffer again
19:33:37kugelI mean it could restart it whatever way it does it in the plugin :)
19:33:49 Join Horscht [0] (
19:33:49 Quit Horscht (Changing host)
19:33:49 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:34:06 Join Alchimysta [0] (
19:34:31gevaertsBut then the playback system wouldn't know it's currently playing, leading to all sorts of funny business
19:35:26gevaertsI'd also expect the main codec thread to be stopped
19:36:01pixelmabtw. the latest mikmod codec doesn't loop so you advance to the next track automatically, depending on the MOD this may sound a bit weird but is quite convenient - and it allows for some seeking it seems. I don't want to lose codec support, especially for this "playlist" handling. Even if you would have to skip manually to the next track it's still better than chosing a new file everytime
19:36:20kugelgevaerts: it could do audio_get_buffer(), then buffer_alloc() the current file, and then restart playback (since buffer_alloc() moves the audio buffer front, or doesn't it do that yet?)
19:36:23gevaertsSo yes, if you want to dive deep into playback and buffering to make them work with seriously different requirements than they were designed for, sure, you can do that :)
19:36:51 Join markun [0] (
19:36:51 Quit markun (Changing host)
19:36:51 Join markun [0] (~markun@rockbox/developer/markun)
19:37:26gevaertskugel: yes, that would work, except for the possible nastyness of freeing the thing reliably again
19:37:33pixelmaI shall write this all to the ml too but doing something else currently and this is quicker :|
19:37:34n1sBuschel: yep
19:37:45gevaertspixelma: no hurry
19:38:01kugelhow does the plugin version handle this?
19:38:46gevaertsIt can just grab the audio buffer until it exits
19:38:53kugelif I see this right it uses malloc. if that's true I expect it to also fail after a few files
19:38:55Buscheln1s: great :) your patch looks fine to me −− especially with the minor output diffs to the reference.
19:39:20 Join franky [0] (
19:39:41Buscheln1s: a question out of interest -> why do you need to saturate "D"-coefs? Is that because of using the topmost 32 bits of the 64 bit result?
19:40:33n1sBuschel: how do you mean saturate?
19:40:34 Quit Alchimysta (Quit: CGI:IRC (EOF))
19:41:01Buscheln1s: you do <<14 to saturate them to a range of +/- 2^31
19:41:04gevaertsI'm not sure about e.g. the case where you'd play a mod file which "stops" playback, buffer_alloc()s the necessary space (which it technically can't, codecs aren't supposed to run when playback is stopped :), then starts playback again, then the user picks a theme with differently-sized album art, which stops playback and buffer_alloc()s a bit and restarts, then the mod file ends. What happens now?
19:41:59kugelit technically can because it's the codec's responsibility to exit on playback stop (if it doesn't do that it continues to run) IIRC
19:42:06gevaertsI really don't think trying to shoehorn that sort of thing in the codec version is a good idea
19:42:24kugelchanging themes involves buffer alloc?
19:42:38gevaertshm, no, wait
19:42:51gevaertsNo, it "just" needs a rebuffer I think
19:42:57n1sdo you mean the <<14 shift?
19:43:16kugelyes, album size changing is a rebuffer only
19:43:16CIA-7New commit by 03moos (r28481): Update the russian translation. ...
19:43:19CIA-7New commit by 03moos (r28482): Update the russian translation. ...
19:44:07 Part franky
19:44:25kugelperhaps we should have a heap for codecs. it bugs me a bit that codecs can use malloc and we can't estimate a suitable codec buffer size anymore
19:44:36gevaertsAnyway, I think we should choose between the codec version with size limitations, and the plugin version without, or possibly have both, but not try to do fancy stuff with buffering. That's already complex enough as it is
19:44:50gevaertscodecs have a heap
19:45:25kugelyes, whatever space is left in the codec buffer
19:45:36CIA-7r28481 build result: All green
19:45:57gevaertswell yes. How else would you handle that? Have an extra fixed chunk somewhere?
19:46:03kugelI mean a heap on the audio buffer, then we could adjust the codec buffer size down since we the know the size again
19:46:03pixelmaI think the current MOD codec for the simple modules and mikmod as a plugin for the more complex formats like .it would be a good compromise
19:46:04gevaertsWe used to have that...
19:46:30gevaertskugel: well, not what you just described
19:46:47gevaertsBut again you're making buffering a lot more complex
19:47:25kugelI think it would be nicer if the first codec that needs a heap makes a codec api call (request_heap()), then the core buffer_allocs() a fixed size heap for all codecs
19:47:41kugelmy idea is a bit separate from the mikmod patch
19:47:41gevaertsWhat if the next codec needs more?
19:47:52 Join fyrestorm [0] (
19:48:00gevaertsAnd how do you maintain gapless playback?
19:48:01kugelthe same thing we do now, tell it to shut up :)
19:48:22gevaertsSo it's not guaranteed to be available? Then how does that help?
19:48:34kugelthe codecs that need a heap would request the heap on the first run, before the first decoding is ever done
19:49:09 Join bmbl [0] (
19:49:12 Quit bmbl (Changing host)
19:49:12 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
19:49:29 Join webguest373 [0] (
19:49:41kugelit'd guaranteed to be made available if there isn't already a heap, just growing wouldn't be supported
19:49:42 Quit webguest373 (Client Quit)
19:50:02gevaertsSo if a codec needs more, you skip that track?
19:50:06 Join Horschti [0] (~Horscht@xbmc/user/horscht)
19:50:21kugelit would help since we can shrink the codec buffer to a suitable size (e.g. the biggest codec)
19:50:24n1sBuschel: sorry, my irc lags as mad, yes i do the shift just as coldfire and the regular c because the smmul/smmla instrs do a 32*32=64>>32 multiply
19:50:38kugelgevaerts: yes, that's what vorbis is doing currently
19:50:58gevaertsFoe very few files
19:51:14kugelit would be the same with my idea
19:51:16gevaertsWhat you're proposing will make some files play and others not, depending on playlist order
19:51:30kugelhow so?
19:51:47 Quit t0rc (Ping timeout: 276 seconds)
19:52:13gevaertsThe only way that I can see to avoid that would be to run test_codec on all files on the disk and remember the largest heap used
19:52:20gevaertsBut that would slow down booting
19:52:24Buscheln1s: ok. I am asking because the mul's should take less cycles if the operand is smaller. so, did you make the experiment to use a smaller shift (e.g. <<6) and post-scale the result (e.g. by another <<6)? Does this have direct influence on the output of the decoder?
19:52:27 Quit Horscht (Ping timeout: 272 seconds)
19:53:09Buscheln1s: nevertheless the patch looks fine so far. would be good, if you would add a comment to the <<14 shift why this is done (e.g. referencing to the 32*32>>32 multiplication)
19:53:12kugelit's not supposed to be big enough for all files, it would be as big as vorbis' effective heap right now, so nothing would change
19:53:23gevaertsSo what's the point?
19:54:12kugelshrinking the codec buffer, potentially not having a heap at all if no codecs need it (memory saving)
19:54:29kugeland knowing the max. size of codecs again
19:54:46gevaertsBut some codecs *do* need it, and you don't know in advance whether or not you'll load those
19:55:01gevaertsAnd how does it tell you that maximum size anyway?
19:55:05kugelI had this desire the other day when playing around (in my mind) with relocatable plugins and a unified codec and plugin buffer
19:55:13n1sBuschel: the arm11 multiplier doesn't have early terminate afaik so i don't think the size of the operands matter there
19:55:50gevaertsSure, you'll know the size of the codec itself, but you still don't know the maximum size of the heap, and you have to allocate it anyway
19:56:07n1sthat's at least according to the system developer's guide (it mentions that arm7tdmi and arm9tdmi multipliers have early termination but arm9e doesn't)
19:56:09kugelgevaerts: the first codec that needs it (that's simply the first codec that has malloc compiled it, i.e. not depending on the files) allocates the heap
19:56:23gevaertsSo at that point you stop playback and rebuffer?
19:56:32kugelyes but as long as the heap is in the codec buffer, plugin and codec buffer can for sure not be unified :)
19:57:01 Join franky [0] (
19:57:06gevaertsSorry, but I don't think that that's at all acceptable
19:57:12 Quit n1s (Read error: Connection timed out)
19:57:16kugelit would be ok since it only happens once, if at all
19:57:36gevaertsA three second gap somewhere halfway your playlist is ok?
19:57:49AlexPIt seems something that would lead people to think playback is broken to me
19:57:56gevaertsAlexP: *think*?
19:58:02gevaertsIt *would* be broken
19:58:08AlexPIf it occasionally randomly stopped and restarted
19:58:14AlexPgevaerts: I mean not by design
19:58:14kugel3 seconds? on my targets playback starts immediately, before buffering is finished. might be worse on hdd though
19:58:56AlexPdisks take a few seconds to spin up
19:59:06gevaertsYou need to spin up the disk before you can even start buffering
19:59:22Buscheln1s: seems you're right. just skimmed through some arm11 documentation -> so, comment the <<14 shift and then commit :)
20:00:35 Join nls_web [0] (
20:00:36kugelthen do the heap unconditionally. it's basically also allocated unconditionally currently
20:00:51gevaertsYes, but then you haven't changed anything
20:00:56nls_webBuschel: I'll do that and commit tomorrow
20:01:27kugelit would enable unifying codec&plugin buffer at the very least
20:01:49gevaertsI'm doubtful about that
20:02:35Buschelnls_web: ok. will you also add the instr-saving change to the ARMv4-patch that you mentioned?
20:03:32gevaertsYou could either make that unified plugin the combined size of both of them, which again isn't very useful on its own, or make it smaller and then have the interesting situation that you can't play certain files if specific plugins are running, or vice versa
20:05:41 Join DerPapst [0] (
20:06:36kugelit would be smaller yes, since to codec is actually bigger than 512k
20:07:31gevaertsIt would be smaller by exactly the size of this heap that you've now just allocated elsewhere
20:07:46kugelyes :9
20:07:57gevaertsSo again, *why*?
20:08:40kugelit could be even smaller, we could test the worst case combination, add a small gap and be done
20:09:18gevaertsAnd what do you do with plugins that grab the remaining bits of the plugin buffer?
20:09:56kugelthat's a case a I haven't thought about yet ;)
20:10:30kugelI'm just playing with the idea once in a while, I'm not claiming I have a definite plan
20:10:34gevaerts"Oh, you can use the text viewer while playing wav just fine. Playback will skip ogg files though"
20:12:03gevaertsReducing heap size and making it predictable are excellent goals. Making playback stop working depending on the exact combination of codecs used, running plugins, and the phase of the moon is not
20:13:51 Join TheLemonMan [0] (
20:15:12nls_webaren't we already in this situation with doom on the 8MB targets? some parts usually work, fewer work if you use dircache and fewer still if you have tagcache loaded to ram and probably fewer still with max playlist and filebrowser sizes
20:15:55nls_webit's not nice, indeed and i agree we shouldn't go this way with core stuff
20:16:33kugeldoom would actually work in the unified buffer
20:16:59gevaertshm, there are other things we could do...
20:17:04*gevaerts checks some stuff
20:17:56gevaertsDoom could actually use the codec buffer right now
20:18:26kugelit doesn't do this already?
20:18:33kugelph the codec buffer
20:18:41kugelwell, it's not accesible for plugins is it?
20:18:57gevaertsit's not supported no
20:19:24gevaertsbut I don't think there's a technical reason why a plugin that grabs the audio buffer shouldn't be allowed to get the codec buffer as well
20:19:42gevaertsIt's not necessarily that simple though
20:19:55gevaertsIsn't the codec buffer in iram on some AMS targets?
20:20:07nls_webyes, on the 2MB one(s)
20:20:14gevaertsok, so not on any fuze?
20:21:13gevaertshm, according to my reading the fuzev2 has that
20:21:47nls_webyes, seems so
20:21:56gevaertsWhich means that the codec buffer isn't adjacent to the audio buffer, which might make malloc() harder to get working properly
20:22:32gevaertsStill not impossible though
20:23:25*kugel wonders if doom's malloc is any better than tlsf
20:25:03gevaertsThat probably depends on what sort of blocks you ask for
20:25:12kugelyou could play with the mmu
20:25:35gevaertsIf you have one, yes
20:28:15gevaertsAnyway, using the codec buffer would gain us about a megabyte
20:28:52AlexPnot to be sniffed at
20:29:00kugelthe ams sansas all have a mmu
20:29:49kugelyou could set it up so that iram after the end of the audio buffer (i.e. physical location of the plugin buffer), and the plugin buffer after that
20:30:16gevaertsYou'd save doom and gain a few enemies in the process :)
20:30:41kugelgain enemies? those who don't like doom?
20:31:11kugelI can just kill them with a shotgun, since doom then works
20:31:54gevaertsah, right
20:32:00gevaertsYes, it seems to be a working solution
20:32:32AlexPalready? That was quick :)
20:34:10Buschelanybody with an iPod 4g, color or nano 1g willing to test fs#11707 ?
20:35:05Strife89Let me see what that FS# is. :)
20:35:40 Join jdut210 [0] (
20:35:59Strife89Oh, it allows for enabling/disabling line out?
20:36:41Strife89I'd need someone to make a build for me. :/
20:36:57Buschelfor which target?
20:36:58Strife89If I can find someone to do it, what exactly would you like tested?
20:37:12kugelgevaerts: I suppose you could then even build doom as overlay, using audiobuffer+codec buffer+plugin buffer
20:37:27***Saving seen data "./dancer.seen"
20:38:13kugelI'm not sure if it's very clean to exploit the adjacency of these buffers
20:38:34gevaertsmaybe not
20:39:01gevaertsBut then it shouldn't be too hard to e.g. have malloc decide on which buffer to use depending on requested size
20:39:04Strife89Buschel: Color/Photo.
20:39:26BuschelStrife89: jsut started the build, will take some minutes now
20:39:32Strife89Buschel: Okay.
20:40:12Strife89Buschel: What exactly would you like me to test for? I don't think I can do it here.
20:40:52Buschel1) mandatory: does on/off affect the headphone output? -> expected: no
20:41:31Buschel2) mandatory: does on/off affect the line out? -> expected: yes
20:41:58 Join Sarcastic_Dude [0] (
20:42:00Buschel3) optional: does on/off affect the battery runtime / battery current -> expected: minor
20:42:25 Quit markun (Read error: Connection reset by peer)
20:43:43bertrikCan anyone with a target containing a tea5767 tuner chip test ?
20:43:53bertrikA list of tea5767 containing targets is at
20:44:14bertrikThe GoGear HDD6330 has already been tested to work
20:47:34 Join markun [0] (
20:47:34 Quit markun (Changing host)
20:47:34 Join markun [0] (~markun@rockbox/developer/markun)
20:47:59 Quit KiwiCam (Quit: Leaving)
20:49:36Strife89Buschel: JSYK, I was planning on leaving on the hour.
20:52:10 Quit jdut210 (Quit: CGI:IRC (EOF))
20:53:04 Join W00fer [0] (
20:54:22Sarcastic_Dude3.7 is great, but I have some SD Card trouble (Fuze v1)
20:56:14 Quit Horschti (Ping timeout: 255 seconds)
20:56:35 Quit moos (Read error: Connection reset by peer)
20:56:48 Join moos [0] (moos@rockbox/staff/moos)
20:57:13Sarcastic_DudeIt only seems to notice my SD card after I unplug the player from the computer, if I turn off/on my Fuze, it fails to recognize it
20:58:02Sarcastic_Dude3.6 never had a problem with it
20:59:04 Quit LambdaCalculus37 (Quit: CGI:IRC 0.5.9 (2006/06/06))
20:59:12Strife89Buschel: My apologies for the impatience, but ... how's it coming?
20:59:13W00ferI saw some improvements in libatrac3. So should it be possible to playback more bitrates now?
20:59:37 Join Horschti [0] (~Horscht@xbmc/user/horscht)
20:59:53AlexPSarcastic_Dude: If you turn on the player withpout the card then insert it when Rockbox is running does it work?
21:00:09balintxgood evening, has anyone experienced a problem on sansa fuze(v2) that when the scrollwheel lights, the lcd itself not works?
21:00:17 Join kazaik [0] (
21:00:20Sarcastic_Dudeonly works *just* after I unplug the player from the computer
21:01:12Strife89Buschel: I need to leave in a few minutes. If the build has not compiled by then, you could e-mail it to me if you like.
21:01:15Sarcastic_DudeAlexP: indeed
21:01:34Sarcastic_Dude(fixed my WPS and it works great btw)
21:01:39AlexPcool :)
21:02:44Strife89I need to scoot. See y'all around! o/
21:02:54 Quit Strife89 (Quit: Vamoose.)
21:03:59AlexPbertrik: FS #11714 seems fine on H100 (it still thinks it has a radio anyway)
21:04:13bertriknice, thanks
21:04:37AlexPyou're welcome
21:05:48bertrikI'm hoping for b0hoon to show up, he has a target that has no tea5767 but could have
21:06:00 Quit moos (Read error: Connection reset by peer)
21:06:35 Join moos [0] (moos@rockbox/staff/moos)
21:07:04BuschelStrife98: sorry, had a private high prio interrupt
21:08:31 Join webguest34 [0] (
21:09:12BuschelStrife98: uploaded the .zip ->
21:09:21 Join webguest87 [0] (
21:09:38pixelmabertrik: my OndioFM has such a TEA5767 too, it's not listed though you can find it in the config file. Here Rockbox has to detect which of the tuner chips it has (older models have a different one)
21:09:43 Quit webguest34 (Client Quit)
21:09:47 Nick webguest87 is now known as CaptainKewllllll (
21:10:04 Join Alchimysta [0] (
21:10:27amiconnpixelma: No detection necessary, just checking the respective hw mask bit., as I already explained y'day
21:10:27 Quit Alchimysta (Client Quit)
21:10:29 Join sideral [0] (~sideral@
21:11:10pixelmaah hmm, so is testing of bertrik's patch necessary?
21:11:18 Join Alchimysta [0] (
21:11:25bertrikamiconn, do you want some #ifdef'ing of the detection routine for the archoses?
21:11:26 Join FOAD [0] (~dok@
21:13:09amiconnbertrik: Would make sense. All archoses with the respective bit set (i.e. all newer Ondio FM's) always have that tuner
21:13:32amiconnOtoh - does the Samsung tuner driver have a presence check? I guess it does
21:14:57pixelmahmm, and what about the Iaudios' alike tuner?
21:15:26kugelr28475 should have been two seperate commits
21:15:28amiconnThat's essentially a TEA5767
21:16:36 Quit pamaury (Remote host closed the connection)
21:16:37bertrikthe tea5767 driver seems to be the only that does not have a detection routine
21:16:55pixelmayes, I wanted to know if the changes in the patch needs checking there too
21:17:44pixelmaas I don't know if Rockbox just uses the TEA driver or something else
21:19:24Sarcastic_Dudeeveryone who contributed to making the Fuze work as a HID gets my compliments (this is really cool)
21:20:39pixelmaok, patch needs testing on Iaudio X5/M3 (or radio modded M5s ;) ) too then
21:20:56amiconnbertrik: Depending on how much binsize that adds, it may stay in. *If* we're going for ifdefing the detection, the TEA5767 driver is not the only one where we could do this (and save some binsize on targets where detection is unnecessary)
21:22:02amiconnThe Samsung detection is necessary on recorder v2 (because it may have a radio - early model), but not on fm recorder or (old) Ondio FM, since those always have it
21:23:22bertrikI would prefer not having to add #ifdefs and do the detection anyway even when not strictly necessary
21:24:19bertrikthe binsize penalty is just 90 bytes for the hdd1630 IIRC
21:25:39*amiconn would be more interested in SH1 binsize
21:25:50kugelwho's got a onda vx767?
21:30:45bertrikamiconn, which sh1 target exactly?
21:31:18amiconnOndio FM, as it's the only one that uses the TEA5767 driver
21:31:29 Part sideral
21:31:51amiconnHmm, I could try myself...
21:32:16 Join fml [0] (
21:32:57 Join Lear [0] (chatzilla@rockbox/developer/lear)
21:33:11bertrik+78 bytes for archos ondio fm
21:33:26 Join yawny [0] (user36@2610:150:c229::2)
21:33:36fmlkugel: you make android builds, right? What do you think about then?
21:33:49kugelanyone can :)
21:34:19 Quit elcan (Ping timeout: 240 seconds)
21:34:31kugeloh I didn't realize he already committed
21:35:55kugelI think your points are valid, sure
21:35:55 Quit Alchimysta (Quit: CGI:IRC (Ping timeout))
21:38:17fmlkugel: I won't change anything because I can't verify the changes. But I think someone who can (you?) should do it.
21:38:22 Quit Buschel (Ping timeout: 240 seconds)
21:38:49kugelwell, it's skin engine code, it doesn't have much to do with android
21:43:32 Join Sings_0f_Failure [0] (amoral.ncu@
21:44:19fmlYes, but the proposed changes affect hosted envs, e.g. android
21:48:32 Join Alchimysta [0] (
21:52:36fmlAlexP: could you please (as a native speaker) look at FS #11724 (seperate -> separate)?
21:54:40AlexPYeah, the correct spelling is separate
21:58:48AlexPI didn't look at the patch or anything :)
22:04:27Sings_0f_Failurei have some problems with Cue Support on my Sansa Clip+
22:04:47 Nick Sings_0f_Failure is now known as Songs_0f_Failure (amoral.ncu@
22:05:12fmlI'll leave it to JdGordon (FS #11724)
22:05:38 Quit S_a_i_n_t (Ping timeout: 265 seconds)
22:05:46 Quit W00fer (Quit: CGI:IRC (Ping timeout))
22:10:01 Join noamsml_ [0] (
22:12:05 Quit Horschti (Ping timeout: 252 seconds)
22:13:14 Quit noamsml (Ping timeout: 260 seconds)
22:13:17Sarcastic_Dudewell, I guess I'll go back to 3.6 until the issue is solved...
22:13:22Sarcastic_Dudelater people of Rockbox
22:13:25 Quit Sarcastic_Dude (Quit: Sarcastic_Dude)
22:13:30 Join n1s [0] (~n1s@rockbox/developer/n1s)
22:14:05 Join Horschti [0] (~Horscht@xbmc/user/horscht)
22:14:43 Quit evilnick_B (Quit: Page closed)
22:16:30 Quit Alchimysta (Quit: CGI:IRC (EOF))
22:16:48 Quit TheLemonMan (Quit: Help me, i got shot! *DIES*)
22:18:15 Quit Judas_PhD (Quit: This is a quitting message)
22:20:27 Quit fml (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
22:29:10 Quit Dreamxtreme (Ping timeout: 240 seconds)
22:31:52 Join MethoS- [0] (~clemens@
22:34:59 Quit esperegu (Read error: Connection reset by peer)
22:35:07 Join Dreamxtreme [0] (~Dre@
22:37:29***Saving seen data "./dancer.seen"
22:38:40 Join Strife89TX [0] (~cstrife89@
22:39:31 Quit csu (Ping timeout: 240 seconds)
22:39:43 Join csu [0] (
22:44:14 Part Songs_0f_Failure
22:44:54 Join toffe82_ [0] (
22:45:41 Quit bmbl (Quit: Verlassend)
22:45:51 Join toffe82__ [0] (
22:45:55 Quit benedikt93 (Quit: Bye ;))
22:46:18 Quit komputes (Quit: I haven't slept for ten days, because that would be too long.)
22:46:31 Quit toffe82 (Ping timeout: 240 seconds)
22:46:48 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
22:49:53 Quit toffe82_ (Ping timeout: 276 seconds)
22:50:21 Join toffe82 [0] (
22:51:16 Quit jgarvey (Quit: Leaving)
22:51:28 Quit JdGordon (Ping timeout: 252 seconds)
22:54:26 Quit toffe82__ (Ping timeout: 276 seconds)
22:55:16 Join y4n [0] (y4n@unaffiliated/y4ndexx)
22:55:39y4nHello. Is chained ogg planned? Or somehow supported?
22:55:49 Quit MethoS- (Remote host closed the connection)
22:55:50y4nMy file just won't play.
22:56:14LearNot supported.
22:56:31 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
22:57:18saratogay4n: its not planned and probably won't happen since its difficult to do on a portable
22:57:51y4nah well ok then, thanks :)
22:59:57 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
23:01:19 Join RoC_MasterMind [0] (
23:04:10 Quit Lear (Quit: ChatZilla 0.9.86 [Firefox 4.0b8pre/20101103031807])
23:04:11 Quit Horschti (Ping timeout: 255 seconds)
23:06:06 Join Horschti [0] (~Horscht@xbmc/user/horscht)
23:07:57 Join Ramsey[LC] [0] (~RamseyLC]
23:09:53 Join toffe82_ [0] (
23:10:02 Quit toffe82_ (Client Quit)
23:11:31 Quit toffe82 (Ping timeout: 240 seconds)
23:13:07 Quit saratoga (Quit: Page closed)
23:16:20 Quit anewuser ()
23:17:25 Quit Strife89TX (Quit: Low battery + no-electronics-at-dinner-policy)
23:20:34CIA-7New commit by 03jdgordon (r28483): Cleanup and oops spotted by Alexander Levin
23:21:17 Join anewuser [0] (anewuser@unaffiliated/anewuser)
23:21:33 Quit domonoky (Read error: Connection reset by peer)
23:22:11CIA-7r28483 build result: All green
23:26:56 Quit csu (Ping timeout: 276 seconds)
23:29:19 Quit y4n (Quit: only amiga makes it possible)
23:32:22 Join csu [0] (~csu@
23:32:47 Quit Horschti (Ping timeout: 276 seconds)
23:33:54kugelJdGordon: last and last->next always point to the same, doesn't it double free in _init()?
23:34:06 Join Horschti [0] (~Horscht@xbmc/user/horscht)
23:34:10kugelah no, nevermind
23:34:48kugelbut I think "else last->next = obj;" is uneeded
23:36:23bertrikDo we have a datasheet describing the portalplayer i2c controller?
23:39:37stacker55How can i add the arm-elf-eabi-gcc compiler to a debian install ? (not in the apt case ; main/contrib)
23:41:18bertrikIf you want the rockbox cross compiler, you can just run the cross-compiler script and it will install it in /usr/local/bin
23:42:38bertrikIs that what you want?
23:42:41stacker55that could work, except that it is not the spot where debian places everything
23:42:57stacker55so i'm unsure whether that would work
23:43:16gevaertsThat's exactly why it will work
23:43:37stacker55gevaerts: because it won't conflict is what you're saying ?
23:43:43gevaertsInstalling stuff that doesn't come from .deb packages in places like /usr/bin is a *bad* idea
23:44:00Bagderbut easy to do too if you want to...
23:44:29kugelJdGordon: just ignore what I said
23:44:54stacker55let's just try it and see where it goes :)
23:45:42bertrikIn my experience it's a painless process, just run the script, wait a bit and it magically works without further setup
23:46:38stacker55where is the script ?
23:46:42bertrikYou will need some packages for running the tools/ script but it will tell you what you're missing
23:46:57stacker55thank you :)
23:47:49 Join lixus001 [0] (
23:51:42 Join atlanx [0] (
23:51:57 Nick atlanx is now known as Atlanx (
23:52:19Atlanxhello ^^/
23:53:02AtlanxI have a little question about the font renaming of Rockbox v3.7
23:53:56AtlanxThere ( ) is mentioned that "nimbus-14" got renamed to "14-Nimbus".
23:54:34AtlanxBut what is with "nimbus+uming-14.fnt"?
23:54:57AtlanxI need it for displaying Japanese Kanjis.
23:56:38 Join Judas_PhD [0] (
23:56:44gevaertsWas that an official rockbox font?
23:57:14pixelmano, that is a version of the font where someone added the uming font to get CJK support
23:57:45kugelAtlanx: the font renaming is *ages* ago, it's not a 3.7 thing
23:57:54 Join Horscht [0] (~Horscht@xbmc/user/horscht)
23:58:33 Quit ender` (Quit: It's amazing how the the human mind does not process the the fact I used the the word "the" twice each time.)
23:58:50pixelmabut you can still use the one you found, only the names changed. The only thing you'll notice that a font without digits in the beginning of their name will be found at the end of the list when browsing for fonts... but you only have to set it once

Previous day | Next day